Physiological and Ecological Adaptations to Feeding in Vertebrates
Editors
J. Matthias Starck Department of Biology 11, University of Munich (LMU), Germany
Tobias Wang Department of Zoophysiology University of Aarhus, Denmark
Science Publishers, Inc. Enfield (NH),USA
Plymouth, UK
SCIENCEPUBLISHERS, INC. Post Office Box 699 Enfield, New Hampshire 03748 United States of America Internet site :http://www.scipub.net
[email protected](marketing department)
[email protected](editorial department)
[email protected](for all other enquiries) Library of Congress-in-Publication Data Physiological and ecological adaptations to feeding in vertebrates / editors, J. Matthias Starck, Tobias Wang p. ; cm. Includes bibliographical references and index. ISBN 1-57808-246-3 1.Digestion. 2. Physiology, Comparative. 3. Adaptation (Biology). 4. Animal Feeding. I. Starck, J. Matthias, 1958-11 Wang, Tobias. [DNLM: 1. Adaptation, Physiological. 2. Vertebratesphysiology. 3. Digestive Physiology. 4. Digestive Systemanatomy & histology. 5. Feeding Behavior--physiology. QP 82P5783 20041 QP145.P46 2004 573.3'16--dc22
ISBN 1-57808-246-3
O 2005, Copyright Reserved All right reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying or otherwise, without the prior permission of the publisher. The request to produce certain material should include a statement of the purpose and extent of the reproduction. Published by Science Publishers Inc., Enfield, NH, USA Printed in India
All animals must eat to obtain energy for resting metabolism and to cover the energetic expenses associated with reproduction and behavioral activities. Utilization of the ingested meals requires digestion and subsequent absorption of nutrients. Given the extreme diversity in foods selection and food abundance between different environments and the large differences in metabolic needs among different animals, it is not surprising that feeding habits and strategies differ enormously among vertebrates. Some species which inhabit environments where food abundance is either scarce or fluctuates on a seasonal basis exhibit extraordinary adaptations to long-term fasting. Other species show extraordinary specialization for procuring, subduing and digesting particular foods items. Such morphological and physiological specializations have evolved in response to environmental conditions, but may be physiologically or functionally constrained. Apart from the need to acquire energy and absorb nutrients, the form and functionsof the gastrointestinal organs have been shaped to reject a plethora of antigens, bacteria and viruses that attempt to invade the body through this open barrier. In recent years interest in the gastrointestinal tract has come in to focus for studies on physiological and evolutionary adaptations to the environment. Physiological ecology and functional ecological morphology have increasingly recognized the model character of the gastrointestinaltract for studies in physiological and ecological adaptation to fluctuating environmental conditions. Apart from studies seeking to understand the basic biological questions of how animals interact with their environment, a number of comparative studies have attracted model organisms to investigate particular physiological mechanisms shared among all animals. Given the recent surge in interest pertaining to the physiology, morphology and function of the gastrointestinal organs and the process of digestion, it seemed timely to us to summarize the current state of the knowledge taking an integrative view. We intended to present a perspective that focuses on the gastrointestinaltract from an integrative ecological and evolutionary perspective, which is based on a physiological foundation where basic mechanisms are understood and quantified. We have brought together experts from a variety of specialized fields from comparative morphology through ecological and molecular physiology, immunology and ecology.All are involved in studies of different aspects of the gastrointestinal tract, but
iv
Physiological and ecological adaptations t o feeding in vertebrates
each applies different techniques and takes a different intellectual and experimental approach to understanding the gastrointestinalfunction. We are fully aware that this book cannot cover all aspects of physiological and ecological adaptations to feeding in vertebrates, but we hope it stimulates discussion and future interest in the question of how organisms adjust their form and function to fluctuating external conditions. The book opens with a comparative morphological chapter that outlines the evolution of the feeding apparatus in vertebrates. The following three chapters provide reviews of concepts of digestive efficiency, absorption models, the role of adaptation and constraints in the evolution of the gastrointestinal tract. The remainder of the book comprise a selection of chapters by experts delving into ecological questions to nutrient absorption in different taxa of vertebrates. Although the book could not cover all topics relevant for the adaptation of the gastrointestinaltract, we think it provides an overview of our present state of knowledge. Also, each chapter provides a perspective paragraph that will hopefully stimulate future research. The book was compiled as a state of the art document and is addressed to all those seriously interested in physiological and ecological adaptations of the gastrointestinal system of vertebrates. This includes graduate student as well as professionals from such fields as animal science, vertebrate biology, veterinary science, animal nutrition and medical gastroenterology. On behalf of all authors we would like to thank Paul Andrews, Victor Apanius, Michael V. Bell, Walter Bock, Colin Brauner, Dominique Burrau, Ian Gibbins, Kirrtberly Hammond, M.M. Hemphiers, Susanne Holrngren, Ian Hume, William Karasov, Marek Konarzewski, Mads Lomholt, Carlos Martinez del Rio, Adam Summers, Scott McWilliams, Mike Rust for reviewing individual chapters of the book. The reviewers provided valuable comments and advise for the individual chapters. Munich, J. Matthias Starck June 2004
Aarhus, Tobias Wang
LIST OF CONTRIBUTORS
Augusto S. Abe Departamento de Zoologia, c.p. 199, Universidade Estadual Paulista, 13506-900, Rio Claro, SP, Brasil. JohnnieB. Andersen Department of Zoophysiology, University of Aarhus, Universitetsparken, Aarhus, Denmark. Denis V. Andrade Departamento de Zoologia, c. p. 199, Universidade Estadual Paulista, 13506900, Rio Claro, SP, Brasil. Phil F. Battley Department of Mathematics and Statistics, University of Otayo, Dunedin, Nezu Zealand Hannah V. Carey Department of Comparative Biosciences, University of Winconsin, School of Veterinary Medicine, 2015, Linden Dr. Madison, W I 53076, U S A . Luis E.C. Conceiqiio CCMAR- Centro de Clencias do Mar, Universidade do Algarve, Campus de Gambela, P-8000-117 Faro, Portugal. Ariovaldo P. Cruz-Neto Departamento de Zoologia, c. p. 199, Universidade Estadual Paulista, 13506900, Rio Claro, SP, Brasil. JamesW. Hicks Department of Ecology and Evolutionary Biology, University of California, Irvine, Irvine, C A . Anna Holmberg Department of Zoophysiology, University of Goteborg, Box 463, SE 405 30 Goteborg, Sweden. Susanne Holmgren Department of Zoophysiology, University of Goteborg, Box 463, SE 405 30 Goteborg, Sweden. Ian Hume University of Sydney, Biological Sciences A08, NS W 2006, Australia.
vi
Physiological and ecological adaptations t o feeding in vertebrates
William H. Karasov Department of Wildlife Ecology, 226 Russell Labs, 1630 Linden Drive, University of Wisconsin, Madison, W I 53706, U S A . Kirk Klasing Department of Animal Science, University of Calijornia, Davis, C A 9561 6 , U S A . Carlos Martinez del Rio Department of Zoology and Physiology, University of Wyoming, Laramie, W Y 82071 -31 66, U S A . David J. McKenzie Centre de Recherche sur les ~ c o s ~ s t 2 mMarins es et Aquacoles de L'Houmeau, UMR 10 CNXS-lfvemer, Place du Stminaire, B.P 5,171 37 L'Houmeau, France. Todd J. McWhorter Department of Wildlife Ecology, 226 Russell Labs, 1630 Linden Drive, University of Wisconsin, Madison, W I 53706, U S A . Scott R. McWilliams Department of Natural Resourses, 116 Coastal Institute in Kingston, University of Rhode Island, Kingston, RI 02881, U S A . Theunis Piersma Department of Animal Ecology, Centrefor Ecological and Evolutionary Studies, University of Groningen, PO Box 14,9750 AA Haren, The Netherlands, and Department of Marine Ecology, Royal Netherlands Institute for Sea Research (NIOZ),PO Box 59,1790 A B Den Burg, Texel, The Netherlands Ivar Rsnnestad Department of Zoology, University of Bergen, Allegatan 41, N-5007 Bergen, Norway. Margaret Rubega Department of Ecology and Evolutionary Biology, University of Connecticut, S ~ O W CT S , 06269-3043, U S A . Kurt Schwenk Department of Ecology and Evolutionary Biology, University of Connecticut, Storrs, CT 06269-3043, U S A . J. Matthias Starck Department of Biology 11, University of Munich (LMU),Grophaderner Str. 2, D-82152, Planegg-Martinsried, Germany. Tobias Wang Department of Zoophysiology, Building 131, University of Aarhus, Universitetsparken,Aarhus, Denmark. Blair 0.Wolf Biology Department, 167 Castetter Hall, The University of New Mexico, Albuquerque, N M 87131-1 091, U S A .
CONTENTS
Preface List of Contributors 1
The Diversity of Vertebrate Feeding Systems Kurt Schwenk and Margaret Rubega
2
Concepts of Digestive Efficiency Ian D. Hume
3
Carbohydrate Hydrolysis and Absorption: Lessons from Modeling Digestive Function Todd J. Mc Whortev
4
Digestive Constraints in Mammalian and Avian Ecology William H. Karasov and Scott R. Mc Williams
5
Paracellular Intestinal Absorption of Carbohydrates in Mammals and Birds Todd J.Mc Whovtev
6
Mass Balance Models for Animal Isotopic Ecology Carlos Martinez del Rio and Blair 0. Wolf
7
Structural Flexibility of the Digestive System of Tetrapods -Patterns and Processes at the Cellular and Tissue Level J. Matthias Starck
8
Adaptive Interplay between Feeding Ecology and Features of the Digestive Tract in Birds Phil F. Butt ley and Theunis Piersma
9
Gastrointestinal Responses to Fasting in Mammals Lessons from Hibernators Hannah V Carey
229
10
Interplay between Diet, Microbes, and Immune Defenses of the Gastrointestinal Tract Kirk Klasing
255
11
Effects of Digestion on the Respiratory and Cardiovascular Physiology of Amphibians and Reptiles Tobias Wang,Johnnie B. Andersen and James W . Hicks
279
Physiological and ecological adaptations t o feeding i n vertebrates
viii
12
Specific Dynamic Action in Ectothermic Vertebrates: A General Review on the Determinants of Post-Prandial Metabolic Response in Fishes, Amphibians, and Reptiles Denis I? Andrade, Ariovaldo I? Cruz-Neto,Augusto S. Abe, and Tobias Wang
13
Control of Gut Motility and Secretion in Fasting and Fed Non-Mammalian Vertebrates Susanne Holmgren and Anna Holmberg
14
Effects of Dietary Fatty Acids on the Physiology of Environmental Adaptation in Fish. David J. McKenzie
15
Aspects of Protein and Amino Acids Digestion and Utilization by Marine Fish Larvae Ivar Rsnnestad and Luis E.C. Concei@io
Index
Diversity of Vertebrate Feeding Systems Kurt Schwenk* and Margaret Rubega Department of Ecology and Evolutionary Biology, University of Connecticut, Storrs, CT, USA
SYNOPSIS The vertebrate gut tube can be divided into "front-end and "back-end" components according to topology, function, and research tradition. The purpose of front-end feeding systems is to acquire food to be delivered to the back-end for chemical digestion and assimilation. In accomplishing this, the feeding system faces as many as five separate mechanical tasks recognized as "feeding stages:" capturelsubjugation, ingestion, transport, processing, and swallowing. In general, aquatic species exploit the high density of water to manipulate food items by modulating water flow through the mouth and pharynx. In contrast, terrestrial vertebrates typically employ some form of hyolingual feeding in which movements of the tongue and hyobranchial apparatus take the place of water flow in capturirrg, supporting and manipulatingfood. The condition of the bolus, when swallowed, varies markedly among taxa. Most ectotherms process their food little, if at all, whereas mammals and birds typically reduce their food to small particles (in mammals, by oral mastication; in birds, within the "gastric mill" of the gizzard). This fundamental difference probably relates to the need of endotherms to increase gut passage rates.
The study of vertebrate feeding proceeded historically along two lines. On the one hand, functional morphologists examined the myriad mechanisms by which vertebrates procure, process and swallow food. On the other, physiologists studied the structure and function of the gut during digestion and absorption of the food swallowed. These distinct fields of endeavor reflect not only different topological foci ("front end" vs "back end" of the gut tube),
2
Physiological and ecological adaptations t o feeding in vertebrates
but a long-standing division of the life sciences into morphology and physiology (Owen, 1866; Gegenbauer, 1878; Woodger, 1929; Russell, 1916; see Schwenk, 2000d for discussion). Although the modern separation of physiological and morphological approaches in the study of vertebrate feeding is justified to some extent by differences in research methodology, their continued isolation seems to be more a manifestation of historical inertia than of biological reality. It is clear that form and function of the gut's front end and back end are inseparately related and that evolutionary changes in one are likely to have consequences for the other. This statement is so obvious as to be nearly superfluous. Nonetheless, it is worth emphasizing that the traditional partitioning of the field obscures the reality of feeding system evolution. Feeding function Sensu Lato requires the functional and anatomical integration of many separate components. The evolution of such integrated systems poses a major challenge in current evolutionary theory and is poorly understood in the context of traditional atomistic or reductionist approaches (Wagner and Schwenk, 2000; Schwenk, 2001b; Schwenk and Wagner, 2001).While some kind of rapprochement is necessary in the study of vertebrate feeding systems is widely acknowledged-indeed, it is a theme central to this volume-a truly synthetic understanding of the evolution of the feeding system still eludes us. Given the functional and evolutionary integration that necessarily exists between front and back ends of the gut, it is not surprising that many common factors can be found to have influenced their phenotypes-ectothermy vs endothermy, and diet. Nonetheless, it is apparent that common environmental "problems" (e.g.herbivory) have often resulted in clade-specificsolutions and that similarities among taxa in one end of the system (e.g. longer relative gut length)are not always paralleled in the other (e.g.different mechanisms of reduction). Thus, integration does not imply that front and back ends must always evolve in lock step-only that evolutionary changes in one are likely to have consequencesfor the other. Indeed, the extent to which feeding mechanisms and gut physiology are deterministically coupled remains an open and important question. In this chapter we provide an overview of the mechanical tasks faced by vertebrates when they feed and consider some of the diverse ways these tasks are executed in different taxa. The feeding systems of larvae are not considered in this review due to limited space and the fact that many larval forms are exceptionally small, particularly in fishes, and operate within the different physical paradigm of low Reynolds numbers where viscous forces predominate. Clearly the question of how such organisms feed and metamorphose into a high Reynolds number world is of great interest, but one we cannot treat here. Readers are referred to reviews provided by Sanderson and Kupferberg (1999)and Wassersug and Yamashita (2001).
Vertebrate feeding systems
3
MORPHOLOGY OF THE FEEDING APPARATUS A complete account of feeding system morphology is beyond the scope of this chapter but an introduction to relevant structures and terminology will help to clarify the following sections. Overviews of trophic morphology in fishes can be found in Lauder (1985a), Vandewalle et al. (1994), and Motta and Wilga (2001), and for tetrapods in Bramble and Wake (1985), H"iiemae and Crompton (1985),Schwenk (2000a), and in the taxon-specific chapters within Schwenk (2000b).General references that include excellent sections on feeding structure include Liem et al. (2001)and Hildebrand and Goslow (2001).As discussed, a traditional consideration of feeding form and function begins at the head and ends at the esophagus, so we confine our discussion to front-end components of the feeding system. The anterior end of the gut tube in deuterostome embryos opens through the stomadeurn to form the mouth. Between the opening of the mouth and the entrance to the esophagus lies a cavity somewhat arbitrarily divided into an anterior buccal cavity and a posterior pharynx. A hallmark of vertebrate evolution was the origin of a series of U-shaped skeletal arches supporting the pharynx. These so-called visceral arches form from a novel embryonic tissue, the neural crest (e.g. Thorogood, 1993).Ancestrally the pharyngeal skeleton formed a kind of "basket" that functioned as a filter to trap suspended food particles brought into the mouth by ciliary currents. The evolution of joints and associated branchiomeric musculature led to flexion of the skeletal elements and active pumping of water for feeding and respiratory function (Mallatt, 1996).This system is retained more or less unchanged in larval lampreys. The origin of jaws from an anterior visceral arch was probably associated with increasingly active and predaceous behavior (Northcutt and Gans, 1983)and a transition from suspension feeding to prehension of individual food particles (Mallatt, 1984).Arguably, jaws were a key innovation in the vertebrate lineage, leading to an explosion of gnathostome (jawed vertebrate) diversity and ultimately the demise of most jawless clades. We can reasonably infer that a great deal of this diversity was engendered by the trophic flexibility of jaw-based feeding systems, which permitted the invasion of new adaptive zones. Thus the jaws and their associated teeth and musculature, were established early in vertebrate evolution as the central elements of vertebrate feeding system evolution. They are the focus of most feeding studies. Vertebrate jaws are complex structures with multiple evolutionary and developmental sources. They develop from cartilages of the first visceral arch, i.e. the palatoquadrate in the upper jaw and the mandibular (or Meckel's) cartilage in the lower jaw. In most adult vertebrates the jaws are primarily composed of dermal (membrane)bones that invest the cartilages during later development.A few parts ossify directly as endochondral bones, notably at
4
Physiological and ecological adaptations t o feeding i n vertebrates
the jaw joint, to form an upper quadrate bone and a lower articular, but the cartilages usually atrophy and in many species disappear. In basal bony fishes and tetrapods, the upper jaw fuses to other dermal bones of the facial skeleton,but retains independence, or at least limited mobility, in many taxa, especially fishes (e.g. Lauder, 1985a; Motta and Wilga, 2001). Indeed, the ability to protrude the jaws is an essential component of feeding in many fishes, particularly in suction feeders (Fig. 1.1) (Lauder, 1985a; Westneat, 1990; Motta and Wilga, 2001; Wilga et al., 2001). In tetrapods, the jaws are never protrusible; however, several lineages, notably birds and squamate reptiles, have evolved kinetic joints in the dermal skull so that the upper jaws and other skull elements can flex relative to the braincase (Beecher, 1962; Frazzetta, 1962;Bock, 1964; Zusi, 1993;Herrel et al., 1999;Hoese and Westneat, 1996; Arnold, 1998; Bout and Zweers, 2001; Metzger, 2002). Such "cranial kinesis" is most highly developed in advanced (macrostomatan)snakes, in which upper and lower jaw and palatal bones are independently and unilaterally mobile (Gans, 1961; Cundall and Greene, 2000). This increases gape and generates ratchet movements of the toothed elements, one side at a time, to pull the snake's head and body over a prey item. Mammals are distinguished by an akinetic and generally robust skull (Davis, 1961).The endochondral jaw joint bones of other vertebrates have been miniaturized and displaced to the middle ear in mammals, where they contribute to the auditory apparatus (Allin, 1975; Novacek, 1993; Rowe, 1996).A new jaw joint has evolved between two dermal bones, the dentary of the lower jaw and squamosal of the upper (thelatter element is usually fused with others to form the temporal bone).A diagnostic feature of living mammals and closely related fossil taxa is the presence of only a single bone, the dentary in the lower jaw. The dentary is, itself, a developmental composite, comprising the fusion of six separate Anlage (Atchley, 1993). The dentary bones of each side are joined anteriorly by a fibrous symphysis to form the mandible.The strength of the symphysis and the extent to which it transmits forces from one half of the mandible to the other varies among species (e.g. Beecher, 1979;Lieberman and Crompton, 2000). Teeth evolved in association with jaws. They are composed primarily of dentine and enamel, ancient hard tissues that invested the armor plates of ancestral jawless fishes (Butler and Joysey, 1978; Reif, 1982). Primitively, teeth were found throughout the buccal cavity and pharynx on various elements of the palate and pharyngeal skeleton, but palatal and pharyngeal teeth are often lost, especially in tetrapods.The marginal teeth of the jaws are restricted to the dentary of the lower jaw, and the maxilla and premaxilla of the upper jaw in bony vertebrates. In mammals and crocodilians, the teeth are rooted in deep sockets within the bone, but in most vertebrates they are cemented to the apical or medial jaw surfaces. The exposed, or crown portion of the tooth varies extensively in form, even among closely related species in some cases, variation that may be functionally related to
Vertebrate feeding systems
s IM
IHG
IHG
Fig. 1.1. Feeding in aquatic vertebrates usually involves manipulation of food particles indirectly through the modulation of water flow (left), whereas in tetrapods, the tongue and hyobranchial apparatus take over this role (right). Suction feeding in bony fishes (left) results from an explosive expansion of the mouth and pharynx caused by protrusion of the jaws, hyoid retraction, and opercular abduction. The negative pressure generated within the mouth causes an inrush of water that drags prey in. Tetrapods often capture food with the tongue which also supports and manipulates it within the mouth (right). Cyclical movements of the hyolingual apparatus transport the food item back to the pharynx for swallowing. ASHG: anterior suprahyoid muscle group; Bh: basihyal element of hyobranchium; Cb: ceratobranchial element of hyobranchium; DM: depressor mandibulae muscle; EAC: external adductor musles; ECC: epaxial cervical muscles; IAC: internal adductor muscles; IHG: infrahyoid muscle group; IM: intermandibularis muscle. Left side figures from Karel F. Liem (1979). Reprinted by permission of WileyLiss, Inc., a subsidiary of John Wiley and Sons, Inc. Right side figures from Bramble and Wake (1985), reprinted by permission of the publisher and President and Fellows of Harvard College.
6
Physiological and ecological adaptations to feeding in vertebrates
the types of food eaten and the manner in which it is procured and processed in the mouth (Fig. 1.2).This is especially true for mammals because they masticate their food (see later).The pharyngeal teeth of some derived teleost fishes show similar, diet-based adaptations in crown form (e.g. Liem, 1973; Sibbing, 1991).Tooth reduction or loss is relatively commonplace among many clades. Turtles and birds have lost their teeth altogether and replaced them with a horny (keratinous)investment of the jaws called a beak or bill.
MORTAR-PESTLE
SHEARING BLADES
w 0
.
0 .
,
m SERIAL ARRAYS -LOW PROFILE BLADES
Fig. 1.2. Crown form in mammalian teeth is closely tied to the nature of the food eaten. Functional specialization of teeth is part of a suite of derived mammalian traits associated with the evolution of mastication. Mastication results in comminution of food, i.e. its reduction to a slurry of fine particles mixed with saliva. From Hiiemae (2000), reproduced with permission of Elsevier Science.
Vertebrate feeding systems
7
In most vertebrates, lower jaw motion is mostly limited to dorsoventral movements. Jaw closing is effected by adductor mandibulae musculature that is variously subdivided in different taxa. Adductor muscles run from the cranium to the lower jaw and are innervated by the trigerninal nerve (cranial nerve V). In most fishes, jaw opening is caused by the action of ventromedial hypobranchial muscles that run anteriorly from the pectoral girdle to the lower jaw (Wilga et al., 2000). These are innervated by spinooccipital nerves and/or the hypoglossal (c.n. XII). In nonmammalian tetrapods, a depressor mandibulae takes over jaw opening. It runs from the back of the cranium and neck to the retroarticular process of the mandible, depressing it by pulling up behind the jaw joint. Since this muscle is developmentally and evolutionarily derived from the superficial constrictor musculature of the throat, it is innervated by the facial nerve (c. n. VII). Jaw opening is often accompanied by elevation of the cranium caused by contraction of the epaxial neck musculature. Mammals use a novel muscle, the digastric, to depress the lower jaw. Its name derives from the fact that in many species it comprises two distinct bellies separated by a short tendon, each innervated by a different nerve (c.n.V and VII).It runs from the paroccipital process of the skull base to the anterior end of the mandible. During the evolution of mastication (see below) mandibular movements in most mammals became complex. They are mostly dorsoventral in carnivorous species but in herbivores include dramatic mediolateral and/or anteroposterior movements. The jaw joint is variously modified to accommodate such mobility. Mastication in mammals is associated with the evolution of a novel adductor muscle, the masseter, running from the zygomatic arch (cheek bone) to the lateral surface of the mandible. The masseter adds lateral and anterior components to jaw movement that are balanced by the dorsal and medial components of pterygoideus and temporalis adductors. The pharyngeal skeleton is an essential part of the feeding apparatus in both fishes and tetrapods. Ancestrally, there were seven visceral arches constituting the pharyngeal skeleton (splanchnocranium) of gnathostomes. Each arch is composed of several jointed elements joined in the ventral midline. The first, most anterior arch is the mandibular, comprising upper and lower jaws. The second is the hyoid arch and the remaining five the branchial (or gill) arches. The upper part of the hyoid arch in fishes (hyomandibula)runs from the jaw joint to the neurocranium and variously supports, braces or suspends the jaws. It is homologous to the columella (stapes),a middle ear ossicle, in tetrapods. The lower part (ceratohyal)is usually highly mobile. When at rest, the paired ceratohyals lie within the arc of the lower jaw, but when pulled back by hypobranchial muscles, the hyoid arch swings posteroventrally, depressing the floor of the pharynx and increasing its volume. Rapid hyoid retraction, along with elevation of the neurocranium and lateral movement of the opercular bones, are used by many fishes to generate suction within the mouth and pharynx to create currents for suspension
8
Physiological and ecological adaptations t o feeding i n vertebrates
feeding,prey capture, and/or prey manipulation (seebelow).Many elements of the pharyngeal skeleton in bony fishes bear teeth (as do palatal and neurocranial bones) which are used to hold, grasp, manipulate, transport, and process food. In some teleosts, upper and lower tooth-bearing elements of the pharyngeal arches interact to form an internal set of "pharyngeal jaws" used in elaborate manipulatory and processing behavior (Liem, 1973; Liem and Osse, 1975; Liem and Greenwood, 1981; Lauder, 198313, 1985a; Sanford and Lauder, 1989;Sibbing, 1991; Vandewalle et al., 1994; Galis and Drucker, 1996),a condition known as "pharyngognathy" (Liem and Greenwood, 1981;Fig. 1.3). Fish use the pharyngeal skeleton, in particular the hyoid arch, to create feeding currents through volumetric changes of the pharynx. This is only possible in water because the density of prey is closely matched by the surrounding density of the medium. Rapid water flow is thus able to overcome
Fig. 1.3. Schematic representation of the pharyngeal jaw apparatus in four families of teleostean fish. The pharyngeal jaws are lightly stippled. Letters refer to muscle groups that act on the jaws to produce complex crushing, grinding, and transport movements. From Liem and Greenwood (1981), reproduced with permission of the Society for Integrative and Comparative Biology.
Vertebrate feeding systems
9
the prey's inertia. Tetrapods could not employ such inertial suction when they first began to feed in the terrestrial environment, but they nonetheless exploited the hyoid mobility inherited from their piscine ancestors when feeding on land (Shaffer and Lauder, 1988; Gillis and Lauder, 1994, 1995; Reilly, 1996).They accomplished this primarily through the evolution of a novel structure, the mobile, muscular tongue. The tongue evolved by elaboration of hypoglossal muscles associated with the hyoid and the first two or three branchial arches (Kallius, 1901). The reduced (compared to fishes) pharyngeal skeleton of tetrapods is called the hyobranchial apparatus or hyobranchium, and it supports the tongue and throat musculature (e.g. Fiirbringer, 1922; Weissengruber et al., 2003). The hyobranchium of nonmammalian tetrapods is often inaccurately referred to as the "hyoid apparatus," a term appropriately applied only to mammals in which the branchial arch contribution is lost or greatly reduced (Schwenk, 2000a). In combination,the tongue and hyobranchium are called the hyolingual apparatus. Instead of modulating water flow, movements of the tetrapod hyobranchium move the tongue which, in effect, takes the place of water in capturing, supporting, and manipulating food particles (Fig. 1.1).Tongue movement that is extrinsically generated by hyobranchial movment is enhanced in many tetrapods by intrinsically generated shape changes of the tongue's soft tissues (seebelow).In secondarily aquatic tetrapods that revert to suction feeding, the hyobranchial apparatus is once again used to modulate the flow of water by changing pharyngeal volume (Lauder, 1985a;Van Damme and Aerts, 1997; Deban and Wake, 2000; Aerts et al., 2001; Lemell et al., 2002). It is noteworthy that in these species the tongue is almost always reduced or even lost -an indication of its uniquely terrestrial role in feeding (Brambleand Wake, 1985). The tongue is a critical element of the tetrapod feeding system, largely overlooked in earlier studies of feeding. In many taxa, it participates in all stages of feeding, from prey capture to swallowing. Its morphology is diverse, ranging from little more than an epithelium covered part of the hyobranchium, to an astoundingly complex muscular organ capable of extreme changes in length and shape. The protean nature of tongue form in some taxa (notably mammals and some squamate reptiles) arises from its unusual biomechanical properties. The tongue is one of the very few vertebrate organs capable of hydrostatic deformation (Owen, 1868;Kier and Smith, 1985; Smith and Kier, 1989).Such so-called "muscular hydrostats" comprise solid muscle masses with a complex histology in which fiber systems are arrayed orthogonally, sometimes includinghelical systems as well (Kier and Smith, 1985; Smith and Kier, 1989; Schwenk, 2001a). Because the organ retains a constant volume and the intracellular fluid within it is incompressible, local or global reductions in diameter cause elongation and/or shape change. For example, myrmecophagous mammals use extreme length changes in their serpentine tongues to probe ant and termite nests (Reiss,
10
Physiological and ecological adaptations t o feeding in vertebrates
2000); generation of intraoral suction within the buccal cavity during suckling in infant mammals is made possible by an oropharyngeal seal at the back of the mouth created by shape changes in the tongue (German and Crompton, 2000);lizards, terrestrial turtles and mammals form humps and cup-shaped depressions in the tongue to hold and push the bolus during hyolingual transport through the mouth (Brambleand Wake, 1985;Hiiemae, and hydrostatic elongation of the tongue in combi2000; Schwenk, 2000~); nation with a uniquely shaped hyobranchium, provides the explosive trigger that projects the chameleon's tongue out of its mouth (Wainwright and Bennett, 1992).
OVERVIEW OF VERTEBRATE FEEDING Feeding Mode, Sensory Biology and Foraging Strategy Before a feeding bout can begin, it is necessary for an animal to find a potential food item and to identify it as appropriate. This requires the use of various sensory systems and some kind of foraging strategy. The evolutionary interplay of sensory function, foraging, and feeding mode has rarely been explored in detail, but the need for their coordinated function is obvious. Vitt et al. (2003), for example, provide a case study of squamate reptiles showing that historical changes in the feeding system have had cascading effects on the evolution of sensory systems and ultimately, community ecology and geographic distribution. The study illustrates that an understanding of the integration of feeding mode, sensory biology, and foraging mode can potentially lead to compelling explanations for higher level patterns, in this case, the structuring of squamate communities on a global scale (Vitt et al., 2003). These patterns are necessarily taxon specific but the important role other systems and behavior play in feeding biology is noteworthy. The remainder of this chapter primarily concerns feeding function from the moment an appropriate prey item is within striking distance until it passes into the esophagus. In studying this behavior, several stages of feeding are formallyrecognized. Vertebrate Feeding Stages In order to acquire and digest food successfully, it is necessary for an animal to proceed through a series of different mechanical tasks, concluding with passage of the food bolus into the esophagus. These tasks are identified technically as "feeding stages" to highlight their different functions and to facilitate description and comparisons of feeding across taxa (Bramble and Wake, 1985; Hiiemae and Crompton, 1985; Schwenk and Throckmorton, 1989;Schwenk,2000a).However, it is necessary to state several caveats about
Vertebrate feeding systems
11
feeding stages before describing them. First, as noted, their recognition is based on the differing functional roles they play during a feeding bout. As such, feeding stages can be thought of as phenotypic "solutions" to a series of mechanical "problems" encountered during the course of getting food from the environment into the esophagus.There is potentially more than one solution to any given mechanical problem, so different taxa sometimes carry out the same feeding stage in mechanistically different ways. This leads to the second point, namely that use of a common name for the "same" feeding stage in different species should not be taken as an assertion of evolutionary homology. The homology of feeding stages among taxa is an open and critical research question, as is, for that matter, the extent to which the stages can be considered evolutionary "characters" at all (Reilly and Lauder, 1990; Smith, 1994; Schwenk, 2000a; Wainwright and Friel, 2001; McBrayer and Reilly 2002b). Third, any given species or individual might exhibit only a subset of all possible feeding stages (described below). Individuals might also vary in the particular feeding stages manifested during one feeding bout compared to another, or might vary the mechanism employed during a particular feeding stage depending on the nature of the fqod eaten or other local circumstances.Finally although some stages must necessarily precede other stages (e.g. capture must precede swallowing), some do not always occur sequentially. For example, capture/subjugation and ingestion are often combined into a single stage (ingestion), whereas processing and transport cycles are occasionally alternated or interspersed (e.g. Hiiemae and Crompton, 1985; Schwenk and Throckmorton, 1989; Hiiemae, 2000; Schwenk, 2000c; McBrayer and Reilly, 2002a). In all vertebrates, feeding emerges from the complex coordinationof skull, jaw, tongue, and hyobranchial movements. In tetrapods especially these movements are rhythmic and cyclic, leading to the suggestion that they are controlled by pattern generators in the central nervous system (Dellow and Lund, 1971; Thexton, 1973; Hiiemae, 2000). Bramble and Wake (1985)proposed that there is a basic or "model" feeding cycle that we might infer represents the ancestral or primitive pattern of coordinated movements and motor patterns in tetrapod feeding. There has been a great deal of discussion regarding the utility of the Bramble-Wake model in predicting the feeding kinematics of different taxa and during different feeding stages (e.g.Schwenk and Throckmorton, 1989;Reilly and Lauder, 1990; Delheusy and Bels, 1992; Bels et al., 1994;Lauder and Gillis, 1997;Schwenk, 2000c; Herrel et al., 2001), with no consensus emerging. Regardless, the critical point is that tetrapod feeding, at least, results from rhythmic, cyclical, and coordinated movements of the aforementioned parts. The fundamental unit of this behavior is the "gape cycle," representing a single excursion of the jaws from closed to open and back to closed (Bramble and Wake, 1985;Hiiemae and Crompton, 1985; Schwenk, 2000a).Movements of all other components of the feeding system are usually described relative to the gape cycle to facilitate comparisons,
12
Physiological and ecological adaptations t o feeding in vertebrates
although in some cases jaw movements are relatively trivial compared to movements of other parts, such as the hyobranchial apparatus. A single feeding bout represents a variable number of sequential gape cycles, with modulation of kinematic patterns occurring throughout, depending on the mechanical task at hand (the feeding stage) and the position and condition of the bolus. Although the different feeding stages outlined below can sometimes be differentiated qualitatively as well as quantitatively, only rarely are the kinematic transitions between them sharply defined. Active prey must first be captured and subjugated before it can be eaten. Once subdued, it can be brought into the mouth (ingestion).These actions represent nominally the first and second stages of feeding. However, in most vertebrates food is captured with the mouth so that capture and subjugation of prey occurs at the same time that it is moved into the oral cavity. Thus, in most vertebrates the single stage of ingestion accomplishes all three functions. However, in many species, particularly mammalian predators that feed on relatively large prey, a separate capture/subjugation stage is necessary before the food can be ingested. Such mammals typically run down and attack prey with the forelimbs and jaws to subdue and kill them (Ewer, 1973; Van Valkenburgh, 1996). Once quiescent, the prey can be consumed and ingestion initiated. Primates also typically use the forelimbs to grab, manipulate and sometimes kill a food item before placing it in the mouth. Although less common, a separate capture/subjugation stage occasionally occurs in nonmammalian taxa, such as fish-spearing with the bill in some wading birds, or raptorial capture and killing with the feet in other birds. The venomous crotalid snakes (rattlesnakes and kin) are particularly illustrative because after an envenomating bite they release their prey and allow it to die before ingesting it (Cundalland Greene, 2000). Rarely, capture and subjugation are performed as separate tasks, as in constricting snakes that capture a prey item with the jaws, but hold and subjugate it in coils of the body until it can be transported (Cundall and Greene, 2000).
Ingestion Ingestion refers to the transfer of a food item from the environment into the oral cavity.As noted, in most vertebrates this is accomplished with the mouth so that capture/subjugation and ingestion are combined into a single ingestion stage. In aquatic vertebrates, ingestion occurs in one of three ways: suspension feeding (sometimesinaccurately called "filter feeding"), suction feeding or jaw prehension. In suspension feeding, small food particles are collected from the water surface, water column or the benthos when water is passed through a porous structure, such as the gill rakers of the pharyngeal skeleton in many fishes or the baleen plates of mysticete whales (e.g. Northcott and
Vertebrate feeding systems
13
Beveridge, 1988;Sanderson and Wassersug, 1993;Goodrich et al., 2000; Werth, 2000b).Many aquatic birds use elaborations at the margins of the bill to filter algae and other suspended food particles from the water (Zweerset al., 1977, 1995;Kooloos et al., 1989).Some freshwater turtles draw small, floating food particles into the mouth (Belkin and Gans, 1968), trapping them in esophageal papillae when the water is expelled (Vogt et al., 1998). Suspension feeders either actively pump water into the mouth and pharynx, or engulf suspended food particles as they move their bodies forward with mouth agape. Many fishes can shift between these two modes of suspension feeding. However, drawing larger food items into the mouth requires the rapid generation of substantial negative pressure within the mouth and pharynx. This is referred to as suction feeding. As water is accelerated into the mouth, its rapid flow draws the prey along with it. There is a continuum among taxa in the amount of suction used and the extent to which the prey moves into the mouth versus the mouth over the prey (Norton and Brainerd, 1993;Liem, 1993; Nemeth, 1997;Van Damme and Aerts, 1997; Summers et al., 1998;Wainwright et al., 2001). In "inertial suction" the prey item is accelerated relative to a fixed point and moved into the mouth along with the water, whereas in "compensatory suction" the negative pressure generated by the predator is only enough to overcome the effects of its own bow wave as it moves forward, engulfing a stationary prey item. Such "ram feeders" attempt to mitigate the pressure wave by means of a large gape, capacious pharynx and unilateral flow of water. In bony fishes, suction is generated by an explosive expansion of the mouth and pharynx caused by expansive movements of the jaws, neurocranium, opercular bones and pharyngeal skeleton, especially retraction and depression of the hyoid arch (e.g. Lauder, 1983a, b, 1985a, b; Miiller and Osse, 1984; Bemis and Lauder, 1986;Bemis, 1987; Lauder and Shaffer, 1993; Liem, 1993; Gillis and Lauder, 1995; De Visser and Barel, 1998; Ferry-Graham and Lauder, 2001; Grubich, 2001; Westneat, 2001; Sanford and Wainwright, 2002). Upper jaw protrusion and kinesis create a small, round, anteriorly directed gape that increases water velocity and targets prey directly in front of the fish (Fig. 1.1). The cartilaginous elasmobranchs obviously lack the opercular apparatus and other elements of the bony fish skull and rely primarily on retraction of the hyoid arch to generate suction (Wu, 1994; Wilga and Motta, 1998; Edmonds et al., 2001; Motta and Wilga, 2001). A labial cartilage functions to restrict gape and accelerate flow. Many aquatic tetrapods, especially salamanders and turtles, have secondarily reverted to suction feeding (Erdman and Cundall, 1984;Lauder, 1985; Lauder and Shaffer, 1985; Elwood and Cundall, 1994; Lauder and Reilly, 1994; Van Damme and Aerts, 1997; Deban and Wake, 2000; Aerts et al., 2001; Deban and Marks, 2002; Lemell et al., 2002). As in fish, suction is generated by rapid expansion of the pharynx (Fig.l.4).These taxa typically have an elaborate and robust hyobranchial apparatus that is articulated in
14
Physiological and ecological adaptations t o feeding in vertebrates
Fig. 1.4. Skull (white) and hyobranchial apparatus (gray) of a larval aquatic salamander in lateral view. At left, the hyobranchium is at rest, lying flat within the throat. When retracted, at right, the downward pivoting of the ceratohyal and the rest of the hyobranchial apparatus causes a massive expansion of buccal and pharyngeal cavities to generate suction for the modulation of water flow. BP: branchial plate; CH: ceratohyal; EB1: first epibranchial. From Deban and Wake (2000), reproduced with permission of Elsevier Science.
such a way that at rest it lies flat in the floor of the mouth and throat, but when retracted it unfolds, dropping dramatically to vastly increase pharyngeal volume (e.g. Van Damme and Aerts, 1997; Deban and Wake, 2000). Esophageal expansion during the strike in some turtles helps prevent back pressure, suggesting that suction may be compensatory rather than inertial in these species (Lauder and Prendergast, 1992; Lemell et al., 2000). Longnecked turtles use a snake-like strike to propel the head toward the prey (Weisgram and Splechtna, 1992; Van Damme and Aerts, 1997,2002; Aerts et al., 2001). Patent gill slits allow fish and larval salamanders to draw water through the pharynx in one direction during suction feeding (unidirectional flow), whereas in metamorphosed salamanders, turtles, and mammals the water must exit the mouth during the compressive stage immediately following capture (bidirectionalflow), leading to a putative mechanical inefficiency (Lauderand Shaffer, 1986).Nonetheless, feeding performance in some highly aquatic adult salamanders is superior to larval forms with unidirectional flow, suggesting that morphological and behavioral adaptations can overcome this potential handicap (Miller and Larsen, 1989). A very few salamanders feed underwater using the terrestrial mechanism of tongue prehension (Deban and Wake, 2000). All neonate mammals (with the possible exception of some monotremes) use suction for ingestion of milk (e.g.German and Crompton, 2000) but suction feeding is uncommon in adult mammals. Walruses retract the tongue in the mouth like a piston to suck molluscs from the substrate (Kasteleinet al., 1994) and several other pimipeds may also use suction (Werth, 2000b). Some blunt-headed cetaceans, such as pilot whales, probably use suction for ingestion of food items such as squid (Werth,2000a, b). Jaw prehension of prey during ingestion is less common in aquatic vertebrates than suction and suspension feeding. Large, predaceous sharks, such as larnniforms and carchariniforms tend to overcome a prey item by
Vertebrate feeding systems
15
rapidly overtaking it, either engulfing it within the mouth or biting it. Some specialized biters, however, remove mouth-sized chunks of flesh from their prey using protrusible jaws and rows of sharp, serrated teeth (Frazzetta and Prange, 1987; Frazzetta, 1988, 1994; Motta and Wilga, 2001; Wilga et al., 2001).Some bony fish use the jaws and marginal teeth to grasp prey directly or to scrape food, such as algae, off the substrate (e.g.Liem, 1980; Turingan and Wainwright, 1993; Wainwright and Turingan, 1993; Alfaro and Westneat, 1999; Wainwright et al., 2000; Alfaro et al., 2001). Many species bite pieces from large prey in much the way some sharks do (Liem, 1980; Alfaro et al., 2001). Jaw feeding fish typically retain the use of suction to manipulate and transport prey within the pharynx. Crocodilians use the jaws to capture prey in water, typically with a rapid, sideways jerk of the head (Busbey, 1989; Davenport et al., 1990; Cleuren and De Vree, 2000). Some species, such as gavials, have long, narrow snouts lined with needlelike teeth specialized for the prehension of fish in water (Cleuren and De Vree, 2000). Either the tongue or the jaws are used as prehensile organs during ingestion in the vast majority of terrestrial vertebrates (Schwenk,2000b). During metamorphosis, salamanders with aquatic larvae shift from suction feeding to lingual prehension (Lauder and Shaffer, 1988; Shaffer and Lauder, 1988; Reilly, 1996; Deban and Marks, 2002), a transition associated with remodeling of the hyobranchial apparatus, closure of the gill slits and development of the tongue (Wake, 1982;Wake and Deban, 2000).Some aquatic sala-manders remain specialized suction feeders as adults, but postmetamorphic terrestrial species feed in water only infrequently and inefficiently. Tongue protrusion during lingual prehension is coupled to hyobranchial protraction, which varies from modest to extreme (e.g. Larsen et al., 1989; Findeis and Bemis, 1990;Lauder and Reilly, 1994;Wake and Deban, 2000; Deban et al., 2001).In plethodontids the hyobranchial apparatus is folded into a cylinder and protruded or projected out of the mouth along with the tongue (Fig. 1.5;Lombard and Wake, 1976; Deban et al., 1997; Wake and Deban, 2000). Prey items adhere to the sticky tongue pad. Virtually all terrestrial salamanders ingest prey with lingual prehension, but some, in particular larger species, resort to jaw prehension for large prey (Wake and Deban, 2000). Most frogs are also obligate lingual feeders, although some species occasionally approach large or difficult to capture prey closely enough to use the jaws (Anderson, 1993; Nishikawa, 2000). Frogs use three different mechanisms of tongue protrusion, i.e. mechanical pulling, inertial elongation,and hydrostatic elongation, each manifesting a characteristic suite of functional traits. Although hydrostatic elongation of the tongue is commonplace in mammals and some lizards, it is exceptional in frogs and restricted tospecies in two families with unusual tongue morphology (Ritter and Nishikawa, 1995; Nishikawa et al., 1999; Nishikawa, 2000). Unlike
16
Physiological and ecological adaptations t o feeding in vertebrates
Ton~wepad
h
Fig. 1.5. Schematic representation of tongue projection during lingual prey capture (ingestion) in a plethodontid salamander. Note that the hyobranchial apparatus (tongue skeleton) is folded and projected from the mouth along with the tongue. DNIP: depressor mandibulae muscle; RCP: rectus cervicis profundus muscle; SAR: subarcualis rectus muscle. From Deban and Dicke (1999), reproduced with permission of The Company of Biologists.
salamanders and lizards, the hyobranchial apparatus in these frogs participates only indirectly in tongue protrusion. Among lepidosaurian reptiles, tuatara and iguanian lizards rely on lingual prehension (Schwenk and Throckmorton, 1989; Bels et al., 1994; Schwenk, 2000~).A few species are obligate tongue feeders but most decrease tongue protrusion distance as prey become larger, eventually shifting to jaw prehension (Gorniaket al., 1982;Schwenk and Throckrnorton, 1989; Schwenk, 2000c; Kardong and Bels, 2001). Tongue protrusion during feeding is coupled to hyobranchial protraction, as in salamanders, but movement is much more limited and the hyobranchium never leaves the mouth. The tongue curls around the margin of the lower jaw and prey adhere to its sticky, papillose surface. Chameleons have modified this basic system by inserting a ballistic, projection phase in which the tongue is launched off a supporting process of the hyobranchium out of the mouth (Schwenk and Bell, 1988; Wainwright and Bennett, 1992; Schwenk, 2000~). The tongue surface is actively dimpled during prey prehension, generating suction to provide extra adhesion for relatively large prey (Herrelet al., 2000). With very few exceptions, the remaining lizards are obligate jaw feeders. Many of these have kinetic skulls that improve the speed and precision of jaw capture by allowing simultaneous movement of upper and lower jaws in a pincer-like action (Frazzettta, 1983; Schwenk, 2000c; Metzger, 2002). Snakes have taken jaw prehension and cranial kinesis to its most extreme form (Cundall and Greene, 2000).
Vertebrate feeding systems
17
Lingual ingestion in turtles is restricted to terrestrial species (Summerset al., 1998; Wocheslander et al., 1999).Tongue protrusion is limited, with the tongue usually making contact with the food item at or near the jaw margins, almost within the mouth. Tongue retraction is typically accompanied by a bite. Semiaquatic species feeding on land use the jaws for prehension. The highly reduced tongue of most birds makes lingual ingestion impracticable for many species. Nectivores, however, are specialized lingual feeders. Hummingbirds, for example, probe flower nectaries with their very long tongues, using narrow channels in the tongue to acquire nectar by capillarity. Other nectivorous species, such as lories, have keratinous, brush-like tongue tips to increase surface area for nectar retrieval (McLelland, 1979).Woodpeckers use exceptionally long tongues to probe holes and crevices, using them to ingest larval insects (McLelland, 1979; Zweers and Berkhoudt, 2001) and crossbills pull seeds from cones using the tongue (Benkrnan, 1987). The extent to which mammals use lingual ingestion is relatively unstudied, but we suspect it is more common than supposed. Certainly specialized myrmecophages are well known to use their long, extensible tongues to probe ant and termite nests for prey (Reiss, 2000). Many nectar- and fruit-eating bats ingest liquid or soft food by lapping, often evincing brush-like tongue tips (like some nectar-feedingbirds) to maximize adherent food (Griffiths, 1982).Giraffes and related okapi use exceptionally protrusible tongues to strip leaves off trees (Owen, 1868; Kingdon, 1979)and some grazing bovids pull grass into the mouth for cropping using a prehensile tongue (KS, pers. obs.). Jaw prehension of food is common and general among tetrapods. Crocodilians lack a protrusible tongue and are obligate jaw feeders (see above). Similarly, gymnophione amphibians (caecilians) lack protrusible tongues and use the jaws to capture prey (Bemis et al., 1983; O'Reilly, 2000). The majority of birds are specialized for jaw prehension. They have modified the bill and rhamphotheca (the keratinous part) in myriad ways for this purpose (Fig. 1.6; Zweers, 1985;Zweers et al., 1994,1997;Rubega, 2000; Zweers and Berkhoudt, 2001).Scleroglossanlizards and snakes have modified the tongue for chemoreception and, with very few exceptions, rely on the jaws for ingestion (Cundall and Greene, 2000; Schwenk, 2000c; Kley, 2001). Most mammals are jaw feeders as well. Except for mammals, most jaw-feeding vertebrates have reduced, simplifiedor immobile tongues. Infrequently, structures other than the jaws and tongue are used for ingestion in tetrapods. The forelimbs are used in some mammals, such as primates and rodents, and in some frogs (Gray et al., 1997).Rarely, the hind limbs are used (as in some birds such as raptors), or other structures, such as an elephant's mobile trunk or the prehensile lips of black rhinoceroses and some other mammals (Kingdon, 1979).Whatever specific mechanism is used, ingestion results in the placement of the food item in the mouth where it is
18
Physiological and ecological adaptations t o feeding i n vertebrates
Fig. 1.6. Diversity of the jaws and beak in birds. Birds lack teeth and do little oral food processing, but the beak is specialized in various ways for food acquisition. (A) hyacinth macaw; (B) southern giant petrel; (C) parakeet auklet; (D) wrybill; (E) Andean avocet; (F) whippoorwill; (G) African spoonbill. From Rubega (2000), reproduced with permission of Elsevier Science.
Vertebrate feeding systems
19
positioned for processing or for immediate transport to the pharynx for swallowing. In some cases, a killing bite and/or head shake is interposed here (see processing,below).
