New Foundation of Biology Neuroimmune Biology, Volume 1
Neuroimmune Biology Series Editors
I. Berczi, A. Szentivanyi
Advisory Board
B.G. Arnason, Chicago, IL E. Artzt, Buenos Aires, Argentina P.J. Barnes, London, UK T. Bartfai, La Jolla, CA L. Bertok, Budapest, Hungary H.O. Besedovsky, Marburg, Germany J. Bienenstock, Hamilton, Canada C.M. Blatteis, Memphis, TN J. Buckingham, London, UK Ch. Chawnshang, Rochester, NY R. Dantzer, Bordeaux, France M. Dardenne, Paris, France N. Fabris, Ancona, Italy R.C. Gaillard, Lausanne, Switzerland Ch. George, Bethesda, MD R. Good, Tampa, FL R.M. Gorczynski, Toronto, Canada C. Heijnen, Utrecht, The Netherlands T. Hori, Fukuoka, Japan H. Imura, Kyoto, Japan
M.D. Kendall, Cambridge, UK E.A. Korneva, St. Petersburg, Russia K. Kovacs, Toronto, Canada G. Kunkel, Berlin, Germany L.A. Laitinen, Helsinki, Finland B. Marchetti, Catania, Italy L. Matera, Turin, Italy H. Ovadia, Jerusalem, Israel C.P. Phelps, Tampa, FL L.D. Prockop, Tampa, FL R. Rapaport, New York, NY S. Reichlin, Tucson, AZ R. Schmidt, Hannover, Germany A. Shmakov, Novosibirsk, Russia K. Skwarlo-Sonta, Warsaw, Poland E.M. Sternberg, Bethesda, MD D.W. Talmage, Denver, CO S. Walker, Columbia, MO A.G. Zapata, Madrid, Spain
New Foundation of Biology Neuroimmune Biology, Volume 1
Volume Editors Istvan Berczi Reginald M. Gorczynski
University of Manitoba, Winnipeg, Canada and University of Toronto, Toronto, Canada
2001 ELSEVIER AMSTERDAM
-
LONDON
-
NEW
YORK
-
OXFORD
-
PARIS
-
SHANNON
- TOKYO
ELSEVIER SCIENCE B.V. Sara Burgerhartstraat 25 P.O. Box 211, 1000 AE Amsterdam, The Netherlands
o 2001 Elsevier Science B.V. All rights reserved.
This work is protected under copyright by Elsevier Science, and the following terms and conditions apply to its use: Photocopying Single photocopies of single chapters may be made for personal use as allowed by national copyright laws. Permission of the Publisher and payment of a fee is required for all other photocopying, including multiple or systematic copying, copying for advertising or promotional purposes, resale, and all forms of document delivery. Special rates are available for educational institutions that wish to make photocopies for non-profit educational classroom use. Permissions may be sought directly from Elsevier Science Global Rights Department, PO Box 800, Oxford OX5 1DX, UK; phone: (+44) 1865 843830, fax: (+44) 1865 853333, e-mail:
[email protected]. You may also contact Global Rights directly through Elseviefs home page (http://www.elsevier.nl), by selecting 'Obtaining Permissions'. In the USA, users may clear permissions and make payments through the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA; phone: (+1) (978) 7508400, fax: (+1) (978) 7504744, and in the UK through the Copyright Licensing Agency Rapid Clearance Service (CLARCS), 90 Tottenham Court Road, London W1P 0LP, UK; phone: (+44) 207 631 5555; fax: (+44) 207 631 5500. Other countries may have a local reprographic rights agency for payments. Derivative Works Tables of contents may be reproduced for internal circulation, but permission of Elsevier Science is required for external resale or distribution of such material. Permission of the Publisher is required for all other derivative works, including compilations and translations. Electronic Storage or Usage Permission of the Publisher is required to store or use electronically any material contained in this work, including any chapter or part of a chapter. Except as outlined above, no part of this work may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission of the Publisher. Address permissions requests to: Elsevier Science Global Rights Department, at the mail, fax and e-mail addresses noted above. Notice No responsibility is assumed by the Publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made. First edition 2001 Library of Congress Cataloging in Publication Data A catalog record from the Library of Congress has been applied for.
ISBN: Series ISSN:
0 444 50754 X 1567 7443
~ T h e paper used in this publication meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). Printed in The Netherlands.
Acknowledgements This volume contains the presentations by invited speakers at the first Canadian Symposium on Neroimmune Biology, held in Winnipeg, June 9-11, 2000. Scientific Committee: Berczi I (Chair), Anisman H, Baines MB, Befus AD, Bienenstock J, Chow AD, Gorczynski RM, Moldofsky H, Nance D, Pittman Q, Pomerantz DK, Rivest S. Organizing Committee: Berczi I, Chow DA, Nance D, Baral E, Dawood M, Kisil FT, Kroeger E, Nagy E, Paraskevas F, Sabbadini ER, Warrington RJ. The symposium was followed by a workshop (June 12) on which the Canadian Network for Neuroimmune Biology (CANIB) was initiated. Website: http://cyboard.com/canib/ This conference has been supported by the Canadian Institutes of Health Research through the CIHR Opportunity Program. The University of Manitoba, the Faculty of Medicine and the Faculty of Graduate studies, University of Manitoba provided additional funding. We are grateful to Mrs. Carol Funk, who provided excellent service as secretary to the conference and also assisted us with the preparation of this volume. Ms. Valentina Tautkus, served as secretary and Technical Editor for this volume with much skill, diligence and devotion.
This Page Intentionally Left Blank
vii
Editorial to Volume I: Why Neuroimmune Biology? The importance of proper mindset for the maintenance of health and for general well-being has been known since prehistoric times [1]. Jesus Christ actually practised, perhaps unknowingly, healing of the sick and miserable, simply by giving them hope for recovery. References to faith healing are also present in the Koran and in other religious texts. Similar practices exist in primitive societies, where the "medicine man" provides spiritual and physical support to the sick. Darwin described the theory of evolution of the species over a hundred years ago [2], which is now regarded as scientifically proven [3], yet religion is still going strong, satisfying the "spiritual need" of enormous masses of people, especially in poor societies. Although there are suggestions for emotions similar to religion in animals, it is reasonably safe to suggest that the true religious mindset is only present in homo sapiens. Why this seemingly obligatory dependence on religion? Why spiritual satisfaction seems to be a compulsive need for so many people? The most fundamental difference between higher animals and man lies in intellectual capacity. Only man has to survive and prosper with the knowledge of certain death. Even today for most people on our planet life poses enormous difficulties that may include starvation, homelessness, devastating diseases, and no hope for improvements in the future. One may suggest that religion was, and still is, essential for providing hope for all those people who need help to maintain a balanced mindset that enables them to cope with the harsh realities of life. It appears that an optimistic mindset for these people is only possible through believing in God and Heaven, where there is eternal life and happiness without any suffering. Throughout history severe crisis situations, such as war, created terrible epidemics of infectious disease. Although not proven, the epidemics of deranged mindset may have contributed significantly to the spread of disease under these conditions. It is now emerging that emotional crisis may lead to severe depression, which is associated with disturbed neuroendocrine and immune functions. If these conditions persist, disease may follow [4-7]. Therefore, there is scientific evidence to indicate that the "spiritual need" of many people may actually stem from the enormous regulatory power of the human neuroimmune regulatory system over bodily functions. It needs to be set properly, in spite of unfavourable circumstances, so that maintenance of health and survival is maximally supported. That a strong belief in recovery from a serious illness has survival value stands the rigor of scientific scrutiny. Modern clinical trials of new drugs are conducted with control groups of patients that receive ineffective substances (placebo). Repeatedly it has been observed on the basis of objective parameters, that a significant percentage of placebo-treated patients show clinical improvements [8]. This may be interpreted as proof for the healing power of the proper mindset. Pathologists observed first that emotional factors and hormonal alterations have a major influence on the size of the thymus [9]. In 1936 Hans Selye discovered that noxious agents, when injected into rats, activate the ACTH-adrenal axis, which leads to the shrinkage of the thymus and of lymphoid organs [11, 12]. He produced evidence that glucocorticoids released by the adrenal gland caused the thymus atrophy. A similar "stress response" could be observed in rats by the emotional upset of being restrained from movement. Selye established that the hypothalamus-pituitary-adrenal-thymus axis was always activated
viii
under stressful conditions. Therefore, it was postulated by him that this axis was involved in the adaptation of animals to survive life-threatening challenges by various 'nocuous' agents [10-12]. It is only now that we are beginning to understand that indeed he was right. What he saw was the acute phase response (APR), which may indeed be regarded as an emergency defence reaction. Selye's legacy has been extended by our studies on the role of the pituitary gland in the regulation of the immune system. Growth hormone and prolactin was shown to maintain immunocompetence, whereas the ACTH-adrenal axis was found to exert an immunosuppressive effect. Sex steroid hormones have been designated as immunomodulators. These conclusions were made in reference to the adaptive immune response [13, 14]. It is now emerging that during febrile illness there is immunoconversion from the adaptive mode of immune reactivity to boosting natural immune mechanisms. The activation of the hypothalamus-pituitary-adrenal axis and the sympathetic outflow actually help the production of acute phase proteins by the liver and of natural antibodies by CD5+ B lymphocytes, which in turn command the immune system during acute illness [15, 16]. These developments fully support Selye's conclusion that the bodies defence mechanisms are mobilized after stress. It was discovered in 1949 by Szentivanyi and colleagues that the hypothalamus regulates the anaphylactic response in guinea pigs. Subsequent observations revealed that in laboratory animals, where the hypothalamus was imbalanced by lesions or by electrical stimulation, anaphylactic reactivity and antibody formation were altered significantly [17-21]. These experiments revealed that the nervous system has a dominant regulatory power over immune reactivity. In 1964 Korneva and Khai made similar observations [22]. The potential of sensory nerves to induce inflammation has been discovered by Jancso and co-workers [23]. This discovery ties in nicely with the above findings, indicating that the nervous system is capable of both causing and inhibiting inflammatory reactions. A compelling body of experimental evidence is available today, indicating the regulatory role of nerves in the inflammatory process. There is little doubt that inflammatory diseases have a significant input from the nervous system. The task is now to understand the mechanisms involved and to use the insights gained to the benefit of patients. The work of Pavlov called attention to the role of the mind in alimentary physiology by demonstrating that in dogs the expectation of receiving food leads to salivation (conditioned reflex). Later, the phenomenon of conditioning has been extended to numerous other bodily functions. In 1926 Metalnikov and Chorine showed that the Pavlovian rules of conditioning also apply to the immune system [24]. In modern times Ader and co-workers [25], MacQuin et al., [26] and Gorczynski and colleagues [27] provided rigorous scientific proof, indicating that the expectation of an immunological insult has a significant modulatory effect on subsequent immune responses. Therefore, immune responses may be conditioned in the classical Pavlovian sense. Moreover, it is now emerging that saliva itself has major immunoregulatory substances. In laboratory rodents the submandibular gland is a major site of production of these substances, which participate in the regulation of both mucosal and systemic immune reactions [28, 29]. In Persia, in Egypt and in the Roman Empire a healing power was attributed to fever. This view, which was supported by empirical observations, persisted till modern times. During the early nineteen hundreds an active search has been done by scientists for pyrogenic substances that could be used for fever therapy of diseases [30]. Such a substance was isolated by Boivin and colleagues from gram-negative bacteria [31], which is now known as bacterial lipopolysaccharide (LPS), or endotoxin. Now it is clear that LPS, a harmless substance by itself, is instantaneously recognized by the immune system. LPS induces cell activation, proliferation,
ix
cytokine production and the activation of immune-effector mechanisms. It also affects directly the central nervous system. If given systemically, LPS induces APR and boosts host defence. Clearly, there is evidence to support the idea that LPS has many beneficial effects, and that it can be used to good advantage in many life-threatening situations [15]. Similar homologous epitopes (homotopes) that are capable of instantaneous activation of the innate immune system exist in other microorganisms and in self-components [ 15, 16]. In 1975 Wannemacker and co-workers isolated the leukocyte endogenous mediator (LEM) of fever [32]. This was the first immune-derived molecule that mediated feedback signals towards the central nervous system. Later LEM was found to be identical with interleukin-1. It is now clear that IL-1 also serves as a feedback signal for pituitary hormone release [33-38]. Subsequently other cytokines, especially IL2, IL6, TNF-~t and interferon gamma were shown to regulate the secretion of pituitary hormones during systemic immune/inflammatory reactions [39]. It is also clear by now that the nerves have immunoregulatory function and provide feedback signals from lymphoid organs and from sites of immune/inflammatory reactions towards the central nervous system [40-43]. The Science of Immunology has evolved from observations that higher animals and man will acquire specific immunity after previous exposure to an infectious agent or toxin. Naturally microbiologists were most interested in this phenomenon as their major concern has been to fight infectious diseases. In order to take advantage of the body's phenomenal capability to develop specific resistance against pathogenic microbes after exposure, Jenner developed vaccination. This was a major advance in preventing infectious diseases and even today, still is a very important part of preventive medicine. Therefore, the traditional thinking in Immunology has revolved around the specific stimulus (antigen) that is capable of inducing immunity and it's interaction with the cells (lymphocytes) that are able to produce antibodies [44, 45]. It took some time to realize that cells, not antibodies, mediated some forms of immunity. With the advent of Cellular Immunology it has been discovered that lymphocytes are capable of producing antibodies in culture systems [46]. This fortified the view that the immune system was a largely autonomous system that went about the business of fighting 'foreign' intruders, while sparing 'self' from immune attack [45]. Seemingly there was no need for other control mechanisms, nor did it occur to the scientists pre-occupied with the prevention of infectious disease, that higher regulation of the immune system is in order, or actually it is required for normal function. Clearly, this system was mysteriously intelligent, capable of deciding with remarkable precision what to do. No other tissues/organs/systems were capable of self-non-self discrimination with such a remarkable precision and to display memory when stimulated by the same antigen/pathogen for the second time. However, the case for Neuroimmune interaction, which was first advanced by pathologists a century ago, grew stronger and stronger and by the mid-seventies half a dozen, or so, laboratories were preoccupied with studies in this area. The term 'bi-directional communication' between the Nervous and Immune Systems has been coined by Blalock and accepted enthusiastically by many people in the field. At the same time, it became obvious that both the immune system (which was watching self integrity) and the nerveous system, which innervated all tissues and organs, including the immune system, were in fact communicating with the entire organism. Indeed, it seems clear by now, there is much more to this interaction than 'bi-directional'. It is emerging, that we are dealing with a truly multi-directional, all-inclusive systemic regulatory network formed by the nervous-, endocrine- and immune systems, which controls all bodily functions of higher animals and man. This system is involved in conception and in the entire process of reproduction, in the growth and development of the fetus and of the newborn, in aging, in the process of daily life rhythms, in the sleep-wake cycle, in seasonal adjustments
and in most, if not all, pathological conditions, where defense, healing and regeneration are all influenced [47-49]. Clearly, the entire biology of higher organisms is based on this highly evolved and incredibly sophisticated regulatory system that is able to sense outside stimuli, including danger signs as well as to monitor and patrol the body for intruders, abnormalities and aberrations and correct, protect, heal and regenerate the organism as it may be required for the optimal maintenance of health and recovery from disease. Historic observations, the healing power of God and Jesus Christ, as well as every day events indicating the association of emotional difficulties and ill health maintain a very strong popular belief in the importance of mind-body interaction. In contrast, scientists pride themselves to only accept phenomena as true when sufficient scientific data are available in support of their validity. So far the scientific community at large does not fully appreciate the fundamental importance of the neuroimmune regulatory network. However, the time has arrived, when the role of this fundamental regulatory system may be submitted to scientific scrutiny. The human genome has been mapped and the experimental tools and sophistication, as well as the capacity of handling the vast amount of information that needs to be evaluated, are all available to undertake this task. There is little doubt, that fitting together the puzzle will soon become the next, and perhaps the last, frontier of Vertebrate Biology. Clearly, what is also required is to organize and interpret the scientific data as we go along. This is especially important because the relevant information is published in diverse specialty journals. The Science of Neuroimmune Biology deals with this systemic regulatory network, coordinating, organizing and interpreting the rapidly accumulating knowledge. The ultimate goal is to understand the function of higher organisms, including man, in their entire complexity. The objective of the book series, Neuroimmune Biology, is to provide regular assessments and interpretation of accumulated experimental evidence. It is hoped that this publication will enable the scientific community to keep abreast with essential advancements of our knowledge in a quest for understanding the Biology of higher organisms. We are pleased to present to the interested readers the introductory volume of this publication series and our plans for the forthcoming issues. We feel that it is high time to turn our attention to the organization and interpretation of the knowledge that has been accumulated in Biology. A new scientific field called Genomics has emerged recently, as attention is focused on the interaction of individual genes in the genome. In contrast with Genomics that still deals largely with events at the molecular and cellular level, our interest focuses on Integrative Physiology and Pathophysiology, never forgetting the milieu in which the cells (and their genes) of the body have to exert their functions. The term Neuroimmune Biology expresses this overall objective.
Istvan Berczi and Andor Szentivanyi
REFERENCES
1. 2. 3.
Sternberg EM. The Balance Within: The Science Connecting Health and Emotions. 2000; WH Freeman and Company, New York. Darwin C. The origin of species. Popular current edition. 1999; Bantam Classic. Mayr C. Darwin's influence on modern thought. Sci Amer. 2000; 283: 79-83.
xi
.
5. .
.
10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.
21. 22. 23. 24. 25. 26. 27.
Selye H. Stress and disease. Science 1955; 122: 625-31. Menkes DB, MacDonald JA. Interferons, serotonin and neurotoxicity. Psychol. Med. 2000; 30: 259-68. Young EA, Lopez JF, Murphy-Weinberg V, Watson SJ, Akil H. Hormonal evidence for altered responsiveness to social stress in major depression. Neuropsychopharmacology. 2000; 23:411-8. Arborelius L, Owens MJ, Plotsky PM, Nemeroff CB. The role of corticotropin-releasing factor in depression and anxiety disorders. J. Endocrinology 1999; 160: 1-12. Thompson WG Placebos: a review of the placebo response. Am J Gastroenterol. 2000; 95:1637-43. Hammar JA. The new views as to the morphology of the thymus gland and their bearing on the problem of the function of the thymus. Endocrinology 1921; 5: 543-73,731-60. Selye H. A syndrome produced by diverse nocuous agents. Nature (Lond) 1936; 138: 32. Selye H. Thymus and adrenals in the response of the organism to injuries and intoxication. Brit. J. Exp Path 1936; 17: 234-248. Selye H. The general adaptation syndrome and the diseases of adaptation. J Clin Endocrinology 1946; 6:117-230. Nagy E, Berczi I. Immunodeficiency in hypophysectomized rats. Acta. Endocrinology 1978; 89: 530-537. Berczi I. Immunoregulation by pituitary hormones. In pituitary function and immunity, Berczi I, Ed., CRC Press, Boca Raton, FL, 1986, pp. 227-240. Berczi I. Neurohormonal host defence in endotoxin shock. Ann NY Acad Sci 1998; 840: 787-802. Berczi I, Bertok L, Chow D. Natural immunity and neuroimmune host defence. Ann NY Acad Sci (in press). Filipp G, Szentivanyi A, Mess B. Anaphylaxis and nervous system. Acta. Med. Hung. 1952; 2: 163-173. Szentivanyi A, Filipp G. Anaphylaxis and the nervous system. Part II, Ann Allergy 1958; 16: 143-51. Filipp G, Szentivanyi A. Anaphylaxis and the nervous system. Part III, Ann Allergy 1958; 16: 306-11. Townley RG, Trapani IL, Szentivanyi A. Sensitization to anaphylaxis and to some of its pharmacological mediators by blockade of the beta adrenergic receptors. J Allergy 1967; 39: 177-97. Szentivanyi A. The beta adrenergic theory of the atopic abnormality in bronchial asthma. J Allergy 1968; 42: 203-232. Korneva EA and Khai LM. Effect of destruction of hypothalamic areas on immunogenesis. Fed. Proc. 1964; 23, T88. Jancso N. Role of the nerve terminals in the mechanism of inflammatory reactions. Bull. Millard Fillmore Hosp. (Buffalo. N.Y.) 1960: 7-53. Metal'nikov S, Chorine V. Roles des reflexes conditionels dans l'immunite. Ann Inst. Pasteur 1926; 40: 893-900. Ader R. Letter: Behaviorially conditioned immunosuppression. Psychosom Med 1974; 36:183-4. MacQueen G, Marshall J, Perdue M, Siegel S, Bienenstock J. Pavlovian conditioning of rat mucosal mast cells to secrete rat mast cell protease II. Science 1989; 243: 83-5. Gorczynski RM, Macrae S, Kennedy M. Conditioned immune response associated with allogeneic skin grafts in mice. J Immunol 1982; 129: 704-9.
xii
28. 29.
30. 31. 32.
33. 34. 35. 36. 37.
38. 39.
40. 41.
42. 43.
44. 45. 46. 47.
48. 49.
Sabbadini E, Berczi I. The submandibular gland: a key organ in the neuro-immunoregulatory network? Neuroimmunomodulation 1995; 2: 184-202. Befus DA, Forsythe P, D6ry RE, Mathison R, Davison JS. Immunomodulation by the Submandibular Gland. In Neuroimmmune Biology: New Foundation of Biology. Berczi I, Gorczynski R, Eds. Elsevier (in press). Nowotny A. In search of the active sites in endotoxins. In: Nowotny (Ed): Beneficial effects of endotoxins. Plenum Press, New York, 1993; 1-55. Boivin A, Mesrobeanu J, Mesrobeanu L. Techniques pour la preparation des polysaccharides microbiens sp6cifiques. CR soc. Biol 1933; 113: 490-492. Wannemacher RW Jr, Pekarek RS, Thompson WL, et al. A protein from polymorphonuclear leukocytes (LEM), which affects the rate of hepatic amino acid transport and synthesis of acute-phase globulins. Endocrinology 1975; 96: 651-661. Besedovsky HO, del Rey A, Sorkin E. Lymphokine-containing supernatants from Con A-stimulated cells increase corticosterone blood levels. J Immunol 1981; 126: 385-7. Berkenbosch F, van Oers J, del Rey A, Tilders F, Besedovsky H. Corticotropin-releasing factor-producing neurons in the rat activated by interleukin-1. Science 1987; 238: 524-526. Sapolsky R, Rivier C, Yamamoto G, Plotsky P, Vale W. Interleukin-1 stimulates the secretion of hypothalamic corticotropin-releasing factor. Science 1987; 238: 522-4. Nakamura H, Motoyoshi S, Kadokawa T. Anti-inflammatory action of interleukin 1 through the pituitary-adrenal axis in rats. Eur J Pharmacol 1988; 141: 67-73. Uehara A, Gottschall PE, Dahl RR, Arimura A. Interleukin-1 stimulates ACTH release by an indirect action which requires endogenous corticotropin releasing factor. Endocrinology 1987; 121: 1580-1582. Bernton EW, Beach JE, Holaday JW, Smallridge RC, Fein HG. Release of multiple hormones by a direct action of interleukin-1 on pituitary cells. Science 1987; 238: 519-521. McCann SM, Karanth S, Kamat A, Les Dees W, Lyson K, Gimeno M, Rettori V. Induction by cytokines of the pattern of pituitary hormone secretion in infection. Neuroimmunomodulation 1994; 1: 2-13. Pletsityi DF, Averi'anova LL. The influence of vagotomy on the development of experimental endocarditis. Dokl. Akad. Nauk. SSSR. 1966; 167: 238-40. Russian. Williams JM, Petersen RG, Shea PA, Schemdtie JF, Bauer DC, Felten DL. Sympathetic innervation of murine thymus and spleen: evidence for a functional link between the nervous and immune systems. Brain Res Bull 1981; 6: 83-94. Nance DM, Burns J. Innervation of the spleen in the rat: evidence for absence of afferent innervation. Brain Behav Immun 1989; 3: 281-90. Fleshner M, Goehler LE, Hermann J, Relton JK, Maier SF, Watkins LR. Interleukin-1 beta induced corticosterone elevation and hypothalamic NE depletion is vagally mediated. Brain Res Bull 1995;37: 605-10. Ehrlich P. Collected studies in immunity. John Wiley and Sons, New York, 1905. Burnet FM. The clonal selection theory of acquired immunity. The University Press, Cambridge, 1959. Nossal GJV, Lederberg J. Antibody production by single cells. Nature 1958; 181: 1419-1420. McCann SM, Lipton JM, Sternberg EM, Chrousos GP, Gold PW, Smith CC (Editors). Neuroimmunomodulation: Molecular aspects, integrative systems, and clinical advances. Ann NY Acad Sci 1998; 840. Csermely P (Editor). Stress of life, from molecules to man. Ann NY Acad Sci 1998; 851. Berczi I, Gorczynski R (Eds). Neuroimmune Biology: New Foundation of Biology. Elsevier, 2001.
xiii
Preface Observations indicating the dominant role of the central nervous system in the maintenance of health have been made since prehistoric times and the proverb "healthy body -- healthy mind" exists in many languages. Hans Selye was the first to study experimentally this mind-body interaction at McGuill University in Montreal. While attempting to isolate a hormone from the placenta, he injected his fractions into rats and observed the enlargement of the adrenal glands, the shrinkage of the thymus and of lymphoid organs. However, despite repeated attempts to purify the putative hormone, the activity was invariably lost. Finally, in 1936, he came to the conclusion that he was dealing with a non-specific reaction whose nature remained unknown (Nature, 138; 32, 1936). Supporting this conclusion, he even found that when he injected formalin into rats a similarly dramatic reaction occurred that included gastrointestinal haemorrhages. All 'noxious' agents and even emotional upset (restraining rats on the laboratory bench) could elicit this syndrome. He called this phenomenon "stress", and the eliciting agents/situations "stressors". Selye argued that stress elicited a defence reaction, which he named "the general adaptation syndrome". After an initial enthusiastic response from scientists to these ideas, contradictions and confusion prevailed and regrettably Selye's achievements went unrecognised during the 50 years he dedicated to the understanding of this phenomenon. It has now been recognized that he was the first to "discover" the existence of a hypothalamicpituitary-adrenal-thymus axis, and he consistently maintained till the end of his life that this axis played an important role in the adaptation of higher animals and man to various physical, chemical, biological and emotional challenges. Only over the past 2-3 decades has Selye's work been appreciated and interpreted. Although he knew little about the immune system, he discovered its conversion from the adaptive mode of reactivity to the development of the so-called acute phase response, which can be understood as an amplification of natural immune defence. The neuroendocrine response he observed is fundamental to this conversion. He was fully correct in concluding that this reaction is a general and adaptive defence reaction. The first Canadian Conference on Neuroimmnune Biology, and this volume which reports the papers presented at that conference, are dedicated to Selye's memory and to his life-time achievements. Andor Szentivanyi was the guest of honour and gave a conference-opening lecture entitled: "Studies on the hypothalamic regulation of histamine synthesis". In this discussion he reported the demonstration, by contemporary scientific methodology, of the mechanism(s) for a fundamentally important discovery he and his colleagues made more than 50 years ago. In 1949 his group discovered that the central nervous system seemed to have broad regulatory power over immune reactions (Acta. Med. Hungarica 3(2): 163, 1952). As a young medial student, Szentivanyi observed catecholamine-resistance in an asthmatic patient, who died in spite of aggressive treatment with adrenaline. This incident inspired him to do animal experiments and to dedicate his entire research career to the clarification of the role of the central nervous system in immune and inflammatory reactions. His subsequent experiments, published in a wide range of Internationally acclaimed journals, demonstrated the important role of the beta-adrenergic receptor in the regulation of immune and inflammatory conditions. It was humbling to hear from this distinguished guest that finally, after a long (over 50 years) career in science, he was able,
xiv
for the first time, to present his experimental results to an audience that was genuinely interested in the subject. The conference and delegates were privileged to honour Dr. Szentivanyi for his fundamental discoveries and remarkable achievements in Nueroimmune Biology, and to welcome him as an Editor for the Proceedings. He has been instrumental to the formulation of the idea of this series as well as to its realization. Only now, some 65 years after Selye's discovery of the stress reaction and 50 years after Szentivanyi's unequivocal demonstration of immunoregulation by the CNS, are we beginning to understand so-called "mind-body" interactions at a cellular and molecular level. There is now a growing consensus amongst the general scientific community that the nervous-, endocrine-, and immune systems form a systemic regulatory network. This network is fundamental to the maintenance of the entire life cycle of higher animals and man in health and disease. It seems clear this regulatory network coordinates and maintains all physiological functions, including reproduction, and further commands host defence mechanisms in life-threatening circumstances and in disease. The term, Neuroimmune Biology, has been adopted to define this new scientific discipline. The realization that the immune system is part of the systemic regulatory network that regulates the function of higher organisms provides important new foundations to Biology. The objective of this book, and the book series it has spawned, termed Neuroimmune Biology, is to present a coordinated and integrated view of the growing body of knowledge which is rapidly accumulating in this area. Our ultimate goal is to achieve a more thorough understanding of higher organisms in their entire complexity.
Istvan Berczi Reginald Gorczynski .,
.:
! Hans Selye
Andor Szentivanyi
XV
List of Corresponding Authors
Jack P. Antel
Montreal Neurological Institute, Department of Neurology and Neurosurgery, McGill University, Montreal QC, Canada Malcom G. Baines
Department of Microbiology & Immunology, McGill University, Room 44, 3775, University Street, Montreal QC, Canada H3A-2B4 A. Dean Befus
Pulmonary Research Group, Department of Medicine, The University of Alberta, Room 574 Heritage Medical Research Centre, Edmonton, AB, Canada T6G-2S2 lstvan Berczi
Department of Immunology, Faculty of Medicine, The University of Manitoba, 795 McDermot Avenue, Winnipeg, MB, Canada R3E-OW3 John Bienenstock
Faculty of Health Sciences, McMaster University, 1200 Main Street West, Room 2El, Hamilton, ON, Canada L8N-3Z5 Peter Bretscher
The University of Saskatchewan, Department of Microbiology and Immunology College of Medicine, A231 Health Sciences Building, 107 Wiggins Road, Saskatoon, SK, Canada S7N-5E5 Donna A. Chow
Department of Immunology, Faculty of Medicine, The University of Manitoba, 795 McDermot Avenue, Winnipeg, MB, Canada R3E-OW3 Joe S. Davison
Department of Physiology, Faculty of Medicine, The University of Calgary, 3330 Hospital Drive N.W., Calgary, AB, Canada T2N-4N1 Judah A. Denburg
Director, Division of Clinical Immunology & Allergy, Department of Medicine, Faculty of Health Sciences, McMaster University, Hamilton, ON, Canada Gordon Ford
University of Calgary, 4A186 Holy Cross Ambulatory Care Centre, Rocky View, General Hospital 7007-14th Street, Calgary, AB, Canada T2V-1P9
xvi
Reginald M. Gorczynski Department of Surgery & Immunology, University of Toronto, The Toronto Hospital, CCRW 2-855, 200 Elizabeth Street, Toronto, ON, Canada M5G-2C5 Kent HayGlass Department for Immunology, Faculty of Medicine, The University of Manitoba, 730 William Avenue, Winnipeg, MB, Canada R3E-OW3 Teresa Krukoff Department of Cell Biology, Faculty of Medicine & Dentistry, The University of Alberta, Edmonton, AB, Canada T6G-2H7 Alexander Kusnecov Department of Psychology, Biopsychology and Behavioral Neuroscience Program Rutgers University, 152 Frelinghuysen Road, Piscataway, NJ, USA 08855 Julio Licinio UCLA Department of Psychiatry and Biobehavioral Sciences, 3357A Gonda (Goldschmied) Center, Los Angeles, CA, USA 90095-1761 Giamal N. Luheshi Department of Neuroscience, Douglas Hospital Research Centre, 6875 Boulv. LaSalle, Verdun, QC Canada H4H-1R3 Brian MacNeil Department of Pathology, P220 Pathology Building, Faculty of Medicine, The University of Manitoba, Winnipeg, MB, Canada R3E-OW3 Zul Merali Psychology & Molecular Medicine, 11 Marie Curie Room, 214 Vanier Building, Ottawa, ON, Canada K1N-6N5 Harvey Moldofsky The University of Toronto, Centre for Sleep & Chronobiology, Toronto Western Hospital University Health Network, 399 Bathurst Street, Room MP14-308, Toronto, ON, Canada M5T-2S8 Eva Nagy Department of Immunology, Faculty of Medicine, The University of Manitoba, 795 McDermot Avenue, Winnipeg, MB, Canada R3E-OW3 Dwight Nance Department of Pathology, University of Manitoba, P220 Pathology Bldg., Winnipeg, MB, Canada R3E-OW3 Trevor Owens Neuroimmunology Unit, Montreal Neurological Institute, 3801 University Street, Montreal, QC, Canada H3A-2B4
xvii
Quentin Pittman Neuroscience Research Group and Department of Medical Physiology, The University of Calgary, 3330 Hospital Drive N.W., Calgary, AB, Canada T2N-4N1 David K. Pomerantz Department of Physiology, University of Western Ontario, London, ON Canada, N6A-3K7 Robert J. Rapaport Mount Sinai Diabetes Center, 1200 Fifth Avenue (101 st Street), First Floor, New York, NY, USA 10029 Serge Rivest Molecular Endocrinol Lab., CHUL Res Ctr., Laval University, 2705 Boul. Laurier, Quebec, QC, Canada G1V-4G2 Edris R. Sabbadini Department of Immunology, Faculty of Medicine, The University of Manitoba, 795 McDermot Avenue, Winnipeg, MB, Canada R3E-OW3 Vijendra Singh Department of Biology, Utah State University, 5305 Old Main Hill Logan, UT USA 84322-5305 Andrzej Stanisz HSC-3N5C, McMaster University, 1200 Main Street West, Hamilton, ON, Canada L8N-3Z5 Lucia Stefaneanu Division of Pathology, St. Michael's Hospital, 30 Bond Street, Toronto, ON Canada M5B-1 W8 Esther M. Sternberg Section of Neuroendocrine Immunology and Behavior, National Institutes of Health, Bldg 10, Room 2D46, Bethesda, MD, USA 20892-1284 Andor Szentivanyi Department of Internal Medicine, University of South Florida, Box 9, 12901 Bruce B. Downs Blvd., Tampa, FL, USA 33612-4799 Richard J. Warrington Departments of Internal Medicine & Immunology, Faculty of Medicine, The University of Manitoba, Winnipeg, MB, Canada R3E-OW3
This Page Intentionally Left Blank
xix
Contents
Acknowledgements ..........................................................
v
Editorial . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
vii
Andor Szentivanyi and Istvan Berczi Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
xiii
Istvan Berczi and Reginald M. Gorczynsld
List of C o r r e s p o n d i n g A u t h o r s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
I.
xv
Introduction
Neuroimmune Biology -- An introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3
Istvan Berczi Studies on the Hypothalamic Regulation of Histamine Synthesis . . . . . . . . . . . . . . . . . .
45
Andor Szentivanyi, Istvan Berczi, Denyse Pitak, Allen Goldman
II. N e u r o i m m u n e R e g u l a t o r y M e c h a n i s m s
Introduction: II. Neuroimmune Regulatory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
57
Reginald M. Gorczynsld Dynamics of Immune Responses: Historical Perspectives in our Understanding of Immune Regulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
61
Kent T. HayGlass Cell-to-cell Interaction and Signaling within the Immune System: Towards Integrating Mechanism and Physiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
71
Peter A. Bretscher, Nahed Ismail, Nathan Peters, Jude Uzonna Regulation of the Immune Response within the Central Nervous System . . . . . . . . . . . .
87
Jack Antel Regulatory Circuits of the Pituitary Gland . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Lucia Stefaneanu
99
XX
Neuroendocrine Stress and Inflammatory Disease: From Animal Model to Human Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
115
Esther M. Sternberg, Mojdeh Moghaddam Immunoregulation by the Sympathetic Nervous System . . . . . . . . . . . . . . . . . . . . . . . . .
121
Dwight M. Nance, Brian J. MacNeil Behavioral and Central Neurochemical Consequences of Cytokine Challenge: Relationship to Stressors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
141
Zul Merali, Hymie Anisman, Shawn Harley Proinflammatory Signal Transduction Pathways in the CNS During Systemic Immune Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
163
Serge Rivest, Sylvain Nadeau, Steve Lacroix, Nathalie Laflamme Nitric Oxide in Neuroimmune Feedback Signaling . . . . . . . . . . . . . . . . . . . . . . . . . . . .
175
Teresa L. Krukoff, Wen@ W. Yang
III. Neuroimmune Mechanisms in Physiology Introduction: III. Neuroimmune Mechanisms in Physiology . . . . . . . . . . . . . . . . . . . . .
207
Reginald M. Gorczynski A Model of Neuroimmune Communication: Mast Cells and Nerves . . . . . . . . . . . . . . .
195
John Bienenstock Immunomodulation by the Submandibular Gland . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
203
A. Dean Befus, Paul Forsythe, Rene E. Dgry, Ronald Mathison, Joseph S. Davison Glandular Kallikrein in Immunoregulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
225
Edris Sabbadini, Eva Nagy, Alexander Viir6s, Gertrude V6r6sova, Fred T. Kisil, Istvan Berczi Understanding Classical Conditioning of Immune Responses . . . . . . . . . . . . . . . . . . . .
237
Reginald M. Gorczynski Sleep, Health and Immunocompetence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
255
Harvey Moldofsky, Wah-Ping Luk, Jodi Dickstein Interactions Between the Immune System and the Testis . . . . . . . . . . . . . . . . . . . . . . . .
269
David K. Pomerantz Leptin and Cytokines: Actions and Interactions in Fever and Appetite Control . . . . . .
Giamal N. Luheshi
283
xxi
IV. Neuroimmune Host Defence Introduction: IV. Neuroimmune Host Defence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
295
Reginald M. Gorczynski Fever and Antipyresis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
297
Quentin J. Pittman, Abdeslam Mouihate, Marie-Stephanie Clerget The Salivary Gland Peptides: Their Role in Anaphylaxis and Lipopolysaccharide (LPS)-Induced Inflammation . . . . . . . . . . . . . . . . . . . . . . . . . .
307
Joe S. Davison, Dean Befus, Ronald Mathison Olfactory Stimuli and Allo-Recognition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
313
Malcolm G. Baines Natural Immune Regulation of Activated Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
331
Donna A. Chow, Ricky Kraut, Xiaowei Wang
V. Neuroimmune Pathology Introduction: V. Neuroimmune Pathology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
347
Istvan Berczi Stress, Health and the Immune Response: Reciprocal Interactions Between the Nervous and Immune Systems . . . . . . . . . . . . . . .
351
Alexander W. Kusnecov, Alba Ross#George, Scott Siegel Cytokines in the Brain: From Localization and Function to Clinical Implications . . . . . . . . . . . . . . . . . . . . . . . .
365
Julio Licinio, Ma-Li Wong Neurogenic Inflammation: Role of Substance P . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
373
Andrew M. Stanisz Lupus as a Model of Neuroimmune Interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
379
Judah A. Denburg, Boris SaMc, Henry Szechtman, Susan D. Denburg The Pathogenesis of Encephalitis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Trevor Owens, Elise H. Tran, Mina Hassan-Zahraee, Alicia Babcock, Michelle L. Krakowski, Sylvie Fournier, Michael B. Jensen, Bente Finsen
387
xxii
VI. Clinical Neuroimmune Biology Introduction: VI. Clinical Neuroimmune Biology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
401
Istvan Berczi Growth Hormone Therapy and Immune Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
403
Robert Rapaport, Robert Moghaddas The Role of Prolactin in Systemic Lupus Erythematosus . . . . . . . . . . . . . . . . . . . . . . . .
409
Richard Warrington, Tim McCarthy, Eva Nagy, Kingsley Lee, Istvan Berczi Combination Immunotherapy of Cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
417
Eva Nagy, Istvan Berczi, Edward Baral, John Kellen The Influence of Reproductive Hormones on Asthma . . . . . . . . . . . . . . . . . . . . . . . . . .
433
Gordon T. Ford, Candice L. Bjornson, lan Mitchell, M. Sarah Rose Neuro-Immunopathgenesis in Autism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
447
Vijendra K. Singh Skin inflammation and Immunity After Spinal Cord Injury . . . . . . . . . . . . . . . . . . . . . .
459
Brian J. MacNeil, Dwight M. Nance
Subject Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
475
INTRODUCTION
This Page Intentionally Left Blank
New Foundation of Biology Edited by I. Berczi and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
N e u r o i m m u n e B i o l o g y -- A n I n t r o d u c t i o n
ISTVAN BERCZI
Department of Immunology, Faculty of Medicine, The University of Manitoba, Bannatyne Campus, 32-795 McDermot Avenue, Winnipeg, Manitoba R3E OW3, Canada
ABSTRACT That a healthy mind is fundamental to general well being has been recognized since prehistoric times and proverbs analogous to "Healthy body - healthy mind" exist in many languages. A century ago pathologists noted first that the size of the thymus was profoundly influenced by emotional events and by neuroendocrine aberrations. Hans Selye discovered first (1936) that the hypothalamus-pituitary-adrenal axis, which is activated by diverse 'nocuous' stimuli, leads to the rapid involution of the thymus. He coined this phenomenon as the 'stress' response. Selye established that stress results in the development of the general adaptation syndrome which is characterized by elevated resistance to diverse insults. Andor Szentivanyi and colleagues discovered (1949) that hypothalamic lesions prevent anaphylactic death in guinea pigs. This is the first experimental evidence for the sweeping regulatory power of the nervous system over violent, life threatening immune reactions. That the nervous system also controls the inflammatory response was first demonstrated by Milos Jancso and co-workers (1964). These fundamental discoveries were not followed by a burst of research activity. Progress has been slow because of the lack of basic knowledge and because of the immense technical difficulties encountered. In the seventies a handful of laboratories started to re-examine various aspects of neuroimmune-interaction. It was established that pituitary hormones have the capacity to stimulate, inhibit and modulate immune responses. Placental and pituitary hormones are also involved in the development of the immune system and maintenance of immunocompetence. It was also described that lyphoid organs are innervated and that neurotransmitters and neuropeptides are important immunomodulators. It became gradually apparent that immune derived cytokines and nerve impulses serve as feedback signals towards the neuroendocrine system. Compelling evidence was produced, indicating that immune reactions may be conditioned in the classical pavlovian sense and that emotions affect immune function of various organs and tissues, and in reproduction. It is also becoming obvious that Selye's general adaptation syndrome really corresponds to the acute phase response. This is a multi-faceted and highly co-ordinated systemic defence reaction, which involves the conversion of the immune system from a specific, adaptive mode of reactivity to a rapidly amplifiable polyspecific reaction mediated by natural immune mechanisms. Immunological (poly)specificity is assured by profoundly elevated levels of natural antibodies and liver derived proteins.
Much has been learned about the regulation of cell activation, growth and function from immunological studies. Burnet's clonal selectional theory designates antigen as the sole activator. Bretcher and Cohn recognised first that at least 2 signals are required. This was followed by numerous studies on cell-to-cell interaction within the immune system and led to our current understanding of the importance of cell adhesion molecules and cytokines in cell activation and proliferation. This, coupled with the available information about the mechanisms of action of hormones and neurotransmitters, of signal transduction and nuclear regulatory pathways paves the way to understanding how higher organisms function in their entire complexity. It is now apparent that the Nervous- Endocrine- and Immune-systems form a systemic regulatory network, which is capable of regulating all aspects of bodily functiuons in health and disease. Thus, Neuroimmune Biology provides new foundations to Biology.
1.
INTRODUCTION
Observations indicating that the central nervous system has a fundamental role in the maintenance of health has been made since prehistoric times and is referred to in proverbs of many languages. The healing power of mind and faith provides one of the important foundations of religion and is described in many religious texts. These phenomena are also observed in modern medicine and is known as the placebo effect. It has been demonstrated repeatedly by exact scientific methodology that patients treated with placebo in controlled medical trials do in fact show significant improvement clinically in the absence of effective treatment. In ancient Persia, Egypt and in the Roman Empire fever has been regarded as a reaction with healing power. This view was maintained until modern times and during the early nineteen hundreds pyrogenic substances have been developed for the purposes of fever therapy [1-3]. About a century ago pathologists observed that acromegaly was frequently associated with thymic hyperplasia. Hammar [4] described that the thymus frequently showed involution under the influence of environmental or emotional factors. In contrast, thymic hyperplasia was associated with castration, Graves' Disease, Addison's Disease and acromegaly. Smith described in 1930 that in hyposectomyzed (Hypox) rats the thymus regressed in weight to less than half of that of controls. In partially Hypox rats there was no involution [5]. In 1936 Hans Selye documented that the pituitary-adrenal-thymus axis was activated by various nocuous stimuli, which led to the involution of the thymus and of the lymphoid organs [6, 7]. Moreover, Selye has established that the bursa of Fabricius in chickens was extremely sensitive to steroid hormones [8]. Within ten years Selye has proposed the theory of general adaptation syndrome (GAS) [9] on the basis of his experiments. He pointed out that this is a general reaction that leads to resistance of the organism to various insults. Selye's scheme of GAS is shown in Figure 1, updated with current information. In 1949 Selye discovered that the inflammatory response is regulated by corticosteroids [10]. In his article entitled "Stress and Disease" he proposed that deficient host defense due to abnormalities of neuroendocrine factors may lead to disease [11 ]. Selye recognized the importance of mast cells in pathology and performed numerous studies in this respect. He summarized the knowledge about mast cell in a book [ 12], which is a lasting contribution on the subject.
Clinical shock Loss of body weight+N Gastrointestinal ulcers Temporary rise in plasma potassium level Temporary fall in plasma
4
Nonspecific damage
Pathway unk . . . .
."1 ." . .." I .-'" ~"I ...." ~ j o
hut not involving hypophysealand/or adrenal hyperactivity
ch,or~.~,eve,
...-"
~
~ :~
~ ~~
.." : from injured ('ells
9
i
Cytokines
: :
.4
(ells, endothelium, platelets, leukocvtes
.
........
...... ~.
..-""
Inflammatory
cytokines
"
:: 9
'
T
/ (~ M S H ~/.~l~ END
iP/ k
~
F
amus, CRF
)
;
~.
~ ~ "~"
on~. o
~ oc oI
~ E o .~ ~ ~1
Testis
Involutionof male accessory
T~ ~ ~ J: ~
.....
Anomalies of sexual cycle
~
~
~. ,e ................................ ~
cathecolamines
....... ~
Carbohydrate metabolism insulin i
/
Thymusand olher
i
~ymphatic
~|
/ ~ ~.
/
organs
/
/
/ /
gans
Ovary
.....
:
I~: Liver
gland
Cessationof lactation
.t .~0 | - m e d u l l a
:
:
9
~ ] "5. I ~
'- ~ 3 .~ ~.1=
Adrenal cortex Fnlarge,.~t andi hormones sed producti(m of corticoid
x: Brain- 9I f v/ nr"o t h a l - I/ "
Increased production of corficotropi( hormone
.."
from mast
! :
,-
Hypophysis-~-~ Mammary ~ ~|---~ oc
C h e m o k i n e s ~'"
:
"9
Inhibition
.: ..~ Q,
:i
:
Somatic growth
~ ~~
?
Rlucagon i
Electrolyte metabolism
sodi.... t~,,,ion
Involution ~'? i
Circulating
lymphocytes Lymphopenia increased g.globulin and antibodies in blood natural antibodies
Blood ve Periarterilisnodosa
oints M~tis Aschoff's nodules
~Nephros,derosis ~
Polyarthritis
~?
Blood pressure Hyperlension
Figure 1. Functional interrelations during the general adaptation syndrome. This figure is modified from Selye [9] by updating it with recent information. Solid arrows and the two broken arrows on the top with bold text is Selye's original figure. Recently identified pathways are indicated with dotted arrows and with text in italics. The text below is also from Selye. "Schematized drawing indicating that non-specific damage causes clinical shock, loss of body weight and nitrogen, gastro-intestinal ulcers, temporary rise in plasma potassium with fall in plasma C1, through unknown pathways (nervous stimulus?, deficiency?, toxic metabolites?) but manifestly not through the stimulation of the hypophyseoadrenal mechanism. This is proven by the fact that the above manifestations are not prevented either by hypophysectomy or by adrenalectomy; they even tend to be more severe in the absence of either or both of these glands. Non-specific damage, again through unknown pathways, also acts upon the hypophysis and causes it to increase corticotropic hormone production at the expense of a decreased gonadotropic, lactogenic and growth hormones. The resulting corticotropic hormone excess causes enlargement of the adrenal cortex with signs of increased corticoid hormone production. These corticoids in turn cause changes in the carbohydrate (sugar active corticoids) and electrolyte metabolism (salt-active corticoids) as well as atrophy of the thymus and the other lymphatic organs. It is probable that the cardiovascular, renal, blood pressure and arthritic changes are secondary to the disturbances in electrolyte metabolism since their production and prevention are largely dependent upon the salt intake. The changes in y-globulin, on the other hand, appear to be secondary to the effect of corticoids upon the thymicolymphatic apparatus. We do not know as yet, whether the hypertension is secondary to the nephrosclerosis or whether it is a direct result of the disturbance in electrolyte metabolism caused by the corticoids. Similarly, it is not quite clear, as yet, whether corticoids destroy the circulating lymphocytes directly, or whether they influence the lymphocyte count merely by diminishing lymphocyte formation in the lymphatic organs. Probably both these mechanisms are operative".
Selye made all his contributions without knowing the function of the thymus, lymph nodes or the bursa of Fabricius. The function of these organs was understood in the sixties and early seventies, decades after he published his seminal papers on stress. With the advent of the science of Immunology it became clear that stress has a profound immunosuppressive effect and increases the susceptibility to infectious disease. These findings seemed to contradict Selye's conclusion that the response to stress was an adaptive defense reaction which increased the resistance of the body to various noxious agents. Andor Szentivanyi and his colleagues were the first to document that the nervous system has a dominant regulatory power over immune reactions. As a medical student Szentivanyi observed that adrenaline treatment was ineffective to alleviate an asthmatic attack in a patient. This clinical observation inspired him to do experiments in guinea pigs using anaphylactic shock as a model system. Hypothalamic lesions inhibited the development of anaphylactic shock in immunized animals [13]. Tuberal lesions (TBL) of the hypothalamus were effective in preimmunized guinea pigs and in later experiments also in rabbits to inhibit anaphylactic reactions elicited by the intravenous application of the antigen. Antibody production was also inhibited if the lesions were induced prior to immunization. The reaction of antibodies with the specific antigen was not affected by such lesions, nor was the release of tissue material mediating anaphylaxis. TBL temporarily increased the resistance of the animals to histamine and inhibited the anaphylactic reaction even when the animals were provided with passively transferred antibody, which elicited lethal shock in normal animals. The Schultz-Dale test, which was performed with small pieces of intestine in vitro, was also inhibited when the animals were subjected to TBL. The Arthus reaction, turpentine induced inflammation and the Sanarelli-Schwartzmann phenomenon were unaffected by hypothalamic lesions. Lesions inflicted in other areas of the hypothalamus or the central nervous system were ineffective in modulating immune phenomena. Electrical stimulation of the mamillary region of the hypothalamus had an inhibitory effect on the anaphylactic response and increased the resistance of animals to histamine [ 14-16]. Szentivanyi devoted his entire career to the study of allergy and asthma. Animal experiments pointed to the importance of the beta-adrenergic receptor in these reactions [17]. In 1968 Szentivanyi had synthesized the knowledge and all his findings in a review article, entitled, "The beta-adrenergic theory of the atopic abnormality in bronchial asthma" [ 18]. He concluded that bronchial asthma, whether it is due to "extrinsic" or "intrinsic" causes, is ultimately elicited by the same mediators, such as histamine, serotonin, catecholamines, slow reactive substances plus cytokines. These are released during asthmatic reactions and should be considered as additional group of mediators in many tissues and in most species. Glucocorticoids are natural inhibitors of inflammation. He proposed that the atopic abnormality in asthma is due to the abnormal function of the [3-adrenergic system, irrespective of what triggered the reaction: "The beta adrenergic theory regards asthma not as an 'immunological disease' but as a unique pattern of bronchial hypersensitivity to a broad spectrum of immunological, psychic, infectious, chemical and physical stimuli. This gives to the antigen-antibody interaction the same role as that of a broad category of non-specific stimuli which function only to trigger the same defective homeostatic mechanism in the various specialized cells of bronchial tissue". Szentivanyi remained faithful to the idea of beta-adrenergic malfunction in atopy and asthma. This is the common thread that connects the numerous papers reviews, book chapters and books he published. He studied c~- and [3-adrenergic receptors; adenylcyclase, cyclic-AMP and signal transduction; isolated, characterized and pharmacologically modulated phosphodiesterase; observed the systemic effect of immunization and of endotoxin on the adrenergic and cholinergic systems, on metabolism and on immune inflammatory mediators; performed clinical studies on asthma and related conditions. His major observations were:
1. Beta-adrenergic sub-sensitivity did exist in patients with atopic dermatitis who never received adrenergic medication. This indicates that therapeutic desensitization cannot account for the dysfunction of the beta-adrenergic system [19]. 2. The beta- adrenergic reactivity of lung tissue of lymphocytes and bronchocytes from patients with atopic asthma was found to be abnormal and various patterns of drug vs. disease-induced sub-sensitivity could be recognized [20-25]. 3. Bronchial hyper-reactivity to cholinergic agents in asthma was not mediated through cholinergic mechanisms but it was caused by the adrenergic abnormality, which was due to the so called "denervation super-sensitivity" [26-29]. Lymphocytes of asthmatic patients showed a significant decrease in adrenaline binding to beta-adrenergic receptors, which was independent of therapy [21, 22, 25]. Szentivanyi also studied the effects of inflammation on [3-adrenergic receptors [30-33]. .
In 1964 Korneva and Khai [34] described that hypothalamic lesions in commonly used laboratory rodents (e.g. rabbits, guienea pigs, rats) inhibited the production of complement fixing antibodies. In 1960 Miklos Jancso and co-workers reported that capsaicin is a sensory irritant and that repeated local or systemic administration to rats, mice and guinea pigs causes desensitization, which involves interference with pain receptors. Systemic pretreatment of animals with capsaicin or repeated local applications prevented the inflammatory response, indicating the involvement of the nervous system. This was later confirmed by experiments performed on denervated tissues. These observations indicated the existence of a distinct form of inflammation, which depends on sensory nerve innervation. The stimulation of C-fibers was necessary to induce this inflammatory response. The neurogenic inflammatory response was also demonstrated in man [35, 36]. It was known for some time that hormones, including those secreted by the pituitary gland, affect immune reactions [37]. However, only after the publication of systematic studies performed on hypophysectomized rats and in animals treated with bromocriptine [38-42], was the role of pituitary hormones seriously considered in immunoregulation by the scientific community. In 1975 Wannemacker and co-workers isolated the leukocyte endogenous mediator (LEM) of fever [43], which was the first immune-derived molecule identified, that mediated feedback signals towards the central nervous system. Later LEM was found to be identical with interleukin-1. That IL-1 also serves as a signal for pituitary hormone release was shown by a number of investigators in the early 1980's [44-49]. Subsequently other cytokines, especially IL2, IL6, TNF-~t and interferon gamma were shown to regulate the secretion of pituitary hormones during systemic immune/inflammatory reactions [50]. It is also clear by now that the nerves have immunoregulatory function and provide feedback signals from lymphoid organs and from sites of immune/inflammatory reactions towards the central nervous system (CNS) [51-54]. In 1926 Metalnikov and Chorine proposed first the behavioral modification of the immune response [55]. In 1933 Smith and Salinger [56] observed that asthmatic attacks were provoked in some patients with visual stimuli in the absence of the allergen. That immune reactions can be conditioned in the Pavlovian sense was demonstrated by Ader, MacQueen et al and by Gorczynski et al [57-59]. It was also observed that various cells in the immune system produce classical hormones and neurotransmitters. Smith and Blalock, Montgomery et al and DiMathia et al. [60-62] pioneered these observations.
2.
NEUROIMMUNE INTERACTIONS
2.1.
Cell-to-cell interaction
Traditionally the cells of all tissues and organs have been divided into stromal cells, which were thought to provide for the structure of organs and the frame for the functioning cells, which were called parenchymal cells. It is now evident that stromal cells interact actively with parenchymal elements and this interaction leads to functional regulation of the tissue/organ. Moreover, invariably the stroma of all tissues and organs contain immune derived elements such as lymphocytes, macrophages or more specialized cells that include the glia cells in the nervous system, Kupffer cells in the liver, the Langerhans cells in the skin, etc. These cells contribute to function both in health and disease. Blood vessels and endothelial cells lining the blood vessels are also active participants in lymphocyte recirculation and in local immune/inflammatory reactions. These cells interact both with the circulatory elements of the immune system and locally with elements of the tissue/organ. Cell-to-cell regulation in tissues is mediated by adhesion molecules that have complementary binding sites. These molecules are capable of delivering activation or inhibitory signals in a tissue and cell-specific manner [63-76]. Adhesion molecules and other cell membrane receptors have the capacity to co-aggregate within the semi-fluid cell membrane (capping) and allow the interaction of immunoreceptor thyrosin based activation motifs (ITAM) and -inhibitory motifs (ITIM). These motifs promote phosphorylation and dephosphoylation of signal transuding molecules, respectively. The cell may be activated or inhibited depending on the outcome of receptor interactions after capping. The relevance of these regulatory motifs to cell function is especially well established for the antigen receptors of NK cells and of T lymphocytes and for the function of Fc receptors. However, the phenomena of "receptor crosstalk" has been observed in many other systems [77-85]. These developments indicate that numerous receptors are involved in cell signaling, and that these receptors interact by multiple mechanisms that may lead to activation, inhibition or even inactivation (apoptosis) [ 130]. Numerous receptors in immunology and several hormone receptors need to be cross-linked by the ligand in order to deliver an activation signal to a cell. This mechanism provides an important regulatory function in that cross linking may take place only at an optimal concentration of the ligand, whereas low or high concentrations would not be able to signal the cells. When more than one receptor isotype is available, the homo- and hetero-diamers formed by the specific ligand may have different regulatory functions. In addition, cross-linking may be one of the important mechanisms that promotes capping of the receptors prior to activation [77, 87]. The immune system consists of mobile cells that are able to home readily to specific target tissues and also to sites of infection, injury, regeneration and healing. Stromal lymphoid cells play physiological roles and are very important for host defense, regeneration and repair. Adhesion molecules mediate immunocyte homing and lymphocyte recirculation. Blood vessels also provide important barrier function in some tissues and organs that are known as immunologically privileged sites. The blood-brain barrier is very important from the point of view of neuroimmune interaction and is being extensively studied at the present time [71, 86, 88-93].
2.2.
Innervation
The central nervous system has the capacity to deliver neurotransmitters and neuropeptides to all tissues and cells in the body. For a long time the immune system was considered as an exception to this rule. However, it is now clear that the thymus and the spleen and other lymphoid organs are innervated. Interestingly, the spleen contains only sympathetic efferent nerve fibers [94, 95]. Tissue mast cells are also innervated and the formation of synapses with nerve fibers and lymphocytes can be readily demonstrated in tissue culture. Neurogenic inflammation is the direct result of the discharge of inflammatory mediators from mast cells after stimulation by mediators (primarily substance P) released from sensory nerve terminals. Neural mediators, such as growth factors, neurotransmitters, and neuropeptides, (e.g. substance-P, somatostatin) play major roles in the regulation of immune/inflammatory responses. Nerve fibers are capable of rapid and specific local delivery of mediators that are suitable of mounting an instantaneous reaction by initiating inflammation. In other situations nerves may exert an anti-inflammatory effect. The local modulation of immune reactions is equally possible by neurotransmitters and neuropeptides [94-96]. During the acute phase response there is a massive release of catecholamines into the circulation, which is known as "sympathetic outflow". Catecholamines are important regulators in the acute phase response, which is an emergency defense reaction. Sensory nerves provide feedback signals towards the CNS from sites of injury, inflammation, and infection. The vagus nerve carries feedback signals to the CNS from visceral organs [94-97]. 2.3.
Humoral communication
Historically the humoral mediators of cell-communication have been classified as hormones that act at distant targets, neurotransmitters and neuropeptides, and locally produced hormone-like mediators, now called cytokines. One may also include here immunoglobulins, which originate from B-lymphocytes within the immune system. Immunoglobulins have evolved from adhesion molecules. In addition, virtually every cell membrane bound molecule is present in the serum, which includes MHC molecules and receptor-like-binding proteins. By now it is clear that "classical" hormones, neurotransmitters and neuropeptides are widely synthesized at various ectopic sites, including the immune system. Moreover, cytokines, which have been originally discovered within the immune system are now known to be synthesized in other tissues and organs, including the neuroendocrine system. Therefore, the historical definition of hormones, neurotransmitters and neuropeptides no longer applies. Rather, systemic and locally produced mediators complement each other, so that optimal function is assured both under physiological and pathophysiological conditions. In addition to the blood stream, lymphatic drainage of tissues, including the CNS, is important for humoral communication. The immune system receives signals from all tissues via the lymphatic system [86, 93, 98, 99].
3.
NEUROIMMUNE REGULATORY PATHWAYS
3.1.
The TRH-PRL, GH, IGF-I, TSH-thyroid axis
Thyrotropin releasing hormone (TRH) stimulates prolactin (PRL), thyroid stimulating hormone (TSH) and under some pathophysiological conditions, growth hormone (GH) release [ 100, 101 ]. Moreover, GH, PRL and TSH producing pituitary cells share the nuclear regulatory factor,
10
Pit-1 [102]. This suggests that these hormones represent an interdependent regulatory unit. Indeed in rats immunized with sheep red blood cells the increase of TRH mRNA was found in the hypothalamus at 4-24 hours after immunization. Pituitary TRH receptor mRNA and plasma PRL levels were also increased at the same time, while TSH and GH did not change. The hypothalamus-pituitary-adrenal (HPA) axis was activated 5-7 days after immunization. Antisense oligonucleotides complementary to TRH mRNA, given i.c.v, inhibited PRL secretion and decresed the titer of antibodies produced [ 103].
3.1.1. Thyrotropin releasing hormone (TRH) TRH affects directly lymphocyte proliferation and the development of T lymphocytes in the gastrointestinal tract [104, 105]. In man, serum interleukin-2 (IL-2) levels rose significantly during the standard TRH test [106]. The treatment of patients in critical illness repeatedly with TRH increased serum TSH, PRL, GH, T4 and T3 levels, and may correct the euthyroid sick syndrome [ 101 ]. 3.1.2. Growth and Lactogenic Hormones (GLH) Growth hormone, PRL and placental lactogen (PL) are referred to collectively as GLH. All three hormones show molecular heterogeneity and the variant forms of GH and PRL differ in their biological activity. GLH hormones are produced by a variety of cells in the body, including lymphocytes [107-117]. Our recent observations indicate that PRL production in lymphoid tissues is pituitary dependent (Figure 2). GLH and cytokines (e.g. G-CSF, GM-CSF, EPO, IL-2, -3, -4, -5, -6, -7,-9, -11, -13) share receptor structure [ 118-121 ]. Receptors for PRL and GH show heterogeneity and require cross-linking for signal delivery. At high hormone concentrations, cross-linking will not take place, but rather each receptor molecule will be bound to a separate hormone molecule, which leads to the self-inhibition of signal delivery. Homo- and heterodiamerization may take place after receptor-ligand interaction and some of the heterodiamers lead to inhibition, rather than stimulation. More than one signaling pathways play a role in GH and PRL action [87, 118-124]. Both GH and PRL induce the production of insulin-like growth factor-I (IGF-I) in cells of the immune system. IGF-I receptors belong to the transmembrane thyrosine kinase receptor family and are ubiquitously displayed on immunocytes [ 125]. The fetal pituitary gland does not play a role in the development of the immune system. There is evidence to suggest that maternal and placental lactogenic hormones fulfil this role [126-128]. After parturition, the function of the bone marrow, the thymus and the maintenance of immunocompetence all become pituitary dependent. The bone marrow deficiency of hypophysectomized rats can be normalized by treatment with purified PRL, GH or PL [ 129-131 ]. IGF-I plays a role in the mediation of GH action on bone marrow [132, 33]. Colony stimulating factor-1 (GM-CSF) and interleukin-3 are capable of stimulating IGF-I production in bone marrow cells and thus might function similarly to GLH in this organ [134]. GH, PRL, PL and IGF-I all stimulate thymus growth [116, 127, 135-139]. This stimulatory effect is directly related to the maintenance of immunocompetence [ 136]. GH, PRL and PL all promote the antibody response [128, 141]. Human pituitary dwarf individuals have normal immune function, which can be explained by the presence of normal serum PRL levels [140]. The dopaminergic drug, bromocriptine, suppressed humoral immunity which could be reversed by treatment with either GH or PRL. ACTH induced immunosuppression was also reversed by these hormones [142]. PRL enhanced the antibody response in mice in a biphasic manner [ 141 ]. Cell mediated immune reactions, including contact sensitivity reactions, graft rejection, graft versus host reaction, and killer cell activity were
11
PITUITARY DEPENDENCE OF PROLACTIN PRODUCTION IN LYMPHOID TISSUE RIA
E;I r r r
.m
9 9
9
9
B~O BIO + o~ PRL
~
,,....-
BIO + HYPOX
1000
HY
Hypophysis
E
TM
Thymus
o
SP
Spleen
SE
Serum (ng/ml)
c::) r v
I"""I
Z
k-o <
500
._.1
9
O
9
13_
9 , 9
r
s
-
~
"
,.,.-......
HY
TM
...
,
s
,
SP
HY
SE
TM
SP
Figure 2. Pituitary dependence of prolactin production in lymphoid tissue. Female or male Fischer rats (150-170 g) were used as normal controls, or were hypohysectomized (HYPOX). Some rats were treated with a rabbit antiserum against rat prolactin (c~PRL, 50 pl/day s.c.), which was initiated on day 14 after hypophysectomy, and maintained until day 21, when extracts of organs were prepared. For the release of tissue PRL 30 mg/ml of wet tissue was placed in serum-free RPMI 1640 culture medium and was frozen (-20~
and thawed (37~
waterbath) three times.
The tubes were then centrifuged, and the supernatants were tested for PRL immuno- and bioactivity. Radioimmunoassy (RIA) [360] and the Nb-2 lymphoma bioassay (BIO) [360] were used. This figure indicates that significant quantities of immuno- and bio-active PRL was present in the thymus and spleen of normal rats, which far exceeded serum levels. The thymus and spleen of HYPOX rats contained only trace amounts of PRL.
stimulated by GLH [113, 129]. The tumoricidal activity of macrophages was also increased by PRL as was the cytotoxic activity of natural killer (NK) cells. High concentrations of PRL inhibited NK and lymphokine activated killer (LAK) cell activation [143]. Recombinant GH corrected the decreased NK activity in GH deficient children [ 144]. GH and PRL stimulated the activities of monocyte/macropahges, and polymorphonuclear leukocytes [ 145-149]. GH enhanced the production of IL-2 and IL-6 and had variable effects on IL-1 and tumor necrosis factor-a (TNFc~) production. PRL promoted IFN- T and inhibited IL-1 production [ 113, 151, 152]. The age-related decline of immunocompetence may be due, at least in part, to the decline of GH/IGF-I production [135, 151,152]. It is clear from this brief overview that GLH show redundancy as immunostimulatory hormones. Current evidence suggests that GLH will support any function performed by the
12
immune system, including suppressor and killer cell activities, which is compatible with the notion of competence hormones [98]. It has been suggested on the basis of experiments performed in knockout mice, that PRL, GH and IGF-I are not obligate immunoregulators, but rather, affect immune reactions as anabolic and stress modulating hormones [153, 154]. In actual fact the data obtained in knockout mice confirm our original observations that GLH show overlap in the maintenance of the immune system. By no means do these knockout experiments indicate the irrelevance of GH and/or PRL to immune function. In order to prove or disprove the relevance of GLH to immunity, the entire system should be disabled. However, we predict on the basis of our observations that such mutations would have lethal consequences [155]. Much remains to be clarified with regards to the role of the various isoforms of PRL and GH, and of their receptors, in immune function. Because the receptor structure and the Jak-Stat transcription pathway of PRL and GH are shared with interleukins and hemopoietic growth factors [86, 153], some regard PRL and GH as members of the hemopoietic cytokine family. However, functional overlap with cytokines could simply indicate the capacity of systemic GLH to maintain the hemopoietic and immune systems at times when cytokines are in short supply. 3.1.3. TSH and thyroid hormones TSH modulates immune function by the stimulation of thyroid hormones and also by acting on lymphoid cells. TSH receptors are expressed on dendritic cells and on CD45Rb highlymph node T cells. Recombinant TSH significantly enhanced the phagocytic activity of dendritic cells from adult mice and selectively augmented the IL-I[5 and IL-12 cytokine responses following phagocytic activation. TSH also stimulated immunoglobulin secretion and IL-2 production. Human lymphocytes treated with TRH released TSH [98, 158, 159]. Thyroid hormone receptors (TR) are nuclear transcription factors and belong to the steroidthyroid hormone receptor family. TR is encoded by two genes, TRot and TR[3. Multiple isoforms of TR proteins are generated by alternative splicing [160]. Lymphocytes convert thyroxin (T4) to bioactive triiodothironine (T3). The effect of thyroid hormones on immune responses is variable. Enhancement, suppression, or no effect was reported repeatedly. While hypothyroidism is usually, but not always, associated with immunodeficiency, treatment of normal animals with T3 yielded mostly negative results. In TR knockout mice (TRot-/-) thymopoiesis was suppressed. B cell maturation is depressed in mice that cannot respond to thyroid hormones [98, 156, 158]. 3.2.
The CRF-ACTH, r
[5-END, -glucocorticoid axis
The hypothalamus-pituitary-adrenal axis (HPA) and the proopiomelanocortin (POMC) derived peptides (ACTH, c~-MSH, [5-END) act antagonistically to GLH and suppress adaptive immune/ inflammatory responses by acting on the nervous, endocrine and immune systems [98, 162]. 3.2.1. Corticotropin releasing factor (CRF) During acute phase immune responses, cytokines stimulate CRF, which in turn induces ACTH release. CRF integrates the stress response in the CNS and exerts a central immunosuppressive effect by the stimulation of sympathetic outflow. CRF is also produced within the immune system and has a direct regulatory effect on lymphocytes, which is mostly, but not always, immunosuppressive [50, 163, 162].
13 3.2.2. Adrenocorticotropic hormone (ACTH) ACTH is immunosuppressive via the stimulation of glucocorticoid secretion by the adrenal gland. ACTH is produced by lymphocytes and it has a direct regulatory effect on lymphocyte proliferation, immunoglobulin production, and phagocytosis. ACTH exerts an anti-pyretic effect in the CNS [98, 162, 163]. 3.2.3. Beta-endorphin ([3-END) [3-END is produced and secreted by the pituitary gland and also within the brain and immune system. [3-END and opioids in general are immunosuppressive when acting on the /~ and K opioid receptors. Opioids are also capable of immunoregulation by acting on the CNS [165-168]. 4.4.4. Alpha-melanocyte stimulating hormone (c~-MSH) ot-MSH is a very effective antagonist of IL-1, -6, TNF, and IFN-~,. It inhibits fever and inflammation by acting on the CNS and also exerts an antiinflammatory effect on peripheral targets, c~-MSH promoted tolerance induction to contact sensitizing agents which was mediated by IL-10 [164, 169-171]. 3.2.5. Glucocorticoids (GC) Glucocorticoid receptors are nuclear transcription factors and are present in all cells in the body. Glucocorticoids suppress the adaptive immune response, although evidence is increasing that basal physiological levels are actually required for the normal function of lymphocytes. Elevated pathophysiological levels (e.g. during systemic immune/inflammatory reactions, trauma, or other stressful conditions) alter lymphocyte distribution in the body and suppress humoral and cell mediated immunity. Mononuclear and polymorphonuclear phagocyte function and cytokine production are suppressed by elevated glucocorticoid levels. On the other hand, glucocorticoids increased the expression of HLA antigens, and receptors for IFN-~,, IL-1, IL-6, Fc~,. Memory cells and the cells maintaining graft-versus-host reactions are resistant to glucocorticoids. The thymic epithelium is capable of synthesizing GC [98, 162, 172-177]. 3.3.
Gonadotropins and sex hormones
In 5 normoprolactinemic women intravenous bolus injectin of luteinizing hormone releasing hormone (LHRH) and TRH increased plasma IFN-~, levels, with the maximum response at 45 rain after injection. Peak levels of PRL appeared at 15 min; TSH: 30 min; FSH: 30 min; LH: 30 min. Moreover, LHRH and TRH, separately and together, significantly enhanced in vitro IFN-~, production by staphylococcal enterortoxin-A (SEA) and concanavalin (ConA)activated peripheral blood mononuclear cells (PBMC) [198]. Luteinizing hormone (LH) has a direct stimulatory effect on the immune system. Follicle stimulating hormone (FSH) affected lymphocyte proliferation and IL-6 production [ 179-181 ]. Sex hormones play a major role in the regulation of mucosal immune responses [ 182]. 3.3.1. Estradiol (E2) E2 has a suppressive effect on bone marrow function, on the thymus, on T cell function, NK cytotoxicity, neutrophil and mast cell degranulation. Phagocytosis, antibody formation and certain forms of autoimmune disease are stimulated by E2. The cytotoxic activity of CD4 + cells is dependent on estrogen [98, 183-186].
14 3.3.2. Androgens Testosterone exerts a suppressive or moderating effect on the immune system, it antagonizes the enhancing effect of estrogens on various autoimmune diseases and stimulates bone marrow function. Aromatase inhibitors block the effect of testosterone on the thymus [98, 187, 188]. Dehydrotestosterone (DHT) has a stimulatory effect on T lymphocytes and immunoglobulin formation. DHT is generated within the immune system from androstenediol or testosterone by macrophages [ 189, 191 ]. Dehydroepiandrosterone (DHEA) is a weak androgen produced in the adrenal glands. DHT and its metabolites have emerged as major regulators of immune reactions capable of both immunostimulation and immunosuppression [191]. DHEA stimulates type 1 helper T cells (Th-1) for proliferation and IL-2 secretion and promotes cell mediated immunity. It antagonizes immunosuppression by glucocorticoids. Age related immunodeficiency was reversed by DHEA in mice and host resistance was increased against viral, bacterial and parasitic infections. Vaccination and mucosal immunity was potentiated by DHEA. In mice DHEA administration restored the depressed splenocyte proliferation as well as IL-2, IL-3, and IFN-~, production following trauma and hemorrhage. In vitro the stimulatory effect of DHEA on splenocyte proliferation was unaltered by the testosterone receptor antagonist flutamide, while the estrogen antagonist tamoxifen completely abrogated its effect [192]. Serum DHEA levels are decreased with aging, during chronic illness, suppressed by dexamethasone treatment, and restored by ACTH treatment [ 193-196]. 3.3.1. Progesterone (PS) Progesterone is a major immunosuppressive hormone and plays a key role in the harmonization of immune function with reproduction. During pregnancy, activated lymphocytes synthesize a progesterone induced blocking factor (PIBF), which inhibits NK activity and exerts an antiabortive effect. Decidual CD56+ NK cells express PIBF. PS decreases host resistance to viral and fungal infections and inhibits the function of phagocytes [98, 197-203]. 3.4.
1-25-Hydroxy vitamin D3 (VD3)
The liver produces 25-hydroxy vitamin D3, which is further processed in the kidney by 1-hydroxylase. This enzyme is also present in monocyte/macrophages, keratinocytes, bone marrow cells, placenta and in pneumocytes. The receptor for VD3 is of 50 kDa protein and belongs to the superfamily of steroid/thyroid hormone receptors [204, 205]. VD3 promotes the differentiation of macrophages, lymphocytes and of other cell types. Monocyte/macrophage phagocytosis and cytotoxicity is promoted by VD3, whereas antigen presentation and cytokine production by T lymphocytes and cell mediated immune reactions are inhibited. NK cell mediated cytotoxicity is stimulated, B lymphocytes proliferation and immunoglobulin secretion are inhibited by VD3 treatment. Experimental autoimmune reactions are prevented by VD3 treatment. In man the treatment of psoriasis with VD3 analogues has a 100% success rate [204-211 ]. 3.5.
Melatonin
Melatonin (MEL) is secreted by the pineal gland. It regulates seasonal breading in animals and is involved in the regulation of circadian rhythms in vertebrates. Helper T cells express G-protein coupled MEL membrane receptors and, perhaps, MEL nuclear receptors as well.
15
MEL stimulates the release of Th-1 cytokines, such as IFN-y, and IL-2, and of novel opioid cytokines which crossreact both with IL-4 and dynorphin B. MEL was found to enhance the production of IL-1, -6 and -12 in human monocytes. In general MEL exerts an immunostimulatory effect. Hematopoiesis is also influenced, possibly by MEL-induced-opioids acting on K-opioid receptors that are present on stromal bone marrow macrophages. IFN-7 and colony stimulating factors appear to influence the production of MEL in the pineal gland. One intriguing feature of immunomodulation by MEL is that it is effective only if given at the fight time within the circadian rhythm of the animal [212-219]. Much remains to be clarified about MEL as an immunoregulatory factor. 3.6.
Nerve growth factor and neurotrophins
Nerve growth factor (NGF) was first detected in murine submandibular glands as a growth factor for sensory and sympathetic ganglia [220]. NGF belongs to the family of neurotrophins, that include brain derived neuroptrophin (BDNT) and neurotrophin-3 (NT-3). There are low affinity neurotrophic receptors (P25) and high affinity receptors, which are thyrosine kinases (e.g. gpl40trkA for NGF; gp145trkB primarily for BNDF; gp145trkC for NT-3). Human macrophages express trkA and NGF is an autocine growth factor for these cells. The thymus, lymph nodes, express trkA and the spleen trkB, localized primarily to the stroma of these organs. There is some expression also in splenocytes and thymocytes. B lymphocytes and antigen presenting cells (follicular dendritic cells) also express receptors for NGF. NGF stimulates the growth and function of mast cells, B and T lymphocytes, stimulates IgM and IgG production, which is inhibited by IL-4. NGF inhibits the induction of IgE by IL-4 [221-235]. T and B lymphocytes, macrophages and mast cells synthesize biologically active NGF. NGF promoted the development of hemopoietic colonies and stimulated the chemotactic and phagocytic activity of polymorphonuclear leukocytes, which suggest a proinflammatory role for NGF. However, in vivo the suppression of inflammation has also been observed by NGF in several experimental models. Recent observations indicate that immune derived NGF provides protection for the nerveous system and to other host tissues during inflammatory reactions. This phenomenon implies the existence of beneficial 'autoimmune' reactions [236-246]. 3.7.
Leptin
Leptin (LEP) is produced primarily by fat cells (adipocytes). Structurally LEP belongs to the GLH/CTK family and signals by a class I cytokine receptor (Ob-R). Two receptor isoforms are known: Ob-Ra and Ob-Rb. Leptin regulates energy metabolism, reproductive function, lymphoid development and function. Under normal physiological conditions the secretion of LEP is regulated by insulin, cortisol and sex steroids, mainly testosterone. In rats centrally administered LEP suppressed the mitogenic response of splenic lymphocytes. This was mediated through CRF-sympathetic activation. Leptin plays an important role in linking nutritional state and T cell function. In starving mice, which show immunosuppression, treatment with LEP enhanced THl-mediated immune responses, in spite of the catabolic state of the animals. Starving animals have reduced LEP levels and show an increased sensitivity to endotoxin shock. Fasting mice respond to LPS with a blunted corticosterone and exaggerated TNF production. This could be corrected by LEP treatment [247-254]. During acute phase responses (e.g. sepsis) the serum level of LEP rises rapidly. Cytokines, especially TNF-c~, causes this elevation. LEP exerts an inhibitory effect on glucocorticoid
16
and IL-6 production. Blood levels of LEP correlate positively with the survival of patients with septicemia. LEP stimulates the production of IL-1 receptor antagonist (IL-lra), which protects against LPS toxicity in mice. In murine glial cells LEP stimulated the production of IL-1 [5. In animal experiments exogenous LEP upregulated both phagocytosis and the production of proinflammatory cytokines. Leptin is also involved in wound healing and angiogenesis [250, 255-261]. 3.8.
Neurotransmitters and neuropeptides
3.8.1. Catecholamines and acetylcholine Various cells in the immune system express [3-type adrenergic receptors. Beta-adrenergic agents inhibit allergic and asthmatic reactions and in general inhibit various immune phenomena that include lymphoid responses to mitogens and to antigen, histamine release from leukocyctes and mast cells and skin reactions to antigen and histamine. The effect on antibody formation is variable. In vivo adrenalin elicits leukocytosis and eosinophilia which is followed by eosinopenia. At least some of these effects on leukocyte distribution are due to glucocorticoid release. Noradrenaline inhibits the histamine release from leukocytes and the degranulation of mast cells and it has a variable effect on antibody formation. In mice treated with LPS the reduction of sympathetic outflow by reserpine dramatically increased TNF production. Neuronal ct2- and macrophage [5- and ct2-receptors were involved. In healthy volunteers catecholamines down regulate LPS-induced TNFmt, IL-6 and IL-I[5, and increased IL 10. In patients with prolonged sepsis TNF-ct and IL-6 were reduced and IL-1 [5 and IL 10 were not modulated by catecholamines [262-271]. The role of the peripheral and central catecholamine systems on immune regulation is the subject of intense investigations at the present time. Acetylcholine affects immune phenomena by nicotinic and muscerinic receptors. Cholinergic agents enhance immune phenomena, including lymphocyte mitogenesis, cytotoxic reactions, the release of histamine and other mediators from mast cells. These effects are meditated by muscarine receptors. Acetylcholine stimulates the synthesis of complement components by human monocytes through the nicotinic receptor. Allergic patients show an increased sensitivity to cholinergic stimulation. The involvement of cholinergic mechanisms in exercise-induced anaphylaxis has been demonstrated [262, 272-274]. 3.8.2. Substance-P (SP) Substance-P mediates pain sensation in type C sensory nerve fibers and is a major mediator of neurogenic inflammation. Thymocytes, B and T lymphocytes, macrophages, mast cells and astrocytes have SP receptors. SP is capable of inducing degranulation of mucosal and intestinal type of mast cells, can cause plasma extravasation and bronchoconstriction. Substance P has a direct effect on lymphocytes, macrophages, eosinophils and neutrophils. It promotes lymphocyte proliferation, lymphokine production, and it has variable influence on immunoglobulin secretion. On eosinophils, SP increases Fcq, and -e receptors and decreases C3b receptors. SP stimulates the respiratory burst, chemotactic and phagocytic responses in polymorphonuclear leukocytes. Substance P stimulates the release of PGE 2 and collagenase from rheumatoid synoviocytes and of PGE and thromboxane B2 from astrocytes. Platelet cytotoxicity against Schistosoma mansoni larvae is activated by SP. SP induced IL-3 and GM-CSF production by bone marrow cells. This was partially mediated by IL-1 and IL-6, which are also induced by SP in the bone marrow. SP receptor expression is up-regulated by IL-4 and IFNq, in murine peritoneal macrophages. The SP receptor was necessary for the normal granulomatous response to Schistosoma mansoni [275-286].
17 3.8.3. Calcitonin gene related peptide (CGRP) CGRP receptors are functionally coupled to adenylate cyclase and are present on mature lymphocytes, macrophages mast cells and bone marrow cells. CGRP induces mast cell discharge, produces slow onset intense erythma in the skin and vasodilation. In human mononuclear phagocytes CGRP interferes with antigen presentation and with IFN-ct induced H202 production. Lymphocyte proliferation ia slso inhibited by CGRP. Nerve fibers containing CGRP are associated with Langerhans cells in the human skin. CGRP plays an important role in the regulation of the cutaneous immune system. It inhibited antigen presentation by human Langerhans cells, and the induction of contact hypersensitivity reactions to haptens in mice. Topically applied CGRP increased the inflammatory response in the skin to allergens and irritants and boosted the sensitization process. In murine thymocytes CGRP inhibited the expression of NF~B and promoted apoptosis. T lymphocytes from rat thymus and mesenteric lymph nodes sythesized CGRP [281-293]. 3.8.4. Somatostatin (SOM) Receptors are present on T and B lymphocytes and mast cells for SOM. SOM acts as an antagonist of substance P and it has beneficial effects in models of autoimmune disease and of chronic inflammation. SOM inhibits IgE dependent mediator release by human basophils and mast cells. It also inhibits lymphocyte proliferation, endotoxin-induced leukocytosis, IgA secretion, IFN-ct production, and affects macrophages. It has a variable effect of antibody dependent cytotoxicity [294-299]. 3.8.5. Vasoactive intestinal peptide (VIP) and pituitary adenylate cyclase activating peptide
(PACAP) Receptors are present in monocytes and lymphocytes and both peptides are produced within the immune system. VIP regulates T cell homing to mucosal lymphoid tissue, it inhibits lymphocyte proliferation and has a variable effect on immunoglobulin secretion and on NK cell mediated cytotoxicity. IgA secretion was induced by VIP by isotype switching [300-304]. VIP and PACAP inhibit the nuclear translocation of NFKB in stimulated macrophages by inhibiting Jak/Stat phosphorylation and thus antagonize the effect of IFN-y and downregulate the inflammatory response. The production of cytokines, such as TGF-[51, IL-12, IL-4, -6, TNF-c~ and nitric oxide (NO) are inhibited by both peptides [305-310]. IL-6 production was enhanced by VIP/PACAP in unstimulated macrophages [311]. VIP and PACAP inhibited antigen induced apoptosis in CD4+ (but not CD8+) T lymphocytes by downregulating Fas [312]. VIP and c~-MSH contribute to the immunosuppressive properties of aqueous humour in the eye [313, 314]. In man VIP inhibited the development of contact dermatitis to nickel sulphate when injected intracutaneously at the site of challenege [315]. 3.9.
Cytokines
Cytokines have been originally discovered within the immune system as humoral mediators between leukocytes (interleukins). By now it is clear that cytokines are produced in all tissues and organs in the body. Under physiological circumstances cytokines are local regulators of tissue/organ function. However, during acute phase reactions cytokines such as IL-1, TNF-ct and IL6 serve as systemic hormones and induce profound neuroendocrine and metabolic alterations, which serves to boost the natural resistance of the body towards diverse noxious agents. Some other cytokines with major roles in the neuroimmunoregulatory system are interleukin-2, -4, -10, & interferonq, (IFNq,). Redundancy is present within the cytokine system in that
18
these mediators have overlapping functions. This is now well substantiated with experiments preformed in various knockout mice [98, 316-319]. 3.10.
Chemokines (CEM)
These are chemotactic pro-inflammatory mediators which are produced in response to injury, irritants, polyclonal activators, antigens and cytokines. As inflammatory mediators CEM play important roles in host defence as well as in the pathogenesis of inflammatory diseases. Chemokines also serve as mediators of cell-to-cell communication within the immune system and promote humoral and cell-mediated immune reactions, regulate cell adhesion, angiogenesis, leukocyte trafficking and homing and contribute to lymphopoiesis and hematopoiesis. A vast number of CEM have been identified to date which may be categorized, based on their structures, into four major groups: CXC (et), CC (13), C (~,) and CX3C subfamilies. Chemokine receptors are seven-transmembrane G proteins. Chemokines show unprecedented redundancy in receptor utilization and leukocytes express multiple receptors [316, 319]. Chemokines play a fundamental role in cell-to-cell communication throughout the body and enable every cell/tissue to emit signals towards the neuroimmune regulatory network. 3.11.
The neuroexocrine-mucosal system
The mucosal immune system consists of mesenteric lymph nodes, Peyer's patches, the tonsils, mucosa associated lymphoid cells and lymphoid cells associated with various glands (e.g. salivary and lacrimal). It is now apparent that the mucosal immune system does not only defend the body against invading pathogens but also exerts major regulatory effects on systemic immune reactions. The problem faced by the mucosal immune system is that mucous membranes are bombarded by large amounts of antigens continuously, most of which are irrelevant to host defense. It would be counter-productive to spend a lot of immunological energy to respond to harmless antigens. On the other hand pathogenic agents and potentially harmful toxic substances must be dealt with. Immune defense is already mounted on the surface of mucous membranes, which are outside of the body. Therefore, self-non-self discrimination, which is being utilized so efficiently by the immune system inside the body, does not apply to this situation [320-324]. The initial response to antigens falling on mucosal surfaces is frequently the induction of immunological tolerance. This may take place by clonal elimination at high antigen dosage or by active suppression if the dosage is low. Although the mechanisms that regulate tolerance induction has not been fully elucidated, it appears that antigen presentation by specialized cells that induce a distinct class of T cells capable of suppressing immune responses locally and systemically takes place. These T cells have been named as a type 3 T (TH3) cells and produce large amounts of transforming growth factor-[3 (TGF-[3). TH3 cells exert a powerful systemic immunosuppressive effect all over the body at sites of inflammation and on cells of lymphoid tissue [325, 326]. Mucosal mast cells are distinct from those situated in other tissues and play important roles in the physiology and pathophysiology of mucous membranes. The submandibular gland in laboratory rodents has been identified as a neuroendocrine and neuroexocrine organ secreting antimicrobial substances, immunoglobulin, hormones and enzymes that play major roles in mucosal immune reactions as well in the regulation of inflammation, regeneration and repair within mucosal tissues and elsewhere in the body. The sympathetic superior cervical ganglionsubmandibular gland axis has been suggested as one of the major immunoregulatory pathways.
19
The submandibular gland produces, secrets and excretes significant amounts of nerve growth factor, epidermal growth factor and TGF-[3, all of which are powerful immunoregulators. Glandular kallikrein, an enzyme with a potent immunosuppressive effect, is also produced. This enzyme was shown to suppress immune reactions and inflammation when applied parenterally to animals and to play a role in the induction of oral immunological tolerance when given by mouth [322, 323, 326]. It has been demonstrated in the gut that must cells are innervated and that these cells play an important role in intestinal absorption as well as in pathological responses, such as the initiation of inflammation and so on. Lymphocytes exposed to antigen/infectious agents at a particular mucosal site will multiply, differentiate and redistribute to other sites of mucosal membranes through re-circulation, which provides generalized protection. This is known as the "common mucosal immune system" [286, 328].
4.
BASIC CONCEPTS AND PRINCIPLES IN NEUROIMMUNOREGULATION
The rapid accumulation of experimental data in diverse systems that has relevance to neuroimmunoregulation has led to contradictions, misconceptions and confusion. These problems need to be addressed and clarified. Some of the key issues are addressed below both from the theoretical and practical points of view. 4.1.
The concept of competence
It has been known for over a century that GH is capable of promoting the proportional growth of all tissues and organs in the body. Naturally, this includes the immune organs, such as the spleen, thymus and lymph nodes. Receptors for GLH are expressed on every tissue and organ in the body. Current indications are that all cells require some members of this hormone family for normal growth and function. GLH in conjunction with IGF-I exert a growth stimulatory and an anti-apoptotic effect on various cells and thus assure maintenance in a functional state, i.e. the cells are capable of responding to additional stimuli. On this basis GLH may be defined as competence hormones. There is much evidence in the literature indicating that the immune system is dependent on GLH for the maintenance of immunocompetence. This is related to the general function of these hormones to maintain growth and development, and lymphocyte growth is a prerequisite for adaptive immune reactions [98, 129, 136]. On the other hand, a vast amount of experimental evidence has accumulated, indicating the role of tissue specific growth factors, cytokines and adhesion molecules in cell-to-cell signaling and the regulation of growth, differentiation and function of various cells in the body. It is apparent that proportional growth, although ultimately controlled by the systemic level of GH, is achieved through the coordinated interaction of systemic and local growth regulatory signals, many of which are tissue/organ specific and are also function-related. Current evidence indicates that GH in conjunction with IGF-I maintain all cells in a state of competence to respond to additional, function-related stimuli and produce additional regulatory mediators. Prolactin, which may be regarded as a modified growth hormone, is capable of providing competence in most tissues and organs in the body with the exception of the skeleton, where its growth promoting effect is very limited. Like GH, PRL also induces IGF-I in its target cells. The IGF-I signal may be regarded as the ubiquitous cytokine signal that is needed for the survival of competent cells. On this basis one may hypothesise, that competence hormones maintain their target cells in aviable and responsive
20
state by a direct stimulatory effect on the genome and by the induction of IGF-I secretion. Ultimately, the proliferation and functional activation of cells is determined by adhesion molecules, which are capable of delivering non-diffusible cell-to cell or cell-to-matrix signals. Adherence signals are capable of regulating cell function specifically on an individual basis. For example, an antigen-presenting cell delivers regulatory signals to an antigen specific T lymphocyte [ 130]. Prolactin synthesis is detectable in numerous tissues, including lymphoid tissue [98, 154]. There is compelling evidence to suggest that tissue-derived PRL fulfils autocrine/paracrine regulatory functions. Within the immune system, small lymphocytes are in a quiescent state and do not synthesize significant amounts of mediators. They need to be activated in order to do so. Based on current experimental evidence, one may propose that inactive small lymphocytes are dependent for survival on pituitary GH/PRL and IGF-I. The dependence of cell survival in the thymus, spleen and bone marrow on GH/PRL supports this hypothesis. Moreover, pituitary PRL and GH maintain vital bodily functions and thus must act as survival hormones for the entire organism [129, 329]. Animals with joint and total deficiency of GH/PRL do not exist, and such deficiency has not been convincingly demonstrated in man to date [155, 329]. The proposed interaction of neuroendocrine factors with adhesion signals and paracrine circuits in the regulation of bodily functions is summarized in Figure 3.
4.2.
Redundancy
Failure of the neuroimmune regulatory system invariably leads to the death of the organism. In order to avoid frequent failures, the system must have multiple and overlapping regulatory pathways with a high degree of flexibility and plasticity. This is achieved through isologous forms of regulatory molecules, multiple forms of receptors, and by the existence of functionally overlapping or totally interchangeable regulatory pathways. The CNS shows a high degree of plasticity. Moreover, redundancy is present in the function of growth and lactogenic hormones, of the IGF/insulin system, of steroid hormones, neuropeptides, cytokines, chemokines and of the various immunoglobin classes. It is quite common in immunology that unsuspected redundancy is revealed in the system by knocking out a particular gene. Similarly, the disabling of prolactin or growth hormone, or even IGF-I, would not paralyze immune function [86, 153]. These facts and clinical observations indicate clearly that the functional integrity of the neuroimmune regulatory network is maintained, even after very severe insults/deficiencies due to the existence of redundant physiological and pathophysiological mechanisms. 4.3.
Homeostasis
Healthy individuals and animals maintain their body temperature, blood pressure heart rate metabolism and the concentration of various ingredients in the serum and in tissue fluids within standard physiological ranges, which is characteristic of the species. This was first recognized by Claude Bernard over a century ago, who coined the term "milieu interieur" [331], now designated as homeostasis. Under homeostatic conditions two basic forms of immune reactivity can be observed, e.g., innate or natural immunity and adaptive immunity. 4.3.1. Natural immunity The natural immune system consists of some highly specialized cells such as natural killer cells, ~,~ T cells and CD5+ B lymphocytes, that produce natural antibodies. The antigen
21
GH, PRL
IGF-I
T4, SH Figure 3. The interaction of neuroendocrine and autocrine/paracrine regulatory pathways. This figure depicts some basic neuroendocrine and paracrine interactions in immunoregulation with reference to other cells (Please see the text for details). It is proposed that the maintenance of all cells in the body in a functional state is dependent on competence hormones. Additional signals are required for tissue and cell-specific regulation that include adhesion molecules and cytokines. 1. Competence signal: This signal is delivered to lymphocytes and to many other cells in the body by GH and/or PRL, produced in the pituitary gland. GH and PRL are also produced in many tissues ectopically, including the immune system. Ectopic PRL/GH fulfil a local regulatory function during immune reactions. It is suggested that this autocrine/paracrine circuit makes rapid lymphocyte proliferation possible, which is a prerequisite of immune reactions. 2. Stromal/adherence signals: Antigen presentation is best known as an activation signal for lymphocytes. It is an adherence signal delivered by MHC molecules on antigen presenting cells. This is accompanied by additional co-stimulatory adherence signals, which may eventually lead to lymphocyte activation. Adherence signals also play a role in the induction of immunological tolerance, in lymphocyte survival, and in the induction of programmed cell death (PCD). It is proposed that adhesion signals fulfil tissue-, site-and cell-specific regulation in the body, i.e. the function/fate of individual cells are determined at this level. 3. Cytokine signals: Lymphocyte activation, and cell activation in general is completed by cytokine signals, which lead to cell proliferation, differentiation, and functional activation. Cytokines may also perform inhibitory function (e.g. TGF-[3, interferon-y) or cause PCD (e.g. tumor necrosis factors), a. Signal modulation: Some steroid hormones, catecholamines and endorphins/enkephalins are capable of modulating the process of signal delivery from the cell membrane to the nucleus by regulating Ca 2+ influx, cAMP and cGMP. b. Signal regulation: Thyroxin (T4) and steroid hormones (SH) control lymphocyte signaling by the regulation of nuclear transcription factors. Thyroxin, steroid hormones and vitamin D3 play a regulatory role also in cell differentation and in the elimination of unwanted cells via the induction of PCD. c. Local hormone activation: Bioactive thyroid and steroid hormones are locally generated from inactive precursors by immunocytes (e.g. T3, E2, androstenediol, androstenetriol, and vitamin D3) while the primary function of others (corticosteroids, estradiol, progesterone, aldosterone) is systemic immunoregulation. Quiescent lymphocytes do not synthesize DNA and exert minimum metabolic activity. Pituitary GH/PRL, some adhesion signals and serum IGF-I play a key role in the maintenance of these lymphocytes until functional activation occurs. Neurotransmitters and neuropeptides are locally acting functional regulators, basically acting as signal modulators and cytokines.
22
receptors of these cells are germ line coded, which are not subject to somatic mutation. These receptors have evolved to recognize highly conserved homologous epitopes (homotopes) on microorganisms and in self components and react instantaneously to their respective homotopes without the need of previous immunization. Natural antibodies are germ line coded and are polyspecific. Some liver derived proteins, namely C-reactive protein, endotoxin-binding protein, mannose binding protein, are also capable of recognizing homotopes, and to activate immune reactions [321,332, 333]. Non-immune factors contribute to innate resistance. In this context behavioral factors, physico-chemical factors, barriers, mucus, enzymes, anti-microbial substances, (HCL, bile acids, nitric oxide, oxygen radicals) heat shock proteins, non-immune interferons, enzymes, properdin, prostaglandins, leukotrienes, chemokines, blood clotting, species related resistance due to cell surface receptors and other factors may be mentioned. Typically it is not the antigen but cytokines and hormones that are fundamental to the regulation of natural immunity. For instance, in the regulation of NK cell mediated cytotoxicity interleukin-2, interferons, prolactin and growth hormone play important roles [333]. 4.3.2. Adaptive immunity The adaptive immune response is initiated by antigen presenting cells that activate antigen specific thymus derived (T) lymphocytes bearing ot[3-type antigens receptors. Bone marrow derived (B) lymphocytes recognize antigen by surface immunoglobulin molecules and produce antibodies. The antigen receptors of B and T lymphocytes undergo somatic mutation, followed by selection of those cells that do not possess self-reactivity but will react to altered self MHC antigens. Because of the elaborate selection process adaptive immune responses show exquisite specificity. Processed antigenic fragments (epitopes) are presented to T-lymphocytes by antigen presenting cells in association with surface MHC molecules. Antigens may come from the external environment or from within the body (e.g. virus infected cells, cancer cells, autoantigens etc). Externally-derived antigens are presented by MHC class II antigens to CD4+ T cells, whereas endogenous antigens produced by virus infected and cancer cells are presented by MHC class I antigens to CD8+ T lymphocytes [86]. Although the adaptive immune response is initiated by specific epitopes of the antigen, it is very well substantiated that additional "costimulatory signals" are also required for full activation of lymphocytes, for the initiation of cell proliferation, differentiation and for functional performance. In some situations the antigen signal is followed by inhibitory rather than stimulatory adherence signals, which lead to the induction of unresponsiveness, known as immunological tolerance, or anergy [86, 130]. Lymphocyte activation follows the rules of mitosis in general and involves cascades of enzymatic reactions, which is accompanied by Ca 2+ influx and ultimately leads to the phosphorylation of nuclear regulatory proteins. Hormones are capable of modulating the process of signal delivery from the membrane receptor to the nucleus by regulating Ca 2+ influx, or by the modulation of cyclic nucleotide levels, or some enzymes, etc. (e.g. catecholamines, some steroid hormones, [3-END and other opioid peptides) [98, 130]. This is designated as signal modulation (Figure 3). The origin of signal modulatory hormones may be exogenous or endogenous to the immune system. Thyroid and steroid hormones, and vitamins A and D, control nuclear transcription factors as their receptors, and are capable of regulating lymphocyte signals at the nuclear level. Because of the ability of these hormones to cross the cell membrane and home to their cytoplasmic/nuclear receptor molecules, they are capable of bypassing the cytoplasmic signal transduction pathway and amplify, suppress or even cancel completely, certain lymphocyte
23
signals in the nucleus. For instance, glucocorticoids are very efficient in the inhibition of ongoing lymphocyte reactions. Hormones belonging to this category are designated as signal regulators (Figure 3). Some of these nuclear regulatory hormones (e.g. T3, E2, androstenediol and androstenetriol, vitamin D3) are synthesized within the immune system and function in an autocrine/paracrine fashion. These locally produced hormones are required for normal immune function (e.g. T3, GC, DHEA and its metabolites). Others act as powerful immunoregulators (e.g. E2, elevated GC levels, PS) and have the capacity to amplify, suppress/terminate ongoing immune reactions. Moreover, glucocorticoids are able to kill thymocytes and lymphocytes by inducing PCD, whereas other steroid hormones sensitize target cells for killer cell induced PCD [ 184]. The final category of signals that complete the mitogenic stimulus to lymphocytes are delivered by cytokines. If the antigenic signal is not complemented with the proper cytokine response, it is followed by activation-induced programmed cell death. The major cytokines involved in the induction of T cell mediated immunity (delayed hypersensitivity reactions, cytotoxic T lymphocytes) are IL-2 and IFNq,. These are produced in large quantities by type 1-helper T lymphocytes (TH1). In contrast, humoral immunity is stimulated by IL-4, -5, -6 and -10, which are secreted by TH2 cells in large quantities. Although this classification of T cells is very convenient, it is recognized that intermediate cells are not uncommon, which provides plasticity and redundancy in the system. The primary antibody response always starts with IgM, which is followed by switching to other immunoglobin classes (e.g. IgG, IgA and IgE) while maintaining epitope specificity. This way a whole range of antibodies may be produced against the same epitope that have the capacity of activating diverse immune effector reactions, such as phagocytosis, complement fixation, cytotoxicity etc., against the same target. The primary immune response needs 5-10 days to provide efficient protection for the host, whereas secondary responses are much faster and can protect the host within less than 5 days. During the immunization period the organism must rely on natural immune mechanisms for protection [86, 333]. In health the immune system provides protection against infectious disease and diverse insults, while homeostasis is maintained. Typically subclinical infections and insults are contained, the pathogenic agents are eliminated and the injury repaired locally. This, however, does not mean that the neuroendocrine and immune systems do not interact under these conditions. There is ample evidence to indicate that the "homeostatic milieu" with well defined levels of hormones, cytokines, neurotransmitters and neuropeptides is fundamental to this immune homeostasis. Many physiological reactions can be considered as adaptive responses, which are required for coping with altered functional demands. Thus for instance, exercise commands higher blood pressure, heart rate, altered endocrine function, metabolism, and leads to immune alterations. The hypertrophy and atrophy of organs according to functional demands is also commonly observed. Adaptive enzyme synthesis is another example of adaptive responses other than the immune response. Clearly, responses analogous to immunization, tolerance induction and apoptosis are all observable in various tissues and organs of the organism. 4.4.
The acute phase response (APR)
The highly coordinated and multi-faceted defense reaction described by Hans Selye as the
general adaptation syndrome [9], is now known as the APR [335]. Fever is the unmistakable hallmark of APR, which is capable of increasing host resistant to diverse insults within hours. While liver derived proteins and natural antibodies increase rapidly during APR, the thymus
24
undergoes a profound involution. The adaptive immune response is suppressed. At the same time natural immune defense mechanisms are amplified several hundred to a thousand times within 24-48 hours. Immune derived cytokines, primarily ILl, IL6 and TNF-c~ initiate the reaction by activating leukocytes and acting on the central nervous system and on numerous other organs and tissues in the body. This triggers the HPA axis for increased activity. The secretion CRF, ACTH, et-MSH, [3-END and glucocorticoids is rapidly increased. Hormones of this axis suppress the adaptive immune response and regulate fever and inflammation by acting on the nervous-, endocrine- and immune systems [334-338]. Circulating GH and PRL levels quickly rise at the beginning of febrile illness and soon return to normal-to-subnormal levels. The IGF-I response to GH stimulation is impaired and the conversion of thyroxine (T4) to triiodothyronine (T3) in the tissues is also inhibited. Sex hormone levels are suppressed and testosterone levels may stay subnormal for lengthy periods. The levels of insulin and glucogen are consistently elevated, although insulin resistance is present [321,332, 333,334-338]. IL 6 levels are grossly elevated in APR. This is a pleiotropic cytokine which stimulates the production of acute phase proteins (APP) in the liver. Glucocorticoids and catecholamines support the production of APP which rise rapidly in the serum to maximum levels (up to 1, 000 x) within 1-2 days. Natural antibody levels also show an abrupt increase. By this the serum concentration of polyspecific defense molecules, such as natural antibodies, LPS binding protein, C-reactive protein, mannose-binding protein is increased enormously. Complement production is also elevated, potentiating further the efficiency of polyspecific defense molecules. A number of APP function as enzyme inhibitors and inhibitors of inflammation, which are likely to provide damage control during febrile illness [334-339]. All these changes are consistent with the rapid enhancement of polyspecific host resistance to infection and to various other insults as originally observed by Selye (Figure 1). Febrile illness is an emergency defense reaction, which takes over the task of host defense in situations when other defense mechanisms, including adaptive immunity, have failed. During APR the adaptive immune response, which is dependent on T cells, is suppressed and the immune system is placed under the command of natural antibodies and liver derived recognition molecules. These molecules are capable of recognizing homotopes on pathogens and on altered self components and activate various immune mechanisms after combining with their spcific target determinants. In this situation interleukin 6 is likely to function as an emergency competence hormone and insulin may be the principle growth factor fueling elevated leukocyte production and activity. Elevated serum levels of leptin ensure the energy requirements of APR. Inhibitory cytokines, such ILl receptor antagonist, TNF synthesis inhibitor, ILl0 and leukemia inhibitory factor are also elevated and participate in the regulation of inflammatory processes. During APR INFy excess serves as an antagonist of these cytokines [248,256, 257,260, 334-339]. The immunoconversion during APR from the adaptive mode of reactivity to the amplification of natural immune mechanisms provides instantaneous and rapidly increasing defense at the expense of muscles and other tissues and organs, which undergo catabolism. Therefore, the natural immune system provides the first line of host defense during health and it also serves as the last resort of host defense in crisis situations. The acute phase response is a highly coordinated pathophysiological reaction where cyctokines, inflammation and the metabolic activity of various organs and tissues are tightly regulated, all in the interest of host defense [332]. For this reason McEwen has adapted the term "allostasis" in contrast with homeostasis [341]. Indeed evidence is rapidly increasing that the "allostatic milieu" is a prerequisite for the suppression of the adaptive immune response and the amplification of innate immunity.
25
Practical observations indicate that APR is a very effective defense reaction indeed, as in the overwhelming majority of febrile illness recovery is the rule, which is followed by the development of specific immunity.
5.
SUMMARY AND CONCLUSIONS
The stroma of various tissues and organs fulfils an important regulatory function towards the paranchymal cells that perform the specific tasks characteristic of the organ/tissue. Lymphoid cells (monocytes, macrophages, T and B lymphocytes, specialized antigen presenting cells, mast cells) are invariably present in the stroma and contibute to regulation. There is evidence for this in the nervous system, in the gastrointestinal tract, in the pituitary gland, in the adrenals and gonads, breast tissue and in other reproductive organs and in the skin. Cell-to-cell communication takes place within tissues and organs via adhesion molecules, which may be tissue- site- and cell-specific or shared with other organs and tissues. Adherence signals may promote or inhibit function, depending on the local requirements. Matrix components also deliver local regulatory signals. PRL, GH and IGF-I maintain the cells and tissues of the body in a functional competent state. Most tissues have the capacity to synthesize PRL, GH and IGF-I. This local production allows for tissues/organs/systems to amplify locally specific functions and to increase the adaptability of the organism (e.g. the adaptive immune response). Tissue specific growth factors may fulfil the role of competence hormones (possible examples are: IL-2, IL-3, IL-6, GM-CSF, epidermal growth factor, fibroblast growth factor). This remains to be established. Therefore, it is suggested that tissues/organs/systems function as partially independent units, capable of generating all three categories of regulatory signals upon functional demands. This provides flexibility and plasticity for adaptaion to the requirements that need to be fulfilled. Various cells of the immune system home to organs and tissues specifically. This is governed by tissue specific adhesion molecules and by humoral signals such as chemokines, and cytokines. Monocyte/macrophage type cells and mast cells take up residence in the tissues and there is evidence for tissue specific differentiation. Examples of differentiation are the glia cells in the brain, mucosal versus the connective tissue type mast cells, Langerhans cells in the skin, Kupffer cells in the liver, and dendritic cells throughout the body. Endothelial cells mediate the communication between leukocytes and specific tissues via cytokines and adhesion molecules. The endothelium may also serve as a barrier between leukocytes and the tissue (e.g. blood brain barrier), but are also involved in the increase of vascular permeability during inflammatory reactions. The CNS has dominant regulatory powers in the body that includes the regulation of immune phenomena. The regulation of the inflammatory response by nerve impulses, the phenomenon of conditioning immune responses, the intricate and sophisticated neuroimmune mechanisms that are built into the process of reproduction in higher animals, especially mammals, the fact that emotions and stress affect immune reactivity are all examples pointing to CNS control. Immune phenomena show circadian and seasonal variation and this also indicates the existence of neuroendocrine regulatory influences. The sleep-wake cycle is fundamental to the maintenance of health and normal immune function [98, 285, 341-345]. Therefore the neuroimmune regulatory network is fundamental to the maintainence of health. Our initial experiments revealed the important role of hormones secreted or regulated by the pituitary gland in immunodeficiency, in hematopoiesis, in the cytokine response to infectious agents, in autoimmune diseases, in host resistance to cancer and in the overall survival of
26
the organism [38-42, 128, 136, 155, 184, 329, 347-354]. Currently there is much evidence, indicating that abnormalities of the neuroimmunoregulatory network are associated with diseases of the nervous-, endocrine-, and immue systems and indeed, of other tissues and organs. It is beyond the scope of this overview to discuss this subject in detail. Instead, some recent publications are cited for the interested reader in addition to the ones presented in this volume [355-359]. One may conclude on the basis of available evidence that the nervous-, endocrine-, and immune systems form a regulatory network, which is fundamental to the normal development and function of individuals from conception till death. This regulatory system also plays a role in host protection against pathological insults and in regeneration and healing. Therefore, the application of the term Neuroimmune Biology to define this multi-disciplinary and integrative science is fully justified.
ACKNOWLEDGEMENTS I owe special homage to Hans Selye, who was my teacher and sparked my interest in neuroimmune interaction. Over the years many colleagues collaborated/contributed to the acquisition of the knowledge and ideas presented in this paper. I owe special thanks to Drs. Andor Szentivanyi, Eva Nagy, Henry Friesen, Kalman Kovacs, Dwight Nance, Donna Chow, Robert Shiu, Edward Baral, Richard Warrington, Lorand Bertok, John Kellen and Sylvia Asa in this respect. The experimental work discussed in this article was supported in part by MRC of Canada, The Arthritis Society of Canada, The Manitoba Health Research Council, The Manitoba Medical Services Foundation, Cancer Care Manitoba and Orion-Pharmos Corporation of Finland. I am indebted to Carol Funk and Valentina Tautkus for their devoted work on this manuscript.
REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10.
Sternberg EM. The Balance Within: The Science Connecting Health and Emotions. New York: WH Freeman and Company, 2000. Thompson WG. Placebos: A review of the placebo response. Am J Gastroenterol 2000; 95(7): 1637-43. Nowotny A. In: Nowotny, editor. Beneficial effects of endotoxins. New York, Plenum Press, 1993: 1-55. Hammar JA. The new views as to the morphology of the thymus gland and their bearing on the problem of the function of the thymus. Endocrinology 1921; 5: 543-73, 731-60. Smith PE. Effects of hypophysectomy upon the involution of the thymus in the rat. Anat Rec 1930; 47:119-29. Selye H. A syndrome produced by diverse nocuous agents. Nature (Lond) 1936; 138: 32. Selye H. Thymus and adrenals in the response of the organism to injuries and intoxication. Brit J Exp Path 1936; 17: 234-248. Selye H. Morphological changes in the fowl following chronic overdosage with various steroids. J Morph 1943; 73: 401-421. Selye H. The general adaptation syndrome and the diseases of adaptation. J Clin Endocrinology 1946; 6:117-230. Selye H. Effect of ACTH and cortisone upon an "anaphylactoid reaction". Can Med
27
11. 12. 13. 14. 15. 16.
17.
18. 19.
20. 21.
22. 23.
24.
25.
26.
27.
28.
Assoc J 1949; 61: 553-556. Selye H. Stress and disease. Science 1955; 122: 625-31. Selye H. The Mast Cells. Washington: Butterworth, 1965. Filipp G, Szentivanyi A, Mess B. Anaphylaxis and nervous system. Acta Med Hung 1952; 2: 163-73. Szentivanyi A, Filipp G. Anaphylaxis and the nervous system. Part II. Ann Allergy 1958; 16: 143-151. Filipp G, Szentivanyi A. Anaphylaxis and the nervous system. Part III. Allergy 1958; 16:306-11. Szentivanyi A, Szekely J. Effect of injury to, and electrical stimulation of, hypothalamic areas on anaphylactic and histamine shock of the guinea pig. A preliminary report. Ann Allergy 1956; 14: 259-261. Townley RG, Trapani IL, Szentivanyi A. Sensitization to anaphylaxis and to some of its pharmacological mediators by blockade of the beta adrenergic receptors. J Allergy 1967; 39(3): 177-79. Szentivanyi A. The beta adrenergic theory of the atopic abnormality in bronchial asthma. J Allergy 1968; 42: 203-32. Szentivanyi A, Szentivanyi J, Wagner H. Measurement of numbers of adrenoceptors in lymphocytes and lung tissue of patients with reversible obstructive airways disease. Clin Pharmacol Ther 1980; 27: 193-206. Szentivanyi A. The conformational flexibility of adrenoceptors and the constitutional basis of atopy. Triangle 1979; 18(4): 109-15. Szentivanyi A, Heim O, Schultze P. Changes in adrenoceptor densities in membranes of lung tissue and lymphocytes from patients with atopic disease. Ann NY Acad Sci 1979; 332: 295-298. Szentivanyi A. La flexibilite de conformation des adrenocepteurs et la base constitutionelle du terrain allergique. Rev Franc Allergol 1979; 19:205-214. Szentivanyi A, Fitzpatrick DF. The altered reactivity of the effector cells to antigenic and pharmacological influences and its relation to cyclic nucleotides. II. Effector reactivities in the efferent loop of the immune response. In: Filipp G, editor. Pathomechanismmus und Pathogenese Allergischer Reaktionen. Werk-Verlag Dr. Edmund Banachewski, Grafelfing bei Munchen 1980:511-580. Szentivanyi A, Polson JB, Krzanowski JJ. The altered reactivity of the effector cells to antigenic and pharmacological influences and its relation to cyclic nucleotides. I. Effector reactivities in the efferent loop of the immune response. In: Filipp G, editor. Pathomechanismus und Pathogenese Allergischer Reaktionen. Werk-Verlag, Dr. Edmund Banachewski, Grafelfing bei Munchen 1980:460-510. Szentivanyi A, Krzanowski JJ, Polson JB, Anderson WH. Evolution of research strategy in the experimental analysis of the beta adrenergic approach to the constitutional basis of atopy. In: Oehling A, Mathov E, Glazer I, Arbesman C, editors. Advances in Allergology and Clinical Immunology. Oxford, Pergamon Press, 1980:301-308. Szentivanyi A. Effect of bacterial products and adrenergic blocking agents on allergic reactions. In: Samter M, Talmage DW, Rose B, Sherman WB, Vaughan JH, editors. Textbook of Immunological Diseases. Boston, MA: Little, Brown and Co.1971: 356-374. Szentivanyi A, Krzanowski JJ, Polson JB. The autonomic nervous system: Structure, function, and altered effector responses. In: Middleton E, Reed CE, Ellis EF, editors. Allergy: Principles and Practice. St. Louis, MO, The CV Mosby Co, 1978: 256-300. Szentivanyi A, Williams JF. The constitutional basis of atopic disease. In: Bierman
28
29.
30. 31. 32.
33. 34. 35. 36.
37. 38. 39. 40. 41. 42. 43.
44. 45. 46. 47. 48.
49.
CW, Pearlman DS, editors. Allergic Diseases of Infancy, Childhood, and Adolescence. Philadelphia, PA, WB Saunders Co1980: 173-210. Szentivanyi A. Adrenergic and cholinergic receptor studies in human lung and lymphocytic membranes and their relation to bronchial hyperreactivity in asthma. In: Patient Care Publications. Darien, CT, 1982: 175-92.. Hackney JF, Szentivanyi A. The specificity of glucocorticoids in the relaxation of respiratory smooth muscle in vitro. J Allergy Clin Immunol 1975; 55:123 (abstract). Hackney JF, Szentivanyi A. The unique action of glucocorticoid succinates on respiratory smooth muscle in vitro. The Pharmacologist 1975; 17: 271. (abstract). Lowitt S, Szentivanyi A, Williams JF. Endotoxin inhibition of dexamethasone induction of tryptophan oxygenase in suspension culture of isolated rat parenchymal cells. II. Effect of in vivo pretreatment of rats with endotoxin. Biochem Pharmacol 1982; 31: 3403-3406. Szentivanyi A, Szentivanyi J. Mechanisms of action of corticosteroids. In: Proceedings of the International Symposium on Allergy and Immunology. Lima, Peru, 1985: 531-534.. Korneva EA, Khai LM. Effect of destruction of hypothalamic areas on immunogenesis. Fed Proc 1964; 23: T88. Jancso N. Role of the nerve terminals in the mechanism of inflammatory reactions. Bull. Millard Fillmore Hosp (Buffalo NY) 1960; 7-53. Jancso N. Desensitization with capsaicin and related acylamides as tool for studying the function of pain receptors. In: RKS Lim, editor. Pharmacology of Pain. Oxford, Pergamon Press, 1968. Dougherty TF. Effect of hormones on lymphatic tissue. Physiol Rev 1952; 32: 379-407. Nagy E, Berczi I. Immunodeficiency in hypophysectomized rats. Acta Endocrinology 1978; 89: 530-537. Berczi I, Nagy E, Kovacs K, Horvath E. Regulation of humoral immunity in rats by pituitary hormones. Acta Endocrinology 1981; 98:506-513. Nagy E, Berczi I, Friesen HG. Regulation of immunity in rats by lactogenic and growth hormones. Acta Endocrinology 1983; 102: 351-357. Nagy E, Berczi I, Wren GE, Asa SL, Kovacs K. Immunomodulation by bromocriptine. Immunopharmacology 1983; 6: 231-243. Berczi I (author/editor). Pituitary Function and Immunity (with 12 contributing authors). Monograph published by CRC Press, Inc., Boca Raton, FL, 1986. Wannemacher RW Jr, Pekarek RS, Thompson WL, et al. A protein from polymorphonuclear leukocytes (LEM) which affects the rate of hepatic amino acid transport and synthesis of acute-phase globulins. Endocrinology 1975; 96:651-661. Besedovsky HO, del Rey A, Sorkin E. Lymphokine-containing supernatants from Con A-stimulated cells increase corticosterone blood levels. J Immunol 1981; 126: 385-7. Berkenbosch F, van Oers J, del Rey A, Tilders F, Besedovsky H. Corticotropin-releasing factor-producing neurons in the rat activated by interleukin-1. Science 1987; 238: 524-526. Sapolsky R, Rivier C, Yamamoto G, Plotsky P, Vale W. Interleukin-1 stimulates the secretion of hypothalamic corticotropin-releasing factor. Science 1987; 238: 522-524. Nakamura H, Motoyoshi S, Kadokawa T. Anti-inflammatory action of interleukin 1 through the pituitary-adrenal axis in rats. Eur J Pharmacol 1988; 141: 67-73. Uehara A, Gottschall PE, Dahl RR, Arimura A. Interleukin-1 stimulates ACTH release by an indirect action which requires endogenous corticotropin releasing factor. Endocrinology 1987; 121: 1580-1582. Bernton EW, Beach JE, Holaday JW, Smallridge RC, Fein HG. Release of multiple
29
50.
51. 52.
53. 54.
55. 56. 57. 58. 59. 60. 61.
62.
63. 64. 65.
66.
67. 68. 69.
hormones by a direct action of interleukin-1 on pituitary cells. Science 1987; 238:519-521. McCann SM, Karanth S, Kamat A, Les Dees W, Lyson K, Gimeno M, Rettori V. Induction by cytokines of the pattern of pituitary hormone secretion in infection. Neuroimmunomodulation 1994; 1: 2-13. Pletsityi DF, Averi'anova LL. The influence of vagotomy on the development of experimental endocarditis. Dokl Akad Nauk SSSR 1966; Mar 1;167(1): 238-40. Russian Williams JM, Petersen RG, Shea PA, Schemdtje JF, Bauer DC, Felten DL. Sympathetic innervation of murine thymus and spleen: evidence for a functional link between the nervous and immune systems. Brain Res Bull 1981; Jan; 6(1): 83-94. Nance DM. Innervation of the spleen in the rat: evidence for absence of afferent innervation. Burns J Brain Behav Immun 1989; Dec; 3(4): 281-90. Fleshner M, Goehler LE, Hermann J, Relton JK, Maier SF, Watkins LR. Interleukin-1 beta induced corticosterone elevation and hypothalamic NE depletion is vagally mediated. Brain Res Bull 1995; 37(6): 605-10. Metal'nikov S, Chorine V. Roles des reflexes conditionels dans l'immunite. Ann Inst Pasteur 1926; 40: 893-900. Smith GH, Salinger R. Hypersensitivity and the conditioned reflex. Yale J Biol Med 1933; 5: 387-402. Ader R. Letter: Behaviorially conditioned immunosuppression. Psychosom Med 1974, 36(2): 183-184. MacQueen G, Marshall J, Perdue M, Siegel S, Bienenstock J. Pavlovian conditioning of rat mucosal mast cells to secrete rat mast cell protease II. Science 1989; 243: 83-85. Gorczynski RM, Macrae S, Kennedy M. Conditioned immune response associated with allogeneic skin grafts in mice. J Immunol 1982; 129: 704-9. Johnson HM, Smith EM, Torres BA, Blalock JE. Regulation of the in vitro antibody response by neuroendocrine hormones. Proc Nat Acad Sci USA 1982; Sci 79: 4171-4. Montgomery DW, Zukoski CF, Shah GN, Buckley AR, Pacholczyk T, Russell DH. Concanavalin A-stimulated murine splenocytes produce a factor with prolactin-like bioactivity and immunoreactivity. Biochem Biophys Res Commun 1987; Jun 15; 145(2): 692-8. DiMattia, Gellersen B, Duckworth ML, Friesen HG. Human prolactin gene expression. The use of an alternative noncoding exon in decidua and the IM-9-P3 lymphoblast cell line. J Biol Chem 1990; Sept 25; 265(27): 16412-21. Salamonsen LA, Woolley DE. Menstruation: induction by matrix metalloproteinases and inflammatory cells. J Reprod Immunol 1999; Sept; 44(1-2): 1-27. Kim HG, Kassis J, Souto JC, Turner T, Wells A. EGF receptor signaling in prostate morphogenesis and tumorigenesis. Histol Histopathol 1999; Oct; 14(4): 1175-82. Bodey B, Bodey B Jr, Siegel SE, Kaiser HE. Molecular biological ontogenesis of the thymic reticulo-epithelial cell network during the organization of the cellular microenvironment. In Vivo. 1999 May-Jun; 13(3): 267-94. B ianco P, Riminucci M, Kuznetsov S, Robey PG. Multipotential cells in the bone marrow stroma: regulation in the context of organ physiology. Crit Rev Eukaryot Gene Expr 1999; 9(2): 159-73. Robinson GW, Karpf AB, Kratochwil K. Regulation of mammary gland development by tissue interaction. J Mammary Gland Biol Neoplasia 1999; Jan; 4(1): 9-19. Maeurer MJ, Lotze MT. Interleukin-7 (IL-7) knockout mice. Implications for lymphopoiesis and organ-specific immunity. Int Rev Immunol 1998; 16(3-4): 309-22. Fidler IJ. Angiogenesis and cancer metastasis. Cancer J Sci Am 2000; Apr; 6 Suppl 2: S134-41.
30
70. 71. 72. 73. 74. 75. 76. 77.
78. 79. 80.
81. 82. 83.
84. 85.
86. 87. 88.
89. 90. 91.
King A. Uterine leukocytes and decidualization. Hum Reprod Update 2000; Jan-Feb; 6(1): 28-36. Wagner JG, Roth RA. Neutrophil migration mechanisms, with an emphasis on the pulmonary vasculature. Pharmacol Rev 2000; Sept; 52(3): 349-74. Pirtskhalaishvili G, Nelson JB. Endothelium-derived factors as paracrine mediators of prostate cancer progression. Prostate 2000; Jun 15; 44(1): 77-87. Tanaka Y. Integrin activation by chemokines: relevance to inflammatory adhesion cascade during T cell migration. Histol Histopathol 2000; Oct; 15(4): 1169-76. Alford D, Taylor-Papadimitriou J. Cell adhesion molecules in the normal and cancerous mammary gland. J Mammary Gland Biol Neoplasia 1996; Apr; 1(2): 207-18. Kimber SJ, Spanswick C. Blastocyst implantation: the adhesion cascade. Semin Cell Dev Biol 2000; Apr; 11(2): 77-92. Shiosaka S, Yoshida S. Synaptic microenvironments--structural plasticity, adhesion molecules, proteases and their inhibitors. Neurosci Res 2000; Jun; 37(2): 85-9. Renard V, Cambiaggi A, Vely F, B lery M, Olcese L, Olivero S, Bouchet M, Vivier E. Transduction of cytotoxic signals in natural killer cells: a general model of fine tuning between activatory and inhibitory pathways in lymphocytes. Transplant Rev 1997; 155: 205-221. DaEron M. Fc receptor biology. Annu. Rev Immunol 1997; 15: 203-34. Isakov N. Immunoreceptor thyrosine-based activation motif (ITAM), a unique module linking antigen and Fc receptors to their signaling cascades. J Leukoc B iol 1997; 61: 6-16. Venters HD, Tang Q, Liu Q, VanHoy RW, Dantzer R, Kelley KW. A new mechanism of neurodegeneration: a proinflammatory cytokine inhibits receptor signaling by a survival peptide. Proc Natl Acad Sci USA 1999; Aug 17; 96(17): 9879-84. Yee D, Lee AV. Crosstalk between the insulin-like growth factors and estrogens in breast cancer. J Mammary Gland Biol Neoplasia 2000; Jan; 5(1): 107-15. Kawashima K, Fujii T. Extraneuronal cholinergic system in lymphocytes. Pharmacol Ther 2000; Apr; 86(1): 29-48. Ptasznik A, Gewirtz AM. Crosstalk between G protein-coupled receptors and tyrosine kinase signaling: Src take centre stage. Arch Immunol Ther Exp (Warsz) 2000; 48(1): 27-30. Bornfeldt KE, Krebs EG. Crosstalk between protein kinase A and growth factor receptor signaling pathways in arterial smooth muscle. Cell Signal 1999; Jul; 11 (7): 465-77. Folli F, Saad MJ, Velloso L, Hansen H, Carandente O, Feener EP, Kahn CR. Crosstalk between insulin and angiotensin II signalling systems. Exp Clin Endocrinol Diabetes 1999; 107(2): 133-9. Paul WE. The immune system: An introduction. In: Paul WE, editor. Fundamental Immunology, 4 th edition. Lippincott-Raven, 1999: 1-35. Clevenger CV, Freier DO, Kline JB. Prolactin receptor signal transduction in cells of the immune system. J Endocrinol 1998; 157: 187-197. Chang L, Crowston JG, Cordeiro MF, Akbar AN, Khaw PT. The role of the immune system in conjunctival wound healing after glaucoma surgery. Surv Ophthalmol 2000; Jul-Aug; 45(1): 49-68. Schwartz M, Cohen IR. Autoimmunity can benefit self-maintenance. Immunol Today 2000; Jun; 21 (6): 265-8. Neumann H. The immunological microenvironment in the CNS: implications on neuronal cell death and survival. J Neural Transm Suppl 2000; 59: 59-68. Havran WL. A role for epithelial gammadelta T cells in tissue repair. Immunol Res
31
92. 93.
94. 95.
96. 97.
98. 99.
100.
101.
102.
103.
104.
105. 106.
107. 108. 109. 110.
2000; 21(2-3): 63-9. Abbott NJ. Inflammatory mediators and modulation of blood-brain barrier permeability. Cell Mol Neurobiol 2000; Apr; 20(2): 131-47. Antel J. Regulation of the Immune Response within the Central Nervous System. In: Berczi I, Gorczynski R, editors. Neuroimmune Biology: New Foundation of Biology. Amsterdam, Elsevier, 2001 (this volume): 87-98. Stevens-Felten SY, Bellinger DL. Noradrenergic and peptidergic innervation of lymphoid organs. Chem Immunol 1997; 69: 99-131. Nance D. Immunoregulation by the sympathetic nervous system. In: Berczi I, Gorczynski R, editors. Neuroimmune Biology: New Foundation of Biology. Amsterdam, Elsevier, 2001 (this volume): 121-139. McKay DM, B ienenstock J. The interaction between mast cells and nerves in the gastrointestinal tract. Immunology Today 1994; 15: 533-538. Derijk RH, Strijbos PJ, van Rooijen N, Rothwell NJ, Berkenbosch F. Fever and thermogenesis in response to bacterial endotoxin involve macrophage-dependent mechanisms in rats. Am J Physiol 1993 Nov 265(5 Pt 2): Rl179-83. Berczi I, Nagy E. Immune-endocrine interactions. In: Stefaneanu L, Sasano H, Kovacs K editors. Molecular and Cellular Endocrine Pathology. Chapman and Hall, 2000: 425-456. Moldofsky H. Sleep, health and immunocompetence. In: Berczi I, Gorczynski RM, editors. Neuroimmune Biology: New Foundation of Biology. Amsterdam, Elsevier, 2001 (this volume): 255-268. Mtiller EE, Salerno F, Cocchi D, et al. Interaction between the thyrotropin releasing hormone-induced growth hormone rise and dopaminergic drugs: Studies in pathological conditions in the animal and man. Clin Endocrinol 1979; 11: 645-56. Van den Berghe G, de Zegher F, Vlasselaers D, Schetz M, Verwaest C, Ferdinande P, Lauwers P. Thyrotropin-releasing hormone in critical illness: From a dopamine-dependent test to a strategy for increasing low serum triiodothyronine, prolactin, and growth hormone concentrations. Crit Care Med 1996; 24: 590-5. Li S, Crenshaw EB III, Rawson EJ, Simmons DM, Swanson LW, Rosenfeld MG. Dwarf locus mutants lacking three pituitary cell types result from mutations in the POU-domain gene pit-1. Nature 1990; 347: 528-33. Castro PR, Penalva R, Pereda MP, Renner U, Reul JMHM, Stalla GK, Holsboer F, Arzt E. Early activation of thyrotropin-releasing-hormone and prolactin plays a critical role during a T cell-dependent immune response. Endocrinology 1999;140: 690-97. Raiden S, Polack E, Nahmod V, Labeur M, Holsboer F, Larzt E. TRH receptor on immune cells: in vitro and in vivo stimulation of human lymphocyte and rat splenocyte DNA synthesis by TRH. J Clin Immunol 1995; 15: 242-9. Wang J, Klein JR. Thymus-neuroendocrine interactions in extrathymic T cell development. Science 1994; 265: 1860-1862. Komorowski J, Zylinska K, Pawlikowski M, Stepien H. Stimulatory effect of thyrotropin (TSH) on interleukin-2 (IL-2) release from human peripheral blood lymphocytes. A dose-response study in vitro. Hormones Metab Res 1993; 25: 598-599. Nicoll CS. Prolactin and growth hormone: Specialists on one hand and mutual mimics on the other. Perspect Biol Med 1982; 25: 369-381. Berczi I. The immunology of prolactin. Semin Reprod Endocrinology 1992; 10: 196-219. Lewis UJ. Growth hormone - What is it and what does it do. Trends Endocrinol Metab 1992; 3:117-121. Southard JN, Talamantes F. Placental prolactin-like proteins in rodents - Variations on a
32
structural theme. Mol Cell Endocrinology 1991; 79: C133-140. 111. Nickel BE, Kardami E, Cattini PA. The human placental growth hormone variant is mitogenic for rat lymphoma Nb2 cells. Endocrinology 1990; 126: 971-976. 112. Duckworth ML, Robertson MC, Schroedter IC, Szpirer C, Friesen HG. Molecular genetics and biology of the rat placental prolactin family. In: Soares MJ, Handwerger S, Talamantes F, editors. Trophoblast cells: pathways for maternal-embryonic communication. New York: Springer, 1993:169-190. 113. Berczi I, Chalmers IM, Nagy N, Warrington RJ. The immune effects of neuropeptides. Bailliere Clin Rheumatol 1996; 10: 227-257. 114. Rohn WM, Weigent DA. Cloning and nucleotide sequencing of rat lymphocyte growth hormone cDNA. Neuroimmunomodulation 1995; 2: 108-14. 115. Palmetshofer A, Zechner D, Luger TA, Barta A. Splicing variants of the human growth hormone mRNA: detection in pituitary, mononuclear cells and dermal fibroblasts. Mol Cell Endocrinology 1995; 113: 225-34. 116. Sabharwal P, Varma S. Growth hormone synthesized and secreted by human thymocytes acts via insulin-like growth factor I as an autocrine and paracrine growth factor. J Clin Endocrinol Metab 1996; 81: 2663-9. 117. Matera L. Endocrine, paracrine and autocrine actions of prolactin on immune cells. Life Sci 1996; 59:599-614. 118. Kelly PA, Djiane J, Postel-Vinay MC, Edery M. The prolactin/growth hormone receptor family. Endocr Rev 1991 ; 12:235-251. 119. Wells JA, de Vos AM. Structure and function of human growth hormone: Implications for the hematopoietins. Annu. Rev Biophys. Biomol Struct. 1993; 22: 329-351. 120. Kishimoto T, Taga T, Akira S. Cytokine signal transduction. Cell 1994; 76: 253-262. 121. Ihle JN. The Janus kinase family and signaling through members of the cytokine receptor superfamily. Proc Soc Exp Biol Med 1994; 206: 268-72. 122. Rui H, Djeu JY, Evans GA, Kelley PA, Farrar WL. Prolactin receptor triggering: Evidence for rapid tyrosine kinase activation. J Biol Chem. 1992; 267:24076-24081. 123. Clevenger CV, Sillman AL, Prystowsky MB. Interleukin-2 driven nuclear translocation of prolactin in cloned lymphocytes-T. Endocrinology 1990; 127:3151-3159. 124. Horseman ND, Yu-Lee LY. Transcriptional regulation by the helix bundle peptide hormones: growth hormone, prolactin, and hematopoietic cytokines. Endocrine Rev 1994; 15: 627-649. 125. Berczi I. Pituitary hormones and immune function, Acta Paediatr. (Suppl) 1997; 423: 70-75. 126. Potter EL, Craig JM. Pathology of the Fetus and the Infant. 3rd edition. Chicago, Year Book, 1975. 127. Arezzini C, De Gori V, Tarli P, Neri P. Weight increase of body and lymphatic tissues in dwarf mice treated with human chorionic somatomammotropin (HCS). Proc Soc Exp Biol Med 1972; 141: 98-100. 128. Berczi I, Nagy E. Human placental lactogen is a hemopoietic hormone. Br J Haematol. 1991 ; 79: 355-358. 129. Berczi I, Nagy E. The effect of prolactin and growth hormone on hemolymphopoietic tissue and immune function. In: Berczi I, Kovacs K, editors. Hormones and Immunity. Lancaster, England, MTP Press, 1987; 145-171. 130. Berczi I. The role of the growth and lactogenic hormone family in immune function. Neuroimmunomodulation. 1994; 1: 201-216. 131. Bellone G, Geuna M, Carbone A, Silvestri S, Foa R, Emanuelli G, et al. Regulatory action of prolactin on the in vitro growth of CD34+ve human hemopoietic progenitor cells. J
33
Cell Physiol 1995; 163: 221-31. 132. Merchav S, Tatarsky I, Hochberg Z. Enhancement of erythropoiesis in vitro by human growth hormone is mediated by insulin-like growth factor-I. Br J Haematol 1988; 70: 267-271. 133. Merchav S, Tatarsky I, Hochberg Z. Enhancement of human granulopoiesis in vitro by biosynthetic insulin-like growth factor I/somatomedin C and human growth hormone. J. Clin. Invest 1988; 81: 791-797. 134. Kelley KW, Arkins S, Minshall C, Liu Q, Dantzer R. Growth hormone, growth factors and hematopoiesis. Hormone Res 1996; 45: 38-45. 135. Kelley KW, Brief S, Westly HJ, Novakofski J, Bechtel PJ, Simon J, Walker EB. GH 3 pituitary adenoma cells can reverse thymic aging in rats. Proc Natl Acad Sci USA 1986; 83: 5663-5667. 136. Berczi I, Nagy E, De Toledo SM, Matusik RJ, Friesen HG. Pituitary hormones regulate c-myc and DNA synthesis in lymphoid tissue. J Immunol 1991; 146:2201-2206. 137. Beschorner WE, Divic J, Pulido H, Yao X, Kenworthy P, Bruce G. Enhancement of thymic recovery after cyclosporine by recombinant human growth hormone and insulinlike growth factor-I. Transplantation 1991; 52: 879-884. 138. B inz K, Joller P, Froesch P, B inz H, Zapf J, Froesch ER. Repopulation of the atrophied thymus in diabetic rats by insulin-like growth factor I. Proc Natl Acad Sci USA 1990; 87: 3690-3694. 139. Timsit J, Savino W, Safieh B, Chanson P, Gagnerault MC, Bach JF, Dardenne M. Growth hormone and insulin-like growth factor-I stimulate hormonal function and proliferation of thymic epithelial cells. J Clin Endocrinol Metab 1992; 75: 183-188. 140. Kaplan SL, Grumbach MM, Friesen HG, Costom BH. Thyrotropin-releasing factor (TRF) effect on secretion of human pituitary prolactin and thyrotropin in children and in idiopathic hypopituitary dwarfism: further evidence for hypophysiotropic hormone deficiencies. J Clin Endocrinol Metab 1972; 35: 825-830. 141. Spangelo BL, Hall NRS, Ross PC, Goldstein AL. Stimulation of in vivo antibody production and concanavalin-A-induced mouse spleen cell mitogenesis by prolactin. Immunopharmacology 1987; 14:11-20. 142. Nagy E, Berczi I, Wren GE, Asa SL, Kovacs K. Immunomodulation by bromocriptine. Immunopharmacology 1983; 6: 231-243. 143. Cesano A, Oberholtzer E, Contarini M, Geuna M, Bellone G, Matera L. Independent and synergistic effect of interleukin-2 and prolactin on development of T- and NK-derived LAK effectors. Immunopharmacology 1994; 28: 67-75. 144. Bozzola M, Valtorta A, Moretta A, Cisternino M, Biscaldi I, Schimpff RM. In vitro and in vivo effect of growth hormone on cytotoxic activity. J Pediat1990; 117: 596-599. 145. Rovensky J, Ferencikova J, Vigas M, Lukac P. Effect of growth hormone on the activity of some lysosomal enzymes in neutrophilic polymorphonuclear leukocytes of hypopituitary dwarfs. Int J Tissue React 1985; 7: 153-159. 146. Wiedermann CJ, Niedermuhlbichler M, Beimpold, Braunsteiner H. In vitro activation of neutrophils of the aged by recombinant human growth hormone. J Infect Dis 1991; 164: 1017-1020. 147. Manfredi R, Tumietto F, Azzaroli L, Zucchini A, chiodo F, Manfredi G. Growth hormone (GH) and the immune system: impaired phagocytic function in children with idiopathic GH deficiency is corrected by treatment with biosynthetic GH. J Pediat Endocrinol 1994; 7: 245-251. 148. Cecilia-Fornari M, Fernanda-Palacios M, Diez RA, Intebi AD. Decreased chemotaxis
34
149. 150.
151.
152. 153.
154.
155. 156.
157.
158. 159. 160. 161.
162.
163.
164.
165.
of neutrophils in acromegaly and hyperprolactinemia. Eur J Endocrinology 1994; 130: 463-468. Rapaport R. Immunomodulation by growth hormone in humans. In: Berczi I, Szelenyi J, editors. Advances in Psychoneuroimmunology. New York: Plenum Press, 1994: 83-98. Galdiero M, Donnarumma G, Cipollareo de l'Ero G, Marcatili A, Scarfogliero P, Sommese L. Effect of growth hormone, prolactin and insulin on the release of IL-1 alpha, IFN-gamma and IL-4 by staphylococcal enterotoxin A-stimulated splenocytes. Eur Cytokine Netw 1995; 6: 187-94. LeRoith D, Yanowski J, Kaldjian EP, Jaffe ES, LeRoith T, Purdue K, Cooper BD, Pyle R, Adler W. The effects of growth hormone and insulin-like growth factor I on the immune system of aged female monkeys. Endocrinology 1996; 137: 1071-9. Gelato MC. Aging and immune function: a possible role for growth hormone. Hormone Res 1996; 45: 46-9. Dorshkind K, Horseman ND. The roles of prolactin, growth hormone, insulin-like growth factor-I, and thyroid hormones in lymphocyte development and function. Insights from genetic models of hormone and hormone receptor deficiency. Endocrine Rev 2000; 21: 292-312. Bole-Feysot C, Goffin V, Edery M, B inart N and Kelly PA. Prolactin (PRL) and its receptor: Actions, signal transduction pathways and phenotypes observed in PRL-receptor knockout mice. Endocrine Rev 1998; 19: 225-68. Nagy E, Berczi I. Hypophysectomized rats depend on residual prolactin for survival. Endocrinology 1991 ; 146: 2776-2784. Komorowski J, Stepien H, Pawlikowski M. The evidence of thyroliberin/triiodothyronine control of TSH secretory response from human peripheral blood monocytes cultured in vitro. Neuropeptides 1993; 25:31-4. Bagriacik EU, Klein JR. The thyrotropin (thyroid-stimulating hormone) receptor is expressed on murine dendritic cells and on a subset of CD45RBhigh lymph node T cells: functional role for thyroid-stimulating hormone during immune activation. J Immunol 2000; Jun 15; 164(12): 6158-65. Hsu J-H, Brent GA. Thyroid hormone receptor gene knock outs. Trends Endocrinol Metab 1998; 9:103-111. Fabris N, Mocchegiani E, Provinciali M. Pituitary-thyroid axis and immune system: a reciprocal neuroendocrine-immune interaction. Hormone Res 1995; 43: 29-38. Berczi I. The influence of pituitary-adrenal axis on the immune system. In: Berczi I, editor. Pituitary Function and Immunity. Boca Raton, Florida, CRC Press, 1986: 49-132. Besedovsky HO, del Rey A, Klusman I, Furukawa H, Monge-Arditi G, Kabiersch A. Cytokines as modulators of the hypothalamus-pituitary-adrenal axis. J Steroid Biochem. Mol Biol 1991; 40: 613-618. Vamvakopoulos NC, Chrousos GP. Hormonal regulation of human corticotropin-releasing hormone gene expression: implications for the stress response and immune/inflammatory reaction. Endocrine Rev 1994; 15: 409-420. Heijnen CJ, Zijlstra J, Kavelaars A, Croiset G, Ballieux RE. Modulation of the immune response by POMC-derived peptides. I. Influence on proliferation of human lymphocytes. Brain Behav Immun 1987; 1: 284-291. Catania A, Airaghi L, Colombo G, Lipton JM. Alpha-Melanocyte-stimulating Hormone in Normal Human Physiology and Disease States. Trends Endocrinol Metab 2000; Oct 1; 11(8): 304-308. Stein C, Schafer M, Hassan AHS. Peripheral opioid receptors. Ann Med 1995; 27: 219-221.
35
166. Taub DD, Eisenstein TK, Geller EB, Adler MW, Rogers TJ. Immunomodulatory activity of p- and K-selective opioid agonists. Proc Natl Acad Sci USA 1991; 88: 360-364. 167. Alicea C, Belkowski S, Eisenstein TK, Adler MW, Rogers TJ. Inhibition of primary murine macrophage cytokine production in vitro following treatment with the K-opioid agonist U50, 488H. J Neuroimmunol 1996; 64: 83-90. 168. B lalock JE. Proopiomelanocortin and the immune-neuroendocrine connection. Ann N Y Acad Sci 1999; Oct 20; 885: 161-72. 169. Catania A, Lipton JM. The neuropeptide alpha-melanocyte-stimulating hormone: a key component of neuroimmunomodulation. Neuroimmunomodulation. 1994; 1: 93-99. 170. Shimizu T, Streilein JW. Influence of alpha-melanocyte stimulating hormone on induction of contact hypersensitivity and tolerance. J Dermatol Sci 1994; 8:187-193. 171. Grabbe S, Bhardwaj RS, Mahnke K, Simon MM, Schwarz T, Luger TA. Alphamelanocyte-stimulating hormone induces hapten-specific tolerance in mice. J Immunol 1996; 156: 473-478. 172. King LB, Vacchio MS, Dixon K, Hunziker R, Margulies DH, Ashwell JD. A targeted glucocorticoid receptor antisense transgene increases thymocyte apoptosis and alters thymocyte development. Immunity 1995; 3: 647-656. 173. Wiegers GJ, Labeur MS, Stec IEM, Klinkert WEF, Holsboer F, Reul JMHM. Glucocorticoids accelerate anti-T cell receptor-induced T cell growth. J Immunol 1995; 155: 1893-1902. 174. Munck A, Naray-Fejes-Toth A, Guyre PM. Mechanisms of glucocorticoid actions on the immune system. In: Berczi I, Kovacs K, editors. Hormones and Immunity. Lancaster, England: MTP Press Ltd, 1987. 175. Vacchio MS, Papadopoulos V, Ashwell JD. Steroid production in the thymus: implications for thymocyte selection. J Exp Med 1994; 179: 1835-46. 176. Rook GA. Glucocorticoids and immune function. Baillieres Best Pract Res Clin Endocrinol Metab 1999; Dec; 13(4): 567-81. 177. Ashwell JD, Lu FW, Vacchio MS. Glucocorticoids in T cell development and function. Annu. Rev Immunol 2000; 18: 309-45. 178. Grasso G, Massai L, De Leo V, Muscettola M. The effect of LHRH and TRH on human interferon-gamma production in vivo and in vitro. Life Sci 1998; 62(22): 2005-14. 179. Athreya BH, Pletcher J, Zulian F, Weiner DB, Williams WV. Subset-specific effects of sex hormones and pituitary gonadotropins on human lymphocyte proliferation in vitro. Clin Immunol ImmunoPathol 1993; 66: 201-211. 180. Rouabhia M, Chakir J, Deschaux P. Interaction between the immune and endocrine systems: Immunodulatory effects of luteinizing hormone. Prog NeuroendocrinImmunol 1991; 4: 86-91. 181. Komorowski J, Stepien H. FSH and LH induce interleukin-6 (IL-6) release from human peripheral blood monocytes cultures in vitro. A dose-response study. Hormone Metab Res 1994; 26: 438-439. 182. Sullivan DA. Endocrine regulation of the ocular secretory immune system. In: Berczi I, Kovacs K, editors. Hormones and immunity. Lancaster, England, MTP Press, 1987: 54-92. 183. Nelson JL, Steinberg AD. Sex steroids, autoimmunity and autoimmune disease. In: Berczi I, Kovacs, K, editors. Hormones and Immunity. Lancaster, England, MTP Press, 1987:93-119. 184. Nagy E, Baral E, Berczi I. Immune System. In: Oettel M, editor-in-chief. Handbook of Experimental Pharmacology. Estrogens and anti-estrogens. Section IV. Physiology and Pathophysiology of estrogens. Springer-Verlag (in press).
36
185. Lahita RG. The role of sex hormones in systemic lupus erythematosus. Curr Opin Rheumatol 1999; Sept; 11(5): 352-6. 186. Muller D, Chen M, Vikingsson A, Hildeman D, Pederson K. Oestrogen influences CD4 + T-lymphocyte activity in vivo and in vitro in [32-microglobulin-deficient mice. Immunology 1995; 86: 162-7. 187. Ahmed SA, Talal N, Christadoss P. Genetic regulation of testosterone-induced immune suppression. Cell Immunol 1987; 104(1): 91-98. 188. Greenstein BD, de Bridges EF, Fitzpatrick FTA. Aromatase inhibitors regenerate the thymus in aging male rats. Int J Immunopharm 1992; 14: 541-53. 189. Araneo B A, Dowell T, Diegel M, Daynes RA. Dihydrotesterone exerts a depressive influence on the production of interleukin-4 (IL-4), IL-5, and ~,-interferon, but not IL-2 by activated murine T cells. Blood 1991; 78: 688-99. 190. Konstadoulakis MM, Syrigos KN, Baxevanis CN, Syrigou EI, Papamichail M, Peveretos P, Anapliotou M, Golematis BC. Effect of testosterone administration, pre- and postnatally, on the immune system of rats. Horm Metab Res 1995; 27: 275-8. 191. Regelson W, Loria R, Kalimi M. Dehydroepiandrosterone (DHEA) - the "mother steroid." I. Immunologic action. Ann NY Acad Sci 719: 553-63. 192. Catania RA, Angele MK, Ayala A, Cioffi WG, Bland KI, Chaudry IH. Dehydroepiandrosterone restores immune function following trauma-haemorrhage by a direct effect on T lymphocytes. Cytokine 1999; Jun; 11 (6): 443-50. 193. Padgett DA, Loria RM. In vitro potentiation of lymphocyte activation by dehydroepiandrosterone, androstenediol, and androstenetriol. J Immunol 1994; 153:1544-52. 194. Suzuki T, Suzuki N, Daynes RA, Engleman EG. Dehydroepiandrosterone enhances IL2 production and cytotoxic effector function of human T cells. Clin Immunol ImmunoPathol 1991; 61:202-11. 195. Weksler ME. Immune senescence and adrenal steroids: immune dysregulation and the action of dehydroepiandrosterone (DHEA) in old animals. Eur J Clin Pharmacol 1993; 45: $21-3. 196. Loria RM, Padgett DA, Huynh PN. Regulation of the immune response by dehydroepiandrosterone and its metabolites. J Endocrinol 1996; 150: $209-20. 197. Check JH, Tarquini P, Gandy P, Lauer C. A randomized study comparing the efficacy of reducing the spontaneous abortion rate following lymphocyte immunotherapy and progesterone treatment versus progesterone alone in primary habitual aborters. Gynecol Obstet Invest 1995; 39: 257-61. 198. Fleming MW, Gamble HR. Consequences of dose-dependent immunosuppression by progesterone on parasitic worm burdens in lambs. Am J Vet Res 1993; 54: 1299-302. 199. Parr MB, Kepple L, McDermott MR, Drew MD, Bozzola JJ, Parr EL. A mouse model for studies of mucosal immunity to vaginal infection by herpes simplex virus type 2. Lab Invest 1994; 70: 369-80. 200. Kalo-Klein A, Witkin SS. Regulation of the immune response to Candida albicans by monocytes and progesterone. Am J Obstet Gynecol 1991; 164:1351-4. 201. Nohmi T, Abe S, Dobashi K, Tansho S, Yamaguchi H. Suppression of anti-Candida activity of murine neutrophils by progesterone in vitro: a possible mechanism in pregnant women's vulnerability to vaginal Candidiasis. Microbiol Immunol 1995; 39: 405-9. 202. Li ZG, Danis VA, Brooks PM. Effect of gonadal steroids on the production of IL-1 and IL-6 by blood mononuclear cells in vitro. Clin Exp Rheumatol 1993; 11: 157-62. 203. Szekeres-Bartho J, Barakonyi A, Polgar B, Par G, Faust Z, Palkovics T, Szereday The role of gamma/delta T cells in progesterone-mediated immunomodulation during pregnancy: a
37
review. Am J Reprod Immunol 1999; Jul; 42(1): 44-8. 204. Manolagas SC, Yu XP, Girasole G, Bellido T. Vitamin D and the hematolymphopoietic tissue: a 1994 update. Sem Nephrol 1994; 14: 129-43. 205. Bouillon R, Garmyn M, Verstuyf A, Segaert S, Casteels K, Mathieu C. Paracrine role for calcitriol in the immune system and skin creates new therapeutic possibilities for vitamin D analogs. Eur J Endocrinology 1995; 133: 7-16. 206. Yang S, Smith C, Prahl JM, Luo X, DeLuca HF. Vitamin D deficiency suppresses cell-mediated immunity in vivo. Arch Biochem. Biophys. 1993; 303: 98-106. 207. Piemonti L, Monti P, Sironi M, Fraticelli P, Leone BE, Dal Cin E, Allavena P, Di Carlo V. Vitamin D3 affects differentiation, maturation, and function of human monocyte-derived dendritic cells. J Immunol 2000; May 1; 164(9): 4443-51. 208. Daynes RA, Enioutina EY, Butler S, Mu HH, McGee ZA, Araneo B A. Induction of common mucosal immunity by hormonally immunomodulated peripheral immunization. Infect Immun 1996; 64:1100-9. 209. Casteels K, Bouillon R, Waer M, Mathieu C. Immunomodulatory effects of 1, 25-dihydroxyvitamin D3. Curr Opin Nephrol Hypertens 1995; 4: 313-8. 210. Branisteanu DD, Waer M, Sobis H, Marcelis S, Vandeputte M, Bouillon R. Prevention of murine experimental allergic encephalomyelitis: cooperative effects of cyclosporine and 1~, 25-(OH)zD 3. J Neuroimmunol 1995; 61: 151-60. 211. Lemire J, 1, 25-Dihydroxyvitamin D3--a hormone with immunomodulatory properties. Z Rheumatol 2000; 59 Suppl 1: 24-7. 212. Shedpure M, Pati AK. The pineal gland: structural and functional diversity. Indian J Exp Biol 1995; 33: 625-40. 213. Roberts JE. Visible light induced changes in the immune response through an eye-brain mechanism (photoneuroimmunology). J Photochem Photobiol B Biol 1995; 29: 3-15. 214. Morrey KM, McLachlan JA, Serkin CS, Bakouche O. Activation of human monocytes by the pineal hormone melatonin. J Immunol 1994; 153: 2671-80. 215. DiStefano A, Paulesu L. Inhibitory effect of melatonin on production of IFN gamma or TNF alpha in peripheral blood mononuclear cells of some blood donors. J Pineal Res 1994; 17: 164-169. 216. Giordano M, Vermeulen M, Palermo MS. Seasonal variations in antibody-dependent cellular cytotoxicity regulation by melatonin. Int J Neurosci 1993; 68: 109-25. 217. Mocchegiani E, Bulian D, Santarelli L, Tibaldi A, Mozzioli M, Pierpaoli W, Fabris N. The immuno-reconstituting effect of melatonin or pineal grafting and its relation to zinc pool in aging mice. J Neuroimmunol 1994; 53:189-201. 218. Maestroni GJ. MLT and the immune-hematopoietic system. Adv. Exp Med. Biol 1999; 460: 395-405. 219. Nelson RJ, Drazen DL. Melatonin mediates seasonal adjustments in immune function. Reprod Nutr Dev 1999; May-Jun; 39(3): 383-98. 220. Levi-Montalcini R. The nerve growth factor, its mode of action on sensory and sympathetic nerve cells. Harvey Lectures 1966; 60: 217-259. 221. Pezzati P, Stanisz AM, Marshall JS, Bienenstock J, Stead RH. Expression of nerve growth factor receptor immunoreactivity on follicular dendritic cells from human mucosa associated lymphoid tissues. Immunology 1992; 76: 485-490. 222. Lomen-Hoerth C, Shooter EM. Widespread neurotrophin receptor expression in the immune system and other nonneuronal rat tissues. J Neurochem 1995; 64: 1780-9. 223. Cosman D. A family of ligands for the TNF receptor superfamily. Stem Cells Dayt 1994; 12: 440-55.
38
224. Franklin RA, Brodie C, Melamed I, Tereda N, Lucas JJ, Gelfand EW. Nerve growth factor induces activation of MAP-kinase and p90 rsk in human B lymphocytes. J Immunol 1995; 154: 4965-4972. 225. Manning PT, Russell JH, Simmons B, Johnson EM. Protection from guanethidine-induced neuronal destruction by nerve growth factor: effect of NGF on immune function. Brain Res 1985; 340:61-69. 226. Otten U, Ehrhard P, Peck R. Nerve growth factor induces growth and differentiaton of human B lymphocytes. Proc Natl Acad Sci USA 1989; 86: 10059-10063. 227. Kimata H, Yoshida A, Ishioka C, Kusunoki T, Hosoi S, Mikawa H. Nerve growth factor specifically induces human IgG4 production. Eur J Immunol 1991; 21: 137-141. 228. Brodie C, Oshiba A, Renz H, Bradley K, Gelfand EW. Nerve growth factor and anti-CD40 provide opposite signals for the production of IgE in inteleukin-4-treated lymphocytes. Eur J Immunol 1996; 26:171-8. 229. Santambrogio L, Benedetti M, Chao MV, Muzaffar R, Kulig K, Gabellini N, Hochwald G. Nerve growth factor production by lymphocytes. J Immunol 1994; 153: 4488-95. 230. Leon A, Buriani A, DalToso R, et al. Mast cells synthesize, store, and release nerve growth factor. Proc Natl Acad Sci USA 1994; 91: 3739-43. 231. Aloe L, Levi-Montalcini R. Mast cells increase in tissues of neonatal rats injected with the nerve growth factor. Brain Res 1977; 133: 358-366. 232. Bohm A, Aloe L, Levi-Montalcini R. Nerve growth factor enhances precocious differentiation and numerical increase in mast cells in cultures of rat splenocytes. Acad Naz Lineei 1986; 80: 1. 233. Marshall JS, Stead RH, McSharry C, Nielsen L, Bienenstock J. The role of mast cell degranulation products in mast cell hyperplasia. I. Mechanism of action of nerve growth factor. J Immunol 1990; 144:1886-1892. 234. Bischoff SC, Dahinden CA. Effect of nerve growth factor on the release of inflammatory mediators by mature human basophils. Blood 1992; 79: 2662-2669. 235. Garaci E, Caroleo MC, Aloe L, Aqaro S, Piacentini M, Costa N, Amendola A, Micera A, Calio R, Perno CF, Levi-Montalcini R. Nerve growth factor is an autocrine factor essential for the survival of macrophages infected with HIV. Proc Nat Acad Sci USA 1999; 96: 14013-8. 236. Gee AP, Boyle MDP, Munger KL, Lawman MJ, Young M. Nerve growth factor: stimulation of polymorphonuclear leukocyte chemotaxis in vitro. Proc Natl Acad Sci USA 1983; 80: 7215-7218. 237. Boyle MDP, Lawman MJP, Gee AP, Young M. Nerve growth factor: a chemotactic factor for polymorphonuclear leukocytes in vivo. J Immunol 1985; 135: 564-568. 238. Banks BEC, Vernon CA, Warner JA. Nerve growth factor has anti-inflammatory activity in the rat hind paw oedema test. Neurosci Lett 1984; 47: 41-45. 239. Amico-Roxas M, Caruso A, Leone MG, Scifo R, Vanella A, Scapagnini U. Nerve growth factor inhibits some acute experimental inflammations. Arch Int Pharmacodyn Ther 1989; 299: 269-285. 240. Braun A, Appel E, Baruch R, Herz U, Botchkarev V, Paus R, Brodie C, Renz H. Role of nerve growth factor in a mouse model of allergic airway inflammation and asthma. Eur J Immunol 1998; Oct; 28(10): 3240-51. 241. Kerschensteiner M, Gallmeier E, Behrens L, Leal VV, Misgeld T, Klinkert WE, Kolbeck R, Hoppe E, Oropeza-Wekerle RL, Bartke I, Stadelmann C, Lassmann H, Wekerle H, Hohlfeld R. Activated human T cells, B cells, and monocytes produce brain-derived neurotrophic factor in vitro and in inflammatory brain lesions: a neuroprotective role of
39
inflammation? J Exp Med 1999; Mar 1; 189(5): 865-70. 242. Besser M, Wank R. Cutting edge: clonally restricted production of the neurotrophins brain-derived neurotrophic factor and neurotrophin-3 mRNA by human immune cells and Thl/Th2-polarized expression of their receptors. Immunol 1999; Jun 1; 162(11): 6303-6. 243. Simone MD, De Santis S, Vignetti E Papa G, Amadori S, Aloe L. Nerve growth factor: a survey of activity on immune and hematopoietic cells. Hematol Oncol 1999; 1: 1-10. 244. Hohlfeld R, Kerschensteiner M, Stadelmann C, Lassmann H, Wekerle H. The neuroprotective effect of inflammation: implications for the therapy of multiple sclerosis. J Neuroimmunol 2000; Jul 24; 107(2): 161-6. 245. Reinshagen M, Rohm H, Steinkamp M, Lieb K, Geerling I, Von Herbay A, Flamig G, Eysselein VE, Adler G. Protective role of neurotrophins in experimental inflammation of the rat gut. Gastroenterology. 2000; Aug; 119(2): 368-76. 246. Moalem G, Gdalyahu A, Shani Y, Otten U, Lazarovici P, Cohen IR, Schwartz M. Production of Neurotrophins by Activated T Cells: Implications for Neuroprotective Autoimmunity. J Autoimmun 2000; Nov; 15(3): 331-345. 247. Lord GM, Matarese G, Howard JK, Baker RJ, Bloom SR, Lechler RI. Leptin modulates the T-cell immune response and reverses starvation-induced immunosuppression. Nature 1998; Aug 27; 394(6696): 897-901. 248. Bornstein SR, Licinio J, Tauchnitz R, Engelmann L, Negrao AB, Gold P, Chrousos GP. Plasma leptin levels are increased in survivors of acute sepsis: associated loss of diurnal rhythm, in cortisol and leptin secretion. J Clin Endocrinol Metab 1998; Jan; 83(1): 280-3. 249. Wauters M, Considine RV, Van Gaal LF. Human leptin: from an adipocyte hormone to an endocrine mediator. Eur J Endocrinol 2000; Sept; 143(3): 293-311. 250. Hosoi T, Okuma Y, Nomura Y. Expression of leptin receptors and induction of IL-lbeta transcript in glial cells. Biochem Biophys Res Commun 2000; Jun 24; 273(1): 312-5. 251. Faggioni R, Moser A, Feingold KR, Grunfeld C. Reduced leptin levels in starvation increase susceptibility to endotoxic shock. Am J Pathol 2000; May; 1 56(5): 1781-7. 252. Matarese G. Leptin and the immune system: how nutritional status influences the immune response. Eur Cytokine Netw 2000; Mar; 11(1): 7-14. 253. Giovambattista A, Chisari AN, Corro L, Gaillard RC, Spinedi E. Metabolic, neuroendocrine and immune functions in basal conditions and during the acute-phase response to endotoxic shock in undernourished rats. Neuroimmunomodulation 2000; 7(2): 92-8. 254. Okamoto S, Irie Y, Ishikawa I, Kimura K, Masayuki Saito. Central leptin suppresses splenic lymphocyte functions through activation of the corticotropin-releasing hormonesympathetic nervous system. Brain Res 2000; Feb 7; 855(1): 192-7. 255. Chautard T, Spinedi E, Voirol M, Pralong FP, Gaillard RC. Role of glucocorticoids in the response of the hypothalamo-corticotrope, immune and adipose systems to repeated endotoxin administration. Neuroendocrinology 1999; May; 69(5): 360-9. 256. Torpy DJ, Bornstein SR, Chrousos GP. Leptin and interleukin-6 in sepsis. Horm Metab Res 1998; Dec; 30(12): 726-9. 257. Faggioni R, Fantuzzi G, Gabay C, Moser A, Dinarello CA, Feingold KR, Grunfeld C. Leptin deficiency enhances sensitivity to endotoxin-induced lethality. Am J Physiol 1999; Jan; 276(1 Pt 2): R136-42. 258. Takahashi N, Waelput W, Guisez Y. Leptin is an endogenous protective protein against the toxicity exerted by tumor necrosis factor. J Exp Med 1999; Jan 4; 189(1): 207-12. 259. Finck BN, Kelley KW, Dantzer R, Johnson RW. In vivo and in vitro evidence for
40
260.
261. 262. 263. 264.
265.
266.
267.
268.
269.
270.
271.
272. 273. 274.
275.
