Advances in Cancer Research Volume 59

ADVANCES IN CANCER RESEARCH VOLUME 59

This Page Intentionally Left Blank

ADVANCES

IN

CANCERRESEARCH Edited by

GEORGE F. VANDE WOUDE ABL-Basic Research Program NCI-Frederick Cancer Research and Development Center Frederick, Maryland

GEORGE KLEIN Department of Tumor Biology Karolinska lnstitutet Stockholm, Sweden

Volume 59

ACADEMIC PRESS, INC. Harcourt Brace Jovanovlch, Pubtlshers

San Diego New York Boston London Sydney Tokyo Toronto

This book is printed on acid-free paper. @

Copyright 0 1992 by ACADEMIC PRESS, INC. All Rights Reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher.

Academic Press, Inc.

1250 Sixth Avenue, San Diego, California 92101-431 1

United Kingdom Edition published by Academic Press Limited 24-28 Oval Road, London NW17DX

Library of Congress Catalog Number: 52-13360 International Standard Book Number: 0-12-006659-9

PRINTED IN THE UNIlED STATES OF AMEFXA 9 2 9 3 9 4 9 5 % 9 7

QW

9 8 1 6 5 4 3 2 1

CONTENTS

CONTRIBUTORS TO VOLUME 59 ......................................

ix

PREFACE ...............................................................

xi

Lev Zilber. The Personality and the Scientist LEV L. KISSELEV.GARYI . ABELEV. AND FEODOR KISSELJOV I. I1. I11. I v. V. VI . VII . VIII . IX . X. XI . XI1. XI11. XIV.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . First Steps in Science: The First Discovery ......................... The First Victory and the First Arrest . . ........... From Bacteria to Viruses .................................. An Exploit in the Taiga .......................................... Science behind Bars and Barbed Wire ............................. Finding the Family ............................. After the War: Virology and Immunology of Cancer . . . . . . . . . . . . . . . The Last Outburst of Stalin’s Tyranny . .................. ............. Reunification with the World Scientific C Discovery of Pathogenicity of Rous Sarco for Mammals . . . . . In Search of Oncogenic Viruses The Virogenic Concept of the Ori Development of Tumor Immunology ............................. xv. The Last Efforts and the Last Days ............................... ..... .............................. XVI . Legacy . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1 3 5 8 9 13 22 24 27 28 29 30 30 32 33 34 37

The Genetics of Wilms’ Tumor DANIEL A . HABER AND DAVID E . HOUSMAN I . Introduction .................................................... I1 . Histology and Clinical Considerations .............................

V

41 42

vi

CONTENTS

I11. I V. V. VI . VII . VIII .

The Knudson Model . ... Genetic Loci Associated with Wilms’ Tumor ........................ Isolation of the WTI Gene at l l p 1 3 .............................. WTI: Characterization of a Novel Tumor Suppressor Gene ......... Functional Studies and Animal Models ............................ Conclusions ..................................................... References ......................................................

43 46 53 54 61 63 63

p53 Expression in Human Breast Cancer

ADRIANL . HARRIS I. I1. I11. IV. V. VI . VII . VIII . IX . X. XI . XI1. XI11. XIV.

xv.

XVI . XVII . XVIII . XIX .

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Discovery of p53 .... ... ......... Dominant Transforming Oncogene ......................... Recessive Oncogene . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Dominant Negative Function and Gain in Function Mutants . . . . . Normal Function and Regulation of p53 ........................... Mutations in p53 and Interactions with Viral Proteins . . . . . . . . . . . . . . Loss of Heterozygosity and p53 Mutations ............ Different Functionai Mutations and Mutatio Methods to Assay p53 in Human Cancer . Studies in Breast Cancer ......................................... Loss of Heterozygosity Immunochemistry . . . . . . . . . . . . . Mutations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Familial Breast Cancer and Li Fra Expression of p53 with Other On Prognosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . In Situ Lesions . . . . . . ............................... Therapeutic Possibilities .......................................... References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

69 69 70 70 70 71 72 74

77

80

82 83 84 85

c-erbA: Protooncogene or Growth Suppressor Gene?

KLAUSDAMM I . Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I1. Avian Erythroblastosis Virus as a Model for Cooperativity of Oncogenes ...................................................... I11. The Protooncogene c-erbA Encodes a Thyroid Hormone Receptor ... IV. Multiple c-erbA Loci ............................................. V. Structural Differences between v-erbA and c-erbA . . . . . . . . . . . . . . . . . . . VI . Functional Properties of the ErbA Proteins ........................ VII . Mutations Affecting the Biological Activity of v-erbA ................

89

90 93 94 96

97 102

CONTENTS

VIII. IX. X. XI.

c-ErbA Regulation of Erythroid Differentiation and Gene Expression . c-erbA: Protooncogene or Growth Suppressor Gene? ................ Mutations Affecting c-erbA Function ............................... Current Concepts and Open Questions ............................ References ......................................................

vii 104 105 106

108 109

The FGF Family of Growth Factors and Oncogenes

CLAUDIO BASILICO AND DAVID MOSCATELLI Introduction .................................................... Protein Structure ........................................ ...................... The FGF Genes and Their Expression . ........... FGF Receptors ............................ Interaction with Extracellular Matrix ........ ........... Biological Function ...... ...................... ...................... VII. Oncogenic Potential . . . . . . . . . . . .......... VIII. Involvement of FGFs in Tumors . . . . . . . . . . . . ........... IX. Concluding Remarks ...................... References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

I.

11. 111. I v. V. VI.

115 117 126 132 138 140 144 149 155 156

Hepatitis 6 Viruses and Hepatocellular Carcinoma

MARIEANNICKBUENDIA I. .......... 11. ical and Immunological Aspects . . . . . . . . . . . . . . . . . 111. Pathogenicity of Hepadnaviruses: Striking Similarities and Obvious Differences ..... ........................................ I v. Hepadnaviru V. Potential Oncogenic Properties of Viral Proteins ................... VI. Integrated State of Viral DNA in Chronic Infections and Hepatocellular Carcinoma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VII. Genetic Alterations in HBV-Related Hepatocellular Carcinoma . . . . . . VIII. Conclusions ........................... References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

167 170 179

185 195 199

208 211

Cytotoxic T Lymphocytes: Specificity, Surveillance, and Escape ANDREWMCMICHAEL 1. Introduction .................................................... 11. The Molecular Basis of Peptide MHC Association ..................

227 228

...

Vlll

CONTENTS

111. HLA-A2 Interaction with Influenza Matrix Peptide(58-66)

IV. Antigen Processing .............................................. V. CTL Function and Escape from CTL Recognition .................. VI . Conclusions ..................................................... References ......................................................

229 233 235 240 241

Cancer Immunotherapy: Are the Results Discouraging? Can They Be Improved? ELI KEDARAND EVA KLEIN I . Introduction .................................................... I1 . Critical Factors in Cancer Immunotherapy ......................... 111. Current Immunotherapy Strategies ............................... IV. Attempts to Improve Cancer Immunotherapy ..................... V. Conclusions ...................................................... References ......................................................

245 248 255 282 292 294

INDEX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

323

CONTRIBUTORS TO VOLUME 59 Numbers in parentheses indicate the pages on which the authors' contributions begin.

GARYI. ABELEV, Research Institute for Carcinogenesis, Russian Oncologacal Scientific Center, Moscow, Russia (1) CLAUDIO BASILICO, Department of Microbiology, New York University School of Medicine, New York, New York 10016 (1 15) MARIEANNICK BUENDIA, Dipartement des Ritrovirus, Unite' de Recombinaison et Expression Ginitigue, INSERM UI 63, Institut Pasteur, 75724 Paris Cedex 15, France (167) KLAUS DAMM,Gene Expression Laboratory, The Salk Institute for Biological Studies, LaJolla, Calqornia 9203 7, and Department of Neuroendocrinology, Max-Planck-Institutefor Psychiatry, 8000 Munich 40, Germany (89) DANIELA. HABER,Center for Cancer Research, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, and Massachusetts General Hospital Cancer Center, Charlestown, Massachusetts 02129 (4 1) ADRIAN L. HARRIS, Institute of Molecular Medicine, John Radcliffe Hospital, University of Oxford, Oxford OX3 9DU, England (69) DAVIDE. HOUSMAN, Center for Cancer Research, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139 (41) ELI KEDAR,The Lautenberg Center for General and Tumor Immunology, The Hebrew Universily-Hadassah Medical School, Jerusalem 91010, Israel (245)

LEVL. KISSELEV,Engelhardt Institute of Molecular Biology, Academy of Sciences, 117984 Moscow, Russia (1) FEODOR KISSELJOV,Research Institute f o r Carcinogenesis, Russian Oncologzcal Scientific Center, Moscow, Russia (1) EVA KLEIN, Department of Tumor Biology, Karolinska Institute, Stockholm S-104 O I , Sweden (245) ANDREW MCMICHAEL, Institute of Molecular Medicine, John Radcliffe Hospital, University of Oxford, Oxford OX3 SOU, England (227) DAVIDMOSCATELLI, Department of Cell Biology, New York University School of Medicine, New York, New York I0016 (1 15)

This Page Intentionally Left Blank

PREFACE FOUNDATIONS IN CANCER RESEARCH

Many of the young molecular biologists active in cancer research today view the premolecular era of cancer biology with a certain skepticism, if they know anything about it at all; many of them don’t. Those who do rarely know enough to understand the historical significance of major contributions. In addition, they may easily overlook some of the still unexplained and therefore persistingly modern consequences of the earlier studies. In order to provide some historical perspective, Advances in Cancer Research will initiate a new feature, with the intention of reviewing some early studies that made a major impact on the field. Many changes in methodology or in thinking can be traced to the activities of one or a small number of scientists. This will be reflected by the biographical character of some of the articles. This new feature is initiated in this volume, with an atypical article about a major Russian figure in cancer immunology and virology, Lev Alexandrovich Zilber. He and the school he created have played a seminal role in the development of these disciplines in the former Soviet Union. But in addition to Zilber’s remarkable scientific stature, the article written by his two sons, the molecular biologist Lev and the cancer biologist Feodor Kisselev, and by his most eminent disciple, Gary Abelev, also bring the terrible fate of many Russian scientists into focus. After having collected totally new information on vector-borne encephalitis virus, through strenuous and highly hazardous expeditions to the Far East, Zilber was arrested on fabricated charges and was interned for several years in various labor camps of the GULag. Trying to force him to “confess” that he had committed a large number of criminal acts, including the absurd charge that he wanted to spread dangerous viruses over the Soviet Union, his tormentors beat him severely and broke his ribs. But Zilber never signed. When I first met Lev Zilber in the early 1960s, he was out of prison xi

xii

PREFACE

and fully rehabilitated since the end of the Second World War. Nevertheless, and in spite of his widely recognized leadership in the field, he was still kept under strict supervision and his movements were checked and controlled even within the Soviet Union. He was cheerful and full of enthusiasm, bursting with new ideas and future plans. I asked him about his GULag years. He brushed the question aside with a big smile and the comment: “Oh well, those were terrible times.” He did not feel it sufficiently important to mention that he had written his first virogenic theor y of cancer in the GULag, using the only available material he could write on. It was cigarette paper which was smuggled out by visitors. The life of Professor Zilber is a story of courage, resilience, and an unshakable belief in the ultimate power of rationality and justice. Who would have believed, knowing the circumstances, that Zilber will not only survive seven years of imprisonment and heavy abuse, but emerge unbroken to become the dynamic leader of the best research group in experimental oncology in the Soviet Union. Still fewer-perhaps nobody-would have believed that the paranoic, irrational, cruelly inhumane system that tortured and imprisoned him and continued to SUSpect, supervise, and infringe upon his free movement even after his rehabilitation, would collapse like a house of cards. But it did. GEORGE F. VANDEWOUDE GEORGE KLEIN

This Page Intentionally Left Blank

Lev Zilber (1 894-1 966)

LEV ZILBER, THE PERSONALITY AND THE SCIENTIST Lev L. Kisselev,* Gary I. Abelev,t and Feodor Kisseljovt ‘Engelhardt Institute of Molecular Biology, Academy of Sciences, Moscow, Russia Wesearch Institute for Carcinogenesis, Russian Oncological Scientific Center, Moscow, Russia

I. Introduction 11. First Steps in Science: T h e First Discovery 111. T h e First Victory and the First Arrest

IV. V. VI. VII. VIII. IX. X. XI. XII. XIII. XIV. XV. XVI.

From Bacteria to Viruses An Exploit in the Taiga Science behind Bars and Barbed Wire Finding the Family After the War: Virology and Immunology of Cancer T h e Last Outburst of Stalin’s Tyranny Reunification with the World Scientific Community Discovery of Pathogenicity o f Rous Sarcoma Virus for Mammals In Search of Oncogenic Viruses T h e Virogenetic Concept of the Origin of Tumors Development of Tumor Imnlunology T h e Last Efforts and the Last Days Legacy References

I. Introduction Lev Zilber (in German, Silber) became known to the world scientific community after Khruschev’s “thawing” of the mid- 1950s and remained prominent until his sudden death on November 10, 1966. During these years he made numerous friends in the Old and the New World, and tremendously surprised them with his bright and brave ideas, his talent as a polemicist, and the charm of his personality. How could such a man survive during that time? How could he be a most prominent scholar if we recall that for almost 30 years he lived under Stalin’s tyranny? Zilber has appeared in the world arena as a cancer researcher, as one of the founders of cancer immunology, the inventor of the virogenetic concept of tumor origin, not known at that time in the West. However, the scientists who knew Zilber and appreciated his essential contribution

1 ADVANCES I N CANCER RESEARCH, VOL. 59

English translation copyright 0 1992 by Academic Press Inc. All rights of reproduction in any form reserved.

2

LEV L. KISSELEV ET AL.

to experimental oncology did not suspect either his previous discoveries or his tragic fate. This essay is the first attempt to familiarize the English-speaking reader with the contribution made by this extraordinary man to science. At the same time it is the story of his life, since science and his biography are indissoluble and cannot be understood if taken separately. Realizing the responsibility we are taking, we have tried to adhere closely to the facts, so as to avoid partiality for the drumatis permme of this essay. Therefore you will find here many quotations and references to the documents and evidence of eye witnesses. We do not, however, conceal our admiration for this man; it would be unnatural to d o so, as this chapter has been written by a Zilber disciple and successor in the Department of Cancer Virology and Immunology (G.A.) and by his t w o sons, who are fully (F.K.) or partially (L.K.) involved in their father’s domain-tumor biology. Lev Zilber was born on March 15 (28), 1894, in the Medved’ village of the Novgorod guberniya (province) in Russia. His place of birth is far from being evidence of his peasant origin: his father was a conductor, a bandmaster of a brass band, and by the time of his eldest son’s birth (he had two daughters and four sons) his regiment was quartered close to that village. Lev studied at the First Pskov Gymnasium, a century old at that time. Other graduates of that gymnasium were his younger brother Veniamin, the well-known writer Kaverin, his close friend A. Letavet (hygienist, member of the USSR Academy of Medical Sciences, and renowned mountain climber), the eminent physicist and academician I. Kikoin, and his closest friend, the famous writer and literary critic Yu. Tynianov. It was a highly moral and intellectual environment, which probably played an important role in the formation of his personality. In 1912 Lev left Pskov for St. Petersburg University to study at the Department of Natural Sciences of the Faculty of Physics and Mathematics, despite his father, who dreamed of making a musician of his son and who had trained him to play the violin. That training helped him in his life-during hungry and penniless student years Lev played a violin in a cafe for food. His younger brother Alexander, a future composer, accompanied him on the piano. In 1915 Lev moved to the Faculty of Medicine of Moscow University, having been granted the right to continue his training at the Department of Natural Sciences by correspondence. In 1917 he graduated from St. Petersburg University with a bachelor’s degree in natural sciences, and in 1919 from the Faculty of Medicine in Moscow, receiving the diploma of a physician.

LEV ZILBER, T H E PERSONALITY AND T H E SCIENTIST

3

T h e in-depth biological and medical training he had received at the two best universities in Russia shaped his range of interests and formed his ability to work fruitfully at the borderline of biology and medicine. II. First Steps in Science: The First Discovery During the civil war following his graduation from university, Zilber made a military medical career before taking up research. In 1920 he began work close to a big southern Russian town, Rostov-on-Don, as chief of the medical unit of the division, almost a general’s position. However, as soon as he had a chance to demobilize he did, changing his high military position for the very modest one of a laboratory assistant at the bacteriological laboratory of the medical unit of the Front. He dreamed of joining the Department of Microbiology at Rostov University, where Professor V. Barykin, an eminent bacteriologist, worked. The hospital, to which the laboratory was attached, had many spotted fever (louse-borne typhus) patients whom he attempted to treat. He injected patients subcutaneously with their own serum, preheated to inactivate the typhus agent. Patients began to feel better. Zilber was inspired by the outcome, which laid the basis for his first research paper (Zilber, 192 1). He reported the results of treatment to the Military Medical Commission of the Front, chaired by V. A. Barykin. But instead of the enthusiastic reception that he anticipated, he experienced a complete fiasco, since he did not offer either in-depth theoretical substantiation for his method or control tests. Lev argued, trying to prove he was right, but alas. However, Barykin liked him and invited Lev to work at his department. Unfortunately, a severe case of spotted fever ruined this most desirable prospect (see Zilber, 1968). It was only after recovery and in Moscow, where Barykin had organized the Institute for Microbiology of the Narkomzdrav (People’sHealth Commissariat), that Lev began to work at Barykin’s laboratory from the end of 192 1. The years 1921- 1928 marked Zilber’s rapid development into a mature scientist. His works of that period were devoted to so-called reactions of “paraimmunity,” in which an immune response was spearheaded against “foreign” infection rather than a causal agent. He studied the immunogenic nature of metals and demonstrated that iron and gold acquire antigenicity in a colloidal state. His other works were devoted to complement, showing that any suitable colloids, not only an antigen-antibody complex, may fix complement (see Zilber, 1928). His most interesting study of that period was of the hereditary transformation of Proteus vulgar& into a variant that is agglutinated by the serum of the guinea pig infected by louse-borne typhus virus. These

4

LEV L. KISSELEV ET AL.

findings were first published (Zilber, 1923) in the German language, long before Griffith’s widely known, classical experiments on the transformation of pneumococci (Griffith, 1928), but at that time Zilber’s study remained unnoticed. However, it was extremely dear to the author, evidence of which is that his last publication in this series was 26 years after the first one (Zilber and Korshunova, 1949), a case unique for Zilber, who did not like to work too long on one subject. Zilber reported the results of his works (1923-1927) in Vienna at the Congress of Microbiology. He stated in his paper that persistent conservation of new properties for 5 years points to their hereditary nature. The strain described by Zilber was last seen in 1941. It preserved its altered properties, but was lost during the war. Zilber’s findings were confirmed (Minervin and Sotskaya, 1935). Before the war Zilber joined 0. Korshunova to launch a large-scale study (published only in 1949 because of his arrests) in which he fully confirmed the initial findings of 1923, this time with abundant material and at a higher methodological level. He concluded that “those experiments showed the possibility of experimentally changing the antigen structure of Protew. Apparently in using the material that contains large amounts of the louse-borne typhus virus (the lungs of infected mice) a substance can be derived that induced the mutation of the Proteus vulgaris to the louse-borne typhoid one” (Zilber and Korshunova, 1949). It is quite evident that Zilber interpreted his experiments at that time in the light of the study by Avery, MacLeod, and McCarty (1944) of DNA as a transforming factor of pneumococci and gave up his initial interpretation of 1923-1928 (Zilber, 1928) originating from the theory of paraagglutination. Regrettably, the serological transformation of Proteus, described by Zilber 5 years before Griffith, has not been studied (as far as we know) with pure DNA preparations, using modern immunological and molecular biological techniques. Could it be that a young scholar, reading these lines, would be able to bridge this gap? We failed to find any recollections of the mid-1920s in Zilber’s memoirs and therefore our attempts to interpret his motives and concepts are inevitably subjective. We may only assume that his headlong progress was accompanied by discomfort and dissatisfaction. The more involved he was in the physicochemical interpretation of immunity, the more he contradicted the genuine situation in immunology, and the more artificial was the stand that he had to protect as a proponent and disciple of Barykin. He was looking for arguments in the exceptions to the rules, which was alien to Zilber, who usually ignored exceptions rather than relied on them. This concept led nowhere. The break was

LEV ZILBER, T H E PERSONALITY AND T H E SCIENTIST

5

inevitable and when in 1929 Lev Zilber was invited to move to Baku, he did not hesitate a minute, as he later put it.

Ill. The First Victory and the First Arrest In 1929 Zilber was elected head of the Department of Microbiology in Baku, and became director of the Institute for Microbiology. Zilber could not rest satisfied only with experiments. His romantic and heroic personality thrived on struggle and the overcoming of difficulties. Fate gave him all of that in abundance. In the winter of 1930-1931 plague was reported in Nagorno-Karabakh, not far from the border with Iran, in the settlement of Gadrut and adjoining areas. Zilber led the medical team, sent to the focus of the plague. Nagorno-Karabakh is a mountain area, which at that time did not have convenient communication facilities with Baku. The population of Nagorno-Karabakh consisted of peasants, mainly Armenians. At that time it was an absolutely wild, uncivilized region. None of the Armenians spoke Russian. Medieval traditions reigned in the area. Zilber left beautifully written memoirs describing the plague outbreak, which were published a month after his death (Zilber, 1966). These memoirs laid the foundations for our essay. Zilber was neither a specialist in plague, nor a practical physician, nor an epidemiologist. That is why his readiness to help may be explained by his drive to take risks, the wish to test himself in an extreme situation, and, undoubtedly, to help people in trouble, rather than by his professional qualifications. He read about plague on his way to Gadrut in a thick German handbook on plague, available in Baku at that time, and in two volumes of the report of the Manchuria expedition of 1910, headed by the prominent Russian epidemiologist D. Zabolotny, who had treated the young Zilber kindly and presented the volumes to him previously. After a walk of many hours from the railway station to Gadrut, the team finally reached its destination point. The situation was dramatic: local doctors had failed to diagnose plague at the beginning, having at first confused it with pneumonia, and therefore lost time stopping the spread of the disease. A correct diagnosis was made by a young military doctor, Lev Margolin, who sent the cable to Baku. His fate was tragic. This is how Lev Zilber described it: Suddenly there was a sharp knock on the door and without waiting for permission a young military doctor entered the room. He looked around and with unsteady steps headed for the desk. “I am Margolin, I contracted plague,” he whispered and fell down. I grasped a gas mask from the wall, put it on and bent over Margolin, who was

6

LEV L. KISSELEV E T AL.

lying on the floor. He was unconscious. His face was dark, his eyes were retracted, he had a rapid pulse, and sterterous and rapid respiration. Orderlies put him on a stretcher, I placed gauze on his face and he was carried away to the plague hut. Margolin was infused with the maximum of antiplague serum, aides were at his bed round the clock, doing their utmost to save him. He died in forty hours.

T h e conditions of work were not only difficult, they were lethal. When Zilber infected guinea pigs with the plague culture, as should have been done to isolate the causal agent, he and his assistant had only one pair of rubber gloves. He gave the gloves to his colleague and worked with bare hands, placing them from time to time into a disinfecting solution. Further events became even more tense and we cannot describe them better than was done by Zilber himself Once late at night I was visited by an NKVD (secret police) man. I lived in a small rooni not far from the school. ‘‘I must have a serious talk with you, professor,” he said, taking the only chair in the room at my invitation. “Things are looking bad. We have all the proof that all this was done by the saboteurs from abroad. They dissect plague corpses, remove the heart and the liver, and cut them into small pieces, and by this way spread the infection. This information is absolutely accurate,” he added, having noticed mistrust in my face. “You know, comrade,” I said, “a plague microbe could be easily cultivated on nutrient medium. In several days you could grow enough of these microbes in a laboratory to infect hundreds of thousands of people. Why should saboteurs cut out these organs from corpses when they could have plague cultures?” “We should not discuss that now, we have to see whether the buried corpses are intact or not. Could you right now organize exhumation and examine all the recently buried corpses? We will have to do it at night because local people would consider it a defilement of graves and there might be disturbances.” I responded that everything would be ready in an hour, but that we did not have spades and crowbars. We made an arrangement that in an hour he would come to the school building with five armed soldiers (to protect us, just in case, as he put it) and that he would bring spades and other instruments. All this seemed to me to be a fantasy, since the saboteurs who dissected corpses and cut out the heart and liver would have inevitably contracted plague, providing they were not themselves bacteriologists or doctors who knew how to protect themselves from infection. No, it could not be the way he told me! It was quiet and dark at the cemetary. An oil torch dimly illuminated a small space. We covered it so that people in the settlement would not see the light at the cemetary. The grave was shallow and soon we saw the lid ofthe coffin. Inside was a woman in her forties, and there were no signs of dissection. “Seems to be all right,” I told the NKVD man. “Unbutton her blouse and examine the chest and the abdomen,” he said dryly and sharply. We unbuttoned the blouse and cut the skirt. The skinny body, already decompos-

LEV ZILBER, T H E PERSONALITY A N D T H E SCIENTIST

7

ing, was not dissected. We were almost nauseated from the unbearable smell coming from the corpse. I stepped aside to take a breath of fresh air. The moon illuminated strange things happening: some ghosts clad in white with rubber gloves and boots, bending over the grave, were lowering the lid of the coffin onto it. We could not restrain ourselves from exclamation when the third grave was opened. The head was separated from the body and was lying by its side. The clothes were cut, the chest and the abdomen were opened, and there was no heart or liver. Three out of ten graves opened that night had corpses with cut-off heads, without the heart and the liver. It was horrifying not only because it was unusual, but mainly because of the consequences. A plague microbe, dried in tissues, remains alive for years. If pieces of plague-infected organs were kept by the local people, then how can we find them to save people? The history of plague epidemics knew nothing of the kind. I did not sleep a wink that night (Zilber, 1966).

This riddle was solved by chance. A schoolmaster told Zilber that the local people believed that if the first member of the family dies and then other members of the family begin to die, it means that the first one is still alive and pulls his kin into the grave. To prevent it, they have to open the grave, cut off the head, take the heart and the liver, cut them into pieces, and give them to all members of the family so that they may eat them. This story of the schoolmaster explained the situation. There were no subversive acts. It was an ancient and dreadful tradition. It became clear that the only way out was first to burn the corpses to liquidate the source of infection, and second to take away the cut pieces of organs that were hidden in houses. Zilber managed to d o it. People were driven out of their houses, the infected were separated from the healthy, the houses were disinfected, and the outbreak of plague was liquidated. When Zilber returned to Baku, having brought with him cultures of plague bacilli (that was a must in any epidemic, for further analysis at a laboratory), the local authorities elected Zilber an alternate member of the AzTZIK (the highest body of power in Azerbaijan) and intended to award him with the Order of the Red Banner. However, instead of these honors Lev found himself behind bars, where he spent several months. He was lucky to be let free and in 1931 he returned to Moscow. What was the reason for making a criminal out of the hero who conquered plague in two weeks? Local bodies of the NKVD alleged that Zilber concealed the “subversive” origin of the plague epidemics and brought the plague culture to Baku to infect its residents. Luckily the absurdity of these accusations was proved and Zilber was freed. However, he would not stay in Baku after all those developments.

8

LEV L. KISSELEV ET AL.

T h e “ore” operation (the word “plague” was prohibited from use in cables and all documents; instead, the word “ore” was recommended) highlighted like a searchlight much in the Zilber personality: brilliant organizing abilities, manifested in most unusual situations, courage in the face of lethal danger and in taking responsibility, the ability to act fast and resolutely, and, finally, a specific “epidemiological gift.” It should be mentioned that this epidemiological episode was not the only one in Zilber’s life: in 1933 he was one of the main participants in eradicating an outbreak of smallpox in Kazakhstan, where he organized the manufacture of smallpox detritus on site. Somewhat later he was in charge of liquidating a large-scale outbreak of typhoid. Mention should be made of the fact that he identified the source of infection in two days. The sewer system of an infection hospital was, during repairs, connected by accident with the water pipeline to one of the districts of the city. These events, to say nothing of the encephalitis campaign, which will be described later, would be enough to rank Zilber among eminent epidemiologists; however, it was only his peculiar “hobby.” IV. From Bacteria to Viruses Having returned to Moscow, Zilber headed the Department of Microbiology at the Institute for Advanced Training of Physicians and worked at the Mechnikov Institute for Microbiology, where he continued to study the problems that were raised back in the Barykin period, but this time on a new, natural basis, finally admitting the existence of antibodies. He was mainly involved in solving two problems (1931-1934)-the preparation of vaccines of a new type and the interaction of rickettsia and viruses with other microorganisms. T h e basis for the first problem was his previous observation that the threshold of thermodenaturation of protein antigens can be substantially increased in the presence of antidenaturants, for instance, sucrose, whereas bacteria in the meantime were destroyed. That is how sucrose vaccines were manufactured, some of which were put to use by Zilber (see Zilber and Wostruchova, 1931, 1932; Zilber et al., 1933). Zilber’s study of the interaction between viruses and microbes demonstrated his willingness to take on a completely ignored problem, since at that time the information on viruses was extremely scarce. Zilber, together with his colleagues, intensively studied the interaction of vaccinia virus with yeast (Zilber and Wostruchova, 1933, 1934; Zilber and Dosser, 1934). At first he identified the ability of yeast to adsorb the virus, apparently rather selectively, and then derived yeast cultures in which, as was then viewed by researchers, the virus penetrated the yeast cells and

LEV ZILBER, T H E PERSONALITY A N D THE SCIENTIST

9

propagated there. Zilber called this phenomenon ~Zlobiophoria.Generally speaking, from the contemporary point of view there are no theoretical prohibitions for the replication of the DNA-containing vaccinia virus in the unicellular eukaryote yeast, although it is not clear in which way the virus penetrates through the extremely dense and almost impenetrable outer yeast membrane. We can make many assumptions: the culture could have consisted of singular cells with a destroyed outer membrane, allowing the virus to penetrate and replicate; the vaccinia virus may have had the ability to penetrate through the outer membrane with the help of a specific mechanism; and, finally, we must admit that it might not have been the whole virus that penetrated inside, but its DNA or nucleoprotein. It is quite clear that the problem raised by Zilber, in a very precise form, has not been solved even after 60 years; allobiophoria lives in wait of its researchers. These were the first studies that Zilber conducted in virology, the area which at that time most closely tallied with his character. Danger, risk, big problems, and the promise of outstanding discoveries-that was the world of Zilber-and he entered that world full of energy, ardor, and enthusiasm. V. An Exploit in the Taiga In 1935 Zilber organized the General Viral Laboratory, the first scientific virological unit in the Soviet Union. Among the most outstanding achievements of that laboratory and Russian virology per se, which we recall with gratitude and of which we are proud, is the discovery of viral tick-borne encephalitis (virus is the causal agent of this disease of the nervous system) and its vector. The history of this discovery is completely unknown to the world of science; therefore, we shall try to briefly recall it. From the beginning of the 1930s in the Soviet Union, in a number of areas of the far east, physicians registered severe disease(s) that often resulted in the death of patients: the central nervous system was affected. The disease was not studied at all and was mainly classified by local physicians as “toxic flu.” In 1935 Panov, who worked in the far east, established that the disease was an encephalitis; he considered it to be a Japanese encephalitis, already known at that time. In 1936 physicians of the local Pasteur Laboratory infected mice with brain emulsions of people who had died of the encephalitis, trying to isolate the causal agent; however, that attempt was unsuccessful. It became clear that local medics were unable to cope with this

10

LEV L. KISSELEV ET AL.

unknown disease and that it was necessary to organize an expedition. Several years afterward Zilber recalled, When the Narkomzdrav (Ministry of Health) formed the expedition it wanted to form a complex team which would have ten professors. I refused point blank to take part in the expedition and said that they should make their choice-either I take all the responsibility and form the expedition or do as they please. After a long dispute I was refused the pleasure. The Military-Sanitary Department was vitally interested in encephalitis control. Try to recall that time-that was the period not only of a largescale assimilation of the Far East, but also the time when we were forced to keep big army units in the taiga. Therefore the Military-Sanitary Department appealed to the Defense Minister and on his direct order I was appointed the head of the expedition. 1 was given the right to take my own choice and to work the way we thought it necessary. I employed only young people and did so on purpose. N o doubt 1 warned them of the dangers and difficulties and what not, but young people had a huge advantage-they were not ridden by the old concepts with respect to this disease. Barykin with a small expedition was in the taiga before us, but failed to do anything; local neuropathologists stated that the disease was Japanese sunimer encephalitis and even our official documents read when we were heading for the far east that we were to study summer encephalitis. I was not convinced of it and we made three research plans. The first one in case it was really summer encephalitis, the second plan if it was any other encephalitis, and finally, the third plan, in case it was not encephalitis at all. These plans had been worked out in detail. From the very beginning I wanted them to work simultaneously. We did it in such a way that two groups of my colleagues were divided into two teams that were doing the same things, to be confident of the final results, and to reduce the time of research. This system, in those concrete conditions when we had to solve the problem fast, had fully justified itself (Zilber, 1969). During the first trip to the taiga on May 19, 1937 to the northern area of the disease outbreak, I came across the facts that made me doubt the concept of the epidemiology of the disease. In a small hospital of a timber-cutting company in the taiga I found medical reports for the last three years. A thorough examination of these reports showed that the encephalitis was contracted mainly in the spring and only by those people who work in the taiga and who very often do not have any contact with each other. This information disagreed with the theory of contact or droplet infection. At the same taiga hospital I found a woman with en. epiAtis on May 19 who had fallen ill on May 4 and was already recovering by the time of my visit. She was the first patient of that season and the establishment of the source of her infection could be of paramount significance for further studies. The patient was a housewife who had not left the taiga settlement for two years and did not have any contact with any other patients or their families. For a long time I failed to guess the origin of the disease. It refuted both the contact theory and summer season nature of the disease, and the assumption of the transfer of the disease by mosquitoes, since there were no mosquitoes at all at that time of the year. After a long cross-examination, the woman casually recalled that 10 to 14 days before the disease she collected last year’s cedar nuts; when she returned home she found ticks feeding on her. This single fact, which could explain her disease, attracted my attention (Zilber, 1945). I took a plane to Vladivostok to find out something about ticks (ofwhich I knew nothing at that time). . . . There I found the work of one veterinarian that showed that the curve depicting the bites of cows by ticks, fully coincided with the curve of an

LEV ZILBER, T H E PERSONALITY AND T H E SCIENTIST

11

increase of the number of cases of the disease in people, but delayed by two weeks. It became clear to me that that was an incubation period (Zilber, 1969). The probability of the transfer of the disease in that way was so evident to me that by the end of May I sent a number of doctors, including members of the expedition to the taiga, to those teams working in the taiga to warn them of the danger of being bitten by a tick. Later it turned out that only one man fell ill of those warned, although in the past these were the most affected groups. Parallel with the collection of epidemiological information, we experimentally tested the tick theory. Appropriate experiments, which I requested Chumakov to conduct, were a complete success and he proved the possibility of an experimental transfer of the disease by ticks. At the same time we isolated virus out of spontaneously infected ticks. These and other works, particularly the studies by Academician Pavlovsky and his colleagues, fully confirmed my theory of the transfer of the disease by ticks and at present nobody doubts that ticks are vectors of encephalitis (Zilber, 1945).

The tick theory put forward by Zilber, substantiated by his colleagues and himself, amazes us 50 years after these events by the circumstances related with them. T h e theory appeared on May 19, two days after the beginning of work on the focus of the disease and twenty days afterward Zilber took responsibility “to propose at a specially convened conference of local health authorities on June 10 to radically change all the measures to control these diseases, by concentrating the main attention on antitick prophylaxis” (Zilber, 1945). It is only the man who combines many qualities-scientific intuition, resoluteness of action, a feeling of responsibility, quickness of thought, inner conviction, and humanitywho could in such a fantastically short period of time traverse the path from a scientific hypothesis that arose literally from nowhere to energetic measures aimed at saving people in the taiga. No doubt it was the working style and character of Zilber, always apt to make rapid generalizations. However, it would be naive to think that only one patient and one guess promoted the success of the expedition. Zilber himself thought that “. . . a huge role in research is played by the preparation of the study. When I think of the role that is played by this preparation I always recall our expedition of 1937. We were equipped with everything, we were only short of tropical monkeys. I asked to get them. And then monkeys, urgently bought in Japan, were sent to the expedition. They were necessary for final experiments” (Zilber, 1969). The tick theory answered the question of the vector of the disease and the paths by which it spread and, in this way, was of huge theoretical and practical significance. However, it did not answer the question of the nature of the causal agent of the disease: ticks, in principle, could transfer both bacteria (rickettsia) and viruses. Only thoroughly performed

12

LEV L. KISSELEV ET AL.

experiments (it is expedient to recall that they were conducted in the wild taiga, without proper roads, in wooden houses, and not in sterile boxes) could answer that question. In an effort to be historically accurate we give the floor again to Zilber. . . . the very first lethal cases gave information that helped me and Shubladze in the south and Levkovich and Chumakovl in the north to isolate the causal agent, which turned out to be an ultravirus that was somewhat similar to the viruses of the Japanese and American encephalitis. Somewhat later, strains were isolated by Soloviev. In June and July Shubladze and I experimented on infecting monkeys with the cerebral emulsions of people who had died of encephalitis, and passaged virus that was isolated at that time. These experiments also confirmed the etiological significance of the strains, isolated by us, but the neutralization of those strains with reconvalescent sera was abortive for a long time, which prevented the recognition of the isolated virus as the causal agent of the disease. . . . Only after experiments were staged with sera of the later period of reconvalescence did we obtain distinct positive results and it became evident that we had isolated the causal agent of the disease (Zilber, 1945). In general the whole work on the study of taiga encephalitis was an exploit of our scientists. An exploit became their everyday phenomenon. But still we would like to say a few words about these heroic acts. Floods began right in the middle of their work. The river broke through the dam. The water penetrated into the vivarium with its experimental animals. They had to save them at all costs, to save themselves. All hands to the pump! Working waist deep in water they carried cages with frightened mice and monkeys to dry places. The animals had been saved. Soon Dr. Chumakov fell ill. Despite severe muscle pain and weakness he continued to work, but he was running a high temperature. The first symptoms of the brain disease appeared. Chumakov could not fight the disease anymore. His friends were alarmed, but he reassured them. “Rubbish, everything will be okay,” he would say, “this is my old rheumatism that does not leave me in peace.” But it was not so: he contracted encephalitis. Chumakov willed himself to life. He courageously looked into the eyes of danger and asked his friends only one thing: to bring their common cause to an end (Kassirsky, 1949).

Luckily, Chumakov not only combated the disease and its consequences, but later successfully worked in virology. Anothel- participant in the expedition, W. Soloviev, had also contracted encephalitis, but fortunately in a milder form. Zilber, with the laconic brevity characteristic of him, summed up the results of the expedition: By August 15 the work of the expedition was over. During those three months we established the presence of a new, previously unknown form of encephalitis and isolated its causal agent-29 strains; we established the epidemiology of the disease, its vector, and studied, in the main, the pathological anatomy and histology of the disease. This success was clouded by laboratory infections of researchers: Chumakov, 1 M. P. Chumakov, an eminent virologist, academician, and one of the inventors of the polio vaccine in the Soviet Union.

LEV ZILBER, T H E PERSONALITY AND THE SCIENTIST

13

Soloviev, and Gnevysheva, a laboratory assistant, contracted encephalitis. It is rather difficult to amplify the circumstances in which they contracted the disease. All of them thoroughly followed the rules of work with infectious material. T h e most dangerous experiments, involving nasal infection of monkeys, were personally conducted by myself and Shubladze. It was impossible to know that the virus was extraordinarily infectious. We were the pioneers in that domain and we were the first people on our planet who studied that formerly unknown virus. . . . Of certain importance were the relatively primitive conditions in which the study was performed and great fatigue, resulting from daily twelve-hour work for three months with only one day off. However, I could not keep my fellows away from that tense work: they all worked with exceptional involvement and genuine enthusiasm. Later, lethal cases were registered during the work with o u r virus in Moscow at special virological laboratories, and even in those cases specially developed measures had been taken to prevent infections. These facts make us suspect the highly infectious nature of o u r virus and it is not surprising that our first acquaintance with it resulted in infections. These losses could have been even more substantial (Zilber, 1945).

We would like to complete the story of the taiga exploit of Zilber and his colleagues with the statement of one of its main participants, M. Chumakov, made many years later: “Lev Zilber has all the right to enter the history of medical science first and foremost as the pioneer and the discoverer of the viral origin and tick transfer of tick-borne encephalitis in the far east. T h e discovery and the study of that disease in the USSR, and later in many countries of Europe and Asia, represent an important stage in the development of world virology” (Chumakov, 1985).

VI. Science behind Bars and Barbed Wire It would seem that people who for many days and every hour risked not only their health, but their lives as well, have the right to recognition and gratitude for what they have done. But that was the year 1937 and as a result of an absurd and blasphemous delation Zilber and his t w o close associates, A. Sheboldaeva and T. Safonova, were arrested. Here, we would like to deviate from the chronological order of events to make brief comments on the situation inside the Soviet Union from the mid-1930s to the beginning of 1950. During that time Stalin and his satraps established a dictatorship, a power without limits and without any moral boundaries, based on massive physical, psychological, economical, and other types of terror. The single goal of this state terrorism was to maintain the absolute power within his hands, to keep up the personal, individual totalitarian dictatorship of Stalin. The “Great Terror” of 1937 was against the whole population of the country, since potentially only the people were able to deprive Stalin of his unlimited

14

LEV L. KISSELEV E T AL.

power. All those who had some individual features, who were even slightly above the average unified level, who tried to keep their own views and opinions, were deliberately either “isolated” (put into jail) or shot. Genocide against the whole nation, done by the NKVD, did not require any justification, any facts, documents, or other means. The main method used by NKVD investigators was to put their victims on the rack in an attempt to force them to slander themselves for actions (or even thoughts) never committed by the given person. Selfslandering was taken as a unique and self-consistent proof of guilt; no other evidence was required. Since the tortures were highly sophisticated and extremely cruel practically all prisoners (with very rare exceptions) signed their confessions in the illusive hope that their suicidal action would help them’ to escape further suffering. One of the most widespread ways used by NKVD officers to achieve a prisoner’s self-slandering was to threaten the victims with imprisonment of all the members of their families (mother, father, children, wife, etc.) if they did not sign the confession. It was impossible to seek justice because there was no court. All decisions were made by the so-called troika (a meeting of three carefully preselected NKVD officers), which had no connection with or similarity to a normal court examination. Verdicts of guilty were brought against all those presented before the troika, independently of the essence of their affair. Obviously, for Western readers living in democratic societies it is very difficult (if not impossible) to imagine this system, created by the Communist party headed by Stalin. Evidently, Zilber, even though absolutely not involved in any political activity, but only deeply engaged in the struggle against human disease (not against the regime), was nevertheless dangerous to those in power mostly because of his extraordinary nature, nonconformism, and other qualities. His arrest in 1937 was no doubt inevitable. The first documented story about the taiga expedition, written on the basis of information received from its participants, reads as follows (Sharov, 1963): An investigator asked Tamara Safonova to sign fabricated evidence suggesting the expedition, under the disguise of scientific work, spread Japanese encephalitis. “This is slander, seen with the unaided eye,” responded Safonova, scornfully adding, “and at the same time it is absolutely illiterate.” The investigator shrugged his shoulders. “Correct it, from the scientific point of view, and your term of imprisonment will be short, and you will not destroy your life.” ‘‘I lived my life. . . and a happy one,” answered Safonova. “1 will not give it up. And what did you have?”

LEV ZILBER, T H E PERSONALITY AND T H E SCIENTIST

15

T h e first scientific report (see Zilber, 1971) on the etiology of tickborne encephalitis was published in 1938 in the absence of Zilber and Sheboldaeva, and they were not listed among the authors. Twenty years later, the coauthors of that work wrote a special letter to the journal Voprosy Virwologzi (Volume 3, p. 191, 1957), in which they asked that the historical and scientific truth be restored by including Zilber and Sheboldaeva as authors of that article. Zilber never talked about the period from 1937 to 1939. The following names will not ring any bells for an English-speaking reader-Lefortovo, Lubyanka, Butyrki, and Sukhanovo, but for hundreds of thousands of citizens of our country these names are synonyms of terrible physical and moral suffering, inevitable and dreadful death. Zilber experienced it all to the fullest but did not sign a confession of nonexistent crimes. Once, many years later, he was undergoing an annual medical examination and the doctor, examining his X ray, exclaimed, “Lev Alexandrovich! Your ribs have been broken! And there is nothing in your case report.” “True,” answered Zilber, “before the war I had an automobile accident.” He was very pleased with the way he deceived the young, trustful doctor. In June 1939 Zilber was discharged from prison. We cannot be certain what it was that helped to set him free-the absurdity of the accusations, the energetic and fearless actions of his loyal friends (the microbiologist Z. Ermolyeva, the writer Yu. Tynyanov, and his brother V. Kaverin, also a writer), or a “change over” in the NKVD when the bloodthirsty minister Ezhov was replaced by a butcher, Beriya, who began his activity by setting free a small number of the prisoners (later the majority of them were once again arrested). After his release from prison Zilber published a classical and fundamental work on tick-borne encephalitis (Zilber, 1939), written right after the expedition back in 1937. He also wrote a monograph on encephalitis and gave it to the publishers in December 1939. The book was composed and was to see light the next year, but it did not happen. Luckily the proofs were preserved. One of us (L.K.) remembers the next, third arrest of Zilber in 1940. It happened in the same apartment where the Zilbers live now. Three strangers came early in the morning around four or five o’clock. They did not talk much. L.K. recalls that his father kissed him on the forehead, embraced him, and then lightly pushed him away and looked at him for what seemed a long time. It seemed to L.K. at that time that his father was calm; in any case he did not make any abrupt movements, he did not raise his voice, and it seemed as if nothing particular was

16

LEV L. KISSELEV ET AL.

happening. He was pale but his eyes seemed to be alight. At that time his son was four years old and his wife Valeriya Kisseleva was expecting their second child. A number of the participants in the second and third far east expeditions of 1938- 1939 (Pavlovsky, Smorodintsev, and Petrischeva) were awarded the highest state award of that time-Stalin’s prize of the first degree. Zilber, Sheboldaeva, and Safonova were not among the prize winners. Thus the third arrest after fourteen months of hope. It would seem that there would be no future for him; once again he underwent interrogations and accusations concerning violations of several counts of Article 58, conviction of which would mean the death penalty. Nevertheless, even these unbelievable tribulations were unable to break his will. He was found guilty and sentenced to (only) ten years of imprisonment. This term of imprisonment was minimal at that time. Apparently the logic that Zilber used to defend himself and his refusal to admit himself as guilty influenced the indictment of the “judges,” and he once more avoided the death penalty. After that, fate turned her face to Zilber. He was sent to a camp in the northern European part of the country, at the basin of the Pechora river. He was given a job as a doctor in the camp and thus he avoided the heavy physical work that killed the majority of the prisoners. Life became somewhat easier for him-no beatings, although he was constantly hungry, but he was able to think and help people. It inspired Zilber and he became his former self, a genuine investigator doing research in a place where even thoughts about research should have been impossible. A striking example of this research was the development of a preparative-scale cultivation of yeast on moss extracts. The yeast had a therapeutic effect in cases of pellagra (general profound avitaminosis). This disease was a curse for all prisoners in the north. Zilber improvised, creating a healing preparation out of moss (abundant in the north), and in this way saved the lives of hundreds of prisoners. Zilber received a patent for this technique of manufacturing yeast (the patent is kept in the family archives). It seems unbelievable-an arrested doctor receiving a patent, the application for which was sent from the Pechora camp with the name of the inventor-but it was Stalin’s epoch and you wouldn’t be surprised if it happened. Zilber not only developed this procedure, he convened a conference of doctors and nurses working in other camps and disseminated his acquired experience and new technique. T h e family archives contain many letters of former prisoners whom Zilber supported by his courageous behavior and medical help. Here are

LEV ZILBER, T H E PERSONALITY A N D T H E SCIENTIST

17

excerpts from only two of them. The first letter refers to the period of interrogation and it was written by the late engineer G. Men’shikov,who spent more than 15 years in jails and camps. “I associate the image of Lev Zilber with the concept of honor, humanism, and loyalty to the cause. I became acquainted with Lev Zilber during the most difficult period for him. At that time he was accused on seven counts of Article 58, each of which meant the death penalty for him. . .. I appreciated him simply as a courageous person, I loved him for his optimism, humor, and rare readiness, although hungry himself, to share with his neighbor a piece of herring, the last piece of frozen swede. I . . . helped him during agonizing attacks of asphyxia, bandaged his back, which was bleeding after beatings in Sukhanovka,2 but I . . . never heard a word of complaint from him.” The second letter was written by the late S. Vasyukhnov, Zilber’s jail mate: “I met many people in my life-good, attentive, and responsive people. . . but Lev Zilber was a miracle man . . . . He would always lend his hand in trouble, would always share the last piece of bread and give advice. He will always stay in my heart as man of strong will, as a courageous, strong, beautiful, and unsurpassable person.” And those who managed to survive remained friends of Lev Zilber and were always welcomed as guests in his house. While working in the camp, Zilber did not give up the struggle for his discharge. He appealed several times to the NKVD with a request to revise his case. After the development of the healing yeast preparation he was suddenly summoned to Moscow. Zilber was fairly confident that his case would be revised-he was taken to Moscow without an armed convoy. However, in Moscow Zilber’s hopes were not fulfilled. He was not released, nor was his term of imprisonment reduced . . . They did not return Zilber to the camp, but instead offered him a job in the chemical laboratory of a Moscow jail, a so-called “sharashka.” The “Great Terror” of 1937-1940 saw many scientists in jails and camps. Many of them died by the middle of the war, but some of them managed to survive. T h e authorities invented an intricate form for using their intellect-special laboratories, sections, and bureaus were set up for arrested scientists, aircraft designers, chemists, medics, and engineers. The work of such establishments has been thoroughly described in Solzhenitsyn’s novel “In the First Circle.” Zilber had been sent to such a chemical laboratory attached to one of the Moscow prisons. He was given the assignment of producing alcohol out of moss. And that is the beginning of the happiest (if such a word can One of the most dreadful interrogation prisons in Moscow.

18

LEV L. KISSELEV ET AL.

be used in that situation) period of Zilber’s life in jail. He was given the opportunity to work, and there was no problem with the literature. Although his every movement was under stringent control (a “laboratory assistant” on staff each day reported Zilber’s behavior), Zilber was quite happy. He was behind bars but he was given the opportunity to see his family. However, once again Fate turned her back on him. He found out that his whole family (his wife, two sons, and his wife’s sister) had disappeared in November 1941 and nobody knew anything about it. Everything seemed hopeless, but even this was unable to break Zilber’s will. He had the opportunity to pursue science and it saved him in that dreadful situation. A surprising paradox-a prison, the most stringent control on the part of spies, but on the other hand the opportunity to take up new and interesting scientific problems. It is rather difficult for us to visualize all this but Zilber found the opportunity to study the problems he thought to be important, interesting, and prospective. Having come to virology in the mid-1930s with the experience of having isolating virus of the far east encephalitis, Zilber returned to the idea of an important role of viruses in inducing cancer in humans, which he voiced back at the All-Union Conference on Viruses in 1935. We could hardly say that that idea originated from nowhere. By that time several viruses had been isolated that induced tumors in animals, among them chicken Rous sarcoma virus (Rous, 191 l), mammary tumor virus in mice (Bittner, 1936),and Shope papilloma virus in rabbits (Shope, 1932). It was no doubt clear to Zilber that he had to begin with laboratory animals. But where could he obtain them? Nonetheless Zilber found them-he made an arrangement with prisoners and they caught mice and rats for him for tobacco. Let us cast a look at oncology of the early 1940s. What was known at that time? It was possible to induce neoplasia by carcinogenic substances and by a few viruses, and it was possible to induce tumors in animals by inoculation with cells. In fact, that was all. If a researcher wanted to prove the involvement of a viral agent in tumor induction, then he used cell-free extracts for inoculation, the extracts having been passed through a Zeitz filter. Zilber took that path. Everything that was done by Zilber at that time was later described by him (Zilber, 1945). Very few people even now know the real conditions under which Zilber conducted his experiments. Zilber induced tumors with carcinogenic substances in mice caught by prisoners, and then injected adult mice with cell-free extracts of these tumors. In all cases but one he failed to induce the formation of a tumor, whereas homogenates that did

LEV ZILBER, THE PERSONALITY AND T H E SCIENTIST

19

not pass through the Zeitz filter induced tumors in both young and adult mice. Two exceptions served as a turning point in these studies. T h e first exception occurred when a pregnant rat, selected randomly for an experiment, was squeezed to death in a cage door accident. Autopsy revealed a small tumor node (a “young” tumor); its extract was used in the experiment and induced tumor in the recipient. The second exception occurred when an extract was prepared from a mouse tumor, which sharply changed its virulence. This tumor was also, to a certain extent, a “young” tumor. On the basis of these singular, occasional, and, as it would seem, insignificant observations, Zilber arrived at a very provocative assumption-virus is identified in “young” tumors, since it only triggers the neoplastic transformation; later there is no need for it. This hypothesis has been tested by large-scale experiments on mice and rabbits. Carcinogens are applied to animals and the nodes that appear are disrupted, filtered through a Zeitz filter under a nitrogen atmosphere, and then infused into recipient animals sensitized with the same carcinogen. In these later experiments a positive effect was observed in fifteen percent of the cases. Quite convincing results! These data gave a new impetus to a new concept of the role of viruses in the induction of tumors. Zilber nurtured the idea that all tumors are of viral origin and that the virus is only an inducing agent. He felt that the results of his first experiments and later theoretical generalizations were of paramount significance and that the scientific community should be informed of them. How could he do it in jail? In his memoirs, written in 1964 but published only recently, Zilber (1989) described it this way. He appealed to the jail authorities and requested permission to talk to them: “I have read your application and did not understand what you wanted. You have been sentence by very serious counts and I cannot understand, in general, why you were given the opportunity to work at a laboratory.” “Citizen commissar, the crux of the matter is not myself; it is the results of my work. Please try to understand that it is cancer that matters.” “The war is on now and nobody is interested in your cancer. And what in particular have you done? Have you learned how to treat cancer?” “The war will be over, but people are dying and will die of cancer although in smaller numbers than at war. I have not learned how to treat cancer, but my experiments show that chemical substances that induce cancer, in fact only promote the genuine cause-the virus-to manifest its effect, like cold promotes the appearence of tuberculosis. Virus only triggers the disease, it transforms in a hereditary way a normal cell into a tumor cell and later the tumor grows without its [the virus] involvement.”

20

LEV L. KISSELEV ET AL.

“Write in detail what you have done and we shall send your report to the Narkomzdrav [Ministry of Health].” ‘‘I will not do it, citizen commissar. . . . In 1937, when my colleagues and I discovered a virus causing tick-borne encephalitis I was arrested several months after that, and my detailed reports to the Narkomzdrav were used in publications by those people who tried to make my discovery their own. Now I have information that is of even greater significance.” “Does it mean that you place your own interests, your scientific ambition, above the interests of Soviet science?” “No, citizen commissar, that is not the way to put it. . . . I ask to publish the results of my study not under my name, but under a fictitious one so that Soviet researchers would be able to use my findings and at the same time nobody would claim these data to be their own.” “How do you want me to publish this work of yours, in the Izz~estzjaor in the Pravda?3 Take him back to the cell.”

This episode discloses another feature of Zilber as a scholar, which was very typical of him-hatred of scientific slavery, the very idea of which was unbearable for him. Zilber was always irreproachable in scientific ethics, he was the author or coauthor only of those publications to which he had essentially contributed. He always popularized and pushed ahead the papers of his colleagues and associates. So, the first attempt was a failure. However, Zilber did not give up and decided to take action-to smuggle his manuscript outside. But how could he do it? Let us turn to his memoirs again (Zilber, 1989). I do not know from where or why we had in our laboratory a cigarette paper of very high quality, on which one could write very small letters with a pencil. . . . Two or even three sheets of this paper could be folded in such a way that it would not be bigger than an average button. This paper roll could be slipped into the hand of a relative during his visit. This was extremely difficult work not only because you had to write microscopic letters, but mainly because you had to do it in a manner that nobody would discover; not only the guards who every minute looked through the peephole of the cell door, but also other people who worked in the laboratory. The cigarette paper was placed in a small package made of wax, then put into a gel of very dark agar, which was available in the laboratory and which I made dark by adding black paint. It was most difficult t o find the opportunity to write in such a way that nobody would see me doing it. The “peephole” was not dangerous. I could pretend to write protocols of experiments, I always had them handy. I wrote only when there was no one in the laboratory room with me. That was not a frequent case and that is why my progress was very slow. I hurried to finish my writing in time for a regular meeting with my relatives, which took place once every 2-3 months. How can one slip such a small roll, button big? The rendezvouz was always attended by one, or, more often, two guards. They not only listened to what is said, but also watched us very closely. It looked almost Daily newspapers, governed by the state and the Communist party, respectively.

LEV ZILBER, T H E PERSONALITY AND THE SCIENTIST

21

impossible to deceive them. . . . I decided that it was necessary to distract their attention. But how can I do it? I had to distract the attention only of the guards, not of my visitors! I was usually visited by my brother V. Kaverin and my old and very close friend Zinaida Ermolieva, who organized a really titanic campaign for my discharge. To whom shall I slip my roll? And, how could I distract the attention of the guards while drawing closer the attention of my friends? It would be unexpected for them as well. I made four bread rolls. Using them I tried to stand in such a way that the guards would not see my right or left hands or at least my wrists . . . Here came visiting day. I had the roll with my manuscript between the middle and the ring fingers of my left hand; it could be easily slipped by the big finger of the same hand. T h e visit went as usual. When parting, 1 stood with my left side closest to Ermolieva. One of the guards was behind my back, another one behind her back. I dropped the handkerchief that I held in my right hand and immediately slipped the roll into Ermolieva’s hand. T h e hand closed and she did not even wink. T h e guard handed me the handkerchief, which he had closely examined. Ermolieva and Kaverin had left; everything was okay. It seemed I did it! (Zilber, 1989.)

This is what Kaverin (1 989) recalls: These were the last minutes of our visit. . . . We embraced each other with a feeling of helplessness and dismay and suddenly Lev took out a handkerchief from his pocket and dropped it o n the floor. In no time the officer lifted it, examined it carefully, and silently returned it to Lev. T h e last parting words. Stunned and dismayed Ermolieva and 1 walked together through the darkened streets of Moscow. “He slipped me a message,” she whispered. . . . I d o not remember how we got to her apartment. T h e message in the roll was folded ten to twenty times. Ermolieva slowl) and carefully unfolded i t . . . and in front of us we SAW a sheet of paper covered with microscopic letters, entitled-and you could read it without a magnifying glass-Viral Theory of Cancer Origin. Ermolieva had a big magnifying glass. We began to read and we read a long time because, although his handwriting was very good and w e could read each letter, we did not understand a single word. Everything was about viruses and the statements were so complicated that Ermolieva sometimes reread certain phrases several times. There was not a single word about false accusations for which he was kept in jail the fifth year; there was not a single hint o n his part about what we should do, to whom we should appeal, and who, according to him, could help. . . . In front of us we had a viral theory of cancer origin stated with the utmost briefness, almost in formulas.

Thus Zilber did it. His theory was in reliable hands and if something happened to him, his works would be read by others. Fortunately, Fate once again turned her face to him and he was discharged in March 1944 on the eve of his fiftieth anniversary. But why? Could it be the result of his talk with the NKVD commissar? Apparently the genuine reason for his sudden discharge was a different one. On the eve of his discharge a letter was written to Stalin that said that Zilber was not guilty and that it was necessary to let him free. An initiator of that letter was Ermolieva, a

22

LEV L. KISSELEV ET AL.

well-known Russian microbiologist, who selected the new strain of penicillin, the production of which played a huge role during the war. The letter was signed by General Nikolai Burdenko, Chief Surgeon of the Red Army, academician Leon-Orbeli, Vice President of the USSR Academy of Sciences, the writer Kaverin, and others. We may assume that this letter did not reach Stalin, but it caused commotion in the NKVD and they preferred to release Zilber before the term expired. We think that the liberation of Zilber was not a “restoration ofjustice,” or admission of fault, or rehabilitation of truth and rights. It became possible during the war only due to two critical circumstances: a solidarity of many eminent intellectuals (which was highly risky for themselves and was definitely a civil feat) and the enormously strong will of Zilber not to confess himself guilty under constant psychological and physical pressure. It was not a gift of fate o r power; it was the fight of a few human beings against the Stalinist NKVD machinery of annihilation, and in fact was a victory of humanism over totalitarianism. An extremely rare case and therefore even more staggering. . . For all of us, Zilber’s life is predominantly a lesson in morality and that is why it has great value forever. The history of Zilber’s three arrests and three discharges totally disproves the notion still widespread in the West that people arrested by the NKVD were “slightly” guilty, had some disagreement with those in power, had “some faults,” etc., but that they were treated and punished lightly. This is a harmful illusion, a dangerous myth justifying the terrorist NKVD actions. No traces of fault, no signs of guilt, no shadow of any, even minute, deviation from the law and nonetheless three arrests during ten years of the most productive period of life. He was released occasionally but put in jail purposely. The regime could not coexist with a man who had a creative mind, behaved honestly, did not denounce his colleagues, was not a member of the VKP(b) (Communist party), and was an aboveaverage Soviet citizen in intellect. VII. Finding the Family

There were twists and turns in Zilber’s life, and the same happened to his family. When the war broke out, Zilber’s wife fled from the bombardment, managing to take her children and their nurse to the summer house of her friends south of Moscow. When German troops approached Moscow all the roads were closed and it was impossible to return home. At that time, by chance, Valeriya Kisselyova was there with her children and her sister, who had arrived from Moscow. T h e whole

LEV ZILBER, THE PERSONALITY AND THE SCIENTIST

23

family was in occupied territory when German troops began their retreat, and were taken away with them by force. In the long run the family found itself in Germany, where in May 1945 they were freed by the Red Army in the city of Hemnitz. After liberation Valeriya was able to send a postcard home, addressed to her mother, who lived in Moscow. T h e postcard was received late in the evening and was immediately brought to Zilber, who had done his utmost to find out about his family. All the military and civil establishments to which he had appealed had answered that they had no information and that his wife and two children were lost without a trace. Therefore, when he saw the postcard it was a heavy, although joyful, shock for him. Clutching the postcard in his hand, he rushed to Miterev, the USSR Minister of Health who was aware of the fact that Zilber’s family was lost. It happened in July 1945, since the postcard, sent from Germany in May, reached Moscow in a month and a half. Government establishments at that time worked nights and Zilber managed to visit the minister in his office at a late hour. According to the rules of that time it was prohibited to write the return address on a postcard: it had only the number of the field post box. Therefore the first thing that Miterev did was to send the postcard to a decipherer who determined, in only a few minutes, that the family was not far from Breslau (Wrotzlav, now in Poland). Zilber immediately requested the minister to send him on a business trip to Germany. The minister thought for a minute and then found the way out of the situation. He retrieved from his safe a special order with a facsimile of Stalin’s signature. By filling in Zilber’s name, Zilber was made responsible for inspecting the hygienic state of the Soviet troops located in that territory of Germany. It was a powerful document and one could only admire the resourcefulness of the minister and his courage, if we take into account that Zilber was not a hygienist, was not a military man, and that only a little more than a year had passed since he had been discharged from jail. With the precious document in the breast pocket of his jacket, Zilber, without going home, headed straight for the aerodrome in the minister’s car and, showing the document, demanded his immediate departure for Berlin. No doubt the document produced a strong impression, as the military obeyed and he took the first plane to Berlin. A day had not yet passed since receiving the postcard before he was in Berlin and arrived at the Chief Hygienic Department, where he found himself friendlily embraced by the general, one of his numerous disciples of the prewar period at the Central Institute for Advanced Medical Studies. The general, having read the document, also saluted him, but soon Zilber

24

LEV L. KISSELEV ET AL.

explained the situation and told him the story of the disappearance and resurrection of his family. The general ordered that the professor be given any car he liked, together with a driver, armed with a machine gun and a box of cartridges (it was not quiet at that time on the roads). Zilber was a good driver; he had learned to drive a car during the eradication of the smallpox outbreak in Kazakhstan, where he could drive anywhere along the steppes, since there were no traffic signs there. Zilber and his driver changed with each other every two hours. It is almost impossible to describe the reunion of the family after five years of separation, jail, and dreadful war. It was there that he saw his youngest son (F.K.) for the first time, when he was four and a half years old. After the liberation, Zilber’s family took a train from Hemnitz to Dresden. The train got into a heavy railway accident and they survived it only by miracle. T h e plane that was to take Zilber and his family from Berlin to Moscow crashed because of a thunderstorm. At the last minute before take-off, while on the landing strip, the family was asked to take the next plane. Zilber, when asked about what happened to him and his family, used to say, “I am a fatalist.”

VIII. After the War: Virology and immunology of Cancer In 1945-1946 Zilber had to have been happy. Life and liberty. ’[he family found and saved. No one killed. He returned to work and not with empty hands. His book was being published, a small one but very purposeful, devoted to the new theory of‘ the origin of tumors (Zilber, 1946). T h e book on encephalitis, which had been preserved in proofs, was published as well (Zilber, 1945). He had been awarded Stalin’s prize not for the discovery of the causal agent and vector of tick-born encephalitis, but for the monograph on encephalitis. He was also elected a member of the newly founded Academy of Medical Sciences. And, once again, full of new ideas and energy, Lev Zilber took up the study of the virology of cancer. His scientific credo-a new look at the nature of cancer. To quote Zilber (1946), . . . the role of the virus is that it alters the hereditary properties ofthe cell, transfbrming i t from a normal into a tumoroiis one, and the cell that is formed in this way is the source of the tumor growth, whereas the virus that caused this transformation is either eliminated from the tumor due to the altered cell being a nonpermissive medium for its replication or loses its pathogenicity and therefore cannot be identified with a further growth of tumor.

LEV ZILBER, T HE PERSONALITY A N D TH E SCIENTIST

25

It is quite possible that the principle, similar to the m e that is effective during the serological transformation of microorganisms [i.e., transformation of bacteria under the action of DNA-uuthoi-s’ note] is effective during the transformation with viruses. We may assume that mutation that occurs with the transformation o f a normal cell into a tumorous one is conditioned by the penetration of the agent, causing this process [virus] to enter into the nuclei of cells and exert its impact on the genetic mechanism. In tumors the main pathological process is caused not by the virus, whose role is to transform a normal cell into a tumorous one, but*by the tumor cell per se, which triggers the growth of tumor. If some nontunior viruses are capable of inducing cellular proliferation it does not mean at all that they are capable of inducing inherited changes in the properties of. cells. T h e latter is typical only of tumor viruses and their efrect differs in principle from the character of the ef€ect of infectious agents, which mainly cause inflammatory and necrotic changes.

Zilber’s viral theory in its first version postulated a number of new concepts, the first among them being the assumption that the virus induces hereditary changes in cells by interacting with their genetic apparatus. The first popular description of the new theory appeared in the Zzvestzya, the central government newspaper (Zilber, 1945). I t is ironic to recall Zilber’s talk with the NKVD commissar (“You want me to publish your findings in the Iz-ciestaya or in the Pruudu?). Zilber published the results of his experiments done in prison (Zilber, 1945, 1946) and tried to reproduce then1 under more accurate and suitable conditions. He performed experiments on rats instead of mice and rabbits and stronger carcinogens were used. However, the results were poorly reproduced and with great difficulty. Here, in parallel studies he used a new approach in attempting to identify an oncogenic virus not by its tumorigenic activity, but as an alien protein-as a new antigen for the cell, as a specific tumor antigen. He launched a new attack together with his old and reliable associates 2. Baidakova and N. Nartzissov. Zilber obtained heterologous immune sera to tumors and studied them in immunoprecipitation and complement fixation reactions; he looked for the virus not in tissue extracts, but in the nucleoprotein fraction; he studied sera of tumor-bearing rats, hoping to reveal antitumor antibodies in the sera. T h e results were promising (Zilber et al., 1948; Zilber, 1950). At that time it was assumed that nucleoproteins were weak antigens or not antigenic at all. It did not fit well with Zilber’s approach, and in one series of experiments he purposefully, together with nucleic acids researchers I . Zbarsky and S. Debov, tested the immunogenicity of nucleoproteins in the anaphylaxis reaction of guinea pigs (Zilber et al., 1949). Zilber was deeply familiar with anaphylaxis from the time when he worked with Barykin-it was

26

LEV L. KISSELEV ET AL.

the subject of his privat-docent lecture at Moscow University and the subject of a detailed analysis in the manual on immunity. Guinea pigs were adequately sensitized with tumor nucleoprotein, responding to the challenging injection with classical anaphylactic shock but similarly to tumor preparations or preparations from the normal tissue. Zilber desensitized the guinea pigs to “normal” nucleoproteins and then injected a tumor antigen-a severe shock! And only to a tumor antigen! It may be said without exaggeration that this elegant demonstration of anaphylaxis with desensitization reaction (ADR) was destined to play a specific role in the scientific evolution of Lev Zilber and in the development of cancer immunology. Soon almost all studies in the laboratory involved ADR, always with steady success. Lev Zilber appealed to the Presidium of the Academy of Medical Sciences, reporting his findings and requesting them to send to his laboratory the most stringent commission to check his findings, to see if he was right and, irrespective of what they found out, to publish the results. The commission was made up of serious and, more importantly, skeptical investigators, such as the experimental oncologist L. Shabad, an ardent opponent of the viral theory of cancer; N. Medvedev, a pedantic researcher, a specialist in mouse genetics; and N. Kosyakov, an extremely skeptical immunologist, a specialist on tissue antigens. For several months the whole laboratory worked “for the commission.” All the results were fully reproduced and confirmed. T h e fact of antigenic differences between malignant and normal tissues was confirmed by the prestigious commission, which, however, did not consider it as evidence of the presence of virus in tumors (Shabad and Medvedev, 1950). For several years-a triumphant march of anaphylaxis with desensitization-tumors of animals and humans, leukoses and solid tumors, viral and “nonviral” tumors all demonstrated antigenic differences between malignant and homologous normal tissues. In 1955 Zilber’s department was visited by a United States delegation-the first after the start of the “cold war.” Members included R. Shope, M. Shimkin, and a number of other scholars. Both Zilber and his work produced a strong impression on the Americans, and they offered to publish his findings in the leading international journals, in which they appeared shortly afterward (Zilber, 1957, 1958). Anaphylaxis entered the world arena and, undoubtedly, promoted the rebirth or, to be more accurate, the second birth, of cancer immunology, the very possibility of which by that time was essentially discredited (see, e.g., Hauschka, 1952). P

LEV ZILBER, T H E PERSONALITY AND T H E SCIENTIST

27

In assessing the contribution of ADR to the development of cancer immunology we must ascribe to it at present mostly historical importance. T h e method is too complicated for a simple interpretation of the data. Does it distinguish qualitative differences from quantitative ones? What do the differences in the histological structure of the organ, heterogeneous in composition compared to the monoclonal tumor, which contains necrotic lesions and inflammation foci besides the connective tissue stroma, tell about results? It is difficult to answer these questions even now. Nevertheless, the most important statement remains unquestionable-anaphylaxis brought back the problem from nonexistence, it promoted the development and use of more simple and purposeful approaches to identify tumor-specific antigens-immunization in the singenic system, serology with singenic hyperimmune sera, and immunodiffision and immunohistochemical comparison of malignant and normal tissues. The latter two approaches were widely used in Zilber’s department, due directly to difficulties and failures with the use of ADR to isolate and characterize the individual antigens responsible for the differences between malignant and normal tissues (Zilber and Abelev, 1962). The successes of ADR inspired Zilber to begin experiments on antitumor vaccination, which he started at the beginning of the 1950s with his colleagues Z. Baidakova and R. Radzikhovskaya. They used BrownPierce rabbit carcinoma, which could be maintained for long periods of time in the nonsingenic condition, and spontaneous mammary cancer of mice-a primary tumor, induced by the virus with vertical transmission. Positive results have been obtained in both cases: strong results in the first system and marginal, but statistically meaningful, results in the second system (Zilber and Baidakova, 1955). If the significance of the results with the Brown-Pierce tumor model is not valuable from the contemporary standpoint, the effect of vaccination on the process of the emergence of primary viral tumors is undoubtedly referred to as one of the most important proofs of the existence of antitumor immunity.

IX. The Last Outburst of Stalin’s Tyranny These experiments played another role that cannot be considered a scientific one. The end of the 1940s and the early 1950s was a time of mounting struggle against the “admiration for the West,” a campaign that was a mixture of pseudopatriotism, chauvinism, and outright antisemitism. In 1952 the campaign was reaching its climax, with medical

28

LEV L. KISSELEV ET AL.

doctors, “killers in white gowns,” arrested and accused of murdering, by deliberately incorrect treatment, top government and party officials. Lev Zilber was too distinguished a personality not to become a victim of that campaign, which involved science, art, and the intellectual and economic life of the country. “Antipatriots” and “cosmopolites” were exposed one after another, and laboratories and departments were being closed. A wave of hostility and hatred was coming closer to the department that Zilber headed. An article had already been prepared for publication in the newspaper Meditzinsky Rabotnik against Zilber and his “ideologically alien” viral theory of cancer. At that time the appearance of such an article was a signal for ruthless persecution, which inevitably resulted in the complete destruction of the laboratory, department, and school. Zilber prepared for another (fourth) arrest, gave recommendations to his colleagues on how to behave during interrogations, and destroyed memos containing the names of other scientists and friends so that they would not be arrested. One day at the Ministry of Health, Zilber reported the results of his experiments on vaccination against cancer to the minister and members of the board, and apparently stated that he was on the way to creating an anticancer human vaccine. It is true that he was absolutely confident about this and tried to do his best to create such a vaccine. The reaction was fast! The campaign was halted at the threshold of Zilber’s department, the article was not published, and the persecution was lifted. His department was expanded to include several new positions and an automobile was provided to aid in collecting operation material from Moscow clinical hospitals. Baidakova and Radzikhovzkaya began to work separately and a curtain of top secrecy was drawn over their studies. The death of Stalin (March 1953) stopped this snowballing horror. Medics were set free in exactly one month and Lev Zilber was congratulated by his happy friends and colleagues.

X. Reunification with the World Scientific Community It was a happy time at first: Khruschev’s “thawing period.” Lev Zilber was once again on the upgrade. In the autumn of 1957, after a 25-year interval, he went abroad to attend the International Symposium on Tolerance, held in Prague. Here he made acquaintance with the late Milan Hasek and Jan Svoboda, who launched his brilliant scientific career. Later he attended international congresses and conferences in London (1958), Berlin (1959),Amsterdam (1960),and Warsaw (1961). He struck

LEV ZILBER, T H E PERSONALITY A N D T H E SCIENTIST

29

up acquaintances with Grabar, Graffi, Dux, Muhlbock, Haddow, Maisin, Klein, and Lvov. In 1955 Peter Medavar came to Moscow with reports on tolerance that he had recently made. Zilber was impressed by his work. Together with a young embryologist, Irina Kryukova, who had recently joined his department, he began a set of experiments on tolerance. These studies were based on the idea of using tolerance to obtain antitumor-specific sera. It was planned to induce tolerance to healthy tissues with a subsequent injection of tumor antigens. It was expected that by using such a scheme of inducing tolerance with subsequent immunization, heterologous antisera to tumor-specific antigens could be raised. XI. Discovery of Pathogenicity of Rous Sarcoma Virus for Mammals Experiments were conducted, using chickens infected with Rous sarcoma virus as donors and rats and rabbits as recipients. Unexpectedly, the rats that received homogenates of the chicken viral sarcoma in the neonatal period developed a specific pathology-tumor-like hemorrhagic cysts-whereas rabbits developed fibromas (Zilber and Kryukova, 1957, 1958; Zilber, 1961). At the same time Svet-Moldavsky and Skorikova (1957) showed that Rous virus induced tumors in adult rats. Later, tumors were induced in various mammals, including monkeys (Zilber et al., 1964, 1965, 1966). These data exerted great influence on the further development of tumor virology. It became clear that RNA-containing tumor viruses were able to bypass interspecies barriers and consequently the virus had to contain certain elements presumably of genetic origin; the pathway of its “oncogenic” manifestation should be similar in the cells of different origin, because the final result of this action was identical-tumor formation. Concerning the mechanism of viral action, the subsequent experiments of Svoboda (1960) showed very clearly that mammalian cells transformed by Rous sarcoma virus did not produce virus but did contain it in a cryptic form, since after cocultivation of transformed cells with susceptible avian cells the latter became infected by Rous virus. The direct contact between viable donor and recipient cells was obligatory for this virus rescue. This phenomenon might be greatly enhanced by the formation of somatic cell hybrids between tumor cells permissive for viral replication. In essence, these data were the first valuable evidence for the persistence of the whole viral genome in “nonviral” tumors, originally induced by virus. Later this cryptic genome was identified in

30

LEV L. KISSELEV ET AL.

the integrated form and called provirus. Ten years later a hypothetical “oncogenic” element in the Rous virus genome was also identified and called the src oncogene. All these data opened a new field in oncology: it became evident that genes responsible for malignant transformation of the cells do exist. XII. In Search of Oncogenic Viruses

At this time Zilber, with his colleagues, continued efforts to isolate oncogenic viruses from tumor cells or to activate them by means of various agents. He succeeded in two cases. Zilber and Stepina (1963) received quite an unexpected result-in an attempt to isolate virus similar to Bittner’s virus, from the cultured mammary tumor cells of mice by adsorption on erythrocytes, they received a new strain of leukemia virus that until then had not yet been well characterized. In other work, with Postnikova (1 966), he managed to isolate a leukemogenic agent in inbred mice by treating them with chemical carcinogens. It should be stressed that Zilber’s studies exerted a large influence on a number of other groups and laboratories in the Soviet Union. Among them a special place is taken by Mazurenko, who in 1958 isolated a mouse leukemia virus by injecting a smallpox virus into mice of the CC57B strain (Mazurenko, 1962). These data, as well as the results of other studies, enabled Zilber to assume that viruses in a latent state can persist in tumor cells and be activated under the action of some factors. These data, his own considerations, and analogies with the process of lysogenization in bacteria allowed Zilber to put forward the idea on the integration of viral and cellular genomes during malignant transformation of cells (Zilber, 1958, 1961). This assumption laid the basis for the virogenetic theory (see Zilber, 1968). XIII. The Virogenetic Concept of the Origin of Tumors Undoubtedly the idea of integrative interaction between viral and cellular genomes with the malignant transformation of cells was on the tips of people’s tongues at the beginning of the 1960s. In particular, this hypothesis was made by Vogt and Dulbecco (1960) but Zilber, back in 1958, made an assertion on the interaction of the two genomes, and in 1961 clearly coined the concept of the integration of virus genome into the genome of the malignant cell. The basic concepts of the virogenetic theory were formulated in a

LEV ZILBER, T H E PERSONALITY AND T H E SCIENTIST

31

final form by Zilber in his last book, written in 1965-1966 and published in 1968: 1. Naturally occurring tumors are induced by viruses. 2. T h e tumor-inducing effect o f viruses on the cell differs in principle from an infectious effect. 3. Tumor-inducing viruses d o not differ from the viruses inducing infectious diseases in their other characteristics. 4.T h e effect of tumor-inducing viruses on cells is accompanied by alterations in the hereditary properties of cells. 5. Tumor conversion of cells is not induced by virus, but by its nucleic acid. 6. New genetic information, imported by the nucleic acid of the virus into the cell, is partially or completely incorporated by the genome of the cell. 7. Hereditary changes, induced by this process, disturb the interrelationship between cells and the systems of the organism that regulate and control cellular division, as a result of which cells d o not subordinate to the latter, which triggers an uncontrolled replication of cells, resulting in the formation of tumors. 8. T h e virus that induced the tumor conversion of cells is not involved in the replication of the already formed malignant cells. Tumor cells either d o not produce a mature virus at all or form its immature forms. In those cases when tumor cells produce a mature virus, it turns out to be a “passenger” and does not affect the cell growth. 9. T h e problem of the involvement of viruses in carcinogenesis induced by chemical and physical factors needs to be studied further. T h e available information enables us to assume the presence of indirect carcinogenesis (Zilber, 1968).

Most important for Zilber at that period (early 1960s) was to find indisputable evidence for the virogenetic theory. He anticipated it in molecular-biological approaches and strove to organize such studies in his department. He invited a microbiologist, a specialist in phage genetics, to his laboratory, entertaining the hope that he would become involved in the study of cancer. He also employed a young and active biochemist, a specialist in nucleic acids, and formed a group of researchers for him in the hope that he would take up the problem of the integration of oncogenic viruses. However, having worked for a couple of years, the biochemist left for another institute to work on more traditional problems. Zilber also invited to his department several gifted young molecular biologists, but they hesitated to take up such an unusual subject. Let us state in passing that those young biochemists began 5-7 years later to actively study the molecular biology of oncogenic viruses. Regrettably, at that time Zilber failed to make them take up the problem. Nowadays, looking back, we see that that time-the beginning of the 1960s-was a golden age for launching studies on molecular oncovirology. Zilber’s department was ready: it had a large-scale biological

32

LEV L. KISSELEV ET AL.

base, including tissue cultures, oncogenic viruses like polyoma, SV40, murine leukemia, chicken sarcoma, and rabbit papilloma. But Zilber himself did not have enough time to change the traditional thinking of either the virologists or biochemists around him or to initiate the investigations that he considered to be most important. During this period of great enthusiasm for the search and isolation of oncogenic viruses from tumor cells, including human tumors, Zilber proposed that viruses can play a role in other pathological processes of unknown etiology, first and foremost in the so-called slow neurological diseases. He took up lateral amyotrophic sclerosis (apparently because his best friend Tynyanov died of a similar disease) and tried, jointly with Gardashyan, Bunina, and Konovalov, to isolate this presumptive virus from patients. Although they managed to transfer this disease to monkeys with the help of cell-free extracts (Zilber et al., 1962, 1963), these studies were not continued after Zilber’s death, and from the contemporary standpoint they lack convincing evidence of the isolation of virus. However, the idea itself turned out to be correct: recently, genetic determinants of human T lymphotropic viruses have been identified in two diseases of the nervous system-spastic trophic paraparesis (Gessain et al., 1985) and multiple sclerosis (Koprowskii et al., 1985). XIV. Development of Tumor Immunology

During the same period (the end of the 1950s and the beginning of the 1960s) Zilber’s department successfully developed immunological studies. Immunodiffusion analysis was widely introduced into the study of the antigenic structure of tumors. The first tumor-embryonal antigen, a-fetoprotein, was discovered, which later led to the immunodiagnosis of cancer (see Abelev, 1983, 1989). The antigen that was associated with carcinoma of the stomach in humans was identified (Zilber, 1962; Zilber and Ludogovskaya, 1967). Antibodies were revealed, in a strictly singenic system, to specific antigens of methylchloranthrene sarcomas in mice (Lejneva et al., 1965). That was the first study reliably demonstrating the presence of humoral antibodies to tumor-specific antigens in tumor-bearing animals. At the same time a study performed in Zilber’s laboratory was the first to reveal the existence of a group-specific antigen of mammary tumor virus in mice (Lejneva, 1961). The monograph “Virology and Immunology of Tumors” was written (Zilber and Abelev, 1962), the first in that area, which also generalized the whole experience of the laboratory and which was later translated in English (Zilber and Abelev, 1968).

LEV ZILBER, T H E PERSONALITY AND T H E SCIENTIST

33

During those years Lev Alexandrovich vividly felt the birth of the new immunology and did his best to create the basis in his department for the development of basic immunology, which was nonexistent at that time in the Soviet Union. In 1961 he invited Gurvich to Moscow and organized a laboratory for him, to study the chemistry and biosynthesis of antibodies; he also extended an invitation to Fontalin to organize the Laboratory of Immunological Tolerance, and to Kulberg to set u p the Laboratory of Immunochemistry. He appointed Abelev to run the Laboratory of Cellular Antigens, and encouraged Brondz, a postgraduate at that time, to launch his active studies of cellular cytotoxicity. His department acquired a different name, becoming the Department of Basic Immunology and Oncology. Its structure was not preserved for long. Soon laboratories headed by Gurvich, Fontalin, and Kulberg became structurally independent, but that was not importantthese units had already been set up and there was effective cooperation between them. For many years they were almost the only laboratories that dealt with the problems of fundamental immunology in the Soviet Union. XV. The Last Efforts and the Last Days In 1964 Lev Zilber was the first to give a lecture at the newly created Department of Virology at Moscow State University. Recognition and respect marched hand in glove to its climax. They were vividly manifest at the International Symposium on Tumor-Specific Antigens, organized by Zilber in 1965 in Sukhumi, at the institute headed by B. Lapin. That brilliant symposium was characterized by a spirit of enthusiasm, and open and lofty cooperation. World-renowned scholars from the United States (Sabin, Huebner, Koprowski, Melnick, Southam), Grabar and Burtin from France, Klein and Sjogren from Sweden, Harris from the United Kingdom, Sachs from Israel, Svoboda and Koldovsky from Czechoslovakia, Hirai and Aoki from Japan, and scientists from other countries took part in it together with the researchers from Zilber’s department and other laboratories of the Soviet Union. T h e whole work of the symposium was permeated with a feeling of respect and gratitude to Zilber. T h e proceedings of that remarkable symposium, published in 1967, were dedicated to his memory (Harris, 1967). T h e most important development for Zilber at that time was his involvement in human cancer immunology. However, it was difficult for him to permanently convince his colleagues, who did not want to tear away from model systems, that it was a golden age for the experimental

34

LEV L. KISSELEV ET AL.

immunology of cancer. Moreover, the studies of Zilber’s laboratory were at the threshold of human cancer immunodiagnostics, and an attempt was yet planned to hold an international experiment in western Africa on the immunodiagnosis of primary hepatocarcinoma. He felt distressed during the last two years of his life because of relations with the director of the institute. Zilber had promoted his breath-taking career, and now expected help and understanding from him. However, the man was doing the opposite and although he could not do too much harm, it irritated Zilber. All this was happening against the background of all-around recognition and world-wide respect! On November 10, 1966, Lev Zilber, who still did not feel well after another attack of influenza, came as usual to the institute to hold a weekly conference of the department. He walked up without stopping as was his custom, to the fourth floor, since the elevator did not function in the (newly) built building, and felt bad. He died of gross myocardial infarction at 10:30 a.m. in his own office, surrounded by his colleagues, who tried to help him. XVI. Legacy

He died the way he lived-swiftly, at work, knowing no weakness and depression. The day before he had given his secretary the last page of the manuscript of his book “The Virogenetic Theory of the Origin of Tumors” (Zilber, 1968). How can w e assess the theory today, and the role it played in the development of cancer virology? The general biological significance of the virogenetic theory was that the principle of integration of viral genomes and cells laid the basis for the malignant transformation of normal cells into malignant ones. This principle was valid for all tumors induced by viruses and was later validated after the discovery of integration of retroviruses through reverse transcription (Temin and Mizutani, 1970; Baltimore, 1970) and the discovery of the possible etiological role of such viruses as human T cell leukemia virus (HTLV 1) (Poiesz et al., 1980; Hinuma et al., 1981), human papilloma virus (HPV) (zur Hausen, 1980), Epstein-Barr virus (Epstein, 1985), and hepatitis B virus (HBV) (Szmuness, 19’78) in certain human neoplasias. However, in subsequent years it was shown that although the integration of genomes is necessary, it is far from being sufficient for malignant transformation. With the general attractiveness of the concept of compulsory integration of viral genetic material it has become evident that only part of viral genome is necessary for the malignant transformation of cells. The main role is played by special genes-oncogenes-that have

LEV ZILBER, THE PERSONALITY AND T H E SCIENTIST

35

been identified for the first time in oncogenic viruses, and which play a key role in these processes (see Weiss et al., 1985). Apparently an auxiliary role in this process could be played by other viral genes, which we may refer to as ( 1 ) gene immortalizers, which make cells immortal (Rassoulzadegan et al., 1982), and (2) gene transactivators, which are capable of making other viral and cellular genes function (Nyborg et al., 1988) and causing parts of the viral genome to interact with nuclear transcription factors of cellular origin (see Leonardo and Baltimore, 1989). In this way, the idea of the integration of the genetic material of a virus and a cell was fully confirmed, and then led to the identification of viral oncogenes, responsible for the process of malignant transformation. That helped to make another step toward the discovery of cellular protooncogenes-a fundamental element of the general theory of normal and malignant growth. T h e virogenetic theory of the origin of tumors became a stepping stone to contemporary virology and experimental oncology. Being in favor of experimentation, and experimenting almost all of his life, Lev Alexandrovitch had a rare working capacity, orderliness, and memory; he had the capacity for generalization and classification and this helped him to leave a large and diverse legacy. The first small monograph by Zilber, entitled “Paraimmunity,” saw light in 1928. In 1937 he published the prestigious book, “Immunity,” written jointly with V. Lyubarsky. That book was based on lectures he presented at the Central Institute for Advanced Medical Studies and it immediately became a textbook on immunology for a whole generation of immunologists and microbiologists. After the war Zilber revised the manuscript and in 1948 it was published under the title, “The Fundamentals of Immunity”; it has no equal in world literature at that time. The book became a handbook for the postwar generation of scholars and students. In 1958 Lev Alexandrovitch published “The Fundamentals of Immunology,” and mentioned in the Foreword that “Immunity,” “The Fundamentals of Immunity,” and “The Fundamentals of Immunology” are a single exposition of the fundamentals of general immunology. According to immunologists, the last book, written 45 years ago, still preserves its significance and is used in training. This seems unbelievable if we take into account that during the last decades immunology has made gigantic steps forward; it would seem that the book would soon lose its significance. But virtually every Russian-speaking immunologist, for not less than 40 years (starting from 1937), studied immunology “by Zilber.” A very important role for 15 years was also played by the manual

36

LEV L. KISSELEV ET AL.

“Immunochemical Analysis,”written by disciples and colleagues of Zilber, and of which he was editor. It was published in 1968, after his death. Zilber’s contribution to virology is also fundamental. In 1945 he published the monograph “Epidemic Encephalitis,” the tragic effects of which have been dwelt on in this essay. In 1946 the book “The Viral Theory of the Origin of Malignant Tumors” was published, small in volume but prophetic in nature. “The Teaching on Viruses (General Virology)”-the first textbook on virology in the Soviet Union-was published in 1956. The first monograph in worid literature on virology and immunology of cancer, written jointly with Abelev, appeared in 1962. And finally, just before his death (literally one day before), Zilber managed to finish his last book, “The Virogenetic Theory of the Origin of Tumors.” It was prepared for publication by L. K. and V. P. Kisseljova, with the active assistance and participation of Zilber’s colleagues, and saw light in 1968. The “Selected Works” appeared in 1971, in which one of us (G.A.) wrote the first essay about Lev Zilber. The book by Zilber, Irlin, and F.K., “The Evolution of the Virogenetic Theory of the Origin of Tumors,” was published posthumously in 1975. It presented experimental data that confirmed Zilber’s concept. We regret only that all this intellectual wealth, created by decades of tremendously hard work, remained inaccessible for the world scientific community as a result of the isolation of the Soviet Union from world science in the epoch of Stalinism and the “iron curtain.” Only one book, “The Virology and Immunology of Cancer,” which was written with G.A. at the time of Khruschev’s “thawing” period, was published in 1968 by Pergamon Press in London. Besides that, the same publisher published the book “Etiology and Pathogenesis of Tumors” in 1959, which was a translation of a collection of articles by Zilber’s colleagues and disciples, and which was published in commemoration of his sixtieth birthday in 1956 in the Russian language. However, it would be wrong to state that Zilber was unknown to the West: it is sufficient to see the list of his reviews and articles, published in the 1960s in international publications (Zilber, 1962a,b, 1965; Zilber and Postnikova, 1966; Zilber and Ludogovskaya, 1967). Completing the enumeration of Zilber’s vast literary legacy, we cannot help drawing attention to one specific feature so typical of Lev Zilber. Many of his books had dedications and from them a reader may discern the circle of people particularly dear and close to him. To Yu. N. Tynyanov, the author of “Kyukhlya,” in memory of our youth (“Paraimmunity,” 1928).

LEV ZILBER, THE PERSONALITY AND THE SCIENTIST

37

To the Blessed memor) of my mother, Anna Grigorievna Desson (“The Fundamentals of Immunity,” 1948). To Zinaida Vissarionovna Ermolieva (“Epidemic Encephalitis,” 1945). To my wife, a truthful and loyal friend in trouble and in happy days, I dedicate my last book (“The Virogenetic Theory of the Origin of Tumors,” 1968).

Finally, several words about Zilber’s school. T h e impetus received from him was so powerful that all his disciples and colleagues continued their work in the direction selected by him. At present the former department has developed into several interacting and cooperating laboratories, actively studying the problems of cancer immunology (Abelev et d), general immunology (Brondz et d), and retroviruses (Kryukova, Ilyin, and others). Other institutes in the country also have laboratories, headed by former colleagues from Zilber’s laboratory, that are actively working in virology and the immunology of cancer. Twice a year, on the days of Zilber’s birth and death, his family, disciples, and colleagues meet at his grave at the Novodevichye Cemetary in Moscow. We should like to complete this essay with the words of Lev Zilber’s friend, Wladiinir Engelhardt, taken from the foreword of Zilber’s posthumously published book “The Virogenetic Theory of the Origin of Tumors”: “Let his last book not only be the result of the brilliant research work, let it stir up recollections of his bright personality as a citizen, a scholar and an unwearing pioneer of the truth! He could, like Cyrano d e Bergerac, be proud of the white plume of his combat helmet: he waved it spotless through hot battles in the fields of scientific disputes and through Stalinist torture-chambers.” (The last words had been crossed out during publication, but were preserved in the original.)

ACKNOWLEDGMENTS T h e authors express their sincere and deep gratitude to Prof. G. Klein, who made valuable suggestions after reading the manuscript, and to all those who helped us with their advice and technical assistance in the preparation of this essay. ~ s Rrsearrli for their proposal that we We are thankful to the Editors of A d ~ ~ iini Crrricpr write Zilber’s niemorial essay.

REFERENCES Abelev, G. I. (1983). on cod^?^. BioL Men. 4, 371-398. Abelev, G. I. (1989). Tumor Bzol. 10, 63-8 1 . Avery, 0. T., MacLeod, C. M., and McCarty, M. J. (1944).J. Exp. Med. 79, 137-156. Baltimore, D. (1 970). Nature (London) 226, 1209- 12 1 1. Bittner, J. J. (1936). Sczence 84, 162-164.

38

LEV L. KISSELEV ET AL.

Chumakov, M. P. (1985). Mol. Biol. (Moscow) 18, 1703-1705. Epstein, A. (1985). IARC Sci. Publ. n60, 17-31. Gessain, A,, Barin, F., Vernant, J. C., Cont, O., Maurs, I., Calendar, A,, and de The, G. (1985). Lancet 2, 407-415. Griffith, E. (1928).J. Hyg. 27, 113-129. Harris, H., and Watkins, J. F. (1965). Nafure (London) 205, 640-642. Harris, R. J. C., ed. (1967). “Specific Tumor Antigens.” Munksgaard, Copenhagen. Hauschka, T. S. (1952). Cancer Res. 12, 615-620. Hinuma, Y., Nagata, K., Hanaoka, M., Nakai, M., Matsumoto, T., Kinoshita, K., Shikawa, S., and Miyoshi, I. (1981). Proc Natl. Acad. Sci. U.S.A. 78,6476-6481. Kassirsky, I. A. (1949). In: “Problemy i Utchionyie (Problems and Scientists),”pp. 256-297. Medgiz, Moscow. Kaverin, V. A. (1989). Druzba Narodov (Moscow) n4, 5-28. Koprowski, H., de Freitas, E., Harper, M. E., Sqnberg-Wollheim, M., Sheremata, W. A,, Robert-Gurrof, M., Saxinger, C. W., Feinberg, M. B., Wong-Staal, F., and Gallo, R. (1985). Nature (London) 318, 154-157. Leonardo, M. J., and Baltimore, D. (1989). Cell 58, 277-28 1. Lejneva, 0. M. (1961). Vopr. Onkol. (Moscow) 7, 53-59. Lejneva, 0. M., Ievleva, I. E., and Zilber, L. A. (1965). Nature (London) 206, 1163-1 165. Lejneva, 0. M., Zilber, L. A., and Ievleva, E. S. (1965). Dokl. Akad. Nauk SSSR (MOSCOW) 162, 1440-1443. Mazurenko, N. P. (1962). “Rol Virusov v Etiologii Leykosov (Role of Viruses in Ethiology of Leukemia).” Kiev. Minervin, S. M., and Sotskaya, 2. Ya. (1935). Zh. Epidem. Microbiol. Immunobiol. (Moscow) n3, 422-430. Nyborg, J. K., Dynan, W. S., Chen, 1. S. Y., and Wachsman, W. (1988). Proc. Natl.Acad. Sci. U.S.A. 85, 1457-1462. Poiesz, B., Ruscetti, F. W., Cazdar, A. F., Bunn, P. A., Minna, J. D., and Gallo, R. (1980). Proc. Natl. Acad. Sci. U.S.A. 77, 7415-7419. Rassoulzadegan, M., Cowie, A., Carr, A,, Glaichenhaus, N., Kamen, R., and Cuzin, F. ( 1982). Nature (London) 300, 7 13-7 15. Rous, P. J. (1911). J. Am. Med. Assoc. 56, 198-211. Shabad, L. M., and Medvedev, N. N. (1950). Vestn.Akad. Med. NaukSSSR (Moscow)n6,28-32. Sharov, A. (1963). In: “Pervoye Srajeniye (The First Battle),” p. 207. Molodaya Guardia, Moscow. Shope, R. E. (1932).J. Exp. Med. 56, 803-810. Svet-Moldavsky, G., and Skozikova, A. (1957). Vopr. Oncol. n3, 673-677. Svoboda, J. (1960). Nature (London) 16, 980-983. Szmuness, W. (1978). Prog. Med. Virol. 24, 46-74. Temin, H., and Mizutani, S. (1970). Nature (London) 226, 121 1-1214. Vogt, M., and Dulbecco, R. (1960). Pruc. Natl.Acad. Sci. U.S.A. 46, 365-370. Weiss, R., Teich, N., and Varmus, H., eds (1985). “RNA Tumor Viruses,’’ Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Zilber, L. A. (1921). I n : “Epidemitcheski Sbornik (Reviews on Epidemiology),” pp. 91-96. Rostov-na-Donu. Zilber, (Silber) L. A. (1923). Zbl. Bact. 89, 250-260. Zilber, L. A. (1928). “Paraimmunitet (Paraimmunity),”Moscow. Zilber, L. A. (1939). Archiv. Biol. Nauk (Moscow) 56, 9-37. Zilber, L. A. (1945a). Izvcstija (Moscow) 1114,3. Zilber, L. A. (1945b). Zh. Epidem. Microbial. Immunobiol. (Moscow) n9, 43-52.

LEV ZILBER, T H E PERSONALITY A N D T H E SCIENTIST

39

Zilber, L. A. (1945~).Zh. Epidem. Microbiol. Imniunobiol. (Moscow) 114-5, 16-25. Zilber, L. A. (1945d). In: “Epydemitscheskye Enzephality (Epidemic Encephalitis),” pp. 125-128. Medgiz, Moscow. Zilber, L. A. (1946). “Virusnaya Teorija Proishojdenya Zlokatchestvennych Opucholey (The Viral Theory of Tumor Origin.)” Medgiz, Moscow. Zilber, L. A. (1948). “Osnowy Immuniteta (The Fundamentals of Immunity).” Medgiz, Moscow. Zilber, L. A. (1949). Usp. Sovrem. B i d . (Ada. Mod. Biol.)(Moscow) 27, 185-210. Zilber, L. A. (1950). Usp. Sovrem. Biol. (Adv. Mod. Biol.) (Moscow) 30, 188-221. Zilber, L. A. (1957).J. Natl. Cancer Inst. 18(3), 341-358. Zilber, L. A. (1958a). “Osnowy lmmunologii (The Fundamentals of Immunology).” Medgiz, Moscow. Zilber, L. A. (1958b). Adu. Cancer Rrs. 5 , 291-329. Zilber, L. A. ( 1958~).Pathologicheskaya Phyxiologia i Experinienlalnay~ilerapaa (Moscow) n3, 3-1 1. Zilber, L. A. (1961a). Vopr. Virusol. ( M o ~ c o w )1, 3-9. Zilber, L. A. (1961b).J. Natl. Cnncer Inst. 25, 131 1-1319. Zilber, L. A. (1 961c). J . Nall. Cancer Inst. 6 , 1295- 1305. Zilber, L. A. (1962a). Ann. N.Y. Acad. Sci. 101, 264-270. Zilber, L. A. (196213). Cold Spring Harbor Sjnip. Quant. Bzol. 27, 5 13-5 17. Zilber, L. A. (1965). Prog. Exp. Tumor Res. 7, 1-48. Zilber, L. A. (1966). Nrmka i Zhizn. (Science and Lifp) (Moscow) nl2, 55-63. Zilber, L. A., ed. (1968a). “Immunochimitchesky Analys (Immunochemical Analysis).” Medgiz, Moscow. Zilber, L. A. (1968b). “The Virogenetic Theory of the Origin of Tumors.” Nauka, Moscow. PrometqY (Molodaya guardia, Moscow) 5 , 296-309. Zilber, L. A. (1968~). Zilber, L. A. (1969). Pn‘roda (Naturu, Moscow) n10, 48-53. Zilber, L. A. (1971). In: “Isbranyie Trudy (Selected Works),” p. 188. Meditsyna, Moscow. Zilber, L. A. (1989). Ogonyok (Moscow) n21, 10-12. Zilber, L. A., and Abelev, G. I. (1962). “The Virology and Immunology of Cancer.” Medgiz, Moscow. Zilber, L. A,, and Abelev, G. I. (1968). “The Virology and Immunology of Cancer.” Pergamon Press, Oxford. Zilber, L. A,, and Baidakova, Z. L. (1955). Vopr. Oncol. (Mowow) 1, 14-20. Zilber (Silber), L. A,, and Dosser, E. M. (1934). Zbl. B a t . 131, 222-232. Zilber, L. A., and Korshunova, 0. S. (1949). Zli. Epidem. Microbiol. Imniunobzol. (Moscow) n4, 47-53. Zilber, L. A , , and Kryukova, I. N. (1957). Vopr. V i m o l . (Moscow) n4,239-243. Zilber, L. A,, and Kryukova, I. N. (1958). Vopr. V i m o l . (Moscow) n3, 166-169. Zilber, L. A,, and Ludogovskaya, L. A. (1967). Folia B i d . (Pranka) 13, 331-334. Zilber, L. A., and Lyubarsky, V. A. (1937). “Immunitet (Immunity.)” Biomedgiz, MoscowLeningrad, Zilber, L. A,, and Postnikova, 2. A. (1966). Natl. CancPr I m t . Monogr. 22, 397-401. Zilber, L. A., and Stepina, V. N. (1963). A&. VII Internat. Cancer Congr. (Moscow) 3, 169. Zilber (Silber), 1.. A,, and Wostruchova, E. I. (1931). 2. Immunol. F o r d 70, 239-249. Zilber (Silber), L. A., and Wostruchova, E. I. (1932). 2. Ininiunol. Forsch. 76, 59-69. Zilber (Silber), L. A,, and Wostruchova, E. I. (1933). Zbl. Bact. 129, 389-400. Zilber (Silber), L. A., and Wostruchova, E. I. (1934) Zbl. Bact. 132, 314-320. Zilber (Silber). L. A,, Schafran, A. S., and Demidova, M. V. (1933). Z. Intmunol. Forsch. 79, 110-1 16.

40

LEV L. KISSELEV ET AL.

Zilber, L. A,, Nartsissov, N. V., Rivkind, T. L., and Baidakova, Z. L. (1948). Vestn. Akad. M e d . Nauk (Moscow) 3 , 36-39. Zilber, L. A,, Freiman, V. B., Zbarski, I.B., and Debov, S. S. (1949). DoR1. Akad. Nauk S S S K (Moscow) 65, 97-100. Zilber, L. A., Baidakova, Z. L., Gardashyan, A. N., Konovalov, N. V., Bunina, T. L., and Barabadze, E. M. (1962). Vopr. Vzrwol. (Moscow) n5, 520-528. Zilber, L. A., Baidakova, Z. L., Gardashyan, A. N., Konovalov, N. V., Bunina, T. L., and Barabadze, E. M. (1963).Bull. WHO 29,449-456. Zilber, L. A., Lapin, B. A., and Adzigitov, F. I. (1964). Vopr. Virwol.(Moscou)n4,498-499. Zilber, L. A., Lapin, B. A., and Adzigitov, F. I. (1965). Nature (London) 205, 1123-1 124. Zilber, L. A., Lapin, B. A., and Adzigitov, F. I. (1966). Znt. J. Cancer 1, 395-407. Zilber, L. A., Irlin, 1. S., and Kisseljov, F. L. (1975). “Evolutsiya Virusogeneticheskoi Teorii Proischojdenia Opucholei (Evolution of the Virogenetic Theory of Tumor Origin.)” Nauka, Moscow. zur Hausen, H. (1980).Adz,. Cancer Re.s. 33, 77-109.

THE GENETICS OF WILMS’ TUMOR Daniel A. Haber*t and David E. Housman* ‘Center for Cancer Research, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139 TMassachusetts General Hospital Cancer Center, Charlestown, Massachusetts 02129

I. 11. 111. IV.

V. VI.

VII. VIII.

Introduction Histology and Clinical Considerations T h e Knudson Model Genetic Loci Associated with Wilms’ Tumor A. Chromosome 1 lp13 €3. Chromosome 1 l p l 5 C . Familial Wilms’ Tumor Isolation of the W T l Gene at 1 1p 13 WT1: Characterization of a Novel Tumor Suppressor Gene A. Gene Structure and Alternative Splicing B. Normal Tissue Expression of’WT1 C. WTI Mutations in the Germline and in Wilms’ Tumors Functional Studies and Animal Models Conclusions References

I. Introduction The isolation of the WTI gene involved in the genesis of Wilms’ tumor has provided a new molecular tool to understand the normal and abnormal development of the kidney and related tissues. Wilms’ tumor is pediatric kidney cancer, which arises in 1 in 10,000 children. It can present in both a common sporadic and a rare hereditary form, along with various congenital abnormalities. The existence of both gross chromosomal abnormalities as well as more subtle molecular deletions has led to the genetic characterization of a number of loci involved in the development of Wilms’ tumor. Within one of these loci, band 13 on the short arm of chromosome 11, we have recently isolated a gene, WTI, which is specifically inactivated in Wilms’ tumors. In this article we discuss the complex genetics of Wilms’ tumor, and the initial studies characterizing the role of the WTI gene product in tumorigenesis.

41 ADVANCES IN CANCER RESEARCH, VOL. 5Y

Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form resemed.

42

DANIEL A. HABER AND DAVID E. HOUSMAN

II. Histology and Clinical Considerations Wilms’ tumor constitutes some 10% of all pediatric cancer, and is the most common intraabdominal solid tumor in children. The peak incidence is between 3 and 4 years of age, with most children presenting with a palpable abdominal mass. Some 5-10% of patients with Wilms’ tumor have bilateral cancers, and these children tend to present between 2 and 3 years of age (Matsunaga, 1981). Rarely, children with Wilms’ tumor will show evidence of genetic malformations and chromosomal abnormalities (see below). T h e malignant transformation in Wilms’ tumor is thought to originate in cells of the metanephric blastema (Bennington and Beckwith, 1975). This fetal structure is thought to give rise to the genitourinary system, and Wilms’ tumors are characterized by their histologic diversity (see Fig. 1). Most tumors have the typical “triphasic” histology, consisting of primitive or blastemal cells, more differentiated or epidermal cells,

FIG. 1. Wilms’ tumor histology. Wilms’ tumor is characteristically composed of‘ primitive blastemal cells (B), epithelial cells (E), and strornal cell components (S). This “triphasic” histology is occasionally more complex with evidence of further cellular differentiation. WTI expression is found primarily within the epithelial and blastemal cell types. [Photomicrograph (hematoxylin and eosin stain) provided by Dr. Nancy Harris, Department of Pathology, Massachusetts General Hospital, Boston, Massachusetts.]

THE GENETICS OF WILMS’ TUMOR

43

along with a stromal cell component. Occasionally, regions of neural or muscle differentiation are found. One variant of Wilms’ tumor, the anaplastic cell type, is distinguished by the presence of large, grossly abnormal cells, and carries a worse clinical prognosis than the majority of tumors (Breslow et al., 1986; Douglass et al., 1986). In addition, tumors with the characteristic triphasic histology can arise outside the kidney, particularly in the retroperitoneum or elsewhere in the genitourinary tract. These extrarenal Wilms’ tumors may reflect malignant transformation in similar precursor cells in the genitourinary developmental pathway (Coppes et al., 1991). Wilms’ tumor can arise within a setting of premalignant renal lesions. Persistent metanephric blastema, so-called nodular renal blastema or nephrogenic rests, is seen in a significant number of kidneys harboring Wilms’ tumors, and in virtually all bilateral cases. These lesions may point to a genetic susceptibility that predisposes to tumor formation (Bove and McAdams, 1976). The treatment of Wilms’ tumor has advanced dramatically since the initial report by Wilms in 1899, when it was uniformly fatal (D’Angio et al., 1989; National Wilms’ Tumor Study Committee, 1991; Grundy et al., 1989). Currently, cure rates of 90% are reported by the National Wilms’ Tumor Study Group, involving multimodality treatment. Most early-stage tumors are treated by surgical resection of the affected kidney, exploration of the inferior vena cava, which can be involved with tumor cells, lymph node dissection, and examination of the contralateral kidney for any evidence of synchronous tumor. This is followed by chemotherapy using actinomycin D and vincristine, with radiation therapy reserved for tumors with adverse prognostic indicators. Wilms’ tumor is very sensitive to chemotherapy, and even patients with advanced metastatic disease have an excellent cure rate. Studies of long-term Wilms’ tumor survivors have shown a low incidence of secondary malignancies, consisting primarily of osteochondromas and sarcomas within the radiation field, and acute leukemias attributed to chemotherapy and radiation. Other than these treatment-associated malignancies, there is no convincing evidence that predisposition to Wilms’ tumor also confers susceptibility to other tumor types (Bryd and Levine, 1984). With current medical therapy, most children with Wilms’ tumor reach reproductive age and remain fertile. The risk of progeny similarly affected by Wilms’ tumor is observed to be low (Li et al., 1987). Ill. The Knudson Model Many of the recent advances that have led to the characterization of tumor suppressor genes and their role in carcinogenesis can be

44

DANIEL A. HABER AND DAVID E. HOUSMAN

understood within the framework laid by Knudson in studies of retinoblastoma, Wilms’ tumor, and neuroblastoma (Fig. 2) (Knudson, 1971; Knudson and Strong, 1972a,b). By using a mathematical model to analyze epidemiologic data on the incidence of these pediatric tumors, Knudson calculated the number of rate-limiting steps required for tumorigenesis. In the case of retinoblastoma, Knudson was able to compare the incidence of unilateral versus multiple tumors in patients with a positive family history of this tumor. The data fit the Poisson distribution for a single rare event, suggesting that in these predisposed individuals, only one genetic lesion was required for tumorigenesis. Knudson proposed that these individuals inherited one genetic mutation, and that one additional “genetic hit” in the target tissues led to the development of a retinoblastoma. Given the number of cells at risk for the second genetic event, the likelihood of tumor development is very high and multiple tumors are common. In contrast, sporadic tumors are exclusively unilateral, reflecting the requirement for two independent rare genetic events for tumor formation. The age of incidence of retinoblastoma also supports the Knudson model. Individuals with an inherited predisposition to tumor formation develop these tumors 1 to 2 years earlier than the sporadic cases. Furthermore, the age of onset of new tumors declines at an exponential rate in susceptible individuals, consistent with the exponential rate of differentiation of predisposed retinoblasts. On the other hand, in sporadic cases, the timing of the second genetic lesion is dependent on the variable timing of the initial mutation, thus producing a more delayed decline in the incidence of tumors over time. T h e predictions of the Knudson model in the case of retinoblastoma have been borne out by the cloning of the RBZ gene (Friend et al., 1986; Lee et al., 1987; Fung et al., 1987). This gene maps to a locus on chromosome 13q14, which had been linked to retinoblastoma formation by genetic analyses (Sparkes et al., 1983; Benedict et al., 1983; Cavenee et al., 1983, 1985; Dryja et al., 1984, 1986), and inactivation of the two alleles of RBI appear to comprise the two genetic hits predicted by Knudson (Dunn et al., 1988; Yandell et al., 1989). Wilms’ tumor shares a number of features with retinoblastoma. Five to 10%of cases are bilateral at presentation, and these tend to arise at a younger age than the unilateral tumors (Matsunaga, 1981). Knudson and Strong (1972a) were able to show that the two-hit model was also compatible with the incidence of Wilms’ tumor, but their analysis was limited by the small number of documented cases of familial Wilms’ tumor. With an estimated 8% incidence of bilateral tumors, Poisson sta-

45

THE GENETICS OF WILMS’ TUMOR

4

I*

3

G

FIG. 2. Schematic representation of Knudson model in Wilms’ tumor. T h e model proposed by Knudson and Strong (l972a) predicts that bilateral Wilms’ tumors result from a genetic predisposition. An initial mutation is present in the germline of the child, either as a result of parental transmission o r a de iiouo germline event. Two genetic events are rate limiting in tumorigenesis, the second event typically consisting of the loss of the wild-type allele in somatic tissues. T h e probability of a second somatic event is high in children with genetic susceptibility to Wilms’ tumor, resulting in bilateral tumors and an earlier age of onset. T h e somatic loss of the wild-type allele can occur by diverse mechanisms (chromosome recombination, nondisjunction events, as well as more subtle deletions). In Wilms’ tumors showing allelic losses, the maternal gene appears to be preferentially lost. Of note, the two Wilms’ tumor loci that have been mapped are both on the short arm of chromosome 11, allowing a single chromosomal event to inactivate both of these loci.

46

DANIEL A. HABER A N D DAVID E. HOUSMAN

tistics would predict a 30% incidence of familial cases, rather than the observed 1-2% incidence (Cochran and Froggatt, 1967; Knudson and Strong, 1972a). These observations imply that the majority of bilateral tumors may result from de novo germline events or that the penetrance of the inherited lesion may be variable. The second possibility is less likely, given the high penetrance of Wilms’ tumor in patients with congenital syndromes such as WAGR (see below). An important component of the Knudson hypothesis, as it is now interpreted, is that inactivation of one allele of a tumor suppressor gene is phenotypically silent, and abnormal growth results only following loss of the second allele (Knudson, 1985). This appears to be the case in retinoblastoma: individuals with a germline RBI mutation who do not develop a tumor during the years in which they are at risk have no detectable ocular abnormalities as adults. In contrast, patients with genetic susceptibility to Wilms’ tumor often show nephroblastosis, preneoplastic kidney lesions, even if they do not develop Wilms’ tumor (Bove and McAdams, 1976; Beckwith et al., 1990). These observations suggest that a heterozygous germline mutation is capable of inducing an abnormality in developmental growth. Whether such an initial genetic lesion can enhance the probability of an additional genetic event is unknown. The Knudson model is based on the statistics of “hit kinetics,”and therefore measures only the number of events that are rate limiting in tumor formation. Genetic events that are necessary for tumor formation but have a higher frequency than the rate-limiting steps, or are dependent on these earlier events, will not be detected in this analysis (see Haber and Housman, 1991). In the case of Wilms’ tumor, a number of genetic loci have been implicated in tumor formation, and how these loci interact with each other remains to be elucidated. IV. Genetic Loci Associated with Wilms’ Tumor

The Knudson model, as exemplified by retinoblastoma, suggests that the same locus may be,inactivated in the germline of susceptible individuals and in the somatic tissues from which a tumor arises (Knudson, 1985). Indeed, the study of both germline and tumor material, using karyotype analysis, genetic mapping with molecular markers, as well as clinical observations on patients with congenital abnormalities, led to the identification of the key genetic loci involved in Wilms’ tumorigenesis. Currently there is evidence supporting three distinct loci for Wilms’ tumor, two on the short arm of chromosome 11, and one still unidentified.

THE GENETICS OF WlLMS’ TUMOR

47

A. CHROMOSOME 1lp13 A seminal contribution to understanding the genetics of Wilms’ tumor was made by Miller and co-workers in 1964, who noted the association between Wilms’ tumor and aniridia. Aniridia, malformation or absence of the iris, occurs in 1 in 70,000 children, while Wilms’ tumor arises in 1 in 10,000 children. Despite the rarity of these two conditions, aniridia is detected in 1 in 70 children with Wilms’ tumors, and 1 in 3 children with aniridia develop such tumors. These observations were the first to imply a physical linkage between two genes responsible for two distinct phenotypes, so-called “contiguous gene syndromes” (see Table I). In addition, Wilms’ tumors arising in the context of aniridia are frequently bilateral and develop at an earlier age than sporadic tumors. Based on the Knudson model, these individuals could thus be suspected of carrying a heterozygous germline deletion, affecting both the aniridia and Wilms’ tumor genes. In the development of the eye, a hemizygous state appears sufficient to confer the aniridia phenotype, whereas in the kidney a second mutation may be necessary for the development of Wilms’ tumor. Rare individuals with Wilms’ tumor have been found to have a number of congenital abnormalities in addition to aniridia. These patients have a number of developmental abnormalities of the genitourinary tract, ranging from common conditions such as hypospadias and TABLE I CHARACTERISTICS OF Two CONGENTIAL SYNDROMES ASSOCIATED WITH WILMS’ TUMOR^ Characteristic

WAGR syndrome

Chromosomal locus Wilms’ tumor incidence Associated features

llp13 >50% Aniridia Genitourinary defects Mental retardation

~

Beckwith-Wiedemann syndrome 1 lp15 <5%

Macroglossia Organomegaly/hemihypertrophy Umbilical hernia Neonatal hypoglycemia Additional tumors Adrenocortical carcinoma Hepatoblastoma

~

WAGR and Beckwith-Wiedemann syndromes have been critical in defining the 1 lp13 and 1 lp15 Wilms’ tumor loci, respectively. Gross chromosomal abnormalities have been described in both of these syndromes, with deletions being characteristic of WAGR and segmental duplications being reported in Beckwith-Wiedemann syndrome. At least two distinct but contiguous genes are affected in WAGR syndrome: the WTI gene and the putative aniridia gene. The 1 lp15 locus contains the IGF I1 gene, but its role in Beckwith-Wiedemann and Wilrns’ tumor remains unknown. a

48

DANIEL A. HABER A N D DAVID E. HOUSMAN

cryptorchidism to more extreme abnormalities such as renal hypoplasia or agenesis, horseshoe kidney, ureteral atresia, or bifid ureters (Pendergrass, 1976; Breslow and Beckwith, 1982). Varying degrees of mental retardation are also described in these individuals, leading to the acronym WAGR syndrome (Wilms’tumor, aniridia, genitourinary malformations, mental retardation). The WAGR syndrome has been of particular importance in the genetic analysis of Wilms’ tumor because it is associated with a gross chromosomal deletion involving the short arm of chromosome 11, at band p13 (Riccardi et al., 1978; Francke et al., 1979). This observation provided the first clue to the location of a Wilms’ tumor gene (Fig. 3). T h e genes contributing to the WAGR phenotype could also be studied in patients with more discrete chromosomal abnormalities (Nakagome et al., 1984; Lavedan et al., 1989; Mannens et al., 1991). Based on such analyses, the gene responsible for aniridia could be distinguished from the Wilms’ tumor gene. In two families with hereditary aniridia, a translocation disrupting 1 lp13 was identified, with the translocated chromosome segregating with aniridia (Simola et al., 1983; Moore et al., 1986). Wilms’ tumors did not arise in affected individuals from these families, suggesting that the Wilms’ tumor gene is distinct from the aniridia gene, and hence unaffected by the translocation. The gene responsible for genitourinary malformations appears to be more closely linked to the Wilms’ tumor gene. In patients with Wilms’ tumor and severe genitourinary defects such as in Drash syndrome (pseudohermaphroditism and early renal failure), no gross chromosomal abnormalities are seen (Drash et al., 1970). Germline inactivation of one allele of the Wilms’ tumor gene itself may be directly responsible for more moderate genitourinary abnormalities, such as hypospadia or cryp-

CAT

f

W T1

centromere

AN2

FSHB

telomere

-w

WAGR Deletion FIG. 3. Schematic map of the W A G R region. Deletions of chromosome band 1 lp13 found in W A G R patients are large and cytologically evident. The chromosomal region deleted includes the WT1 gene associated with Wilms’tumor (WTl) and the AN2 locus implicated in aniridia. The W A G R deletion is defined at its centromeric border by the catalase gene (CAT) and at its telomeric boundary by the gene for the p chain of folliclestimulating hormone (FSHB).