Intraoral Transport Intraoral transport (or simply transport) usually refers to posterior movement of the food item through the oral cavity to the pharynx where it can be swallowed, but more generally it can be taken to mean any intraoral movement and manipulation of the food item once held in the mouth. For example, food that is chewed is often laterally repositioned so that it lies between upper and lower tooth rows. Side-switching during chewing is common in mammals and in many lizards (Hiiemae and Crompton, 1985; Hiiemae, 2000; Schwenk, 2000c; Reilly et al., 2001; McBrayer and Reilly, 2002a).A food item is sometimes transported anteriorly if, after ingestion, it comes to lie too far back in the mouth for processing (Schwenk and Wake, 1993; Schwenk, 2000a, c). In mammalian studies, two discrete types of transport are often distinguished (Hiiemae et al., 1978; Hiiemae and Crompton, 1985;Hiiemae, 2000).In stage I transport, food is moved from the incisive area at the front of the mouth to the postcanine region for processing. In stage 11transport, liquids and reduced food are moved posteriorly through the fauces (the posterior border of the oral cavity demarcated by the vertical columns of the palatoglossal muscles), either for bolus formation or for immediate swallowing. In aquatic vertebrates, transport is usually accomplished hydrodynamically (hydraulically)by creating pressure gradients within the oropharyngeal (and opercular) cavities (e.g. Lauder, 1985a; Bemis, 1987; Liem, 1990; Gillis and Lauder, 1994,1995; Lauder and Gillis, 1997).In some teleost fishes, tooth-bearing pharyngeal jaws are used to manipulate prey, moving it toward the esophagus (e.g. Liem and Greenwood, 1981; Sibbing, 1982; Lauder, 1983, 1985a; Sibbing et al., 1986; Vandewalle et al., 1994). Many cetaceans and pinnipeds also use hydrodynamic transport of captured food (Werth, 2000a, b). In terrestrial vertebrates, most transport and manipulation of food is hyolingual (Bramble and Wake, 1985; Hiiemae and Crompton, 1985; Schwenk, 2000a),meaning that it is mediated by coordinated, cyclical movements of the tongue and the hyobranchial skeleton (Fig. 1.1).During transport the food item sits on the tongue while cyclical motions of the tongue and hyobranchium move it toward the pharynx for swallowing,or reposition it in the mouth for processing (see next section). In taxa with muscular or fleshy tongues (mammals, many lizards, some turtles, parrots, possibly waterfowl), the tongue forms itself around the food item during transport, cupping it, or humps up in front of it, pushing it. In one bird lacking a fleshy tongue (and probably others), the tongue is bent sharply downward at an intrinsic hyobranchial joint and the food item is transported on the tongue
20
Physiological and ecological adaptations t o feeding in vertebrates
behind the peak (Rubega et al., submitted manuscript ). Among various tetrapods, palatal teeth, palatal rugae, or other keratinous projections on the palate prevent the bolus from moving forward while the hyolingual apparatus protracts beneath it in preparation for the next transport cycle (Bramble and Wake, 1985;Hiiemae and Crompton, 1985; Zweers, 1985;Hiiemae, 2000). In frogs, hyolingual transport may be unnecessary due to the brevity of the pharynx in this virtually neckless group. Except in a few species that use hydrostatic tongue elongation, the frog tongue is attached at the front of the mandible so that when it flips back into the mouth during ingestion, the adherent prey item is placed at the rear of the throat in position for immediate swallowing. Many tetrapods sometimes replace hyolingual transport with inertial transport, in which the food item is released or tossed by the jaws (and the head repositioned over it) so that it comes to lie farther back in the mouth (Gans, 1969).Inertial transport is especially typical of many reptiles, including some lizards, crocodilians, and birds (Gans, 1969; Smith, 1986; Zweers et al., 1994; Cleuren and de Vree, 2000; Schwenk, 2000c; Tomlinson, 2000), but it is also exhibited by ancestral mammals such as opossums, tenrecs, and tree shrews (Tupaia),as well as carnivoran species that bolt chunks of flesh (Hiiemae and Crompton, 1985;Van Valkenburgh, 1996). In snakes, the hyolingual apparatus is so specialized for chemosensory function that its role in feeding is entirely lost (Schwenk, 2000~).Intraoral transport is accomplished with movements of the highly kinetic skull, including unilateral movements of toothed jaw and palatal bones that alternately "grasp" a prey item on one side and then the other. Although small prey items are potentially pulled through the mouth and into the pharynx, most snakes feed on relatively large prey. In these species the kinetic skull mechanism is more accurately said to pull the snake's head and body over a stationary food item stabilized by its own mass. Thus, the snake transport mechanism is considered to be a type of inertial feeding (Cundall and Greene, 2000).Scolecophidiansnakes employ unique mechanisms of upper and lower jaw kinesis to "rake" prey into the mouth and push it into the pharynx (Kley 2001). Axial bending of the anterior trunk supplements cranial transport once the bolus is far enough back (Moon, 2000; Kley and Brainerd, 2002). Surface tension transport is a specialized mechanism of intraoral transport characteristic of shorebirds. Tiny prey items are suspended within a drop of water between the jaws while surface tension drives the drop along the bill as upper and lower jaws are spread apart (Rubega and Obst, 1993; Rubega, 1996,1997). In all cases, the outcome of transport is placement of food in the pharynx, ready to be swallowed. In many taxa, however, the food is mechanically reduced or otherwise processed before transport is completed.
Vertebrate feeding systems
Processing Processing refers to any mechanical reduction or preparation of the food before it is swallowed. However, many taxa, including most fish, most amphibians, many birds and snakes, do not process their food at all (other than lubrication with saliva) -they simply swallow it whole directly after ingestion and transport. And although most processing occurs within the oral cavity, some vertebrates do a significant amount of food preparation before or during the act of ingestion. Carnivoran mammals, for example, often rend pieces from their prey with the jaws, sometimes aided by the forelimbs (Ewer, 1973). Many terrestrial turtles do something similar by pinning a food item against the substrate with the forelimbs while tearing pieces off with the beak (KS, pers. obs.), as do many raptorial birds. Mammals, especially rodents and primates, hold a food item in the forelimbs and bite off small pieces for further processing within the mouth. Galapagos land iguanas often use their forelimbs to scrape the spines off the prickly pear cactus fruit they favor (H. K. Snell, pers. comm.). Crocodilians rend chunks from large prey by grasping the prey in the jaws and spinning violently on their axes (Cott, 1961; Pooley and Gans, 1976; Taylor, 1987).They also sometimes cache dispatched prey underwater to store it for later consumption and possibly to soften it before ingestion.Similarly,shrikes (passerine birds) impale prey on thorns or barbed wire, returning to feed on it later. Many sharks and other predatory fish tear or bite pieces from larger prey (Frazzetta and Prange, 1987; Motta and Wilga, 2001), as do some bony fish. Although processing of any kind is otherwise unknown in snakes, two species tear apart freshly molted crabs by pulling them through a loop of the body (Jayne et al., 2002). Granivorousbirds sometimes hold hard seeds with the feet while cracking them with the beak, or husk them directly in the bill, ingesting only the inner kernel (Ziswiler and Farner, 1972; Zweers et al., 1994; Nuijens and Zweers, 1997). Parrots employ an elaborate shelling behavior involving the beak and tongue (Homberger, 1980,1986).Kingfishers beat a captured fish against a perch, both to subdue and to soften it by breaking bones, then swallow it whole. Uniquely, many birds process food after it is swallowed, in specialized partitions of the esophagus and stomach (seelater). In tetrapods, most processing occurs within the mouth by crushing or biting with the teeth. Sometimes this is restricted to cropping during ingestion, or killing bites and head shakes immediately upon capture, but usually prey is further chewed with the teeth. Chewing involves repeated, cyclical biting movements that crush, puncture, shear and/or grind the food item, mechanically reducing it in preparation for swallowing. In nonmammalian taxa that chew, this behavior is referred to descriptively as puncture-crushing, as typified by lizards (Schwenk, 2000c; McBrayer and Reilly, 2002). Food is pierced by sharp, pointed teeth or crushed between blunt, molariform teeth, but there is little, if any, fragmentation of the bolus
22
Physiological and ecological adaptations t o feeding in vertebrates
and certainly no true comminution (see below).Puncture-crushingserves to soften the food item, to lubricate it with copious saliva, and potentially to introduce salivary enzymes into the bolus, initiating chemical digestion. Durophagous species feeding on snails and large arthropods may use temporal summation of pulsatile adductor contractions to increase bite force (Gans and De Vree, 1986). Chewing in nonmammalian taxa is often erroneously referred to as "mastication", but this term is accurately applied only to mammals (Davis, 1961; Schwenk, 2000a). Mastication is a derived and specialized form of chewing associated with a suite of mammalian novelties, including functional specialization of teeth along the tooth row (heterodonty),precise, unilateral occlusion of upper and lower teeth, a masticatory cycle including lateral and/or anteroposterior movements of the lower jaw, a derived tongue morphology, and the evolution of a muscular pharynx associated with a unique form of swallowing (Hiiemae and Crompton, 1985;Crompton, 1989, 1995;Smith, 1992; Herring, 1993; Weijs, 1994; Thexton and Crompton, 1998; Hiiemae, 2000; Schwenk, 2000a, 2001a). The important feature of mastication, in contrast to puncture-crushing, is that it reduces ingested food to a fine slurry of tiny particles mixed with saliva, a process referred to as comminution (Fig. 1.2). Food in this semiliquid state is moved during stage 2 transport into the pharynx where it is temporarily held or swallowed immediately. Uniquely, mammals often interpose swallow cycles amidst a series of masticatory cycles (Thexton and Crompton, 1998;Hiiemae, 2000), whereas other tetrapods only swallow a bolus once chewing is completed. Some mammals have secondarily reduced or lost their ability to masticate. This is usually correlated with modification, reduction or loss of the teeth associated with specialized diets such as insectivory and piscivory (e.g. odontocete cetaceans) or secondary reversion to suspension feeding (e.g. mysticete cetaceans).Modern monotremes have lost their teeth altogether, substituting keratinized structures on the tongue and palate to rasp their food (Owen, 1868; Doran and Baggett, 1972; Griffiths, 1978). Turtles and modern birds entirely lack teeth and rely on the keratinous rhamphotheca for whatever oral processing they do (Fig. 1.6).In turtles the apical edges of the beak are sharp and the lower jaw fits snugly within the upper forming an effective shearing mechanism. Some turtles also crush food between upper and lower plates (trituratingsurfaces) at the beak's front end (Gaffney, 1979).Birds, in general, do very little oral food processing, for reasons discussed in a subsequent section; however some birds shear food with the sharp edges of the beak. Owls and raptors, for example, often hold prey with the feet and use the beak to tear it into bits and some frugivorous parrots similarly shear off pieces of fruit (Zweersand Berkhoudt, 2001). Most sharks do no processing with the marginal teeth other than killing bites and/or excision of pieces from larger prey (Frazzetta and Prange, 1987; Frazzetta, 1994;Motta and Wilga, 2001).Some elasmobranchs,however, crush
Vertebrate feeding systems
23
hard-bodied prey, such as mollusks, with plates of flattened teeth (Moss, 1977; Summers, 2000; Wilga and Motta, 2000). In these taxa the cartilaginous jaws may be strengthened by unusual "trabecular cartilage" (Summers, 2000).Relatively few bony fish use the jaws and marginal teeth for processing, probably because without cheeks and lips, reduced food particles would be lost in the water (Vandewalleet al., 1994).However, some taxa do manage to reduce food in the marginaljaws (e.g.Hernandez and Motta, 1997),but in most bony fish that process their food, it is crushed, ground or pierced by intraoral and/or intrapharyngeal teeth on the palate and hyobranchial skeleton. Indeed, in derived taxa, such as the Cichlidae and Labridae, teeth are restricted solely to tooth plates on the pharyngeal jaws which are used to process the food before it is swallowed (e.g.Liem, 1973; Sibbing, 1982,1991; Liem and Sanderson, 1986;Vandewalle et al., 1994; Galis and Drucker, 1996; Grubich, 2000). Some aquatic taxa pump water in and out of the mouth, lacerating prey as it is raked across the marginal teeth (Bemis, 1987;Elwood and Cundall, 1994). Although salamanders virtually never process their food, one group (thedesmognathine plethodontids) routinely delivers crushing bites using "head tucking"behavior in which force is transmitted to the lower jaw via a ligamentous connection to the cervical vertebrae (Schwenk and Wake, 1993).
Swallowing During swallowing (also called pharyngeal emptying; Smith, 1992;Schwenk, 2000a), the bolus is moved from the pharynx into the esophagus where peristalsis takes over the task of its transport through the remainder of the gut. Depending on the particular mechanism of swallowing used and/or the relative length of the prey item compared to the pharynx, the transition between intraoral transport and swallowing is often blurred. An extreme example is evident in macrostomatan snakes, which typically eat relatively large or elongate prey (Cundall and Greene, 2000). One end of a prey item often extends from the mouth even as the other end enters the esophagus! Initially, unilateral, alternating movements of the kinetic skull and jaws are used to move the prey toward the esophagus. These are supplemented and then replaced by axial bending movements of the trunk as the food item moves farther into the esophagus (Moon, 2000; Kley and Brainderd, 2002), so that one feeding stage blends imperceptibly into the next. A similar situation occurs in some seabird chicks feeding on relatively long fish. In most taxa there is a somewhat gradual transition between transport and swallowing cycles rather than a sharp demarcation. In mammals, as noted, swallowing cycles are often interposed among a series of masticatory cycles so there may not be a terminal swallowing stage, per se. Swallowingin fishes is poorly understood, but in many teleosts, at least, it is accomplished with manipulatory movements of the pharyngeal jaws that push the bolus into the esophagus (Lauder, 1983a, b, 1985a).How it
24
Physiological and ecological adaptations t o feeding in vertebrates
occurs in taxa lacking pharyngeal jaws is not clear (see Lauder, 1983a).Some suspension feeding fish trap small food particles in mucous strands that move into the esophagus with water flow (Sandersonet al., 1996).Recently, Sanderson et al. (2001)showed that suspension-feedingteleosts do not trap particles directly in the gill rakers as previously thought ("sieving"), but rather capture them using "crossflow filtration" in which food particles are concentrated in the oral cavity and then swept across the rakers toward the throat, thus solving "the mystery of particle transport to the oesophagus" (Sanderson et al., 2001). Teleosts have an esophageal sphincter that putatively prevents them from swallowing too much water (Stevens and Hume, 1995). Most tetrapods use cyclical movements of the tongue and hyobranchial skeleton, as well as compression of the pharynx with the superficial constrictor musculature, to push or squeeze food into the esophagus. In lizards and terrestrial turtles, for example, the posterior end of the tongue is used to tamp food into the throat ("pharyngeal packing") and this is followed or in some cases, replaced with pharyngeal compression (Fig. 1.7; Smith, 1984, 1986; Bels et al., 1994; Schwenk, 2000~).In amphibians, swallowing is accomplished primarily with pharyngeal compression, often accompanied by retraction of the eyeballs, which may help to force the bolus into the esophagus (Duellman and Trueb, 1986; Deban and Wake, 2000). In taxa with reduced tongues, such as crocodilians and birds, pharyngeal compression is supplemented with inertial movements of the head, as well as gravity (Cleuren and De Vree, 2000; Tomlinson, 2000; Zweers and Berkhoudt, 2001). Some birds also use sinuous contractions of the floor of the pharynx ("properistalsis") to move food into the esophagus (Zweers, 1985; Zweers and Berkhoudt, 2001). Mammals employ a uniquely derived form of swallowing called deglutition (Thexton and Crompton, 1998; Hiiemae, 2000). Like the term "mastication", deglutition is often incorrectly applied to nonrnammalian taxa. Mammalian deglutition is associated with a derived tongue morphology and especially the presence of a soft palate and pharyngeal musculature (Smith, 1992;Hiiemae, 2000).Mammals can form a sphincter-like seal at the base of the tongue (at the fauces), functionally subdividing the buccal and pharyngeal space (Hiiemae,2000). Masticated food is passed through this seal during stage 2 transport to accumulate in the oropharynx.When a bolus is sufficientlylarge, an explosive contraction of the tongue base and soft palate, in conjunction with peristaltic waves of contraction in the pharyngeal musculature, propel food into the esophagus. CONSEQUENCES OF FEEDING-TAXONOMIC HIGHLIGHTS Vertebrate feeding systems were surveyed above according to the mechanical task associated with each feeding stage. The review revealed a great deal
Vertebrate feeding systems
Fig. 1.7. Swallowing in a terrestrial turtle based on individual frames from cineradiographic film. Numbers indicate frame number. Note shape changes in the tongue as it moves in front of the bolus and forms a seal with the palate. It then pushes the bolus posteriorly, into the esophagus, where peristalsis takes over. Tongue movement is accompanied by hyobranchial movement. ASHG: anterior suprahyoid muscles. From Bramble and Wake (1985), reprinted by permission of the publisher and President and Fellows of Harvard College.
of diversity in the manner that vertebrates procure, process, transport, and swallow food. Despite this diversity, all systems share the common outcome of moving the bolus into the esophagus for further digestion and nutrient assimilation. Therefore, the mechanical and chemical condition of the bolus at the point of swallowing, as determined by the front-end feeding mechanism, has obvious significance for physiological functioning in the remainder of the gut and it is this inescapable connection we turn to here.
Fishes Obviously, fishes are phylogenetically and taxonomically disparate, but within any clade the nature of the food swallowed devolves to one of five
26
Physiological and ecological adaptations t o feeding in vertebrates
types: particulate food collected by suspension feeding; tiny food items taken individually; relatively larger, whole, unprocessed food items; minimally processed food consisting of large pieces or chunks excised by the marginal teeth; and food that is processed, sometimes sigruficantly,by the marginal or pharyngeal jaws. Whether they trap suspended food from the water column or substrate, or ingest it one bit at a time, microphagous fishes swallow food that does not require further reduction and is ready for passage into the gut and assimilation. Whole food particles, however small, may nonetheless pose a different digestive challenge compared to food that has been milled, and therefore ruptured, by the pharyngealjaw apparatus. Pharyngognathy opens the door to significant food processing potentially rivaling that of mammals (e.g. Sibbing, 1991),but the nature of the processed food is rarely studied. Stomach content analyses of labrids by Wainwright (1987, 1988) showed that snail and crab shells were crushed and fragmented by the pharyngeal.jaws, with most of the shell fragmentswinnowed out before swallowing.Parrotfish process algae (and the coral with which it is associated) into a fine paste (Bellwoodand Choat, 1990;Bellwood, 1996,Choat et al., 2002) and presumably this is why they do not require a stomach (Horn, 1989).There seems to be a strong correlation between the degree of pharyngeal processing and gut form, at least in herbivorous species (Horn, 1989, 1992). Some seedeating species may use pulsatile contractions of the pharyngeal jaw musculature to increase crushing forces by means of temporal summation (Irish, 1983).Pharyngognathy may be particularly important to herbivorous and durophagous species, as might be expected, but this is an untested assumption. Contrastingly, in fishes that swallow entire prey or large pieces thereof, the gut must complete digestion chemically. Such species possess an extensible stomach for storage and face the potentially daunting digestive problem of the food item's high volume-to-surface area ratio. These issues are less problematic for carnivorous species because flesh is more easily digested. Herbivorous fishes are either inefficient in digestion, highly selective in diet, restricted to microphagy, or employ significant pharyngeal processing.
Amphibians Amphibians are nearly universal in their lack of prey processing and thus, with few exceptions, swallow whole, unreduced prey items. Notable exceptions are aquatic salamanders that shred prey items against the teeth as they are sucked in and out the mouth (Elwood and Cundall, 1994),desmognathine salamanders that routinely crush prey between the jaws (Schwenk and Wake, 1993),and caecilians that shear off pieces of prey with specialized dentition and/or axial rotation of the body while withdrawing into a burrow (Bemiset al., 1983; O'Reilly, 2000). Virtually no adult amphibians are herbivorous,
Vertebrate feeding systems
27
although two frog species are reported to consume more than incidental quantities of fruit or leaves (da Silva et al., 1989; Das, 1995). Despite the lack of herbivory, amphibian diets are diverse, particularly in prey size, which ranges from mites to vertebrates.Disparity in prey size and type, coupled to constraints on gut length imposed by the anuran body plan, make frogs an ideal group in which to examine diet-gut relationships. The only known folivorous frog, for example, was found to have an unusually long gut that may contain fermentative microorganisms (Das, 1995).A number of frog groups share a suite of characters related to microphagy, including cutaneous sequestering of dietary alkaloids (Vences et al., 1997/98). Microphagy increases the surface-area-to-volume ratio of the food particles swallowed (circumventing the problem of no chewing), but the arthropod prey eaten (mostlyants) tend to be noxious and/or low in nutritive value. A termite specialist, for example, was found to have a relatively longer gut than other species, putatively because termites are relatively hard to digest (Das, 1995).
Nonavian Reptiles Most reptiles do little or no processing and swallow food largely intact, either whole or in large pieces. Crocodilians exemplify this mode of feeding: small prey are simply tossed into the back of the mouth and swallowed (sometimes after a crushing or killing bite). Larger prey are dismembered or tom apart and the pieces bolted. Both feeding modes are enhanced by an exceptionally powerful bite (Ericson et al., 2003). The stomach is divided into a cranial glandular part and a caudal muscular part (the pylorus) that appears as a separate chamber when the stomach is empty (Richardson et al., 2002). The pylorus is often referred to as a "gizzard". Its homology to the bird gizzard is assumed and not certain. In any case, crocodilians, like birds, often swallow grit or small stones called gastroliths that are held within the gizzard and presumed to aid in the mechanical reduction of food. However, the evidence for this in crocodiliansis weak, gastroliths may also function as ballast (Taylor, 1993).Mechanical processing in the muscular gizzard, possibly aided by gastroliths, may help to mitigate the problem of digesting large, mostly intact pieces of food. Crocodiliansare also reputed to have exceptionally acidic gastric secretions that promote rapid digestion (Richardsonet al., 2002). Turtles do little or no intraoral processing, but do crop bite-sized pieces of food with their sharp beaks. Small, cropping bites may be particularly important in herbivorous species (Bjomdal and Bolten, 1992). Carnivorous and durophagous species generate more powerful bites than species with other diets (Herrel et al., 2002) but even extreme dietary specialization is possible without attendant specializationof the feeding system and gut (e.g. Meylan, 1988). Suction feeding aquatic species potentially swallow relatively large, whole prey.
28
Physiological and ecological adaptations t o feeding in vertebrates
Snakes, except for the two crab-eatingspecies mentioned above, virtually always swallow whole prey with no intraoral processing. However, the venom that crotalid snakes (rattlesnakes and their relatives) inject into their prey contains proteolytic digestive enzymes that initiate digestion from inside-out (Thomas and Pough, 1979; Cundall and Greene, 2000). Relative prey size is exceptionally large in this group, sometimes exceedingthe snake's own mass, and the prey's relatively small surface area could challenge the gut's ability to digest it extrinsically before it putrefies, This problem is exacerbated by the fact that the gut takes time to be upregulated after a period of quiescence (e.g. Secor and Diamond, 1998; Starck and Beese, 2001). Some snakes are able to crush bird eggs within the gut, retaining the liquid contents while regurgitating the indigestible shell (Gans, 1952). Although many lizards swallow small prey whole, lepidosaurs are exceptional among reptiles in the extent of their intraoral processing. Tuatara and many lizards initially crop larger food items into mouth-sized pieces. This is particularly true for herbivores whose teeth are often specialized for this purpose (Schwenk, 2000~).Sometimes inertial shaking is used to dismember prey. Most food is then chewed between marginal tooth rows, sometimes aided by palatal teeth. Chewing typically crushes, pierces, and softens the food item, but rarely results in significant trituration, even in herbivores. However, it is very likely that chewing introduces salivary enzymes and presumably the soft, well-lubricated bolus is more easily swallowed and digested.
Birds Although many birds reduce prey to some extent before ingestion, loss of teeth precludes significant intraoral processing. Consequently, birds often swallow whole, large, hard, or refractory food particles such as seeds without the benefit of front-end processing. Instead, they have adopted a novel strategy for processing that relies on specialization of the anterior gut. The esophagus is extremely extensible in species that swallow large prey and in many taxa it includes an expanded region where food can be temporarily stored. When evident as a distinct diverticulum, this structure is called a crop (Ziswiler and Farner, 1972;McLelland, 1979).The principal function of the crop is to store excess food and to regulate its rate of delivery to the stomach. In pigeons the crop produces and stores a liquid slurry of shed cells that is regurgitated to feed chicks. In many other species, the crop stores food and water for later regurgitation feeding of chicks. Although no significant chemical digestion occurs here, grain and other hard foods are moistened and softened. The single exception is the folivorous hoatzin in which the muscular crop serves as the site of fermentative digestion (Grajal, 1995). The stomach comprises two principal chambers, a cranial proventriculus and a caudal gizzard or ventriculus (Fig. 1.8, Ziswiler and Farner, 1972; McLelland, 1979). The proventriculus is glandular and the site of most
29
Vertebrate feeding systems
chemical secretion and digestion while the gizzard is extremely muscular in most species and responsible for mechanical processing. The gizzard is especially well developed in taxa whose food requires mechanical reduction, i.e. omnivores, insectivores,herbivores, and granivores (in some carnivores it is reduced or virtually absent). In these species, the gizzard is lined with a thick, hard cuticle forming dorsal and ventral "grinding plates" covered with ridges. Food is ground or crushed between the plates, with reduction facilitated by grit intentionally swallowed. The amount of grit in the gizzard has been correlated with diet and found to be more prevalent in granivorous birds, while grit size correlates with body size (Gionfriddo and Best, 1996).The crushing action of the gizzard in some species is considerable, i.e. whole, hard-shelled nuts are fragmented by turkeys within hours and there are even reports of metal objects being folded and ground into fragments (Welty, 1975).The gizzard apparatus is often referred to as a "gastric mill" and has been compared to the mammalian masticatory apparatus (e.g. King and King, 1979).
Papilla proventricularis
Isthmus gastris -
r, Lateral muscle
G* N Y
b,
Saccus cauda Fig. 1.8. The stomach of a granivorous bird (rock dove), showing its division into a cranial proventriculus and a caudal gizzard. The gizzard is highly muscular and lined with a thick, abrasive cuticle which, in conjunction with swallowed grit, mechanically processes food, reducing it to small particles. From N. S. Proctor and P. J. Lynch (1993), reproduced with permission of Yale University Press.
30
Physiological and ecological adaptations t o feeding in vertebrates
Mammals Evolution of mastication is a key theme in the history of mammals. It is at the heart of a sweeping reorganization of the skull, palate, dentition, tongue, pharynx, jaw muscles, and ear, and is intimately associated with the evolution of endothermy. It uniquely distinguishesmammals from all other vertebrates. The masticatory apparatus evolved incrementally within the synapsid stem lineage and was present essentially in the modern form once a fully functional dentary-squamosaljaw joint replaced the increasingly weak, ancestral quadrate-articular joint (Crompton, 1989,1995). The critical feature of the masticatory system is its ability to reduce food quickly to a mash of tiny particles mixed with copious enzyme-containing saliva (Fig. 1.2). Compared to other vertebrates, the bolus is virtually "predigested when swallowed. This has obvious relevance for passage rates in the gut. Although there are many secondary departures from the fundamental pattern of mastication, these usually occur in taxa that consume flesh or other types of food relatively more easily digested.Conversely the masticatory mills of herbivores and taxa that consume other types of refractory foods are elaborate. General Patterns One noteworthy pattern that emerges from the preceding consideration of the bolus condition is that in ectothermic vertebrates, food is processed relatively little, if at all, whereas in birds and mammals, food entering the intestine is usually extensively triturated. This distinction probably relates to the greater energetic demands of endotherms and their need to increase gut passage rates. Increased passage rates, as well as greatly increased gut surface area, are necessary in endotherms because they are no more efficient in extracting energy from their food than are ectotherms(Karasov, 1987;Karasov and Diamond, 1985,1988).By essentially "predigesting" their food, birds and mammals decrease the time necessary to hold it in the gut for chemical breakdown. In contrast, ectotherms can afford the low gut passage rates required for adequate digestion by virtue of their modest energetic demands. Mammals employ a complex oral masticatory system to process their food, whereas birds rely on a "gastric mill". In birds, the anterior gut effectively functions as part of the front-end feeding system. The evolutionary transfer of food processing from the oral apparatus to the gizzard, along with the loss of teeth, shifted the mass of the prey-reduction apparatus toward the center of gravity, an obvious advantage for a volant animal. These general patterns potentially have had important consequences for patterns of evolution in front-end phenotypes, a topic we explore elsewhere (Schwenk and Rubega, in litt.).
Vertebrate feeding systems
Acknowledgments
We thank the editors of this volume, Matthias Starck and Tobias Wang, for their invitation to contribute and Adam Summers for his critical review of the manuscript. We are grateful to Chris Elphick, Harry Greene, JulieHagelin, George Lauder, Peter Wainwright, and Mark Westneat for answering our questions and assisting with the literature. KS and MR were supported by grants from the University of Connecticut Research Foundation. REFERENCES Aerts P., van Damme J. and Herrel A. 2001. Intrinsic mechanics and control of fast cranio-cervical movements in aquatic feeding turtles. Amer. Zool. 41: 1299-1310. Alfaro M. E. and Westneat M. W. 1999. Motor patterns of herbivorous feeding: electromyographic analysis of biting in the parrotfishes Cetoscarus bicolor and Scarus iseri. Brain Behav. Evol. 54: 205-222. Alfaro M. E., Janovetz, J. and Westneat M W. 2001. Motor control across trophic strategies: muscle activity of biting and suction feeding fishes. Amer. Zool. 41: 1266-1279. Allin E. G. 1975. Evolution of the mammalian ear. 1. Morph. 147: 403438. Anderson C. W. 1993. The modulation of feeding behavior in response to prey type in the frog Rana pipiens. 1. Exp. Biol. 179: 1-12. Arnold E. N. 1998. Cranial kinesis in lizards: variations, uses, and origins. Evol. Biol. 30: 323-357. Atchley W. R. 1993. Genetic and developmental aspects of variability in the mammalian mandible. In: The Skull, vol. 1. Development. J. Hanken and B. K. Hall, (eds.). Univ. Chicago Press, Chicago, IL (USA), pp. 207-247. Beecher R. M. 1979. Functional significance of the mandibular symphysis. 1. Morph. 159: 117-130. Beecher W. J. 1962. The bio-mechanics of the bird skull. Bull. Chicago Acad. Sci. 11: 10-33. Belkin D. A. a n d G a n s C. 1968. An unusual chelonian feeding niche. Ecology 49:768-769. Bellwood D. R. 1996. Production and reworking of sediment by parrotfishes (family Scaridae) on the Great Barrier Reef, Australia. Mar. Biol. 125: 795-800. Bellwood D. R. and Choat J. H. 1990. A functional analysis of grazing in parrotfishes (family Scaridae): the ecological implications. Environ. Biol. Fishes 28: 189-216. Bels V. L., Chardon M. and Kardong K. V. 1994. Biomechanics of the hyolingual system in Squamata. In: Biomechanics of Feeding in Vertebrates (Advances i n Comparative b Environmental Physiology 18). V. L. Bels, M. Chardon, and P. Vandewalle, (eds.). Springer-Verlag, Berlin, pp. 198-240. Bemis W. E. 1987. Feeding systems of living Dipnoi: anatomy and function. 1. Morph. (suppl.) 1: 249-275. Bemis W. E. and Lauder G. V. 1986. Morphology and function of the feeding apparatus of the lungfish, Lepidosiren paradoxa (Dipnoi). 1. Morph. 187: 81-108. Bemis W. E., Schwenk K. and Wake M. H. 1983: Morphology and function of the feeding apparatus in Dermophis mexicanus (Amphibia: Gymnophiona). Zool. 1. Linn. SOC.77: 75-96. Benkman C. W. 1987. Crossbill foraging behavior, bill structure, and patterns of food profitability. Wilson Bull. 99: 351-368. Bjorndal K. A. and Bolten A. B. 1992. Body size and digestive efficiency in a herbivorous freshwater turtle: advantages of small bite size. Physiol. Zool. 65: 1028-1039. Bock W. J. 1964. Kinetics of the avian skull. 1. Morph. 114: 1-41. Bout R. G. and Zweers G. A. 2001. The role of cranial kinesis in birds. Comp. Biochem. Physiol. A 131: 197-205.
32
Physiological and ecological adaptations t o feeding in vertebrates
Bramble D. M. and Wake D. B. 1985. Feeding mechanisms of lower tetrapods. In: Functional Vertebrate Morphology. M. Hildebrand, D. M. Bramble, K. F. Liem, and D. Wake (eds.). Harvard Univ. Press, Cambridge, MA, pp. 230-261. Busbey 111, A. B. 1989. Form and function of the feeding apparatus of Alligator mississippiensis. J. Morph. 202: 99-127. Butler P. M. and Joysey K. A. 1978. Development, Function and Evolution of Teeth. Acad. Press, New York, NY. Choat J. HI Clements K. D. and Robbins, W. D. 2002. The trophic status of herbivorous fishes on coral reefs, I: Dietary analyses. Mar. Biol. 140: 613-623. Cleuren J. and De Vree F. 2000. Feeding in crocodilians. In: Feeding. Form, Function and Evolution in Tetrapod Vertebrates. K. Schwenk (ed.). Acad. Press. San Diego, CA (USA), pp. 337-358. Cott H. B. 1961. Scientific results of a n inquiry into the ecology and economic status of the Nile crocodile (Crocodylus niloticus) in Uganda and northern Rhodesia. Trans. Zool. Soc. Lond. 29: 211-356. Crompton A. W. 1989. The evolution of mammalian mastication. In: Complex Organismal Functions: integration and Evolution in Vertebrates. D. B. Wake and G. Roth (eds.). John Wiley & Sons, Chichester, UK, pp. 23-40. Crompton A. W. 1995. Masticatory function in nonmammalian cynodonts and early mammals. In: Functional Morphology in Vertebrate Paleontology. J. Thomason (ed.). Cambridge Univ. Press, Cambridge, MA, pp. 55-75. Cundall D. and Greene H. W. 2000. Feeding in snakes. In: Feeding. Form, Function and Evolution in Tetrapod Vertebrates. K. Schwenk (ed.). Acad. Press, San Diego, CA, USA, pp. 293-333. da Silva H. R., de Britto-Pereira M C. and Caramaschi U. 1989. Frugivory and seed dispersal by Hyla truncata, a Neotropical treefrog. Copeia 1989: 781-783. Das I. 1995. Comparative morphology of the gastrointestinal tract in relation to diet in frogs from a locality in south India. Amphibia-Reptilia 16: 289-293. Davenport J., Grove D. J., Cannon J., Ellis T. R. and Stables R. 1990. Food capture, appetite, digestion rate and efficiency in hatchling and juvenile Crocodylus porosus. J. Zool., Lond. 220: 569-592. Davis D. D. 1961. Origin of the mammalian feeding mechanism. Amer. Zool. 1: 229-234. De Visser J. and Barel C. D. N. 1998. The expansion apparatus in fish heads, a 3-D kinetic deduction. Neth. 1. Zool. 48: 361-395. Deban S. M. and Dicke U. 1999. Motor control of tongue movement during prey capture in plethodontid salamanders. J. Exp. Biol. 202: 3699-3714. Deban S. M. and Wake D. B. 2000. Aquatic feeding in salamanders. In: Feeding. Form, Function and Evolution in Tetrapod Vertebrates. K. Schwenk (ed.). Acad. Press, San Diego, CA (USA), pp. 65-94. Deban S . M. and Marks S . B. 2002. Metamorphosis and evolution of feeding behaviour in salamanders of the family Plethodontidae. Zool. J. Linn. Soc. 134: 375400. Deban S. M., Wake D. B. and Roth, G. 1997. Salamander with a ballistic tongue. Nature 389: 27-28. Deban S. M., OrReilly J. C. and Nishikawa K. C. 2001. The evolution of the motor control of feeding in amphibians. Amer. Zool. 41: 1280-1298. Delheusy V. and Bels V. 1992 Kinematics of feeding behaviour in Oplurus cuvieri (Reptilia: Iguanidae). !. Exp. Biol. 170: 155-186. Dellow P. G. and Lund J. P. 1971. Evidence for central timing of rhythmical mastication. 1. Physiol. 215: 1-13. Doran G. A. and Baggett H. 1972. The specialized lingual papillae of Tachyglossus aculeatus. I. Gross and light microscopic features. Anat. Rec. 172: 157-166. Duellman W. E. and Trueb L. 1986. Biology of Amphibians. McGraw Hill, New York, NY. Edmonds M A., Motta P. J. and Hueter R. E. 2001. Food capture kinematics of the suction feeding horn shark, Heterodontus francisci. Environ. Biol. Fishes 62: 415427.
Vertebrate feeding systems
33
Elwood J. R. L. and Cundall, D. 1994. Morphology and behavior of the feeding apparatus in Cryptobranchus alleganiensis (Amphibia: Caudata). J. Morph. 220: 47-70 Erdman S. and Cundall D. 1984. The feeding apparatus of the salamander Amphiunza tridactylum: morphology and behavior. J. Morph. 181: 175-204. Erickson G. M., Lappin A. K. and Vliet K. A. 2003. The ontogeny of bite-force performance in American alligator (Alligator mississippiensis). J. Zool. Lond. 260: 317-327. Ewer R. F. 1973. The Carnivores. Comstock Publ., Ithaca, New York. Ferry-Graham L. A. and Lauder G. V. 2001. Aquatic prey capture in ray-finned fishes: a century of progress and new directions. J. Morph. 248: 99-119. Findeis E. K. and Bemis W. E. 1990. Functional morphology of tongue projection in Taricha torosa (Urodela: Salamandridae). Zool. J. Linn. Soc. 99: 129-157. Frazzetta T. H. 1962. A functional consideration of cranial kinesis in lizards. J. Morph. 111: 287-320. Frazzetta T. H. 1983. Adaptation and function of cranial kinesis in reptiles: a time-motion analysis of feeding in alligator lizards. In: Advances in Herpetology and Evolutionary Biology. Essays in Honor of Ernest E. Williams. A. G. J . Rhodin and K. Miyata (eds.). Museum of Comparative Zoology, Cambridge, MA, pp. 222-244. Frazzetta T. H. 1988. The mechanics of cutting and the form of shark teeth (Chrondrichthyes, Elasmobranchii). Zoonzorphology 108: 93-107. Frazzetta T. H. 1994. Feeding mechanisms i n sharks a n d other elasmobranchs. In Biomechanics of Feeding in Vertebrates (Advances in Comparative 6 Environmental Physiology 18). V. L. Bels, M. Chardon, and P. Vandewalle (eds.). Springer-Verlag, Berlin, pp. 31-57. Frazzetta T. H. a n d Prange C. D. 1987. Movements of cephalic components during feeding in some requiem sharks (Carcharhiniformes: Carcharhinidae). Copeia 1987: 979-993. Fiirbringer M. 1922. Das Zungenbein der Wirbeltiere insbesondere der Reptilien und Vogel. Abh. der Heidelberger Akad., math.-naturw. Kl. 11: 1-164. Gaffney E. S. 1979. Comparative cranial morphology of recent and fossil turtles. Bull. Amer. Mus. Nut. Hist. 164: 65-376. Galis F. and Drucker E. G. 1996. Pharyngeal biting mechanics in centrarchid and cichlid fishes: insights into a key evolutionary innovation. J. Evol. Biol. 9: 641-670. Gans C. 1952. The functional morphology of the egg-eating adaptations in the snake genus Dasypeltis. Zoologica 37: 209-244. Gans C. 1961. The feeding mechanism of snakes and its possible evolution. Amer. Zool. 1: 217-227. Gans C. 1969. Comments on inertial feeding. Copeia 1969: 855-857. Gans C. and De Vree F. 1986. Shingle-back lizards crush snail shells using temporal summation (tetanus) to increase the force of the adductor muscles. Experientia 42: 387-389. Gegenbauer C. 1878. Elements of Comparative Anatomy. MacMillan and Co., London (2nd. ed). German R. Z. and Crompton A. W. 2000. The ontogeny of feeding in mammals. In: Feeding. Form, Function and Evolution in Tetrapod Vertebrates. K. Schwenk (ed.). Acad. Press, San Diego, CA, (USA) pp. 449457. Gillis G. B. and Lauder G. V. 1994. Aquatic prey transport and the comparative kinematics of Ambystoma tigrinum feeding behaviors. J. Exp. Biol. 187: 159-179. Gillis G. B. and Lauder G. V. 1995. Kinematics of feeding in bluegill sunfish: is there a general distinction between aquatic capture and transport behaviors? J. Exp. Biol. 198: 709-720. Gionfriddo J. P. and Best L. P. 1996. Grit-use patterns in North American birds: The influence of diet, body size, and gender. Wilson Bull. 108: 685-696. Goodrich J. S., Sanderson S. L., Batjakas I. I. and Kaufman, L. S. 2000. Branchial arches of suspension-feeding Oreochromis esculentus: sieve or sticky suspension? 1. Fish Biol. 56: 858-875.
34
Physiological and ecological adaptations t o feeding i n vertebrates
Gorniak G. C., Rosenberg H. I. and Gans, C. 1982. Mastication in the tuatara, Sphenodon punctatus (Reptilia: Rhynchocephalia): structure and activity of the motor system. J. Morph. 171: 321-353. Grajal A. 1995: Structure and function of the digestive tract of the hoatzin (Opisthocomus hoazin): a folivorous bird with foregut fermentation. Auk 112: 20-28. Gray L. A., O'Reilly J. C. and Nichikawa K. C. 1997. Evolution of forelimb movement patterns for prey manipulation in anurans. J. Exp. Zool. 277: 417424. Griffiths M. 1978: The Biology of the Monotremes. Acad. Press, New York, NY. Griffiths T. A. 1982. Systematics of the New World nectar-feeding bats (Mammalia, Phyllostomidae), based on the morphology of the hyoid and lingual regions. Amer. Mus. Nov. 2742: 1 4 5 . Grubich J. R. 2000. Crushing motor patterns in drum (Teleostei: Sciaenidae): functional novelties associated with molluscivory. J. Exp. Biol. 203: 3161-3176. Grubich J. R. 2001. Prey capture in actinopterygian fishes: a review of suction feeding motor patterns with new evidence from an elopomorph fish, Megalops atlanticus. Amer. ZOO^. 41: 1258-1265. Hernandez L. P. and Motta P. J. 1997. Trophic consequences of differential performance: ontogeny of oral jaw-crushing performance in the sheepshead, Archosargus probatocephalus (Teleostei, Sparidae). J. Zool., Lond. 243: 737-756. Herrel A., De Vree F., Delheusy V. and Gans C. 1999. Cranial kinesis in gekkonid lizards. J. Exp. Biol. 202: 3687-3698. Herrel A., O'Reilly J. C. and Richmond A. M. 2002. Evolution of bite performance in turtles. J. Evol. Biol. 15: 1083-1094. Herrel A., Meyers J. J., Aerts P. and Nishikawa K. C. 2000. The mechanics of prey prehension in chameleons. J. Exp. Biol. 203: 3255-3263. Herrel A., Meyers J. J., Nishikawa K. C. and De Vree F. 2001: The evolution of feeding motor patterns in lizards: modulatory complexity and possible constraints. Amer. ZOO^. 41: 1311-1320. Herring S. W. 1993. Functional morphology of mammalian mastication. Amer. Zool. 33: 289-299. Hiiemae K. M. 2000. Feeding in mammals. In: Feeding. Form, Function and Evolution in Tetrapod Vertebrates. K. Schwenk (ed.). Acad. Press, San Diego, CA, pp. 411448. Hiiemae K. M. and Crompton A. W. 1985. Mastication, food transport and swallowing. In: Functional Vertebrate Morphology. M. Hildebrand, D. M. Bramble, K. F. Liem, and D. Wake (eds.). Harvard Univ. Press, Cambridge, MA, pp. 262-290. Hiiemae K. M, Thexton A. J. and Crompton A. W. 1978. Intraoral food transport: the fundamental mechanism of feeding. In: Muscle Adaptation in the Craniofacial Region. Monograph No. 8, Craniofacial Growth Series. D. Carlson and J. MacNamara (eds.). Univ. Michigan, Ann Arbor, MI, USA. pp. 181-208. Hildebrand M. and Goslow Jr. G. E. 2001. Analysis of Vertebrate Structure. John Wiley & Sons, New York, NY, USA. (Sh ed.). Hoese W. J. and Westneat M. W. 1996. Biomechanics of cranial kinesis in birds: testing linkage models in the white-throated sparrow (Zonotrichia albicollis). J. Morph. 227: 305-320. Homberger D. G. 1980 Funktionell-morphologische Untersuchungen zur Radiation der Ernahrungs- und Trinkmethoden der Papageien (Psittaci). Bonn. Zool. Monogr. 13, 1-192. Homberger D. G . 1986. The lingual apparatus of the African grey parrot, Psittacus erithacus Linne (Aves: Psittacidae): description and theoretical mechanical analysis. Ornith. Monogr. (Amer. Ornith. Union) 39: 1-233. Horn M. H. 1989. Biology of marine herbivorous fishes. Oceanogr. Mar. Biol. Ann. Rev. 27: 167-272. Horn M. H. 1992. Herbivorous fishes: Feeding and digestive mechanisms. Plant-Animal Interactions in Marine Benthos, vol. 46. D. M. John, S. J. Hawkins, and J. H. Price (eds.). Clarendon Press, Oxford, UK, pp. 339-362.
Vertebrate feeding systems
35
Irish F. 1983. The dynamics of seed-eating in the characiform fish Colossoma. Amer. Zool. 23: 1027. Jayne B. C., Voris H. K. and Ng P. K. L. 2002. Snake circumvents constraints on prey size. Nature 418: 143. Kallius E. 1901. Beitrage zur Entwickelung der Zunge. I. Teil. Amphibien und Reptilien. Anatomischen Institut zu Gottingen, Anat. Heft. 52/53: 532-760. Karasov W. H. 1987. Nutrient requirements and the design and function of guts in fish, reptiles, and mammals. In: Conlparative Physiology: Life in Water and on Land. Fidia Research Series, IX. P. Dejours, L. Bolis, C. R. Taylor, and E. R. Weibel (eds.). Liviana Press, Padovas, Italy, pp. 181-191. Karasov W. H. and Diamond J. M. 1985. Digestive adaptations for fueling the cost of endothermy. Science 228: 202-204. Karasov W. H. and Diamond J. M. 1988. Interplay between physiology and ecology in digestion. BioScience 38: 602-611. Kardong K. V. and Bels V. L. 2001. Functional morphology and evolution of the feeding apparatus in squamates. In: Vertebrate Functional Morphology. Horizons of Research in the 21" Century. H. M. Dutta and J. S. Datta Munshi (eds.). Science Publ., Enfield, NH, (USA), pp. 173-219. Kastelein R. A., Muller M. and Terlouw A. 1994. Oral suction of a Pacific walrus (Odobenus rosmarus divergens) in air and under water. Zeitschr. Saugetierk. 59: 105-115 Kier W. M. and Smith K. K. 1985. Tongues, tentacles and trunks: the biomechanics of movement in muscular-hydrostats. Zool. J. Linn. Soc. 83: 307-324. King A. S. and King D. Z. 1979. Avian morphology: general principles. In: Form and Function in Birds., vol. 1, A. S. King and J. McLelland (eds.). Acad. Press, New York, NY, USA, pp. 1-38. Kingdon J. 1979. East African Mammals. An Atlas of Evolution in Africa, vol. IIIb. Large Mammals. Univ. of Chicago Press, Chicago, IL. Kley N. J. 2001. Prey transport mechanisms in blindsnakes and the evolution of unilateral feeding systems in snakes. Amer. Zool. 41: 1321-1337. Kley N. J. and Brainerd E. L. 2002. Post-cranial prey transport mechanisms in the black pinesnake, Pituophis melanoleucus lodingi: a n x-ray videographic study. Zoology 105: 153-164. Kooloos J. G. M., Kraaijeveld A. R., Langenbach G . E. J. and Zweers G . A. 1989. Comparative mechanics of suspension feeding in Anas platyrhynchos, Anas clypeata, and Aythya fuligula (Aves, Anseriformes). Zoomorphology 108: 269-290. Larsen Jr., J. H., Beneski Jr., J. T. and Wake D. B. 1989. Hyolingual feeding systems of the Plethodontidae: comparative kinematics of prey capture by salamanders with free and attached tongues. J. Exp. Zool. 252: 25-33. Lauder G . V. 1983a. Food capture. In: Fish Biomechanics. P. W. Webb and D. Weihs (eds.). Praeger Publ, New York, NY, USA, pp. 280-311. Lauder G. V. 198313. Functional design and evolution of the pharyngeal jaw apparatus in euteleostean fishes. Zool. J. Linn. Soc. 77: 1-38. Lauder G . V. 1985a. Aquatic feeding in lower vertebrates. In: Functional Vertebrate Morphology. M. Hildebrand, D. M. Bramble, K. F. Liem, and D. B. Wake (eds.). Harvard Univ. Press, Cambridge, MA, USA, pp. 210-229. Lauder G . V. 198513. Functional morphology of the feeding mechanism in lower vertebrates. In: Functional Morphology in Vertebrates (Fortschr. Zool. 30). H.-R. Duncker and G. Fleischer (eds.). Gustav Fischer Verlag, Stuttgart, pp. 179-188. Lauder G . V. and Shaffer H. B. 1985. Functional morphology of the feeding mechanism in aquatic ambystomatid salamanders. J. Morph. 185: 297-326. Lauder G . V. and Shaffer H. B. 1986. Functional design of the feeding mechanism in lower vertebrates: unidirectional and bidirectional flow systems in the tiger salamander. Zool. J. Linn. Soc. 88: 277-290. Lauder G . V. and Shaffer H. B. 1988. Ontogeny of functional design in tiger salamanders (Ambystoma tigrinum): are motor patterns conserved during major morphological transformations? J. Morph. 197: 249-268.