276. 277.
the involvement of tumor necrosis factor-alpha in the induction of leptin by lipopolysaccharide. Endocrinology 1998; May; 139(5): 2278-83. Loffreda S, Yang SQ, Lin HZ, Karp CL, Brengman ML, Wang DJ, Klein AS, Bulkley GB, Bao C, Noble PW, Lane MD, Diehl AM. Leptin regulates proinflammatory immune responses. FASEB J 1998; Jan; 12(1): 57-65. Fantuzzi G, Faggioni R. Leptin in the regulation of immunity, inflammation, and hematopoiesis. J Leukoc Biol 2000; Oct; 68(4): 437-46. Barach EM, Nowak RM, Lee TG, Tomlanovich MC. Epinephrine for treatment of anaphylactic shock. JAMA. 1984; 251(16): 2118-2122. Nelson HS. Adrenergic therapy of bronchial asthma. J Allergy Clin Immunol 1986; 77(6): 71-785. Livnat S, Felten SY, Carlson DL, Bellinger FL, Felten DL. Involvement of peripheral and central catecholamine systems in neural-immune interactions. J Neuroimmunol 1985; 10: 5-30. Cross RJ, Jackson JC, Brooks WH, Sparks DL, Markesbery WR, Roszman TL. Neuroimmunomodulation: Impairment of humoral immune responsiveness by 6-hydroxydopamine treatment. Immunology 1986; 57(1): 145-152. Berczi I, Nagy E. Neurohormonal control of cytokines during injury. In: Rothwell NJ, Berkenbosch F, editors. Brain Control of Responses to Trauma. Cambridge University Press, 1994: 32-107. Schauenstein K, Rinner I, Felsner P, et al. The role of the adrenergic/cholinergic balance in the immune-neuroendocrine circuit. In: Berczi I, Szelenyi J, editors. Advances in psychoneuroimmunology. New York, Plenum Press, 1994: 349-356. Nagatomi R, Kaifu T, Okutsu M, Zhang X, Kanemi O, Ohmori H. Modulation of the immune system by the autonomic nervous system and its implication in immunological changes after training. Exerc Immunol Rev 2000; 6: 54-74. Szelenyi J, Kiss JP, Vizi ES. Differential involvement of sympathetic nervous system and immune system in the modulation of TNF-alpha production by alpha2- and betaadrenoceptors in mice. J Neuroimmunol 2000; Feb 1; 103(1): 34-40. Bergmann M, Gornikiewicz A, Sautner T, Waldmann E, Weber T, Mittlbock M, Roth E, Fugger R. Attenuation of catecholamine-induced immunosuppression in whole blood from patients with sepsis. Shock 1999 Dec; 12(6): 421-7. Schauenstein K, Rinner I, Felsner P, et al. The role of the adrenergic/cholinergic balance in the immune-neuroendocrine circuit. In: Berczi I, Szelenyi J, editors. Advances in psychoneuroimmunology. New York, Plenum Press, 1994: 349-356. Druce HM, Wright RH, Kossoff D, Kaliner MA. Cholinergic nasal hyperreactivity in atopic subjects. J Allergy Clin Immunol 1985; 76: 445-452. Coffey RG, Hadden JW. Neurotransmitters, hormones, and cyclic nucleotides in lymphocyte regulation. Fed Proc 1985; 44(1): 112-117. Szelenyi J, Ha NTH. Expression of cholinergic structure on lymphocytes. In: Berczi I, Szelenyi J, editors. Advances in psychoneuroimmunology. New York, Plenum Press, 1994:271-277. Payan DG, McGillis JP, Renold FK, Mitsuhashi M, Goetzl EJ. The immunomodulating properties of neuropeptides. In: Berczi I, Kovacs K, editors. Hormones and immunity. Lancaster, England, MTP Press, 1987. Foreman JC. Neuropeptides and the pathogenesis of allergy. Allergy 1987; 42(1): 1-11. Shanahan F, Lee TDG, B ienenstock J, Befus AD. Mast cell heterogeneity: Effect of anti-allergic compounds on neuropeptide-induced histamine release. Int Arch Allergy
41
Appl Immunol 1986; 80(4): 424-426. 278. Jancso G, Obal F Jr, Toth-Kasa I, Katona M, Husz S. The modulation of cutaneous inflammatory reactions by peptide-containing sensory nerves. Int J Tissue React 1985; 7(6): 449-457. 279. Serra M, Bazzoni F, Della Bianca V, Greskowiak M, Rossi F. Activation of human neutrophils by substance P. Effect on oxidative metabolism, exocytosis, cytosolic Ca 2+ concentration and inositol phosphate formation. J Immunol 1988; 141:2118-2124. 280. Lotz M, Carson DA, Vaughan JH. Substance P activation of rheumatoid synoviocytes: Neural pathway in pathogenesis of arthritis. Science 1987; 235(4791): 893-895. 281. Hartung HP, Heininger K, Schafer B, Toyka KV. Substance P and astrocytes: stimulation of the cyclooxygenase pathway of arachidonic acid metabolism. Fed Am Soc Exp B iol J 1988; 2: 48-51. 282. Damonneville M, Monte D, Auriault C, Capron A. The neuropeptide substance-P stimulates the effector functions of platelets. Clin Exp Immunol 1990; 81: 346-351. 283. Rameshar P, Ganea D, Gascon P. Induction of IL-3 and granulocyte-macrophage colonystimulating factor by substance P in bone marrow cells is partially mediated through the release of IL-1 and IL-6. J Immunol 1994; 152: 4044-4054. 284. Mariott I, Bost KL. IL-4 and IFN-gamma up-regulate substance P receptor expression in murine peritoneal macrophages. J Immunol 2000; Jul 1; 165(1): 182-91. 285. Blum AM, Metwali A, Kim-Miller M, Li J, Qadir K, Elliott DE, Lu B, Fabry Z, Gerard N, Weinstock JV. The substance P receptor is necessary for a normal granulomatous response in murine schistosomiasis mansoni. J Immunol 1999; May 15; 162(10): 6080-5. 286. Bienenstock J. A model of neuroimmune communication: Mast cells and nerves. In: Neuroimmune Biology: New foundation of Biology. Berczi I, Gorczynski R, editors. Amsterdam, Elsevier (this volume): 2001; 195-202. 287. Mullins MW, Ciallella J, Rangnekar V, McGillis JP. Characterization of a calcitonin gene-related peptide (CGRP) receptor on mouse bone marrow cells. Regul Pept 1993; 49: 65-72. 288. Nong YH, Titus RG, Ribeiro JMC, Remold HG. Peptides encoded by the calcitonin gene inhibit macrophage function. J Immunol 1989; 145: 45-49. 289. Wang X, Fiscus RR, Tang Z, et al. CGRP in the serum of endotoxin-treated rats suppresses lymphoproliferation. Brain Behav Immun 1994; 8: 282-292. 290. Hosoi J, Murphy GF, Egan CL, et al. Regulation of Langerhans cell function by nerves containing calcitonin gene-related peptide. Nature 1993; 363: 159-63. 291. Asahina A, Hosoi J, Beissert S, Stratigos A, Granstein RD. Inhibition of the induction of delayed-type and contact hypersensitivity by calcitonin gene-related peptide. J Immunol 1995; 154: 3056-61. 292. Millet I, Phillips RJ, Sherwin RS, Ghosh S, Voll RE, Flavell RA, Vignery A, Rincon M. Inhibition of NF-kappaB activity and enhancement of apoptosis by the neuropeptide calcitonin gene-related peptide. J Biol Chem 2000; May 19; 275(20): 15114-21. 293. Wang X, Xing L, Xing Y, Tang Y, Han C. Identification and characterization of immunoreactive calcitonin gene-related peptide from lymphocytes of the rat. J Neuroimmunol 1999; Feb 1; 94(1-2): 95-102. 294. Streilein JW, Alard P, Niizeki H. A new concept of skin-associated lymphoid tissue (SALT): UVB light impaired cutaneous immunity reveals a prominent role for cutaneous nerves. Keio J Med 1999; Mar; 48(1): 22-7. 295. Foris G, Gyimesi E, Komaromi I. The mechanism of antibody-dependent cellular cytotoxicity stimulation by somatostatin in rat peritoneal macrophages. Cell Immunol
42
1985; 90(1): 217-225. 296. Muscettola M, Grasso G. Somatostatin and vasoactive intestinal peptide reduce interferon gamma production by human peripheral blood mononuclear cells. Immunobiology. 1990; 180: 419-430. 297. Church MK, E1-Lati S, Caulfield JP. Neuropeptide-induced secretion from human skin mast cells. Int Arch Allergy Appl Immunol 1991; 94:310-318. 298. Bokum AM, Hofland LJ, van Hagen PM. Somatostatin and somatostatin receptors in the immune system: a review. Eur Cytokine Netw 2000; Jun; 11(2): 161-76. 299. Szolcsanyi J, Helyes Z, Oroszi G, Nemeth J, Pinter E. Release of somatostatin and its role in the mediation of the anti-inflammatory effect induced by antidromic stimulation of sensory fibres of rat sciatic nerve. Br J Pharmacol 1998; Mar;123(5): 936-42. 300. Ottaway CA. In vitro alterations of receptors for vasoactive intestinal peptide changes the in vivo localization of mouse T cells. J Exp Med 1984; 160(4): 1054-1069. 301. O'Dorisio MS. Biochemical characteristics of receptors for vasoactive intestinal polypeptide in nervous endocrine and immune systems. Fed Proc 1987; 46(1): 192-195. 302. Kimata H, Fujimoto M. Vasoactive intestinal peptide specifically induces human IgA1 and IgA2 production. Eur J Immunol 1994; 24: 2262-2265. 303. Xin Z, Tang H, Ganea D. Vasoactive intestinal peptide inhibits interleukin (IL)-2 and IL-4 production in murine thymocytes activated via the TCR/CD3 complex. J Neuroimmunol 1994; 54: 59-68. 304. Boirivant M, Fais S, Annibale B, Agostini D, Delle Fave G, Pallone F. Vasoactive intestinal polypeptide modulates the in vitro immunoglobulin A production by intestinal lamina propria lymphocytes. Gastroenterology. 1994; 106: 576-82. 305. Delgado M, Ganea D. Inhibition of IFN-gamma-induced janus kinase-l-STAT1 activation in macrophages by vasoactive intestinal peptide and pituitary adenylate cyclase-activating polypeptide. J Immunol 2000; Sep 15; 165(6): 3051-7. 306. Sun W, Tadmori I, Yang L, Delgado M, Ganea D. Vasoactive intestinal peptide (VIP) inhibits TGF-betal production in murine macrophages. J Neuroimmunol 2000; Jul 10; 107(1): 88-99. 307. Xin Z, Sriram S, Vasoactive intestinal peptide inhibits IL-12 and nitric oxide production in murine macrophages. J Neuroimmunol 1998; Aug 14; 89(1-2): 206-12. 308. Kawamura N, Tamura H, Obana S, Wenner M, Ishikawa T, Nakata A, Yamamoto H. Differential effects of neuropeptides on cytokine production by mouse helper T cell subsets. Neuroimmunomodulation 1998; Jan-Apr; 5(1-2): 9-15. 309. Martinez C, Delgado M, Pozo D, Leceta J, Calvo JR, Ganea D, Gomariz RP. Vasoactive intestinal peptide and pituitary adenylate cyclase-activating polypeptide modulate endotoxin-induced IL-6 production by murine peritoneal macrophages. J Leukoc B iol 1998; May; 63(5): 591-601. 310. Tang H, Sun L, Xin Z, Ganea D. Down-regulation of cytokine expression in murine lymphocytes by PACAP and VIP. Ann N Y Acad Sci 1996; Dec 26; 805: 768-78. 311. Martinez C, Delgado M, Pozo D, Leceta J, Calvo JR, Ganea D, Gomariz RP. VIP and PACAP enhance IL-6 release and mRNA levels in resting peritoneal macrophages: in vitro and in vivo studies. J Neuroimmunol 1998; May 15; 85(2): 155-67. 312. Delgado M, Ganea D. Vasoactive intestinal peptide and pituitary adenylate cyclaseactivating polypeptide inhibit antigen-induced apoptosis of mature T lymphocytes by inhibiting Fas ligand expression. J Immunol 2000; Feb 1; 164(3): 1200-10. 313. Taylor AW, Streilein JW, Cousins SW. Immunoreactive vasoactive intestinal peptide contributes to the immunosuppresssive activity of normal aqueous humor. J Immunol
43
1994; 153: 1080-1086. 314. Ferguson TA, Fletcher S, Herndon J, Griffith TS. Neuropeptides modulate immune deviation induced via the anterior chamber of the eye. J Immunol 1995; 155: 1746-56. 315. Lundeberg L, Mutt V, Nordlind K. Inhibitory effect of vasoactive intestinal peptide on the challenge phase of allergic contact dermatitis in humans. Acta Derm Venereol 1999; May; 79(3): 178-82. 316. Frontiers FB, editor. The Cytokine Network. In: Hames B D, Glover DM, series editors. Oxford in Molecular Biology Series University Press, 2000. 317. Grignani G. Maiolo, Cytokines and hemostasis. A Haematologica 2000; Sept; 85(9): 967-972. 318. Warren JL. Type I cytokines, interferons and their receptors. In: Paul WE, editor. Fundamental Immunology, 4 th edition. Lippincott-Raven, 1999: 741-74. 319. Krauker T, Vilcek J, Oppenheim JJ. Proinflammatory cytokines. In: Paul WE, editor. Fundamental Immunology4 th edition. Lippincott-Raven, 1999:775-811. 320. Mathison RT, Davison JS, Befus AD. Neuroendocrine regulation of inflammation and tissue repair by submandibular gland factors. Immunology Today 1994; 14: 527-532. 321. Berczi I, Chow DA, Sabbadini ER. Neuroimmunoregulation and natural immunity. Domestic Anim Endocrin 1998; 15: 273-281. 322. Sabbadini ER and Berczi I. Immunoregulation by the salivary glands. Biomed Rev 1998; 9: 79-91. 323. Befus D et al. Immunomodulation by the submandibular gland. In: Berczi I, Gorczynski R, editors. Neuroimmmune Biology: New Foundation of Biology. Amsterdam, Elsevier (this volume): 203-224. 324. Sabbadini et al. Immunoregulation by glandular kallikrein. In: Berczi I, Gorczynski R, editors. Neuroimmmune Biology: New Foundation of Biology. Amsterdam, Elsevier (this volume): 225-235. 325. Brandtzaeg P, Farstad IN, Haraldsen G, Jahnsen FL. Cellular and molecular mechanisms for induction of mucosal immunity. Dev Biol Stand 1998; 92: 93-108. 326. Chen YH, Weiner HL. Dose-dependent activation and deletion of antigen-specific T cells following oral tolerance. Ann N Y Acad Sci 1996; Feb 13; 778: 111-21. 327. Davison J. The Salivary Gland Peptides: Their Role in Anaphylaxis and Lipopolysaccharide (LPS)-Induced Inflammation. In: Berczi I, Gorczynski R, editors. Neuroimmmune Biology: New Foundation of Biology. Ireland, Elsevier 2001; 307-311. 328. McDermott MR, Clark DA, Bienenstock J. Evidence for a common mucosal immunologic system. II. Influence of the estrous cycle on B immunoblast migration into genital and intestinal tissues. J Immunol 1980; Jun; 124(6): 2536-9. 329. Nagy E, Berczi I. Pituitary dependence of bone marrow function. Brit J Haematol 1989; 71: 457-462. 330. Berczi I, Nagy E. Effects of hypophysectomy on immune function. In: Ader R, Felten DL, Cohen N, editors. Psychoneuroimmunology II. Academic Press 1991: 339-375. 331. Bernard C. Lecons sur les phenomenes de la vie communs aux animaux et aux vegetaux. Balliere, Paris 1878. 332. Berczi I. Neurohormonal host defence in endotoxin shock Ann NY Acad Sci 840: 787-802, 1998. 333. Berczi I, Bertok L, Chow D. Natural immunity and neuroimmune host defence. Ann NY Acad Sci (in press). 334. Medzhitov R, Janeway C Jr. Innate immune recognition: mechanisms and pathways. Immunol Rev 2000; Feb;173: 89-97.
44
335. Berczi I. The stress concept and neuroimmunoregulation in modern biology. Ann NY Acad Sci 1998; 851: 3-12. 336. Berczi I. Neuroendocrine defence in endotoxin shock. (A review). Acta Microbiol Hun 1993; 40(4); 265-302. 337. Berczi I, Nagy E. Neurohormonal control of cytokines during injury. In: Rothwell NJ, Berkenbosch F, editors. Brain Control of Responses to Trauma. Cambridge University Press, 1994: 32-107. 338. Berczi I. Neuroendocrine response to endotoxin. Ann NY Acad Sci 1998; 851: 411-415. 339. Haeryfar SMM, Berczi I. The thymus and the acute phase response. Mol Cell Biol (in press). 340. Pittman QJ, Chen X, Mouihate A, Martin S. Vasopressin-induced atipyresis. Sex- and experience-dependent febrile response. Ann NY Acad Sci 1998; 856: 53-61. 341. McEwen BS. Allostasis, allostatic load, and the aging nervous system: role of excitatory amino acids and excitotoxicity. Neurochem Res 2000; Oct; 25(9-10): 1219-31. 342. Anisman H, Baines MG, Berczi I, Bernstein CN, Blennerhassett MG, Gorczynski RM, Greenberg AH, Kisil FT, Mathison RD, Nagy E, Nance DM, Perdue MH, Pomerantz DK, Sabbadini ER, Stanisz A, Warrington RJ. Neuroimmune mechanisms in health and disease: 1. Health. Can Med Assoc J 1996; 155(7): 867-874. 343. Gorczinsky R. Understanding classical conditioning of immune responses. In: Berczi I, Gorczynski R, editors. Neuroimmune Biology: New Foundation of Biology. Amsterdam, Elsevier 2001 ; 237-254. 344. Pomerantz D. Interactions Between the Immune and the Reproductive Systems. In: Berczi I, Gorczynski R, editors. Neuroimmune Biology: New Foundation of Biology. Amsterdam, Elsevier 2001; 237-254. 345. Luheshi G. Leptin and Cytokines: Actions and Interactions in Fever and Appetite Control. In: Berczi I, Gorczynski R, editors. Neuroimmune Biology: New Foundation of Biology. Amsterdam, Elsevier 2001; 283-292. 346. Baines M. Olfactory Stimuli and Allo-Recognition. In: Berczi I, Gorczynski R, editors. Neuroimmune Biology: New Foundation of Biology. Amsterdam, Elsevier 2001; 313-329. 347. Berczi I, Nagy E, Asa SL, Kovacs K. The influence of pituitary hormones on adjuvant arthritis. Arthritis Rheum. 1984; 27: 682-688. 348. Nagy E, Friesen HG, Sehon AH, Berczi I. Immunomodulation in rats by transplantable anterior pituitary tumors. Endocrinology 1985; 116:1117-1122. 349. Nagy E, Berczi I. Immunomodulation by tamoxifen and pergolide. Immunopharmacology 1986; 12: 145-153. 350. I. Berczi I, Cosby H, Hunter T, Baragar F, McNeilly AS, Friesen HG. Decreased bioactivity of circulating prolactin in patients with rheumatoid arthritis. Brit J Rheum 1987; 26: 433-436. 351. Nagy E, Chalmers IM, Baragar FD, Friesen HG, Berczi I. Prolactin deficiency in rheumatoid arthritis. J Rheumatol 1991 ; 18:1662-1668. 352. Ramachandra RN, Sehon AH, Berczi I. Neuro-hormonal host defence in endotoxin shock. Brain Behav Immun 1992; 6: 157-169. 353. Baral E, Nagy E, Berczi I. Target cells are sensitized for cytotoxic T lymphocyte mediated destruction by estradiol and tamoxifen. Int J Cancer 1994; 58: 64-68. 354. Baral E, Nagy E, Berczi I. Modulation of natural killer cell mediated cytotoxicity by tamoxifen and estradiol. Cancer 1995; 75: 591-599. 355. Anisman H., Baines MG, Berczi, Bernstein CN, Blennerhassett MG, Gorczynski RM, Greenberg AH, Kisil FT, Mathison RD, Nagy E, Nance DM, Perdue MH, Pomerantz
45
356.
357. 358. 359.
DK, Sabbadini ER, Stanisz A., Warrington RJ. Neuroimmune mechanisms in health and disease: 2. Disease. Can Med Assoc J 1996; 155(8): 1075-1082. McCann SM, Lipton JM, Sternberg EM, Chrousos GP, Gold PW, Smith CC, editors. Neuroimmunomodulation: Molecular aspects, integrative systems, and clinical advances. Ann NY Acad Sci 1998; 840. Csermely P, editor. Stress of life, from molecules to man. Ann NY Acad Sci 851:1998. Dickstein J, Moldofsky H. Sleep, cytokines and immune function. Sleep Medicine Reviews 1999; 3(3) (in press). S Rivest, S Lacroix, L Vallieres, S Nadeau, J Zhang, N Laflamme. How the blood talks to the brain parenchyma and the paraventricular nucleus of the hypothalamus during systemic inflammatory and infectious stimuli. Proc Soc Exp Biol Med 2000; 223: 22-38.
This Page Intentionally Left Blank
New Foundation of Biology Edited by I. Berczi and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
47
Studies on the Hypothalamic Regulation of Histamine Synthesis History and some current information ANDOR SZENTIVANYI, ISTVAN BERCZI, DENYSE PITAK and ALLAN GOLDMAN Department of Internal Medicine, University of South Florida, Box 9, 12901 Bruce B. Downs Blvd., Tampa, FL. 33612-4799, USA.
ABSTRACT As described earlier, L-histidine decarboxylase (HDC) activity in bone marrow is markedly increased on incubation with interleukin-3 (IL-3) or granulocyte macrophage colony stimulating factor (GM-CSF). This is due to the induction of the de novo synthesis of HDC in hemopoiefic progenitor cells. Chemical axotomy in adult mice, chemical sympathectomy in neonatal mice or catecholamine depletion in adult rats by 6-hydroxydopamine inhibited the IL-3 or GM-CSF induced histamine synthesis. Stereotaxic lesions of the pre-optic and anterior hypothalamic area (parasympathetic representation) or electric stimulation of the posterior hypothalamus, enchanced the same effect. Likewise, chemical anterior lesions produced by N-methyl-DL-aspartate, or surgical antero-lateral deafferentation show the same result. Conversely, posterior hypothalamic lesions (sympathetic representation) or electric stimulation of the anterior hypothalamus have no effect on HDC synthesis. Thus the de novo synthesis of HDC induced by IL-3 and GM-CSF required [3-adrenergic activation of progenitor hemopoietic cells that sensitizes these cells to the cytokines. The [5-adrenergic input can be completely replaced in the induction of HDC synthesis by in vitro addition of Interleukin-1 ~t.
1.
HISTORY
The role of the hypothalamic influences in the induction and expression of immunologic inflammation, immunity and hypersensitivity was first discovered in the fall of 1951 at the University of Debrecen School of Medicine in Hungary. The rationale behind the decision for a systematic exploration of the hypothalamus was as follows. Historically, the interpretation of the symptomatology and the underlying reaction sequence of human asthma was patterned after those of the anaphylactic guinea pig. However, the range of atopic responsiveness in asthma included a variety of stimuli that are non-immunologic in nature. Foremost among these is a broad range of pharmacologically active mediators that today could be considered as the chemical organizers of central and peripheral autonomic regulation. Therefore it was believed that anaphylaxis could not be used as a model for the investigation of the constitutional basis of atopy in asthma. It was postulated that such a model, if it was to be meaningful, must be able to imitate both the immunologic and autonomic abnormalities of the disease. Consequently, hypothalamically imbalanced anaphylactic animals were used. These were
48
produced by the electrolytic lesions, and conversely, by the electrical stimulation of various nuclear groupings in the hypothalamus through permanently implanted depth electrodes placed stereotaxically into the hypothalamus.
3.
INTRODUCTION
L-Histidine decarboxylase (HDC) activity is detectable in normal bone marrow cell lysates. Michel Dy and associates showed in a series of experiments (1981-1987) that this level of HDC activity is markedly increased following incubation of the cells with interleukin-3 (IL-3) or granulocyte macrophage colony stimulating factor (GM-CSF). The cell mass that predominantly responded to these cytokines with enchanced HDC activity were those containing most of the hematopoietic progenitor cell types such as colony forming cells and mast cell precursors. In keeping with our group' s long-time interest and studies of the role of histamine in mammalian physiology, and the many aspects of the inverse, reciprocal histamine-catecholamine counterregulatory interplay in health and disease, we have examined the effects of (1) chemical sympathetic ablation by 6-hydroxydopamine hydrobromide (6-OHDA) and (2) hypothalamic lesions on the de n o v o synthesis of histidine decarboxylase in hematopoietic progenitor cells induced by IL-3 or GM-CSF in mice and rats. The study that follows shows the role of hypothalmic influences on histamine synthesis.
4.
MATERIALS AND METHODS
4.1.
Animals and cell cultures
Bone marrow cells (BMC) were obtained from femurs of female C57BL/6 mice. They were incubated at a final concentration of 2.5 x 106/ml in minimum essential medium with Earle's salts supplemented with 1% sodium pyruvate 100 x, 1% L-glutamine, 100 IU/ml penicillin and 100 pg/ml streptomycin. No serum was added. Murine recombinant GM-CSF and IL-3 were incubated with BMC for 24 hours at 37 ~ in a humidified atmosphere of 95% air and 5% CO 2. The cytokines were added to the BMC suspension at a concentration of 50 U HCSA/ml. In most experiments, bone marrow cells were separated on a discontinuous Ficoll gradient as described [ 1, 2]. Briefly, the gradient was prepared from Ficoll 400 at a concentration of 10., 14.6, 16.1, 17.7, 19.2 and 23% (w/w) in 0.1 M sodium phosphate buffer (pH 7.4). Bone marrow cells from 10 mice were layered on top of the gradient (1.2 ml/layer) and then centrifuged for 30 min at 23, 500 x G at 4~ Layer O was defined as the interphase between culture medium and 10% Ficoll and subsequent interphases were numbered sequentially. Cells from the different layers were then collected, washed, and incubated with rGM-CSF or rlL-3 (Genzyme, Canbridge, MA) at a final concentration of 106 cells/ml. After incubation, cell suspensions were centrifuged, supernatants were collected for histamine determination and cell pellets were stored at-20 ~ for HDC assay. 4.2.
Histamine determination and HDC assay
Histamine concentrations in bone marrow supernatants or in cell lystates were determined by the fluorometric assay of [3] with some modifications as described by [4, 5]. HDC activity was measured by the radiochromatographic assay of [6] with minor modifications [1]. Briefly, cell
49
pellets (2.5 x 1 0 6 cells) were frozen and thawed, resuspended in 50 mM ice-cold phosphate buffer (pH 7.4) containing a final concentration of 10 p M pyridoxal 5'-phosphate, 100 p M of unlabeled histamine, and 0.1 pM L-[3H] histidine. Incubations were performed under conditions of initial velocity measurement and were stopped after 60 min by the addition of perchloric acid (0.4 N, final concentration) containing 0.3 M unlabeled histidine to minimize possible nonspecific decarboxylation of remaining L-[3H]-histamine. After centrifugation, the synthesized L-[3H]-histamine was separated from L-[3H]-histidine by ion exchange chromatography on Amberlite CG-50 columns in the early studies, and later on Bio-Rex 70 resin (Bio-ad Laboratories, Rockville, N.Y.) and Aminex HPX-72-S column (Bio-Rad Laboratories, Rockville, N.Y.). Blanks were obtained from parallel incubations performed in the presence of 10 -5 MS-c~-fluoromethylhistidine hydrochloride hemihydrate (10 -5 M) an irreversible HDC inhibitor. 4.3.
Peripheral and central (hypothalamic and extrahypothalamic) chemical and surgical interventions
For a detailed review of surgical and chemical procedures, see [7]. 4.4.
Axotomy
To produce peripheral axotomy, 8 to 10 weeks old A/J mice were injected i.p. daily for 10 days with 0.2 ml 6-OHDA (6-hydroxydopamine hydrobromide, a neurotoxin specific for adrenergic nerve endings) at a dose of 100 mg/kg body weight. 6-OHDA was dissolved in 0.9% NaC1 containing 0.1 mg/ml ascorbic acid. Experiments were performed 1 week after the last 6-OHDA treatment. 4.5.
Sympathectomy
Sympathectomy was achieved by injecting newborn CBA/J x A/J F 1 mice daily with the dose and preparation of 6-OHDA mentioned above. Parental mice were bred in the laboratory. The F1 strain was used because the maternal care provided by CBA/J female mice permitted increased litter sizes. The volumes injected started at 0.02 ml and increased proportionately with body size up to 0.1 ml making the appropriate adjustments in 6-OHDA concentration to keep the dose constant. Experiments were performed 7 to 10 weeks after the last 6-OHDA treatment. 4.6.
Anterolateral deafferentation of hypotalamic and extrahypothalamic nuclei
This involves bilateral knife cuts anterior and lateral to the nuclear grouping involved. This is carried out by the stereotaxic placement of an extendable Halasz knife lowered 5.5 mm into the brain through a burr hole made in the median saggital stuture. After lowering the 20 g knife carrier, the sharpened blade is extruded in the midline until its tip comes to rest at rostral border of the targeted nucleus. The knife is turned (radius-l.5 mm) 180 ~ right and left 10-12 times with the blade tip coming through the periosteum of the sphenoid to sever all anterior nervous connections. The knife is then turned 90 ~ to the left and the entire carrier is moved 2.0 mm posterior and then anterior with the knife blade positioned so that the left lateral nucleus connections are severed. The blade is then returned to the midline and turned 90 ~ to the right so that a second 2.0 mm anterior-posterior excursion will result in the knife blade severing fight lateral nervous connections. The blade is then returned to the midline,
50
retracted into the carrier, and removed from the brain. The completeness of the deafferentation is verified post-mortem by histology. 4.7.
Sham-deafferentation controls
Control animals in this study received only midline (5.5 mm) descent of the carrier and the 2.0 mm anterior-posterior excursion. 4.8.
Sham bilateral NMA lesions
Lowering of syringe barrels as above and injection of CSF vehicle. 4.9.
Bilateral hypothalamic and extrahypothalamic brain lesions in adult rats
Each animal receives two lesions, one 0.5 mm anterior to the other, bilaterally in the areas of various hypothalamic and extrahypothalamic nuclei through stereotaxically placed and permanently implanted depth electrodes (tungsten electrodes insulated except for 0.2 mm at the tip using 1.5 mA DC for 15 seconds at each coordinate; [8]. Anchoring of electrodes in place with dental cement and machine screws placed in burr holes made in the skull. Permanent sealing of the skull's trephine opening by plexiglass. Post-mortem verification of lesions was done by histology. 4.10.
Sham bilateral electrolytic lesions
Stereotaxic placement of permanently implanted depth electrodes as above but without passing current. Post-mortem histologic localization of electrode position. 4.11.
Central catecholamine depletion by intrathecal administration of 6-OHDA
Catecholamine depletion of central adrenergic neurons through intrathecally administered 6-OHDA to adult rats involving 1 single application in a total volume of 0.1 ml with the proportionate dose and 6-OHDA preparation as above. Studies were performed 1 week after treatment. Catecholamine depletion verified by the fluorometric-measurement of catecholamines according to [9] in brain following sacrifice. 4.12.
Sham central catecholamine depletion
Intrathecal administration of vehicle without 6-OHDA and with post-mortem catecholamine determination in brain. 4.13.
Bilateral hypothalamic and extrahypothalamic chemical brain lesions produces by N-methyl-DL-aspartate (NMA)
These lesions aimed at various hypothalamic and extrahypothalamic nuclei are made by intracranial injections of axon-sparing NMA (0.3 M) diluted in artificial Cerebrospinal fluid (CSF) adjusted to pH 7.4. CSF consists (in M) of the following: 0.13 NaCL, 0.025 NaHCO 3, 0.0005 Na2HPO 4, 0.0029 KCL, 0.0008 MgCL, and 0.00013 CaCL 2. Two 1 pl Hamilton syringes are mounted with barrel tips 1.0 mm apart in a stereotaxic electrode holder to
51
enable an injection of 0.15 /~1 o v e r 15 m i n u t e s f r o m both anterior and p o s t e r i o r syringes. B o t h syringes r e m a i n in place for 5 m i n u t e s after injections to m i n i m i z e b a c k f l o w of N M A t h r o u g h the n e e d l e track. 4.14.
Bilateral electric s t i m u l a t i o n of h y p o t h a l a m i c and e x t r a h y p o t h a l a m i c nuclei
Sterotaxic p l a c e m e n t of a single p l a t i n u m (90% p l a t i n um, 10% iridium) e l e c t r o d e bilaterally into neclei, and a n c h o r i n g t h e m in place with dental c e m e n t and m a c h i n e screws p l a c e d in burr holes m a d e in the skull with p e r m a n e n t p l e x i g l a s s sealing of the trephine opening. S t i m u l a t i o n p e r i o d is 30 m i n with c o n t i n u o u s l y m o n i t o r e d current p a r a m e t e r s (50 H Z , 200 ~tA, 0.5 m s e c width, 30 sec on-off, tip-cathode n e g a t i v e [10, 11]. S t i m u l a t i o n sessions c o n t i n u e e v e r y 4 days for a period of 12 days. P o s t - m o r t e m localization of e l e c t r o d e posi t i on by histology. 4.15.
S h a m bilateral electric s t i m u l a t i o n of h y p o t h a l a m i c and e x t r a h y p o t h a l a m i c neclei
E l e c t r o d e p l a c e m e n t as a b o v e w i t h o u t current passage. P o s t - m o r t e m l ocal i zat i on of e l e c t r o d e position.
Table I
Histamine production and histidine decarboxylase activity in discontinuous ficoll gradient layers of bone marrow cells derived from 6-OHDA axotomized, sympathectomized of centrally catecholaminedepleted mice in response to recombinant GM-CSF or recombinant IL-3.
6-OHDA Treatment of Animals
BMC Treatment 1,2
Histamine Production 3 (ng/106 cells)
Histidine Decarboxylase Activity3 (dpm/hour/106 cells)
None (adult mice)
rGM-CSF2 rlL-3
340 + 283 530 + 47
4, 250 + 370 6, 600 + 490
None (newborn mice)
rGM-CSF rIL-3
772 + 63 1,133 + 84
5,900 + 395 13, 120 _+455
Axotomy (adult mice)
rGM-CSF rlL-3
34 __6 11 _+3
395 __45 325 _+20
Sympathectomy (newborn mice)
rGM-CSF rIL-3
11 + 3 14 + 2
325 + 20 905 + 75
1 Bone marrow cells were incubated with rGM-CSF and rIL-3 for 24 hours. 2 Added to BMC suspensions at a concentration of 50 U HCSA/ml, one unit representing the amount of cytokine that produces a 100% increase in histamine production. 3 Data are expressed as means + SEM of four separate determinations.
52
Table II
The effect of bilateral electric stimulation of hypothalamic and extrahypothalamic nuclear groupings on histamine production and histidine decarboxylase activity in discontinous ficoll gradient layers of bone marrow cells of rats in response to recombinant GM-CSF or recombinant IL-3.
Site of stimulation
Sham stimulation
BMC
Histamine
treatmentl
production 2
activity 2
(ng/10 6 cells) 2
(dpm/hour/10 6 cells)
274 • 30
2, 211 • 670
None
Histdine decarboxylase
untreated 3
Sham stimulation
rGM-CSF 4
680 • 78
5, 102 • 670
treated 5
rlL-34
998 • 121
6, 320 _+ 614
Pre-Optic
rGM-CSF
504 • 69
4, 947 +_ 511
rlL-3
598 • 71
4, 588 +_ 534
Anterior
rGM-CSF
447 • 59
5,475 • 498
Hypothalamic
rlL-3
512 • 62
5, 136 • 602
Posterior
rGM-CSF
4, 348 • 461
41,637 • 637
Hypothalamic
rlL-3
5 , 9 6 3 • 574
66, 974 • 851
Hippocampal
rGM-CSF
496 • 57
5,301 • 683
rlL-3
422 • 49
5,732 • 598
Amygdaloid
rGM-CSF
591 + 67
5,064 ___610
rlL-3
622 • 71
5,934 • 729
Bone marrow cell were incubated with rGM-CSF and rlL-3 for 24 hours. 2 Data are expressed as means + SEM of four separate determinations. 3 Electrode placement without current passage and no treatment with cytokines. 4 Added to BMC suspension at a concentration of 50 U HCSA/ml, one unit representing the amount of cytokine that produces a 100% increase in histamine production. 5 Electrode placement without current passage and treatment with cytokines.
53
Table III
The effect of bilateral hypothalamic and extrahypothalamic brain lesions produced by central and peripheral 6 - O H D A , N M A , or anterolateral deafferentation on histamine synthesis and histidine decarboxylase activities in discontinuous gradient layers of bone m a r r o w derived from adult rats.
Bone marrow intervention
None
6-OHDA
Sham 6-OHDA
NMA
Sham NMA
B M C treatment 1,2
Histamine synthesis 3
(ng/106 cells)
(dpm/hour/10 6 cells)
rGM-CSF
284 + 313
6, 100 + 415
rIL-3
452 + 43
9, 855 _+ 620
rGM-CSF
28 + 4
351 + 602
rIL-3
37 + 5
612 +_ 5
rGM-CSF
219 + 34
6, 422 + 602
rIL-3
481 + 51
9, 336 + 598
Histidine decarboxylase 3
rGM-CSF
30 + 7
294 _+ 33
rIL-3
39 + 8
487 +_ 49
rGM-CSF
251 + 29
6, 054 + 721
rIL-3
372 +_ 48
8 , 9 8 5 _ 832
Anterolateral
rGM-CSF
32 + 6
285 + 41
Deafferentation +
rIL-3
35 + 8
418 + 55
Total S y m p a t h e c t o m y
S h a m Anterolateral
rGM-CSF
348 + 92
6, 131 + 785
Deafferentation
rIL-3
403 + 84
9, 020 _+ 816
No S y m p a t h e c t o m y
6-OHDA
r G M - C S F + EPI
475 + 54
6, 593 + 671
EPI In g i t F o 4
rIL-3 + EPI
532 + 68
9, 122 _+ 784
6-OHDA
r G M - C S F + rIL -1 c~
483 + 51
6, 969 + 549
IL-lc~ 5
rIL-3 + rIL-lc~
498 + 63
9, 073 + 708
1 B M C were incubated with cytokines for 24 hours. 2 A d d e d to B M C suspension at a concentration of 50 U H C S A / m l , one unit representing the a m o u n t of cytokine that produces a 100% increase in histamine production. 3 M e a n s + SEM of 4 separate determinations. 4 100 n M epinephrine added to B M C suspension incubated with r G M - C S F or rIL-3. 5 A m o u n t corresponds to the ECs0 of the most established other biochemical effect of the cytokine.
54
Table IV
The effect of bilateral hypothalamic and other brain lesions on histamine production and histidine decarboxylase activity in discontinuous ficoll gradient layers of bone marrow cells of rats in response to recombinant G M - C S F or recombinant IL-3.
Localization of lesions
BMC
Histamine
treatment I
production 2
Histidine decarboxylase activity 2
(ng/106 cell) 2
(dpm/hour/106 cells)
Sham lesion
None
230 +_ 19
2, 624 _+ 282
untreated 3
None
308 + 27
2, 145 + 230
Sham lesion
rGM-CSF 4
850 _+ 69
6, 492 + 423
treated 5
rIL-34
1, 114 + 95
7, 108 + 471
Pre-Optic
rGM-CSF
4, 372 + 103
46, 892 +_ 621
rIL-3
6, 954 + 141
7 l, 324 + 984
Anterior
rGM-CSF
3, 951 _ 174
49, 514 + 732
Hypothalamic
rIL-3
6, 678 -4-_165
79, 898 + 794
Posterior
rGM-CSF
163 +_ 20
1,543 + 88
Hypothalamic
rIL-3
171 + 21
l, 729 + 98
Hippocampal
rGM-CSF
134 + 12
5 , 8 6 3 + 481
rlL-3
155 + 14
6, 998 + 499
rGM-CSF
201 + 17
5 , 9 4 7 + 392
rIL-3
241 + 21
7, 223 + 441
Amygdaloid
Bone marrow cells were incubated with rGM-CSF and rIL-3 for 24 hours. 2 Data are expressed as means + SEM of four separate determinations. 3 Electrode placement without electrolysis and no treatment with cytokines. 4 Added to BMC suspensions at a concentration of 50 U HCSA/ml, one unit representing the amount of cytokine that produces a 100% increase in histamine production. 5 Electrode placement without electrolysis and treatment with cytokines.
5.
CONCLUSIONS
Chemical axotomy in adult mice, chemical sympathectomy in neonatal mice, or central catecholamine depletion in rats by 6-OHDA, inhibit the IL-3 or GM-CSF induced histamineproducing cell-stimulating activity, whereas electrolytic lesions of the pre-optic and anterior hypothalamic area (representation of parasympathetic control) enhances the same activity in adult rats. Conversely, posterior hypothalamic lesions (representation of sympathetic control) inhibit the histamine-producing cell stimulating activity and those in the hippocampus and amygdala have no effect. The findings indicate that in the biochemical sequence of reactions, induction of the d e n o v o synthesis of HDC in hematopoietic progenitor cells produced by IL-3
55
or GM-CSF there appears to be a step that requires the presence of normal adrenergic input both centrally as well as peripherally.
REFERENCES 1.
2.
3. 4. 5.
6. 7. 8. 9.
10. 11.
Schneider E, Pollard H, Lepault F, Guy-Grand D, Minkoweski M, Dy M. Histamineproducing cell-stimulating activity. Interleukin3 and granulocyte macrophage colonystimulating factor induce d e n o v o synthesis of histidine decarboxylase in hemopoietic progentior cells. J Immunol 1987; 139(11): 3710. Szentivanyi A, Reiner S, Schwartz ME, Heim O, Szentivanyi J, Robicsek S. Restoration of normal beta adrenoceptor concentrations in A549 lung adenocarcinoma cells by leukocyte protein factors and recombinant interleukin- 1c~ (I1-1c~). Cytokine 1989; 1: 118 Shore PA, Burkhalter A, Cohn Jr VH. A method for the fluorometric assay of histamine in tissues. J Phamacol Exptl Ther 1959; 127:182 Katsch S and Szentivanyi A. Dissociation of the histamine-sensitizing and histidine decarboxylase-enchancing activities of the Bordetella pertussis cell. J Allergy 1968;41:106 Szentivanyi A, Katsch S, McGarry B. Lack of correlation between the histamine sensitizing and histidine decarboxylase activating effects of Bordetella pertussis. Federation Proc 1968; 27: 268. Baudry M, Martres MP, Schwartz JC. The subcellular localization of histidine decarboxylase in various regions of rat brain. J Neurochem 1973; 21: 1301 Jonsson G. Lesion methods in neurobiology. Techniques in Neuroanatomical Research, Ch Heym and W G Forssmann, eds., Springer-Verlag, Heidelberg, 1981, pp. 71-99. Szentivanyi A, Szekely J. Anaphylaxie und Nerven system II. Orv Hetil (Budapest) 1952; 19:1193-8. Szentivanyi A, Fischel CW, Talmage DW. Adrenaline mediation of histamine serotonin hyperglycemia in normal mice and the absence of adrenaline-induced hypoglycemia in pertussis-sensitized mice. J Infect Dis 1963; 113: 86. Szentivanyi A, Szekely J. Effect of injury to, and electrical stimulation ofhypothalamic areas on the anaphylactic and histamine shock of the guinea pig. Ann Allergy 1956; 14: 259. Szentivanyi A, Szekely J. Anaphylaxis and the nervous system. Part IV Ann Allergy 1958; 16: 389.
This Page Intentionally Left Blank
I1.
NEUROIMMUNE REGULATORY MECHANISMS
This Page Intentionally Left Blank
New Foundation of Biology Edited by I. Berczi and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
59
Introduction
REGINALD M. GORCZYNSKI
Department of Surgery & Immunology, University of Toronto, The Toronto Hospital, CCRW 2-855, 200 Elizabeth Street, Toronto, ON, Canada M5G-2C5
This series of chapters brings together a number of manuscripts presented by individuals who were asked to address, in broad terms, the potential regulatory mechanisms involved in neuroendocrine control of the immune system. However, before any significant discussion of these mechanisms can begin, it is crucial to understand at least some essentials of the manner in which autoregulation occurs in the immune system. This is elegantly done in the first two chapters of this section by Bretscher et al., and by Hayglass, both prominent Immunologists in their own right. Bretscher takes a contemporary problem, and leads us through a detailed consideration of cytokine production by CD4 + T cells within the immune system as a means to illustrate his points, while Hayglass adopts a more historical perspective. He considers "the role that changing perceptions of antigen specificity and self tolerance have played in shaping our understanding of the immune system, its regulation and its interactions with other biological systems" as a tool to introduce us to the view that our understanding of key issues in immunology, and even their identification as important concepts, has been critically shaped by the prevailing paradigms in the science. This provides a useful starting point from which to consider how other physiological systems might be integrated within the regulation of the immune system per se. Antel discusses how understanding the commonality between cell surface and soluble molecules both produced, and responded to by cells within the CNS and immune system can help highlight potential novel strategies for manipulations which might have therapeutic efficacy. As an example quoted by Antel, T cells may be an important source of cytokines and neurotrophins of benefit to a diseased or injured CNS, or gentic engineering (introducing such immune-derived molecules directly into the CNS) might be benefical. The chapter by Moghaddam and Sternberg (and also some of the later sections of this book) expands upon this issue-in this particular case examining in animal models and humans the interrelationship between inflammatory states and autoimmune disorders. Rivest takes this important problem even further with an elegant analysis of the molecular events which unfold in the CNS following endotixin challenge. They discuss in some detail the role of TLR4 and CD14 expression in immunoregulation and protection (of the CNS) from inflammatory processes, including how their expression is altered in response to changes in the blood brain barrier. Stefaneanu provides a detailed review of regulatory circuits within the pituitary as an introduction to the complexity of integration of neurohormonal and cytokine regulation of immunity. Hypothalamic hormones, peripheral hormones and local mediators are all believed important in control of pituitary cells. In addition to the role of pituitary hormones in stimulating peripheral
60
endocrine glands, and growth and metabolism, it is known that they can stimulate (GH, PRL, TSH) or depress (ACTH) peripheral immunity. Furthermore it has now been realized that the pituitary synthesizes some important cytokines (classically believed to be immunological mediators!). Even this complexity is not enough, however, as witnessed by independent contributions from Nance and Krukoff and their colleagues. Nance et al., showed that while the HPA axis was clearly implicated in control of the peripheral immune response (in their case, analysed by spleic macrophage function), nevertheless acute suppression of such macrophage function by central inflammatory stimuli and stress were still observed in adrenalectomized animals. This was abrogated by surgically cutting the sympathetic nerve fibres innervating the spleen, documenting unequivocally that the sympathetic nervous system constituted an important pathway for neural regulation of peripheral immunity. The work they have described document an essential role for the PVN and its connections in this regulation. Krukoff et al., take as their starting point an examination of the role of that uniquitous molecule, NO, in CNS-immune system interactions. They show that it may be the NOS isoenzyme stimulated, and level of NO itself released within the brain, which is crucial. Modest inflammatory challenge activates nNOS and eNOS, while more intense inflammatory stimuli activates iNOS itself, and greater NO production. While release of NO from all three forms of NOS occurs within the PVN, most of the available literature suggests that NO produced from nNOS or eNOS inhibits acticity of the HPS axis, while NO from iNOS has direct toxic/inflammatory effects within the brain. They propose that nNOS and eNOS activation are most likely involved in a feedback regulation of the HPA axis in response to chronic stimuli, returning the system to homeostasis through central inhibition of immunity. This discussion of Krukoff's bears upon a theme introduced also by Anisman and colleagues. While not questioning the role of acute stressors (physical and/or psychological) on the inflammatory responses discussed by other workers, and indeed the role of cytokine changes per se as potential mediators of those responses, there is a need to consider the role of chronic stress-induced changes in such parameters. Firstly, many insults themselves are protracted in nature; secondly, many of the cytokines produced by acute insults are persistent for long periods. Many years ago Anisman documented compensatory changes in amine synthesis in animals subjected to chronic environmental challenges-do similar changes occur at the immunological (and or neuroimmunological level) and what might be the effect of such changes to behaviour and the immune system? This compendium of discussions highlights what we know to date concerning the mediators implicated in CNS: immune system interactions, and provides "food for thought" on how their integration is crucial for normal physiological functioning of both the CNS and immune system independently, as well as together. In later chapters the reader will be confronted with the many physiological processes whose "normal" functioning is itself evidence of how tightly controlled these CNS immune system interactions really are.
New Foundationof Biology
61
Edited by I. Berczi and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
Dynamics of Immune Responses" Historical Perspectives in our Understanding of lmmune Regulation
KENT T. HAYGLASS
Department of Immunology, University of Manitoba, Winnipeg, Manitoba, Canada R3E-OW3
ABSTRACT This presentation provides a historical analysis of major developments that have led to our current understanding of immune regulation. Particular emphasis is placed on the role that changing perceptions of antigenic specificity and self tolerance have played in shaping our understanding of the immune system, its regulation and its interactions with other biological systems. Four specific cases are discussed. I review (i) the initial debate between cell biologists favoring a pre-eminent role for cell mediated immunity vs. immunochemists as proponents of specific immunity as "the" key defining characteristic of immunity; (ii) the concept that immune regulation was largely attributable to the actions of antigen-specific T cell factors; (iii) the controversy concerning the initiation and expansion of studies on cytokines and their roles in ontogeny, control and extinction of immune responses, (iv) the birth, rapid acceptance and subsequent scepticism of the utility of the Thl/Th2 hypothesis as a paradigm to encompass immune regulation. The aim in considering these four cases is to highlight how changing perspectives of the scientists working at the time and the then dominant models of immune regulation act to influence development of the immunological models used to interpret these findings. The presentation concludes with a personal view of some of the issues that need to be considered in order to allow a broadly accepted synthesis of biological systems, such as the immune and the neuroendocrine systems, that have largely been studied in isolation.
1.
INTRODUCTION
This presentation provides a historical overview of developments in immune regulation in contemporary immunology. The underlying theme is that not only are immune responses dynamic in their interactions with the broad variety of positive and negative stimuli which they encounter but so is our understanding of their regulation. In particular, in this review, I will consider how the then dominant views of antigenic specificity and tolerance have shaped our understanding of immunity over the last 125 years. Four specific examples-which have been deliberately selected from popular models that have come and largely gone-are chosen to illustrate the frequently made observation that few things are more appealing to a population of critically minded scientists than an all encompassing hypothesis. My intent is to explore
62
how the thinking that dominates scientific thinking at a given time acts to shape the scientific community's view of what is "valid" and to encourage scientists interested in dissecting the complex relationship between the nervous system, the endocrine system and the immune system in their efforts to obtain a broader view of the physiology of host defense.
.
PROTECTIVE IMMUNITY" CELL MEDIATED VS. HUMORAL IMMUNE RESPONSES AS KEY
The first case study I examine begins in the 1880's. During this period, generally regarded as the birth of modern immunology, there was a protracted, often bitterly personal debate about whether cell mediated or humoral immunity (mutually exclusive options being implied) was more important for host defence. Inflammation had long been recognized and had been a topic of active enquiry by pathologists for the preceding 40 years. One school of thought was represented by the thinking of the great German pathologist Rudolf Virchow (1821-1902), a founder of cellular pathology, who contributed to nearly every branch of medicine. This school viewed inflammation as an undesirable response to a pathologic agent that resulted from disturbed nutrition of host cells in the area of infection. This in turn led to intensified local proliferation of parenchymal cells and in serious cases to the tumor, which he considered the most significant component of the process. Julius Cohnheim, on the other hand, concluded from his famous experiments that inflammation was mostly due to lesions of the walls of the blood vessels. This permitted passive leakage of all of the components (many of which are humoral) that were then recognized as part of the inflammatory response. So, although the field of general pathology was divided on which of these two mechanisms was most important for the inflammatory reaction, almost all agreed that inflammation was a deleterious reaction of no benefit to the host. It was a purely passive response on the part of the insulted organism. Indeed, phagocytic cells were thought to be one of the main mechanisms that pathogens used to traffic to other parts of the body, widening the infection. In this context of inflammation being a strictly undesirable response of the host to an environmental insult (much as delayed hypersensitivity from poison ivy is viewed today), Elie Metchnikoff (1845-1916, Russian zoologist and bacteriologist) dared to suggest that the phagocytic cells seen in inflammation, far from being harmful, in fact constitute an active first line of defense. He believed that their ability to ingest and digest invading organisms was the key characteristic of self defense by the organism. Metchnikoff was advised to proceed with great caution, since "most pathologists do not believe in the protective role of inflammation". Silverstein, in his seminal History of Immunology defines the historical context that makes it so understandable that Metchnikoff's radical views would have a poor reception. They not only challenged the very foundation of then-current dogma, they also were advanced by an individual who was: (1) not a member of the confraternity of pathologists (Metchnikoff was a zoologist); (2) not even a physician (the chemist Pasteur had encountered similar problems); and (3) a Russian (a people then traditionally considered somewhat backward by many Western Europeans). (Silverstein, 1989). Metchnikoff is both a great and a tragic figure. At one stage in his life, having undergone great personal and professional tribulations, he attempted (for the second time) to end his own life. In order to save his wife and others embarrassment, he decided to do this by means of the scientific experiment of inoculating himself with relapsing fever to find out whether it was transmissible by the blood. The attack of relapsing fever that followed was severe, but
63
it did not kill him. He went on to carry out a series of brilliant studies in Italy and Russia. Metchnikoff was appointed Director of an Institute established in 1886 in Odessa to carry out Pasteur's vaccine treatment of rabies, but there was much local hostility to this treatment. Metchnikoff found that, partly because he was not a medical man, circumstances became so difficult that, in 1888, he left Odessa and went to Paris to ask Pasteur for his advice. Pasteur gave him a laboratory and an appointment in the Pasteur Institute. Here he remained for the rest of his life. His studies on the role that phagocytosis and inflammation might play in host defence were ultimately to win him the Nobel Prize (1908). (http://www.nobel.se/laureates). Much of the scientific community was most unhappy with Metchnikoff's provocative ideas. Photographs taken of him when he was working at the Pasteur Institute show him with long hair and an unkempt beard. It is said of him that at this time he usually wore overshoes in all weathers and carried an umbrella, his pockets being overfull with scientific papers, and that he always wore the same hat, and often, when he was excited, sat on it. As he felt more and more isolated he became increasingly strident in his arguments for phagocytes, arguing that in addition to immunity, phagocytes are the chief agents of the aging process, due to active phagocytosis of neurons leading to senility and the phagocytosis of hair pigment to causing graying. At roughly the same time, discoveries by British and Prussian workers (most strikingly yon Behring, later Paul Ehrlich) focussed attention on an alternative explanation for adaptive immunity: humoral factors. This work sprang from studies carried out developing anti-toxins, leading to a series of studies demonstrating vigorous protective efficiency by serum in the absence of any cells (i.e. Ab+Complement). Ehrlich, Director of Institute for the Control of Therapeutic Sera in Berlin, made important breakthroughs in immunology, especially on hemolysins antibodies (Abs). He demonstrated that the toxin-antitoxin reaction, like most classical chemical reactions, is accelerated by heat and retarded by cold and that the content of antitoxin in antitoxic sera varied so much for various reasons that it was necessary to establish a standard by which the antitoxin content could be exactly measured. He accomplished this with von Behring's antidiphtheritic serum and thus made it possible to standardize this serum in units related to a fixed and invariable standard. The methods of doing this that Ehrlich established formed the basis of all future standardization of sera. This work and his other immunological studies led Ehrlich to formulate his famous side-chain theory of immunity. Thus, the application of reproducible, quantitative methodology paved the way for development of an immunochemically based theory of immunity in a period that followed closely on the advances in chemistry that characterized the end of the 19th century. The positive result of this rivalry was that both sides were forced to test and continuously refine their theories in light of the opposition of the other camp. Unfortunately this debate took place during a period of rabid European nationalism between the Franco-Prussian war and humiliating defeat of France in 1870 and at the end of World War I (1918) with the collapse of the Central Powers. Given that the chief protagonists of specific immunity were in Prussia, and the major proponent of non-specific cellular immunity was a Russian living in Paris, it is easy to see how few were interested in the compromise position that BOTH forms of biological response might be essential contributors. The phagocytic theory of immunity was not the only dispute in which objective science was compromised by the after effects of the Franco-Prussian War. In the aftermath of the siege and destruction of Paris in 1870, Louis Pasteur, who in 1868 had received an honorary M.D. degree from the University of Bonn, returned his honors in anger (Table I). Compromise was no more the order of the day in science than it was in politics.
64
Table I
Politics, science and compromise at the birth of modern immunology.
Pasteur wrote to the Dean of the Faculty of Medicine at Bonn that "Now the sight of that parchment is odious to me, and I feel offended at seeing my name, with the qualification of virum
clarissimum that you have given it, placed under a name which is henceforth an object of execration to my country, that of Rex Gulielmus. I am called upon by my conscience to ask you to efface my name from the archives of your faculty, and to take back that diploma, as a sign of the indignation inspired in a French scientist by the barbarity and hypocrisy of him who, in order to satisfy his criminal pride, persists in the massacre of two great nations" ~
In response, Pasteur received a reply from the Principal of the Faculty of Medicine of Bonn, "I am requested to answer the insult which you have dared to offer to the German nation in the sacred person of its august emperor, King Wilhelm of Prussia, by sending you the expression of its entire contempt" 2. [1, 2] Silverstein Arthur M. A History of Immunology. San Diego: Academic Press, Inc. Harcourt Brace Jovanovich, Publishers, 1989. pg. 45.
The cellular vs. humoral debate continued, and the Nobel committee attempted to show balance by the joint award of the 1908 prize to Ehrlich and Metchnikoff. However the humoralists won the day and came to dominate immunology research. This state persisted for some 60 years as chemists and antigenic specificity dominated development of the science. The dominant focus was on structure rather than function of different components of the immune response, firstly as immunochemistry then as genetics. Most scientists were much more comfortable with antibody than they were with the difficult phagocyte, in part because of the rigor, reproducibility and quantitative nature of the assays rather than a focus on phenomenologic data.
3.
"HOW IS THE IMMUNE RESPONSE REGULATED"?
The second case I would like to discuss concerns our attempts to understand the very basic question of how immunity is regulated. Until the mid-1970' s, most of the focus in immune regulation had been directed towards: "How does one turn ON an immune response"? Whether it be following natural infection or in extensive empirical efforts towards vaccine development. Responses were generally thought to be turned OFF by default. Once the supply of antigen was exhausted, presumably by the success of the immune system in removing the pathogen, the response was thought to just stop, leaving circulating antibodies remaining for the next exposure and the development of the memory response. Studies in the mid-1970's by Dick Gershon and many others argued against this view. They reported that there were active negative mechanisms of regulation that played a key role in homeostasis of the immune response. Cells termed suppressor T cells (Ts) were found that could strikingly inhibit adaptive immune responses and that were seen as key negative regulators of immunity. A number of characteristics become widely accepted (Table II). Ts cells recognized antigen directly, much as does Ab. In this way they are different from all other T cells. A dominant hypothesis of the time was that the idiotype (a specific marker of antigen binding receptors of T or B cells) was a key player. This concept resulted in a Nobel prize for Neils Jerne for "theories concerning the specificity in development and control of the immune
65
system". This hypothesis argued that an immune response engendered an immune response against itself (anti-idiotype). This anti-id then engendered a response against itself (anti-anti-id) and so forth like a series of mirrors. This concept of id anti-id regulation was integrated into the view of suppressor T cell biology. Thus, different populations of suppressor T cells (expressing idiotypes, anti-id and anti-anti-id for Ts 1, 2, 3 respectively) were seen to form a web of regulatory cells that controlled virtually all immune responses. This view dominated immune regulatory models for over 15 years. What the hypothesis had in its favour was: 9 the concept of active regulation of immune responses (both positive and negative) was inherently appealing, 9 it integrated the information known at the time about T cells and antigen recognition structures, and 9 it fit most of the available data into a global view of immune regulation. Table II
Suppressor T cell characteristics.
9
A distinct class of T lymphocyte, capable of binding Ag directly (no need for Ag processing): The biological
9
They control a wide variety of immune responses (DTH, Ab, CTL...) directed against foreign and self Ags.
9
Predominately CD8 populations restricted by class II major histocompatibility gene complex elements (I-J).
9
Are themselves part of a regulatory cascade (Tsl, Ts2, Ts3) that is based on the notion of a network of idiotype:
9
Inhibit (i) the initial development or (ii) recall of immune responses in an Ag-specific manner, largely by release
counterpart to "helper" T cells.
anti-idiotype interactions. of soluble factors (TsF) that are Ag-specific.
However, as time went by the model had to be stretched further and further to encompass emerging data until it ultimately became untenable. This area of investigation eventually led to discovery of suppressors of suppressors (contra-suppressors). Contra-suppressors were said to be required to allow a response to progress in the presence of suppressor cells. However, beyond the discomfort that many in the immunologic community felt about an increasingly arcane model, the most significant problem for the model of Ts cells as elaborated in the 1970's and 1980's was inconsistencies pointed out by simultaneous findings in molecular biology. The key MHC restriction elements for Ts cells did not exist in the genome; some of the T cell receptor hybridomas did not express TcR or did not have rearranged T cell receptor genes (c~ or TS); and a number of other observations all represented significant problems that left many scientists unsettled with the current model. At the same time, it opened up a field that had itself been plagued with poorly reproducible and highly complex findings up to that time: cytokines.
4.
CYTOKINES IN IMMUNE REGULATION
At the same time that Ts were very much the dominant paradigm, work was going on in other labs examining a large family of secreted molecules termed interleukins. The pivotal initial finding was the discovery that tissue culture supernatants of activated lymphocytes could substitute for the intact cell population. Thus, cells did not need to be physically in contact-
66
soluble mediators produced by activated cells could replace multiple biological functions. One of the first examples of this came from Robert Gallo's laboratory who discovered in 1977's that fresh T cells-when stimulated with specific antigen or polyclonally-produced a "factor" called T C G F (T cell growth factor), now known as IL-2. This was one of the first definitive clues that cells of the i m m u n e system communicated with one another via soluble molecules. This rather revolutionary finding drove others to begin looking for "factors" in tissue culture supernatants. Scores were found. Several observations soon became obvious (Table III). As is well known, every one of these key tenets of i m m u n e regulation in 1980 are false: our bias for specificity-the legacy of the battles in the first part of the 1900's, greatly coloured our view. Table III
Cytokinesin immune regulation: 1980 vs. Today.
1980 Immunoregulatory factors are lineage specific in activity (a T cell activating factor can not activate B cells). Factors are functionally specific (A factor inducing proliferation is distinct from one which would induce differentiation or activation or cell death). Factors are produced by and act specifically on cells of the immune system. 2000
9
,,
Cytokine receptor expression, hence the capacity of cytokines to convey a particular signal, is virtually never unique to a particular cell type. Dependingon the circumstances, the same cytokine can have multiple opposing effects. IL-2 greatly assists in T cell proliferation. It is also required for apoptosis, the key immunologic mechanism for maintaining homeostasis. Cytokines, chemokines and the multitude of other regulatory factors that influence ontogeny, initiation and maintenance of immune responses are made by--and act on--a wide variety of cell types.
The successes that followed the identification of cytokines as antigen-non-specific regulatory molecules led to the emergence of new problems. Ehrlich's side chain theory to explain antibody formation had worked well when it was assumed that the i m m u n e system might respond to perhaps a couple of thousand different bacteria or viral pathogens. Once organic chemists became involved and demonstrated that a specific antibody response could be induced to virtually any chemical they could synthesize, then an apparent contradiction was created. How could the body contain a sufficient variety of antibody producing cells and how could each possibly contain enough genetic material to encode 109 specificities: a paradox that took the work of Burnet and T o n e g a w a to resolve. A similar problem developed for cytokines. As more and more of these immunoregulatory molecules were discovered, and it was learned that each had multiple activities, multiple targets and could synergise or antagonize the activity of others, the system appeared to become impossibly complex.
67
Table IV
The type 1/type 2 immunity paradigm.
1. Assets Provides an integrative model of immune regulation that allowed one to causally relate. 9 inductionof immune responses (i.e. IL-12 vs. IL-4 production), 9 the nature of the subsequent T cell response (IFNg vs. IL-4, 5, 13), 9 immuneeffector mechanisms (IgE vs. IgG2a/DTH dominated) and 9 clinicalstatus. 2. Liabilities 9 9
Overinterpreted. Overextended.
3. Popularity Year 1986-1987 1990-1991 1994-1995 1996-1997 1998-1999
5.
Th 1/Th2 Citations 29 238 1307 2231 2750
THE TH1/TH2 HYPOTHESIS
The next major leap forward addressed this problem. In the mid-80's, Tim M o s m a n n and Bob Coffman at D N A X made a striking observation that led to their development of the T h l / T h 2 paradigm. When examining cytokine synthesis by a large panel of murine T cell clones, they did not find a random distribution but rather observed that most clones fell into one of two major patterns. With minor exceptions, only T h l clones synthesized m R N A for IL-2, IFN~, and lymphotoxin, and only Th2 clones synthesized m R N A for IL-4, IL-5, and another induced gene, P600 (later found to be IL-13). Some cytokines are made by both populations (i.e. IL-3, GM-CSF). Clones established from less differentiated T cells made a more primitive pattern and came to be called Th0 precursor cells. With increasing attention from the scientific community a number of other features emerged. This model rapidly came to dominate studies of immune regulation from the late 1980's into the beginning of the new millennium. W h y was it so successful? I believe its primary appeal was that it provided a unifying view of this area of biology. It was well established that excessive IFN~, production was often associated with the process of inflammation and maintenance of chronic inflammatory diseases. It was becoming clear that elevated production of IL-4 and IL-5 was associated with increased IgE, eosinophilia and clinical allergy. Thus, for perhaps the first time, this model allowed us to causally relate alterations in immunoregulatory cell function and cytokine production (IL-4,
68
IL-5, IL-13 dominated responses vs. IFNT dominated responses) with the kind of effector immune responses that dominated in vivo (i.e. IgE vs. DTH) and hence to clinical status and prognosis (protective immunity, hypersensitivity, or tolerance). To be able to tie biology together from the most mechanistic level of molecular interactions between cells through to identifying which one of ten randomly selected children will ultimately develop asthma upon exposure to cats, is very appealing. As one might anticipate, this sparked a flood of enthusiasm for examining the usefulness of this model in a variety of experimental and clinical situations. Thl/Th2 engendered an almost religious fervor. Indeed, it became difficult to find an immune response where the data was not attributed to Thl/Th2 based immunoregulatory interactions. A proportion of this was reasonable, because it provided a very useful working framework within which to interpret data and test hypotheses. However, it must be said that a substantial proportion of the citations could also be attributed to overly enthusiastic adherence to a currently popular model (i.e. identifying "Thl cell" activation solely on the basis of measuring IL-2 production). Already by the early 1990' s, a time at which the Thl/Th2 hypothesis was gaining widespread acceptance, research initiated by Anne Kelso then of WEHI, indicated that T cell differentiation and activation did not yield a bimodal distribution of type 1 or type 2 cytokine responses but rather that Thl and Th2 clones in vitro represented extremes at the end of the spectrum. This work continues today in many laboratories with important questions about the reversibility of type 1 vs. type 2 committed T cells, the critical factors that determine whether a naive T cell seeing antigen for the first time will differentiate into a cell with an IFNT or IL-4/IL-13 dominated cytokine response, and characterization of the intracellular signalling components that mediate and control this process, still to be resolved. Type 1 vs. type 2 immunity is now generally accepted to encompass CD4/8 T cells, many other cells of the innate immune response (mast cells, T8 T cells, NK T cells), other antigen-specific populations (i.e. mast cells or eosinophils bearing specific Ab via Fc receptor expression) and perhaps even to cells outside of the classical immunological repertoire (i.e. members of the neuroendocrine system). A balance of commitment to type-1 vs. type-2 immunity is often predictive of effector immune responses and clinical status. Consequently, much work is directed towards better understanding how these many factors interact to regulate host responsiveness to chronic antigenic stimulation.
6.
CONCLUDING REMARKS
In summary, in each of the instances discussed, there was a difficult struggle by researchers in convincing their colleagues that the current, usually excessively restricted, model was no longer sufficient to explain the biology. Once that point was passed, enthusiasm for the new paradigm often became excessive as many jumped on the bandwagon, repeating the cycle. The concept of identifying regulatory interactions between the nervous system, the endocrine system and the immune system is very appealing to anyone with an interest in biology. The idea that these physiologic systems function in splendid isolation seems most unreasonable. At the same time, a significant proportion of research in this difficult multi-disciplinary area has been viewed as phenomenologic. Reports of striking, often extremely provocative biological effects are numerous but definitive mechanistic experiments, and a broad integrative, testable view of the biology, are often lacking. In this regard, the field sometimes seems rather close to that of cytokines in the late 1970's, as "factors": were discovered that generated considerable data and papers but limited comprehensive understanding. In hindsight, those working on
69
cytokines were stuck in a limbo where to be working on "interleukins or factors" at that time was an almost embarrassing admission. This was not because other scientists disputed the existence of lymphokines but because the experiments we were designing were for ourselves-we were generating more data and phenomena for the converted while the rest of the immunologic community saw us as living in a ghetto. Few would dispute that the immune and the neuroendocrine systems interact. The challenge will be to develop research strategies that can definitively address those connections and convince others that these are experimentally approachable problems.
REFERENCES Delves PJ, Roitt IM. (eds.). Encyclopedia of Immunology. 2nd edition, volume one. Academic Press: San Diego and London. 1998. Janeway CA, Travers P, Walport M, Capra JD. Immunobiology: The Immune System in Health and Disease. 4th Edition. Elsevier Sciences Ltd/Garland Publishing: London and New York. 1999. Paul WE (ed.). Fundamental Immunology. 4th edition. Lippincott-Raven Publishers: Philadelphia and New York. 1999. Samter M(ed.). Excerpts from Classics in Allergy. Edited for the 25th Anniversary Committee for the American Academy of Allergy. Ross Laboratories: Columbus, Ohio. 1969. Silverstein AM. A History of Immunology. Academic Press, Inc.: San Diego and London. 1989. Website: http: //www.nobel.se. The official Nobel website with excellent brief histories of prizewinners, the significance of their discoveries and the text of their acceptance speeches.
This Page Intentionally Left Blank
New Foundationof Biology
71
Edited by I. Berczi and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
Cell-to-cell Interactions and Signaling within the Immune System" Towards Integrating Mechanism and Physiology
PETER A. BRETSCHER, NAHED ISMAIL, NATHAN PETERS and JUDE UZONNA
Department of Microbiology and Immunology, University of Saskatchewan, 107 Wiggins Road, Saskatoon SK., Canada S7N-5E5
"There are more things in heaven and hell, Horatio, than are dreamt of in your philosophy". Shakespeare in Hamlet.
ABSTRACT Signals to cells provide the means of controlling their fate by the environment in which they exist. The integrity of multicellular organisms requires the fate of individual cells to serve the overall purpose of the organism's physiology. The changing needs of the organism is reflected by changes in its internal environment, thus regulating the fate of cells. Signaling has been recognized as central in accounting for how higher level properties or attributes of the immune system are achieved. The first attribute is self-nonself discrimination, i.e. the attribute allowing the immune system to attack foreign invaders while not attacking self. The importance of this attribute is evident when its integrity is lost, as occurs in autoimmune diseases such as Hashimoto's disease and diabetes. The second attribute depends upon the two facts that the adaptive immune system has many distinct ways of fighting foreign invaders, i.e. can produce many different classes of immunity, and that it chooses to produce different classes in different circumstances. These decision making processes can be referred collectively to as immune class regulation. The importance of the integrity of this attribute is again evident when it is lost. For example, about 95% of people infected with Mycobacterium tuberculosis contain the pathogen, by producing a sufficiently strong cell-mediated response, and therefore do not suffer from TB. The others have an inappropriate response, most probably because the cell-mediated response is too weak or because it contains a substantial and detrimental, antibody, Th2 component. We discuss what features of these two attributes are physiologically important, and how desirable features might be attained. We focus on whether the perceived need to explain certain features imposes significant restrictions on the nature and generation of the signals required to activate lymphocytes to divide, to differentiate or to undergo apoptosis. In particular, it seems likely that immune responses are usually independent, such that inhibiting an anti-self response does not result in inhibition of an anti-foreign response, or that induction of an anti-foreign response does not result in induction of an anti-self response. Somewhat similarly, we expect
72
the decision to mount a cell-mediated, Thl response to a foreign antigen F1, should not affect the kind of response induced simultaneously to another foreign antigen F2. Evidence for such independence will be reviewed. The implications of such independence will be examined for what they might tell us about how signals for the activation and differentiation of lymphocytes are generated.
1.
INTRODUCTION
Cellular signals are a means of communicating from the outside to the inside of a cell. The nature of the signals generated governs the fate/function of the receiving cell. The fate of individual cells of the immune system must, under the best of physiological circumstances, be governed in a manner consonant with the attributes of the immune system that reflect its overall functioning. Thus the environment of a cell must be carefully controlled to achieve such integration. We outline here a scheme of cellular interactions, mediated by distinct signals, that provides our best, provisional guess as to what is the nature of some of these interactions and signals; our guide is that the postulated signals and interactions should not only be consonant with observation but also with central attributes/features of the immune system. A driving consideration in developing the scheme proposed has been to account for two attributes of the immune system: self-nonself discrimination and immune class regulation. We consider this approach to be an experiment in theory-making. We can only judge its usefulness by where it leads us, and it is certainly only at best a clumsy start. Considerations about how the immune system functions should acknowledge at the start that "the immune system" is an abstraction made for simplicity, and that the process of simplification may invalidate the analysis. "The immune system" is part of an integrated organism, so we expect there to be physiological interactions between components of what we artificially delineate, for example, as the "endocrine", the "immune" and the "nervous" systems. We therefore take for granted the provisional nature of the picture we draw. We hope this picture provides a starting point for those who wish to "integrate" how the endocrine, immune and nervous systems might interact, but we are also aware that attempts at integration may show up the inadequacy of our first attempts in which we consider the immune system in isolation. We can anticipate a dialectical process. A simple illustration indicates why we should be cautious. There are reports that the number of mice in a cage can affect the nature of the immune response generated to a given antigenic stimulus. For the present, our approach is knowingly to ignore such observations, without denying their validity and potential pertinence to understanding the functioning of the individual as a whole. We feel it appropriate, before embarking on the main subject matter of this contribution, to make a plea to the reader to allow that the immune system may be much more complex or, to put it more positively, more sophisticated, than what is commonly discussed in our public dialogue. Most neurophysiologists, aware of the elaborate architecture of the brain and of the topological relationships between neural cells, feel these spatial relationships must have physiological significance. Valid explanations of neurophysiological function must "account for" neuroanatomy. This appears more than obvious. Most of the primary cells of the immune system, lymphocytes and phagocytic cells, are mobile cells that are often not obviously engaged in intimate contact and communication with each other. This contrasts with neurons. Lymphocytes and phagocytic cells have surveillance functions and hence need to be mobile. This should not preclude us from envisioning that they may interact in as sophisticated a fashion
73
as neurons. We have found it remarkable that there has been so much general scepticism against the idea that lymphocytes can have inhibitory or suppressive activities. A positive regulatory activity of one subset of lymphocytes on the activity of another subset will be unstable without any means of controlling it. Nature is surely much more elaborate than is our current understanding.
THE FOCUS OF ATTENTION: PERIPHERAL TOLERANCE AT THE LEVEL OF CD4 HELPER T CELLS AND THE NATURE OF THE DECISION CRITERION CONTROLLING THE TH1/TH2 NATURE OF THE RESPONSE An overview of how the immune system functions in terms of signals and cellular interactions is a vast subject, and we hope it will help the reader if we delineate at the start the two main areas we shall focus upon, with a brief explanation of why we consider them so central. We take it as established that processes related to self tolerance occur in primary lymphoid organs, such as the bone marrow and thymus. Lymphocytes, specific for self antigens that are present in these organs at sufficient levels, are silenced one way or another. This is referred to as central tolerance. Lymphocytes specific for self antigens not present in the thymus emigrate [ 1]. Their existence is central to understanding the occasional development of organ-specific autoimmunity. The rarity and devastating nature of organ-specific autoimmunity, once it does occur, leads most immunologists to suppose that there are not only ways of activating mature lymphocytes in the secondary lymphoid organs to produce effector cells, but ways of silencing those mature lymphocytes with the ability to recognize organ-specific self antigens, and that these silencing processes are effective in most people. The process of inactivating mature lymphocytes implies a mechanism of "peripheral tolerance". Self antigens, present outside the thymus in a form accessible to lymphocytes, but not present in the thymus at the levels required to establish central silencing, are called peripheral antigens. Most resting B cells, and many resting CD8 T cells, require activated CD4 T cells to be fully activated. In the absence of such activated helper T cells, antigen interacts with these resting B and CD8 T cells to render them non-functional, i.e. refractory to subsequent attempts to activate them. Although there are some exceptions to these rules, they are relatively minor. It thus appears that, if there are no CD4 T cells specific for peripheral self antigens, no B and CD8 T lymphocytes (whose activation is CD4 T cell dependent) will be induced, but rather their interaction with peripheral antigens will lead to their silencing. The significance of ensuring that CD4 T cells specific for anti-self peripheral antigens are not activated is apparent, as CD4 T cells appear to be the guardians over the activation/inactivation of other lymphocytes. It is no accident that effector CD4 T cells are called helper T cells. Therefore, the first area we shall focus on will be what determines whether resting precursor T helper (pTh) CD4 cells are inactivated, or activated to multiply and to differentiate to produce effector T helper (eTh) cells, upon interacting with antigen. We perceive this as being central to understanding peripheral tolerance and the etiology of organ-specific autoimmunity. It is interesting to observe that most clinically relevant autoimmunity (with the exception of systemic lupus erythematosus) is organ-specific and is believed to reflect a failure in peripheral tolerance. We believe this failure most often reflects a failure in tolerance at the CD4 T cell level. The second area we shall focus upon is the mechanism that determines whether the activation of pTh cells results in the generation of Thl or Th2 cells. The importance of this is reflected by the current prominence of discussions concerning its nature. We refer to this mechanism as the "decision criterion" controlling the Thl/Th2 nature of the response. It is clear that the
74
two problems we shall discuss, namely how pTh cells are activated/inactivated, and whether pTh cells, if activated, give rise to Thl or Th2 cells, are closely related. Indeed, if there really is a common pTh cell for Thl and Th2 cells (which for simplicity we shall assume), one could combine the two questions: what three different sets of circumstances of antigen encounter result in the inactivation of a pTh cell or its activation to produce Thl or Th2 cells? This second area concerning the nature of the decision criterion is central to understanding how the class of immunity is determined following exposure to antigen. An understanding of the mechanisms underlying peripheral tolerance and immune class regulation together should allow the rational design of strategies to: vaccinate against intracellular parasites; vaccinate against and treat cancer; achieve transplantation of foreign organs; prevent and treat allergies, and to prevent autoimmunity.
3.
THE ACTIVATION OF pTh CELLS: SOME CURRENT IDEAS
Both subject areas we focus upon involve activation of pTh cells, and it is useful to have some terminology in hand that reflects both what we know and what we speculate concerning the activation of these T cells. We know that an antigen Q can be processed in various antigen presenting cells (APC) into peptides, q l, q2, q3 ..... qn, that bind to intracellular class II MHC molecules of the APC and that these peptide/class II MHC complexes are then presented on the APC's surface. These q/MHC complexes have the potential for interacting with pTh cells specific for the nominal antigen Q. Similarly, a nominal antigen P is processed into peptides p 1, p2, p3 ..... pn, and these peptides have the potential to interact with anti-P pTh cells. We say P and Q do not crossreact if the populations of anti-P and anti-Q pTh cells do not significantly overlap. In addition, we say peptide q2 has companion peptides q l, q2, q3 ..... qn. The significance of this will be more apparent later, when we argue that the fate of pTh cells specific for q2 are not independent of the fate of pTh cells specific for q 1, q3 ..... qn. In addition, it is very well documented that the activation of a pTh cell requires the generation of at least two signals. The first follows the engagement of the T cell receptor (TcR) with the peptide/class II MHC complex on the surface of the APC. This signal is called signal 1. Activation also requires the generation of a costimulatory signal, or signal 2, via the interaction of a receptor/counter receptor interaction between the pTh cell and the APC, see Figure 1. One such potential signal would follow the engagement of B7 molecules on the APC and CD28 molecules on the pTh cell. A central question concerning the initiation of an immune response is what controls the expression of the costimulatory molecule by the APC? (The counter receptor (CD28) is believed to be constitutively expressed by resting pTh cells). The importance of this question can be appreciated by considering various possibilities. The first is that it is constitutively expressed on all APC. This would mean that any peripheral self antigen or foreign antigen processed by an APC, would be immunogenic if there were any pTh cells available. This model is not seriously entertained, because it would be difficult to explain why we do not all suffer from autoimmune diabetes and Hashimoto's disease. Both Janeway [2, 3] and Matzinger [4, 5] have suggested that the expression by APC of costimulatory molecules requires the generation of a third signal, signal 3, that activates the APC to express costimulatory functions. Janeway proposes that this third signal follows an interaction of a component of an infectious agent with receptors on the APC that are a reflection of innate resistance. He predicts that "benign" proteins of non-microbiological origin will in general be non-immunogenic when given without adjuvant. Matzinger suggests that "danger" or stress signals not only stimulate the production of "heat shock" proteins but generate the third signal required to
75
activate APC. These are perhaps the two most influential current views. We shall later consider different alternatives.
TcR counter ~ receptor to ( costimulat~ J molecule~
&"~'"
.-.
~t25,~
~'~)-"_' -",~'~ (;;ell /
1~~ ~ ' ~ I ~ j~ \
pTh cell
_~~
!
class !i MHC molecule
constitutive costimulator molecule Figure 1. Contemporary models for the activation of pTh cells. The generation of signal 1 alone leads to inactivation; of both signal 1 and signal 2 to activation.
4.
IMPORTANT ATTRIBUTES/FEATURES OF THE IMMUNE SYSTEM
Before discussing mechanistic questions, we would like to describe some real or anticipated features/attributes of the immune system that we consider important. These features/attributes, we shall argue, put considerable constraints on how cells interact and communicate through signals, if such interactions and signals are to be consonant with such attributes and features. 4.1.
Independence of concurrent activation and inactivation of specific pTh cells by "non-crossreacting" antigens P and Q
It seems natural that the inactivation of pTh cells specific for a peripheral self antigen P by P should not be affected or, most critically, not be deviated into activation, by the circumstance that an immune response is occurring against a foreign, non-crossreacting antigen Q. Conversely, it is important that ongoing inactivation of anti-P pTh cells by P should not interfere with the activation of anti-Q pTh cells by the foreign antigen Q. If such independence were not operative, it would seem that autoimmunity would often arise, or immune responses to foreign invaders would often fail. This conjecture is relevant to the plausibility of third signal models. There are ongoing immune responses all the time in mice and presumably in humans to "foreign antigens". The spleen and other secondary lymphoid organs of an adult mouse, not
76
immunized by an immunologist, contain many antibody producing cells and activated T cells. These "on-going" immune responses presumably involve the continuous induction of lymphocytes. Thus if, as Janeway and Matzinger seem to suggest, the induction of lymphocytes requires third signals to activate APC, such activated APC must always be present. How then is organ-specific autoimmunity prevented? We shall return to this. 4.2.
Independence of the Thl/Th2 nature of concurrent immune responses to non-crossreacting antigens
Clinical observation demonstrates that the class of immunity generated against many infections is critical to clinical outcome. Thus about 95% of individuals infected by Mycobacterium tuberculosis, the pathogen responsible for tuberculosis, mount a relatively strong and exclusive cell-mediated response that contains the pathogen, and no disease is apparent. Disease probably occurs in people who either make too weak a cell-mediated response or who make a response that contains a too large and detrimental, antibody, Th2 component. On the other hand, an antibody, Th2 response is protective against infection by Ascaris braziliensis. We suppose that it often happens that individuals are simultaneously infected with more than one microorganism or parasite requiring different classes of immunity to contain them, and that usually an appropriate and effective response is generated. We think that there are obvious physiological advantages if the Thl/Th2 nature of immune responses to unrelated (noncrossreactive) pathogens are normally independently determined. We refer to this suggestion as the Independence Hypothesis. For example, it seems unlikely that the existence of an effective Th2 response against Ascaris braziliensis would generally result in the deviation of what would otherwise be an effective and exclusive cell-mediated, Thl response against Mycobacterium tuberculosis, so that the anti-mycobacterial response contained a substantial Th2, antibody component, was therefore ineffective, and therefore lead to tuberculosis. We recognize that under certain circumstances, such as when overwhelmingly large immune responses occur, such independence of the Thl/Th2 nature of responses can be lost. We regard such loss as reflecting pathology rather than physiology. We recently tested the Independence Hypothesis. Different mice were immunized with two non-crossreacting antigens R and Q, in such a manner that the spleen of the Q-immunized mice generated a virtually exclusive Thl response, whereas the spleen of the R-immunized mice produced a predominant Th2 response. We also examined the responses in mice immunized simultaneously with both the antigens R and Q in exactly the same manner as the singly immunized mice. The responses in the doubly immunized mice were indistinguishable from that in the corresponding singly immunized mice [6]. These observations support our conjecture as reflected in the Independence Hypothesis. 4.3.
The Historical Postulate: The property of peripheral self antigens that the immune system relies upon, in distinguishing them from foreign antigens and thus favoring unresponsiveness to the peripheral antigens, is their early and continuous presence
Most of the originators of Clonal Selection Theory, Lederberg, Burnet and Jerne (but perhaps not Talmage) attempted to suggest a mechanism that could account for self-nonself discrimination. They envisaged that self antigens were present before the immune system was competent to respond to them, and that under such circumstances the immune system adapted to be tolerant of these antigens, so long as they were continually present thereafter. We call this idea The Historical Postulate, because whether an animal responds to an antigen Q depends upon
77
the animal's past history of exposure/non-exposure to Q. An overwhelming amount of evidence supports The Historical Postulate. However, the recognition that there are mechanisms of central and peripheral tolerance means we must be careful in deciding what the evidence implies. We illustrate this by discussing the issue in terms of T cells. We know that the presence of sufficient antigen Q in the thymus results in the lack of export of mature, functional, Q-specific T cells to the periphery, by a thymic process of deleting or anergizing the Q-specific cells from which these mature Q-specific T cells could be derived [7]. No mature T cells are therefore generated for those antigens present in the thymus before the thymus becomes functional, and which are continuously present thereafter. Our modern observations thus fit in precisely with Lederberg's clearly enunciated vision [8]. However, how does this bear on the fate of T cells specific for a self antigen P that is present on organ-specific tissues but not present at a sufficient level in the thymus to prevent the generation and emigration of Q-specific T cells to the periphery? There is very strong evidence that such T cells are generated [ 1]. Consider the following scenario. Two similar CD4 T cells are exported from the thymus at the same time, but they have different specificity. One pTh cell is specific for an organ-specific nominal self antigen P, the other for a nominal foreign antigen F. Consider the situation where both F and P are present in similar amounts and form. In general, we would wish that P inactivates its corresponding pTh cell, and that F can activate its corresponding pTh cell. How can we ensure this? We believe this is one of the most critical questions in immunology. The only satisfactory way of ensuring this, as far as we can envisage, is by invoking. The Historical Postulate. The only difference between P and F is that P has been present during the past history of the animal. If the anti-P and anti-F pTh cells are to undergo different fates, there must be something in the environment that allows P and F to interact with their respective pTh differently. Moreover, this difference in the environment must be antigen specific, so that it affects the anti-P and anti-F pTh interactions with their respective antigens differently. We shall discuss possible realisations of The Historical Postulate later. Nevertheless, we wish to stress one point at this time. The third signal models of Janeway and Matzinger state that only the circumstances at a particular time are critical to whether a nominal antigen Q can activate its corresponding pTh cells, and therefore they in principle violate The Historical Postulate. There are experimental approaches available to test The Historical Postulate, as it might apply to peripheral tolerance as discussed elsewhere [9, 10, 11], but we are not aware that the critical experiments have been performed. Nevertheless, the violation of The Historical Postulate by the third signal models of Janeway and Matzinger makes us sceptical of their validity. 4.4.
The relationship between physiological significance of distinct classes of immunity and the decision criterion
It is important to try to be aware of the limitations of an analysis before one embarks upon it. In this context, we wish to point out in this preamble some of the limitations in the analysis that will follow. One might ask quite generally why there are distinct classes of immunity and such a range of different antibody isotypes. One advantage of having many different classes and subclasses of immunity is that they can be differentially effective in different circumstances and, only by the virtue of being different, can they be differentially regulated. This general view has much to support it, both because different classes and subclasses of immunity are differentially regulated, and because many conditions are known where different classes/subclasses are differentially effective. This is clearly a complex problem given the number of different
78
isotypes of antibody; in addition, too few observations have been made on the differential induction of different subclasses of cell-mediated immunity, such as cytotoxic T cells and delayed-type hypersensitivity, for us to get a feeling whether they are coordinately regulated. Nevertheless, we would anticipate that an understanding why different classes of immunity are differentially effective is very important on two accounts. Firstly, it would provide insight into the evolutionary forces driving the development of different classes of immunity and why cell-mediated immunity is sometimes required and a particular isotype of antibody is best able to resolve an insult from a foreign invader. Secondly, it would seem that there must be a meaningful relationship between why a certain class of immunity is most appropriate to contain an insult and the nature of the decision criterion controlling which class of immunity is induced. Only if there is a meaningful relationship will an appropriately effective response usually be mounted. It is noteworthy that, although tuberculosis is the greatest killer world-wide of all infectious diseases, about 95% of those infected with Mycobacterium tuberculosis do not suffer any overt disease because an appropriate immune response is made. We shall shortly consider one such potentially meaningful relationship between the nature of the different classes of immunity and the postulated decision criterion. However, we believe there must be many such relationships. The discussion that follows is based only upon the distinction between cell-mediated and humoral immunity, which is clearly inadequate in the long run and can only be tolerated as the beginning of an analysis. It was shown in the 60s that the binding of two IgG molecules close together on the surface of a cell was required to activate complement to lyse the cell. Observations suggested that the binding of several 100,000 IgG molecules to a red blood cell were required in order that there was a 50% chance of an appropriate IgG doublet forming, leading to complementdependent cell lysis [12]. Somewhat similarly, IgG-mediated cellular cytotoxicity was found only to be effective against target cells that had a very large number of sites recognized by IgG antibody [13]. These observations suggest that antibody is ineffective against cells with only a few (less than 100,000) recognizable sites. Consider a cancer cell. There will be about 100-1000 major peptides able to bind class I MHC molecules generated inside a cancer cell, and there are about 100,000 class I MHC molecules present on a typical cell's surface. There will therefore be about 100 to a 1000 class I MHC molecules bearing a given major peptide on the cancer cell's surface. Most of these peptides will of course be self-peptides. Even if the cancer cell bore several "foreign" peptides, and even if antibody recognized peptide/MHC class I complexes (as has been observed [14]), it seems IgG mediated mechanisms would be ineffective. This makes sense in that cell-mediated immunity is known to be required to contain most tumors, and slowly growing intracellular parasites, which would also have low presentation of antigen on the surface of infected cells. What could be the advantage of this requirement for the formation of an IgG doublet to activate complement? It certainly makes antibody ineffective at times. An advantage might be that in many cases IgG autoantibody, even if generated, is benign. We have suggested that cell-mediated effector mechanisms are effective against cells with a low number of recognized sites, whereas antibody is not. A cell bearing few recognized sites should therefore induce cell-mediated immunity if the immune system is to be effective; antibody, if induced, would not only be ineffective, but might mask the antigen and thereby block cell-mediated effector functions. Consider a cellular antigen with many foreign, recognizable sites, sufficient in number to be susceptible to IgG-dependent attack. In this case, antibody can be effective; the advantage of an IgG response would be that, if any autoantibodies are induced, they are less likely to be damaging than if a corresponding cell-mediated response was generated. Note that this scheme explains both why cancers and infections caused by slowly growing intracellular parasites are only contained by cell-mediated, Thl responses, and why
79
autoimmune reactions tend to be much more damaging if of Thl than Th2 type [ 15-17]. Given this potential reason why cell-mediated and humoral immunity are desirable under different circumstances, it is natural to ask whether we can envisage a decision criterion up to the job of selecting the most appropriate response. The job would be to ensure that cells with a low density of foreign, recognizable sites induce cell-mediated responses, whereas those with a high density can induce antibody responses. This is something to which we shall return. 4.5.
Coherence in the regulation of the immune response to an antigen: the Thl/Th2 nature of the response to the diverse peptides derived from one antigen and recognized by CD4 T cells tends to be coordinately regulated.
There is so much we do not know about how different classes and subclasses of immunity are controlled. Given this ignorance, it is perhaps interesting to ask what would be the best advice to give an individual who wishes to raise different types of immunity to the same hapten. He/she wishes to separately raise hapten-specific IgA, IgE, IgG1, IgG2a antibody, delayed type hypersensitivity and cytotoxic T cells. The best advice we could give this person would be: go to the literature and find conditions under which an antigen of one sort or another can induce the class of immunity desired. Couple the hapten to this carrier antigen and immunize as described. Proceed without thinking. The basis for this advice is that in most cases the isotype of the anti-hapten antibody induced is the same as that of the anti-carrier antibody. In this sense, the response to the different epitopes of the hapten-carrier (h-C) conjugate are coherently regulated. The reason for this coherence is understandable within our current conceptual framework (see Figure 2) primarily due to Lanzavechhia. The Ig receptors of both an anti-hapten and an anti-carrier B cell will bind the hapten carrier conjugate, leading to its endocytosis and processing, giving rise to peptides that associate with class II MHC molecules, and ultimately with the presentation of c l, c2 ..... cn peptides at the B cell's surface. These peptide complexes will be recognized by eTh cells specific for the nominal antigen C, and the isotype induced will depend upon the Thl/Th2 nature of the eTh cells present, this dependence being the same for the anti-h and for the anti-C B cell. Hence we can understand why the immune response to the different B cell epitopes of an antigen consists of antibody of the same isotype.
C-derived peptide~.
hapten 4-~~ " ~!iM
-h class II MHC molecul~
Figure 2. The MHC-restricted model for B cell/Th cell interaction.
carrier(C)
80
Consider a complex antigen Q that yields peptides q l, q2, q3 ..... qn recognized by CD4 Th cells. Many observations suggest the Thl/Th2 nature of the response to these diverse peptides are controlled so that they tend to be similar. A very interesting, instructive example comes from a report made decades before the Thl/Th2 paradigm was formulated. Raffel and Pearson made the observation that antigens with few foreign sites (in modern terms able to produce few q l, q2 ..... qn foreign peptides) could only induce cell-mediated immunity in the form of delayed-type hypersensitivity but were unable to induce antibody. They pointed out, however, that antibody could be raised to such antigens if they were coupled to another antigen with many foreign sites that was itself immunogenic for antibody production. Similarly, we see that the dose employed for immunization of a complex antigen, such as a xenogeneic red blood cell (RBC), is critical in determining the Thl/Th2 nature of the response. Changes in dose of a 100 fold can dramatically change the Thl/Th2 nature of the response (e.g. 6). Given that the anti-RBC response is likely to be pretty heterogeneous in terms of stimulatory peptides, the coordinate change in the Thl/Th2 nature of the response depending on dose would seem to reflect coherence. It is perhaps noteworthy that if all the peptides of a xenogeneic red blood cell, or those generated from an intracellular protozoan parasite such as Leishmania major, were to induce pTh independently, it would be hard to imagine circumstances where there would be responses relatively uniform with respect to their Thl/Th2 phenotype. Such observations suggest that the Thl/Th2 nature of the response to a peptide q2, derived from a nominal antigen Q, is coordinately controlled to be similar to the Thl/Th2 nature of the response to its companion peptides, p l, p3 ..... pn, also derived from Q.
.
5.1.
INTEGRATING MECHANISMS OF pTh CELL ACTIVATION/INACTIVATION WITH ATTRIBUTES/FEATURES OF THE IMMUNE SYSTEM Implications of the Historical Postulate for Peripheral Tolerance at the CD4 T cell level
The original two signal model of lymphocyte activation, proposed in 1970, provided a description for how pTh cells can be activated and inactivated by antigen [18]. This model accounted for peripheral tolerance within the context of The Historical Postulate. Although the original model, conceived in a framework in which T cells directly recognized the nominal antigen, cannot be accepted in detail as originally proposed. We would like to discuss some of the ideas that inspired it as an introduction to a discussion of current models. The basic idea was that a single CD4 pTh cell, specific for an antigen A, was inactivated on interacting with A, through the generation of signal 1 alone, following the interaction of its receptor with antigen. The activation of this resting CD4 pTh cell required the presence of other CD4 T cells specific for A. In one illustrative formulation of the model, the two CD4 T cells interacted via an antigen bridge, see Figure 3. The pTh cell being activated was "informed" of the presence of the second CD4 T cell by the generation of a second signal, signal 2. This resulted in the activation of the pTh cell. Consider the situation with respect to the generation of T cells specific for a peripheral self antigen P that is present early in development and continuously thereafter. The first pTh cell generated will be inactivated upon interacting with P, as there is no second CD4 T cell with which it can cooperate. CD4 T cells specific for P will be inactivated as they are generated one or a few at a time by virtue of the continuous presence of P. Consider a foreign antigen F, that is not continuously present. In this case, CD4 T cells specific for F can accumulate in the absence of F, and once F impinges upon the immune system, F can mediate the cooperation required for activation of the pTh cells.
81
Note a critical point here. The second signal must be very short range, preferably involving a membrane/membrane interaction. If it were not, the cooperation between the anti-F CD4 T cells, involving the release of long range-acting second signal, allow the activation of newly generated pTh cells specific for P in the presence of P. This would violate the first feature we discussed concerning the independence of the inactivation of pTh cells specific for one antigen P from the activation of a pTh cell specific for a non-crossreacting antigen Q above. Is this idea for how peripheral tolerance at the CD4 level can be achieved, embedded in the original two signal model, still pertinent and realizable in modern terms?
receptor antigen
receptor Figure 3. The original two signal model.
We have shown that the primary induction of T cells mediating delayed-type hypersensitivity (DTH) to xenogeneic red blood cells (XRBC) can be helped by T cells specific for a protein P in the presence of the conjugate P-XRBC, but not when P and the XRBC are both present but unlinked [19]. Thus activation appears to require cooperation between T cells that is mediated by the recognition of linked epitopes. Furthermore, univalent peptides recognized by CD4 T cells appear to inactivate them when administered to mice without adjuvant. All this seems very reminiscent of the original two signal model. Our attempt at a modern formulation for the processes of activating and inactivating pTh cells incorporates the ideas on peripheral tolerance and satisfies The Historical Postulate. It is also consistent with modern findings, and is shown in Figure 4 [20]. According to this scheme, the full activation of pTh cells specific for an antigen Q to give rise to eTh cells takes place in a minimum of two steps. In step 1, the APC is a mature dendritic cell or a macrophage which bears constitutively expressed costimulatory molecules. As a result of step 1, pTh cells multiply. They will however die in time unless they complete step 2. In step 2, the step 1 primed pTh cells interact with an APC that is a Q-specific B cell. This B cell can expresses inducible costimulatory molecules. The degree to which they are expressed depends upon the number and state of activation of eTh cells specific for Q and the degree to which such Q-specific eTh cells recognize q/MHC class II molecules on the surface of the B cell, thereby causing the expression of the inducible costimulatory molecules. Thus pTh cell activation requires CD4 T cell/CD4 T cell cooperation, mediated by the recognition of linked epitopes, as we shall see. Suppose the
82
pTh cell recognizes the peptide q derived from Q. Under what circumstances can its activation be helped by an eTh cell that recognizes the peptide r derived from the nominal antigen R that does not crossreact with Q? The B cell will have to be specific for either R or Q, and in order for this B cell to present q and r it will have to endocytose the conjugate Q-R. Thus this model accounts for independence of activation/inactivation of pTh cells belonging to non-crossreacting antigens (see (4.2) above); it explains why there is an operational requirement for the recognition of linked epitopes in the CD4 T cell/CD4 T cell collaboration; and it provides an explanation for peripheral tolerance that is in accord with The Historical Postulate. As discussed elsewhere, the conditions resulting in the induction of autoimmunity are accounted for in a very natural way by the model, and the phenomenon of epitope spreading in T cell autoimmunity seems an inescapable consequence of the original two signal model and of the two step, two signal model, which can be regarded as its contemporary formulation. A significant question not addressed here is where the first eTh cells come from. It would appear that the generation of eTh cells require the activation of pTh cells, but that such pTh cell activation itself requires eTh cells. We have recently discussed this "priming problem", and so we refer the interested reader elsewhere [20]. Finally, we would like to address what the significance of the first step in the activation of pTh cells could be. We have always been somewhat concerned by the postulate that the initiation of a primary immune response requires the interaction of two rare T cells. We have referred to this problem as the "scarcity problem". The first step results in an expansion of the pTh population, and thus perhaps reduces/overcomes the scarcity problem as it is potentially encountered in the second step [20]. 5.2.
The physiological significance of distinct classes of immunity and its relationship to the decision criterion determining the Thl/Th2 nature of the response: The Threshold Hypothesis
We have argued above that antigenic cells with very few "foreign" sites, such as cancer cells and cells infected by slowly growing parasites, can only be contained by a cell-mediated response, and therefore should only induce such a response, see (4.4) above. On the other hand, cellular antigens with very many foreign sites, such as bacteria, can be contained by antibody. Thus antigens minimally foreign should only induce cell-mediated immunity, whereas more foreign antigens can be contained by antibody and so may induce such a response. If the decision criterion is to make physiological sense, there must be a way of measuring the foreignness of an antigen. How could this be achieved? There are few if any pTh or eTh cells specific for peptides derived from nominal self antigens. On the other hand, there are significant numbers of such T cells specific for peptides derived from nominal foreign antigens. In other words, there will be many CD4 T cells specific for very foreign antigens and fewer CD4 T cells specific for minimally foreign antigens. Consider the consequences of these considerations in terms of the model depicted in Figure 4. Figure 4 represents a model limited by its "static" nature. This model suggests that, in the presence of few CD4 T cells, the inducible costimulatory molecules of the APC involved in step 2 will be poorly expressed. In the presence of many CD4 T cells they will be better expressed. We can satisfy the proposition that minimally foreign antigens induce exclusive cell-mediated, Thl responses, if we postulate that the generation of such cells requires the expression of relatively few helper T cell-dependent costimulatory molecules on the antigen-presenting B cell in step 2. The generation of antibody and Th2 cells requires a higher level of expression of such T helper cell dependent costimulatory molecules. This hypothesis is referred to as The Threshold Hypothesis [ 15-17]. This hypothesis is probably more realistic when we consider
83
TcR counter ~ receptorto ( costimulat~) molecule ~)~
\--"--';;':= J Cell ] "~-f~ ~/~
... ~ ~ " ~
___
1~~ if k
~ pTh cell
'
ClarnSo~/lculuC lHe
costimulation
:ive costimulator molecule
STEPONE
TcR counter ~ receptorto ~ c~
/( •
classli \ h~=~rinn ~"~-~;;'= ~C~) ~'="
class !1 / MHcmOlecule / ~ /
~'J
"~
::; ~
~1(~ !
activation~
costimulation'~~ A P C ~ MHC molecule
inducible costimulator molecule
STEPTWO Figure 4. The two step, two signal model. For detailed description, see text.
t/
84
events from a dynamic rather than static standpoint. Thus in the presence of few pTh cells, few eTh cells are generated, and the few eTh cells can only modestly help in the further activation of pTh cells. If there are many pTh cells, they will yield more eTh cells even if only modestly induced, and these greater numbers of eTh cells can act more effectively in the further activation of pTh cells, resulting in a cooperative spiral. Our recent studies have provided striking support for The Threshold Hypothesis in two respects. We find that the number of CD4 T cells present in vivo determines the Thl/Th2 nature of the response when the amount of antigen is held constant. The presence of few CD4 T cells results in an exclusive Thl response, whereas the presence of more results in a mixed Thl/Th2 or predominant Th2 response. Moreover, the Thl/Th2 nature of the response also depends on the size of the antigen challenge. Lowering the antigen dose results in a more exclusive Thl response. Most interestingly, the number of CD4 T cells and the amount of antigen are inter-dependent variables in determining the Thl/Th2 nature of the response: higher numbers of CD4 T cells and a higher amounts of antigen favor the generation of Th2 cells. This interdependence is most naturally explained if the Thl/Th2 nature of the response depends upon CD4 T cell/CD4 T cell interactions mediated through recognition of antigen (N. Ismail and P.A. Bretscher, submitted). This is a cardinal feature of The Threshold Hypothesis. 5.3.
Independence of the Th 1/Th2 nature of concurrent immune responses to non-crossreacting antigens, and the coherence of the immune response to the peptides derived from the same nominal antigen
I remember a remark of Francis Crick's made during a seminar on the nature of scientific enquiry. He said it is important to develop a conceptual framework when making a theory that does not rely upon all the observations available for its formulation. If the framework starts accounting for non-incorporated observations, you have some reason for believing you are getting somewhere. I hope that it is apparent, after a bit of reflection, that The Threshold Hypothesis accounts for the independence of the processes determining the Th 1/Th2 nature of concurrent responses to two non-crossreacting antigens. It also accounts for the coherence seen in the immune response to the different peptides that are derived from each of the nominal antigens. Coherence is explained as the Thl/Th2 nature of the response to q2 depends upon the number/nature of the T cells specific for the peptides q l, q2, q3 ..... qn, as does the Thl/Th2 nature of the response to its companion peptides.
ACKNOWLEDGEMENT This article was supported by a grant from the MRC of Canada. We are grateful to T. Strutt for critically reading the manuscript and her comments.
REFERENCES 1. 2.
Penhale WJ, Farmer A, McKenna RP, Irvine WJ. Spontaneous thyroiditis in thymectomised, irradiated Wistar rats. Clin. Exp Immunol 1973; 15: 225-234. Janeway CA. Approaching the Asymptote? Evolution and Revolution in Immunology. Cold Spring Harbour Lab. Press. 1989; Vol LIV: 1-13.
85
.
.
.
.
.
9. 10. 11. 12. 13.
14. 15. 16.
17.
18. 19.
20.
Medzhitov R, Preston-Hurburt P, Janeway CA. A human homologue of the Drosophila Toll protein signals activation of adaptive immunity. Nature 1997; 388: 394-397. Matzinger P. Tolerance, Danger, and the Extended Family. Ann Rev Immunol 1994; 12: 991-1045. Ridge JP, Fuchs EJ, Matzinger P. Neonatal tolerance revisited: turning on newborn T cells with dendritic cells. Science 1996; 271: 1723-1726. Ismail N, Bretscher PA. The Thl/Th2 nature of concurrent immune responses to unrelated antigens can be indeperndent. J Immunol 1999; 163: 4842-4850. Kappler JW, Roehm N, Marrack P. T cell tolerance by clonal elimination in the thymus. Cell 1987; 49: 273-280. Lederberg J. Genes and antibodies. Science 1959; 129: 1649-1653. Adams TE, Alpert S, Hanahan D. Non-tolerance and autoantibodies to transgenic self antigen expressed in pancreatic beta-cells. Nature 1987; 325: 223-228. Hanahan D. Peripheral antigen-presenting cells in thymic medulla: in self-tolerance and autoimmunity. Curr Opin Immunol 1998; 10: 656-662. Bretscher PA. Contemporary models for peripheral tolerance and the classical "historical postulate". Seminars in Immunology, 2000; in press., 9 pp. Humphrey J, Dourmashkin R. The lesions in cell-membranes caused by complement. Adv Immunol 1969; 11: 75-115. Wiedermann G, Denl H, Stemberger H, Eckersforfer R, Tappeiner G. Influence of antigenicity of target cells on the antibody-mediated cytotoxicity of nonsensitised lymphocytes. Cell Immunol 1975; 17: 440-446. Froscher BG, Klinman NR. Immunisation with SV40-transformed cells yields mainly MHC-restricted monoclonal antibodies. J Exp Med 1986; 164: 196-210. Bretscher PA. Hypothesis: On the Control between Cell-Mediated, IgM and IgG Immunity. Cell Immunol 1974; 13: 171-194. Bretscher PA. An Integration of B and T cells in Immune Activation, in B and T cells in Immune Recognition. F Loor and G Roelants, Eds, John Wiley and Son, 1977; 457-498. Bretscher PA. Quantitative considerations in the design of vaccination strategies against pathogens uniquely susceptible to cell-mediated attack, in Concepts in Vaccine Developments. SHE Kaufmann, Ed, Walter de Gruyer, Publ, 1996; 187-204. Bretscher PA, Cohn M. A theory of self-nonself discrimination. Science 1970; 169: 1042-1049. Tucker MJ, Bretscher PA. T cells cooperating in the induction of delayed-type hypersensitivity act via the linked recognition of antigenic determinants. J Exp Med 1982; 155: 1037-1049. Bretscher PA. A two step, two signal model for the primary activation of precursor helper T cells. Proc Natl Acad Sci 1999; 96:185-190.
This Page Intentionally Left Blank
New Foundation of Biology Edited by I. Berczi and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
87
Regulation of the Immune Response within the Central Nervous System
JACK ANTEL Montreal Neurologic Institute, McGill University, Montreal PQ, Canada
ABSTRACT The human disease post vaccination (or acute disseminated) encephalomyelitis (ADEM) and its animal counterpart experimental autoimmune encephalomyelitis (EAE) demonstrate that the CNS can be the selective target of a self-antigen directed immune response. These disorders are dependent on systemic CD4+ T cell sensitization to CNS antigens. In contrast to ADEM, the human disorder multiple sclerosis (MS), also postulated to reflect CNS directed immune responses, is characterized by its recurrent and or progressive disease course. The above clinical disorders raise issues regarding the role that resident cells of the CNS play in regulating CNS directed immune responses, under physiologic and pathologic conditions. Such participation could occur at the level of the blood brain barrier (BBB) and/or within the parenchyma of the CNS. BBB-lymphocyte interaction-the molecular events that regulate lymphocyte access to the CNS include those involved in adhesion, chemoattraction, and migration through the cellular and extracellular matrix components of the BBB. Using a Boyden chamber assay system as an in vitro model of lymphocyte migration, we could show an increased rate of migration of lymphocytes derived from MS patients compared to controls, through a barrier comprised either of fibronectin alone or of endothelial cells (EC) derived from adult human CNS microvessels. Migration could be partially inhibited by matrix metalloproteinase (MMP) inhibitors and antibodies to MCP-1, the major lymphocyte chemoatractant produced by the ECs. Although the ECs can be induced to express both MHC class II and co-stimulatory molecules (B7-1), they favor induction of T cell anergy rather than proliferation. The perivascular microglia are the fully functional antigen presenting cells (APCs) at the level of the BBB. Parenchymal cell-lymphocyte interactions-within the human adult CNS, microglia can express both MHC class II and co-stimulatory molecules; in vitro studies indicate their capacity to process and present antigen. In contrast, adult human astrocytes can be induced to express only MHC class II molecules. They do not support classical antigen induced T cell proliferation but can support super-antigen induced responses. Parenchymal microglia are a source of the cytokine IL-12 that biases the T cell response toward a Thl phenotype. In context of primary immune-mediated disease, the immune-glial cell network of interactive events is likely initiated by the former (e.g. via CD40-CD40L signaling). In context of neurodegenerative or chronic inflammatory CNS disorders, neural cells may play the central role in initiating or sustaining the response.
88
1.
INTRODUCTION
Although the central nervous system (CNS) continues to be considered a site of relative immune privilege, this compartment can be the site of acute, recurrent, or chronic immune responses. The inflammatory demyelinating disorder recognized to follow systemic administration of the neural tissue containing vaccine developed by Pasteur in the 1880s is considered the prototype of such disorders. This entity is termed post vaccination encephalomyelitis or acute disseminated encephalomyelitis (ADEM). The extensive studies of the animal disorder experimental autoimmune encephalomyelitis (EAE), developed to model ADEM, have established that auto-reactive CD4 T cells either induced by systemic immunization with CNS antigens or adoptively transferred can initiate a CNS inflammatory-demyelinating disease. Similar mechanisms are postulated to underlie the development of the human relapsing/chronic CNS demyelinating disease multiple sclerosis (MS). The above disorders are examples in which the systemic immune system initiates the neuro-immune network. The interactions of the infiltrating lymphoid cells with the resident parenchymal glial cells are central events involved both in the persistence or recurrent aspect of any of these disorders and for the development of the immune effector responses that result in subsequent tissue injury. There is increasing recognition of an immune response contributing to the overall pathology of a number of clinical and experimental disorders that are initiated in the absence of any initial breach of the BBB. Examples would include the neurodegenerative disorders, including Alzheimer's disease. The experimental facial axotomy lesion has provided an experimental model for neuronal degeneration without breach of the B BB. This lesion is associated with a microglia response and recognizable presence of T cells [43]. The leukodystrophies, e.g. adreno-leukodystrophy, characterized by accelerated myelin breakdown consequent to inborn errors of metabolism, also feature an inflammatory response. These observations support the concept that there is ongoing physiologic immune surveillance within the CNS. Activation of the neuro-immune network under these circumstances would presumably, have been initiated by the neural arm of the network. The extent to which these presumed secondary inflammatory responses contribute to the disease course remains to be established. Their observed presence has already led to clinical trials using anti-inflammatory medications. This presentation will describe how the interactions of the constituents of the neural-immune network regulate each other's state of activation. The state of activation of the immune and neural cells within the CNS are important determinants of their contribution to tissue injury either in the context of diseases of presumed primary immuno-pathogenic origin or as a secondary component. Both could be amenable to therapeutic intervention. Throughout, we will also consider that neuro-immune interactions may contribute in a positive manner to tissue repair and regeneration within the CNS. Not to be overlooked is that many of the molecular mediators such as cytokines implicated in contributing to disease development, may have important functions under physiologic conditions where they are present at much lower concentrations. Such dual roles (physiologic and pathologic) add complexity to interpreting the significance of experimental models in which disease is induced in animals in whom such molecules (or their receptors) have been totally depleted. The emphasis of this presentation will be on direct neural immune interactions that occur within the CNS. We will not discuss the important issue of indirect CNS-systemic immune interactions. The latter would include CNS regulation of the systemic immune system via molecules such as neurohormones or neurotransmitters that can act at a distance and systemic immune feedback to the CNS via soluble mediators such as cytokines acting on the hypothalamic pituitary axis. We will consider how immune response within the CNS is regulated by interaction of the
89
immune system with resident cells of the CNS at the level of the BBB and within the CNS parenchyma.
2.
REGULATION OF THE IMMUNE RESPONSE AT THE BLOOD BRAIN BARRIER
The BBB provides a functional barrier that impacts on the movement of cells and soluble molecules from the systemic circulation into the CNS. The major cell constituents of the BBB are the endothelial cells (ECs), the perivascular microglia, and the astrocytes. Serial studies, using gadolinium-enhanced magnetic resonance based imaging of patients with MS, indicate that acute or new lesion formation is associated with disruption of the BBB. The limited available histology of such acute lesions indicates that they are characterized by infiltration with lymphocytes, presumably migrating from the systemic compartment into the CNS. A similar sequence of events can be demonstrated in the EAE model. Furthermore, animal models using adoptive transfer of T cell lines reactive with antigens such as ovalbumin, that are not present within the CNS indicate that such cells can traffic to the CNS. These cells, in contrast to cell lines whose antigens are present in the CNS, such as myelin reactive T cells, do not persist there. We thus need to consider the dual issues of regulation of lymphocyte trafficking and antigen presentation that occurs at the level of the BBB.
3.
REGULATION OF LYMPHOCYTE TRAFFICKING
We have utilized lymphocytes derived from the peripheral blood and endothelial cells (ECs) derived from brain microvessels of adult humans to characterize the properties of these cells that contribute to the process of migration under physiologic and pathologic conditions [39]. Our method for preparing dissociated cultures of such human brain ECs (HBECs) from surgically resected temporal lobe tissue has been described [39] and is similar to that previously utilized by others [16]. The ECs express an array of markers, as for example Von Willebrand factor and Ulex europaeus type 1 (UEA-1) lectin binding sites, expected of this cell type. These ECs grow to confluency and form tight junctions between each other as also expected for this cell type. The permeability to soluble molecules of the barrier formed by the ECs, as tested by rate of passage of radio-labeled albumin, can be further restricted by exposing the ECs to supernatant derived from astrocytes and microglia [42]. We have grown HBECs to confluency on the fibronectin-coated membrane of a Boyden chamber. The Boyden chamber is a dual compartment chamber system separated by a membrane containing different size pores (3 ~t in our system). This system can be used as a functional assay to assess the rate and molecular basis of migration of cells and soluble factors through the barrier created by the ECs. The molecular events involved in the process of lymphocyte extravasation from microvessels into tissue compartments, have been analyzed in detail [3]. The sequence of events includes chemoattraction, adhesion of the lymphocytes to ECs, and migration of cells through or between the ECs and then across the basement membrane/extracellular matrix (ECM). Many of the molecules involved in these processes are differentially expressed under basal and inflammatory conditions. Pro-inflammatory cytokines such as interferon T (IFNT) are potent inducers of adhesion molecules on lymphocytes (LFA-1, VLA-4) and of their ligands on the ECs (ICAM, VCAM), respectively. Antibodies directed against these molecules will inhibit the rate of lymphocyte migration through an EC barrier [1]. The functional importance of these molecules has also been demonstrated using the EAE model in animals in which specific adhesion molecules were
90
selectively depleted by genetic manipulation or by administration of antibody [2, 10]. In MS, there is an apparent up-regulation of adhesion molecules on circulating lymphocytes, as well as of their ligands on ECs in the regions of inflammation in the CNS [28]. Anti-adhesion molecule antibodies, such as VLA-4 antibody (Antegren), are in clinical trial in MS [50]. Our analysis of chemokine expression in HBECs indicates, that under basal culture conditions, the HBECs predominately express the chemokines MCP-1 and IL-8 [41]. The former is a potent chemoattractant for lymphocytes, the latter for neutrophils. After the HBECs are exposed to pro-inflammatory cytokines or pro-inflammatory cytokine producing T cells (Th 1), we could detect a much wider array of chemokine expression. These observations would support the concept that the initial participants in the initial phase of the immune response in an autoimmune disease such as MS are more limited and thus potentially more amenable to therapy. Anti-MCP-1 antibodies have been used to inhibit the development of EAE [26]. In MS, there is an apparent up-regulation of chemokines in the CNS and of chemokine receptors on lymphocytes [4, 30]. The actual migration process of lymphocytes from the vessel into the parenchyma of the CNS requires that there be disruption of the basement membrane surrounding the ECs and of the ECM. We found that the rate of migration of lymphocytes derived from MS patients with active disease, either relapsing or secondary progressive, through either a fibronectin coated membrane or a combined EC/fibronectin coated membrane in a Boyden chamber assay was increased compared to lymphocytes derived from control donors [40, 46]. This increased rate reflected, at least in part, the increased rate of production of matrix metalloproteinases (MMPs) by the MS donor derived lymphocytes [51]. Activated glial cells are also an important source of MMPs, with production being increased in response to chemokine exposure [14]. One action of interferon [3 (IFN[3), the most frequently used therapy of MS, is inhibition of MMP production [51 ].
4.
ANTIGEN PRESENTATION AT THE BBB
As mentioned previously, for T cells to persist in the CNS, they must be presented with their antigen. Since ECs would be the first cell type encountered by T cells destined to migrate from the microvessels into the CNS, there has been interest in establishing whether ECs are capable of serving as competent antigen presenting cells (APCs). Activation of previously naive or resting CD4 T cells in response to peptide antigens requires dual signals. One of these signals is delivered by the antigen/major histocompatibility complex (MHC) class II antigen complex engaging the T cell receptor; the other by co-stimulatory molecules, particularly CD80/CD86 that interact with their receptors B27/CTLA-4 [13, 20]. Under basal culture conditions, there is little or no MHC class II expression on HBECs. These molecules can be induced with proinflammatory cytokines [39]. In situ, MHC class II molecules can be detected on ECs in active MS lesions [49, 53]. The CD86 co-stimulatory molecule is constituitively expressed on HBECs whereas CD80 expression must be induced with pro-inflammatory cytokines [36, 39]. Despite expression of both the requisite MHC and co-stimulatory molecules on HBECs in vitro, we found that these cells were unable to support proliferation of immediately ex vivo heterologous CD4 T cells in a mixed lymphocyte reaction [39]. The T cells seemed to have entered a state of anergy in that addition of IL-2 could restore their proliferative capacity. Our initial data suggests that these results reflect the effects of an active inhibitory factor produced or expressed by the HBECs. Activated T cells have less requirement for co-stimulatory signals. We found that HBECs could support continued proliferation of pre-activated T cells. The constituent cells of the B BB that appear to be the most competent APCs are the perivascular microglial cells. These cells are derived from systemic monocytes. These cells have
91
a relatively high turnover rate with continued replacement by systemic monocytes. In situ studies of MS tissues indicate that these perivascular cells express both MHC class II molecules and co-stimulatory molecules CD80 and 86 [56]. Experimental studies conducted using chimeric animals indicate that histocompatibility between T cells and the perivascular microglia is required for antigen presentation in the CNS and development of EAE [22]. Depletion of perivascular microglia results in systemic T cells accumulating in the perivascular regions of the CNS but failing to migrate into the parenchyma [48]. Pericytes, a cell likely derived from mesenchymal origins, are also implicated as potential APCs.
5.
REGULATION OF THE IMMUNE RESPONSE WITHIN THE CNS PARENCHYMA
The initial studies related to regulation of the immune response within the CNS tended to focus on the role of glial cells as participants in neural immune networks and as immune regulatory cells. There is now also recognition of a role of neurons and possibly oligodendrocytes (OLs), cells usually regarded as targets of the immune response, in such processes.
6.
PARENCHYMAL MICROGLIA
These cells are usually regarded as being bone marrow derived cells that populate the CNS early in development [44]. The rate of turnover and re-population of these cells remains to be established. These can contribute to immune reactivity within the CNS in multiple ways. These include:
7.
MICROGLIA AS REGULATORS OF THE ADAPTIVE IMMUNE RESPONSE
This function refers to the role of microglia as APCs and as sources of cytokines that shape the cytokine production profile of ct[3 receptor bearing T cells. As previously discussed, APCs are required to deliver dual signals (MHC/antigen and co-stimulatory molecules) to produce T cell activation. MHC class II molecules can be detected on parenchymal microglia in situ even in the apparently normal adult human CNS (reviewed in [5]). These molecules are also expressed on microglia immediately ex vivo and under basal culture conditions. This contrasts to microglia derived from the adult rodent CNS or fetal human CNS. Whether these differences reflect altered regulation of gene ex-pression in the adult human CNS or the effects of repeated infectious and non-infectious insults encountered by humans over time remains speculative. MHC class II expression is significantly increased on parenchymal microglia in an inflammatory environment such as is present in an active MS lesion and in vitro upon exposure to proinflammatory cytokines. The co-stimulatory molecule CD86 is constituitively expressed on microglia; CD80 is induced by presence of inflammatory mediators in situ and in vitro [6]. Our functional in vitro studies have shown that adult human microglia can support proliferation of immediately ex vivo CD4 T cells in response to allo-antigen stimulation. The microglia are also able to take up and process antigens such as myelin constituents. In this regard, Katz-Levy et al were able to demonstrate that MBP peptide could be recovered from the MHC class II groove of microglia isolated from animals in the demyelinating phase of disease induced by Theiler murine encephalomyelitis virus [27]. Ford et al, however, have argued that in the graft-versus-host (GVH) model microglia favored induction of T cell apoptosis rather than T cell survival and proliferation [ 18].
92
Polarization of T cells into the cytokine defined phenotypes Thl (IFN~,, TNFc~, IL-2) and Th2 (IL-4, IL-5) is itself subject to cytokine regulation. IL-12 is a potent inducer of the Thl phenotype. IL-12 can be detected in the active lesions of MS and levels are increased in the CSF of such patients [34, 57]. Depletion of IL-12 will prevent development of EAE [12]. Microglia are the apparent resident cells of the CNS that are capable of producing IL-12. Such production can be induced via signaling through receptors expressed on microglia that interact with infiltrating immune cells and also, as discussed later, by primary events occurring within the CNS. Signaling via CD40 expressed on microglia and its ligand CD154 expressed on lymphocytes is a central event in microglia: T cell interaction. In our studies of adult human CNS derived microglia in vitro, we could detect CD40 expression on these cells even under basal culture conditions [7]. Levels were significantly up regulated on microglia that were exposed to pro-inflammatory cytokines. CD154 is expressed by activated but not resting T cells. We found that co-incubating activated T cells with microglia induced IL-12 production and that this production could be blocked with anti-CD40 antibodies. Blocking CD40:CD154 interactions has been shown to inhibit development of EAE [11, 23]. Anti-CD154 antibodies are now being tested in clinical trials of MS. Microglia are a source of multiple other cytokines including IL-10 and IL-15, as well as chemokines, that regulate the response of cells of the adaptive immune system [ 11].
8.
MICROGLIA AS CONSTITUENTS OF THE INNATE IMMUNE SYSTEM
Constituents of the innate immune system recognize antigens via germ line encoded receptors rather than via rearranged receptors, as is the case with cells (ct[3 T cells, B cells) of the adaptive immune system. Members of the innate immune system that include monocytes and macrophages can respond rapidly to novel events in the environment without the need for pre-existent memory. Microglia, in a manner akin to that of their monocyte/macrophage counterparts can interact with both endogenous and exogenous molecules present in their environment. Microglia express CD14, the receptor for lipopolysaccharide, a component of the cell wall of a number of gram-negative bacteria [8]. We could demonstrate that exposing human adult CNS derived microglia to LPS in vitro induced IL-12 production. Our data further suggested that LPS, as well as T cell induced, IL-12 production by microglia required a second signal that could be provided by TNF since IL-12 production was inhibited by soluble TNF receptor. Microglia also have receptors for apoptotic cells and for an array of proteins such as [3 amyloid that accumulate in specific neurodegenerative disease states. Scavenger receptors are implicated in lhese processes although animals lacking such receptors do not appear to be deficient in clearing [3 amyloid [15, 17, 24]. Microglia are able to actively phagocytose debris within the CNS, process the material, and present the processed antigen to immune cells. This capacity of microglia to function as members of the innate immune system provides the CNS with the capacity to initiate the neuroimmune cascade.
9.
MICROGLIA AS EFFECTORS OF THE IMMUNE RESPONSE
This function a n be viewed in terms of direct and indirect effects. The former refers to the array of molecules that these cells produce that can potentially injure resident CNS populations including OLs and neurons. These molecules include effector cytokines such as TNF, proteases,
93
excitotoxins, complement components, and reactive oxygen species. Although such effector molecules lack specificity, selective target injury could still occur as a consequence of the properties of the target cells. For example, we have generated data, consistent with that reported by others, that OLs are relatively more susceptible to injury mediated by TNF than are neurons (reviewed in [38]). Axons and neuronal cell bodies may be differentially susceptible to excitotoxins mediated injury dependent on distribution of receptors for such mediators. The indirect effects of microglia on tissue injury refers to the capacity of molecules produced by microglia to up-regulate receptors for potentially injurious mediators on target cells. In this regard we have found that TNF will increase fas expression on OLs making these cells more susceptible to fas mediated injury [37]. Microglia express a number of cell surface receptors that promote their interaction with constituents of the adaptive immune system and thus result in a means whereby the non specific effector molecules produced by microglia may have target selectivity. This is exemplified by the interaction of microglia via their Fc receptors with either antibody or immune complexes. Antibodies to an array of antigens associated with oligodendrosytes, particularly myelin oligodendrocyte glycoprotein (MOG), are present in MS lesions [19]. The specific antibody would bind via its hypervariable regions to the specific target cell while the Fc region of the molecule would engage the microglia via the latters' Fc receptors. This process is termed antibody dependent cell cytotoxicity (ADCC). Engagement of the Fc receptors on microglia by antibody can result in activation of the cells and release of an array of effector molecules. We could demonstrate this by use of immune complexes comprised of red blood cells (RBC) and anti-RBC antibodies or by use of myelin coated with antibody. In both circumstances, we could show marked elevation in levels of cytokine production by microglia and production of reactive oxygen species [52]. Again in context of MS, both myelin debris and anti-myelin antibodies are detected in the lesion sites as are activated macrophages/microglia. Microglia also contain receptors for complement.
10.
MICROGLIA AS PROMOTERS OF REGENERATION
The previous discussion has focussed on the potential contribution of microglia to immune mediated injury response in the CNS. The converse consideration is that microglia may contribute to re-modeling and regeneration following disease or injury. There are experimental paradigms in which microglia or their soluble factors can promote such occurrences. Studies in which microglia are selectively depleted in the CNS without depletion of systemic macrophages remain to be carried out. Applying inducible knock-out techniques to these slowly or non-dividing cells remains problematic. As previously mentioned, anti-inflammatory agents are increasingly being considered for use as therapies for neurodegenerative diseases, particularly Alzheimer's disease. Similar strategies are being explored in cases of stroke and trauma. The effects of these therapies on long-term recovery will need to be carefully evaluated. Whether the profile of molecules produced by microglia differs in response to immune or CNS environmental stimuli remains to be established. Specific chemokines such as fractalkine [47, 58] which are shown to have neuroprotective properties, down-regulate the activity of microglia. IFN [3 and ~, induce production of different chemokines by microglia [31 ].
94
11.
ASTROCYTES AND IMMUNE REGULATION
Astrocytes have been implicated to contribute to immune reactivity in a parallel manner to that described for microglia. Variables related to species, age, and state of activation again need be considered. Although astrocytes can be induced to express MHC class II molecules in situ, in the inflamed human CNS expression is less apparent than on microglia. CD80/86 co-stimulatory molecules have been detected on rodent but not human CNS derived astrocytes [25, 55]. In our studies with human fetal astrocytes, we found that these cells were not able to support proliferation of immediately ex vivo allogenic T lymphocytes. They could support proliferation of previously activated lymphocytes. In contrast to the results obtained with conventional antigens, we found that the fetal human astrocytes could present superantigens to T cells with resultant proliferation [21]. A previous report suggested murine astrocytes could not fulfill this function [45]. Astrocytes are an acknowledged source of cytokines with differences being noted between fetal and adult CNS derived cells. For example, TNF production is more readily apparent in fetal cells compared to their adult counterparts. Recently signaling via fas has been linked with cytokine production in fetal astrocytes in contrast to other cell types in which such signaling usually induces a cell death program [29]. Astrocytes can contribute to protection and recovery from immune mediated injury by several mechanisms. These cells can protect OL and neuron targets from injury mediated by free radicals by actively removing such molecules from the environment [35]. Astrocytes are a source of growth factors including CNTF which is shown to be neuro-protective for TNF mediated injury of OLs. The cytokine IFN[3 is an inducer of NGF on astrocytes [9].
12.
NEURONS AND OLS
These cells are usually considered as targets of the immune response within the CNS rather than regulators of the response. Recent studies indicate these cells may have a more active role. As mentioned there is recognition that neuronal injury in context of an intact BBB can still be associated with a microglial response. There is evidence that the signaling may involve cytokine production by the neurons in addition to cell-cell dependent interactions. Neuronal expression of MHC class I molecules has been linked with interruption of their electrical activity indicating how the state of the neurons can determine its susceptibility to immune mediated responses [33]. In a similar context OLs under pathologic conditions may express antigens that can underlie development of on going autoreactive immune responses. One candidate family of antigens of this type would be members of the heat shock or stress family. One such antigen, named alpha [3 crystallin is implicated in the MS disease process [54].
13.
S UMMARY
The resident cells of the CNS and the cells of the immune system are increasingly recognized to produce and respond to a wide array of common cell surface and soluble molecules. These shared molecular properties promote the capacity of these cell types to interact and regulate each other' s activities. These neural immune interactions can be initiated by either the cells of the immune system or by neural cells responding to events within the central nervous system. Although initial emphasis has been on the role of these interactions in promoting immune
95
mediated injury within the CNS, the converse that these responses promote neural repair and regeneration must also be considered. The observation that myelin basic protein (MBP) reactive T cells can promote repair from CNS injury indicates the contrasting spectrum of effects that such cells can exert [32]. The opportunities now exist to manipulate the neural immune interactions for therapeutic purposes. Given the capacity of autoreactive T cells to reach the site of injury within the CNS, the challenge would be to have such cells deliver molecules exerting positive effects. T cells are already shown to be sources of cytokines and neurotrophins that may be of potential benefit to the diseased or injured CNS. Even higher levels could be genetically engineered into these cells. Advances in neuroimaging will provide the means to monitor the effect of therapy.
REFERENCE 1. 2. 3. 4.
5. 6. 7.
8. 9. 10.
11. 12.
13.
14.
Antel JP, Biernacki K. Migratory Behavior of Thl and Th2 T Cells across Human Brain Derived Endothelial Cells. Neurology 2000; 54. Antel JP, Owens T. The attraction of adhesion molecules [editorial]. Ann Neurol 1993; 34, 123-124. Baggiolini M. Chemokines and leukocyte traffic. Nature 1998; 392, 565-568. Balashov KE, Rottman JB, Weiner HL and Hancock WW. 1999; CCR5(+) and CXCR3(+) T cells are increased in multiple sclerosis and their ligands MIP-lalpha and IP-10 are expressed in demyelinating brain lesions. Proc Natl Acad Sci USA 1999; 96, 6873-6878. Becher B, Prat A, Antel JP. The brain-immune connection: immunoregulatory properties of CNS-resident cells. Glia 2000; 29: 293-304. Becher B and Antel JP. Comparison of phenotypic and functional properties of immediately ex vivo and cultured human adult microglia. Glia 1996; 18, 1-10. Becher B, Blain M. and Antel J. CD40 engagement stimulates IL-12 p70 production by human microglial cells: basis for Thl polarization in the CNS. J Neuroimmunol 2000; Ref Type: In Press. Becher B, Fedorowicz V and Antel JP. Regulation of CD14 expression on human adult central nervous system-derived microglia. J Neurosci Res 1996; 45, 375-381. Boutros T, Croze E and Yong VW. Interferon-beta is a potent promoter of nerve growth factor production by astrocytes. J Neurochem 1997; 69, 939-946. Brocke S, Piercy C, Steinman L, Weissman IL and Veromaa T. Antibodies to CD44 and integrin alpha4, but not L-selectin, prevent central nervous system inflammation and experimental encephalomyelitis by blocking secondary leukocyte recruitment. Proc Natl Acad Sci USA 1999; 96, 6896-6901. Chabot S., Williams G, Hamilton M, Sutherland G and Yong VW. Mechanisms of IL-10 production in human microglia-T cell interaction. J Immunol 1999; 162, 6819-6828. Constantinescu CS, Wysocka M, Hilliard B et al.. Antibodies against IL-12 prevent superantigen-induced and spontaneous relapses of experimental autoimmune encephalomyelitis. J Immunol 1998; 161, 5097-5104. Cross AH, Girard TJ, Giacoletto KS et al.. Long-term inhibition of murine experimental autoimmune encephalomyelitis using CTLA-4-Fc supports a key role for CD28 costimulation [see comments]. J Clin Invest 1995; 95, 2783-2789. Cross AK and Woodroofe MN. Chemokine modulation of matrix metalloproteinase and TIMP production in adult rat brain microglia and a human microglial cell line in vitro. Glia 1999; 28, 183-189.
96
15.
16.
17.
18. 19.
20.
21. 22. 23.
24.
25.
26.
27.
28.
29. 30.
31. 32.
Devitt A, Moffatt OD, Raykundalia C, Capra JD, Simmons DL and Gregory CD. Human CD14 mediates recognition and phagocytosis of apoptotic cells [see comments]. Nature 1998; 392, 505-509. Dorovini-Zis K and Huynh HK. Ultrastructural localization of factor VIII-related antigen in cultured human brain microvessel endothelial cells. J. Histochem. Cytochem 1992; 40, 689-696. E1 Khoury J, Hickman SE, Thomas CA, Cao L, Silverstein SC and Loike JD. Scavenger receptor-mediated adhesion of microglia to beta-amyloid fibrils [see comments]. Nature 1996; 382, 716-719. Ford AL, Foulcher E, Lemckert FA and Sedgwick JD. Microglia induce CD4 T lymphocyte final effector function and death. J Exp Med 1996; 184, 1737-1745. Genain CP, Cannella B, Hauser SL and Raine CS. Identification of autoantibodies associated with myelin damage in multiple sclerosis [see comments]. Nat Med 1999; 5, 170-175. Girvin AM, Dal Canto MC, Rhee L et al.. A critical role for B7/CD28 costimulation in experimental autoimmune encephalomyelitis: a comparative study using costimulatory molecule-deficient mice and monoclonal antibody blockade. J Immunol 2000; 164, 136-143. Hassan-Zahraee M, Ladiwala U, Lavoie PM et al..Superantigen presenting capacity of human astrocytes. J Neuroimmunol 2000; 102, 131-136. Hickey WF and Kimura H. Perivascular microglial cells of the CNS are bone marrowderived and present antigen in vivo. Science 1988; 239, 290-292. Howard LM, Miga AJ, Vanderlugt CL et al.. Mechanisms of immunotherapeutic intervention by anti-CD40L. CD154) antibody in an animal model of multiple sclerosis. J Clin Invest 1999; 103, 281-290. Huang F, Buttini M, Wyss-Coray T et al.. Elimination of the class A scavenger receptor does not affect amyloid plaque formation or neurodegeneration in transgenic mice expressing human amyloid protein precursors. Am J Pathol 1999; 155, 1741-1747. Karandikar NJ, Vanderlugt CL, Eagar T, Tan L, Bluestone JA and Miller SD. Tissuespecific up-regulation of B7-1 expression and function during the course of murine relapsing experimental autoimmune encephalomyelitis. J Immunol 1998; 161,192-199. Karpus WJ and Kennedy KJ. MIP-lalpha and MCP-1 differentially regulate acute and relapsing autoimmune encephalomyelitis as well as Thl/Th2 lymphocyte differentiation. J Leukoc Biol 1997; 62, 681-687. Katz-Levy Y, Neville KL, Girvin AM et al.. Endogenous presentation of self myelin epitopes by CNS-resident APCs in Theiler's virus-infected mice [see comments]. J Clin Invest 1999; 104, 599-610. Keszthelyi E, Karlik S, Hyduk S et al.. Evidence for a prolonged role of alpha 4 integrin throughout active experimental allergic encephalomyelitis. Neurology 1996; 47, 1053-1059. Lee SJ, Zhou T, Choi C, Wang Z and Benveniste EN. Differential regulation and function of Fas expression on glial cells. J Immunol 2000; 164, 1277-1285. McManus C, Berman JW, Brett FM, Staunton H, Farrell M and Brosnan CF. MCP-1, MCP-2 and MCP-3 expression in multiple sclerosis lesions: an immunohistochemical and in situ hybridization study. J Neuroimmunol 1998; 86, 20-29. McManus CM, Liu JS, Hahn MT et al.. Differential induction of chemokines in human microglia by type I and II interferons. Glia 2000; 29, 273-280. Moalem G, Leibowitz-Amit R, Yoles E, Mor F, Cohen IR and Schwartz M. Autoimmune
97
T cells protect neurons from secondary degeneration after central nervous system axotomy. Nat Med 1999; 5, 49-55. 33. Neumann H and Wekerle H. Neuronal control of the immune response in the central nervous system: linking brain immunity to neurodegeneration. J Neuropathol Exp Neurol 1998; 57, 1-9. 34. Nicoletti F, Patti F, Cocuzza C et al.. Elevated serum levels of interleukin-12 in chronic progressive multiple sclerosis. J Neuroimmunol 1996; 70, 87-90. 35. Noble PG, Antel JP and Yong VW. Astrocytes and catalase prevent the toxicity of catecholamines to oligodendrocytes. Brain Res 1994; 633, 83-90. 36. Omari K and Dorovini-Zis K. Expression of the co-stimulatory molecules B7 and LFA-3 by cerebral endothelium is important for effective T cell priliferation. J Neuroimmunol 1998; 90. 37. Pouly S, Becher B, Blain M, Antel JP. 2000; Interferon modulates human oligodendrocyte susceptibility to fas-mediated apoptosis. J Neuropathology Exp Neurology 59, 280-286. 38. Pouly S and Antel JP. Multiple sclerosis and central nervous system demyelination. J Autoimmun 1999; 13,297-306. 39. Prat A, Biernacki B, Becher B, Antel JP. B7 expression and antigen presentation by human brain endothelial cells: requirement for pro-inflammatory cytokines. J Neuropathology Exp Neurology 2000; 59, 129-136. 40. Prat A, A1 Asmi A, Duquette P and Antel JP. Lymphocyte migration and multiple sclerosis: relation with disease course and therapy [In Process Citation]. Ann Neurol 1999; 46, 253-256. 41. Prat A, B iernacki K, Poirier J, Duquette P and Antel JP. submitted; Migration of Multiple sclerosis lymphocytes through brain endothelium. Brain. 42. Prat A, Biernacki K, Pouly S, Nalbantoglu J, Couture R and Antel JP Kinin B 1 Expression and Function on Human Brain Endothelial Cells. J Neuropathol Exp Neurol-in press 43. Raivich G, Jones LL, Kloss CU, Werner A, Neumann H and Kreutzberg GW. Immune surveillance in the injured nervous system: T-lymphocytes invade the axotomized mouse facial motor nucleus and aggregate around sites of neuronal degeneration. J Neurosci 1998; 18, 5804-5816. 44. Rezaie P and Male D. Colonisation of the developing human brain and spinal cord by microglia: a review. Microsc Res Tech 1999; 45,359-382. 45. Rott O, Tontsch U and Fleischer B. Dissociation of antigen-presenting capacity of astrocytes for peptide-antigens versus superantigens. J Immunol 1993; 150, 87-95. 46. Stuve O, Dooley NP, Uhm JH et al.. Interferon beta-lb decreases the migration of T lymphocytes in vitro: effects on matrix metalloproteinase-9. Ann Neurol 1996; 40, 853-863. 47. Tong N, Perry SW, Zhang Q et al.. Neuronal fractalkine expression ~n HIV-1 enc~ halitis: roles for macrophage recruitment and neuroprotection in ~he central nervous sy~ -m. J Immunol 2000; 164, 1333-1339. 48. Tran EH, Hoekstra K, van Rooijen N, Dijkstra CD and O,~-ns T. Immune invasion of the central nervous system parenchyma and experiment ~" ,lergic encephalomyelitis, but not leukocyte extravasation from blood, are prevented i~ ~acrophage-depleted mice. J Immunol 1998; 161, 3767-3775. 49. Traugott U, Scheinberg LC and Raine CS. On the presence of !a-positive endothelial cells and astrocytes in multiple sclerosis lesions and its relevance to antigen presentation. J Neuroimmunol 1985; 8, 1-14. 50. Tubridy N, Behan PO, Capildeo R et al.. The effect of anti-alpha4 integrin antibody on
98
51.
52.
53.
54.
55.
56. 57.
58.
brain lesion activity in MS. The UK Antegren Study Group [see comments]. Neurology 1999; 53,466-472. Uhm J, Dooley NP, Stuve O, Francis G, Duquette P, Antel JP, Young VW. Migratory behavior of lymphocytes isolated from MS patients: effects of interferon 115 therapy. Ann Neurol 1999; 46, 319-324. Ulvestad E, Williams K, Matre R, Nyland H, Olivier A and Antel J. Fc receptors for IgG on cultured human microglia mediate cytotoxicity and phagocytosis of antibody-coated targets. J Neuropathol Exp Neurol 1994; 53, 27-36. van der MK, Hinojoza JR and Sobel RA. Endothelial cell class II major histocompatibility complex molecule expression in stereotactic brain biopsies of patients with acute inflammatory/demyelinating conditions. J Neuropathol Exp Neurol 1999; 58, 346-358. van Noort JM, van Sechel AC, van Stipdonk MJ and Bajramovic JJ. The small heat shock protein alpha B-crystallin as key autoantigen in multiple sclerosis. Prog Brain Res 1998; 117,435-452. Williams K, Dooley N, Ulvestad E, Blain M, Young VW, Antel JP. Antigen presentation by human fetal astrocytes with the cooperative effects of microglia or the microglial-cell derived cytokine II-1. J Neurosci 1995; 15, 1869-1875. Williams K, Ulvestad E and Antel JP. B7/BB-1 antigen expression on adult human microglia studied in vitro and in situ. Eur J Immunol 1994; 24, 3031-3037. Windhagen A, Newcombe J, Dangond F et al.. Expression of costimulatory molecules B7-1 (CD80), B7-2 (CD86), and interleukin 12 cytokine in multiple sclerosis lesions. J Exp Med 1995; 182, 1985-1996. Zujovic V, Benavides J, Vige X, Carter C and Taupin V. Fractalkine modulates TNF-alpha secretion and neurotoxicity induced by microglial activation. Glia 2000; 29, 305-315.
New Foundation of Biology Edited by I. Berczi and R.M. Gorczynski 9 2001 ElsevierScience B.V. All rights reserved
99
Regulatory Circuits of the Pituitary Gland
LUCIA STEFANEANU
Department of Laboratory Medicine, St. Michael's Hospital, University of Toronto, Toronto, Ontario
ABSTRACT Hormones are messengers that enable the communication among the nervous, endocrine and immune systems, in order to maintain homeostasis. The pituitary gland produces hormones with multiple functions, including stimulation of peripheral endocrine glands, i.e. thyroid, adrenals, and gonads, of body growth, lactation, and several metabolic processes. Pituitary hormones are also playing an integrating role in the function of the immune system. According to the classic concept, the six anterior pituitary hormones, namely growth hormone (GH)~ prolactin (PRL), adrenocorticotropin (ACTH), thyroid stimulating hormone (TSH), and gonadotropins (FSH and LH) are produced by five pituitary cell types represented by somatotrophs, lactotrophs, corticotrophs, thyrotrophs, and bihormonal gonadotrophs. The hormone production and proliferation of pituitary cells are controlled by hypothalamic releasing and inhibiting hormones as well as peripheral target hormones. It is well established that GH, PRL, and TSH are involved in the stimulation of immune responses, whereas ACTH in the depression of immune responses. The GH production by somatotrophs is stimulated by growth hormone-releasing hormone (GHRH) and inhibited by somatostatin (SRIF), both produced by hypothalamus. GH is released into circulation and stimulates the liver and other tissues including hematopoietic cells to produce insulin-like growth factor I (IGF-I). IGF-I has a stimulation effect on the size of lymphoid organs. Pituitary PRL secretion by lactotrophs is under tonic inhibition by hypothalamic dopamine. Several candidates for PRL releasing factor (PRF) such as vasoactive intestinal peptide (VIP), thyrotropin stimulating hormone (TRH), galanin, oxytocin and prolactin-releasing peptide have been proposed, but a physiologic PRF has not been identified. TSH production by thyrotrophs is stimulated by hypothalamic TRH, and inhibited by SRIF. ACTH production by corticotrophs is stimulated by corticotropin releasing hormone (CRH), and in some species by arginin vasopressin (AVP), which is co-localized with CRH in the hypothalamus. In response to host stress, corticotrophs integrate peripheral and brain signals and release ACTH that stimulates adrenal glucocorticoid release, followed by immunosuppression. Besides hormones, pituitary cells synthesize many growth factors including cytokines known to regulate growth and differentiation of hematopoietic and inflammatory cells. They comprise interleukins, leukemia inhibitory factor, macrophage migration inhibitory factor, epidermal growth factor, transforming growth factors, fibroblastic growth factors, nerve growth factor, galanin, IGFs, activin, and inhibins. The cytokines may be released into circulation or locally
100
acting directly on hormone producing cells and adding an additional level of pituitary control. An intrapituitary network of cytokines is induced in the acute phase of septic shock, in addition to the circulating, peripherally derived cytokines. Recently, mice lacking hormones, cytokines, or their receptors have been produced by genetic manipulation, helping to better understand their role in the cross-talk between immune and neuroendocrine system.
1.
INTRODUCTION
The nervous, endocrine and immune systems interact in order to maintain homeostasis. Among endocrine glands, the pituitary occupies a key position by integrating central and peripheral signals, allowing it to exercise its "master" role. The pituitary or hypophysis is a small, bean-shaped endocrine gland weighing 500-600 mg. It is situated at the base of the brain in a depression of the sphenoid bone, called the sella turcica. The pituitary is divided into the anterior pituitary (adenohypophysis) and the posterior pituitary (neurohypophysis). The adenohypophysis, representing 80% of the gland, is composed mainly of hormone producing cells that secrete growth hormone (GH), prolactin (PRL), adrenocorticotropin (ACTH), thyroid stimulating hormone (TSH), follicle stimulating hormone (FSH) and luteinizing hormone (LH). A special cell type, the folliculo-stellate cells named after their shape, do not contain hormones, and may play multiple roles, including paracrine functions. The neurohypophysis stores two hormones produced by the hypothalamus, arginine vasopressin and oxytocin. Pituitary hormones exert multiple functions, including stimulation of peripheral endocrine glands, i.e. thyroid, adrenals, and gonads, of body growth, lactation, and several metabolic processes. Pituitary hormones also play an integrating role in the function of the immune system. It is well established that GH, PRL, and TSH are involved in the stimulation of immune responses, whereas ACTH can cause depression of immune responses. The hormone production and proliferation of pituitary cells are controlled by hypothalamic releasing and inhibiting hormones as well as peripheral target hormones. During the last decade, it became evident that besides the five classic hormones, the pituitary produces some of its own regulators, including hypothalamic and peripheral hormones, as well as an expanding list of growth factors. It also synthesizes cytokines- molecular mediators of inflammatory responses. These locally produced mediators have the capacity to alter pituitary cell functions in an autocrine or paracrine manner, or can be released into the circulation to act as hormones on distant target cells. Accumulating evidence indicates that pituitary hormones are also produced by target cells, including immune cells, where they act as growth factors in an autocrine or intracrine fashion. Based on the present knowledge, the regulation of pituitary cells is extremely complex, and not completely understood, involving hypothalamic hormones, peripheral hormones and locally produced mediators. In this review, the focus is on regulatory mechanisms governing the pituitary, with special emphasis on the present knowledge on the control of pituitary hormones and pituitary cytokines involved in the immune function. The regulation of hormones and cytokines produced by immune cells are not included in this review.
101
2.
PITUITARY HORMONES
2.1.
GH
GH is a single 191-amino acid chain that belongs to the GH/PRL family, composing at least 18 distinct genes expressed in pituitary, uterus, placenta and other tissues [ 1]. In the pituitary, GH is produced by somatotrophs and released into circulation. GH stimulates the liver to secrete IGF-I, which mediates in part the GH effects on target tissues. Although recent studies showed that the pituitary is not the only source of GH, the contribution of extrapituitary tissues to blood GH must be very low. The hypothalamic control of somatotrophs is exercised by the stimulating hormone, growth hormone releasing hormone (GHRH), and the inhibiting hormone, somatostatin (SRIF). GHRH is secreted in two forms, a 44-and 40-amino acid peptide encoded by a gene that has been sequenced and localized in humans to chromosome 20. GHRH belongs to the glucagon family of peptides [2]. The receptors for these peptides are also related and grouped into family B of the G protein-coupled receptor superfamily. The coupling of GHRH with its receptor increases the intracellular level of cAMP [3]. The human GHRH receptor (GHRH-R) gene was found to generate 3 receptor isoforms, and in rat 2 isoforms are generated by differential splicing [4, 5]. In the rat pituitary, only the short isoform, which is predominant, signals through a cAMPmediated pathway, leading to activation of protein kinase A, followed by phosphorylation of a cAMP responsive element binding protein (CREB). CREB transactivates the pituitary-specific transcription factor (Pit-1) gene promoter. Pit-1 transactivates the GH gene promoter and provides the final pathway that leads to GH secretion [6, 7]. The signaling by which GHRH induces somatotroph proliferation is less clear. GHRH inhibits the production of its own receptor by a receptor-mediated, cAMP-dependent reduction of GHRH-R mRNA level. A series of structurally diverse growth hormone-releasing substances were synthesized during last years that are different from the naturally occurring GHRH. They include GH releasing peptides (GHRPs) and mimetics such as MK-0677. GH releasing substances proved to act via a common receptor, GH secretagogue receptor (GHS-R) which is distinct from the GHRH-R. GHS-R is present in the hypothalamus and pituitary. The determination of the coding sequence of GHS-R led to the identification of a family of related receptors highly conserved in evolution [8], and recently of the natural ligand of GHS-R. Kojima et al. isolated the ligand from stomach extract, and found it to be a 28 amino acid peptide in which the serine 3 is n-octanoylated [9]. The identification of this peptide, called ghrelin, raises many questions related to the regulation of GH, and its significance in GH-related disorders. SRIF is produced by the anterior periventricular nuclei of the hypothalamus as a cyclic peptide of 14 (SRIF-14) and 28-amino acids (SRIF-28). The mechanisms by which SRIF inhibits GH secretion and somatotroph proliferation are complex and poorly understood. The biologic effects of somatostatin are mediated by specific membrane-bound, high affinity G-protein coupled receptors. Five distinct somatostatin receptor subtypes encoded by different genes [10] are present in the normal rat pituitary [11-13], whereas in the human gland only SSTR1, 2 and 5 are found [14]. SSTR1-4 bind both SRIF-14 and SRIF-28 with high affinity, while SSTR5 has high affinity for SRIF-28 [14-16]. The role of each SSTR subtype in the activity of pituitary cells is just emerging. In an in vitro system, the absence of SSTR1 did not influence GH release [14]. The GH response to SRIF did not correlate with the presence or absence of SSTR5 mRNA [17], despite the fact that both SSTR2 and 5 inhibited GH release from cultured human GH secreting adenomas and rat tumor cells [ 18, 19].
102
IGF-I modulates pituitary GH production by acting at hypothalamic and pituitary levels. IGF-I acting via IGF-I subtype I receptor inhibits GHRH release, and pituitary GH synthesis and secretion [20-22]. GH regulates its own secretion by acting directly on central nervous system areas, and on hypothalamic neurons producing GHRH and SRIF via a specific receptor (GH-R). GH-R has a long and a short isoform and belongs together with PRL-receptor to the family of helix bundle peptide (HBP) cytokine receptors. GH-R mRNA and protein were demonstrated in rat somatotrophs. However there is no proof that GH affects GH secretion by acting directly on somatotrophs [23]. Glucocorticoids act at multiple levels of the GH pathway [24]. They act on SRIF and GHRH producing nuclei of the hypothalamus [25]. Peripherally, glucocorticoids reduce circulating IGF-I level, and decrease IGF-I and GH receptors. At the pituitary level, GH gene transcription is stimulated or inhibited by glucocorticoids, depending on the species, duration of treatment, and experimental conditions. A single injection of dexamethasone causes an early stimulation of GH secretion followed by a late inhibitory effect in normal subjects [26]. Chronic exposure to glucocorticoids exerts a growth-suppressive effect [27]. Children chronically treated with glucocorticoids have decreased growth rate. However, in vitro, pituitary cells from adrenalectomized rats have down-regulated GHRH-R binding sites, and dexamethasone restores GHRH binding sites [28]. A positive element in the promoter region of GHRH-R gene was found [29], explaining the capability of glucocorticoids to enhance pituitary responsiveness to GHRH. Thyroid hormones are essential for growth, and they affect the GH axis at different levels [30]. Thyroidectomy decreases markedly pituitary GH and GHRH-R mRNAs, which can be partially reversed by T 4 t r e a t m e n t . No putative thyroid hormone receptor-response element was yet identified in the GHRH-R 5'-flanking region. Sex differences are apparent in many species regarding the rate of somatic growth, and in GH axis. A negative estrogen receptor-responsive element was found in GHRH-R, explaining the differences in GHRH-R mRNA levels between sexes. Estrogen can also affect GH gene transcription interacting with transcription factor Spl. [31]. Numerous data support the view that testosterone enhances GH secretion. Castration diminishes GH secretion in rats, and the effect is reversed by testosterone replacement [32]. The presence of androgen receptor in GH immunoreactive cells suggests a direct effect of testosterone on somatotrophs. The role of GH in the immune system was deduced from the parallels between thymus changes and circulating GH and IGF-I levels. The maximum thymic size and highest blood GH levels are attained at puberty, and both decline with age. In the aged rat, the implantation of pituitary GH secreting GH 3 cells reversed age-related thymic atrophy and increased the number and function of T cells in the thymus [33]. Transgenic mice overexpressing bovine GH or human GHRH, have elevated blood levels of GH and IGF-I, and develop gigantism with enlargement of internal organs, particularly of the thymus and spleen. The mitogenic responses of splenocytes to immune challenges are significantly increased in such mice [34]. However in humans the role of GH in immune function is controversial. The main argument against it is the fact that GH-deficient children are not immunodeficient and GH administration has no effect of lymphoid tissues [35]. The discrepancy between rodents and humans may be due to differences in the level of circulating GH and locally produced GH. It may be that in man, where the blood GH level is much lower than in rat, the locally produced GH may compensate for the lack of endocrine GH [36]. GH promotes the antibody response, and antagonizes the immunosuppressive effect of ACTH. IGF mediates in part the biological effects of GH. IGF-I is produced almost ubiquitously. This is why it is difficult to distinguish between the effects of circulating and locally produced IGF-I [36].
103
2.2.
PRL
PRL is a 198 single-chain peptide encoded by a gene located on chromosome 6 in man. PRL belongs to the family of GH/PRL hormones. Pituitary production of PRL is performed by lactotrophs. The control of PRL secretion is complex and not completely understood [37]. PRL secretion is under tonic inhibition by hypothalamic dopamine (DA). DA acts via DA subtype 2 receptor (D2R). Activation of lactotroph D2R results in suppression of PRL gene transcription, PRL synthesis, PRL release, and lactotroph proliferation [38]. D2R is a member of the G-protein coupled receptor superfamily and is highly expressed in pituitary [39, 40]. DA binding to D2R produces multiple intracellular responses, such as inhibition of adenylyl cyclase (AC) activity, reduction of cytosolic Ca 2+ levels, and activation of a K + channel [41 ]. D2R RNA is alternatively spliced by inclusion or exclusion of exon 6 to produce two isoforms of mRNA encoding proteins differing by 29 amino acids, called D2444 (D2L) and D2415(D2S)receptors [42, 43]. In the human pituitary both isoforms are found in equal amounts, whereas in the rat pituitary the longer form is predominantly expressed [43, 44]. As the extra sequence of D2L is situated within a putative region that binds to G protein, the isoforms may be important in determining the coupling to different G-proteins [42, 45]. SRIF inhibits basal and stimulated PRL release by cultured rat pituitary cells. The inhibitory effect of somatostatin occurs only in the presence of estradiol that increases the number of somatostatin receptors on lactotrophs [46]. Thyrotropin releasing hormone (TRH) is a tripeptide produced by the hypothalamus, initially shown to stimulate the release of thyrotropin (TSH), and subsequently proven to stimulate PRL synthesis and release in rat. However in humans such a role is not yet established. Vasoactive intestinal peptide (VIP), originally isolated from porcine intestine, is abundant in the hypothalamus and is another candidate for prolactin releasing factor. Oxytocin produced by hypothalamus, stimulates PRL release by acting on lactotrophs via oxytocin receptor [47]. Recently, a 31 amino acid hypothalamic PRL-releasing peptide (PrRP) that specifically stimulates PRL production by pituitary lactotrophs was reported in several species including man [48, 49]. PrRP acts through a specific transmembrane receptor coupled to G-protein. Human pituitary transcribes PrRP gene as well, suggesting an autocrine or paracrine regulatory role for this peptide [50]. PRL autoregulates its own secretion through both the long and intermediate PRL receptor isoforms. There is some evidence that PRL negative feedback is exercised at the hypothalamic and pituitary level [51 ]. Estrogen is a powerful stimulator of lactotrophs and acts via an estrogen receptor (ER) to increase PRL gene transcription, PRL synthesis and release, and lactotroph multiplication [52]. The ER recognizes a nonpalindromic DNA sequence in the 5'-flanking region designated the distal enhancer of PRL gene. Other DNA elements are necessary for the response of the PRL gene to estrogen, such as Pit-1, a tissue specific-transcription factor that binds to multiple sites in both the proximal and distal enhancer of the PRL gene. The interaction between Pit-1 and ER appears to be necessary for an estrogen response of the rat PRL gene [53]. Recently, a second ER gene designated ER[3 (the original one became ERa), was discovered [54-56]. ER[3 binds physiological ligands similarly to ERc~, but has different ability of binding antiestrogens, selective ER modulators, and potential environmental estrogens [56, 57]. ER[3 mRNA was demonstrated in rat and human pituitary but not in mouse pituitary [54-56, 58]. Estrogen can stimulate the lactotrophs indirectly by increasing the number of TRH receptors [59]. The maximum binding activity of TRH receptors occurs at proestrus in the pituitaries of female rats [60]. The up-regulation of TRH receptor mRNA by estrogen seems to occur at both the transcription rate and stability [61 ].
104
PRL regulates reproduction, behavior, and fluid and electrolyte metabolism. PRL acts via PRL receptor, which has three isoforms. PRL-R is a member of the cytokine/hematopoietic receptor family. The role of PRL in immune function was recently reviewed [36, 62]. As for GH, PRL functions as a "competence hormone" for the immune system [63]. PRL counteracts the effects of corticosteroids. It seems that in the rat extrapituitary PRL plays an important role in hematopoiesis [64], whereas PRL of pituitary origin is necessary for antibody formation in response to immune challenge [65]. Lymphocyte-derived PRL/GH probably has an autocrine function. However ultimately immunocompetence depends on normal pituitary function. 2.3.
ACTH
ACTH is a 39 amino acid peptide produced by pituitary corticotrophs from a precursor molecule-proopiomelanocortin (POMC). POMC is a glycoprotein processed into different biologically active peptides by proprotein convertases [66]. POMC gene is located on chromosome 2 in man. Corticotrophs are stimulated by corticotropin-releasing hormone (CRH) produced by the hypothalamus via CRH receptor [67]. CRH, a 41 amino acid peptide is the most potent ACTH secretagog in man and rat. Its action is potentiated several fold by arginine vasopressin (AVP), oxytocin (OT), angiotensin II, norepinephrin (NE), and epinephrine (EPI). CRH receptor is linked to the adenylate cyclase complex, and has two isoforms, designated CRH-R1 and CRF-R2, in which 29 amino acids are spliced in the first intracellular loop of CRH-R1 [68]. Pituitary CRH-R is down-regulated in response to adrenalectomy, glucocorticoid treatment, and chronic stress. The existence of a hypothalamic corticotropin release-inhibitory factor is suspected, but has not yet been identified [67]. Vasopressin exerts its stimulatory effect via the V1 receptor, represented by V la and V lb receptors, and V2 receptor subtypes [69, 70]. V lb receptor expression is positively regulated by glucocorticoids and it is coupled to the phosphatidylinositol (PI) pathway. Oxytocin stimulates the release of ACTH by an indirect mechanism, since oxytocin receptor was not identified in corticotrophs [47]. Corticotrophs are inhibited by circulating glucocorticoids, via type II glucocorticoid receptor (GR) [71]. In the human pituitary, the inhibited corticotrophs are characterized by accumulation of cytoplasmic microfilaments, the so-called Crooke's hyaline change. ACTH acts on the adrenal cortex to stimulate glucocorticoid production and proliferation of cortical cells. ACTH action is mediated by an ACTH receptor, which was cloned and included in the type 2 among the melanocortin receptor family. The role of the hypothalamicpituitary-adrenal axis in the function of immune system was extensively investigated [72]. Glucocorticoids have antiinflammatory and immuno-suppressive effects. There is also evidence that they can stimulate certain immune mechanisms. Glucocorticoid effects on the immune system were comprehensively reviewed [73, 74]. ACTH and other POMC derived peptides such as [3-endorphin and ct-MSH have the ability to regulate directly immune reactions.
2.4.
TSH
TSH[3 gene is located on chromosome 19 in humans. The c~-subunit chain is common for TSH and gonadotropins and is encoded by one gene on chromosome 6 in man. Thyrotropin releasing
105
hormone (TRH) released by hypothalamus stimulates TSH synthesis and release, as well as thyrotroph multiplication. TRH acts via a specific receptor. TRH effect is counterbalanced by thyroid hormones acting directly on thyrotrophs T 3 receptor [75]. Hypothalamic somatostatin and dopamine have inhibitory role on TSH secretion [76]. TSH stimulates the thyroid to synthesize thyroglobulin (Tg), enhances the proteolysis of Tg with the release of thyroid hormones, and stimulates the proliferation of thyroid follicular cells and thyroid vascularization. TSH acts on some immune cells, and thyroid hormones have stimulatory effect on thymus growth and hormone production and on bone marrow. Hypothyroidism in man is associated with immunodeficiency. 2.5.
FSH and LH
FSH and LH are composed of two subunits, ct and [3, each consisting of a peptide with branched carbohydrate side chains. The [5-subunits are encoded by separate genes located on chromosome 11 for FSH and chromosome 19 for LH in humans. Gonadotropin releasing hormone (GnRH) produced by hypothalamus exerts a stimulatory effect on gonadotrophs through GnRH receptor [77]. Estrogen may stimulate LH release during mid-cycle LH surge, an effect that may be prolonged by progesterone. Estrogen has an inhibitory effect on FSH synthesis and release. Inhibin and activin produced by ovaries and pituitary selectively regulate FSH synthesis. FSH stimulates ovarian follicular growth and testicular spermatogenesis. LH promotes ovulation and luteinization of the ovarian follicle, stimulates Leydig cells, and enhances steroid production in both ovary and testis. Sex hormones are not necessary for immune function, but they play important roles as immunomodulators [63]. They determine a clear gender dimorphism on the immune system, and affect the susceptibility and progression of numerous autoimmune diseases [78]. Thus lymphocytic hypophysitis, a rare inflammatory disorder assumed to be of autoimmune origin occurs predominantly in women [79].
3.
PITUITARY CYTOKINES
Inflammatory responses are accomplished by participation of numerous cell types that communicate via cytokines. An increasing number of cytokines have been characterized, and proven to be pleiotropic and functionally redundant. In the broadest definition, cytokines are cellular products that alter target cell function in an autocrine, paracrine or endocrine manner. This definition includes polypeptides, neuropeptides, lipids, vasoactive amines, nucleotides, and metabolites of oxygen and nitrogen [80]. Herein only the cytokines produced by pituitary are discussed. 3.1.
Interleukins
The interleukins are leukocytes-derived peptides with roles in regulating immune responses. Interleukins and their receptors are constitutively expressed or induced in the pituitary [81]. 3.1.1. Interleukin-1 (IL-1) IL-1 is produced by many cells, including monocytes, and stimulates B and T-lymphocytes to produce other cytokines. Two distinct genes encoding IL-lct and IL-I[3 were identified and showed 26% homology. They bind the same membrane receptor, and it seems that only IL-I[3 is active. In the rat pituitary, IL-I[3 was localized in TSH cells by immunocytochemistry [82].
106
Both type I and type II receptors for IL-1 were reported in mouse and rat adenohypophysis [83, 84]. Since administration of bacterial lipopolysaccharides (LPS) to rats induced a marked increase of pituitary IL-I[3 mRNA, it was suggested that this cytokine may play a role in paracrine or autocrine regulation of pituitary function during infectious challenge [82]. The effect of IL-1 on pituitary hormones is controversial, depending on the in vitro or in vivo system, and duration of treatment. 3.1.2. Interleukin-2 (IL-2) IL-2 is a potent immunoregulatory T-cell derived cytokine which also acts on the hypothalamicpituitary-adrenal axis. IL-2 and its receptor mRNA were reported in pituitary tumors, such as human corticotroph adenomas and mouse AtT-20 pituitary tumor cell line [85]. An IL-2 receptor was shown in ACTH cells of cultured rat pituitary [85]. In vitro, IL-2 stimulates POMC gene expression in rat pituitary cells and AtT-20 cell line. 3.1.3. Interleukin-6 (IL-6) IL-6 shares biologic action with IL-1. IL-6 was localized in the folliculo-stellate cells of rat and mouse pituitary [86]. IL-6 gene expression was also shown in human pituitary and adenohypophysial tumors. IL-6 binding sites are present in rat and human pituitary [87]. IL-6 stimulated GH, PRL, and LH release by cultured rat pituitary cells [88]. In vivo, the main pituitary effect of IL-6 is the release of ACTH probably by acting on hypothalamic CRH producing neurons [89]. 3.1.4. Interleukin 11 (IL- 11) IL-6 and IL-11 utilize the gp 130 protein as an initial cellular signal transducer. IL-11 and its receptor mRNAs were reported in human pituitary, corticotroph and nonfunctioning pituitary adenomas and AtT-20 pituitary cell line [90]. In AtT-20 cells IL-11 stimulates POMC gene transcription and ACTH secretion [90]. In vitro, IL-11 stimulated the proliferation of a folliculo-stellate cell line (TtT/GF) and GH 3 cells, as well as GH and PRL secretion by GH 3 cells [91]. Ciliary neurotropic factor (CNTF), another cytokine of the gpl30 signal-mediated cytokines, is expressed by TtT/GF and GH 3 cell lines. CNTF treatment of these cells had the same effects as IL- 11 [91 ]. 3.2.
Leukemia-inhibitory factor (LIF)
LIF, a 1053 amino acid glycoprotein, is a pleiotropic cytokine with multiple functions on diverse tissues. LIF acts via a specific receptor subunit, which forms a heterodimeric complex with gpl30, a signal transduction glycoprotein shared with IL-6, IL-11, CNTF and oncostatin M. LIF is produced by the pituitary. In the human fetal and adult adenohypophysis, LIF was localized by immunocytochemistry in ACTH immunoreactive cells, and some GH immunoreactive cells, and less frequently in other cell types [92]. LIF stimulates ACTH synthesis and release by AtT-20 cells. The factors controlling LIF expression within the pituitary are not yet known. Mice transgenic for LIF structural gene under the GH promoter are dwarfs and their pituitaries contain decreased number of GH and PRL cells and a significantly increased number of ACTH cells. A special change is the presence of cystic cavities formed by invaginations from the anterior wall of Rathke's cleft, suggesting failure of Rathke's epithelium to differentiate into hormone-secreting cells [93]. LIF knockout mice exhibit retarded growth and suppression of hematopoietic stem cells [94]. LIF and
107
IL-6 stimulated prohormone convertases 1 protein and mRNA in AtT-20 tumor cells. PLS administration to rats increased pituitary PC1 and POMC mRNAs [95], suggesting LIF role in inflammatory stress. 3.3.
Macrophage migration inhibitory factor (MIF)
MIF, an initiator of the inflammatory response was found to be abundant in the mouse pituitary tumor cell line AtT-20, and its release was stimulated by bacterial lipopolysaccharides [96]. In both murine and human pituitaries, MIF immunoreactivity was localized mainly in ACTH cells and some TSH cells [97, 98]. Pituitary-derived MIF was found to contribute to circulating MIF during endotoxemia [99]. The available data support the role of MIF to act at an inflammatory site or lymph node to counterbalance the inhibitory effects of steroids on the immune system. 3.4.
Activins and inhibins
These dimeric proteins and their receptors are members of the transforming growth factor-j3 family [100]. They were initially isolated and characterized based on their ability respectively to enhance or inhibit pituitary FSH secretion in vitro. Inhibins are heterodimers composed of an c~-chain and either a [3A or a [3B-chain, and activins are homodimers of the inhibin [3-chains. They play an important role in the regulation of the hypothalamic-pituitary-gonadal axis. Activins and inhibins are produced in many tissues including the pituitary gonadotrophs. In vitro, IL-I[3 attenuated FSH secretion induced by activin-A, possibly by influencing the local balance of activin-B and follistatin (activin-binding protein) [101]. Besides the effects on FSH, they modulate other pituitary cell types, such as the somatotrophs, lactotrophs and corticotrophs. Activin-A inhibits basal and GHRH-stimulated GH secretion as well as GH synthesis [ 102]. Inhibin suppresses plasma GH levels, by acting at a hypothalamic level where it increases SRIF mRNA level and decreases GHRH mRNA content in the arcuate nucleus [103]. Recent evidence indicates that activin A acts as both a pro and anti-inflammatory cytokine, which is released early following LPS administration [ 104]. The origin of this activin remains to be determined. Other cytokines, such as basic fibroblast growth factor (bFGF), epidermal growth factor, transforming growth factor-c~, and transforming growth factor-or necrosis factor are induced or constitutively produced by pituitary, and they may influence hormone secretion and cell multiplication in an autocrine or paracrine fashion acting via specific receptors. These factors are not discussed here and were recently reviewed [81].
4.
CONCLUSIONS
The last decade has brought us overwhelming evidence that the same substance can act as a hormone, cytokine or growth factor, making possible the communication between the neuroendocrine and immune system. The shared molecular networks of ligands and receptors converge to maintain homeostatic responses to adverse factors such as stress, injuries, infections, and other diseases. The presence of cytokines in many tissue types makes them strong candidates as drugs, or drug targets in infection, autoimmune diseases, allergies, and other disorders. It is likely that an understanding of the complex mechanisms regulating pituitary function will lead to novel modalities of treatment of pituitary and immune system diseases.
108
REFERENCES
.
.
.
.
.
10. 11.
12. 13.
14.
15. 16.
17.
18.
Ishibashi K, Imai M. Identification of four new members of the rat prolactin/growth hormone gene family. Biochem Biophys Res Commun 1999; 262: 575-578. Bell GI. The glucagon superfamily: precursor structure and gene organization. Peptides 1986; 7 Suppl 1: 27-36. Mayo KE. Molecular cloning and expression of a pituitary-specific receptor for growth hormone-releasing hormone. Mol Endocrinology 1992; 6:1734-1744. Tang J, Lagace G, Castagne J, Collu R. Identification of human growth hormone-releasing hormone receptor splicing variants. J Clin Endocrinol Metab 1995; 80: 2381-2387. Miller TL, Godfrey PA, Dealmeida VI, Mayo KE. The rat growth hormone-releasing hormone receptor gene: structure, regulation, and generation of receptor isoforms with different signaling properties. Endocrinology 1999; 140:4152-4165. Lin C, Lin SC, Chang CP, Rosenfeld MG. Pit-l-dependent expression of the receptor for growth hormone releasing factor mediates pituitary cell growth. Nature 1992; 360: 765-768. Lin SC, Li S, Drolet DW, Rosenfeld MG. Pituitary ontogeny of the Snell dwarf mouse reveals Pit-l-independent and Pit-l-dependent origins of the thyrotrope. Development 1994; 120: 515-522. Smith RG, Palyha OC, Feighner SD, Tan CP, McKee KK, Hreniuk DL, Yang L, Morriello G, Nargund R, Patchett AA, Howard AD. Growth hormone releasing substances: types and their receptors. Horm Res 1999; 51 Suppl 3: 1-8. Kojima M, Hosoda H, Date Y, Nakazato M, Matsuo H, Kangawa K. Ghrelin is a growthhormone-releasing acylated peptide from stomach. Nature 1999; 402: 656-660. Reisine T, Bell GI. Molecular biology of somatostatin receptors. Endocr Rev 1995; 16: 427-442. O'Carroll AM, Lolait SJ, Konig M, Mahan LC. Molecular cloning and expression of a pituitary somatostatin receptor with preferential affinity for somatostatin-28. Mol Pharmacol 1992; 42: 939-946. O'Carroll AM, Raynor K, Lolait SJ, Reisine T. Characterization of cloned human somatostatin receptor SSTR5. Mol Pharmacol 1994; 46: 291-298. Kumar U, Laird D, Srikant CB, Escher E, Patel YC. Expression of the five somatostatin receptor (SSTR1-5) subtypes in rat pituitary somatotrophes: quantitative analysis by doublelayer immunofluorescence confocal microscopy. Endocrinology 1997; 138:4473-4476 Miller GM, Alexander JM, Bikkal HA, Katznelson L, Zervas NT, Klibanski A. Somatostatin receptor subtype gene expression in pituitary adenomas. J Clin Endocrinol Metab 1995; 80: 1386-1392. Greenman Y, Melmed S. Heterogeneous expression of two somatostatin receptor subtypes in pituitary tumors. J Clin Endocrinol Metab 1994; 78: 398-403. Greenman Y, Melmed S. Expression of three somatostatin receptor subtypes in pituitary adenomas: evidence for preferential SSTR5 expression in the mammosomatotroph lineage. J Clin Endocrinol Metab 1994; 79: 724-729. Murabe H, Shimatsu A, Ihara C, Mizuta H, Nakamura Y, Nagata I, Kikuchi H, Nakao K. Expression of somatostatin receptor (SSTR) subtypes in pituitary adenomas: quantitative analysis of SSTR2 mRNA by reverse transcription- polymerase chain reaction. J Neuroendocrinology 1996; 8: 605-610. Shimon I, Yan X, Taylor JE, Weiss MH, Culler MD, Melmed S. Somatostatin receptor (SSTR) subtype-selective analogues differentially suppress in vitro growth hormone and prolactin in human pituitary adenomas. Novel potential therapy for functional pituitary
109
19.