THE GENETICS OF WILMS’ TUMOR

49

torchidism, which are seen in some cases of Wilms’ tumor (see below) (Pelletier et al., 1991b). The complex developmental abnormalities manifested by individuals with WAGR syndrome and related disorders result from the hemizygosity of genes within this locus, and the resulting reduction in gene dosage. It is possible, therefore, that the variable severity of the aniridia and genitourinary defects reflects differences in expression levels of the remaining wild-type alleles during critical periods in the development of the target organs. The association of chromosome 1lp13 with Wilms’ tumor is not restricted to germline abnormalities in patients with susceptibility to this cancer. Indeed, as in retinoblastonia, genetic evidence could link the somatic “second hit” to the same locus as the initial germline mutation. Chromosome abnormalities in sporadic Wilms’ tumor can be complex, but frequently involve the short arm of chromosome 11 (Kondo et al., 1984; Douglass et al., 1985; Solis el al., 1988; Wang-Wuu et al., 1990). Molecular analyses can be used to identify genetic changes present in virtually all tumor cells, and therefore linked to the initiation of tumorigenesis. Using polymorphic DNA markers, a number of groups were able to show loss of heterozygosity at markers on the short arm of chromosome 11 in sporadic Wilms’ tumors (Fearon et al., 1984; Koufos et al., 1984; Orkin et al., 1984; Reeve et al., 1984). While some Wilms’ tumors were subsequently shown t o have allelic losses limited to the more distal 1 lp15 locus (see below), some 15520% of tumors appear to have loss of heterozygosity, including l l p 1 3 (Mannens et nl., 1988; Glaser et al., 1989). These observations suggested that loss of the gene residing at the 1lp13 locus is a feature both of tumor predisposition in the germline as well as sporadic mutation leading to tumorigenesis. llp15 B. CHROMOSOME Susceptibility to Wilms’ tumor is associated with a number of other congenital syndromes in addition to aniridia and WAGR. One of these was described by Beckwith (1969) and Wiedemann (1964), consisting of abnormally enlarged organs, including the tongue and abdominal viscera, resulting in an umbilical hernia, and neonatal hypoglycemia. The organomegaly can affect additional organs and can be unilateral, leading to the designation hemihypertrophy. Beckwith-Wiedemann syndrome is associated with an increased risk of pediatric tumors, notably Wilms’ tumor, adrenocortical carcinoma, and hepatoblastoma. These tumors are seen in some 7.5% of patients with Beckwith-Wiedemann syndrome, although the true risk is difficult to ascertain since the penetrance of Beckwith-Wiedemann syndrome is variable, and the clinical

50

DANIEL A. HABER A N D DAVID E. HOUSMAN

diagnosis is often uncertain (Sotelo-Avilaet al., 1980; Wiedemann, 1983). Unlike aniridia and WAGR syndrome, Beckwith-Wiedemann syndrome has been associated with a genetic locus closer to the telomere of chromosome 11, at band p15.5. This is based both on linkage analyses of rare families with this syndrome (Koufos et al., 1989; Ping et al., 1989), as well as gross chromosomal abnormalities that are occasionally seen in the germline of sporadically affected individuals (Waziri et al., 1983; Turleau et al., 1984). Abnoriiial karyotypes in these patients most commonly show duplications of chromosome band 1lp15, although in two cases a ring chromosome has been reported (Romain et al., 1983). In BeckwithWiedemann patients with 1lp15 trisomy, the duplicated chromosome is of paternal origin (Brown et al., 1990), whereas in individuals with the familial syndrome, penetrance is more severe when the disease is transmitted by the mother (Lubinsky et al., 1974; Niikawa et al., 1986; Koufos et al., 1989). Recently, uniparental chromosome 11 isodisomy has been reported in a number of individuals with Beckwith-Wiedemann syndrome (Henry et al., 1991; Grundy et al., 1992). In these patients, analysis with polymorphic markers indicated that the affected child had inherited two copies of chromosome 11 from the father, and none from the mother. Such an unusual genetic mechanism could reflect two underlying molecular events. It is possible that these cases represent transmission of a mutated paternal allele, rendered homozygous in the child’s germline by duplication of the paternal chromosome and loss of the maternal chromosome. Alternatively, these chromosomal events could reflect imprinting events, which have been noted in genes residing on chromosome 1lp15 (See Haig and Graham, 1991). The insulin-like growth factor I1 (IGF 11), which maps to l l p 1 5 in humans, has been shown to be expressed uniquely in paternally derived alleles in the mouse. The introduction of an inactivated IGF I1 into the mouse germline results in small-sized offspring if the disrupted gene is transmitted by the father, but it is phenotypically silent if transmitted by the mother (DeChiara et al., 1990, 1991). If such genomic imprinting were also observed in the human gene, the organomegaly associated with Beckwith-Wiedemann syndrome could be explained by the doubling of IGF I1 expression resulting from the presence of two actively transcribed paternal alleles. The link between IGF I1 and Wilms’ tumorigenesis, however, remains unclear. In one case of Wilms’ tumor, a gross DNA rearrangement affecting the IGF I1 gene was reported (Irminger et al., 1989). In a number of Wilms’ tumors, IGF I1 mRNA levels were found to be elevated (Reeve et al., 1985; Scott et al., 1985), a finding consistent with the high levels of IGF I1 expression in the devel-

THE GENETICS OF WILMS’ TUMOR

51

oping kidney. However, no specific mutations in IGF I1 have been reported in Wilms’ tumors, and genetic analysis of the l l p 1 5 locus supports a “loss of function” as a genetic events, rather than increased expression of a potential tumor gene, It is of interest that in Wilms’ tumors, which show loss of heterozygosity at either the 1lp13 or 1lp15 loci, the maternal allele appears to be preferentially lost (Schroeder et al., 1987; Mannens et al., 1988). This genetic imbalance could result from an increased mutation rate during male gametogenesis, with absence of the maternal allele reflecting loss of the wild-type gene. If, on the other hand, genomic imprinting rather than mutation is invoked as the underlying mechanism, then the putative tumor suppressor gene should be expressed by the maternal allele and imprinted in the paternal COPY. A possible relationship between the 1 lp13 and 1lp15 Wilms’ tumor genes is suggested by molecular studies of tumor specimens. Much as the 11p13 Wilms’ tumor locus appears to be involved in both germline predisposition and somatic mutations, sporadic Wilms’ tumors have been shown to have allelic losses at 1lp15. The demonstration of tumors that have deleted markers within 1lp15, but retained heterozygosity at 1lp13, supports the existence of a distinct tumor suppressor locus (Mannens et al., 1988; Henry et al., 1989; Reeve et al., 1989). It is unclear whether the two Wilms’ loci on chromosome 11 are involved in distinct populations of tumors, or whether they both contribute to tumorigenesis within individual tumors. Beckwith has proposed that the preneoplastic nephrogenic rests associated with Wilms’ tumors differ histologically depending on whether they arise in the setting of WAGR syndrome (1 lp13) or Beckwith-Wiedemann syndrome (1lp15) (Beckwith et al., 1990). Wilms’ tumors linked to abnormalities at the 1lp13 genetic locus contain nephrogenic rests predominantly localized within the kidney lobules (intralobar), while those associated with the 1lp15 disease locus have preneoplastic lesions at the periphery of the kidney lobules (perilobular). Such an observation implies that developmental arrest at different stages in kidney development could be associated with distinct genetic lesions. On the other hand, individual tumors have been described in which genetic evidence points to involvement of both chromosome 11 loci. In one case, a patient with a germline 1lp13 deletion in the setting of WAGR syndrome developed a Wilms’ tumor which showed loss of heterozygosity at 1lp15 but not 1lp13 (Henry et al., 1991). In another case, a sporadic tumor showed evidence of chromosome 11 loss and reduplication prior to a mutation at the 1lp13 gene (Haber et al., 1990). These observations suggest potential interactions between the two chromosome 11 Wilms’ loci.

52

DANIEL A. HABER A N D DAVID

E.

HOUSMAN

Unlike the l l p 1 3 locus, which has been linked almost exclusively to the development of Wilms’ tumor, the chromosome l l p 1 5 locus has been associated with a number of tumor types. Adrenocortical carcinoma, breast cancer, lung cancer, and acute myelogenous leukemia have been shown to have reduction to homozygosity at the H-ras or globin markers within 1lp15 (Ali et al., 1987; Henry et al., 1989; Weston et al., 1989; Ahuja et al., 1990). These observations suggest that the putative tumor suppressor gene on chromosome 11p 15 may be involved in the normal development of a wide range of different tissues, in apparent contrast to the 1lp13 gene (see below). C. FAMILIAL WILMS’TUMOR Familiar susceptibility to Wilms’ tumor is rare, with only 1% of cases having evidence of hereditary transmission (Cochran and Froggat, 1967; Knudson and Strong, 1972). This is in contrast with retinoblastoma, in which the high number of familial cases is consistent with the incidence of bilateral tumors (Knudson, 1971). To date, one case report has identified the transmission of a mutated 1 lp13 gene in a family with hereditary Wilms’ tumor, and two cases with bilateral tumors have been shown to have de novo germline mutations in the 11p 13 gene (Huff et al., 1991; Pelletier et al., 1991b) (see below). However, the analysis of three large family pedigrees with Wilms’ tumor has pointed to the existence of another Wilms’ tumor locus, distinct from both the I lp13 and 1 lp15 genes. In these studies of families with Wilms’ tumor in the absence of gross congenital abnormalities, genetic linkage excluded the chromosome 11 loci from transmission of disease susceptibility (Grundy et al., 1988; Huff et al., 1988; Schwartz et al., 1991). These findings suggest the existence of a third Wilms’ locus, as yet unmapped, which is responsible for these cases of hereditary susceptibility. T h e genetic locus involved in familial Wilms’ tumor may interact with the two other Wilms’ loci in a number of ways. The putative familial Wilms’ tumor gene may be a third disease gene, inactivated in both the germline of affected individuals and in somatic tissues, leading to tumor development independently of the 11p 13 and 11p 15 genes. Alternatively, a germline heterozygous mutation in this gene could confer tumor susceptibility without requiring loss of the wild-type allele as a second mutational event. It is interesting to note that polymorphic markers at the chromosome 5q familial polyposis coli locus do not show reduction to homozygosity in the polyps of affected individuals, suggesting that predisposition to these premalignant lesions is conferred by the heterozygous germline lesion (see Fearon and Vogelstein, 1990). By

THE GENETICS OF WILMS’ TUMOR

53

analogy with polyposis coli, the putative familial Wilms’ tumor locus could also interact with the other Wilms’ genes in contributing to tumor initiation. In support of this notion is a tumor from one individual with familial Wilms’ tumor that was found to have lost heterozygosity at the 1lp15 locus, implying that inactivation of this second Wilms’ locus can contribute to the tumorigenic process initiated by the familial gene (Grundy et al., 1988).

V. Isolation of the WTI Gene at 11p13 T h e identification of the Wilms’ tumor gene at 1lp13 was the culmination of many years of effort in many laboratories. These studies included both the generation of somatic cell hybrids and the analysis of patients with chromosomal deletions and translocations. An initial crucial reagent in the mapping of 1lp13 was the generation of hybrid cell lines containing fragments of human chromosome 11 in a hamster cell background. T h e initial hybrid cell, J 1, maintained the entire human chromosome 11 in the absence of selection, apparently complementing the glycine requirement of the parental hamster cell (Kao et al., 1976; Gusella et al., 1982). T h e presence of a cell surface antigen, MIC 1, mapping to 1lp13 made it possible to select derivative hybrid cells retaining this segment of chromosome 11, following X irradiation of the parental cells by the Goss-Harris technique (Goss and Harris, 1975). One such derivative cell line, Goss-Harris 3A, retained only some 3 megabases of DNA, encompassing the WAGR region (Glaser et al., 1987, 1990a). Hybrid cell lines containing small amounts of human genomic DNA from the 11p region were also generated by chromosome-mediated gene transfer (Porteous et al., 1087). Cells from patients with WAGR deletions proved to be particularly useful in the generation of somatic cell hybrids. T h e development of these hybrids allowed the segregation of the deleted chromosome 11 from its normal counterpart, providing valuable mapping reagents (Davis et al., 1988; Couillin et al., 1989; Gessler et al., 1989). By analyzing DNA from hybrid cell lines using pulsed field gel electrophoresis, a number of groups were able to generate detailed physical maps of 1 lp13 and the WAGR region (Compton et al., 1988; Davis et al., 1988, 1990; Gessler and Bruns, 1989; Glaser et al., 1989; Rose et al., 1990). T h e molecular analysis of clinical material from patients with aniridia or WAGR syndrome provided a complementary approach to the somatic hybrid data. Much as biochemical studies of the esterase D enzyme initially linked the retinoblastoma locus to chromosome 13q14 (Benedict et al., 1983), the erythrocyte catalase (CAT) enzyme proved critical in the

54

DANIEL A. HABER AND DAVID E. HOUSMAN

initial mapping of the WAGR deletion. Junien and co-workers (1980) showed that levels of catalase activity were reduced in individuals with WAGR syndrome, indicating that this gene was located within the 11 p 13 region hemizygously deleted in these patients. This gene was subsequently shown to flank the WAGR deletion on the centromeric side (van Heyningen et al., 1985). The gene encoding the p subunit of folliclestimulating hormone (P-FSH) was also found to be hemizygously deleted in WAGR patients and to mark the telomeric extent of this deletion (Glaser et al., 1986). Within the WAGR deletion, the Wilms’ tumor locus could be mapped with respect to the aniridia locus by the analysis of smaller chromosomal deletions associated with Wilms’ tumor but not aniridia, and which placed the WT locus centromeric of the aniridia locus (Davis et al., 1988). Whereas germline karyotype analysis of individuals with Wilms’ tumor syndromes was critical in establishing the map of the markers within 1 lp13, these chromosome abnormalities were large. The discovery of a sporadic Wilms’ tumor with two nested deletions within 1lp13 (Lewis et al., 1988) made it possible to identify a genomic fragment of 350 kb that was homozygously deleted in this tumor (Rose e l al., 1990), and within which the 1 lp13 gene must reside. Using a library of human genomic DNA derived from a somatic cell hybrid, we identified clones that were homozygously deleted in this tumor, and which showed cross-species hybridization, consistent with the presence of an evolutionarily conserved transcription unit. These genomic clones were then used to isolate a series of cDNAs, encoding a gene we now call W T l (Call et al., 1990).This gene was independently isolated by another group of investigators, who used a chromosomejumping cloning technique based on the presence of CpG islands known to occur at the 5’ end of many transcription units (Gessler et al., 1990). A distinct transcript that is adjacent to W T I , Wit-1, has also been isolated (Bonetta et al., 1990; Huang et al., 1990). This gene has an open reading frame of only 276 base pairs, and it is currently unclear whether it encodes a functional polypeptide.

VI. WT7: Characterization of a Novel Tumor Suppressor Gene A. GENESTRUCTURE AND ALTERNATIVE SPLICING

The WTZ gene spans 50 kb of DNA and is encoded by 10 exons (Call et al., 1990; Haber et al., 1991). The W T l transcript is 3 kb, with a predicted polypeptide of 46-49 kDa, reflecting the presence or absence of two alternative splices (see Fig. 4). Two potential functional domains

55

THE GENETICS OF WILMS’ TUMOR

aItcrui1vc splice I

a l t c r u t i v c splice I1

FIG.4. Functional domains of the WTl gene product. The predicted protein encoded by WTI contains a proline-rich domain at the amino terminus and four Cys-His zinc finger regions at the carboxy terminus (Call et al., 1990; Gessler et al., 1990). The zinc finger domains have a high degree of amino acid homology with those of the early growth response genes, EGR 1 and 2. Two alternative splices are present in the WTl transcript: splice I inserts 17 amino acids between the proline-rich domain and the first zinc finger, while splice I1 inserts 3 amino acids in the knuckle between zinc fingers 3 and 4. Four distinct mRNA species are present in all cell types that express W T l . These have been defined as follows: splice variant A (no alternative splice present), splice variant B (splice I present, splice I1 absent), splice variant C (splice I absent, splice I1 present), and splice variant D (both splices present). Splice variant D is the most prevalent WTI transcript, and splice variant A the least common (Haber et al., 1991).

are apparent on the basis of amino acid homology: the carboxy terminus contains four zinc finger domains of the cysteine-histidine type, and the amino terminus includes a stretch of amino acids extremely rich in prolines and glutamines. Zinc finger domains have been described in a number of transcription factors and have been shown to mediate DNA recognition (see Evans and Hollenberg, 1988). The WT1 zinc finger domains have a striking amino acid homology with those of the early growth response genes (EGR 1 and 2) (Sukhatme et al., 1988; Joseph et al., 1988), and in fact the polypeptide encoded by the WT1 transcript lacking alternative splice I1 has been shown to bind the EGR 1 DNA recognition site with high affinity (Rauscher et al., 1990). The EGR genes are expressed in a number of tissues (including the renal tubule), and are induced following mitogenic stimuli such as serum stimulation. Homology to the WT1 zinc finger domains has also been reported for the yeast gene MZGl, which is involved in the regulation of genes involved in sugar metabolism (Nehlin and Ronne, 1990). The role of the prolinerich domain of WT1 is not as well defined as that of the zinc finger region. Such domains are found in other transcription factors, including the prototype transcription factor Kruppel, and may be involved in mediating the trans-activation signal (see Mitchell and Tijan, 1989). Recent evidence has suggested that the amino terminus of the WTl protein has properties of a transcriptional repressor, following binding of the protein to the EGR 1 promoter (Madden et d., 1991). However, the in vivo targets of WT1 have yet to be defined.

56

DANIEL A. HABER AND DAVID E. HOUSMAN

Four WTI mRNA species can be distinguished, reflecting the presence or absence of two alternative splices (Haber et al., 1991). Splice I is encoded by a single exon and inserts 17 amino acids between the proline-rich amino terminus and the first zinc finger domain. These amino acids include five serines and one threonine, suggesting that the alternatively spliced domain may be a target for protein phosphorylation. Alternative splice I1 results from the use of an alternative splice donor site and introduces three amino acids (lysine, threonine, serine) in the “knuckle” between zinc fingers 3 and 4. Insertion of splice 11 disrupts the high degree of homology between the WTl and EGR 1 zinc fingers and abolishes sequence-specific binding to the EGR 1 DNA consensus (Rauscher et al., 1990). It is unclear, however, whether the zinc finger domain containing splice 11 has a distinct DNA recognition site, or whether it acts to block DNA binding by the alternative zinc finger domain. The gene products of all four alternative splice variants are localized in the cell nucleus following transfection into Cos cells (Pelletier et al., 1991a). All four W T 1 splice variants are found in tissues that express WT 1, both in humans and in the mouse (Haber et al., 1991). The relative ratio of these splice forms is remarkably constant in these tissues as a function of development, and in normal kidney tissue and Wilms’ tumors, with splice form D (both alternative splices present) 5 to 10-fold more prevalent than splice form A (both splices absent), and the other species (splice forms B and C) intermediate in prevalence. The role of these splice variants in WTl function has yet to be clarified, but the remarkable level of conservation, both in sequence and relative proportion, suggests that each form contributes to the normal function of WT1.

B. NORMAL TISSUE EXPRESSION OF WTI Normal tissue expression of W T l is restricted to a small number of organs and may shed light on the normal role of this gene during differentiation. This expression pattern is in marked contrast with the wellstudied tumor suppressor genes RBI and p53, which are expressed in virtually all tissues (Friend et al., 1986; Lee et al., 1987; Bernards et al., 1989) and appear to affect basic cellular mechanisms such as progression through the cell cycle (Buchkovich et al., 1989; Chen et al., 1989; DeCaprio et al., 1989; Ludlow et al., 1989). In the mouse and in the baboon, W T l mRNA is readily detectable by Northern blot in kidney, spleen, gonads, and uterus (Call et al., 1990; Buckler et al., 1991; Pelletier et al., 1991). Low levels of expression can also be demonstrated in thymus, heart, and lung (Buckler et al., 1991). In the kidney, the developmental

57

THE GENETICS OF WILMS' TUMOR

time course of expression is dramatic (see Fig. 5). In the mouse, WTI expression is detectable at day 8 of gestation, rises to a peak between day 17 and birth, and then rapidly declines to low adult levels of expression by day 17 (Buckler et al., 1991). In human tissues, high levels of WTI expression have been demonstrated in fetal tissue (day 20), while adult human kidney tissue does not contain detectable W T l mRNA (Haber et al., 1990; Pritchard-Jones et al., 1990). RNA in .ritu hybridization studies show that in the developing kidney, WT1 expression is restricted to the condensed mesenchyme, renal vesicle, and glomerular epithelium

BIRTH

1

I el3 el7 0

3

8

15

24

37-->Adult

Age (days)

Fic. 5. Developmental time course of W T l expression. T h e time course of WTI expression has been studied in the mouse kidney (Buckler et nl., 1991). W T l mRNA is detectable by Northern blot by day 8 of gestation. T h e level of expression peaks around the time of birth, and then declines rapidly to adult levels. In the human kidney, WTI is highly expressed at 20 weeks, and is n o longer detectable in the adult organ (Haber et al., 1990; Pritchard-Jones et al., 1990). Other tissues with high W T l mRNA levels, such as the gonads, d o not show this dramatic pattern of expression. Instead, WTI expression increases during fetal development and remains high in the adult tissue (Pelletier ef al., 1991).

58

DANIEL

A.

HABER AND DAVID E. HOUSMAN

(Pritchard-Jones et al., 1990). In other tissues, W T l expression is also limited to specific cell types. In the ovary, W T l is expressed in the granulosa cells lining the follicles, while in the testis, expression is limited to the Sertoli cells. In the gonads, however, W T l expression increases during development, and then remains elevated during adult life, rather than peaking sharply as it does during kidney development (Pelletier et al., 1991a).

C. W T l MUTATIONSIN THE GERMLINE A N D I N WILMS’ TUMORS The majority of Wilms’ tumors express high levels of WTI mRNA, consistent with malignant cells thought to originate in the fetal kidney (Haber et al., 1990). W T l mRNA is detectable by i n situ hybridization primarily in the blastemal cells and epithelial cells of the tumors. The stromal cell component does not express W T l , and tumors with a predominant stromal component express low levels of W T l , as do anaplastic tumors (Pritchard-Jones et al., 1990; Zabel et al., 1992; Gerald et al., 1992). As predicted by the Knudson model, mutations in WTI can be demonstrated within Wilms’ tumors. While DNA rearrangements detectable by Southern blotting and transcripts of grossly altered size are rare in Wilms’ tumors, small deletions within the W T l gene can be identified by more sensitive polymerase chain reaction (PCR)-based techniques (Haber et al., 1990, 1991). To date, Wilms’ tumors have been found to have homozygous deletions that extend from adjoining DNA into the 5‘ exons of the gene (Ton et al., 1991) or the 3’ exons of the gene (Cowell et al., 1991), as well as small mutations that are entirely contained within W T l (Haber et al., 1990; Huff et al., 1991; Pelletier et al., 1991b) (Fig. 6). The presence of these mutations clearly identifies WTI as a gene specifically inactivated in Wilms’ tumor and supports its identification as a tumor suppressor gene. Not all Wilms’ tumors appear to have mutations in W T I , however, suggesting that the genetic complexity of Wilms’ tumor may be reflected in the different genetic lesions that lead to tumorigenesis. To date, we have found one in eight sporadic Wilms’ tumors to have a mutation in WTI (Haber et aE., 1990). Homozygous mutations in W T l do not appear to be the only mechanisms by which this gene contributes to tumorigenesis. For instance, one sporadic tumor was found to have a normal W T l allele along with a mutated gene (Haber et al., 1990). The mutation, a 25-base pair deletion, resulted in loss of both the third zinc finger domain and alternative splice I1 from the encoded polypeptide. The mutation, however, maintained a distal open reading frame, thus encoding a viable but shortened

-

59

THE GENETICS OF WILMS’ TUMOR

A

B

D

+*C

U

u-

E

alternative

/

AT6

h

proline rich

splices

\

\

J

4 1

4 2

TGA

c

4 . C 3

4

zinc fingers

Fic. 6. Mutations affecting the WTI gene. W T I is encoded by 10 exons, spanning 50 kb of genomic DNA. Alternative splice I is encoded by a discrete exon, while alternative splice I1 results from the use of either of two splice donor sequences in exon 9. A number of Wilms’ tumors have been found to have mutations involving W T I , including the following: A, homozygous deletion extending into exon 1 (Ton et al., 1991); B, homozygous mutation in exon 4 (Pelletier et al., submitted); C, homozygous deletion of exon 6 (Huff et al., 1991); D, heterozygous deletion within exon 9 (Haber et al., 1990); E, homozygous deletion extending into exon 10 (Cowell et al., 1991).

polypeptide, with reduced DNA binding affinity (Rauscher et al., 1990). The presence in a tumor of both wild -type and mutated W T l alleles suggests the possibility that the altered W T l gene product is able to suppress the function of the normal protein, by a “dominant negative” mechanism (see Herskowitz, 1987), as has been demonstrated for p53 (Finlay et al., 1989; Lavigueur et al., 1989) and v-Erb A (Damm et al., 1989). We have recently demonstrated that this mutated W T l allele, encoding an altered DNA binding domain, can cooperate with the viral oncogene E1A in transforming baby rat kidney cells (Haber et al., 1992). T h e existence of dominant negative mutations in W T l implies that, in some cases, a single genetic “hit” is sufficient to inactivate WT1 function. Such a specific mutation, however, would be far more infrequent than the simple inactivating mutations which are the basis of the Knudson “two hit” model. The Knudson model predicts that individuals with bilateral Wilms’ tumors carry a germline mutation, thus ensuring that one second somatic mutational event is sufficient to induce tumorigenesis. Indeed, in one patient with bilateral Wilms’ tumors, a heterozygous deletion internal to W T l could be demonstrated in the patient’s germline, with each tumor having lost the remaining normal W T l allele by a different mechanism (Huff et al., 1991). In this patient, the mutation was a de nouo germline event, with neither parent carrying the mutation. Familial transmission

60

DANIEL A. HABER AND DAVID E. HOUSMAN

of a mutated WTI gene does exist, although it appears not to be as common as predicted based on the retinoblastoma model. A child with a unilateral Wilms’ tumor and genitourinary abnormalities (hypospadias and cryptorchidism) was found to have a germline WTI mutation that was transmitted from his father (Pelletier et al., 1991b). The boy’s father had also recovered from a unilateral Wilms’ tumor, but did not show any genitourinary defects. This case, and that of a second child with bilateral Wilnis’ tumors and similar genitourinary defects also associated with a germline W T l mutation, suggests that constitutional inactivation of one WTI allele may be directly responsible for genitourinary defects (Pelletier et al., 1991). Since genitourinary abnormalities are relatively common in the general population, more patients will have to be analyzed before a firm conclusion can be reached that the W T l gene is solely responsible for all the genitourinary abnormalities seen in WAGR syndrome (van Heyningen et al., 1990). T h e variable penetrance of these genitourinary defects, as well as their relative infrequency in female patients, suggests that individuals who are hemizygous for the WT1 gene may have variable levels of expression from the remaining wild-type allele. Alternatively, the variable expressivity of the genitourinary abnormalities may reflect a level of redundancy in this complex developmental system. T h e potential disruption of normal genitourinary development by a mutated germline W T l allele is best demonstrated in Drash Syndrome. Children with this condition suffer from pseudohermaphroditism, early renal failure, and a high risk of Wilms’ tumor. WTI mutations in Drash Syndrome appear to be highly specific, with amino acid substitutions either in zinc finger 2 or 3 (Pelletier et al., 1991~).T h e developmental abnormalities conferred by these Drash mutations are far more severe than those resulting from simple inactivation of one W T l allele, and hence are suggestive of a dominant negative effect, resulting in more complete suppression of WT1 function. Individuals who are predisposed to Wilms’ tumor by virtue of a germline WTI mutation do not appear to have an increased risk of other malignancies, other than those induced by chemotherapy or radiation therapy (Bryd and Levine, 1984). W T l mutations therefore appear to be rate limiting only in Wilms’ tumor development, although it is possible that they contribute to tumorigenesis in other tissues. In an analogous situation, germline RBI mutations confer predisposition only to retinoblastoma and osteosarcoma, but a diverse group of sporadic tumor types can be shown to contain lesions in R B I . Tumors derived from the small number of tissues in which W T l is normally expressed during development are currently being analyzed for evidence of W T l mutations. To

THE GENETICS OF WILMS’ TUMOR

61

date, however, W T l mutations have been detected only in Wilms’ tumor specimens.

VII. Functional Studies and Animal Models Much as the Knudson hypothesis provided the conceptual framework leading to the isolation of tumor suppressor genes, somatic cell fusion experiments first suggested that suppression of cell growth was a dominant phenotype. T h e fusion between malignant cells and normal fibroblasts resulted in the stable loss of tumorigenicity (Harris et al., 1969; Stanbridge, 1976). In these experiments, the ability to form tumors in nude mice, which was lost following cell fusion, could be distinguished from other criteria of transformed cell growth, which were unaffected (Stanbridge and Wilkinson, 1978). T h e use of microcell-mediated chromosome transfer allowed a more detailed genetic analysis of the tumor suppression phenotype. Weissman and co-workers (1987) used a Wilms’ tumor-derived cell line, G40 1, which was deficient in hypoxanthine phosphoribosyltransferase (HPRT), to introduce a t(X; 11) translocation chromosome. T h e X chromosome provided the HPRT gene, thus allowing for drug selection, while chromosome 11 contained both the 1 lp13 and 1lp15 Wilms’ loci (1 lq23-11 pter). Tumorigenicity of the hybrid cells was markedly diminished, although no changes were detected in other parameters, such as soft-agar cloning efficiency. The specificity of this tumor suppressor effect was demonstrated by the inability of the X chromosome alone, or chromosome 13 (carrying the RBI gene), to affect tumorigenicity (Weissman et al., 1987). Similarly, tumor suppression by chromosome 11 could be demonstrated only in certain tumors, such as cervical carcinoma and rhabdomyosarcoma, but not in others, such as adult renal carcinoma and carcinogen-induced rat nephroblastoma (Oshimura et al., 1990). Tumor suppression studies involving the introduction of an entire chromosome 11 cannot distinguish between the two Wilms’ tumor loci that are present on the short arm. The highly anaplastic G401 cell line does not express detectable levels of W T l mRNA and does not have any gross lesions within this gene (unpublished data). It is therefore possible that the tumorigenicity of this cell line does not result from a WTI mutation, but rather may reflect a defect in the 1lp15 locus. Indeed, a recent report has linked the reversal of tumorigenicity in G401 cells to a chromosomal fragment including 1lp15, not 1 1p13 (Dowdy et al., 1991). Studies are currently underway to determine if the W T l gene itself is capable of reversing tumorigenicity in an appropriate Wilms’ tumor cell line. Animal models of Wilms’ tumor may offer the potential for studying

62

DANIEL A. HABER AND DAVID E. HOUSMAN

the development of this malignancy in an experimental system. The best studied animal models are the rat nephroblastomas (Hasgekar et al., 1989; Ohaki, 1989). These tumors may arise either spontaneously, or following induction by X irradiation o r by transplacental carcinogens such as n-ethylnitrosourea. While these rat nephroblastomas have been used as models for chemotherapy, they do differ from the typical human nephroblastomas in being composed primarily of mesenchymal elements with few blastemal cells (Deshpande et al., 1989). Perhaps the closest animal model of Wilms’ tumor is a spontaneous nephroblastoma in the Japanese eel, which demonstrates the classic triphasic histology, along with occasional muscle differentiation (H.I.H. Prince Masahito, personal communication). The role of W T l in the etiology of these animal Wilms’ tumors is currently unknown. In the mouse, kidney tumors histologically similar to human Wilms’ tumor are exceedingly rare. Mouse models may be attempted by transgenic techniques, but an existing mutant mouse strain suggests a different level of genetic complexity in the mouse. The small eye mouse mutation appears to be analogous to the aniridia phenotype in humans (Theiler et al., 1978; Hogan et al., 1986), and it appears to map within mouse chromosome 2, in a region syntenic with human chromosome 1lp13 (Glaser et al., 1990b). One variant of this mutation is the Sey/Dey strain, which carries a heterozygous deletion analogous to the human WAGR deletion, including the mouse homolog of W T l (Glaser et al., 1990b; Buckler et al., 1991). T h e heterozygous Sey/Dey phenotype includes small eyes, a small body size, and a white belly patch, while homozygosity for the Sey/Dey allele appears to be embryonic lethal. Despite the similarity between this mutant mouse strain and WAGR patients, Sey/Dey mice do not develop nephroblastomas, nor do they have any detectable abnormalities of renal or genital development. This discrepancy is all the more striking, given the extensive homology between the human and mouse W T l genes, both in terms of amino acid composition and developmental expression (Buckler et al., 1991). It is possible that the smaller number of target cells in the mouse compared with the human kidney makes it less likely that a Sey/Dey mouse would acquire the “second hit” in the W T l gene required for Wilms’ tumorigenesis. Alternatively, the mouse kidney developmental pathways may have more redundancy than the human, thus allowing for silent mutations in W T l . An intriguing observation is that, in humans, two Wilms’ loci are located on the short arm of chromosome 11, making it possible for a single chromosomal recombinational event to cause loss of both genes. In the mouse, however, two separate genetic events would be required since the l l p 1 3 locus maps to mouse chromosome 2, while l l p 1 5 is

THE GENETICS OF WILMS’ TUMOR

63

syntenic with mouse chromosome 7 (Glaser et al., 1990b). Thus, while the mouse Sey/Dey mutation and the human WAGR syndrome appear to be genetically similar, the dramatically different phenotype between mouse and human may underlie a major functional divergence between the two organisms. Understanding these differences may well provide insight into the mechanism of Wilms’ tumorigenesis in humans.

VIII. Conclusions The genetic events underlying the development of Wilms’ tumor are complex, and the isolation of the W T l gene has provided an initial molecular key to decipher this process. The identification of the 1 lp15 and familial Wilms’ tumor genes should allow a more complete understanding of the pathways involved in Wilms’ tumorigenesis and their potential interactions. Wilms’ tumor presents a genetic pattern that may be intermediate in complexity, between the single-locus disease exemplified by retinoblastoma (Friend et al., 1988) and the multistep tumorigenesis model proposed for colon cancer and other adult tumor types (Fearon and Vogelstein, 1990). Nonetheless, the presence of clearly defined genetic syndromes that confer predisposition to Wilms’ tumor have made it possible to dissect the critical genetic loci that are tied to the initiation of tumorigenesis. The clinical and molecular approach to study these individuals with rare genetic defects not only led directly to the isolation of the WTZ gene, but also provided initial evidence that deletion of this gene can be the initial event leading to Wilms’ tumor. In Wilms’ tumor, as in retinoblastoma, the prophetic model that Knudson and Strong proposed based on epidemiologic analyses has stood the test of time, and continues to provide a unifying theme for the molecular study of human neoplasia. REFERENCES Ahuja, H. G., Foti, A., Zhou, D. J., and Cline, M. J. (1990). Blood 75, 819-822. Ali, 1. U., Liderau, R., Theillet, C., and Callahan, R. (1987).Science 238, 185-188. Beckwith, J. B. (1969). Birth Defects, Orig. Artic. Ser. 5 , 188-196. Beckwith, J. B., Kiviat, N. B., and Bonadio, J. F. (1990). Pediatr. Pathol. 10, 1-36. Benedict, W. F., Murphree, A. L., Banerjee, A., Spina, C. A , , Sparkes, M. C., and Sparkes, R. S. (1983). Science 219, 973-975. Bennington, J. L., and Beckwith, J. P. (1975). “Atlas of Tumor Pathology,” 2nd Ser., Fasc. 12. Armed Forces Institute of Pathology, Washington, D.C. Bernards, R., Schackleford, F., Gerber, M., Horowitz, J., Friend, S., Schartl, M., Bogenmann, E., Rapaport, J., McGee, T., Dryja, T., and Weinberg, R. (1989). Proc. Nutl. Acad. Sci. U.S.A. 86, 6474-6478.

64

DANIEL A. HABER AND DAVID E. HOUSMAN

Bonetta, L., Kuehn, S. E., Huang, A., Law, D. J., Kalikin, L. M., Koi, M., Reeve, A. E., Brownstein, B. H., Yeger, H., Williams, B. R. G., and Feinberg, A. P. (1990). Science 250, 994-997. Bove, K. E., and McAdams, A. J. (1976). Perspect. Pediatr. Pathol. 3, 185-223. Breslow, N. E., and Beckwith, J. B. (1982). JNCI, J. Natl. Cancer Inst. 68, 429-436. Breslow, N. E., Churchill, G., Nesmith, B. S., Thomas, P. R. M., Beckwith, J. B., Othersen, H. B., and DAngio, G. J. (1986). Cancer (Philadelphia) 58, 2501-2511. Brown, K. W., Williams, J. C., Maitland, N. J., and Mott, M. G. (1990). A m . ] . Hum. Genet. 46, 1000-1001. Buchkovich, K., Duffy, L. A., and Harlow, E. (1989). Cell (Cambridge, Mass.) 58, 1097--1105. Buckler, A. J., Pelletier, J., Haber, D. A., Glaser, T., and Housman, D. E. (1991). Mol. Cell. BZol. 11, 1707-1712. Bryd, R. L., and Levine, A. S. (1984). In “Wilms’Tumor: Clinical and Biological Manifestations” (C. Pochedly and E. S. Baum, eds.), pp. 347-360. Elsevier, New York. Call, K. M., Glaser, T., Ito, C. Y., Buckler, A. J., Pelletier, J., Haber, D. A,, Rose, E. A,, Kral, A., Yeger, H., Lewis, W. H., Jones, C., and Housman, D. E. (1990). Cell (Cambridge, MUSS.)60, 509-520. Cavenee, W. K., Dryja, T. P., Phillips, R., Benedict, W., Codbout, R., Gallie, B., Murphree, A,, Strong, L., and White, R. (1983). Nature (London) 305, 779-784. Cavenee, W. K., Hansen, M., Nordenskjold, M., Kock, E., Maumenee, I., Squire, J., Phillips, R., and Gallie, B. (1985). Science 228, 501-503. Chen, P.-L., Scully, P., Shew, J.-Y., Wang, J. Y. I., and Lee, W.-H. (1989). Cell (Cambn’dge, Mass.) 58, 1193-1198. Cochran, W., and Froggatt, P. (1967).J. Urol. 97, 216-220. Compton, D.A., Weil, M. M., Jones, C. A., Riccardi, V. M., Strong, L. C., and Saunders, C. F. (1988). Cell (Cambridge, Mass.)55, 827-836. Coppes, M. J., Wilson, P. C. G., and Weitzman, S. (1991).J. Clin. Oncol. 9, 167-174. Couillin, P., Azoulay, M., Henry, I., Ravise, N., Grisdrd, M. C., Jeanpierre, C., Barichard, F., Metezeau, F., Chandelier, J. J., Lewis, W., van Heyningen, V., and Junien, C. (1989). Hum. Genet. 82, 171-178. Cowell, J. K., Wadey, R. B., Haber, D. A., Call, K. M., Housman, D. E., and Prichard, J. (1991). Oncogene 6, 595-599. Damm, K., Thompson, C. C., and Evans, R. M. (1989). Nature (London) 339, 593-597. DAngio, G. J., Breslow, N., Beckwith, J. B., Evans, A,, Baum, E., DeLorimier, A., Fernbach, D., Hrabovsky, E., Jones, B., Kelalis, P., Othersen, H. B., Tefft, M., and Thomas, P. R. M. (1989). Cancer 64, 349-360. Davis, L. M., Byers, M. G., Fukushima, Y., Qin, S., Nowak, N. J., Scoggin, C., and Shows, T. B. (1988). Genomics 3, 264-272. Davis, L. M., Senger, G., Ludecke, H. J., Claussen, U., Hortshemke, B., Zhang, S. S., Metzroth, B., Hohenfellner, K., Zabel, B., and Shows, T. B. (1990). Proc. Natl. Acad. Sci. U.S.A. 87, 7005-7009. DeCaprio, J. A,, Ludlow, J. W., Lynch, D., Furukawa, Y., Griffin, J., Piwnica-Worms, H., Huang, C.-M., and Livingston, D. M. (1989). Cell (Cambridge, Mass.) 58, 1085-1095. DeChiara, T. M., Efstradiadis, A., and Roberson, E. J. (1990). Nature (London) 3 4 5 , 7 8 4 0 . DeChiara, T. M., Robertson, E. J., and Efstradiadis, A. (1991). Cell (Cambridge, Mass.) 64, 849-859. Deshpande, R. B., Hasgekar, N. N., Chitale, A. R., and Lalitha, V. S . (1989). Pedzatr..Pathol. 9, 141-151. D o u g h s , E. C., Wilimas, J. A., Green, A. A., and Look, A. T. (1985). Cancer Genet. Cytogenet. 14, 331-338.

THE GENETICS OF WILMS’ TUMOR

65

Douglass, E. C., Look, A. T., Webber, B. Parham, D., Wilimas, J. A., Green, A. A,, and Robertson, P. K. (1986).J. Clin. Oncol. 4, 975-981. Dowdy, S. F., Fasching, C. L., Aravjo, D., Lai, K.-M., Livanos, E., Weissman, B. E., and Stanbridge, E. J. (1991). Science (in press). Drash, A., Sherman, F., Hartmann, W. H., and Blizzard, R. M. (1970).5.Pedintr. 76,585-593. Dryja, T. p., Cavenee, W., White, R., Rapaport, B., Petersen, R., Albert, D., and Bruns, G. (1984). N . Engl. J . Med. 310, 550-553. Dryja, T. P., Rapaport, J., Joyce, J., and Petersen, R. (1986).Proc. Nntl. Acad. Sci. U.S.A. 83, 739 1-7394. Dunn, J., Phillips, R., Becker, A., and Gallie, B. (1988). Sczence 241, 1797-1800. Evans, R., and Hollenberg, S . (1988). Cell (Cambridge, Matr.) 52, 1-3. Fearon, E. R.,and Vogelstein, B. (1990).Cell (Cnnihridge,Mass.) 61, 759-767. Fearon, E. R., Vogelstein, B., and Feinberg, A. (1984). Nature (London) 309, 176-178. Finlay, C. A., Hinds, P. W., and Levine, A. J. (1989). Cell (Canibdgr, mas^.) 57, 1083-1093. Francke, U., Holmes, L. B., Atkins, L., and Riccardi, V. M. (1979). C~togenet.Cell Geiiet. 24, 185-192. Friend, S. H., Bernards, H . R., Rogelj, S., Weinberg, R. A., Rapaport, J. M., Albert, D. M., and Dryja, T. P. (1986). Natuw (London) 323, 643-646. Friend, S. H., Dryja, T. P., and Weinberg, R. A. (1988). N . Engl. J. Med. 318, 618-622. Fung, Y. D. T., Murphree, A. L., T A n g . A., Quian, J., Hinrichs, S. H., and Benedict, W.F. (1987). Science 236, 1657-1661. Gerald et al. (1992). Submitted for publication. Gessler, M., and Bruns, G. A. P. (1989). C;enomirs 5, 43-55. Gessler, M., Thomas, G. H., Couillin, P., Junien, C . , McFillivray, B. C., Hayden, M., Jaschek, G.. and Bruns, G. A. P. (1989). Ain. J . Hum. Genet. 44,486-495. Gessler, M., Poustka, A., Cavenee, W., Neve, R. L., Orkin, S. H., and Bruns, G. A. P. (1990). Nature (London) 343, 774-778. Glaser, T., Lewis, W. H., Bruns, G. A. P., Waltkins, P. C., Rogler, C . E., Shows, T. B., Powers, V.’E., Willard, H. F., Goguen, J. M., Siniola, K. O., and Housman, D. E. (1986). .Va/ure (London) 321, 882-887. Glaser, T.,Jones, C., Call, K. M., Lewis, W. H., Bruns, G. A. P., Junien, G , b’aziri, M., and Housman, D. E. (1987). Cy/ogt.net. Cell Gene/. 46, 620. Glaser, T., Jones, C., Douglass, E. C., and Housman, D. E. (1989). In “Molecular Diagnostics of Human Cancer,” pp. 253-277. Cold Spring Harbor Lab., Cold Spring Harbor, New York. Glaser, T., Rose, E., Morse, H., Housman, D. E., and Jones, C. (1990a). Genunizcs 6,48-64. Glaser, T., Lane, J., and Housman, D. E. (1990b). Science 250, 823-827. Goss, S. J., and Harris, H . (1975). Nature (London) 255, 680-684. Grundy, P., Kougos, A., Morgan, K., Li, F. P., Meadows, A , , and Cavenee, W. K. (1988). Nature (London) 336, 374-376. Grundy, P., Breslow, N., Green, D. M., Sharples, K., Evans, A., and D’Angio, G.J. (1989).J. C h i . Oncol. 7, 638-647. Grundy, P. et al. (1992). Submitted for publication. Gusella, J. F., Jones, C., Kao, F. T., Housman, D., and Puck, T. T. (1982). Pmr. Natl. A m d . Sci. U.S.A. 79, 7804-7808. Haber, D. A , , and Housman, D. E. (1991). Cell (Cambridge, Mus.~.)64, 5-8. Haber, D. A., Buckler, A. J., Glaser, T., Call, K. M., Pelletier,J., Sohn, R. L., Douglass, E. C., and Housman, D. E. (1990). Cell (Cambridge, M m . ) 61, 1257-1269. Haber, D. A., Sohn, R. L., Buckler, A. J., Pelletier, J., Call, K. M., and Housman, D. E. (1991). Proc. Natl. Acad. Sci. U.S.A. (Cambridge, MaJs.) 88, 9618-9622.

66

DANIEL A. HABER AND DAVID E. HOUSMAN

Haig, D., and Graham, C. (1991). Cell 64, 1045-1046. Harris, H., Miller, 0.J., Klein, G., Worst, P., and Tachibana, T. (1969). Nature (London) 223, 363-366.

Hasgekar, N. N., Pendse, A. M., and Lalitha, V. S. (1989). Pediutr. Pathol. 9, 131-139. Henry, I., Grandjouan, S., Couillin, P., Barichard, F., Huerre-Jeanpierre, C., Glaser, T., Philip, T., Lenoir, G., Chaussain, J., and Junien, C. (1989). Proc. Natl. Acad. Sci. U.S.A. 86, 3247-325 1.

Henry, I., Bonaiti-Pellie, C., Chehensse, V., Beldjord, C., Schwartz, C., Utermann, G., and Junien, C. (1991). Nature (London) 351,665-667. Herskowitz, I. (1987). Nature (London) 329, 219-222. Hogan, B. L. M., Horsburgh, G., Cohen, J., Hetherington, C. M., Fisher, G., and Lyon, M. F. (1986).J. Embryol. Exp. Morphol. 97, 95-110. Huang, A., Campbell, C. E., Bonetta, L., McAndrews-Hill, M. S., Chilton-MacNeil, S., Coppes, M. J., Law, D. J., Feinberg, A. P., Yeger, H., and Williams, B. R. B. (1990). Science 250, 99 1-994. Huff, V., Comptom, D., Chao, L., Strong, L., Geiser, C., and Saunders, G. (1988). Nuture (London) 336, 377-378. Huff, V., Miwa, H., Haber, D. A., Call, K. M., Housman, D. E., Strong, L. C., and Saunders, G. F. (1991). Am. J. Hum. Genet. 48,997-1003. Irminger, J. C., Schoenle, E. J., Briner, J., and Humbel, R. E. (1989). Eur. J . Ped. 148,620623.

Joseph, L. J., LeBeau, M. M., Jamieson, G. A,, Acharya, S., Shows, T., Rowley, J. D., and Sukhatme, V. P. (1988). Proc. Natl. Acad. Sci. U.S.A. 85, 7164-7168. Junien, C., Turleau, C., de Grouchy, J., Said, R., Rethore, M. O., Tenconi, R., and Dufier, J. L. (1980). Ann. Genet. 23, 165-168. Kao, F. T., Jones, C., and Puck, T. T. (1976). Proc. Natl. Acad. Scz. U.S.A. 73, 193-197. Knudson, A. G. (1971). Proc. Natl. Acad. Sci. U.S.A. 68, 820-823. Knudson, A. G. (1985). Cancer Res. 45, 1437-1443. Knudson, A. G., and Strong, L. C. (1972a). J. Natl. Cancer Inst. ( U S . )48, 313-324. Knudson, A. G., and Strong, L. C. (1972b). Am. J. Hum. Genet. 24, 514-532. Kondo, K., Chilcote, R. R., Maurer, H. S., and Rowley, J. D. (1984). Cancer Res. 44, 53765681.

Koufos, A., Hansen, M. F., Lampkin, B. C., Workman, M. L., Copeland, N. G., Jenkins, N. A., and Cavenee, W. K. (1984). Nuture (London) 309, 170-172. Koufos, A., Grundy, P., Morgan, K., Aleck, K. A,, Hadro, T., Lampkin, B. C., Kalbakji, A,, and Cavenee, W. K. (1989). Am. J . Hum. Genet. 44,711-719. Lavedan, C., Barichard, G., Azoulay, M., Couillin, P., Molina, Gomex, D., Nicolas, H., Quack, B., Rethore, M. O., Noel, B., and Junien, C. (1989). Cytogenet. Cell Genet. 50, 70-74. Lavigueur, A., Maltby, V., Mock, D., Rossant, J., Pawson, T., and Bernstein, A. (1989). Mol. Cell. Biol. 9, 3982-3991, Lee, W.-H., Bookstein, R., Hong, F., Young, L.-J., Shew, J.-Y., and Lee, E. Y.-H. (1987). Science 235, 1394-1 399. Lewis, W. H., Yeger, H., Bonetta, L., Chan, H. S. L., Kang, J., Junien, C., Cowell,J., Jones, C., and Dafoe, L. A. (1988). Genomics 3, 25-31. Li, F. P., Gimbrere, K., Gelber, R. D., Sallan, S. E., Flamant, F., Green, D. M., Heyn, R. M., and Meadows, A. T. (1987).JAMA, J. Am. Med. Assoc. 257, 216-219. Lubinsky, M., Hermann, J., Kosseff, A. L., and Opitz, J. M. (1974). Lancet 1, 932. Ludlow, J. W., DeCaprio, J. A., Huang, C.-M., Lee, W.-H., Pauch, E.,and Livingston, D. M. (1989). Cell (Cambridge, Mass.) 56, 57-65.

THE GENETICS OF WILMS’ TUMOR

67

Madden, S. L., Cook, D. M., Morris, J. F., Gashler, A,, Sukhatme, V. P., and Rauscher 111, F. .J. (1991). Science 253, 1550-1553. Mannens, M., Slater, R. M., Heyting, C., Bliek, J., d e Kraker, J., Coad, N., de PagterHolthuizen, P., and Pearson, P. L. (1988). Hum. Genet. 81, 41-48. Mannens, M., Hoovers, J., Bleeker-Wagemakers, E. M., Redeker, E., Blick, J., OverbeekeMelkert, M., Huff, V., Williams, B., van Heyningen, V., Junine, C., Haber, D., Speleman, F., Heyting, C., Slater, R. M., and Westerveld, A. (1992). Cytogenet. CellGenet. (in press). Matsunaga, E. (1981). Hum. Genet. 57, 231-246. Miller, R. W., Fraumeni, J. F., and Manning, M. D. (1964). N . Engl. J. Med. 270, 922. Mitchell, P. J., and Tijan, T. (1989). Science 245, 371-378. Moore, J. W., Hyman, S., Antonarakis, S. E., Mules, E. H., and Thomas, G. H. (1986).Hum. Genet. 72, 297-302. Nakagome, Y., Ise, T., Sakurai, M., Nakajo, T., Okamoto, E., Takano, T., Nakahori, Y., Tsuchida, Y., Nagahara, N., Takada, Y., Ohsawa, Y., Sawaguchi, S., Toyosaka, A,, Kobayashi, N., Matsunaga, E., and Saito, S. (1984). Hum. Genet. 67, 245-248. National Wilms’ Tumor Study Committee. (1991).J. Clin. Oncol. 5 , 877-887. Nehlin, J. O., and Ronne, H. (1990). EMBO J. 9, 2891-2898. Niikawa, N., Ishikiriyama, S., Takahashi, S., Inagawa, A,, Tonoki, H., Ohta, Y., Hase, N., Karnei, T., and Kajii, T. (1986). Am. J. Med. Genet. 24, 41-55. Ohaki, Y. (1989). Pediatr. Pathol. 9, 19-33. Orkin, S. H., Goldman, D. S., and Sallan, S. E. (1984). Nature (London) 309, 172-174. Oshirnura, M., Kugoh, H., Koi, M., Shimizu, M., Yamada, H., Satoh, H., and Barrett, J. C. (1990). J. Cell. Biochem. 42, 135-142. Pelletier, J., Schalling, M., Buckler, A. J., Rogers, A,, Haber, D. A., and Housman, D. E. (1991a). Genes Deu. (Philadelphia) 5, 1345-1356. Pelletier, J.. Bruening, W., Li, F. P., Haber, D. A., Glaser, T., and Houseman, D. E. (1991b). Nature (London) 353, 431-434. Pelletier, J., Bruening, W., Kashton, C. E., Maver, S. M., Manivel, J. C., Striegel, J. E., Houghton, D. C., Junien, C., Habib, R., Fouser, L., Fine, R. N., Silverman, B. L., Haber, D. A,, and Houseman, D. E. ( 1 9 9 1 ~ )Cell . 67, 437-447. Pendergrass, T. W. (1976). Cancer (Philadelphia) 37, 403-409. Ping, A. J., Reeve, A. E., Law, D. J., Young, M. R., Boehnke, M., and Feinberg, A, P. (1989). Am. J. Hum. Genet. 44,720-723. Porteous, D. J., Bickmore, W., Christie, S., Boyd, P. A., Cranston, G., Fletcher, J. M., Gosden, J. R., Rout, D., Sewwright, A., Simola, K. 0.J., van Heyningen, V., and Hastie, N. D. (1987). Proc. Natl. Acad. Sci. U.S.A. 84, 5355-5359. Pritchard-Jones, K., Fleming, S., Davidson, D., Bickmore, W., Porteous, D., Gosden, C., Bard, J., Buckler, A., Pelletier, J., Housman, D., van Heyningen, V., and Hastie, N. ( 1990). Nature (London) 346, 194- 197. Rausher, F. J., 111, Morris, J. F., Tournay, 0.E., Cook, D. M., and Curran, T. (1990).Science 250, 1259-1262. Reeve, A. E., Housiaux, P. J., Gardner, R. J., Chewing, W. E., Grindley, R. M., and Millow, L. J. (1984). Nature (London) 309, 174-176. Reeve, A. E., Eccles, M. R., Wilkins, R. J., Bell, G. I., and Millow, L. J. (1985). Nature (London) 317, 258-260. Reeve, A. E., Sih, S. A., Raizis, A. M., and Feinberg, A. P. (1989). Mol. Cell. B i d . 9, 17991803. Riccardi, V. M., Sujansky, E., Smith, A. C., and Francke, U. (1978). Pediatrics 61,604-610.

68

DANIEL A. HABER AND DAVID E. HOUSMAN

Rornain, D. R., Gebbie, 0. B., Parfitt, R. G., Columbano-Green, L. M., Smythe, R. H., Chapman, C. J., and Kerr, A. (1983).J. Med. Genet. 20, 380-382. Rose, E. A., Glaser, T., Jones, C., Smith, C. L., Lewis, W. H., Call, K. M., Minden, M., Champagne, E., Lewis, W., Bonetta, L., Yager, H., and Housman, D. E. (1990). Cell (Cambridge, Mass.) 60, 495-508. Schroeder, W. T., Chao, L. Y., Dao, D. D., Strong, L. C., Pothak, V., Riccardi, V., Lewis, W., and Saunders, G. F. (1987). Am. J. Hum. Genet. 41, 413-420. Schwartz, C. E., Haber, D. A., Stanton, V. P., Strong, L. C., Skolnick, M. H., and Housman, D. E. (1991). Genomics (in press). Scott, J., Cowell, J., Robertson, M. E., Priestley, L. M., Wadey, R., Hopkins, B., Pritchard, J., Bell, G. I., Rall, L. B., Graham, C. F., and Knott, T. J. (1985). Nature (London) 317,260262. Simola, K. 0.J., Knuutila, S., Kaitila, I., Pirkola, A,, and Pohja, P. (1983). Hum. Genet. 63, 158-161. Solis, V., Pritchard, J.. and Cowell, J. K. (1988). Cancer Genet. Cytogenet. 34, 223-234. Sotelo-Avila, C., Gonzalez-Crussi, F., and Fowler, J. W. (198O).J. Pudiatr. 96, 47-50. Sparkes, R. S., Murphree, A. L., Lingua, R. W., Sparkes, M. C., Fields, L. L., Funderburk, S. J., and Benedict, W.F. (1983). Science 219, 971-973. Stanbridge, E. J. (1976). Nature (London) 260, 17-20. Stanbridge, E. J., and Wilkinson, J. (1978). Proc. Natl. Acad. Sci. U.S.A. 75, 1466-1471. Sukhatme, V. P., Cao, X., Chang, L. C., Tsai-Morris, C. H., Stamenkovich, D., Ferreira, P. C. P., Cohen, D. R., Edwards, S. A., Shows, T. B., Curran, T., LeBeau, M. M., and Adamson, E. D. (1988). Cell (Cambndge, Mass.) 53, 37-43. Theiler, K., Varnum, D. S., and Stevens, L. C . (1978). Anat. Embryol. 155, 81-86. Ton, C. C., Huff, V., Call, K. M., Cohn, S., Stong, L. C., Housman, D. E., and Saunders, G. F. (1991). Genomics 10, 293-297. Turleau, C., de Grouchy, J., Nihoul-Fekete, C., Chavin-Colin, F., and Junien, C. (1984). Hum. Genet. 67,455-456. van Heyningen, V., Boyd, P. A,, Seawright, A., Fletcher, J. M., Fantes. J. A., Buckton, K. E.. Spowart, G., Porteous, F. J., Hill, R. E., Newton, M. S., and Hastie, N. D. (1985). Proc. Natl. Acad. Sci. U.S.A. 82, 8592-8596. van Heyningen, V., Bickmore, W. A., Seawright, A,, Fletcher, J. M., Maule, J . , Fekete, G., Gessler, M., Bruns, G. A. P., Huerre-Jeanpierre, C., Junien, C., Williams, B. R. G . , and Hastie, N. D. (1990). Proc. Natl. Acad. Scz. U.S.A. 87, 5383-5386. Wang-Wuu, S., Soukup, S., Bove, K., Gotwals, B., and Lampkin, B. (1990). Cancet-Rex. 50, 2786-2793. Waziri, M., Patil, S., Hanson, J., and Bartley, S. A. (1983).J. Pedzatr. 102, 873-876. Weissman, B. E., Saxon, P. J., Pasquale, S. R.,Jones, G. R. Geiser, A. G., and Stanbridge, E. J. (1987). Science 236, 175-180. Weston, A., Willey, J. C., Modali, R., Sugimura, H., McDowell, E. M., Resau, J., Light, B., Haugen, A., Mann, D. L., Trump, B. F., and Harris, C. C. (1989). Proc. Natl. Acad. Scz. U.S.A. 86, 5099-5103. Wiedemann, H. R. (1964).J. Genet. Hum. 13, 223-232. Wiedemann, H. R. (1983). Eur. J . Pedzatr. 141, 129. Yandell, D. W., Campbell, T. A., Dayton, S. H., Petersen, R., Walton, D., Little, J. B., McConkie-Rosell, A., Buckley, E. G., and Dryja, T. P. (1989). N . Engl. J . Med. 321, 16891695. Zabel et al. (1992). Submitted for publication.

p53 EXPRESSION IN HUMAN BREAST CANCER Adrian

L. Harris

Institute of Molecular Medicine, John Radcliffe Hospital, University of Oxford, Oxford, OX3 9DU England

I. Introduction 11. Discovery of p53

111. IV. V. VI. VII. VIII. IX. X. XI. XII. XIII. XIV. XV. XVI. XVII. XVIII. XIX.

Dominant Transforming Oncogene Recessive Oncogene Dominant Negative Function and Gain in Function Mutants Normal Function and Regulation of p53 Mutations in p53 and Interactions with Viral Proteins Loss of Heterozygosity and p53 Mutations Different Functional Mutations and Mutational Profiles Methods to Assay p53 in Human Cancer Studies in Breast Cancer Loss of Heterozygosity Immunochemistry Mutations Familial Breast Cancer and Li Fraumeni Syndrome Expression of p53 with Other Oncogenes and Receptors Prognosis In Situ Lesions Therapeutic Possibilities References

I. Introduction Rapid progress has been made in the last 2 years in understanding the function of p53 in human cancer (Levine et al., 1991). From the finding that it was a cellular protein binding to the viral oncogene large T antigen, to the observation that mutations in p53 are the commonest genetic change in human cancer, has been a major step in understanding tumor biology. This article covers normal and abnormal functions of p53 and considers its role in primary breast cancer. Potential therapeutic implications are also discussed. II. Discovery of p53

p53 was initially discovered (Lane and Crawford, 1979) as a normal cellular protein bound to the viral transforming oncogene large T anti69 ADVANCES IN CANCER RESEARCH. VOL. 59

Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.

70

ADRIAN L. HARRIS

gen (Linzer and Levine, 1979). Immunoprecipitations of large T antigen brought down a nuclear phosphoprotein of M , 53,000. Initial characterization and sequencing of the gene was reported by Lamb and Crawford (1986). The gene is 20 kilobases (kb) long, and has 11 exons. The chromosome location is 17~13.1(van Tuinen et al., 1988). Ill. Dominant Transforming Oncogene

Experiments with the initial cloned p53 genes demonstrated that p53 could immortalize or transform primary cells in culture and also demonstrated a cooperation with other oncogenes such as ras (Eliyahu et al., 1984; Jenkins et al., 1984). These experiments suggested that p53 was a dominant transforming oncogene.

IV. Recessive Oncogene However, there were discrepancies in that loss of p53 also seemed to be associated with transformation. Work with transformed cells produced by Friend leukemia virus insertions showed loss of both copies of p53 in erythroleukemia cell lines (Mowat et al., 1985).Also, some isolated genes had sequence differences from originally isolated p53 and these were shown to compete with the original p53 genes, which had dominant phenotypes. The original p53 clones are now known to have mutations (Hinds et al., 1989; Eliyahu et al., 1989). Thus, the wild-type gene was able to compete with the mutant and prevent transformation. The wild type can also block transformation by the cooperating oncogenes adenovirus EIA or c-my and p21 rus (Finlay et al., 1989; Eliyahu et al., 1989).

V. Dominant Negative Function and Gain in Function .Mutants More complex is the potential interaction between different degrees of expression of wild-type and mutant proteins in the same cell (Levine et al., 1991). If the wild type is able to compete with the mutant and arrest cell growth, then there will be strong selection pressure to lose the wild-type gene. However, the mutant p53 is able to complex itself and the wild-type p53 and thus exert a dominant negative role. Recently, cotranslation of activated mutant p53 with the wild type has been shown to drive the wild-type p53 into the mutant conformation (Milner and Medcalf, 1991). Thus outcome will depend on the relative level of ex-

p53

EXPRESSION I N HUMAN BREAST CANCER

71

pression of mutant and wild-type p53. Loss of wild-type p53 may occur through allelic deletion. If loss of heterozygosity then results in loss of wild-type p53, what is the selective pressure to retain mutant p53? Various investigators have studied a possible gain in function of mutant p53, as well as its negative role in interfering with normal p53 function. Some cell lines have neither p53 gene and transfections into these with wild-type p53 blocks growth (Chen et al., 1990; Diller et al., 1990), while transfection of mutant does have some further effect in changing morphology and growth rate (Chen et al., 1990). It would appear that mutations in p53 result in a loss of the normal function but there can be some gain in function, which may be an aberrantly regulated function of the wild-type p53.

VI. Normal Function and Regulation of p53 p53 is a nuclear protein with putative roles in DNA replication, transcription, and cell cycle control. The wild-type p53 can interfere with simian virus 40 (SV40) DNA replication in uitro and competes with DNA polymerase ci binding to large T antigen. Mutant p53 does not have this effect (Gannon and Lane, 1987; Friedman et al., 1990). Recently the colocation of p53 in intact cells with replicating SV40 virus has been studied; it was found that p53 colocalizes with replication complexes (Wilcock and Lane, 1991). When expressed as a fusion protein with Gal-4, activation of a test gene can be demonstrated (Fields and Jang, 1990; Raycroft et al., 1990; O’Rourke et al., 1990). However, this may demonstrate the ability to interact with other proteins rather than the role as an independent transcription factor. The first gene sequence reported to be regulated by p53 is the MCK enhancer (Weintraub et al., 1991). It has also been demonstrated that p53 may be on the pathway of transforming growth factor P (TGFP) signal transduction. TGFP inhibits cells at the G, phase of the cell cycle and cells transfected with viral oncogenes that complex p53 are less responsive to this inhibitory growth factor (Wyllie et al., 1991).Thus, a potential mechanism of action of p53 is in coupling inhibitory signal transduction to the regulation of DNA replication. The normal p53 protein has a rapid half-life of a few minutes and is located in the nucleus. Usually it is at a too-low level to be detected in normal tissues and cells by immunochemical methods. It is low after mitosis, but increases in G,. During S phase p53 becomes phosphory-

72

ADRIAN L. HARRIS

lated (Bischoff et al., 1990).At least one enzyme involved in casein kinase I1 (Meek et al., 1990). RNA is covalently linked to the same residue (ser-389) that is phosphorylated by casein kinase 11. The role of this RNA is unknown by p53 may also be involved in RNA metabolism, or bind to heterogeneous nuclear RNA fibrils (Samad and Carroll, 1991). The cell cycle-dependent kinase cdc-2 kinase, which has an important role in regulating the entry of cells into mitosis, has a consensus phosphorylation sequence in its target genes. This consensus has been demonstrated in p53 and indeed p53 has now been shown to be a substrate for cdc-2 kinase (Milner et al., 1990; Sturzbecher et al., 1990). This implies another important link between cell cycle regulation and the function of p53, although an effect of changing phosphorylation of p53 has not been demonstrated. However, the nuclear location signal at the C terminus is contiguous with the cdc-2 kinase site (Dang and Lee, 1989; Addison et al., 1990). Nuclear location of mutant p53 may be important for its transforming function (Shaulsky et al., 1990).This is a controversial area, as others have shown cytoplasmic location of mutant p53 (see Sections VII and VIII below).

VII. Mutations in p53 and Interactions with Viral Proteins Many mutant p53 genes have now been sequenced and the mutations cluster in four conserved domains (Soussi et al., 1990). These domains were analyzed by sequence homology between human, mouse, rat, chicken, and Xenopw p53 genes. There are 11 exons, with 5 highly conserved, homologous domains on exons 2, 5, 7, and 8, which specify domains I, 111, IV, and V, respectively. Domain I1 is specified by both exons 4 and 5 (see Fig. 1). These mutations result in a change of conformation of p53 and many of the mutant p53 proteins can interact with heat shock protein 70 (hsp70) (Finlay et al., 1988; Sturtzbecher et al., 1987). Since there is a major cytoplasmic localization in hsp70 there may be a shift of the p53 protein from nucleus to cytoplasm. This has indeed been shown with a temperature-sensitive p53 mutation; at the permissive temperature the protein is in the nucleus and, as the temperature is changed and the mutant form predominates, there is a shift to a cytoplasmic location (Martinez et al., 1991; Gannon and Lane, 1991). The mutations also result in a much longer half-life of p53 of several hours (Matlashewski et al., 1986; Benchimol et al., 1982). This results, therefore, in an elevation of intracellular level of p53, which can now be immunologically detected. The above work relates to mutational forms of p53, but p53 was

p53

73

EXPRESSION IN H U M A N BREAST CANCER

1

200

100

I

II

Ill

IU

u

300

393

11111

I1

-

32

79

PablSOl

1

t Li Fraumeni 245-258 I

212

217

Pab 240

-

370 378

Pab 421

FIG. 1. Diagram of p53 conserved regions, mutation sites, and antibody epitopes. T h e hatched bars represent the five conserved domains in p53. Mutations occur in four of them (11, 111, IV, and V), as shown by the vertical bars. Numbers 1-393 represent the amino acids in p53. T h e three arrows below the line represent hot spots that coniprise onethird of all reported mutations. T h e solid bars represent the epitopes that react with the antibodies designated PAb 1801, 240, and 42 1.

discovered as a protein interacting with viral transforming oncogenes. Large T, papilloma virus E6 protein, and adenovirus E I B proteins can all interact with p53. In the case of the large T antigen there is prolongation of the p53 half-life and high expression of p53 as a result of this. However, interestingly, the papilloma virus protein E6 has been shown to increase the rate of degradation of p53 (Werness et al., 1990). This could also be related to transformation if it resulted in loss of sufficient p53 to inhibit its normal function. It is also noteworthy that the papilloma virus E7 protein, large T antigen, and adenovirus EIA transforming proteins interact also with the retinoblastoma (RB) gene product. The latter has a similar putative role in the regulation of the G , / S transition via TGFP and a normal role in inhibiting the entry of the cell into the DNA replication phase of the cell cycle (Laiho et al., 1990; Pietenpol el al., 1990). It may be that coordinate loss of function of-both of these genes is very important in certain tumor types. However, it does seem to vary between different tumor types whether both genes are affected o r not. Thus, in lung cancer, there is clear evidence that coordinate p53 and RB mutations or loss of genes is very common, although this is much less so in the case of breast cancer. T h e regions in normal p53 that interact with the viral proteins are contained within the conserved boxes where mutations occur. Thus, a strong possibility exists that there are endogenous cellular proteins similar to the viral proteins that interact normally with p53. A variety of mechanisms therefore can interact with p53 to change its half-life, its function, and its cellular

74

ADRIAN L. HARRIS

location. In the case of the viral proteins the p53 is either sequestered from its normal role or degraded more rapidly. Yet another way in which elevated p53 expression could occur is via failure of degradation as a result of change of the proteolytic pathways involved in its normal destruction. This has not been demonstrated directly but ubiquitination is an important mechanism in the destruction of p53 (Ciechanover et al., 1991) and it can be postulated that changes in this pathway may lead to high p53 accumulation without mutation. The effect of this would depend on understanding more about the biology of the rapid turnover of wild-type p53. So far, all mechanisms that lead to elevation of p53 either involve sequestration of the protein or mutation of the protein. However, high expression of wild-type p53 alone inhibits growth of cell lines with mutant p53 or with complete loss of p53.

VIII. Loss of Heterozygosity and p53 Mutations Loss of heterozygosity is a common observation in human cancer and such deletions may indicate sites of putative recessive oncogenes. This strategy was successfully followed by Vogelstein et al., who demonstrated that loss of heterozygosity at 17p13 was very common in colon cancers (Vogelstein et al., 1989). Since p53 was at this locus and a candidate oncogene, they sequenced the remaining gene and found point mutations. This was the major stimulus to the further work in human p53 mutations in cancer (Baker et al., 1989). They then went on to investigate other primary tumors and cell lines and demonstrated that mutations occurred at many sites in the p53 gene in the conserved domains in several types of cancer, including breast cancer (Nigro et al., 1989).Thus, loss of heterozygosity in colon cancer was associated with p53 mutation in the remaining gene. This has been a common observation in several tumor types but is not always the case. Loss of heterozygosity may occur after p53 mutation since there are so many cases reported of tumors in which there is not a loss of heterozygosity but there is mutation in p53. It may be that there are different effects in different genetic backgrounds with mutant and wild-type p53 and that for mutant p53 to exert its full effect, loss of the wild type is necessary in some circumstances but in others expression of mutant p53 with wild type still allows for a transforming function. In some types of tumors, in particular cell lines from leukemias, there may be loss of both p53 genes. Explanation for these different genotypes is not clear but it may be that p53 has different roles, particularly in cell cycle regulation, differentiation, and mitosis, in different types of tissue.

p53

EXPRESSION I N HUMAN BREAST CANCER

75

IX. Different Functional Mutations and Mutational Profiles Different carcinogens produce different patterns of mutation. Different mutations in p53 may be found in different types of cancer, depending on the predisposing factors. Mutational profiles may be useful evidence for the effects of putative carcinogens. Also, specific functions may be lost or gained by particular mutations in p53, so there may be selection of a particular mutation in a particular type of cancer. There is evidence of both mechanisms. In lung cancers G to T transversions are the common change but they involve at least 16 different codons. T h e commonest codon affected occurs only 4 out of 25 reported mutations. G to T transversions are the changes expected for bulky adducts typically produced by carcinogens in cigarette smoke, e.g., benz [alanthracene. More recently, in hepatomas occurring in South Africa and China, a hot spot in codon 249 has been reported (Hsu et al., 1991; Bressac et al., 1991). In all cases with p53 mutations this hot spot was found. This is the only tumor type so far described with this pattern. It is noteworthy that the pattern can be ascribed to aflatoxin, which damages selectively certain bases, including those on codon 249. However, the hot spot is difficult to explain by this mechanism alone. It is possible that the mutant protein interacts with viral proteins from hepatitis B virus and is part of the selection (Harris, 1991). This is an important interface between viral and chemical carcinogenesis. In a survey of 94 primary tumors with mutant p53 genes, the majority clustered between codons 130 and 290 (Levine et al., 1991). There were three mutational hot spots, codons 175, 248, and 273, accounting for about one-third of the cases. There are biological differences between these mutants. Thus, mutants for codon 273 are more like the wild type in that they do not bind hsp70, and are not so well recognized by antibodies to mutant p53 proteins, and can trans-activate a test gene. In contrast, mutants at codon 175 are much more active in transformation assays, do bind hsp70, and are recognized by antibodies to mutant p53 (Hinds et al., 1990). Similar results have been found with mutant murine p53 genes (Halevy et al., 1990). A comprehensive review of mutations in human cancers analyzed the mutational profile in 280 cases with base substitutions (Hollstein et al., 1991). In breast cancers (n = 31) G:C to A:T transitions at CpG sites are rare compared to colon cancers (13 vs 67 percent). G to T transversions are common in breast cancer compared with colon cancer (23 vs 0 percent).

76

ADRIAN L. HARRIS

X. Methods to Assay p53 in Human Cancer Several methods have been used to assess the role of p53 in human cancers and essentially they are complementary techniques. Loss of heterozygosity provides putative evidence for loss of wild-type p53 but there are other genes that may be involved on chromosome 17. Sequencing from RNA should show most mutations but not if the mutations lead to loss of expression. Genomic sequencing by polymerase chain reaction (PCR) from DNA may give similar information, with less degradation than mRNA in tumors. However, since normal tissues make up at least 50% of the tumor mass, representative samples may not be taken. Microdissection from slides can overcome this problem if necessary. SSCP (single strand conformational polymorphism of DNA) seems to be the most sensitive method and has picked up mutations that were not detected by sequencing, probably because the latter method was contaminated with normal tissue and SSCP may be sensitive enough to detect 10% mutant sequences among wild-type cDNA. Touch preparations are useful to obtain relatively pure sources of cancer cells for PCR (Kovach et al., 1991). Immunochemistry has become widely used with antibodies that react to both mutant and wild-type p53, although there are antibodies that preferentially react with mutant protein (PAb240) (Gannon et al., 1990). Some antibodies cross-react with cytokeratin (Ab42 l), which makes cytoplasmic location difficult to assess. The epitope map for the most commonly selected antibodies is shown in Fig. 1. Normal tissues are reported not to stain with the antibodies and high levels of stable, mutated proteins produce differential staining of p53 in tumors, with antibodies that react with either mutant or wild-type forms. It is important t o point out that the antibodies that have been assessed against the mutant p53 have been shown to differentiate in immunoprecipitation assays, which depend on the mutant protein being in its native conformation (Gannon el al., 1990). However, after tissue fixation or even frozen sectioning and light fixation, there may be conformational changes that mean that the antibodies are no longer specific for mutant forms but will pick up any denatured p53. It may be the high level of expression of the mutant p53 that allows it to be detected with antibodies reacting with native wild-type and mutant protein. The advantages of immunochemistry are, of course, demonstration of high expression of presumptively mutated protein within the tumor, heterogeneity of expression, coexpression or other tumor markers, and ready availability. If only some epitopes are detected with each antibody, a panel may be necessary. However, different epitopes may be associated with different

p53

EXPRESSION I N HUMAN BREAST CANCER

77

biological behavior and prognosis and this remains to be assessed. There is evidence that different epitopes may be detectable during the cell cycle, possibly relating to different functions of p53 (Milner, 1991; Arai et al., 1986). XI. Studies in Breast Cancer With the preceding background it is clear that investigations in breast cancer need to look at loss of heterozygosity of the p53 locus, mutations in the p53 gene, and imrnunochemistry. Furthermore, studies on familial breast cancer and early in sztu lesions versus invasive lesions will be of major interest. T h e prognostic importance and coexpression with other oncogenes could allow for selection of high-risk subgroups.

XII. Loss of Heterozygosity Several groups have analyzed loss of heterozygosity on chromosome 17p in breast cancer (Table I) (Mackay et al., 1988; Prosser et al., 1990; Varley et al., 1991; Sato et al., 1990; Cropp et al., 1990; Devilee et al., 1991; Davidoff et al., 1991b; Chen et al., 1991). This is a frequent occurrence but there are at least two loci on chromosome l7p-one at p53 and one at locus YNZ-22-which are commonly deleted (Coles et al., 1990; Sato et al., 1990). There is a possibility that YNZ-22 regulates the p53 locus (Thompson et al., 1990a). In addition, there is an association of

Loss

OF

Number of cases 50 98 81

76 40 25 74 52 28 49

TABLE I HETEROZYGOSITY ON CHROMOSOME 1 7 I~N BREAST CANCER

LOH' probes UNZ-22 p144 D6 YNZ-22 pBHP 53 YNZ-22 YNZ-22 BHP 53 YNZ-22 YNZ-22 p53/MC735.1 HF 12-2 YNZ-22

= LOH, Loss of heterozygosity.

Loss (%)

Ref.

48

Cropp et al. (1990)

58 27 58 48 32 69 52 61

Coles et al. (1990)

61

Thompson et al. (1990b) Sato et al. (1990) Varley et al. (1991) Chen et al. (1991) Davidoff et al. ( 199 la) Devilee et al. (1991)

78

ADRIAN L. HARRIS

loss of heterozygosity at 17p with erbB-2 amplification (Sato et al., 1990). Loss of heterozygosity at this second locus has been related to high proliferative capacity and aneuploidy in breast cancer (Chen et al., 1991).

XIII. lmmunochemistry Cattoretti et al. (1988a,b) initially demonstrated the frequent expression immunochemically of p53 in tumors and absence from normal tissue (Table 11). This was also associated with epidermal growth factor (EGF) receptor expression. Although there were potential problems with the antibodies used, in that one of them also reacts with cytokeratins to some extent, nevertheless nuclear staining with antibodies to p53 was demonstrated in a high proportion of cases. The prognostic importance of this has not yet been followed up. Cattoretti et al. (1988a,b) also noted nuclear staining of varying intensity and also a few cases where single scattered positive cells were stained. With PAb1801 a higher proportion of cases was stained than with PAb42 1. Varley et al. ( 1991) used PAb240 and PAb 1801 and found nuclear staining in most cases. Some showed only cytoplasmic staining. In a few cases there were scattered positive cells only. In general, there was a good concordance of PAb240 and 1801, but in some cases staining was clearer with the former. Thus, three patterns of staining appear to be detectable. Bartek et al. (1990a) also described three patterns-nuclear staining in the majority, a few cases with scattered positive cells, and some with mainly cytoplasmic staining. We studied 111 primary tumors. Fifty-nine tumors stained positively with the monoclonal antibody PAb240 (Horak et al., 1991). In 58 cases staining was concordant with a polyclonal antibody JG8. Three of the 52 monoclonal antibody-negative cases stained with the polyclonal antibody. The staining was heterogeneous, ranging from 10 to 70% of cells. The majority of tumors were in the range of 30-70%. Tumors that were negative clearly had no cell staining with the antibodies. In the majority of cases, staining was cytoplasmic and nuclear. In 5 of 59 cases the staining was predominantly nuclear. In several cases, only cytoplasmic staining was seen. In addition, in some tumors, there was focal staining of only some cells of the population. To assess whether the cytoplasmic location of proteins that are normally nuclear was a general phenomenon, we studied the localization of the ER. This is normally nuclear and we found in all cases that this was so. Thus, the cytoplasmic localization of p53 was not due to nuclear disruption. The differences in relative frequency of the different patterns may be related to fixation meth-

TABLE I1 p53 ASSOCIATION WITH ER, EPIDERMAL GROWTHFACTORRECEPTOR,NODESTATUS, AND TUMOR GRADEQ

n

Antibody

304 90 111 73 200 76 122

1801 1801 240 240/1801 42 1/ 1801 mRNA 421/804, immunoassay 1801 240/ 1801/42 I 1801 CM- 1 1801

72 62 108 102 73

Percentage positive

Inverse association with ER

22 36 53 58 15/45 57 24

Yes, p < 0.0001 Yes, p < 0.1 Yes, p < 0.2 Yes, p < 0.01 Yes, p < 0.001 Yes, p < 0.05 ND

26 55 88 37 23

ND ND ND

nND, Not done; NS, not significant.

Yes, p < 0.002 Yes, p < 0.01

Association with EGFr

Association with high-grade tumor (Ki67)

ND ND

p < 0.0001 p < 0.01

Yes, p < 0.02 ND Yes, p < 0.001 ND ND ND ND ND ND ND

(p

ND Yes < 0.001) NS Yes ND ND ND

p < 0.01

Nodes

Ref.

NS NS NS NS NS NS NS

Thor et al. (1991) Ostrowski et al. (1991) Horak et al. (1991) Varley et al. (1991) Cattoretti et al. (1988a) Thompson et al. (1990a) Crawford et al. (1984)

ND ND ND NS NS

Davidoff et al. (1991a) Bartek et al. (1990a) Martinazzi et al. (1990) Barnes and Fisher Iwaya et al. (1991)

80

ADRIAN L. HARRIS

odology, since p53 is able to move from a nuclear to a cytoplasmic location as described above. However, in primary tumors much more extensive correlations are necessary since there may be other mechanisms regulating p53 expression. Thus, in normal human peripheral lymphocytes, stimulation with phytohemagglutinin (PHA) increases p53 expression severalfold and it is readily detectable immunochemically. Also, immunochemical detection in dysplastic areas near p53-positive tumors has been noted (Bartek et al., 1990a). XIV. Mutations Several breast tumor mutations have now been sequenced and so far there is a good correlation of immunochemistry with mutated p53. All cases that show positive immunochemistry and have been sequenced show mutations (Table 111) (Varley et al., 1991; Davidoff et al., 1991a). However, more extensive studies are necessary and there may be mechanisms other than mutations leading to high expression assessed immunochemically. Mutations in breast cancer differ markedly in the frequency of G to T transversion compared to lung cancer. G to A transitions are the commonest change in breast cancer and may be related to deamination of 5-methyl cytosine at CpG sites (Rideout et al., 1990). The hot spots 175,248, and 273 occur in breast cancer but do not appear particularly common compared to their overall frequency. In addition, several breast cancer cell lines have been sequenced and have a range of mutations (Bartek et al., 1990b). There is no correlation of loss of heterozygosity (LOH) for chromosome 17p and mutations in p53 (Chen et al., 1991; Davidoff et al., 1991b). This contrasts with colon cancer, where they usually occur together (Baker et al., 1990). XV. Familial Breast Cancer and Li Fraumeni Syndrome Li Fraumeni syndrome is associated with familial breast cancer, sarcomas, adrenal carcinomas, and leukemia. Recently mutations in p53 in the germ line have been demonstrated in codons 245,248,252, and 258 (Malkin et al., 1990; Srivastava et al., 1990). It appears that this mutant protein behaves differently from several other mutant proteins. Most of the mutations previously described result in proteins that bind with hsp70 but this is not the case for Li Fraumeni syndrome. It appears to be more like the codon 273 than the codon 175-mutated proteins. The importance of the Li Fraumeni syndrome is the demonstration that only

p53

81

EXPRESSION IN HUMAN BREAST CANCER

TABLE 111 p53 MUTATIONS I N BREAST CANCER PRIMARY TUMORS AND CELLLINES Codon Cell lines BT20 BT549 BT474 MDA23 1 MDA468 T47D BT123 Tumors

A+C

132 249 285 280

Ref. Bartek et al. (1990b)

G+C G+A

G+A G+A C+T ND In-frame deletion A+G G-A G-A G-A G-tT T-A C-G G-T

273 194 NDa 175-180 163 237 245 248 266 254 278 282

Nigro et al. (1989)

Davidoff el al. (1991a,b)

175 167

G-A Deletion

Varley et al. (1991)

164 136 187 175 179 152 157

A-C C-G G-T G-tT C-+G C-T G-T

Singh et al. (199 1)

C+T

Chen et al. (1991)

14-base pair frameshift deletion G-C 1-base pair f-rarneshift deletion

Kovach et al. (1991)

149 175 235-239 280 329

a

Base change

3

ND, Not detected.

certain tumors are predisposed to it by the germ line mutation in p53. T h e implication is that p53 is more important in the biology of certain tumors than others. It is generally accepted that at least five changes are required to produce a cancer based on the exponential frequency rates in the incidence of adult cancers. The question therefore arises, why should mutations in p53 predispose to breast cancer at a far greater

82

ADRIAN L. HARRIS

frequency than to colon cancer, when in the primary sporadic tumors, mutations in p53 are even more common in colon than in breast tumors? One possibility is that p53 has different roles in different tissues-for example, apart from stopping cells proliferating, also having a role in differentiation or apoptosis. Thus, if p53 had a role early on in the differentiation of breast cancer, mutations in p53 could trap cells in a proliferating phase. In contrast, if p53 is important at a late stage of differentiation of the normal colon, mutations here would have very little effect until preceding events had accumulated. This suggests that more detailed investigation of the normal role of p53 in the tissues in which tumors occur very early on should be carried out. Familial breast cancer without the Li Fraumeni syndrome is also being investigated for mutation of p53, and although there are some negative studies (Prosser et al., 1991) there are preliminary reports that they occur at a low frequency. XVI. Expression of p53 with Other Oncogenes and Receptors Cattorstti et al. initially reported that EGF receptor (EGFr) was coexpressed with p53 in breast cancer (Cattoretti et al., 198813). We have similarly studied a series of breast tumors with quantitative estimation of EGF receptor and estrogen receptor versus p53 assessed by immunochemistry. Our results showed that EGFr-positive tumors more frequently expressed mutant p53 protein. Twenty-eight of 65 EGFr-negative tumors expressed p53, as did 31 of 46 EGFr-positive cases ( p < 0.02). A similar analysis was carried out for ErbB-2 but there was no significant association. ErbB-2 expression wa assessed by initially comparing gene amplification with ErbB-2 prot in expression by immunochemistry on frozen sections. This then allowed standardization of the immunochemistry for the whole series. All series so far show that p53 is inversely related to ER expression (Table 11). Although this is not always significant, the trend is there and the relative importance may be related to different ER assays and cut-off points. A possible interaction between high EGFr loss of ER and p53 may relate to synergistic regulation of growth-stimulatory pathways.

t

XVII. Prognosis Three reports have appeared, one in preliminary form, suggesting that tumors expressing high levels of p53 by immunochemistry show a higher rate of relapse (Thor et al., 1991; Iwaya et al., 1991; Ostrowski et

p53

EXPRESSION I N HUMAN BREAST CANCER

83

al., 1991). This has been carried out on paraffin sections in the two former studies and frozen sections in the latter. T h e paraffin series showed a lower frequency of positivity but one may presume that those that are positive have a particularly high level of p53 expression. In 304 cases (Thor et al., 1991), 22% were positive with Ab1801 on archival material. This was an independent prognostic factor for overall and disease-free survival in node-positive and -negative patients. Other independent significant factors in this series were ER and lymph node status. In the series of Ostrowski et al. (1991), 36% were positive on frozen sections. There was a worse survival for p53-positive patients, but this did not reach significance. Iwaya et al. (1991) showed an association with ErbB-2 expression, and a poorer survival. However, this was not significant and patients undergoing palliative surgery and curative surgery were isolated in the assessments. Clearly large series with adequate follow-up are needed to further assess the prognostic value of p53 expression. The association of p53 with other standard poor prognostic factors has been assessed. These include node status and differentiation. Most groups have found an association with high-grade tumors, but no association with lymph node (Table 11) status. This again may suggest an important early role of p53 mutation in tumor proliferation. It has been shown that expression of p53 in rodent cell lines can increase the metastatic potential (Pohl et al., 1988), so it is perhaps not surprising that there may be an association of p53 expression with a more aggressive phenotype, although the direct metastatic effect has not been confirmed in human breast cancer. XVIII. In Sifu Lesions

Assessing p53 by immunochemistry has shown that it is frequently expressed (13-22%) (Davidoff et al., 1991a; Bartek et al., 1990a) in in situ lesions and that for the same tumor in situ and invasive lesions express the same mutation. It is also present in lymph node metastases and there is concordance of expression (Bartek et al., 1990a).Again the same mutation was expressed in the primary and secondary lesions (Davidoff et al., 1991a). p53 seems to be expressed in all stages of breast cancer. This fits with its potential importance in the early steps of differentiation. Another mechanism to consider by which p53 might be particularly expressed with certain growth factors and receptors is the following: p53 might be a rate-limiting step in proliferation; since the high expression of growth factors or receptors alone is insufficient to transform cells, removing the rate-limiting effect with p53 or retinoblastoma gene

84

ADRIAN L. HARRIS

product mutations may allow rapid proliferation in an uncontrolled manner. XIX. Therapeutic Possibilities Because mutant p53 is so widely expressed in human cancer (e.g., breast, colon, lung, bladder, brain, cervical, esophageal, carcinomas, melanoma, sarcoma, leukemia, and hepatoma) it is an attractive target for therapy. Whether it is a late change producing a more aggressive phenotype or an early change involved in carcinogenesis, it is still tumor specific. Thus, screening cervical smears, breast aspirates, and bladder sediments may be very useful. Although negative results will not exclude cancer, positive results would be a strong indication for further investigation and treatment. Since the protein is mutant, it is possible that an immune response could be generated. Antigen presentation of HLA class I molecules is essential for cytotoxic T cell responses (Tanaka et al., 1988). Endogenous protein is presented by this route. It will be important to demonstrate that epitopes presented from mutant p53 differ from wild-type p53 or, alternatively, that there is a much higher level of expression of such epitopes on the surface of cancer cells. It may be that several different peptides will be required to generate an immune response or that denatured p53 is differentially degraded and presented. There are several new methods by which T cell cytotoxic responses could be generated. These include peptide vaccination with synthetic lipopeptide vaccine (Deres et al., 1989), use of recombinant vaccinia virus (Murray et al., 1990), and recombinant BCG (Stover et al., 1991). If HLA class I presentation is essential then one way in which tumor growth could be enhanced would be loss of HLA expression (Sobel, 1990). Several authors have demonstrated suppression of HLA class I expression in human tumors (Smith et al., 1989; Natali et al., 1989). Loss of function of a transcription factor interacting with the HLA class I gene enhancer may be a mechanism of suppression (Singh, 1988; Henseling et al., 1990). We therefore evaluated HLA class expression and p53 to see if the presence of mutant p53 resulted in the selective loss of class I molecules. T h e results show that loss of HLA expression was common in breast cancer but is independent of p53 (Table IV). In many cases p53 and HLA class I expression were concomitant, suggesting that therapy would be appropriate in these cases. Antibody responses to mutant p53 have been reported in some patients (Crawford et al., 1982), although

p53

Loss

OF

TABLE IV HLA-A2 EXPRESSION IN PRIMARY BREASTCANCER AND DETECTION OF MUTANTp53a Number p53 positive*

Sample number and characterization ~

85

EXPRESSION IN HUMAN BREAST CANCER

~

~

~

~~

HLA-A2 stroma+, tumor+ (66) HLA-A2 a r o m a + , tumor- (22) HLA-A2 stroma-, tumor- (73)

~~

Number p53 negative* ~

43 (65) 1 3 (59) 31 (42)

~~

23 (35) 9 (41) 42 (57)

a Tumors were stained with AhMA2. 1 to HLA-A2 and samples with loss of expression compared to normal breast epithelium and stroma (HLA-AP-positive cases only) are designated stroma+, tumor-. Eightyeight of 161 cases were A2 positive. p53 expression was assessed by Ab240. b Numbers in parentheses represent percentages of total.

different epitopes would be involved in the B cell response. This may have been due to release of mutant p53 from necrotic tissue. Other possibilities may include drugs that bind specifically to the mutant form, thus providing selective localization and tumor toxicity. In the case of viral carcinogenesis, p53 is normal but may be important in human cervical cancer and interaction with papilloma virus proteins. In this case drugs that could block binding of viral proteins of p53 may also be effective.

ACKNOWLEDGMENTS I would like to thank Diana Barnes and Christine Fisher for making data available before publication.

REFERENCES Addison, C., Jenkins, J. R., and Sturzbecher, H.-W. (1990). Oncogene 5 , 423-426. Arai, N., Nornura, D., Yokota, K., Wolf, D., Brill, E., Shohat, O., and Rotter, V. (1986). Mol. Cell. Biol. 6, 3232-3239. Baker, S. J., Fearon, E. R., Nigro,]. M., Hamilton, S. R., Preisinger, A. C., Jessup,J. M., van Tuinen, P., Ledbetter, D. H., Barker, D. F., Nakamura, Y., White, R.,and Vogelstein, B. ( 1989). Science 244, 2 17-22 1. Baker, S. J., Preisinger, A. C., Jessup, J. M., Paraskeva, C., Markowitz, S., Willson, J. K. V., Hamilton, S., and Vogelstein, B. (1990). Cancer Res. 50, 7717-7722. Bartek, J., Bartkova, J., Vojtesek, B., Staskova, Z., Rejthar, A., Kovarik, J.. and Lane, D. P. (I99Oa). Int. J . Cancer 46, 839-844. Bartek, J., Iggo, R., Gannon, I., and Lane, D. P. (1990b). Oncogene 5 , 893-899. Benchimol, S., Pim, D., and Crawford, L. (1982). E M B O J . 1, 1055-1062.

86

ADRIAN L. HARRIS

Bischoff, J. R., Friedman, P. N., Marshak, D. R., Prives, C., and Beach, D. (1990).Proc. Natl. Acad. Sci. U.S.A. 87, 4766-4770. Bressac, B., Kew, M., Wands, J., and Ozturk, M. (1991). Nature (London) 350, 429-430. Cattoretti, G., Andreola, S., Clemente, C., DAmato, L., and Rilke, F. (1988a).Sr.1. Cancer 57,353-357. Cattoretti, G., Rilke, F., Andreola, S., DAmato, L., and Delia, D. (1988b).Int.1. Cancer 41, 178-1 83. Chen, L.-C., Neubauer, A., Kurisu, W., Waldman, F. M., Ljung, B.-M., Goodson, W., 111, Goldman, E. S., Moore, D., Balazs, M., Liu, E., Mayall, B. H., and Smith, H. S. (1991). Proc. Natl. Acad. Sci. U.S.A. 88, 3847-3851. Chen, P.-L., Chen, Y.,Bookstein, R., and Lee, W.-H. (1990). Science 250, 1576-1580. Ciechanover, A., DiGiuseppe, J. A., Bercovich, B., Orian, A., Richter, J. D., Schwartz, A. L., and Brodeur, G. M. (1991). Proc. Null. Acad. Sci. U.S.A. 88, 139-143. Coles, C., Thompson, A. M., Elder, P. A., Cohen, B. B., MacKenzie, I. M., Cranston, G., Chetty, U., MacKay, J., MacDonald, M., Nakamura, Y., Hoyheim, B., and Steel, C. M. (1990). Lancet 336, 761-763. Crawford, L. V., Pim, D. C., and Bulbrook, R. D. (1982). Znt. J . Cancer 30, 403-408. Crawford, L. V., Pim, D. C., and Lamb, P. (1984). Mol. Biol. Med. 2, 261-272. Cropp, C. S., Lidereau, R., Campbell, G., Champene, M. H., and Callahan, R. (1990).Proc. Natl. Acad. Sci. U.S.A. 87, 7737-7741. Dang, C. V., and Lee, W. M. F. (1989).J. Biol. Chem. 264, 18019-18023. Davidoff, A. M., Humphrey, P. A., Iglehart, J. D., and Marks, J. R. (1991a).Proc. Natl. Acad. Sci. U.S.A. 88, 5006-5010. Davidoff, A. M., Kerns, B.-J. M., Iglehart, J. D., and Marks, J. R. (1991b). Cancer Res. 51, 2605-2610. Deres, K., Schild, H., Wiesmuller, K.-H., Jung, C., and Rammensee, H.-G. (1989). Nature (London) 342, 561-564. Devilee, P., van den Broek, M., Mannens, M., Slater, R., Cornelisee, C. J., Westerveld, A,, and Khan, P. M. (1991).Int. J. Cancer 47, 817-821. Diller, L., Kassel, J., Nelson, C. E., Gryka, M. A., Litwak, G., Cebhardt, M., Bressac, B., Ozturk, M., Baker, S. J., Vogelstein, B., and Friend, S. H. (1990). Mol. Cell. Biol. 10, 5772-5781. . Eliyahu, D., Raz, A., Gruss, P., Givol, D., and Oren, M. (1984). Nature ( L o d o n ) 312,646649. Eliyahu, D., Michalovitz, D., Eliyahu, S., Pinhasi-Kimhi, O., and Oren, M. (1989). Proc. Nutl. Acad. Sci. U.S.A. 86, 8763-8767. Fields, S., and Jang, S. K. (1990). Science 249, 1046-1049. Finlay, C. A., Hinds, P. W., Tan, T.-H., Eliyahu, D., Oren, M., and Levine, A. J. (1988).Mol. Cell. Biol. 8, 531-539. Finlay, C. A., Hinds, P. W., and Levine, A. J. (1989). Cell ( C a m b d g e , Mass.) 57, 1083-1093. Friedman, P. N., Kern, S. E., Vogelstein, B., and Prives, C. (1990). Proc. Natl. Acad. Sci. U.S.A. 87, 9275-9279. Cannon, J. V., and Lane, D. P. (1987). Nature (London) 329, 456-458. Cannon, J. V., and Lane, D. P. (1991). Nature (London) 349, 802-806. Cannon, J. V., Greaves, R., Iggo, R., and Lane, D. P. (1990). EMBOJ. 9, 1595-1602. Halevy, O., Michalovitz, D., and Oren, M. (1990). Science 250,113-116. Harris, A. L. (1991). Nature, News €3Views 350, 377-378. Henseling, U., Schmidt, W., Scholer, H. R., Gruss, P., and Hatzopoulos, A. K. (1990).Mol. Cell. Biol. 10(8), 4100-4109. Hinds, P. W., Finlay, C. A., and Levine, A. J. (1989).J . Virol. 63, 739-746.

p53

EXPRESSION IN HUMAN BREAST CANCER

87

Hinds, P. W., Finlay, C. A., Quartin, R. S., Baker, S. J., Fearon, E. R., Vogelstein, B., and Levine, A. J. (1990). Cell Growth Dzffer. 1, 571-580. Hollstein, M., Sidransky, D., Vogelstein, B., and Harris, C. C. (1991). Science 253, 49-53. Horak, E., Smith, K., Bromley, L., LeJeune, S., Greenall, M., Lane, D. P., and Harris, A. L. (1991). Oncogene 6, 2277-2284. Hsu, I. C., Metcalf, R. A., Sun, T.,Welsh, J. A., Wang, N. J., and Harris, C. C. (1991). Nature (London) 350, 427-428. Iwaya, K., Tsuda, H., Hiraide, El., Tamaki, K., Tamakuma, S., Fukutomi, T., Murai, K., and Hirohashi, S. (1991). Jpn.1. Cancer Res. 82, 835-840. Jenkins, J. R., Rudge, K., and Currie, G. A. (1984). Nature (London) 312, 651-654. Kovach, J. S., McGovern, R. M., Cassady, J. D., Swanson, S. K., Wold, L. E., Vogelstein, B., and Sommer, S . S. (1991).J. Natl. Cancer Inst. 83, 1004-1009. Laiho, M., DeCaprio, J. A., Ludlow, J. W., Livingston, D. M., and Massague, J. (1990). Cell (Cambndge, Mass.) 62, 175- 185. Lamb, P., and Crawford, L. V. (1986). Mol. Cell. Biol. 6(5), 1379-1385. Lane, D. P., and Crawford, L. V. (1979). Nature (London) 278, 261-263. Levine, A. J., Momand, J., and Finlay, C. A. (1991). Nature (London) 351, 453-460. Linzer, D. I. H., and Levine, A. J. (1979). Cell (Cambridge, Mass.) 17, 43-52. Mackay, J., Steel, C. M., Elder, P. A., Forrest, A. P. M., and Evans, H. J. (1988). Lancet 2, 1384-1385. Malkin, D., Li, F. P., Strong, L. C., Fraumeni, J. F., Jr., Nelson, C. E., Kim, D. H., Kassel, J., Gryka, M. A., Bischoff, F. Z., Tainsky, M. A., and Friend, S. H. (1990). Science 250, 1233-1238. Martinazzi, M., Crivelli, F., Grass, C., and Zampatti. (1990). Znt.J. Biol. Markers 5, 138-144. Martinez, J., Georgoff, I., Martinez, J., and Levine, A. J. (1991). Genes Dev. 5, 151-159. Matlashewski, G., Banks, L., Pim, D., and Crawford, L. (1986). Eur.1. Biochem. 154,665672. Meek, D. W., Simon, S., Kikkawa, U., and Eckhart, W. (1990). EMBOJ. 9(10), 3253-3260. Milner, J. (1991). Curr. Opin. Cell Biol. 3, 282-286. Milner, J., and Medcalf, E. A. (1991). Cell (Cambridge, Mass.)65, 765-774. Milner, J., Cook, A,, and Mason, J. (1990). EMBOJ. 9(9), 2885-2889. Mowat, M. A., Cheng, A., Kimura, N., Bernstein, A., and Benchimol, S. (1985). Nature (London) 314, 633-636. Murray, R. J., Kurilla, M. G., Griffin, H. M., Brooks, J. M., Mackett, M., Arrand, J. R., Rowe, M., Burrows, S. R., Moss, D. J., Kieff, E., and Rickinson, A. B. (1990). Proc. Nutl. Acad. Sci. U.S.A.87, 2906-2910. Natali, P. G., Nicotra, M. R.,Bigotti, A,, Venturo, I., Marcenaro, L., Giacomini, P., and Russo, C. (1989). Proc. Acud. Sci. U.S.A. 86, 6719-6723. Nigro, J. M., Baker, S. J., Preisinger, A. C., Jessup, J. M., Hostetter, R., Cleary, K., Bigner, S . H., Davidson, N., Baylin, S., Devilee, P., Glover, T., Collins, F. S., Weston, A., Modali, R., Harris, C. C., and Vogelstein, B. (1989). Science 342, 705-708. ORourke, R. W. O., Miller, C. W., Kato, G. J., Simon, K. J., Chen, D.-L., Dang, C. V., and Koeffler, H. P. (1990). Oncogene 5, 1829-1832. Ostrowski, J. L., Henry, L., Wright, C., Henry, J. A,, Hennessy, C., Lennard, T. J. W., Angus, B., and Horne, C. H. W. (1991).J. Puthol. 164, 75-82. Pietenpol, J. A., Stein, R. W., Moran, E., Yaciuk, P., Schlegel, R., Lyons, R. M., Pittelkow, M. R., Munger, K., Howley, P. M., and Moses, H. L. (1990). Cell (Cambridge, Muss.) 61, 777-785. Pohl, J., Goldfinger, N., Radler-Pohl, A., Rotter, V., and Schirrmacher, V. (1988). Mol. Cell. Biol. 8(5), 2078-208 1.

88

ADRIAN L. HARRIS

Prosser, J., Thompson, A. M., Cranston, G., and Evans, H. J. (1990). Oncogene 5, 15731579.

Prosser, J., Elder, P. A., Condie, A,, MacFadyen, I., Steel, C. M., and Evans, H. J. (1991). Br. J. Cancer 63, 18 1- 184. Raycroft, L., Wu, H., and Lozano, G. (1990). Science 249, 1049-1051. Rideout, W. M., 111, Coetzee, G. A., Olumi, A. F., and Jones, P. A. (1990). Science 249, 1288- 1290.

Samad, A., and Carroll, R. B. (1991). Mol. Cell. Biol. 11(3), 1598-1606. Sato, T., Tanigami, A., Yamakawa, K., Akiyama, F., Kasumi, F., Sakamoto, G., and Nakamura, Y. (1990). Cancer Res. 50, 7 184-7 189. Shaulsky, G., Goldfinger, N., Ben-Ze'ev, A,, and Rotter, V. (1990). Mol. Cell. B i d . 10(12), 6565-6577.

Singh, H., LeBowitz, J. H., Baldwin, A. S., Jr., and Sharp, P. A. (1988). Cell (Cambridge, M a s . ) 52, 415-423. Singh, S., Simon, M., Meybohm, I., Jonat, W., Mass, H., and Goedde, H. W. (1991). Proc. Am. Assoc. Cancer Res. 32, 307. Smith, M. E. F., March, S. G. E., Bodmer, J. G., Gelsthorpe, K., and Bodmer, W. F. (1989). Proc. Natl. Acad. Sci. U.S.A. 86, 5557-5561. Sobel, M. E. (1990). J. Natl. Cancer Inst. 82, 267-276. Soussi, T., Caron d e Fromentel, C., and May, P. (1990). Oncogene 5, 945-952. Srivastava, S., Zou, Z., Pirollo, K., Blattner, W., and Chang, E. H. (1990). Nature (London) 348, 747-748.

Stover, C. K., d e la Cruz, V. F., Fuerst, T. R., Burlein, J. E., Benson, L. A , , Bennett, L. T., Bansel, G. P., Young, J. F., Lee, M. H., Hatfull, G. F., Snapper, S. B., Barletta, R. G., Jacobs, W. R., Jr., and Bloom, B. R. (1991). Nature (London) 351, 456-460. Sturzbecher, H.-W., Chumakov, P., Welch, W. J., and Jenkins, J. R. (1987). Oncogene 1,20121 1. Sturzbecher, H.-W., Maimets, T., Chumakov, P., Brain, R., Addison, C., Simanis, V., Rudge, K., Philip, R., Grimaldi, M., Court, W., and Jenkins, J. R. (1990). Oncogene 5 , 795-801.

Tanaka, K., Yoshioka, T., Bieberich, C., and Jay, G. (1988). Annu. Rev. Immunol. 6,359-380. Thompson, A. M., Steel, C. M., Chetty, U., Hawkins, R. A., Miller, W. R., Carter, D. C., Forrest, A. P. M., and Evans, H. J. (1990a). Br. J . Cancer 61, 74-78. Thompson, A. M., Steel, C. M., Foster, M. E., Kerr, D., Paterson, D., Deane, D., Hawkins, R. A,, Carter, D. C., and Evans, H. J. (1990b). Br. J. Cancer 62, 78-84. Thor, A., Edgerton, S., Moore, D., Chun-Chen, L., and Smith, H. S. (1991). Proc. Am. Soc. Clin. Oncol. (ASCO) 10, 45. van Tuinen, P., Dobyns, W. B., Rich, D. C., Summers, K. M., Robinson, T. J., Nakamura, Y., and Ledbetter, D. H. (1988). Am. J. Hum. Genet. 43, 587-596. Varley, J. M., Brarnmar, W. J.. Lane, D. P., Swallow, J. E., Dolan, C., and Walker, R. A. (1991). Oncogene 6, 413-421. Vogelstein, B., Fearon, E. R., Kern, S. E., Hamilton, S. R., Preisinger, A. C., Nakamura, Y., and White, R. (1989). Science 244, 207-21 1. Weintraub, H., Hauschka, S., and Tapscott, S. J. (1991). Proc. Natl. Acad. Scz. U.S.A. 88, 4570-457 1 . Werness, B. A,, Levine, A. J., and Howley, P. M. (1990). Science 248, 76-79. Wilcock, D., Lane, D. P. (1991) Nature (London) 349, 429-431. Wyllie, F. S., Dawson, T., Bond, J. A., Goretzki, P., Game, S., Prime, S., and WynfordThomas, D. (1991). Mol. Cell Endocrinol. 76, 13-21.

c-erbA: PROTOONCOGENE OR GROWTH SUPPRESSOR GENE? Klaus Damm Gene Expression Laboratory, The Salk Institute for Biological Studies, La Jolla, California 92037, and Department of Neuroendocrinology, Max-Planck-Institute for Psychiatry, 8000 Munich 40, Germany

I. Introduction Avian Erythroblastosis Virus as a Model for Cooperativity of Oncogenes T h e Protooncogene c-ei-bA Encodes a Thyroid Hormone Receptor Multiple c-erbA Loci Structural Differences between v-erbA and c-erbA Functional Properties of the ErbA Proteins A. Hormone Binding B. DN.4 Binding C. Oncogene (v-erbA) and Protooncogene (c-erbA) Act as Transcription Factors D. Cotransfection and Competition VII. Mutations Affecting the Biological Activity of v-erbA VIII. c-ErbA Regulation of Erythroid Differentiation and Gene Expression IX. c-erbA: Protooncogene o r Growth Suppressor Gene? X. Mutations Affecting c-erbA Function XI. Current Concepts and Open Questions References 11. I1 I. IV. V. VI.

I. Introduction

Cellular growth and differentiation are mutually exclusive and cancer has been commonly linked to aberrant differentiation and a failure of the tumor cells to differentiate. Tumor cells usually show multiple genetic lesions, including gene amplification, chromosomal translocations, and point mutations. In a number of cases cooperativity between oncogenes has been observed and a multistep transformation process may involve a primary oncogene providing a continuous proliferation signal with a second oncogene blocking the ability of these cells to differentiate. In a distinct mechanism cellular transformation may also be achieved by the inactivation of genes whose normal function is to constrain cell growth by either suppressing proliferation or inducing differentiation, thereby acting normally in a fashion opposite to the oncogenes, and it is the loss or inactivation of these genes that would trigger cancer development. The efficient cellular transformation induced by retroviruses carrying a pair of cooperating oncogenes might be achieved by a combination of these 89 ADVANCES IN CANCER RESEARCH, VOL. 59

Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.

90

KLAUS DAMM

two mechanisms. One well-studied example is the v-erbA and v-erbB oncogenes of the avian erythroblastosis virus (AEV), where v-mbA function is needed to provide a complete block in differentiation of erythroid precursor cells transformed by v-erbB. In this article, the molecular mechanisms by which v-erbA potentiates transformation and influences the growth requirements of cells will be discussed in the context of normal c-erbAlthyroid hormone receptor function.

II. Avian Erythroblastosis Virus as a Model for Cooperativity of Oncogenes The avian erythroblastosis virus (AEV) represents an acute chicken retrovirus capable of inducing lethal erythroleukemia and sarcomas in vivo, properties paralleled by its transformation of hematopoietic cells and fibroblasts in vitro (for review, see Graf and Beug, 1978, 1983). Erythroblastosis induced by AEV is characterized by the appearance of large, blastlike cells in the peripheral blood as early as 5 days after intravenous injection. Infection of chicken bone marrow cells with AEV in uitro leads to the appearance of rapidly proliferating blastlike cells. In contrast to normal erythroid progenitor cells, AEV-transformed erythroblasts in culture are capable of extensive self-renewal in the absence of terminal differentiation throughout their life span of 10-40 generations. This suggests that the virus transforms its target cells either by arresting their differentiation or by inducing their proliferation. The cells that are susceptible to transformation by AEV have been characterized by using a series of antisera specific for cells of various hematopoietic cell lineages. These studies indicate that most of the target cells are at the burst forming unit-erythroid (BFU-E) stage. The molecular cloning of AEV revealed that the viral genome contains two oncogenes derived from the chicken genome (Fig. 1; Vennstrom et al., 1980; Vennstrom and Bishop, 1982). The first oncogene, v-erbB, is transduced from the gene that encodes the cell surface receptor for epidermal growth factor (Downward et al., 1984; Ullrich et al., 1984),whereas the second oncogene, v-erbA, exhibits extensive structural similarities to genes encoding intracellular receptors for steroid hormones and was subsequently shown to be a mutated derivative of the chicken thyroid hormone receptor a gene (Sap et al., 1986; Weinberger et al., 1986). An analysis of the transforming capacities of viral deletion mutants in both the v-erbA and the v-erbB genes led to the following conclusions (Fig. 2). Neither of the two mutant types is capable of inducing a typical AEV

-08kb+-

LTR

Agag

1 z k b __c

- -

erbA

lSkb

erbB

LTR

FIG. 1. Structure and expression of the proviral genome of AEV. In AEV the retroviral pol and e m genes have been replaced by two cell-derived oncogenes, erbA and erbB. The genome is transcribed into two RNAs, a genomic length RNA coding for the 75-kDa GagErbA fusion protein and a subgenomic RNA encoding the 68-kDa ErbB glycoprotein.

v-erbA + v-erbB

ftc

c) v-erbB

v-erbA

-0

ftc

FIG. 2. Contribution of v-erbA and v-erbB to the leukemic phenotype. Normal erythroblasts exhibit only a limited self-renewal capacity and differentiate to mature erythrocytes. v-erbE induces infected erythroid progenitors to self-renew and abrogates the requirements of these cells for erythroid growth factors, e.g., erythropoietin. However, v-erbB does not completely block erythroid differentiation and causes only a weak erythroleukemia in chicken. v-erbA cooperates with v-erbB by completely arresting their spontaneous differentiation and bypassing the complex growth requirements of v-erbBtransformed cells. Acting by itself, v-erbA is sufficient to arrest differentiation of normal erythroid progenitors but is unable to induce sustained self-renewal.

92

KLAUS DAMM

erythroblastosis in uiuo or in uitro. However, v-erbB is the primary transforming gene of AEV since it is sufficient for transformation of fibroblasts and erythroid progenitor cells, the two types of target cells that AEV interacts with in uiuo and in uitro (Frykberg et al., 1983; Fung et al., 1983; Sealy et al., 1983; Yamamoto et al., 1983). The described effects of v-erbB in chicken cells include its ability to abrogate the requirement for mitogens such as epidermal growth factor (EGF), transforming growth factor a (TGFa), or the erythroid growth factor erythropoietin, the induction of an abnormal self-renewal in erythroid precursor cells, and phenotypical changes typical for transformed fibroblasts (Beug et al., 1982, 1985; Royer-Pokora et al., 1978; Khazaie et al., 1988; Pain et al., 1991). However, leukemia cells induced by v-erbB alone resemble normal CFU-E in that they exhibit a significant potential to differentiate spontaneously. Furthermore, they grow only under culture conditions similar to those that promote the differentiation of normal erythroid cells in uitro (Beug et al., 1982, 1985). T h e v-erbA oncogene, on the other hand, does not exhibit a strong transforming capacity by itself. In chicken fibroblasts v-erbA elicits only modest biological effects such as enhancement of agar colony formation, in vitro life span, and production of extracellular matrix proteins (Frykberg rt al., 1983; Jansson ct ul., 1987; Grandrillon et al., 1987). In hematopoietic cells, v-erbA by itself induces an aberrant, largely immature phenotype, characterized by the coexpression of erythroblast arid erythrocyte differentiation antigens and indicative of its ability to arrest differentiation (Frykberg et a[., 1983; Grandrillon et al., 1989; Schroeder et al., 1990). Important clues as to the respective roles of v-erbA in erythroblast transformation came from the analysis of two AEV derivatives, AEVtdSSY,a transformation defective mutant of AEV unable to transform erythroblasts (Royer-Pokora et al., 1979), and its revertant AEV' I Z , obtained after in uzuo passage of the AEVtclS.iSmutant (Damn1 et al., 1987). Nucleotide sequence analysis of the cloned viral genomes revealed that the v-ErbA protein of AEVtdS5gcontained among other substitutions one amino acid change at position 144 of c-ErbA, a Pro Arg change (Damm et al., 1987). The v-ErbA protein of AEV'12 had reverted to a proline in this position. T h e v-ErbB genes of both the mutant and the revertant carry identical deletions in their 3' ends: a 306nucleotide deletion removed most of the region located after the domain of v-ErbB homologous to tyrosine kinases. To analyze the role of these mutations in v-ErbA and v-ErbB for erythroblast transformation, recombinant viruses containing all possible combinations of AEV"', AEVtd35g, and AEVrl* v-erbA and v-erbB genes were constructed. T h e results of subsequent bone marrow transformation experiments with the resulting

c-erbA:

PROTOONCOGENE OR GROWTH SUPPRESSOR GENE?

93

viruses showed that the mutant v-erbB genes were unable to transform erythroblasts in the absence of an active v-erbA gene: v-erbA of AEV'" or AEVwtbut not AEVtd359was able to restore the erythroblast-transforming activity to the mutant v-erbB genes, confirming that the reversion in the AEVr12 genome had occurred in its v-erbA oncogene (Damm et al., 1987). These and other studies showed that v-ErbA contributes to the transformed phenotype in two ways. First, v-ErbA enhances the transforming activity of v-ErbB, and restores the erythroblast-transforming potential of transformation-defective v-ErbB mutants (Damm et al., 1987; Jansson et al., 1987) by blocking the residual, spontaneous differentiation program in erythroid precursors. In addition, v-ErbA enables a variety of tyrosine kinase-encoding oncogenes (e.g., v-src, v-fps, v-sea, and v-fms) as well as v-Ha-ras to transform erythroblasts, all of which by themselves are unable to do so (Kahn et al., 1986). Second, v-ErbA abrogates the complex growth requirements of transformed erythroblasts and enables these cells to proliferate under a wide range of pH and HC0,-/Na+ concentration (Beug et al., 1985; Damm et al., 1987; Kahn et al., 1986). III. The Protooncogene c-erbA Encodes a Thyroid Hormone Receptor

The cloning of the human glucocorticoid receptor (hGR) cDNA represented a major breakthrough in our understanding of the mechanisms by which v-ErbA exerts its oncogenic activity. A comparison of the amino acid sequences revealed a remarkable homology between the carboxyterminal half of hGR and the v-ErbA oncoprotein (Hollenberg et al., 1985; Weinberger et al., 1985). The relationship of hormone receptors to v-ErbA was independently confirmed by the cloning of receptors for all major classes of steroid hormones as well as retinoic acid (vitamin A) and vitamin D, receptors (for review, see Evans, 1988; Green and Chambon, 1988; Beato, 1989). T h e region with maximal homology between hGR and v-ErbA corresponds to a cysteine-rich region; 9 out of 10 cysteine residues of hGR are conserved in this high-identity region with v-ErbA (Weinberger et al., 1985; Green et al., 1986). The presence of this structure, which embodies the DNA-binding domain in the hormone receptors, suggests that the v-ErbA oncoprotein itself might bind to DNA and may function in oncogenesis by inappropriately modulating transcription of specific target genes in the host cell. Significant homology was also found in the ligand-binding domain, which suggests that the cellular homolog of v-erbA, the protooncogene c-erbA, encodes a receptor for a steroid-related ligand. Two groups succeeded in the molecular cloning

94

KLAUS DAMM

and characterization of the c-erbA protooncogene product: Sap et al. (1986) cloned the authentical c-erbA protooncogene from chicken embryo and Weinberger et al. (1986) identified a c-erbA gene from a human placenta library. To the surprise of most of the “receptorologists,” both groups could demonstrate the c-erbA protooncogene products do not bind steroids but bind thyroid hormones [3,5,3’-triiodo-~-thyronine (T3), 3,5,3’,5’-tetraiodo-~-thyronine (TJ, 3,5,3’-triiodo-~-thyroacetic acid (TRIAC)] with high affinity and specificity and may therefore be the chicken and human receptors for these hormones. The identity of c-erbA as a thyroid hormone receptor (TR) was further supported by the nuclear localization of the chicken c-erbA product (Sap et al., 1986).