36
Physiological and ecological adaptations t o feeding in vertebrates
Lauder G. V. and Prendergast T. 1992. Kinematics of aquatic prey capture in the snapping turtle Chelydra serpentina. J. Exp. Biol. 164: 55-78. Lauder G. V. and Shaffer H. B. 1993. Design of feeding systems in aquatic vertebrates: major patterns and their evolutionary interpretations. In: The Skull, vol. 3. Functional and Evolutionary Mechanisms. J . Hanken and B. K. Hall (eds.). Univ. Chicago Press, Chicago, IL, pp. 113-149. Lauder G. V. and Reilly S. M. 1994. Amphibian feeding behavior: comparative biomechanics and evolution. In: Biomechanics of Feeding in Vertebrates (Advances in Comparative 6 Environmental Physiology 18). V. L. Bels, M. Chardon, and P. Vandewalle (eds.). Springer-Verlag, Berlin, pp. 163-195. Lauder G. V. and G i l l i s G. B. 1997. Origin of the amniote feeding mechanism: experimental analyses of outgroup clades. In: Amniote Origins. S. S. Sumida and K. L. M. Martin (eds.). Acad. Press, San Diego, CA, USA, pp. 169-206. Lemell P., Beisser C. J. and Weisgram, J. 2000. Morphology and function of the feeding apparatus of Pelusios castaneus (Chelonia; Pleurodira). 1. Morph. 244: 127-135. Lemell P., Lemell C., Snelderwaard P., Gumpenberger M., Wocheslander R. and Weisgram J. 2002. Feeding patterns of Chelus fimbriatus (Pleurodira: Chelidae). J. Exp. Biol. 205: 1495-1506. Lieberman D. E. and Crompton A. W. 2000. Why fuse the mandibular symphysis? A comparative analysis. Amer. 1. Phys. Anth. 112: 517-540. Liem K. F. 1973. Evolutionary strategies and morphological innovations: cichlid pharyngeal jaws. Syst. Zool. 22: 425-441. Liem K. F. 1979. Modulatory multiplicity in the functional repertoire of the feeding mechanism in cichlid fishes I. Piscivores. J. Morph. 158: 323-360. Liem K. F. 1980. Adaptive significance of intra- and interspecific differences in the feeding repertoires of cichlid fishes. Amer. Zool. 20: 295-314. Liem K. F. 1990. Aquatic versus terrestrial feeding modes: possible impacts on the trophic ecology of vertebrates. Amer. Zool. 30: 209-221. Liem K. F. 1993. Ecomorphology of the teleostean skull. In: The Skull, vol. 3. Functional and Evolutionary Mechanisms. J. Hanken and B. K. Hall (eds.), Univ. Chicago Press, Chicago, IL, pp. 422-452. Liem K. F. and Osse J. W. M. 1975. Biological versatility, evolution, and food resource exploitation in African cichlid fishes. Amer. Zool. 15: 427-454. Liem K. F. and Greenwood P. H. 1981. A functional approach to the phylogeny of the pharyngognath teleosts. Amer. Zool. 21: 83-101. Liem K. F. and Sanderson S. L. 1986. The pharyngeal jaw apparatus of labrid fishes: a functional morphological perspective. J. Morph. 187: 143-158. Liem K. F., Bemis W. E., Walker Jr., W. F. and Grande L. 2001. Functional Anatomy of the Vertebrates. Harcourt College Publ., Fort Worth, TX, USA, (Yded). Lombard R. E. and Wake D. B. 1976. Tongue evolution in the lungless salamanders, family Plethodontidae I. Introduction, theory and a general model of dynamics. 1. Morph. 148: 265-286. Mallatt J. 1984. Feeding ecology of the earliest vertebrates. Zool. J. Linn. Soc. 82: 261-272. Mallatt J. 1996. Ventilation and the origin of jawed vertebrates: a new mouth. Zool. J. Linn. SOC.117: 329-404. McBrayer L. D. and Reilly S. M. 2002a. Prey processing in lizards: behavioral variation in sit-and-wait and widely foraging taxa. Can. J. Zool. 80: 882-892. McBrayer L. D. and Reilly S. M. 2002b. Testing amniote models of prey transport kinematics: a quantitative analysis of mouth opening patterns in lizards. Zoology 105: 71-81. McLelland J. 1979. Digestive system. In: Form and Function in Birds, Vol. 1. A. S. King and J. McLelland (eds.). Acad. Press, New York, NY, pp. 69-181. Metzger K. 2002. Cranial kinesis in lepidosaurs: skulls in motion. In: Topics in Functional and Ecological Vertebrate Morphology. P. Aerts, K. D'AoQt, A. Herrel, and R. Van Damme (eds.). Shaker Publ., Maastricht, Netherlands, pp. 15-46.
Vertebrate feeding systems
37
Meylan A. 1988. Spongivory in hawksbill turtles: a diet of glass. Science 239: 393-395. Miller B. T.and Larsen Jr., J. H., 1989. Feeding performance in aquatic postmetamorphic newts (Urodela: Salamandridae): are bidirectional flow systems necessarily inefficient? Can. J. Zool. 67: 2414-2421. Moon B. R. 2000. The mechanics of swallowing and the muscular control of diverse behaviours in gopher snakes. J. Exp. Biol. 203: 2589-2601. Moss S. A. 1977. Feeding mechanisms in sharks. Amer. Zool. 17: 355-364. Motta P. J. and Wilga C. D. 2001. Advances in the study of feeding behaviors, mechanisms, and mechanics of sharks. Env. Biol. Fishes 60: 131-156. Miiller M. and Osse J. W. M. 1984. Hydrodynamics of suction feeding in fish. Trans. Zool. Soc. Lond. 37: 51-135. Nemeth D. H. 1997. Modulation of buccal pressure during prey capture in Hexagrammos decagarmmus (Teleostei: Hexagrammidae). J. Exp. Biol. 200: 2145-2154. Nishikawa K. C. 2000. Feeding in frogs. In: Feeding Form, Function and Evolution in Tetrapod Vertebrates. K. Schwenk (ed.). Acad. Press, San Diego, CA, USA, pp. 117-147. Nishikawa K. C., Kier W. M. and Smith K. K. 1999. Morphology and mechanics of tongue movement in the African pig-nosed frog Hemisis marmoratum: a muscular hydrostatic model. J. Exp. Biol. 202: 771-780. Northcott M. E. and Beveridge M. C. M. 1988. The development and structure of pharyngeal apparatus associated with suspension feeding in tilapias (Oreochromis niloticus). J. Zool., Lond. 215: 133-149. Northcutt R. G. and Gans C. 1983. The genesis of neural crest and epidermal placodes: a reinterpretation of vertebrate origins. Quart. Rev. Biol. 58: 1-28. Norton S. F. and Brainerd, E. L. 1993. Convergence in the feeding mechanics of ecomorphologically similar species in the Centrarchidae and Cichlidae. J. Exp. Biol. 176: 11-29. Novacek M. J. 1993. Patterns of diversity in the mammalian skull. In: The Skull, vol. 2. Patterns of Stuctural and Systematic Diversity. J. Hanken and B. K. Hall (eds.). Univ. Chicago Press, Chicago, IL, pp. 438-545. Nuijens F. W. and Zweers G. A. 1997. Characters discriminating two seed husking mechanisms in finches (Fringillidae: Carduelinae) and estrildids (Passeridae: Estrildinae). J. Morph. 232: 1-33. O'Reilly J. C. 2000. Feeding in caecilians. In: Feeding. Form, Function and Evolution in Tetrapod Vertebrates. K. Schwenk (ed.). Acad. Press, San Diego, CA, pp. 149-166. Owen R. 1866. On the Anatomy of Vertebrates, Vol. I. Fishes and Reptiles. Longmans, Green, and Co., London, UK. Owen R. 1868 On the Anatomy of Vertebrates, Vol. 111. Mammals. London: Longmans, Green and Co., London, UK. Pooley A. C. and Gans C. 1976. The Nile crocodile. Sci. Amer. 234: 114-124. Proctor N. S. and Lynch P. J. 1993: Manual of Ornithology. Avian Structure and Function. Yale Univ. Press, New Haven, CT (USA). Reif W. E. 1982. Evolution of dermal skeleton and dentition in vertebrates: the odontode regulation theory. Evol. Biol. 15: 287-368. Reilly S. M. 1996. The metamorphosis of feeding kinematics in Salamandra salamandra and the evolution of terrestrial feeding behavior. J. Exp. Biol. 199: 1219-1227. Reilly S. M. and Lauder G. V. 1990. The evolution of tetrapod feeding behavior: kinematic homologies in prey transport. Evolution 44: 1542-1557. Reilly S. M., McBrayer L. D. and White T. D. 2001. Prey processing in amniotes: biomechanical and behavioral patterns of food reduction. Comp. Biochem. Physiol. A 128: 397-415. Reiss K. Z. 2000. Feeding in myrmecophagous mammals. In: Feeding Form, Function and Evolution in Tetrapod Vertebrates. K. Schwenk (ed.) Acad. Press, San Diego, CA, USA, pp. 459-485.
38
Physiological and ecological adaptations t o feeding i n vertebrates
Richardson, K. C., Webb G. J. W. and Manolis S. C. 2002. Crocodiles: lnside Out. A Guide to the Crocodilians and Their Functional Morphology. Surrey Beatty & Sons, Chipping Norton, NSW, Australia. Ritter D. A. and Nishikawa K. C. 1995. The kinematics and mechanism of prey capture in the African pig-nosed frog (Hemisis marmoratum): the description of a radically divergent anuran tongue. J. Exp. Biol. 198: 2025-2040. Rowe T. 1996. Coevolution of the mammalian middle ear and neocortex. Science 275: 683-684. Rubega M. A. 1996. Sexual size dimorphism in red-necked phalaropes and functional significance of nonsexual bill structure variation for feeding performance. J. Morph. 228: 45-60. Rubega M. A. 1997. Surface tension prey transport in shorebirds: how widespread is it? Ibis 139: 488-493. Rubega M. A. 2000. Feeding in birds: approaches and opportunities. In: Feeding. Form, Function and Evolution in Tetrapod Vertebrates. K. Schwenk (ed.). Acad. Press, San Diego, CA, USA, pp. 395-408. Rubega M. A. and Obst B. S. 1993. Surface-tension feeding in phalaropes: discovery of a novel feeding mechanism. The Auk 110: 169-178. Russell E. S. 1916. Form and Function. A Contribution to the History of Animal Morphology. John Murray, London, UK. Sanderson S. L. Cheer A. Y., Goodrich J. S., Graziano J. D. and Callan W. T. 2001. Crossflow filtration in suspension-feeding fishes. Nature 412: 439-441. Sanderson S. L., and Kupferberg S. J . 1999. Development and evolution of aquatic larval feeding mechanisms. In: The Origin and Evolution of Larval Forms. B. K. Hall and M. H. Wake (eds.). Acad. Press, San Diego, CA, USA, pp. 301-377. Sanderson S. L., Stebar M. C., Ackermann K. L., Jones S. H., Batjakas J. E. and Kaufman L. 1996. Mucus entrapment of particles by a suspension-feeding Tilapia (Pisces: Cichlidae). J. Exp. Biol. 199: 1743-1756. Sanderson S. L. and Wassersug R. 1993. Convergent and alternative designs for vertebrate suspension feeding. In: The Skull, vol. 3. J. Hanken and B. K. Hall (eds.). Univ. Chicago Press, Chicago, IL, pp. 37-112. Sanford C. P. and Lauder G. V. 1989. Functional morphology of the "tongue-bite" in the osteoglossomorph fish Notopterus. J. Morph. 202: 379-408. Sanford C. P. J. and Wainwright P. C. 2002. Use of sonomicrometry demonstrates the link between prey capture kinematics and suction pressure in largemouth bass. J. Exp. Biol. 205: 3445-3457. Schwenk K. 2000a. An introduction to tetrapod feeding. In: Feeding. Form, Function and Evolution in Tetrapod Vertebrates. K. Schwen (ed.). Acad. Press, San Diego, CA, USA, pp. 175-291. Schwenk K. 2000b. Feeding. Form, Function and Evolution in Tetrapod Vertebrates. Acad. Press, San Diego, USA, CA. Schwenk K. 2000c. Feeding in lepidosaurs. In: Feeding. Form, Function and Evolution in Tetrapod Vertebrates. K. Schwenk (ed.). Acad. Press, San Diego, CA, USA, pp. 459-485. Schwenk K. 2000d. Tetrapod feeding in the context of vertebrate morphology. In: Feeding. Form, Function and Evolution in Tetrapod Vertebrates. K. Schwenk (ed.). Acad. Press, San Diego, CA, USA, pp. 3-20. Schwenk K. 2001a. Extrinsic versus intrinsic lingual muscles: a false dichotomy? Bull. Mus. Comp. Zool. (Harvard) 156: 219-235. Schwenk K. 2001b. Functional units and their evolution. In: The Character Concept in Evolutionary Biology. G. P. Wagner (ed.). Acad. Press., San Diego, CA, USA, pp. 165-198. Schwenk K. and Bell D. A. 1988. A cryptic intermediate in the evolution of chameleon tongue projection. Experientia 44: 697-700.
Vertebrate feeding systems
39
Schwenk K. and Throckmorton G . S. 1989. Functional and evolutionary morphology of lingual feeding in squamate reptiles: phylogenetics and kinematics. 1. Zool., Lond. 219: 153-175. Schwenk K. and Wake D. B. 1993. Prey processing in Leurognathus marmoratus and the evolution of form and function in desmognathine salamanders (131ethodontidae).Biol. 1. Linn. Soc. 49: 141-162. Schwenk K. and Wagner G . P. 2001. Function and the evolution of phenotypic stability: connecting pattern to process. Amer. Zool. 41: 552-563. Secor S. M. and Diamond J. 1998. A vertebrate model of extreme physiological regulation. Nature 395: 659-662. Shaffer H. B. and Lauder G . V. 1988. The ontogeny of functional design: metamorphosis of feeding behaviour in the tiger salamander (Ambystoma tigrinum). 1. Zool., Lond. 216: 437-454. Sibbing F. A. 1982. Pharyngeal mastication and food transport in the carp (Cyprinus carpio): a cineradiographic and electromyographic study. 1. Morph. 172: 223-258. Sibbing F. A. 1991. Food processing by mastication in cyprinid fish. In: Feeding and the Texture of Food. J. F. V. Vincent and P. J. Lillford (eds.). Cambridge: Cambridge Univ. Press, Cambridge, UK, pp. 57-92. Sibbing F. A., Osse J. W. M. and Terlow A. 1986. Food handling in the carp (Cyprinus carpio): its movement patterns, mechanisms and limitations. 1. Zool., Lond. 210: 161-203. Smith K. K. 1984. The use of the tongue and hyoid apparatus during feeding in lizards (Ctenosaura similis and Tupilzambis nigropunctatus). 1. Zool., Lond. 202: 115-143. Smith K. K. 1986. Morphology and function of the tongue and hyoid apparatus in Varanus (Varanidae, Lacertilia). 1. Morph. 187: 261-287. Smith K. K. 1992. The evolution of the mammalian pharynx. Zool. 1. Linn. Soc. 104: 313-349. Smith K. K. 1994. Are neuromuscular systems conserved in evolution? Brain Behav. Evol. 43: 293-305. Smith K. K. and Kier W. M. 1989. Trunks, tongues, and tentacles: moving with skeletons of muscle. Amer. Sci. 77: 29-35. Starck J. M. and Beese K. 2001. Structural flexibility of the intestine of the Burmese python in response to feeding. 1. Exp. Biol. 204: 325-335. Stevens C. E. and Hume I. D. 1995. Comparative Physiology of the Vertebrate Digestive System. Cambridge Univ. Press, Cambridge, UK. Summers A. P. 2000. Stiffening the stingray skeleton-an investigation of durophagy in myliobatid stingrays (Chondrickhthyes, Batoidea, Myliobatidae). 1. Morph. 243: 113-126. Summers A. P., Darouian K. F., Richmond A. M. and Brainerd E. L. 1998. Kinematics of aquatic and terrestrial prey capture in Terrapene carolina, with implications for the evolution of feeding in cryptodire turtles. 1. Exp. Zool. 281: 280-287. Taylor M. A. 1987. How tetrapods feed in water: a functional analysis by paradigm. Zool. 1. Linn. Soc. 91: 171-195. Taylor M. A. 1993. Stomach stones for feeding or buoyancy? The occurrence and function of gastroliths in marine tetrapods. Phil. Trans. Roy. Soc. Lond B 341: 163-175. Thexton A. J. 1973. Some aspects of neurophysiology of dental interest. Part I. Theories of oral function. 1. Dent. 2: 49-54. Thexton A. J. and Crompton A. W. 1998. The control of swallowing. In: The Scientific Basis of Eating. Frontiers of Oral Biology Series, vol 9. R. A. Linden (ed.). Karger, Basel, pp. 168-222. Thomas R. G . and Pough F. H. 1979. The effect of rattlesnake venom on digestion of prey. Toxicon 17: 221-228. Thorogood P. 1993. Differentiation and morphogenesis of cranial skeletal tissues. In: The Skull, vol. 1. Development J. Hanken and B. K. Hall (eds.), Univ. Chicago Press, Chicago, IL, pp. 112-152.
40
Physiological and ecological adaptations t o feeding in vertebrates
Tomlinson C. A. B. 2000. Feeding in paleognathous birds. In: Feeding. Form, Function and Evolution in Tefrapod Vertebrates. K. Schwenk (ed.). Acad. Press, San Diego, CA, USA, pp. 359-394. Turingan R. G. and Wainwright P. C. 1993. Morphological and functional bases of durophagy in the queen triggerfish, Balisfes vefula (Teleostei: Tetradontiformes). J. Morph. 215: 101-118. Van Damme J. and Aerts, P. 1997. Kinematics and functional morphology of aquatic feeding in Australian snake-necked turtles (Pleurodira; Chelodina). J. Morph. 233: 113-125. Van Damme J. and Aerts P. 2002. Cervical movements during prey capture in the Australian snake-necked turtle, Chelodina sp. (Pleurodira, Chelidae). In: Topics i n Funcfional and Ecological VerfebrafeMorphology. P. Aerts, K. D'Aotit, A. Herrel, and R. Van Damme (eds.). Shaker Publ., Maastricht, pp. 77-94. Van Valkenburgh B. 1996. Feeding behavior in free-ranging, large African carnivores. J. Mamm. 77: 240-254. Vandewalle P., Huyssene A., Aerts P. and Verraes W. 1994. The pharyngeal apparatus in teleost feeding. In: Biomechanics of Feeding in Verfebrafes (Advances in Comparafive 6 Environmenfal Physiology 18). V. L. Bels, M. Chardon, and P. Vandewalle (eds.). SpringerVerlag, Berlin, pp. 59-92. Vences M., Glaw F. and Bohme W. 1997/98. Evolutionary correlates of microphagy in alkaloid-containing frogs (Amphibia: Anura). Zool. Anz. 236: 217-230. Vitt L. J., Pianka E. R., Cooper W. E. and Schwenk K. 2003. History and the global ecology of squamate reptiles. Amer. N a f . 162: 44-60. Vogt R. C., Sever D. M. and Moreira G. 1998. Esophageal papillae in pelomedusid turtles. J. Herp. 32: 279-282. Wagner G. P. and Schwenk K. 2000. Evolutionarily stable configurations: functional integration and the evolution of phenotypic stability. Evol. Biol. 31: 155-217. Wainwright P. C. 1987. Biomechanical limits to ecological performance: mollusc-crushing by the Caribbean hogfish, Lachnolaimus maximus (Labridae). J . Zool., Lond. 213: 283-297. Wainwright P. C. 1988. Morphology and ecology: functional basis of feeding constraints in Caribbean labrid fishes. Ecology 69: 635-645. Wainwright P. C. and Bennett A. F. 1992. The mechanism of tongue projection in chameleons. 11. Role of shape change in a muscular hydrostat. J. Exp. Biol. 168: 2340. Wainwright P. C., Ferry-Graham L. A., Waltzek T. B., Carroll A. M., Hulsey C. D. and Grubich J. R. 2001. Evaluating the use of ram and suction during prey capture by cichlid fishes. J. Exp. Biol. 204: 3039-3051. Wainwright P. C. and Friel J. P. 2001. Behavioral characters and historical properties of motor patterns. In: The Character Concepf in Evolufionary Biology G. P. Wagner (ed.). Acad. Press, San Diego, CA, pp. 285-301. Wainwright P. C. and Turingan R. G. 1993 Coupled versus uncoupled functional systems: motor plasticity in the queen triggerfish Balisfes vefula. J. Exp. Biol. 180: 209-227. Wainwright P. C., Westneat M. W. and Bellwood D. R. 2000. Linking feeding behaviour and jaw mechanics in fishes. In: Biomechanics in Animal Behaviour. P. Domenici and R. W. Blake (eds.). BIOS Scientific Publ., Oxford, UK, pp. 207-221. Wake D. B. 1982: Functional and developmental constraints and opportunities in the evolution of feeding systems in urodeles. In: Environmenfal Adapfafion and Evolution. D. Mossakowski and G. Roth (eds.). Fischer, Stuttgart, pp. 51-66. Wake D. B. and Deban S. M. 2000. Terrestrial feeding in salamanders. In: Feeding. Form, Funcfion and Evolufion i n Tefrapod Vertebrates. K. Schwenk (ed.). Acad. Press, San Diego, CA, pp. 95-116. Wassersug R. J. and Yamashita M. 2001. Plasticity and constraints on feeding kinematics in anuran larvae. Comp. Biochem. Physiol. A 131: 183-195.
Vertebrate feeding systems
41
Weijs W. A. 1994. Evolutionary approach of masticatory motor patterns in mammals. In: Biomechanics of Feeding in Vertebrates (Advances in Comparative & Environmental Physiology 18). V. L. Bels, M. Chardon, and P. Vandewalle (eds.). Springer-Verlag, Berlin, pp. 281-320. Weisgram J. and Splechtna H. 1992. Cervical movement during feeding in Chelodina novaeguinaeae (Chelonia, Pleurodira). Zool. Jb. Anat. 122: 331-337. Weissengruber G. E., Forstenpointner G., Peters G., Kiibber-Heis A. and Fitch W. T. 2003. Hyoid apparatus and pharynx in the lion (Panthera leo), jaguar (Panthera onca), tiger (Panthera tigris), cheetah (Acinonyx jubatus) and domestic cat (Felis selvestris f. cafus). J. Anat. 201: 195-210. Welty J. C. 1975. The Life of Birds. W. B. Saunders, Philadelphia, PA, USA, (2nded.). Werth A. 2000a. A kinematic study of suction feeding and associated behavior in the long-finned pilot whale, Globicephala melas (Traill). Mar. Mam. Sci. 16: 299-314. Werth A. 2000b. Feeding in marine mammals. In Feeding. Form, Function and Evolution in Tetrapod Vertebrates. K. Schwenk (ed.). Acad. Press, San Diego, CA, USA, pp. 487-526. Westneat M. W. 1990. Feeding mechanics of teleost fishes (Labridae): a test of four-bar linkage models. J Morph. 205: 269-295. Westneat M. W. 2001. Ingestion in fishes. In Encyclopedia of Life Sciences. Macmillan Publ., London, UK. Wilga C. D. and Motta P. J. 1998. Conservation and variation in the feeding mechanism of the spiny dogfish Squalus acanthias. J. Exp. Biol. 201: 1345-1358. Wilga C. D. and Motta P. J. 2000. Durophagy in sharks: feeding mechanics of the hammerhead Sphyrna tiburo. J. Exp. Biol. 203: 2781-2796. Wilga C. D., Wainwright P. C. and Motta P. J. 2000. Evolution of jaw depression mechanics in aquatic vertebrates: insights from Chondrichthyes. Biol. J. Linn. Soc. 71: 165-185. Wilga C. D., Hueter R. E., Wainwright P. C . and Motta P. J. 2001. Evolution of upper jaw protrusion mechanisms in elasmobranchs. Amer. Zool. 41: 1248-1257. Wocheslander R., Hilgers H. and Weisgram J. 1999. Feeding mechanism of Tesfudo hermanni boettgeri (Chelonia, Cryptodira). Nefh. J. Zool. 49: 1-13. Woodger J. H. 1929. Biological Principles. A Critical Study. Kegan Paul, Trench, Trubner, New York, NY. Wu E. H. 1994. Kinematic analysis of jaw protrusion in orectolobiform sharks: a new mechanism for jaw protrusion in elasmobranchs. J. Morph. 222: 175-190. Ziswiler V. a n d Farner D. S. 1972. Digestion a n d the digestive system. In: Avian Biology, vol. 11. D. S. Farner and J. R. King (eds.). Acad. Press, New York, NY, pp. 343-430. Zusi R. L. 1993. Patterns of diversity in the avian skull. In: The Skull, Vol. 2. Patterns of Structural and Systematic Diversify. J. Hanken and B. K. Hall (eds.). Univ. Chicago Press, Chicago, IL, pp. 391-437. Zweers G. A. 1985. Generalism and specialism in avian mouth and pharynx. In: Functional Morphology in Vertebrates (Fortschr. Zool. 30). H.-R. Duncker and G. Fleischer (eds.). Gustav Fischer Verlag, Stuttgart, pp. 189-201. Zweers G. A. and Berkhoudt H. 2001. Ingestion in birds. In: Encyclopedia of Life Sciences. Macmillan Publ., London, LTK. Zweers G. A., Gerritsen A. F. C. and van Kranenburg-Vood P. J. 1977. Mechanics of feeding of the mallard (Anas platyrhynchos L.; Aves, Anseriformes). Contrib. Vert. Evol. Karger, Basel. Zweers G. A., Berkhoudt H. and Vanden Berge J. C . 1994. Behavioral mechanisms of avian feeding. In: Biomechanics of Feeding in Vertebrates (Advances in Comparative & Environmental Physiology 18). V. L. Bels, M. Chardon, and P. Vandewalle (eds.). SpringerVerlag, Berlin, pp. 241-279. Zweers G. A., Vanden Berge J. C. and Berkhoudt H. 1997. Evolutionary patterns of avian trophic diversification. Zoology 100: 25-57. Zweers G . A., d e Jong F., Berkhoudt H. and Vanden Berge J. C. 1995. Suspension feeding in flamingos (Phoenicopferus ruber). Condor 97: 297-324.
Concepts of Digestive Efficiency Ian D. Hume University of Sydney, Biological Sciences, NSW, Australia
SYNOPSIS This chapter defines digestibility (digestive efficiency) and addresses three issues associated with the concept of digestive efficiency: the problem of terminology and confusion by some workers of digestibility with metabolizability;the trade-off between maximizing digestive efficiency and rate of net energy gain; and the various factors that affect digestive efficiency.The main animal factors include mean retention time of food in the digestive tract, food particle size and effectiveness of mastication, level of food intake, digestive tract capacity and morphology, hydrolytic enzyme activities, and absorptive capacities.
The potential value of a food for supplying energy or a particular nutrient can be determined by chemical analysisbut the actual value of the food to the animal can be arrived at only after allowing for the inevitable losses that occur during digestion, absorption, and metabolism. Figure 2.1 shows the various avenues of loss for food energy.Of the three main avenues of nutrient loss, that in the feces is usually the greatest by far and also the easiest to measure. Hence, digestibility, the proportion of food not appearing in the feces (Fig.2.1),is used as an attribute of a food almost as much as its chemical composition. Stated another way, digestibilityis the ratio of food absorbed to food ingested. Ratios such as this may be the only meaningful way of interpreting some biological processes (Sokal and Rohlf, 1995),but there can be statistical problems with their use, especially in correlation analysis (Raubenheimer, 1995). Transformation of the data to render them approximately normally distributed can be used to overcome some of the
44
Physiological and ecological adaptations t o feeding in vertebrates
statistical drawbacks (Sokal and Rohlf, 1995).Digestibility (or digestive efficiency) is an example of a biologically meaningful ratio. It has important implications in studies of nutritional ecology, resource exploitation, and energy flow through ecosystems. Its measurement is also critical in the design of feeding systems to maximize production efficiency in commercial livestock feeding programs. Thus there are two distinct contexts in which digestive efficiency can be discussed: the nutritional ecology of wild animals and livestock production systems. The term "digestibility" is more commonly used in livestock production circles, and "digestive efficiency" in the nutritional ecology of wild animals. They are equivalent terms.
U+G=U,+G, +Ue+Ge
- Heat increment of feeding (HIF) - Foraging - Tissue synthesis
Fig. 2.1. Relationships among energy terms used in this chapter, modified from Kleiber (1961). Fe, Ue, and Ge denote energy of endogenous origin in feces, urine, and gases respectively. F,, U,, and G, denote energy of food origin in feces, urine, and gases respectively. The total heat production is divided into basal metabolic rate (BMR) which has a fixed and a variable (flexible) component, and other processes and activities. The efficiencies depicted are apparent efficiencies; for true efficiencies F, U, and G in these equations should be replaced by F,, U,and G, respectively.
Concepts of digestive efficiency
45
In the ecological context two points should be considered before exploring factors that affect digestive efficiency. The first is the choice of the most appropriate unit of measurement. In most cases it is probably energy rather than protein or some other specific nutrient because animals tend to eat to meet their requirements for energy under most circumstances.This is particularly so for birds and mammals because of the high energetic costs of endothermy. However, it is not always convenient or possible to measure the energy content of food and feces and dry matter may be more appropriate. The second point is the question of whether animals attempt to maximize digestive efficiency, or try to maximize the rate of digestible energy intake or some more specific nutrient. Karasov and Hume (1997)concluded that the structure and function of the gastrointestinaltract did not necessarily evolve toward maximal digestive efficiency. They proposed that a more appropriate evolutionary goal might be maximization of the rate of extraction of nutrients, be it in terms of energy or some other limiting resource. Penry and Jumars (1987) in their application of chemical reactor theory to animal digestive systems, defined their design objective as the maximum conversion of ingested food to assimilableproducts in the minimum of time and gut-reactor volume. This chapter accepts the premise that for most foraging animals the objective is to maximize the rate of net energy gain. This premise is the basis for the simple model of digestion introduced by Sibly (1981;Fig. 2.2). In this model the rate of release of net energy is initially negative as energy is expended by the animal in overcoming the physical defenses of the ingested food, such as the chitinous exoskeletonof arthropods or tough seed coats. This is followed by a period of rapid digestion (e.g. of hemolymph and soft tissues of arthropods, and the contents of plant cells), but thereafter digestion rate falls as digestion is progressively confined to less tractable food components such as the structural proteins of animal tissues and the structural carbohydrates (cellulose and hemicelluloses) of plant cell walls. Optimal digestion time is given by the straight line from the origin tangential to the curve. A
?
Maximal rate of digestion Energy
food
B
/
t
high
Optimal digestion ;/time short
+ Mean retention time of food
Fig. 2.2. Model of digestion in a continuous-flow system for a high-quality (A) and a lowquality (B) food. Modified from Sibly (1981) by Hume (1989).
46
Physiological and ecological adaptations t o feeding in vertebrates
Several predictions arise from the Sibly model. First, optimal digestion time will vary among foods, being longer for poor-quality foods (e.g. hardbodied adult insects) than those of higher quality (soft-bodied insect larvae) (Fig.2.2). Second, because of longer digestion times, animals routinely eating poorer quality foods should have larger digestive tracts (e.g.herbivores versus carnivores, larger versus smaller animals of the same species, and larger-size species of similar gastrointestinal tract morphology such as colon fermenters; Hume, 1989). Third, if gastrointestinal tract capacity is limiting (as in young animals), the optimal strategy is to maximize digestion rate by selecting only high-quality foods. Finally, at any given level of intake, an animal should maximize retention of food in order to maximize the amount of energy absorbed (Sibly, 1981). This is particularly so when food availability is limited, although clearly there is a holding time beyond which net loss occurs. These predictions are explored in the rest of this chapter.
DEFINITIONS AND TERMINOLOGY There is some confusion in the literature as to what is actually being measured by digestive efficiency, partly because synonyms for digestive efficiency are loosely employed. Efficiency implies a ratio, usually between output and input. Digestive efficiency is the ratio (often expressed as a percent) between output in terms of dry matter, energy, or a nutrient absorbed (i.e.not eliminated in the feces) and its intake. Feces consist only of undigested food residues and metabolic products (including bacteria) of the gut and not renal excretory products. Absorption is usually taken to be synonymous with digestion, that is, an item digested is assumed to be absorbed. Thus equivalent terms for digestive efficiency are absorbability (Kleiber, 1961) and absorption efficiency (Speakman, 1987). The assumption of synonymy between digestion and absorption is probably valid in a goodly majority of cases, but instances in which the absorptive capacity of the gut can be exceeded are known (see below). Other alternative terms for digestive efficiency are extraction efficiency (Karasov, 1996;Karasov and Hume, 1997), assimilable mass coefficient, and utilization efficiency (Guglielmo and Karasov, 1993). "Digestibility", "digestion coefficient" or "digestive efficiency" are the preferred terms because they clearly exclude considerations of the efficiency with which nutrients are utilized once they have been absorbed. Also eliminated in the feces are materials not directly of dietary origin but from the animal itself. Of metabolic origin, they are referred to as metabolic fecal dry matter or metabolic fecal nitrogen (MFN) for example. An alternative to and perhaps more precise term than MFN is nondietary fecal nitrogen (NDFN)(Mason,1969)because the microbial component of MFN is not strictly of endogenous (animal) origin. If metabolic fecal material is not accounted
Concepts o f digestive efficiency
47
for, digestibility is "apparent digestibility", as distinct from "true digestibility" if allowance is made for metabolic fecal material. Unless reported digestibilities are defined as true digestibilities they can be assumed to be apparent digestibilities.True digestibility is the ratio of quantity absorbed to quantity ingested (Sibly, 1981) and is always greater than apparent digestibility by the amount of metabolic fecal material (Van Soest, 1994). Measurement of metabolic fecal material is fraught with difficulties. For dry matter and energy it means measurement of fecal output at zero intake of dry matter. Long collection periods from starving animals would be necessary but not ethically acceptable. Also, loss of metabolic fecal material at zero food intake may not reflect losses at higher intakes. Determination of metabolic fecal nitrogen is based on measurement of fecal output on a nitrogen-free diet, but animals usually refuse to eat such unbalanced food. Alternatively, fecal output at zero intake can be estimated by extrapolation from a range of diets of decreasing nitrogen concentration, but rarely can this be done with any degree of confidence. Whenever true digestibilities of energy or dry matter have been estimated in animals at or above maintenance (bear in mind the difficulties in doing this) the differencebetween true and apparent digestibility has usually been small, less than two percentage units (Miller and Reinecke, 1984). However, at low levels of food intake it can be greater, which leads to erroneous ctsnclusions about the efficiency with which energy is extracted from a particular food (Guglielmo and Karasov, 1993).The problem is most acute for protein because fecal metabolic losses of nitrogen are so large, equivalent to about 4 percentage units of crude protein (nitrogen times 6.25). Thus when dietary crude protein falls to 4% of dry matter, as it does in some cereal straws, apparent digestibilityof nitrogen falls to zero, even though its true digestibility may be 85-90% (Van Soest, 1994).For some minerals it is essential to distinguish in the feces that portion which is unabsorbed material from that portion excreted from the body, perhaps by labeling the body pool of that mineral with a radioactive isotope. This is especially so for calcium. Feces are the principal means of elimination of calcium from the body. Apparent digestion coefficients for this mineral are quite meaningless (McDonald et al., 1995). In amphibians, reptiles and birds, feces cannot readily be separated from urine. Inclusion of urine in the excreta yields "apparent metabolizability" rather than apparent digestibility, a distinction often not made in the literature. The problem is greatest with the total collection method of determining digestibility. To minimize the problem, Speakman (1987) used the ratio of natural ash content (ash equivalent) of excreta and food of birds to calculate apparent digestive efficiency, with the assumption that negligible dietary ash was excreted in the urine. However, the result is a compromise at best and does not negate the need for using "metabolizability" whenever urine is included in the excreta.
48
Physiological and ecological adaptations to feeding in vertebrates
True metabolizability takes into account endogenous urinary losses as well as metabolic fecal losses. Methods for estimating endogenous urinary losses are analogous to methods for estimating metabolic fecal losses and suffer from similar shortcomings. Endogenous urinary nitrogen is particularly sensitive to the energy and protein status of the animal. For instance, catabolism of protein to meet essential energy needs elevates endogenous urinary nitrogen (Kleiber,1961).
-
Iguana SNAIL
rusta ace ins
Horse
-
Fish
(LIVEI
-
Orthoptera
. I
Lepidoptera
Change
DRIED
~n
FRESH
61rd(Chicken) Oiptera
Carcass
Hornoptera
Hymenoptera Beef
I
..** oleop pi era
-
Henegg * * * * * ' mollusc^
~soptera
**,
-
~=I!JNG~L
MYCELLA Soyabean
Legumes Pea Peanuts
Otl seed
Cereals
RE PLANTS
Cheslrmts
Acorns
Leaves
I
I
I
I
20
40
60
80
FM FLUID PHLOEM FI UID
Water content (% wet mass)
Fig. 2.3. Materials used by animals as food arranged according to their water (abscissa) and total nitrogen (ordinate) contents, two measures of quality. The nitrogen contents for phloem and xylem are percent of fresh mass; all other nitrogen contents are percent of dry mass from Chivers and Langer (1994) and Slansky and Scriber (1985).
Concepts o f digestive efficiency
FACTORS AFFECTING DIGESTIVE EFFICIENCY The wide variation in digestive efficienciesobserved in nature can be attributed to both dietary and animal effects.
Dietery Effects The diets of vertebrates range from microbial through animal to plant and fungal tissues (Fig. 2.3).In general, animal tissues are more digestible than plant tissues but exceptions include the low digestibility of the exoskeleton of invertebrates and the endoskeleton of vertebrate prey. Plant material can be divided into four major components: cell contents, cell walls, exudates, and secondary metabolites. The digestibility of plant material is often determined by the ratio of cell walls to cell contents (Hume, 1989). Cell contents include cytoplasmic elements that are largely soluble (e.g. enzymes, organic acids, simple sugars) and storage forms of energy (e.g.starch, triacylglycerols). Fruits, seeds and young leaves are high in cell contents and are therefore generally of high digestibility. Cell walls contain (in order of increasing difficulty to digest)pectin, hemicelluloses, cellulose, and lignin. Ligrun is virtually indigestible in the anaerobic conditions of the vertebrate digestive system and reduces the digestibility of hemicelluloses and cellulose by forming a physical barrier between them and hydrolytic enzymes secreted mainly by symbiotic microbes in the digestive tract. Mature leaves and other structural parts of plants (stems,petioles) are high in cell wall content and are therefore highly variable but relatively low in digestibility. Roots may be highly fibrous (and hence of low digestibility)or may contain storage forms of carbohydrate (starch, oligosaccharides)and hence be of higher digestibility. Plant exudates include (in increasing difficulty to digest) nectar, saps, gums and resins. Nectar and saps consist mainly of water and sugars, gums are mainly polymers of sugars and approach hemicellulose in degree of difficulty of digestion, and resins are virtually indigestible. Plant secondary metabolites interfere with digestion (e.g.tannins) or with metabolism of the animal and inhibit food ingestion (e.g.terpenes). Dietary characteristics thus play an important role in determining digestive efficiency. The effects of diet are often clear but in other cases differences in digestive efficiency among animals can be subtle, but nevertheless significant. It is therefore essential that comparisons in digestive efficiency across animal taxa are made on a common diet. This is often not done, leading to misleading statements about taxonomic differences that are confounded by dietary effects. Animal Effects Once potential differences in digestibility of foods are controlled, the greatest differences in digestive efficiency among vertebrates are between taxa. The
50
Physiological and ecological adaptations t o feeding in vertebrates
most potent factor affecting digestive efficiency is likely to be mean retention time (MRT),which is the average time that food stays within the gastrointestinal tract. MRT determines the time that dietary substrates are subjected to attack by digestive enzymes (Fig.2.2). These enzymes come principally from the animal itself in the case of carnivores and some exudivores (animalsthat feed on some form of plant exudate), but in omnivores and herbivores digestive enzymes come from both the animal and mutualistic microorganisms (bacteria, protozoa and sometimes anaerobic fungi) in the gastrointestinal tract. Stevens and Hume (1998) emphasized the ubiquitous role that gut microbes play in the conversion and conservation of nutrients in all vertebrates. Mean retention times are determined by factors such as particle sizes of solid digesta and thus the effectiveness of mastication, level of food intake, and gastrointestinal tract capacity and morphology (Stevens and Hume, 1995), as well as the suite of hydrolytic enzymes in the small intestine, nutrient absorption rates, and the location and size of the microbial population in the hind gut (cecum and proximal colon) and/or foregut (forestomach) of herbivores. The relationship between digestive efficiency and MRT depicted in the Sibly model (Fig.2.1) is well illustrated by the study of Hilton et al., (2000a)in which eight North Atlantic seabird species were offered two fish species commonly found in the diet of wild seabirds. There were small but statistically significant differences in digestive efficiency among the seabirds. True metabolizable energy coefficients (Miller and Reinecke, 1984)varied between 75% in black-legged kittiwakes and 83%in northern fulmars on a common diet of sand eel, and between 77% and 84% in the same two species on a common diet of whiting. Correlation was positive between digestive efficiency and mean retention time, which Hilton et al. (2000a)suggested represented a trade-off between the conflicting benefits of maximizing digestive efficiency and minimizing time spent feeding and digesting food. Retention time of digesta in the stomach was the most important component of total tract MRT and was greater in species with larger stomachs. Mean retention time of digesta depends not only on gut capacity but also on the rate of digesta flow, so that MRT (h)is proportional to digesta volume (mL) / digesta flow (mL/h-') (Karasov, 1996). The factor most strongly influencing digesta flow rate is food ingestion rate. Thus, in the absence of any change in gut capacity (digesta volume), an increase in food intake will reduce MRT and, in turn, digestive efficiency. However, intake of metabolizable energy may be maximized in the process. Thus there is an apparent trade-off between digesta MRT and digestive efficiency, and animals do not always attempt to maximize digestive efficiency. Hilton et al. (2000b)concluded that rapid (and thus less complete) digestion is likely to be favored when the energy costs of commuting between feeding and nesting sites are large and there is selective advantage in maximizing metabolizable energy intake. In contrast, slow (but more complete) digestion is preferable when
Concepts of digestive efficiency
51
commuting costs are small. In raptors, rapid digestion appears to be associated with a pursuit foraging mode, where the weight savings that can be achieved through rapid digestion (smaller gut capacity, smaller digesta load) exceed the costs in reduced digestive efficiency (Hilton et al., 1999). In contrast, slow digestion tends to be found in species with a searching foraging mode. These species may be able to exploit a wider range of food types, including lower quality prey, than raptors that actively pursue aerial prey. If juveniles are poorer foragers than adults and thus not able to always select high-quality foods (as predicted above from the Sibly model), they may be expected to maximize efficiencyby having slower passage rates and larger guts (Castro et al., 1989; Jackson, 1992; Barton and Houston, 1993). Particle size and mastication Oral processing is the first stage of digestion in most vertebrates. In most fishes, amphibians, reptiles, and birds, oral processing may result in only limited comminution of food, but in mammals the teeth act together with soft tissues to fracture food particles inside the mouth before swallowing (Lucas, 1994). The more finely the food is ground, the greater the surface area available for attack by digestive enzymes and bacteria. The efficiencywith which food is comminuted in the mouth of mammals depends malnly on the degree of tooth wear, and thus with age of the animal. For instance, Gipps and Sanson (1984) showed that as common ringtail possums (Pseudocheirus peregrinus) aged their teeth became more worn. As a result, the ratio of small particles (< 280 pm) to large particles (> 560 pm) of Eucalyptus leaf in their stomach decreased from 0.85 to 0.55, and the digestibility of dry matter decreased by 7% and that of cell walls by 46%. In koalas (Phascolarctos cinereus), both very young and very old animals were found to be less efficient in comminuting eucalypt leaves, indicating that some tooth wear is necessary in order to maintain maximal total length of cutting edges of the molars (Lanyon and Sanson, 1986). The implication is that digestive efficiency is likely to be greatest in young adult koalas (Hume, 1999). Level of food intake If digestive tract capacity remains constant, increased rates of food intake lead to shorter MRTs and, in turn, lowered digestive efficiency. The effect is likely to be greatest on low-quality (low-digestibility)foods. Thus, on a highfiber synthetic diet of 41% neutral-detergent fiber (NDF),common brushtail possums (Trichosurus vulpecula) eating 10 g dry matter.kg-o-75. dayldigested 88% of dry matter and 77% of NDF consumed, but those eating 50 gkg-0.75dayldigested only 71% dry matter and 42% NDF (Wellard and Hume, 1981). However, digestible energy intake remained constant, supporting the contention that animals tend to maximize the rate of extracting energy rather than digestive efficiency. On a low-fiber (17% NDF) diet there was no effect of feeding level on digestive efficiency. Similar responses were reported in ruminants fed early- and late-cut hays (Blaxter, 1962).
52
Physiological and ecological adaptations t o feeding in vertebrates
Digestive tract capacity Animals are often able to accommodate increased food loads if given time to acclimatize to the new condition. There appear to be two ways that animals meet increases in energy requirements and thus food intake: 1)more rapid digesta flow through a digestive tract of unchanged dimensions, which results in lower digestive efficiencies because ingested food is exposed to digestive processes for shorter periods; or 2) more rapid digesta flow through an enlarged digestive tract, resulting in similar turnover times and MRT, with no change in digestive efficiency. Karasov and Hurne (1997)concluded that for modest increases in energy requirements, the first alternative probably operates, with a small reduction in digestive efficiency but an increase in metabolizableenergy intake. The response of meadow voles with thyroxine implants eating 40% more than maintenance followed this strategy (Derting and Bouge, 1993). For larger increases in energy requirements, the second alternative appears to hold; more rapid digesta flow through a larger digestive tract so that MRT does not change and digestive efficiency remains the same, with a significant increase in the rate of metabolizable energy intake. The response patterns of mice (Toloza et al., 1991)and house wrens (Dykstraand Karasov, 1992)exposed to cold most closely fitted this strategy; digestive tract capacity increased, MRT did not change, and feeding rate more than doubled with no decrease in digestive efficiency. A special case of this latter response was reported in the garden warbler (Sylvia borin), a long-distancemigratory species. Bairlein (1985) and Hume and Biebach (1996) found that the 33% increase in daily food intake during the fattening period just prior to the autumnal migratory flight from Europe to Africa was associated not just with maintenance of energy metabolizability but with increased energy metabolizability. Hume and Biebach (1996) showed that MRT did not change, indicating that digestive tract capacity must have increased substantially to accommodate greater gut loading. Why the birds do not retain this peak digestive efficiency outside the premigration fattening period is a question that requires an answer (Biebach, 1996).It may be related to the costs of maintaining such a large mass of gut tissue; the gastrointestinal tract is responsible for a disproportionately high fraction of whole-body protein turnover and energy utilization (McBride and Kelly 1990).During the migratory fight itself there is atrophy of the digestive tract, which saves weight, reduces the maintenance cost of a tissue not being used, and partially fuels the flight. After a 48-hour period of starvation of fattened birds to partially simulate a migratory flight over desert, digestive tract tissue mass fell by 50% and that of the small intestine by 63% (Hume and Biebach, 1996). Note that the small intestine is by far the most active tissue of the whole tract; in alpine marmots (Marmota marmots), mitotic indices in the small intestine ranged from 40% to 60% during the active (feeding)season, but in the stomach, cecum, and colon mitotic indices never rose above 4%
Concepts of digestive efficiency
53
(Hume et al., 2002). Thus the disadvantage of maintaining peak digestive efficiency during long migratory flights is clear. It may be that at other times of the year when energy demands are lower, the costs of maintaining a large gut tissue mass outweigh the energetic benefits of maintaining digestive efficiency at a peak level. Another way an increased food load may come about is through a decrease in food quality. Increases in digestive tract capacity have been reported in a range of types and sizes of mammalian herbivores in response to seasonal or experimental increases in the plant cell wall (fiber) content of the diet (e.g. Gross et al., 1985; Green and Millar, 1987; Yahav and Choshniak, 1990; Nagy and Negus, 1993; Hammond and Wunder, 1995; Loeb et al., 1991; Bozinovic, 1995; Castle and Wunder, 1995). In each case, although digestive efficiency was depressed by the increased fiber content of the diet, higher food intakes meant that intakes of metabolizable energy either remained constant or declined only slightly. The greatest increases in volumetric capacity were in parts of the tract associated with microbial fermentation of plant cell walls, namely the cecum in the small hind-gut fermenters studied. Morphological features that retard digesta flow In addition to increased capacity of the digestive tract, numerous morphological features of the tract also serve to retard digesta flow. Often this affects one phase of the digesta more than another. For instance, the stomach of all vertebrates, as far as we know, selectively retards flow into the duodenum of large food particles relative to fluid, solutes, and small particles (Stevens and Hume, 1995). Selective retention of large food particles is enhanced in the stomach of foregut fermenting herbivores by compartmentalization of the forestomach into several chambers, and morphological specializations such as the reticulo-omasal orifice of ruminants and the haustrated "colon-like" forestomach of kangaroos (Hume, 1989).Large hindgut fermenting herbivores (colon fermenters) also selectively retain large particles, enhancing microbial breakdown in a large haustrated proximal colon. In contrast, many small hindgut fermenters, the cecum fermenters, feature selective retention not of the large particles but the small particles (including bacteria), along with fluid and solutes, in an enlarged cecum. This strategy enhances microbial breakdown of food in the cecum while concomitantly facilitating passage through the colon of large, less digestible food particles. This clears the digestive tract of indigestible bulk and allows much higher intakes of plant material than would otherwise be the case. As a result, cecum fermenters, most of which are below 10kg in body mass, are able to process plant material of much higher fiber content than would be predicted on the basis of body size and thus digestive tract size alone. The mechanisms responsible for selective retention of fluid, solutes and small particles in the cecum have been termed "colonic separation mechanisms" and are reviewed by Bjornhag (1987, 1994).
54
Physiological and ecological adaptations t o feeding i n vertebrates
Table 2.1. Digestive efficiency in four rodents on a common diet of commercial rabbit pellets containing 35% neutral-detergent fiber (Hume et al., 1993) Species
Colonic separation mechanism
Body mass (8)
Digestibility of dry matter ( "10
Yellow pine chipmunk Columbian ground squirrel
Yes No No
61 55 629
Hoary marmot
No
2,522
+ 1.0" + 3.0b + l.lb 50.3 + 1.2"
Townsend's vole
>
50.9 40.6 40.9
The interplay between digestive tract capacity and morphology on digestive efficiency is illustrated by the example in Table 2.1. The 25% greater digestive efficiency in the vole (Microtustownsendii)than a sciurid rodent (the yellow pine chipmunk, Eutarnias arnoenus ) of similar body size is due to the presence in the hindgut of all microtine rodents (voles and lemmings) of a colonic separation mechanism (CSM). In contrast, no sciurid rodent (the squirrels)has been found to have a CSM. A 10-fold increase in body size, and thus of absolute gut capacity in the Columbian ground squirrel (Sperrnophilus colurnbianus)did not compensate for the lack of a CSM in the sciurid digestive tract (digestive efficiency was still 25% greater in the vole), but a 40-fold increase (in the hoary marmot, Marrnota caligata) did. Thus greater gut capacity can compensate for the lack of a CSM-but only at the body sizes reached by marmots.