20.
21. 22.
23. 24.
25.
26.
27. 28.
29.
30. 31.
32.
33.
34.
tumors. J Clin Invest 1997; 100: 2386-2392. Rohrer SP, Birzin ET, Mosley RT, Berk SC, Hutchins SM, Shen DM, Xiong Y, Hayes EC, Parmar RM, Foor F, Mitra SW, Degrado SJ, Shu M, Klopp JM, Cai SJ, Blake A, Chan WW, Pasternak A, Yang L, Patchett AA, Smith RG, Chapman KT, Schaeffer JM. Rapid identification of subtype-selective agonists of the somatostatin receptor through combinatorial chemistry. Science 1998; 282: 737-740. Berelowitz M, Szabo M, Frohman LA, Firestone S, Chu L, Hintz RL. Somatomedin-C mediates growth hormone negative feedback by effects on both the hypothalamus and the pituitary. Science 1981; 212: 1279-1281. Tannenbaum GS, Guyda HJ, Posner BI. Insulin-like growth factors: a role in growth hormone negative feedback and body weight regulation via brain. Science 1983; 220: 77-79. Yamashita S, Melmed S. Insulin-like growth factor I action on rat anterior pituitary cells: suppression of growth hormone secretion and messenger ribonucleic acid levels. Endocrinology 1986; 118: 176-182. Mertani HC, Waters MJ, Jambou R, Gossard F, Morel G. Growth hormone receptor binding protein in rat anterior pituitary. Neuroendocrinology 1994; 59: 483-494. Wajchenberg BL, Liberman B, Giannella ND, Morozimato MY, Semer M, Bracco LO, Salgado LR, Knoepfelmacher M, Borges MH, Pinto AC, Kater CE, Lengyel AM. Growth hormone axis in Cushing's syndrome. Horm Res 1996; 45: 99-107. Senaris RM, Lago F, Coya R, Pineda J, Dieguez C. Regulation of hypothalamic somatostatin, growth hormone-releasing hormone, and growth hormone receptor messenger ribonucleic acid by glucocorticoids. Endocrinology 1996; 137: 5236-5241. Casanueva FF, Burguera B, Muruais C, Dieguez C. Acute administration of corticoids: a new and peculiar stimulus of growth hormone secretion in man. J Clin Endocrinol Metab 1990; 70: 234-237. Thakore JH, Dinan TG. Growth hormone secretion: the role of glucocorticoids. Life Sci 1994; 55: 1083-1099. Seifert H, Perrin M, Rivier J, Vale W. Growth hormone-releasing factor binding sites in rat anterior pituitary membrane homogenates: modulation by glucocorticoids. Endocrinology 1985; 117: 424-426. Petersenn S, Rasch AC, Presch S, Beil FU, Schulte HM. Genomic structure and transcriptional regulation of the human somatostatin receptor type 2. Mol Cell Endocrinology 1999; 157: 75-85. Giustina A, Wehrenberg WB. Influence of thyroid hormones on the regulation of growth hormone secretion. Eur J Endocrinology 1995; 133: 646-653. Leone TC, Cresci S, Carter ME, Zhang Z, Lala DS, Strauss AW, Kelly DP. The human medium chain acyl-CoA dehydrogenase gene promoter consists of a complex arrangement of nuclear receptor response elements and Spl binding sites. J Biol Chem 1995; 270: 24622. Ho KY, Thorner MO, Krieg RJJ, Lau SK, Sinha YN, Johnson ML, Leong DA, Evans WS. Effects of gonadal steroids on somatotroph function in the rat: analysis by the reverse hemolytic plaque assay. Endocrinology 1988; 123: 1405-1411. Kelley KW, Brief S, Westly HJ, Novakofski J, Bechtel PJ, Simon J, Walker EB. GH3 pituitary adenoma cells can reverse thymic aging in rats. Proc Natl Acad Sci USA 1986; 83: 5663-5667. Dialynas E, Brown-Borg H, Bartke A. Immune function in transgenic mice overexpressing growth hormone (GH) releasing hormone, GH or GH antagonist. Proc Soc Exp Biol Med 1999; 221: 178-183.
110
35. 36. 37. 38. 39.
40. 41. 42. 43.
44. 45.
46.
47.
48.
49.
50.
51. 52. 53. 54.
Auernhammer CJ, Strasburger CJ. Effects of growth hormone and insulin-like growth factor I on the immune system. Eur J Endocrinology 1995; 133: 635-645. Clark R. The somatogenic hormones and insulin-like growth factor-l: stimulators of lymphopoiesis and immune function. Endocr Rev 1997; 18: 157-179. Benker G, Jaspers C, Hausler G, Reinwein D. Control of prolactin secretion. Klin Wochenschr 1990; 68:1157-1167. Maurer RA. Dopaminergic inhibition of prolactin synthesis and prolactin messenger RNA accumulation in cultured pituitary cells. J Biol Chem 1980; 255: 8092-8097. Grandy DK, Marchionni MA, Makam H, Stofko RE, Alfano M, Frothingham L, Fischer JB, Burke-Howie KJ, Bunzow JR., Server AC. Cloning of the cDNA and gene for a human D2 dopamine receptor. Proc Natl Acad Sci USA 1989; 86: 9762-9766. Bevan JS, Webster J, Burke CW, Scanlon MF. Dopamine agonists and pituitary tumor shrinkage. Endocr Rev 1992; 3: 220-240. Vallar L, Meldolesi. J Mechanisms of signal transduction at the dopamine D2 receptor. Trends Pharmacol Sci 1989; 10: 74-77. Giros B, Sokoloff P, Martres MP, Riou JF, Emorine LJ, Schwartz JC. Alternative splicing directs the expression of two D2 dopamine receptor isoforms. Nature 1989; 342: 923-926. Dal Toso R, Sommer B, Ewert M, Herb A, Pritchett DB, Bach A, Shivers BD, Seeburg PH. The dopamine D2 receptor: two molecular forms generated by alternative splicing. EMBO J 1989; 8: 4025-4034. Gandelman KY, Harmon S, Todd RD, O'Malley KL. Analysis of the structure and expression of the human dopamine D2A receptor gene. J Neurochem 1991; 56: 1024-1029. DeKeyser J, Walraevens H, De BJ, Ebinger G, Vauquelin G. D2 dopamine receptors in the human brain: heterogeneity based on differences in guanine nucleotide effect on agonist binding, and their presence on corticostriatal nerve terminals. Brain Res 1989; 484: 36-42. Kimura N, Hayafuji C, Konagaya H, Takahashi K. 17 beta-estradiol induces somatostatin (SRIF) inhibition of prolactin release and regulates SRIF receptors in rat anterior pituitary cells. Endocrinology 1986; 119: 1028-1036. Breton C, Pechoux C, Morel G, Zingg HH. Oxytocin receptor messenger ribonucleic acid: characterization, regulation, and cellular localization in the rat pituitary gland. Endocrinology 1995; 136: 2928-2936. Hinuma S, Habata Y, Fujii R, Kawamata Y, Hosoya M, Fukusumi S, Kitada C, Masuo Y, Asano T, Matsumoto H, Sekiguchi M, Kurokawa T, Nishimura O, Onda H, Fujino M. A prolactin-releasing peptide in the brain. Nature 1998; 393: 272-276. Fujii R, Fukusumi S, Hosoya M, Kawamata Y, Habata Y, Hinuma S, Sekiguchi M, Kitada C, Kurokawa T, Nishimura O, Onda H, Sumino Y, Fujino M. Tissue distribution of prolactinreleasing peptide (PrRP) and its receptor. Regul Pept 1999; 83: 1-10. Zhang X, Danila DC, Katai M, Swearingen B, Klibanski A. Expression of prolactinreleasing peptide and its receptor messenger ribonucleic acid in normal human pituitary and pituitary adenomas. J Clin Endocrinol Metab 1999; 84: 4652-4655. Devost D, Boutin JM. Autoregulation of the rat prolactin gene in lactotrophs. Mol Cell Endocrinology 1999; 158: 99-109. Maurer RA, Kim KE, Day RN, Notides AC. Regulation of prolactin gene expression by estradiol. Prog Clin Biol Res 1990; 322: 159-169. Nowakowski BE, Maurer RA. Multiple Pit-l-binding sites facilitate estrogen responsiveness of the prolactin gene. Mol Endocrinology 1994; 8: 1742-1749. Kuiper GG, Enmark E, Pelto-Huikko M, Nilsson S, Gustafsson JA. Cloning of a novel receptor
111
55. 56.
57.
58.
59. 60.
61.
62. 63. 64. 65. 66.
67.
68. 69. 70.
71.
72. 73. 74.
expressed in rat prostate and ovary. Proc Natl Acad Sci USA 1996; 93: 5925-5930. Mosselman S, Polman J, Dijkema R. ER beta: identification and characterization of a novel human estrogen receptor. FEBS Lett 1996; 392: 49-53. Tremblay JJ, Lanctot C, Drouin J. The pan-pituitary activator of transcription, Ptxl (pituitary homeobox 1), acts in synergy with SF-1 and Pit 1 and is an upstream regulator of the Lim-homeodomain gene Lim3/Lhx3. Mol Endocrinology 1998; 12: 428-441. Paech K, Webb P, Kuiper GG, Nilsson S, Gustafsson J, Kushner PJ, Scanlan TS. Differential ligand activation of estrogen receptors ERalpha and ERbeta at AP1 sites. Science 1997; 277:1508-1510. Kuiper GG, Carlsson B, Grandien K, Enmark E, Haggblad J, Nilsson S, Gustafsson JA. Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptors alpha and beta. Endocrinology 1997; 138: 863-870. Gershengorn MC, BE, Geras E. Estrogens increase the number of thyrotropin-releasing hormone receptors on mammotropic cells in culture. Endocrinology 1979; 105: 171-176. Lean AD, Garon M, Kelly PA, Labile F. Changes of pituitary thyrotropin releasing hormone (TRH) receptor level and prolactin response to TRH during the rat estrous cycle. Endocrinology 1977; 100:1505-1510. Kimura N, Arai K, Sahara Y, Suzuki H. Estradiol transcriptionally and posttranscriptionally up-regulates thyrotropin-releasing hormone receptor messenger ribonucleic acid in rat pituitary cells. Endocrinology 1994; 134: 432-440. Chikanza IC. Prolactin and neuroimmunomodulation: in vitro and in vivo observations. Ann NY Acad Sci 1999; 876: 119-130. Berczi I, Nagy A, Stefaneanu L, Sasano H, Kovacs K, eds. Molecular and cellular endocrine pathology. London: Arnold, 2000: 425-456. Berczi I, Nagy E. Placental lactogen is a haemopoietic hormone. Br J Haematol 1991; 79: 355-358. Nagy E. Regulation of immunity in rats by lactogenic and growth hormones. Acta Endocrinol (Copenh) 1983; 102: 351-357. Benjannet S, Rondeau N, Day R, Chretien M, Seidah NG. PC1 and PC2 are proprotein convertases capable of cleaving proopiomelanocortin at distinct pairs of basic residues. Proc Natl Acad Sci USA 1991; 88: 3564-3568. Engler D. The corticotropin-release inhibitory factor hypothesis: a review of the evidence for the existence of inhibitory as well as stimulatory hypophysiotropic regulation of adrenocorticotropin secretion and biosynthesis. Endocr Rev 1999; 20: 460-500. Chen R, Lewis KA, Perrin MH, Vale WW. Expression cloning of a human corticotropinreleasing-factor receptor. Proc Natl Acad Sci USA 1993; 90: 8967-8971. Morel A, O'Carroll AM, Brownstein MJ, Lolait SJ. Molecular cloning and expression of a rat Vla arginine vasopressin receptor. Nature 1992; 356: 523-526. Birnbaumer M, Seibold A, Gilbert S, Ishido M, Barberis C, Antaramian A, Brabet P, Rosenthal W. Molecular cloning of the receptor for human antidiuretic hormone. Nature 1992; 357: 333-335. Hollenberg SM, Weinberger C, Ong ES, Cerelli G, Oro A, Lebo R, Thompson EB, Rosenfeld MG, Evans RM. Primary structure and expression of a functional human glucocorticoid receptor cDNA. Nature 1985; 318: 635-641. Besedovsky HO, del Rey A. Immune-neuro-endocrine interactions: facts and hypotheses. Endocr Rev 1996; 17: 64-102. Bateman A. The immune-hypothalamic-pituitary-adrenal axis. Endocr Rev 1989; 10: 92-112. Wick G. Immunoendocrine communication via the hypothalamo-pituitary-adrenal axis in
112
75.
76.
77. 78. 79. 80. 81. 82.
83.
84.
85.
86.
87.
88. 89. 90.
91.
92.
autoimmune diseases. Endocr Rev 1993; 14: 539-563. Nakai A, Sakurai A, Bell GI, DeGroot LJ. Characterization of a third human thyroid hormone receptor coexpressed with other thyroid hormone receptors in several tissues. Mol Endocrinology 1988; 2: 1087-1092. Scanlon MF, Toft AD. Regulation of thyrotropin secretion. In: Barverman LE, Utiger RD, eds. The thyroid. Seventh Edition Ed. Philadelphia: Lippincott-Raven Publishers, 1996: 220-240. Kakar SS. Molecular structure of the human gonadotropin-releasing hormone receptor gene [see comments]. Eur J Endocrinology 1997; 137: 183-192. Da Silva JA. Sex hormones and glucocorticoids: interactions with the immune system. Ann NY Acad Sci 1999; 876: 102-117. Sautner D, Saeger W, Ludecke DK, Jansen V, Puchner MJ. Hypophysitis in surgical and autoptical specimens. Acta Neuropathol. (Berl.) 1995; 90: 637-644. Chensue SW, Ward PA. Inflammation. In: Damjanov I, Linder J, eds. Anderson's pathology. 10th Ed St Louis: Mosby, 1996; 387-415. Ray D, Melmed S. Pituitary cytokine and growth factor expression and action. Endocr Rev 1997; 18: 206-228. Koenig JI, Snow K, Clark BD, Toni R, Cannon JG, Shaw AR, Dinarello CA, Reichlin S, Lee SL, Lechan RM. Intrinsic pituitary interleukin-1 beta is induced by bacterial lipopolysaccharide. Endocrinology 1990; 126:3053-3058 Cunningham ETJ, Wada E, Carter DB, Tracey DE, Battey JF, De Souza EB. Distribution of type I interleukin-1 receptor messenger RNA in testis: an in situ histochemical study in the mouse. Neuroendocrinology 1992; 56: 94-99. Parnet P, Brunke DL, Goujon E, Mainard JD, Biragyn A, Arkins S, Dantzer R, Kelley KW. Molecular identification of two types of interleukin-1 receptors in the murine pituitary gland. J Neuroendocrinology 1993; 5:213-219. Arzt E, Stelzer G, Renner U, Lange M, Muller OA, Stalla GK. Interleukin-2 and interleukin-2 receptor expression in human corticotrophic adenoma and murine pituitary cell cultures. J Clin Invest 1992; 90:944-1951. Vankelecom H, Carmeliet P, Van Damme J, Billiau A, Denef C. Production of interleukin-6 by folliculo-stellate cells of the anterior pituitary gland in a histiotypic cell aggregate culture system. Neuroendocrinology 1989; 49:102-106. Ohmichi M, Hirota K, Koike K, Kurachi H, Ohtsuka S, Matsuzaki N, Yamaguchi M, Miyake A, Tanizawa O. Binding sites for interleukin-6 in the anterior pituitary gland. Neuroendocrinology 1992; 55:199-203. Spangelo BL, Judd AM, Isakson PC, MacLeod RM. Interleukin-6 stimulates anterior pituitary hormone release in vitro. Endocrinology 1989; 125: 575-577. Mastorakos G, Magiakou MA, Chrousos GP. Effects of the immune/inflammatory reaction on the hypothalamic-pituitary-adrenal axis. Ann NY Acad Sci 1995; 771: 438-448. Auernhammer CJ, Melmed S. Interleukin-11 stimulates proopiomelanocortin gene expression and adrenocorticotropin secretion in corticotroph cells: evidence for a redundant cytokine network in the hypothalamo-pituitary-adrenal axis. Endocrinology 1999; 140: 1559-1566. Perez CC, Nagashima AC, Pereda MP, Goldberg V, Chervin A, Largen P, Renner U, Stalla GK, Arzt E. The gpl30 cytokines interleukin-ll and ciliary neurotropic factor regulate through specific receptors the function and growth of lactosomatotropic and folliculostellate pituitary cell lines. Endocrinology 2000; 141: 1746-1753. Akita S, Webster J, Ren SG, Takino H, Said J, Zand O, Melmed S. Human and murine
113
93.
94.
95. 96. 97.
98.
99. 100. 101.
102. 103.
104.
pituitary expression of leukemia inhibitory factor. Novel intrapituitary regulation of adrenocorticotropin hormone synthesis and secretion. J Clin Invest 1995; 95:1288-1298. Akita S, Readhead C, Stefaneanu L, Fine J, Tampanaru-Sarmesiu A, Kovacs K, Melmed S. Pituitary-directed leukemia inhibitory factor transgene forms Rathke's cleft cysts and impairs adult pituitary function. A model for human pituitary Rathke's cysts. J Clin Invest 1997; 99: 2462-2469. Escary JL, Perreau J, Dumenil D, Ezine S, Brulet P. Leukaemia inhibitory factor is necessary for maintenance of haematopoietic stem cells and thymocyte stimulation. Nature 1993; 363: 361-364. Li QL, Jansen E, Friedman TC. Regulation of prohormone convertase 1 (PC1) by gpl30-related cytokines. Mol Cell Endocrinology 1999; 158: 143-152. Bucala R. Neuroimmunomodulation by macrophage migration inhibitory factor (MIF). Ann NY Acad Sci 1998; 840: 74-82. Nishino T, Bernhagen J, Shiiki H, Calandra T, Dohi K, Bucala R. Localization of macrophage migration inhibitory factor (MIF) to secretory granules within the corticotrophic and thyrotrophic cells of the pituitary gland. Mol Med 1995; 1: 781-788. Tampanaru-Sarmesiu A, Stefaneanu L, Thapar K, Kovacs K, Donnelly T, Metz CN, Bucala R. Immunocytochemical localization of macrophage migration inhibitory factor in human hypophysis and pituitary adenomas. Arch. Pathol Lab. Med 1997; 121:404-410. Bernhagen J. MIF is a pituitary-derived cytokine that potentiates lethal endotoxaemia. Nature 1993; 365: 756-759. Mather JP, Moore A, Li RH. Activins, inhibins, and follistatins: further thoughts on a growing family of regulators. Proc Soc Exp B iol Med 1997; 215: 209-222. Bilezikjian LM, Turnbull AV, Corrigan AZ, Blount AL, Rivier CL, Vale WW. Interleukinl beta regulates pituitary follistatin and inhibin/activin betaB mRNA levels and attenuates FSH secretion in response to activin-A. Endocrinology 1998; 139: 3361-3364. B ilezikjian LM, Corrigan AZ, Vale W. Activin-A modulates growth hormone secretion from cultures of rat anterior pituitary cells. Endocrinology 1990; 126: 2369-2376. Carro E, Senaris RM, Mallo F, Dieguez C. Regulation of hypothalamic somatostatin and growth hormone releasing hormone mRNA levels by inhibin. Brain Res Mol Brain Res 1999; 66: 191-194. Jones KL, Brauman JN, Groome NP, de Kretser DM, Phillips DJ. Activin A release into the circulation is an early event in systemic inflammation and precedes the release of follistatin. Endocrinology 2000; 141: 1905-1908.
This Page Intentionally Left Blank
New Foundation of Biology Edited by I. Berczi and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
115
Neuroendocrine Stress and Inflammatory Disease" From Animal Model to Human Disease
MOJDEH MOGHADDAM and ESTHER M. STERNBERG
Section on Neuroendocrine Immunology and Behavior, Integrative Neural Immune Program, National Institute of Mental Health/NIH, Bldg. 10, Rm. 2D-46, 10 Center Dr., MSC 1284, Bethesda, MD 20892-1284 USA
ABSTRACT This review will outline scientific advances in understanding the communication networks between the nervous and immune systems: the scientific underpinning of the popular mindbody interaction. The idea that the mind and negative or positive states of mind, such as psychological stress or well-being, can influence health and disease has been in the popular culture for thousands of years. Recent scientific advances prove that there is a molecular, cellular, neuroanatomical and neurohormonal basis for communication between the brain and the immune system. Through such communications the nervous and immune systems interact and modify each other's functions. Interruptions of this interaction, on a genetic, drug-induced or surgical basis, lead to enhanced susceptibility to inflammatory disease. Over-activity of the neuroendocrine component of these interactions, such as occurs during stress, is associated with exacerbations of, or increased susceptibility to, infectious disease. The presence of cytokines in the brain, and their role in neuronal cell death and survival, help explain the role of immune molecules in degenerative brain diseases like dementia seen in Alzheimer's and AIDS. Cytokines expressed within the nervous system also play a role in nerve damage and recovery from nerve trauma. On the basis of such findings, new drug treatments are currently being developed, such as the use of anti-inflammatory drugs in Alzheimer's or neurotransmitter related drugs for improving immune responses associated with aging.
1.
INTRODUCTION
As one glances back through the early history of medicine, it is apparent that a prevalent notion for millennia was the idea that the mind plays a crucial role in the manifestation of illness. It is ironic that more recent investigations in the field of inflammatory and infectious diseases rejected the concept that mind can affect state of disease. In other words, the interaction between the neural and immune systems was significantly ignored, largely because technologies were insufficient to prove these connections. Until recently, this assumption continued amongst most scientists. However, growing evidence clearly indicates that the brain and immune systems continuously signal each other, along multiple hormonal, neuronal, cellular and molecular
116
pathways [1, 2]. This information now provides a basis for understanding how state of mind can influence our health. To date, we know that interactions between the central components of the stress response and the immune system play a significant role in susceptibility and resistance to inflammatory disease. This communication occurs via both hormonal and neural mechanisms. Peripheral cytokines released during inflammation or infection activate the hypothalamic-pituitary-adrenal (HPA) axis. The stimulated HPA axis in turn regulates immune responses through the immunosuppressive effects of glucocorticoids. The sympathetic nervous system also plays an important role in regulating immune responses, through interaction with immune cells in immune organs such as the spleen or thymus. Furthermore, neuropeptides released from peripheral nerves at sites of inflammation also play a role in local inflammation, and which may be either pro-inflammatory or anti-inflammatory. Rapid progress in the fields of molecular neuroscience, neuroimmunology, and neuroendocrinology, have enabled delineation of these principles in a quantitative manner in animal models as well as in related human diseases.
2.
STRESS AND NEUROENDOCRINE FACTORS
Stress involves a range of emotional, physiological, behavioral and neuroendocrine responses evoked by a threatening stimulus, and can cause a reduction in health [3]. Although necessary and adaptive in acute situations, when the stress response persists after the threat has diminished, the response itself can become maladaptive. The total load of repeated diverse stressors has been termed "allostatic load" [4]. Indeed, a large scientific literature exists, emphasizing the fact that exposure to stress is one predisposing factor in the development of a number of psychopathological disorders. Alternately, an inappropriate neuroendocrine stress response may be secondary to other etiological factors in these illnesses. For instance, depression, anxiety, and/or post-traumatic stress disorder (PTSD) are among those psychiatric diseases that can manifest themselves after exposure to stress [5, 6, 7] and that have also been associated with dysregulation of the neuroendocrine stress response. Finally, stress hormones can modify immune responses, and therefore inflammatory or infectious disease. Contemporary investigation has revealed that during stressful situations, specific neuronal and neuroendocrine response pathways are activated by different stressors, which induce different patterns of CNS response [8]. Following exposure to a stressful event, central corticotropin-releasing hormone (CRH) is released from the paraventricular nucleus of the hypothalamus. The locus coeruleus is also activated, resulting in behavioral and physiological responses known as the fight-or-flight response [7]. The sub-cortical hypothalamic-pituitary-adrenal axis hormonal cascade is the final common pathway of the neuroendocrine stress response. The HPA axis response is activated almost immediately after exposure to stressful stimuli. This causes the secretion of corticotropin releasing hormone (CRH) from the hypothalamus, which subsequently stimulates the pituitary gland to secrete adrenocorticotropic hormone (ACTH) which in turn stimulates the adrenal glands to release glucocorticoid hormones [7]. Glucocorticoid feed back suppresses the hypothalamic-pituitary-adrenal HPA axis cascade at every level. Another part of brain that responds to stressful stimuli is the brain stem. Activation of this area results in an increased sympathetic nervous system outflow to the periphery [8]. Hypothalamic CRH can stimulate brain stem noradrenergic areas and elevate their sympathetic activation.
117
As a result, brain stem adrenergic pathways signal the hypothalamus to secrete CRH. Recent investigations indicate that neuronal pathways activated during the stress response vary according to the nature of the stressful stimulus whether physiological, immunological or pharmacological [8]. Furthermore, the response to psychological stressors is dependent on individual perception of the event [9].
3.
ANIMAL MODELS OF INFLAMMATORY AND NEUROENDOCRINE DISEASE
The immune and central nervous systems are capable of communicating and influencing each other at many levels [10, 11, 12]. A variety of studies have defined patterns of expressions of cytokines and their receptors in brain [ 13, 14]. In general cytokines expressed within brain tissue act either to promote cell death or as neurotrophic factors, preventing neuronal death. Peripheral infection or inflammation can also send signals to the hypothalamus and pituitary to secrete CRH and ACTH through the activation of cytokines [15, 16, 17]. Some important cytokines that can signal the CNS are tumor necrosis factor (TNF), TNF-c~, TNF-[5, interleukin-1 (IL-1), IL-2, I1-3, IL-6, IL-10, and platelet activating factor (PAF). Many different types of animal models have been used to investigate the role of the stress response, particularly the HPA axis and sympathetic nervous system, in the pathophysiology of autoimmune disease.
4.
LEWIS AND FISCHER RATS
Lewis (LEW/N) and Fischer (F344/N) rats have been used intensively to study the role of the HPA axis in inflammatory disease. LEW/N rats have been used to study different autoimmune inflammatory conditions such as arthritis and uveitis [12]. The neuroendocrine response was initially investigated in these rats following their exposure to group A streptococcal cell wall peptidoglycan polysaccharide (SCW). After SCW injection, female LEW/N rats develop a severe erosive arthritis resembling human RA clinically, histologically, and radiologically [18]. In contrast to Lewis rats, inflammation resistant F344/N rats lie at the other end of the inflammatory severity spectrum, showing only minimal disease expression in response to this and other inflammatory stimuli [ 18]. It has been shown that F344/N rats are resistant not only to progress of chronic erosive arthritis but also are less susceptible to acute carrageenan-induced exudative inflammation than are LEW/N rats [19, 20]. Administration of low dose corticosteroids reduces the acute inflammatory response of LEW/N rats to a level similar to that observed in the F344/N strain. On the other hand, treatment of F344/N rats with the glucocorticoid antagonist RU486 increases their inflammatory response to levels close as LEW/N rat [ 18]. The LEW/N rat animal model has also been used to study other autoimmune diseases. Following administration of myelin basic protein, rats develop experimental allergic encephalomyelitis (EAE) a disease that in many ways resembles multiple sclerosis (MS). Exogenous corticosteriods have been shown to reduce the intensity of EAE in these animals, while adrenalectomy results in increased disease severity [21, 22].
118
.
OTHER ANIMAL MODELS OF HPA AXIS HYPOACTIVITY AND AUTOIMMUNE DISEASE
The association between HPA axis hypoactivity and autoimmune disease has also been shown in chickens prone to thyroiditis; mice prone to lupus; and in other rat models as well as in humans. A variety of mouse strains have been used as models for SLE. Some of these strains have been found to have different HPA axis activation responses. In comparison to non-lupus-prone mice, NZB, NZW, and MRL/MP-Lpr mice have elevated baseline corticosteroids levels with a blunted elevation following IL-I[3 injection [23]. This HPA axis phenotype differs from that observed in the arthritis-susceptible LEW/N rats. Since diseases collectively referred to as autoimmune, differ greatly in etiology as well as in pathogenesis and clinical features, these differences in the mechanism or location of HPA axis dysregulation are not surprising.
HUMAN AUTOIMMUNE AND NEUROIMMUNE AND NEUROENDOCRINE RHEUMATOID ARTHRITIS DISEASE Several human autoimmune/inflammatory diseases have been associated with blunted HPA axis responses. These include rheumatoid arthritis [24], atopic dermatitis, allergic asthma [25], chronic fatigue syndrome [26], and fibromyalgia. Children with juvenile rheumatoid arthritis show both a relatively blunted HPA axis response and perturbations in sympathoneuronal responses to stress Cobi Heijnen [27]. Adults with rheumatoid arthritis show relatively blunted HPA axis responses to the stress of surgery. Other clinical syndromes associated with blunted HPA axis responses included psychiatric syndromes such as atypical depression and seasonal affective disorder [28]. In chronic inflammatory disease, chronic inflammation itself can act as a stressor to the HPA axis and can alter acute HPA axis responses to other stressors [29]. According to Schmidt et al., [30] cytokine stimulation of HPA axis and inflammation induces a long-lived shift in the stress response from one primarily driven by CRH to another one driven by arginine vasopressin (AVP). The importance of the sympathetic nervous system in addition to the HPA axis is underlined in human studies as well as animal findings. Juvenile RA patients show altered sympathetic responsiveness as measured using orthostatic stress [27]. In addition, peripheral mononuclear white blood cells in juvenile RA show decreased responses to [3-2 adrenergic stimulation that may possibly result in altered T cell activation and monocyte function. This alteration could possibly contribute to differential inflammatory responses.
7.
CONCLUDING REMARKS AND FUTURE DIRECTION
In reviewing the English language literature that pertains to the field of neuroimmune modulation and stress, it is clear that the last decade has witnessed a rapid advancement and expansion in the study of cytokine biology, neuroendocrinology, and the interface between these fields. Cytokines were initially considered to have their main role as messengers between immune cells in inflammation. However these proteins communicate with many tissues and organs and are now recognized to have new roles outside their traditionally assumed functions. One of their new roles recently recognized is their action as neuromodulators within the brain.
119
As such, cytokines affect important brain functions such as neuroendocrine regulation, sickness behavior, sleep, etc. and when expressed within the brain play a role in neuronal cell death and survival. In turn neuronal and neuroendocrine responses alter immune function and play a role in immune disease. These new discoveries are opening up new avenues of treatment for many illnesses including neurodegeneration, nerve trauma, and neuro-AIDS. Conversely, studies of the effects of many different neuroendocrine and neuronal responses on autoimmune inflammatory diseases have introduced new approaches to management of these diseases with agents directed at the central nervous system. Our knowledge and understanding of the role of cytokines, particularly in regards to human brain activities is still in its infancy. Much work must still be done to investigate the interface between genetic and environmental factors in diseases affected by neural-immune interactions. However, a better understanding of the role of cytokines, neuroendocrine and neuronal responses on various immune activities will enhance our knowledge of specific psychobiological factors involved in health and disease. It will also provide us with promising opportunities for novel therapeutic interventions in the illness.
REFERENCES 1. 2. 3. 4. 5. 6. 7. 8.
9. 10. 11. 12. 13. 14.
Sternberg EM. Emotions and disease: from balance of humors to balance of molecules. Nat Med 1997; 3: 264-267. Kronfol Z, Remick D. Cytokines and the brain: implications for clinical psychiatry. Am J Psychiatry 2000; 157: 683-694. Huether G. The central adaptation syndrome: Psychosocial stress as a trigger for adaptive modifications of brain structure and brain function. Prog Neurobiol 1996; 48: 569-612. McEwen BS. Seminars in medicine of the Beth Israel deaconess medical center: protective and damaging effects of stress mediators. NEJM 1998; 338: 171-179. Anisman H, Zacharko RM.Depression as a consequence of inadequate neurochemical adaptation in response to stressor. Br. J Psychi. 1992; Suppl. 15, 160: 36-43. Berridge CW, Dunn AJ. Restraint-stress-induced changes in exploratory behavior appear to be mediated by norepinephrine stimulated release of CRF. J Neurosci 1989; 9:3513-3521. Chrousos GP, Gold PW. The concepts of stress and stress system disorders. JAMA 1992; 267(9): 1244-1252. Sawchenko PE, Brown ER, Chan RK, Ericsson A, Li HY, Roland BL, Kovacs KJ. The paraventricular nucleus of the hypothalamus and the functional neuroanatomy of visceromotor responses to stress. Prog Brain Res 1996; 107: 201-222. Ehlert U, Straub R. Physiological and emotional response to psychological stressors in psychiatric and psychosomatic disorders. Ann NY Acad Sci 1998; 851: 477-486. Sternberg EM, Chrousos GP, Wilder RL, Gold PW. The stress response and the regulation of inflammatory disease. Ann Intern Med 1992; 117: 854-866. Chrousos GP. The hypothalamic-pituitary-adrenal axis and immune-mediated inflammation. N Engl J Med 1995; 332: 1351-1362. Sternberg EM. Neural-immune interactions in health and disease. J Clin Invest 1997; 100:2641-2647. Besedovsky HO, del Rey A. Immune-neuro-endocrine interactions: facts and hypotheses. Endocr Rev 1996; 17: 64-102. Turnbull AV, Rivier CL. Regulation of the hypothalamic-pituitary-adrenal axis by cytokines: actions and mechanisms of action. Physio Rev 1999; 79: 1-71.
120
15. 16. 17. 18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29. 30.
Berkenbosch F, Oers VJ, Rey DA, Tilders F, Besedovsky H. Corticotropin-releasing factorproducing neurons in the rat activated by interleukin-1. Science 1987; 238: 524-536. Bernton EW, Beach JE, Holaday JW, Smallridge RC, Fein HG. Release of multiple hormones by a direct action of interleukin-1 on pituitary cells. Science 1987; 238: 519-521. Sapolsky R, Rivier C, Yamamoto G, Plotsky P, Vale W. Interleukin-1 stimulates the secretion of hypothalamic corticotropin releasing factor. Science 1987; 238: 522-524. Sternberg EM, Hill JM, Chrousos GP, Kamilaris T, Listwak SJ, Gold PW, Wilder RL. Inflammatory mediator-induced hypothalamic-pituitary-adrenal axis activation is defective in streptococal cell wall arthritis-susceptible Lewis rats. Proc Nati Acad. Sci USA 1989; 86: 2374-2378. Karalis K, Crofford L, Wilder RL, Chrousos GP. Glucocorticoid and/or glucocorticoid antagonist effect in inflammatory disease-susceptible Lewis rats and inflammatory diseaseresistant Fischer rats. Endocrinology 1995; 136:3107-3112. Misiewicz B, Zelazowska E, Raybourne RB, Cizza G, Sternberg EM. Inflammatory responses to carrageenan injection in LEW/N and F344/N rats: LEW/N rats show sex-and age-dependent changes in inflammatory reactions. Neuroimmunomodulation 1996; 3:93-101. Reder AT, Thapar M, Jensen MA. A reduction in serum glucocorticoids provokes experimental allergic encephalomyelitis: implications for treatment of inflammatory brain disease. Neurology 1994; 44: 2289-2294. Mason D, MacPhee I, Antoni F. The role of the neuroendocrine system in determining genetic susceptibility to experimental allegic encephalomyelitis in the rat. Immunology 1990; 70: 1-5. Lechner O, Hu Y, Jafarian-Tehrani M, Dietrich H, Schwarz S, Herold M, Haour F, Wick G. Disturbed immunoendocrine communication via the hypothalamo-pituitary-adrenal axis in murine lupus. Brain Behav Immun 1996; 10: 337-350. Cash JM, Crofford LJ, Gallucci WT, Sternberg EM, Gold PW, Chrousos GP, Wilder RL. Pituitary-adrenal axis responsiveness to ovin corticotropin releasing hormone in patients with rheumatoid arthritis treated with low dose prednisone. J Rheumatol 1992; 19:1692-1696. Buske-Kirschbaum A, Jobst S, Psych D, Wustmans A, Kirschbaum C, Rauh W, Hellhammer D. Attenuated free cortisol response to psychosocial stress in children with atopic dermatitis. Psychosom Med 1997; 59:419-26. Demitrack MA, Dale JK, Straus SE, Laue L, Listwak SJ, Kruesi MJ, Chrousos GP, Gold PW. Evidence for impaired activation of the hypothalamic-pituitary-adrenal axis in patients with chronic fatigue syndrome. J Clin Endocrinol Metab 1991; 73: 1224-1234. Kavelaars A, de Jong-de Vos van Steenwijk T, Kuis W, Heijnen CJ. The reactivity of the cardiovascular system and immunomodulation by catecholamines in juvenile chronic arthritis. Ann NY Acad Sci 1998; 840: 698-704. Gold PW, Licinio J, Wong ML, Chrousos GP. Corticotropin releasing hormone in the pathophysiology of melancholic and atypical depression and in the mechanism of action of antidepressant drugs. Ann NY Acad Sci 1995; 771: 716-729. Harbuz MS, Rees RG, Lightman SL. HPA axis responses to acute stress and adrenalectomy during adjuvant-induced arthritis in the rat. Am J Physiol 1993; 264: 179-185. Schmidt ED, Jansen AW, Wouterlood FG, Tilders FJ. Interleukin-l-induced long-lasting changes in hypothalamic corticotropin-releasing hormone (CRH)-neurons and hyperresponsiveness of the hypothalamus-pituitary-adrenal axis. J Neurosci 1995; 15: 7417-7426.
New Foundationof Biology
121
Edited by I. Berczi and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
Immunoregulation by the Sympathetic Nervous System
DWIGHT M. NANCE and BRIAN J. MACNEIL
Department of Pathology, University of Manitoba, Winnipeg, MB., Canada
ABSTRACT The neuroendocrine and autonomic nervous systems constitute efferent pathways through which the nervous system modulates peripheral immune responses. Regulation of glucocorticoid levels by the HPA axis is a primary component of the neural-immune regulatory system, whereas the role of the autonomic nervous system in the regulation of immune function is not fully elucidated. We have identified two paradigms, central inflammatory stimuli and stress, for which an immunosuppressive role of the sympathetic nervous system has been demonstrated. Since the spleen is exclusively innervated by sympathetic nerve fibers and is accessible for experimental manipulation, we have utilized this secondary immune organ as a model system for analyzing brain-immune interactions. Central injections of inflammatory stimuli (IL-1 or PGE2), as well as stress, produce an acute suppression of splenic macrophage function. Although the HPA axis and the sympathetic nervous system are jointly activated by these treatments, we have shown that the acute suppression of splenic macrophage function by central inflammatory stimuli and stress are still observed in adrenalectomized animals. Abrogation of this adrenal-independent immunosuppression in splenic immune function by surgically cutting the sympathetic nerve fibers innervating the spleen illustrates that the sympathetic nervous system constitutes an important pathway for the neural regulation of peripheral immune function. Although both stress and immune stimuli activate the same efferent system, they access this regulatory system via different neural pathways. The paraventricular nucleus (PVN) is proposed as an essential component of this regulatory network and a nodal region for the integration and regulation of both neuroendocrine and autonomic responses. Brain stem knife cuts or posterolateral deafferentation of the PVN indicate that activation of the PVN by immune stimuli are primarily, if not exclusively, mediated by ascending brain stem afferents to the PVN. These same brain stem knife cuts have a minimal effect on the activation of the PVN by stress and loss of posterior and lateral connections of the PVN only partially attenuates the activation of the PVN by stress. These results indicate that, in contrast to immune related stimuli, rostral inputs to the PVN mediate a major portion of the activational effects of stress on the PVN. Thus, the HPA axis and the sympathetic nervous system are the two primary output pathways utilized by the neural-immune regulatory system to regulate peripheral immune responses.
122
1.
INTRODUCTION
Selye observed that a diverse number of stressful challenges generate a common pattern of physiological responses and adaptive changes. Among the numerous physiological responses produced by stress, some of the most dramatic changes were observed in the immune system. Selye demonstrated the fundamental role of adrenal steroids in mediating many of the changes producd by stress and therefore identified the hypothalamic-pituitary-adrenal (HPA) axis as a major pathway for the brain to regulate peripheral immune responses. Although the HPA axis is a primary component of the brain-immune regulatory system, it is not the sole mediator of brain-immune interactions. As will be reviewed here, the autonomic nervous system constitutes a separate and independent pathway for the regulation of immune function and together with the neuroendocrine system, they constitute the efferent pathways through which the nervous system modulates peripheral immune responses. The hypothesis that these brain-immune pathways might be components of a more complex regulatory feedback network was first stated by Besedovsky [1, 2, 3] who proposed the bi-directional communication between the brain and the immune system. In their model, products of the immune system released from stimulated immune cells (lymphokines) signal the brain which consequently induced a response that downregulated immune function. In support of this regulatory feedback model, they demonstrated the concurrent activation of the HPA axis and sympathetic nervous system produced by an immune challenge and cytokines. In addition to verifying a central role for the HPA axis in the bi-directional communication between the brain and immune system, their demonstration of a relationship between splenic norepinephrine (NE) and the magnitude of the immune response to antigen stimulation focused attention on the autonomic nervous system as an independent and significant immunoregulatory pathway. We have tested the model system proposed by Besedovsky and have examined the functional organization of this proposed neural-immune regulatory system. Illustrated in Figure 1 is our summary of the functional and neuroanatomical organization of the neural-immune regulatory system with specific reference to the sympathetic regulation of splenic immune function. A diverse number of multidisciplinary approaches and techniques have been applied to the analysis of this regulatory network and we summarize here some of the data that supports our working model of the neural-immune regulatory system. A first requirement of this model system is that immune organs must be innervated by nerve fibers.
2.
INNERVATION OF IMMUNE ORGANS
Not only are all lymphoid organs richly innervated, these nerve fibers are distributed to specific cellular compartments in immune organs [4, 5]. In addition to the vascular innervation of immune organs, sympathetic fibers are located throughout the parenchyma and distributed to macrophage, dendritic, T and B cell specific compartments of immune organs. The origin of these sympathetic fibers have been identified for the spleen by neuroanatomical tract-tracing studies and demonstrate that the prevertebral celiac-mesenteric ganglia provide a major source of the sympathetic innervation to the spleen and the paravertebral sympathetic chain ganglia provide a second and comparable input [6]. Cutting the splenic nerve eliminated all nerve fibers and terminals in the spleen, establishing that the splenic nerve constitutes the final common neural pathway. No parasympathetic or afferent input to the spleen was observed. Thus, direct neural modulation of splenic function, and most likely that of the thymus, must be mediated entirely via the sympathetic nervous system [7]. This singular source of efferent
123
innervation, combined with the distribution of nerve fibers to immune cells in the spleen and the accessability of this immune organ to experimental manipulations and analysis make it an ideal organ system with which to analyze further the neural-immune regulatory system.
3.
PHYSIOLOGICAL EFFECTS OF IMMUNE STIMULI
As indicated in the model (Figure 1), products generated by immune responses must be able to signal the brain in order to initiate counterregulatory responses. Most thoroughly examined of the lymphokines shown to produce behavioral and central effects is interleukin-1 [8]. This cytokine is produced primarily by activated macrophages and has a primary role in acute phase reactions during infections and in amplification of T cell responses. IL-1 also induces fever and slow wave sleep, alters hypothalamic norepinephrine (NE) turnover and releases hypothalamic corticotrophin releasing factor (CRF) and ACTH [9, 10, 11], responses that are similar to those observed during immune responses [2]. The paraventricular hypothalamic nucleus (PVN) plays an important role in CRF release and is a target for IL-1 and antigen induced alterations in NE metabolism and neuronal firing patterns [12, 13]. An immune response to sheep erythrocytes (SRBC) and IP injection of IL-1 also alter the level of NE, serotonin and metabolites in other brain areas [11, 13]. As illustrated in Figure 1, forebrain areas providing input to the hypothalamus, such as the amygdala, septal and hippocampal areas, are implicated in the control of immune function [14, 15, 16] as well as targets for corticosterone and IL-1 [17] feedback. IL-1 receptors have been described in the hypothalamus [17], but in situ hybridization studies indicate that IL-1 receptors are primarily expressed and widely distributed on cerebral vascular endothelial cells [ 18]. Further evidence that endothelial cells provide a critical interface between immune stimuli and the central nervous system is that the synthetic enzyme for prostaglandins, a proposed intermediatary signal of inflammatory stimuli, is also primarily localized to vascular endothelial cells and both message and protein are rapidly induced throughout the cerebral microvasculature and leptomeninges following LPS or inflammatory cytokine injections [19, 20, 21, 22]. Thus immune-dependent stimuli produce a pattern of physiological responses which supports the concept that immune signals act upon a neural-immune regulatory system composed of the hypothalamus and it's connections. Further identification of the specific hypothalamic nuclei and neurochemical pathways that compose the neural-immune regulatory system has been provided by functional neuroanatomical and immunocytochemical studies.
4.
ACTIVATION OF THE CENTRAL NEURAL-IMMUNE REGULATORY SYSTEM
The expression of immediate early response genes such as c-fos has become a powerful tool for structural and functional analysis of the nervous system [23, 24]. For example, c-fos is expressed in dorsal horn neurons of the spinal cord following noxious stimulation of the foot [25], in the supraoptic nucleus of the hypothalamus following water deprivation and the hippocampus following seizures [23]. Physiological and psychological stressors and noxious stimuli produce a rapid induction of c-fos protein in neuroendocrine and autonomic regions of the hypothalamus and brain stem [26]. In addition, specific limbic forebrain structures, such as the amygdala, bed nucleus of the stria terminalis, lateral sepatal region and frontal cortex were also activated by stress. We have tested whether ICV injections of endotoxin (lipoplysaccharide; LPS), a cell membrane constituent of gram-negative bacteria, would activate c-fos protein in the brain. Relative to vehicle injected controls, we found that 3 hour following LPS infusions
124
LIMBIC SYSTEM* SEPTAL AREA/AMYGDALA*
~G receptors?
C-FOS GLUTAMATE
"A" NITRIC OXIDE
DESCENDING HYPOTHALAMIC INPUTTO BRAINSTEM AND SYMPATHETIC PREGANGLIONICS
~
* PGE2 ml JUTS
~
I Centre: ~ o n ~
I
~
MEDULLARY
C j A ~ T] E C H O /L A M I N E : INPUTTO THE PVN (PG receptors?) (CYTOKINES) / '
I
1
~-.~,-,-,-.,.~,.J~: "" ~Intraperltoneal DESCENDINGMEDULLARY ::~p~~i,i~,' . ..,r .9 ~ INPUTTOSYMPATHETIC ~ "J, ...........
I | T
/
~ ' - ! ' ~ * GLUTAMATE
/ CYTOKINES(IL-1, TNF,IL-6) ,~ T
~,',":'.'::-':'.~k ~ ~.":"-.":"~:.':.~4Y
I
INTRATHECAL
*
|I THORACICSYMPATHETIC PREGANGLIONICS 7
SPLENIC NERVECUTS
~ ~
SYMPATHETIC
"A" Denotes points In the circuitry that we have demonstrated to be critical for the expression of c-fos in the PVN, altered splenic immune function and/or splenic nerve activity.
Figure 1. Illustrated is our model of the neural-immune regulatory system which is focused on the primary components of the regulatory network that regulate the sympathetic control of splenic immune function. The neuroendocrine component of this regulatory system is not shown. The PVN is proposed as a critical integrative brain region in nerual-immune interactions and in the regulation of immune-related neuroendocrine and autonomic responses.
125
there was a localized expression of c-fos protein in the paraventricular nucleus of the hypothalamus (PVN) and dorsal medulla as well as elevated splenic levels of NE and VIP in rats [27]. A more widespread pattern of c-fos induction was observed following IP and IV injections of LPS which produced a dose and time dependent activation of c-fos protein in the PVN, supraoptic nucleus, arcuate nucleus, as well as the monoamine cell groups located in the dorsal, ventrolateral and ventromedial regions of the medulla [26, 27]. Consistent with the concept that some of the actions of LPS may be mediated via cytokine release, central and peripheral injections of IL-1 produce a neuronal pattern of c-fos mRNA and/or protein in the brain comparable to what we reported for LPS [28, 29, 30]. Thus, the highly localized distribution of neurons expressing c-fos protein following LPS or IL-1 continues to provide a neuroanatomical correlate of the central activational effects of these potent immune stimuli. Other activity dependent cellular markers, such as NGFI-B, have verified the central activational effects of various immune stimuli [18]. Double labelling immunocytochemical studies indicate that numerous neuropeptide producing neurons in the hypothalamus are activated by LPS, and include CRF, vasopressin, oxytocin, and nitric oxide (NO) producing cells [31, 32]. These same chemically specific cell groups in the hypothalamus provide direct input to autonomic premotor neurons in the medulla, as well as sympathetic preganglionic neurons in the thoracic spinal cord [33, 34, 35], underscoring the potential involvement of these neuropeptides in the sympathetic regulation of the immune system. We demonstrated that pretreatment with the prostaglandin synthesis inhibitor indomethacin blocked the activation of c-fos protein in the brain by both IP and IV injections of LPS [26]. Indomethacin blocks most of the physiological effects of LPS (fever, gastric secretion, corticosterone release, changes in central catecholamine metabolism, etc.) as well as those of IL-1 and TNF-~t [36, 37, 38, 39, 40]. The localization of prostaglandin receptors in hypothalamic and brain stem nuclei implicated in autonomic and endocrine control suggest further that prostaglandins may be a primary mediator of brain-immune interactions [20, 21, 41, 42]. Importantly, the EP4-PGE2 receptor has been localized to the same hypothalamic nuclei activated by central PGE2 injections, and may mediate the central effects of this prostaglandin. The localization of endothelial cells, a primary source of cytokines/LPS-induced prostaglandin production, in relation to neurons bearing specific prostaglandin receptors has not been systematically examined. However, the hypervascularity of the neuroendocrine hypothalamic nuclei (PVN and supraoptic nucleus), relative to all other brain regions, is well documented [43], and suggest that there is ample opportunity for vascular endothelial cell derived prostaglandins to act directly upon hypothalamic neurons that are targets for the functional effects of prostaglandins. For example, Ray and Choudhury [44] observed that hypothalamic vasopressin cell bodies and dendritic processes were located either adjacent to blood vessels or on the endothelium, and occasionally, the neurons were located in the lumen of hypothalamic capillaries. The relationship between other hypothalamic neuropeptide producing neurons, especially oxytocin and CRF, and the vascular supply has not been systematically examined. Consistent with prostaglandin production being a mediator of the central effects of LPS [26], we have found that ICV prostaglandin E2 (PGE2) activates c-fos protein in the hypothalamus. An additional factor, nitric oxide production, has been linked with both glutamate neural transmission and the activation of c-fos protein [45]. In support of a functional role for nitric oxide producing neurons in the regulation of the neural-immune regulatory system, we have preliminary data that indicates central inhibition of nitric oxide production with the NOS inhibitor L-NAME attenuates the central induction of c-fos protein by IV LPS (Figure 2) as well as by ICV PGE2 (Figure 3) [46]. As illustrated in our central signal cascade (Figure 1),
126
e
" 9
"/'
.
"-
"'In'
i.
.
~ ,
".
,"
.
.
,,,"
.
We '
"
. 9 .
'" , , t ' - .
"g't
9 -
O
i
r 9 ,
" .
,,,;
-~"nI~ , I,,. -
~'o
,
alp O -
lIT ,..,q~4t
", .,, 21.., ,.
.
"
.9
s
j
.~
g.
-,,,.,
-
LI ~
f .
Q
~.',
B
~
Q
.., .
V
"~
'
l
t"
B
Figure 2. High contrast video digitized micrographs of immunocytochemically identified c-fos positive neurons in the hypothalamus demonstrating the inhibition of i.v. LPS (20/~g) induced c-fos protein in the PVN by pretreatment i.c.v. with the NOS inhibitor L-NAME (A). Illustrated in B is the absence of any effects of pretreatment with the inactive isomer, D-NAME on the induction of c-fos protein in the PVN by LPS.
NO appears to act distal to PGE2. Since the PVN exerts control over pituitary function, autonomic regulatory centers in the medulla, and preganglionic sympathetic neurons in the spinal cord (splenic innervation), the PVN remains the primary candidate for being a nodal region for mediating neural-immune interactions. Finally, we also demonstrated that the NMDA glutamate antagonist, MK801, blocked the activation of c-fos protein in the brain by IP and IV injections of LPS, thereby implicating glutamate neural transmission in this central circuitry activated by LPS [26]. In this regard, intranuclear excitatory glutamate neurotransmission has been proposed to account for the excitatory effects of ascending noradrenergic inputs to the PVN from the medulla [47]. While both IV and IP injection routes for immune stimuli result in similar patterns of central expression of c-fos in specific nuclei, the neuroanatomical route by which these immune stimuli reach this common central neural-immune network are different. We identified a primary role for visceral afferents in mediating the central activational effects of IP injections of LPS and showed that subdiaphragmatic vagotomy blocked the induction of c-fos protein in the brain following IP injections of LPS [48, 26]. Since this observation was reported, many of the
127
,0
qe.
..
.p
i,
.*~f
.~.
~
O
.~
Figure 3. High contrast video digitized micrographs of immunocytochemically identified c-fos positive neurons in the hypothalamus demonstrating the inhibition of i.c.v. PGE2 induced c-fos protein in the PVN by pretreatment i.c.v. with the NOS inhibitor L-NAME (A). Illustrated in B is the absence of any effects of pretreatment with the inactive isomer, D-NAME on the induction of c-fos protein in the PVN by i.c.v. PGE2.
behavioral and physiological effects of IP injections of LPS and/or IL-1 have been shown to be eliminated by subdiaphragmatic vagotomy or selective hepatic deafferentation [49, 50, 51]. The likely importance of peripheral afferents in the central processing of immune signals has forced a reexamination of the widely held concept that all immune system related signals reach the CNS directly or via circumventricular organs [52]. However, since we showed that vagal afferents do not mediate CNS activation produced by blood borne immune signals (i.e., IV LPS/IL-1) or stress, we hypothesized the existence of multiple and separate input pathways to endocrine and autonomic regulatory brain regions [26].
5.
REGULATION OF SPLENIC SYMPATHETIC NERVE ACTIVITY
Although the localization of activity dependent cellular markers, like c-fos, continue to provide a powerfull tool to characterize neural-immune pathways, electrophysiological experiments can reveal dynamic and rapid information on the neural modulation of organ systems.
128
Therefore, we have utilized peripheral nerve recordings as an additional index of the central activation of the neural-immune regulatory system and the sympathetic regulation of the spleen (Figure 4). We have reported that IV injections of LPS produce a dramatic and sustained increase in splenic nerve electrical activity [53, 54, 55, 56]. Since renal nerve activity was unchanged, or even reduced, during this same period, these results illustrate that the sympathetic nervous system can be regulated in an organ specific manner. Just as has been noted for the induction of c-fos protein in the brain [26], we found that both systemic and central injections of indomethacin attenuated the activation of the splenic nerve by LPS (Figure 4). In support of our hypothesis that prostaglandins are a mediator of the central effects of LPS, we found that ICV injections of PGE2 produced a rapid increase in splenic nerve activity. The role of neuropeptide producing hypothalamic neurons in mediating the central actions of PGE2 has not been thoroughly examined. However, we have tested the effects of several neuropeptide receptor antagonists on the activation of the splenic and renal nerves by central PGE2 [56]. Activation of the renal nerve by PGE2 was partially and similarly blocked by pretreatment ICV with selective CRF and V 1 vasopressin antagonists. However, with regards to splenic nerve activity, an oxytocin antagonist produced a selective and complete inhibition of the facilitation in splenic nerve activity produced by central PGE2, whereas the CRF and vasopressin antagonists did not significantly alter the effects of central PGE2 on splenic nerve electrical activity. In support of a unique and functional relationship between PGE2 and oxytocin in neuroimmune regulation, it has been shown that ICV injections of PGE2 induced c-fos in the PVN primarily in oxytocin neurons, as well as CRF cell bodies, but not in vasopression cells [32]. Similarly, the neuroanatomical location of hypothalamic neurons that showed c-fos induction following ventrolateral medulla injections of PGE2 corresponded to oxytocin neurons [57]. Lastly, we have found that IV injections of LPS produced an increase in oxytocin immunostaining in the rostral PVN, as well as for NADPHd activity, but not for VP [58]. We have also found that intrathecal injections of AP-5, a NMDA glutamate receptor antagonist, administered at the time of the peak effect of LPS, reduced splenic nerve activity to baseline (Figure 4). Taken together, these results indicate that LPS acts centrally to activate the splenic nerve, and a descending glutamate pathway acting upon spinal NMDA receptors (spinal preganglionic neurons) transmits this central action to the splenic nerve. These studies provide independent and convergent physiological evidence that prostaglandin is a central mediator of the sympathetic control of splenic immune function. These results demonstrate further that a signal cascade generated by the immune system activates the hypothalamus and specific neuropeptide and transmitter systems which regulate the subsequent efferent sympathetic and neuroendocrine responses to immune challenges. Given that the immune system can signal the brain and produce a specific pattern of neuroendocrine and autonomic responses, do these proposed counterregulatory changes alter peripheral immune responses?
6.
FEEDBACK INHIBITON OF IMMUNE FUNCTION
We have verified that there are two immunoregulatory pathways that exert inhibitory control over peripheral immune function. In our experiments in which we showed that in vitro splenic macrophage IL-1 secretion was suppressed by ICV IL-1 injections [10], we found that this immune suppression was coincidental with increases in the secretion of ACTH and corticosterone. Adrenalectomy, which removes plasma corticosterone and epinephrine, reversed the immunosuppressive effect of central IL-1. Significantly, cutting the splenic nerve, which selectively removes the sympathetic input to the spleen, was as effective as adrenalectomy.
129
Injection
200t 100
0
20ot
100
0
LPS i.v. (Renal)
200t 100
0
indo i.v. + LPS i.v.
(Splenic)
200t 100
0
Indo i.c.v. + LPS i.v.
(Splenic)
200 1
,o:1 AP-5 2 i.c.v. (Splenic)
200 nmol
i.t. injection LPS 100 pg i.v.
o~ ZOO] Lr3
o
1 0 rain Effect of i.t. injection of 200 nmoles of AP-5 during peak response of l,l'S-induced increase in splenic nerve activity. Figure 4. Splenic and renal nerve recordings in rats following 100 pg i.v. LPS or 1.0 pg i.c.v. PGE2 injections. Demonstrated are the differential effects of LPS on splenic and renal nerve activity, the inhibitory effects of i.v. and i.c.v, indomethacin injections on the splenic nerve response to LPS as well as the immediate activational effects of central PGE2 injections on splenic nerve activity. Also shown is the inhibitory effects of intrathecal (i.t.) injections of the NMDA receptor antagonist AP-5 on LPS-induced increase in splenic nerve activity.
130
The effects of nerve section combined with adrenalectomy were additive and resulted in an enhanced responsiveness of LPS stimulated macrophages. Thus, nerve section enhanced LPS responses despite the presence of high levels of serum corticosterone, verifying that the effects of the sympathetic nervous system on immune function can be independent of the HPA axis. Sundar [59] identified a similar immunosuppressive effect of ICV IL-1 on in vitro NK cell activity, PHA responsiveness and IL-2 production. This effect could be blocked by c~-MSH, an endogenous IL-1 antagonist, but was only partially reversed by adrenalectomy. These results illustrate the existence of a cytokine-responsive central regulatory system that regulates peripheral immune responses via definable pathways (endocrine and autonomic). In further support for the counterregulatory role for the sympathetic nervous system on immune function, NE has been shown to suppress in vitro IL-1 production in splenic macrophages, and the NE content of the spleen decreases during immunization at the exponential phase of the immune response [60, 61]. Spleen NE turnover is increased after exposure to antigen and following ICV IL-1 injections [62]. NE can change migration of lymphocytes in the spleen and whether it can stimulate or inhibit immune activity is a function of cell type, adrenoreceptor subtype and temporal parameters [63]. Chemical sympathectomy is reported to alter T cell activity and antigen presentation in macrophage. Neuropeptides, such as Neuropeptide Y (NPY) are co-localized with NE [64] and may enhance or inhibit transmitter activity. Many of the physiological effects of tumor necrosis factor (TNF-ct), another cytokine produced by macrophage, are comparable to IL-1. It is typically produced and secreted before IL-1 and is a key mediator of sepsis and shock. Importantly, in vitro data indicates that macrophage secretion of TNF-et is under noradrenergic control [65, 66, 67]. Macrophages express adrenergic receptors [68] and selective or- and [3- agonist can potentiate and inhibit macrophage cytokine secretion, respectively [65, 66, 67] and modify macrophage cytokine mRNA expression. However, most of the studies examining neurotransmitter modulation of macrophage function have utilized in vitro preparations. However, since interactions between nerve fibers and immune cells occur primarily within specific compartments of immune organs, in vitro studies may not model accurately neuroimmune interactions.
EFFECTS OF CENTRAL INFLAMMATORY STIMULI AND STRESS ON I N VIVO IMMUNE FUNCTION We have focused much of our research efforts on testing the functional significance of the sympathetic innervation of immune organs. We have identified two experimental paradigms, central inflammatory stimuli and stress, for which an immunosuppressive role of the sympathetic nervous system has been clearly demonstrated. Previously we showed that the in vitro immunosuppressive effects of central IL-1 injections, as well as psychological stress, can be abrogated by prior sectioning of the splenic nerve [10, 69]. To test this relationship in an in vivo model system, we utilized ICV injections of PGE2, which we have identified as a primary activator of the neural-immune regulatory system, in conjunction with a systemic injection of LPS [70]. Preliminary data show that ICV injections of PGE2 dramatically decreased the in vivo production of splenic TNF-ct mRNA and protein following an IV injection of endotoxin (Figures 5 & 6). Importantly, cutting the splenic nerve attenuates the suppression in splenic TNF-et production produced by central PGE2 (Figures 7 & 8). That cutting the splenic nerve reduced the suppression of splenic TNF-ct production in adrenal intact animals supports the concept that the sympathetic nervous system and HPA axis exert independent inhibitory control over immune cell function. This same relationship was observed in a separate and independent
131
6.0
Z
I
T
TNF IL-1
71//I
_
~
<
I / t l
I
//11
/ A
" i
i
i
t
:I
/////A
I,.//'t2/A lil./I//+'~ t./t/'/.J
.I1
! I I! i J i / i . . . ~+.,i
4.0
. "
E O 0
3.0
~
2.0
.,4
1////A
5.0
Z
E]
*
=
o
Ili
""
/;'774
1,0
................... ~ ~ / ~ 0,0
..............................
.. . . . . . .
L
f
PGE2
SAL GROUPS
Figure 5. Inhibitory effects of central PGE2 (4/xg) injections on in vivo splenic cytokine mRNA levels 1 hour following i.v. LPS (0.1/xg). PGE2 or saline were injected i.c.v. 30 min. prior to the systemic injection of LPS. Cytokine mRNA levels were measured by quantitative Northern blotting with dig-labelled RNA probes. The blots were analyzed by densitometry and the mRNA levels expressed as a ratio relative to the loading control.
,= o~
4
v
o) r c r 03
.,,.=
.,..
3
J:: r . . =
2 z I-o r ~ m
r 03
PGE2
Saline
Group Figure 6. Inhibitory effects of central PGE2 (4/xg) injections on in vivo splenic TNF-alpha protein levels 1 hour following i.v. LPS (0.1 /xg). PGE2 or saline were injected i.c.v. 30 min. prior to the systemic injection of LPS. Splenic TNF protein levels were indexed by means of quantitative immunocytochemistry and digitalmicroscopy.
132
mo
-9
[-]
TNF-a
I~
IL-1
I
I 6
~(
o O
]
~///-//
"/.
.......
r ////A
,_= "o m
<
V///A
z
E o o
2
Sham
N e r v e cut GROUP
Figure 7. Cutting the splenic nerve attenuates the inhibitory effects of central PGE2 (4 pg) injections on in vivo splenic TNF-alpha mRNA levels 1 hour following i.v. LPS (0.1 pg). PGE2 or saline were injected i.c.v. 30 min. prior to the systemic injection of LPS. Splenic nerve cuts or sham operations were performed by microsurgical procedures 7-10 days prior to the experiment.
0.30 A b
m ,~ 0 . 2 0
,i t~
~p L 0.10 Z t-"
0.00
Nerve cut
Sham GROUPS
Figure 8. Cutting the splenic nerve attenuates the inhibitory effects of central PGE2 (4 pg) injections on in vivo splenic TNF-alpha protein levels 1 hour following i.v. LPS (0.1 pg). PGE2 or saline were injected i.c.v. 30 min. prior to the systemic injection of LPS. Splenic nerve cuts or sham operations were performed by microsurgical procedures 7-10 days prior to the experiment.
133
paradigm in which we examined the effects of stress on LPS induced splenic cytokine production. We found that 15 minutes of intermittent footshock suppressed LPS-induced in vivo splenic TNF-c~ and IL-1 mRNA and protein production, relative to nonstressed controls [71, 72]. Significantly, the immunosuppressive effects of stress on splenic macrophage cytokine production was also observed following adrenalectomy, verifying the existence of an adrenalindependent immunosuppressive pathway. Finally, evidence that the sympathetic nervous system represents this/other pathway was shown by the fact that cutting the splenic nerve abrogated the immunosuppressive effects of stress in adrenalectomized animals. Taken together, these diverse set of experimental results provide convergent evidence that the sympathetic innervation of immune organs constitutes a functionally significant pathway for regulating the immune system and this modulation of immune function can be independent from the actions of the HPA axis.
.
STRESS AND IMMUNE STIMULI ACTIVATE THE NEURAL-IMMUNE REGULATORY SYSTEM VIA SEPARATE PATHWAYS
While both immune stimuli and stress impact on the in vivo function of splenic macrophage, we have shown using c-fos immunocytochemistry that these diverse stimuli can activate the same central neural circuit via separate and independent pathways [26]. For example, subdiaphragmatic vagal afferents are critical for the behavioral, physiological and central effects of IP injections of cytokines or endotoxin [49, 50, 51, 26], but vagotomy has no effect of the activational effects of footshock or restraint stress. Also, the central effects of IV injections of endotoxin do not require vagal afferents, demonstrating independent signaling pathways. Finally, indomethacin pretreatment completely blocked the central activational effects of IP and IV LPS injections, but had no effect on the central induction of c-fos protein by stress. Therefore, we proposed an additional signalling pathway through which stress activates a common neural-immune regulatory network via extrahypothalamic forebrain structures [26]. In support of this hypothesis, unilateral knife cut studies indicate that deafferentation of ascending brain stem catecholamine pathways to the hypothalamus unilaterally reduces the number of catecholamine fibers and terminals in the PVN and attenuate the central activational effects of endotoxin, but has no significant effect on the activational effects of footshock on these hypothalamic neurons (Figure 9). However, if the unilateral knife cuts are located posterolateral to the hypothalamus, a procedure which unilaterally eliminates all ascending catecholamine fibers in the hypothalamus, the activational effects of LPS on the PVN is dramatically reduced on the cut side of the brain (Figure 10). However, these same hypothalamic knife cuts had only a modest effects on the activation of the PVN by stress. These results indicate that immune related signals reach the PVN via ascending brain stem/mesopontine pathways, whereas the effects of stress on the PVN only depend in part upon this ascending pathway [73]. We propose that rostrolateral inputs to the PVN from limbic forebrain structures, which are the only connections to the hypothalamus spared by these posterior hypothalamic knife cuts, mediate the activational effects of stress on the PVN (Figure 1). As illustrated in the model, the immunosuppressive and antiinflammatory effects of stress are dependent upon extrahypothalamic inputs from limbic forebrain structures. However, the ability of inflammatory and immune-dependent stimuli and central inflammatory signals to activate the PVN and to generate the appropriate counterregulatory neuroendocrine and autonomic responses, is mediated by ascending monoamine neural pathways. Although the functional activity of this neural-immune regulatory network can continue in the absence of neural inputs from limbic
134
forebrain structures, there is no doubt that these same forebrain structures exert powerful modulatory control over this hypothalamic homeostatis system.
700"
Unilateral brain stem knife cuts [--I D
T
600 (jr) ..!
-J 500
CUT CONT
i
7
,.,,-V/.,-V;/,I
m
0
400
0
300
W/,:.v;4,,:i
x .." i /
, . " t l /'~t I I i x
/ i " ," II,, /,0"/. ,,"
/;. :;-'>--:,,> /
T
,
/.i"//-.-,. ,,",..
,, I / / / A
,i
,. f
///I;"X/',,I i 9t / .
z 200
,,'X.,.,Y-:>,>;I
100
;?,;;/.>,,",,;">??;1
/
",,.o / ' i "
.."/,
./;4/,;,~,/,,
"I
i I ///"
/ / .,'
,...... l>- ,,,, / I / ,. I. /ill~ i,I/. ftttl
;.,vx::-//>.l 9
x,
,p,':,::;:s, .+. >,<., , /,.,/.,%, ,,
/, ///,,,;/.
I,,
I/////,"A
"ft/i
t
"/.'.....,3:'3,1 i/i//,.
i,!
i
ix..- ,., o.
fp:,,, y//,', ,, , .. z ! i / f Z;.':i;;" , ,.,
".';/,.1 p?,,/,.,,lJ
"/ /'//'i"
;,,v';;'/Y-'q
II/I1///
LPS
/,."
FS GROUPS
Figure 9. Effects of unilateral brain stem knife cuts on the bilateral distribution of c-fos positive cells in the PVN 2 hour following an i.v. injection of LPS or 1 hour following intermittent footshock (FS). Note the reduction in the number of LPS-induced c-fos positve cells on the cut side of the PVN, relative to the contralateral intact side of the brain. The effects of footshock on the bilateral induction of c-fos protein in the PVN was unaltered by these same brain stem knife cuts.
Unilateral hypothalamic knife cuts
I
I
600 [-]
03 ..I . j 500 ill 0 400
CUT CONT
/ //t///f //p" //'.>/i "./," /'~ / " / . . " t
x,,:,..,:,<,-;,.,., -x...<,:P.;'~"<,
1
O i,
V..,"W ,/..//,/
I
(.1, 300 0 z 200 ,=l: ui
100
r::p?;<,,;:>-:::i F?-"s
.";:),S:!:S%"
'il.........v /,,/; .,'/;q,.,
.... ,.I,..,. ,.';;~ //-/-/x _...-/- A ////
i
," ...."i"....."//.i..,..,." ,." / / / " / , / . . / ..
.,:,,,>:;:,,.,./,.,';. ,'~/VI',:,'~/; ;
,',,'/,,-,~,,-,, ~./"/'
//..,"///,,".,
/ / f 11 I / :,':,-;;:.:,-,.m.,,,,;
,'V,,";, /-';.d //VV,.'/,":I 'i
,$#.?./,,;,:,
/,,.-//,-/,-} L//,,',' ,",.............
/ ..
";,".,'7,':,."i,.'5-',{71
I_PS
FS
GROUPS Figure 10. Effects of unilateral posteriolateral hypothalamic knife cuts on the bilateral distribution of c-fos positive cells in the PVN 2 hour following an i.v. injection of LPS or 1 hour following intermittent footshock. Note the dramatic reduction in the number of LPS-induced c-fos positve cells on the cut side of the PVN, relative to the contralateral intact side of the brain. These same cuts had no significant effect on the activation of the PVN by footshock.
135
9.
CONCLUSION
There is now a substantial body of literature which supports the existence of a central neuroimmune regulatory network which mediates the interactions between the brain and peripheral immune system. The PVN and it's connections appear to constitute a central component of this regulatory system. Stress and central inflammatory stimuli acutely produce neuroendocrine and sympathetic responses that are antiinflammatory and inhibitory on splenic immune responses. The activation of this common central regulatory network by stress and immune stimuli is via separate and neuroanatomically distinct pathways. Our research has provided evidence that the sympathetic fibers provide direct, immediate, and potent regulatory control over a number of critical cellular and regulatory components of the immune system. The functional significance of this regulatory system in the maintenance of health through it's proposed role in mediating integrated and adaptive responses to injury, infection, disease, or stress, has not been fully determined. Further analysis of this regulatory network will establish the functional mechanisms and neuroanatomical pathways that mediate the immunoregulatory effects of the sympathetic nervous system.
ACKNOWLEDGEMENTS Research supported by National Institute of Mental Health Grant MH-43778-04A2 and Medical Research Council of Canada.
REFERENCES 1. 2. 3. 4. 5.
6. 7. 8. 9.
10.
Besedovsky H, del Rey A, Sorkin E, Da Prada M, Burri R, and Honegger C. The immune response evokes changes in brain non-adrenergic neurons. Science 1983; 221: 564-6. Besedovsky H, Del Rey AE, Sorkin E. Immune-neuroendocrine interactions. J Immunol 1985; 135: 750-4. Besedovsky H, del Rey A, Sorkin E, Dinarello CA. Immunoregulatory feedback between interleukin-1 and glucocorticoid hormones. Science 1986; 233,: 652-4. Felten SY, Felten DL. Innervation of lymphoid tissue. In: Psychoneuroimmunology, 2nd Edition, Edited by R Ader, DL Felten and N Cohen, Academic Press, pp 27-61,1991. Meltzer JC, Grimm PC, Greenberg AH, Nance DM. Enhanced immunohistochemical detection of autonomic nerve fibers, cytokines and inducible nitric oxide synthase by light and fluorescent microscopy in rat spleen. J. Histochem. Cytochem 1997; 45: 599-610. Nance DM, Burns J. Innervation of the spleen in the rat: Evidence for absence of afferent innervation. Brain Behav Immunity 1989; 3:281-90. Nance DM, Hopkins DA, Bieger D. Re-investigation of the innervation of the thymus gland in mice and rats. Brain Behav Immunity 1987; 1: 134-47. Dinarello CA. Biology of interleukin-1. FASEB J., 1988; 2: 108-15. Berkenbosch F, van Oelers J, del Rey A, Tilders F, Besedovsky H. Corticotrophin releasing factor producing neurons in the rat activated by interleukin-1. Science 1987; 238: 524-6. Brown R, Zuo L, Vriend C, Nirula R, Janz L, Falk J, Nance D, Dyck D, Greenberg AH. Suppression of splenic macrophage IL-1 secretion following intracerebroventricular injection of interleukin-l: Evidence for pituitary-adrenal and sympathetic control.
136
11.
12. 13. 14.
15.
16.
17. 18.
19.
20.
21.
22.
23. 24. 25. 26.
27.
28.
Cell Immunol 1991; 132: 84-93. Zalcman S, Green-Johnson JM, Murray L, Nance DM, Dyck D, Anisman H, Greenberg AH. Cytokine-specific central monoamine alterations induced by interleukin (IL)-I, IL-2 and IL-6. Brain Res 1994; 643: 40-9. Saphier D, Abramsky O, Mor G, Ovadia H. Multiunit electrical activity in conscious rats during an immune response. Brain Behav Immunity 1987; 1:40-51. Carlson SL, Felten DL, Livnat S, Felten SY. Alterations of monoamines in specific central autonomic nuclei following immunization in mice. Brain Behav Immunity 1987: 52-63. Nance DM, Rayson D, Carr RI. The effects of lesions in the lateral septal and hippocampal areas on the humoral immune response of adult female rats. Brain Behav Immunity 1987; 1: 292-305. Wetmore L, Nance DM. Differential and sex specific effects of kainic acid and domoic acid lesions in the lateral septal area on immune function and body weight regulation. Exper Neurol 1991; 13: 226-36. Wetmore L, Green-Johnson J, Gartner JG, Sanders V, Nance DM. The effects of kainic acid induced lesions in the lateral septal area on cell mediated immune function. Brain Behav Immunity 1994; 8:341-54. Farrar WL, Kilian PL, Ruff MR, Hill JM, Pert CB. Visualization and characterization of interleukin- 1 receptors in brain. J Immunol 1987; 139: 459-63. Ericsson A, Liu C, Hart RP, Sawchenko PE. Type 1 interleukin-1 receptor in the rat brain: distribution, regulation, and relationship to siets of IL-1-induced cellular activation. J Comp Neurol 1995; 361 : 681-698. Cao C, Matsumura K, Yamagata K, and Watanabe Y. Induction by lipopolysaccharide of cyclooxygenase-2 mRNA in rat brain; its possible role in the febrile response. Brain Res 1995; 697:187-96. Matsumura K, Watanabe Y, Imai-Matsumar K, Connolly M, Koyama Y, Onoe H, Watanabe Y. Mapping of prostaglandin E2 binding in rat brain using quantitative autoradiography. Brain Res 1992; 581 : 292-8. Matsumura K, Cao C, Ozaki M, Morii H, Nakadate K, and Watanabe Y. Brain endothelial cells express cyclooxygenase-2 during lipopolysaccharide-induced fever: Light and electron microscopic immunocytochemical studies. J Neurosci 1998; 18: 6279-89. Lacroix S, Rivest S. Effect of acute systemic inflammatory response and cytokines on the transcription of the genes encoding cyclooxygenase enzymes (Cox-1 and Cox-2) in the rat brain. J Neurochem 1998; 70: 452-66. Morgan JI, Curran T. Stimulus-transcription coupling in the nervous system: Involvement of the inducible proto-oncogenes fos and jun. Ann Rev Neurosci 1991; 14: 421-51. Sheng M, Greenberg ME. The regulation and function of c-fos and other immediate early genes in the nervous system. Neuron 1990; 4: 477-85. Hunt SP, Pini A, Evans G. Induction of c-fos-like protein in spinal cord neurons following sensory stimulation. Nature 1987; 328: 632-4. Wan W, Wetmore L, Sorensen CM, Greenberg AH, Nance DM. Neural and biochemical mediators of endotoxin and stress-induced c-fos expression in the rat brain. Brain Res Bull 1994; 34: 7-14. Wan W, Janz L, Vriend CY, Sorensen CM, Greenberg AH, Nance DM. Differential induction of c-fos immunoreactivity in hypothalamus and brain stem nuclei following central and peripheral administration of endotoxin. Brain Res Bull 1993; 32:581-7. Rivest S, Torres G, Rivier C. Differential effects of central and peripheral injection interleukin-l[3 on brain c-fos expression and neuroendocrine functions. Brain Res 1992;
137
29.
30. 31.
32.
33.
34. 35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
587: 13-23. Ericsson A, Kovacs KJ, Sawchenko PE. A functional anatomical analysis of central pathways subserving the effects of interleukin-1 on stress-related neuroendocrine neurons. J Neurosci 1994; 14:897-913. Brady LS, Lynn AB, Herkenham M, Gottesfeld Z. Systemic interleukin-1 induces early and late patterns of c-fos mRNA expression in brain. J Neurosci 1994; 14: 4951-64. Jackson ATK, Janz L, Greenberg AH, Nance DM. Central nitric oxide synthase inhibition blocks endotoxin-induced c-fos protein in the hypothalamus. Society for Neuroscience, Nov., San Diego, CA, 1995. Lacroix S, Vallieres L, Rivest S. C-fos mRNA pattern and corticotropin-releasing factor neuronal activity throughout the brain of rats injected centrally with a prostaglandin of E2 type. J Neuroimnumol 1996; 70: 163-79. Holets V, Elde E. The differential distribution and relationship of serotoninergic and peptidergic fibers to sympathoadrenal neurons in the intermediolateral cell column of the rat: A combined retrograde axonal transport and immunofluorescence study. Neurosci 1982; 7:1155-74. Cechetto DF, Saper CB. Neurochemical organization of the hypothalamic projection to the spinal cord in the rat. J Comp Neurol 1988; 272: 579-604. Appel NM, Elde RP. The intermediolateral cell column of the thoracic spinal cord is comprised of target-specific subnuclei: Evidence from retrograde transport studies and immunohistochemistry. J Neurosci 1988; 8: 1767-87. Derijk RH, Van Kampen M, Van Rooijen N, Berkenbosh F. Hypothermia to endotoxin involves reduced thermogenesis, macrophage-dependent mechanisms, and prostaglandins. Amer J Physiol 1994; 35: R1-R8. Niijima A, Hori T, Aou S, Oomura Y. The effects of interleukin-lon the activity of adrenal, splenic and renal sympathetic nerves in the rat. J Auton Nerv System 1991; 36:183-92. Terao A, Oikawa M, Saito M. Cytokine-induced change in hypothalamic norepinephrine turnover: involvement of corticotropin-releasing hormone and prostaglandins. Brain Res 1993; 622: 257-61. Masana MI, Heyes MP, Mefford IN. Indomethacin prevents increased catecholamine turnover in rat brain following systemic endotoxin challenge. Prog Neuropsychopharmacol Biol Psychiatry 1990; 14:609-21. Endres S, Cannon JG, Ghorbani R, Dempsey RA, Sisson SD, Lonnemann G, Van der Meer JW, Wolff SM, Dinarello CA. In vitro production of IL-1 beta, IL-2-alpha, TNF and IL-2 in healthy subjects: Distribution, effect of cycloxygenase inhibition and evidence of independent gene regulation. Eur J Immunol 1989; 19: 2327-33. Zhang J, Rivest S. Distribution, regulation and colocalization of the genese encoding the EP2- and EP4-PGE2 receptors in the rat brain and neuronal responses to systemic inflammation. Eur J Neurosci 1999; 1: 2651-68. Zhang J, Rivest S. A functional analysis of EP4 receptor-expressing neurons in mediating the action of prostaglandin E2 within specific nuclei of the brain in response to circulating interleukin-lb. J Neurochem 2000; 74: 2134-45. Gross PM, Sposito NM, Pettersen SE, Fenstermacher JD. Differences in function and structure of the capillary endothelium in the supraoptic nucleus and pituitary neural lobe of rats. Evidence for the supraoptic nucleus as an osmometer. Neuroendocrinology 1986; 44: 401-7. Ray PK, Choudhury SR. Vasopressinergic neurons and the associated blood vessels
138
45. 46.
47.
48.
49.
50.
51. 52.
53.
54.
55.
56.
57.
58. 59.
60. 61.
in the rat anterior hypothalamus: an immunohistochemical study. Histol Histopathol 1990; 1: 73-82. Peunova N, Enikolopov G. Amplification of calcium-induced gene transcription by nitric oxide in neuronal cells. Nature 1993; 364: 450-3. Jackson ATK, Greenberg AH, Nance DM. Neurochemical specificity of endotoxininduced c-fos expressing neurons in the hypothalamus. Society for Neuroscience, Nov., Miami, FL. 1994. Daftary SS, Boudaba C, Szabo K, Tasker JG. Noradrenergic excitation of magnocellular neurons in the rat hypothalamic paraventricular nucleus via intranuclear glutamatergic circuits. J Neurosci 1998; 18:10619-28. Nance DM, Wan W, Wetmore L, Greenberg AH. Biochemical and neural mediation of endotoxin induced c-fos protein in the brain. Res Perspect. Psychoneurolmmunol IV. Boulder, CO., April, 1993. Watkins LR, Goehler LE, Relton JK, Tartaglia N, Silbert L, Martin D, Maier SF. Blockade of interleukin-1 induced hyperthermia by subdiaphragmatic vagotomy: evidence for vagal mediation of immune-brain communication. Neurosci Lett 1995; 183:27-31. Watkins LR, Wiertelak EP, Goehler LE, Mooney-Heiberger K, Martinez J, Furness L, Smith P, Maier SF. Neurocircuitry of illness-induced hyperalgesia. Brain Res 1994; 639: 283-99. Watkins R, Wiertelak EP, Goehler LE, Smith KP, Martin D, Maier SF. Characterization of cytokine-induced hyperalgesia. Brain Res 1994; 654: 15-26. Quan N, Whiteside M, Herkenham M. Time course and localization patterns of interleukinl beta messenger RNA expression in brain and pituitary after peripheral administration of lipopolysaccharide. Neuroscience 1998; 83:281-93. MacNeil B J, Jansen AH, Greenberg AH, Nance DM. Activation and selectivity of splenic sympathetic nerve electrical activity response to bacterial endotoxin. Amer J Physiol 1996; 270: R264-R270. MacNeil B J, Jansen AH, Janz LJ, Greenberg AH, Nance DM. Peripherial endotoxin increases splenic sympathetic nerve activity via central prostaglandin synthesis. Am J Physiol 1997; 273:R609-R614. MacNeil B J, Jansen AH, Greenberg AH, Nance DM. The effect of acute adrenalectomy on sympathetic activation induced by peripheral lipopolysaccharide or central PGE2. Am J Physiol 2000; 278: R1321-1328. MacNeil B J, Jansen AH, Greenberg AH, Nance DM. Neuropeptide specificity of PGE2induced activation of splenic and renal sympathetic nerves in the rat. Am J Physiol 2000 (Submitted). Ericsson A, Arias C, Sawchenko PE. Evidence for an intramedullary prostaglandindependent mechanism in the activation of stress-related neuroendocrine circuitry by intravenous interleukin- 1. J Neurosci 1997; 15:7166-79. Jackson ATK. Chemical specificity of endotoxin-induced c-fos expressing neurons in the rat hypothalamus. M.Sc. Thesis, Dept. Physiology, Univ. Manitoba, April, 1999. Sundar SK, Cierpial A, Kilts C, Ritchie JC, Weiss JM. Brain IL-l-induced immunosuppression occurs through activation of both pituitary-adrenal axis and sympathetic nervous system by corticoropin-releasing factor. J Neurosci 1990; 10:3701-6. Del Rey A, Besedovsky HO, Sorkin E, Da Prada M, Arrenbrecht S. Immunoregulation mediated by the sympathetic nervous system, II. Cell Immunol 1981; 63: 329-34. Green-Johnson JM, Zalcman S, Vriend CY, Nance DM, Greenberg AH. Role of norepinephrine in suppressed IgG production by eplipsy-prone mice. Life Sci 1996; 59: 1121-32.
139 62.
Vriend CY, Zuo L, Dyck D, Nance DM, Greenberg AH. Central administration of Interleukin-l[3 increases norepinephrine turnover in the spleen. Brain Res Bull 1993; 31: 39-41. 63. Livnat S, Madden KS, Felten DL, Felten SY. Regulation of the immune system by sympathetic neural mechanisms. Prog NeuropsychoPharmacol Biol Psychiatry 1987; 1: 145-52. 64. Romano TA, Felten SY, Felten DL, Olschowka JA. Neuropeptide-Y innervation of the rat spleen: Another potential immunomodulatory neuropeptide. Brain Behav Immunity 1991; 5:116-31. 65. Spengler RN, Allen RM, Remick DG, Strieter RM, Kunkel SL. Stimulation of a-adrenergic receptor augments the production of macrophage-derived tumor necrosis factor. J Immunol 1990; 145: 1430-4. 66. Spengler RN, Chensue SW, Giacherio DA, Blenk N, Kunkel SL. Endogenous norepinephrine regulates tumor necrosis factor-a production from macrophages in vitro. J Immunol 1994; 152: 3024-30. 67. Hu X, Goldmuntz EA, Brosnan CF. The effect of norepinephrine on endotoxin-mediated macrophage activation. J Neuroimmunol 1991; 31: 35-42. 68. Abrass CK, O'Connor SW, Scarpace PJ, Abrass IB. Characterization of the [3-adrenergic receptor on the rat peritoneal macrophage. J Immunol 1985; 135: 1338-41. 69. Wan W, Vriend CY, Wetmore L, Gartner JG, Greenberg AH, Nance DM. The effects of stress on splenic immune function are mediated by the splenic nerve. Brain Res Bull. 1993; 30: 101-5. 70. Pan L, Pylypas S, Sanders V, Greenberg AH, Nance DM. Central injections of prostaglandin activate the hypothalamus and suppress splenic cytokine mRNA production. Society for Neuroscience, Nov., 1997, New Orleans, LA. 71. Meltzer JC, MacNeil B J, Sanders V, Vriend CAY, Jansen AH, Greenberg AH, Nance DM. Effects of stress on in vivo splenic TNF-c~ and IL-I[3 mRNA levels. Society for Neuroscience, Oct., 1997, New Orleans, LA. 72. Meltzer JC, MacNeil B J, Sanders V, Grimm PC, Vriend J, Jansen AH, Greenberg AH, Nance DM. The immunosuppressive effects of stress on splenic cytokine mRNA levels are mediated by adrenal dependent and independent mechanisms. Society for Neuroscience, Nov, 1998, Los Angeles, CA. 73. Nance DM, Greenberg AH, Jackson ATK. Central catecholamine involvement in the hypothalamic induction of c-fos after endotoxin treatment. Society Neuroscience, Nov., 1995, San Diego, CA.
This Page Intentionally Left Blank
New Foundation of Biology
141
Edited by I. Berczi and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
Behavioral and Central Neurochemical Consequences of Cytokine Challenge: Relationship to Stressors
HYMIE ANISMAN 1, SHAWN HAYLEY 1 and ZUL MERALI 2
~Institute of Neurosciences, Carleton University and 2School of Psychology and Department of Cellular and Molecular Medicine, University of Ottawa, Ottawa, Canada
ABSTRACT Multidirectional communication likely occurs between the immune, neuroendocrine, autonomic and central nervous systems. In this respect, it has been proposed that signaling molecules (cytokines) released from activated immune cells may serve as messengers (immunotransmitters) stimulating neuroendocrine and central neurotransmitter processes. Further, given that cytokines, such as interleukin-l[3 (IL-I[5) and tumor necrosis factor-or (TNF-c~), elicit neurochemical changes reminiscent of those provoked by stressors, the view has been expressed that the central nervous system (CNS) interprets immune activation much as if it were a stressor. Among other things, these cytokines stimulate hypothalamic-pituitary-adrenal activity, provoking the release of corticotropin releasing hormone (CRH) from the paraventricular nucleus of the hypothalamus, and hence secretion of pituitary ACTH and adrenal glucocorticoids. As well, like stressors, cytokines provoke increased utilization of norepinephrine, dopamine and serotonin within various brain regions. This includes the hypothalamus, which influences neuroendocrine processes, as well as extrahypothalamic regions, such as the amygdala and prefrontal cortex, which are involved in fear and anxiety. In addition to their independent actions, various cytokines may act synergistically in modulating neuroendocrine functioning. Moreover, paralleling the actions of stressors, acute cytokine exposure may have protracted effects on CNS processes. Having been exposed to a cytokine, subsequent reexposure to the cytokine (or to a stressor) may influence neurochemical functioning, so that even sub-effective cytokine doses have marked effects (sensitization). Interestingly, such effects increase with the passage of time, possibly reflecting diverse processes including such things as phenotypic changes of CRH neurons within the external zone of the median eminence so that the terminals coexpress both CRH and arginine vasopressin. Inasmuch as cytokines induce stressor like effects, efforts have been devoted to the analysis of the behavioral changes exerted by these immunotransmitters. The best known of these effects is the sickness profile induced by proinflammatory cytokines and bacterial endotoxins (e.g., reduced activity, ptosis, piloerection, signs of disturbed blood flow, curled body posture, reduced food consumption, and reduced intake of highly palatable food substances). In part, these effects may reflect the malaise engendered by the challenges, but there is reason to suppose that they also induce anhedonia (diminished reward obtained from otherwise rewarding stimuli),
142
as well as anxiety. Moreover, chronic administration of some cytokines (e.g., IL-2) provoke cognitive disturbances, such that working memory processes are disrupted. It is provisionally suggested that these cytokines may play a role in the provocation or maintenance of depressive disorders, as well as illnesses involving affective components.
1.
INTRODUCTION
It's a jungle out there! If it's not one thing, then its another! On a day-to-day basis humans and infrahumans face a series of challenges of varying types and severity, some needing immediate attention, while the response to others can be held in abeyance. Some challenges, termed processive stressors, involve higher order sensory processing (e.g., in animals these take the form of conditioned fear cues, exposure to a novel environment or predators, painful or uncomfortable events such as footshock, cold water immersion or restraint) and are classified as being either neurogenic (i.e., of physical nature) or psychogenic (psychological origin). These may occur on an acute (single events) or on a chronic (repeated) basis, they may be either predictable or unpredictable, and appear as either continuous, discontinuous or intermittent events. A second type of frequently encountered threat involves insults of a physiological nature (systemic stressor), including immunological challenges. It was suggested that challenges to the immune system affect neuroendocrine and neurotransmitter functioning, in much the same way that processive stressors induce such an outcome. In effect, it was argued that in addition to its other functions, the immune system serves in a sensory capacity informing the brain of antigenic challenge [1]. In this respect, cytokines released from activated macrophages may serve as signaling molecules between the immune system and the CNS, and the CNS may interpret immune activation as if it were a stressor [2, 3]. In response to stressors animals adopt a series of behavioral strategies to escape or mitigate the impact of the challenge. Concurrently, a series of neurochemical changes occur which presumably have the effect of either enhancing the animal's physical or cognitive functioning (to deal with the insult), blunt the physical or psychological impact of the stressor, or limit the disturbances that might occur if physiological responses were not held in check (e.g., glucocorticoid inhibition of excessive immune activation) [4]. Among other things, stressors will promote the utilization and synthesis of monoamines (norepinephrine, dopamine, serotonin) amino acids, acetylcholine, endorphins and various other peptides (corticotropin releasing hormone, arginine vasopressin) in numerous brain regions [5]. Some of these may be essential for the elicitation of emotional responses or even emotional memories (e.g., various aspects of the amygdala), whereas others, such as the prefrontal cortex (PFC), may be involved in the interpretation or appraisal of environmental stimuli that activate the amygdala. Still other brain regions, such as the nucleus of the solitary tract (NTS), may be involved in activation of autonomic functioning, or activation of the hypothalamus, which in turn stimulates pituitary secretion of ACTH and hence adrenal glucocorticoid release. It is assumed that when the stressor is of moderate severity and, is applied acutely on a fairly predictable basis, the utilization of central neurotransmitters is met by increased synthesis, and hence the stores of the transmitters are maintained. Under some conditions (e.g., uncontrollable events) the levels may decline owing to utilization exceeding synthesis, and the animal may be rendered more vulnerable to adverse effects of the insults. Interestingly, if the stressor continues (chronic) a compensatory increase of synthesis may evolve (possibly owing to alterations of autoreceptor sensitivity) resulting in elevations of the transmitters, and presumably vulnerability
143
to pathology may be averted. Of course, there are conditions wherein the adaptation is less readily observed (e.g., within some strains of mice, or following an unpredictable, intermittent, chronic schedule), and hence vulnerability to pathology is maintained. It ought to be underscored, as well, that if the stressor is sufficiently sustained, then this continuous strain (termed allostatic load) may ultimately lead to pathological outcomes of one sort or another, depending on the weak link within the system [6].
2.
CYTOKINE-ELICITED NEUROCHEMICAL ALTERATIONS
As will be seen shortly, peripheral cytokine activation elicits a wide range of neuroendocrine and central neurotransmitter alterations. One of the best documented series of changes elicited by central and systemic interleukin-l[3 (IL-I[3) administration (and to a lesser extent TNF-c~ and IL-6) is the activation of HPA axis. Ordinarily, IL-1 [3 potently stimulates pituitary ACTH and adrenal corticosterone release. This appears to stem from secretion of CRH from the paraventricular nucleus (PVN), as the pituitary and adrenal hormonal effects elicited by the cytokine could be prevented by passive immunization with antisera to CRH and by CRH receptor antagonists (see Dunn, 1995). There is reason to believe that NE may promote IL-I[3 induced HPA changes, while nitric oxide restrains hypothalamic-pituitary-adrenal (HPA) responses to proinflammatory stimuli [7]. However, while IL-1[3 increased c-fos expression of CRH-immunoreactive neurons within the PVN, it appears that blood-borne cytokines (after i.v. injection) exert their effects through nerve terminals in the median eminence, whereas i.c.v, administration of IL-I[3 activates CRH neurons of the parvocellular portion of the PVN [8, 9]. The process by which peripherally released cytokines influence CNS functioning has yet to be fully elucidated, and several potential hypotheses have been offered. According to one view, cytokines released from activated immune cells may serve to stimulate cytokine receptors present on the dendrites of the vagal nerve. Indeed, both IL-I[3 and LPS induce c-los expression at the nodose ganglion and provoke electrical activity of neurons within the gastric branch of the vagus [ 10, 11]. After peripheral activation, the vagus may transmit neural information, via the nodose ganglion, to brainstem regions such as the NTS, which in turn stimulates the hypothalamus. An alternative explanation (although not necessarily a mutually exclusive one) is that macrophage-derived cytokines, such as IL-I[3, IL-6, and tumor necrosis factor-~ (TNF-c~), as well as IL-2 secreted from T-helper cells, may act as mediators (immunotransmitters) in this respect. Despite the fact that peripheral cytokines comprise fairly large molecules, entry into brain may occur where the blood-brain barrier is less restrictive (e.g., the organum vasculosum laminae terminalis), or by a saturable transport system [12, 13]. As well, blood brain barrier permeability may be compromised under some pathological conditions (e.g., seizure) culminating in greater accessibility to the brain. In addition, it is possible that the cytokines may interact with receptors at circumventricular sites, leading to either central cytokine production or activation of prostaglandins which serve to excite parenchymal cells [ 14, 15]. Although circulating immune cells represent a source of cytokines that can potentially influence CNS processes, several cytokines and their receptors have also been documented within the brain [ 16, 17]. Cytokine synthesis can be elicited by a number of factors, including systemic bacterial endotoxins [18], central endotoxin application [14], as well as by brain injury, cerebral ischemia, and seizure [19, 20]. The functional significance of alterations in central cytokine levels remains to be elucidated. Proinflammatory cytokines, such as
144
IL-I[3, may function in a reparative capacity in neurological diseases or brain trauma, or alternatively may actually promote neuronal damage following central insults [19]. In any case, acute events (e.g., stroke), chronic degenerative disease (e.g., Alzheimer's) or central autoimmune neuropathology (e.g., multiple sclerosis) all have common elements of inflammation characterized by abnormal accumulation of immune components (e.g., complement and infiltrating leukocytes) coupled with an upregulation of central cytokines such as IL-I[3 and TNF-ct [19, 21]. Of course, it is possible that cytokines may simply represent bystanders, not involved in neurodegenerative processes at all. However this is unlikely given that cytokines likely orchestrate the infiltration of immune cells into brain, as well as modulating their actions within the brain parenchyma, much like their welldocumented effects in periphery. Moreover, de novo synthesized brain cytokines may activate apoptotic pathways, as well as induce the production of potentially toxic free radicals or arachidonic acid metabolites [15, 22]. Central cytokine alterations are not only elicited by insults associated with brain injury, but may be affected by stressful events. To be sure, there have been few studies that assessed whether processive stressors influence central cytokine levels or functioning. However, it does appear that stressors influence IL-I[3 mRNA or protein levels within the brain and pituitary [20, 23, 24, 25, 26]. Moreover, stressors induce an acute phase-like response, including the reduction of acute-phase negative reactant (corticosteroid binding globulin), increased serum levels of acute phase positive reactants (haptoglobin and alpha 1-acid glycoprotein), as well as increased body temperature [27]. Furthermore, consistent with the proposition that IL-I[3 plays a fundamental role in the provocation of behavioral disturbances associated with stressors (e.g., the depression-like behavioral impairment typically evident in a shuttle escape test), intracebroventricular pretreatment with IL-1 receptor antagonist (IL-lra) blocked the development of behavioral disturbances ordinarily elicited by uncontrollable stressor application [11 ]. Likewise, IL-lra pretreatment was shown to attenuate the ACTH, as well as the hypothalamic monoamine alterations ordinarily elicited by stressors [28]. As already alluded to, like processive stressors, the administration of proinflammatory cytokines, such as IL-I[3 and TNF-ct, as well as bacterial endotoxins, such as lipopolysaccharide (LPS), are particularly effective in provoking HPA activation. In particular, cytokines readily promote the release of hypothalamic monoamines [2, 29], CRH release from the paraventricular nucleus (PVN) of the hypothalamus, and hence pituitary ACTH release and adrenal glucocorticoid secretion [2]. However, in several respects the neural circuitry associated with cytokine treatments are distinguishable from that associated with more traditional stressors. For instance, although both processive stressors and systemic (metabolic) stressors increase HPA activation, processive stressors appear to involve activation of limbic forebrain regions, which in turn, may stimulate GAB Aergic neurons at an intervening synapse (e.g., bed nucleus of the stria terminalis) before acting upon corticotropic PVN cells. In contrast, there is reason to suspect that HPA alterations elicited by a systemic stressor (e.g., cytokines) may reflect the stimulation of brainstem nuclei which directly innervate the PVN [30]. This is not to say that cytokines are without effect on amygdaloid activity, as IL-I[3 increased Fos-immunoreactivity in the central amygdala and bed nucleus of the stria terminalis, as well as in catecholaminergic neurons of the NTS and ventrolateral medulla (the latter projecting to regions of the PVN) [8]. Interestingly, bilateral lesions of the central amygdala reduced HPA activity, as well as c-fos expression within the bed nucleus and in the aforementioned brainstem regions [31]. Moreover, IL-I[3 and TNF-c~ influenced monoamine activity within several extrahypothalamic sites, including the central amygdala [29, 32].
145
3.
INTERLEUKIN- 1[3: ACUTE EFFECTS
Following IL-1[3 or LPS challenge c-fos mRNA expression may follow a biphasic temporal course across brain regions [33, 34]. The central circuitry activated following systemic cytokine administration may involve different mechanisms. The initial changes observed may stem from actions associated with cytokine entry into the brain (i.e., promotion of neuroendocrine changes and the neural transduction of peripheral signals), whereas the subsequent effects reflect the cellular activation associated with IL-I[3 along various diffusion routes or effects secondary to de novo synthesis [33, 34]. In addition to the neuropeptide variations, systemic administration of either LPS or IL-I[3 promoted increased norepinephrine (NE) activity and/or affected the levels of the amine in several hypothalamic nuclei, including the PVN, median eminence, medial basal, and lateral nuclei [35]. As seen in Figure 1, within the locus coeruleus, arcuate nucleus plus median eminence, and the PFC, IL-I[3 increased the utilization of NE, as reflected by elevated MHPG accumulation. Paralleling such postmortem studies, in vivo analyses confirmed that systemic IL-I[3 increased the release of hypothalamic NE and dopamine (DA) [28, 36]. These effects, it will be recognized, were reminiscent of those frequently associated with neurogenic or psychogenic stressors. The impact of cytokines and bacterial endotoxins are not limited to variations of NE and DA functioning, and marked variations have been observed with respect to 5-HT activity. For instance, IL-I[3 provoked elevations of tryptophan in brainstem, hypothalamus and PFC [2], and the accumulation of the serotonin (5-HT) metabolite, 5-HIAA, was increased following systemic administration of LPS and IL-I[3 [32, 35]. Interestingly, while the ganglionic blocker, chlorisondamine, prevented the effects of LPS on 5-HT turnover, changes of central NE utilization were largely unaffected [2]. Furthermore, nitric oxide synthase inhibition did not affect central NE activity induced by IL-I[3 and LPS, but precluded the effects ordinarily provoked with respect to 5-HT turnover [2, 37]. Thus, it appears that independent processes are responsible for the effects of IL-1 [3 and LPS on central NE and 5-HT activity. For several reasons, postmortem analyses may not provide a sufficiently clear view of neurotransmitter alterations associated with environmental challenges. Among other things, postmortem samples offer an index of amine and metabolite levels at only a single time point, which is obviously problematic given the dynamic, time-dependent variations of neurotransmitter functioning that occur following cytokine treatment. Furthermore, the uncertain fate of amine metabolites over lengthy periods following cytokine treatment may not permit accurate appraisal of utilization rate, and hence amine functioning may be more clearly reflected by in vivo analyses. In fact, in vivo microdialysis samples revealed that IL-I[3 stimulated hypothalamic [36] and hippocampal [3, 38] 5-HT release, and increased in vivo 5-HIAA at the nucleus accumbens. Interestingly, as seen in Figure 2, the superimposition of a mild stressor (air-puff) in IL-1 [3 treated rats appreciably enhanced the serotonergic changes in what appeared to be a synergistic fashion [3, 39]. Although the neuroendocrine and neurotransmitter changes elicited by IL-I[3 are reminiscent of those elicited by stressors, the effects of these treatments are not entirely congruent [40]. While stressors reliably enhance NE activity within the amygdala and DA within the PFC and nucleus accumbens [5], IL-113 treatment did not elicit such effects in either postmortem or in vivo analyses [3, 32, 39]. However, when accumbal DA was assessed in vivo following LPS treatment marked amine variations were evident (see Figure 3). Approximately 90 min following systemic LPS administration, in vivo elevations of accumbal DA were apparent which persisted for about 90 min. Accumbal 5-HIAA were also elicited by the treatment; these variations occurred sooner and were far more persistent. While the source for the divergent courses of DA and 5-HT release has
146
PVN 5-HIAA t4 O
12 zl/p'.*'f/.s:~'/#
Q. Ob
E r
Ff.,~..-/~j.u.p_z~ ,,,-zz / ~,-z.s-.z,,,- f ~
10 8
v
< <
6
I! LO
4
t1.1#/lJ_,r
r
SAL
0.025
0.1
0.05
AMY 5-HIAA r..--.
o Q.
~r
25
/i ..,.,,~..
,-. . . . .
E v
10
.
.... -
-7 NB
15
c
< < -r" ! t~
J, ~'~'~r..,
20
3-
5
SAL
0.025
U///////~ r
NN NN 0.05
0.1
0.05
0.1
PFC 5 - H I A A
.=_
10
o x.. Q.
8
E Cv
< < -r1 LD
SAL
0.025
Dose (ug) Figure 1. Mean (+ S.E.M.) 5-HIAA accumulation in the paraventricular nucleus of the hypothalamus (PVN), central amygdala (AMY) and the medial prefrontal cortex (PFC) 1 hour following intraperitoneal administration of low doses of IL-I[3. * p <.05 relative to saline treated animals. (from Brebner et al., 2000).
147
Hippocampal 5-HT
250-
200-
c (D 69
.Q
o
150IL-1
v
Saline
I
50-
................................ l Ti baseline Inj
stress
0
-2
0
2
4
6
8
10
12
14
Sample #
t 50-
Hippocampal 5-HIAA 140' Z: ,,,,.,,.
t30.
:il
0
....~ 120,< <: ,,,.,.,.,
3"
~
IL-1
tlo-
100Saline
I
90:
I
................... lnj
i
0
2
4
stress 6
8 Sample
10
12
14
#
Figure 2. Extracellular 5-HT and 5HIAA at the dorsal hippocampus, expressed as a percent of baseline, over 30 min dialysate samples. Following 4 baseline samples, rats were injected with saline (squares) or IL-I[3 (circles) and 5 samples (30 min) collected. Air-puff stress (5 puffs of 1 sec duration) was then delivered and dialysates collected over 5 additional periods. Filled-in symbols denote significant differences form the average of the baseline period, asterisks denote significant differences from saline treated rats (from Merali et al., 1997).
148
yet to be deduced, it seems likely that different neuronal networks within a particular brain region may be dissimilarly responsive to endotoxin challenge [41]. Furthermore, in evaluating the effects of LPS (relative to those elicited by IL-I[3) it ought to be recalled that LPS stimulates the release of IL-1 [3, IL-6 and TNF-c~ from macrophages, and thus the central actions of LPS may involve the synergistic effect of several cytokines, rather than IL-I[3 alone. Further to this point, IL-I[3 may also promote other cytokine alterations which could elicit behavioral and neurochemical changes. As" well, it was reported that at doses sufficient to attenuate the effects of IL-1 [3, the administration of the antagonist IL-lra was ineffective in precluding the brain NE, ACTH or corticosterone activation elicited by LPS [37]. Given the marked effects of stressors on NE activity within the central amygdala [40], the finding that IL-1 [3 did not elicit such an effect was somewhat surprising. Accordingly, we assessed whether amygdala NE variations would be detected in vivo, which as indicated earlier, yields a better view of the effects stemming from endogenous or exogenous challenges. Moreover, inasmuch as marked interindividual differences have been reported concerning the impact of stressors, in collaboration with D. McIntyre, we evaluated the effects of systemic IL-1 [3 on amine variations in two rat strains selectively bred for differences of their amygdala excitability (realized by either fast or slow development of kindled epileptic seizures elicited by focal amygdala stimulation) [42]. In addition to differences in seizure susceptibility, these strains also exhibit marked differences in their behavioral profiles in response to stressors. The SLOW seizing rats tend to be more fearful, while the FAST seizing rats are somewhat hyper-reactive and show disturbed habituation [42]. In response to 1.0 ~tg of IL-1 ~ (i.p.) a marked in vivo release of NE was elicited from the central amygdala of the emotional SLOW rats, whereas hardly any change from baseline was seen in the less emotional FAST rats. Whether the observed effects were related to amygdala excitability or to emotionality differences, remains to be determined. Whatever the case, these data make it eminently clear that the central responses to cytokines, as in the case of other challenges such as psychogenic stressors, are largely dependent on characteristics of the organism being studied.
4.
TNF-ot: ACUTE EFFECTS
The neurochemical consequences of TNF-c~ have not been assessed as extensively as those associated with IL-1 [3. Yet, it is believed that this cytokine plays a role in numerous pathological states (anorexia, reduced social exploration) [10, 29], autoimmune disorders (rheumatoid arthritis, multiple sclerosis) and neurodegenerative conditions (Alzheimer's disease) [43], and may be involved in mood disorders, including depressive illness [44]. While apparently less potent that IL-I[3, it has been demonstrated that TNF-ct stimulates plasma corticosterone secretion [29, 45, 46]. Likewise, the cytokine provoked central monoamine alterations, including an increase of central tryptophan and MHPG accumulation [45], a reduction of NE within the locus coeruleus and dorsal hippocampus, and increased 5-HT turnover within the PVN and central amygdala [29]. The effects of TNF-ct, however, could be distinguished from those of other cytokines. For instance, while both central IL-I[3 and IL-2 administration influenced hippocampal 5-HT activity, TNF-~ did not promote such an outcome [47].
5.
SYNERGISTIC EFFECTS OF PRO-INFLAMMATORY CYTOKINES
As indicated earlier, since IL-I[3 elicits several effects similar to those ordinarily provoked by LPS, it has been assumed that at least some of the endotoxin effects involve IL-I[3. However,
149
160
r
140
03 co s
120 -
--o--
Saline
--*--
LPS
-,*
" " " -'-//"'l"
LPS
'
0 '~
100
80 0 - - - ' 1 1I 2
~
T .............
3
4
~ ................. [ . . . . . .
~
5
7
6
~
'
-
'
~
'
8
-
~
r
-
9
'
-
-
~
10
T
11
--~
12
I30
120
I
C @ t~ 0
• I
I
110
!00
~
9 0 -
i ~
-80-
-
T
T,-"I
7
"
T___!
70 ,0 .... 7 . . . . . T ............ 1 1 2 3
1
"f
T .........
4
5
6
-'1
7
"'
I
" "
8
" t .......
9
"l.
10
.
.
.
.
1'1
.
~ ....
12
30 minute samples Figure 3. Mean (+ S.E.M) interstitial DA (upper panel) and 5-HIAA (lower panel) at the nucleus accumbens as a percentage of baseline (four samples) and after i.p. administrationof either saline or LPS (100 ~tg). Points represents 30 min dialysate samples. * P < 0.05 relative to saline. (from Borowski et al., 1998).
the finding that IL-lra did not attenuate several actions elicited by LPS has led to the view that factors other than, or in addition to, IL-I~ are responsible for the effects of LPS [37]. Indeed, there is evidence indicating that IL-I[3 acts conjointly or synergistically with IL-6 and/or TNF-ot [46, 48]. Synergistic effects between IL-I[3 and IL-6 have been reported with respect to the enhanced release of adrenal corticosterone [48] and pituitary ACTH [49], while IL-I[3 and TNF-o~ synergistically stimulated IL-11 through the production of prostaglandin-E2 in rheumatoid synovial fibroblasts [50]. As well, IL-I[3 and T N F - ~ synergistically influenced signs of illness, including reduced social exploration, food intake, and consumption of a highly palatable food source [10, 46, 51]. As well, IL-I[3 plus T N F - ~ elicited synergistic effects with respect to corticosterone and illness were not evident in response to IL-[3 plus IL-6, or IL-6 plus TNF-ot [46].
150
-(3
30
,-'.,,"//.."A
v tO
6O
20
O O
!E o O
10
E
O3
Veh Veh
03
IL-1 Veh
I L-6 Veh
IL-1 IL-6
30
v (:1) c" 0
20 03 0 0
-T-
0
0
10
E
09
13_ Veh Veh
IL-6 Veh
TNF Veh
IL-6 TNF
72 23 v
30
(1) C O k_
(1)
,r O3 O O
20 --T-
O
O ~3 E O3 e~
10
--F-
~;...".:;//.y...-" ~ . , ./,
~;.'Z "~.~/~
s Veh Veh
1L-1 Veh
TNF Veh
IL-1 TNF
Figure 4. Plasma corticosterone concentrations 1 hour after administration of a single cytokine or coadministration of two cytokines.
The upper panel show the individual and combine effects of IL-1[3 and IL-6; the middle
panel shows the effects of IL-6 and TNF-cz; the lowest panel shows the synergistic effects of IL-I[3 and TNF-ot (from Brebner et al., 2000).
151
Centrally, IL-I[3 and mild stressors synergistically increased monoamine variations within the PFC, nucleus accumbens and dorsal hippocampus [3]. Moreover IL-I[3 plus TNF-c~ acted synergistically to influence neurotoxicity in mixed neuronal/glial cell cultures containing IFN~, [52], and along with IL-6 to provoke TNF-~ mRNA expression in rat C6 glioma cells [53]. However, synergistic effects of IL-I[3, IL-6 and TNF-c~ were not detected with respect to NE, DA or 5-HT activity in several brain regions where IL-1 and TNF-c~ were found to elicit acute effects [46]. However, in these studies multiple dose and time-dependent variations of the cytokines were not explored, and thus pending detailed in vivo analyses it may be premature to exclude synergistic central actions of these proinflammatory cytokines.
6.
SENSITIZATION OF CYTOKINE-INDUCED NEUROCHEMICAL CHANGES
In addition to immediate neurochemical alterations, stressors may proactively influence the central neurochemical changes introduced by subsequently applied stressors. Ordinarily, the hormonal and neurotransmitter alterations elicited by stressors persist for only a brief period; however, upon reexposure to a stressor (even of reduced severity) marked monoamine variations may be induced in several brain regions [54]. These effects are not necessarily stressor-specific and may be evident even when the reexposure stressor differs from the initial stressor experience (cross-sensitization) [5] or involves a pharmacological challenge that affects monoamine functioning [54]. It is interesting that the sensitization of a variety of neurochemical systems may be subject to time-dependent changes wherein the magnitude of the response increases progressively with the passage of time since initial stressor inception [55]. As in the case of the monoamine sensitization, time-dependent hypothalamic peptide changes are observed following stressor exposure, as well as in response to IL-I[3 [56, 57]. Specifically, with the passage of time following the initial challenge a phenotypic change occurred within CRH neurons of the PVN that terminated in the external zone of the median eminence. Ordinarily, most terminals within the external zone of the median eminence contain CRH, while only about 20% contain arginine vasopressin (AVP). However, over time following stressor or IL-I[3 challenge the colocalization of AVP and CRH was greatly increased. Since CRH and AVP synergistically stimulate pituitary ACTH secretion, the elevated levels of the peptides might account for the increased HPA responsivity observed upon subsequent exposure to a stressor or to cytokine challenge. While not necessarily permanent, this effect was long lasting, still being evident at about 6 weeks following initial treatment. As in the case of IL-113, administration of TNF-c~ not only elicited immediate behavioral and neurochemical consequences, but also provoked profound time-dependent sensitization effects. Ordinarily, systemic administration of recombinant human TNF-c~ (1.0, 2.0 and 4.0 ~tg) dosedependently influenced sickness behaviors, as reflected by reduced consumption of a highly favored solution (chocolate milk), social interaction, locomotor activity, and general appearance (e.g., lethargy, ptosis, ruffled fur), as well as elevated plasma corticosterone concentrations [29, 46]. Following administration of human TNF-c~ (4.0 ~tg), reexposure to a modest dose of TNF-c~ (1.0 ~tg) was associated with a pronounced time-dependent sensitization with respect to sickness behavior, as well as circulating corticosterone concentrations [29]. Figure 5 shows sickness behavior as reflected by general appearance of the animals (lethargy, ptosis, ruffled fur) rated on a 4-point scale (other aspects of sickness were similarly affected, e.g., locomotor functioning and social interaction). When acutely administered, TNF-c~ (1.0 ~tg) had only modest effects on sickness behaviors measured 1 hour later, as did a higher dose (4.0 ~tg) administered 2 weeks earlier. Likewise, among mice treated with the 4.0 ~g dose, subsequent
152
reexposure to the lower dose of the cytokine 1 and 7 days later induced little evidence of sickness. However, if the cytokine reexposure occurred 14 or 28 days after initial treatment, then marked signs of illness were evident. Plasma corticosterone variations associated with TNF-c~ reexposure were also found to be progressively greater with time. Acute administration of 1.0 ~tg of TNF-ct administered 1 hour earlier had a modest effect on plasma corticosterone, whereas 4.0 gg administered 14 days previously had no carryover effects in this respect. If mice had received TNF-c~ (4.0 ~g) followed subsequently by a low dose of the cytokine (1.0 gg) one day after initial treatment, then a desensitization effect was induced (i.e., corticosterone levels were lower than those seen 1 hour after cytokine administration in animals that had not previously received the TNF-c~ treatment). With progressively longer reexposure intervals the plasma corticosterone response increased, so that at a 28 day interval corticosterone levels exceeded those observed after a single injection (see Figure 5). Such an outcome was not unique to the human recombinant form of this cytokine, as we have observed a similar sensitization effect at the 28 day interval using the murine form of the cytokine [58]. Thus, the sensitization effects were not due to immune changes related to cross-species cytokine effects (i.e., the use of the human recombinant cytokine in the mouse). Furthermore, the outcome was probably not a reflection of effects unique to the murine form of the cytokine, including activation of both the p55 and the p75 receptor, as opposed to just the p55 receptor activation induced by human TNF-et. In fact the sensitization of sickness behaviors was slightly greater using the human form of TNF, suggesting that the p55 receptor may be more important than p75 in this respect. Indeed, it has been reported that the p55 receptor may mediate the preponderance of the neurotoxic effects of TNF-ct and may also play a more prominent role than the p75 subtype in the activation of nuclear factor KB and a host of intracellular second messengers (e.g., particularly caspases) [59]. In addition to the corticosterone alterations, TNF-et was also associated with time-dependent monoamine alterations, but the nature of these effects were dependent upon the specific amine and brain region being assessed. The utilization of NE within the PVN, as reflected by MHPG accumulation, was subject to a sensitization effect that paralleled the corticosterone alterations, although the magnitude of this effect was less pronounced (see Figure 5). Specifically, the metabolite accumulation was greater when TNF-ct was re-administered at longer intervals following initial cytokine treatment. Inasmuch as NE may play a fundamental role in the provocation of CRH release at the PVN, which then comes to affect adrenal glucocorticoid release, it is possible that the sensitization with respect to corticosterone changes were related to the NE alterations. A very different profile was apparent within the mPFC, central amygdala, and hippocampus. When administered on a single occasion TNF-c~ did not influence NE activity within the central amygdala, and produced only a modest increase of NE utilization within the mPFC. However, within both regions, a marked sensitization was evident upon reexposure to the cytokine 1 day later (Figure 6). At longer intervals, TNF-ct increased NE utilization within the PFC, but the sensitization was entirely absent. Interestingly, in contrast to NE variations, the accumulation of 5-HIAA within the mPFC increased over the course of the reexposure interval, peaking at 2 weeks following initial treatment. Evidently, TNF-ct may induce a sensitization effect, such that the amine changes exerted by the cytokine are augmented upon cytokine reexposure, but the time dependent profile is readily distinguishable from that seen at the hypothalamus, and is also dependent upon the specific transmitter being determined.
153
4 :3
3 v 9 O
2
--J O >
O
0
Sat Sal
Sal TNF TNF Sat
1 7 14 28 TNF-cz reexposure interval (days)
"'~ t:~
!.
i
35
F.7
30oJ
C 0
25 20
0
.o
l::t 0 0
15
E
~~
"1
~ a.
s
0
Sal Sal
Sal TNF TNF Sat
i 7 t4 28 TNF-o: reexposure interval (days)
8 ,.....,,,
.c9 7
v..4 .....
i~"
/ A
.t / .#
/
O Q,
E 5
.,.,.,i k.,,,.oH~
x ~,-~ / .,4
/ .'" z /
i/z ~! -'
z A , -
r
C
/"
'''
o~/.,~ / ,.. ,o"/
I Z
>
r ,, ~ !
i 0
Sal
Sal
Sal
TNF Sa~ .
TNF
1
7
/., ...o-
r r
14
......
28
TNF-c~ reexposure interval (days)
Figure 5. Overall sickness profile (upper panel), plasma corticosterone concentrations (middle panel), and MHPG within the PVN (lowest panel) among mice exposed to various T N F - ~ regimens. All mice received two i.p. injections; the three groups to the left (solid bars) received either saline only, saline followed two weeks later by the low dose of TNF-c~ (1.0 ~tg) or a high dose of TNF-c~ (4.0 ~tg) followed two weeks later by saline treatment. The four groups on the fight (hatched bars) received two injections of TNF-cq an initial 4.0 ~tg dose followed by a second 1.0 ~tg dose either 1, 7, 14 or 28 days later. * P < 0.05 relative to saline treated mice, ** P < 0.05 relative to mice that received acute T N F - ~ (modified from Hayley et al., 1999).
154
lo
*w
P r e f r o n t a l Cortex
0
E
s
t
.~c 4
(.9 a 3:: 2
Sal Sal
Sal TNF
TNF Sai
14
7
28
TNF-c~ reexposure interval
Central Amygdala // // //
10 ,
~
(D O
/
C~
E c 9
/
// // // // // // //
8
6 4
/
// /
/
//
// //
I
c
.
Sal Sal
,
............... :
Sal TNF
TNF Sal
/
//
,/i
/
/
/
/
/ /
. .. f...,
1
7
14
28
TNF-~ reexposure interval
Figure 6. Mean (+ S E M ) concentrations of M H P G within the prefrontal cortex and the central amygdala as a function o f the T N F - treatments mice received. Mice received either saline on o c c a s i o n s , saline f o l l o w e d by TNF-ot (1.0 ~tg), or TNF-ot (4.0 ~tg) f o l l o w e d two weeks later by saline. The four groups on the right (hatched bars) received an initial 4.0 ~tg dose o f TNF-ot f o l l o w e d by a s e c o n d 1.0 ~tg dose either 1, 7, 14 or 28 days later. * P < 0.05 relative to
saline treated ( f r o m H a y l e y et al., 1999).
7.
NEUROPEPTIDE VARIATIONS
As indicated earlier, Tilders and his associates demonstrated that IL-1 [3 administration provoked time-dependent variations of CRH and AVP coexpression within the external zone of the median eminence. Thus, it was of interest to establish whether TNF-ot would likewise promote such an outcome. Moreover, given that stressors markedly affected amygdala functioning, in our experiments we also assessed the changes of CRH that occurred within this region. In these
155
experiments TNF- (4.0 ~tg) was administered to mice, and then at 1, 7, 14 or 28 days afterward half the mice received saline and half received reexposure to the cytokine (1.0 ~tg). Thus, we could determine whether the reexposure session was essential for particular peptide variations. As seen in Figure 7, within the external zone of the median eminence, immunoreactivity for both CRH and AVP was increased at 7 and 14 days following the initial TNF-ct treatment. Interestingly, these neuropeptide alterations were not further augmented upon reexposure to TNF-~, suggesting that the passage of time rather than sensitized cellular responses were responsible for these variations. Interestingly, while initial TNF-~ treatment itself did not markedly affect c-fos-immunoreactive (c-fos-ir) measured after 7 and 14 day intervals, reexposure to TNF-ct at these times provoked a sensitization of the immediate early gene. Moreover, double labeling revealed that the c-fos-ir cells also were immunoreactive for AVP within the neuroendocrine regulatory PVN and supraoptic nucleus, suggesting that the cytokine may have been having protracted actions on magnocellular AVP circuits [58]. Unlike the effects observed in the hypothalamus, within the central amygdala, increased staining of CRH immunoreactive fibers was only evident if mice were reexposed to TNF-c~ after the 1 and 7 day intervals following the initial cytokine treatment. In effect, the passage of time alone was a necessary, but not sufficient condition, to elicit the sensitization. Yet, it appeared that amygdaloid c-fos immunoreactivity was increased upon reexposure to TNF-~ 7 and 14 days following initial challenge with the cytokine. Thus, it is likely that neural responses within the central amygdala involving non-CRH processes are also subject to more protracted time-dependent sensitization.
Figure 7. Immunoreactivity for AVP and CRH (top and bottom panels, respectively) within the median eminence of mice receiving saline (A and D) or those treated with 4.0 ~tg of rhTNF- and sacrificed 7 (B or E) or 14 (C and F) days later. Pretreatment with a single dose of TNF-ct maximally increased stores of the neuropeptides in the external terminals after the 7-day interval (B and E). In contrast to the increased AVP within the external terminals, immunoreactivity for the peptide was reduced within the internal zone of the median eminence 14 days following the single cytokine exposure (C).
The neuroplasticity of the central CRH and AVP systems to both IL-1 and TNF-ct raises the possibility that these peptides may be responsible for the long term actions on hormonal and behavioral processes associated with cytokine or stressor treatments. Moreover, there is evidence to suggest that a cross-sensitization between immune (e.g., cytokine or infection) and non-immune (e.g., stressor) stimuli may exist with respect to behavioral and neuroendocrine
156
activity [57, 60]. It might be noted at this juncture that the effects of chronic stressors on CRH/AVP co-expression were comparable to those associated with the passage of time following an acute stressor. This should not be misconstrued to suggest, however, that the two have comparable outcomes with respect to other systems, nor that chronic unpredictable and intermittent events have effects like those associated with predictable stressors. Furthermore, while cytokine administration may influence neuroendocrine responses to later acute stressors, the effects elicited upon exposure to chronic stressors remain to be established. Likewise, data are unavailable concerning the impact of chronic stressors on the response to later cytokine or endotoxin challenge. The protracted CRH and AVP variations associated with acute and chronic challenges may be important for a wide array of pathological states, including depression, anxiety, neurodegenerative diseases and drug abuse, wherein neuropeptide involvement is suspected and some degree of neural memory may underlie the disorder [61 ]. Indeed, the possibility ought to be considered that individuals suffering from viral or bacterial infections may be at increased risk for subsequent stressor-related pathology. In this respect, it may be important to consider the frequency and timing of the stressors relative to the period of infection, as well as the number of infections previously encountered. In parallel with such neuroendocrine changes, repeated release of acute phase reactants (which might possibly include 13-amyloid) after multiple infections, may ultimately contribute to the progression of Alzheimer's disease [21].
8.
BEHAVIORAL EFFECTS OF CYTOKINE TREATMENTS
It will be recalled that proinflammatory cytokines, such as IL-I[5 and TNF-c~, induce a behavioral profile which reflects sickness. These behaviors have been thought to minimize energy expenditure, conserve body temperature, and generally allow the organism to maximize its ability to contend with viral and bacterial insults [10, 11]. In effect, such behaviors may reflect the manifestation of a highly organized motivational state critical to the survival of the organism [10]. In addition, it seems that cytokine treatments may engender anhedonic-like effects, wherein ordinarily reinforcing stimuli are not perceived as rewarding as they might otherwise be. Thus, when animals exhibit a decline in the consumption of palatable snacks, it may represent the anorexic effects associated with illness, but it may also be indicative of a general anhedonia. Unfortunately, it is exceedingly difficult to disentangle the anorexic and anhedonic effects of cytokine treatments. However, it has been shown that endotoxin administration may provoke a disruption of responding for rewarding brain stimulation (a behavior independent of the anorexic actions of cytokines), and this effect could not simply be attributed to illness [41 ]. Although psychogenic stressors and cytokines influence CNS functioning through different neural circuitry, the cytokines might be part of a regulatory loop that, by virtue of their effects on the CNS, contribute to the symptoms of behavioral pathologies, including mood and anxiety-related disorders [40, 44]. Indeed, depressive illness is associated with elevated levels of IL-I[5, IL-2, IL-6, TNF-c~, and soluble cytokine receptors [44]. Furthermore, the effects of endotoxin on consumption of palatable food substances could be attenuated by chronic antidepressant administration [62], possibly indicating that the treatment acts to temper the anhedonia, a fundamental characteristic of depressive illness. It might be noted at this juncture that humans u~ ~lergoing immunotherapy (e.g., with IL-2 or with interferon-~t) display numerous adverse neuropsychological, neurologic and psychiatric disturbances, including depression, which may be sufficiently severe to necessitate discontinuation of treatment [63, 64, 65].
157
Interestingly, it has been reported that the selective serotonin reuptake inhibitor, paroxetine, attenuated the depressive-like effects associated with interferon-or treatment of malignant melanoma [65]. Although TNF-ot has been used in cancer treatment, its extreme toxicity and shock inducing properties limit the systemic application of this cytokine. There have also been suggestions that elevated circulating TNF-ot may be associated with psychiatric illness [44] and such effect could come about by virtue of the cytokine's effects on central monoamine turnover [66]. Indeed, antidepressant medication has been reported to alter central levels of TNF-c~ [66].
9.
SUMMARY
Both IL-I[5 and TNF-c~ engender a profile of stressor-like effects. In addition to eliciting sickness-like behaviors, these cytokines may also promote depressive and anxiogenic-like responses, and elicit marked neuroendocrine and central monoamine changes. The latter effects were not restricted to the hypothalamus, but also occurred at mesolimbic sites. As in the case of stressors, cytokine challenge may engender a sensitization effect, wherein the response to subsequent cytokine and stressor challenges are augmented [57]. Despite the similarity between stressor and cytokine actions, analysis of the behavioral and neurochemical alterations associated with cytokines have been largely limited to the effects of acute treatments. Yet, viral and bacterial insults are sustained over protracted periods, and it is likely that the effects of traumatic events may engender at least some central cytokines alterations that are fairly persistent [57]. To assess adequately the impact of cytokine challenge on psychological processes it is essential to evaluate the impact of protracted treatments. In the case of processive stressors, chronic insults may promote a compensatory increase of amine synthesis, permitting the organism to deal with environmental challenges e.g., [5]. Yet, if the stressor is sufficiently protracted the wear and tear induced by attempts to adapt (allostatic load) may have adverse behavioral repercussions [6]. Although it is unclear what behavioral and central neurochemical alterations are introduced by sustained cytokine administration, analyses regarding immunebrain interactions and the ramifications for psychopathology need to consider the impact of continuous and sustained insults.
ACKNOWLEDGEMENTS Supported by the Medical Research Council of Canada. H.A. is a Senior Research Fellow of the Ontario Mental Health Foundation.
REFERENCES 1. 2.
3.
B lalock JE. The syntax of immune-endocrine communication. Immunol Today 1994; 15,504-544. Dunn AJ. Interactions between the nervous system and the immune system. In FE Bloom & DJ Kupfer DJ, Psychopharmacology: The Fourth Generation of Progress. Raven Press, New York 1995; 719-731. Merali Z, Lacosta S, Anisman H. Effects of interleukin-l[5 and mild stress on central monoamine alterations: A regional microdialysis study. Brain Research 1997; 761, 225-235.
158
.
10. 11.
12.
13. 14.
15.
16.
17. 18.
19. 20.
21. 22.
Sapolsky RM, Romero LM, Munck AU. How do glucocorticoids influence stress responses? Integrating permissive, suppressive, stimulatory, and preparative actions. Endocr Rev 2000; 21, 55-89. Anisman H, Zalcman S, Zacharko RM. The impact of stressors on immune and central neurotransmitter activity: Bidirectional communication. Rev Neurosci 1993; 4, 147-180. McEwen BS, Seeman T. Protective and damaging effects of mediators of stress. Elaborating and testing the concepts of allostasis and allostatic load. Ann NY Acad Sci 1999; 896, 30-47. Turnbull AV, Rivier C. Cytokine effects on neuroendocrine axes: Influence of nitric oxide and carbon monoxide. In NJ Rothwell (Ed.), Cytokines in the Nervous System. RG Landes, London, 1996; 93-116. Ericsson A, Kovacs KJ, Sawchenko PE. A functional anatomical analysis of central pathways subserving the effects of interleukin-1 on stress-related neuroendocrine neurons. J Neurosci 1994; 14, 89-91. Rivest S, Torres G, Rivier C. Differential effects of central and peripheral injection of interleukin-1 beta on brain c-fos expression and neuroendocrine functions. Brain Research 1992; 587, 13-23. Dantzer R, Bluthe RM, Laye S, Bret-Dibat JL, Parnet P and Kelley KW Cytokines and sickness behavior. Ann NY Acad Sci 1998; 840, 586-590. Maier SF, Watkins LR. Cytokines for psychologists: Implications of bidirectional immuneto-brain communication for understanding behavior, mood, and cognition. Psychol. Rev 1998; 105, 83-107. Banks WA, Ortz L, Plotkin SR, Kasten AJ. Human interleukin (IL) 1 alpha, murine IL-1 alpha and murine IL-2 beta are transported from blood to brain in the mouse by a shared saturable mechanism. J Pharmacol Exp Therap 1991; 259, 988-996. Gutierrez EG, Banks WA, Kastin AJ. Murine tumor necrosis factor alpha is transported from blood toi brain in the mouse. J Neuroimmunol 1993; 47, 169-176. Quan N, Sundar SK, Weiss JM. Induction of interleukin-1 in various brain regions after peripheral and central injections of lipopolysaccharide. J Neuroimmunol 1994; 49, 125-134. Zhang J, Rivest S. A functional analysis of EP4 receptor-expressing neurons in mediating the action of prostaglandin E2 within specific nuclei of the brain in response to circulating interleukin- 1 beta. J Neurochem 2000; 74, 2134-2135. Breder CD, Tsujimoto M, Terano Y, Scott DW. Saper CB. Distribution and characterization of tumor necrosis factor-a-like immunoreactivity in the murine central nervous system. J Comp Neurol 1993; 337,543-567. Schobitz B, De Kloet ER, Holsboer F. Gene expression and function of interleukin 1, interleukin 6, and tumor necrosis factor in the brain. Prog Neurobiol 1994; 44, 397-432. Pitossi F, del Rey A, Kabiersch A, Besedovsky H. Induction of cytokine transcripts in the CNS and pituitary following peripheral administration of endotoxin to mice. J Neurosci Res 1997; 48, 287-298. Rothwell NJ. Cytokines-killers in the brain. J Physiol 1999; 514, 3-17. Yabuuchi K, Maruta E, Minami M, Satoh M. Induction of interleukin-l[3 mRNA in the hypothalamus following subcutaneous injections of formalin into the rat hind paws. Neurosci Lett 1996; 207, 109-112. Patterson PH. Cytokines in Alzheimer's disease and multiple sclerosis. Curr Opin Neurobiol 1995; 5, 642-646. Chao CC, Hu S, Ehrlich L, Peterson PK. Interleukin-1 and tumor necrosis factor-alpha
159
23. 24.
25.
26
27.
28.
29.
30. 31.
32.
33. 34. 35. 36.
37.
38.
39.
synergistically mediate neurotoxicity: involvement of nitric oxide and of N-methyl-Daspartate receptors. Brain Behav Immun 1995; 9, 355-65. Minami M, Kuraishi Y, Yamaguchi T, Nakai S, Hirai Y, Satoh M. Immobilization stress induces interleukin-lb mRNA in rat hypothalamus. Neurosci Lett 1991; 123,254-256. Nguyen KT, Deak T, Owens SM, Kohno T, Fleshner M, Watkins LR, Maier SF. Exposure to acute stress induces brain interleukin-lb protein in the rat. J Neurosci 1988; 18, 2239-2246. Nguyen KT, Deak T, Will MJ, Hansen MK, Hunsaker BN, Fleshner M, Watkins LR, Maier SF. Time course and corticosterone sensitivity of the brain, pituitary, and serum interleukin-lbeta protein response to acute stress. Brain Res 2000; 859, 193-201. Takao T, Tojo C, Nishioka T, Kurokawa H, Takemura T, Hashimoto K. De Souza EB. Reciprocal modulation of corticotropin-releasing factor and interleukin-1 receptors following ether-laparotomy stress in the mouse. Brain Research 1994; 660, 170-174. Deak T, Meriwether JL, Fleshner M, Spencer RL, Abouhamze A, Moldawer LL, Grahn RE, Watkins LR, Maier SF. Evidence that brief stress may induce the acute phase response in rats. Am J Phsyiol. 1997; 273, R1998-2004. Shintani F, Kanba S, Nakaki T, Nibuya M, Kinoshita N, Suzuki E, Yagi G, Kato R, Asai M. Interleukin-113 augments release of norepinephrine, dopamine and serotonin in the rat anterior hypothalamus. J Neurosci 1993; 13, 3574-3581. Hayley S, Brebner K, Lacosta S, Merali Z, Anisman H. Sensitization to the effects of tumor necrosis factor-c~: Neuroendocrine, central monoamine and behavioral variations. J Neurosci 1999; 19, 5654-5665. Herman JP, Cullinan WE. Neurocircuitry of stress: Central control of hypothalamopituitary-adrenocortical axis. Trends Neurosci 1997; 20, 78-84. Xu Y, Day TA, Buller KM. The central amygdala modulates hypothalamic-pituitaryadrenal axis responses to systemic interleukin-1 beta administration. Neuroscience 1999; 94, 175-183. Lacosta S, Merali Z, Anisman H. Influence of interleukin-1 on exploratory behaviors, plasma ACTH and cortisol, and central biogenic amines in mice. Psychopharmacology 1998; 137, 351-361. Brady LS, Lynn AB, Herkenham M, Gottesfeld Z. Systemic interleukin-1 induces early and late patterns of c-fos mRNA expression in brain. J Neurosci 1994; 14, 4951-4964. Rivest S, Laflamme, N. Neuronal activity and neuropeptide gene transcription in the brains of immune-challenged rats. J Neuroendocrinology 1995; 7,501-525. Lacosta S, Merali Z, & Anisman H. Behavioral and neurochemical consequences of lipopolysaccharide in mice: anxiogenic-like effects. Brain Research 1999; 818, 291-303. Smagin GN, Swiergiel AH, Dunn AJ. Peripheral administration of interleukin-1 increases extracellular concentrations of norepinephrine in rat hypothalamus: Comparison with plasma corticosterone. Psychoneuroendocrinology 1996; 21, 83-93. Dunn AJ. Effects of the IL-1 receptor antagonist on the IL-1- and endotoxin- induced activation of the HPA axis and cerebral biogenic amines in mice. Neuroimmunomodulation. 2000; 7, 36-45. Linthorst ACE, Flachskamm C, Muller-Preuss P, Holsboer F, Reul JMHM. Effect of bacterial endotoxin and interleukin-l[5 on hippocampal serotonergic neurotransmission, behavioral activity, and free corticosterone levels: An in vivo microdialysis study. J Neurosci 1995; 15, 2920-2934. Song C, Merali Z, Anisman H. Variations of nucleus accumbens dopamine and serotonin following systemic interleukin-1, interleukin-2 or interleukin-6 treatment. Neuroscience
160
40. 41. 42.
43. 44. 45.
46.
47.
48.
49.
50.
51.
52.
53.
54. 55. 56.
1999; 88, 823-836. Anisman H, Merali Z. Anhedonic and anxiogenic effects of cytokine exposure. Adv. Exp Med. Biol 1999; 461,199-233. Borowski T, Kokkinidis L, Merali Z, Anisman H. Lipopolysaccharide, central in vivo biogenic amine variations, and anhedonia. Neuro. Report. 1998; 9, 3797-3802. McIntyre DC, Anisman H. Anxiety, impulse control in rats selectively bred for seizure sensitivity. In Mysblodsky M, Weiner, I (eds.), Contemporary Issues in Modeling Psychopathology. Kluwer Academic, New York 2000; (in press). Beutler B A. The role of tumor necrosis factor in health and disease. J Rheumatol. 26 suppl. 1999; 57: 16-21. Maes M. Major depression and activation of the inflammatory response system. Adv. Exp Med. Biol 1999; 461, 25-46. Ando T, Dunn AJ. Mouse tumor necrosis factor-alpha increases brain tryptophan concentrations and norepinephrine metabolism while activating the HPA axis in mice. Neuroimmunomodulation. 1999; 6, 319-329. Brebner K, Hayley S, Merali Z, Anisman H. Synergistic effects of interlwukin-lb, interleukin-6 and tumor necrosis facotr-a: Central monoamine, corticosterone and behavioral variation. Neuropsychopharmacology. 2000; 22, 566-580. Pauli S, Linthorst ACE, Reul JMHM. Tumor necrosis factor-a and interleukin-2 differentially affect hippocampal serotonergic neurotransmission, behavioral activity, body temperature and hypothalamic-pituitary-adrenocortical axis activity in the rat. Eur J Neurosci 1998; 10, 868-878. Zhou D, Shanks N, Riechman SE, Liang R, Kusnecov AW, Rabin BS. Interleukin 6 modulates interleukin-1 and stress-induced activation of the hypothalamic- pituitaryadrenal axis in male rats. Neuroendocrinology 1996; 63, 227-236. Perlstein RS, Moughey EH, Jackson WE, Neta R. Interleukin-1 and interleukin-6 act synergistically to stimulate the release of adrenocorticotropic hormone in vivo. Lymphokine Cytokine Res 1991 ; 10, 141 - 146. Mino T, Sugiyama E, Taki H, Kuroda A, Yamashita N, Maruyama M, Kobayashi M. Interleukin-lalpha and tumor necrosis factor alpha synergistically stimulate prostaglandin E2-dependent production of interleukin-ll in rheumatoid synovial fibroblast. Arthritis Rheum 1998; 41, 20004-2013. Plata-Salaman CR, Sonti G, Borkoski JP, Wilson CD, Ffrench-Mullen, JMH. Anorexia induced by chronic central administration of cytokines at estimated pathophysiological concentrations. Physiol Behav 1996; 60, 867-875. Jeohn GH, Kong LY, Wilson B, Hudson P, Hong JS. Synergistic neurotoxic effects of combined treatments with cytokines in murine primary mixed neuron/glia cultures. J Neuroimmunol 1998; 85, 1-10. Gayle D, Ilyin SE, Miele ME, Plata-Salman CR. Modulation of TNF-alpha mRNA production in rat C6 glioma cells by TNF-alpha, IL-1 beta, IL-6 and IFN-alpha: in vitro analysis of cytokine-cytokine interactions. Brain Res Bull 1998; 47, 231-235. Kalivas PW, Stewart J. Dopamine transmission in the initiation and expression of drug-and stress-induced sensitization of motor activity. Brain Res Rev 1991; 16, 223-244. Antelman SM, Soares JC, Gershon S. Time-dependent sensitization-possible implications for clinical psychopharmacology. Behav Pharmacol 1997; 8,505-514. Schmidt ED, Janszen AWJW, Wouterlood FG, Tilders FJH. Interleukin-1 induced longlasting changes in hypothalamic corticotropin-releasing hormone (CRH) neurons and hyperresponsiveness of the hypothalamic-pituitary-adrenal axis. J Neurosci 1995; 15, 7417-7426.
161
57. 58.
59. 60.
61. 62. 63.
64.
65.
66.
Tilders FJH, Schmidt ED. Interleukin-l-induced plasticity of hypothalamic CRH neurons and long-term stress hyperresponsiveness. Ann NY Acad Sci 1998; 840, 65-73. Hayley S, Merali Z, Staines WA, Anisman H. Immunohistochemical evidence for a time-dependent sensitization of neuroendocrine regulatory peptides in response to tumor pecrosis factor-alpha. Brain Behav Immun Abst 2000a; 14, 100. Beutler B, van Huffel C. Unraveling function in the TNF ligand and receptor families. Science 1994; 264(5159): 667-8. Hayley S, Dodd C, Merali Z, Anisman H. Lipopolysaccharide pretreatment influences behavioural and neuroendocrine responses to subsequent cytokine endotoxin or stressor exposure: cross-sensitization effect. Soc Neurosci Abst 2000b; in press. Griffiths J, Ravindran AV, & Merali Z, Anisman H. Dysthymia: Neurochemical and behavioral perspectives. Molecul Psychiat 2000; 26, 1179. Yirmiya R. Endotoxin produces a depressive-like episode in rats. Brain Research 1996; 711,163-167. Capuron L, Ravaud A, Radat F, Dantzer R, Goodall G. Affects of interleukin-2 and alpha-interferon cytokine immunotherapy on the mood and cognitive performance of cancer patients. Neuroimmunomodulation 1998; 5, 9. Denicoff KD, Rubinow DR, Papa MZ, Simpson L, Seipp LA, Lotze MT, Chang AE, Rosenstein D, Rosenberg SA. The neuropsychiatric effects of treatment with interleukin-2 and lymphokine-activated killer cells. Ann Inter Med 1987; 107, 293-300. Miller A, Musselman D, Penna S, Su C, Pearce B, Nemeroff C. Pretreatment with the antidepressant paroxetine, prevents cytokine-induced depression during IFN-alpha therapy for malignant melanoma. Neuroimmunomod Abst 1999; 6, 237. Ignatowski TA, Spengler RN. Tumor necrosis factor-alpha: presynaptic sensitivity is modified after antidepressant drug administration. Brain Res 1994; 665,293-299.
This Page Intentionally Left Blank
New Foundation of Biology Edited by I. Berczi and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
163
Proinflammatory Signal Transduction Pathways in the CNS During Systemic Immune Response
SERGE RIVEST, SYLVAIN NADEAU, STEVE LACROIX and NATHALIE LAFLAMME
Laboratory of Molecular Endocrinology, CHUL Research Center and Department of Anatomy and Physiology, Laval University, 2705, boul. Laurier, Qudbec, Canada, G1V 4G2
ABSTRACT Circulating lipopolysaccharide (LPS) causes a rapid transcriptional activation of its transmembrane receptor mCD14 within the circumventricular organs (CVOs), brain regions that contain a rich vascular plexus with specialized arrangements of the blood vessels. Parenchymal cells located in the anatomical boundaries of the CVOs exhibit a delayed response, which is followed by a positive signal for CD 14 transcript in microglia across the brain parenchyma. The constitutive expression of the toll-like receptor 4 (TLR4) in the CVOs is likely to be a key element allowing the proinflammatory signal transduction pathways (MyD88/IRAK/NIK/NF-KB) to take place rapidly in these organs in response to circulating LPS. These results strongly suggest that the endotoxin first reaches organs devoid of the blood brain barrier (BBB) to induce the transcription of its own receptor and thereafter increases CD14 biosynthesis within parenchymal structures surrounding the CVOs and then the entire brain of severely challenged animals. Brain CD 14 expression may be a key step in the transcription of proinflammatory cytokines primarily within accessible structures from the blood and subsequently through scattered parenchymal cells during severe sepsis. However, CD14 synthesis in parenchymal cells of the brain is also dependent on the production of proinflammatory cytokines. Of interest is the data that systemic injection of the bacterial endotoxin induces a strong expression of CD14 mRNA in a pattern that is closely related to the induction of tumor necrosis factor alpha (TNF-ct) transcript with a rapid and delayed response. Although there is a large body of evidence that CD 14 (and now TLR4) is necessary for the role of LPS on the induction of cytokine transcription from different myeloid cells, the possibility remains that the cytokine itself acts as an autocrine and paracrine factor to up regulate the LPS receptor. The binding of TNF to its type I receptor (p55) leads to the activation and translocation of p50/65 NF-~:B into the nucleus, which seems a key player in activating CD 14 transcription in the CNS. Central injection of recombinant rat TNF-ct causes a robust expression of the genes encoding IKBct, TNF-~t and CD14 in microglial cells of the brain parenchyma. The time-related induction of these transcripts suggested a potential role of NF-KB in mediating TNF-induced transcriptional activation of the LPS receptor. Systemic injection with the endotoxin LPS provoked a similar microglial activation that was
164
prevented in inhibiting the biological activity of the proinflammatory cytokine in the CNS. Together these data provide the evidence that centrally-produced TNF-c~ plays an essential autocrine/paracrine role in triggering parenchymal microglial cells during severe endotoxemia. These events may be determinant for orchestrating the neuroinflammatory responses that take place in a well coordinated manner to activate the resident phagocytic population of cells in the brain. The physiological outcomes of this innate immune response of the CNS are likely to include a rapid elimination of LPS particles via an increased opsonic activity of the transmembrane CD 14 receptor to prevent potential detrimental consequences on neuronal elements during blood sepsis.
1.
INTRODUCTION
The clinical manifestations of endotoxemia are characterized by hyperventilation, hypercoagulation, pain, fever, cachexia, tachycardia, hypotension, somnolence, change in oxygen consump, tion and multiple organ failure [1 ]. Most of these symptoms can be mimicked by the stimulation of host monocytes/macrophages in the presence of the endotoxin lipopolysaccharide (LPS) [2] or prevented with anti-CD14 antibody [3, 4, 5]. The endotoxin is released by the outer membrane of the Gram-negative bacteria during sepsis and is detected by cells of myeloid origin, which bear the LPS receptor CD 14 at their membrane surfaces [6]. CD 14 is considered as the key player in the induction of the septic shock provoked by Gram-negative bacteria. Two forms of the CD14 receptor can be found; the first one is present at the surface of myeloid cells (mCD14) and acts as a glycosylphosphatidylinositol (GPI)-anchored membrane glycoprotein. The other form is soluble in the serum (sCD14) and lacks the GPI properties, although it can bind LPS to activate cells devoid of mCD14, such as endothelial and epithelial cells, astrocytes and vascular smooth muscle cells [7]. The spontaneous binding of the endotoxin to its transmembrane and soluble receptor occurs slowly, but with high affinity [8]. However, the binding rate is accelerated by the presence of LPS-binding protein (LBP), a serum protein that binds the endotoxin [9, 10]. Although LBP is not essential for the LPS signaling, the LPS/LPB complex is particularly powerful to activate cells of myeloid origin including monocytes, macrophages, neutrophils, and microglia. One of the well known consequences of such activation is the production of proinflammatory molecules, namely interleukin-1 (IL-1), IL-6, tumor necrosis factor (TNF) and different prostaglandins (PGs). It is not well known how cell activation is triggered after binding between the LBP-LPS complex and the GPI-anchored mCD14, although there is now evidence that activation of tyrosine kinase leads to signal transduction and cytokine gene transcription through nuclear factor kappa B (NF-KB). The recent characterization of human homologues of Toll, especially the Toll-like receptor 4, may be the missing link for the transduction events leading to NF-KB activity in response to the LPS/mCD14 interaction (see below). NF-~zB is normally present in the cytoplasm forming an inactive complex with an inhibitor known as I~;Bc~ (see Figure 1). Following extracellular stimulation by growth factors, mitogens and cytokines that activate mitogen-activated protein (MAP) kinases, I~:Bc~ is phosphorylated by NF-KB-inducible kinase (NIK)/IKBc~ kinases (IKK), ubiquitinated and degraded by cytoplasmic proteasomes [11, 12]. Free active NF-KB (the commonest complex is the p50/p65 heterodimer) is then translocated into the nucleus where it is able to regulate transcription of various genes by binding to an ~;B consensus sequence. Following its degradation, I~:Bc~ is rapidly re-synthesized to act as an endogenous inhibitory signal for NF-~zB, and monitoring its d e n o v o expression is
165
Figure 1. The proinflammatory signal transduction pathways evolving the nuclear factor kappa B (NF-•B). p50 and p65 are the two DNA-binding subunits of the NF-KB dimer that is a potent transcription factor for numerous proinflammatory genes. See text for details and abbreviations.
a powerful tool to investigate the activity of the transcription factor within the CNS (for reviews, see [ 13, 14]).
2.
DO THESE EVENTS TAKE PLACE IN THE BRAIN?
We have recently reported that circulating LPS causes a rapid expression of CD14 mRNA within the circumventricular organs (CVOs), brain regions that contain a rich vascular plexus with specialized arrangements of the blood vessels [15]. The tight junctions normally present between the endothelial cells are shifted in part to the ventricular surface and partly to the boundary between the CVOs and adjacent structures, explaining the diffusion of large molecules into the perivascular region [16]. Parenchymal cells located in the anatomical boundaries of the CVOs exhibited a delayed response, followed by a positive signal for CD14 transcripts in microglia throughout the brain parenchyma [15]. These results strongly suggest that endotoxin first reaches organs devoid of the blood brain barrier (BBB) to induce the transcription of its own receptor and thereafter increased CD 14 biosynthesis occurs within parenchymal structures surrounding the CVOs and subsequently the entire brain of severely challenged animals. Brain CD14 expression is likely to be a key step in the transcription of proinflammatory cytokines [17, 18, 19] first within accessible structures from the blood and thereafter through scattered parenchymal cells during severe sepsis.
166
Like systemic phagocytes [20], CD14 synthesis in parenchymal cells of the brain may be dependent on the production of proinflammatory cytokines. Of interest is the data that systemic injection of the bacterial endotoxin induced strong expression of CD14 mRNA [15] in a pattern that was closely related to the induction of TNF-c~ transcript [17] with a rapid and delayed response. Although there is a large body of evidence that CD14 is necessary for an effect of LPS on the induction of cytokine transcription from different myeloid cells, the possibility remains that the cytokine itself acts as an autocrine and paracrine factor to upregulate the LPS receptor. The binding of TNF to its type I receptor (p55) leads to the activation and translocation of p50/65 NF-KB into the nucleus [21], an event that has been reported to modulate CD14 expression. Indeed, TNF-c~ is able to induce a transient increase in plasma CD14 levels with a peak at 6-8 hour and this elevation in levels of CD14 antigen was shown to be accompanied by increased levels of CD14 mRNA in lung, liver and kidney [22]. Pretreatment of mice with an anti-TNF-c~ antibody significantly prevented LPS-induced mCD14 transcription [22, 23]. We have recently investigated the role of microglial-derived TNF in the regulation of CD14 in the brain during endotoxemia [24]. Central injection of recombinant rat TNF-c~ caused a robust expression of the genes encoding IKB~t, TNF-c~ and CD14 in microglial cells of the brain parenchyma. The time-related induction of these transcripts suggested a potential role of NF-KB in mediating TNF-induced transcriptional activation of the LPS receptor. Systemic injection with the endotoxin LPS provoked a similar microglial activation that was prevented by inhibiting the biological activity of the proinflammatory cytokine in the CNS. Together these data provide the evidence that centrally-produced TNF-c~ plays an essential autocrine/paracrine role in triggering parenchymal microglial cells during severe endotoxemia. These events may be determinant for orchestrating the neuroinflammatory responses that take place in a well coordinated manner to activate the resident phagocytic population of cells in the brain. The physiological outcomes of this innate immune response of the CNS are likely to include a rapid elimination of LPS particles via an increased opsonic activity of the transmembrane CD14 receptor to prevent potential detrimental consequences to neuronal elements during blood sepsis.
3.
THE AUTOCRINE/PARACRINE ACTION OF TNF-c~ ACROSS THE CNS
As mentioned, systemic LPS injection induced mCD14 expression first within structures devoid of BBB and thereafter throughout the brain parenchyma of animals that received high doses of LPS [ 15]. Microscopic analysis of emulsion-dipped slides revealed that CD 14-positive cells spread over the anatomical boundaries of the CVOs in a migratory-like pattern during the course of endotoxemia. The direct action of LPS on myeloid-derived cells expressing mCD14 and present in structures that are accessible from the systemic circulation may allow a rapid production of proinflammatory cytokines within these organs. The rapid induction of TNF-~ mRNA in the CVOs by i.p. LPS clearly indicates that such events take place in specific populations of cells in the brain. Like CD14, small scattered TNF-expressing cells can be found across the brain parenchyma in response to LPS, although this depends on the dose of the endotoxin and the route of administration [17]. Indeed, the CVOs displayed low but positive signal as soon as 1 hour after the LPS challenge and increased to reach maximal levels at 6 hour, but positive cells gradually became apparent in the boundary of these organs and spread over the entire brain from time 6 to 24 hour during severe endotoxemia. However, this phenomenon is only provoked by a high dose of bacterial LPS (2.5 mg/kg i.p.); a lower
167
dose (250/,tg/kg i.p.) caused a more restricted transcriptional activation of the gene encoding the proinflammatory cytokine within the CVOs and the adjacent areas [17]. The same dose injected i.v. is capable of inducing CD14 and TNF across the brain parenchyma, whereas these transcripts are localized to the CVOs and choroid plexus (chp) in response to 5 jug LPS i.v. (N. Laflamme and S. Rivest, unpublished data). This clearly indicates that the endotoxin first reaches available targets devoid of BBB, which in return, depending on the severity of the challenge, primes adjacent cells within the parenchymal brain to stimulate TNF-ot transcription. The bacterial endotoxins are among the most powerful agents known to stimulate circulating monocytes and tissue macrophages, which leads to the synthesis and release of a variety of proinflammatory cytokines [7]. The most important target of macrophage-derived secretory products is the macrophage itself [1]. The early production of TNF may be an essential step in this autocrine and paracrine loops, as this cytokine is able to induce its own production by an autocrine stimulation that is followed by the synthesis of other proinflammatory cytokines, such as IL-113 and IL-6 [7]. TNF has also been found to be a major factor inducing shock, and passive immunization against the cytokine can attenuate the appearance of IL-I[3 and IL-6 [25]. Surprisingly however, cytokine production and circulating IL-6 and IL-I[3 induced by i.p. LPS injection is intact in TNF-deficient mice [26], which suggests that TNF production is certainly not the sole primary event that leads to production of subsequent cytokines in response to endotoxin and other models of inflammation. In the brain, the cytokine seems to act as a key ligand to activate parenchymal microglia in a paracrine manner during endotoxemia. It is suggested here that circulating LPS targets its transmembrane receptor in CVOs/chp resident macrophages and microglia, which may stimulate the NF-KB signaling events and trigger TNF transcription (see Figure 2). The cytokine may in return bind to its cognate p55 receptor and lead to the formation of the TNF-Rl-associated death domain (TRADD)/TRAF2 complex, which activates the NF-KB signaling events in adjacent microglia. TNF-~ is actually one of the most potent effectors of NF-KB activity through the 55 kd TNF type I receptor [11, 12]. Such events are likely to contribute to the transcriptional activation of both CD14 and TNF genes in the brain of endotoxin-treated animals. Central production of TNF-c~ is a key mechanism controlling CD 14 expression in the brain parenchyma, but not in the CVOs and chp. The anti-TNF did not significantly change the relative CD14 mRNA levels in these organs, but prevented quite specifically the parenchymal expression of the LPS receptor during endotoxemia. The constitutive expression of CD14 in regions that can be reached by the bloodstream may allow a rapid production of TNF that in return acts as the endogenous ligand to activate adjacent myeloid cells. An important question however is what mechanisms control the spreading of the message from regions close to the CVOs to deep parenchymal elements. As mentioned, the spreading of CD14 and TNF-c~-expressing cells depends on the severity of the endotoxin challenge [15, 17]. It is therefore possible that the paracrine influence of the cytokine remains localized in response to low circulating levels of LPS, but takes place across the cerebral tissue during severe endotoxemia. Such effects of TNF in activating CD14 expression have previously been reported in different systemic organs and like the CNS, pretreatment of mice with anti-TNF-~ antibody significantly prevented LPS-induced mCD14 transcription in the lung, liver and kidney [22, 23]. IL-113 has also been reported to stimulate CD 14 expression in different organs and anti-IL-113 antibody attenuated the induction of the LPS receptor in response to the endotoxin [22, 23]. IL-I[3 and TNF-c~ are known to have numerous overlapping activities and inhibiting one cytokine may frequently be associated with redundant mechanisms because of the presence of
168
Figure 2. The lipopolysaccharide (LPS)/LPS-binding protein (LBP) complex has the ability to trigger accessible organs through its membrane CDI4 (mCD14) receptor. The constitutive expression of mCD14 together with Toll-like receptor 4 (TLR4) may allow the signaling events to take place in the circumventricular organs (CVOs) and activate transcription of proinflammatory cytokines. The subsequent release of tumor necrosis factor alpha (TNF-ct) acts as an autocrine and paracrine factor for the synthesis of the LPS receptor CD14 in the brain microglial cells during blood endotoxemia. See text for details.
the other cytokine [27]. Although IL-l[3 may have the ability to stimulate CD14 in the brain microglia, its involvement depends most likely on the prior production of TNF as the anti-TNF completely inhibited LPS-induced CD14 transcription. On the other hand, IL-1[3 is the key inflammatory signal in the brain to stimulate the production of growth factors by astrocytes during brain trauma, while TNF is not essential for such response. Indeed, we have recently observed a strong and rapid production of numerous proinflammatory molecules in cells lining the lesion site that was followed by a robust increase in ciliary neurotrophic factor (CNTF) biosynthesis [28]. The release of CNTF was completely inhibited in IL-1 [3-deficient mice while TNF-ct was still produced by microglial cells lining the corticectomy [28]. Circulating IL-I[3 has also been recently shown to be a key mediator of the NF-KB activity and COX-2 transcription in cells of the BBB during a systemic model of inflammation [29]. These data, together with the present study, support the concept that despite the recognized overlapping activities of both NF-KB-signaling cytokines in the systemic immune system, IL-I[3 and TNF-ct seem to have a distinct role in orchestrating the inflammatory events that take place in the brain. The possibility that circulating TNF may influence CNS CD14 expression was also investigated, as the cytokine is rapidly detected into the bloodstream during endotoxemia [30]. Intravenous injection of recombinant rat TNF-ct caused up regulation mCD14 transcript
169
quite selectively in the CVOs and not within parenchymal cells adjacent to these organs. Therefore, circulating TNF may not contribute to the robust transcriptional activation of the LPS receptor in parenchymal microglia of endotoxin-challenged rats. The positive autoregulatory loop is therefore not always associated with a paracrine influence of the cytokine to trigger CD14 in surrounding cerebral tissues. This unexpected result provides the evidence that microglial-derived TNF-c~ only is responsible for activating the biosynthesis of the LPS receptor in deep parenchymal cells of the brain in response to high circulating levels of the bacterial endotoxin.
4.
TOLL-LIKE RECEPTORS IN THE CNS
Host organisms detect the presence of infection by recognizing specific elements produced by micro-organisms, such as Gram-negative bacteria, Gram-positive bacteria and mannans of fungi [31]. These elements are called the pathogen-associated molecular patterns (PAMPs) that are recognized by specific cells of the immune system as innate mechanisms to mount a rapid response to bacterial infection. The endotoxin LPS is a major component of the outer membranes of Gram-negative bacteria, which is the best characterized example of innate recognition associated with a robust inflammatory response by phagocytic cells [32]. As mentioned, secretion of cytokines by circulating monocytes/neutrophils and tissue macrophages by LPS requires a series of mechanisms in cascade, the first step being the binding of the endotoxin with the serum proteins LPS-binding protein (LBP) or septins. The newly formed complex may then activate different populations of cells in binding to its CD 14 receptor [5]. Until recently, the exact mechanisms involved in the activation of the proinflammatory signal transduction pathways after binding between the LBP-LPS complex and the GPI-anchored mCD14 was unknown. Indeed, studies in CD14-deficient mice suggested the existence of a coreceptor to mediate LPS-induced NF-•B activity and cytokine gene transcription [2, 33]. The recent characterization of human homologues of Toll may be the missing link for the transduction events leading to NF-~:B activity and cytokine production in response to bacterial cell wall components. A large family of Toll-like receptors (TLRs) has already been characterized, which share similar extracellular and cytoplasmic domains [31]. The extracellular domains include 18-31 leucine-rich repeats (LRRs), whereas the cytoplasmic domains are similar to the cytoplasmic portion of the IL-1 receptor and is named the Toll/IL-l-receptor homologous region [31, 34]. Distinct TLRs have now been proposed as the key molecules to recognize quite selectively one of the major PAMPs produced by either Gram-negative or Gram-positive bacteria. The data that mutation of the mouse Lps locus abolishes the LPS response and that Lps encodes the TLR4 provided the first evidence that this particular receptor may play a key role in the innate immune response to Gram-negative bacteria (for review, see [35]). Further support for this concept comes from the TLR4-deficient mice that are unresponsive to LPS, whereas TLR2-deficient mice exhibit a normal inflammatory response to the endotoxin [36]. These results demonstrate that while TLR2 makes no contribution to LPS signaling, TLR4 is critical to recognize the PAMP produced by Gram-negative bacterial cell wall components. It is not yet known how LBP, CD14 and TLR4 interact together to function as the LPS signal transducer leading to activation of NF-KB and MAP kinases. It is possible that CD 14 acts as the principal LPS binding protein on the surface of monocytic cells and the newly formed complex reaches adjacent TLR4 receptors, which transduce the LPS signal via the general adaptor protein MyD88 [35, 37]. These events may also take place in the brain because
170
mRNAs encoding mCD14 and TLR4 are present in structures that can be reached by the bloodstream, namely the CVOs, chp and leptomeninges [38]. In contrast to the profound transcriptional activation of the LPS receptor CD 14 and the indicator of NF-KB activity I~:Bc~, the endotoxin and circulating IL-l[3 caused a significant decrease of TLR4 transcript in most of the constitutively-expressing parenchymal and non-parenchymal regions of the brain [38]. The basal expression of CD14 and TLR4 in the CVOs is likely to be a key mechanism in the proinflammatory signal transduction events that originate from these structures during innate immune response. Indeed, cell wall components of the Gram-negative bacteria may be selectively recognized by the TLR4/CD14-bearing cells of the CVOs, which allows the LPS signaling and then the rapid transcription of proinflammatory cytokines; the subsequent microglial activation in the brain parenchyma is, however, dependent on TNF-c~ (Figure 2). Therefore, TLR4 may be essential in this innate immune reaction that originates from the CVOs in response to cell wall components of Gram-negative bacteria. Although a strong increase in CD14 transcription is generally detected after systemic LPS injection, the endotoxin failed to stimulate the gene encoding TLR4 [38]. CD14-expressing cells were clearly devoid of TLR4 transcript in microglia across the brain parenchyma during moderate and severe endotoxemia. It is possible that TLR4 is the recognizing molecule for Gram-negative bacterial components only in response to systemic infection, whereas CD14 has a more complex role in the proinflammatory signal transduction events in the brain parenchyma. Nomura and colleagues have recently reported that TLR4 mRNA expression in mouse peritoneal macrophages significantly decreased within a few hours of LPS treatment and returned to the original level at 24 hours [39]. A rapid decrease of surface TLR4 expression was seen as early as 1 hour and remained suppressed over 24 hours in cells pre-exposed with LPS. These authors suggested that the down-regulation of the surface TLR4 expression may be responsible for the decrease in inflammatory cytokine production in tolerant macrophages, which may explain one of the mechanisms for LPS tolerance [39]. These data obtained from systemic macrophages are in complete agreement with our study that shows convincing down regulation of TLR4 gene in response to a single LPS bolus [38]. The phenotype of TLR4 cells in the CVOs was not determined due to the rather low levels of TLR4 transcript, making interpretation of the agglomeration of silver grains within immunoreactive cells arbitrary. Because LPS has the ability to increase CD14 mRNA in these organs, it was possible to perform the dual-labeling for the LPS receptor and numerous resident macrophages/microglia were positive for the transcript [ 15]. Although both transcripts may not be expressed within the same cells, we speculate here that TLR4 is located at the surface of the phagocytic population of cells of the CVOs, chp as well as the leptomeninges. TLR4 transcript levels are quite low in the cerebral tissue under basal conditions [38]. The signal was nevertheless specific, as we did perform numerous controls to ensure that what was being seen may not be related to an artifact of the in situ hybridization procedure. Actually, we had to adjust and maximize the hybridization conditions to detect this transcript in situ by generating the riboprobe just after the pre-hybridization step on freshly mounted brain sections. This very low level in the brain fits however quite well with the fact that the copy number of TLR4 is extremely low in systemic phagocytes compared to the more abundant membrane protein CD14 [35]. It is nevertheless remarkable that so few TLR4 receptors (perhaps 1000 or fewer per cell), residing on macrophages alone, have such an important influence in the LPS signaling and the coordination of the biological responses to Gram-negative infections [35]. It is expected that CVO TLR4 acts as a sensor for engaging the cerebral innate immune response in case of invasion during such systemic bacterial infection that may have detrimental consequences for the neuronal material.
171
5.
CONCLUDING REMARK
Our data strongly suggest that endotoxin first reaches organs devoid of the BBB to induce the transcription of its own receptor and thereafter increased CD 14 biosynthesis within parenchymal structures surrounding the CVOs and then the entire brain of severely challenged animals. TLR4 is likely to play a key role in LPS signaling and the innate immune response that is triggered in a very well organized manner from specific structures of the brain during endotoxemia. The constitutive expression of CD 14 in the CVOs and its up regulation in the brain parenchyma suggest a potential role in protecting the neural elements against LPS particles. The resident macrophages and microglia in the CVOs are strategically well positioned to respond rapidly to circulating endotoxin or bacteria, while parenchymal microglia are the phagocytic population of cells in the brain in case of invasion. There is alteration of the BBB during severe endotoxemia [40], which may allow diffusion of molecules that normally have no access to the parenchymal elements and be detrimental for neurons. Activation of the microglial cells across the CNS may rapidly eliminate this foreign material, although a sustained activity of these cells is not suitable as it may have opposite effects and be associated with neurodegenerative disorders [41 ]. A better understanding of this innate immune response in the cerebral tissue may lead us to the fundamental mechanisms underlying how the brain is capable of mounting an inflammatory response that either protects or contributes to damage neurons.
ACKNOWLEDGMENTS Our work on this subject is currently supported by Canadian Institutes of Health Research (CIHR) formely the the Medical Research Council (MRC) of Canada. Serge Rivest is a Canadian MRC Scientist. Sylvain Nadeau is supported by a Ph.D. studentship from the Canadian MRC, whereas Steve Lacroix was supported by a Ph.D. studentship from the Natural Sciences and Engineering Research Council of Canada. Dr. Lacroix is currently an MRC postdoc fellow at UCSD in La Jolla, California.
REFERENCES 1. 2.
3.
4.
5. 6.