IV. Multiple c-erbA Loci

A surprising observation in the comparison of the chicken and human c-ErbA amino acid sequences was the relatively low (91%) homology in the DNA-binding domain, which is in contrast to human and chick estrogen receptors or human and rat glucocorticoid receptors, where the DNA-binding domains are entirely conserved. The two c-erbA products may thus correspond to similar but not identical TRs, suggesting the possibility of multiple TRs and corresponding genes in a given species. Indeed, previous studies localized a v-erbA related sequence to human chromosome 17 (17q11.2-17q21) Jansson et al., 1983; Dayton et al., 1984; Spurr et al., 1984) whereas the cloned human c-erbA is localized to chromosome 3 ( 3 ~ 2 2 - 3 ~ 2 4 . 1(Weinberger ) et al., 1986; Drabkin et al., 1988). Based on amino acid homologies and chromosomal localizations, thyroid hormone receptors from human, rat, mouse, chicken, and Xenopus laevis have been classified into two groups, a and (Fig. 3; Sap et al., 1986; Weinberger et al., 1986; Thompson et al., 1987; Benbrook and Pfahl, 1987; Nakai et al., 1988a,b; Lazar et al., 1988, 1989a; Murray et al., 1988; Koenig et al., 1988; Izumo and Mahdavi, 1988; Prost et al., 1988; Hodin et al., 1989; Forrest et al., 1990a; Yaoita et al., 1990). The c-erbAITRa gene locus produces different mRNAs that give rise to multiple divergent receptor proteins (Thompson et al., 1987; Benbrook and Pfahl, 1987; Nakai et al., 1988a; Lazar et al., 1989b; Murray et al., 1988; Mitsuhashi et al., 1988; Izumo and Mahdavi, 1988). The proteins encoded are identical for the first 370 amino acids and then diverge completely. This digression of TRa-2 from the sequence in rat and human T R a spans the ligand-binding domain, probably accounting for the fact that the protein encoded by TRa represents an authentic receptor whereas the protein encoded by TRa-2 does not bind hor-

c-erbA:

95

PROTOONCOGENE OR GROWTH SUPPRESSOR GENE?

hormone binding

DNA binding

transactivation

+

TRa2

TRa

+

+

+

TRP

+

+

+

TRP2

+

+

+

FIG.3. Comparison of c-ErbAIthyroid hormone receptor isoforms. Homology of different domains of the TRs are compared; numbers refer to the percentage similarity at the amino acid level and domains with little or no homology are indicated with different shading. Functional properties such as hormone binding, DNA binding, and trans-activation, are listed.

mone. Both proteins, however, contain the DNA-binding domain and can interact with putative TREs (Koenig et al., 1988; Izumo and Mahdavi, 1988; Lazar et al., 1988). Furthermore, in rat and human there seem to be multiple splicing variants of the TRa-2 (Izumo and Mahdavi, 1988); at least one of these forms differs in its amino acid sequence by a deletion of 39 amino acids at the beginning of the divergent carboxyterminal region (Mitsuhashi et al., 1988; C . Thompson, K. Damm, and R. M. Evans, unpublished results). In the chicken, no TRa-2 has been found and the 4.5-kb mRNA is apparently expressed ubiquitously, with elevated levels during the late stages of erythrocyte differentiation (Sap et al., 1986; Hentzen et al., 1987; Forrest et al., 1990b).In rat and human, Rev-ErbAa, a non-T,-binding variant of the hormone receptor family, is also encoded at this genomic locus (Miyajima et al., 1989; Lazar et al., 1989a). The DNA strand coding for Rev-ErbAa is opposite of that encoding T R a proteins. As a result, the mRNAs encoding Rev-ErbA and TRa-2 are complementary for 269 nucleotides whereas T R a and RevErbA mRNAs are derived from nonoverlapping regions (Miyajima et al., 1989; Lazar et al., 1990). The c-erbA/TRP gene locus encodes at least two receptors, TRP, the human c-erbA clone described by Weinberger et al. (1986), and TRP-2, which differs completely at the N terminus but represents a bona fide T R by the criteria of T, and DNA binding (Hodin et al., 1989). TRP is expressed in most tissues of rat and humans, with somewhat elevated levels in liver, kidney, and hypothalamus. An intriguing aspect of the rat TRP-2 form is that it is expressed only in the pituitary gland and that its

96

KLAUS DAMM

expression is down regulated by T, (Hodin et al., 1989, 1990). In the chicken, temporal and tissue-specific expression has been reported, suggesting specific developmental functions for TRP (Forrest et al., 1990a). In human, the syndrome of generalized thyroid hormone resistance has been tightly linked to the TRP gene and multiple mutations in T R P have been reported in affected kindreds (Sakurai et al., 1989; Usala et al., 1990, 1991a,b). V. Structural Differences between v-erbA and c-erbA

Examination of the v-ErbA amino acid sequence reveals that the v-ErbA protein represents a heavily mutated version of the chicken c-ErbA/TRa (Fig. 4; Sap et al., 1986). In the AEV viral genome the erbA-

A 1

51

118

408

c-erbA 1

252

291

356

639

v-erbA B c-erbA v-erbA

1 235

c-erbA v-erbA

55 GDKATGYHYR CITCEGCKGF FRRTIQKNLH PTYSCKYDGC CVIDKITRNQ CQLCRFKKCI 295 GDKATGYHYR CITCEGCKSF FRRTIQKNLH PTYSCTYDGC CVIDKITRNQ CQLCRFKKCI

c-erbA v-erbA

115 SVGMAMDLVL DDSKRVAKRK LIEENRERRR KEEMJKSLQH RPSPSAEEWE LIHVVTEAHR 355 SVGMAMDLVL DDSKRVAKRK LIEENRERRR KEEMIKSLQH RPSPSAEEWE LIHVVTEAHR

c-erbA v-erbA

1 7 5 STNAQGSHWK QKRKFLPEDI GQSPMASMPD GDKVDLFAFS EFTKIITPAI TRVVDFAKKL 415 STNAQGSHWK QRRKFLLEDI GQSPMASMLD GDKVDLEAFS EFTKIITPAI TRVVDFAKNL

c-erbA v-erbA

235 PMFSELPCED QIILLKGCCM EIMSLRAAVR YDPESETLTL SGEMAVKREQ LKNGGLGWS

c-erbA v-erbA

295 DAIFDLGKSL SAFNLDDTEV ALLQAVLLMS SDRTGLICVD KIEKCQETYL LAF'EHYINYR 535 DAIFDLGKSL SAFNLDDTEV ALLQAVLLMS SDRTGLICVD KIEKCQESYL LAFEHYINYR

c-erbA v-erbA

355 KHNIPHFWPK LLMKVTDLRM IGACHASRFL HMKVECPTEL FPPLFLEVFE DQEV 595 KHNIPHFWSK LLMKVADLRM IGAYHASRFL HMKVECPTEL Sip-------- -QEV

MEQK PSTLDPLSEP EDTRWLDGKR KRKSSQCLVK SSMSGYIPSY LDKDEQCWC EDTRWLDGKE KRXSSQCLVK SSMSGYIPSC LDKDEQCWC

y v m w r m EGPAWTPLEP

475 PMFSELPCED QIILLKGCCM EIMSLRAAVR YDPESETLTL SGEMAVKREQ LKNGGLGWS

FIG. 4. (A) Schematic organization of chicken c-ErhA and v-ErhA protein sequences. DNA and T3/T4 refer to the DNA- and hormone-binding domains, respectively. (B) Amino acid differences between c-ErhA and v-ErbA. Amino acid sequences of chicken c-ErbA and v-ErbA are compared in the one-letter code. Differences are indicated by hold letters and shading;-, deletions.

c-erbA:

PROTOONCOGENE OR GROWTH SUPPRESSOR GENE?

97

specific sequences form one long open reading frame together with 5' gag sequences to code for a protein of 75 kDa (p75gUg-"lbA). The chicken c-erhA/TRa gene product is a protein of 46 kDa. In p75gag-erbAthe 253 N-terminal amino acids derived from the retroviral gag gene are fused in frame to chicken T R a sequences starting at amino acid 13. A comparison between the c-erbA/TRa coding sequence (Debuire et al., 1984; Damm et al., 1987; Sap et al., 1986) and the gag sequences of Rous sarcoma virus (Schwartz et al., 1983) demonstrated a region with 19 of22 nucleotide identities that bridges the junction between the gag and erbA domains in p75ga:afi-erhA. This suggests that c-erbA/TRa was fused to gag either by homologous recombination at the DNA level or during retrotranscription of c-erhA/TRa mRNA packaged into virions. The coding sequence of the erbA-specific part of the molecule shows 17 point mutations in v-erbA as compared to the c-rrbA/TRa sequence (Damm et al., 1987; Sap et al., 1986). These mutations lead to 12 amino acid substitutions, 2 of which are located between the gag and the DNA-binding domains, two within the DNA-binding domain, and the remaining eight in the region corresponding to the ligand-binding domain. Finally, ~75g"g-~*b* exhibits a nine-amino acid deletion three amino acids from the carboxy terminus. As a functional consequences of the mutations in the ligand-binding domain, the v-ErbA product is defective in binding thyroid hormones (Sap et al., 1986).

VI. Functional Properties of the ErbA Proteins Steroid/thyroid hormone receptors act as transcriptional regulatory proteins whose ability to control gene expression is dependent on the binding of their specific. ligand (for review, see Evans, 1988; Beato, 1989). Regulation of transcription results from the specific interaction of the hormone-receptor complex with "hormone response elements" in the promoter region of target genes. T h e cloning of the c-erhA gene provided the first opportunity to dissect the structure and functional properties of both the oncogene (v-erbA) and protooncogene (c-erbA /TR). Using cotransfection assays, the thyroid hormone receptors encoded by the c-erbA genes have been shown to act as hormone-inducible trans-acting factors similar to other hormone receptors of this class (Umesono et al., 1988; Koenig et al., 1988; Damm et al., 1989; Thompson and Evans, 1989). T h e most exciting development in our understanding of v-ErbA and c-ErbA/TR activity, however, has been the demonstration that v-ErbA and unliganded thyroid hormone receptors can bind thyroid hormone response elements (TREs) and may act as constitutive negative regulators of genes containing these elements (Damm et d.,

98

KLAUS DAMM

1989; Sap et al., 1989). Mutational analysis and structural comparisons of the different members of the steroid hormone receptor family identified specific domains responsible for hormone binding, DNA binding, and transcriptional activation (for review, see Evans, 1988; Green and Chambon, 1988; Beato, 1989). Similar domains and properties can be found in the v-erbAlc-erbAITR subfamily. A. HORMONE BINDING The initial characterization of in vitro-translated c-ErbA protein demonstrated that it can bind thyroid hormones with affinities (K = 0.2 nM) similar to the values obtained with thyroid hormone receptors in whole cells and in nuclear extracts (Sap et al., 1986; Weinberger et al., 1986). Sap et al. also tested chicken embryo fibroblasts containing either c-ErbA or v-ErbA for hormone binding. The nuclei from cultures of c-ErbA-expressing cells bound T, at levels similar to that found in GH1 cells, a rat pituitary cell line expressing endogenous thyroid hormone receptors. In contrast, only a low amount of specific T, binding was found in v-ErbA-expressing cells, demonstrating the defectiveness of the oncogene product for binding the ligand (Sap et al., 1986). To identify which of the mutations in v-ErbA conferred the inability to bind hormone, Mufioz et al. (1988) characterized a series of recombinants between the viral and cellular genes. Recombinant proteins synthesized in reticular lysates revealed that as expected the c-ErbA product binds T, 12 times better than v-ErbA. A progressive reduction in hormonebinding ability results from introducing into c-ErbA the mutations present in the ligand-binding domain of v-ErbA. These results reveal that the point mutations and the nine-amino acid deletion close to the C terminus act together in abolishing hormone binding and suggest that hormone-independent action by v-ErbA provided a selective advantage to AEV during its selection as a highly and acutely oncogenic virus strain (Mufioz et al., 1988). However, the cellular environment or external factors may also influence binding activity since it was recently shown that in the yeast Saccharomyces cerevisiae the v-ErbA protein responds to high concentrations of the thyroid hormone derivative TRIAC (Privalsky et al., 1990). Thus, cellular components influencing the ability of the v-ErbA product to respond to hormone may be important in the oncogenicity of v-ErbA.

-

B. DNA BINDING The identification of thyroid hormone response elements (TREs) in the promoter regions of regulated genes made it possible to demonstrate

c-erbA:

PROTOONCOGENE OR GROWTH SUPPRESSOR GENE?

99

sequence-specific DNA binding by both c-ErbA and v-ErbA proteins. Putative TREs have been identified in the rat growth hormone (Glass et al., 1987; Koenig et al., 1987; Bonde and Privalsky, 1990; Sap et al., 1990), a-myosin heavy chain (Izumo and Mahdavi, 1988), Moloney leukemia virus long terminal repeat (LTR) (Sap et al., 1989), the malic enzyme (Petty et al., 1990), and in the erythrocyte specific carbonic anhydrase I1 (Disela et al., 1991) genes. A variation of the rat growth hormone TRE that is characterized by the palindromic motif 5’-TCAGGTCATGACCTGA-3’ represents a very effective TRE (Glass el al., 1988; Umesono et al., 1988) and was used in the experiments by Damm et al. (1989) described below. When whole-cell extracts of COS cells expressing c-ErbA were incubated with a 32P-labeled TRE and separated in a nondenaturing acrylamide gel, a protein-DNA complex migrating more slowly than the free DNA was observed, regardless of whether the cells were cultured in the absence or presence of T,. Similarly, T, had no effect on the appearance of the retarded complex when it was added to extracts of transfected COS cells that were not exposed to the hormone in uivo. Hormone-independent formation of retarded complexes was also observed with extracts from cells expressing v-ErbA (Damm et al., 1989; Sap et al., 1989). Thus, the v-ErbA protein is able to selectively bind to specific DNA sequences in uitro, demonstrating that the amino acid changes in the DNA-binding domain of v-ErbA do not grossly interfere with DNA binding activity. However, these amino acid differences might nevertheless influence binding affinity and target gene specificity of v-ErbA (Damm et al., 1989; Sap et al., 1989; Bonde and Privalsky, 1990) because the first of the amino acid substitutions in the DNAbinding domain is located just after the first zinc finger and represents one of the three amino acids that are important in discriminating the sequence of the hormone response element (Umesono and Evans, 1989; Mader et al., 1989; Danielsen et al., 1989). The second amino acid substitution is in between the two cysteine residues at the base of the second zinc finger, a region that stands as a potential interface of receptor dimerization (Umesono and Evans, 1989). C. ONCOGENE (v-erbA) AND PROTOONCOGENE (c-erbA) ACT AS TRANSCRIPTION FACTORS The transcriptional activity of both c-ErbA and v-ErbA products was assessed by their ability to regulate expression of thyroid hormoneresponsive reporter genes (Damm et al., 1989). These constructs contain oligonucleotides corresponding to previously identified thyroid hormone response elements linked to a heterologous promoter and the

100

KLAUS DAMM

chloramphenicol acetyltransferase (CAT) gene (Umesono et al., 1988). In a cotransfection assay, expression plasmids containing c-erbA (rTRa cDNA) or the v-erbA oncogene under the transcriptional control of the Rous sarcoma virus (RSV) long terminal repeat were cotransfected with one of the reporter plasmids into CV1 cells, which lack significant levels of endogenous TR. Transfection of RSV-rTRa together with AMTVTREp-CAT, containing a palindromic response element, resulted in a hormone-dependent 20-fold stimulation of reporter gene expression, demonstrating that the c-ErbA/TR product is, as expected, a potent, hormone-dependent transcriptional activator (Fig. 5A).This functional assay enabled also a direct determination of the putative transcriptional activity of the v-erbA oncogene product. Since v-ErbA has lost its ability to bind thyroid hormone but retains an intact DNA-binding domain, it seems logical that it would function as a constitutively active TR. Unexpectedly, no constitutive, hormone-independent stimulation of transcription could be observed when AMTV-TREp-CAT was cotransfected with a v-erbA expression plasmid. However, when a different reporter construct, containing the same response element (TREp) but a different promoter (tk), was used in this cotransfection paradigm, novel regula-

A

C

B

1

100

SO.

T3

- +

- +

c-erbA

v-erbA

*' d8

&"

''

T3

c-erbA v-erbA

1

-

+ 1

-

+ 1

0.3

+

+

+

1 1

1

1 10

3

FIG. 5. (A) Trans-activation assay for ErbA proteins. CVl cells were transfected with the reporter construct AM-TREp-CAT and the respective expression plasmid. Thyroid hormone was added as indicated. Induction values are typically 20- to 30-fold using either c-ErbA/TRa or while no activity was observed with v-ErbA. (B) Repression of basal promoter level. CV 1 cells were transfected with the reporter construct tk-TREpP-CAT and the respective expression plasmid. All experiments were performed in the absence of thyroid hormone. c-ErbA and v-ErbA caused an -80% reduction of basal promoter activity. (C) Competition of T3 induction. The reporter gene AM-TREp-CAT was cotransfected into CVl cells with 1 pg c-ErbA expression vector and increasing quantities of the non-hormone-binding competitor v-ErbA. An increase in v-ErbA expression leads to a decrease in the hormone-dependent trans-activation by c-ErbA.

c-erbA:

PROTOONCOGENE OR GROWTH SUPPRESSOR GENE?

101

tory properties of both v-ErbA and c-ErbA were revealed (Fig. 5B). Transfection of tk-TREp-CAT alone resulted in a relative high basal level of CAT activity whereas cotransfection with the v-erbA expression vector resulted in an 80% decrease in basal CAT activity that could not be relieved by the addition of T,. This marked effect on tk-TREp-CAT expression demonstrated that v-ErbA provokes a ligand-independent inhibitory effect on transcription and can act as a constitutive repressor of TRE-containing genes. Surprisingly, the c-ErbA protein, in the absence of ligand, induced a similar repression of the reporter construct. In the case of c-ErbA, however, addition of thyroid hormone resulted in a 20-fold induction of transcription, suggesting that the primary effect of the hormone is to trigger a conformational change in the DNA-bound receptor that relieves the repression activity and reveals or induces the activation function. In this context, v-ErbA acts as a constitutive repressor of gene transcription because the point mutations in the ligand binding domain as well as the C-terminal deletion impaired hormone binding and transcriptional activation functions. D. COTRANSFECTION A N D COMPETITION Chicken erythroid progenitor cells were shown to contain c-ErbA products, suggesting that these proteins are involved in erythroid differentiation (Hentzen et al., 1987; Bigler and Eisenman, 1988). Therefore, in AEV-infected erythroid cells, v-erbA and c-erbA might be coexpressed. It seems likely that v-ErbA occupies c-ErbA binding sites on the DNA, thereby interfering with the function of the endogenous c-ErbA and acting as a constitutive repressor of gene transcription because it has lost the ability to bind and thus to be regulated by T,. This type of dominant negative oncogene function has been described on theoretical grounds by Herskowitz (1987) and was proposed by Bishop (1986) as a possible mechanism of v-ErbA function. To evaluate this property of v-ErbA, CV 1 cells were cotransfected with the AMTV-TREp-CAT reporter plash i d and a c-erbAlrTRol-expression construct (Damm et al., 1989). The levels of expression of v-ErbA"' competitor proteins were varied by titrating the amounts of the respective expression plasmids transfected into the recipient cells. These experiments revealed that v-ErbA serves as a negative regulator of thyroid hormone action since the transcriptional response to thyroid hormone is drastically reduced as the proportion of v-ErbA protein is increased (Fig. 5C). A 1: 1 ratio of c-ErbA and v-ErbAwtresulted in a 70% inhibition and a 3: 1 ratio completely blunted the hormonal response (Damm et al., 1989). Using a TRE identified in the Moloney leukemia virus LTR, Sap et al. (1989) also found a v-ErbA-

102

KLAUS DAMM

induced inhibition of the T, response. Because an intact DNA-binding domain is essential for v-ErbA function and biological activity (Damm et al., 1989; Privalsky et al., 1988), the competition or antagonism apparently occurs at the level of the response element by blocking the binding of c-ErbA to its template. T h e effectiveness of v-ErbA in blocking c-ErbA function may be a consequence of altered DNA-binding properties, e.g., a higher affinity for or a lower off rate from the DNA. On the other hand, c-ErbA is thought to bind DNA as a dimer and thus might form heterodimers with the v-ErbA product. If these c-Erb/v-ErbA heterodimers are inactive then, at a 1:l ratio, 75% of the homo- and heterodimers in a cell would be of the inhibitory type. Since the DNA-binding domain has been implicated in dimer formation of the receptor molecules it might well be a combination of the two mechanistic explanations that make v-ErbA function so effectively. These results place v-er6A as a dominant negative oncogene, dominant because its phenotype is manifested in the presence of the wild-type gene, and as it inactivates at least one wild-type gene function, it acts negatively. VII. Mutations Affecting the Biological Activity of v-erbA The functional properties of a biologically inactive v-erbA gene derived from the mutant AEVtd359virus (Damm et al., 1987) directly link the transforming potential of v-er6A to its ability to act as a negative regulator of transcription (Damm et al., 1992). First, when v-ErbAfdwas tested for its ability to act as a repressor of the basal promoter level, the mutant protein, in contrast to v-ErbAwt,failed to reduce promoter activity significantly. Second, v-ErbAtd also showed a reduced ability to act as a dominant negative inhibitor of the hormone-activated c-erbA. A 1: 1 plasmid ratio of c-erbA and the mutant v-erbAtd introduced into CV1 cells resulted in no significant reduction of the transcriptional response, whereas cotransfection of v-er6AWtunder identical conditions resulted in a 70% inhibition. Furthermore, v-ErbAr12, a natural revertant of v-ErbAtd which regained its biological activity (Damm et al., 1987), antagonizes the c-er6A activation as well as v-ErbAwfdoes. T h e defectiveness of the v-ErbAtd protein was not due to an impairment of its ability to interact with a thyroid hormone response element (TRE), since in an electrophoretic mobility shift assay retarded protein-DNA complexes were observed with v-ErbAfd-, v-ErbAwf-,and c-ErbA/TRa-containing extracts but not with untransfected control extract. Thus, the Pro+ Arg change that is responsible for the biological defectiveness of v-ErbAtd, and which is reverted in v-ErbAr12 (Damm et al., 1987), severely and

c-erbA:

103

PROTOONCOGENE OR GROWTH SUPPRESSOR GENE?

specifically affects the negative regulatory functions and the ability of v-ErbA to act as a dominant negative inhibitor of the thyroid hormone receptor. To identify the contribution of the different mutations in v-ErbAwtto the altered properties of this protein, to confirm the deleterious effect of the arginine mutation in v-ErbAtd, and to further dissect the processes of activation and repression, chimeric receptors of the v-erbA oncogene and the rat c-erbA/TRa were constructed and analyzed (Fig. 6; Damm et al., 1989). From amino acids 154 to 410, v-ErbAwfdiffers from rat c-ErbA/TRa in 26 amino acids and a deletion of 9 amino acids close to the carboxy terminus. Substitution of this region between rTRa and v-ErbA"' gives rise to the hybrid TR( 1 54)Awt,whose properties are virtually identical to the v-ErbAWthomolog. Thus, the mutations of the DNA-binding domain and flanking amino acids are not crucial to the v-ErbA phenotype. The mutations responsible for the functional conversion of c-ErbA into v-ErbA are localized to the ultimate C terminus, since replacement of the carboxy-terminal93 amino acids of rat c-ErbA/TRa with the corresponding sequence of v-ErbAwtyielded the hybrid protein TR(31 7)Awt,with suppressor properties identical to the viral oncogene product. These observations were supported in several subsequent transformation studies that demonstrated the importance of the v-ErbAwtC terminus for biological function and transcriptional repression in erythroid cells (Privalsky et al., 1988; Boucher and Privalsky,

T

R

C

c-erbA/TR

+

+

+

v-erbA

-

+

+

v-erbA-td

-

( 154/3 16)A

+

+

+

(1 54/31 6)A-td

+

-

-

-

FIG.6. Structure and activity of c/v-ErbA chimeric proteins. The origin of the different domains in the chimeras is indicated by different shading; R depicts the P-to-R mutation in v-ErbALd.Hormone-dependent trans-activation (T), repression of basal promoter activity (R), and competition of hormone induction (C) were performed as described in Fig. 5. Ts-binding derivatives were tested in the C-assay for competition against the retinoic acid receptor.

104

KLAUS DAMM

1990; Forrest et al., 1990b; Zenke et al., 1990). However, other regions in the ligand-binding domain have also been implicated in v-ErbA transcriptional and biological control functions. The hybrid TR( 154)Atd, containing the Pro + Arg mutation found in v-ErbAfd but otherwise structurally similar to TR( 154)Awt,exhibited properties that are in striking contrast to TR(154)Awt. Similar to v-ErbAtd, TR(154)Atd did not lower the basal transcription level of tk-TREp2-CAT and did not inhibit the hormone-dependent trans-activation by c-ErbA, demonstrating that the structural integrity of the region harboring the mutation, the socalled hinge region linking the DNA to the ligand-binding domain, seems to be critical for the correct function of the ErbA proteins. This is further supported by additional chimeric constructs in which a substitution of an internal region of the ligand-binding domain, amino acids 154-3 16, creates T,-responsive hybrids [TR(154/316)Awt and TR( 154/316)Atd]that exhibit activation properties similar to c-ErbA. However, in the absence of ligand TR( 154/316)Atd does not lower the basal promoter level. This is again in contrast to the equivalent TR( 154/316)Awtproduct and demonstrates clearly that the Pro + Arg change selectively abolishes the negative regulatory properties of the ErbA proteins (Damm et al., 1992). VIII. c-ErbA Regulation of Erythroid Differentiation and Gene Expression

T h e question of how v-ErbA, c-ErbA, and chimeras of both exert specific effects in erythroblasts was approached by using recombinant retroviruses to introduce the various ErbA derivatives into target cells (Zenke et al., 1988, 1990; Boucher and Privalsky, 1990; Privalsky et al., 1988; Forest et al., 1990a; Schroeder et al., 1990; Gandrillon et al., 1989; Pain et al., 1990). These studies revealed that v-ErbA by itself is sufficient to arrest differentiation (Gandrillon et al., 1989; Schroeder et al., 1990) and to selectively suppress the activity of three erythrocyte genes: the erythrocyte-specific carbonic anhydrase I1 (CAII) gene, the anion transporter gene (Band 3), and the erythrocyte version of the &aminolevulinic acid synthase (ALA-S) gene (Zenke et al., 1988; Schroeder et al., 1990). When the c-erbA gene was expressed as a Cag-c-ErbA fusion protein in erythroblasts, two effects were observed. In the absence of T,, Gag-c-ErbA arrested differentiation and reduced the levels of CAII, Band 3, and ALA-S mRNA. In the presence of T,, however, the erythroblasts expressed elevated levels of the three mRNAs and underwent abnormal differentiation (Zenke et al., 1990). Combination of the DNA-binding domain of v-ErbA with the hormone-binding domain of

c-erhA:

PROTOONCOGENE OR GROWTH SUPPRESSOR GENE?

105

c-ErbA resulted in a protein that bound T, with affinity similar to that of the Gag-c-ErbA protein and led to a similar T,-dependent regulation of the three erythrocyte-specific genes (Zenke et al., 1990). However, the transformation phenotypes induced were distinct since this chimeric protein caused only a partial inhibition of differentiation in the absence of hormone, while permitting normal or even accelerated differentiation into red cells in the presence of T,. This suggests that the amino acid substitutions in the v-ErbA DNA-binding region affect its activity in erythroid cells. A chimeric protein carrying the C-terminal deletion of v-ErbA in the context of the c-ErbA hormone-binding domain was unresponsive to T, but retained the ability to suppress differentiation and erythrocyte gene expression in a constitutive fashion, thus confirming the results of the hybrid chimeras studied in the transient transfection assays (Damn1 et al., 1989; Zenke et al., 1990). That the C-terminal domain is critical for v-ErbA biological and biochemical properties has also been demonstrated in erythroblasts and functional assays (Privalsky et al., 1988; Boucher and Privalsky, 1990; Forrest et al., 1990a). A specific role for the N terminus is suggested by the fact that c-ErbA and v-ErbA proteins might also be susceptible to posttranslational modifications such as phosphorylation by CAMP-dependent kinases or protein kinases C that modulate their function as transcriptional regulators (Goldberg et al., 1988; Glineur el al., 1989, 1990).

IX. c-erbA: Protooncogene or Growth Suppressor Gene? T h e results of the transient transfection experiments described above generated a new model of how the v-~rbAoncoprotein acts to block differentiation: it may actively repress the transcription of differentiation-specific genes as well as inhibit the function of its normal endogenous counterpart, resulting in a loss of hormone responsiveness and probably hormone-induced differentiation. Here we have a parallel to the recessive oncogene (or growth suppressor gene), where it is also a loss of function that induces the transformation process (for review, see Klein, 1987). Among the natural targets of c-erbA appear to be genes that play important parts during erythroid differentiation (Zenke et al., 1988, 1990). Thyroid hormones reveal or induce the activating function of c-erbA, causing the respective genes to be transcribed at maximum efficiency and allowing the differentiation to proceed. In this way, c-erbA acts as a growth suppressor since the resulting differentiated cells irreversibly lose proliferative potential. c-erbA might also fulfill another criterion usually implicated with growth suppressor genes: if both alleles of

106

KLAUS DAMM

the c-erbA gene are inactivated, the result would be a loss of hormone responsiveness and most likely a failure of the cells to differentiate. On the other hand, if one allele were converted by somatic mutation into a dominant negative version, differentiation would be blocked also and the cells would be free to proliferate. Thus, gain of function, such as the activation of c-erbA as a dominant negative similar to v-erbA, and the loss of function by inactivation of both alleles, might both be involved in neoplastic transformation. If that model is true, are there somatic mutations that inactivate the c-erbA gene, or convert one allele into a dominant negative derivative? A series of experiments, described in the next section, were performed to substantiate this idea (Damm et al., 1991).

X. Mutations Affecting c-erbA Function The constitutive negative phenotype of v-ErbA is due to mutations found in the ultimate C terminus, a region where v-ErbA and the thyroid hormone receptors exhibit a high degree of homology. The removal of as few as 14 amino acids from the 3’ end of the full-length c-ErbA/TRa and c-ErbA/TRP created proteins that exhibit the “v-ErbA phenotype” in cotransfection assays: they act as potent repressors of the basal transcription level, lose all hormone-dependent trans-activation properties, and act as highly effective dominant negative inhibitors of the c-ErbA-mediated hormone response (Fig. 7). C-terminal deletions extending up to about 30 amino acids into the ligand-binding domain do not change this phenotype whereas derivatives with more severe truncations are inactive. Thus, despite an intact DNA-binding domain and the ability to bind to TREs in vitro, these mutants are unable to repress basal promoter level or inhibit the trans-activation through c-ErbA. Interestingly, and in contrast to the glucocorticoid and other steroid receptors (Evans, 1988; Beato, 1989), none of these receptor derivatives showed hormone-independent, constitutive trans-activation functions. Characterization of the mutant derivatives revealed that amino acid sequences implicated in receptor dimerization (Glass et al., 1989, 1990)are absolutely required for the negative regulatory properties of c-ErbA. Therefore, the dominant negative phenotype may, in part, be mediated through the formation of inactive heterodimers between wild-type and mutant receptors. The C-terminal amino acids, which are deleted in v-ErbA and the dominant negative mutants, seem to be involved in hormone binding and in the hormone-dependent relief of the negative regulatory properties (Damm et al., 1989; MuAoz et al., 1988; Forrest et al., 1990b; Zenke et al., 1990). More importantly, this domain might be involved in hormone-dependent transcriptional regulatory functions

c-erbA:

PROTOONCOGENE OR GROWTH SUPPRESSOR GENE?

Z

R

C

TR-A1

-

+

+

TRA2

-

+

+

TR-A3

-

_

_

TR-mi

+

+

+

TR-rn2

-

+

+

TR-rn3

-

+

+

TR-rn4

+

+

*

TR-m5

- - *

107

FIG.7. Schematic structure and activity of c-ErbA/TR mutant proteins. In-frame linker insertions (arrows) in different parts of the ligand-binding domain were analyzed for functional properties as described in Fig. 5. Deletion of up to 30 amino acids (TR-Al) or more than 35 amino acids (TR-A2 and -A3) from the C terminus of c-ErbA results in mutant products with the indicated functional properties.

and thus in the interaction with other factors of the transcription machinery. In the natural splicing variant TRol2, which is identical to TRa in the first 370 amino acids, this domain is not present. Using the TREp as response element, TRa2 does not exhibit any repression ability and does not inhibit the hormone response (Damm et al., 1991), results that are not consistent with those reported by Koenig et al. (1989) and Lazar et al. (1989a). It seems likely that the variant C-terminal region may alter DNA binding by allosteric or other conformational effects and thus change DNA-binding specificity and affinity. In summary, these data demonstrate that most of the ligand-binding domain of c-ErbA is needed for the negative regulatory properties and that the ultimate C terminus is involved in hormone-mediated transcriptional activation processes. However, a single mutational event, e.g., the conversion of a coding nucleotide triplet near the C terminus into a termination codon, might lead to the creation of a mutant c-ErbA product with a dominant negative phenotype and potential oncogenic properties. What effects can we expect from possible point mutations in other regions of the c-erbA gene? The characterization of the biologically inactive revealed that a single amino acid change can have profound effects on transcriptional regulatory functions (Damm et d.,1987; Damm et al., 1992). By using the technique of oligonucleotide-linker mutagenesis a series of mutant c-ErbA proteins with small in-frame

108

KLAUS DAMM

insertions at different positions in the ligand-binding domain were created (Damm et al., 1992). In the cotransfection assays these mutant proteins exhibited a wide range of biological activity, from mutants indistinguishable from c-ErbA/TR to those completely defective in the functional assays (Fig. 7). An insertion in the N-terminal part of the ligand-binding domain, close to where the arginine mutation in v-ErbAtd is located, had no measurable effect on receptor function whereas a linker insertion closer to the C terminus completely inactivated the resulting protein. Insertions at two different positions within the ligand-binding domain induced the dominant negative phenotype exemplified by v-ErbA, that is, a loss of hormone-dependent transcriptional activation but the retention of the inhibitory potential. Another mutant showed the v-ErbAtd phenotype in that the protein retained hormone-binding and trans-activation capability but lost its negative regulatory properties. This mutational analysis supports the essential role of the C terminus in c-ErbA/TR activity and provides further evidence suggesting that this “ligand-binding domain” is a complex structure exhibiting hormone binding, dimerization, as well as positive and negative transcriptional regulatory functions. XI. Current Concepts and Open Questions In the last years a number of receptors for retinoid acid (RA, vitamin A) and 1,25-dihydroxycholecalciferol(vitamin D,) have been characterized (Giguere et al., 1987; Petkovich et al., 1987; Baker et al., 1988; Brand et al., 1988; Benbrook et al., 1988; Krust et al., 1989; Ishikawa et al., 1990; Mangelsdorf et al., 1990) and together with v-ErbA and c-ErbA/TR classified as a subgroup of the steroid hormone receptor superfamily. This classification is based on structural similarities between the receptor proteins and the finding that they display cross-recognition of their response elements. For example, the palindromic thyroid hormone response element (Glass et al., 1988)described above also serves as an efficient RA response element for the three isoforms of the retinoic acid receptors (RARs) and the retinoid X receptors (Umesono et al., 1988; Ishikawa et al., 1990; Mangelsdorf et al., 1990). The ability of v-ErbA to form nonproductive protein-DNA complexes with this response element suggests that v-ErbA might also interfere with RAR activity. Indeed, in transient transfection experiments v-ErbA acts as a potent inhibitor of the RAR-mediated retinoic acid response (Damm et al., 1992). Retinoic acid is known to be involved in hematopoiesis and thus it is tempting to speculate that v-ErbA might possibly interfere with RA-induced differentiation in vivo. Final proof, however, awaits the

c-erbA:

PROTOONCOGENE OR GROWTH SUPPRESSOR GENE?

109

identification of RA-induced hematopoietic genes and the demonstration of their regulation by v-ErbA. Alternatively, the RAR genes might turn out to be protooncogenes themselves. Rearrangements and altered expression patterns of RARa have been reported in several cases of acute promyelocytic leukemia (APL) (de The et al., 1990; Borrow et al., 1990). How these mutated receptors act in APL is unclear at present; however, in embryonal carcinoma cells and in vitro cotransfection assays it was recently demonstrated that mutant RAR proteins might act in a fashion similar to v-ErbA as dominant negative inhibitors of their normal counterparts (Pratt et al., 1990; Damm et al., 1992). v-ErbA and mutated RARs are not the only examples of dominant negative inhibition. A growing number of oncogenes turn out to encode mutated transcription factors that act through a dominant negative mechanism (Ballard et al., 1990; Inoue et d., 1991; Nakabeppu and Nathans, 1991); however, whether their normal counterparts may also act as growth suppressor genes remains to be demonstrated. ACKNOWLEDGMENTS I would like to thank the many colleagues who have provided stimulating discussions and comments while these data were generated, with special thanks to Drs. Hartmut Beug, Thomas Graf, and Bjorn Vennstrom, in whose laboratories at the European Molecular Biology Laboratory the AEV work originated, and Dr. Ron Evans at the Salk Institute for his generous support and many insightful discussions in the course of deciphering the molecular basis of v-erbA oncogenicity.

REFERENCES Baker, A. R.,McDonnell, D. P., Hughes, M., Crisp, T. M., Mangelsdorf, D. J., Haussler, M. R.,Pike, J. W., Shine, J., and O’Malley, B. (1988). Proc. Natl. Acad. Sci. U.S.A. 85, 3294-3298. Ballard, D. W., Walker, W. H., Doerre, S., Sista, P., Molitor,J.A., Dixon, E. P., Peffer, N. J., Hannink, M., and Greene, W. (1990). Cell (Cambridge, Mass.) 63, 803-814. Beato, M. (1989). Cell (Cambridge, Mass.) 56, 335-344. Benbrook, D., and Pfahl, M. (1987). Science 238, 788-791. Benhrook, D., Lernhardt, E., and Pfahl, M. (1988). Nature (London) 333, 669-672. Palmieri, S., Freudenstein. C., Zentgraf, H., and Graf, T. (1982). Cell (Cambridge, Beug, H., Mass.) 28, 907-919. Beug, H., Kahn, P., Doderlein, G., Hayman, M. J., and Graf, T. (1985). Haematol. Blood Transfw. 29,290-297. Bigler, J., and Eisenman, R. N. (1988). Mol. Cell. Biol. 8, 415.5-4161. Bishop, J. M. (1986). Nature (London) 321, 112-113. Bonde, B. G., and Privalsky, M. L. (199O).J. Virol. 64, 1314-1320. Borrow, J., Goddard, A. D., Sheer, D., and Solomon, E. (1990). Science 249, 1577-1580. Boucher, P., and Privalsky, M. L. (1990). Oncogene 5, 1303-131 1.

110

KLAUS DAMM

Brand, N. J., Petkovich, M., Krust, A., Chambon, P., de The, H., Marchio, A., Tiollais, P., and Dejean, A. (1988). Nature (London) 332, 850-853. Damm, K., Beug, H., Graf, T., and Vennstrom, B. (1987). E M B O J . 6, 375-382. Damm, K., Thompson, C. C., and Evans, R. M. (1989). Nature (London) 339, 593-597. Damm, K. et al. (1992). Submitted for publication. Danielsen, M., Hinck, L., and Ringold, G. M. (1989).Cell (Cambridge, Mass.) 57,1131-1 138. Dayton, A. I., Selden, J. R.,Laws, G., Dorney, D. J., Finan, J., Tripputi, P., Emanuel, B. S., Roveera, G., Nowell, P. C., and Croce, C. (1984).Proc. Natl. Acad. Sci. U.S.A. 81,44954499. Debuire, B., Henry, C., Benaissa, M., Biserte, G., Claverie, J. M., Saule, S., Martin, P., and Stehelin, D. (1984). Science 224, 1456-1459. de The, H., Chomienne, C., Lanotte, M., Degos, L., and Dejean, A. (1990).Nature (London) 347, 558-56 1. Disela, C . , Dodgson, J., Beug, H., and Zenke, M. (1991). Genes Dev. 5, 2033-2047. Downward, J., Yarden, Y., Mayes, E., Scrace, G., Totty, N., Stockwell, P., Ullrich, A,, Schlessinger, J., and Waterfield, M. D. (1984). Nature (London) 307, 521-527. Drabkin, H., Kao, F. T., Hartz, J., Hart, I., Gazdar, A., Weinberger, C., Evans, R. M., and Gerber, M. (1988). Proc. Natl. Acad. Sci. U.S.A. 85, 9258-9262. Evans, R. M. (1988). Science 240, 889-895. Forrest, D., Sjoberg, M., and Vennstrom, B. (1990a). E M B O J . 9, 1519-1528. Forrest, D., Mutioz, A., Raynoschek, C., Vennstrom, B., and Beug, H. (1990b).Omogene5, 309-316. Frykberg, L., Palmieri, S., Beug, H., Graf, T., Hayman, M. J., and Vennstrom, B. (1983). Cell (Cambridge, Mass.) 32, 227-238. Fung, Y. K. T., Lewis, W. G., Kung, H. J., and Crittenden, L. B. (1983). Cell (Cambridge, M ~ s . 33, ) 357-368. Gandrillon, O.,Jurdii., P., Benchaibi, M., Xiao, J. H., Ghysdael, J., and Samarut, J. (1987). Cell (Cambridge, Mass.) 49, 687-697. Gandrillon, O., Jurdii., P., Pain, B., Desbois, B., Madjar, J. J., Moscovoci, M. G., Moscovici, C., and Samarut, J. (1989). Cell (Cambridge, Mass.) 58, 115-121. Giguere, V., Ong, E. S., Segui, P., and Evans, R. M. (1987).Nature (London) 330,624-629. Glass, C. K., Franco, R., Weinberger, C., Albert, V. R., Evans, R. M., and Rosenfeld, M. G. (1987). Nature (London) 329, 738-74 1. Glass, C. K., Holloway,J. M., Devary, 0. V., and Rosenfeld, M. G. (1988). Cell (Cambridge, M a s . ) 54, 313-323. Glass, C . K., Lipkin, S. M., Devary, 0. V., and Rosenfeld, M. G. (1989). Cell (Cambridge, Mass.) 59, 697-708. Glass, C. K., Devary, 0. V., and Rosenfeld, M. G. (1990). Cell (Cambridge, Mass.) 63, 729738. Glineur, C., Bailly, M., and Ghysdael, J. (1989). Oncogene 4, 1247-1254. Glineur, C., Zenke, M., Beug, H., and Ghysdael, J. (1990). Genes Dev. 4, 1663-1676. Goldberg, Y., Glineur, C., Gesquire, J. C., Ricouart, A., Sap, J., Vennstrom, B., and Ghysdael, J. (1988). E M B O J . 7, 2425-2433. Graf, T., and Beug, H. (1978). Biochim. Bzophys. Acta 516, 269-299. Graf, T., and Beug, H. (1983). Cell (Cambridge, Mass.) 34, 7-9. Green, S., and Chambon, P. (1988). Trends Genet. 4, 309-314. Green, S., Walter, P., Kumar, V., Krust, A., Bornet, J. M., Argos, P., and Chambon, P. (1986). Nature (London) 320, 134-139. Hentzen, D., Renucci, A., le Guellec, D., Benchaibi, M., Jurdii., P., Gandrillon, O., and Samarut, J. (1987). Mol. Cell. Biol. 7, 2416-2424.

c-erbA:

PROTOONCOGENE OR GROWTH SUPPRESSOR GENE?

11 1

Herskowitz, I. (1987). Nature (London) 329, 219-222. Hodin, R. A., Lazar, M. A., Wintman, B. I., Darling, D. S., Koenig, R. J., Larsen, P. R., Moore, D. D., and Chin, W. W. (1989). Science 244, 76-79. Hodin, R. A,, Lazar, M. A,, and Chin, W. W. (199O).J. Clin. Invest. 85, 101-105. Hollenberg, S. M., Weinberger, C., Ong, E. S., Cerelli, G., Oro, A., Lebo, R., Thompson, E. B., Rosenfeld, M. G., and Evans, R. M. (1985). Nature (London) 318, 635-641. Inoue, J.-I., Kerr, L. D., Ransone, L. J., Bengal, E., Hunter, T., and Verma, I. M. (1991). Proc. Natl. Acad. Sci. U.S.A. 88, 3715-3719. Ishikawa, T., Umesono, K., Mangelsdorf, D. J., Aburatani, H., Stanger, B. Z., Shibasaki, Y., Imawari, M., Evans, R. M., and Takaku, F. (1990). Mol. Endocrinol. 4, 837-844. Izumo, S., and Mahdavi, V. (1988). Nature (London) 334, 539-542. Jansson, M., Philipson, L., and Vennstrom, B. (1983). E M B O J . 2, 561-565. Jansson, M., Beug, H., Gray, C., Graf, T., and Vennstrom, B. (1987). Oncogene 1, 167-173. Kahn, P., Frykberg, L., Brady, C., Stanley, I., Beug, H., Vennstrom, B., and Graf, T. (1986). Cell (Cambridge, Mass.) 45, 349-356. Khazaie, K., Dull, T. J., Graf, T., Schlessinger, J., Ullrich, A., Beug, H., and Vennstrom, B. (1988). EMBOJ. 7, 3061-3071. Klein, G. (1987). Science 238, 1539-1545. Koenig, R. J., Brent, G. A., Warne, R. L., Larsen, P. R., and Moore, D. D. (1987).Proc. Natl. Acad. Sci. U.S.A. 84, 5670-5674. Koenig, R. J., Warne, R. L., Brent, G. A., Harney, J. W., Larsen, P. R., and Moore, D. D. (1988). Proc. Null. Acad. Sci. U.S.A. 85, 5031-5035. Koenig, R. J., Lazar, M. A,, Hodin, R. A., Brent, G. A., Larsen, P. R.,Chin, W. W., and Moore, D. D. (1989). Nature (London) 337, 659-66 1 . Krust, A., Kastner, P., Petkovich, M., Zelent, A., and Chambon, P. (1989).Proc. Natl. Acad. Sci. U.S.A. 86, 5310-5314. Lazar, M. A., Hodin, R. A,, Darling, D. S., and Chin, W. W. (1988).Mol. Endocrznol. 2,893901. Lazar, M. A., Hodin, R. A,, and Chin, W. W. (1989a). Proc. Natl. Acad. Sci. U.S.A. 86, 77717774. Lazar, M. A,, Hodin, R. A., Darling, D. S., and Chin, W. W. (1989b).Mol. Cell. Biol. 9, 11281136. Lazar, M. A., Hodin, R. A., Cardona, G., and Chin, W. W. (1990). J. Biol. Chem. 265, 12859-12863. Mader, S., Kumar, V., d e Verneuil, H., and Chambon, P. (1989).Nature(London) 338,27 1-274. Mangelsdorf, D. J., Ong, E. S., Dyck, J. A., and Evans, R. M. (1990). Nature (London) 345, 224-229. Mitsuhashi, T., Tennyson, G. E., and Nikodem, V. M. (1988).Proc. Natl. Acud. Sci. U.S.A. 85,5804-5808. Miyajima, N., Horiiuchi, R., Shibuya, Y., Fukushige, S. I., Matsubara, K. I., Toyoshima, K., and Yamamoto, T. (1989). Cell (Cambridge, Mass.) 57, 31-39. Muiioz, A., Zenke, M., Gehring, U., Sap, J., Beug, H., and Vennstrom, B. (1988).EMBOJ. 7, 155-159. Murray, M. B., Zilz, N. D., McCreary, N. L., MacDonald, M. J ., and Towle, H . C. (1988).J. B i d . Chem. 263, 12770-12777. Nakabeppu, Y., and Nathans, D. (1991). Cell (Cambridge, Mass.) 64,751-759. Nakai, A., Sakurai, A., Bell, G. I., and DeGroot, L. J. (1988a). Mol. Endocrinol. 2, 10871092. Nakai, A , , Seino, S., Sakurai, A,, Szilak, I., Bell, G. I., and decroot, L. J. (1988b).Proc. Natl. Acad. Sci. U.S.A. 85, 2781-2785.

112

KLAUS DAMM

Pain, B., Melet, F., JurdiC, P., and Samarut, J. (1990).Nature (London) New B i d . 2,284-294. Pain, B., Woods, C. M., Saez, J., Flickinger, T., Raines, M., Peyrol, S., Moscovici, C., Moscovici, M. G., Kung, H. J., JurdiC, P., Lazarides, E., and Samarut, J. (1991). Cell (Cambridge, Mass.) 65, 37-46. Petkovich, M., Brand, N. J., Krust, A., and Chambon, P. (1987).Nature (London) 330,444450. Petty, K. J., Desvergen, B., Mitsuhashi, T., and Nikodem, V. M. (1990).J,Biol. Chem. 265, 7395-7400. Pratt, M. A. C., Kralova, J., and McBurney, M. W. (1990).Mol. Cell. Biol. 10, 6445-6453. Privalsky, M. L., Boucher, P., Koning, A., and Judelson, C. (1988). Mol. Cell. Biol. 8,451045 17. Privalsky, M. L., Sharif, M., and Yamamoto, K. R. (1990). Cell (Cambridge, Mass.) 63, 12771286. Prost, E., Koenig, R. J., Moore, D. D., Larsen, P. R., and Whalen, R. G. (1988).Nzicleic Acids Res. 16, 6248. Royer-Pokora, B., Beug, H., Claviez, M., Winkhardt, H. J., Friis, R. R., and Graf, T. (1978). Cell (Cambridge, Mass.) 13, 751-760. Royer-Pokora, B., Grieser, S., Beug, H., and Graf, T. (1979).Nature (London) 282,750-752. Sakurai, A., Takeda, K., Ain, K., Ceccarelli, P., Nakai, A,, Seino, S., Bell, G. I., Refetoff, S., and DeGroot, L. J. (1989). Proc. Natl. Acad. Sci. U.S.A. 86, 8977-8981. Sap, J., Munoz, A., Damm, K., Goldberg, Y., Ghysdael, J., Leutz, A,, Beug, H., and Vennstrom, B. (1986). Nature (London) 324, 635-640. Sap, J., MuAoz, A., Schmitt, J., Stunnenberg, H., and Vennstrom, B. (1989). Nature (London) 340, 242-244. Sap, J., d e Magistris, Stunnenberg, H., and Vennstrom, B. (1990). E M B O J . 9, 887-896. Schroeder, C., Raynoschek, C., Fuhrmann, U., Damm, K., Vennstrom, B., and Beug, H. (1990). Oncogene 5, 1445-1453. Schwartz, D. E., Tizard, R., and Gilbert, W. (1983). Cell (Cambridge, Mass.) 32, 853-869. Sealy, L., Privalsky, M. L., Moscovici, G., Moscovici, C., and Bishop, J. M. (1983). Virology 130, 155-178. Spurr, N. K., Solomon, E., Jansson, M., Sheer, D., Goodfellow, P. N., Bodmer, W. F., and Vennstrom, B. (1984). E M B O J . 3, 159-163. Thompson, C. C., and Evans, R. M. (1989). Proc. Natl. A c d . Sci. U.S.A. 86, 3494-3498. Thompson, C. C., Weinberger, C., Lebo, R., and Evans, R. M. (1987). Science 237, 16101614. Ullrich, A., Coussens, L., Hayflick, J. S., Dull, T. J., Gray, A,, Tam, A. W., Lee, J., Yarden, Y., Liberman, T. A,, Schlessinger, J., Downward, J., Mayes, E. L. V., Whittle, N., Waterfield, M. D., and Seeburg, P. H. (1984). Nature (London) 309,418-425. Umesono, K., and Evans, R. M. (1989). Cell (Cambridge, Mass.) 57, 1139-1 146. Umesono, K., Giguere, V., Glass, C. K., Rosenfeld, M. G., and Evans, R. M. (1988). Nature (London) 336, 262-265. Usala, S. J., Tennyson, G. E., Bale, A. E., Lash, R. W., Gesundheit, N., Wondisford, F. E., Accili, D., Hauser, P., and Weintraub, B. D. (1990).J. Clzn. Invest. 85, 93-100. Usala, S. J., Menke, J. B., Watson, T. L., Berard, J., Bradley, W. E. C., Bale, A. E., Lash, R. W., and Weintraub, B. (1991a).J. Cell. Endocrinol. Metab. 72, 32-38. Usala, S. J., Menke, J. B., Watson, T. L., Wondisford, F. E., Weintraub, B. D., Berard, J., Bradley, W. E. C., Ono, S., Mueller, 0. T., and Bercu, B. B. (I99lb).Mol. Er2docri?zol. 5, 327-335. Vennstrom, B., and Bishop, J. M. (1982). Cell (Cambridge, Mass.)28, 135-143. Vennstrom, B., Fanshier, L., Moscovici, C., and Bidhop, J. M. (198O).j.Virol. 36,575-585.

c-erbA:

PROTOONCOGENE OR GROWTH SUPPRESSOR GENE?

1 13

Weinberger. C., Hollenberg, S. M., Rosenfield, M. G . , and Evans, R. M. (1985). Nature (London) 318, 670-672. Weinberger, C., Thompson, C. C., Ong, E. S., Lebo, R., Gruol, D., and Evans, R. M. (1986). Nature (London) 324, 641-646. Yamamoto, T., Hihara, H., Nishida, T., Kawai, S.,and Toyoshima, K. (1983). Cell (Cambridge, Mass.) 34, 22.5-232. Yaoita, Y., Shi, Y. B., and Brown, D. D. (1990). Proc. Natl. A d . Scz. U.S.A. 87, 7090-7094. Zenke, M., Kahn, P., Disela, C., Vennstrom, B., Leutz, A., Keegan, K., Hayman, M. J., Choi, H . R., Yew, N., Engel, J. D., and Beug, H. (1988). Cell ( C a d r i d g e , Mass.) 52, 107119. Zenke, M., MuAoz, A,, Sap, J., Vennstriim. B., and Beug, H. (1990). Crll (Crtmbrzdp, Mass.) 61, 103.5-1049.

This Page Intentionally Left Blank

THE FGF FAMILY OF GROWTH FACTORS AND ONCOGENES Claudio Basilica* and David Moscatellit 'Department of Microbiology and +Department of Cell Biology, New York University School of Medicine, New York, New York 10016

I. Introduction 11. Protein Structure A. Basic FGF (FGF-2) B. Acidic FGF (FGF-I) C. INT-2 (FGF-3) D. K-FGF/HST (FGF-4) and FGF-6 E. FGF-5 F. Keratinocyte Growth Factor (FGF-7) 111. T h e FGF Genes and Their Expression Molecular Regulation of FGF Expression IV. FGF Receptors V. Interaction with Extracellular Matrix VI. Biological Function VII. Oncogenic Potential A. K-FGF/HST B. INT-2 C. FGF-5 D. bFGF and aFGF E. FGF-6 and KGF VIII. Involvement of FGFs in Tumors A. INT-2 and K-FGF B. bFGF and aFGF C. FGFs and Tumor Angiogenesis IX. Concluding Remarks References

I. Introduction

During the past few years the family of fibroblast growth factors (FGF) has emerged as perhaps the largest family of growth factors involved in soft-tissue growth and regeneration. It presently includes seven members that share a varying degree of homology at the protein level, and that, with one exception, appear to have a similar broad mitogenic spectrum, i.e., they promote the proliferation of a variety of cells of mesodermal and neuroectodermal origin and are angiogenic. While their genes have similar organization and in some cases map to contiguous regions on human and mouse chromosomes, their pattern of 115 ADVANCES IN CANCER RESEARCH, VOL. 59

Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.

116

CLAUD10 BASILICO A N D DAVID MOSCATELLI

expression is very different, ranging from extremely restricted expression in some stages of development, to rather ubiquitous expression in a variety of tissues and organs. In addition, some of these factors are efficiently secreted by producer cells, while others are not. They all appear to bind heparin and heparan sulfate proteoglycans and glycosaminoglycans and strongly concentrate in the extracellular matrix. Thus they have also been called heparin-binding growth factors (HBGF) (reviewed in Baird and Bohlen, 1990; Burgess and Maciag, 1989; Rifkin and Moscatelli, 1989). “Fibroblast growth factor” was originally identified as an activity in pituitary extracts that stimulated the proliferation of mouse 3T3 cells (Armelin, 1973; Gospodarowicz, 1974). The mitogenic activity was partially purified from bovine pituitary and brain and was found to be due to a molecule with a molecular weight of 14K-16K with a basic isoelectric point (Gospodarowicz, 1975; Gospodarowicz et al., 1978a). Shortly afterward evidence accumulated that a second 3T3 cell mitogen with an acidic isoelectric point occurred in brain (Maciag et al., 1979; Thomas et al., 1980). The mitogenic activity of these FGFs was not restricted to fibroblasts, and stimulated many cell types, including endothelial cells and chondrocytes (Gospodarowicz et al., 197813). Although small amounts of both the basic and the acidic mitogen were purified by conventional chromatography methods (Bohlen et al., 1984; Thomas et al., 1984), further characterization was aided by observations made by investigators purifying angiogenesis factors, agents that stimulate new blood vessel growth. Since both angiogenic factors and the FGFs were mitogenic for endothelial cells, they were suspected to be closely related. The discovery that an endothelial cell mitogen from a tumor extract bound strongly to heparin (Shing et al., 1984) suggested that heparin affinity columns might be used for the purification of the FGFs. Indeed, both basic and acidic FGFs bound strongly to heparin affinity columns, with acidic FGF eluting at 1 M NaCl and basic FGF eluting at 1.5 M NaCl (Gospodarowicz et al., 1984; Klagsbrun and Shing, 1985; Maciag et al., 1984). Complete characterization of the FGFs was greatly facilitated by the use of heparin affinity columns, which, by permitting a simple and rapid purification of FGFs, also helped demonstrate that the mitogenic and angiogenic activities identified in a variety of tissues were mostly due to either basic or acidic FGF. This reduced a large list of poorly characterized mitogens to synonyms for basic and acidic FGF. However, the apparent simplicity of the FGF family did not last, and the family has grown to seven members. Three other members of the family (K-FGF/HST, FGF-5, INT-2) were identified originally as oncogenes, while the two latest additions, FGF-6 and keratinocyte growth

FGF FAMILY OF GROWTH FACTORS A N D ONCOGENES

117

factor (KGF), were isolated by sequence homology or factor purification and cloning. The related molecules that have been characterized have mitogenic activity and also bind tightly to heparin affinity columns. Further complexity arises from the fact that some members of the family also exist in multiple forms arising from initiation of translation at alternative codons (Table I). Genes or proteins corresponding to the same FGF often appear in the literature under different names, reflecting their origin, isolation, etc. To alleviate this cause of confusion, a new nomenclature has been proposed (Aaronson et d,1991). In this proposal FGFs are numbered consecutively (i.e., acidic FGF = FGF-1, basic FGF = FGF-2, etc.). Also, this proposal recommends that the name fibroblast growth factor, which implies a spectrum of action specific for fibroblasts, be discontinued and replaced simply by FGF. We have continued in this article to use the old nomenclature, but we have indicated also the proposed new names. The proposal for a new nomenclature also suggests a numbering system for the amino acids of the FGFs. This numbering system designates the amino acid immediately following the initiation methionine as amino acid 1. We have adopted the new numbering system in the discussion of FGF structure. The genes and cDNAs for most FGFs have been cloned, often in difference species, and some knowledge about their receptors is starting to emerge. In spite of the large body of information that is accumulating about these growth factors, little is known about their physiological and pathological role. While in tissue culture they stimulate the growth and proliferation of a wide variety of cell types and often seem to exhibit an identical spectrum of action, it is not clear whether this apparent lack of specificity is also reflected in vivo. In a similar vein, although it is clear that some of the FGFs can be potently oncogenic in model systems, their involvement in human tumors remains to be clarified. We will review in this article the protein structure, localization, gene organization, and regulation of expression of these growth factors in an attempt to understand their mechanism of action, interaction with their receptors, and ultimately their function and role in physiology and development, as well as in pathological conditions. 11. Protein Structure

A. BASICFGF (FGF-2)

Basic FGF (bFGF) was originally purified from bovine pituitary as a 146-amino acid protein with a molecular weight of 16.5K and an iso-

TABLE I PROPERTIES OF HUMAN FGFsa ~~

Growth factor aFGF (FGF-1) bFGF (FGF-2)

INT-2 (FGF-3) K-FGF/HST (FGF-4) FGF-5 FGF-6 KGF (FGF-7)

Primary translation productb

Size of secreted form*

155 155 196d 20 Id 210d 239 271d 206 267 198 (murine) 194

Heparin affinityc

Known high-affinity receptors

Gene mapping (human)

Not secreted Not secreted

1.0 M 1.6 M

FGFR-1 R-2, R-3, R-4 FGFR- 1, FGFR-2

5q31-33 4q25

NDe

NDe

176 ND' NDe -163

1.2-1.3 M 1.0-1.5 M NDe 0.6 M

Unless otherwise noted, data are for human FGFs. Pertinent references are in the text. Number of amino acids. c NaCl concentration required for elution from heparin affinity columns. d Products of translation initiation at upstream CUG. e ND, Not determined. a

b

1lq13 FGFR-1, FGFR-2

FGFR-2 (variant)

llq13 4q2 1 12~13

FGF FAMILY OF GROWTH FACTORS AND ONCOGENES

119

electric point of 9.6 (Esch et al., 1985a). Molecules with identical properties were purified from bovine brain, retina, and adrenal, and from human brain (Baird et al., 1985; Bohlen et al., 1985; Gimenez-Gallego et al., 1986; Gospodarowicz et al., 1984, 1986). Smaller forms containing truncations at the amino-terminal end, but which still retained biological activity, were isolated from corpus luteum, adrenal, and testes (Gospodarowicz et al., 1985; Ueno et al., 1987). It is not clear whether all of these forms occur in i~ivo.At least some of the truncated forms seem to arise from proteolysis that occurs during the extraction procedure (Klagsbrun et al., 1987), and, with careful isolation in the presence of protease inhibitors, bFGF molecules even larger than 146 amino acids have been obtained (Story et al., 1987; Ueno et al., 1986). From these results, it is unclear how much proteolytic processing of bFGF occurs in viva When both bovine and human bFGF cDNAs were cloned, an AUG codon was found in the proper context to initiate translation of a protein of 155 amino acids, and no in-frame AUG codons were found upstream (Abraham et al., 1986a,b). Therefore, translation was predicted to initiate at this AUG codon and to result in an 18-kDa protein. However, larger forms of bFGF were purified from human placenta and from guinea pig brain (Moscatelli et al.,1987; Sommer et al., 1987). These higher molecular weight forms of bFGF have been shown to arise from the use of three upstream CUG codons as alternate initiation codons for translation. Thus, human bFGF is expressed in four forms, an 18-kDa form (155 amino acids) generated by initiation at the AUG codon and 22-, 22.5-, and 24-kDa forms (196, 201, and 210 amino acids) arising from the CUG codons (Florkiewicz and Sommer, 1989; Prats et al., 1989). The high molecular weight forms of bFGF contain the same amino acid sequence as the 18K form but have additional N-terminal extensions of varying lengths. Both 18K and higher molecular weight forms are expressed in brain (Moscatelli et al., 1987; Presta et al., 1988) and a number of different cell lines (Ensoli et al., 1989a; Iberg et al., 1989; Renko et al., 1990; Tsuboi et al., 1990). The different forms of bFGF seem to be correlated with differences in subcellular distribution. Both the 155-amino acid bFGF and the higher molecular weight forms lack a typical signal sequence for secretion and seem to be retained within the cell. The 155-amino acid form is primarily located in the cytosol while the higher molecular weight forms are present in the nuclear and ribosomal fractions (Renko et al., 1990; Florkiewicz et al., 1991). These results suggest that the higher molecular weight forms of bFGF contain a nuclear translocation sequence. Indeed, when the N-terminal extension of the higher molecular weight forms is

120

CLAUD10 BASILICO AND DAVID MOSCATELLI

grafted to normally cytoplasmic proteins, it is able to drive these recombinant proteins to the nucleus (Bugler et al., 1991; Quarto et al., I991a). T h e N-terminal extensions in the higher molecular weight forms of bFGF contain several stretches of alternating glycine and arginine residues. Some of the arginine residues in these sequences are methylated (Burgess et al., 1991; Sommer et al., 1989), as has been described for other nuclear proteins. The significance of nuclear forms of bFGF is presently not clear. In addition to endogenous nuclear forms of bFGF, translocation of exogenous 18-kDa bFGF to the nucleus has been described (Bouche et al., 1987). This translocated bFGF associates primarily with the nucleolus. A putative nuclear translocation sequence has also been identified in amino acids 26 to 3 1. Mutant 18-kDa bFGF lacking this sequence retains full mitogenic activity but has a reduced ability to induce plasminogen activator in endothelial cells (Isacchi et al., 1991). While all of these findings are very provocative, a conclusive interpretation of these data will require additional work. T h e amino acid sequence of 18-kDa bFGF is highly conserved among species with 89-95% identity among human, bovine, ovine, and rat bFGFs (Abraham et al., 1986a,b; Kurokawa et al., 1988; Shimasaki et al., 1988; Simpson et al., 1987). X e n o w bFGF is more divergent, but still shares -80% homology with human bFGF (Kimelman et al., 1988). This low level of divergence suggests that there may be functional importance for all regions of bFGF. There seems to be more variation in the Nterminal extensions of the high molecular weight forms of bFGF (Brigstock et al., 1990; Sommer et al., 1989), suggesting that these regions are less functionally restricted. Additional information about biologically active sequences in bFGF can be obtained by comparing it to other members of the FGF family. However, the homology to other members,of the family extends over almost the entire sequence of bFGF, from amino acids 27 to 149. bFGF contains four cysteine residues, and two of these are conserved among all members of the FGF family, suggesting that they may have important functions in the biology of the FGFs. However, none of the cysteines are strictly necessary for biological activity, since in vitro mutagenesis of the cysteine residues to serine does not alter the mitogenic activity of bFGF (Fox et al., 1988; Seno et al., 1989; Arakawa et al., 1989). T h e heparin-binding and receptor-binding regions of bFGF have been mapped based on the ability of synthetic peptides representing different amino acid sequences in bFGF to bind radiolabeled heparin, to block binding of radiolabeled bFGF to its receptor, and to act as agonists or antagonists of bFGF biological activity (Baird et al., 1987, 1988). The

FGF FAMILY OF GROWTH FACTORS AND ONCOGENES

121

receptor-binding activity was found in two regions, amino acids 32-76 and 114-123. The inclusion of C-terminal sequences in the peptide containing amino acids 114-123 increased its potency. In these studies, heparin binding was strongest in these same regions, but lower heparinbinding capacity was found in other sequences, suggesting that heparinbinding sites are distributed throughout the molecule. Other experiments suggest that the receptor-binding and heparin-binding regions of bFGF are distinct. Antibodies that bind to bFGF and inhibit its interaction with its receptor have no affect on the heparin affinity of the molecule (Kurokawa et al., 1989). Furthermore, Sen0 et al. (1990) found that deleting 42 amino acids from the C terminus of bFGF abolished the heparin affinity of bFGF, but the molecule still retained some biological activity, although it was about lo4 times less potent than the intact molecule. These studies also showed that mutant forms of bFGF lacking amino acids 1-48 had normal affinity for heparin, while forms lacking more than 6 amino acids from the C terminus had reduced affinity for heparin, suggesting that the carboxy-terminal structure was important for heparin binding (Seno et al., 1990). X-Ray crystalographic studies of the three-dimensional structure of bFGF show a cluster of basic residues in this region. In addition, the crystals contain ordered sulfate ions forming ionic contacts with Lys- 127, Lys- 137, Lys- 133, and Arg- 128, which may mimic the contacts made by sulfate moieties in heparin (Eriksson et al., 1991; Zhang et al., 1991). The major structural characteristics of bFGF are shown in Fig. 1 . conserved cvstelnes

~

arginine heparin binding

receptor bindlng

FIG. 1 . Schematic representation of the bFGF molecule. The schematic diagram represents the amino acid sequence of both 18K and higher molecular weight bFGFs. The scale on the bottom marks the location of the amino acid residues with the methionine that initiates 18K bFGF designated amino acid - 1 . The relative locations of structural features described in the text are indicated.

122

CLAUD10 BASILICO AND DAVID MOSCATELLI

bFGF has also been reported to be a substrate for phosphorylation by protein kinase C and protein kinase A (Feige and Baird, 1989). Phosphorylation by protein kinase C occurs on Ser-72 while phosphorylation by protein kinase A occurs at Thr-120. Phosphorylation by protein kinase A occurs in one of the putative receptor-binding regions and seems to slightly increase the affinity of bFGF for its receptor, while phosphorylation by protein kinase C has no effect on receptor affinity. Although the biological significance of phosphorylation of bFGF is not clear, phosphorylated bFGF has been identified in extracts of cultured endothelial cells, and, thus, appears to occur naturally in these cells. T h e cellular source of bFGF is uncertain. It has been found in all organs and tissues examined except serum (Baird et al., 1986). It is synthesized by cultured fibroblasts, endothelial cells, glial cells, and smooth muscle cells (Connolly et al., 1987; Gospodarowicz et al., 1988; Hatten et al., 1988; Moscatelli et al., 1986a; Schweigerer et al., 1987a; Vlodavsky et al., 1987; Weich et al., 1990), and, since these cell types are ubiquitous, they may be the source of bFGF in the organs. B. ACIDICFGF (FGF-1) Acidic FGF (aFGF) was isolated originally as a 154-amino acid protein in addition to truncated forms of 140 and 134 amino acids (Burgess et al., 1986; Esch et al., 1985b; Gimenez-Gallego et al., 1985; Harper et al., 1986). T h e N terminus of the 154-amino acid form is acetylated (Crabb et al., 1986), but acetylation seems to have no effect on the biological activity of the molecule. Several groups have shown that recombinant nonacetylated aFGF made in bacteria, recombinant acetylated aFGF made in yeast, and natural aFGF all have equivalent mitogenic and angiogenic potencies (Barr et al., 1988; Jaye edal., 1987; Linemeyer et al., 1987). T h e primary translation product for aFGF is, however, a 155-amino acid protein (Jaye et al., 1986). There appear to be no N terminalextended forms of aFGF, since a termination codon is found at position - 1 to the AUG initiation codon. However, the existence of alternate 5'untranslated exons in aFGF messengers has been described (Chiu et al., 1990; Crumley et aZ., 1990). T h e role of these untranslated sequences is unknown but might be involved in the differential regulation of translation of the molecule. T h e aFGF protein, like bFGF, has no signal sequence for secretion and is inefficiently released by cells that product it. aFGF has 55% amino acid sequence identity to 18-kDa bFGF. Homology extends over the entire sequence of the molecule except for the 18 N-terminal amino acids and a 2-amino acid insert at positions 116 and

FGF FAMILY OF GROWTH FACTORS AND ONCOGENES

123

117. aFGF contains three cysteine residues that, like the cysteine residues in bFGF, do not seem to be necessary for biological activity. Crabb et al. (1986) reported that aFGF in which these residues have been reduced and carboxymethylated still retains biological activity. In contrast, Jaye et al. (1987) have reported that quantitative alkylation of cysteine residues abolishes the receptor-binding activity. Finally, the studies of Linemeyer et al. ( 1 990) have shown that disulfide bond formation between the cysteines drastically reduces the biological activity of aFGF. These studies showed that cysteine residues in brain-derived and recombinant aFGF are normally reduced, that formation of intramolecular disulfide bonds results in an inactive molecule, and that reduction of the disulfide bonds restores activity. In addition, site-directed mutation of any of the cysteine residues to serine results in an aFGF with high biological activity (Linemeyer et al., 1990). Indeed, substitution of any two of the three cysteines with serine residues produces aFGF that is less heparin dependent and more stable in the absence of heparin (Ortega et al., 1991). Together these investigations suggest that the cysteine residues are not necessary for biological activity, but modification of the cysteines may hinder the formation of biologically active conformations of aFGF. Immunolocalization studies have also identified nuclear forms of aFGF (Sano et al., 1990; Speir et al., 1991).A putative nuclear localization sequence has been identified in residues 2 1-27 of the protein. Deletion of this sequence results in a molecule of reduced potency, while substitution of the nuclear translocation sequence from yeast histone 2B restores biological activity (Imamura et al., 1990). Although this basic sequence seems to be necessary for full biological activity of aFGF, its role as a nuclear targeting sequence has not been directly demonstrated. Thus, it is not clear whether the biological effects of deletion of this sequence are mediated through effects on the ability of aFGF to be translocated to the nucleus. Attempts to map binding regions of aFGF have identified two sites. A synthetic peptide corresponding to residues 49-72 of the primary translation product competed with aFGF for heparin binding (Mehlman and Burgess, 1990). This region is homologous to one of the regions identified for heparin binding in bFGF. A second binding region was indicated by chemical modification experiments. Methylation of lysine residues in aFGF caused a reduction in its affinity for heparin, receptor affinity, and biological potency (Harper and Lobb, 1988). The alteration in activity was correlated with the modification of Lys- 132. Site-directed mutation of this lysine to glutamic acid resulted in an aFGF with a lower affinity for heparin and reduced mitogenicity but with no alteration in its receptor affinity or ability to stimulate early responses in cells (Burgess et al.,

124

CLAUD10 BASILICO A N D DAVID MOSCATELLI

1990a). Thus, this residue seems to be important in heparin interactions but not in receptor binding.

C. INT-2 (FGF-3) The INT-2 gene was originally identified as a site of frequent insertion of the mouse mammary tumor virus (Dickson et al., 1984). Viral insertion activated the transcription of the cellular gene, leading to tumor formation. The predicted product of this gene, the INT-2 protein, is 245 amino acids in the mouse or 239 amino acids in humans (Brookes et al., 1989a). T h e core of the protein has 44% amino acid sequence homology to bFGF, while the N-terminal and C-terminal sequences are unique (Dickson and Peters, 1987). The protein contains a short stretch of hydrophobic amino acids at the N terminus that may serve as an atypical secretory signal sequence. Indeed, INT-2 protein has been detected in the endoplasmic reticulum and Golgi compartments (Acland et al., 1990), but release into the medium seems to be inefficient. The primary translation product has a molecular weight of 28.5K3,and posttranslational modifications, including glycosylation and a presumed cleavage of the signal peptide, give rise to 30.5K and 3 1.5K forms (Dixon et al., 1989). INT-2 protein can also initiate from a CUG codon upstream from the normal AUG codon, giving rise to a 27 l-amino acid human protein or a 274-amino acid (31.5 kDa) mouse protein (Acland et al., 1990). T h e N-terminal extended form of INT-2, like the N-terminal extended forms of bFGF, is localized in the nucleus. INT-2 seems to be expressed primarily during development, and has not been detected in any normal adult tissue. D. K-FGF/HST (FGF-4) AND FGF-6 T h e HST/K-FGF gene was discovered by screening for genes present in human stomach tumors or Kaposi’s sarcoma that are able to transform NIH 3T3 cells (Delli Bovi and Basilico, 1987; Sakamoto et al., 1986). The human HST/K-FGF gene encodes a 206-amino acid protein (K-FGF) with a classical signal sequence for secretion (Delli Bovi et al., 1987; Taka et al., 1987). T h e murine K-FGF protein is only 202 amino acids long, but otherwise has about 85% identity to the human protein (Brooks et al., 1989b). While the N-terminal 80 amino acids of the protein are unique, the remaining 126 amino acids of the human protein have about 40% homology to human bFGF. Thirty o r 3 1 N-terminal amino acids containing the signal sequence are cleaved during posttranslational processing, giving rise to a final product of 175 or 176 amino acids (Delli Bovi et al.,

FGF FAMILY OF GROWTH FACTORS AND ONCOGENES

125

1988).The protein contains one site for N-linked glycosylation,and glycosylation increases themolecularweighttoabout 23K(Delli Bovietal., 1988). The protein is efficiently secreted by cells that express it. K-FGF contains the two conserved cysteines of all FGFs and also in this case they do not appear essential for function (D. Talarico and C. Basilico, unpublished). K-FGF seems to be expressed only at limited times during development. In animal models, inappropriate expression of the growth factor in adult tissues leads to tumor formation through autocrine activation of FGF receptors on the cell surface (Talarico et al., 1990; Talarico and Basilico, 1991). The FGF-6 gene was discovered after screening a library with probes to the HST/K-FGF gene (Marks et al., 1989). The FGF-6 protein is closely related to K-FGF and seems to be structurally very similar. Little is known about the biology of FGF-6, but it has also been shown to be a secreted protein (deLapeyriere et al., 1990).

E. FGF-5 The FGF-5 gene also was identified by screening for genes present in tumors that are able to transform NIH 3T3 cells (Zhan et al., 1988). The gene encodes a 267-amino acid protein. Like INT-2, the central core of the protein has about 50% homology to bFGF, but the N-terminal and Cterminal sequences are unique. The molecule contains a classical signal sequence and is efficiently secreted. T h e molecular weight of the primary translation product is 29.5K. Posttranslational processing, including cleavage of the signal sequence, N-linked glycosylation at one site, and possible O-linked glycosylation, yields molecules of 32.5-38.5 kDa (Bates et al., 1991). FGF-5 is expressed in a site-specific manner at limited times during development. No FGF-5 mRNA could be detected in extracts of a number of adult tissues, but low levels of FGF-5 mRNA have been detected in localized areas of adult brain by in situ hybridization (Haub et al., 1990). F. KERATINOCYTE GROWTH FACTOR (FGF-7) Keratinocyte growth factor (KGF) was isolated as a mitogen for a cultured murine keratinocyte line (Rubin et al., 1989). Unlike the other members of the FGF family, it has little activity on mesenchyme-derived cells but stimulates the growth of epithelial cells. The keratinocyte growth factor gene encodes a 194-amino acid polypeptide (Finch et al., 1989). T h e N-terminal64 amino acids are unique, but the remainder of

126

CLAUD10 BASILICO AND DAVID MOSCATELLI

the protein has about 30% homology to bFGF. All in all, KGF is the most divergent member of the FGF family. T h e molecule has a hydrophobic signal sequence and is efficiently secreted. Posttranslational modifications include cleavage of the signal sequence and N-linked glycosylation at one site, resulting in a protein of 28 kDa. Keratinocyte growth factor is produced by fibroblasts derived from skin and fetal lung (Rubin et al., 1989). The keratinocyte growth factor mRNA was found to be expressed in adult kidney, colon, and ilium, but not in brain or lung (Finch et al., 1989). The conserved regions within the FGF protein family are shown in Fig. 2.

111. The FGF Genes and Their Expression Consistent with their origin from a common ancestral gene, the FGF genes studied so far appear to have a similar organization. They consist of three exons, separated by two introns of variable length. Typically the second exon is very short and in many cases the third exon includes a very long (2-3 kb) 3’-untranslated sequence. Although we did not conduct a detailed analysis, there seems to be very little homology at the DNA level among FGF genes outside of the coding regions. T h e FGF genes map on several chromosomes. The two prototypes of the family, basic and acidic FGF, map on human chromosomes 4q25 and 5q31-33, respectively (Fukushima et al., 1990; Jaye et al., 1986). INT-2 and K-FGF are very closely linked on llq13 (Huebner et ad., 1988; Nguyen et al., 1988; M. C. Yoshida et al., 1988). In the mouse, these genes are arranged head to tail and separated only by about 20 kb of DNA (Brookes et al., 1989b). FGF-5 maps on human chromosome 4q21 (Nguyen et al., 1988) and FGF-6 on 1 2 ~ 1 3(Marks et al., 1989). FGF genes have been found in all mammals, birds, and amphibians. We are not aware of any report of FGF-like genes in Drosophila or yeast. It is likely that FGF-like molecules exist in Drosophih and may have a role in development, because a putative Drosphila FGF receptor has just been reported (Glazer and Shilo, 1991). T h e pattern of expression of each FGF family member is quite distinct. Basic and acidic FGF are quite ubiquitous and are present in most tissues at relatively high concentration, generally bound to the extracellular matrix (ECM). At the RNA level, their expression is particularly high in the brain. Presumably these factors are produced at low levels by a variety of tissues and cells, and accumulate in the ECM. As already mentioned, the situation is further complicated by the fact that bFGF and aFGF do not possess a signal peptide and are very inefficiently secreted by producer cells.

127

FGF FAMILY OF GROWTH FACTORS AND ONCOGENES

........

K-FGF FGF-6 FGF-5 aFGF bFGF KGF INT-2

MS.GPGTAAV ALLPAVLLAL LA .PWAGRGGAA APTAPNGTLE MSRGAGRVQG TLQALVF’LGV LV........ .GMVVPSPAG AR..ANGTLL MSL SFLLLLFFSH LILSAWAHGE KRLAPKGQPG PAATDRNPIG

K-FGF FGF-6 FGF-5 aFGF bFGF KGF INT-2

AELERRWESL VALSLARLPV AAQPKEAAVQ SGAGDYLLGI D..SRGWGTL ..LSRSRAGL AGEISGVNWE SG...YLVGI SSSRQSSSSA MSSSSASSSP AASLGSQGSG LEQSSFQWSL MAEGE ITTFTALTEK FN ...LPPGN MAAGS ITTLPALPED GGSGAFPPGH LVGTISLACN DMTPEQMATN VNCSSPERHT RSYDYMEGGD GLIWLLLLSL LEPGWPAAGP GARLRRDAGG RGGVYEHLGG

....... .................................................. .................................................. ........................... MHK WILTWILPTL LYRSCFHIIC ................................................. M

...............

...............

.DTRDSLLEL

K-FGF FGF-6 FGF-5 aFGF bFGF KGF INT-2

89 81 93 32 35 72 51

.KRL

.

G Y F I AP

138 130 142 81 84 120 98

SYKYPGM. SDLYRGT. SAIHRTEK TG..... ... SKKHAEKN SRKYT..S SAKWTHNG GE SRLYRTVS STPGARRQPS

..........

171 169 184 121 122 162 148

.......... ............ SE QPELSFTVTV SD .......... s. .......... ............

206 198 232 155 155 194 196

..........

.......... .......... ........

.....FIALS KN .....YIALS K . .REWYVALN K ....WFVGLK K ....WYVALK R

.. NRVSPTMKVT .. SKVSPIMTVT CS PRVKPQHIST .. PRTHYGQKAI .. SKTGPGQKAI .. KKTKKEQKTA .. FKTRRTQKSS

RL....

K-FGF FGF-6 FGF-5 aFGF bFGF KGF INT-2

MEVALN Q AERLWYVSVN G

K-FGF FGF-6 FGF-5 aFGF bFGF KGF INT-2

........................................... ........................................... PEKKNPPSPI KSKIPLSAPR KNTNSVKYRL KFRFG ........ ........................................... ........................................... ........................................... GLPRPPGKGV QPRRRRQKQS PDNLEPSHVQ ASRLGSQLEA SAH

....

23 1

FGVASRFFVA FGVKSALFIA RGVF SNKFLA KSTETGQYLA KGVCANRYLA KGVESEFYLA RGLFSGRYLA

.ENPYSLLEI .ANMLSVLEI RSDQHIQLQL KSDPHIKLQL MKNNYNIMEI NS.AYSILE1

K-FGF FGF-6 FGF-5 aFGF bFGF KGF INT-2

40 39 43

HR DHEMVRQLQS

267

239

FIG.2. Protein sequence homology within the FGF family. The deduced amino acid sequences of the seven FGFs are aligned with some conserved motifs highlighted. The arrows bracket the conserved “core”region common to all FGFs. Sequences are for human FGFs, with the exception of FGF-6, which is murine. The N-terminal extensions produced by CUG initiation of bFGF and INT-2 mRNAs are not shown.

128

CLAUD10 BASILICO AND DAVID MOSCATELLI

bFGF has been found in most tissues and organs examined. A variety of cultured cells, including a number of tumor cell lines (Klagsbrun et al., 1986; Moscatelli et al., 1986a; Presta et al., 1986; Schweigerer et al., 1987b), synthesize bFGF, suggesting that expression of bFGF is widespread in vivo. There is some evidence that expression of bFGF increases as an adaptation to culture (Speir et al., 1991), raising the possibility that conclusions about the source of bFGF based on cultured cells may be misleading. However, immunolocalization studies of normal adult tissue have shown that bFGF is indeed produced in vivo by a variety of cell types, including skeletal, cardiac, and smooth muscle, and epithelial cells of sweat glands, trachea, bronchi, colon, and endometrial glands (Cordon-Cardo et al., 1990). bFGF is also found in the endothelial cells of some blood vessels, but not in all (Cordon-Cardo et al., 1990; Hanneken et al., 1989). In the adult, the mRNA is expressed in much higher levels in brain than in other tissues (Shimasaki et al., 1988). In normal brain, high-level expression is restricted to discrete areas, where it is present mainly in neuronal cell bodies, but lower level expression is found throughout the tissue, where it seems to be associated with glial cells (Emoto et al., 1989; Finklestein et al., 1988; Pettmann et al., 1986). In injured brain the expression of bFGF increases in the area of injury and seems to be associated with reactive astrocytes (Finklestein et al., 1988). aFGF appears to have a more limited distribution than bFGF. It has been found in neural tissue, kidney, prostate, and cardiac muscle (Casscells et al., 1990; Crabb et al., 1986; D’Amore and Klagsbrun, 1984; Gautschi-Sova et al., 1987; Quinkler et al., 1989; Sasaki et al., 1989; Thomas et al., 1984). lmmunolocalization studies have detected aFGF in neurons in discrete regions of the cerebrum and cerebellum (Wilcox and Unnerstall, 1991). It has also been identified in cultured vascular smooth muscle cells (Weich et al., 1990; Winkles et al., 1987). T h e physiological distribution of INT-2 is much more restricted. INT-2 is generally not expressed in adult tissues, including the mammar y gland, although it appears to be produced in precise steps of embryo development (Jakobovits et al., 1986; Wilkinson et al., 1988). INT-2 transcripts were detected by in situ hybridization in extraembryonic endoderm, localized to the parietal endoderm. INT-2 RNA is also present in cells of the primitive streak and later detected in the hind brain and pharingeal pouches. Sometime beyond day 10 of gestation INT-2 expression appears to be switched off. Similarly, K-FGF is also undetectable in adult tissues and in “normal” cell lines (Hebert et al., 1990; Velcich et al., 1989), but is present in the mouse blastocyst (inner cell mass), primitive streak, and myotomes (Niswander and Martin, 1992). Interestingly, INT-2 and K-FGF have a mirror-like pattern of expression in embryonal

FGF FAMILY OF GROWTH FACTORS AND ONCOGENES

129

carcinoma (EC) cells. K-FGF is expressed in undifferentiated EC cells, and induction of differentiation shuts off its expression (Curatola and Basilico, 1990; Velcich et al., 1989). INT-2 is transiently turned on after induction of differentiation (Jakobovits et al., 1986). Thus the pattern of expression of these two genes suggest that their physiological role may be related to development. I n Xenopzcs embryonic development, both bFGF and K-FGF appear to be expressed (Kimelman et al., 1988; Slack and Isaacs, 1989; J. C. Slack, personal communication). Low levels of maternal messages are present, and they increase in abundance at the blastula level. Both these factors can induce mesoderm formation in Xenopus animal pole explants, strongly suggesting a role for FGFs in mesodermal induction (Kimelman et al., 1988; Paterno et al., 1989; Slack et al., 1987). FGF-5 is also expressed at specific development stages in the mouse embryo (Haub and Goldfarb, 1991), but is also present in adult brain (Haub et al., 1990) and many “normal” and tumor cell lines (Zhan et al., 1988). During development, FGF-5 RNA is not detected in the blastocyst, but it later appears in the postimplantation epiblast, splanchnic mesoderm, somatic mesoderm, myotomes, limb mesenchymes, and acoustic ganglia (days 12-14). Interestingly, at several of these sites, expression is spatially restricted within the tissue (Haub and Goldfarb, 1991). Thus, although it appears that FGF-5 may be in some cases expressed at similar times and sites as INT-2 and K-FGF, it is quite possible that different cells in the developing organs express different types of FGFs. Thus, these observations strongly suggest a role for INT-2, K-FGF, and FGF-5 in development, but the precise meaning of these findings remains to be elucidated. Not much is known yet about FGF-6 and KGF, except that the latter is produced by a variety of epithelial tissues and stromal cells. MOLECULAR REGULATION OF FGF EXPRESSION The cis- and trans-acting elements involved in regulating FGF transcription are just beginning to be elucidated. Perhaps the FGF gene whose control of transcription is best known is K-FGF. T h e K-FGF gene has a canonical TATA box located about 30 nucleotides upstream of the transcription initiation site. However, the KFGF promoter and upstream DNA sequences can promote only very low levels of transcription when placed upstream of a reporter gene [such as chloramphenicol acetyltransferase (CAT)] on transfection in a variety of cell types, including undifferentiated EC cells, which express high levels

130

CLAUD10 BASILICO AND DAVID MOSCATELLI

of endogenous K-FGF transcripts (Curatola and Basilico, 1990). Juxtaposition of generalized enhancers [such as that of simian virus 40 (SV40)] to the K-FGF promoter results in ubiquitous, nonspecific expression, and a similar mechanism (juxtaposition of general enhancer elements) is likely to be responsible for activation of the K-FGF oncogene. Physiological K-FGF transcription depends instead on the presence of an enhancer element that is located in the 3’-untranslated region of the third exon and is present in a comparable location in the murine and human gene (Curatola and Basilico, 1990). This enhancer promotes transcription only in undifferentiated EC cells, but not in their differentiated counterpart or in HeLa or 3T3 cells, thus mimicking the pattern of expression of the endogenous gene. This indicates that the K-FGF enhancer interacts with specific trans-acting factors whose expression is also developmentally regulated. T h e mouse K-FGF enhancer has been narrowed down to a minimum 270-nucleotide fragment with full activity that contains a series of consensus binding sites for several known transcription factors, including Spl and AP4. Site-directed mutagenesis of these binding sites decreases somewhat enhancer strength, but mutagenesis of an octamer-binding site completely abolishes enhancer activity (A. M. Curatola, Daaka, Dailey, and C. Basilico, unpublished). Thus the critical factor(s) for KFGF transcription are likely to belong to the family of octamer-binding proteins, some of which have been recently known to be developmentally regulated (Scholer et al., 1989). The K-FGF transcript is a single RNA species of 3.4 Kb (Velcich et al., 1989). There is no evidence of posttranscriptional or translational control. An mRNA of about 1.1 kb is, however, strongly expressed in 3T3 cells transfected with the K-FGF oncogene as originally isolated (Delli Bovi et al., 1987). This RNA is identical to the physiological RNA species in the coding region, but is prematurely terminated approximately 250 nucleotides downstream of the translation stop codon. This is likely to result from the 3’ rearrangement found in the K-FGF oncogene, which occurs upstream of and eliminates the normal termination site (Delli Bovi and Basilico, 1987). The 1.1-kb RNA is somewhat more stable than the normal transcript, and this could contribute to the activation of the K-FGF oncogene, which, however, appears to be mainly transcriptional. T h e protein is efficiently secreted. Thus the main regulat,ion of this gene appears to be transcriptional. Not much is known about transcriptional regulation of the bFGF and aFGF genes. Both genes produce multiple transcripts, likely to result from alternative splicing and polyadenylation (Abraham et al., 198613; Crumley et al., 1990). The human bFGF promoter does not appear to

FGF FAMILY OF GROWTH FACTORS AND ONCOGENES

131

have a TATA box, but contains several potential SP1 and one APl binding site. Negative regulatory elements, whose deletion increases gene expression, appear to be present 5' of the promoter region (Shibata et al., 1991). As mentioned above, the bFGF mRNA can be translated from four initiation sites: a canonical AUG, which results in the synthesis of the 155-amino acid protein, and three upstream CUGs (Florkiewicz and Sommer, 1989). T h e N-terminally extended forms appear to localize predominantly in the cell nucleus. T h e significance of this finding remains to be elucidated. Also, some forms of acidic FGF are found in the nucleus, but this localization does not appear to require the synthesis of higher molecular weight forms (Imamura et al., 1990). An interesting mechanism of regulation of bFGF expression has been reported in Xenopus oocytes (Kimelman and Kirschner, 1989). In addition to a transcript encoding bFGF, an antisense transcript is present in large excess. T h e antisense transcript hybridizes to bFGF mRNA, but surprisingly does not appear to inhibit its translation and could be involved in regulation of bFGF mRNA stability. As mentioned above, aFGF and bFGF are not secreted. Thus, a further control on their expression could be exerted at the level of cell release. INT-2 transcription appears also to be restricted to specific steps of development, but not much is known about its regulation. As mentioned above, INT-2 is expressed in EC cells only after induction of differentiation (Jakobovits et al., 1986; Mansour and Martin, 1988). Regulation appears to be mainly transcriptional, and the combination of three distinct promoters and two alternative polyadenylation sites generates six different RNA species, which, however, all have the same coding capacity (Grinberg et al., 1991; Smith et al., 1988). T h e main cis-acting elements necessary for INT-2 transcription appear to map to a 1.7-kb fragment of INT-2 DNA, which includes the three promoters, and about 1 kb of upstream sequences (Grinberg et al., 1991). Translation of INT-2 RNA can begin at two sites: from an AUG that produces a secreted protein (although apparently not very efficiently) and from an upstream CUG. The product of upstream initiation partially localizes to the cell nucleus (Acland et al., 1990). In addition, INT-2 RNA contains an out-of-frame AUG in its 5' region, and this AUG apparently interferes with correct translation initiation (Dixon et al., 1989). Thus it appears that INT-2 expression is controlled at at least two main levels: transcription and translation initiation. T h e human FGF-5 gene is transcribed into two main RNA species of 1.6 and 4 kb, likely to result from the use of alternative polyadenylation sites (Zahn et al., 1988). T h e regulatory elements of FGF-5 transcription are not yet identified, but there is evidence of translational control. T h e

132

CLAUDIO BASILICO A N D DAVID MOSCATELLI

FGF-5 mRNA contains a short out-of-frame open reading frame (ORF) upstream of the ORF coding for the growth factor. Deletion of the upstream ORF enhances FGF-5 translation and transforming ability (Bates et al., 1991). FGF-5 is efficiently secreted as a glycosylated protein. Not much is known about regulation of FGF-6 and KGF expression. They both appear to be secreted proteins. In conclusion, regulation of FGF expression is complex and takes place at many levels. T h e data available so far are compatible with the hypothesis that release of these very potent and broad spectrum mitogens must be tightly regulated physiologically. Thus K-FGF has a very tight transcriptional control. Basic and acidic FGF, which are more ubiquitously transcribed, have a complex regulation of translation and are normally not secreted. INT-2 expression is controlled transcriptionally and translationally. Perhaps the broader the spectrum of action of the growth factors, the tighter is the regulation of their production. IV. FGF Receptors The discovery of seven growth factors, many of which seem to have a very similar spectrum of action, raises the question of what could be the evolutionary advantage for the organism in producing many growth factors with similar target specificity. While a partial answer to this question can be glimpsed from the pattern of expression of these genes, which is quite different, the final answer will require the identification and characterization of FGF receptors, their tissue distribution, and specificity of interaction with the various members of the FGF family. A number of laboratories have studied binding of basic and acidic FGFs to cellular plasma membranes. These studies led to the conclusion that these growth factors bound to two types of cell surface receptors. A low-affinity receptor, which is widely distributed with a large number of sites per cell (1-2 x lo6) and with a binding affinity of 2 to 10 nM. These receptors are likely to be heparan sulfate proteoglycans (HSPGs). In addition, studies demonstrate the presence of high-affinity receptors ( K d 10-100 pM), with a lower number of sites per cell (10,000-100,000) and a molecular weight on the order of 110,000-160,000. These receptors are glycosylated proteins with intrinsic tyrosine kinase activity, a characteristic of many growth factor receptors (reviewed in Baird and Bohlen, 1990; Burgess and Maciag, 1989; Rifkin and Moscatelli, 1989). T h e elucidation of the nature of these receptors has taken great impulse from the first cloning of the complete cDNA for an FGF receptor, which was performed by Lee et al. (1989) after receptor purification from chicken embryos. This receptor turned out to be highly homolo-

-

FGF FAMILY OF GROWTH FACTORS AND ONCOGENES

133

gous to the protein encoded in a partial cDNA (flg) isolated from a human endothelial cell cDNA library by its homology to a tyrosine kinase receptor (Ruta et al., 1988). Another previously identified partial cDNA of mouse origin, bek, also isolated on the basis of its encoding a tyrosine kinase (Kornbluth et al., 1988), was quickly seen to have substantial homology to the Flg receptor. Subsequent or concurrent efforts by a number of laboratories have led to an explosion in this field (Dionne et al., 1990; Hattori et al., 1990; Johnson et al., 1990; Keegan et al., 1991; Mansukhani et al., 1990; Musci et al., 1990; Partanen et al., 1991; Pasquale, 1990; Pasquale and Singer, 1989; Ruta et al., 1989; Safran et al., 1990), with results that at present can be summarized in the following manner: the high-affinity FGF receptors also constitute a gene family, which includes at least four members. They share a common structure consisting (from the N terminus to the C terminus) of a signal peptide, three immunoglobulin-like loops flanked by characteristic cysteines, and a hydrophobic transmembrane region. There is a characteristic acidic region between the first and second immunoglobulin (Ig) loop. The intracellular domain includes the catalytic tyrosine kinase domain, which is split by a short kinase insert, as is present in the platelet-derived growth factor (PDGF) receptor (reviewed in Ullrich and Schlessinger, 1990), and a rather long C-terminal tail (Fig. 3). The degree of homology of the receptors varies, with the highest homology in the tyrosine kinase domain, and the lowest in the first Ig loop (Fig. 3). T h e nomenclature of these receptors is starting to be as confusing as that of the FGFs, reflecting the isolation of these clones (sometimes fortuitous), the species of origin, etc. Thus, FGF-R1 has been called Flg, Cek-1, etc. FGF-R2 has been called Bek, Cek-3, etc. FGF-R3 has been called also Cek-2. FGF-R4 possibly has no other name. The primary transcripts of these genes are unusually prone to alternative splicing (Hou et al., 1991). Thus cDNAs encoding soluble receptors, truncated receptors, Ig loop variants, etc., have been isolated (Johnson et al., 1990). Their physiological and pathological significance remains to be elucidated and with some exceptions will not be discussed here. Since these receptors are often coexpressed in tissue culture cell lines, it becomes clearly necessary to express each molecule in receptornegative cells to test their specificity of binding, receptor activation, etc. To date, the following facts have emerged. 1. The FGFR-1 (Flg) has been expressed in Chinese hamster ovary (CHO) cells, N l H 3T3 cells, and in Xenopus oocytes (Dionne et al., 1990; Johnson et al., 1990; Mansukhani et al., 1990). Chinese hamster ovary

134

CLAUD10 BASILICO A N D DAVID MOSCATELLI

TK2

CT I

I

I I I

I

I

s

II

I

I I

I

I I I I

I

I

I

I I I I I

I

II

I

586

i 47

i322

I I

,

overall identity

X identity FGFR-I/FGFR-2 FGFR-I/FGFR-4 FGFR-IIFGFR-3 FGFR-2/FGFR-4 FGFR-Z/FGFR-S FGFR-3/FGFR-4

38 22 21 21 27 26

81 63 66 70 72 78

79 74 82 72 81 74

66 32 41 31 45 34

88 75 83 713 87 78

50 14 43 29 50 36

92 06 91 84 92 86

60 38 44 48 56 46

71 56 61 57 65 60

FIG. 3. Structure of the FGF receptors and protein homology among the various domains of the four human FGF receptors known. The “wild-type”form of receptors is shown. Variant forms produced by alternative splicing are not shown. Amino acid numbers on the schematic are those for human FGFR-1 (flg). Loops 1-111 represent the extracellular Ig-like domains. TM, transmembrane region; JM, juxtamembrane region; TK, tyrosine kinase domains; KI, kinase insert; CT, carboxy-terminal tail, Stippled box, signal peptide; dark box, acidic region.

cells are essentially FGF receptor negative, although some differences between the various CHO strains and clones are likely to exist. NIH 3T3 cells are receptor positive and necessitate large levels of expression. Testing in Xenopus oocytes (following microinjection of RNA) allows measuring calcium effluxes in response to FGFs, but does not allow measurements of binding affinity, etc. FGFR-1 appears to bind acidic and basic FGF with similar high affinity and K-FGF with about 15-fold lower affinity (Dionne et al., 1990; Mansukhani et al., 1990). The receptor kinase is activated following growth factor binding and a significant proliferative response is obtained. Results in other systems confirm these conclusions. Interestingly, a murine Flg receptor lacking the first immunoglobulin loop behaves in this assay like the wild-type receptor (Mansukhani et al., 1990), indicating that this domain of the protein is not important for FGF binding or receptor activation. 2. The FGFR-2 (Bek) cDNA has also been expressed in CHO and NIH 3T3 cells (Dionne et al., 1990; Mansukhani et al., 1992). It appears to bind aFGF, bFGF, and K-FGF with similar high affinity. Interestingly, the recently described KGF receptor (Miki et al., 1991) appears to be a

FGF FAMILY OF GROWTH FACTORS AND ONCOGENES

135

Bek variant that differs from prototype Bek in lacking the first Ig-like domain and the acidic region and considerably diverges in the third Ig loop. This receptor appears to bind KGF and acidic FGF with high affinity, but binds bFGF with about 20 times lower affinity (Miki et al., 1991). Considering what was said above, this strongly points to the third Ig loop as an important domain for growth factor binding. 3. T h e third and fourth cDNAs encoding distinct but related FGF receptor molecules have been isolated by screening cDNA libraries from K562 erythroleukemia cells for tyrosine kinases (Keegan et al., 1991; Partanen et al., 1991). This was somewhat surprising, since hematopoietic cells are generally not thought to respond to FGFs, but is in agreement with the finding that human hematopoietic stem cells respond to basic FGF with increased survival and proliferative ability (Gabbianelli et al., 1990). FGFR-3 is found expressed in brain, lung, and kidneys, while FGFR-4 is prevalent in adrenals, lung, and pancreas. A detailed analysis of binding specificities of FGFR-3 and -4 is not yet available. Preliminary results indicate that both FGFR-3 and -4 bind acidic FGF with higher affinity than basic FGF (Partanen et al., 1991; D. Ornitz, personal communication). Although these results are still rather preliminary, they clearly suggest a complex picture of many receptors with overlapping specificity of binding, yet with preferential affinity for one ligand or another. The physiological significance of a low binding affinity (e.g., FGFR-1 for KFGF) remains unknown in the absence of a precise knowledge of what are the real concentrations of these growth factors in tissues and organs. It could represent a purely evolutionary conservation of structure, with no physiological implications: in other words, in the example chosen, Flg would not be the physiological target for K-FGF. On the other hand, this could be another mechanism of regulation of growth factor action, with activation taking place only at relatively high growth factor concentrations. The interpretation of the final effectiveness of one or another of the FGFs vis a vis their affinity for a specific receptor is complicated by what has recently emerged on the low-affinity receptors. Until recently the function of these receptors was substantially unknown. A variety of biochemical evidence indicated that they consisted of heparan sulfate-like glycosaminoglycans, present on the cell surface and in the extracellular matrix. It was also known that binding to low-affinity receptors could be competed by high concentrations of soluble heparin. Since heparin stabilizes and somewhat potentiates the action of most FGFs, an essential role of low-affinity receptors in receptor activation and signal transduc-

136

CLAUD10 BASILICO AND DAVID MOSCATELLI

tion was generally not postulated. The situation has changed dramatically during the last year, as several lines of evidence emerged to indicate an essential role of low-affinity receptors in FGF action. Barr’s group reported the cloning of a low-affinity hamster receptor (Kiefer et al., 1990). T h e predicted protein structure is that of an HSPG with a short cytoplasmic tail. Expression of this molecule can make lymphoid cells capable of binding to FGF-coated dishes. Interestingly, the protein is highly homologous to the mouse syndecan HSPG, which has been implicated in cell adhesion and other processes (Saunders et al., 1989). The primary amino acid sequence predicts a proteolytic cleavage site separating most of the extracellular portion from the transmembrane and cytoplasmic regions. Thus this portion of the molecule could become part of the ECM. Evidence pointing to the importance of this or similar molecules comes from a diverse set of experiments. When the Flg receptor was transfected into CHO cell mutants deficient in the production of heparan sulfates, it was found that these cells did not bind bFGF (while wild-type cells did), but could bind bFGF in the presence of heparin (Yayon et al., 1991). It was also shown that treatment of various cell types with heparitinase or culture in the presence of sodium chlorate, which blocks sulfation, resulted in drastic reduction of FGF binding to its receptors, and again binding was restored by the presence of heparin (Rapraeger et al., 1991). In addition, transfection of the Flg or Bek receptor into 32D cells, an IL-3-dependent murine hematopoietic cell line that does not express significant levels of HSPG, creates cells that now can grow in the presence of bFGF and K-FGF, but only in the presence of heparin. Low doses of heparin are sufficient (0.5-1 Fg) to show this effect and it can be shown that the action of heparin does not consist in stabilization of FGF against proteolytic degradation (Mansukhani et al., 1992). All these results suggest that FGFs must interact with low-affinity receptors in order to be able to activate the high-affinity receptors, and that the role of low-affinity receptors can be replaced by heparin. The fact that soluble heparin can substitute for the low-affinity receptors suggests that it modifies somewhat the structure of the growth factor and makes it functionally capable of binding and activating the highaffinity receptors (Fig. 4). It does not support the hypothesis that lowand high-affinity receptors oligomerize and create a higher affinity structure. Much about this novel mechanism of “presentation” of a growth factor to its receptors remains to be elucidated. The nature and diversity of low-affinity receptors will surely be investigated. It will have to be conclusively demonstrated that the heparin requirement in the systems de-

FGF FAMILY OF GROWTH FACTORS AND ONCOGENES

137

FIG.4. bFGF interaction with heparin (Hep) is necessary for its binding to receptors. The diagram represents the proposal that bFGF in the absence of heparin exists in a conformation that is not able to interact with its receptor. Interaction with heparin in solution or heparan sulfates (HSPG) on the cell surface convert the bFGF to a conformation that is recognized by the receptor. [Reprinted from Yayon et al. (1991).]

scribed above can be alleviated by the expression of low-affinity receptors. Finally, it is evident from what is discussed above that modifications of FGF “potency” could be achieved by a factorial combination of their ability to bind low- and high-affinity receptors. T h e modification of FGF structure or function following heparin binding, and how that affects receptor binding, will have to be studied. Much of this work will probably require studying the crystal structure of FGFs with or without heparin, that of the extracellular domain of the receptors, etc. T h e mechanism of signal transduction by FGF receptors is only beginning to be studied. Like many other receptor tyrosine kinases, FGF receptors probably oligomerize following ligand binding (Ullrich and Schlessinger, 1990; Yarden and Ullrich, 1988), resulting in activation of tyrosine kinase activity and trans- and autophosphorylation. One of the substrates of the FGF receptor tyrosine kinase appears to be phospholipase Cy and a 90-kDa protein can also be easily detected in mouse cells by anti-phosphotyrosine antibodies following receptor activation (Burgess et al., 1990b; Coughlin et al., 1988). T h e nature of the 90-kDa protein is still unknown. The significance of the frequent coexpression of some of the receptors (e.g., Flg and Bek) is yet unknown. Recent evidence indicates that different FGF receptors can create heterodimers following ligand binding (Bellot et al., 1991). It will be interesting to know whether this leads to different patterns or efficiency of activation, i.e., whether coexpression of two distinct receptors with similar affinity for a given growth factor leads to different patterns of oligomerization and receptor activation and whether that in any way modifies or regulates the interaction of activated receptors with their substrates.

138

CLAUD10 BASILICO AND DAVID MOSCATELLI

V. Interaction with Extracellular Matrix

Although signal transduction occurs through binding to receptors, the actions of FGFs are also influenced through their interactions with extracellular matrix. As expected from its high affinity for heparin, bFGF added exogenously to cultured cells binds to h'eparin-like molecules produced by the cells (Baird and Ling, 1987; Moscatelli, 1987; Vlodavsky et al., 1987). These molecules have been identified as HSPGs present on the cell surface and in the extracellular matrix (Saksela et al., 1988; Saksela and Rifkin, 1990). bFGF interacts with the heparan sulfate moieties of these molecules (Saksela et al., 1988) and efficient binding depends on the presence of N-sulfate groups (Bashkin et al., 1989). In endothelial cell cultures, the cell surface and extracellular matrix bFGFbinding HSPGs appear to be distinct molecules, the cell surface form having a molecular weight of 250K and the matrix-associated form having a molecular weight greater than 800K (Saksela and Rifkin, 1990). As mentioned above, the cell surface bFGF-binding HSPG of hamster kidney cells was recently cloned and was found to be homologous to the mouse HSPG syndecan (Kiefer et al., 1990). Cloned syndecan has also been shown to bind bFGF. However, syndecan is probably not the only cell surface HSPG that binds bFGF, and other bFGF-binding HSPGs are likely to be present in other cell types. bFGF has also been shown to bind to heparin-like molecules in basement membranes in vivo (Jeanny et al., 1987) and to be present in isolated basement membranes (Folkman et al., 1988), providing further evidence that bFGF-binding HSPGs exist in addition to the cell surface ones. bFGF has a lower affinity for these heparan sulfates (2-600 X 10-gM) than for its cell surface high-affinity receptors (2-5 x 10- M) (Bashkin et al., 1989; Dionne et al., 1990; Moenner et al., 1986; Moscatelli, 1987; Vigny et al., 1988). Binding to the HSPGs does not preclude bFGF from binding to receptors (Moscatelli, 1987, 1988). Rather, interaction with HSPGs seems to confer specific biological advantages to bFGF. First, the HSPG-bound bFGF can act as a reservoir of growth factor for the cells. For endothelial cells, it has been shown that when the cells are exposed briefly to 10 ng/ml bFGF, most of the growth factor binds to HSPGs. If the growth medium is changed, leaving the cells with only the HSPG-bound bFGF, biological responses measured 24 hr later are the same as in cultures continuously exposed to bFGF-containing medium. However, if bFGF is stripped from its HSPG-binding sites, biological responses measured 24 hr later are greatly reduced (Flaumenhaft et al., 1989). These results suggest that cells can use the HSPGs as a temporary storage site for bFGF and can draw on this reserve of growth factor to

FGF FAMILY OF GROWTH FACTORS AND ONCOGENES

139

greatly increase the response to a brief exposure to bFGF. Second, bFGF bound to heparin or heparan sulfates is protected from thermal denaturation and proteolytic degradation (Gospodarowicz and Cheng, 1986; Saksela et al., 1988; Sommer and Rifkin, 1989). Thus, HSPGbound bFGF is more stable. Third, soluble heparan sulfates can act as carriers for bFGF, increasing its radius of diffusion. Since bFGF interacts strongly with HSPGs fixed in place on the surface of cells and in the extracellular matrix, it does not diffuse freely from its site of release (Flaumenhaft et al., 1990). Soluble heparan sulfates bind bFGF, neutralizing its interaction with fixed HSPGs and increasing its availability to cells distant from its site of release (Flaumenhaft et al., 1990). Finally, as described above, Yayon et al. (1991) showed in an elegant series of experiments that bFGF interaction with heparin or heparan sulfates is necessary for its interaction with receptors. The simplest interpretation of these experiments is that binding to heparin or heparan sulfates alters the conformation of bFGF so that it can then interact with its binding site on the receptor. Since all members of the FGF family bind strongly to heparin, it is likely that other members of the family also interact with heparin-like molecules in the extracellular matrix. Indeed, aFGF has also been shown to bind to extracellular matrix molecules produced by cells (Gordon et al., 1989; Kan et al., 1988) and to basement membranes in uiuo (Jeanny et al., 1987). Furthermore, more than other members of the family, the biological activity of aFGF is greatly protentiated by the addition of heparin (Mueller et al., 1989; Schreiber et al., 1985; Uhlrich et al., 1986) or HSPGs (Gordon et al., 1989). It is not entirely clear how this potentiation occurs, but it could be related to some of the effects observed for the interaction of bFGF with heparin and heparan sulfates. Like bFGF, aFGF is protected from thermal denaturation and proteolytic degradation by heparin (Gospodarowicz and Cheng, 1986; Rosengart et al., 1988) and therefore may be more stable in .a complex with heparin or HSPG. In addition, interaction of aFGF with heparin seems to alter the conformation of the protein since addition of heparin increases the binding of aFGF to specific monoclonal antibodies (Schreiber et al., 1985). This change in conformation may also be responsible for the approximately twofold greater affinity of aFGF for its receptor in the presence of heparin (Kaplow et al., 1990; Schreiber et al., 1985). The interaction of other members of the family with extracellular matrix molecules has not been studied in detail. Interactions with the extracellular matrix may be especially important in regulating the action of the secreted members of the FGF family. Despite their efficient release from cells, interactions with the extracellular matrix may restrict the

140

CLAUD10 BASILICO AND DAVID MOSCATELLI

distribution of these factors to areas in the immediate vicinity of the producing cell and thereby modulate their activity. VI. Biological Function

bFGF and aFGF are potent mitogens for a variety of cells of mesodermal, ectodermal, and endodermal origin (Gospodarowicz et al., 1987). However, the absence of a signal sequence for secretion in these molecules makes it difficult to understand their role and mode of action in viuo. How are these molecules released to exert their effects in viuo? It has been proposed that they are released from dead or dying cells and, thus, may be primarily involved in responses to tissue destruction. In a variation on this, it has been suggested that bFGF is released through small, nonlethal disruptions of the plasma membrane (McNeil et al., 1989). However, there is increasing evidence that, despite the lack of a signal sequence, low levels of the growth factor are released by healthy cells. In cultured endothelial cells, which both synthesize and have receptors for bFGF, basal levels of protease production and DNA synthesis are inhibited by neutralizing antibodies to bFGF (Sakaguchi et al., 1988; Sato and Rifkin, 1988). These results suggest that the cells release small amounts of bFGF that can activate their own FGF receptors in an autocrine manner. An alternative explanation is that, in mass culture experiments, the death of a minute percentage of the cells releases enough bFGF to cause these results and that healthy cells release no bFGF. This possibility was addressed in experiments in which the migration of single cells expressing different amounts of bFGF was investigated (Mignatti et al., 1991). Cell movement was shown to be proportional to the content of bFGF and could be inhibited by antibodies to bFGF, indicating that the cells were responding to their o w n bFGF, which was released to a space accessible to the antibodies. Since only a single cell was present in each well during this experiment, the bFGF had to be released from the responding cell. Thus, it appears that healthy cells are able to release small but biologically significant amounts of bFGF. The mechanism of bFGF release is still unclear. One of the major roles proposed for the FGFs in uiuo is in the induction of new blood vessel growth or angiogenesis (Folkman and Klagsbrun, 1987). Angiogenesis occurs physiologically in the development of the vascular system during embryonic, fetal, and adolescent growth, and in the growth of the uterine lining during the menstrual cycle. Angiogenesis also contributes to several pathologies either directly, as in diabetic retinopathy, or indirectly by supporting the growth of pathologic tissues, as in rheumatoid arthritis and tumor growth. Neovasculariza-

FGF FAMILY OF GROWTH FACTORS AND ONCOGENES

141

tion occurs from capillaries and is initiated when the capillary endothelial cells break through their basement membranes, migrate toward the source of angiogenic inducer, and proliferate, forming new cords of endothelial cells that eventually develop into capillaries (Ausprunk and Folkman, 1977). T h e FGFs have effects on cultured endothelial cells that are consistent with a role in this process. bFGF has been shown to induce an invasive phenotype in cultured endothelial cells, enabling them to penetrate basement membranes in vitro (Mignatti et al., 1989). T h e ability to penetrate the basement membrane is dependent on the increased production of the proteolytic enzymes plasminogen activator and collagenase in response to bFGF (Mignatti et d . , 1989; Moscatelli et al., 1986b; Presta et al., 1986). aFGF and K-FGF are likely to have similar effects, since they also stimulate the production of proteolytic enzymes in cultured endothelial cells (Delli Bovi et al., 1988). In addition, both aFGF and bFGF are chemotactic for endothelial cells (Moscatelli et al., 1986b; Terranova et al., 1985), suggesting that these factors support the directed growth of capillaries during angiogenesis. Finally, aFGF, bFGF, K-FGF, and FGF-5 stimulate endothelial cell proliferation (Delli Bovi et al., 1988; Gospodarowicz et al., 1987; Zhan et al., 1988). Thus, members of the FGF family have properties expected of angiogenic factors, and, indeed, aFGF and bFGF have been shown to induce angiogenesis in vzvo in a number of model systems (Hayek et al., 1987; Lobb et al., 1985; Shing et al., 1985; Thomas et al., 1985). However, the roles of the FGFs in viuo have been difficult to sort out, not only because of the overlapping biological properties of the members of the FGF family, but also because similar biological effects are also induced by unrelated growth factors. Because of their initial isolation as fibroblast growth factors and angiogenesis factors, bFGF and aFGF have been proposed to have a major role in wound healing, especially in the formation of granulation tissue and the accompanying neovascularization. This hypothesis was supported by experiments that showed that application of exogenous bFGF increased granulation tissue formation and the breaking strength of incisional wounds (Davidson et al., 1985; McGee et al., 1988). Furthermore, bFGF greatly improved the normally impaired healing of dermal wounds in diabetic mice, restoring the response to levels seen in normal littermates (Tsuboi and Rifkin, 1990). However, these experiments, like the experiments demonstrating angiogenic effects of purified FGFs, show only that bFGF and aFGF have the capacity to promote wound healing and angiogenesis. Other growth factors have also been shown to have wound healing and angiogenic properties (Folkman and Klagsbrun, 1987), and it is not clear which factors are involved in the natural processes. Indeed, one study has

142

CLAUD10 BASILICO AND DAVID MOSCATELLI

suggested that bFGF is not involved in some instances of tumor angiogenesis, since high-level expression of antibodies to bFGF in mice did not reduce the angiogenesis of tumors in these animals (Matsuzaki et al., 1989). The role of endogenous bFGF in wound healing and angiogenesis has been addressed by Broadley et al. (1989). In these experiments, collagen sponges were implanted subcutaneously in rats, and the effect of antibodies to bFGF on the subsequent formation of granulation tissue in these sponges was investigated. When a pellet that slowly released neutralizing antibody to bFGF was included in the sponge, vascularization of the sponge and granulation tissue formation were inhibited. Thus, endogenous bFGF seems to be required for wound repair and neovascularization in at least some instances. The FGFs also seem to have important functions in early development. bFGF has been shown to behave as an embryonic inducer in early Xenopus embryos. Induction of mesoderm is a result of signals generated from the endoderm. When the animal pole region is dissected away from a Xenopw blastula, the isolated animal pole forms only ectoderm. If the animal pole tissue is incubated with bFGF, mesodermal structures are formed (Kimelman et al., 1988; Slack et al., 1987). bFGF is expressed in the oocyte and early embryo and receptors that recognize bFGF are present in the blastula, consistent with its proposed role in early development (Gillespie et al., 1989; Kimelman et a/., 1988; Slack and Isaacs, 1989). However, aFGF,K-FGF, and INT-2 also are able to induce mesodermal structures (Grunz et al., 1988; Paterno et al., 1989; Slack et al., 1988), and one of these molecules, especially one of the secreted family members, may be the actual inducer in vivo. Xenopus blastulas produce other potent mesoderm inducers, called activins, that are members of the transforming growth factor f3 (TGF-P)family (Thomsen et al., 1990). While the FGF family member that acts as the physiological inducer has not yet been identified, an elegant demonstration of the importance of FGFs in the formation of mesoderm in Xenopw embryos has just been provided (Amaya et al., 1991). Expression of a truncated, tyrosine kinase-deficient FGF receptor in Xenopus embryos makes them unable to induce mesoderm in response to FGFs and causes specific defects in gastrulation and development. The truncated FGF receptor can be shown to act as a dominant negative mutant that abolishes wild-type receptor function. Thus FGFs not only can induce mesoderm in Xenopus, but their signaling pathway appears to be essential in the formation of posterior and lateral mesoderms. Perhaps another indication that FGFs are involved in cell differentiation is the report that aFGF causes bladder carcinoma cells to lose their epithelial character and to acquire some properties typical of mes-

FGF FAMILY OF GROWTH FACTORS AND ONCOGENES

143

enchymal cells (Valles et al., 1990). On exposure to aFGF, the bladder carcinoma cells lost their desmosomal contacts and transformed into elongated fibroblastoid cells. This shape change was accompanied by a decrease in cell surface staining for desmosomal proteins and an induced expression of vimentin intermediate filaments (Boyer et al., 1989; Valles et al., 1990). Interestingly, bFGF was ineffective in altering the morphology of the bladder carcinoma cells, suggesting that these cells express a receptor that recognizes aFGF but not bFGF. T h e idea that the FGFs may have important functions in development is also supported by the highly tissue- and time-specific expression of members of this family during murine embryonic growth. At least two of the secreted members of the family, INT-2 and K-FGF, seem to be present only in the embryo or fetus and are not found in normal adult tissue. INT-2, K-FGF, and FGF-5 are expressed at precise steps of development: high levels of K-FGF are detected only in very early embryos, and INT-2 and FGF-5 are produced at later stages of embryonic and fetal growth, each in specific tissues (Hebert et al., 1990; Wilkinson et al., 1988, 1989). T h e secreted members of the FGF family may act as differentiation signals at specific steps in development. bFGF is expressed at high levels at later stages of embryonic and fetal development (Hebert et al., 1990) and has been implicated in some differentiation processes. For example, several lines of evidence suggest that bFGF may be involved in muscle differentiation. In the chicken embryo, bFGF is abundant in the myocardium, somite myotome, and developing limb bud muscle (Joseph-Silverstein et al., 1989). In the mouse fetus, bFGF is also detected in high levels in cardiac, skeletal, and smooth muscle (Gonzalez et al., 1990). The amount of bFGF present decreases as the tissues mature (Joseph-Silverstein et al., 1989). This seems to correlate with the observation that cultured myocytes remain undifferentiated when they are maintained in bFGF-containing medium, but differentiate into myotubes on withdrawal of bFGF (Clegg et al., 1987). Differentiation into myotubes is accompanied by a decrease in bFGF and aFGF mRNA expression (Moore et al., 1991). In addition, on differentiation, the expression of FGF receptors is down regulated (Moore et al., 1991; Olwin and Hauschka, 1988). Nevertheless, it is still unclear whether the high levels of bFGF in developing muscle are directly involved in stimulating the proliferation of myocytes or are simply a reflection of higher expression of bFGF mRNA in myocytes than in myotubes. The FGFs may also be involved in the differentiation and maintenance of nervous tissue. Both aFGF and bFGF as well as K-FGF cause rat PC 12 pheochromocytoma cells to send out neurites and to differentiate

144

CLAUD10 BASILICO A N D DAVID MOSCATELLI

into sympathetic neuron-like cells (Rydel and Greene, 1987; Schubert et al., 1987; Togari et al., 1983, 1985; Wagner and D’Amore, 1986). In quail embryos, bFGF is expressed by young neurons of the neural tube and the neural crest, and, at later stages, by neurons of the spinal cord and dorsal root ganglia (Kalcheim and Neufeld, 1990). In the spinal cord, bFGF levels reach a peak on embryonic day 10 and then decrease by hatching time (Kalcheim and Neufeld, 1990), suggesting that bFGF may be important in neural cell movements and formation of connections in the embryo. However, expression of aFGF, bFGF, and FGF-5 persists in adult brain, and brain contains much higher levels of mRNA for these growth factors than other adult tissues (Alterio et al., 1988; Haub et al., 1990; Shimasaki et al., 1988), suggesting that the FGFs are also important in neural physiology in adults. High-level expression of aFGF and bFGF and low-level expression of FGF-5 is limited to neurons in localized ares of the brain. Interestingly, aFGF and bFGF are expressed differently in neighboring areas of specific brain regions: aFGF is expressed in fields CA1 and CA3 of the hippocampus, while bFGF is expressed in field CA2; aFGF is preferentially localized in layers 3 and 5 of the cerebral cortex, while bFGF is expressed in layers 2 and 6 (Emoto et al., 1989; Wilcox and Unnerstall, 1991). In addition to being localized to specific neurons, both aFGF and bFGF seem to be expressed by glial cells (Emoto et al., 1989; Wilcox and Unnerstall, 1991). Both neural and glial-derived FGFs may have important neurotrophic effects on surrounding neurons. These growth factors have been shown to support the survival in culture of neurons isolated from numerous sites in the central nervous system, including the hippocampus, cerebral cortex, ciliary ganglion, spinal cord, and cerebellum (Hatten et al., 1988; Morrison et al., 1986; Unsicker et al., 1987; Walicke et al., 1986). Furthermore, the ability of glial cells to support the survival of neurons in culture has been shown to be due to glial-derived bFGF (Hatten et al., 1988). Finally, the FGFs have been shown to promote the survival of lesioned nerves in uiuo (Anderson et al., 1988). These results have been interpreted as evidence that the endogenous FGFs in the brain may be necessary for the survival of intact neurons in vivo or in establishing neuronal contacts. VII. Oncogenic Potential

It has become almost obvious that any gene encoding a growth factor has the potential to be an oncogene. Constitutive expression of such a gene in a cell that expresses specific receptor(s) for the growth factor can create an autocrine growth loop, resulting in self-sustained aberrant growth. It is therefore not surprising that three members of the FGF

FGF FAMILY OF GROWTH FACTORS A N D ONCOGENES

145

family were originally identified as oncogenes. INT-2 was identified as a gene frequently activated in mouse mammary carcinomas. K-FGF/HST and FGF-5 were originally isolated as genes capable of transforming NIH 3T3 cells in culture. Cloning of their cDNAs revealed the homologies of their predicted gene products to FGFs. T h e transforming potential of other FGFs is not equally clear, and in most cases the importance of FGFs as “natural” oncogenes, i.e., their involvement in the etiology or progression of “spontaneous” human or animal malignancies, remains to be established. No FGF has yet been found as a retroviral oncogene, i.e., an oncogene acquired by the many naturally occurring transforming retroviruses. Information about the oncogenic potential of the FGFs is variable from member to member. We will discuss first the molecular aspects of their oncogenic potential in uitro, and then their involvement in tumors. A. K-FGF/HST T h e human HST/K-FGF gene was isolated first in Japan by transfection of stomach cancer DNA into NIH 3T3 cells, and at about the same time in New York by transfection of DNA from Kaposi’s sarcoma (Sakamoto et ul., 1986; Taira et ul., 1987; Delli Bovi and Basilico, 1987; Delli Bovi et al., 1987). In both cases, the demonstration that the oncogene was activated in the original tumor is still lacking. While efforts at proving this point were probably hampered by the fact that the original DNA used for transfection came from biopsies and not a cultured cell line, it is probably safe to assume that the isolation of this gene was accidental, i.e., the gene was activated during transfection. Irrespective of this point, it is noteworthy that both the HST and K-FGF oncogenes, as originally described, carried DNA rearrangements (Sakamoto et al., 1986; Delli Bovi and Basilico, 1987). These rearrangements were probably important for “activation” of this oncogene. Dominant oncogenes are “activated” versions of their normal counterpart (i.e., protooncogenes) and the mechanism of activation has been shown to fall into two broad categories; mutations, ranging from point mutations to deletions and fusions with other coding sequences, and changes in the regulation of expression. T h e mechanism of activation of K-FGF/HST clearly consists of the latter. Cloning and expression of the human protooncogene revealed that the proteins encoded by the oncogene and protooncogene were identical (Yoshida et al., 1987; Delli Bovi et al., 1988). Thus the mechanism of activation must result from changes in the regulation of gene expression. This conclusion is in line with the finding (see above) that K-FGF expression is extremely re-

146

CLAUD10 BASILICO AND DAVID MOSCATELLI

stricted in normal tissues and organs, being probably limited to early stages of development (Hebert et al., 1990). It is therefore quite likely that the rearrangement found outside the coding sequences in the original oncogene isolates played a role in activating gene expression. It is likely (Curatola and Basilico, 1990) that this results from the juxtaposition of general enhancer sequences to the K-FGF gene, resulting in ubiquitous expression, but the possibility of deletion of negative regulatory sequences cannot yet be ruled out. A puzzling observation that has been made in several laboratories (Sakamoto et al., 1988; M. Goldfarb, personal communication; A. M. Curatola and C. Basilico, unpublished) is that the cloned K-FGF protooncogene has low but significant transforming ability when transfected into NIH 3T3 cells. In protooncogene-transformed cells the transfected gene is expressed while the endogenous K-FGF gene remains silent. The transforming ability of the K-FGF protooncogene could be due to integration next to cellular enhancer sequences or to the deletion of negative regulatory sequences on cloning or transfection. Neither hypothesis has yet been fully investigated. Curatola and Basilico (1990) found no evidence of the presence of negative regulatory sequences in the 5' portion of the K-FGF oncogene when these DNA regions were used to drive CAT expression in a transient assay. Transformation by K-FGF/HST appears to result from the creation of an autocrine growth loop in which constitutive expression of K-FGF in a cell that also expresses its receptors leads to secretion of the growth factor, activation of the receptor, and continuous stimulation of its signal transduction pathway (Talarico and Basilico, 1991). Cells are thus constantly induced to proliferate. It appears that secretion of the growth factor is required for transformation, i.e., the mitogenic pathway can be activated only at the cell surface. Anti-K-FGF neutralizing antibodies cause reversion of the transformed phenotype, and inhibit the proliferation of K-FGF-transformed cells in serum-free media. To investigate the question of why the phenotype of some K-FGF transformed lines was only partially reverted by antibodies, as well as to test the hypothesis (Keating and Williams, 1988) that interaction of the growth factor with the receptor could also occur intracellularly, Talarico and Basilico (1991) constructed K-FGF cDNA mutants encoding proteins with impaired secretion. These mutants had extremely reduced transforming ability and the rare transformed cells they could produce appeared to secrete minute amounts of the growth factor. Their phenotype was fully reverted by anti-K-FGF antibodies. Thus it appears that intracellular activation of the receptor cannot occur in this system, perhaps because FGFs must interact with heparan sulfates (see above) before activating the receptor.