Hydrolytic enzymatic activity Most vertebrates can digest disaccharides in the small intestine, but with one notable, well-researched exception:American robins ( Turdus rnigratorius) and all other birds in the sturnid-muscicapid lineage (starlings, thrushes and Old World flycatchers) examined to date lack expression of the intestinal enzyme sucrase (Levey and Martinez del Rio, 2001).Consequently, those birds are unable to hydrolyze sucrose. This extreme example serves to illustrate how the digestibility of certain foods can be zero in some species but close to 100%in others. Within these extremes there are numerous examples of differences among vertebrate species in their capacity to digest fruit diets. However, the bases for the differences invariably seem not to lie in levels of hydrolytic enzyme activity or rates of nutrient uptake per unit of small intestine, but in increases in small intestinal length, mass, and surface area (McWilliams et al., 1999). Absorptive capacities Comparative aspects of the mechanisms of nutrient uptake in vertebrates were reviewed by Stevens and Hume (1995)and Karasov and Hume (1997). Absorptive capacity is the product of absorption rate per unit surface area of intestine (which depends on the transport properties of the intestinal tissue) and total surface area available for absorption. From data available at that
Concepts o f digestive efficiency
55
time, Karasov and Hume (1997)concluded that absorptive capacity among vertebrates was most strongly determined by differences in absorptive surface area rather than area-specific uptake rates. Within fish, reptiles, mammals, and birds there does not appear to be any dependence of nutrient uptake per unit surface area of intestine on body size or taxon. However, small intestinal nominal surface areas scale to body mass in all vertebrate groups to a common slope of 0.71, suggesting a match between absorptive area and food intake that scales in a similar fashion. Proportionality coefficients of nominal surface areas calculated by Karasov and Hume (1997) were 1.06,0.63,1.08,1.43,and 2.47 for fish, amphibians, reptiles, birds, and mammals respectively. Thus the higher absorptive capacities of mammals and birds than of fish, amphibians and reptiles can be explained largely by their longer small intestines (Stevens and Hume, 1995) and thus greater nominal surface areas. Total surface area for absorption includes the contribution of villi and microvilli as well, which increase the mucosal area of the intestine several orders of magnitude above the nominal area. So far, no pattern has emerged between villus or microvillus multiplication factors and taxon. On average, the villi increase the potential absorptive area of the mammalian small intestineby 6.7 times, and the microvilli by 53 times. However, nutrient uptake takes place mostly by epithelial cells closer to the villus tip, so actual absorptive surface areas are likely to be much less than potential uptake surface areas. The matches among the capacity for nutrient absorption and the capacity for chemicalbreakdown, digesta MRT, and overall digestive efficiency are of great interest. The principle of symmorphosis (Weibel et al., 1998)states that there should be close matches among these component parts of the system for delivery of nutrients to animal tissues. That is, there should not be gross overenpeering of any one component of the system, otherwise energy would be wasted in the maintenance of unused capacity and there would be a waste of physical space within the organism. However, with increased environmental stochasticitythe need for a higher than average capacity in one or more components of the delivery system might be expected in order to cope with peaks in nutrient supply. The existence of such "safety margins" (Diamond, 1998)will inevitably obscure symmorphosis. The size of any safety margin is likely to be related to a) the expected frequencyof peaks in nutrient supply, and b) the consequences of overload that is unprocessed. Digestive and/or absorptive capacity of the intestine can be exceeded, as seen in grain overload caused by fermentationof undigested grain in the hindgut of horses.
CONCLUSION Digestive efficiencyis a central concept in the nutritional ecology of animals, and digestibility (or digestion coefficient), the proportion of food not
56
Physiological and ecological adaptations t o feeding in vertebrates
appearing in the feces, is an important measurement of food quality and animal digestive performance.However, from theoretical models of the digestive tract and from accumulating evidence in the literature, it appears that animals only rarely seek to maximize digestibility. Instead, for most foragmg animals, the objective seems to be maximizing the rate of net energy gain. That is, it can be costly in terms of net energy gain to maximize digestibility. Also, because of the high maintenance costs of the gastrointestinal tract, particularly the small intestine, it may be an optimal strategy to maintain digestive efficiency at somethingbelow peak performance except at those times of the year when energy demands are especially high. Examples include the premigratory fattening period of some long-distance migratory birds, the short summer period for reproduction, growth and fattening of some rodents that hibernate at high altitudes and/or high latitudes, and the periods of intense cold faced by some high latitude nonmigratory avian species. That digestive efficiency on a given diet is not a fixed parameter attests to the highly flexible nature of the vertebrate digestive tract. Acknowledgment I thank Marcel Klaassen for his valuable insights into concepts of digestive efficiency and for discussions on how we might include the costs of maintaining the tissues of the gastrointestinaltract in future models of digestion.
REFERENCES Bairlein F. 1985. Efficiency of food utilization during fat deposition in the long-distance migrating garden warbler, Sylvia borin. Oecologia 68: 118-125. Barton N.W.H. and Houston D.C. 1993. The influence of gut morphology on digestion time in raptors. Comp. Biochem. Physiol. 105A: 571-578. Biebach H. 1996. Energetics of winter and migratory fattening. In: Avian Energetics and Nutritional Ecology. C. Carey (ed.). Chapman and Hall, New York, NY, pp. 280-323. Bjornhag G. 1987. Comparative aspects of digestion in the hindgut of mammals. The colonic separation mechanism (CSM) (a review). Dtsch. Tierarztl. Wschr. 94: 33-36. Bjornhag G. 1994. Adaptations of the large intestine allowing small animals to eat fibrous foods. In: The Digestive System in Mammals: Food, Form and Function. D.J. Chivers and P. Langer (eds.). Cambridge Univ. Press, Cambridge, UK, pp. 287-309. Blaxter K.L. 1962. The Energy Metabolism of Ruminants. Hutchinson, London, U.K. Bozinovic F. 1995. Nutritional energetics and digestive responses of an herbivorous rodent (Octodon degus) to different levels of dietary fiber. J. Mammal. 76: 627-637. Castle K.T. and Wunder B.A. 1995. Limits to food intake and fiber utilization in the prairie vole, Microtus ochrogaster: Effects of food quality and energy need. J. Comp. Physiol. B164: 609-617. Castro G., Stoyan N. and Myers J.P. 1989. Assimilation efficiency in birds: a function of taxon or food type? Comp. Biochem. Physiol. 92A: 271-278. Chivers D. J. and Langer, P. (eds.) 1994. The Digestive System in Mammals: Food, Form and Function. Cambridge Univ. Press, Cambridge, UK. Derting T.L. and Bouge B.A. 1993. Responses of the gut to moderate energy demands in a small herbivore (Microtus pennsylvanicus). J. Mammal. 74: 58-68.
Concepts of digestive efficiency
57
Diamond J. M. 1998. Evolution of biological safety factors: a cost/benefit analysis. In: Principles of Animal Design. E.R. Weibel, C.R. Taylor, and L. Bolis (eds.). Cambridge Univ. Press, Cambridge, UK, pp. 21-27. Dykstra C. R. and Karasov W.H. 1992. Changes in gut structure and function of house wrens (Troglodytes aedon) in response to increased energy demands. Physiol. Zool. 65: 422-442. Gipps J.M. and Sanson G.D. 1984. Mastication and digestion in Pseudocheirus. In: : Possums and Gliders. A.P. Smith and I.D. Hume (eds.). Australian Mammal Soc., Sydney, NSW, Australia, pp. 237-246. Green D.A. and Millar J.S. 1987. Changes in gut dimensions and capacity of Peromyscus maniculatus relative to diet quality and energy needs. Can. J. Zool. 65: 2159-2162. Gross J.E., Wang Z. and Wunder B.A. 1985. Effects of food quality and energy needs: changes i n gut morphology and capacity of Microtus ochrogaster. J. Mammal. 66: 661-667. Guglielmo C.G. and Karasov W.H. 1993. Endogenous mass and energy losses in ruffed grouse. Auk 110: 386-390. Hammond K.A. and Wunder B.A. 1995. Effect of cold temperatures on the morphology of gastrointestinal tracts of two microtine rodents. J. Mammal. 76: 232-239. Hilton G.M., Furness R.W. and Houston D.C. 2000a. A comparative study of digestion in North Atlantic seabirds. J. Avian Biol. 31: 3646. Hilton G.M., Furness R.W. and Houston D.C. 2000b. The effects of switching and mixing on digestion in seabirds. Funct. Ecol. 14: 145-154. Hilton G.M., Houston D.C., Barton N.W.H., Furness R.W. and Ruxton G . D . 1999. Ecological~constraintson digestive physiology in carnivorous and piscivorous birds. J. Exp. ZOO/.283: 365-376. Hume I.D. 1989. Optimal digestive strategies in mammalian herbivores. Physiol. Zool. 62: 1145-1163. Hume I.D. 1999. Marsupial Nutrition. Cambridge Univ. Press: Cambridge, U.K. Hume I.D. and Biebach H. 1996. Digestive tract function in the long-distance migratory garden warbler, Sylvia borin. J. Comp. Physiol. B166: 388-395. Hume I.D., Morgan K.R. and Kenagy G.J. 1993. Digesta retention and digestive performance in sciurid and microtine rodents: Effects of hindgut morphology and body size. Physiol. Zool. 66: 396411. Hume I.D., Beiglboeck C., Ruf T., Frey-Roos F., Bruns U. and Arnold W. 2002. Seasonal changes in morphology and function of the gastrointestinal tract of free-living alpine marmots (Marmota rnarmota). J. Comp. Physiol. B172: 197-207. Jackson S. 1992. Do seabird gut sizes and mean retention times reflect adaptation to diet and foraging method? Physiol. Zool. 65: 674-697. Karasov W.H. 1996. Digestive plasticity in avian energetics and feeding ecology. In: Avian Energetics and Nutritional Ecology. C. Carey (ed.). Chapman Hall, New York, NY, pp. 61-84. Karasov W.H. and Hume I.D. 1997. Vertebrate gastrointestinal system. In: Handbook of Physiology. Sec. 13: Comparative Physiology. W. H. Dantzler (ed.). Oxford Univ. Press, New York, NY, pp. 409-480. Kleiber M. 1961. The Fire of Life. Wiley, New York, NY. Lanyon J.M. and Sanson G.D. 1986. Koala (Phascolarctos cinereus) dentition and nutrition. 11. Implications of toothwear in nutrition. J. Zool. Lond. (A) 209: 169-181. Levey D.J. and Martinez del Rio C. 2001. It takes guts (and more) to eat fruit: lessons from avian nutritional ecology. Auk 118: 819-831. Loeb S.C., Schwab R.G. and Demment M.W. 1991. Responses of pocket gophers (Thomomys bottae) to changes in diet quality. Oecologia 86: 542-551. Lucas P.W. 1994. Categorisation of food items relevant to oral processing. In: The Digestive System in Mammals: Food, Form and Function. D.J. Chivers and P. Langer (eds.). Cambridge Univ. Press, Cambridge, LTK, pp.197-218.
58
Physiological and ecological adaptations t o feeding in vertebrates
Mason V.C. 1969. Some observations on the distribution and origin of nitrogen in sheep feces. J. Agric. Sci. 73: 99-111. McBride B.W. and Kelly J.M. 1990. Energy cost of absorption and metabolism in the ruminant gastrointestinal tract and liver: a review. J. Anim. Sci. 68: 2997-3010. McDonald P., Edwards R.A., Greenhalgh J.F.D. and Morgan C.A. 1995. Animal Nutrition, Longman: Harlow, Essex, UK, (5thed.). McWilliams S.R., Caviedes-Vidal E. and Karasov W.H. 1999. Digestive adjustments in cedar waxwings to high feeding rate. J. Exp. Zool. 283: 394407. Miller M.R. and Reinecke K.J. 1984. Proper expression of metabolizable energy in avian energetics. Condor 86: 396-400. Nagy T.R. and Negus N.C. 1993. Energy acquisition and allocation in male collared lemmings (Dicrostonyx groenlandicus): Effects of photoperiod, temperature and diet quality. Physiol. Zool. 66: 537-560. Penry D.L. and Jumars P.A. 1987. Modeling animal guts as chemical reactors. Amer. Nat. 129: 69-96. Raubenheimer D. 1995. Problems with ratio analysis in nutritional studies. Funct. Ecol. 9: 21-29. Sibly R.M. 1981. Strategies of digestion and defecation. In: Physiological Ecology: An Evolutionary Approach to Resource Use. C.R. Townsend and P. Calow (eds.). Sinauer, Sunderland, MA (USA), pp. 109-139. Slansky F. and Scriber J.H. 1985. Food consumption and utilization. In: Comprehensive lnsect Physiology, Biochemistry and Pharmacology, Vol. 4: Regulation: Digestion, Nutrition, Excretion. G.A. Kerkut and L.I. Gilbert (eds.). Pergamon: Oxford, U.K. pp.87-163. Sokal R.R. and Rohlf F.J. 1995. Biometry: T? Principles and Practice of Statistics in Biological Research, Freemans, New York, NY, (3 ed.). Speakman J.R. 1987. Apparent absorption efficiencies for redshank (Tringa totanus L.) and oystercatcher (Haematopus ostralegus L.): implications of optimal foraging models. Amer. Nat. 130: 677-691. Stevens C.E. and Hume I.D. 1995. Comparative Physiology of the Vertebrate Digestive System. Cambridge Univ. Press, Cambridge, UK, (2nded.). Stevens C.E. and Hume I.D. 1998. Contributions of microbes in vertebrate gastrointestinal tract to production and conservation of nutrients. Physiol. Rev. 78: 393427. Toloza E.M., Lam M. and Diamond J. 1991. Nutrient extraction by cold-exposed mice: a test of digestive safety margins. Amer. J. Physiol. 261: G6084620. Van Soes5P.J. 1994. Nutritional Ecology of the Ruminant. Cornell Univ. Press, Ithaca, IVY, (2 ed.). Weibel E.R., Taylor C.R. and Bolis L., (eds.) 1998. Principles of Animal Design. Cambridge Univ. Press, Cambridge, UK. Wellard G.A. and Hume I.D. 1981. Digestion and digesta passage in the brushtail possum, Trichosurus vulpecula (Kerr). Ausr. J. Zool. 29: 147-156. Yahav S. and Choshniak I. 1990. Response of the digestive tra to low quality dry food in the fat jird (Microtus crassus) and the levant vole (Microtus guentheri). J. Arid Environ. 19: 209-215.
Carbohydrate Hydrolysis and Absorption: Lessons from Modeling Digestive Function Todd J. McWhorter
Department of Wildlife Ecology University of Wisconsin, Madison, WI, USA
SYNOPSIS Nutrient assimilation is a complex phenomenon involving numerous enzymatic and transport pathways taking place in a variety of organs within the gastrointestinal tract (GIT). The structure and surface area of the GIT, factors affecting motility and thus the throughput of digesta, and the GIT's ability to chemically break down or ferment macromolecules and then absorb the resultant products affect an animal's ability to extract nutrients from its food. Digestive physiologists tend either to focus narrowly (working at the cellular, biochemical and molecular levels), or to analyze digestive performance at the whole-animal level (measuring digestive efficiency or retention time of food in the gut). As a consequence of this dichotomy, we have understood very little about how the fine details of nutrient digestion and uptake affect whole organism digestive efficiency and food intake rates until relatively recently. Since the pioneering work of Penry and Jumars and Martinez del Rio and Karasov, digestive physiologists have borrowed concepts from chemical reactor theory to model gut function. These models have provided the theoretical framework needed to integrate digestive processes with gut morphology and the chemical properties of food. This chapter reviews the application of chemical reactor theory to the study of digestion and digestive capacities in vertebrates, using recent studies of sucrose hydrolysis and hexose absorption in nectar- and fruit-eating birds as examples.
MODELING GUT FUNCTION: LESSONS FROM NECTAR- AND FRUIT-EATING BIRDS*
In the past two decades, models adopted from optimality theory (Sibly,1981) and chemical reactor theory (see for example Froment and Bischoff,1990; * See Table 3.1 for explanation of symbols and units
60
Physiological and ecological adaptations t o feeding in vertebrates
Levenspiel, 1998;Carberry, 2001) have provided the theoretical framework needed to integrate digestive processes with gut morphology and the chemical properties of food. Such models are useful because they: (1)clarify the relationships of gastrointestinal tract (GIT)attributes to one another, (2)identify those aspects of the GIT that determine the rate and efficiency of nutrient extraction from food, and (3)reduce the complexity of GIT systems within and between phyletic lines (Karasov and Hume, 1997). Models are important tools in mechanistic and comparativestudies of complex systems. They also provide a means for making inferences about whole-animal function from digestive processes, and are thus of ecological importance. Penry and Jumars (1986,1987)first recognized that gut function could be modeled using analogies to man-made chemical reactors. The task of the digestive physiologist parallels that of the chemical engineer evaluating the performance of reactor designs (gut functional morphologies)with the goal of maximizing yield (energy or nutrient assimilation, the optimizationcriterion or design objective),given a series of chemical reactions (digestivehydrolysis and nutrient uptake) taking place within these reactors (Martinez del Rio et al., 1994). Dade et al. (1990) and Martinez del Rio and Karasov (1990) extended the approach to include both hydrolysis and absorption and to predict the ingestion rate that maximizes net rate of absorptive gain to the animal. The first generation of chemical reactor models were "strategical" (sensu Levins, 1966),providing primarily qualitativepredictions, sacrificing precision to realism and generality (Martinez del Rio et al., 1994).This approach differed from previous "tactical" attempts (Levins, 1966)to model gut function, which emphasized precision, were complex and rich in detail but had limited generality. Newer generations of reactor models include significantly more physiological and ecological detail but the accompanying gain in precision comes at a loss of generality. The application of reactor theory to digestive processes has significantly broadened our perspective and allowed us to recognize two important things. First, that the tubular portions of the gut are not merely connectors, but reaction vessels in their own right. Second, that considering animal guts as cornbinationsof different kinds of reaction vessels (e.g.tubular, mixing, batch) raises important questions about the relative function of these vessels, how they interact, and what combination of vessels is optimal under different physiological and ecological conditions (see for example, Alexander, 1994). Reactor theory has allowed us to appreciate the significance of guts with widely ranging proportions of mixing to nonmixing compartments. The approach can be used to model alternate ways in which foods can be processed (Sibly, 1981; Penry and Jumars, 1987)and thus aid in our understanding of GIT evolution (Prop and Vulink, 1992; Karasov and Cork, 1996;Karasov and Hume, 1997). Chemical reactor theory is general and flexible: it permits the study of digestion in animals with diverse diets and digestive modes under a common paradigm (Martinez del Rio et al., 1994). It has been applied to
Modeling digestive function Table 3.1. Symbols and units Symbol Definition
Units
pmol-pL-' substrate or dietary sugar concentration pmol.pL-I glucose concentration per unit volume of digesta sucrose concentration per unit volume of digesta pmol.pL-I pmol-pL-I initial sucrose concentration per unit volume of digesta pmol.pL-' final sucrose concentration per unit volume of digesta min-' coefficient of intestinal passive permeability of hexoses pmol-pL-' Michaelis constant (K,) of intestinal uptake of glucose Michaelis constant (K,) of intestinal hydrolysis of sucrase pmol-pL-' reaction or uptake rate pmol.(pL.min)-' pmol.(pL.min)-' rate of glucose uptake in the intestine rate of hexose (glucose and fructose) uptake in the intestine pmol.(pL.min)-' rate of sucrose hydrolysis in the intestine pmol.(pL.min)-' intestinal throughput or retention time min optimal intestinal throughput or retention time min min intestinal throughput time that maximizes extraction efficiency pL-mini digesta flow rate in the intestine or food intake rate gut volume PL maximal rate (Vmax) of glucose uptake in the intestine pmol.(pL.min)-' pmol.(pL.min)-' maximal rate (Vmax) of hexose uptake in the intestine maximal rate (Vmax) of sucrose hydrolysis in the intestine pmol.(pL.min)-' reaction efficiencv
understanding the relationship between ingestion and downstream digestive processes in a wide variety of taxa, including marine invertebrates (e.g. Penry and Jumars, 1990;Plante et al., 1990),herbivorous mammals (e.g.Hwne, 1989),carnivores (Cochran, 1987),nectar- and fruit-eating birds (e.g.Martinez del Rio and Karasov, 1990; Lopez-Calleja et al., 1997; Levey and Martinez del Rio, 1999; McWhorter and Martinez del Rio, 2000), insectivorous birds (Dykstra and Karasov, 1992), and marine herbivorous fishes (Horn and Messer, 1992).Indeed, chemical reactor theory is so general and flexible that it is also widely applied by biologists, hydrologists, and geochemists in examining transport time scales (e.g. residence or retention time of water or solutes in bodies of water, see Monsen et al., 2002). The purpose of this chapter is not to provide a comprehensive review of the application of reactor engineering concepts to the study of digestion, nor to provide detailed theoretical derivations for reactor models. Interested readers are referred to the fine reviews by Martinez del Rio et al. (1994) and Karasov and Hume (1997) for the former, and to Penry and Jumars (1987), Dade et al. (1990), Martinez del Rio and Karasov (1990),Horn and Messer (1992),Jumars and Martinez del Rio (1999),and Jumars (2000a, 2000b) for the latter. In addition,
62
Physiological and ecological adaptations t o feeding in vertebrates
any chemical engineering textbook (e.g. Levenspiel, 1998) will provide a formal exposition of reactor performance analysis. Rather, the purposes of this chapter are to: I ) provide readers with a brief introduction to chemical reactor models and 2) review the recent and rapidly accelerating use of nectar- and fruit-eating birds as valuable test cases for the description of guts as chemical reactors and of optimization premises.
Guts as Chemical Reactors Building mathematical models of the digestive process using the reactor approach relies on three steps. (1)An optimization criterion (or design objective, Penry and Jumars, 1987)must be defined. Researchers have proposed that maximization of the rate of nutrient and/or energy extraction is an appropriate criterion, although many others are possible (e.g.extraction efficiency may be useful for examiningboth mechanistic and evolutionary questions; see Prop and Vulink, 1992;Karasov and Cork, 1996). The former will be a primary focus of my discussion, as it was the criterion adopted in most early models of gut function in nectar- and fruit-eating birds (e.g.Martinez del Rio and Karasov, 1990; Martinez del Rio et al., 1994).(2) Explicit analogies between gut morphologies and reactor designs must be established. Penry and Jumars (1987)identified three basic types of reactors analogous to GIT organs: batch reactors, continuous-flow stirred tank reactors (CSTR), and plug-flow reactors (PFR).Complex gut functional morphologies can be modeled by setting two or more reactors in series or parallel (Levenspiel, 1998). (3)Physiological measurements such as affinities and maximal rates of enzymes and transporters must be employed as parameters in performance equations to predict reactor configurations and/ or digestive behaviors that maximize the optimization criterion (Hume, 1989; Dade et al., 1990; Martinez del Rio et al., 1994; Karasov and Hume, 1997). Modeling guts as chemical reactors requires one to make, and hence eventually validate, explicit assumptions about the digestive process (see Martinez del Rio et al., 1994),but it also allows one to make crisp falsifiablepredictions about wholeorganism outcomes from measurements done in vitro at the organ and tissue level (Martinez del Rio and Karasov, 1990). The important interrelationships in digestive function that facilitate use of reactor theory are outlined below, followed by a description of the morphological analogies that have been established between guts and reactors. The physiological measurements employed as parameters in performance equations are discussed in the section Reactor Models Applied to Nectar- and Fruit-eating Birds. General features of digestion models Three distinct phases of digestion can be distinguished from a simple graphical model conceived by Sibly (1981)and refined by Karasov and Hume (1997) (Fig. 3.1). First, time and energy are invested in extraction (mechanicaland chemical breakdown and fermentation);next, there is a phase of rapid absorption, and third the rate of absorption declines as digestion is completed.
Modeling digestive function
63
Net energy or nutrient gain (defined as the amount absorbed minus the amount invested in digestion, Sibly, 1981)is a positive function of the time that digesta is retained in the GIT (the reaction chamber).For a given retention time, z, net gain is a positive function of reaction rate, r. For the GIT, r is equal to the rate of extraction and absorption (essentially the instantaneous slope of the curve f(z) in Fig. 3.1).The efficiency of any given reaction, X, is defined as the amount of product formed or nutrient absorbed relative to the initial amount of substrate present, C. Gut functional morphology affects digesta retention time, or the time required to process one reactor volume of digesta input (Penry and Jumars, 1987). Given that digesta flow, v,,, is constant, as the volume, V, of any region of the gut increases the retention time in that region also increases. As gut motility, and hence flow of digesta, increases, the retention time decreases given that V is constant.
I
Digestion corr~plete Slope = f(~)/a,or net rate of extraction
I
Ingestion Fig. 3.1. A simple graphical model of digestion. The net amount of energy or nutrient obtained (f(z), on the y-axis) is a positive function of retention time (7, on the x-axis). f(z) may decline with time during the initial phases of extraction (between the time of ingestion and absorption of breakdown products) but then increases rapidly during the phase of maximal absorption. As digestion nears completion, the rate of absorption declines. f(z) has the same form as digestibility (extraction efficiency) plotted as a function of retention time. The maximum net rate of extraction, f(z)/z, occurs at the retention time z* where a line through the origin is tangent to f(z)),while the efficiency of extraction is maximized at z" (adapted from Sibly, 1981; Karasov and Hume, 1997).
64
Physiological and ecological adaptations t o feeding i n vertebrates
These relationships can be integrated by a general model presented by Karasov and Hume (1997): ( X . C ) / r rn z rn V/v, (1) The relationships in Figure 3.1 and eqn. (1)apply to the digestion of whole meals, as well as to specific reactions (e.g. breakdown of disaccharides or uptake of hexoses) in the GIT or portions there of, and to models of different kinds and combinations of reactor types (below).
Morphologicalanalogies Modeling gut function is facilitated by establishing analogies between digestive organs and one or a combination of several reactor types. Chemical engineers recognize three basic types of reactors: batch reactors, continuous-flow stirred tank reactors (CSTR),and plug-flow reactors (PFR). The ideal forms of these reactors and their performance characteristics are briefly described here and examples of gut structures analogous to each reactor type are provided. Batch reactors process substrates (reactants) in discrete batches. Reaction periods alternate with idle periods during which the reactor is emptied of reaction products and unreacted components and reloaded. This interruption of material flow may result in a low overall extraction rate capacity unless the volume of the reactor is very large. The contents of an ideal batch reactor are perfectly mixed (i.e. spatially homogeneous at any given time); changes in reactant concentration occur only with respect to time. Figure 3.2a shows the change in concentration of products and reactants over time in an ideal batch reactor. Examples of batch reactors are found in coelenterates (Yonge, 1937), and in the blind compartments of vertebrate guts (e.g. ceca of herbivorous birds or the stomachs of carnivores, Cochran, 1987;Duke, 1989),assuming that meals are processed as separate batches. Continuous-flowstirred tank reactors are characterized by a continuous flow of reactants into and products out of a perfectly mixed reaction chamber. The composition of reactants within a CSTR operating at steady state are spatially homogeneous and constant with time. Incoming substrates are immediately diluted by recirculating materials upon entry, which reduces reaction rate, but efficiency can be high if the flow rate is low enough (i.e. retention time is long enough). Figure 3.2b shows that efficiency and rate of product formation are a function of flow rate through a CSTR. These conditions are thought to occur in the reticulorumen of the ruminant stomach, the camelid forestomach,and the sacciform region of the rat-kangaroo forestomach (Hume, 1989). These examples deviate from ideal CSTR conditions in that input is not continuous but outflow is modulated and probably more continuous. Plug-flow reactors consist of (usually) tubular reaction vessels through which there is a continuous orderly flow of material. Perfect radial mixing and negligible axial mixing (along the length of the reactor) are assumed;
Modeling digestive function
- Product
Contents mixed
Reactant md OUQU?3!
Holding time (b)
8
continuous
.H
I8'
Continuous
d 0
Low flaw rate
1P
2
output
High flow rate
Interior of reactor
&
\ I
0
I
Exit
_ - - _Low _ flow rate
."Y
D
High flow rate
Continuous input and output
_I.#--
#
','
/;
*I
t,
High flow rate Low flow rate
4IxIIlcornposition'uniform in cross-sectionat steady state, unchanging with time.
Exit Position along the reactor
0
Fig. 3.2. Ideal chemical reactors analogous to digestive organs. (a) In batch reactors the concentration of reactants and products changes over time. (b) In continuous-flow stirred tank reactors (CSTR), flow of reactants into and products out of the reactor maintains constant concentrations and reaction rates. (c) In plug-flow reactors (PFR) there is continuous flow and a steady state gradient in the concentration of reactants, products, and reaction rates along the length of the reactor (adapted from Penry and Jumars, 1987; Martinez del Rio et al., 1994; Karasov and Hume, 1997).
reactants are thus uniform in any given cross section and form a constant gradient along the length of the reactor during steady-state function. Figure 3 . 2 illustrates ~ efficiency and rate of product formation as a function of flow rate through an ideal PFR. Reactant concentration, and thus reaction rate, decline along the length of the vessel, although PFRs provide the highest rate of reaction in the minimum time and volume under most conditions (see Penry and Jumars, 1987). The small intestine of many vertebrates, assuming that they exhibit little axial mixing, and the simple tubular guts of deposit feeders probably function as PFRs. With few exceptions, the GI% of vertebrates are complex and contain elements of many reactor types whose function may deviate significantly
66
Physiological and ecological adaptations t o feeding in vertebrates
from the ideal conditions described above. Penry and Jumars (1986,1987), Martinez del Rio et al. (1994), Karasov and Hurne (1997) and Jumars (2000a, 2000b) extensively discuss reactor performance and the application of chemical reactor theory to the study of digestion in a wide variety of vertebrate and nonvertebrate taxa. Penry and Jumars (1987) and Horn and Messer (1992)provide analytical derivations for mass-balance equations for the ideal reactor types. The discussion herein is focused on the recent use of chemical reactor theory to model gut function in nectar- and fruit-eating birds.
Reactor Models Applied to Nectar- and Fruit-eating Birds The energy in nectar and the pulp of many fruits is primarily in the form of simple sugars (disaccharides and hexoses, Baker and Baker, 1983; Martinez del Rio et al., 1992), and nectarivorous and frugivorous animals tend to possess relatively simple guts. The digestive processes of animals feeding on nectars and sugary fruits (vs. waxy fruits, see Snow, 1981)may therefore be simpler to study than those of animals feeding on more complicated foods. Martinez del Rio and Karasov (1990) recognized the value of nectar- and fruit-eatingbirds as test cases for the description of guts as chemical reactors and of optimization premises. They extended the approach advocated by Penry and Jumars (1987)to include both hydrolysis and absorption and to predict the ingestion rate that maximizes net rate of absorptive gain to the animal. The two areas they addressed are: (1)how digestive constraints influence sugar and hence resource preferences and (2)how digestive traits and strategies are modified by the sugar composition and concentration of food. It has often been suggested that the chemical characteristics of the "rewards" (e.g. sugars) offered by plants reflect the dietary preferences of pollinators (Baker and Hurd, 1968;Howell, 1974; Calder, 1979),and ingestion rate is a functional response that connects individuals with populations and ecosystem processes (Karasov,1990;Jurnars and Martinez del Rio, 1999). Thus, studying the digestive strategiesof birds feeding on sugars can provide ecological and evolutionary insight. The theoretical framework provided by Martinez del Rio and Karasov (1990)has been employed by researchers at an accelerating pace during the past decade (Karasov and Cork, 1996; Downs, 1997; Lopez-Calleja et al., 1997; McWilliams and Karasov, 1998b; Levey and Martinez del Rio, 1999). The general conclusion from these empirical tests of the first generation of reactor models is that the results do not match predictions under the optimization premise of maximizing net rate of energy gain (Jumars and Martinez del Rio, 1999).The second generation of chemical reactor models, applied outside the context of maximization of net rate of energy or nutrient gain to estimate digestive capacities in nectar-eating birds, have yielded some remarkably accurate predictions. The digestive limitations identified by these models are informing recent tests of the sugar composition and
Modeling digestive function
67
concentrationpreferences of nectar-eatingbirds (e.g.Schondube and Martinez del Rio, 2003; Fleming et al., 2004.)and community level studies of resource use (Schondube,2003). In this section Martinez del Rio and Karasov's (1990) models are described:the morphological analogies drawn, the physiological parameters and performance equations used, and the resultant predictions. Mismatches between the predictions of these models and empirical results are then briefly described, focusing on how digestive traits and strategies are modified by the sugar composition and concentration of food and foraging costs. This is followed by a discussion of the second generation of chemical reactor models, their assumptions, and how the accuracy of their predictions differsbetween hummingbirds and nectar-eating passerine birds. The section concludes with suggestions for future directions and a discussion of the potential for chemical reactor models to test the ecological and evolutionary consequences of digestive constraints under natural conditions. Although Martinez del Rio and Karasov (1990) make explicit predictions about the diet preferences of nectar- and fruit-eating birds, the matches between these predictions and empirical results are not discussed. The sugar composition and concentrationpreferences of nectar- and fruit-eatingbirds seem to be the result of a complex interaction of digestive and osmoregulatory constraints on the part of the animal, and plant nectar characteristics;interested readers are referred to Martinez del Rio and Karasov (1990),Martinez del Rio (1990b), Martinez del Rio et al. (2001),Nicolson (2002),and Schondube and Martinez del Rio (2003). Modeling from the bottom up: tractability from digestive simplicity Martinez del Rio and Karasov (1990)modeled the guts of nectar- and fruiteating birds as PFRs because the digestion and absorption of sugars takes place entirely within a relatively simple tubular intestine. They included parameters describing both carrier-mediated and diffusive absorption of hexose sugars in their performance equations. Carrier-mediated absorption (active transport for glucose, facilitated diffusion for fructose, Karasov and Diamond, 1983; Alpers, 1987) is described by Michaelis-Menten kinetics with parameters of maximal absorption rate (Vp)and Michaelis constant (k 8' the sugar concentration at which absorption is equal to Vg/2, which is reciprocally related to the affinity of the carrier system for the sugar). Simple diffusive absorption is described as the product of luminal glucose concentration (G) and a permeability coefficient (k,). The rate of intestinal absorption of a single hexose such as glucose can be modeled as: rg= (G.Vg)/(kg + G) + (k,.G) (2) with units of p o l (pL&)-I, p o l pL-Iand min-lfor Vg,kg and k, respectively. In this case, digestion can be modeled simply as the rate of absorption: d G / d ~= -up (3)
68
Physiological and ecological adaptations t o feeding i n vertebrates
where z is a time constant (in units of min) equal to the ratio of the volume of the intestine to the rate of entrance of food into the intestine (v,).In an ideal PFR, equals the mean retention time of the reactor (also called throughput time, see above and Martinez del Rio and Karasov, 1990; Levenspiel, 1998). When Gis initially high, absorption rate is maximal (equalto Vg+ k,.G if G >> $). As glucose is absorbed, the lumenal concentration, and hence absorption rate drops, at an accelerating rate once concentration is in the range of kg. These kinetic features determine the form of the relationship between glucose (or energy) absorbed as a function of z (Fig. 3.1) and can be seen as boundary conditi&s in an optimization model wherein the control variable is z. ~ h e i that maximizes the net rate of energy intake (z*)is that at which a straight
a, V)
2 0
3 V)
u
s (CI
h
-s (b) a,
x
(CI
w
Q 3
Concentration of hexoses Fig. 3.3. Predicted absorption kinetics as a function of hexose concentration for Martinez del Rio and Karasov's (1990) (a) "hummingbird" and (b) "frugivore" models. Curve C represents the absorption of hexoses resulting exclusively from carrier-mediated processes. Curve C + P shows the sum of carrier-mediated uptake and passive diffusion of hexoses. Rates of sucrose hydrolysis (rs) are shown as lines parallel to the concentration axes for scenarios in which sucrase activity is limiting (2Vs < V,) and hexose uptake is limiting (2Vs> V,). Predictions were based on kinetic parameters resembling those of a hummingbird and a representative frugivore (adapted from Martinez del Rio and Karasov, 1990).
Modeling digestive function
69
line passing through the origin is tangent to the absorption curve. Modeling the digestion and absorption of the disaccharide sucrose, which must be hydrolyzed prior to the absorption of its hexose components glucose and fructose, requires an additional equation for the hydrolysis step (seeequations 3 and 4 in Martinez del Rio and Karasov, 1990).The overall hexose absorption curve in the case of a PFR with sucrose as the feed therefore depends on the relative rates of sucrose hydrolysis and hexose absorption (see Fig. 3.3b and Martinez del Rio and Karasov, 1990). Martinez del Rio and Karasov (1990) analyzed the behavior of two hypothetical systems: one in which all intestinaltransport is carrier-mediated, and a second which includes a passive component to absorption. Hummingbirds were thought to show no passive absorption of glucose (Diamondet al., 1986),while the frugivorous birds studied to date did absorb glucose passively (Karasovand Levey 1990);hence the systems were dubbed the "hummingbird model" and the "frugivore model", respectively. The authors assumed that sucrose was the primary dietary substrate for hummingbirds and considered both sucrose and hexose meals for frugivores. Figure 3.3 summarizes the main difference between these models. In the hummingbird model, the rate of hexose absorption (r,, includingboth glucose and fructose absorption)tends asymptotically to twice the maximum rate of sucrose hydrolysis (2VJ if sucrose hydrolysis is the rate-limitingstep (i.e.2Vs < V,, the maximal rate of hexose absorption)and to Vhif absorption of hexoses is the rate-limiting step (2Vs> V,), provided that the initial concentration of sucrose is higher than the Michaelis constant of sucrase (probably valid for hummingbirds, see Martinez del Rio and Karasov, 1990; Fig. 3.3a). The rate of hexose absorption from sucrose is therefore constant and approximately equal to the value of the rate-limiting step (see Fig. l b and Appendix A in Martinez del Rio and Karasov, 1990). In the frugivore model with hexose meals, the rate of hexose absorption depends on the lumenal concentration of hexoses (see example for glucose above). The maximal rate of absorption for a frugivore feeding on sucrose is equal to 2Vs (Fig. 3.3b). In this case sucrose hydrolysis is the rate-limiting step unless the Vq/V, ratio is extremely high. Because of the high sugar concentrations found in most fruit pulps and nectars (Lee et al., 1970; Levey, 1987), it is more likely for a bird with substantial passive absorption of hexoses to be limited by sucrose hydrolysis than a bird that relies only on carrier-mediated uptake (Martinezdel Rio and Karasov, 1990). Martinez del Rio and Karasov (1990)explored optimal digestion strategies for animals feeding on sucrose or hexoses and possessing intestines with the characteristics described above for the hummingbird and frugivore models. Their optimization criterion was maximization of the rate of net energy gain. These authors and Martinez del Rio et al. (1994) discuss the mechanics of including foraging, food processing, and maintenance costs in optimal digestion cost-benefit analyses, so these subjects are not discussed
70
Physiological and ecological adaptations t o feeding in vertebrates
here. In the hummingbird model absorption curves are approximately linear except at very low sugar concentrations (see Fig. 2 in Martinez del Rio and Karasov, 1990).The optimal value of z for the hummingbird model is that at which almost all sugar is absorbed; thus hummingbirds should exhibit nearly 100%efficiency in digestion of sugars. The value of z* is always a linearly increasing function of the initial sugar concentration, which implies that throughput time should increase with increasing nectar sugar concentration and animals should prefer energy concentrated foods. Given a choice between foods with equicaloric concentrations of sucrose and hexoses (the energy per mole of the former being twice that of the latter),energy-maximizing birds should prefer hexoses when sucrose hydrolysis is limiting (2Vs< Vh) and be indifferent if hexose absorptionis limiting (2Vs> V,). As a consequence of the linearity of the cost-benefit curve in the hummingbird model, T*, and thus the time invested in digesting a meal, is independent of the cost of obtaining it. In the frugivore model when sucrose is the main food and 2Vs < V,, the absorption curves are approximately linear, T* is a linearly increasing function of sucrose concentration in food, and optimal sucrose digestibility is near 100%. As in the hummingbird model, time invested in digestion is independent of cost of acquisition. When hexoses are the main sugar in food, the maximal rate of absorption occurs at the beginrung of the absorptive process and decreases thereafter (see above and Fig. 3.1). In this case, both the concentration of hexose in food and the cost of feeding influence T*. Increasing the cost of feeding increases T*:food that costs more to acquire should be retained longer and digested more thoroughly. Increasing the concentration of hexoses in food decreases z*:energy-rich foods should be retained for a shorter time and digested less thoroughly. Optimal energy intake rate increaseswith sugar concentration as in the hummingbird model and animals should thus prefer concentrated over dilute nectars. Martinez del Rio and Karasov (1990)predicted that birds with an important passive component to nutrient absorption and feeding on sucrose are more likely to be limited by sucrose hydrolysis, and thus should prefer hexoses to equicaloric sucrose. They predicted that this preference should increase with increasing energy density in food. Mismatch between predictions about digestive strategies and empirical data Martinez del Rio and Karasov (1990) used explicit assumptions about digestive processes and invitro measurements done at the organ and tissue level to generate predictions about whole-organism digestive (and thus behavioral) outcomes. How well did their first generation models and predictions about digestive strategies conform to reality? The general conclusion drawn from early empirical tests of reactor-based digestion models in nectar- and fruit-eating birds is that the results do not match predictions
Modeling digestive function
71
under the optimization premise of maximizing net rate of energy gain (Jumars and Martinez del Rio, 1999; Levey and Martinez del Rio, 2001). The mismatches between predictions and empirical results and the lessons learned from early applications of guts-as-reactor models to hexose- and sucrose-eatingbirds are briefly described in this section. Karasov (1999)also summarized some of these tests and provides a useful discussion of methodological details. Tests of optimal digestion models in sucrose- and hexose-eating birds have involved manipulation of food quality (sugar concentration,Karasov and Cork, 1996;Downs, 1997; Lopez-Calleja et al., 1997;Witmer, 1998;Levey and Martinez del Rio, 1999)and energetic demands (ambient temperature and foraging costs, McWilliams and Karasov, 1998b). Three observations seem to apply broadly to the results of food-quality manipulation studies. First, increasing the sugar concentration in food leads to decreased food intake rates. Second, assimilationefficiency is very high (>90%)and appears to be independent of sugar concentration in food. Third, net assimilation rate is relatively constant in spite of variation in sugar concentration in food (Karasov and Cork, 1996; Downs, 1997; Witmer, 1998). In other words, gut retention time either remains constant or increases with increment in sugar concentration (Levey and Martinez del Rio, 1999). In addition, tests of the predictions that extraction efficiency and retention time should increase with foraging costs (e.g.longer intermeal intervals)have also failed to support the model (McWilliamsand Karasov, 1998a, 1998b). Taken together, these results call into question the optimization premise and/or the physiological assumptions of the model (Jumarsand Martinez del Rio, 1999; Levey and Martinez del Rio, 2001). One possibility is that the optimizationpremise is incorrect. Nectar- and fruit-eating birds may not maximize their net rate of energy assimilationbut rather maintain constant rates of energy intake or minimize feeding time by maximizing digestive efficiency. Lopez-Calleja et al. (1997)investigated the effect of varying sucrose concentration on feeding patterns, gut function, and energy management in hummingbirds. They found that hummingbirds exhibited almost complete assimilation of sugars and increased meal retention times and intermeal intervals with increased sugar concentration, in agreement with the predictions of Martinez del Rio and Karasov's (1990) model. However, hummingbirds showed no significant differences in daily energy intake when fed different sugar concentrations. This constant daily rate of energy intake can seemingly be interpreted as two alternatives:(1)at the scale of a day, hummingbirds were not acting as "energy maximizers" (sensu Schoener, 1971),or (2)energy intake was constrained by maximal gut processing rates (Karasov et al., 1986; Levey and Martinez del Rio, 1999). The feeding patterns and energy intake rates observed by Lopez-Calleja et al. (1997) in hummingbirds indicated that they were not using their guts to maximal capacity. The authors concluded that energy maximization is
72
Physiological and ecological adaptations to feeding in vertebrates
probably an inappropriate assumption for birds that are not growing, storing fat, or reproducing (see also Karasov and Cork, 1996; Karasov, 1999),and proposed an alternativemodel (or rather an alternative optimizationpremise) which assumes that birds which vary their food intake maintain constant energy intake. This model also predicts complete sugar assimilation and increased retention time with increasing sugar concentration,but in contrast to Martinez del Rio and Karasov's (1990) model, it predicts that as sugar concentration increases the gut spends increasing amounts of time idle. McWhorter and Martinez del Rio (2000)and Martinez del Rio et al. (2001) developed and tested models which differentiatebetween constant energy intake and digestive constraints to energy assimilation (see below: Second Generation Reactor Models: Gut Function in Nectar-eating Birds). The hypothesis that birds minimize feeding time by maximizing digestive efficiency is not mutually exclusive to these alternatives. They may reduce their need to forage (andthus expend energy and risk predation)by thoroughly digesting everything they consume (Levey and Martinez del Rio, 2001). McWilliams and Karasov (1998b) found that cold-acclimated frugivorous cedar waxwings (Bombycilla cedrorum) behaved as time minimizers while maintaining maximal glucose extraction efficiency. A second, and by no means mutually exclusive possibility, is that one or more of the physiological assumptions of the model are incorrect. These assumptions include descriptions of food acquisition and processing costs, constancy of reactant volumes, and reaction kinetics or mixing (Jumarsand Martinez del Rio, 1999). The reaction kinetics assumed by Martinez del Rio and Karasov's (1990)model for frugivorous birds feeding on hexoses appears to be correct: a substantial body of research supports the notion that there is significant passive, presumably paracellular, absorption of glucose in passerine birds (reviewedby Afik et al., 1997). It remains to be seen whether absorption of hexoses by hummingbirds is entirely carrier mediated as previously thought (Diamond et al., 1986);recent evidence suggests that this may not be the case (see Paracellular Intestinal Absorption of Carbohydrates in Mannals and Birds in chapter 5by McWhorter, present volume). Preingestional food processing costs that rise with sugar concentration (e.g.increasing nectar viscosity; see Kingsolver and Daniel, 1983)may overturn predictions about feeding rates but have not been included in models to date, which instead assume fixed costs per time that are proportional to volumetric ingestion rates (Jumarsand Martinez del Rio, 1999). Functional response models (see Jeschke et al., 2002) may provide a useful framework for linking digestive and preingestional limitations on intake rate. The assumption of constant gut volume may be violated by water-balance constraints: birds feeding on solutions of hexoses at concentrations that exceed the osmolality of plasma dilute incoming food with intestinal secretions (Chang and Rao, 1994), and some nectar-eating passerines may modulate water absorption across the gut more or less depending on food sugar concentration
Modeling digestive function
73
(McWhorter et al., 2003). A crucial assumption of the optimal digestion model in frugivores is that animals can modulate assimilation efficiency. However, the consistently high sugar assimilation efficiencies exhibited by these birds casts doubt on this assumption. Indeed, placing a high concentration of undigested, osmotically active substrates (i.e. sugars) in the lower gut may cause osmotic diarrhea and impair ability to reabsorb water and electrolytes (Malcarneyet al., 1994;Levey and Martinez del Rio, 1999).Lastly, Levey and Martinez del Rio (1999) found evidence of significant axial (i.e. longitudinal) mixing in the guts of cedar waxwings, invalidating the simplifying assumption of Martinez del Rio and Karasov (1990) that no axial mixing takes place. A theoretical framework is useful to the extent that it spawns empirical tests of its predictions and facilitates interpretation of experimental results. The first generation of chemical reactor models applied to nectar- and fruiteatingbirds were fertile at generating empirical work, but this work led to the conclusion that the optimization premise and/or physiological assumptions of the models were incorrect. Optimization studies often cycle between the process of model building and empirical performance tests (Seger and Stubblefield, 1996);modeling gut function in nectar-and fruit-eating birds is certainly an example of the cyclical nature of this process. Levey and Martinez del Rio (1999) proposed alternative models of gut function in frugivores, modifying assumptions from Martinez del Rio and Karasov (1990) for a better match between predictions and empirical data. McWhorter and Martinez del Rio (2000),Martinez del Rio et al. (2001),and McWhorter (2002) tested several of those alternatives in nectar-eating birds. This second generation of chemical reactor models is discussed in the following section.