Sweet MJ and Hume DA. Endotoxin signal transduction in macrophages. J Leukoc Biol 1996; 60: 8-26. Haziot A, Ferrero E, Kontgen F, Hijiya N, Yamamoto S, Silver J, Stewart CL and Goyert SM. Resistance to endotoxin shock and reduced dissemination of gram- negative bacteria in CD14-deficient mice. Immunity 1996; 4: 407-414. Dentener MA, Bazil V, Von Asmuth EJ, Ceska M and Buurman WA. Involvement of CD 14 in lipopolysaccharide-induced tumor necrosis factor-alpha, IL-6 and IL-8 release by human monocytes and alveolar macrophages. J Immunol 1993; 150:2885-2891 Wright SD, Ramos R, Hermanowski-Vosatka A, Rockwell P and Detmers PA. Activation of the adhesive capacity of CR3 on neutrophils by endotoxin: dependence on lipopolysaccharide binding protein and CD14. J Exp Med 1991; 173: 1281-1286. Wright SD, Ramos RA, Tobias PS, Ulevitch RJ and Mathison JC. CD14, a receptor for complexes of lipolysaccharide (LPS) and LPS binding protein. Science 1990; 249: 1431-1433. Pugin J, Heumann ID, Tomasz A, Kravchenko VV, Akamatsu Y, Nishijima M, Glauser MP,
172
.
o
.
10.
11. 12. 13. 14.
15.
16. 17.
18.
19.
20.
21.
22. 23. 24.
Tobias PS and Ulevitch RJ. CD14 is a pattern recognition receptor. Immunity 1994; 1: 509-516. Andersson J, Nagy S, Bj6rk L, Abrams J, Holm S and Andersson U. Bacterial toxininduced cytokine production studied at the single-cell level. Immunological Rev 1992; 127: 69-96. Tobias PS, Soldau K and Ulevitch RJ. Isolation of a lipopolysaccharide-binding acute phase reactant from rabbit serum. J Exp Med 1986; 164: 777-793. Hailman E, Lichenstein HS, Wurfel MM, Miller DS, Johnson DA, Kelley M, Busse LA, Zukowski MM and Wright SD. Lipopolysaccharide (LPS)-binding protein accelerates the binding of LPS to CD 14. J Exp Med 994; 179: 269-277. Schumann RR, Leong SR, Flaggs GW, Gray PW, Wright SD, Mathison JC, Tobias PS and Ulevitch RJ. Structure and function of lipopolysaccharide binding protein. Science 1990; 249:1429-1431. Baeuerle PA. Pro-inflammatory signaling: last pieces in the NF-kappa B puzzle. Curr Biol 1998; 8: R19-R22. Baeuerle PA and Baltimore D. NF kappa B: ten years after. Cell 1996; 87: 13-20. Rivest S. What is the cellular source of prostaglandins in the brain in response to systemic inflammation? facts and controversies. Mol Psychiat 1999; 4: 501-507. Rivest S, Lacroix S, Vallieres L, Nadeau S, Zhang J and Laflamme N. How the blood talks to the brain parenchyma and the paraventricular nucleus of the hypothalamus during systemic inflammatory and infectious stimuli. Proc Soc Exp B iol Med 2000; 223: 22-38. Lacroix S, Feinstein D and Rivest S. The bacterial endotoxin lipopolysaccharide has the ability to target the brain in upregulating its membrane CD14 receptor within specific cellular populations. Brain Pathol 1998; 8: 625-640. Oldfield BJ and Mckinley MJ. Circumventricular organs. In: The rat nervous system (Paxinos G, ed.) 1995; pp 391-403. San Diego: Academic Press. Nadeau S and Rivest S. Regulation of the gene encoding tumor necrosis factor alpha in the rat brain and pituitary in response to different models of systemic immune challenge. J Neuropathol Exp Neurol 1999; 58: 61-77. Quan N, Whiteside M and Herkenham M. Time course and localization patterns of interleukin-115 mRNA expression in the brain and pituitary after peripheral administration of lipopolysaccharide. Neuroscience 1997; 83:281-293. Vallieres L and Rivest S. Regulation of the genes encoding interleukin-6, its receptor, and gpl30 in the rat brain in response to the immune activator lipopolysaccharide and the proinflammatory cytokine interleukin- 1[5. J Neurochem 1997; 69:1668-1683. Fearns C, Kravchenko VV, Ulevitch RJ and Loskutoff DJ. Murine CD14 gene expression in vivo: extramyeloid synthesis and regulation by lipopolysaccharide. J Exp Med 1995; 181: 857-866. Beg AA, Finco TS, Nantermet PV and Baldwin AS. Tumor necrosis factor and interleukin-1 lead to phosphorylation and loss of I~:Bc~: a mechanism for NF-~zB activation. Mol Cell Biol 1993; 13: 3301-3310. Fearns C and Loskutoff DJ. Role of tumor necrosis factor alpha in induction of murine CD14 gene expression by lipopolysaccharide. Infect Immun 1997; 65: 4822-4831. Fearns C and Ulevitch RJ. Effect of recombinant interleukin-lbeta on murine CD 14 gene expression in vivo. Shock 1998; 9: 157-163. Nadeau S and Rivest S. Role of microglial-derived tumor necrosis factor in mediating CD14 transcription and NF-KB activity in the brain during endotoxemia. J Neurosci 2000; 20: 3456-3468.
173
25.
26.
27.
28.
29.
30.
31. 32. 33.
34.
35. 36.
37. 38.
39.
40. 41.
Fong Y, Tracey KJ, Moldawer LL, Hesse DG, Manogue KB, Kenney JS, Lee AT, Kuo GC, Allison AC, Lowry SF and et al. Antibodies to cachectin/tumor necrosis factor reduce interleukin 1 beta and interleukin 6 appearance during lethal bacteremia. J Exp Med 1989; 170: 1627-1633. Marino MW, Dunn A, Grail D, Inglese M, Noguchi Y, Richards E, Jungbluth A, Wada H, Moore M, Williamson B, Basu S and Old LJ. Characterization of tumor necrosis factordeficient mice. Proc Natl Acad Sci USA 1997; 94: 8093-8098. Baumann H, Morella KK and Wong GH. TNF-alpha, IL-1 beta, and hepatocyte growth factor cooperate in stimulating specific acute phase plasma protein genes in rat hepatoma cells. J Immunol 1993; 151: 4248-4257. Herx L, Rivest S and Yong W. Inflammatory cytokine regulation of neurotrophism following CNS trauma: a requirement for IL-I[3 in CNTF production. J Immunol 2000; 165: 2232-2239. Laflamme N, Lacroix S and Rivest S. An essential role of interleukin-l[3 in mediating NF-KB activity and COX-2 transcription in cells of the blood-brain barrier in response to systemic and localized inflammation, but not during endotoxemia. J Neurosci 1999; 19: 10923-10930. Basu S, Dunn AR, Marino MW, Savoia H, Hodgson G, Lieschke GJ and Cebon J. Increased tolerance to endotoxin by granulocyte-macrophage colony- stimulating factordeficient mice. J Immunol 1997; 159: 1412-1417. Anderson KV. Toll signaling pathways in the innate immune response. Curr Opin Immunol 2000; 12: 13-19. Wright SD. Toll, A New Piece in the Puzzle of Innate Immunity. J Exp Med 1999; 189: 605-609. Perera PY, Vogel SN, Detore GR, Haziot A and Goyert SM. CD 14-dependent and CD 14independent signaling pathways in murine macrophages from normal and CD 14 knockout mice stimulated with lipopolysaccharide or taxol. J Immunol 1997; 158: 4422-4429. Muzio M, Polentarutti N, Bosisio D, Prahladan MK and Mantovani A. Toll-like receptors: a growing family of immune receptors that are differentially expressed and regulated by different leukocytes. J Leukoc Biol 2000; 67: 450-456. Beutler B. Tlr4: central component of the sole mammalian LPS sensor. Curr Opin Immunol 2000; 12: 20-26. Takeuchi O, Hoshino K, Kawai T, Sanjo H, Takada H, Ogawa T, Takeda K and Akira S. Differential roles of TLR2 and TLR4 in recognition of gram-negative and gram-positive bacterial cell wall components. Immunity 1999; 11:443-451. Kawai T, Adachi O, Ogawa T, Takeda K and Akira S. Unresponsiveness of MyD88deficient mice to endotoxin. Immunity 1999; 11: 115-122. Laflamme N and Rivest S. Toll-like receptor 4: the missing link of the cerebral innate immune response triggered by circulating gram-negative bacterial cell wall components. FASEB J. 2001; 15: 155-163. Nomura F, Akashi S, Sakao Y, Sato S, Kawai T, Matsumoto M, Nakanishi K, Kimoto M, Miyake K, Takeda K and Akira S. Cutting edge: endotoxin tolerance in mouse peritoneal macrophages correlates with down-regulation of surface toll-like receptor 4 expression. J Immunol 2000; 164: 3476-3479. Mayhan WG. Effect of lipopolysaccharide on the permeability and reactivity of the cerebral microcirculation: role of inducible nitric oxide synthase. Brain Res 1998; 792: 353-357. Gonzalez-Scarano F and Baltuch G. Microglia as mediators of inflammatory and degenerative diseases. Annu. Rev Neurosci 1999; 22: 219-240.
This Page Intentionally Left Blank
New Foundationof Biology Edited by I. Berczi and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
175
Nitric Oxide in Neuroimmune Feedback Signaling
TERESA L. KRUKOFF and WENDY W. YANG
Department of Cell Biology and Division of Neuroscience, Faculty of Medicine and Dentistry, University of Alberta, Edmonton, AB, Canada T6G 2H7
ABSTRACT The gaseous neurotransmitter, nitric oxide (NO), has been implicated in regulation of the hypothalamo-pituitary-adrenal (HPA) axis. NO donors attenuate lipopolysaccharide (LPS)-induced release of corticotropin releasing factor (CRF) in vitro and NO synthase (NOS) inhibitors potentiate and prolong activation of the HPA axis by LPS in vivo. Changes in activities of the NO synthase isoforms, neuronal NOS (nNOS), endothelial NOS (eNOS), and inducible NOS (iNOS), have been reported in response to immune challenge. High doses of LPS administered either intravenously or intraperitoneally lead to increased activity of iNOS in perivascular microglia and in endothelial cells of microvascular origin. The NO produced by iNOS may in turn stimulate release of pro-inflammatory cytokines from cells of the brain. At much lower doses of endotoxin, however, where septic shock is not induced nor is the blood-brain barrier disrupted, nNOS and/or eNOS may play more important roles in NO production and signaling. Our work has shown that, in rats receiving 100 pg/kg intravenous LPS, blockade of NO production in the brain leads to elimination of the drop in body temperature and increased neuronal activation (including NO-producing neurons) in the paraventricular nucleus of the hypothalamus (PVN). The location of activated neurons in functionally distinct subdivisions of the PVN suggests that NO is involved in inhibition of the HPA axis, of sympathetic output, and of vasopressinand/or oxytocin-producing neurons in response to LPS. We also found that inhibition of NO production leads to increased gene expression of the cytokine, interleukin-1 ct (IL-1 ct), in non-neuronal cells of the PVN 4 hours after LPS injection and a return to baseline levels at 8 hours. While IL-1 t~ affects secretion of corticotropin releasing factor (CRF) from the PVN, the mechanism is controversial since neither IL-1 ct receptors nor IL-1 t~ binding have been reported in the PVN. NO's inhibition of IL-1 ct gene expression may, therefore, be mediated through an intermediate molecule (e.g. prostaglandins, cytokines, NO). Using inhibitors of the NOS isoforms, we provide evidence that eNOS is the isoform responsible for the effects described above. While it is most likely that eNOS found in the vasculature of the brain is responsible for these effects, other possible sources include neurons or glial cells because eNOS has been shown to be present in hippocampal neurons and in astrocytes activated by cytokines. In conclusion, NO is a strong messenger candidate as a mediator of signaling between the immune system and the brain, both in septic and non-septic conditions. Our data suggest that NO from eNOS inhibits neuronal
176
activation and IL-1 ot gene expression in the PVN and affects temperature regulation in response to relatively mild levels of immune stress.
1.
INTRODUCTION
The gas, nitric oxide (NO), is recognized as a versatile signalling molecule which has a wide array of effects throughout the body depending on its site and level of production, and the molecules with which it interacts. In the presence of oxygen and nicotinamide adenine dinucleotide phosphate, NO is produced from L-arginine by NO synthase (NOS). NO diffuses in aqueous and lipid environments and, because of spread in three dimensions, can potentially influence activity in many nearby cellular elements. Three isoforms of NOS have been identified. Neuronal NOS (nNOS/NOS I) and endothelial NOS (eNOS/NOS III) are constitutively expressed, calcium-dependent, and commonly regarded as being produced in neurons and endothelial cells, respectively. Activity of inducible NOS (iNOS/NOS II) is calcium-independent and almost undetectable under basal conditions; appropriate stimuli induce this enzyme at the transcriptional level, nNOS, eNOS, and iNOS are encoded by genes on chromosomes 7, 12, and 17, respectively [74]. nNOS and eNOS produce NO in the nanomolar range in response to intermittent increases in calcium concentration; iNOS can produce NO in the micromolar range for extended periods of time [14, 74]. This review will address two primary issues. First, the effects of immune challenges on activity of each NOS isoform within the brain will be discussed. Second, the current state of knowledge about the role of brain NO in regulating activity of the hypothalamo-pituitary-adrenal axis in response to immune challenges will be presented.
2.
IMMUNE CHALLENGE AFFECTS CENTRAL NO
2.1.
nNOS
Neurons producing NO are found in many autonomic centers in the brain, including in neurons of the parvocellular PVN (pPVN) which have the capacity to directly affect activity of the HPA axis through their projections to the median eminence. Enkephalin and corticotropinreleasing factor (CRF) are co-localized with nNOS in neurons of the pPVN [reviewed in 40]. In addition, NO neurons in the PVN possess the NMDA R~ subunit [4], supporting a role for glutamate as a regulator of NO neurons in this hypothalamic nucleus. Intravenous (i.v.) injections of lipopolysaccharide (LPS) at doses which do not cause endotoxic shock (100 pg/kg) have been shown to stimulate nNOS gene expression and NO content in the PVN [44; 79]; intracerebroventricular (i.c.v.) injections of IL-I[3 stimulate nNOS expression within the PVN [43]. Relatively large numbers of NO neurons in the pPVN are also activated by LPS (100 pg/kg, i.v.) as assessed by expression of the immediate early gene, c-fos [86]. In addition to its neuroimmune effects, this dose of LPS leads to changes in body temperature and arterial pressure [85] and to activation of neurons in the pPVN [41] including NO neurons [55]. Therefore, the potential exists for LPS to affect the central neuronal NO system through several routes, directly through neuroimmune pathways and indirectly through other autonomic pathways. As will be discussed below, one of the important roles that NO plays in the pPVN is to affect activity of the HPA axis.
177
High doses of LPS which cause endotoxic shock have been shown to stimulate or have no effect on nNOS gene expression in the hypothalamus. Using homogenates of hypothalamus, no differences in nNOS mRNA were found after 20 mg/kg intraperitoneal (i.p.) LPS [30; 66]. More selective assessment of nNOS mRNA with in situ hybridization, however, showed that i.p. LPS (25 mg/kg) stimulated nNOS gene expression in the PVN with a peak in expression at 5 hours after LPS injection [24]. These neurons were found primarily in the magnocellular PVN suggesting an interaction with vasopressin-and/or oxytocin-producing neurons. In addition, smaller numbers of neurons with increased levels of nNOS were found in the parvocellular PVN and about one-third of these neurons also expressed the CRF gene [24]. 2.2.
eNOS
Like nNOS, eNOS is a constitutive enzyme whose activity depends on the presence of Ca 2+. The primary source of eNOS in the brain is the vasculature where it is produced by endothelial cells. It has been claimed that neurons and glia do not normally produce eNOS [68; 72]. On the other hand, eNOS has been demonstrated in hippocampal [18; 32; 76], cortical [76] and thalamic neurons [76], and in astrocytes [2; 82; 10]. Viral infection has been shown to stimulate eNOS gene expression in glia both in vivo and in vitro [2], suggesting that cells other than endothelial cells may act as sources of eNOS in the brain during periods of immune challenge. Because of the association of astrocytes with blood vessels, astrocytic production of eNOS may provide a means through which blood-borne signaling molecules affect NO production which, in turn, may affect activity of nearby neurons. Until our recent study, surprisingly little attention had been paid to the effects of immune stimuli on eNOS activity. Studies focussed on identifying the source of NO which affects activity of the HPA axis often ignored eNOS [24; 30, 66]. Only in one study using i.v. injections of NOS inhibitors was it shown that a constitutive form of NOS, likely eNOS, suppresses the ACTH response to systemic injections of IL-I[3 [78]. Because of the systemic injections of inhibitors and IL-1 [3, however, this study did not differentiate whether the effects were central in nature or whether the effects were mediated at the level of the anterior pituitary gland. Our study using i.c.v, injections of NOS inhibitors has provided data which implicate eNOS in the brain as the source of NO which inhibits neuronal activation (as assessed by expression of Fos, the protein product of the immediate early gene, c-fos) and IL-I[3 gene expression in the PVN in response to i.v. LPS [85]. In addition, NO from eNOS is involved in mediating the drop in body temperature that occurs approximately 80 minutes after i.v. injection 100 ktg/kg of LPS [85]. A recent study has shown that brain astrocytes may be one of the sources of this eNOS as i.p. LPS (2.5 mg/kg) stimulated eNOS, but not iNOS, activity in these cells [29]. It is interesting that in vivo inhibition of NO production led to the conclusion that NO inhibits LPS-induced IL-I[3 gene expression in the PVN [85] while an in vitro study showed that NO stimulated IL-I[3 release from hypothalamic explant cultures under basal and KCl-stimulated conditions [49]. This discrepancy requires further investigation but, as pointed out in Section 3, may be related to inherent differences between studies using whole animals and those using hypothalamic explants. 2.3.
iNOS
iNOS is transcriptionally regulated and, after expression of the gene, iNOS is active for 4 to 24 hours to generate levels of NO which are 100 to 1000-fold greater than those produced by nNOS or eNOS [83]. Various cellular elements within the brain have the ability to produce iNOS.
178
LPS or cytokine-induced iNOS production occurs in perivascular cells [83], microglia [53, 11], neonatal glial cell cultures [36], adult astrocyte cultures [22, 70], and endothelial cells of microvascular origin [8]. High levels of NO have been implicated in neurotoxicity in vitro [16, 27]. I.c.v injections of LPS lead to iNOS-associated neuronal death in the vicinity of the injection site and to spatial memory deficits [87]. In primary cultures of hippocampal neurons, IL-I[3, TNF-c~, and interferon attenuated astrocytic high-affinity glutamate uptake through an NO-dependent process [88]. The resulting excess of glutamate at the synaptic cleft is, in turn, responsible for causing widespread neurotoxicity [ 12]. Furthermore, production of free radicals and subsequent damage and disruption of metabolic processes also contribute to the toxicity of high levels of NO [7]. Finally, TNF-ot and LPS-induced apoptosis in cultured PC12 cells was found to be dependent on NO from iNOS [26]. In addition to the direct damage to neurons and glia by NO from iNOS, it has been hypothesized that recurrent bouts of systemic infection may play a role in the pathogenesis of neuronal disease associated with aging and may impair the brain's ability to respond appropriately to stressors and infection [52]. Increases in the permeability of the blood-brain barrier (BBB) occur during systemic infection [6] and L-NAME administration attenuated LPS-induced changes in the BBB [9, 69]. That NO from iNOS is specifically important in this response was shown when topical LPS (200 ng/ml) applied onto pial arterioles caused opening of the BBB which could be blocked by aminoguanidine, a specific inhibitor of iNOS [51]. High doses of LPS associated with endotoxic shock stimulate iNOS production and activity within the brain. After 20 mg/kg of i.p. LPS, perivascular cells increased their iNOS gene expression 6 hours after injection [83]. High-dose i.p. LPS (20-25 mg/kg) induced gene expression of iNOS in homogenates of the hypothalamus [30, 66] and in tissue sections [24]. iNOS mRNA was first detected at 2 hours, was significantly elevated at 3 hours, and returned to basal levels at 12 hours. Gene expression of iNOS in the brain after i.p. LPS was attenuated by i.p. administration of dexamethasone, suggesting that glucocorticoids are directly or indirectly involved in the iNOS response [75]. LPS (10 mg/kg, i.v.) has also been shown to lead to increased iNOS expression in the brainstem nucleus of the tractus solitarius beginning at 5 hours after LPS administration [46]. iNOS expression after LPS treatment is dependent on time of exposure. LPS at 0.25 mg/kg i.p. stimulated a significant increase in iNOS mRNA at 8 hours but not at 3 hours [66]. iNOS was expressed in cells of the circumventricular organs, preoptic area, hippocampus, arcuate nucleus, and nucleus of the tractus solitarius [37]. Angiotensin II, a neuropeptide known best as a neurotransmitter and renal hormone, attenuates LPS and IL-1 [3-induced increases in iNOS activity in cultures of adult rat astrocytes [38]. As angiotensin II is regarded as an important regulator of central cardiovascular responses and of body fluid balance it may, therefore, participate in the integration of autonomic and neuroimmune responses to immune challenges. 2.4.
Molecular mechanisms of neuroimmune signaling by NO
NO activates guanylyl cyclase to catalyze production of the second messenger, cGMP, in target cells [20, 21, 47]. cGMP can then affect ion channel function or phosphodiesterase activity, or activate cGMP-dependent protein kinases to affect other cellular events [1] including gene transcription [56]. NO has been shown to inhibit constitutive NOS activity in cerebellar extracts, possibly through indirect inhibition of NMDA receptors or through interactions with ferric heme which is required for NOS activity [reviewed in 40].
179
NO affects activity of the DNA binding factor, NF-~B, which is essential for activation of several inflammatory mediators including TNF-c~, IL-I[3, IL-2, IL-6, IL-8, and interferon-j3 [5]. NO's effects may depend on the cell type in question. In glial cultures, NO from nNOS inhibited NF-~cB activity [77]. In cultured microglia, NO donors prevented LPS and TNF-c~-inducible NO synthesis from iNOS [13]. These effects were thought to be mediated through inhibition of NK-KB activity [13] since the iNOS promoter has a NK-KB binding site [84]. On the other hand, exposure of primary neuronal cultures of rat striatum to NO donors led to increased nuclear expression of the NK-KB subunits, p50 and p65, within 30 minutes [71]. As these subunits dimerize to form the NK-~zB which binds to DNA, these results were interpreted to indicate that NO has the potential to stimulate NK-KB activity [71], although a direct relationship to immune-stimulated changes in NK-KB activity in neurons remains to be demonstrated. Basal levels of NO have been measured in several types of tissue including endothelial cells and invertebrate ganglia [48]. Because addition of L-NAME resulted in a decrease in this basal level, it was assumed that the NO produced is physiologically relevant [48; 74]. It has been speculated that "tonal" NO from constitutive NOS tonically inhibits NF-~:B under basal conditions and that NO has, therefore, the ability to inhibit proinflammatory responses [74]. These investigators further hypothesized that, when the stimulus is sufficiently strong, a threshold of activation is surpassed, iNOS induction is not prevented, and relatively large amounts of NO are produced [74] to stimulate the effects described for iNOS above.
NITRIC OXIDE AFFECTS THE RESPONSE OF THE HPA AXIS TO IMMUNE CHALLENGE 3.1.
Effects of lipopolysaccharide and interleukin- 1[3
In the 1990's, investigators began to turn their attention to the role of NO in regulating activity of the HPA axis. In vitro studies have yielded contradictory results and have demonstrated both inhibitory and stimulatory roles for NO. In hypothalamic explants, it was shown that NO donors attenuated endotoxin-induced release of CRF but not basal CRF secretion [15]. On the other hand, inhibition of NO with Na-nitro-L-arginine or hemoglobin (a NO scavenger) in explant cultures attenuated the stimulatory effect of IL-1 on CRF secretion [64]. Finally, further confusion has been added to the area by a study in which inhibition of NO production reversed the inhibitory effects of LPS on release of CRF in hypothalamic explants [39]. In favor of an inhibitory role for NO on activity of the HPA axis, in vivo experiments showed that systemic L-NAME, a non-specific inhibitor of NOS, potentiated and prolonged the ACTH response to systemic LPS, IL-I[3, IL-6, or tumor necrosis factor-c~ [63; 62; 78; 34]. The proposed inhibitory role for NO on LPS-stimulated activity of the HPA axis proposed by the Rivier group was later contradicted by the same group when they showed that systemic L-NAME attenuated the ACTH response to i.v. LPS [79]. The reasons for this difference in results are not apparent at this time. Systemic L-NAME was shown to have no effect on the ACTH response to peripherally injected CRF or i.c.v. IL-I[3 which is thought to stimulate the HPA axis through CRF neurons in the PVN [61]. CRF mRNA levels were also unaffected [63], arguing against a direct action of NO on CRF neurons in the PVN. It has been suggested that NO's effects may occur at the level of the median eminence where CRF terminals are found or at the anterior pituitary [34]. In contradiction of the latter possibility, however, ACTH release stimulated by IL-I[3, CRF, vasopressin, or phorbol myristate acetate from anterior pituitary cell cultures was not affected
180
by application of the NOS inhibitor, N omega-Nitro-L-arginine (Nitro-arg) [25]. A further possible explanation [24] is that NO activates both guanylyl cyclase and cyclo-oxygenase [23] and that these enzymes have opposing effects on CRF secretion [24]. Some of the confusion with NOS inhibitor studies may arise from their systemic application, so that NO inhibition likely occurs, not only in the brain, but in peripheral sites including the pituitary gland. For example, i.v. or i.p. administration of L-NAME is well-known for its hypertensive effect due to systemic NO depletion. Activity in the PVN and the HPA axis may then be affected by changes in cardiovascular activity which have not been well controlled. Indeed, our results using i.c.v, inhibitors of NOS showed that, while arterial pressure responses were not affected, NO inhibited LPS-induced neuronal activation (Fos expression) in those subdivisions of the PVN which regulate activity of the HPA axis and inhibited IL-I[3 gene expression in the PVN [85]. Because IL-1[3 is stimulatory to CRF production and release [3, 65, 54, 31 ], our results support an inhibitory role for NO on the HPA axis. Finally, because NOS inhibition led to increased numbers of magnocellular neurons which were activated in response to LPS, we speculate that NO from eNOS inhibits LPS-induced secretion of vasopressin and/or oxytocin [85]. NO's effects on activity of the HPA axis may depend on the state of activation of the system. In the non-stimulated state, for example, direct injection of the NO donor, 3-morpholinosydnonimine (SIN-l), stimulated ACTH release and the response was blunted by injections of CRF or vasopressin antibodies [45]. In addition, increased gene expression of NGFI-B, CRF, and vasopressin were also noted. The authors hypothesized that, during unstimulated states, exogenous application of NO stimulates the HPA axis [45]. It is, however, probably too early to determine if this hypothesis is correct. First, the authors discounted the possibility that SIN-l-induced early increases in arterial pressure may have been responsible for some of the changes in gene transcription. On the contrary, increased pressure has been shown to be associated with changes in CRF gene expression in the PVN [42]. Second, the amount of NO released by SIN-1 is not known [45] and may have reached levels which are not characteristic of those released during responses to non-septic levels of immune challenge. The type of stressor may also affect NO's effects on activity of the HPA axis. Although outside the scope of this review, NO has been reported to have inhibitory [81] and stimulatory [35; 78] effects on activation of the HPA axis to non-immune physical stressors. 3.2.
Effects of other cytokines
NO has been implicated as a mediator of the IL-2-stimulated release of vasopressin and CRF from hypothalamic and amygdalar slices [57, 58]. Application of the NO donor, sodium nitroprusside, stimulated vasopressin and CRF release from these areas and application of the NOS inhibitor, NC-methyl-L-arginine (L-NMA), blocked the stimulatory effect of IL-2 on their release [57]. Similarly, the substrate for NOS, L-arginine, has been shown to enhance interleukin 2-induced CRF release in hypothalamus explants [33]. In explant cultures of median eminence, IL-10 stimulated NO production and CRF release through an NO-dependent process that was demonstrated by blocking NO production using L-NAME [73]. 3.3.
Vagal transmission
Information transfer about inflammation and cytokine levels in the peritoneum is mediated by the vagus nerve [80]. Little information is available about NO's role in transmission of
181
these signals, although intraperitoneal IL-I[3 stimulation of NO production in the PVN was significantly attenuated by subdiaphragmatic vagotomy [28]. 3.4.
NO and LPS-induced changes in body temperature
The effects of LPS on body temperature very much depend on the dose and route of administration. While low doses of i.v. LPS (5 or 10 pg/kg) stimulated increases in body temperature 2 to 6 hours after injection [19, 67], higher doses (100 or 125 pg/kg i.v.) produced a drop in temperature 80 to 100 minutes after injection [19; 85]. Intraperitoneal LPS at 100 jug/kg caused increased body temperature with a peak at about 5 hours [50]; this effect could be blocked with vagotomy [80], suggesting that the fever response is related to peritoneal inflammation. Most of the studies focussed on the role of NO in LPS-induced changes in body temperature have utilized systemic injections of NOS inhibitors; pyretic, antipyretic, or no effects have been described [59, 67, 60]. As systemic L-NAME may have its primary effects on LPS-induced changes in body temperature through its actions on brown fat thermogenesis [ 17], these earlier results may not be relevant to the role of brain NO in thermal regulation. On the other hand, we have used i.c.v. NOS inhibitors to show that central NO mediates the drop in body temperature which occurs 80 to 100 minutes after i.v. LPS (100 pg/kg) [85].
eNOS
.
PVN nNOS
iNOS
Anterior Pituitary
Adrenal glands Figure 1. Nitric oxide (NO) is produced from neuronal NO synthase (nNOS), endothelial NOS (eNOS), or inducible NOS (iNOS) within several cellular elements found in the paraventricular nucleus of the hypothalamus (PVN) including neurons (N), astrocytes (A), endothelial cells of blood vessels (B), and microglia (M). Most available data support the hypothesis that NO from nNOS and eNOS inhibits the hypothalamo-pituitary-adrenal axis as a feedback mechanism to limit chronic activation of the immune system.
182
4.
CONCLUSIONS
While it is clear that NO participates in communication between the immune system and the brain, the precise role(s) for NO in neuroimmune signaling appear(s) to depend on the relative amounts of NO released in the brain. Low to moderate levels of immune challenge stimulate NO production from nNOS and eNOS whereas more intense immune challenges stimulate production of larger amounts NO from iNOS. All three types of NO release occur within the PVN (Figure 1). Most of the data published to date support the hypothesis that NO from nNOS and/or eNOS inhibits activity of the HPA axis. NO from iNOS, on the other hand, has toxic or inflammatory effects within the brain which may or may not lead to tissue damage. Challenges to homeostasis by immune stressors lead to activation of the HPA axis as part of the "host defence response", and recovery of an animal from disease is dependent on the acute neural and hormonal activation of the immune system through this pathway. On the other hand, chronic stimulation of these potent systems is deleterious to the long-term survival of the animal, making a negative feedback mechanism necessary. The current knowledge about the central NO system in the brain leads to the hypothesis that the inhibitory effects of NO from nNOS and eNOS on the HPA axis are related to the feedback loop which returns the animal to homeostatic balance through central inhibition of the immune system.
REFERENCES 1.
2. 3.
4.
5.
6. 7.
8.
9.
Arnold WP, Mittal CK, Katsuki S, Murad F. Nitric oxide activates guanylyl cyclase and increases guanosine 3': 5'-cyclic monophosphate levels in various tissue preparations. Proc Natl Acad Sci USA 1977; 74: 3203-3207. Barna M, Komatsu T, Reiss CS. Activation of type III nitric oxide synthase in astrocytes following a neurotropic viral infection. Virology 1996; 223:331-343. Berkenbosch F, VanOers J, Del Ray A, Tilders F, Besedovsky H. Corticotropin-releasing factor-producing neurons in the rat activated by interleukin-1. Science 1987; 238: 524-526. Bhat GK, Mahesh VB, Lamar CA, Ping L, Aguan K, Brann DW. Histochemical localization of nitric oxide neurons in the hypothalamus: association with gonadotropinreleasing hormone neurons and co-localization with N-methyl-D-aspartate receptors. Neuroendocrinology 1995; 62: 187-197. Blackwell TS, Blackwell TR, Holden EP, Christman BW, Christman JW. In vivo antioxidant treatment suppresses nuclear factor-kappa B activation and neutrophilic lung inflammation. J Immunol 1996; 157: 1630-1637. Boje KM. Inhibition of nitric oxide synthase attenuates blood-brain barrier disruption during experimental meningitis. Brain Res 1996; 720: 75-83. Bolanos JP, Almeida A, Stewart V, Peuchen S, Land JM, Clark JB, Heales SJ. Nitric oxide-mediated mitochondrial damage in the brain: mechanisms and implications for neurodegenerative diseases. J Neurochem 1997; 68: 2227-2240. Bonman E, Suschek C, Spranger M, Kobl-Bachofen V. The dominant role of exogenous or endogenous interleukin-1 beta on expression and activity of inducible nitric oxide synthase in rat microvascular brain endothelial cells. Neurosci Lett 1997; 230:109-112. Buster BL, Weintrob AC, Townsend GC, Scheld WM. Potential role of nitric oxide in the pathophysiology of experimental bacterial meningitis in rats. Infect Immunol 1995; 63: 3835-3839.
183
10.
11.
12. 13.
14. 15.
16.
17. 18.
19.
20. 21.
22. 23.
24.
25.
26.
Caillol M, Devinoy E, Lacroix MC, Schirar A. Endothelial and neuronal nitric oxide synthases are present in the suprachiasmatic nuclei of syrian hamsters and rats. Eur J Neurosci 2000; 12: 649-661. Casamenti F, Prosperi C, Scali C, Giovannelli L, Faussone-Pellegrini MS, Pepeu G. Interleukin-1 beta activates forebrain glial cells and increases nitric oxide production and cortical glutamate and GABA release in vivo: implications for Alzheimer's disease. Neuroscience 1999; 91: 831-842. W. Choi, Glutamate neurotoxicity and diseases of the nervous system. Neuron. 1988; 1: 623-634. Colasanti M, Persichini T, Menegazzi M, Mariotto S, Giordano E, Caldarera CM, Sogos V, Lauro GM, Suzuki H. Induction of nitric oxide synthase mRNA expression. Suppression by exogenous nitric oxide, J Biol Chem, 1995; 270:26731-26733. Cook JP, Dzau VJ, Nitric oxide synthase: role in the genesis of vascular disease, Ann Rev Med 1997; 48: 489-509. Costa A, Trainer P, Besser M, Grossman A. Nitric oxide modulates the release of corticotropin-releasing hormone from the rat hypothalamus in vitro. Brain Res 1993; 605: 187-192. Dawson VL, Brahmbhatt HP, Mong JA, Dawson TM. Expression of inducible nitric oxide synthase causes delayed neurotoxicity in primary mixed neuronal-glial cortical cultures. Neuropharmacology 1994; 33: 1425-1430. De Luca B, Monda M, Sullo A. Changes in eating behavior and thermogenic activity following inhibition of nitric oxide formation. Amer J Physiol 1995; 268: R1533-R1538. Dinerman JL, Dawson TM, Schell MJ, Snowman A, Snyder SH. Endothelial nitric oxide synthase localized to hippocampal pyramidal cells: implications for synaptic plasticity. Proc Natl Acad Sci USA 1994; 91: 4214-4218. Elmquist JK, Scammell TE, C.D. Jacobson, Saper CB. Distribution of Fos-like immunoreactivity in the rat brain following intravenous lipopolysaccharide. J Comp Neurol 1996; 371: 1-19. F6rsterman U, Kleinert H. Nitric oxide synthase: expression and expressional control of the three isoforms. Naunyn-Schmied Arch Pharmacol 1995; 352: 351-364. F6rsterman U, Closs El, Pollock JS, Nakane M, Schwarz P, Gath I, Kleinert H Nitric oxide synthase isozymes. Characterization, purification, molecular cloning, and functions. Hypertension 1994; 23: 1121-1131. Galea E, Reis DJ, Feinstein DL. Cloning and expression of inducible nitric oxide synthase from rat astrocytes. J Neurosci Res 1994; 37:406-414. Grossman AB, Costa A, Forsling ML, Jacobs R, Kostoglou-Athanassiou I, Nappi G, Navarra P, Satta MA. Gaseous neurotransmitters in the hypothalamus. Horm Metab Res 1997; 29: 477-482. Harada S, Imaki T, Chikada N, Naruse M, Demura H. Distinct distribution and timecourse changes in neuronal nitric oxide synthase and inducible NOS in the paraventricular nucleus following lipopolysaccharide injection. Brain Res 1999; 821: 322-332. Hashimoto K, Nishioka T, Tojo C, Takao T. Nitric oxide plays no role in ACTH release induced by interleukin-1 beta, corticotropin-releasing hormone, arginine vasopressin and phorbol myrsitate acetate in rat pituitary cell cultures. Endocrine J 1995; 42: 435-439. Heneka MT, Loschmann PA, Gleichmann M, Weller M, Schulz JB, Wullner U, Klockgether T. Induction of nitric oxide synthase and nitric oxide mediated apoptosis in neuronal PC12 cells after stimulation with tumor necrosis factor-alpha/lipopolysaccharide. J Neurochem 1998; 71: 88-94.
184
27. 28.
29.
30.
31.
32.
33.
34. 35.
36.
37.
38.
39.
40. 41.
42.
43.
Hewett SJ, Csernansky CA, Choi DW. Selective potentiation of NMDA-induced injury following induction of astrocytic iNOS. Neuron 1994; 13: 487-494. Ishizuka Y, Ishida Y, Jin Q-J, Shimokawa A, Saita M, Kato K, Kunitake T, Hanamori T, Mitsuyama Y, Kannan H. Abdominal vagotomy attenuates interleukin-l[3-induced nitric oxide release in the paraventricular nucleus region in conscious rats. Brain Res 1998; 789: 157-161. Iwase K, Miyanaka K, Shimizu A, Nagasaki A, Gotoh T, Mori M, Takiguchi M. Induction of endothelial nitric-oxide synthase in rat brain astrocytes by systemic lipopolysaccharide treatment. J Biol Chem 2000; 275:11929-11933. Jacobs RA, Satta MA, Dahia PL, Chew PL, Grossman AB. Induction of nitric oxide synthase and interleukin-lbeta, but not heme oxygenase, messenger RNA in rat brain following peripheral administration of endotoxin. Molec Brain Res 1997; 49: 238-246. Kakucska I, Qi Y, Clark B D, Lechan RM. Endotoxin-induced corticotropin-releasing hormone gene expression in the hypothalamic paraventricular nucleus is mediated centrally by interleukin- 1. Endocrinology 1993; 133:815-821. Kantor KB, Lanzrein M, Stary SJ, Sandoval GM, Smith WB, Sullivan BM, Davidson N, Schuman EM. A role for endothelial NO synthase in LTP revealed by adenovirusmediated inhibition and rescue. Science 1996; 472:1744-1748. Karanth S, Lysin K, McCann SM. Role of nitric oxide in interleukin-2-induced corticotropin-releasing factor release from incubated hypothalamus. Proc Natl Acad Sci USA 1993; 90: 3383-3387. Kim CK, Rivier C. Influence of nitric oxide synthase inhibitors on the ACTH and cytokine responses to peripheral immune signals. J Neuroendocrinol 10 (1998) 353-362. Kim CK, Rivier C. Nitric oxide and carbon monoxide have a stimulatory role in the hypothalamic-pituitary-adrenal response to physico-emotional stressors in rats. Endocrinol 2000; 141 : 2244-2253. Kitamura Y, Takahashi H, Nomura Y, Taniguchi T. Possible involvement of Janus kinase Jak2 in interferon-gamma induction of nitric oxide synthase in rat glial cells. Eur J Pharmacol 1996; 306: 297-306. Konsman JP, Kelley K, Dantzer R. Temporal and spatial relationships between lipopolysaccharide-induced expression of Fos, interleukin-l[3 and inducible nitric oxide synthase in rat brain. Neuroscience 1999; 89: 535-548. Kopnisky KL, Sumners C, Chandler LJ. Cytokine- and endotoxin induced nitric oxide synthase in rat astroglial cultures: differential modulation by angiotensin II. J Neurochem 1997; 68: 935-44. Kostoglou-Athanassiou I, Costa A, Navarra P, Nappi G, Forsling ML, Grossman AB. Endotoxin stimulates an endogenous pathways regulating corticotropin-releasing hormone and vasopressin release involving the generation of nitric oxide and carbon monoxide. J Neuroimmunol 1998; 86: 104-109. Krukoff TL. Central actions of nitric oxide in regulation of autonomic functions. Brain Res Rev 1999; 30: 52-65. Krukoff TL, MacTavish D, Jhamandas JH. Activation by hypotension of neurons in the hypothalamic paraventricular nucleus that project to the brainstem. J Comp Neurol 1997; 385: 285-296. Krukoff TL, MacTavish D, Jhamandas JH. Hypertensive rats exhibit heightened expression of corticotropin-releasing factor in activated central neurons in response to restraint stress. Molec Brain Res 1999; 65: 70-79. Lee S, Rivier C. Interaction between corticotropin-releasing factor and nitric oxide in
185
mediating the response of the rat hypothalamus to immune and non-immune stimuli. Molec Brain Res 1998; 57: 54-62. 44. Lee S, Barbanel G, Rivier C. Systemic endotoxin increases steady-state gene expression of hypothalamic nitric oxide synthase comparison with corticotropin-releasing factor and vasopressin gene transcripts. Brain Res 1995; 705: 136-148. 45. Lee S, Kim K, Rivier C. Nitric oxide stimulates ACTH secretion and the transcription of the genes encoding the NGFI-B, corticotropin-releasing factor, corticotropin-releasing factor receptor type I, and vasopressin in the hypothalamus of the intact rat. J Neurosci 1999; 19: 7640-7647. 46. Lin H-C, Wan F-J, Kang B-H, Wu C-C, Tseng C-J. Systemic administrations of lipopolysaccharide induces release of nitric oxide and glutamate and c-fos expression in the nucleus tractus solitarii of rats. Hypertension 1999; 33: 1218-1224. 47. L6pez-Figueroa MO, Day HEW, Akil H, Watson SJ. Nitric oxide in the stress axis. Histol. Histopathol. 1998; 13: 1243-1252. 48. Magazine HI, Liu Y, Bilfinger TV, Fricchione GL, Stefano GB. Morphine-induced conformational changes in human monocytes, granulocytes, and endothelial cells and in invertebrate immunocytes and ganglia are mediated by nitric oxide. J Immunol 1996; 156: 4845-4850. 49. Mancuso C, Tringali G, Grossman A, Preziosi P, Navarra P. The generation of nitric oxide and carbon monoxide produces opposite effects on the release of immunoreactive interleukin-l[5 from the rat hypothalamus in vitro: evidence for the involvement of different ignaling pathways. Endocrinology 1998; 139: 1031-1037. 50. Matsumura K, Cao C, Ozaki M, Morii H, Nakadate K, Watanabe Y. Brain endothelial cells express cyclooxygenase-2 during lipopolysaccharide-induced fever: light and electron microscopic immunocytochemical studies. J Neurosci 1998; 18: 6279-6289. 51. Mayhan WG. Effect of lipopolysaccharide on the permeability and reactivity of the cerebral microcirculation: role of inducible nitric oxide synthase. Brain Res 1998; 792: 353-357. 52. McCann SM, Licinio J, Wong ML, Wong WH, Karanth S, Rettori V. The nitric oxide hypothesis of brain aging. Exper. Gerontol. 1998; 33: 813-826. 53. Minghetti L, Levi G. Microglia as effector cells in brain damage and repair: focus on prostanoids and nitric oxide. Prog Neurobiol 1998; 54: 99-125. 54. Navarra P, Tsagarakis S, Faria MS, Rees LH, Besser GM, Grossman AB. Interleukins- 1 and -6 stimulate the release of corticotropin-releasing hormone-41 from rat hypothalamus in vitro via the eicosanoid cyclooxygenase pathway. Endocrinology 1991; 128: 37-44. 55. Petrov T, Harris KH, MacTavish D, Krukoff TL, Jhamandas JH. Hypotension induces Fos immunoreactivity in NADPH-diaphorase positive neurons in the paraventricular and supraoptic hypothalamic nuclei of the rat. Neuropharmacology 1995; 34:509-514. 56. Peunova N, Enikolopov G. Amplification of calcium-induced gene transcription by nitric oxide in neuronal cells. Nature 1993; 364: 450-453. 57. Raber J, Bloom FE. IL-2 induces vasopressin release from the hypothalamus and the amygdala: role of nitric oxide-mediated signaling. J Neurosci 1994; 14: 6187-6195. 58. Raber J, Koob GF, Bloom FE. Interleukin-2 (IL-2) induces corticotropin-releasing factor (CRF) release from the amygdala and involves a nitric oxide-mediated signaling: comparison with the hypothalamic response. J Pharmacol Exp Therapeut 1995; 272: 815-824. 59. Redford J, B ishai I, Coceani F. Pyrogen-prostaglandin coupling in the pathogenesis of fever: evidence against a role for nitric oxide. Can. J Physiol Pharmacol 1995; 73:
186
60. 61.
62.
63.
64. 65. 66.
67. 68. 69. 70. 71.
72.
73.
74.
75.
76.
77.
1466-1474. Riedel W. Antipyretic role of nitric oxide during endotoxin-induced fever in rabbits. Int J Tissue React 1997; 19: 171-178. Rivest S, Torres G, Rivier C. Differential effects of central and peripheral injection of interleukin-l[3 on brain c-los expression and neuroendocrine functions. Brain Res 1992; 587: 13-23. Rivier C. Blockade of nitric oxide formation augments adrenocorticotropin released by blood-borne interleukin-1 beta: role of vasopressin, prostaglandins, and alpha 1-adrenergic receptors. Endocrinology 1995; 136: 3597-3603. Rivier C, Shen GH. In the rat, endogenous nitric oxide modulates the response of the hypothalamic-pituitary-adrenal axis to interleukin-1, vasopressin, and oxytocin. J Neurosci 1994; 14: 1985-1993. Sandi C, Guaza C. Evidence for a role of nitric oxide in the corticotropin-releasing factor release induced by interleukin-1 beta. Eur J Pharmacol 1995; 274: 17-23. Sapolsky R, Rivier C, Yamamoto G, Plotsky P, Vale W. Interleukin-1 stimulates the secretion of hypothalamic corticotropin-releasing factor. Science 1987; 238: 522-524. Satta MA, Jacobs RA, Kaltsas GA, Grossman AB. Endotoxin induced interleukin-lbeta and nitric oxide synthase mRNA in rat hypothalamus and pituitary. Neuroendocrinology 1998; 67:109-116. Scammell TE, Elmquist JK, Saper CB. Inhibition of nitric oxide synthase produces hypothermia and depresses lipopolysaccharide fever. Amer J Physiol 1996; 271: R333-R338. Seidel B, Stanarius A, Wolf G. Differential expression of neuronal and endothelial nitric oxide synthase in blood vessels of the rat brain. Neurosci Lett 1997; 239: 109-112. Shukla A, Dikshit M, Srimal RC. Nitric oxide modulates blood-brain barrier permeability during infections with an inactivated bacterium. Neuroreport 1995; 6:1629-1632. Simmons ML, Murphy S. Induction of nitric oxide synthase in glial cells. J Neurochem 1992; 59: 897-905. Simpson CS, Morris BJ. Actiation of nuclear factor KB by nitric oxide in rat striatal neurones: differential inhibition of the p50 and p65 subunits by dexamethasone. J Neurochem 1999; 73: 353-361. Stanarius A, Topel I, Schulz S, Noack H, Wolf G. Immunocytochemistry of endothelial nitric oxide synthase in the rat brain: a light and electron microscopical study using the tyramide signal amplification technique. Acta Histochem. 1997; 99: 411-429. Stefano GB, Prevot V, Beauvillain JC, Hughes TK. Interleukin-10 stimulation of corticotropin releasing factor median eminence in rats: evidence for dependence upon nitric oxide production. Neurosci Lett 1998; 256:167-170. Stefano GB, Goumon Y, Bilfinger TV, Welters ID, Cadet P. Basal nitric oxide limits immune, nervous and cardiovascular excitation: human endothelia express a mu opiate receptor. Prog Neurobiol 2000; 60:513-530. Suzuki E, Shintani F, Nakaki T, Nagata K, Yamazoe Y, Fujita N, Asai M, Kanba S. Dexamethasone and nitric oxide synthase gene expression in brain. J Psychiat Neurosci 1997; 22: 105-110. Teichert AM, Miller TL, Tai SC, Wang Y, Bei X, Robb B, Phillips MJ, Marsden PA. In vivo expression profile of an endothelial nitric oxide synthase promoter transgene. Am J Physiol 2000; 278: H1352-H1361. Togashi H, Sasaki M, Frohman E, Taira E, Ratan RR, Dawson TM, Dawson VL. Neuronal (type I) nitric oxide synthase regulated nuclear factor kappa B activity and immunologic (type II) nitric oxide synthase expression. PNAS 1997; 94: 2676-2680.
187
78.
79.
80. 81.
82. 83.
84. 85.
86.
87.
88.
Turnbull AV, Rivier C. Selective inhibitors of nitric oxide synthase (NOS) implicate a constitutive isoform of NOS in the regulation of interleukin-l-induced ACTH secretion in rats. Endocrine 1996; 5: 135-140. Uribe M, Lee S, Rivier C. Endotoxin stimulates nitric oxide production in the paraventricular nucleus of the hypothalamus through nitric oxide synthase I: correlation with hypothalamic-pituitary-adrenal axis activation. Endocrinology 1999; 140:5971-5981. Watkins LR, Maier SF, Goehler LE. Cytokine-to-brain communication: a review and analysis of alternative mechanisms. Life Sci 1995; 57:1011-1026. Weidenfeld J, Feldman S, DeKeyser FG, Ovadia H. Effect of exogenous nitric oxide and inhibitors of nitric oxide synthase on the hypothalamic pituitary adrenal axis responses to neural stimuli. Neuroendocrinology 1999; 70: 153-159. Wiencken AE, Castagrande VA. Endothelial nitric oxide synthase (eNOS) in astrocytes: another source of nitric oxide in neocortex. Glia 1999; 26: 280-290. Wong ML, Bongiorno PB, al-Shekhlee A, Esposito A, Khatri P, Licinio J. IL-1 beta, IL-1 receptor type I and iNOS gene expression in rat brain vasculature and perivascular areas. Neuroreport 1996; 7: 2445-2448. Xie Q, Nathan C. The high-output nitric oxide pathway: role and regulation. J Leukocyte Biol 1994; 56: 576-582. Yang WW, Krukoff TL. Nitric oxide regulates body temperature, neuronal activation and interleukin-1 [5 gene expression in the hypothalamic paraventricular nucleus in response to immune stress. Neuropharmacology 2000; 39: 2075-2089. Yang WH, Oskin O, Krukoff TL. Immune stress activates putative nitric oxide-producing neurons in rat brain: cumulative effects with restraint. J Comp Neurol 1999; 405: 380-387. Yamada K, Komori Y, Tanaka T, Senzaki K, Nikai T, Sugihara H, Kameyama T, Nabeshima T. Brain dysfunction associated with an induction of nitric oxide synthase following an intracerebral injection of lipopolysaccharide in rats. Neuroscience 1999; 88:281-294. Ye ZC, Sontheimer H. Cytokine modulation of glial glutamate uptake: a possible involvement of nitric oxide. NeuroReport 1996; 7: 2181-2185.
This Page Intentionally Left Blank
Ill.
NEUROIMMUNE MECHANISMS IN PHYSIOLOGY
This Page Intentionally Left Blank
New Foundationof Biology Edited by I. Berczi and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
191
Introduction
REGINALD M. GORCZYNSKI
Department of Surgery & Immunology, University of Toronto, The Toronto Hospital, CCRW 2-855, 200 Elizabeth Street, Toronto, ON, Canada M5G-2C5
In this following section, the reader will find a series of papers in which authors have examined different physiological systems in order to uncover the mechanisms(s) whereby CNS control of function may be mediated. We have already encountered some preliminary thoughts in this area with the previous chapters on inflammation and CNS: immune system regulation (see for instance, Nance et al., (chapter II-7) and Merali et al., Rivest et al., (chapters II-8 and II-9 respectively). The section begins with a timely review from B ienenstock and colleagues of their elegant work, over nearly two decades, which has focused on the communication between mucosal mast cells and the nervous system. There follows the first in a number of discussion to be found in this volume of an issue many readers will find one of the most provocative and exciting in this field at the present time, namely the key role of molecules produced by the submandibular gland (SMG) in neuroimmunology. Much of this work stems from the laboratory of Befus in Edmonton. In this chapter, the first to introduce the concepts, Forsythe et al., characterize the structure, including the peptides produced by the SMG, and regulation of this gland by hormones (e.g. by androgens and the thyroid gland especially), documenting the integration of the gland within the context of the body's endocrine system. There is also evidence for autonomic system control of the SMG, which leads immediately to consideration of its role in immunoregulation, given the knowledge that surgical sympathectomy has long been favoured as a way to study CNS: immune system interactions. The authors discuss the role for the SMG in inflammation, and regulation of mast cell function, and proceed to characterize some of the peptides, and the active moieties thereof, which are implicated in these functions (see also Davison et al., (Chapter II-3)). They conclude by stressing that investigations of this gland "have the potential to provide valuable insights into the mechanisms underlying psychological and neuroendocrine control of the immune system". This chapter is followed, appropriately, by a discussion by Sabbadini et al., on the evidence for the importance of another salivary gland peptide, kallikrein, in CNS-mediated immunoregulation. This molecule, a member of the serine protease family of enzymes, was active in modifying lymphocyte proliferation in vitro, and suppressing a variety of immune responses in vivo, including allograft rejection and contact hypersensitivity reactions-in all cases, activity was dependent upon enzyme activity being retained. These data suggest that there is a crucial substrate, in vivo, which when acted upon by kallikrein, releases immunoregulatory peptides. Even more provocative is the data from this group concerning the potential role for kallikrein in oral tolerance induction. Arguing that its high levels in saliva might have a physiological
192
role, the authors tested the effect of removal of the salivary gland on oral tolerance induction to collagen in rats. No tolerance developed, but oral replacement of kallikrein fully restored tolerance induction. The target molecules for these effects remain unknown. The two following chapters represent somewhat of a recursion to more phenomenological studies of CNS: immune interactions, though not wholely so. Gorczynski attempts to provide some logic, and even molecular understanding, of studies which have documented so-called "classical conditioning" of immune responses. There are a number of such experiments in the literature, all of which might lead the reader to believe that in principle, any immune response system might be "brought under" the control of the CNS (by biobehavioural techniques). If this is the case, Gorczynski asks, what might be happening within the CNS? Taking production of inflammatory cytokines (TNFet and IL-1) in response to LPS (or its perceived presence, in conditioned mice) as a measuring system, and using monoclonal antibodies and quantitative PCR technology as a measuring instrument, Gorczynski argues that both peripherally, and in the CNS, the same two peptides, somatostatin and substance P, play important regulatory roles. This can be seen as supporting of those studies, which have documented redundancy in the mediators used within the CNS and immune system (see the previous section of this book). Nevertheless their relative importance in the conditioned mice is quite different from that seen in mice where immunity is induced by conventional immunological tools. This discussion of TNFc~ and IL-1 induction in the CNS is timely given the thrust of the next chapter, by Moldofsky and colleagues, concerning an investigation of the role of sleep, and perturbations of it, in general health and immunocompetence. It has become increasingly evident that disruption of sleep has profound consequences to the function of the immune system as a whole, in addition to its effects on other physiological systems. As an example, sleep deprived rats die from systemic sepsis. Furthermore, data from Krueger and co-workers over the last decade have clearly established a role for TNFc~ and IL-1 production, within the CNS itself, in regulation of sleep induction. The potential communication between such cytokines and the CNS/immune system, and the additional role of the HPA axis and neuroendocrine circuitry, in sleep physiology, offers an important avenue of approach for those interested in manipulating those disruptions in sleep which have been implicated in human pathology and disease. Few areas of contemporary research are likely to provoke as much interest to the lay and scientific community as those in which the subject matter involves the genital organs, or obesity. These topics are introduced by our last two chapters in this section. Pomerantz lays to rest the notion that all physiological regulation of the testis can be considered in terms of the HPA axis, and instead forces us to consider the importance of the nervous system, the immune system and even a number of intragonadal factors. As an example, a number of groups have reported on peptides contained in Sertoli-cell-conditioned-medium with immunoregulatory properties; the expression of FasL by Sertoli cells in rats has been implicated as the critical factor explaining (in this species) the immunoprivilige accorded to this site for tumor growth. And finally, Pomerantz's group has also documented an important role for iNOS induction and NO production by Sertoli cells (like immune macrophages) in response to inflammatory stimuli, which can in turn decrease androgen production by neighbouring Leydig cells. Luheshi concludes this section with a discussion of the interactions between leptin and cytokines in both fever and control of obesity. Leptin was believed to be produced predominantly by adipocytes in relation to body mass, to gain access to the brain via a saturable transport mechanism, and to act on hypothalamic receptors to suppress appetite. IL-1 (in the periphery) is implicated in leptin production, while secondary mediators of leptin action in the CNS include members of the pro-opiomelanocortin family, CRF and neuropeptide-Y. Most interestingly, however, Luheshi suggests that leptin can induce IL-1 in the hypothalamus, and this in turn leads
193
him to speculate that actions of leptin on food intake might be directly related to production and action of IL-1 within the CNS. In addition, doses of leptin which induce appetite suppression are pyrogenic (a known effect of IL-1 within the CNS), suggestive of widespread ramifications for leptin as a mediator of neuroimmune regulation in human disease.
This Page Intentionally Left Blank
New Foundation of Biology
195
Edited by I. Berczi and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
A Model of Neuroimmune Communication Mast Cells and Nerves
JOHN BIENENSTOCK
Departments of Medicine, Pathology and Molecular Medicine, McMaster University, Hamilton, Ontario
ABSTRACT Communication between mast cells and nerves is perhaps the best current example of neuroimmune interaction. This occurs in a bidirectional fashion and has been demonstrated in a variety of animal and human tissues in vitro, in vivo and ex vivo. Examples of this interaction and its importance have been produced in experimental models as well as in human diseases, and the unit of mast cell-nerve is thought to play an important homeostatic role in several tissues such as skin, lung, intestine and urinary bladder. The unit is involved in complex interactions which include psychological stress in which corticotropin releasing factor (CRF) plays a role. Nerve growth factor (NGF), a member of the neurotrophin family, which is synthesized by many cell types, both structural and inflammatory, also plays an important role in maintaining the physiological state, and in addition is involved in as yet unknown fashion in inflammation and disease. NGF has an enormously pleiotropic activity and can be both pro and antiinflammatory. Since NGF is synthesized by mast cells and nerves, and because of its neuronal and non-neuronal effects, it plays a significant role in the maintenance of health, and an as yet unknown role in disease.
1.
INTRODUCTION
The purpose of this review is to outline some of the interactions between the nervous and immune systems as seen through the eyes of myself and my many colleagues who have helped me in my research over the years. This short review is far from exhaustive, but nevertheless serves to highlight some of the experimental evidence which supports Blalock's general thesis of bidirectional communication between the nervous and immune systems [ 1].
2.
NERVE-MAST CELL COMMUNICATION
While there had been, in the literature, some previous references to occasional connections between mast cells and nerves, we carried out the first detailed morphometric study in this area [2].
196
We showed that mast cells and nerves were commonly in association in the small intestinal villi of rats. These nerves contained substance P (SP) and calcitonin gene related peptide (CGRP). The analysis showed that the association was not by chance alone and therefore was purposeful. In many subsequent studies, others have shown this association to hold true in tissues from different animal species and various body sites which include small and large bowel, mesentery, portal tracts of the liver, skin, lung, blood vessels, urinary bladder, etc. [3, 4]. That this anatomical association had some functional attributes was shown by Baird and Cuthbert in guinea pigs sensitized to lactoglobulin [5]. In an in vitro system, antigen was shown to have a pronounced physiological affect on the epithelium. These observations have been extended enormously since that time mainly through the use of the Ussing chamber. Thus, in mouse, rat, guinea pig, human and primate tissues a number of investigators have shown that antigen is capable of causing short circuit current changes whether placed on the luminal or serosal side of the tissue, using pharmacological approaches [6-9]. In this model, antigen causes chloride and water secretion by the epithelium as a result of interaction between mast cells and nerves. Various mediators derived from mast cells have been shown to be involved in whole or in part, and these include histamine and serotonin. The process can be inhibited by atropine and substance P antagonists. Thus, an axon reflex is thought to be initiated in both intestine and lung [10] which involves antigen, sensitized mast cells, local nerves and target tissue. Mast cells are an obligatory component of this reaction, as shown by Perdue et al., using the mast cell deficient w/w v strain [7]. Conclusive evidence followed from studies of the repopulation of these animals with mast cells derived from bone marrow culture, which restored the response to antigen in Ussing chambers to its former level. Cooke and co-workers [6] showed in guinea pig intestine, again using the Ussing chamber, that in exactly the same time course as the development of a short circuit current response to antigen, acetylcholine was released, indicating that antigen-dependent neurotransmitter release had occurred. We have performed co-culture using sympathetic superior cervical ganglion neurons which change their phenotype in the culture conditions used, to those of cholinergic neurones. Mast cells (both peritoneal and rat basophil leukemic cells, an exemplar of mucosal mast cells) have been shown to purposefully associate with such nerves in culture, with both trophic and tropic effects [11-14]. There are electrical consequences to mast cell association with nerves, once contact has been made. The contacts have been maintained for 72 hours in vitro and associated mast cells matured, increased their granulation and released more mediator on challenge. In the associated nerve, the appearance of dense core vesicles as well as their accumulation, again suggested a meaningful and significant association. The contacts between the neuron and mast cell were intimate (less than 20 nm). However no specialized synaptic structural changes occurred either in the associated neurone or mast cell. We have used this model to further study communication between mast cells and nerves and have co-cultured RBL-2H3 with superior cervical ganglion neurites which, in the culture conditions employed, synthesize and release substance P (SP). Cells were loaded with the calcium fluorphore fluo-3 and examined by confocal laser scanning microscopy [15]. Scorpion venom and bradykinin, in a dose response manner showed calcium to increase in associated neurones invariably before it increased in the associated RBL cell. Neither scorpion venom or bradykinin had any effect on RBL alone. The most important communication seemed to occur via SP since a neutralizing antibody to it as well as an NK-1 receptor antagonist (CP90-994-1) but not an NK2 antagonist (SR48968), interrupted this direct neurite mast cell communication. An antibody to the IgE receptor caused an appropriate lag response with an increase in calcium in the associated neurite. Previous work by Cooke had shown the presence of NK-1 receptors on RBL [16].
197
While substance P has generally been shown to have some physiological effect on mast cells, its role has been questioned since the concentrations required to degranulate the mast cell are relatively high, i.e., 10 -5 M. We have shown in patch clamp studies of mast cells and RBL cells that stimulation with very low concentrations of substance P (picomolar) have been able to promote whole cell current oscillations and even degranulation after a very significant and prolonged lag period [17]. The lag period was proportionate to the concentrations used, i.e., >25 minutes in the case of repeated 5 pM concentrations of substance P. Similar results were obtained by initial priming of mast cells with 5 pM substance P followed by subthreshold amounts of anti-IgE. These experiments revealed that the effects described were not just in vitro artifacts with neuropeptides, but had some biological and physiological meaning.
3.
NERVE GROWTH FACTOR
Nerve growth factor is involved in neuroimmune interactions in a variety of ways and I will review some of these briefly. N G F is found in secretions, especially those in the male submandibular gland but probably in all secretions [18]. It is essential for the growth, survival and differentiation of sympathetic sensory afferent and some cortical neurons. The predominant message for its up regulation is IL-1. It is made by many different cell types which include structural cells in the nervous system, a variety of immune cells which include mast cells [19], lymphocytes (especially TH2) [20], and eosinophils [21]. N G F has been shown to be synthesized by fibroblasts, keratinocytes [22] as well as constitutively by various epithelial cells.The effects of nerve growth factor are pleiotropic (Table I) and they range from synergy with G M - C S F and IL-5 in terms of promotion of human stem cell colony growth [23, 24], to the priming of basophils and mast cells for subsequent degranulation by mast cell secretogogues [25, 26]. In these latter experiments N G F was shown to be a significant molecule involved in priming of cells for subsequent agonist activity in the formation and release of leukotrienes and histamine. In addition, N G F has an effect on B-cells [27], promotes synthesis by human B-cells of IgG4 [28], promotes wound healing [29], repair of human corneal ulceration [30], and prevents mast cell, eosinophil and neutrophil apoptosis [31-33]. Table I
9 9 9 9 9 ~ 9 9
Non-neuronaleffects of nerve growth factor.
Promoteswound healing, [58], prevents carrageenin-induced inflammation [61] Heals corneal ulcers [75] Promotesgrowth, differentiation, proliferation and survival of B cells [27, 51]. EnhancesIgG4 production [28]. Prevents apoptosis in mast cells, eosinophils and neutrophils [31-33]. Strongmastopoietic effect in rodents, partly through mast cell degranulation [34, 52]. Primes basophils and mast cells for agonist action (e.g., C5a) [25, 26]. Enhances human haematopoietic colony growth of basophils and eosinophils; synergistic with GM-CSF and IL-5 [23, 24].
N G F has been shown to be one of the most potent mastopoietic substances known. Injections of small amounts of NGF, especially in the neonatal period, promote significant increases in
198
mast cells in peripheral tissues [34]. The effects are in part due to mast cell degranulation, since treatment of animals with disodium cromoglycate which stabilizes connective tissue mast cells, prevented this increase, while leaving unaffected the increase in mucosal mast cells. Interestingly, NGF causes upregulation of substance P message and synthesis in autonomic ganglia [35] and causes airway hyper-reactivity in a variety of experimental models. These include transgenic mice which over-express NGF in the lung [36]. In human nasal secretions NGF has been shown to be released upon specific antigen challenge in perennial rhinitis patients [37] and surprisingly, antibody to NGF to abrogate at least one model of experimental asthma and the brochial hyperreactivity which occurs in it [38]. We have shown that NGF causes an increase in IL-6 production by rat peritoneal mast cells at the same time as it inhibits TNFc~ synthesis and release [39]. This effect appears to be mediated via PGE2 through autocoid release since indomethacin treated peritoneal mast cells failed to respond in the fashion described. The response was restored by the addition of PGE2 to indomethacin treated mast cells. These results offer a possible explanation for why NGF may have protective (anti-inflammatory) activities (Table II) since the inhibition of TNF~t synthesis and secretion would be expected to have decidedly anti-inflammatory effects through inhibition of its known pro-inflammatory biological role such as the promotion of neutrophil influx into inflammatory sites. It is possible that this may also be responsible for the reduced inflammatory response of animals treated with NGF in the experimental model of autoimmune encephalomyelitis [40]. Table II
Protective effects of nerve growth factor.
9 Promotes wound healing [29]. 9 Prevents carrageenan inflammation [53]. 9 Repairs human corneal ulceration [30]. 9 Inhibits intestinal inflammation in hapten induced colitis [50]. 9 Inhibits autoimmune encephalomyelitis [40].
One of the many examples of neuroimmune interactions which appears to involve the mast cell-nerve physiological unit is that of stress. It has been well established now that stress effects cause dura mater intracerebral mast cell degranulation through the release and direct action of corticotrophin releasing factor (CRF) [41 ]. This molecule is essential for the signal to the pituitary gland to synthesize and release ACTH, and subsequently influence the up regulation and discharge of cortisone from the adrenal cortex. CRF causes mast cell degranulation in vivo and in vitro [41-43]. Both acute and chronic stress causes mast cell activation and degranulation in intestinal tissue and the effect on mucus discharge by goblet cells, as well as the changes in motility and increase in chloride ion secretion by the epithelium appear all to be mediated through this interaction [43-46]. The intestinal effects of stress can be interrupted by pharmacological blockade of a variety of autonomic and ganglionic nerves through the use of bretylium, atropine and hexamethonium [47]. The whole process can also be interrupted by NK1 antagonists which block the effect of substance P, and also by blockade of neurotensin [43, 45]. Finally, the intracerebral or peripheral injection of CRF can mimic most of the effects of stress on the intestine [43]. These complex interactions affect epithelial function, barrier integrity and smooth muscle motility and involve mast cells and nerves. We wished to test
199
whether NGF was involved in any of these activities and chose to look at epithelial function using the Ussing chamber as a read-out. In brief, the injection of anti-NGF before and at the time of acute stress caused a significant change (increase) of short circuit current in the colon of Wistar Kyoto animals. This preliminary observation suggests a role for NGF in stabilizing or protecting the epithelium and may likely fit the pattern of protection and anti-inflammatory effects which were alluded to earlier. Most recently, we have examined intestinal epithelial cells for their constitutive production of NGF. We have shown that human intestinal epithelial T84 cells have a low constitutive NGF expression, but message for NGF is markedly up regulated if the cells are incubated for 1 hour with physiologic concentrations of human recombinant IL-10. Furthermore, the message is translated and T84 cells synthesize NGF protein. This action of IL-10 is reciprocated since 10 or 100 ng of NGF markedly and selectively up regulated IL-10 synthesis, and again, IL-10 protein was increased in the cells subsequently. Thus, there appears to be a reciprocal autocoid up regulation of IL-10 by NGF, and in turn of NGF by IL-10 in this intestinal epithelial cell line. These actions appear to be very selective since no effects by these molecules on other cytokines have been seen with respect to TGF[3, IL-6, IL-8 or themselves in the test systems employed. Interestingly, Brodie et al., [48] have shown that IL-10 similarly causes astrocytes to increase their synthesis of NGF. These experiments promote further support for the protective effects of NGF in addition to giving an additional emphasis to the possible role of IL-10 in providing protection to the intestine. It has been known for some time that transgenic IL-10 knockout mice develop spontaneously a form of colitis that is caused by conventional intestinal flora. Colonization of IL-10 KO mice with lactococci which synthesize IL-10 prevents the colitis [49]. Furthermore, in a model of hapten induced colitis, treatment of mice with anti-NGF strikingly increased the inflammation induced by challenge with the hapten, confirming the likely protective effect of NGF in this model [50]. The whole way in which the brain and the nervous system interact, regulate and in turn are regulated by the cells of the immune and inflammatory systems is being slowly defined. Perhaps the best example is one given in terms of stress where even injection of CRF into the brain of experimental rats can mimic the peripheral effects of stress in animals. It can reasonably be expected that we will begin to have the same sense of molecular events and an understanding of them in many, if not all, physiologic and disease conditions and in this manner, may manage to offer new alternative therapeutic possibilities where they are needed.
REFERENCES 1. 2.
3. 4. 5. 6.
Blalock JE. The immune system as a sensory organ. J Immunol 1984; 132: 1067-1070. Stead RH, Tomioka M, Quinonez G, Simon GT, Felten SY, Bienenstock J. Intestinal mucosal mast cells in normal and nematode-infected rat intestines are in intimate contact with peptidergic nerves. Proc Natl Acad Sci USA 1987; 84: 2975-2979. McKay DM, B ienenstock J. The interaction between mast cells and nerves in the gastrointestinal tract. Immunol Today 1994; 15: 533-538. Theoharides TC, Pang X, Letourneau R, Sant GR. Interstitial cystitis: a neuroimmunoendocrine disorder. Ann NY Acad Sci 1998; 840: 619-634. Baird AW, Cuthbert AW. Neuronal involvement in type 1 hypersensitivity reactions in gut epithelia. Br J Pharmacol 1987; 92: 647-655. Cooke HJ. Neuroimmune signaling in regulation of intestinal ion transport. Am J Physiol 1994; 266: G167-G178.
200
10.
11. 12.
13. 14. 15.
16. 17.
18. 19.
20.
21.
22.
23.
24.
Perdue MH, Masson S, Wershil BK, Galli SJ. Role of mast cells in ion transport abnormalities associated with intestinal anaphylaxis. Correction of the diminished secretory response in genetically mast cell-deficient W/Wv mice by bone marrow transplantation. J Clin Invest 1991; 87: 687-693. Perdue MH, McKay DM. 1994. Integrative immunophysiology in the intestinal mucosa. Am J Physiol 1994; 267: G 151-G 165. Bienenstock J. Cellular communication networks. Implications for our understanding of gastrointestinal physiology. Ann NY Acad Sci 1992; 664: 1-9. Sestini P, Bienenstock J, Crowe SE, Marshall JS, Stead RH, Kakuta Y, Perdue MH. Ion transport in rat tracheal epithelium in vitro. Role of capsaicin-sensitive nerves in allergic reactions. Am Rev Respir Dis 1990; 141 : 393-397. Blennerhassett MG, Bienenstock J. Sympathetic nerve contact causes maturation of mast cells in vitro. J Neurobiol 1998; 35: 173-182. Blennerhassett MG, Janiszewski J, Bienenstock J. Sympathetic nerve contact alters membrane resistance of cells of the RBL-2H3 mucosal mast cell line. Am J Respir Cell Mol Biol 1992; 6: 504-509. Blennerhassett MG, Tomioka M, Bienenstock J. Formation of contacts between mast cells and sympathetic neurons in vitro. Cell Tissue Res 1991 ; 265:121-128. Blennerhassett MG, Bienenstock J. Apparent innervation of rat basophilic leukaemia (RBL-2H3) cells by sympathetic neurons in vitro. Neurosci Lett 1990; 120: 50-54. Suzuki R, Furuno T, McKay DM, Wolvers D, Teshima R, Nakanishi M, Bienenstock J. Direct neurite-mast celI(RBL) communication in vitro occurs via the neuropeptide substance P. J Immunol 1999; 163:2410-2415. Cooke HJ, Fox P, Alferes L, Fox CC, Wolfe SAJ. Presence of NK1 receptors on a mucosal-like mast cell line, RBL-2H3 cells. Can J Physiol Pharmacol 1998; 76: 188-193. Janiszewski J, Bienenstock J, Blennerhassett MG. Picomolar doses of substance P trigger electrical responses in mast cells without degranulation. Am J Physiol 1994; 267: C 138-C 145. Bienenstock J. Mast cell-nerve interactions: possible significance of nerve growth factor. In Mast cells and basophils. G Marone, editor. Academic Press, 2000; 313-323. Leon A, Buriani A, Dal TR, Fabris M, Romanello S, Aloe L, Levi-Montalcini R. Mast cells synthesize, store, and release nerve growth factor. Proc Natl Acad Sci USA 1994; 91: 3739-3743. Lambiase A, Bracci-Laudiero L, Bonini S, Starace G, D'Elios MM, De CM, Aloe L. Human CD4+ T cell clones produce and release nerve growth factor and express highaffinity nerve growth factor receptors. Allergy Clin Immunol 1997; 100: 408-414. Solomon A, Aloe L, Pe'er J, Frucht-Pery J, Bonini S, Levi-Schaffer F. Nerve growth factor is preformed in and activates human peripheral blood eosinophils. J Allergy Clin Immunol 1998; 102: 454-460. Tron VA, Coughlin MD, Jang DE, Stanisz J, Sauder DN. Expression and modulation of nerve growth factor in murine keratinocytes (PAM 212). J Clin Invest 1990; 85: 1085-1089. Matsuda H, Coughlin MD, B ienenstock J, Denburg JA. Nerve growth factor promotes human hemopoietic colony growth and differentiation. Proc Natl Acad Sci USA 1988; 85:6508-6512. Tsuda T, Wong D, Dolovich J, Bienenstock J, Marshall J, Denburg JA. Synergistic effects of nerve growth factor and granulocyte-macrophage colony-stimulating factor on human basophilic cell differentiation. Blood 1991; 77: 971-979.
201
25. 26. 27.
28. 29.
30.
31.
32. 33. 34.
35. 36.
37.
38.
39.
40.
41.
Bischoff SC, Dahinden CA. c-kit Ligand: A Unique Potentiator of Mediator Release by Human Lung Mast Cells. J Exp Med 1992; 175: 237-244. B ischoff SC, Dahinden CA. Effect of nerve growth factor on the release of inflammatory mediators by mature human basophils. Blood 1992; 79: 2662-2669. Torcia M, Bracci-Laudiero L, Lucibello M, Nencioni L, Labardi D, Rubartelli A, Cozzolino F, Aloe L, Garaci E. Nerve growth factor is an autocrine survival factor for memory B lymphocytes. Cell 1996; 85: 345-356. Kimata H, Yoshida A, Ishioka C, Kusunoki T, Hosoi S, Mikawa H. Nerve growth factor specifically induces human IgG4 production. Eur J Immunol 1991; 21: 137-141. Matsuda H, Koyama H, Sato H, Sawada J, Itakura A, Tanaka A, Matsumoto M, Konno K, Ushio H, Matsuda K. Role of nerve growth factor in cutaneous wound healing: accelerating effects in normal and healing-impaired diabetic mice. J Exp Med 1998; 187: 297-306. Lambiase A, Rama P, Bonini S, Caprioglio G, Aloe L. Topical treatment with nerve growth factor for corneal neurotrophic ulcers [see comments]. N Engl J Med 1998; 338: 1174-1180. Kawamoto K, Okada T, Kannan Y, Ushio H, Matsumoto M, Matsuda H. Nerve growth factor prevents apoptosis of rat peritoneal mast cells through the trk proto-oncogene receptor. Blood 1995; 86: 4638-4644. Kannan Y, Usami K, Okada M, Shimizu S, Matsuda H. Nerve growth factor suppresses apoptosis of murine neutrophils. Biochem Biophys Res Commun 1992; 186: 1050-1056. Hamada A, Watanabe N, Ohtomo H, Matsuda H. Nerve growth factor enhances survival and cytotoxic activity of human eosinophils. Br J Haematol 1996; 93: 299-302. Marshall JS, Stead RH, McSharry C, Nielsen L, Bienenstock J. The role of mast cell degranulation products in mast cell hyperplasia. I. Mechanism of action of nerve growth factor. J Immunol 1990; 144:1886-1892. Hart RP, Shadiack AM, Jonakait GM. Substance P gene expression is regulated by interleukin- 1 in cultured sympathetic ganglia. J Neurosci Res 1991; 29:282-291. Hoyle GW, Graham RM, Finkelstein JB, Nguyen KP, Gozal D, Friedman M. Hyperinnervation of the airways in transgenic mice overexpressing nerve growth factor. Am J Respir Cell Mol Biol 1998; 18: 149-157. Sanico AM, Stanisz AM, Gleeson TD, Bora S, Proud D, Bienenstock J, Koliatsos VE, Togias A. Nerve growth factor expression and release in allergic inflammatory disease of the upper airways. Am J Respir Crit Care Med 2000; 161: 1631-1635. Braun A, Appel E, Baruch R, Herz U, Botchkarev V, Paus R, Brodie C, Renz H. Role of nerve growth factor in a mouse model of allergic airway inflammation and asthma. Eur J Immunol 1998; 28:3240-3251. Marshall JS, Gomi K, Blennerhassett MG, Bienenstock J. Nerve growth factor modifies the expression of inflammatory cytokines by mast cells via a prostanoid-dependent mechanism. J Immunol 1999; 162: 4271-4276. Villoslada P, Hauser SL, Bartke I, Unger J, Heald N, Rosenberg D, Cheung SW, Mobley WC, Fisher S, Genain CP. Human nerve growth factor protects common marmosets against autoimmune encephalomyelitis by switching the balance of T helper cell type 1 and 2 cytokines within the central nervous system [see comments]. J Exp Med 2000; 191: 1799-1806. Theoharides TC, Spanos C, Pang X, Alferes L, Ligris K, Letourneau R, Rozniecki JJ, Webster E, Chrousos GP. Stress-induced intracranial mast cell degranulation: a corticotropin-releasing hormone-mediated effect. Endocrinology 1995; 136: 5745-5750.
202
42.
43.
44.
45.
46.
47.
48. 49.
50.
51. 52. 53.
Theoharides TC, Singh LK, Boucher W, Pang W, Letourneau R, Webster E, Chrousos G. Corticotropin-releasing hormone induces skin mast cell degranulation and increased vascular permeability, a possible explanation for its proinflammatory effects. Endocrinology 1998; 139: 403-413. Castagliuolo I, LaMont JT, Qiu B, Fleming SM, Bhaskar KR, Nikulasson ST, Kornetsky C, Pothoulakis C. Acute stress causes mucin release from rat colon: role of corticotropin releasing factor and mast cells. Am J Physiol 1996; 271: G884-G892. Castagliuolo I, Wershil B K, Karalis K, Pasha A, Nikulasson ST, Pothoulakis C. Colonic mucin release in response to immobilization stress is mast cell dependent. Am J Physiol 1998; 274:G1094-G1100. Castagliuolo I, Leeman SE, Bartolak-Suki E, Nikulasson S, Qiu B, Carraway RE, Pothoulakis C. A neurotensin antagonist, SR 48692, inhibits colonic responses to immobilization stress in rats. Proc Natl Acad Sci USA 1996; 93: 12611-12615. Santos J, Benjamin M, Yang PC, Prior T, Perdue MH. Chronic stress impairs rat growth and jejunal epithelial barrier function: role of mast cells. Am J Gastrointest. Liver Physiol 2000; 278: G847-G854. Santos J, Saunders PR, Hanssen NP, Yang PC, Yates D, Groot JA, Perdue MH. Corticotropin-releasing hormone mimics stress-induced colonic epithelial pathophysiology in the rat. Am J Physiol 1999; 277:G391-G399. Brodie C. Differential effects of Thl and Th2 derived cytokines on NGF synthesis by mouse astrocytes. FEBS Lett 1996; 394:117-120. Madsen KL, Doyle JS, Jewell LD, Tavernini MM, Fedorak RN. Lactobacillus species prevents colitis in interleukin 10 gene-deficient mice [see comments]. Gastroenterology 1999; 116:1107-1114. Reinshagen M, Rohm H, Steinkamp M, Lieb K, Geerling I, Von Herbay A, Flamig G, Eysselein VE, Adler G. Protective role of neurotrophins in experimental inflammation of the rat gut. Gastroenterology 2000; 119: 368-376. Otten U, Ehrhard P, Peck R. Nerve growth factor induces growth and differentiation of human B lymphocytes. Proc Natl Acad Sci USA 1989; 86: 10059-10063. Aloe L, Levi-Montalcini R. Mast cells increase in tissues of neonatal rats injected with the nerve growth factor. Brain Res 1977; 133: 358-366. Banks BE, Vernon CA, Warner JA. Nerve growth factor has anti-inflammatory activity in the rat hindpaw oedema test. Neurosci Lett 1984; 47: 41-45.
New Foundationof Biology
203
Edited by I. Berczi and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
Immunomodulation by the Submandibular Gland
PAUL FORSYTHE 1, RENE E. DIARY1, RONALD MATHISON 2, JOSEPH S. DAVISON 2 and A. DEAN BEFUS 1
1pulmonary Research Group, Department of Medicine, Faculty of Medicine, University of Alberta, Edmonton, Alberta, Canada, T6G 2S2 2Department of Physiology and Biophysics, Faculty of Medicine, University of Calgary, Calgary, Alberta, Canada, T2N-4N1
ABSTRACT We have established that decentralization (cutting sympathetic nerve trunk) of the superior cervical ganglia bilaterally reduces the magnitude of allergic inflammation in the airways of rats. The magnitude of anaphylactic and endotoxic hypotension, and of gastrointestinal inflammation was also reduced. This anti-inflammatory activity was dependent upon intact submandibular glands. Reconstitution of sialadenectomized (removal of submandibular glands) rats with soluble extracts of the submandibular glands identified two polypeptides with antiinflammatory activities. The sequences of these polypeptides were found within a prohormone, submandibular gland rat 1 (SMR1). The C-terminal peptide TDIFEGG, has been studied extensively. Using sequential amino acid substitutions and systematic removal of C terminal or N terminal amino acids, we established that the tripeptide FEG is biologically active. Modification of FEG to the D-isomeric feG, enhances its activity in some assay systems. We postulated that feG would inhibit airways inflammation, and tested this using a model of allergic asthma, namely the Brown Norway rat sensitized to ovalbumin (OA). Sensitized rats were challenged 14 to 21 days later with aerosolized OA. This challenge markedly increased numbers of inflammatory cells recovered from the airways after 24 hour (29 x 106, n = 23) compared to saline controls (1 x 106, n = 4). The infiltrating cells included macrophages (10 x 106), neutrophils (9 x 106) and eosinophils (9 x 106). Intravenous (0.25 mg/kg) or oral feG (1 mg/kg) given 30 min prior to OA significantly inhibited influx of inflammatory cells by 50 to 70%. feG reduced inflammatory cell infiltration when given 30 min before to 3 to 6 hour post allergen exposure. Oral feG reduced the numbers of macrophages, neutrophils and eosinophils. One of the mechanisms underlying the effects of feG may be its ability to inhibit PAF-induced expression of CD1 l b on purified human neutrophils. It is possible that feG may be useful in the treatment of allergic asthma, given either as an oral prophylactic, or as a post exposure therapeutic.
204
The cervical sympathetic nerve trunk-submandibular gland axis of neuroendocrine regulation of inflammation may be dysfunctional in inflammatory diseases and provide opportunities for new therapeutic intervention. This axis may be sensitive to modulation by central and peripheral neural mechanisms that influence its function/dysfunction. Supported by MRC, Salpep Biotechnology Inc and Heart & Stroke Fdn of Canada.
1.
INTRODUCTION
To most, the salivary glands are simply exocrine organs concerned with aiding digestion and maintaining oral health. This view is held despite the fact that endocrine secretion from salivary glands was first reported almost 50 years ago. Since then it has emerged that salivary gland factors aid in the maintenance and integrity of the esophageal and gastrointestinal mucosa, promote hepatic regeneration and mammary gland tumorigenesis, are essential for the maintenance of the reproductive system and have regulatory effects on the immune system. Similarily, while regulation of salivary gland function by the autonomic nervous system has been well known since Pavlov's classical experiments, the significance of the cervical sympathetic trunk-submandibular gland (CST-SMG) axis as an effector of neuroendocrine-immune regulation has only recently come to light [1]. What follows is a description of the immunoregulatory activities of the submandibular gland (SMG), the endocrine factors responsible for these effects and mechanisms that control the production and release of such factors. The potential implications of the CST-SMG axis in neuroendocrine immunology and psychoneuroimmunology are also discussed.
2.
STRUCTURE OF THE SUBMANDIBULAR GLAND
Structurally the SMG comprises four major epithelial compartments: acinar cells, intercalated ducts, granular convoluted tubule (GCT) cells and striated excretory ducts. The acini are connected to each other by intercalated ducts that lead to the GCT, which in turn join into the striated secretory ducts [2] (Figure 1). The acinar cells secrete the important digestive enzyme amylase, and produce saliva. The intercalated duct cells include stem cells that produce acinar and GCT cells during development. The cells of the striated excretory ducts regulate water content and ionic composition of saliva. GCT cells represent a major source of bioactive polypeptides produced in the SMG [3]. The potential importance of GCT cells in homeostatic mechanisms is reflected in the number of hormonal systems that exert control over their development and content of biologically active polypeptides.
3.
BIOLOGICALLY ACTIVE POLYPEPTIDES
A large number of biologically active polypeptides have been identified in the SMG many of which have been localized to the GCT [3, 4]. These factors, released as both exocrine and endocrine agents, can be classified into three groups: a) growth factors such as nerve growth factor (NGF), epidermal growth factor (EGF) and transforming growth factor-J3 (TGF-[3), b) processing enzymes such as kallikrein-like proteinases and renin and c) regulatory peptides including glucagon, insulin, erythropoietin, somatostatin, angiotensin II,
205
Acinar cells Intercalated ducts
i
Granulated Convoluted tubule
i
1
L
o 0
0 0
0
0
0
~
o o _o
0
0
o
0 010 10101010101 101010101011
1010101010101q 01010101010101
Striated excretory duct
Figure 1. A schematic representation of the relationship between the four major epithelial compartments of the submandibular gland.
vasoactive intestinal peptide and neuropeptide Y. In the following sections a number of wellcharacterized SMG derived peptides with biological functions relevant to immunoregulation will be described. 3.1.
Nerve growth factor
Nerve growth factor was first described in the mouse SMG and identified as a neurotrophic agent. Subsequent investigations have revealed properties of NGF that suggest it may play a role in immunoregulation [5, 6]. NGF increases the number and size of mast cells in tissues of neonatal mice [7], and stimulates histamine release from mast cells, both in vivo and in vitro [8-10]. NGF also increases phagocytosis and chemotaxis of neutrophils, promotes the development of hemopoietic colonies and stimulates lymphocyte growth in vitro [11-15]. Other effects include the up-regulation of IgM and IgG 4 production by human B cells [ 16, 17]. NGF may also aid in neuronal control of the immune system as it stimulates the growth of sympathetic ganglia that innervate immune organs [5]. While these effects suggest a proinflammatory role for NGF the mediator has been shown to suppress in vivo inflammatory reactions in several models [ 14]. The reason for this apparent paradox is, as yet, unknown. 3.2.
Transforming growth factor [5
Transforming growth factor [5 (TGF-[3) is a multifunctional regulator of cell growth and differentiation in a wide variety of normal and neoplastic systems [18]. A homodimeric
206
polypeptide originally purified from human platelets, TGF-[5 shares structural and functional homology with epidermal growth factor (EGF). These factors share a receptor [19]. Binding of TGF-[5 or EGF to the 170 kDa plasma membrane receptor on the target cell stimulates intrinsic tyrosine kinase activity. The subsequent phosphorylation cascade leads to increased proliferation and differentiation of skin tissues, corneal epithelium, lung and tracheal epithelium [20-22]. TGF-[5 exerts a range of effects on inflammatory and immune responses, acting as a chemoattractant for monocytes, neutrophils and lymphocytes and activating monocytes to secrete cytokines and growth factors [23-25]. TGF-[5 is a stimulatory factor in the early stages of inflammation but later supports its resolution and contributes to healing. CD8+ T cells are stimulated, while activated CD4+ T cells are suppressed [26]. B cell proliferation and the production of IgG and IgM are also suppressed [27]. 3.3.
Epidermal growth factor
One of the most widely studied aspects of SMG function is the exocrine release of epidermal growth factor (EGF) and its regenerative and reparative properties on oral, oesophageal and gastric mucosa. The importance of EGF in tissue repair is emphasized by its diverse effects that include gastric epithelial cell division, DNA synthesis, collagen synthesis, matrix deposition, neovascularization and stimulation of protein and hyaluronic acid synthesis in epithelial cells [28]. Through these multiple activities SMG-derived EGF promotes the healing of gastric ulcers and tongue lesions and maintains the integrity of oesophageal mucosa [29, 30]. EGF can also be released into the blood stream and its actions extend beyond the oral and gastrointestinal mucosa to other organs. EGF enhances liver regeneration after partial hepatectomy and is important in maintaining uterine growth and fertility [31, 32]. Immunoregulatory functions of EGF include stimulation of T cell proliferation, up-regulation of interferon-~, production and reduction of T suppressor cell activity [33-35]. EGF also stimulates macrophage chemotaxis and phagocytosis [36, 37]. These actions suggest that EGF stimulates the immune system and is proinflammatory. However, as with NGF, immunosuppressive activity of EGF has been reported [38]. 3.4.
Proteases and convertases
The polypeptides produced by the SMG are synthesized as inactive precursors that have the potential to become active following proteolytic processing. The submandibular glands contain an array of enzymes capable of hydrolysing peptide bonds including members of the prohormone convertase and kallikrein families [3, 39]. These proteases act to modulate the biological activities of growth factors and regulatory peptides. In addition to being vital for the production of biologically active polypeptides it is also apparent that at least one family of proteases found in the SMG may be directly involved in the modulation of immune responses. These enzymes, the kallikreins, will be considered here in more detail. 3.5.
Kallikreins
The SMG of the rat contains large amounts of kallikrein, a serine protease with the ability to release the vasodilatory peptide kallidin (Lys-bradykinin) from the plasma protein kininogen. Kallikrein is secreted into the saliva and blood. In the mouse, 14 genes have been confirmed to have the potential to encode functional kallikrein proteins. At least three of these are EGF binding proteins and can cleave EGF to its active form [40].
207
The rat SMG expresses at least 6 kallikrein genes [41]. Glandular or tissue kallikreins differ in molecular weight and enzymatic activity from plasma kallikreins. Mouse glandular kallikrein strongly enhances spontaneous and mitogen induced proliferation of lymphocytes [42]. This function is independent of EGF as non-EGF binding forms of kallikrein also exhibit this activity. Serine proteinase inhibitors can block the response suggesting it is related to enzymatic activity of kallikreins. This is in accordance with reports that trypsin, chymotrypsin, thrombin and other proteinases mitogenically stimulate many kinds of cells including lymphocytes. In vitro the addition of kallikrein and other serine proteases to B cells stimulated with LPS and IL-4 enhances the production of IgE, IgG 1 and IgG 3 [43]. Rat glandular kallikrein has also been reported to suppress the DTH response to picryl chloride in mice [44].
4.
HORMONAL CONTROL OF THE SMG
It has been clearly demonstrated that androgens, thyroid and adrenocortical hormones are necessary for the normal development of the SMG and for the production of biologically active polypeptides [45-47]. Perhaps the most striking aspect of hormonal control of the SMG is the sexual dimorphism of the organ. This is reflected in the androgen-induced increase in the number of GCT cells and content of biologically active peptides [4]. Assays have revealed much greater concentrations of many polypeptides (e.g. EGF, NGF, kallikrein and renin) in the glands of male compared to female rodents [48-50]. The levels of these factors are androgen dependent. The NGF content of the SMG is reduced by castration and increased by administration of testosterone to castrated male or female animals. NGF levels also increase during pregnancy and lactation [51]. Similar observations have been made regarding EGF levels in the SMG. The gland of the female contains about 1/10 the EGF detected in the male. Administration of testosterone to both male and female animals leads to complete reversal of the drastic reduction in EGF levels observed following removal of the pituitary gland (hypophysectomy) [52]. Androgen receptors have been detected in mouse and rat SMG. Specific androgen binding is higher in homogenates of female SMG compared to males in both rats and mice [53, 54] despite the fact that glands of male mice contain three times as much total androgen receptor capacity as those of females. The reason for this apparent conflict was determined by Kyakamoto et al. They showed that in males 74% of receptor capacity was in nuclei (occupied) while in females 94% was in the cytosol (unoccupied). Castration results in a female distribution of the receptor that can be reversed by testosterone administration [53]. There is a significant decrease in plasma lutenizing hormone (LH) levels following sialectomy but this procedure does not significantly modify hypophyseal LH content [55]. The decrease in plasma LH causes changes in Leydig cells and reduces testosterone production. This observation indicates that while testosterone plays a crucial role in maintenance of the SMG there is also a feedback relationship between the SMG and the testis via hypophyseal LH secretion. Several studies have established that in addition to androgens, thyroid hormones are important in the development and maintenance of the GCT cells [4]. Aloe and Levi-Montalcini reported that thyroxine caused a precocious differentiation of cells [56]. Thyroid hormones have also been demonstrated to increase the levels of NGF and EGF in the SMG of the adult female mouse [57-61]. The SMG of Tfm/Y mice lacks specific androgen binding capacity [59] and subsequently these mice are deficient in NGF and EGF [62]. Administration of thyroxine but not testosterone greatly increases the level of both growth factors in the SMG of Tfm/Y mice. Thyroxine also restores EGF levels to normal in hypothyroid animals.
208
Hypophysectomy induces marked atrophy of the SMG and reduces both in vitro lymphocytestimulating activity and in vivo immunosuppressive activity. The important role of thyroid hormones and androgens in maintaining SMG function is further emphasized by the fact that thyroid stimulating hormone (TSH) and LH restore the ability of the SMG to modulate lymphocytes in hypophysectomized rats. Prolactin was also required for full restoration [63]. Feedback between the SMG and the hypothalamus-pituitary-adrenal (HPA) axis has been suggested, based on observations that NGF and EGF are able to stimulate release of ACTH and glucocorticoids [64, 65] while, in contrast, TGF-13 depresses acetylcholine induced CRH release from the hypothalamus [66]. These demonstrations of SMG dependence on pituitary function and the feedback relationships with other glands indicate that the SMG is a fully integrated component of the body's endocrine system (Figure 2). Stimulation -- i
- Inhibition
~hypo~t~alamu.q~ Cot,..
Exogenous LH
Immunomodulation
Exogenous TSH
Figure 2. The SMG as a fully integrated component of the endocrine system (see text for details).
5.
AUTONOMIC CONTROL OF THE SMG
In addition to endocrine control, the autonomic nervous system also participates in homeostatic responses to inflammation through the regulation of cardiovascular function and stimulation of release of certain stress hormones such as glucagon and renin. Given the role of the SMG in controlling inflammation it is not surprising that it is also under the control of the autonomic nervous system (ANS). It has been known for some time that the size of SMG is influenced by the ANS and that in adult rats prolonged electrical stimulation of the sympathetic branch of the ANS via the superior cervical ganglion (SCG) causes enlargement of SMG by increasing
209
both cell size and number [67]. Similarily the [3-adrenoreceptor agonist isoprenaline (IPR) induces hypertrophic and hyperplastic enlargements of rodent SMG and induces the expression of a number of genes. The SMG receives both sympathetic and parasympathetic innervation. Both branches regulate the volume and composition of saliva. The parasympathetic system stimulates the glandular acinus leading to the secretion of large volumes of saliva with low concentrations of biologically active polypeptides. [3-adrenergic stimulation increases the synthesis and release of these polypeptides, c~-adrenergic agents stimulate the secretion of growth factors and homeostatic proteases by the GCT cells and lead to the appearance of large amounts of kallikrein NGF, EGF and renin in the saliva, whereas [3-adrenergic agents exert a significantly lesser effect [68, 69]. The parasympathetic and sympathetic nerves differentially regulate the compartments into which kallikrein is secreted. Local parasympathetic nerve stimulation selectively releases kallikrein into the saliva (exocrine secretion) while activation of the sympathetic nerves increases kallikrein levels in the saliva and the blood through a process mediated by adrenergic receptors [70]. The kallikreins released from the SMG exert both local and distant effects in that they dilate veins of the glands as well as decrease systemic blood pressure in response to heat stress. The release of EGF is stimulated by both ct and [3 adrenergic mechanisms, although the ct-adrenergic response is more intense and prolonged. The sympathetic nerves also regulate gene expression for NGF and EGF [1, 2]. The increase in the release of kallikrein and EGF from SMG upon activation of the sympathetic nerves, suggests that autonomic regulation of inflammatory responses includes stimulation of the release of glandular factors involved in cardiovascular control and tissue repair. Cystatin S is a cysteine proteinase inhibitor that regulates proteolysis by endogenous and/or exogenous cysteine proteases such as the cathepsins. Rat cystatin S gene expression is tissue specific and occurs temporally during normal development [71]. The steady state level reaches a maximum level at 28 days of age and is not observed in the adult animal reaching barely detectable levels at 32 days of age. However, the cystatin S gene can be induced in the adult SMG by IPR [71, 72]. Data suggests that expression of the rat cystatin S gene is also controlled by tissue specific factors. This is reflected in a much greater IPR induced increase of cystatin S mRNA in the SMG compared to the parotid gland. Induction of cystatin mRNA is also more pronounced in the SMG of female compared to male rats. This difference is evident in 15 day old animals and is therefore thought to be linked to gender genotype rather than to circulating levels of steroid hormones [71 ]. Autonomic regulation of protease inhibitor levels offers another potential control mechanism for the production of biologically active polypeptides in the SMG. Changes in the ratios of polypeptides, proteases and protease inhibitors following hormonal or neuronal stimulation of the SMG have yet to be assessed. However, such investigation may provide important insights into, what is undoubtedly, a complex regulatory system (Figure 3).
6.
THE CERVICAL SYMPATHETIC TRUNK-SUBMANDIBULAR GLAND AXIS
Axons project down the cervical sympathetic trunk to the inferior and superior cervical ganglia (SCG). The postganglionic neurons leaving the inferior cervical ganglia predominately innervate the lungs and heart while axons leaving the SCG provide sympathetic innervation to the upper thorax, neck and skull as well as to facial structures [73]. The number of endocrine organs found in these areas, including the pineal, thyroid, parathyroid and salivary glands is an indication of the relevance of the SCG to the neuroendocrine system.
210
Physiological- Pathological- Psychological Stimuli
Central Processing
Sympathetic _ / ~L
SMG ProteasesCm=~Protease/ ~-~ inhibi7 Regulatory/ Bloodstreamrelease
Immunomodulation Figure 3. Mechanisms controlling bioactive peptide release from the SMG (see text for details).
Surgical sympathectomy has been used as an approach to study neural regulation of the immune system. These surgical denervations involve either SCG ganglionectomy (SCGx) or severing the connection between the inferior and superior cervical ganglia, so called decentralization. Unilateral removal of the SCG enhances contact hypersensitivity and delayed type reactions in the denervated submandibular lymph nodes [74]. These altered responses of the immune system indicate a direct modulation of inflammatory events by the sympathetic nervous system. 6.1.
A role for the CST-SMG axis in anaphlaxtic and endotoxic reactions
Rats sensitized by infection with the nematode Nippostrongylus brasiliensis have been used to investigate the modulatory role of the sympathetic nervous system in pulmonary inflammation. Within 8 hour of intravenous challenge with sensitizing allergen these rats develop a pronounced influx of macrophages and neutrophils into the lumen of the airway [75]. The cellular responses
211
can be used as a read out to determine the effects of surgical SCGx and decentralization. We have shown these surgical interventions have dramatic anti-inflammatory effects on lifethreatening anaphylaxis and subsequent pulmonary inflammation when compared to sham operated animals [75]. Macrophage and neutrophil influx into the lumen of the airways is markedly reduced. Peripheral blood neutrophils from treated animals exhibit a decreased phagocytotic ability and respiratory burst [76]. Chemotaxis to N-formyl-methionyl-leucylphenyalanine is also depressed, while TNFc~ production by alveolar macrophages is similarly compromised [77]. In marked contrast to the anti-inflammatory effects of SCGx and decentralization, the hypotensive effects of endotoxin are increased following these interventions in an animal model of endotoxic shock [78]. While the cells responsible for the modified responses to endotoxin in decentralized or SCGx animals have not been identified, neutrophils, platelets and monocytes/macrophages could all be involved. Neutrophils are known to play a significant role in determining the severity of hypersensitivity reactions, endotoxemia and postoperative hypoxia in the heart [79-81]. In addition a correlation exists between the severity of the response to endotoxin and neutrophil activity to nitroblue tetrazolium [82]. Platelets may attenuate the cytolytic activity of neutrophils through H202 scavenging by a glutathione cycle dependent process [83] and the monocyte/macrophage has a tremendous capacity to release mediators implicated in the immunophysiological effects of endotoxemia e.g. HzO2 and TNFc~ [84, 85]. At the time we first made these observations they highlighted the involvement of the SCG in modulation of responses to endotoxic and anaphylactic shock. However, given that the field of innervation of the SCG is limited to the upper thoracic and head regions it was deemed unlikely that the nerves affected by SCGx were innervating the organs primarily responsible for the immunophysiological reactions. These results suggested the involvement of an intermediary gland or organ. Subsequent experiments identified this intermediary as the SMG. Prior removal of the SMG prevented the depression of pulmonary inflammation and down regulation of macrophage and neutrophil function seen following surgical denervation [ 1]. These observations suggest that the SMG is a direct source of factors, which down regulate inflammation or control the release of anti-inflammatory factors from elsewhere in the body. Under normal circumstances the SCG exerts an inhibitory influence preventing the release of these factors from the gland while removal of the inhibitory tone through sympathetic decentralization or SCGx allows for an increased anti-inflammatory action as observed in the animal model of pulmonary inflammation. The enhanced hypotensive effects of endotoxin observed following sympathetic decentralization and SCGx were also observed following sialadectomy suggesting that the CST-SMG axis normally protects against potential hypotensive effects of endotoxin [78]. Taken together these results suggested a model in which postganglionic fibres arising from the SCG innervate cells within the SMG that synthesise both anti-hypotensive and anti-inflammatory factors. Under normal circumstances sympathetic signaling differentially regulates the release of these factors, stimulating the anti-hypotensive factor while inhibiting the release of the anti-inflammatory agents (Figure 4). 6.2.
CST-SMG control of mast cell function
Mast cell function is also modulated by the CST-SMG axis. However, the regulatory mechanism appears distinct from that involved in controlling neutrophil and macrophage responses [86]. TNF-ct-dependent cytotoxic activity of peritoneal mast cells (PMC) is reduced in rats following
212
Superior Cervical
A)
Ganglia decentralization
SCG stimulates release of factors leading to inhibition of endotoxic response
Q G
B) SCG exerts inhibitory tone on release of anti-inflammatory factors
G Figure 4. Immunomodulatory actions of the cervical sympathetic trunk-submandibular gland axis. A) Superior cervical ganglion decentralisation increases the hypotensive responses to endotoxin. This suggests that under normal circumstances the cervical sympathetic nerves stimulates the submandibular gland to release factors that down-regulate the response of immune cells to endotoxin. B) Superior cervical ganglion decentralisation down-regulatesinflammatory responses in sensitized rats. This suggests that intact sympathetic nerves suppress the release of anti-inflammatory agents from the submandibular gland.
decentralization but not SCGx. This suggests that the neural regulation of mast cell function probably occurs at the level of the SCG unlike neutrophils and macrophages where neural structures within the thoracic spinal cord are responsible for depressing function. Removal of the SMG also inhibits TNF-ot production by PMC indicating that salivary glands constitutively release a factor that upregulates mast cell function. However, since a combination of sialadenectomy and decentralization has no effect on PMC cytotoxicity, glands or organs other than the SMG probably participate in the regulation of mast cell activity. Taken as a whole these observations indicate that there are multiple mechanisms by which the CST-SMG can regulate immunological functions. 6.3.
A novel class of regulatory peptides
In an attempt to determine the SMG derived factors responsible for the modulation of endotoxic and anaphylaxtic reactions, extracts of SMG subjected to molecular weight cut-off filtration and high-performance liquid chromatography (HPLC) purification were tested for their ability to reduce the severity of endotoxin induced hypotension. Our initial expectation was that the agents involved would be well characterized regulatory factors such as NGF and EGF. However, utilizing these methods two novel peptides, which could attenuate the severity of endotoxin-induced hypotension, were isolated from purified extracts, sequenced and
213
synthesized. These peptides were: a pentapeptide, submandibular gland peptide S (SGP-S), with the sequence SGEGV and a heptapeptide, with the sequence TDIFEGG, named SGP-T [44, 87]. When given intravenously SGP-T can attenuate hypotension during cardiovascular anaphylaxsis, inhibit the disruption of intestinal myoelectric activity of the intestine and development of diarrhea during intestinal anaphylaxis, and downregulate neutrophil chemotaxis [88, 89]. These activities were also evident in the C-terminal fragment of the peptide, FEG. The D-isomeric form of FEG, denoted feG, was an effective inhibitor in both models of anaphylaxis when administered orally. SGP-S has not yet been assessed in a biological assay other than the endotoxic shock model. 6.4.
The VCS gene family
SGP-T was identified as a carboxy-terminal fragment (residues 138-144) of the submandibular gland rat 1 (SMR1) protein, while SGP-S is found closer to the amino terminal of the same polypeptide [44] (Figure 5). The sequence of the SMR1 protein was deduced from the cDNA sequence of the SMR1-VA1 gene, which encodes the prohormone-like protein in rat SMG [90]. The SMR1 polypeptide is found almost exclusively in the salivary glands and prostate and is one of several peptides generated by the variable coding sequence (VCS) multigene family that has been localized to chromosome 14 bands p21-p22 [91] [92]. The gene family has at least 10 members. Three of these, VCSA1 (SMR1 gene), VCSA2 and VCSA3 belong to the VCSA subclass and are found exclusively in the rat. These genes encode SMR1 and SMRl-related polypeptides that contain potential recognition sites for proteolytic enzymes. They can be considered potential preprohormones [93]. The structure of the VCSA1 gene is similar to the structure of several genes encoding prohormones such as genes for preprothyrotrophin-releasing hormone, preproenkephalins or preproopiomelanocortin [94]. It is not known whether this structure reflects the existence of a common ancestor, or convergent evolution. Seven genes belong to the VCSB subclass. The B subclass is found in several species and encodes a family of proline rich proteins found in rats, mice and humans [95]. A major characteristic of the VCS gene family is the presence of a hypervariable region inside the coding sequence [91]. In intraspecies pairwise comparisons a higher level of sequence divergence is observed in the hypervariable region than in the adjacent exonic or intronic sequences. Furthermore most of the mutations at the nucleotide level lead to amino acid substitutions. As a consequence the VCS family encodes proteins that are diverse in amino acid content, structure and probably function. Like many SMG derived polypeptides, the accumulation of mature SMR1 peptides appears to be dependent on the integrity of the hypothalamic-pituitarygonad axis. In 4 week old hypophysectomized or gonadectomized male rats the levels of mature peptide are greatly reduced, being close to two orders of magnitude less than in the SMG of sham operated animals [90].
6.5.
Processing of SMR1
The SMR1 prohormone contains an amino-terminal putative secretory signal sequence and a tetrapeptide (QHNP), located between dibasic amino acids, that constitutes the most common signal for prohormone processing. The proteolytic processing of SMR1 has been partially characterized by Rougeot et al. [96]. Cleavage at pairs of arginine residues close to the amino-terminus of the SMR1 polypeptide generates three structurally related peptides. The undecapeptide (23VRGPRRQHNPR33) is
214
SMR1 Prohormone and Peptide Fragments 146
1
SIGNALI~IPEPTI~RR DERF~ NH2
/
',,
19-SGEGV-23 (SGP-S)
23-VRGPRRQHNPR-33 28-RQHNPR-33 29-QHNPR-33
~
COOH
138-TDIFEGG-144 (SGP-T) 141-FEG-143
Figure 5. Schematic representation of the SMR1 preprotein and the peptides derived from it through proteolytic cleavage. Dibasic cleavage sites are indicated by small arrows and single-letter amino acid symbols.
generated by selective endoproteolysis at the Arg33-Arg34 bond and at the signal sequence. The hexapeptide (28RQHNPR33) and the pentapeptide (29QHNPR33) are generated by selective cleavages at both the Arg27-Arg28 and Arg33-Arg34 bonds (Figure 5). The biosynthesis of these peptides is subject to distinct regulatory pathways depending on the organ, sex and age of the rat. The peptides are differentially distributed within the SMG and in resting or epinephrine-elicited salivary secretions, suggesting distinct proteolytic pathways are involved in their maturation. In the male rat SMG the hexapeptide and the undecapeptide are found at increasing levels at 6-10 weeks of postnatal life [96]. This corresponds to the differentiation of acinar cells where the SMR-1 protein has been shown to localize. The 6 week old rat contains mostly the undecapeptide form, while the hexapeptide predominates in 10 week old animals. In 14 week old female SMG the undecapeptide is the major form. Protease activities are very low in the glands of newborn rats and mice and a rapid increase in enzyme activity coincides with the onset of puberty. Therefore the changes in ratio of processed peptides may be explained by the sex hormone dependant accumulation of SMR-1, processing enzymes or both, with fully active processing enzymes only being present in the male rat from 10 weeks onward. Although generated in the gland of both the male and female rats under basal conditions, the undecapeptide is only released into the saliva of the male. The hexapeptide is produced in large amounts in the gland of adult male rats and released into the saliva under both resting and epinephrine stimulated conditions. The pentapeptide appears only in the male saliva and is present mostly under stimulated conditions. Administration of epinephrine also induces the release of the hexapeptide into the blood stream. Therefore, it appears that the SMG can act as both an exocrine and endocrine organ for the SMR-1 derived peptides (Figure 6) [96]. Although no biological function has been ascribed to these peptides Rougeot et.al, were able to identify specific binding sites of the pentapeptide at physiological concentrations using radiolabelled peptide coupled to quantitative image analysis of whole rat body sections [97]. The peptide was detected in the renal outer medulla, bone and dental tissue glandular gastric mucosa and pancreatic lobules. Pentapeptide binding was localized to selective portions of the
215
Unstimulated
Stimulated
Epinephrine Saliva VRGPRRQHNPR RQHNPR
/
Saliva VRGPRRQHNPR
SMG
QHNPR
)
PRRQH
RQHNPR
-
/
RQHNPR
Figure 6. Differential processing and release of SMR 1-derived peptides in the SMG of male rats (ref. 96).
male rat nephron and in bone exclusively accumulated within the trabecular bone, remodeling unit. Based on these studies it has been suggested that the SMR-1 derived pentapeptide is primarily involved in the modulation of mineral balance between at least four systems: kidney, bone, tooth and circulation [97]. The enzymatic cleavage processes that generate SGP-T have not been identified. However, non-arginine-dependent serine proteases are abundant in salivary glands and could generate this peptide [98]. It has generally been believed that the immunomodulatory effects of the SMG are mediated predominantly by growth factors released into saliva and blood. However, more recent studies, described here, indicate that small peptides of salivary gland origin are also capable of modulating inflammatory reactions. The possibility that other pro-hormones in the salivary gland may yield small peptides capable of immunomodulation must be considered. Indeed, salivary glands contain chromogranin A and B, members of a family of highly acidic proteins, the chromogranins [99]. Chromogranins are prohormones and biological functions have been ascribed to many chromogranin-derived peptides. There are reports of antibacterial and antifungal activities and immunoregulatory functions including induction of monocyte chemotaxis and modulation of mast cell activity [100-103]. In rats, chromogranin A has been localized to the GCT and chromogranin-like immunoreactivity is detected in the saliva following stimulation with acetylcholine and noradrenaline [104]. Further investigations may reveal that these peptides also have a place in the immunomodulatory arsenal of the SMG.
216
7.
STRESS AND THE SUBMANDIBULAR GLAND
The widely held belief that the mind exerts significant effects on the health of an individual has stimulated research in an area termed psychoneuroimmunology. A major focus of psychoneuroimmunology has been a study of the effects of stress on the immune system. It is regarded as common knowledge that stressful experiences suppress the ability of the immune system to respond to antigenic challenge and thus increase susceptibility to infectious and neoplastic disease [105, 106]. Data from human and animal studies have confirmed the immunosuppressive effect of stress. However, there is evidence that stress may also enhance immune function, suggesting that the relationship between stress and the immune system is complex [107, 108]. These differences in the immune response to stress may be related to the stress inducing agent or the duration of stress (acute vs. chronic). During fighting or when under attack from a predator, immunological challenge in the form of a wound is likely. It would be advantageous for an organism if the capacity to respond to immunological challenge were enhanced in these acutely stressful situations. Increasing immune surveillance and preparing an organism for potential immune challenges arising from the action of stress inducing agents makes evolutionary sense. Indeed, antigen-specific cell mediated immunity can be significantly enhanced through activation of the physiological stress response [109]. The SMG may be one effector of the immunomodulatory actions of stress. Fighting is known to increase plasma NGF in mice, while other types of stress such as cold-water swim, escapable and inescapable footshock or physical restraint, do not affect plasma levels of this growth factor. Plasma NGF levels are related to the number of fighting episodes [110-112]. The NGF released from the SMG of adult male mice following intraspecific fighting is biologically active and causes degranulation of peritoneal mast cells. Administration of NGF antibodies or sialadenectomy prior to fighting blocks the mast cell degranulation [113]. Bilateral removal of the SMG leads initially to a significant increase in aggressive behaviour during social encounters. These behavioral changes decreased significantly during subsequent social encounters while defensive behaviors and elements of arrested flight increased progressively. These results suggested that sialadenectomy, perhaps by removing salivary NGF, interferes with the ability of mice to cope in stressful situations [113]. The exposure of rats to saturated ether vapour for 2 minutes is used as a model of acute stress. In adult male rats secretion of mature SMR1 peptides is rapidly stimulated in response to exposure to ether fumes (2 min), while no detectable increase in release is observed following long term exposure to stressful ambient temperatures (120 min at 4~ or 37~ [95]. The response to acute stress stimuli in conscious male rats results in an SMR-1 increase similar to that observed under pharmacologically induced adrenergic stimulation, suggesting that the surge in circulating SMR-1 is mediated by an endogenous adrenergic secretory response to acute stress. It would be of interest to examine changes in SMR-1 levels induced by behaviorally relevant stress such as fighting among males. It has been shown that adrenalectomy fails to block NGF release in fighting mice and injections of ACTH or extracts of adrenal gland or hypophysis do not result in NGF release into the bloodstream [112]. This suggests that the HPA axis is not involved in NGF release by the SMG. However, there are likely to be several mechanisms that control differential expression of biologically active peptides in response to various stimuli and the role of the HPA axis in modulating SMR-1 release should be examined.
217
8.
CONCLUSIONS
Some SMG factors show marked species related differences. For example, the serine proteinase, renin, was found in murine SMG but not in the rat [114]. Tonin, a member of the kallikrein family is present in large amounts in the rat SMG but has not been found in other species. High levels of NGF are detectable in both murine and rat SMG, whereas the human gland contains much smaller amounts [115]. Other factors such as kallikrein and EGF have been found in several species including humans [3, 4]. Immunoregulatory effects of the SMG have been observed in many species. However, the role of the human SMG in this regard is not clear. The qualitative and quantitative differences in protein content, coupled with the fact that many of the biologically active peptides can be synthesised elsewhere in the body, prompts the question; how necessary is the SMG for immune homeostasis? The possible redundancy of the gland's immunomodulatory actions is unclear, even in rodents. However, there are a number of observations that encourage further investigation of immuoregulation by the SMG in man. Sj6gren's syndrome (SS) is an inflammatory disease of the salivary and lacrimal glands. The disease results in failure of the glands leading to xerostomia and keratoconjunctivitis sicca [116, 117]. SS may occur as an isolated disease, but, it is frequently associated with systemic immune disorders, such as systemic lupus erythematous, rheumatoid arthritis, and sarcoidosis [118, 119]. These associations suggest that immunoregulation by salivary glands plays an important role in modulating systemic immunity. Strenuous exercise is followed by lymphopenia, neutrophilia, depressed NK cell function and lymphocyte proliferative responses to mitogens, and impaired natural immunity [120-122]. These exercise-induced immune changes may provide the physiological basis of altered resistance to infections. The mechanisms underlying exercise-induced immune changes are thought to be multifactorial and include neuroendocrinological and metabolic mechanisms. The contribution of the SMG to the immunological changes observed following exercise is unknown. However, it is interesting to note that salivary composition changes following strenuous exercise. Salivary IgA levels are decreased, while amylase, EGF and total protein have all been demonstrated to increase [123-125]. Further investigation of salivary and plasma levels of other SMG derived regulatory peptides may provide evidence for involvement of the gland in the immunological changes which follow exercise induced stress. As stated previously the gene encoding SMR1 (VCSA1) is found only in rats, while human salivary glands express genes from the VCSB subfamily. However it is possible that these, or as yet undiscovered, genes encode a human homologue of the SMR1. The observation that human neutrophils respond to, and thus may express receptors for, the tri-peptide FEG lends support to this hypothesis [44]. The use of degenerative primers for SMR1 mRNA to screen for humans homologues and the development of sensitive assay systems for FEG or SGP-T are two approaches which may lead to the discovery of novel anti-inflammatory peptides in man. From the information presented above it is clear the SMG constitutes a fully integrated component of the body's neuroimmunoendocrine network. The gland is under control of the CNS and regulated by the hypothalamus and sympathetic nervous system. The extent of the role played by the SMG in controlling systemic immune function remains to be seen. However, it is clear that investigations of what Barka describes as "that somewhat neglected "appendix" of the digestive system" [3] have the potential to provide valuable insights into the mechanisms underlying psychological and neuroendocrine control of the immune system.
218
REFERENCES
.
.
.
.
6. o
.
10. 11. 12.
13.
14.
15. 16. 17. 18. 19. 20. 21.
Mathison R, Hogan A, Helmer D, Bauce L, Woolner J, Davison JS, Schultz G, Befus D. Role for the submandibular gland in modulating pulmonary inflammation following induction of systemic anaphylaxis. Brain Behav Immun 1992, 6:117-129. Mathison R, Davison JS, Befus AD. Neuroendocrine regulation of inflammation and tissue repair by submandibular gland factors. Immunol Today 1994, 15: 527-532. Barka T. Biologically active polypeptides in submandibular glands. J Histochem Cytochem 1980; 28: 836-859. Gresik EW. The granular convoluted tubule (GCT) cell of rodent submandibular glands. icrosc Res Tech 1994; 27: 1-24. Levi-Montalcini R. The saga of the nerve growth factor. Neuroreport. 1998; 9: R71-83. Levi-Montalcini R, Skaper SD, Dal Toso R, Petrelli L, Leon A. Nerve growth factor: from neurotrophin to neurokine. Trends Neurosci 1996; 19: 514-520. Aloe L, R. Levi-Montalcini. Mast cells increase in tissues of neonatal rats injected with the nerve growth factor. Brain Res 1977; 133: 358-366. Marshall JS, Stead RH, McSharry C, Nielsen L, Bienenstock J. The role of mast cell degranulation products in mast cell hyperplasia. I. Mechanism of action of nerve growth factor. J Immunol 1990; 144:1886-1892. Pearce FL, Thompson HL. Some characteristics of histamine secretion from rat peritoneal mast cells stimulated with nerve growth factor. J Physiol (Lond) 1986; 372: 379-93: 379-393. Bischoff SC, Dahinden CA. Effect of nerve growth factor on the release of inflammatory mediators by mature human basophils. Blood 1992; 79: 2662-2669. Boyle MD, Lawman MJ, Gee AP, Young M. Nerve growth factor: a chemotactic factor for polymorphonuclear leukocytes in vivo. J Immunol 1985; 134: 564-568. Gee AP, Boyle MD, Munger KL, LawmanMJ, Young M. Nerve growth factor: stimulation of polymorphonuclear leukocyte chemotaxis in vitro. Proc Natl Acad Sci USA 1983; 80:7215-7218. Kannan Y, Ushio H, Koyama H, Okada M, Oikawa M, Yoshihara T, Kaneko M, Matsuda H. 2.5S nerve growth factor enhances survival, phagocytosis, and superoxide production of murine neutrophils, lood 1991; 77: 1320-1325. Amico-Roxas M., Caruso A, Leone MG, Scifo R, VanellaA, Scapagnini U. Nerve growth factor inhibits some acute experimental inflammations. Arch Int Pharmacodyn Ther. 1989; 299: 269-85: 269-285. Brodie C, Gelfand EW. Functional nerve growth factor receptors on human B lymphocytes. Interaction with IL-2. J Immunol 1992; 148: 3492-3497. Otten U., Ehrhard P, Peck R. Nerve growth factor induces growth and differentiation of human B lymphocytes. Proc Natl Acad Sci USA 1989; 86: 10059-10063. Kimata H., Yoshida A, Ishioka C, Kusunoki T, Hosoi S, Mikawa H. 1991. Nerve growth factor specifically induces human IgG4 production. Eur J Immunol 21: 137-141. Massague J, Cheifetz S, Laiho M, Ralph DA, Weis FM, Zentella A. 1992. Transforming growth factor-beta. Cancer Surv 12: 81-103: 81-103. Gill GN, Bertics PJ, Santon JB. Epidermal growth factor and its receptor. Mol Cell Endocrinol 1987; 51: 169-186. Tsutsumi O, Tsutsumi A, Oka T. Epidermal growth factor-like, corneal wound healing substance in mouse tears. J Clin Invest 1988; 81: 1067-1071. Partridge M, Gullick WJ, Langdon JD, Sherriff M. Expression of epidermal growth
219
22.
23.
24.
25.
26.
27.
28.
29.
30. 31. 32. 33.
34. 35. 36.
37.
38.
factor receptor on oral squamous cell carcinoma. Br J Oral Maxillofac Surg 1988; 26: 381-389. Nordlund L, Hormia M, Saxen L, Thesleff I. Immunohistochemical localization of epidermal growth factor receptors in human gingival epithelia. J Periodontal Res 1991; 26: 333-338. Wahl SM, Hunt DA, Wakefield LM, McCartney-Francis N, Wahl LM, Roberts AB, Sporn MB. Transforming growth factor type beta induces monocyte chemotaxis and growth factor production. Proc Natl Acad Sci USA 19887; 84: 5788-5792. Brandes ME, Mai UE, Ohura K, Wahl SM. Type I transforming growth factor-beta receptors on neutrophils mediate chemotaxis to transforming growth factor-beta. J Immunol 1991; 147: 1600-1606. Adams DH, Hathaway M, Shaw J, Burnett D,Elias E, Strain AJ. Transforming growth factor-beta induces human T lymphocyte migration in vitro. J Immunol 1991; 147: 609-612. Salgame P, Abrams JS, Clayberger C, Goldstein H, Convit J, Modlin RL, Bloom BR. Differing lymphokine profiles of functional subsets of human CD4 and CD8 T cell clones. Science 1991; 254: 279-282. Kehrl JH, Thevenin C, Rieckmann P, Fauci AS. Transforming growth factor-beta suppresses human B lymphocyte Ig production by inhibiting synthesis and the switch from the membrane form to the secreted form of Ig mRNA. J Immunol 1991; 146: 4016-4023. Konturek PC, Konturek SJ, Brzozowski T, Ernst H. Epidermal growth factor and transforming growth factor-alpha: role in protection and healing of gastric mucosal lesions. Eur J Gastroenterol. Hepatol. 1995; 7: 933-937. Wingren U, Brown TH, Watkins BM, Larson GM. Delayed gastric ulcer healing after extirpation of submandibular glands is sex-dependent, cand J Gastroenterol 1989; 24: 1102-1106. Noguchi S, Ohba Y, Oka T. Effect of salivary epidermal growth factor on wound healing of tongue in mice. Am J Physiol 1991; 260: E620-5. Noguchi S, Ohba Y, Oka T. Influence of epidermal growth factor on liver regeneration after partial hepatectomy in mice. J Endocrinol 1991; 128:425-431. Tsutsumi O, Taketani Y, Oka T. The uterine growth-promoting action of epidermal growth factor and its function in the fertility of mice. J Endocrinol 1993; 138: 437-444. Acres RB, Lamb JR, Feldman M. Effects of platelet-derived growth factor and epidermal growth factor on antigen-induced proliferation of human T-cell lines. Immunology 1985; 54: 9-16. Johnson HM, Torres B A. Peptide growth factors PDGF, EGF, and FGF regulate interferon-gamma production. J Immunol 1985; 134: 2824-2826. Aune TM. Inhibition of soluble immune response suppressor activity by growth factors. Proc Natl Acad Sci USA 1985; 82: 6260-6264. Laskin DL, Laskin JD, Weinstein IB, Carchman RA. Modulation of phagocytosis by tumor promoters and epidermal growth factor in normal and transformed macrophages. Cancer Res 1980; 40: 1028-1035. Laskin DL, Laskin JD, Weinstein IB, Carchman RA. Induction of chemotaxis in mouse peritoneal macrophages by phorbol ester tumor promoters. Cancer Res 1981; 41: 1923-1928. Roberts ML, Freston JA, Reade PC. Suppression of immune responsiveness by a submandibular salivary gland factor. Immunology 1976; 30:811-814.
220
39.
40. 41. 42. 43. 44. 45. 46.
47. 48. 49.
50.
51. 52.
53.
54.
55.
56. 57. 58.
Farhadi H, Pareek S, Day R, Dong W, Chretien M, Bergeron JJ, Seidah NG, Murphy RA. Prohormone convertases in mouse submandibular gland: co-localization of furin and nerve growth factor. J Histochem Cytochem 1997; 45: 795-804. Fuller PJ, Funder JW. The cellular physiology of glandular kallikrein. Kidney Int 1986; 29: 953-964. Wines DR, Brady JM, Pritchett DB, Roberts JL, MacDonald RJ. Organization and expression of the rat kallikrein gene family. J Biol Chem 1989; 264: 7653-7662. Hu ZQ, Murakami K, Ikigai H., Shimamura T. Enhancement of lymphocyte proliferation by mouse glandular kallikrein. Immunol Lett 1992; 32: 85-89. Matsushita S, Katz DH. Biphasic effect of kallikrein on IgE and IgG1 syntheses by LPS/IL-4-stimulated B cells. Cell Immunol 1993; 146: 210-214. Mathison R. Submandibular gland peptides and the modulation of anaphylactic and endotoxic reactions. Biomedical Reviews 1998; 9: 101-109. Gresik EW. Postnatal developmental changes in submandibular glands of rats and mice. J Histochem Cytochem 1980; 28: 860-870. Kumegawa M, Takuma T, Takagi Y. Precocious induction of secretory granules by hormones in convoluted tubules of mouse submandibular glands. Am J Anat 1977; 149:111-114. Sagulin GB, Roomans GM. Effects of thyroxine and dexamethasone on rat submandibular glands. J Dent Res 1989, 68:1247-1251. Bhoola KD, Figueroa CD, Worthy K. Bioregulation of kinins: kallikreins, kininogens, and kininases. Pharmacol Rev 1992; 44: 1-80. Aloe L, Levi-Montalcini R. Comparative studies on testosterone and L-thyroxine effects on the synthesis of nerve growth factor in mouse submaxillary salivary glands. Exp Cell Res 1980; 125:15-22. Wagner D, Metzger R, Paul M, Ludwig G, Suzuki F, Takahashi S, Murakami K, Ganten D. Androgen dependence and tissue specificity of renin messenger RNA expression in mice. J Hypertens 1990; 8: 45-52. Ishii DN, Shooter EM. Regulation of nerve growth factor synthesis in mouse submaxillary glands by testosterone. J Neurochem 1975, 25: 843-851. Hiramatsu M, Kashimata M, Takayama F, Minami N. Developmental changes in and hormonal modulation of epidermal growth factor concentration in the rat submandibular gland. J Endocrinol 1994, 140: 357-363. Kyakumoto S, Kurokawa R, Ota M. Effect of castration and administration of testosterone on cytosol and nuclear androgen receptor in mouse submandibular gland. Biochem Int 1985, 11: 701-707. Kyakumoto S, Kurokawa R, Ohara-Nemoto Y, Ota M. Sex difference in the cytosolic and nuclear distribution of androgen receptor in mouse submandibular gland. J Endocrinol 1986; 108: 267-273. Boyer R, Escola R, Bluet-Pajot MT, Arancibia S. Ablation of submandibular salivary glands in rats provokes a decrease in plasma luteinizing hormone levels correlated with morphological changes in Leydig cells. Arch Oral Biol 1990, 35: 661-666. Aloe L, Levi-Montalcini R. Enhanced differentiation of sexually dimorphic organs in L-thyroxine treated Tfm mice. Cell Tissue Res 1980; 205: 19-29. Cabello G, Wrutniak C. Thyroid hormone and growth: relationships with growth hormone effects and regulation. Reprod Nutr Dev 1989; 29: 387-402. Walker P. Thyroxine increases submandibular gland nerve growth factor and epidermal growth factor concentrations precociously in neonatal mice: evidence for thyroid hormone-
221
59. 60.
61.
62. 63. 64.
65. 66.
67.
68.
69. 70. 71.
72.
73. 74.
75.
76.
mediated growth factor synthesis. Pediatr Res 1986; 20: 281-284. Verhoeven G. Androgen binding proteins in mouse submandibular gland. J Steroid Biochem 1979; 10: 129-138. Fujieda M, Murata Y, Hayashi H, Kambe F, Matsui N, Seo H. Effect of thyroid hormone on epidermal growth factor gene expression in mouse submandibular gland. Endocrinology 1993; 132: 121-125. Walker P, Weichsel Jr. ME, Hoath SB, Poland RE, . Fisher DA. Effect of thyroxine, testosterone, and corticosterone on nerve growth factor (NGF) and epidermal growth factor (EGF) concentrations in adult female mouse submaxillary gland: dissociation of NGF and EGF responses. Endocrinology 1981; 109: 582-587. Gresik EW, Chung KW, Barka T, Schenkein I. Immunocytochemical localization of nerve growth factor, submandibular glands of Tfm/Y mice. Am J Anat 1980; 158: 247-250. Nagy E, Berczi I. Sabbadini E. Endocrine control of the immunosuppressive activity of the submandibular gland. Brain Behav Immun 1992; 6:418-428. Luger A, Calogero AE, Kalogeras K, Gallucci WT, Gold PW, Loriaux DL, Chrousos GP. Interaction of epidermal growth factor with the hypothalamic-pituitary-adrenal axis: potential physiologic relevance. J Clin Endocrinol Metab 1988; 66: 334-337. Otten U, Baumann JB, Girard J. Stimulation of the pituitary-adrenocortical axis by nerve growth factor. Nature 1979; 291: 358-366. Raber J, Koob GF, Bloom FE. Interferon-alpha and transforming growth factor-beta 1 regulate corticotropin-releasing factor release from the amygdala: comparison with the hypothalamic response. Neurochem. Int 1997; 30: 455-463. Muir TC, Pollock D, Turner CJ. The effects of electrical stimulation of the autonomic nerves and of drugs on the size of salivary glands and their rate of cell division. J Pharmacol Exp Ther 1975; 195: 372-381. Martinez JR, Quissell DO, Wood DL, Giles M. Abnormal secretory response to parasympathomimetic and sympathomimetic stimulations from the submaxillary gland of rats treated with reserpine. J Pharmacol Exp Ther 1975; 194: 384-395. Tsukitani K, Mori M. Immunohistochemistry and radioimmunassay of EGF in submandibular glands of mice treated with secretagogues. Cell Mol Biol 1986; 32: 677-683. Orstavik TB, Carretero OA, Scicli AG. Kallikrein-kinin system in regulation of submandibular gland blood flow. Am J Physiol 1982; 242: H1010-4. Shaw PA, Barka T, Woodin A, Schacter BS, Cox JL. Expression and induction by beta-adrenergic agonists of the cystatin S gene in submandibular glands of developing rats. Biochem. J. 1990; 265:115-120. Chaparro O., Yu WH, Shaw PA. Effect of sympathetic innervation on isoproterenolinduced cystatin S gene expression in rat submandibular glands during early development. ev. Neurosci 1998; 20: 65-73. Cardinali DP, Romeo HE. Peripheral neuroendocrine interrelationships in the cervical region. NIPS 1990; 5:100-104. Alito AE, Romeo HE, Bajer R, Chuluyan HE, Braun M, Cardinali DP. Autonomic nervous system regulation of murine immune responses as assessed by local surgical sympathetic and parasympathetic denervation. Acta Physiol Pharmacol Latinoam. 1987; 37:305-319. Ramaswamy K, Mathison R, Carter L, Kirk D, Green F, Davison JS, Befus D. Marked antiinflammatory effects of decentralization of the superior cervical ganglia. J Exp Med 1990; 172: 1819-1830. Carter L., Ferrari JK, Davison JS, Befus D. Inhibition of neutrophil chemotaxis and
222
77.
78. 79.
80. 81. 82.
83. 84. 85.
86.
87. 88.
89. 90.
91.
92.
93.
activation following decentralization of the superior cervical ganglia. J Leukoc Biol 1992; 51: 597-602. Mathison R, Carter L, Mowat C, Bissonnette E, Davison JS, Befus AD. Temporal analysis of the anti-inflammatory effects of decentralization of the rat superior cervical ganglia. Am J Physiol 1994; 266: R1537-43. Mathison R, Befus D, Davison JS. 1993. Removal of the submandibular glands increases the acute hypotensive response to endotoxin. Circ. Shock 39: 52-58. Wilkinson JR, Crea AE, Clark TJ, Lee TH. Identification and characterization of a monocyte-derived neutrophil-activating factor in corticosteroid-resistant bronchial asthma. J Clin Invest 1989; 84:1930-1941. Grisham MB, Everse J, Janssen HF. Endotoxemia and neutrophil activation in vivo. Am J Physiol 1988; 254: H1017-22. Kraemer R, Mullane KM. Neutrophils delay functional recovery of the post-hypoxic heart of the rabbit. J Pharmacol Exp Ther 1989; 251: 620-626. Waddell SC, Davison JS, Befus AD, Mathison RD. Role for the cervical sympathetic trunk in regulating anaphylactic and endotoxic shock [published erratum appears in J Manipulative Physiol Ther 1992 Mar-Apr; 15(3): following Table of Contents]. J. Manipulative. Physiol Ther. 1992; 15: 10-15. Dallegri F., Ballestrero A, Ottonello L, Patrone F. Platelets as inhibitory cells in neutrophil-mediated cytolysis. J. Lab Clin Med 1989; 114: 502-509. Tsunawaki S., Sporn M, Ding A, Nathan C. Deactivation of macrophages by transforming growth factor-beta. Nature 1988; 334: 260-262. Zuckerman SH, Evans GF, Snyder YM, Roeder WD. Endotoxin-macrophage interaction: post-translational regulation of tumor necrosis factor expression. J Immunol 1989; 143: 1223-1227. Mathison R, Bissonnette E, Carter L, Davison JS, Befus D. The cervical sympathetic trunk-submandibular gland axis modulates neutrophil and mast cell functions. Int Arch Allergy Immunol 1992; 99:419-421. Mathison RD, Befus AD, Davison JS. A novel submandibular gland peptide protects against endotoxic and anaphylactic shock. Am J Physiol 1997; 273: R1017-23. Mathison R, Lo P, Moore G, Scott B, Davison JS. Attenuation of intestinal and cardiovascular anaphylaxis by the salivary gland tripeptide FEG and its D-isomeric analog feG. Peptides 1998; 19: 1037-1042. Mathison R, Tan D, Oliver M, Befus D, Scott B, Davison JS. Submandibular gland peptide-T (SGP-T) inhibits intestinal anaphylaxis. Dig Dis Sci 1997; 42: 2378-2383. Rosinski-Chupin I, Tronik D, Rougeon F. High level of accumulation of a mRNA coding for a precursor-like protein in the submaxillary gland of male rats. Proc Natl Acad Sci USA 1988; 85: 8553-8557. Courty Y, Singer M, Rosinski-Chupin I, Rougeon F. Episodic evolution and rapid divergence of members of the rat multigene family encoding the salivary prohormone-like protein SMR1. Mol Biol Evol. 1996; 13: 758-766. Rosinski-Chupin I., Kuramoto T, Courty Y, Rougeon F, Serikawa T. Assignment of the rat variable coding sequence (VCS) gene family to chromosome 14. Mamm. Genome 1995; 6: 153-154. Rosinski-Chupin I., Rougeot C, Courty Y, Rougeon F. Localization of mRNAs of two androgen-dependent proteins, SMR1 and SMR2, by in situ hybridization reveals sexual differences in acinar cells of rat submandibular gland. J Histochem Cytochem 1993; 41: 1645-1649.
223
94.
95.
96.
97.
98.
99.
100.
101.
102.
103.
104.
105. 106. 107. 108.
109. 110. 111.
Rosinski-Chupin I, F. Rougeon. The gene encoding SMR1, a precursor-like polypeptide of the male rat submaxillary gland, has the same organization as the preprothyrotropinreleasing hormone gene. DNA Cell Biol 1990; 9: 553-559. Rougeot C, Rosinski-Chupin I, Rougeon F. Novel genes and hormones in salivary glands: from the gene for the submandibular rat 1 protein (SMR1) precursor to receptor sites for SMR1 mature peptides. Biomedical Reviews 1998; 9: 17-32. Rougeot C, Rosinski-Chupin I, Njamkepo E, Rougeon F. Selective processing of submandibular rat 1 protein at dibasic cleavage sites. Salivary and bloodstream secretion products. Eur J Biochem. 1994; 219: 765-773. Rougeot C, Vienet R, Cardona A, Le Doledec L, Grognet JM, Rougeon F. Targets for SMRl-pentapeptide suggest a link between the circulating peptide and mineral transport. m. J Physiol 1997; 273: R1309-20. Berg T., Wassdal I, Sletten K. Immunohistochemical localization of rat submandibular gland esterase B (homologous to the RSKG-7 kallikrein gene) in relation to other serine proteases of the kallikrein family. J Histochem Cytochem 1992; 40: 83-92. Letic-Gavrilovic A, Shibaike S, Niina M, Naruse S, Abe K. Localization of chromogranin A and B, beta-endorphin and enkephalins in the submandibular glands of mice. Shika. Kiso. Igakkai. Zasshi. 1989; 31: 453-462. Lugardon K, Raffner R, Goumon Y, Corti A, Delmas A, Bulet P, Aunis D, MetzBoutigue. Antibacterial and antifungal activities of vasostatin-1, the N-terminal fragment of chromogranin A. J Biol Chem 2000; Apr. 14; 275(15): 10745-53. 275: 10745-10753. Kong C, Gill BM, Rahimpour R, Xu L, Feldman RD, Xiao Q, McDonald TJ, Taupenot L, Mahata SK, Singh B, O'Connor DT, Kelvin DJ. Secretoneurin and chemoattractant receptor interactions. J Neuroimmunol 1998; 88: 91-98. Reinisch N, Kirchmair R, Kahler CM, Hogue-Angeletti R, Fischer-Colbrie R, Winkler H, Wiedermann CJ. Attraction of human monocytes by the neuropeptide secretoneurin. FEBS Lett 1993, 334: 41-44. Forsythe P, Curry WJ, Johnston CF, Harriott P, MacMahon J, Ennis M. 1The modulatory effects of WE-14 on histamine release from rat peritoneal mast cells. Inflamm. Res 1997; 46 Suppl 1: S13-4: S13-4. Kanno T, Asada N, Yanase H, Iwanaga T, Ozaki T, Nishikawa Y, Iguchi K, Mochizuki T, Hoshino M, Yanaihara N. Salivary secretion of highly concentrated chromogranin a in response to noradrenaline and acetylcholine in isolated and perfused rat submandibular glands. Exp Physiol 1999; 84: 1073-1083. Borysenko JZ. Behavioral-physiological factors in the development and management of cancer. Gen. Hosp. Psychiatry 1982; 4: 69-74. Kort WJ. The effect of chronic stress on the immune response. Adv Neuroimmunol 1994; 4:1-11. Dhabhar FS, Miller AH, McEwen BS, Spencer RL. Effects of stress on immune cell distribution. Dynamics and hormonal mechanisms. J Immunol 1995; 154:5511-5527. Dhabhar FS, Miller AH, Stein M, McEwen BS, Spencer RL. Diurnal and acute stressinduced changes in distribution of peripheral blood leukocyte subpopulations. Brain Behav Immun 1994; 8: 66-79. Dhabhar FS. Stress-induced enhancement of cell-mediated immunity. Ann N. Y. Acad. Sci 1998; 840: 359-72: 359-372. Alleva E, Aloe L, Bigi S. An updated role for nerve growth factor in neurobehavioural regulation of adult vertebrates. Rev Neurosci 1993; 4: 41-62. Maestripieri D, De Simone R, Aloe L, Alleva E. Social status and nerve growth factor
224
serum levels after agonistic encounters in mice. Physiol Behav 1990; 47:161-164. 112. Aloe L., Alleva E, Bohm A, Levi-Montalcini R. Aggressive behavior induces release of nerve growth factor from mouse salivary gland into the bloodstream. Proc Natl Acad Sci USA 1986; 83: 6184-6187. 113. De Simone R, Alleva E, Tirassa P, Aloe L. Nerve growth factor released into the bloodstream following intraspecific fighting induces mast cell degranulation in adult male mice. Brain Behav Immun 1990; 4: 74-81. 114. Morris B J, de Zwart RT, Young JA. Renin in mouse but not in rat submandibular glands. Experientia 1980; 36: 1333-1334. 115. Levi-Montalcini R, Dal Toso R, della Valle F, Skaper SD, Leon A. Update of the NGF saga. J Neurol Sci 1995; 130:119-127. 116. St. Clair EW. New developments in Sjogren's syndrome, urr Opin Rheumatol 1993; 5:604-612. 117. Rhodus NL. Sjogren's syndrome. Quintessence Int 1999; 30: 689-699. 118. Lois M, Roman J, Holland W, Agudelo C. Coexisting Sjogren's syndrome and sarcoidosis in the lung. Semin Arthritis Rheum 1998; 28:31-40. 119. McDonagh JE, Isenberg DA. Development of additional autoimmune diseases in a population of patients with systemic lupus erythematosus. Ann Rheum Dis 2000. Mar; 59(3): 230-2. 2000; 59: 230-232. 120. Pedersen BK, Bruunsgaard H, Jensen M, Toft AD, Hansen H, Ostrowski K. Exercise and the immune system-influence of nutrition and ageing. J Sci Med Sport 1999; 2: 234-252. 121. Pedersen BK, Bruunsgaard H, Jensen M, Krzywkowski K, Ostrowski K. Exercise and immune function: effect of ageing and nutrition. Proc Nutr Soc 1999; 58: 733-742. 122. Nieman DC. Nutrition, exercise, and immune system function. Clin Sports Med 1999; 18: 537-548. 123. Nexo E, Hansen MR, Konradsen L. Human salivary epidermal growth factor, haptocorrin and amylase before and after prolonged exercise. Scand. J Clin Lab Invest 1988; 48: 269-273. 124. Mackinnon LT, Ginn E, Seymour GJ. Decreased salivary immunoglobulin A secretion rate after intense interval exercise in elite kayakers. Eur J Appl Physiol 1993; 67: 180-184.
125. McDowell SL, Hughes RA, Hughes RJ, Housh TJ, Johnson GO. The effect of exercise training on salivary immunoglobulin A and cortisol responses to maximal exercise. Int J Sports Med 13: 577-580.
New Foundationof Biology
225
Edited by I. Berczi and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
Glandular Kallikrein in Immunoregulation
EDRIS SABBADINI 1, EVA NAGY 1, ALEXANDER FRED T. KISIL 1 and ISTVAN BERCZI 1
V(~RI~S 2, GERTRUDE VOROSOVA 3,
1Department of Immunology, University of Manitoba, Winnipeg, Manitoba, Canada 2Department of Physiology, University of Manitoba, Winnipeg, Manitoba, Canada 3Department of Anatomy, University of Manitoba, Winnipeg, Manitoba, Canada
ABSTRACT Glandular Kallikrein (GK) is an enzyme of the serine protease family capable of generating biologically active peptides by partially degrading various substrates. It is found in several tissues with particularly high concentrations in salivary glands, pancreas, kidney and the prostate gland. The physiological functions of this enzyme appear to vary according to the tissue in which it acts and the substrate(s) available in such tissues. Our interest in GK arose when we demonstrated that an immunoregulatory factor isolated from the salivary submandibular (SM) gland of rats, was rat GK (rGK). When added to cultures of lymphoid tissues, rGK induced a significant increase in lymphocyte proliferation manifested either in unstimulated cell cultures or in cultures of cells stimulated with suboptimal amounts of T cell mitogens. When injected subcutaneously, rGK induced a marked, albeit short lived, immunosuppressive effect in various experimental models including contact hyper-sensitivity to picryl chloride, allograft rejection, the production of antibody plaque-forming cells in the spleen of animals immunized with sheep red blood cells, and collagen-induced arthritis. All of the above effects were species-nonspecific inasmuch as murine GK and pig GK induced identical effects to rGK either in mice or in rats. Moreover, for these effects to take place, the enzymatic activity of GK had to be preserved. In the presence of inhibitors of proteolytic activity, both the in vivo and the in vitro immunoregulatory effects were abolished. This indicated that GK acts on a protein substrate to generate immunoregulatory peptides. If this is the case, the apparent conflict between in vitro stimulation and the suppression induced by subcutaneously injected GK can be explained as due to the accumulation of lymphocyte stimulating peptides in the in vitro cultures, while in vivo injection probably results in the rapid dispersal and degradation of such peptides leaving the organism depleted of substrate and unable to produce amounts of active peptides sufficient for a normal immune response. The effects described up to this point refer to in vitro phenomena or to the subcutaneous administration of GK. The fact that GK is found in high concentration in salivary glands and is secreted in saliva suggests that a significant physiological function of GK may occur following external salivary secretion. For this reason, we also tested the effect of orally administered GK in rats. Oral GK was shown to be required for the induction of tolerance to orally administered antigens. Rats subjected to the surgical removal of the SM gland (SMX) did
226
not develop tolerance to oral collagen given in doses that induced significant tolerance in normal controls. The oral administration of GK in the SMX animals fully restored the capacity to develop oral tolerance. This suggests that salivary GK is responsible for homeostasis in the mucosa associated lymphoid tissue which is responsible for the development of tolerance to oral antigens. The observation that GK may be used to enhance the induction of oral tolerance holds promises for the therapuetic application of oral GK in autoimmune diseases.
1.
INTRODUCTION
Kallikreins are a family of serine proteases capable of cleaving various substrates and generating biologically active peptides. In spite of the identity of names, tissue or glandular kallikreins should be distinguished from plasma kallikrein. They differ from plasma kallikrein in their genes of origin, molecular weight, amino acid sequences, substrates, peptide products and most probably physiological functions. There are at least 20 genes for tissue kallikrein in rodents [1], while in humans only 4 genes have been so far described. Of these, only one gene in each species codes for true glandular kallikrein. Of the rat genes at least 6-7 appear to be expressed in the submandibular (SM) gland [2]. These include true glandular kallikrein, tonin, c~ and ~, NGF, and the EGF-binding protein (EGF-BP), type A, B, and C. Here the term glandular kallikrein (GK) will apply to true GK only, while the general term kallikrein(s) will be used for any unspecified members of the tissue kallikrein family. True GK has been designated in various species as kallikrein-1 (K1) [3]. The best known substrates for GK action are hepatic-derived kininogens [3] which occur in two forms, low molecular weight kininogen (50 kDa) and high molecular weight kininogen (120 kDa). From the action of plasma kallikrein on high molecular weight kininogen a nonapeptide, bradikinin, is generated, while in most species GK gives rise to a decapeptide, kallidin (lys-bradikinin) from either low or high molecular weight kininogen. Kallidin is biologically active in itself but may also be further processed into bradikinin. An exception may be the GK of the rat SMG which was reported to produce bradikinin [4]. While the action of GK on kininogen is particularly well studied, the full range of GK substrates has not yet been investigated. Since GK is, and most of the times remains, localized in certain tissues, physiological substrates are likely to vary from tissue to tissue. Of particular interest here is the possibility that GK may activate or in some way regulate other immunologically active factors of the SM gland, including NGF, EGF/TGF~ and TGF[3. Thus, salivary gland GK may exert its immunological effects either via the production of classical kinins or via other immunologically active factors. Moreover, salivary GK is actively secreted in saliva [5] and would be expected to reach various points in the gastro-intestinal tract and act on various substrates there. Several in vitro effects of kallikreins on cells of the immune system have been reported. Thus, several authors have described mitogenic and co-mitogenic effects of kallikrein and other serine proteases. Such mitogenic effects were observed with thymocytes [6], T cells and B cells [7]. Although bradikinin may also have mitogenic effects [8], the involvement of this kinin in kallikrein-induced mitogenesis is not well investigated. Moreover, several proteases, including kallikrein, were shown to be involved in immunoglobulin isotype control. Thus Ishizaka described a kallikrein-like factor called glycosylation-enhancing factor, which induced CD4+ T cells to produce an IgE-potentiating factor and to favour the production of IgE by memory B cells [9]. Serine proteases from Schistosoma mansoni schistosomula were reported to enhance IgE production [10]. Moreover, the addition of kallikrein and other serine proteases in various
227
concentrations to cultures of B cells stimulated with LPS and IL 4 enhanced the production of IgE, IgG 1, or IgG3, depending on the enzyme concentration used [ 11 ]. Our interest in GK arose from studies on immunosuppressive factors in the SM gland of rats. The addition of crude extracts from rat SM glands to murine spleen and lymph node cultures stimulated with concanavalin A (Con A) induced either suppression (at high concentrations) or further stimulation (at lower concentrations) of proliferative activity [12]. This suggested that these extracts contained factors with suppressive effects as well as factors with the ability to enhance lymphocyte proliferation. Gel filtration of the crude extracts revealed that the in vitro suppressive activity was due to factors with molecular weight higher than 50 kilo Daltons (kDa), while stimulation was due to factors with molecular weight lower than 50 kDa (Figure 1). We tested the in vivo activity of both the higher and lower molecular weight fractions in the skin allograft, direct plaque forming cell response and in the delayed-type hypersensitivity (DTH) models [13]. As shown in Table I, and contrary to what one might have expected in view of their in vitro suppressive activity, the high molecular weight fractions did not have any significant effect in these models. On the other hand, the lower molecular weight fractions produced significant suppression in all three models. .I 6 0
-
120 =
,,7 o v
"(
80
ci.
40
!
!
o z's 8. 3 z- 8 0.93 0:3, o.,o Fresh hssue
(/J..g/ml)
Figure 1. Effects of pooled Sephacryl fractions of submandibular ( 0 , II) and parotid (O, F-I) glands on the Con A stimulated lymph node cell proliferation: ( 0 , 9 MW > 50 kDa; (m, D) MW < 50 kDa. Thesolid horizontal line is the value of mean c.p.m, for the controls and the shaded area represents the 95% confidence limits. From Ref. 12.
Fractionation of the lower molecular weight pool of fractions through successive steps of hydrophobic interaction, anion exchange chromatography and finally gel filtration, led to the isolation a single protein (Figure 2) which retained the properties of in vitro stimulation of lymphocyte proliferation and in vivo immunosuppression (results not shown). The amino acid sequence of the isolated protein was determined [14] using an automated Edman degradation procedure [15]. Figure 3 shows the partial N-terminal amino acid sequence of the 40 kDa protein and of the members of the kallikrein family represented in the rat SMG. The x's (unidentified amino acids) in our sequence are probably cysteines which are destroyed in the Edman degradation process. If this is taken into account, the first 25 amino acids of our protein
228
has identical sequence
with that of true GK and differ from those of other members
of the
kallikrein family. In view of the fact that no rat genes with the same amino acid sequence are k n o w n , it w a s c o n c l u d e d t h a t t h e i s o l a t e d p r o t e i n w a s t r u e r a t G K ( r G K ) . Table I
Effects of high molecular weight (HMW) and low molecular weight (LMW) pools of gel filtration fractions in three in vivo immunological assays. Modified from Ref. 13.
Exp Model
Treatment a
Groups
Results ___SD b
Skin transplantation
10 daily doses
PBS
12.2 + 0.37
CBA/2J to C57B1/6J
Days 0 to 9
HMW
13.0 • 0.44
NS
LMW
14.7 _+0.76
p < 0.05
Direct PFC
5 daily doses
PBS
237.0 • 19.7
Days 1 to 3
HMW
193.0 + 12.0
NS
DTH (A/J mice)
2 daily doses
Days 4 and 6
HMW
LMW
119.6 • 10.0
PBS
19.0 • 0.70
18.2 + 0.49
NS
LMW
9.0 • 1.00
Significancec
p < 0.05
p < 0.01
a. The animals received the subcutaneous injection of 200 ~tl of PBS or PBS containing either the HMW or the LMW fractions. The doses corresponded to one-half of a SM gland (0.965 mg or 0.53 mg, respectively) and to one gland in the other models (1.93 mg or 1.06 mg, respectively). b. Skin transplantation results are expressed as mean survival time; PFC response is expressed as the number of IgM anti-SRBC PFC per one million splenocytes; DTH results are expressed as increases of ear thickness in 1/10 mm units 24 hours after challenge. c. Significance was determined by two-sample t-test, using the PBS-treated group as the control; NS, not significant.
. . . .
97.4 kD 66.2 kD
I~
42.7 kD
~.
31kD
~,
21.5 kD
I~
i i 84184184 51 i.
14.4 kD
1
2
Figure 2. SDS-PAGE of the purified protein from rat SM gland (lane 2). Molecular weight standards are shown in lane 1. From Ref. 14.
229
Protein
40
1
i0
15
20
25
WGGYNxEMNSQPWQVAVYYFGEYLx
kDa
Gland.
5
Kallikrein
VVGGYNCEMNSQPWQVAVYYFGEYLC
GYNy~ ~ NSQPWQVA~
KLP-S3
K sQpwQvA
Tonin Antigen 7
:I VGG'i K rd~ Kt~SQPWQV'~
T-kininogenase
I VGG~ K ZE K NSQPWQV~
Proteinase
I
B
Proteinase A Figure 3. Partial N-terminal amino acid of the 40 kDa protein isolated from rat SM gland compared with those of members of kallikrein family expressed in the rat SM gland. The boxed areas represent regions of identity with the 40 kDa protein. Blank spaces are used to align homologous sequences in different proteins, x = not identified. From Ref. 14.
The esterase activity of the isolated rGK was approximately the same as that of a commercially obtained porcine GK (pGK) when measured in the 2-N-benzoyl-arginine ethyl estez (BAEE) assay [16]. Different concentrations of aprotinin induced different degrees of inhibition. Figure 4 demonstrates the effects of rGK and pGK in the presence or in the absence of aprotinin on the proliferative activity of Con A stimulated murine lymph node cells. The same concentrations of pGK induced similar co-mitogenic effects. The Con A concentration used in these experiments was such as to induce only suboptimal mitogenic effects, suitable for the demonstration of the co-mitogenic activity of rGK. The addition of different concentrations of aprotinin to the co-stimulated cultures induced dose dependent suppression. It should be noted that the highest concentration of aprotinin used in this experiment (1.5 ~tg/ml or 6 ~tg/culture) was capable of inducing some 90% inhibition in the B AEE assay. On the other hand, the lowest concentration of aprotinin (1.5 ~tg/culture) induced approximately 40% inhibition of the enzymatic activity and partial inhibition of the co-mitogenic activity. The results of an in vivo experiment performed along the same lines are presented in Figure 5. A DTH reaction was induced in mice sensitized with picryl chloride and challenged with the same agent six days later. The injection of rGK 24 hour before challenge resulted in nearly complete (~ 87%) suppression of the response. Similar suppression was obtained with pGK (-72%). The dose of rGK used in this experiment (57 ~tg/animal) was based on our previous experience. The higher of the two aprotinin doses (190 ~tg/animal) was selected so as to provide, after dilution in the blood stream, a concentration similar to that used in the in vitro experiments. The suppressive effects of rGK and pGK were almost completely removed by the injection of this higher dose of aprotinin given immediately before GK injection (inhibition nearly -8%). On the other hand, the lower dose (95 ~tg) of aprotinin induced only partial suppression of the rGK effects (inhibition still -60%). These two experiments clearly demonstrate that
230
the enzymatic activity of GK must be preserved in order to retain its in vivo and i n vitro immunological effects.
cpm xl,000 10 15 I
20
I
!
PBS Ill
rGK
II1|
II
I
-t-
I
§
Apr. (6 lag) rGK + Apr. (6 I~g) rGK + Apr. (3 I~g)
-q-
rGK + Apr. (1.5 I~g)
II
III
I
I
I
pGK
I
I
I
-t_ff
pGK + Apr. (6 Ixg)
Figure 4. Increased proliferative activity of Con A stimulated A/J lymph node cells induced by rGK and pGK and reversal of this effect with aprotinin. The rGK and pGK doses were 1.78 g/culture (final concentration 0.22 ~tM/1). Aprotinin (Apr.) doses were: 6 gg, 3 gg, or 1.5 gg per culture 4.6 M/i, 2.3 M/l, and 1.15 M/l, respectively). Results are expressed as counts per minute (C.P.M.) in triplicate cultures (+S.D.). From Ref. 14.
A Ear thickness ~rn rn) 0.5 1.0 I
I
+
PBS
+
rGK Apr. (190 I~g) rGK + Apr. (190 I~g)
II
el
rGK + Apr. (95 I~g) pGK pGK + Apr. (190 Jag)
II
, I
, , II
I
"t-
II
-I-
Figure 5. Effects of rGK and pGK on the DTH response of A/J mice immunized with picryl chloride and reversal of such effects with aprotinin. Aprotinin (Apr.) was injected subcutaneously as a full dose (190 pg per animal) or as a half dose (95 ~tg per animal). Fifteen minutes later the animals received a further subcutaneous injection of rGK or pGK in 0.2 ml of PBS or PBS only. Results are expressed in terms of the increase of the thickness of the challenged ear over the pre-challenge values. Unimmunized controls (not shown) did not show any increases of the ear thickness. From Ref. 14.
231
Table II demonstrates the effects of varying the time of GK injection with respect to the time of immunization or of challenge in the DTH model in mice. If given before immunization, GK had a suppressive effect lasting for at least fourteen days. This suggested that the animals did not develop any immunity when presented with the antigen. On the other hand, if GK was given after the development of immunity, it induced a short lived suppression of the skin reaction with a full recovery of reactivity a week after GK administration. This demonstrated that GK did not affect the state of immunity of an animal and suppressed the skin reaction with a mechanism that may be either immunological or anti-inflammatory. It should be noted that the half life of GK is such that 24 hours after injection, i.e. at the time of antigen administration for either immnunization or challenge, none of the injected GK is present in the animal. Thus, GK mediated inhibition probably occurs indirectly, following production of mediators dependent on GK enzymatic action. Table II
Effects of the intradermal injection o f r G K given before or after immunization in the DTH model.
Treatment a
None
Test on day: b
7
14
18.4 + 0.76
17.6 + 0.92
rGK, Day-1
6.0 + 1.2"
7.6 + 0,98*
rGK, Day+6
4.4 + 0.63*
16.8 + 1.12
rGK, Day+13
19.0 + 1.24
5.6 + 0.78*
a. A/J mice were immunized by the application of 0.1 ml of a 5% solution of picryl chloride in ethanol to the skin of the abdomen; the day of immunization is referred to as day 0; the animals a single dose of 60 ~tg of rGK in 200 ~tl of PBS on the days indicated. b. Animals were tested with the application of a 1% solution of picryl chloride in olive oil on one ear. Results are expressed in terms of the increase if the challenged ear over the pre-challenge values using units equivalent to 1/10 ram. * p < 0.05.
Figure 6 shows the results of an experiment in the collagen arthritis (CA) model in rats. A single injection of GK given at the time when the arthritic reaction begins to flare induced an almost complete suppression which lasted 4-5 days, followed by the return to central levels of arthritis. In contrast, repeated injections maintained suppression for the entire duration of the experiment. Thus, this experiment confirmed that the effects of GK in immune animals are short lived and do not reduce the state of immunity of the animals. The effects described up to this point refer to in vitro phenomena or to the subcutaneous administration of GK. The fact that GK is found in high concentration in salivary glands and is secreted in saliva suggests that a significant physiological function of GK may occur following external salivary secretion. For this reason, we tested the effects of orally administered GK in rats. Experiments of this nature were carried out in rats using the collagen arthritis model. This involves the injection of Type II collagen in an oil based adjuvant which induces an arthritic reaction beginning two weeks after immunization, which lasts for the next 3-4 weeks.
232
10 x --
8 6
9 w,,,,,i
4
i
"
12
18
9
24
u
30
9
36
Days Figure 6. Effects of a single (V, day 14) or multiple ( I , days 14, 18, and 24) injections of a semi-purified preparation of rGK in the CA model. Controls (O) received PBS only.
The variables we tested included the effects of a pre-treatment with oral collagen to induce tolerance to the subsequent immunization; the effects of the surgical removal of the submandibular gland; and the effect of oral administration of GK, given before tolerization and/or before immunization to normal and sialoadenectomized rats. Figure 6 demonstrates the results of one of such experiments. The oral pre-treatment with Type II collagen significantly reduced the arthritic response in normal rats but had no such effect in sialoadenectomized animals. In contrast, the oral administration of GK significantly reduced the arthritic reaction. Figure 7 shows the results of another experiment which confirmed that sialoadenectomy interferes with tolerance induction. Moreover, this experiment demonstrated that the oral administration of GK to sialoadenectomized rats restores the ability of the mucosal immune system of these animals to develop tolerance upon oral collagen administration. These results point to similarities and differences in the action of GK depending on whether it is used in vitro or in vivo, and whether it is injected or given orally. The most prominent in vitro effect was to induce stimulation of lymphocyte proliferation, while the most prominent in vivo effect was an immunosuppressive one. One hypothesis to explain this apparent discrepancy in vivo suggests that immune deviation occurs following administration of GK in vivo, involving a reduction in the responses we assessed, but a stimulation of other responses which we did not resume. One example of such an immune deviation would be an altered balance of TH1 to TH2 responses with a decrease of TH1 activity. The responses we found to be suppressed by GK treatment were cell- mediated ones (DTH, allograft rejection and CA) or T-dependent IgM production (direct PFC response). These responses would be suppressed by any mechanisms that reduce TH1 activity or favor the switch from IgM to any other immunoglobulin class. This explanation is consistent with the observation that oral GK favors the induction of oral tolerance, a reaction thought to involve a deviation from cell-mediated responses to IgA production and induced by increased activity of TGF- (and IL-4) producing
233
4.5
A
IE IE nr"
uJ I--
4.0
I..U
< a
3.5
o lo
2'0 DAYS AFTER IMMUNIZATION
Figure 7. Effects of SMX and semi-purified oral rGK on the arthritic response and on the induction of oral tolerance in Sprague Dawley rats. (O) immunized controls; ( 9 oral collagen followed by immunization; (A) SMX followed by immunization; (m) SMX, oral collagen, followed by immunization; (@) oral rGK, followed by immunization.
TH3_ type cells. Alternately suppressor cells may be activated by GK. Yet another possible explanation for the apparent discrepancy of in vitro versus in vivo effects would suggest that in vitro stimulation of lymphocyte proliferation was due to the formation of stimulatory peptides, formed by the action of GK on some substrate(s) contained in the foetal calf serum present in our cultures. Under in vivo conditions, the proteolytic action of GK may result in decreased concentration of the same substrate(s) and in a deficient production of stimulatory peptides for a few days, until the substrates are reconstituted in full. Differences in GK action were also observed when the route of GK administration changed from subcutaneous to oral. In both situations, some suppression of immune reactions was observed if the treatment was simultaneous with or shortly preceded, antigen administration. On the other hand, the oral administration of GK appeared to enhance the induction of tolerance if given together with the oral antigen. The ensuing suppression of the arthritic reaction exceeded in duration and in magnitude the "immunosuppressive" effect of GK alone, administered either by injection or by mouth. The same experiments also demonstrated that oral tolerance could not be induced by antigen alone in sialoadenectomized animals. Oral tolerance is a state of antigen specific hyporesponsiveness which follows oral delivery of an antigen. It represents a protective reaction by the gut associated mucosal tissue (GALT) to prevent unnecessary and potentially harmful reactions to food antigens. It involves more than one mechanism. High antigen doses induce clonal deletion, while lower doses induce an active form of suppression or immune deviation, mediated by TGF- and IL-4 producing T cells,
234
6.5
6.0
E E v nLU i'LU
5.5
< a
.,,..=
n<<:: 5.0
4.5 4.3
..j/; 0 14 1'61'8 2'1'2'3;~5 28 31
,g :35:36
4'2 Days
DAYS AFTER IMMUNIZATION Figure 8. Effects of SMX and oral pGK on the induction of oral tolerance in Lewis rats. (0) immunized controls; (O) oral collagen followed by immunization; (A) SMX followed by immunization; (I) SMX, oral collagen, followed by immunization; (I-]) SMX, oral collagen, oral pGK, followed by immunization.
referred to by some authors as TH3 cells. The action of these cells suppresses TH1 responses and favors TH2 responses [17-19]. From Peyer's Patches, these regulatory cells migrate to periferal lymphoid organs rendering systemic the modified responsiveness. Thus, our results suggest that the SM gland plays a significant role in maintaining the normal responsiveness of GALT and that GK secretion in saliva is one of the factors involved in this function. The immunological properties of GK make this molecule an interesting candidate in the treatment of autoimmune diseases.
ACKNOWLEDGMENTS The experiments described here were carried out in collaboration with A. Kemp, L. Mellow, M. Abdelhaleem (experiments with crude SM gland extracts and with semi-purified rGK). Financial support was from the Medical Reasearch Council of Canada.
235
REFERENCES
.
.
.
.
.
.
10.
11. 12. 13. 14. 15. 16. 17.
18. 19.
Fuller PJ, Funder JW. The cellular physiology of glandular kallikrein. Kidney 1986; 29: 953. Wines DR, Brady JM, Pritchett DB, Roberts JL, MacDonald RJ. Organization and expression of the rat kallikrein gene family. J Biol Chem 1989; 264: 7653. Bhoola KD, Figueroa CD, Worthy K. Bioregulation of kinins: kallikreins, kininogens, and kininases. Pharmacol Rev 1992; 44: 1. Alhenc-Gelas F, Marchetti J, Allegrini J, Corvol P, Menanrd J. Measurement of urinary kallikrein activity species differences in kinin production. Biochim Biophys. Acta 1981; 677: 477. Simpson JAV, Chao J, Chao L. Localization of kallikrein gene family proteases in rat tissues. Agents and Actions-Suppl 1992; 38 (Pt. 1): 595. Naughton MA, Geczy C, Bender V, Hoffman H, Hamilton E. Esteropeptidase and thymotropic activity of a protein isolated from the mouse submaxillary gland. B iochim Biophys Acta 1972; 263: 106. Hu ZQ, Murakami K, Ikigai H, Shimamura T. Enhancement of lymphocyte proliferation by mouse glandular kallikrein. Immunol Lett 1992; 32: 85. Perris AD, Whitfield JF. The mitogenic action of bradykinin on thymic lymphocytes and its dependence on calcium. Proc Soc Exp Biol Med 1969; 130: 1198. Ishizaka K. Twenty years with IgE: from the identification of IgE to regulatory factors for the IgE response. J Immunol 1985; 135: i-x. Verwaerde C, Auriault C, Neyrinck JL, Capron A. Properties of serine proteases of Schistosoma mansoni. Shistosomula involved in the regulation of IgE synthesis. Scand J Immunol 1988; 27: 17. Matsushita S, Katz DH. B iphasic effect of kallikrein on IgE and IgG1 syntheses by LPS/IL4 stimulated B cells. Cell Immunol 1993; 146: 210. Kemp A, Mellow L, Sabbadini E. Suppression and enhancement of in vitro lymphocyte reactivity by factors in rat submandibular gland extracts. Immunology 1985; 56: 261. Abdelhaleem M, Sabbadini E. Identification of immunosuppressive fractions from the rat submandibular salivary gland. Immunology 1992; 76: 331. Nagy E, Berczi I, Sabbadini E. Immunoregulatory effects of glandular kallikrein from the salivary submandibular gland of rats. Neuroimmunomodulation 1997; 4: 107. Edman P, Begg G. A protein sequenator. Eur J B iochem 1967; 1: 80. Trautschold I. Assay methods in the kinin system. In Erdos G: Handbook of experimental pharmacology, Bradikinin, kallidin, and kallikrein. Berlin, Springer, 1970; 25: 52-81. Weiner HL, Friedman A, Miller A, et al. Oral tolerance: immunologic mechanisms and treatment of animal and human organ-specific autoimmune diseases by oral administration of autoantigens. Annu Rev Immunol 1994; 12: 809. Weiner HL. Oral tolerance: immune mechanisms and treatment of autoimmune diseases. Immunol Today 1997; 18: 335. Friedman A, Weiner HL. Induction of anergy or active suppression following oral tolerance is determined by antigen dosage. Proc Natl Acad Sci 1994; 91: 6688.