FGF FAMILY OF GROWTH FACTORS AND ONCOGENES

147

Alternatively, this interaction could occur, but in such rare circumstances as to have no practical effect on the regulation of cellular proliferation.

B. INT-2 T h e INT-2 gene was originally identified and cloned as a frequent site of insertion of mouse mammary tumor virus (MMTV) proviral DNA in mouse mammary carcinomas. Insertion of the MMTV long terminal repeat (LTR) upstream or downstream of the INT-2 gene leads to activation of its expression (Dickson et al., 1984; Moore et al., 1986). Here again, therefore, the mechanism of INT-2 oncogenic activation results from induction of transcription, and the gene is not expressed in normal breast tissue (Dickson et al., 1984). Thus induction of INT-2 expression is probably one of the early events (but certainly not the only one; see below) in mouse mammary carcinogenesis. Our knowledge of the mechanism of transformation by INT-2 is limited, particularly because INT-2 is not a strong transforming gene in tissue culture systems. NIH 3T3 cells can be transformed by INT-2 but transformation is inefficient and appears to require large amounts of the protein (Goldfarb et al., 1991). One of the reasons for this poor efficiency may be the fact that INT-2 is not efficiently secreted, although it possesses a signal peptide (Dixon et al., 1989), possibly because upstream initiation of translation at an in-frame CUG results in the synthesis of a protein with 29 additional amino acids, which seems to localize to the nucleus (Acland et al., 1990). However, mutagenesis of the INT-2 cDNA clearly indicates that only the secreted form of INT-2 is transforming, while no discernible phenotype is associated with the expression of an Nterminally extended protein that has the signal peptide deleted and is almost exclusively localized in the nucleus (Dickson et al., 1991). Thus the general conclusions outlined for K-FGF, i.e., that transformation requires growth factor secretion and extracellular activation of the receptor(s), probably hold true for INT-2 also. It cannot, however, be ruled out that INT-2 is also a weak mitogen for the cells tested so far, possibly because its affinity for their receptor is low. Most of these questions will be answered when large amounts of pure INT-2 protein are available. C. FGF-5

FGF-5 was also isolated as an oncogene by transfection of DNA from a tumor cell line into NIH 3T3 cells. T h e mechanism of activation was clearly shown to result from the juxtaposition of enhancer sequences 5'

148

CLAUD10 BASILICO AND DAVID MOSCATELLI

to the FGF-5 gene. The rearrangement had occurred during transfection as the enhancer sequences were those of the pLTR-Neo plasmid, which had been cotransfected together with tumor cell DNA (Zhan et al., 1988). FGF-5-transfected cells were selected for growth in serum-free medium. It is thus likely that for FGF-5, as well, activation results from changes in the regulation of expression. This could be due to different mechanisms, as it has been reported that the FGF-5 mRNA contains a short open reading frame, upstream of that coding for the growth factor. Translation of this ORF decreases translation initiation from the AUG of FGF-5 (Bates et al., 1991). Expression of FGF-5 has been detected in several tumor cell lines (Zhan et al., 1988). The significance of this finding remains to be elucidated. FGF-5 is also a secreted protein (Bates et al., 1991), and although not much is known about its mechanism of transformation, it is quite likely that, as in the case of I(-FGF and INT-2, transformation requires secretion and activation of the receptor at the cell surface.

D. bFGF

AND

aFGF

bFGF and aFGF were not identified as oncogenes and to our knowledge they have never been isolated in a transformation assay such as NIH 3T3 transfection. This is not surprising since basic and acidic FGF cDNAs under the control of a constitutive promoter do not display a significant transforming ability in tissue culture (Sasada et al., 1988; Quarto et al., 1989; Rogelj et al., 1988; Jaye et al., 1988; Neufeld et al., 1988). As discussed, these two proteins do not contain a signal peptide and are not efficiently released from producer cells. It is therefore likely that their lack of oncogenic potential in tissue culture stems from lack of secretion. In line with the hypothesis, it has been shown that recombinant bFGF cDNAs in which a sequence encoding a signal peptide had been inserted in the 5’ region can efficiently transform cells in culture (Rogelj et al., 1988; Blam et al., 1988). A number of papers have reported that high levels of expression of native bFGF o r aFGF can confer to NIH 3T3 cells a transformed phenotype, but these cells are generally not tumorigenic in animals (Sasada et al., 1988; Rogelj et al., 1988; Jaye et al., 1988; Quarto et al., 1989). On the other hand, the studies of Wellstein et al. (1990) showed that SW-13 adrenal cells, which constitutively produce large amounts of bFGF, do not grow in agar and are not tumorigenic. Growth in agar and tumorigenicity followed transfection with K-FGF cDNA, It can probably be concluded that the oncogenic potential of bFGF and aFGF is very

FGF FAMILY OF GROWTH FACTORS A N D ONCOGENES

149

limited, and can only be manifested when very high levels of expression are achieved. A related question is whether the transformed phenotype of the rare cells transformed by bFGF and aFGF is due to small amounts of protein that these cells may release. These proteins are not detectable by biochemical methods, but experiments using blocking antibodies have in some cases demonstrated a reversion of the transformed phenotype (Sasada et al., 1988). We consider it therefore likely that transformation by bFGF or aFGF can be achieved, when, due to very high levels of expression of these proteins, a minute amount of growth factor is released into the medium, and can thus activate cell surface receptors. An alternative hypothesis is that this low level of transformation results from the synthesis of FGF molecules that are not released into the extracellular compartments, but are targeted to the nucleus. However, experiments using mutant bFGFs capable of expressing only the N-terminally extended forms associated with nuclear localization have failed to substantiate this hypothesis (Quarto et al., 1991b), and the same was true for INT-2, in which only the secreted forms are capable of transformation (Dickson et al., 1991) (see above).

E. FGF-6 AND KGF FGF-6 expression can transform NIH 3T3 cells, and apparently also in this case secretion of the growth factor is important (Marics et al., 1989; deLapeyriere et al., 1990). Nothing is known so far about the oncogenic potential of KGF. VIII. Involvement of FGFs in Tumors FGFs are generally angiogenic, and thus the first demonstration of the possible involvement of these growth factors in tumor formation or progression was the identification of TAF (tumor angiogenesis factor; later proved to be bFGF) from several tumors (Folkman et al., 1971). Thus the role of FGFs in tumor formation has been investigated mainly from two angles, that of factors necessary to promote tumor vascularization, and as primary oncogenes. A. INT-2

AND

K-FGF

The involvement of INT-2 in mouse mammary carcinomas produced by MMTV infection is quite clear and has been demonstrated in a number of laboratories. As the name indicates, INT-2 was the second

150

CLAUD10 BASILICO A N D DAVID MOSCATELLI

gene clearly identified as a very frequent site of integration of the MMTV proviral DNA in mammary tumors (Dickson et al., 1984; Moore et al., 1986). This does not imply, however, that insertion near INT-2 is a preferential attribute of MMTV, but likely results from random integration events in widespread viral infection, and selection by tumor growth of cells in which the INT-2 gene has been activated. It is not clear whether activation of INT-2 is all that is required for mammary carcinogenesis. Mice transgenic for INT-2 under the control of the MMTV enhancerlpromoter develop a high incidence of mammary hypertrophia following pregnancy, presumably reflecting the increased hormonal stimulation of the MMTV regulatory elements (Muller et al., 1990). However, these hypertrophic manifestations generally regress, indicating that INT-2 expression alone is not sufficient for tumor formation. K-FGF has also been found to be activated by MMTV insertion in some mouse tumors (Peters et al., 1989), and a recent publication suggests that activation of both INT-2 and K-FGF is important for the metastatic ability of these tumors (Murakami et al., 1990). These and other findings have led investigators to study possible rearrangements or expression of INT-2 in human breast carcinomas. A high percentage (-20%) of these human tumors appear to carry amplifications of both the INT-2 and K-FGF genes (Tsutsumi et al., 1988; Tsuda et al., 1989; Theillet et al., 1989; Adnane et al., 1989; Ali et al., 1989) (this is not surprising, since as already discussed these genes map very close one to another), but with some exception no clear-cut evidence of their expression (at the RNA level) was found. The interpretation of this finding remains to be provided. Since in general gene amplification is not detectable unless selected for, and selection could not be expected to occur without gene expression, it could be that INT-2 and K-FGF are passive bystanders in a large amplification event that involves a yet undiscovered oncogene mapping in their vicinity on chromosome 11. Alternatively, INT-2 and K-FGF amplification could represent a past event in the evolution of the tumors, which is no longer essential for tumor growth because other genetic alterations have taken place. This possibility seems, however, quite unlikely. It should be mentioned that other cases of amplification of INT-2 together with K-FGF have been reported, particularly in squamous carcinomas of the head and neck (Tsutsumi et al., 1988; Theillet et al., 1989; Tsuda et al., 1988). Again, no clear evidence of INT-2 or K-FGF expression has been reported. In human tumors, K-FGF has been found to be expressed in a number of teratocarcinomas and germ cell tumors (T. Yoshida et al., 1988; Schofield et al., 1991),but the significance of this finding in regard

F'GF FAMILY OF GROWTH FACTORS A N D ONCOGENES

151

to the etiology of these tumors is doubtful, as K-FGF is physiologically expressed in undifferentiated embryonal carcinoma cells and in the blastocyst. Thus K-FGF expression in teratocarcinomas is likely to reflect simply their stage of differentiation, rather than being responsible for their tumorigenic phenotype. Amplification has also been found in several tumors, and is discussed above. K-FGF was isolated by transfection of Kaposi's sarcoma (KS) DNA as well as from stomach cancer DNA. As already discussed, however, it is likely that the isolation was coincidental and did not reflect the activation of this oncogene in the original tumors. No evidence has been found so far of K-FGF expression in KS biopsies (A. Friedman-Kien and C. Basilico, unpublished), and although these experiments could have been hampered by the peculiar characteristics of KS, it is unlikely that K-FGF plays an important role in KS. Similar conclusions were reached about stomach carcinomas (Tsuda et al., 1988). Future experiments will undoubtedly answer these questions more conclusively. On the other hand, K-FGF can clearly be an oncogene in mice. Evidence for this conclusion comes from experiments showing that K-FGF is one of the oncogenes that can be activated by MMTV in mouse mammary carcinomas (Peters et al., 1989),and by the occurrence of a fibrosarcoma in a transgenic mouse that, although carrying a human K-FGF transgene, did not generally express it in any tissue. T h e tumor, which arose after a long latency, clearly expressed K-FGF RNA and protein (D. Talarico and C. Basilico, unpublished). Talarico et al. (1992) have tried to gain information on the cellular targets of the K-FGF oncogene in uzvo by constructing a recombinant retrovirus expressing the K-FGF protein. This virus, when injected into newborn immunocompetent mice, originally produced no detectable pathologies, with the exception of long-latency T cell leukemias, due to the helper leukemia virus that had been coinjected to allow multiplication of the defective transforming virus. One animal, however, developed a fibrosarcoma that, when grown in tissue culture, was found to produce high titers of transforming K-FGF virus, together with helper Moloney leukemia virus. This virus was highly tumorigenic when injected into newborn immunocompetent mice, as well as in nude mice. The virus differed from the original construct, and appeared to be the result of a recombination event between the K-FGF-containing retrovirus and the helper. The new virus contained the gag, pol, and m u genes of Mo-MuLV, with the e m gene fused to the K-FGF cDNA sequences. The resulting protein is an Env-K-FGF fusion protein that maintains most of the K-FGF sequence and -300 amino acids of Env. It is secreted (presumably utilizing the Env signal peptide) and has a molecular weight

152

CLAUD10 BASILICO AND DAVID MOSCATELLI

more than double that of native K-FGF. In spite of this drastic rearrangement, this protein appears to have maintained all of the properties of native K-FGF. T h e pathologies produced by this novel virus in mice are of two types: fibrosarcomas, which generally occur at or near the site of injection, and an unusual form of meningioma, or meningeal fibrosarcoma, originating from the dura mater. These tumors, which have a multifocal (polyclonal) origin, result in massive hyperproliferation of the meninges that surround the brain and the spinal cord. They cause early hydrocephalus, which eventually kills the animal. These intriguing findings raise the possibility that meningeal cells may be exquisitely susceptible to the action of K-FGF, possibly because they express highly specific K-FGF receptors. While further investigations should clarify these issues, these findings clearly underline K-FGF oncogenic potential in vivo. How these observations can be translated into similar pathological situations in humans remains to be determined.

B. bFGF

AND

aFGF

bFGF has been implicated in several human cancers, both as angiogenic factor and as a growth factor capable of sustaining autonomous cell proliferation. bFGF RNA has been found frequently to be highly expressed in malignant melanomas (Halaban et al., 1988a). This, together with the finding that bFGF (but also other FGFs) is a potent mitogen for melanocytes (Halaban et al., 1987, 1988b), has led to the hypothesis that bFGF may be an oncogene whose activation is important in the etiology of human melanomas. Halaban et al. (1988a) measured bFGF RNA levels in melanomas and normal melanocytes and found them undetectable in normal melanocytes and detectable, albeit at low levels, in many melanomas. In addition, synthetic peptides that act as bFGF antagonists reduced the growth of melanoma cells in chemically defined medium. Addition of neutralizing antibodies against bFGF, however, had no substantial effect. On the other hand, introduction of a bFGF cDNA into normal murine melanocytes rendered them capable of growth without bFGF in tissue culture, but the cells were not tumorigenic (Dotto et al., 1989). In line with these findings, Becker et al. ( 1989) found that antisense oligodeoxynucleotides targeted against human bFGF mRNA inhibited the proliferation of melanoma cells and their ability to form colonies in soft-agar medium. All of these results point to an important role of bFGF in melanomas. T h e mechanism of activation of bFGF expression in melanomas or the question of whether extracellular release of bFGF was important was not addressed in these studies.

FGF FAMILY OF GROWTH FACTORS AND ONCOGENES

153

An elevated level of bFGF RNA has also been found in cell lines derived from Kaposi's sarcoma (Ensoli et al., 1989a,b; Werner et al., 1989a). The levels of RNA corresponding to other growth factors (e.g., IL-1) are also substantially elevated and, in addition, some KS cell lines apparently secrete an unknown heparin-binding mitogen (Ensoli et al., 1989b; Werner et al., 198913). These findings seem to indicate that KS cells have activated the expression of several cytokines and growth factors. Activation of bFGF production could be related to the high vascularization of these tumors, or could be more directly involved in their growth. On the other hand, production of bFGF is observed in many normal and tumor cell lines, and thus the high levels of expression of many growth factors may reflect the cell of origin of KS (still not conclusively identified) rather than the oncogenic potential of FGF. Furthermore, it must be pointed out that the nature of the KS cell lines has never been clearly established; in other words, it is not totally certain that they are derived from KS cells. An intriguing observation has been made by Kandel et al. (1991), studying the insurgence of dermal fibrosarcomas in mice that have been made transgenic for bovine papilloma virus. These tumors grow through definite stages of progression, which can be classified as mild fibromatosis, aggressive fibromatosis, and finally fibrosarcomas. These two latter stages are highly vascularized. Cell lines derived from the various tumor stages differ in many properties in vitro and in viuo, but in particular mild fibromatosis cell lines produced large amounts of bFGF that is cell associated, while aggressive fibromatosis and particularly fibrosarcoma lines released most of the bFGF-like activity in the culture medium. Thus transition from benign to a malignant and highly vascularized state seems to be associated with changes in the secretion potential of bFGF. The molecular mechanism by which this phenomenon occurs has not yet been elucidated, and the question of how a molecule that does not possess a signal peptide and that has been shown in many laboratories to be incapable of secretion can be converted to an efficiently secreted protein remains to be answered. Since the identification of bFGF as the secreted growth factor was mainly immunological, it is also conceivable that the secreted bFGF-like activity may not be bFGF, but perhaps a closely related but distinct new member of the FGF family. C. FGFs

AND

TUMOR ANGIOCENESIS

bFGF and other members of the FGF family have also been implicated in tumor vascularization. There is little doubt that solid tumors need angiogenesis to grow in mass, and a variety of classical experiments

154

CLAUD10 BASILICO AND DAVID MOSCATELLI

have demonstrated that tumors can release angiogenic factors (Folkman and Klagsbrun, 1987). bFGF was one of the first angiogenic factors isolated from tumors, but many other factors with angiogenic properties also are produced by neoplastic cells (Folkman and Klagsbrun, 1987). It is not yet clear which of the factors is most relevant for tumor vascularization. The hypothesis that increased expression of FGF (at the RNA or protein level) is a general characteristic of tumors has not yet been demonstrated, and similarly nothing is known about the mechanism of this activation, if it exists. If FGFs are involved in tumor angiogenesis, in the case of bFGF and aFGF the problem of their inefficient release from producer cells remains to be solved (see, however, the results of Kandel et al.,1991, discussed above). FGFs accumulate in the ECM, and perhaps tumor cells possess the ability of mobilizing these growth factors from their storage sites, rather than directly releasing FGFs produced by the cells themselves. Since endothelial cells also produce bFGF, it has been suggested (Schweigerer et d.,1987a) that tumors may secrete a factor that increases bFGF production in endothelial cells, leading to autocrine activation of these cells. Thus, the assessment of the precise involvement of FGFs in tumor vascularization will require further experiments. The possibility that FGFs may be essential for solid tumor growth, and in general the recognition that angiogenesis is necessary for the increase in tumor mass, has prompted several investigators to study whether substances that are inhibitors of vascularization, or are specific antagonists of growth factors, could be used in cancer therapy. Suramin, a polyanion that strongly inhibits the interaction of FGFs with their receptors (Yayon and Klagsbrun, 1990; Wellstein et al., 1991), has been used in clinical trials on cancer patients and in some cases appears to have had remarkable effects on tumor regression (LaRocca et al., 1991; Walz et al., 1991). Suramin, however, binds to a variety of proteins, including other growth factors, and thus could block autocrine growth loops as well as induction of angiogenesis (Yayon and Klagsbrun 1990; Wellstein et al., 1991; Kim et al., 1991). The elucidation of the respective role of various growth factors in tumor vascularization will certainly be made easier by experiments using specific growth factor antagonists that are being developed at a fast pace. In conclusion, establishing the precise role of FGFs in tumor formation will require further data. This is not very surprising, since the oncogenic potential of FGFs has been discovered quite recently, and even in the case of oncogenes that have been studied for a longer period of time (e.g., the ras family) it is not yet clear whether their activation is specific for certain tumors or is just another factor in an ill-defined

FGF FAMILY OF GROWTH FACTORS AND ONCOGENES

155

multistep pathway to malignancy. Although it is clear that at least the secreted members of the FGF family have oncogenic potential, a critical role in in uivo carcinogenesis has only been established for INT-2 in mouse mammary tumors. T h e probability that a protooncogene becomes a relevant oncogene is determined by a variety of factors, including the likelihood of its activation in cells that respond to this oncogene, and the probability that other important steps in carcinogenesis, such as the inactivation of a tumor suppressor gene, occur in the same cells. Finally, the host immune response to the phenotype of the cell carrying the activated oncogene may well determine whether such cells will produce a tumor or not. It is quite likely in our view that FGF activation will turn out to be one of the many factors contributing to the malignancy of many solid tumors, but whether specific tumors will turn out to involve FGF activation more than others remains to be determined. IX. Concluding Remarks To review a field of research that has undergone a recent “explosion” is fraught with difficulties. Since it would have been impossible to cover all that has been published in the field, we have been trying to concentrate on the aspects of those growth factors that we felt were potentially of higher interest for future development. Without any doubt this will reveal our biases, and we would not be surprised if a number of observations that we overlooked would reveal themselves to have been of great importance in the future years. T h e FGF field offers at present a fascinating series of questions to scientists interested in many diverse areas, far and beyond the interest in growth factor action on cell proliferation. Students of regulation of gene expression can find all possible mechanisms operating on the expression of FGFs and their receptors: transcriptional controls, posttranscriptional regulation involving alternative splicing, alternative translation starts resulting in proteins with different properties, and control affecting the secretion of these proteins. Students of development will undoubtedly be attracted by the strong evidence that these growth factors play a role in development, although the precise role is not yet totally clear. Students of angiogenesis and oncogenesis will no doubt be interested in the possible role of FGFs in physiological and pathological angiogenesis, as well as in their oncogenic potential. Finally, the existence of a family of growth factors as well as a family of receptors whose specificity of interaction is only beginning to be elucidated should be of interest to students of signal transduction and of the mechanisms by which growth factors influence cell proliferation, survival, and differentiation. It will

156

CLAUD10 BASILICO AND DAVID MOSCATELLI

be quite apparent from reading this article that a great many questions in this field remain unanswered. We do not know the exact physiological function of FGFs, we do not know whether their involvement in oncogenesis is only potential or real, we do not know why there are seven growth factors (so far) with an apparently similar spectrum of action. The available data, however, clearly support a number of verifiable hypotheses and thus we are confident that most of these questions will be answered in the years to come.

REFERENCES Aaronson, S., Abraham, J., Baird, A., Basilico, C., Birnbaum, D., Bohlen, P., Burgess, W., Dickson, C., Fiddes, J., Goldfarb, M., Gospodarowicz, D., Klagsbrun, M., Maciag, T., Martin, G., Peters, G., Rubin, J., Thomas, K., Terada, M., and Yoshida, T. (1991). Ann. N.Y. Acad. Sci. 638, 13-16. Abraham, J. A., Mergia, A., Whang, J. L., Tumolo, A., Friedman, J., H,jerrild, K. A,, Gospodarowicz, D., and Fiddes, J. C. (1986a). Science 233, 545-548. Abraham, J. A., Whang, J. L., Tumolo, A., Mergia, A., Friedman, J., Gospodarowicz, D., and Fiddles, J. C. (1986b). EMBO J. 5 , 2523-2528. Acland, P., Dixon, M., Peters, G., and Dickson, C. (1990). Nature (London) 343, 662-665. Adnane, J., Gaudray, P., Simon, M.P., Simony-Lafontaine, J., Jeanteur, P., and Theillet, C. (1989). Oncogene 4, 1389-1395. Ah, 1. U., Merlo, G., Callahan, R., and Lidereau, R. (1989). Oncogene 4, 89-92. Alterio, J., Halley, C., Brou, C., Soussi, T., Courtois, Y., and Laurent, M. (1988). FEBS Ldt. 242, 4 1-46. Amaya, E., Musci, T. J., and Kirschner, M. W. (1991). Cell (Cambridge, Mass.) 66, 257-270. Anderson, K. J., Dam, D., Lee, S., and Cotrnan, C. W. (1988). Nature (London) 332, 360361. Arakawa, T., Hsu, Y.-R., Schiffer, S. G., Tsai, L. B., Curless, C., and Fox, G. M. (1989). Biochem. Biophys. Res. Commun. 161, 335-341. Armelin, H. A. (1973). Proc. Natl. Acad. Sci. U.S.A. 70, 2702-2706. Ausprunk, D. H., and Folkman, J. (1977). Mzcrouasc. Res. 14, 53-65. Baird, A,, and Bohlen, P. (1990). In ”Handbook of Experimental Pharmacology” (M. B. Sporn and A. B. Roberts, eds.), Vol. 95, pp. 369-418. Springer-Verlag, Berlin. Baird, A,, and Ling, N. (1987). Biochem. Bzophy. Res. Commun. 142, 428-435. Baird, A,, Esch, F., Gospodarowicz, D., and Guillemin, R. (1985). Bzochemistv 24, 78557860. Baird, A., Esch, F., Mormede, P., Ueno, N., Ling, N., Bohlen, P., Ying, S.-Y., Wehrenberg, W. B., and Guillemin, R. (1986). Recent Prog. Horm. Res. 42, 143-205. Baird, A,, Ueno, N., Esch, F., and Ling, N. (1987). J . Cell. Phys. Suppl. 5, 101-106. Baird, A., Schubert, D., Ling, N., and Guillemin, R.(1988). Proc. Nutl. Acad. Sci. U.S.A. 85, 2324-2328. Barr, P. J., Cousens, L. S., Lee-Ng, C. T., Medina-Selby, A., Masiarz, F. R., Hallewell, R. A., Chamberlain, S. H., Bradley, J. D., Lee, D., Steimer, K. S., Poulter, L., Burlingame, A. L.,Esch, F., and Baird, A. (1988). J. Biol. Chem. 263, 16471-16478. Bashkin, P., Doctrow, S., Klagsbrun, M., Svahn, C. M., Folkman, J., and Vlodavsky, I. (1989). Biochemktty 28, 1737-1743.

FGF FAMILY OF GROWTH FACTORS A N D ONCOGENES

157

Bates, B., Hardin, J., Zhan, X., Drickamer, K., and Goldfarb, M. (1991).Mol. Cell. Biol. 11, 1840-1845. Becker, D., Meier, C. B., and Herlyn, M. (1989). E M B O J 8, 3685-3691. Bellot, F., Crurnley, G., Kaplow, J. M., Schlessinger, J., Jaye, M., and Dionne, C. A. (1991). EMBO J. 10, 2849-2854. Blam, S. B., Mitchell, R., Tischer, E., Rubin, J. S., Silva, M., Silver, S., Fiddes, J. C., Abraham, J. A,, and Aaronson, S. A. (1988). Oncogene 3, 129-136. Bohlen, P., Baird, A,, Esch, F., Ling, N., and Gospodarowicz, D. (1984). Proc. Nutl. Acud. Sci. L’.S.A. 81, 5364-5368. Bohlen, P., Esch, F., Baird, A,, Jones, K. L., and Gospodarowicz, D. (1985).FEBS Lett. 185, 177-181. Bouche, G., Gas, N., Prats, H., Baldin, V., Tauber, J.-P., Teissie, J., and Amalrick, F. (1987). Proc. Natl. h a d . Sci. U.S.A. 84, 6770-6774. Boyer, B., Tucker, G. C., Valles, A. M., Franke, W. W., and Thiery, J. P. (1989).J. Cell Biol. 109, 1495-1509. Brigstock, D. R., Klagsbrun, M., Sasse, J., Farber, P. A., and Iberg, N. (1990).GrowthFuctors 4, 45-52. Broadley, K. N., Aquino, A. M., Woodward, S. C., Buckely-Sturrock, A,, Sato, Y., Rifkin, D. B., and Davidson,J. M. (1989). Lab. Invest. 61, 571-575. Brookes, S., Smith, R., Casey, G., Dickson, C., and Peters, G. (1989a). Oncogene 4,429-436. Brookes, S., Smith, R., Thurlow, J., Dickson, C., and Peters, G. (198913).Nucleic Acids Reg. 17,4037-4045. Bugler, B., Amalric, E., and Prats, H. (1991). Mol. Cell. Biol. 11, 573-577. Burgess, W.H., and Maciag, -1.. (1Y89). Annu. Rev. Biochem. 58, 575-606. Burgess, W.H., Mehlman, T., Marshak, D. R., Fraser, B. A., and Maciag, T. (1986). Proc. Natl. Acad. Sci. U.S.A. 83, 7216-7220. Burgess, W.H., Shaheen, A. M., Ravera, M., Jaye, M., Donohue, P. J., and Winkles, J. A. (1990a).J. Cell Biol. 111, 2129-2138. Burgess, W. H., Dionne, C. A., Kaplow, J., Mudd, R., Friesel, R., Zilberstein, A., Schlessinger, J., and Jaye, M. (1990b). Mol. Cell. Biol. 10, 4770-4777. Burgess, W. H., Bizik, J., Mehlman, T., Quarto, N., and Rifkin, D. B. (1991). Cell Kegul. 2, 87-93. Casscells, W., Speir, E., Sasse, J., Klagsbrun, M., Allen, P., Lee, M., Calvo, B., Chiba, M., Haggroth, L., Folkman, J.. and Epstein, S. E. (1990).J . Clin. Invest. 85, 433-441. Chiu, I.-M., Wang, W.-P., and Lehtoma, K. (1990). Oncogene 5, 755-762. Clegg, C. H., Linkhart, T. A., Olwin, B. B., and Hauschka, S. D. (1987).J. Cell Biol. 105, 949-956. Connolly, D. T., Stoddard, B. L., Harakas, N. K., and Feder,J. (1987).Biochem. Bzophys. Res. Commun. 144, 705-712. Cordon-Cardo, C., Vlodavsky, I., Haimovitz-Friedman, A,, Hicklin, D., and Fuks, Z. (1990).Lab. Invest. 63, 832-840. Coughlin, S. R., Barr, P. J.. Cousens, L. S., Fretto, L. J., and Williams, L. T. (1988).J. Biol. Chem. 263, 988-993. Crabb, J. W., Armes, L. G., Carr, S. A., Johnson, C. M., Roberts, G. D., Bordoli, R. S., and McKeehan, W. L. (1986). Biochemistry 25, 4988-4993. Crumley, G., Dionne, C. A., and Jaye, M. (1990).Biochem. Biop/tys. Res. Commun. 171,7-13. Curatola, A. M., and Basilico, C. (1990). Mol. Cell B i d . 10, 2475-2484. DAmore, P. A., and Klagsbrun, M. (1984).J. CellBiol. 99, 1545-1549. Davidson, J. M., Klagsbrun, M., Hill, K. E., Buckley, A., Sullivan, R., Brewer, P. S., and Woodward, S. C. (1985).J. Cell Biol. 100, 1219-1227.

158

CLAUD10 BASILICO AND DAVID MOSCATELLI

delapeyriere, O., Rosnet, O., Benharroch, D., Raybaud, F., Marchetto, S., Planche, J., Galland, F., Mattei, M.-G., Copeland, N. G., Jenkins, N. A., Coulier, F., and Birnbaum, D. (1990). Oncogene 5, 823-83 1. Delli Bovi, P., and Basilico, C. (1987). Proc. Nutl. Acud. Sci. U.S.A. 84, 5660-5664. Delli Bovi, P., Curatola, A. M., Kern, F. G., Greco, A., Ittmann, M., and Basilico, C. (1987). Cell (Cambridge, Mass.) 50, 729-737. Delli Bovi, P., Curatola, A. M., Newman, K. M., Sato, Y., Moscatelli, D., Hewick, R. M., Rifkin, D. B., and Basilico, C. (1988). Mol. Cell. Biol. 8, 2933-2941. Dickson, C., and Peters, G. (1987). Nature (London) 326, 833. Dickson, C., Smith, R., Brookes, S., and Peters, G. (1984). Cell (Cambridge, Mass.) 37,529536. Dickson, C., Fuller-Pace, F., Kiefer, P., Acland, P., MacAllen, D., and Peters, G. (1991).Ann. N.Y. Acad. Sci. 638, 18-26. Dionne, C. A., Crumley, G., Bellot, F., Kaplow, J. M., Searfoss, G., Ruta, M., Burgess, W. H., Jaye, M., and Schlessinger, J. (1990). EMBO J. 9, 2685-2692. Dixon, M., Deed, R., Acland, P., Moore, R.,Whyte, A., Peters, G., and Dickson, C. (1989). Mol. Cell. B i d . 9, 4896-4902. Dotto, G. P., Moellmann, G., Ghosh, S., Edwards, M., and Halaban, R. (1989).J. Cell Bzol. 109, 3115-3128. Emoto, N., Gonzalez, A.-M., Walicke, P. A., Wada, E., Simmons, D. M., Shimasaki, S., and Baird, A. (1989). Growth Factors 2, 21-29. Ensoli, B., Nakamura, S., Salahuddin, S. Z., Biberfeld, P., Larson, L., Beaver, B., WongStaal, F., and Gallo, R. C. (1989a). Science 243, 223-226. Ensoli, B., Salahuddin, S. Z., and Gallo, R. C. (1989b). “Cancer Cells,” No. 1042; pp. 93-96. Cold Spring Harbor Lab., Cold Spring Harbor, New York. Eriksson, A. E., Cousens, L. S., Weaver, L. H., and Matthews, B. W. (1991).Proc. Nutl. Acad. Sci. U.S.A. 88, 3441-3445. Esch, F., Baird, A., Ling, N., Ueno, N., Hill, F., Denoroy, L., Klepper, R., Gospodarowicz, D., Bohlen, P., and Guillemin, R. (1985a). Proc. Natl. Acud. Sci. U.S.A. 82, 6507-651 1. Esch, F., Ueno, N., Baird, A., Hill, F., Denoroy, L., Ling, N., Gospodarowicz, D., and Guillemin, R. (1985b). Biochem. Biophys. Res. Commun. 133, 554-562. Feige, 3.-J., and Baird, A. (1989). Proc. Nutl. Acad. Scz. U.S.A. 86, 3174-3178. Finch, P. W., Rubin, J. S., Miki, T., Ron, D., and Aaronson, S. A. (1989). Science 245, 752755. Finklestein, S. P., Apostolides, P. J., Caday, C. G., Prosser,J., Philips, M. F., and Klagsbrun, M. (1988). Bruin Res. 460, 253-259. Flaumenhaft, R., Moscatelli, D., Saksela, O., and Rifkin, D. B. (1989).J. Cell. Physiol. 140, 75-81. Flaumenhaft, R., Moscatelli, D., and Rifkin, D. B. (1990).J. Cell B i d . 111, 1651-1659. Florkiewicz, R. Z., and Sommer, A. (1989). Proc. Natl. Acud. Scz. U.S.A. 86, 3978-3981. Florkiewicz, R. Z., Baird, A., and Gonzalez, A.-M. (1991). Growth Factors 4, 265-275. Folkman, J., and Klagsbrun, M. (1987). Science 235, 442-447. Folkman, J., Merler, E., Abernathy, C., and Williams, G. (1971).J. Exp. Med. 133,275-288. Folkman, J., Klagsbrun, M., Sasse, J., Wadzinski, M., Ingber, D., and Vlodavsky, 1. (1988). Am. J . Pathol. 130, 393-400. Fox, G. M., Schiffer, S. G., Rohde, M. F., Tsai, L. B., Banks, A. R.,and Arakawa, T. (1988). J. B i d . Chem. 263, 18452-18458. Fukushima, Y., Byers, M. G., Fiddes, J. C., and Shows, T. B. (1990). Cytogenet. Cell Genet. 54, 159- 160.

FGF FAMILY OF GROWTH FACTORS AND ONCOGENES

159

Gabbianelli, M., Sargiacomo, M., Pelosi, E., Testa, U., Isacchi, G., and Peschle, C. (1990). Science 249, 1561- 1564. Gautschi-Sova, P., Jiang, Z., Frater-Schroder, M., and Bohlen, P. (1987). Biochemistry 26, 5844-5847. Gillespie, L. L., Paterno, G. D., and Slack, J. M. W. (1989). Development (Cambridge, UK) 106, 203-208. Gimenez-Gallego, G., Rodkey, J., Bennett, C., Rios-Candelore, M., DiSalvo,J., and Thomas, K. (1985). Science 230, 138551388, Gimenez-Gallego, G., Conn, G., Hatcher, V. B., and Thomas, K. A. (1986). Biochem. Biophys. Res. Commun. 135, 541-548. Glazer, L., and Shilo, B.-Z. (1991). Genes Dev. 5, 697-705. Goldfarb, M., Deed, R., MacAllan, D., Walther, W., Dickson, C., and Peters, G. (1991). Oncogene 6 , 65-7 1. Gonzalez, A.-M., Buscaglia, M., Ong, M., and Baird, A. (1990).J. Cell Biol. 110, 753-765. Gordon, P. B., Choi, H. U., Conn, G., Ahmed, A., Ehrmann, B., Rosenberg, L., and Hatcher, C: B. (1989).J . Cell. Physiol. 140, 584-592. Gospodarowicz, D. (1974). Nature (London) 249, 123-127. Gospodarowicz, D. (1975).J. Biol. Chem. 250, 2515-2520. Gospodarowicz, D., and Cheng, J. (1986).J. Cell. Physiol. 128, 475-484. Gospodarowicz, D., Bialecki, H., and Greenburg, G. (1978a). J. B i d . Chem. 253, 37363743. Gospodarowicz, D., Greenburg, G., Bialecki, H., and Zetter, B. R. (1978b). In Vitro 14,85118. Gospodarowicz, D., Cheng, J., Lui, G. M., Baird, A., and Bohlen, P. (1984).Proc. Natl. Acad. SCZ. U.S.A. 81, 6963-6967. Gospodarowicz, D., Cheng, J., Lui, G. M., Baird, A., Esch, F., and Bohlen, P. (1985). Endocrinology (Baltimore) 117, 2382-2391. Gospodarowicz, D., Baird, A., Cheng, J., Lui, G. M., Esch, F., and Bohlen, P. (1986). Endocrinology (Baltimore) 118, 82-90. Gospodarowicz, D., Ferrara, N., Schweigerer, L., and Neufeld, G. (1987). Endocr. Rev. 8, 95-1 14. Gospodarowicz, D., Ferrara, N., Haaparanta, T., and Neufeld, G. (1988). Eu7.J. Cell Biol. 46, 144-151. Grinberg, D., Thurlow, J., Watson, R., Smith, R., Peters, G., and Dickson, C. (1991). Cell Growth Dafft.7. 2, 137-143. Grunz, H., McKeehan, W. L., Knochel, W., Born, J.. Tiedemann, H., and Tiedemann, H. (1988). Cell Dzffer. 22, 183- 190. Halaban, R., Ghosh, S., and Baird, A. (1987). In Vitro Cell. Den. Biol. 23, 47-52. Halaban, R., Kwon, B. S., Ghosh, S., Delli Bovi, P., and Baird, A. (1988a). Oncogene Res. 3, 177- 186. Halaban, R., Langdon, R., Birchall, N., Cuono, C., Baird, A., Scott, G., Moellmann, G., and McGuire, J. (1988b).J. Cell Biol. 107, 1611-1619. Hanneken, A., Lutty, G. A., McLeod, D. S., Robey, F., Harvey, A. K., and Hjelmeland, L. M. (1989).J. Cell. Physiol. 138, 115- 120. Harper, J. W., and Lobb, R. R. (1988). Biochemist? 27, 671-678. Harper, J. W., Strydom, D. J., and Lobb, R. R. (1986). Biochemistry 25, 4097-4103. Hatten, M. E., Lynch, M., Rydel, R. E., Sanchez, J., Joseph-Silverstein, J., Moscatelli, D., and Rifkin, D. B. (1988). Dpu. Biol. 125, 280-289. Hattori, Y., Odagiri, H., Nakatani, H., Miyagawa, K., Naito, K., Sakamoto, H., Katoh, O.,

160

CLAUDIO BASILICO AND n A v m MOSCATELLI

Yoshida, T., Sugimura, T., and Terada, M. (1990).Proc. Natl. Acad. Sci. U.S.A. 87,59835987. Haub, O., and Goldfarb, M. (1991). Deuelupment (Cambridge, U K ) 112, 397-406. Haub, O., Drucker, B., and Goldfarb, M. (1990). Proc. Natl. Acad. Sci. U.S.A. 87, 80228026. Hayek, A., Culler, F. L., Beattie, G. M., Lopez, A. D., Cuevas, P., and Baird, A. (1987). Biochem. Biophys. Res. Commun. 147, 876-880. Hebert, J. M., Basilico, C., Goldfarb, M., Haub, O., and Martin, G. R. (1990).Deu. Biol. 138, 4.54-463. Hou, J., Kan, M., McKeehan, K., McBride, G., Adams, P., and McKeehan, W. L. (1991). Science 251, 665-668. Huebner, K., Ferrari, A. C., Delli Bovi, P., Croce, C. M., and Basilico, C. (1988). Oncogene Res. 3, 263-270. Iberg, N., Rogelj, S., Fanning, P., and Klagsbrun, M. (1989).J. Biol. Chem. 264, 1995119955. Imamura, T., Engleka, K., Zhan, X., Tokita, Y., Forough, R., Roeder, D., Jackson, A,, Maier, J. A. M., Hla, T., and Maciag, T. (1990). Science 249, 1567-1570. Isacchi, A., Statuto, M., Chiesa, R., Bergonzoni, L., Rusnati, M., Sarminetos, P., Ragnotti, G., and Presta, M. (1991). Proc. Natl. Acad. Sci. U.S.A. 88, 2628-2632. Jakobovits, A., Shackleford, G. M., Varmus, H. E., and Martin, G. R., (1986). Proc. Natl. Acad. Sci. U.S.A. 83, 7806-7810. Jaye, M., Howk, R., Burgess, W., Ricca, G., Chiu, I.-M., Ravera, M. W., O’Brien, S. J., Modi, W. S., Maciag, T., and Drohan, W. N. (1986). Science 233, 541-545. Jaye, M., Burgess, W. H., Shaw, A. B., and Drohan, W. N. (1987).J.Biol. Chem. 262,1661216617. Jaye, M., Lyall, R. M., Mudd, R., Schlessinger, J., and Sarver, N. (1988).E M B O J . 7, 963969. Jeanny, J.-C., Fayein, N., Moenner, M., Chevallier, B., Barritault, D., and Coutois, Y. (1987). Exp. Cell Res. 171, 63-75. Johnson, D. E., Lee, P. L., Lu, J., and Williams, L. T. (1990).Mol. Cell. Biol. 10,4728-4736. Joseph-Silverstein, J., Consigli, S. A., Lyser, K. M., and Ver Pault, C. (1989).J.CellBiol. 108, 2459-2466. Kalcheim, C., and Neufeld, G. (1990). Development (Cambrulge, U K ) 109, 203-215. Kan, M., DiSorbo, D., Hou, J., Hoshi, H., Mansson, P.-E., and McKeehan, W. L. (1988).J. Biol. Chem. 263, 1 1306- 1 13 13. Kandel, J., Bossy-Wetzel, E., Radvanyl, F., Klagsbrun, M., Folkman, J., and Hanahan, D. (1991). Cell (Carnbndge, Mass.) 66, 1-20. Kaplow, J. M., Bellot, F., Crumley, G., Dionne, C. A., and Jaye, M. (1990).Biochem. Biophys. Res. Commun. 172, 107-1 12. Keating, M. T., and Williams, L. T. (1988). Science 239, 914-916. Keegan, K., Johnson, D. E., Williams, L. T., and Hayman, M. J. (1991). Proc. Natl. Acad. Sci. U.S.A. 88, 1095-1099. Kiefer, M. C., Stephans, J. C., Crawford, K., Okino, K., and Barr, P. J. (1990).Proc. Natl. Acad. Sci. U.S.A. 87, 6985-6989. Kim, J. H., Sherwood, E. R., Sutkowski, D. M., Lee, C., and Kozlowski,J. M. (1991).J.Urol. 146, 171-176. Kimelman, D., and Kirschner, M. W. (1989) Cell (Cambridge, Mass.)59,687-696. Kimelman, D., Abraham, J. A., Haaparanta, T., Palisi, T. M., and Kirschner, M. W. (1988). Science 242, 1053-1056. Klagsbrun, M., and Shing, Y. (1985). Proc. Natl. Acad. Scz. U.S.A. 82, 805-809.

FGF FAMILY OF GROWTH FACTORS AND ONCOGENES

161

Klagsbrun, M., Sasse, J., Sullivan, R., and Smith, J. A. (1986). Proc. Natl. Acad. Sci. U.S.A. 83,2448-2452. Klagsbrun, M., Smith, S., Sullivan, R., Shing, Y., Davidson, S . , Smith, J. A., and Sasse, J. (1987). Proc. Natl. Acad. Sci. U.S.A. 84, 1839-1843. Kornbluth, S., Paulson, K. E., and Hanafusa, H. (1988). Mol. Cell. Biol. 8, 5541-5544. Kurokawa, M., Doctrow, S. R., and Klagsbrun, M. (1989).J. Biol. Chem. 264, 7686-7691. Kurokawa, T., Seno, M., and Igarashi, K. (1988). Nucleic Acids Res. 16, 5201. LaRocca, R. V., Danesi, R., Cooper, M. R., Jamis-Dow, C. A., Ewing, M. W., Linehan, W. M., and Myers, C. E. (1991).J. Urology 145, 393-398. Lee, P. L., Johnson, D. E., Cousens, L. S., Fried, V. A., and Williams, L. T. (1989).Science 245, 57-60. Linemeyer, D. L., Kelly, L. J., Menke, J. G., Gimenez-Gallego, G., DiSalvo,J., and Thomas, K. A. (1987). BaolTechnoloby 5, 960-965. Linemeyer, D. L., Menke, J. G., Kelly, L. J., DiSalvo, J., Soderman, D., Schaeffer, M.-T., Ortega, S., Gimenez-Gallego, G., and Thomas, K. A. (1990). Growth Factors 3,287-298. Lobb, R. R., Alderman, E. M., and Fett, J. W. (1985). Biochmistry 19, 4969-4973. Maciag, T., Cerundolo, J., Ilsley, S . , Kelley, P. R., and Forand, R. (1979). Proc. Natl. Acad. Sci. U.S.A. 76, 5674-5678. Maciag, T., Mchlman, T., Friesel, R., and Schreiber, A. B. (1984). Science 225, 932-935. Mansour, S. L., and Martin, G. R. (1988). E M B O J . 7, 2035-2041. Mansukhani, A., Moscatelli, D., Talarico, D., Levytska, V., and Basilico, C. (1990). Proc. Natl. Acad. Sci. U.S.A. 87, 4378-4382. Mansukhani, A., Dell’Era, P., Moscatelli, P., Kornbluth, S . , Hanafusa, H., and Basilico, C. (1992). Pror. Natl. Acad. Sci. U.S.A. (in press). Marics, I., Adelaide, J., Raybaud, F., Mattei, M., Coulier, F., Planche, J., DeLapeyriere, O., and Birnbaum, D. (1989). Oncogene 4, 335-340. Matsuzaki; K., Yoshitake, Y., Matuo, Y., Sasaki, H., and Nishikawa, K. (1989).Proc. ,Vatatl. A c ~ d Sci. . U.S.A. 86, 991 1-9915. McGee, G. S., Davidson, J. M., Buckley, A,, Sommer, A., Wooclward, S. C., Aquino, A. M., Barbour, R., and Demetriou, A. A. (1988).J. Surg. Res. 45, 145-153. McNeil, P. L., Muthukrishnan, L., Warder, E., and D’Amore, P. A. (1989).J. Cell Bzol. 109, 8 1 1-822. Mehlman, T., and Burgess, W. H. (1990). Anal. Biochem. 188, 159-163. Mignatti, P., Tsuboi, R., Robbins, E., and Rifkin, D. B. (1989).J . Cell Biol. 108, 671-682. Mignatti, P., Morimoto, T., and Rifkin, D. B. (1991). Proc. Natl. Acad. Sci. U.S.A. 88, 1100711011. Miki, T., Fleming, T. P., Bottaro, D. P., Rubin, J. S., Ron, D., and Aaronson, S. A. (1991). Science 251, 72-75. Moenner, M., Chevallier, B., Badet, J., and Barritault, D. (1986). Proc. Natl. Acad. Sci. U.S.A. 83, 5024-5028. Moore, J. W., Dionne, C., Jaye, M., and Swain, J. L. (1991). Deuelofiment (Cambridge, U K ) 111, 741-748. Moore, R., Casey, G., Brookes, S., Dixon, M., Peters, G., and Dickson, C. (1986). EMBOJ. 5, 919-924. Morrison, R. S., Sharma, A., deVellis, J., and Bradshaw, R. A. (1986). Pmc. Natl. Acad. Sci. U.S.A. 83, 7537-7541. Moscatelli, D. (1987). J. Cell. Physiol. 131, 123-130. Moscatelli, D. (1988).J. Cell Biol. 107, 753-759. Moscatelli, D., Presta, M., Joseph-Silverstein, J., and Rifkin, D. B. (1986a).J. Cell. PhYSZOl. 129, 273-276.

162

CLAUD10 BASILICO A N D DAVID MOSCATELLI

Moscatelli, D., Presta, M., and Rifkin, D. B. (1986b). Proc. Natl. Acad. Sci. U.S.A. 83, 20912095. Moscatelli, D., Joseph-Silverstein, J., Manejias, R., and Rifkin, D. B. (1987).Proc. Natl. Acad. Sci. U.S.A. 84, 5778-5782. Mueller, S. N., Thomas, K. A., Di Salvo, J., and Levine, E. M. (1989).J. Cell. Physiol. 140, 439-448. Muller, W. J., Lee, F. S., Dickson, C., Peters, G., Pattengale, P., and Leder, P. (1990).EMBO J. 9, 907-913. Murakami, A., Tanaka, H., and Matsuzawa, A. (1990). Cell Growth Differ. 1, 225-231. Musci, T.J., Amaya, E., and Kirschner, M. W. (1990). Proc. Natl. Acad. Sci. U.S.A. 87,83658369. Neufeld, G., Mitchell, R., Ponte, P., and Gospodarowicz, D. (1988).J. Cell Biol. 106, 13851394. Nguyen, C., Roux, D., Mattei, M.-G., delapeyriere, O., Goldfarb, M., Birnbaum, D., and Jordan, B. R. (1988). Oncogene 3, 703-708. Niswander, L., and Martin, G. R. (1992). Development (in press). Olwin, B. B., and Hauschka, S. D. (1988).J. Cell Biol. 107, 761-769. Ortega, S., Schaeffer, M.-T., Soderman, D., DiSalvo, J., Linemeyer, D. L., GimenezGallego, G., and Thomas, K. A. (1991).J. Biol. Chem. 266, 5842-5846. Partanen, J., Makela, T. P., Eerola, E., Korhonen, J., Hirvonen, H., Claesson-Welsh, L., and Alitalo, K. (1991). E M B O J . 10, 1347-1354. Pasquale, E. B. (1990). Proc. Natl. Acad. Sci. U.S.A. 87, 5812-5816. Pasquale, E. B., and Singer, S. J. (1989). Proc. Natl. Acad. Sci. U.S.A. 86, 5449-5453. Paterno, G. D., Gillespie, L. L., Dixon, M. S., Slack, J. M. W., and Heath, J. K. (1989). Development (Cambridge, U K ) 106, 79-83. Peters, G., Brookes, S., Smith, R., Placzek, M., and Dickson, C. (1989).Proc. Natl. Acad. Sci. U.S.A. 86, 5678-5682. Pettmann, B., Labourdette, G., Weibel, M., and Sensenbrenner, M. (1986). Neurosci. Lett. 68, 175-180. Prats, H., Kaghad, M., Prats, A. C., Klagsbrun, M., Lelias, J. M., Liauzun, P., Chalon, P., Tauber, J. P., Amalric, F., Smith, J. A., and Caput, D. (1989). Proc. Natl. Acad. Sci. U.S.A. 86, 1836- 1840. Presta, M., Moscatelli, D., Joseph-Silverstein, J., and Rifkin, D. B. (1986).Mol. Cell. Biol. 6, 4060-4066. Presta, M., Rusnati, M.,Maier, J. A. M., and Ragnotti, G. (1988). Biochem. Biophys. Res. Commun. 155, 1161-1 172. Quarto, N., Talarico, D., Sommer, A., Florkiewicz, R., Basilico, C., and Rifkin, D. B. (1989). Oncogene Res. 5 , 10 1 - 1 10. Quarto, N., Finger, F. P., and Rifkin, D. B. (1991a).J. Cell. Physiol. 147, 31 1-318. Quarto, N., Talarico, D., Florkiewicz, R., and Rifkin, D. B. (1991b). Cell Regul. 2,699-708. Quinkler, W., Maasberg, M., Bernotat-Danielowski, S., Luthe, N., Sharma, H. S., and Schaper, W. (1989). Eur. J. Biochem. 181, 67-73. Rapraeger, A. C., Krufka, A,, and Olwin, B. B. (1991). Science 252, 1705-1708. Renko, M., Quarto, N., Morimoto, T., and Rifkin, D. B. (199O).J. Cell. Physiol. 144, 108114. Rifkin, D. B., and Moscatelli, D. (1989).J. Cell Biol. 109, 1-6. Rogelj, S., Weinberg, R. A., Fanning, P. and Klagsbrun, M. (1988). Nature (London) 331, 173-175. Rosengart, T. K., Johnson, W. V., Friesel, R., Clark, R., and Maciag, T. (1988). Biochem. Biophys. Res. Commun. 152, 432-440.

FGF FAMILY OF GROWTH FACTORS A N D ONCOGENES

163

Rubin, J. S., Osada, H., Finch, P. W., Taylor, W. G., Rudikoff, S., and Aaronson, S. A. (1989). Proc. Natl. Acad. Sci. U.S.A. 86, 802-806. Ruta, M., Howk, R., Ricca, G., Drohan, W., Zabelshansky, M., Laureys, G., Barton, D. E., Francke, U., Schlessinger, J., and Givol, D. (1988). Oncogene 3, 9-15. Ruta, M., Burgess, W., Givol, D., Epstein, J., Neiger, N., Kaplow, J., Crumley, G., Dionne, C., Jaye, M., and Schlessinger, J. (1989). Proc. Natl. Acad. Sci. U.S.A. 86, 8722-8726. Rydel, R. E., and Greene, L. A. (1987).J. Neurosci. 7, 3639-3653. Safran, A,, Avivi, A,, Orr-Urtereger, A., Neufeld, G., Lonai, P., Givol, D., and Yarden, Y. (1990). Oncogene 5 , 635-643. Sakaguchi, M., Kajio, T., Kawahara, K., and Kato, K. (1988). FEES Lett. 233, 163-166. Sakamoto, H., Mori, M., Taira, M., Yoshida, T., Matsukawa, S., Shimizu, K., Sekiguchi, M., Terada, M., and Sugimura, T. (1986). Proc. Natl. Acad. Sci. U.S.A. 83, 3997-4001. Sakamoto, H., Yoshida, T., Nakakuki, M., Odagiri, H., Miyagawa, K., Sugimura, T., and Terada, M. (1988). Biochem. Biophys. Res. Commun. 151, 965-972. Saksela, O., and Rifkin, D. B. (1990).J. Cell Biol. 110, 767-775. Saksela, O., Moscatelli, D., Sommer, A,, and Rifkin, D. B. (1988).J. Cell Biol. 107,743-75 1. Sano, H., Forough, R., Maier, J. A. M., Case, J. P., Jackson, A., Engleka, K., Maciag, T., and Wilder, R. L. (199O).J. Cell B i d . 110, 1417-1426. Sasada, R., Kurokawa, T., Iwane, M., and Igarashi, K. (1988). Mol. Cell. Biol. 8 , 588-594. Sasaki, H., Hoshi, H., Hong, Y.-M., Suzuki, T., Kato, T., Sasaki, H., Saito, M., Youki, H., Karube, K., Konno, S., Onodera, M., Saito, T., and Aoyagi, S. (1989).J. Biol. Chem. 264, 17606-1 7612. Sato, Y., and Rifkin, D. B. (1988).J. Cell Biol. 107, 1199-1205. Saunders, S., Jalkanen, M., OFarrell, S., and Bernfield, M. (1989).J. Cell Biol. 108, 15471556. Schofield, P. N., Ekstrom, T. J., Granerus, M., and Engstrom, W. (1991).FEBSLett. 280,810.

Scholer, H. R., Hatzopoulos, A. K., Balling, R., Suzuki, N., and Gruss, P. (1989). EMBO J . 8, 2543-2550. Schreiber, A. B., Kenney, J., Kowalski, W. J., Friesel, R., Mehlman, T., and Maciag, T., (1985). Proc. Natl. Acad. Sci. U.S.A. 82, 6138-6142. Schubert, D., Ling, N., and Baird, A. (1987).J . Cell Bzol. 104, 635-643. Schweigerer, L.. Neufeld, G., Friedman, J., Abraham, J. A., Fiddes, J. C., and Gospodarowicz, D. (1987a). Nature (London) 325, 257-259. Schweigerer, L.. Neufeld, G., Mergia, A., Abraham, J. A., Fiddes, J. C., and Gospodarowicz, D. (1987b). Proc. Natl. Acad. Sci. U.S.A. 84, 842-846. Seno, M., Sasada, R., Iwane, M., Sudo, K., Kurokawa, T.,-Ito, K., and Igarashi, K. (1988). Biochem. Biophys. Res. Commun. 151, 701-708. Seno, M., Sasada, R., Kurokawa, T., and Igarashi, K. (1990). Eur. J . Biochem. 188, 239245. Shibata, F., Baird, A,, and Florkiewicz, R. 2. (1991). Growth Factors 4, 277-287. Shimasaki, S., Emoto, N., Koba, A., Mercado, M., Shibata, F., Cooksey, K., Baird, A,, and Ling, N. (1988). Biochem. Biophys. Res. Cornrnun. 157, 256-263. Shing, Y., Folkman, J., Sullivan, R., Butterfield, C., Murray, J., and Klagsbrun, M. (1984). Science 223, 1296- 1299. Shing, Y., Folkman, J., Haudenschild, C., Lund, D., Crum, R., and Klagsbrun, M. (1985).J. Cell. Biochem. 29, 275-287. Simpson, R. J., Moritz, R. L., Lloyd, C. J., Fabri, L. J., Nice, E. C., Rubira, M. R., and Burgess, A. W. (1987). FEBS Lett. 224, 128-132. Slack, J. M. W., and Isaacs, H. V. (1989). Development (Cambridge, U K ) 105, 147-153.

164

CLAUDIO BASILICO AND DAVID MOSCATELLI

Slack, J. M. W., Darlington, B. G., Heath, J. K., and Godsave, S. F. (1987). Nature (London) 326, 197-200. Slack, J. M. W., Isaacs, H. V., and Darlington, B. G. (1988). Development (Cambridge, U K ) 103, 581-590. Smith, R., Peters, G., and Dickson, C. (1988). E M B O J . 7, 1013-1022. Sommer, A., and Rifkin, D. B. (1989).J. Cell. Physiol. 138, 215-220. Sommer, A., Brewer, M. T., Thompson, R. C., Moscatelli, D., Presta, M., and Rifkin, D. B. (1987). Biochem. Biophys, Res. Commun. 144, 543-550. Sommer, A., Moscatelli, D., and Rifkin, D. B. (1989). Biochem. Biophys. Res. Commun. 160, 1267-1274. Speir, E., Sasse, J., Shrivastav, S., and Casscells, W. (1991). J. Cell. Physiol. 147, 362-373. Story, M. T., Esch, F., Shimasaki, S., Sasse, J., Jacobs, S. C., and Lawson, R. K. (1987). Baochem. Biophys. Res. Commun. 142, 702-709. Taira, M., Yoshida, T., Miyagawa, K., Sakamoto, H., Terada, M.,and Sugimura, T. (1987). Proc. Natl. Acad. Sci. U.S.A. 84, 2980-2984. Talarico, D., and Basilico, C. (1991). Mol. Cell. Biol. 11, 1138-1 145. Talarico, D., Ittmann, M., Balsari, A., Delli Bovi, P., Basch, R. S., and Basilico, C. (1990). Proc. Natl. Acad. Sci. U.S.A. 87, 4222-4225. Talarico, D., Ittmann, M. M., Bronson, R., and Basilico, C. (1992). Submitted for publication. Terranova, V. P., DiFlorio, R., Lyall, R. M., Hic, S., Friesel, R., and Maciag, T. (1985).J. Cell Biol. 101, 2330-2334. Theillet, C., LeRoy, X., DeLapeyriCre, O., Grosgeorges, J., Adnane, J., Raynaud, S. D., Simony-Lafontaine, J., Goldfarb, M., Escot, C., Birnbaum, D., and Gaudray, P. (1989). Oncogene 4, 915-922. Thomas, K. A., Riley, M. C., Lemmon, S. K. Baglan, N. C., and Bradshaw, R. A. (1980).J. Biol. Chem. 255, 55 17-5520. Thomas, K. A., Rios-Candelore, M., and Fitzpatrick, S. (1984). Proc. Natl. Acad. Scz. U.S.A. 81,357-361. Thomas, K. A., Rios-Candelore, M., Gimenez-Gallego, G., DiSalvo, J., Bennett, C., Rodkey, J., and Fitzpatrick, S. (1985). Proc. Natl. Acad. Sci. U.S.A. 82, 6409-6413. Thomsen, G., Woolf, T., Whitman, M., Sokol, S., Vaughan, J.. Vale, W., and Melton, D. A. (1990). Cell (Cambridge, Mass.) 63, 485-493. Togari, A., Baker, D., Dickens, G., and Guroff, G. (1983). Biochem. Biophys. Res. Commun. 114, 1189-1 193. Togari, A., Dickens, G., Kuzuya, J., and Guroff, G. (1985).J. Neurosci. 5, 307-316. Tsuboi, R., and Rifkin, D. B. (1990).J. Exp. Med. 172, 245-251. Tsuboi, R., Sato, Y., and Rifkin, D. B. (199O).J. CeLLBiol. 110, 511-517. Tsuda, T., Nakatani, H., Matsumura, T., Yoshida, K., Tahara, E., Nishihira, T., Sakamoto, H., Yoshida, T., Terada, M., and Sugimura, T. (1988).Jpn. J . Cancer Res. 79, 584-588. Tsuda, H., Hirohashi, S., Shimosato, Y., Hirota, T., Tsugane, S., Yamamoto, H., Miyajima, N., Toyoshima, K., Yamamoto, T., Yokota, J., Yoshida, T., Sakamoto, H., Terada, M., and Sugimura, T. (1989). Cancer Res. 49, 3104-3108. Tsutsumi, M., Sakamoto, H., Yoshida, T., Kakizoe, T., Koiso, K., Sugimura, T., and Terada, M. (1988). Jpn. J . Cancer Res. 79, 428-432. Ueno, N., Baird, A., Esch, F., Ling, N., and Guillemin, R. (1986). Biochem. Biophys. Res. Commun. 138,580-588. Ueno, N., Baird, A,, Esch, F., Ling, N., and Guillemin, R. (1987) Mol. Cell. Endocrinol. 49, 189- 194.

FGF FAMILY OF GROWTH FACTORS AND ONCOGENES

165

Uhlrich, S., Lagente, O., Lenfant, M., and Courtois, Y. (1986). Biochem. Biophys. Res. Commun. 137, 1205-1213. Ullrich, A., and Schlessinger, J. (1990). Cell (Cambridge, Mass.) 61, 203-212. Unsicker, K., Reichert-Preibsch, H., Schmidt, R., Pettmann, B., Labourdette, G., and Sensenbrenner, M. (1987). Proc. Natl. Acad. Sci. U.S.A. 84, 5459-5463. Valles, A. M., Boyer, B., Badet, J., Tucker, G. C., Barritault, D., and Thiery, J. P. (1990). Proc. Natl. Acad. Sci. U.S.A. 87, 1124-1128. Velcich, A., Delli Bovi, P., Mansukhani, A,, Ziff, E. B., and Basilico, C. (1989). Oncogene Reg. 5, 31-37. Vigny, M., Ollier-Hartmann, M. P., Lavigne, M., Fayein, N., Jeanny, J. C., Laurent, M., and Courtois, Y. (1988).J. Cell. Physiol. 137, 321-328. Vlodavksy, I., Folkman, J., Sullivan, R., Fridman, R., Ishai-Michaeli, R.,Sasse, J., and Klagsbrun, M. (1987). Proc. Natl. Acad. Sci. U.S.A. 84, 2292-2296. Wagner, J. A., and D'Amore, P. A. (1986).J. Cell Biol. 103, 1363-1367. Walicke, P., Cowan, W. M., Ueno, N., Baird, A., and Guillemin, R. (1986).Proc. Natl. Acad. Sci. U.S.A. 83, 3012-3016. Walz, T. M., Abdiu, A., Wingren, S., Smeds, S., Larsson, S.-E., and Wasteson, A. (1991). Cancer Res. 51, 3585-3589. Weich, H. A,, Iberg, N., Klagsbrun, M., and Folkman, J. (1990). Growth Factors 2,313-320. Wellstein, A,, Lupu, R., Zugmaier, G., Flamm, S. L., Cheville, A. L., Delli Bovi, P., Basilico, C., Lippman, M. E., and Kern, F. G. (1990). Cell Growth Difer. 1, 63-71. Wellstein, A,, Zugmaier, G., Califano, J. A,, 111, Kern, F., Paik, S., and Lippman, M. E. (1991).JNCI, J . Natl. Cancer Inst. 83, 716-720. Werner, S., Hof'schneider, P. H., and Roth, W. K. (1989a). Inf. J. Cancer 43, 1137-1144. Werner, S., Hofschneider, P. H., Stiirzl, M., Dicke, I., and Roth, W. K. (1989b).J . Cell. Phys. 141, 490-502. Wilcox, B. J., and Unnerstall, J. R. (1991). Neuron 6 , 397-409. Wilkinson, D. G., Peters, G., Dickson, C., and McMahon, A. P. (1988). EMBOJ. 7,691-695. Wilkinson, D. G., Bhatt, S., and McMahon, A. P. (1989).Development (Cambridge, U K ) 105, 131-136. Winkles, J. A., Friesel, R., Burgess, W. H., Howk, R., Mehlman, T., Weinstein, R.,and Maciag, T. (1987). Proc. Natl. Acad. Sci. U.S.A. 84, 7124-7128. Yarden, Y., and Ullrich, A. (1988). Annu. Rev. Biochem. 57, 443-478. Yayon, A., and Klagsbrun, M. (1990). Proc. Natl. Acad. Sci. U.S.A. 87, 5346-5350. Yayon, A., Klagsbrun, M., Esko, J. D., Leder, P., and Ornitz, D. M. (1991). Cell (Cambridge, Mass.) 64, 841-848. Yoshida, M. C., Wada, M., Satoh, H., Yoshida, T., Sakamoto, H., Miyagawa, K., Yokota, J., Koda, T., Kakinuma, M., Sugirnura, T., and Terada, M. (1988). Proc. Natl. Acad. Sci. U.S.A. 85, 4861-4864. Yoshida, T., Miyagawa, K., Odagiri, H., Sakamoto, H., Little, P. F. R.,Terada, M., and Sugirnura, T. (1987). Proc. Natl. Acad. Sci. U.S.A. 84, 7305-7309. Yoshida, T., Tsutsumi, M., Sakamoto, H., Miyagawa, K., Teshima S., Sugimura, T., and Terada, M. (1988). Biochem. Bzophys. Res. Commun. 155, 1324-1329. Zhan, X., Botes, B., Hu, X., and Goldfarb, M. (1988).Mol. Cell. B i d . 8, 3487-3495. Zhang, J., Cousens, L. S., Barr, P. J., and Sprang, S. R. (1991). Proc. Natl. Acad. Sci. U.S.A. 88, 3446-3450.

This Page Intentionally Left Blank

HEPATITIS B VIRUSES AND HEPATOCELLULAR CARCINOMA Marie Annick Buendia Departement des RBtrovirus, Unit6 de Recombinaison et Expression Gbnbtique, INSERM U163, lnstitut Pasteur, 75724 Paris Cedex 15, France

I. Introduction 11. Epidemiology: Clinical and Immunological Aspects

111.

IV.

V.

VI.

VII. VIII.

A. Prevalence of HBV Infections and Modes of Transmission B. Progression to Chronicity C. Epidemiologic Association of HBV with Hepatocellular Carcinoma Pathogenicity of Hepadnaviruses: Striking Similarities and Obvious Differences A. Mammalian Hepadnaviruses B. Avian Hepadnaviruses Hepadnavirus Genomes A. Genetic Organization of the HBV Genome B. Genome Structure and Replication C. Virion Assembly D. Regulated Expression of Viral Genes Potential Oncogenic Properties of Viral Proteins A. Surface Glycoproteins B. H B A Transcriptional Trans-activator Integrated State of Viral DNA in Chronic Infections and Hepatocellular Carcinoma A. Integrated Sequences: Physical and Functional Aspects B. Cellular Targets for Viral Integration in Human Hepatocellular Carcinoma C. Insertional Activation of inyc Family Genes in Woodchuck Hepatocellular Carcinoma Genetic Alterations in HBV-Related Hepatocellular Carcinoma Conclusions References

I. Introduction

Primary hepatocellular carcinoma (HCC), one of the most common cancers in many parts of the world, is also one of the rare human cancers showing seroepidemiologic association with a viral infection. The role of hepatitis B virus (HBV) as a causal agent of HCC has been clearly established, and the increased risk of developing HCC, estimated to be 100fold for chronic carriers of the virus as compared with noninfected individuals, places HBV in the first rank among known human carcinogens (Szmuness, 1978; Beasley and Hwang, 1991). Besides epidemiologic evidence, the existence of related animal viruses that form 167 ADVANCES IN CANCER RESEARCH. VOL. 59

Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.

168

MARIE ANNICK BUENDIA

with HBV the hepadnavirus group, and induce acute and chronic infections of the liver and eventually HCC (reviewed by Schodel et al., 1989; Robinson, 1990), add substantial weight to theconcept that HBV stands among the few recognized human oncogenic viruses. A potential role of HBV as an insertional mutagen, suggested by the constant finding of integrated viral sequences in the cellular DNA of HBV-associated HCC, has been described in rare cases but appears now highly improbable in a majority of liver tumors (reviewed by Matsubara and Tokino, 1990). The major unsolved question is whether HBV acts through any already known oncogenic mechanism, either directly or indirectly. T h e problems encountered in hepatitis research are better understood by the long difficulty in identifying the viral agent for hepatitis B and in setting u p convenient experimental systems for genetic and biological studies. After the initial discovery of the “Australia” antigen by Blumberg et al. in 1965, and its later identification with an envelope determinant of the infectious agent for hepatitis B, hepatitis research has long been hampered by the strict host specificity of HBV, which infects only humans and chimpanzees, somewhat uneasy experimental models, and by the absence of tissue culture systems capable of carrying out the complete HBV life cycle. Molecular cloning and sequencing of the HBV genome (Galibert et al., 1979), and the concomitant discovery of animal hepadnaviruses (Summers et al., 1978; Mason et al., 1980; Marion et al., 1980), have represented a major breakthrough by allowing intensive studies of the viral genetic organization and replication pathway. During the last decade, the outcome of cell cultures supporting viral replication and the construction of mouse lines carrying viral transgenes have made it possible to better delineate the contribution of individual proteins to the replicative machinery, the regulation of viral gene expression, and the role of their protein products in liver pathogenesis. These advances, and parallel studies of HBV DNA integration patterns into host cell DNA, have led to a different hypothesis on the contribution of HBV to hepatocarcinogenesis. It has been generally admitted for a long time that HBV has no direct oncogenic or even cytopathic effect on the infected hepatocyte; indeed, viral hepatitis appears to be an immunologic disease. Malignant transformation, which occurs after a long period of chronic liver disease and which is frequently associated with cirrhosis, might be triggered in a nonspecific manner by the immune response against infected hepatocytes, which induces a chronic inflammation of the liver and causes cell killing and consequent cell proliferation-known risk factors for cancer (Ames, 1989). In a still indirect, but more specific pathway, persistent production of viral proteins with potential cytotoxic effects might modify endogenous metabolic processes

HBV AND HEPATOCELLULAR CARCINOMA

169

and sensitize liver cells to endogenous or exogenous mutagens. Alternatively, the virus might play a direct role as a mutagen, through insertion of its DNA into host cell chromosomes, causing direct activation of protooncogenes or, in a roundabout way, secondary chromosomal aberrations. Integrated HBV sequences might also alter the host cell growth control through unregulated expression of native or modified viral proteins. Until1 recently, little attention has been paid to the underlying cellular pathway leading to malignancy in response to viral induction. To date, although information on risk factors causally linked to HCC has accumulated, the role of viral agents and carcinogenic cofactors is only partially elucidated. No unifying model, accounting for the contribution of viral and cellular factors to liver oncogenesis, has been proposed. In this context, our recent studies of the mechanisms linking woodchuck HCC with chronic infection by woodchuck hepatitis virus (WHV), a virus closely related to HBV, present an amazing contrast. In about 50% of the woodchuck tumors analyzed, we found activation of myc family genes (c-myc and N-myc) by nearby insertion of WHV DNA (Hsu et al., 1988; Fourel et al., 1990; Y. Wei, A. Ponzetto, and M. A. Buendia, unpublished results). This is the first case of a DNA virus producing insertional activation events at such a frequency. Although genetically related viruses showing similar pathobiological properties might be expected to develop common oncogenic strategies, insertional activation of myc protooncogenes by HBV DNA has never been observed in human liver cancer, in which other potentially oncogenic cellular targets for viral integration have been described only rarely (Dejean et al., 1986; Wang et al., 1990). Whether this striking difference is related to intrinsic properties of WHV, or to genetic or epigenetic variability between humans and rodents, remains to be determined. In addition, overexpression of myc genes, either through genetic alterations or by a still unknown trans-acting mechanism, has been observed frequently in liver tumors from woodchucks and ground squirrels infected with hepadnaviruses, identifying myc gene activation as a key step in the oncogenic process induced by hepatitis virus infection in rodents (Fourel et al., 1990; Moroy et al., 1986; Transy et al., 1992). In human HCCs, there is no published evidence for a predominant role of activated myc genes, albeit amplification of c-myc has been occasionally described (Trowbridge et al., 1988), suggesting a different transformation pathway. Rather than presenting an exhaustive survey of the recent advances in HBV-associated liver cancer research, this article will explore and compare different possible mechanisms by which hepadnaviruses may trigger liver cell proliferation and/or transformation, and considers the

170

MARIE ANNICK BUENDIA

factors that may influence the primacy of some oncogenic pathways over others in tumors induced by different viruses of the same family. II. Epidemiology: Clinical and Immunological Aspects Epidemiologic research has been and remains a main contributor to our understanding of the etiology of HCC. In the late 1970s, it became evident that chronic HBV infection was by far the major risk factor of liver cancer (Szmuness, 1978). These conclusions have rapidly invigorated the search for the molecular mechanisms linking HBV and HCC, and point out the importance of vaccination against HBV infection as the appropriate strategy to prevent HCC. Other carcinogenic factors frequently associated with HBV infection, like exposure to dietary aflatoxins and excessive alcohol intake, have also been implicated in human hepatocarcinogenesis (Bosch and Munoz, 1991). More recently, preliminary epidemiologic data also support a correlation between chronic infection with a quite different virus, the human hepatitis C virus (HCV), cirrhosis, and HCC (Saito et al., 1990). Early studies of the prevalence of HBV chronic carriers have been greatly facilitated by the presence of large amounts of empty viral particles, carrying the viral surface antigen (HBsAg) in the serum of many infected patients, that provide a stable and easily detectable marker of chronic hepatitis. Other serological markers, like the soluble antigen e related to the capsid (HBeAg) and the viral-associated DNA polymerase, allow the distinction, in many cases, of those patients who support active viral replication. The introduction of HBV DNA into the panel of commonly used HBV markers has been an important step in the improvement of diagnosis assays. DNA hybridization techniques and polymerase chain reaction (PCR) amplification procedures allow a more accurate, and even semiquantitative estimation of the rate of viral replication (Brechot et al., 1981a; Bonino et al., 1981; L a r d et al., 1988; Gerken et al., 1991b); these assays have played an essential role in the detection of HBV infections unrecognized by conventional assays (Brechot et al., 1985; Liang et al., 1991b) and in the identification of HBV variants lacking HBs antigenicity or presenting precore region defects (Thiers et al., 1988; Carman et al., 1989; Wands et al., 1986). However, epidemiologists must face major difficulties: the absence of clinical symptoms during most primary infections with HBV (inapparent or subclinical disease is the rule in young children) and the limited number worldwide of cancer registries, which sometimes do not discriminate between primary and secundary liver neoplasms (Bosch and Munoz, 1991). Within these constraints, the epidemiologic associa-

HBV AND HEPATOCELLULAR CARCINOMA

171

tion has brought compelling evidence that chronic HBV infection plays an important role in development of HCC. The modes of spread of the virus, which differ in endemic and nonendemic regions, and the factors that influence the establishment of persistent infections and modulate the relative risk of cancer development, have been studied in great detail (reviewed by Beasley, 1987; Hollinger, 1990a), providing a valuable basis for biological investigations of the oncogenic properties of HBV.

A. PREVALENCE OF HBV INFECTIONS AND MODESOF TRANSMISSION Persistent HBV infections are widely spread over the world, but are unevenly distributed, with prevalence rates ranging from 10 to 20% in coastal regions in China, in Southeast Asia, and subsaharan Africa, to less than 0.5% in northwest Europe, North America, and Australia. It has been recently estimated that worldwide there are nearly 300 million actively infective carriers of HBV markers (Ayoola et al., 1988), which is even more than in a previous estimation by Szmuness in 1978. Since 1985, however, a slight, gradual decrease in the incidence of hepatitis B has been observed in high endemic areas as well as in intermediate- and low-prevalence regions (Hollinger, 1990b; Goudeau, 1990). It may be attributed to the current hepatitis B control strategies, and particularly to the availability of new HBV vaccines. The unusual stability of infectious HBV virions, present mainly in the blood but also in other body fluids like saliva, urine, and semen, renders hepatitis B highly contagious. Transmission of viral hepatitis B can be achieved in many different ways, and varies greatly between regions of high and low endemicity. In endemic areas, the viral infection is most often acquired early in life, and the maintenance of an HBV carrier population has been correlated with perinatal transmission from infected mothers to their offspring, and with contact-associated transmissions during the first years of life. In low-risk areas, where parenteral transmission and sexual contacts appear to be the predominant modes of spread, hepatitis B is mostly confined to teenagers and adults. This epidemiologic distinction is essential for the establishment of control measures aimed to limit the spread of the virus; it is also important to understand the variations observed in the relative risk to develop persistent infections and eventually HCC among different populations. In highly endemic areas in Asia and Africa, as well as in endemic pockets that occur in low-prevalence regions within specific subgroups or limited areas (for instance, oriental immigrants in the United States,

172

MARIE ANNICK BUENDIA

native Alaskans, and populations of Mediterranean countries in southern Europe and the Middle East), the majority of HBV infections occur in early childhood. Extensive epidemiologic studies have delineated the relative importance of vertical and horizontal spread, but the precise modes of maternal-neonatal transmission remain incompletely understood. This is mainly due to the failure of the currently used immunological methods (i.e., detection of HBsAg and HBeAg) to allow a direct appraisal of the HBV status in potentially infected infants. The hypothesis of true vertical transmission of HBV, raised by the presence of infectious virions in the semen and the finding of integrated HBV sequences in spermatozoa of patients with active hepatitis (Hadchouel et al., 1985),is difficult to confirm in humans. Although the prevalence of maternal-fetus transmission is poorly documented, it is generally considered to be very low, whereas perinatal transmission, occurring during labor or delivery or during the first months after birth, might be the predominant mechanism of viral spread in certain areas of the world, particularly in Asia (Beasley et al., 1977; Merrill et al., 1972; Schweitzer et al., 1973). However, transplacental transmission has been described on rare occasions (Beasley and Hwang, 1984; Stevens et al., 1984). Horizontal transmission of hepatitis B is considered as relatively common among siblings in the same household, or through child-tochild contact within nurseries in endemic regions (Hollinger 1990a; Whittle et al., 1983). In Africa, horizontal transmission during the first years of life may account for the majority of hepatitis B cases (Ayoola, 1988). Horizontal plus perinatal transmission may also coexist in other populations (Machado et al., 1988). The reasons for the seroepidemiological differences between modes of spread in several ethnic groups are now apparent: they reflect differences in the HBV status among HBsAg carriers in these populations. Women who are chronic carriers of HBeAg, a marker of productive HBV infection, almost invariably transmit the virus to their offspring (Okada et al., 1976), but perinatal transmission among HBsAg- antiHBeAg-positive women is much less frequent; infants with a high transplacental anti-HBeAg titer at delivery are at lower risk to become HBV carriers (Chen et al., 1991). The protective effect of maternal antibodies is prolonged for about 6 months; thereafter, the children become susceptible to HBV infection (Ayoola, 1988). The importance of the HBeAg/anti-HBeAg status in transmission is further emphasized in the establishment of persistent infections, as discussed later. The introduction of HBV vaccination as part of routine programs, already decided by several organizations and governments, should prove to rapidly decrease the incidence of hepatitis B in endemic re-

HBV AND HEPATOCELLULAR CARCINOMA

173

gions. In Western, industrialized countries, major steps toward prevention of parenteral and contact-associated transmission were made recently through improved sensitivity in the screening of blood donors for HBsAg and relatively successful vaccination programs within a number of high-risk target groups like health care workers and paramedics, hemophiliacs, and persons living in institutions for the developmentally disabled (Hollinger, 1990b; Goudeau, 1990). Screening of pregnant women for the presence of HBsAg, and vaccination'of newborns from carrier mothers, should also further reduce the low rates of perinatal transmission. Persons at greater risk for contracting hepatitis B are now parenteral drug abusers, male homosexuals, and sexually active heterosexuals and the development of persistent infections is observed only in about 5% of primarily infected, immunocompetent adults (Aldershvile et al., 1980; McMahon et al., 1985). The differences observed in the mode of HBV transmission and in the rate of persistent infections between endemic and nonendemic countries are illustrated in Fig. 1. B. PROGRESSION TO CHRONICITY T h e establishment of persistent HBV infections remains incompletely understood; different mechanisms appear to be involved, most of them being related to the immunologic status of the host at the time of infection or to genetic heterogeneity of HBV. Most infections acquired at birth lead to a chronic carrier state; the risk drops rapidly within the first years of life and the frequency of chronicity decreases with increasing age at the time of infection (Beasley and Hwang, 1984; Hollinger, 1990a). A clear inverse correlation has been observed between increasing age and the outcome of clinical signs of acute hepatitis (McMahon et al., 1985) (Fig. 1). Immunocompetent adults are much less prone to the development of chronic hepatitis than young children, whose immune system is still immature. In adult life, increased risk of progression to chronicity is associated with immune deficiencies, as seen in renal transplant recipients, in patients with prior human immunodeficiency virus (HIV) infection, and in leukemic patients treated with chemotherapy (Degos et al., 1988; Taylor et al., 1988; Melegari et al., 1991). Hepatitis B virus replication in liver cells is associated with the production of three distinct antigens (HBsAg, HBcAg, and HBeAg) that elicit both cell-mediated and humoral immune responses (reviewed by Milich, 1988; Schodel et al., 1990). Anti-HBsAg is the main neutralizing antibody and its appearance signifies termination of HBV infection. HBsAg is also a potential target for the immune attack of infected hepatocytes by cytotoxic T lymphocytes ( C T L s ) . Studies of transgenic mice

174

MARIE ANNICK BUENDIA

Endemic countries v e r t i c a l transmission (Asia)

Nonendemic countries

contact-associated transmission (Africa)

parenteral and sexual transmission

$. newborns infants

r

young children

chronic c a r r i e r s ( 10-20%o f the t o t a l population)

I

teenagers adults

3

5-10%

chronic c a r r i e r s (0.5%of the t o t a l

I

40%

I I

1

25%

cirrhosis I

hepatocellular carcinoma

I

hepatocel l u l a r carcinoma

FIG. 1 . Comparative description of the modes of HBV transmission, the rate of progression to chronicity, and the HCC incidence in endemic and nonendemic countries.

expressing HBsAg under the control of HBV or albumin promoter indicate that immune tolerance and the absence of liver disease results from in utero exposure to HBsAg (Babinet et al., 1985; Moriyama et al., 1990). Cytolysis of HBsAg-positive hepatocytes is observed in one of these models after adoptive transfer of spleen cells from immunized

175

H B V A N D HEPATOCELLULAR CARCINOMA

congenic mice, indicating that cell-mediated immune response directed against the s antigen can be implicated in the pathogenesis of liver cell injury in human hepatitis B (Moriyama et al., 1990). It has recently been shown that tolerant animals can be induced to make antibodies against HBsAg group and subtype epitopes when injected with large amounts of HBsAg particles (Mancini et al., 1991). In humans, it can be correlated with the efficient humoral response of neonates to HBsAg carrier mothers following HBsAg vaccination (Beasley and Hwang, 1984). The predominant role of the HBeAgIanti-HBeAg status of the infected host in the outcome of chronic infections is now firmly established. The viral e antigen, first described in 1972 by Magnius and Espmark, has been identified with a processed product of the precore-core region, which differs from the capsid protein (HBcAg) by 10 additional amino-terminal residues originating from the pre-C region, after cotranslational cleavage of a signal peptide, and by a deletion of 34 amino acids from the carboxy terminus (Takahashi et al., 1983; Uy et al., 1986; Standring et al., 1988; Weimer et al., 1987) (Fig. 2). The t w o proteins share common antigenic determinants, but present distinct antigenic properties (Milich et al., 1978; Salfeld et al., 1989). These structural changes are also associated with different biological properties: the capsid proteins are linked to the viral genome through their phosphorylated carboxy-terminal region, self-assembled into core particles, and

--

pre C mRNA C mRNA

stop

HBV DNA 1814

2450

1901

174aa

precore precursor ( P25 1

212 aa

processingat signal sequewa

precore derivatlve (PZ)

HBe Ag ( pis-17e)

193aa

4

proteolyticcleavage 149-159 aa

FIG.2. The synthesis of HBV core antigen (HBcAg) and e antigen (HBeAg) in infected hepatocytes is controlled by different regulatory signals and by different posttranslational modifications.

176

MARIE ANNICK BUENDIA

incorporated into the envelope of secreted virions, whereas the e antigen is presented on the hepatocyte membrane and excreted as a dimer into the bloodstream (Ou et al., 1968; Jean-Jean et al., 1989; Schlicht and Schaller, 1989). T h e question arises of the biological function of HBeAg, which appears to be dispensable for viral replication in an animal model (C. Chang et al., 1987; Schlicht et al., 1987; Pugh et al., 1989). It has been shown that HBeAg/HBcAg is the major target of the cellmediated immune response and plays an important part in liver damage during acute and chronic acute hepatitis B, as well as in virus clearance (Mondelli et al., 1982; Schlicht et al., 1991). Usually, seroconversion to anti-HBe represents a crucial step in the course of the disease, and signals extensive elimination of the virus. However, persistence of antiHBe has also been observed in a subset of HBV carriers with chronic liver disease (Bonino et al., 1981; Hadziyannis et al., 1983), and has recently been correlated with a genomic variation of HBV that prevents the secretion of HBeAg (Brunetto et al., 1989; Carman et al., 1989; Tong et al., 1990). Infection with HBV precore mutants, already detected in different patients from many parts of the world, now appears to prevail among persistent carriers with anti-HBe, showing that genetic variations are used by HBV as a strategy to evade the immune system and persist in the infected host (Okamoto et al., 1990; Bonino et al., 1991).HBe-defective HBV causes a form of hepatitis with severe pathogenicity and even fulminant hepatitis, but the factors involved in liver damage are still unknown (Brunetto et al., 1991; Carman et al., 1991; Liang et al., 1991a; Omata et al., 1991). It seems probable that different mutations arising in the pre-C sequence during long-term chronic HBV infections confer selective advantage to variant HBV and lead to the emergence of antiHBe-positive diseases (Tran et al., 1991). The spread of e-negative HBV mutants might be limited, in endemic regions, by the low rates of transmission from anti-HBeAg-positive mothers. In contrast, virtually all infants born to HBe-positive mothers become chronic carriers, and many of them remain HBe positive all their lives. A tolerogenic role of the e antigen in the establishment of these persistent infections has been recently proposed (Milich et al., 1990). The hypothesis that HBe may cross the placenta and induce tolerance in utero is supported by experimental data obtained in HBe-expressing transgenic mice; it is consistent with human epidemiological and serological observations. Other general mechanisms allowing a virus to persist in a host (Oldstone, 1989) may also operate in chronic hepatitis B. Avoidance of immune surveillance is achieved by “capping” of HBcAg by anti-HBc antibodies on the surface of infected hepatocytes (Mondelli et al., 1982).

HBV A N D HEPATOCELLULAR CARCINOMA

177

Abrogation of lymphocyte/macrophage function may result from HBV infection of the immunocytes-HBV DNA and antigens have been found in bone marrow cells and in peripheral blood lymphocytes (Elfassi et al., 1984; Pontisso el al., 1984; Pasquinelli et al., 1986; Laure et al., 1987), from selective killing of HBsAg-specific B cells by class I-restricted CTLs (Barnaba et al., 1990), and from deficient production or blunted action of a-interferon (INF), which reduces HLA class I protein synthesis and prevents recognition of infected cells by specific CTLs (Ikeda et al., 1986; Onji et al., 1989; Twu et al., 1988). In addition, males are at higher risk than females of becoming chronic carriers (Szmuness et al., 1978). The importance of steroid hormones in the development of chronicity has been established in clinical assays showing that treatment of acute hepatitis by steroids increases the frequency of subsequent persistent infections (Blum et al., 1969). The role of steroids in enhancing viral gene expression has been demonstrated in HBsAg-expressing transgenic mice (Farza et al., 1987). Multiple, interrelated mechanisms allowing for viral persistence may therefore contribute to the progression to chronicity. Progress in understanding these mechanisms would allow us to improve the treatment of chronic hepatitis B, a major health problem owing to the gravity of its sequellae, cirrhosis and primary liver cancer. C. EPIDEMIOLOGIC ASSOCIATION OF HBV WITH HEPATOCELLULAR CARCINOMA Primary liver cancer, mainly HCC, ranks among the most frequent cancers of males in many countries. In a recent estimation (Bosch and Munoz, 1991), it represents the eighth most common cancer, with about 250,000 new cases each year, 70% of which occur in Asia. Several lines of evidence associate chronic HBV infection with the development of HCC. 1. T h e incidence of HCC and the prevalence of HBV serological markers follow the same general geographic pattern of distribution. Hepatocellular carcinoma is common in regions where HBV is endemic, but comes far behind other types of cancer in regions where HBV infection is uncommon (Szmuness, 1978; Tabor, 1991; Hollinger, 1990a). 2. Serologic evidence of HBV infection is detected in about 70% of HCC patients in Africa, and more than 90% in mainland China, as compared with 10 to 20% of the total population residing in the same areas (Tabor, 1991). 3. A marked increase risk of HCC has been shown among HBsAg carriers, compared with noncarriers [risk factors up to 200 have been

178

MARIE ANNICK BUENDIA

reported in different ethnic or social groups, using different methodologies of investigation (Beasley et al., 1981; Hall et al., 1985; Chen et al., 1991; Obata et al., 1980; Beasley and Hwang, 1991)]. The long period of chronic HBV infection that generally precedes the onset of liver tumors has led to the proposal that “the oncogenic sword of HBV is terrible, but not swift” (Ganem, 1990). It might be added that it is unevenly hanging over carrier individuals. Hepatocellular carcinoma usually develops after a 20- to 50-year period of chronic carriage, but it may also affect, although unfrequently, HBsAg-positive children under 12 years of age. In contrast, a large number of HBsAg carriers remain anaware of their carrier status before they die at old ages (Beasley and Hwang, 1984; Chen et al., 1991; Lok et al., 1991).Epidemiological data have clearly shown that HBV is causally related with the development of HCC, but also that there is a great variation in HCC incidence among different carrier populations. This variation may be attributed both to differences in the intrinsic properties of HBV infection patterns observed among chronic carriers, and to additional genetic and environmental determinants. Chronic infections resulting from material-neonatal transmission present a greater risk of HCC than those acquired as adults (Beasley and Hwang, 1984; Popper et al., 1987a). In the Far East, early detection of HCC is now frequent in asymptomatic carriers. Among HBsAg carriers infected at an early age, additional HCC risk has been associated with HBeAg carriage, with significant liver damage and a high level of antiHBcAg antibodies in chronic active hepatitis, and with the presence of cirrhosis (Beasley and Hwang, 1984; Chen et al., 1991). A gender discrepancy (males incur a two- to eight-fold elevated risk of developing HCC than do females) and familial tendency (familial clusters of HCC are common in Asia) have also been documented as factors involved in the frequency of tumor development (Obayashi et al., 1972; Lok et al., 1991). In addition, inconsistent geographical variations in HCC mortality and HBsAg prevalence have been observed in endemic regions, suggesting that other independent or cooperative factors might be implicated. In highly endemic regions, particularly in South Africa and in southern provinces of mainland China, an association of dietary aflatoxins and primary HCC has been recognized in several reports (Bosch and Munoz, 1991; Harris, 1990; Kew, 1990). The carcinogenic potential of aflatoxin B1 in liver cells is well known in many species (Newberne and Butler, 1969). Excessive alcohol intake also increases the risk of HCC, in HBV carriers as well as in cirrhotic males of advanced age in regions of

HBV AND HEPATOCELLULAR CARCINOMA

179

low prevalence for HBV (Austin, 1991; Chen et al., 1991). However, the coexistence of HBV infection, often undetectable by conventional serologic assays, and/or of HCV infection in more than 90% of patients from various countries, has called in question the prevalence of chronic hepatitis induced by alcohol (BrCchot et al., 1985; Takase et al., 1991). T h e potential role of cigarette smoking and of long-term use of oral contraceptives is still debated (Austin, 1991). In addition, preliminary data indicate that infection with the human hepatitis delta virus (HDV), which causes extremely severe hepatic injury and cirrhosis, might be associated with a more rapid onset of liver tumors (Oliveri et al., 1991). It is noticeable, however, that HBV infection, alone or associated with cooperative factors, is clearly implicated in only 20% of HCC cases in low-endemic regions (North America and Europe) and in Japan. In this country, the incidence of liver cancer has been continuously increasing during the past decade, but the number of cases related to HBV remains constant. It seems now probable that infection with hepatitis C virus (HCV), a human RNA virus related to the Flaviviridae and Pestiviridae, plays an increasing part in the development of HCC in these regions, as well as in countries highly endemic for HBV, such as in China. With the development of HCV markers, evidence is now increasing for an association between HCV infection, cirrhosis, and HCC (Okuda, 1991). Some differences have been noted between HBV- and HCV-associated tumorigenic processes: HCV-related HCC develops after a longer incubation period and frequently presents more benign histological features. T h e relationship between cirrhosis and HCC appears to be complex and the degree of correlation varies with the etiology of cirrhosis. Macronodular cirrhosis precedes or accompanies a majority of HBVassociated HCC (over 80% in Asia and 40-60% in Africa), in children as well as at older ages (M. H. Chang el al., 1988). T h e risk of HCC is much lower in HBsAg-negative micronodular cirrhosis observed in alcohol hepatitis and in HCV infections (Beasley and Hwang, 1984; Kew and Popper, 1984; Craig el al., 1991). Whether cirrhosis and carcinogenesis result from the action of common factors, or whether increased risk of HCC can be related to some mechanism like regenerative stimulation, which occurs in cirrhotic livers, remains an unsolved problem.

Ill. Pathogenicity of Hepadnaviruses: Striking Similarities and Obvious Differences As in many infectious diseases in humans, progress in the study of viral hepatitis has been critically dependent on the development of animal models, consisting either in experimental systems like HBV-inoculated

180

MARIE ANNICK BUENDIA

chimpanzees and mouse strains carrying viral transgenes, or in naturally occurring models. Hepatitis B virus is the prototype member of the hepadnaviruses family, which includes a small number of enveloped DNA viruses with liver tropism and restricted host range (Summers, 1981). These viruses share a common genetic organization and replication pathway (as discussed in Section IV) and the ability to induce acute and persistent infections in their natural hosts. Animal hepadnaviruses have been isolated from rodents and birds and form two groups that have largely diverged during evolution; they will be considered separately. A. MAMMALIAN HEPADNAVIRUSES

Mammalian viruses include woodchuck hepatitis virus (WHV),which infects eastern woodchucks (Mamnota monax) in several states on the eastern coast in North America (Summers et al., 1978),and ground squirrel hepatitis virus (GSHV), a virus present in Beechey ground squirrels (Spermophilus beecheyi) in a limited area bordering the Stanford campus in California, Richardson squirrels (Spermophilus richarhonyii) at Picture Butte in Alberta (Canada), and possibly tree squirrels (Marion et al., 1980; Minuk et al., 1986; Feitelson et al., 1986). All these species belong to the Sciuridae family of rodents, and derived about 10 million years ago from a common ancestor marmotini. The divergence time between WHV and GSHV has been estimated as 10,000years, suggesting that the evolution of the hepadnavirus family was independent of host-species divergence (Orito et al., 1989).Ground squirrel hepatitis virus can infect other members of the Sciuridae family, as shown by experimental transmission of GSHV to woodchucks and chipmunks, but not to other rodent species (Seeger et al., 1978, 1991; Ganem et al., 1982a; Trueba et d., 1985; Marion et al., 1983; Chomel et al., 1984). Hepadnaviruses might also be associated with the development of liver tumors in related species (Snyder, 1979). Several WHV and GSHV isolates have been cloned and their nucleotide and amino acid sequences do not differ more than those of two different HBV subtypes (Cohen et al., 1988; Etiemble et al., 1986; Galibert et al., 1982; Ganem et al., 1982b; Girones et al., 1989; Kodama et al., 1985; Seeger et al., 1984b; Siddiqui et al., 1981). Their surface and core antigens are highly cross-reactive, and also cross-react with the corresponding HBV antigens (Feitelson et al., 1981, 1982; Cote and Gerin, 1983; Gerlich et al., 1980). Chimpanzees immunized with WHV surface antigen (WHsAg)can be protected after challenge with HBV (Cote et al., 1986). However, substantial variations have been noted in the pathological properties of WHV and GSHV, not only in their natural hosts, but

HBV AND HEPATOCELLULAR CARCINOMA

181

also in a common host, the experimentally infected newborn woodchuck (Ganem et al., 1982a; Marion et al., 1983; Popper et al., 1981; Seeger et al., 1991; Snyder and Summers, 1980). The mode of spread of mammalian hepadnaviruses in the wild has not been firmly established. Vertical transmission from carrier dams to their offspring might prevail, as suggested by the presence of WHV DNA in adult woodchuck ovaries and testis, and in livers and sera of several woodchuck fetal litters (Korba et al., 1988b; Kulonen and Millman, 1988). Experimental infection of newborn woodchucks with WHV progresses almost invariably to chronicity, whereas most animals infected at older ages develop acute hepatitis and efficient immune response that leads to viral clearance (Popper et al., 198713). Chronic WHV infection of woodchucks usually results in mild portal hepatitis, in which limited portal inflammation is associated with minimal liver damage and the absence of hepatocyte necrosis and fibrosis, mimicking the human “healthy” carrier state (Snyder and Summers, 1980; Popper et al., 1981, 1987b; Toshkov et al., 1990; M. A. Buendia, unpublished observations). Extensive studies of the tissue tropism of WHV in chronically infected woodchucks have shown that active viral replication occurs mainly in the liver, with high levels (500-2000 genome units per cell) of replicative intermediates and viral RNA in hepatocytes (Korba et al., 1988b). Abundant viral replicative forms and RNA transcripts have also been observed in the spleen, whereas nonreplicating WHV DNA has been found in peripheral blood lymphocytes (PBLs) and replicative forms in scattered foci of cells within the pancreas, kidney, thymus, and transiently in ovary and testis (Korba et al., 1986, 1987, 1988b, 1989a, 1990; Ogston et al., 1989). Studies of the natural history of WHV infections have revealed that lymphoid cells of the bone marrow are first infected, followed by the liver, spleen, PBLs, lymph nodes, and thymus (Korba et al., 1989a, 1990). In addition, WHV DNA replication can be induced in PBLs on in zritro activation with a mitogen (Korba et al., 1988a). These findings, which can be related to the presence of HBV DNA in human bone marrow and PBLs, address the question of the mechanisms targeting the oncogenic potential of hepadnaviruses almost exclusively toward liver cells. T h e clinical and histological signs of acute and chronic viral infection are even less apparent in GSHV carrier ground squirrels (Ganem et al., 1982a; Marion et al., 1980, 1983; Seeger et al., 1984a). T h e animals appear to be relatively healthy in captivity and, on histological examination, their livers show very mild portal hepatitis characterized by periportal inflammation and proliferation of bile ductules without evidence of hepatocellular necrosis o r significant disruption of hepatic architec-

182

MARIE ANNICK BUENDIA

ture. Young animals are highly susceptible to horizontal infection, but neither clinical nor pathological signs of acute hepatitis have been demonstrated, despite the presence of high levels of viral replicative forms in the liver (600-6000 genome units per cell) and of abundant viral particles in the serum. Most primary infections transmitted by the percutaneous route are self-limited, the rate of persistence being as low as for HBV infections in adults. At variance with WHV-infected woodchucks, GSHV DNA has been detected only in livers of chronically infected ground squirrels. Finally, virtually all WHV carrier woodchucks succumb to HCC after 2-4 years. A much lower risk (5.4%) has been established between the development of HCC and past WHV infection with seroconversion to anti-woodchuck hepatitus surface antigen (anti-WHs) (Korba et al., 1989b; Popper et al., 1987b). Like human HCC, woodchuck HCCs are primarily of the well-differenciated trabecular type and occasionally of the pseudoacinar type, but they differ in their inability to produce metastasis. At the time HCCs are detected, pericarcinomatous livers show moderate or acute hepatitis, inflammation, proliferation of bile ducts and connective tissues, necrosis, and degeneration of hepatocytes. Ground-glass cells have been occasionally observed. These clinical and histological features are highly similar to those associated with chronic hepatitis B in humans and can be classified using the same criteria as medical pathologists. Whether these signs of necroinflammation and acute hepatitis precede the tumor onset by a short period of time or result from compression of adjacent liver tissues by expanding tumor masses has not been determined. A similar question has been raised about the sequence of events leading to cirrhosis and HCC (often detected simultaneously) in asymptomatic human carriers (Beasley and Hwang, 1984). In addition, regenerative hepatocellular nodules and hepatocellular adenomas, which are believed to represent precursor lesions to carcinoma, are characteristically observed in WHV-infected woodchuck livers as during experimental carcinogenesis induced by chemicals in rats and mice (Abe et al., 1988; Roth et at., 1985; Popper et al., 1981). The frequency of tumor incidence in humans and rodents is generally correlated with the fractional life span in a similar manner. The average life span of captive healthy woodchucks, which never develop primary liver tumors, is about 15 years. The viral-induced neoplastic process appears therefore relatively more rapid in woodchucks than in humans, suggesting that WHV might be a more oncogenic virus than HBV. Moreover, HCC occurs as well in captive woodchucks, infected at birth and kept under strictly controlled conditions, ruling out the contribution of exogenous carcinogenic cofactors (Popper et al., 198713).

H B V A N D HEPATOCELLULAR CARCINOMA

183

In sharp contrast, chronic GSHV infection induces a delayed onset of liver tumors, with a much lower incidence in naturally infected ground squirrels as well as in experimentally infected woodchucks (Marion et al., 1986, 1987; Seeger et al., 1991). Hepatocellular carcinoma develops only in about 50% of GSHV carrier animals after a 4- to 8-year latency period. A similar tumor incidence has recently been reported in a study of Richardson squirrels infected with a virus closely related to GSHV (Tennant et al., 1991). Furthermore, the occurrence of liver tumors in convalescent ground squirrels after seroconversion, and in aging animals showing no marker of past or ongoing GSHV infection, indicates a weaker epidemiologic association between GSHV carriage and HCC than in the M7HV/woodchuck model. Within the same period of time, the risk of HCC development has been estimated to be threefold higher in WHV-infected woodchucks than in GSHV-infected ground squirrels, a rather surprising observation considering that woodchucks and squirrels are members of the same Sciuridae family, and that WHV and GSHV are closely related to each other. The importance of viral and host determinants in the striking differences observed between the two systems has recently been investigated and it has been established that GSHV and WHV differ in oncogenic determinants that can affect the kinetics of HCC development (Seeger et al., 1991). Determining the precise nature of these oncogenic factors should be of crucial importance, not only for understanding viral-induced carcinogenesis in rodents, but also to get some insight into the mechanisms linking HBV infection and HCC development in humans. A main difference between human and rodent hepatitis B resides in the absence of associated cirrhosis in woodchuck and squirrel livers, even after prolonged viral infection. In the rodent species, hyperplastic nodules might play a role as a precursor lesion during transition to HCC. These nodules are clonal and present morphological and metabolic aberrations resembling those observed during chemically induced hepatocarcinogenesis in rats (Rogler et al., 1987; Toshkov et al., 1990). It is important to notice that rodents in general are not prone to the development of cirrhosis; the rat model of liver fibrosis, which reproduces some features of the human cirrhotic process, can be obtained only after treatment with carbon tetrachloride or long-term feeding with a high-fat diet containing ethanol (Hall et al., 1991; McLean et al., 1969; Tsukamoto et al., 1986). Species-specific factors like the rapid onset of hepatocytic proliferation following liver damage in rodents, rather than intrinsic properties of the different hepadnaviruses, might therefore account for this discrepancy. Similarly, the presence of high levels of viral replicative intermediates in woodchuck and squirrel livers at the

184

MARIE ANNICK BUENDIA

tumor onset, as compared with the marked decline of HBV markers in human patients with HCC, might reflect a more rapid development of HCC following initiation and promotion in rodents (Farber, 1984). I n conclusion, clinical and histological studies have shown that both WHV and GSHV represent potent, albeit different, animal models for liver diseases and hepatocarcinogenesis in humans. Advantage might be taken of the substantial variations in oncogenic properties between the mammalian hepadnaviruses to approach the molecular mechanisms connecting HBV and HCC.

B. AVIANHEPADNAVIRUSES Duck hepatitis virus (DHBV) was first discovered in brown domestic ducks residing in China and thereafter in flocks of Pekin ducks and in other wild duck populations from most parts of the world; more recently, a related virus was isolated in gray herons in Germany (heron hepatitis virus, HHBV) (Zhou, 1980; Mason et al., 1980; Sprengel et al., 1988). These viruses are more distantly related to HBV in their genetic organization than are mammalian hepadnaviruses, but they share with other members of the hepadna group similar replication mechanisms and biological properties, and represent therefore interesting models. Experimental systems using infection of ducklings with cloned viral DNA, in uitro infection of primary duck hepatocytes, and transfection with cloned DNA of different cell lines capable of producing infectious virions have been successfully developed to study the hepadnaviral life cycle, as discussed in Section IV (Tuttleman et al., 1986b; Sprengel et al., 1984; Galle et al., 1988). Vertical transmission in DHBV-infected ducklings has been demonstrated as the prevalent mode of natural transmission in this model (Urban et al., 1985). Duck hepatitis virus is transmitted from the viremic dam to the yolk sac, probably through passive transfer with liver-derived yolk proteins, and then to the developing embryos. Viral replication starts in embryonic livers at 6-8 days of incubation (Urban et al., 1985; Tagawa et al., 1987). Like mammalian hepadnaviruses, DHBV is mainly hepatotropic, but extrahepatic viral replication has also been observed, particularly in kidney and pancreas (Halpern et al., 1983). Generally, chronic DHBV infection is not associated with apparent liver disease. Most infected duck livers produce high levels of viral replicative intermediates and excrete large amounts of viral particles into the circulating blood, but show only very mild hepatitis, which does not usually evolve to carcinogenesis (Marion et al., 1984). Variations in the severity of liver lesions and intensity of viremia have been noted among wild mallard and domestic Pekin DHBV infections (Lambert et al., 1991). The devel-

HBV AND HEPATOCELLULAR CARCINOMA

185

opment of cirrhosis and HCC has been observed only in a particular Chinese flock of DHBV-infected ducks, originating from the province of Qidong in mainland China (Zhou, 1980; Yokosuka et al., 1985; Imazeki et al., 1988), but not in Pekin ducks infected in Western countries. This difference might be related to genetic variations among DHBV strains and/or among susceptible hosts. Alternatively, HCC development might not be strictly correlated with DHBV infection but rather be due to exposure to dietary afatoxins (L. Cova, personal communication), also demonstrated in the human food in the same area (Sun and Chu, 1984). Introduction of aflatoxin B1 in the diet of DHBV-infected Pekin ducks kept in laboratory facilities has also proven to play a critical part in the incidence of duck HCC (Uchida et al., 1988; Cova et al., 1990). The study of viral and chemical factors contributing to HCC development in ducks might help to clarify some of the basic mechanisms of liver carcinogenesis. IV. Hepadnavirus Genomes

Since the earliest studies of the nucleotide sequence of cloned HBV genomes and of the viral replication by reverse transcription of an RNA intermediate (Galibert et al., 1979; Pasek et al., 1979; Summers and Mason, 1982), there has been constant interest in the unique genetic organization and replicative pathway of hepatitis B viruses. Virological and molecular studies have outlined the structural organization of the HBV genome, its coding potential, the mode of transcription of individual viral genes, and their functional capacities; the main aspects of viral DNA replication and virion assembly within infected hepatocytes have been unraveled. This article does not attempt to give an exhaustive review of the physical and biological properties of hepadnaviruses, as several reviews have recently appeared (Tiollais et al., 1985; Howard, 1986; Ganem and Varmus, 1987; Robinson et al., 1987; Chisari et al., 1989a; Mason and Taylor, 1989; Schodel et al., 1989; Robinson, 1990; Schroder and Zentgraf, 1990; Tiollais and Buendia, 1991). Less is known about the viral-cellular interractions that control virus attachment, uncoating, and entry into susceptible cells (the cell-surface receptor for HBV has not been identified with certainty), as well as the regulation of viral gene transcription and the integration of viral DNA in the host cell genome. OF THE HBV GENOME A. GENETIC ORGANIZATION

Nucleotide and deduced amino acid sequences of cloned HBV DNA from different virus subtypes (adw, adr, ayw, ayr) have revealed a genome

186

MARIE ANNICK BUENDIA

size of 3.2 kb and the presence of four open reading frames, localized on one viral strand in the same transcriptional orientation (Galibert et al., 1979; Pasek et al., 1979; Valenzuela et al., 1980; Fujiyama et al., 1983; Ono et al., 1983; Kobayashi and Koike, 1984; Bichko et al., 1985; Okamoto et al., 1986; Vaudin et al., 1988; Loncarevic d al., 1990) (Fig. 3). Two of them, the C and S regions, specify structural proteins of the virion core and surface (or envelope); the longest one, P, encodes a

FIG. 3. Genetic organization of the HBV genome. Four open reading frames encoding seven peptides are indicated by large arrows. Regulatory sequences [promoters, enhancers, and glucocorticoid responsive element (GRE)] are marked. Only the two major transcripts (corelpregenome and S mRNAs) are represented. DRl and DR2 are two directly repeated sequences of 1 1 bp at the 5’ extremities of the minus- and plus-strand DNA.

HBV AND HEPATOCELLULAR CARCINOMA

187

polyprotein necessary for viral replication that binds to DNA at its amino terminus, and presents DNA polymerase, reverse transcriptase, and RNase H activities; the smallest, X, codes for a transcriptional transactivator. T h e entire viral genome is coding, even in two different reading frames on a large portion of the genome. More striking than the overlapping of the P gene with the other viral genes, a feature common to all structures presenting reverse transcriptase activity (Toh et al., 1983), is the constant and unusual overlapping of coding and regulatory sequences (promoters, enhancers, and termination signal). The genomes of different HBV subtypes differ mainly by single nucleotide substitutions or by addition of multiples of three nucleotide blocks, preserving the reading frames and leading primarily to conservative amino acid changes. Genomic variability among viruses of the same subtype have been characterized in different patients and also in the same patient at different times during chronic infection; annual mutation rates for viral DNA have been estimated at 2.6 X mutations/site/year or less, a low value compared to those of RNA viruses (Okamoto et al., 1987b). These genetic variations can be attributed to errors during synthesis of the minus-strand DNA by reverse transcriptase, an enzyme that lacks polymerase-associated proofreading functions. A number of mutations, however, lead to the outcome of HBV variants with modified immunological and pathological properties: altered immunological response has been correlated with HBV-related variants (Blum et al., 1991; Wands et al., 1986; Thiers et al., 1988; Tran et al., 1991), and severe liver injury with e-negative mutant-associated hepatitis (Raimondo et al., 1990; Liang et al., 1991a; Omata et al., 1991). Coinfection with different HBV isolates has been shown to induce homologous recombination between the different viral genomes, and defective viral particles carrying a deleted genome have been described (Gerken et al., 1991a; Okamoto et al., 1987a; Terri. et al., 1991), showing that HBV shares with other viruses, and notably with retroviruses, the property of developing defective variants in the natural host. In addition, the presence of an in-frame ATG upstream of the X translation initiation codon in some HBV subtypes, suggesting the existence of a pre-X open reading frame comparable to pre-S and pre-C, has been correlated with increased replication rate (Loncarevic et al., 1990). Although deleted viral forms have been observed in a human HCC, creating core/polymerase fusion proteins (Will et al., 1986), there is no experimental evidence that free defective HBV genomes might present oncogenic properties, or that particular HBV subtypes might be more oncogenic than others. Comparison of the HBV genome with those of animal hepadnaviruses reveals extensive homologies among mammalian hepadnaviruses, which

188

MARIE ANNICK BUENDIA

share a basically identical genomic organization. The slightly longer size of rodent viral genomes results from additional sequences in the pre-S1 region, in the P gene immediately upstream of pre-S1, and in the carboxy-terminal part of the C gene (see Schodel et al., 1989). The best conserved regions are located in the C and S genes and in the viral DNA polymerase. The extent of homology is weaker in the pre-S1 region, which, however, retains the same general conformation and hydrophobicity profile. The rather strict host range of the different hepadnaviruses suggests that the cellular receptor for these viruses, which has not been fully characterized, might be poorly conserved during evolution. Accordingly, viral sequences implicated in the interaction of the virion with the cellular membrane are located in the amino-terminal part of the pre-S1 domain (Neurath et al., 1986), a variable region among hepadnaviral genomes. Finally, the internal domain of the X protein is the less conserved region. Two strong homology blocks in the amino- and carboxy-terminal parts of X might correspond to a conservative pressure for functional activity of the X trans-activator, and of the viral RNase H encoded by C-terminal P sequences, which overlap with the 5‘ end of X (Radziwill et al., 1990; Takada and Koike, 1990).Another striking feature is the shorter size of WHx and GSHx, which lack the homologous counterpart for the 9- 12 carboxy-terminal residues of HBx, abolishing the overlap with the pre-C region that occurs in all HBV subtypes, and creating a noncoding region of 11 nucleotides in GSHV. In WHV, the presence of an additional ATG upstream of the precore translation initiation codon has not been clearly correlated with the synthesis of a longer pre-C/C protein product. Sequence analysis of five WHV isolates has shown that the extent of nucleotide variation was lower than between different HBV subtypes, ranging from 0.5% among isolates from the same geographical area to 3.1 % among other isolates (Girones et al., 1989). Accordingly, the mutation rate of WHV DNA in a chronically infected woodchuck, inoculated with an infectious WHV clone has been estimated to be 2.3 x base substitutions/site/year, a value comparable to that found for HBV (Girones and Miller, 1989). Other studies have shown that a significant proportion of replicative intermediates in woodchucks is deleted and defective (Etiemble et al., 1988; Miller et al., 1990). Highly rearranged viral forms (“novel forms”) have also been described in chronic infections (Rogler and Summers, 1982). Replication defective WHV variants do not appear to play a part in the establishment of chronic infections (Miller et al., 1990). Whether they contribute to hepatocarcinogenesis has not been investigated, but it seems probable that novel forms can also integrate into the host genome (Hsu et al., 1988; Ogston et al., 1982).