Second Generation Reactor Models: Gut Function in Nectareating Birds Nectar-eating birds generally respond to experimentally increased sugar concentration in food by decreasing their food intake rates (Collins, 1981; Downs, 1997; L6pez-Calleja et al., 1997;Lotz and Nicolson, 1999; McWhorter and Lopez-Calleja, 2000; McWhorter and Martinez del Rio, 2000; Martinez del Rio et al., 2001; McWhorter et al., 2003).Similar reciprocal relationships have been described for a wide variety of species (Montgomery and Baumgardt, 1965; Batzli and Cole, 1979; Simpson et al., 1989; Nagy and Negus, 1993; Castle and Wunder, 1995) and have often been attributed to compensatory feeding (Simpsonet al., 1989).According to this explanation, animals regulate food intake to maintain a constant flux of assimilated energy or nutrients (Montgomeryand Baumgardt, 1965;Slansky and Wheeler, 1992). If the energy of nutrient density of food is decreased, animals compensate by increasing intake. Indeed, this inverse relationship often leads to relatively constant energy intake by nectar-eating birds (e.g. Beuchat et al., 1979; Lopez-Calleja et al., 1997; Lotz and Nicolson, 1999; McWhorter and
74
Physiological and ecological adaptations t o feeding in vertebrates
Martinez del Rio, 2000). An alternative explanation often proposed is that intake may be constrained by the ability of animals to process the nutrients contained in food (see above and Karasov et al., 1986;Levey and Martinez del Rio, 1999).Two complementary approaches may be used to differentiate compensatory feeding from physiological constraint. The first is descriptive and relies on examining the functional structure of the intake response, in nectar-eating birds usually a power function of the form v, = (4) where v, is food intake rate, C equals sugar concentration, and a and b are empirically derived constants (McWhorterand Martinez del Rio, 1999,2000). Because v, decreasesas a power function of C, the amount of energy ingested is also a power function of C. Animals exhibiting values of b equal to 1show "perfect" compensation for the energy density of food, and thus sugar intake that is independent of concentration. This approach provides only inferential evidence of constraintson intake, that is, a slope (b) less than one implies but does not demonstrate constraint. Martinez del Rio et al. (2001) discuss this approach in detail and provide a general review of the intake responses of nectar-eating birds. The second approach, described below, is experimental and relies on determining the effect of changing energetic demands on the intake response. The chemical reactor models described in this section predict how capacity for sucrose digestion, and thus food intake, changes with food sucrose concentration.These models provide a mechanistic bridge between gut function and feeding behavior. They allow one to determine whether the intake response observed in sucrose-eating nectarivorous birds is the result of constraints imposed by digestive processes or the result of compensatory feeding. In this section the experimental approach to differentiating between compensatory feeding and constraint is outlined, followed by a brief description of the digestive capacity model developed by McWhorter and Martinez del Rio (2000).Explicit predictions made by Martinez del Rio et al. (2001), based on this model, about how compensatory feeding and physiological constraint shape the intake responses of nectar-eating birds are then discussed. Lastly, tests of the model made by McWhorter (2002)in nectar-eatingpasserine sunbirds and Schondube (2003)in a community of nectar-eating birds are described. Apparent mismatches in the accuracy of the predictions of this model between hummingbirds and passerine birds are pointed out. McWhorter and Martinez del Rio (2000)addressed the question of whether the intake-responserelationship observed in hummingbirds is the result of compensatory feeding or a digestive constraint to energy assimilation.They exposed broad-tailedhummingbirds (Selasphorus platycercus) to ambient temperatures of 22°C and 10°Cand fed them diets ranging in sucrose concentration from 292 to 1,168mM. Because chronic cold exposure in endotherms is
Modeling digestive function
75
often accompanied by increased digestive and metabolic capacities (Konarzewski and Diamond, 1994 and references therein), hummingbirds were acutely exposed to the lower temperature. The resting metabolic rate of broad-tailed hummingbirds is considerably higher at 10°C than at 20°C (Bucher and Chappell, 1988).Based on this observation, McWhorter and Martinez del Rio (2000)hypothesized that for a given food energy density birds exposed to lower temperatures would show increased food intake. An increase in sugar intake under energetically demanding conditions would support the compensatory feeding explanation. Conversely, the opposite effect would provide evidence that a physiological process limits sugar assimilation. Although the birds exhibited the expected intake-responserelationship, they did not significantly increase food consumption when exposed to the lower ambient temperature (McWhorter and Martinez del Rio, 2000). In addition, at 10°C birds lost significantly more body mass, were often observed emerging from torpor in the morning, and exhibited other behavior commonly associated with energy conservation (ptiloerection,decreased flying time, feet held close to the body in flight, Gass and Montgomerie, 1981; Udvardy, 1983). Regardless of any energy conserving mechanisms employed, it appeared that acutely cold-exposed hummingbirds could not assimilate energy fast enough to compensate for their higher energy demands. These observationswere interpreted as evidence of a physiologcal constraint to energy assimilation. The apparent inability of hummingbirds to increase energy assimilation when acutely subjected to higher energetic demands led McWhorter and Martinez del Rio (2000) to speculate about the factors imposing an upper limit to food intake. Because hummingbirds obtain the vast majority of their energy from the sugars contained in sucrose-rich floral nectars, physiological processes that determine rates of sucrose assimilation were identified as potential limiting factors. Sucrose ingestion can be limited by rates of sucrose hydrolysis or transport of the resulting monosaccharides (Karasov et al., 1986;Martinez del Rio, 1990a),and by rates of sugar catabolism or biosynthetic processes (Suarez et al., 1988; Suarez et al., 1990). McWhorter and Martinez del Rio (2000)focused on the potential role of digestive processes in limiting energy assimilation and assumed that sucrose hydrolysis in the small intestine is the limiting step in sucrose assimilation (or that digestive and metabolic processes such as hydrolysis and uptake are matched to one another, Hamrnond and Diamond, 1997). They used in vitro data on enzyme (sucrase-isomaltase)activity and gut volume to predict maximum sucrose digestive capacity as a function of sugar concentration in food. They assumed that the guts of hummingbirds function as PFRs (see above and Penry and Jumars, 1987)and included significantlymore physiological detail than previous attempts to estimate hydrolytic capacity in animal guts (e.g. Diamond and Hammond, 1992). Most significantly their model takes into account the decline in sucrose concentration along the length of the gut that
76
Physiological and ecological adaptations t o feeding in vertebrates
accompanies hydrolysis. Previous analyses of guts as chemical reactors used models to predict throughput times that maximize the net rate of nutrient or energy assimilation (Dade et al., 1990;Martinez del Rio and Karasov, 1990; Jumars and Martinez del Rio, 1999). McWhorter and Martinez del Rio's (2000) model was designed outside this context of optimality. Following Levey and Martinez del Rio (1999),they assumed that hummingbirds must show high assimilation efficiencies to prevent osmotic imbalances in the lower gut, and used this physiological detail as a constraint on the model to estimate intake rates. McWhorter and Martinez del Rio (2000)made two critical assumptions in their model: (1)that digesta flows unidirectionally (Jumars and Martinez del Rio, 1999) and (2) that the rate at which sucrose is hydrolyzed in the intestine (rs)follows simple Michaelis-Menten kinetics: r5= Vs-S.(ks + S)-' (5) where Vsequals the maximum rate of hydrolysis along the intestine (in pmol min-I pLal),ks the Michaelis constant of sucrase (in pmol pL-I), and S the concentration of sucrose (inpnol pL-l)down the intestine or over time (Jumars and Martinez del Rio, 1999). Equation (5) can be integrated to yield the throughput time (7) required to reduce the initial sucrose concentration (So) to a given final value (Sf,based on assimilation efficiency): In plug flow reactors if one knows z and the volume of gut contents (Vin pL), intake rate (voin pL min-I)can be estimated as: v = V.z-'
The model predicted that the relationship between maximal food intake rate and sugar concentration should follow a power function with a slope (b) lower than one. The intake rates predicted for broad-tailed hummingbirds slightly overestimated observed intake rates (by a m a r p that increased from 15%at the lowest concentration to 35% at the highest) but there was a remarkable qualitative match between the model's output and bird behavior (Fig. 3.4a). The model also predicted that sugar assimilation rate should increase with increased concentration in food. This prediction was upheld in broad-tailed hummingbirds. Although there was no significant increase in sucrose intake between the two ambient temperatures,there was a signlficant effect of concentration on sucrose intake, a consequence of lower average hydrolysis rates at lower food concentrations (McWhorter and Martinez del Rio, 2000). Sucrose apparent assimilation efficiency was high (>95%)and independent of sugar concentration. Taken together, these observations provide compelling evidence for a physiological constraint to energy assimilation. Martinez del Rio et al. (2001)extended the approach of McWhorter and Martinez del Rio (2000)to make explicit predictions as to the conditions that
Modeling digestive function
Sucrose concentration (mM)
Fig. 3.4. (a) Food intake rate (v,) of broad-tailed hummingbirds at 22 "C (solid circles) and 10°C (open circles) decreased as a common power function of diet sugar concentration (v, = 1000.C-0.77, rZ = 0.87, dashed line). Hummingbirds were unable to increase food
consumption when abruptly exposed to low ambient temperature (and hence increased metabolic demands). Solid line represents predicted maximal food intake rate (vo= 653.CRedrawn from McWhorter and Martinez del Rio (2000). (b) Hypothetical relationship between daily food intake rate, sucrose concentration in food, and energy expenditures for magnificent hummingbirds. Thick line represents predicted maximal food intake rate Observed food based on McWhorter and Martinez del Rio's (2000) model (v, = 4030.C-0.82). intake rates for birds feeding under mild ambient conditions were lower than predicted maximal intake rates, but the exponent of this relationship was higher than that of the r2 = 0.98; data not shown). Other lines represent predicted relationship (v, = 7229~C-O.~~, hypothetical intake responses for three levels of daily energy expenditure (DEE). Note that at higher levels of energy expenditures, this hypothesis predicts "broken" intake responses described by two power functions with different slopes. Redrawn from Martinez del Rio et al. (2001). (c) Food intake rate of Palestine sunbirds at 30°C (open crosses), 15°C (solid circles), and 5°C (open circles) decreased as power functions of diet sugar concentration (v, = 966.C-076, r2 = 0.91 for 30°C, lower dashed line, and v, = 3296.C-0.92, 12 = 0.97 for 5 and 15"C, upper dashed line). Solid line represents predicted maximal food intake rate (v, = 1472.C-075). Redrawn from McWhorter (2002). Note that all axes are logarithmic and vary in scale among panels.
78
Physiological and ecological adaptations t o feeding in vertebrates
determine whether compensatory feeding or physiological constraint shape the intake response of nectar-eating birds. They applied McWhorter and Martinez del Rio's (2000) model to magnificent hummingbirds (Eugenes fulgens)and similarly found that the model's output and the actual intake response of birds were both well described by power functions with negative exponents. In this case, however, the slope of the predicted intake response (0.822)was significantly lower than that of the observed relationship (0.942 + 0.047 SD), suggesting that under the mild conditions of the experiment birds were exhibiting compensatory feeding (Martinez del Rio et al., 2001). The birds appeared to possess digestive "spare capacity" (Diamond, 1991). The digestive safety factors (defined as the ratio of capacity to load, Diamond and Hammond, 1992) for magnificent hummingbirds were modest and ranged from 1.09 to 1.26 from the lowest to the highest food sugar concentration. McWhorter and Martinez del Rio (2000)reported similarly modest safety factors for broad-tailed hummingbirds (averaging 1.2 0.2 SD). Although Martinez del Rio et al. (2001) did not expose magnificent hummingbirds to low ambient temperatures, they predicted that acutely increasing energy demands would reduce digestive safety margins and force birds to shift from an intake response attributable to compensatory feeding to one shaped by digestive constraints. Figure 3.4b shows the hypothetical effect of increasing energy demands on the intake response of magnificent hummingbirds.Note that at higher levels of energy expenditures this model predicts a "broken" intake response described by two power functionswith different slopes (Martinez del Rio et al., 2001). McWhorter (2002) examined the interplay between compensatory and constrained feeding in a nectar-eating passerine bird, the Palestine sunbird (Nectarinia osea). Sunbirds were fed on sucrose solutions of varying concentration and exposed to two ambient temperatures within their acclimatized range (15and 30°C),and acutely to one temperature well below this range (5°C).The intake responses of sunbirds were compared with maximal intake rates predicted using McWhorter and Martinez del Rio's (2000)model. As expected, surlbirds decreased their food intake rates in response to sugar concentration in relationships described by power functions with negative exponents (Fig. 3.4~). At 15°C and 30°C, they were able to compensate for differences in food energy density and increased metabolic demands. When exposed to a relatively sudden drop in ambient temperature (to 5°C) and hence to an acute increase in thermoregulatory and food-warming energy expenditures, surlbirds were unable to increase their rates of food and energy intake. Predicted maximal sucrose and food intake rates matched actual intake rates at low food sucrose concentrations (292 mM) and low )~ that under temperatures (5 and 15°C) very closely (Fig. 3 . 4 ~suggesting these conditions intestinal sucrose hydrolysis rates were near maximal. Although the intake response at 5°C was not statistically distinguishable from that at 15"C, data for 5°C considered separately suggest a broken intake
*
Modeling digestive function
79
response as predicted by Martinez del Rio et al. (2001). Unlike the broadtailed hummingbirds studied by McWhorter and Martinez del Rio (2000), Palestine sunbirds exposed to ambient temperatures below their acclimatized range and subjected to digestive limitations to energy assimilation gained body mass. They appeared to balance their energy budgets using behavioral reductions in energy expenditures (McWhorter,2002). Taken together the patterns of food and energy intake exhibited by Palestine sunbirds and modeling of their digestive capacities provide compelling evidence for a physiological constraint to food intake when they feed on dilute diets under energetically demanding conditions. Like the hummingbirds studied by McWhorter and Martinez del Rio (2000) and Martinez del Rio et al. (2001), sunbirds appear to operate with small digestive spare capacities.Safety factors for sunbirds feeding at 15OCwere modest and ranged from 1.05to 1.5from the lowest to the highest sucrose concentration.At 30°C, safety factors were slightly larger, ranging from 1.62 to 1.66. McWhorter and Martinez del Rio's (2000)model overestimated maximal intake rates for broad-tailed hummingbirds (between 10% and 35'10, depending on sugar concentration),but appeared to very accurately predict maximal intake rates for Palestine sunbirds. What might explain this difference? One possibility is differences in the mechanism of hexose absorption between hummingbirds and sunbirds. Hummingbirds are thought to absorb glucose entirelyby mediated uptake (Diamond et al., 1986)whereas passive, presumably paracellular absorption of glucose may be very significant in fruit- and nectar-eating passerine birds (see Afik et al., 1997 and Chapter 5 by McWhorter, this volume). McWhorter and Martinez del Rio (2000)did not include the kinetics of glucose and fructose absorption in their model and assumed that sucrose hydrolysis was the limiting step in digestion. However, they found more glucose and fructose than sucrose in the excreta of broad-tailed hummingbirds, suggesting that hexose uptake capacities in hummingbirds may be more limiting than sucrose hydrolysis. It is possible that the inclusion of hexose uptake in the model would yield lower predicted intake rates for hummingbirds. Unfortunately, the method most widely used to estimate intestinal uptake of hexoses in vitro, the intestinal everted sleeve (Karasov and Diamond, 1983),may cause serious damage to intestinal tissues and lead to large underestimates of in vivo uptake rates (Starck et al., 2000). Indeed, glucose uptake rates measured in vitro using everted sleeves in rufous hummingbirds were approximately four times lower than glucose absorption rates observed in vivo (see Karasov et al., 1986).Note that this also presents a technical challenge for modeling digestive capacities in frugivores feeding on hexose diets. The passive absorption of glucose has not been measured in any species of sunbird, but it is likely to be similar to that in other nectar-eating passerines. Martinez del Rio and Karasov (1990)predicted that sucrose hydrolysis should be more limiting for birds with significantpassive components to hexose absorption.
80
Physiological and ecological adaptations t o feeding in vertebrates
McWhorter and Martinez del Rio's (2000)assumption that sucrose hydrolysis is the limiting step may therefore be correct for nectarivorous passerines feeding on sucrose, but incorrect for hummingbirds.Indeed, Schondubeand Martinez del Rio (2003)recently found that nectar-eating passerine flowerpiercers (Diglossa baritula) were able to consume and assimilate 10%more of a hexose (glucoseand fructose) solution than an equicaloric sucrose solution. A second possibility is that the discrepancy in accuracy of the model's predictions is due to differences in the roles of the guts of hummingbirds and sunbirds in osmoregulation.McWhorter and Martinez del Rio (2000)assumed that the concentration of sucrose in the intestinal lumen changed simply as a result of hydrolysis. In reality, sucrose concentration changes with hydrolysis and the addition or removal of water by secretion and absorption into and from the intestinal lumen (Chang and Rao, 1994). McWhorter and Martinez del Rio (1999) showed that hummingbirds absorb essentially all ingested water across the intestine, but McWhorter et al. (2003)found that Palestine sunbirds shunt up to 60%of ingested water through their intestines. The model's simplifying assumptions about the decline in sucrose concentration and the constancy of digesta volume may therefore be more correct for sunbirds than for hummingbirds. Data on the concentration of solutes in the intestines of nectar-eating birds can help evaluate the validity of these assumptions (see Ferraris et al., 1990). Chemical reactor models of digestive capacity have the useful feature of generating precise predictions about the form of the intake response based on energy expendituresand the magnitude of physiological traits. They allow probing the relative importance of behavioral changes and physiological mechanisms for balancing the energy budgets of small endotherms over short timescales. The picture emerging from recent tests of these models is that physiologicalconstraints and compensatory feeding are complementary mechanisms shaping the behavioral responses of nectar-eating birds to varying food energy density (Martinez del Rio et al., 2001). Rapid-exposure experiments such as those employed by McWhorter and Martinez del Rio (2000)and McWhorter (2002))used within the framework of modeling, are informative because they reveal the immediate short-term digestive spare capacity of an animal. Small digestive safety factors may be a general trait of nectar-eating birds, which appear to have considerable flexibility in modulating their energy expenditureswhen faced with digestive constraints and/or increased thermoregulatory energy demands (Martinez del Rio et al., 2001; McWhorter, 2002).The approach described here may also be useful in determining how digestive safety factors change during growth, reproduction, and migration (McWhorterand Lopez-Calleja, 2000).Chapter 4 in this volume (Karasov and McWilliams) presents a general discussion of the adaptationof digestivecapacities to the amounts and types of food eaten, and the use of cold-exposuretests to measure digestive capacities. McWhorter and Lopez-Calleja (2000)review the factors that may impose ceilings on the energy budgets of hummingbirds during chronic cold exposure.
Modeling digestive function
81
Models of digestive capacity are also useful tools in community-level studies of resource use. Schondube (2003) used McWhorter and Martinez del Rio's (2000)approach to predict the maximal sucrose digestive capacities of a community of 10nectar-feedingbirds (7hummingbirds, 2 warblers, and 1 flowerpiercer). He compared the birds' predicted maximal rate of energy intake with their field metabolic rates (FNIRs) estimated from published allometric equations, and with their food intake rates. Predicted maximal energy intake rates for hummingbirds were either equal to or higher than their expected FMRs and exceeded observed food intake by 10 to 70%. Predicted maximal energy intake values for passerines were either equal to or lower than their predicted FMRs and digestion of sucrose seemed to limit ingestion rate. Knowledge of interspecific variation in digestive capacities can clearly lead to testable predictions about resource use on a community scale (but see caveats regarding the differences in digestive performance among hummingbirds and passerines above).
Chemical reactor models have provided the theoretical framework needed to integrate digestive processes with gut morphology and the chemical properties of food. Models force us to make o w assumptions explicit, provide falsifiable predictions, and locate areas where research is needed. They provide a means to make inferences about whole-animal function from digestive processes, and are thus of ecological importance. The approach can also be used to model alternate ways in which foods can be processed and thus augment our understanding of GIT evolution. The cyclical, iterative process of model building, empirical testing, and reformulation has been fruitful (see Levey and Martinez del Rio, 1999;Levey and Martinez del Rio, 2001). Although many of the predictions of early guts-as-reactor models have been proven false, much has been learned in the process. Secondgeneration models in nectar-eating birds, developed outside the context of optimization, have generated some remarkable predictions. Digestive limitations identified by these approaches are already informing recent tests of the sugar composition and concentration preferences of nectar-eating birds (e.g. Nicolson, 2002; Schondube and Martinez del Rio, 2003; Fleming et al., 2004) and community level studies of resource use (Schondube,2003). Future gut function models in frugivores will likely not emphasize energy intake maximization, but instead explore efficient gut designs that permit digestive processes to take place at the rate dictated by metabolic demands (Jumars, 2000b). These models may include design objectives such as optimizing levels of expression and distributions of enzymes and transporters along the gut (Levey and Martinez del Rio, 2001). Modeling digestive capacities in frugvores feeding on hexoses using McWhorter and Martinez del Rio's (2000)
82
Physiological and ecological adaptations t o feeding in vertebrates
approach described above will require overcoming technical problems in estimating the rate of hexose transport in the intestine. Functional response models (see Jeschke et al., 2002) are consistent with chemical reactor theory and may provide a useful framework for linking digestive and preingestional limitations on intake rate. Integrating pre- and postingestional constraints with foraging costs and predation risks may increase the value of digestive models for interpreting ecological and behavioral patterns (see Bednekoff and Houston, 1994). Nectar- and fruit-eating birds present an unparalleled opportunity to explore the interaction between gut and metabolic function, and to test the notion that digestive constraints have ecological and evolutionary consequences for animals (and plants) under natural conditions. The techniques necessary to test the predictions of the model presented by McWhorter and Martinez del Rio (2000)and refined by Martinez del Rio et al. (2001)in the field are readily available: daily energy expenditures can be measured using standard methods (Powers and Nagy, 1988; Tiebout and Nagy, 1991) and digestive capacities can be estimated from physiological measurements and estimates of sugar concentration and composition of floral nectars.
Acknowledgements Supported by NSF (IBN - 02 16709)to William H. Karasov. REFERENCES Afik D., McWilliams S. R. and Karasov W. H. 1997. A test for passive absorption of glucose in yellow-rumped warblers and its ecological implications. Physiol. Zool. 70: 370-377. Alexander Mc. R. 1994. Optimum gut structure for specified diets. In: The Digestive System in Mammals: Food, Form and Function. D. J . Chivers and P. Langer (eds.). Cambridge Univ. Press, Cambridge, pp. 54-62. Alpers D. H. 1987. Digestion and absorption of carbohydrates and proteins. In: Physiology of the Gastrointestinal Tract, vol. 2. L. R. Johnson (ed.) Raven Press, New York, NY, pp. 1469-1487. Baker H. G. and Baker I. 1983. Chemical constituents of nectar in relation to pollination mechanisms and phylogeny. In: Handbook of Experimental Pollination Biology, pp. 131-171. Baker H. G. and Hurd P. H. 1968. Intrafloral ecology. Annu. Rev. Enfomol. 13: 385414. Batzli G . 0. and Cole F. R. 1979. Nutritional ecology of microtine rodents: digestibility of forage. Mammal 60: 740-750. Bednekoff P. A. and Houston A. I. 1994. Avian daily foraging patterns: effects of digestive constraints and variability. Evol. Ecol. 8: 36-52. Beuchat C. A., Chaplin S. B. and Morton M. L. 1979. Ambient temperature and the daily energetics of two species of hummingbirds, Calypte anna and Selasphorus rufus. Physiol. Zool. 52: 280-295. Bucher T. L. and Chappell M. A. 1988. Energy metabolism and patterns of ventilation in euthermic and torpid hummingbirds. In: Physiology of Cold Adaptation in Birds C. Bech and R. E. Reinersten (eds.). Plenum Press, New York, NY, pp. 187-195.
Modeling digestive function
83
Calder W. A., 111. 1979. On the temperature-dependency of optimal nectar concentrations for birds. J. Theor. Biol. 78: 185-196. Carberry J. J. 2001. Chemical and Catalytic Reaction Engineering. Dover Publ. New York, NY. Castle K. T. and Wunder B. A. 1995. Limits to food intake and fiber utilization in the prairie vole, Microtus orchrogaster: effects of food quality and energy need. J. Comp. Physiol. B 164: 609-617. Chang E. B. and Rao M. C. 1994. Intestinal water and electrolyte transport. In: Physiology of the Gastrointestinal Tract L. R. Johnson (ed.). Raven Press, New York, NY, pp. 2027-2081. Cochran P. A. 1987. Optimal digestion in a batch-reactor gut: the analogy to partial prey consumption. Oikos 50: 268-270. Collins B. G. 1981.Nectar intake and water balance for two species of Australian honeyeater, Lichmera indistincta and Acanthorynchus superciliosis. Physiol. Zool. 54: 1-13. Dade W. B., Jumars P. A. and Penry D. L. 1990. Supply-side optimization: maximizing absorptive rates. In: Behavioral Mechanisms of Food Selection, R. N. Hughes (ed.). Springer, Berlin, pp. 531-556. Diamond J. 1991. Evolutionary design of intestinal nutrient absorption: enough but not too much. News Physiol. Sci. 6: 92-96. Diamond J. M. and Hammond K. A. 1992. The matches, achieved by natural selection, between biological capacities and their natural loads. Experientia 48: 551-557. Diamond J. M., Karasov W. H., Phan D. and Carpenter F. L. 1986. Digestive physiology is a determinant of foraging bout frequency in hummingbirds. Nature 320: 62-63. Downs C. T. 1997. Sugar digestion efficiencies of Gurney's sugarbirds, malachite sunbirds, and black sunbirds. Physiol. Zool. 70: 93-99. Duke G. E. 1989. Relationship of cecal and colonic motility to diet, habitat, and cecal anatomy in several avian species. J. Exp. Biol. Suppl. 3: 3847. Dykstra C. R. and Karasov W. H. 1992. Changes in gut structure and function of house wrens (Troglodytes aedon) in response to increased energy demands. Physiol. Zool. 65: 422-442. Ferraris R. P., Yasharpour S., Lloyd K. C. K., Mirzayan R. and Diamond J. M. 1990. Luminal glucose concentrations in the gut under normal conditions. Amer. J. Physiol. 259: G822-G837. Fleming P. A., Hartman Bakken B., Lotz C. N. and Nicolson S. W. 2004. Concentration and temperature effects on sugar intake and preferences in a sunbird and a hummingbird. Funcf. Ecol. 18: 223-232. Froment G. E. and Bischoff K. B. 1990. Chemical Reactor Analysis and Design. John Wiley & Sons, New York, NY, (2nded.). Gass C. L. and Montgomerie R. D . 1981. Hummingbird foraging behavior: decisionmaking and energy regulation. In: Foraging Behavior: Ecological, Ethological and Physiological Approaches. A. C. Kamil and T. D. Sargent (eds.). Garland STPM, New York, NY, pp. 159-194. Hammond K. A. and Diamond J. 1997. Maximal sustained energy budgets in humans and animals. Nature 386: 457-462. Horn M. H. and Messer K. S. 1992. Fish guts as chemical reactors: a model of the alimentary canals of marine herbivorous fishes. Marine Biol. 113: 527-535. Howell D . J . 1974. Bats and pollen: physiological aspects of the syndrome of chiropterophily. Comp. Biochem. Physiol. 48: 263-276. Hume I. D. 1989. Optimal digestive strategies in mammalian herbivores. Physiol. Zool. 62: 1145-1163. Jeschke J. M., Kopp M. and Tollrian R. 2002. Predator functional responses: discriminating between handling and digesting prey. Ecol. Monogr. 72: 95-112. Jumars P. A. 2000a. Animal guts as ideal chemical reactors: maximizing absorption rates. Amer. Nat. 155: 527-543.
84
Physiological and ecological adaptations t o feeding in vertebrates
Jumars P. A. 2000b. Animal guts as nonideal chemical reactors: partial mixing and axial variation in absorption kinetics. Amer. Nat. 155: 544-555. Jumars P. A. and Martinez del Rio C. 1999. The tau of continuous feeding on simple foods. Physiol. Biochem. Zool. 72: 633-641. Karasov W. H. 1990. Digestion in birds: chemical and physiological determinants and ecological implications. S fudies Avian Biol. 13: 391415. Karasov W. H. 1999. Optimal digestive responses to changing diet and foraging costs. In: Proc. 22nd Inf. Ornithological Congress, Durban. N . J. Adams and R. H. Slotow (eds.). BirdLife South Africa, Johannesburg, pp. 2247-2258. Karasov W. H. and Diamond J. M. 1983. A simple method for measuring intestinal solute uptake in vitro. J. Comp. Physiol. 152: 105-116. Karasov W. H. and Levey D. J. 1990. Digestive system trade-offs and adaptations of frugivorous passerine birds. Physiol. Zool. 63: 1248-1270. Karasov W. H. and Cork S. J. 1996. Test of a reactor-based digestion optimization model for nectar-eating rainbow lorikeets. Physiol. Zool. 69: 117-138. Karasov W. H. and Hume I. D. 1997. Vertebrate gastrointestinal system. In: Handbook of Comparative Physiology. W. Dantzler (ed.). Amer. Physio. Soc., Bethesda, MD, pp. 409-480. Karasov W. H., Phan D., Diamond J. M., and Carpenter F. L. 1986. Food passage and intestinal nutrient absorption in hummingbirds. Auk 103: 453464. Kingsolver J. and Daniel T. 1983. Mechanical determinants of feeding strategy in hummingbirds: energetics, tongue morphology, and licking behavior. Oecologia 60: 214-226. Konarzewski M. and Diamond J. 1994. Peak sustained metabolic rate and its individual variation in cold-stressed mice. Physiol. Zool. 67: 1186-1212. Lee C. Y., Shallenberg R. S. and Vittum M. T. 1970. Free sugars in fruits and vegetables. Food Life Sci. Bull. 1: 1-12. Levenspiel 0. 1998. Chemical Reaction Engineering, John Wiley & Sons, New York, NY (3rded.). Levey D. J. 1987. Sugar-tasting ability and fruit selection in tropical fruit-eating birds. A u k 104: 173-179. Levey D. J. and Martinez del Rio C. 1999. Test, rejection, and reformulation of a chemical reactor-based model of gut function in a fruit-eating bird. Physiol. Biochem. Zool. 72: 369-383. Levey D. J. and Martinez del Rio C. 2001. It takes guts (and more) to eat fruit: lessons from avian nutritional ecology. Auk 118: 819-831. Levins R. 1966. The strategy of model building in population biology. Amer. Sci. 54: 421-431. Lopez-Calleja M. V., Bozinovic F. and Martinez del Rio C. 1997. Effects of sugar concentration on hummingbird feeding and energy use. Comp. Biochem. Physiol. 118A: 1291-1299. Lotz C. N. and Nicolson S. W. 1999. Energy and water balance in the lesser doublecollared sunbird (Nectarinia chalybea) feeding on different nectar concentrations. J. Comp. Physiol. B 169: 200-206. Malcarney H. L., Martinez del Rio C. and Apanius V. 1994. Sucrose intolerance in birds: simple nonlethal diagnostic methods and consequences for assimilation of complex carbohydrates. Auk 111: 170-177. Martinez del Rio C. 1990a. Dietary, phylogenetic, and ecological correlates of intestinal sucrase and maltase activity in birds. Physiol. Zool. 63: 987-1011. Martinez del Rio C. 1990b. Sugar preferences in hummingbirds: the influence of subtle chemical differences on food choice. Condor 92: 1022-1030. Martinez del Rio C. and Karasov W. H. 1990. Digestion strategies in nectar- and fruiteating birds and the sugar composition of plant rewards. Amer. Nat. 136: 618-637. Martinez del Rio C., Baker H. G . and Baker I. 1992. Ecological and evolutionary implications of digestive processes: bird preferences and the sugar constituents of floral nectar and fruit pulp. Experientia 48: 544-551.
Modeling digestive function
85
Martinez del Rio C., Cork S. J. and Karasov W. H. 1994. Modelling gut function: an introduction. In: The Digestive System in Mammals: Food, Form and Function. D. J. Chivers and P. Langer (eds.). Cambridge Univ. Press, Cambridge, pp. 25-53. Martinez del Rio C., Schondube J. E., McWhorter T. J. and Herrera L. G . 2001. Intake responses in nectar-feeding birds: digestive and metabolic causes, osmoregulatory consequences, and coevolutionary effects. Amer. Zool. 41: 902-915. McWhorter T. J. 2002. The integration of digestive, metabolic and osmoregulatory processes in nectar-eating birds. PhD Diss; Dept. Ecol. Evol. Biol., University Arizona, Tucson, AZ. McWhorter T. J. and Martinez del Rio C. 1999. Food ingestion and water turnover in hummingbirds: how much dietary water is absorbed? J. Exp. Biol. 202: 2851-2858. McWhorter T. J. and L6pez-Calleja M. V. 2000. The integration of diet, physiology, and ecology of nectar-feeding birds. Rev. Chil. Hist. Nut. 73: 451460. McWhorter T. J. and Martinez del Rio C. 2000. Does gut function limit hummingbird food intake? Physiol. Biochem. Zool. 73: 313-324. McWhorter T. J., Martinez del Rio C. and Pinshow B. 2003. Modulation of ingested water absorption by Palestine sunbirds: evidence for adaptive regulation. J. Exp. Biol. 206: 659-666. McWilliams S. R. and Karasov W. H. 1998a. Test of a digestion optimization model: effect of variable-reward feeding schedules on digestive performance of a migratory bird. Oecologia 114: 160-169. McWilliams S. R. and Karasov W. H. 1998b. Test of a digestion optimization model: effects of costs of feeding on digestive parameters. Physiol. Zool. 71: 168-178. Monsen N. E., Cloern J. E., Lucas L. V. and Monismith S. G . 2002. A comment on the use of flushing time, residence time, and age as transport scales. Limnol. Oceanog. 47: 1545-1553. Montgomery M. J. and Baumgardt B. R. 1965. Regulation of food intake in ruminants. 1. Pelleted rations varying in energy concentration. J. Dairy Sci. 48: 569-574. Nagy K. A. and Negus N. C. 1993. Energy acquisition and allocation in male collared lemmings Dicrostonyx groenlandicus: effects of photoperiod, temperature, and diet quality. Physiol. Zool. 66: 537-560. Nicolson S. W. 2002. Pollination by passerine birds: why are the nectars so dilute? Comp. Biochem. Physiol. B 131: 645 -652. Penry D. L. and Jumars P. A. 1986. Chemical reactor analysis and optimal digestion. Biosci. 36: 310-315. Penry D. L. and Jumars P. A. 1987. Modeling animal guts as chemical reactors. Amer. Nut. 129: 69-96. Penry D. L. and Jumars P. A. 1990. Gut architecture, digestive constraints and feeding ecology of deposit-feeding and carnivorous polychaetes. Oecologia 82: 1-11. Plante C. J., Jumars P. A. and Baross J. A. 1990. Digestive association between marine detritivores and bacteria. Ann. Rev. Ecol. and System. 21: 93-127. Powers D. R. and Nagy K. A. 1988. Field metabolic rate and food consumption by freeliving Anna's hummingbirds (Calypte anna). Physiol. Zool. 61: 500-506. Prop J. and Vulink T. 1992. Digestion by barnacle geese in the annual cycle: the interplay between retention time and food quality. Funct. Ecol. 6: 180-189. Schoener T. W. 1971. Theory of feeding strategies. Ann. Rev. Ecol. System. 2: 369-404. Schondube J. E. 2003. Flowerpiercers and hummingbirds: a comparative study of nectar feeding strategies in birds. PhD Diss. Dept. Ecol. Evol. Biol. Univ. Arizona, Tucson, AZ. Schondube J. E. a n d Martinez del Rio C. 2003. Concentration-dependent sugar preferences in nectar-feeding birds: mechanisms and consequences. Funct. Ecol. 17: 445-453. Seger J. and Stubblefield J. W. 1996. Optimization and adaptation. In: Adaptation, M. R. Rose and G. V. Lauder (eds.). Academic Press, New York, NY, pp. 93-124.
86
Physiological and ecological adaptations t o feeding in vertebrates
Sibly R. M. 1981. Strategies of digestion and defecation. In: Physiological Ecology: A n Evolutionary Approach to Resource Use. C. R. Townsend and P. Calow (eds.). Blackwell Sci. Publ., Oxford, UK, pp. 109-139. Simpson S. J., Barton-Browne L. and van Genven A. C. M. 1989. The patterning of compensatory feeding in the Australian sheep blowfly. Physiol. Entomol. 14: 91-105. Slansky F. and Wheeler G. S. 1992. Caterpillars compensatory feeding response to diluted nutrients leads to toxic allelochemical dose. Entomol. Exper. Applic. 65: 171-186. Snow D. W. 1981. Coevolution of birds and plants. In: The Evolving Biosphere. P. L. Forey (ed.). Cambridge Univ. Press, Cambridge, UK, pp. 169-178. Starck J. M., Karasov W. H. and Afik D. 2000. Intestinal nutrient uptake measurements and tissue damage: validating the everted sleeves method. Physiol. Biochem. Zool. 73: 454-460. Suarez R. K., Brownsey R. W., Vogel W., Brown G. S. and Hochachka P. W. 1988. Biosynthetic capacity of hummingbird liver. Amer. J. Physiol. 255: R699-R702. Suarez R. K., Lighton J. R. B., Moyes C. D., Brown G. S., Gass C. L. and Hochachka P. W. 1990. Fuel selection in rufous hummingbirds: ecological implications of metabolic biochemistry. Proc. Natl. Acad. Sci. U S A 87: 9207-9210. Tiebout H. M., I11 and Nagy K. A. 1991. Validation of the doubly labeled water method (3HH1BO) for measuring water flux and CO, production in the tropical hummingbird Amazilia saucerottei. Physiol. Zool. 64: 362-374. Udvardy M. D. F. 1983. The role of the feet in behavioral thermoregulation of hummingbirds. Condor 85: 281-285. Witmer M. C. 1998. Ecological and evolutionary implications of energy and protein requirements of avian frugivores eating sugary diets. Physiol. Zool. 71: 599-610. Yonge C. M. 1937. Evolution and adaptation in the digestive system of metazoa. Biol. Rev. 12: 87-115.
Digestive Constraints in Mammalian and Avian Ecology William H. Karasovl and Scott R. McWilliams2
'University Wisconsin, Department of Wildlife Ecology, Madison, WI, USA University Rhode Island, Department of Natural Resources, Kingston, RI, USA
SYNOPSIS The difference between the rate of nutrient intake for maintenance and the maximum rate of digestion, termed spare digestive capacity, potentially limits energy allocation. Because the maximum digestion rate can be adjusted upward in relation to factors such as diet quality and quantity there are both immediate and long-term spare capacities. We review their quantitation and time course for change, which are both ecologically important. A critical design feature of most studies measuring immediate spare capacity is that they quickly challenge animals to increase rate of digestion through cold stimuli, forced activity, or reduction in feeding time. A rapid time course is important because within a few days adjustments occur in the digestive tract that increase digestive capacity, in which case immediate spare capacity is no longer measured. Technical reasons why biochemical measures of spare capacity may not necessarily establish limitation at the whole-animal level are discussed herein. The majority of species studied had quite modest immediate spare capacities (range 950%). But in the same species the long-term spare capacity was about 100-125% above routine rates of nutrient intake or digestion. In laboratory mice digestive capacity increased to match any demand put on it, but whether the gut sometimes ultimately limits the energy budget is unknown for most animals. We review examples in which digestive limits are apparently dictated by the volumetric capacity of the gut or the rates at which food is either mechanically or biochemically broken down, but we know of no examples of limiting absorption.
88
Physiological and ecological adaptations t o feeding in vertebrates
DIGESTIVE CONSTRAINTS: SIGNIFICANCE AND GENERAL PRINCIPLES Wild mammals and birds undergo a range of food intake from hyperphagia during low temperature acclimation or periods of production (e.g.lactation, migratory fattening), which may require an enlarged gut, to restricted feeding or fasting which may cause disuse atrophy (see examples below and in Chapters 8,9, and 13 of this volume). Omnivorous species encounter different types of food on a daily and/or seasonal basis, and this may require biochemical adjustments for breaking down and absorbing different substrates (Karasov,1996;Karasov and Hume, 1997). Our interpretation that the attendant changes in gastrointestinal (GI) tract structure and biochemistry are "adaptive" rests on our assumption that the GI tract digestive characteristics (e.g.size, enzymes, etc.) are matched to the prevailing diet composition and feeding rate, and that these characteristics do not provide a digestive capacity in great excess of what is necessary for the prevailing diet and feeding rate. The first part of this idea is very well supported by many studies that show a positive correlation between size and enzyme content of the GI tract and daily feeding rate, a positive correlation between enzyme levels and the diet concentration of the enzyme primary substrates, and a positive correlationbetween diet nutrient density and retention time of digesta in the gut (Karasov and Hume, 1997).The second part of the idea is more often asserted than actually demonstrated. We assume that when load is increased, the animal's feeding may be constrained until digestive capacity is increased via the aforementioned adjustments in digestive characteristics. The idea of digestive limitation, besides operating as an important interpretive paradigm, could also be important if digestive processing limited energy flow or other ecological processes (e.g. diet selection). Wild animals do appear to have maximal sustained metabolic rates and if the limit is not imposed by food availability, three physiological hypotheses about the proximate factor(s)have been proposed (Karasov, 1986; Weiner, 1992;Hammond and Diamond, 1997). The central limitation hypothesis suggests that the bottleneck resides in physiological processes and systems, including the digestive system, that are involved in acquiring, processing, and distributing energy to energy-consumingorgans such as muscle or mammary glands. The peripheral limitation hypothesis suggests that processes (such as thermoregulation, lactation, activity)within the energy-consuming organs each have their own metabolic ceilings and this determines the maximum sustained metabolic rate. Finally, the idea of "symmorphosis" proposes that capacities of several of these potentially Limiting factors might be matched to each other and to natural loads (Taylor and Weibel, 1981;Weibel, 2000). One theme of interest here is that maximum sustained metabolic rate in many wild vertebratesmay be determined by the capacity of their digestive system.
Digestive constraints in mammalian and avian ecology
89
The concept of a digestive limitation can be traced back at least to the work of Max Kleiber (Kleiber, 1933, cited in Kleiber, 1961) on the maximal food capac:ty of domestic animals, and more recently updated by several authors (Kendeigh, 1949;West, 1960;Kirkwood, 1983;Karasov, 1986;Weiner, 1992).Most recently we have come to appreciate that digestion may represent a flexible limit because there is considerable evidence in birds (Karasov, 1996)and mammals (Karasov and Hume, 1997)that digestive features that may limit food processing are adjusted in relation to factors such as diet quality and quantity.Digestion rate for a particular food or substrate can be greatly increased through changes in digestive organ size, changes in the complement of enzymes and transport mechanisms for breaking down and absorbing food and substrate, and changes in alimentary tract muscular activity that affect the contact time between food or substrates and the gastrointestinal (GI)processes. The relative differences (orratios) between either the current or the absolute maximal digestion rate and the current food intake rate are measures of an animal's "safety margin" (Diamond, 1991) or "reserve capacity" (Diamondand Hamrnond, 1992)for responding to changes in environmentalconditions over different timescales.These concepts of GI flexibility and spare capacity are illustrated in Fig. 4.1. Two points deserve highlighting in Fig. 4.1: (1)at any given time an animal has some limited spare capacity (called "immediate spare capacity") but this decreases in extent as the GI system reaches its long-term capacity (Hammond et al., 1994);and (2)phenotypic flexibility of the GI organs is primarily responsible for an animal's ability to change food intake and diet (i.e.it represents most of the "long-term capacity"); however, such phenotypic flexibility requires acclimation time. Explicit references in ecology to possible digestive constraints actually predate most of the works on digestive constraint. For example, C.S. Holling criticized early predator-prey models because they assumed a linear relationship between an individual predator's consumption rate of prey and the prey's density. He proposed instead a "functional response" whereby the consumption rate increased with prey density but reached a plateau value beyond which consumption would not increase (Holling, 1959). Though many ecologists today associate the plateau value with a handling time that defines maximal intake in Holling's "disc equation" or with an herbivore's maximum rate of cropping and chewing (Gross et al., 1993),some ecologists, even in Holling's time, have recognized that maximal digestion rate could also dictate the plateau value (Jeschkeet al., 2002).For example, Mook (1963), observing clear satiation of wild bay-breasted warblers feeding on spruce budworms, modeled predation by including a functional response that included a digestive pause of two hours. Besides potentially limiting energy intake and thus growth, storage and reproductive rates, digestive limitations can be important in behavioral models of optimal diet, models of
90
Physiological and ecological adaptations t o feeding in vertebrates
(baseline nutrient load)
. . . . . . . . . , . . . . .
a
m
.
1
.
d
Acclimation time
Fig. 4.1. Immediate spare capacity and long-term capacity (phenotypic flexibility plus immediate spare capacity) for a hypothetical animal exposed to increasing energy demands (e.g. during migration, during cold weather). The solid lower line represents the nutrient load from feeding. Its baseline corresponds to the animal's routine energy demands (e.g. not during migration or at thermoneutral temperatures). The solid upper line represents the capacity of the gut for processing that nutrient load. Capacity on the y-axis could be total digestion rate, volumetric intake, nutrient uptake capacity, rate of digestive enzyme activity or some other performance measure of the animal. The x-axis is time since the start of an increase in energy demand or change in diet quality. At any given time, an animal can increase its food intake only within the limits set by the level of immediate spare capacity, which decreases as the animal approaches its longterm capacity. When energy and nutrient demands increase, and if the animal has been given time to fully acclimate to these elevated energy demands, then phenotypic flexibility in the digestive system of the animal enables increased energy intake (shown as the increase in solid lower line above the baseline nutrient load). These changes in digestive capacity are critically important in allowing animals to overcome the challenges associated with changing diet quality or quantity (adapted from Diamond, 1991; Diamond and Hammond, 1992).
territorial defense, daily foraging patterns, and optimal migration strategy (examples in Bednekoff and Houston, 1994; Karasov, 1996).
Digestive constraints in mammalian and avian ecology
91
The easiest way to detect a digestion-limited animal is to directly measure food collection rate (or handling time) and digestion rate (or digestion time) and compare them, although there are some other methods as well (Jeschke et al., 2002). Thus, insectivorous house wrens (Troglodytes aedon) were found capable of collecting at least 21 grams dry mass arthropods/day (Dykstra and Karasov, 1993),three times as much as their maximal digestion rate (Dykstra and Karasov, 1992). Voles can consume herbage at a rate of 0.15 g min-', at least ten times faster than they can process it (Zynel and Wunder, 2002).Classic examples of digestive limitation are provided by some avian herbivores (Kenward and Sibly 1977;McWilliams and Raveling, 2004) and by ruminant herbivores, for whom rate of intake may be limited by biting, chewing, and rumination, and not plant abundance (e.g. Spalinger et al., 1986;Spalinger et al., 1988).Similar kinds of arguments have been made for nectar-feeding hummingbirds (Diamond et al., 1986),and for bivalve- or crab-consuming birds (Zwarts and Dirksen, 1990; Kersten and Visser, 1996; Guillemette, 1994;Guillemette, 1998).Granivores ought to provide other examples of digestive limitation because they forage on a sometimes highly available food resource yet have a relatively long digestive processing time (Karasov, 1990),but to our knowledge no one has adduced an example to date. According to Jeschke et al. (2002), all animals for which measures of both digestion and handling time are available are digestion limited. One goal here is to review the quantitation of digestive capacity. There have been far more estimates of long-term than immediate spare capacity and the two are sometimes not distinguished. Further, we think some current estimates of spare capacity are inaccurate and have illustrated how it can be more accurately estimated. While digestive Limitation can have clear ecological significance, its mechanistic basis is rarely defined. The magnitude of the limit might be dictated by the volumetric capacity of the gut or the rates at which food are either mechanically or biochemically broken down or absorbed. Whichever feature(s)dictates the digestion limit, its time course for change is also rarely defined, though that too has important ecological implications. Without knowing such details, the quantitative integration of digestion with postabsorptive metabolism in the overall scheme of nutrient processing cannot be completely achieved in a fashion analogous to that achieved for respiratory and metabolic physiology (Weibel,2000). To further this endeavor, and in light of their ecological significance, we therefore focus on these mechanistic details. We think that the magnitude of immediate and long-term spare capacity, and the time course over which digestive capacity can be increased, are the two keys to understanding the digestive challenges that animals face under a variety of interesting ecological situations.
Physiological and ecological adaptations t o feeding in vertebrates
92
DEFINING THE LIMITS: MAGNITUDE OF OVERALL DIGESTIVE CAPACITY AND I T S MEASUREMENT I N FEEDING TRIALS Immediate and long-term spare capacity can be estimated in balance trials in which maximum feeding and digestion rates are measured in animals highly motivated to feed, presumably at maximal levels. The spare capacity is the relative difference (or ratio) between the rates measured under those conditions and the rates measured under more routine or baseline conditions. The method is exemplified in Weiner's (1987)study of energetics of Djungarian hamsters. He took hamsters acclimated to room temperatures (22°C) and switched them to cold conditions (-2°C) either quickly or gradually over many days. In the cold, hamsters must eat more to balance higher heat loss or they will catabolize their body tissues to supply the extra energy. Hamsters switched quickly increased their feeding and digestion rate only 15% and lost body mass, whereas hamsters acclimated slowly increased their rates 92% and maintained body mass. Thus, hamsters switched quickly experienced an energy deficit and should have been motivated to eat more, but did not. Presumably, they did not have the spare digestive capacity to do so; their "immediate spare capacity" was only 15% above what they needed routinely for energy balance at 22°C. Hamsters switched more gradually were able to increase their digestive capacity, but their 92% increase in response to cold acclimation was still less than their long-term spare capacity. Weiner (1987) found that hamsters at peak lactation could increase their digestion rate 116 O/O compared with nonreproductives, so this would be a closer estimate of their long-term spare capacity. There have been many measurements of feeding and digestion rates in mammals and birds highly motivated to feed, presumably at near-maximal levels. Typically they involve animals acclimated to very low temperatures, ideally at their limit of thermal tolerance, high levels of forced activity or hyperphagic animals during lactation, storing energy for migration and hibernation, or engaged in rapid growth. Some of these data have been Table 4.1. Relationships between near maximum metabolizable energy intake (MEmax, kJ/d) and body mass (m in g). Group
Energy demanding situation1 Mammals and birds C, L, G Mammals L Birds C Passerine birds M Shorebirds M
Allometric equation
Reference
MEmax= 11.84 m0.72 (Kirkwood, 1983) MEmax= 18.49 nzO66 (Weiner, 1992) (Karasov, 1990) MEmax= 16.42 MEmax= 16.09 m0.70(Lindstrom and Kvist, 1995) MEmax= 11.7 moR2 (Kvist, 2001)2
Energy demanding situations: C = cold acclimation; L = lactation; G = growth; M migratory fattening. Equation calculated by us.