This Page Intentionally Left Blank
New Foundation of Biology
237
Edited by I. Berczi and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
Understanding Classical Conditioning of Immune Responses
REGINALD M. GORCZYNSKI
Departments of Surgery & Immunology, University of Toronto and The University Health Network, CCRW 2-855, The Toronto Hospital, 200 Elizabeth Street, Toronto, ON., Canada M5G-2C5
ABSTRACT There is now a large body of data showing quite convincingly that a variety of immune responses can be regulated using a classical (Pavlovian) conditioning, taste aversion, paradigm. Amongst the models used are those described initially by Ader, Cohen and co-workers, studying suppression of antibody responses in rats exposed to saccharin in their drinking water, following trials of repeated exposure to saccharin in association with the non-specific immunosuppressant cyclophosphamide [1]; an augmented frequency of cytotoxic cells (CTL) following sham skin grafting in mice receiving pre-test trials of repeated skin allografts [2]; environmentally "cued" inhibition of antibody responses following rotational stress-induced immunosuppression in mice [3]; conditioned degranulation of mucosal mast cells; and altered NK activity in mice again receiving gustatory stimuli paired with an NK inducing stimulus [3, 4]. In a number of such model systems researchers have used these conditioned responses to explore immune regulation in clinically relevant disorders, including control of tumor growth [4]; autoimmunity [5]; and the process of aging [6]. Some effort has been expended in trying to understand the mechanism(s) behind conditioning of immune responses. Our group reported evidence favouring both an altered environment (in the conditioned animal), and some change in the responder cell pool which might explain these findings [6]. We predicted that at least some of the effects seen might reflect an alteration in recirculation of responder cells, and there is now also evidence to support this hypothesis [7]. Others have focused their attention on alterations in the molecular milieu which might contribute to conditioned immunoregulation. There is evidence that neuropeptides, neurohormones, conventional neurotransmitters and neurotrophins, growth factors which have been documented to regulate the development of nervous tissue cells, also participate in immune regulation [8-12], though their role(s) in conditioning of immunity remains to be explored in detail. Following an interest in taste aversion conditioning paradigms and oral immunization [13], my laboratory has explored cytokine production (TNFc~, IL-1) following conditioned oral exposure to LPS. Conditioned induction of cytokines occurred in parallel with augmented local (in the gut) and distal (in the brain) expression of mRNAs for substance P (SP) and somatostatin (SOM). Antagonists of the latter molecules blocked the increased brain mRNA cytokine
238
expression in conditioned mice, while systemic administration of anti-cytokine antibodies did not produce inhibition. We conclude that neuropeptides, rather than the cytokines per se, are of key importance in the regulatory loop producing altered expression of mRNAs for certain cytokines in the CNS.
INTRODUCTION: MULTIPLE LINES OF EVIDENCE FOR AN INTERACTION BETWEEN THE CNS AND THE IMMUNE SYSTEM 1.1.
Neurotransmitters and neurohormones in immunity
For many years conventional wisdom was that the immune system and the nervous system were independent of one another for their normal functioning [reviewed in [14, 15]]. Data showing that cells of the immune system express receptors for a number of CNS-active molecules, including catecholamines, histamine, endorphins etc. challenged this idea, and led to speculation that this may not be the case under normal physiological conditions [9-11, 16]. Indeed some early studies by Sklar and Anisman documented a correlation between catecholamine and steroid stress hormone levels in animals experiencing inescapable or escapable shock with growth of a transplantable tumor implant [17]. Further extensions to these studies came from analyses of the functional significance of sympathetic innervation of lymphoid organs [18-20], and the application of the information so obtained to investigations of an animal model of multiple sclerosis (MS), experimental allergic encephalomyelitis (EAE) [21], as well as to clinical studies in MS patients themselves [22]. In addition to affecting the peripheral immune system, sympathetic innervation has been documented to be of importance for the development of mucosal immunity [23]. For an extensive review of the role played by neuropeptides and other neurotransmitters in the interactions of the CNS and immune system the reader is referred to Goetzl and Spector, 1989 [8]-see also below. In addition to conventional neurohormones and neurotransmitters, molecules known to be produced primarily by cells within the lymphoid system, cytokines and chemokines, have also been found both to be produced within the CNS (e.g. by glial cells and astrocytes, as well as conventional lymphocytes within the CNS) and, in the case of IL-1 and TNFc~ [24], to modulate the development of nerve tissue cells. Responses of nervous tissue cells can also be regulated by cytokines, and as but one example of the latter, IL-4, IL-10 and TGF[3 have recently been documented to modify LPS induction of inflammatory cytokines by glial cells [25]. 1.2.
Neuropeptides and immunity: the role of cytokine changes
As already indicated, there is now abundant evidence that not only nerve tissue, but also lymphoid tissue, can express mRNAs for a number of different neuropeptides. Included in these lymphoid tissues are those known to be crucial for the initiation of most immune responses, namely macrophages and dendritic cells, sparking interest in the notion that neuropeptides might be implicated not merely in modulation of ongoing immune responses, but in their initation. Throsby, analysing expression of a number of neuropeptides, including prosomatostatin polypeptide in thymic extracts, thymic cells, and thymic cell lines from C57BL/6 mice (neonatal and 2-, 4-, and 8-week-old animals), documented a 10-to 20-fold enrichment in expression of peptides in low buoyant density cells, shown to stain for the surface markers F4/80 (macrophage) or DEC205 (dendritic cells) [26]. Neuropeptides were differentially expressed by dendritic cells and macrophages in these normal mice. In a separate anatomical
239
study, the relationships between immunocytochemically identified nerve fibers and MHC class II-expressing antigen presenting dendritic cells were investigated in the rat hepatobiliary system [27]. Contacts between nerve fibers staining for SP, calcitonin generelated peptide, calretinin, and vasoactive intestinal polypeptide (VIP) and dendritic cells were observed. The authors concluded that antigen presentation is modulated in this tissue by the autonomic nervous system. In gut tissue, the neuropeptides SP, VIP and somatostation (SOM) are present in the nerve endings. SP is also present in nerve endings in the skin and SP is thought to be present at abnormal concentrations in atopic dermatitis (AD) patients, which have also been reported to show a marked imbalance in type-1 vs. type-2 cytokine production, believed in turn to be responsible for some of the pathophysiology in this disease. In a study designed to investigate the effect of neuropeptides themselves on cytokine profiles, SP was found to have an enhancing effect on production of both IFNT and IL-4 at physiological concentrations (10 -1~ 10 -6 mol/L), while VIP had inhibitory effects over this same range [28]. In an analogous study in a contact hypersensitivity model, an SP agonist, GR73632 or delta-Aminovaleryl (Pro (9), N-Me-Leu (10))-substance P7-11 and an SP antagonist, spantide I, were injected intradermally to modify contact hypersensitivity (CH) to locally applied haptens. The SP agonist enhanced CH induced by conventional, but not optimal, sensitizing doses of hapten, while the SP antagonist inhibited the induction of CH by optimal sensitizing doses of hapten. These and other data were taken to infer once again that SP agonist enhanced the generation of hapten-specific immunogenic signals from the dermis, though the role of local cytokine production in this model was not investigated [29]. Given that at least some of the immunomodulatory effects produced by macrophages are in turn a reflection of immunomodulation occurring secondary to altered release of low molecular weight molecules from those cells (reactive oxygen and nitrogen intermediates), some studies have focused on the effect of neuropeptide mediators on inducible nitric oxide synthase (iNOS) activity in macrophages. In LPS-activated macrophages, SP stimulated NO production in time and concentration dependent manners, which were blocked by a specific receptor antagonist [30]. Moreover, SP stimulation increased the levels of both iNOS mRNA and iNOS protein, documenting that SP can increase LPS induced NO production in macrophages by augmenting the induction of iNOS expression. In an extension of these data this group examined the role of SP on acute-cold stress induced altered production of NO by peritoneal macrophages, finding that SP enhanced the LPSinduced macrophages NO production from stressed mice relative to the non-stressed mice [30]. These results suggest that SP may have an important modulatory role in production of NO by macrophages, which may then be in part, at least, responsible for some of the changes documented by the other groups cited above. Is there any independent evidence that SP, or other neuropeptides, is implicated under physiological conditions in gastrointestinal immunity? Mast cell hyperplasia in the gut is a feature of inflammatory bowel disease (IBD), but the role of mast cells in this disease is still unclear. Since it is believed that mast cell-nerve interactions might impact on intestinal inflammation, one group has investigated whether SP could cause histamine secretion from human gut mucosal mast cells [31]. No induction of mast cell activation was seen in histologically normal mucosa from controls, while mucosal specimens taken from inflamed IBD tissue or from uninvolved Crohn's disease tissue showed enhanced histamine secretion in the presence of SP, either alone or in combination with anti-IgE. These data argue that mast cell-nerve interactions are indeed involved in histamine-releasing processes in the gut in IBD.
240
1.3.
Neurotrophins in CNS: immune system interactions
More recently the field of molecular communication between the CNS and the immune system has been expanded further by documentation that neurotrophins, structurally related growth factors which include nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF) and neutotrophins-3 and -4 (NT-3, NT-4), all of which have been reported to play roles in development, differentiation and survival of neuronal subsets, can also be produced by, and can act upon, cells within the immune system. Under inflammatory conditions, for example, BDNF synthesis has been detected in activated T and B cells, macrophages and neurons [32]. NGF has been shown to promote growth and differentiation of mast cells and basophils [33], and a receptor for NGF, tyrosine kinase receptor B (trkB) has been found to be expressed predominantly in neonatal thymic tissue [34], consistent with reports that neurotrophin transcripts were highest during fetal life. Analysis of the effect of NGF on responses of mature lymphocytes has provided evidence that it may modulate the threshold of response to conventional immunologic stimuli, including those inducing IgE synthesis and type-2 cytokines [35, 36]. Finally, there are a number of reports concerning altered levels of NGF and other neurotrophins in a variety of autoimmune and/or inflammatory diseases, including systemic lupus erythematosus (SLE) and multiple sclerosis (MS) [32], as well as in bronchiolar lavage fluid of asthmatic patients after allergic challenge [37]. 1.4.
Behavioural changes and altered immune responses
Perhaps even more provocative than these studies, however, have been those which have focused directly on the ability of behavioural changes themselves to modulate immunity. In one of the more recent and striking examples of such analyses, Heijnen et al have explored the induction of EAE in rat strains preselected for their response to apomorphine as a selection criterion (so-called APO-sus and APO-unsus). The latter rats were also found to have a lower response to novelty stress as determined by lower locomotor activity and decreased activation of the hypothalamic-pituitary-adrenal (HPA) axis. In addition, these rats were susceptible to induction of experimental allergic encephalomyelitis (EAE) after immunization with myelin basic protein in Freund's Adjuvant, while APO-sus rats were not [38]. Since EAE is believed to be a type-1 cytokine mediated phenomenon, it was of interest to find also that APO-sus rat cells in vitro were more likely to develop type-2 cytokine responses than cells from APO-unsus rats [38, 39]. A disparity in type-1 vs. type-2 cytokine production has been observed in schizophrenia, though the relationship of such changes to disease remains unexplored [40]. My own laboratory has concentrated on an alternative model of behavioural regulation of immunity, namely that which follows application of a classical conditioning paradigm to immune responses. These studies were spurred by the initial findings of Ader and Cohen, using a taste aversion model in rats. Animals which had received immunization with sheep erythrocytes (SRBC) initially in the context of the unconditioned stimulus cyclophosphamide (a non-specific immunosuppressant) and a conditioned stimulus, saccharin, subsequently demonstrated decreased antibody responses to saccharin after re-exposure to SRBC in the context of saccharin [1]. This suppressed response occurred independently of triggering of the HPA axis. The same group went on to report that the conditioned immune response could be extinguished by repeated exposure to the conditioned stimulus in the absence of the unconditioned stimulus (an antigen challenge). A number of groups, including my own, were able to reproduce these results in a variety of models, in which conditioned NK cell activation [3, 4] and even skin graft rejection responses were studied [2]. There have in addition
241
been a number of investigators who have applied this conditioning paradigm to manipulate responses in animal models of clinical disease, including (SLE) [5], malignancy [3], and infectious disease [12]. It is important to note that conditioned compensatory responses (rather than attenuated responses) have often been observed using similar paradigms to those used by other groups to observe conditioned immunosuppression [41]. This may in part reflect the physiological operation of adaptive drug responses, as discussed at length by Eikelboom and Stewart [42], or represent the less paradoxical triggering of counter immunoregulatory responses seen frequently in immunity.
MECHANISM(S) IMPLICATED IN CONDITIONING OF IMMUNE RESPONSES IN ANIMALS 2.1.
Environmental vs. cell-derived changes in conditioned mice
Following documentation of the phenomenology of conditioning of immune responses, it was natural that there should follow studies designed to explore the mechanism(s) behind the effects observed. My own laboratory adopted a conventional immunological approach to this problem, asking initially whether conditioning could be adoptively transferred to naive animals by cells, or whether the environment of the conditioned animal was necessary for documentation of a conditioned response. Perhaps not surprisingly we observed that there was evidence for both being important to observe a conditioned immune suppression, using animals in which conditioning had been achieved using the SRBC/cyclophosphamide/saccharin model described by Ader and Cohen [43]. Moreover, alterations in both cells and/or environmental factors seemed to underly the poorer conditioned response observed when aged animals were used in these studies [6]. One explanation for the observation that there was something unique about the cells in conditioned mice which might help explain conditioning phenomena lies in the data that lymphocytes (and/or other cells of the immune system) can express receptors for molecules known to be products of cells in the nervous system (see above), including neuro-transmitters, neurohormones, molecules of the HPA axis, members of the NGF family etc. Altered expression of such receptors could help explain altered immunity in conditioned mice. In addition, since immune responses generally depend upon cell: cell interactions, we speculated that alterations in cell trafficking in conditioned mice, perhaps a reflection of altered expression of so-called lymphocyte homing receptors (integrins/selectins etc.) might also help explain altered immune responses in these animals [44-46]. Note that altered release of chemoattractants in conditioned mice (chemokines-see below) might also contribute to differences in cell migration in these animals. Recent data has now confirmed that CNS control of cell migration is likely a fundamentally important mechanism for integration of the CNS: immune system axis [7, 47]. 2.2.
The role of altered chemokine and/or cytokines production in conditioning
Other molecules are also implicated in regulating immunity, in particular chemokines and cytokines. While it had been thought that chemokines function primarily to regulate cell trafticking [48-50], it is now clear that their function is considerably more complex than this. There is a large body of evidence implicating their importance in HIV infection [51 ]; in cell activation [52-54]; in autoimmune dieases and neoplasia [55, 56]; and even in atherosclerosis [57].
242
A large body of data now indicates clearly that synthesis of these molecules is not restricted to cells of the immune system [58], and indeed that some chemokines and chemokine receptors have a significant impact on the CNS. While initially these latter models involved studying chemokines involved in the pathogenesis of important neuroinflammatory diseases, ranging from multiple sclerosis and stroke to HIV encephalopathy, more-recent studies indicate that, in addition to their role in pathological states, chemokines and their receptors have an important role in cellular communication in the developing and the normal adult CNS. Thus, stromal-cell-derived factor-l, which is synthesized constitutively in the developing brain, seems to play an obligatory role in neurone migration during the formation of the granule-cell layer of the cerebellum. In addition, many other chemokines are capable of directly regulating signal-transduction pathways involved in a variety of cellular functions within the CNS, ranging from synaptic transmission to growth [59]. The potential role of altered chemokine synthesis in classical conditioning has yet to be investigated. In a similar fashion to the complex nature by which altered expression of chemokines and chemokine receptors might affect immune responses, there is a wealth of data concerning immunoregulation occurring as a function of altered cytokine production, subsequently signaling cells through JAK/STAT pathways [60]. Much of this work was sparked by the observations of Mossmann et al, who suggested that CD4 + cells could be subdivided into different subsets, so-called type-1 vs. type-2 cytokine producing cells, according to the pattern of cytokines they produce [61, 62]. The former synthesized mainly IL-2, IFN~,, TNFc~, while the latter produced mainly IL-4, IL-10 and IL-13. It is now evident that this polarization is not as clear-cut as we might first have hoped [63]; for example, further definition of a TGF[3/IL-10 producing type-3 CD4 + cells has been postulated by Weiner [64]. Nevertheless, interest was spurred by the observation that polarization of cytokine production seemed to correlate with distinct clinically relevant conditions [65]. Thus tolerance to tissue allografts seemed distinguishable by type-2 cytokine production [66-68]; inflammatory cytokines are implicated in a number of autoimmune conditions [69]; and polarization to type-2 cytokine production was one of the early hallmarks of chronic infection with cutaneous leishmaniasis [70]. In terms of considering how cytokines might be implicated in CNS: immune system interactions, it is important to note that there is independent evidence that a number of cytokines, including IL-1 and TNFc~, can affect the HPA axis [71], even following peripheral cytokine synthesis. Furthermore, and complementary to the data discussed above concerning a potential role for chemokines in CNS: immune system interactions, there are a number of studies indicating that cytokines can affect tissue chemokine production [72], including CNS tissue [73]. Perhaps even more striking are recent reports implicating CNS production of the chemokine MCP-1 in the type-1 cytokine production contributing to the pathophysiology of EAE [74]. This interaction between MCP-1 and subsequent cytokine production has been confirmed by my own group in a model of xenotransplantation [75], though whether such changes underly alterations seen in conditioned mice remains open to investigation. Few studies have examined direct evidence for altered cytokine production in conditioned mice. We were able to document in orally immunized mice that following exposure to a conditioned stimulus alone altered cytokine production was seen, along with diminished antibody synthesis [13]. In a classical model of murine leishmaniasis, where as indicated above, type-2 cytokine production is believed to be critically involved in persistence of parasitic infection, modification of cytokine production was achieved using a conditioning paradigm of saccharin paired with cyclophosphamide. We were then able to show that re-exposure to (immunologically) inert cues could subsequently dramatically affect parasite growth, in a manner entirely consistent with the cytokine production in conditioned animals [12].
243
In the studies that follow, the mechanism(s) which might be brought into play to regulate cytokine production in animals conditioned to respond to oral immunization have been investigated in more detail.
SPECIFIC STUDIES ANALYZING A REGULATORY ROLE FOR NEUROPEPTIDES AND/OR CYTOKINES IN CONDITIONING OF ORALLY IMMUNIZED MICE 3.1.
A conditioned local, and systemic production of cytokines following oral LPS
In the model system used a (conditioned: UCS+CS) group of 30 C57BL/6 mice were exposed to oral LPS (300 mg/mouse-an unconditioned stimulus, UCS), with additional exposure to chocolate milk in the drinking supply (conditioned stimulus, CS). A control group received the UCS alone. After 3 such trials at 21 day intervals, groups of 10 conditioned mice (UCS+CS) now received either the UCS alone, the CS alone, or water only (referred to as "NIL" or "H20" in subsequent Figures). 10 mice previously receiving the UCS alone received exposure to the CS on the test trial. A further control group (untreated) received no exposure to UCS or CS at any time. Beginning at 1 hour post exposure, and at 6, 24 and 48 hours thereafter, blood was sampled from the tail of individual animals, and serum measurements of TNFot performed using a bioassay (inhibition of growth of Wehi 1643 cells, as assessed by proliferation in vitro by addition of 3HTdR after 40 hours). The assay was standardized using recombinant mouse TNFcz (obtained from Endogen, USA). Typical data for such a study are shown in Figure 1. It is clear that following conditioning trials, re-exposure to the CS alone, but not water, does indeed lead to serum production of TNFot to levels nearly 50% of those induced by re-exposure to the UCS (LPS). This induction of TNFcz is not seen in unconditioned mice (CS in Figure). Peak production of TNFez occurs in both the UCS and CS groups by some 24-48 hours post exposure (to either the UCS or CS). In further studies we examined the source of TNF~z produced in these mice, asking whether mesenteric lymph node cells (MLN) or peripheral blood cells (PBL) taken from mice at 6 or 24 hours after exposure to the UCS or CS would spontaneously release TNFcz in vitro, in the absence of further stimulation. Once again the same control groups were used as in Figure 1. These data are shown in Figure 2, panels a and b. They document that 2 x 106 cells/culture of either PBL and MLN from conditioned mice exposed to the CS spontaneously produce some 100-300 pg/ml of TNFot in vitro, levels again some 50% of control levels produced by cells from conditioned mice exposed to the UCS. 3.2.
Altered CNS expression of cytokine mRNAs in conditioned mice
As noted above, there is now evidence suggesting that a number of cytokines, released peripherally, and within the CNS, can modify CNS signaling, at least in part by effects on the HPA axis [71]. There are other data confirming a critical role for CNS synthesized cytokines, in particular IL-1 and TNFot, in a number of phenomena, including nerve tissue development and sleep (see review by Moldofsky et al. in this volume). In order to investigate whether CNS induction of cytokines also occurred in conditioned mice re-exposed to conditioning cues, we used animals conditioned with LPS in the paradigm described in Figures 1 and 2, and at different time following challenge with the UCS alone (LPS) or the CS alone, we sacrificed mice and snap froze brain samples. These same animals were used for some of the data shown
244
Stimulation of TNF~ production by conditioned stimulus 3000
m
~-~=~ T N F ~ - t 1 ~ TNF~-t6
O
o
qp,,
Q.
2000
c =m
~
TNF~-t24
~
TNFc~-t48
Z
k.E
lOOO
u~ Untreated
UCS
CS
Nil
CS (UCS§
T r e a t m e n t of mice (31group) Figure 1. 0 mice/group were conditioned (UCS+CS) as described in the text, and thereafter received UCS, CS or water (NIL). Unconditioned mice (exposed to UCS only throughout) received the CS at trial. Untreated mice received water only throughout.
Stirr~alion of TNFc~production in PBL of conditioned mice
S'dnulation of TNF~ production in MLN of condilJoned mice 500
400-
,oo ,1~
_[ '
~ I
I
"g'qFo~-t6
"lN:~-t6 r--3 Tl~o~-t24
300-
1T1~or
1
200
,ool
1oo o
Untreated
UCS
CSonly
NIL
CS(UCS+CS)
Treatrrent of mice (3/group)
Untreated
UCS
Trea~ett
CS0nly
of rrice
NIL
CS(UCS+CS)
(3#~~)
Figure 2. Spontaneous production of TNFc~ in MLN/PBL from groups of mice shown in Figure 1.
in Figure 2. RNA was extracted from brain tissue, cDNA prepared, and the latter tested at serial 4-fold dilutions in PCR with 32p dATP and primer pairs for IL-1, TNFct and GAPDH (control primer set). Data are shown in Figure 3, after normalizing all samples to an equivalent signal for
245
GAPDH. Data to the far left of the Figure show the signal obtained from tissue of (UCS+CS) mice re-exposed to water only (NIL in previous Figures)-only data for the 0 hour time point is shown, but equivalent data was obtained at all time points for this control tissue. Interestingly, the conditioning paradigm used clearly led to induction of mRNA for IL-1 and TNFct in the CNS, as well as in peripheral tissue (see Figures 1 and 2). This induction occurred rapidly for IL-1 (peak 12-24 hours), with a significantly slower induction for TNF~t (peak 48-72 hours). Once again peak levels induced following conditioning were comparable (mRNA levels~10-fold less) to those seen following exposure to the UCS alone, and significantly greater (by 20 to 30-fold) than those seen in unexposed (to conditioning cues) conditioned mice. We also found (data not shown) similar differences in mRNA induction for IL-1 and TNFct in gut tissue in conditioned vs. unconditioned mice.
Conditioning induces cytokine mRNA (by PCR)in brain LPS challenge alone
m r
= n,, tJ
a.
6
I
~ I L - 1
t--
CS
t~
->9
I TNFct
5 CS+UCS)
4
r O
"-e.-
3
O .m
_= .=.
"o
2
z
c3 e~
Untreated
1
O --.I
0,
6
12
24
48
72
6
12
24
48
72
0
Time (hrs)after challenge with LPS (UCS) or saccharin (S) Figure 3. mRNA (by quantitative PCR) for TNFcdlL-1 in brain tissue of conditioned mice.
3.3.
Altered expression of mRNAs for neuropeptides precedes changes in cytokines
It has been shown that neurons of the autonomic nervous system innervating the gastrointestinal tract contain a number of distinct peptide mediators, including SP, SOM and VIP, and it has been postulated that SOM and SP exert mutually antagonistic functions, at least in some mucosal tissue, by exerting opposing effects on mediator release by mast cells [76]. Furthermore, it has been reported that SP, in nanomolar concentrations, can stimulate IL-1 and TNFot production from human PBL [77]. In the following study mRNA levels for SP and SOM in gut tissue of conditioned mice (see previous Figures) were investigated at 18 hours following exposure to the UCS or CS. Marked induction of expression of both SOM and SP mRNAs was observed both in UCS and CS exposed conditioned mice, but not in conditioned animals given water to drink or in non-conditioned mice exposed to conditioned cues (panel a, Figure 4).
246
Increased mRNA levels were seen when quantitative PCR analysis of brain tissue was assayed (panel b, Figure 4). 3.4.
Altered neuropeptide expression regulates changes in cytokine expression
In an attempt to examine the causal, rather than mere temporal, relationship between altered expression of cytokines and neuropeptides peripherally, and in the CNS, in conditioned mice, we exposed conditioned mice to cues as before (UCS or CS), and in addition treated them with mAbs to IL-1 or TNFc~, or with known antagonists of SOM and SP. In these latter cases mAbs (250 ~tg/mouse) or peptide antagonists (50 lag/mouse) were injected intravenously to animals immediately before challenge with CS or UCS. Animals were sacrificed as before and tissue extracted for mRNA analysis (PCR) and for TNFc~ cytokine expression (by bioassay). Data shown in Figure 5, panel a, clearly document that SP antagonists, but not SOM antagonists, block CNS induction of cytokine mRNA following either UCS or CS exposure-SOM antagonists led to a small increase in mRNA levels, consistent with the notion put forward earlier that SP and SOM may act as mutual antagonists in this system. In contrast (see panel b, Figure 5), mAbs to IL-1 or TNFc~ have no effect on increased CNS expression of cytokine mRNAs following exposure to UCS or CS. Thus it seems that altered CNS expression of cytokine mRNAs even following administration of the UCS in conditioned mice is brought about by perturbations in the expression of the neuropeptides SP and SOM, and does not reflect altered CNS signaling occurring secondary to peripheral changes in cytokine express ion/production. 3.5.
Altered intracellular signaling after exposure to a UCS or CS in conditioned mice
Mice and/or cells pretreated with low doses of LPS show a diminished production of cytokines, most noteably TNFc~, rather than, for instance IL-1, on re-exposure to LPS [78]. A number of groups have examined the mechanism for this tolerance, and found that it occurs via signaling downstream of binding to the LPS receptor, and before the intracellular activation of MAP kinase and NFkB [79, 80]. There are reports that this tolerance is itself mediated via production of TGF[3 and/or IL-10 [81], and that it can be blocked by blocking PI3 kinase signaling (by wortmannin), but not by inhibition of protein tyrosine kinases (PTK), using genestein or herbimycin A, or by blocking PKC activity, using staurosporine or bisindolylmaleimimde [82]. More recently LPS tolerance has been associated with a decreased expression of the pattern recognition receptor of innate immunity, Toll-like receptor 4 (TLR4), on the cell surface [83]. In the experiments described earlier, stimulation of TNFc~ production by the UCS or CS in conditioned mice seemed to be equivalent in all assays tested, with the exception that the CS induced quantitatively less TNFc~. In order to investigate whether the cell signaling induced by the CS in conditioned animals really was equivalent to that provided by LPS itself (the UCS), conditioned (or control) mice were exposed to the UCS or CS as earlier, and 24 hours later MLN cells were prepared from these animals and stimulated in vitro with LPS (500 ng/ml), in the presence or absence of wortmannin (0.3 ~tM). TNFot production in the supernatant was assayed at 18 hours, with data shown in Figure 6. As expected, there was no difference in TNFot production in all control groups used: cells from naive mice, from mice receiving only the CS thoughout, or conditioned mice allowed only water to drink (NIL). Cells from conditioned mice rechallenged with the UCS (LPS) showed a 3-4 fold diminished production of TNFot on stimulation with LPS (tolerance), which was abolished by concomitant culture
247
mRNAs induced in gut tissue by conditioning me ucs (after UCS+CS) CS only
mRNAs induced in brain tissue following conditioning 5
-1
UCS (alter UCS+CS)
[
=== CS only
9HzO only
H20 only
i~ CS (z[fter UCS+CS)
GAPDH
Somatostatin
Substance P
E ~ CS (after UCS+CS)
L~APUH
mRNAs detected in assay
Somatostatin
Substance P
rnRNAs detected in assay
Figure 4. mRNA expression (quantitative PCR) for SP or SOM in gut/brain tissue of conditioned mice.
Substance P (somat0s~n) anatagonists block induction of cytokine ~ in brain UCS ( a ~ CS+UCS)
g
Failure of anti-TNF'~ or anti-IL-1 to block conditioned induction of cytokine ~ in brain i UC~ (afterCSKr
c s (~t~ c s . t ~ )
l
T
9
.T_ J._
'B
None SubP Som 50pg a n a t ~
None SubP Som
None SubP Som" Note TNF I~L-1
injected iv immdial~ly before challenge
None TNF IL-1
None TNF IL-1
250~g antibodies injected iv imTec:ie~ly before challenge
Figure 5. SP anatagonists, but not SOM antagonists or mAbs to TNFcdIL-1, block conditioning.
with wortmannin. Cells from conditioned ice rechallenged with the CS also showed a tolerance effect to deliberate stimulation with LPS, but this was unmodified by culture with wortmannin. This suggests that intracellular signaling elicited by the CS w a s n o t e q u i v a l e n t to that triggered by the UCS in this model.
248
2000 0 0
LPS stimulation of TNF(~ production following conditioning I"-'1 Wortmannin No drug
01 D. C I.I. Z I--
1000
*1
E L_
Naive
ii
UCS
CS
Nil
CS (UCS+CS)
Treatment of mice (31group) Figure 6. Wortmannin blocks LPS tolerance induced by the UCS, but not that induced by the CS.
4.
DISCUSSION
Behavioural conditioning of altered immune responses represents perhaps one of the more profound illustrations of the functional interactions between the CNS and the immune system. As is evident from the number of examples cited above, evidence for such phenomena is quite compelling. Moreover, a number of groups have gone on to confirm that such conditioned immune responses show many of the other features of classical conditioning models in general, including extinction in the face of repeated presentation of CS cues in the absence of UCS reinforcement. The physiological mechanism(s) behind conditioned immunity nevertheless remains much of an enigma. To a large degree there remains a body of thought that explicitly or implicitly assumes that conditioned suppression of immunity reflects the operation of stress-induced responses (of the HPA axis). Glucocorticoid levels are elevated in, and often equated with, stress, and have in addition been shown independently to decrease a number of immune responses. Thus the idea that this should be (the sole) explanation for conditioning is perhaps an understandably attractive one. However there are a number of pieces of evidence which can be cited to refute this notion. Included in these are: reports of conditioned immunosuppression occurring in the absence of altered adrenocortical responses [84]; studies examining the effect of circadian rhythmicity on the efficacy of conditioning, which showed that conditioned suppression was most easily elicited in animals at the trough (not peak) of their endogenous circulating corticosterone levels [85]; data showing that stimulation of animals to elicit increased adrenocortical stimulation leads to enhanced (not suppressed) immune responses [86]; experiments indicating that LiC1, an agent which, like cyclophosphamide, evokes a profound taste aversion and elevated corticosterone levels, is ineffective as a UCS for conditioning antibody responses [1]; and data showing that eliminating the differential, likely stress-inducing, components of taste aversion paradigms does not eliminate conditioned immunosuppression [87]. However, none of these arguments necessarily negates the notion that conditioned immuno-suppression reflects the operation of a conditioned neuroendocrine change per se!
249
If alterations in the HPA axis are not primarily responsible for the conditioned changes observed, what other factors might be implicated? Multiple processes are probably involved, and different factors may thus take on primary importance in different model systems/physiological situations. Indeed, it is possible that different mechanisms are involved when conditioning is superimposed on a resting versus an activated immune system. This is particularly likely given that there is a plethora of potential neuroimmune mediators, and it is known already that the same immunomodulating agents bind to, and activate differentially, resting versus activated cells. Given the evidence (above) that the immune system is innervated; that leukocytes and neurons share numerous receptors for neuropeptides/neurotransmitters; that lymphocytes and tissues of the nervous system produce and respond to many different common molecules, including cytokines, chemokines, neurotrophins and conventional neuropeptides and/or neurotransmitters, it should not be surprising that there are indeed many potential mechanisms by which CNS: immune interactions occur in conditioned animals. There may be conditioned changes in neuroendocrine activation pathways which affect lymphocytes and their activation. There may alternatively be conditioned changes in the release of lymphocyte products which in turn affect cells within the nervous system. Using a model system investigating conditioned alterations in cytokine production occurring following pairing of LPS with a novel taste in the drinking water, the data above have explored elements of this CNS: immune system interaction, with particular regard to the role of the neuropeptides SP and SOM. It is clear from these studies that the conditioned alteration in cytokine production seen, which is evident both locally (in gut tissue) and in the CNS of conditioned mice, is regulated primarily by locally produced SP and SOM (probably acting in a mutually antagonistic fashion), and does not reflect signaling (in the CNS) from peripherally produced cytokines. As evidence for this hypothesis, infusion of an excess of mAbs to IL-1 and/or TNF~ produced no effect on CNS mRNA expression of IL-1/TNFct in conditioned mice, while infusion of SP antagonists inhibited this completely (and SOM showed a trend towards some increased levels of IL-1/TNFct). Thus in this system at least, it seems that CNS: immune system interactions in conditioned mice receiving oral antigen challenge are primarily regulated by the neuropeptides SP and SOM. In addition, using an indirect method to examine intracellular signaling induced by the UCS or CS in conditioned mice, namely the induction of tolerance to deliberate restimulation with LPS itself, the data reported suggest that, perhaps paradoxically, the CS does not trigger TNFct production in this system in the same fashion as the UCS (in this case LPS). There has been a surge in recent interest in the potential dual role of cytokines in the CNS acting both as CNS stimuli and modulators of inflammation. As an example, transgenic mice overexpressing IFNot demonstrated many of the changes phenotypically characteristic of human neurodegenerative disorders [88], though they were (paradoxically) resistant to numerous neurotropic viruses, while IL-6 has been found to promote production of argininevasopressin, in turn leading to SIADH, not uncommon in chronic inflammation. The dual action of other brain derived hormones has also been well documented (note for instance that ~-melanocyte-stimulating hormone inhibits an anti-inflammatory response, likely by regulation of NF-kB/IkB [89]. The wheel turns full circle once more with evidence that norepinephrine can induce differential cytokine gene expression from lymphocytes [90], and that the sensitivity of lymphocytes to such signaling is critically dependent upon their prior activation state (see Chambers and Schauenstein, 2000). Integration of our understanding of CNS: immune system interactions will likely become an increasingly more important issue for basic and clinical scientists.
250
5.
SUMMARY
An overview of CNS: immune system interactions, with particular attention to behavioural conditioning of immunity, is presented. Attention is focused on studies which have examined the mechanism(s) whereby conditioning of immune responses may take place. There is now concrete evidence for regulation at the level of cell migration which in urn may reflect changes in chemokines, chemokine receptors, or other ("addressing") receptors on leukocytes controlling cell migration in conditioned animals. In addition, regulation in conditioned animals might reflect modulation of cell activation, in particular in the context of an alteration in the cytokine milieu in which cells become activated in conditioned mice. Data is presented that, at least in a model where conditioning occurs following oral exposure to antigen (in this case LPS, a stimulator of, amongst others, inflammatory cytokines), the primary agents implicated in the CNS modulation of the immune response are the neuropeptides SP and SOM, and the signaling pathways engaged by the CS are not necessarily the same as those triggered by the UCS.
ACKNOWLEDGEMENTS My laboratory was supported throughout the course of these studies by the Medical Research Council (MRC) of Canada, and Clinique la Prairie (Montreux, Switzerland).
REFERENCES 1. 2. 3. 4. 5. 6.
7. 8. 9. 10. 11.
Ader R, Cohen N. Behaviourally conditioned immunosuppression. Psychosom Med 1975; 37: 333. Gorczynski RM, MacRae S, Kennedy M Conditioned immune response associated with allogeneic skin grafts in mice. J Immunol 1982; 129: 704. Greenberg AH, Dyck DG, Sandier LS, Pohazdak B, Dresel KM, Grant DJ. Neurohormonal modulation of natural resistance to a murine lymphoma. J Natl Cancer Inst 1984; 72: 653. Gorczynski RM, Kennedy M. Associative learning and regulation of immune responses. Prog Neuro-PsychoPharmacol and Biol Psychiat 1984; 8: 593. Ader R, Cohen N Behaviourally conditioned immunosuppression and murine systemic lupus erythematosus. Science 1982; 215: 1534. Gorczynski RM Conditioned immunosuppression in young versus aged mice: Differences in cells and responses to environmental stimuli lead to altered conditioning. Brain Behav Immun 1987; l: 307. Ottaway CA, Husband AJ. Central nervous system influences on lymphocyte migration. Brain Behav Immun 1992; 6: 97. Goetzl EJ, Spector NH. Neuroimmune networks: Physiology and Disease. New York: Alan Liss, 1989. Hall N, Goldstein AL. Neurotransmitters and the immune system. In: Ader R, ed. Psychoneuroimmunology. New York: Academic Press, 1981: 521. Besodovsky H, Sorkin E. Network of immune: endocrine interactions. Clin Exp Immunol 1977; 27: 1. Besedovsky HO, del Ray A. Immune-neuro-endocrine interactions: facts and hypotheses. Endocrine reviews 1996; 17: 64.
251
12. 13. 14. 15. 16. 17. 18. 19.
20. 21.
22.
23. 24. 25.
26.
27.
28.
29. 30.
31. 32.
Gorczynski RM. Conditioned immunity to L. Major in young and aged mice. In: Klein HFaT, ed. Psychoneuroimmunology, Stress and Infection. Florida: CRC Press, 1995. Morato EF, GerbaseDelima M, Gorczynski RM. Conditioned immunosuppression in orally immunized mice. Brain Behav Immun 1996; 10: 44. Anisman H, Baines MG, Berczi I, et al. Neuroimmune mechanisms in health and disease: 1. Health. Can Med Assoc J 1996; 155: 867. Anisman H, Baines MG, Berczi I, et al. Neuroimmune mechanisms in health and disease.2. Disease. Can Med Assoc J 1996; 155: 1075. Blalock JE. The immune system as a sensory organ. J Immunol 1984; 132: 1067. Sklar LS, Anisman H. Stress and Cancer. Psychol Bull. 1981; 89: 369. Schorr EC, Arnason B. Interactions between the sympathetic nervous system and the immune system. Brain Behav Immun 1999; 13:271. Bellinger DL, Felten SY, Lorton D, Felten DL. Innervation of lymphoid organs and neurotransmitter-lymphocyte interactions. In: WFHickey RWKa, ed. Immunology of the nervous system. New York: Oxford University Press, 1997. Madden KS, Felten DL. Experimental basis for neural-immune interactions. Physiol Rev 1995; 75: 77. Wiegmann K, Muthyala S, Kim D, Arnason BG, Chelmick-Schorr E. "-adrenergic agonists suppress relapsing experimental allergic encephalomyelitis (CREAE) in LEW rats. J Neuroimmunol 1995; 56: 201. Karazewski JW, Reder AT, Anlar B, Kim WC, Arnason BGW. Increased lymphocyte beta-adrenergic receptor density in progressive multiple sclerosis is specific for the CD8 § CD28-suppressor cell. Ann Neurol. 1991; 30: 42. GonzalezAriki S, Husband AJ. Ontogeny of IgA(+) cells in lamina propria: effects of sympathectomy. Develop Comp Immunol 2000; 24: 61. Dziegielewska KM, Moiler JE, Potter AM, Ek J, Lane MA, Saunders NR. Acute-phase cytokines IL-1 beta and TNF-alpha in brain development. Cell Tissue Res 2000; 299: 335. Ledeboer A, Breve JJP, Poole S, Tilders FJH, VanDam AM. Interleukin-10, interleukin-4, and transforming growth factor-beta differentially regulate LPS-induced production of pro-inflammatory cytokines and nitric oxide in co-cultures of rat astroglial and microglial cells. Glia 2000; 30: 134. Throsby M, HomoDelarche F, Chevenne D, Goya R, Dardenne M, Pleau JM. Pancreatic hormone expression in murine thymus: Localization in dendritic cells/macrophages. Endocrinol 1998; 139: 2399. Markus AMA, Kockerling F, Neuhuber WL. Close anatomical relationships between nerve fibers and MHC class II- expressing dendritic cells in the rat liver and extrahepatic bile duct. Histochemistry Cell Biol 1998; 109: 409. Gordon DJ, Ostlere LS, Holden CA. Neuropeptide modulation of Thl and Th2 cytokines in peripheral blood mononuclear leucocytes in atopic dermatitis and non-atopic controls. Br J Dermatol 1997; 137: 921. Niizeki H, Kurimoto I, Streilein JW. A substance p agonist acts as an adjuvant to promote hapten-specific skin immunity. J Invest Dermatol 1999; 112: 437. Jeon HK, Jung NP, Choi IH, Oh YK, Shin HC, Gwag BJ. Substance P augments nitric oxide production and gene expression in murine macrophages. ImmunoPharmacol 1999; 41: 219. Raithel M, Schneider HT, Hahn EG. Effect of substance P on histamine secretion from gut mucosa in inflammatory bowel disease. Scand J Gastroenterol 1999; 34: 496. Kerschensteiner M J Exp Med 1999; 189: 865.
252
33. 34.
35. 36. 37. 38.
39. 40. 41.
42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52.
53. 54. 55. 56. 57.
Burgi B, Otten UH, Ochensberger B, et al. Basophil priming by neurotrophic factors. Activation through the trk receptor. J Immunol 1996; 157: 5582. Maroder M, Bellavia D, Meco D, et al. Expression of trkB neurotrophin receptor during T cell development: role of brain-derived NTF in immature thymocyte survival. J Immunol 1996; 157: 2864. Levi Montalcini R. Trends in Neurosci 1996; 19:514. Braun A. Europ J Immunol 1998. Virchow JC. Am J Respir Crit Care Med 1998; 158: 2002. Kavelaars A, Heijnen CJ, Ellenbroek B, van Loveren H, Cools AR. Apomorphine-susceptible and apomorphine-unsusceptible Wistar rats differ in their susceptibility for inflammatory and infectious diseasses: A study on rats with group specific differences in structure and reactivity of the hypothalamic-pituitary-adrenal axis. J Neurosci 1997; 17: 2580. Heijnen CJ. Who believes in "communication". Brain Histol. Histopathol. 2000; 14: 2. Schwarz MJ, Ackenheil M, Reidel M, Mueller N. Ratio of type-I/type-2 cytokines in schizophrenia. Neuroimmunomodulation 1999; 6: 463. Krank MD, McQueen GM. Conditioned compensatory responses elicited by environmental signals for cyclophosphamide-induced immunosuppression of antibody production in mice. Psychobiol 1988; 16: 229. Eikelboom R, Stewart J. Conditioned drug-induced physiological response. Physiol Rev 1982; 89: 507. Gorczynski RM. Analysis of lymphocytes in, and host environment of, mice showing conditioned immunosuppression to cyclophosphamide. Brain Behav Immun 1987; 1: 21. Butcher EC. Leukocyte-endothelial cell recognition: three (or more) steps to specificity and diversity. Cell 1991; 67: 1033. Issekutz TB. Lymphocyte homing to sites of inflammation. Curr Opin Immunol 1992; 4: 287. Butcher EC, Picker LJ. Lymphocyte homing and homeostasis. Science 1996; 272: 60. Husband AJ. The immune system and integrated homeostasis. Immunol Cell Biol 1995; 73 (4): 377. Baggiolini M. Chemokines and leukocyte traffic. Nature 1998; 392: 565. Cyster JG. Chemokines and cell migration in secondary lymphoid organs. Science 1999; 286: 2098. Zlotnik A, Morales J, Hedrick JA. Recent advances in chemokines and chemokine receptors. Crit Rev Immun 1999; 19: 1. Zagury D, Lachgar A, Chams V, et al. C-C chemokines, pivotal in protection against HIV type 1 infection. Proc Natl Acad Sci USA 1998; 95: 3857. Mantovani A, Allavena P, Vecchi A, Sozzani S. Chemokines and chemokine receptors during activation and deactivation of monocytes and dendritic cells and in amplification of Thl versus Th2 responses. Int J Clin Lab Res 1998; 28: 77. Sallusto F, Lanzavecchia A, Mackay CR. Chemokines and chemokine receptors in T-cell priming and Thl/Th2- mediated responses. Immunol Today 1998; 19: 568. Ward SG, Westwick J. Chemokines: understanding their role in T-lymphocyte biology. Biochem. J 1998; 333: 457. Rottman JB. Key role of chemokines and chemokine receptors in inflammation, immunity, neoplasia, and infectious disease. Vet Pathol 1999; 36: 357. Wang JM, Deng XY, Gong WH, Su SB. Chemokines and their role in tumor growth and metastasis. J Immunol Methods 1998; 220: 1. Terkeltaub R, Boisvert VA, Curtiss LK. Chemokines & atherosclerosis. Curr Opin Lipidol 1998; 9: 397.
253
58. 59. 60. 61. 62.
63. 64. 65. 66. 67. 68.
69. 70. 71. 72.
73. 74.
75.
76.
77. 78. 79.
Gale LM, McColl SR. Chemokines: extracellular messengers for all occasions? Bioessays 1999; 21: 17. Asensio VC, Campbell IL. Chemokines in the CNS: plurifunctional mediators in diverse states. Trends Neurosci 1999; 22: 504. Schindler C. Cytokines and JAK-STAT signaling. Exp Cell Res 1999; 253: 7. Mosmann TR, Coffman RL. Thl and Th2 cells: different patterns of lymphokine secretion lead to different functional properties. Ann Rev Immuno 1989; 7: 145. Mosmann TR, Li L, Hengartner H, et al. Differentiation and functions of T cell subsets. In: Bock GR, Goode JA, eds. olecular Basis of Cellular Defence Mechanisms. John Wiley & Sons Lt, 1997: 148. Swain SL. Helper T cell differentiation. Curr Opin Immunol 1999; 11:180. Weiner HL. Oral tolerance: Immune mechanisms and treatment of autoimmune diseases. Immunol Today 1997; 18: 335. Kelso A. Cytokines: Principles and prospects. Immun Cell Biol 1998; 76: 300. Dallman MJ. Cytokines and transplantation: Thl/Th2 regulation of the immune response to solid organ transplants in the adult. Curr Opin Immunol 1995; 7 (5): 632. Gorczynski RM. Immunosuppression induced by hepatic portal venous immunization spares reactivity in IL-4 producing T lymphocytes. Immunol Lett 1992; 33: 67. Gorczynski RM, Cohen Z, Fu XM, Hua Z, Sun YL, Chen ZQ. Interleukin-13, in combination with anti interleukin-12, increases graft prolongation after portal venous immunization with cultured allogeneic bone marrow-derived dendritic cells. Transplantation 1996; 62:1592. Falcone M, Sarvetnick N. Cytokines regulating autoimmunity. Curr Opin Immunol 1999; 11: 670. Liew FY, Xu DM, Chan WL. Effector mechanisms in parasitic infections. Immunol Lett 1999; 65: 101. Turnbull AV, Rivier CL. Regulation of the hypothalamic-pituitary-adrenal axis by cytokines: Actions and mechanisms of action. Physiol Rev 1999; 79: 1. Teran LM, Mochizuki M, Bartels J, et al. Thl- and Th2-type cytokines regulate the expression and production of eotaxin and RANTES by lung fibroblasts. Amer J Respir Cell Molec Biol 1999; 20: 777. Oh JW, Schwiebert LM, Benveniste EN. Cytokine regulation of CC and CXC chemokine expression by human astrocytes. J. Neurovirology. 1999; 5: 82. Karpus WJ, Kennedy KJ, Kunkel SL, Lukacs NW. Monocyte chemotactic protein 1 regulates oral tolerance induction by inhibition of T helper cell 1-related cytokines. J Exp Med 1998; 187: 733. Ramakrishna V, Hu J, Lei J, Li X, Gorczynski RM. Alterations in chemokine mRNA expression in animals receiving portal vein immunization and renal allo-or xenotransplantation precede altered cytokine production. J Surg Res 1999; 87: 62. Goetzl EJ, Turck CW, Sreedharan SP. Production and recognition of neuropeptides by cells of the immune system. In: Ader R, Felten, D, Cohen, N., ed. Psychoneuroimmunology, 2nd. Edition: Academic Press, 1991: 263. Lotz M, Vaughan JH, Carson DA. Effects of neuropeptides on production of inflammatory cytokines by human monocytes. Science 1988; 241: 1218. Rayhane N, Fitting C, Cavaillon JM. Dissociation of IFN-gamma from IL-12 and IL-18 production during endotoxin tolerance. J Endotoxin Res 1999; 5: 319. Nakano M, Tominaga K, Saito S, et al. Lipopolysaccharide- and paclitaxel (Taxol)induced tolerance in murine peritoneal macrophages. J Endotoxin Res 1999; 5: 102.
254
Kraatz J, Clair L, Rodriguez JL, West MA. Macrophage TNF secretion in endotoxin tolerance: Role of SARK, p38, and MAPK. J Surg Res 1999; 83: 158. 81. Randow F, Docke WD, Bundschuh DS, Hartung T, Wendel A, Volk HD. In vitro prevention and reversal of lipopolysaccharide desensitization by IFN-gamma, IL-12, and granulocyte-macrophage colony-stimulating factor. J Immunol 1997; 158:2911. 82. Bowling WM, Flye MW, Qiu YY, Callery MP. Inhibition of phosphatidyl inositol-3'-kinase prevents induction of endotoxin tolerance in vitro. J Surg Res 1996; 63: 287. 83. Nomura F, Akashi S, Sakao Y, et al. Cutting edge: Endotoxin tolerance in mouse peritoneal macrophages correlates with down-regulation of surface Toll-like receptor 4 expression. J Immunol 2000; 164: 3476. 84. Kusnecov AV, Sivyer M, King MG, Husband AJ, Cripps AW, Clancey RL Behaviourally conditioned suppression of the immune response by antilymphocyte serum. J Immunol 1983; 130:2117. 85. Gorczynski RM, MacRae S, Kennedy M. Factors involved in the classical conditioning of antibody responses in mice. In: Ballieux RE, Fielding JF, Abbate AL, eds. Breakdown in Human adaptation to stress. Utrecht: Martinhus Nijhof, 1984: 704. 86. Bovbjerg D, Cohen N, Ader R. Behaviourally conditioned enhancement of delayed type hypersensitivity in the mouse. Brain Behav Immun 1987; 1: 64. 87. Klosterhalfen S, Klosterhalfen W. Classically conditioned effects of cyclophosphamide on white blood cell counts in rats. Ann New York Acad Aci 1987; 496: 569. 88. Campbell IL, et al. Structural and functional neuropathology in transgenic mice with CNS expression of IFNct. Brain Res 1999; 835: 46. 89. Ichiyama T, et al. Systemically administered alpha-melanocyte-stimulating peptides inhibit NF-kappaB activation in experimental brain inflammation. Brain Res 1999; 836: 31. 90. Kalinichenko VV, Mokyr M, Graf LH, Cohen RL, Chambers DA. Norepinephrine-mediated inhibition of antitumor cytotoxic T lymphocyte generation involves a [3-adrenergic receptor mechanism and decreased TNFct gene expression. J Immunol 1999; 163: 2492. 91. Chambers DA, Schauenstein K. Mindful immunology: neuroimmunomodulation. Immunol Today, 2000; 21 : 168.
80.
New Foundation of Biology Edited by I. Berczi and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
255
Sleep, Health and Immunocompetence
HARVEY MOLDOFSKY, WAH-PING LUK and JODI DICKSTEIN
The Centre of Sleep and Chronobiology, University of Toronto, 399 Bathurst St., MP 14-308, Toronto, Ontario, M5T 2S8
ABSTRACT Basic and applied research on the sleep/wake-related cytokine-immune-endocrine functions and their implications for human health and disease are reviewed. The interaction of the circadian sleeping-waking brain and the cytokine-immune/endocrine system are integral to preserving homeostasis. As with the animal studies, there may be host defence implications for altered immune and endocrine functions in sleep-deprived humans. Activation of cytokines and sleepiness occur during the acute phase response to bacterial or viral disease. There are disturbances in sleep and cytokine-immune functions in chronic protozoal and viral disease e.g., trypanosomiasis, human immune deficiency viral disease. Sleep-related physiological disturbances may play a role in autoimmune diseases, primary sleep disorders and major mental illnesses.
1.
INTRODUCTION
Over the past two decades various immunomodulatory cytokines and neuroendocrine substances are reported to affect the sleep-wake behaviour of animals. Such peptides as interleukin-1 (IL-1), interferon (IFN) alpha, Factor S, vasoactive intestinal peptide (VIP), tumour necrosis factor alpha (TNF-c~) not only promote sleep, but also exert effects on the immune system. Prostaglandin D2 and E2 have reciprocal influence on sleep via IL-1. Hormones that stimulate IL-1 and facilitate sleep include: insulin, growth hormone releasing factor (GHRH), growth hormone, somatostatin and melatonin. Corticotrophin-releasing factor (CRF), adrenocorticotropic hormone (ACTH), alpha melanocyte stimulating hormone and glucocorticoids inhibit sleep. These immunologically active cytokines and neuroendocrinedel factors interact and regulate sleep-wake behaviour [1, 2]. Such findings have important implications not only in advancing knowledge of how the immune-neuroendocrine systems interact to regulate the sleeping/waking brain, but also how these systems influence peripheral functions of the body in health and disease. The purpose of this review is to summarise the knowledge derived from basic and applied research on the sleep/wake-related cytokine-immune-endocrine functions and their implications for matters related to immunocompetence in humans.
256
2.
IMMUNE FUNCTIONS AND SLEEP REGULATION
At the turn of the century, Ishimori in Japan and Pieron in France discovered that the cerebrospinal fluid of sleep-deprived dogs when injected intraventricularly into normal recipient dogs promoted their sleep. In 1971, Pappenheimer and colleagues showed that cerebrospinal fluid (CSF) transferred from sleep-deprived goats induced sleep in the recipient rats [3]. Subsequently, Krueger at al. isolated a sleep-promoting substance that was termed Factor S. This factor proved to be a muramyl peptide, which is a component of bacterial cell walls [4]. These researchers showed that intraperitoneal or intraventricular administration of low doses of muramyl peptides and lipopolysaccharides in rabbits and cats enhanced slow wave sleep (SWS) and increased the amplitude of delta slow waves (0.5-4 Hz) for one to six hours [5]. The likely mediators for the sleep-inducing effects of muramyl peptides are the cytokines IL-1 and TNF-ct. Not only do both of these cytokines occur in the brain, but also they serve to regulate peripheral immune functions. Therefore, both of these cytokines have been the subject of considerable interest. They enhance non-rapid eye movement (NREM) SWS and suppress rapid eye movement (REM) sleep after intravenous or intraventricular injection [6, 7]. Moreover, IL-1 (is elevated in the cat CSF during SWS [33]. Experimental inactivation of IL-1 or TNF-~t inhibits spontaneous sleep, sleep rebound after sleep deprivation and microbialproduct induced sleep [8, 9]. Furthermore, both IL-1 and TNF-ct are produced in the brain and fluctuate with the circadian day. IL-1 mRNA [10] and TNF-ct mRNA exhibit a diurnal rhythm in the rat, peaking during periods of greatest NREM sleep, just after lights were turned on. TNF-ct protein levels are 10 fold greater during NREM sleep in the rat [11]. IL-1 and TNF-ct interact with each other and are part of a biochemical cascade of events leading to sleep [2]. IL-1 and TNF-ct induce nitric oxide (NO). Inhibition of NO synthase inhibits normal sleep and IL-1 induced sleep [12, 13] suggesting that NO is a downstream mediator. Although it is becoming clear that cytokines (such as IL-1 and TNF) have hypnogenic actions in both pathological and physiological conditions, there has been little investigation of how these substances act at the neuronal level to induce sleep. A search of Medline reveals that of the 127 papers pertaining to IL-1 and sleep published since 1987 only four examine the central loci of action of IL-1 [14, 15, 16, 17]. As with any emerging field the current literature concerning IL-1 hyponogenic actions in the CNS are often contradictory and confusing. This review illustrates the truly pleotropic nature of IL-1, by examining the possible neural mechanism by which IL-1 acts to induce slow wave sleep.
3.
INTERLEUKIN-1 MRNA LEVELS VARY ACCORDING TO TIME OF DAY
Numerous studies have observed the presence of IL-1 (and its mRNA) in the normal brain [for review see 18, 19]. However, if IL-1 is involved in the normal physiological sleep process, then it can be reasoned that its levels in the CNS should vary with the time of day. Recently Taishi and collegues [10], provide evidence supporting this idea. With RT-PCR their group replicated previous findings, detecting IL-1 mRNA in the hypothalamus, hippocampus, cerebral cortex and brainstem. But, more importantly, they demonstrate that these levels are differentially expressed, with the highest and lowest levels occurring during the end of the wake and end of sleep periods, respectively.
257
.
I N T E R L E U K I N - I : SLEEP E F F E C T S OF L O C A L I Z E D APPLICATION IN T H E CNS
Walters, Meyers and Krueger, [14] utilizing microinjections of IL-1 into different regions of rabbit brains, while recording brain activity, were the first to attempt a localization of the site of action within the central nervous system of IL-1 with respect to sleep. Although they were unable to identify any parenchymal CNS region with IL-1 inducible sleep enhancing properties, they noticed that injection of IL-1 into the aqueduct of Sylvius produced an enhancement of slow wave sleep (SWS). De Sarro et al [15] observed that microinjections of f M levels of recombinant IL-1 into the rat locus coeruleus, a noradrenaline containing neuronal group whose activation is associated with waking and R E M sleep, produced increased behavioural sedation and/or sleep associated with E c o G synchronization that was preventable by pretreatment with anti-IL-1 antibodies. More recently, Terao et al [16] undertook a comprehensive study examining the effect of IL-1 infusion into seven different locations in the ventricular and subarachnoid systems of the brain in freely moving rats. Although infusions into all areas increased SWS, the greatest enhancement was seen after IL-1 was infused into the prostaglandin D2 sensitive sleep-promoting zone located within the ventral surface of the rostral basal forebrain. These effects were blocked by the co-administration of diclofenac, indicating the involvement of the PG cascade in the sleep promoting pathway of IL-1. These experiments indicate that IL-1 can act to promote SWS in both the parenchymal and leptomeninges areas. It should be noted that although Walters et al [14] did not observe activity in parenchymal areas, they may have been limited by the fact that only a non-rat specific early extract of IL-1 was available at that time. Table I
IL-1 Injection Sites and Sleep Response.
IL-1 Injection Site
Action
Ventral surface of Rostral basal forebrain [16]. Third Ventricle apposed to the medial preoptic area [16]. Lateral Ventricle [16]. Subarachnoid space underlying posterior hypothalamus [16]. Third Ventricle near the aquaduct of Sylvius [16]. Between aquaduct of Sylvius and the forth ventricle, Apposed to the locus coeruleus [16]. Cisterna magna [16]. Aquaduct of Sylvis [14]. Locus Coeruleus [15]. Hypothalmic lateral paraventricular area [17].
t 1't SWS, inhibit PS t t t sws t t t sws t t t sws t t t sws t t t sws t t t sws Increase SWS Increase SWS Inhibit SWS
IL-l-interleukin-1, t SWS -- increase of slow wave sleep While these studies demonstrate that IL-1 applied to multiple sites in the CNS can induce somnergenic and related effects, the question still remains as to how this substance works at the neuronal level. To investigate this question, we will briefly review the basal forebrain, an area central to the regulation of SWS, and focus on the possible mechanisms by which IL-1 could act in this area.
258
5.
THE BASAL FOREBRAIN: AN AROUSAL AND HYPNOGENIC CENTER
The basal forebrain (BF) is a heterogenous CNS region, that encompasses the entire basal telencephalon, including the preoptic area, in addition to the medial septum, diagonal band nuclei and substantia innominata. Long implicated as an area important in the control of sleep and waking [for a detailed anatomical examination of the BF see 20], the basal forebrain for our purposes can be characterized into anatomically distinct neocortical cholinergic "arousal" regions and "somnergenic" (non-cholinergic) regions, both of which project to cortical areas. It should be noted that there are areas within the BF that have functions independent of arousal and sleep. This issue will not be addressed here as it is beyond the scope of this review.
.
IDENTIFICATION OF THE BASAL FOREBRAIN SLEEP PROMOTING REGIONS: DEACTIVATION OF THE CORTEX
Sterman and Clemente in 1962 [21] were first to demonstrate that low and high frequency trains of electrical stimulation given in the magnocellular preoptic area, horizontal diagonal band of Broca, and rostral substantia innominata BF regions of cats elicited sleep. Along the same lines electrolytic [22] and neurotoxic [23] lesions applied to these areas have also been shown to produce chronic insomnia. These areas are now known to be rich in cortically projecting cholinergic neurons, local GABAergic interneurons and cortically projecting GABAergic neurons [20, 24]. These features suggest that stimulation in these areas elicit inhibition of projecting cholinergic components via local inhibitory circuits in addition to recruiting projecting inhibition to the cortex, leading to EEG synchronization characteristic of SWS. The detailed examination of the BF via in-vivo single unit recordings [20] and immediate early gene expression [25] have localized neurons exhibiting selective activation during sleep. These sleep active neurons were detectable in the medial preoptic/anterior hypothalamic areas and within the ventrolateral preoptic region increase their discharge rates several seconds prior to the first signs of EEG synchronization when recorded from chronically implanted cats [20]. While they display their most active state in NREM sleep, this activity tapers off during wake and REM sleep. Such physiological changes suggest that these neurons are involved in the transition from periods of sustained wakefulness to sustained sleep. Recently, these "sleep active" neurons in the ventral lateral preoptic area were shown to have distinct electrophysiological properties, such that they are inhibited by wake active substances and are GABAergic in nature [26].
7.
AROUSAL CENTERS OF THE BASAL FOREBRAIN
In contrast, chemical and electrical stimulation of the magnocellular cholinergic regions of the BF, which are the major cholinergic innervators of the limbic telencephalon and neocortex result in behavioral arousal. These cells have little GABAergic overlap [20] and display their most active states during wakefulness and REM sleep, and become quiescent in NREM sleep. Therefore, these BF neurons are termed "wake-active".
259
.
SLEEP ENHANCING SUBSTANCES" ACTIVATION OF SLEEP ACTIVE SYSTEMS WITH CONCURRENT INHIBITION OF WAKE ACTIVE CENTERS
B a s e d on the a n a t o m i c a l outlay of the B F we w o u l d expect that sleep-related substances excite sleep active n e u r o n s and inhibit arousal neurons. Table II
Sleep enhancing agents.
Substance
Somnergenic or Arousal Mediating
Action on BF Wake Neurons
Action on BF Sleep Neurons
Sleep Wake Wake Wake Sleep Wake Sleep
Inhibit Excite Excite Excite 9 Excite Inhibit
Excite Inhibit Excite Inhibit Excite 9 9
Adenosine 5-HT Noradrenaline Acetylcholine PGD2 Glutamate GABA
5HT -- serotonin, PGD2 -- prostaglandin D2, GABA -- gamme aminobutyric acid F r o m this table it can be n o t e d that there is overall a g r e e m e n t with this hypothesis. Substances that are i m p l i c a t e d in the arousal state will g e n e r a l l y activate B F wake active neurons while inhibiting B F sleep active neurons, while the opposite is true for sleep-related substances [20, 27, 28, 29].
9.
B A S A L F O R E B R A I N A N D IL-I: P O S S I B L E S O M N E R G E N I C M E C H A N I S M S
H o w does IL-1 p r o m o t e sleep? A l t h o u g h a c o m p l e t e p a t h w a y has yet to be elucidated, one hypothesis w o u l d be that IL-1 p r o m o t e s SWS by activating or m o d u l a t i n g pathways of sleep related substances, while inhibiting arousal m e d i a t i n g ones. F r o m the table listed below we can see that there is an overall a g r e e m e n t with this hypothesis. W h a t r e m a i n s to be done is to verify at the cellular level that IL-1 i n d e e d does inhibit wake active neurons while exciting sleep active ones [14, 30]. Table III
Possible somnergenic mechanisms. Substance
Linkage to IL-1
PGD2 NF-•B Adenosine C-fos Acetylcholine GABA Noradrenine Glutamate
IL-1 increases expression IL-1 increases expression IL-1 increases endogenous adenosine output IL-1 causes induction IL-1 inhibits release IL-1 potentiates IL-1 increases turnover IL-1 inhibits release
See footnotes Tables I and II. NF-~zB - nuclear factor KB C-for - early gene signaling cell actination
260
10.
CIRCADIAN REGULATION OF CYTOKINES
If we postulate that levels of cytokines and other sleep factors build up during wakefulness and dissipate during sleep, then there must be some level of circadian control which allows these levels to be built up, without immediate dissipation. For example, for sleep to operate under a purely homeostatic mechanism, then animals would sleep intermittently throughout the day whenever levels of sleep factors build up. This is exactly what is seen in animals that have had their suprachiasmatic nucleus (SCN) ablated [31]. Therefore, the regulation of cytokine-mediated operations in controlling the sleep/wake cycle depends upon the efficiency of intrinsic regulatory biological rhythms that operate in part through SCN regulatory control. This would account for the circadian variations in the levels of IL-1 mRNA. Because other CNS substances have sleep modulating effects how these substances interact to function in a circadian manner needs to be determined.
11.
THE CENTRAL NERVOUS SYSTEM AND PERIPHERAL ACTIVITIES OF CYTOKINES
While it is unknown how sleep-inducing cytokines influence the peripheral immune system, pathways have been identified which link brain cytokines to the peripheral immune system. Cytokines introduced into the brain gain access to the peripheral circulation. Radiolabel tracer experiments have demonstrated the transport of IL-lct [32], IL-l[3 [33], IL-2 (34], IL-6 (35, 36] and TNF-ct (37, 38) from the CSF into the venous blood by way of the arachnoid villi draining into the superior sagittal sinus, the principle venous drainage route of the CSF and brain. In addition, cytokines introduced into the brain drain into the cervical lymphatics. Studies have shown that a tracer injected into the CSF is transported along the extensions of the subarachnoid space around specific cranial nerves, primarily the olfactory nerve, to the cribriform plate. Tracer then passes through the cribriform plate and the nasal submucosa to the cervical lymph nodes [39]. Recently we have demonstrated the transport of TNF-~ from the CSF to the cervical lymphatics [40]. This direct connection between the brain and lymphatic system allows for immuno-regulatory cytokines produced in the brain to selectively modify local immune responses in regional cervical nodes [41]. Brain cytokines may also influence the expression of cytokines and immune function in the periphery. Centrally administered IL-1 causes marked increases in IL-6 levels in the blood [42] and produces a rapid suppression of natural killer cell activity [43]. The potential for brain cytokines to drain into the blood, lymphatics or to induce the expression of additional cytokines in the periphery allows the brain (behaviour, physiological and pathological) to regulate the immune system. The function and development of primary and secondary lymphoid organs, lymphocytes, natural killer cells and mononuclear and polymorphonuclear phagocytic cells are dependent on the presence of cytokines. Sleep inducing cytokines can also modulate the peripheral immune system by their interaction with the Hypothalamic-pituitary adrenal axis (HPA). Central administration of IL-1 activates the HPA axis by stimulating the secretion of corticotrophin-releasing factor [44], plasma ACTH [45] and plasma cortisol [46].
261
Pathways by which sleep-inducing cytokines can interact with the periphery 1. 2. 3. 4. 5.
12.
Transport of CSF cytokine into cerebral venous drainage. Drainage of brain extracellular fluid and cytokine into neck lymphatics. Central cytokine induction of peripheral cytokines. Hypothalamic pituitary axis. Sympathetic nervous system.
CYTOKINES AND IMMUNE FUNCTIONS IN NORMAL SLEEP/WAKE BEHAVIOUR
In humans cytokine levels are also intimately linked to the sleep/wake cycle. IL-1 activity and IL-I[5 are elevated in the CSF of cats during slow wave sleep [47]. Moldofsky et al. (1986) [48] were the first to show elevations in IL-1 and IL-2-1ike activity in the blood that were related to sleep onset and SWS in young healthy males. Gudewill et al. (1992) [49], Covelli et al. (1992) [50], and Hohagen et al. (1993) [51] observed similar increases in plasma IL-1 and IL-1 production during sleep in addition to increases in IL-6 and INF- T. Darko et al. (1995) [52] found a strong correlation of EEG delta sleep and plasma TNF-~. Born et al. (1997) [53] demonstrated that sleep has a stimulatory effect on IL-2 production by CD3+ T cells but did not influence the production of IL-I[5, TNF-~ or IL-6. Because cytokines, like neuroendocrines, show pulsatile secretory behaviour, the differences observed in sleep-related cytokine functions may be related to the 3-hour interval in blood sampling techniques that were used in this study. Aspects of cellular functions of the immune system show circadian variations [54]. Lymphocyte distribution and activity vary over the 24-hour sleep/wake cycle. Sheep lymph flow and efferent prescapular lymphocyte output are reduced during sleep [55]. In humans, Moldofsky et al. (1986, 1996) [48, 56] showed changes in cellular immune functions during nocturnal sleep that differ from daytime wakefulness. Hourly sampling during daytime wakefulness and half hourly sampling during nocturnal sleep revealed decrease in natural killer (NK) cell activity and the proportion of NK cells in peripheral blood during SWS. Lymphocyte proliferation responses to pokeweed mitogen (PWM) and the proportion of T cells in the blood increase during sleep time compared to wakefulness. Similarly, Born et al. (1997) [53] showed sleep-related decreases in the number of circulating monocytes, NK cells and lymphocytes in peripheral blood. The circadian sleep/wake-related circulating T and NK cell are in harmonious inter-relationships to some neuroendocrines (growth hormone, melatonin, cortisol, prolactin) and temperature [57]. Changes in the number and proportion of circulating peripheral blood mononuclear cells (PBMC) in the blood during periods of sleep and wakefulness likely reflect a redistribution of these cells to tissues or organs such as the spleen and the lymphatic system. These activities are important in host defence mechanisms because they facilitate the detection and elimination of pathogens and the dissemination of immunologic memory. These functions are consistent with the theory that sleep contributes to host defences.
262
13.
GENDER DIFFERENCES IN SLEEP, NEUROIMMUNE AND ENDOCRINE SYSTEMS
The sleep physiology of women differs from that of men. Females have more SWS and higher amplitude delta activity than males across all ages [58]. During ovulatory menstrual cycles, there are prominent changes in reproductive hormones and core body temperature, which are accompanied by changes in sleep physiology and immune functions. Sleep/wake diaries from women aged 27-51 show that the late luteal phase is associated with a longer time to fall asleep, reduced sleep efficiency and sleep quality [59]. Polysomnographic studies report changes in sleep architecture throughout the menstrual cycle [60]. Moldofsky et al. (1995) [61] studied sleep-wake physiology and aspects of the immune system during high (> 5 ng/ml) and low (< 2 ng/ml) progesterone phases of the menstrual cycle of healthy women. Similar to men, women showed increased IL-1 activity during sleep but also during the mid-afternoon when there is an increased predisposition to sleep. However, during the high progesterone phase, SWS was delayed and reduced, as is the nocturnal sleep-related decline in NK activity. The response to the T-cell mitogen, phytohaemaglutinin (PHA), peaked at a later hour at night during the high progesterone phase. During the perimenstrual interval PBMC produced less IFN-gamma and more IL-10, resulting in a decreased IFN-gamma: IL-10 ratio compared with the mid-cycle interval. The peri-menstrual decrease in the IFN-gamma: IL-10 ratio was observed in women not taking oral contraceptives, but not in women taking oral contraceptives. Furthermore, the oral contraceptives group had a lower mid-cycle IFN-gamma: IL-10 ratio compared with the control group. Finally, subjects reported increased levels of distress during the perimenstrual interval compared with the mid-cycle interval. Overall these data suggest that healthy women have a perimenstrual shift in the type-I/type-2 cytokine balance toward a type-2 response that is blunted in women taking oral contraceptives [62].
14.
SLEEP DEPRIVATION AND IMMUNE FUNCTIONS
14.1. Human Studies Sleep deprivation is used to determine the effects of altered sleep physiology on psychological and physiological functions. Sleep deprivation causes sleepiness, fatigue, negative mood, and impairment in a variety of intellectual functions. Prolonged wakefulness also affects specific facets of the immune response. However, the studies do not report consistent results. Palmblad et al. (1976) [63] demonstrated decreased phagocytic activity and increased interferon production in 8 women deprived of sleep for 77 hour. Subsequently, they showed that 48 hours of wakefulness reduced the proliferative response of lymphocytes to mitogenic stimulation [64]. The stressful effects of the procedure and reliance upon single morning blood samples confound these studies. The effect of 40 hours of wakefulness on immune function was assessed in 10 healthy males [65]. Serial venous blood samples were obtained every 2 hours from 0800 to 2400 hour and half hourly from 2400 to 0730 hour. Forty hours of wakefulness did not show any physiological effect of stress as indicated by the lack of change in the diurnal pattern of plasma cortisol. Sleep deprivation led to enhanced nocturnal plasma IL-l-like and IL-2-1ike activities, which returned to baseline with recovery sleep. The nocturnal rise of PWM response during baseline sleep was suppressed during the night of wakefulness. NK activity declined progressively throughout the study and remained reduced during the night of resumed sleep.
263
These findings were recently confirmed by Heiser et al [66]. They found that sleep deprivation and recovery sleep lead to changes in the distribution of peripheral leukocytes, especially in a reduction of NK cells. Moreover, the cortisol rhythm was not affected by the sleep deprivation or recovery sleep. Diurnal patterns of immune functions were disrupted during 64 hours of wakefulness in five subjects [67]. Within a few days of this sleep deprivation study, two of the subjects reported upper respiratory infections. One of the subjects had asthmatic symptoms for the first time. The disruption of the regular diurnal rhythmic patterns of sleep-wakefulness and their coincident immune functions may have predisposed these two people to illness when affected by an infectious challenge to the body. Similarly, partial sleep deprivation in the early part of the night (2200-0300 hour) or in the later part of the night (0300-0700) decreased NK cell activity which returned to baseline levels after recovery sleep [68, 69]. In contrast, Dinges et al. (1994) [70] showed an increase in granulocytes, monocytes, NK activity and the proportion of lymphocytes in the S phase of the cell cycle in response to 64 hours of wakefulness in 13 males and 7 females. These results were based on single daily blood samples collected at 2200 hours. None of the subjects became ill during and up to 9 days after the study. Born et al. (1997) [53] compared the effects of sleep and sleep deprivation on immune function in 10 healthy men. Sustained wakefulness increased the number of circulating monocytes, NK cells and lymphocytes in peripheral blood. Furthermore, IL-2 production by T cells was significantly decreased during sleep deprivation. Heiser et al. (1997) [71] reported that blood samples taken at 0700, 1300, 1900 hours following acute sleep deprivation resulted in higher evening levels and lower morning levels of IL-I[~ release vs. baseline conditions. From these studies it is difficult to conclude if sleep plays a role in host defence mechanisms. Each study cited employed different blood sampling protocols and methods to measure immune indices. Furthermore, it is difficult to determine whether a change in a particular immune parameter is harmful, beneficial or inconsequential for the host. Human studies rely on the assessment of cells from peripheral blood samples, which represent about 1% of the total number of lymphocytes. However, the redistribution in PBMC and changes in cytokine production during sleep deprivation suggest a disruption to the regulatory influence of sleep on normal immune function. 14.2. Animal Studies Evidence that supports the hypothesis that sleep influences host defence mechanisms originates from sleep deprivation experiments in animals. Studies by Rechtschaffen et al. (1983) [72] showed that rats subjected to prolonged sleep deprivation (>30 days) died without any specific cause. A subsequent study by Everson (1993) [73] employing the same paradigm showed viable opportunistic bacteria in blood cultures from 5 of the 6 rats deprived of sleep. Fever or the acute phase response did not accompany the bacteremia. Subsequent research showed that chronic sleep deprivation led to the accumulation of microbes in mesenteric lymph nodes and subsequently translocation to adjacent organs such as the liver, kidney and lungs. The presence of pathogenic microorganisms and their toxins in tissues likely constitutes a septic burden and chronic antigenic challenge for the host, which is a serious challenge to the health of the animal [74]. These results are consistent with the theory that sleep deprivation leads to immunosuppression and vulnerability to disease. In contrast to this theory, using the same sleep deprivation paradigm, splenic cell mitogen responses and B-cell derived antibody responses did not show differences between sleepdeprived and control rats [75]. Furthermore, Bregmann et al (1996) [76] demonstrated that sleep deprivation enhances certain aspects of the immune system. Rats subjected to 10 days
264
of sleep deprivation showed less tumour growth and an increased rate of tumour regression following injection of Walker carcinoma cells when compared to the control rats who were not deprived of sleep. Experimental studies assessing the effects of short-term sleep deprivation on host defenses also provide conflicting results. Experiments by Brown et al. (1989) [77] suggest that shortterm sleep deprivation suppresses host defenses. In these experiments, mice were orally immunized with influenza and challenged intranasally one week later. One group of mice was sleep-deprived for 7 hours following the challenge. The sleep-deprived group showed reduced influenza-specific immunoglobulins and increased viral titres in the lung homogenates indicating the presence of a diseased state. In their attempt to duplicate these studies, Renegar et al. (1998 a, b) [78, 79] and Toth and Rehg (1998) [80] were unable to demonstrate any effect of short-term sleep deprivation on pre-existing mucosal and humoral immunity or on viral clearance and antibody titre. Subtle differences in the experimental protocols of the studies including the sex of the mice, the influenza strain, the route of immunization, the immunization period prior to viral challenge and stress response to the procedure could account for the differences reported in these studies. The findings by Everson and Toth [74] of the prolonged effects of sleep deprivation on the microbial invasion from the lumen of the gut into the body serves to emphasize that more studies are required to assess the differing mechanisms that are involved in short-term vs. long term sleep deprivation on mucosal and humoral immunity.
15.
SUMMARY
Bi-directional communication pathways exist between the brain and the cytokine-immuneendocrine systems. Cytokines, in particular IL-1, are involved in sleep/wake-related functions that operate via a variety of neurotransmitters and neuronal pathways involving the basal forebrain. The hypothalamic-pituitary adrenal axis, the efferent neuronal hypothalamus-autonomic nervous system axis, and the direct drainage of macromolecules from the brain into the blood and the lymphatic system provide a network by which the sleeping/waking brain influences bodily functions. Similarly, changes in cytokine levels in the periphery modulate the central nervous system either directly or via the vagal nerve, and influence the sleeping/waking brain. In humans circadian patterns of nocturnal sleep and daytime wakefulness are associated with changes in peripheral cytokines, cellular immune functions, and endocrines. Progesterone levels influence sleep and cellular immune functions during the menstrual cycle. The interaction between the circadian sleeping-waking brain and the cytokine-immune-endocrine system are integral to preserving homeostasis. Disorganisation or loss of sleep disrupts the harmonious integration of the circadian cytokine-immune-endocrine system and may result in a loss of host defense against microbial invasion. The mechanisms that control circadian sleep/wakefulnessrelated cytokine-immune-endocrine functions in the host defence against disease remain to be unravelled.
REFERENCES 1. 2.
Krueger JM, Takahashi S, Kapas L, Bredow S, Roky R, Fang J et al. Cytokines in sleep regulation. Adv Immunol 1995; 5: 171-188. Krueger JM, Obal Jr E Sleep regulatory substances. In: Schwartz WJ, editor. Sleep science: Integrating basic research and clinical practice. Basel: Karger, 1997: 175-194.
265
3. 4. 5. 6. 7.
8.
9. 10.
11. 12. 13. 14. 15.
16.
17.
18. 19.
20. 21.
22.
Fencl V, Koski G, Pappenheimer JR. Factors in cerebrospinal fluid from goats that affect sleep and activity in rats. J Physiol 1971; 216: 565-589. Krueger JM, Pappenheimer JR, Karnovsky ML. The composition of sleep-promoting factor isolated from human urine. J Biol Chem 1982; 257: 1664-1669. Krueger JM, Pappenheimer JR, Karnovsky ML. Sleep-promoting effects of muramyl peptides. PNAS 1982; 79: 6102-6106. Krueger JM, Walter J, Dinarello CA, Wolff SM, Chedid L. Sleep-promoting effects of endogenous pyrogen (interleukin-1). Am J Physiol 1984; 246: R994-R999. Shoham S, Davenne D, Cady AB, Dinarello CA, Krueger JM. Recombinant tumor necrosis factor and interleukin 1 enhance slow wave sleep. Am J Physiol 1987; 253: R142-R149. Takahashi S, Fang J, Kapas L, Wang Y, Krueger JM. Inhibition of brain interleukin-1 attenuates sleep rebound after sleep deprivation in rabbits. Am J Physiol 1997; 273: R677-R682. Takahashi S, Kapas L, Seyer JM, Wang Y, Krueger JM. Inhibition of tumor necrosis factor attenuates physiological sleep in rabbits. NeuroReport 1996; 7: 642-646. Taishi P, Bredow S, Guha-Thakurta N, Obal FJ, Krueger JM. Diurnal variations of interleukin-1 beta mRNA and beta-actin mRNA in rat brain. J Neuroimmunol 1997; 75: 69-74. Floyd RA, Krueger JM. Diurnal variation of TNFc~ in the rat brain. NeuroReport 1997; 8: 915-918. Kapas L, Fang J, Krueger JM. Inhibition of nitric oxide synthesis inhibits rat sleep. Brain Res 1994; 664: 189-196. Kapas L, Shibata M, Kimura M, Krueger JM. Inhibition of nitric oxide synthesis suppresses sleep in rabbits. Am J Physiol 1994; 266: R151-R157. Walter JS, Meyers P, Krueger JM. Microinjection of Interleukin-1 Into Brain: Seperation of Sleep and Fever Responses. Physiology and Behavior 1989; 45: 169-176. De Sarro G, Gareri P, Sinopoli VA, David E, Rotiroti D. Comparative, Behavioral and Electrocortical Effects of Tumor Necrosis Factor-or and Interleukin-1 Microinjected into the Locus Coeruleus of Rat. Life Sciences 1996; 60(8): 555-564. Terao A, Matsumura H, Saito M Interleukin-1 Induced Slow Wave Sleep at the Prostaglandin D2- Sensitive Sleep Promoting Zone in the Rat Brain. Journal of Neuroscience 1998; 18(6): 6599-6607. Slisli Y, De Beaurepaire R. Interleukin-l[5 and calcitonin, but not corticotropin-releasing factor, alter sleep cycles when injected into the rat hypothalamic lateral paraventricular area. Neuroscience Letters 1999; 265: 29-32. Vitkovia L, Bockaert Jacque C. "Inflammatory" Cytokines: Neuromodulators in Normal Brain? Journal of Neurochemistry. 2000; 74:457-471 Schobitz B, Holsboer F, De Kloet E. Gene Expression and function of Interleukin-1, interleukin-1 and tumor necrosis factor in the brain. Progress in Neurobiology 1994; 44: 397-432. Szymusiak R. Magnocellulad Nuclei of the Basal Forebrain: Substrates of Sleep and Arousal Regulation. Sleep 1995; 18(6): 478-500. Sterman MB, Clemente CD. Forebrain inhibitory mechanisms: sleep patterns induced by basal forebrain stimulation in the behaving cat. Experimental Neurology 1962; 6: 103-107. Lucas EA, Sterman MB. Effect of a forebrain lesion on the polycyclic sleep-wake cycle and sleep-wake patterns in the cat. Experimental Neurology 1975; 46: 368-388.
266
23. 24.
25. 26.
27. 28.
29.
30. 31.
32. 33.
34. 35. 36.
37. 38.
39.
40. 41.
Szymusiak R, McGinty D. Sleep suppression following kainic acid-induced lesions of the basal forebrain. Experimental Neurology 1986; 94: 598-614. Gritti I, Mainville L, Jones BE. Codistribution of GABA -- with acetylcholinesynthesizing neurons in the basal forbrain of the rat. Journal of Comparative Neurology 1993; 329: 438-457. Sherin JE, Shiromani PJ, McCarley RW, Saper CB. Activation of ventrolateral preoptic neurons during sleep. Science 1996; 271: 216-219. Gallopin T, Fort P, Eggermann E, Cauli B, Luppi PH, Rossier J, Audinat E, Muhlethaler M, Serafin M. Identification of sleep-promoting neurons in vitro Nature 2000; 404: 992-995. Garcia-Garcia F, Drucker-Colin R. Endogenous and Exogenous Factors on Sleep Wake Cycle Regulation. Progress in Neurobiology 1999; 58: 297-314. De Simoni MG, Imeri L, Dr Matteo W, Perego C, Simard S, Terrazzino S. Sleep Regulation: Interactions among cytokines and classical neurotransmitters. Advances in Neuroimmunology 1995; 5: 189-200. Morairty SR, Rainnie DG, McCarley RW, Greene RW. Adenosine Mediated Disinhibition of the Ventrolateral Preoptic Area 1998. On-line Proceedings of the 5th Internet World Congress on Biomedical Sciences '98 at McMaster University, Canada (available from URL: http://www.mcmaster.ca/inabis98/semba/morairty 0461/index.html). Krueger JM, Obal F Jr, Fang J. Humoral regulation of physiological sleep: cytokines and GHRH. Journal of Sleep Research 8 Suppl 1999; 1: 53-9. Edgar DM, Dement WC, Fuller CA. Effect of SCN lesions on sleep in squirrel monkeys: evidence for opponent processes in sleep-wake regulation. Journal of Neuroscience 1993; 13: 1065-1079. Banks WA, Kastin JA, Durham DA. Bidirectional transport of interleukin-1 alpha across the blood-brain barrier. Brain Res Bull 1989; 23: 433-7. Banks WA, Ortiz L, Plotkin SR, Kastin AJ. Human interleukin 1-c~, murine IL-lc~ and murine IL-I[5 are transported from the blood to brain in the mouse by a shared saturable mechanism. J Pharmac Exp Ther 1991; 259: 988-96. Banks WA, Kastin AJ. The interleukins-1 alpha, -1 beta and -2 do not acutely disrupt the murine blood-brain barrier. Int J ImmunoPharmacol 1992; 14: 629-36. Banks WA, Kastin A J, Gutierrez EG. Penetration of interleukin-6 across the murine blood-brain barrier. Neurosci Lett 1994; 179: 53-6. Chen G, Castro WL, Chow HH, Reichlin S. Clearance of ~25I-labeled interleukin-6 from brain into blood following intracerebroventricular injection in rats. Endocrinology 1997; 138: 4830-6. Gutierrez EG, Banks WA, Kastin AJ. Murine tumor necrosis factor alpha is transported from blood to brain in the mouse. J Neuroimmunol 1993; 47: 169-76. Chen G, Reichlin S. Clearance of [125I] tumor necrosis factor-or from the brain into the blood after intracerebroventricular injection in rats. Neuroimmunomodulation 1998; 5: 261-9. Kida S, Pantazis A, Weller RO. CSF drains directly from the subarachnoid space in nasal lymphatics in the rat. Anatomy, histology and immulogical significance. Neuropathology and Applied Neurobiology 1993; 19: 480-8. Dickstein JB, Moldofsky H, Lue FA, Hay JB. Intracerebroventricular injection of TNF-c~ promotes sleep and is recovered in cervical lymph. Am J Physiol 1998; 276: R1018-22. Cserr HF, Knopf PM. Cervical lymphatics, the blood-brain barrier and the immunoreactivity of the brain: a new view. Immunol Today 1992; 13: 507-12.
267
42.
43.
44. 45.
46.
47.
48. 49. 50.
51.
52.
53. 54. 55. 56. 57. 58. 59. 60. 61.
DeSimoni MG, De Luigi A, Gemma L, Sironi M, Manfridi A, Ghezzi E Modulation of systemic interleukin-6 induction by central interleukin-1. Am J Physiol 1993; 265: R739-42. Sundar SK, Becker KJ, Cierpial MA, Carpenter MD, Rankin LA, Fleener SL, Ritchie JC, Simson PE, Weiss JM. Intracerebroventricular infusion of interleukin 1 rapidly decreases peripheral cellular immune responses. Proc Natl Acad Sci 1998; 86: 6398-402. Sapolsky R, Rivier C, Yamamoto G, Plotsky P, Vale W. Interleukin-1 stimulates the secretion of hypothalamic corticotropin-releasing factor. Science 1987; 238: 522-4. Bernton EW, Beach JE, Holaday JW, Smallridge RC, Fein HG. Release of multiple hormones by a direct action of interleukin-1 on pituitary cells. Science 1987; 238: 519-21. Vellucci SV, Parrott RF, da Costa AC, Ohkura S, Kendrick KM. Increased body temperature, cortisol secretion, and hypothalamic expression of c-fos, corticotrophin releasing hormone and interleukin-l[5 mRNAs following central administration of interleukin-l[5 in the sheep. Mol Brain Res 1995; 29: 64-70. Lue FA, Bail M, Jephthah-Ochola J, Carayanniotis K, Gorczynski R, Moldofsky H. Sleep and cerebrospinal fluid interleukin-l-like activity in the cat. Int J Neurosci 1988; 42: 179-183. Moldofsky H, Lue FA, Eisen J, Keystone E, Gorczynski R. The relationship of interleukin-1 and immune function to sleep in humans. Psychosom Med 1986; 48: 309-318. Gudewill S, Pollmacher T, Vedder H, Schreiber W, Fassbender K, Holsboer E Nocturnal plasma levels of cytokines in healthy men. Eur Arch Psy Clin N 1992; 242: 53-56. Covelli V, Massari F, Fallarca C, Muno I. Interleukin-1 beta and beta endorphin circadian rhythms are inversely related in normal and stress-altered sleep. Int J Neurosci 1992; 63: 299-305. Hohagen F, Timmer J, Weyerbrock A, Fritsch-Montero R. Cytokine production during sleep and wakefulness and its relationship to cortisol in healthy humans. Neuropsychobiol 1993; 28: 9-16. Darko DF, Miller JC, Gallen C, White J, Koziol J, Brown SJ, et al. Sleep electroencephalogram delta frequency amplitude, night plasma levels of tumor necrosis factor c~, and human immunodeficiency virus infection. Proc Natl Acad Sci 1995; 92: 12080-12084. Born J, Lange T, Hansen K, Molle M, Fehm HL. Effects of sleep and circadian rhythm on human circulating immune cells. J Immunol 1997; 158: 4454-4464. Haus E, Smolensky MH. Biologic rhythms in the immune system. Chronobiology International 1999; 16: 581-622. Dickstein JB, Hay JB, Lue FA, Moldofsky H. The relationship of lymphocytes in the blood and in lymph to sleep wake states in sheep. Sleep 2000: 23: 185-90. Moldofsky H, Jiang CG, Lue FA, Poplonski L, Gorczynski RM. Circadian sleep-wake related changes in natural killer cells and T cells in humans. J Sleep Res 1996; 5: 144. Dickstein JB, Moldofsky H. Sleep, Cytokines and Immune Function. Sleep Medicine Reviews 1999; 3: 219-228. Mourtazev MS, Kemp B, Zwinderman AH, Kampuisen HAC. Age and gender affect different characteristics of slow waves in the sleep EEG. Sleep 1995; 18: 557-564. Manber R, Bootzin RR. Sleep and the menstrual cycle. Health Psychol 1997; 16: 209-214. Driver HS, Baker FC. Menstrual factors in sleep. Sleep Medicine Reviews 1998; 2: 213-229. Moldofsky H, Lue F, Shahal B, Cheng J, Gorczynski R. Diurnal sleep/wake immune functions during the menstrual cycle of healthy young women. J Sleep Res 1995; 4: 150-159.
268
62. 63.
64. 65. 66.
67.
68.
69.
70.
71.
72. 73. 74. 75. 76. 77. 78. 79. 80.
Agarwal SK, Marshall GD Jr. Perimenstrual alterations in type-I/type-2 cytokine balance of normal women. Annals of Allergy, Asthma, & Immunology 1999; 83: 222-228. Palmblad J, Cantell K, Strander H, Froberg J, Karlsonn CG, Levi L, et al. Stressor exposure and immunological response in man: Interferon -- producing capacity and phagocytosis. J Psychosom Res 1976; 20: 193-199. Palmblad J, Petrini B, Wasserman J, Akerstedt T. Lymphocyte and granulocyte reactions during sleep deprivation. Psychosom Med 1979; 41: 273-278. Moldofsky H, Lue FA, Davidson JR, Gorczynski R. Effects of sleep deprivation on human immune functions. FASEB 1989; 3: 1972-1977. Heiser P, Dickhaus B, Schreiber W, Clement HW, Hasse C, Hennig J, Remschmidt H, Krieg JC, Wesemann W, Opper C. White blood cells and cortisol after sleep deprivation and recovery sleep in humans. European Archives of Psychiatry & Clinical Neuroscience. 2000; 250: 16-23. Moldofsky H, Lue FA, Jephthah-Ochola J, Carayanniotis K, Gorczynski R. The effect of 64 hour wakefulness on immune functions and plasma cortisol in humans. In: Horne J, editor. Sleep 88. Stuttgard, New York: Gustav Fischer Verlag, 1989: 185-187. Irwin M, Mascovich A, Gillin JC, Willoughby R, Pike J, Smith TL. Partial sleep deprivation reduces natural killer cell activity in humans. Psychosom Med 1994; 56: 493-498. Irwin M, McClintick J, Costlow C, Fortner M, White J, Gillin JC. Partial night sleep deprivation reduces natural killer and cellular immune responses in humans. FASEB 1996; 10: 643-653. Dinges DF, Douglas SD, Zaugg L, Campbell DE, McMann JM, Whitehouse WG, et al. Leukocytosis and natural killer cell function parallel neurobehavioural fatigue induced by 64 hours of sleep deprivation. J Clin Invest 1994; 93: 1930-1939. Heiser P, Dickhaus B, Opper C, Schreiber W, Clement HW, Hasse C et al. Platelet serotonin and interleukin-1 beta after sleep deprivation and recovery sleep in humans. J Neural Transm 1997; 104: 1049-1058. Rechtschaffen A, Gilliland MA, Bergmann BM, Winter JB. Physiological correlates of prolonged sleep deprivation in rats. Science 1983; 221: 182-184. Everson CA. Sustained sleep deprivation impairs host defense. Am J Physiol 1993; 265:R1148-R1154. Everson CA, Toth LA. Systemic bacterial invasion induced by sleep deprivation. American Journal of Physiology 2000; 278: R905-916. Benca RM, Kushida CA, Everson CA, Kalski R, Bergmann BM, Rechtschaffen A. Sleep deprivation in the rat: VII Immune function. Sleep 1989; 12: 47-52. Bregmann BM, Rechtschaffen A, Gilliland MA, Quintans J. Effect of extended sleep deprivation on tumor growth in rats. Am J Physiol 1996; 271: R1460-R1464. Brown R, Pang G, Husband AJ, King MG. Suppression of immunity to influenza virus infection in the respiratory tract following sleep disturbance. Reg Immunol 1989; 2: 321-325. Renegar K, Floyd R, Krueger JM. Effects of sleep deprivation on serum influenza specific IgG. Sleep 1998; 21: 19-24. Renegar KB, Floyd RA, Krueger JM. Effects of short-term sleep deprivation on murine immunity to influenza virus in young adult and senescent mice. Sleep 1998; 21: 241-248. Toth LA, Rehg JE. Effects of sleep deprivation and other stressors on the immune and inflammatory responses of influenza-infected mice. Life Sci 1998; 63: 701-709.
New Foundation of Biology Edited by I. Berczi and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
269
Interactions Between the Immune System and the Testis ~
DAVID K. POMERANTZ
Departments of Physiology, Obstetrics and Gynaecology, University of Western Ontario, London, Canada N6A-5C1
EVIDENCE FOR UNIQUE IMMUNE FUNCTION ASSOCIATED WITH THE TESTIS The classic view of physiologic regulation of the testis gave the hypothalamic-pituitary unit a dominant role. Control of testosterone secretion by the Leydig cell was achieved by luteinizing hormone (LH), regulation of the spectrum of functions of the Sertoli cell was ascribed to follicle-stimulating hormone (FSH), and negative feedback control of secretion both gonadotropins was accomplished by steroid and protein products of the testis. The model continues to provide adequate understanding of the control of parameters of testis function such as total testicular output of testosterone. We and others have noted that the above view is incomplete; it ignores the regulation of steroidogenesis and gametogenesis by the nervous system, immune system and intragonadal factors. Relatively recent studies of the regulation of steroidogenesis have pursued the theme that the seminiferous tubule directly influences androgen production in adjacent Leydig cells; the supporting evidence has been reviewed [1, 2, 3, 4, 5, 6]. Work from our laboratory showed that Leydig cells from rats treated in utero to eliminate germ cells, were more responsive to LH in vitro than control Leydig cells and that this difference was not acquired until the first wave of spermatogenesis was under way [7, 8]. We also found that disruption of spermatogenesis in adult rats by cryptorchidism or efferent-duct-ligation (EDL) enhanced Leydig cells responsiveness to LH and unilateral cryptorchidism or EDL caused mainly, but not solely, ipsilateral alterations in Leydig cells function [9]. The combined data led us to hypothesize that the Leydig cell's response to disrupted spermatogenesis was mediated in part by local factors [6]. Papadopoulos' group was the first to identify such a factor when they described a procathepsin L/TIMP-1 complex of Sertoli cell origin that stimulated testosterone production [10]. Many of the model systems used to disrupt gametogenesis in the seminiferous tubules also elicit an inflammatory-like response. In such situations immune activation usually occurs. Thus, it was possible that the changes in testosterone production noted above could be attributed to changes in activity of the local, intragonadal components of the immune system. Inspection of existing literature reveals abundant examples of the potential for interactions between the immune system and reproductive tracts. An elegant summary was presented by Medawar in his Nobel lecture, in which he addressed the apparent enigma of the fetal allograft [11]. Here, I confine the citations to limited number of the extensive examples encountered in the male. 1 Researchfrom the author's laboratory was funded by the Medical Research Council of Canada (Grant MT-12593).
270
Studies performed first in 1898 by Calzolari are among the earliest cited [12]. Castration of rabbits was associated with increased size of the thymus. This is consistent with observations made nearly a century later by Viselli et al. who reported increased T-cell production of IL-2 and IFN~, after orchidectomy in mice [13]. Spermatic cord torsion produces ischemia, stasis of blood flow and severe pain. In the past treatment involved untwisting of the affected testis and orchidopexy with the hope of reperfusion and regeneration of function. However, torsion of only a few hours duration not only causes severe degeneration of the affected testis, but tubular diameter in the contralateral testis and later fertility can be reduced [14, 15]. Prompt removal of the torsed testis or suppression of T-lymphocyte function [16, 17, 18] alleviated many of the morphologic changes in the contralateral testis and improved fertility in rats. Post-torsion damage was more severe and its "immune" management more effective in pubertal rather than adult rats [19, 20]. Thus, torsion-induced autoimmune damage may be communicated to the contralateral gonad in part by circulating T-cells and the behaviour of the immune system may depend on the state of sexual maturity. Fertility impairment occurs in 75% of men who had bilateral cryptorchidism and in 50% who had only unilateral cryptorchidism as youngsters [21]. It is accepted that increased testicular temperature elicits depletion of germ cells in the heated testis, however the mechanism for contralateral damage is unknown. Macrophage numbers in the cryptorchid testis increase [22, 23] and the incidence of apoptosis increases in primary spermatocytes concurrent with the testicular atrophy [24]. As well, unidentified testicular material that inhibits in vitro T-cell activation increases in abundance after induction of cryptorchidism [25]. Again, immunomodulatory factors induced by insult to the testis could influence steroid and germ cell production. The immune status of the testis may play an insidious role in leukemia. Relapse of acute lymphoblastic leukemia occurs at a significantly higher rate in males and occurs often in the testis. These testicular relapses are most common before completion of the initial wave of spermatogenesis and are very rare in adults [26]. In this case, the immune state of the testis can have catastrophic consequences and these may be a function of sexual maturity. Clearly, the immune status of the testis is dynamic and can be altered by normal development as well as disease. Andrologists and immunologists are intrigued by the fact that the testis of rodents is an "immunologically privileged tissue", i.e., one in which presentation of antigen elicits a subdued immune response in the presence of an otherwise competent immune system. Obviously an understanding of this naturally occurring immune tolerance is of great importance to understanding autoimmune disease and to improving success in organ transplantation. Reviews and pivotal reports are available [27, 28, 29, 30, 31, 32] and establish: that spermatogenic cells possess autoantigens; that most germ cell development (with production of autoantigens) occurs after the perinatal period in which the immune system has set its criteria for recognition of "self' and "non-self'; that these autoantigenic cells are tolerated in the normal gonad; that the afferent and efferent limbs of the immune response in the testis are intact because both antibody and cell mediated immune reactions can be induced; that allografts survive and function in the testis; and finally, that the cryptorchid testis is a more hospitable site for transplants than the scrotal testis. In other immunologically suppressed or privileged sites, such as the eye and brain, avascularity, lack of lymphatics, or presence of a blood-tissue barrier contributes to the state of "privilege". The interstitium of the testis is well perfused with blood and has a notable lymphatic drainage, therefore the blood-testis barrier, provided by adjacent Sertoli cells, is the primary means of isolating the majority of auto-immunogenic germ cells from the immune system. Nevertheless, this is an incomplete explanation. First, because successful intratesticular grafts are most likely placed outside the blood-testis barrier in the interstitial space [33]. Therefore, the immune system has access to such grafts. Second, because
271
activated T-cells from mice immunized with testis homogenate can transfer the experimental autoimmune orchitis to normal mice, indicating that immune responses to spermatogenic cells can occur in the presence of an intact blood-testis barrier and that the afferent arm of the immune response is normally suppressed [34, 35]. Third, because preleptotene spermatocytes and spermatogonia, which lie outside the blood-testis barrier and are exposed to the immune system, possess autoantigens and are tolerated [36, 37, 38]. In summary, the necessary elements to mount an immune response are present in or have access to the interstitial space, but this tissue is locally "immunosuppressed". The suppression could be due to failure of antigen presentation to T-lymphocytes, or inhibition of the normal proliferative and secretory responses of activated T-lymphocytes. The mechanisms underlying these phenomena remain an enigma. An enigma as perplexing as that of the "fetal allograft". A final example of interactions between testicular and immune function is found in the sexual dimorphism of the immune system's response to systemic trauma and hemorrhage. In a series of papers from Chaudry et al., [39, 40, 41, 42, 43] it was found that hemorrhagic shock alone or combined with soft tissue trauma caused depressed responses in both specific and innate arms of the immune system. It was later reported that gonadectomy prior to soft tissue trauma and hemorrhagic shock prevented the immune depression [44]. This result raised the intriguing possibility that the effectiveness of treatment of prostatic carcinoma by castration might involve multiple mechanisms. The traditional view would be that androgen deprivation aided regression of the androgen dependent tumor. However a second possibility is that the androgen deficient state following castration allowed a more vigorous immune attack on the malignant tissue.
2.
SELECTED CURRENT STUDIES
The innate, or non-specific, arm of the immune system consists of anatomic and physiologic barriers as well as cells which have endocytic, phagocytic and cytotoxic activity. Macrophages are well known as phagocytic cells, cytotoxic cells, and active secretors of a host of cytokines. Importantly, macrophages play an obligatory role in the specific component of the immune system because after digesting antigen within phagosomes they complex the fragments with class II MHC molecules for presentation to T-cells. This antigen presenting process is absolutely required for the development of the T-cell-mediated specific responses. In order for the phagocytic, cytotoxic and antigen presenting activities of macrophages to be expressed, a process of activation occurs and this is associated with secretion of several compounds. This non-specific component of the immune system has been given less attention by andrologists compared to the specific component. Given their known role in inflammatory processes, antigen presentation, and potential for contributing to the state of immune privilege in the testis, the testicular macrophage has begun to attract the attention of investigators. Macrophages are present in the interstitium of the testes of a number of species. They constitute up to 25% of the interstitial cells and are in direct contact with Leydig cells via cytoplasmic extensions of Leydig cells into invaginations of the macrophage plasma membrane. They are thought to be authentic macrophages because they have phagocytic properties and express class II MHC markers [22, 23, 45, 46, 47, 48]. Hume et al., [49] reported the presence of the macrophage marker F4/80 on cells that comprised 20% of cells in the interstitium of the mouse. Finally, the testis of the ram is bereft of macrophages and, based on attempts to transplant thyroid tissue to the testis, is not an immunologically privileged site [50].
272
In rodents, testicular macrophages are first detected in late gestation and their numbers increase steadily to plateau by adulthood. As macrophage numbers stabilize, interdigitation of Leydig cells and macrophage cell membranes is initiated and this is complete by adulthood [45, 48, 51]. Khan and Dorrington have suggested that macrophages may play a role in the proliferation of Leydig cells that occurs during puberty [52]. Procedures which disrupt spermatogenesis (vasectomy, cryptorchidism, fetal irradiation) have been associated with increased numbers and/or size of macrophages [22, 23, 53]. Apparently testicular macrophages can respond to gonadotropin because treatment of prepubertal rats with hCG causes premature increases in interstitial macrophage numbers [54]. FSH in vitro is reported to alter macrophage function [55, 56]. In the testis of the seasonally breeding hamster, macrophage responsiveness to FSH is influenced by photoperiod [57]. The potential for FSH effects on macrophages was recently been thrown into doubt when Hutson's laboratory reinvestigated their own studies on FSH-macrophage interactions and stated that mRNA for the FSH receptor was not detectable in testicular macrophages by in situ hybridization, nor by northern blot analysis of RNA obtained from purified macrophages [58]. Nonetheless, testicular macrophages appear to undergo changes which can be related to the functional milieu of the testis. Several lines of data obtained from in vivo experiments have been cited to support the assertion that testicular macrophages can alter function of the adjacent Leydig cells. Utilizing a model of surgically induced unilateral cryptorchidism, Berg [22] reported that the number and size of Leydig cells and testicular macrophages changed in concert. More direct demonstrations of the potential for Leydig-macrophage interactions were achieved by destruction of macrophages using silica powder [59] or liposome encapsulated clodronate [60]. In both cases a decreased testosterone production by Leydig cells was reported. An alternative approach was used by Cohen et al., [61]. These investigators used the osteopetrotic mouse model in which colony-stimulating factor 1 (CSF-1) production is defective. Associated with this defect is a reduction in numbers of all peripheral macrophages, including those residing in the testis. Testosterone production was reduced in these animals and could be restored toward normal by exogenous CSF-1. These three studies may be taken to show that macrophages are required to maintain normal steroidogenic activity by Leydig cells. However, an opposite outcome was obtained by Gaytan and co-workers [62] who also used clodronate to destroy macrophages, but found this intervention increased testosterone production. These in vivo studies may be complicated by alterations in feedback regulation of secretion of LH that would occur when testosterone synthesis changes. There are abundant examples of attempts to study macrophage-Leydig cell interactions with in vitro model systems. Yee and Hutson [56] were the first to report that media harvested from culture of testicular macrophages stimulated steroidogenesis in Leydig cells. Subsequent reports yielded variable results ranging from no effect [63] to an inhibitory effect of macrophage conditioned medium on testosterone production by Leydig cells [58, 64, 65, 66]. The reasons for such diverse results is unclear, however it is likely that the various approaches to isolation and culture of macrophages elicited varying degrees of activation [67]. The majority of workers have found that the macrophage products Interleukin-1 and 2, Interferon-c~, [3, and ~,, and TNFc~ inhibit androgen secretion. For a review see [68]. However, some have reported stimulatory effects of Interleukin or TNF~t on basal and/or stimulated androgen secretion [69, 70, 71]. The basis of these inconsistencies has not been resolved, but variations in experimental design can account for most discrepancies. It has also been reported that testicular macrophages produce a lipophilic compound that stimulates testosterone secretion [72] and recently it was determined that this compound is 25-hydroxycholesterol [73]. Its mechanism of action appears to be independent of the steroidogenic acute regulatory protein,
273
StAR [74]. Another important macrophage product, nitric oxide (NO), has recently received attention. It will become evident from this review that NO has the potential to mediate both normal and pathologic processes within the testis. In brief, NO is synthesized from L-arginine in a reaction catalysed by NO synthase (NOS). The enzyme exists in several isoforms, but in mammalian systems only three have been identified. In neuronal and endothelial cells two calcium/ calmodulin-dependent constitutive forms (cNOS) are found and thought to produce modest amounts of NO, the endothelial form (eNOS) is also found in non-vascular tissues [75, 76]. These isoforms are sensitive to intracellular calcium levels and believed to participate in neuronal function, regulation of blood flow, and other normal physiologic processes. The third isoform, inducible NOS (iNOS), which exists in macrophages and other cell types, is inducible by several cytokines and binds calmodulin with such high avidity to render iNOS expression independent of intracellular calcium levels [77, 78]. Macrophage activation results in prolonged release of large amounts of NO in the presence of cytokines or endotoxin until arginine substrate is exhausted or macrophage death [79, 80]. The cytotoxic and cytostatic activity of macrophages against bacteria, fungi and tumour cells is mediated in part by iNOS catalysed NO [81]. In addition NO is a known agent for apoptosis in pancreatic [3-cells, chondrocytes, and macrophage themselves [82, 83, 84]. As this field developed, a variety of arginine analogs were found which inhibit both isoforms of NOS; two often used compounds are NG-nitro-L arginine methyl ester (L- NAME) and amino-guanidine (AG). The latter is more selective for iNOS [85, 86, 87, 88, 89]. The importance of NO in male reproduction is recognized. The molecule is a neuronal mediator of penile erection [90]. In the human testis, the neuronal and endothelial constitutive isoforms of NOS protein have been detected by immunohistochemistry in Sertoli, Leydig and degenerating germ cells [91, 92]. A single report claimed that bacterial lipopolysaccharide (LPS) stimulated iNOS activity in cultured rat Sertoli cells [93]. In non-reproductive tissue NO contributes to endothelial damage occurring after ischemia/reperfusion and may interact with superoxide free radicals in this situation [94, 95]. It follows that NO may be involved in the tissue damage noted after induction of cryptorchidism, EDL or testicular torsion. We [96] have reported that both peritoneal macrophages and a murine macrophage cell line (RAW 264.7), when activated in vitro by sequential exposure to interferon gamma (IFN) and then bacterial lipopolysaccharide (LPS) for 24 hours, produced significant quantities of NO. When Leydig cells were introduced into the culture dishes at the end of the activation period (and the activating mixture removed), we found co-culture of Leydig cells with quiescent macrophages altered neither basal, nor LH-stimulated androgen production. Co-culture with either type of activated macrophage did not alter basal, but significantly reduced (by 50%) LH-stimulated androgen production. The inhibitory effect of activated macrophages on testosterone secretion by co-cultured Leydig cells was blocked by the nitric oxide synthase inhibitor N4-Nitro-LArginine Methyl Ester (L-NAME) as well as by the NO scavenger 2-(4-Carboxyphenyl)-4, 4, 5, 5-tetramethylimidazoline-l-oxyl-3-oxide (C-PTIO). The NO donor S-Nitroso-N-acetyl penicillamine (SNAP) significantly inhibited LH-stimulated androgen production by purified Leydig cells. Further studies led to the conclusion that the inhibition occurred, at least in part, at cholesterol-side chain cleavage (P450scc) as well as at 17~-hydroxylase/C17/20 lyase (P450c17) [96, 97]. We concluded that activation of macrophages results in decreased androgen production by co-cultured Leydig cells. The inhibition is mediated in part by macrophage-derived NO acting directly on the Leydig cells via inhibition of the P450 steroidogenic enzymes. Although it remains to be determined whether testicular macrophages can be similarly activated to inhibit testicular steroidogenesis, the results cited above extend the notion that molecules produced by the immune system have important effects on endocrine cells.
274
The T and B-lymphocytes of the active or specific component of the immune system are endowed with the ability to respond to the presentation of foreign antigens. This results in both cell mediated cytotoxic attack of foreign and altered "self" cells by an expanding (proliferating) clonal cohort of CD8 + cytotoxic T-cells and of CD4 + helper T-cells (T~-cells) which, via cytokines, stimulate B-cell proliferation and antibody production. Only T-lymphocytes are considered here. In attempts to understand the physiology of the immune cells within the testis, known testicular compounds have been evaluated for their effects on T-cell proliferation in an attempt to explain the immunosuppressed state. The effect of steroids, POMC-derived peptides, and arachadonic acid metabolites on these immune cells is equivocal with stimulation, inhibition, and no effect being reported; see [32] for specific citations. Importantly, undefined compounds also have immunoregulatory activity. Emoto et al., [98, 99] reported the presence of material in whole tissue extracts of mouse testes which inhibited in vitro proliferation of lymphocytes. Later P611~nen and coworkers [100, 101] described protein(s) of MW > 100 kDa in rat testicular extracts that inhibited in vitro lymphocyte proliferation and which they named "protectin". Selawry et al., [102] have reported that Sertoli cell-conditioned medium (SCM) contains a factor that inhibits T-cell proliferation. Conversely, tubule conditioned medium and testicular intersitial fluid contain an "interleukin-l-like" protein which augments lymphoproliferation [103, 104] and testicular interstitial fluid contains material of >30 kDa that stimulates lymphoproliferation [32]. Hedger et al., showed in vitro stimulation of lymphocyte proliferation by inhibin and reduction by activin [105]. A coherent picture does not arise from these studies; the testis may contain a mixture of immuno-stimulatory and inhibitory factors. The normal immuno-suppressed state may represent the net effect of these factors on the immune system of the testis and may be influenced by the functional state of the gonad. Fas (CD95, Apo-1) is a membrane protein which, upon contact with crosslinking antibodies or the natural Fas ligand (FasL) elicits rapid apoptosis in the FAS expressing cell. This has led to the notion of "death factors" and the "death gene" as components of this pathway for programmed cell death [106]. It is believed that the Fas/FasL system plays an important role in regulating clonal expansion and contraction in the immune system. Several groups have suggested that Fas mediates both activation induced T-cell death (viewed by some as a mechanism to eliminate autoreactive T-cells) as well as the cytolytic effects of CD8 + T-cells [107, 108, 109, 110, 111]. In papers submitted within weeks of one another, two groups suggested that Fas/FasL might be expressed in non-lymphoid tissue and be involved in immune privilege. Griffith et al., [112] noted that Fas+, but not Fas-, cells were killed by apoptosis when placed in the anterior chamber of the eye. Bellgrau's group [113] demonstrated that expression of FasL by Sertoli cells was necessary to prevent their rejection as allografts. Both groups contend that immune privilege can be explained in part by Fas/FasL systems mediating death of immune cells in these privileged sites. It is already known that NO can induce apoptosis in several cell types [82, 83, 84]. It was recently discovered that NO can up regulate Fas and apoptosis in vascular smooth muscle. Given that the Sertoli cell may express both constitutive and inducible NOS as well as FasL, we embarked on a series of experiments that examined the potential for NO involvement in Sertoli-Leydig cell interactions. We hypothesized that activation of Sertoli cells with cytokines and bacterial lipopolysaccharide would result in increased expression of induceable nitric oxide synthase (iNOS) and production of nitric oxide (NO); this in turn would act directly on adjacent Leydig cells to alter synthesis of androgens. Sertoli cells from immature mice were activated in vitro with a mixture of 500 U/mL, Interferon% 1 ~g/mL, E.
275
coli lipopolysccharide, and 500 U/mL Tumor-Necrosis Factor-~t (I/L/T). The Sertoli cells responded with increased expression of iNOS mRNA and increased production of NO. I'--!
Control
BB LH 6.0
m
(9
o
4.0
e~ o .,m
E
~
z.0
01
0.0 None
Quiescent
Activated
Quiescent
Activated
L-NAME Figure 1. The effect of activation of Sertoli cells on androgen production by mouse Leydig cells in co-culture. Sertoli cells were untreated or activated for 24 hour with a mixture of I/L/T which was then removed at the same time Leydig cells were added. Activated Sertoli cells caused a significant (P < 0.001) inhibition of LH-stimulated androgen production. This inhibition was completely reversed by inhibition of NO production with L - N A M E .
As shown in Figure 1, Sertoli cells that were induced to NO production had an inhibitory effect on the production of testosterone by Leydig cells that were co-cultured with these activated Sertoli cells. These results have led us to suggest that exposure of Sertoli cells to agents that are associated with inflammation and immune activation result in induction of iNOS expression, and an attendant NO release. This NO production may account for a significant component of the decreased androgen production noted in neighboring Leydig cells. Of particular interest is the similarity of response of a somatic cell-the Sertoli cell, and an immune cell-the macrophage to activation. These observations provide an additional entry in the lengthening catalogue of examples of interactions and similarity of physiologic regulation between immune cells and cells of the reproductive and endocrine systems.
REFERENCES 1. 2.
3.
Aoki A, Fawcett DW. Is there a local feedback from the seminiferous tubules affecting activity of the Leydig cells? Biol Reprod. 1978; 19: 144-158. Bergh A. Local differences in Leydig cell morphology in the adult rat testis: Evidence for a local control of Leydig cells by adjacent seminiferous tubules. Int J Androl 1982; 5: 325-330. de Kretser DM. Sertoli cell-Leydig cell interaction in the regulation of testicular function. Int J Androl Suppl 1982; 5: 11-17.
276
.
.
.
.
.
.
10.
11. 12.
13. 14. 15.
16. 17. 18.
19. 20. 21. 22. 23. 24.
Sharpe RM, Cooper I. Testicular interstitial fluid as a monitor for changes in the intratesticular environment in the rat. J Reprod Fertil 1983; 69: 125-135. Sharpe RM. Intratesticular factors controlling testicular function. Biol Reprod. 1984; 30: 29-49. Pomerantz DK, Jansz GF. Evidence for intratesticular factors which mediate the response of Leydig cells to disruption of spermatogenesis. In: Endocrinology and Physiology of Reproduction. P.C.K. Leung, D.T. Armstrong, K.B. Ruf, W.H. Moger and H.G. Friesen, eds. Plenum Press, New York & London 1987; 233-241. Jansz GF, Pomerantz DK. Fetal irradiation increases androgen production by dispersed Leydig cells of the rat. J Androl 1984; 5: 344-350. Jansz G.F, Pomerantz DK. The effect of prenatal treatment with busulfan on in vitro androgen production by testes from rats of various ages. Can J Physiol Pharmacol 1985; 63:1155-1158. Jansz GF and Pomerantz DK. A comparison of Leydig cell function after unilateral and bilateral cryptorchidism and efferent-duct-ligation. Biol Reprod 1986; 34:316-321. Boujrad N, Ogwuegbu SO, Garnier M, Lee CH, Martin B M and Papadopoulos V. Identification of a stimulator of steroid hormone synthesis isolated from testis. Science 1995; 268: 1609-1612. Medawar PB. The Nobel Lectures in Immunology. The Nobel Prize for Physiology or Medicine, 1960. Immunological tolerance. Scand J Immunol 1991; 33: 337-344. Angele MK, Ayala A, Cioffi WG, Bland KI and Chaudry IH. Testosterone: the culprit for producing splenocyte immune depression after trauma hemorrhage. Am J Physiol 1998; 274: Pt 1: C1530-6. Viselli SM, Stanziale S, Shults K, Kovacs WJ and Olsen NJ. Castration alters peripheral immune function in normal male mice. Immunology 1995; 84: 337-342. Cosentino MJ, Rabinowitz R, Valvo JR and Cockett ATK. The effect of prepubertal spermatic cord torsion on subsequent fertility in rats. J Androl 1984; 5: 93-98. Cosentino MJ, Nishida M, Rabinowitz R and Cockett ATK. Histological changes occurring in the contralateral testes of prepubertal rats subjected to various durations of unilateral spermatic cord torsion. J Urol 1985; 133:906-911. Nagler HM, White RD. The effect of testicular torsion on the contralateral testis. J Urol 1982; 128: 1343-1348. Madarikan B A. Testicular salvage following spermatic cord torsion. J Pediatr Surg. 1987; 22: 231-234. Pakyz RE, Heindel RM, Kallish M and Cosentino MJ. Spermatic cord torsion: Effects of cyclosporine and prednisone on fertility and the contralateral testis in the rat. J Androl 1990; 11: 401-408. Turner TT. On unilateral testicular and epididymal torsion: no effect on the contralateral testis. J Urol 1987; 138: 1285-1290. Heindel RM, Pakyz RE and Cosentino MJ. Spermatic cord torsion. Contralateral testicular degeneration at various ages in the rat. J Androl 1990; 11:506-513. Kogan SJ. Fertility in cryptorchidism. An overview in 1987. Eur J Pediatr 1987; 146: $21- $24. Bergh A. Effect of cryptorchidism on the morphology of testicular macrophages: Evidence for a Leydig cell-macrophage interaction in the rat testis. Int J Androl 1985; 8: 86-96. Niemi M, Sharpe RM and Brown WR. Macrophages in the interstitial tissue of the rat testis. Cell Tissue Res 1986; 243: 337-344. Shikone T, Billig H and Hsueh AJ. Experimentally induced cryptorchidism increases
277
apoptosis in rat testis. Biol Reprod 1994; 51: 865-872. Sainio-Prll~inen S, Prll~inen P and Setchel BP. Testicular immunosuppressive activity in experimental hypogonadism and cryptorchidism. J Reprod Immunol 1991; 20: 59-72. 26. Cosentino MJ, Nishida M, Rabinowitz R and Cockett ATK. Histopathology of prepubertal rat testes subjected to various durations of spermatic cord torsion. J Androl 1986; 7: 23-31. 27. Tung KSK, Teuscher C and Meng AL. Autoimmunity to spermatozoa and the testis. Immunol Rev 1981; 55: 217-255. 28. Head JR, Billingham RE. Immune priviledge in the testis. I. Basicparameters of allograft survival. Transplantation 1983a; 36: 423-431. 29. Head JR, Billingham RE. Immune privilege in the testis. II. Evaluation of potential local factors. Transplantation 1985; 40: 269-275. 30. Alexander NJ, Anderson DJ. Immunology of semen. Fertil Steril 1987; 47: 192-205. 31. Hedger MP. The testis: an "immunologically suppressed" tissue? Reprod Fertil Dev 1989; 1: 75-79. 32. Hedger MP, Qin JX, Robertson DM and de Kretser DM. Intragonadal regulation of immune system functions. Reprod Fertil Dev 1990; 2: 263-280. 33. Head JR, Neaves WB and B illingham RE. Reconsideration of the lymphatic drainage of the rat testis. Transplantation 1983b; 35: 91-95. 34. Mahi-Brown CA, Yule TD and Tung KSK. Adoptive transfer of murine autoimmune orchitis to naive recipients with immune lymphocytes. Cell Immunol 1987; 106: 408-419. 35. Tung KSK, Yule TD, Mahi-Brown CA and Listrom MB. Distribution of histopathology and Ia positive cells in actively induced and passively transferred experimental autoimmune orchitis. J Immunol 1987; 138: 752-759. 36. Yule TD, Montoya GD, Russell LD, Williams TM and Tung KSK. Autoantigenic cells exist outside the blood testis barrier. J Immunol 1988; 141: 1161-1167. 37. Yule TD, Mahi-Brown CA and Tung KSK. Role of testicular autoantigens and influence of lymphokines in testicular autoimmune disease. J Reprod Immunol 1990; 18: 89-103. 38. Saari T, Jahnukainen K and Prll~inen P. Autoantigenicity of the basal compartment of the seminiferous tubules in the rat. J Reprod Immunol 1996; 31: 65-79. 39. Stephan RN, Kupper TS, Geha AS, Baue AE and Chaudry LH. Hemorrhage without tissue trauma produces immunosuppression and enhances susceptibility to sepsis. Arch Surg 1987; 122: 62-68. 40. Ayala A, Perrin MM and Chaudry LH. Defective macrophage antigen presentation following haemorrhage is associated with the loss of MHC class II (Ia) antigens. Immunology 1990; 70: 33-39. 41. Chaudry IH, Ayala A, Ertel W and Stephan RN. Hemorrhage and resuscitation: immunological aspects. Am J Physiol 1990; 259: Pt 2: R663-78. 42. Ayala A, Lehman DL, Herdon CD and Chaudry IH. Mechanism of enhanced susceptibility to sepsis following hemorrhage. Interleukin-10 suppression of T-cell response is mediated by eicosanoid-induced interleukin-4 release. Arch Surg 1994; 129: 1172-1178. 43. Zellweger R, Ayala A, DeMaso CM and Chaudry IH. Trauma-hemorrhage causes prolonged depression in cellular immunity. Shock 1995; 4: 149-153. 44. Wichmann MW, Zellweger R, DeMaso CM, Ayala A and Chaudry IH. Mechanism of immunosuppression in males following trauma-hemorrhage. Critical role of testosterone. Arch Surg 1996; 131: 1186-1191. 45. Christensen AK, Gillim SW. The correlation of fine structure and function in steroid secreting cells, with emphasis on those of the gonads. In: The gonads. KW McKerns, ed. 25.
278
46.
47. 48. 49.
50. 51. 52. 53.
54. 55. 56. 57.
58. 59.
60.
61.
62.
63. 64.
Appleton-Century-Crofts, New York, 1996; 415-488. Fawcett DW, Neaves WB and FloRes MN. Comparative observations on interbular lymphatics and the organization of the interstitial tissue of the mammalian testis. B iol Reprod 1973; 9: 500-532. Dirami G, Poulter LW and Cooke B A. Separation and characterization of Leydig cells and macrophages from rat testes. J Endocrinol 1991; 130: 357-365. Hutson JC. Development of cytoplasmic digitations between Leydig cells and testicular macrophages of the rat. Cell Tissue Res 1992; 267: 385-389. Hume DA, Halpin D, Charlton H and Gordon S. The mononuclear phagocyte system of the mouse defined by immunohistochemical localization of antigen F4/80: Macrophages of endocrine organs. Proc Natl Acad Sci USA 1984; 81: 4174-4177. Maddocks S, Setchell BP. The physiology of the endocrine testis. Oxf. Rev Reprod Biol 1988; 10: 53-123. Hutson JC. Changes in the concentration and size of testicular macrophages during development. Biol Reprod 1990; 43: 885-890. Khan SA, Khan SJ and Dorrington JH. Interleukin-1 stimulates deoxyribonucleic acid synthesis in immature rat Leydig cells in vitro. Endocrinology 1992; 131: 1853-1857. Geierhaas B, Bornstein SR, Jarry H, Scherbaum WA, Helrmann M and Pfeiffer EF. Morphological and hormonal changes following vasectomy in rats, suggesting a functional role for Leydig-cell associated macrophages. Horm Metab Res 1991; 23: 373-378. Raburn DJ, Coquelin A and Hutson JC. Human chorionic gonadotropin increases the concentration of macrophages in neonatal rat testis. Biol Reprod 1991; 45: 172-177. Yee JB, Hutson JC. Effects of testicular macrophage-conditioned medium on Leydig cells in culture. Endocrinology 1985b; 116: 2682-2684. Yee JB, Hutson JC. Biochemical consequences of FSH binding to testicular macrophages. Biol Reprod 1985a; 32: 872-879. Mayerhofer D, Mayerhofer A and Bartke A. Isolation and culture of testicular macrophages from a seasonally breeding species, Phodopus sungorus. Evidence for functional differences between macrophages from active and regressed testes. Int J Androl 1992; 15: 263-281. Carpenter AM, Lukyanenko YO, Lee VH and Hutson JC. FSH does not directly influence testicular macrophages. J Androl 1998; 19: 420-427. Huhtaniemi I, Nikula H, Lehtonen E and Hovatta O. Testicular morphology and endocrine function after silica-induced damage of interstitial macrophages. In: Molecular and Cellular Endocrinology of the Testis. M Stefanini, M Conti, R Geremia and E Ziparo, eds. Elsevier, Amsterdam 1986; 57-62. Bergh A, Damber JE and van Rooijen N. Liposome-mediated macrophage depletion: an experimental approach to study the role of testicular macrophages in the rat. J Endocrinol 1993; 136: 407-413. Cohen PE, Chisholm O, Arceci RJ, Stanley ER and Pollard JW. Absence of colonystimulating factor-1 in osteopetrotic (csfmop/csfmop) mice results in male fertility defects. Biol Reprod 1996; 55: 310-317. Gaytan F, Bellido C, Morales C, Garcia M, van Rooijen N and Aguilar E. In vivo manipulation (depletion versus activation) of testicular macrophages: central and local effects. J Endocrinol 1996; 150: 57-65. Bruot BC, Clemens JW. Regulation of testosterone production in the adjuvant- induced arthritic rat. J Androl 1992; 13: 87-92. Sun XR, Risbridger GP. Site of macrophage inhibition of luteinizing hormone-stimulated testosterone production by purified leydig cells. Biol Reprod 1994; 50: 363-367.
279
65.
66. 67. 68. 69. 70.
71.
72.
73.
74.
75. 76. 77.
78.
79. 80. 81. 82. 83. 84.
Watson ME, Newman RJ, Payne AM, Abdelrahim M and Francis GL. The effect of macrophage conditioned media on Leydig cell function. Ann Clin Lab Sci 1994; 24: 84-95. Onami S, Matsuyama S, Nishihara M and Takahashi M. Splenic macrophages can modify steroidogenesis of Leydig cells. Endocr J 1996; 43: 477-485. Hutson JC. Interactions between testicular macrophages and Leydig cells. J Androl 1998; 19: 394-398. Hales DB. Leydig Cell-Macrophage Interactions: An Overview. In: The Leydig Cell. AH Payne, MP Hardy and LD Russell, eds. Cache River Press, Vienna, IL, 1996; 451-465. Verhoeven G, Cailleau J, Van-Damme J and Billiau A. Interleukin-1 stimulates steroidogenesis in cultured rat Leydig cells. Mol Cell. Endocrinol 1988; 57:51-60. Warren DW, Pasupuleti V, Lu Y, Platler BW and Horton R. Tumor necrosis factor and interleukin-1 stimulate testosterone secretion in adult male rat Leydig cells in vitro. J Androl 1990; 11: 353-360. Moore C, Moger WH. Interleukin-l[3-induced changes in androgen and cyclic adenosine 3', 5'-monophosphate release in adult rat Leydig cells in culture. J Endocrinol 1991; 129:381-390. Hutson JC, Garner CW and Doris PA. Purification and characterization of a lipophilic factor from testicular macrophages that stimulates testosterone production by Leydig cells. J Androl 1996; 17: 502-508. Nes WD, Lukyanenko YO, Jia ZH, Quideau S, Howald WN, Pratum TK, West RR and Hutson JC. Identification of the lipophilic factor produced by macrophages that stimulates steroidogenesis. Endocrinology 2000. Mar; 141 (3): 953-958. Lukyanenko YO, Carpenter AM, Brigham DE, Stocco DM and Hutson JC. Regulation of Leydig cells through a steroidogenic acute regulatory protein-independent pathway by a lipophilic factor from macrophages. J Endocrinol 1998; 158: 267-275. Moncada S, Palmer RM and EA. Higgs. Nitric oxide: physiology, pathophysiology, and pharmacology. Pharmacol Rev 1991; 43: 109-142. Knowles RG, Moncada S. Nitric oxide synthases in mammals. [Review]. Biochem J 1994; 298: 249-258. Ding AH, Nathan CF and Stuehr DJ. Release of reactive nitrogen intermediates and reactive oxygen intermediates from mouse peritoneal macrophages. Comparison of activating cytokines and evidence for independent production. J Immunol 1988; 141: 2407-2412. Cho HJ, Xie QW, Calaycay J, Mumford RA, Swiderek KM, Lee TD and Nathan C. Calmodulin is a subunit of nitric oxide synthase from macrophages. J Exp Med 1992; 176: 599-604. Stuehr DJ, Marletta MA. Synthesis of nitrite and nitrate in murine macrophage cell lines. Cancer Res 1987a; 47: 5590-5594. Stuehr DJ, Marletta MA. Induction of nitrite/nitrate synthesis in murine macrophages by BCG infection, lymphokines, or interferon-gamma. J Immunol 1987b; 139:518-525. Hibbs JB Jr, Taintor RR and Vavrin Z. Macrophage cytotoxicity: Role for L- arginine deiminase and imino nitrogen oxidation to nitrite. Science 1987; 235: 473-476. Sarih M, Souvannavong V and Adam A. Nitric oxide synthase induces macrophage death by apoptosis. Biochem Biophys Res Commun 1993; 191: 503-508. Blanco FJ, Ochs RL, Schwarz H and Lotz M. Chondrocyte apoptosis induced by nitric oxide. Am J Pathol 1995; 146: 75-85. Kaneto H, Fujii J, Seo HG, Suzuki K, Matsuoka T, Nakamura M, Tatsumi H, Yamasaki
280
Y, Kamada T and Taniguchi N. Apoptotic cell death triggered by nitric oxide in pancreatic beta-cells. Diabetes 1995; 44: 733-738. 85. Griffiths MJ, Messent M, MacAllister RJ and Evans TW. Aminoguanidine selctively inhibits inducible nitric oxide synthase. Br. J Pharmacol 1993; 110: 963-968. 86. Hasan K, Heesen BJ, Corbett JA, McDaniel ML, Chang K, Allison W, Wolffenbuttel BH, Williamson JR and Tilton RG. Inhibition of nitric oxide formation by guanidines. Eur J Pharmacol 1993; 249:101-106. 87. Moncada S and Higgs A. The L-arginine-nitric oxide pathway. [Review]. N Engl J Med 1993; 329: 2002-2012. 88. Joly GA, Ayres M, Chelly F and Kilbourn RG. Effects of NG-methyl-L- arginine, NG-nitro-L-arginine, and aminoguanidine on constitutive and inducible nitric oxide synthase in rat aorta. Biochem Biophys Res Commun 1994; 199: 147-154. 89. Nathan C, Xie QW. Nitric oxide synthases: roles, tolls, and controls. [Review]. Cell 1994; 78: 915-918. 90. Burnett AL, Lowenstein CJ, Bredt DS, Chang TS and Snyder SH. Nitric oxide: a physiologic mediator of penile erection. Science 1992, 257:401-403. 91. Davidoff MS, Middendorff R, Mayer B and Holstein AF. Nitric oxide synthas (NOS-I) in Leydig cells of the human testis. Arch Histol Cytol 1995; 58: 17-30. 92. Zini A, O'Bryan MK, Magid MS and Schlegel PN. Immunohistochemical localization of endothelial nitric oxide synthase in human testis, epididymis, and vas deferens suggests a possible role for nitric oxide in spermatogenesis, sperm maturation, and programmed cell death. Biol Reprod 1996; 55: 935-941. 93. Stdphan JP, Guillemois C, J6gou B and Bauch6 F. Nitric oxide production by Sertoli cells in response to cytokines and lipopolysaccharide. Biochem Biophys Res Commun 1995; 213: 218-224. 94. Beckman JS, Beckman TW, Chen J, Marshall PA and Freeman BA. Apparent hydroxyl radical production by peroxynitrite: implications for endothelial injury from nitric oxide and superoxide. Proc Natl Acad Sci USA 1990; 87:1620-1624. 95. Lipton SA, Choi YB, Pan ZH, Lei SZ, Chen HV, Sucher NJ, Loscalzo J, Singel DJ and Stamler JS. A redox-based mechanism for the neuroprotective and neurodestructive effects of nitric oxide and related nitroso-compounds. Nature 1993; 364: 626-632. 96. Pomerantz DK, Pitelka V. Nitric oxide is a mediator of the inhibitory effect of activated macrophages on production of androgen by the Leydig cell of the mouse. Endocrinology 1998; 139:922-931. 97. del Punta K, Charreau EH and Pignataro OP. Nitric oxide inhibits Leydig cell steroidogenesis. Endocrinology 1996; 137: 5337-5343. 98. Emoto M, Yagyu Y, Nishikawa F, Katsui N, Kita E and Kashiba S. Effects of mouse testicular extract on immunocompetent cells. Am J Reprod Immunol 1989; 21: 61-66. 99. Emoto M, Nishikawa F, Oku D, Hamuro A, Kita E and Kashiba S. Suppressive effect of a mouse testicular extract on lymphocyte activation. Int J Androl 1991; 14: 291-302. 100. P611~inen P, S6der O and Uksila J. 1988. Testicular immunosuppressive protein. J Reprod Immunol 1988; 14: 125-138. 101. P611finen P, S6der O, Uksila J, Nikula H, Kaipia A, Kangasniemi M, Punnonen J, Huhtaniemi I and Parvinen M. Testicular immunosuppressive peptide. Transplant. Proc 1989; 21: 1144-1146. 102. Selawry HP, Kotb M, Herrod HG and Lu ZN. Production of a factor or factors, suppressing IL-2 production and T cell proliferation by Sertoli cell-enriched preparations. A potential role for islet transplantation in an immunologically priviliged site. Transplantation 1991;
281
52: 846-850. 103. Khan SA, S6der O, Syed V, Gustafsson K, Lindh M and Ritz6n EM. The rat testis produces large amounts of an interleukin-l-like factor. Int J Androl 1987; 10: 495-503. 104. Gustafsson K, S6der O, P611~inen P and Ritz6n EM. Isolation and partial characterization of an interleukin-l-like factor from rat testis interstitial fluid. J Reprod Immunol 1988; 14: 139-150. 105. Hedger MP, Drummond AE, Robertson DM, Risbridger GP and de Kretser DM. Inhibin and activin regulate [3H]thymidine uptake by rat thymocytes and 3T3 cells in vitro. Mol Cell. Endocrinol 1989; 61: 133-138. 106. Nagata S, Golstein P. The Fas death factor. [Review]. Science 1995; 267: 1449-1456. 107. Kagi D, Vignaux F, Ledermann B, Burki K, Depraetere V, Nagata S, Hengartner H and Golstein P. Fas and perforin pathways as major mechanisms of T cell-mediated cytotoxicity. Science 1994; 265: 528-530. 108. Lowin B, Hahne M, Mattmann C and Tschopp J. Cytolytic T-cell cytotoxicity is mediated through perforin and Fas lytic pathways. Nature 1994; 370: 650-652. 109. Alderson MR, Tough TW, Davis-Smith T, Braddy S, Falk B, Schooley KA, Goodwin RG, Smith CA, Ramsdell F and Lynch DH. Fas ligand mediates activation-induced cell death in human T lymphocytes. J Exp Med 1995; 181: 71-77. 110. Brunner T, Mogil RJ, LaFace D, Yoo NJ, Mahboubi A, Echeverri F, Martin SJ, Force WR, Lynch DH, Ware CF and et al. Cell-autonomous Fas (CD95)/Fas-ligand interaction mediates activation-induced apoptosis in T-cell hybridomas. Nature 1995; 373: 441-444. 111. Dhein J, Walczak H, Baumler C, Debatin KM and Krammer PH. Autocrine T- cell suicide mediated by APO-1/(Fas/CD95). Nature 1995; 373: 438-441. 112. Griffith TS, T. Brunner T, S.M. Fletcher SM, D.R. Green DR and T.A. Ferguson TA. 1995. Fas ligand- induced apoptosis as a mechanism of immune privilege. Science 1995; 270:1189-1192. 113. Bellgrau D, Gold D, Selawry H, Moore J, Franzusoff A and Duke RC. A role for CD95 ligand in preventing graft rejection. Nature 1995; 377: 630-632.
This Page Intentionally Left Blank
New Foundation of Biology Edited by I. Berczi and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
283
Leptin and Cytolines: Actions and Interactions in Fever and Appetite Control
GIAMAL N. LUHESHI
The Douglas Hospital Research Center, Department of Psychiatry, McGill University, Quebec, Canada H4H-1R3
ABSTRACT Leptin, the 16 kDa ob gene product is an important regulator of energy balance via direct action on the brain. Since its discovery in 1994, there has been an explosion in this area of research aimed mainly at leptin's role in the regulation of appetite and by inference its use as a potential treatment for obesity. This hormone is synthesised mainly by adipocytes in relation to body mass and is released into the circulation from which it gains access to the brain via a saturable transport mechanism. In the brain leptin acts on its hypothalamic receptors to suppress appetite and increase energy expenditure thus collectively resulting in weight loss. On activating its receptors, leptin has been proposed to act via the induction of a number of secondary mediators including members of the pro-opiomelanocortin (POMC) family, corticotrophin releasing factor (CRF) and neuropeptide-Y (NPY). More recently interaction between leptin and the proinflammatory cytokine interleukin (IL)-I has been proposed, suggesting that this cytokine is involved in leptin production in the periphery. Our own recent findings in rodent brain have suggested the converse by demonstrating that leptin can induce IL-I[3 production in the hypothalamus. These findings have led us to propose that actions of leptin on food intake are mediated by the production and action of IL-I[5 in the brain. We have also demonstrated, that leptin, at the same dose that induces appetite suppression, is pyrogenic and that this effect is also mediated by IL-I[5. These results suggest that leptin may be an important mediator of neuroimmune interactions which activates CNS responses to disease, and reveal novel mechanisms of leptin action in the brain that depend on the synthesis and action of IL-115.
1.
OBESITY AND LEPTIN
Obesity affects at least 20% of the adult population in the Western world and is fast becoming the leading cause of illness world wide. A major factor that can lead to the development of obesity is dysregulation of energy balance, which is dependent on the level, and control of food intake (appetite) and energy expenditure (thermogenesis). Energy balance is controlled by the brain, through actions and interactions of a variety of peptides and neurotransmitters acting mainly in the hypothalamus to regulate appetite, satiety and thermogenesis [45]. Leptin appears
284
to be particularly important in the regulation of this process and since its discovery by Zhang et al. in 1994 [46], has received a great deal of attention relating to its role as a principle regulator of appetite and energy expenditure [16, 46]. Leptin, (the product of the ob gene), is produced mainly by adipocytes and acts as a hormonal link between peripheral fat mass and the appetite regulating centres in the brain. Leptin does this by conveying information to the brain on the amount of energy stored as fat, reflecting the nutritional state of the individual [7] thus helping to maintain weight stability by modulating food intake and energy expenditure. It is not surprising therefore that dysfunction in the leptin system could lead to the development of obesity as dramatically demonstrated in mice with recessive mutations in either genes encoding leptin (ob/ob) or its receptor (db/db), and rats possessing a dysfunctional leptin receptor (fa/fa). These animals exhibit an abnormal increase in body weight, several fold greater than normal controls and approximately five fold increase in body fat content [16]. Injection of leptin in ob/ob mice and in normal rodents results in a reduction in food intake and body weight [7]; inhibiting the actions of endogenous leptin with antibodies [6] or antagonists in the shape of inactive leptin mutant forms [41 ] results in an increase in body weight. The weight loss induced by leptin is specific to the depletion of adipose tissue [21], which appears to be mediated by apoptotic mechanisms [35]. This is qualitatively distinct from the weight loss resulting from restriction of food intake, which includes loss of both fat and lean body mass [21 ]. These results suggest that leptin would perhaps present a way of treating obesity in humans. However, the vast majority (90-95%) of obese humans exhibit normal plasma leptin levels [29] relative to their body weight [15]. It is therefore likely that the development of obesity could be associated with failure of leptin transport into the brain (reduced in some obese individuals, [8]), or to insensitivity to leptin at the level of the receptor or post-receptor pathways, which probably present the best targets for the development of therapeutic strategies.
2.
SITES AND MECHANISMS OF ACTION OF LEPTIN
The primary site of leptin action on appetite and energy metabolism is the hypothalamus, particularly in the arcuate nucleus [20, 40], though actions in the brain stem have been reported [11]. The hypothalamus is rich in leptin receptor expression and lesions of hypothalamic areas involved in appetite control [e.g. ventromedial hypothalamus (VMH) and the arcuate (ARC) nuclei] results in obesity (see [12]). Circulating leptin gains entry to the brain by crossing the blood brain barrier (BBB) via a saturable transport mechanism [2]. This form of transport is mediated via the short (non-signalling) form of the leptin receptor that is expressed in blood vessels associated with the choroid plexus, meninges, hypothalamus and cerebellum [4]. This transport mechanism acts by endocytosis of the leptin molecule and is a specific and temperature-dependent system [ 19]. In the brain the specific mechanisms of leptin action have not been elucidated fully. A recent study [ 17] demonstrated that leptin influences the state of the brain reward circuitry resulting in the reduction of appetite, probably via the induction of mediators implicated in the control of feeding. The most prominent of these mediators include NPY and CRF [43] but others have also been implicated and include insulin, POMC and its cleavage product c~-melanocyte-stimulating hormone (et-MSH), urocortin, orexins/hypocretins, melanin-concentrating hormone (MCH), noradrenergic and serotonergic pathways, cocaine and amphetamine-regulated transcript (CART), agouti and agouti-related peptide, glucagon-like peptide-1 (GLP), and galanin (see [16] for review). Leptin may of course stimulate food intake and energy expenditure directly, or may act via as yet undetermined mediators such as the product of the TUB gene [24].
285
Alternatively, other molecules known to be involved in regulating energy balance such as cytokines may mediate actions of leptin (see Figure 1, for overview of leptin expression and actions). NPY .................................................. ~:>
CNS
c.,.
c 0
-~
POMC~
~
TLIB -~'~i~' ~' ~' :~~' ~' ~' i~............ ~' ~' ~' '~:'~...............
Hypothalamic Output
,~-MSH
9Food Intake 9M e t a b o l i s m 9S y m p a t h e t i c NS 9Energy Expenditure 9R e p r o d u c t i o n
Cytokines
"~
Soluble
r ,--
Serum Protein
........
Receptor
I!11)]
• •
Free
...e.. .............OB ...... !~................ ............
'H'='YD"r;"n~"~'" r=
Negative
:'............ : ................. OB 'i
................ :,,O
oB
:...................................
F i g u r e 1. S c h e m a t i c r e p r e s e n t a t i o n of e x p r e s s i o n a n d a c t i o n of leptin.
3.
CYTOKINES AND NEUROIMMUNE INTERACTIONS
Cytokines are a heterogeneous family of endogenous proteins that are produced in response to a variety of physiological and pathophysiological stimuli. These molecules which include the interleukins, interferons, colony stimulating factors growth factors and tumor necrosis factors induce a variety of effects and are generally considered to influence cell growth, differentiation, survival and in some cases cell death [38]. Cytokines are produced by virtually all cell types in the body but are associated particularly with the motile cells of the peripheral immune system. More diverse actions have now been reported and cytokines are now known to be released by and to affect a variety of organs including the brain. Association with the brain has become an area of intense study and interest, and the neurobiology of cytokines is now a major research area, in particular their role as neuroimmune modulators of host defence responses in disease. Via direct action on the brain cytokines can elicit fever, sickness behavior, reduced food intake, increased energy expenditure and cachexia (see [22, 26, 37] for comprehensive reviews of this area). Perhaps the best known and most intensely studied member of the cytokine family is interleukin (IL)-I. This cytokine exists in two forms (c~ & [5) both of which act on the same receptor (IL-1RI) and induce identical biological responses. This molecule was originally described as a heat labile protein that induced fever when administered to experimental animals or humans [ 1], and consequently was named "endogenous pyrogen". Other members of the IL-1 family include the naturally occuring receptor antagonist (IL-lra), which acts by limiting the action of IL-1 via competitively preventing its actions on the IL-1RI. In the brain IL-1 is expressed primarily by microglia but is also found in astrocytes [10] and it is activated in these cells after local (brain) or systemic injury or infection. Although it acts on a number of different brain sites, the primary area of I L - l ' s action in mediating the
286
host defense response to disease is the hypothalamus. Application of exogenous IL-1 directly to the hypothalamus or into the cerebral ventricles proximal to it results in the induction of fever and other sickness like behaviours including appetite suppression [36]. Studies on the role of endogenous IL-1 either by investigating its induction in the brain following a systemic stimulus or by inhibiting its action using IL-lra have confirmed that this cytokine is a primary mediator of sickness responses via direct action on the hypothalamus [37]. The nature of the afferent signal responsible for the induction of brain IL-1 is however still controversial. A role for IL-1 in this respect is somewhat controversial owing to a lack of significant biologically active amounts in the circulation of sick e.g. febrile humans or experimental animals. Unlike IL-1, however another cytokine, IL-6, increases dramatically in the circulation of febrile subjects and this increase correlates favorably with the development of fever. Our recent studies in rats have demonstrated the importance of this cytokine in fever by using a neutralizing antiserum raised against rat IL-6 which abolished the febrile response to systemic LPS (Figure 2 [9]).
A
r
39.0-
-o- PIS / Saline -e- PIS / LPS -=- Saline / LPS
38.5-
-4-IL-6 AS/LPS
o.. L.
=3 38.0L
(1) Q.
E 37.5.
tn
37.0
36.51
l
-60
0
Time (hours)
Figure 2. Systemic LPS (100 ~tg/kg) induces fever, which is abolished in the presence of IL-6 antiserum, also injected systemically. Changes in body temperature measured by remote biotelemetry.
Other studies favour a neuronal link between the periphery and the brain namely the vagus nerve. A number of recent reports have demonstrated a role for the vagus nerve in transmitting cytokine signals to the brain to elicit an array of host defence responses. These responses, which included fever, were shown to be significantly inhibited in vagotomized experimental animals injected peripherally with inflammatory stimuli including IL-1 (see [30, 42] for reviews of this area). This evidence would suggest that a number of mechanisms are involved in relaying cytokine induced signals from the periphery to the brain during infection injury or inflammation. The relative importance of a humoral versus a neuronal signal remains to be determined. Recent evidence has linked circulating leptin with cytokines, a relationship that could be an important one in neuroimmune inetraction.
287
5.
INTERACTIONS BETWEEN LEPTIN AND CYTOKINES
It has been suggested that leptin is a cytokine [44]. The primary leptin receptor resembles gpl30, the common signal transducing subunit of the IL-6 receptor [3, 33] and leptin shares some actions with pro-inflammatory cytokines, such as suppression of appetite, and stimulation of thermogenesis [16]. Leptin production by adipose tissue (and its release into circulation) in rats is stimulated by administration of bacterial LPS, which is a potent stimulus for cytokine production in vivo [31, 32]. Leptin release is also induced by proinflammatory cytokines such as IL-1[3 and tumor necrosis factor (TNF)-~t [39], which themselves inhibit food intake and stimulate metabolic rate [37]. IL-1 has also been shown to mediate leptin induction during inflammation and LPS fails to increase leptin levels in mice lacking the gene expressing IL-I[3 [14]. Conversely, exogenous leptin has been demonstrated to up-regulate LPS-induced phagocytosis and proinflammatory cytokine expression in ex vivo mouse macrophages [27]. Additionally, leptin-deficient (ob/ob) mice and the obese (fa/fa) Zucker rats exhibit attenuated levels of serum TNF-c~ and IL-6 in response to LPS administration [27]. Our own recent studies [28] revealed that peripheral or central injection of leptin not only suppresses appetite but also induces marked fever in rats at doses similar to, or lower than, those which inhibit food intake [21, 34]. We also found that obese Zucker rats which posses a dysfunctional leptin receptor fail to develop fever in response to injection of recombinant leptin. These previously unreported effects of leptin appeared to be IL-1 mediated in the brain since co-administration of leptin systemically or into the brain with intracerebroventricular IL-lra abrogated the response on fever and food intake (Figure 3). In addition, experiments on genetically modified IL-1RI receptor knock-out mice showed that these animals do not respond to leptin's effect on food intake. These observations, along with the fact that leptin induces hypothalamic IL-I[3 production, further verified the involvement of IL-1 in leptin's action on appetite control and fever [28]. We subsequently demonstrated that the effects of leptin on body temperature, like those of IL-1, are inhibited by administration of a cyclo-oxygenase inhibitor [28]. Since cyclo-oxygenase inhibitors do not modify effects of leptin on food intake [28], different pathways appear to mediate the effects of leptin on body temperature and food intake. Pyrogenic cytokines also depend on PG's for induction of fever [13], but not suppression of food intake. Both of these responses (fever and appetite suppression) also form part of generalized sickness behavior responses to disease in which brain IL-I[3 is involved [23]. Interestingly, there are no reports associating leptin with defined sickness behaviors other than fever, which would suggest a divergence in the pathways controlling food intake and fever and other behaviors such as depressed social interaction, which is also induced by brain IL-I[3 [23]. Our recent observations would support this hypothesis, since we demonstrated that the effect of leptin on appetite involves a direct action of CRF [18] a neuropeptide shown not to be involved in sickness behavior [5]. In addition, we have now demonstrated that IL-6, which mediates the pyrogenic action of IL-1, does not itself induce sickness behavior [25]. This would suggest that IL-6 could play an important part in the mediation of leptin's action on appetite and fever either directly or through IL-1 (see Figure 4 for schematic representation). In fact IL-6 has previously been shown to be involved in body weight control [15, 35] however this evidence is largely circumstantial and no direct evidence exists to link brain IL-6 with appetite regulation.
6.
SUMMARY
The original suggestion that leptin could be a possible treatment for obesity through its appetite suppressing actions has led to a great deal of interest from scientist across many fields.
288
(A) 30i
A 25-
tJii!i,i i i fi i i i i i!
0
r 20-
##
::::::::::::::::::::::::::: ::::::::::::::::::::::::::: ::::::::::::::::::::::::::: ........................... ............................ ~::::::::::=::::=:::::::::: ::::::::::::::::::::::::::: ............ . ............... ::::::::::::::::::::::::::: ............................ :::::::::::::::::::::::::::
"0 0 0
u. 15-
I::::1::11:::111:1:::1:1::1 :::::::::::::::::::::::::::: ::::::::::::::::::::::::::: ::::::::::::::::::::::::::: ............................
~-:--~
:::::::::::::::::::::::::::: :::::::::::::::::::::::::::: :::::::::::::::::::::::::::: :::::::::::::::::::::::::::: :::::::::::::::::::::::::::: ............................
lO
. 9- _ - - _ - = _ = - = _ = - _ = = = -
i!!il!iiiiiiiiiiiiii!iiii!!
o 1"
IL-1 ra
Vehicle
Leptln I IL-1 ra
Leptln
(B) 39
***
I
38.5. A
0
,- 38.0" (g t,,
0
=" 37.5-
E 0 I'-
.
37.0 -7
36.5 oT -1
- 0 - Vehicle ~
: t)
--qF- I L - l r a
1
Leptin
- 0 - LeptlnllL-lra
:2 Time
:3
,t
(hours)
Figure 3. IL-lra (200 ~tg, icv) inhibits suppression of appetite (A) and stimulation of body temperature (B) by leptin (4 ~tg, icv). ***P,0.001 leptin vs. controls and # # P < 0.01 leptin in the presence of IL-lra vs. leptin only treatment.
Many aspects of the mechanisms of action of leptin were investigated resulting in the discovery that apart from regulating energy balance leptin is also involved in other biological processes making it pleiotropic in nature. Leptin has now been implicated in for example reproduction and development and in the pathogenesis and progress of a number of diseases, most notable diabetes. It is interesting to note that some of these processes are also influenced by the immune system, which has now been described to be a major target for leptin. This association and particularly the interaction with cytokines will no doubt receive more attention in the future.
289
LEPTIN
Other sickness behaviours
1 IL-1
?
IL-6
CRF
'~ FOOD INTAKE
PGEz
~ BODY TEMPERATURE
Figure4. Hypotheticalschematicrepresentationof possible pathwaystriggeredby leptinto inducefever and reducedappetite.
REFERENCES 1. 2. 3.
4.
5. 6.
7.
8.
Atkins E. Fever: its history, cause, and function. Yale J.Biol.Med 55: 283-289, 1982. Banks WA, Kastin AJ, Huang W, Jaspan JB, Maness LM. Leptin enters the brain by a saturable system independent of insulin. Peptides 1996; 17:305-311. Baumann H, Morella KK, White DW, Dembski M, Bailon PS, Kim H, Lai CF, Tartaglia LA. The full-length leptin receptor has signaling capabilities of interleukin 6-type cytokine receptors. Proc Natl Acad Sci USA 1996; 93: 8374-8378. Bjorbaek C, Elmquist JK, Michl P, Ahima RS, van Bueren A, McCall AL, Flier JS. Expression of leptin receptor isoforms in rat brain microvessels. Endocrinology 1998; 139:3485-3491. Bluthe RM, Crestani F, Kelley KW, Dantzer R. Mechanisms of the behavioral effects of interleukin 1. Role of prostaglandins and CRF. Ann NY Acad Sci 1992; 650: 268-275. Brunner L, Nick HP, Cumin F, Chiesi M, Baum HP, Whitebread S, Stricker-Krongrad A, Levens N. Leptin is a physiologically important regulator of food intake. Int J Obes Relat Metab Disord 1997; 21: 1152-1160. Campfield LA, Smith FJ, Guisez Y, Devos R, Burn P. Recombinant mouse OB protein: evidence for a peripheral signal linking adiposity and central neural networks. Science 1995; 269: 546-549. Caro JF, Kolaczynski JW, Nyce MR, Ohannesian JP, Opentanova I, Goldman WH, Lynn RB, Zhang PL, Sinha MK, Considine RV. Decreased cerebrospinal-fluid/serum leptin ratio in obesity: a possible mechanism for leptin resistance Lancet 1996; 348:159-161.
290
10.
11. 12. 13. 14. 15.
16. 17.
18. 19.
20.
21.
22. 23. 24.
25.
26. 27.
Cartmell T, Poole S, Turnbull AV, Rothwell NJ, Luheshi GN. Circulating interleukin (IL)-6 mediates the febrile response to localized inflammation in rats. J Physiol 2000; 526: 653-661. Davies CA, Loddick SA, Toulmond S, Stroemer RP, Hunt J, Rothwell NJ. The progression and topographic distribution of interleukin-lbeta expression after permanent middle cerebral artery occlusion in the rat. J Cereb Blood Flow Metab 1999; 19: 87-98. Elmquist JK, Ahima RS, Maratos-Flier E, Flier JS, Saper CB. Leptin activates neurons in ventrobasal hypothalamus and brainstem. Endocrinology 1997; 138: 839-842. Elmquist JK, Elias CF, Saper CB. From lesions to leptin: hypothalamic control of food intake and body weight. Neuron 1999; 22: 221-232. Elmquist JK, Scammell TE, Saper CB. Mechanisms of CNS response to systemic immune challenge: the febrile response. Trends Neurosci 1997; 20: 565-570. Faggioni R, Fantuzzi G, Fuller J, Dinarello CA, Feingold KR, Grunfeld C. IL-1 beta mediates leptin induction during inflammation. Am J Physiol 1998; 274: R204-R208. Frederich RC, Hamann A, Anderson S, Lollmann B, Lowell BB, Flier JS. Leptin levels reflect body lipid content in mice: evidence for diet-induced resistance to leptin action. Nat Med 1995; 1: 1311-1314. Friedman JM, Halaas JL. Leptin and the regulation of body weight in mammals. Nature 1998; 395: 763-770. Fulton S., B. Woodside and P. Shizgal. Modulation of brain reward circuitry by leptin [published erratum appears in Science 2000 Mar 17; 287 (5460): 1931]. Science 2000; 287:125-128. Gardner J. D., N. J. Rothwell and G. N. Luheshi. Leptin affects food intake via CRFreceptor-mediated pathways. Nat Neurosci 1998; 1: 103. Golden PL, Maccagnan TJ, Pardridge WM. Human blood-brain barrier leptin receptor. Binding and endocytosis in isolated human brain microvessels. J Clin Invest 1997; 99: 14-18. Hakansson ML, Hulting AL, Meister B. Expression of leptin receptor mRNA in the hypothalamic arcuate nucleus: relationship with NPY neurones. Neuroreport 1996, 7: 3087-3092. Halaas JL, Gajiwala KS, Maffei M, Cohen SL, Chait BT, Rabinowitz D, Lallone RL, Burley SK, Friedman JM. Weight-reducing effects of the plasma protein encoded by the obese gene. Science 1995; 269: 543-546. Hopkins SJ, Rothwell NJ. Cytokines and the nervous system. I: Expression and recognition. Trends Neurosci 1995; 18: 83-88. Kent S, Bluthe RM, Kelley KW, Dantzer R. Sickness behavior as a new target for drug development. Trends Pharmacol Sci 1992; 13: 24-28. Kleyn PW, Fan W, Kovats SG, Lee JJ, Pulido JC, Wu Y, Berkemeier LR, Misumi DJ, Holmgren L, Charlat O, Woolf EA, Tayber O, Brody T, Shu P, Hawkins F, Kennedy B, Baldini L, Ebeling C, Alperin GD, Deeds J, Lakey ND, Culpepper J, Chen H, GlucksmannKuis MA, Moore KJ. Identification and characterization of the mouse obesity gene tubby: a member of a novel gene family. Cell 1996; 85: 281-290. Lenczowski MJ, Bluthe RM, Roth J, Rees GS, Rushforth DA, Van Dam AM, Tilders FJ, Dantzer R, Rothwell NJ, Luheshi GN. Central administration of rat IL-6 induces HPA activation and fever but not sickness behavior in rats. Am J Physiol 1999; 276: R652-R658. Liles WC, Van Voorhis WC. Review: nomenclature and biologic significance of cytokines involved in inflammation and the host immune response. J Infect Dis 1995; 172: 1573-1580. Loffreda S, Yang SQ, Lin HZ, Karp CL, Brengman ML, Wang DJ, Klein AS, Bulkley
291
GB, Bao C, Noble PW, Lane MD, Diehl AM. Leptin regulates proinflammatory immune responses. FASEB J 1998; 12: 57-65. 28. Luheshi GN, Gardner JD, Rushforth DA, Loudon AS, Rothwell NJ. Leptin actions on food intake and body temperature are mediated by IL-1. 1999; Proc Natl Acad Sci USA 96: 7047-7052. 29. " Maffei M, Halaas J, Ravussin E, Pratley RE, Lee GH, Zhang Y, Fei H, Kim S, Lallone R, Ranganathan S. Leptin levels in human and rodent: measurement of plasma leptin and ob RNA in obese and weight-reduced subjects. Nat Med 1995; 1: 1155-1161. 30. Maier SF, Goehler LE, Fleshner M, Watkins LR. The role of the vagus nerve in cytokineto-brain communication. Ann NY Acad Sci 1998; 840: 289-300. 31. Miller AJ, Hopkins SJ, Luheshi GN. Sites of action of IL-1 in the development of fever and cytokine responses to tissue inflammation in the rat. Br J Pharmacol 1997; 120: 1274-1279. 32. Miller AJ, Luheshi GN, Rothwell NJ, Hopkins SJ. Local cytokine induction by LPS in the rat air pouch and its relationship to the febrile response. Am J Physiol 1997; 272:R857-R861. 33. Nakashima K, Narazaki M, Taga T. Overlapping and distinct signals through leptin receptor (OB-R) and a closely related cytokine signal transducer, gpl30. FEBS Lett 1997; 401: 49-52. 34. Pelleymounter MA, Cullen MJ, Baker MB, Hecht R, Winters D, Boone T, Collins F. Effects of the obese gene product on body weight regulation in ob/ob mice. Science 1995; 269: 540-543. 35. Qian H, Azain MJ, Compton MM, Hartzell DL, Hausman GJ, Baile CA. Brain administration of leptin causes deletion of adipocytes by apoptosis. 1998; Endocrinology 139: 791-794. 36. Rothwell NJ. Functions and mechanisms of interleukin 1 in the brain. Trends Pharmacol Sci 1991; 12: 430-436. 37. Rothwell NJ, Hopkins SJ. Cytokines and the nervous system II: Actions and mechanisms of action. Trends Neurosci 1995; 18:130-136. 38. Sachs L, Lotem J. The network of hematopoietic cytokines. Proc Soc Exp B iol Med 1994; 206: 170-175. 39. Sarraf P, Frederich RC, Turner EM, Ma G, Jaskowiak NT, Rivet DJ, Flier JS, Lowell BB, Fraker DL, Alexander HR. Multiple cytokines and acute inflammation raise mouse leptin levels: potential role in inflammatory anorexia. J Exp Med 1997; 185: 171-175. 40. Schwartz MW, Seeley RJ, Campfield LA, Burn P, Baskin DG. Identification of targets of leptin action in rat hypothalamus. J Clin Invest 1996; 98:1101-1106. 41. Verploegen SA, Plaetinck G, Devos R, van der Heyden J, Guisez Y. A human leptin mutant induces weight gain in normal mice. FEBS Lett 1997; 405: 237-240. 42. Watkins LR, Maier SF, Goehler LE. Cytokine-to-brain communication: a review & analysis of alternative mechanisms. Life Sci 1995; 57: 1011-1026. 43. Wettstein JG, Earley B, Junien JL. Central nervous system pharmacology of neuropeptide Y. Pharmacol Ther 1995; 65: 397-414. 44. White DW, Tartaglia LA. Leptin and OB-R: body weight regulation by a cytokine receptor. Cytokine Growth Factor Rev 1996; 7: 303-309. 45. Wilding J, Widdowson P, Williams G. Neurobiology. Br Med Bull 1997; 53: 286-306. 46. Zhang Y, Proenca R, Maffei M, Barone M, Leopold L, Friedman JM. Positional cloning of the mouse obese gene and its human homologue. Nature 1994; 372: 425-432.
This Page Intentionally Left Blank
IV.
NEUROIMMUNE HOST DEFENCE
This Page Intentionally Left Blank
New Foundation of Biology Edited by I. Berczi and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
295
Introduction
REGINALD M. GORCZYNSKI
Department of Surgery & Immunology, University of Toronto, The Toronto Hospital, CCRW 2-855, 200 Elizabeth Street, Toronto, ON, Canada M5G-2C5
In the preceding chapters of this volume we have been introduced to the role of cytokines, and neurohormones, as mediators of CNS: immune system interactions. We have seen evidence that these mediators can influence behaviour and vice-versa. Particular attention has already focused on inflammatory cytokines (TNFc~, IL-1 etc), given the evidence already extant which supports their role in phenomena as widely separate as sleep induction (Moldofsky et al., Chapter III-6), appetite control (Luheshi, Chapter III-8), and even autoimmune disease (Sternberg and Moghaddam, Chapter II-6). In the section which follows four authors present reviews which develop this theme of CNS: immune system interactions further, and show how such interactions play an important part in the regulation of breaches in host defence, the sine qua non of immunity. In an interesting and thought-provoking discussion, Pittman et al., review recent work from the author's laboratory which implicates the neuropeptides alpha melanocyte stimulating hormone (etMSH) and arginine vasopressin (AVP) in particular, in reduction of pyresis. Sex-related differences in fever regulation are suggested by the authors to be related in turn to a decreased utilization of AVP in females. An intriguing possibility is that increased AVP production near term may also underlie the suppression of fever in response to peripheral pyrogens (such as IL-1 and LPS) in pregnant animals. Nevertheless, as the authors indicate if, as current dogma suggests, fever has a survival value, what is the rationale for production of these endogenous anti-pyretics? The paper that follows from the Befus laboratory (Davison et al.,) should not be read in isolation, but in the context of the earlier discussion by Forsythe et al., (Chapter III-3). This current review provides more detail on the intriguing role of salivary gland peptides in anaphylaxis and LPS-induced inflammation, and the functional activity of a tripeptide analogue (FEG) of the critical submandibular gland (SMG) peptide. It seems these molecules inhibit expression of important regulatory integrins (such as CD14), thus blocking recruitment and activation of neutrophils and eosinophils, responsible for tissue damage during inflammation. Since secretion from the SMG is itself under sympathetic control, these studies in turn begin to describe, at the molecular level, the pathway whereby innervation by the sympathetic chain might directly regulate inflammation in vivo. As mentioned earlier in this book, few discussions so enliven an audience as those on sex or appetite. One might add, on reflection, cancer (and probably heart disease). Two of the former are covered again in the last two chapters of this section. Baines provides a lucid and thoughtful review of a large literature on olfactory control of allorecognition, particularly as it pertains to
296
mate selection in animals. Perhaps the most well-known reproductive responses to pheromonal stimuli are the so-called Whitten effect and the Bruce effect. The former describes the onset of oestrus in unmated female mice by the presence of the odour of an alien male, while the latter documents the pre-implantation block of pregnancy in mated females by the odour of an alien male. Both effects are apparently a reflection of release of neuroendocrine mediators, and both are believed to result from a "down-stream" regulation of the (female) host cell-mediated uterine response to the potential fetal graft. From the male perspective, few could find fault with this as a method of improving survival of one's own gene pool! Chow and her colleagues conclude this section with an insightful consideration of the role of neuroendocrinological regulation of "natural immunity" in host resistance to tumors, and perhaps even to inflammation and autoimmune disease. Activation of the acute-phase response invariably increases the titre of polyclonal natural antibody (Nab), and activity of activated macrophages, both of which are likely important in immune surveillance against tumors, especially NK-resistance tumors. Interestingly, major epitopes recognized by Nabs seem to be those believed to be primarily associated with T cell activation (CD25 and asialo-forms of CD45RA). This raises intriguing possibilities concerning the mechanism(s) by which intravenous Ig might prove efficacious in the treatment of a myriad of autoimmune disorders, and Chow provides evidence that indeed natural human serum used for iv infusion does recognize CD45RA. The authors conclude that the epitope recognized on CD45RA "may be a highly conserved homologous epitope or homotope of the neuroimmune system implicated in health and disease".
New Foundationof Biology
297
Edited by I. Berczi and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
Fever and Antipyresis
QUENTIN J PITTMAN, ABDESLAM MOUIHATE and MARIE-STEPHANIE CLERGET
Neuroscience Research Group and Department of Physiology and Biophysics, Faculty of Medicine, University of Calgary, 3330 Hospital Dr NW, Calgary, Alberta, Canada T2N-4N1
ABSTRACT One of the most common manifestations of disease is a regulated elevation in body temperature known as fever. A variety of experiments now indicate that fever is an integral part of the host defense response, which acts in synergy with other participants to combat disease. In the light of its importance, the sequence of events leading to development of the fever has been intensively studied. It is thought that lipopolysaccharide (LPS) released by bacteria, or other antigenic substances are phagocytosed by cells of the immune system, which in turn synthesize and secrete a number of cytokine molecules, including Interleukin-lbeta, Interleukin-6, and tumour necrosis factor-oz. These cytokines then activate the brain in various ways, including activation of sensory afferent nerves such as the vagus, direct access to the brain via circumventricular organs where the blood brain barrier is leaky and direct activation of endothelial cells in the microvasculature of the brain. Subsequent elaboration of prostaglandins of the E series from endothelial cells and glia appears to be an obligatory step for most experimental fevers; in keeping with this idea is the ability of prostaglandin synthesis inhibitors such as aspirin to reduce fever. Prostaglandins act largely in the anterior, ventral hypothalamus to cause activation of heat production and conservation mechanisms through sympathetic, hormonal and behavioral outputs. There appears also to be within the brain secondary synthesis of some cytokines, but the involvement of these molecules in the neural response is not well understood. The reduction of febrile body temperature appears also to be an active process. At least two transmitter molecules have been implicated as neurotransmitters within the brain to lower fever. These molecules include arginine vasopressin (AVP) and alpha-melanocyte stimulating hormone (alpha MSH); interference with the synthesis, release or action of these molecules results in elevated fevers. In addition, peripherally released corticosteroids suppress fever. Such molecules have been called endogenous antipyretics or cryogens. There are times in an animal's life when the ability to develop a fever is compromised; such times include the early neonatal period, the peri-parturient period in the mother, and some hypertensive states. The fever appears to be suppressed due to a central (neural) mechanism, a part of which may involve endogenous antipyretics. In the light of the survival value of fever, one questions why there are periods in an animal's life in which fever suppressed.