HBV AND HEPATOCELLULAR CARCINOMA

189

T h e avian hepadnavirus genomes are shorter (3.0 kb) and show marked divergence at the nucleotide and amino acid level with mammalian viruses (Mandart et al., 1984; Sprengel et al., 1988). An X open reading frame is missing and the avian C gene is significantly larger than its mammalian counterpart. Sequence comparisons between avian hepadnavirus genomes (DHBV and HHBV) have revealed significant genomic variations, as among HBV subtypes and among rodent viruses (Sprengel et al., 1988, 1991). The strict host range of DHBV and HHBV, which differ notably in the pre-S1 region, might allow the identification of pre-S sequences responsible for recognition of the cell-surface receptor for avian hepadnaviruses.

B. GENOME STRUCTURE AND REPLICATION The hepadnaviral genome, isolated from infectious extracellular virions (Dane particles), is made of two complementary DNA strands of different length, maintained in a circular configuration by base pairing at their 5’ extremities (Summers el al., 1975; Hruska et al., 1977; Sattler and Robinson, 1979) (Fig. 3). The viral replication pathway, virtually identical for all hepadnaviruses, takes place in the nucleus and cytoplasm of infected cells, and although it can be instructively compared with the retroviral life cycle, it is entirely extrachromosomal (Seeger et al., 1986; Will et al., 1987) (Fig. 4).On virus entry in hepatocytes, the partially double-stranded DNA is converted to doublestranded DNA and then to covalently closed circular DNA (cccDNA, also designated supercoiled DNA). This process, requiring removal of the primers covalently linked to 5’ DNA extremities and of the seven- to eight-base terminal redundancy of the minus strand, is achieved by a set of cellular enzymes that have not been clearly identified. cccDNA, localized primarily in the cell nucleus, serves as a template for viral transcription, a step that coincides with the multiplication of replicative forms. At this stage, several viral RNAs are produced: messenger RNAs specifying surface, pre-C/C, and X protein products, and the so-called “pregenome” RNA, identified as a replicative intermediate as well as the message for C and P genes (Summers and Mason, 1982; L. J. Chang et al., 1989; Schlicht et al., 1989). The contribution of viral proteins to subsequent steps of the replicative process has been largely delineated using mutational analysis of DHBV and in vitro culture systems. Briefly, the asymmetric synthesis of viral minus- and plus-strand DNA can be schematized as follows. 1. Packaging of selected genomic RNA into subviral cores, a cotranslational process mediated by the polymerase gene product, using

190

MARIE ANNICK BUENDIA

+ strand envelope

- Strand

capsid

receptor

polyprotein polymerase

ccc DNA 3.5 m pmgenome RNA

L

- strand

d

2.4 kb mRNA

-3-

2.1 kbmRNA

3 .

large S middle

Nucleus 22 nm particle

0

J \

FIG. 4. Schematic illustration of the hepadnavirus life cycle in hepatocytes. "Polyprotein" designates the primary product of translation of the P gene, which encodes the terminal protein and the polymeraselreverse transcriptase. cccDNA, Covalently closed circular DNA.

a cis-acting encapsidation signal recently identified near DRl (Bartenschlager et al., 1990; Hirsch et al., 1990, 1991;Junker-Niepman et al., 1990; Ou et al., 1990; Roychoudhury et al., 1991) 2. Initiation of DNA synthesis at a DR1 repeat, probably in the 3' end

of pregenome RNA, using as a primer the N-terminal portion of the P gene polyprotein product that covalent binds to the 5' extremity of the growing minus-strand DNA (Molnar-Kimber et al., 1983, 1984; Bartenschlager and Schaller, 1988) 3. Degradation of the RNA template by the viral RNase H activity while complementary DNA is produced (Radziwill et al., 1990) 4. Initiation of plus-strand DNA synthesis at the DR2 repeat, using as

H B V A N D HEPATOCELLULAR CARCINOMA

191

a primer a capped oligoribonucleotide of 17 bases derived from the 5’ end of the pregenome RNA (Lien et al., 1986, 1987), and elongation of the plus-strand DNA by the viral DNA polymerase. Nucleocapsids containing replicati,ve intermediates at different stages of maturation, even as early as during synthesis of the minus-strand DNA, are then coated and excreted from infected cells, as indicated by the variable length of the short viral strand DNA and the presence of RNA/DNA duplexes in extracellular virions (Miller et al., 1984; Scotto et al., 1985). Finally, amplification of cccDNA through an intracellular process involving transport of newly synthesized viral intermediates to the nucleus allows the establishment of a pool of transcriptional templates in persistently infected cells (Tuttleman et al., 1986a). The viral pre-SIS proteins participate in the control of cccDN A copy numbers in persistent hepadnaviral infections (Summers et al., 1990, 1991).

C. VIRIONASSEMBLY In many chronic hepadnavirus infections, liver cells produce huge amounts of empty viral envelopes, that are assembled in spherical or tubular particles with a diameter of 22 nm and released into the bloodstream through the lumen of the endoplasmic reticulum. HBsAg particles are made of major and middle surface proteins (encoded by the S gene and the pre-S2IS region) embedded in cellular lipids in glycosylated and unglycosylated forms. They differ mainly from the envelope of infectious virions in the almost complete absence of large surface glycoproteins, encoded by the pre-S 1lpre-S2lS region (reviewed by Tiollais et al., 1985). Excretion through the membrane bilayer is governed by the major S protein, which contains three hydrophobic domains and topogenic signal sequences for directing transmembrane orientation and translocation of virions and HBsAg particles across the bilayer (Bruss and Ganem, 1991; Eble et al., 1986, 1987). The hypothetical transmembrane configuration of the middle S protein is shown in Fig. 5. Only the major S protein is required to form 22-nm particles, whereas all three envelope proteins are necessary for production of infectious virions (Persing et al., 1985; Ueda et al., 1991). Several lines of evidence indicate that the pre-S1 polypeptide plays a predominant role in the secretion of Dane particles and in the control of cccDNA amplification during persistent infections. Mutant viruses that do not produce the large S protein are unable to form infectious virions, whereas cccDNA and core particles carrying viral DNA accumulate intracellularly (Summers et al., 1991; Ueda et al., 1991). These mutants have probably lost

192

MARIE ANNICK BUENDIA

.OW.

= ER-Signal

FIG. 5. Hypothetical conformation of the middle protein (encoded by the preS2IS region) in the endoplasmic reticulum bilayer. Truncation of this protein at residues 77 to 210 generates a novel transcriptional trans-activating activity. (From A. Kekule, Ph.D. thesis.)

the ability to envelope the newly synthesized nucleocapsids. Whether the absence of large S proteins modifies the conformation of budding envelope shells, preventing the envelope from encircling core particles, or whether unique sequences in the pre-Sl domain interact with the viral core has not been determined. It has been shown that retention of the large S protein in the endoplasmic reticulum is governed by signals within the myristilated amino-terminal pre-S1 domain (Kuroki et al., 1989; Persing et al., 1987; Prange et al., 1991). In the absence of a large excess of major and middle S proteins, the large S protein is retained in the endoplasmic reticulum and blocks efficient secretion of HBsAg particles (Chisari et al., 1986; McLachlan et al., 1987; Ou and Rutter, 1987; Persing et al., 1986; Standring et al., 1986). Furthermore, secretion of major and middle S proteins is inhibited in cells producing more than 20% of large S protein (Molnar-Kimber et al., 1988).Therefore, a tight regulation of the relative amounts of the three envelope proteins appears to be an important feature in viral assembly and in control of persistent hepadnaviral infections.

D. REGULATED EXPRESSION OF VIRALGENES In productive hepadnavirus infections, a strict balance in the amount of individual viral gene products is necessary for active viral replication

HBV AND HEPATOCELLULAR CARCINOMA

193

and release of infectious virions as well as for survival of infected host cells. Despite the small size of the HBV genome and a very compact organization of coding sequences, the expression of the different viral genes is subjected to a complex regulation at various levels, both transcriptionally and posttranscriptionally. Each individual HBV gene is controlled by an independent set of regulatory signals, probably acting in cooperation with HBV elements that coordinate the relative level of viral gene expression. All HBV regulatory signals are contained within coding sequences, and for some of them in regions encoding two different polypeptides from overlapping reading frames. In chronically infected livers and in cell lines supporting active viral replication, two major HBV transcripts of molecular size 3.5 and 2.1 kb are produced from cccDNA template at roughly similar levels (Cattaneo et al., 1984; Gough, 1983) and minor 2.4- and 0.8-kb transcripts have been described (Ou and Rutter, 1985; Siddiqui et al., 1986, 1987; Treinin and Laub, 1987). These polyadenylated RNAs are colinear with the viral genome and complementary to the DNA minus strand. They have heterogeneous 5‘ ends and a common termination site in the core gene, downstream of a variant polyadenylation signal shared by all hepadnaviruses. The transcription patterns of WHV and GSHV in chronically infected livers are strikingly similar, whereas that of DHBV differs mainly by higher levels of 2.4-kb RNA (Buscher et al., 1985; Enders et al., 1985; Moroy et al., 1985). Two distinct 3.5-kb RNAs serving different functions start around the pre-C initiator codon (see Fig. 2), the longer one specifying the pre-C/C polypeptide (HBeAg) and the shorter the nucleocapsid protein (HBcAg) and the viral P gene products. The shorter RNA species, designated “pregenome,” is also implicated in the viral replicative process (Seeger et al., 1987; Will et al., 1987). The 3.5-kb RNAs are longer than genome size by about 120 nucleotides, and their synthesis requires RNA polymerase reading through the polyadenylation signal without cleavage at the first passage, a feature probably associated with a strong secondary structure at the 5’ end of the nascent transcript as in other retroid elements (Russnak and Ganem, 1990). In contrast with that of retroviruses, translation of the hepadnaviral polymerase from the 3.5-kb pregenome RNA does not involve ribosomal frame shifting, but more probably de n o w translational initiation (L. J. Chang et al., 1989, 1990; Ou et al., 1990; Schlicht et al., 1989).The liver-specific promoter for 3.5kb RNAs is made of a basal AT-rich element and of upstream promoter sequences that bind different nuclear factors (Lopez-Cabrera et al., 1990; Yaginuma and Koike, 1989). Two enhancer elements stimulate transcription from the core/pregenome promoter: EN I, positioned

194

MARIE ANNICK BUENDIA

about 450 bp upstream of the core promoter and EN 11, located in the X gene coding region (Shaul et at., 1985; Yee, 1989) (see Figure 3). The HBV enhancers I and I1 bind multiple liver-specific factors as well as ubiquitous transcriptional activators and exhibit a preferred activity in differentiated hepatocytes or hepatoma-derived cell lines (Ben-Levy et al., 1989; Jameel and Siddiqui, 1986; Lopez-Cabrera et al., 1991; Shaul and Ben-Levy, 1987; Y. Wang et al., 1990; Yuh and Ting, 1990). Enhancer I stimulates expression of all viral genes (Antonucci and Rutter, 1989; Hu and Siddiqui, 1991). A strong enhancer element has also been identified upstream of the pregenome start site of DHBV (CrescenzoChaigne et al., 1991). Recently, two spliced RNA species of about 2 kb, which share 5'- and 3'-terminal sequences with the 3.5-kb RNAs and present different coding capacities, have been described (Chen et al., 1989; Su et al., 1989; Suzuki et at., 1989). These relatively abundant transcripts appear to be dispensible for HBV replication, but they are detected in most natural infections, suggesting that they may serve some biological function(s) (Wu et al., 1991). One of the spliced RNAs has been shown to be efficiently packaged and reverse transcribed in vivo, giving rise to defective viral particles (Terre et al., 1991). The major 2. l-kb and minor 2.4-kb RNAs direct the synthesis of the three envelope proteins (Cattaneo et al., 1983).Expression of the hepatitis B surface antigen gene has been extensively studied in cultured cells (Pourcel et al., 1982; Roossinck et al., 1986; Siddiqui et al., 1986; Standring et al., 1984) and in transgenic mouse models (Babinet et al., 1985; Burk et al., 1988; DeLoia et al., 1989).It presents one of the most striking examples of the efficient organization of the compact HBV genome, in that it involves a complex set of regulatory elements embedded in two overlapping coding sequences. The major and middle surface proteins are synthesized from the major 2.1-kb transcripts, which initiate heterogeneously 5' and 3' of the pre-S2 ATG codon. The longer 2.1-kb species encode both the major and middle S proteins, but the shorter one encodes only the major S. The levels of expression of the pre-S2/S and S genes are controlled by a dual promoter located in a 200-bp sequence of the pre-S2 region, and by differential use of the pre-S2 and S translational initiator codons. The pre-S2/S promoter, highly active in a variety of cultured hepatoma cells, has no TATA motif and shares homology with the simian virus 40 (SV40) origin-late promoter (Cattaneo et al., 1983). Transcription from this promoter is regulated by a complex interplay between positive and negative elements and is stimulated by the downstream glucocorticoid responsive element (GRE) and by the viral enhancer I (Bulla and Siddiqui, 1988; H. K. Chang et al., 1987; De-

HBV AND HEPATOCELLULAR CARCINOMA

195

Medina et al., 1988; Farza et al., 1988; Raney et al., 1989, 1991; Shaul et al., 198613; Tur-Kaspa et al., 1986). In contrast, the pre-S1 promoter is markedly weaker and shows a strict cell-type specificity attributed to the binding of liver-specific transcription factors, like HNF- 1. This might account partially for the hepatotropism of HBV (Bulla and Siddiqui, 1989; Chang and Ting, 1989; H. K. Chang et al., 1989; Courtois et al., 1988; Nakao et al., 1989; Raney et al., 1990; Zhou and Yen, 1991). Finally, the production of a 0.9-kb mRNA encoding the viral X gene product is better seen in cultured cells transfected with viral subgenomic fragments than in natural HBV infections, indicating that the HBX promoter might be down regulated during the viral replicative process, which is associated with very low concentrations of X protein in infected human livers (Asselsbergs et al., 1986; Gough and Murray, 1982; Saito et al., 1986; Treinin and Laub, 1987). V. Potential Oncogenic Properties of Viral Proteins

It has been admitted for a long time that prolonged expression of viral genes might have no direct cytotoxic effects on the infected hepatocytes. Transfection of cultured cells with HBV DNA has not been usually associated with tumorigenic conversion, and most transgenic lines of mice bearing the full-length HBV genome or subgenomic constructs never show any sign of liver cell injury. However, recent studies in particular experimental systems have indicated that abnormal overexpression of different viral proteins, including the surface proteins and the X gene product in native or modified forms, might play a part in malignant transformation of infected hepatocytes. A. SURFACE GLYCOPROTEINS In natural HBV infections, the production of infectious virions and HBsAg particles depends on a tight regulation of the relative levels of the three envelope glycoproteins, as shown in Section IV,C. Neither liver lesions nor HCC have been observed in any of the published transgenic lineages that produce the middle and major surface proteins from HBVderived regulatory sequences (Araki et al., 1989; Babinet et al., 1985; Burk et al., 1988; Farza et al., 1988). However, it has been shown that the appearance and rate of production of preneoplasic nodules and primary tumors following carcinogen administration are slightly increased in HBsAg-positive transgenic mouse livers as compared to negative littermates, suggesting that HBsAg expression might enhance the effects of the hepatocarcinogens (Dragani et al., 1989).When the endogenous preS1 promoter is replaced by an exogenous promoter (the metallothionein

196

MARIE ANNICK BUENDIA

or the albumin promoter) the production of roughly equimolar ratios of large S protein with respect to middle and major S leads to intracellular accumulation of nonsecretable filamentous envelope particles within the endoplasmic reticulum of transgenic mouse hepatocytes (Chisari et al., 1986, 1987). This provokes histological and ultrastructural features of “ground-glass” hepatocytes, which have been described in some cases of chronic human liver diseases and are considered to be typical of chronic hepatitis B (Gerber et al., 1974; Stein et al., 1972), ultimately killing the cells. I n the transgenic mouse lineages, mild persistent hepatitis was followed by the development of regenerative nodules and eventually HCCs by 12 months of age (Chisari et al., 1989b; Dunsford et al., 1990). T h e preneoplastic nodules and tumors display a marked reduction in transgene expression, suggesting that hepatocytes that express low levels of the large S polypeptide would have a selective survival advantage. Exogenous, chemical cofactors are not required for tumorigenic induction in this model, but exposure of adult transgenic mice to hepatocarcinogens produced more rapid and extensive development of preneoplastic lesions and HCC, under conditions that do not alter the liver morphology of nontransgenic controls (Teeter et al., 1990). These data show that inappropriate expression of the large S protein has the potential to be directly cytotoxic to the hepatocyte and may initiate a cascade of events that ultimately progress to malignant transformation, although the molecular mechanism connecting viral and host factors in this process has not been elucidated. Studies of integrated HBV sequences in human liver tumors have also suggested a possible role for abnormal expression of rearranged viral S genes in HCC development. It has been shown that deletion of the carboxy-terminal region of the S protein generates a novel transcriptional trans-activation activity (Caselmann et al., 1990; Kekul6 et al., 1990) (see Fig. 5). Integrated HBV sequences from a human tumor and a hepatoma-derived cell line, as well as different constructs bearing similarly truncated pre-S2/S sequences, can stimulate the SV40 promoter in transient transfection assays; trans-activation occurs at the transcriptional level and is dependent on the SV40 enhancer. The c-myc P2 promoter is also activated in trans. These findings support the hypothesis that accidental 3’ truncation of integrated pre-S2/S genes could be a causative factor in HBV-associated oncogenesis. B. HBx: A TRANSCRIPTIONAL TRANS-ACTIVATOR

T h e smallest of the four HBV open reading frames was initially designated “ X because it was unclear at that time (and remained as such for

HBV A N D HEPATOCELLULAR CARCINOMA

197

several years) that it might encode a protein produced during HBV infection, and that this potential protein might have a predictable function in the viral life cycle (Galibert et al., 1979). Examination of the nucleotide and amino acid sequences of HBx has led to the prediction of a regulatory function for the deduced X polypeptide and has revealed a codon usage similar to that used in eukaryotic cell genes, suggesting that X might have been transduced by the HBV genome (Miller and Robinson, 1986). The strong conservation of X sequences among HBV subtypes (Lo et al., 1988) and the presence of homologous reading frames in the WHV and GSHV genomes suggest that their products may be of importance for viral replication. The absence of an X open reading frame in the avian hepadnavirus genomes, despite some weak sequence homologies in the DHBV core gene (Feitelson and Miller, 1988), raises questions about an essential contribution of this protein to the viral life cycle. Evidence for expression of the HBV X gene has been obtained by Moriarty et al. (1985) and by.Kay et al. (1985),who reported that the sera of HBV-related HCC patients recognize synthetic peptides made on X sequences. Expression of the X reading frame in prokaryotic and eukaryotic cells, using various vectors, has allowed the identification a 16.5kDa polypeptide that reacts with serum samples from a number of HBVinfected individuals (Elfassi et al., 1986; Meyers et al., 1986; Pfaff et al., 1987; Schek et al., 1991b). Anti-HBx antibodies have been detected in a minor proportion of acutely infected patients, about 3-4 weeks after the onset of clinical signs, and more frequently in chronic HBsAg carriers showing markers of active viral replication. Very few patients show antiHBx antibodies after seroconversion to anti-HBs and at the time HCC develop (Levrero et al., 1991). However, conflicting results have been obtained regarding the association of anti-HBx antibodies with other viral markers and with HCC. These problems may be related to the weak antigenicity of HBx or to its sequestration into cellular compartments that render it inaccessible to the host immune system. The hepatitis B x antigen (HBxAg) has been detected in livers and sera of HBsAg carriers and has been correlated with ongoing viral replication and chronic liver disease (Haruna et al., 1991; Levrero et al., 1990; Wang et al., 1991).The X protein is localized mainly in the cytoplasm of in vivo-infected cells, at or near the plasma membrane and at the nuclear periphery (Levrero et al., 1990; Vitvitski et al., 1988; Wang et al., 1991; Zentgraf et al., 1990). The X protein has been detected in the nuclear compartment only in transfected cell lines (Hohne et al., 1990; Seifer et al., 1990). The recent finding that the X gene product can trans-activate transcription from a number of HBV and heterologous promoters is of

198

MARIE ANNICK BUENDIA

considerable importance in defining its role in viral replication and in pathogenesis. T h e X trans-activator stimulates transcription from the the C, S, or X promoters coupled to HBV enhancer (Colgrove et al., 1989; Raney et al., 1990; Spandau and Lee, 1988), from heterologous viral promoters like the SV40 early promoter, the HSV-thymidine kinase (TK) promoter and the HIV-1 long terminal repeat (LTR) (Spandau and Lee, 1988; Twu and Robinson, 1989; Twu et al., 1989; Zahm et al., 1988), and from cellular promoters including @-interferon, HLA-DR, and c - m y promoters (Hu et al., 1990; Twu and Schloemer, 198’7; A. KekukC, personal communication). Trans-activation of RNA polymerase I11 promoters has also been reported (Aufiero and Schneider, 1990). Conflicting results on X-mediated trans-activation of several cis-acting regulatory sequences in different transfected cell lines might be related to protein interactions between X and cellular factors. Indeed, the X protein has been shown to interact, directly or indirectly, with transcriptional activators such as AP1, AP2, CREB, ATF2, NFKBand possibly Spl (Maguire et al., 1991; Seto et al., 1990; Twu et al., 1989; A. Kekule, personal communication). T h e subcellular localization of the X polypeptide makes it unlikely that it might directly bind to its DNA target sequences; activation of cellular and viral genes, which occurs at the level of transcription (Colgrove et al., 1989), might be mediated via the protein kinase C (PKC) signal transduction pathway, as for tumor promoters (A. Kekuke, personal communication). It has been shown that HBx can be phosphorylated in vivo and that it displays an intrinsic serinel threonine protein kinase activity (Schek et al., 1991a; Wu et al., 1990), two features also shared by a number of proteins implicated in intracellular signalling pathways. HBx might therefore directly activate cellular transcription factors. Although present at very low levels in chronically infected livers, the X protein has been shown to stimulate the production of viral particles in transient transfection assays (Yaginuma et al., 1987). However, it has not been established that the expression of cellular genes showing Xresponsive promoters is stimulated in HBV-infected hepatocytes in vivo. More clues to the possible role of HBx in HBV-associated pathogenesis have been provided by three recent lines of studies, including both in nitro and in vivo studies and direct analysis of human liver and HCC biopsy samples. It has been shown that high levels of X expression may induce malignant transformation of certain cultured cells, like the NIH 3T3 cell line (Shirakata et al., 1989), immortalized hepatocytes expressing the SV40 large tumor antigen (Hohne et al., 1990), and primary embryo fibroblasts cotransfected with X and ras expression vectors (A. Kekuld, per-

HBV AND HEPATOCELLULAR CARCINOMA

199

sonal communication). Studies of transgenic mice carrying the X reading frame controlled by its natural HBV enhancer/promoter sequences or by heterologous liver-specific promoters have given rise to conflicting results. In two lines of mice derived from the inbred CDl strain, carrying a 1.15-kb HBV fragment (spanning the enhancer, the complete X coding region, and the polyadenylation signal), the development of preneoplastic lesions has been observed in the liver, followed by malignant carcinomas at 8-10 months of age (Kim et al., 1991). In contrast, other transgenes, in which the X coding domain was placed under the control of the a-l-antitrypsin or the antithrombin-3 regulatory region, failed to induce serious liver damage in ICR x B6C3F1 or C57BL/6 X SJL/J transgenic animals, although X mRNAs were detected in liver tissues (Lee et al., 1990; P. Briand, personal communication). Analysis of integrated viral sequences in tumor DNA has shed new light on one of the mechanisms leading to overexpression of HBx in chronically infected livers and in HCCs. It has been shown that HBV sequences are frequently interrupted between the viral direct repeats DR1 and DR2 on integration into host cell DNA (see Section VI,A) and that overproduction of hybrid viral/host transcripts may result from HBV DNA integration in a hepatoma cell line (Freytag von Loringhoven et al., 1985; Ou and Rutter, 1985). T h e presence of viral/host transcripts containing a 3'truncated version of the X coding region fused with flanking cellular sequences and retaining trans-activating capacity was first described in a human HCC (Wollersheim et al., 1988). Moreover, enhanced trans-activating capacity of the integrated X gene product has been related to the substitution of viral carboxy-terminal residues by cellular amino acids (Koshy and Wells, 1991). Trans-activating ability of similarly truncated X products made from fusion of integrated HBV sequences with adjacent cell DNA has also been shown in many chronic hepatitis tissues (Takada and Koike, 1990). This suggests that the integrated X gene might be essential for maintaining the tumor phenotype that develops at the early stages of carcinogenesis. Consistent with this model, viral/host junctions have been mapped in the carboxy-terminal region of X in a majority of human HCCs (Nagaya et al., 1987; Shih et al., 1987). Further studies are now necessary to better delineate the contribution of X gene product to malignant transformation in persistent HBV infections. VI. Integrated State of Viral DNA in Chronic Infections and Hepatocellular Carcinoma

The hepadnavirus replication pathway in infected cells has been shown to take place within nuclear and cytoplasmic compartments (see

200

MARIE ANNICK BUENDIA

Section IV,B) and more recent studies have indicated that it can be entirely cytoplasmic (Zhou and Standring, 1991). Unlike retroviruses, this process does not require viral DNA integration in the host cell genome. However, hepadnaviruses share with other retroelements of common evolutionary origin-transposons and retroviruses (Xiong and Eickbush, 1990)-the ability to integrate their DNA into cellular chromosomes. The molecular events leading to the invasion of cell DNA by hepadnaviral DNA have not been fully elucidated. The main question is whether viral integrations might play a part in the virally induced transformation process, either by conferring a selective growth advantage to targeted cells, leading to the onset of preneoplastic nodules, or by providing an additional step in tumor progression. Whereas insertional activation of protooncogenes now emerges as a common event in WHV-induced woodchuck HCC, a related mechanism has been observed only in rare examples of human HCCs, in which different activities for integrated HBV sequences have been proposed. A. INTEGRATED SEQUENCES: PHYSICAL AND FUNCTIONAL ASPECTS Analyses of Southern profiles have allowed the identification and primary characterization of integrated HBV sequences in established hepatorna cell lines and in about 80% of human HCCs (BrCchot et al., 1980, 1981b, 1982; Chakraborty et al., 1980; Edman et al., 1980; Koshy et al., 1981; Shafritz et al., 1981; for reviews, see Tiollais et al., 1985; Matsubara and Tokino, 1990). Hepatitis B virus DNA integrations occur at early stages in natural acute infections and in experimental infections of cultured cells (Scotto et al., 1983; Lugassy et al., 1987; Ochiya et al., 1989; Yaginuma et al., 1987). As a result of multiple integrations in chronic hepatitis tissues (Boender et al., 1985; Brechot et al., 1981b; Shafritz et al., 1981; Tanaka et al., 1988),integrated HBV sequences have been detected in most HBV-related HCCs that arise from clonal outgrowth of one or a few transformed liver cells (see Matsubara and Tokino, 1990). Single HBV insertions are common in childhood HCCs but are rather uncommon later in life, suggesting that multiple integrations occurring during the course of long-standing HBV infections might accumulate within single cells (Chang et al., 1991), as also indicated by sequence divergence among HBV inserts in the same tumor (Imai et al., 1987). Integrated WHV sequences have been similarly detected in woodchuck liver and in a majority of woodchuck HCCs (Ogston et al., 1982; Rogler and Summers, 1984; Hsu et al., 1990), but

HBV AND HEPATOCELLULAR CARCINOMA

20 1

viral integrations appear to be less frequent in GSHV- and DHBV-related tumors (Marion et al., 1986; Yokosuka et al., 1985; L. Cova, personal communication). Studies of the organization of cloned HBV inserts in liver tissues and HCCs have shown that HBV sequences are fragmented and rearranged and that integration and recombination sites are dispersed over the viral genome, indicating that HBV integration does not occur through a unique mechanism, as in the case of other retroelements and retroviruses (Dejean et al., 1984; Koch et al., 1984b; Shaul et al., 1984; Ziemer et al., 1985). T h e absence of complete genomes in virtually all HBV inserts, which consist either of linear subgenomic fragments or of rearranged fragments in different orientations, indicates that these integrated sequences cannot serve as a template for viral replication. Rearrangements of integrated HBV sequences may take place during the integration process as well as after the formation of viral inserts (Mizusawa et al., 1985; Nagaya et al., 1987; Tokino et al., 1987). Integrated forms made of a continuous genome or subgenomic fragment, which are frequent in HCC and hepatitis tissues from children (Yaginuma et al., 1987), are believed to represent primary products of integration. They are of particular interest in the study of the molecular mechanisms responsible for HBV DNA integration. Highly preferred integration sites have been mapped in the HBV genome within the “cohesive ends” region, that lies between two 1 l-bp direct repeats (DRl and DR2) highly conserved among hepadnaviruses (Koshy et al.. 1983; Nagaya et al., 1987). A narrow region encompassing DR1 has been shown to be particularly prone to recombination (Hino et al., 1989; Nakamura et al., 1988; Shih et al., 1987; Yaginuma et al., 1987). This region coincides with a short terminal redundancy of the minusstrand DNA, which confers a triple-stranded structure to the circular viral genome (Seeger et al., 1986; Will et al., 1987). Integration sites are tightly clustered both at the 5‘ and 3’ ends of minus-strand DNA, suggesting that replication intermediates and specially relaxed circular DNA might be preferential preintegration substrates (Nagaya et al., 1987; Shih Pt al., 1987). Invasion of cellular DNA by single-stranded HBV DNA, using mainly free 3‘ ends, might take place through a mechanism of illegitimate recombination, also suggested by frequent patch homology between HBV and cellular sequences at the recombination break points (Matsubara and Tokino, 1990). Although different minor changes in flanking cellular DNA have been associated with viral integration (both microdeletions and short duplications have been described) (Berger and Shaul, 1987; Dejean et al., 1986; Hino et al., 1989; Nakamura et al., 1988; Yaginuma et al., 1985), more precise mechanisms have been proposed. T h e recombination-proficient region spanning

202

MARIE ANNICK BUENDIA

DRl is located close to a U5-like sequence highly conserved between hepadnaviruses, suggesting that sequences necessary for precise recombination with cellular DNA have been retained from a common ancestor with retroviruses, despite the absence of a gene coding for an integrase in the HBV genome (Miller and Robinson, 1986). A specific mechanism based on sequence polarity in DR1 and DR2 and on deletion of 1-2 bp at the 5' end of the viral DRs has been proposed (Dejean et al., 1984), but collection of data on the structure of about 40 HBV inserts has not confirmed this model. Mapping of a set of preferred topoisomerase I (TopoI) sites near DR1 and DR2 and in nitro studies of WHV DNA integration into cloned cellular DNA have sustained the hypothesis that TopoI might promote illegitimate recombination of hepadnavirus DNA in viuo (Wang and Rogler, 1991). A role for the eukaryotic enzyme TopoI has been proposed not only in nonhomologous recombinations that lead to integration in other DNA virus systems (Bullock et al., 1984, 1985) but also in the life cycle of different retroviruses (Priel et al., 1991). The mechanisms underlying HBV DNA integration remain to be fully identified, but it seems probable that similar events lead to the integration of all hepadnaviruses, although only a limited number of WHV and DHBV insertions and no integrated forms of GSHV have been analyzed until now (Fourel et al., 1990; Hsu et al., 1988; Imazeki et al., 1988; Ogston et al., 1982; Rogler and Summers, 1984). As a consequence of the viral integration process, sequences of the S and X genes and of the enhancer I element are almost systematically present in HBV inserts, whereas those of the C gene are less frequently represented. It has been shown that the pre-SP/S promoter was transcriptionally active in its integrated form in human and woodchuck HCCs (Caselmann et al., 1990; Freytag von Loringhoven et al., 1985; Ou and Rutter, 1985; Y. Wei and M. A. Buendia, unpublished results) and that HBsAg might be produced from viral inserts (Dejean et al., 1984; Zhou et al., 1987). Highly rearranged HBV inserts show virus junctions scattered throughout the viral genome, and in some of them, recombination break points have been mapped in the S coding region (Nagaya et al., 1987). It has been recently shown that truncation of the S gene between residues 77 and 22 1 confers a transcriptional activation activity to the mutated pre-SPIS products (Caselmann et al., 1990; Kekule et al., 1990). T h e shorter pre-SP/S protein lacks carboxy-terminal signals for translocation through the endoplasmic reticulum membrane and should be retained in the bilayer. Activation of the c-my oncogene promoter, demonstrated in in vitro assays, might result from an indirect transacting action of the truncated viral proteins (Kekule et al., 1990). Whether this o r some related mechanism participates in liver cell trans-

HBV A N D HEPATOCELLULAR CARCINOMA

203

formation remains to be determined (see Section V,A). Other studies have shown that an important percentage of viral junctions is localized in the carboxy-terminal part of the viral X gene, predicting a fusion of the X open reading frame to flanking cellular sequences in a way that might preserve the functional capacity of the X trans-activator. Evidence for transcriptional activity at integrated X sequences has been provided in tumors and chronically infected livers (Hilger et al., 1991; Miyaki et al., 1968; Takada and Koike, 1990; Wollesheim et al., 1988) and might be correlated with the detection of HBxAg in a number of human HCCs (Moriarty et al., 1985; Kay et al., 1985). It has also been shown that a number of viral X/cell fusion peptides harbor transcriptional activation activity. The contribution of downstream cellular sequences to activated expression and/or to enhanced trans-activating capacities of the integrated HBV sequences has been suggested in two cases (Freytag von Lorinhoven et al., 1985; Wollesheim et al., 1988). These data indicate that abnormal expression of integrated and truncated X gene might play a part in HBV-associated oncogenesis, by deregulating the normal expression of cellular genes in trans (see Section V,B). B. CELLULAR TARGETS FOR VIRALINTEGRATION IN HUMAN HEPATOCELLULAR CARCINOMA Studies of different viral insertions in many human HCCs have revealed that integration can take place at multiple sites on various chromosomes (25 insertion sites have been mapped on 16 different human chromosomes) (Matsubara and Tokino, 1990). These studies failed to demonstrate the presence of any known dominant oncogene or tumor suppressor gene in the immediate vicinity of any integration site. It has been reasoned that integration of HBV DNA occurs at random in the human genome and that it has no direct mutagenic effect on growth control genes in most cases. However, contrary to a widely held opinion, integration of retroelements and viruses might not be entirely random; it has been shown that the possible sites for retroviral integration in eukaryotic DNA are numerous but not unlimited (about 500-1000) (Shih et al., 1988). It has been proposed that simple repetitive elements are hot spots for HBV insertion in the human genome (Berger and Shaul, 1987). Indeed, Alu-type repeats, minisatellite-like, satellite 111, or VNTR sequences have frequently been identified near HBV insertion sites (Berger and Shaul, 1987; Nagaya et al., 1987; Shaul et al., 1986a), suggesting that chromosomal regions accessible to specific families of mobile repeated sequences are also preferential targets for HBV insertion. A small cellular DNA compartment (H3) characterized by a base

204

MARIE ANNICK BUENDIA

composition close to that of HBV DNA and a high concentration of Alu repeats has been designated as a major target for stable HBV integration (Zerial et al., 1986). It has been shown that HBV DNA integration may enhance chromosomal instability: in many tumors, large inverted duplications, deletions, amplifications, or chromosomal translocations have been associated with HBV insertions, suggesting that this process may function as a random mutagen, promoting chromosomal defects in hepatocytes (Hatada et al., 1988; Hino et al., 1986; Koch et al., 1984a; Mizusawa et al., 1985; Rogler et al., 1985; Tokino et al., 1987; Yaginuma et al., 1985). It has also been shown that HBV DNA promotes homologous recombination at a distance from the insertion site (Hino et al., 1991). However, a role for most of these chromosomal abnormalities has not been assigned as yet, although in a few cases the p53 or hstl loci have been altered as a sequel to HBV integration in the same chromosomal region (Hatada et al., 1988; Slagle et al., 1991; Zhou et al., 1988). Evidence for a direct cis-acting promoter insertion mechanism has been provided in two independent HCCs (Dejean et al., 1986; de The et al., 1987; Dejean and de The, 1990; J. Wang et al., 1990). These investigators have analyzed early tumors that developed in noncirrhotic livers from clonal proliferation of a cell containing a single specific viral integration. In one case, the HBV insertion occurred in an exon of the retinoic acid receptor P gene (RARP) and fused the amino-terminal domain of the viral pre-S1 gene to the DNA-binding and hormonebinding domains of RARP (Dejean et al., 1986). Retinoic acid and retinoids are vitamin A-derived substances that have striking effects on differentiation and proliferation in a large variety of systems (Brockes, 1990). Retinoic acid receptors are members of the steroid thyroid hormone receptors family, which includes the c-er6A protooncogene (Graf and Berg, 1983). Recently, the chromosomal translocation t( 15;17) fusing the RARa gene to a cellular gene termed PML has been implicated in acute promyelocytic leukemias (de The et al., 1990, 1991). In the human HCC, it seems most probable that inappropriate activation of RARP resulted in expression of a chimeric HBV/RARP protein at greater levels than that of native RARP protein, and participated in the tumorigenic process. In the second case, HBV sequences were found to be integrated in an intron of the human cyclin A gene, resulting in the production of spliced HBV/cyclin A fusion mRNAs initiated at the preS2/S promoter (J. Wang et al., 1990; C. Brechot, personal communication). In the deduced polypeptide, the amino-terminal domain of cyclin A, a target for proteolytic degradation of cyclin A at the end of the M phase, was replaced by residues of the viral S region. Cyclins are impor-

H B V AND HEPATOCELLULAR CARCINOMA

205

tant in the control of cell division, and disruption of the cyclin A gene by viral insertion probably contributed to oncogenesis. It should be interesting to further analyze other tumors that, like in the two cases reported above, are at an early stage of cancer development to better delineate the relative contribution of insertional mutagenesis and other mechanisms, associated or not with HBV DNA insertion, in the virally induced transformation pathway. C. INSERTIONAL ACTIVATION OF myc FAMILY GENES I N WOODCHUCK HEPATOCELLULAR CARCINOMA T h e availability of naturally occurring animal models for HBVinduced liver disease and cancer has been largely exploited for a better understanding of the viral/host interactions, as shown in Section I11 and IV. In the WHV/woodchuck model particularly, the natural history of viral infections, the presence and state of viral DNA, and the patterns of viral gene expression have been extensively investigated. Experimental inoculation of newborn woodchucks with infectious virions has given conclusive information on the oncogenic activity of WHV: this virus now appears to be the most potent inducer of liver cancer among the hepadnavirus group (Popper et al., 1987b; Seeger et al., 1991). A search for transcriptional activation of already known protooncogenes and for viral insertion sites in tumor cell DNA has revealed that WHV acts as an insertional mutagen, activating myc family genes (c-myc or N-myc) in more than one-half of the tumors examined (Fourel et al., 1990; Hsu et al., 1988; Moroy et al., 1986; Y. Wei, A. Ponzetto, P. Tiollais, and M. A. Buendia, unpublished results). Analysis of the mutated c-myc alleles in two individual tumors has shown integration of WHV sequences in the vicinity of the c-myc coding domain, either 5' of the first exon or in the 3'-untranslated region (Hsu et al., 1988). Deregulated expression of the oncogene driven by its normal promoters resulted from deletion or displacement of c-myc regulatory regions known to exert a negative effect on c-myc expression, and their replacement by viral sequences encompassing the enhancer I element. Such a mechanism is highly reminiscent of that previously reported for c-myc activation in murine T cell lymphomas induced by murine leukemia viruses (MuLV) (Corcoran et al., 1984; Selten et al., 1984). Evidence for a direct role of WHV DNA integration into c-myc in hepatocyte transformation has been provided by the development of hepatocellular carcinoma in transgenic lines of mice bearing WHV and myc sequences from the mutated allele of a woodchuck HCC (J. Etiemble, C. Babinet, P. Tiollais, and M. A. Buendia, unpublished results). Transient ex-

206

MARIE ANNICK BUENDIA

pression of the transgene in liver cells after birth was correlated with the onset of primary liver tumors at 5-12 months of age in 100% of cases. More frequently observed was the insertional activation of N-myc genes. In contrast with human and mouse, the woodchuck genome contains two distinct N-myc genes: N-myc 1, the homolog of known mammalian N-myc genes similarly organized into three exons, and N-myc 2, a functional processed pseudogene or “retroposon,”which has retained extensive coding and transforming homology with parental N-myc (Fourel et al., 1990). In woodchuck HCCs, N-myc 2 represents by far the most frequent target for WHV DNA integrations. As shown in Fig. 6, in about one-half of cases viral inserts were detected either upstream of the gene or in a short sequence of the 3’-untranslated region, also identified as a unique hot spot for retroviral insertions into the murine N-myc gene in T cell lymphomas (Dolcetti et al., 1989; Setoguchi et al., 1989; van Lohuizen et al., 1989). Activated expression of the N-myc 2 retroposed oncogene, frequently correlated with overexpression of c-fos and c-jun,

N-myc2

t

If

t 5

9

7

N-mycl

t 1 c-myc

t 1

t 1

t 1

FIG. 6. Insertion sites of WHV DNA in the c-myc and N-my genes in 49 woodchuck HCCs. The processed N-my2 oncogene, generated by retrotransposition of the parental woodchuck N-myc gene (N-mycl), is specific to the Sciuridae family of rodents. The number of individual WHV insertions found in each region of the myc genes in different woodchuck HCCs is indicated in bold characters. Insertional activation of myc family genes was observed in about 50% of the tumors analyzed.

HBV A N D HEPATOCELLULAR CARCINOMA

207

was observed in a large majority of woodchuck HCCs (Hsu et al., 1990). Significantly enhanced expression of the human N-myc, c-fos, or c-jun protooncogenes has been observed only occasionally in HBV-associated HCCs (M. A. Buendia, unpublished results), underlying the importance of host cell factors in the differences observed between the tumorigenic processes induced by hepadnaviruses in humans and woodchucks. Further evidence that myc family genes are predominantly implicated in rodent liver tumors associated with hepadnavirus infection comes from other studies of woodchuck and ground squirrel HCCs. In two independent woodchuck tumors, a genetic rearrangement fusing the coding domain of c-myc with the promoter and 5’-translated sequences of a cellular locus termed ‘‘hcr”has been observed (Etiemble et al., 1989; Moroy et al., 1986, 1989; 0. Hino, personal communication). The colocalization of c-myc and hcr on the same woodchuck chromosome (Y, Mizuno, personal communication) suggests that a large intrachromosomal deletion is responsible for the rearrangement, which apparently does not involve the direct interaction of integrated viral sequences. In a recent study of ground squirrel HCCs, frequent amplifications of c-myc were found in tumor cell DNA (6 of 14 cases examined) and were associated with enhanced expression of the oncogene (C. Transy, G. Fourel, P. Marion, and M. A. Buendia, unpublished results). Integration of GSHV DNA into host cell genome, which occurs only rarely in squirrel tumors, has not been correlated with the observed genetic alterations of c-myc. It is interesting to note that such alterations have also been described in some cases of chemically induced liver tumors in rodents (Chandar et al., 1989; Hayashi et al., 1984; Tashiro et al., 1986). Although amplification of c-myc has been observed on rare occasions in HBV-positive human liver tumors (Trowbridge et al., 1988; M. A. Buendia, unpublished results), there is no experimental demonstration, until now, that deregulated expression of myc genes might be generally associated with HBV-induced tumorigenesis in human livers, by any known cis- or trans-acting mechanism. The strategy used by WHV in liver cell transformation now appears strikingly comparable to that of some nonacute retroviruses, like Moloney murine leukemia virus (MoMuLV), which induce disease (usually leukemias) slowly, emphasizing the described similarities between hepadnaviruses and retroviruses (Wain-Hobson, 1984; Miller and Robinson, 1986). These conclusions ask two different, albeit related, questions. What are the factors driving the oncogenic potential of WHV exclusively toward hepatocytes, as we know that this virus can infect a wide variety of woodchuck tissues? How can the apparent differences in

208

MARIE ANNICK BUENDIA

the strategies of the closely related mammalian hepadnaviruses be explained at the molecular level? To address these issues and identify the genomic variations responsible for such discrepancies, it would probably be instructive to reexamine the retroviral genetic elements influencing disease specificity and latency of different MuLVs (Li et al., 1987; Portis et al., 1991; Thiesen et al., 1988) or the oncogenic properties of human papilloma virus (HPV) strains in different tissues (Miinger et al., 1989; Romanczuk et al., 1991; zur Hausen and Schneider, 1987). VII. Genetic Alterations in HBV-Related Hepatocellular Carcinoma

Different genetic alterations, which cannot be clearly associated to a direct effect of viral infection, have been observed in human HCCs. These somatic changes include allele losses on several chromosomal regions, mutation and activation of cellular genes showing oncogenic potential, and deletion or mutation of a tumor suppressor gene. Search for activated oncogenes using the NIH3T3 cell transformation assay has not been conclusive for most HCC DNAs analyzed. In rare cases, a transforming DNA called “Eca” has been obtained in the transformation assays (Ochiya et al., 1986). This novel oncogene, located on human chromosome 2, is expressed at a proliferative stage in fetal liver, and its activation in liver cancer has not been associated with gross rearrangements of the gene (Matsubara et al., 1987; Shiozawa et al., 1988). Conflicting results have been obtained with human HCC DNA samples using the NIH3T3 cells transformation assay. Barbacid and co-workers have failed to isolate a transforming gene, whereas others report a high incidence of activated N-ras gene (Pulciani et al., 1982; Gu et al., 1986). Indeed, a very low incidence of point mutations in the c-Ha-rm, c-Ki-ras, and N-rm genes has been described in liver tumors (Ogata et al., 1991; Tsuda et al., 1989). Loss of heterozygosity on the distal l p region and on chromosomes 4q, 1 lp, 13q, and 16q appears frequently in human liver tumors, and it has been suggested that these parts of the human genome might contain some genes whose functional loss might be involved in hepatocellular carcinogenesis (Buetow et al., 1989; Pasquinelli et al., 1988; Simon et al., 1991; Tsuda et al., 1990; Wang and Rogler, 1988). Because the largescale chromosomal alterations that arise in cancer cells occur infrequently in normal cells, it is probable that control mechanisms that safeguard chromosomal integrity are abrogated in the development of malignancy (Wright et ul., 1990). These changes might represent secondary events linked to tumor progression and reflect a general property of

HBV A N D HEPATOCELLULAR CARCINOMA

209

transformed cells; whether HBV DNA integration, which has been shown to promote genetic instability (Hino et al., 1991), contributes to these events has not been elucidated. Allele loss of the short arm of chromosome 17, which includes the p53 gene, has been commonly observed in human HCCs and hepatomaderived cell lines (Slagle et al., 1991; Fujimori et al., 1991). Studies of the p53 gene at the DNA, RNA, and protein levels have revealed abnormal structure and expression in most established HCC cell lines; these alterations, including partial deletions or DNA rearrangements and abnormal expression patterns, might not occur as a late in uitro event and were not correlated with integration of HBV DNA (Bressac et al., 1990). Although viral insertions in chromosome 17p have been described in some liver tumors (Hino et al., 1986; Tokino et al., 1987; Zhou et al., 1988), it seems most probable that the genetic alterations observed in a majority of human HCCk are not due to a direct action of the virus. The wild-type p53 gene seems to regulate negatively the cellular growth and was therefore designated as a tumor suppressor gene or “antioncogene” (see Levine et al., 1991). Mutant forms of p53 frequently gain a growthstimulatory function, and genetic alterations of the gene, a common feature in human neoplasms, generally consist in the deletion of one p53 allele and in the mutation of the second allele. Point mutations that would alter the functional properties of p53 have recently been reported: liver-specific hotspot mutations at the third base of codon 249, consisting of G + T transversions, have been observed in more than 50% of liver cancers from patients residing in South Africa and in the Qidong province of China (Bressac et al., 1991; Hsu et al., 1991). Other studies providing similar results have also shown that two-thirds of the patients with the specific mutation had lost the remaining p53 allele (B. Slagle, personal communication). In these countries, exposure to high levels of dietary aflatoxin is well documented (see Section II,C) and the G T transversions are known to be induced by aflatoxin B 1 (Puisieux et al., 1991). However, this carcinogen also binds other G residues, particularly at codon 248, a mutational hotspot in other tumors but not in HCC. This suggests that additional factors might contribute to the selective targeting to Godon 249. Activation of aflatoxin B1 to metabolites having mutagenic o r DNA-binding activities is mediated by several forms of the human hepatic cytochrome P450 (Aoyama et al., 1991). Whether HBV infection plays part in the induction of p53 mutations is not clear. T h e p53 mutation at codon 249 has not been observed in HBV-related HCCs from patients that did not accumulate high exposure to aflatoxin B 1 (Hayward et al., 1991; Ozturk et al., 1991), or only in rare cases (M. Y. Lai, personal communication). Further studies have

-

210

MARIE ANNICK BUENDIA

identified point mutations in other p53 codons in 5 of 55 tumors from Taiwan, a region highly endemic for HBV but not for aflatoxin (M. Y. Lai, personal communications). However, a link between HBV infection and inactivation of a tumor-suppressor gene has not been established.

VIII. Conclusions The exponential relationship of HCC incidence to age indicates that multiple steps, probably involving independent genetic lesions, are required as in other human cancers. In particular, the long latency of HCC development after the initial HBV infection may be interpreted as a sign of an indirect action of the virus: a long-term toxic effect of viral gene products and/or the immune response against infected hepatocytes would trigger continuous necrosis and cell regeneration, which would in turn favor the accumulation of genetic alterations (Chisari et al., 1989b). In this model, productive HBV infections might potentiate the action of exogenous carcinogenic factors, like aflatoxins and alcohol. It might also be speculated that the latency period depends on the occurrence of a decisive HBV integration event, which would promote genetic instability or lead to cis- or trans-activation of relevant genes (Caselmann et al., 1990; Dejean et al., 1986; Hino et al., 1991; Kekule et al., 1990; Wollesheim et al., 1988;J. Wang et al., 1990).Recent investigations of the functional and pathological properties of HBV gene products and of the consequences of HBV integration in the liver DNA suggest that various and probably cooperative mechanisms may operate in the development of liver cancer, and that HBV may share with other human oncogenic viruses a number of basic strategies. The HBV genome encodes at least seven different polypeptides; none of them seems to act as a strong, dominant oncogene, but several lines of evidence indicate that the surface glycoproteins and the viral X trans-activator might participate in carcinogenesis, in a native or modified state. In this respect, it is noteworthy that in oncogenic viruses such as human T cell lymphotropic virus type I (HTLV-I), Epstein-Barr virus (EBV),and human papilloma virus (HPV) 16 and 18, transforming capacity cannot be dissociated from transcriptional trans-activation activity of viral gene products (Cohen et al., 1991; Hawley-Nelson et al., 1989; Henderson et al., 1991; Inoue et al., 1986; Fuji et al., 1991; Kieff and Liebowitz, 1990; Munger et al., 1989). Comparative analyses of the different viral trans-activators might help to guide future work in this field. Studies of mammalian HBV-related viruses have revealed strikingly different mechanisms that might be irrelevant to elucidate the molecular basis of HBV-induced tumorigenesis. However, they have revealed that activation of myc family genes lies at

HBV AND HEPATOCELLULAR CARCINOMA

21 1

the meeting point of the oncogenic pathways triggered by hepadnavirus infections in rodents, as for several human oncogenic viruses. Human papilloma virus infection has been correlated with amplifications and rearrangements of c-myc in cervical tumors and in squamous cell carcinoma of the anus and insertional mutagenesis of c-my or N - m y by HPV DNA has been occasionally observed (Couturier et al., 1991; Crook et al., 1991; Ocadiz et al., 1987); EBV has been associated with the development of Burkitt lymphomas, a tumor characterized by chromosomal translocations joining c-my to different immunoglobulin gene regions (reviewed by Magrath, 1990). So far, a role for HBV in activating the cm y oncogene, suggested by in uitro assays, has not been established in uivo and the identification of the cellular effectors in the HBV-induced transformation pathway remains the main unsolved question. ACKNOWLEDGMENTS I am grateful to C. Brechot, P. Briand, L. Cova, 0. Hino, A. Kekule, and Y. Mizuno for communicating recent unpublished data and to F. Chisari, K. Matsubara, M. Roggendorf, H. Schaller, C. Schroder, C. Seeger, B. Slagle, and H. Will for sending preprints. I also wish to thank M. Robertson for reading the manuscript, P. Tiollais and my collaborators for their critical comments. and L. M. Da for secretarial assistance.

REFERENCES Abe, K., Kurata, T., Shikata, T., and Tennant, B. C. (1988). Gann 79, 466-472. Aldershvile, J., Frosner, G. G., Nielsen, J. O., Hardt, F., Dienhardt, F., and Skinhoj, P. (1980). J. Infect. Dis. 141, 293-298. Ames, B. N. (1989). Envzron. Mol. Mutagen. 14, 66s-77s. Antonucci, T. K., and Rutter, W. J. (1989).J . Virol. 63, 579-583. Aoyama, T., Yamano, S., Guzelian, P. S., Gelboin, H. V., and Gonzalez, F. J. (1991). Hepatology 13, 1003-1005. Araki, K., Miyazaki,J. I., Hino, O., Tonita, N., Chisaka, O., Matsubara, K., and Yamamura, K. I. (1989). Proc. Natl. Acud. Scz. U.S.A. 86, 207-21 1 Asselsbergs, F. A. M., Will, H., Wingfield, P., and Hirshi, M. (1986).J. Mol. Bzol. 189, 401411.

Aufiero, B., and Schneider, R. J. (1990). EMBOJ. 9,497-504. Austin, H. (199 1). Adv. A w l . Biotechnol. Ser. 13, 57-59. Ayoola, E. A. (1988). In “Viral Hepatitis and Liver Disease” (A. J. Zuckerman, ed.), pp. 161-169. Alan R. Liss, New York. Ayoola, E. 4., and Members of the WHO Committee (1988). Bull. W.H.O. 66, 443-455. Babinet, C., Farza, H., Morello, D., Hadchouel, M., and Pourcel, C. (1985). Science 230, 1 160- 1663. Barnaba, V., Franco, A., Alberti, A., Benvenuto, R.,and Balsano, F. (1990).Nature (London) 345, 258-260. Bartenschlager, R., and Schaller, H. (1988). EMBOJ. 7, 4185-4192. Bartenschlager, R.,Junker-Niepman, M., and Schaller, H. ( 1 990).J. Virol. 64, 5324-5332. Beasley, R. P. (1987). Cancer (Philadelphia) 61, 1942-1956.

212

MARIE ANNICK BUENDIA

Beasley, R. P., and Hwang, L. Y. (1984). In “Viral Hepatitis and Liver Disease” (G. N. Vyas, J. Dienstag, and J. Hoofnagle, eds.), pp. 209-224. Grune & Stratton, New York. Beasley, R. P., and Hwang, L. Y. (1991). In “Viral Hepatitis and Liver Diseases” (F. B. Hollinger, S. M. Lemon, and M. Margolis, eds.), pp. 532-535. Williams & Wilkins, Baltimore, Maryland. Beasley, R. P., Trepo, C., Stevens, C. E., and Szmuness, W. (I977).Arn.J. Epidemiol. 105,94-98. Beasley, R. P., Lin, C. C., Hwang, L. Y., and Chien, C. S. (1981). Lancet 2, 1129-1133. Ben-Levy, R., Faktor, O., Berger, I., and Shaul, Y. (1989). Mol. Cell. Biol. 9, 1804-1809. Berger, I., and Shaul, Y. (1987).J. Virol. 61, 1180-1186. Bichko, V., Pushko, P., Dreilina, D., Pumpen, P., and Gren, E. (1985).FEBS Lett. 185,208212. Blum, A. L., Stutz, R., Haemmerli, V. P., Schmid, P., and Crady, G. F. (1969). Am.J. Med. 47, 82-92. Blum, H. E., Liang, T. J., Galun, E., and Wands, J. R. (1991). Hepatology 14, 56-63. Blumberg, B. S., Alter, H. J., and Visnich, S. E. (1965).JAMA,J. Am. Med. Assoc. 191,541546. Boender, P. J., Schalm, S. W., and Heijtink, R. A. (1985).J. Med. Virol. 16, 47-54. Bonino, F., Hoyer, B., Nelson, J., Engle, K., Verme, G., and Gerin, J. (1981). Hepatology 1, 386-391. Bonino, F., Brunetto, M. R., Rizzetto, M., and Will, H. (1991). Gastroenterology 100, 11381141. Bosch, F. X., and Munoz, N. (1991). Adv. Appl. Biotechnol. Ser. 13, 35-56. Brechot, C., Pourcel, C., Louise, A., Rain, B., and Tiollais, P. (1980). Nature (London) 286, 533-535. Brechot, C., Hadchouel, M., Scotto, J., Degos, F., Charnay, P., Trepo, C., and Tiollais, P. (1981a). Lancet 2, 765-768. Brechot, C., Hadchouel, M., Scotto, J., Fonck, M., Potet, F., Vyas, G. N., and Tiollais, P. (1981b). Proc. Natl. Acad. Sci. U.S.A. 78, 3906-3910. Brechot, C., Pourcel, C., Hadchouel, M., Dejean, A., Louise, A., Scotto, J., and Tiollais, P. (1982). Hepatology 2, 27s-33s. Brechot, C., Degos, F., Lugassy, C., Thiers, V., Zafrani, S., Franco, D., Bismuth, H., Trepo, C., Benhamou, J. P., Wands, J., Isselbacher, K., Tiollais, P., and Berthelot, P. (1985). N. Engl. J. Med. 312, 270-276. Bressac, B., Calvin, K. M., Liang, T. J., Isselbacher, K. J., Wands, J. R., and Ozturk, M. (1990). Proc. Natl. Acad. Sca. U.S.A. 87, 1973-1977. Bressac, B., Kew, M., Wands, J., and Ozturk, M. (1991). Nature (London) 350, 429-431. Brockes, J. (1990). Nature (London) 345, 766-768. Brunetto, M. R., Stemler, M., Schodel, F., Will, H., Ottobrelli, A., Rizzetto, M., Verme, G., and Bonino, F. ( 1989). Itul. J . Gastroenterol. 2 I, 151- 154. Brunetto, M. R., Giarin, M. M., Oliveri, F., Chiaberge, E., Baldi, M., Alfarano, A., Serra, A., Saracco, G., Verme, G., Will, H., and Bonino, F. (1991). Proc. Natl. Acad. Sci. U.S.A. 88,4186-4190 Bruss, V., and Ganem, D. (1991).J. Virol. 65, 3813-3820. Buetow, K. H., Murray, J., Israel, J., London, W., Smith, M., Kew, M., Blanquet, V., Brkchot, C., Redeker, A., and Govindarajah, S. (1989). Proc. Natl. Acad. Sci. U.S.A. 86, 8852-8856. Bulla, G. A., and Siddiqui, A. (1988).J. Virol. 62, 1437-1441. Bulla, G. A., and Siddiqui, A. (1989). Virology 170, 251-260. Bullock, P., Forrester, W., and Botchan, M. (1984).J. Mol. Biol. 174, 55-84. Bullock, P., Champoux, J. J., and Botchan, M. (1985). Science 230, 954-958.

HBV AND HEPATOCELLULAR CARCINOMA

213

Burk, R. D., De Loia, J. A., Elawady, M. K., and Gearhart, J. D. (1988).J. Viral. 62, 649654. Biischer, M., Reiser, W., Will, H., and Schaller, H. (1985). Cell (Camhdge, Mass.) 40, 717724. Carman, W. F.,Jacyna, M. R., Hadziyannis, S., Karayiannis, P., McGarvey, M. J., Markis, A,, and Thomas, H. C. (1989). Lancet 2,588-591. Carman, W. F., Fagan, E. A., Hadziyannis, S., Karayiannis, P., Tassopoulos, N. C., Williams, R., and Thomas, H. C. (1991). Hepatology 14, 219-222. Caselmann, W. H., Meyer, M., Kekul6, A. S., Lauer, U., Hofschneider, P. H., and Koshy, R. (1990). Proc. Natl. Acad. Sci. U.S.A. 87, 2970-2974. Cattaneo, R., Will, H., Hernandez, N., and Schaller, H. (1983). Nature (London) 305, 336340. Cattaneo, R., Will, H., and Schaller, H. (1984). E M B O J . 3, 2191-2196. Chakraborty, P. R., Ruiz-Opazo, H., Shouval, D., and Shafriz, D. A. (1980).Nature (London) 286, 531-533. Chandar, N., Lombardi, B., and Locker, J. (1989).Proc. Natl. Acad. Sci. U.S.A. 86, 27032707. Chang, C., Enders, G., Sprengel, R., Peters, N., Varmus, H. E., and Ganem, D. (1987).J. Viral. 61, 3322-3325. Chang, H. K., and Ting, L. P. (1989). Virology 170, 176-183. Chang, H. K., Chou, C. K., Chang, C., Su, T. S., Hu, C., Yoshida, M., and Ting, L. P. (1987). Nucleic Acids Res. 15, 2261-2268. Chang, H. K., Wang, B. Y., Yuh, C. H., Wei, C. L., and Ting, L. P. (1989).Mol. Cell. Biol. 9, 5189-5197. Chang, L. J., Pryciak, P., Ganem, D., and Varrnus, H. E. (1989).Nature (London)337,364368. Chang, L. J., Ganem, D., and Varmus, H. E. (1990).Proc. Natl. Acad. Scz. U.S.A. 87,51585162. Chang, M. H., Chen, D. S., Hsu, H. C., Wu, M. Z., Hsu, H. Y., and Lee, C. Y. (1988). I n “Viral Hepatitis and Liver Disease” (A. J. Zuckerman, ed.), pp. 726-727. Alan R. Liss, New York. Chang, M. H., Chen, P. J., Chen, J. Y., Lai, M. Y., Hsu, H. C., Lian, D. C., Liu, Y. G., and Chen, D. S. (1991). Hepatology 13, 316-320. Chen, C. J., Liang, K. Y., Chang, A. S., Chang Y. C., Lu, S. N., Liaw, Y. F., Chang, W. Y., Sheen, M. C., and Lin, T. M. (1991). Hepatology 13, 398-406. Chen, P. J., Chen, C. R., Sung, J. L., and Chen, D. S. (1989).J. Viral. 63, 4165-4171. Chisari, F. V., Filippi, P., McLachlan, A,, Milich, D. R., Riggs, M., Lee, S., Palmiter, R. D., Pinkert, C. A , , and Brinster, R. L. (1986).J. Virol. 60, 880-887. Chisari, F. V., Filippi, P., Buias, A., McLachlan, A., Popper, H., Pinkert, C. A., Palniter, R. D., and Briuster, R. L. (1987). Proc. Natl. Acad. Scz. U.S.A. 84, 6909-6913. Chisari, F. V., Ferrari, C., and Mondelli, M. U. (1989a). Microb. Pathog. 6, 311-325. Chisari, F. V., Klopchin, K., Moriyama, T., Pasquinelli, C., Dunsford, H. A,, Sell, S., Pinkert, C. A., Brinster, R. L., and Palmiter, R. D. (1989b). Cell (Cambridge, Mass.) 59, 1145-1 156. Chomel, B., Trkpo, C. H., Pichoud, C. H., Jacquet, C., Boulay, P., and Joubert, L. (1984). Comp. lmmunol. Microbial. Infect. Dis. 7, 179- 194. Cohen, J. I., Miller, R. H., Rosenblum, B., Denniston, K., Gerin, J. L., and Purcell, R. H. (1988). Virology 162, 12-20. Cohen, J. I., Wang, F., and Kieff, E. (1991).J. Virol. 65, 2545-2554. Colgrove, R., Simon, G., and Ganem, D. (1989).J. Viral. 63, 4019-4026.

214

MARIE ANNICK BUENDIA

Corcoran, L. M., Adams, J. M., Dunn, A. R., and Cory, S. (1984).Cell (Cambridge, Mass.) 37, 113-122. Cote, P. J., and Gerin, J, L. (1983).J. Virol. 47, 15-23. Cote, P. J., Shapiro, M., Engle, R. E., Popper, H., Purcell, R. H., and Gerin, J. L. (1986).J. Virol. 60,895-901. Courtois, G. S., Baumhueter, S., and Crabtree, G. R. (1988). Proc. Nutl. Acad. Scz. U.S.A. 85, 7937-7941. Couturier, J., Sastre-Garau, X., Schneider-Maunoury, S., Labib, A., and Orth, G. (1991).J. Virol. 65, 4534-4538. Cova, L., Wild, C. P., Mehrotra, R,. Turusov, V., Shirai, T., Lambert, V., Jacquet, C., Tomatis, L., Trtpo, C., and Montesano, R. (1990). CuncerRes. 50, 2156-2163. Craig, J. R., Klatt, E. C., and Yu, M. (1991). Adv. Appl. Biotechnol. Ser. 13, 177-190. Crescenzo-Chaigne, B., Pillot, J., Lilienbaum, A., Levrero, M., and Elfassi, E. (1991). J. Varol. 65, 3882-3886. Crook, T., Wrede, D., Tidy, J., Scholefield, J., Crawford, L., and Vousden, K. H. (1991). Oncogene 6, 1251-1257. Degos, F., Lugassy, C., Degott, C., Debure, A., Carnot, F., Thiers, V., Tiollais, P., Kreis, H., and Brtchot, C. (1988). Gastroenterology 94, 151-156. Dejean, A., and de ThC, H. (1990). Mol. Biol. Med. 7, 213-222. Dejean, A., Sonigo, P., Wain-Hobson, S., and Tiollais, P. (1984).Proc. Nutl. Acad. Sci. U.S.A. 81,5350-5354. Dejean, A., Bougueleret, L., Grzeschik, K. H., and Tiollais, P. (1986). Nature (London) 322, 70-72. DeLoia, J. A., Burk, R. D., and Gearhart, J. D. (1989).J. Virol. 63,4069-4073. De-Medina, T., Faktor, O., and Shaul, Y. (1988). Mol. Cell. Biol. 8, 2449-2455. de The, H., Marchio, A., Tiollais, P., and Dejean, A. (1987).Nature (London)330,667-670. de T h t , H., Chomienne, C., Lanotte, M., Degos, L., and Dejean, A. (1990). Nature (London) 347, 558-561. de T h t , H., Lavau, C., Marchio, A., Chomienne, C., Degos, L., and Dejean, A. (1991). Cell (Cambndge, Mass.) 66, 675-684. Dolcetti, R., Rizzo, S., Viel, A., Maestro, R., De Re, V., Feriotto, G., and Boiocchi, M. (1989). Oncogene 4, 1009-1014. Dragani, T. A., Manenti, G., Farza, H., Della Porta, G., Tiollais, P., and Pourcel, C. (1989). Carcinogenesis (London) 11, 953-956. Dunsford, H. A., Sell, S., and Chisari, F. V. (1990). Cancer Res. 50, 3400-3407. Eble, B. E., Lingappa, V. R., and Ganem, D. (1986). Mol. Cell. Bzol. 6, 1454-1463. Eble, B. E., MacRae, D. R., Lingappa, V. R., and Ganem, D. (1987). Mol. Cell. Biol. 7,35913601. Edman, J. C., Gray, P., Valenzuela, P., Rall, L. B., and Rutter, W. J. (1980). Nature (London) 286, 535-538. Elfassi, E., Romet-Lemonne, J. L., Essex, M., Frances-McLane, M., and Haseltine, W. A. (1984). Proc. Natl. Acud. Sci. U.S.A. 81, 3526-3528. Elfassi, E., Haseltine, W. A., and Dienstag, L. (1986).Proc. Natl. Acad. Scz. U.S.A. 83,22192222. Enders, G. H., Ganem, D., and Varmus, H. E. (1985). Cell (Cambridge, Mars.) 42,297-302. Etiemble, J., Moroy, T., Trtpo, C., Tiollais, P., and Buendia, M. A. (1986). Gene 50, 207214. Etiemble, J., Moroy, T., TrCpo, C., Tiollais, P., and Buendia, M. A. (1988). In “Viral Hepatitis and Liver Disease” (A. J. Zuckerman, ed.), pp. 484-487. Alan R. Liss, New York.

HBV A N D HEPATOCELLULAR CARCINOMA

215

Etiemble, J., M(iroy, T., Jacquemin, E., Tiollais, P., and Buendia, M. A. (1989). Oncogene 4, 51-57. Farber, E. (1984). Cancer Res. 44,42 17-422 1 . Farza, H., Salmon, A. M., Hadchouel, M., Moreau, J. L., Babinet, C., Tiollais, P., and Pourcel, C. (1987). Proc. Natl. Acud. Sci. V.S.A. 84, 1187-1191. Farza, H., Hadchouel, M., Scotto, J., Tiollais, P., Babinet, C., and Pourcel, C. (1988).J . Virol. 62, 4144-4152. Feitelson, M. A,, and Miller, R. H. (1988). Proc. Natl. Acad. Sci. U.S.A. 85, 6162-6166. Feitelson, M. A,, Marion, P. L., and Robinson, W. S. (1981).J. Virol. 39, 447-454. Feitelson, M. A,, Marion, P. L., and Robinson, W. S. (1982).J. Virol. 43, 687-696. Feitelson, M., Millman I., Halbherr, T., Simmons, H., and Blumberg, R. (1986).Proc. Natl. Acad. Sci. U.S.A. 83, 2233-2237. Fourel, G., Trtpo, C., Bougueleret, L., Henglein, B., Ponzetto, A., Tiollais, P., and Buendia, M. A. (1990). Nature (London) 347, 294-298. Freytag von Loringhoven, A., Koch, S., Hofschneider, P. H., and Koshy R. (1985).EMBOJ. 4, 249-255. Fujii, M., Niki, T., Mori, T., Matsuda, T., Matsui, M., Nomura, N., and Seiki, M. (1991). Oncogene 6, 1023-1029. Fujimori, M., Tokino, T., Hino, O., Kitagawa, T., Imamura, T., Okamoto, E., Mitsunobu, M., Ishikawa, T., Nakagama, H., Harada, H., Yagura, M., Matsubara, K., and Nakamura, Y. (1991). Cancer Res. 51, 89-93. Fujiyama, A,, Miyanohara, A., Nozaki, C., Yoneyama, T., Ohtomo, N., and Matsubara, K. (1983). Nucleic Acids Res. 11, 4601-4610. Galibert, F., Mandart, E., Fitoussi, F., Tiollais, P., and Charnay, P. (1979). Nature (London) 281, 646-650. Galibert, F., Chen, T. N., and Mandart, E. (1982).J. Virol. 41, 51-65. Galle, P. R., Schlicht, H. J., Fischer, M., and Schaller, H. (198Q.J. Virol. 62, 1736-1740. Ganem, D. (1990). Nature (London) 347, 230-232. Ganem, D., and Varmus, H. E. (1987). Annu. Rev. Biochem. 56,651-693. Ganem, D., Weiser, B., Barchuk, A,, Brown, R. J., and Varmus, H. E. (1982a).J. Vzro2. 44, 366-373. Ganem, D., Greenbaum, L., and Varmus, H. E. (1982b).J . Virol. 44,374-383. Gerber, M. A,, Hadziyannis, S., Vissoulis, C., Schaffner, F., Paronetto, F., and Popper, H. (1974). Am. J. Pathol. 75, 489-502. Gerken, G., Kremsdorf, D., Capel, F., Petit, M. A., Dauguet, C., Manns, M. P., Meyer zum Buschenfelde, K. H., and BrCchot, C. (1991a). Virology 183, 555-565. Gerken, G., Paterlini, P., Manns, M., Housset, C., Terre, S., Dienes, H. P., Hess, G., Gerlich, W. H., Berthelot, P., Meyer zum Biischenfelde, K. H., and Brechot, C. (1991b). Hepatology 13, 158-166. Gerlich, W. H., Feitelson, M. A., Marion, P. L., and Robinson, W. S. (1980).J. Virol. 36,787795. Girones, R., and Miller, R. H. (1989). Virology 170, 595-597. Girones, R., Cote, P. J., Hornbuckle, W. E., Tennant, B. C., Gerin, J. L., Purcell, R. H., and Miller, R. H. (1989). Proc. Natl. Acad. Sci. V.S.A. 86, 1846-1849. Goudeau, A. (1990). Vuccine 8, 113-1 16. Cough, N. M. (1983).J. Mol. Biol. 165, 683-699. Cough, N. M., and Murray, K. (1982).J. Mol. B i d . 162, 43-67. Graf, T., and Berg, H. (1983). Cell (Cambndge, Mass.) 34, 7-9. Gu, J. R., Hu, L. F., Cheng, Y. C., and Wan, D. F. (1986). J . Cell. Physiol. 4, Suppl., 13-20.

216

MARIE ANNICK BUENDIA

Hadchouel, M., Scotto, J., Hinet, J. L., Molinie, D., Villa, E., Degos, F., and Brkchot, C. (1985).J . Med. Virol. 16, 61-66. Hadziyannis, S. J., Lieberman, H. M., Karvountzis, .G. G., and Shafritz, D. A. (1983). Hepatology 3, 656-662. Hall, A. J., Winter, P. D., and Wright, R. (1985). Lancet 1, 91-93. Hall, P., Plummer, J. L., Ilsley, A. H., and Cousins, M. J. (1991). Hepatology 13, 815-819. Halpern, M. S., England, J. M., Deery, D. T., Petcu, D. J., Mason, W. S., and MolnarKimber, K. L. (1983). Proc. Nutl. Acad. Scz. U.S.A. 80, 4865-4869. Harris, C. C. (1990). Cancer Cells 2, 146-148. Haruna, Y., Hayashi, N., Katamaya, K., Yuki, N., Kasahara, A., Sasaki, Y., Fusamoto, H., and Kamada, T. (1991). Hepatology 13, 417-421. Hatada, I., Tokino, T., Ochiya, T., and Matsubara, K. (1988). Oncogene 3, 537-540. Hawley-Nelson, P., Vousden, K. H., Hubbert, N. L., Lowy, D. R., and Schiller,J. T. (1989). EMBO J. 8, 3905-3910. Hayashi, K., Makino, R., and Sugimura, T. (1984). Gann 75, 475-478. Hayward, N. K., Walker, G. J., Graham, W., and Cooksley, E. (1991).Nature (London) 352, 764. Henderson, S., Rowe, M., Gregory, C., Croom-Carter, D., Wang, F., Longnecker, R., Kieff, E., and Rickinson, A. (1991). Cell (Cambrzdge, Mass.)65, 1107-1 115. Hilger, C., Velhagen, I., Zentgraf, H., and Schriider, C. H. (1991).J. Virol. 65,4284-4291. Hino, O., Shows, T. B., and Rogler, C. E. (1986). Proc. Nutl. Acad. Sci. U.S.A. 83, 83388342. Hino, O., Ohtake, K., and Rogler, C. E. (1989).J . Virol. 63, 2638-2643. Hino, O., Tabata, S., and Hotta, Y. (1991). Proc. Nutl. Acad. Sci. U.S.A. 88, 9248-9252. Hirsch, R. C., Lavine, J. E., Chang, L. J., Varmus, H. E., and Ganem, D. (1990). Nature (London) 344, 552-555. Hirsch, R. C., Loeb, D. D., Pollack, J. R., and Ganem, D. (1991).J. Virol. 65, 3309-3316. Hohne, M., Schaefer, S., Seifer, M., Feitelson, M. A., Paul, D., and Gerlich, W. H. (1990). EMBO J . 9, 1137-1 145. Hollinger, F. B. (1990a).I n “Viral Hepatitis” (F. B. Hollinger, W. S. Robinson, R. H. Purcell, J, L. Gerin, and J. Ticehurst, eds.), pp. 73-138. Raven Press, New York. Hollinger, F. B. (1990b). Vaccine 8, 122-128. Howard, C. R. (1986).J . Gen. Virol. 67, 1215-1235. Hruska, J. F., Clayton, D. A., Rubenstein, J. L. R., and Robinson, W. S. (1977).J. Virol. 21, 666-672. Hsu, H. C., Chiou, T. J., Chen, J. Y., Lee, C. S., Lee, P. H., and Peng, S. Y. (1991).Hepatology 13,923-928. Hsu, T. Y., Moroy, T., Etiemble, J., Louise, A., Trepo, C., Tiollais, P., and Buendia, M. A. (1988). Cell (Cambridge, Mass.) 55, 627-635. Hsu, T. Y., Fourel, G., Etiemble, J., Tiollais, P., and Buendia, M. A. (1990). Gastroenterol. Jpn. 25,43-48. Hu, K. Q., and Siddiqui, A. (1991). Virology 181, 721-726. Hu, K. Q., Vierling, J. M., and Siddiqui, A. (1990). Proc. Natl. Acad. Sci. U.S.A. 87, 71407144. Ikeda, T., Lever, A. M. L., and Thomas, C. H. (1986). Hepatology 6, 962-965. Imai, M., Hoshi, Y., Okamoto, H., Matsui, T., Tsurimoto, T., Matsubara, K., Miyakawa, Y., and Mayumi, M. (1987).J . Virol. 61, 3555-3560. Imazeki, F., Yaginuma, K., Omata, M., Okuda, K., Kobayashi, M., and Koike, K. (1988).J. Virol. 62, 861-865.

HBV A N D HEPATOCELLULAR CARCINOMA

217

Inoue, J., Seiki, M., Taniguchi, T., Tsuru, S., and Yoshida, M. (1986). E M B O J . 5, 28832888. Jameel, S., and Siddiqui, A. (1986). Mol. Cell. B i d . 6, 710-715. Jean-Jean, O., Levrero, M., Will, H., Perricaudet, M., and Rossignol, J. M. (1989). Virology 170, 99-106. Junker-Niepman, M., Bartenschlager, R., and Schaller, H. (1990).EMBOJ. 9, 3389-3396. Kay, A., Mandart, E., TrCpo, C., and Galibert, F. (1985). EMBOJ. 4, 1287-1292. KekulC, A. S. (1989). Ph.D. Thesis. Freie Universitat, Berlin, Germany. KekulC, A. S., Lauer, U., Meyer, M., Caselmann, W. H., Hofschneider, P. H., and Koshy, R. ( I 990). Nature (London) 343, 457-461. Kew, M. C. (1990). In “Hepatology: A Textbook of Liver Disease” (D. Zakin and T. D. Boyer, eds.), pp. 1206- 1240. Saunders, Philadelphia, Pennsylvania. Kew, M. C., and Popper, H. (1984). Semin. Liver Dir. 2, 136-145. Kieff, E., and Liebowitz, D. (1990).In “Virology”(B. N. Fields, D. M. Knipe, R. M. Chanok, M. S. Hirsch, J. L. Melnick, T. P. Monath, and B. Roizman, eds.), 2nd ed., pp. 18891920. Raven Press, New York. Kim, C. M., Koike, K., Saito, I., Miyamura, T.,and Jay, G. ( 1 991). Nature (London)351,3 17320. Kobayashi, M., and Koike, K. (1984). Gene 30, 227-232. Koch, S., Freytag von Loringhoven, A., Hofschneider, P. H., and Koshy, R. (19844. EMBO J . 3,2185-2189. Koch, S., Freytag von Loringhoven, A., Kahmann, R., Hofschneider, P. H., and Koshy, R. (1 984b). Nucleic Acids Res. 12, 687 1-6876. Kodama, K., Ogasawara, N., Yoshikawa, H., and Murakami, S. (1985).J. Virol. 56,978-986. Korba, B. E., Wells, F., Tennant, B. C., Yoakum, G. H., Purcell, R. H., and Gerin, J. L. (1986).J. Vzrol. 58, 1-8. Korba, B. E., Wells, F., Tennant, B. C., Cote, P. J., and Gerin, J. L. (1987).J. Vzrol. 61, 13181324. Korba, B. E., Cote, P. J., and Gerin, J. L. (198th). Science 241, 1213-1216. Korba, B. E., Gowans, E. J.. Wells, F. V., Tennant, B. C., Clarke, R., and Gerin, J. L. (l988b). Virology 165, 172-181. Korba, B. E., Cote, P. J., Wells, F. V., Baldwin, B., Popper, H., Purcell, R. H., Tennant, B. C., and Gerin, J. L. (1989a).J. Vzrol. 63, 1360-1370. Korba, B. E., Wells, F. V., Baldwin, B., Cote, P. J., Tennant, B. C., Popper, H., and Gerin, J. L. (1989b). Hepatology 9, 461-470. Korba, B. E., Brown, T. L., Wells, F. V., Baldwin, B., Cote, P. J., Steinberg, H., Tennant, B. C., and Gerin, J. L. (1990). J. Virol. 64,4499-4506. Koshy, R.,and Wells, J. (1991). Adv. in Appl. Biolechnol. Ser. 13, 159-170. Koshy, R.,Maupas, P., Muller, R., and Hofschneider, P. H. (198 I ) . J . Gen. Virol. 57,95- 102. Koshy, R., Koch, S., Freytag von Loringhoven, A., Kahmann, R., Murray, K., and Hofschneider, P. H. (1983). Cell (Cambridge, Mass.) 34, 215-223. Kulonen, K., and Millman, I. (1988).J. Med. Virol. 26, 233-242. Kuroki, K., Russnak, R., and Ganem, D. (1989). Mol. Cell. B i d . 9, 4459-4466. Lambert, V., Cova, L., Chevallier, P., Mehrotra, R., and TrCpo, C. (1991).J. Gen. Virol. 72, 41 7-420. Larzul, D., Guigue, F., Sninsky,J. J., Mack, D. H., Brkchot, C., and Guesdon, J. L. (198S).J. Varol. Methods 20, 227-237. Laure, F., Chatenoud, L., Gazengel, C., Beaurain, G., Torchet, M. F., Zagury, D., Bach, J. F., and BrCchot, C. (1987). Br.J. Huematol. 65, 181-185.

218

MARIE ANNICK BUENDIA

Lee, T. H., Finegold, M. J., Shen, R. F., DeMayo, J. L., Woo, S. L. C., and Butel, J. S. (1990). J. Virol. 64, 5939-5947. Levine, A. J., Momand, J., and Finlay, C. A. (1991). Nature (London) 351, 453-456. Levrero, M., Jean-Jean, O., Balsano, C., Will, H., and Perricaudet, M. (1990). Virology 174, 299-304. Levrero, M., Stemler, M., Pasquinelli, C., Alberti, A,, Jean-Jean, O., Franco, A., Balsano, C., Diop, D., Brkchot, C., Melegari, M., Villa, E., Barnaba, V., Perricaudet, M., and Will, H. (1991). Hepatology 13, 143-149. Li, Y., Golemis, E., Hartley, J. W., and Hopkins, N. (1987).J. Virol. 61, 693-700. Liang, T. J., Hasegawa, K., Rimon, N., Wands, J. R., and Ben-Porath, E. (1991a).N . Eng1.J. Med. 324, 1705- 1709. Liang, T. J., Baruch, Y., Ben-Porath, E., Enat, R., Bassan, L., Brown, N., Rimon, N., Blum, H. E., and Wands, J. R. (1991b). Hepatology 13, 1044-1051. Lien, J. M., Aldrich, C. E., and Mason, W. S. (1986).J. Virol. 57, 229-236. Lien, J. M., Petcu, D. J., Aldrich, C. E., and Mason, W. S. (1987).J . Virol. 61, 3832-3840. Lo, S. J., Chien, M. L., and Lee, Y. H. (1988). Virology 167, 289-295. Lok, A., Lai, C., Chung, H., Lau, J., Leung, E., and King, L. (1991).Hepatology 13, 834837. Loncarevic, I. F., Zentgraf, H., and Schroder, C. H. (1990). Nucleic Acids Res. 18, 4940. Lopez-Cabrera, M., Letovsky,J., Hu, K. Q., and Siddiqui, A. (1990). Proc. Natl. Acad. Sci. U.S.A. 87, 5069-5073. Lopez-Cabrera, M., Letovsky, J., Hu, K. Q., and Siddiqui, A. (1991). Virology 183,825-829. Lugassy, C . , Bernuau, J., Thiers, V., Krosgaard, K., Degott, C., Wantzin, P.,Schalm, S. W., Rueff, B., Benhamou, J. P., Tiollais, P., and Brechot, C. (1987).J.Infect. DW.155,64-71. Machado, I. V., Carvajal, J., Mondolfi, A., Marcano, N., Yarzabal, L., and Bianco, N. E. (1988).I n “Viral Hepatitis and Liver Disease” (A.J. Zuckerman, ed.), pp. 174-176. Alan R. Liss, New York. Magnius, L. O., and Espmark, J. A. (1972).J.Immunol. 109, 1017-1021. Magrath, I. (1990). Adv. Cancer Res. 55, 134-270. Maguire, H. F., Hoeffler, J. P., and Siddiqui, A. (1991). Science 252, 842-844. Mancini, M., Pourcel, C., Tiollais, P., and Michel, M. L. (1991). In “Viral Hepatitis and Liver Diseases” (F. B. Hollinger, S. M. Lemon, and H. Margolis, eds.), pp. 273-276. Williams & Wilkins, Baltimore, Maryland. Mandart, E., Kay, A., and Galibert, F. (1984).J. Virol. 49, 782-792. Marion, P. L., Oshiro, L. S., Regnery, D. C., Scullard, G. H., and Robinson, W. S. (1980). Proc. Natl. Acad. Sci. U.S.A. 77, 2941-2945. Marion, P. L., Knight, S. S., Salazar, F. H., Popper, H., and Robinson, W. S. (1983). Hepatology 3, 519-527. Marion, P. L., Knight, S. S., Ho, B. K., Guo, Y. Y., Robinson, W. S., and Popper, H. (1984). Proc. Natl. Acad. Sci. U.S.A. 81, 898-902. Marion, P. L., Van Davelaar, M. J., Knight, S. S., Salazar, F. H., Garcia, G., Popper, H., and Robinson, W. S. (1986). Proc. Natl. Acad. Sci. U.S.A. 83, 4543-4546. Marion, P. L., Popper, H., Azcarraga, R. R., Steevens, C., Van Davelaar, H. J., Garcia, G., and Robinson, W. S. (1987). I n “UCLA Symposia on Molecular and Cellular Biology, New Series: Hepadna Viruses” (W. S. Robinson, K. Koike, and H. Will, eds.), pp. 337348. Alan R. Liss, New York. Mason, W. S., and Taylor, J. M. (1989). Hepatology 9, 635-645. Mason, W. S., Seal, G,. and Summers, J. (1980).J. Virol. 36, 829-836. Matsubara, K., and Tokino, T. (1990). Mol. Baol. Med. 7, 243-260.

HBV A N D HEPATOCELLULAR CARCINOMA

219

Matsubara, K., Ochiya, T., Hatada, I., Shiozawa, M., Fukushige, S., Araki, M., Fujiyama, A., and Tsurimoto, T. (1987). Oncogenes and Cancer. Jpn. Sci. SOC.Press, Tokyo, and VNU Sci. Press BV, Utrecht, T h e Netherlands. McLachlan, A., Milich, D. R., Raney, A. K., Riggs, M. G., Hughes, J. L., Sorge, J., and Chisari, F. V. (1987).J. Virol. 61, 683-692. McLean, E. K., McLean, A. E. M., and Sutton, P. M. (1969).Br. J. Exp. Pathol. 50,502-506. McMahon, B. J., Alward, W. L., Hall, D. B., Heyward, W. L., Bender, T. R., Francis, D. P., and Maynard, J. E. (1985).J. Infect. Dis. 151, 599-603. Melegari, M., Jung, M. C., Schneider, R., Santantonio, T., Bagnulo, S., Luchena, N., Pastore, G., Pape, G., Scaglioni, P., Villa, E., and Will, H. (1991).J. Hepatol. (in press). Merrill, D. A., Dubois, R. S., and Kohler, P. F. (1972). N . Engl. J. Med. 287, 1280-1282. Meyers, M. L., Vitvitski Trtpo, L., Nath, N., and Sninsky, J. J. (1986).J. Virol. 57, 101-109. Milich, D. R. (1988). Immunol. Today 9, 380-386. Milich, D. R., McLachlan, A., Moriarty, A., and Thornton, G. B. (1987).J. Immunol. 139, 1223- 1232. Milich, D. R., Jones, J. E., Hughes, J. L., Price, J., Ranay, A. K., and McLachlan, A. (1990). Proc. Natl. Acad. Sci. U.S.A. 87, 6599-6603. Miller, R. H., and Robinson, W. S. (1986). Proc. Natl. Acad. Sci. U.S.A. 83, 2531-2535. Miller, R. H., Girones, R., Cote, P. J., Hornbuckle, W. E., Chestnut, T., Baldwin, B. H., Korba, B. E., Tennant, B. C., Gerin, J. L., and Purcell, R. H. (1990).Proc. Natl. Acad. Sci. U.S.A. 87, 9329-9332. Miller, T., Tran, C. T., and Robinson, W. S. (1984). Virology 139, 53-63. Minuk, G. Y., Shaffer, E. A., Hoar, D. I., and Kelly, J. (1986). Liver 6, 350-356. Miyaki, M., Sato, C., Gotanda, T., Matsui, T., Mishiro H., Imai, M., and Mayumi, M. (1986). J . Gen. Vzrol. 67, 1449-1454. Mizusawa, H., Taira, M., Yaginuma, K., Kobayashi, M., Yoshida, E., and Koike, K. (1985). Proc. Natl. Acad. Sci. U.S.A. 82, 208-212. Molnar-Kimber, K. L., Summers, J.,Taylor, J., and Mason, W. S. (1983).J . Virol. 45,165-172. Molnar-Kimber, K. L., Summers, J., and Mason, W. S. (1984).J. Virol. 51, 181-191. Molnar-Kimber, K. L., Jarocki-Witek, V., Dheer, S. K., Vernon, S. K., Conley, A. J., Davis, A. R., and Hung, P. P. (1988).J . Virol. 62, 407-416. Mondelli, M., Mieli-Vergani, G., Alberti, A., Vergani, D., Portman, B., Eddleston, A. L., and Williams, R. (1982).J. Immunol. 129, 2773-2778. Moriarty, A. M., Alexander, H., Lerner, R. A., and Thornton, G. B. (1985). Science 227, 429-433. Moriyama, T., Guilhot, S., Klopchin, K., Moss, B., Pinkert, C. A., Palmiter, R. D., Brinster, R. L., Kanagawa, O., and Chisari, F. V. (1990). Science 248, 361-364. Moroy, T., Etiemble, J., Trepo, C., Tiollais, P., and Buendia, M. A. (1985). EMBO J . 4, 1507- 15 14. Moroy, T., Marchio, A,, Etiemble, J., Trepo, C., Tiollais, P., and Buendia, M. A. (1986). Nature (London) 324, 276-279. Moroy, T., Etiemble, J., Bougueleret, L., Hadchouel, M., Tiollais, P., and Buendia, M. A. (1989). Oncogene 4, 59-65. Munger, K., Phelps, W. C., Bubb, V., Howley, P. M., and Schlegel, R. (1989).J. Vzrol. 63, 44 17-4421. Nagaya, T., Nakamura, T., Tokino, T., Tsurimoto, T., Imai, M., Mayumi, T., Kamino, K., Yamamura, K., and Matsubara, K. (1987). Genes Dev. 1, 773-782. Nakamura, T., Tokino, T., Nagaya, T., and Matsubara, K. (1988). Nucleic AcidF Res. 16, 4865-4873.

220

MARIE ANNICK BUENDIA

Nakao, K., Miyao, Y., Ohe, Y., and Tamaoki, T. (1989). Nucleic Acids Res. 17, 9833-9842. Neurath, A. R., Kent, S. B. H., Strick, N., and Parker, K. (1986). Cell (Cambridge, Mass.) 46, 429-436. Newberne, P. M., and Butler, W. H. (1969). Cancer Res. 29, 236-250. Obata, H., Hayashi, N., Motoike, Y., Hisamitsu, T., Okuda, H., Kobayashi, S., and Nishioka, K. (1980). 1nt.J. Cancer 25, 741-747. Obayashi, A., Okochi, K., and Mayumi, M. (1972). Gastroenterology 62, 618-625. Ocadiz, R., Sauceda, R., Cruz, M., Graef, A. M., and Gariglio, P. (1987). Canter Res. 47, 4173-4177. Ochiya, T., Fujiyama, A., Fukushige, S., Hatada, I., and Matsubara, K. (1986). Proc. Natl. Acud. Sci. U.S.A. 83, 4993-4997. Ochiya, T., Tsurirnoto, T., Ueda, K., Okubo, K., Shiozawa, M., and Matsubara, K. (1989). Proc. Natl. Acaa!. Sci. U.S.A. 86, 1875-1879. Ogata, N., Karnimura, T., and Asakura, H. (1991). Hepatology 13, 31-37. Ogston, C. W., Jonak, G. J., Rogler, C. E., Astrin, S. M., and Summers, J. (1982). Cell (Cambridge, Mass.) 29, 385-394. Ogston, C. W., Schechter, E. M., Humes, C. A., and Pranikoff, M. B. (1989). Virology 169, 9-14. Okada, K., Kemiyama, I., Inomato, M., Imai, M., Miyakawa, Y., and Mayumi, M. (1976).N . Engl. J. Med. 294, 746-749. Okamoto, H., Imai, M., Shimozaki, M., Hoshi, Y., Lizuka, H., Gotanda, T., Tsuda, F., Miyakawa, Y., and Mayumi, M. (1986).J. Gen. Virol. 67, 2305-2314. Okamoto, H., Tsuda, F., and Mayumi, M. (1987a).Jpn.J. Exp. Med. 57, 217-221. Okamoto, H., Imai, M., Kametani, M., Nakamura, T., and Mayumi, M.(1987b).Jpn.J. Exp. Med. 57, 231-236. Okamoto, H., Yotsumoto, S., Akahane, Y., Yamanaka, T., Miyazaki, Y., Sugai, Y., Tsuda, F., Tanaka, T., Miyakawa, Y., and Mayumi, M. (199O).J. Vzrol. 64,1298-1303. Okuda, K: (1991). Adv. A@l. Biotechnol. Ser. 13, 119-126. Oldstone, M. B. A. (1989). Cell (Cambridge, Mass.) 56, 517-520. Oliveri, F., Brunetto, M. R., Baldi, M., Piantino, P., Ponzetto, A,, Forzani, F., Smedile, A., Verme, G., and Bonino, F. (1991). In “The Hepatitis Delta Virus” L. Gerin, R. H. Purcell, and M. Rizzetto, eds.), pp. 217-222. Wiley-Liss, New York. Omata, M., Ehata, T., Yokosuka, O., Hosoda, K., and Ohto, M. (1991).N . Eng1.J. Med. 324, 1699-1704. Onji, M., Lever, A. M. L., Saito, I., and Thomas, H. C. (1989). Hepatology 9, 92-96. Ono, Y., Onda, H., Sasada, R., Igarashi, K., Sugino, Y., and Nishioka, K. (1983). Nucleic Ac& R ~ s 11, . 1747-1757. Orito, E., Mizokami, M., h a , Y., Moriyama, E. N., Kameshima, N., Yamamoto, M., and Gojobori, T. (1989). Proc. Natl. Acaa!. Scz. U.S.A. 86, 7059-7062. Ou, J. H., and Rutter, W. J. (1985). Proc. Natl. Acud. Sci. U.S.A. 82, 83-87. Ou, J. H., and Rutter, W. J. (19871.1. Virol. 61, 782-786. Ou, J. H., Laub, O., and Rutter, W. J. (1986). Proc. Natl. Acad. Sci. U.S.A. 83, 1578-1582. Ou, J. H., Bao, H., Shih, C., and Tahara, S. M. (199O).J. Virol. 64,4578-4581. Ozturk, M., et al., (1991). Lancet 338, 1356-1359. Pasek, M., Goto, T., Gilbert, W., Zink, B., Schaller, H., MacKay, P., Leadbetter, G., and Murray, K. (1979). Nature (London) 284, 575-579. Pasquinelli, C., Laure, F., Chatenoud, L., Beaurin, G., Gazengel, C., Bismuth, H., Degos, F., Tiollais, P., Bach, J. F., and Brkhot, C. (1986).J. H e p o l . 3, 95-103. Pasquinelli, C., Garreau, F., Bougueleret, L., Cariani, E., Grzeschik, K. H., Thiers, V., Croissant, O., Hadchouel, M., Tiollais, P., and Brtchot, C. (1988).J. Virol. 62,629-632.

u.

HBV A N D HEPATOCELLULAR CARCINOMA

22 1

Persing, D. H., Varmus, H. E., and Ganem, D. (1985). Proc. Natl. Acad. Sci. U.S.A. 82,34403444. Persing, D. H., Varmus, H. E., and Ganem, D. (1986).J. Viral. 60, 177-184. Persing, D. H., Varmus, H. E., and Ganem, D. (1987).J. Viral. 61, 1672-1677. Pfaff, E., Salfeld, J., Gmelin, K., Schaller, H., and Theilmann, L. (1987). Virology 158,456460. Pontisso, P., Poon, M. C., Tiollais, P., and Brechot, C. (1984). Br. A4ed.j. 288, 1563-1566. Popper, H., Shih, J. W. K., Gerin, J. L., Wong, D. C., Hoyer, B. H., London, W. T., Sly, D. L., and Purcell, R. H. (1981). Hepatology 1, 91-98. Popper, H., Shafritz, D. A., and Hoofnagle, J. H. (1987a). Hepatology 7, 764-772. Popper, H.. Roth, L., Purcell, R. H., Tennant, B. C., and Gerin, J. L. (1987b).Proc. Natl. Acad. Sci. U.S.A. 84, 866-870. Portis, J. L., Perryman, S., and McAtee, F. J. (1991).J. Viral. 65, 1877-1883. Pourcel, C., Louise, A., Gervais, M., Chenciner, N., Dubois, M. F., and Tiollais, P. (1982).J. Viral. 42, 100-105. Prange, R., Clernen, A,, and Streeck, R. E. (1991).J. Viral. 65, 3919-3923. Priel, E., Showalter, S. D., Roberts, M., Oroszlan, S., and Blair, D. G. (1991).]. Vzral. 65, 4137-4 14 1 . Pugh, J., Zweidler, A., and Summers, J. (1989).J. Viral. 63, 1371-1376. Puisieux, A., Lim, S, Groopman, J., and Ozturk, M. (1991). Cancer Res. 51, 6185-6189. Pulciani, S., Santos, E., Lauver, A. V., Long, L. K., Aaronson, S. A,, and Barbacid, M. (1982). Nature (London) 300, 539-542. Radziwill, G., Tucker, W., and Schaller, H. (199O).J. Viral. 64, 613-620. Raimondo, G., Stemler, M., Schneider, R., Wildner, G., Squadrito, G., and Will, H. (1990). J. Hepatol. 11, 374-380. Raney, A. K., Milich, D. R., and McLachlan, A. (1989).J. Viral. 63, 3919-3925. Raney, A. K., Milich, D. R., Eason, A. J., and McLachlan, A. (1990).J. Viral. 64,2360-2368. Raney, A. K., Milich, D. R., and McLahlan, A. (1991).J. Viral. 65, 4805-481 1 . Robinson, W. S. (1990).I n “Viral Hepatitis” (F. B. Hollinger, W. S. Robinson, R. H. Purcell, J. L. Gerin, and J. Ticehurst, eds.), pp. 39-71. Raven Press, New York. Robinson, W. S., Miller, R. H., and Marion, P. L. (1987). Hepatology, 7, 64s-73s. Rogler, C. E., and Summers, J. (1982).J. Viral. 44, 852-863. Rogler, C. E., and Summers, J. (1984).J . Viral. 50, 832-837. Rogler, C. E., Sherman, M., Su, C. Y., Shafritz, D. A., Summers, J., Shows, T. B., Henderson, A., and Kew, M. (1985). Science 230, 319-322. Rogler, C. E., Hino, O., and Su, C. Y. (1987). Hepatology 7, 74s-78s. Romanczuk, H., Villa, L. L., Schlegel, R., and Howley, P. M. (1991).j.Viral. 65,2739-2744. Roossinck, M., Jameel, S., Loukin, S., and Siddiqui, A. (1986).Mol. Cell. Biol. 6, 1393-1400. Roth, L., King, J. M., Hornbuckle, W. E., Harvey, H. J., and Tennant, B. C. (1985). Vet. Pathol. 22, 338-343. Roychoudhury, S., Faruqi, A. F., and Shih, C. (1991).J . Viral. 65, 3617-3624. Russnak, R., and Ganem, D. (1990). Genes Dev. 4, 764-776. Saito, I., Oya, Y., and Shimojo, H. (1986).J. Viral. 58, 554-560. Saito, I., Miyamura, T., Ohbayashi, A., Harada, H., Katayama, T., Kikuchi, S., Watanabe, Y., Koi, S., Onji, M., Ohta, Y., Choo, Q. L., Houghton, M., and Kuo, G. (1990). Prof. Natl. Acad..Sci. U.S.A. 87, 6547-6549. Salfeld, J., Pfaff, E., Noah, M., and Schaller, H. (1989). J. Viral. 63, 798-808. Sattler, F., and Robinson, W. S. (1979). J. Vzrol. 32, 226-233. Schek, N., Bartenschlager, R., Kuhn, C., and Schaller, H. (1991a). Oncogene 6, 17351744.

222

MARIE ANNICK BUENDIA

Schek, N., Fischer, M., and Schaller, H. (1991b). In “The Hepadnaviral Protein X” (A. McLachlan, ed.), pp. 181-192. CRC Press, Boca Raton, Florida. Schlicht, H. J., and Schaller, H. (1989).J. Virol. 63, 5399-5404. Schlicht, H. J.,,Salfeld, J., and Schaller, H. (1987).J. Virol. 61, 3701-3709. Schlicht, H. J., Radziwill, G., and Schaller, H. (1989). Cell (Cambridge, Mass.) 56, 85-92. Schlicht, H. J., von Brunn, A., and Theilmann, L. (1991). Hepatology 13, 57-61. Schodel, F., Sprengel, R., Weimer, T., Fernholz, D., Schneider, R., and Will, H. (1989). Adv. Viral Oncol. 8, 73-102. Schodel, F., Weimer, T., and Will, H. (1990).Biotest Bull. 4, 63-83. Schroder, C. H., and Zentgraf, H. (1990). Biochim. Biophys. Acta 1032, 137-156. Schweitzer, I. L., Dunn, A. E. G., Peters, R. L., and Spears, R. L. (1973). Am.J. Med. 55, 762-77 1. Scotto, J., Hadchouel, M., Hery, C., Alvarez, F., Yvart, J., Tiollais, P., and Brechot, C. (1983). Gut 24, 618-624. Scotto, J., Hadchouel, M., Wain-Hobson, S., Sonigo, P., Courouck, A. M., Tiollais, P., and Brechot, C. (1985).J. Infect. Dir. 151,610-617. Seeger, C., Ganem, D., and Varmus, H. E. (1984a). Proc. Natl. Acad. Scz. U.S.A. 81, 58495852. Seeger, C., Ganem, D., and Varmus, H. E. (1984b).J . Virol. 51, 367-375. Seeger, C., Ganem, D., and Varmus, H. E. (1986). Science 232,477-484. Seeger, C., Marion, P. L., Ganem, D., and Varmus, H. E. (1987).J. Virol. 61, 3241-3247. Seeger, C., Baldwin, B., Hornbuckle, W. E., Yeager, A. E., Tennant, B. C., Cote, P., Ferrell, L., Ganem, D., and Varmus, H. E. (1991).J. Virol. 65, 1673-1679. Seifer, M., Herman, K. H., and Gerlich, W. H. (1990). Virology 179, 287-299. Selten, G., Cuypers, H. T., Zijlstra, M., Melief, C., and Berns, A. (1984).EMBOJ. 3,32153222. Seto, E., Mitchell, P. J., and Yen, T. S. B. (1990). Nature (London) 344, 72-74. Setoguchi, M., Higuchi, Y., Yoshida, S., Nasu, N., Miyazaki, Y., Akizuki, S. I., and Yamamoto, S. (1989). Mol. Cell. Biol. 9, 4515-4522. Shafritz, D. A., Shouval, D., Sherman, H., Hadziyannis, S., and Kew, M. C. (1981). N. Engl. J . Med. 305, 1067-1073. Shaul, Y., and Ben-Levy, R. (1987). EMBOJ.6, 1913-1920. Shaul, Y., Ziemer, M., Garcia, P. D., Crawford, R., Hsu, H., Valenzuela, P., and Rutter, W. J. (1984).J. Vzrol. 51, 776-787. Shaul, Y.,Rutter, W. J., and Laub, 0. (1985). EMBO J . 4, 427-430. Shaul, Y., Garcia, P. D., Schonberg, S., and Rutter, W. J. (1986a). J. VzroL. 59, 731-734. Shaul, Y., Ben-Levy, R., and De Medina, T. (1986b). EMBOJ. 5, 1967-1971. Shih, C., Burke, K., Chou, M. J., Zeldis, J. B., Yang, C. S., Lee, C. S., Isselbacher, K. J., Wands, J. R., and Goodman, H. (1987).J. Virol. 61, 3491-3498. Shih, C. C., Stoye, J. P., and Coffin, J. M. (1988). Cell (Cambridge, Mass.) 53, 531-537. Shiozawa, M., Ochiya, T., Hatada, I., Imamura, T., Okudaira, Y., Hiraoka, H., and Matsubara, K. (1988). Oncogene 2, 523-526. Shirakata, Y., Kawada, M., Fujiki, O., Sano, H., Oda, M., Yaginuma, K., Kobayashi, M., and Koike, K. ( 1989). Jpn. J . Cancer Res. 80, 6 17-625. Siddiqui, A., Marion, P. L., and Robinson, W. S. (1981).J. Virol. 38, 393-397. Siddiqui, A., Jameel, S., and Mapoles,J. E. (1986). Proc. Natl. Acad. Sci. U.S.A. 83,566-570. Siddiqui, A.,Jameel, S., and Mapoles,J. (1987). Proc. Natl. Acad. Sci. U.S.A. 84,2513-2517. Simon, D., Knowles, B. B., and Weith, A. (1991). Oncogene 6, 765-770. Slagle, B. L., Zhou, Y. Z., and Butel, J. S. (1991). Cancer Res. 51,49-54.

HBV A N D HEPATOCELLULAR CARCINOMA

223

Snyder, R. L. (1979).In “Verhandlungsbericht des XXI Internationalen Symposiums uber die Erkrankungen der Zootiere,” pp. 325-334. Akademie-Verlag, Berlin. Snyder, R. L., and Summers, J. (1980). Cold Spring Harbor Conf. Cell Prolferation 7, 447457. Spandau, D. F., and Lee, C. H. (1988).J. Virol. 62, 427-434. Sprengel, R., Kuhn, C., Manso, C., and Will, H. (1984).J. Virol, 52,932-937. Sprengel, R., Kaleta, E. F., and Will, H. (1988).J. Virol. 62, 3832-3839. Sprengel, R., Schneider, R., Marion, P. L., Fernholz, D., Wildner, G., and Will, H . (1991). Nucleic Aczdv Res. 19, 4289. Standring, D. N., Rutter, W. J., Varmus, H. E., and Ganem, D. (1984). J. Virol. 50, 563571. Standring, D. N., Ou, J. H., and Rutter, W. J. (1986). Proc. Natl. Acad. Sci. U.S.A. 83,93389343. Standring, D. N., Ou, J. H., Masiarz, F. R., and Rutter, W. J. (1988). Proc. Natl. Acad. Sci. U.S.A. 85, 8405-8409. Stein, 0..Fainaru, M., and Stein, Y. (1972). Lab. Invest. 26, 262-269. Stevens, C. E., Taylor, P. E., Tong, M. J., Toy, P. T., and Vyas, G. N. (1984). I n “Viral Hepatitis and Liver Disease” (C. N. Vyas, J. L. Dienstag, and J. H. Hoofnagle, eds.), pp. 275-291. Grune & Stratton, New York. Su, T. S, Lai, C. J., Huang, J. L., Lin, L. H., Yank, Y. K., Chang, C., Lo, S. J., and Han, S. H. (1989).J. Virol. 63, 4011-4018. Summers, J. (1981). Hepatology 1, 179-183. Summers, j.,and Mason, W. S. (1982). Cell (Cambridge, Mass.) 29, 403-415. Summers, J., OConnell, A., and Millman, I. (1975). Proc. Natl. Acad. Sci. U.S.A. 72,4597460 I . Summers, J., Smolec, J. M., and Snyder, R. (1978). Proc. Natl. Acad. Sci. U.S.A. 75, 45334537. Summers, J., Smith, P. M., and Horwich, A. L. (199O).J. Virol. 64, 2819-2824. Summers, j.,Smith, P. M., Huang, M., and Yu, M. (1991).J. Virol. 65, 1310-1317. Sun, T. T., and Chu, Y. Y. (1984).J. Cell. Physiol., Suppl. 3, 39-44. Suzuki, T., Masui, N., Kajino, I., Saita, I., and Miyamura, T. (1989). Proc. Natl. Acad. Sci. U.S.A. 86, 8422-8426. Szmuness, W. (1978). Prog. Med. Virol. 24, 40-69. Szmuness, W., Harley, E. j.,Ikram, H., and Stevens, C. E. (1978).In “Viral Hepatitis” (G. N. Vyas, S. N. Cohen, and R. Schmid, eds.), pp. 297-320. Franklin Institute Press, Philadelphia, Pennsylvania. Tabor, E. (1991). Adv. Appl. Biotechnol. Ser. 13, 107-117. Tagawa, M., Robinson, W. S., and Marion, P. L. ( 1 987). J. Virol. 61, 2273-2279. Takada, S., and Koike, K. (1990). Proc. Natl. Acad. Sci. U.S.A. 87, 5628-5632. Takahashi, K., Machida, A., Funatsu, G., Nomura, M., Usuda, S., Aoyagi, S., Tachibana, K., Miyamoto, H., Imai, M., Nakamura, T., Miyakawa, Y., and Mayumi, M. (1983). J. Immunol. 130, 2903-2907. Takase, S., Takada, N., Enomoto, N., Yasuhara, M., and Takada, A. (1991). Hepatology 13, 876-88 1 . Tanaka, Y., Esumi, M., and Shikata, T. (1988). J. Med. Virol. 26, 7-14. Tashiro, F., Morimura, S., Hayashi, K., Makino, R., Kawamura, H., Horikoshi, N., Nemoto, K., Ohtsubo, K., Sugimura, K., and Ueno, Y. (1986). Biochem. Biophys. Res. Commun. 138,858-864. Taylor, P. E., Stevens, C. E., De Cordoba, S. R., and Rubinstein, P. (1988). In “Viral

224

MARIE ANNICK BUENDIA

Hepatitis and Liver Disease” (A. J. Zuckerman, ed.), pp. 198-200. Alan R. Liss, New York. Teeter, L. D., Becker, F. F., Chisari, F. V., Li, D., and Kuo, M. T. (1990). Mol. Cell. B i d . 10, 5728-5735. Tennant, B. C., Mrosovsky, N., McLean, K., Cote, P. J., Korba, B. E., Engle, R. E., Gerin, J. L., Wright, J., Michener, G. R., Uhl, E., and King, J. M. (1991). Hepatology 13, 12151222. T e d , S., Petit, M. A., and Brechot, C. (1991).J. Viral. 65, 5539-5543. Thiers, V., Nakajima, E., Kremsdorf, D., Mack, D., Schellekens, H., Driss, F., Goudeau, A., Wands, J., Sninsky, J., Tiollais, P., and Brechot, C. (1988).Lancet 2, 1273-1276. Thiesen, H. J., Bosze, Z., Henry, L., and Charnay, P. (1988).J. Virol. 62, 614-618. Tiollais, P., and Buendia, M. A. (1991). Sci. Am. 264, 47-54. Tiollais, P., Pourcel, C., and Dejean, A. (1985).Nature (London) 317, 489-495. Toh, H., Hayashida, H., and Miyata, T. (1983).Nature (London) 305, 827-829. Tokino, T., Fukushige, S., Nakamura, T., Nagaya, T., Murotsu, T., Shiga, K., Aoki, N., and Matsubara, K. (1987).J . Virol. 61, 3848-3854. Tong, S. P., Li, J. S., Vitvisky, L., and TrCpo, C. (1990). Virology 176, 596-603. Toshkov, I., Hacker, H. J., Roggendorf, M., and Bannasch, P. (1990).J. Cancer Res. Clin. Oncol. 116, 581-590. Tran, A., Kremsdorf, D., Capel, F., Housset, C., Dauguet, C., Petit, A. M., and Brechot, C. (1991).J. Virol. 65, 3566-3574. Transy, C., Fourel, G., Robinson, W. S., Tiollais, P., Marion, P. L., and Buendia, M. A. (1992).Proc. Natl. Acad. Sci. U.S.A. (in press). Treinin, M., and Laub, 0. (1987). Mol. Cell. Biol. 7, 545-548. Trowbridge, R., Fagan, E. A., Davison, F., Eddleston, A. L. W. F., Williams, R., Linskens, M. H. K., and Farzaneh, F. (1988).In “Viral Hepatitis and Liver Disease” (A. J. Zuckerman, ed.), pp. 764-768. Alan R. Liss, New York. Trueba, D., Phelan, M., Nelson, J., Beck, F., Pecha, B. S., Brown, R. J., Varmus, H. E., and Ganem, D. (1985). Hepatology 5, 435-439. Tsuda, H., Hirohashi, S., Shimosato, Y., Ino, Y., Yoshida, T., and Terada, M. (1989).Jpn.J. Cancer Res. 80, 196- 199. Tsuda, H., Zhang, W., Shimosato, Y., Yokota, J., Terada, M., Sugimura, T., Miyamura, T., and Hirohashi, S. (1990). Proc. Natl. Acad. Sci. U.S.A. 87, 6791-6794. Tsukamoto, H., Towner, S. J., Ciofalo, L. M., and French, S. W. (1 986). Hepatology 6, 8 14822. Tur-Kaspa, R., Burk, R. D., Shaul, Y., and Shafritz, D. A. (1986). Proc. Natl. Acad. Sci. U.S.A. 83, 1627-1631. Tuttleman, J. S., Pourcel, C., and Summers, J. W. (1986a).Cell (Cambridge, M a s . ) 47,451460. Tuttleman, J. S., Pugh, J. C., and Summers, J. W. (1986b). J. Virol. 58, 17-25. Twu, J. S., and Robinson, W. S. (1989).Proc. Natl. Acad. Sci. U.S.A. 86, 2046-2050. Twu, J. S., and Schloemer, R. H. (1987).J . Virol. 61, 3448-3453. Twu, J. S., Lee, C. H., Lin, P. M., and Schloemer, R. H. (P988).Proc. Natl. Acad. Sci. U.S.A. 85,252-256. Twu, J. S., Rosen, C. A., Haseltine, W. A., and Robinson, W. S. (1989).J. Viral. 63, 28572860. Uchida, T., Suzuki, K., Esumi, M., Arii, M., and Shikata, T. (1988). Cancer Res. 48, 15591565. Ueda, K., Tsurimoto, T., and Matsubara, K. (1991).J. Viral. 65, 3521-3529. Urban, M. K., OConnell, A. P., and London, W. T. (1985).J. Viral. 55, 16-22.

HBV AND HEPATOCELLULAR CARCINOMA

225

Uy, A,, Bruss, V., Gerlich, W. H., Kochel, H. G., andThomssen, R. (1986). Virology 155,8996. Valenzuela, P., Quiroga, M., Zaldivar, J., Gray, P., and Rutter, W. J. (1980). In ”Animal Viruses Genetics” (B. Fields, R. Jaenisch, and C. F. Fox,eds.), pp. 57-70. Academic Press, New York. van Lohuizen, M., Breuer, M., and Berns, A. (1989). E M B O J . 8, 133-136. Vaudin, M., Wolstenholme, A. J., Tsiquaye, K. N., Zuckerman, A. J., and Harrison, T. J. (1988).J. Gen. Virol. 69, 1383-1389. Vitvitski, L., Meyers, M. L., Sninsky,J. J., Berthillon, P., Chevalier, P., Sells, M. A., Acs, G., and Trepo, C. (1988). I n “Viral Hepatitis and Liver Disease” (A. J. Zuckerman, ed.), pp. 341-344. Alan R. Liss, New York. Wain-Hobson, S. (1984). In “Advances in Hepatitis Research (F. V. Chisari, ed.), pp. 4953. Masson, New York. Wands, J. R., Fujita, Y. K., Isselbacher, K. J., Degott, C., Schellekens, H., Dazza, M. C., Thiers, V., Tiollais, P., and Brechot, C. (1986). Proc. Natl. Acad. Sci. U.S.A. 83,660866 12. Wang, H. P., and Rogler, C. E. (1988). Cytogenet. Cell Genet. 42,72-78. Wang, H. P., and Rogler, C. E. (1991).J. Virol. 65,2381-2392. Wang, J., Chenivesse, X., Henglein, B., and Brechot, C. (1990). Nature (London) 343,555557. Wang, W., London, T., Lega, L., and Feitelson, M. A. (1991). Hepatology 14,29-37. Wang, Y., Chen, P., Wu, X., Sun, A. L., Wang, H., Zhu, Y. A,, and Li, Z. P. (199O).J. Virol. 64,3977-398 1. Weimer, T., Salfeld, J., and Will, H. (1987).J. Virol. 61,3109-31 13. Whittle, H. C., Bradley, A. K., McLauchlan, K., Ajdukiewicz,A. B., Howard, C. R., Zuckerman, A. J., and Mc Gregor, I. A. (1983). Lancet 1, 1203-1206. Will, H., Salfeld, J., Pfaff, E., Manso, C., Theilmann, L., and Schaller, H. (1986). Science 231, 594-596. Will, H., Reiser, W., Weimer, T., Pfaff, E., Biischer, M., Sprengel, R., Cattaneo, R.,and Schaller, H. (1987).J. Vzrol. 61,904-91 1. Wollersheim, M., Debelka, U., and Hofschneider, P. H. (1988). Oncogene 3, 545-552. Wright, J. A., Smith, H. S., Watt, F. M., Hancock, C., Hudson, D. L., and Stark, G. R. (1990). Proc. Natl. Acad. Sci. U.S.A. 87, 1791-1795. WU, H. L., Chen, P. J . , Tu, S. J., Lin, M. H., Lai, M. Y., and Chen, D. S. (1991).J. Virol. 65, 1680-1686. Wu, J. Y., Zhou, Z. Y., Judd, A., Cartwright, C. A,, and Robinson, W. S. (1990). Cell (Cambridge, Mass.) 63, 687-695. Xiong, Y., and Eickbush, T. H. (1990). EMBOJ. 9, 3353-3362. Yaginuma, K., and Koike, K. (1989).J. Virol. 63,2914-2920. Yaginuma, K., Kobayashi, M., Yoshida, E., and Koike, K. (1985).Proc. Natl. Acad. Sci. U.S.A. 82,4458-4462. Yaginuma, K., Kobayashi, H., Kobayashi, M., Morishima, T., Matsuyama, K., and Koike, K. (1987).J. Vzrol. 61, 1808-1813. Yee, J. K. (1989). Science 246, 658-661. Yokosuka, O., Omata, M., Zhou, Y. Z. Imazeki, F., and Okuda, K. (1985). Proc. Natl. Acad. Sci. U.S.A. 82, 5180-5184. Yuh, C. H., and Ting, L. P. (199O).J. Virol. 64,4281-4287. Zahm, P., Hofschneider, P. H., and Koshy, R. (1988). Oncogene 3, 169-177. Zentgraf, H., Herrmann, G., Klein, R., Schianz, P., Loncarevic, I., Herrman, D., Hiibner, K., and Schriider, C. H. (1990). Oncology 47, 143-149.

226

MARIE ANNICK BUENDIA

Zerial, M., Salinas, J., Filipski,J., and Bernardi, G. (1986).Nucleic Acids Res. 14,8373-8385. Zhou, D. X., and Yen, T. S. B. (1991). Mol. Cell. B i d . 11, 1353-1359. Zhou, S., and Standring, D. N. (1991).J. Virol. 65, 5457-5464. Zhou, Y. Z. (1980). Shanghai Med. J . 3, 641-644. Zhou, Y. Z., Butel, J. S., Li, P. J., Finegold, M. J., and Melnick, J. L. (1987).JNCI, J. Natl. Cancer Inst. 79, 223-231. Zhou, Y. Z., Slagle, B. L., Donehower, L. A., van Tuinen, P., Ledbetter, D. H., and Butel, J. S. (1988).J. Virol. 62, 4224-4231. Ziemer, M., Garcia, P., Shaul, Y., and Rutter, W. J. (1985).J. Virol. 53, 885-892. zur Hausen, H., and Schneider, A. (1987). I n “The Papovaviridae” (N. P. Salzman and P. M. Howley, eds.), pp. 245-263. Plenum, New York.