=
Digestive constraints in mammalian and avian ecology
93
summarized in allometric equations that express maximum metabolizable as a function of body mass (Table 4.1). As a general energy intake (MEmax) rule, MEmxscales with body mass in a fashion similar to other metabolic rates (i.e.with mass2/3-3/4; note though that none of the estimates control for phylogenetic association among the data) and is 4 - 7 times basal metabolic rate (Hammond and Diamond, 1997).The highest value we know of, measured in lactating mice exposed to low temperature, is 7.7X resting metabolism (Johnsonand Speakman, 2001). MEmxcan depend on the nature of the food. For example, shorebirds that must crush hard shellfish in their gizzards cannot sustain the very high rates they achieve when eating commercially prepared, soft trout food or mealworms (below). Also, Kvist and Lindstrom (2000)made an important point that the absolute amount of food digested per day is influenced not just by the hourly rate, but also the total hours available for feeding (also see McWilliams and Raveling, 2004). The digestive adjustments of mammals and birds acclimated to high feeding rate almost always include increased gut size (though see Johnson and Speakman,2001) and consequently increased amounts of digestive enzymes and nutrient transporters (Karasov and Hume, 1997; McWilliams and Karasov, 2001). Unfortunately, there is no published study for any vertebrate of both rapid and gradual adjustment of feeding and digestion to high energy demand that includes corresponding changes in gut size and biochemistry. The rapid-adjustment experiments, which have been least often performed, are perhaps most interestingbecause they reveal the immediate spare digestive capacity of the animal. We designed a comprehensive study with white-throated sparrows (Zonotrichia albicollis) to determine their response to both rapid and gradual increase in energy demand in order to estimate the level of spare capacity and phenotypic flexibility in their digestive system in response to changes in feeding rate. The experiment involved manipulating ambient temperature, which caused changes in the metabolic rate of sparrows (i.e.increased metabolic rate with lower ambient temperature) and thus induced changes in their food intake to maintain their body temperature constant. By random assignment, sparrows were either held continuously at +21°C,switched rapidly from +21°Cto -20°C, or gradually acclimated to -20°C over 50 days. We measured daily food intake and digestive efficiency of starch (the primary nutrient in their semisynthetic diet) in the three groups of sparrows. The prediction was that sparrows switched rapidly from warm to cold temperatures would maintain digestive efficiency constant only if some safety margin of nutrient absorption capacity over nutrient intake existed before the temperature switch. White-throated sparrows at -20°C required 83% more food than birds at +21°C, as indicated by the comparison of feeding rates of acclimated sparrows in steady state at -20°C and +21°C (Fig.4.2).When birds were switched rapidly from +21°Cto -20°C they increased feeding rate only 45%, a level of
ysiological and ecological adaptations t o feeding
rtebr
h
u
A
Fig. 4.2. Food intake, body mass change, retention time of digesta, and digestive efficiency of starch in white-throated sparrows that were either acclimated to +21°c or - 2 0 ' ~ or switched immediately from +21°c to -20'~. Sparrows acclimated to -20°c ate more than sparrows acclimated at +21°c although both groups of sparrows maintained similar body mass. Sparrows in all three treatment groups had similar digestive efficiency and retention times. Thus, sparrows acclimated to +21°c have a limited spare capacity of about 45% as indicated by an increase in food intake of this magnitude for birds switched rapidly to colder temperatures. However, this limited increment in food intake did not suffice to satisfy the energy demands imposed by a rapid switch from +21°c to - 2 0 ' ~ given these birds lost body mass. This indicates that phenotypic flexibility in digestive features is necessary for sparrows to achieve their long-term capacity.
food intake which was not sufficient to satisfy the extra energy demands, as evidenced by body mass loss (Fig.4.2). Interestingly, birds in all three treatment groups had similar digestive efficiency and retention times (Fig.4.2). Thus, sparrows have some spare capacity (of about 45%) but it did not suffice to satisfy the energy demands imposed by a rapid switch from +21°Cto -20°C. If given enough time for acclimation to the cold, however, sparrows can satisfy the elevated energy demands associated with living in the cold, as evidenced by their ability to maintain body mass after 50 days of acclimation at -20°C. The digestive adjustments to increased feeding rate that occurred during acclimation to the cold included an increase in size of small intestine (Fig. 4.3),large intestine, and liver but not gizzard and pancreas. We are currently completing analysis of digestive enzyme activity and nutrient uptake rates to determine whether adjustments in these digestive features are involved along with changes in gut size. Notice that the 57% increase in
Digestive constraints in mammalian and avian ecology
Fig. 4.i. Small intestine mass (g) of white-throated sparrows that were either acclimated to +21 C or - 2 0 ' ~ or switched immediately from +21 C to -20'~. Sgarrows acclimated to - 2 0 ' ~ had larger small intestines than sparrows acclimated at +21 C, whereas sparrows switched immediately from +21°C to - 2 0 ' ~ had similar small intestine mass as sparrows acclimated to +21°C. See text for a discussion of how these increases in gut size along with the results shown in Fig. 4.2 can be used to estimate the immediate spare capacity and long-term capacity of white-throated sparrows (depicted hypothetically in Fig. 4.1).
small intestine sufficed to accommodate the 83'10 higher feeding rate in birds acclimated at -20°C. This is apparent because mean retention time, efficiency digesting starch, and body mass did not decline significantly with cold acclimation (Fig.4.2). If one considers that sparrows acclimated to +21°Chad a spare capacity of 45% to start with, adding an increase in gut size of 57% to that can more than account for the 87% increased ability to process food. The two measures together imply that sparrows acclimated to -200C probably still had some immediate spare capacity and therefore their long-term digestive capacity was higher than 87% above their digestion rate when held at +21°C. This makes sense because it is known that captive white-throated sparrows can tolerate temperatures down to -29°C when feeding rates are 2.26 (126%)times higher than at +21°C (Kontogiannis,1968). Interestingly, white-throated sparrows engaged in forced activity could tolerate temperatures down to only -5°C but their digestion rates were similar to those of birds acclimated to -29"C, which is consistent with a central limitation set by nutrient processing rather than a peripheral limitation set by heat generation. Thus, the results from the experiment with white-throated sparrows, along with those of Kontogiannis (1968), conform nicely to the model presented in Fig. 4.1 and imply that immediate spare capacity is around 45%
96
Physiological and ecological adaptations t o feeding i n vertebrates
but that after long term acclimation the long-term capacity is around 126% above "baseline". Another method to estimate the immediate spare capacity is to measure intake/digestion rate during periods of restricted feeding. The approach is exemplified in the study by Winter (1998)on a single nocturnal nectarivorous bat (Glossophaga longirostris, 16.4 g), and other examples are provided below. Winter (1998)manipulated the ratio of day-to-night length and forced the bat to eat and digest relatively large amounts of sugar in relatively short amounts of time. He pointed out that after the first hour of feeding during which the gut becomes essentially filled, there is a steady-state period during which food intake can be no faster than the rate of processing, which includes both sugar (sucrose + glucose + fructose) digestion/absorption, postabsorptive processing, and excess water excretion. When feeding time was decreased after 7 d at 12 h / d to 2 h/d, the rate of glucose assimilation during the steady-state period of the night increased to a level 73% higher than during 12-hnights. But the bat feeding for just 2 h/d lost mass and so after 2 d it was switched to 4 h / d and then to 6 h / d feeding throughout which it maintained mass and continued to have an elevated hourly feeding rate as when given 2 h / d to feed. As the bat's energy budget was more and more stressed, it reduced its energy expenditures by reducing flight time. The results thus implied that when motivated to feed, the bat utilized its immediate spare digestive capacity to the maximum, which was apparently at least 73%. With the data it cannot be decided whether this is characteristic of the species (only one individual was studied), nor whether food intake was limited by sugar absorption capacity or water clearance capacity. In a third method to infer immediate spare digestive capacity, which we term a hybrid method, results of feeding trials and in vitro assays are combined to yield estimates of spare capacity. A study on migratory yellowrumped warblers (Dendroica coronata) can be used to illustrate the method (Leeet al., 2002). The birds were captured during migration and habituated to a diet of fruit mash and mealworms. Controlbirds were fed ad libitum but experimentalbirds were food restricted for 3 days by providing 44% of the ad libitum level of food. One purpose of the food restriction was to increase their motivation to feed maximally, and once they were again provided food ad libitum they increased their feeding and digestion rate by 18% compared with controls and the birds gained body mass. This suggests a spare digestive capacity of at least 18%,but other measures of digestive enzymes indicated that it was actually greater than this. The authors showed that the food restriction caused a 20% decline in intestine mass, declines of about 40% in intestinal enzymatic capacities (sucrase, maltase, and aminopeptidase activities were measured along the intestine), and pancreatic enyzme levels were not significantlyaffected (trypsin and chymotripsin) except for a 36% reduction in amylase. If the previously food-restricted birds could increase their digestion rate by 18% compared with controls while concomitantly
Digestive constraints in mammalian and avian ecology
97
possessing around 40% less enzyme activity than controls, apparently the immediate spare capacity of the control birds must have been approximately 58% (= 18 + 40). There is some support for this estimate. McWilliams and Karasov (1998b)employed a time-limited intermittent feeding protocol and also found evidence of immediate spare digestivecapacity in yellow-rumped warblers. In that study, yellow-rumped warblers were capable of increasing their food intake by 50% in a matter of hours with no change in digestive efficiency or mean retention time. With longer acclimation time warblers increase their feeding and digestion rates more than this during the migratory phase which can be induced by changes in light cycle. For example, warblers on long daylength (16L:8D)had hourly feeding/digestion rates 130%higher than warblers on short daylength (10L:14D;McWilliams and Karasov, 1998a). We are aware of only seven balance studies that permit estimation of immediate spare digestive capacity in mammals and birds (Table 4.2). For now we have excluded a number of studies that estimated immediate spare capacity solely on the basis of biochemical measures (e.g. Buddington and Diamond, 1990, 1992; Toloza et al., 1991; Toloza and Diamond, 1992;Jackson and Diamond, 1995;Weiss et al., 1998)because they do not necessarily establish limitation at the whole-animal level (i.e. the chosen biochemical measure may have higher spare capacity compared with some other step of digestion) and because we have concerns (discussed below) about their accuracy. Of the balance trials, four were discussed above and the others will be considered shortly. The critical design feature of most of these studies is that they quickly challenged animals to increase rate of feeding and digestion, either through cold stimulus, forced activity, or reduction in feeding time. The last, hybrid method essentially inferred the immediate spare capacity by comparing digestive responses of control animals feeding at routine levels with experimental animals with reduced digestive tracts. Another method, never tried but which might be considered, is experimental ablation of the brain's food intake control center (ventromedial hypothalamus)which rapidly brings about hyperphagia.Whatever the method, a rapid time course is important because within a few days adjustmentsoccur in the GI tract that increase the digestive capacity (discussedbelow), in which case immediate spare capacity is no longer measured. All the species studied had quite modest immediate spare capacities (range 9-5O0/0), excluding the measurement on a single bat. This implies that in the wild sudden larger increases in energy needs due to increased activity or thermoregulatory costs cannot be immediatelycompensated by increased food intake even if food is abundant; instead, behavior patterns must be altered to save energy or energy stores must be recruited. But in the same species the long-term spare capacity, achieved partly through adjustments in the GI tract over the course of several days (below) is about 100-125% above routine rates of feeding/digestion (much higher in mice; Table 4.2).
26.4
24
16.4
51
11
3.3
Mus rnusculus
White-throated sparrow
Glossophaga Iongiros tris
Prairie vole
Yellow-rumped warbler
Broad-tailed hummingbird
Daily digestion rate at 22OC
Daily digestion rate at 23°C Daily digestion rate at 21°C
Daily digestion of lactating females at 21°C Daily digestion rate at 21°C Hourly digestion rate for 12 h/d
Daily digestion rate at 22°C
"Baseline" conditions
50
58%
20%
-
9%
73?hd
45%
10%
15%
126%
300hb 488%"
116°/0
1
Cold acclimation, 4 lactation Migratory mode 5 induced by increasing daylength 6
Lactating females acclimated to 8 "C 7 Acclimated to -29 "C 2
Peak lactation
Long-term spare capacity Ref ." Increase Method of over baseline determination
Reduced feeding Not determined time to 2 - 4 h/d Reduced feeding time 95% and feeding bout duration Hybrid estimate 130% and reduced feeding time Measurement Not of sucrase and determined switched to 10°C
Lactating females switched to 8OC Switched to -20°C
Switched to -2OC
Immediate spare capacity Increase Method of over baseline determination
"eferences: 1 Weiner (1987); 2 McWilliams and Karasov (2002); Kontogiamis (1968); 3 Winter (1998); 4 Zynel and Wunder (2002); 5 McWilliams and Karasov (1998a); Lee et al. (2002); 6 McWhorter and Martinez del Rio (2000); 7 Johnson and Speakman (2001) bincrease compared with lactating females at 21°C 'increase compared with nonreproductive females at 21°C donly a single bat studied
35
Mass (g)
Djungarian hamster
Species
Table 4.2. Immediate and long-term spare capacity estimated in balance trials in which maximum feeding and digestion rates were measured in animals highly motivated to feed, presumably at maximal levels. The critical design feature of most of these studies is that they quickly challenged animals to increase rate of feeding and digestion, either through cold challenge, forced activity, or reduction in feeding time.
Digestive constraints in mammalian and avian ecology
99
DEFINING MECHANISMS: DIGESTIVE FEATURESTHAT MIGHT LIMIT OVERALL DIGESTIVE CAPACITY The magnitude of the digestive limit might be dictated by the volumetric capacity of the gut or the rates at which food are either mechanically or biochemically broken down or absorbed. There are plausible examples of most of these.
Limited Gastrointestinal Tract Volume as a Digestive Constraint Zynel and Wunder (2002),employing a protocol of reduced feeding time (see above), described an apparent gut volume limitation in captive, nonreproductive herbivorous prairie voles (Microtus ochrogaster). They held the animals at 23°C and fed the controls ad libitum and the experimentalseither in a single 3-hour time block per day or in six half-hour time blocks spread every 4 h through the day (still 3 h total feeding time). Three hours of feeding was chosen for the experimental voles because this was more than enough time for them to ingest and chew their daily food requirement of 7.7 g d-'. Voles in both the experimental groups rapidly filled their stomachs with up to 1.4g dry food, the maximum stomach capacity determined in earlier studies. Voles fed in a single time block could not maintain body mass constant whereas voles fed in multiple time blocks could. In a single 3-h time block voles could apparently process at most 2 g if they continually "topped-off" as digesta moved from the stomach through the distal GI tract. In contrast, if voles filled their stomach with 1.4g once every 4 h, which is apparently time enough to clear the stomach, they potentially could digest 8.4 g d-' (= 1.4 g x 24 h/4 h), which suggests an immediate spare capacity of 9% (8.4/7.7 = 1.09). Voles can increase their feeding and digestion rate much more than this when chronically acclimated to low temperature or during lactation. Their long-term spare capacity is about double the routine digestion rate of the controls in this study (Zynel and Wunder, 2002). Though we have described this as an example of a volumetric constraint, perhaps it would be more accurate to say that the bottleneck might lie in the volumetric turnover in g/h, or rate of emptying in g/h of the stomach. This expression puts the bottleneck in the same units as the rate being limited (feeding rate in g/h). On the one hand the distinction seems moot if voles exhibit near instantaneous stomach filling time in relation to stomach emptying time, but on the other hand it begs the question of whether subsequent digestive processes (breakdown,absorption, etc.) are too slow to permit more rapid emptying of the stomach into the small intestine and thus signal the stomach by negative feedback. In any event, an important ecological interpretation of this putative bottleneck is framed in terms of time, i.e. that this bottleneck causes optimally spaced rest bouts between feeding and thus is the primary cause of the observed ultradian rhythm in voles (see discussion in Zynel and Wunder, 2002).
100
Physiological and ecological adaptations t o feeding in vertebrates
Limited Rate of Mechanical Breakdown as a Digestive Constraint Birds that consume and crush shellfish provide a compelling example of this kind of limitation. The limitation is suggested for red knots (Calidris canutus) by the fact that they achieve lower rates of MErnx(by less than half) when consuming whole bivalves, whose shells they must crush and excrete via the cloaca, than when consuming the flesh alone which has been removed from shells (T. Piersma, pers. comm.).The increase in gizzard muscularity when knots are transited from soft food to whole bivalves (Piersma et al., 1993; Dekinga et al., 2001) is consistent with the idea that mechanical breakdown is an important limitation to overall digestion rate. The higher maximum rate on flesh perhaps reflects limits in digestive or postabsorptive biochemical processing of the primarily proteinaceous material. Another example of an apparent bottleneck caused by limiting rate of mechanical breakdown might be rumen clearance in ruminants (Van Soest, 1994). The orifice between the rumenoreticulum and omasum functions like a particle size or density sieve so that particles do not escape the rumen until they are sufficiently reduced in size. Reduction is achieved partly through mechanical means (muscular activity of the rumen in conjunction with rumination and chewing) and partly through biochemical means (fermentation rate). Intake of additional food must be matched to the rate of clearance from the rumen. There are other interesting cases of possible limitation by mechanical breakdown that beg to be studied. As mentioned above, among birds granivores have relatively long digesta processing times but the possibility of this being a mechanical digestion limitation has not been systematically explored. Insectivores have been little studied, but Hanski (1984)reported that apparent digestive pauses became more evident when shrews were fed heavily chitinized beetles than when fed lightly chitinized insect pupae. It seems reasonable to apply the same research approach to these situations as described above for red knots: present the same food either intact or mechanically preprocessed under conditions that motivate the animals to feed maximally. Limited Rate of Biochemical Breakdown as a Digestive Constraint McWhorter and Martinez del Rio (2000)proposed that food intake by migratory broad-tailed hummingbirds (Selasphorus platycercus) is limited by rates of hydrolysis. These birds digest mainly sucrose and so sucrase activity in the intestine's brush border was measured in vitro. The in vitro measurement was made with homogenates of tissues collected along the length of the intestine under conditions that saturate the enzyme(s) so that the maximal reaction velocity (V_J could be integrated along the length to yield a total hydrolytic capacity. This capacity was about 120% higher than the
Digestive constraints in mammalian and avian ecology
101
observed rates of sucrose intake and digestion, implying that the immediate spare capacity was quite high. But, as the authors pointed out, the common procedure of using the VmaX over the entire intestinal length is physiologically unrealistic because the sucrose concentration progressively lowers as the digesta flows distally along the gut during digestion. Using a more sophisticated model of the gut as a plug-flow chemical reactor, Jumars and Martinez del Rio (1999)calculated a lower digestive capacity that was only 15-35% higher than observed rates of sucrose intake/digestion. They considered this to be the more accurate estimate of the immediate spare digestive capacity of the broad-tailed hummingbird. Support for their argument came in trials in which they rapidly exposed the hummingbirds to low temperature. The birds did not (could not?) increase their intake but instead reduced their expenditure by utilizing torpor. In a similar kind of experiment rufous hummingbirds (Selasphorus rufus, 3.2 g) switched suddenly to low temperature did not (could not?) sufficiently increase their intake and lost body mass (Gass et al., 1999). The study by McWhorter and Martinez del Rio (2000)underscored some important considerationsin estimating digestive capacity by extrapolation from measures of maximum enzymatic breakdown rate in vitro. First, the method assumes that hydrolysis rates measured in vitro correspond to actual rates in vivo. This may apply best for digestion of sucrose for which hydrolysis depends only on an enzyme bound to the intestine's brush border that is easily measured. Second, for most foods, besides sucrose-rich nectars, there are multiple substrates (e.g. starch, protein, fat) whose digestion is much more complex involving gastric and/or pancreatic enzymes that act in addition to multiple intestinal brush border enzymes. This complexity far exceeds our current abilities to model the overall process. Third, the hydrolysis rate is concentration dependent over some substrate range. Though most other studies estimating hydrolytic capacity (e.g.Hammond et al., 1994; Weiss et al., 1998; Martinez del Rio et al., 2001) have assumed constant saturating substrate concentrations, the newer, more physiologically realistic approach by McWhorter and Martinez del Rio (2000)showed that the aforesaid studies surely overestimated the hydrolytic capacity.
Limited Rate of Nutrient Absorption as a Digestive Constraint We know of no published study that provides strong evidence of nutrient absorption acting as a digestive bottleneck. Earlier studies that stimulated much interest in digestive bottlenecks (Karasov et al., 1986;Diamond et al., 1986)may be used to illustrate the problem. The intestinal glucose uptake capacity of rufous hummingbirds (SeIasphorusrufous, 3.2 g) was estimated to be 87.7 pmol h-' based on in-vitro measurement. How does this compare with actual intake? A 3-g hummingbird held at room temperature digested 1.5 - 2 g sucrose d-' (McWhorter and Martinez del Rio, 2000) or 4.4
102
Physiological and ecological adaptations t o feeding in vertebrates
- 5.8 mmol/d-'. Assuming that all this was digested in 16 h and that half was glucose, the bird's actual glucose absorption rate was thus at least 138 pmol h-I, 56% higher than the maximal uptake capacity in vitro! In some other studies when the in vitro measurement of D-glucose absorption was less than what animals actually achieved, the authors argued that glucose that was not absorbed by the intestine was later fermented in the hind gut. But this kind of explanation cannot apply to a hummingbird (no hind gut). The simplest explanation is that the in vitro measurement underestimated actual glucose absorption rate. There is valid concern that in other studies a similar underestimation occurred but was overlooked by invoking cecal fermentation. Consider some of the problems that plague estimates of nutrient absorption capacity, which can easily lead to either over- or underestimation. Overestimation of absorption rate, as for hydrolysis rate, is possibly caused by improper assumptions about lumenal nutrient concentrations.For example, when Toloza and Diamond (1992)estimated the immediate spare absorptive capacity of adult laboratory rats they found that mediated glucose absorption was 130% higher than daily glucose intake rate when they assumed lumenal concentrationwas 50 mM, but only 20% higher when they assumed the lower actual determinations of lumenal glucose concentration because absorption rate is much lower at low concentration (Fig.4.4).Several factors can lead to underestimation of absorption rate. It is possible that absorption rates measured in vitro are less than the rates in vivo because isolated tissue may become damaged and lead to underestimation of active transport rates (Starck et al., 2000). Also, measures of absorption with isolated intestinal tissue apparently fail to incorporate processes that may function in the intact animal such as trafficking of additional glucose transporters (GLUT 2) to the brush border stimulated by the presence of lumenal sugar (Kellett and Helliwell, 2000), and an important passive absorption pathway that seems very important, at least in birds (Karasov and Cork, 1994; Caviedes-Vidal and Karasov, 1996; Chediack et al., 2001) and probably in mammals (Pappenheimer, 1998; Fig. 4.4). Other kinds of absorption measures in vivo in anesthetized animals may be suspect because the anesthesia can influence the rates of absorption (Uhing and Kimura, 1995).Weber and Ehrlein (1998) arguably misestimated spare capacity by overlooking the very real physiological constraint that animals do not excrete a large amount of unabsorbed solute (see Mc Whorter and Martinez del Rio 2000),and by assuming that the apparent maximum absorption rate at their test concentration would represent the maximum absorption rate at higher test concentrations. Estimation of whole-animal glucose absorptive capacity by in vitro methodology has rarely been validated and in one of the earlier studies applying it, Toloza and Diamond (1992) pointed out that the calculation, which has numerous approximations, should be considered meaningful to an order of magnitude. Setting aside the issue of quantitative accuracy, we
Digestive constraints in mammalian and avian ecology
1 0
de+;*
METHOD in vitro
A
10 20 30 40 50 60 70 80 90 100 Glucose concentration (mM)
Fig. 4.4. Estimation of nutrient absorption capacity depends on method used and concentration assumed. This is illustrated in the comparison of measures in jejunum of adult laboratory rats. Many studies have applied the everted sleeve method (Karasov and Diamond, 1983) which was used by Debnam et al., (1988) to measure mediated Dglucose uptake ("in-vitro", open triangles, solid line). These researchers also measured mediated D-glucose absorption in perfused jejunum of anesthetized rats ("in-vivo A", open circles, dashed line). At low concentrations the rate is lower than in the in-vitro preparation because of unstirred layer effects, but maximal mediated uptake (plateau values) is fairly similar. Both measures, as well as the single highest reported maximal mediated in-vitro uptake in rats (431 nmol min-' ~ m -Toloza ~; and Diamond, 1992) are lower than absorption rates measured in chronically perfused, unanesthetized adult rats (Ugolev et al., 1986)) interpreted by Pappenheimer, 1998) ("in vivo C", open squares, dotted line). This latter measurement may include the effect of recruitment of additional glucose transporters (GLUT 2) that have lower affinity than the brush border glucose transporter (SGLT 1)(Kellett and Helliwell, 2000) and includes passive absorption, typically neglected in calculations of absorption capacity but which becomes especially important as concentration increases.
do think that in vitro measures are very useful for indicating qualitative changes in digestive capacity. Furthermore, when in vitro biochemical measures are made in conjunction with other whole-animal measures, perhap they can lead to useful hybrid estimates of immediate spare capacity, as described above.
HOW QUICKLY DOES DIGESTIVE CAPACITY INCREASE?
In the wild when energy needs suddenly increase, the digestive system could act as a bottleneck over some short term even if it eventually adjusts to permit
104
Physiological and ecological adaptations t o feeding in vertebrates
a higher rate of energy flow. The period of time over which this digestive constraint operates is dictated by the time it takes to increase digestive organ size or tissue-specific levels of digestive enzymes and nutrient absorption mechanisms. Relying on a rather limited number of studies, we can assemble a picture of the time course of digestive adjustment starting with turnover time of intestinal enzymes and epithelial cells and proceeding through rates of change of entire tissues to whole-animal feeding responses. The picture that emerges is that biochemical changes may occur faster than structural changes and changes may occur faster in small than in large animals. Starting with the most basic level, birds and mammals switched from carbohydrate-free diet to high carbohydrate diet could at least double the enzymes and/or nutrient specific activity of their ~arboh~drate-d;~estin~ transporter within 1-2 days of the dietdswitch (Karasov and Hume, 1997). An important mechanism is the replacement of intestinal cells with new cells possessing more copies of particular digestive enzymes (Karasov and Hume, 1997). In birds the rate of cell proliferation, indexed by the length of the S-phase (phase of DNA replication, measured by labeling in vivo) was measured in two different-sized species during growth (Starck, 1996).This rate did not differ markedly by age or species and so given a rather invariant S-phase (average 6 hours), the intestinal turnover time of small birds (replacementtime of intestinal cells)was 2-3 days compared with 8-12 days in larger birds (Starck, 1996). Among the six mammal species studied by Smith et al. (1984), however, there was no marked body-size dependent variation in enterocyte life span and, as in birds, enterocyte turnover rate was independent of age in mice (Ferraris and Vinakota, 1995). In laboratory rats, which have a one-day enterocyte turnover (Karasovand Diamond, 1987), following a fast the villi returned to their normal length within a day after initiation of feeding (Butset al., 1990;Hodin et al., 1994).The first responses of the atrophied gut of starved rats to initiation of feeding occurred as early as two hours after the first meal, when genes such as c-fos and c-jun, which represent the mitogenic response in many types of tissues, were first expressed in intestinal crypt cells (Hodin et al., 1994). Several studies, especially in birds, have monitored progressive changes in organ sizes following diet switches using destructive or nondestructive sampling methods. Fasted blackcaps (Sylvia atricapilla) that had reduced intestinal mass grew back their small intestine in two days or less once they were provided with food ad libitum (Karasovet al., 2004).Red knots switched from soft food to hard shellfish increased gizzard mass 147'/0 within 6 days (Dekinga et al., 2001). Japanese quail switched to high fiber diet increased gizzard mass 110% within 6 days, but significant increases were already apparent 1 day after the diet switch (Starck, 1999a). Reversible changes in gut length in response to changes in diet composition have been reported to occur within 3 4 weeks in grouse and quail (Moss and Parkinson, 1972;
Digestive constraints in mammalian and avian ecology
105
Savory and Gentle, 1976a, b) and ducks (Miller, 1975),with significant responses within 5 days in ducks (Drobney, 1984; Kehoe et al., 1988). Whole-animal feeding trials gave a similar picture of the time course for adjustment. American robins (Turdus migratorius) and European starlings (Sturnisvulgaris) switched from fruit to insect diets progressively increased digestive efficiency within three days of the diet switch (Levey and Karasov, 1989).Fasted blackcaps and thrush nightingales that had been food restricted, progressively increased their digestion rates to a maximum over the course of 3 days after returning to ad libitum feeding (Karasov and Pinshow, 2000; Kvist and Lindstrom, 2000). Red knots delayed accepting a new shellfish diet for at least 2 days when switched from soft food (Piersma et al., 1993; Dekinga et al., 2001). In summary, the response of the digestive system to changes in diet composition and feeding rate seems rapid. Even for structural measures (e.g. gizzard or intestine mass) that may respond more slowly than biochemical measures, statistically significant changes of a magnitude of 2040% are apparent in most species within 1-2 days of a change in diet (Starck, 199913). But systematic studies within and across species of correlated rates of change in digestive biochemistry and structure in response to whole-animal dietary adjustment are generally lacking. Long-term Digestive Capacity-How High can it Go? If birds are given adequate time to acclimate, then increases of at least two times in food intake and digestion rate are possible (Karasov,1996). Doubling food intake occurs commonly in birds preparing for migration (Berthold, 1975;Blem, 1980; Karasov, 1996)and in birds at cold temperatures (Dawson et al., 1983; Karasov, 1990; Dykstra and Karasov, 1992; McWilliams et al., 1999). Many mammals exhibit increases of similar magnitude (e.g. Tables 4.1 and 4.2) but some truly extraordinary increases have been recorded in laboratory mice (Hamrnond et al., 1994).For example, nonreproductiveSwissWebster female mice doubled their intake/digestion rate when switched from 23 to 5°C and could still increase it 3.3times more at peak lactation with very large litters. The net long-term digestive capacity was thus about 6.7 times the rate under routine conditions. The relative increase was similarly high, 5.9 times, in the MF1 strain of Mus musculus (Johnsonand Speakman, 2001). Are mice exceptional in this regard because of selection for high reproductive rate? In domesticated birds an important digestive change obtained as a result of artificial selection for more rapid growth was an increase in the relative size of the digestive organs (Lilja et al., 1985; Jackson and Diamond, 1996) which presumably permits relatively high digestion rates. As mentioned above, the digestive adjustmentsof mammals and birds to long-term acclimation to high feeding rate almost always include increased gut size and consequently increased amounts of digestive enzymes and
106
Physiological and ecological adaptations t o feeding in vertebrates
nutrient transporters (Karasov and Hume, 1997). Interestingly,the processing time of each meal, measured as mouth-to-cloaca total mean retention time (MRT; an index of turnover), and digestive efficiency do not change markedly (Dykstra and Karasov, 1992; McWilliams et al., 1999; Fig. 4.2). Probably, feedback mechanisms in the digestive tract ensure that the rate food enters .the intestine from the stomach and travels distally along the intestine does not exceed the rate at which it is broken down and absorbed. What permits higher food intake (inflow)even though turnover time is held fairly constant, is the larger volumetric capacity (Karasov, 1996). The primary instance in which NIRT is altered is when food richness is altered, in which case MRT changes in a corresponding fashion with the result that movement of digesta is matched to breakdown and absorption rates and digestive efficiency is maintained (Karasov, 1996). Thus, for these cases in which the intestine's rate of breakdown and absorption is limiting, if the feeding rate or food richness is to increase, then the biochemical features (enzyme levels, nutrient absorption rates) must be increased through an increase in activity per unit tissue or an increase in total amount of tissue. Both kinds of adjustments occur in mammals (Weiss et al., 1998) and birds (McWilliams and Karasov, 2001).This kind of integrated analysis of how the gut functions and adjusts has not been performed for the types of feeders whose intake is possibly limited by the rate of physical breakdown of the food (e.g.feeders on bivalves and crabs; see below). Whether the digestive capacity can be increased to match any demand put on it or whether the gut sometimes ultimately limits the energy budget is not known for most animals. The issue has been thoroughly studied in laboratory mice challenged during cold acclimation, lactation, and a combination of these factors (Hammond et al., 1994; Johnson and Speakman, 2001). With each increasing energetic challenge Swiss-Webstermice increased gastrointestinal mass and hydrolytic and absorptive capacity and, for the highest load of lactation in the cold, the energy budget limit was not set by the digestive system but more likely by lactational performance (Hammond et al., 1996).For the MF1 strain Johnson and Speakman (2001)doubted that even lactational performance was a limit, at least during a female's first lactation. They speculated that in that strain females may limit themselves during their first reproduction perhaps to maximize lifetime reproduction. How can we test whether the gut limits the energy budget for an animal in the field? The method used so far has been to measure the long-term limit in laboratory studies and compare it with the field energy budget. Studies on house wrens (Dykstra and Karasov, 1992, 1993) and yellow-eyed Juncos (Juncophaeonotus) (Weathers and Sullivan, 1989), for example, rejected the hypothesis that rate of digestion might limit brood size proximally because parental energy expenditure,measured with doubly labeled water, was below the longer term digestive capacity.
Digestive constraints in mammalian and avian ecology
107
Should we generalize from these results with laboratory mice and two passerine species and conclude that for all mammals and birds that digestive capacity can be increased to match any demand put on it and will not be linliting in the ecological setting? This would be premature we think. As described above, there are interesting plausible examples of digestive bottlenecks involving animals eating foods quite differentfrom the formulated laboratory chow fed to mice. There are other energy intensive points in the life cycle, such as growth (Karasov and Wright, 2002) and migration, during which digestion may prove to be the limiting factor in the energy budget. Also, there may be situations in which the immediate spare digestive capacity may be ecolo~callyimportant even if over the longer term digestive capacity increases and the long-term capacity is not limiting. For example, a gutlimitation hypothesis for many migratory birds suggests that the initially slow rate of mass gain at stopover sites occurs because birds lose digestive tract tissue and hence function during fasting, and rebuilding of the gut takes time and resources and itself restricts the supply of energy and nutrients from food (McWilliamsand Karasov, 2001). For birds that fly, the size of the digestive tract is likely ultimately limited by mass balance requirements for flight (i.e.big guts can't fly; Piersma and Gill, 1998).
FUTURE DIRECTIONS
(1) The idea of a digestive constraint is most plausible when food collection rate and digestion rate are both measured and the former is higher than the latter. Granivores are good candidates for such digestive limitation but to our knowledge no one has yet provided an example. (2) There is no published study for any vertebrate of both rapid and gradual adjustment of feeding and digestion to high energy demand that includes corresponding changes in gut size and biochemistry. (3) Rapid-adjustment experiments, rarely performed (only seven studies that we know of),are perhaps most interesting because they reveal the immediate spare digestive capacity of the animal.
(4) Granivores and insectivores eating heavily chitinized prey, provide interesting cases of possible digestivelimitation by mechanicalbreakdown and beg to be studied. (5) Estimation of whole-animal hydrolytic and absorptive capacity by invitro methodology has rarely been validated which undercuts their application for quantitative estimation of spare digestive capacity. (6) More integrative studies are needed that simultaneously measure adjustments in gut anatomy, retention time of digesta, enzyme hydrolysis
108
Physiological and ecological adaptations t o feeding i n vertebrates
rates, nutrient absorption rates, and digestive efficiency in response to changes in food quantity and quality. (7) Our understanding of the time course of digestive adjustment starting with turnover time of intestinal enzymes and epithelial cells and proceeding through rates of change of entire tissues to whole-animal feeding responses is based on very few studies. (8) Can wild species or much larger species achieve the increases in longterm digestive capacity achieved by small rodents such as laboratory mice (6.7times the digestion rate under routine conditions)? (9) Do laboratory mice reflect the norm or are they exceptional in their ability to match digestive capacity to any demand put on it? (10) More integrative studies are needed that compare immediate and longterm digestive capacity with rates of energy flow in free-living animals at energy intensive points in their life cycle. Are there other ways to test the hypothesis that digestion proximally limits energy budgets in the field? Acknowledgments
Supportedby N.S.F. (IBN-9723793,IBN-0216709 to W.H.K. and IBN-9984920 to S.R.M.).
Bednekoff P.A. and Houston A.I. 1994. Avian daily foraging patterns: effects of digestive constraints and variability. Euol. Ecol, in, 8: 36-52. Begon M., Harper J.L., and Townsend C.R. 1986. Ecology. Sinauer Assoc., Sunderland, MA. Berthold P. 1975. Migration: control and metabolic physiology. In: Avian Biology, D. S. Farner and J.R. King (Eds.), Acad. Press, New York, NY (pp. 77-128). Blem C. R. 1980. The energetics of migration. In: Animal Migration, Orientation, and Navigation, S.A. Gauthreaux (Ed.), Acad. Press, New York, NY, pp. 175-224. Buddington R.K. and Diamond J. M. 1990. Ontogenetic development of monosaccharide and amino acid transporters in rabbit intestine. Amer. 1. Physiol. 259: G544-G555. Buddington R.K. and Diamond J. 1992. Ontogenetic development of nutrient transporters in cat intestine. Amer. Physiol. 263: G605-G616. Buts J.P., Vivjerman V., Barudi C., De Keyser N., Maldague P., and Dive C. 1990. Refeeding after starvation in the rat: comparative effects of lipids, proteins and carbohydrates on jejunal an dileal mucosal adaptation. European 1. Clinical Investigation 20: 441-452. Caviedes-Vidal E. and Karasov W.H. 1996. Glucose and amino acid absorption in house sparrow intestine and its dietary modulation. Amer. 1. Physiol. 271: R561-R568. Chediack J.G., Caviedes-Vidal E., Karasov W.H., and Pestchanker M. 2001. Passive absorption of hydrophilic probes by house sparrows, Passer domesticus. 1. Exp. Biol. 204: 723-731. Dawson W.R., Marsh R.L., and Yacoe M.E. 1983. Metabolic adjustments of small passerine birds for migration and cold. Amer. 1. Physiol. 245: R755-R767.
Digestive constraints in mammalian and avian ecology
109
Debnam E.S., Karasov W.H., and Thompson C.S. 1988.Nutrient uptake by rat enterocytes during diabetes mellitus; evidence for an increased sodium electrochemical gradient. J. Physiol. (Lond.) 397: 503-512. Dekinga A., Dietz M.W., Koolhaas A., and Piersma T. 2001.Time course and reversibility of changes in the gizzards of red knots alternately eating hard and soft food. J. Exp. Biol. 204: 2167-2173. Diamond J. 1991.Evolutionary design of intestinal nutrient absorption: enough but not too much. News Physiol. Sci. 6: 92-96. Diamond J. and Hammond K. 1992.The matches, achieved by natural selection, between biological capacities and their natural loads. Experientia 48: 551-557. Diamond J., Karasov W.H., Phan D., and Carpenter EL. 1986. Hummingbird digestive physiology, a determinant of foraging bout frequency. Nature 320: 62-63. Drobney R.D. 1984. Effect of diet on visceral morphology of breeding wood ducks. Auk 101: 93-98. Dykstra C.R. and Karasov W.H. 1992.Changes in gut structure and function of house wrens (Troglodytes aedon) in response to increased energy demands. Physiol. Zool. 65: 422-442. Dykstra C.R. and Karasov W.H. 1993. Nesting energetics of House Wrens (Troglodytes aedonl in relation to maximal rates of energy flow. A u k 110: 481-491. Ferraris R.P. and Vinakota R.R. 1995.The time course of adaptation of intestinal nutrient uptake in mice is independent of age. J. Nutr. 125: 2172-2182. Gass C.L., Romich M.T., and Suarez R.K. 1999. Energetics of hummingbird foraging at low ambient temperature. Canadian J. Zool. 77: 314-320. Gross J.E., Hobbs N.T., Spalinger D.E., and Wunder B.A. 1993. Functional response of herbivores in food-concentrated patches: tests of a mechanistic model. Ecology 74: 778-791. Guillemette M. 1994. Digestive-rate constraint in wintering common eiders (Somateria mollissima): implications for flying capabilities. A u k 111: 900-909. Guillemette M. 1998.The effect of time and digestion constraints in Common Eiders while feeding and diving over blue mussel beds. Funct. Ecol. 12: 123-131. Hammond K. and Diamond J. 1997.Maximal sustained energy budgets in humans and animals. Nature 386: 457-462. Hammond K., Lloyd K.C.K., and Diamond J. 1996. Is mammary output capacity limiting to lactational performance in mice? J. Exp. Biol. 199: 337-349. Hammond K.A., Konarzewski M., Torres R.M., and Diamond J. 1994.Metabolic ceilings under a combination of peak energy demands. Physiol. Zool. 67: 1479-1506. Hanski I. 1984. Food consumption, assimilation and metabolic rate in six species of shrew (Sorex and Neomys). A n n . Zool. Fennici 21: 157-165. Hodin R.A., Graham J.R., Meng S., and Upton M.P. 1994.Temporal pattern of rat small intestinal gene expression with refeeding. Amer. J. Physiol. 266: G83-G89. Holling C.S. 1959.The components of predation as revealed by a study of small mammal predation of the European pine sawfly. Can. Entom. 91: 293-320. Jackson S. and Diamond J. 1995. Ontogenetic development of gut function, growth, and metabolism in a wild bird, the Red Jungle Fowl. Amer. J. Physiol. 269: R1163-R1173. Jackson S. and Diamond J. 1996.Metabolic and digestive responses to artificial selection in chickens. Evolution, 50: 1638-1650. Jeschke J.M., Kopp M., and Tollrian R. 2002.Predator functional responses: discriminating between handling and digesting prey. Ecol. Mono. 72: 95-112. Johnson M.S. and Speakman J.R. 2001. Limits to sustained energy intake. V. Effect of cold-exposure during lactation in M u s musculus. J. Exp. Biol. 204: 1967-1977. Jumars P.A. and Martinez del Rio C. 1999. The tau of continuous feeding on simple foods. Physiol, and Biochem. Zool. 72: 633-641. Karasov W.H. 1986. Energetics, physiology, and vertebrate ecology. Trends Ecol. and Evol. 1: 101-104.
110
Physiological and ecological adaptations t o feeding in vertebrates
Karasov W.H. 1990. Digestion in birds: chemical and physiological determinants and ecological implications. Studies Avian Biol. 13: 391-415. Karasov W.H. 1996. Digestive plasticity in avian energetics and feeding ecology. In: Avian Energetics and Nutritional Ecology C. Carey (ed.), Chapman and Hall, New York, NY, (pp. 61-84). Karasov W.H. and Cork S.J. 1994. Glucose absorption by a nectarivorous bird: the passive pathway is paramount. Amer. J. Physiol. 267: G18-G26. Karasov W.H. and Diamond J.M. 1983. A simple method for measuring solute uptake in vitro. J. Comp. Physiol. B152: 105-116. Karasov W.H. and Diamond J.M. 1987. Adaptation of intestinal nutrient transport. In: Physiology of the Gastrointestinal Tract L.R.Johnson (ed.). Raven: New York. NY, pp. 1489-1497. Karasov W.H. and Hume I.D. 1997. Vertebrate gastrointestinal system. In: W. Dantzler (ed.). Handbook of Comparative Physiology Amer. Physiol. Soc. Bethesda, MD, pp. 409-480. Karasov W.H., Phan D., Diamond J.M., and Carpenter F.L. 1986. Food passage and intestinal nutrient absorption in hummingbirds. Auk 103: 453-464. Karasov W.H. and Pinshow B. 2000. Test for physiological limitation to nutrient assimilation in a long-distance passerine migrant at a springtime stopover site. Physiol. Biochem. Zool. 73: 335-343. Karasov W. H., Pinshow B., Starck J. M. and Afik D. 2004. Anatomical and histological changes in the alimentary tract of migrating blackcaps (Syliva atricapilla): A comparison among fed, fasted, food-restricted and refed birds. Physiol. Biochem. Zool. 77: 149-160. Karasov W.H. and Wright J. 2002. Nestling digestive physiology and begging. In: The Study of Nestling Begging: Current State and Future Growth J. Wright and M.Leonard (eds.), Kluwer Acad. Publ. Dordrecht, Netherlands. Kehoe F.P., Ankney C.D., and Alisauskas R.T. 1988. Effects of dietary fiber and diet diversity on digestive organs of captive mallards (Anas platyrhynchos). Can. J. Zool. 66: 1597-1602. Kellett G . and Helliwell P.A. 2000. The diffusive component of intestinal glucose absorption is mediated by the glucose-induced recruitment of GLUT2 to the brushborder membrane. Biochem. J. 350, 155-162. Kendeigh S.C. 1949. Effect of temperature and season on energy resources of the English Sparrow. A u k 66: 470-487. Kenward R.E. and Sibly R.M. 1977. A woodpigeon (Columbia palumbus) feeding preference explained by a digestive bottleneck. J. Appl. Ecol. 14: 815-826. Kersten M. and Visser W. 1996. The rate of food processing in the oystercatcher: food intake and energy expenditure constrained by a digestive bottleneck. Func. Ecol. 10: 440-448. Kirkwood J.K. 1983. A limit to metabolizable energy intake in mammals and birds. Comp. Biochem. Physiol. 75A: 1-3. Kleiber M. 1933. Tiergrope und Futterverwertung. Tierernahrung 5: 1-12. Kleiber M. 1961. The Fire of Life. John Wiley & Sons, New York, NY. Kontogiannis J.E. 1968. Effect of temperature and exercise on energy intake and body weight of the white-throated sparrow, Lonotrichia albicollis. Physiol. Zool. 41: 54-64. Kvist A. 2001. Fuel and fly: adaptations to endurance exercise in migrating birds. Lund, Sweden: PhD diss. University Lund, Lund, Sweden. Kvist A. and Lindstrom A. 2000. Maximum daily energy intake: it takes time to lift the metabolic ceiling. Physiol. Biochem. Zool. 73: 30-36. Lee K.A., Karasov W.H., and Caviedes-Vidal E. 2002. Digestive responses to restricted feeding in migratory Yellow-rumped warblers. Physiol. and Biochem. Zool. 75: 314-323. Levey D.J. and Karasov W.H. 1989. Digestive responses of temperate birds switched to fruit or insect diets. Auk 106: 675-686.
Digestive constraints in mammalian and avian ecology
111
Lilja C., Sperber I., and Marks H.L. 1985. Postnatal growth and organ development in Japanese quail selected for high growth rate (Coturnix coturnix japonica). Growth 49: 51-62. Lindstrom A. and Kvist A. 1995. Maximum energy intake rate is proportional to basal metabolic rate in passerine birds. Proc. Roy. Soc. (London) B 261: 337-343. Martinez del Rio C., Schondube J.E., McWhorter T.J., and Herrera L.G. 2001. Intake responses in nectar feeding birds: digestive and metabolic causes, osmoregulatory consequences, and coevolutionary effects. Amer. Zool. 41: 902-915. McWhorter T.J. and Martinez del Rio C. 2000. Does gut function limit hummingbird food intake? Physiol. Biochem. Zool. 73: 313-324. McWilliams S.R. and Karasov W.H. 1998a. Test of a digestion optimization model: effect of variable-reward feeding schedules on digestive performance of a migratory bird. Oecologia 114: 160-169. McWilliams S.R. and Karasov W.H. 1998b. Test of a digestion optimization model: effects of costs of feeding on digestive parameters. Physiol. Zool. 71: 168-178. McWilliams S.R. and Karasov W.H. 2001. Phenotypic flexibility in digestive system structure and function in migratory birds and its ecological significance. Comp. Biochem. Physiol. 128A: 579-593. McWilliams S. R. and Karasov W. H. 2002. Spare capacity in the digestive system of a migratory songbird and its ecological significance. Integ. Comp. Biol. 42: 1277. McWilliams S.R. and Raveling D.G. 2004. Energetics and time allocation of cackling Canada geese during spring. In: Proc. International Canada Goose Symposium R.Lien (ed.). Madison, WI. McWilliams S.R., Caviedes-Vidal E., and Karasov W.H. 1999. Digestive adjustments in Cedar Waxwings to high feeding rate. I. Exp. Zool. 283: 394-407. Miller M. 1975. Gut morphology of mallards in relation to diet quality. 1. Wildlife Mgrnt. 39: 168-173. Mook L.J. 1963. Birds and the spruce budworm. In: The Dynamics of Epidemic Spruce Budworm Populations R.F. Morris (ed.). Mem. Entom. Soc. Can. 31. (pp. 268-271). Moss R. and Parkinson J.A. 1972. Digestion of heather (Calluna vulgaris) by red grouse (Lagopus lagopus scoticus). Brit. I. Nutr. 27: 285-298. Pappenheimer J.R. 1998. Scaling of dimensions of small intestines in non-ruminant eutherian mammals and its significance for absorptive mechanisms. Comp. Biochem. Physiol. A 121: 45-58. Piersma T. and Gill R.E.J. 1998. Guts don't fly: small digestive organs in obese bartailed godwits. A u k 115: 196-203. Piersma T., Koolhaas A., and Dekinga A. 1993. Interactions between stomach structure and diet choice in shorebirds. A u k 110: 552-564. Savory C.J. and Gentle M.J. 1976a. Changes in food intake and gut size in Japanese quail in response to manipulation of dietary fiber content. Brit. I. Poultry Sci. 17: 571-580. Savory C.J. and Gentle M.J. 1976b. Effects of dietary dilution with fibre on the food intake and gut dimensions of Japanese quail. Brit. I. Poultry Sci. 17: 561-570. Smith M.W., Paterson J.Y.F., and Peacock M.A. 1984. A comprehensive description of brush border membrane development applying to enterocytes taken from a wide variety of mammalian species. Comp. Biochem. Physiol. 77A: 655-662. Spalinger D.E., Hanley T.A., and Robbins C.T. 1988. Analysis of the functional response in foraging in the Sitka black-tailed deer. Ecology 69: 1166-1175. Spalinger D.E., Robbins C.T., and Hanley T.A. 1986. The assessment of handling time in ruminants: the effect of plant chemical and physical structure on the rate of breakdown of plant particles in the rumen of mule deer and elk. Can. I. Zool. 64: 312-321. Starck J.M. 1996. Intestinal growth in the altricial European starling (Sturnus vulgaris) and the precocial Japanese quail ( C o t u r n i x c o t u r n i x japonica). A morphometric and cytokinetic study. Acta Anat.(Basel) 156: 289-306. Starck J.M. 1999a. Phenotypic flexibility of the avian gizzard: rapid, reversible and repeated changes of organ size in response to changes indietary fibre content. 1. Exp. Biol. 202: 3171-3179.
112
Physiological and ecological adaptations t o feeding in vertebrates
Starck J.M. 1999b. Structural flexibility of the gastro-intestinal tract of vertebratesimplications for evolutionary morphology. Zool. A n z . 238: 87-101. Starck J.M., Karasov W.H., and Afik D. 2000. Intestinal nutrient uptake measurements and tissue damage. Validating the everted sleeves method. Physiol. Biochem. Zool. 73: 454-460. Taylor C.R. and Weibel E.R. 1981. Design of the mammalian respiratory system. I. Problem and strategy. Respir. Physiol. 44: 1-10. Toloza E.M. and Diamond J. 1992. Ontogenetic development of nutrient transporters in rat intestine. Amer. J. Physiol. 263: G593-G604. Toloza E.M., Lam M., and Diamond J. 1991. Nutrient extraction by cold-exposed mice: A test of digestive safety margins. Amer. J. Physiol. 261: G608-G620. Ugolev A.M., Zaripov B.Z., Iezuitova N.N., Gruzdkov A.A., et al. 1986. A revision of current data and views on membrane hydrolysis and transport in the mammalian small intestine based on a comparison of techniques of chronic and acute experiments: experimental re-investigation and critical review. Comp. Biochem. Physiol. 85A: 593-612. Uhing M.R. and Kimura R.E. 1995. The effect of surgical bowel manipulation and anesthesia on intestinal glucose absorption in rats. J. Clin. Invest. 95: 2790-2798. Van Soest P.J. 1994. Nutritional Ecology of the Ruminant. Cornell Univ. Press, Ithaca, NY. Weathers W.W. and Sullivan K.A. 1989. Juvenile foraging proficiency, parental effort, and avian reproductive success. Ecol. Monog. 59: 223-246. Weber E., Ehrlein J. 1998. Reserve capacities of the small intestine for absorption of energy. A m . J. Physiol. 275: R300-R307. Weibel E.R. 2000. Symmorphosis, on Form and Function Shaping Life. Harvard Univ. Press, Cambridge, MA, (USA). Weiner J. 1987. Limits to energy budget and tactics in energy investments during reproduction in the Djungarian hamster (Phodopus sungorus sungorus Pallas 1770). Symp. Zool. Soc. Lond. 57: 167-187. Weiner J. 1992. Physiological limits to sustainable energy budgets in birds and mammals: ecological implications. Trends Ecol. Evol. 7: 384-388. Weiss S.L., Lee E.A., and Diamond J. 1998. Evolutionary matches of enzyme and transporter capacities to dietary substrate loads in the intestinal brush border. Proc. Natl. Acad. Sci. 95: 2117-2121. West G.C. 1960. Seasonal variation in the energy balance of the tree sparrow in relation to migration. A u k 77: 306-329. Winter Y. 1998. In vivo measurement of near maximal rates of nutrient absorption in a mammal. Comp. Biochem. Physiol. 119A: 853-859. Zwarts L. and Dirksen S. 1990. Digestive bottleneck limits the increase in food intake of whimbrels preparing for spring migration from the Banc D'Arguin, Mauritania. Ardea 78: 257-278. Zynel C.Y. and Wunder B.A. 2002. Limits to food intake by the Prairie Vole: effects of time for digestion. Funct. Ecol. 16: 58-66.