298
1.
INTRODUCTION
While fever has been studied for well over a hundred years, the recognition that there are endogenous systems that limit or reduce fever (antipyresis) came only in the last 30 years. While the focus of this brief review is to provide an understanding of antipyresis, it is important to know the basics of the cascade of events underlying the fever response. Several recent reviews cover this field in depth [1-5], but the following provides a relatively brief overview.
2.
FEVER
When microorganisms (gram negative infections are best understood) invade our bodies, they expose our immune system to large lipopolysaccharide molecules (LPS) often called exogenous pyrogens or endotoxins. LPS binds to a soluble, circulating LPS binding protein and this complex binds to the CD 14 surface receptor found on certain monocytes and macrophages. These in turn synthesize and release a variety of endogenous proteins; those thought to be most important in fever are interleukin-l[3 (IL-I[3), IL-6 and tumour necrosis factor (TNF). During fever, a number of humoral changes take place, collectively called the acute phase response. However, the regulated rise in body temperature characteristic of fever involves the CNS and the mechanism by which peripherally generated cytokines or other peptides activate the brain has been intensively investigated. Evidence exists for several possible avenues, depending upon the route and dose of administration of cytokines (reviewed in [3, 6]). These include direct transport across the blood brain barrier, entry at circumventricular organs, local stimulation of perivascular and meningeal cells, and activation of peripheral nerves. Whatever the route of administration, it appears that most cytokines activate an inducible cyclooxygenase (COX 2), most likely in glia, to cause intracerebral synthesis of prostaglandins, largely of the E series (PGE; reviewed in [3, 7]). Peripheral immune stimuli activate many autonomic and endocrine nuclei, as revealed by Fos expression [8-10], but it is difficult to distinguish which pathways are involved in the fever response and which are involved in the many other autonomic responses (cardiovascular, gastrointestinal, etc) associated with immune activation, especially at the high doses often employed in these studies. Prostaglandins are known to act in several sites to activate central sympathetic pathways (reviewed in [7]), but the most sensitive of these for the purposes of fever generation appears to be a small nucleus in the ventral medial preoptic area (VMPOA) [3, 11]. Among other projections of this nucleus, that to the paraventricular nucleus (PVN) and nearby perifornical area appear to be particularly important sites for activation of heat conservation and thermogenesis to cause fever. In addition to the prostaglandin link, intense (i.e. high dose) peripheral immune activation causes synthesis within the brain of a variety of cytokines and certain transcription factors [1]. While application of IL-1 or TNF directly to the brain by icv injection will cause a fever, and receptors of such cytokines are present in the brain, the involvement of this brain cytokine system in the responses to peripheral immune stimuli is not well understood. Nonetheless, for some models of fever, particularly those with long latencies, injection of IL-1 receptor antagonist (IL-lra) into the brain will inhibit fever due to peripheral inflammation [ 12, 13].
299
3.
ANTIPYRESIS
Fevers subside, either naturally by inactivation of causative organisms, or due to active pharmacological intervention (i.e. aspirin inhibits PGE synthesis). There is now good evidence that defervescence and fever suppression is a controlled process involving release and action of the neuropeptides, arginine vasopressin (AVP) and alpha melanocyte stimulating hormone (MSH). While the evidence for MSH is not as extensive as that for AVP, its icv injection reduces fever and an antagonist elevates fever [14, 15]. Salient facts supporting a role for AVP as an antipyretic arise from several laboratories [16, 17], including our own and are summarized as follows: 1. AVP, introduced into the brain (ventral septal area (VSA) or amygdala) reduces febrile, but not normal body temperature via an action at V 1 type receptors. 2. Activation of endogenous AVP pathways causes antipyresis. 3. Interference with AVP release or action results in elevated, prolonged fevers. 4. During defervescence, increased quantities of AVP are released into the brain. While these observations provide support of AVP's role, for the remainder of this review we will discuss possible avenues for future research and unresolved questions, drawing upon additional observations that implicate AVP in the reduction of fever.
4.
WHAT IS THE STIMULUS FOR THE RELEASE OF AVP?
As work with AVP antagonists has implicated endogenous release of AVP in the control of fever height and duration, we know that it is released during fever. In keeping with this, studies using immunohistochemistry for the immediate early onset protein Fos indicate that cell bodies in the bed nucleus of the terminalis (BST) are activated during fever induced by LPS [18]. Unfortunately, we do not know if these Fos-immunoreactive neurons are AVP in nature, as immunohistochemistry of AVP in this nucleus usually requires pretreatment with colchicine to allow visualisation of the peptide and this would interfere with fever. We considered the possibility that it was the rise in body temperature that was responsible for its release. To test this we took advantage of the fact that one can manipulate body temperature in urethane anaesthetised rats. We found that AVP was released during PGE induced fever even when the body temperature was initially at relatively low levels, i.e. 35~ Administration of PGE at high body temperatures was not associated with release if temperature was not allowed to rise. Similarly, elevating temperature alone, without the pyretic stimulus, was not associated with AVP release as measured by push-pull perfusion. Thus we concluded that both the pyretic stimulus and the rise in body temperature were the triggers [19]. It should be emphasized that these studies with PGE as the pyretic stimulus may not be representative of what occurs after administration of a peripheral pyrogen. Indeed, there is some evidence that AVP may be involved in hypothermic responses to cytokines when body temperature is never elevated [20]. Within the BST itself, there appear to be neurons responsive to cytokines that activate AVP neurons [21]. We were able to carry out push-pull studies in the terminal areas of BST vasopressinergic cells and demonstrate that micro infusion of IL-I[3 caused release of AVP in the VSA [22]. Thus, one could envision a scenario where intracerebral synthesis of IL-I[3, possibly followed by PGE synthesis, activates vasopressinergic neurons in the BST, which then initiate antipyresis.
300
5.
HOW DOES AVP INITIATE ANTIPYRESIS?
We do not know the exact locus of AVP action within the brain, nor the mechanism by which it acts to reduce fever. There are two ill-defined loci that have been identified-the VSA [23] and the medial amygdala [24]. They are ill-defined because, in working with conscious animals, the precision of the micro injection techniques previously utilized did not facilitate identification of a specific locus; thus hence the description of this region as a "VS Area". Although there may be more than one site and mechanism of action, a likely possibility is that AVP may interfere with the ability of PGE to activate the cell group in the VMPOA identified by Saper et al as a "hot spot" of PGE sensitivity. In recent unpublished experiments we have observed that local application of AVP in nanoliter quantities interferes with the pyretic action of locally applied PGE, making this a likely possibility. The febrile response is but one of a host of reactions to LPS which include elaboration of hepatic derived proteins (host defence response) [25]; induction of a type of social withdrawal termed "sickness behaviour" [26]; changes in food intake [27]; and alterations in hormone secretion [28, 29]. Most of these responses can occur independently. Just as it attenuates fever, AVP also attenuates sickness behaviour via an action in the VSA [30], but it is not known if it has an action upon other components of the response to LPS. One such action could occur within the brain where AVP may act on neural pathways important in controlling the hypothalamic-pituitary axis. In support of such a possibility is the observation that brain vasopressin is involved in stress-induced suppression of immune function in the rat [31]. It is also important to note that the source of the AVP involved in brain antipyresis is a group of AVP immunoreactive cell bodies located in the bed nucleus of the stria terminalis (BST) and is not the AVP found in the magnocellular PVN and supraoptic nuclei (SON). The latter cell bodies project to the pituitary where they release AVP into the circulation to regulate renal and cardiovascular function. It is interesting that, in addition to the release of AVP within the brain that we have described [19], it is also released during fever from the pituitary into the circulation [29, 32]. It would be important to know if this circulating AVP interacts with the neuroimmune response resulting from LPS exposure. It is possible that AVP's well known actions on cardiovascular and renal systems are important to counteract the cardiovascular collapse that can occur after high doses of LPS. Several other actions have also been proposed relating to an action of AVP to "restore homeostasis" [33]. Current dogma is that plasma AVP does not affect fever [34]. However, as fever is associated with high circulating levels of AVP, receptors are most likely saturated and it would appear unlikely that exogenously applied AVP would have much of an effect. However, experiments using peripheral AVP antagonists may reveal such an action. One could predict a possible action on ACTH, as AVP is a releasing factor in conjunction with CRH. Another possibility is a direct action on the adrenal cortex to enhance corticosterone secretion. V I receptors have been localized to the adrenal [35] and AVP increases corticosterone secretion of the isolated perfused adrenal gland [36].
6.
GENDER DIFFERENCES
It has now been reported in many laboratories, including our own, that fever size and duration can change dramatically between males and females [37] and also in association with the reproductive cycle [38]. While the mechanism for these changes is not known, there is circumstantial evidence that varying antipyretic action may underlie these changes n particular,
301
we have evidence that females have higher fevers in response to central PGE than do males, and this appears to be associated with a reduced utilization of AVP as an antipyretic. We found that an AVP antagonist did not elevate PGE fever in females unlike the effect in males, and also did not display the elevation in VSA AVP release associated with fever. This is possibly because the AVP innervation arising from the BST is sexually dimorphic and much reduced in females [39]. As other substances exist in the brain with potential antipyretic action, it is possible that females may use a different strategy to lower body temperature. For example, an antagonist of MSH has recently become available and was shown to prolong fever in male rats [14]. It's function merits testing in females. Of course, it is also possible that females are simply more responsive to PGE because of differences in PGE receptor or numbers. There are at least 4 such receptors identified by binding studies and they are designated EP1-4 [40]. cDNAs encoding these receptors have been cloned and a number of isozymes are known to exist [41, 42]. There is evidence that gene expression for at least some of these receptors is enhanced by estrogen [43]. The EP3 subtype appears to be the predominant form expressed in rodent brain [44] and is found in high concentration in the medial POA [45]. Other data indicate a role for the EP1 receptor in fever [46]. Studies need to be carried out to ask if PGE receptors are more numerous or show different affinities in females. As pointed out above, PGE 1 induced fevers are higher in females than in males. Now it is necessary to look earlier in the fever cascade to ask if IL-I[3 and LPS fevers are similarly different in males and females. While this might seem obvious, there are reasons to suspect that it may not be so. Responses to peripherally injected pyrogens are known to be modulated by circulating steroids [47]; females have higher circulating glucocorticoids and enhanced steroid secretion in response to some kinds of stress [48].
7.
ENDOGENOUS ANTIPYRESIS
Central AVP may also be important in causing a condition we have called "endogenous antipyresis". This is a state in which the normal febrile response to a pyrogen is reduced. It can be seen in certain neonates; in some types of hypertension; in acute hypotension; and in parturient animals (reviewed in [49]). However, we still do not know what is responsible for the antipyresis seen at parturition. Suppression of fever due to peripheral pyrogens such as LPS, IL-I[3 and to centrally administered PGE in pregnant animals including rats has been reported by several labs, including our own [50-54]. The phenomenon is most evident when pregnancy is close to term. While the fact that PGE fever is suppressed at term [54] suggests a central mechanism, the fact that the suppression is actually most profound after injection of LPS [53] raises the possibility of more than one locus. Two possibilities which are not necessarily mutually exclusive may account for the antipyresis of parturition: 1. It may be due to a general reduction in central sympathetic drive, including that to sympathetic organs involved in thermogenesis and heat conservation. In favour of this possibility is the well known reduction in peripheral vascular responsiveness and baroreflex during pregnancy [55-57]. In addition, we have been able to show that cardiovascular responses to centrally injected PGE are also reduced at term [58]. 2. There may b e a specific endogenous antipyretic activity which manifests itself at term. In favour of this is our previous observation that there appears to be more AVP in the VSA in pregnant and post-parturient rats compared to non-pregnant females [59]. Several studies have tested this possibility. We were unable to demonstrate enhanced vasopressinergic
302
'tone' at the time of parturition [58], but another laboratory reported that an AVP antagonist was effective in elevating fever at term, but not at other times [60]. However, this same laboratory was unable to find evidence for AVP involvement in the suppression of interleukin fever, which is also thought to involve a PGE step [61]. We have recently looked at the expression of the AVP receptor in the brain around parturition and have also found no differences (Clerget, unpublished observations).
8.
PERSPECTIVES
It is now appreciated that fever is an important part of the host defense response, but that it must be regulated at an optimum level to fight infection without deleterious effects upon the organism. It therefore makes good physiological sense that there are endogenous mechanisms to limit its height and duration. There are many questions which have been raised in the preceding discussion which will require investigation to resolve the mechanisms by which AVP acts as an endogenous antipyretic. With respect to the suppression of fever at certain times, one must ask why this takes place when fever is thought to have survival value.
9.
ACKNOWLEDGEMENTS
This work supported by MRC/CIHR.
REFERENCE 1. 2. 3. 4. 5. 6. 7.
8. 9.
10.
11.
Hopkins S J, Rothwell NJ. Cytokines and the nervous system I: Expression and recognition. Trends Neurosci 1995; 18: 83-88. Rothwell NJ, Hopkins SJ. Cytokines and the nervous system II: Actions and mechanisms of action. Trends Neurosci 1995; 18: 130-136. Elmquist JK, Scammell TE, Saper CB. Mechanisms of CNS response to systemic immune challenge: the febrile response. Trends Neurosci 1997; 20: 565-570. Moltz H. Fever: Causes and Consequences. Neurosci Biobehav Rev 1993; 17: 237-269. Kluger MJ. Fever: Role of Pyrogens and Cryogens. Physiol Rev 1991; 71: 93-127. Blatteis CM, Sehic E. Fever: How may circulating pyrogens signal the brain? News Physiol Sci 1997; 12:1-9. Blatteis CM, Sehic E. Prostaglandin E2: A putative fever mediator. In: Mackowiak PA (Ed.), Fever: Basic Mechanisms and Management. Lippincott-Raven Publishers 1997; pp. 117-145. Rivest S, Laflamme N. Neuronal activity and neuropeptide gene transcription in the brains of immune-challenged rats. J Neuroendocrinol 1995; 7:501-525. Wan W, Wetmore L, Sorensen CM, Greenberg AH, Nance DM. Neural and Biochemical Mediators of Endotoxin and Stress-Induced c-fos Expression in the Rat Brain. Brain Research Bulletin 1994; 34(1): 7-14. Elmquist JK, Scammell TE, Jacobson CD, Saper CB. Distribution of Fos-like immunoreactivity in the rat brain following intravenous lipopolysaccharide administration. J Comp Neurol 1996; 371: 85-103. Scammell TE, Elmquist JK, Griffin JD, Saper CB. Ventromedial preoptic prostaglandin E2
303
activates fever-producing autonomic pathways. J Neurosci 1996; 16: 6246-6254. Luheshi G, Hopkins SJ, Lefeuvre RA, Dascombe MJ, Ghiara P, Rothwell NJ. Importance of Brain IL-1 Type II Receptors in Fever and Thermogenesis in the Rat. American Journal of Physiology 1993; 265: E585-E591. 13. Luheshi G, Miller AJ, Brouwer S, Dascombe MJ, Rothwell NJ, Hopkins SJ. Interleukin-1 receptor antagonist inhibits endotoxin fever and systemic interleukin-6 induction in the rat. Am J Physiol Endocrinol Metab 1996; 270:E91-E95. 14. Huang QH, Entwistle ML, Alvaro JD, Duman RS, Hruby VJ, Tatro JB. Antipyretic role of endogenous melanocortins mediated by central melanocortin receptors during endotoxin-induced fever. J Neurosci 1997; 17:3343-3351. 15. Rajora N, Boccoli G, Burns D, Sharma S, Catania AP, Lipton JM. alpha-MSH modulates local and circulating tumor necrosis factor-alpha in experimental brain inflammation. J Neurosci 1997; 17:2181-2186. 16. Zeisberger E. Role of Vasopressin in Fever Regulation and Suppression. TIPS 1985; 6(11): 428-430. 17. Cridland RA, Kasting NW. A Critical Role for Central Vasopressin in Regulation of Fever During Bacterial Infection. Am J Physiol 1992; 263: R1235-R1240. 18. Hare AS, Clarke G, Tolchard S. Bacterial lipopolysaccharide-induced changes in FOS protein expression in the rat brain: correlation with thermoregulatory changes and plasma corticosterone. J Neuroendocrinol 1995; 7:791-799. 19. Landgraf R, Malkinson T, Veale WL, Lederis K, Pittman QJ. Vasopressin and oxytocin in the rat brain in response to prostaglandin fever. Amer J Physiol 1990; 259: R1056-R1062. 20. Derijk RH, Berkenbosch F. Hypothermia to Endotoxin Involves the Cytokine Tumor Necrosis Factor and the Neuropeptide Vasopressin in Rats. Am J Physiol 1994; 266: R9-R14. 21. Wilkinson MF, Mathison WB, and Pittman QJ. Interleukin-I[5 has excitatory effects on neurons of the bed nucleus of the stria terminalis. Brain Res 1993; 625: 342-346. 22. Wilkinson MF, Horn TFW, Kasting NW, Pittman QJ. Central interleukin-l[3 stimulation of vasopressin release into the rat brain: Activation of an antipyretic pathway. J Physiol (Lond.) 1994; 481: 641-646. 23. Naylor AM, Ruwe WD, Veale WL. Thermoregulatory actions of centrally- administered vasopressin in the rat. Neuropharmacology 1986; 25: 787-794. 24. Federico P, Veale WL, Pittman QJ. Vasopressin-induced antipyresis in the medial amygdaloid nucleus of conscious rats. Amer J Physiol: Regulatory, Integrative and Comparative Physiology 1992; 262: R901-R908. 25. Long NC. Evolution of infectious disease: How evolutionary forces shape physiological responses to pathogens. News Physiol Sci 1996; 11: 83-90. 26. Hart BL. Biological Basis of the Behavior of Sick Animals. Neurosci Biobehav Rev 1988; 12: 123-137. 27. Kent S, Bret-Dibat JL, Kelley KW, Dantzer R. Mechanisms of sickness-induced decreases in food-motivated behavior. Neurosci Biobehav Rev 1996; 20: 171-175. 28. Rivier C, Rivest S. Mechanisms mediating the effects of cytokines on neuroendocrine functions in the rat. Ciba Found Symp 1993; 172: 204-220. 29. Landgraf R, Neumann I, Holsboer F, Pittman QJ. Interleukin-l[5 stimulates both central and peripheral release of vasopressin and oxytocin in the rat. Eur J Neurosci 1995; 7: 592-598. 30. Dantzer R, Bluthe RM, Kelley KW. Androgen-dependent vasopressinergic neurotransmission attenuates interleukin-l-induced sickness behavior. Brain Res 1991; 557: 115-120. 12.
304
31.
Shibasaki T, Hotta M, Sugihara H, Wakabayashi I. Brain vasopressin is involved in stress-induced suppression of immune function in the rat. Brain Res 1998; 808: 84-92. 32. Kasting NW, Carr DB, Martin JB, B lume H, Bergland R. Changes in Cerebrospinal Fluid and Plasma Vasopressin in the Febrile Sheep. Can J Physiol Pharmacol 1983; 61: 427-431. 33. Martin, SM, Malkinson TJ, Veale WL, Pittman QJ. Prostaglandin Fever in Rats is Altered by Kainic Acid Lesions of the Ventral Septal Area. Brain Res 1988; 455: 196-200. 34. Cooper KE, Kasting NW, Lederis K, Veale WL. Evidence Supporting a Role for Endogenous Vasopressin in Natural Suppression of Fever in the Sheep. J Physiol 1979; 295: 33-45. 35. Balla T, Enyedi P, Spat A, Antoni FA. Pressor-type vasopressin receptors in the adrenal cortex: properties of binding, effects on phosphoinositide metabolism and aldosterone secretion. Endocrinology 1985; 117: 421-423. 36. Hinson JP, Vinson GP, Porter ID, Whitehouse BJ. Oxytocin and arginine vasopressin stimulate steroid secretion by the isolated perfused rat adrenal gland. Neuropeptides 1987; 10: 1-7. 37. Chen X, Landgraf R, Pittman QJ. Differential ventral septal vasopressin release is associated with sexual dimorphism in PGE 2 fever. Am J Physiol Regul Integr Comp Physiol 1997; 272: R 1664-R 1669. 38. Mouihate A, Chen X, Pittman QJ. Interleukin-lbeta fever in rats: gender difference and estrous cycle influence. Am J Physiol 1998; 275: R1450-R1454. 39. DeVries GJ, Buijs RM, van Leeuwen FW, Caffe AR, Swaab DF. The Vasopressinergic Innervation of the Brain in Normal and Castrated Rats. Journal of Comparative Neurology 1985; 233: 236-254. 40. Negishi M, Sugimoto Y, Ichikawa A. Prostanoid receptors and their biological actions. Prog Lipid Res 1993; 32:417-434. 41. Kawamura T, Yamauchi T, Koyama M, Maruyama T, Akira T, Nakamura N. Expression of prostaglandin EP2 receptor mRNA in the rat spinal cord. Life Sci 1997; 61: 2111-2116. 42. Manba T, Sugimoto Y, Negishi M, Irie, A, Ushikubi F, Kakizuka A, Ito S, Ichikawa A, Narumiya S. Alternative splicing of prostaglandin E receptor subtype EP 3 determines Gprotein specificity. Nature 1993; 365: 166-170. 43. Rage F, Lee BJ, Ma YJ, Ojeda SR. Estradiol enhances prostaglandin E 2 receptor gene expression in luteinizing hormone-releasing hormone (LHRH) neurons and facilitates the LHRH response to PGE 2 by activating a glia-to-neuron signaling pathway. J Neurosci 1997; 17: 9145-9156. 44. Sugimoto Y, Shigemoto R, Namba T, Negishi M, Mizuno N, Narumiya S, Ichikawa A. Distribution of the messenger RNA for the prostaglandin E receptor subtype EP3 in the mouse nervous system. Neuroscience 1994; 62: 919-928. 45. Ericsson A, Arias C, Sawchenko PE. Evidence for an intramedullary prostaglandindependent mechanism in the activation of stress-related neuroendocrine circuitry by intravenous interleukin- 1. J Neurosci 1997; 17:7166-7179. 46. Oka T, Hori T. EPl-receptor mediation of prostaglandin E2-induced hyperthermia in rats. Am J Physiol 1994; 267: R289-94. 47. Coelho MM, Luheshi G, Hopkins SJ, Pel~i IR, Rothwell NJ. Multiple mechanisms mediate antipyretic action of glucocorticoids. Am J Physiol Regul Integr Comp Physiol 1995; 269: R527-R535. 48. Kant GJ, Lenox RH, Bunnell BN, Mougey EH, Pennington LL, Meyerhoff JL. Comparison
305
49. 50.
51. 52. 53. 54. 55. 56. 57. 58. 59.
60. 61.
of stress response in male and female rats: pituitary cyclic AMP and plasma prolactin, growth hormone and corticosterone. Psychoneuroendocrinology 1983; 8:421-428. Pittman QJ, Wilkinson MF. Central arginine vasopressin and endogenous antipyresis. Can. J Physiol Pharmacol 1992; 70: 786-790. Cooper KE, B lahser S, Malkinson TJ, Merker G, Roth J, Zeisberger E. Changes in body temperature and vasopressin content of brain neurons, in pregnant and non-pregnant guinea pigs, during fevers produced by PolyI: Poly C. Pflugers Arch 1995; 412: 292-296. Simrose RL, Fewell JE. Body temperature response to IL-1 beta in pregnant rats. Am J Physiol 1995; 269: R 1179-82. Stobie-Hayes KM, Fewell JE. Influence of pregnancy on the febrile response to intracerebroventricular administration of PGE 1 in rats. J Appl Physiol 1996; 81: 1312- 1315. Martin SM, Malkinson TJ, Veale WL, Pittman QJ. Fever in pregnant, parturient, and lactating rats. Am J Physiol Regul Integr Comp Physiol 1995; 268: R919-R923. Martin SM, Malkinson TJ, Veale WL, Pittman QJ. Prostaglandin fever in rats throughout the estrous cycle, late pregnancy and parturition. J Neuroendocrinol 1996; 8: 145-151. Brooks VL, Quesnell RR, Cumbee SR, Bishop VS. Pregnancy attenuates activity of the baroreceptor reflex. Clin Exp Pharmacol Physiol 1995; 22: 152-156. Deng YM, Kaufman S. Effect of pregnancy on activation of central pathways following atrial distension. Am J Physiol Regul Integr Comp Physiol 1995; 269: R552-R556. Heesch CM, Rogers RC. Effects of pregnancy and progesterone metabolites on regulation of sympathetic outflow. Clin Exp Pharmacol Physiol 1995; 22:136-142. Pittman QJ, Chen X, Mouihate A, Hirasawa M, Martin S. Arginine vasopressin, fever and temperature regulation. Prog Brain Res 1998; 119: 383-92: 383-392. Landgraf R, Neumann I, Pittman QJ. Septal and Hippocampal Release of Vasopressin and Oxytocin During Late Pregnancy and Parturition in the Rat. Neuroendocrinology 1991; 54: 378-383. Eliason HL, Fewell JE. AVP mediates the attenuated febrile response to administration of PGE1 in rats near term of pregnancy. Am J Physiol 1998; 275: R691- R696. Eliason HL, Fewell JE. Arginine vasopressin does not mediate the attenuated febrile response to intravenous IL-lbeta in pregnant rats. Am J Physiol 1999; 276: R450-R454.
This Page Intentionally Left Blank
New Foundationof Biology
307
Edited by I. Berczi and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
Salivary Gland Peptides: Their Role in Anaphylaxis and Lipopolysaccharide (LPS)-Induced Inflammation
J.S. DAVISON, D. BEFUS 1 and R. MATHISON
Department of Physiology and Biophysics, Faculty of Medicine, University of Calgary, 3330 Hospital Drive NW, Calgary, Alberta, Canada T2N-4N1 1Asthma Research Institute, University of Alberta, Edmonton, Alberta
ABSTRACT About a decade ago, we published the first of a series of studies, which eventually led us to develop the concept of a sympathetic-superior cervical ganglion-submandibular gland axis, which appeared to regulate the release of immunosuppressive substances from the submandibular glands of rats. These putative mediators could suppress anaphylaxis and LPS-induced shock. We eventually discovered that the agents released from the salivary glands capable of producing the suppression of these inflammatory responses were novel, small molecular weight peptides and our current work has focused on the actions of one of these (submandibular gland peptide-T: SGP-T) and its analogs on animal models of anaphylaxis and LPS-induced inflammation. SGP-T, as well as the C terminal tri-peptide FEG, are both potent inhibitors of intestinal and cardiovascular anaphylaxis in egg albumen-sensitized Hooded-Lister or Sprague Dawley rats. They also inhibit endotoxin-induced hypotension in Sprague Dawley rats. These results are a striking demonstration of the ability of these salivary gland peptides to inhibit early phase immune responses. We have shown that the D-isomer of FEG (feG) prevents the infiltration of leukocytes following injection of LPS into the peritoneum. Similarly, in another presentation, at this meeting, we show that feG can also block late phase responses in anaphylaxis by preventing infiltration of pulmonary tissue by leukocytes. In other models, we have been able to show that these peptides inhibit carrageenan-induced neutrophilia within the skin and inhibit leukocyte rolling and adhesion in mesenteric venules. Current work is focusing on the molecular mechanisms which lead to recruitment and activation of leukocytes into intestinal tissue following anaphylaxis and LPS activation. We will present data showing inhibition of expression of identified cell markers involved in chemotaxis and activation in both these models, which reveals some interesting differences between the two, suggesting that these peptides may act on different receptor subtypes. In summary, our collective work to date implies that there is an important sympathetic pathway, involving the superior cervical ganglion, that regulates the release of novel peptides from the submandibular glands which play an important part in early and late phase immune responses in anaphylaxis and LPS-induced inflammation. In future work we hope, not only
308
to reveal the molecular mode of action of these peptides, but also to return to the point where we began and use our increasing molecular knowledge to study the neuroregulation of their production and release.
1.
INTRODUCTION
The salivary glands subserve a number of physiological functions [1]. Besides their wellrecognized role in carbohydrate digestion through the secretion of salivary amylase, they play an important role in the growth and development of the digestive tract in the young, and in the maintenance and repair of mucosal integrity in the adult through the secretion of trophic factors such as epidermal growth factor, nerve growth factor, and transforming growth factor [3. It has been long recognized that saliva plays an important role in wound healing involving not only the growth factors, but also suppression of local infection through the antibacterial properties of lyzozymes. It has become increasingly evident however, in the past decade, that the salivary glands also play a role in the regulation of the immune system at both a local and systemic level, due to the release of immunosuppressive agents that modulate the activity of leukocytes, in particular granulocytes, and perhaps other cells of the immune system and that the sympathetic nervous system regulates the release of these salivary gland immunoregulatory agents. Much of the work upon which this concept of a sympathetic nervous system-salivary gland-immune system axis is based was carried out in our laboratories [2]. The present paper will review some of the key findings which led to this concept and will present some new preliminary data that provide an explanation at the molecular level of the way in which immunosuppressive agents released from the salivary glands might suppress inflammatory responses.
THE ROLE OF THE SUPERIOR CERVICAL GANGLION AND THE SALIVARY GLANDS IN THE REGULATION OF ANAPHYLACTIC AND ENDOTOXIN-INDUCED SHOCK About a decade ago we published the first in a series of studies which eventually led us to develop the concept of a sympathetic-superior cervical ganglion-submandibular gland axis, which appeared to regulate the release of immunosuppressive substances from the submandibular glands of rats. These putative mediators could suppress the late phase pulmonary inflammation following anaphylaxis as well as LPS-induced shock [3, 4]. We showed that cutting the superior cervical sympathetic nerve, thereby decentralizing the superior cervical ganglion, modified the late phase pulmonary inflammation induced in rats previously parasitized with Nippostronglylus brasiliensis (Nb). The effect of the denervation procedure was eliminated by extirpation of the submandibular glands, suggesting that the cervical sympathetic nervous system was, in some way, regulating the release of immunomodulatory substances from the submandibular glands, which then modified the ability of leukocytes to infiltrate the pulmonary epithelium [3]. Subsequently we showed that endotoxin (LPS)-induced shock was exacerbated by cutting the superior cervical sympathetic trunks or by bilateral extirpation of the submandibular glands [4]. These observations again supported the view that the sympathetic nervous system regulated the immune system by modulating the release of immunoreactive agents from the salivary glands. Specifically we postulated that these agents were immuno-suppressive substances, whose secretion could either be inhibited or excited by the sympathetic nervous system, depending on the physiological, pathophysiological or experimental conditions pertaining at the time.
309
.
ISOLATION AND IDENTIFICATION OF NOVEL SUBMANDIBULAR GLAND PEPTIDES
On the basis of these earlier studies, we postulated that rat submandibular glands contained factors that would reduce the severity of endotoxin or anaphylaxis-induced hypertension. Therefore we carried out a series of studies using classical peptide isolation procedures in which extracts of rat submandibular glands were subjected to molecular weight cut off filtration followed by preparative, reverse phase high performance liquid chromatography (HPLC) and finally analytical HPLC purification. At each step in the process isolated fractions were tested for their ability to reduce the severity of endotoxin-induced hypertension. As a result, a novel heptapeptide was isolated from these extracts, which was subsequently sequenced and synthesized and shown to attenuate the severity of endotoxin-induced hypotension [5]. This peptide had the sequence TDIFEGG and will subsequently be referred to as submandibular gland peptide T (SGP-T). In this and subsequent studies, we confirmed that SGP-T would reduce the severity of cardiovascular and intestinal anaphylaxis in Nb and ovalbumen sensitized rats, as well as the severity of hypotension and fever induced by lipopolysaccharide [6-10]. Subsequent structure-activity relationship studies revealed that the inhibition of intestinal anaphylaxis required only the tripeptide FEG localized at the carboxyl terminal of the parent heptapeptide [11]. Our current studies, therefore, have focused on the parent molecule SGP-T and its tripeptide analogue FEG and the D-isomeric form-the tripeptide leG. During the period we were sequencing the active peptide, a search of the gen bank database revealed that TDIFEGG is a COOH-terminal fragment of the submandibular gland rat 1 (SMR1) protein at positions 138-144 of this 146 amino acid protein. The structure of this prohormone was deduced from the cDNA sequence of the SMR1-VA1 gene in the submandibular glands of Wistar rats (12). The SMR1-VA1 protein in Sprague-Dawley or Fischer rats differs by only one amino acid from that found in Wistar rats (13). This does not affect the TDIFEGG sequence, which is present in all rat strains studied thus far. The SMR1-VA1 gene is one of the variable coding sequence (VCS) multigene family of genes of which 3 belong to the VCSA subgroup and are found only in the rat. The gene coding for the anti-shock peptide SGP-T is one of these and our work provided the first description of a biological activity for a peptide product of a VCSA gene.
4.
MECHANISM OF ACTION
Since isolating SGP-T we have tested it in various models of shock and inflammation several of which have been alluded to above and another is reported in an article in these proceedings (Befus et al.). As a result we know that SGP-T and certain analogues such as FEG and feG can attenuate both immediate and later phases of shock or inflammation. Our studies of the mechanism of action of these novel antishock/antiinflammatory peptides has focused on the recruitment of inflammatory cells during the early to middle periods of inflammation. These cells are the source for the development of the late phase reaction. We have shown that SGP-T and analogues inhibit leukocyte rolling (14) an important first step in the recruitment and subsequent activation of leukocytes. This process of recruitment is initiated by chemotactic agents and inhibition of recruitment was one of the first properties we identified for our putative anti-shock hormone prior to its isolation (15). Recruitment and activation is regulated by a large family of integrin-associated proteins and some of our recent work has focused on the actions of feG on integrin expression. We have found that
310
intraperiteonal injection of LPS into rats increases the expression of CD18 on mesenteric tissue macrophages by 4-fold and that a single treatment with feG (100 ~tg/kg) abrogated the increase in this cell activation marker [16]. In addition, we found that feG reduces the expression of CDllb and CD16b (Fc~,RIIIb) on isolated human neutrophils provoked by platelet activating factor (PAF) [17]. Since the peptide did not have noticeable effects on CD43, CD62L or CD162, we are actively exploring the role of feG in regulating the expression and activity of the [3-integrins and IgG-receptors. These results are consistent with our original hypothesis and provide us with the beginning of an explanation for the action of these peptides. By blocking the expression of important regulatory integrins, including the LPS receptor CD14, these salivary gland peptides are able to inhibit recruitment and activation of leukocytes such as neutrophils and eosinophils which are responsible for tissue damage in late phase inflammation. In summary, our collective work to date implies that there is an important sympathetic pathway through the superior cervical ganglion, that regulates the release of novel peptides from the submandibular glands which play an important role in immediate and later phase immune responses during anaphylaxis and LPS-induced inflammation. In future work we hope to determine, not only the molecular mode of action of these peptides, but also to return to the point where we began and use our increasing knowledge of molecular mechanisms to develop the tools to study the neuroregulation of the production and release of these immunoregulatory hormones.
REFERENCES 1. 2. 3.
4. 5.
6. 7. 8. 9.
10.
11.
Barka T. Biologically active polypeptides in submandibular glands. J Histochem Cytochem 1980; 28: 836-859. Mathison R, Befus D, Davison JS. Neuroendocrine regulation of inflammation and tissue repair by submandibular gland factors. Immunology Today 1994; 15: 527-532. Ramaswamy K, Mathison R, Carter L, Kirk D, Green F, Davison JS, Befus D. Marked antiinflammatory effects of decentralization of the superior cervical ganglia. J Exp Med 1990; 172: 1819-1830. Mathison R, Befus D, Davison JS. Removal of the submandibular glands increases the acute hypotensive response to endotoxin. Circ Shock 1993; 39: 52-58. Mathison RD, Befus AD, Davison JS. A novel submandibular gland peptide protects against endotoxic and anaphylactic shock. Am J Physiol 1997; 273 (Regulatory Integrative Comp Physiol 42): R1017-R1023. Mathison RD, Befus AD, Davison JS. Attenuation of cardiovascular anaphylaxis by submandibular gland peptide-T (SGP-T). Proc West. Pharmacol Soc 1997; 40: 5-7. Mathison RD, Tan D, Oliver M, Befus AD, Davison JS, Scott B. A novel peptide from submandibular glands inhibits intestinal anaphylaxis. Dig Dis Sci 1997; 442: 2378-2383. Mathison RD, Davison JS, Moore G. Submandibular Gland Peptide-T (SGP-T). Modulation of endotoxic and anaphylactic shock. Drug Discovery Research 1997; 42: 164-171. Mathison RD, Malkinson T, Cooper KE, Davison JS. Submandibular glands: Novel structures participating in thermoregulatory responses. Can J Physiol Pharmacol 1997; 75: 407-413. Mathison R, Kubera M, Davison JS. Submandibular Gland Peptide-T (SGP-T) modulates ventricular function in response to intravenous endotoxin. Pol J Pharmacol 1999; 51: 331-339. Mathison RD, Lo P, Davison JS, Scott B, Moore G. Attenuation of intestinal and
311
12.
13.
14. 15.
16. 17.
cardiovascular anaphylaxis by the salivary gland tripeptide FEG and its D-isomeric analogue feG. Peptides 1998; 19: 1037-1042. Rosinski-Chupin I, Tronik D, Rougeon F. High level of accumulation of a mRNA coding for a precursor-like protein in the sub-maxillary gland of male rats. Proc Natl Acad Sci USA 1988; 85: 8553-8557. Rosinki-Chupin I, Rougeon F. One amino acid change in rate SMRI polypeptide induces a lkDa difference in its apparent molecular mass determined by electrophonetic analysis. FEBS Lett 1990; 267: 147-149. Mathison R, Sank C, Davison JS. Inhibition of leukocyte rolling by submandibular gland peptide-T (SGP-T). Proc West. Pharmacol Soc 1999; 42: 39-40. Carter L, Ferrari JK, Davison JS, Befus D. Inhibition of neutrophil chemotaxis and activation following decentralization of the superior cervical ganglia. J Leukoc Biol 1992; 51: 597-602. Mathison RD, Lo P. Attenuation of intestinal endotoxemia in rats by the salivary gland tripeptide FEG and its d-isomeric analog feG. INABIS Symposium. Mathison RD, Teoh D, Woodman R, Lo P, Davison JS, Befus D. Regulation of neutrophil function by SMR1 C-terminal peptides. Shock 2000; 13 (Suppl): 52.
This Page Intentionally Left Blank
New Foundation of Biology
313
Edited by I. Berczi and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
Olfactory Stimuli and Allo-recognition
MALCOLM G. B AINES,
Dept. Microbiology and Immunology, McGill University, 3775 University St. Montreal, QC., Canada H3A-2B4
ABSTRACT A number of species have demonstrated the ability to recognise related individuals by scent alone in a manner which may relate to their major histocompatibility (MHC) genotype. Further, the scent of the members of the opposite sex can activate aggressive responses, affect mating preference, induce oestrus, induce implantation delay or loss, augment early embryo survival and even alter the quality of the recipients immune defences. Two well known reproductive responses to pheromonal stimuli include the Whitten and Bruce effects. The "Whitten effect" is initiated by the presence of odours of an alien male, resulting in the onset of oestrus in un-mated female mice. The "Bruce effect" defines a complete pre-implantation block of pregnancy in mated females exposed to odours of an alien male. Conversely, if an alien male is introduced to a pregnant mouse on the day after implantation the incidence of spontaneous early embryo losses may be reduced. Therefore, neuroendocrine mediators induced by pheromonal messages derived from the resident male can alter the maternal cell-mediated immune response in the uterus to the fetal graft, dramatically affecting the outcome of pregnancy.
1.
INTRODUCTION
Reproduction in mammals has long intrigued researchers in both basic and clinical immunology because the long-term acceptance of the semi-allogeneic fetal graft throughout gestation appears to violate the basic principles of transplantation biology. While it is obvious that mate selection is governed by the senses of the participants, recent data suggests that the sense of smell is arguably the most important of the senses (Table I). The senses of vision and hearing, which operate from a distance, provide only a general assessment of the attractions of a mate. However, the sense of smell can define the general health, the readiness of the subject to mate and the genetic suitability of that individual as a mate. The senses of taste and touch, which require much closer contact, may serve to confirm this evaluation. Most importantly, the sense of smell can define kinship as defined by the expression of antigens of the major histocompatibility complex (MHC). Part of the sense of smell resides in the vomeronasal organ (VNO). The VNO sensory tissue is located in a tubular organ in the base of the anterior part of the nasal septum where it is supplied with bipolar sensory neurones, which are connected to the accessory olfactory bulb [1]. This organ appears to be important but not essential for
314
detection of MHC related odours. Endocrine hormones augment the activity of the VNO in gonadectomized animals [2] and test odours trigger neuronal activity [3]. Removal of the vomeronasal organ reduced but did not completely eliminate odour discrimination. However, the anterior bilateral transection of the lateral olfactory tract eliminated the identification of mice by odour, suggesting the involvement of the accessory olfactory system in the transmission of pheromonal stimuli [4, 5]. Table I
Involvement of the senses in reproduction.
Senses
Sensory organs
Reproduction
Vision
Eyes
General physical health
Sound
Ears
Acoustic attractants
Touch
Skin
Tactile stimulation
Taste
Tongue
Oral stimulation
Smell
Nose
Gender, health, kinship, oestrus, receptiveness
2.
OLFACTORY RESPONSES AND REPRODUCTION
Apanius et al and Penn have published recent reviews on odour discrimination, mating preference and MHC selection [6, 7]. As a general introduction to the sources of data on this subject, it is necessary to know that only some experiments involve the observation of wild or inbred mice in controlled 'semi-wild' natural settings. Most researchers have focused on the training of inbred mice and rats to recognise MHC specific odours in various olfactory testing devices in which the odours of 'donor' mice are alternately blown to, or sensed by, the recipient whose 'correct' responses are noted. Early experiments used a passive testing box where untrained subjects could move towards preferred odours and mate with preferred partners (Figure 1) [8, 9]. Subjects can now be placed in automated training and testing chambers which are supplied with test odours and responses are compiled by computer (Figure 2) [10, 11]. Whether these two basic types of experiments entirely accurately reflect the natural mating preferences of mice in the wild is not certain although the data obtained confirms responses observed in 'semi-wild' experiments and helps explain the unusual distribution of MHC alleles in wild populations. Further, there can be many confounding elements in experimental studies, which have to be controlled or accounted for in the final analysis. Such factors include the genetic selection against allo-MHC preference that is inevitably associated with the breeding of MHC inbred mouse strains which might alter responses of the subjects. Using 'outbred' mice avoids this problem. The olfactory 'noise' from non-MHC genes for gender and other genetic factors combined with exogenous factors such as diet in 'semi-wild' mice can influence responses [12]. The effects of gene dosage on donor odour concentration and the inherent strength or hierarchical dominance of odours must also be taken into account. However, it is clear that prior to mating, all animals tested thus far can detect the foreign scents of the presumptive mate and this can influence willingness to mate, the implantation of the blastocyst and the successful development of the implanted embryo.
315
PASSIVE APPARATUS FOR TESTING ODOR PREFERENCE OF MICE.
30 cm
@ @ @ A and C contain the test and control odors. B is where the test subject starts the test.
Figure 1. A passive apparatus for testing odor preference of mice. This device is primarily used for assessing odor preferences of untrained mice without a reward to encourage correct choices. The test subject always starts at position "B" and has the choice of attractants "A" or "C" and the preference for "A" over "C" is noted. The attractant may be bedding scent marked by another mouse or it may be a mouse tethered so that it can not stray from the "home" cubicle. Adapted from Egid and Brown Anim. Behav 38 (548), 1989.
3.
ODOURS: G E N E R A L AND MHC SPECIFIC
A m o n g most animals, natural odours are used for kin recognition, scent marking of territory, aggression/dominance, defence/flight and reproduction. The basis for odour recognition or preference could be instinctive, learned or by serendipity (Table II). Such scents are found in the volatile fraction of sweat, saliva, urine and faeces. Odours can be defined by both exogenous and endogenous factors and are most certainly expressed simultaneously as a complex multifactorial signal. Exogenous factors affecting the odour of an individual can originate from diet, microbial flora, general body secretions and interactions between these factors. The diet of mice, rats and humans, can add m a n y volatile factors to urine (e.g. garlic, asparagus, meat diet versus vegetarian diet) [13]. In fact, animals can learn to detect the dietary differences between individuals quicker than they can learn strain specific differences and therefore, diet cues may occasionally mask M H C associated kinship scents [12, 14]. The familiar c o m m e n s a l microbial flora colonising the individual may also be detectable and could vary inversely with infection by pathogenic microbes. The health of the mate is important as infected mates are not accepted
316
L.
0
L.
0
0
gin._
E
0
0
N
0
ira__
E
0
w
Computerized Olfactometric Apparatus for Odor Identification. 1
e
m
2
~ m l]
Y
1
,--t,t..
c-.l
c~
-4
Z
i
Beauchamp et a1 P.N.A.S. 92 (2617), 1995 .c>~9
"~
9,.~ m c~ ~m m
i
;>
9,.~
c~
.~~
"io
9~.~
~ ~o ~
9
c~
Figure 7. Computerized olfactometric apparatus for odor identification. This device is primarily used for assessing odor preferences of mice or rats that have been trained with a reward to e n c o m g e correct choices. The volatile scents are injected into the air stream and when the subject identifies the correct odor. a reward of water is supplied. The rdt move5 forward in the chamber, breaking a light beam (Inset 1 and 2) and auto~naticallyrecording its response for the computer. Adapted from Beauchamp et al
317
even when they express an allogeneic MHC [15, 16]. Healthy males will avoid infected females and healthy females will avoid infected males [17]. When both are infected there is less discrimination against infection and kinship indicating that olfactory imprinting can be learned from sensing odours related to self [18]. Further, the scent of an infected male appears to 'suppress' the sensitivity or receptivity of the female to allogeneic male scents and even heat stimuli [15]. It has also been suggested that the action of commensal microbes may be required for the processing of both exogenous and endogenous macromolecules into the volatile fragments that are detected by the vomeronasal organ. However, as germ-free mice can identify MHC associated scents, the microbial component may complement but is not critical for recognition [19]. Endogenous factors detected by the sense of smell are genetically controlled factors specific for the individual, some of which may be related to products of the sex chromosomes and endocrine factors related to gender and maturity [20]. Species related factors are also detectable in the excretions of animals (e.g. rat versus mouse). Males are more interested in unrelated females or their bedding (non-kin) [21], which contains both common and specific factors [22]. Common female factors are those that indicate that the female is receptive or has commenced oestrus. While males strongly respond to the indicators of oestrus, it is also clear that the difference between female oestrus versus dioestrus scents is learned by juvenile male mice [23]. Whereas many reproduction related scents appear to relate to the MHC specificity of the individual, specific scents from specific genes such as the lethal t-complex gene can been detected, as +/+ females prefer +/+ males to +/t males [9]. However, the fact that both male and female can distinguish relatives from strangers and choose the latter, is perhaps one of the most interesting facets of mate selection. The role of MHC associated volatile compounds in defining the odourtype of the individual, still lacks many details. MHC associated odours and proteins have been shown to be present in urine and other secretions of individual mice but are not dominant over all other factors, such as diet and oestrus related odours [22]. Further, mouse serum and purified mouse MHC proteins or synthetic MHC peptides are not specific attractants and MHC proteins can be separated from the more volatile active attractants in natural excretions [24, 25]. The suggestion that MHC associated serum proteins may bind or carry volatile carboxylic acid odourants that could be liberated by cellular and/or microbial action, was supported by the observation that pronase digestion of MHC or serum proteins produced volatile attractants [11, 26, 27]. The source of MHC associated odours has been shown to be the bone marrow. Radiation chimeras were produced by repopulating the hemopoietic system of mice of one strain with the marrow of a second. The MHC chimerism was detectable in the urine of these mice by odour response tests [28]. Using the "Y"-maze habituation dis-habituation training process, mouse or rat males could learn to identify MHC specific odours from congenic donors with 80-90% accuracy (Figure 3) [22]. MHC detection can be trained for a difference at a single MHC class-I or MHC-II allele [29, 30]. Even the lack of an MHC allele in an MHC deletion mutant could be detected by trained rats [31]. In fact, the specificity of the MHC detection can be trained for a few amino acid differences in a single mutation in a class-I MHC allele H2 b bm or b m l [32]. In recent experiments, rats trained to detect MHC related odours in human urine, could even identify paternal odours derived from the fetal tissues in the uterus of pregnant donors [10]. The major remaining questions concern the nature of the normal endogenous process of 'volatilisation' of the MHC macromolecules, the chemical structure of the scents and whether microbial mediators are required or simply complement the process.
318
Y-MAZE APPARATUS FOR TESTING ODOR PREFERENCES Air
*---Fan
......
I 9
Odor Box Petri Dish Containing Urine or ~rum
Water Fence
L
R
~I~J/~/7
Gates
Gate
Figure 3. The "Y"-maze apparatus is also primarily used for assessing odor preferences of mice or rats that have been trained with a reward to encourage correct choices. The volatile scents are injected into the air stream of one arm of the "Y" and when the subject identifies the correct odor by moving towards it, a reward of water is supplied. Each test is a choice of two odors. Adapted from Yamazaki et al., P.N.A.S 96 (1522), 1999.
319
Table II
Basis and effects of odour recognition in reproduction.
Odour Response
Effect of allogeneic odour recognition.
Instinct
Avoidance of genetically related mice (littermates)?
Learned
Avoidance of MHC pheromones previously sensed?
Curiosity
Select an allogeneic mate who smells different or more interesting?
4.
MHC BASED MATE PREFERENCE SELECTION
It is now apparent that volatile derivatives or components associated with MHC are contained in the body secretions and excretions of all species tested thus far. Males prefer the odours of females with an MHC which is allogeneic versus syngeneic in tests using congenic females [33, 34]. However, some studies of actual mating in the passive chamber device appear to contradict this point as males occasionally mate without MHC preference [8]. One explanation is that in these tests, interest of a recipient in a donor scent is a one-way response that may be totally under the control of the recipient while mating is a two-way event that requires some physical co-operation by both parties. Consequently, in mating tests using tethered females in the passive test chamber, males seemed to show no preference for MHC, whereas females continued to show allo-MHC preference when the males were tethered [8]. Therefore, even though female preference for allogeneic MHC versus syngeneic MHC in congenic males appears stronger than male preference, this may be an artefact of the assay format [35]. Males show no preference for MHC-identical congenic females or MHC-identical females with different non-MHC background genes implying that non-MHC associated factors are either non-existent or are much weaker in this type of experiment [8]. Finally, the question as to the genetic v e r s u s environmental basis for self or kinship recognition was answered by cross-fostering experiments. Cross-fostering allogeneic mouse pups to MHC different dams changes the specificity of MHC based mate selection, supporting a mechanism based on familial imprinting [33, 34, 36, 37]. All the female mice derived from cross-fostered familial settings avoided mating with males expressing MHC alleles identical to the original a n d the cross-fostered members of the litter. The benefits of mating with healthy, allogeneic males provides a significant advantage both for the individual female and for the population by creating offspring with heterozygous resistance to disease and preserving existing and new MHC polymorphism in the gene pool (Table III). In conclusion, most mates instinctively prefer partners with an MHC different from their parents and their littermates. However, opportunity to mate may over-rule choice in situations where the female can not resist, or the signal indicating the oestrus state or receptiveness is stronger than the allo-recognition signal.
320
5.
PREGNANCY: PHEROMONE-INDUCED EVENTS FOLLOWING MATING
5.1.
Induction of oestrus: the "Whitten effect"
Female receptivity is also altered in response to a number of pheromonal stimuli related to the Y chromosome, health, ovulation, endocrinology and other factors. While the male can naturally detect oestrus-related scents from female mice, the oestrous cycle of female mice can also be altered by the presence of male mice. The Whitten effect defines the induction and synchrony of oestrus by exposure of grouped female mice to corralled males. The volatile factor is present in male urine and is effective even when direct contact between male and female is prevented [38]. The response of female mice is also directly related to the allogeneicity of the males as previously described. Therefore, male scent induces females to ovulate and female scent of 'heat' attracts the males. 5.2.
Pre-implantation pregnancy block: the "Bruce Effect"
Once mating has taken place, the MHC alloantigens of the male can still have significant effects on pregnancy outcome. The most prominent pre-implantation example is known as the Bruce effect which describes a virtually total pregnancy block or the prevention of implantation by exposure to alien males so that the female mouse can return to oestrus (Figure 4) [39]. The corpora lutea degenerate and all fertilised ova fail to implant and are lost. The apparent value of this response would be to allow the female to quickly block the progression of the current pregnancy, so that she can mate with the new and more successful dominant male. Pregnancy block optimally requires from 48 to 72 hours of contact with the new male (Figure 5) and can be induced at any time up to the time of implantation (Figure 6) [40, 41]. The induction of the Bruce effect requires direct contact with the urine, soiled bedding or other male derived factors (Figure 7) [42]. The effect is greater for MHC different alien mice, whether they are male or female though responses to male factors appear stronger [43, 44, 45]. A castrated male does not induce these effects implying that testosterone may be required. However, castrated males injected with testosterone induce pregnancy block in mated mice, confirming this requirement. Direct contact with the 'resident' male is required for this effect, suggesting a poorly volatile factor [42]. If a cage within a cage format physically separates the male and female from each other, the volatile scents are less effective in proportion to the degree of separation (Figure 8). Urinary odours or factors associated with differences in the X and Y-chromosomes also induce pregnancy block. However, there is no convincing evidence of gender selective embryo loss that could affect the sex ratio of the offspring [20, 46, 47]. Pregnancy block occurs in entirely untrained mice, indicating that the fine discrimination of MHC associated odours is a natural event. Early studies by Bruce and Dominic clearly demonstrated that the alien male elicited the implantation blocking effect via androgendependent pheromones which appeared in male urine and acted via a maternal neuroendocrine pathway involving decreased prolactin and progesterone secretion (Figure 9) [44, 48, 49, 50]. The augmentation of prolactin or stimulation of endogenous prolactin secretion by reserpine, reduced pregnancy block (Figure 9) [50]. Similarly, supplementing progesterone levels also reduced pregnancy block in some experiments but not in others depending on the time of injection (Figure 9) [44, 51, 52, 53].
321
MALE INDUCED PREGNANCY BLOCK THE "BRUCE" EFFECT: RETURN TO ESTRUS. Day of return to estrus
DAY 2 DAY 3 DAY4 DAY 5
B
DAY 6 I TOTAL 0
10
20
30
40
50
60
PERCE NT BLOCK 24 Hours with CBAJG on day 0 Figure 4. Male induced pregnancy block, the "Bruce" effect: Return to oestrus. Outbred Parkes strain albino female mice were mated to albino-P males and then introduced to a CBA/G mouse for 24 hours on the first day after the mating plug was detected. This normally causes about half the females to return to oestrus, mostly within two days. Adapted from Bruce, J Reprod Fert 2 (138-142) 1961.
MALE INDUCED IMPLANTATION BLOCK THE "BRUCE" EFFECT" INDUCTION TIME. HOURS OF STRESS Stud AIb-P male New male 12 Hrs New male 24 Hrs New male 48 Hrs New male 72 Hrs
. 0
20
.
. 40
. 60
80
100
PERCENT BLOCK Figure 5. Male induced implantation block, the "Bruce" effect: Induction time. Outbred Parkes strain albino female mice were mated to albino-P males and then introduced to a new CBA/G mouse for variable lengths of time starting on the first day after the mating plug was detected. This greatest incidence of pregnancy block was achieved by two or more days of exposure. Adapted from Parkes and Bruce, J Reprod Fert 4 (303-308) 1963.
322
MALE INDUCED PREGNANCY BLOCK THE " B R U C E " EFFECT" S E N S I T I V E PERIOD. 90 80700 60~50 z40 uJ 30uJ 20 100
2
4
6
8
DAYALIEN MALEINTRODUCED Figure 6. Male induced pregnancy block, the "Bruce" effect: The sensitive period extends from the first day after mating to the point of implantation. Once the blastocyst has implanted in the uterine wall, pregnancy block does not occur. Bruce, J Reprod Fert 2 ( 138-142) 1961.
MALE INDUCED IMPLANTATION BLOCK T H E " B R U C E " EFFECT: ( C O N T A C T F A C T O R ) STRESS FACTOR Stud AIb-P male New AIb-P male New CB.A~ male Castrated CBMG Testosterone S/C New female/none 0
20
40
60
80
100
PE RCENT PREGNANCY BLOCK Figure 7. Male induced implantation block, the "Bruce" effect: Contact with the factor. Outbred Parkes strain albino female mice were mated to albino-P males and then introduced to a new albino-P or a CBA/G male starting on the first day after the mating plug was detected and continued until the female returned to oestrus or was obviously pregnant. If the CBA/G male had been castrated, the incidence of pregnancy block was reduced but could be partly restored by injections of testosterone to the castrated male. Adapted from Bruce, J Reprod Fert 1 (96-103) 1960.
323
MALE INDUCED IMPLANTATION B L O C K THE "BRUCE" EFFECT: (ISOLATED FACTOR). FACTOR Stud AIb-P male
STRESS
!!i!i!ii!ilililii!!iii!!!i!~!i !iii!i!!!ii~i!i~i!!ii
New AIb-P male New CBA/G male [Caged female] + CBA/G Female + [Caged CBA/G] [Female] + [CBA/G male] 0
20
40
60
80
100
PE RCENT BLOCK
Figure 8. Male induced implantation block, the "Bruce" effect: Isolated from the factor. If either the male or the female is placed in an internal cage, which prevents direct contact, pregnancy block is reduced. If both are kept apart in separate mini-cages, pregnancy block is further reduced indicating that the smell of the volatile component is less effective than direct contact with the scent marked bedding. Adapted from Bruce J Reprod Fert 1 (96-103) 1960.
MALE INDUCED IMPLANTATION BLOCK THE "BRUCE" EFFECT: ENDOCRINOLOGY.
Stud AIb-P male Alien CBA/G male
i - -
CBA male urine (AMU) AMU + Prolactin I/M
!
AMU + Reserpine I/M AMU + Progesterone
R
0
20
40
60
80
100
PERCENT P R E G N A N C Y B L O C K Figure 9. Male induced implantation block, the "Bruce" effect: Endocrinology. Outbred Parkes strain albino female mice were mated to albino-P males and then introduced to a new CBA/G male or CBA/G urine starting on the first day after the mating plug was detected. Two other groups of mated albino-P mice were injected with prolactin or reserpine, which increased prolactin secretion. A final group was injected with progesterone. Dominic, J Reprod Fert 11(415-421), 1966.
324
6.
POST-IMPLANTATION ENHANCING EFFECTS ON PREGNANCY
The presence of MHC related odours from allogeneic males also affects post-implantation embryo survival. While most strains of inbred mice show a very low incidence of postimplantation spontaneous early embryo loss (<10%), losses for certain mating models such as the CBA/j ~? x DBA/2 ~, can approach 40% [54]. The survival of litters of implanted embryos is enhanced by the presence of the pheromones of an allogeneic or "alien" male or its soiled bedding (Figure 10) [55]. The relatively high incidence of embryo loss was not affected by removal and replacement of the original DBA/2 6' with another DBA/2 or a CBA/J ~, possibly indicating that the female was familiar with the scent of DBA and CBA MHC alleles. However, the replacement of the original DBA/2 6' by a BALB/c ~ (H2d), dramatically reduced the incidence of early embryo loss in pregnant CBA/J $ mice and the introduction of a C57B1/6 6' also reduced embryo losses but to a lesser extent. While C57B1/6 mice are H2 b, BALB/c and DBA/2 ~ both express the same H2 d MHC alleles, indicating that other allogeneic factors may also be sensed. In contrast, the introduction of a DBA/2 6' to a CBA/J, which had been mated by a B ALB/c ~, did not increase the normally low incidence of early embryo loss in this mating. While this 'alien ~ ' effect is not identical for all allogeneic 6', the lack of MHC specificity appears to indicate the involvement of non-MHC related genes. Further, the reduction of spontaneous embryo loss in CBA x DBA (HU x H2 d) pregnancies by exposure to a B ALB/c 6' (H2 d) was not testosterone dependent as castrated B ALB/c ~' were as effective as intact males at preventing embryo loss. Histologically, there was a significant reduction in the infiltration of uterine implantation sites by maternal macrophages and natural killer cells coincident with the reduction in early embryo losses. Therefore, the maternal cell-mediated immune response in the uterus to the fetal graft can be altered by neuroendocrine mediators induced by pheromonal messages derived from the resident male, normally resulting in an improved survival of the implanted embryos. The result of this response would be the successful completion of a pregnancy to which the pregnant female has already committed considerable resources. Therefore, male derived pheromonal stimuli may play a role in both eliciting and preventing early pregnancy loss.
7.
HUMAN DETECTION OF MHC ASSOCIATED ODOURS
As previously found for rodent species, human MHC associated proteins were found in human urine, saliva and sweat. Urinary MHC in female urine increases at oestrus and around ovulation in humans [56]. Human MHC associated proteins may also carry volatile carboxylic acid odourants [26, 27]. While the sense of smell is not as well developed in humans, in odour preference tests, humans can distinguish differences between the odours of different mouse strains [57]. Further, 74% of humans can identify their own sweaty clothes by smell alone [58]. Human mothers can identify the odour of their own offspring even when the presence of cosmetic scents and child-care products are eliminated [59]. Humans can also detect HLA dis-similarity by smell and associate that with a pleasant impression, and they are more likely to pair with MHC differing partners, particularly in respect to maternal MHC haplotype [60, 61]. Progesterone containing birth control pills appear to block or even reverse MHC associated allorecognition [61]. Whereas mating preferences and pregnancy block can significantly influence population MHC heterozygosity in rodents, which have short gestation periods, there may be evidence, which supports the concept of selection for heterozygous human offspring through spontaneous abortion of homozygous embryos where the parents share MHC alleles [62].
325
EARLY PREGNANCY ENHANCEMENT: THE ALLEN MALE EFFECT. No male change DBM2 to DBAJ2 DBA to C57BI/6 DBA to BALB/C 0
5
10
15
20
25
PE RCE NT RE SORBE D CBAJJ X DBAJ2 MALES Figure 10. Early pregnancy enhancement, The 'Alien male' effect. In CBA/j females (HZk) mated with DBA/2 males (HZd), 20-30% of the implanted embryos resorb (die) by the 12th day of gestation. In contrast, CBA/j females mated to BALB/c males lose only 5-10% of the embryos by day 12. The introduction of a BALB/c male (HZd) or its scent marked bedding to the cage containing the pregnant CBA/j female results in a significant reduction in embryo loss. C57B1/6 males (HZb) induce an intermediate beneficial effect. Adapted from Baines et al., J.Reprod. Fert 102, (221) 1994.
In terms of f e m a l e r e s o u r c e utilisation and effective r e p r o d u c t i v e strategies for g e n e transmission, the t e r m i n a t i o n o f a 3 - w e e k m u r i n e p r e g n a n c y seems less a d v a n t a g e o u s t h a n h a l t i n g a 9 - m o n t h single h u m a n p r e g n a n c y at an e a r l y stage. In conclusion, M H C a s s o c i a t e d m a t i n g preferences also a p p e a r to o c c u r in h u m a n s t h o u g h all the m e c h a n i s m s l e a d i n g to M H C d i s - a s s o r t m e n t o f o f f s p r i n g are not yet clear.
Table III
Benefits of MHC selection in reproduction.
Benefit Disease resistance
Mechanism or effect Heterozygosity favoured: increased genetic diversity for individual resistance genes and immunity.
Decreased birth defects
Suppression of inbreeding: Reduced occurrence of recessive genetic anomalies.
Genetic evolution
Increased gene polymorphism: new or rare genes are positively selected during reproduction.
326
8.
CONCLUSIONS
In summary, there appears ample evidence that MHC-associated mate selection by odour exists and significantly affects the heterozygosity of the MHC genes and, by gene linkage, many other desirable traits related to resistance and immunity to pathogens. The basis for odour recognition by both males and females could be genetic, environmental and psychological though post-natal familial imprinting appears the most important factor. The results of MHC-associated mate selection by the sense of smell are most evident in the enormous polymorphism of the MHC locus and the bias towards heterozygosity of the MHC gene pool. By selecting against mating with members of the opposite sex that share some or all of the MHC antigens with us, we inadvertently select against common MHC alleles and favour rarer MHC alleles. This has the effect of amplifying rare MHC alleles and reducing the expression of more common alleles. The mechanisms of regulation of mate selection and reproductive success may differ somewhat from one species to another depending on their instinctive or customary strategies for reproduction. Whether we are aware of them or not, MHC related odours have a lot more to do with our everyday reproductive lives than we may have previously anticipated.
REFERENCES 1. 2.
3.
4.
5.
6.
7. 8.
9. 10.
11.
Keverne EB. The vomeronasal organ. [Review] [56 refs]. Science 1999; 286: 716-720. Paredes RG, Lopez ME, Baum MJ. Testosterone augments neuronal Fos responses to estrous odors throughout the vomeronasal projection pathway of gonadectomized male and female rats. Hormones & Behavior 1998; 33: 48-57. Krauel K, Pause B M, Mueller C, Sojka B, Mueller-Ruchholtz W, Ferstl R. Central nervous correlates of chemical communication in humans. Annals.of the. New York. Academy. of Sciences 1998; 855: 628-631. Petrulis A, Peng M, Johnston RE. Effects of vomeronasal organ removal on individual odor discrimination, sex-odor preference, and scent marking by female hamsters. Physiol Behav 1999; 66: 73-83. Rajendren G, Dominic CJ. Effect of bilateral transection of the lateral olfactory tract on the male-induced implantation failure (the Bruce effect) in mice. Physiol Behav 1986; 36: 587-590. Apanius V, Penn D, Slev PR, Ruff LR, Potts WK. The nature of selection on the major histocompatibility complex. [Review] [388 refs]. Critical Reviews in Immunology 1997; 17:179-224. Penn D, Potts W. How do major histocompatibility complex genes influence odor and mating preferences?. [Review] [107 refs]. Advances in Immunology 1998; 69: 411-436. Eklund A, Egid K, Brown JL. The major histocompatibility complex and mating preferences of male mice. Animal Behaviour 1991; 42: 693-694. The chemosensory recognition of individuality in rats. Physiol Behav 1987; 40: 73-79. Lenington S, Egid K, Williams J. Analysis of a genetic recognition system in wild house mice. Behavior Genetics 1988; 18: 549-564. Beauchamp GK, Katahira K, Yamazaki K, Mennella JA, Bard J, Boyse EA. Evidence suggesting that the odortypes of pregnant women are a compound of maternal and fetal odortypes. Proceedings. of the National. Academy. of Sciences. of the. United.States. of America 1995; 92: 2617-2621. Yamazaki K, Beauchamp GK, Singer A, Bard J, Boyse EA. Odortypes: their origin and
327
12.
13.
14.
15.
16.
17.
18. 19.
20. 21. 22. 23. 24. 25. 26.
27. 28.
29.
composition. Proceedings. of the National. Academy. of Sciences. of the. United.States. of America 1999; 96: 1522-1525. Schellinck HM, Slotnick BM, Brown RE. Odors of individuality originating from the major histocompatibility complex are masked by diet cues in the urine of rats. Animal Learning & Behavior 1997; 25:191-199. Schellinck HM, West AM, Brown RE. Rats can discriminate between the urine odors of genetically identical mice maintained on different diets. Physiology & Behavior 1992; 51: 1079-1082. Brown RE, Schellinck HM, West AM. The influence of dietary and genetic cues on the ability of rats to discriminate between the urinary odors of MHC-congenic mice. Physiol Behav 1996; 60: 372-378. Kavaliers M, Colwell DD, Choleris E. Analgesic responses of male mice exposed to the odors of parasitized females: effects of male sexual experience and infection status. Behavioral Neuroscience 1998; 112:1001-1011. Klein SL, Gamble HR, Nelson RJ. Trichinella spiralis infection in voles alters female odor preference but not partner preference. Behavioral Ecology & Sociobiology 1999; 45: 322-329. Kavaliers M, Colwell DD, Ossenkopp KP, Perrot-Sinal TS. Altered responses to female odors in parasitized male mice: Neuromodulatory mechanisms and relations to female choice. Behavioral Ecology & Sociobiology 1997; 40: 373-384. Kavaliers M, Colwell DD, Choleris E. Parasitized female mice display reduced aversive responses to the odours of infected males. Proc R. Soc Lond. 1998; 265:1111-1118. Yamazaki K, Beauchamp GK, Imai Y, Bard J, Phelan SP, Thomas L, Boyse EA. Odor types determined by the major histocompatibility complex in germfree mice. Proceedings. of the. National. Academy. of Sciences. of the United. States. of America. 87: 1990; 8413-8416. Yamazaki K, Beauchamp GK, Bard J, Boyse EA. Chemosensory identity and the Y chromosome. Behavior Genetics 1990; 20: 157-165. Krackow S, Matuschak B. Mate choice for non-siblings in wild house mice: Evidence from a choice test and a reproductive test. Ethology 1991; 88: 99-108. Eggert F, Hoeller C, Luszyk D, Mueller-Ruchholtz W. MHC-associated and MHCindependent urinary chemosignals in mice. Physiol Behav 1996; 59: 62-70. O'Connell RJ. Estrous cycle modulation of the attraction to odors in female golden hamsters. Behavioral & Neural Biology 1983; 37: 328-325. Singer AG. A chemistry of mammalian pheromones. [Review] [56 refs]. Journal. Of. Steroid Biochemistry & Molecular. Biology 1991; 39: 627-632. Singer AG. A chemistry of mammalian pheromones. [Review] [56 refs]. Journal of Steroid Biochemistry & Molecular Biology 1991; 39: 627-632. Singer AG, Beauchamp GK, Yamazaki K. Volatile signals of the major histocompatibility complex in male mouse urine. Proceedings.of the. National. Academy. of Sciences. of the United States of America 1997; 94: 2210-2214. Yamazaki K, Singer A, Beauchamp GK. Origin, functions and chemistry of H-2 regulated odorants. [Review] [59 refs]. Genetica 1998; 104: 235-240. Eggert F, Uharek L, Mueller-Ruchholtz W, Ferstl R. MHC-associated and MHCindependent urinary chemosignals in mice are expressed via the hematopoietic system. Neuropsychobiology 1994; 30: 39-45. Brown RE, Roser B, Singh PB. Class I and class II regions of the major histocompatibility complex both contribute to individual odors in congenic inbred strains of rats. Behavior
328
30. 31. 32.
33.
34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44.
45.
46. 47.
48. 49. 50. 51.
Genetics 1989; 19: 659-674. Brown RE, Singh PB, Roser B. The Major Histocompatibility Complex and the chemosensory recognition of individuality in rats. Physiol Behav 1987; 40: 73-79. Yamazaki K, Beauchamp GK, Shen FW, Bard J, Boyse EA. A distinctive change in odortype determined by H-2D/L mutation. Immunogenet 1991; 34:129-131, 1991. Yamazaki K, Beauchamp GK, Egorov IK, Bard J, Thomas L, Boyse EA. Sensory distinction between H-2b and H-2bml mutant mice. Proc Natl Acad Sci (USA) 1983; 80: 5685-5688, 1983. Beauchamp GK, Yamazaki K, Bard J, Boyse EA. Preweaning experience in the control of mating preferences by genes in the major histocompatibility complex of the mouse. Special Issue: Kin recognition in animals. Behavior Genetics 1988; 18: 537-547. Yamazaki K, Beauchamp GK, Kupniewski D, Bard J. Familial imprinting determines H-2 selective mating preferences. Science 1988; 240:1331-1332. Egid K, Brown JL. e major histocompatibility complex and female mating preference in mice. Animal Behaviour 1989; 38: 548-550. Eklund A. The effect of early experience on MHC-based mate preferences in two B 10.W strains of mice (Mus domesticus). Behavior Genetics 1997; 27: 223-229. Penn D, Potts W. MHC-disassortative mating preferences reversed by cross-fostering. Proceedings. of the Royal Society of London- Series B. Whitten WK. Pheromones and mammalian reproduction. Adv Reprod Physiol 1966; 1: 155-177. Bruce HM. An exteroceptive block to pregnancy in the mouse. Nature 184: 105-108, 1959. Bruce HM. Block to pregnancy in the mouse caused by proximity of strange males. J Reprod Fert 1960; 1: 96-103. Bruce HM. Time relations in the pregnancy-block induced in mice by strange males. J Reprod Fert 1961; 2: 138-142. Parkes AS, Bruce HM. Pregnancy block in mice placed in boxes soiled by males. J Reprod Fert 1962; 4: 303-308. Dominic CJ. Observations on the reproductive pheromones of mice. I. Source. J Reprod Fert 1966a; 11 : 407-414. Dominic CJ. Observations on the reproductive pheromones of mice. II. Neuro-endocrine mechanisms involved in the olfactory block to pregnancy. J Reprod Fert 1966b; 11: 415-421. Yamazaki K., Beauchamp GK, Wysocki CJ, Bard J, Thomas L, Boyse EA. Recognition of H-2 types in relation to the blocking of pregnancy in mice. Science 1983; 221: 186-188. Krackow S. Sex ration manipulation in wild house mice: the effect of fetal resorption in relation to the mode of reproduction. Biol Reprod 1992, 47: 541-548. Yamazaki K, Beauchamp GK, Bard J, Boyse EA. Sex-chromosomal odor types influence the maintenance of early pregnancy in mice. Proc Natl Acad Sci (USA) 1989; 86: 9399-9401. Bruce HM, Parrott DVM. Role of olfactory sense in pregnancy block induced by strange males. Science 1960; 131: 1526-1527. Dominic CJ. The origin of the pheromones causing pregnancy block in mice. J Reprod Fert 1965; 10: 469-472. Dominic CJ. Reserpine: inhibition of olfactory blockage of pregnancy in mice. Science 1966; 152: 1764-1765. Huck UW, Lisk RD, Miller KS, Bethel A. Progesterone levels and socially-induced
329
52.
53.
54.
55.
56.
57.
58. 59.
60. 61.
62.
implantation failure and fetal resorption in golden hamsters (Mesocricetus auratus). Physiology & Behavior 1988; 44: 321-326. Rajendren G, Dominic CJ. e male-induced pregnancy block (the Bruce effect) in mice: re-evaluation of the ability of exogenous progesterone in preventing implantation failure. Exp Clin Endocrinol 1987; 89:188-196. Rajendren G, Dominic CJ. The male-induced implantation failure (the Bruce effect) in mice: effect of exogenous progesterone on maintenance of pregnancy in male-exposed females, p. Clin Endocrinol 1993; 101: 356-359. Baines MG, Gendron RL. Natural and experimental animal models of reproductive failure. In: Immunology of Pregnancy, edited by G. Chaouat. Boca Raton: CRC Press, Inc 1993, p. 173-203. Baines MG, Haddad EK, Pomerantz DK, Duclos AJ. Effects of sensory stimuli on the incidence of fetal resorption in a murine model of spontaneous abortion: the presence of an alien male and postimplantation embryo survival. J Reprod Fert 1994; 102: 221-228. Wobst B, Zavazava N, Luszyk D, Lange K, Ussat S, Eggert F, Ferstl R, Muller-Ruchholtz W. Molecular forms of soluble HLA in body fluids: potential determinants of body odor cues. Genetica 1998; 104: 275-283. Gilbert AN, Yamazaki K, Beauchamp GK, Thomas L. Olfactory discrimination of mouse strains (Mus musculus) and major histocompatibility types by humans (Homo sapiens). Journal of Comparative Psychology 1986; 100: 262-265. Lord T, Kasprzak M. Identification of self through olfaction. PerceptuaL & Motor Skills 1989; 69:219-224. Fleming AS, Corter C, Surbey M, Franks P, Steiner M. Postpartum factors related to mother's recognition of newborn infant odours. Journal of Reproductive & Infant. Psychology 1995; 13: 197-210. Ober C, Weitkamp LR, Cox N, Dytch H, Kostyu D, Elias S. HLA and mate choice in humans [see comments]. American. Journal of Human. Genetics 1997; 61: 497-504. Wedekind C. and S. Furi. Body odour preferences in men and women: do they aim for specific MHC combinations or simply heterozygosity? Proc R Soc Lond 1997; 264: 1471-1479. Ober C, Elias S, Kostyu DD, Hauck WW. Decreased Fecundability in Hutterite Couples Sharing HLA-DR. Am J Hum Gen 1992; 50: 6-14.
This Page Intentionally Left Blank
New Foundation of Biology Edited by I. Berczi and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
331
Natural Immune Regulation of Activated Cells
DONNA A. CHOW, RICKY KRAUT and XIAOWEI WANG
Department of Immunology, University of Manitoba, Winnipeg, MB., Canada R3E-OW3
ABSTRACT The natural immune system provides a prompt first-line of defense against invading pathogens. It is comprised of both cellular and humoral mediators including macrophages, NK and T cells, natural antibodies, complement and interferon. Unlike the exquisitely specific, adaptive immune system, natural immunity is polyspecific, recognizing highly conserved homologous or crossreactive epitopes (homotopes), and does not require previous exposure to antigen to generate a response. This evolutionarily conserved, innate system is also considered important in homeostatic regulation of the organism through a highly connected receptor network and through the removal of aberrant cells, including nascent tumor cells. It provides the immunological environment for the development of the adaptive T cell response. Furthermore, activation of the acute phase response (APR) through immune derived cytokines when adaptive immunity has failed to eradicate an invader, provides a massive stimulus to the natural immune system including the polyspecific acute phase mediators in an all-out final natural defense. Several in vivo and in vitro syngeneic murine tumor models of natural resistance have provided evidence to support polyclonal natural antibody (NAb), NK cell and activated macrophage surveillance of developing tumors. A study of NAb, those antibodies in the circulation of normal individuals that are not intentionally immunized, revealed a consistent inverse correlation between NAb binding by tumors and tumor incidence in a threshold s.c. tumor inoculum assay of natural resistance. Considering that LPS induces APR, a role for neuroendocrine regulation of NAb surveillance was suggested by correlating LPS-induced increases in NAb levels and reductions in tumorigenicity of threshold s.c. inocula of an NK-resistant tumor. Furthermore, a strong influence of sex hormones on the natural resistance was evident in the reduced tumor incidence of threshold s.c. tumor inocula in female versus male mice. T lymphomas treated with the tumor promoter 12-O-tetradecanoylphorbol-13-acetate (TPA) to generate variants and repeatedly selected for high serum NAb IgG- plus IgM-binding, exhibited a reduced s.c. tumorigenicity. Surface expression analysis also revealed that the selected cells bound more monoclonal antibodies against activationassociated molecules including CD25, the IL2-Ret chain, and asialo forms of CD45RA, a transient marker of T cell activation. This argues that NAb also regulates T cell activation. The observations in the murine models raised the possibility that direct down-regulation of T cell activation may contribute to the beneficial effects of intravenous immunoglobulin (IVIg) against inflammatory and autoimmune diseases. In order to identify the T cell surface targets of IVIg in humans, we produced a model human T cell through TPA treatment of Jurkat T
332
leukemia cells with repeated selection for high serum IgG- plus IgM-binding which mimics the conditions for IVIg action in patients. Cells produced from the 3rd and 4th sequential selections exhibited stable increases in serum IgG- and IgM-binding of 28% for J3.1 and 45% for J4.1. The J4.1 bound at least 45% more IVIg than the parental cells consistent with their increased binding of pooled human IgG-plus IgM. The J4.1 exhibited a 120% increase in expression of CD45RA, a marker which correlates with cytokine and hormone expression in women. The results provide the first evidence that human serum natural IgG-plus IgM antibodies and IVIg react with a CD45RA epitope. The conserved nature of many NAb and the contribution of endogenous NAb and passively administered IVIg to the control of immune activation within the context of neuroendocrine regulation, raise the possibility that the epitope on CD45RA may be a highly conserved homologous epitope or homotope of the neuroimmune system involved in health and disease.
1.
NATURAL IMMUNITY
1.1.
Function
In the last decade, the increased focus of research on the natural or innate immune system has expanded our knowledge regarding the plethora of mediators which contribute to this system, their characteristics, their differences compared with those of the adaptive immune response and the relationship of the natural to the adaptive response. Questions regarding the function of the natural immune system are being revisited again. It has chiefly been considered to be a first line of defense against microbial invaders. However, other roles are seen in the removal of dead and dying cells (and thus in morphological development); in the regulation of the immune system and as a platform for the development of the adaptive T cell response; in the defense against tumor development; and, finally, in a last line of defense against invading pathogens when all else has failed [1-4]. 1.2.
Mediators
Macrophages and neutrophils employ phagocytosis, an ancient mechanism for removing effete cells and invading pathogens, and for providing nutrients. Natural killer (NK) cells, the prototypical killer lymphocyte of the natural immune system differ from the adaptive response cytotoxic T cells in their lack of MHC restriction, and in their requirement for "missing self" (absence of MHC) for killing target cells. Natural antibody, those immunoglobulins found in individuals which are not intentionally immunized, do not require an intact thymus and some originate from CD5 + B cells. Mast cells, basophils and eosinophils also contribute to natural immunity. ~,6T cells contribute particularly to the natural defense of surface epithelia and T lymphocytes in general provide a non-specific defense when they are activated polyclonally by a superantigen. Humoral mediators include the components of the complement cascade and the acute phase response proteins [1-4]. 1.3.
Specificity
The natural or innate immune response is characterized by the target antigens which it recognizes. In contrast with the peptide/MHC complexes of the adaptive response these are more broadly occurring epitopes. Janeway has referred to the invariant molecular structures in pathogens
333
which are distinct from self and would be the targets of the innate immune system as pathogenassociated molecular patterns (PAMPs) [1]. Their non-clonal receptors which have a broad specificity and recognize a range of different ligands are then pattern-recognition receptors (PRRs). These receptors are germline encoded, their specificities having been selected during evolution based on survival against pathogens. This is somewhat similar to Berczi's concept of homotopes and their receptors [2]. However, the highly conserved homologous or crossreactive epitopes (homotopes) may also be expressed on self tissue [3] suggesting an additional contribution from self-recognition in the evolutionary development and conservation of the reactivity. Recent reports suggest that CD14, the cellular receptor for LPS, the prototypical homotope or pathogen-associated pattern expressing molecule, also binds apoptotic cells [4]. As summarized in Table I, LPS binding to LPS-binding protein (LBP) and subsequently CD14 leads to macrophage activation and proinflammatory cytokine production. This is thought to occur through signaling via the Toll-like receptor Tolr4. However, binding of apoptotic cells to CD 14 does not lead to the release of proinflammatory cytokines from macrophages [4] and thus may not activate the Toll pathway. Nevertheless, binding of apoptotic cells by a region of CD14 which is identical to or overlapping with the LPS binding site, extends the binding of PRRs to self (apoptotic self) approaching closer to the homotope concept. Furthermore, a distinction in the macrophage responses of CD 14 to non-self and self raises a question regarding the ability of CD14-self interactions to activate the Toll pathway. In Drosophila, the local antimicrobial peptide response is distinguished from the classically described systemic response to wounding by its independence from the Toll pathway [5]. Considering the relationships which have been suggested between the insect systemic immune response and the acute phase response in mammals, and between the insect local immune response of epithelial cells and vertebrate mucosal immunity, this raises a question regarding the evolutionary conservation of a distinction in signaling between the response to self and non-self. Table I
Characteristics of LPS-binding receptors.
LPS/homotope Receptor
Ligand/Homotope
Outcome
LBP
LPS, phospholipids
binding to CD 14
CD14
LPS lipid A, & numerous microbial products
macrophage activation & proinflammatory cytokines (binding to Tolr)
apoptotic cells
no inflamation
LPS
macrophage activation and proinflammatory cytokines
Toll-like receptor 4 (Yolr4)
2.
NEUROENDOCRINE REGULATION OF NATURAL IMMUNITY
The importance of the neuroendocrine system to natural immune function is evident in the extensive regulation of natural immune mediators by neuroendocrine factors. These include,
334
growth and lactogenic hormones (GLH), hormones of the hypothalamus-pituitary-adrenal (HPA) axis, steroid hormones and other mediators [3]. Factors which regulate the proliferation of lymphocytes of the adaptive response include; growth hormone (GH) and prolactin (PRL) for immunocompetence, specific receptor recognition of MHC/peptide complexes and adhesion molecules for cell signaling, and cytokines as soluble mediators for amplification of proliferation and for differentiation [6]. While regulation of natural immune lymphocytes likely depends similarly on pituitary growth hormone and prolactin plus cytokines, recognition of homotopes or molecular patterns on pathogens or self is necessary for signaling. Phagocytosis is promoted by PRL and GH. CD5 + B cells and/or NAb production is promoted by the GLH hormone, thyroxin (T3), HPA axis hormones glucocorticoid (GC) and ot-melanocyte-stimulating hormone (c~-MSH) and steroid hormones estradiol (E2) and dehydroepiandrosterone (DHEA) [1]. Innervation of lymphoid organs and the often anatomically close contact between nerves and mast cells ensures regulation of the natural immune response by neurotransmitters and neuropeptides. The inflammatory response is highly regulated by hormones, neurotransmitters and neuropeptides, being promoted by PRL and GH and diminished by HPA hormones acting on leukocytes and the endocrine and nervous systems [3]. In the response to LPS, macrophages are activated producing tumor necrosis factor (TNF), interleukin-1 (IL-1), interleukin-6 (IL-6) and interleukin-8, with the latter two also released from epithelial cells. Depending on the amount of cytokines released, the inflammatory response induced will be local at low levels or systemic at moderate levels. The local response involves leukocytes and epithelial cells, while the systemic inflammatory response also involves the brain, liver and bone marrow. TNF acts on the hypothalmic region of the brain to induce fever. TNF and/or IL-1 in conjunction with IL-6 act on the liver to produce the acute phase proteins (APP). TNF suppresses stem cell division in the bone marrow. Together, this response generated by severe wounding or infection and characterized by fever, is termed the acute phase response. The substantial regulation of the APR by the neuroendocrine system is evident beginning wih the early increases in PRL and GH which support the rapid increase in polyspecific natural immune mediators. Catecholamines and glucocorticoids dramatically amplify APE Subsequently, cytokine production and the inflammatory response are suppressed by the elevation of the HPA axis hormones [3].
3.
NATURAL ANTIBODY FUNCTION AND SPECIFICITY
Polyclonal NAb is present in the circulation of normal, not intentionally immunized individuals of a wide range of vertebrate species. Evidence suggests the importance of NAb in the defense against bacterial and parasitic infections, autoregulation of the immune system, homeostasis of the organism [7] and in the early defense against tumors [8]. The multispecificity [9] and synergistic nature of NAb was shown by the binding of individual monoclonal NAb to several autoantigens, in particular highly conserved cytoskeletal molecules, dsDNA, denatured DNA, cytokines and immunoglobulins, the last implicating NAb in a highly connected anti-idiotypic network [ 10]. Anti-carbohydrate reactivity recognized blood type antigens and tumor cells [11]. One percent of human IgG exhibits anti-c~-galactosyl reactivity against non-human mammalian cells and human trophoblast, senescent red blood and mammary carcinoma cells [12]. IVIg, a source of immune and NAb reacts with cell surface molecules including CD4, CD5, the T cell receptor [13] and Fas antigen [14]. Thus there is extensive evidence of the autoreactivity of NAb but no unifying concept regarding NAb reactivity.
335
.
NAb C O N T R O L S GROWTH OF N E O P L A S T I C AND P R E N E O P L A S T I C CELLS I N VIVO
In vivo and in vitro syngeneic murine tumor models have provided evidence from a variety of approaches to support a role for polyclonal NAb, NK cells and activated macrophages in the surveillance of small tumor foci [ 15, 16]. While some tumors were resistant to NK cytolysis, every tumor tested bound syngeneic NAb. A consistent inverse correlation was observed between tumor binding of NAb and the tumor incidence following a threshold s.c. tumor challenge. Our early reports described how LPS, a polyclonal B cell activator in mice raised anti-tumor NAb levels assayed as complement-mediated cytotoxicity, and concordantly reduced the tumorigenicity of a tumor challenge, including that of the NK-resistant L5178Y-F9 [16]. This data and observations from numerous other approaches including the passive transfer of serum NAb from untreated normal mice which reduced the incidence of a tumor challenge, in B-cell, IgM and IgG 3 deficient xid-bearing mice, demonstrated the contribution of NAb to tumor surveillance. However, LPS also initiated the acute phase response as was evident in the ruffled fur and hunched appearance of the LPS-treated mice. This argues for t h e beneficial contribution to tumor resistance of acute phase proteins including, C-reactive protein and mannose binding protein, and activation of the neuroendocrine system with increases in HPA axis hormones GC and c~-MSH and possibly the sex hormone E2, all of which stimulate B cells. Additionally, many other natural immune cells are activated by LPS and likely participate in the tumor elimination. Sex hormones strongly influenced natural immune resistance against a threshold s.c. tumor challenge. This was evident in the reduced tumor incidence in age matched female versus male C3H/HeN mice after inoculation of NK- and NAb-sensitive variants of the I3T2.1, v-Ha-ras transformant of the syngeneic 10T 89fibroblast line (Table II).
Table II
Syngeneictumor incidence at day 30 in male versus female C3H/HeN micea.
Tumorb I3T2.1 I3T2.1SVX I3T2.1SVX/MYC
Male
Female
2/3 6/9 8/9
0/3 1/9 0/9
a Threshold tumor inocula of 2.5 X 103 cells were injected into a shaved area on the back of each syngeneic C3H/HeN mouse. b I3T2.1 is a clone of ras-transformed C3H 10T 89fibroblasts. I3T2.1SVX and I3T2.1SVX/MYC are populations of cells obtained after incubation of I3T2.1 with the G418-resistance converting SVX plasmid and the same plasmid bearing v-myc respectively.
Further evidence of NAb contributing to tumor surveillance was generated by a variety of approaches. In an in vitro model aimed at reversing progression and generating tumors with increased susceptibility uniquely to NAb, T-lymphomas were treated with the tumor promoter TPA to generate variants and were then selected for high NAb binding through fluorescenceactivated cell-sorting (FACS). Three cycles of TPA treatment and sorting yielded the high NAb binding L5178Y-F9 TPA/NAb+3 (LYNAb+), which exhibited a reduced s.c. tumorigenicity [17]. In addition, ras transformation of 10T 89fibroblasts which increases their protein kinase C (PKC)
336
activity, increased NAb binding [18, 19] with an inverse correlation between NAb binding and tumorigenicity [20], providing evidence for an NAb-susceptible phase of oncogene-induced tumor development. In a model employing non-tumorigenic cells, preneoplastic, rat PKC-[51 overexpressing 10T 89cells also exhibited higher NAb binding and a more rapid elimination in vivo of 131I-dUrd-labeled cells 21. Thus, constitutive increases in the basal activity of PKC seen in both the ras-transformation and PKC preneoplasia models argues for PKC upregulation of NAb binding structures and for NAb recognition of incipient neoplasia. In a direct assay preincubation of tumor cells in serum NAb versus serum NAb depleted of tumor-binding antibodies, reduced their tumorigenicity [22]. Furthermore, i.v. pretreatment of mice with syngeneic serum NAb, ammonium sulfate precipitated fractions of NAb, or human IVIg reduced the incidence and/or increased the latency of syngeneic tumors injected s.c. [19, 23] and reduced the metastasis of tumors injected i.v. [24].
5.
IMPACT OF NAb ON TUMOR CELL GROWTH IN VITRO
Growth in syngeneic NAb produced a concentration-dependent, heat-inactivation independent reduction in the cloning efficiency of the L5178Y-F9 in sloppy agar and 0.25 mg/ml IgG-plus 0.1-0.25 mg/ml IgM reduced the growth of ras-transformed 10T 89cells with a higher proportion in G0/G1 after 2 days [25].
6.
NAb BINDING TO APOPTOTIC CELLS
Since NAb has been implicated in the immune surveillance of small tumor inocula, we examined whether expression of the apoptosis-inducing antioncogene, p53, might influence the regulation of the NAb binding essential for this defense against tumor development. Erythroleukemia cells transfected with the temperature sensitive p53 TM ~35 gene suggested that increased wildtype p53 expression led to increased NAb binding. Flow cytometry using a viable Hoechst stain and propidium iodide DNA probes showed that cells judged to be apoptotic bound more NAb suggesting that NAb binding to apoptotic cells contributes to their surveillance. Furthermore, the NAb treated samples exhibited smaller percentages of apoptotic cells and/or fragments suggesting that NAb contributes to the clearance of apoptotic cells (Jensen P, Chow DA. In preparation). Increased binding of NAb to apoptotic cells might be related to NAb which is specific for neoantigens expressed during apoptosis.
7.
INTRACELLULAR SIGNALING INDUCED BY NAb
Cell surface binding by NAb argues for a direct role in controlling cell function possibly through signaling. An instability in syngeneic NAb binding to C3H 10T 89fibroblast variants was observed at 37~ and this was partially reduced by H7, an inhibitor of the pivotal signaling serine/threonine kinase PKC [25]. Pre-coating cells with purified NAb at 4~ to maximize NAb binding, followed by a rise to physiological temperature, 37~ produced an increase in membrane expression of introduced rat PKC-[31 and endogenous PKC-ot, in the PKC-131-overexpressing PKC-4 and v-H-ras- producing I3T2.1 respectively. A concurrent marked reduction was seen in tyrosine phosphorylation of membrane-associated 60-KDa protein including the tyrosine kinase src. In addition, both the pre-coated NAb and numerous membrane
337
molecules with molecular weights from 20 to 220 KDa were released into the supernatant, including the receptor-like protein tyrosine phosphatase (RPTPc~). Purified NAb added to cells in culture reduced the growth of I3T2.1 cells assessed as a decrease in total cell numbers and an increase in the proportion of cells in the G0/G1 phase of the cell cycle. Taken together, these results support the idea that the NAb interaction with cell surface structures initiates a series of intracellular signaling events leading to the release of membrane molecules and over time the suppression of cell proliferation [25]. This process suggests a biological mechanism for the direct NAb control of activated cells in both physiological and pathological conditions.
8.
TUMOR CELL TARGETS OF NAb BINDING
The flow cytometry assessment of polyclonal NAb binding to tumors has provided an invaluable tool for investigating the role of NAb against tumors. Every approach taken has consistently revealed an inverse relationship between tumorigenicity and the tumor's ability to bind polyclonal NAb. 8.1.
Saccharide specificity
Inhibitors of N-linked and O-linked oligosaccharide production and glycolipid synthesis, monosaccharide inhibition, neuraminidase and pronase treatments suggested that NAb binds asialo forms of cell surface N-linked and O-linked oligosaccharides and glycolipids on L5178Y-F9 cells [26](Chow & Reese, in preparation). 8.2.
Signal regulating molecules
Since NAb may regulate T-lymphoma growth, it was important to assess NAb binding to cell surface molecules known to participate in T-cell activation, proliferation, differentiation and development or in the downregulation of these activities. Anti-transferrin receptor and anti-Thy 1.2 monoclonal antibody (MAb) reactivity did not correlate with NAb binding [27]. Our panel of L5178Y- F9 variants demonstrated a correlation between NAb binding and expression of the interleukin-2 receptor-c~ (IL-2R-c~) (p55) (7D4) and CD45RA (RA3-2C2). 8.3.
CD45
CD45RA, CD45RB and CD45RC are determinants on high molecular weight species of the leukocyte common antigen and are restricted to isoforms expressing exons 4, 5 and 6 respectively, of the CD45 variable expression region. CD45 exhibits protein phosphotyrosine phosphatase activity and functions in cell-cell interactions and intracellular signaling [28] by modifying the IL-2/IL-2R pathway [29]. Expression of CD45 determinants distinguishes discrete functional populations of T cells with most neonatal T cells exhibiting CD45RA and no CD45RO and the transition from adult naive unprimed T cells to memory T cells characterized by a switch in surface expression of CD45RA isotype to CD45RO. Anti-CD45 monoclonal antibodies can alter T-cell activation either in a positive or negative way depending upon the antibody employed or the time of antibody addition during the proliferative response. High CD45RA and CD45RC expression and low tumorigenicity accompanied FACS selection of the LYNAb + for high NAb binding, and outgrowth of the LYNAb + in vivo reversed each parameter [30]. Neuraminidase treatment increased binding by NAb, anti-CD45RA and
338
anti-CD45RC. Thus, NAb probably binds preferentially, asialo isoforms of CD45 expressing exons A and C. Anti-CD45 immunoprecipitation of lysates detected corresponding changes in CD45 isoform expression with more variable exon expressing, higher molecular weight products of mRNA splicing on the LYNAb +. Thus, asialo high molecular weight isoforms of the CD45 were selected via NAb binding and were reduced during tumor development suggesting they participate in NAb mediated anti-tumor mechanisms. Results from several experimental approaches suggest that NAb binds CD45 directly (Zhang Z, Ostergaard H, Chow DA. In preparation). NAb binding to the A strain YAC T-lymphoma correlated with the presence of CD45 in the wild type, CD45- and a CD45RABC-transfectant. More specifically, in an ELISA assay NAb eluted from LYNAb + cells incubated in normal syngeneic serum, demonstrated significant direct binding to purified CD45 containing the products of the variably-expressed exons. IgG NAb partially reduced the binding of antiCD45RA and anti-CD45RB relative to anti-pan CD45. Purified CD45 containing the variably expressed exon products inhibited NAb binding. Others have generated a natural hybridoma autoantibody to CD45 through the fusion of unstimulated neonatal Balb/c spleen cells [29]. Collectively, these results support CD45, particularly asialo high molecular weight forms, as one of the cell surface molecules that NAb binds. Furthermore, they implicate NAb in the direct regulation of CD45 function on lymphoid tumor and normal cells [31]. 8.4.
IL-2 Receptor
Interleukin-2 (IL-2) is the major growth factor for T cells and also modulates the activation, proliferation and differentiation of other immune system cells including monocytes, B cells and NK cells. Flow cytometry analysis of the binding by anti-IL-2 receptor (IL-2R) 7D4 (p55, chain) mAb, showed a consistent correlation with NAb binding for our panel of L5178Y-F9 variants [27]. An L5178Y-F9 variant FACS-selected for high anti-IL-2R binding (TPA/-IL-2R+2) bound 30% more NAb, implicating NAb in the regulation of the IL-2/IL-2R pathway and in signal transduction. The variant selected for high anti-IL-2R mAb binding did not exhibit an increase in anti-CD45RA binding, distinguishing CD45RA expression from IL-2R expression. Together, these results suggest that NAb binds the IL-2R, probably asialo forms, independent of binding to CD45RA which is also increased on cells selected for high NAb binding. The IL-2R chain assayed here is the (55 KD) ct glycoprotein (CD25) which is expressed upon T cell activation and is required to form the high affinity IL-2R with the [3 (75 KD) (CD122) and 5, (64 KD) chains [32]. Monoclonal anti-IL-2R antibodies have been shown to inhibit IL-2 binding of the IL-2R and to abolish T cell proliferation in vitro and delayed type hypersensitivity in vivo [33].
9.
NAb BINDING TO ACTIVATED CELLS
In our ras and PKC-[31 models, upregulation of NAb binding structures with the elevation of a dominant signaling activity, PKC, argued for NAb binding to activated cells in general. Natural monoclonal IgM bound to phytohemagglutinin-stimulated normal human CD3 + T cells [34] and normal serum IgG bound to autologous phytohemagglutinin-activated human T lymphocytes [35]. The identification of a transitional, higher density CD45RA + naive adult T cell phenotype 1 to 2 days following mitogenic stimulation of T cells [36] coincides with CD45RA expression on the most highly proliferating cells [37]. This suggests that highly proliferating cells which appear briefly, early after activation might be exquisitely sensitive to NAb regulation.
339
This is consistent with the origin of the LYNAb + via selection for high NAb binding after TPA treatment which rapidly increases CD45RA expression in human T cells [38]. Increased neuraminidase activity 24-48 hours after T-cell activation should ensure that the NAb-reactive asialo forms of CD45 are expressed. Similarly, IL-2R-c~ is expressed on activated T cells and IL-2R[3 [32], is increased on activated T cells. Thus, natural immunoglobulin likely exerts its function by binding high asialo-carbohydrate expressing molecules including CD45 and the IL-2R on T cells in an early transition phase of cellular activation. These observations open a new phase of research toward an understanding of NAb and IVIg function from all perspectives including biochemical, physiological, medical and pharmaceutical.
10.
IVIG THERAPY FOR INFLAMMATION AND AUTOIMMUNE DISEASE
10.1. Mechanisms of Immunoglobulin Action Pooled immunoglobulin from normal humans prepared for intravenous injection is used in the treatment of infectious, inflammatory, and autoimmune diseases. IVIg provides a benefit against Kawasaki Syndrome, idiopathic thrombocytopenic purpura (ITP) and Guillain-Barr6 syndrome (GBS), which are all characterized by increased helper T cells activation and cytokine production. Proposed mechanisms of IVIg action [39, 40] include antigen-specific antibodies which neutralize the causative microbe or toxin and thus focus on the initiating agent. Mechanisms which focus on modulating the immune system include, Fc receptor (FcR) blockade and modulation, anti-idiotypic antibody inhibition of autoreactive antibodies and restoration of the idiotypic network, modulation of cytokine production, complement uptake by target cells, regulation of T-cell function and induction of T-cell suppressor activity. Accelerated catabolism of IgG not protected by the FcRn IgG transport receptor has recently been added to the list. However, the mode of IVIg reduction of immune activation is not clear and probably involves more than one mechanism. Furthermore, high doses of IVIg, 0.5-2 gm/Kg are currently required to achieve a benefit.
11.
RATIONALE FOR IVIG REGULATION OF ACTIVATED CELLS
Activation of T cells contributes to inflammation and has been implicated in many human and animal autoimmune diseases [39, 40]. Thus, IVIg regulation of T cells in early stages of activation may provide a rationale for some of the reported benefits of IVIg by reducing the recruitment of cells into the response or preventing their full expression of the activated phenotype including the production and release of cytokines. This is consistent with: i) the demonstration of activated T cells in diseases including Kawasaki's, ITP and GBS for which IVIg provides a therapeutic benefit, ii) the observation of high CD45RA and high IL-2R in addition to high HLA class II expression on these cells and, iii) decreases in the activated cells after IVIg treatment. The identification of molecules contributing to signal transduction, the IL-2 receptor and CD45 as IVIg targets and characterization of IVIg-sensitive cells as early activated cells will suggest mechanisms of IVIg action. Moreover, it will provide a basis for:
340
1) The development of adjunct treatments to increase cell surface expression of IVIg targets and, 2) The purification of an anti-'activated' T cell fraction of IVIg. Such innovative approaches would increase IVIg target sensitivity and specific IVIg activity, expanding the applications for IVIg in the control of pathological immune processes and reducing the IVIg protein dosage required and undesirable side effects.
12.
GENERATION AND SELECTION OF HIGH IVIG-BINDING CELLS
In order to identify the cell surface targets of intravenous immunoglobulin on human T cells, a model human T cell line which exhibited an increased binding of human immunoglobulin relative to the parental cell was generated. In order to mimic the conditions for IVIg action in patients, we selected first for high IgG and IgM binding in the physiological context of human serum. Different individual human sera (recalcified plasma) exhibited reproducibly different abilities to bind human Jurkat T lymphoma cells, possibly related to different antiHLA reactivity. Pooled serum from normal donors yielded an intermediate binding capacity. Following the success in generating well-selected and stable variants from cells treated with TPA in our previous murine models, the selection process was carried out using TPA treated cells and pools of serum from healthy donors. Jurkat cells treated with 100 ng/ml TPA for at least one week in culture to generate variants were incubated in pooled serum and selected via magnetic separation of the highest binders following incubation with goat anti-human IgG and IgM-coupled BioMag magnetic beads. The 1st and 2nd selections yielded the top 18% binders using sera pooled from 4 normal donors. Cells produced from the 3rd and 4th sequential selections of the 4% highest binders using sera pooled from 13 normal donors exhibited a 28% (J3.1) and 45% (J4.1) increase in human serum IgG- plus IgM binding respectively compared with the parental Jurkat line (Table III). The increase in human serum IgG- plus IgM binding of the J4.1 remains stable for cells in culture for over 4 months. Our initial assessment of IVIg binding showed that the human serum IgG- plus IgM-selected J4.1 bound at least 45% more IVIg than the parental cells assayed near saturating conditions of 15 mg/ml and considerably more, on average a 123% increase at 10 mg/ml (Figure 1, Table III). These observations are consistent with the increase in pooled human IgG- plus IgM-binding for the J4.1. This suggests that the selection based on high serum IgG- plus IgM-binding yielded a high IVIg binding phenotype which will be particularly useful for revealing the T cell surface targets which IVIg can bind in the context of human serum. In addition, it will be important to carry out a titration for all IVIg assessments as the concentration of the IVIg appears to have a great effect on its differential binding ability. A suboptimal concentration of IVIg, 5 mg/ml, chosen to minimize the impact of the IVIg on the cell provides a viable selected cell population for cell growth in tissue culture. Under UV microscopy Jurkat cells treated with 5 mg/ml IVIg followed by FITC-coupled antihuman IgG antibody all at 4~ showed a speckled fluorescence pattern across the cell surface with a faint ring around the edge of the cell.
13.
HIGH IVIG-BINDING VARIANT EXPRESSION OF CD45RA
A brief assessment of the J3.1 population showed a marked increase in binding of anti-CD45RA (HI100) (Table IV). Consistent with the observations using the J3.1, our analysis of the J4.1
341
IVIG Binding by Jurkat and Jurkat 4.1 Cells (6000) CELLS PliT2
(SO00) CELLS PIIT2
m m
~D
Jurkat
I
Jurkat 4.1
0
Z:
" 0
t
t
t
i
i
t
t
I
I
1023
FITC
HIe
F I T C Fluorescence using IVIG at 15 mg/mi Figure 1. IVIg binding by Jurkat and J4.1 cells assayed as liner fluorescence after sequential incubation in IVIg followed by FITC-coupled antihuman IgG, all at 4~
Table III Expt #
Binding of serum Igs and IVIg to Jurkat and human serum IgG- plus IgM-selected Jurkat cells. Cells
Mean Channel Fluorescence + SE Serum Ig
1
2
Jurkat
185.1
J3.1
236.8 _+ 37.2
% Increase
IVIg 15 mg/ml
_+ 3 1 . 5 a
--
28
Jurkat
154.6 _+ 14.0 b
J4.1
222.8 + 152
% Increase
IVIg 10 mg/ml % Increase
_
-308.5 _+ 53.5 c
45
446.1 -+ 42.0
70.1 _+ 16.2 d 45
156.6 + 16.2
123
a Eight experiments were performed at a 89or 1/3 serum dilution. b Six experiments were performed at a 89serum dilution using pooled serum from 13 normal donors. c Six experiments were performed with Ptd < 0.02. d Three experiments were performed.
Table IV
Binding of anti-CD45RA mAbs to Jurkat and human serum IgG- plus IgM-selected Jurkat cells.
Mean M C F + SE b MAbs a
Specificity
Jurkat
J3.1
% Increase
Jurkat
J4.1
% Increase
HI100
CD45RA
98.5+5.8
316.0_+39.8
220 (3)
163.8_+19.6
335.8_+40.7
120 (8)
(IgGzb)
isotype control
--
--
24.5_+3.9
29.4_+4.7
a
MAb was used at 1 ug/ml with J3.1 and at saturating concentrations, 1-7.5 ug/ml, with J4.1.
b The number of experiments performed is shown in parentheses.
342
has shown a marked increase in expression of CD45RA relative to the parental Jurkat line amounting to a 120% increase. The data are consistent with our previous observations using syngeneic murine systems in that selection of a human T lymphoma for high human serum NAb IgG-plus IgM binding yielded high CD45RA expressing cells arguing that human serum NAb, similar to murine NAb react with CD45RA a marker of cells undergoing activation [30, 31]. Furthermore, the correspondence of high IVIg binding with this selection provides the first evidence that CD45RA is also recognized by IVIg and may provide a target for IVIg regulation of cells in an early stage of activation. Several pieces of evidence suggest that IVIg may control activated T cells via CD45-initiated apoptosis. 1) CD45 is abundant on T cells with some 105 CD45 molecules covering about 23% of the total cell surface area [42]. 2) Purified high molecular weight isoforms of CD45 inhibited the binding of purified IgG NAb binding to a relatively high extent, 25-30%, in our murine system. 3) Cross-linking of CD45 was recently shown to lead to apoptosis of thymocytes along a pathway distinct from the typical Fas-dependent pathway [43]. 4) Others have observed that galectin-1 which induces apoptosis of activated human T cells and thymocytes by an unknown mechanism binds to a restricted set of T cell surface molecules and that only CD45, CD43 and CD7 appear to participate directly in galectin-1 induced apoptosis [44]. 5) The known anti-carbohydrate activity of many natural antibodies further supports the idea that IVIg, like galectin-1, may be capable of inducing apoptosis via a pathway which involves CD45 and is distinct from the Fas-mediated pathway. This provides important information for a future examination of direct mechanisms of IVIg cell regulation. A recent report on galectin-l-induced apoptosis revealed that after galectin-1 binding, CD45 and CD3 were colocalized with externalized phosphatidyl serine on large islands on apoptotic blebs protruding from the cell surface [44]. In contrast CD7 and CD43 were colocalized on small patches away from the membrane blebs. The receptor segregation was not seen on cells that did not die, arguing that the spatial redistribution of cell surface molecules into specific microdomains was required for apoptosis induction.
14.
CONCLUSIONS
These results provide the first evidence that human serum natural IgG-plus IgM antibodies and IVIg react with a CD45RA epitope. They confirm our evidence from the syngeneic murine system and combined with the reported transient high expression of CD45RA on human T cells soon after activation, they support the idea that T cells early after activation express a serum IgG- plus IgM, and- IVIg-sensitive phenotype. The conserved nature of many NAb and the contribution of endogenous NAb and passively administered IVIg to the control of immune activation within the context of neuroendocrine regulation, raise the possibility that the epitope on CD45RA may be a highly conserved homologous epitope or homotope of the neuroimmune system involved in health and disease.
343
ACKNOWLEDGMENTS We wish to thank Dr. Ed Rector for operating the Beckman Coulter Epics Altra Fluoresence Activated Cell Sorter. This work was supported by the Manitoba Medical Services Foundation and The Bayer, Canadian Blood Services, H6ma Qu6bec Partnership Fund.
REFERENCES 1. 2. 3. 4. 5.
6. 7.
8. 9. 10. 11. 12. 13.
14.
15.
16. 17.
Medzhitov R, Janeway Jr C. Innate immunity: impact on the adaptive immune response. Current Opinion in Immunology 1997; 9: 4-9. Berczi I. Neurohormonal host defense in endotoxin shock. In: Annals New York Acad Sci 1998; 840: 787-801. Berczi I, Bertok L, Chow DA. Natural immunity and neuroimmune host defense. Acad Press NY In press. Devitt A, Moffatt OD, Raykundalia C, Capra JD, Simmons DL, Gregory CD. Human CD 14 mediates recognition and phagocytosis of apoptotic cells. Nature 1998; 392: 505-509. Ferrandon D, Jung AC, Criqui M-C, Lemaitre B, Uttenweiler-Joseph S, Michaut L, Reichhart J-M, Hoffmann JA. A drosomycin-GFP reporter transgene reveals a local immune response in Drosophila that is not dependent on the Toll pathway. EMBO J 1998; 17: 1217- 1227. Berczi I, Chow DA, Sabbadini ER. Neuroimmunoregulation and natural immunity. Domestic Animal Endocrinology 1998; 15: 273-281. Avrameas S. Natural Autoantibodies: Self Recognition and Physiological Autoimmunity In: Natural Autoantibodies Their Physiological Role and Regulatory Significance Eds Shoenfeld & Isenberg CRC Press London p 1, 1993. Miller V, Chow DA, Greenberg AH. Natural Antibodies In: "Natural Immunity" Acad Press Australia, ed. Nelson DS, 1989: p. 172-217 Ternyck T, Avrameas S. Murine natural monoclonal autoantibodies: a study of their polyspecificities and their affinities. Immunol Rev 1986; 94:99-112. Coutinho A. Beyond clonal selection and network. Immun Rev 1989; 110: 63-87. Sela B-A, Edelman GM. Isolation by cell-column chromatography of immunoglobulins specific for cell surface carbohydrates. J Exp Med 1977; 145: 443-449. Galili U. Evolution and pathophysiology of the human natural anti-ct-galactosyl IgG (anti- Gal) antibody. Springer Seminars in Immunology 1993; 15: 155-171. Marchalonis JJ, Kaymaz H, Dedeoglu F, Schluter SF, Yocum DE, Edmundson AB. Human autoantibodies reactive with synthetic autoantigens from T-cell receptor beta chain. Proc Natl Acad Sci USA 1992; 89: 3325-3329. Prasad NKA, Papoff G, Zeuner A, Bonnin E, Kazatchkine MD, Ruberti G, Kaveri SV. Therapeutic preparations of normal polyspecific IgG (IVIg) induce apoptosis in human lymphocytes and monocytes: A novel mechanism of action of IVIg involving the Fas apoptotic pathway. J Immunol 1998; 161:3781-3790. Chow DA, Miller VE, Carlson GA, Pohajdak B, Greenberg AH. Natural resistance to tumors is a heterogeneous immunological phenomenon. Invasion and Metastasis 1981; 1: 205-219. Chow DA, Wolosin LB, Greenberg AH. Murine natural anti-tumor antibodies II. The contribution of natural antibodies to tumor surveillance. Int J Cancer 1981; 27: 459-469. Tough DF, Chow DA. Tumorigenicity of murine lymphomas selected through fluorescence-
344
18. 19.
20. 21.
22.
23. 24.
25. 26. 27. 28. 29. 30.
31. 32. 33.
34.
35. 36.
37.
detected natural antibody binding. Cancer Res 1998; 48: 270-275. Tough DF, Chow DA. Natural antibody recognition of v-H-ras induced 10T 89transformation. Natural Immun Cell Growth Regul 1991; 10: 83-93. Tough DF, Haliotis T, Chow DA. Regulation of natural antibody binding and susceptibility to natural killer cells through Zn++-inducible ras oncogene expression. Int J Cancer 1992; 50: 423-430. Tough DF, Feng X, Chow DA. p21 ras independent down-regulation of ras-induced increases in natural antibody binding during tumor progression. Nat Immun 1995; 14: 20-34. Wang H, Chow DA. PKC expression in 10T 89cells associated increased acquisition of natural antibodies with surveillance of preneoplasia and homeostasis of the organism. Scand J Immunol Scand J Immunol 1999; 49:381-390. Chow DA, Wolosin LB, Greenberg AH. Murine natural antitumor antibodies II. The contribution of natural antibodies to tumor surveillance. Int J Cancer 1981; 27: 459-469. Chow DA. Reduced tumorigenicity of threshold tumor inocula in xid-bearing mice treated with natural antibodies. Int J Cancer 1995; 60: 848-853. Chow DA, Yuan X-y, Tough DF. Polyclonal natural antitumor antibody binding dynamics: Preferential release of surface membrane molecules and increased metastasis. Inv & Met 1992; 12:218-232. Wang H, Chow DA. Natural antibody-induced intracellular signaling and growth control in C3H 10T 89fibroblast variants. Immunol 2000; 101: 458-467. Chow DA, Reese RM. Multifaceted saccharide reactivity in T lymphoma selection for high natural antibody binding. In preparation. Chow DA, Wang H, Zhang Z. Natural antibody surveillance of neoplastic and activated cells. J Biotecnologia Aplicada 1999; 16:18-19. Trowbridge IS, Thomas ML. CD45: An emerging role as a protein tyrosine phosphatase required for lymphocyte activation and development. Annu Rev Immunol 1994; 12: 85-116. Glotz D, Zanetti M. A natural neonatal hybridoma antigen autoantibody to the T200 antigen. J Immunol 1989; 142: 439-443. Zhang Z, Chow DA. Tumor progression: Inverse correlation between natural antibody binding and tumorigenicity with differential CD45 isoform expression. J Immunol 1997; 159: 344-350. Zhang Z, Ostergaard H, Chow DA. Natural antibody recognition of CD45RA. Abstracts for the Can. Soc Immunol 14th Spring Meeting, Bromont Quebec, 2000 #157. Karnitz LM, Abraham RT. Interleukin-2 receptor signaling mechanisms. Adv Immunol 1996; 61: 147-199. Francois C, Dantel J, Sorel M, Miyasaka M, Soulillou JP, Jaques Y. Antibodies directed at mouse IL-2R alpha and beta chains act in synergy to abolish T-cell proliferation in vitro and delayed type hypersensitivity in vivo. Transpl Int 1996; 9: 46-50. Bohn J, Roggenbuck D, Settmacher U, Docke W, Volk HD, Von-Baehr R, Jahn S. Binding of natural human IgM auto-antibodies to human tumor cell lines and stimulated normal T lymphocytes. Immunol Lett 1994; 39: 187-194. Wolf-Levine R, Azuma T, Aoki K, Yagami Y, Okada H. Specific IgG autoantibodies are potent inhibitors of autoreactive T cells. J Immunol and 1993; 151: 5864-5877. Deans JP, Serra HM, Shaw J, Shen YJ, Torres RM, Pilarski LM. Transient accumulation and subsequent rapid loss of messenger RNA encoding high molecular mass CD45 isoforms after T cell activation. J Immunol 1992; 148: 1898-1905. Johannisson A, Festin R. Phenotype transition of CD4 + T cells from CD45RA to CD45RO
345
38.
39.
40. 41. 42.
43.
44.
is accompanied by cell activation and proliferation. Cytometry 1995; 19: 343-352. Yamada A, Streuli M, Saito H, Rothstein DM, Schlossman SF, Morimoto C. Effect of activation of protein kinase C on CD45 isoform expression and CD45 protein tyrosine phosphatase activity in T cells. Eur J Immunol 1990; 20: 1655-1660. Mouthon L, Kaveri SV, Spalter SH, Lacroix-Desmazes S, LeFranc C, Desai R, Kazatchkine MD. Mechanisms of action of intravenous immune globulin in immunemediated diseases. Clin Exp Immunol 1996; 104 Suppl 1: 3-8. Ballow M. Mechanisms of action of intravenous immune serum globulin in autoimmune and inflammatory diseases. J Allergy Clin Immunol 1997; 100:151-7. Kumar V, Kono DH, Urban JL, Hood L. The T cell receptor repertoire and autoimmune diseases. Annu Rev Immunol 1989; 7: 657-682. Barclay AN, et al. Editors. The architecture and interactions of leucocyte surface molecules. In: The leucocyte antigen facts book 2nd Ed. Acad Press, Harcourt Brace & Co, Publishers, 1997. Lesage S, Steff AM, Philippoussis F, Page M, Trop S, Mateo V, Hugo P. CD4+ CD8+ thymocytes are preferentially induced to die following CD45 cross-linking through a novel apoptotic pathway. J Immunol 1997; 159:4762-4771. Pace KE, Lee C, Stewart PL, Baum LG. Restricted receptor segregation into membrane microdomains occurs on human T cells during apoptosis induced by galectin-1. J Immunol 1999; 163:3801-3811.
This Page Intentionally Left Blank
V I
NEUROIMMUNE PATHOLOGY
This Page Intentionally Left Blank
New Foundationof Biology
349
Edited by I. Berczi and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
Introduction
ISTVAN BERCZI
Department of Immunology, Faculty of Medicine, The University of Manitoba, Bannatyne Campus, 32-795 McDermot Avenue, Winnipeg, Manitoba R3E OW3, Canada
The notion of an interplay between the CNS and the immune system, in popular parlance, the "mind-body" issue, has intrigued the lay population over the ages. What holds particular fascination is not necessarily the basic science of the "hardware" and "software" of this interaction, however, which has in many ways been the focus of this text to date, but the phenomenology as it pertains to regulation of human disease. The issue has been touched upon in some of the earlier chapters, but in this, and the following, sections of the book are highlighted several key areas where clear evidence exists for an impact of the neuroimmune network on pathology. Kusnecov begins the discussion in this particular section by addressing the issue of the relationship between stress, health and the immune system. He emphasises the adaptive value of stress. The immunomodulatory effect of stress need not be harmful, but rather the maintenance of host defence is the goal. For instance exposing mice to an intense stressor is protective against experimental autoimmune encephalitis. These observations conform to the original observation of "general adaptation syndrome" of Hans Selye. Licinio and Wong present the localization and function of cytokines in the brain as detected by in situ hybridization. IL-1 is produced in the brain vasculature and also in the parenchyma in response to various stimuli, such as middle cerebral artery occlusion or the i.p. administration of LPS. The production of IL-1 receptor antagonist in the brain provided neuroprotection in a rat stroke model. The CNS produced far more IL-l-beta in response to peripheral inflammation when compared to those cytokines which antagonize IL-1 bioactivity. Central IL-l-beta regulates the beneficial effects of inflammation and leads to the production of nitric oxide synthase-2. Stanisz provides a brief overview of neurogenic inflammation (NI). The role of Substance-P (SP) is emphasized as a key initiator of NI. SP affects smooth muscle contraction, blood vessel permeability, neutrophil and macrophage extravasation, induces mast cell discharge, and activates monocytes and lymphocytes to release their mediators, such as histamine, IL-1, IL-2 and immunoglobulins. These mediators in turn propagate further inflammatory processes. Pain perception associated with inflammation is also mediated by SP in many systems. All acute and chronic inflammatory conditions are associated with increased levels of SP. If SP or it's receptor are blocked, the progress of inflammation is inhibited. SP plays a role in inflammatory diseases, such as asthma, inflammatory bowel disease, rheumatoid arthritis, nephritis, parasitic infections and in various skin disorders. Other tachykinins, vasoactive intestinal peptide, somatostatin, neuropeptide Y and bradykinins also regulate various functions in inflammation.
350
Autoimmune reactions directed against the neuroendocrine system may lead to neuropsychiatric disorders in experimental animals and man. Denburg and co-workers discuss the case of autoimmune MRL strain of mice, which show immunoglobulin deposits and cellular infiltration in the choroid plexus and brain parenchyma. This is associated with profound disturbances of emotional reactivity and motivated behaviour, before any other symptoms of disease would occur. The presence of "Neuropsychiatric" involvement in systemic lupus erythematosus (SLE) patients is also reviewed. The role of phospholipid autoantibodies in cognitive dysfunction has been established, which may be reversed by immunosuppressive therapy with low dose of glucocorticoids. The pathogenesis of encephalitis is presented by Owens et al. Activated T lymphocytes (Thl) cross the blood-brain barrier and provide protection against viral infections. Myelin auto-reactivity can occur as a consequence of epitope spreading from anti-viral responses. Naive T cells can enter the CNS during experimental autoimmune encephalitis (EAE) and may be activated by microglia that express MHC-II and co-stimulatory molecules (e.g. B7). IFNgamma can induce glial cells to produce a number of mediators, such as TNF and NO that are cytopathic for oligodendrocytes in vitro. TNF is also implicated in repair and regeneration. TNF is produced in IFN-gamma deficient mice, which is amplified by IFN-gamma. In IFN-gamma deficient mice EAE is lethal and the macrophage dominated perivascular infiltrates of EAE are replaced by disseminated invasive neutrophilia. The neutrophil attractant chemokines MIP2 and TCA3 become prominent. These observations demonstrate that immunological reactions in the CNS may fulfil both inflammatory and protective functions and that the immune system is capable of interpreting and amplifying CNS-derived signals.
New Foundation of Biology Edited by I. Berczi and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
351
Stress, Health and the Immune Response'Reciprocal Interactions Between the Nervous and Immune Systems
ALEXANDER W. KUSNECOV, ALBA ROSSI-GEORGE and SCOTT SIEGEL
Department of Psychology, Rutgers, The State University of New Jersey, New Brunswick, N J, USA
ABSTRACT Both the nervous (CNS) and immune systems are charged with maintaining biological integrity in the face of constant physical and psychological challenges from the environment. These challenges generate changes in each system that are distinguished as "stressful". However, while the term 'stress' has traditionally suggested negative health consequences, the experience or physiological situation known as "stress", is largely adaptive and activated for the good of the organism. Therefore, if we view biological integrity as a state of positive health, the psychological response (e.g., anxiety) to a stressor, is not necessarily unhealthy. Similarly, the neural, behavioral and biological consequences of an immune response to microbial organisms need hardly be considered unhealthy-in spite of the discomfort rendered by inflammatory changes. Moreover, although there is a large body of evidence demonstrating that psychological or physical stress can modify immune function, this does not necessarily imply that the immune system is compromised in maintaining protection against infectious agents or in maintaining self-tolerance. This dynamic relationship between the nervous and immune systems was very likely selected to serve some advantage in preventing disease. One instructive example is neuroendocrine feedback during an ongoing immune response. For example, the HPA response to interleukin-1 modifies exaggerated humoral and cellular immune function. Alternatively, such neuroendocrine feedback was unable to alter interleukin-2 production induced in vivo by staphylococcal enterotoxin B (SEB). This suggests that neuroendocrine immunoregulation during the immune response may be selective. Ultimately, it may be important to identify disrupted or ineffective regulatory interactions between the nervous and immune systems, which may reveal the reasons behind the appearance or persistence of immunopathology. One critical issue will be the degree to which chronicity of stressor exposure and/or impaired immune-mediated neural signaling modify stress circuits in the brain and impair normal nervous-immune functional interactions.
352
1
INTRODUCTION
It is now well documented that personality and emotions are in some way related to health, with psychological distress, depression, anxiety, and social support being strongly correlated with changes in susceptibility and/or resistance to infection, autoimmune disease, and cancer. The fundamental concept that binds this general theory together is that the immune system is very sensitive to neuroendocrine and autonomic nervous system changes imposed by fluctuations in the internal physiological milieu of which the immune system is an inseparable entity. Physiological fluctuations outside of some biologically circumscribed boundary may be viewed as a state of "stress", but one that need not imply an imperiled state for the organism. In the neurosciences the concept of plasticity has served to underscore this notion, emphasizing that stress-as a physiological state-is a natural, adaptive shift in function. Generally, this shift warrants correction, based on whether it is considered unacceptable (e.g., represents a biological threat) and/or incurs a prolonged energetic expense. Interestingly, the correction (through regulatory mechanisms that can also be used as measures of "stress") of the circumstances that induced the stressful state may also be seen a s the "stress response" (a fact which has plagued the definition of 'stress'). Ultimately, if we focus on stress as the behavioral and physiological response to some challenging event, it is difficult not to view this as being in the service of maintaining health and survival. However, with respect to the immune system, the problem that exists is understanding when psychological stress is exerting a health-promoting or health-impairing influence. For it is quite evident from studies discussed below, that stressor exposure can both increase and decrease the immune response. One aspect of stress research that has not been seriously addressed is the current understanding that the immune system is not only affected by events originating from within the brain (i.e. in response to "processive" or psychological stimuli), but that systemic influences arising from activated immune cells can also elicit stress-like responses within the central nervous system (CNS). This signifies the immune systems embrace of neuroendocrine factors that may serve to regulate in a feedback manner ongoing immune processes. If this is the case (and there is evidence to suggest this), then neuroendocrine responses induced by processive stimuli may not necessarily represent a deleterious event for the immune system. In what follows, we will review some of the animal and human literature on the effects of stressor exposure on immune function. It will not be our intention to review the stress literature in detail (which we have already done elsewhere-e.g, see (1). Instead, we will attempt to focus on the bidirectional communication existing between the brain and the immune system. To this end we will conclude in the latter half of this chapter with a consideration of recent data from our laboratory on the effects of the potent T cell stimulus, staphylococcal enterotoxin B, on the neural and behavioral reactivity to novel stressors. This and other research serves to emphasize that a true appreciation of the health implications of stress and immune function is likely to be found in considering the simultaneous, reciprocal elements of the CNS-immune system relationship.
2.
STRESSOR-INDUCED IMMUNOMODULATION
Understanding of nervous-immune interactions has been aided significantly by observations over the past few decades that physical and psychological stressors can alter basic parameters of immune function. It has been demonstrated that the characteristics of the stressor, host, and the immune-stimulating agent (i.e. antigen) all influence modulation of the immune response.
353
The stressor, for example, can vary in both severity and duration. Furthermore, the type of stressor used, the timing employed (e.g., before challenge with antigen vs. after challenge), coping and avoidance opportunities provided by the stressor, and the frequency with which it is delivered are all relevant issues that have been raised in the literature. Additionally, the age, sex, species, and strain of the host appear to play an important role. In human research, the social support available, coping style, and personality type of the subject are all proving to be very important in their relevance to immune function. We will briefly address studies that have primarily involved in vivo immune measures. For a thorough review of in vitro immune measures see (1).
3.
ANIMAL STUDIES
3.1.
Humoral Assays
Most animal studies examining stressor effects on immune function have used rats and mice, although some have examined the impact of infrahuman primates. Stressors of acute duration (footshock, cold exposure, or isolated housing) several days following immunization with sheep erythrocytes (i.e. SRBC) can suppress the number of B cells producing antibody (2, 3). However, application of the same stressor within two days, prior to or following immunization produced no alterations on the number of B cells producing antibody (2). In other studies, stressor application at the time of immunization was shown to augment or diminish antibody production to a commonly used protein keyhole limpet hemocyanin (KLH) and SRBC (4, 5). Other studies have shown that short durations of restraint stress (2 days) reduced antibody to SRBC while longer periods (4 days) showed no effect (6). In yet another study, acute conflict stress resulted in no change to the antibody response but more chronic exposure had an inhibitory effect on the antibody response to antigen (7). It is likely that the seemingly inconsistent changes reported in these studies are attributable to variations in stressor duration, type of stressor and the intensity of the stressor. An important factor that has not received systematic attention is the chronicity of stressor exposure (although see below). However, it is evident that differences exist in humoral immune measures between stressor-exposed and control groups, although the direction and conditions under which such differences are observed presently elude reliable prediction. 3.2.
Cellular Assays
Recent work has shown an enhancement of the DTH response to a single session (or acute) exposure to restraint stress in mice (8). By adding an additional characteristic to the stress paradigm (viz., shaking the restraint apparatus), a further augmentation of the DTH response was observed. These findings are in agreement with a much earlier study (9) that demonstrated an increase in the DTH response following short-term (or acute) stressor exposure. Alternatively, more prolonged (or chronic) stressor exposure exerts a suppressive influence on DTH reactivity (8). While this suggests that T cell mediated immune responses are suppressed by chronic stress, it should be noted that acute stressor exposure was initiated prior to challenge for DTH reactivity, whereas the chronic stress regimens were initiated prior to sensitization and continued up to the time of challenge. Therefore, it is not clear where the chronic stress exposure is acting (during priming of naive T cells or elicitation of memory T cell reactivity). Nevertheless, the study by Dhabar and McEwen is perhaps the first systematic attempt to
354
deal with an important issue of potential health relevance when considering the impact of stress on immune function. 4.
Human Studies
Demonstrations of stressor-mediated immunomodulation in human subjects is common, though such research comes with various limitations. For example, stressors may force people to engage in poor health practices such as smoking, poor nutritional intake, increased drug intake (e.g., alcohol) and impaired sleeping patterns see (10). All these factors may impact on immune function, thereby making it difficult to isolate pure neural and hormonal mechanisms in the effects of psychosocial stressors. In most experimental studies involving human subjects, the laboratory stressors are acute (no more than 20 minutes in duration), being administered only once. Some examples of the stressors used include making a speech, performing mental arithmetic, and attempting the Stroop color word interference task. The most commonly reported findings occurring in response to these stressors include elevations in natural killer (NK) and CD8 cell number and suppression of mitogen-mediated proliferative responses by peripheral blood leukocytes reviewed by (11). Findings from both animal and human research are compelling in that they suggest neuroimmunomodulation. However, one very important issue remains unresolved. This pertains to how closely laboratory stressors simulate ethological stressors encountered by animals and the every-day stressors encountered by people. Recognizing this, much of the work in human research involves 'naturalistic stressors'. For example, Glaser & Kiecolt-Glaser in a series of studies showed immunological alterations in medical students after taking medical school examinations (12). Decreases in NK cell activity, lymphocyte proliferation, and an increase in antibody to the herpes virus were documented (12). Other work showed 'negative' daily events, such as having an argument, were associated with decreased antibody production to a novel protein, while 'positive' daily events such as accomplishing a goal were correlated with increases in antibody production (13). These investigations suggest that daily, frequently occurring events may have the capacity to influence immune function. Other more chronic stressors such as having to care for relatives with Alzheimer's disease may reduce the cellular and humoral response to influenza vaccinations (14, 15). Certainly, these findings of immunological changes associated with naturalistic stressors corroborate the animal and human research using laboratory-imposed stressors.
5.
STRESS AND INFECTIOUS DISEASE
In discussing the impact of stress on immune function, it is important to remember that the immune system is a large multifaceted amalgam of cells and tissues operating in concert to detect and eliminate pathogenic microorganisms. Many studies of stress and immune function focus on a specific measure, such as antibody production or lymphocyte proliferation. However, measuring only one element of immunological behavior does not accurately depict the true state of the immune system. Therefore, general descriptions of augmented or suppressed immune "function" while convenient, are nonetheless inaccurate in describing the state of the immune system. Whereas in the aforementioned studies, differences between experimental and control groups in a particular measure serves to indicate some kind of alteration, it does not necessarily allow for a conclusion to be made about the competence of the immune system. Competence of the immune system is generally considered in terms of ability to defend against
355
infection. However, the concept of competence can be broadened to include self-regulation and the prevention of autoimmune disease. Moreover, in the realm of neural-immune interactions, it could be argued that immunocompetence involves the ability of activated immune cells to signal the CNS. This issue we will come back to later, since the ability of the immune system to communicate to the CNS is perhaps more reflective of the functional status of the CNS. Numerous animal studies have been conducted demonstrating that bacterial and viral infections in mice are affected by exposure to restraint, novel housing and electric footshock see (16). However, the most exciting set of findings in human subjects was reported by (17). In this study, several questionnaires were completed by nearly 400 subjects to access degrees of psychological stress from the prior year. Following assessment, nasal drops containing various viruses capable of producing respiratory illness (i.e. the common cold) were administered to the experimental groups, while saline was given to the control group. All groups were quarantined and monitored for presence of infection (measured by an antibody response to the virus) and for clinical symptoms. The results indicated a dose-dependent relationship between stress and rates of both symptoms and infection. Low stress groups had rates of infection at 75% while high stress groups were found to be approximately 90% infected. Similar differences, though not as significant, were found with respect to the presence of clinical symptoms. Other studies have replicated these findings, suggesting that psychological stress fosters the development of the common cold (e.g. 18). Numerous studies have also documented a psychosocial influence on the recurrence and duration of latent viral infections. Psychiatric illness, life events, mood state, and unhappiness have all been shown to have some role in increasing genital herpes recurrences (19).
6.
PSYCHOSOCIAL FACTORS AND AUTOIMMUNE DISEASE
Considering the role that the CNS plays in modulating immunological processes the effect of psychosocial factors on autoimmune disease has received considerable attention. The two most commonly researched autoimmune diseases believed to be influenced by stress are rheumatoid arthritis (RA) and multiple sclerosis (MS). The most consistent finding is that stressful life events are correlated with an exacerbation of autoimmune disease (20). For example, in a recent study by Ackerman (21) it was found that mild periods of stress resulted in an increased incidence of relapse in MS patients. Indeed, depression and anxiety are common in MS patients, which makes it difficult to determine whether stress arises from the disease and immune dysregulation (22). In animal models that approximate clinical conditions of autoimmunity it has been possible to assess more directly the influence of stressor exposure on the autoimmune process. Experimental autoimmune encephalomyelitis (EAE) closely resembles MS, and is an inflammatory disease of the CNS mediated by autoreactive T lymphocytes that target various components of myelin. EAE is most often studied in the Lewis rat or certain strains of mice (SJL, PL/J, B10.PL) and may be induced through a variety of procedures. All of these procedures involve injection of myelin basic protein (MBP), or the encephalitogenic peptides contained in myelin, such proteolipid protein (PLP) and myelin oligodendrocyte glycoprotein (MOG). Activated CD4+ T-helper cells traverse the blood brain barrier and are associated with the appearance of perivascular lesions in both the brain and spinal cord, and are linked to neurological deficits such as overt paralysis. In the Lewis rat, there is considerable evidence that daily sessions of restraint for periods of one hour augmented the severity of EAE, whereas more prolonged periods of restraint
356
suppressed the course of the disease (23). Similarly, restraint during remission can prevent relapses of EAE (24). Furthermore, it has been documented that hypothalamic-pituitary-adrenal activation is necessary for recovery from EAE in the Lewis rat (25). These findings suggest that elevation of GC hormones may be responsible for immunosuppression in EAE, and hence diminished clinical symptoms. Furthermore, it has been suggested that the stress animal experience as a result of paralysis facilitates the GC response and the associated remission of disease (24). 6.1.
Concluding comments on Stress and Immune Function
In short, there is considerable evidence that there are important interactions between the behavioral reactions of the organism to psychological stressors and subsequent immunological events, as well as disease states with an underlying immunological pathology. At present we understand very little about the precise mechanisms involved in mediating stressor effects on immune function. This work is in progress, although a number of mechanisms (e.g., glucocorticoids, catecholamines and opioid peptides) have been addressed in the past, and will continue to be pursued further in the future see (1). The above sections on infection and autoimmune disease have served to introduce an important issue supported by one of the more active areas in neuroscience: the effects of cytokines on the CNS. Infection and autoimmune disease represent extreme stress states that involve host invasion and organ dysfunction, as well as pronounced production of a number of proinflammatory and T cell derived cytokines. Activation of stress circuits in the brain is likely to arise as a result of perceptual and sensory elements arising from the soma, as well as from signaling molecules from the immune system. Such signaling solicits protective neuroendocrine and sympathetic activity that regulate immunological processes and metabolic processes. Indeed, recent evidence revealed that interleukin-1 a potent activator of the HPA axis and other CNS processes-is responsible for the increased glucocorticoid response during EAE in rats (26). Thus, adequate resources at the level of the CNS may be important in responding to signals from the immune system, thereby providing an essential supportive function to allow successful recovery from autoimmune disease.
7.
THE ROLE OF THE IMMUNE SYSTEM IN ELICITING THE STRESS RESPONSE
7.1.
Neural-Immune Interactions and the Role of Cytokines
As already mentioned, the nervous and immune systems share a mutually interactive relationship that promotes various forms of physiological and behavioral adaptations in the face of pathogenic challenges. This view has been based on the study of cytokines, in particular the proinflammatory cytokines IL-1, IL-6 and tumor necrosis factor-c~ (TNFct). These cytokines have been shown to alter neuroendocrine activity, increase neurotransmitter release, induce regional activation of immediate early genes, and modify basic behaviors, such as food ingestion, locomotion and sleep (27). In addition, viral and bacterial agents capable of inducing cytokine production in vivo display similar effects. Of these, the predominant model for assessing immunological activation of central nervous system (CNS) function is administration of lipopolysaccharide (LPS), which chiefly activates mononuclear phagocytic cells (e.g., macrophages), a major source of IL-1, IL-6, and TNF.
357
The neuromodulatory effects of LPS and proinflammatory cytokine administration are complemented by evidence for cytokine receptors in CNS tissue, both on microglial and neuronal cells, with de novo synthesis of cytokines occurring largely in astrocytes and oligodendrocytes (28). This suggests that local cytokine receptor-ligand interactions may take place in the brain independently of the immune system. Alternatively, engagement of central cytokine receptors may ensue from the release of cytokines by activated infiltrating lymphocytes and/or mononuclear phagocytic cells, as well as by cerebrovascular endothelial cells (29). Finally, direct effects of circulating cytokines on the CNS may occur through blood-brain saturable transport mechanisms or emanate from circumventricular regions where the bloodbrain barrier is highly permeable to large molecules (30). In addition to these humoral mechanisms, afferent neural pathways exist to mediate the influence of cytokines on CNS function (27). It has been suggested that immunologically provoked behavioral repertoires, such as anorexia, lethargy and sleep represent adaptive strategies (27). However, given the nature of the neural effects induced by cytokine treatment or LPS administration, there may be circumstances where adaptation may be compromised. For example, neurochemical-mapping studies have identified immune activation as a stressor that engages neural substrates of emotional reactivity (31). Consequently, in conjunction with psychological stress, immunological activation of the brain may impose additional and potentially subversive demands on neurochemical resources supporting emotional and cognitive functions. As a case in point, amygdaloid and hypothalamic neurons that synthesize the anxiogenic neuropeptide, corticotropin releasing hormone (CRH), are highly responsive to immunological stimuli (32). This introduces the possibility that fundamental alterations in mood and cognition may occur independently of, or prior to, more obvious signs of infectious illness. In apparent support of this, IL-1 and LPS have been shown to reduce exploratory behavior in novel or precarious test environments, in some cases dependent on central CRH receptors (33), but in others influenced by depressive illness-like effects on locomotion (34). Additionally, mice progressing into developmentally regulated stages of autoimmune reactivity display increased emotionality (35) that may be due to hyperactivation of the immune system. 7.2.
T Cell Function and the CNS: The Superantigen Model
It is notable that while the foregoing information represents increasing experimental support for immune-mediated influences on adaptive behavioral functions (27), there is a profound gap concerning the impact of T cell mediated immune responses on cognitive and emotional functions. This is in contrast to strong clinical evidence of disturbances in memory and emotion in autoimmune disease, including multiple sclerosis and following IL-2 immunotherapy (36, 37), reinforcing the possibility that T helper cytokines may alter affective and neuropsychological processes. Recently, a class of protein antigens has been characterized that have "super" antigenic properties (and hence, are referred to as 'superantigens'), in that they induce pronounced cytokine production by T cells in vivo, and subsequent to this, drive activated T cells into a lymphoproliferative phase that increases two-fold the initial percentage of the relevant T cell pool. This is quite a marked and impressive effect, since most benign protein antigens, such as ovalbumin and keyhole limpet hemocyanin, and various other immunogenic proteins do not stimulate sufficient numbers of T cells to allow for detectable levels of proliferation and cytokine production in vivo.
358
Given the wide number of superantigenic molecules of both bacterial and viral origin, and the similarity of the immune effects observed to some autoimmune processes, some investigators have suggested that these molecules may serve not only as models for examining normal immune processes, but also that surreptitious and stealth-like presence of microbial superantigenic proteins in the host may cause molecular mimicry that disrupts immune tolerance and initiates autoimmune disease. As such, superantigens may be useful in assessing the impact of T cell activation on behavior and its neural substrates. 7.3.
Staphylococcal Enterotoxin B
Exotoxins produced by the gram positive bacteria staphylococcus aureus are among the most commonly studied bacterial superantigens. Staphylococcal enterotoxin B (SEB) has received the greatest attention (38), having been shown to induce major production in vivo of CD4+ T cell cytokines (39, 40). SEB activation of T cells occurs via the variable region of the [3 chain (V[3) of the T cell receptor. This stimulation is dependent on the presence of MHC class II molecules, resulting in a massive cytokine response, that includes the rapid appearance in plasma of TNFct and nanogram quantities of IL-2 and IFN~, (39). The immunological effects of SEB have been well characterized in the B ALB/c mouse, which possesses a high percentage of CD4+/V[38+ T cells that respond to this superantigen. Furthermore, it has been shown that T cells activated with SEB mediate a dose-dependent stimulation of the hypothalamic-pituitary-adrenal (HPA) axis, as measured by increased ACTH and corticosterone elevations in plasma (41). 7.4.
Activation of corticotropin releasing hormone (CRH) neurons in the brain by SEB
Corticotropin releasing hormone (CRH) promotes numerous endocrine, autonomic, and behavioral functions in the brain. It is synthesized in large amounts in neurosecretory cells in the PVN, as well as in the central nucleus of the amygdala (ceA), where it is considered to play an important role in mediating fear and anxiety (42). BALB/c mice were shown to display increased CRH gene transcription in the PVN and ceA 4-6 hours after challenge with SEB (43). These changes were associated with elevated circulating IL-2 protein and ACTH, although examination of the brains from all animals did not reveal any hybridization for IL-2 mRNA. This suggests that within the time period examined (0-6 hour), T cell activation was confined to lymphoid compartments, such as the spleen. Although the changes in CRH mRNA were observed several hours after increased plasma ACTH, which occurs by 2 hours, it is likely they reflect compensatory responses to synthesis and release of neuropeptide. For instance, after LPS treatment or suppression of glucocorticoid synthesis, a rapid increase in CRH prespliced heteronuclear RNA indicates early activation of CRH-containing neurons, which is reflected much later by increased CRH mRNA (44). In support of the CRH mRNA data, systemic immunoneutralization of CRH using sheep anti-rat serum abrogated increased plasma ACTH measured at 2 hours after SEB challenge. This implies that although CRH mRNA in the PVN did not display a statistically significant increase until 6 hours after SEB challenge, CRH release was induced at 2 hours. This CRH was most probably released by neurosecretory cells of the PVN, since in situ hybridization conducted on spleens failed to show any CRH mRNA hybridization (43). Therefore, activated T cells or other components of the immune system (e.g., MHC class II positive APC) were unlikely to be the source of immunoneutralizable CRH.
359
7.5.
The contribution of IL-2 to CNS modulation by SEB
Immunoneutralization of IL-2 failed to affect the ACTH response to SEB. This is consistent with other evidence showing that acute administration of human recombinant IL-2 to mice does not elevate plasma corticosterone elevations. However, failure of IL-2 administration to affect plasma corticosterone does not necessarily imply failure to influence extrahypothalamic functions in the brain. Indeed, it has been shown in mice and rats that acute IL-2 injection increases hypothalamic noradrenaline turnover and mesolimbic dopaminergic function (45, 46). Consistent with this, immunoneutralization of IL-2 production by SEB challenged mice revealed a marked reduction in c-Fos immunoreactivity in the ceA, PVN, dorsomedial hypothalamus and suprachiasmatic nucleus (Kusnecov et al: unpublished observations). The significance of this attenuation is still to be determined. Nonetheless, these results suggest that reduction of endogenous circulating IL-2 results in attenuated excitation of neuronal populations in the ceA following SEB challenge. Because this occurred in the face of HPA activation unaffected by IL-2 immunoneutralization, it is consistent with the view that neuronal activation in the central nucleus of the amygdala is not a necessary prerequisite for HPA activation (47). 7.6.
Behavioral Assessment Following SEB Challenge
An important and consistent finding among studies examining LPS is the induction of sickness behavior, such as lethargy, anorexia and reduced exploration. The neurobiological effects of SEB may reflect alterations due to illness. In addition to endogenous elaboration of cytokines, the response to somatic illness may be driving the neural changes, contributing to the increased c-Fos immunoreactivity and CRH transcription and release. Therefore, we conducted a series of tests using the dose of SEB (50 ~tg) that consistently produced the above mentioned neural changes. Testing for consumption of food as well as for the development of conditioned taste aversion, it was found that SEB challenge was ineffective in reducing food intake of a familiar food substance nor in producing an aversion to novel foods paired with SEB injection. Therefore, it did not appear that excitation of CRH-containing and other neurons in the hypothalamus and amygdala was associated with malaise. We then decided to focus on the possibility that these neurobiological changes may be associated with altered perception of novel stimuli, specifically neophobia (i.e. fear of novelty). This was based on previous findings that the amount of food or water consumed in a novel environment can be modulated by CRH, as well as anxiogenic and anxiolytic drugs (42). In addition, seemingly normal appetitive behavior can raise CRH activity in the ceA (48), demonstrating that normal eating bouts may have considerable emotional valency. Since activation of T cells with SEB increased CRH mRNA in the central nucleus of the amygdala (ceA), and the amygdala mediates aversive reactions, in part through CRH production (42), we tested in a series of experiments whether SEB challenged animals would display neophobic responses to a novel drinking stimulus (43). Previously unhandled, group-housed animals were challenged with 50 ~tg SEB or with an equivalent volume of Saline, and 2 hours later removed from their cages and transferred to an individual holding cage equivalent in size, shape and bedding as the home cage. They were immediately exposed for 1 hour to a drinking bottle containing Prosobee liquid formula solution, to which they had no previous exposure. At the end of the 1 hour exposure, all animals were returned to their home cages and original cage mates. The results revealed a dramatic reduction in consumption of the novel tasting solution in the group of animals that had been challenged with SEB. This suggested that the CNS alterations in CRH mRNA
360
after immunological challenge with SEB may indeed influence "emotional" reactivity, perhaps through an enhancement of anxiogenic or fear processes. The procedure used to assess taste neophobia in the above experiment involved a temporary shift of housing to a new cage in which animals were exposed to a novel solution. This type of contextual manipulation can serve as a stressor, which was confirmed by sacrificing animals 15 and 30 minutes after placement into the new cage, and measuring plasma ACTH. The results showed a significant elevation of plasma ACTH concentration 30 minutes after the contextual manipulation (43). Therefore, the enhanced neophobic response in SEB challenged animals may have been due to an interaction with the stressor properties of a novel context. This was confirmed in a subsequent experiment, which showed a loss of the neophobic response in SEB challenged animals that received the novel solution in a familiar context (one to which animals had been habituated by daily pre-exposure). This result suggested that immunological challenge sensitizes animals to psychological stressors, which then augment neophobic responses to discrete and novel constituents of a novel environment. Alternatively, the immunological challenge may sensitize stressors to alter basic motivational properties, such as appetite. Consequently, in a third experiment, we manipulated the novelty of the taste solution. On two separate days, two different groups of animals were pre-exposed in their home cage to either a bottle of Prosobee or Water. On the third day, animals in each group were either challenged with 50 ~tg SEB or an equal volume of Saline. Two hours later, all animals were moved into the novel testing cage as in previous experiments, and exposed for 1 hour to Prosobee solution. Therefore, for half of the animals Prosobee was a novel taste solution, whereas for all animals the relocation to a new cage was a unique experience. The results showed that the level of consumption of SEB challenged animals previously experienced with the taste of Prosobee, did not differ from similarly experienced Salineinjected controls. However, if SEB-challenged animals were na'~'ve to the taste of Prosobee, consumption was significantly reduced compared to similarly na'~'ve Saline-injected control animals. Therefore, in spite of the novel (and presumably stressful) contextual conditions, consumption of the preferred drinking solution was affected only if it was novel. This shows that taste novelty is an important property determining whether the interaction between exposure to a novel environment and SEB challenge induces a behavioral avoidance of the drinking solution in that novel environment. Viewed another way, the novel solution was just one constituent of the generally novel context to which SEB challenged animals displayed an augmented neophobic reaction. Failure of SEB challenge to affect consumption of Prosobee if it was familiar, provided additional proof that it did not disrupt appetitive functions. The results of the foregoing behavioral experiments (for more detail see (43)), and the conclusion that psychological stress interacts with immunological challenge to promote enhanced gustatory neophobia, is consistent with the effects of SEB challenge on increased c-Fos immunoreactivity in the amygdala, as well as increased transcriptional activity of the CRH gene in the central nucleus of the amygdala (ceA). It remains to be determined whether CRH gene activation is important to the behavioral changes observed, and whether other potential neurochemical mediators may be involved. Neurons synthesizing CRH are highly responsive to immunological stimuli raising the possibility that fundamental alterations in mood and cognition may occur independently of or prior to more obvious signs of infectious illness. Furthermore, given that reduced levels of circulating IL-2 in SEB challenged mice resulted in a significant reduction of c-Fos immunoreactivity in the ceA, suggests that endogenous IL-2 may contribute to the behavioral effects of SEB.
361
8.
CONCLUSIONS
The general significance of the above findings and those of others who have examined the relationship between cytokines and stress (49, 50) relates to issues involving not just how stress may affect immune function and ultimately infectious and autoimmune disease, but also how stressor-induced immunomodulation may ultimately affect CNS function. As reviewed above, stress can influence infectious processes and/or autoimmune disease, which can result in elevated cytokine production and impact on the CNS. If augmented infection or autoimmunity are in part related to a prior history of stress, it is likely that elaborated cytokines are impacting on neural functions already overburdened by adaptational demands. Therefore, the demands of processive (i.e. psychological) and systemic (i.e. cytokine) influences in the nervous system can quickly deplete neuro-transmitter resources and promote dysregulation of central processes mediating adaptive responses to stress, resulting in heightened sensitivity and distorted cognitive and perceptual functions. Such arguments have been put forth to account for aspects of depression. That is, excessive or unchecked inflammation in general may contribute to affective symptoms (51). For example, Maes (51) has argued that depression is accompanied by an activation of the inflammatory response system, consisting of increased numbers of neutrophils, monocytes, and activated T cells; increased serum concentrations of pro-inflammatory cytokines (e.g., IL-1, IL-6, and IFN-7); increased secretion of neopterin, a marker of cell-mediated immunity induced by IFN-~,; and an increased secretion of prostaglandin E2. It is argued that the increased production of pro-inflammatory cytokines (IL-1, IL2, IL-6, and IFN-7) may be orchestrating the activation of the inflammatory response system in depression. Additionally, these cytokines might also play a role in the etiology of depression, which is consistent with the effects of IL-1, IL-2, and IFN-~, on motivational systems typically affected in depression. Examples of these alterations include anhedonia, anorexia, social withdrawal, anergy, irritability, sleep disturbances, and malaise (51 ). It is interesting that stress has been considered an etiologic factor in the onset of clinical depression. Therefore, given what we know about the enhancing effects of stress on immune function, and the relationship of cytokines to motivational and stress-reactive behaviors, it will be particularly interesting to consider whether psychiatric conditions involving disrupted affect and cognition may ultimately be related to influences arising from the immune system.
ACKNOWLEDGEMENTS Supported by NIMH grant 51051 and the Charles and Johanna Busch Memorial Fund.
REFERENCES 1. 2. 3.
Kusnecov AW, Rabin BS. Stressor-induced alterations of immune function: mechanisms and issues. International Archives of Allergy & Immunology 1994; 105:107-121. Zalcman S, Anisman H. Acute and chronic stressor effects on the antibody response to sheep red blood cells. Pharmacology, Biochemistry & Behavior 1993; 46: 445-452. Bhatnagar S, Shanks N, Meaney MJ. Plaque-forming cell responses and antibody titers following injection of sheep red blood cells in nonstressed, acute, and/or chronically stressed handled and nonhandled animals. Dev Psychobiol 1996; 29:171-181.
362
.
.
.
10. 11. 12. 13.
14.
15.
16. 17. 18.
19. 20. 21. 22. 23.
Wood PG, Karol MH, Kusnecov AW, Rabin BS. Enhancement of antigen-specific humoral and cell-mediated immunity by electric footshock stress in rats. Brain Behavior & Immunity 1993; 7: 121-134. Shanks N, Kusnecov AW. Differential immune reactivity to stress in BALB/cByJ and C57BL/6J mice: in vivo dependence on macrophages. Physiol Behav 1998; 65: 95-103. Millan S, Gonzalez-Quijano MI, Giordano M, Soto L, Martin AI, Lopez-Calderon A. Short and long restraint differentially affect humoral and cellular immune functions. Life Sci 1996; 59:1431-1442. Lyte M, Nelson SG, Thompson ML. Innate and adaptive immune responses in a social conflict paradigm. Clin Immunol Immunopathol 1990; 57: 137-147. Dhabhar FS, McEwen BS. Acute stress enhances while chronic stress suppresses cellmediated immunity in vivo: a potential role for leukocyte trafficking. Brain Behav Immun 1997; 11: 286-306. Blecha F, Barry RA, Kelley KW. Stress-induced alterations in delayed-type hypersensitivity to SRBC and contact sensitivity to DNFB in mice. Proc Soc Exp Biol Med 1982; 169: 239-246. Kiecolt-Glaser JK, Glaser R. Methodological issues in behavioral immunology research with humans. Brain Behav Immun 1988; 2: 67-78. Herbert TB, Cohen S. Stress and immunity in humans: a meta-analytic review. Psychosom Med 1993; 55: 364-379. Kiecolt-Glaser JK. Norman Cousins Memorial Lecture 1998. Stress, personal relationships, and immune function: health implications. Brain Behav Immun 1999; 13: 61-72. Stone AA, Neale JM, Cox DS, Napoli A, Valdimarsdottir H, Kennedy-Moore E. Daily events are associated with a secretory immune response to an oral antigen in men. Health Psychol 1994; 13: 440-446. Kiecolt-Glaser JK, Glaser R, Gravenstein S, Malarkey WB, Sheridan J. Chronic stress alters the immune response to influenza virus vaccine in older adults. Proc Natl Acad Sci USA 1996; 93: 3043-3047. Vedhara K, Cox NK, Wilcock GK, Perks P, Hunt M, Anderson S, Lightman SL, Shanks NM. Chronic stress in elderly carers of dementia patients and antibody response to influenza vaccination. Lancet 1999; 353: 627-631. Sheridan JF. Norman Cousins Memorial Lecture 1997. Stress-induced modulation of anti-viral immunity. Brain Behav Immun 1998; 12: 1-6. Cohen S, Tyrrell DA, Smith AP. Psychological stress and susceptibility to the common cold. N Engl J Med 1991; 325: 606-612. Stone AA, Bovbjerg DH, Neale JM, Napoli A, Valdimarsdottir H, Cox D, Hayden FG, Gwaltney JM Jr. Development of common cold symptoms following experimental rhinovirus infection is related to prior stressful life events. Behav Med 1992; 18: 115-120. Cohen S, Williamson GM. Stress and infectious disease in humans. Psychol Bull 1991; 109: 5-24. Walker JG, Littlejohn GO, McMurray NE, Cutolo M. Stress system response and rheumatoid arthritis: a multilevel approach. Rheumatology 1999; 38: 1050-1057. Ackerman KD, Rabin BS, Heyman R, Baum A. Stressful life events and disease activity in multiple sclerosis. Brain Beh Immun 2000; 14, 77. Riether AM. Anxiety in patients with multiple sclerosis. Semin Clin Neuropsychiatry 1999; 4: 103-113. Whitacre CC, Dowdell K, Griffin AC. Neuroendocrine influences on experimental autoimmune encephalomyelitis. Ann NY Acad Sci 1998; 840: 705-716.
363
24. 25.
26.
27.
28. 29. 30. 31. 32.
33.
34.
35. 36. 37.
38.
39.
40.
41.
42.
Levine S, Saltzman A. Nonspecific stress prevents relapses of experimental allergic encephalomyelitis in rats. Brain Behav Immun 1987; 1: 336-341. MacPhee IA, Antoni FA, Mason DW. Spontaneous recovery of rats from experimental allergic encephalomyelitis is dependent on regulation of the immune system by endogenous adrenal corticosteroids. J Exp Med 1989; 169:431-445. Del Rey A, Klusman I, Besedovsky HO. Cytokines mediate protective stimulation of glucocorticoid output during autoimmunity: involvement of IL-1. Am J Physiol 1998; 275:R1146-R1151. Maier SF, Watkins LR. Cytokines for psychologists: implications of bidirectional immuneto-brain communication for understanding behavior, mood, and cognition. Psychological Review 1998; 105: 83-107. Hopkins SJ, Rothwell NJ. Cytokines and the nervous system. I: Expression and recognition. Trends Neurosci 1995; 18: 83-88. Merrill JE, Benveniste EN. Cytokines in inflammatory brain lesions: helpful and harmful. Trends Neurosci 1996; 19: 331-338. Banks WA, Kastin AJ, Broadwell RD. Passage of cytokines across the blood-brain barrier. Neuroimmunomodulation 1995; 2: 241-248. Haas HS, Schauenstein K. Neuroimmunomodulation via limbic structures--the neuroanatomy of psychoimmunology. Prog Neurobiol 1997; 51: 195-222. Laflamme N, Barden N, Rivest S. Corticotropin-releasing factor and glucocorticoid receptor (GR) gene expression in the paraventricular nucleus of immune-challenged transgenic mice expressing type II GR antisense ribonucleic acid. J Mol Neurosci 1997; 8: 165-179. Dunn AJ, Antoon M, Chapman Y. Reduction of exploratory behavior by intraperitoneal injection of interleukin-1 involves brain corticotrophin-releasing factor. Brain Research Bulletin 1991; 26, 539-542. Montkowski A, Landgraf R, Yassouridis A, Holsboer F, Schobitz B. Central administration of IL-1 reduces anxiety and induces sickness behaviour in rats. Pharmacol B iochem Behav 1997; 58: 329-336. Schrott LM, Crnic LS. Increased anxiety behaviors in autoimmune mice. Behavioral Neuroscience 1996; 110: 492-502. Rao SM. Neuropsychology of multiple sclerosis. Curr Opin Neurol 1995; 8: 216-220. Walker LG, Walker MB, Heys SD, Lolley J, Wesnes K, Eremin O. The psychological and psychiatric effects of rIL-2 therapy: a controlled clinical trial. Psycho-Oncology 1997; 6: 290-301. Marrack P, Winslow GM, Choi Y, Scherer M, Pullen A, White J, Kappler JW. The bacterial and mouse mammary tumor virus superantigens; two different families of proteins with the same functions. Immunol Rev 1993; 131: 79-92: 79-92. Bette M, Schafer MK, Van Rooijen N, Weihe E, Fleischer B. Distribution and kinetics of superantigen-induced cytokine gene expression in mouse spleen. J Exp Med 1993; 178: 1531-1539. Florquin S, Amraoui Z, Abramowicz D, Goldman M. Systemic release and protective role of IL-10 in staphylococcal enterotoxin B-induced shock in mice. J Immunol 1994; 153: 2618-2623. Shurin G, Shanks N, Nelson L, Hoffman G, Huang L, Kusnecov AW. Hypothalamicpituitary-adrenal activation by the bacterial superantigen staphylococcal enterotoxin B: role of macrophages and T cells. Neuroendocrinology 1997; 65:18-28. Koob GF, Heinrichs SC. A role for corticotropin releasing factor and urocortin in
364
43.
44. 45.
46.
47. 48.
49.
50.
51.
behavioral responses to stressors. Brain Res 1999; 848: 141-152. Kusnecov AW, Liang R, Shurin G. T-lymphocyte activation increases hypothalamic and amygdaloid expression of CRH mRNA and emotional reactivity to novelty. J Neurosci 1999; 19: 4533-4543. Rivest S, Laflamme N. Neuronal activity and neuropeptide gene transcription in the brains of immune-challenged rats. Journal of Neuroendocrinology 1995; 7: 501-525. Lacosta S, Merali Z, Anisman H. Influence of acute and repeated interleukin-2 administration on spatial learning, locomotor activity, exploratory behaviors, and anxiety. Behav Neurosci 1999; 113: 1030-1041. Petitto JM, McCarthy DB, Rinker CM, Huang Z, Getty T. Modulation of behavioral and neurochemical measures of forebrain dopamine function in mice by species-specific interleukin-2. J Neuroimmunol 1997; 73:183-190. Gray TS, Bingaman EW. The amygdala: corticotropin-releasing factor, steroids, and stress. Critical Reviews in Neurobiology 1996; 10: 155-168. Merali Z, McIntosh J, Kent P, Michaud D, Anisman H. Aversive and appetitive events evoke the release of corticotropin-releasing hormone and bombesin-like peptides at the central nucleus of the amygdala. Journal of Neuroscience 1998; 18: 4758-4766. Connor TJ, Song C, Leonard BE, Merali Z, Anisman H. An assessment of the effects of central interleukin-1 beta, -2, -6, and tumor necrosis factor-alpha administration on some behavioural, neurochemical, endocrine and immune parameters in the rat. Neuroscience 1998; 84: 923-933. Merali Z, Lacosta S, Anisman H. Effects of interleukin-lbeta and mild stress on alterations of norepinephrine, dopamine and serotonin neurotransmission: a regional microdialysis study. Brain Research 1997; 761 : 225-235. Maes M. Major depression and activation of the inflammatory response system. Adv Exp Med B iol 1999; 461: 25-46.
New Foundationof Biology
365
Edited by I. Berczi and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
Cytokines in the Brain: From Localization and Function to Clinical Implications
JULIO LICINIO and MA-LI WONG
Laboratory of Pharmacogenomics, Neuropsychiatric Institute, University of California, Los Angeles, School of Medicine, 3357A Gonda (Goldschmied) Neuroscience and Genetics Research Center, 695 Charles E. Young Dr South, Box 951761, Los Angeles, CA 90095-1761
ABSTRACT A highly complex network of central and peripheral cytokines regulates important functions of the body that are essential to the responses to infection, inflammation and stress. This article has a focus on a new developments on the neurobiology of interleukin-1 (IL-1). IL-1 is a primary cytokine that affects key functions of the brain such as neuroendocrine regulation, temperature control, sleep, food intake and body weight. It is now clear that events that occur within the brain, particularly those that have an inflammatory component are associated with central IL-1 induction. Such events included kindled seizures, trauma, stroke, CNS infection or inflammation. It is still not clear whether events that do not exert a direct injury to the CNS such as psychological stress can activate the brain IL-1 system. Future directions for research in this field include the question of whether endogenous brain IL-1 occurring physiologically outside of the context of inflammation or acute neurogenic injury has role in the regulation of key CNS functions, such as the response to psychological stress, learning and memory, food intake and body weight regulation. Additionally, a role for brain cytokines in the biology of psychiatric disorders has not yet been fully established. A promising avenue for applied for applied research in this area is the development of IL-1 based interventions for the treatment of human CNS disorders.
CYTOKINES IN THE BRAIN: FROM LOCALIZATION AND FUNCTION TO CLINICAL IMPLICATIONS Pleiotropism and redundancy are the hallmarks of cytokine biology. Each cytokine has multiple and distinct effects in the periphery and in the central nervous system (CNS). Moreover, the effects of various cytokines can have substantial overlaps. A highly complex network of central and peripheral cytokines regulates key functions of the body that are essential to the responses to infection, inflammation and stress (1). Several key concepts have emerged in the last several years. Our work has supported the
366
hypothesis that there are distinct central and peripheral cytokine compartments, which are integrated but differentially regulated. It is interesting that the same cytokine molecules serve as informational substances in the immune and nervous systems, which are highly distinct in their organization, but which share an important common feature: memory. Where are we now in terms of our understanding of cytokine neurobiology? In the past decade our knowledge base in this area has greatly expanded. We now know that CNS disorders that used to be classified in separate categories share an inflammatory component (2). Those conditions include not only multiple sclerosis, the prototypic immune disorder of the CNS, but also diseases in the following categories: ischemic (stroke), neurodegenerative (Alzheimer's disease), traumatic (brain injury), and possibly psychiatric (depression and schizophrenia). In spite of considerable advances in our knowledge, many important points remain to be elucidated. Key questions that still need to be addressed include the following: 9 If caspase 1 is not required for IL-1 ct bioactivity, why do caspase 1 knockout animals only have 25% of the IL-lct response to systemic inflammation? 9 How can one explain the mismatches between the cells that express cytokines and cells that express cytokine receptors? 9 Do neurons synthesize interleukin-1 related ligands? 9 What are the precise endogenous roles of cytokines in the CNS? 9 Are cytokines involved in the biology of psychiatric disorders, such as schizophrenia and depression? 9 Is there a role for therapeutic interventions for CNS disorders aimed at modulating cytokine bioactivity? Our own work has been focused mostly on the IL-1 system, which consists of a complex network of ligands and binding sites (see Table I). Of the two bioactive ligands, IL-113 has been the most studied. IL-113 is synthesized as an inactive precursor molecule which requires enzymatic cleavage by caspase 1 for conversion into the bioactive 17 kD molecule. In contrast IL-lot does not require cleavage for bioactivity. However for unknown reasons caspase 1 knockout animals have only 25% of the IL-lct response to inflammation (and no bioactive IL-113) (3). A key feature of the IL-1 system is the presence of a third ligand, IL-1 receptor antagonist (IL-lra), which binds to the signal transducing IL-1 receptor, but exerts no activity other than to serve as a pure endogenous receptor antagonist (4-7). The levels of IL-1 bioactivity are therefore a reflection of the ratio of IL-let + IL-113 in relation to those of IL-lra. Levels of IL-Ira that are at least fifty-fold higher than those of IL-l~t + IL-113 are required for adequate counter regulation of IL- 113 bioactivity. We described the localization of IL-lra gene expression in the brain in 1991 (7), and subsequently we showed in collaboration with Nancy Rothwell's group that in an animal model of stroke, endogenous IL-Ira mRNA is induced in the area of lesion and that immune neutralization of IL-lra increases the area of stroke (8). Thus, we concluded that IL-lra is an endogenous neuroprotective agent. That conclusion is further substantiated by the findings that temporary focal cerebral ischemic injury in the mouse is attenuated following transfection with a interleukin-1 receptor antagonist adenovirus vector (9). It is clear that the brain can synthesize IL-113, and that inflammatory events involving IL-113 are a key element in the biology of several types of brain disorders. This review will focus on some of the most exciting emerging aspects on the biology and clinical implications of brain IL- 1.
367
Table I
Elementsof the IL-1 System.
Ligands: IL-lot-alpha IL- 1[5-beta IL-1 receptor antagonist Binding Sites: IL-1 receptor type I (signal transducing) IL-1 receptor type II (decoy receptor) IL-1 receptor accessory protein I Soluble receptors Autoantibodies
THE CENTRAL AND PERIPHERAL CYTOKINE COMPARTMENTS ARE INTEGRATED BUT DIFFERENTIALLY REGULATED Our work in 1997 showed an integration and differential counter regulation of the central and peripheral cytokine compartments. Specifically, we demonstrated that in the context of systemic inflammation, peripheral tissues such as the pituitary gland had expression of IL-lra mRNA in levels that far exceeded those of IL-1 [5. Such excess is required for effective counter regulation of IL-I[5 bioactivity. In contrast in the CNS we did not find induction of IL-lra mRNA in levels that exceed those of IL-I[5 (10). The sequence of events linking peripheral inflammation, peripheral IL-1 synthesis, brain induction of IL-1 as a primary cytokine in the CNS causing the induction of secondary cytokines and CNS effects has been further supported by the elegant work of Laye et al., (11). Briefly, the group at Bordeaux has showed that peripheral LPS treatment of the mouse resulted in induction of IL-1 [5 in the circulation and IL-1 [5 mRNA induction in the brain, along with central induction of IL-6 mRNA and TNF-~t mRNA. Physiologically such changes occurred in the context of anorexia. Intraventricular (icv) treatment with IL-lra preceding LPS administration did not affect the induction of plasma IL-1, but it blocked acute central induction of IL-6 mRNA and TNF-c~ mRNA and attenuated LPS-induced anorexia. These data show a sequence of events that require the effects of IL-1 in the brain both for the central induction of IL-6 mRNA and TNF-c~ mRNA and for the full CNS effects of peripheral inflammation on food intake. The recent work of Eriksson and colleagues (12) at the Karolinska Institute confirmed the findings that in the context of inflammation, central IL-lra expression does not exceed that of IL-1 [3, and that consequently the counter regulation of IL-1 [3 bioactivity in the brain is different from that occurring in the periphery. However, following systemic kainic acid administration there is robust induction of IL-lra in the hippocampus. This may suggest that different stimuli elicit different patterns of central IL-1 counter regulation (13).
368
3.