CYTOTOXIC T LYMPHOCYTES: SPECIFICITY, SURVEILLANCE, AND ESCAPE Andrew McMichael Institute of Molecular Medicine, John Radcliffe Hospital, University of Oxford, Oxford, OX3 9DU England

I. Introduction 11. The Molecular Basis of Peptide MHC Association 111. HLA-A2 Interaction with Influenza Matrix Peptide(58-66) IV. Antigen Processing V. CTL Function and Escape from CTL Recognition VI. Conclusions References

I. Introduction

Cytotoxic T lymphocytes (CTLs) are a major subpopulation of T lymphocytes that carry the CD8 glycoprotein and recognize foreign antigen presented by class I molecules of the major histocompatibility complex (MHC). They were first recognized in vitro as vigorous responders to allogeneic MHC molecules (Cerottini and Brunner, 1974) and then as cells that could lyse virus-infected target cells (Gardner et al., 1974). Specificity studies on the latter response led to the discovery of MHC restriction, in which virus-specific cytotoxic T lymphocytes recognize foreign antigens only in the context of self-MHC class I molecules (Zinkernagel and Doherty, 1975).Similar restrictions were demonstrated for CTLs specific for minor transplantation antigens (Fischer-Lindhal et al., 1989) A better understanding of MHC restriction came when it was shown that CTLs recognize peptide fragments of foreign antigens presented by class I MHC molecules (Townsend et al., 1986). This finding was followed by determination of the three-dimensional structure of HLA A2 (Bjorkman et al., 1987a,b). Foreign peptide material cocrystalized in a groove on the outer surface of the molecule (Bjorkman et al., 1987a,b). Cytotoxic T lymphocytes, as their name implies, lyse target cells. There is debate as to whether this is primarily achieved by release of enzymes in cytotoxic granules or by delivery of signals that direct target cells to apoptose (Martz and Howell, 1989). They can also release cytokines, particularly interferon-? (Morris et al., 1982), which can have 227 ADVANCES Ih’ CANCER RESEARCH, VOL. 59

Copyright 0 1992 hy Academic Press, Inc. All rights of reproduction in any form reserved.

228

ANDREW McMICHAEL

direct effects on the target cell (e.g., reduction in viral replication) as well as enhancing antigen processing and expression. Interferon-y may also recruit nonspecific mediators of immune responses such as natural killer cells. Other mediators may also be released and reduce virus replication (Brinchmann et al., 1990; Walker et al., 1986). Cytotoxic T lymphocytes react predominantly with internally derived proteins, processed within the cell and expressed on the surface as small peptides in class I MHC molecules (Townsend et al., 1986). Molecules within the cell that represent allelic differences, even by a single amino acid, can stimulate CTL responses (Fischer-Lindhal et al., 1989). Similarly, mutations within the cell, as are known to occur in tumors, could also trigger CTLs. Boon and colleagues (De Plaen et al., 1988) have demonstrated this for rejectable transplantable tumor variants of the mastocytoma P815 cell line in BALB/c mice (Maryanski et al., 1982; Uttyenhove et al., 1983; De Plaen et al., 1988; Degiovanni et al., 1988; Van den Eynde et al., 1989). These findings revive the concept of immune surveillance and immune rejection of tumors. Tumor-specific transplantation antigens are likely to be peptides bound to MHC molecules, inaccessible to, or poorly recognized by, antibodies. Thus in the modern synthesis of immune surveillance, cytotoxic T lymphocytes would scan the surfaces of many different types of cells, destroying cells bearing class I MHC molecules with peptides not expressed as part of the selfrepertoire. Such peptides could be derived from viruses, mutant selfproteins, o r abnormally expressed or processed self-proteins. This article explores the nature of the MHC peptide association, which has now reached a detailed level of understanding. T h e evidence for immune surveillance, particularly the strategies by which target cells escape CTL recognition, is examined. II. The Molecular Basis of Peptide MHC Association

T h e demonstration that cytotoxic T lymphocytes recognized peptide fragments of viral proteins presented by class I MHC molecules was first made by Townsend et al. (1986). Using transfected cells to express recombinant influenza virus proteins, virus-specific CTLs were shown to recognize influenza proteins that were expressed inside the cells, such as the polymerase PB2 and nucleoprotein (NP) (Bennink et al., 1982; Townsend and Skehel, 1982). Sites within NP that such CTLs recognize were partially mapped by the finding that fragments of NP, transfected into target cells, were recognized (Townsend et al., 1985). From here, it was shown that CTLs recognized short synthetic peptides representing appropriate epitopes (Townsend et al., 1985). Since then a large number

CYTOTOXIC T LYMPHOCYTES

229

of peptide epitopes have been described (McMichael et al., 1991). In general, it is clear that different class I MHC molecules present different peptides. Although there are some examples where the same peptide is presented by two class 1 molecules (Culmann et al., 1989), the peptide is usually long enough to contain two overlapping epitopes. It is also strikingly clear from human studies that MHC type (HLA type) determines the peptide to which T cells respond (McMichael et al., 1986). Thus, all individuals with HLA-A2 make a CTL response to the influenza matrix peptide(56-68) (McMichael et al., 1988); even mice transgenic for HLAA2 selectively respond to this peptide when challenged with influenza virus (Engelhard et al., 1991). Similar selection of antigenic determinants has been shown for HLA-B37 (McMichael et al., 1986) and -B27 (Huet et al., 1990) and it is reasonable to argue that this is a general rule. The implication is that different class I MHC molecules selectively bind to different peptides and these then trigger specific immune responses, provided the peptide differs from self. It is reasonable to assume that class I MHC molecules are filled with a variety of self-peptides (Rotzschke et al., 1990), but if these are normal in sequence, they should not trigger T cell responses because of T cell tolerance. The diversity of selfpeptides that fit into each class I molecule probably accounts for the strong allostimulation, where responding T cell clones are highly heterogeneous (Matzinger and Bevan, 1977). In contrast, analysis of T cell receptors for a single peptides plus self-MHC reveals that they can show restricted heterogeneity in both cx and P chains (Aebischer et al., 1990; Moss et al., 1991). Ill. HLA-A2 Interaction with Influenza Matrix Peptide(58-66)

There is now a good understanding of how HLA-A2 binds to an antigenic peptide epitope derived from influenza A matrix protein and it is instructive to consider this in some detail. The HLA-A2 crystal, containing unknown peptides has been resolved to the level of 2.6 A (Saper et al., 1991). This molecule has also been crystalized with the influenza A matrix peptide but X-ray diffraction data have not yet been obtained (Silver et al., 1991). However, a large number of functional experiments that have used mutant HLA-A2 molecules and peptides of altered sequence have revealed the basis of peptide binding. The groove in HLA-A2 is bordered by the cx cx helix and the cxp a helix and f3 sheet on the floor. Resolution at 2.6 reveals that there are six pockets within the groove, labeled A to F (Saper et al., 1991) (Fig. 1). The A and F pockets are almost invariant between different HLA types

1

45 116

Falk eta/. (1991) motif: 1 2 3 4 -L M E K matrix(58-66): G L L G B NP(84-92): K L G E HIV RT(481-485): I L K E

NP(383-391)

5 6 7 8 9 V

K

l!

L F V F T I F Y N Q M P V H G Y

S _R Y W A I R T B

FIG. I . Pocket structure of HLA class I molecules and orientation of peptides in the antigen-binding cleft. Top: The location of the six pockets, A-F, in HLA-A2 is indicated on the C, backbone structure of the a1 and a2 domains of HLA A2 (Bjorkman el al., 1987a,b; Saper e6 al., 1991). The positions of the critical and polymorphic residues 45 and 116 are shown. Amino acid 45 is methionine in A2 and glutamic acid in B27. Amino acid 116 is tyrosine in A2 and aspartic acid in B27. Middle: The predicted orientation of the influenza matrix peptide(58-66) in the cleft of HLA A2, with the isoleucine (I) at position 59 pointing its side chain into the B pocket and the side chain of leucine at position 66 pointing toward residue 116 of A2. To the right of the A2 structure is shown the amino acid motif described by Falk et al. (1991), showing the dominant amino acids in the eluted peptides from HLA A2 underlined and the weaker common residues below. Below the

CYTOTOXIC T LYMPHOCYTES

23 1

and each contains conserved hydrogen bond donors, particularly tyrosines and threonines. In recent experiments we have shown that if two of the conserved tyrosines in the A pocket are converted to phenylanines, peptide presentation is grossly impaired (Latron et al., 1991). The recently determined three-dimensional structure of HLA-B27 has revealed peptides in the groove that, unlike those in HLA-A2, were relatively homogeneous and could be partially resolved: it appeared that the carboxyl terminus of the peptide was in the F pocket and the amino terminus in the A pocket (Madden et al., 1991).As each of these pockets is conserved between HLA-A2 and B27 (and other class I molecules), this orientation is likely to be true for HLA-AP. Support comes from an experiment in which an important residue in HLA-A2, tyrosine at position 116, which lies in the floor of the cleft near the F pocket, was mutated to aspartic acid, which is smaller and acidic. This created an additional negatively charged pocket in the floor of the groove (Latron et al., 1991). This mutant A2 molecule was able to present to CTLs influenza matrix peptide(56-68) with positively charged arginine or ornithine at position 66; these peptides were only poorly presented by normal HLA-A2, requiring 100- to 1000-fold highler concentrations (Latron et al., 1991).This implies that residue 66 in the peptide, which is near, or at, its carboxyl terminus, binds close to the F pocket. These two lines of evidence, one structural and one functional, orientate the peptide with its N terminus in the A pocket and C terminus in the F pocket (Fig. 1). Further information comes from a recent analysis of peptides eluted from class I molecules. Rammensee and colleagues (Rotzschke et al., 1990) and Van Bleck and Nathenson (1990) have shown that peptides can be eluted from purified class I MHC molecules and that such peptides are eight or nine amino acids in length. Falk et al. (1991) sequenced these mixed peptides and were able to identify locus-specific “anchor” residues, at particular positions, which differ in peptides eluted from different class I molecules. For HLA-A2 the anchors were at residues 2 and 9. At position 2, the predominant amino acid was leucine and at 9, valine or leucine. We have tested nonamer peptides, within the residues motif are shown the sequences of three well-substantiated A2-restricted epitopes, all of which fit the motif pattern well. BNP,Influenza B virus nucleoprotein; HIV RT, human immunodeficiency virus reverse transcriptase. Bottom: The predicted position of the NP(383-392) peptide in HLA B27. The single-letteramino acid code is used in this figure: A, alanine; C, cysteine; D, aspartic acid; E, glutamic acid; F, phenylalanine; G, glycine; H, histidine; I, isoleucine; K, lysine; L, leucine; M, methionine; N,asparagine;P,proline; Q, glutamine; R, arginine; S, serine; T, threonine; V, valine; W, tryptophan; Y , tyrosine.

232

ANDREW McMICHAEL

57-68 sequence of the influenza A matrix protein, for recognition by HLA-A2-restricted CTLs. Peptide(58-66) was found to sensitize target cells for specific CTL recognition at picomolar concentrations, which is at least three orders of magnitude lower than any other matrix peptide tested (R. M. Moots, J. L. Strominger, and A. J. McMichael, unpublished results). Therefore, this is very probably the natural peptide epitope (Rotzschke et al., 1990).This would place the carboxyl leucine, at residue 9 in the peptide or 66 in the protein, into the F pocket. The isoleucine at position 2 in the peptide (59 in the sequence) would fit into the B pocket in HLA-A2. This pocket, which extends under the cxl helix, has the methionine at residue 45 at its apex and has been shown to contain electron density, probably leucine, in the A2 crystal (Saper et al., 1991). This pocket is important because, when methionine at residue 45 was changed to glutamic acid, a residue found in some B locus molecules, CTL recognition of the matrix peptide was abrogated (F. Latron et al., unpublished). The matrix peptide has isoleucine at the second position, which is one of the anchor positions of Falk et al. (1990), and consistent with the observed electron density in the B pocket of the A2 crystal. Therefore, our view is that the influenza A matrix peptide(58-66) would fit into the groove as an extended chain with the amino terminus in the A pocket, the isoleucine side chain in the B pocket, and the carboxyl terminus in the F pocket (Fig. 1). The valine at position 6 in the peptide might also be an anchor residue because valine was dominant at this position in eluted peptides (Falk et al., 1990). This leaves the side chains of amino acids 1, 3,4,6, 7,8, and 9 with the potential to point out of the groove to make contact with the T cell receptor. However, glycine at the amino terminus has no side chain, nor does glycine at position 4 and the leucine at position 9 probably points downward because it is affected by changing residues at position 116 in the floor (Latron et al., 1991). Thus a maximum of three side chains point out of the groove to make contact with the T cell receptor. As indicated above, these are restricted in heterogeneity (Moss et al., 1991). This understanding of how a peptide fits in the groove in HLA-A2 has parallels in the understanding of certain other peptides fitting into other class I molecules. HLA-B27 has been crystalized and resolved at high resolution (Madden et al., 1991) and it was possible to obtain structural information on the peptides bound because of their limited heterogeneity. The peptides had the amino termini in the A pocket and carboxyl termini in the F pocket, with an arginine in the negatively charged B pocket. In this molecule residue 116 is aspartic acid and it is likely that a positively charged side chain at the carboxyl terminus could fit here. The best defined peptide presented by HLA-B27 is influenza nucleopro-

CYTOTOXIC T LYMPHOCYTES

233

tein residues 383-391 (Huet et al., 1990). This has an arginine at position 2 in the B pocket and an arginine at the carboxyl terminus, probably close to the F pocket. A second HLA-B27 peptide from human immunodeficiency virus (HIV) Gag has structural homology with this peptide (Nixon et al., 1988).Similarly, four peptides presented by HLA B8 have a common sequence motif with arginine or lysine at positions 3 and 5 (J. Sutton, S. Rowland-Jones, and A. J. McMichael, unpublished results); one of these might fit into the D pocket which is negatively charged in B8. An analysis of the influenza nucleoprotein peptide(89- 101) presented by HLA-A68 also has a positively charged residue at position 9 that might approximate the aspartic acid at position 116, near the F pocket (like the arginine in the altered influenza matrix peptide interacting with the same amino acid in the mutated A2 molecule described above) (Cerundolo et al., 1991). These examples illustrate how detailed knowledge of peptide association with class I MHC molecules is becoming. This kind of information should lead to confident predictions of peptide epitopes in the near future. From the work described above and that of Falk et al. (1991), this will mean looking for amino acid sequence motifs, to identify anchor residues. The relative position of these will be different for different class I MHC molecules. Once these motifs have been defined, peptide predictions can be tested in readily accessible model systems, such as the CTL response to influenza virus. Then it should be possible to look for such sequence patterns in, for instance, tumor-associated proteins that are mutated or abnormally expressed, such as p53 (Fearon and Vogelstein, 1990), and test whether they have stimulated CTL responses in cancer patients. IV. Antigen Processing

The peptide epitopes defined above are derived from virus or other internal proteins. When peptides are added to cells for CTL recognition experiments, it is very likely that they bind to class I MHC molecules that have already reached the surface. The evidence for this is that presentation of virus, but not peptide, can be blocked by brefeldin A (BfA) (Yewdell and Bennink, 1989), an inhibitor of transport to and through the Golgi, and that an anti-A2 antibody binding to the cell surface can enhance peptide presentation (Bodmer et al., 1989). More direct evidence came from a number of mutant cell lines that fail to express stable class I MHC molecules, although they are synthesized in the endoplasmic reticulum (ER). In the cell line RMA-S, H2 Db heavy chain is found in the ER but, although it folds and reaches the cell surface, such

234

ANDREW McMICHAEL

complexes are unstable at 37°C (Ljunggren et al., 1990). Townsend et al. (1989) found that expression could be rescued by adding the H-2 Dbrestricted peptide NP(365-379) to these cells and that NP-specific CTLs could then recognize these cells. However, NP-specific CTLs failed to recognize these cells when infected with influenza A virus or appropriate recombinant vaccinia virus. This result implied a block in processing or transport of virus (or other cytoplasmic) protein or peptide into the ER for association with class I MHC molecules. More recently other cell lines with a similar phenotype have been identified. The first, 721.174, or its derivative T2, gives similar results to those described above (Cerundolo et al., 1990). Of great interest was the fact that these cell lines have a deletion in the class I1 region of the MHC. Spies and DeMars (1991) have recently defined another mutant, in which the phenotype is very similar, which has only a point mutation in the MHC class I1 region. Spies has shown that this mutation is one of the putative transporter genes recently mapped to the class I1 region of the HLA complex. There are two peptide transporters here and a component of a protease complex, all of which could be involved in antigen processing and presentation (Spies et al., 1990; Trowsdale et al., 1990; Deverson et al., 1990; Monaco et al., 1990). It is clear that peptide and class I MHC molecules associate in the ER, the peptides stabilizing or inducing folding of newly synthesized class I molecules and p2 microglobulin (Townsend et al., 1990). It is possible that the class I molecules fold and then peptide is inserted; alternatively, peptide may trigger folding of the heavy chain and p2 microglobulin might then bind. More likely, both routes could occur. The experimentally derived mutants have been extremely powerful tools in analyzing antigen processing and presentation. Recently there have been indications that there are naturally occurring polymorphisms in antigen processing that affect peptide antigen presentation. In rats, Livingstone et al. (1989) have described a gene or genes mapping into the class I1 region that determined which (putative) peptides were presented by identical rat class I MHC molecules. T cells specific for alloantigen or for the minor transplantation antigen HY demonstrated these differences. We have recently identified a human family where an HLAB27 variant molecule, B2702, presents neither influenza N P nor the peptide(381-393) epitope to CTLs, when both are presented by other cells with identical HLA-B2702 molecules (Pazmany et al., 1991). When B2702 was cloned from these cells and transfected into third party cells, presentation of peptide was restored. Therefore this polymorphism, which may be linked to the HLA-A3-B2702 haplotype, affects the function of this class I molecule. This result may represent polymorphisms in

CYTOTOXIC T LYMPHOCYTES

235

the same genes as are implicated in the rat by Livingstone et al. (1989) and that are deleted or altered in the mutant cells. Such polymorphisms would divide class I molecules into functional subtypes that provoke different immune responses. These could be important in determining whether disease-associated immune responses occur.

V. CTL Function and Escape from CTL Recognition Cytotoxic T lymphocytes are known to be important in the control of virus infections in vivo. Evidence comes from experiments in which CTL lines or clones have been transferred to infected animals and have facilitated recovery from lethal infection (Yap and Ada, 1978; Lin and Askonas, 1981). Mice, in which CD4+ T cells were depleted to levels that grossly impaired antibody responses but left CTL responses intact, recovered from virus infections (Buller et al., 1987; Lightman et al., 1987). In humans CTL activity has been correlated with clearance of infecting virus in vivo (McMichael and Askonas, 1978).In vitro, human CTLs limit the outgrowth of B lymphoblastoid cell lines after infection with Epstein-Barr virus (Moss et al., 1978) and similarly control HIV infection in growing ‘r cells (Walker et al., 1986; Kannagi et al., 1988; Tsubota et al., 1989). Cytotoxic T lymphocytes are also likely to be important in the control of neoplasia. Cytotoxic T lymphocytes specific for viruses that induce tumors in rodents have been demonstrated. These include Friend leukemia virus (Blank et al., 1976), Moloney murine sarcoma virus (Cerundolo et al., 1987), Moloney-murine leukemia virus (Collavo et al., 1982), simian virus 40 (SV40; Gooding and O’Connell, 1983), oncornavirus (Leclerc et al., 1973), and adenovirus (Kast et al., 1989). In a striking experiment Kast et al. (1989) showed that adenovirus-specific CTLs transferred to mice with large tumors caused complete regression of the malignancy. There is one human example in which CTLs specific for the adult T cell leukemia-associated virus HTLV- 1 have been described (Kannagi et al., 1984). Cytotoxic T lymphocytes specific for a variety of solid tumors in humans that are not known to be caused by viruses have also been described. T h e vast majority are specific for melanomas and have been derived in vitro from both peripheral blood and infiltrating lymphocytes (Mukherji and McAlister, 1983; Herin et al., 1978; Degiovanni et al., 1988; Itoh et al., 1988; Anichini et al., 1989; Knuth et al., 1989a; Wolfel et al., 1989; Crowley et al., 1991). At least two other types of solid tumor have been shown to elicit CTLs, ovarian cancer (Ferini et al., 1985) and sarcomas (Slavin et al., 1986). There is some experimental evidence in

236

ANDREW McMICHAEL

animals that CTLs can control solid tumor growth. In addition to the adenovirus-induced tumor described by Kast et al. (1989), rejection of the tum- mutant of the mastocytoma P815 in mice was mediated by CTLs (Maryanski et al., 1982; De Plaen et al., 1988). Overall these demonstrations of tumor-specific CTLs, which have been made repeatedly, despite considerable technical difficulties, are strongly indicative that their role in uiuo is important. Some of the most convincing evidence of the importance of CTLs in the control of both virus infections and tumors comes from investigation into escape mechanisms. There are two ways by which tumor cells or viruses can escape CTL recognition. The first occurs if the immune response is grossly suppressed, the second occurs by mutation in the virus or tumor. It is well recognized that generalized immunosuppression can be associated with an increased frequency of tumors (e.g., Cohen, 1991). This is clearly apparent in patients with HIV infection, where many types of tumor are increased in frequency, particularly Kaposi’s sarcoma and lymphomas associated with Epstein-Barr virus (Crawford et al., 1980; Beral et al., 1991). The latter is clearly infectious in origin and the former may be, although no agent responsible has been identified and there are other possibilities (Vogel et al., 1988). Persistent Epstein-Barr virus infection is ubiquitous in adults and tumors growing in HIV-seropositive and other immunocompromised patients are frequently large-cell lymphomas of the B lymphoblastoid cell line phenotype expressing eight latent gene products (Young et al., 1989; Thomas et al., 1990; Cohen, 1991). This contrasts with the expression of only EBNA 1 in Burkitt lymphomas, a phenotype found rarely in these patients (Rowe et al., 1987). The frequency of these tumors in HIVseropositive and other immunodeficient patients may reflect the fact that persistent infection with Epstein-Barr virus (EBV) is common, so that they are the first tumors to occur. It is relevant that this type of lymphoma also occurs when lymphocytes from normal EBV-seropositive donors are used to reconstitute mice with severe combined immunodeficiency (Mosier et al., 1988; Rowe et al., 1991). Human immunodeficiency virusseropositive patients have early impairment of T helper cell function, nonspecific activation of B cells, and later impairment of CTL function. Therapeutic immunosuppression is also associated with other tumors, particularly skin cancers, some of which are papilloma virus associated (Pecqueux et al., 1990). Thus the associations between immunosuppression and at least some forms of cancer implicate the immune system in immune surveillance and specifically in an antitumor function. However, virus-associated cancers, in which T cell control would be expected to be important, appear to dominate the scene and it is likely that T cell

CYTOTOXIC T LYMPHOCYTES

237

immunity is not the whole story, not least because athymic nude mice are not unusually susceptible to spontaneous tumors. Natural killer cells, a detailed discussion of which is beyond the scope of this article, constitute one additional defense mechanism that may be as important as specific T cells. The second type of escape mechanism involves changes in the target antigens or in other molecules associated with CTL recognition of targets. Again, lessons come from parallel studies on CTL responses to viruses. Virus-specific CTLs tend to recognize conserved internal virus proteins (Gotch et al., 1987). These are probably conserved because there is little or no immune selection (because when cells are infected by thousands of virions, all would have to mutate to escape from specific CTLs). This contrasts with surface glycoproteins of viruses such as influenza, which are strongly selected by neutralizing antibodies; one escape mutant may be enough to initiate the infection. For CTLs to exert selective pressure on viral proteins, responding T cell populations would have to be limited in heterogeneity, preferably monoclonal, and cells would need to be infected with single virus copies. These conditions are unusual in vi-clo, although they have been contrived in mice transgenic for a lymphocytic choriomeningitis (LCMV)-specificCTL receptor (Pircher et al., 1990). A natural example, however, has recently been found (Phillips et ul., 1991). Cytotoxic T lymphocytes specific for HIV concentrate on relatively few epitopes, with Gag being a major target antigen. Several peptide epitopes have been mapped in this and other proteins and, in one series, there was clear heterogeneity in epitope sequences, presented by HLA-B8, in Gag. Furthermore, when peptides were made for these sequences, CTLs discriminated between them and in some instances were unable to respond to sequences known to be present in the same patients at the same time. These results strongly imply that CTLs drive escape mutation within this virus, HIV. However, in contrast to HLA-B8-restricted epitopes, no escape mutants have been found to be associated with an epitope in HIV Gag that is presented by HLA-B27 (Phillips et al., 1991;Wain-Hobson, personal communication). Therefore the ability of the virus to escape depends on the epitope responsible, which in turn will depend on HLA type. There is good evidence that tumor cell lines grown in nitro can escape from CTL control by mutation of the antigen recognized. This has been shown by analyzing tumor variants that escape rejection and by incubating mutagen-treated tumor cell lines with tumor-specific CTL clones (Uttyenhove et al., 1983; Degiovanni et al., 1988; Knuth et al., 198913). This technique has been used as the first step in identification of the internal antigens recongnized by the CTLs. These examples indicate

238

ANDREW McMICHAEL

that this kind of mutation can occur. It is worth noting that for the mutation to be successful to the tumor it should not have a deleterious effect on the transformed phenotype. An alternative escape mechanism is to change expression of HLA molecules. Adenovirus types 5 and 12 down-regulate expression of class I MHC antigens and so escape from CTL recognition (Signas et al., 1982; Bernards et al., 1983; Schrier et al., 1983; Anderson et al., 1985). There is substantial evidence that HLA gene expression is often abnormal in tumor cells and tumor cell lines (Lampson et al., 1983; Doyle et al., 1985; Smith et al., 1988, 1989; Momberg et al., 1989a,b; Vanky et al., 1990). For instance, in colonic cancer cells, class I antigens such as HLAA2 are frequently down regulated or absent from tumor cells in vivo (Momberg et al., 1989a; Smith et al., 1989). In a recent study, several types of abnormal expression were identified; these included single allele loss, locus loss, total heavy chain loss, and total p2 microglobulin loss (Hill et al., 1991). Many of the phenotypes observed require mutations or deletions on both chromosomes, implying that there is very strong selective pressure to achieve the common end results. However, it is important that these mechanisms be worked out. Disordered transcriptional or posttranscriptional control might be secondary to more generally disrupted gene regulation in tumor cells and particular HLA alleles may be more sensitive. In the extravillous trophoblast, HLA-A and -B antigens are not expressed but HLA-G is present and a similar phenotype occurs on the choriocarcinoma cell line Bewo (Ellis et al., 1990); perhaps some of the loss of HLA-A and -B expression in other tumors may reflect this type of control, which may be at the DNA level. Also there are indications that folding, assembly, and transport of all class I molecules do not occur at the same rate (Williams et al., 1985; Cerundolo et al., 1990; Hill et al., 1991; H. Ploegh, personal communication). Different class I alleles may therefore be differently susceptible to changes in the cell environment. Nevertheless, whether abnormal MHC class I expression is due to direct effects on MHC genes o r is a consequence of other changes in the cell, the end result should give the cells the advantage of resistance to recognition by CTLs. Overall, it seems unlikely that impaired expression of class I MHC could be a neutral effect. The enormous outgrowth of cells with bizarre HLA phenotypes strongly implies selective pressure and this is very likely to be mediated by CTLs. If tumors have down-regulated class I MHC to escape CTL recognition, class I-negative cells should be more tumorigenic. This has been found in murine models in which class I-negative cells transfer or metastasize more frequently, and this effect has been reversed by transfection of class I genes into such cells (Alon et al., 1987; Gopas et al., 1989). However, this picture is not always that simple, because Alon et al. (1987)

CYTOTOXIC T LYMPHOCYTES

239

also showed that transfection of other H-2 genes could enhance metastatic potential, probably by an effect on TUS oncogene expression. An additional complicating factor is the enhanced susceptibility of class Inegative tumor cells to natural killer cells (Karre et al., 1986). Thus total class I MHC loss may result in tumors for which there is a back-up protective mechanism. Such a safety net might not apply to selective allele loss, which seems to be more common in some series of tumors examined. Other mechanisms for CTL escape is impaired expression of accessory molecules. Expression of the adhesion molecule ICAM-1 is often abnormal on a variety of tumors (Vanky et al., 1990). Colonic cancer cells that fail to express LFA-3, another important adhesion molecule, have been described (Smith et al., 1989). This could be achieved by regulatory effects acting on both LFA-3 genes. Alternatively, if low expression is due to changes in the structural gene both chromosomes need to be affected, implying strong selective pressure. On the other hand, increased expression of certain adhesion molecules may be associated with metastasis (Johnson et al., 1988). It is perhaps significant that, while Epstein-Barr virus enhances expression of LFA-3 and ICAM-1 (Gregory et al., 1988) as well as other surface markers, including the virus receptor CD21 (Aman et al., 1990), if all EBV-transforming genes are expressed in Burkitt’s lines in which only EBNA-1 is expressed, these are not up regulated (Gregory et al., 1988). Thus, a potential protective mechanism, which might favor chronic low-grade persistence of a virus, is bypassed in this form of malignancy. One consequence of CTLmediated control of tumors should be that HLA type would have an influence, because as discussed above HLA molecules select epitopes. If CTL responses were confined to mutant sequences in oncogenic proteins such as Ras or p53, only individuals with certain HLA types should respond and these might be more resistant to cancer. However, such mutations are relatively few, and if CTL responses were limited to these sequences, relatively few individuals would make antitumor CTL responses. In fact, analysis of the specificity of antimelanoma CTL clones (Degiovanni et al., 1988; Knuth et al., 1989a; Van den Eynde et al., 1989) has revealed that there are several distinct epitopes presented by different HLA antigens. Of these epitopes some could be immunoselected against but one was resistant, possibly because it was involved in the transformed phenotype. It may be relevant that p53 expression is altered by mutation and this might expose the whole molecule to a different route of processing and thus normal sequences could be epitopes, although no CTL epitope has yet been defined in this protein. Similarly, a gene that is not normally expressed in a tissue could provide normal sequence epitopes. The latter has been

240

ANDREW McMICHAEL

found to be true for one of the antigens detected by CTLs in one of the p815-turn- mutants (Van den Eynde, 1991). If this type of antigen were common, HLA type would have less influence on whether or not a CTL response was made; most HLA types would be capable of responding and antitumor CTLs would be a common defense against cancer. These considerations also have implications for association between HLA type and tumor susceptibility. If epitopes are limited to mutant sequences, there should be strong HLA associations. If a variety of proteins are abnormally processed in tumor cells, HLA type associations should be less prominent. So far there have been relatively few convincing studies of HLA associations with tumor susceptibility and virtually none for tumor resistance. The HLA-B5, 35, 18, 15 group has been associated with susceptibility to Hodgkin’s disease but only very marginally (Amiel, 1967). There are no substantial associations with resistance. This may be because it is very difficult to demonstrate HLA associations with resistance, particularly where protective HLA alleles are rare, because they have not been selected for. In a study by Hill et al. (1991), an association between HLA-Bw53 and protection from severe malaria was demonstrated, but the study required 2000 children (including controls) to be HLA typed. In this case the protective HLA type had an antigen frequency of 25%, probably because it was indeed selected in this population. A similar study for tumor resistance might require several thousand patients to obtain statistically clear data that would demonstrate an association with resistance for a rare HLA allele. Therefore such associations might well exist but have yet to be demonstrated. VI. Conclusions

The cytotoxic T lymphocyte is a powerful component of the immune response. The molecular details of specificity are being unraveled and this may lead to prediction of epitopes and thus ready demonstration of CTL to any self or foreign protein. However, as the specificity is studied in more detail it is clear that targeted viruses and cells have developed strategies to escape recognition. This implies strong selective pressure in uiuo, which is a clear indicator that these cells exert a major protective role for the host. Their role, in the broadest sense, is clearly that of immune surveillance. They explore cell surfaces, monitoring MHC class I molecules with their inserted peptides. Self-reactive CTLs are obviously deleted or anergized. Those that bind above a threshold affinity to class I molecules containing foreign peptides react vigorously and destroy those cells. The diversity of a class I molecule ensures that the species can cope with most foreign invaders and probably most mutant

CYTOTOXIC T LYMPHOCYTES

24 1

self proteins. However, some mutant proteins may fail to evoke an immune response because no new petides are inserted into class I HLA molecules. On other occasions, a CTL response is stimulated but cells of viruses escape CTL recognition. This is probably an important part of tumorigenesis.

ACKNOWLEDGMENTS I am grateful to my colleagues for many helpful discussions, particularly Jack Strominger, Rob Moots, and Frances Gotch, and to Ann Hill and Nick Murray, who also critically read the manuscript. My thanks also go to Margaret Fletcher, who helped to prepare the manuscript and to the MRC for grant support.

REFERENCES Aebischer, T., Oehen, S., and Hengartner, H. (1990). Eur.J. Zmmunol. 20, 523-531. Alon, Y., Hammerling, G. J., Segal, S,. and Bar-Eli, M. (1987). CancerRes. 47, 2553-2557. Aman, P., Rowe, M., Kai, C., Finke, J,, Rymo, L., Klein, E., and Klein, G. (1990). Int. 1. Cancer 45, 77-82. Amiel, J. L. (1967). Hidocompat. Test. 3, 14-24. Andersson, M., Paabo, S., Nilsson, T., and Peterson, P. A. (1985). Cell (Cambridge, Mass.) 43, 215. Anichini, A,, Mazzocchi, A., Fossati, G., and Parmiani, G. (1989).J. Zmmunol. 142, 3692370 1. Bennink, J. R., Yewdell, J. W., and Gerhard, W. (1982). Nature (London) 296, 75-76. Beral, V., Peterman, T., Berkelman, R., and Jaffe, H. (1991). Lancet 337, 805-809. Bernards, R., Schrier, P. I., Houweling, A., Bos, J. L., van der Eb, A. J., Zisjltra, M., and Melief, C. J. M. (1983). Nature (London) 305, 776. Bjorkman, P., Saper, M., Samraoui, B., Bennett, W., Strominger, J., and Wiley, D. (1987a). Nature (London) 329, 506-5 1 1. Bjorkman, P., Saper, M., Samraoui, B., Bennett, W., Strominger, J., and Wiley, D. (1987b). Nature (London) 329, 512-519. Blank, K. J., Freedman, H. A,, and Lilly, F. (1976). Nature (London) 260, 250-252. Bodmer, H., Ogg, G., Gotch, F., and McMichael, A. (1989). Nature (London) 342,443-446. Brinchmann, J. E., Gaudernack, G., and Vartdal, F. (1990). J. Immunol. 144, 2961-2966. Buller, R. M. L., Holmes, K. L., Hugin, A., Frederickson, T. N., and Morse, H. C. (1987). Nature (London) 328, 77-79. Cerottini, J.-C., and Brunner, K. T. (1974). Adv. Zmmunol. 18, 67-132. Cerundolo, V., La Haye, T., Horvath, C., Zanovello, P., Collavo, D., and Engers, H. D. ( 1987). Eur. J. Immunol. 17, 173- 178. Cerundolo, V., Alexander, J., Anderson, K., Lamb, C., Cresswell, P., McMichael, A., Gotch, F., and Townsend, A. (1990). Nature (London) 345, 449-452. Cerundolo, V., Tse, A. G. D., Salter, R. D., Parham, P., and Townsend, A. (1991). Proc. R . SOC.London, B Ser. 244, 169-177. Cohen, J. I. (1991). Medicine (Baltimore) 70, 137-160. Collavo, D., Ronchese, F., Zanovello, P., Biasi, G., and Chieco-Bianchi, L. (19821.1. Immunol. 128, 774-783. Crawford, D. H., Thomas, J. A., Janossy, G., Sweny, P., Fernando, 0. N., Moorgead, J. F., and Thompson, J. H. (1980). Lancet 1, 135551356,

242

ANDREW McMICHAEL

Crowley, N. J., Darrow, T. L., Quinn-Allen, M. A., and Seigler, H. F. (1991).J. Immunol. 146, 1692-1699. Culmann, B., Gomard, E., Kieny, M.-P., Guy, B., Dreyfus, F., Saimot, A.-G., Sereni, D., and Levy, J.-P. (1989). Eur. J. Immunol. 19, 2383-2386. Degiovanni, G., Lahaye, T., Herin, M., Hainaut, P., and Boon, T. (1988). Eur. J. Immunol. 18,671-676. De Plaen, E., Lurquin, C., Van Pel, A., Mariame, B., Szikora, J. P., Wolfel, T., Sibille, C., Chomez, P., and Boon, T. (1988). Proc. Natl. Acad. Scz. U.S.A. 85, 2274-2278. Deverson, E. V., Gow, J. R., Coadwell, W. J., Monaco, J. J., Butcher, G. W., and Howard, J. C. (1990). Nature (London) 348, 738-74 1. Doyle, A., Martin, W. J., Funa, K., Gazdar, A., Carney, D., Martin, S. E., Linnoila, I., Cuttitta, F., Mulshine, J., Bunn, P., and Minna, J. (1985).J. Exp. M e d . 161, 1135-1151. Ellis, S. A., Palmer, M. P., and McMichael, A. J. (1990).J. Immunol. 144, 731-735. Engelhard, V. H., Lacy, E., and Ridge, J. P. (1991).J. Immunol. 146, 1226-1232. Falk, K., Rotzschke, O., and Rammensee, H.-G. (1990). Nature (London) 348, 248-251. Falk, K., Rotzschke, O., Stevanovic, S., Jung, G., and Rammensee, H.-G. (1991). Nature (London) 351, 290-296. Fearon, E. R., and Vogelstein, B. (1990). Cell 61, 759-767. Ferrini, S., Biassoni, R., Moretta, A,, Buzzone, M., Nicolin, A., and Moretta, L. (1985).Int. J . Cancer 36, 337-343. Fischer-Lindhal, K., Hermel, E., Loveland, B., Richards, S., Wamg, C.-R., and Yonkawa, H. (1989). Cold Spring Harbor Symp. Qmnt. Bwl. 54,563-569. Gardner, I., Bowern, N. A., and Blanden, R. V. (1974). EUT.J. Immunol. 4, 68-72. Gooding, L. R., and OConnell, K. A. (1983).J. Immunol. 131, 2580-2586. Gopas, J., Rager-Zisman, B., Bar-Eli, M., Hammerling, G. J., and Segal, S. (1989). Ada. Cancer Res. 53, 89-115. Gotch, F. M., McMichael, A. J., Smith, G. L., and Moss, B. (1987).J. Exp. M e d . 165, 165. Gregory, C. R., Murray, R. J., Edwards, C. F., and Rickinson, A. B. (1988).J. Exp. M e d . 167, 1811-1824. Herin, M., Lemoine, C., Weynants, P., Vessiere, F., Van Pel, A., Knuth, A,, Devos, R., and Boon, T. (1987). Int. J. Cancer 39, 390-396. Hill, A. V. S., Allsopp, C. E. M., Kwiatkowski, D., Anstey, N. M., Twumasi, P., Rowe, P. A., Bennett, S., Brewster, D., McMichael, A. J., and Greenwood, B. M. (1991). Nature (London) 352, 595-600. Huet, S., D. F. Nixon, Rothbard, J., Townsend, A. R. M., Ellis, S. A., and McMichael, A. J. (1990). Int. Immunol. 2, 311. Itoh, K., Platsoucas, C. D., and Balch, C. M. (1988).J. Exp. M e d . 168, 1419-1441. Johnson, J. P., Stade, B. G., Holzman, B., Schwable, W., and Reithmuller, G. (1988). Proc. NatE. Acad. Sci. U.S.A. 86, 641-645. Kannagi, M., Sugamura, K., Kinoshita, K. I., Uchino, H., and Hinuma, Y. (1984). J. Immunol. 133, 1037-1 041. Kannagi, M., Chalifoux, L. V., Lord, C. I., and Letvin, N. L. (1988).J. Immunol. 140,22372242. Karre, K., Ljunggren, H. G., Piontek, G., and Kiessling, R. (1986). Nature (London) 319, 675. Kast, W. M., Offringa, R., Peters, P. J., Voordouw, A. C., Meloen, R. H., van der Eb, A. J., and Melief, C. J. (1989). Cell (Cambridge, Muss.) 59, 603-614. Knuth, A., Danowski, B., Oettgen, H. F., and Old, L. J. (1989a).Proc. Natl. Acad. Sci. U.S.A. 81, 3511-3515. Knuth, A., Wolfel, T., Klehmann, E., Boon, T., and M. z. B. K-H., (1989b).Proc. Natl. Acad. Sci. U.S.A. 86, 2804-2808.

CYTOTOXIC T LYMPHOCYTES

243

Lampson, L. A,, Fisher, C. A,, and Whelan, J. P. (1983).J . Zmmunol. 130, 2471-2478. Latron, F., Moots, R., Rothbard, J., Garrett, T., Strominger, J. L., and McMichael, A. J. (1991). Proc. Natl. Acad. Scz. U.S.A. 88, 11325-11329. Leclerc, J. C., Gomard, E., Plata, F., and Levy, J.-P. (1973). Znt. J. Cancer 11, 426-432. Lightman, S., Cobbold, S., Waldmann, H., and Askonas, B. A. (1987). Immunology 62, 139-144. Lin, Y., and Asknoas, B. A. (1981).J. Exp. Med. 154, 225-234. Livingstone, A. M., Powis, S. J., Diamond, A. G., Butcher, G. W., and Howard, J. C. (1989). J. Exp. Med. 170, 777-795. Ljunggren, H. G., Stam, N. J., Ohlen, C., Neefjes, J. J., Hoglund, P., Heemels, M. T., Bastin, J., Schumacher, T. N., Townsend, A., Karre, K., et al. (1990). Nature (London) 346, 476-480. Madden, D. R., Gorga, J. C., Strominger, J. L., and Wiley, D. C. (1991). Submitted for publication. Martz, E., and Howell, D. M. (1989). Zmmunol. Today 10, 79-85. Maryanski, J. L., Van Snick, J., Cerottini, J.-C., and Boon, T. (1982). Eur. J. Zmmunol. 12, 40 1-406. Matzinger, P., and Bevan, M. (1977). Cell. Immunol. 29, 1 . McMichael, A. J., and Askonas, B. A. (1978). Eur. J. Immunol.8, 705. McMichael, A. J., Gotch, F. M., and Rothbard, J. (1986).J. Exp. Med. 164, 1397-1406. McMichael, A. J., Gotch, F. M., Santos-Aguado, J., and Strominger, J. L. (1988).Proc. Natl. Acad. Sci. U.S.A. 85, 9194-10006. McMichael, A. J., Askonas, B. A,, and Frelinger. (1991). In “Antigen Processing and Presentation by Class I MHC” (J. McCluskey, ed.). CRC Press, Boca Raton, Florida. Momberg, F., Degener, T., Bacchus, E., Moldenhauer, G., and Hammerling, G. J. (1989a). Znt. J. Cancer 37, 179-184. Momberg, F., Ziegler, A., Harpprecht, J., Moller, P., Moldenhauer, G., and Hammerling, G . J. (198913).J . Immunol. 142, 352-358. Monaco, J., Cho, S., and Attaya, M. (1990). Science 250, 1723-1726. Morris, A. G., Lin, Y.-L., and Askonas, B. A. (1982). Nature (London) 295, 150-152. Mosier, D. E., Gulizia, R. J., Bavid, S. M., and Wilson, D. B. (1988). Nature (London) 335, 256-259. Moss, D. J., Rickinson, A. B., and Pope, J. H. (1978). Znt. J. Cancer 22, 662-668. Moss, P. A. H., Moots, R.J., Rosenberg, W. M. C., Rowland-Jones, S. J., McMichael, A. J., and Bell, J. I. (1991). Proc. Natl. Acad. Scz. U.S.A. 88, 8987-8991. Mukherji, B., and McAlister, T. J. (1983).J. Exp. Med. 158, 240-245. Nixon, D. F., Townsend, A. R. M., Elvin, J. G., Rizza, C. R., Gallwey, J., and McMichael, A. J. (1988). Nature (London)336, 484-487. Pazmany, L., Rowland-Jones, S., Huet, S., Sutton, J., Brooks, J., and McMichael, A. J. (1991). Submitted for publication. Pecqueux, J. C., Schwarz, A., Dieckmann, K. P., and Offermann, G. (1990). Urol. Znt. 45, 290-292. Phillips, R., Rowland-Jones, S., Nixon, D. F., Gotch, F. M., Edwards, J. P., Ogunlesi, A. O., Elvin, J. G., Rothbard, J., Bangham, C. R. M., Rizza, C. R., and McMichael, A. J. (1991). Nature (London) 354, 453-459. Pircher, H., Moskphidis, A., Rohrer, U., Burki, K., Hengartner, H., and Zinkernagel, R. M. (1990). Nature (London) 346, 629-233. Rotzschke, O., Falk, K., Deres, K., Schild, H., and Rammensee, H. G. (1990). Nature (London) 348, 252. Rowe, M., Rowe, D. T., Gregory, C. D., Young, L. S., Farrelli, P. J., Rupani, H., and Rickinson, A. B. (1987). EMBOJ. 6, 2743.

244

ANDREW McMICHAEL

Rowe, M., Young, L. S., Crocker, J., Stokes, H., Henderson, S., and Rickinson, A. B. (1991). J . Exp. Med. 173, 147-158. Saper, M. A., Bjorkman, P. J., and Wiley, D. C. (1991).J. Mol. B i d . (in press). Schrier, P. L., Bernards, R., Vaessen, R. T. M. J., Houweling, A,, and van der Eb, A. J. (1983). Nature (London) 305, 771-775. Signas, C., Katze, M.G., Persson, H., and Philipsson, L. (1982). Nature (London) 299, 175178. Silver, M. I., Parker, K. C., and Wiley, D. C. (1991). Nature (London) 350, 619-623. Slavin, S. F., Lackman, R. D., Ferrone, S., Kiely, P. E., and Mastrangelo, M. J, (1986).J. Immunol. 137, 3042-3048. Smith, M. E. F., Bodmer, W. F., and Bodmer, J. G. (1988). Lancet 1, 823-824. Smith, M. E. F., Marsh, S. G. E., Bodmer, J. G., Gelsthorpe, K., and Bodmer, W. F. (1989). Proc. Natl. Acad. Sci. U.S.A. 86, 5557-5561. Spies, T., and DeMars, R. (1991). Nafure (London) 351, 323-325. Spies, T., Bresnahan, M., Bahram, S., Arnold, D., Blanck, G., Mellins, E., Pious, D., and DeMars, R. (1990).Nature (London) 348, 744-746. Thomas, J. A,, Hotchin, N. A,, Allday, M. J., Amlot, P., Rose, M., Yacoub, M., and Crawford, D. H. (1990). Transplantation 49, 944. Townsend, A. R. M., and Skehel, J. J. (1982). Nature (London) 300, 655. Townsend, A. R. M., Gotch, F. M., and Davey, J. (1985). Cell (Cambridge, Mars.) 42,457467. Townsend, A., Rothbard, J., Gotch, F., Bahadur, B., Wraith, D., and McMichael, A. (1986). Cell (Cambridge, Mars.) 44,959-968. Townsend, A., Ohlen, C., Bastin, J., Ljunggren, H. G., Foster, L., and Karre, K. (1989). Nature (London) 340, 443-448. Townsend, A., Elliott, T., Cerundolo, V., Foster, L., Barber, B., and Tse, A. (1990). Cell 62, 285-295. Trowsdale, J., Hanson, I., Mockridge, I., Beck, S., Townsend, A. R. M., and Kelly, A. (1990). Nature (London) 348, 74 1-744. Tsubota, H. C., Lord, C. I., Watkins, D. I., Morimoto, C., and Letvin, N. L. (1989).J. Erp. Med. 169, 1421-1434. Uttyenhove, C., Maryanski, J., and Boon, J. (1983).J. Exp. Med. 157, 1040-1052. Van Bleek, G. M., and Nathenson, S. M. (1990). Nature (London) 348, 213. Van den Eynde, B., Hainaut, P., Herin, M., Knuth, A., Lemoine, C., Weynants, P., van der Bruggen, P., Fauchet, R.,and Boon, T. (1989). Znt. /. Cancer 44, 634-640. Vanky, F., Wang, P., Patarroyo, M., and Klein, E. (1990). Cancer Immunol. Immunother. 31, 19-27. Vogel, J., Hinrichs, S. H., Reynolds, R. K., Luciw, P. A., and Jay, G. (1988). Nature (London) 335, 606. Walker, C. M., Moody, D. J., Stites, D. P., and Levy, J. A. (1986). Science 234, Klein review. Williams, D. B., Swiedler, S. J., and Hart, G. W. (1985).J. Cell Bzol. 101, 725-734. Wolfel, T., Klehmann, E., Muller, C., Schutt, K.-H., Meyer zum Buschenfelde, K.-H., and Knuth, A. (1989).J. Exp. Med. 170, 797-810. Yap, K. L., and Ada, G. L. (1978). Nature (London) 273, 238-240. Yewdell, Y. W., and Bennink, J. R. (1989). Science 244, 1072. Young, L., Alfieri, C., Henessey, K., Evans, H., OHara, C., Anderson, K. C., Ritz, J., Shapiro, R. S., Rickinson, A., Kieff, E., and Cohen, J. I. (1989). N . Eng1.J. Med. 321, 1080.

Zinkernagel, R. M., and Doherty, P. C. (1975).J. Exp. Med. 141, 1427-1436.

CANCER IMMUNOTHERAPY: ARE THE RESULTS DISCOURAGING? CAN THEY BE IMPROVED? Eli Kedar* and Eva Kleint T h e Lautenberg Center for General and Tumor Immunology, the Hebrew University-Hadassah Medical School, Jerusalem, Israel 'Department of Tumor Biology, Karolinska Institute, Stockholm, Sweden

I. Introduction 11. Critical Factors in Cancer Immunotherapy A. Tumor Immunogenicity and T Cell Response B. Tumor Burden and Location C. Heterogeneity of the Tumor Cell Population 111. Current Immunotherapy Strategies A. Active Specific Immunotherapy B. Active Nonspecific Immunotherapy: Cytokines C. Adoptive Immunotherapy: Lymphokine-Activated Killer (LAK) Cells and Tumor-Infiltrating Lymphoctyes (TIL) D. Chemoimmunotherapy IV. Attempts to Improve Cancer Immunotherapy A. Gains from Experimental Models B. Selection of Patients C. Tumor Debulking D. Active Specific Immunotherapy E. Elimination of Suppressor Cells/Factors F. Active/Adoptive Immunotherapy V. Conclusions References

1. Introduction Over the past three decades, cancer immunotherapy has passed through many cycles of enthusiasm and disappointment. Whereas much progress has been made with animal models and in uitro systems, the clinical experience did not live up to the expectations. During the search for effective therapeutic measures, treatment strategies have changed several times. Initially, active immunization with autologous or allogeneic tumor tissue or extracts, and nonspecific stimulation of the immune system with crude agents (e.g., bacterial toxins, BCG, MER, Clostridium parvum) were attempted. Recently, a new approach-adoptive cellular immunotherapy-evolved as a consequence of developments in in uitro activation and propagation of lymphocytes. This ap245 ADVANCES I N CANCER RESEARCH, VOL. 59

Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.

246

ELI KEDAR AND EVA KLEIN

proach was generated with optimism and received much attention. However, the difficulty of obtaining from the patients sufficient numbers of lymphoid cells with seemingly specific reactivity with their tumor redirected the attention to nonspecific measures (reviewed in Weiss, 1980a; Kedar and Weiss, 1983; Rosenberg et al., 1989a; Borden and Sondel, 1990). The identification and large-scale production of immunostimulatory and growth-inhibitory lymphokines/cytokines, such as interleukin-2 (IL2), interferons (IFN), and tumor necrosis factor (TNF), and the development of monoclonal antibodies directed against cancer cells have led to new avenues (Oldham, 1986; Mitchell, 1988; Talmadge, 1988; Baldwin and Byers, 1989; Balkwill and Burke, 1989; Foon, 1989; Fridman, 1989; Janson et al., 1989; Kelso, 1989; Malkovska et al., 1989; Rosenberg et al., 1989a,b; Borden and Sondel, 1990; Gilewski and Golomb, 1990; DeVita et ‘al., 1991; Oldham, 1991). In particular, the IL2-induced activation and propagation of nonselective cytotoxic lymphocytes [lymphokine-activatedkiller (LAK) cells], with the capacity to damage cancer cells i n vitro, was met with enthusiasm and such cells were administered to several groups of patients (Adler et al., 1984; Rosenberg, 1984, 1986, 1988, 1990, 1991a; Herberman, 1987, 1989; Hersey and Bolhuis, 1987; Rosenberg et al., 1989a,b; Smith, 1988; Yagita and Grimm, 1988; Lotzova, 1989; Sondel and Hank, 1989; Stevenson, 1989; West, 1989; Borden and Sondel, 1990; Lotze and Finn, 1990; Masucci and Mellstedt, 1990; Parkinson, 1990; Rees and Wiltrout, 1990; Robertson and Ritz, 1990; Semenzato, 1990a; Sosman et al., 1990; Dillman et al., 1991b). While the preclinical results with cytokines, with and without LAK cells, were remarkable, the majority of cancer patients did not benefit from such treatments, and the toxic effects (e.g., with TNFa or IL-2) outweighed the slight therapeutic gains (Siege1 and Puri, 1991). The clinical results with LAK cells were rather disappointing and the interest returned to putatively tumor-specific T cells. It was assumed that such cells can be isolated from the tumor tissue [tumor-infiltrating lymphocytes (TIL)],and when amplified in number and activated by I L 2 they can be reinjected (Rosenberg et al., 1986, 1988, 1989b; Rosenberg, 1991a; Itoh et al., 1988; Whiteside et al., 1988; Kradin et al., 1989a; Maleckar et al., 1989; Radrizzani et al., 1989; F’armiani et al., 1990; Shimizu et al., 1990; Dillman et al., 1990a, 1991a,c). Other recent endeavors use combinations of cytokines (e.g., IL-2, IFN, TNF) (Brunda et al., 1987; Winkelhake et al., 1987; Agah et al., 1988; Watanabe et al., 1988; McIntosh et al., 1989; Rosenberg et al., 1989c; Fox et al., 1990), and both new and already well-known non-

CANCER IMMUNOTHERAPY

247

specific immunomodulators (e.g., muramyl ddtripeptide, levamisole, bacterial products, and “synthetic” adjuvants), given alone and in conjunction with chemotherapy or tumor vaccines (Watanabe and Iwa, 1987; Mitchell et al., 1988a, 1990; Foon, 1989; Hoover and Hanna, 1989; Lise and Audibert, 1989; Rosenberg et al., 1989a; Sarosdy and Lamm, 1989; Bauer et al., 1990; Moertel et al., 1990; Urba et al., 1990; Kleinerman, 1991). This short survey of the therapeutic attempts shows the steady search for new strategies and returns with modifications, indicating thus the discontent. However, when the therapeutic effects are evaluated separately on the various types of malignancies, the view is somewhat less gloomy. Immunobiotherapy was indeed beneficial in certain types of malignancies. Good results were obtained in several hematological malignancies with IFNa (response rate >50%) (reviewed in Balkwill, 1989; Foon, 1989; Rosenberg et al., 1989a; Billard and Wietzerbin, 1990). The 25% response rate in metastatic melanoma and renal cell carcinoma patients (with up to 10% of the patients brought to a tumor-free state for at least several months) treated with IL-2, with or without LAK cell administration, can also be regarded as success (Rosenberg et al., 1989b,Rosenberg, 1990).Anecdotal responses to IL2-based treatments have been reported in several other cancers (Oliver, 1988; Rosenberg et al., 1989a,b; Rosenberg, 1990; Foon, 1989; West, 1989; Borden and Sondel, 1990; Lotze and Finn, 1990; Clamon et al., 1991; Dillman et al., 1991a,b). The vast majority of patient groups with the frequently occurring tumors (i.e., colon, lung, and breast carcinoma) did not benefit, however, from the current immunotherapeutic manipulations. Within the melanoma or renal cell carcinoma patient groups, the individual response is highly variable. While some patients responded to various immunotherapeutic modalities, with most or all tumor foci completely regressing, others did not respond at all. The differences may be ascribed to tumor subtypes, to variations in the antigenicity/ immunogenicity of the tumor, and to differences in the tumor-reactive T cell repertoire of the host, influenced by the HLA phenotype. After reviewing the experimental and clinical results, we conclude that the existence of an immune response against the tumor is a prerequisite for the therapeutic effects, and the biological response modifiers (BRMs) act by improving it. We discuss here both immunotherapy in the classical sense and biotherapy with BRMs, as the majority of these agents are immunopotentiators (reviewed in Oldham, 1983, 1986, 1991; Mitchell, 1988; Talmadge, 1988; Foon, 1989; Hadden, 1989; Rosenberg et al., 1989a).

248

ELI KEDAR AND EVA KLEIN

Several reviews and books on various aspects of cancer immunology and immunotherapy have been published recently (Greenberg et al., 1988; Mitchell, 1988; Rosenberg, 1988, 1990, 1991a; Rosenberg et al., 1989a; Talmadge, 1988; Foon, 1989; Fridman, 1989; Herberman, 1989; Schreiber, 1989; Borden and Sondel, 1990; Bystryn, 1990; Lotze and Finn, 1990; Oettgen, 1990, 1991a; Osband and Ross, 1990; Parmiani, 1990; DeVita et al., 1991; Greenberg, 1991; Longo, 1991; Old, 1991; Oldham, 1991; Vanky and Klein, 1991; Wadler, 1991; Melief, 1992). We have restricted our survey to recent pertinent studies on nonhematologic neoplasms, and to the relatively newly proposed strategies. The article certainly bears the marks of experimentalists without experience at the bedside.

II. Critical Factors in Cancer lmmunotherapy A large body of studies in experimental models suggests the importance of the following parameters for the outcome of immunological intervention: (1) immunogenicity of the tumor, to which the expression of the major histocompatibility complex (MHC) antigens contribute, (2) the size and location of the tumor, (3) heterogeneity of the tumor cell population, and (4) the immunocompetence status of the host and, most importantly, its ability to mount cell-mediated responses. The studies have demonstrated that: (1) the therapeutic efficacy of a regimen is determined by the potentiation of an existing tumor-specific cellular immunity; (2) combination treatment modalities are more effective than single ones; and (3) intense treatments can lead to adverse effects (reviewed in Kedar and Weiss, 1983; Greenberg et al., 1988; Mitchell, 1988; Talmadge, 1988; Bergmann, 1989; Fridman, 1989; Herberman, 1989; Rosenberg et al., 1989a; Borden and Sondel, 1990; Osband and Ross, 1990; Parmiani, 1990; Parmiani et al., 1990; Rosenberg, 1990). A. TUMOR IMMUNOGENICITY AND T CELLRESPONSE 1. Experimental Tumors The importance of tumor immunogenicity and a T cell response for the success of immunotherapy is clearly demonstrated in the experimental models (reviewed in Kedar and Weiss, 1983; Rosenberg, 1984, 1986; Robins and Baldwin, 1985; Greenberg et al., 1988; Rosenberg et al., 1989a; Schreiber, 1989; Greenberg, 1991). The evidence is as follows: (1) Adoptive transfer of tumor-reactive helper (Th) or cytotoxic (CTL) T cell populations, or CD4 T h and/or CD8 CTL clones (together with

CANCER IMMUNOTHERAPY

249

IL2), led to rejection of tumor grafts of high or low immunogenicity (Eberlein et al., 1982; Engers et al., 1984; Palladino et al., 1984; Chou et al., 1988; Greenberg et al., 1988; Kast et al., 1989; Klarnet et al., 1989; Ellenhorn et al., 1990). (2) Marked effects were obtained in mice with established tumors of low immunogenicity with either IL-2-based cellular (LAK/TIL) immunotherapy, or with cytokines (IL2, IFNa, TNFol), with and without adjunct chemotherapy; however, the same regimens were not effective against nonimmunogenic tumors (Rosenberg et al., 1986; Mule et al., 1987; Spiess et al., 1987; Cameron et al., 1988; Kedar et al., 1988a, 1989; Papa et al., 1988; McIntosh et al., 1989; Krosnick et al., 1989a). ( 3 ) In some cases, immunotherapy of murine tumors with proven immunogenicity was more successful if initiated relatively late after tumor inoculation, when the antitumor response has already developed, than if the treatments began earlier (Rosenberg et al., 198513; Thompson et al., 1986; Maas et al., 1989; Maekawa et al., 1989; Ciolli et al., 1991). (4) However, a tumor with poor immunogenicity could be controlled only when IL-2 was introduced shortly after tumor inoculation (Slavin et al., 1989; Ackerstein et al., 1991). (5) Mice in which weakly immunogenic tumor grafts regressed after treatment with I L 2 with or without LAK cells or chemotherapy, acquired long-lasting specific immunity (Mule et al., 1986; Formelli et al., 1988; Hornung et al., 1988; Maas et al., 1989; Kedar et al., 1990, 1992). (6) I n vivo depletion of either CD4 or CD8 T cells [but not of natural killer (NK) cells] prior to cytokine (IL-1, IL-2, or IL-4) administration reduced the therapeutic effects in tumor-bearing mice (Mule et al., 1987; Peace and Cheever, 1989; Bosco et al., 1990; Ciolli et al., 1991; Kedar et al., 1992). (7) Splenocytes of mice taken 3-6 months after cyclophosphamide and I L 2 treatment, which led to complete regression of weakly immunogenic tumors, conferred specific immunity to naive recipients, with most of the protective activity residing in the CD8 T cells. The frequency of antitumor CTL precursors in such splenocyte populations was 5 to 20 times higher compared to that of control mice (Kedar et al., 1990, 1992). (8)The therapeutic effects of I L 2 , I L 4 , I L 6 , or IFN in tumor-bearing mice were markedly reduced when applied after immunosuppression with corticosteroids, cyclosporin A, or radiation (Papa et al., 1986; Rosenberg, 1988; Cameron et al., 1988; Bosco et al., 1990; Mule et al., 1990). (9) Combination treatments consisting of chemotherapy, with I L 2 , with or without IFNa, induced tumor regression in euthymic mice, but were relatively ineffective in nude mice carrying human tumor xenografts. Addition of LAK cells to these regimens increased the therapeutic efficacy in nude but not in conventional mice (Kedar et al., 1988a, 1990, 1992; Gazit et al., 1992), suggesting that the LAK cells are efficient only when T cell substitution

250

ELI KEDAR A N D EVA KLEIN

is needed. Combination treatment with IFNa and LFNy was successful in euthymic but not in nude mice carrying murine tumor grafts (Sayers et al., 1990). The general experience is that only rarely can immunotherapy cure nude mice when the tumor grafts are already well established. 2. Patients A key question is why mainly patients with melanoma and renal cell carcinoma respond to IL2, with or without LAK cell therapy, and why only about one-fourth of these patients respond (Oliver, 1988; Margolin et al., 1989; West, 1989; Rosenberg, 1990, 1991a; Rosenberg et al., 1989b; Borden and Sondel, 1990; Sosman et al., 1990). Similar response rates (i.e., 20-40%, albeit generally less durable) were obtained with other immunologic manipulations, such as active specific immunotherapy in melanoma (Berd et al., 1990a; Mitchell et al., 1990; Morton et al., 1990, 1991a),or treatment with IFNa, with or without chemotherapy, in melanoma and renal cell carcinoma (McLeod et al., 1987; Bergmann, 1989; Guillou et al., 1989; Kellokumpu-Lehtinen and Nordman, 1990; Mickiewicz et al., 1990; Mulder at al., 1990; Falkson et al., 1991). It is likely that the patients for which the therapy is successful have (1) immunogenic tumors, (2) a certain level of existing antitumor response, and (3)a relatively higher capacity to be influenced by immunostimulatory measures. The immunogenicity of melanomas and the role of cellular immunity in the control of these tumors is supported by the following findings: (1) Spontaneous tumor regression (in 1-2% of the patients), and recurrences occurring late (>5 years) after surgical removal of the primary tumor (Cochran et al., 1988; Rosenberg et al., 1989a); (2) the existence of a large proportion of patients with antibodies (Old, 1981) and cytotoxic cells (reviewed in Anichini, 1989; Rosenberg et al., 1989a; Parmiani et al., 1990; Boon, 1992) reactive with their tumors in vztro; (3) tumor regression sometimes occurring weeks or months after IL-2 treatment is discontinued (Rosenberg, 1988); (4) T cells with specific cytotoxicity detectable within the regressing tumor tissue (Cohen et al., 1987; Parmiani et al., 1990); (5) generation (from about 25% of the patients) of MHCrestricted cytotoxic T cell lines or clones derived from tumor-infiltrating lymphocytes (TIL), blood, or lymph nodes, which can damage ex vzvo autologous tumor cells (Itoh et al., 1988; Slingluff et al., 1988; Maleckar et al., 1989; Topalian et al., 1989; Mukherji et al., 1990; Parmiani et al., 1990); ( 6 )preferential localization of readministered autologous TIL in tumor sites (Griffith et al., 1989); (7) the more frequent lesional (subcutaneous metastases) response to IL-2-based immunotherapy in pa-

CANCER IMMUNOTHERAPY

25 1

tients whose tumor cells expressed a high level of HLA-DR antigens during treatment, and infiltration of T cells and macrophages (but not NK cells) in the regressing tumor tissue (Rubin et al., 1989; Clark et al., 1990); (8) correlation between the clinical response and the frequency of melanoma-specific CTL precursors in the blood lymphocyte population following immunization with a tumor vaccine (Mitchell et al., 1990; Mitchell, 1991). Some of these parameters indicate that about 25% of the patients with melanoma (and probably renal cell carcinoma as well) carry immunogenic tumors. This corresponds to the approximately 25% response rate to immunotherapy in these patient groups. There is no evidence for a role of the nonselective cellular effectors, such as NK/LAK cells. In the majority of the clinical trials with melanoma patients treated with IL-2 o r other immunotherapeutic modalities, no correlation was found between the clinical response and the levels of cytotoxicity in NK and LAK cell assays with blood lymphocytes (Boldt et al., 1988; Eberlein et al., 1989; Ghosh et al., 1989; Favrot et al., 1990; Dillman et al., 1991b; Isacson et al., 1992). In uitro assays indicated the immunological recognition of tumor cells in a proportion of patients with other malignancies as well. Blood lymphocytes were found to be cytotoxic and/or were stimulated by autologous ex uiuo tumor cells (Vanky et al., 1976, 1982, 1983a,b, 1987; Vose et al., 1977; Vose and Bonnard, 1982; Vanky and Klein, 1982a,b, 1989; Uchida et al., 1987, 1990; Allavena et al., 1988; Uchida and Mizutani, 1989; Parmiani et al., 1990; Ioannides et al., 1991). The analysis of the cytotoxic response indicated that it represents a T cell-mediated immunity (Vanky and Klein, 1991) and it correlated with a relatively favorable prognosis in a group of patients with adeno- and squamous cell carcinoma of the lung and with mesenchymal cancers (Vanky et al., 1986, 1987; Uchida and Mizutani, 1989; Uchida et al., 1990) (see Section IV,B). 3 . MHC Antigen Expression on Tumor Cells

Major histocompatibility complex molecules on the cell surface carry and present the antigenic peptides that are recognized by cellular immunity. Appropriate expression on the tumor cells is therefore decisive for an efficient antitumor response (Tanaka et al., 1988; Van Duinen et al., 1988; Gopas et al., 1989; Elliott et al., 1989; Vanky 1986; Vanky et al., 1986, 1988, 1990; Greenberg, 1991; Vanky and Klein, 1991). Low MHC antigen expression on the tumor cells can thus be the cause of low o r no immunogenicity. In many, but not in all, experimental models deficiency in certain class I and/or class I1 antigens was associated with increased

252

ELI KEDAR AND EVA KLEIN

take of tumor grafts and metastasizing tendency. However, for the majority of human tumor types there seems to be no solid evidence for a correlation between altered MHC antigen expression, quantitative or qualitative, and increased malignancy, defined by metastasis or survival (reviewed in Tanaka et al., 1988; Anichini, 1989; Elliott et al., 1989; Gopas et al., 1989; Schreiber, 1989; Wintzer et al., 1990; Moller et al., 1991; Ruiz-Cabello et al., 1991). In melanoma and renal cell carcinoma patients, an association between the expression of high levels of MHC class I antigens on the tumor cells and favorable prognosis was seen in some studies (Van Duinen et al., 1988; Tomita et al., 1990; reviewed in Ruiter et al., 1991). In contrast, high expression of MHC class I1 antigens was associated with unfavorable prognosis in melanoma patients (reviewed in Ruiter et al., 1991).The presence of HLA-DR and D Q antigens on primary breast carcinoma was correlated with several distinct parameters of good prognosis (Brunner et al., 1991). Selective down regulation of some alleles on the malignant cells, including often the HLA-A2 allele (present in approximately 40% of the Caucasian population), has been detected by immunofluorescence and by biochemical methods in a proportion of patients with several types of cancers (Momburg et al., 1989; Natali et al., 1989; Smith et al., 1989; Ruiz-Cabello et al., 1991; Wang et al., 1991). In a group of melanoma patients, the HLA phenotype of the patients seemed to influence the response to active immunotherapy with a vaccine prepared from cultured allogeneic melanoma cells. A higher response rate was obtained in patients with the phenotypes HLA-A2 and -28, B12s (B12, B44, and B45), and C3, particularly in combination, or with HLA-DR4 in combination with A2 or C3 (Mitchell, 1990, 1991). In patients with melanoma and renal cell carcinoma treated with I L 2 and LAK cells or I L 2 and IFNa, the frequency of HLA-AS, B44, or DR4 was higher in responders than in nonresponders (Scheibenbogen et al., 1991). In patients with colon cancer, immunization with autologous tumor cells led to better clinical response if the tumor cells expressed MHC class I1 antigens (Ransom et al., 1991). T h e effect of the changes in MHC antigen expression on melanoma cells during immunotherapy was also investigated. The lesional response (subcutaneous metastases) to IL-2-based immunotherapy correlated with DR antigen expression on the tumor cells after but not before treatment (Rubin et al., 1989). In some (Atkins et al., 1988), but not in other studies (Schwartzentruber et al., 1990),a high incidence of autoimmune thyroid dysfunction was associated with the clinical response to IL2, with and without LAK cell administration, in patients with various types of neoplasms. A similar correlation between the development of

CANCER IMMUNOTHERAPY

253

hypothyroidism and the clinical response was observed in melanoma and renal cell carcinoma patients treated with IL-2 and IFNa (Scalzo et al., 1990). This effect may be ascribed to increased MHC antigen expression on the tumor and thyroid cells, induced by the cytokines (e.g., TNFa, IFNy) produced after IL-2 administration (Siege1 and Puri, 1991). Increased MHC class I antigen expression on a murine tumor conferred increased responsiveness to IL-2 therapy (Weber et al., 1987). Several studies reported that human tumors with high levels of MHC class I antigens had more intense T lymphocyte infiltration (Kornstein et al., 1983; Van Duinen et al., 1988; Vanky et al., 1988). T h e significance of MHC class I antigen expression on the tumor cells for their sensitivity to immune effectors was shown directly in uitro. Blood lymphocytes of patients with solid tumors damaged the autologous ex vzuo tumor cells only if the latter expressed appreciable levels of MHC class I antigens and the intercellular adhesion molecule- 1 (ICAM- 1) molecules (Vanky et al., 1988, 1990). Tumor cells acquired sensitivity concurrent with induction or elevation of these molecules on the tumor cells by exposure to TNFa and IFNy. In several experiments, cytotoxic lymphocytes generated in mixed cultures with the cytokine-treated tumor cells also damaged the unmodified, low MHC class I expressor tumor cell aliquots (Vanky et al., 1988, 1989, 1990). T h e role of adhesion molecules in tumor-T cell interaction was also demonstrated by Anichini et al. (1990). Among tumor cell clones isolated from a human metastatic melanoma, susceptibility to lysis by autologous CTL clones was associated with high expression of IGAM- 1 and the very late activation antigens (VLA)-1, -2, -3, -4, and -6.

4. Conclusions It is very likely that the MHC antigenic make-up and the expression of adhesion molecules on the tumor cells are important for the response to immunotherapy (Braakman et al., 1990; Parmiani et al., 1990; Vanky and Klein, 1991; Melief, 1992). While T cells appear to be pivotal in the development and execution of antitumor response, other cell types may participate in tumor destruction. Regressing tumors are often infiltrated by macrophages, neutrophils, and eosinophils (Pretlow et al., 1983; Fidler, 1985; Musiani et al., 1989; Bosco et al., 1990), cells that do not possess antigen-specific properties. They represent the second line of defense, presumably mobilized by the lymphokines released by the antigen-specific T cells. Thus, BRM-stimulated nondiscriminative effector cells, such as macrophages (Fidler, 1985, 1988a,b; Schreiber, 1989; Mantovani, 1990; Whitworth et al., 1990; Greenberg, 1991) and NK/LAK cells (Herberman, 1987,

254

ELI KEDAR AND EVA KLEIN

1989; Rosenberg, 1988, 1990; Rosenberg et al., 1989a),may contribute to the therapeutic effect. The preclinical studies (and to some extent the clinical studies as well) suggest that antigenicity of the tumor and the capacity to mobilize a T cell response are prerequisites for the success of immunotherapy. In extensive studies, the patients (except those with advanced disease) were not found to be immunodeficient. Therefore, the antigenicity of the tumor seems to be the main property that determines whether therapeutic effects can be achieved by measures that act on the immune system. B. TUMOR BURDEN AND LOCATION Experiments in animals demonstrated that the capacity of the immune system to cope with a growing tumor is limited-up to a certain tumor size-and there is a critical time after tumor implantation when immunotherapy should be initiated. Therefore, it is advisable to reduce the tumor load-by surgery, chemotherapy, or radiotherapy-prior to administration of immunotherapy (Fefer, 1974; Fefer et al., 1976, 1982; Kedar and Weiss, 1983; Mitchell, 1988). Usually it is not possible to eliminate all the tumor cells; it may not even be an advantage because the residual cells may serve as antigen source (Hamblin, 1989). Evidence for this is seen in leukemia patients treated with high-dose chemo/ radiotherapy followed by allogeneic bone marrow transplantation, where the residual leukemic cells evoke an immune response and are destroyed subsequently by the graft versus leukemia (GVL) response (Slavin and Kedar, 1988; Butturini and Gale, 1989; Gottlieb et al., 1989; Sullivan et al., 1989; Horowitz et al., 1990; Porwit et al., 1990).There are examples in animal models in which a larger tumor responded better to the immunological manipulation than the smaller one. In mice carrying transplanted immunogenic or weakly immunogenic tumors, responses were better when I L 2 was administered 5-10 days after rather than earlier following tumor inoculation (Rosenberg et al., 198513; Thompson et al., 1986; Maas et al., 1989). Most likely, the tumor had to reach a certain size for provision of critical amounts of antigens, and time was needed to build up the antitumor immunity, which could then be amplified by the lymphokine. In mice with metastases of weakly immunogenic tumors, TNFa was more effective against 5- to 6-mm tumor foci than against smaller ones (Mu16 et al., 1988),although this finding can also be explained by the fact that TNFa action depends, in part, on newly formed capillaries in the tumor (Manda et al., 1990).On the other hand, a complete response to immunomodulators in ovarian carcinoma patients with tumors confined to the peritoneal cavity occurred only when the tumors were smaller than 0.5 cm (Berek, 1990).

CANCER IMMUNOTHERAPY

255

An important parameter is the location of the tumor and the extent of its dissemination. As with chemotherapy, patients with “widespread disease’’ and those with metastases in certain sites (e.g., in the brain, liver, o r bones), do not respond as well to immunotherapy as patients with a “limited disease” o r patients with skin, lymph node, lung, or soft-tissue metastases (Lotze et al., 1986; West et al., 1987; Mitchell et al., 1988b; Bergmann, 1989). Locoregional immunotherapy can be applied and showed some effects in patients with localized disease (Markman, 1987; Bubenik, 1989), such as head and neck (Forni et al., 1988), brain (Yagita and Grimm, 1988; Yoshida et al., 1988), bladder (Pizza et al., 1984), and abdominal tumors (Ottow et al., 1987; Urba et al., 1989). C. HETEROGENEITY OF THE TUMOR CELL POPULATION Tumor cell populations are heterogeneous, comprising cells with variable growth, metastatic and immunogenic properties, and sensitivities to chemoradiotherapy and to immunological effector mechanisms (Heppner, 1984; Nicolson, 1984; Poste, 1986; Schnipper, 1986; Clark et al., 1988; Fidler, 1988a; Woodruff, 1988; Parmiani et al., 1990). Tumor samples collected from different sites of one patient, and parallel tumor cell lines derived from one tumor, often react differently with monoclonal antibodies, cytotoxic T cell clones, LAK and TIL populations (Anichini et al., 1989; Rivoltini et al., 1989; Van den Eynde et al., 1989; Notter and Schirrmacher, 1990; Parmiani et al., 1990; Topalian el al., 1990). Variants that can escape immune attacks may be present in the tumor cell populations. It is therefore likely that immunotherapy using one monoclonal antibody, one clonal effector cell population, or one cytokine, can affect only part of the cell population, and the variant cells can take over. Therefore, “cocktails” of monoclonal antibodies and/or several T cell clones directed against different antigenic epitopes on the tumor may be required for the control of the tumor. Ill. Current lmmunotherapy Strategies

A. ACTIVESPECIFIC IMMUNOTHERAPY Immunization with tumor cells or tumor extracts preventing the growth of subsequently grafted cells provided the proof for immunogenicity of some animal tumors. Active specific immunotherapy in patients was attempted by several investigators (reviewed in Foon, 1989; Rosenberg et al., 1989a; Bystryn, 1990; Livingston, 1991a), using various vaccines (with and without adju-

256

ELI KEDAR A N D EVA KLEIN

vants), such as irradiated autologous tumor cells (Schulof et al., 1988; Hoover and Hanna, 1989; Wiseman et al., 1989; Berd et al., 1990a; McCune et al., 1990); allogeneic (single or pooled) ex uzuo or cultured tumor cells, or extracts, of the corresponding histologic type (Mitchell et al., 19$8a, 1990; Mitchell, 1991; Bystryn, 1990; Morton et al., 1991a); tumor oncolysates (tumor cells infected with lytic viruses) (Cassel et al., 1983); or purified tumor antigens (Hollinshead et al., 1989; Knuth et al., 1991; Livingston, 1990, 1991b; Srivastava, 1991). Encouraging results have been reported in melanoma and carcinoma (colon and lung) patients with progressive disease (Berd and Mastrangelo, 1988a; Berd et al., 1990a; Mitchell et al., 1988a, 1990; Mitchell, 1991; Morton et al., 1990, 1991a) or as adjuvant treatment after surgery (Cassel et al., 1983, 1986; Hoover and Hanna, 1989; Hollinshead et al., 1989, 1990; Bystryn et al., 1988, 1990; Morton et al., 1991a). In some of these trials, the patients were pretreated with low-dose cyclophosphamide (to deplete suppressor cells), and the tumor vaccines were given together with BCG or other adjuvants. In some cases, indomethacin or cimetidine were also used in order to act against suppressor cells (see Section 111,D). The therapeutic benefit of these auxiliary measures has not been proven, however, in randomized trials. T h e decision whether to use autologous or allogeneic tumor cells for vaccination has not yet been reached. It can be expected that only the autologous tumor carries the relevant individually distinct epitopes (Old, 1981; Notter and Schirrmacher, 1990; Parmiani et al., 1990). However, due to heterogeneity of the tumor cell population, a proportion of cells may lack the antigens present in the autologous vaccine (Bystryn, 1990). Several arguments can be mentioned for the use of pooled allogeneic tumors o r tumor cell lines: (1) Shared tumor-associated antigens may be expressed in various quantities on the tumors (Old, 1981; Crowley et al., 1990; Mitchell, 1991). (2) Some of the contributing tumors may share HLA specificities with the patient and thus the vaccine may provide the proper antigenic peptide against which the patient can respond. This may have no importance when extracts are used; the antigens then are expected to be processed and presented by the patient’s cells. (3) The alloantigens may exert “help” that can potentiate the response against the putative tumor antigens (Mitchell, 1991). (4)A pool can be produced in large quantities, thus providing a standard immunogen. It is difficult to say which of these considerations are valid, therefore a combined preparation consisting of both autologous and allogeneic source could be preferred, when possible. T h e majority of tumor types in humans are nonimmunogenic (reviewed in Sulitzeanu, 1985; Rosenberg et al., 1989a). Various manipula-

CANCER IMMUNOTHERAPY

257

tions, however, can enhance the expression of weak antigens or impose new antigens on the cell. Such cells can then be used for elicitation of an immune response, which may act also on the resident nonaltered cells. Human and murine tumor cells have been modified in vitro by (1) chemical/enzymatic treatment (Kedar and Lupu, 1978; Kedar et al., 1979; Wunderlich et al., 1985; Berd et al., 1990b; Wiseman et al., 1989, 1990; Ramakrishna and Shinitzky, 199l), (2) xenogenization with viruses (Kobayashi, 1986; Schirrmacher, 1989; Schirrmacher et al., 1986, 1989; Bash et al., 1990; Bohle et al., 1990; Freedman et al., 1990; Lehner et al., 1990), ( 3 ) mutagenization (Boon, 1983, 1992; Boon et al., 1989), (4) transfection of MHC class I or class I1 genes (Gopas et al., 1989; Isobe et al., 1989; Ostrand-Rosenberg et al., 1990), and (5) cytokine treatment (IFNy, TNFa, IL-4), which u p regulates MHC antigen, adhesion molecule, and perhaps also tumor-associated antigen expression (Vanky et al., 1990; Wiebke et al., 1990; Hoon et al., 1991). Chemically/enzymatically altered (Skornick et al., 1986; Berd et al., 1990b; Wiseman et al., 1989, 1990) and virus-infected (Cassel et al., 1983, 1986; Schirrmacher, 1989, 1991; Freedman et al., 1990; Bohle et al., 1990) autologous tumor cells have recently been used for immunization in patients. Results cannot be evaluated yet because a relatively small number of patients have been treated, and the observation period has been short. T h e image of tumor antigens in the form of antiidiotypic antibodies has been employed as immunogen both in animals and patients. T h e preliminary clinical results are encouraging (Herlyn et al., 1987; Campbell et al., 1988, 1990; Bhattacharya-Chatterjee and Kohler, 1989; Chatterjee et al., 1990; Levy and Miller, 1990; Mittelman et al., 1990; Frodin et al., 1991; reviewed in Lee and Hellstrom, 1988; Sikorska, 1988; Foon, 1989; Rosenberg et al., 1989a; Mellstedt, 1990; Stevenson et al., 1990). These various manipulations were often successful in eliciting a T cell response in uitro or in uivo against the unmodified tumor cells also and led to regression of experimental tumors. In some of the clinical trials, the patients that responded to active specific immunotherapy with antibodies, CTL, or delayed-type hypersensitivity (DTH) reactive with the immunizing antigen, showed a better clinical response (in patients with advanced disease), and a longer disease-free survival and fewer recurrences (in the adjuvant setting) than those that responded poorly o r not at all in these immunological tests (Berd et al., 1990a; Bystryn, 1990; Frodin et al., 1991; Mitchell, 1991; A. Mittelman and V. Schirrmacher, personal communications). Obviously, the first step of a well-planned active specific irnmunotherapy requires the identification of the relevant antigens. Recently, the group of T. Boon exploited tumor cell-reactive CTL clones and

258

ELI KEDAR AND EVA KLEIN

defined (and sequenced) a family of molecules (“MAG,”) expressed on a human melanoma cell line. While the antigens were not detected on normal cells, they were expressed on other (HLA-A1 positive) melanoma lines and also on other tumor types (Van der Bruggen et al., 1991). These, and other similarly defined antigens, can be used as a raccine source in patient groups with the appropriate class I MHC molecules. B. ACTIVENONSPECIFIC IMMUNOTHERAPY: CYTOKINES

I. Introduction Over 20 cytokines (among them 12 interleukins) are known and functionally defined. The majority of these are now produced by recombinant DNA technology and are available for therapy. The cytokines are part of a complex system; each of them can be produced by more than one cell type and one cell can produce several cytokines, in a tightly regulated way (Lebendiker et al., 1987; Sariban et al., 1988; Ketzinel et al., 1990). They have pleiotropic biological effects, and different cytokines can share certain functions. Usually these soluble factors act at a short range, in low concentrations, in both autocrine and paracrine fashions. They can influence the immunological host-tumor relationship; some of them can also affect tumor growth and differentiation and can regulate inflammatory responses (reviewed in Balkwill, 1988; Balkwill and Burke, 1989; Foon, 1989; Kelso, 1989; Mizel, 1989; Paul, 1989; Rosenberg et al., 1989a; Borden and Sondel, 1990; DeVita et al., 1991; Oettgen, 1991b; Oldham 1991; Street and Mosmann, 1991; Zwierzina, 1991).

The response to cytokines may be influenced by the genetic background of the host. Furthermore, various cell types differ in susceptibility, mainly due to differences in the expression of the relevant receptors, in the signal transduction mechanism, and in the activity of specific transcription factors (Korber et al., 1988; Landolfo et al., 1988; Steiniger et al., 1988; Gribaudo et al., 1990). Cytokines can be useful in cancer treatment by (1) exerting direct effects on the tumor (cytolysis, cytostasis, vasculature damage, terminal differentiation), (2) enhancing the expression of MHC antigens, cell adhesion molecules, and other surface moieties on the tumor cells including tumor-associated antigens, (3) recruiting, expanding, and stimulating endogenous effector cells, and (4) maintaining and even enlarging adoptively transferred lymphocyte populations.

CANCER IMMUNOTHERAPY

259

Because of the highly complex network of cytokine and cellular interactions, the mechanisms of antitumor actions obtained in several experiments are not clear. Although thousands of cancer patients have been treated with IFNs and therapeutic effects were achieved, it is unknown whether the antiproliferative, the immunomodulatory, or the antiviral effects, alone or in combination, were important. Likewise, it is unclear what the mechanisms are of the remissions seen in IL2-treated patients (Quirt and Tannock, 1990; Sasaki et al., 1991; Siege1 and Puri, 1991). Experiments in animal models point to the possibility that cytokine-mediated antitumor effects may also be caused, in part, by a decreased food intake and alterations in host substrate metabolism (Gelin et al., 199lb).

2. Antitumor Effects by Application of One Cytokine Interleukin 2, TNFa, IFNa, and IFNy have been found effective against a wide spectrum of experimental tumors, whereas in patients mainly IL-2 and IFNa were beneficial and only in certain malignancies (reviewed in Foon, 1989; Rosenberg et al., 1989a,b; Borden and Sondel, 1990; DeVita et al., 1991; Oldham, 1991). Interleukin 2-based immunotherapy is discussed later (Section 111,C). The growth inhibitory and cytotoxic effects of IFN and TNF both in vitro and in vivo for various types of experimental and human tumor cells is well established. Tumor necrosis factor can also induce hemorrhagic necrosis in the tumor through the blockade of the blood supply, and it facilitates the influx of inflammatory cells to the tumor tissue (North and Havell, 1988; Shimomura et al., 1988; Mule et al., 1988; Balkwill et al., 1990; McIntosh et al., 1990; Semenzato, 1990b). Both IFN and TNF up regulate the expression of MHC class I and class I1 antigens, the cell adhesion molecules, and perhaps even putative tumorassociated antigens, thereby increasing immunogenicity and susceptibility of the tumor cells to T cell-mediated damage (Borden, 1988; Weber and Rosenberg, 1988; Guadagni et al., 1989; Maio et al., 1989; Stotter et al., 1989; Vanky et al., 1989, 1990; Wiebke et al., 1990). Moreover, they can stimulate T cells, macrophages, and N K cells (Faltynek and Oppenheim, 1988; Havell et al., 1988; Jaffe and Herberman, 1988; Paul and Ruddle, 1988; Talmadge et al., 1988; Asher et al., 1989; Kunkel et al., 1989; Rosenblum and Donato, 1989). Interleukin 2 probably lacks direct antitumor effects, but it is an efficient stimulator of several immune functions (Mertelsmann and Welte, 1986; Smith, 1988; Rosenberg et al., 1989a; Swain, 1991). The effect of cytokines on tumor growth may differ in vivo and in vitro. For example, TNFa (Tomazic et al., 1988; Balkwill et al., 1990;

260

ELI KEDAR AND EVA KLEIN

Semenzato, 1990b) and IFN (Morikawa et al., 1989; reviewed in Wadler and Schwartz, 1990) inhibited the growth of some tumors either in vitro or in Y ~ Z I Obut , not under both conditions. IFNy stimulated the growth of a human T cell lymphoma in vitro but inhibited its growth when grafted in nude mice (Jemma et al., 1989). Adverse effects were also reported. Treatment with IFNy seemed to increase the relapse rate of patients with stage I1 cutaneous melanoma (Meyskens et al., 1990). Interleukin 1 (Giavazzi et al., 1990; Bani et al., 1991), IL-2 (Kedar et al., 1989), T N F a (Balkwill et al., 1990; Malik et al., 1990), and IFNy (Kelly et al., 1991) enhanced the local growth and experimental metastases of both murine and human tumor grafts in euthymic and athymic mice. 3 . Cytokine Combinations: Additive and Synerptic Effects

Cytokines initiate a cascade; they trigger the production of other cytokines and the expression of their receptors. For example, patients treated with high IL-2 doses have high IL-1, I L 6 , TNFa, and IFNy plasma levels (Gemlo et al., 1988; Kasid et al., 1989; Boccoli et al., 1990; Dupere et al., 1990). Treatment with combinations of cytokines differing in their mode of action, using each at subtoxic doses, may therefore improve the therapeutic index. Cytokines have been used for tumor therapy in various combinations, alone and together with chemotherapy, with and without TIL or LAK cells. Additive o r synergistic effects have been obtained with cytokine combinations in experimental systems: IL- 1 plus IL-2 (Belardelli et al., 1989; Crump et al., 1989; Ciolli et al., 1991), IL-2 plus TNFa (Winkelhake et al., 1987; Yang et al., 1989), I L 2 plus IFNa (Brunda et al., 1987; Cameron et al., 1988), IL-2 plus IFNy (Agah et al., 1988), T N F a plus IFNy (Watanabe et al., 19SS), and IL-2 plus IL-5 (Aoki et al., 1989) synergized in the activation of NK/LAK cells in vitro and/or in the induction of antitumor effects in vivo. Additive or synergistic therapeutic effects for animal tumors have also been reported for IL-1 plus IL4 (Forni et al., 1989), TNFa plus IL-6 (Mule et al., 1990),and TNFa plus IL-2 plus IFN (Agah et al., 1988; McIntosh et al., 1989). Administration of 1L-2 and IFNa to mice enhanced the proliferation of lymphoid cells (Puri et al., 1990). In our studies (Kedar et al., 1990, 1992), the cojoint treatment with IL-2 and IFNa had additive therapeutic effects in mice with pulmonary metastases or intraperitoneal growths of weakly immunogenic tumors. This cytokine combination synergized with cyclophosphamide. I L 2 plus TNFa and IL-2 plus macrophage-colony stimulating factor (M-CSF) were considerably less efficient, alone and with chemotherapy.

CANCER IMMUNOTHERAPY

26 1

Additive therapeutic effects were also obtained in nude mice carrying human colorectal carcinoma and melanoma grafts with IL-2 plus IFNa, particularly in combination with chemotherapy [5-Fluorouracil (5-FU) or dacarbazine (DTIC), respectively] (Kedar et al., 1990, 1992; Gazit et al., 1992). Impressive clinical results (an approximately 40% response) were reported in patients with metastatic melanoma and renal cell carcinoma treated with high-dose IL-2 plus IFNa (Rosenberg et al., 1989b,c). T h e effects were apparently additive, and the response correlated with the cytokine doses. This was confirmed in another study, in which 50% of the patients with renal cell carcinoma responded (Hirsh et al., 1990). In other studies, however, a lower (525%) response rate was obtained with this cytokine combination (Legha et al., 1990; Stahel et al., 1990; Bukowski et aE., 1990; Sznol et al., 1990; Atkins et al., 1991; Ilson et al., 1991; Pichert et al., 1991). In patients with metastatic renal cell carcinoma, low doses of IL-2 and IFNa, administered subcutaneously, was less toxic than, and as effective as, the high-dose intravenous regimen of IL-2 (Atzpodien et al., 1990a,b). The combination of IFNa and TNFa gave a 43% response in patients with renal cell carcinoma (Otto et al., 1990), but was ronsiderally less effective in other types of solid tumors (Fuchimoto et al., 1990). Other combinations of two cytokines, employing IL-2, IFNy, and TNFa, were also of low potency in patients with various solid tumors (Rosenberg et al., 1989b; Redman et al., 1990; Sohn et al., 1990; Dexeus et al., 1991; Dillman et al., 1991a; Smith et al., 1991b; Weiner et al., 1991; Yang et al., 1991). T h e experience in experimental models suggests that sequential administration of cytokines may be more effective than concurrent treatments. The antitumor effects in tumor-bearing mice were usually stronger when TNFa or IFNa was given 1-3 days prior to IL-2, compared to either concurrent administration or sequential treatment but in the reverse order (McIntosh et al., 1988; Zimmerman et al., 1989b; Kedar et al., 1992). The optimal sequence may be different in protocols combining these cytokines and chemotherapy. In our studies using both euthymic and athymic mice carrying mouse or human tumor grafts, respectively, the most effective combination regimen, yielding a synergistic effect, was IFNa (6 days) + chemotherapy + IL-2 (6 days), with a l-day interval between IFNa and chemotherapy and 2-3 days between chemotherapy and IL-2. The second best regimen was chemotherapy -+ I L 2 --+ IFN. T h e following schedules were less efficient: IL-2 + chemotherapy + IFN, chemotherapy + IFN + I L 2 , and chemotherapy --+ IL-2 plus IFN (Kedar et al., 1992; Cazit et al., 1992).

262

ELI KEDAR A N D EVA KLEIN

The animal models suggest that the synergistic effects depend largely on the presence of tumor-specific T cells that are activated by the cytokines. The synergistic effects of I L 2 plus TNFa or I L 2 plus IFNa were obtained only with immunogenic tumors; moreover, if X irradiation or i n vivo depletion of Lyt2 T cells preceeded immunotherapy, the therapeutic effects did not occur (Cameron et al., 1988; McIntosh et al., 1988). Similarly, I L l plus I L 2 treatment was inefficient in nude mice and in euthymic mice depleted of Thy 1.2+ cells (Belardelli et al., 1989). +

4 . Cytokine Combinations: Antagonistic Effects Combinations of certain cytokines can lead to adverse effects because one cytokine may counteract the function of another. Evidence for such effects stems from i n vitro experiments. Interleukin 4 (Spits et al., 1988; Ebina et al., 1990) and transforming growth factor P (TGFP) (Grimm et al., 1988) were shown to inhibit the induction of human LAK cells by IL-2; I L l O inhibited lymphokine production by activated T cells (K. W. Moore et al., 1990; Fiorentino et al., 1991); IFNy suppressed IL4-induced IgE production (ChrCtien et al., 1990); and TNFa abolished the immunosuppressive activity of TGFP (Ranges et al., 1987).

5 . Toxicity The currently used cytokines can cause moderate or severe toxicity. This can be attributed to the extremely large doses administered, which are several orders of magnitude higher than the physiologic levels. Most serious is the vascular leak syndrome, induced mainly by I L 2 and TNFa (S. A. Rosenberg et al., 1987a, 1989a,b; Margolin et al., 1989; Semenzato, 1990b; Dillman et al., 1991b; Siege1 and Puri, 1991). High doses of I L 2 or IFN also caused transient immunosuppression in both animals and patients (Talmadge et al., 1987; Kedar et al., 1988b; Wiebke et al., 1988; Kradin et al., 1989b; Hank et al., 1990b). Several attempts were made to search for less toxic cytokines. Impressive antitumor effects, similar to those seen with high-dose IL-2, but with negligible toxicity, were obtained in mice treated with I L 6 (Mule et al., 1990). A nontoxic derivative of I L 1 maintained the immunomodulatory activity (Forni et al., 1989). Chemically altered recombinant human TNFa had a broader cytotoxic spectrum for human and mouse tumor cells in vitro and stronger antitumor effects in mice, with fewer side effects, than the unmodified cytokine (Noguchi et al., 1991). It is also possible to administer agents that counteract toxicity. Indomethacin (inhibits prostaglandin synthesis) and bismuth subnitrite (blocks oxygen-free radical production) have been shown to reduce TNFa toxicity in rodents (Haranaka, 1988; Talmadge, 1988; Krosnick et

CANCER IMMUNOTHERAPY

263

al., 1989b). Corticosteroids decreased I L 2 toxicity in patients (Vetto et al., 1987; Mier et al., 1990). Their effect on the tumor response has not been determined; in mice, however, the therapeutic efficacy was also reduced (Papa et al., 1986). Certain cytokines can counteract the toxicity of other cytokines. Interleukin 1 administered together with I L 2 reduced the vascular leakage, without impairing the antitumor activity in mice (Puri et al., 1989). In culture, pretreatment of human endothelial cells with I L 1, TNFa, or IFNy increased their resistance to LAK cell-mediated lysis (Mier et al., 1989). 6. Dose Response

The majority of the cytokines show a direct antitumor dose-response relationship in vivo. In studies with mice carrying weakly immunogenic tumors and also in patients, the beneficial response to systemic I L 2 treatment was usually dose dependent and required high doses (Rosenberg, 1986; Bradley et al., 1987; West, 1989). This was also the case with TNFa in experimental models (McIntosh et al., 1988). For systemic treatment of mice carrying immunogenic tumors (Talmadge et al., 1987; Talmadge, 1988)and for local treatment of murine and human tumors (Vaage, 1987; Forni et al., 1988; Musiani et al., 1989),even very low I L 2 doses (25- 1000 U/day) were sufficient to cause partial or complete tumor regression. Interferon? was shown to have a bell-shaped therapeutic and immunomodulating dose-response curve, both in the experimental and clinical studies (Jaffe and Herberman, 1988; Maluish et al., 1988; Talmadge, 1988; Frick and Aulitzky, 1990; Wadler and Schwartz, 1990). Its optimal therapeutic and immunomodulatory doses were usually lower than the maximum tolerated dose.

7. Mode of Cytokine Administration Because of the nonspecific biodistribution and the short plasma halflives (several minutes in mice and up to approximately 4 hr in humans), the in vivo effects of systemically administered cytokines depend on the mode and route of administration. It was suggested that the antitumor effects of IL-2 correlate more with the duration of exposure to a certain level than with the peak plasma level, whereas toxicity was mainly correlated with the latter (Cheever et al., 1985; Ettinghausen and Rosenberg, 1986; Zimmerman et al., 1989a). Bolus intravenous (iv) administration leads rapidly both to high plasma levels and clearance. In melanoma patients treated with high-dose I L 2 by either bolus or continuous iv infusion, the former mode appeared slightly more toxic and less immunostimulatory, but more efficient therapeutically, although

264

ELI KEDAR A N D EVA KLEIN

the total I L 2 dose administered per day was usually higher with the bolus administration (Rosenberg et al., 1987a; West et al., 1987; Sosman et al., 1988; Thompson et al., 1989; Clark et al., 1990; Dutcher et al., 199 1). Comparison of the therapeutic efficacy and toxicity between bolus and continuous infusion of IL-2 has been carried out in only a few randomized clinical trials. T h e two ways of administration were equally effective in renal cell carcinoma, whereas the bolus treatment was more beneficial in melanoma patients (reviewed in Parkinson, 1990). In mice carrying intraperitoneal (ip) tumor grafts, three daily ip injections of IL-2 for 7 days was significantly more efficient therapeutically than continuous administration with the Alzet microosmotic pump (E. Kedar, unpublished observations). Bolus iv administration of IL-2 was also more efficient than continuous infusion in other experimental models (Zimmerman et al., 1990; S. Slavin, personal communication). Additional studies are also warranted to evaluate single or a few highdose courses versus multiple, low-dose treatments extended in time. A possible drawback of the latter schedule is that the therapeutic efficacy may gradually decline due to increasing serum levels of soluble cytokine receptors (Lotze et al., 1987) and/or the appearance of antibodies against the cytokines (Oberg et al., 1989; Kirchner et al., 1991). It may be possible to reduce levels of such antibodies by alternate cycles of recombinant and natural cytokines (Wussow et al., 1990). In one study, local continuous infusion of natural IL-2, but not recombinant I L 2 , led to complete regressions of advanced bladder carcinomas in several patients (Huland et al., 1990). It is still unknown whether or not the soluble cytokine receptors (Fernandez-Botran, 1991) or anti-cytokine antibodies jeopardize the therapeutic efficacy in patients. Due to the short plasma half-life, continuous iv infusions or frequent repeated bolus iv administrations of high (and toxic) doses of cytokines (in particular, IL-2) have been commonly used in patients. Therefore, in order to reduce toxicity and also to simplify the treatment, alternative routes were tested, such as ip or subcutaneous (sc) administrations, already shown to be effective in the prolonged maintenance of high cytokine levels (Gustavson et al., 1989; Urba et al., 1989; Cebon et al., 1990). High ip cytokine levels may be advantageous for patients with tumors confined to the peritoneal cavity, such as in ovarian and colon carcinomas (Berek, 1990; Steis et al., 1990). However, a common side effect of the ip infusion of several cytokines is the formation of local fibrosis (reviewed in Kovacs, 1991). The sc route, on the other hand, may facilitate production of neutralizing anti-cytokine antibodies (Kolitz et al., 1988; Oberg et al., 1989; Atzpodien et al., 1990a; Whitehead et al., 1990). T h e use of infusion pumps, similar to those introduced to pa-

CANCER IMMUNOTHERAPY

265

tients for the local continuous instillation of chemotherapy, may be considered also for cytokines. T h e impact of the route of cytokine administration was studied in tumor-bearing mice. Comparing the ip and iv routes for IL-2, the former was more effective for ip tumors, whereas for pulmonary metastases the latter was superior (Kedar et al., 1988a; R. Zimmerman, personal communication). In mice with spontaneous pulmonary metastases, TNFa was effective when administered iv but not ip (Talmadge, 1988). These observations can be explained by the fact that the route of administration markedly influences the tissue distribution, and by the different mechanisms of action of the various cytokines. For tumors confined to accessible compartments, a locoregional administration (intratumoral, periturnoral, perilymphatic) may be effective. Periturnoral injections of IL-1, IL-2, and I L 4 , individually or in various combinations, caused marked responses in mice (Vaage, 1988; Bubenik, 1989; Forni et al., 1989; Bosco et al., 1990). In patients with head and neck tumors, local administration of IL-2 caused transient tumor regression (Musiani et al., 1989) and local activation of specific and nonspecific killer cells (Rivoltini et al., 1990). Beneficial effects were also reported with I L 2 with or without LAK cells administered ip (in colon and ovarian carcinoma; Urba et al., 1989; Stewart et al., 1990; Steis et al., 1990), intrapleurally (in mesothelioma; Yasumoto et al., 1987; Eggermont et al., 1991), intravesically (in bladder carcinoma; Pizza et al., 1984; Huland and Huland, 1989; Huland et al., 1990), and intracerebrally (in brain tumors; Yagita and Grimm, 1988; Yoshida et al., 1988). Local administration of TNFa, IFNP, or IFNy was also found effective as a palliative treatment in patients with different types of solid neoplasms (Bezwoda and Dansey, 1990; Schmid et al., 1990; Wildfang et al., 1990; Boutin et al., 1991; Pujade-Lauraine et al., 1991; Rath et al., 1991). For disseminated diseases, the local treatment may be supported by systemic treatment. 8. New Methods f o r Delivery of Cytokines and Other Biologzcal Response Modifiers

The problems raised by the short plasma half-lives, requiring continuous infusion or frequent bolus administration of high doses of cytokines, as well as the nonspecific biodistribution and systemic toxicity, may be alleviated by resorting to controlled-release vehicles, using techniques already available for the delivery of conventional drugs. These include ( 1) incorporation into microvesicles (phospholipid liposomes) that are introduced systemically, (2) entrapment within polymeric materials (minipellets) or pumps, which are placed in direct contact with the

266

ELI KEDAR A N D EVA KLEIN

tumor, and (3) modification by chemical means (reviewed in Langer, 1990). As opposed to other slow-release vehicles, the distribution of liposomes in the various tissues can be controlled by altering their lipid composition and size and by using various administration routes (Gregoriadis, 1990). “Classical” liposomes (negatively charged and larger than 0.1 pm) are rapidly taken up by phagocytes (mainly in liver, spleen, and lungs) and thus can be effective for tumors in these organs through local activation of macrophages (Schroit et al., 1986; Brodt et al., 1989; Phillips, 1989; Whitworth et al., 1990; Nii et al., 1991). In contrast with the classical liposomes, the recently introduced “stealth” liposomes [containing specific glycolipids and phospholipids of high phase-transition temperature ( T J , or polyethylene glycol (PEG)-derivatized phospholipids and lipids of either high or low T,, with or without cholesterol, and smaller than 0.1 pm] can evade the reticuloendothelial system, thereby achieving prolonged circulation time and enhanced accumulation in tumors localized in various body compartments (Gabizon, 1989, 1991; Gabizon and Papahadjopoulos, 1988; Ranade, 1989; Gabizon et al., 1990; Gregoriadis, 1990; Papahadjopoulos and Gabizon, 1990; Mayhew and Lasic, 1991; Papahadjopoulos et al., 1991). The entrapped cytokines may even be taken up through endocytosis by cells that lack surface cytokine receptors (Fidler et al., 1985).The poor extravasation capacity of certain types of liposomes (Matzku et al., 1990) can be overcome by administering them simultaneously with cytokines that increase vascular permeability, such as TNFa (Suzuki et al., 1990) and I L 2 (Schultz et al., 1991).Other agents with vasoactive activity [e.g., prostaglandin E, (PGE,), histamine, leukotrienes] (Hennigan et al., 1991), or hyperthermia, may also be useful. Even if large “stealth” liposomes fail to extravasate, they can still function as circulating microreservoirs, slowly releasing the entrapped BRM. The availability of small stealth liposomes with prolonged circulation time and good extravasation capacity opens the possibility of targeting liposomes to tumors in various locations by using them equipped with monoclonal antibodies [or the F(ab’) fragments] directed against tumor antigens (immunoliposomes)(Hashimoto et al., 1983; Freeman and Mayhew, 1986; Papahadjopoulos and Gabizon, 1987; Bankert et al., 1989; Singh et al., 1989; Akaishi et al., 1990; Kumai et al., 1990; Yemul et al., 1990). The immunoliposomes can be administered systemically or locally (such as in ovarian carcinoma; Nassander, 1991). In addition to the “classical”putative tumor-associated antigens, other molecules expressed on the tumor cells can be exploited for targeting. Theoretically, for tumor cells “overexpressing”growth factor receptors

CANCER IMMUNOTHERAPY

267

[e.g., for epidermal growth factor (EGF), platelet-derived growth factor (PDGF), estrogen, transferrin, bombesin] (reviewed in Sat0 and Sato, 1989; Herlyn et al., 1990; Cantley et al., 1991; Hunter, 1991; Lippman, 1991), targeting may be enhanced by coupling to the liposomes the respective growth factor or growth factor analog (e.g., tamoxifen in breast carcinoma), or antibodies directed to the receptor. Antibodies to cell surface adhesion molecules may also be used for targeting. Increased binding to tumor cells in vitro was demonstrated with nerve growth factor (M. B. Rosenberg et al., 1987) and anti-laminin receptor antibody (Rahman et al., 1989)-conjugated liposomes. The same possibility applies to tumor cells displaying on their surface high levels of oncogene products (e.g., HER-2lneu) (Hellstrom and Hellstrom, 1989; Yarden, 1990; Tecce et al., 1991; Xu et al., 1991). Thus, compared with free, soluble BRMs, the treatment with BRMcontaining targeted liposomes for site-specific delivery is likely to be more efficient, more cost effective, and less toxic. Other advantages of liposome-encapsulated cytokines are their relatively long shelf life (at 4"C), allowing preparation of large batches sufficient for several months (E. Kedar and Y. Barenholz, unpublished observations), and protection from antibodies in uivo (Debs et al., 1989). Liposome-encapsulated cytokines and other BRMs have already been utilized in experimental systems and in patients (Fidler, 1988b; Brenner, 1989; Whitworth et al., 1990; Kleinerman, 1991). The antiproliferative activity of IFNa on human tumor cell lines in uitro was markedly enhanced when added in liposomes (Killion et al., 1989; Shin et d., 1990). In animal models, better antitumor responses were obtained with encapsulated TNFa (Debs et al., 1990) and IL-2 (Anderson et al., 1990; Loeffler et al., 1991a; E. Kedar and Y. Barenholz, unpublished results) than with the soluble ones. Similarly, encapsulated muramyl dipeptide (MDP; Phillips et al., 1987; Fidler, 1988b) and muramyl tripeptidephosphatidylethanolamine (MTP-PE, MacEwen et al., 1989; Kleinerman, 1991), exhibited stronger therapeutic effects and the effects were obtained with much lower doses, as compared with the unmodified agents. In an attempt to further enhance therapeutic efficacy, combinations of cytokines and encapsulated BRMs have been tested recently. Treatment with liposomal MTP-PE and granulocyte/macrophage-CSF (GM-CSF) increased the survival of nude mice carrying human ovarian cancer xenografts, compared to mice given either agent singly (Malik et al., 1991). A synergistic antitumor effect was also found in mice treated with liposomal MTP-PE and IL-2 (Dinney et al., 1991). In clinical trials, encapsulated MDP was beneficial in patients with prostate cancer (Vosika et al., 1990), and phase 1/11 trials with encapsu-

268

ELI KEDAR A N D EVA KLEIN

lated MTP-PE began recently in patients with advanced cancer, alone (Kleinerman et al., 1989; Creaven et al., 1990; Urba et al., 1990; Frohmiiller et a!., 1991; Kleinerman, 1991) and in combination with IFNy (Lautersztain et al., 1991). Encouraging results were recently obtained in children with osteosarcoma receiving encapsulated MTP-PE as postsurgery adjuvant treatment (I. Fidler, personal communication; Mihich, 1991). Another novel approach for regional delivery of cytokines to tumor sites is the use of solid-phase cytokines. Interleukin 2 coupled to polystyrene beads was more efficient than soluble I L 2 in vivo in the induction of cytotoxic activity and local tumor growth inhibition in rats (Crum and Kaplan, 1991). Prolongation of the IL-2 (Katre et al., 1987) and TNFa (Noguchi et al., 1991) half-life in uiuo has been achieved by coupling them to a carrier molecule. In mice, polyethylene glycol (PEG)-conjugated IL-2 had a 20fold longer plasma half-life, compared to soluble IL-2. Moreover, the antitumor effect of PEG-IL2 was superior, at equitoxic doses, and it could be obtained with fewer and less frequent administrations (Katre et al., 1987; Zimmerman et al., 1989a;J. C. Yang et al., 1990b; E. Kedar, unpublished results). Recently, w e have found that the LAK cell activity of peritoneal cells and splenocytes of mice inoculated ip or iv with PEGIL-2, or liposomal I L 2 (each given in two doses, 3-4 days apart) was 2030 times greater than that of cells derived from mice given the same total amount (1-2 x lo5 Cetus units) of the unmodified cytokine (introduced once daily over 5 days) (E. Kedar and Y. Barenholz, unpublished observations). Polyethylene glycol-IL-3 and polyethylene glycol-G-CSF were recently found to increase the blood leukocyte count in animals more efficiently compared to the unmodified cytokines (Bree et at., 1991; Tanaka et al., 1991). C. ADOPTIVE IMMUNOTHERAPY: LYMPHOKINE-ACTIVATED KILLER(LAK) CELLS AND TUMOR-INFILTRATING LYMPHOCYTES (TIL) The combination treatment with I L 2 and LAK cells (Rosenberg et al., 1985a, 1989b), which led to complete regression of advanced tumors in some patients with melanoma and renal cell carcinoma, raised criticism because of the severe toxicity and the low rate of success (Moertel, 1987; Quirt and Tannock, 1990). T h e treatment with activated autologous lymphocytes (LAK or TIL) entails complicated logistics and its practice is limited. Therefore it is

CANCER IMMUNOTHERAPY

269

important to summarize the available experience in order to ascertain whether such efforts are justified.

I . Lymphokine-Activated Killer Cells: Experimental Models Lymphokine-activated killer cells (or TIL) may inflict tumor damage directly and/or indirectly by releasing immunopotentiating (e.g., I L 2 , I L 4 , IL-6, IFNy) and antiproliferativeicytolytic (e.g., TNFa, TNFP, TGFP, perforin, serine esterases) molecules (Henkart and Yue, 1988; Berke, 1989; Hersh et al., 1989; Larisch-Bloch et al., 1989, 1990; Mazzocchi et al., 1990; Barth et al., 1991). In mice with a relatively small tumor burden, LAK cells and IL-2 were shown to act synergistically; treatment with LAK cells without IL-2 had no or minimal therapeutic effects (reviewed in Rosenberg, 1986, 1988, 1991a; Rosenberg et al., 1989a). In an experimental model of postsurgical adjuvant immunotherapy, treatment with LAK cells improved the results in mice receiving low, but not high, doses of IL-2 (Rodolfo et al., 1990). In mice with advanced sarcoma or carcinoma, both of low immunogenicity, cojoint administration of LAK cells did not improve the cure rate achieved by a combination of cyclophosphamide and intermediate doses of IL-2; moreover, treatment with IL-2 and LAK cells without cyclophosphamide had no effect at all (Kedar et al., 1988a, 1989). Thus in animals with bulky disease, the therapeutic gain from LAK cell administration seems to be minimal, particularly when given in combination with high-dose I L 2 or with I L 2 and cytotoxic chemotherapy.

2. Lymphkine-Activated Killer Cells: Clinical Trials In many of the clinical trials, mostly nonrandomized, the objective response rate [partial and complete remissions (PR and CR, respectively)] was about equal for IL-2 given with and without LAK cells (Boldt et al., 1988; Bergmann, 1989; West, 1989; Margolin et al., 1989; Parkinson, 1990; IMlman et al., 1991b; McCabe et al., 1991; Oliver, 1991). In a few studies, LAK cell infusions to IL-2-treated patients slightly increased the incidence of complete remission and the duration of response in patients with melanoma and renal cell carcinoma (Nkgrier et al., 1989; Rosenberg, 1990, 1991a; Rosenberg et al., 1989a,b; Parkinson, 1990; Dillman et al., 1991a; Escudier et al., 1991). In patients with melanoma and renal cell carcinoma, the success with IL-2 treatment, with and without LAK cells, differs in various trials. T h e relatively high response rate (CR plus PR, 20-35%) reported in several studies (Rosenberg et al., 198913; Rosenberg, 1990; Parkinson et al.,

270

ELI KEDAR AND EVA KLEIN

1990b; Thompson et al., 1991a) was not confirmed (0-20% response, or no CR) in other studies (Dutcher et al., 1989, 1991; Sondel and Hank, 1989; Abrams et al., 1990; Bar et al., 1990; Clark et al., 1990; Gaynor et al., 1990; Hercend et al., 1990; Parkinson et al., 1990a; Dillman et al., 1991b; Kim and Louie, 1991; McCabe et al., 1991). Since the patient groups were heterogeneous and they were treated with different I L 2 doses, administered at various schedules, sometimes even in combination with other agents (e.g., cyclophosphamide, indomethacin), comparisons and judgments for the optimal treatment conditions are not possible. The higher response rates reported in several studies may be attributed to more aggressive treatments (i.e., higher amounts of I L 2 andlor LAK cells; bolus vs continuous I L 2 infusion) and to differences in the criteria for selection of the patients. The review of these reports leads to the conclusion that the therapeutic value of LAK cells as an adjunct to I L 2 treatment is questionable. In fact the cells may even aggravate the toxicity (Ettinghausen et al., 1988; Albertini et al., 1990; Siegel and Puri, 1991), by damaging normal tissues. Mouse and human LAK cells were shown to be moderately cytotoxic in uitro for normal cells, such as lymphocytes, endothelial cells, and monocytes (Kedar et al., 1982, 1983; Damle et at., 1987; Djeu and Blanchard, 1988; Borden and Sondel, 1990; Zambello et al., 1990; Siegel and Puri, 1991). In spite of the discouraging clinical results with LAK cell therapy, the use of nonspecifically activated cells has not been abandoned. Recently, patients with metastatic renal cell carcinoma were treated with autologous blood lymphocytes, activated in uitro by crude supernatants of OKT3 antibody (a T cell mitogen)-stimulated autologous lymphocytes. A response rate of 2 1%, a clear survival benefit, and minimal toxicity were reported (Osband et al., 1990, 1991).In another approach, patients with metastatic carcinoma and melanoma were treated with autologous blood-derived monocytes briefly activated in uitro with IFNy (Keller, 1989). In several patients with malignant ascites, ip infusions of the cells led to disappearance of the local tumor (Andreesen et al., 1990). 3 . Tumor-InfiltratingLymphocytes: Experimental Systems and Clinical Trials

The limited therapeutic capacity of blood-derived LAK cells motivated the search for more potent effector cells. It can be assumed that T cell populations collected from the tumor tissue (TIL) would be enriched in lymphocytes with specificity for the tumor (Vose and Moore, 1985; Rosenberg et al., 1986, 1989a,b; Rosenberg, 1991a; Itoh et al., 1988; Whiteside et al., 1988; Maleckar et al., 1989; Topalian et al., 1989;

CANCER IMMUNOTHERAPY

27 1

Parmiani et al., 1990). In mice with tumors of low immunogenicity treated with low-dose I L 2 and TIL or LAK cells, the former cells were 50- 100 times more potent in eradicating micrometastatic disease (Rosenberg et al., 1986). Such an effect was not seen, however, in mice with nonimmunogenic tumors (Spiess et al., 1987). In a model for the clinical situation, treatment of mice carrying spontaneous metastases (after resection of the primary tumor) with IL-2 and tumor-specific T cells, generated from splenocytes of immune mice in mixed lymphocyte-tumor cell cultures, was considerably more effective than with LAK cells (Rodolfo et al., 1990). In the limited number of clinical trials, the results with TIL were disappointing. In melanoma and renal carcinoma patients, as well as in other malignancies, the response rate to I L 2 and TIL (10-30%) was not better than that reported for I L 2 and LAK cells (Kradin et al., 1989a; Dillman et al., 1990a, 1991a,c; Hanson et al., 1991; Markowitz et al., 1991; Thompson et al., 1991b). A higher response rate (55%)was obtained in 20 melanoma patients by Rosenberg et al. (1988), but the responses were mainly partial and of short duration. In addition to the questionable therapeutic benefit of TIL, the possibility for their use is very limited. Tumor-infiltrating lymphocytes in sufficient numbers can be collected only rarely, and extensive enlargement of the cell population in uitro must be interposed before their use. This step is not always successful. Prolonged cultivation of the lymphocytes with IL-2 (usually 1-2 months is required) can abrogate selectivity for the tumor. The “contaminating” nonselective killer cells (NK) expand also (Vanky et al., 1982; Whiteside et al., 1988; Topalian et al., 1989; Lotzova et al., 1990), and even the tumor-selective T cells can acquire broader reactivity when exposed to IL2. Moreover, only a small fraction of the readministered human TIL reaches the tumor (Griffith et al., 1989). In control mice (i.e., without tumors), iv-administered T I L localized preferentially in the liver and lungs (Wong et al., 1991). In our studies, iv-injected cultured TIL and spleen-derived LAK cells accumulated mainly in the liver and lungs of both control and tumor-bearing mice (with pulmonary metastases), while the corresponding freshly explanted lymphoid cells localized mainly in the spleen (E. Kedar, unpublished results; Gazit et al., 1992). Although broadening of the cytotoxic potential for allogeneic tumor cells (which is often seen in T cell cultures) is not relevant for the therapeutic effects, damage inflicted on normal cells and poor tumor localization impose serious limitations for the use of lymphocytes activated and expanded in culture. In recent studies, a small fraction of readministered TIL was traced in the tumor tissue of mice and patients for several weeks or months.

272

ELI KEDAR A N D EVA KLEIN

Human TIL transfected in uitro with the neomycin-resistance gene and returned to the patients intravenously could be detected by the highly sensitive polymerase chain reaction (PCR) technique. In three of five melanoma patients, the cells were traced for 6 to >60 days in cutaneous tumor biopsies (Rosenberg et al., 1990). In mice with pulmonary metastases, readministered TIL were found in the lungs 4 months later (Alexander and Rosenberg, 1991). In an attempt to use TIL on a rational basis, their in vitro performance was tested in a group of melanoma patients treated with TIL and IL-2. T h e cells of those who showed a clinical response had a stronger autotumor cytotoxicity in vitro than did TIL of the nonresponding patients (median cytotoxicity 18 vs. 4.5%,respectively) (Aebersold et al., 1991). In similar murine studies, however, the therapeutic effect of TIL correlated better with secretion of IFNy and TNFol on exposure to the relevant tumor cells than with the autotumor cytotoxic activity in vitro (Barth et al., 1991). It seems therefore that the recognition of tumor cells in nitro by the TIL population has a predictive value. T h e employment of tumor-reactive cells can be further refined. Cytotoxic or helper T cell clones specific for the tumor cells could represent the ideal tools for immunotherapy (Greenberg et al., 1988; Greenberg, 1991; Melief, 1992). However, due to the difficulty of establishing such clones from patients and the limitations in their expansion, this strategy may be possible only in highly specialized centers and for only a small number of patients.