Paracellular Intestinal Absorption of Carbohydrates in Mammals and Birds Todd J. McWhorter Department of Wildlife Ecology, University of Wisconsin, Madison, WI, USA
SYNOPSIS In most animals and for most dietary substrates, digestion comprises two steps: hydrolysis into smaller molecules and absorption across the gut wall. This chapter deals with the latter of these inextricably interrelated processes. Intestinal absorption and paracellular permeability are subjects of considerable interest, both because they determine capacities for nutrient uptake and because they have important consequences for exposure to natural and man-made water-soluble compounds including drugs, toxins, and toxicants. These subjects are reviewed in the context of our current understanding of intestinal carbohydrate absorption and some important, and sometimes controversial, differences among mammals and birds are highlighted. A short description of mediated carbohydrate absorption, outlining the roles played by specific transporters, is given and the modulation of solute uptake capacities is briefly discussed. The heated and ongoing debate regarding the relative importance of paracellular uptake (via diffusion and solvent drag) as a pathway for carbohydrate absorption in mammals is taken up next, followed by a discussion of the current evidence regarding paracellular absorption of carbohydrates in birds, and a critical examination of the experimental approaches that have been used in resolving the relative importance of this process. Lastly recent studies correlating functional and mechanistic aspects of paracellular flux in birds are reviewed.
INTESTINAL ABSORPTION OF CARBOHYDRATES Epithelial uptake of carbohydrates in the intestine is known to occur through both carrier-mediated (active or facilitative passive) and nonmediated mechanisms (Hopfer, 1987). Carrier-mediated transport is effected by
114
Physiological and ecological adaptations t o feeding i n vertebrates
specific membrane-associated carriers, and its rate follows saturation kinetics. By contrast, the rate of nonmediated passive uptake in the intestine varies linearly with solute concentration and does not obey saturation kinetics. This chapter commences with a discussion of each of these processes in turn and a brief outline of the modulation of solute uptake capacities. Carrier-mediated Transport As early as the 1960s it was recognized that active, carrier-mediated processes were important in carbohydrate absorption (Crane, 1960). The transcellular route for active D-glucose absorption involves carrier-mediated transport at both the lumenal (brush-border) and basolateral membranes. The concept of Na+-gradient-driven, secondary active transport was contrived by Crane et al., (1961) to explain concentrative absorption of D-glucose in the intestine. This insight was particularly remarkable, considering the paucity of data on cytosolic composition and intracellular processes available at that time (Hopfer, 1987).Crane et al. (1961)correctly inferred that the concentrative step is located at the brush-border membrane and is a result of cotransport with sodium. The specific protein transporter responsible was cloned by Wright and colleagues in 1987 and is now known as SGLTl (Hediger et al., 1987).It is a high-affinity, low-capacity transporter in which D-glucose translocation is coupled to the cotransport of two Na' (Hediger, 1994).The absorption of D-glucose across intestinal epithelial cells occurs against a concentration gradient by SGLTl in the brush-border membrane. The Na+/K+-ATPase(Skou and Esmann, 1992) located in the basolateral membrane maintains the electrochemical gradient required to drive uphill glucose transport. The other well-characterized member of the SGLT family is a low-affinity, high capacity brush-border Na+-glucose cotransporter (SGLT2, stoichiometry 1 Na' : 1 glucose) that is principally involved in glucose reabsorption in the kidney (Kanai et al., 1994). SGLT2 has not been documented in the intestine (Halaihel et al., 1999). D-glucose moves out of intestinal epithelial cells via GLUT2 in the basolateral membrane (Hediger, 1994). GLUT2 is a low-affinity member of the GLUT family of facilitated glucose cotransporters that permit movement of glucose across plasma membranes down its concentration gradient (Thorens, 1993;Hediger, 1994). Recent studies in perfused rat jejunum have described a mechanism for gut glucose (and fructose) absorption that involves GLUT2-mediated facilitated diffusion in the apical membrane (Helliwell et al., 2000a; Kellett and Helliwell, 2000; Kellett, 2001). This mechanism is initiated by the active transport of glucose via SGLT1, which causes the translocation of GLUT2 to the lumenal surface of the gut. Regulation appears to be via protein kinase C-dependent and mitogen-activatedprotein kinase signaling pathways (see Kellett, 2001 for review). This novel finding has direct bearing on the debate regarding paracellular glucose absorption (see below).
Paracellular carbohydrate absorption
115
Some researchers have suggested that a heterogeneity of D-glucose transport systems exists in the intestinal epithelial brush-border (Halaihel et al., 1999 and references therein). The proposed second active transport system, dubbed system 2 (S2), is different from SGLTl and basolateral GLUT2. Halaihel et al. (1999) used pig jejunal brush-border membrane vesicles to confirm that S2 is a low-affinity, high-capacity, D-glucose and D-mannose transporter, distinct from any previously known. The existence of 52 is somewhat controversial, having been questioned by researchers who affirm that a single system (i.e. SGLTl plus some passive up take) completely explains glucose uptake in the intestine (Malo, 1993; Hediger and Rhoads, 1994). Halaihel et al. (1999) carefully reexamined the heterogeneity question to determine whether previous experimental procedures could have caused the apparent presence of nonexistent transport systems, and to determine the minimum range of substrate concentrations needed to correctly fit saturation curves. Their reanalysis indicated that D-glucose contaminated with D-sorbitol cannot cause the spurious existence of transport systems (a criticism of Malo, 1993) and that a large range of substrate concentrations (2 50 mM) is necessary to correctly distinguish the kinetics of D-glucose transport processes (Halaihel et al., 1999). They affirmed that their results distinguish two kinetically distinct systems: high-affinity, low-capacity SGLTl and low-affinity, high capacity S2, which is not believed to be a member of the SGLT family. The debate continues and it remains to be seen whether a new transporter will be cloned (see for example, Doege et al., 2000). It is possible that this proposed S2 relies on the recruitment of GLUT2 to the brush-border membrane (see above). Regardless, most of the active D-glucose uptake at low concentrations appears to mediated by SGLTl (Halaihel et al., 1999). Fructose transport by the brush-border epithelia of the mammalian intestine is accomplished by another member of the GLUT family, GLUT5 (Burant et al., 1992; Rand et al., 1993). GLLTT5 is a sodium-independent, facilitative transporter that has a very poor affinity for D-glucose (Rand et al., 1993and references therein). Fructose absorption appears to be much more concentration-dependent than D-glucose transport (Holdsworth and Dawson, 1964), although there has been some suggestion of active fructose transport associated with sucrose digestion and disaccharidase activity (Shi et al., 1997and references therein). Transport of fructose across the basolateral membrane (and possibly also across the brush-border membrane, see above and Helliwell et al., 2000b) is carried out by the facilitative glucose transporter GLUT2, which transports galactose as well (Burant et al., 1992). Functional studies have demonstrated significant differences in the biochemical properties of rat and human GLLTT5, so it is possible that there are interspecific differencesin the regulation of dietary fructose uptake (Rand et al., 1993).
116
Physiological and ecological adaptations t o feeding in vertebrates
Modulation of Solute Uptake Capacities Regulation of solute uptake capacities has been a well-known phenomenon for nearly 70 years (Ferraris et al., 1990). It has been described under a wide range of conditions, including pregnancy, lactation, exercise, exposure to a cold environment, and varying dietary carbohydrate levels (reviewed by Karasov and Diamond, 1983, see also Chapter 4 by Karasov and McWilliams, this volume; Ferraris and Diamond, 1989).Several functional considerations predict the pattern of uptake capacity regulation (termed the "adaptive modulation hypothesis" Karasov and Diamond, 1983; Karasov, 1992). Two of these considerations are relevant to the discussion here of carbohydrate absorption. First, transporters should be repressed if the biosynthetic costs of producing and maintaining the transporter exceed the benefits that the transporter provides. Second, transporters for nutrients that yield calories should be up-regulated by their substrates since metabolizable nutrients yield calories in proportion to amount of nutrient. Indeed, it has repeatedly been shown in a wide variety of animals that dietary glucose levels modulate uptake capacity (Karasovand Diamond, 1983; Diamond and Karasov, 1984; Ferraris and Diamond, 1989; Karasov, 1992).Similar modulations are known to occur as the result of varying dietary fructose levels (Ferraris and Diamond, 1989; Corpe et al., 1996; Shu et al., 1997). Carbohydrate uptake appears to be regulated primarily in those species that encounter significant and varying carbohydrate levels in their diets (Ferraris and Diamond, 1989; Afik et al., 1995). Such modulation is physiologically important in overcoming potential digestive bottlenecks (Ferrarisand Diamond, 1989)while avoiding waste of biosynthetic activity on unutilizable capacity (Diamond, 1991). The dietary modulation of transport capacity has recently been extensively studied in birds (e.g. Afik et al., 1995; Martinez del Rio et al., 1995; Levey et al., 1999). Nonmediated Uptake of Carbohydrates The paracellular component of the epithelial barrier is the pathway between adjacent epithelial cells. Transport across this pathway is restricted by the junctional complex and the lateral intercellular spaces (Ballard et al., 1995). The most important component of the junctional complex for restricting passage of small solutes through the paracellular pathway appears to be the tight-junction (Anderson and van Itallie, 1995;Anderson, 2001).This barrier is created where transmembrane protein strands from adjacent cells converge in the paracellular space (Tsukita and Furuse, 2000).There is thought to be a positive correlation between the number of strands and paracellular electrical resistance (the inverse of permeability).Claudins have been identified as the major protein constituents of tight-junction strands and these strands appear to be reorganized very dynamically (Sasaki et al., 2003). The standing osmotic gradient theory for solute and liquid absorption proposes that active transport of glucose and sodium drives paracellular liquid
Paracellular carbohydrate absorption
117
uptake by establishing a solute osmotic gradient across the enterocytes (Diamond and Bossert, 1967).Glucose is actively transported across the brushborder (apical)membrane and then moves across the basolateral membrane into the intercellular space (see above and Hediger, 1994). Sodium is actively transported across the basolateral membrane by the Na+/K+-ATPase.This buildup of osmotically active solutes drives net absorption of water (and theoretically accompanying nutrients via solvent drag) across intestinal epithelia (Diamond and Bossert, 1967). It is presently unclear whether water moves predominantly across cell membranes or through tight-junctions (Ballard et al., 1995), but see discussion below. In contrast with carriermediated transport, the rate of passive transport varies linearly with solute concentration and does not obey saturation kinetics (see Chap. 3, Fig. 3.3b). RELATIVE IMPORTANCE OF PARACELLULAR CARBOHYDRATE UPTAKE I N MAMMALS
Until just recently it was widely accepted that carbohydrate transport in the mammalian intestine occurs principally through active, carrier-mediated mechanisms (Crane, 1960,1975;Hopfer, 1987).Pappenheimer and coworkers proposed in the late 1980s that most glucose uptake occurs passively via the paracellular pathway instead (Madara and Pappenheimer, 1987; Pappenheimer, 1987;Pappenheimer and Reiss, 1987; Pappenheimer, 1990). This hypothesis was based on the observation that the permeability of the small intestine epithelia was increased when exposed to high concentrations of glucose in the lumen. It was controversial because it challenged conventional theories of intestinal solute and water absorption. The subject has been one of considerable debate ever since. In this section the evidence presented by Pappenheimer and coworkers in support of the notion that paracellular flux is the major pathway for the absorption of hexoses and other hydrosoluble nutrients is outlined, followed by a discussion of several criticisms of their hypothesis and a review of recent evidence bearing on this controversial subject. In a series of three papers published in 1987, Pappenheimer and colleagues presented evidence in support of their novel hypothesis (Madara and Pappenheimer, 1987; Pappenheimer, 1987; Pappenheimer and Reiss, 1987). The main tenet of this hypothesis is that glucose and amino acids taken up by enterocytes serve as intracellular signals to decrease tight-junction integrity and thereby facilitate the absorption of hydrophilic solutes by paracellular bulk flow. Based on the clearances of inert solutes (steady-state transepithelial fluxes per unit concentration) from the small intestines of anesthetized animals, Pappenheimer and Reiss (1987) calculated the fraction of fluid absorption which passes paracellularly. The addition of 25 mM
118
Physiological and ecological adaptations to feeding in vertebrates
D-glucose to the perfusion solution of an in vivo rat preparation doubled the estimated rate of paracellular liquid absorption (to approximately 50% of total). They concluded that above a lumenal concentration of 250 mM D-glucose (which may or may not be exceedingly high, see below), paracellular solvent drag is the principal route for intestinal absorption of glucose. They further proposed that Na+-coupledtransport of organic solutes from the lumen to intercellular spaces provides the principal osmotic force for fluid absorption and modifies tight-junction permeability. Pappenheimer (1987)addressed the question of whether Na+-coupledtransport increases solvent drag solely by its effects on fluid absorption, or whether it also opens epithelial.junctions. He found that the addition of small concentrations of D-glucose or amino acids to isolated intestinal segments greatly decreased transepithelial impedances. Impedance was interpreted in terms of junctional and lateral space resistance and surface area of basolateral membranes (i.e.decreased impedance = decreased resistance = increased permeability). The addition of 25 mM D-glucose decreased lateral space resistance by more than 50%. These results provided support for the theory that the surface area of lateral membranes and dimensions of epithelial cell junctions are regulated by the concentrationof nutrients in the intestinal lumen (Pappenheimer, 1987). Pappenheimer (1987) further supposed that Na+coupled solute transport triggers contraction of circumferential actomyosin fibers in the terminal web of the microvillar cytoskeletal system, thereby pulling apart tight-junctions to allow the paracellular absorption of solutes by solvent drag. Madara and Pappenheimer (1987)explored the structural correlates of the impedance and permeability changes induced by glucose and amino acids. Profound changes in ultrastructure were observed using light and electron microscopy and freeze-fracture techniques. Specifically, the addition of D-glucose or amino acids induced junctional dilation, expansion of lateral spaces, and condensation of actomyosin in the perijunctional ring. The authors interpreted these observations as providing support for the hypothesis that Na+-coupledtransport triggers the physical opening of intercellular channels and thus facilitates paracellular flux (Madara and Pappenheimer, 1987). Pappenheimer (1990) emphasized the importance of paracellular glucose transport by comparing the daily carbohydrateintake of several species with maximum predicted rates of transcellular Na+-coupledtransport. He estimated that total ingestion of sugars exceeds the capacity for active transcellular transport threefold in rats and fivefold in rabbits. The discrepancy between average rates of ingestion and maximum rates of transport increased exponentially with body weight. In humans, the ingestion-absorptionrate was predicted to be 10-20times greater than active transport. The discrepancy between these estimates of maximal active transport and ingested carbohydrate loads, and the recovery of >8O0/0of [3H]mannitol (a marker supposedly not absorbed by any carrier-mediatedprocess) in urine
Paracellular carbohydrate absorption
119
or other body fluids, led Pappenheimer to reiterate the conclusion that paracellular transport plays the major role in intestinal nutrient absorption (Pappenheimer, 1990). In fact, according to the hypotheses of Pappenheimer and coworkers, the primary role of Na+-coupled transport of solutes is to provide the osmotic force for convective flow and to effect the decrease in epithelial permeability necessary for increased paracellular flux (Pappenheimer and Reiss, 1987; Pappenheimer, 1990). Recent experiments using phloridzin blocking in dogs (Pencek et al., 2002; Pencek et al., 2003) confirm that SGLT1-mediated glucose uptake is indeed required to activate paracellular absorption of glucose; however, these studies also suggest that paracellular absorption is a relatively minor component of total uptake (see below). As is the case with any hypothesis that challenges conventional thought, there has been intense scrutiny of Pappenheimer's work. Some of these criticisms are discussed below. The alternative view to paracellular uptake, advanced by Ferraris and Diamond (1989; 1997),is that adaptation of brush-border membrane carbohydrate transporters (i.e.SGLT1, GLUT5, and based on recent evidence possibly the recruitment of GLUT2) is matched to dietary intake. A significant body of literature reports measurements of intestinal glucose transport as a function of lumenal concentration (usually based on perfusions). Knowledge of the glucose concentrations normally present in the small intestine of animals is essential for assessing the physiological meaning of these measurements. Ferraris et al. (1990) criticized previous work reporting exceptionally high lumenal glucose concentrations and subsequent studies that have based their conclusions on such observations. They argued that knowing the true physiological glucose concentration in the lumen is critical for determining: (1)whether or not the intestine possesses reserve absorptive capacity, (2)whether regulation of glucose transport at the membrane level has physiological significance, (3) the predominant mechanism of glucose absorption (i.e. paracellular vs. transcellular uptake), and lastly (4) the significance of Michaelis-Menten constants (K,) reported for glucose transport (Ferraris et al., 1990). Lumenal glucose concentrations in the small intestine were widely assumed to be in the range of 50-500 mM (Crane, 1975; Alpers, 1987; Pappenheimer and Reiss, 1987). Ferraris et al. (1990) pointed out that although reports of lumenal glucose concentrations appear to be unanimous, these exceptionally high values have several surprising implications. First, they imply that lumenal contents are very hypertonic, because in addition to glucose the lumen contains many other osmotically active substances (ions, amino acids, etc.). Since the small intestine is extremely permeable to water (Fromter and Diamond, 1972; Chang and Rao, 1994),this high concentration would presumably be quickly diluted to isotonicity. Second, the reported prevailing high glucose concentrations are nearly two orders of magnitude higher than Kmvalues for glucose absorption measured in vivo or in vitro. The authors suggested that this would be unusual, as most
120
Physiological and ecological adaptations t o feeding in vertebrates
enzyme and transporter Kmvalues are in the range of prevailing substrate concentrations, permitting the efficient regulation of transport rates through modulation of Km(Hochachkaand Somero, 1984; Ferraris et al., 1990). Third, the quoted values imply that the small intestine's capacity for glucose absorption is up to two orders of magnitude higher than daily intakes. Ferraris et al. (1990) argued that this is also unusual, based on the observation that of enzymes tend to be in the range of premaximal reaction velocities (VmaX) vailing fluxes. However, if pa&kellular absorption is indeed a significant route of absorption (Pappenheimer, 1990),lumenal concentrations and abof an enzyme syssorptive fluxes considerably higher than the Kmand Vmax tem respectively,would not be surprising. Fourth, the reported values would make upregulation of glucose absorption pointless, since absorptive capacity would already be present in enormous excess. Modulation of absorptive capacity in response to energetic demands or dietary substrate concentration has been observed in many animals (Karasov and McWilliams, Chapter 4 this volume). Finally the authors argued that cited values imply that the contribution of carrier-mediated transport to total glucose absorption is increasingly minor as lumenal concentrations increase. This is in fact exactly the point made by Pappenheimer (1990; also see above), based on his estimates of maximal uptake and review of lumenal glucose concentrations (mostly > 250 mM). These implications motivated Ferraris et al. (1990) to measure glucose concentrations in the gut under "physiological" conditions. They allowed animals to consume diets typical to their species, and also created artificial, supraphysiological diets. Lumenal glucose concentrations ranged from 0.2 mM to 48 mM under all physiological conditions and did not exceed 100 mM even in rats fed 65% (wt/wt) glucose diets. Because concentrations were found to be relatively low, they concluded that glucose contributes only a small percentage of total lumenal osmolality. Direct measurements of lumenal osmolality confirmed that even at their peak values were only moderately hypertonic. The lumenal glucose concentrations observed by Ferraris et al. (1990) were considerably lower than those assumed by earlier studies (Crane, 1975; Pappenheimer and Reiss, 1987). They resolve this discrepancyby pointing out that earlier studies often infused high concentrations of glucose directly into the gut, used nonspecific assay methods and omitted controls. Reported lumenal glucose concentrationshave declined dramatically as a function of year of publication (i.e.with the use of modern specific assays and controls, Ferraris et al., 1990). Studies using modern methods (Olsen and Ingelfinger, 1968;Murakami et al., 1977; Ilundain et al., 1979)show lumenal glucose concentrations that agree well with the measurements of Ferraris et al. (1990). These much lower levels eliminate the potential of unlikely osmotic scenarios in the small intestine and also answer the apparent paradoxes presented by high lumenal glucose concentrations. Apparent Kmvalues (uncorrectedfor the effects of unstirred layers, USL, see Meddings and
Paracellular carbohydrate absorption
121
Westergaard, 1989) measured by Ferraris et al. (1990) were between 6 mM and 23 mM in vivo, values which accord well with prevailing lumenal glucose concentrations reported in mammals. The conclusion that lumenal glucose concentrations in mammals were not as high under normal conditions as previously believed (> 250 mM) undermined Pappenheimer's arguments that paracellular transport is the primary route for intestinal absorption of hydrosoluble nutrients. Pappenheimer (1990)may have also underestimated maximal active transport capacities. His calculations of maximal active transport rates were based on assumptions regarding the uniformity of uptake capacity throughout the intestine. It is unlikely these assumptions hold across all species examined (see McWhorter and Lopez-Calleja, 2000). In addition, Uhing and Kimura (199513) showed that in studies using anesthesia and surgical manipulation in rats active glucose transport may be inhibited by up to 86%. Ferraris et al. (1990) estimated that carrier-mediated D-glucose uptake capacity exceeds glucose intake by only about twofold in rats fed normal diets. Since there appears to be only a modest excess of uptake capacity over intake, upregulation serves to match capacity to demand (see discussion of adaptive modulation above). The considerable adaptability for glucose uptake shown by vertebrate intestines would be superfluous if such a grossly enormous margin in uptake capacity existed. Pappenheimer (1993)argued that the adaptive modulation of glucose transport capacity does not refute a major role for paracellular transport of nutrients. He contended that these changes alter absorptive capacity by regulating the permeability of tightjunctions and the concentration gradients available to drive fluid and nutrient uptake. Until it is possible to distinguish between active regulation of tight-junctions via cytoskeletal contraction and secondary dilation due to transported fluid, or the alteration of tight-junction structure for other purposes (e.g. translocation of GLUT2 to the apical membrane or immune-related responses) this argument is not testable. The conclusions of Pappenheimer and Reiss (1987) regarding the importance of paracellular flux were based on physiological glucose concentrations they believed to be >250 mM. In a later paper Pappenheimer states explicitly that measurements of glucose flux in vivo show that passive flux begins to exceed active transport at concentrations in the range of 25 to 50 mM (Pappenheimer, 1990 and references therein). Because of their finding of much lower lumenal glucose concentrations, Ferraris et al. (1990) argued that the conclusion of a significant paracellular component to glucose absorption is unwarranted under normal conditions. They did concede, however, that in animals fed supraphysiological glucose loads paracellular transport may play a significant role. Pappenheimer (1993) claimed that the lumenal glucose concentrations measured by Ferraris et al. (1990)are rnisleadmg because most glucose present in the small intestine is liberated from a-limit dextrins and disaccharides
122
Physiological and ecological adaptations t o feeding in vertebrates
cleaved by membrane-bound enzymes. His argument was that the glucose concentration in the pericellular USL may be far greater than measured in the intestinal chyme and may exceed 300 mM close to the enterocytes (Pappenheimer, 1993,1998). Ballard et al. (1995)estimated that if this concentration of glucose was present in the USL, the osmolality could be up to 600 mOsm(kg H20)-',a value about twice that of plasma or intestinal chyme (Ferraris et al., 1990).They reiterated that the maintenance of such a large osmotic gradient is unlikely given the high osmotic water permeability of the intestinal epithelial barrier (Fromter and Diamond, 1972; Chang and Rao, 1994). Ballard et al. (1995) conducted a mathematical assessment of Pappenheimer's (1990) argument and confirmed that intestinal glucose concentrations in vivo (by recent measurements) do not reach levels at which paracellular transport could account for more than about 30% of total glucose uptake. Pappenheimer and Reiss (1987) argued that the principal role of carriermediated transport of solutes from the lumen to intercellular spaces is to provide the osmotic driving force for paracellular fluid absorption. The proposed mechanics of this process were discussed earlier. The addition of D-glucose to lumenal contents has been shown to cause significant increases in net active Na+absorption and in liquid absorption across intestinal epithelia (Pappenheimer and Reiss, 1987; Atisook et al., 1990). Furthermore, the addition of phloridzin (Heaton and Code, 1969),a selective inhibitor of Na+glucose cotransport, or replacement of Na' with other cations (Csaky and Zollicoffer,1960)blocks this increased liquid absorption. Ballard et al. (1995) pointed out that although the observation of increased fluid movement is not in dispute, it is debatable whether the water moves predominantly across cell membranes or through the tight-junctions. A study with Necturus gall bladder found that about 30% of osmotically driven water absorption passes transjunctionally (implying that up to 70% passes transcellularly; Ballard et al., 1995and references therein). Further, Loo et al. (1996)estimated that the translocation of each glucose molecule by SGLTl is coupled with the transport of up to 260 water molecules. In hummingbirds, the uptake of water associated with the hydration sphere of actively transported D-glucose alone can account for complete absorption of their immense daily water fluxes (McWhorter and Martinez del Rio, 1999).If a significant portion of water absorption occurs via a transcellular route, the importance of paracellular water (and thus hydrosoluble nutrient) absorption hypothesized by Pappenheimer and Reiss (1987) is questionable. The observation of phloridzin blocking water absorption (Heaton and Code, 1969)is consistent with the idea that significant water absorption occurs via active cotransport, although phloridzin would also block the building of osmotic gradients in intercellular spaces by actively transported glucose. Indeed, Pencek et al. (2003) recently found that phloridzin blocked both passive and active uptake of glucose (measured as uptake of radiolabeled L-glucose and
Paracellular c a r b o h y d r a t e absorption
123
D-glucose respectively) in chronically catheterized dogs and concluded that mediated glucose transport by SGLTl was necessary to activate passive transport. Ballard et al. (1995) made several criticisms of the conclusions of Pappenheimer's group regarding solute-induced changes in tight-junction permeability. First, they pointed out that because the concentration of ferrocyanide (animpermeant osmolyte)used by Madara and Pappenheimer (1987) only reduced liquid absorption by 20°/0, their conclusion that the morphological changes observed in tight-junctions resulted from contraction of cellular cytoskeletal elements and not from liquid absorption was unjustified. It is interesting to note, however, that if a significant volume of water is absorbed by Na+-coupledactive transport as discussed in the previous paragraph, this criticism may not be valid. Second, Ballard et al. (1995)pointed out that although several lines of evidence indicate that solute-induced changes in cellular cytoskeleton may modulate tight-junction integrity (Madara et al., 1987;Madara and Pappenheimer, 1987; Pappenheimer, 1987; Madara and Carlson, 1991; Pappenheimer and Volpp, 1992),some authors have challenged the notion that glucose-stimulated liquid absorption increases tight-junction permeability to hydrophilic solutes. Fine et al. (1993) found no changes in tight-junction permeability associated with glucosestimulated liquid absorption. It is not clear whether these differences are due to experimental or interspecifc differences. The problems of separating the direct effects of D-glucose and Na' transport make it difficult to determine whether changes in tight-junctionmorphology occur prior to or as a consequence of liquid absorption. Several recent studies appear to confirm the notion that paracellular transport plays only a minor role in intestinal carbohydrate absorption in mammals. Fine at al. (1993)found that in human in vivo jejunal perfusions the fraction of total glucose absorption that could be attributed to a passive mechanism averaged 5%. No passive absorption was detected in the human ileum in vivo. As already mentioned, Fine et al. (1993) likewise did not observe an increase in tight-junction permeability associated with Nai-dependent nutrient transport. Several recent studies in rats (O'Rourke et al., 1995;Schwartz et al., 1995;Uhing and Kimura, 1995a;Shi and Gisolfi, 1996) and dogs (Lane et al., 1999; Pencek et al., 2002; Pencek et al., 2003) have also concluded that paracellular flux is not a major absorptive pathway for hexoses, despite observations of increased fluid absorption (consistent with Pappenheimer's ideas) in many cases. Lane et al. (1999) used Thiry-Vella perfusion loops to quantify paracellular (i.e. L-glucose) uptake in dog jejunum. This preparation is highly controllableand thought to mimic natural digestive conditions very closely. At physiological concentrations of D-glucose (1-50 mM), the fractional absorption of L-glucose was only 4 to 7% of total glucose absorption. Even under supraphysiological conditions (perfusion of 150mM D-glucose, D-maltose, or D-mannitol),fractional absorption
124
Physiological and ecological adaptations t o feeding in vertebrates
of L-glucose was low (2 to 5'/0). There was significant fractional water absorption in all experiments, which is a prerequisite for solvent drag (but see discussion above). Despite experimental conditions designed to maximize paracellular transport, their results were interpreted as evidence that paracellular transport of glucose plays a minor role in the dog (Lane et al., 1999). Pencek et al. (2003)recently found that the contribution of passive absorption to total glucose uptake in dogs was slightly higher than previously reported and increased after exercise relative to rested controls (from -11% to 18%of total). Interestingly, Schwartz et al. (1995)found that a significant fraction (71%) of administered L-glucose was absorbed in rat intestine, but only after Dglucose concentrations were reduced to negligible levels. Based on these measurements they concluded that L-glucose has a weak affinity for the Dglucose carrier and is therefore not an adequate marker for paracellular uptake. These authors cite differences among the absorption of L-glucose and other nonmetabolized hexoses as additional evidence that L-glucose interacts with an active transport system (Schwartzet al., 1995). Several older studies seem to show active transport of L-glucose (Caspary and Crane, 1968;Neale and Wiseman, 1968;Bihler, 1969;Hopfer et al., 1975);however, it is possible that in these older studies the L-glucose used could have been contaminated with D-glucose or other carbohydrates. A considerable volume of recent work, including some basolateral membrane vesicle uptake studies, indicates that L-glucose does not interact with the SGLTl transporter (Wright et al., 1980; Ikeda et al., 1989; Fine et al., 1993; Uhing and Kimura, 1995a; Lane et al., 1999). An additional criticism of Schwartz et al. (1995)is that gut absorption of hexoses was determined by their subsequent collection in urine, an indirect method subject to the effects of differential renal handling of hexoses (Ullrich and Papavassiliou, 1985). The absorption of these carbohydrateprobes also varies greatly with structure and molecular weight (e.g.Chediack et al., 2003). A criticism of the mammalian studies described above is that in most cases hexoses were perfused directly into the intestinal lumen and were therefore subject to the diffusion dampening effects of USL (Barry and Diamond, 1984;Meddings and Westergaard, 1989). This potential restriction to the diffusion could reduce the absorption of markers and thus lead to underestimates of paracellular uptake. Indeed, Michaelis-Menten constants for glucose transport are higher in vivo than in vitro (attesting to the effects of thicker USL in vivo; Meddings and Westergaard, 1989; Ferraris et al., 1990). It does appear, however, that diffusion-limited probes can be utilized to measure USL resistance in the intestine of a live animal so that absolute transport parameters can be determined in vivo in experimental animals (Westergaard et al., 1986). Such corrections and derivations of corrected kinetic constants (Meddings and Westergaard, 1989) would allow direct comparison among species and experimental protocols, remove this
Paracellular carbohydrate absorption
125
potential criticism, and assist in elucidating the true contribution of paracellular solvent drag to the absorption of dietary carbohydrates. Kellet and colleagues recently provided new evidence that an apparent passive component of glucose uptake indeed exists, but that it is mediated by GLUT2 in the brush-border membrane (see above and Helliwell et al., 2000a; Kellett and Helliwell, 2000; Kellett, 2001). They argued that this finding explains the discrepancy between the saturation of SGLTl at lumenal concentrations of 30-50 mM D-glucose and the long-established pattern of a linear increase in glucose uptake up to concentrations of several hundred millimolar (Kellett, 2001 and references therein). They further argued that the very presence of a facilitative transporter in the brush-border membrane invalidates the theory of paracellular solvent drag proposed by Pappenheimer and colleagues because this would prevent the concentration of glucose in intercellular spaces. They explained the direct relationship between water transport and glucose absorption at higher glucose concentrations (e.g. Fullerton and Parsons, 1956; Pappenheimer and Reiss, 1987)by invoking the idea that glucose transporters can act as low conductance water channels (see above and Fischbarg et al., 1990; Loo et al., 1999). Their conclusions were based on the premise that GLUT2 is the primary glucose transporter at high lumenal glucose concentrations. Their results confirmed that glucose absorption via SGLTl is necessary to activate passive absorption, consistent with Pappenheimer 's predictions, and that SGLTl plays an important regulatory role in regulating tight-junction structure. They argued that alterations in tight-junction permeability mediated by glucose transport may be associated with the translocation of GLUT2. Given the debate surrounding in vivo lumenal glucose concentrations (see above), the technical difficulties in quantifying recruitable brush-border GLUT2 activity (see Helliwell et al., 2000b; Kellett, 2001), and the fact that recruitable GLUT2 activity has only been studied in rats (which apparently have very low paracellular intestinal permeability, Schwartz et al., 1995), the significance of their findings remains unclear. Recent evidence suggests, for example, that GLUT2 recruitment is unlikely to account for the significant passive absorption of glucose observed in birds (seebelow and Chang, 2002), although there may be differences in paracellular permeability and uptake mechanisms between birds and mammals.
ABSORPTION OF CARBOHYDRATES BY THE AVIAN INTESTINE The evidence to date indicates that paracellular flux plays a relatively minor role in total dietary carbohydrate uptake in mammals. Paracellular permeability may be slightly increased in mammals after exercise (e.g.Pencek et al., 2003). There is a general consensus, however, that at high lumenal glucose
126
Physiological and ecological adaptations t o feeding in vertebrates
concentrations (>50mM) paracellular transport may become more important (Ferrariset al., 1990; Pappenheimer, 1993). Although the diets of most animals do not routinely include carbohydrate levels high enough to create lumenal glucose concentrations >50 rnM (but see discussion above), nectarand fruit-eating animals may be an exception. The average nectar concentration found in the floral nectars of hummingbird pollinated plants for example, is about 23% sucrose (wt/vol; approximately 670 mM; Pyke and Waser, 1981),yielding an equal glucose concentration upon hydrolysis. The relative contribution of passive, paracellular hexose uptake (Karasov et al., 1986; Karasov and Cork, 1994; Karasov et al., 1996; Levey and Cipollini, 1996;Afik et al., 1997)and the correlation between functional and mechanistic aspects of paracellular absorption (Chediack et al., 2001; Chang, 2002; Chediack et al., 2003) have been extensively investigated in birds over the past decade. The findings of these studies, the methods employed, and their implicit assumptions are critically examined below. The spare capacity hypothesis proposes that the ability of the intestine to absorb nutrients is slightly greater than required to meet loads determined by food intake (Diamond, 1991; Diamond and Hammond, 1992). Nectarand fruit-eating birds routinely consume several times their body mass in food per day to meet energy demands (e.g. Rooke et al., 1983; Powers and Nagy, 1988; Beuchat et al., 1990; Williams, 1993; Powers and Conley 1994; Goldstein and Bradshaw, 1998; McWhorter and Martinez del Rio, 1999; Fleming and Nicolson, 2003). Although nectar diets consist mainly of water, these birds ingest and assimilate enormous carbohydrate loads. Humrningbirds by and large assimilate all of the carbohydrates in their diets, regardless of concentration (Hainsworth, 1974; Karasov et al., 1986; McWhorter and Martinez del Rio, 1999). Capacity therefore appears to be well matched to load in these animals (but see Second Generation Reactor Models: Gut Function in Nectar-eating Birds in chapter 3, by McWhorter, this volume). Until recently, it was not known whether nectar- and fruit-eating birds have exceptionally high active transport capacity or rely extensively on paracellular routes of carbohydrate uptake. It might be expected that given potentially high lumenal glucose concentrations, these birds could save the energy required for active transport if the intestine's passive permeability were sufficiently high. Karasov et al. (1986) evaluated both carrier-mediated active transport and passive permeability of the intestinal epithelia in hummingbirds. Their results indicated that the passive permeability of hummingbird intestines was immeasurably low but that capacity for carrier-mediatedsugar absorption was the highest measured for any animal to date. They calculated that active transport alone was sufficient to account for all glucose absorption over a brief experimental period. Since passive absorption was apparently unnecessary to absorb sugars, they speculated that the intestine's low passive permeability might be an adaptation to high rates of fluid transit,
Paracellular c a r b o h y d r a t e absorption
127
protecting the animal from losing valuable solutes by dialysis (Karasov et al., 1986). It is important to point out, however, that the validity of the everted intestinal sleeve method used to measure these uptake parameters has recently come into question. The method employs an in vitro preparation where a section of intestine is removed, everted, and placed on a glass or metal rod. The tissue is then immersed in a rapidly stirred nutrient bath to measure ~ (but apparently not in mammals) uptake. It has been discovered that i r birds the shearing force of the stirred bath and tissue manipulation during preparation significantlycompromise the epithelia (Starck et al., 2000), and tissue bioactivity may also be reduced (Karasov and Debnam, 1987).In addition, the everted sleeve method has been criticized for ignoring physiological conditions relevant to accurately characterizing passive transport (Pappenheimer, 1990; Karasov et al., 1996). Powers and Nagy (1988) estimated the field metabolic rate (FMR)of Anna's hummingbirds (Calypteanna, one of the species studied by Karasov et al., 1986)to be 32 kJd-'. This energy requirement is far above the maximum estimates of transport based on active mechanisms alone (6.2 kJ d-l, calculated from Karasov et al., 1986). Two possibilities arise from this line of reasoning: (1)the active transport rates in hummingbird intestines were underestimated due to methodological problems and (2)paracellular absorption of nutrients plays a major role in hummingbirds. These possibilities are, of course, not mutually exclusive. A discussion of more recent studies of passive transport in birds and analysis of the methods employed provides additional insight. Karasov and Cork (1994)tested the hypothesis that most glucose absorption across the small intestine's brush border is by active cotransport with sodium in nectar-eating rainbow lorikeets (Trichoglossus haema todus). Maximal mediated D-glucose uptake summed along the entire intestine using the in vitro everted sleeve method was an order of magnitude too low to explain observed rates of glucose assimilation in vivo. This result implied a predominant role for passive glucose absorption in these birds. These authors applied a pharmacokinetic technique to measure passive absorption in vivo and found that 80% of ingested L-glucose was absorbed, providing evidence for significant nonrnediated transport (Karasov and Cork, 1994). Several additional studies using Karasov and Cork's (1994)method provided similar estimates of L-glucose absorption: 52% to 92% in Northern bobwhite quail (Levey and Cipollini, 1996);91% in omnivorous yellow-rumped warblers (Afik et al., 1997);75% in granivorous house sparrows (Caviedes-Vidal and Karasov, 1996); and 79% in frugivorous cedar waxwings (D.J. Levey pers. comm.). In addition, McWhorter (unpub. data), using this method, recently found that between 37% and 85% of ingested L-glucose was absorbed by nectarivorous broad-tailed hummingbirds. There seems to be considerable evidence that paracellular absorption is a significant route for carbohydrate absorption in the avian intestine. The consistency of this finding in birds with diverse diets and taxonomic associations suggests a general
128
Physiological and ecological adaptations t o feeding in vertebrates
phenomenon in birds. Because all of the abovementioned studies rely on the same model, however, careful examination of the method and its assumptions is prudent. Karasov and Cork's (1994)method relies on determining the steady-state concentration of labeled L-glucose (P)in the plasma of freely feeding birds, the marker ingestion rate (I),the marker distributionspace (D),and the elimination constant for removal of L-glucose from plasma (k,). Fractional absorption (F)is then calculated as: (1) F = (P. k;D)/I The L-glucose pool size and elimination constant are measured simultaneous with the steady-state feeding experiment by injecting L-glucose with an alternate label and taking consecutive blood samples. Although the model is relatively simple, it relies on several important assumptions. A primary one is that absorbed L-glucose is not metabolized but entirely excreted and that its excretion can be approximated by single pool kinetics. Karasov and Cork (1994) cited two lines of evidence in support of the assumption that L-glucose is not metabolized. First, injected Lglucose was 94%recovered in quantitative collections of excreta, which implies little or no metabolism or deposition in tissues. Second, the ratio of activity of excreta counted before and after drying was one, implying that label had not been transferred to water during metabolism. Further, CaviedesVidal and Karasov (1996)and Chang et al. (2004) found that essentially all radioactivity (95-98%) remained associated with L-glucose in house sparrow (Passerdomesticus)plasma, using thin-layer chromatography to separate [3H]L-glucoseand high-performance liquid chromatography (HPLC)to separate [14C]L-glucose respectively. Examination of plasma label decay curves followinginjection of labeled L-glucose led Karasov and Cork (1994)to conclude that excretion of L-glucose was approximated by single pool kinetics. Specifically,semilog plots of specific activity per gram excreta vs time were approximately linear, which implies single pool kinetics. McWhorter (unpubl. data) also observed linear semilog plots of [14C]L-glucoseactivity in excreta in hummingbirds. Karasov and Cork (1994) argued that changing the assumption about the number of compartments causes counterbalancing changes in the slope (-ke,)and intercept (the inverse of pool size)of the slowest components of the curve that are plotted at longer time points. In other words, a second compartment with slower turnover would lead to a shallower slope (and thus lower elimination constant), but a larger pool size. The authors argued that these changes would cancel each other out, which seems reasonable based on a brief graphical analysis of the model (Karasov and Cork, 1994). A second major assumption of the model is that L-glucose absorption can be equated with passive D-glucose absorption. This involves both the assumption that the diffusion coefficients across the epithelia are the same and
Paracellular carbohydrate absorption
129
that L-glucose is not recognized by a transporter. Karasov and Cork (1994) admitted that the assumption of equal diffusion coefficients is not strictly correct in every circumstance.The difference is due to the mediated transport of D-glucose into the intercellular space, potentially leading to a small or even negative diffusive term for D-glucose (whilethat of L-glucose remained positive). They pointed out, however, that if this is the case then L-glucose absorption actually underestimates D-glucose passive absorption and the conclusion of significant paracellular transport holds. Although Schwartz et al. (1995)concluded that L-glucose seems to have a weak affinity for SGLT1, most experimental evidence suggests that it does not (Wright et al., 1980; Ikeda et al., 1989; Fine et al., 1993; Uhing and Kimura, 1995a; Lane et al., 1999;Chang, 2002). It appears that the assumptions of Karasov and Cork's (1994)model are satisfied or that deviations from such do not impact conclusions significantly. If anything, deviations from the assumptions lead to underestimates of paracellular glucose absorption. Fractional absorption measurements using alternative markers and calculation methods suggest that previous measurements of L-glucose absorption are not artifacts of isotope separation or affinity for active transport systems (Chediack et al., 2001; Chediack et al., 2003). Chediack et al. (2001) extended the observations of hydrophilic absorption in birds using a pharmacokinetic technique that relies on the appearance of probes in the blood after feeding and injection (Caviedes-Vidal and Karasov, 1996). The absorption of carbohydrate probes (D-mannitoland L-arabinose,both thought to be abso:rbedonly via nonmediated processes, Dawson et al., 1987;Krugliak et al., 1994)was calculated by assuming a single compartment and first-order kinetics (Welling,1986). Inspection of plasma decay curves following injection of probes and comparative fitting of the data to one-compartment and two-compartment models lead the authors to conclude that this assumption was appropriate. Following typical procedures in pharmacokinetics, fractional absorption (F)was calculated using the areas under the post-gavage and -injection curves (AUC) of marker concentration vs time: = (AUCgavage/Dosegavage)/(AUCiniection/Doseiniection) (2). The authors found substantial absorption of D-mannitol and L-arabinose (69 -+ 3%, no difference between probes or among administered concentrations) in house sparrows using this method. Absorption of both probes correlated directly with their orally administered doses, indicating that uptake was nonmediated. These results are consistent with an earlier report of substantial (75%)absorption of [3H]L-glucosein house sparrows (Caviedes-Vidaland Karasov, 1996),providing an important confirmation of a substantial pathway for passive absorption in birds.
130
Physiological and ecological adaptations t o feeding in vertebrates
Mechanistic Examinations of Paracellular Uptake Recent studies of the correlation between functional and mechanistic aspects of paracellular uptake in birds are providing further confirmation of the importance of this pathway for nutrient absorption and will certainly invite and inform more mechanistic future studies. These studies are briefly reviewed below. Chediack et al. (2003)used the pharmacokineticmethod described above (see eqn 2) to examine two mechanistic aspects of paracellular absorption of carbohydrates. The first objective of their study was to examine the affect of probe molecular size on absorption. Previous studies had shown a decline in absorption with increase in molecular weight of probes that was more rapid than expected based on their free aqueous diffusion coefficients (Hamilton et al., 1987;Meehye, 1996;Ghandehari et al., 1997;He et al., 1998). Chediack et al. (2003) recognized that this pattern is consistent with movement through effective pores in epithelia ("sieving", Chang et al., 1975;Friedman, 1987);such size selectivity partly determines the size range of hydrophilic nutrients or toxins that might be absorbed passively. Absorption of water-soluble compounds in vivo seems more likely to occur via solvent drag through intercellular channels than by diffusion through the cytoplasm of enterocytes (Pappenheimer,2001). The second objective of the Chediack et al. (2003) study was to test for modulation of paracellular absorption. Paracellular permeability may be altered by both endogenous and exogenous agents (the latter including nutrients such as glucose, amino acids and medium chain fatty acids, as well as toxins, see Chediack et al., 2003 and references therein). The mechanisms by which such modulation occurs are not known, but may include increased solvent drag and/or cytoskeletal contractions (Madara et al., 1986; Madara and Pappenheimer, 1987; Pappenheimer, 1987;Pappenheimer and Reiss, 1987;Madara et al., 1988)or protein strand alterations that change tight-junction effective pore size (see Sasaki et al., 2003). Chediack et al. (2003) measured the fractional absorption of metabolically inert carbohydrate probes ranging in molecular weight from 150.13to 342.2 Da in the presence and absence of both lumenal food and glucose to meet these objectives. They found that (1)fractional absorption decreased significantlywith increase in molecular weight (61 9% for L-arabinose, 64 13% for L-rhamnose, 46 2 8% for perseitol, and 15 + 5% for lactulose, averaged over all treatments), and (2) fractional absorption was significantly greater in the presence of food or D-glucose than in the presence of D-mannitol (not absorbed by mediated processes). The latter result was confirmed by Chang et al. (2004)who found that the fractional absorption of L-glucose was significantlyincreased in the presence of 3-0-methylD-glucose (3-OMG, an analogue of D-glucose that is transported but not metabolized). A theoretical understanding of how hydrophilic molecules cross epithelia can be applied profitably to the interpretation of such data
*
Paracellular carbohydrate absorption
131
(see Chediack et al., 2003 and references therein). Such interpretation may clarify the relative effects of changes in effective pore size and solvent flux for determining the magnitude of paracellular absorption within and among species, and may be important for predicting the oral bioavailability of water-soluble nutrients and natural and man-made xenobiotics (toxins,plant secondary compounds, toxicants, drugs, etc.). Differences in effective pore size among species or due to dietary or other modulation (He et al., 1998; Chediack et al., 2003) discovered using the aforesaid methods may also invite further comparative studies on the mechanism(s)of paracellular uptake. Kellet and colleagues recently suggested an alternative mechanism by which D-glucose absorption might be enhanced by mediated nutrient uptake: recruitment of GLUT2 to the brush-border membrane (see above and Helliwell et al., 2000b; Kellett and Helliwell, 2000). The apparent Kmof GLUT2 is higher than that of SGLT1; thus the authors argued that the effect may falselygive the appearance of an increase in nonmediated uptake. However, Chang et al. (2004)argued that observed increases in the absorption of L-glucose, which is not transported by GLUT2 expressed in Xenopus oocytes (Burant and Bell, 1992),and observations of increased absorption of other metabolically inert carbohydrates (Chediack et al., 2003),in the presence of lumenal glucose refutes this assertion, at least in birds. Chang et al. (2004)tested the hypothesis that paracellular absorption is an important component of glucose absorption in birds by comparing the apparent rate and extent of absorption of D- and L-glucose in vivo in house sparrows. These authors further predicted that the absorption of labeled Lglucose would not be depressed (by competitive inhibition, Malo and Berteloot, 1991)when measured in the presence of a high sugar concentration (i.e. 100 mM unlabeled D- or L-glucose) in vitro. The in vivo predictions were tested simultaneouslyby measuring the fractional absorption and apparent absorption rate of radiolabeled 3-OMG (see above) and L-glucose from the intestine. In vitro uptake experiments employed the everted sleeve method; however, the gross morphological damage and reduced bioactivity found in other avian species (Starck et al., 2000) were apparently not observed. The authors found that (1) uptake of L-glucose in vitro was not inhibited by high concentrations of unlabeled sugars, indicating that it does not interact with a transporter, and (2) the apparent rate of absorption and extent of D- and L-glucose uptake were similar in vivo and indicated that >7O0/0of glucose absorption was passive. Schwartz et al. (1995) suggested that simply comparing the fractional absorptions of D- and L-glucose was inappropriate because these isomers appeared to be absorbed in different regions in rats. Chang et al. (2004)asserted that this explanation does not apply to house sparrows; D- and L-glucose had similar apparent absorption rates throughout all experimental sampling time points, and absorption of L-glucose was not prolonged compared to D-glucose (it in fact occurred faster in the presence of lumenal D-glucose). These experiments confirmed
132
Physiological and ecological adaptations t o feeding in vertebrates
that the vast majority of glucose is absorbed by a passive route in house sparrows. Because the fractional absorptions of D- and L-glucose were similar and L-glucose is apparently not transported by GLUT2 (Burant and Bell, 1992),it seems implausible that GLUT2 recruitment to the brush-border is entirely responsible for the passive component of glucose uptake in birds (see also Kellett, 2001). Chang and Karasov (2004)attempted to further verify and visualize the paracellular pathway in intact house sparrows using sodium fluorescein, which distributes only to the extravascular interstitial space (Nugent and Jain, 1984; Hurni et al., 1993; Sakai et al., 1997) and is widely used as a hydrophilic marker for in vitro paracellular permeability studies (Chao et al., 1998; Lindmark et al., 1998; Stagni et al., 1999; Clausen and BernkopSchnurch, 2000; Gaillard and de Boer, 2000). These authors predicted that the absorption of fluorescein would be only via the paracellular pathway and would be increased in the presence of lumenal D-glucose. Pharmacokinetic analysis indicated that the fractional absorption of fluorescein was approximately 40% and that of L-glucose was approximately 80%. Fluorescein absorption peaked faster in the presence of lumenal D-glucose (at 6 rnin vs 8 min in the D-mannitol control group), but fractional absorption was not sigruficantly different among treatments. Visualization using confocal laser scanning microscopy showed that fluorescein was found primarily in the paracellular space and villous core. Some fluorescence at the apical membrane was attributed to binding (Braginskaja et al., 1994).Three-dimensional image reconstruction confirmed that fluorescein was distributed extravascularly. A positive correlation between fluorescein and L-glucose absorption was interpreted as suggesting that these molecules share a common uptake pathway. Chang and Karasov (2004) argued that alternate explanations (transport of fluorescein into enterocytesfollowed by metabolism or rapid expulsion into intercellular spaces) are unlikely. Mediated transporters of fluoresceinmay be expressed in rat small intestine (see Walters et al., 2000; Cattori et al., 2001; Sun et al., 2001), however there is no direct evidence that fluorescein is transported in intestinal enterocytes.