BRAIN IL-1 INDUCTION FOLLOWING NEUROGENIC INTERVENTIONS
Several recent lines of evidence appear to indicate that microglia may be the major, if not the sole source of brain IL-I[5 after acute events (12, 14). Importantly, neurotrophic factor induction after brain trauma appears to be dependent on the induction of IL-1 in the brain, as IL-1 deficient mice do not synthesize ciliary neurotrophic factor (CNTF) in response to trauma as wildtype animals do. The ability to secrete CNTF is restored in those animals after central IL-1 [5 treatment. Miles Herkenham's group (15) recently showed that if LPS is administered directly into the brain there is an intense pattern of CNS cytokine expression that is far more accentuated than that seen after peripheral LPS injection. This would indicate that an intact blood-brain barrier is an element for the differentiation of central cytokine responses during the course of peripheral inflammation and according to the authors this might represent a mechanism for protection of the brain during the course of acute immune challenges.
.
IL-115 INDUCTION CAN DIFFER FOLLOWING NEUROGENIC OR PSYCHOLOGICAL EVENTS
Plata-Salaman and colleagues found that amygdala kindling in rats results in acute (2 hour) but not chronic (3 weeks) upregulation of the transcripts for IL-I[5 and TNF-~t, and TGF-[3, including a 30-fold induction of IL-I[5 and TNF-ot mRNA in the prefrontal cortex, but no regulation of IL-lra (16). Those data further support the concept that different stimuli elicit different levels of IL-1 counter regulation. Having shown that neurogenic events regulate IL-I[5 mRNA levels the same authors examined whether a psychological stressor would regulate IL-I[5 mRNA. They looked at the effects of acute (1 session) and chronic (30 one-day session) restraint or exposure to a predator (ferret) and found no induction of brain cytokines (16). These results differ from those of Nguyen et al., who showed induction of IL-1 immunoreactivity in rat brain following inescapable shock (17). Tanebe et al., (18) showed that repeated cold stress induced IL-115 and IL-2 mRNA expression in the hypothalamus of ovariectomized rats. The differences between those results could be explained by the differences in the type of stimuli (restraint or predator exposure versus inescapable shock or cold stress) and in the methods used (RNase-protection assay versus immunoassay). In conclusion, the issue of IL-1 induction in the brain by stress has not been fully resolved yet.
6.
TECHNICAL ISSUES
Parker et al. (19) recently showed that perfusing tissue to remove blood cells affected the expression of IL-I[5 in the brain in response to icv IL-I[5 administration. This study supports the logical assumption that tissue sources other than brain have to be taken into account for the results of highly-sensitive RT-PCR reactions performed with homogeneized tissue. Those reactions can now actually quantify absolute copy numbers by use of the TaqMan real-time PCR procedure, which has been optimized for the study of I1-115 gene induction in the CNS (20). Moreover, precision of dissection is another potential confounding variable. Additional factors that can contribute to IL-lgene expression in the brain and that therefore need to be well controlled in experimental situations include stimulation of the afferent vagus nerve (21).
369
On the other hand, data from other groups question the role of the vagus: van Dam et al., (15) found no changes in the patterns of central IL-I[3 expression after LPS administration in vagotomized rats that were compared with sham-operated controls.
7.
IL-113IN HUMAN CNS TUMORS AND INFECTIONS
The role of brain cytokines in tumors such as astrocytomas is now being explored. Ilyin et al., examined astrocytoma cells obtained shortly after tumor neurosurgical resection. Those cells respond to in vitro treatment of human IL-I[3 with a significant upregulation of IL-lc~, IL-I[3, IL-1RI, and TNF-c~ mRNAs. However, IL-lra mRNA was not upregulated. In a control experiment those authors showed that IL-I[3 application did not modulate any cytokine component in acutely resected and dissociated primitive neuroectodermal tumor cells. The data suggest the existence of a positive autoregulatory IL-I[3 feedback system and synergistic IL-I[3-TNF-c~ interactions, which can be involved in the biology and growth of pilocytic astrocytomas. The lack of IL-lra counter regulation might be a new feature of those tumors. Future studies should test the biological significance of that event. If the lack of IL-lra counter regulation is found to be biologically relevant, then it might be possible to use IL-lra or a compound with similar cytokine inhibitory activity for brain immunotherapy of astrocytomas (22). In cerebral infections such as cerebral malaria, meningitis, and encephalitis, expression of IL-I[3 and TNF-c~ was observed by Brown et al., (23). Thus, in severe diseases in which human brain tissue is made available as surgical specimens (astrocytomas) or postmortem (cerebral malaria, meningitis, and encephalitis), cytokine induction can be demonstrated as part of the disease process.
8.
FUNCTIONAL SIGNIFICANCE V E R S U S THERAPEUTIC RELEVANCE
The issue of functional significance has to be taken into consideration in translational studies. For example, if the levels of a cytokine are increased in the brain in a pathological state it would be intuitive to assume that such induction might open up new therapeutic strategies. However, before getting carried away it is important to document whether or not the cytokine affects outcome. The interesting work of Knoblach and Faden (24) showed that IL-I[3 was induced in the cerebral cortex in the context of traumatic brain injury. However, icy treatment with either soluble IL-1 receptors or with IL-lra did not affect outcome in a series of motor tasks. These results indicate that caution is warranted in the generation of hypotheses on the role of IL-1 in the treatment of CNS disorders that are based on data of local brain IL- 1 induction.
9.
CONCLUSIONS
There are several very recent pieces of evidence that add to the complexity of the role of IL-1 [3 in brain function. Importantly, it has become clear that different types of stressors such as cold stress, inescapable shock, and restraint are associated with different types of IL-1 response. Moreover, different types of neurogenic events such as kindling, infection, and tumor can cause different patterns of brain IL-1 induction and counter regulation. Several key questions
370
remain answered. For example, the role of brain IL-l[5 in psychiatric disorders is still unclear. Moreover, it has not yet been demonstrated that IL-1 based therapeutic interventions will be of use in the treatment of human disease. Future studies will continue to expand the frontiers of existing knowledge on the functional and therapeutic implications of brain cytokines such as IL-1.
ACKNOWLEDGEMENTS The work of J.L. is supported by NIH grants U01GM61394, RO1DK58851 and K30HL04526, by the Stanley Foundation, and by Sanofi-Synthelabo, Italy. The work of M.-L.W. is supported by NH grant 5P50AT00151 and by NARSAD-National Alliance for Research on Schizophrenia and Depression.
REFERENCES 1. 2. 3.
4.
5.
6. 7. 8.
9.
10.
11.
12.
Licinio J, Wong ML. Pathways and mechanisms for cytokine signaling of the central nervous system. J Clin Invest 1997; 100:2941-7. Sternberg EM. Neural-immune interactions in health and disease. J Clin Invest 1997; 100:2641-2647. Li P, Allen H, Banerjee S, Franklin S, Herzog L, Johnston C, et al. Mice deficient in IL-l[5 converting enzyme are defective in production of mature IL-l[5 and resistant to endototoxic shock. Cell 1995; 80:401-411. Hannum CH, Wilcox CJ, Arend WP, Joslin FG, Dripps DJ, Heimdal PL, et al. Interleukin-1 receptor antagonist activity of a human interleukin-1 inhibitor. Nature 1990; 343: 336-340. Eisenberg SP, Evans RJ, Arend WP, Verderber E, Brewer MT, Hannum CH, et al. Primary structure and functional expression from complementary DNA of a human interleukin-1 receptor antagonist. Nature 1990; 343:341-346. Ohlsson K, Bjork P, Bergenfeldt M, Hageman R, Thompson RC. Interleukin-1 receptor antagonist reduces mortality from endotoxin shock. Nature 1990; 348: 550-552. Licinio J, Wong ML, Gold PW. Localization of interleukin-I receptor antagonist mRNA in rat brain. Endocrinology 1991; 129: 562-564. Loddick SA, Wong ML, Bongiorno PB, Gold PW, Licinio J, Rothwell NJ. Endogenous interleukin-1 receptor antagonist is neuroprotective. Biochem Biophys Res Commun 1997; 234:211-215. Yang GY, Mao Y, Zhou LF, Ye W, Liu XH, Gong C, et al. Attenuation of temporary focal cerebral ischemic injury in the mouse following transfection with interleukin-1 receptor antagonist. Brain Res Mol Brain Res 1999; 72: 129-137. Wong ML, Bongiorno PB, Rettori V, McCann SM, Licinio J. Interleukin (IL) lbeta, IL-1 receptor antagonist, IL-10, and IL-13 gene expression in the central nervous system and anterior pituitary during systemic inflammation: pathophysiological implications. Proc Natl Acad Sci U S A 1997; 94: 227-232. Laye S, Gheusi G, Cremona S, Combe C, Kelley K, Dantzer R, et al. Endogenous brain IL-1 mediates LPS-induced anorexia and hypothalamic cytokine expression. Am J Physiol 2000; 279: R93-98. Eriksson C, Nobel S, Winblad B, Schultzberg M. Expression of interleukin 1 alpha and
371
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
beta, and interleukin 1 receptor antagonist mRNA in the rat central nervous system after peripheral administration of lipopolysaccharides. Cytokine 2000; 12: 423-431. Eriksson C, Tehranian R, Iverfeldt K, Winblad B, Schultzberg M. Increased expression of mRNA encoding interleukin-1 beta and caspase-1, and the secreted isoform of interleukin-1 receptor antagonist in the rat brain following systemic kainic acid administration. J Neurosci Res 2000; 60: 266-279. Herx LM, Rivest S, Yong VW. Central nervous system-initiated inflammation and neurotrophism in trauma: IL-1 beta is required for the production of ciliary neurotrophic factor. J Immunol 2000; 165: 2232-2239. Stern EL, Quan N, Proescholdt MG, Herkenham M. Spatiotemporal induction patterns of cytokine and related immune signal molecule mRNAs in response to intrastriatal injection of lipopolysaccharide. J Neuroimmunol 2000; 106:114-129. Plata-Salaman CR, Ilyin SE, Turrin NP, Gayle D, Flynn MC, Romanovitch AE, et al. Kindling modulates the IL-lbeta system, TNF-alpha, TGF-betal, and neuropeptide mRNAs in specific brain regions. Brain Res Mol Brain Res 2000; 75: 248-258. Nguyen KT, Deak T, Owens SM, Kohno T, Fleshner M, Watkins LR, et al. Exposure to acute stress induces brain interleukin-lbeta protein in the rat. J Neurosci 1998; 18: 2239-2246. Tanebe K, Nishijo H, Muraguchi A, Ono T. Effects of chronic stress on hypothalamic lnterleukin-lbeta, interleukin-2, and gonadotrophin-releasing hormone gene expression in ovariectomized rats. J Neuroendocrinol 2000; 12: 13-21. Parker LC, Rushforth DA, Rothwell NJ, Luheshi GN. IL-lbeta induced changes in hypothalamic IL-1R1 and IL-1R2 mRNA expression in the rat. Brain Res Mol Brain Res 2000; 79: 156-158. Li X, Wang X. Application of real-time polymerase chain reaction for the quantitation of interleukin-lbeta mRNA upregulation in brain ischemic tolerance. Brain Res Brain Res Protoc. 2000; 5:211-217. Hosoi T, Okuma Y, Nomura Y. Electrical stimulation of afferent vagus nerve induces IL-lbeta expression in the brain and activates HPA axis. Am J Physiol 2000; 279: R141-147. Ilyin SE, Plata-Salaman CR. An efficient, reliable and inexpensive device for the rapid homogenization of multiple tissue samples by centrifugation. J Neurosci Methods 2000; 95: 123-125. Brown H, Turner G, Rogerson S, Tembo M, Mwenechanya J, Molyneux M, et al. Cytokine expression in the brain in human cerebral malaria. J Infect Dis 1999; 180: 1742-1746. Knoblach SM, Faden AI. Cortical interleukin-1 beta elevation after traumatic brain injury in the rat: no effect of two selective antagonists on motor recovery. Neurosci Lett 2000; 289: 5-8.
This Page Intentionally Left Blank
New Foundation of Biology Edited by I. Berczi and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
373
Neurogenic Inflammation" Role of Substance P
ANDREW M. STANISZ
Department of Pathology and Molecular Medicine, McMaster University, Hamilton, Ontario and AJBioCom, Dundas, Ontario.
ABSTRACT The term "neurogenic inflammation" was introduced some 50 year ago by a Hungarian group of Jancso and Szolcsanyi. In their initial experiments skin injection of an extract from chili pepper (capsaicin) resulted in local inflammation. It was suggested that capsaicin caused sensory nerves to release substance(s) which in turn activated inflammatory cells. The neuropeptide, substance P (SP) seems to be a prime candidate for these effects. All classical signs of inflammation (heat, redness, swelling, pain and consequent loss of function) may be affected by SP. Later, it became evident that many other neuropeptides including remaining tachykinins, vasoactive intestinal peptide, somatostatin, neuropeptide Y or bradykinins could also effect various functions of inflammation. However, for the purpose of this presentation we will concentrate on SP. Substance P affects inflammation both in a direct and indirect way. Directly, SP affects smooth muscle contraction, epithelial permeability, and neutrophile and macrophages traffic. In addition SP affects immune cells by activating mast cells, monocytes or lymphocytes to release their mediators such as histamine, IL-1, IL-2 and immunoglobulins respectively. Indirectly, these mediators are released from inflammatory cells and in turn activate additional surrounding cells to further propagate the inflammatory process. Similarly, pain perception associated with inflammation is in many systems mediated by a SP. Therefore, SP has been identified as a pro-inflammatory neuropeptide. Increased levels of SP have been found in many inflammatory diseases such as asthma, inflammatory bowel disease, rheumatoid arthritis, nephritis, parasitic infections or various skin disorders. In fact it is safe to conclude that in all chronic and acute inflammatory conditions increased levels of SP are present. Moreover, it is possible by blocking SP release or SP receptors to interfere with the progress of inflammation. Obviously, this is a very complicated system in which various neuropeptides differentially interact and affect inflammatory cells. It is also clear that inflammation remains under strict neuronal control. We are only starting now to understand these interactions. Further investigations of the nerve/immune cell network could lead to new therapeutic strategies in management of inflammatory diseases. Indeed, some agents which act upon SP, have been recently successfully included in clinical trials.
374
1.
INTRODUCTION
The term "Neurogenic Inflammation" was introduced some 50 years ago by a Hungarian group [1 ]. In their initial experiments skin injection of an extract from chili pepper (capsaicin) resulted in local inflammation. Inflammation was not induced in denervated skin, therefore, it was suggested that capsaicin caused the sensory nerves to release substance(s) which in turn activated inflammatory cells. Since then the concept of neurogenic inflammation has received growing attention in the literature. It has been discussed extensively in the context of various inflammatory diseases affecting the eye, skin, respiratory or gastrointestinal tract and in association with several chronic diseases such as rheumatoid arthritis, Crohn's disease or asthma. In its initial historical definition neurogenic inflammation refers to reactions that include vasodilatation, plasma extravasation and smooth muscle contraction as a result of neuronal activation and mediator release from sensory endings. One of such mediators, neuropeptide substance P, seems to be a prime candidate for these effects and was suggested to have a direct role in smooth muscle contraction, vasodilatation and plasma extravasation [2]. We will not discuss here these responses. However, all cardinal signs of inflammation (heat, redness, swelling, pain and consequent loss of function) are affected during the release of substance P from nerve endings as it happens during neurogenic inflammation. In particular, traffic and cellular response of all inflammatory cells is affected directly or indirectly by the nervous system.
2.
SUBSTANCE P AND INFLAMMATORY CELLS
For in depth analysis of substance P as a neurotransmitter and its role in classical inflammation the reader is referred to excellent reviews by [3, 4] respectively. Substance P is an 11 amino acids peptide with positive charge. Its main biological activity resides in its carboxyl sequence which is sheared with the rest of the tachykinin family peptides, such as neuromedin K or substance K. Their biological activity is similar and often attributed to substance P. Therefore, various activities of substance P may be shared with other tachykinins and limited only by neuropeptide availability and specific receptors on target cells. Substance P, a product of the nervous system, is found in particular in unmyelinated sensory C fibers. Interestingly substance P can also be produced by non-neuronal inflammatory cells such as macrophages [5] and eosinophils [6, 7]. The effects of substance P on target cells are mediated via specific surface receptors [8]. These receptors, being different for each of the tachykinins, show a high degree of cross-reactivity [9]. In addition, some of the effects of substance P, for example, effects on mast cell degranulation, are thought to be non receptor mediated, but rather to be the result of direct insertion of substance P into the cell membrane bilipid layer and consequent activation of the protein kinase system [10, 11 ]. It is now evident that substance P and other tachykinins affect the function of lymphocytes, macrophages, neutrophils, eosinophils and mast cells, all of which are primary responding cells during inflammatory reaction. For this to be true, substance P must be readily available for these cells. Indeed it has been shown that substance P is localized within the mucosa and around blood vessels [12, 13]. In fact substance P containing nerves are found in close proximity, often adjacent to mast cells or lymphocytes and direct membrane/membrane contact between these cells and nerves has been shown [14]. Subsequently, substance P has been identified as a prime mediator of neurogenic inflammation [4]. In this brief presentation we will discuss the evidence suggesting that all cellular aspects of the inflammatory response, in
375
particular the reaction of lymphocytes, mast cells and macrophages and the subsequent cytokine release/synthesis, are affected by substance P. In addition, we will discuss the role of substance P in chronic inflammatory diseases. Substance P activates very broad spectrum of response in various cells. For example substance P enhances macrophage phagocytosis [15] and neutrophil activation [ 16]. Tromboxane production and prostaglandin synthesis are altered by substance P [17]. Mast cells release histamine and TGF alpha in the presence of substance P [18]. These effects are particularly important in asthma, as broncchoconstriction and broncchosecretion are main clinical manifestations of this disease [19]. Hematopoiesis, cytokine production [20] and lymphocyte migration [21] are all affected by substance P. Over the years our own laboratory has been involved in studying the role that substance P plays in modulating immunity [22]. We were able to confirm most of the above findings and significantly contribute to the field. In early experiments we have shown that substance P stimulates lymphocyte proliferation and immunoglobulin synthesis both in vitro and in vivo [23, 24]. These effects were mediated via specific substance P receptors on both T and B lymphocytes [25]. Similarly natural killer cell activity and cytokine production was enhanced by substance P [26]. Indirectly these mediators released from lymphocytes are able to activate surrounding cells and further affect the progress of the inflammatory process.
3.
SUBSTANCE P AND INFLAMMATORY DISEASES
Substance P plays a major role in the initiation and/or potentiating of inflammatory response. Firstly, during chronic inflammatory conditions local levels of substance P at the site of inflammation are significantly increased, as is the expression of substance Preceptors on inflammatory cells [27]. Secondly, as shown by Levine, and co-workers [28], denervation, or the use of anti-substance P antibody, inhibit the progress of inflammation. There is now a vast body of evidence linking substance P to various inflammatory diseases. Its proinflammatory role has been suggested in the pathophysiology of asthma [29, 30], urticaria, rheumatoid arthritis [31], irritable bowel disease [32], and so on. In our own studies synovial fluid levels of SP were significantly elevated in rheumatoid arthritis patients, as compared to controls, and this correlated with the severity of the disease and the degree of activation of inflammatory cells [33]. In animal models it has been shown that depleting substance P containing nerves by the neurotoxin, capsaicin, and the use of specific anti-substance P antibodies, resulted in inhibition of inflammation [34]. In contrast, intra-articular injection of substance P dramatically enhanced inflammation. In similar experiments using the Trichinella spiralis model of intestinal inflammation, we were able to confirm elevated levels of substance P during the progression of the disease [35]. We could inhibit the inflammatory process by specific anti-substance P antibodies or by antagonists of substance P [36].
4.
CONCLUSIONS
Substance P is synthesized and released at the site of inflammation from both neuronal and non-neuronal sources. It can affect inflammation both directly and indirectly. Direct effects include activation of smooth muscle cells, changes in endothelial permeability and vasodilatation. Indirect effects are associated with activation of a spectrum of inflammatory cells which in turn release and/or promote the synthesis of various cytokines and mediators, which
376
induce inflammation. In fact it is safe to conclude that in all acute and chronic inflammatory conditions increased levels of substance P are present. Moreover, it is possible to interfere with the progress of inflammation by blocking substance P release, activity, or receptors to interfere with. Obviously, this is a very complicated system in which various neuropeptides and cytokines play a role and affect inflammatory cells. It is also clear that inflammation remains under strict neuronal control. We are only starting now to understand these interactions. Further investigations of the nerve/inflammatory cell interactions could lead to new therapeutic strategies in management of inflammatory diseases. Indeed, some agents which act upon substance P have been recently successfully tested in clinical trials.
REFERENCES 1.
2.
3. 4. 5. 6.
7. 8. 9. 10.
11. 12. 13.
14.
15.
Jancso N, Jancso-Gabor A, Szolocsanyi J. Direct evidence for neurogenic inflammation and its prevention by denervation and by pretreatment with capsaicin. Br J Pharmacol 1967; 1: 138-51. Lembeck F, Holzer P. Substance P as a neurogenic mediator of antidromic vasodilatation and neurogenic plasma extravasation. Naunyn-Schmiedeberg's Arch Pharmacol 1979; 310: 175-186. Pernow B. Substance P Pharmacol Rev 1983; 35: 85-103. Payan DG. Neuropeptides and inflammation: the role of substance P. Annu Rev Med 1989; 40:341-325. Pascual DW, Bost KL. Substance P production by macrophage cell lines. Immunology 1990; 71 : 52-57. Neil GA, Blum A, Weinstock JV. Substance P but not vasoactive intestinal peptide modulates immunoglobulin secretion in murine schistosomiasis. Cell Immunol 1991; 135:394-401. Weinstock JV, Blum J, Walder J, Walder R. Eosinophils from granulomas in murine Schistosomasis Mansoni produce substance P. J Immunol 1988; 141: 961-966. Payan DG, Brewster DR, Goetzel EJ. Stereo-specific receptors for substance P on cultured IM-9 lymphoblasts. J Immunol 1984; 133: 3260-3265. Parnet P, Mitsuhashi M, Turck CW, Kerdelhue B, Payan DG. Tachykinin receptor cross-talk. BBI 1991 ; 5:73-81. Duplaa H, Convert O, Sautereau AM, Tocanne JF, Chassaing G. Binding of substance P to monolayers and vesicle made of phosphatidylcholine and/or phosphatidylserine. Biochim Biophys Acta 1992; 1107: 12-19. Mousli M, Bueb JL, Bronner C, Rouot B, Landry Y. G protein activation. TiPS 1990; 11: 358-364. Stead RH, Bienenstock J, Stanisz AM. Neuropeptide regulation of mucosal immunity. Immunol Rev 1987a: 100: 333-359. Stead RH, Tomioka M, Quinonez G, Simon GT, Felten SY, Bienenstock J. Intestinal mucosal mast cells in normal and nematode-infected rat intestine are in intimate contact with peptidergic nerves. Proc Natl Acad Sci USA 1987b; 84: 2975-2979. Stead RH, Dixon MF, Bramwell NH, Riddell RH, Bienenstock J. Mast cells are closely apposed to nerves in the human gastrointestinal mucosa. Gastroenterology 1989; 97: 575-579. Bar-Shavit Z, Goldman R, Stabinsky Y, Gottlieb P, Fridkin M, Teichborg VI, Blumberg S. Evidence of phagocythosis m a newly found activity of substance P residing in its
377
16.
17. 18. 19. 20.
21. 22. 23.
24. 25.
26.
27.
28.
29. 30. 31. 32. 33. 34. 35.
N-terminal tetrapeptide sequence. Biochem Biophys Res Com 1980; 94: 1445-1449. Hafstrom I, Gyllenhammar H, Palmblad J, Ringertz B. Substance P activates and modulates neutrophil oxidative metabolism and aggregation. J Rheumatol 1989; 16: 1033-1038. Hartung HP, Walters K, Tonka KV. Substance P: binding properties and studies on cellular responses in guinea-pig macrophages. J Immunol 1986; 136: 3856-3861. Ansel JC, Brown JR, Payan DG, Brown MA. Substance P selectively activates TNF-alpha gene expression in murine mast cells. J Immunol 1993; 150: 4478-85. Barnes JP, Chung FK, Page, CP. Inflammatory mediators in asthma. Pharmacol Rev 1988; 40: 49-63. Laurenzi MA, Person MA, Dalsgaard CJ, Haegerstrand A. The neuropeptide substance P stimulates production of interleukin 1 in human blood monocytes. Scand J Immunol 1990; 31: 529-533. Moore TC, Lami JL, Spruck SU. Substance P increases lymphocyte traffic and lymph node flow through peripheral lymph nodes of sheep. Immunology 1989; 67: 109-114. Stanisz AM, Bienenstock J, Agro A. Neuromodulation of mucosal immunity. Regional Immunol 1989; 2:414-418. Stanisz AM, Befus D, Bienenstock J. Differential effects of vasoactive intestinal peptide, substance P and somatostatin on immunoglobulin synthesis and proliferation by lymphocytes from Peyer's patches, mesenteric lymph nodes and spleen. J Immunol 1986; 136: 152-156. Scicchitano R, Bienenstock J, Stanisz AM. In vivo immunomodulation by the neuropeptide substance P. Immunology 1988; 63: 733-735. Stanisz AM, Scicchitano R, Dazin P, Bienenstock J, Payan DG. Distribution of substance P receptors on murine spleen and Peyer's patch T and B cells. J Immunol 1987; 139: 749-754. Croitoru K, Ernst PB, Bienenstock J, Padol I, Stanisz AM. Selective modulation of the natural killer activity of murine intestinal intraepithelial leucocytes by the neuropeptide substance P. Immunology 1990; 71: 196-201. Gronblad M, Konttinen YK, Korkala O, Liesi P, Hukkanen M, Plack JM. Neuropeptides in synovium of patients with rheumatoid arthritis and osteoarthritis. J Rheumatol 1988; 15: 1807-1814. Levine JD, Collier DH, Basbaum AI, Moskowitz MA, Helms CA. Hypothesis: the nervous system may contribute to the pathophysiology of rheumatoid arthritis. J Rheumatol 1985; 12: 406-423. Barnes PJ. Neurogenic inflammation and asthma. J Asthma 1992; 29: 165-180. Barnes PJ, Belvisi MG, Rogers DF. Modulation of neurogenic inflammation. Trends Pharmaceut Sci 1990; 11:185-189. Marchall KW, Chiu B, Inmann RD. Substance P and arthritis: analysis of plasma and synovial fluid levels. Arthritis Rheumat 1990; 33: 87-90. Reinshagen M, Patel A, Sottili M, Davis W, Eysselein VE. Regulation of substance P gene expression in experimental colitis. Gastroenterology 1992; 102: A505. Agro A, Stanisz AM. Are lymphocytes target for substance P modulation in arthritis? Semin Arthritis Rheum 1992; 21: 252-258. Levine JD, Clark DH, Devor M, Helms CA, Moskowitz MA, Basbaum AI. Intraneuronal substance P contributes to the severity of experimental arthritis. Science 1984; 32: 369-376. Swain MG, Agro A, Blennerhassett P, Stanisz AM, Collins SM. Increased levels of substance P in the myenteric plexus of Trichinella-infected rats. Gastroenterology 1993;
378
36.
102: 1913-1917. Agro A, Stanisz AM. Depletion of substance P levels reduces intestinal inflammation and restores lymphocyte reactivity to substance P in Trichinella spiralis infected mice. Regional Immunol 1993; 5" 120-129.
New Foundationof Biology Edited by I. Berczi and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
379
Lupus as a Model o f Neuroimmune Interactions
JUDAH A. DENBURG, BORIS SAKIC, HENRY SZECHTMAN and SUSAN D. DENBURG
Lupus Research Group, Departments of Medicine and Psychiatry and Behavioural Neurosciences, McMaster University, Hamilton, Ontario
ABSTRACT Nervous system or "neuropsychiatric" (NP) involvement in SLE has been recognized since the first descriptions by Kaposi (1925), but estimates of its prevalence have varied widely. In the absence of major NP events (eg., stroke, psychosis), the nervous system was presumed to be unaffected. Owing to the multiple "minor" NP symptoms endorsed by SLE patients, including cognitive and mood problems, neuropsychological assessment was introduced into the study of NP-SLE in the 1980s to systematically study brain function in patients both with and without overt NP manifestations. In human SLE, there are now multiple studies documenting a high prevalence of cognitive dysfunction. Mechanisms that lead to abnormal cognition in SLE, however, are not yet fully understood. A prevailing notion has been that autoantibodies which target neuronal or cross-reactive lymphocytic antigens directly cause brain damage and result in cognitive impairment; proof of this direct pathogenic role is lacking, although positive clinical correlations have been found. The potential biological importance of phospholipid antibodies (a marker of a hypercoagulable state) in the development and prediction of cognitive decline in specific domains, has provided a clue to one mechanism (thrombosis) to be targeted therapeutically. Another relevant and informative approach to pathogenesis derives from studies in lupus-prone MRL/lpr mice, who develop an autoimmune-associated behavioural syndrome (AABS), characterized by significant abnormalities in both "emotional" and "cognitive" tasks; some of these abnormal behaviours can be induced by the cytokine, IL-6, and/or be positively associated with lymphoid infiltrates in the brain. In humans, we have shown: the positive effects of immunosuppressive therapies on cognitive function and concomitant reversal of abnormalities on brain imaging by positron emission tomography (PET), and of single patient drug/placebo trials of low dose corticosteroids; the contributions of generalized disease activity, mood problems, pain and fatigue to cognitive dysfunction; the types of cognitive dysfunction that have been documented or proposed to be associated with specific autoantibodies or proinflammatory cytokines; and, the course of cognitive dysfunction over time, including risk factors. Future understanding of the complexity of neuroimmune interactions could utilise information obtained from both human and animal models of NP-SLE.
380
1.
INTRODUCTION
Nervous system or "neuropsychiatric" (NP) involvement in SLE has been estimated to occur in approximately 50% of patients over the course of their disease and is typically diagnosed on the basis of major neurologic and psychiatric events or syndromes, including stroke, seizures, neuropathies, transverse myelitis, organic brain syndrome, psychosis and less consistently, headache and depression [1]. NP-SLE remains a descriptive diagnosis, since proposed etiopathogenetic mechanisms, which include brain autoantibodies, cerebral vasculopathy, and/or cytokine-mediated brain inflammation, require definitive confirmation. The documentation of more subtle abnormalities of central nervous system functioning, as evidenced by cognitive testing, has added to the complexity of the syndrome of NP-SLE.
2.
UTILITY OF NEUROPSYCHOLOGICAL APPROACHES IN NP-SLE
Neuropsychological techniques, which yield data about cognitive function and allow for inferences about the underlying neural substrate, provide a means to study systematically brain involvement in SLE. Studies of cognitive function in SLE have yielded prevalence estimates for quantitatively-defined cognitive dysfunction ranging from 21% to 59% [2, 3]. These studies have involved diverse patient samples, utilized different neuropsychological tests, and have applied different criteria for defining impairment. Nevertheless, they collectively suggest that a sizeable proportion of SLE patients can be shown to have significant cognitive problems, unrelated to disease activity, organ involvement or corticosteroid dose or frequency [2, 3]. In fact, in a series of single-patient drug/placebo trials, using relatively low dose steroids (0.5 mg/kg) in patients without active NP-SLE but typically with cognitive complaints, we documented significant improvement in cognition and/or mood in five of eight patients [4].
3.
RELATIONSHIP OF AUTOANTIBODIES TO COGNITIVE PROFILES IN SLE
The various studies of cognitive function suggest considerable diversity in the type of cognitive problems demonstrated in SLE patients as a whole, as well as in NP subgroups. Problems that have been identified include attention and concentration, various aspects of verbal and nonverbal memory including working memory, verbal fluency/productivity, visuospatial skills, psychomotor speed and cognitive flexibility. 3.1
Cognitive Heterogeneity: Association of Cognitive Deficits with Pathogenetic Mechanisms
The absence of an "SLE pattern" of deficits is consistent with the clinical heterogeneity of NP- SLE and the multiple mechanisms that have been proposed to underlie it [1-3]. Attempts to reduce this heterogeneity have included examining the relationship between cognitive dysfunction and the presence of specific autoantibodies. For example, significant associations have been documented between positivity for neuronal, lymphocytotoxic, and antiphospholipid (aPL) antibodies and an overall designation of cognitive impairment, as well as between sequential changes in neuronal antibodies and cognitive function; negative findings in this area have also been reported, using different criteria for cognitive impairment (reviewed in 3). Of greater interest, from the point of view of subgrouping patients, are the findings of a specific
381
association between lymphocytotoxic antibody positivity and visuospatial deficits [5, 6], and between aPL positivity and a pattern of cognitive dysfunction characterized by difficulties with verbal memory, productivity, and speeded output [7]. While hypotheses regarding the mechanism(s) of cognitive dysfunction can emerge from these cross-sectional studies (for example, that ongoing aPL-related microthrombotic events or vasculopathy can lead to CNS compromise manifested as cognitive dysfunction), longitudinal data are essential for their confirmation. Studies relating antibodies and cognitive function have typically identified antibody status and examined cognitive function in relation to it. Other target mechanisms which could be examined in this manner include cytokines such as IL-6, which has been recently shown to be sufficient to produce behavioural impairment in mice [8] (see below).
4.
NATURAL HISTORY OF COGNITIVE DYSFUNCTION IN SLE
Data on the natural history of cognitive deficits in SLE are limited. Recently, several investigators reported fluctuating cognitive impairment within SLE patients [9, 10]. However, the relatively brief follow-up periods, together with a number of methodological issues leave open the interpretation of their findings. It is quite possible that NP-SLE, with its associated cognitive problems, represents a range of reversible and non-reversible brain abnormalities. In an ongoing longitudinal study of 41 SLE patients without neuropsychiatric involvement (Never NP), we have data showing that current (retest) performance was significantly worse in those who had an event than in those who continued to be Never NP (p < .009), with more positivity for both lupus anticoagulant (p <.01) and anticardiolipin antibodies (p < .03). A more recent assessment of our longitudinal data further emphasizes the predictive value of cognitive impairment in aPL-positive SLE patients in the expression, over time, of psychiatric (almost exclusively mood) disorders in SLE patients [ 11 ].
5.
MOOD AND COGNITION IN SLE
Psychosis and depression are the most frequently cited clinical psychiatric disorders in SLE, although both their frequency and pathophysiological relationship to SLE remain controversial. Support for the idea that depression reflects primary nervous system involvement in SLE comes from studies relating its occurrence to neurologic events or to specific autoantibodies with potential pathogenetic significance, such as anti-ribosomal P antibodies [12]. Examination of the quality of depression in SLE, rather than its prevalence, might yield more clues to underlying mechanisms. In this vein, we have gathered pilot data on depressed SLE outpatients and compared them to non-depressed SLE and non-medical depressed outpatients. The three patient groups were matched for age, premorbid IQ, disease activity (SLE groups), and level of depression (depressed groups). No differences were found between the two depressed groups with respect to average severity ratings on the cognitive, affective, or somatic item subsets, suggesting that the clinical presentation in the two groups may be very similar. In contrast to the similarity in pattern of clinical symptomatology, the SLE depressed patients performed more poorly than the non-medical depressed and/or SLE non- depressed groups on tasks involving sustained mental effort, verbal and nonverbal learning and visuospatial planning [ 13].
These data point to the potential utility of neuropsychological measures for differentiating cognitive deficits associated with depression of possibly diverse etiologies.
382
6.
ANIMAL MODEL OF NP-SLE: THE MRL MOUSE
The availability of an animal model of behavioural changes in SLE has begun to help to resolve some of these issues, in part through identification of immune (or other) mechanisms underlying altered behaviour, utilising two congenic substrains of MRL mice, differing less than 0.1% in their genome. The MRL/MpJ-lpr/lpr (MRL-lpr) substrain, which has a defective fas gene encoded by 1 pr, resulting in disruption of normal cellular apoptosis [14], spontaneously develops an accelerated form of SLE-like disease with an average life span of five to six months. Mice of the congenic MRL/MpJ +/+ (MRL +/+) substrain develop lupus-like symptoms much more slowly, and have a life span of up to two years. The autoimmune disease of MRL mice resembles human lupus in that there is hyperproduction of autoantibodies, including brain-reactive antibodies, antinuclear antibodies, anticardiolipin and anti-Sm antibodies, as well as organ-specific pathology (arthritis, nephritis, dermatitis). Although there exist other autoimmune murine strains (NZB, NZB/W and BXSB mice), the availability of two congenic MRL substrains permits comparisons between diseased and non-diseased groups which are relatively unconfounded by the influence of genetic background [15].
7.
BRAIN PATHOLOGY IN MRL MICE AND HUMAN NP-SLE
Alexander and colleagues were the first to observe that, while the brains of age-matched MRL +/+ mice appeared normal, mononuclear cell infiltrates in the choroid plexus and meninges were abundant in six-month old MRL-lpr mice (reviewed in 15). However, by twelve months, MRL +/+ mice also developed inflammatory lesions which varied from perivascular cuffing of smaller vessels to occasional invasion and destruction of the walls of larger vessels. The majority of the lesions affected the hippocampus, and in contrast to MRL-lpr mice, there was sparing of the choroid plexus [15]. In both strains, the infiltrates consisted of mononuclear cells and plasma cells. More recently, inflammatory infiltrates were reported in the brains of MRL-Ipr mice younger than six months, while no infiltrates were seen in congenic MRL +/+ controls [16]. Immunohistochemical staining revealed that cells were predominantly CD4 + T-cells, but some CD8 + T-cells were also present, along with cells which bore surface immunoglobulin (presumed to be B cells). We have observed infiltration by both CD45 and CD45R-positive cells in the choroid plexus and parenchyma of 8-11 week old MRL-lpr mice [16]. Other neuropathological changes in MRL-Ipr mice include ventricular enlargement, presence of lymphoid cells, reduced brain weight, as well as significant neurodegenerative changes [15, 16] (see Chapter by Sakic et al). In human SLE, there is a paucity of controlled studies of the nature of brain pathology in relation to clinical manifestations of disease. Some of the classic histopathological findings that have been reported [17] reveal combinations of microgliosis, microinfarction and fibrin deposition, compatible with the end stages of either a chronic inflammatory or thrombotic process, or both. Thus, an animal model which might relate brain pathology, including aspects of cellular and molecular inflammation, to behaviour, could potentially provide much needed insight into the pathogenesis and treatment of human NP-SLE, and perhaps allow a better understanding of events which might initiate NP-SLE.
383
8.
BEHAVIOUR AND AUTOIMMUNITY IN MRL MICE
The onset of autoimmunity in MRL-lpr mice correlates with a variety of changes in behaviour [ 1, 18], many of which appear to reflect emotional reactivity or depressive-like presentations, as found frequently in human SLE [3, 19]. To establish the usefulness of the MRL strain as a model of behavioural changes found in lupus, it is necessary to demonstrate, first, that the behaviour of mice with lupus is different from the behaviour of mice without the disease, and second, that any behavioural difference is a specific effect of the autoimmune process. The first issue was addressed by comparing the performance of MRL-lpr and MRL +/+ mice on a variety of tests, ranging from those that measure spontaneous activity to those that examine competence in learning and memory. The second issue was addressed by correlating the severity of autoimmune disease with behavioural change, and by using therapeutic agents to arrest progress of autoimmune disease and prevent appearance of lupus-related behavioural changes. We compared the performance of lupus-prone MRL-lpr mice with that of their congenic MRL +/+ controls on several tests, at three different time points: prior to onset of recognizable (serologic) disease in MRL-lpr mice (4-6 weeks of age); at the time of appearance of serologic autoimmunity but before the development of overt pathology (8-11 weeks of age); and during florid disease (12-18 weeks) [18, 20]. With disease onset, MRL-lpr mice not only locomote less, but also explore a novel environment differently [18]. In addition, MRL-lpr mice are slower to approach a novel object and spend less time exploring it than MRL +/+ mice; this substrain difference is present only when there is evidence of serologic autoimmunity in MRL-lpr mice [15]. In all, there is ample evidence to satisfy the criterion of substantial behavioural differences between MRL-lpr and MRL +/+ substrains [15]. Further evidence which suggests that the behavioural profile of MRL-lpr mice is the result of an autoimmune disease process includes: studies showing that the behaviour of MRL-lpr mice changes contemporaneously with emergence of serologic autoimmunity [18]; studies that demonstrate that within the MRL-lpr substrain there is a relationship between the severity of the autoimmune disease process and the magnitude of the behavioural effect [ 16, 21 ]; and, studies relating behaviour to cellular infiltration in the brain, and their parallel reversal on treatment with the immunosuppressive drug, cyclophosphamide (CY) [15, 16]. CY was successful in blocking the appearance of serologic signs of autoimmunity in MRL-lpr mice, in significantly diminishing the cellular infiltrate in MRL-lpr brain and in eliminating the substrain difference in two distinct behavioural tests [22].
.
AUTOIMMUNITY-ASSOCIATED BEHAVIOURAL SYNDROME (AABS) IN MRL-LPR MICE: A DEPRESSIVE STATE?
Most of the results discussed above are consistent with the hypothesis that AABS reflects a depressive-like behavioural state, which has features resembling those induced by repeated, inescapable stress in rodents. For example, chronic exposure to inescapable stress results in decreased spontaneous activity, altered exploration of a novel environment, increased floating in the forced swim test, perseveration of a learned response, impaired exploration of the open arms in the plus maze or a novel object, submissive behaviour in the resident-intruder test and attenuation in intake of a palatable sucrose solution [15, 23, 24]. All of these deficits are observed from 5 to 20 weeks of age in autoimmune MRL-lpr mice [18, 22], suggesting a common mechanism for the effects of external stressors and autoimmunity. Some of AABS
384
appears consistent with behaviours observed as a result of changes to the hippocampus during repeated stress [25], which may relate to alterations in neuroendocrine function.
10.
IL-6 AND AABS IN MRL MICE
Several pro-inflammatory or immunologically active cytokines, measured either systemically or (rarely) within the neural substrate, have been related to SLE and/or NP-SLE, including: IL-1, IL-2/IL-2R, IL-3, IL-4, IL-6, IL-8, IFN-et, IFN-T and TNF-ct [26-28]. Many of these cytokines can be expressed in or act upon neuronal or glial cell populations and influence behaviour [29]. Of potential relevance regarding a primary immune-mediated mechanism in lupus-related depression is our recent demonstration of the role of systemic (i.e., present in serum) IL-6, a pro-inflammatory cytokine and B-cell growth factor, in the development of behaviours suggesting altered emotional reactivity in lupus-prone MRL mice [8], as well as indicators of chronic illness such as decreased food and water intake [30]. Some of these behaviours, observed spontaneously in IL-6 hyper-secreting MRL/lpr mice, could be elicited using an adenoviral vector to transiently increase gene expression for IL-6 in the systemic circulation of naive (i.e., non-autoimmune) C3H or MRL +/+ mice [8, 30]; "anhedonic" behaviour could be ameliorated by immunosuppressive therapy which ablates serum IL-6 levels [8, 30]. Given that one of the earliest immune abnormalities in MRL mice is overproduction of IL-6, which can stimulate autoantibody production secondarily, we have postulated a primary role for systemic IL-6 in the AABS of these mice [8, 30]. These findings buttress a growing literature documenting direct injurious effects of IL-6 on neuronal populations in vivo: effects on cerebellar function in IL-6 transgenic, non-lupus mice [31]; the production of IL-6 within the neural substrate by both astrocytes and neuronal populations [29, 31-33]; the regulatory roles of IL-6 in neuroendocrine and neurotransmitter pathways [31, 34]; and, elevations of IL-6 in the blood and cerebrospinal fluid [27, 35] of patients with NP-SLE. Recent studies have identified higher serum levels of IL-6 in non- medical depressed patients and in subsets' of these patients, for example, those non-responsive to antidepressant therapy [36]. While elevations of peripheral blood monocyte production and gene expression of IL-6 have been noted in human SLE, in association with activation of the immune system, neither systemic nor central nervous system IL-6 has been studied with respect to the presence of major depression in human SLE. IL-6, present systemically, may provide ongoing, chronic stimulation of both B-cells and the neural substrate, leading to concurrent autoantibody production and behavioural changes; specifically, systemic IL-6 up regulation may be sufficient to trigger (directly or indirectly) a primary, autoimmunity-related mood disorder. The weight of available evidence thus points to IL-6 as an important factor in AABS, and possibly in NP-SLE as well.
REFERENCES 1. 2. 3.
Denburg JA, Denburg SD, Carbotte RM, Sakic B, Szechtman H. Nervous system lupus: pathogenesis and rationale for therapy. Scand. J Rheumatol 1995; 24: 263-273. Carbotte RM, Denburg SD, Denburg JA. Prevalence of cognitive impairment in systemic lupus erythematosus. J Nerv Ment Dis 1986; 174: 357-364. Denburg SD, Denburg JA. Cognitive dysfunction in systemic lupus erythematosus. In: Lahita RG, editor. Systemic Lupus Erythematosus. 3 ed. San Diego: Academic Press,
385
.
.
.
.
10. 11.
12.
13. 14.
15. 16. 17.
18. 19. 20.
21. 22.
1999. p. 611-629. Denburg SD, Carbotte RM, Denburg JA. Corticosteroids and neuropsychological functioning in patients with systemic lupus erythematosus. Arthritis Rheum 1994; 37:1311-1320. Denburg SD, Behmann SA, Carbotte RM, Denburg JA. Lymphocyte antigens in neuropsychiatric systemic lupus erythematosus: Relationship of lymphocyte antibody specificities to clinical disease. Arthritis Rheum 1994; 37: 369-375. Denburg SD, Carbotte RM, Long AA, Denburg JA. Neuropsychological correlates of serum lymphocytotoxic antibodies in systemic lupus erythematosus. Brain Behav Immun 1988; 2: 222-234. Denburg SD, Carbotte RM, Ginsberg JS, Denburg JA. The relationship of antiphospholipid antibodies to cognitive function in patients with systemic lupus erythematosus. J Int Neuropsychol Soc 1997; 3: 377-386. Sakic B, Szechtman H, Braciak T, Richards C, Gauldie J, Denburg JA. Reduced preference for sucrose in autoimmune mice: a possible role of interleukin-6. Brain Res Bull 1997; 44: 155-165. Hay EM, Huddy A, Black D, Mbaya P, Tomenson B, Bernstein RM, Holt PJL, Creed F. A prospective study of psychiatric disorder and cognitive function in systemic lupus erythematosus. Ann Rheum Dis 1994; 53: 298-303. Hanly JG, Fisk JD, Sherwood G, Eastwood B Clinical course of cognitive dysfunction in systemic lupus erythematosus. J Rheumatol 1994; 21: 1825-1831. Boulos P, Denburg JA, Goldsmith CH, Stewart K, Denburg SD. The predictive value of neurocognitive deficits for neuropsychiatric events in antiphospholipid antibody positive lupus patients. J Rheumatol 2000; in press (abstract). Bonfa E, Golombek SJ, Kaufman LD, Skelly S, Weissbach H, Brot N, Elkon KB. Association between lupus psychosis and anti-ribosomal P protein antibodies. N Engl J Med 1987; 317:265-271. Denburg SD, Carbotte RM, Denburg JA. Cognition and mood in SLE: evaluation and pathogenesis. Ann N Y Acad Sci 1997; 823: 44-59. Watanabe-Fukunaga R, Brannan CI, Copeland NG, Jenkins NA, Nagata S Lymphoproliferation disorder in mice explained by defects in Fas antigen that mediates apoptosis. Nature 1992; 356: 314-317. Sakic B, Szechtman H, Denburg JA. Neurobehavioral alterations in autoimmune mice. NeuroSci Biobehav Rev 1997; 21: 327-340. Farrell M, Sakic B, Szechtman H, Denburg JA. Effect of cyclophosphamide on leukocytic infiltration in the brain of MRL/lpr mice. Lupus 1997; 6: 268-274. Ellis SG, Verity MA Central nervous system involvement in systemic lupus erythematosus: a review of neuropathologic findings in 57 cases, 1955-1977. Semin Arthritis Rheum 1979; 8: 212-221. Sakic B, Szechtman H, Keffer M, Talangbayan H, Stead R, Denburg JA. A behavioral profile of autoimmune lupus-prone MRL mice. Brain Behav Immun 1992; 6: 265-285. Carbotte RM, Denburg SD, Denburg JA. Cognitive deficit associated with rheumatic diseases: neuropsychological perspectives. Arthritis Rheum 1995; 38: 1363-1374. Sakic B, Szechtman H, Talangbayan H, Denburg S, Carbotte R, Denburg JA. Behavior and immune status of MRL mice in the postweaning period. Brain Behav Immun 1994; 8: 1-13. Millward JM. T-lymphocyte infiltration into brains of MRL-lpr mice: a model for systemic lupus erythematosus. 1999. Biology Senior Thesis, McMaster University. Sakic B, Denburg JA, Denburg SD, Szechtman H. Blunted sensitivity to sucrose in
386
23.
24.
25.
26.
27.
28. 29.
30. 31.
32.
33. 34.
35.
36.
autoimmune MRL-lpr mice: a curve-shift study. Brain Res Bull 1996; 41: 305-311. Papp M, Willner P, Muscat R. An animal model of anhedonia: attenuation of sucrose consumption and place preference conditioning by chronic unpredictable mild stress. Psychopharmacology 1991; 104: 255-259. Monleon S, D'Aquila P, Parra A, Simon VM, Brain PF, Willner P. Attenuation of sucrose consumption in mice by chronic mild stress and its restoration by imipramine. Psychopharmacology 1995; 117: 453-457. McEwen BS, Gould EA, Sakai RR. The vulnerability of the hippocampus to protective and destructive effects of glucocorticoids in relation to stress. Br J Psychiatry Suppl 1992; 18-23. Frei K, Leist TP, Meager A, Gallo P, Leppert D, Zinkernagel RM, Fontana A. Production of B cell stimulatory factor-2 and interferon ~ in the central nervous system during viral meningitis and encephalitis: evaluation in a murine model infection and in patients. J Exp Med 1988; 168: 449-453. Hirohata S, Miyamoto T Elevated levels of interleukin-6 in cerebrospinal fluid from patients with systemic lupus erythematosus and central nervous system involvement. Arthritis Rheum 1990; 33: 644-649. Shiozawa S, Kuroki Y, Kim M, Hirohata S, Ogino T. Interferon-alpha in lupus psychosis. Arthritis Rheum 1992; 35: 417-422. Norris JG, Tang LP, Sparacio SM, Benveniste EN. Signal transduction pathways mediating astrocyte IL-6 induction by IL-1 beta and tumor necrosis factor-alpha. J Immunol 1994; 152:841-850. Sakic B, Szechtman H, Gauldie J, Denburg JA. Behavioral effects of infection with IL-6 adenovector. Brain Behav Immun 2000; (in press). Campbell IL, Abraham CA, Masliah E, Kemper P, Inglis JD, Oldstone MB, Mucke L. Neurologic disease induced in transgenic mice by cerebral overexpression of interleukin 6. Proc Natl Acad Sci USA 1993; 90: 10061-10065. Ringheim GE, Burgher KL, Heroux JA. Interleukin-6 mRNA expression by cortical neurons in culture: evidence for neuronal sources of interleukin-6 production in the brain. J Neuroimmunol 1995; 63:113-123. Spangelo BL, MacLeod RM, Isakson PC. Production of interleukin-6 by anterior pituitary cells in vitro. Endocrinology 1990; 126: 582-586. Lyson K, McCann SM. Alpha-melanocyte-stimulating hormone abolishes IL-1- and IL-6- induced corticotropin-releasing factor release from the hypothalamus in vitro. Neuroendocrinology 1993; 58:191-195. Hagiwara E, Gourley MF, Lee S, Klinman DM. Disease severity in patients with systemic lupus erythematosus correlates with an increased ratio of interleukin-10: Interferongamma-secreting cells in the peripheral blood. Arthritis Rheum 1996; 39: 379-385. Sluzewska A, Rybakowski JK, Laciak M, Mackiewicz A, Sobieska M, Wiktorowicz K. Interleukin-6 serum levels in depressed patients before and after treatment with fluoxetine. Ann N Y Acad Sci 1995; 762: 474-476.
New Foundation of Biology Edited by I. Berczi and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
387
The Pathogenesis of Encephalitis
TREVOR OWENS 1, ELISE H. TRAN 1, MINA HASSAN-ZAHRAEE 1, ALICIA BABCOCK 1, MICHELLE L. KRAKOWSKI 1, SYLVIE FOURNIER 2, MICHAEL B. JENSEN ~' 3 and BENTE FINSEN 3. 'Montreal Neurological Institute, eUniversit( de Sherbrooke, 30dense Universitet, Denmark.
ABSTRACT One of the most fundamental neuroimmune interactions is that involving immune responses in and against the brain. Although the CNS is immunologically-privileged relative to other organs, activated T lymphocytes are known to cross the blood-brain barrier. Entry of virusspecific T cells, usually a host-protective event, can induce encephalitis. The pathology of viral encephalitis is associated with inflammatory (Thl) immune responses against infected cells, such as in Theiler's virus infection of oligodendrocytes. Myelin reactivity can occur as a consequence of epitope spreading from anti-viral responses. Myelin-specific CD4+ T cells induce autoimmune encephalomyelitis. The inflammatory, demyelinating pathology of experimental autoimmune encephalomyelitis (EAE) is similar in many respects to that of Multiple Sclerosis, including axonal damage. We find that naive T cells can enter the CNS during EAE, and may become activated there if costimulator molecules such as B7 are expressed on MHC II+ microglia. Indeed, B7 is known to be induced by viral infection, thus linking infection to CNS autommunity. Although initiated by infiltrating T cells, many of the inflammatory mediators detected in the CNS in MS or EAE are produced by CNS-resident glial cells. Interferon-gamma (IFN~,), an immune cytokine not normally expressed in the adult CNS, can induce glial cells to produce a variety of mediators, including tumor necrosis factor (TNF) and nitric oxide, that are cytopathic for oligodendrocytes in vitro. TNF is also implicated in repair/regenerative responses, in vivo. We find that IFN7 amplified but did not affect the kinetics of microglial TNF production, induced in response to axonal lesioning in MBP promoter/IFN~, transgenic mice. TNF, whether induced by EAE or by axonal damage, was nevertheless produced in IFNT-deficient mice. This indicates that there are endogenous programs of glial response, which are amplified by IFN~,. The macrophage-dominated, perivascular infiltrates that are characteristic of EAE were replaced by a disseminated, invasive neutrophilia in IFN~,-deficient mice, with lethal consequence. The EAE-associated enzyme NOS2, the cytokine interleukin-10, and chemokines MCP-1 and RANTES were undetectable in IFNT-deficient mice with EAE, whereas the neutrophil-attractant chemokines MIP2 and TCA3 became prominent. CNS glia may interact with immune cells via chemokines to redirect further infiltration. Restriction of NOS2 expression to parenchymal glia, in chimeric mice reconstituted with NOS2-/- bone marrow, conferred protection against EAE. Nitric oxide may play distinct roles when made by microglia/macrophages versus astrocytes. Our observations demonstrate
388
the capacity of the CNS to mediate and direct protective and inflammatory responses, and of the immune system to interpret and amplify CNS-derived signals.
1.
INTRODUCTION
The traditional view of the central nervous system (CNS) has been of an immunologicallyprivileged organ, which in pathological circumstances may be invaded by self-reactive immune cells, with deleterious consequence. This 'brain passive, immune active' perspective has guided research on CNS inflammation and autoimmunity for decades. Even the more informed realization that the immune system must exert a surveillance role for protection against CNS infections has maintained the principle of immune mediation in a passively-infected tissue. Recent work from our own and other laboratories have provided a modified perspective, whereby the CNS plays a more active role in what were termed 'immune events'. Encephalitis necessarily involves immune cell entry to the CNS. Because of the pathologic consequences, and given the prevailing perspective, this is frequently termed 'immune invasion'. Immune invasion of the CNS occurs in host-protection when virus-specific T cells that have encountered their cognate antigen initiate the cascades of cytokine and cellular interactions that are characteristic of DTH or Type IV hypersen sitivity responses. These are well-described for immune responses in the CNS against Herpesvirus and Theiler's Murine Encephalitis Virus (TMEV) [1, 2]. These host-protective responses can evolve into pathogenic autoimmune responses via Epitope Spreading (see below). This has been invoked to explain the etiology of Acute Disseminated Encephalomyelitis (ADEM) and Multiple Sclerosis (MS), and can be directly demonstrated in the generation of demyelinating pathology in animal models such as Experimental Auto immune Encephalomyelitis (EAE) [3]. Immune invasion can also be provoked by events within the CNS. In cerebral ischemia the initial hypoxic insult is followed by massive infiltration by waves of inflammatory leukocytes, and the subsequent reperfusion injury contributes significantly to tissue damage and the final infarct [4]. Globoid cell/myeloleukodystrophy illustrates a different etiology, where a metabolic defect in oligodendrocytes induces macrophage infiltration and demyelination, as modeled in the Twitcher mouse [5]. The microglial reactivity in Alzheimer's Disease may represent a similar response to inherent CNS defects [6]. In the latter cases, there is no evidence for T cell infiltration, showing that immune events or their CNS analogues are not necessarily T cell-directed, and can include elements of innate immune responses, as are modeled by the systemic administration of lipopolysaccharide (LPS) [7].
2.
IMMUNE CELL ENTRY TO AND ACTIVATION IN THE CNS
In most of the clinical illustrations of the capacity for entry of immune cells to the CNS, entry is of 'reactive' cells, and represents an immune response to already-initiated or ongoing infection or dysfunction. In these cases, lymphocytes or leukocytes were probably initially activated in lymphoid tissue, via presentation there of antigen draining from cerebral vessels, or from cerebrospinal fluid that drains through the lamina cribosa and down to the deep cervical lymph nodes [8], or of soluble mediators such as cytokines released from sites of damage in the CNS. This situation is modeled in EAE, where induction of myelin-specific CD4+ T lymphocytes occurs in adjuvant-primed lymph nodes. Although EAE is used as an animal model for MS, the etiology of MS is unknown, and may involve either intra-cerebral or
389
peripheral T cell immunization. The question arises then, whether immune reactivity can be induced within the CNS. Similarly, the question whether T cells with novel specificities are induced by antigen presented within the CNS (viz. Epitope Spreading) arises in anti-virus responses, where viruses infect CNS tissue. For that to occur, naive lymphocytes would have to access the CNS. This focuses attention on processes of leukocyte entry to the CNS, barriers to that entry, and the capacity of naive cells to enter and be activated within the CNS. We have carried out a number of studies of these aspects, and this article will attempt to summarize our findings and their relevance to the issues described above. The association between infection and autoimmunity is well-known, although imperfectly understood. One key finding that guides thinking in this area is the phenomenon of Epitope Spreading. This refers to the observation that immune responses in CNS that were initiated by immunization against defined, mono-determinant peptide epitopes evolve with time to include T cell reactivities against other epitopes of the same myelin protein, or even to epitopes of other proteins. In a clinically pertinent example, an anti-viral T cell response broadened to include myelin protein specificity e.g. infection led to autoimmunity [2]. A major dogma that provides a theoretical underpinning for Epitope Spreading is that of Molecular Mimicry. By this model, some individuals, or strains of laboratory animal, contain T cells amongst those with TCR's specific for pathogenic epitopes, whose TCR's cross-reactively recognize self epitopes e.g. in myelin proteins. Induction of a host-protective immune response then inadvertently, but unavoidably, triggers autoimmunity in those individuals. There are by now numerous descriptions of TCR cross-reactivities, e.g. between Borrelia burgdorferi and myelin or neurons, that can explain linkage between infection and autoimmunity [9]. However, other studies show that for induction of autoimmunity, the pathogen must infect the target tissue. Thus, Horwitz et al., report that myocarditis is only induced in mice where virus can productively infect cardiac tissue, whereas a limited infection which induces an immune response was insufficient [10]. This does not negate the concept of cross-reactivity between host-protective and self-reactive TCR's, but introduces an important modification for the overall dogma. This modification bears on the antigen presentation requirements for induction of an immune response. One of first requirements for induction of a response from a pool of naive CD4+ T cells is that antigenic peptides be appropriately presented, by cells expressing MHC II and costimulator ligands (so-called Professional APC). Costimulator ligand expression is induced by viral infection, by bacterial products (e.g. LPS), as well as by other stimuli deriving from already-initiated immune responses [11]. This suggests that the requirement for pathogen to infect the target tissue reflects a need for costimulator induction, as is indeed observed in infected tissues. This further suggests that induction of an autoimmune response is induced by antigen-presenting cells in the target tissue or organ. In the context of CNS-directed autoimmunity, this raises two questions: Are there potential APC's in the CNS which might express B7 to induce an autoimmune response? And, can naive, previously unactivated T cells enter the CNS for induction by B7-positive APC's in the CNS? We have addressed both these questions. In the absence of systems for analysis of spontaneous autoimmunity, even though this is presumed to underlie diseases such as MS, we recourse to study of experimentally-induced CNS inflammation, in the EAE model. In this system, immune cells are deliberately immunized in the periphery and then they migrate to the CNS and initiate cytokine production, accumulate as parenchymal and perivascular infiltrates, and both initiate and mediate neural cytotoxicity [12, 13]. Interest in the entry of non-myelin-specific T cells into the mouse CNS derived from experiments where we analyzed the memory phenotype of infiltrating ovalbumin-specific T
390
cell lines in SJL/J mice. Ovalbumin (OVA) is an antigen which is not expressed in mouse. These analyses complemented our demonstration that the T cells which accumulated in CNS in mice in which EAE was transferred with MBP-specific T cell lines were predominantly of a memory-effector phenotype, arguing for accumulation in the CNS through antigen recognition [14]. Although OVA-specific T cell lines could also enter the CNS, they did not accumulate there, nor did they show a population shift to a memory-effector phenotype, consistent with lack of an antigen stimulus. These experiments, like those performed in other laboratories in the same era, of necessity, used in vitro-cultured T cells, in order to determine their antigen specificity. With the advent of transgenic technology, TCR transgenic mice allowed the experiment to be repeated using naive T cells. Our experimental design was to transfer fluorescently-labelled MBP-specific T cells to mice whose T cells all expressed TCR specific for OVA. This led to the interesting and somewhat unexpected finding that equivalent numbers of OVA-specific and MBP-specific T cells entered the CNS at the initiation of disease. As expected, the OVAspecific T cells expressed a naive phenotype and did not secrete cytokines [ 15]. We assume that if OVA were presented to these T cells, they would have become activated and this would have led to cytokine production and accumulation of even greater numbers of cells. The relative lack of OVA-specific T cells in CNS in our previous experiments likely reflects the fact that in those experiments, there was no ongoing inflammatory response. It may also be the case that use of a TCR transgenic mouse disallowed entry and subsequent activation of T cells of specificities other than the disease-inducing myelin basic protein epitope, e.g. Epitope Spreading, although this remains to be established. The critical point emerging from this experiment was that naive T cells, regardless of specificity, are present at the initiation of CNS inflammation, and are therefore available for recruitment to immune inflammatory responses. This has implications for generation of anti-viral and autoimmune responses in the CNS, in particular, to the consideration of whether the phenomenon of epitope spreading might draw on a pool of naive T cells for extension of the inflammatory repertoire. It also raises questions whether 'opportunistic' immune reactions might be induced whenever inflammation occurs, for instance in a pro-inflammatory milieu such as the ischemic penumbra (see below). The healthy CNS is devoid of MHC expression, and the best-defined costimulator ligands, members of the B7 family, are usually not detectable. Both MHC II and B7 molecules are upregulated in MS and EAE [13, 16]. In EAE this may be considered a consequence of adjuvant+peptide disease induction, whereas the root cause for B7 induction is not known for MS. Interestingly, B7 upregulation has not been described in stroke [17], and indeed autoimmune sequelae are not prominent in ischemia [4]. We are studying whether B7 induction might lead to an autoimmune pathology in CNS. Our guiding hypothesis is described in Figure 1, which attempts to reconcile various observations on encephalitis. These include the often-noted association between infection and autoimmunity, and epitope spreading so that anti-viral or anti-bacterial immune responses have the potential to turn into anti-self responses. Critical aspects that require testing for such a model to have validity include whether endogenous APC's normally express self-antigens, whether there are cells within the healthy CNS that can function as professional APC, and whether they can express MHC or B7. We have obtained data for the first points, in that previous data showed the ability of ex-vivo MNC's, predominantly microglia, to support proliferation of anti-MBP T cell lines, and that this included weak activity of microglia to which no antigen was added in vitro [18]. Analogous observations have been made by Katz-Levy et al., and others [2]. The implication is that if these cells should be induced to express B7, then an anti-MBP response should occur, and we are actively engaged in studying this question, using transgenic over-expression of B7 as a tool.
o~
|
o~
o~
o~ m
m
o~
b-. Z
_
_
"-I
Zl
Normal _
L
_
-
_
_
_
_
_
_
_
_
L
_
-
C
-
-
-
-
-
-
_
_
_
_
-
1
_
_
_
_
r
|
e~
o~
o~
Peptide-specific Myelin peptides T cells on microglia MHC IIhigh B7+ Autoimmune Response o~
m m
y
o~
E
o~
m
Pathologic
_
I I [ I
I I I I I I I I I [ I I I I I [ [ I I
Trauma Infection
_
391
9
+,,a
9
..=
Z
9
Figure 1. B7 expression in CNS determines autoimmune disease.
392
3.
PERIVASCULAR ANTIGEN-PRESENTING CELLS: GATEKEEPERS OF THE CNS
Further observations on the role of antigen-presenting cells, in particular those in the perivascular space, come from experiments in which peripheral macrophages were targeted for depletion by use of cladronate-loaded liposomes. A seminal observation in this simplified description of events was that activity of peripheral macrophages was essential for immune invasion of the CNS in MBP-induced EAE. In macrophage-targeted mice where cladronate-loaded liposomes depleted and/or crippled peripheral phagocytic populations, leukocytes accumulated in the perivascular space, and did not enter the CNS [19]. In these circumstances, there was no induction of cytokines or other mediators within the CNS, despite their expression at the exterior of the glia limitans. This exemplified the critical role for a macrophage or macrophagelike cell in directing immune traffic to the CNS, and for that traffic to direct inflammatory events within the same tissue. In asking whether macrophage subpopulations intrinsic to the CNS contributed to this direction of immune invasion, we have used two approaches. One has been to transfer T cells specific for ovalbumin to mice depleted of macrophages through liposome treatment. In this way we could assess the role of antigen presentation in initiating perivascular accumulation of leukocytes, by making it unavailable. In these experiments there was no accumulation of leukocytes in the perivascular space, a result which taken at face value suggests that antigen presentation is a critical component of the events leading to leukocyte accumulation [16]. It is possible that perivascular macrophages play an important role in immune infiltration of CNS, and were targeted by liposomes. We term this the Gatekeeper Hypothesis. It should be kept in mind that presentation of CNS antigen, whatever its role, cannot be essential for immune cell entry to CNS, given that OVA-reactive T cells may enter under certain circumstances. Perivascular macrophages are a recognised early player in immune inflammation of the CNS, e.g. as shown by the Campbell laboratory in their GFAP promoter IL-3, TNFc~ and IL-12 transgenic mice [20]. They have shown production of the chemokine C10 as an early event in leukocyte invasion to initiate pathology [21]. Interestingly, preliminary data shows that C10 is not produced in liposome-treated mice (Tran, Asensio, Campbell, Owens, unpublished observations). These observations place limits on the extent to which the CNS directs events, and identify requirements for peripheral leukocyte involvement in encephalitic pathology.
.
SELECTIVE EXPRESSION OF MEDIATORS IN CNS VERSUS PERIPHERY-BONE MARROW CHIMERAS AND TRANSGENIC MICE
To address the role of mediators expressed by invading immune cells, versus those that are induced or pre-programmed in the CNS, tissue-specific promoters can be used to drive selective expression in transgenic mice. A number of cytokines, including TNFct, IL-12, IL-3 and IFNT have been over-expressed in CNS using neuronal, astroglial and/or oligodendroglial promoters [16, 20], with the general result that pathology either ensued, or was promoted in cases where spontaneous disease did not occur [22]. In some cases, notably that of IFNT, these findings are at odds with inhibitor or antibody studies [23]. Furthermore, gene disruption by homologous recombination, an effective technique for systemic ablation of function, in socalled knockout mice, has resulted in many cases (e.g. IFNT, iNOS) in enhanced susceptibility or exacerbated EAE, findings that are hard to reconcile with results from transgenic or many of the inhibitor studies [24]. The role of nitric oxide is of particular interest. Nitric oxide inhibitor studies have shown both prevention and exacerbation of EAE, and knockout mice are reported
393
to develop more severe disease than wild-type controls [25]. Our own data show expression of iNOS by astrocytes distal from lesions in EAE in SJL/J mice, which may indicate a protective role [26]. Protection has been inferred from many of the reported studies and it has been suggested that either timing or location of nitric oxide production may determine its final effect. To resolve the confusion in this field, it becomes desirable to selectively express, or prevent expression of, cytokines or soluble mediators in CNS. At present, inducible promoter and tissue-specific knockout technology is in its infancy, so tissue-specific knockouts are not an option for most laboratories. An approach which facilitates investigation has been to use bone marrow chimeras, reconstituting irradiated mice with gene-deficient bone marrow, or reconstituting irradiated knockout mice with intact bone marrow, so obtaining selective expression in those tissues which are either reconstituted, or not, depending on the design of the experiment. Interpretation of such experiments relies on the fact that all bone marrowderived cells express the CD45 marker, and are ablated by irradiation, and that perivascular macrophages but not parenchymal microglia are replaced by bone marrow-derived cells, within the time frame of these experiments. Using this approach, the role of IFN~, in EAE has been examined, and this has led to some new insights on the role of nitric oxide [27]. Interferon-gamma, a classic pro-inflammatory cytokine is one of the primary inducers of iNOS expression. IFN~, is associated with EAE and MS by correlation of levels of expression with disease severity, but curiously, its ablation if anything exacerbates disease [16, 23]. In mice lacking IFN~,, iNOS expression, in CNS or elsewhere, is undetectable [28]. One of the principal effects of IFN• is activation of macrophages, and it has also been shown to activate microglia. In EAE in IFN~,-deficient mice, the CNS is infiltrated by neutrophils rather than macrophages, and the profile of chemokines that are detectable as mRNA in CNS is shifted from one associated with macrophage/monocyte recruitment to one with activity for neutrophils [28]. This identifies IFN~, as a 'director' of immune invasion, promoting (in its presence) a macrophage infiltration with microglial activation. Opinions vary as to whether IFN~, is disease-promoting or protective. Our interpretation of data such as those described is that IFN~, shapes the immune response in CNS, rather than determining it. Willenborg has shown that IFN~, is required for downregulation of EAE, and that this involves induction of iNOS by a macrophage population, which then inhibits T cell proliferation [27]. Because this same effect was noted in chimeric mice made using IFN~,R-/- bone marrow, it was concluded that the inhibition must also occur within the CNS [27]. However, CD45- cells in peripheral tissues, not deriving from bone marrow and so retaining the phenotype of the irradiated recipient mouse e.g. expressing IFN~,R, may also have played a role in those studies. We have also investigated the mechanism of regulation of immune response in the CNS, focusing initially on nitric oxide, a mediator whose production is characteristic of activated macrophages and which has been implicated as both pro-inflammatory and protective (for review, see [25]). We have used irradiation, bone marrow chimeras to selectively express iNOS (NOS2) in non-bone marrow-derived cells. In chimeric mice, EAE is significantly delayed compared to either intact mice, or mice in which NOS2 has been ablated through gene knockout (Tran, Hassan-Zahraee, Owens, unpublished). The mechanism whereby this selective expression of an enzyme confers protection, compared to the situation in its global presence or total absence, may relate to where iNOS was selectively expressed. The spleen has been identified as a site for regulation of T cell responses in mice with knockout of the enzyme NADPH oxidase, which cannot make superoxide. In these mice, the spleen becomes an inhibitory environment for T cell responses [29]. It is possible that iNOS-expressing
394
cells also play a significant role in such inhibition, although macrophages cannot be responsible in those bone marrow chimeras where iNOS-expressing CD45+ cells are depleted.
5.
INTERFERON-GAMMA, THE ONLY IMMUNE-SPECIFIC CYTOKINE?
Thus far, the only mediator associated with immune responses, inflammation or encephalitis in CNS that is not known to be produced by CNS-resident cells is Interferon-gamma. There is some dispute on this point, and 2 labs have reported production of IFN~, by neurons and microglia [30, 31 ]. Only one report describes IFN~, expression (by immunostaining) in situ [32]. The majority view is that IFN~, is not a natural mediator within the CNS, and that its expression reflects immune pathology. The cell types that are best known to make IFN~, are T cells, both CD4+ and CD8+, and Natural Killer cells. Reports of IFN~,-positive macrophages have become more numerous in recent years [33] and it is possible that such cells may contribute to IFN~, titers within the CNS, although that would still be an inflammation-associated event. Our own analyses have failed to detect IFN~, in the adult mouse CNS, including at sites of axonal lesioning. We assume that this cytokine is indeed an indicator of immune involvement. That raises questions about why glial cells in CNS express IFN~,-receptors, whether this represents an evolutionarily selected mechanism to ensure response in cases of infectioninduced immune infiltration, or whether there is a deeper meaning. It is notable that the nitric oxide production which is implicated in inhibition of encephalitogenic T cell responses is itself dependent on IFN~,, and also that chemokine profiles in CNS and in the periphery are profoundly influenced by IFN~,. At present this issue is unresolved.
6.
ENDOGENOUS PROGRAMS OF CNS RESPONSE
Although there are data for direct activation of microglia by IFN~,, in vitro evidence suggests that IFN~, is only optimal as a macrophage stimulant in concert with other ligands, especially LPS or interaction with previously-activated T cells. Consistent with this, we find that overexpression of IFN~, in CNS, in MBP promoter/IFN~, transgenic mice, does not in itself drive a microglial response [22, 34]. However, it does amplify TNFc~ production in response to axonal lesioning. Importantly, this amplification did not include any affect on the kinetics of the TNF response, which differed from those of the reactive MBP-promoter-driven IFN~, response [34]. This identifies TNFc~ production as an endogenous program of glial response to CNS injury, which IFN~, amplifies rather than induces. The nature of the inducing stimulus for glial reactivity in axonal lesions in the hippocampus is unknown, but it is of interest that this is an immune-independent event. The implication of this is that the TNFc~ production, and in our transgenic experiment, its amplification by IFN~,, are programs of CNS responses that occur independently of immune invasion. This is significant, because exactly similar responses are detectable and indeed define the pathology of encephalitis, raising the question of how much these pathological indices are endogenous CNS responses, induced by the immune system, rather than responses of the invading immune cells themselves. It is also of interest whether analogous programs of response exist with respect to other mediators, such as chemokines. These small molecules play a critical role in directing leukocyte migration, but have also been shown to be produced in the CNS in response to injury or trauma, in some cases without any discernible immune involvement [35]. Both fractalkine and its CXCR3 receptor are upregulated in the rat facial nucleus after axonal lesioning [36].
395
Other chemokines such as RANTES and MCP-1 are also implicated in glial reactivity, and may be implicated in injury responses too [35].
7.
CONCLUSIONS
The CNS is accessible to naive T lymphocytes and this makes it possible that B7 and MHC II-positive APC's within either perivascular space or the CNS parenchyma may induce autoimmune responses. Entry of immune cells is regulated by a variety of mediators, including TNFc~ and NO, most if not all of which can be produced endogenously and so the CNS plays a role in regulating immune invasion. There are endogenous programs of CNS response to injury which are amplified by IFNy, the only cytokine that is immune-specific, arguing for immune-CNS interplay at the level of response amplification, distinct from induction. These processes are all implicated in encephalitis and, through Epitope Spreading, can initiate autoimmune pathology. Chemokines are produced by cells within the CNS in the absence of infiltration by immune cells, further supporting a role for the CNS in directing immune infiltration which leads to extension of host-protective immune responses to frank autoimmunity, demyelination and neuroautoimmune pathology.
8.
ACKNOWLEDGEMENTS
Research in Trevor Owens' laboratory was funded by the Multiple Sclerosis Society of Canada, and the Medical Research Council of Canada. Research in Bente Finsen's laboratory was funded by the Danish MS Society.
REFERENCES 1.
2.
3.
4. 5.
6. 7.
Soldan SS, Berti R, Salem N, Secchiero P, Flamand L, Calabresi PA, Brennan MB, Maloni HW, McFarland HF, Lin HC, Patnaik M, Jacobson S. Association of human herpes virus 6 (HHV-6) with multiple sclerosis: increased IgM response to HHV-6 early antigen and detection of serum HHV-6 DNA. Nat Med 1997; 3:1394. Katz-Levy Y, Neville KL, Girvin AM, Vanderlugt CL, Pope JG, Tan LJ, Miller SD. Endogenous presentation of self myelin epitopes by CNS-resident APCs in Theiler's virus-infected mice. Clin Invest 1999; 104: 599. Tuohy VK, Yu M, Yin L, Kawczak JA, Kinkel RP. Spontaneous regression of primary autoreactivity during chronic progression of experimental autoimmune encephalomyelitis and multiple sclerosis. J Exp Med 1999; 189: 1033. Iadecola C. Bright and dark sides of nitric oxide in ischemic brain injury. Trends. NeuroSci 1997; 20: 132. Matsushima GK, Taniike M, Glimcher LH, Grusby MJ, Frelinger JA, Suzuki K, Ting JP. Absence of MHC class II molecules reduces CNS demyelination, microglial/macrophage infiltration, and twitching in murine globoid cell leukodystrophy, ell 1994; 78: 645. McGeer PL, McGeer EG. Inflammation of the brain in Alzheimer's disease: implications for therapy. J Leukoc Biol 1999; 65: 409. Nadeau S, Rivest S. Role of Microglial-Derived Tumor Necrosis Factor in Mediating CD 14 Transcription and Nuclear Factor kappa B Activity in the Brain during Endotoxemia.
396
.
.
10.
11. 12. 13. 14.
15. 16. 17.
18.
19.
20.
21.
22.
23. 24.
J NeuroSci 2000; 20: 3456. Cserr HF, Knopf PM. Cervical lymphatics, the blood-brain barrier and the immunoreactivity of the brain: a new view. Immunol Today 1992; 13: 507. Hemmer B, Gran B, Zhao Y, Marques A, Pascal J, Tzou A, Kondo T, Cortese I, Bielekova B, Straus SE, McFarland HF, Houghten R, Simon R, Pinilla C, Martin R. Identification of candidate T-cell epitopes and molecular mimics in chronic Lyme disease. Nat Med 1999; 5: 1375. Horwitz MS, La Cava A, Fine C, Rodriguez E, Ilic A, Sarvetnick N. Pancreatic expression of interferon-gamma protects mice from lethal coxsackievirus B3 infection and subsequent myocarditis. Nat Med 2000; 6: 693. Chambers CA, Allison JP. Costimulatory regulation of T cell function. Curr. Opin Cell Biol 1999; 11: 203. Martin R, McFarland HF, McFarlin DE. Immunological aspects of demyelinating diseases. Annu Rev Immunol 1992; 10: 153. Owens T, Sriram S. The immunology of multiple sclerosis and its animal model, experimental allergic encephalomyelitis. Neurol Clin 1995; 13:51. Zeine R, Owens T. Direct demonstration of the infiltration of murine central nervous system by Pgp-1/CD44high CD45RB (low) CD4+ T cells that induce experimental allergic encephalomyelitis. J Neuroimmunol 1992; 40: 57. Krakowski M1, Owens T. Naive T lymphocytes traffic to inflamed central nervous system, but require antigen recognition for activation. Eur J Immunol 2000; 30: 1002. Owens T, Tran E, Hassan-Zahraee M, Krakowski M. Immune cell entry to the CNS--a focus for immunoregulation of EAE. Res Immunol 1998; 149: 781. Windhagen A, Newcombe J, Dangond F, Strand C, Woodroofe MN, Cuzner ML, Hailer DA. Expression of costimulatory molecules B7-1 (CD80), B7-2 (CD86), and interleukin 12 cytokine in multiple sclerosis lesions. J Exp Med 1995; 182: 1985. Krakowski ML, Owens T. The central nervous system environment controls effector CD4+ T cell cytokine profile in experimental allergic encephalomyelitis. Eur J Immunol 1997; 27: 2840. Tran EH, Hoekstra K, van Rooijen N, Dijkstra CD, Owens T. Immune invasion of the central nervous system parenchyma and experimental allergic encephalomyelitis, but not leukocyte extravasation from blood, are prevented in macrophage-depleted mice. J Immunol 1998; 161 : 767. Campbell IL, Stalder AK, Akwa Y, Pagenstecher A, Asensio VC. Transgenic models to study the actions of cytokines in the central nervous system. Neuroimmunomodulation 1998; 5: 126. Asensio VC, Lassmann S, Pagenstecher A, Steffensen SC, Henriksen SJ, Campbell IL. C10 is a novel chemokine expressed in experimental inflammatory demyelinating disorders that promotes recruitment of macrophages to the central nervous system. Am J Pathol 1999; 154:1181. Renno T, Taupin V, Bourbonniere L, Verge G, Tran E, De Simone R, Krakowski M, Rodriguez M, Peterson A, Owens T. Interferon-gamma in progression to chronic demyelination and neurological deficit following acute EAE. Mol Cell NeuroSci 1998; 12: 376. Popko B, Corbin JG, Baerwald KD, Dupree J, Garcia AM. The effects of interferongamma on the central nervous system. Mol Neurobiol 1997; 14:19. Drakesmith H, Chain B, Beverley P. How can dendritic cells cause autoimmune disease? Immunol Today 2000; 21: 214.
397
25. 26.
27.
28. 29.
30.
31. 32.
33.
34. 35.
36.
Willenborg DO, Staykova MA, Cowden WB. Our shifting understanding of the role of nitric oxide in autoimmune encephalomyelitis: a review. J Neuroimmunol 1999; 100:21. Tran EH, Hardin-Pouzet H, Verge G, Owens T. Astrocytes and microglia express inducible nitric oxide synthase in mice with experimental allergic encephalomyelitis. J Neuroimmunol. 1997; 74: 121. Willenborg DO, Fordham SA, Staykova MA, Ramshaw IA, Cowden WB. IFN-gamma is critical to the control of murine autoimmune encephalomyelitis and regulates both in the periphery and in the target tissue: a possible role for nitric oxide. J Immunol 1999; 163: 5278. Tran E.H., Prince EN, Owens T. IFN-gamma shapes immune invasion of the central nervous system via regulation of chemokines. J Immunol 2000; 164: 2759. van der Veen RC, Hinton DR, Incardonna F, Hofman FM. Extensive peroxynitrite activity during progressive stages of central nervous system inflammation. J Neuroimmunol 1997; 77: 1. Neumann H, Schmidt H, Wilharm E, Behrens L, Wekerle H. Interferon gamma gene expression in sensory neurons: evidence for autocrine gene regulation. J Exp Med 1997; 186: 2023. De Simone R., G. Levi G, F. Aloisi F. 1998. Interferon gamma gene expression in rat central nervous system glial cells. Cytokine 10:418. Olsson T, Kristensson K, Ljungdahl A, Maehlen J, Holmdahl R, Klareskog L. Gammainterferon-like immunoreactivity in axotomized rat motor neurons. J NeuroSci 1989; 9: 3870. Klocker U, Schultz U, Schaller H, Protzer U. Endotoxin stimulates liver macrophages to release mediators that inhibit an early step in hepadnavirus replication. J Virol. 2000; 74: 5525. Jensen MB, Hegelund IV, Lomholt ND, Finsen B, Owens T. IFNgamma Enhances Microglial Reactions to Hippocampal Axonal Degeneration. J. NeuroSci 2000; 20: 3612. Mennicken F, Maki R, de Souza EB, Quirion R. Chemokines and chemokine receptors in the CNS: a possible role in neuroinflammation and patterning. Trends. Pharmacol. 1999; 20: 73. Harrison JK, Jiang Y, Chen S, Xia Y, Maciejewski D, McNamara RK, Streit WJ, Salafranca MN, Adhikari S, Thompson DA, Botti P, Bacon KB, Feng L. Role for neuronally derived fractalkine in mediating interactions between neurons and CX3CR1expressing microglia. Proc. Natl. Acad. Sci USA 1998; 95: 10896.
This Page Intentionally Left Blank
VI.
CLINICAL NEUROIMMUNE BIOLOGY
This Page Intentionally Left Blank
New Foundation of Biology
401
Edited by Berczi I and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
Introduction
ISTVAN BERCZI
Department of Immunology, Faculty of Medicine, The University of Manitoba, Bannatyne Campus, 32-795 McDermot Avenue, Winnipeg, Manitoba R3E OW3, Canada
In many ways this section can be considered a continuum from the preceding one, extending further an investigation into the evidence for linkage between the basic science of neuroimmunology and clinical disease. Many of the ideas will kindle a spark of d~j~ vu, though there are in addition some more provocative points made, as in the chapters by Singh (on autism) and MacNeil et al., on spinal cord injury. In the first chapter of this section, Rapaport gives a brief overview on growth hormone (GH). He discusses observations that immune functions are not altered significantly in GH deficient children. During GH treatment a significant decrease occurs in B cells, which is transient. Immunoglobulin levels remain normal. In some patients a substantial transient decrease in PHA responsiveness as well as a T-helper/suppressor ratio has been described. Lymphocytes of GH-deficient children produce normal amounts of GH. The role of prolactin in systemic lupus erythematosus (SLE) is presented by Warrington et al. SLE is predominant in females and is modulated by estrogens and androgens. Suppression of pituitary prolactin by bromocriptine delayed the onset of disease in the NZB/W F1 murine model of SLE, whereas hyperprolactinemia induced by pituitary grafts accelerated mortality. In SLE patients elevated PRL levels were demonstrated in certain subgroups, but this elevation did not always correlate with increased disease activity. Unstimulated peripheral blood mononuclear cells from lupus patients contain a prolactin-like protein at significantly higher levels than do unstimulated cells from normal donors. The immunotherapy of cancer is still experimental. Here Nagy and co-workers present evidence indicating that the triphenyl-ethyl antiestrogens, tamoxifen or toremifene, enhance the cytotoxic effect of killer cells against syngeneic tumor targets. These drugs sensitise the target cells to the apoptotic signals delivered by killer cells trough the Fas-FasL pathway. The membrane attack pathway of cell killing is also enhanced. Antiestrogens enhanced the therapeutic effect of killer cells in syngeneic murine tumor-host systems. Moreover, the cytotoxic effect of autologous killer cells on human ovarian and lung carcinoma targets was also significantly enhanced by combination treatment with IL-2, tamoxifen or toremifene and interferon-alpha. Ford discusses the evidence, including his own data, indicating that the menstrual cycle and reproductive hormones have a significant influence on clinical symptoms in asthma. Premenstrual and menstrual exacerbation is frequent and oral contraceptives attenuate these changes. Treatment with GnRH antagonist analogues has a similar effect. Leukotrienes are important mediators of exacerbation. Changes in asthma severity also occur during pregnancy and at menopause. The age-adjusted incidence is lowest in postmenopausal woman who never
402
received hormone replacement therapy. The use of conjugated estrogens enhances the risk of asthma. The previous use of oral contraceptives is associated with a small increase in risk of developing asthma. Singh presents the neuroimmunopathogenesis of autism. The data indicate that autistic children have elevated antibody levels against measles virus and against myelin basic protein. On this basis it was suggested that autism is the result of an autoimmune response to the brain triggered by an atypical measles infection. These observations concur very well with the epitope spreading hypothesis discussed by Owen in the previous chapter. The disregulation of immune function in denervated areas of skin and in the spleen due to spinal cord injury (SCI) in rats is presented by MacNeil et al. In the spleen of SCI animals the production of TNF, but not IL-1, was elevated in response to endotoxin. No change was observed in the cytokine response of the liver. In the skin turpentine-induced inflammation was delayed by 12-24 hours in SCI rats. The production of TNF and monocyte chemotactic protein was dramatically reduced below the SCI, but not above. These observations demonstrate that spinal cord injury has a significant modulatory effect on immune responses, which is highly relevant to the management of patients with such injury.
New Foundation of Biology Edited by Berczi I and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
403
Growth Hormone Therapy and Immune Function
ROBERT RAPAPORT, and ROBERT MOGHADDAS
Mount Sinai School of Medicine, Mount Sinai Diabetes Center, 1200 Fifth Avenue-Box 1616 New York, NY, USA 10029
ABSTRACT Multiple lines of evidence have documented interactions between the Growth Hormone/InsulinLike Growth Factor axis and the Immune System. These interactions have been documented both in vitro and in vivo, both in animal models as well as in humans. We have been studying this interaction in children with growth hormone insufficiency. In the baseline state, we, and others, have shown that children with growth hormone deficiency do not have significantly altered immune functions. The total levels of white blood cells, as well as red blood cells as well as most immune parameters tend to be normal. During treatment with exogenous human growth hormone, total levels of white blood cells do not change significantly. We and others have shown that there is a significant decrease of percent B-cells during the course of treatment, which in most patients is transient. This has been shown during treatment with both pituitary derived growth hormone as well as biosynthetic human growth hormone. Of interest however, serum levels of immunoglobulin including immunoglobulin A, immunoglobulin M, and immunoglobulin G have remained normal. In some patients a substantial transient decrease in PHA responsiveness as well as a T-helper/suppressor ratio has also been described. In as much as most of the effects of exogenous human growth hormone treatment have affected B-cells we attempted to describe growth hormone receptors on circulating peripheral lymphocytes. By the use of two-color flow cytometry, we were able to report hat monocytes along with B-cells are the cells that exhibit growth hormone receptors. Growth hormone receptors on B-cells were normal in a variety of clinical conditions characterized by short stature in whom growth hormone receptors are expected to be normal. In preliminary studies, we have also been able to examine the production of growth hormone by lymphocytes and were able to detect significant growth hormone levels produced by cells derived from even growth hormone deficient children. In vitro, numerous studies have shown that growth hormone as well as IGF-I can increase red blood cell production as well as increase the proliferation of leukemic cell lines. There has been a concern that some patients with growth hormone deficiency have developed leukemia during the course of growth hormone treatment. We have shown that several patients identified as having growth hormone deficiency who have not received any exogenous growth hormone also developed leukemia suggesting that the association between growth hormone treatment and leukemia is as best tenuous. In growth hormone deficient children and adults there has been a suggestion that acute institution of growth hormone therapy has lead to an increase in hemoglobin levels. We are in the process of studying the interaction
404
between hemoglobin and insulin like growth factor I (an index of growth hormone production and treatment). In a series of more than 30 growth hormone deficient children treated for up to 8 years, we have found no significant asssociation between Hemoglobin and IGF-I, suggesting that the effects of growth hormone on red blood cell production in children may be an acute adaptational effect which does not lead to a long-term association between the two parameters.
1.
INTRODUCTION
Multiple lines of evidence, in vitro and in vivo, in both animals and humans, have documented interactions between the growth hormone (GH), Insulin-like growth factor (IGF-I) axis and the immune system. Many studies have demonstrated that GH has a profound influence on both cellular, and humoral-immunity. GH and IGF-I have been been implicated as important factors in regulating the immune system and as anabolic and/or stress mediators with effects on the immune system [1, 2, 3, 4, 5, 6].
2.
ANIMAL STUDIES
Much information regarding the role of growth hormone in immune system modulation has come from the study of congenitally hypopituitary mice: the Snell-Bagg (dw) and Ames (df) mice. They lack growth hormone and prolactin producing cells in the anterior pituitary gland. In addition to growth defects, they have multiple profound defects in immunity, many of which can be reversed by GH [7, 8]. In dwarf mice and hypophysectomized rats, GH administration results in full restoration of immuno-competence, including increase of splenic and thymus resident T and B-lymphocytes [9, 10, 11, 12]. However, T-cell and B-cell numbers, distribution and function appear normal in growth hormone knockout mice [13]. In burned mice, GH improved immune function, in part by modulating Th-1 and Th-2 cytokine response [14, 15]. It has recently been shown that development of the immune system of an amphibian is also dependent on the pituitary gland [16].
3.
STUDIES IN MAN
In humans, the clinical syndrome of growth hormone deficiency is not accompanied by increased susceptibility to infections. In the base line state, we and others have shown that most children with growth hormone deficiency do not have significantly altered immune functions. The total levels of white blood cells, red blood cells, as well as most immune parameters tend to be normal [17, 18]. During treatment with exogenous human growth hormone, total levels of white blood cells do not change significantly. In GH-deficient patients a substantial, but transient decrease in phytohemagglutinin (PHA)-induced lymphoblast responsiveness and in the T-helper/suppressor ratio has been described during treatment with exogenous GH [18, 19, 20, 21]. We found the number of natural killer (NK) cells to be normal [22], but there are some reports of decreased NK cells in GH deficient adult patients [23]. Decreased NK cell function has also been reported in GH deficient states [24, 25]. The therapy of GH-deficient children with recombinant GH enhanced maturational steps in immune lymphoid tissues [26]. We, and
405
others, have shown that there is a significant decrease of B-cells during the course of treatment, which in most patients is transient. This has been shown during treatment with both pituitary derived growth hormone as well as biosynthetic human growth hormone [18, 19, 22, 27, 28, 29]. Of interest, however, serum levels of immunoglobulins including immunoglobulin-A, immunoglobulin-M, and immmunoglobulin-G remain normal. In vitro, when peripheral blood lymphocytes derived from control and GH deficient children were exposed to GH, the percent of cells exibiting B cell surface markers decreased compared to the lymphocytes grown without GH [30]. This effect has also been confirmed by others [31]. In patients with GH excess, acromegaly, decreased peripheral and intramural lamina propria B-cell expression has been observed [32]. Growth hormone receptors have been identified on both cultured and circulating peripheral lymphocytes [36, 37, 38]. By the use of two-color flow cytometry, we and another group, were able to report that monocytes along with B-cells are the cells that exhibit growth hormone receptors most consistently [39, 40]. Growth hormone receptor expression on circulating B-cells were normal in a variety of clinical conditions characterized by short stature in whom growth hormone receptors were expected to be normal. The percentage of B-cell receptors correlated well with Growth Hormone Binding Protein (GHBP) levels in the blood. GHBP reflects the extra-cellular portion of the GH receptor (39]. IGF-I receptors have been reported on human peripheral blood T-cells, B-cells and NK cells [33]. Receptors for IGF-I have also been found on lymphocytes [34]. IGF-I has been found to bind preferentially to monocytes and B-cells [35]. The production of both GH and IGF-I have been described by cells of the immune system. Some evidence suggests that the effects of GH on the immune system may be in an autocrine or paracrine fashion, perhaps by way of locally produced IGF-I [41, 42]. Cells of the immune system can produce diverse hormones but more specifically GH and IGF-I. The production of GH by rat leukocytes [43] and that of IGF-I by human B-lymphocytes [44] have also been reported. GH production by lymphocytes of normal adult volunteers has been reported [45]. It was also demonstrated that GH, but not IGF-I, up-regulated GH secretion by human lymphocytes [46]. In a preliminary study, we were able to detect significant growth hormone levels produced by lymphocytes derived from even growth hormone deficient children [47]. GH and IGF-I have profound effects on hematopoiesis [48]. Increase in red blood cell production and proliferation of leukemic cell lines (myeloid progenitor cells) by GH, as well as IGF-I, has been documented [49]. A correlation between IGF-I and IGF-binding protein 3 and hemoglobin-hematocrit levels has been suggested in short children [50]. There has been a concern that some patients with growth hormone deficiency have developed leukemia during the course of growth hormone treatment. For children without previous risk factors, GH treatment does not seem to increase the risk of developing leukemia [51, 52]. We have described that several patients identified as having growth hormone deficiency who have not received any exogenous growth hormone, who developed leukemia suggesting that the association between growth hormone treatment and leukemia is at best tenuous [53]. In growth hormone deficient children and adults, there has been a suggestion that acute institution of growth hormone therapy has lead to an increase in hemoglobin levels. We are in the process of studying the interaction between hemoglobin and IGF-I. In a series of more than thirty growth hormone deficient children treated for up to eight years, we have found no significant association between hemoglobin and IGF-I [54]. This suggests that the effects of growth hormone on red blood cell production in children may be an acute adaptation effect, which does not lead to a long term association between the two parameters.
406
REFERENCES 1.
Kelly KW. The role of growth hormone in modulation of the immune response. Ann NY Acad sci 1991; 95-103. 2. Gala RR. Prolactin and Growth hormone in the regulation of the immune system. Proc Soc Exp Biol Med 1991; 198: 513-527. 3. Rapaport R. Effects of growth hormone on immune functions. Advances in psychoneuroimmunology. J Szelenyi and Berczi I. Springer Verlag pp. 83-98, 1994. 4. Auernhammer CJ, Strasburger CJ. Effect of growth hormone and insulin-like growth factor I on the immune system. Eur J Endocrinology 1995; 83: 5663-5667. 5. Clark R. The somatogenic hormones and insulin-like growth factor-I: stimulators of lymphopoiesis and immune function. Endocrine Rev 1997; 18:157-179. 6. Chappel S. Growth hormone in immune reconstitution. J Acquired Immune Defic Syndrome, Human Retrovirol 1999; 20:423-431. 7. Duquesnoy RJ. Immunodeficiency of the thymus-dependent system of the Ames dwarf mouse. J Immunol 1972; 108:1578. 8. Murphy WJ, Durum SK, Longo DL. Role of neuro-endocrine hormones in murine T-cell development: growth hormone exerts thymopoietic effects in vivo. J Immunol 1992; 149:851-3857. 9. Nagy E, Berczi I, Friesen HC. Regulation of immunity in rats by lactogenic and growth hormones. Acta Endocrinology 1983; 102: 351. 10. Davila DR, Brief S, Sun J, Hammer RE, Brinster RL, Kelly KW. Role of the growth hormone in regulating T-dependent immune event in aged, nude, and transgenic rodents. J Neurosci Res 1987; 18:108-116. 11. Nagy E, Berczi I. Pituitary dependence of bone marrow function. Br J Haematol 1989; 71: 457-462. 12. Murphy WJ, Rui H, Longo DL. Effect of GH and prolactin on immune development and function. Life Sci 1995; 57( 1) 1-4. 13. Foster MP, Jensen ER, Montecino-Rodriguez E, Leathers H, Horseman N, Dorshkind K. Humoral and cell-mediated immunity in mice with genetic deficiencies of prolactin, growth hormone, insulin-like growth factor-I and thyroid hormone. Clinical Immunology 2000; 96:140-149. 4] Takagi K, Suzuki F, Barrow RE, et al. Growth hormone improves immune function and survival in burned mice infected with herpes simplex virus type I. J Surg Res 1997; 69: 166-170. 15. Takagi K, Suzuki F, Barrow RE, et al. Recombinant human growth hormone modulates th-1 and th-2 cytokine response in burned mice. Ann Surg 1998; 228: 106-111. 16. Rollins-Smith LA, Tray Davis A, Reinert LK. Pituitary involvement in T-cell renewal during development and metamorphosis of Xenopus Laevis. Brain, behavior and Immunity 2000; 14: 185-197. 17. Abbassi V, Ballanti JA. Humoral, and cell-mediated immunity in growth hormonedeficient children: effect of therapy with human growth hormone. Pediatr Res 1985; 19:299-301. 18. Rapaport R, Oleske J, Ahdieh H, Solomon S, Delfaus C, Denny T. Suppression of immune function in growth hormone deficient children during treatment with human growth hormone. J Pediatr 1986; 109: 434-439. 19. Rapaport R, Petersen B, Skuza KA, Heim M, Goldstein S. Immune functions during the treatment of growth hormone-deficient children with biosynthetic human growth hormon.
407
20. 21. 22.
23. 24. 25. 26.
27. 28. 29.
30.
31.
32.
33.
34.
35.
36. 37.
Clin Pediatr 1991; 30: 22. Kiess W, Holtman H, Butenandt O, Eife R. Modulation of lymphoproliferation by human GH. Eur J Pediatr 1983; 140: 47-50. Bozzola M, Valtorta A, Moretta A, Montagna D, Maccario R, Burgio GR. Modulating effect of GH on PHA-induced lymphocyte proliferation. Thymus 1988; 12: 157-165. Peterson B H, Rapaport R, Henry DP, Huseman C, Moore WV. Effect of treatment with biosynthetic human growth hormone on peripheral blood lymphocyte populations and functions in growth hormone deficient children. J Clin Endocrinol Metab 1990; 70: 1756-1760. Span JP. Number and percentage of NK-cells are decreased in growth hormone- deficient adults. Clin Immunol ImmunoPathol 1995; 78: 90-92. Bozzola M, Valtoria A, Moretta A, Cisternino M, Biscaldi I, Schimpff RM. In vitro and in vivo effect of growth hormone on cytotoxic activity. J Pediatri 1990; 117: 596. Kiess W, Doerr H, Eisl E, Butenandt O, Belohradsy B H. Llymphocyte subsets and natural killer activity in growth hormone deficiency. N Engl J Med 1986; 314: 321. Bozzola M, Cisternino M, Valtorta A, Moretta A, Biscaldi I, Maghnie M, De Amici M, Schimpff RM. Effect of biosynthetic methionyl growth hormone therapy on the immune function in GH-deficient children. Hormone Res 1989; 31: 153. Spadoni GL, Rossi P, Ragno W, et al. Immune function in growth hormone-deficient children treated with biosynthetic growth hormone. Acta Paediatric 1991; 80: 75-79. Rapaport R, Oleske J. Effects of growth hormone therapy on immune functions. J Pediatr 1987; 110: 663-664(letter). Balducci R, Toscano V, Pasquino AM, et al. Lymphocyte subsets in adults with panhypopituitarism before and during GH treatment abstract]. Proc of the 76 th annual meet of The Endocrine Society 1994; 276. Rapaport R, Oleske J, Ahdieh H, et al. Effects of human growth hormone on immune function: in vitro studies on cells of normal and growth-deficient children. Life Sci 1987; 41: 2319-2324. Fornari MC, Scolnik MP, Palacios MF, Intebi AD, Diez RA. Growth hormone inhibits normal B-cell differentiation and neutrophil chemotaxis in vitro. Int J ImmunoPharmacol 1994; 16(8): 667-673. Colao A, Balzano A, Ferone D, Panza N, Grande G, Marzullo P, Bove A, Iodice G, Merola B, Lombardi G. Increased prevalence of colonic polyps and altered lymphocyte subset pattern in colonic lamina propria in acromegaly. Clin Endocrinol (oxf) 1997; 47(1): 23-28. Kooijman R, Willems M, Dehaas JC, Rijkers GT, Schuurmans ALG,VanBuul-offers SC, Heihens CJ, Zegers BJM. Expression of type-1 insulin-like growth factor receptors on human peripheral blood mononuclear cells. Endocrinology 1992; 131: 2244-2249. Tapson VF, Boni-Schnetzler M, Pilch PF, Center DM, Berman JS. Structural and functional characterization of the human T-lymphocyte receptor for IGF-I in vivo. J Clin Invest 1998; 82: 950-957. Stuart CA, Meehan RT, Neale LS, Cintron NM, Furlanetto RW. IGF-I binds selectively to human peripheral blood monocytes and B-lymphocytes. J Clin Endocrinol Metab 1991; 72: 1117-1122. Leung DW, Spencer SA, Cachianes G, et al. Growth hormone receptor and serum binding protein: purification,cloning, and expression. Nature 1987; 330: 537-543. Kiess W, Butenandt O. Specific growth hormone receptors on human peripheral mononuclear cells: re-expression, identification, and characterization. J Clin Endocrinol
408
38. 39.
40.
41.
42.
43. 44.
45.
46.
47.
48. 49.
50.
51. 52. 53. 54.
Metab 1985; 60: 740-746. Eshet R, Manheimer S,Chosbseing P, Laron Z. Human GH receptor in circulating lymphocytes. Horm Metab Res 1985; 7: 352-352. Rapaport R, Sills IN, Green L, Barrett P, Labus J, Skuza KA, Chartoff A, Goode L, Stene M, Petersen BH. Detection of human growth hormone receptors on IM-9 cells and peripheral blood mononuclear cell subsets by flow cytometry: correlation with growth hormone-binding protein levels. J Clin Endocrinol Met 1995; 80:2612-2619. Badolato R, Bond H, Valerio G, et al. Differential expression of surface membrane growth hormone receptor on human peripheral blood lymphocytes detected on dual fluorochrome flow cytometry. J Clin Endocrinol Metab 1994; 79: 984-990. Sabharwal P, Varma S. GH synthesized and secreted by human thymocytes acts via IGF-I as an autocrine and paracrine growth factor. J of Clinical Endo and Metab 1996; 81: 120-125. Geffner ME, Bersch N, Lippe BM, Rosenfeld RG, Hintz RL, Golde DW. Growth hormone mediates the growth of T-lymphoblast cell lines via locally generated insulin-like growth factor-I. J Clin Endocrinol Metab 1990; 71: 464-469. Weigent DA, Blalock JE. The production of Growth hormone by subpopulations of rat mononuclear leukocytes. Cell Immunol 1991 ; 134: 55-65. Merimee TJ, Grant MB, Broder CM, Cavalli-sforza LL. IGF-secretion by human B-lymphocytes: a comparison of cells from normal and pygmy subjects. J Clin Endocrinol Metab 1989; 69: 978-984. Hattori N, Ikekubo K, Ishihara T, Moridera K, Hino M, Kurahachi H. Spontaneous GH secretion by unstimulated human lymphocytes and the effect of GH-releasing hormone and somatostatin. J Clin Endocrinol Metab 1994; 79: 1678-1680. Hattori N, Shimomura K, Ishihara T, et al. Growth hormone secretion from human lymphocytes is up-regulated by GH but not affected by IGF-I. J Clin Endocrinol Metab 1993; 76: 937-939. Rapaport R, Stein D, Louzano A, Goode L, Garibaldi L, Stene M. Growth hormone secretion by lymphocytes of growth hormone deficient children. Presented at 10 th international congrress of Endocrinology. San Francisco, CA, USA, June 12-15, 1996. Merchov S, Tatarsky I, Hochberg Z. Enhancement of erythropoesis in vitro by human growth hormone is mediated by IGF-I. Brit J Haematol 1988; 70: 267-271. Christ ER, Cummings MH, Westwood NB, Sawyer BM, Pearson TC, Sonken PH, RusselJones DL. The importance of growth hormone in the regulation of Erythropoiesis, red cell mass, and plasma volume in adults with growth hormone deficiency. Journal of Clin Endocrinol and Metab 1997; 82,9: 1-10. Vihervuori E, Virtanen M, Koistinen H, Koistinen R, Sepalla M, Simmes MA. Hemoglobin level is linked to growth hormone-dependent proteins in short children. Blood 1996; 87:2075-2081. Watanabe S, Tsunematsu Y, Fujimoto J, et al. Leukemia in patients treated with growth hormone. Lancet 1998; 331:1159-1160. Allen DB, Chen Rundle A, Graves DA, Blethen SL. Risk of leukemia in children treated with growth hormone: review and reanalysis. J of Pediatrics 1997; 131: 1-8. Rapaport R, Oberfield SE, Robinson L, Salisbury S, David R, Rao J, Redmond GP. Relationship of growth hormone deficiency and leukemia. J Pediatr 1995; 126: 759-761. Rapaport R, Davidow A, Eisenstein D. IGF-I and Hemoglobin in children with growth hormone deficiency during treatment with human growth hormone. Abstract number 1983. 82 "d annual meeting of The Endocrine Society. Toronto, Canada.
New Foundationof Biology Edited by Berczi I and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
409
The Role of Prolactin in Systemic Lupus Erythematosus
RICHARD WARRINGTON, TIM MCCARTHY, EVA NAGY, KINGSLEY LEE and ISTVAN BERCZI
Departments of Medicine and Immunology, University of Manitoba, Winnipeg, Manitoba, Canada
ABSTRACT Systemic lupus erythematosus is a disease that emphasizes the role of the endocrine system in autoimmunity. This condition primarily affects females, and during pregnancy, a state that shifts the balance of the immune response from TH1 to TH2, flares in disease activity occur in approximately 40% of patients. That oestrogen and androgen can modulate disease activity in SLE has been clearly shown in the experimental murine model of lupus that develops in NZB/W F1 mice, but oestrogen also stimulates prolactin secretion from the pituitary gland and prolactin is a stimulator of both humoral and cell-mediated immunity. It was shown in 1991 that suppression of pituitary prolactin secretion by bromocriptine delays the onset of disease in NZB/W F1 mice with SLE, while hyper-prolactinaemia induced by pituitary tranplantation accelerates mortality. In human lupus, it has been possible to demonstrate elevations of serum prolactin levels in patient groups, but the presence of an elevated prolactin does not always correlate with increased disease activity; although interesting associations with renal disease and neuropsychiatric lupus have been described. In males with lupus there appears to be a clearer relationship between hyper-prolactinaemia and disease. Bromocriptine has been shown to have some therapeutic benefit in human lupus. The cytokines IL-1, IL-2 and IL-6 stimulate production of prolactin, which is itself structurally related to GM-CSF and cytokines IL-2 through IL-7. A role for this hormone in the immune response is implied by the presence of prolactin receptors on B lymphocytes, T lymphocytes, monocytes and thymic epithelial cells. Interaction of the ligand induces activation via jak kinases and the phosphorylation of latent STAT proteins. But the level of pituitary hormone may not limit the effects of prolactin. Unstimulated peripheral blood mononuclear cells from lupus patients secrete a prolactin-like protein and at significantly higher levels than do unstimulated cells from normal donors. The similarity of this protein to pituitary prolactin has been shown in vitro in the Nb2 assay, by Western blot analysis using polyclonal antibodies to prolactin and by southern blot analysis. The study of lupus therefore emphasizes the role of neuroimmune mechanisms in autoimmunity. The role of prolactin (PRL) in Systemic lupus erythematosus (SLE) remains unclear but its interactions in this disease emphasize the interrelationships that exist between the neuroimmune system and human autoimmunity.
410
SLE is an autoimmune disease characterized by enhancement of the activities of the humoral immune system, in which antibody responses occur against nuclear, cytoplasmic, membrane and soluble antigens. Hormonal influences are manifest by the predominance of the disease in females during the reproductive years, where the female to male ratio is 5: 1. Before menarche and after the menopause this ratio is only 2:1 [1]. There are definite genetic influences at work in this disease, with the presence of B cell hyperactivity in family members and an increase in the prevalence of other autoimmune diseases in lupus families [2]. As in many autoimmune conditions, these genetic influences are limited, with concordance of SLE in identical twins being about 35% [3]. One of the unique characteristics of lupus is the presence of antibodies against double stranded (ds) or native DNA. Other anti-nuclear antibodies are common in SLE and the anti-Sm antibody, which is directed against snRNP proteins, is also a unique marker for this disease [4]. Since dsDNA itself is not immunogenic, considerable energy has been spent in determining how this autoantibody might be generated in lupus. It now appears that it may appear through epitope spreading from an initial antibody response directed against nuclear particle structures that include protein determinants such as are found on the nucleosomes [5]. These particles, which include both nuclear proteins and DNA, can be found expressed on the surfaces of apoptotic cells such as keratinocytes, after exposure to ultra violet light in the skin. The phagocytosis of these particles by antigen presenting cells and the subsequent expression of the processed peptides is thought to activate autoreactive T cells, which then act as help for B cells with specificity for nuclear determinants, including dsDNA [6, 7]. This results in an autoantibody response that begins with specificity for nuclear proteins and spreads to include dsDNA. Abnormalities such as enhanced apoptosis have been found to occur in lupus and these may result in the appearance of cryptic or previously hidden epitopes derived from processed nuclear proteins [8]. Evidence for a T cell driven response is derived from the presence of somatic mutations in the combining regions of anti-dsDNA antibodies from lupus patients [9]. There are probably many ways in which lupus can develop, and several quite different animal models of this disease exist. In terms of its female predominance, the NZB/W F1 mouse is the closest to human lupus, with a high female predominance in terms of the severity and rapidity of onset of the disease, and evidence of potent hormonal influences [ 10]. While estrogen exacerbates and testosterone ameliorates the disease in this model, lupus in NZB/W F1 mice is also accelerated by the presence of hyperprolactinemia induced by pituitary transplants [11]. Interestingly, treatment of the pituitary-transplanted mice with bromocriptine, which blocks prolactin secretion, results in disease remission. It also appears that estrogen exacerbates disease in lupus mice via prolactin. The administration of estrogen does not exacerbate the disease if the prolactin-stimulating properties of this hormone are blocked by bromocriptine [11]. There has been a great deal of effort spent in examining the role of pituitary prolactin in human lupus [12-20]. Some data suggest that the presence of hyperprolactinemia itself is associated with the development of SLE, but it has been difficult to show an association between high serum prolactin levels and disease activity in groups of lupus patients. B lanco-Favela et al., [21] have suggested that this may be due to a lack of power in such studies. They re-analyzed the pooled data from 3 studies and found that hyperprolactinemia appeared to be associated with disease activity (OR = 3.89, CI 1.43- 11.3). McMurray et al., [22] treated 7 lupus patients with bromocriptine, with subsequent suppression of disease activity in six. In a double blind placebo controlled study, Alvarez-Nemegyel et al., [23] found that bromocriptine 2.5 mg daily for a mean duration of 12.5 months resulted in a fall in serum PRL levels from 24.9 + 18.4 to 5.8 + 9.0 and a reduction in disease activity as measured by SLEDAI.
411
In the active treatment group the SLEDAI was 0.9 + 1.4 vs. 2.6 + 4.5 (p < 0.05). Although the question as to whether or not a high serum prolactin predisposes to increased disease activity in SLE remains unanswered, there is some data suggesting that PRL may increase the risk of disease involvement of certain organs, such as the kidney and the brain. Miranda et al., [24] found an association between active renal disease and serum and urinary prolactin levels, while Jara et al., [25] described the presence of increased levels of prolactin in the cerebrospinal fluid of lupus patients with neuropsychiatric lupus. There was a strong correlation between the levels of IL-6 in the CSF and the levels of prolactin. This raises the important question of the interactions between cytokines and prolactin, which itself exhibits some similarities to cytokines. PRL is structurally related to GM-CSF, erythropoitin (EPO) and the cytokines IL-2 through to IL-7 [26]. Prolactin induces IL-2 receptor expression on sp|enocytes and enhances IL-2 responsiveness [27], an interesting effect when one considers that a defect in IL-2 responsiveness is a common abnormality in human and murine lupus [28, 29] and IL-2 may protect lymphocytes from apoptosis [30]. A direct effect of prolactin on cells of the immune system is supported by the presence of prolactin receptors on T cells, B cells and monocytes [31, 32]. These receptors express homology with receptors for Growth Hormone, EPO, IL-2 and IL-6 [26]. When it comes to the influences of cytokines on PRL secretion itself, IL-1, IL-2 and IL-6 stimulate PRL secretion while IFN- T inhibits PRL release [33]. In lupus, the most consistent cytokine abnormalities associated with disease activity are an enhancement of IL-6 and IL-10 production and impaired IL-2 responsiveness, both of which would tend to maintain the balance toward the TH-2 responses typical of SLE [34-36]. It is possible that PRL itself might antagonize such a deviation towards TH-2. The TH-2 predominance may explain the effects of pregnancy on lupus. Normally TH-2 responses are enhanced in pregnancy, because this is of immunological benefit to the fetus, making it more likely that a pregnancy will go to term. Maternal cytokine profiles in pregnancies that miscarry or abort show enhanced TH-1 responses. Increases in the TH-2 cytokines IL-4 and IL-10 are found during normal pregnancy, while those cytokines associated with TH-1 responses are reduced [37, 38]. The enhancement of IL-10 production that occurs at this time appears to correlate with increases in serum 17-beta estradiol levels [39]. Abnormally enhanced TH-1 responses in pregnancy tend to be associated with miscarriages or failure to reach term [38]. So a number of factors are present during pregnancy that would in theory exacerbate lupus and influence PRL production. There is a reduction in IL-2 synthesis and IFN-T synthesis, which would accentuate the defect in IL-2 responsiveness present in lupus but would also enhance prolactin levels. Increases in IL-10 synthesis and estradiol would suppress TH-1 responses, enhance TH-2 responses and also increase PRL secretion. Such changes may explain why in SLE, a TH-2 disease, exacerbation during pregnancy occurs in about 40% of patients, while in Rheumatoid arthritis, a TH-1 disease, there is remission in about 75% of cases [40, 41]. So far, we have limited our discussions to pituitary-derived prolactin and its potential effects on disease activity in lupus, including how these might be mediated. But in 1995, Gutierrez et al., [42] reported the release of PRL-like immunoreactive proteins from peripheral blood lymphocytes in lupus. These proteins were 11 kD and 24-27 kD in size. Subsequently Larrea et al., [43] demonstrated constitutive and Con-A induced PRL-like activity secreted by peripheral blood mononuclear cells from lupus patients and normal subjects. The levels of production by mononuclear cells from lupus patients' cells were significantly enhanced compared to normal subjects (p < 0.001). In contrast to the data from Gutierrez et al., the PRL-like material detected by Larrea and associates was 60 kDa in size. Nevertheless, this material stimulated proliferation of Nb2 cells in cultures, and was blocked by inhibitory
412
anti-prolactin antibodies. The prolactin-like protein was bound by anti-prolactin antibodies on Western blots and showed homology with prolactin on Southern blots. By Southern blot the prolactin was identified as a 600-bp band. We have examined the constitutive levels of prolactin-like proteins (assayed using the Nb-2 cell line in vitro) extracted from peripheral blood mononuclear cells from lupus patients and compared this to production in normal control subjects. Proliferative activity of the cell extracts as determined in the Nb-2 assay was specifically neutralized by polyclonal anti-PRL antibody. As shown in Figure 1, the mean level of mononuclear cell-derived PRL was markedly and significantly increased in lupus patients, compared to normal controls (3.78 _+ 1.1 ~tg/ml compared to 27.25 + 6.43 ~tg/ml, p = 0.001). However, when the levels of mononuclear cell prolactin were analyzed to determine if these were raised in patients with active disease, as determined by the Systemic lupus erythematosus disease activity index (SLEDAI), no correlation was found. Indeed, there was a trend, which was not statistically significant, for the levels of mononuclear cell-derived prolactin to be increased in those patients with lower SLEDAI scores or less active disease. (Figure 2) The mean SLEDAI for the group was 9.308 + 9.15, with a range from 0 to 34. The differences in the mononuclear cell-derived-PRL between individual patients showed no relationship to the dose of oral prednisone (or equivalent) being administered (Figure 3). The possibility must therefore be considered that under certain circumstances, mononuclear cell-derived PRL might act to ameliorate the immune defects of lupus, or alternatively, that PRL levels in mononuclear cells do not reflect disease activity. To confirm the latter, it is necessary also to assess mononuclear cell PRL mRNA levels and their correlation with the SLEDAI. Finally, the specificity of the increases in mononuclear cell PRL must be examined in comparison to disease controls, particularly because lymphocyte PRL has been reported to be elevated in synovial tissues of patients with Rheumatoid arthritis [44]. Systemic lupus erythematosus is a complex condition that has both multiple genetic and environmental factors involved in its pathogenesis. Nevertheless, it is a condition that clearly demonstrates the interrelationships that exist between the neuroimmune system and this human autoimmune disease. There is evidence that prolactin may, under certain circumstances, influence disease activity in SLE. There is an obvious effect of hormones such as estrogen and testosterone on the condition, and there is a relationship between pregnancy, a TH-2 state and the exacerbation of lupus, a TH-2 disease. But it is unlikely, given the complexity of the immune system and human disease that it will be possible to demonstrate a clear pathogenic role for prolactin in lupus, whether pituitary in origin or mononuclear cell derived. In some circumstances prolactin may act as an exacerbating factor and in others as an ameliorating influence. That would not be an unusual situation, given the close similarities between prolactin and the cytokines of the immune system, where redundancy is common and cytokines may exert stimulatory or inhibitory effects, depending upon circumstances. Perhaps our confusion over the role of the neuroimmune system in autoimmunity is more a reflection of our inability to understand this complex and non-linear system and the ways in which a cytokine or a mediator like PRL can both predispose to a disease and under other circumstances, act to ameliorate the disease.
413
0
Normal
+
SLE
1 O0
80
p = 0.001 60
I,,i
40 9
i
20
2
Figure 1.
A comparison of lymphocyte prolactin levels in normal donors (+) and SLE patients (o), with mean and
standard deviation of the means for the two groups, which differ significantly (p = 0.001).
o
100
"~ =
-,
SLE PATIENTS
9
NORMALS
o
8o O
R =-0.578
p=0.0148
z ~o
i4o
O
20 O~
m~
0
o o
ol
o
f
10
20
SLEDAI
Figure 2.
o
ol
30
o
40
SCORE
The relationship between disease activity in lupus patients, determined by SLEDAI score, and lymphocyte
prolactin levels in ug/ml. The block on the Y-axis shows the mean and standard deviation for normal subjects and the horizontal line is 3 standard deviations from the normal mean.
414
2.8
.
21
i,~
14
01 1
60"
l~r~ eti~ t~tt~l Figure 3.
,/
100 80"
~
d
~
The relationship between prednisone dosage equivalent in mg (X-axis), lymphocyte prolactin levels in
ug/ml (Y-axis), and disease activity by SLEDAI score (Z-axis), plotted as a 3-dimensional mesh. There is a positive correlation between prednisone dosage and disease activity (correlation coefficient = 0.46, p = 0.061) and a negative correlation between lymphocyte prolactin levels and disease activity (correlation coefficient = -0.578, p = 0.0148). No relationship exists between prednisone dosage and lymphocyte prolactin levels.
REFERENCES 1. 2. 3.
4. 5.
6.
7. 8.
Klippel. JH. Systemic lupus erythematosus: demographics, prognosis and outcome. J Rheumatol Suppl 1997; 48: 67-71. Reveille JD, Bias WB, Winkelstein JA et al. Familial systemic lupus erythematosus: immunogenetic studies in eight families. Medicine 1983; 62: 21-35. Grennan DM, Parfitt A, Manolios N, Huang Q, Hyland V, Dunckley H, Dovas T, Gatenby P, Badiock C. Family and twin studies in systemic lupus erythematosus. Dis Markers 1997; 13: 93-98. Tan EM, Cohen AS, Fries JF et al., The 1982 revised criteria for the classification of systemic lupus erythematosus (SLE). Arth Rheum 1982; 25: 1271-77. Burlingame RW, Rubin RL, Balderas RS, Theofilopoulos AN. Genesis and evolution of anti-chromatin autoantibodies in murine lupus implicates T-dependent immunization with self-antigen. J Clin Invest 1993; 91: 1687-96. Casciola-Rosen L, Anhalt G, Rosen A. Autoantigens targeted in systemic lupus erythematosus are clustered in two populations of surface structures on apoptotic keratinocytes. J Exp Med 1994; 179: 1317-30. Bach JF, Koutouzov S. New clues to systemic lupus. Lancet 1997; Suppl III: 11. Andrade F, Casciola-Rosen L, Rosen A. Apoptosis in systemic lupus erythematosus. Clinical implications. Rheum Dis Clin North Am 2000; 26: 215-27.
415
10.
11. 12. 13.
14.
15. 16. 17. 18.
19.
20. 21.
22.
23.
24.
25.
26. 27.
Winkler TH, Fehr H, Kalden JR. Analysis of immunoglobulin variable region genes from human IgG anti-DNA hybridomas. Eur J Immunol 1992; 22: 1719-28. Elbourne KB, Keisler D, McMurray RW. Differential effects of estrogen and prolactin on autoimmune disease in the NZB/NZW F1 mouse model of systemic lupus erythematosus. Lupus 1998; 7: 420-27. McMurray R, Keisler D, Kanuckel K, Izui S, Walker SE. Prolactin influences autoimmune disease activity in the female B/W mouse. J Immunol 1991; 147: 3780-87. Pauzner R, Urowitz MB, Gladman DD, Gough JM. Prolactin and systemic lupus erythematosus. J Rheumatol 1994; 21: 2064-7. Neidhart M. Elevated serum prolactin or elevated prolactin/cortisol ratio are associated with autoimmune processes in systemic lupus erythematosus and other connective tissue diseases. J Rheumatol 1996; 23: 476-81. Buskila D, Lorber M, Neuman L, Flusser D, Shoenfeld Y. No correlation between prolactin levels and clinical activity in patients with systemic lupus erythematosus. J Rheumatol 1996; 23:2185-6. Mok CC, Lau CS. Lack of association between prolactin levels and clinical activity in patients with systemic lupus erythematosus. J Rheumatol 1996; 23: 2185-6. Huang CM, Chou CT. Hyperprolactinemia in systemic lupus erythematosus. Chung Hua I. Hsueh. Tsa Chih (Tapei) 1997; 59: 37-41. Mok CC, Lau CS, Tam SC. Prolactin profile in a cohort of Chinese systemic lupus erythematosus patients. Br J Rheumatol 1997; 36: 986-9. Rovensky R, Jurankova E, Rauova L, B lazickova S, Veselkova Z, Jezova D, Vigas M. Relationship between endocrine, immune and clinical variables in patients with systemic lupus erythematosus. J Rheumatol 1997; 24: 2330-4. Ostendorf B, Fischer R, Santen R, Schmitz-Linneweber B, Specker C, Schneider M. Hyperprolactinemia in systemic lupus erythematosus. Scand J Rheumatol 1996; 25: 97-102. Jimena P, Aguirre MA, Lopez-Curbelo A, de Andres M, Garcia-Courtay C, Cuadrado MJ. Prolactin levels in patients with systemic lupus erythematosus. Lupus 1998; 7: 383-6. B lanco-Favela F, Quintal-Alvarez GR, Leanos-Miranda A. Association between prolactin and disease activity in systemic lupus erythematosus. Influence of statistical power. J Rheumatol 1999; 26: 55-9. McMurray RW, Weidensaul D, Allen SH, Walker SE. Efficacy of bromocriptine in an open label therapeutic trial for systemic lupus erythematosus. J Rheumatol 1995; 22:2084-2091. Alvarez-Nemegyei J, Cobarrubias-Cobos A, Escalante-Triay F, Sosa-Munoz J, Miranda JM, Jara LJ. Bromocriptine in systemic lupus erythematosus: a double blind, randomized, placebo-controlled study. Lupus 1998; 7: 414-9. Miranda JM, Prieto RE, Paniagua R, Garcia G, Amato D, Barile L, Jara LJ. Clinical significance of serum and urine prolactin levels in lupus glomerulonephritis. Lupus 1998; 7: 387-91. Jara LJ, Irigoyen L, de J Ortiz M, Zazueto B, Bravo G, Espinoza LR. Prolactin and Interleukin-6 in Neuropsychiatric lupus erythematosus. Clin Rheumatol 1998; 17: 110-114. Bazan JF. Haematopoietic receptors and helical cytokines. Immunol Today 1990; 11: 350-54. Mukherjes P, Mastro AM, Hymer WC. Prolactin induction of interleukin 2 receptors on rat splenic lymphocytes. Endocrinology 1990; 126: 88-94.
416
28. 29. 30. 31. 32. 33. 34. 35.
36.
37.
38.
39.
40. 41. 42.
43.
44.
Warrington RJ, Sauder PJ, Homik J, Ofosu-Appiah W. Reversible Interleukin-2 response defects in systemic lupus erythematosus. Clin Exp Immunol 1989; 77: 163-7. Crispin JC, Alcocer-Varela J. Interleukin-2 and systemic lupus erythematosus -Fifteen years later. Lupus 1998; 7:214-22. Kosac FJ, Cornell DL, Lipke AB, Graham LJ, Bear HD. Protective role of IL-2 during activation of T cells with byostatin 1. Int J ImmunoPharmacol 2000; 22: 645-53. Cosman D. The hematopoietin receptor superfamily. Cytokine 1993; 5: 95-106. Leite de Moraes MC, Touraine P, Gagnerault MC, Savino W, Kelly PA, Dardenne M. Prolactin receptors and the immune system. Ann Endocrinol (Paris) 1995; 56: 567-70. Chikanza IC. Prolactin and neuroimmunomodulation: in vitro and in vivo observations. Ann NY Acad Sci 1999; 876:119-30. Cross JT, Benton HP. The roles of interleukin-6 and interleukin-10 in B cell hyperactivity in systemic lupus erythematosus. Inflamm Res 1999; 48: 255-61. Crondal G, Kristjansdottir H, Gunnlaugsdottir B, Arnason A, Lundberg I, Klareskog L, Steinsson K. Increased numbers of interleukin-10-producing cells in systemic lupus erythematosus patients and their first degree relatives and spouses in Icelandic multi-case families. Arthritis Rheum 1999; 42:1649-54. Jones BM, Liu T, Wong RW. Reduced in vitro production of interferon gamma, interleukin-4 and interleukin-12 and increased production of interleukin-6, interleukin-10 and tumor necrosis factor-alpha in systemic lupus erythematosus. Weak correlations of cytokine production with disease activity. Autoimmunity 1999; 31:117-24. Wegmann TG, Lin H, Guilbert L, Mosmann TL. B i-directional cytokine interactions in the maternal-fetal relationship: is successful pregnancy a TH2 phenomenon. Immunol Today 1993; 14: 353-8. Marzi M, Vigano A, Trabattoni D, Viha ML, Salvaggio A, Clerici E, Clerici M. Characterization of Type 1 and Type 2 cytokine production profile in physiologic and pathologic human pregnancy. Clin Exp Immunol 1996; 106: 127-133. Kruse N, Grief M, Moriabaldi NF, Marx L, Toyka KV, Rieckmann P. 2000 Variations in cytokine mRNA expression during normal pregnancy. Clin Exp Immunol 2000; 119: 317-22. Sittiwangkul S, Louthrenoo W, Vithayasai P, Sukitaut W. Pregnancy outcome in Thai patients with systemic lupus erythematosus. Asian Pac J Allergy Immunol 1999; 17: 77-83. Ostensen M. Sex hormones and pregnancy in rheumatoid arthritis and systemic lupus erythematosus. Ann NY Acad Sci 1999; 876: 131-43. Gutierrez MA, Molina JF, Jara LJ, Cuellar ML, Garcia C, Gutierrez-Urena S, Gharavi A, Espinoza LR. Prolactin and systemic lupus erythematosus: Prolactin secretion by SLE lymphocytes and Proliferative (autocrine) activity. Lupus 1995; 4: 348-52. Larrea F, Martinez-Castillo A, Cabrera V, Alcocer-Varela J, Queipo G, Carino C, Alarcon-Segovia D. A bioactive 60-kilodalton Prolactin species is preferentially secreted in cultures of mitogen-stimulated and nonstimulated peripheral blood mononuclear cells from subjects with systemic lupus erythematosus. J Clin Endocrinol Metab 1997; 82: 3664-69. Nagafuchi H, Suzuki N, Kaneko A, Asai T, Sakane T. Prolactin locally produced by synovium infiltrating T lymphocytes induces excessive synovial cell functions in patients with rheumatoid arthritis. J Rheumatol 1999; 26:1840-1900.
New Foundationof Biology
417
Edited by Berczi I and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
Combination Immunotherapy of Cancer
NAGY E 1, BERCZI I ~, BARAL E 2 and J. KELLEN 3
Department of Immunology, Faculty of Medicine, The University of Manitoba, Winnipeg, Manitoba, R3E O W3, Canada. Cancer Care Manitoba 2, 100 Olivia Street, Winnipeg, Manitoba, R3E O V9, Canada. Department of Laboratory Medicine 3, Sunnybrook Health Sciences Center, University of Toronto, Toronto, Ontario Canada.
ABSTRACT The interaction of cancer with the immune system has long been established. Modern approaches to cancer immunotherapy concentrate on boosting host resistance by vaccines, killer cells or cytokines, such as interleukin-2 or interferons. Little attention has been paid to the possibility of sensitizing tumor cells to immune mediated attack. The non-steroidal antiestrogenic agents, tamoxifen (TX) and toremifene (TO) are widely used for the treatment of estrogen receptor positive mammary carcinomas and other sex hormone dependent tumors. We discovered that TX and TO sensitize tumor cells for killing by natural killer (NK), lymphokine activated killer (LAK) and cytotoxic T lymphocytes (CTL). We also demonstrated that TX and TO potentiated the immunotherapy of lethal cancer in syngeneic murine tumor-host system. DBA/2 and C3H mice were injected with lethal doses of tumor cells subcutaneously. Combination therapy with NK, LAK or CTL effector cells and antiestrogens could cure lethal cancer in up to 75% of mice. Lymphocytes freshly isolated from the ascites of patients with ovarian cancer had no lytic effect on autologous tumor cells. Activation of tumor associated lymphocytes (TAL) or tumor infiltrating lymphocytes (TIL) with hrIL-2 in the presence of autologous tumor cells induced detectable cytotoxicity in most of the cell cultures. A highly significant increase of tumor cell lysis was found when both target and effector cells were treated with antiestrogens. Additional treatment with interferon-alpha resulted in the further enhancement of cytotoxicity in a significant number of cases. It is clear from our results that TX and TO are capable of increasing the susceptibility of human ovarian and lung cancer cells to autologous killer cells in about 60% of the cases so examined. Sensitization requires the active metabolic participation of the target cells suggesting the amplification of programmed cell death as the underlying mechanism. Both the Fas/Fas-L and perforin/granzyme pathways, which can trigger apoptosis, are affected by antiestrogens. This was further supported by the observations that Fas receptor expression on ovarian carcinoma cells and the Fas-antibody mediated killing of ovarian carcinomas were upregulated by antiestrogen treatment. In contrast, TX and TO did not stimulate the Fas-L expression on killer cells. The lack of correlation between Fas receptor expression and drug induced increase
418
of killer cell mediated cytolysis suggest the participation of the perforin/granzyme pathway. K562 cells are Fas negative and are not susceptible to anti-Fas mediated cytolysis. However, antiestrogen treatment significantly increased the NK cell mediated cytotoxicity of K562 cells. The total abrogation of cell death by chelation of extracellular Ca++ indicates that NK cell mediated cytolysis of K562 cells was dependent on perforin/grenzyme pathway. Similar results were obtained with the Fas receptor positive ovarian carcinomas. It is clear from our results that antiestrogens render the cancer cells more susceptible for killer cell mediated destruction, and thereby potentiate immune defence against cancer. On the basis of our combination immunotherapy experiments in tumor bearing mice we now conduct feasibility trials for the development of treatment protocols for the combination immunotherapy of cancer in man.
1.
INTRODUCTION
Breast cancer is the leading cause of death among women. Ovarian carcinoma is the fifth most diagnosed cancer among Canadian women. In 1999 there were about 2600 new cases and 1500 deaths due to ovarian cancer and both incidence and mortality rates increase with advancing age [1]. The clinical burden of prostate cancer in Canada is substantial and rising. Prostate cancer will develop in over 12% of Canadian men [2]. The traditional treatment of cancers (surgery, irradiation, chemotherapy), all followed by serious side effects. In order to improve the quality of life for patients with cancer and to increase the number of disease free survivals, we need to develop new treatment protocols. The interaction of cancer with the immune system has long been demonstrated. The classical experiments of Klein and his coworkers [3] showed that immunity to tumors can be transferred to non-immune animals by lymphocytes but not by serum. Almost three decades ago in our laboratory it was found that a guinea pig sarcoma was infiltrated by lymphocytes which had a cytotoxic effect on autologous tumor cells in vitro [4], but were inefficient when administered to tumor-bearing animals in immunotherapeutic experiments [5]. Studies of Mule and his coworkers [6] with isotope-labelled lymphocytes derived from immunized mice revealed that T lymphocytes localize specifically in the tumor which has been used for immunization. Modern approaches to cancer immunotherapy involve a wide range of manipulations of the immune system. Vaccines, killer cells and soluble mediators of cellular immunity, such as interleukin-2 or interferons are all being investigated [7, 8]. However, insufficient attention has been paid to the possibility of sensitizing tumor cells to direct immune-mediated attack.
2.
IMMUNE-DERIVED KILLER CELLS AND ANTIESTROGENS
2.1.
Natural killer (NK) cells
Natural killer cells originate in the bone marrow and are able to kill susceptible target cells without prior immunization. They do not recognize target cells in a strictly antigen specific manner thereby enabling a first line of defence quickly and without antigen presentation [9]. NK cells also express Fc~,RIII (CD 16); as a result, they are able to kill antibody-coated targets by antibody dependent cell-mediated cytotoxicity [10]. The lytic activity of NK cells can be enhanced by cytokines like interleukin (IL)-2, IL- 12, interferons [ 11-15].
419
2.2.
Lymphokine activated killer (LAK) cells
The treatment of normal lymphoid cells with high dose of IL-2 leads to the generation of LAK cells. These cytotoxic cells are capable to kill a wide range of targets, including tumor cells. The origin of LAK cells generated from peripheral blood lymphocytes (PBLs) is contraversial. Herrera [ 16] used low dose IL-2 and granulocyte-macrophage colony-stimulating factor (GM-CSF) to stimulate peripheral blood stem cells from cancer patients. LAK activity was significantly reduced by depletion of both CD4+ and CD8+ T cells and almost completely abolished after depletion of both subsets, suggesting that T cells and not NK cells were the main LAK precursors. In contrast LeFever [17] found that after T cell depletion of peripheral blood mononuclear cells (PBMCs) activated in lymphokine conditioned medium (LCM) and IL-2, the LAK cells lytic activity was enhanced. LAK cells have been tested in recent years for the treatment of various experimental and human tumors with encouraging results [8, 18, 19]. 2.3.
Cytotoxic T lymphocytes (CTLs)
Both CD4+ Thl and CD8+ cytotoxic T lymphocytes deliver a cytotoxic signal upon activation and getting in contact with target cells through T cell receptors [20]. Cytotoxic T lymphocytes are capable of attacking the membrane of target cells through the secretion of complement-like molecules called cytolysin or perforin. These molecules insert into the lipid plasma membrane of the target where Ca++-dependent polymerization takes place, leading to the creation of channels that are permeable for ions and other small molecules. CTL-s also secrete enzymes (granzymes or serine esterase) that contribute to cytotoxcicity. Moreover, CTL can cause programmed cell death (apoptosis) in target cells. Induction of apoptosis requires the active metabolic participation of the target [21, 22]. CTL-s are killing their targets in a major hystocompatibility complex (MHC) restricted manner. CD4+ and CD8+ cytotoxic T lymphocytes recognize the antigens in the context of MHC II and MHC I molecules, respectively. 2.4.
Antiestrogens
The non-steroidal antiestrogens, tamoxifen (TX) and toremifene (TO) [23, 24] belong to the triphenylethylene group of antiestrogens. These drugs are widely used for the therapy of breast cancer, prostatic and renal cell carcinomas. It is accepted that these drugs inhibit tumor growth by competing for the estrogen receptor [25]. However, these antiestrogens are known to affect the neuroendocrine and immune systems in vivo, which in turn may influence host resistance to cancer [26-28]. Clinical observation indicate that TX treatment can have a beneficial effect on breast carcinomas lacking the classical estrogen receptor [29]. The effect of TX on NK cells has been studied in several laboratories. Warri and Kangas [30] found that low doses of TO (0.1-10 mg/kg) did not interfere with the in vivo activation of NK cells by interferon-ct (IFN-ct) in NZB/NZW mice, but high doses (50 mg/kg for six weeks) were inhibitory. Rotstein [31] studied 23 breast cancer patients treated with TX for two years. A significantly lower NK activity was found against K562 target cells. In contrast, Berry [32] found in 17 breast cancer patients treated with TX for one month, that NK activity was significantly increased. Mandeville [33] suggested first that TX is capable of enhancing cell mediated cytotoxicity in vitro using human peripheral blood lymphocytes as effectors and K562 erythroleukemia cell line as target. TX was used in 10 -7 M and 10-8 M concentrations and was present during the assay.
420
Augmentation of cytotoxicity was observed when the cells were exposed simultaneously to IFN-alpha and TX. Baral and Vanky [34] isolated PBL-s from 20 patients with various types of cancer, both K562 and autologous tumor cells were used as targets. Lymphocytes were pretreated with TX (50, 100, 200 and 400 ng/ml) overnight. Untreated lymphocytes of four patients lysed autologous tumor cells, while after TX treatment lymphocytes from eight patients lysed autologous tumor cells. We also studied the effect of TX on killer cell mediated tumor cell lysis. NK cells in the spleen of Fischer female rats were used against the Yac-1 murine lymphoma target in the 51Cr-release assays. TX was used in different concentrations ( l n M - l p M ) . Target, effector or both cell population were treated with TX for 4-24 hours. The results indicated that NK cell-mediated target cell destruction can be enhanced by TX primarily through sensitization of the target for lysis [35]. In vitro experiments also revealed that TX and TO sensitize the SL2-5 lymphoma for NK cell mediated lysis. Lymphocytes prepared from DBA/2 mice were activated with 500 IU IL-2 for 24 hours before being used (Table I). The enhancement of NK activity by low doses of IL-2 treatment for 24-48 hours has also been observed by others [ 12, 13]. Table I
The effect of TX and TO on IL-2 activated NK cell mediated killing of the SL2-5 lymphoma.
No. of exp.
Percent specific 5~CRrelease Treatment None TX 1 ~tM(4h) TO 5 ~tM (4 h) TO I ~tM(4h)
Target treated 9+ 68• 65 • 55•
Effector treated
Both treated
9• 1 2• 0.5 • 0 0•
9• 60• c 61 • 7c 51+2 c
i c 8c c
Target/effector ratio: 1"25. 5~Cr release was determined after 4 hours of incubation at 37"C, 5% CO2. c = p < 0.001 (t test)
The effect of TX on IL-2 activated killer cell mediated cytotoxicity was also examined using the P815 murine mastocytoma cells as targets. We found that TX sensitized P815 cells for LAK cells destruction [36]. Albertini [37] also studied the effect of TX on LAK cell mediated lysis of ER+ and ER-human breast cancer cell lines. They concluded that treatment of the ER+ MCF-7 cells were modestly sensitized by TX for lysis by LAK cells. In other experiments we examined the effect of TX on CTL-mediated target cell lysis [38]. Fischer rat lymphocytes were used as responders and female Wistar rat cells as stimulators. Concanavalin-A stimulated Wistar lymphoblast were used as targets. CTL-s were harvested on day 5. The results indicate that CTL mediated cytotoxicity is amplified by pretreatment of target cells with TX, treatment of effector cells under the same conditions led to inhibition of cytoyoxicity. When both target and effector cells were pretreated, the amplification of target cell lysis was similar to the experiment where target cells were treated alone. We also performed in vivo experiments with the SL2-5 lymphoma and the P815 mastocytoma tumor in syngeneic DBA/2 mice. The mice were injected with lethal dose (107) of SL2-5 tumor cells s.c. and left untreated until the tumor grew to approximately 0.5 cm in diameter.
421
Relevant groups were fed with TX or TO (10 mg/kg/day) from day 14 until the end of the experiment. Two days later, 25-50 x 107 IL-2 stimulated spleen lymphocytes were given i.p., which was repeated twice 6 days apart. Drug treatment alone and immunotherapy alone exerted the inhibition of tumor growth. This antitumor effect was further enhanced when the drug and killer cell treatments were combined [39]. The results illustrate that amplification of NK cell-mediated tumor cell destruction by TX and/or TO, observed in vitro, can also be achieved in animals by oral administration of these drugs. On the basis of this observation one may suggest that TX and TO can also amplify the efficiency of killer cells that are present in tumor-bearing hosts [4, 17]. In 1980 Rosenberg [18] and his group described the generation of LAK cells and their lytic activity for fresh, autologous or syngeneic cells in patients and in mice. LAK cells did not destroy normal cells. In the animal models after LAK cells and IL-2 treatment they observed the regression of metastasis in liver and lung [40, 41]. Kim [42] and his coworkers studied the effect of TX (2 pg/ml in drinking water for 18 days) and IL-2 (50.000 IU i.p. 2x a day) on MCA-106 (methylcholantrene induced fibrosarcoma) bearing mice. They found that the number of metastasis in the lung was reduced by 66% in IL-2 group, by 30% in Tx group and by 95% in group treated with IL-2 and TX. We also studied the therapeutic effect of both LAK cells and antiestrogens on the P815 murine mastocytoma in syngeneic DBA/2 mice. Mice were injected with 10 4 tumor cells s.c. When the tumors reached 5 mm in size treatment was started with TX and TO and maintained till the end of experiment. Antiestrogens were given by gavage. Two days later they received 25 x 10 4 LAK cells i.p., which was repeated twice, 4 days apart. We found that LAK cells alone could not protect the animals; however, TX and TO alone were able to induce tumor regression, which was further enhanced by LAK treatment. In this latest group, 50% of the animals were tumor-free [43]. Many animal and human tumors are infiltrated by killer cells. Recent studies have shown that the stimulation of such killer cells with IL-2 in the presence of autologous tumor cells resulted in the generation of CTL-s [44]. On the other hand CTL-s were detected in the spleen of mice 12-14 days after tumor inoculation by the susceptibility of the cytotoxic action to treatment with anti-Thy 1.2 antibodies and complement [45]. Mononuclear cells isolated from the spleen of P815 tumor bearing mice on day 12-14 were capable of lysing 6-9% of untreated P815 target cells. Antiestrogen treatment significantly increased the lytic reactions. This increased cytotoxicity was abolished after the treatment of spleen cells with anti Thy-1.2 antibodies. On the basis of this observation, we treated tumor bearing mice with TX and TO and killer cells from tumor-bearing host. The results of in vivo experiments indicate that CTL injection alone induced partial regression of tumor growth when 5 x 105 spleen cells were injected i.p. and total regression when 1 x 10 6 cells were used. Antiestrogen treatment alone induced total regression of tumor in 25% of animals. Tumor growth was further inhibited when drug treatment was combined with CTL injection [45]. On the basis of our in vitro and in vivo (animal) experiments we extended our research towards patients with ovarian or lung cancers. An overwhelming body of evidence indicates that tumor-infiltrating lymphocytes (TIL) and tumor-associated lymphocytes (TAL) have a powerful killing effect on autologous tumor cells in vitro after activation by IL-2. Tumor cells and lymphoid cells were isolated from ascites of patients with ovarian cancer, using the discontinuous Ficoll-Hypaque density gradient (75%-100%). The tumor associated lymphocytes were than cultured in the presence of autologous tumor cells (TAL: tumor cell ratio of 10: 1) in RPMI medium supplemented with human recombinant IL-2 (100 IU/ml). The cytotoxicity of TAL derived cells were measured on days 0, 6 and 14. No cytotoxicity was observed using
422
the freshly isolated lymphocytes against the autologous tumor cells as found by others [46-51]. The cultured lymphocytes became cytotoxic by day 6 which increased by day 14. A significant enhancement of tumor cell lysis occured in 5 of 13 patients on day 6 and in 8 of 13 patients on day 14 when the autologous tumor cells were treated with TO. TX also enhanced tumor cell lysis but less efficiently. The killer cells were examined for the CD4 and CD8 markers and 13 out of 15 cell cultures turned out to be mixtures of CD4+ and CD8+ T lymphocytes. A low level of markers was detected on two cell cultures [52]. Cells of patients with ovarian cancer have also been treated with human interferon (IFN)-c~. IFN-ot alone did not sensitize target cells for lysis mediated by CTL-s, however, TX and TO treatment had a significant enhancing effect. When both target and effector cells were treated with IFN-ot a significant increase was detected in autologous tumor cell lysis mediated by CTL-s and this tumor cell destruction was further enhanced with joint treatment by IFN-c~ and TX or TO (Figure 1).
(16 patients) 300
b
TARGET TREATED
b
200
OA" 100
H
rr"
Z O C) u.. O
EFFECTOR TREATED 200
,oo
c
,00-
BOTH TREATED
c
_T_-
o
O 300 200
100
M
C IFN Tx To IFN IFN Tx To
Figure 1. The effect of tamoxifen, toremifene and interferon-a on the lysis of autologous ovarian carcinoma cells by tumor associated lymphocytes. C" control, IFN: interferon (100IU), TX: tamoxifen ( l p M ) , TO: toremifene (5pM), Tumor associated lymphocytes were activated with hrIL-2 for 14 days. Statistics: a = P < 0.05, b = P < 0.01, c = P < 0.001, T: S.E.
423
The cytolytic activity of tumor infiltrating lymphocytes (TIL) derived from patients with lung cancer was also studied. A total of 23 patients were involved. Some tumor did not grow lymphocytes and from others there was not sufficient amount of viable target cells to do detailed experiments. Fresh tumor tissue removed from patients during surgery was rinsed with cold tissue culture medium and cut into small pieces (2-3 mm3), placed in RPMI containing 0.1% collagenase type IV, 0.01% hyaluronidase type 5 and stirred for 2-6 hours at room temperature. The cells were separated from coarse tissue chunks and washed with Hank's balanced salt solution without Ca++ and Mg++. Tumor cells were separated on a 75-100% discontinuous Ficoll-Hypaque density gradients by centrifugation at 800 x G for 20 minutes [44]. Lymphocytes were cultured with autologous tumor cells in complete medium supplemented with 100 IU/ml of hr IL-2. The cytolytic activity of CTL-s was measured on days 14 and 21 (Figure 2). On day 14 a significant increase of tumor cell lysis was observed with the treatment of target cells with TX, TO or by combination treatment with IFN-ct. When both target and effector cells were treated with single agent, the amplification of tumor cell lysis was seen to the same degree as when target cells were treated alone. Combination treatment with antiestrogens and IFN-ct resulted in highly significant increase of tumor cell lysis by killer cells. On day 21 similar results were obtained [52]. •'
300
TARGET TREATED
DAY 14 (15
patients)
DAY 21 (19 patients)
a a
200 100
or" l-Z
300
nl EFFECTOR TREATED
8 200 100
400 -
BOTH TREATED
b
o 300 200 100
C IFN Tx To IFN IFN Tx To
n
C IFN Tx To IFN IFN Tx To
Figure 2. The effect of interferon-a tamoxifen and toremifene on the lysis of lung carcinoma cells by tumor infiltrating lymphocytes. C: control, IFN: interferon (100IU), TX: tamoxifen (1MM), TO: toremifene (5MM), Tumor associated lymphocytes were activated with hrIL-2 for 14 and 21 days. Statistics: a = P < 0.05, b = P < 0.01, T: S.E.
424
Lung cancer cell killing by CTL-s generated from draining lymph nodes was also studied. Lymph nodes removed during surgery were mechanically dissociated, and lymphocytes and tumor cells were separated with the same method as used for lung tissue. The killing activity of in vitro generated CTL was tested on days 14 and 21. The treatment of tumor cells with IFNalpha, TX or TO were all capable to increase tumor cell lysis in some patients, but not the others. However, when both target and effector cells were treated with antiestrogens jointly with IFN-alpha, the CTL-mediated target cell lysis was significantly increased [52]. At the time of the surgery heparinized blood was also taken from the patients with lung cancer. Lymphocytes were separated on Ficoll-Hypaque and cultured at 2 x 106/ml in medium supplemented with 500 IU of hr IL-2 in order to generate LAK cells. The cytotoxic activity of LAK cells could be influenced the least by TX, TO and IFN-c~. IFN-c~ increased cytotoxicity only when both target and effector cells were treated. The best lytic activity was achieved by joint treatment with IFN-c~ and TX or TO on day 21 [52]. It is clear from this overview, that the original observation of Mandeville [33] that TX is capable of enhancing the lysis of target cells by NK cells in vitro has been confirmed in our laboratory and we found that the enhancing effect is due to the sensitization of target cells to killer cell mediated lysis [38, 52, 53]. The sensitizing effect of TX and TO could be prevented by metabolic inhibitors (actinomycin-D and cycloheximide) and the sensitized cells showed an accelerated release of the nuclear label 3H-Thymidine into the culture medium [38, 54]. These results suggest that the antiestrogens activate metabolic pathways in the target cell, which lead to priming for the killer cell induced apoptosis. It has been established that CTL can induce target cell lysis, either by signalling the cell to commit suicide or by the release of cytolysins which initiate the lytic process through a membrane attack mechanism [55]. However, it is not clear how these pathways are activated and whether certain targets stimulate preferentially one and/or the other pathway. The suicide signal is delivered through the Fas membrane receptor on the target cell [56-58]. We found that antiestrogens have the potential to affect the Fas pathway in some tumors. This was demonstrated by a significant increase of lysis of human ovarian carcinoma cells induced by anti-human Fas (CH-11) monoclonal antibody [59]. Both TX and TO increased the mean percentage of cytotoxicity (Figure 3). These experiments indicate that in selected cases antiestrogens are capable of potentiating Fas receptor mediated tumor cell lysis. Further studies revealed that both IFN-c~ and antiestrogens stimulate Fas expression on tumor cells. In contrast, Fas-L expression was not stimulated by TX, whereas TO and combination treatment of TO and IFN-c~ increased the expression on killer cells (Table II). Fas receptor expression and drug-induced increase in tumor cell lysis do not correlate in every case. This lack of correlation suggest that other mechanisms than Fas/Fas-L must be involved in the modification of cytotoxicity by antiestrogens and IFN-~. The fact that TO is more efficient for sensitizing cancer cells for CTL but not for antibody-induced lysis, again, suggests that there are other additional mechanisms involved in sensitization for killer cell-induced lysis, not just signalling through the Fas receptor. In order to clarify the role of Fas receptor in sensitization we initiated experiments with the K562 target cells. According to the literature, K562 cells do not express the Fas receptor and even transfection with Fas does not make them susceptible to anti-Fas mediated lysis [59]. The lack of Fas receptor on K562 has been confirmed in our laboratory as well. Moreover, pretreatment with antiestrogens did not change Fas expression on K562 despite increased NK-mediated cytolysis. It was further confirmed in our laboratory that the K562 cytolysis and its antiestrogen enhancement is due to the Ca++-dependent perforin/granzyme pathway [60]. This has been demonstrated by total abrogation of cell death when extracellular Ca++ is chelated by EGTA/Mg++ (Figure 4).
425
4OO
g r 8
-[
300
o
2OO
o
Ir
100
r
F = Ab
TX + F ~ Ab
TO + F = Ab
Figure 3. The effect of antiestrogenic treatment in anti-Fas killing of human ovarian carcinomas. Fas Ab: Fas antibody, TX: tamofixen, TO: toremifene. Statistics: * P < 0.05, ** P < 0.01, T: S.E.
Table II
The effect of TX, TO and IFN-~t on Fas receptor and Fas-Ligand expression of human ovarian carcinomas and of autologous CTL.
Surface receptor/ligand expression Treatment FasL+ 2
Fas+ 1
100
(12)3
100
(11) 3
IFN-c~
132 + 14
(8)
122 + 10
TX
127 + 19
(8)
(8) (8)
TO
120 + 10
(12)
128 + 19
(11)
IFN-c~ + TO
133 + 15
(8)
134 _+ 18
(8)
Untreated
1. Fas+: Flow cytometric analysis of human Fas receptors on tumor cells. 2. FasL+: Flow cytometric analysis of human Fas Ligand on CTL-s. 3. No. of patients analyzed.
99+8
426
30
.~. r
20
m
x o o
O 10
0 No treatment K562 treated PBLs treated
Both treated
Figure 4. The effect of Ca++ chelation on NK cell-mediated K562 lysis with or without toremifene treatment. Results obtained from 1 experiment representative of 3 independent experiments are shown as mean (+ SEM) in triplicate wells. Solid and crossed bars represent the absence and the presence of EGTA/Mg++, respectively.
Recent studies revealed that ovarian tumor cell lines can be sensitized by cisplatin (CDDP) to anti-Fas antibody mediated cytotoxicity [61]. Others found that CDDP and etoposid (VP-16) sensitized the PC-3 and DU145 prostate cancer cell lines to the CTL mediated Fas/Fas-L cytotoxic pathway [56]. The sensitization was not due to the up regulation of Fas receptor on tumor cells. They found that IL-2 activated TIL and LAK cells are able to kill the untreated prostate cancer cells and this lytic activity was abolished by EGTA/MgC12 treatment, which indicates the perforin pathway of killing. After the sensitization of target cells with CDDP, TIL and LAK cells killed the target cells a Ca++ independent way. The conclusion was that the Fas/Fas-L pathway is masked by the perforin pathway which is a potent way and rule the overall cytotoxic activity in the case of drug-resistant tumors. The mechanism of antiestrogen activity has traditionally been ascribed to the ability to bind to and translocate the classical estrogen receptor (ER-c~) to the nucleus where it inhibits estrogen-mediated events. However, responses to antiestrogen treatment have been observed in patients with both ER-negative and ER-positive tumors. Moreover, antiestrogenic enhancement of cell-mediated cytotoxic reactions has been observed in both ER-negative and positive tumor targets. It is clear from our earlier studies that the classical ER is not a prerequisite for the antiestrogenic enhancing effect on immune cytolysis [36, 54]. A second estrogen receptor (ER-[3) has been described recently which can bind estradiol and can transactivate estrogenregulated reporter genes, although less efficiently than ER-c~. Also, ER-[3 has an overlapping
427
but non-identical tissue distribution to ER-ct [62]. It is possible that ER-[3 also plays a role in the enhanced cytolysis by antiestrogens. The sensitization of tumor cells against killer cell mediated lysis may modify gene expression and either upregulate death signal proteins or downregulate antiapoptotic genes. It is known that overexpression of Bcl-2 is protective against the mitochondrial cascade-mediated programmed cell death (PCD) [63]. Cheng and coworkers [64] described the conversion of Bcl-2 to Bax-like death effector by the enzyme CAS3 (CPP32), which took place after Fas ligation by antibody or interleukin-3 withdrawal. The carboxy-terminal cleavage product of Bcl-2 triggered PCD. Similar cleavage has been described during virus-induced PCD of Bcl-2 over expressing cells [65] and during drug-induced PCD [66]. This pathway of PCD was facilitated, rather than inhibited by the overexpression of Bcl-2. Chiu [67] has shown that Bcl-2 had no effect on anti Fas antibody or CTL mediated cytotoxicity. Berchem [68] found that increased Bcl-2 expression may be responsible for drug resistance of prostate cancer. Our preliminary experiments with human ovarian carcinoma cell targets and autologous CTL-s revealed, that TO treatment increased the expression of Bcl-2 and suppressed p53 in carcinoma cells. The retinoblastoma (RB) gene and the expression of classical estrogen receptor was unaffected by TO. Neither Fas-L nor the expression of the above genes were affected in killer T cells. Table III
Toremifene enhances Bcl-2 and suppresses p53 expression in human ovarian carcinoma cells.
Gene expr. % Positive cells
OV26
OV27
C
TO
Bcl-2
tr
p53
20
OV28
OV29
OV30
C
TO
C
TO
C
TO
C
TO
20
tr
15
tr
30
0
tr
20
tr
tr
0
-
r
0
20
0
50
tr
9_+0.3
4.1
8_+0.9
3.8
8_+1
2.1
7 _+2
3.5
0.8
1.8
1.0
2.1
2.6
3.2
2.8
3.9
9-+0.3
0.6
8_+0.9
0.7
8_+1
0.6
7_+2
0.5
Cytotoxicity TT: OV IFN ET: CTL IFN BT: CTL + OV IFN TT: aFas IFN + aFas
16_+0.6 0.7
1.1 1.8
16_+0.6
0.1 0.8
1.3
1.5
1.1
1.1
1.0
0.8
1.0
0.8
1.1
9-+0.3
2.7
8__.0.9
3.1
8__.1
3.2
7-+2
5.1
16-+0.6
2.8
1.7
3.0
1.3
3.3
4.1
8.1
5.2
5.8
1.3
4.5
18-+0.3
1.6
21_+0.3
1.3
7_+0.6
4.0
3.0
13_+0.3
2.0
0.8
2.0
0.6
1.7
3.8
8.8
3.7
1.3
3.5
13_+0.6 2.4
OV26-30 =ovarian; carcinomas; C=control; IFN=interferon; CTL=cytotoxic T lymphocyte; aFas=Fas antibody; TT=targets treated; ET=effectors treated; BT=both treated; tr =trace;- =not done Mean % cytotoxicity __ S.E. is given for control experiments. For other experiments the numbers indicate the cytotoxicity relative to control (Calculation: % release in E / % release in C).
The results shown in Table. 3 indicate that treatment of 5 ovarian carcinoma cells with TO increased or induced Bcl-2 expression in all cases, whereas p53 was suppressed in all 4 tumors so tested. Highest cytotoxicity values were found with the highest increase of Bcl-2 expression (30% of the cells became positive by immunochemistry after TO treatment) with no
428
detectable p53. "Wild" p53 is a differentiation and tumor suppressing gene, which has a proapoptotic function through the suppression of Bcl-2 and increased expression of Bax [69, 70]. Moreover, mutant P53, which frequently occurs in cancer cells, is known to interfere with the pro-apoptotic function of wild type P53.
3.
CONCLUSION
Our results showed that antiestrogens, TX and TO have the capacity to sensitize target cells for lysis by NK, LAK and CTL-s. Moreover TX and TO potentiated the immunotherapy of lethal cancer in syngeneic murine tumor host systems. The classical estrogen receptor (ER-c~) is not involved in target cell sensitization [71, 72] and it is assumed that an alternate receptor (ER-[5) is responsible. Antiestrogens also sensitized a significant proportion of human ovarian and lung carcinoma cells for lysis by autologous killer cells. Sensitization requires the active participation of the target cells and manifests itself in enhanced DNA degradation after the cytotoxic insult. Sensitization involves the overexpression of Bcl-2 protein. Both the Fas/Fas-L and the perforin/granzyme pathways of immune cytolysis are affected by antiestrogens. It is estimated that more than 130,000 new cases of cancer will be diagnosed and approximately 60,000 deaths will occur in one year in Canada. Cancer therapy still relies on the use of drugs that damage many other cells in the body in addition to the cancer and for this reason there are serious side effects. Only the immune system has the capacity to kill tumor cells with no side effects. Our discovery that antiestrogens render the cancer cells more susceptible for killer cell induced destruction was already applied to tumors in mice and a significant enhancement of cures resulted from joint therapy with killer cells and antiestrogens. On the basis of our results now we are conducting feasibility trials for the development of treatment protocols for the combination immunotherapy of cancer in man.
REFERENCES 1. 2.
3.
4. 5. 6.
7.
National Cancer Institute of Canada. Canadian cancer statistics (1999) Toronto: The National Cancer Institute: 1999. Grover SA, Coupal LC, Zowall H, Goldenberg L. The economic burden of prostate cancer in Canada: Forecasts from the Montreal Prostate Cancer Model. CMAJ 2000; 162: 977. Klein G, Sjogren HO, Klein E, Hellstrom KE. Demonstration of resistance against methylcholantrene-induced sarcomas in the primary autochthonous host. Cancer Res 1960; 20:1561. Berczi I, Strausbauch P, Sehon AH. Rejection of tumor cells in vitro. Science 1973; 180: 1289. Berczi I, Sehon AH. Effector and suppressor lymphoid cells in tumor bearing guinea pigs. Int J Cancer 1979; 23: 274. Mule JJ, Hellstrom I, Hellstrom KE. Cell surface phenotypes of radiolabelled immune long-lived lymphocytes that selectively localize in syngeneic tumors. Am J Pathol 1982; 107: 142. Austin FC, Finerty JF, Sogn JA. Cancer immunology: highlight of the NCI extramural immunology program. Cancer Invest 1990; 8: 207.
429
10.
11. 12.
13. 14. 15. 16.
17. 18.
19.
20. 21. 22. 23.
24.
25.
Hellstrom I, Hellstrom KE. Tumor immunology: an overview. Ann NY Acad Sci 1993; 690: 24. Yamaguchi A, Taga K, Mostowski HS, Bloom ET. Target cell induced apoptosis of interleukin-2 activated human natural killer cells: Roles of cell surface molecules and intracellular events. Blood 1996; 87: 5127. Vivier E, Morin P, O'Brien C, Druker B, Schlossman FS, Anderson P. Tyrosine phosphorylation of the Fc gamma R I I I (CD16): Zeta complex in human natural killer cells. Induction by antibody dependent cytotoxicity but not by natural killing. J Immunol 1991; 146: 206. Welsh RM. Regulation of virus infections by natural killer cells: a review. Nat Immun Cell Growth Regul 1986; 5: 169. Maghazachi AA, Vujanovic NL, Herberman RB, Miserodt JC. Lymphokine activated killer cells in rats, IV: developmental relationships among large agranular lymphocytes, large granular lymphocytes and lymphokine activated killer cells. J Immunol 1988; 140: 2846. Sarin A, Adler WH, Saxena RK. Lack of optimal activation of natural killer levels by interleukin-2 in rat spleen cells: evidence of suppression. Cell Immunol 1989; 122: 548. Cho D, Lee WJ, Halloran PJ, Trinchieri G, Kim YB. Enhancement of porcine natural killer cell activity by recombinant human and murine IL-2, Cell Immunol 1996; 172: 29. T. Dao T, Mehal WZ, Crispe IN. IL-18 augments perforin dependent cytotoxicity of liver NK-T cells. J Immunol 1998; 161: 2217. Herrera C, Garcia-Perez MJ, Ramirez R, Martin C, Alvarez MA, Martinez F, Gomez P, Garcia-Castellano JM, Torres A. Lymphokine activated killer (LAK) cell generation from peripheral blood stem cells by in vitro incubation with low-dose interleukin-2 plus granulocyte-macrophage colony-stimulating factor. Bone Marrow Transpl 1997: 19: 545. LaFever AV, Piaskowski VD, Casper JT, Truitt RL. Kinetic analysis of human IL-2 activated cytotoxic cells. ImmunoPharmacol Immunotoxicol 1991; 13:147. Rosenberg SA, Lotze MT, Muul LM, Chang AE, Avis FP, Leitman S. A progress report of 157 patients with advanced cancer using lymphokine activated killer cells and interleukin-2 or high dose interleukin-2 alone. New Engl J Med 1987; 316: 889. Rosenberg SA, Lotze MT, Muul LM, Leitman S, Chang AE, Ettinghausen SE. Observation on the systemic administration of autologous lymphokine activated killer cells and recombinant interleukin-2 to patients with metastatic cancer. New Engl J Med 1985; 313: 1485. E1-Khatib M, Stanger Z, Dogan H, Cui H, Ju S. The molecular mechanism of FasLmediated cytotoxicity by CD4+ Thl clones. Cell Immunol 1995; 163: 237. Russel JH. Internal disintegration model of cytotoxic lymphocytes induced target damage. Immunol Rev 1983; 72: 97. Zychlinsky A, Zheng LM, Liu C, Young JD. Cytolytic lymphocytes induce both apoptosis and necrosis in target cells. J Immunol 1991; 146: 393. Kallio S, Kangas L, Blanco G, Johansson R, Karjalainen A, Perila M, Pippo I, Sundquist H, Sodervall M, Toivola R. A new triphenylethylene compound, Fc-1157a. Hormonal effects. Cancer Chemother Pharmacol 1986; 17: 103. Kangas L, Nieminen AL, B lanco G, Gronroos M, Kallio S, Karjalainen A, Perila M, Sodervall M, Toivola R. A new triphenylethylene compound, Fc-1157a. Antitumor effects. Cancer Chemother Pharmacol 1986; 17: 109. Lerner LJ, Jordan VC. Development of antiestrogens and their use in breast cancer: 8th
430
26.
27.
28. 29.
30. 31. 32.
33. 34. 35. 36. 37.
38. 39. 40.
41.
42. 43. 44.
Cain memorial award lecture. Cancer Res 1990; 50: 4177. Paavonen T, Anderson LC. The estrogen antagonists, tamoxifen and Fcl157a, display estrogen like effects on human lymphocyte function in vitro. Clin Exp Immunol 1985; 61: 467. Valavaara R, Touminen J. A. Toivanen. The immunological status of breast cancer patients during treatment with a new antiestrogen, toremifene. Cancer Immunol Immunother 1990; 31: 381. Baral E, Nagy E, Berczi I. The effect of tamoxifen on the immune response. In: Tamoxifen: Beyond the antiestrogens (Kellen JA, ed.) Boston Birkhauser, 1996, pp. 137-178. Baral E, Glas U, Rutquist LE, Skoog L. Adjuvant therapy in postmenopausal patients with operable breast cancer. In: Fundamental Problems in Breast Cancer. Amsterdam, M Nijhoff, 1987, p 335. Warri A, Kangas L. Effect of toremifene on the activity of NK-cells in NZB/NZW mice. J Steroid Biochem 1990; 36: 207. Rotstein S, Blomgren H, Petrini B, Wasserman J, Von Stedingk VL. Influence of adjuvant tamoxifen on blood lymphocytes. Breast Cancer Res Treat 1988; 12: 75. G. Berry G, Green B J, Matheson DS. Modulation of natural killer cell activity by tamoxifen in stage I post-menopausal breast cancer. Eur J Cancer Clin Oncol 1987; 23: 517. Mandeville R, Galli SS, Chausseau JP. In vitro stimulation of human NK activity by an estrogen antagonist (tamoxifen). Eur J Cancer Clin Oncol 1984; 20: 983. Baral E, Vanky F. Effect of tamoxifen on the cell-mediated auto tumor lysis. J Clin Lab Immunol 1987; 22: 97. Baral E, Nagy E, Berczi I. Modulation of natural killer cell-mediated cytotoxicity by tamoxifen and estradiol. Cancer 1995; 75:591. Baral E, Nagy E, Berczi I. Modulation of lymphokine-activated killer cell-mediated cytotoxicity by estradiol and tamoxifen. Int J Cancer 1996; 66: 214. Albertini MR, Gibson DFC, Robinson SP, Howard SP, Tans KJ, Lindstrom MJ, Robinson RR, Tormey DC, Jordan VC, Sondel PM. Influence of estradiol and tamoxifen on susceptibility of human breast cancer cell lines to lysis by lymphokine activated killer cells. J Immunother 1992; 11: 30. Baral E, Nagy E, Berczi I. Target cells are sensitized for cytotoxic T-lymphocytes-mediated destruction by estradiol and tamoxifen. Int J Cancer 1994; 58: 64. Baral E, Nagy E, Kangas L, Berczi I. Immunotherapy of the SL2-5 murine lymphoma with natural killer cells and tamoxifen or tornemifene. Anticancer Res 1997; 17: 77. Lafneiere R, Rosenberg SA. Successful immunotherapy of murine experimental hepatic metastases with lymphokine activated killer cells and recombinant interleukin-2. Cancer Res 1985; 45: 3735. Rosenberg SA, Mule JJ, Spiess PJ, Reicher CM, Schwartz SL. Regression of established pulmonary metastases and subcutaneous tumor mediated by the systemic administration of high dose recombinant interleukin-2. J Exp Med 1985; 161:1169. Kim B, Varnaka P, Konrad C. Tamoxifen potentiates in vivo antitumor activity of interleukin-2. Surgery 1990; 108: 139. Baral E, Nagy E, Berczi I. Anti-estrogens enhance the therapeutic effect of lymphokineactivated killer cells on the P815 murine mastocytoma. Int J Cancer 1996; 67: 580. Ioannides CG, Platsoucas CD, Rashed S, Wharton TJ, Edwards CL, Freedman RS. Tumor cytolysis by lymphocytes infiltrating ovarian malignant ascites. Cancer Res 1991; 51: 4257.
431
45.
46.
47.
48.
49.
50.
51.
52.
53.
54. 55. 56.
57. 58.
59.
60. 61.
62.