4. Genetically Enganeered Tumor-In.h a t i n g Lymphocytes Readministered T cells may be exploited as vehicles for genes encoding cytokines/cytotoxins that can be inserted into the cells by the retroviral-mediated gene transfer technique (Kohn et al., 1987; Blease, 1991). Even if only a small proportion of the cells localize in the tumor, the engineered cells may release large amounts of the products of the transfected genes for extended periods (Kasid et al., 1990; Rosenberg et al., 1990; Rosenberg, 1991a; Morecki et al., 1991; Culver et al., 1991a; reviewed in Kinnon and Levinsky, 1990; Russell, 1990; Friedmann, 1991). It can be expected that such lymphocytes may release quantities of the cytokine in the vicinity of the tumor that highly exceed the local concentration achievable by systemic administration, without causing severe systemic toxicity. However, because the majority of the cells do not reach tumor sites (Griffith et al., 1989), the cytokines released constitutively at other sites may inflict damage on normal cells as well. Theoretically, toxicity could be minimized if the transduced gene could be regulated

CANCER IMMUNOTHERAPY

273

and expressed only if the lymphocytes are activated through confrontation with the tumor cells. T h e provision of cytokines inducing or potentiating endogenous effector cells at the tumor site, and of cytokines with a direct effect on tumor growth, may even help generate a response against poorly immunogenic and “nonimmunogenic” tumors. Several animal models suggest that “local help” can induce efficient antitumor responses (Fearon et al., 1990; reviewed in Kinnon and Levinsky, 1990; Russell, 1990; Blankenstein et al., 1991a). Cells from an antitumor CTL clone transfected with the IFNy gene had a stronger antitumor effect in mice than the unmodified lymphocytes (Miyatake et al., 1990). Tumor rejection was reported in euthymic and athymic nude mice when lymphocytes or fibroblasts transfected with the I L 2 (Bubenik et al., 1988, 1990) or the IFNy gene (Ogura et al., 1990) were deposited in its vicinity. This new approach must be exercised with caution, however, because it can lead to adverse effects. Tumors may become more aggressive under the influence of certain cytokines, as it has recently been shown in I L 2 (Kedar et al., I989), IL-1, or TNFa (Bani et al., 1991) treated euthymic and athymic mice, and also with tumor cells transfected with the TNFa gene (Malik et al., 1990). The growth-promoting activity of TNFa in tumor-bearing mice has also been shown in other studies. Antibodies against TNFa inhibited local tumor growth (Gelin et al., 1991a) and metastasis (D. Mannel, personal communication). 5. Adoptive Immunotherapy with Monoclonal Antibodies

The clinical experience with antitumor monoclonal antibodies (MAbs) in patients with solid tumors has been, generally, disappointing (reviewed in Catane and Longo, 1988; Foon, 1989; Rosenberg et al., 1989a; Chapman et al., 1991; Mach et al., 1991). However, MAbs can be useful in combination with LAK cells, T cells, and cytokines in inducing antibody-dependent cellular cytotoxicity (ADCC). Treatment with IL-2 or other cytokines, with and even without exogenous LAK cell administration, was more efficient in tumor-bearing euthymic and athymic mice when MAbs were also administered (Eisenthal et al., 1987, 1988; Kawase et al., 1988; Eisenthal and Rosenberg, 1989; Gill et al., 1989; Junghans, 1990; Schultz et al., 1990; Pendurthi et al., 1991; Van Dijk et al., 1991; Gazit et al., 1992). In patients, systemic treatment with tumorreactive antibodies and IL-2 was not effective (Rosenberg et al., 1989b; Bajorin et al., 1990; Ziegler et al., 1991), whereas local administration of antibody-coated LAK cells (but not untreated LAK cells) caused complete tumor regression in several patients with glioma (Nitta et al., 1990).

274

ELI KEDAR A N D EVA KLEIN

This strategy depends on the availability of tumor-reactive antibodies with the appropriate isotype (Vuist et al., 1990).High levels of circulatory soluble tumor antigens can interfere with this treatment. It is possible that once satisfactory MAbs are available, the administration of effector cells may be superfluous, unless the patient’s cellular functions are deficient.

D. CHEMOIMMUNOTHERAPY 1. Introduction Efficient chemotherapy must eradicate all the tumor cells. This cannot be achieved even in cases where the tumor cells are sensitive to the drug. There is a limitation to the tolerated dose and, frequently, variant cells that are resistant to the drugs arise in the population. Immunotherapy triggers the host defense mechanisms. These modalities rarely have overlapping toxicities, therefore a combination of the two can increase the therapeutic index (Mitchell, 1988; LoRusso et al., 1990). Moreover, drug-resistant tumor cells are still sensitive to an immunological attack (Gambacorti-Passerini et al., 1988). There are several differences between patient response to chemotherapy and to immunotherapy: (1) In general, only those patients achieving a “complete response” to chemotherapy experience a significant increase in survival, whereas even partial responses to immunotherapy can result in prolonged survival; (2) while the effect of chemotherapy is over with the cessation of treatment, clinical responses can occur weeks or months after immunotherapy (Hamblin, 1989); (3) response to chemotherapy, in most cases, is dose dependent, whereas with immunotherapy doseresponse relationships are less marked (Talmadge, 1988); and (4) chemotherapy is more effective than immunotherapy for bulky disease, whereas immunotherapy is probably more effective for minimal residual disease (Wadler, 1991). The goal is thus to reduce the tumor load by chemotherapy, and perhaps also to diminish suppressor cell activity, whereby the chance for the endogenously generated or adoptively transferred effector cells of coping with the residual disease is increased (Fefer, 1974; Fefer et al., 1976, 1982; Kedar and Weiss, 1983; Mitchell, 1988; Greenberg, 1991; Longo, 1991). Since many of the chemotherapeutic agents are immunosuppressive, the protocols of chemoimmunotherapy need to be carefully designed. With a few exceptions (see below), the cytoreductive therapy must be given prior to the immunological measures, at doses close to the maximal tolerance, with intervals allowing the recovery of immunohematopoietic functions but avoiding tumor regrowth (Mitchell,

CANCER IMMUNOTHERAPY

275

1988). However, in many clinical trials, chemotherapy and immunostimulating agents have been applied concurrently, and when sequentially, without attention to the optimal intervals and/or sequences. Therefore immunohematopoietic recovery must be followed in order to design the treatment schedule optimally.

2. Experimental Systems: Chemoimmunotherapy with IL-2 Additive or synergistic effects of chemotherapy (mostly cyclophosphamide) followed by IL-2, with or without LAK cells or CTLs, in tumor-bearing mice were reported by several groups (Kedar et al., 1984a, 1988a, 1989, 1990; Rosenberg et al., 1986; Silagi and Schaefer, 1986; Greenberg et al., 1988; Wiltrout and Salup, 1988; Eggermont and Sugarbaker, 1988; Papa et al., 1988; Formelli et al., 1988; LoRusso et al., 1990; Greenberg, 1991). T h e importance of the sequence and timing in chemoimmunotherapy protocols was studied by us in mice with advanced, weakly immunogenic carcinomas and sarcomas. T h e treatment consisted of cyclophosphamide (100-150 mg/kg) and I L 2 (5-10 x lo4 Cetus U/day, for 5-6 days). Synergistic effects were observed only when IL-2 was given after cyclophosphamide, optimally 2-4 days later, whereas the reverse sequence was ineffective (Kedar et al., 1988a, 1989, 1990). Similar findings were reported by other investigators (Hosokawa et al,, 1988). It is likely that with the former schedule, chemotherapy reduced the tumor load (and perhaps also eliminated suppressor cells), thereby allowing the IL-2-stimulated effector cells to act more efficiently on the smaller number of targets, whereas in the reverse sequence the IL-2induced lymphocyte activation and proliferation was counteracted by the chemotherapy. In other reports, however, a potentiation of antitumor response was shown when I L 2 was given before but not after chemotherapy (Rinehart et al., 1990; Wolmark et al., 1990). The contradictory findings may be due to differences in several parameters, such as the dosages, the treatment schedules, the drugs, and the tumor models, and show the difficulties in the design of treatment protocols in the clinic. T h e immunosuppressive effect of the given chemotherapy regimen should thus be considered when planning the sequences and intervals. Thus, cytokines whose lymphoproliferative effect is exploited should not be given immediately prior to immunosuppressive chemotherapy. When the chemotherapy drugs are weakly immunosuppressive (e.g., doxorubicin, dacarbazine) (Mitchell, 1988), or when cytokines with other effects are used (e.g., I L 1 ) (Nakamura et al., 1991), the sequence and timing may be less important.

276

ELI KEDAR A N D EVA KLEIN

3 . Experimental System: Chemoimmunotherapy with Interferon and Tumor Necrosis Factor In several animal models (using, however, mostly tumors with proven immunogenicity), sequential treatment with chemotherapeutic agents and TNFa o r IFN gave additive or synergistic effects. These cytokines do not induce lymphoproliferation, and in mice they were found almost always to be more effective when applied shortly before chemotherapy (Borden et al., 1988; Krosnick et al., 1989a; Kedar et al., 1992). In our experiments with mouse and human tumor grafts in euthymic and athymic mice, respectively, the best effects were obtained with IFNa given before and I L 2 given after chemotherapy (Gazit et al., 1992; Kedar et al., 1992). It can be assumed therefore that IFN and TNF may act better in patients if applied prior to chemotherapy rather than concomitantly (which is, however, the current procedure). These cytokines can also protect the patient from myelosuppression because they inhibit the proliferation of hematopoietic cells (Mitchell, 1988; Talmadge, 1988; Richman et al., 1990; Wadler and Schwartz, 1990). Pretreatment with low-dose IFN or TNF could thus allow the delivery of cytotoxic agents more frequently and/or in higher doses. While the inhibitory effect on cell proliferation can decrease sensitivity to drugs acting on cycling cells, tumor cells exposed to IFN (Elias and Sandoval, 1989; Yoneda et al., 1989; Wadler and Schwartz, 1990; Scala et al., 1991) o r to leukoregulin (Baker and Evans, 1990) had increased sensitivity to chemotherapy. This effect seems to depend on (1) increased drug uptake due to down regulation of the multidrug resistance-associated p 170 glycoprotein, and (2) modifications in intracellular drug metabolism. 4 . Clinical Trials with Various Cytokines

In several clinical trials, chemoimmunotherapy was not more effective than immunotherapy alone. The response rate (approximately 25%) in patients with metastatic melanoma treated with low- or high-dose I L 2 in combination with either DTIC (Shiloni et al., 1989; Dillman et al., 1990b; Flaherty et al., 1990a; Stoter et al., 1991; Isacson et al., 1992), doxorubicin (F'aciucci et al., 1990), or low-dose cyclophosphamide (used mainly as an immunomodulator) (Mitchell et al., 1988b; Lindemann et al., 1989; Rosenberg et al., 1989b)was similar to that achieved with high-dose l L 2 alone. Other combinations were more effective, however. In melanoma patients, approximately 40% responded to I L 2 combined with cisplatin (Atkins et al., 1990) or cisplatin and DTIC (Flaherty et al., 1990b; Blair et al., 1991), and to IFNa combined with DTIC (Breier et al., 1990; Mulder

CANCER IMMUNOTHERAPY

277

et al., 1990; Falkson et al., 1991). Higher response rates (60433%) were recently reported when I F N a was combined with DTIC and 5-FU (Mulder et al., 1991) and for IL-2 plus I F N a together with combination chemotherapy (Hamblin et al., 1991; Richards et al., 1991). T h e results are also encouraging in patients with colon cancer. A higher response rate and longer disease-free survival, compared to chemotherapy or biotherapy alone, were achieved with two combinations: (1) 5-FU plus I F N a , which gave a response rate of 20-40% (and up to 76%), including a few complete responses in metastatic disease (Wadler et al., 1989, 1991; Kemeny et al., 1990; Pazdur et al., 1990; Wadler and Schwartz, 1990; Inoshita et al., 1991; Wadler, 1991), and (2) 5-FU plus levamisole as postsurgery adjuvant treatment in Duke’s C colon cancer (Moertel et al., 1990). T h e combination of I F N a and 5-FU or other drugs was also effective for urothelial tumors (Logothetis et al., 1991; Ruther et al., 1991), stage IV head and neck cancer (Vokes et al., 1991), and follicular lymphoma (Solal-Celigny et d., 1991). A response rate of 30-40% was obtained in patients with renal cell carcinoma treated with IFNa and vinblastine (Dal Ri et al., 1990; Nordman and Kellokumpu-Lehtinen, 1990; Wadler and Schwartz, 1990). These responses appeared to be at least additive, as compared with single-modality treatments. In large groups of patients with various tumors, alternating chemotherapy with LAK cells and IL-2 gave better results than without chemotherapy (Dillman p t al., 1991a).

5. Immunomodulatory Efects of Chemotherapy Drugs used for chemotherapy can inhibit but also stimulate immune functions. Their effect is strongly influenced by the dose and the treatment schedules. Enhanced I L 2 production and NK/LAK cell activity, and amplification of T cell responses, were observed in mice treated with low doses of cyclophosphamide, doxorubicin, or cisplatin (Ehrke et al., 1982, 1986, 1988; Kedar et al., 1984b, 1986; Mitchell, 1988; Lafreniere et al., 1989). In our recent experiments (Kedar and Gazit, 1992), high-dose (100 o r 200 mg/kg) cyclophosphamide suppressed the in uitro proliferative and cytotoxic T cell responses to allogeneic cells and reduced the number of LAK cell precursors in mouse splenocytes, tested 14 days after its administration. On the other hand, these responses were markedly enhanced (2-20 times the normal levels) when tested 67 and 9-10 days, respectively, later. A transient decrease of LAK cell activity followed by an increase in vitro and in viuo was seen in athymic nude mice treated with 100 mg/kg of 5-FU or 100 mg/kg of dacarbazine. T h e postchemotherapy “overshot” effect, lasting for 2-4 days in mice, might be the optimal time window for administration of immunother-

278

ELI KEDAR AND EVA KLEIN

apy. Indeed, the therapeutic effect was found to be maximal when I L 2 was administered daily to tumor-bearing euthymic mice on days 3-8 after treatment with 100 mg/kg cyclophosphamide (Kedar et al., 1989, 1990). Similar chemotherapy-induced fluctuations in immune functions were also reported by other groups in animals and humans (Hosokawa et al., 1988; Cramer et al., 1989; Kim et al., 1989; Sarneva et al., 1989; Allavena et al., 1990; Karimine et al., 1990; Katsanis et al., 1990; Avner et al., 1990; Sensi et al., 1990). When lymphocytes of cancer patients were collected 5-7 days after treatment with mitomycin C and cultured with I L 2 they had a stronger cytotoxic activity than pretreatment lymphocytes cultured the same way (Nanbara et al., 1989).Ex vivo blood lymphocytes of patients 1 month after chemotherapy (without IL2) had LAK cell-like activity (Kiyohara et al., 1988) and an increased capacity to be triggered for cytotoxicity in culture, to respond to mitogens, and to produce I L 2 (Onodera et al., 1990). Changes in the composition of the cell population, including the elimination of suppressor cells (see the next section), may explain the elevation of certain immune functions (Berd et al., 1984; North, 1984; Mokyr and Dray, 1987; Awwad and North, 1989; Hoover et al., 1990). 6. Effects of Chemotherapy and Other Drugs on Suppressor Cells

Low-dose chemotherapy can eliminate suppressor cells or counteract their activation (Fefer et al., 1982; Kedar et al., 1984b, 1986; Livingston et al., 1987; Mokyr and Dray, 1987; Berd and Mastrangelo, 1988b; North et al., 1989; Hoon et al., 1990; reviewed in Naor et al., 1989). The existence of chemosensitive tumor-specific suppressor T cells and nonspecific suppressor cells has been documented in animals with advanced tumors. Depletion of such cells by low-dose cytoreductive therapy (primarily by cyclophosphamide), prior to adoptive cellular immunotherapy, enhanced the antitumor response in mice (Fefer et al., 1976, 1982; North, 1984; Awwad and North, 1988a,b), and in some studies (Berd et al., 1990a,b; Hoon et al., 1990), but not others (Morton et al., 1991a) appeared to improve the clinical response in patients treated with a melanoma cell vaccine. It is still not clear whether suppressor cells have any impact on immunotherapy in patients. Patients with solid tumors usually do not show impaired immune functions and they do not have circulating active suppressor cells (reviewed in Naor et al., 1989). Suppressor cells may be present, however, in the tumor or in the draining~lymphnodes (Parmiani et al., 1990).

CANCER IMMUNOTHERAPY

279

The role of suppressor cells in adoptive immunotherapy was assessed in mice treated with IL-2 and TIL as an adjunct to cyclophosphamide, whole-body low-dose irradiation, o r local tumor irradiation (Cameron et al., 1990). T h e immunotherapeutic effect was largely correlated with the reduction of the tumor load rather than with the elimination of suppressor cells. Prior treatment with cyclophosphamide or low-dose irradiation was essential for the effect of specifically sensitized T cells but not of LAK cells (Cameron et al., 1990). In mice carrying weakly immunogenic tumors treated with several chemotherapeutic agents at various doses in combination with IFNa and/or I L 2 , or with tumor-reactive T cells, additive/synergistic effects were obtained only when the chemotherapy itself had an antitumor effect (Formelli et al., 1988; Papa et al., 1988; E. Kedar, unpublished observations). Other drugs that were found to counteract suppressor cell functions in experimental models, in particular the type-2 histamine (H2) receptor antagonist, cimetidine (which inhibits the activation of H2 receptor-bearing suppressor cells), and the cyclooxygenase inhibitor, indomethacin (which blocks the production of the immunosuppressive mediator PGE,), were also employed in therapy. In mice, cimetidine was efficient in several tumor systems (Gifford et al., 1981; Osband et al., 1981), and it enhanced the therapeutic efficacy of I L 2 (Nakajima and Chu, 1991). Also, in patients with various types of solid tumors, cimetidine combined with histamine was beneficial (Burtin et al., 1988). In patients with metastatic melanoma, cimetidine improved the antitumor effect of IFNa in some (Flodgren et al., 1983) but not in other (Creagan et al., 1985) trials. In melanoma patients immunized with a melanoma cell vaccine, pretreatment with cimetidine augmented cellular and humoral responses to melanoma cells and appeared to improve the clinical response (Morton et al., 1991b). Indomethacin had therapeutic effects in several experimental tumor systems (Lynch et al., 1978; Fulton, 1988; Gelin et al., 1991b). It also enhanced the antitumor effects of I L 2 and IFNa, administered with and without chemotherapy, in euthymic and athymic mice (Kedar et al., 1984a; Lala and Parhar, 1988; Kim and Warnaka, 1989; Lala et al., 1990). However, patients treated with I L 2 (Sosman et al., 1988) or IFNa (Miller et al., 1989) did not benefit from indomethacin. It is possible that continuous, extended treatment with indomethacin is required (Lala et al., 1990), which was not the case in these clinical studies. Indeed, in a recent study on 25 patients with advanced melanoma receiving indomethacin (together with ranitidine) for several days, 2 patients showed an objective response (1 CR, 1 PR) even before I L 2 was administered (Mertens et al., 1991). The same treatment regimen with I L 2 was not more

280

ELI KEDAR A N D EVA KLEIN

effective, however, than treatment with 1 L 2 alone in patients with renal cell carcinoma (Bramwell et al., 1991). Indomethacin was also found to potentiate IL-2 production (Kedar et al., 1986), LAK cell induction, and ADCC activity in murine splenocytes (Eisenthal, 1990). Interestingly, indomethacin and other cyclooxygenase inhibitors (e.g., aspirin) diminished the toxicity of high doses of I L l , I L 2 , IFNy, or TNFa in tumor-bearing mice and rats without reducing the therapeutic efficacy (Haranaka, 1988; Talmadge, 1988; E. Kedar, unpublished observations). 7. Protection against MyelofImmunosu#n-ession with Cytokines

Certain cytokines can protect patients from the myelo/immunosuppression caused by high-dose chemotherapy, or radiotherapy. By accelerating the recovery of immunohematopoietic functions, they can shorten the intervals of cytoreductive therapy o r immunotherapy administrations, and may even allow a dose increment of chemotherapy, thereby enhancing antitumor effects. Administration of colony-stimulating factors (CSFs)-IL-3, GM-CSF, G-CSF, and M-CSF-as well as IL-1, IL-6, and IFN, individually or combined, to normal and tumor-bearing animals shortly before and/or after chemol radiotherapy was found to facilitate immunohematopoietic reconstitution and to enhance the therapeutic efficacy (Neta, 1988; Neta and Oppenheim, 1988; Kedar et al., 1988c; Slavin and Kedar, 1988; Futami et al., 1990; M. A. S. Moore et al., 1990). In cancer patients treated with cytoreductive therapy and CSFs, the number and severity of infections and hospitalization time were decreased, the patients required fewer blood or platelet transfusions, and chemotherapy could be administered more frequently or at higher doses, compared to patients not receiving CSFs (reviewed in Griffin, 1989; Groopman et al., 1989; Laver and Moore, 1989; Metcalf, 1989a,b; Bronchud, 1990; Demetri and Griffin, 1990; Ganser et al., 1990; Gianni et al., 1990; Golde, 1990; Kelso and Metcalf, 1990; Monroy et al., 1990; Richman et aL, 1990; Antman, 1991; Moore, 1991). In addition to the myeloprotective effects, CSFs and other cytokines (e.g., TNFa, IFNy) can amplify antitumor effector mechanisms by augmenting cytotoxic and phagocytic activities of monocytesfmacrophages and granulocytes. T h e beneficial effect of levamisole in combination with 5-FU in patients with colon cancer has been attributed, in part, to improved immunohematopoietic recovery, probably through inducing production of CSFs and other cytokines (Grem and Allegra, 1989). It is thus likely that optimal treatment with certain cytokines will

CANCER IMMUNOTHERAPY

28 1

provide both myeloprotection and immunostimulation, allowing the use of more intensive cytoreductive therapy, which may lead to a higher response rate. It should be noted, however, that CSFs must be employed with caution, since they may stimulate residual tumor growth, not only leukemias (Laver and Moore, 1989; Aglietta et al., 1990; Glaspy and Golde, 1990; Moore, 1991) but, as shown thus far in vztro, some types of solid tumors as well (Foulke et d.,1990; Joraschkewitz et al., 1990; Marshall and von Hoff, 1990; Pedrazzoli et al., 1990; Vellenga et al., 1991). 8. Conclusions

In addition to tumor debulking, the elimination of suppressor cells, and the transient increase in certain immune functions, the chemotherapeutic agents may also potentiate immunotherapy by: (1) increasing the sensitivity of‘ tumor cells to immunological attack, by arresting them in the G,/G, phase and/or by inducing biochemical alterations of the cell membrane, (2) imposing antigenic changes on the cells by acting as a hapten, (3) facilitating infiltration of effector cells to tumor sites, (4) potentiating stimulation of effector cells as a consequence of massive release of tumor antigens, (5) making “space” for adoptively transferred effector cells, and (6) protecting against toxicity induced by some cytokines (e.g., I L 2 ) (Giampietri et al., 1981; Mokyr and Dray, 1987; Mitchell, 1988; Hosokawa et al., 1988, 1990; Lee et al., 1988; Papa et al., 1988; Wiltrout and Salup, 1988; Kovach, 1991). Although not yet proven in clinical studies, chemotherapy may also reduce production of neutralizing antibodies to exogenously administered recombinant cytokines, and suppress the human anti-mouse antibody (HAMA) response in patients treated with mouse monoclonal antitumor antibodies. In order to best exploit these various actions of chemotherapy, the drugs must be carefully selected and their schedule and dosage should be adjusted to the individual, with evaluation of the immunosuppressive and immunostimulatory effects. During treatment with either cytokines or effector cells, repeated low-dose chemotherapy courses may also be required to maintain the reduced tumor burden. Better clinical results with chemoimmunotherapy may be obtained with new combination chemotherapy protocols based on the above considerations. It should be noted that no systematic clinical studies have yet been carried out to determine the optimal schedule, sequence, time interval, and duration of treatment for combined chemoimmunotherapy regimens, which may explain, in part, the meager success in many trials.

282

ELI KEDAR AND EVA KLEIN

IV. Attempts to Improve Cancer lrnrnunotherapy

Based on the new developments in biotechnology and on the results in experimental systems and in clinical trials, we summarize the considerations that can lead to improvement of cancer immunotherapy. These are outlined below and rely on the data reviewed in the previous sections. Considerations in Cancer Immunotherapy A. Basis for patient selection 1 . Tumor load/location

2. Karnofsky performance 3. MHC antigen expression on the tumor cells 4. Cell-mediated recognition of the tumor cells in vitro

J. B. Tumor debulking 1. Surgery 2. Radiotherapy/chemotherapy a. Supported by administration of colony stimulating factors (CSF) b. Periodic monitoring of general immune competence C. Active immunotherapy 1. Tumor vaccines a. Supported by adjuvants b. Inhibition of suppressor cells 2. Cytokines and other BRMs a. Combinations b. Targeted delivery D. Passive/adoptive immunotherapy 1 . Tumor-reactive monoclonal antibodies (MAbs) 2. Tumor-reactive T-cells

A. GAINSFROM EXPERIMENTAL MODELS

The rationale of immunotherapy was based on results obtained in experimental systems. It can be argued that the majority of animal models do not represent clinical situations (Hewitt, 1978, 1982; Weiss, 1980b; Herberman, 1983a,b). In many of the preclinical studies, long-passaged immunogenic tumors were selected and usually grafted at sites other than the tissue of origin. Moreover, immunotherapy was frequently initiated concomitantly or soon after tumor grafting. Immunotherapeutic manipulations were usually not effective in animals with established tumors, particularly if the latter were poorly or nonimmunogenic. This criticism is, however, only partly valid, because the relevant model system

CANCER IMMUNOTHERAPY

283

is hardly available (Scott, 1991). In addition, a considerable body of information has been collected in the animal experiments. Among these are the importance of tumor immunogenicity and the T cell response for the outcome of immunotherapy, the potential of BRMs, and the analysis of critical factors in combined immunotherapy regimens. T h e natural history of the majority of human tumors, and the fact that the common cancers do not show higher incidence in immunodeficient individuals, indicates that these tumors are rarely recognized by the immune system. T h e tumor cells may lack immunogenic determinants; and if they carry such, the host may have developed tolerance. Therefore, animal models using low or nonimmunogenic metastatic tumors must be concentrated upon (McIntosh et al., 1990; Rodolfo et al., 1990; Sakai et al., 1990), and attempts should be made to induce the immunological recognition of tumor cells. Human tumors can be studied in athymic nude mice (Giovanella and Fogh, 1985; Giavazzi et al., 1986; Ortaldo et al., 1986; Fidler, 1990; Kedar et al., 1990; Gazit et al., 1992) and in severe combined immunedeficient (SCID) mice (Mueller and Reisfeld, 1990; Waller et al., 1990; Mule et al., 1991; Schmidt-Wolf et at., 1991), in which the function of autologous effector cell populations can also be tested (Crowley et al., 1992). Although these xenogeneic systems may not permit optimal interaction between the host and the grafted cells due to differences in species-specific adhesion molecules, homing receptors, and extracellular matrix proteins (Albeda and Buck, 1990; Michl et al., 1991; Van Seventer et al., 1991), these models can have advantages over the test tube experiments. Many experimentalists (and clinicians) use the reduction in tumor size or in number of metastases as parameters for assessing the efficacy of immunotherapy. Although these are indicative of antitumor effects, critical end-point evaluation of treatment benefit should be based on prolongation of survival and lack of disease progression (Osband and Ross, 1990). Despite all the efforts involved in creating “relevant” animal models, even these may not predict the value of therapeutic manipulations in patients, because (1) the response to, and the toxicity of, various BRMs (e.g., TNFa) in animals and humans may be different (Rosenberg, 1991a), and (2) the patient groups may be immunobiologically heterogeneous. Extrapolation from the preclinical models to the clinical setup can be facilitated by better characterization and grouping of the patients, and by identification of prognostic factors predictive of response to immunotherapy (reviewed in Osband and Ross, 1990; Parkinson, 1990).

2 84

ELI KEDAR AND EVA KLEIN

B. SELECTION OF PATIENTS Apart from general parameters, such as the physical condition of the patient and the type of tumor and disease stage, specific, reliable criteria for selection of candidates for immunotherapy do not exist. Discovery of meaningful predictive parameters is therefore an important goal. Patients in general good physical condition (e.g., with a Karnofsky performance status of 270%) are expected to withstand the toxic effects accompanying most of the combined immunotherapeutic regimens. Patients with severe leukopenia, severe cardiovascular, respiratory, renal, or liver disorders, neurological disorders, or brain metastases, cannot receive certain types of treatment (e.g., high-dose IL2 or IFN). With age, both the capacity for immune response and the tolerance for chemotherapy decline. When priority must be exercised, patients presenting with a small tumor burden, o r those whose large tumor load can be decreased by conventional measures, are to be selected. Accordingly, patients with metastatic renal cell carcinoma likely to respond to I L 2 therapy are those in risk group 1, whose primary tumor had been removed by nephrectomy, and who have only pulmonary metastases and not a bulky disease (reviewed in Parkinson, 1990).A considerably higher response rate to various forms of immunotherapy and longer remissions have been obtained in melanoma patients with subcutaneous nodules than in patients with visceral metastases. Since partial responses and long-lasting disease stabilization have also been achieved with immunotherapy in some patients with advanced bulky disease, protocols for these patients should be designed according to their special needs and evaluated separately. Even for the assumed immunogenic tumors, melanoma and renal cell carcinoma, no specific parameters are available that can predict the response to immunotherapy. Demonstration of a specific antitumor T cell response may be valuable. This can be tested in (1) cytotoxicity assays [autologous lymphocyte cytotoxicity (ALC)],using ex vivo tumor cells and blood- or lymph node-derived lymphocytes, (2) proliferation assays [autologous tumor stimulation (ATS)](see Section II,A), and (3) a skin test (i.e., DTH response) with autologous tumor cells or extracts. In view of their importance for recognition by T cells, the level of MHC class 1/11 antigen expression on the tumor cells can also be of predictive value. These assays cannot be performed regularly, however, due to the requirement for tumor cells in sufficient quantity and good quality. Furthermore, because of the heterogeneity of the tumor cell population, the results may not cover reactivities to all tumor cells. As mentioned before (Section II,A), a relatively favorable prognosis

CANCER IMMUNOTHERAPY

285

was associated with the autotumor cytotoxicity (ALC), tested at the time of surgery, in patients with squamous cell- and adenocarcinoma of the lung o r with malignant mesenchymal tumors without apparent metastases (Vanky et al., 1983a,b, 1986, 1987; Uchida and Mizutani, 1989; Uchida et al., 1990, 1991; Ortaldo and Wiltrout, 1990). There was no correlation between the autotumor cytotoxicity and other in vitro immune functions of blood lymphocytes, including N K activity, proliferation to autologous fresh tumor cells, to mitogens, and to allogeneic leukocytes, and production of cytokines (Uchida et al., 1990). In order to establish whether the cytotoxicity of blood lymphocytes against autologous tumor cells could be used for patient selection, it would be important to correlate the results of this test with the immunotherapeutic effects. Coexpression of high levels of MHC class I antigens and the adhesion molecules ICAM-1 and certain VLA antigens on the tumor cells (Vanky and Klein, 1989, 1991; Anichini et al., 1990; Parmiani et al., 1990; RuizCabello et al., 1991) may also be indicative for the possible response to immunotherapy. Patients with particular HLA haplotypes may be more likely to respond, as recently demonstrated for melanoma patients treated with a tumor vaccine (Mitchell, 1990, 1991) or with IL-2 based immunotherapy (Scheibenbogen et al., 1991) (see Section 11,A). It would be of interest, therefore, to determine whether such a correlation exists in patients treated with other immunotherapy modalities. Usually cancer patients do not have impaired N K and LAK activities, and these functions do not correlate with the clinical response in IL-2treated patients (Boldt et al., 1988; Eberlein et al., 1989; Ghosh et al., 1989; Favrot et al., 1990; Dillman et al., 1991b). We tested a group of patients with metastatic melanoma who received chemotherapy (dacarbazine) and IL-2 in alternating cycles for the composition of the lymphocyte populations in the blood, mitogenic responsiveness, IL- 1, IL-2, TNFa, and soluble I L 2 receptor levels in the serum, and N K and LAK cell activity, before and during treatment (Isacson et al., 1992). None of these had any prognostic value. The only parameter that correlated with the clinical response was the greater increase (relative to pretreatment level) in the proportion of lymphocyte populations expressing the CY chain of the IL-2 receptor (CD25) in the responders, 2 days after the IL-2 infusion. T h e number of patients (CR + PR = 4/18) was small, however. A similar observation was made in patients with renal cell carcinoma who responded to IL-2 and indomethacin (Banerjee et al., 1991). In another group of patients with renal carcinoma, the clinical response to IL-2-based immunotherapy was correlated with the I L 1 and

286

ELI KEDAR AND EVA KLEIN

TNFa serum levels and the capillary leak syndrome (Blay et ul., 1991). It should be noted that in all these studies, the differences in the nonspecific immune parameters between responders and nonresponders were seen only posttreatment, usually shortly after the first one to three cycles of IL-2, but not after additional cycles. Differences in these and other nonspecific parameters before treatment were not observed. It is thus highly desirable to identify specific immunological parameters that can be used for patient selection. C. TUMOR DEBULKING In patients with a large tumor burden, immunotherapy must be applied after surgery and/or chemo/radiotherapy. There is no guideline, however, for the optimal schedules. It is not known whether (1) immunotherapy should be given after the completion of cytoreductive therapy, or (2) chemo/radiotherapy and immunotherapy should be applied in several alternating cycles. Obviously, if the cytoreductive treatments are immunosuppressive, immunohematopoietic recovery is essential. In addition to blood counts, composition of the lymphocyte populations, proliferative responses to allogeneic cells and to mitogens, skin tests with common test antigens [e.g., Candida, purified protein derivative (PPD), dinitrochlorobenzene (DNCB)], antibody response to recall antigens (e.g., Tetanus toxoid), and phagocytic cell functions can be performed in order to decide the timing of immunotherapy. Testing for the presence of suppressive factors (e.g., TGFP and others) (Itoh et al., 1985; Ebert et al., 1990; Hirte and Clark, 1990; Reynolds et al., 1990) in the serum can also be informative. For efficient tumor reduction, high-dose (lethal) chemotherapy may be administered followed by autologous bone marrow or derived blood stem-cell transplantation (Dicke et al., 1989; Antman, 1991; Henon et al., 1991; Kessinger and Armitage, 1991; Spitzer et al., 1991). However, since this procedure is burdened with risks (with a mortality rate of approximately lo%), it can be justified only in patients with a poor prognosis whose tumors have a good response to chemotherapy. A supportive treatment with CSFs and other hematopoietic-stimulating cytokines (e.g., I L l , IL4,ILS)could alleviate myelotoxicity and potentiate the therapeutic effects (Demetri and Griffin, 1990; Ganser et al., 1990; Gianni et al., 1990; Glaspy and Golde, 1990; Golde, 1990; Gutterman et al., 1990; Monroy et al., 1990; Antman, 1991; Demetri et al., 1991; Grem et al., 1991; Luikart et al., 1991; Moore, 1991; Nemunaitis et al., 1991; Postmus et al., 1991; Shank and Balducci, 1991). A promising additive to the arsenal of hematopoietic factors is the

CANCER IMMUNOTHERAPY

287

recently discovered stem cell factor (also known as c-kit ligand, or mast cell growth factor), which appears to act on primitive hematopoietic cells, particularly in combination with other cytokines (de Vries et al., 1991; Scadden et al., 1991). D. ACTIVESPECIFIC IMMUNOTHERAPY

Immunization with various tumor vaccines has been performed in patients with several types of cancer (Section 111,A). Active immunization is expected to stimulate the tumor-reactive T cell population as well as the nonspecific effector cells. These populations can be enlarged and further activated by administration of cytokines. In patients treated with high-dose chemotherapy, an interval of several weeks to months may be needed before active immunization (Ahlert et al., 1990). In contrast, lowdose chemotherapy given prior to active immunization may improve vaccine efficacy, as discussed earlier. Several measures are known that can elevate or impose immunogenicity on the tumor cells. We mentioned earlier manipulations that could augment immunogenicity (see Section 111,A). Further possibilities that can elicit or amplify the response have been tested in experimental systems, also with tumor antigens; some of these have not yet been applied in patients: 1. Antigens encapsulated in liposomes (Gregoriadis, 1990; Harding et al., 1991): In patients with stage 111 melanoma, immunization with autologous tumor material incorporated in liposomes led to complete or partial tumor regressions in 5 of 13 patients (Phillips et al., 1990). 2. Administration of the immunogen together with cytokines, such as IL-1 (McCune and Marquis, 1990), IL2 (Naito et al., 1988; Freedman et al., 1990; Thiele et al., 1990), liposomal IL-2 (Sencer et al., 1991), IL6 (Naito et al., 1991), IFNy (Giovarelli et al., 1986), or IL-2 plus IFNol (Arroyo et al., 1990). 3 . Administration of the immunogen together with improved adjuvants [e.g., detoxified endotoxin (DETOX)] (Mitchell et al., 1990; Mitchell, 1991) or with BCG (Hoover and Hanna, 1989; Berd et al., 1990a; Russel et al., 1990; Morton et al., 1991a): Other BRMs that may potentiate the effect of specific immunization are muramyl di/tripeptide (Kleinerman et al., 1989), bestatin (Inoue et al., 1990; Sawada et al., 1990), bryostatin (Schuchter et al., 1991), levamisole (Moertel et al., 1990), OK-432 (Watanabe and Iwa, 1987; Nakagami et al., 1990), PSK (Koike et al., 1990; Matsushima et al., 1990; Mitomi and Noto, 1990; Mitomi et al., 1990; Torisu et al., 1990), and AS101 (Sredni et al., 1987, 1990, 1991; Tichler et al., 1990; Kalechman et al., 1991).

288

ELI KEDAR AND EVA KLEIN

4. Oncogene products: Some of the oncoproteins, such as Ad E l and HER-2/neu, were shown to elicit T cell responses in vitro and in vivo (Bernards et al., 1987; Hellstrom and Hellstrom, 1989; Kast et al., 1989; Fendly et al., 1990; Melief and Kast, 1990a,b; Talarico et al., 1990; Jung and Schluesener, 1991). Many tumors carry mutated p53 and p21 (ras) protooncogenes (Bos, 1989; Bishop, 1991), which may also evoke an antitumor response. In a recent study, T cells specific for the p21 ras protein were detected in mice immunized with synthetic peptides corresponding to the mutated regions of activated p21 ras proteins (Peace et al., 1991). 5 . Genetically engineered tumor cells: Immunogenicity of the tumor cells may be potentiated by introduction of cytokine-encoding genes with the help of retroviral vectors (reviewed in Kinnon and Levinsky, 1990; Russell, 1990; Blankenstein et al., 1991; Rosenberg, 1991a). The tumor cells can then release cytokines that can act on an autocrine or paracrine basis. In this way the tumor cells can induce “local help” for amplification of an immune response (Fearon et al., 1990). Mice grafted with weakly immunogenic tumor cells transfected with the genes encoding IL-1 (Blease, 1990; Douvdevani et al., 1992), IL-2 (Gansbacher et al., 1990b; Fearon et al., 1990; Ley etal., 1991), IL-4 (Tepper et al., 1989; Li et al., 1990), IFNy (Watanabe et al., 1989; Gansbacher et al., 1990a), TNFa (Asher et al., 1991; Blankenstein et al., 1991b; Vanhaesebroeck et al., 1991), o r G-CSF (Colombo et al., 1991) rejected the tumor and developed immunity against a subsequent challenge with the nontransduced tumor cells. T h e regression of such tumor cells correlated with the amounts of cytokine they produced. In a recent study, even an established wild-type tumor was rejected following administration of the transduced tumor cells (Golumbek et al., 1991). When used for vaccination, care must be taken, because such tumor cells may acquire a higher growth or metastatic potential (Malik et al., 1990; Blankenstein et al., 1991a; see also Section 111,B). 6. Tumor cells transfected with exogenous DNAs specifying foreign antigens (e.g., viral antigens, MHC antigens, tumor antigens) (reviewed in Nowak et al., 1991). Such tumor cells demonstrated increased immunogenicity in animals. 7. Recombinant viruses (e.g., vaccinia virus constructs) that express human tumor antigens (Estin et al., 1988; Kahn et al., 1991).

Thus, new approaches for cancer vaccines have been developed, but the majority of these have not yet been tested in patients. Active immunotherapy can be followed by adoptive measures. The population of tumor-reactive T cells generated by active immunization may be first en-

CANCER IMMUNOTHERAPY

289

larged in vitro and readministered to the patient. In a presently conducted clinical trial at the National Institutes of Health (Bethesda, MD), viable, autologous melanoma cells transduced with the TNFa gene are injected subcutaneously and intradermally, in an attempt to enlarge the number of tumor specific T cells in the regional lymph nodes. The latter are removed 2-3 weeks later, and the lymphocytes expanded in culture with IL2 for readministration (R Rosenberg, personal communication).

E. ELIMINATION OF SUPPRESSOR CELLS/FACTORS Although tumor reduction may decrease the levels of suppressor cells and/or soluble factors (see Section III,D), additional measures could be applied prior to or during immunotherapy. Inhibitors of the cyclooxygenase pathway (indomethacin, ibuprofen, aspirin) (Kedar et al., 1984a,b, 1986; Israel et al., 1990; Khoo et al., 1990; Lala et al., 1990), the histamine H2 receptor blocker, cimetidine (Burtin et al., 1988), low-dose cyclophosphamide or doxorubicin (Berd and Mastrangelo, 1988b; Mitchell, 1988; North et al., 1989), and monoclonal antibodies directed to T suppressor cells (North et al., 1989) had beneficial therapeutic effects in rodents. However, their therapeutic value in patients is still not substantiated (see Section 111,D). It is possible that administration of monoclonal antibodies reactive with immunosuppressive cytokines (e.g., TGFP) (Sporn and Roberts, 1990; Wahl et al., 1990; Tada et al., 1991), which are secreted in large quantities by certain cancer cells, or removal of such factors by plasmapheresis may be beneficial.

F. ACTIVE/ADOPTIVE IMMUNOTHERAPY I . Cpokines After the reduction of tumor burden (and putative suppressor cells), and following active specific immunization (if possible), further potentiation of effector mechanisms may be achieved with cytokines, singly o r combined (see Section 111,B).Based largely on the results in experimental systems, the most effective combinations tested thus far are IL-2 and IFNa (Cameron et al., 1988; Rosenberg et al., 1989b,c; Kedar et al., 1990, 1992) and IL-2, IFNa and TNFa (McIntosh et al., 1989). I n recent clinical trials, tumor regressions were also achieved with IL-1 in melanoma (Smith et al., 1991a; Starnes et al., 1991), and with I L 4 in lymphomas (Davis et al., 1991). Other cytokines that may prove effective in patients are IFNP, IL-6, and IL7. Another novel approach to potentiate cytokine activity is the use of

290

ELI KEDAR A N D EVA KLEIN

hybrid or chimeric molecules, composed of parts of several cytokines, such as the consensus IFNa and the GM-CSF/ILS (PIXY 321) molecules (Curtis et al., 1991; Johnston, 1991). Combination of IL-2 (or other cytokines) and the monoclonal antiCD3 reagent can activate both T and LAK cells. This approach has been successful in animal models, and clinical trials have begun recently (Anderson et al., 1988, 1989; Yun et al., 1989; Ellenhorn et al., 1990; Gallinger et al., 1990; Lafreniere et al., 1990; Schoof et al., 1990; Stohl et al., 1990; S. C. Yang et al., 1990; Gambacorti-Passeriniet al., 1991; Loeffler et al., 1991b; Ochoa et al., 1991; Weil-Hillman et al., 1991). A novel approach for damaging tumor cells that express high levels of cytokine receptors is the use of relevant cytokine molecules conjugated to cytotoxic agents (Strom et al., 1990; Pastan and FitzGerald, 1992). A recombinant fusion protein, made by replacement of the diphtheria toxin gene receptor-binding domain with the gene for human IL2, was found effective in patients with hematologic malignancies that express the IL2 receptor (LeMaistre et al., 1991a,b). The cytokines IL4,IL-6, and TGFa fused with the Pseudomonas exotoxin exhibited impressive antitumor effects in mice (Pai et al., 1991; Puri et al., 1991; Siegall et al., 1991). Patients with localized tumors (e.g., bladder carcinoma, mesothelioma, abdominal carcinomatosis, head/neck, and brain tumors) could benefit from locoregional administration, concurrent with systemic treatment with cytokines. The combined 1ocaUsystemic regimen has not yet been practiced in clinical trials. The therapeutic effects may be improved and toxicity may be reduced by delivery of the cytokines in liposomes or by their chemical modification (see Section 111,B). 2 . Cells The clinical experience of the past 8 years with several thousands of patients does not support the use of LAK cells (see Section 111,C). A possibility yet to be explored is the targeted cellular cytotoxicity. Lymphokine-activated killer cells (or T cells), with or without IL2, could be administered together with monoclonal antitumor antibodies active in ADCC, or with bispecific, heteroconjugate, or hybrid antibodies directed against molecules on the plasma membrane of effector cells (such as CD2, CD3, CD16, CD28, CD59) and the tumor cells. Such antibodies synergized in antitumor effects with adoptively transferred cells in animal models, and enhanced the cell-mediated lysis of human tumor cells in vitro (Titus et al., 1987; Eisenthal et al., 1988; Donohue et al., 1990; Fanger et al., 1990, 1991; Hank et al., 1990a; Kerr et al., 1990; Segal et al., 1990; Goldenberg, 1991; Nitta et al., 1991; Reid et al., 1991a,b). Encour-

CANCER IMMUNOTHERAPY

29 1

aging results in glioma patients were reported with bispecific (antiCD3/anti-glioma) antibody-coated LAK cells (Nitta et al., 1990). Such cell-antibody combinations may be much more efficient when administered together with IFN or TNF, because these can enhance the expression of tumor antigens and adhesion molecules on the tumor cells, thereby potentiating the interaction between the effector and tumor cells (Murray et al., 1990).The use of “humanized”or human MAb instead of murine MAb should improve these possibilities. The results with TIL do not appear to be superior to those with LAK cells in clinical trials (see Section 111,C). Moreover, TIL-based immunotherapy can be carried out only in a limited number of patients. Tumor cell-selectivesyngeneic T cell clones gave excellent results in animal models, and they could exert the most efficient antitumor effects. However, even if such human T cells were selected in culture, they could rarely proliferate to reach sufficient numbers for readministration. The aims are usually to develop CD8 cytotoxic cells, but it is possible that cytokine-producing CD4 clones may give better therapeutic effects (Greenberg, 1991). Attention should be directed to recent methodological improvements that may render this approach feasible. These include: (1) selection of T cell populations from the tumor tissue by using anti-T cell antibodies bound to beads (Morecki et al., 1990;J. C. Yang et al., 1990a),(2) repeated in uitro stimulation of the T cells with irradiated tumor cells and lowdose IL2 (McKinnon et al., 1990; Skornick et al., 1990), or by monoclonal anti-CD3 antibodies (Nijhuis et al., 1990; Yoshizawa et al., 1991), and (3) use of several cytokines (e.g., IL-2 and IL-4, IL-2 and TNFa, or IL2, and TNFa and IFNa or IFNy), with and without anti-CD3, instead of IL-2 alone in the T cell cultures (Finke et al., 1991;Jadus et al., 1991; Shimizu et al., 1991). Expansion of adoptively transferred tumor-specific T cells was obtained in mice by repeated administrations of the tumor antigen together with low-dose IL-2 (Chen et al., 1990). It is possible that such a manipulation can be effective in patients. Propagation of mouse T cells for an extended period of time, with or without I L 2 , and without the need for repeated exposure to the antigen, was recently achieved by introduction of the protein kinase C y gene, using a retroviral vector (Finn et al., 1991). Such transduced CTL clones maintained their specific cytotoxic activity in uitro and antitumor effect in uiuo without being tumorigenic (Chen et al., 1991). The therapeutic effects of T cells transduced with cytokine-encoding genes are presently being tested, based on the assumption that they can specifically accumulate and release large amounts of cytokines at tumor sites (Rosenberg et al., 1990; Culver et al., 1991a) (see Section 111,C).

292

ELI KEDAR A N D EVA KLEIN

Melanoma patients are now being treated with autologous TIL transduced with the TNFa gene (Rosenberg, 1991a,b). The value of this approach has not yet been substantiated, however, in animal models. In mice, IL1 gene-carrying CD8 CTL had enhanced tumor cytotoxicity in vitro, but the therapeutic effects were not superior, compared with the unmanipulated T cell population. In other experiments, TNFa genecarrying T cells lacked antitumor activity in uiuo (Culver et al., 1991b; Fox et al., 1991). T h e failure may be ascribed to a low level of cytokine secretion o r poor localization of the cells in tumor sites. Thus, this approach needs to be tested further in animal models for efficacy and toxicity before it can be applied to patients. V. Conclusions

T h e demonstration of antigenicity of experimental tumors and the beneficial effects of immunotherapy in animal models generated great expectations for cancer treatment. Various immunotherapy protocols have been reported to be moderately effective in patients with certain types of cancers. T h e value of many of these trials cannot be judged, however, because though they could stand the scrutinies of evaluation, they were not repeatable by other groups. This may be attributed to differences in patient selection and in the details of treatment protocols. T h e question can be posed whether at the present state of knowledge cancer immunotherapy can be improved. Some recommendations can be proposed. T h e experimentalists may focus on models relevant to the clinical situation, such as metastatic tumors of low or no immunogenicity, and exploit the possibility of studying the interaction of human tumor and effector cells in immunologically deficient mice. The message for the clinicians is that immunotherapy, generally, is not effective in patients with high tumor load. Therefore, it can only be an adjunct to traditional cytoreductive treatments. Tests for new treatment regimens in phase 1/11 trials ought to be performed in patients that have low tumor burden, but poor prognosis, rather than in end-stage patients with bulky disease. Imposing experimental immunotherapy regimens on such patients is hard to justify, however. New regimens should be tested first in well-designed experimental models. Obviously, priority should be given to protocols without major toxicity and with the possibility for application to large patient groups. New modalities may first be tested in melanoma and renal cell carcinoma patients, because these have been already shown to be capable of response. Since only a low proportion (<20%) of patients have been shown to benefit from immunotherapy, it would be important to identify mean-

CANCER IMMUNOTHERAPY

293

ingful parameters predictive of response. Assuming that specific T cells are crucial for the antitumor response, it would be desirable to test the patients for the existence of these lymphocytes. Technically, this is not possible in cases, however. Several studies have shown that the nonspecific cellular effector functions have no predictive value. Throughout this article we emphasized the potential of the combined treatment modalities. These include tumor debulking, followed by active immunization with tumor material, and by administration of cytokines, other BRMs, and antitumor antibodies in various combinations. In special cases, readministration of autologous, selected T cell populations may add to this treatment arsenal. Considering the multitude of variables in such strategies, the planning of clinical protocols is difficult. Therefore, only general guidelines are available for a rational design of the type of combinations and timing schedules. Obviously, they can be a burden on patients and clinicians alike. It appears that the immunotherapeutic measures must be applied repeatedly over an extended period of time. Monitoring the general immunocompetence and, if possible, the “specific”immune parameters as well, can provide guidelines for the continued application. The new ideas for immunotherapy propose readministration of autologous tumor-reactive T cells transfected with cytokine-encoding genes, and the use of encapsulated cytokines directed by antibodies (immunoliposomes) to the tumor cells. These may home in on the tumor tissue and deliver large quantities of cytokines. The therapeutic efficacy of these two modalities is not yet known, but the latter seems to be more practical. Measures that impose or increase the immunogenicity of the tumor cells can be applied in vitro, or, depending on the localization of the tumor, even in viuo. It is possible that the manipulated cells administered as vaccine can elicit an immune response that may act on the resident unmodified cells as well. We cannot escape the conclusion that these approaches may not change the prospects of cancer immunotherapy significantly, and that dramatic improvements of generally applicable immunotherapy cannot be expected without conceptual and technical innovations. ACKNOWLEDGMENTS E. Kedar is supported by grants from the US.-Israel Binational Foundation (BSF 8600267); the Israel Cancer Association; the National Council for Research and Development, Israel and the DKFZ, Heidelberg (CA 43/724); Concern Foundation, Inc., and Concern 11, Los Angeles, California; the Harold €3. Abramson Memorial Fund; the Society

294

ELI KEDAR AND EVA KLEIN

of Research Associates of the Lautenberg Center; and the Lautenberg Endowment Fund of the American Friends of the Hebrew University. E. Klein is supported by PHS Grant No. 5 RO1 Ca 25250-12, National Cancer Institue, DHHS, and the Swedish Cancer Society. We are indebted to Drs. R. Isacson, R. W. Baldwin, R. Catane, P. Sondel, G. Parmiani, M. Mitchell, V. Schirrmacher, Y. Barenholz, and F. Wnky for advice and suggestions.

REFERENCES Abrams, J. S., Rayner, A. A,, Wiernik, P. H., Parkinson, D. R., Eisenberger, M., Aronson, F. R., Gucalp, R., Atkins, M. B., and Hawkins, M. J. (1990).J. Natl. Cancer Inst. 82, 1202- 1206. Ackerstein, A., Kedar, E., and Slavin, S. (1991). Blood 78, 1212-1215. Adler, A., Stein, J. A., Kedar, E., Naor, D., and Weiss, D. W. (1984).J. Biol. RespomeModif 3, 49 1-500. Aebersold, P., Hyatt, C., Johnson, S., Hines, K., Korcak, L., Sanders, M., Lotze, M., Topalian, S., Yang, J., and Rosenberg, S. A. (1991).J. Natl. Cancer Inst. 83, 932-937. Agah, R., Malloy, B., Sherrod, A., and Mazumder, A. (1988). Cancer Res. 48,2245-2248. Aglietta, M., De Felice, L., Stacchini, A., Sanavio, F., Severino, A., Simone, F., Piacibello, W., and Mandelli, F. (1990). Leuk. Lymphoma 2, 207-214. Ahlert, T., Kaufmann, M., Bastret, G., and Schirrmacher, V. (1990).J. Cancer Res. Clin. Oncol. 116, Suppl. 1 , 80 (Abstr. B 1.019.03). Akaishi, S . , Kobari, M., and Matsuno, S. (199O).J.CancerRes. Clin. Oncol. 116, Suppl.l,604 (Abstr. A5.221.09). Albeda, S. M., and Buck, C. A. (1991).f. Immunoder. 10,389-397. Albertini, M. R., Sosman, J. A., Hank, J. A., Moore, K. H., Borchert, A,, Schell, K., Kohler, P. C., Bechhofer, R., Storer, B., and Sondel, P. M. (1990). Cancer (Philadelphia) 66, 2457-2464. Alexander, R. B., and Rosenberg, S. A. (1990). Proc. AACR 31, 264 (Abstr. 1566). Allavena, P., Lo Presti, P., Di Bello, M., Lucchini, V., Lissoni, A., Zanetta, G., Mangioni, C., and Mantovani, A. (1988). Cancer Immonol. Immunother. 27, 69-76. Allavena, P., Pirovano, P., Bonazzi, C., Colombo, N., Mantovani, A,, and DIncalci, M. (1 990). J. Natl. Cancer Inst. 82, 139- 142. Anderson, P. M., Bach, F. H., and Ochoa, C. (1988). Cancer Immunol. Immunother. 27, 8288. Anderson, P. M., Blazar, B. R., Bach, F. H., and Ochoa, A. C. (1989).J. Immunol. 142, 1383-1394. Anderson, P. M., Katsanis, E., Leonard, A. S., Schow, D., Loeffler, C. M., Goldstein, M. B., and Ochoa, A. C. (1990). Cancer Res. 50, 1853-1856. Andreesen, R., Scheibenbogen, C., Brugger, W., Krause, S., Meerpohl, H. G., Leser, H. G., Engler, H., and Lohr, G. W. (1990). CancerRes. 50, 7450-7456. Anichini, A. (1989). Cum. Opzn. Immunol. I, 917-921. Anichini, A,, Mazzocchi, A., Fossati, G., and Parmiani, G. (1989).J. Immunol. 142, 36923701. Anichini, A., Mortarini, R.,Supino, R., and Parmiani, G. (1990).Int. J. Cancer46,508-515. Antman, K. H. (1991). Cancer Res. (Suppl.) 51, 5060-5064. Aoki, T., Kikuchi, H., Miyatake, S. I., Oda, Y., Iwasaki, K., Yamasaki, T., Kinashi, T., and Honjo, T. (1989).J. Exp. Med. 170, 583-588.

CANCER IMMUNOTHERAPY

295

Arroyo, P. J., Bash, J. A., and Wallack, M. K. (1990). Cancer Immunol. Immunothr. 31, 305311. Asher, A. L., Mule, J. J., and Rosenberg, S. A. (1989). Cancer Immunol. Immunother. 28, 153156. Asher, A. L., Mule, J. J., Kasid, A., Restifo, N. P., Salo, J. C., Reichert, C. M., Jaffe, G., Fendly, B., Kriegler, M., and Rosenberg, S. A. (1991). J. Immunol. 146, 3227-3234. Atkins, M. B., Mier, J. W., Parkinson, D. R., Could, J. A., Berkman, E. M., and Kaplan, M. M. (1988). N . Engl.J. Med. 318, 1557-1563. Atkins, M., Demchak, P., Mier, J., Robert, N., and Sznol, M. (1990). Proc. Am. Soc. Clin. Oncol. 9, 186 (Abstr. 719). Atkins, M. B., Sparano, J., Fisher, R. I., Weiss, G. R., Margolin, K. A., Aronson, F. R., and Sznol, M. (1991). Proc. Am. SOC.Clin. Oncol. 10, 166 (Abstr. 526). Atzpodien, J., Korfer, A., Franks, C. R., Poliwoda, H., and Kirchner, H. (1990a). Lancet 335, 1509-1512. Atzpodien, J., Korfer, A., Franks, C. R., Loriaux, E., Palmer, P. A,, Gessner, S., Dallmann, I., Goldmann, U., Benter, T., Poliwoda, H., and Kirchner, H. (I990b). J. Cancer Res. Clin. Oncol. 116, Suppl. 1, 333 (Abstr. A3.112.20) Avner, B. P., Avner, B., Kalman, L., and Thomthwaite, J. T. (1990). Proc. Am. Assoc. Cancer Res. 31, 272 (Abstr. 1613). Awwad, M., and North, R. J. (1988a). Immunology 65, 87-92. Awwad, M., and North, R. J. (1988b). Cancer Immunol. Immunother. 26, 55-60. Awwad, M., and North, R. J. (1989). Cancer Res. 49, 1649-1654. Bajorin, D. F., Chapman, P. B., Wong, G., Coit, D. G., Kunicka, J., Dimaggio, J., CordonCardo, C., Urmacher, C., Dantes, L., Templeton, M. A,, Liu, J., Oettgen, H. F., and Houghton, A. N. (1990). Cancer Res. 50,7490-7495. Baker, P. D., and Evans, C. H. (1990). Proc. UCLA Symp. Mol. Pathways Cytokine Action, Park City Abstr. CD 300. Baldwin, R. W., and Byers, V. S. (1989). Curr. Opin. Immunol. 1, 891-894. Balkwill, F. (1988). Cum. Opin. Immunol. 1, 241-249. Balkwill, F. R. (1989). Lantet 1, 1060-1063. Balkwill, F. R., and Burke, F. (1989). Immunol. Today 10, 299-304. Balkwill, F. R., Naylor, M. S., and Malik, S. (1990). Eur. J . Cancer 26, 641-644. Banerjee, D., Mertens, W., Bramwell, V., and Lala, P. K. (1991).Proc. Am. Assoc. Cancer Res. 32, 247 (Abstr. 1471). Bani, M. R., Garofalo, A,, Scanziani, E., and Giavazzi, R. (1991). J. Natl. Cancer Inst. 83, 119-123. Bankert, R. B., Yokota, S., Ghosh, S. K., Mayhew, E., and You, J. H. (1989). CancerRes. 49, 301-308. Bar, M. H., Sznol, M., Atkins, M. B., Ciobanu, N., Micetich, K. C., Boldt, D. H., Margolin, K. A., Aronson, F. R., Rayner, A. A., Hawkins, M. J., Mier, J. W., Paietta, E,. Fisher, R. I., Weiss, G. R., and Doroshow, J. H. (199O).J. Clin. Oncol. 8, 1138-1147. Barth, R. J., Mule, J. J., Spiess, P. J., and Rosenberg, S. A. (1991).J. Exp. Med. 173,647-658. Bash, J. A., Arroyo, P. J., and Wallack, M. K. (1990). FASEBJ. 4, A1718 (Abstr. 135). Bauer, R., Kopald, K., Lee, J., Wong, J., and Morton, D. (1990). Proc. Am. Sac. Clin. Oncol. 9, 276 (Abstr. 1068). Belardelli, F., Ciolli, V., Testa, U., Montesoro, E., Bulgarini, D., Proietti, E., Borghi, P., Sestili, P., Locardi, C., Peschle, C., and Gresser, 1. (1989).Int. J. Cancer44, 1108-11 16. Berd, D., and Mastrangelo, M. J. (1988a). Cancer Invest. 6, 337-349. Berd, D., and Mastrangelo, M. J. (1988b). Cancer Res. 48, 1671-1675. Berd, D., Maguire, H. C., and Mastrangelo, M. J. (1984). Cancer Res. 44,5439-5443.

296

ELI KEDAR AND EVA KLEIN

Berd, D., Maguire, H. C., McCue, P., and Mastrangelo, M. J. (1990a). J. Clin. Oncol. 8, 1858-1867. Berd, D., Maguire, H. C., Mastrangelo, M. J., and Murphy, G. (1990b). Proc. Am. Assoc. Cancer Re$.31, 279 (Abstr. 1654). Berek, J. S. (199O).J. Clin. Oncol. 8, 1610-1612. Bergmann, L. (1989). Eur.J. Cancer Clin. Oncol. 25, Suppl. 3, S31-S36. Berke, G. (1989). In “Fundamental Immunology” (W. E. Paul, ed.), 2nd ed., pp. 735-764. Raven Press, New York. Bernards, R., Destree, A,, McKenzie, S., Gordon, E., Weinberg, R. A,, and Panicali, D. (1987). Proc. Natl. Acad. Sci. U.S.A. 84, 6854-6858. Bezwoda, W. R., and Dansey, R. (199O).J. CancerRes. Clin. Oncol. 116, Supl.11, 1121 (Abstr. 20.19.04). Bhattacharya-Chatterjee, M., and Kohler, H. (1989). Zn “Imrnunobiology of Proteins and Peptides” (M. Z. Attasi, ed.), Vol. 5, pp.113-127. Plenum, NewYork. Billard, C., and Wietzerbin, J. (1990). Eur. J. Cancer 26, 67-69. Bishop, J. M. (1991). Cell (Cambridge, Mass.) 64, 235-248. Blair, S., Flaherty, L., Valdivieso, M., and Redman, B. (1991). Proc. Am. SOC.Clin. Oncol. 10, 294 (Abstr. 1031). Blankenstein, T., Rowely, D. A,, and Schreiber, H. (1991a). Curr. Opin. Immunol. 3, 694698. Blankenstein, T., Qin, Z., Uberla, K., Muller, W., Rosen, H., Volk, H. D., and Diamantstein, T. (1991b).J. Exp. Med. 173, 1047-1062. Blay, J. Y., Combaret, V., NCgrier, S., Franks, C. R., Mercatello, A,, Philip, T., and Favrot, M. (1991). Proc. Am. SOC.Clin. Oncol. 10, 217 (Abstr. 732). Blease, R. M. (1990). Proc. SOC.Biol. Ther. 9 (Abstr. 18). Blease, R. M. (1991). Clin. Immunol. Immunopath. 61, S47-S55. Boccoli, G., Masciulli, R., Ruggeri, E. M., Carlini, P., Giannella, G., Montesoro, E,. Mastroberardino, G., Isacchi, G., Testa, U., Calabresi, F., and Peschle, C. (1990).CancerRes. 50,5795-5800. Bohle, W., Schlag, P., Liebrich, W., Hohenberger, P., Mantasterski, M., Moller, P., and Schirrmacher, V. ( 1990). Cancer (Philadelphia) 66, 15 17- 1523. Boldt, D. H., Mills, B. J., Gemlo, B. T., Holden, H., Mier, J., Paietta, E., McMannis, J. D., Escobedo, L. V., Sniecinski, I., Rayner, A. A., Hawkins, M. J., Atkins, M. B., Ciobanu, N., and Ellis, T. M. (1988). Cancer Res. 48, 4409-4416. Boon, T. (1983). Adv. Cancer Res. 39, 121-151. Boon, T. (1992). Adv. Cancer Res. 58, 177-210. Boon, T., van Pel, A., and d e Plaen, E. (1989). Cancer Cells 1, 25-28. Borden, E. C. (1988).J. Natl. Cancer Inst. 80, 148-149. Borden, E. C., and Sonde], P. M. (1990). Cancer (Philadelphia) 65, 800-814. Borden, E. C., Sidky, Y. A., Hatcher, J. F., and Bryan, G. T. (1988). Cancer Res. 48,23292334. Bos, J. L. (1989). Cancer Res. 49, 4682-4689. Bosco, M., Giovarelli, M., Forni, M., Modesti, A,, Scarpa, S., Masuelli, L., and Forni, G. (199O).J. Immunol. 145, 3136-3143. Boutin, C., Viallet, J. R., Van Zandwijk, N., Douillard, J. T., Paillard, J. C., Guerin, J. C., Mignot, P., Migueres, J, Varlet, F., Jehan, A,, Delepoulle, E., and Brandely, M. (1991). Cancer (Philadelphia) 67,2033-2037. Braakman, E., Goedegebuure, P. S., Vreugdenhil, R. J., Segal, D. M., Shaw, S., and Bolhuis, R. L. H. (1990). Int. J. Cancer 46, 475-480.

CANCER IMMUNOTHERAPY

297

Bradley, E. C., Louie, A. C., Paradise, C. M., Carlin, D. A., Bleyl, K. L., Groves, E. S., a n d Rudolph, A. R. (1987). Proc. Am. Soc. Clin. Oncol. 8, 133 (Abstr. 519). Bramwell, V. H . C., Mertens, W. C., and Lala, P. K. (1993). Proc. Am. Soc. Clin. Oncol. 10, 171 (Abstr. 546). Bree, A., Shaw, G., Veldman, T., Timony, G., Metzger, M., Horgan, P., Garnick, M., and Stoudemire, J. (1991). Proc. Am. Assoc. Cancer Res. 32, 347 (Abstr. 2058). Breier, S., Pensel, R., Roff6, C., Giangiacomo, G., and Traine, G. (1990). Proc. Am. Soc. Cbn. Oncol. 9, 281 (Abstr. 1090). Brenner, D. E. (1989).J. Natl. Cancer Inst. 81, 1436-1438. Brodt, P., Blore, J., Phillips, N. C., Munzer, J. S., and Rioux, J. D. (1989). Cancer Immunol. Immunother. 28, 54-58. Bronchud, M. H . (1990). Eur.J. Cancer 26,928-929. Brunda, M. J.. Bellantoni, D., and Sulich, V. (1987). Int. J. Cancer 40, 365-371. Brunner, C. A,, Gokel, J. M., Riethmuller, G., and Johnson, J. P. (1991). Eur.J. Cancer 27, 411-416. Bubenik, J. (1989). Immunol. Lett. 21, 267-274. Bubenik,J., Voitenok, N. N., Kieler, J., Prassolov, V. S., Chumakov, P. M., Bubenikova, D., Simovd, J., and Jandlova, T. (1988). Immunol. Lett. 19, 279-282. Bubenik,J., Simovl, J., and JandlovA, T. (1990). Immunol. Lett. 23, 287-292. Bukowski, R. M., Murthy, S., Sergi, J., Thomas, B., McKeever, S., Medendorp, S. V., Tubbs, R., Gibson, V., and Finke, J. (1990).J. Biol. Response Mod$ 9, 538-545. Burtin, C., Noirot, C., Scheinmann, P., Galoppin, L., Sabolovic, D., and Bernard, P. (1988). Eur. J . Cancer Clin. Oncol. 24, 16 1- 167. Butturini, A,, and Gale, R. P. (1989). Br. J. Haematol. 72, 479-485. Bystryn, J. C. (1990). Cancer Metastasis Rev. 9, 81-91. Bystryn, J. C., Ortatz, R., Harris, M. N., Roses, D. F., Golomb, F. M., and Speyer, J. L. (1988). Cancer (Philadelphia) 61, 1065-1070. Bystryn, J. C., Ortatz, R.,Roses, D. F., Harris, M. N., Henn, M., and Lew, R. (1990). Proc. Soc. Biol. Ther. 12 (Abstr. 6). Cameron, R. B., McIntosh, J. K., and Rosenberg, S. A. (1988). Cancer Res. 48, 581058 17. Cameron, R. B., Spiess, P. J., and Rosenberg, S. A. (1990). J . Exp. Med. 171, 249-263. Campbell, M. J., Esserman, L., and Levy, R. (1988). J. Immunol. 141, 3227-3233. Campbell, M. J., Esserman, L., Bayars, N. E., Allison, A. C., and Levy, R. (1990).J.Immunol. 145, 1029-1036. Cantley, L. C., Auger, K. R., Carpenter, C., Duckworth, B., Graziani, A,, Kapeller, R.,and Soltoff, S. (1991). Cell (Cambridge, Mass.) 64, 281-302. Cassel, W. A., Murray, D. R., and Phillips, H. (1983). Cance7 (Philadelphza) 52, 856-860. Cassel, W. A., Weidenheim, K. M., Campbell, W. G., and Murray, D. R. (1986). Cancer (Philadelphiu) 57, 1302- 13 12. Catane, R., and Longo, D. L. (1988). Isr. J . Med. Sci. 24, 471-476. Cebon, J. S., Bury, R. W., Lieschke, G. L., and Morstyn, G. (1990).Eur. J. Cancer26, 10641069. Chapman, P. B., Scheinberg, D. A,, DiMaggio, J. J., and Houghton, A. N. (1991).Immunol. Allergy Clan. North Am. 11, 257-275. Chatterjee, M., Biddle, W., Foon, K. A., and Kohler, H. (1990). J. Cancer Kes. Clin. Oncol. 116, Suppl.1, 288 (Abstr. A3.103.63). Cheever, M. A., Thompson, J. A., Kern, D. E., and Greenberg, P. D. (1985).J. Immunol. 134, 3895-3900.

298

ELI KEDAR AND EVA KLEIN

Chen, W., Reese, V. A,, and Cheever, M. A. (1990). J. Immunol. 144, 3659-3666. Chen, W., Schweins, E., Finn, 0.J., and Cheever, M. A. (1991). FASEB J. 5 , A1333 (Abstr. 5579). Chou, T., Bertera, S., Chang, A. E., and Shu, S. (1988).J. Immunol. 141, 1775-1781. ChrCtien, I., P h e , J., Briere, F., De Waal Malefijt, R., Rousset, F,. and De Vries, J. E. (1990). Eur. J. Immunol. 20,243-251. Ciolli, V., Gabriele, L., Sestili, P., Varano, F., Proietti, E., Gresser, I., Testa, U., Montesoro, E., Bulgarini, D., Mariani, G., Peschle, C., and Belardelli, F. (1991). J. Exp. Med. 173, 313-322. Clamon, G., Goutsou, M., Perry, M. C., Lyss, A., and Green, M. R. (1991). Proc. Am. Soc. Clin. Oncol. 10, 267 (Abstr. 925). Clark, D. A., Hirte, H. W., and Buick, R. N. (1988). Br. J. Cancer 58, 415-418. Clark, J. W., Smith, J. W., Steis, R. G., Urba, W. J., Crum, E., Miller, R., McKnight, J., Beman, J., Stevenson, H. C., Creekmore, S., Stewart, M., Conlon, K., Sznol, M., Kremers, P., Cohen, P., and Longo, D. (1990). Cancer Res. 50, 7343-7350. Cochran, A. J., Wen, D. R., Stene, M., Fairhurst, M. V., and Hoon, D. S. (1988). Prog. Clin. Biol. Res. 256, 597-618. Cohen, P. J., Lotze, M. T., Roberts, J. R., Rosenberg, S. A., and Jaffe, E. B. (1987). Am. J. Pathol. 129, 208-216. Colombo, M. P., Ferrari, G., Stoppacciaro, A,, Parenza, M., Rodolfo, M., Mavilio, F., and Parmiani, G. (1991). J. Exp. Med. 173, 889-897. Cramer, D. V., Harnaha, J. B., and Herberman, R. B. (1989). Transplant. Proc. 21, 32753276. Creagan, E. T., Ahmann, D. L., Green, S. J., Long, H. J., Frytak, S., and Itri, L. M. (1985).J. Clin. Oncol. 3, 977-981. Creaven, P. J., Cowens, J. W., Brenner, D. E., Dadey, B. M., Han, T., Huben, R., Karakousis, C., Frost, H., LeSher, D., Hanagan, J., Andrejcio, K., and Cushman, M. K. (1990). J. Biol. Response Mod$ 9,492-498. Crowley, N. J., Slingluff, C. L., Darrow, T. L., and Seigler, H. F. (1990). CancerRes. 50,492498. Crowley N. J., Vervaert, C. E., and Seigler, H. F. (1992). Cancer Res. 52, 394-399. Crum, E. D., and Kaplan, D. R. (1991). CancerRes. 51,875-879. Crump, W. L., Owen-Schaub, L. B., and Grimm, E. A. (1989). Cancer Res. 49, 149-153. Culver, K., Cornetta, K., Morgan, R., Morecki, S., Aebersold, P., Kasid, A., Lotze, M., Rosenberg, S. A., Anderson, W. F., and Blease, R. M. (1991a). Proc. Natl. Acad. Sci. U.S.A. 88, 3155-3159. Culver, K. W., Chiang, Y., Anderson, W. F., and Blease, R. M. (1991b). FASEB J. 5, A1712 (Abstr. 7789). Curtis, B. M., Williams, D. E., Broxmeyer, H. E., Dunn, J., Farrah, T., Jeffery, E., Clevenger, W., DeRoss, P., Martin, U., Friend, D., Craig, V., Gayle, R., Price, V., Cosman, D., March, C. J., and Park, L. S. (1991). Proc. Natl. Acad. Sci. U.S.A. 88,5809-5813. Dal Ri, P., Graffer, U., Bosetti, S., Reich, S., and Menichelli, E. (1990). J. Cancer Res. Clzn. Oncol. 116, Suppl.1, 534 (Abstr. B4.208.03). Damle, N. K., Doyle, L. V., Bender, J. R., and Bradley, E. C. (1987).J. Immunol. 138, 17791785. Davis, I., Maher, D., Cebon, J., Dyer, W., McKendrick, J., Begley, G., Woodruff, R., Green, M., Fox, R., Rallings, M., Bonnem, E., Boyd, A., and Morstyn, G. (1991). Proc. Am. SOC. Clin. Oncol. 10, 287 (Abstr. 1006). Debs, R. J., Duzgunes, N., Brunette, E. N., Frendley, B., Patton, J., and Philip, R. (1989).J. Immunol. 143, 1192-1 197.

CANCER IMMUNOTHERAPY

299

Debs, R.J., Fuchs, H. J., Philip, R., Brunette, E. N., Duzgiines, N., Shellito, J. E., Liggitt, D., and Patton, J. R. (1990). Cancer Res. 50, 375-380. Demetri, G. D., and Griffin, J. D. (1990).J. Clin. Oncol. 8, 761-764. Demetri, G. D., Younger, J., and McGuire, B. W. (1991).Proc. Am. SOC.Clin. Oncol. 10, 70 (Abstr. 153). DeVita, V. T., Hellman, S., and Rosenberg, S. A., eds. (1991). “Biologic Therapy of Cancer.” Lippincott, Philadelphia. d e Vries, P., Brasel, K. A,, Eisenman, J. R., Alpert, A. R., and Williams, D. E. (1991).J.Exp. Med. 173, 1205-1211. Dexeus, F. H., Kilbourn, R. G., Striegel, A,, Sella A,, Amato, R. J., Bui, C., and Logothetis, C. J. (1991). Proc. Am. Assoc. Cancer Res. 3 2 , 267 (Abstr. 1590). Dicke, K. A., Spitzer, G., Jagannath, S., and Evinger-Hodges, M. J., eds. (1989). “Autologous Bone Marrow Transplantation.” University of Texas, M. D. Anderson Cancer Center. Dillman, R. O., Oldham, R. K., Birch, R., Barth, N. M., Cohen R. J., Minor, D. R., Maleckar, J., Yannelli, J., and West, W. H. (1990a). Proc. Am. Soc. Clin. Oncol. 9, 276 (Abstr. 1069). Dillman, R. O., Oldham, R. K., Barth, N. M., Birch, R., Arnold, J., and West, W. H. (1990b). J. Natl. Cancer Inst. 8 2 , 1345-1349. Dillman, R. O., Church, C., Barth, N. M., Oldham, R. K., Schwartzberg, L., Arnold, J., Birch, R.,and West, W. H. (1991a). Proc. Am. SOC.Clin. Oncol. 10, 212 (Abstr. 711). Dillman, R. O., Oldham, R. K., Tauer, K. W., Orr, D. W., Barth, N. M., Blumenschein, G., Arnold, J., Birch, R., and West, W. H. (1991b).J. Clin. Oncol. 9, 1233-1240. Dillman, R. O., Oldham, R. K., Barth, N. M., Cohen, R. J., Minor, D. R., Birch, R., Yannelli, Cancer68, 1-8. J. R., Maleckar, J. R., Sferruzza, A,, Arnold, J., and West, W. H. (1991~). Dinney, G . P. N., Utsagi, T., Bucana, G. D., Fidler, I. J., and Killion, J. J. (1991).Proc. Am. Assoc. Cancer Res. 3 2 , 249 (Abstr. 1478). Djeu, J. Y., and Blanchard, D. K. (1988). Cell. Immunol. 111, 55-65. Donohue, J. H., Ramsey, P. S., Kerr, L. A., Segal, D. M., and McKean, D. J. (1990). Cancer Res. 50,6508-6514. Douvdevani, A., Hulihel, M., Zoller, M., Segal, S., and Apte, R. N. (1992).Int. J. Cancer (in press). Dupere, S., Obiri, N., Lackey, A., Emma, D., Yannelli, J., Orr, D., Birch, R., and OConnor, T. E. (1990).J. B i d . Response Modif 9, 140-148. Dutcher, J. P., Creekmore, S., Weiss, G. R., Margolin, K., Markowitz, A. B., Roper, M., Parkinson, D., Ciabanu, N., Fisher, R. I., Boldt, D. H., Doroshow, J. H., Rayner, A. A., Hawkins, M., and Atkins, M. (1989).J. Clin. Oncol. 7, 477-485. Dutcher, J. P., Gaynor, E. R., Boldt, D. H., Doroshow, J. H., Bar, M. H., Sznol, M., Mier, J., Sparano, J., Fisher, R. I., Weiss, G., Margolin, K., Aronson, F. R., Hawkins, M., and Atkins, M. (1991).J. Clin. Oncol. 9, 641-648. Eberlein, T. J., Rosenstein, M., and Rosenberg, S. A. (1982).J. Exp. M e d . 156, 385-397. Eberlein, T. J., Rodrick, M. L., Massaro, A. F., Jung, S. E., Mannick, J. A., and Schoof, D. 1). (1989). Cancer Immunol. Immunother. 30, 145- 150. Ebert, E. C., Roberts, A. I., Devereux, D., and Nagasi, H. (1990). Cancer Res. 50, 61586161. Ebina, N., Gallardo, D., Shau, H., and Golub, S. H. (1990). Br. J . Cancer 6 2 , 619-623. Eggermont, A. M. M., and Sugarbaker, P. H. (1988). B r . J . Cancer 58,410-414. Eggermont, A. M. M., Goey, S. H., Slingerland, R., Jansen, R. L. H., Oostrom, R., Bolhuis, R. L. H., and Stoter, G. (1991). Proc. Am. Assoc. Cancer Res. 3 2 , 206 (Abstr. 1228). Ehrke, M. J., Cohen, S. A., and Mihich, E. (1982). Immunol. Rev. 65, 55-78.

300

ELI KEDAR A N D EVA KLEIN

Ehrke, M. J., Maccubbin, D., Ryoyama, K., Cohen, S. A., and Mihich, E. (1986).Cancer Res. 46, 54-60. Ehrke, M. J., Maccubin, D., Mace, F. K., and Mihich, E. (1988).Proc. Am. Assoc. Cancer Res. 29, 1640. Eisenthal, A. (1990). Cancer Immunol. Zmmunother. 31, 342-348. Eisenthal, A., and Rosenberg, S. A. (1989). Cancer Res. 49,6953-6959. Eisenthal, A., Lafreniere, R., Lefor, A. T., and Rosenberg, S. A. (1987). Cancer Res. 47, 2771-2776. Eisenthal, A., Cameron, R. B., Uppenkamp, I., and Rosenberg, S. A. (1988).CancerRes. 48, 7 140-7 145. Elias, L., and Sandoval, J. M. (1989). Biochem. Biophys. Res. Commun. 163, 867-874. Ellenhorn, J. D. I., Schreiber, H., and Bluestone, J. A. (199O).J.Zmmunol. 144,2840-2846. Elliott, B. E., Carlow, D. A., Rodricks, A. M., and Wade, A. ( 1 989).Adv. Cancer Res. 53, 18 1245. Engers, H. D., Lahaye, T., Sorenson, G. D., Glasebrook, A. L., Horvath, C., and Brunner, K. T. (1984).J . Zmmunol. 133, 1664-1670. Escudier, B., Farace, F., Droz, J. P., Antoun, S., Triebel, F., Leclercq, B., Nitenberg, G., and Hercend, T. (1991). Proc. Am. SOC. Clin. Oncol. 10, 166 (Abstr. 527). Estin, C. D., Stevenson, U. S., Plowman, G. D., Hu, S. L., Sridhar, P., Hellstrom, I., Brown, J. P., and Hellstrom, K. E. (1988). Proc. Natl. Acad. Scz. U.S.A. 85, 1052-1056. Ettinghausen, S. E., and Rosenberg, S. A. (1986). Cancer Res. 46, 2784-2792. Ettinghausen, S. E., Puri, R. K., and Rosenberg, S. A. (1988).J. Natl. Cancer Znst. 80, 177188. Falkson, C. I., Falkson, G., and Falkson, H. C. (1991).J.Clin. Oncol. 9, 1403-1409. Faltynek, C. R., and Oppenheim, J. J. (1988).J. Natl. Cancer Znst. 80, 151-153. Fanger, M. W., Segal, D. M., and Wunderlich, J. R. (1990).FASEB J. 4, 2846-2849. Fanger, M. W., Segal, D. M., and Romet-Lemonne, J. L. (1991). Immunol. Today 12,51-54. Favrot, M. C., Combaret, V., Negrier, S., Philip, I., Thiesse, P., Freydel, C., Bijmann, J. T., Franks, C. R., Mercatello, A., and Philip, T. (199O).J.B i d . ResponseModif. 9, 167-177. Fearon, E. R., Pardoll, D. M., Itaya, T., Golumbek, P., Levitsky, H. I., Simons, J. W., Karasuyama, H., Vogelstein, B., and Frost, P. (1990). Cell (Cambridge, Mass.) 60, 397403. Fefer, A. (1974). Zn “Antineoplastic and Immunosuppressive Agents” (A. C. Sartorelli and D. G. Johns, eds.), Vol. 1 , pp. 528-554. Springer-Verlag. Berlin and New York. Fefer, A., Einstein, A. B., Cheever, M. A., and Berenson, J. R. (1976). Ann. N.Y. Acad. Scz. 276, 573-580. Fefer, A., Cheever, M. A,, and Greenberg, P. D. (1982).In “Immunological Approaches to Cancer Therapeutics” (E. Mihich, ed.), pp. 334-362. Wiley, New York. Fendly, B. M., Kotts, C., Vetterlein, D., Lewis, G. D., Winget, M., Carver, M. E., Watson, S. R., Sarup, J., Saks, S., Ullrich, A,, and Shepard, M. (199O).J.B i d . Response Modif. 9, 449-455. Fernandez-Botran, R. (1991). FASEB J. 5, 2567-2574. Fidler, I. J. (1985). Cancer Res. 45, 4714-4726. Fidler, I. J. (1988a). Zsr. J . Med. Scz. 24, 456-463. Fidler, I. J. (1988b). Adu. Drug Delzvety Rev. 1, 69-106. Fidler, I. J. (1990). Cancer Res. 50, 6130-6138. Fidler, I. J., Fogler, W. E., Kleinerman, E. S., and Saiki, I. (1985).J. Immunol. 135, 42894296. Finke, J., Stanley, J., Rayman, P., Alexander, J., Edinger, M., Tubbs, R., Sergi, J., and Bukowski, R. (1991). Proc. Am. Assoc. Cancer Res. 32, 273 (Abstr. 1626).

CANCER IMMUNOTHERAPY

30 1

Finn, 0.J., Persons, D. A,, Bendt, K. M., Pirami, L., and Ricciardi, P. (1991).J. Immunol. 146, 1099-1103. Fiorentino, D. F., Zlotnik, A., Yieira, P., Mosmann, T. R., Howard, M., Moore, K. W., and OGarra, A. (1991).J. Immunol. 146, 3444-3451. Flaherty, L. E., Redman, B. G., Chabot, G. G., Martino, S., Gualdoni, S. M., Heilbrun, L. K., Valdivieso, M., and Bradley, E. C. (1990a). Cancer (Philadelphia) 65, 2471-2477. Flaherty, I-. E., Robinson, W., Redman, B., Gonzales, R., Martino, S., Kraut, M., Valdivieso, M., and Rudolph, A. (1990b). Proc. Am. Soc. Clin. Oncol. 9, 187 (Abstr. 723). Flodgren, P., Borgstrom, S., Jonsson, P. E., Lindstrom, C., and Sjogren, H. 0. (1983).Int.J. Cancer 32, 657-665. Foon, K. A. (1989). CancerRes. 49, 1621-1639. Formelli, F., Rossi, C., Sensi, M. L., and Parmiani, G. (1988). Int. J. Cancer 42, 952957. Forni, G., Cavallo, G. P., Giovarelli, M., Benetton, G., Jemma, C., Barioglio, M. G., De Stefani, A,, Forni, M., Santoni, A,, Modesti, A., Cavallo, G., Menzio, P., and Cortesina, G. (1988). Prog. Exp. Tumor Res. 32, 187-212. Forni, G., Giovarelli, M., Bosco, M. C., Caretto, P., Modesti, A., and Boraschi, D. (1989). Int. J. Cancer, Suppl. 4, 62-65. Foulke, R. S., Marshall, M. H., Trotta, P. P., and von Hoff, D. D. (1990). Cancer Res. 50, 6264-6267. Fox, B. A., Spiess, P. J., Kasid, A,, Puri, R., Mule, J. J., Weber, J. S., and Rosenberg, S. A. (1990).J. Biol. Response Mod$ 9, 499-5 1 1. Fox, B. A,, Culver, K. W., Freeman, S., Koeplin, D., and Blease, R. M. (1991). FASEBJ. 5, A1463 (Abstr. 6333). Freedman, R. S., Patenia, R., Rashed, S., Platsoucas, C. D., and Ioannides, C. G. (1990). FASEBJ. 4, A1718 (Abstr. 136). Freeman, A. I . , and Mayhew, E. (1986). Cancer (Philadelphia) 58, Suppl. 2, 573-583. Frick, J., and Aulitzky, W. (1990). J. Cancer Res. Clin. Oncol. 116, Suppl.11, 1167 (Abstr. 21.18.06). Fridman, W. H. (1989). Eur. J. Cancer Clin. Oncol. 25, 1525-1528. Friedmann, T. (1991). Cancer Cells 3, 271-274. Frodin, J.-E., Faxas, M.-E., Hagstrom, B., Lefvert, A.-K., Masucci, G., Nilsson, B., Steinitz, M., Unger, P., and Mellstedt, H. (1991). Hybridoma 10, 421-431. Frohmiiller, S., Schlag, P., Dueck, M., Betzler, M., Friederich, P., Bradbury, I., and Frost, H. (1991). Proc. Am. Assoc. Cancer Res. 32, 156 (Abstr. 492). Fuchimoto, S., Hamada, H., Gohchi, A., Kamikawa, Y., Okanobu, K., and Orita, K. (1990). J. Cancer Res. Clin. Oncol. 116, Suppl.1, 276 (Abstr. A3.103.13). Fulton, A. M. (1988). Prostaglandins 35, 413-425. Futami, H., Jansen, R., Macphee, M. J., Keller, J., McCormick, K., Longo, D. L., Oppenheim, J. J., Ruscetti, F. W., and Wiltrout, R. H. (1990).J. Immunol. 145,4121-4130. Gabizon, A. (1989). Horiz. Biochem. Biophys. 9, 185-21 1. Gabizon, .4. (1991). Submitted for publication. Gabizon, A., and Papahadjopoulos, D. (1988). Proc. Natl. Acud. Sci. U.S.A. 85, 69496953. Gabizon, A., Price, D. C., Huberty, J., Bresalier, R. S., and Papahadjopoulos, D. (1990). Cancer Res. 50, 6371-6378. Gallinger, S., Hoskin, D. W., Mullen, J. B. M., Wong, A. H. C., and Roder, J. C. (1990). Proc. Am. Assoc. Cancer Res. 31, 276 (Abstr. 1637). Gambacorti-Passerini, C., Rivoltini, L., Supino, R., Rodolfo, M, Radrizzani, M., Fossati, G., and F’armiani, C . (1988). Cancer Res. 48, 2372-2376.

302

ELI KEDAR A N D EVA KLEIN

Gambacorti-Passerini, C., Hank, J., Surfus, J., Tans, K., Moore, K., Borchert, A., Albertini, M., and Sondel, P. M. (1991). Proc. Am. Assoc. Cancer Res. 32, 248 (Abstr. 1476). Gansbacher, B., Bannerji, R., Daniels, B., Zier, K., Cronin, K., and Gilboa, E. (1990a). Cancer Rex 50, 7820-7825. Gansbacher, B., Zier, K., Daniels, B., Cronin, K., Bannerji, R., and Gilboa, E. (1990b).J . Exp. Med. 172, 1217-1224. Ganser, A,, Lindemann, A., Seipelt, G., Ottmann, 0.G., Herrmann, F., Eder, M., Frisch, J., Schulz, G., Mertelsmann, R., and Hoelzer, D. (1990). Blood 76, 666-676. Gaynor, E. R., Weiss, G. R., Margolin, K. A,, Aronson, F. R., Sznol, M., Demchak, P., Grima, K. M., Fisher, R. I.,.Boldt, D. H., Doroshow, J. H., Bar, M. H., Hawkins, M. J., Mier, J. W., and Caliendo, G. (1990).J . Natl. Cancer Iwt.82, 1397-1402. Gazit, Z., Weiss, D. W., Shouval, D., Yechezkeli, M., Schirrmacher, V., Notter, M., Walter,J., and Kedar, E. (1992). Immunol. Immunother. (in press). Gelin, J., Moldawer, L. L., Lonnroth, C., Sherry, B., Chizzonite, R., and Lundholm, K. (1991a). Cancer Res. 51, 415-421. Gelin, J., Anderson, C., and Lundholm, K. (1991b). Cancer Res. 51, 880-885. Gemlo, 8. T., Palladino, M. A., Jaffe, H. S., Espevik, T. P., and Rayner, A. A., (1988). Cancer Res. 48, 5864-5867. Ghosh, A. K., Dazzi, H., Thatcher, N., and Moore, M. (1989). Int.J. Cancer 43,410-414. Giampietri, A,, Bonmassar, A., Puccetti, P., Circalo, A., Goldin, A,, and Bonmassar, E. (1981). Cancer Rex 41, 681-687. Gianni, A. M., Bregni, M., Siena, S., Orazi, A,, Stern, A. C., Gandola, L., and Bonadonna, G . (1990).J. Clzn. Oncol. 8, 768-778. Giavazzi, R., Campbell, D. E., Jessup, J. M., Cleary, K., and Fidler, I. J. (1986). Cancer Res. 46, 1928-1933. Giavazzi, R., Garofalo, A., Bani, M. R., Abbate, M., Ghezzi, P., Boraschi, D., Mantovani, A., and DeJana, E. (1990). Cancer Res. 50, 4771-4775. Gifford, R. R., Ferguson, R. M., and Voss, B. V. (1981). Lancet 1, 638-640 Gilewski, T. A., and Golomb, M. H. (1990). Semzn. Oncol. 17, Suppl. 1, 3-10. Gill, I., Agah, R., Hu, E., and Mazumder, A. (1989). Cancer Res. 49,5377-5379. Giovanella, B. C., and Fogh, J. (1985). Adu. Cancer Res. 44,69-120. Giovarelli, M., Cofano, F., Vecchi, A., Forni, M., Landolfo, S., and Forni, G. (1986). Int. J . Cancer 37, 141-148. Glaspy, J. A,, and Golde, D. W. (1990). Exp. Hematol. 18, 1137-1141. Golde, D. W. (1990). Semzn. Hematol. 27, Suppl. 3, 1-7. Goldenberg, D. M. (1991).J. Natl. Cancer Inst. 83, 78-79. Golumbek, P. T,, Lazenby, A. J., Levitsky, H. I., Jaffee, L. M., Karasuyama, H., Baker, M., and Pardoll, D. M. (1991). Science 254, 713-716. Gopas, J., Rager-Zisman, B., Bar-Eli, M., Hammerling, G. J., and Segal, S. (1989). Adu. Cancer Res. 53, 89-1 15. Gottlieb, D. J., Brenner, M. K., Heslop, H. E., Bianchi, A. C. M., Bello-Fernandez, C., Mehta, A. B., Newland, A. C., Galazka, A. R., Scott, E. M., Hornrand, A. V., and Prentice, H. G. (1989). I3r.J. Cancer 60, 610-615. Greenberg, P. D. (1991). Adu. Immunol. 49,281-355. Greenberg, P. D., Klarnet, J. P., Kern, D. E., and Cheever, M. A. (1988). Prog. Exp. Tumor Res. 32, 104-127. Gregoriadis, G. (1990). Immunol. Toduy 11, 89-97. Grem, J. L., and Allegra, C. J. (1989).J. Natl. CancerIwt. 81, 1413-1417. Grem, J. L., McAtee, N., Murphy, R. F., Balis, F. M., Sorensen, J. M., Hamilton, J. M., Sartor, O., and Allegra, C. J. (1991). Proc. ASCO 10, 94 (Abstr. 246).

CANCER IMMUNOTHERAPY

303

Gribaudo, G., Gariglio, M., Cavallo, G., and Landolfo, S. (1990). Eur.J. Immunol. 20, 12431249. Griffin, J. D. (1989).J. Clin. Oncol. 7, 151-155. Griffith, K. D., Read, E. J., Carrasquillo, J. A., Carter, C. S., Yang, J. C., Fisher, B., Aebersold, P., Packard, B. S., Yu, M. Y., and Rosenberg, S. A. (1989).J. Natl. Cancer Inst. 81, 1709-1717. Grimm, E. A., Crump, W. L., Durett, A., Hester, J. P., Lagoo-Deenadalayan, S., and OwenSchaub, L. B. (1988). Cancer Immunol. Immunother. 27, 53-58. Groopman, J. E., Molina, J. M., and Scadden, D. T. (1989). N. Eng1.J. Med. 321, 14491459. Guadagni, F., Schlom, J., Johnston, W. W., Szpak, C. A., Goldstein, D., Smalley, R., Simpson, J. F., Borden, E. C., Pestka, s., and Greiner, J. W. (1989). J . Natl. Cancer Inst. 81, 502-5 12. Guillou, P. J., Somers, S. S., and Sedman, P. C. (1989). Interferons Cytokznes 11, 6-9. Gustavson, L. E., Nadeau, R. W., and Oldfield, N. F. (1989).J. Biol. Response Modif. 8,440449. Gutterman, J., Vadhan-Raj, S., Logothetis, C., Anaissle, E., Estey, E., Talpaz, M., and Kurzrock, P. (1990). Semin. Hematol. 27, Suppl. 3, 15-24. Hadden, J. W. (1989). CUT. Opin. Immunol. 2, 258-263. Hamblin, T. J. (1989). Eur. J. Cancer Clin. Oncol. 25, Suppl.3, S7-S9. Hamblin, T. J., Davies, B., Sadullah, S., Oskam, R., Plamer, P., and Franks, C. R. (1991). Proc. Am. SOC.Clin. Oncol. 10, 294 (Abstr. 1029). Hank, J. A,, Robinson, R. R., Surfus, J., Mueller, B. M., Reisfeld, R. A,, Cheung, N. K., and Sondel, P. M. (1990a). Cancer Res. 50, 5234-5239. Hank, J. A,, Sosman, J. A., Kohler, P. C., Bechhofer, R., Storer, B., and Sondel, P. M. (1990b).J. Biol. Response M o d f . 9, 5-14. Hanson, J., Petit, R., Walker, M., Kurtz, J., Rohloff, C., Kabler-Babbitt, C., Aleem, J., and Rausch, C. (1991). Proc. Am. Soc. Clin. Oncol. 10, 295 (Abstr. 1035). Haranaka, K. (1988). J . Biol. Response Modif. 7,525-534. Harding, C. V., Collins, D. S., Slot, J. W., Geuze, H. J., and Unanue, E. R. (1991). Cell (Cambrzdge, Mass.) 64, 393-401. Hashimoto, Y.,Sugawara, M., Masuko, T., and Hojo, H. (1983).CancerRes. 43,5328-5334. Havell, E. A., Fiers, W., and North, R. J. (1988).J. Exp. Med. 167, 1067-1085. Hellstrom, K. E., and Hellstrom, I. (1989). FASEBJ. 3, 1715-1722. Henkart, P., and Yue, C. C. (1988). Prog. Allergy 40, 82-1 10. Hennigan, T. W., Begent, R. H. J., and Allen-Mersh, T. G. (1991). Br. J . Cancer 64,872874. Henon, P. R., Butturini, A., and Gale, P. R. (1991). Lancet 337, 961-963. Heppner, G. H. (1984). Cancer Res. 44,2259-2265. Herberman, R. B. (1983a).J. Biol. Response Mod$ 2, 39-46. Herberman, R. B. (1983b).J. Biol. Response M o d f . 2, 217-226. Herberman, R. B. (1987). Cancer Bull 39, 6-13. Herberman, R. B. (1989).J. Clin. Oncol. 17, 1-4. Hercend, T., Farace, F., Baume, D., Charpentier, F., Droz, J. P., Triebel, F., and Escudier, B. (1990). J. Biol. Response Modv. 9, 546-555. Herlyn, D., Wettendorff, M., Schmoll, E., Iliopoulos, D., Schedel, I., Dreikhausen, U., Rabb, R., Ross, A. H., Jaksche, H., Scriba, M., and Koprowski, H. (1987). Proc. Natl. Acad. S C ~U.S.A. . 84, 8055-8059. Herlyn, M., Menrad, A,, and Koprowski, H. (199O).J. Natl. Cancer Inst. 82, 1883-1889. Hersey, P., and Bolhuis, R. (1987). Ittzmunol. Today 8,233-239.

304

ELI KEDAR A N D EVA KLEIN

Hersh, E. M., Scuderi, P., Grimes, W. J., Chong, A,, Brailey, J. L., Gschwind, C. R., and Salmon, S. E. (1989). Cancer Immunol. Immunother. 30, 65-70. Hewitt, H. B. (1978). Adv. Cancer Res. 27, 149-200. Hewitt, H. B. (1982).J. Biol. Response Modif. 1, 107-119. Hirsh, M., Lipton, A., Harvey, H., Givant, E., Hopper, K., Jones, G., Zeffren, J., and Levitt, D. (1990).J. Clin. Oncol. 8, 1657-1663. Hirte, H. W., and Clark, D. A. (1990). Proc. Soc. Biol. Ther. 16, (Abstr. 26). Hollinshead, A,, Takita, H., and Stewart, T. (1989). In “Cancer Metastasis” (V. Schirrmacher and R. Schwartz-Albeiz, eds.), pp. 199-205. Springer-Verlag, Berlin. Hollinshead, A,, Stewart, T., Elias, G., and Arlen, M. (1990). Proc. Am. Soc. Clin. Oncol. 9, 120 (Abstr. 467). Hoon, D. S. B., Forshag, L. J., Nizze, A. S., Bohman, R., and Morton, D. L. (1990). Cancer Res. 50,5358-5364. Hoon, D. S. B., Banez, M., Okun, E., Morton, D. L., and hie, R. F. (1991). Cancer Res. 51, 2002-2008. Hoover, H. C., and Hanna, M. G. (1989). In “Cancer Metastasis” (V. Schirrmacher and R. Schwartz-Albeiz, eds.), pp. 192-1 98. Springer-Verlag, Berlin. Hoover, S. K., Barrett, S. K., Turk, T. M. T., Lee, T. C., and Bear, H. D. (1990). Camer Immunol. Immunother. 31, 121-127. Hornung, R. L., Back, T. C., Zaharko, D. S., Urba, W. J., Longo, D. L., and Wiltrout, R. H. (1988). J . Immunol. 141, 3671-3679. Horowitz, M. H., Gale, R. P., Sondel, P. M., Goldman, J. M., Kersey, J., Kolb, H.-J., Rimm, A. A,, Rindgen, O., Rozman, C., Speck, B., Truitt, R. L., Zwaan, F. E., and Bortin, M. M. (1990). Blood 75, 555-562. Hosokawa, M., Sawamura, Y., Morikage, T., Okada, F., Xu, Z. Y., Morikawa, K., Itoh, K., and Kobayashi, H. (1988). Cancer Immunol. Immunother. 26, 250-256. Hosokawa, M., Kawata, A., Sawamura, Y., and Kobayashi, H. (1990). J . Cancer Res. Clin. Oncol. 116, Suppl. 11, 1140 (Abstr. 20.30.03). Huland, E., and Huland, H. (1989). Cancer Res. 49, 5469-5474. Huland, E., Huland, H., and Meier, T. (199O).J. Cancer Res. Clzn. Oncol. 116, Suppl.1, 336 (Abstr. A3.112.33). Hunter, T. (1991). Cell (Cambridge, Mass.) 64,249-270. Ikon, D., Motzer, R., Kradin, R., Vogelzang, N., Bajorin, D., Scher, H., OMoore, P., Zeffren, J., Marathias, K., Lazarus, D., Ochoa, M., and Bosi, G. (1991). Proc. Am. Assoc. Cancer Res. 32, 187 (Abstr. 1 1 16). Inoshita, G., Yalavarthi, P., Murthy, S., Budd, G. T., Andresen, S., Netaji, B., Gibson, V., Sergi, J., and Bukowski, R. M. (1991). Proc. Am. Soc. Clin. Oncol. 10, 152 (Abstr. 475). Inoue, K., Higuchi, K., Tsukishiro, T., Miyagiwa, M., Okada, K., Watanabe, A., and Nishino, K. (1990). J . Cancer Res. Clzn. Oncol. 116, Suppl.1, 292 (Abstr. A3.103.78). Ioannides, C. G., Freedman, R. S., Platsoucas, C. D., Rashed, S., and Kim, P. L. (1991).J. Immunol. 146, 1700-1707. Isacson, R., Kedar, E., Barak, V., Gazit, Z., Yurim, O., Kalichman, I., Ben-Basat, H., Biran, S., Schlesinger, M., and Shiloni, E. (1992). Immunol. Lett. (in press). Isobe, K. I., Hasegawa, Y., Iwamoto, T., Hasegawa, T., Kawashima, K., Ding, L. N., and Nakashima, I. (1989).J. Natl. Cancer Inst. 81, 1823-1828. Israel, L., Breau, J. L., Morere, J. F., Boaziz, C., and Bouillet, T. (1990). Proc. Am. Assoc. Cancer Res. 31, 207 (Abstr. 1234). Itoh, K., Tilden, A. B., and Balch, C. M. (1985). Cancer Res. 45, 3173-3178. Itoh, K., Platsoucas, C. D., and Balch, C. M. (1988).J. Exp. Med. 168, 1419-1441.

CANCER IMMUNOTHERAPY

305

Jadus, M. R., Good, R. W., Crumpacker, D. B., and Yannelli, J. R. (1991).J. Leuk. Biol. 49, 139-15 1. Jaffe, H. S., and Herberman, R. B. (1988). J. Nutl. Cancer Imt. 80, 616-618. Janson, C. H., Tehrani, M., Wigzell, H., and Mellstedt, H. (1989). Leuk. Res. 13, 1039-1046. Jemrna, C., Giovarelli, M., Faltynek, C. R., Benetton, G., Forni, M., Caretto, P., Brunda, M. J., Varesio, L., Geuna, M., and Forni, G. (1989).J. Immunol. Res. 1, 11 1-1 19. Johnston, J. (l99l).J. NIH Res. 3, 36-38. Joraschkewitz, M., Depenbrock, H., Freund, M., Erdmann, G., Meyer, H. J., De Riese, W., Neukam, D., Hanauske, U., Krurnwieh, M., Poliwoda, H., and Hanauske, A. R. (1990). Eur. J. Cancer 26, 1070-1074. Jung, S., and Schluesener H. J. (1991).J. Exp. Med. 173, 273-276. Junghans, R. P. (1990). Cancer Immunol. Immunother. 81, 207-212. Kahn, M., Sugawara, H., McGowan, P., Okuno, K., Nagoya, S., Hellstrom, K. E., Hellstriirn, I., and Greenberg, P. (199l).J. Immunol. 146, 3235-3241. Kalechrnan, Y., Albeck, M., Oron, M., Sobelman, D., Gurwith, M., Horwith, G., Kirsch, T., Maida, B., Sehgal, S. N., and Sredni, B. (1991). CancerRes. 51, 1499-1503. Karirnine, N., Arinaga, S., Abe, R., Inoue, H., Watanabe, D., Matsuoka, H., Takamuku, K., Ueo, H., and Akiyoshi, T. (1990). J. Cancer Res. Clzn. Oncol. 116, Supl.1, 283 (Abstr. A3.103.43). Kasid, A,, Director, E. P., and Rosenberg, S. A. (1989).J. Immunol. 143, 736-739. Kasid, A,, Morecki, S., Aebersold, P., Cornetta, K., Culver, K., Freeman, S., Director, E., Lotze, M. T., Blease, R. M., Anderson, W. F., and Rosenberg, S. A. (1990). Proc. Nutl. Acad. Sri. U.S.A. 87, 473-477. Kast, W. M., Offringa, R., Peters, P. J., Voordouw, A. C., Meloen, R. H., Van der Eb, A. J., and Melief, J. M. (1989). Cell (Cambndge, Mass.) 59, 603-614. Katre, N. V., Knauf, M. J., and Laird, W. J. (1987). Proc. Natl. A d . Scz. U.S.A. 84, 14871491. Katsanis, E., Ochoa, A. C., Bausero, M. A,, Loeffler, C. M., and Anderson, P. M. (1990). Proc. Am. A s ~ o c .Cancer Res. 31, 270 (Abstr. 1597). Kawase, I., Komutd, K., Hara, H., Inoue, T., Hosoe, S., Ikeda, T., Shirasaka, T., Yokota, S., Tanio, Y., Masuno, T., and Kishirnoto, S. (1988). Cancer Res. 48, 1173-1179. Kedar, E., and Lupu, T. (1978).J. Immunol. Methods 21, 35-50. Kedar, E., and Weiss, D. W. (1983). Adv. Cancer Res. 38, 171-287. Kedar, E., Lupu, T., Schwartzbach, M., and Avraham, Y. (1979). J . Immunol. Methods 26, 151-171. Kedar, E., lkejiri, B., Gorelik, E., and Herberman, R. B. (1982). Cancer Immunol. Immunother. 13, 14-23. Kedar, E., Ikejiri, B. L., Tirnonen, T., Bonnard, G. D., Reid, J., and Herberman, R. B. (1983). Eur. ,I. Cancer Clin. Oncol. 19, 757-773. Kedar, E., Chriqui-Zeira, E., and Kyriazis, A. P. (1984a).J. Biol. ResponseMod$ 3, 517-526. Kedar, E., Chriqui-Zeira, E., and Mitelrnan, S. (1984b). Cancer Immunol. Immunother. 18, 126-134. Kedar, E., Bercowitz, H., Katz-Gross, A., and Chriqui-Zeira, E. (1 986). Immunopharmacology 11, 61-65. Kedar, E., Ben-Aziz, R., and Shiloni, E. (1988a). Isr. J . Med. Sci. 24, 494-504. Kedar, E., Rezai, A. R., Giorgi, J., Gale, R. P., Charnplin, R. E., Mitsuyasu, R.T., and Fahey, J. L. (1988b). Nat. Immun. Cell Growth Regul. 7, 13-30. Kedar, E., Tsuberi, B., Landesberg, A., Anafi, M., Leshem, B., Gillis, S., Urdal, D. L., and Bone Marrow Transplant. 3, 297-314. Slavin, S. (1988~).

306

ELI KEDAR A N D EVA KLEIN

Kedar, E., Ben-Aziz, R., Epstein, E., and Leshem, B. (1989). Cancer Immunol. Immunother. 29, 74-78. Kedar, E., Lebendiker-Gazit, Z., Tsuberi, B., Ben-Aziz R., Epstein, E., Weisz, A., Rutkowski, Y., Shiloni, E., Shouval, D., Weiss, D. W., Leshem, B., and Slavin, S. (1990). In “Recent Advances in Chemotherapy” (A. Adam and E. Rubinstein, eds.), pp. 840.1848.1. Lewin-Epstein Ltd., Bat-Yam. Kedar, E., Rutkwoski, Y., and Leshem, B. (1992). Cancer Immunol. Immunother. (in press). Kedar, E., and Gazit, Z. (1992). Submitted for publication. Keller, R. (1989). Eur. J. Cancer Clin. Oncol. 25, 599-601. Kellokumpu-Lehtinen, P., and Nordman, E. (1990). J. B i d . Response Mod$. 9, 439444. Kelly, S. A., Gschmeissner, S., East, N., and Balkwill, F. R. (1991). Cancer Res. 51, 40204027. Kelso, A. (1989). Cum. Opin. Immunol. 2, 215-225. Kelso, A., and Metcalf, D. (1990). Adv. Zmmunol. 48, 69-105. Kemeny, N., Kelsen, D., Derby, S., Sammarco, P., Adams, L., and Murray, P. (1990). Proc. Am. Soc. Clin. Oncol. 9, 109 (Abstr. 420). Kerr, L., Huntoon, C., Donohue, J., Leibson, P. J., Bander, N. H., Chose, T., Luner, S. J., Vessella, R., and McKean, D. J. (1990). J. Immunol. 144, 4060-4067. Kessinger, A., and Armitage, J. 0. (1991). Blood 77, 211-213. Ketzinel, M., Arad, G., Tal, C., Schlesinger, M., Nussinovich, R., Reshef, A., Sayar, D,. Gerez, L., and Kaempfer, R. (1990). Mol. Immunol. 27, 1325-1330. Khoo, N. K. S., Chan, F. P. H., Saarloos, M. N., and Lala, P. K. (1990). Proc. Am. Assoc. Cancer Res. 31, 275 (Abstr. 1628). Killion, J. J., Fan, D., Bucana, C. D., Frangos, D. N., Price, J. E., and Fidler, I. J. (19891.J. Natl. Cancer Imt. 81, 1387-1392. Kim, B., and Louie, A. (1991). Proc. Am. Soc. Clin. Oncol. 10, 164 (Abstr. 521). Kim, B., and Warnaka, P. (1989). Surgety (St. Louis) 106, 248-256. Kim, R., Lafreniere, R., Borkenhagen, K., and Bryant, L. D. (1989). Cancer Immunol. Immunother. 30, 283-288. Kinnon, C., and Levinsky, R. J. (1990). Eur.1. Cancer 26, 638-640. Kirchner, H., Korfer, A., Evers, P., Szamel, M. M., Kniiver-Hopf, J., Mohr, H., Franks, C. R., Pohl, U., Resch, K., Hadam, M., Poliwoda, H., and Atzpodien, J. (1991). Cancer (Philadelphia) 67, 1862- 1864. Kiyohara, T., Taniguchi, K., Kubota, S., Koga, S., Sakuragi, T., and Saitoh, Y. (1988).J. Exp. Med. 168,2355-2360. Klarnet, J. P., Kern, D. E., Okumo, K., Holt, C., Lilly, F., and Greenberg, P. D. (1989).J. Exp. Med. 169, 457-467. Kleinerman, E. S. (1991).J. NIH Res. 3, No. 1, 48-50. Kleinerman, E. S., Murray, J. L., Snyder, J. S., Cunnigham, J. E., and Fidler, I. J. (1989). Cancer Res. 49,4665-4670. Knuth, A., Wolfel, T., and Meyer zum Biischenfelde, K.-H. (1991). Curr. @in. Immunol. 3, 659-664. Kobayashi, H. (1986).J. Bzol. Response Mod$ 5 , 1-1 1. Kohn, D. B., Kantoff, P. W., Eglitis, M. A., McLachlin, J. R., Moen, R. C., Karson, E., Zwiebel, J. A., Nienhuis, A., Karlsson, S., OReilly, R., Gillio, A,, Bordignon, C., Gilboa, E., Zanjani, I. D., Blease, R. M., and Anderson, W. F. (1987). Blood Cells 13, 285-298. Koike, A., Nakazato, H., and Saji, S. (1990). J. Cancer Res. Clzn. Oncol. 116, Suppl.11, 872 (Abstr. 06.04.08). Kolitz, J. E., Wong, G. Y.,Welte, K., Merluzzi, V. J., Engert, A., Bialas, T., Polivka, A.,

CANCER IMMUNOTHERAPY

307

Bradley, E. C., Konrad, M., Gnecco, C., Oettgen, H. F., and Mertelsmann, R. (1988).J. Biol. Response Mod$ 7,457-472. Korber, B., Mermod, N., Hood, L., and Stroynowski, I. (1988). Science 239, 1302-1306. Kornstein, M. J., Brooks, J. S. J., and Elder, D. E. (1983). Cancer Res. 43, 2749-2753. Kovach, J. S. (1991). J. Nutl. Cancer Inst. 83, 2-3. Kovacs, E. J. (1991). Immunol. Toby 12, 17-23. Kradin, R. L., Kurnick, J. T., Lazarus, D. S., Preffer, F. I., Dubinett, S. M., Pinto, C. E., Gifford, J., Davidson, E., Grove, B., Callahan, R. J., and Strauss, H. W. (1989a). Lancet 1, 577-580. Kradin, R. L., Kurnick, J. T., Preffer, F. I., Dubinett, S. M., Dickersin, G. R., and Pinto, C. (1989b). Clin. Immunol. Immunopathol. 50, 184- 195. Krosnick, J. A,, Mule, J. J., McIntosh, J. K., and Rosenberg, S. A. (1989a). Cancer Res. 49, 3729-3733. Krosnick, J. A., McIntosh, J. K., Mule, J. J., and Rosenberg, S. A. (1989b). CancerImmunol. Immunother. 30, 133-138. Kumai, K., Suzuki, H., Takahashi, T., Hirahata, S., Uyama, I., Shimizu, H., Ishibiki, K., and Abe, 0. (1990). J. Cancer Res. Clin. Oncol. 116, Suppl.1, 603 (Abstr. A5.221.06). Kunkel, S. L., Remick, D., Strieter, R. M., and Larrick, J. W. (1989). CRC Crit. Rev. Immunol. 9, 93-117. Lafreniere, R., Borkenhagen, K., Bryant, L. D., and Ng, E. (1989).J. Biol.RespomeMod$ 8, 238-25 1. Lafreniere, R., Borkenhagen, K., and Bryant, L. D. (1990). FASEB J . 4, A1810 (Abstr. 683). Lala, P. K., and Parhar, R. S. (1988). CancerRes. 48, 1072-1079. Lala, P. K., Elkashab, M., Kerbel, R. S., and Parhar, R. S. (1990). Int. Immunol. 2, 11491158. Landolfo, S., Gariglio, M., Gribaudo, G., Jemma, C., Giovarelli, M., and Cavallo, G. (1988). Eur. J. Immunol. 18, 503-509. Langer, R. (1990). Science 249, 1527-1533. Larisch-Bloch, S., Sterman, B., Weiss, D. W., and Sulitzeanu, D. (1989). Immunol. Lett. 23, 27-30. Larisch-Bloch, S . , Sterman, B., and Sulitzeanu, D. (1990). Immunol. Lett. 26, 189192. Lautersztain, J., Kleinerman, J. E., Fidler, J., Benjamin, R. S., Krakoff, I., and Gutterman, J. U. (1991). Proc. Am. Assoc. Cancer Res. 32, 202 (Abstr. 1204). Laver, J., and Moore, M. A. S. (1989).J. Nail. Cancer Inst. 81, 1370-1382. Lebendiker, M. A., Tal, C., Sayar, D., Pilo, S., Eilon, A., Banai, Y., and Kaempfer, R. (1987). EMBO J. 6, 585-589. Lee, K., ODonnell, R. W., Marquis, D. and Cockett, A. T. K. (1988).J. Biol. Response Mod$ 7, 32-42. Lee, V. K., and Hellstrom, K. E. (1988). Adv. Drug Delivery Rev. 2, 271-296. Legha, S. S., Gutterman, J. U., and Benjamin, R. S. (1990). J. Cancer ReJ. Clin. Oncol. 116, Suppl.1, 288 (Abstr. A3.103.61). Lehner, B., Schlag, P., Liebrich, W., and Schirrmacher, V. (1990). Cancer Immunol. Immunother 32, 173- 178. LeMaistre, C. F., Rosenblum, M. G., Reuben, J. M., Parkinson, D. R., Meneghetti, C. M., Parker, K., Shaw,J. P., Deisseroth, A. B., and Woodworth, T. (1991a). Lancet 337, 11241125. LeMaistre, C. F., Von Hoff, D., Meneghetti, C., Adkins, D., Rosenblum, M., Reuben, J., Deisseroth, A,, Parker, K., Shaw, J., and Woodworth, T. (1991b). Proc. Am. SOC.Clin. Oncol. 10, 280 (Abstr. 975).

308

ELI KEDAR AND EVA KLEIN

Levy, R., and Miller, R. A. (1990). Mongr. Natl. Cancer Inst. 10, 61-68. Ley, V., Langlade-Demoyen, P., Kourilsky, P., and Larsson-Sciard, E. L. (1991). Eur. J. Immunol. 21, 851-854. Li, W., Diamantstein, T., and Blankenstein, T. (1990). Mol. Immunol. 27, 1331-1337. Lindemann, A,, Hoffken, K., Schmidt, R. E., Diehl, V., Kloke, O., Gamm, H., Hayungs, J., Oster, W., Bohm, M., Kolitz,J. E., Franks, C. R., Herrmann, F., and Mertelsmann, R. H. ( 1989). Cancer Immunol. Immunother. 28, 275-28 1. Lippman, M. E. (1991).J. NIH Res. 3, No. 2, 59-62. Lise, L. D., and Audibert, F. (1989). Cum. Opin. Immunol. 2, 269-274. Livingston, P. 0. (1990). Proc. SOC.Bzol. Ther. 11 (Abstr. 26). Livingston, P. (1991a). Immunol. Allergy Clin. North Am. 11, 401-423. Livingston, P. (1991b). Proc. Am. Assoc. Cancer Res. 32, 491-492. Livingston, P. 0.. Cunningham-Rundles, S., Marfleet, G., Gnecco, C., Wong, G. Y., Schiffman, G., Enker, W. E.,and Hoffman, M. K. (1 987).J. Biol. Response Mod$ 6,392403. Loeffler, C. M., Platt, J. L., Anderson, P. M., Katsanis, E., Ochoa, J. B., Urba, W. J., Longo, D. L., Leonard, A. S., and Ochoa, A. C. (1991a). Cancer Res. 51,2127-2132. Loeffler, C. M., Tribble, H., Tasse, J., Harvey, L., Urba, W. J., Longo, D. L., and Ochoa, A. C. (1991b). Proc. Am. Assoc. Cancer Res. 32, 245 (Abstr. 1459). Logothetis, C. J., Hossan, E., Sella, A. Dexeus, F. H., and Amato, R. J. (1991).J. Natl. Cancer Imt. 83, 285-288. Longo, D., ed. (1991). “Cancer Chemotherapy and Biological Response Modifiers Annual.” Elsevier, Amsterdam. LoRusso, P. M., Aukerman, S. L., Polin, L., Redman, B. G., Valdivieso, M., Biernat, L., and Corbett, T. H. (1990). Cancer Res. 50, 5876-5882. Lotze, M. T., and Finn, 0.J., eds. (1990). “Cellular Immunity and the Immunotherapy of Cancer.” Wiley-Liss, New York. Lotze, M. T., Chang, A. E., Seipp, C. A., Simpson, C., Vetto, J. T., and Rosenberg, S. A. (1986). J . Am. Med. Assoc. 256, 3 117-3 124. Lotze, M. T., Custer, M. C., Sharrow, S. O., Rubin, L. A,, Nelson, D. L., and Rosenberg, S. A. (1987). CancerRes. 47, 2188-2195. Lotzova, E. (1989). Cum. Opm. Immunol. 1, 904-909. Lotzova, E., Savary, C. A., Freedman, R. S., Edwards, C. L., and Morris, M. (1990). Cancer Immunol. Immunother. 31, 169-175. Luikart, S. D., MacDonald, M., Herzan, D., Modeas, C., Goutsou, M., Clamon, G., Maurer, H., Perry, M. C., and Green, M. R. (1991). Proc. Am. Soc. Clzn. Oncol. 10, 242 (Abstr. 825). Lynch, N. R., Castes, M., Astoin, M., and Salomon, J. C. (1978). Br.1. Cancer38, 503-512. Maas, R. A,, Dullens, H. F. J., De Jong, W. H., and Den Otter, W. (1989). Cancer Res. 49, 7037-7040. MacEwen, E. G., Kurzman, I. D., Rosenthal, R. C., Smith, B. W., Manley, P. A., Roush, J. K., and Howard, P. E. (1989).J . Natl. Cancer Inst. 81, 935-938. Mach, J.-P., Pklegrin, A., and Buchegger, F. (1991). Cum. Opin. Immunol. 3 , 685-693. Maekawa R., Kitagawa, T., Koizumi, K., Sato, K., and Homma, M. (1989).J. Bzol. Response Modif. 8, 676-690. Maio, M., Gulwani, B., Morgano, A,, and Ferrone, S. (1989). Int. J. Cancer 44, 554-559. Maleckar, J. R., Friddell, C. S., Sferruzza, A,, Thurman, G. B., Lewko, W. M. West, W. H., Oldham, R. K., and Yannelli, J. R. (1989).J. Natl. Cancer Inst. 81, 1655-1660. Malik, S. T. A., Naylor, S., East, N., Oliff, A,, and Balkwill, F. R. (1990). Eur. J. Cancer 26, 103I- 1034.

CANCER IMMUNOTHERAPY

309

Malik, S. T. A., Martin, D., Hart, I., a n d Balkwill, F. (1991). L3r.J. Cancer 63, 399-403. Malkovska. V., Sondel, P. M., a n d Malkovsky, M. (1989). CUT. Opin. Immunol. 1, 883-890. Maluish, A. E., Urba, W. J., Longo, D. L., Overton, W. R., Coggin, D., Crisp, E. R., Williams, R., Sherwin, S. A., Gordon, K., a n d Steis, R. G. (1988).J. Clin. Oncol. 6,434445. Manda, T., Nishigaki, F., Mukumoto, S., Masuda, K., Nakarnura, T., a n d Shimomura, K. (1990). Eur. J. Cancer 26, 93-99. Mantovani, A. (1990). CUT. Opin. Immunol. 2, 689-692. Margolin, K. A., Rayner, A. A., Hawkins, M. J., Atkins, M. B., Dutcher, J. P., Fisher, R. I., Weiss, G. R., Doroshow, J. H., Jaffe, H . S., Roper, M., Parkinson, D. R., Wiernik, P. H., Creekmore, S. P., a n d Boldt, D. H. (1989).J. Clin. Oncol. 7, 486-498. Markman, M. (1987).J. Biol. Response M o d f . 6,404-41 1 . Markowitz, A,, Parkinson, D., Itoh, K., Legha, S., von Eshenbach, A,, Logothetis, C., Yeomans, A., Thielvoldt, D., Kilbourn, M., Hunter, C., Gutterman, J., a n d Balch, C. (1991). Proc. Am. Soc. Clin. Oncol. 10, 216 (Abstr. 726). Marshall, M. H., a n d von Hoff, D. D. (199O).J. Cancer Res. Clin. Oncol. 116, Suppl.1, 279 (Abstr. A3.103.27). Masucci, G., a n d Mellstedt, H. (1990). Leukemia 4, 81-82. Matsushima, K., Hirose, K., a n d Oppenheim, J. (1990). J. Cancer Res. Clin. Oncol. 116, Suppl. 11, 871 (Abstr. 06.04.04). Matzku, S., Krernpel, H., Weckenmann, H. P., Schirrmacher, V., Sinn, H., a n d Stricker, H. (1990). Cancer Immunol. Immunother. 31, 285-291. Mayhew, E., a n d Lasic, D. (1991). Submitted for publication. Mazzocchi, A., Anichini, A., Castelli, C., Sensi, M., Poli, F., Russo, C,. a n d Parrniani, G. (1990). Cancer Immunol. Immunother. 32, 13-21. McCabe, M. S., Stablein, D., a n d Hawkins, M. J. (1991).Proc. Am. SOC.Clzn. Oncol. 10, 213 (Abstr. 714). McCune, C. S., a n d Marquis, D. M. (1990). Cancer Res. 50, 1212-1215. McCune, C. S., ODonnell, R. W., Marquis, D. M., a n d Sahasrabudhe, D. M. (1990).Cancer Immunol. Immunother. 32, 62-66. McIntosh, J. K.. Mule, J. J., Merino, M. J., a n d Rosenberg. S. A. (1988). Cancer Res. 48, 4011-4017. McIntosh, J. K., Mule, J. J., Krosnick, J. A,, a n d Rosenberg, S. A. (1989). Cancer Res. 49, 1408- 14 14. McIntosh, J. K., Mule, J. J., Travis, W. D., a n d Rosenberg, S. A. (1990). Cancer Res. 50, 2463-2469. McKinnon, J. G., Hoover, S. K., Inge, T. H., a n d Bear, H. D. (1990). Cancer Immunol. Immunother. 32, 38-44. McLeod, G. R. C., Thomson, D. B., a n d Hersey, P. (1987). 1nt. J. Cancer Suppl. 1, 31-35. Melief, C. J. M. (1992). Ada. Cancer Res. 58, 143-175. Melief, C. J. M., a n d Kast, W. M. (1990a). Cancer Cells 2, 116-120. Melief, C. J. M., a n d Kast, W. M. (1990b).J. Cancer Rex. Clin. Oncol. 116, Suppl. 11, 1157 (Absir. 21.11.03). Mellstedt, H. (1990). Curr. Opin. Immunol. 2, 708-713. Mertelsmann, R., a n d Welte, K. (1986). lmmunobiology 172, 400-419. Mertens, W. C., Bramwell, V. H. C., a n d Lala, P. K. (1991). Proc. Am. SOC.Clzn. Oncol. 10, 209 (Abstr. 697). Metcalf, D. (1989a). Lancet 1, 825-827. Metcalf, D. (1989b). Lancet 1, 885-887.

310

ELI KEDAR A N D EVA KLEIN

Meyskens, F. L., Kopecky, K., Samson, M., Hersh, E., Macdonald, J., Jaffe, H., Crowley, J., and Coltman, C. (1990).J. Natl. Cancer Inst. 82, 1071. Michl, J., Qiu, Q.-Y., and Kuerer, H. M. (1991). Cum Opin. Immunol. 3, 373-382. Mickiewicz, E., Estevez, R., Rao, F., Breier, S., Balbiani, L., Cabalar, E., Cazap, E., Cassini, J., Gerchunoff, M.,Temperley, G., and Capparos, C. (1990).Proc. Am. Sac. Clin. Oncol. 9, 281 (Abstr. 1089). Mier, J. W., Brandon, E. P., Libby, P., Janicka, M. W., and Aronson, F. R. (1989).J. Immunol. 143, 2407-2414. Mier, J. W., Vachino, G., Klempner, M. S., Aronson, F. R., Noring, R., Smith, S., Brandon, E. P., Laird, W., and Atkins, M. B. (1990). Blood 76, 1933-1940. Mihich, E. (1991). Cancer Res. 51, 745-747. Miller, R. L., Steis, R. G., Clark, J. W., Smith, J. W., Crum, E., McKnight, J. E., Hawkins, M. J., Jones, M. J., Longo, D. L., and Urba, W. J. (1989). CancerRes. 49, 1871-1876. Mitchell, M. S. (1988).J. Natl. Cancer Inst. 80, 1445-1450. Mitchell, M. S. (1990). Proc. Sac. Biol. Thm. 11 (Abstr. 25). Mitchell, M. S. (1991). Ink Rev. Immunol. 7 , 331-347. Mitchell, M. S., Kan-Mitchell, J., Kempf, R. A., Harel, W., Shau, H., and Lind, S. (1988a). Cancer Res. 48, 5883-5893. Mitchell, M. S., Kernpf, R. A., Harel, W., Shau, H., Boswell, W. D., Lind, S., and Bradley, E. C. (1988b).J. Clin. Oncol. 6, 409-424. Mitchell, M. S., Harel, W., Kempf, R. A., Hu, E., Kan-Mitchell, J., Boswell, W. D., Dean, G., and Stevenson, L. (1990).J. CZin. Oncol. 8, 856-869. Mitomi, T., and Noto, T. (1990). J. Cancer Res. Clin. Oncol. 116, Suppl. 11,872 (Abstr. 06.04.09). Mitomi, T., Tsuchiya, S., I@ma, N., Aso, K., Suzuki, K., Noto, T., and Ogawa, N. (1990).J . Cancer Res. Clin. Oncol. 116, Suppl. 1,189 (Abstr. A2.099.26). Mittelman, A., Chen, Z. J., Kageshita, T., Yang, H., Yamada, M., Baskind, P., Goldberg, N., Puccio, C., Ahrned, T., Arlin, Z., and Ferrone, S. (1990).J. Clin. Invest. 86, 2136-2144. Miyatake, S., Nishihara, K., Kikuchi, H., Yamashita, J., Namba, Y., Hanaoka, M., and Watanabe, Y.(1990).J. Natl. Cancer Inst. 82, 2 17-220. Mizel, S. B. (1989). FASEB J. 3, 2379-2388. Moertel, C. G. (1987). N . Engl. J. Med. 316, 274-275. Moertel, C. G., Fleming, T. R., MacDonald, J. S., Haller, D. G., Laurie, J. A., Goodman, P. J., Ungerleider, J. S., Emerson, W. A., Tormey, D. C., Click, J. H., Veeder, M. H., and Mailliard, J. A. (1990). N . Engl. J . Med. 322, 352-358. Mokyr, M. B., and Dray, S. (1987). Cancer Invest. 5, 31-38. Moller, P., Momburg, F., Koretz, K., Moldenhauer, G., Herfarth, C., Otto, H. F., Hammerling, G. J., and Schlag, P. (1991). Cancer Res. 51, 729-736. Momburg, F., Ziegler, A., Harpprecht, J., Moller, P., Moldenhauer, G., and Hammerling, G. J. (1989).J. Immunol. 142, 352-358. Monroy, R. L., Davis, T. A., and MacVittie, T. J. (1990). Clin. Immunol. Immunopathol. 54, 333-346. Moore, K. W., Vieira, P., Fiorentino, D. F., Trounstine, M. L., Khan, T. A,, and Mosmann, T. R. (1990). Science 248, 1230-1234. Moore, M. A. S. (1991). Annu. Rev. Immunol. 9, 159-191. Moore, M. A. S., Stolfi, R. L., and Martin, D. S. (1990).J. Natl. CancerInrt. 82, 1031-1037. Morecki, S., Topalian, S. L., Myers, W. W., Okrougly, D., Okarma, T. B., and Rosenberg, S. A. (1990).J. Biol. Response Modif. 9, 463-474. Morecki, S., Karson, E., Cornetta, K., Kasid, A., Aebersold, P., Blaese, R. M., Anderson, W. F., and Rosenberg, S. A. (1991). Cancer Immunol. Zmmunother. 32, 342-352.

CANCER IMMUNOTHERAPY

31 1

Morikawa, K., Fan, D., Denkins, Y. M., Levin, B., Gutterman, J. U., Walker, S. M., and Fidler, I. J. (1989). Cancer Res. 49, 799-805. Morton, D., Nizze, A., Foshag, L., Hoon, D., Famatiga, E., Gupta, R., and Irie, R. (1990). Proc. Am. Assoc. Cancer Res. 31, 281 (Abstr. 1664). Morton, D. L., Hoon, D. S. B., Foshag, L. J., Nizze, J. A., Gupta, R. K., Famatiga, E., and Irie, R. F. (1991a). Proc. Am. Assoc. Cancer Res. 32, 492-494. Morton, D. L., Hoon, D., Nizze, A,, Foshag, L., Gupta, R., and Irie, R. (1991b). Proc. ASCO 10, 295 (Abstr. 1033). Mueller, B. M., and Reisfeld, R. A. (1990). Proc. AACR 31, 299 (Abstr. 1775). Mukherji, B., Chakraborty, N. G., and Sivanandham, M. (1990). Immunol. Rev. 116, 33-62. Mulder, N. H., Willemse, P. H. B., Schraffordt Koops, H., de Vries, E. G. E., and Sleijfer, D. T. (1990). Br. J. Cancer 62, 1006-1007. Mulder, N. H., Schraffordt Koops, H., Sleijfer, D. T., de Vries, E. G. E., and Willemse, P. H. B. (1991). Proc. Am. SOC.Clzn. Oncol. 10, 292 (Abstr. 1020). MulC, J. J., Yang, J., Shu, S., and Rosenberg, S. A. (1986). J. Immunol. 136, 3899-3909. Mule, J. J., Yang, J. C., Lafreniere, R., Shu, S., and Rosenberg, S. A. (1987).J. Zmmunol. 139, 285-294. MulC, J. J., Asher, A., McIntosh, J., Lafeniere, R., Shiloni, E., Lefor, A., Reichert, C. M., and Rosenberg, S. A. (1988). Cancer Immunol. Immunother. 26, 202-208. MulC, J. J., McIntosh, J. K., Jablons, D. M., and Rosenberg, S. A. (199O).J. Exfi. Med. 171, 629-636. MulC, J. J., Jicha, D. L., Aebersold, P. M., Travis, W. D., and Rosenberg, S. A. (1991).J. Natl. Cancer Inst. 83, 350-355. Murray, J. L., Zukiwski, A. Z., Mujoo, K., and Rosenblum, M. G. (1990). J. Biol. Response ModiJ 9, 556-563. Musiani, P., De Campora, E., Valitutti, E., Castellino, F., Calearo, C., Cortesina, G., Giovarelli, M., Jemma, C., De Stefani, A., and Forni, G. (1989).J. Biol. ResfionseModg. 8, 571-578. Naito, K., Pellis, N. R., and Kahan, B. D. (1988). Cancer Res. 48, 101-108. Naito, K., Ueda, Y., Nomi, S., Yamagishi, H., Oka, T. Komichi, H., and Kahan, B. D. (1991). Proc. Am. Assoc. Cancer Res. 32, 337 (Abstr. 1408). Nakagami, Y. J., Kishimoto, T., and Matsumoto, K. (1990). J . Cancer Res. Clzn. Oncol. 116, Suppl. I, 541 (Abstr. B4.202.20) Nakajima, I., and Chu, T. M. (1991). Cancer Immunol. Immunother. 33, 9-14. Nakamura, S., Kashimoto, S., Fajikawa F., and Nakata, K. (1991). Cancer Res. 51, 215221. Nanbara, S., Arinaga, S., and Akiyoshi, T. (1989). Cancer Immunol. Immunother. 29, 237241. Naor, D., Klein, B. Y., Tarcic, N., and Duke-Cohen, J. S. (eds.). (1989). “Immunosuppression and Human Malignancy.” Humana Press, Clifton, New Jersey. Nassander, U. K. (1991). Ph.D. Thesis, University of Utrecht, T h e Netherlands. Natali, P. G., Nicotra, M. R., Bigotti, A,, Venturo, I., Marcenaro, L., Giacomini, P., and Russo, C. (1989). Proc. Natl. Acud. Scz. V.S.A. 86, 6719-6723. Negrier, S., Philip, T., Stoter, G., Fossa, S. D., Janssen, S., Iacone, A., Cleton, F. S., Eremin, O., Israel, L., Jasmin, C., Rugarli, C., Masse, H. V. D., Thatcher, N., Symann, M., Bartsch, H. H., Bergmann, L., Bijman, J. T., Palmer, P. A., and Franks, C. R. (1989). Eur. J. Cancer Clin. Oncol. 25, Suppl. 3, S21-S28. Nemunaitis, J., Rabinowe, S. N., Singer, J. W., Bierman, P. J., Vose, J. M., Freedman, A. S., Onetto, N., Gillis, S., Oette, D., Gold, M., Buckner, C. D., Hansen, J. A,, Ritz, J.,

312

ELI KEDAR A N D EVA KLEIN

Appelbaum, F. R., Armitage, J. O., and Nadler, L. M. (1991). N . Engl. J. Med. 324, 1773- 1778. Neta, R. (1988). Biotherapy 1, 41-45. Neta, R.,and Oppenheim, J. J. (1988). Blood 72, 1093-1095. Nicolson, G. L. (1984). Cancer Metastasis Rev. 3,25-42. Nii, A., Fan, D., and Fidler, I. J. (1991).J. Immunother. 10, 13-19. Nijhuis, E. W. P., Wiel-van Kemenade, E. V. D., Figdor, C. G., and van Lier, R. A. W. (1990). Cancer Zmmunol. Immunother. 32, 245-250. Nitta, T., Sato, K., Yagita, H., and Okumura, K. (1990). Lancet 335, 368-371. Nitta, T., Nakata, M., Yagita, H., and Okurnura, K. (1991). Immunol. Lett. 28, 31-38. Noguchi, K., Inagawa, H., Tsuji, Y., Morikawa, A., Mizuno, D. I., and Soma, G. I. (1991).J. Immunother. 10, 105- 1 1 1. Nordrnan, E., and Kellokumpu-Lehtinen, P. (199O).J. Cancer Res. Clin. Oncol. 116, Suppl. 1,302 (Abstr. A3.106.11). North, R. j.(1984). Adv. Immu?zol. 35, 89-155. North, R.J., and Havell, E. A. (1988).J. Exp. Med. 167, 1086-1099. North, R. J., Awwad, M., and Dunn, P. L. (1989). Prog. Immunol. 7, 1097-1 103. Notter, M., and Schirrmacher, V. (1990). Int. J . Cancer 45, 834-841. Nowak, J., Cohen, E. P., and Graf, L. H. (1991). Cancer Immunol. Immunother. 33, 91-96. Oberg, K., A h , G., Magnusson, A., Lundquist, G., Theodorsson, E., Wide, L., and Wilander, E. (1989).J. Natl. Cancer Imt. 81, 531-535. Ochoa, A. C., Conlon, K., Loeffler, C., Koop, W., Strobl, S., Warner, I., Smith, K., Urba, W. J., and Longo, D. L. (1991). Proc. Am. Assoc. Cancer Res. 32, 251 (Abstr. 1494). Oettgen, H. F., ed. (1990). Immunol. Allergy Clin. North Am. 10, no. 4. Oettgen, H. F., ed. (1991a). Zmmunol. AllPrgy Clin. North Am. 11, no. 2. Oettgen, H. F. (1991b). Cuw. Opzn. Zmmunol. 3, 699-705. Ogura, H., Tani, K., Ozawa, K., Nagata, S., Asano, S., and Takaku, F. (1990). Cancer Res. 5095102-5106. Old, L. J. (1981). Cancer Reg. 41, 361-375. Old, L. J., ed. (1991). Cuw. Opin. Immunol. 3, 643-707. Oldam, R. K. (1983).J. Natl. Cancer Inst. 70, 789-795. Oldham, R. (1986). J. Cell. Physiol., Suppl. 4, 91-99. Oldam, R. K. (ed.) (1991). “Principles of Cancer Biotherapy”, 2nd edition. Marcel Dekker, Inc., New York. Oliver, R. T. D. (1988). Br. J. Cancer 58, 405-409. Oliver, R. T. D. (1991). Eur.J. Cancer 27, 1168-1172. Onodera, H., Somers, S. S., and Guillou, P. J. (1990). Br. J. Cancer 62, 1042-1046. Ortaldo, J. R.,and Wiltrout, R. H. (199O).J. Natl. CancerZnst. 82, 1663-1665. Ortaldo, J. R., Porter, H. R.,Miller, P., Stevenson, H. C., Ozols, R. F., and Hamilton, T. C. ( 1986). Cancer Res. 46, 44 14-44 19. Osband, M. E., and Ross, S. (1990). Immunol. Today 11, 193-195. Osband, M. E., Hamilton, D., Shen Y. J., Cohen, E., Shlesinger, M., Lavin, P., Brown, A,, and McCaffrey, R. (1981). Lancet 1, 636-638. Osband, M. E., Lavin, P. T., Babayan, R. K., Graham, S., Lamm, D. L., Parker, B., Sawczuk, I., Ross, S., and Krane, R. J. (1990). Lancet 335, 994-998. Osband, M., Ross, S., Celis, E., Lavin, P., and Krane, R. (1991). Proc. Am. Soc. Clin. Oncol. 10, 164 (Abstr. 519). Ostrand-Rosenberg, S., Thakur, A., and Clements, V. (199O).J. Zmmunol. 144,4068-4071. Otto, U., Conrad, S., Schneider, A. W., Kemperi, J., Schlick, E., and Klosterhalfen, H. (199O).J. Cancer Res. Clin. Oncol. 116, Suppl. II,1057(Abstr. 16.11.05).

CANCER IMMUNOTHERAPY

313

Ottow, R. T., Steller, E. P., Sugarbaker, P. H., Wesley, R. A,, and Rosenberg, S. A. (1987). Cell. Immunol. 104, 366-376. Paciucci, P. A,, Ryder, J. S., and Holland, J. F. (1990). Proc. Am. SOC. Clzn. Oncol. 9, 201 (Abstr. 777). Pai, L. H., Gallo, M. G., FitzGerald, D. J., and Pastan, I. (1991). CancerRes. 51,2808-2812. Palladino, M.A., Welte, K., Carroll, A. M., and Oettgen H. F. (1984). Cell. Immunol.86, 299-307. Papa, M. Z., Vetto, J. T., Ettinghausen, S. E., Mule, J. J., and Rosenberg, S. A. (1986).Cancer Res. 46, 56 18-5623. Papa, M. Z., Yang, J. C., Vetto, J. T., Shiloni, E., Eisenthal, A., and Rosenberg, S. A. (1988). Cancer Res. 48, 122-129. Papahadjopoulos, D., and Gabizon, A. (1987). Ann. N.Y. Acad. Scz. 507, 64-74. Papahadjopoulos, D., and Gabizon, A. (1990). Prog. Bzol. Clin. Res. 343, 85-93. Papahadjopoulos, D., Allen, T. M., Gabizon, A., Mayhew, E., Matthay, K., Huang, S. K., Lee, K. D., Woodle, M. C., Lasic, D. D., Redemann, C., and Martin, F. J. (1991). Proc. Natl. Acad. Sci. U.S.A. 88, 11460-11464. Parkinson, D. R. (1990).J. Natl. Cancer Iwt. 82, 1374-1376. Parkinson, D. R., Fisher, R. I., Rayner, A. A,, Paietta, E., Margolin, K. A,, Weis, G. R., Mier, J. W., Sznol, M., Gaynor, E. R., Bar, M. H., Gucalp, R., Boldt, D. H., Mills, B., and Hawkins, M. J. (1990a).J. Clin. Oncol. 8, 1630-1636. Parkinson, D. R., Abrams, J. S., Wiernik, P. H., Rayner, A. A., Margolin, K. A., Van Echo, D. A,, Sznol, M., Dutcher, J. P., Aronson, F. R., Doroshow, J. H., Atkins, M. B., and Hawkins, M. J. (1990b).J. Clzn. Oncol. 8, 1650-1656. Parmiani, G. (1990).Immunol. Toduy 11, 113-115. Parmiani, G., Anichini, A,, and Fossati, G. (1990).J. Natl. Cancer Inst. 82, 361-370. Pastan, I., and FitzGerald, D. (1991). Science 254, 1173-1 177. Paul, N. L., and Ruddle, N. H. (1988). Annu. Rev. Immunol. 6, 407-438. Paul, W. E. (1989). Cell (Cambridge, Mass.) 57, 521-524. Pazdur, R., Ajani, J. A,, Patt, Y. Z., Winn, R., Jackson, D., Shepard, B.,*DuBrow, R., Campos, L., Quaraishi, M., Faintuch, J., Abbruzzese, J. L., Gutterman, J., and Levin, B. ( 1990).J. Clin. Oncol. 8, 2027-203 1. Peace, D. J., and Cheever, M. A. (1989).J. Exp. Med. 169, 161-173. Peace, D. J., Chen, W., Nelson, H., and Cheever, M. A. (1991).J.Immunol. 146,2059-2065. Pedrazzoli, P., Zibera, C., Gibelli, N., Cazzola, M., and Robustelli della Cuna, G. (1990).J. Cancer Res. Clzn. Oncol. 116, Suppl. 1,313 (Abstr. A3.108.06). Pendurthi, T. K., Schlom, J., and Primus, F. J. (1991).J. Immunother. 10, 2-12. Phillips, N. C. (1989). Cancer Metastasis Rev. 8, 23 1-252. Phillips, N. C., Chedid, L., Bernard, J. M., Level, M., and Lefrancier, P. (1987). J . B i d . Response Mod$ 6, 678-691. Phillips, N. C., Loutfi, A,, A-Kareem, A. M., Shibata, H. R., and Baines, M. G. (1990). Cancer Detect. Prev. 14, 491-496. Pichert, G., Jost. L. M., Fierz, W., and Stahel, R. A. (1991). Br. J. Cancer 63, 287-292. Pizza, G., Severini, G., Menniti, D., De Vinci, C., and Corrado, F. (1984). 1nt.J. Cancer 34, 359-369. Porwit, A., Palucka, K., and Reizenstein, P. (1990). Leuk. Lymphoma 2, 171-178. Poste, G. (1986). Cancer Treat. Rep. 70, 183-199. Postmus, P. E., Gietema, J. A., Damsma, O., Willemse, P. H. B., d e Vries, E. G. E., and Vellenga, E. (1991). Proc. Am. SOC.Clin. Oncol. 10, 248 (Abstr. 851). Pretlow, T. P., Keith, E. F., Keith Crayer, A., Bartolucci, A. A,, Pitts, A. M., Pretlow, T. G., Kimball, P. M., and Boohaker, E. A. (1983). Cancer Res. 43, 2997-3000.

314

ELI KEDAR A N D EVA KLEIN

Pujade-Lauraine, E., Guastella, J. P., Colombo, N., Namer, N., Fumoleau, P., Monnier, A,, Nooy, M. A., Falkson, G., Mignot, L., Bugat, R., Oliveira, C. M. D., Mousseau, M., Netter, G., Oberling, F., Coiffier, B., and Brandely, M. (1991).Proc. Am. Soc. Clin. Oncol. 10, 195 (Abstr. 642). Puri, R. K., Travis, W. D., and Rosenberg, S. A. (1989). Cancer Res. 49,969-976. Puri, R. K., Travis, W. D., and Rosenberg, S. A. (1990). Cancer Res. 50, 5543-5550. Puri, R. K., Ogata, M., Leland, P., Feldman, G. M., FitzGerald, D., and Pastan, I. (1991). Cancer Res. 51, 3011-3017. Quirt, I. C., and Tannock, I. F. (199O).J. Clin. Oncol. 8, 1125-1127. Radrizzani, M., Gambacorti-Passerini, C., Parmiani, G., and Fossati, G. (1989). Cancer Immunol. Immunother. 28, 67-73. Rahman, A., Panneerselvam, M., Guirguis, R., Castronovo, V., Sobel, M. E., Abraham, K., Daddona, P. E., and Liotta, L. A. (1989).J. Natl. Cancer Inst. 81, 1794-1800. Ramakrishna, V., and Shinitzky, M. (1991). Cancer Immunol. Immunother. 33, 1-8. Ranade, V. V. (1989). 1. Clin. Phannacol. 29, 685-694. Ranges, G. E., Figari, 1. S., Espevik, T., and Palladino, M. A. (1987).J. Exp. Med. 166,991998. Ransom, J. H., Pelle, B., and Hanna, M. G. (1991). Proc. Am. Assoc. Cancer Res. 32, 255 (Abstr. 1516). Rath, U., Kaufmann, M.,Schmid, H., Hofmann, J., Wiedenmann, B., Kist, A., Kempeni, J., Schlick, E., Bastert, G., Kommerell, B., and Mannel, D. (1991). Eur. J. Cancer. 27, 12 1-125. Redman, B. G., Flaherty, L., Chou, T. H., Al-Katib, A., Kraut, M., Martino, S., Chen, B., Kaplan, J., and Valdivieso, M. (199O).J. Clin. Oncol. 8, 1269-1276. Rees, R. C., and Wiltrout, R. H. (1990). Immunol. Today 11, 36-38. Reid, I., Lundy, J., Nelson, H., and Donohue, J. (1991a).Proc. Am. Assoc. CancerRes. 32,259 (Abstr. 1543). Reid, I., Lundy, J., Ramsey, P., Nelson, H., and Donohue, J. (1991b).Proc. Am. Assoc. Cancer Res. 32, 263 (Abstr. 1567). Reynolds, J. V., Somers, S., and Guillou, P. J. (1990). Br. J. Cancer 62, Suppl.11, 17 (Abstr. 039). Richards, J. M., Priest, E., Hamasaki, V., Skosey, P., and Ramming, K. (1991).Proc. Am. SOC. Clin. Oncol. 10, 298 (Abstr. 1046). Richman, C. M., Slapak, C. A., and Toh, B. (1990).J. B i d . Respome Modif. 9, 570-575. Rinehart, J. J., Triozzi, P. L., Lee, M. H., and Aldrich, W. (1990).Proc. Am. Assoc. CancerRes. 31, 277 (Abstr. 1641). Rivoltini, L., Gambacorti-Passerini, C., Supino, R., and Parmiani, G. (1989). Int. J. Cancer 43,880-885. Rivoltini, L., Gambacorti-Passerini, C., Squadrelli-Saraceno, M., Grosso, M. I., Cantu, G., Molinari, R., Orazi, A., and Parmiani, G. (1990). Cancer Res. 50, 5551-5557. Robertson, M. J., and Ritz, J. (1990). Blood 76, 2421-2438. Robins, R. A,, and Baldwin, R. W. (1985). Immunol. Today 6, 55-58. Rodolfo, M., Salvi, C., Bassi, C., and Parmiani, G. (1990). CancerImmunol. fmmunother. 31, 28-36. Rosenberg, M. B., Breakefield, X. O., and Hawrot, E. (1987).J. Neurochem. 48, 865-875. Rosenberg, S. A. (1984). Cancer Treat. Rep. 68, 233-255. Rosenberg, S. A. (1986). In “Important Advances in Oncology 1 9 8 6 (V. T, DeVita, S. Hellman, and S. A. Rosenberg, eds.), pp. 55-91. Lippincott, Philadelphia, Pennsylvania. Rosenberg, S. A. (1988). Immunol. Today 9, 58-62.

CANCER IMMUNOTHERAPY

315

Rosenberg, S. A. (1990). Sci. Am. 262, 34-41. Rosenberg, S. A. (1991a). Cancer Res. (Suppl.) 51, 5074s-5079s. Rosenberg, S. A. (1991b). Proc. Adv. Biol. Clin. Management Melanoma (Abstr.) p. 61. Houston. Rosenberg, S. A., Lotze, M. T., Muul, L. M., Leitman, S., Chang, A. E., Ettinghausen, S. E., Matory, Y. L., Skibber, J. M., Shiloni, E., Seipp, C. A., Simpson, C., and Reichert, C. M. (1985a). N . Eng1.J. Med. 313, 1485-1492. Rosenberg, S. A,, MulC, J. J., Spiess, P. J., Reichert, C. M., and Schwartz, S. L. (1985b).J. Exp. Med. 161, 1169-1183. Rosenberg, S. A., Spiess, P., and Lafreniere, R. (1986). Science 233, 1318-1321. Rosenberg, S. A., Lotze, M. T., Muul, L. M., Chang, A. E., Avis, F. P., Leitman, S., Linehan, M., Robertson, C. N., Lee, R. E., Rubin, J. T., Seipp, C. A., Simpson, C. G., and White, D. E. (1987). N . Engl. J. Med. 316, 889-897. Rosenberg, S. A., Packard, B. S., Aebersold, P. M.,Solomon, D., Topalian, S. L., Toy, S. T., Simon, P., Lotze, M. T., Yang, J. C., Seipp, C. A,, Simpson, C., Carter, C., Bock, S., Schwartzentruber, D., Wei, J. P., and White, D. E. (1988). N . Eng1.J. Med. 319, 16761680. Rosenberg, S. A., Longo, D. L., and Lotze, M. T. (1989a). In “Cancer: Principles and Practice of Oncology” (V. T. DeVita Jr., S. Hellman, and S. A. Rosenberg, eds.), 3rd ed. pp. 302-347. Lippincott, Philadelphia, Pennsylvania. Rosenberg, S. A., Lotze, M. T., Yang, J. C., Aebersold, P. M., Linehan, W. M., Seipp, C. A., and White, D. W. (1989b). Ann. Surg. 210, 474-485. Rosenberg, S. A., Lotze, M. T., Yang, J. C., Linehan, W. M., Seipp, C., Calabro, S., Karp, S. E., Sherry, R. M., Steinberg, S., and White, D. E. (1989c).J. Cldn. Oncol. 7,1863-1874. Rosenberg, S . A., Aebersold, P., Cornetta, K., Kasid, A,, Morgan, R. A., Moen, R., Karson, E. M.,Lotze, M. T., Yang, J. C., Topalian, S. L., Merino, M. J., Culver, K., Miller, A. D., Blease, R. M., and Anderson, W. F. (1990). N . Engl. J. Med. 323, 570-578. Rosenblum, M.,and Donato, R. J. (1989). CRC Crit. R w . Zmmunol. 9, 21-44. Rubin, J. T., Elwood, L. J., Rosenberg, S. A., and Lotze, M. T. (1989). Cancer Res. 49,70867092. Ruiter, D. J., Mattijssen, V., Broecker, E. B., and Ferrone, S. (1991). Semin. Cancer Biol. 2, 35-45. Ruiz-Cabello, F., Klein, E., and Garrido, F. (1991). Zmmunol. Lett. 29, 181-190. Russel, C., Romics, I., and Bach, D. (1990).J . Cancer Res. Clin. Oncol. 116, Suppl. I, 540 (Abstr. B4.209.16). Russell, S. J. (1990). Immunol. T&y 11, 196-200. Riither, U., Schmidt, A., and Bub, P. (1991). Proc. Am. SOC.Clin. Oncol. 10, 177 (Abstr. 573). Sakai, K., Chang, A. E., and Shu, S. (1990). Cell. Zmmunol. 129, 241-255. Sariban, E.,Imamura, K., Luebbers, R., and Kufe, D. (1988).J. Clin. Invest. 81, 1506-1510. Sarneva, M.,Vujanovic, N. L., Van Den Brink, M.R. M., Herberman, R. B., and Hiserodt, J. C. (1989). Cell. Immunol. 118, 448-457. Sarosdy, M. F., and Lamm, D. L. (1989). J. Urol. 142, 719-722. Sasaki, A., Melder, R. J., Whiteside, T. L., Herberman, R. B., and Jain, R. K. (1991).J. Natl. Cancer Imt. 83,433-437. Sato, G . H., and Sato, J. D. (1989).J. Natl. CancerZnst. 81, 1600-1601. Sawada, T., Iwai, M., and Furuta, I. (199O).J. Cancer Res. Clin. Oncol. 116, Suppl. I, 292 (Abstr. A3.103.77). Sayers, T. J., Wiltrout, T. A., McCormick, K., Husted, C., and Wiltrout, R. H. (1990). Cancer Res. 50, 54 14-5420.

316

ELI KEDAR A N D EVA KLEIN

Scadden, D. T., Feder, D., and Groopman, J. E. (1991). Curr. Opin. Cell B d . 1, 301303. Scala, S., Murphy, L. D., Tortora, G., ChoChung, Y., and Bates, S. (1991).Proc. Am. Assoc. Cancer Res. 34, 377 (Abstr. 2240). Scalzo, S., Gengaro, A., Boccoli, G., Masciulli, R., Giannella, G., Salvo, G., Marolla, P., Carlini, P., Massirnini, G., Holdener, E. E., Testa, U., Calabresi, F., and Peschle, C. ( 1990). Eur. J. Cancer 26, 1152- 1156. Scheibenbogen, C., Keilholz, U., Mytilineos,J., Manasterski, U., Tilgen, W., and Hunstein, W. (1991).Proc. Sac. Biol. Ther. (Abstr.) 19. Schirrmacher, V. (1989). Interdwcip. Sci. Rev. 14, 291-303. Schirrmacher, V. (1991). Cancer Sum. 13 (in press). Schirrmacher, V., Ahlert, T., Heicappell, R., Appelhans, B., and von Hoegen, P. (1986). Cancer Rev. 5, 19-49. Schirrmacher, V., Hoegen, P. V., Schlag, P., Liebrich, W., Lehner, B., Schumacher, K., Ahlert, T., and Bastert, G. (1989). In “Cancer Metastasis” (V. Schirrmacher and R. Schwartz-Albeiz, eds.), pp. 157- 170. Springer-Verlag, Berlin. Schmid, H., Kaufrnann, M., Grischke, E. M., Kempeni, J., and Bastret, G. (199O).J. Cancer Res. Clzn. Oncol. 116, Suppl.11, 1058 (Abstr. 16.11.07). Schmidt-Wolf, I. G. H., Negrin, R. S., Kiem, H.-P., Blume, K. G., and Weissman, I. L. (1991).J. Exp. Med. 174, 139-149. Schnipper, L. E. (1986). N. Eng1.J. Med. 314, 1423-1431. Schoof, D. D., Selleck, C. M., Massaro, A. F., Jung, S. E., and Eberlein, T. J., (1990).Cancer Res. 50, 1138-1143. Schreiber, H. (1989). In “Fundamental Immunology” (W. E. Paul, ed.), 2nd ed. pp. 923955. Raven Press, New York. Schroit, A. J., Madsen, J., and Nayer, R. (1986). Chem. Phys. Lipids 40, 373-393. Schuchter, L. M., Esa, A. H., May, W. S., Laulis, M. K., Pettit, G. R., and Hess, A. D. (1991). Cancer Res. 51, 682-687. Schulof, R. S., Mai, D., Nelson, M. A., Paxton, H. M., Cox, J. W., Turner, M. L., Mills, M., Hix, W. R., Nochomovitz, L. E., Peters, L. C., and Hanna, M. G. (1988).Mol. Bzother. 1, 30-36. Schultz, K. R., Klarnet, J. P., Peace, D. J., Cheever, M. A., Badger, C. C., Bernstein, I. D., and Greenberg, P. D. (1990). Cancer Res. 50,5421-5425. Schultz, K. R., Badger, C. C., Dombi, G. W., Greenberg, P. D., and Bernstein, I. D. (1991). Proc. Am. Assoc. Cancer Res. 32, 264 (Abstr. 1572). Schwartzentruber, D. J., Zweigg, M. H., Weintraub, B. D., White, D. E., and Rosenberg, S. A. (1990). Proc. ASCO 9, 184 (Abstr. 71 1). Scott, 0. C. A. (1991). Cancer Res. 51, 757-763. Segal, D. M., Garrido, M. A., Qian, J. H., Mezzananica, D., Andrew, S., Perez, P., Kurucz, I., Valdayo, M. J., Titus, J. A., Winkler, D. F., and Wunderlich, J. R. (1990). Mol. Inamunol. 27, 1339- 1342. Semenzato, G. (1990a). Leukemza 4, 71-80. Semenzato, G. (1990b).Br. J. Cancer 61, 354-36 1. Sencer, S., Rich, M., and Anderson, P. (1991).Proc. Am. Assoc. Cancer Res. 34, 255 (Abstr. 1515). Sensi, M., Bergami, M., Formelli, F., and Parmiani, G. (1990).Imt.J. Cancer46,1088-1094. Shank, W., and Balducci, L. (1991). Proc. Am. Sac. Cbn. Oncol. 10, 326 (Abstr. 1150). Shiloni, E., Pouillart, P., Janssens, J., Splinter, T., Di Peri, T., Symann, M., Roest, G. J., Palmer, P. A., and Franks, C. R. (1989). Eur. J. Cancer Clin. Oncol. 25, Suppl. 3, S 45s 49.

CANCER IMMUNOTHERAPY

317

Shimizu, Y., Iwatsuki, S., Herberman, R. B., and Whiteside, T. L. (1990). Int.1. Cancer 46, 878-883. Shimizu, Y., Iwatsuki, S., Herberman, R. B., and Whiteside, T. L. (1991). Cancer Immunol. Immunother. 32, 280-288. Shirnornura, K., Manda, T., Mukurnoto, S., Kobayashi, K., Nakano, K., and Mori, J. (1988). Int. J. Cancer 41, 243-247. Shin, D. M., Fidler, I. J., Bucana, C. D., Fan, D., Hong, W. K., and Killion, J. J. (1990).J. Biol. Response Mod$ 9, 355-360. Siegall, C. B., Kreitman, R. J., FitzGerald, D. J., and Pastan, I. (1991).Cancer Res. 51,28312836. Siegel, J. P., and Puri, R. K. (1991).J. Clin. Oncol. 9, 694-704. Sikorska, H. M. (1988).J. Biol. Response Modif 7 , 327-358. Silagi, S., and Schaefer, A. E. (1986).J. B i d . Response Modif. 5 , 41 1-422. Singh, M.,Chose, T., Faulkner, G., Kralovec, J., and Mezei, M. (1989). Cancer Res. 49, 3978-3984. Skornick, Y. G., Rong, G. H., Sindelar, W. F., Richert, L., Klausner, J. M., Rozin, R. R., and Shinitzky, M. (1986). Cancer (Philadelphia) 58, 650-654. Skornick, Y., Topalian, S., and Rosenberg, S. A. (1990).J. B i d . Response Mod$ 9,431-438. Slavin, S., and Kedar, E. (1988). Blood Rev. 2,259-269. Slavin, S., Weiss, L., Eckerstein, A., Kedar, E., Morecki, S., Samuel, S., Mumcuoglu, M., Hardan, I., Or, R., and Naparstek, E. (1989). Bone Marrow Transplant 4, 80-81. Slingluff, C. L., Darrow, T., Vervaert, C, Quinn-Allen, M. A., and Seigler, H. F. (1988).J. Natl. Cancer Inst. 80, 1016-1026. Smith, K. A. (1988). Science 240, 1169-1 175. Smith, J., Urba, W., Steis, R., Janik, J., Sharfman, W., Fenton, R., Conlon, K., Schoen, D., Sznol, M., Miller, L., Holrnlund, J., Creekmore, S., Shimuzu, M., Oppenheim, J., and Longo, D. (1991a). Proc. Am. SOC.Clin. Oncol. 10, 293 (Abstr. 1025). Smith, J. W'., Urba, W. J., Clark, J. W., Longo, D. L., Farrell, M., Creekmore, S. P., Conlon, K. C., Jaffe, H., and Steis, R. G. (1991b).J. Immunother. 10, 355-362. Smith, M. E. F., Marsh, S. G. E., Bodmer, J. G., Gelsthorpe, K., and Bodmer, W. F. (1989). Proc. Natl. Acad. Sci. U.S.A. 86, 5557-5561. Sohn, M., Levens, W., Riibben, H., Richter, R.,Rath, U., Kempeni, J., and Jakse, G. (1990). J . Cancer Res. Clzn. Oncol. 116, Suppl.11, 1058 (Abstr. 16.1 1.06). Solal-Celigny, P., Lepage, E., Browse, N., Peuchmaur, M., Haioun, C., Coiffier, B., Leporrier, M., Reyes, F., and Gisselbrecht, C. (1991). Proc. Am. SOC.Clin. Oncol. 10,275 (Abstr. 955). Sondel, P. M., and Hank, J. A. (1989). Cancer Invest. 7 , 437-442. Sosman, J. A., Kohler, P. C., Hank, J. A., Moore, K. H., Bechhofer, R., Storer, B., and Sondel, P. M. (1988).J. Natl. Cancer Inst. 80, 1451-1461. Sosman, J. A., Hank, J. A., and Sondel, P. M. (1990). Semin. Oncol. 17, Suppl.1, 22-30. Spiess, P. J., Yang, J. C., and Rosenberg, S. A. (1987).J. Natl. Cancer Inst. 79, 1067-1075. Spits, H., Yssel, H., Paliard, X., Kastelein, R., Figdor, C., and de Vries, J. E. (1988).J. Immunol. 141,29-36. Spitzer, T. R.,Swain, S. M., Lippman, M. E., and Deeg, H. J. (1991). CancerZnuest. 9,93-97. Sporn, M. B., and Roberts, A. B. (1990). Ann. N.Y. Acad. Sci. 593, 1-6. Sredni, B., Caspi. R. R., Klein, A., Kalechman, Y., Danziger, Y., Ben Yaakov, M., Tamari, T., Shalit, F., and Albeck, M. (1987). Nature (London) 330, 173-176. Sredni, B., Shani, A., Catane, R., Tichler, T., Gupwith, M. J., Burde, B., Haim, N., Levi, E., Kalechman, Y., Michlin, H., Schlesinger, M., Rozenzain, R., and Albeck, M. ( 1990).J . Cancer Res. Clin. Oncol. 116, Suppl. 1, 341 (Abstr. A3.112.54).

318

ELI KEDAR AND EVA KLEIN

Sredni, B., Tichler, T., Shani, A., Catane, R., Schlesinger, M., Horwith, G., Kirsh, T., Maida, B., Farbstein, M., Farbstein, H., and Albeck, M. (1991). Proc. Am. Assoc. Cancer Res. 32, 255 (Abstr. 1517). Srivastava, P. K. (1991). Curr. Ofin. Immuml. 3, 654-658. Stahel, R., Pichert, G., Jost, L., Ludwig, C., and Sess, C. (1990).J. Cancer Res. Clin. Oncol. 116, Suppl. 1, 286 (Abstr. A3.103.53). Starnes, H. F., Hartman, G., Torti, F., Carlson, R. W., Tewari, A., and Buhles, W. C. (1991). Proc. Am. SOC.Clin. Oncol. 10, 292 (Abstr. 1023). Steiniger, B., Falk, P., and Van der-Meide, P. H. (1988). Eur. J. Imnzunol. 18, 661-669. Steis, R. G., Urba, W. J., VanderMolen, L. A., Bookman, M. A., Smith, J. W., Clark, J. W., Miller, R. L., Crum, E. D., Beckner, S. K., McKnight, J. E., Ozols, R. F., Stevenson, H. C., Young, R. C., and Longo, D. L. (1990).J. Clin. Oncol. 8, 1618-1629. Stevenson, F. K., George, A. J. T., and Glennie, M. J. (1990). C h .Immunol. 48, 126-166. Stevenson, H. C., ed. (1989). “AdoptiveCellular Immunotherapy of Cancer.” Dekker, New

York. Stewart, J. A., Belinson, J. L., Moore, A. L., Dorighi, J. A., Grant, B. W., Haugh, L. D., Roberts, J. D., Albertini, R. J., and Branda, R. F. (1990). Cancer Res. 50, 6302-6310. Stohl, W., Tovar, Z., and Talal, N. (199O).J. Immunol. 144, 3718-3725. Stoter, G., Aamdal, S., Rodenhuis, S., Cleton, F. J., Iacobelli, S., Franks, C. R., Oskam, R., and Shiloni, E. (1991).J. Clin. Oncol. 9, 1687-1691. Stotter, H., Wiebke, E. A., Tomita, S., Belldegrun, A., Topalian, S., Rosenberg, S. A., and h z e , M. (1989).J. Immunol. 144, 1767-1773. Street, N. E., and Mosmann, T. R. (1991). FASEB J. 5, 171-177. Strom, T. B., Anderson, P. L., Rubin-Kelly, V. E., Williams, D. P., Kiyokawa, T., and Murphy, J. R. (1990). Sem. Immunol. 4,467-479. Sulitzeanu, D. (1985). Adv. Cancer Res. 44, 1-42. Sullivan, K. M., Storb, R., Buckner, C. D., Fefer, A., Fisher, L., Weiden, P. L., Witherspoon, R. P., Appelbaum, F. R., Banaji, M., Hansen, J., Martin, P., Sanders, J. E., Singer, J., and Thomas, E. D. (1989). N.EngL. J. Med. 340,828-834. Suzuki, S., Ohta, S., Takashio, K., Nitanai, H., and Hashimoto, Y. (1990).Int.J. Cancer 46, 1095-1 100. Swain, S. L. (1991). Curr. Ofin. ImmunoL. 3, 304-310. Sznol, M., Mier, J. W., Sparano,J., Gaynor, E. R.,Weiss, G. R., Margolin, K. A., Bar, M. H., Hawkins, M. J., Atkins, M. B., Dutcher, J. P., Fisher, R. I., Boldt, D. H., Doroshow,J. H., Louie, A., and Aronson, F. R. (199O).J. B i d . Respome Modif. 9, 529-537. Tada, T., Ohzeki, S., Utsumi, K., Takiuchi, H., Muramtsu, M., Li, X. F., Shimizu, J., Fujiwara, H., and Hamaoka, T. (1991).J. Immunol. 146, 1077-1082. Talarico, D., Ittmann, M., Balsari, A., Delli-Bovi, P., Basch, R.S., and Basilico, C., (1990). Proc. Natl. Acad. Sci. U.S.A. 87,4222-4225. Talmadge, J. E. (1988). Prog. Exp. Tumor RR(. 32, 154-173. Talmadge, J. E., Phillips, H., Schindler, J.. Tribble, H., and Pennington, R. (1987). Cancer Res. 47,5725-5732. Talmadge, J. E., Phillips, H., Schneider, M., Rowe, T., Pennington, R., Bowersox, O., and Lenz, B. (1988). Cancer Res. 48,544-550. Tanaka, K., Yoshioka, T., Bieberich, C., and Jay, G. (1988).Annu. Rev.Immunol. 6,359-380. Tanaka, H., Satake-Ishikawa, R., Ishikawa, M., Matsuki, S., and Asano, K. (1991). Cancer Res. 51, 3710-3714. Tecce, R., Digiesi, G., and Natali, P. G. (1991). Proc. Am. Assoc. CancmRes. 34, 260 (Abstr. 1545).

CANCER IMMUNOTHERAPY

3 19

Tepper, R. I., Pattengale, P. K., and Leder, P. (1989). Cell (Cambridge, Mass.)57, 503-512. Thiele, R., Pomer, S., Staeler, G., and Schirrmacher, V. (199O).J. Cancer Res. Clin. Oncol. 116, Suppl. 1, 80 (Abstr. B1.019.01). Thompson, J. A., Peace, D. J., Klarnet, J. P., Kern, D. E., Greenberg, P. D., and Cheever, M. A. (1986).J. Zmmunol. 137, 3675-3680. Thompson, J., Lee, D. J., Lindgren, C. G., Benz, L. A., Collins, C., Shuman, W. P., Levitt, D., and Fefer, A. (1989). Cancer Res. 49,235-240. Thompson, J. A., Benyunes, M., Benz, L., Lindgren, C., and Fefer, A. (1991a).Proc. Am. SOC.Clin. Oncol. 10, 179 (Abstr. 581). Thompson, J. A,, Lindgren, C. G., Benz, L. A,, Benyunes, M., and Fefer, A. (1991b).Proc. Am. Assoc. Cancer Res. 32, 270 (Abstr. 1609). Tichler, T., Shani, A., Catane, R., Brenner, H., Schlesinger, M., Kalechman, Y., Farbstein, H., Albeck, M., and Sredni, B. (1990). Proc. Am. SOC.Clin. Oncol. 9, 198 (Abstr. 765). Titus, J. A., Perez, P., Kaubisch, A., Garrido, M. A,, and Segal, D. M. (1987).J. Immunol. 139, 3153-3158. Tomazic, V. J., Farha, M., Loftus, A, and Elias, E. G. (1988).J. Immunol. 140, 4056-4061. Tomita, Y., Nishiyama, T., Fujiwara, M., and Sato, S. (1990). Br. J. Cancer 62, 354-359. Topalian, S. L., Solomon, D., and Rosenberg, S. A. (1989).J. Immunol. 142, 3714-3725. Topalian, S. L., Kasid, A., and Rosenberg, S. A. (1990).J. Zmmunol. 144, 4487-4495. Torisu, M., Hayashi, Y., Inhimitsu, T., Fujimura, T., Iwasaki, K., Katano, M., Yamamoto, H., Kimura, Y.,Takesue, M., Kondo, M., and Nomoto, K. (1990). Cancer Immunol. Immunother. 81,261-268. Uchida, A., and Mizutani, M. (1989). In “New Horizons in Tumor Immunotherapy” (Y. Torisu and T. Yoshida, eds.), pp. 201-213. Elsevier, Amsterdam. Uchida, A., Moore, M., and Klein, E. (1987). Int. J. Cancer 40, 165-170. Uchida, A., Kariya, Y., Okamoto, N., Sugie, K., Fujimoto, T., and Yagita, M. (199O).J.Natl. Cancerlmt. 82, 1697-1701. Uchida, A,, Kariya, Y., Okamoto, N., Sugie, K., and Fujimoto, T. (1991). Proc. Am. A550C. Cancer Res. 32, 237 (Abstr. 141 1). Urba, W. J., Clark, J. W., Steis, R. G., Bookman, M. A., Smith, J. W., Beckner, S., Maluish, A. E., Rossio, J. L., Rager, H., Ortaldo, J. R., and Longo, D. L. (1989).J.Natl. Cancerlnst. 81,602-611. Urba, W. J., Hartmann, L. C., Longo, D. L., Steis, R. G., Smith, J. W., Kedar, I., Creekmore, S., Sznol, M., Conlon, K., Kopp, W. C., Huber, C., Herold, M., Alvord, W. G., Snow, S., and Clark, J. W. (1990). Cancer Res. 50,2979-2986. Vaage, J, (1987). Cancer Res. 47, 4296-4298. Vaage, J. (1988). Cancer Res. 48,2193-2197. Van den Eynde, B., Hainaut, P., Herin, M., Knuth, A., Lemoine, C., Weynants, P., van der Bruggen, P., Pauchet, R., and Boon, T. (1989). Int. J. Cancer 44, 634-640. Van der Bruggen, P., Traversari, C., Chomez, P., Lurquin, C., De Plaen, E., Van den Eynde, B., Knuth, A., and Boon, T. (1991). Science 254, 1643-1647. Van Dijk, J.. Beniers, A. J. M. C., Fleuren, G. J., Schalken, J. A., and Warnaar, S. 0.(1991). Proc. Am. Assoc. Cancer Res. 32, 270 (Abstr. 1604). Van Duinen, S. G., Ruiter, D. J., Broecker, E. B., van der Velde, E. A., Sorg, C., Welvaart, K., and Ferrone, S. (1988). Cancer Res. 48, 1019-1025. Vanhaesebroeck, B., Mareel, M., Van Roy, F., Grooten, J., and Fiers, W. (1991). CancerRes. 51,2229-2238. Vinky, F. (1986). Biochim. Biophys. Acta 865, 253-265. VAnky, F., and Klein, E. (1982a). Cancer Zmmunol. Immunother. 14, 73-77.

320

ELI KEDAR AND EVA KLEIN

Vanky, F., and Klein, E. (1982b). Int. J. Cancer 29, 547-553. Vanky, F., and Klein, E. (1989). Cancer Immunol. Immunother. 29, 125-132. Vhnky, F., and Klein, E. (1991). Semin. Cancer Biol. 2, 55-62. Vanky, F., Klein, E. and Stjernsward, J. (1976). In “Clinical Tumor Immunology” (J. Wybran and M. Staquet, eds.), pp. 55-58. Pergamon, Oxford. Vanky, F., Gorsky, T., Gorsky, Y., Masucci, M. G., and Klein, E. (1982).J. Exp. Med. 155,8395. Vanky, F., Willems, J., Kreicbergs, A,, Aparisi, T., AndrCen, M., Brostrom, L. A,, Nilsonne, U., Klein, E., and Klein, G. (1983a). Cancerlmmunol. Immunother. 16, 11-16. Vanky, F., Peterffy, A., Book, K., et al. (1983b). Cancer Immunol. Immunother. 16, 17-22. Vanky, F., Klein, E., Willems, J., Book, K., Ivert, T., P&erffy, A., Nilsonne, U., Kreicbergs, A., and Aparisi, T. (1986). Cancer Immunol. Immunother. 21, 69-76. Vanky, F., Klein, E., Willems, J., Book, K., Ivert, T., Pkterffy, A., Nilsonne, U., and Kreicbergs, A. (1987). Cancer Immunol. I m m u n o t k . 24, 180-183. Vanky, F., Stuber, G., Willems, J., Sjowall, K., Larsson, B., B%k, K., Ivert, T., Peterffy, A., and Klein, E. (1988). Cancer Immunol. Immunother. 27, 213-222. Vanky, F., Stuber, G., Rotstein, S., and Klein, E. (1989). Cancer Immunol. Immunother. 28, 1721. Vanky, F., Wang, P., Patarroyo, M., and Klein, E. (1990). Cancerlmmunol. Immunother. 31, 19-27. Van Seventer, G. A., Shimizu, Y., and Shaw, S. (1991). C u m Opin. Immunol. 3, 294-303. Vellenga, E., Biesma, B., Meyer, C., Wagteveld, L., Esselink, M., and d e Vries, E. G. E. (1991). Cancer Res. 51, 73-76. Vetto, J. T., Papa, M. Z., Lotze, M. T., Chang, A. E., and Rosenberg, S. A. (1987).J. Clin. Oncol. 5, 496-503. Vokes, E. E., Weichselbaum, R. R., Ratain, M. J., McEvilly, J. M., Ervin, T. J., Wenig, B., Berezin, F., Haraf, D., Schilsky, R. L., Kozloff, M. F., and Panje, W. R. (1991). Proc. Am. Soc. Clin. Oncol. 10, 208 (Abstr. 693). Vose, B. M., and Bonnard, G. D. (1982). Nature (London) 296, 359-361. Vose, B. M., and Moore, M. (1985). Semin. Hematol. 22, 27-40. Vose, B. M., Vanky, F., and Klein, E. (1977). Int. J. Cancer 20, 512-519. Vosika, G. J., Cornelius, D. C., Gilbert, C. W., Bennek, J. A., Sadlik, J. R., and Doyle, A. (1990). Proc. Am. SOC.Clin. Oncol. 9, 199 (Abstr. 770). Vuist, W. M. J., Buitenen, F. V., Hekman, A,, and Melief, C. J. M. (1990). CancerRes. 50, 5767-5772. Wadler, S. (1991). Semin. Oncol. 18, Suppl. 1, 27-38. Wadler, S., and Schwartz, E. L. (1990). Cancer Res. 50, 3473-3486. Wadler, S., Schwartz, E. L., Goldman, M., Lyver, A., Rader, M., Zimmerman, M., Itri, L., Weinberg, V., and Wiernik, P. H. (1989). J. Clin. Oncol. 7 , 1769-1775. Wadler, S., Lembersky, B., Kirkwood, J., Atkins, M., and Petrelli, N. (1991). Proc. Am. SOC. Clzn. Oncol. 10, 136 (Abstr. 41 1). Wahl, S . M., McCartney-Francis, N., Allen, J. B., Dougherty, E. B., and Dougherty, S. F. (1990). Ann. N.Y. Acad. Sci. 593, 188-196. Waller, E. K., Lieberman, M., Hansteen, G., Sen, A., and Weissman, I. (1990). Proc. Am. Assoc. Cancer Res. 31, 300 (Abstr. 1777). Wang, P., Wnky, F., Li, S.-L., Vkgh, Z. S., Persson, U., and Klein, E. (1991). Int. J. Cancer SUPPI.6, 106-1 16. Watanabe, N., Niitsu, Y., Yamauchi, N., Umeno, H., Sone, H., Neda, H., and Urushizaki, I. (1988). J. Biol. Response Modif. 7 , 24-31. Watanabe, Y., and Iwa, T. (1987).J. Biol. Response Mod$ 6, 169-180.

CANCER IMMUNOTHERAPY

32 1

Watanabe, Y., Kuribayashi, K., Miyatake, S., Nishihara, K., Nakayama, E. I., Taniyama, T., and Sakata, T. A. (1989).Proc. Natl. Acud. Sci. U.S.A. 86, 9456-9460. Weber, J. S., and Rosenberg, S. A. (1988).Cancer Res. 48, 5818-5824. Weber, J. S.,Jay, G., Tanaka, K., and Rosenberg, S. A. (1987).J. Exp. Med. 166, 1716-1733. Weiner, L. M., Padavic-Shaller, K., Kitson, J., Watts, P., Krigel, R. L., and Litwin, S. (1991). Cancer Res. 51, 3910-3918. Weiss, D. W. (1980a).Cum. Ti@. Microbiol. Immunol. 89, 1-83. Weiss, D. W. (1980b).Cancer Treat. Rep. 64, 481-485. West, W. H. (1989).Eur.1. Cancer Clin. Oncol. 25, Suppl. 3, Sll-S15. West, W. H., Tauer, K. W, Yannelli, J. R., Marshall, G. D., Orr, D. W., Thurman, G. B., and Oldham, R. K. (1987).N. Engl. J. Med. 316, 898-905. Whitehead, R. P., Ward, D., Hemingway, L., Hemstreet, G. P., Bradley, E., and Konrad, M. (1990).Cancer Res. 50,6708-6715. Whiteside, T. L., Heo, D. S., Takagi, S., Johnson, J. T., Iwatsuki, S., and Herberman, R. B. (1988).Cancer Immunol. Immunother. 26, 1- 10. Whitworth, P. W., Pak, C. C., Esgro, J., Kleinerman, E. S., and Fidler, J. J. (1990).Cancer Metastasis Rev. 8, 319-35 1. Wiebke, E. A., Rosenberg, S. A., and Lotze, M. T. (1988).J.Clin. Oncol. 6, 1440-1449. Wiebke, E. A., Custer, M. C., Rosenberg, S. A,, and Lotze, M. T. (1990).J . Biol. Response Mod$ 9, 113-126. Wiel-Hillman, G., Schell, K., Segal, D. M., Hank, J. A,, Sosman, J. A,, and Sondel, P. M. (1991).J. Immunother. 10, 267-277. Wildfang, J., Rathmann, R., Emminger, A,, and Schmoll, H. J. (199O).J.Cancer Res. Clin. Oncol. 116, Suppl. 1, 306 (Abstr. A3.106.27). Wiltrout, R. H., and Salup, R. R. (1988).Prog. Exp. Tumor Res. 32, 128-153. Winkelhake, J. L., Stampfl, S., and Zimmerman, R. J. (1987).Cancer Res. 47, 3948-3953. Wintzer, H.O.,Benzing, M., and von Kleist, S. (1990).Br. J. Cancer 62, 289-295. Wiseman, C. L.,Rao, V. S., Kennedy, P. S., Presant, C. A., Smith, D., and McKenna, R. J. (1989).West.J. Med. 151, 283-288. Wiseman, C., Rao, V. S., Williams, T., Mazumder, A., Kennedy, P., Presant, C., Smith, J., and McKenna, R. (1990).Proc. Am. SOC.Clin. Oncol. 9, 192 (Abstr. 741). Wolmark, N., Sicker, D., Yee, L., Hiserodt, J. C., and Salup, R. R. (1990).Proc. AACR 31, 268 (Abstr. 1585). Wong, R. A., Alexander, R. B., Puri, R. K., and Rosenberg, S. A. (1991).J.Immunother. 10,

120- 130. Woodruff, M. F. A. (1988).Adu. Cancer Res. 50, 197-229. Wunderlich, M., Schiessel, R., Rainer, H., Rauhs, R., Kovats, E., Schemper, M., Dittrich, Ch., Micksche, M., and Sedlacek, H. H. (1985).Br.J. Surg. 72, S107-SI10. Wussow, P. V., Pralle, H., Jakschies, D., Freund, M., Hochkeppel, H. K., and Deicher, H. (199O).J.Cancer Res. Clin. Oncol. 116, Suppl. 11, 1 1 14 (Abstr. 20.15.05). Xu, F. J., Rodriguez, G. C., Whitaker, R., Boente, M., Berchuck, A. McKenzie, S., Houston, L., Boyer, C. M., and Bast, R. C. (1991).Proc. Am. Assoc. Cancer Res. 32, 260 (Abstr.

1544).

Yagita, M., and Grimm, E. A. (1988).Prog. Exp. Tumor Res. 32, 213-226. Yang, J. C., Perry-Lalley, D., and Rosenberg, S. A. (1990a).J.Biol. Response Mod$ 9, 149-

159. Yang, J. C.,Schwarz, S., Perry-Lalley, D., and Rosenberg, S. A. (1990b).Proc. Am. Assoc. Cancer Res. 31, 410 (Abstr. 2434). Yang, S. C . , Owen-Schaub, L., Grimm, E. A., and Roth, J. A. (1989).Cancer Immunol. Immunother. 29, 193- 198.

322

ELI KEDAR AND EVA KLEIN

Yang, S. C., Grimm, E. A., and Roth, J. A. (1990). Proc. AACR 31, 237 (Abstr. 1399). Yang, S., Harris, B., Grimm, E., Hong, W., and Roth, J. A. (1991).Proc. Am. Sac. Clin. Oncol. 10, 215 (Abstr. 723). Yarden, Y. (1990). Mol. Immunol. 27, 1319-1324. Yasumoto, K., Miyazaki, K., Nagashima, A., Ishida, T., Kuda, T., Yano, T., Sugimachi, K,. and Nomoto, K. (1987). Cancer Res. 47, 2184-2187. Yemul, S., Berger, C., Katz, M., Estabrook, A., Edelson, R.,and Bayley, H. (1990). Cancer Immunol. Immunothm 30, 3 17-322. Yoneda, K., Yamamoto T., and Osaki, T. (1989). Int. J. Cancer 44,483-488. Yoshida, S., Tanaka, R.,Takai, N., and Ono, K. (1988). CancmRes. 48, 5011-5016. Yoshizawa, H., Sakai, K., Chang, A. E., and Shu, S. (1991). Cell. Immunol. 134,473-479. Yun, Y. S., Hargrove, M. E.,and Ting, C. C. (1989). Cancer Res. 49,4770-4774. Zambello, R., Trentin, L., Fergulio, C., Bulian, P., Masciarelli, M.,Cipriani, A., Agostini, C., and Semenzato, G. (1990). Cancer Res. 50, 1768-1773. Ziegler, L., Palazzolo, P., Janus, M., Cunnigham, J., Itoh, K., Hayakawa, K., Hellstrom, I., Nicaise, C., Dennin, R., and Murray, J. L. (1991). Proc. Am. Sac. CZin. Oncol. 10, 210 (Abstr. 703). Zimmerman, R.J., Aukerman, S. L., Katre, N. V., Winkelhake, J. L., and Young, J. D. (1989a). Cancer Res. 49,6521-6528. Zimmerman, R. J., Gauny, S., Chan, A., Landre, P., and Winkelhake, J. L. (1989b).J. Natl. Cancer Inst. 81,227-231. Zimmerman, R. J.. Winkelhake, J. W., Katre, N. V., Young, J. D., and Bauer, R. J. (1990). Proc. Sac. Biol. T k . 14 (Abstr. 18). Zwierzina, H. (1991). Eur.J. Cancer 27, 789-795.

INDEX

A Acetylation, fibroblast growth factors and, 122 Acidic fibroblast growth factor (FGF-l), 116 biological function, 140-144 extracellular matrix, 139 gene expression, 126, 128, 130-132 oncogenic potential, 148-149 protein structure, 122-124 receptors, 132, 134-1 35 tumors, 152-154 Active nonspecific immunotherapy, 258-268 Active specific immunotherapy, 255-258, 282,287-292 Activins, fibroblast growth factors and, 142 Acute promyelocytic leukemia (AF’L), c-erbA and, 109 Additive effects, cancer immunotherapy and, 260-262 Adenovirus, cytotoxic T lymphocytes and, 236,238 Adoptive immunotherapy, 293 current strategies, 268-274 improvement attempts, 282, 284, 289-292 Aflatoxin breast cancer, p53 expression in, 75 hepatitis B viruses and, 170, 178, 185, 209-210 Age, Wilms’ tumor and, 44, 47 Alcohol intake, hepatitis B viruses and, 178-179, 210

Alleles breast cancer, p53 expression in, 71 c-erbA and, 106 cytotoxic T lymphocytes and, 228, 238, 240 hepatocellular carcinoma and, 205, 208-209 Wilms’ tumor and, 44, 49-51, 58-60 Allobiophoria, Zilber and, 9 Alternative splicing fibroblast growth factors and, 130, 155 Wilms’ tumor and, 54-56 Amino acids c-erbA and, 92-93,99, 105-107 multiple loci, 94-95 mutations, 103-104 structure, 9 6 9 7 cytotoxic T lymphocytes and, 228, 231-233 fibroblast growth factors and, 117, 131, 147, 151 protein structure, 117, 119-122, 124-125 hepatitis B viruses and, 175, 180 genomes, 185, 187, 189 oncogenesis of viral proteins, 197, 199 Wilms’ tumor and, 55-56, 60, 62 6 -Aminolevulinic synthetase ( U - S ) gene, c-erbA and, 104 Anaphylaxis with desensitization reaction (ADR),Zilber and, 26-27 Angiogenesis, fibroblast growth factors and, 115-116, 122, 155 biological function, 140-142 tumors, 149, 153-155 323

324

INDEX

Aniridia, see also WAGR syndrome Wilms’ tumor and, 47-50,53-54,62 Antagonistic effects, cancer immunotherapy and, 262 Antibodies breast cancer, p53 expression in, 75-76, 78-79, 84 cancer immunotherapy and, 250,293 chemoimmunotherapy, 281 current strategies, 257, 264, 266, 270, 273 improvement attempts, 286,290-291 cytotoxic T lymphocytes and, 228, 233, 235 fibroblast growth factors and, 121, 137, 152 biological function, 140, 142 oncogenic potential, 146, 149 hepatitis B viruses and, 172, 175-176, 178, 197 Zilber and, 8, 32-33 Antibody-dependent cellular cytotoxicity (ADCC), cancer immunotherapy and, 273, 279, 290 Antigens breast cancer, p53 expression in, 69-71, 73,84 c-erbA and, 92 cancer immunotherapy and, 247, 292 critical factors, 248, 251-255 current strategies, 256-259, 266, 273, 281 improvement attempts, 284-288, 291 cytotoxic T lymphocytes and, 227-229 function, 237-240 processing, 233-235 hepatitis B viruses and, 168, L80, 202 epidemiology, 170, 172-178 genomes, 191-192, 194 oncogenesis of viral proteins, 195, 197-198 Wilms’ tumor and, 53 Zilber and, 8,25-27,29,32-33 Antioncogenes, hepatocellular carcinoma and, 209 Antitumor effects, cancer immunotherapy and, 293 chemoimmunotherapy, 275, 279-280 current strategies, 259-261, 263, 267, 272-273 improvement attempts, 283, 291

Atresia, Wilms’ tumor and, 48 AUG codons, fibroblast growth factors and, 119, 124,131, 148 Autologous lymphocyte cytoxicity, cancer immunotherapy and, 284 Autologous tumor stimulation, cancer immunotherapy and, 284 Avian erythroblastosis virus, c-erbA and, 90-93,96,98, 101-102 Avian hepadnaviruses genomes, 189 pathogenicity, 184-185

B cells breast cancer, p53 expression in, 85 cancer immunotherapy and, 285-286 cytotoxic T lymphocytes and, 236 hepatitis B viruses and, 177 Bacteria, Zilber and, 8, 30 Basement membranes, fibroblast growth factors and, 138-139, 141 Basic fibroblast growth factor (FGF-2), 116 biological function, 140-144 extracellular matrix, 138-139 gene expression, 126, 128-132 oncogenic potential, 148-149 protein structure, 117, 119-124 receptors, 132, 134-136 tumors, 152-153 BCG, cancer immunotherapy and, 256,287 Beckwith-Wiedemann syndrome, Wilms’ tumor and, 47, 49-51 Biological response modifiers (BRMs), cancer immunotherapy and, 247,293 critical factors, 253, 255 current strategies, 266267 improvement attempts, 283,287 Bladder carcinoma, cancer immunotherapy and, 264265 Bone marrow c-erbA and, 90 cancer immunotherapy and, 254,286 hepatitis B viruses and, 177, 181 Brain, fibroblast growth factors and, 144, 152 Breast cancer immunotherapy and, 252 p53 expression in, 69

INDEX

assay methods, 76-77 discovery, 69-70 dominant oncogene, 70 function, 71-72 function mutants, 70-71 immunochemistry, 78-80 lesions, in situ, 83-84 Li Fraumeni syndrome, 80-82 loss of heterozygosity, 74, 77-78 mutations, 75,80-81 prognosis, 82-83 receptors, 82 recessive oncogene, 70 studies, 77 therapeutic possibilities, 84-85 viral proteins, 72-74 Wilms’ tumor and, 47 Brefeldin A, cytotoxic T lymphocytes and, 233 Burkitt’s lymphoma, hepatocellular carcinoma and, 21 1

C C-mbA, 89-90, 108-109 avian erythroblastosis virus, 90-93 function, 105-106 mutations, 106-108 gene expression, 104-105 hepatocellular carcinoma and, 204 multiple loci, 94-96 mutations, biological activity, 102-104 protein function, 97-98 competition, 101-102 DNA binding, 98-99 hormone binding, 98 transcription factors, 99-101 structure, 96-97 thyroid hormone receptor, 93-94 c-myc, hepatitis B viruses and, 198,202, 205, 207, 21 1 Calcium, fibroblast growth factors and, 134 Cancer immunotherapy, 245-248, 292-293 critical factors, 248, 255 T cell response, 248-254 tumor location, 254255 current strategies active nonspecific immunotherapy, cytokines, 258-268 active specific immunotherapy, 255-258

325

adoptive immunotherapy, 268-273 chemoimmunotherapy, 274-281 improvement attempts, 282 active immunotherapy, 289-292 active specific immunotherapy, 287-289 elimination of suppressor cells, 289 experimental models, 282-284 patient selection, 284-286 tumor debulking, 286-287 Carbonic anhydrase, c-erbA and, 99 Carbonic anhydrase I1 gene, c-erbA and, 104 Carcinogenesis breast cancer, p53 expression in, 75, 84 fibroblast growth factors and, 147, 150, 155 hepatitis B viruses and, 167, 169, 188, 209-210 epidemiology, 170, 178-179 oncogenesis of viral proteins, 195-196, 199 pathogenicity, 182-185 Wilrns’ tumor and, 43, 61-62 Zilber and, 19 Carcinomas, see also specific carcinomas cancer immunotherapy and, 275 Casein kinase 11, breast cancer, p53 expression in, 72 Catalase, Wilrns’ tumor and, 53 CD4, cancer immunotherapy and, 248-249,291 CD8, cancer immunotherapy and, 248-249,291-292 cdc-2 kinase, breast cancer, p53 expression in, 72 cDNA breast cancer, p53 expression in, 76 c-erbA and, 93, 100 fibroblast growth factors and, 117, 119 oncogenic potential, 145-148 receptors, 132-1 35 tumors, 151-152 hepatitis B viruses and, 189 Wilms’ tumor and, 54 Cell adhesion molecules, cancer immunotherapy and, 259 Central nervous system fibroblast growth factors and, 144 Zilber and, 9 Centromeres, Wilms’ tumor and, 54

326

INDEX

Chemoimmunotherapy, cancer, 274-281 Chemotherapy cancer immunotherapy and, 247 chemoimmunotherapy, 274277,281 critical factors, 249-250, 254-255 current strategies, 260-261, 264, 269 immunomodulatory effects, 277-278 immunosuppression, 280 improvement attempts, 284-287 suppressor cells, 278-280 hepatitis B viruses and, 173 Wilms’ tumor and, 43, 60, 62 Chinese hamster ovary cells, fibroblast growth factors and, 133-134, 136 Chloramphenicol acetltransferase (CAT) C-erbA and, 100-101,104 fibroblast growth factors and, 129 Chromosomes breast cancer, p53 expression in, 70, 76-77 chromosome 17p, 77 c-erbA and, 89,94 cytotoxic T lymphocytes and, 239 fibroblast growth factors and, 115, 126, 150 hepatitis B viruses and, 169 genetic alterations, 208-209 viral DNA,200,203-204,207 Wilms’ tumor and, 41-42 functional studies, 61-63 genetic loci, 46-53 Knudson model, 44,47 llp13,47-49,51-54,61-62 1 1 ~ 1 549-53,61 , 17p, 47 W l gene, 53-54 Chronic infection, hepatocellular carcinoma and epidemiology, 173-177 viral DNA, 199-208 Cimetidine, cancer immunotherapy and, 256,279,289 Cirrhosis, hepatitis B viruses and, 170, 178179,182-183, 185 Cisplatin, cancer immunotherapy and, 276277 Clinical trials with cytokines, cancer immunotherapy and, 276-277 Clones breast cancer, p53 expression in, 70

c-erbA and, 90,92-95,97 cancer immunotherapy and, 248,253, 255,272-273,291 cytotoxic T lymphocytes and, 229, 239 fibroblast growth factors and, 117, 119, 138 oncogenic potential, 145-146 receptors, 132, 134 hepatitis B viruses and, 168 genomes, 185,188 pathogenicity, 180, 184 viral DNA, 200-201,204 Wilms’ tumor and, 44, 54, 61 Colon cancer breast cancer, p53 expression in, 75,80, 82 cytotoxic T lymphocytes and, 238 Colon carcinoma, cancer immunotherapy and, 252 chemoimmunotherapy, 276-277,280 current strategies, 256,264-265 Colony-stimulating factors, cancer immunotherapy and, 280, 286 Competition, c-erbA and, 101-102 Complete remission (CR), cancer immunotherapy and, 269 Contiguous gene syndromes, Wilms’ tumor and, 47 Corticosteroids, cancer immunotherapy and, 249,262 COS cells, c-erbA and, 99 Cotransfection c-erbA and, 100-102,106,109 hepatitis B viruses and, 198 Covalently closed circular DNA,hepatitis B viruses and, 189, 191, 193 CpG sites, breast cancer, p53 expression in, 75,80 Cryptorchidism, Wilms’ tumor and, 48-49, 60 CUG codons, fibroblast growth factors and, 119,124,131,147 CV1 cells, c-erbA and, 100-102 Cyclic AMP,c-erbA and, 105 Cyclin A,hepatocellular carcinoma and, 204-205 Cyclooxygenase, cancer immunotherapy and, 279,289 Cyclophosphamide, cancer immune therapy and, 249

INDEX

chemoimmunotherapy, 275-278 current strategies, 256, 260, 269 Cyclosporin A, cancer immunotherapy and, 249 Cytochrome P450, hepatocellular carcinoma and, 209 Cytokeratin, breast cancer, p53 expression in, 76, 78 Cytokines, cancer immunotherapy and, 246,293 chemoimmunotherapy, 275-277, 280-281 critical factors, 248, 253 current strategies, 257-268,272-273 improvement attempts, 286292 Cytoplasm breast cancer, p53 expression in, 72, 80 fibroblast growth factors and, 120, 136 hepatitis B viruses and, 189, 197, 199-200 Cytoplasmic staining, breast cancer, p53 expression in, 78 Cytoreductive factors, cancer immunotherapy and, 280, 286, 292 Cytotoxic T cells, cancer immunotherapy and, 255,272 Cytotoxic T lymphocytes, 227-228, 240-241 antigen processing, 233-235 cancer immunotherapy and, 248,257, 273,275,291-292 function, 235-240 hepatitis B viruses and, 173,177 influenza matrix peptide, 229-233 peptide MHC, 228-229 Cytotoxicity, cancer immunotherapy and chemoimmunotherapy, 276,278, 280 current strategies, 259, 269, 271-272 improvement attempts, 285, 292

D Dacarbazine, cancer immunotherapy and, 277 Delayed-type hypersensitivity (DTH), cancer immunotherapy and, 257 Differentiation c-erbA and, 89,95, 108 avian erythroblastosis virus, 92-93

327

function, 105-106 gene expression, 104-105 fibroblast growth factors and, 129, 142-143,151,155 Dimerization, c-erbA and, 99, 106 Direct repeats, hepatitis B viruses and, 199, 201-202 DNA breast cancer, p53 expression in, 71, 73, 76 c-erbA and, 93, 106108 gene expression, 104-105 multiple loci, 94-95 mutations, 102-104 protein function, 98-102 structure, 97 cancer immunotherapy and, 258, 288 cytotoxic T lymphocytes and, 238 fibroblast growth factors and, 140, 150 gene expression, 126, 129, 131 oncogenic potential, 145-148 hepatitis B viruses and, 168-169, 210-211 epidemiology, 170, 177 genetic alterations, 208-209 genomes, 185-187,189-191, 193 oncogenesis of viral proteins, 195, 198-199 pathogenicity, 180-182, 184 viral DNA, 199-208 Wilms’ tumor and, 49-50,53-56,58 Zilber and, 4, 9, 25 DNA polymerase, hepatitis B viruses and, 170, 187-188,191 Dose response, cancer immunotherapy and, 263,274 Double-stranded DNA, hepatitis B viruses and, 189 Doxorubicin, cancer immunotherapy and, 276277,289 Drash syndrome, Wilms’ tumor and, 48, 60 LhomphiZa, fibroblast growth factors and, 126 DTIC, cancer immunotherapy and, 276 Duck hepatitis virus (DHBV) genomes, 189, 193-194 oncogenesis of viral proteins, 197 pathogenicity, 184-185 viral DNA, 201-202

INDEX

E Early growth response genes, Wilms’ tumor and, 55-56 EBNA, cytotoxic T lymphocytes and, 236, 239 Embryonal carcinoma cells c-erbA and, 109 fibroblast growth factors and, 128-131, 151 Encephalitis, Zilber and, 9-13, 15, 18,20, 24 Endocytosis, cancer immunotherapy and, 266,270 Endoplasmic reticulum breast cancer, p53 expression in, 78-79, 82-83 cytotoxic T lymphocytes and, 233-234 fibroblast growth factors and, 124 hepatitis B viruses and, 191-192, 196, 202 Endothelium, fibroblast growth factors and, 116,128, 154 biological function, 140-141 protein structure, 120, 122 mu genes, fibroblast growth factors and, 151 Enzymes breast cancer, p53 expression in, 72 c-erbA and, 99 cancer immunotherapy and, 257 cytotoxic T lymphocytes and, 227 fibroblast growth factors and, 141 hepatitis B viruses and, 187, 189, 202 Wilms’ tumor and, 53 Epidemiology hepatitis B viruses and, 170-171 hepatocellular carcinoma, 177-179 progression, 173-177 transmission, 171-1 74 Zilber and, 5, 8 Epidermal growth factor, c-erbA and, 92 Epidermal growth factor receptor breast cancer, p53 expression in, 78-79, 82 c-erbA and, 90 Epithelium fibroblast growth factors and, 125, 128129 Wilms’ tumor and, 57-58

Epitopes breast cancer, p53 expression in, 76-77, 84 cancer immunotherapy and, 255-256 cytotoxic T lymphocytes and, 240 antigen processing, 233-234 function, 237, 239 influenza matrix peptides, 232-233 peptide MHC, 228-229 Epstein-Barr virus cytotoxic T lymphocytes and, 235-236, 239 hepatocellular carcinoma and, 210-21 1 ErbA proteins, c-erbA and, 97-102 ErbB-2, breast cancer, p53 expression in, 82-83 Erythroblasts, see also Avian erythroblastosis virus c-wbA and, 92-93, 104-105 Erythrocytes c-mbA and, 92.95, 104-105 Wilms’ tumor and, 53 Zilber and, 30 Erythroid cells, c-wbA and, 92-93, 101, 103-105 Esterase D, Wilms’ tumor and, 53 Eukaryotes hepatitis B viruses and, 197,202 Zilber and, 9 Experimental systems, cancer immunotherapy and, 275-276 Experimental tumors, cancer immunotherapy and, 248-250 Extracellular matrix, fibroblast growth factors and, 116, 126, 136, 138-140, 154 Eye, Wilms’ tumor and, 47

F Familial breast cancer, p53 expression in, 80-82 Familial hepatocellular carcinoma, 178 Familial Wilms’ tumor, 52-53 Fibroblast growth factor-1, see Acidic fibroblast growth factor Fibroblast growth factor-2, see Basic fibroblast growth factor Fibroblast growth factor-3, see INT-2

INDEX

Fibroblast growth factor-4, see HST/K-FGF Fibroblast growth factor-5 biological function, 141, 143-144 gene expression, 126, 129, 131-132 oncogenic potential, 145, 147-148 protein structure, 125 Fibroblast growth factor-6 gene expression, 126, 129, 132 oncogenic potential, 149 protein structure, 125 Fibroblast growth factor-7, see Keratinocyte growth factor Fibroblast growth factors, 115-1 18, 155-156 biological function, 140-144 extracellular matrix, 138-140 gene expression, 126, 128-132 oncogenic potential, 144-149 protein structure acidic FGF, 122-124 basic FGF, 117, 119-122 FGF-5, 125 HST/K-FGF, 124-1 25 INT-2,124 keratinocyte growth factor, 125-127 receptors, 132-137 tumors, 149-155 Fibroblasts c-erbA and, 90,92,98 Wilms’ tumor and, 61 Fibromatosis, fibroblast growth factors and, 153 Fibrosarcomas, fibroblast growth factors and, 152-L53 Flg receptor, fibroblast growth factors and, 133-1 36 P-Follicle-stimulating hormone, Wilms’ tumor and, 54 5-FU, cancer immunotherapy and, 276-277, 280

G GCSF, cancer immunotherapy and, 268, 288 Gag, cytotoxic T lymphocytes and, 237 gag gene c-erbA and, 97

329

fibroblast growth factors and, 151 Gender, hepatitis B viruses and, 178 Gene expression c-erbA and, 97, 100, 104-105 cytotoxic T lymphocytes and, 238 fibroblast growth factors and, 150, 155 oncogenic potential, 145-146 protein structure, 126, 128-132 hepatitis B viruses and, 177, 192-195 Gene transfer, cancer immunotherapy and, 272 Genes cancer immunotherapy and, 257 fibroblast growth factors and, 117, 133, 150 Genetic engineering, cancer immunotherapy and, 272-273, 288 Genetics breast cancer, p53 expression in, 69-71, 74 c-erbA and, see c-erbA cancer immunotherapy and, 258,292 hepatitis B viruses and, 169, 178, 210 alterations, 208-210 genomes, 185-189 pathogenicity, 180, 184 viral DNA, 207-208 of Wilms’ tumor, see Wilms’ tumor, genetics of Genitourinary malformation, see also WAGR syndrome Wilms’ tumor and, 43,47-49,60 Genomes, hepatitis B viruses and, 185 gene expression, 192-195 organization, 185-1 89 structure, 189-19 1 virion assembly, 191-192 Genotype, breast cancer, p53 expression in, 74 Glucocorticoid receptors, c-erbA and, 93-94, 106 Glycoprotein cancer immunotherapy and, 276 cytotoxic T lymphocytes and, 227, 237 Glycosaminoglycans, fibroblast growth factors and, 116, 135 Glycosylation fibroblast growth factors and, 124-126, 132 hepatitis B viruses and, 191

330

INDEX

GM-CSF, cancer immunotherapy and, 267, 290 Golgi cytotoxic T lymphocytes and, 233 fibroblast growth factors and, 124 Goss-Harris technique, Wilms’ tumor and, 53 Granulation tissue formation, fibroblast growth factors and, 141-142 Ground squirrel hepatitis virus (GSHV) genomes, 188, 193 oncogenesis of viral proteins, 197 pathogenicity, 180, 182-184 viral DNA, 201-202,207 Ground squirrel hepatocellular carcinoma, 207 Growth suppressor genes, see c-erbA

H H2 receptor, cancer immunotherapy and, 279,289 HBX, hepatitis B viruses and, 195-199 Heat shock protein 70, breast cancer, p53 expression in, 70, 72, 76 Hematopoietic cells c-erbA and, 90,92, 108 cancer immunotherapy and, 276, 286-287 fibroblast growth factors and, 135-136 Hemizygosity,Wilms’ tumor and, 49,54,60 Hepadnaviruses, 168-169,211 genomes, 185 gene expression, 192-195 organization, 185-189 structure, 189-191 viral DNA, 199-200,202,205,207-208 virion assembly, 191-192 pathogenicity, 179-180 avian, 184-185 mammalian, 180-184 Heparan sulfate, fibroblast growth factors and, 135-136, 139,146 Heparan sulfate proteoglycans (HSPGs), fibroblast growth factors and, 116, 132, 136,138 Heparin, fibroblast growth factors and, 116-118, 153 extracellular matrix, 138-139 protein structure, 120-121, 123-124

receptors, 135-137 Heparin-binding growth factor (HBGF), 116 Hepatitis, breast cancer, p53 expression in, 75 Hepatitis B c antigen (HBcAg), epidemiology, 173,175-176 Hepatitis B e antigen (HBeAg), epidemiolO ~ Y ,170,172-173,175-176 Hepatitis B surface antigen (HBsAg) epidemiology, 170,172-175,177-179 genomes, 191-192, 194 oncogenesis of viral proteins, 195, 197 viral DNA, 202 Hepatitis B viruses, 167-170,210-211 epidemiology, 170-171 hepatocellular carcinoma, association with, 177-179 progression to chronicity, 173-177 transmission, 171-174 genetic alterations, 208-210 genomes gene expression, 193-195 organization, 185-189 oncogenesis of viral proteins, 195-199 pathogenicity, 179-185 viral DNA, 200-208 Hepatitis B x antigen (HBxAg) oncogenesis of viral proteins, 197 viral DNA, 203 Hepatitis C virus, 170, 199 Hepatitis delta virus (HDV), 179 Hepatocellular carcinoma, 167-170, 210 epidemiology, 170-171, 177-179 genetic alterations, 208-210 genomes, 187 oncogenesis of viral proteins, 195-199 pathogenicity, 182-185 viral DNA, 199-200 cellular targets, 203-205 integrated sequences, 200-203 myc genes, 205-208 Heron hepatitis B virus (HHBV) genomes, 189 pathogenicity, 184 Heterodimers c-erbA and, 102 fibroblast growth factors and, 137 Heterozygosity loss of, see Loss of heterozygosity Wilms’ tumor and, 52, 59, 62

33 1

INDEX

Hippocampus, fibroblast growth factors and, 144 Histamine, cancer immunotherapy and, 279,289 Hit kinetics, Wilms’ tumor and, 46 HLA breast cancer, p53 expression in, 84-85 cancer immunotherapy and, 247, 251252,256,285 cytotoxic T lymphocytes and, 227, 229-234 function, 237-240 hepatitis B viruses and, 177, 198 HLA-A2 cancer immunotherapy and, 252 cytotoxic T lymphocytes and, 229-233, 238 Hodgkin’s disease, cytotoxic T lymphocytes and, 240 Homology breast cancer, p53 expression in, 72 c-erbA and, 92-94,97, 103, 106 cytotoxic T lymphocytes and, 233 fibroblast growth factors and, 115, 117, 126,138 protein structure, 120, 123-127 receptors, 132-133 hepatitis B viruses and genomes, 187-188, 194 oncogenesis of viral proteins, 197 viral DNA, 201, 204, 206 Wilms’ tumor and, 55-56,62 Homozygosity, Wilms’ tumor and, 50,52, 54,58 Hormone receptors, c-&A and, 93, 95, 97 Hormone response elements, c-erbA and, 97,99, 106108 Hormones, c-erbA and, 90, 106, 108 function, 105-106 gene expression, 104-105 mutations, 102, 104 protein function, 98-101 HST/K-FGF (Fibroblast growth factor-4) oncogenic potential, 145-147 protein structure, 124-125 Human anti-mouse antibody (HAMA), cancer immunotherapy and, 281 Human breast cancer, p53 expression in, see Breast cancer, p53 expression in Human glucocorticoid receptor, c-erbA and, 93

Human immunodeficiency virus cytotoxic T lymphocytes and, 233, 235237 hepatitis B viruses and, 173, 198 Human papilloma virus, hepatitis B viruses and, 210-211 Hybridization, see also in situ hybridization c-erbA and, 103-105 cancer immunotherapy and, 290 hepatitis B viruses and, 170, 199 Wilms’ tumor and, 53-54,57-58 Hyperthermia, cancer immunotherapy and, 266 Hypospadias, Wilms’ tumor and, 47-48,60 Hypoxanthine phosphoribosyltransferase (HPRT), Wilms’ tumor and, 61

1 ICAM-1, cancer immunotherapy and, 253, 285 Immune response cancer immunotherapy and, 247,254, 284,288,293 fibroblast growth factors and, 155 hepatitis B viruses and, 175-176, 181 Immune responses, cytotoxic T lymphocytes and, 229, 235, 240-241 Immune surveillance, cytotoxic T lymphocytes and, 228, 236, 240 Immunization cancer immunotherapy and, 245, 251, 257,293 current strategies, 287-289 hepatitis B viruses and, 180 Zilber and, 27 Immunodeficiency, cancer immunotherapy and, 254,283 Immunogenicity, cancer immunotherapy and, 292-293 chemoimmunotherapy, 281 critical factors, 251, 254-255 current strategies, 255-257, 259-261, 263,270 improvement attempts, 282-284, 287-288 Immunoglobulin fibroblast growth factors and, 133-135 hepatitis B viruses and, 211

332 Immunolocalization, fibroblast growth factors and, 128 Immunology breast cancer, p53 expression in, 72, 76-77 hepatitis B viruses and, 168, 172 Zilber and, 25-26, 32-35, 37 Immunomodulation, cancer immunotherapy and, 258,262-263,277-278 Immunosuppression cancer immunotherapy and, 249,286 chemoimmunotherapy, 275, 279-281 current strategies, 263 cytotoxic T lymphocytes and, 236 Immunotherapy, cancer, see Cancer immunotherapy in situ hybridization fibroblast growth factors and, 125, 128 Wilms’ tumor and, 57-58 in sttu lesions, breast cancer, p53 expression in, 83-84 Indomethacin, cancer immunotherapy and, 256,262,279,286 Inflammation cancer immunotherapy and, 258-259 hepatitis B viruses and, 168, 181-182 Influenza A virus, cytotoxic T lymphocytes and, 234 Influenza matrix peptide, cytotoxic T lymphocytes and, 229-233 Influenza virus, cytotoxic T lymphocytes and, 237 Influenza virus proteins, cytotoxic T lymphocytes and, 228-229 Inhibitors breast cancer, p53 expression in, 71-72, 74 c-erbA and, 101-109 cancer immunotherapy and, 246, 289 chemoimmunotherapy, 276,279 current strategies, 259-260, 262, 268, 273 fibroblast growth factors and, 131, 142, 146 protein structure, 119, 121 tumors, 152, 154 Insertional activation of myc genes, hepatitis B viruses and, 205-208 Insulin-like growth factor 11, Wilms’ tumor and, 50

INDEX

INT-2 (Fibroblast growth factor-3) biological function, 142-143, 145 gene expression, 126, 128,131-132 oncogenic potential, 147-148 protein structure, 124 tumors, 149-152, 155 Interferon cancer immunotherapy and, 246,249, 29 1 chemoimmunotherapy, 275-276,280 current strategies, 258-259, 261-262 hepatitis B viruses and, 177, 198 Interferon a,cancer immunotherapy and, 247 chemoimmunotherapy, 276277,279 critical factors, 249-250, 252-253 current strategies, 259-261, 267 improvement attempts, 287, 289-291 Interferon p, cancer immunotherapy and, 265,290 Interferon T cancer immunotherapy and adoptive immunotherapy, 269-270, 272-273 chemoimmunotherapy, 280 critical factors, 250, 253 current strategies, 257, 259-263, 265, 267 improvement attempts, 287-299.291 cytotoxic T lymphocytes and, 227-228 Interleukin-1, cancer immunotherapy and chemoimmunotherapy, 275,280 current strategies, 260, 262-263, 265 improvement attempts, 286-289, 292 Interleukin-2, cancer immunotherapy and, 246-247 chemoimmunotherapy, 275-277, 279-281 critical factors, 249-253 current strategies, 259-273 improvement attempts, 285-288, 290-291 Interleukin-3, cancer immunotherapy and, 268,290 Interleukin-4, cancer immunotherapy and current strategies, 257, 260, 262, 265, 269 improvement attempts, 288,290-291 Interleukin-6, cancer immunotherapy and, 260,262,269,280,287,290

INDEX

Iris, Wilms’ tumor and, 47 Irradiation, cancer immunotherapy and, 278, 291

K K-FGF biological function, 141-143 gene expression, 126, 128-130 oncogenic potential, 145-148 protein structure, 124-125 receptors, 135-136 tumors, 149-152 Kaposi’s sarcoma cytotoxic T lymphocytes and, 236 fibroblast growth factors and, 151, 153 Karyotype, Wilms’ tumor and, 46,50,54 Keratinocyte growth factor (FGF-7), 116117 gene expression, 132 oncogenic potential, 149 protein structure, 125-126 receptors, 134-135 Kidney cancer, see Wilms’ tumor Knudson model, Wilms’ tumor and, 43-46, 58-59,61,63

L Lesions, in situ, breast cancer, p53 expression in, 83-84 Leukemia breast cancer, p53 expression in, 74, 80 c-er6A and, 109 fibroblast growth factors and, 151 hepatitis B viruses and, 173 Wilms’ tumor and, 43, 52 Leukemia viruses, Zilber and, 30 Leukopenia, cancer immunotherapy and, 284 Levamisole, cancer immunotherapy and, 277,280, 288 Li Fraumeni syndrome, breast cancer, p53 expression in, 80-82 Ligands c-erbA and, 93-94,97, 106108 mutations, 104 protein function, 97-98, 101

333

fibroblast growth factors and, 135, 137 Lipids cancer immunotherapy and, 265-266 hepatitis B viruses and, 191 Liposomes, cancer immunotherapy and, 265-268,287,290 Liver, see ako Hepatitis B viruses; Hepatocellular carcinoma cancer immunotherapy and, 271 Long terminal repeats c-et6A and, 99, 101 fibroblast growth factors and, 147 hepatitis B viruses and, 198 Loss of heterozygosity breast cancer, p53 expression in, 71, 74, 76-78,80 hepatitis B viruses and, 208 Wilms’ tumor and, 47, 49, 51, 53 Lung cancer breast cancer, p53 expression in, 80 immunotherapy and, 271 Lung carcinoma, cancer immunotherapy and, 256 Lymph nodes breast cancer, p53 expression in, 83 cancer immunotherapy and, 250, 255, 278,284,289 Lymphocytes, see also Cytotoxic T lymphocytes; Peripheral blood lymphocytes; Tumor-infiltrating lymphocytes (TILs) breast cancer, p53 expression in, 80 cancer immunotherapy and, 246, 293 critical factors, 251, 253 current strategies, 268, 270, 278 improvement attempts, 285-286 hepatitis B viruses and, 173, 177, 181 Lymphocytic choriomeningitis virus, cytotoxic T lymphocytes and, 237 Lymphoid cells, cancer immunotherapy and, 246,260,271 Lymphokine-activated killer (LAK) cells, cancer immunotherapy and, 246-247 chemoimmunotherapy, 275,277, 279 critical factors, 249-252, 254-255 current strategies, 262-263, 265, 268-271, 273 improvement attempts, 285, 290-291 Lymphokines, cancer immunotherapy and, 246,253-254,262

334

INDEX

Lymphoma, see also Burkitt’s lymphoma; T cell lymphoma cancer immunotherapy and, 277,290 cytotoxic T lymphocytes and, 236

M Macrophages cancer immunotherapy and, 251, 253, 259,266,280 hepatitis B viruses and, 177 Major histocompatibility complex cancer immunotherapy and, 285 critical factors, 248, 250-255’ current strategies, 257-259 cytotoxic T lymphocytes and, 227-229, 231, 233,240 antigen processing, 233-234 function, 238-239 Mapping cytotoxic T lymphocytes and, 228, 234, 237 fibroblast growth factors and, 115, 118, 120, 123,131, 150 hepatitis B viruses and, 199,202-203 Wilms’ tumor and, 46,53-54 Me1anoma cancer immunotherapy and, 247, 293 chemoimmunotherapy, 276,278-279 critical factors, 250-253 current strategies, 256, 260-261, 264, 268-272 improvement attempts, 284-285, 287, 289, 292 cytotoxic T lymphocytes and, 235 fibroblast growth factors and, 152 Mental retardation, see also WAGR syndrome Wilms’ tumor and, 48 Messenger RNA breast cancer, p53 expression in, 76 c-erbA and, 94-95,97, 104 fibroblast growth factors and, 148 biological function, 143-144 gene expression, 128, 131 protein structure, 125-126 hepatitis B viruses and, 189, 195, 199, 204 Wilms’ tumor and, 50,56-58, 61

Metastases, cancer immunotherapy and, 255,292 chemoimmunotherapy, 277 current strategies, 260-261, 264-265, 270-271,273 improvement attempts, 283-285, 288 p p Microglobulin, cytotoxic T lymphocytes and, 234,238 Mitogens c-erbA and, 92 cancer immunotherapy and, 270,278, 285-286 fibroblast growth factors and, 115-117, 132,140,147 protein structure, 120, 122-123, 125 tumors, 152-153 hepatitis B viruses and, 181 Wilms’ tumor and, 55 Mitomycin C, cancer immunotherapy and, 278 Mitosis, breast cancer, p53 expression in, 71-72,74 Molecular markers, Wilms’ tumor and, 46 Moloney leukemia virus c-erbA and, 99, 101 fibroblast growth factors and, 151 Moloney murine leukemia virus, hepatitis B viruses and, 207 Monoclonal antibodies cancer immunotherapy and, 246, 255 current strategies, 266, 273-274 improvement attempts, 289-291 fibroblast growth factors and, 139 Mouse m a m m q tumor virus (MMTV), fibroblast growth factors and, 124, 147, 149-151 mRNA, see Messenger RNA MTP-PE, cancer immunotherapy and, 267 AMTV, c-erbA and, 100-101 Muramyl dipeptide (MDP), cancer immunotherapy and, 267 Murine leukemia virus, hepatitis B viruses and, 205,208 Mutagenesis fibroblast growth factors and, 130 hepatitis B viruses and, 169, 203-205, 21 1 Mutation, see also Point mutation breast cancer, p53 expression in, 69-77, 80-85 c-erbA and, 89,106-109

335

INDEX

avian erythroblastosis virus, 90, 92-93, 96 mutations, 102-104 protein function, 97-98, 101 structure, 96-97 cancer immunotherapy and, 288 cytotoxic T lymphocytes and, 228, 233-234,236241 fibroblast growth factors and, 136, 142 oncogenic potential, 145-146, 149 protein structure, 120-121, 123 hepatitis B viruses and, 176, 187, 189, i9i,202,208-210 Wilms’ tumor and functional studies, 61-63 genetic loci, 47, 49, 51-52 Knudson model, 44,46 W T I gene, 57-61 Zilber and, 4 m9c genes, hepatitis B viruses and, 169, 205-208, 210 Myelosuppression, cancer immunotherapy and, 276,280-281 Myocytes, fibroblast growth factors and, 143

N N-myc, hepatitis B viruses and, 206207, 21 1 N-ras gene, hepatitis B viruses and, 208 Natural killer cells cancer immunotherapy and chemoimmunotherapy, 277 critical factors, 251, 254 current strategies, 259-260, 271 improvement attempts, 285 cytotoxic T lymphocytes and, 228, 237 Negative regulatory elements, fibroblast growth factors and, 131 Negative regulatory sequences, fibroblast growth factors and, 146 Neoplasms cancer immunotherapy and, 253, 265 fibroblast growth factors ahd, 154 hepatitis B viruses and, 170, 209 Wilms’ tumor and, 63 Neovascularization, fibroblast growth factors and, 141-142 Nephroblastomas, Wilms’ tumor and, 62

Neuroblastoma, Wilms’ tumor and, 44 Neurological diseases, Zilber and, 32 Neurons, fibroblast growth factors and, I44 Neutralizing antibodies cancer immunotherapy and, 281 cytotoxic T lymphocytes and, 237 Nucleocapsids, hepatitis B viruses and, 191-193 Nucleoprotein cytotoxic T lymphocytes and, 228, 232-234 Zilber and, 9, 25-26 Nucleotides c-erbA and, 92,97, 107 fibroblast growth factors and, 129 hepatitis B viruses and, 180, 185, 187-189,193, 197

0 Oligomerization, fibroblast growth factors and, 136-137 Oligonucleotides, c-erbA and, 99, 107 Oncogenes breast cancer, p53 expression in, 69-70, 73-74, 77,82 c-erbA and, 89-93, 107, 109 function, 105 mutations, 103 protein function, 97-102 cancer immunotherapy and, 267,288 cytotoxic T lymphocytes and, 239 fibroblast growth factors and, 117, 130, 155-156 potential, 144-149 tumors, 149-155 hepatitis B viruses and, 168-170, 210-211 epidemiology, 171, 178 pathogenicity, 181-184 viral DNA, 203 viral proteins, 195-199 Wilms’ tumor and, 58 Oncogenic viruses, Ziiber and, 30-32, 34-35 Open reading frames c-erbA and, 97 fibroblast growth factors and, 132, 148

336

INDEX

Open reading frames (continued) hepatitis B viruses and, 186187, 189, 197,203 Wilms’ tumor and, 54 Osteosarcoma, Wilms’ tumor and, 60 Ovarian carcinoma, cancer immunotherapy and, 254,264-265 Ovary, Wilms’ tumor and, 57 Overshot effect, cancer immunotherapy and, 277

P p53 expression in human breast cancer, see Breast cancer, p53 expression in P53 gene, hepatocellular carcinoma and, 209 Partial remission (PR), cancer immunotherapy and, 269 PC12 cells, fibroblast growth factors and, 143 Peptides breast cancer, p53 expression in, 84 cancer immunotherapy and, 251, 256, 288 cytotoxic T lymphocytes and, 227-234, 237,240-241 fibroblast growth factors and, 126, 133 oncogenic potential, 147-148 protein structure, 120-121, 123-124 tumors, 151-153 hepatitis B viruses and, 175, 197,203 Peripheral blood lymphocytes, hepatitis B viruses and, 181 Phagocytes, cancer immunotherapy and, 266,280,286 Phenotype breast cancer, p53 expression in, 70, 83-84 c-erbA and, 92-93,103,105-108 cancer immunotherapy and, 247,252 cytotoxic T lymphocytes and, 234, 236, 238-239 fibroblast growth factors and, 141, 146149,151,155 hepatitis B viruses and, 199 Wilms’ tumor and, 47-48,50, 61-63 Phospholipids, cancer immunotherapy and, 266

Phosphorylation breast cancer, p53 expression in, 71-72 fibroblast growth factors and, 122, 137 hepatitis B viruses and, 198 Wilms’ tumor and, 56 Phytohemagglutinin, breast cancer, p53 expression in, 80 Plague, Zilber and, 5-8 Plasma, cancer immunotherapy and, 263-265,268 Plasma membrane cancer immunotherapy and, 290 fibroblast growth factors and, 132, 140 hepatitis B viruses and, 197 Plasmids, c-erbA and, 100 Plasminogen activator, fibroblast growth factors and, 120, 141 Pneumococci, Zilber and, 4 Point mutation c-erbA and, 98, 101, 107 fibroblast growth factors and, 145 hepatocellular carcinoma and, 208-210 pol genes, fibroblast growth factors and, 151 Polyadenylation, hepatitis B viruses and, 193 Polyethylene glycol (PEG), cancer immunotherapy and, 266, 268 Polymerase chain reaction breast cancer, p53 expression in, 76 hepatitis B viruses and, 170 Wilms’ tumor and, 58 Polymorphism, cytotoxic T lymphocytes and, 234-235 Polypeptides fibroblast growth factors and, 125 hepatitis B viruses and, 204,210 genomes, 191,193 oncogenesis of viral proteins, 196198 Wilms’ tumor and, 54-55,58 Polyposis coli, Wilms’ tumor and, 52-53 Preneoplastic lesions, hepatitis B viruses and, 195-196, 199-200 Protease, cytotoxic T lymphocytes and, 234 Protein breast cancer, p53 expression in, 69-76, 80,82 c-erbA and, 92, 106108 function, 97-102 gene expression, 104-105

INDEX

multiple loci, 94-95 mutations, 102-104 structure, 96-97 cancer immunotherapy and, 283, 288, 290 cytotoxic T lymphocytes and, 228, 240-241 antigen processing, 233-234 function, 237, 239 influenza matrix peptides, 232-233 fibroblast growth factors and, 115, 117, 155 biological function, 143 gene expression, 131-132 oncogenic potential, 146149 receptors, 132-134, 136137 structure, 117, 119-127 tumors, 151-154 hepatitis B viruses and, 168-169, 184, 202 genomes, 186,189, 191-195 oncogenesis of viral proteins, 195-199 Wilms’ tumor and, 55-56,58 Zilber and, 28 Protein kinase A, fibroblast growth factors and, 122 Protein kinase C c-erbA and, 105 cancer immunotherapy and, 291 fibroblast growth factors and, 122 hepatitis B viruses and, 198 Proteolysis breast cancer, p53 expression in, 74 fibroblast growth factors and, 119, 136, 139 Proteus vulgaris, Zilber and, 3-4 Protooncogenes, see also c-erbA fibroblast growth factors and, 146, 155 hepatitis B viruses and, 169, 200, 204, 207

R Radiation, cancer immunotherapy and, 249 Radiation therapy, Wilms’ tumor and, 43, 60 Radiotherapy, cancer immunotherapy and, 254-255, 280, 286

337

ras, cytotoxic T lymphocytes and, 239 Rat growth hormone, c-erbA and, 99 R B I gene, Wilms’ tumor and, 44 Recombination, hepatocellular carcinoma and, 201-202,204 Remission, cancer immunotherapy and, 269 Renal cell carcinoma, cancer immunotherapy and, 247, 293 chemoimmunotherapy, 277,279 critical factors, 250-253 current strategies, 261, 264, 268-271 improvement attempts, 284-286 Replication breast cancer, p53 expression in, 71, 73 cytotoxic T lymphocytes and, 228 hepatitis B viruses and, 168 epidemiology, 170, 176 genomes, 185-192,194-195 oncogenesis of viral proteins, 197-198 pathogenicity, 180-182, 184 viral DNA, 199, 201 Retinoblastoma breast cancer, p53 expression in, 73, 83 Wilms’ tumor and, 63 genetic loci, 49, 52 Knudson model, 44,46 WTl gene, 53,60 Retinoic acid c-erbA and, 93, 108-109 hepatocellular carcinoma and, 204 Retinoic acid receptor p gene, hepatocellular carcinoma and, 204 Retinoic acid receptors (RARs), c-erbA and, 108-1 09 Retroposons, hepatocellular carcinoma and, 206 Retroviruses c-erbA and, 89, 104 cancer immunotherapy and, 272, 288, 29 1 fibroblast growth factors and, 145, 151 hepatitis B viruses and, 187, 189, 193 hepatocellular carcinoma and, 200-203, 206208 Reverse transcription, hepatitis B viruses and, 185, 187, 194 Ribosomes fibroblast growth factors and, 119 hepatitis B viruses and, 193

338

INDEX

RNA breast cancer, p53 expression in, 72.76 c-erbA and, 91 fibroblast growth factors and gene expression, 126, 128-131 receptors, 134 tumors, 150-154 hepatitis B viruses and, 178, 181 genomes, 187, 189-191, 193-194 Wilms' tumor and, 57 Zilber and, 29 RNA polymerase, hepatitis B viruses and, 193, 198 RNase H, hepatitis B viruses and, 190 Rous sarcoma virus c-erbA and, 97, 100 Zilber and. 29-30

S S protein, hepatitis B viruses and, 191-192, 194, 196 Saccharomyces cerevisiae, c-erbA and, 98 Selective pressure, cytotoxic T lymphocytes and, 238-240 Sequences breast cancer, p53 expression in, 70-72, 74, 76, 80 c-erbA and, 92-97,99 cancer immunotherapy and, 261, 274275 cytotoxic T lymphocytes and, 229-230, 233,237,239-240 fibroblast growth factors and, 117, 136, 151 gene expression, 126, 129, 131 oncogenic potential, 145-148 protein structure, 119-127 hepatitis B viruses and, 168-169,172, 180 genomes, 185,187-189, 193-194 integration, 200-203 oncogenesis of viral proteins, 195-199 viral DNA, 200,205,207 Wilms' tumor and, 56 Signal transduction cancer immunotherapy and, 258 fibroblast growth factors and, 135-138, 146,155

hepatitis B viruses and, 198 Somatic mutation, c-erbA and, 106 Spinal cord, fibroblast growth factors and, 144, 152 Spleen cancer im-munotherapyand, 249,268, 277, 279 hepatitis B viruses and, 174,181 Squamous cell carcinoma, cancer immunotherapy and, 251 SSCP, breast cancer, p53 expression in, 76 Steroid hormone receptors, c-erbA and, 90, 93,97-98,106,108 Steroids, hepatitis B viruses and, 177 Suppressor cells, cancer immunotherapy and, 278-281,289 Surface glycoprotein cytotoxic T lymphocytes and, 237 hepatitis B viruses and, 195-196,210 Survival, cancer immunotherapy and, 274, 276,288 Susceptibility,cancer immunotherapy and, 258-259 SV40 breast cancer, p53 expression in, 71 hepatitis B viruses and, 194, 196, 198 Syndecan, fibroblast growth factors and, 138 Synergistic effects, cancer immunotherapy and, 260-262,275

T T antigens, breast cancer, p53 expression in, 69-71, 73 T cell growth factor j3, fibroblast growth factors and, 142 T cell leukemias, fibroblast growth factors and, 151 T cell lymphoma cancer immunotherapy and, 260 hepatocellular carcinoma and, 205-206 T cell receptor, cytotoxic T lymphocytes and, 229,232 T cells breast cancer, p53 expression in, 84 cancer immunotherapy and, 246-247, 293 chemoimmunotherapy, 277-279

339

INDEX

critical factors, 248-254 current strategies, 257,259,261-262,

270-273 improvement attempts, 283-284,

286-292 cytotoxic T lymphocytes and, 229,

235-237 T lymphocytes, see also Cytotoxic T lymphocytes Teratocarcinomas, fibroblast growth factors and, 151 Therapy, breast cancer, p53 expression in,

84-85 Thymidine kinase, hepatitis B viruses and,

198 Thyroid hormone receptor c-erbA and, 90,106-108 multiple loci, 94-96 mutations, 102-104 protein function, 97-98,100-101 protooncogene, 93-94 structure, 97 hepatocellular carcinoma and, 204 Thyroid hormone response elements (TREs),c-erbA and, 97-99,101-102,

106,108 Tick-borne encephalitis, Zilber and, 10-11,

13,15,20,24 Tolerance, cancer immunotherapy and,

284 Toxicity, cancer immunotherapy and chemoimmunotherapy, 281 current strategies, 262-265,268,270,

272 improvement attempts, 283-284,292 Trans-activation, hepatitis B viruses and,

Transcription factors, fibroblast growth factors and, 130 Transfection breast cancer, p53 expression in, 71 c-erbA and, 99-101,105,108 cancer immunotherapy and, 257,

272-273.288 cytotoxic T lymphocytes and, 228,234,

238 fibroblast growth factors and, 129,

145-148 hepatitis B viruses and, 195,198 Wilms’ tumor and, 56 Transforming growth factora, c-erbA and,

92 Transgenes, hepatitis B viruses and, 168,

180,195-196,199,206 Translation fibroblast growth factors and, 155 gene expression, 130-132 oncogenic potential, 147-148 protein structure, 118-119,123,125 hepatitis B viruses and, 187,193-194,

207 Translocation c-erbA and, 89 fibroblast growth factors and, 119-120,

123 hepatitis B viruses and, 191,204 Wilms’ tumor and, 48,53,61 TREp, c-erbA and, 100-101,104,107 TRIAC, c-erbA and, 94,98 3,5,3’-Triiodo-~-thyronine (Ts) , c-erbA and,

94-96,98-99,101-102,104-105

Tumor angiogenesis factor, fibroblast growth factors and, 149 196-199,203,210 Tumor burden, cancer immunotherapy Transcription and, 254255 breast cancer, p53 expression in, 71,84 Tumor debulking, cancer immunotherapy c-erbA and, 97-98,100-108 and, 282,286-287,293 cytotoxic T lymphocytes and, 238 Tumor grafts, cancer immunotherapy and fibroblast growth factors and, 124,133, chemoimmunotherapy, 276 147,155 critical factors, 249-250,254 gene expression, 128-132 current strategies, 255,261,264 hepatitis B viruses and, 180,210 improvement attempts, 282 genomes, 185-187,189,191,193-195 Tumor immunogenicity, cancer immunooncogenesis of viral proteins, therapy and, 248-254 Tumor-infiltrating lymphocytes (TILs), 196-199 viral DNA, 202-203,205 cancer immunotherapy and, 246 Wilms’ tumor and, 54-55,58 critical factors, 249-250,255

340

INDEX

Tumor-infiltrating lymphocytes (continued) current strategies, 260, 268, 270-273, 278 improvement attempts, 289,291-292 Tumor necrosis factor, cancer immunotherapy and, 246, 259, 276, 291 Tumor necrosis factor a,cancer immunotherapy and, 283,286,288-289, 291-292 chemoimmunotherapy, 275, 280 critical factors, 253-254 current strategies, 247, 259-263, 265-269,272-273 Tumor suppressor genes hepatocellular carcinoma and, 203, 208-210 Wilms’ tumor and, 43-44, 52, 59-61 Tumor vaccines, cancer immunotherapy and, 247,251,256,287 Tumors, see also Wilms’ tumor, genetics of breast cancer, p53 expression in, 69, 73-76,82-83 immunochemistry, 78-81 therapy, 8 4 8 5 c-erbA and, 89 cancer immunotherapy and, 245-247, 292-293 chemoimmunotherapy, 274-281 critical factors, 248-255 current strategies, 255, 257-261, 264273 improvement attempts, 282,284-292 cytotoxic T lymphocytes and, 228, 233, 241 function, 235-240 fibroblast growth factors and, 116, 149-153 angiogenesis, 153-155 biological function, 140 gene expression, 128-129 oncogenic potential, 145, 148 protein structure, 124-125 hepatitis B viruses and, 168, 170, 208, 210-211 epidemiology, 178-1 79 oncogenesis of viral proteins, 195-196,198-199 pathogenicity, 180, 182 viral DNA, 200,203-207

Zilber and, 18-19, 21, 25-27, 29,34 development of immunology, 32-33 oncogenic viruses, 30 origin, 30-32 Typhoid, Zilber and, 8 Tyrosine, c-erbA and, 92-93 Tyrosine kinase, fibroblast growth factors and, 132-133,135,137, 142

v V-UbA, 90-93,109 function, 105-106 mutations, 102-104 protein function, 97-98, 100-102 structure, 9 6 9 7 v-ErbA, 93-94, 108 function, 97-102 gene expression, 104105 mutations, 102-104 v-erbB, function, 90-93 Vaccination breast cancer, p53 expression in, 84 cancer immunotherapy and, 247, 251, 293 current strategies, 256, 279 improvement attempts, 287-288 hepatitis B viruses and, 170-1 71, 173 Zilber and, 8, 27-28 Vaccinia virus breast cancer, p53 expression in, 84 Zilber and, 8-9 Vascular leak syndrome, cancer immunotherapy and, 262-263 Vascularization, fibroblast growth factors and, 153-154 Vimentin, fibroblast growth factors and, 143 Viral DNA, hepatocellular carcinoma and, 199-208 Viral genes, hepatitis B viruses and, 192-195,210 Viral proteins breast cancer, p53 expression in, 72-75. 85 hepatitis B viruses and, 195-199 Virion assembly, hepatitis B viruses and, 191-192

341

INDEX

Viruses, Zilber and, 8-9, 12-13, 18-19, 24-25,27, 29 legacy, 34-36 oncogenes, 30-32 origin of tumors, 30-32 Vitamin A, hepatocellular carcinoma and, 204 Vitamin D,, c-erbA and, 93, 108

W WAGR syndrome, Wilms’ tumor and functional studies, 62-63 genetic loci, 48-51 Knudson model, 46-47 Wl gene, 53-54,60 Wilms’ tumor, genetics of, 41, 6 3 functional studies, 61-63 genetic loci, 46-53 histology, 42-43 Knudson model, 43-46 W T l gene characterization, 54-61 isolation of, 53-54 Woodchuck hepatitis B virus (WHV), 169 genomes, 188, 193 oncogenesis of viral proteins, 197 pathogenicity, 180-184 viral DNA, 200, 202, 205-207 Woodchuck hepatocellular carcinoma, viral DNA, 205-208 Wound healing, fibroblast growth factors and, 141-142 W T I gene, Wilms’ tumor and, 41,63 characterization, 54-61 isolation, 53-54

X X protein, hepatitis B viruses and, 195, 197-198 Xenopus, fibroblast growth factors and, 120, 129, 131, 133-134, 142 Xenopus tamis, c-erbA and, 94

Y Yeast c-erbA and, 98 fibroblast growth factors and, 122-123, 126 Wilms’ tumor and, 55 Zilber and, 8, 16-17

Z Zilber, Lev, 1-3 arrest of, 5-8 exploit in Taiga, 9-13 family of, 22-24 first discovery of, 3-5 immunology of cancer, 24-27 imprisonment of, 13-22 last efforts, 33-34 legacy of, 34-37 oncogenic viruses, 30 origin of tumors, 30-32 reunification with world scientific community, 28-29 Rous sarcoma virus, 29-30 Stalin, 27-28 tumor immunology, 32-33 viruses, study of, 8-9 Zinc fingers, Wilms’ tumor and, 55-56, 58, 60

This Page Intentionally Left Blank