Nagy E, Baral E, Kangas L, Berczi I. Anti-estrogens potentiate the immunotherapy of the P815 murine mastocytoma by cytotoxic T lymphocytes. Anticancer Res 1997; 17: 1083. Topalian SL, Muul LM, Solomon D, Rosenberg SA. Expansion of human tumor infiltrating lymphocytes for use in immunotherapy trials. J Immunol Methods 1987; 102: 127. Ioannides CG, Freedman RS, Platsoucas CD, Rashed S, Kim YP. Cytotoxic T cell clones isolated from ovarian tumor infiltrating lymphocytes recognize multiple antigenic epitopes on autologous tumor cells. J Immunol 1991; 146:1700. Ioannides CG, Fisk B, Jerome K, Irimura T, Wharton JT, Finn OJ. Cytotoxic T cells from ovarian malignant tumors can recognize polymorphic epithelial mucin core peptides. J Immunol 1993; 151: 3693. Vaccarello L, Y. Wang Y, Witeside TL. Sustained outgrowth of auto-tumor reactive T lymphocytes from human ovarian carcinomas in the presence of tumor necrosis factor and IL-2. Hum Immunol 1990; 28: 216. Kimura H, Yamaguchi Y, Fujisawa T. Cytotoxicity of autologous and allogeneic lymphocytes against cultured human lung cancer cells: optimal conditions for the production of cytotoxic lymphocytes. Gann 1984; 75: 1006. Kurnick JT, Kradin RL, B lumberg R, Schneeberger EE, Boyle LA. Functional characterization of T lymphocytes propagated from human lung carcinomas. Clin Immunol Immunopathol 1986; 38: 367. Baral E, Nagy E, Krepart GV, Lotocki RJ, Unruh HW, Berczi I. Antiestrogens sensitize ovarian and lung carcinomas for lysis by autologous killer cells. Anticancer Res 2000; 20: 2027. Berczi I, Nagy E, Baral E. The amplification of cell-mediated cytolysis by estradiol and tamoxifen, 8th International Congress of Immunology, August 23-28, 1992, Budapest, Hungary. Baral E, Nagy E, Kangas L, Berczi I. Immunotherapy of the SL2-5 murine lymphoma with natural killer cells and tamoxifen or toremifene. Anticancer Res 1997; 17: 77. Baral E, Nagy E, Berczi I. Enhancement of natural killer cell mediated cytotoxicity by tamoxifen. Cancer 1995; 75: 591. Uslu R, Borsellino N, Frost P, Garban H, NG CP, Mizutani Y, Belldegrun A, Bonavida B. Chemosensitization of human prostate carcinoma cell lines to anti-Fas-mediated cytotoxicity and apoptosis. Clin Cancer Res 1997; 3: 963. Ramsdell F, Seaman MS, Miller RE, Tough TW, Anderson MR, Lynch DH. Gld/gld mice are unable to express a functional ligand for Fas. Eur J Immunol 1994; 24: 928. Takahashi T, Tanaka M, Brannan CI, Jenkins NA, Copeland NG, Suda T, Nagata S. Generalized lymphoproliferative disease in mice, caused by a point mutation in the Fas ligand. Cell 1994; 76: 969. Munker R, Marini F, Jiong S, Savary C, Owen Schaub L, Andeeff M. Expression of CD95 (Fas) by gene transfer does not sensitize K562 to Fas-killing. Hematol. Cell Ther 1997; 39: 75. Haeryfar SMM, Nagy E, Baral E, Krepart G, Lotocki R, Berczi I. Antiestrogens affect both pathways of killer cell-mediated oncolysis. Anticancer Res 2000; 20:1849. Uslu R, Jewett A and Bonavida B. Sensitization of human ovarian tumor cells by subtoxic CDDP to anti-Fas antibody-mediated cytotoxicity and apoptosis. Gynecol Oncol 1996; 62: 282. Dotzlaw H, Leygue E, Watson PH, Murphy LC. Expression of estrogen receptor-beta in
432
63. 64. 65.
66. 67. 68.
69.
70.
71.
72.
human breast tumors. J Clin Endocrinol Metab 1997, 82:2371. Kroemer G. The proto-oncogene Bcl-2 and its role in regulating apoptosis. Nature Med 1997; 3: 614. Cheng EHY, Kirsch DG, Clem RJ, Ravi R, Kastan MB, Bedi A, Ueno K, Hardwick JM. Conversion of Bcl-2 to a Bax-like death effector by caspases. Science 1997; 278: 1966. Grandgirard D, Studer E, Monney L, Besler T, Fellay I, Borner C, Michel MR. Alphaviruses induce apoptosis in Bcl-2 overexpressing cells: Evidence for a caspasemediated, proteolytic inactivation of Bcl-2. EMBO J 1998; 17: 1268. Fujita N, Tsuruo T. Involvment of Bcl-2 cleavage in the acceleration of VP-16-induced U937 cell apoptosis. Biochem Biophys Res Comm 1998; 246: 484. Chiu UK, Walsh CM, Liu C, Reed JC, Clark WR. Bcl-2 blocks degranulation but not Fas-based cell-mediated cytotoxicity. J Immunol 1995; 154: 2023-9. Berchem JG, Bosseler M, Sugars LY, Voeller HY, Zeitlin S, Gelmann EP. Androgens induce resistance to Bcl-2 mediated apoptosis in LnCAP prostate cancer cells. Cancer Res 1995; 55: 735. Valassiadou KE, Stefanaki K, Tzardi M, Datseris G, Georgoulias V, Melissas J, Tsiftsis DD, Delides G, Kanavaros P. Immunohystochemical expression of p53, bcl-2, mdm2 and wafl/p21 proteins in colorectal adenocarcinomas. Anticancer Res 1997; 17: 251. Jonathan J, Herod O, Eliopoulos AG, Warwick J, Niedobitek G, Young LS, Kerr DJ. Prognostic significance of Bcl-2 and P53 expression in ovarian carcinoma. Cancer Res 1996; 56: 2178. Gu WZ, Chen Z, Tahir SK, Rosenberg SH, Ng SC. Synergistic effect of paclitaxel and 4-hydroxytamoxifen on estrogen receptor-negative colon cancer and lung cancer cell lines. Anticancer Drugs 1999; 10: 895. Toma S, Ugolini D, Palumbo R. Tamoxifen in the treatment of metastatic malignant melanoma: still a controversy? (Review), Int J Oncol 1999; 15: 321.
NewFoundationof Biology
433
Edited by Berczi I and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
The Influence of Reproductive Hormones on Asthma
GORDON T. FORD 1, CANDICE L. BJORNSON 2, IAN MITCHELL 3 and M. SARAH ROSE 4
1Department of Medicine, University of Calgary 2Department of Pediatrics, University of Calgary 3Department of Pediatrics, University of Calgary 4Department of Community Health Sciences, University of Calgary
ABSTRACT Changes in reproductive hormones have important influences on asthma in women. In early childhood, asthma is less common in girls (the female to male ratio is about 1: 2), but by menarche, the gender ratio is equal. Perimenstrual exacerbation(s) of asthma occurs in about 40% of women. Asthma symptoms and exacerbation(s) may increase perimenstrually with considerable reduction in expiratory airflow or change in non-specific airways hyperreactivity. Oral contraceptives may attenuate these changes, especially in those women who experience more severe exacerbations with menstruation compared to those not receiving oral contraceptives. This appears to be related to suppression of the luteal phase rise in reproductive hormones caused by oral contraceptives. Case reports demonstrating a reduction in the frequency and severity of asthma exacerbation(s) during treatment with continuous gonadotropin releasing hormone antagonist analogues may also provide a clue to the mechanism(s) of perimenstrually exacerbated asthma. Leukotrienes are implicated in the pathogenesis of perimenstrually associated asthma and agents which block the effects of leukotrienes may be useful in the treatment of perimenstrually associated asthma. Further support for a relationship between reproductive hormones and asthma is derived from studies of asthma in pregnancy. Changes in asthma severity ranging from improvement to deterioration to no change may occur. Women generally experience a similar pattern from one pregnancy to the next, whatever the change. At menopause, the epidemiology of asthma changes again. The age-adjusted risk of asthma is lower in post-menopausal women who have never received hormone replacement therapy compared to pre-menopausal women or to post-menopausal women who have used hormone replacement therapy. The risk of asthma is greatest in women with the longest duration of hormone replacement therapy and those currently using conjugated estrogens. Previous exposure to an oral contraceptive agent(s) is independently associated with a small increased risk of asthma. These data suggest that female reproductive hormones play a significant role in the pathophysiology of asthma in women. The findings may have significant implications in the management of asthma in women.
434
1.
INTRODUCTION
Asthma is a common chronic disease that presents clinically with wheezing, chest tightness, shortness of breath and/or cough. Physiologically asthma is a disease of the airways characterized by paroxysmal, reversible airflow obstruction secondary to airway inflammation and/or hyper responsiveness to a variety of stimuli. Pathologically, asthma appears as an eosinophilic bronchitis with inflamed airways, shedding of the airway epithelial lining and thickening of airway walls with various stages of remodeling from recurrent airway injury and repair [1]. Asthma exemplifies a medical disorder that exhibits marked variability of severity due to multiple different precipitating factors. A number of clinical observations and emerging basic and clinical research suggests that reproductive endocrine status, oral contraceptive agents (OCA) and reproductive hormone replacement therapy (HRT) can influence the natural history and severity of a number of diseases, including asthma. Support for a relationship between reproductive factors and asthma is derived from a number of studies of asthma at the time of menarche [2-9], perimenstrually [ 10-25], while receiving OCA [26-31 ], at the time of pregnancy [32-34], at menopause [3, 6-8, 35, 36], and those who have received HRT [36]. In addition, asthma severity may be greater in women as suggested by their longer duration of hospital stay during an exacerbation [7]. Furthermore, female gender has been shown to be the main patient characteristic associated with severe exacerbation(s) [37]. Considerable controversy still exists over the relationship between asthma and reproductive factors. This is because of the episodic natural history of asthma severity and exacerbation(s). While an individual's asthma severity during an exacerbation may coincide with a change in reproductive status, this may simply represent coincidental cycling of two episodic events. The purpose of this chapter is to examine the relationship(s) between asthma incidence, severity and exacerbation(s) as influenced by reproductive hormones in women.
2.
MENARCHE, THE REPRODUCTIVE YEARS AND ASTHMA EPIDEMIOLOGY
It is well known that there is gender difference in prevalence and incidence of asthma in children [6-8]. Many studies have shown that asthma prevalence in young children is twice as common in boys compared to girls (2: 1) which reverts to nearly one to one (1:1) at the time of menarche (Figure 1) [6-8]. In a Finnish Birth Cohort study, boys had a 64% higher cumulative incidence of asthma by the age of seven [5]. Males develop asthma at a younger age with the median onset between three years for males and eight years for females. However, from the age of 15 through 49 years, females have the higher incidence rates [8]. The incidence in girls improves less than boys during the adolescent years [6]. These results were confirmed in a British study which demonstrated a rise in the male to female incidence ratio from 1.23 up to age seven years, to 1.48 in 12 to 16 year olds with a reversal to 0.59 in 17 to 23 year olds [2]. Hospitalizations for asthma are highest in young children and almost two times higher in boys than girls, but the incidence increases in females and becomes higher than in males in the 20-50 year age group [7]. Further, seventy-five percent of all hospitalizations for asthma are women (Figure 2), and their severity of asthma as suggested by length of hospital stay is greater (Figure 3) [7]. Population-based data have demonstrated that this higher incidence and severity cannot be fully explained by diagnostic bias [8]. In the recent Formoterol and Corticosteroids Establishing Therapy (FACET) study [38] the subgroup of patients with severe exacerbations was further examined [37]. It was found that female gender was the main patient characteristic associated with an increased risk of having a severe exacerbation. Other patient characteristics
435
4,000 -'0 0 0
c:f
3,000
--
2,000
~
o
L_
o c
o
_.c <1
1-4
5-9
10-14
15.29
30-49
>50
A g e g r o u p , yr
Figure 1. The annual incidence rates per 100,000 person-years on the ordinate as a function of gender and age for the definite and probable asthma cases among Rochester residents, 1964 through 1983 as shown on the abscissa. Hatched bars = females; shaded bars - males. (Adapted from Yunginger et al., [8]; with permission).
1600" 1400' -o
1200'
.
o
t~ E
.~ 1000 ~ fit E -o < o d z
800 600'
400' 200' ..............
0
~0
do
3'0 ....... 40"....................... -io
60".............................. 70'
Age, y
Figure 2. The total number of asthma admissions by age for males (dotted line) and females (solid line) on the ordinate as a function of time in years from 1986 through 1989 on the abscissa. (Adapted from Skobeloff et al., [7]" with permission).
i n c l u d i n g p e a k e x p i r a t o r y flow v a r i a b i l i t y , i n h a l e d c o r t i c o s t e r i o d d o s e a n d a g e w e r e also p o s i t i v e l y a s s o c i a t e d b u t to a l e s s e r m a g n i t u d e
[37]. T h e s e d a t a a r g u e the p o s s i b i l i t y that
h o r m o n a l or o t h e r b i o c h e m i c a l g e n d e r d i f f e r e n c e s p l a y a s i g n i f i c a n t role in a s t h m a p r e v a l e n c e a n d severity.
436
8
JWq
W
"~
6"
CO '4" o
5"
o') c: n) ._t
4"
,,,L
0
,, ,,,,
,,
10
_
,,
,, i,,,,,,
20
,
4'o " 3"0 Age, y
io
6'0
7'0
Figure 3. The mean length of stay for asthma admissions in days by age for males (dotted line) and females (solid line) on the ordinate as a function of time in years from 1986 through 1989 on the abscissa. (Adapted from Skobeloff et al., [7]; with permission).
3.
PERIMENSTRUAL ASTHMA SEVERITY AND EXACERBATION(S)
In some women, the severity of asthmatic symptoms may increase during the few days preceding the start of the menstrual cycle. A perimenstrual exacerbation has been suggested by numerous case reports [10-13, 18]. Some investigators have reported that just prior to or during menses, up to forty percent of female asthmatics experience an increase in their symptoms of asthma or a deterioration of peak flows [17, 19, 20, 35]. More recent case reports demonstrating a reduction in frequency and severity of asthma exacerbation during treatment with continuous gonadotropin releasing hormone (GnRH) agonist analogues lend further support to the existence of menstrually exacerbated asthma [ 11, 12]. GnRH agonist analogues has been successfully used in the treatment of other chronic medical conditions including premenstrual syndrome, endometriosis, uterine leiomyoma, recurrent anaphylaxis, precocious puberty, prostrate and breast cancers, polycystic ovarian disease, functional bowel disease and in-vitro fertilization [39-42]. Apart from these highly suggestive case reports, the literature supporting the existence of perimenstrually exacerbated asthma has been inconsistent [16, 20, 23, 28]. Previous studies have been limited because they have not accounted for intrasubject correlation of repeated measurements in their statistical analysis. Failure to take this into account may have resulted in potentially unwarranted positive findings [24]. Pauli et al., [20] prospectively followed 11 asthmatic women for three or four consecutive menstrual cycles. They demonstrated a significant deterioration in asthma symptoms (p < 0.001) from the follicular to the luteal phase of the menstrual cycle. However the more objective measure of morning peak expiratory flow demonstrated only a slight but significant (p = 0.045)
437
decline of 15 L/min between seven to ten days premenstrually. This small change is of questionable clinical significance. They also assumed the independence of successive health measurement in the same individual in their statistical analysis. Eliasson et al., [16] studied 17 asthmatic women in a four month double-blind, crossover, placebo controlled study of the prostaglandin synthesis inhibitor sodium meclofenamate. They described a gradual decline in peak expiratory flow during the premenstrual period that reached a nadir during menstruation. A statistical analysis was not applied to confirm the existence of a menstrual cycle phase effect. Meclofenamate treatment was associated with a small, but significant (p < 0.025) improvement in peak expiratory flow of 12-19 L/min only during the four to six days premenstrually. Juniper et al., [28] followed 17 asthmatic women over two consecutive menstrual cycles. Although they demonstrated a worsening of asthma symptoms during menstruation, no difference was found in the mean provocation concentration of methacholine to cause a fall in FEV1 of 20% (PC 20). PC 20 measurements were performed one week before and one week after the start of menstruation. Although their sample size was small, they claimed adequate power (96% to 98%) to detect a PC 20 difference of 0.2 mg/mL should it have existed. A difficulty with their study was that the PC 20 was measured on two of the days of the menstrual cycle in those using oral contraceptives and they might thus have missed the menstrual effect. Weinmann et al., [23] also failed to demonstrate differences in forced vital capacity, FEV1, histamine challenge test or skin responsiveness when these measures were compared between the follicular and luteal phases of the menstrual cycle in nine women with mild asthma. Their failure to observe a menstrual effect might also be explained by the small sample and infrequent measurement. In a survey of 100 females with asthma in India, 23% described the perception of deterioration in asthma symptoms in relation to their menstrual cycle. The same patients showed a statistically significant decrease in their lung function in premenstrual and menstrual weeks as assessed by peak expiratory flow [14]. The mechanism(s) whereby gonadal steroids (or other hormones), fluctuating predictably through the menstrual cycle, alter bronchial reactivity remain the subject of hypothesis [16, 23, 28, 42-50]. Considerable attention has been devoted to progesterone, levels of which decline rapidly on the days before menstruation. This hormone also possesses mild immunosuppressive effects and has smooth muscle relaxant properties. Clinical studies of bronchial reactivity in well controlled asthmatics have failed to establish a progestogenic effect and the effects of progesterone therapy on asthma have been equally variable [16, 23, 28]. Beynon et al., [10] improved the control of asthma in a small group of women, whereas others [12] have shown deterioration when medroxyprogesterone was administered. Further deterioration with progesterone therapy in patients with recurrent idiopathic anaphylaxis suggests that the potential immunomodulatory effects of the hormone may not be beneficial [43]. The formulation and prescription regime of progestins may be critical in determining the bronchial response. However, failure of previous studies to demonstrate differences in bronchial reactivity between the follicular phase (low progesterone state) and luteal phase (high progesterone state) call into question the "progesterone theory" [16, 23, 28]. Support for the influence of hormonal factors on immune response is derived from the observed improvement of patients with recurrent idiopathic anaphylaxis rendered amenorrheic with continuous GnRH agonist analog [42, 43]. Estrogen and progesterone influence free cortisol levels as well as modifying beta-adrenergic receptor site density in rabbit lung [44]. Reproductive steroids have been shown to increase the potency of catecholamines in the
438
pig bronchus [45]. High levels of estrogen and progesterone may modify relative levels of prostaglandin (PG) F2 alpha and PGE2 [46]. Koullapis and Collins' finding that the plasma concentrations of the major metabolite of PGF2 alpha varied with estrogen concentration in women has not been confirmed [47]. Animal studies have demonstrated that estradiol can increase acetylcholine, cholinesterase activity and accelerate the release of secretory material from epithelial cells in the trachea [48-50]. Asthma presentations to the emergency department are least frequent when serum estradiol levels are at a sustained peak [22]. A four fold variation in asthma presentations during the perimenstral interval (days 26-04) was the most common time (46%) for women to present to the emergency department with asthma. This occurred when serum estradiol levels decreased sharply after a prolonged peak (during the postovulatory interval) suggesting that monthly variations in serum estradiol levels may influence the severity of asthma in adult females (Figure 4) [22]. Mast cell activation and degranulation have been shown to be a common feature in the functional layer of the endometrium perimenstrually [51]. Animal studies [52] have demonstrated estrogen receptors on peritoneal mast cells. Estradiol augments nonspecific mast cell degranulation in v i v o . These data suggest that systemic activation of mast cells during the menstrual cycle may be implicated in perimenstrually associated asthma. A possible mechanism for these observations may be related to perimenstral alterations in the type-1/type-2 cytokine balance as demonstrated by Agarwal et al., [53].
500 II
400 tl tl
..J
E
i 1 1
. 300-
t I I
l
E
i
O !
L_ cr
"' E
200#1,,,
,'
,,
' ,
'
1,,'!',,t,
/
i; , I
,
",o
/
i,-,
I
|
t
t
#
I i
I
I
I
1
\' i
I i "
\;
"
,
"~
"
100 -
0
'"i ....
6
i
8
.....
i
10
....
~
....
i
12 14
~
16
~
~
18 20
i
I
22
24
.......... 1
26
......
i
28
"i
......
2
4
Day of Cycle
Figure 4. Asthma presentations to the emergency department on the ordinate as a function of standardized day of the menstrual cycle on the abscissa (dashed line). Data for presentations are superimposed on the rise and fall of serum estradiol levels also shown on the ordinate as a function of standardized day of the menstrual cycle (solid line). The standardized menstrual cycle was divided into four intervals based on known fluctuations of serum estradiol. The intervals were: preovulatory (days 5-11); periovulatory (days 12-18); postovulatory (days 19-25) and premenstrual (days 26-04). (Adapted from Skobeloff et al., [22]; with permission).
439
Nakasato et al., [25] have implicated leukotrienes. These authors demonstrated significantly elevated levels of Leukotriene C4 (LTC4) during perimenstrual exacerbation of asthma than after recovery, as compared to control asthmatic women (those without premenstrual exacerbation). They hypothesized that an agent(s) that blocks the effects of leukotrienes may then be useful in the treatment of perimenstrual exacerbation of asthma. Oral administration of the leukotriene receptor antagonist (LTRA) pranlukast significantly reduced decreases in peak expiratory flow from baseline levels and improved asthma symptoms in their cohort of asthmatic women with perimenstrual exacerbation of asthma [25]. We have conducted a longitudinal study to confirm the relationship between menstrual phase and the exacerbation of asthma in fourteen women with asthma and regular menstrual cycles [24]. All women agreed to keep daily records of morning and evening peak expiratory flows, menstrual history, environmental exposure, medication, symptoms and hospitalization for four standardized consecutive menstrual cycles. A significantly increased risk of an asthma exacerbation occurred during menses [odds ratio = 1.79 (95% CI 1.15-2.79) p = 0.01] using a random effects logistic regression model that controlled for environmental exposures, calendar month and previous day exacerbation. The estimated probability of experiencing an exacerbation for each standardized day of the menstrual cycle demonstrated that the highest estimated prevalence occurred on the first day of menstrual flow and decreased thereafter. Conversely, the lowest prevalence of exacerbation occurred mid-cycle, between days 17 and 19, thereafter increasing until the premenstrual phase, (days 25-28) when the prevalence of exacerbation(s) was approximately the same as the first day of menstruation (Figure 5). There was no interactive effect of self reported environmental exposure, calendar month or oral steroid use on the association between asthma exacerbation and menstrual cycle phase. These results establish the relationship between asthma exacerbation and menstrual phase based on the significantly increased risk of asthma exacerbation perimenstrually.
0.30
i\ i ;
" E
...,..--~ >,
0.25 !!
i-.,
o t-I ,
= 0.20 O ,,_., ID
.., /]
.."// .,..//
,
':',. :
'.,'..
. . . ./. . . . !/: '\ i
E
tu
...
0.15
.........
Estimatecl (_Crude). Probability
Smoothed izsl)mate
,. j ,, .:
]
9
9
0.10
r 1
4
i 8
12
i 16
.
.
20
24
28
Standardized Day
Figure 5. The estimated probability of experiencing an asthma exacerbation as shown on the ordinate as a function of standardized day of the menstrualcycle on the abscissa. The dashed line represents the estimate (crude) probability of experiencing an asthma exacerbation. The solid line represents a smoothed estimate of the probability of experiencing an asthmaexacerbation. (Adaptedfrom Rose et al., [24]; with permission).
440
4.
ORAL CONTRACEPTIVE AGENTS (OCA) AND ASTHMA
There have been conflicting case reports of patients with perimenstrually associated asthma who have been treated with OCA, some showing dramatic improvement [27], while others showed deterioration [29, 30]. Clinical studies of bronchial reactivity in well controlled asthmatics have failed to show a progestogenic effect [28]. More recently, Tan et al., [26] compared airway reactivity to adenosine monophosphate (AMP) in women with asthma and natural menses to those in a second group who were taking combined OCA. The patients receiving the OCA had an attenuated cyclical change in airway reactivity to AMP and reduced diurnal variation in peak expiratory flow compared to those not receiving OCA. This appeared to be related to the suppression of the normal luteal phase rise in reproductive hormones caused by the OCA.
5.
PREGNANCY AND ASTHMA
Pregnancy appears to have variable effects on the severity of asthma, although in many circumstances the condition is worse during pregnancy [32-34]. A summary of nine retrospective studies in which 25 or more asthmatic pregnancies were reported revealed that of the 1,087 pregnancies, 36% improved, 41% remained unchanged and 23% became worse [32]. Prospective data from Gluck and Gluck [32] demonstrated a 43% worsening of their 47 patients with 43% remaining unchanged and 14% improving. In their severe group 83% became worse (significantly different from the mild and moderate groups) with a significant number requiring hospitalization (Figure 6). It appears that in those women whose asthma is most severe, there will likely be no change or a deterioration in asthma severity during pregnancy; however; whatever the change, women generally have consistent pregnancy experiences (a pattern that appears during one pregnancy likely occurs during the next).
I00 i ~ BETTER DUNCHANGED mmWORSE
5
a FREOUENCY IIILO (25) OF ATTACKS <MONTHLY
MOD. [10) SEVERE(12) MONTHLY WEEKLY SEVERITYOF ASTHMA
Figure 6. The percent change in asthma severity due to pregnancy on the ordinate as a function of clinical antepartum asthma severity on the abscissa. Dotted bars = better; hatched bars = unchanged; solid bars - worse. (Adapted from Gluck and Gluck [32]; with permission).
441
6.
MENOPAUSE, HORMONE REPLACEMENT THERAPY (HRT) AND ASTHMA
Troisi et al., [36] have recently described another approach to the study of asthma and reproductive hormones. They evaluated the association of HRT and asthma incidence prospectively in a cohort of pre and postmenopausal registered nurses aged 34 to 68 from the Nurses Health Study Cohort initiated in 1976 [54]. The age-adjusted incidence of asthma was lower in postmenopausal women than in premenopausal women. This difference was greatest amongst postmenopausal women who reported never using HRT compared to women reporting current or past use. This suggests that some of the benefit of menopause was lost with HRT. Post and current postmenopausal HRT was associated with an increased risk of asthma amongst the cohort reporting natural menopause. The risk of development of asthma was greatest in those women reporting long duration or current use of HRT (conjugated estrogens) (Figure 7). Reported past use of oral contraceptives was associated with a modestly increased risk of the development of asthma. These data suggest that estrogen may play a role in the pathophysiology of asthma, especially in the elderly where asthma has been demonstrated to be more severe [35] and that long term use or high dose of HRT may increase the subsequent risk of asthma [3, 35]. Relative Risk (no. of cases) 3 128)
2.5
..........................
2
0.5 0
(~)
................................... ................
141). . . . . . . . . . . . . . . .
........................................................................
Never
0.30
0.625
1.25+
Conjugated Estrogen Doses (mg) Test for trend includes never users; p=O.O07.
Figure 7. The age-adjusted relative risk of new onset asthma (95% confidence interval) as shown on the ordinate as a function of conjugated estrogens as shown on the abscissa. (Adapted from Troisi et al., [36]" with permission).
.
SUMMARY AND POTENTIAL MECHANISM(S) OF HORMONALLY RELATED ASTHMA
Many mechanisms have been proposed for the change in incidence, worsening asthma symptoms and reduced airflow in women. These include psychological, immune, and reproductive hormone mechanisms. We have reviewed the data documenting the change in asthma incidence, severity and exacerbation in females at puberty and as adults [2-9, 37]. This increased incidence remains higher throughout the reproductive years [8]. Adult women
442
represent approximately 75% of all hospital admissions for acute asthma [7] and population-based data have demonstrated that the higher incidence of asthma cannot be fully explained by diagnostic bias [8]. Further, female gender has been shown to be the main patient characteristic associated with an increased risk of severe exacerbation(s) [37]. Asthma severity has shown to change with alteration in short-term or long-term reproductive states including: the menstrual cycle [9-11, 13-22]; use of OCA [26, 27, 29, 31]; pregnancy [32-34]; and at menopause with or without HRT [6-8, 36]. Perimenstrually exacerbated asthma may be attenuated with GnRH agonist analogues [11, 12]. Mast cell activation and degranulation can be influenced by reproductive hormones in animals [52]. Systemic activation of mast cells by reproductive hormones during the menstrual cycle may cause the release of cellular mediators resulting in cyclical airway inflammation [25, 51, 53, 55]. Ultimately, this may be the final pathophysiological pathway in perimenstrually associated asthma. New basic, clinical and multicenter randomized trials will be required to confirm the effects of reproductive hormones on asthma.
ACKNOWLEDGEMENT The authors wish to thank Mr. Indraneel Datta for his help in the preparation of the manuscript.
REFERENCES 1. 2. 3. 4. 5.
6. 7. 8.
9. 10. 11.
Canadian Asthma Consensus Group. Canadian Asthma consensus report. Can Med Assoc J 1999; 161 : S 1-$62. Anderson H, Pottier A, Strachan D. Asthma from birth to age 23: incidence and relation to prior and concurrent atopic disease. Thorax 1992; 47: 537-542. Becklake MR, Kauffmann F. Gender differences in airway behavior over the human life span. Thorax 1999; 54:1119-1138. Dodge RR, Burrows B. The prevalence and incidence of asthma and asthma-like symptoms in a general population sample. Am Rev Respir Dis 1980; 122: 567-575. Gissler M, Jarvelin MR, Louhiala P, Hemminki E. Boys have more health problems in childhood than girls: follow-up of the 1987 Finnish birth cohort. Acta Paediatr 1999; 88:310-314. Martin AJ, McLennan LA, Landau LI, Phelan PD. The natural history of childhood asthma to adult life. BMJ 1980; 280:1397-1400. Skobeloff EM, Spivey WH, St Clair SS, Schoffstall JM. The influence of age and sex on asthma admissions. JAMA 1992; 268: 3437-3440. Yunginger JW, Reed CE, O'Connell EJ, Melton LJ, O'Fallon WM, Silverstein MD. A community-based study of the epidemiology of asthma. Am Rev Respir Dis 1992; 146: 888-894. yon Mutius E. Progression of allergy and asthma through childhood to adolescence. Thorax 1996; 51: s3-s6. Beynon HLC, Garbett ND, Barnes PJ. Severe premenstrual exacerbations of asthma: effect of intramuscular progesterone. Lancet 1988; 2: 370-372. B lumenfeld Z, Bentur L, Yoffe N, Alroy G, Rubin AHE. Menstrual asthma: use of a gonadotropin-releasing hormone analogue for the treatment of cyclic aggravation of bronchial asthma. Fertil Steril 1994; 62: 197-200.
443
12.
13. 14. 15. 16.
17. 18. 19. 20.
21. 22.
23. 24. 25.
26.
27. 28.
29.
30. 31. 32. 33.
Edwards AL, Rose MS, Donovan LE, Ford GT. Premenstrual exacerbation of life-threatening asthma: Effect of gonadotropin releasing hormone analogue therapy. Can Respir J 1996; 3: 203-206. Enright T, Lim DT, Devnani R, Mariano R. Cyclic exacerbation of bronchial asthma (clinical conference). Ann Allergy 1987; 58: 405-406,452-453. Agarwal AK, Shah A. Menstrual-linked asthma. J Asthma 1997; 34: 539-545. Eliasson O, Scherzer HH, Degraff AC. Morbidity in asthma in relation to the menstrual cycle. J Allergy Clin Immunol 1986; 77: 87-94. Eliasson O, Densmore MJ, Scherzer HH, Degraff AC. The effect of sodium meclofenamate in premenstrual asthma: a controlled clinical trial. J Allergy Clin Immunol 1987; 79: 909-918. Gibbs CJ, Coutts II, Lock R, Finnegan OC, White RJ. Premenstrual exacerbation of asthma. Thorax 1984; 39: 833-836. Lenoir RJ. Severe acute asthma and the menstrual cycle. Anaesthesia 1987; 42: 1287-1290. Hanley SP. Asthma variation with menstruation. Br J Dis Chest 1981; 75: 306-308. Pauli BD, Reid RL, Munt PW, Wigle RD, Forkert L. Influence of the menstrual cycle on airway function in asthmatic and normal subjects. Am Rev Respir Dis 1989; 140: 358-362. Rees L. An aetiological study of premenstrual asthma. J Psychomatic Res 1963; 7: 191-197. Skobeloff EM, Spivey WH, Silverman R, Eskin BA, Harchelroad F, Alessi TV. The effect of the menstrual cycle on asthma presentations in the emergency department. Arch Intern Med 1996; 156: 1837-1840. Weinmann GG, Zacur H, Fish JE. Absence of changes in airway responsiveness during the menstrual cycle. J Allergy Clin Immunol 1987; 79: 634-638. Donovan LE, Rose SM, Cowie RL, Edwards AL, Ford GT. Menstrually exacerbated asthma: fact or fiction? Am J Respir Crit Care Med 1994; 149: A914. Nakasato H, Ohrui T, Sekizawa K, Matsui T, Yamaya M, Tamura G, Sasaki H. Prevention of severe premenstrual asthma attacks by leukotriene receptor antagonist. J Allergy Clin Immunol 1999; 104: 585-588. Tan KS, McFarlane LC, Lipworth BJ. Modulation of airway reactivity and peak flow variability in asthmatics receiving the oral contraceptive pill. Am J Respir Crit Care Med 1997; 155: 1273-1277. Matsuo N, Shimoda T, Matsuse H, Kohno S. A case of menstruation-associated asthma: treatment with oral contraceptives. Chest 1999; 116: 252-253. Juniper EF, Kline PA, Roberts RS, Hargreave FE, Daniel EE. Airway responsiveness to methacholine during the natural menstrual cycle and the effect of oral contraceptives. Am Rev Respir Dis 1987; 135: 1039-1042. " Derimanov GS, Oppenheimer J. Exacerbation of premenstrual asthma caused by an oral contraceptive. Ann Allergy Asthma Immunol 1998; 81: 243-246. Horan JD, Lederman JJ. Possible asthmogenic effect of oral contraceptives. Can Med Assoc J 1968; 99: 130-131. Collins LC, Peiris A. Bronchospasm secondary to replacement estrogen therapy. Chest 1993; 104: 1300-1302. Gluck JC, Gluck PA. The effects of pregnancy on asthma: a prospective study. Ann Allergy 1976; 37: 164-168. Schatz M, Harden K, Forsythe A, Chilingar L, Hoffman C, Sperling W, Zeiger RS. The course of asthma during pregnancy, post partum, and with successive pregnancies: a prospective analysis. J Allergy Clin Immunol 1988; 81: 509-517.
444
34.
35. 36.
37.
38.
39.
40. 41.
42.
43. 44.
45. 46. 47.
48.
49. 50. 51.
52.
Juniper EF, Daniel EE, Roberts RS, Kline PA, Hargreave FE, Newhouse MT. Improvement in airway responsiveness and asthma severity during pregnancy. Am Rev Respir Dis 1989; 140:924-931. Bailey WC, Richards JM, Brooks CM, Soong S, Brannen AL. Features of asthma in older adults. J Asthma 1992; 29:21-28. Troisi RJ, Speizer FE, Willett WC, Trichopoulos D, Rosner B. Menopause, post menopausal estrogen preparations, and the risk of adult onset asthma: a prospective cohort study. Am J Respir Crit Care Med 1995; 152: 1183-1188. Tattersfield AE, Postma DS, Barnes PJ, Svensson K, Bauer C, O'Byrne PM, Lofdahl C, Pauwels RA, Ullman A. Exacerbations of Asthma. A descriptive study of 425 severe exacerbations. Am J Respir Crit Care Med 1999; 160: 594-599. Pauwels RA, Lofdahl C, Postma DS, Tattersfield AE, O'Bryne P, Barnes PJ, Ullman A. Effect of inhaled formoterol and budesonide on exacerbations of asthma. N Engl J Med 1997; 337: 1405-1411. Mortala JF, Girton L, Fischer U. Successful treatment of severe premenstrual syndrome by combined use of gonadotropin-releasing hormone agonist and estrogen/progestin. J Clin Endocrinol Metab 1991; 71: 252A-252F. Casper RF. Clinical uses of gonadotropin-releasing hormone analogues. Can Med Assoc J 1991; 144: 153-8. Mathius JR, Ferguson KL, Clench MH. Debilitating "functional" bowel disease controlled by leuprolide acetate, gonadotropin-releasing hormone (GnRH) analog. Digestive Diseases and Sciences 1989; 34: 761-766. Slater JE, Raphael G, Cutler GB, Loriaux DL, Meggs WJ, Kaliner M. Recurrent anaphylaxis in menstruating women: treatment with a luteinizing-releasing hormone agonist-a preliminary report. Obstet Gynecol 1987; 70: 542-546. Meggs WJ, Pescovitz OH, Metcalfe D, Loriaux DL, Cutler G, Kaliner M. Progesterone sensitivity as a cause of recurrent anaphylaxis. N Engl J Med 1984; 311: 1236-1238. Moawad AH, River LP, Kilpatrick SJ. The effect of estrogen and progesterone on B-adrenergic receptor activity in rabbit lung tissue. Amer J Obstet Gynecol 1982; 144: 608-613. Foster PS, Goldie RG, Paterson JW. Effect of steroids on beta-adrenoreceptor-mediated relaxation of pig bronchus. Br J Pharmacol 1983; 78: 441-445. Senna G, Mezzelani P, Andri G, Andri L. Bronchial asthma in women: peculiar aspects. Recenti Prog Med 1989; 80:366-371. Koullapis EN, Collins WP. The concentration of 13, 14-dihydro-15-oxo-prostaglandin F2 alpha in peripheral venous plasma throughout the normal ovarian and menstrual cycle. Acta Endocrinologica 1980; 93:123-128. Abdul-Karim RW, Marshall LD, Nesbitt REL. Influence of estradiol-17 [3 on the acetylcholine content of the lung in the rabbit neonate. Amer J Obstet Gynec 1970; 107:641-644. Abdul-Karim RW, Drucker M, Jacobs RD. The influence of estradiol-17 [3 on cholinesterase activity in the lung. Amer J Obstet Gynec 1970; 108: 1098-1101. Vidic B, Kapur SP, Jenkins DP. Estrogen and tracheal secretion: the effect of estrogen on the epithelial secretory cells of the rat trachea. Cytobiologie. 1978; 18: 10-21. Jeziorska M, Salamonsen LA, Woolley DE. Mast cell and eosinophil distribution and activation in human endometrium throughout the menstrual cycle. Biol Reprod 1995; 53: 312-320. Vliagoftis H, Dimitriadou V, Boucher W, Rozniecki JJ, Correia I, Raam S, Theoharides
445
53. 54.
55.
TC. Estradiol augments while tamoxifen inhibits rat mast cell secretion. Int Arch Allergy Immunol 1992; 98: 398-409. Agarwal SK, Marshall GD. Perimenstrual alterations in type-I/type-2 cytokine balance of normal women. Ann Allergy Asthma Immunol 1999; 83: 222-228. Colditz GA, Stampfer MJ, Willett WC, Rosner B, Speizer FE, Hennekens CH. A prospective study of parental history of myocardial infarction and coronary heart disease in women. Am J Epidemiol 1986; 123: 48-58. Galli SJ. New concepts about the mast cell. N Engl J Med 1993; 328: 257-265.
This Page Intentionally Left Blank
New Foundationof Biology
447
Edited by Berczi I and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
Neuro-immunopathogenesis in Autism
VIJENDRA K. SINGH
Department of Biology & Biotechnology Center, Utah State University, Logan, Utah, USA 84322-5305
ABSTRACT Autism is an idiopathic brain disorder of unknown cause and etiology. It causes severe deficits of higher mental functions, as well as behavioral manifestations. Based on our ongoing research of a reciprocal relationship between nervous system and immune system, we studied autism as a neuro-immune dysfunction syndrome (NIDS) in which autoimmunity to brain was strongly implicated. Because myelination is one of the most important postnatal developmental events, we specifically focused on autoimmunity to brain myelin. We conducted laboratory evaluation of brain autoantibodies and virus serology in approximately 250 autistic children and 150 controls (healthy and non-autistic disease controls). The brain autoantibody study included detection of antibodies to neuron-axon filament proteins (NAFP) and three main myelin components: myelin basic protein (MBP), galactocerebrosides (GC) and 2, 3-cyclic nucleotide 3'-phosphohydrolase (CNP). The virus serology included measurement of IgG antibodies to measles virus, human herpesvirus-6, cytomegalovirus and rubella virus. We found that autoantibodies to MBP were selectively present in up to 80% of the autistic children, but they were only rarely detected in the controls. Autoantibodies to NAFP and GC were also detected, but they were found non-specifically in control subjects also. Autoantibodies to CNP were absent in both groups of children. Thus MBP is potentially a candidate autoantigen in autism. Regarding virus serology, autistic children had a significantly higher level of measles virus antibodies as compared to controls; however, the antibody level of other three viruses did not significantly differ between the two groups. This suggested a temporal link of measles virus with autoimmunity in autism. The examination of brain autoantibody and virus serology data revealed that there was a serological association between measles virus and MBP autoantibodies, i.e., the higher the measles virus antibody level the greater the chance of MBP autoantibody. Collectively, these observations led us to speculate that an autoimmune response, presumably secondary to an atypical measles virus infection, may cause autism. The idea that autism is an autoimmune disorder is further strengthened by the fact that many autistic children respond well to treatment with immune modulating drugs. Considering MBP autoantibodies as an index of autoimmunity to myelin, an open-label trial of oral Sphingolin (myelin containing autoantigen) is under assessment-preliminary results are encouraging with significant improvement of behavioral characteristics in the autistic people. In conclusion, autism involves a neuroautoimmune response that occurs at the neuro-immune biology interface. Clinically, therefore, there is enormous potential for restoring brain function in autistic people through immunology.
448
1.
INTRODUCTION
Autism is an idiopathic developmental disorder of the central nervous system (CNS), which affects an estimated one half million people, mainly children, today in the United States. The disorder is defined by behavioral characteristics, largely because pathognomic biomedical markers have not been identified. Autistic characteristics include behavioral symptoms under four main headings: impairments of speech, language and cognitive capabilities; disturbances of developmental rates and/or sequences; aberrations of responses to sensory stimuli; and disturbances of social interaction with other people. Although autism is generally believed to be a multifactorial disorder, the etiology and pathogenesis of autism is not well known. The contributing factors include genetics, immunity, infections, chemical toxicity and as yet other unknown factors; hence, there is no cure for autism. Historically speaking, all diseases in man can be viewed as a net result of deviant complex interplay of three main factors: population predisposition (genetic factors), resistance to natural environment (immune factors), and mental attitudes (neural factors). This led to the concept of "neuroimmunomodulation" which refers to a reciprocal structural-functional relationship between the immune system and nervous system. The evidence for neuroimmunomodulation now exists at virtually every level: structural, cellular, molecular, and physiological. Previously, we postulated that this relationship is of pathophysiologic significance as it fundamentally denotes the existence of immunologic abnormalities in neurologic or neuropsychiatric diseases [1]. Furthermore, we [2] conceptualized that the diseases of the nervous system can be distinguished into three main categories: neurodevelopmental diseases that result from abnormal development of the brain; neurodegenerative diseases that show a progressive downward course of a neurologic function; and neuropsychiatric diseases which affect behavior or psyche [2]. In each case, we speculated autoimmunity to be a critical pathogenic factor [1, 2], and now autoimmunity is indeed perceived as a prime culprit of neuro-immune dysfunction [3]. In view of our ongoing research of neuroimmunology or neuro-immune biology, we studied autism as a model of neuro-immune dysfunction syndrome (NIDS), in which an abnormal or inappropriate interplay between the immune system and nervous system is important [2]. We hypothesized that brain-specific autoimmunity plays a key role in the pathogenesis of the disorder, hence the neuro-immunopathogenesis of autism [4]. Abnormalities of T cells, NK cells, cytokines, immune activation markers, and serum immunoglobulins are commonly found in children with autism [5-12]. Additionally, some autistic children also respond to treatment with immunomodulating agents [13, 14]. They also have organ-specific autoantibodies to brain [4, 15-18], but the nature of autoantigen is not known. To that end, we found a significantly higher incidence of autoantibodies to myelin basic protein (anti-MBP) that led us to postulate that an autoimmune response to brain myelin, secondary to a prenatal or postnatal viral infection is involved in the immunopathogenesis of autism [4, 17]. Autoimmunity is commonly believed to be triggered by viral infections and certain viral infections have been associated with autism [ 17, 19, 20]. We have now extended this research to identify autoantigen in myelin and to search for viral etiology. In this paper, we describe that MBP is potentially a candidate autoantigen in autoimmune autism, MBP-directed autoimmune response is likely related to an atypical measles infection, and MBP-containing autoantigen alleviates certain autistic characteristics. Thus autism involves an autoimmune pathology of brain myelin, presumably secondary to a measles infection.
449
2.
EXPERIMENTAL METHODS
2.1.
Subjects in the study
The study included mainly two groups of subjects: autistic patients and normal controls. Autistic patients were children (age 2.5 to 12 years) with the diagnosis of autism, including autistic regression. Normal controls were normal children (age 3 to 14 years) and/or normal adults (age 23 to 48 years) with firm physical and mental health. We used serum samples that were procured after proper informed consent from adult subjects or parents/guardians for children in the study. For ethical purposes, we secured approval of the Institutional Review Board (IRB) that oversees human research protocols at each academic institution. Whenever needed, human participation was limited to blood draw by venipuncture only. We collected sera from approximately 250 human subjects and stored them at -20~ until further use. At the time of blood drawing, human subjects taking prescription medications (neuroleptic or antipsychotic drugs), alcohol or drugs of substance abuse were excluded from the study. The clinical diagnosis of autism, including autistic regression, was made essentially according to standard DSM-III-R criteria by child psychiatrists and/or psychologists. 2.2.
Detection of autoantibodies
Brain autoantibodies were analyzed according to our previously published methods [4, 15, 17]. Autoantibodies to MBP, NAFP and CNP were analyzed by western analysis, followed by immunoblotting assays. For this purpose, the screening antigens were bovine brain MBP, bovine spinal cord NAFP, and rat brain CNP. GC, owing to glycolipid nature, could not be resolved by western analysis; hence, an ELISA method was used to detect autoantibodies to GC. For this purpose, the screening antigen was bovine brain GC (Galactocerebrosides, Product code #C-4905, Sigma-Aldrich, St. Louis, MO). 2.3.
Virus serology
Antibodies to different viruses were measured by ELISA methods as described previously [17]. The ELISA kits were purchased commercially: measles-IgG and rubella-IgG kits from Sigma Diagnostics (St. Louis, MO), HHV-6-IgG kit from Advanced Biotechnologies, Inc. (Columbia, MD), and CMV-IgG kit from Whittaker Laboratories (Walkersville, MD). All procedures were performed essentially according to the instructions of the manufacturer of these kits. 2.4.
Data analysis
The statistical significance of the laboratory data was evaluated by the Student's t test using Statview software for the Macintosh computer.
3.
RESULTS
By using an immunoblotting assay, we found positive reactions of autistic sera for autoantibodies to MBP (anti-MBP) and NAFP (anti-NAFP). A representative illustration of these autoantibodies is depicted in Figure 1. The anti-MBP was positive for 18.5-20 kDa band of
450
authentic bovine brain MBP (middle lane) whereas the anti-NAFP showed a positive reaction with high molecular weight NAFP (200 kDa) and sometimes with mid-range molecular weight NAFP (75-90 kDa) (right lane) also. The data analysis revealed that the incidence of anti-MBP was significantly (p = 0.0001) much higher in autistic children as compared to normal children, i.e., up to 84% of autistic children (n = 223), but less than 1% of normal children (n = 60), had anti-MBP. In contrast, the anti-NAFP was found in both autistic and normal children: approximately 50% of autistic children (n = 132) and 27% of normal children (n = 58) had anti-NAFP and the difference between these two groups was only marginally significant (p = 0.05). Because anti-MBP showed a markedly higher incidence in autistic children, this marker was also studied in other control subjects. Figure 2 shows that anti-MBP was present in 65% of autistic children but it was either absent (Down's syndrome children and major depression adults) or present in less than 4% of other controls (normal adults and adults with diabetes and alcoholism). Regarding the myelin markers, in a study of 76 autistic and 20 normal sera (Figure 3), we made three observations: first, approximately 82% of autistic children but none of the normal children had anti-MBP (highly significant); second, none of the autistic or normal children had anti-CNP (no significance at all); and third, a smaller proportion of both autistic children (36%) and normal children (20%) had anti-GC (only marginally significant). Because of a negative result for anti-CNP, the photograph of immunoblots is not included here; however, a major band of CNP (50 kDa) was localized by simultaneously running a positive control of monoclonal CNP antibody (purchased from Sigma, St. Louis, MO). The level of anti-GC in autistic children (34 + 20 Units; n = 76) was higher than the level in normal children (24 + 16 Units; n = 20), but the group difference was only marginally significant (p = 0.043). Thus, the autistic children showed an absolute specificity for autoantibodies to MBP, but not for two other markers (GC or CNP) of the myelin sheath. Furthermore, 56% of autistic children had autoantibodies to both MBP and GC markers, an association that was not found for any other combination of myelin markers (e.g., CNP and MBP or CNP and GC) or for the normal group.
6-
Control Se~
Autisdc Serum
Monoclonal antibody
Figure 1. Photographshowing typical immunoblots of Anti-MBP and Anti-NAFP.
y
451
Figure 2. Distribution of MBP Autoantibodies in different subject populations.
i~!~J~J~!~i~7~i~i~J~i!~y~!~i~!~y2~i~i~i~i~ii~!ii!i!~i~i~!!i~i~ii~ii~!iii~i~!ii~i!i~i~i~i!!T~!ii~!~ii~!i~i~iIT!i~i!i~i~i~ii~i~Tii!~i~!ii!~ii~i~!T!ii~!~i~i~~i!~Ti i~i~~!ii~ii!;i~!~i~i!!~i ~!~!; ~i~i!~i ~i!~!~i ~i;T~~i!i!~~i!~i ~!~!!~i i~!~i~i~i!~ii!i~i~i!!ii~ii!i~~i~ii!ii!i!iii!~i~i~i~!!~ii~ii~ii!ii~i!ii~ii~ii~i!i~ii!~!ii!~ii!ii~iiii~i~!i~i~!i!~ii~ii~i!i!i!~i~i~i~i~i!iii~T!i~ii~
Figure 3. Summary of Myelin Marker Autoantibodies in Autistic Children.
452
Table I
Viral Antibodies in the Sera of Autistic Children.
Virus antibody
Subject group
EIA Units (Mean __SD)
p value
Autistic (n = 42) Normal (n = 26)
3.83 _ 1.23 3.08 __.0.45
0.004
Autistic (n = 45) Normal (n = 37)
2.18 _ 5.35 1.52 __.0.64
0.459
Autistic (n = 31 ) Normal (n = 12)
3.59 __.1.19 2.90 _ 0.81
0.076
Autistic (n = 30) Normal (n = 30)
0.23 __0.32 0.28 ___0.46
0.370
Measles-IgG
HHV-6-IgG
Rubella-IgG
CMV-IgG
The virus serology data are summarized in Table I, which shows antibody levels of four viruses that were measured in the sera of autistic and normal children. As shown, the antibody levels of HHV-6, rubella or C M V did not significantly differ between autistic and normal children. However, the measles antibody level was significantly (p = 0.004) higher in autistic children. Because measles antibody level was significantly higher in autistic children, we sought correlations between virus serology and brain autoantibodies. For the autistic group, we found an interesting association between these two parameters: 90% or greater of measles antibodypositive sera also had positive titers of anti-MBP (Figure 4). This association was not found for any other combination of two parameters, for example, measles antibody and anti-NAFP or HHV-6 antibody and anti-MBP or HHV-6 antibody and anti-NAFP. Considering the existence of autoimmunity to myelin in autistic children, additional evidence of brain myelin pathology was gathered. This effort relied exclusively on unsolicited parental reports. At first, some parents of autistic children, those who already had their brain MRI done, provided copies of brain MRI reports. These reports evidently showed myelin abnormalities such as dysmyelinization, delayed myelination, hypomyelination, immature myelin or patchy diffused white matter myelin (Table II). Secondly, some parents gave their children a myelincontaining nutritional supplement Sphingolin. They observed significant positive effects of Sphingolin and provided a brief account of that. As summarized in Table III, autistic children taking oral Sphingolin showed considerable improvements in speech, language, sociability, sleep, and attention span.
453
Virus Antibodies and Brain Autoantibodies in Autistic Children
m L
> , m
o
Measles!MBP Antibod ies
Measles/NPIP Antibodies
HHV-61MBP Anti bodies
HHV-6tNFP Antibodies
Figure 4. Relationship between Virus Serology and Brain Autoantibodies.
Table II
Patient code
MRI findings of Myelin Abnormality in the Brain of Autistic Children.
Patient age
Patient
Patient
gender
diagnosis
Brain Myelin Abnormality
OB
3 yr
Male
Autism
Corpus callosum myelinated but appears markedly thinned;
BK
3.5 yr
Male
Autism
Myelination pattern appeared abnormal; delayed myelination
KM
3.5 yr
Male
Autism
MRI showed dysmyelination.
ML
3
Male
Autism
Delayed myelinization.
AZ
4 yr
Male
Autism
Delayed myelination.
RT
2.5 yr
Female
Autism
Non-myelinated white matter or white matter was not fully
SD
2.5 yr
Male
Autistic
white matter is less than normal in cerebral hemispheres. when compared to patient's age.
myelinated.
MJ
2.5 yr
SB
4 yr
Female Male
Diffused patchy white matter consistent with slight delay of
regression
myelinization for the age.
Autistic
Delayed myelination within terminal zones of the white
regression
matter.
Autism
MRI showed poor or abnormal myelination.
454
Table III
Behavioral changes after oral administration of Sphingolin in Autistic Children.
Patient code
Patient age
Patient gender
Patient diagnosis
AA
9 yr
Male
Autism
HI
3.5 yr
Male
Autism
SW
3 yr
Male
Autism
DW
2.5 yr
Male
Autism
DG
4.5 yr
Male
Autism
RT
2.5 yr
Female
Autism
MB
3 yr
Male
Autistic
JN
2.5 yr
Male
Autism
AN
2.5 yr
Male
Autism
4.
Sphingolin effects
Significant improvement in language, social interactions and behavior. Significant increase in speech and language; much better eye contact. Initially helped with language; impressive improvements in speech, cognition and sociability; much better sleep pattern. Screaming tantrums down; much better sleep pattern; considerable improvements in language and speech, using many more commands and making phrases and sentences; much more alert. Significantimprovements in language, attention span and social interaction; much calmer and better sleep. Impressive improvement in language, speaking many more commands and making partial to full sentences; better sleep; more eye contact. regression Much improved attention span, eye contact, and language and speech; more social interactions. Milder benefit but more eye contact, speech and communication. Milder effect but more eye contact, speech and communication.
DISCUSSION
T h e field of n e u r o i m m u n o l o g y or n e u r o - i m m u n e biology (NIB) is rapidly expanding. The i m m u n e s y s t e m has a significant impact on the nervous system. Indeed, nearly all nervous s y s t e m diseases or disorders involve i m m u n e p r o b l e m s [1, 2]. Thus the basic n e u r o - i m m u n e interactions and disturbances thereof are e x t r e m e l y important facets of m o d e r n medicine, especially when it concerns brain diseases and mental illnesses. As r e v i e w e d elsewhere [2], they can be treated with i m m u n o m o d u l a t i n g drugs. A l t h o u g h autism is generally considered as a n e u r o d e v e l o p m e n t a l disorder, we studied autism as a n e u r o - i m m u n e dysfunction syndrome ( N I D S ) in which brain-specific a u t o i m m u n i t y played a key role. The possibility that autism is an a u t o i m m u n e disorder has been suggested by e p i d e m i o l o g i c a l as well as i m m u n o l o g i c a l studies. Families with autistic children demonstrate a familial clustering of a u t o i m m u n e diseases [21, 22]. I m m u n e studies of autistic children [4-18] have identified a host of a u t o i m m u n e factors: autism shows an increased frequency of i m m u n e response genes, e.g., H L A , C4B null allele, and e x t e n d e d haplotype; autism displays a family history of a u t o i m m u n e diseases, in particular multiple sclerosis; autism involves a gender factor as it affects males 3 to 5 times
455
more than females; autism shows a microbial association, especially viral infections; autism involves hormonal factors such as beta-endorphins and secretin; autistic patients have immune abnormalities, including organ-specific autoantibodies, that characterize an autoimmune reaction in a disease; and autistic patients show positive responses to immune therapy. Collectively, these findings provide supporting evidence for autoimmune pathology in autism [4-20]. Recent advances have shown that the organ-specific autoimmunity plays a key role in the pathogenesis of autism [4, 6, 10, 11, 15, 18]. Since the brain is the affected organ in autism, the autoimmune reaction will be directed against this organ. In autism, a brain-specific autoimmune response has been determined, the hallmark being the auto-antibodies to brain tissue [4, 15-18]. In the brain, there are two major types of cells: neurons and glia. We examined autoantibodies for each of these two types of cells. We used anti-NAFP as the marker of neurons whereas antiMBP and anti-GFAP were used as the markers of oligodendrocytes and astrocytes, respectively. We found brain-specific autoantibodies to MBP [4, 17], NAFP [5, 15, 17], GFAP [15], and serotonin receptor [16], but the anti-MBP was present in the most significant proportion (58-84%) of children with autism. Our observation of anti-MBP in a significant proportion of autistic children has been independently replicated [14]. Consistent with our previous reports [4, 17], we found that the incidence of anti-MBP was markedly higher when compared to antibodies to other brain antigens; hence, we think that MBP serves as a primary marker of neuroautoimmunity in autism. It should also be noted that anti-MBP was virtually absent in other disease controls. Autoantibodies to other brain antigens therefore are regarded as the secondary markers of autoimmunity in autism. Thus, autism involves a highly specific autoimmune response to myelin sheath in the brain. To establish the specificity of autoantigen involved, we furthermore studied autoantibodies to three major constituents of myelin sheath: MBP, CNP, and GC. As described here, the anti-MBP was present selectively in 82% of autistic children; the anti-CNP was absent, and anti-GC was present non-specifically. Since there was no detectable anti-CNP, this marker was not an autoantigen. GC was most likely a secondary autoantigen because anti-GC was found in low titers, it was also present in normal controls, and it was found in a much smaller number of autistic children. Based on these observations, we infer that MBP is likely a candidate autoantigen in autism. Since myelination is essentially a postnatal event in the developing brain and autism is a neurodevelopmental disorder, autoimmunity to myelin may be causally linked to autism. We hypothesize that the autoimmune response to MBP may impair the development of brain myelin, leading to small anatomical changes or so-called "nicks" in the myelin sheath. Presumably, they differ from degenerative changes that are characteristic of demyelinating diseases such as multiple sclerosis (MS). Some credence for this hypothesis may come from two sets of preliminary observations: first, the MRI findings of myelin abnormalities in autistic children (Table II); and secondly, the therapeutic responses to myelin-containing Sphingolin in autistic children (Table III). In brain, the myelin is elaborated by oligodendrocytes, in which MBP and CNP are the cytoplasmic markers whereas GC is a membrane glycolipid. It was recently shown that oligodendrocytes are equipped with two types of cell-signaling pathways that have opposing effects on the stability of myelin membranes [23]. The first pathway is the MBP/GC pathway that causes destabilization of microtubules whereas the second pathway is the myelin-oligodendrocyte-specific protein (MOSP)/CNP pathway that causes stabilization of microtubules within the myelin membranes. An equilibrium between these two pathways is believed to be essential for the maintenance of normal myelin membrane structure [23]. As described here, autistic children have autoantibodies to MBP and GC, but not to CNP, which we think may potentially influence the MBP/GC pathway of cell-signaling in oligodendrocytes-
456
this could potentially impair the development of myelin in the brain of autistic children especially those prone to autoimmune reaction. Therefore, we speculate that an autoantibodymediated defect of MBP/GC pathway in oligodendrocytes is involved in the pathogenesis of autism. Autoimmunity is commonly triggered by viral infections. Congenital rubella or CMV was co-incidental with autism [19, 20], but there were no laboratory data supporting their link with the disorder. Recently, we performed serological studies of measles virus and HHV-6 [17] and now extended this study to include rubella and CMV as well. As described here, we found that measles virus antibodies, but not the HHV-6, rubella or CMV antibodies, were significantly elevated in children with autism. Furthermore, we found a very important serological association between measles antibody level and anti-MBP, i.e., higher the measles antibody titer the greater the chance of MBP autoantibodies [17]. Because none of the children had a wild-type measles infection but they all had measles-mumps-rubella (MMR) immunization, we suggested that an atypical measles infection is etiologically related to autism. This observation, albeit preliminary, leads us to postulate that a measles-induced autoimmune response to myelin is involved in the pathogenesis of autism. To conclude, our research demonstrate that autism is quite likely a neuro-immune dysfunction syndrome (NIDS) or neuro-immune biology syndrome (NIBS). The neuro-immunopathogenesis of the disorder may involve a measles-induced autoimmune response to brain myelin; however, other viral and immune factors remain to be investigated. Certainly, the genetic factors are important but they will most likely serve as the susceptibility factors rather than causative factors. If and when the specific genes are indeed identified, they are likely to explain autism in only a very small proportion of autistic cases; this is a clear lesson learned from other diseases of the nervous system. Autoimmune mechanism in autism may involve specific autoantibodies and/or it may involve cellular immunity through immunoregulatory T cells. The antigen presenting cells such as macrophages and dendritic cells and B cells, which by themselves can also function as the antigen-presenting cells [24], may also contribute to autoimmune pathology in autism. An autoimmune mechanism is also suggested by the existence of a Th-1 type response in autistic children [ 11 ]. While more remains to be researched, the early clues are that autoimmunity is pivotal to the understanding of pathogenesis of autism.
ACKNOWLEDGMENTS The research was supported, without any conflict of interest, by donations from multiple private sources, including The Yorio Foundation, The Autism Autoimmunity Project, and The BHARE Foundation. The author sincerely thanks students Sheren Lin, David Schubert, Meena Mital and Elizabeth Newell for their assistance in some of the laboratory work. He would also like to thank all those families who participated in this study.
REFERENCES 1. 2. 3.
Singh VK. Neuropeptides as native immune modulators. Progress in Drug Research 1995; 45: 9-31. Singh VK. Immunotherapy for brain diseases and mental illnesses. Progress in Drug Research 1997; 48:129-146. Chambers DA, Schauenstein K. Mindful immunology: neuroimmunomodulation. Immunology
457
Today 2000; 21: 168-170. Singh VK, Warren RP, Odell JD, Cole P, Warren L. Antibodies to myelin basic protein in children with autistic disorder. Brain, Behavior, and Immunity 1993; 7: 97-103. S ingh VK, Fudneberg HH, Emerson D, Coleman M. Immunodiagnosis and immunotherapy in autistic children. Ann NY Acad Sci 549: 602-604. Weizman A, Weizman R, Szekely GA, Wijsenbeek H, Livni E. Abnormal immune response to brain tissue antigen in the syndrome of autis. Am J Psychiatry 1982; 139: 1462-1465. Warren RP, Foster A, Margaretten NC. Reduced natural killer cell activity in autism. J Am Acad Child Adol Psychol 1987; 26: 333-335. Yonk LJ, Warren RP, Burger RA, Cole P, Odell DJ, Warren WL, White E, Singh VK. CD4+ helper T cell depression in autism. Immunolology Letters 1990; 25: 341-346. Plioplys AV, Greaves A, Kazemi K, Silverman E. Lymphocyte function in autism and Rett syndrome. Neuropsychobiology 1994; 29: 12-16. 10. Singh VK, Warren RP, Odell JD, Cole P. Changes of soluble interleukin-1, interleukin-2receptor, T8 antigen, and interleukin-1 in the serum of autistic children. Clin Immunol Immunopathol 1991; 61: 448-455. 11. Singh VK. Plasma increase of interleukin-12 and interferon-gamma: pathological significance in autism. J Neuroimmunol 1996; 66: 143-145. 12. Warren RP, Singh VK, Averett RE, Odell JD, Maciulis A, Burger RA, Daniels WW, Warren WL. Immunogenetic studies in autism and related disorders. Mol Chem Neuropathol 1996; 28: 77-81. 13. Gupta S, Aggarwal S, Heads C. Dysregulated immune system in children with autism: beneficial effects of intravenous immune globulin on autistic characteristics. J Aut Dev Disord 1996; 26:439-452. 14. Fudenberg HH. Dialyzable lymphocyte extract (DlyE) in infantile onset autism: a pilot study. Biotherapy 1996; 9: 143-147. 15. Singh VK, Warren RP, Averett R, Ghaziuddin M. Circulating autoantibodies to neuronal and glial filament proteins in autism. Pediatric Neurology 1997; 16: 88-90. 16. Singh VK, Singh EA, Warren RP. Hyperserotoninemia and serotonin receptor antibodies in children with autism but not mental retardation. Biol Psychiatry 1997; 41: 753-755. 17. Singh VK, Lin SY, Yang VC. Serological association of measles virus and human herpesvirus-6 with brain autoantibodies in autism. Clin Immunol Immunopathol 1998; 89: 105-108. 18. Connelly AM, Chez MG, Pestronk A, Arnold ST, Mehta S, Deuel RK. Serum antibodies to brain in Landau-Kleffner variant, autism, and other neurological disorders. J Pediatrics 1999; 134: 607-613. 19. Chess S, Fernandez P, Korn S. Behavioral consequences of congenital rubella. J Pediatrics 1978; 93: 699-703. 20. Ivarsson SA, Bjerre I, Vegfors P, Ashfors K. Autism as one of several disabilities in two children with congenital cytomegalovirus infection. Neuropediatrics 1989; 21: 102-103. 21. Money J, Bobrow NA, Clarke FC. Autism and autoimmune disease: a family study. J Aut Childh Schizo 1971; 1: 146-160. 22. Comi AM, Zimmerman AW, Frye VH, Law PA, Peeden JN. Familial clustering of autoimmune disorders and evaluation of medical risk factors in autism. J Child Neurol 1999; 14: 388-394. 23. Dyer CA, Philibotte T, Wolf MK, Billings-Gagliardi S. Regulation of cytoskeleton by .
458
24.
myelin components: studies on Shiverer oligodendrocytes carryingMbp transgene. Dev Neuroscience 1997; 19: 395-409. McLachlan SM, Rapoport B. A powerful role for B cells and autoantibodies in disease. Immunology Today 2000; 21" 57.
New Foundationof Biology
459
Edited by Berczi I and R.M. Gorczynski 9 2001 Elsevier Science B.V. All rights reserved
Skin Inflammation and Immunity After Spinal Cord Injury
BRIAN J. MACNEIL and DWIGHT M. NANCE
Department of Pathology, University of Manitoba, 770 Bannatyne Avenue, Winnipeg, Manitoba, Canada R3E-OW3
ABSTRACT Spinal cord injury (SCI) is associated with an elevated risk of infection as well as increased mortality from septicemia. Pressure ulcers of the skin, a frequent complication after SCI, are a major source of these infections. Reports of suppressed immune function in otherwise healthy SCI humans suggests that altered immunity may be part of the sequela of SCI. Therefore, we examined whether impaired inflammation and immune function could be demonstrated in an animal model of complete SCI. In an initial study, rats given a complete SCI at the T1-T2 level were challenged with lipopolysaccharide (LPS) 2 weeks after the SCI. In the spleen, production of tumor necrosis factor-alpha (TNF-c~) mRNA and protein were elevated in SCI rats above that of control animals whereas interleukin-1 beta (IL-I[5) synthesis was not effected. These changes were specific to the spleen, an immune organ, since production of these cytokines in the liver was not effected by SCI. To extend these studies to the site of antigen entry at pressure ulcers, we tested how well the skin of SCI rats could generate an inflammatory response. Injection of turpentine into the skin produced robust inflammation in control animals which was fully established two hours after the injection. In contrast, SCI rats showed little or no sign of inflammation until 12-24 hours after the injection. The cutaneous inflammatory response to turpentine was further characterized by measuring cytokine production two hours after the injection. A significant reduction in the synthesis of TNF-c~, IL-I[5, IL-6, and MCP were seen below the level of the SCI compared to skin from the same site in control rats. These results demonstrate that the changes in skin immune function are limited to skin which is affected by the SCI. We propose that dysregulated neural input to skin following SCI is a primary mediator of the altered inflammatory response in skin and may increase susceptibility to infection.
1.
INTRODUCTION
Spinal cord injury (SCI) is associated with an elevated incidence of infection and a higher rate of mortality due to septicemia relative to control populations [1, 2, 3, 4, 5, 6]. Pressure ulcers account for 20-30% of infection cases seen after chronic SCI [7, 8, 5, 9] and 20% of deaths from septicemia result from infected pressure ulcers [10]. The incidence of pressure ulcers in the SCI population is as high as 60% in countries without specialized SCI centers [9]. However, the incidence of pressure ulcers in community-resident SCI persons remains high (33%) even
460
in North America [11]. Aside from high incidence, pressure ulcers are also a concern due to the potential for extended hospitalizations [6] which are both debilitating to the individual and costly to health care systems. Although several factors result in greater exposure of the SCI population to infectious agents, these may not entirely account for the total incidence of infection. Rather, alterations in immune function may increase the infection rate above that attributed to level of exposure alone and/or more severe or more prolonged disease may occur for a given infectious episode [12, 13, 14]. An increasing body of evidence indicates that immune function is altered by SCI. For example, natural killer cell cytotoxicity, phagocytosis of bacteria, and lymphocyte proliferation are all suppressed in chronic SCI [ 15, 16, 13, 14]. These reports indicate that normal immune function is disrupted after SCI and this may contribute to the high incidence of infection. A potential mechanism of altered immune function after SCI is the loss of regulatory input to immune organs from the central nervous system (CNS). It is known that all immune organs are innervated by the sympathetic nervous system (SNS) [17, 18, 19]. Moreover, functional adrenergic receptors have been identified on immune cells and exposure of immune cells to norepinephrine (NE) modifies numerous parameters of immune function [20]. Specifically, production of the cytokine, tumor necrosis factor- alpha (TNF-ct) is known to be regulated by catecholamines. In vitro studies have demonstrated that while isolated treatment with alpha-adrenergic agonists or beta-adrenergic agonists can enhance or inhibit TNF-ct production, respectively, the net effect of NE treatment is decreased TNF-a synthesis [21, 22, 23]. These findings have been verified in vivo [24]. Thus, an anatomical and biochemical pathway of SNS-immune regulation has been established. SCI, especially when occurring to the upper thoracic and cervical spine will block the neural output of the SNS pathways to peripheral tissues including immune organs. This has been demonstrated in animals in which chemical blockade of residual sympathetic function does not reduce blood pressure in SCI rats, and renal sympathetic nerve activity is minimal when recorded in conscious SCI animals [25, 26]. The loss of sympathetic outflow was observed in animals with lesions as low as spinal segment T4-T5 [25]. Since all immune organs are innervated by the SNS and the input from the SNS can regulate immune responses, it follows that the loss of coordinated sympathetic input to immune organs after SCI should alter immune function. A second neural factor which may influence immune responses, particularly those arising from cutaneous antigen challenge, is the interruption of sensory nerve function. For example, stimulation of sensory nerves results in local release of neurotransmitters such as substance P which cause vasodilation. This process, known as neurogenic inflammation, is mediated by small unmyelinated fibers and is dependent upon mast cell release of histamine which can be modulated by postganglionic sympathetic nerve fibers [27]. Disruption of this process may impair cutaneous inflammatory responses. Substance P and other sensory-related neuropeptides (calcitonin gene-related peptide, CGRP; nerve growth factor, NGF) can also modify specific aspects of immune function such as cytokine and immunoglobulin production [28, 29, 30, 31]. Depletion of substance P fibers with neonatal capsaicin treatment blocked the airway edema and vascular permeability increases induced by several irritants [32] and resulted in more severe infections of Mycoplasma pulmonis [33]. Neonatal capsaicin treatment also led to an 80% decrease in the immune response to subcutaneous antigen which was restored by localized substance P delivery coincident with antigen injection [34]. Langerhans cells resident in skin initiate immune responses by engulfing antigens and then migrating to lymph nodes. CGRPcontaining sensory nerve fibers are closely associated with Langerhans cells in the skin and can modify the function of these cells [35]. Some clinical examples of altered cutaneous responses
461
after nerve injury include an absence of contact dermatitis over the effected skin [36], the selective loss of histamine-induced flare responses below the neurological level of SCI [37] as well as others [38, 39, 40]. The objective of this study was to determine the impact of SCI on cytokine production in an immune organ after systemic challenge with an immune stimulus. In addition, inflammatory responses were also measured in skin above and below the SCI in order to address the potential for altered sensory nerve function to impact on these responses.
2.
METHODS
2.1.
Spinal cord injury (SCI)
Sprague Dawley rats (250-300 g, Charles River, St. Constant, Quebec) were anesthetized (Somnotol, 60 mg/kg, ip) and a SCI induced by clamp compression [41 ]. In short, a laminectomy was performed at the upper thoracic level (T1-T3) and a SCI induced by a 3 minute clamp of the cord at this level. The clamp was calibrated to deliver a closing force of approximately 100 g. The exposed cord was then covered with gel foam, the wound closed, and animals allowed to recover. Sham SCI rats received a laminectomy only. Post-operative care included manual expression of the bladders for 5-7 days (3 x daily) until spontaneous emptying returned. The SCI induced by this protocol resulted in a complete injury as confirmed by histological techniques. The injuries also appeared to be functionally complete since rats forced to swim showed no hindlimb motor function. 2.2
Injections
In some experiments, lipopolysaccharide (LPS, 100 ng) was injected into a lateral tail vein to stimulate splenic cytokine production. Animals were then sacrificed 60 minutes later and blood and spleens collected for analysis. In other experiments, animals were anesthetized with isoflurane and the dorsum of the back shaved at a site below the level of the SCI. At this time animals received intradermal injection of turpentine (50-100 ~tl) and sacrificed two hours later. Skin samples from the injection sites were harvested and rapidly frozen for future analysis of cytokine production. 2.3.
Blood and tissue sampling
At the appropriate time points, animals were deeply anesthetized with Somnotol and skin and spleen samples harvested and immediately frozen in liquid nitrogen. Blood from the right atrium was collected into EDTA treated tubes and centrifuged. The plasma was collected and frozen for future assay. 2.4.
Quantification of cytokine mRNA
The extent of cytokine mRNA induction was determined following our published procedure [42]. Spleen or skin samples were homogenized in Trizol reagent (Life Technologies, Burlington, ON), the RNA extracted and electrophoreised on a denaturing agarose gel, blotted onto a nylon membrane, and hybridized with digoxigenin-labeled riboprobes for TNF-c~, IL-I[5, IL-6 and MCP-1 mRNA. Bound probe was detected with a sensitive chemiluminescence system. Membranes
462
were reprobed with radiolabelled ribosomal RNA probes as loading controls. Image analysis was used to obtain the band densities for cytokines and loading controls. Cytokine mRNA was expressed as a ratio of cytokine to loading control optical densities. 2.5.
Cytokine ELISA
Commercial ELISA kits (Biosource, Camarillo, CA) were used to detect splenic levels of TNF-c~ and IL-1 [3 following the extraction of protein from homogenized spleen samples. 2.6.
Radioimmunoassay (RIA) for plasma corticosterone
A radioimmunoassay was used to determine circulating levels of corticosterone. Rat tritiated corticosterone and anti-rat corticosterone antibody were obtained from ICN (Costa Mesa, CA) and optimized as per manufacturer's instructions for the corticosterone RIA. Plasma samples were diluted 1:500 and heat denatured before incubating overnight (4~ with radiolabelled tracer and antibody. The following day unbound tracer was precipitated by adding a charcoal/dextran mixture followed by centrifugation and removal of the supernatant. The radioactivity of the supernatant was determined and the data entered in a curve-fitting software package (Assay-Zap, Biosoft, Cambridge, UK) to generate a standard curve and calculate sample corticosterone values. 2.7.
HPLC
Catecholamines were extracted directly from plasma and from spleen homogenates and assayed for epinephrine and norepinephrine using standard techniques (alumina extraction, electrochemical detection) as described earlier [43]. 2.8.
Analysis
In experiments utilizing systemic LPS injections, the data were analyzed by one-way ANOVA to assess the between group main effect of sham vs. SCI. In the other experiments using intradermal turpentine injections, a two-way ANOVA assessed main effects of the between groups factor (sham vs. SCI) and the within groups factor of injection site (above vs. below) using Statistica (StatSoft, Inc., Tulsa, OK). All pairwise comparisons were accomplished using an orthogonal contrast procedure. Data are expressed as mean + standard error of the mean.
3.
RESULTS
Initial studies were undertaken to determine whether the SCI procedure was associated with an elevation of plasma corticosterone which would be a possible mechanism of suppressed immune function. Sham operated or SCI rats were sacrificed 1-4 weeks after surgery and compared to home cage controls (Figure 1). At no time was corticosterone elevated by SCI or sham surgery indicating that corticosterone, likely elevated acutely after surgery, returned to basal levels within 1 week of the surgery, even within the SCI group. To assess splenic cytokine production in response to a circulating immune stimulus, sham and SCI animals were injected intravenously with 100 ng LPS and sacrificed 60 minutes later. Relative to sham animals, spleens obtained from SCI rats showed an increased production of
463
400 1
J Sham SCI Home Cage Control
300 c-co x _
o 09
"~ 2 0 0 o
o
E 13_
100 -
.........
..
J
-
1
.......
........................
]
I
I
2
3
I .....
4
Weeks Post-Surgery
Figure 1. Plasma corticosterone levels (means + standard error of means) in sham and SCI rats at various times after surgery. Values (mean + standard error of mean) for unoperated home cage control animals are shown as horizontal lines.
TNF-(~ m R N A (Figure 2). In contrast, cytokine production in the liver from the same animals was not effected by SCI (Figure 2). SCI did not effect IL-I[3 m R N A production in the spleen or liver compared to sham controls. To characterize further the changes in cytokine m R N A production, the relationship between TNF-o~ and IL-I[3 was examined. Production of IL-I[3 was highly correlated to TNF-ot in the spleen of sham animals but not in SCI animals (sham: r = 0.703, p = 0.023; SCI: r = 0.203, p = 0.664). In contrast, correlations of liver cytokine production were not different between the groups, although the relationships were not as strong overall (p ~ 0.09). Measurement of cytokine protein levels in the spleen revealed the same pattern as that for mRNA; enhanced TNF-~x in SCI animals with no effect on IL-I[3 (Figure 3). Thus, SCI had a very selective effect in that only splenic, and not hepatic, TNF-c~ production was altered. Further, the changes in TNF-ot synthesis in SCI rats had no carryover effect on IL-I[3 production which was highly correlated to the magnitude of the TNF-c~ response on the spleens of sham animals. These data also suggest that the SNS may not be a primary regulator of in vivo IL-1 [3 production. Assessment of neuroendocrine factors in these animals revealed that LPS induced a similar elevation of plasma corticosterone between sham and SCI groups (661 _+ 45 vs. 886 _+ 145 ng/ml, respectively; p = 0.123). As expected, plasma levels of catecholamines were drastically reduced in SCI animals confirming that outflow of the SNS was disrupted. However, since SCI does not destroy peripheral nerves, splenic norepinephrine content was not different between sham and SCI groups (Figure 4).
464
A
Sham SCI
r~~ 1.5 l O .,-, c-
1.0~
O
o C~ c-
~5 o 0.5 '~i
J
s
o 0.0
B
i
TNF-alpha
IL- 1 beta
O
1.0a
0 s
0.8-
Sham SCI
c
o 0.6 o r cO O .J
0.4 0.2
O
0~', 0.0
UZ0. . . . . . . . . . . . .
TNF-alpha
IL- 1 beta
Figure 2. Ratios (mean + standard error of mean) of cytokine mRNA to loading control mRNA in spleen (A) and liver (B) samples of sham and SCI rats one hour after iv injection of 100 ng LPS. (* significantly different from sham, p < 0.05).
The next series of experiments were designed to assess whether inflammatory responses in the skin were also effected by SCI. In a pilot study, 4 sham and 2 SCI animals received intradermal injections of 100 Ftl of turpentine and the subsequent inflammatory response was observed. In these studies, SCI animals were maintained for 4-5 weeks after surgery to ensure a chronic stage of SCI was present. Two hours after intradermal injection of turpentine, sham rats developed a necrotic lesion surrounded by edema (Figure 5). In contrast, injection of turpentine below the level of injury in SCI rats failed to induce this reaction. Although edema was present, the central lesion observed in sham rats did not appear until much later in the SCI rats (Figure 5). Based on this initial observation in the skin, an experiment was completed which examined cytokine responses after intradermal injection of turpentine in sham and SCI rats. Importantly, to determine if any changes in cytokine production were a result of local versus systemic factors, animals received injections of turpentine both above and below the injury. Intradermal injection of turpentine induced IL-113, IL-6 and monocyte chemotactic protein (MCP) mRNA expression above and below the level of injury or sham surgery and, thus permitted the use of within group
465
.......
Sham F'////-/~ SCl "E" 6 .I eO
T
0
~:~ 5 Z~
v
4> O~ ...J
c
3-
0 0 2._o
i
'i
C
co
1-I
0
TNF-alpha
IL-1 beta
Figure 3. Cytokine protein levels (mean + standard error of mean) in spleen samples from sham and SCI rats one hour after iv injection of 100 ng LPS. (-k significantly different from sham, p < 0.05).
20 -
~
Sham SCI
- 2.0
18 -
-
1.8
16-
-
1,6
-
1,4
Tt
!412-
E --
- 1 , 2 E
-
10I
8-
I
~ 4~-
-
0.8
-
0.6
-
0.4
i 0,2
0
0,0
NorEPl Plasma
EPI
NorEPl Spleen
Figure 4. Catecholamine levels (mean _+ standard error of mean) in plasma and spleen samples from sham and SCI rats one hour after iv injection of 100 ng LPS. ( ~ significantly different from sham, p < 0.05).
466
SHA:M 2 hr
12 hr
24 hr
Figure 5. Skin response to intradermal injection of 100 ~tl turpentine in sham and SCI rats at 2, 12, and 24 hours after injection. All injections were placed below the level of the SCI or laminectomy.
(repeated measures) comparisons to analyse this data. In the sham group, IL-113, IL-6 and MCP mRNA synthesis was not affected by the sampling site (i.e., above or below injury) (Figure 6). However, production of these same three cytokines was suppressed below the injury level in the SCI group. An unexpected finding was the apparent lack of expression of TNF-c~ mRNA in skin samples taken from above the injury level. This may reflect an anatomical phenomenon since this cytokine was also absent in sham samples from the same location. In contrast, TNF-c~ was readily detected below the injury/surgery level in both groups. Between group comparisons revealed a dramatic reduction in TNF-c~ mRNA production from skin below the injury level
467
in the SCI group compared to sham controls (Figure 6). In addition to TNF-c~, the magnitude of the difference between SCI and sham animals regarding MCP mRNA synthesis below the injury was such that a significant between groups difference was also present. No between group effects were present for data from above the injury level although a trend toward higher IL-I[5 production was present in the SCI group.
....
~6
Above
~'
T
Iii........_21 Sham SCl
~5 ~
o
l
4-
~ 0
3~ 2 detection _E~_~~//,
~~~__
O
TNF
IL-1
IL-6
MCP
Cytokine mRNA
4l
Below
3 o
E~
T
Sham
mII ......
~m
n1:3
O
r
1 -
0-
TNF
IL-1
IL-6
MCP
Cytokine mRNA Figure 6. Ratios (mean + standard error of mean) of cytokine mRNA to loading control mRNA in skin samples from above and below the injury level in sham and SCI rats two hours after intradermal injection of 50 ~tl turpentine. (# significantly different from sham at same injection site, p < 0.05; ~r significantly different from above injury sample within same treatment group, p < 0.05).
468
4.
DISCUSSION
This study assessed the impact of SCI on inflammatory and immune responses. Specifically, responses were measured following immune challenge through two separate routes of injection (i.e., intravenous and intradermal). Regarding the first route of administration, the data show that after intravenous injection of a low dose of LPS, the production in the spleen of TNF-~t mRNA and protein is elevated in SCI animals. This outcome was isolated to the spleen because TNF-ot synthesis in the liver was not effected by SCI. In contrast, the inflammatory response observed after intradermal injection of turpentine below the SCI level was reduced in SCI animals compared to sham controls. This effect was further characterized in a follow up experiment by measuring inflammatory cytokine production following the same challenge. However, the design was expanded to included test injections above and below the SCI. Cytokine mRNA synthesis was not different in sham animals between the two skin sites. In contrast, SCI animals demonstrated a clear reduction of cytokine mRNA production below the SCI regarding the cytokines IL-I[3, IL-6 and MCP. Although not detectible in any skin samples above the injury level, a dramatic reduction in TNF-ct mRNA was seen in the SCI animals below the level of the injury relative to sham controls. These data reveal a distinct effect of SCI on inflammatory and immune responses which is dependent upon the route of stimulus administration. It is proposed that these region-specific effects arise from the different patterns of innervation present in these tissues and the effect of disruption of this innervation after SCI. In the rat, the spleen is exclusively innervated by postganglionic sympathetic fibers [17, 18]. This provides an opportunity to assess the impact of altered sympathetic innervation on immune function without the potential confound of other types of innervation (e.g., sensory, vagal). Using this model, a clear effect of SCI was seen on splenic cytokine production following intravenous injection of LPS. That is, TNF-et mRNA and protein synthesis were elevated in SCI animals relative to sham controls. These data are consistent with the reported actions of NE on TNF-c~ production. Specifically, administration of NE or beta-adrenergic agonists leads to a suppression of TNF-c~ synthesis following LPS stimulation [44, 45]. Therefore, interruption of endogenous release of NE as a result of SCI would be expected to facilitate TNF-~t production. This interpretation is supported by the catecholamine data which shows that splenic tissue levels of NE were high while plasma levels of NE were very low indicating that NE-containing nerve fibers are present in the spleen but no significant amount of NE is being released. Data from liver tissue samples also support a direct sympathetic mechanism regarding elevated splenic TNF-c~ production. That hepatic cytokine production was not altered in SCI rats indicates that an organ specific mechanism is present rather than systemic factors such as circulating hormones, nutritional status, and physical or psychological stress. The exclusive sympathetic innervation of the spleen would suggest the SNS as the primary mediator of the effect of SCI on the TNF-c~ response. Given the neurological level of the SCI used in the present study (T1-T2), one would predict that TNF-c~ production in other immune organs would also be enhanced because sympathetic innervation of these organs would also be disrupted. In fact, SCI in the rat as low as T4-T5 should have the same effect since there is no evidence of meaningful sympathetic activity remaining after a SCI at the T4-T5 level [25]. While this initial study indicated that immune responses arising from systemically administered challenges were altered following SCI, it was also of interest to determine if cutaneous responses would also be effected by SCI since aspects of the sensory nervous system are known to modify inflammation and immune responses. For this, an inflammatory agent, turpentine, was injected intradermally and the resulting skin response observed. This simple test revealed
469
a dramatic effect of SCI on skin responses in that the turpentine-induced skin lesion was markedly blunted in the SCI group. To begin characterization of this effect, a larger study was designed which included skin injections above and below the SCI in the same animals. This design would determine whether local or systemic factors were responsible for the blunted turpentine responses seen in the SCI rats. The data from this study revealed two distinct findings. First, the production of proinflammatory cytokines (TNF-~t, IL-I[3, IL-6, MCP) was clearly reduced in SCI animals. Second, this deficit was restricted to skin below the neurological level of the SCI. Therefore, systemic factors were not responsible for the reduction in cutaneous cytokine production. Given the distinct patterns of cytokine production above and below the SCI, the most likely mechanism would be one of neurological origin. An unexpected outcome of the present study was the finding that TNF-c~ mRNA was not detectable in skin samples taken from the dorsal aspect of the upper trunk and neck. While the precise reason for this is unknown, the finding was consistent between control and SCI animals. In contrast, TNF-~t mRNA was readily detectable in skin from the lower aspect of the dorsal trunk in all animals. However, in these samples, the SCI group had a dramatically lower response compared to the controls, as was the case for all other cytokines measured. The inability to demonstrate the presence of a systemic factor which could account for the effects of SCI on cytokine responses addresses two important issues. First, it indicates that future studies aimed at determining the mechanisms by which SCI alters cytokine production must focus on local factors. Second, and equally important, it demonstrates that the present model of SCI is not undermined by detrimental effects of the SCI procedure on the animals' general health. Specifically, the SCI animals are not in a generalized state of stress since basal plasma corticosterone levels are not elevated in SCI rats. Nor are these animals in a state of generalized debilitation resulting in a global suppression of immune function. Although there is an initial loss of body weight as a result of disuse atrophy of the effected muscles, these animals show a normal food intake and gain weight similar to the control rats. Therefore, a well documented mechanism of immune suppression (stress-mediated activation of the hypothalamicpituitary adrenal axis) can be eliminated as a possible mediator of the effects of SCI on cytokine production. While the most likely explanation of how SCI alters cytokine production may involve a neural mechanism, the exact process is not identified in the present experiments and several possibilities can be considered. Neurotransmitters released from nerve terminals within immune organs can act directly on immune cells via specific receptors. In particular, functional adrenergic receptors are present on immune cells [20] and binding of NE to these receptors is known to modify TNF-c~ production [21, 22, 23, 24]. Such a direct action of NE on immune cells was originally proposed by Besedovsky [46, 47] in that the sympathetic innervation of immune organs provides a tonic regulation of immune responses and the magnitude of this input can be increased or decreased to suppress or enhance, respectively, immune function. The SNS formed part of a bidirectional feedback system which permitted the integration of several factors regarding the status of the host with the magnitude of the immune response to a pathogen. In the case of LPS and related cytokines, the SNS is rapidly activated as indicated by increased electrical activity of the splenic nerve [48] and greater turnover of NE in the spleen [49]. The additional release of NE would provide one mechanism for regulating the magnitude of the cytokine response to prevent harm to the host (septic shock) while allowing adequate expression of immune responses. Because SCI will interrupt the flow of sympathetic activity to the periphery, this branch of neuroimmune regulation will be lost and, as a result, the response to LPS may be exaggerated. This process may explain the increased production of TNF-c~ in the SCI animals since TNF-~ is known to be strongly regulated by NE via adrenergic receptors.
470
A second potential mechanism of altered TNF-ot production in the spleens of SCI rats is a change in cellular trafficking through the spleen. NE and neuropeptide Y, both released from sympathetic nerve terminals, can alter the movement of leukocytes by their actions on chemotaxis and cellular adhesion [50, 51, 52, 53, 54]. The precise manner by which altered migration takes place is not certain. However a direct effect on cellular adhesion molecules is not supported by previous studies [50, 55]. Cellular trafficking is also dependent on the rate of blood flow through the organ; vasoconstriction results in reduced cellular efflux from the spleen and vice versa [56]. Importantly, this study confirmed that, in addition to blood flow, the effects of NE on cell migration included mechanisms which were independent of hemodynamic factors. Thus, a complex interaction takes place in immune organs whereby cellular influx and efflux are determined by several factors including blood flow, adhesion to endothelial cells, and movement in and out of immune cell compartments. It is possible that a greater number of cells which produce TNF-c~ in response to LPS accumulated in the spleens of SCI rats. In skin, the above factors would still be relevant in addition to those contributed by sensory nerves. Again, the net outcome on skin inflammation may be a combination of effects on immune cell function, cell migration, and changes in blood flow. Regardless of the exact mechanism, it remains most likely that the altered neural environment is the underlying reason for the observed changes. This is based primarily on the distinct anatomical pattern of the changes in splenic and cutaneous cytokine production (i.e., an absence of systemic changes in cytokine production in SCI rats). Presumably, the difference in the net outcome of disruption in sympathetic (spleen) versus sympathetic and sensory (skin) nerves accounts for the discrete patterns of cytokine production in the spleen and skin. In an attempt to integrate these data into a meaningful outcome on host resistance in SCI subjects, one must consider the most relevant route of pathogen challenge. In the clinical setting, a primary site from which infections arise is pressure ulcers. Once the physical barrier function of the skin is lost, these ulcers are highly susceptible to infection which can lead to septicemia. Using this working model, we would predict that a pathogen would have an increased chance of establishing a colony in the skin below a SCI since the inflammatory response is blunted in this region. The initial inflammatory response is a critical factor in recruiting and activating immunocompetent cells. The alteration of this process after SCI may permit the early establishment of microbial pathogens prior to the recruitment and infiltration of lymphocytes. In clinical terms, it is possible that attenuated inflammation may increase the likelihood of infectious pathogens establishing themselves at sites of pressure ulcers. Thus, SCI may result in a greater frequency of infections above and beyond that expected even after consideration of the unchecked physical stress placed on the skin as a result of paralysis and loss of sensation. In addition, infections which become more severe and result in blood-borne organisms, may have a greater likelihood of producing septic shock due to the enhanced production in immune organs of TNF-c~, a primary hypotensive factor mediating endotoxic shock. Therefore, at two separate stages, the loss of neural regulation of immune function after SCI may place the individual at greater risk of infection and septic shock, both of which have a highly elevated incidence in persons with SCI [1, 2, 3, 4, 5, 6]. In summary, these initial studies in an animal model of SCI demonstrate that in vivo cytokine responses to inflammatory and immune stimuli are altered after SCI. Further, the anatomical distribution of these changes support the conclusion that the underlying mechanism behind the altered cytokine responses is neural in origin. The net outcome of these changes may be a heightened susceptibility to the establishment of infections at pressure ulcers as well as a greater chance of septic shock in the case of more severe infections. Understanding the precise nature of the neural impact on immune function following SCI may lead to improved clinical
471
management and a better understanding of the role of the nervous system in regulating these responses in uninjured individuals.
ACKNOWLEDGEMENTS Supported by Manitoba Medical Services Fund and Paralyzed Veterans of America (2082).
REFERENCES 1. 2. 3.
4.
5. 6.
7. 8. 9. 10. 11.
12. 13. 14. 15.
16. 17.
DeVivo MJ, Kartus PL, Stover SL, Rutt RD, Fine PR. Cause of death for patients with spinal cord injuries. Arch Intern Med 1989; 149:1761-1766. Jackson AB, Groomes TE. Incidence of respiratory complications following spinal cord injury. Arch Phys Med Rehabil 1994; 75: 270-275. Hartkopp A, Bronnum Hansen H, Seidenschnur AM, Biering Sorensen F. Survival and cause of death after traumatic spinal cord injury. A long-term epidemiological study. Spinal Cord 1997; 35: 76-85. Hartkopp PA, Bronnum-Hansen H, Seidenschnur AM, Biering-Sorensen F. Survival and cause of death after traumatic spinal cord injury. A long-term epidemiological study. Ugeskr Laeger 1998; 160: 6207-6210. Rish BL, Dilustro JF, Salazar JM, Schwab KA, Brown HR. Spinal cord injury: a 25-year morbidity and mortality study. Mil Med 1997; 162:141-148. Vaidyanathan S, Soni BM, Gopalan L, Sett P, Watt JW, Singh G, Bingley J, Mansour P, Krishnam KR, Oo T. A review of the readmissions of patients with tetraplegia to the Regional Spinal Injuries Centre, Southport, United Kingdom, between January 1994 and December 1995. Spinal Cord 1998; 36: 838-846. Montgomerie JZ, Chan E, Gilmore DS, Canawati HN, Sapico FL. Low mortality among patients with spinal cord injury and bacteriemia. Rev Infect Dis 1991; 13: 867-871. Yarkony GM. Pressure ulcers: a review. Arch Phys Med Rehabil 1994; 75: 908-917. Beraldo PSS, Neves ECG, Alves CMF, Khan P, Cirilo ACM, Alencar MRB. Pyrexia in hospitalised spinal cord injury patients. Paraplegia 1993; 31:186-191. DeVivo MJ, Black KJ, Stover SL. Causes of death during the first 12 years after spinal cord injury. Arch Phys Med Rehabil 1993; 74: 248-254. Fuhrer MJ, Garber SL, Rintala DH, Clearman R, Hart KA. Pressure ulcers in communityresident persons with spinal cord injury: Prevalence and risk factors. Arch Phys Med Rehabil 1993; 74:1172-1177. Lyons M. Immune function in spinal cord injured males. J Neurosci Nurs 1987; 19:18-23. Nash MS. Immune responses to nervous system decentralization and exercise in quadriplegia. Med Sci Sports Exerc 1994; 26: 164-171. Campagnolo DI, Keller SE, DeLisa JA, Glick TJ, Sipski ML, Schleifer SJ. Alteration of immune system function in tetraplegics. Am J Phys Med Rehabil 1994; 73: 387-393. Cruse JM, Lewis RE, Bishop GR, Kliesch WF, Gaitan E, Britt R. Decreased immune reactivity and neuroendocrine alterations related to chronic stress in spinal cord injury and stroke patients. Pathobiol 1993; 61:183-192. Campagnolo DI, BartLett. JA, Keller SE, Sanchez W, Oza R. Impaired phagocytosis of staphylococcus aureus in complete tetraplegics. Am J Phys Med Rehabil 1997; 76: 276-280. Felten DL, Felten SY, Bellinger DL, Carlson SL, Ackerman KD, Madden KS, Olschowki
472
18. 19. 20. 21.
22. 23. 24.
25. 26. 27. 28.
29. 30. 31.
32. 33.
34.
35.
36. 37. 38.
JA, Livnat S. Noradrenergic sympathetic neural interactions with the immune system: structure and function. Immunol Rev 1987; 100: 225-260. Nance DM, Burns J. Innervation of the spleen: Evidence for an absence of afferent innervation. Brain Behav Immun 1989; 3: 281-290. Nance DM, Hopkins DA, B ieger D. Re-investigation of the innervation of the thymus gland in mice and rats. Brain Behav Immun 1987; 1: 134-147. Plaut M Lymphocyte hormone receptors. Ann Rev Immunol 1987; 5: 621-669. Spengler RN, Alen RM, Remick DG, Strieter RM, Kunkel SL. Stimulation of ot-adrenergic receptor augments the production of macrophage-derived tumor necrosis factor. J Immunol 1990; 145: 1430-1434. Severn A, Rapson NT, Hunter CA, Liew FY. Regulation of tumor necrosis factor production by adrenaline and beta-adrenergic agonists. J Immunol 1992; 148: 3441-3445. Hu X, Goldmuntz EA, Brosnan CF. The effect of norepinephrine on endotoxin-mediated macrophage activation. J Neuroimmunol 1991; 31: 35-42. Elenkov IJ, Hask6 G, Kov~ics KJ, Vizi ES. Modulation of lipopolysaccharide-induced tumor necrosis factor-or production by selective or- and [3-adrenergic drugs in mice. J Neuroimmunol 1995; 61: 123-131. Maiorov DN, Weaver LC, Krassioukov AV. Relationship between sympathetic activity and arterial pressure in conscious spinal rats. Am J Physiol 1997; 272: H625-H631. Osborn JW, Taylor RF, Schramm LP. Determinants of arterial pressure after chronic spinal transection in rats. Am J Physiol 1989; 256: R666-R673. Basbaum AI and Levine JD. The contribution of the nervous system to inflammation and inflammatory disease. Can J Physiol Pharmacol 1991; 69: 647-651. Torcia M, Bracci-Laudiero L, Lucibello M, Nencioni L, Labardi D, Rubartelli A, Cozzolino F, Aloe L, Garaci E. Nerve growth factor is an autocrine survival factor for memory B lymphocytes. Cell 1996; 85: 345-356. Lotz M, Vaughan JH, Carson DA. Effect of neuropeptides on production of inflammatory cytokines by human monocytes. Science 1988; 241 : 1218-1221. Moore TC. Modification of lymphocyte traffic by vasoactive neurotransmitter substances. Immunol 1984; 52:511-518. Stanisz AM, Befus D, Bienenstock J. Differential effects of vasoactive intestinal peptide, substance P, and somatostatin on immunoglobulin synthesis and proliferations by lymphocytes from Peyer's patches, mesenteric lymph nodes, and spleen. J Immunol 1986; 136: 152-156. Lundberg JM and Saria A. Capsaicin-induced desensitization of airway mucosa to cigarette smoke, mechanical and chemical irritants. Nature 1983; 302:251-253. Bowden JJ, Baluk P, Lefevre PM, Schoeb TR, Lindsay JR, McDonald DM. Sensory denervation by neonatal capsaicin treatment exacerbates Mycoplasma pulmonis infection in rat airways. Am J Physiol 1996; 270: L393-L403. Helme RD, Eglezos A, Dandie GW, Andrews PV, Boyd RL. The effect of substance P on the regional lymph node antibody response to antigenic stimulation in capsaicin-pretreated rats. J Immunol 1987; 139: 3470-3473. Hosoi J, Murphy GF, Egan CL, Lerner EA, Grabbe S, Asahina A, Granstein RD. Regulation of Langerhans cell function by nerves containing calcitonin gene-related peptide. Nature 1993; 363: 159-163. Rebhun J. Effect of innervation on contact dermatitis. Ann Aller 1980; 45:33 Rebhun J, Botvin J. Histamine flare, a neurovascular response. Ann Aller 1980; 45: 59-62. LeVasseur SA, Gibson SJ, Helme RD. The measurement of capsaicin-sensitive sensory
473
39.
40. 41. 42.
43.
44. 45.
46. 47. 48.
49. 50. 51.
52. 53.
54. 55.
56.
nerve fiber function in elderly patients with pain. Pain 1990; 41: 19-25. Helme RD, LeVasseur SA, Gibson SJ. RSI revisisted: evidence for psychological and physiological differences from an age, sex and occupation matched control group. Aust NZ J Med 1992; 22: 23-29. Basile S, Khalil Z, Helme RD. Skin vascular reactivity to the neuropeptide substance P in rats with peripheral neuropathy. Pain 1993; 52:217-222. Rivlin AS and Tator CH. Effect of duration of acute spinal cord injury compression in a new acute cord injury model in the rat. Surg Neurol. 1978; 10: 39-43. Meltzer JC, Sanders V, Grimm PC, Chiasson N, Hoeltke H-J, Garrett KL, Greenberg AH, Nance DM. Non-radioactive northern blotting with biotinylated and digoxingenin-labelled RNA probes. Electrophoresis 1998; 19: 1351-1355. Brown R, Li Z, Vriend CY, Nirula R, Janz L, Falk J, Nance DM, Dyck DG, Greenberg AH. Suppression of splenic macrophage interleukin-1 secretion following intracerebroventricular injection of interleukin-l[3: Evidence for pituitary-adrenal and sympathetic control. Cell Immunol 1991; 132: 84-93. Wu CC, Liao MH, Chen SJ, Chou TC, Chen A, Yen MH. Terbutaline prevents circulatory failure and mitigates mortality in rodents with endotoxemia. Shock 2000; 14: 60-67. Szabo C, Hask6 G, Zingarelli B, Nemeth ZH, Salzman AL, Kvetan V, Pastores SM, Vizi ES. Isoproteronol regulates tumor necrosis factor, interleukin-10, interleukin-6 and nitric oxide production and protects against the development of vascular hyporeactivity in endotoxemia. Immunol 1997; 90: 95-100. Besedovsky HO, Del Rey AE, Sorkin E, Da Prada M, Keller HH. Immunoregulation mediated by the sympathetic nervous system. Cell Immunol 1979; 48: 346-355. Zhou ZZ, Wurster RD, Qi M, Jones SB. Sympathoadrenal activation in sinoaorticdenervated rats following endotoxin. Am J Physiol 1991; 260: R739-R746. MacNeil BJ, Jansen AH, Greenberg AH, Nance DM. Activation and selectivity of splenic sympathetic nerve electrical activity response to bacterial endotoxin. Am J Physiol 1996; 270: R264-R270. Vriend CY, Zuo L, Dyck DG, Nance DM, Greenberg AH. Central administration of interleukin113 increases norepinephrine turnover in the spleen. Brain Res Bull 1993; 31: 39-42. Sung CP, Arleth AJ, Feuerstein GZ. Neuropeptide Y upregulates the adhesiveness of human endothelial cells for leukocytes. Circ Res 1991; 68: 314-318. Madden KS, Felten SY, Felten DL, Hardy CA, Livnat S. Sympathetic nervous system modulation of the immune system. II. Induction of lymphocyte proliferation and migration in vivo by chemical sympathectomy. J Neuroimmunol 1994; 49: 67-75. Carlson SL, Fox S, Abell KM. Catecholamine modulation of lymphocyte homing to lymphoid tissues. Brain Behav Immun 1997; 11: 307-320. Straub RH, Mayer M, Kreutz M, Leeb S, Scholmerich J, Falk W. Neurotransmitters of the sympathetic nerve terminal are powerful chemoattractants for monocytes. J Leukoc Biol 2000; 67: 553-558. Ortega E, Garcia JJ, De La Fuente M. Modulation of adherence and chemotaxis of macrophages by norepinephrine. Mol Cell Biochem 2000; 203:113-117. Carlson SL, Beiting DJ, Kiani CA, Abell KM, McGillis JP. Catecholamines decrease lymphocyte adhesion to cytokine-activated endothelial cells. Brain Behav Immun 1996; 10: 55-67. Rogausch H, Del Rey AE, Oertel J, Besedovsky HO. Norepinephrine stimulates lymphoid cell mobilization from the perfused rat spleen via beta-adrenergic receptors. Am J Physiol 1999; 276: R724-R730.
This Page Intentionally Left Blank
475
Index
A AABS (Autoimmunity Associated Behavioural Syndrome), 383-384 acetylcholine, 438 acromegaly, 4 ACTH (Adrenocorticotropic hormone), 117, 123, 141-143, 149, 151, 177, 179, 180, 198, 208, 216, 300, 358-330 action on adrenal cortex, 104 activation elicited by LPS (lipopolysaccharide), 148 anti-pyretic effect, 13 attenuation, 144 cells, 106-107 derived peptides, 104 immunosuppression and, 10, 12, 13, 24, 60, 99, 100, 102, 104, 128 NGF release and, 216 production, 24, 99, 104 receptor, 104 release, 12, 99, 104, 106, 123, 143, 144, 149, 179, 180, 198, 208, 300 sleep and, 255,260 synthesis, 106 treatment, 14 ACTH response to CRF (corticotropin releasing factor), 179 enterotoxins, 358-360 host stress, 99, 117, 142, 148, 198 IL-l[3, 177, 179 systemic LPS (lipopolysaccharide), 179 ACTH secretion, 24, 100, 104, 106, 117, 128, 141-144, 149, 151,260 pituitary ACTH secretion, 141-144, 149, 151 activins, 107 activity 5-HT activity, 145, 148, 151 activity-dependent cellular markers, 125, 127 amygdaloid activity, 144 antibodies, anti-carbohydrate activity, 342
antipyretic activity, 301 astrocytes, eNOS (nitric oxide synthase) activity, 177 [3-integrins, 310 brain activity during sleep, 257-258, 261 cardiovascular activity, 180 CD 14 receptor, opsonic activity, 164, 166 CD45, protein phosphotyrosine phosphatase activity, 337 Con A, proliferative activity, 229 CRH (corticotropin releasing hormone) activity and appetitive behavior, 359 CRH gene, transcriptional activity, 360 CTLs (cytokine T lymphocytes), cytolytic activity, 412,413 cytokines, 66, 164, 166 dendritic cell, 12 disease activity, 409-402 DNA binding factor, 179 F-KB activity, 164, 167-170, 179 FEG, functional activity, 295 histamine producing cell-stimulating activity, 54 HPA(hypothalamus-pituitary-adrenal)-axis activity, 141, 144, 176, 177, 179, 180 IgG-receptors, 310 IL activity during sleep, 261,262 IL-1 receptor antagonist, 366, 369 immune activity, 130, 133 immune cells, 271 iNOS (nitric oxide synthase) activity in Sertoli cells, 273 leucocytes, regulation of, 15, 24, 308, 384 Leydig cells, steroidogenic activity, 272 L-histidine decarboxylase activity in bone marrow, 47-48, 51, 52, 54 locomotor activity, 151 lymphocyte activity during sleep, 261 lymphocyte, 73 lymphokine activated killer cell activity, 409-410, 413,415 macrophage activity, 11,271,273,296, 383
476
metabolic activity, 24 microglial cells, 171,182 NADPHd activity, 128
allo-MHC (major histocompatibility complex), 314, 319 allostatic load, 116, 143, 157 alpha-melanocyte stimulating hormone (c~-MSH), 13,
natural killer cell activity, 11, 14, 130, 260-263, 354, 375,408-410
249, 284, 334, 345 altered chemokine, 241-242
neuroendocrinal activity, 145, 148, 152, 356 neuronal activity, 143, 177 NOS (nitric oxide synthase) isoforms in the brain, 176-180, 273 phagocytic activity and sleep deprivation, 262 phosphodiesterase activity, 178 protein kinase C, 336
Alzheimer's disease, 88, 93, 115, 144, 148, 156, 354, 366 microglial reactivity during, 388 aminoguanidine, 178, 273 amphibian immune system, 394 amygdala, 54, 123, 141-149, 152, 154-155, 180,
renal nerve activity, 128-129 rGK (rat glandular kallikrein), esterase or co-mitogenic activity, 229 splenic sympathic nerve activity, 127-130 submandibular gland peptides, 309 substance P, biological activity, 374 T cell activity, 130, 336, 339, 340 T cell suppressor activity, 339 transcription factor activity, 165 tumor cells, signaling activity, 338 tumor infiltrating lymphocytes, cytolytic activity, 412 VNO (vomeronasal organ) in gonadectomized animals, 314 acute disseminated encephalomyelitis (ADEM), 87-88, 388 acute phase response, 3.9, 15, 23-25, 24, 255, 263,296, 298, 331-335,345 adaptive immunity/immune response See immune response adhesion molecules, 4, 8-9, 19-21.89-90, 334, 470 adipocytes leptin production by, 15, 192, 283-284 adrenalectomy, 5, 104, 117, 128. 130, 133, 216 adrenergic agonists, 209, 434, 442 input, 47, 55 nerve endings, nerves, 49-50 pathways, 117 receptors, 6-7, 16, 130, 209, 434, 443 responses, 209, 216 stimulation, 209, 216 systems, 6-7 adrenocorticotropic hormone, See ACTH allergic asthma treatment of, 118, 203 ~
299-300, 358-360, 368 Fos immunoreactivity in, 144, 360 HPA (hypothalamus-pituitary-adrenal) activation and, 359 NE (norepinephrine) activity in, 145, 148 release of vasopressin and CRF, 180 anaphylaxis (,-ic), 3, 6, 16, 47, 203, 211-213, 295, 307-3 ! 0 hypothalamic lesions and, 3, 6 role of salivary gland peptides in, 295, 307-310 androgen(s), 14. 191-192, 207-208, 269, 271-275,401, 409 control of the SMG (submandibular gland), 207-208 disease activity in SLE (systemic lupus erythematosus) and, 409 production in Leydig cells. 269, 273-275 androgen depression, 271 androgen receptors, 102, 207 androgen secretion, 272 androgen-dependent pheromones, 320 anergy See immunological tolerance angiotensin II, 104, 178, 204 anhedonia, 141, 156, 361 anorexia, 148, 357, 359, 361,367 LPS (lipopolysaccharide) -induced, 367 antibodies, 3, 6-7, 10, 13, 16-17, 20, 22-24, 64, 66, 71, 75-80, 82, 89-93, 102, 192, 196, 198, 216, 225, 237, 240, 242, 264, 270, 274, 284, 331-339, 353, 402, 409-410, 421,447 anti-adhesion molecules antibodies, 89-90 antibody isotopes, 77-79 anticardiolipin, 381 anti-CD 14 antibodies, 164 anti-CD 154 or CD40 antibodies, 92 anti-CD45 antibodies, 337, 338 anticytokine antibodies, 238
477
anti-hapten antibodies, 79
appetite control, 283-289
anti-human Fas-antibody, 417, 424, 426 anti-IL antibodies, 167, 257, 338
appetite suppression, 192-193,286-289 aprotinin, 229-230
anti-MBP (myelin basic protein) autoantibodies, 390-392, 448-450, 455-456 anti-MCP-1 antibodies, 87, 90 anti-myelin antibodies, 93
arcuate nucleus, 145, 178 c-fos protein activation, 125
anti-nuclear antibodies, 410-411 anti-prolactin antibodies, 412 anti-rat corticosterone antibody, 462 anti-RBC (red blood cells) antibodies, 93 anti-substance P antibody, 375 anti-TNF-c~ antibodies, 166-167 dependent cell toxicity, 93 hemolysins antibodies, 63 IgG or IgM antibodies, 78-79, 332, 340-342 natural polyclonal antibodies, 296, 331,332, 334, 336, 409 NGF antibodies, 198, 216 production, 6, 80, 225, 274, 353-354 response, 10, 23, 66, 78, 102, 237, 240, 248, 263, 353-355, 410 synthesis, 242 vasopressin antibodies, 180 virus antibodies, 456 antiestrogens, 103,401,417, 418-419, 421,424-427 antigen presentation, 14, 17-18, 21,239, 271,389, 392, 418 at the BBB (blood brain barrier), 89-91 in macrophage, 130 antigen presenting cells (APC), 74-75, 81-82, 358, 389-390, 395 role of microglia as, 87, 90-91 anti-inflammatory, 212, 309 anti-MBP (myelin basic protein), anti-NAFP (neuronaxon filament protein), See autoantibodies antipyresis fever and, 297-307 antishock anti-inflammatory peptides, 309 anxiety, 116, 141-142, 156, 351-352, 355, 358 APC, See antigen presenting cells apomorphine, 240 apoptosis, 8, 17, 23, 66, 71, 91, 197, 270, 336, 410-411, 417 CD45-induced, 342 killer cell induced, 419, 421,424 LPS (lipopolysaccharide)-induced, 178 NO-induced, 273-274
GHRH (growth hormone releasing hormone) mRNA content, 107 leptin action, 284 arginine, 176, 213, 215, 229, 273 arginine vasopressin, 100, 104, 118, 141-142, 249, 295 fever suppression and, 297, 299 arousal centers basal forebrain, 258-259 Arthus reaction, 6 Ascaris braziliensis, 76
asthma, 6, 7, 16, 47, 68, 118, 198, 203,240, 262, 349, 373-375,401-402 reproductive hormones and, 433-442 astrocytes, 87, 89, 164, 168, 175, 199, 238, 357, 387, 455 IL-1 in, 285 immunoregulation and, 94 NOS in, 177-178, 181,393 substance P and, 16 artherosclerosis, 242 arthritis, collagen arthritis (CA), 231 juvenile rheumatoid arthritis, 118 rheumatic arthritis (RA), 117-118, 148, 217, 349, 355, 373-375, 411-412 atopic asthma, 6-7, 47 dermatitis, 7, 118, 239 atropine, 196-198 atypical measles infection, 402, 447-448, 456 autoantibodies, 350, 410, 456 detection of, 449 to brain, 367, 447-449, 455 to MBP (myelin basic protein), 390-392, 447-450, 455-456 to NAFP (neuron-axon filament protein), 449-450, 455 autocrine/paracrine, 20-21, 23, 164, action of TNF-c~ across the CNS, 166 autoimmune disease, 13-14, 17-18, 25, 71, 90, 105, 108, 117, 255, 270, 295-296, 331,410, 412, 454 HPA (hypothalamus-pituitary-adrenal) axis hypoactivity and, 118
478
application of oral GK in, 226 psychosocial factors and, 355 stress and, 352, 355-361 IVIG (intravenous immunoglobulin) therapy, 339 autoimmunity, 82, 237, 355,361,388-390, 395,409, 412 Associated Behavioural Syndrome (AABS), 383-384 autism and, 447-456 brain-specific, 389, 447-448, 454 organ-specific, 73-75,455 autonomic nervous system, 121-122, 204, 208, 239, 245,
blood-brain barrier (BBB), 94, 178, 284 endotoxins reaching organs devoid of BBB, 163-134, 166-168, 171 regulation of immune response, 87-90 body temperature, 20, 144, 156, 175-178, 261 effect of leptin, 286-288 febrile, 296-301 NO and LPS-induced changes in, 181 bone marrow, 13-16, 20, 22, 47-48, 73, 91,105, 196, 317, 334, 387, 418 cells, 10, 14-17, 23, 48, 51-52, 54 chimeras, 392-394
264, 352 autoreactive (,-ity) Nab, 338 antibodies, 339 T cells, 88-89, 95, 274, 355,388, 410 immune responses, 94 AVP, See arginine vasopressin axotomy, 47, 49, 54, 88
B
Boyden chamber, 87, 89-90 bradykinin, 196, 206, 349, 373 brain injury cytokine alterations and, 143-144, 366, 369 brain stem, 116-117, 121, 123, 125, 133, 143-145, 178, 256, 284 breast cancer, 418-420 Broca, horizontal diagonal band of, 258 bromocriptine, 7, 10, 401,409-410, 418-420 Bruce effect, 296, 313, 320
B cells, 66, 73, 197, 205,207,226-227, 263, 332, 335, 338, 353,401,410, 456
C
B D N F (brain-derived neutrophic factor) synthesis in,
240 growth hormone therapy and, 403-405 receptors, 64, 197,405, 411
2-(4-Carboxyphenyl)-4, 4, 5, 5-tetramethylimidazoline1-oxyl-3-oxide (C-PTIO), 273 calcitonin gene related peptide (CGRP), 17, 196, 460
B27/CTLA-4, 90 bacteremia, 263 bacteria(l), 66, 82, 256, 263, 273, 358 endotoxins, 141 - 145, 163, 166-167, 169, 358 Gram-negative, 92, 123, 164, 169-170 infections, 14, 156-157, 169-170, 255,334, 355-356 liposaccharides, 106, 107, 169, 273-274, 287, 297,
calcium fluorphore fluo-3, 196 cAMP responsive binding element protein, 101 cancer, 25, 74, 78, 157, 295,352, 417-427 breast cancer, 408-410 prostate cancer, 418, 423 cancer cells, 22, 78, 82, 417-427 cancer immunotherapy, 401, 417-427
389 phagocytosis of, 460 basal forebrain, 257-259, 264 based activation motifs, 8
capsaicin, 7, 373-375,460 caspases, 152, 366 castration, 4, 102, 207, 270-271,320-322, 325 catecholamines, 6, 9, 16, 21-22, 24, 48, 125, 133, 141,
basement membrane, 89-90 Bcl-2, 426-427 bed nucleus, 123, 144, 299-300 behavioural changes, 141, 216, 360, 428 and altered immune response, 240 beta-adrenergic receptor, See adrenergic receptors
238, 334, 356, 460-463,468 depletion, 47, 50-51, 54 CD4 T cells, 22, 59, 68, 79-80, 84, 87-90, 206, 226, 274,
beta-adrenergic theory, 6 beta-endorphin ([3-END), 13, 21,104, 142, 455
334, 355, 394, 421 apoptosis in, 17 auto-reactive, 87-88, 387-389 cytotoxic activity, 13,274, 419 myelin-specific, 387-389
479
peripheral tolerance, 73-74, 80-82 subsets, 242 /V~8+ T cells, 358 CD8 T cells, 17, 22, 65, 68, 73, 206, 394, 421 cytotoxic activity, 274, 419 stress and, 354 CD 11b T cells, 203, 310 CD14 T cells, 59, 92, 163-171,333 CD14 receptor, 163-171,298, 310, 333-334 CD16b T cells, 92, 310, 418 CDI8 cells, 310 CD40 cells, 87, 92 CD45 cells, 296, 331-332, 393-394 NAb binding and, 337-342 CD80/CD86 cells, 90-91, 94 CD 154 cells, 92 cell-to-cell interactions/regulation, 4, 8, 18-19, 25, 71(-84) central inflammatory stimuli effects on immune system, 130-133 central nervous system (CNS), 7, 9, 20, 25, 59, 117, 217, 258, 283, 298 autism and, 448 auto-immunity, 387-395 cytokines and, 156, 238, 243-246, 260, 295, 365-370, 388, 392-394 disorders, 87, 366, 369 gatekeepers, 392 IL in, 256-257, 359, 365-370 immune cell entry and activation, 388 immune system interactions, 60, 87-95, 127, 141-143, 163-171,191-193, 238-250, 295, 298, 349-361,387-395,460 modulation by SEB, 359 neurotrophins in, 240 stress and, 12, 116 T cell function and, 357 cerebral malaria, 369 cerebrospinal fluid, 50, 256, 388, 411 c-fos, 123, 125-128, 133-134, 143-145, 155, 176-177, 259, 359-360 mRNA, 125, 145 chemical axotomy, 47, 49, 54, 88 chemokines, 66, 90, 92-93, 238, 250, 350, 387, 393,395 function, 241-242 chemotactic protein monocyte chemotactic protein (MCP), 402, 462 chronic
fatigue syndrome, 218 inflammation/inflammatory disease, 17, 87, 118, 249, 255, 349, 374-376 insomnia, 258, 263 stress, 60, 104, 142, 156-157, 198, 216, 353-354 ciliary neurotrophic factor (CNTF), 106, 168, 368 circadian sleep-waking brain, 255, 264 variations, 14-15, 25, 261 regulation of cytokines, 260 circumventricular organs, 127, 163, 165, 168, 178, 178, 297-298, 357 cisplatin, 426 classical conditioning of immune resoponse, 192, 237-250 classical estrogen receptor (ER-ct), 419, 426-427 clinical allergy, 67 Clonal Selection Theory, 76 CNP, 447, 449-450, 455 CNS, See central nervous system CNTF, See ciliary neurotrophic factor Coffman, Bob, 67 cognitive deficits, 380-381 Cohnheim, Julius, 62 Collagen, 423,206 arthritis See arthritis oral, 192, 231-234 colony-stimulating factor (CSF), 10, 15, 24, 47-48, 272, 285,419 common mucosal immune system, 19 concanavalin A (ConA), 13, 227, 420 convertase, 104, 107, 206 cortex, 256, 359 prefrontal, 141-142, 146, 154, 368 adrenal, 5, 104, 198, 300 frontal, 123 deactivation of, 258 conditioned immunesuppression, 241,248 corticosteroids, 4, 21, 104, 107, 294 binding globulin, 144 corticosterone, 15, 123, 125, 128, 130, 248, 300, 358-359 cytokines and, 141,143, 148-153 radioimmuinoassay for, 462 corticotropin releasing factor (CRF), 12, 15, 24, 128, 192, 195, 198-199 gene, 177, 180 IL-1 and, 123, 125, 179
480
NO and, 175-180 sleep and, 255 leptin and, 283-284, 287 corticotropin releasing hormone (CRH), 116-118, 208 cytokines and, 141,143-144, 151-152, 154-156, 357
cytotoxic T lymphocytes (CTLs), 23, 77, 79, 274, 332, 417,419
D
activation of CRH neurons by SEB, 358-360 gene, 360 ACTH release and, 99, 104, 106, 116-117, 141,300 receptor, 104 cortisol, 15, 260-262 costimulatory signals, 22, 74 costimulatory molecules, 74, 81-82 CP90-994-1, 196 cribriform plate, 260 cryptorchidism, 269-270, 272-273 CST-SMG (central sympathetic trunk-submandibular gland) axis, 204, 209-212 control mast cell function, 211-212 role in anaphylactic and endotoxic reactions, 210, 211 cutaneous antigen, 460 cyclophosphamide, 237, 240-241,243,248, 252 cystatin S, 209 cytokine(s), behavioural and central neurochemical consequences, 141-157 changes, role of, 60, 238 circadian regulations, 260 expression, 130-131,238, 243,246, 249, 287, 332, 368 function and localization, 349, 365-367 leptin and, 287-288 mRNA, 130-131,243, 246, 461-464, 467-468 neuroimmune interactions, 260-264, 285-286, 356-357 peripheral activities, 367, 260 pleitropism, 24, 105-106, 195, 197, 288, 365 production, 13-14, 17, 59, 67, 91, 93-94, 133, 143, 167, 169-170, 237, 239, 241-243,245-246, 249, 263, 287, 333-334, 339, 356-357, 361,375, 389-390, 459, 461-464, 468-470 production, type-1 vs. type-2, 68, 240, 242, 262 pro-inflammatory, 89-92, 107, 123, 143-144, 148-151, 163-170 protein, receptors, 15, 66, 102, 143, 156, 357, 366 regulatory role, 243-247 sleep and, 255-256, 260-264
danger signal, 74 DBA/2 mice, 324-325, 417, 419-421 deafferentation, 47, 49-50, 53, 121, 127, 133 delayed-type hypersensitivity, 23, 62, 77, 79-81,210, 227, 338 response, 65, 67-68, 81,207, 227-232, 353, 388 depression, 99-100, 116, 118, 144, 156, 352, 355, 361, 366, 450 depressive illness, 148, 156, 357 dermatitis, 7, 17, 118,461 dexamethasone, 14, 102, 178 disodium cromoglycate, 198 dopamine, 10, 99, 103, 105, 141-142, dorsomedial hypothalamus, 359 dysmyelinization, 452
E effector T helper cells (eTh cells), 73, 79, 81-82, 84 embryo survival, 313,324 emotions, 3, 25,352 encephalitis, 349-350, 369 pathogenesis of, 387-395 encephalomyelitis acute disseminated encephalomyelitis (ADEM), 87-88, 388 experimental autoimmune encephalomyelitis (EAE), 87, 117, 198, 238, 240, 351,353,387-388 murine encephalomyelitis virus, 91 endogenous pyrogen, 285 endothelial cells, 8, 25, 87, 90, 123, 125, 164-165, 273, 297, 357, 375,470 NOS in, 175-181 endotoxin, 6, 15, 130, 133, 141, 143-145, 148, 156, 163-171, 175, 179, 273, 298 binding protein, 22 hypotensive response, 211, 212, 307-309 -induced shock, 17, 308-309 lipopolysaccharide, 123, 144, 163-171 shock, 470 enhanced apoptosis, 410
481
eosinophilia, 16, 67 eosinophils, 16, 68, 197, 203, 295, 310, 332, 374 epidermal growth factor, 19, 25, 99, 107, 204, 206, 308 epitope spreading, 82, 350, 387-390, 395,402, 410 erythroleukemia, 336, 419 estradiol (E2), 334, 345, 361,438 estrogen, 13, 14, 103, 301, 401-402, 409-410, 412 asthma and, 433,437-438, 441 receptors (ER-c~, ER-[3), 102-103,417, 419, 426-427 eTh-cells, See effector T helper cells experimental autoimmune encephalomyelitis, See encephalomyelitis extra-cellular matrix (ECM), 89-90
F FACET (Formoterol and Corticosteroids Establishing Therapy), 434 Fas, 17, 93, 94, 148, 192, 334, 342, 401,417-418, 424-427 ligand (fas-L), 274, 401,417, 424-427 membrane receptor, 424-414 fat cells, See adipocytes Fc~,RIIIb,310 fear, 142-143,358-360 febrile illness, 24-25 feedback inhibiton, immune funtion, 128-130, 182, 351 female to male ratio, 410 females autism and, 455 fever regulation, 295-296, 300-302 odor preferences, 313, 319, 321,326 SLE (systemic lupus erythematosus) in, 401,409-410 sleep physiology, 261-263 fever, 4, 17, 13, 23-24, 62-63, 123, 125, 164, 181,192, 263, 297, 334 antipyresis and, 297-302 leptin and cytokines, interaction in, 283-289 Ficoll-Hypaque, 48, 420-424 Fischer (F344/N) rats, 11, 117, 309, 420 follicle stimulating hormone (FSH), 13, 99-100, 105, 107, 269, 272 Fos (c-fos gene protein), 123, 125-128, 133-134, 177, 299 frontal cortex, See also prefrontal cortex, 123
G GABAergic neurons, 144, 258-259 galanin, 99, 284 Gallo, Robert, 66 gender differences, in asthma, 433-442 in autism, 427-417, 454 in fever, 300-301 in sleep, neuroimmune and endocrine systems, 105, 261-262, 317 general adaptation syndrome, 3-5, 23,349 Gershon, Ehrlich, 64 ghrelin, 101 glandular kallikrein, See also kallikrein in immunoregulation, 225-234 glucocorticoids (GC), 6, 13, 106, 141,170, 208, 255, 301,334 antagonist RU486, 117 immunosuppressive effects, 13-14, 23-24, 104, 116, 350, 356 role in GH pathway, 102 glutamate receptor antagonist, 128 glycosylation-enhancing factor, 226 gonads, 25, 99, 208, 270, 274 gonadal steroids, ? gonadotropin, 5, 99, 269, 272, 433 and sex hormones, 13 gonadotropin releasing hormone (GRH), 105,434, 436, 442 Gram-negative bacteria, See bacteria granulocyte macrophage colony stimulating factor (GM-CSF), 10, 47-48, 51-55, 67, 197, 409, 411, 419 growth hormone (GH), 9-12, 19-21, 24-25, 99-104, 106-107, 324, 345,401 axis, 102 deficiency in children, 403-405 lactogenic hormones and, 10-12 receptors, 102, 405 therapy, 403-405 growth hormone releasing hormone (GHRH), 99, 101-102, 107, 255 Guillain-Barrd syndrome, 339 gut associated mucosal tissue, 233
482
human pituitary dwarf, 10 human subjects, 354-355,449 hypersensitivity, 6, 47, 68, 211
H 17c~-hydroxylase, 273 1-25-Hydroxy vitamin D3 (VD3), 14 5-HT activity, 145, 148, 151 6-hydroxydopamine hydrobromide, 49 hapten, 17, 79, 198-199, 239 -carrier (h-C) conjugate, 79 heat shock proteins, 22, 74, (94) helix bundle peptide, 102 hematopoesis, 375,405 hemorrhage, 14, 271 hidden epitopes, 410 hippocampus, 52, 54, 123, 145-148, 175, 177-178, 256 IL- lra induction, 367 lesions, 54, 394 histamine, 6, 16, 196-197, 205, 238-239, 349, 373, 375, 460-461 synthesis history, 47-54 historical postulate, 76-77, 80-82 HIV infection, 241-242 homeostasis, 20-24, 60, 64, 66, 99-100, 108, 134, 182, 195, 208-209, 217, 226, 255, 264, 300, 331,334 mechanisms, 6, 204, 260 milieu, 23 homotopes, 22, 24, 331,333-334 hormonal control of the SMG (submandibular gland), 207-208 Hormone Replacement Therapy, 402, 433-442 hormones, 8-13, 18-20, 22-24, 59, 99, 103, 191,255, 284, 300, 310, 405, 412, 468 action of, 4, 249 anti-shock, 309 competence, 19, 21, 25 cytokines as, 17 endocrine, 314 expression in women, 332 HPA axis, 334-335 pituitary, 3, 7, 59, 99-101, 106 placental, 3 sex hormones, 105,210, 214, 261,331,335,401, 417, 433-442 steroid hormones, 4, 20-23,209, 333-334 stress hormones, 116, 208, 210, 238 therapy, 403-405 thyroid hormones, 12, 105, 207-208, 210 human MHC (major histocompatability complex), 234
delayed, 23, 62, 77, 79-81,210, 227, 338 contact, 17, 191,210, 239 to haptens, 17 hypophysectomy, 5, 11,207-208 hypopituitary mice, 404 hypothalamic control of somatotrophs, 101 homeostasis, 134 hormones, 59, 100, 103-106, 123, 359 lesions, 3, 6-7, 48, 49-54 neurons, 102, 106, 125, 128, 133, 357 nuclei, 49-52, 123, 125, 145, 176 receptors, 192, 283-284 regulation of histamine synthesis, 47-54 hypothalamic-pituitary-adrenal (HPA) axis, 121, 216, 242, 249 activation, 10, 24, 116, 121, 133, 143-144, 240, 260, 356, 358-359 control, 60, 121-122, 130 cytokines and, 242-243 feedback regulation, 60, 208 hormones, 12, 333-335 hypoactivity, 118 regulation of by NO, ! 75-182 response to immune challenge, 60, 116-118, 143, 151, 179, 248, 351 role of in inflammatory disease, 117, 192 hypothalamus, 47-48, 99-105, 123, 126-128, 133, 142-145, 152, 157, 180, 192, 217, 256, 284, 297, 359 -autonomic nervous system axis, 264 CRH release from, 99, 117, 144, 208 interleukine production in, 283, 286, 368 mamillary region, 6 NOS gene expression in, 177-178 paraventricular nucleus, 116, 125, 141, 146, 175, 181
identical twins, 410 idiopathic brain disorder, 447-448 idiopathic thrombocytopenic purpura, 339 IFN (interferons) See also cytokines IFN-et, 16, 249, 255,419-424
483
IFN-[3, 93 IFN-~,, 23, 67, 93, 151,239, 242, 262, 270, 273,350, 358, 361,384, 387, 411 hormones and, 11, 13-17 dominated response, 68 lymphocyte migration and, 89, 92 IGF-I (insulin-like growth factor), 9-12, 25 immune system interactions, 403-405 GH interactions, 9-10, 19-21, 24, 99-102 receptors, 10, 405 IgE, 17, 23, 79, 196-197, 226 induction by IL' s, 15, 67-68, 207 IgG, 23, 67, 78-79, 197, 205, 331-332, 335-336, 338, 340-342 IgM, 23, 228, 331-332, 335-336, 338, 340-342 production, 205-206, 232 IL (interleukins) See also cytokines, 12, 17, 65, 69, 99, 105-106, 285 function, 105-106 IL-1, IL-2 See IL-1 and IL-2 IL-3, 14, 16, 25, 47-48, 51-54, 67, 117, 384, 392 IL-4, 15-17, 23, 67-68, 92, 207,232-233, 238-239, 242, 384 IL-5, 67-68, 92, 197 IL-6, 11, 13, 16-17, 25, 106, 108, 117, 143, 148-151, 156, 164, 167, 179, 198-199, 249, 260-261, 286-287, 298, 334, 356, 361,367, 379-381,384, 411,460, 464, 466, 468-469 IL-7, 409, 411 IL-8, 90, 179, 199, 384 IL-10, 13, 92, 117, 180, 199, 238,242, 246, 262, 411 IL- 11,106, 149 IL-12, 12, 17, 67, 87, 92, 392, 418 IL- 13, 67-68 IL-15, 92 receptors, 16, 105, 106, 123, 149, 169, 175, 285, 367, 369 IL-1 (interleukin-1), 7, 16-17, 53, 141, 155, 164, 242-247, 249, 295, 334, 349, 356-357, 361, 365-370, 402 antagonists, 13, 130 brain induction, 368 effects, 11, 16, 53, 105-106, 117, 123-127, 130, 179, 192-193, 197, 238, 259, 298, 349, 361,365-370 IL- 1ct, 47, 105, 260, 366-369 injection, 121, 123, 125, 127-128, 130 leptin and, 283, 286-287, 289 prolactin and, 409-411
mRNA, 133, 256, 260 neural-immune interactions, 356-357 production, 11, 15, 130, 192, 261, 192, 237, 245, 260-261 receptor antagonist (IL-lra), 16, 144, 285-286, 288, 298, 349, 366-367, 369 receptor, 106, 123, 149, 169, 175, 285, 367, 369 reparative capacity, 143 secretion, 128 sleep effects, 255-257, 259-264 stressor-relationships, 141 system, elements, 367 IL-I[5, 105-107, 176, 260-263, 349, 366-370, 459-463, 466-469 anti-IL- 1[3 antibody, 167 effects, 145-151, 155-157, 167-170 fever and, 258, 298-299, 301 Fos-immunoreactivity and, 143-144 injection, 118, 143, 176-177, 179 interactions, 12, 16 leptin and, 283,287 mRNA/synthesis, 106, 144, 167, 176-180, 367, 463 NO and, 175-181 IL-2, 10-15, 23, 25, 66-68, 90, 92, 106, 117, 142, 148, 156, 242, 345,349, 357-360, 368, 373, 384, 401, 409, 411,406-424, 427-417 effects, 10-15, 106, 117, 180, 411,406-417 immunotherapy, 357 mRNA/expression, 67, 358, 368 NO and, 179-180 production, 11-12, 68, 130, 261,263, 270, 360 receptors, 106, 337-339, 345, 411 secretion, 14, 143 sleep effects, 260-263 treatment, 419-424 immune Function, neural regulation, 119, 121,351 neuroendocrine regulation, 119, 333 role autonomous nervous system, 119, 121-135 role of cytokines in, 100 role of hormones in, 102, 104-105, 401,403-404 sleep and, 255-256, 260-264 spinal cord injury and, 459-471 splenic, 121-122, 124, 128 stress and, 130, 133, 216, 300, 351-354, 356, 361 testis-associated, 269-271 immune privilege, 8, 88, 270-271,274, 387-388 immune regulation
484
of activated cells, 331-342 immune response
peripheral, 238, 285 prolactin and, 411-412
inflammatory, 387
salivary gland and, 308
acute phase, 3, 9, 15, 23-25, 24, 255,263,296, 298, 331-335,345
sleep and, 255, 260-263 stress and, 351-361
adaptive, 13, 19-20, 22, 24-25, 63, 71, 91-93, 331-332
testis interactions, 269-275 immunity,
autoreactive, 94 behavioral modification, 7
adaptive See adaptive immunity autoimmunity See autoimmunity
conditioning, 25, 192, 237-250
cell mediated, 13-14, 23, 61, 77-78, 80, 82, 216, 361,
dynamics of, 61-69
409, 418, 456
genes, 454
central inhibition, 60
glandular kallikrein in, 225
classes, 68, 77-79 conditioning See under immune response
hormones and, 3, 12-13, 15, 19, 99-100, 116, 409 improvement with drugs, 115
cytokine regulation, 59, 192, 238
in CNS (See also autoimmunity), 387-395,448,
genes, 325
455-456
hormones and, 12-14
innate, 24, 68, 92, 164, 169-171,332, 334, 388 interleukins and, 105
humoral, 23, 62, 78, 264, 404, 409 kallikreins and, 230-231
late phase, 309-310
modulation, 240-241
maternal cell-mediated, 313, 324
mucosal, 238
modulation, 206
natural, 22, 24, 217, 246, 296, 331-334
peripheral, control of, 60, 121 - 123, 128, 130, 135 pheromones and, 313
neuropeptides and, 238 organ-specific, 73
pregnancy and, 409
peripheral, 60
primary, 23, 74, 82, 307
protective, 62-63, 68
regulation within CNS, 87-95, 116, 121, 163-171
regulation of, 64
SCI and, 468
substance P and, 375
sleep, health and, 262, 351-361
systemic, 217
stress and, 116, 216, 360
to tumors, 418
systemic, 163-171,333
type 1 vs type 2, 68
testis and, 269-275 Thl/Th2 nature of, 76, 79, 84 immune system adaptive, 331
understanding, 61 immunoreceptor thyrosin based activation motifs See ITAM immunosuppression, 18, 121,227,437
appetite and, 285,288
conditioning, 248
autoregulation, 334 cancer interactions, 417-419, 427
hormones and, 10, 14, 99, 356
cell-to-cell communication and, 8, 9, 18
sleep and, 263
CNS interaction, 238-242, 248-250, 295,349-361,
leptin and, 15 stress-induced See under stress
388, 394, 403-396, 447-456 fever and, 298
therapies, 379 inducible NOS (iNOS) See nitric oxide synthase
hormones and, 3, 4, 10, 12-14, 19-20; 403-396
inflammation
IL-2 and, 338 modulation, 339 mucosal, 18-19, 232 natural, 20, 24, 331-333 neurotransmitters and, 16-19
cyclical airway, 442 IVIG therapy for, 339 inflammatory bowel disease See irritable bowel disease cells, substance P and, 374
485
disease, animal models, 117 disease, substance P and, 375
K
stimuli, 130 inhibins, 99, 105, 107, 274 innate immune system, 20, 24, 68, 74, 164, 166, 169-171,246, 271,331-333, 388
K562 target cells, 418-420, 414 kainic acid, 367 kallikreins, 191, 192, 204, 206-207, 209
microglia as constituents of, 92 innervation, 7, 9, 211,238, 301 immune organs, 122-123, 126, 130, 133,334, 442-443 inputs,
glandular, 225-234 Kawasaki Syndrome, 339 Kelso, Anne, 68 keratinocytes, 14, 410 Korneva and Khai, 7
adrenergic, 47, 55 from limbic forebrain, 133 hypothalamus, 123, 125
L
neural to skin, 433 noranergic, 126
Langerhans cells, 8, 17, 25,460
pathways, 127
late phase, 308-310
regulatory to immune organs, 434, 443 rostral (to the PVN), 121, 133
lateral sepatal region, 123
to spleen, 122, 128
interaction in fever and appetite control, 283-289 lethargy, 151,357, 359
insulin, 15, 24, 255, 284 -like growth factor See IGF-1 IGF/insulin system, 20 interferons, See IFN interleukins, See IL intracellular
leptin, 192, 193
leukemia, 270, 332, 403,405, 419 inhibitory factor (LIF), 24, 99, 103, 106-107 leukocyte, 17-18, 144, 262, 334, 354, 392-393,405,470 activity, 24, 307-310 common antigen, 337
calcium levels, 273 level of cAMP, 101
communication, 25
parasites, 74, 78, 80
distribution, 16
-derived peptides, 105
response, 103
endogenous mediator (LEM), 7
second messengers, 152
entry to the CNS, 389
signaling, 68, 246-247, 249, 336-337 irritable bowel disease, 239, 349, 373,375
inflammatory, 388 polymorphonuclear, 11, 15-16
ischemia, 143,270, 273, 388, 390
receptors, 249-250
ITAM/ITIM (immunoreceptor thyrosin based activation/
rolling, 307-310
inhibitory motifs), 8 IVIg Therapy, 331-332, 334, 336-342
trafficking, 18 leukotrienes, 22, 197, 401 Lewis rats (LEW/N), 117, 234, 355-356 Leydig cells, 105, 192, 207, 269-275 androgen production by, 269, 272, 275
JAK/STAT pathways, 242
L-histidine decarboxylase (HDC), 47-49, 54 limbic forebrain, 123, 133, 144, 258
Jancso, Miklos, 3, 7, 373
lipopolysaccharide (LPS), 106-107, 129, 134, 144, 149,
juvenile rheumatoid arthritis, 118
164, 176, 179, 256, 273, 287, 297, 356, 459, 461 -binding protein (LBP), 168, 298 circulating LPS, 163 -induced inflammation, 125,295, 307-310 -induced release of CRF, 175
486
-induced shock, 308 receptor, 92 responses, 181,239, 273-274, 298 systemic administration, 388 L-NAME (NG-nitro-L-arginine methyl ester), 125-127, 178-181,273-274 lupus anticoagulant, 381 Lymphokine activated killer (LAK) cells, 11, 417, 419-424, 426-427 lymphotoxin, 67
M 3-morpholino-sydnonimine (SIN- 1), 180 macrophages, 8, 11, 14-17, 25, 60, 93, 121-123, 128, 130, 133, 142, 164, 167-171, 192, 203, 210, 212, 238-240, 270-275, 296, 298, 310, 324, 331-335, 349, 350, 356, 373-375,387-388, 392-394, 456 chemotaxis, 206 derived cytokines, 143, 148 migration inhibitory factor (MIF), 99, 107 monocyte/macrophage cell types, 25, 92, 164, 211 major histocompatibility complex (MHC), 9, 21-22, 65, 90-91,332, 334, 390, 419 class II antigen, 22, 90, 313, 345,350, 358, 387, 389-390, 395,419 specific odours, 313-326 MAP (mitogen-activated protein) kinase, 164, 169, 246 mast cell, 6, 9, 15, 17-19, 25, 68, 332, 374 activity, 215, 239, 438 degranulation, 13, 16, 216, 374, 438 differentiation, 240 discharge, 349 function, regulation of, 191, 211-212 hyperplasia, 239 histamine release from, 205, 275,460 mediator release, 245 mucosal, 25, 191,196, 198, 237, 239 -nerve communication, 195-199, 239, 334 peritoneal, 211, 216 precursors, 48 substance P, 373-375 mating preference, 313-314, 319, 324-325 MCA- 106, 420 MCP-1, 87, 90, 242, 387, 395,459, 468-469 mRNA, 461,464, 466-469 measles, 402, 447-448, 456
antibody, 402, 449, 452, 456 -mumps-rubella, 456 melatonin, 14-15,255, 261 membrane attack mechanism, 401,424 menarche, 410, 434 meningitis, 369 menopause, 401, 410 menstrual cycle, 261-262, 264, 401,433-442 Metalnikov and Chorine, 7 metabolites, 5, 23, 105, 123, 145, 155 arachidonic acid, 144, 274 Metchnikoff, Elie, 62-64 MHC, See major histocompatibility complex microglia, 87-91, 163-171,285,350, 357, 368, 387, 390, 394 reactivity in Alzheimer's disease, 388 role in immune response/system, 91-94 parenchymal, 87, 91, 163-171,393 perivascular, 87, 89, 90-91, 175, 178, 181 NO in, 175, 178-179, 181 role in regeneration, 93 milieu interieur, 20 MK801, 126 monocyte, 11, 25, 92, 118, 164, 167, 169, 206, 211,261, 263, 349, 361,373,403,405,409, 411 /macrophage cells, 11, 14, 92, 164, 211,298, 393 chemotactic protein (MCP), 215,402, 464 interleukines and, ! 5, 105,338 receptors, 16-17, 298 systemic, 90-91 mononuclear cell-derived prolactin, 412 cells, 13, 421 phagocytes, 13, 17, 260, 356-357 Mosmann, Tim, 67 Mouse/mice, adenohypophysis, 106 allo-recognition in, 296, 313-324 antibody response, I0 autoimmune reactive, 350, 357 axotomy in See axotomy B cell maturation, 12 BALB/c, 358 bone marrow, 48, 393 Bruce effect in See Bruce effect C3H/HeN, 335 C57BL/6, 48-49, 238, 243 cancer in, 417-418, 421,427
487
capsaicin, systemic administration of in, 7
peritoneal macrophages, 170
catecholamine-depleted, 51
pituitary, 103, 106-107,404, 410
CBA/J, 49
SEB challenged, 359-360
CD 14-deficient, 169 chimeric, 387, 393
Sertoli cells, 274 sham skin grafting, 237
CNS, cytokines in, 390, 394
SJL/J, T cell lines in, 390, 393
conditioned, 169, 192, 238, 241-247, 249-250
SLE models, 118
DBA/2, 417, 420-421 dendritic cells, 12, 239
SMG, 205,207 Snell-Bagg, 404
DHEA administration, 14 DTH response to picryl chloride, 207, 229-231
stress, 216, 237-239, 353
spleen, 75, 102, 421
dwarf, 404
syngeic C3H/HeN mouse, 335
EAE in, 355, 393 haptens in, 17 hematopoietic progenitor cells, 48
T cells in, 81,393
hemopoietic system, 317 hormone-lacking, 100 IFN-gamma deficient, 350-351,387, 393
testes, tissue extracts, 274 Tfm/Y, 207 TLR-deficient, 169 TNF-cz treated, 151-155, 166-168, 192, 244 transgenic, 102, 107, 198, 249, 387, 390-394
IL's and, 168, 359, 368 immune responses, 72, 75, 102, 192
Twitcher, 388 use of human recombinant cytokine in, 152
immunized/immunization of, 76, 243, 263-264, 271,
VCSB subclass, 213 Whitten Effect in See Whitten effect
418 immunodeficiency, age related, 14 infections, 355 ischemic injury, 366 kallikreins, 206, 225 knockout See knockout mice leptin treatment, 15, 284, 287 Leydig cells, 275
Lps locus, 169 LPS response, 169, 192, 335, 367 LPS toxicicy, 16 lupus-prone, 118, 379, 410 lymphoid organs, 75 macrophage targeted, 392 macrophages, cytokine expression in, 287
MRL mice, 118, 350 Multiple Sclerosis (MS), 87-94, 117, 238, 240, 355, 387-390, 455 mucosal immune system, 13-14, 18-19, 232, 238, 264, 308, 333 mucosal tissue, 17, 25,245 mast cells, 25, 191,196, 198, 237, 239 gut associated, 233
Mycobacterium tuberculosis, 71, 76, 78 Mycoplasma pulmonis, 460 myelin autoimmunity to, 447-448, 452, 455-456 basic protein (MBP), 95, 117, 240, 355, 390, 402, 447-448
MRL mice, 118, 350, 379, 382-384
breakdown, 88 coated peptides, 93, 355
myocarditis in, 389
constituents, 91,447
Nab treatment, 336 NGF treatment, 199, 205, 216 NK-resistant tumor, 331
markers, 450 oligodendrocyte glycoprotein (MOG), 93, 355 oligodendrocyte specific protein (MOSP), 455
NZB/W, 409-410
reactive T cells, 89
NZB/W, 410, 419
specific T cells, 388-390
MCA- 106 bearing, 420
obesity, 284 orchidectomy, 270, 274 osteopetrotic model, 272 ovoalbumin in, 390
488
neuroexocrine-mucosal system, 18-19 neurogenic
N 2-N-benzoyl-arginine ethyl estez assay, 229 NAFP, 447-450, 452, 455 natural antibody, 24, 296, 331-334 natural immunity, 20-22, 217, 296, 331-334 neuroendocrine regulation, 333-334 natural killer (NK) cells, 68, 217, 240, 331-332, 335, 338, 404, 417-420, 424 abnormalities, 448
inflammation, 7, 9, 16, 142, 349, 373-376, 460 stressors, 145 injury, 365 events, IL-1 induction after, 368, 369 neurohormones, 88, 237-238, 241,295 neuro-immune dysfunction syndrome, 456 interactions, lupus as a model of, 379-384 network, 88, 135
activity, 130, 179, 237, 261-263, 354, 419-420
neuro-immunopathogenesis
cytotoxicity, 11, 13-14, 17, 22, 418, 414, 427 receptor antagonists, 196, 198
neuro-immunoregulation,
receptors, 8, 196, 405 -resistant tumors, 296, 331,335 naturalistic stressors, 354 neocortex, 258 neophobia, 359-360
in autism, 447-456 basic concepts, 19-25 neuronal cells, 93, 115, 15 I NOS, 60, 175-177, 179, 181-182 activation, 176
neoplasia, 242, 336
networks, 148
neoplastic disease, 205, 216
death, 115, 119
neoplastic cells, 335-336
response pathway, 116-117, 119 degeneration, 88
nerve
activity, 128, 460
damage, 144
damage, 115, 119, 461
receptors, 16
endings, 49, 143, 239, 469 fibers, 16, 60, 121 - 123, 130, 239, 468 growth factor See nerve growth factor impulses, 25 -mast cell communication, 195-199, 239, 334, 373 olfactory, 260
expression of MHC class I molecules, neuropeptides, 3, 9, 16, 20-21, 23, 105, 116, 125, 128, 130, 178, 197, 237-239, 243, 245-246, 249-250, 287, 295,299, 334, 357-358, 373-374, 376, 460 substance P See Substance P variations, 154-156
peripheral, 116, 128, 298,463 section, 130
Y (NPY), 130, 192, 205, 283, 349, 373,444 neurotensin, 198
sensory afferent, 297, 373-374, 460-461
neurotransmitters, 4, 9, 23, 175, 178, 238, 460
splenic nerve See under splenic substance P containing, 374-375 sympathetic, 121 - 122, 203-204, 209, 211-212, 308, 460, 470 tissue, 238, 243 vagal See under vagal nerve growth factor (NGF), 19, 99, 195, 197-199, 204-205, 240, 308,460 and neurotrophins, 15 neurochemical consequences of cytokine challenge, 141 - 157 mapping, 357 mediators, 360 pathways, 123
cytokines and, 141, 143, 145, 151,249, 259, 264, 356, 469 fever-suppressing, 297 immune system and, 3, 7, 9, 16, 21, 88, 130, 142, 145, 178, 237-238, 249, 334 related PAGE drugs, 115 release, 196, 259 Substance P as, 374, 460 neurotrophic agents, 205 factors, 117, 168, 240, 368 receptors, 15 neurotrophins, 15, 59, 95 in CNS-immune systems interactions, 240, 249
489
neutrophils, 13, 16, 164, 169, 197-198, 203, 212, 217, 295, 307, 310, 332, 349-350, 373-375 influx into lumen, 210-211 chemotaxis, 90, 205-206, 213 NF-•B (nuclear factor kappa B), 17, 152, 163-170, 249 ADDIN ENRfu, 246 cytokines and, 152, 164, 168, 259 Nippostrongylus brasiliensis, 210, 308-309, 412 nitric oxide (NO), 17, 22, 60, 125-126, 143, 145,256, 273, 387 role in neuroimmune feedback signaling, 175-182 inhibition, 180, 393 production, 60, 175, 177, 179, 192, 239, 274, 393-394 nitric oxide synthase (NOS), -2, 349 inhibitors, 125-126, 175, 273 isoforms, 175-176 inducible (iNOS), 60, 145, 175-182, 192, 239, 273-276, 349, 393-394 neuronal (nNOS), 60, 175-177, 179, 181-182 endothelial (eNOS), 60, 175-177, 180 NK cells See Natural killer cells NK receptor antagonists, 196 NMDA (N-methyl-DL-aspartate), glutamate antagonist (MK801), 126
central, of amygdala, 358-360 paraventricular See paraventricular nucleus signal delivery from membrane receptor, 21-23 solitary tract, 142 suprachiasmatic, 260, 359 supraoptic of hypothalamus, 123, 125, 155 VMPOA, 298 NZB/NZW mice See under mouse/mice
O obesity, 192, 283 odor preference of mice, 315-316, 318-319 olfactory nerve, 260 stimuli and allorecognition, 295, 313-326 oligodendrocytes, 91,350, 357, 387-388,456 oncostatin M, 106 oncogenes, 336, 426-427 oral contraceptive agents, 262, 401-402 oxytocin, 99-100, 128, 175, 177 receptor, 103-104
P
glutamate receptor antagonist (AP-5), 128-129 R 1 subunit, 176 receptors, 128, 178 NG-nitro-L-arginine methyl ester See L-NAME noradrenergic, 116, 126, 130, 284 norepinephrine (NE), 104, 142, 152, 434 activity, 145, 148, 151 cytokines and, 123, 141,249 hypothalamic, 123 splenic, 122, 125, 130, 462-463 treatment, 468 nNOS See neuronal NOS NOS See nitric oxide synthase NREM (non-rapid eye movement) sleep, 256 nuclear factors See also NF-KB, 10, 12-13, 21-22 nuclear groupings, 48-49, 52 nuclear proteins, 22, 410 nuclear receptors, 14, 22 nucleus, accumbens, 149, 151 arcuate, 107, 125, 145, 178, 284 bed nucleus stria terminalis, 123, 144, 299-300
p53, 336, 426 P815 murine mastocytoma, 420-421 paraventricular hypothalamic nucleus (PVN), 116, 141, 143, 146, 151-152, 155, 298 immunoregulation and, 121,123-128, 133-135 NO in, 60, 175-182 parvocellular, 143, 176-177 pathogen-associated molecular patterns (PAMPs), 169, 333 pathology, 4, 62, 88, 156 autoimmune, 390, 395,448,455-456 peptides See also polypeptides and proteins, 142, 151, 206, 233, 295, 298-299, 317, 355, 374, 410 antigenic, recognition of, 74, 78-82, 84, 90, 333-334, 389-390 biologically active, 191,204, 206-207, 209-210, 212-213, 216-217, 225-226 CGRP See calcitonin gene-related peptide epitopes, 389 GHRP See growth hormone glucagon-like peptide 1,284
490
helix bundle peptide, 102 /MHC complexes, 78, 332-334 muramyl peptide, 256 neuropeptides See neuropeptides opioid, 22, 356 PACAP See pituitary adenylate ... POMC derived, 12, 274 salivary gland, 307-310 SMR 1-derived, 214-215 VIP See vasoactive intestinal peptide peptidoglycan polysaccharide, 117 pericytes, 91 perimenstrual interval, 262 exacerbations, 433-442 Peyer' s patches, 18, 234 PGE2 See prostaglandins EP4-PGE2 receptor, 125 phagocytic, -osis, 12-17, 23, 62-64, 72, 92, 169-171, 206, 211,260, 262, 392 by APC, 410 by dendritic cells, 12 by macrophages, 13-14, 17, 205, 271,332, 356-357, 375 by NK cells, 460 LPS-induced, 287 population of cells in brain, 164, 166, 170-171 stimulation of, 12-13, 205, 334, 375 theory of immunity, 63-64 pheromonal stimuli See also Bruce and Whitten effects reproductive response to, 295, 313-314, 319-320, 324 pheromones, 320, 324 PI3 kinase signaling, 246 Pit-1 See also nuclear factors, 10 pituitary ACTH, 104-105, 116-117, 142-143, 149, 177, 198 adenylate cyclase activating peptide (PACAP), 17 cytokines, 105-106, 144, 177 function, 126 GH, 20-21, 101-102, 334, 403,405 gland, 7, 10, 21, 25, 99-108, 177, 179-180, 206, 208, 210, 300, 367, 404, 409 hormones, 3, 7, 9-10, 59, 99-108, 409 hypothalamic-pituitary-adrenal (HPA) axis See there prolactin, 11,401,409-412 regulatory circuits, 59-60, 99-108 transplants, 410 placental hormones, 3, 10
plasma ACTH, 260, 358, 360 CD 14 levels, 166 corticosterone,128, 148, 150-151,153, 359, 462-463, 468-469 cortisol, 260, 262 cytokine levels (IFN, IL), 13,261-262, 367 extravasation, 16, 374 hormone levels (PRL, GH, LH, NGF, AVP), 10, 107, 207, 216, 300 kallikreins, 207, 226 leptin levels, 284 membrane, 206, 271, 419 platelets, 16, 206, 221 activating factor (PAF), 117, 310 pleiotropism, 24, 105-106, 195, 197, 288, 365 polypeptides See also peptides and proteins, 105, 213 biologically active, 191,204, 206-207, 209-210, 212-213,216-217, 225-226 POMC (precursor molecule-proopiomelanocortin), 12, 104-107, 283-284 derived peptides, 105, 274 post-traumatic stress disorder, 116 precursor T helper (pTh) cells See also T helper cells, 73-77, 80-84 preference odor/mating, 313-316, 318-319, 324-325 prefrontal cortex, 141 - 142, 146, 154, 368 pregnancy, 207, 301,313, 320, 324-325,401,409, 411-412 asthma and, 433-442 block, 313,320-324 preneoplastic ceils, 335-336 preoptic area hippocampus, 257-258, 298 pressure ulcers, 459, 460, 470 'priming problem', 82, 353,424 progesterone (PS), 14, 21,205,262, 264, 320, 323-324, 436-437 theory, 437 proinflammatory cytokines See cytokines prolactin (PRL), 9, 11, 19-20, 22, 99, 208, 261,334, 345,404 -like proteins, 401,412 releasing protein, 99, 103 role in SLE, 401,409-414 secretion, 320, 323,409 prosomatostatin, 238 prostaglandins, 164 D2, 255, 257
491
E series (incl. PGE2), 121,125-132, 198, 297-298, 361 synthesis, 123, 125,297, 375 prostate cancer, 418, 426 proteases, 92, 191,206-207, 209-210, 214-215, 225-226 protein kinase C, 246, 335-336, 338 proteins See also (poly)peptides, 9, 81, 104-107, 123, 130, 170, 206, 215, 217, 227-229, 274, 285, 309, 353,389, 410-411,426, 459, 462 acute phase, 24, 332, 334, 345 antigens, 357-358 C-reactive and mannose-binding, 22, 24, 335 EFG-binding, 226 Fos (c-fos gene), 123, 125-128, 133-134, 177, 299 G-protein, 14, 18, 101, 103
reproduction, 3, 14, 25, 104, 273, 288, 313 olfactory responses and, 314-315,317, 319, 325-326 reproductive hormones, 261, 401 asthma and, 433-442 rheumatoid arthritis, See arthritis rostral substantia innominata, 258 rostrolateral inputs to PVN, 133 rubella, 447, 449, 452, 456
S Sanarelli-Schwartzmann phenomenon, 6 scarcity problem, 82 Schistosoma mansoni, 16, 226
MHC-associated, 317, 324 mitogen-activated, 164
schizophrenia, 366 Schultz-Dale test, 6 scorpion venom, 196 seizure, 123, 143, 148, 365 self-nonself discrimination, 72-73, 76 Selye, Hans, xiii-xiv, 3-6, 23-24, 122, 349
monocyte chemotactic See under monocyte myelin-basic See under myelin neuron-axon filament, 447 prolactin-like, 409, 412
sensitization, 17, 87, 141,353, 417, 420, 424-427 cytokine-induced neurochemical changes, 151-155, 157 sensory nerve, 16, 373-374, 470
proteolipid protein, 355 receptors for, 92, 101,103 regulatory, 272
innervation, 7, 9 function, 460-461 septal area, 123, 299 serotonergic pathways, 145, 284 serotonin, 6, 123, 141-142, 196 receptor, 455 reuptake, 157 Sertoli cells, 192, 269-270, 273-275
heat-shock, 22, 74 kinases, 101,335-336 liver-derived, 3, 22, 24, 300 LPS-binding, 164, 168-169, 298, 333
STAT, 409 submandibular gland rat (SMR1), 213-214, 309 testis, 269 tyrosin kinase, 246 pTh cells See precursor T helper cells pulmonary inflammation, 210-211,307-308 PVN See paraventricular nucleus
R rapid eye movement (REM), 256-258 RBL cell(rat basophilic leukemia cell), 196-197 receptor crosstalk, 8 reduced exploration, 359 redundancy, 11, 17-18, 20, 23, 192, 217, 365, 412 regeneration, 8, 18, 26, 88, 95, 270, 350 hepatic, 204, 206 microglia as promotors of, 93 releasing hormone,
sex differences, 102 sex hormones, 13, 24, 105, 210, 214, 331,335 dependent tumors, 417 sickness behavior, 359 signal modulation, 21-22 signal regulators, 23 signaling, 92, 94, 101,168, 211,242, 333-334 activity, 338 CNS signaling, 243,246, 249 intracellular signaling, 246-247, 336-337, 351 LPS signaling, 164, 169-171 molecules, 141-142, 356 neuroimmune feedback signaling, 175-182 NF-kappaB, 167-168 pathways, 133, 250, 455
492
within the immune system, 71-84
as adaptive response reaction, 3, 4, 6, 122, 349
SL2-5 lymphoma, 420
Bruce effect and, 322-323
SLE (systemic lupus erythematosus), 118, 240-241, 401,
chronic, 104, 198, 353
409-412 'neuropsychiatric' involvement in, 350, 379-384, 409, 411 as a model of neuroimmune interactions, 379-384 sleep, 119, 192, 243, 255-264, 354, 356-357
autoimmune diseases and, 356 cytokines and, 60, 118 disorders, 116 hormones, 12, 116, 208, 238 immune reactivity and, 25, 60, 115-119, 124, 130,
active neurons, 258-259 enhancing substances, 259
immune response, health and, 351-361
deprivation, 255-256, 262-263
immunosuppressive effects of, 6, 12, 99, 121, 130,
cytokines and, 257, 260-261,295 regulation, immune functions and, 256 slow wave sleep, 123, 257, 261
133, 199, 216-217, 239-240, 349, 354
133, 216, 237, 300 mast cell-nerve unit and, 195, 198-199 infectious disease and, 354-355
promoting regions, 258
neuroendocrine, inflammatory disease and, 115-119
gender differences, 261-264
NO production and, 239
/wake cycle, 25,255-256, 258, 260-262, 264 smell allo-recognition and, 313-314, 317, 319, 323-324, 326 S-Nitroso-N-acetyl penicillamine (SNAP), 273 sodium compounds, 48, 50, 180, 198 somatostatin (SOM), 17, 237, 239, 245-250 spermatogenesis, 105, 269-270, 272 Sphingolin administration of in acoustic children, 447, 452, 454-455 splenic,
APC activation and, 74 submandibular gland and, 216 stress response, 12, 104, 107, 122-123, 127, 133, 195, 209, 248, 264, 301,351-352 neuroendocrine, 116-118 role of immune system, 356-360 stressful events, 13, 116, 122, 144, 216, 262, 351 stressors, 60, 116, 118, 123, 141 - 157, 178, 180, 182, 349, 351-354 induced immunomodulation, 352-353 immune activation as, 141 - 157, 182 stria terminalis, 123, 144, 299-300
cytokine production, 461-463,468, 470
stromal-cell-derived factor-1,242
immune function, 121-122, 124, 128
substance K, 374
lymphocytes, 15,402
submandibular gland (SMG),
macrophages, 121, 128, 130, 133 nerve, 122, 127-133,443
Substance P (SP), 192, 196-198, 237, 239, 460
peptide (S),
norepinephrin, 122, 125,463,469
antagonists, 17, 196, 198, 239
TNF-c~, 130-131, 133
neurogenic inflammation and, 9, 16-17, 349,
staphylococcal enterotoxin A (SEA), 13 enterotoxin B (SEB), 351-352, 358-360 steroid hormones, 4, 20, 209 corticosteroids See corticosteroids
373-376, 460 superantigens, 94, 332, 357-358 suppressor T cells, 64, 233,401,403-404 characteristics, 65 activity, 12, 206, 339
effect on immune system, 107, 238, 269-270, 274, 333-334
suprachiasmatic nucleus See under nuclei supraoptic nuclei See under nuclei
function, 15, 21-23, 122, 274, 301
sympathectomy, 47, 49, 51, 53, 130, 191,210
production, 105,272-273 steroid-thyroid receptors, 12, 14 stress, air-puff stress, 147 antigens, 94
sympathetic activity, 356 ganglia, 15, 205 innervation immune organs, 122, 130, 133,209, 238, 468-469
493
nerve (fibres), 9, 60, 121-122, 127, 203, 209, 460, 468, 470 nervous system, 60, 116-118, 121-135, 196, 210, 261,308-309, 434 outflow, 9, 12, 16, 116 pathways, 298, 307, 310 preganglionic neurons, 125 superior cervical ganglion-SMG axis, 18, 196, 307-308 trunk-submandibular gland axis, 204, 209-210, 212 systemic trauma, 271 Szentivanyi, Andor, 3, 6-7
T Tamoxifen (TX), 14, 401, 417, 419 T cells See also CD4, CD8 .... T cells ~,6, 20, 68, 332 apoptosis, 342 autoreactive, 88-89, 95, 274, 355, 388, 410 cytokines and, 59, 92, 233, 263,337-339, 356-358, 394 cytotoxic, 23, 77, 79, 274, 332, 417, 419 encephalitis and, 387-395 function, 24, 67, 76-77, 80-84, 90-91,102, 270, 331, 456 growth factor, 66 helper cells See Th cells myelin reactive, 89, 95 polarization, 92 receptors, 65, 91,405 SEB activation of, 358-359 sleep and, 261 superantigen, 94 suppressor See suppressor T cells thymus, 102 types (CD4-CD...) See CD4 T cells telencephalon, 258 testosterone, 102, 207, 320, 322, 324, 410, 412 function, 14-15, 24 transgenic mice, 390 SLE and, 412 production by Leydig cells, 207, 269, 272-273, 275 TGF-c~ (transforming growth factor-c~), 375 TGF-[3 (transforming growth factor-j3), 17-19, 21,199, 204-206, 208, 226, 232, 2381 246, 368 TGF producing T-cells, 18, 233, 242
Thl (type-1 helper T lymphocytes), 72-84, 409 cytokines and, 15, 23,404, 411,456 clones, 67 Thl/Th2 nature of response, 15, 72-74, 76, 79, 80, 82, 84, 409 hypothesis, 61, 67-68 Th2 (type-1 helper T lymphocytes), 23, 71, 73-78, 82, 84, 92, 197, 409 Th3 (type-3 helper T lympocytes), 18 T helper cells, 80 type 1 See Thl eTh, 73, 79, 81-82, 84 pTh, 73-77, 80-84 thermogenesis,181,283, 287, 298, 301 third signal, signal 3, Threshold Hypothesis, 82, 84 thyroid gland, 99-100, 191,209, 271 hormones (TH), 12, 21-22, 102, 207-210 receptors, 12, 14, stimulating hormone (TSH), 9, 99-100, 105 thyrotropin releasing hormone (TRH), 9-10, 99, 103-104 thyroxine, 24, 207-208 tuberal lesions, 6 TNF (tumor necrosis factor), 92, 117, 141 alpha See TNF-ot beta (TNF-[3), 117 synthesis, 24, 198, 242 stress and, 141,143-144, 148-157 neurochemical alterations by, 143 receptor, 92, 94, 163, 167 fever and, 298, 334effects, 7, 24, 92-93, 117, 130, 148-156, 163-168 production, 11, 15-17, 130, 133, 144, 192, 212, 237, 402 antagonists, 13, 125 TNF-~ (tumor necrosis factor alpha) See also TNF as neuromediators, 143-144 effects of, 17, 24, 92, 117, 130-133, 143, 148-156, 163-168, 192, 238, 242-249, 387, 395 fever and, 298 indomethacin and, 125 interactions with hormones, 7, 11, 13-15, 272 interactions with leptin, 287 interactions with neurotransmitters, 16-17 neural-immune interactions and, 356, 358 NO and, 178-179 production, 11, 15-17, 130, 133, 144, 166-167, 192,
494
198, 212, 237, 242-249, 392, 394, 459-468 regulation of, 367-371 sleep and, 255-261 SMG and, 211-212 splenic, 130, 131-134, 459-468
producing neurons, 177, 299 VCS (variable coding sequence) gene family, 213, 217, 309 ventral lateral preoptic area, 258 viral
Toll-like receptor (TRL), 163-164, 168-169, 246, 333
agents, 356
toremifene (TO), 401,417, 419
antigens, 66, 358
Trichinella spira, 375
encephalitis, 387 infections, 14, 156-157, 177, 255,264, 350, 355,
transforming growth factor 13(TGF-[3), 18, 99, 107, 204-205,308
387, 389, 448, 455-456
tuberal lesions, 6
Virchow, Rudolf, 62
tumor, 78, 101, 106, 192, 237-238, 271,331-337, 417
vomeronasal organ (VNO), 313-314
cell lines, 107 host resistance to, 296, 401 IL- 1[3 in, 369 sex hormone dependent, 417
W
tumor necrosis factor ot See TNF-ct
wake-active, 258
turpentine, 6, 402, 459, 461-462, 464, 466-469 type 1 T helper cells See Thl
Wannemacker, 7 Whitten effect, 296, 313, 320
type-1 vs. type-2 cytokine production, 68, 240, 242, 262
Women
tyrosine
asthma in, 433-442
phosphorylation, 336
breast cancer, 408 hormones, 13, 332
kinase, 164, 206, 240, 246, 336 receptor-like protein phosphatase, 337
sleep physiology, 261-262
U
Y
ultra violet light, 410
Yac-1 murine lymphoma, 338, 420
urocortin, 284 Ussing chamber, 196, 199 urticaria, 375 uveitis, 117
V vagal afferents, 127, 133,442 transmission, 180 nerve (nervus vagus), 9, 143, 180, 264, 286, 297, 368-369 vasoactive intestinal peptide (VIP), 17, 99, 103, 125, 239, 245, 255, 349, 373 vasopressin, 100, 104, 125, 175, 179, 180, 300 antagonists, 128 antibodies, 180 arginine vasopressin, See AVP