A primary observation that led to the hypothesis that paracellular flux may be important for nutrient absorption was that maximal carrier-mediated transport was insufficient to explain rates of glucose assimilation in vivo (Pappenheimer, 1990). Most recent evidence suggests that paracellular flux plays a relatively minor role in the intestinal absorption of carbohydrates in mammals, but may play a much more sigruficant role in birds. This conclusion is robust even if the studies discussed above seriously underestimated mediated uptake capacities (see caveats for the everted sleeve method).
Paracellular carbohydrate absorption
133
Glucose-dependentrecruitment of GLUT2 to the brush-border membrane may constitute a significant pathway for passive transcellular glucose and fructose absorption in mammals (Kellett,2001) but may not play as significant a role in birds. Evidence clearly falsifying Pappenheimer 's paracellular solvent drag hypothesis in mammals, however, has not yet come to light. It remains to be seen whether significant paracellular flux is a general phenomenon among birds, or if rigorous testing of the assumptionsof previously employed models and use of alternative methods (e.g.in vivo blocking of SGLTl and GLUT2 with phloridzin and phloretin respectively) will lead to different conclusions. It is possible that paracellular absorption is a general vertebrate trait that has been particularly enhanced to augment mediated nutrient absorption in flying birds that are known to possess relatively less small intestinal tissue (Karasov and Hume, 1997)and exhibit relatively shorter digesta retention times. A systematic study across birds, mammals, and reptiles using uniform methodology would be an important contribution toward resolving this issue. Recent studies of the correlation among functional and mechanistic aspects of paracellular flux have important bearing on the emerging fields of comparative evolutionary physiology and ecotoxicology. These studies establish important limits on studies quantifying physiological capacity in relation to load (Diamond, 1991,1993). Passive absorption clearly must be considered when attempting to match nutrient absorption capacity and nutrient intake. Pappenheimer and colleagues almost single-handedly advanced the hypothesis of water-soluble solute absorption in a field dominated by the molecular biology of carbohydrate transporters from the late 1980s onward (Pappenheimer,1990,1993,1998,2001). They proposed that nutrient-stimulated water, and thus accompanying solute absorption increases with body size, whereas the importance of transcellular mediated absorption declines and therefore the importance of paracellular absorption must increase with body size. This prediction has important implications for nutrient absorption, oral drug deliver$ and exposure to natural and man-made toxins. Phylogenetic and body-size differences in, and modulation of, paracellular flux are important for understanding and predicting the extent of water-soluble compound absorption (Anderbertet al., 1993;Chang and Rao, 1994; Bjork et al., 1995; Karlsson et al., 1999). Some studies that have compared water-soluble solute permeability among different size mammals seem to support Pappenheimer's predictions (seeChediack et al., 2003 for review). Other studies on humans (Fine et al., 1993),rats (Schwartzet al., 1995;Uhing and Kimura, 1995a), and dogs (Lane et al., 1999; Pencek et al., 2002; Pencek et al., 2003) have concluded that most glucose absorption in mammals is mediated. A theoretical framework based on the flux of solutes through porous epithelia (Kedem and Katchalsky, 1958;Chang, 2002; Chediack et al., 2003) can be used to describe the paracellular absorption of water-soluble
134
Physiological and ecological adaptations t o feeding in vertebrates
molecules across the small intestine's mucosal epithelium and to separate the relative importance of diffusion and solvent drag. Such an understanding will invite and inform future mechanistic and comparative studies of nutrient absorption. The molecular events that result in tight-junctionregulation (Anderson, 2001; Kellett, 2001), phylogenetic differences in tight-junction structure (Sasakiet al., 2003),and the regulation and generality of GLUT2 recruitment to the brush-border membrane (Kellett, 2001), will certainly be important areas of future research. Acknowledgments
Supported by N. S. F. (IBN-0216709)to William H. Karasov. REFERENCES Afik D., Caviedes-Vidal E., Martinez del Rio C., and Karasov W. H. 1995. Dietary modulation of intestinal hydrolytic enzymes in yellow-rumped warblers. Amer. J. Physiol. 269: R413-R420. Afik D., McWilliams S. R. and Karasov W. H. 1997. A test for passive absorption of glucose in yellow-rumped warblers and its ecological implications. Physiol. Zool. 70: 370-377. Alpers D. H. 1987. Digestion and absorption of carbohydrates and proteins. In: Physiology of the Gastrointestinal Tract, vol. 2 L. R. Johnson (ed.). Raven Press, New York, NY, pp. 1469-1487. Anderbert E. K., Lindmark T., and Artusson P. 1993. Sodium caprate elicits dilations in human intestinal tight junctions and enhances drug absorption by the paracellular route. Pharm. Res. 10: 857-864. Anderson J. M. 2001. Molecular structure of tight junctions and their role in epithelial transport. News Physiol. Sci. 16: 126-130. Anderson J. M. and van Itallie C. M. 1995. Tight junctions and the molecular basis for regulation of paracellular permeability. Amer. J. Physiol. 269: G467-G475. Atisook K., Carlson S. and Madara J. L. 1990. Effects of phlorizin and sodium on glucose-elicited alterations of cell junctions in intestinal epithelia. Amer. J. Physiol. 258: C77-C85. Ballard S. T., Hunter J. H., and Taylor A. E. 1995. Regulation of tight-junction permeability during nutrient absorption across the intestinal epithelium. A n n u . Rev. Nutr. 15: 35-55. Barry I? H. and Diamond J. M. 1984. Effects of unstirred layers on membrane phenomena. Physiol. Rev. 64: 763-872. Beuchat C. A., Calder W. A., I11 and Braun E. J. 1990. The integration of osmoregulation and energy balance in hummingbirds. Physiol. Zool. 63: 1059-1081. Bihler I. 1969. Intestinal sugar transport: ionic activation and chemical specificity. Biochim. Biophys. Acta 183: 169-181. Bjork E., Isaksson U., Edman P. and Artusson P. 1995. Starch microspheres induce pulsatile delivery of drugs and peptides by reversible separation of tight junctions. J. Drug Target. 2: 501-507. Braginskaja 0.V., Lazarev V. V., Tikhonova 0.V. and Shneider V, B. 1994. Role of sodium fluorescein binding during its accumulation inside cells. Biofizika 39: 68-73. Burant C. F. and Bell G . I. 1992. Mammalian facilitative glucose transporters: evidence for similar substrate recognition sites in functionally monomeric proteins. Biochemistry 31: 10414-10420.
Paracellular carbohydrate absorption
135
Burant C. F., Takeda J., Brot-Laroche E., Bell G. I., and Davidson N. 0. 1992. Fructose transporter in human spermatozoa and small intestine is GLUT5. J. Biol. Chem. 267: 14523-14526. Caspary W. F. and Crane R. K. 1968. Inclusion of L-glucose within the specificity limits of the active sugar transport system of hamster small intestine. Biochim. Biophys. Acta 163: 395-400. Cattori V., van Montfoort J. E., Stieger B., Meier P. J., and Hagenbuch B. 2001. Localization of organic anion transporting polypeptide 4 (Oatp4) in rat liver and comparison of its substrate specificity with O a t p l , Oatp2 and Oatp3. Eur. J. Physiol. 443: 188-195. Caviedes-Vidal E. and Karasov W. H. 1996. Glucose and amino acid absorption in house sparrow intestine and its dietary modulation. Amer. J. Physiol. 40: R561-R568. Chang E. B. and Rao M. C. 1994. Intestinal water and electrolyte transport. In: Physiology of the Gastrointestinal Tract, L. R. Johnson (ed.). Raven Press, New York, NY, pp. 2027-2081. Chang M. H. 2002. The magnitude and significance of passive absorption in intact house sparrows, Passer domesticus. PhD dis. Dept. Zoology, University Wisconsin, Madison, WI. Chang M. H., Chediack, J. G., Caviedes-Vidal, E. and Kavasov, W.H. 2004. L-glucose absorption in house sparrows (Passer domesticus) is nonmediated. J. Comp. Physiol. B 174: 181-188 Chang M. H. and Kavasov, W. H. 2004. Absorption and paracellular visualization of fluorescein, a hydrosoluble probe, in intact house sparrows (Passer domesticus). Zoology 107: 121-133. Chang R. L. S., Robertson C. R., Deen W. M., and Brenner B. M. 1975. Permselectivity of the glomerular capillary wall to macromolecules. I. Theoretical considerations. Biophys. J. 15: 861-886. Chao A. C., Taylor M. T., Daddona P. E., Broughall M. and Fix J. A. 1998. Molecular weight-dependant paracellular transport of fluorescent model compound induced by palmitoylcarnitine chloride across the human intestinal epithelial cell line Caco-2. J. Drug Target. 6: 37-43. Chediack J. G., Caviedes-Vidal E., Karasov W. H., and Pestchanker M. 2001. Passive absorption of hydrophilic carbohydrate probes by the house sparrow Passer domesticus. J. Exp. Biol. 204: 723-731. Chediack J. G., Caviedes-Vidal E., Fasulo V,, Yamin L. J., and Karasov W. H. 2003. Intestinal passive absorption of water-soluble compounds by sparrows: effect of molecular size and lumenal nutrients. J. Comp. Physiol. B 173: 187-197. Clausen A. E. and Bernkop-Schnurch A. 2000. In: vitro evaluation of the permeationenhancing effect of thiolated polycarbophil. J. Plrarm. Sci. 89: 1253-1261. Corpe C. l?, Basaleh M. M., Affleck J., Gould G., Jess T. J., and Kellett G. L. 1996. The regulation of GLUT5 and GLUT2 activity in the adaptation of intestinal brush-border fructose transport in diabetes. Pfliigers Archiv - European J. Physiol. 432: 192-201. Crane R. K. 1960. Intestinal absorption of sugars. Physiol. Rev. 40: 789-825. Crane R. K. 1975. The physiology of the intestinal absorption of sugars. In: Physiological Effects of Food Carbohydrates, A. Jeanes and J. Hodge (eds.). Amer. Chem. Soc. Washington, DC, pp. 2-19. Crane R. K., Miller D. and Bihler I. 1961. The restrictions on possible mechanisms of intestinal active transport of sugars. In: Membrane Transport and Metabolism, A. Kleinzeller and A. Kotyk (ed.). Acad. Press, London, UK, pp. 439-449. Csaky T, and Zollicoffer L. 1960. Ionic effect on intestinal transport of glucose in the rat. Amer. J. Physiol. 198: 1056-1058. Dawson D. J., Burrows P. C., Lobley R. W., and Holmes R. 1987. The kinetics of monosaccharide absorption by human jejunal biopsies: evidence for active and passive processes. Digestion 38: 124-132.
136
Physiological and ecological adaptations t o feeding i n vertebrates
Diamond J. 1991. Evolutionary design of intestinal nutrient absorption: enough but not too much. News Physiol. Sci. 6: 92-96. Diamond J. 1993. Evolutionary Physiology. In: The Logic of Life, C. A. R. Boyd and D. Noble (ed.). Oxford: Univ. Press, Oxford, UK, pp. 89-111. Diamond J. M. and Bossert W. H. 1967. Standing-gradient flow: a mechanism for coupling of water and solute transport in epithelium. J. Gen. Physiol. 50: 2061-2083. Diamond J. M. and Karasov W. H. 1984. Effect of dietary carbohydrate on monosaccharide uptake by mouse small intestine in vitro. J. Physiol. 349: 419440. Diamond J. M. and Hammond K. A. 1992. The matches, achieved by natural selection, between biological capacities and their natural loads. Experientia 48: 551-557. Doege H., Schurmann A., Bahrenberg G., Brauers A. and Joost H. G. 2000. GLUT8, a novel member of the sugar transport facilitator family with glucose transport activity. J. Biochem. 275: 16275-16280. Ferraris R. P. and Diamond J. M. 1989. Specific regulation of intestinal nutrient transporters by their dietary substrates. Annu. Rev. Physiol. 51: 125-141. Ferraris R. P. and Diamond J. 1997. Regulation of intestinal sugar transport. Physiol. Rev. 77: 257-302. Ferraris R. P., Yasharpour S., Lloyd K. C. K., Mirzayan R. and Diamond J. M. 1990. Lumenal glucose concentrations in the gut under normal conditions. Amer. J. Physiol. 259: G822-G837. Fine K. D., Santa Ana C. A., Porter J. L., and Fordtran J. S. 1993. Effect of D-glucose on intestinal permeability and its passive absorption in human small intestine in vivo. Gastroenterology 105: 1117-1125. Fischbarg J. K., Kuang K., Vera J. C., Arant S., et al. 1990. Glucose transporters serve as water channels. Proc. Natl. Acad Sci U S A 87: 3244-3247. Fleming P. A. and Nicolson S. W. 2003. Osmoregulation in an avian nectarivore, the whitebellied sunbird Nectarinia talatala: response to extremes of diet concentration. J. Exp. Biol. 206: 1845-1854. Friedman M. H. 1987. Principles and Models of Biological Transport. Springer-Verlag, Berlin. Fromter E. and Diamond J. M. 1972. Route of passive ion permeation in epithelia. Nature New Biology 235: 9-13. Fullerton P. M. and Parsons D. S. 1956. The absorption of sugars and water from rat intestine in vivo. Quart. Exp. Physiol. 41: 387-397. Gaillard P. J. and d e Boer, A. G. 2000. Relationship between permeability status of the blood-brain barrier and in vitro permeability coefficient of a drug. Eur. J. Pharm. Sci. 12: 95-102. Ghandehari H., Smith P. L., Ellens H., Yeh P.-Y., and Kopecek J. 1997. Size-dependent permeability of hydrophilic probes across rabbit colonic epithelium. J. Pharmacnl. Exp. Ther. 280: 747-753. Goldstein D. L. and Bradshaw S. D. 1998. Regulation of water and sodium balance in the field by Australian Honeyeaters (Aves: Meliphagidae). Physiol. Zool. 71: 214-225. Hainsworth F. R. 1974. Food quality and foraging efficiency: the efficiency of food assimilation by hummingbirds. J. Comp. Physiol. 88: 425-431. Halaihel N., Gerbaud D., Vasseur M. a n d Alvarado F. 1999. Heterogeneity of pig intestinal D-glucose transport systems. Amer. J. Physiol. 277: C1130-C1141. Hamilton I., Rothwell J., Archer D. and Axon T. R. 1987. Permeability of the rat small intestine to carbohydrate probe molecules. Clin. Sci. 73: 189-196. He Y. L., Murby S., Warhurst G., Gifford L., et al. 1998. Species differences in size discrimination in the paracellular pathway reflected by oral bioavailability of poly(ethy1ene) glycol and D-peptides. J. Pharm. Sci. 87: 626-633. Heaton J. and Code C. 1969. Sodium-glucose relationships during intestinal absorption in dogs. Amer. J. Physiol. 216: 749-755. Hediger M. A. 1994. Structure, function and evolution of solute transporters in prokaryotes and eukaryotes. J. Exp. Biol. 196: 15-49.
Paracellular carbohydrate absorption
137
Hediger M. A*, Coady M. J., Ikeda T. S., and Wright E. M. 1987. Expression cloning and cDNA sequencing of the Na+/glucose co-transporter. Nature 330: 379-381. Hediger M. A. a n d Rhoads D. B. 1994. Molecular physiology of sodium-glucose cotransporters. Physiol. Rev. 74: 993-1026. Helliwell P. A., Richardson M., Affleck J. and Kellett G. L. 2000a. Regulation of GLUT5, GLUT2, and intestinal brush-border fructose absorption by the extracellular signal-regulated kinase, p38 mitogen-activated kinase and phosphatidylinositol 3kinase intracellular signaling pathways: implications for adaptation to diabetes. Biochem. J. 350: 163-169. Helliwell P. A., Richardson M., Affleck J. and Kellett G. L. 2000b. Stimulation of fructose transport across the intestinal brush-border membrane by PMA is mediated by GLUT2 and dynamically regulated by protein kinase C. Biochem. J. 350: 149-154. Hochachka P. W., and Somero G. N. 1984. Biochemical Adaptation. Princeton, NJ, Princeton Univ. Press, USA. Holdsworth C. D. and Dawson A. M. 1964. The absorption of monosaccharides in man. Clin. Sci. 27: 371-379. Hopfer U. 1987. Membrane transport mechanisms for hexoses and amino acids in the small intestine. In: Physiology of the Gastrointestinal Tract, L. R. Johnson (ed.). Raven Press, New York, NY, pp. 1499-1526. Hopfer U., Sigrist-Nelson K. and Murer H. 1975. Intestinal sugar transport: studies with isolated plasma membranes. Ann. N.Y. Acad. Sci. 264: 414-427. Hurni M. A., Noach A. R. J. Blom-Roosemalen M. C. M., Boer A. D., Nagelkerke J. F., and Breimer D. D. 1993. Permeability enhancement in Caco-2 cell monolayers by sodium salicylate and sodium taurodihydrofusidate: assessment of effect-reversibility and imaging of transepithelial transport routes by confocal laser scanning microscopy. J. Pharm. Exp. Tech. 267: 942-950. Ikeda T. S., Hwang E.-S., Coady M. J., Hirayama B. A., Hediger M. A., and Wright E. M. 1989. Characterization of a Na+/glucose cotransporter cloned from rabbit small intestine. J. Membr. Biol. 110: 87-95. Ilundain A., Lluch M. and Ponz F. 1979. Kinetics of intestinal sugar transport, in vivo. Rev. Esp. Fisiol. 35: 359-366. Kanai Y., Lee W.-S., You G., Brown D. and Hediger M. A. 1994. The human kidney low affinity Na' /glucose cotransporter SGLT2. J. Clin. Invest. 93: 397-404. Karasov W. H. 1992. Tests of the adaptive modulation hypothesis for dietary control of intestinal nutrient transport. A m . J. Physiol. 263: R496-R502. Karasov W. H. and Diamond J. M. 1983. Adaptive regulation of sugar and amino acid transport by vertebrate intestine. Amer. J. Physiol. 245: G4434462. Karasov W. H. and Debnam E. S. 1987. Rapid adaptation of intestinal glucose transport: a brush-border or basolateral phenomenon? Am. J. Physiol. 253: G54-G61. Karasov W. H. and Cork S. J. 1994. Glucose absorption by a nectarivorous bird: the passive pathway is paramount. Amer. J. Physiol. 267: G16-G26. Karasov W. H. and Hume I. D. 1997. Vertebrate gastrointestinal system. In: Handbook of Comparative Physiology, W . Dantzler (ed.). Amer. Physiol. Soc. Bethesda, MD, pp. 409-480. Karasov W. H., Afik D. and Darken B. W. 1996. Do northern bobwhite quail modulate intestinal nutrient absorption in response to dietary change? A test of an adaptational hypothesis. Comp. Biochem. Physiol. 113A: 233-238. Karasov W. H., Phan D., Diamond J. M., and Carpenter F. L. 1986. Food passage and intestinal nutrient absorption in hummingbirds. Auk 103: 453464. Karlsson J., Ungell A.-L., Grasjo J. and Artusson P. 1999. Paracellular drug transport across intestinal epithelia: influence of charge and induced water flux. European Journal of Pharmaceutical Sciences 9: 47-56. Kedem 0. and Katchalsky A. 1958. Thermodynamic analysis of the permeability of biological membranes to non-electrolytes. Biochinz. Biophys. Acta 27: 229-246.
138
Physiological and ecological adaptations t o feeding in vertebrates
Kellett G. L. 2001. The facilitated component of intestinal glucose absorption. J. Physiol. 531: 585-595. Kellett G. L. and Helliwell P. A. 2000. The diffusive component of intestinal glucose absorption is mediated by the glucose-induced recruitment of GLUT2 to the brushborder membrane. Biochem. J. 350: 155-162. Krugliak P., Hollander D., Schlaepfer C. C., Nguyen H. and Ma T. Y. 1994. Mechanisms and sites of mannitol permeability of the small and large intestine in the rat. Dig. Dis. Sci. 39: 796-801. Lane J. S., Whang E. E., Rigberg D. A., Hines 0.J., et al. 1999. Paracellular glucose transport plays a minor role in the unanesthetized dog. Arner. J. Physiol. 276: G789G794. Levey D. J. and Cipollini M. L. 1996. Is most glucose absorbed passively in northern bobwhite? Comp. Biochem. Physiol. 113A: 225-231. Levey D. J., Place A. R., Rey P. J. and Martinez del Rio C. 1999. An experimental test of dietary enzyme modulation in pine warblers Dendroica pinus. Physiol. Biochem. Zool. 72: 576-587. Lindmark T., Kimura Y. and Artusson P. 1998. Absorption enhancement through intracellular regulation of tight junction permeability by medium chain fatty acids in Caco2 cells. J. Pharmacol. Exper. Ther. 284: 362-369. Loo D. D. F., Zeuthen T., Chandy G . and Wright E. M. 1996. Cotransport of water by the Na+/glucose cotransporter. Proc. Natl. Acad. Sci. U S A 93: 13367-13370. Loo D. D., Hirayama B. A., Meinild A. K., Chandy G., Zeuthen T. and Wright E. M. 1999. Passive water and ion transport by cotransporters. J. Physiol. 518: 195-202. Madara J. L., Moore R. a n d Carlson S. 1987. Alteration of intestinal tight junction structure and permeability by cytoskeletal contraction. Arner. J. Physiol. 253: C854-C861. Madara J. L. and Carlson S. 1991. Supraphysiologic L-tryptophan elicits cytoskeletal and macromolecular permeability alterations in hamster small intestinal epithelium in vitro. J. Clin. Invest. 87: 454-462. Madara J. L., Barenberg D. and Carlson S. 1986. Effects of cytochalasin D on occluding junctions of intestinal absorptive cells: further evidence that the cytoskeleton may influence paracellular permeability and junctional charge selectivity. J. Cell Biol. 102: 2125-2136. Madara J. L. and Pappenheimer J. R. 1987. Structural basis for physiological regulation of paracellular pathways in intestinal epithelia. J. Mernbr. Biol. 100: 149-164. Madara J. L., Stafford J., Barenberg D. and Carlson S. 1988. Functional coupling of tight junctions and microfilaments in T84 monolayers. Arner. J. Physiol. 254: G416-G423. Malo C. 1993. Ontogeny of Na+/D-glucose cotransport in guinea-pig jejunal vesicles: only one system is involved at both 20°C and 35°C. Biochim. Biophys. Acta 1153: 299-307. Malo C. and Berteloot A. 1991. Analysis of kinetic data in transport studies: new insights from kinetic studies of Na+-D-glucose cotransport in human intestinal brush-border membrane vesicles using a fast sampling, rapid filtration apparatus. J. Mernbr. Biol. 122: 127-141. Martinez del Rio C., Brugger K. E., Rios J. L., Vergara M. E., and Witmer M. 1995. An experimental and comparative study of dietary modulation of intestinal enzymes in European starlings (Sturnus vulgaris). Physiol. Zool. 68: 490-511. McWhorter T. J. and Martinez del Rio C. 1999. Food ingestion and water turnover in hummingbirds: How much dietary water is absorbed? J. Exp. Biol. 202: 2851-2858. McWhorter T. J. and Lopez-Calleja M. V. 2000. The integration of diet, physiology, and ecology of nectar-feeding birds. Rev. Chil. Hist. Nat. 73: 451460. Meddings J. B. and Westergaard H. 1989. Intestinal glucose transport using perfused rat jejunum in vivo: model analysis and derivation of corrected kinetic constants. Clin. Sci. 76: 403-413.
Paracellular carbohydrate absorption
139
Meehye K. 1996. Absorption of polyethylene glycol oligomers (330-1122 Da) is greater in the jejunum than in the ileum of rats. ]. Nufr. 126: 2172-2178. Murakami E., Saito M. and Suda M. 1977. Contribution of diffusive pathway in intestinal absorption of glucose in rat under normal feeding conditions. Experientia 33: 1469-1470. Neale R. J. and Wiseman G. 1968. Active intestinal absorption of L-glucose. Nature 218: 473-474. Nugent L. J. and Jain R. K. 1984. Plasma pharmacokinetics and intestinal diffusion of macromolecules in a capillary bed. Amer. J. Physiol. 246: H129-H137. Olsen W. A. and Ingelfinger F. J. 1968. The role of sodium in intestinal glucose absorption in man. J. Clin. Invest. 47: 1133-1142. O'Rourke M., Shi X., Gisolfi C. and Schedl H. 1995. Effect of absorption of D-glucose and water on paracellular transport in rat duodenum-jejunum. Amer. J. Med. Sci. 309 146-151. Pappenheimer J. R. 1987. Physiological regulation of transepithelial impedance in the intestinal mucosa of rats and hamsters. 1. Membr. Biol. 100: 137-148. Pappenheimer J. R. 1990. Paracellular intestinal absorption of glucose, creatinine, and mannitol in normal animals: relation to body size. Amer. J. Physiol. 259: G2904299. Pappenheimer J. R. 1993. On the coupling of membrane digestion with intestinal absoprtion of sugars and amino acids. A m . J. Physiol. 265: G409-G417. Pappenheimer J. R. 1998. Scaling of dimensions of small intestines in non-ruminant eutherian mammals and its significance for absorptive mechanisms. Comp. Biochem. Physiol. 121A: 45-58. Pappenheimer J. R. 2001. Intestinal absorption of hexoses and amino acids: from apical cytosol to villus capillaries. J. Membr. Biol. 184: 233-239. Pappenheimer J. R. and Reiss K. Z. 1987. Contribution of solvent drag through intercellular junctions to absorption of nutrients by the small intestine of the rat. J. Membr. Biol. 100: 123-126. Pappenheimer J. R. and Volpp K. 1992. Transmucosal impedance of small intestine: correlation with transport of sugars a n d amino acids. Amer. J. Physiol. 263: C480-C493. Pencek R. R., Koyama Y., Lacy D. B., James F. D. 2002. Transporter-mediated absorption is the primary route of entry and is required for passive absorption of intestinal glucose into the blood of conscious dogs. J. Nufr. 132: 1929-1934. Pencek R. R., Koyama Y., Brooks Lacy D., James F. D. 2003. Prior exercise enhances passive absorption of intraduodenal glucose. J. Appl. Physiol. 95: 1132-1138. Powers D. R. and Nagy K. A. 1988. Field metabolic rate and food consumption by freeliving Anna's hummingbirds (Calypte anna). Physiol. Zool. 61: 500-506. Powers D. R. and Conley T, M. 1994. Field metabolic rate and food consumption of two sympatric hummingbird species in Southeastern Arizona. Condor 96: 141-150. Pyke G. H. and Waser N. M. 1981. The production of dilute nectars by hummingbird and honeyeater flowers. Biofropica 13: 260-270. Rand E. B., Depaoli A. M., Davidson N. O., Bell G. I. and Burant C. F. 1993. Sequence, tissue distribution, and functional characterization of the rat fructose transporter GLUT5. Amer. J. Physiol. 264: G1169-G1176. Rooke I. J., Bradshaw S. D. and Langworthy R. A. 1983. Aspects of water, electrolyte and carbohydrate physiology of the silvereye, Zosterops lateralis (Aves). A u s f . J. Zool. 31: 695-704. Sakai M., Imai T., Ohtake H., Azuma H. and Otagiri M. 1997. Effects of absorption enhancers on the transport of model compounds in Caco2 cell monolayers: assessment by confocal laser scanning microscopy. ]. Pharm. Sci. 86: 779-785. Sasaki H., Matsui C., Furuse K., Mimori-Kiyosue Y., Furuse M. and Tsukita S. 2003. Dynamic behavior of paired claudin strands within apposing plasma membranes. P Natl Acad Sci U S A 100: 3971-3976.
140
Physiological and ecological adaptations t o feeding in vertebrates
Schwartz R. M., Furne J. K. and Levitt M. D. 1995. Paracellular intestinal transport of six-carbon sugars is negligible in the rat. Gastroenterology 109: 1206-1213. Shi X. and Gisolfi C. V. 1996. Paracellular transport of water and carbohydrates during intestinal perfusion of protamine in the rat. Am. J. Med. Sci. 311: 107-112. Shi X., Schedl H. P., Summers R. M., Lambert G. P., 1997. Fructose transport mechanisms in humans. Gastroenterology 113: 1171-1179. Shu R., David E. S. and Ferraris R. P. 1997. Dietary fructose enhances intestinal fructose transport and GLUT5 expression in weaning rats. Amer. J. Physiol. 272: G446-G453. Skou J. C. and Esmann M. 1992. The Na, K-ATPase. J. Bioeng. Biomembr. 24: 249-261. Stagni G., O1Donnell D., Liu Y. J., Kellogg Jr., D. L., and Shepard A. M. 1999. Iontophoretic current and intradermal microdialysis recovery in humans. J. Pharmacol. Toxicol. Methods 41: 49-54. Starck J. M., Karasov W. H. and Afik D. 2000. Intestinal nutrient uptake measurements and tissue damage: validating the everted sleeves method. Physiol. Biochem. Zool. 73: 454-460. Sun H., Miller D. W. and Elmquist W. F, 2001. Effect of probenecid on fluorescein transport in the central nervous system using in vitro and in vivo models. Pharm. Res. 18: 1542-1549. Thorens B. 1993. Facilitated glucose transporters in epithelial cells. Annu. Rev. Physiol. 55: 591-608. Tsukita S. and Furuse M. 2000. Pores in the wall: claudins constitute tight junction strands containing aqueous pores. J. Cell Biol. 149: 13-16. Uhing M. R. and Kimura R. E. 1995a. Active transport of 3-0-methyl-glucose by the small intestine in chronically catheterized rats. J. Clin. Invest. 95: 2799-2805. Uhing M. R. and Kimura R. E. 1995b. The effect of surgical bowel manipulation and anesthesia on intestinal glucose absorption in rats. J. Clin. Invest. 95: 2790-2798. Ullrich K. J. and Papavassiliou F. 1985. Contralumenal transport of hexoses in the proximal convolution of the rat kidney in situ. Pfliigers Arch. 404: 150-156. Walters H. C., Craddock A. L., Fusegawa H., Willingham M. C., and Dawson P. A. 2000. Expression transport properties and chromosomal location of organic anion transporter subtype 3. Am. J. Physiol. 279: G1188-G1200. Welling P. G. 1986. Pharmacokinetics: Processes and Mathematics. Amer. Chem. Soc. Washington, D.C. Westergaard H., Holtermiiller K. H. and Dietschy J. M. 1986. Measurement of resistance of barriers to solute transport in vivo in rat jejunum. Am. J. Physiol. 250: G727G735. Williams J. B. 1993. Energetics of incubation in free-living orange-breasted sunbirds in South Africa. Condor 95: 115-126. Wright E. M., van 0 s C. H., and Mircheff A. K. 1980. Sugar uptake by intestinal basolateral membrane vesicles. Biochim. Biophys. Acta 597: 112-124. Zynel C.Y. and Wunder B.A. 2002. Limits to food intake by the Prairie Vole: effects of time for digestion. Funct. Ecol. 16: 58-66.
Mass-Balance Models for Animal Isotopic Ecology Carlos Martinez del Riol and Blair 0.WolfZ University Wyoming, ~ e ~ a r t m eof n tZoology Physiology, Laramie, WY, USA, University New Mexico, Department of Biology, Albuquerque, NM, USA
SYNOPSIS Analysis of natural stable isotope ratios has created a methodological upheaval in animal ecology. Because the distribution of stable isotopes in organisms follows reliable patterns, their analyses have become established useful methods for animal ecologists. However, because animal ecologists have adopted a phenomenological approach to the use of stable isotopes, the mechanisms that create isotope variation patterns remain unexplored. The mass-balance models that can provide a mechanistic, and hence predictive foundation for animal isotopic ecology are presented here. We review and elaborate the current mixing models used to reconstruct animal diets and develop new mathematical models to explain one of the most widely used patterns in animal isotopic ecology: enrichment in 15N observed acrpss trophic levels. Construction of element and isotope budgets is central to testing the mass-balance models described herein. Because the concept of a budget is central to all animal physiological ecology, development of a mechanistic and predictive framework for isotopic animal ecology falls naturally on physiological ecologists. We argue that progress in isotopic animal ecology hinges on laboratory experiments that explore mechanism, documentation of pattern in the field, and theoretical integration of mechanism and pattern.
Democritus was right: living organisms are collections of interacting atoms. We now believe that atoms are made of electrons clouding around a
142
Physiological and ecological adaptations t o feeding in vertebrates
nucleus made up of protons and neutrons. The numbers of charged particles (electrons and protons) within the atom are equal, so the whole atom is electrically neutral. The neutrons stop the nucleus from tearing itself apart. They work as a glue that bonds with the protons and provides cohesion within the nucleus. Elements with the same number of protons but a different number of neutrons are called isotopes and vary in mass. Most of these isotopes are stable (do not undergo radioactive decay) and can be distinguished by their mass. Many physicochemical processes are sensitive to differences in the dissociation energies of molecules, which often depend on the mass of the elements of which these molecules are made [Ball (2002) provides a particularly good introduction to atoms, elements, and isotopes]. The enzymatic pathways that organisms use to manufacture and transform organic molecules for example, can be isotopically discriminating. In general, it is easier to form, or break, bonds that contain lighter isotopes. The result is that molecules that contain the lighter isotope are preferentially incorporated into the products of incomplete reactions. As a result, the unreacted residues become enriched in the heavier isotope (Hoeffs, 1997). These isotopic effects are useful. The isotopic composition of many materials, including the tissues of organisms, often contains a label of the process that created it. Ecologists and physiologists can use these labels or isotopic signatures to detect the imprint of processes at a variety of scales. Plant physiologists, atmospheric scientists, and geochemists have relied on the measurement of natural stable isotope signatures for decades (Lajtha and Michener, 1994).Animal physiologists and ecologists, on the other hand, have been tardy in joining the isotopic research enterprise.Only one chapter in a recent review on the use of stable isotopes to integrate biological, ecological, and geochemical processes deals with animals (Griffith, 1998).Interestingly, the animals that this chapter deals with are extinct (Cerling et al., 1998 in Griffith, 1998).Indeed, paleontologists and archaeologists have been unusual among zoologists in their reliance on stable isotopes as tools in the reconstruction of the diets and habits of extinct animals and ancient humans (Koch et al., 1994 and references therein). Although zoologists have been latecomers, we have recently been active. The number of publications in animal ecology and physiological ecology that use stable isotopes has doubled every 3 years over the last 10 years (Fig. 6.1).This is a phenomenal rate of increase for the incorporation of any scientific methodology,As the following,almost certainly incompletelist attests, a large variety of phenomena in animal ecology can be informed by an isotopic approach. Stable isotopes have been used to reconstruct animal diets (Hobsonet al., 1994),determine patterns of resource allocation to reproduction (O'Brien et al., 2002), track animal migration (Hobson, 1999),assess the flux of materials from the sea into terrestrial food webs (Ben David et al., 1998), assign trophic levels (Post, 2002), and to determine the structure of food webs (France, 1995).
Mass-balance models
year Fig. 6.1. Number of publications on animal ecology and physiological ecology that rely on stable isotopes has increased exponentially (r2 = 0.81) in the last 10 years with a proportional rate of increase of 23%. Number of publications in this data set obtained by searching Biological Abstracts using "stable isotopes" and at least one of the following terms as key words: "animal", "diet", "food web", and "migration".
In geochemistry, plant physiology, physiological ecology, progress in the use of stable isotopes relies on vigorous interaction between theory, laboratory research, and field study [the chapters in Griffiths (1998)volume are superb examples].With few exceptions (someof which are reviewed below), animal ecologistshave adopted a different pathway. The vast majority of our field isotopic studies are phenomenological and a well-developed theoretical edifice does not inform our laboratory experiments. The objective of this chapter is to outline what we believe are some of the elements of a mechanistic theoretical framework for the isotopic ecology of animals. Like work in plant physiology and geochemistry,we too rely on mass-balance models to disentangle the relative importance of the factors that determine animal tissue stable isotopic composition. The two broad themes considered are (1) what is the timescale of incorporation of an isotopic signal into an animal's tissues and (2) why does the isotopic composition of animal tissues often differ from that of the resources they use. We review and elaborate on the current mixing models used to reconstruct animal diets and develop new models to explain one of the most widely used patterns in animal isotopic
144
Physiological and ecological adaptations t o feeding in vertebrates
ecology: enrichment in 15Nobserved across trophic levels. Although we focus on the isotopes of carbon and nitrogen for clarity, mass-balance models described here can be easily applied to other elements. Geochemists developed an arcane, but precise isotopicjargon. Before dealing with mass-balance models, we must introduce the terminology that isotopic ecology shares with the atmospheric and geological sciences.
STABLE ISOTOPES: TERMINOLOGY PRIMER The isotopic composition of a sample is measured as the ratio of one isotope to another: In most cases the abundance of one isotope (generally the lightest) exceeds that of the other by a large margin. For example, 13Cmakes up only 1.l0/0of the total carbon on earth and 15Nconstitutes only 0.37O/0 of the nitrogen (Richardson and McSween, 1989).Consequently this ratio can be a very small number. To make measurements of the relative abundance of two isotopes graspable, geochemists express the isotopic composition of most materials as the normalized ratio of the sample to a standard in parts per thousand (per md, %o):
a=[
sample
-
standard
]xlOOO
(1)
R stan,, where X is an element, and Rsamp, and R,an,a, are the ratios of the heavy to the light isotopes for the sample and standard, respectively, In some cases, it is useful to transform from fractions or percentages to ratios and 6 values with the transformation f
where f, is the fraction of the heavy isotope. For values of fH < 0.1, Rsample can be approximated very closely by f, (RSample=fH). Although some of the standards chosen by geochemists seem capricious to biologists, at this point we have no say in the matter. Amarine belemnite for the Pee Dee Formation (VPDB) and ocean water (standard mean ocean water = SMOW) are used as standards for nitrogen and carbon respectively. Thus, isotope ratios are commonly expressed as %o SMOW or %o VPDB. The words "depleted" and "enriched" refer to the heavy, and often less abundant isotope of a pair: Depleted means a more negative 6 value whereas enriched means a more positive 6 value.
Fractionation As mentioned in the introduction, the natural variation in the relative abundance of stable isotopes in any substance is the consequence of tiny mass differences that cause the isotopes to behave differently in both physical
Mass-balance models
145
processes and chemical reactions. In general, the lighter isotope (I2C,or 14N) tends to form weaker bonds and to react faster than the heavier isotope (13C, or 15N).As a consequence, the abundance of stable isotopes of an element will vary among the reactants and products of a chemical reaction. The change in isotopic abundance between chemical species (i.e. reactants or products) resulting from physical and chemical processes is called fractionation. Fractionation (a,.,) between the chemical species A and B is described in terms of the ratio in delta (6) values between the species:
Values of a are usually very close to 1, so the difference between two delta values is often reported and denoted by the discrimination factor aA.,(aA-, = - 6,). Two types of fractionation have relevance for biologists. Equilibrium fractionation occurs among chemical molecules linked by equilibria as a result of bond strength differences between the isotopic species. For example, carbonate in bone is probably derived from blood bicarbonate. Carbon and oxygen isotopes are rapidly exchanged among blood bicarbonate, dissolved blood carbon dioxide, and body water by the following equilibria:
The isotope equilibrium of bone carbonate is controlled by the composition of dissolved CO,, which is produced by respiration, and fractionation associated with equilibrium exchanges of carbon. Suppose that one is attempting to estimate the isotopic composition of the diet of an extinct mammal from the carbon in the apatite of its teeth. At mammalian body temperatures, the fractionation (E) from C0, to HCOl is about 8%0 (Mook, 1986). Assuming that E between dissolved bicarbonate and carbonate in apatite is l % o 0r2%~ (theE value for calcium carbonate), then apatite carbonate should have a 613Cvalue approximately9%0to 10%ogreater than that of respired CO, which presumably reflects that of diet. Kinetic fractionation effects occur because of differences in the rate of transport or rate of reaction of isotope species. For reactions catalyzed by enzymes, the magnitude of fractionation can be used to approximate the relative affinity of an enzyme for a compound with one isotope or another. An example of a kinetic fractionation is the reaction catalyzed by the enzyme glutamic oxaloacetic transaminase. This enzyme catalyzes the symmetrical reaction that transfers an amino group from glutamic acid to oxalacetic acid to yield a-ketoglutaric acid and aspartic acid (Mackoet al., 1986).Glutamic oxaloacetic transaminase transfers 14NH2from glutamic acid to aspartic acid 1.0083times faster than 15NH2.In the reverse reaction l4NH2is incorporated
146
Physiological and ecological adaptations t o feeding in vertebrates
into a-ketoglutarate 1.0017 times faster than 15NH. Transaminases catabolize nitrogen transfers for 12 other amino acids. Their kinetic discrimination against 15N may explain the observed 15Nenrichment between diet and nonessential amino acids, as well as the progressive enrichment in 15Nacross trophic levels (Gaebler et al., 1966; Post 2002; and this chapter).
Terminology : Caveat Many processes can lead to differencesin composition between an organism's tissues and its diet. For example, the enzymatic fractionation resulting from the action of transaminases described above, produces tissue proteins that most likely tend to be enriched in 15Nrelative to diet. Some of the differences between the isotopic composition of diet and a consumer's tissues are the result of fractionating processes. Others are the result of stoichiometric effects and what has been called isotopic routing (see subsequent sections). Because several processes can lead to differences in the isotopic composition of diet and animal tissues, it is inappropriate to call these differences "fractionation". Cerling and Harris (1999)proposed the term discrimination faca,,,) for the differencebetween the isotopic composition of tor (A,,, = 6t,ssu_diet and that of consumer tissues. Using the term "fractionation" to describe differences between the isotopic composition of a resource and the tissues of a consumer is inappropriate for two reasons: (1)it confuses pattern with process (fractionation is only one of the processes that produce discrimination) and (2) it is inconsistent with usage in other fields.
MIXING MODELS: GUIDE FORTHE PERPLEXED Stable isotopes are widely used to reconstruct animal diets. The basic idea has been summarized in the phrase "animals are what they eat". The isotopic composition of an animal tissue reflects the contribution of dietary components with different isotopic compositions (DeNiro and Epstein, 1978, 1981).Two types of approaches have been used to reconstruct animal diets from isotopic data: Euclidean distance methods and mixing mass-balance models (reviewed by Phillips, 2001). Phillips (2001)demonstrated that Euclidean distance methods do not estimate diet proportions correctly, Thus, we do not deal with these models here. Rather, we review in some detail mass-balancemixing models and their assumptions. Our description of rnixing models relies heavily on the papers by Phillips (2001)and Phillips and Koch (2001).
Linear Mixing Models The simplest of the mass-balance mixing models assumes that the isotopic composition of their tissues equals the weighed average of the isotopic composition of the diet constituents. For two diet constituents:
Mass-balance models
147
where p equals the fraction of diet A and 6XAand 6XBare the isotopic composition of diet components A and B. Provided that the isotopic composition of two elements is used, eqn (4) can be extended to estimate the fraction (pi)of the diet comprised by three types of items (Phillips,2001, Ben David and Schell, 2001). For carbon (C) and nitrogen (N) this requires solving the following system of linear equations in which A, B, and C are three different food types, and p, + p, + p, are the contributions of each food type to the animal diet:
= P A + P B+PC In general, one can use n-1 isotopes to discriminate the contribution of n food sources. Because eqns (4)and (5)depend linearly on p, the mixing relations they depict can be labeled "linear mixing models". Although eqns (4)and (5)look reasonable, they contain a variety of unrealistic assumptions.First, they assume that food types are stoichiometrically identical, i.e. that food type A and B contain exactly the same relative carbon and nitrogen contents. Second, they assume that all dietary items are assimilated with equal efficiency. Lastly, eqns (4) and (5) assume that isotopes are completely homogenized in the consumer's body prior to tissue synthesis. Phillips and Koch (2001)refined mixing models to incorporatedifferences in food stoichiometry and assimilation efficiency.We deal with the homogeneity assumption in a later section.
Concentration-dependent Mixing Models Phillips and Koch's (2001)concentration-dependentmixing models assume that the contribution of a given dietary item to an animal's carbon (or nitrogen) pool depends on how much carbon (or nitrogen)that item contains. The difference in the results of using linear mixing models and concentrationdependent mixing models is best illustrated with one isotope and two diet types. Let us call B the total assimilated biomass, p the fraction of total assimilation contributed by diet 1, (1-p)the fraction of total assimilation contributed by diet 2, and [C,] and [C,] the concentrations of element X in diets 1and 2, respectively.The relative contribution of diet 1to the pool of element X in the consumers tissues will be
and the isotopic composition of the pool of element X will be
148
Physiological and ecological adaptations to feeding in vertebrates
Figure. 6.2 illustrates the potentially large errors that can be committed by assuming a linear mixing model when the two diets differ significantly in elemental composition. The concentration-dependent mixing model can be easily modified for more than one isotope and more than one diet. Again, n-1 isotopes can be use to differentiate among n diets. For more than one isotope, pi is the fraction of total assimilated biomass (B) contributed by item i and Pxirepresents the fraction of assimilated element X Pxi
-
Bpix.
-
p.x.
~f. pix, 2 PjXj
'
j=1
(8)
j= 1
For three food sources, and two elements for example, carbon and nitrogen with concentrations [Ci]and [Nil (i = 1,2, and 3) and isotopic compositions 'Tiand 15Ni,we have that:
Of course, eqn (9)reduces to eqn (5)if all the diet components have the same elemental composition (i.e. [C,] = [C,] = [C,], and [N,]= [N, ] = [N,]). The error caused by neglecting concentration dependence increases as the differences in elemental composition among dietary components increase.With a bit of algebra, eqn (9) can be written in matrix form as a system of 3 linear equations in 3 unknowns: AP=B where
and
r P,I
P =I p,
lP3
roi
1,
1
and B =I 0 1
1
11
Phillips and Koch (2001) provide an algorithm to solve for vector P in eqn (10). The concentration-dependent mixing model proposed by Phillips and Koch (2001)assumes that all elements in a diet are assimilated with the same efficiency,which is not necessarily the case. Fortunately, element-dependent
Mass-balance models
149
variation in assimilation efficiency can be incorporated into the model. Let us call B' the total biomass ingested and eyithe efficiency with which element X is assimilated in diet i. Then eqn (8)must be modified as
j =l
j=1
Equation (9) has to be modified accordingly and the value of eximust be estimated experimentally. Although physiological ecologists estimate the assimilation efficiency for food types and even specific nutrients routinely, there are few accounts of the efficiency with which different elements are assimilated. We emphasize that the term exin eqn (11)represents "true" assimilation efficiency rather than the apparent assimilation efficiency so often reported. True assimilation efficiency is the fraction of the ingested element absorbed (i.e.ex= amount of element x not assimilated/amount of element x ingested), whereas apparent assimilation efficiency includes endogenous fecal losses (apparent assimilation efficiency = [amount of element x not assimilated + endogenous fecal losses]/amount of element x ingested). Readers can find a lucid explanation of the difference between true and apparent assimilation in Karasov (1990). Incorporating food stoichiometry in mixing models requires more empirical work. It requires analyzing (or at least estimating)the food's elemental composition and may require determining the efficiency with which different elements in each diet are assimilated. Field researchers may understandably complain that concentration-dependent models require more additional data and assumptions than simple linear mixing models (Robbins et al. 2002). However, simple models that make seriously wrong assumptions can yleld seriously erroneous results. The "collect-combust-and infer" approach that has characterized animal isotopic ecology so far has been fruitful. Although it will probably remain the approach of choice for some problems that can be solved by qualitativeapproaches, it has serious limitations. The simple linear mixing models that dominate the literature are misleading if the elemental composition of diet components differs substantially (Fig.6.2).Considering the potential effect of food's elemental composition and differential assimilation on isotopic incorporation adds realism to mixing models. In some cases, however, even the detail provided by concentration-dependentmodels may not suffice and additional assumptions may need to be incorporated.
Isotopic Routing Mass-balance mixing models make a crucial assumption which is almost certainly wrong in many animals. They assume that the isotopes of the elements contained in all dietary sources are completely homogenized ("mixed")
150
Physiological and ecological adaptations t o feeding in vertebrates
Fig. 6.2. Isotopic composition of a homogeneous mixture of two materials depends on two factors: 1) the fraction of each material in the mixture and 2) the elemental composition of the two materials. In the example depicted by the family of curves, p equals the fraction of material 1 in the mixture and 1-p is the fraction of material 2. The isotopic composition of materials 1 and 2 is 615N,=13.2and 615N, = -0.9 respectively. To construct the curves we maintained C, constant (C, = 0.12) and varied C, from 0.01 (punctate curve) to 0.12 (thick line). The values for C, in the remaining curves, from top to bottom are 0.02, 0.04, and 0.08. A linear mixing model [see eqn (4)]assumes that C, = C , and hence always predicts a straight line. Mixing models that account for the elemental composition of the mixture yield curves rather than straight lines if C, z C,. The isotopic and elemental compositions of this artificial example correspond to the values of salmon (material 1) and plants (material 2) ingested by brown bears (Ursus arctos, after Phillips and Koch, 2001).
in the animal body before tissues are synthesized. The animals that best fit this assumption are foregut fermenters in which nutrients are homogenized to the common denominator of volatile fatty acids and bacterial protein before being absorbed. However, even in ruminants many nutrients escape the fermentative chamber and are absorbed intact in the lower gut (Van Soest, 1994).Once absorbed, nutrients enter a variety of metabolic pathways and the elements in them can undergo varying degrees of mixing (Fig. 6.3).The mixing assumption is problematic whenever diet components differ in macronutrient content. Synthesis of one macronutrient from another can be difficult (e.g. glucose and glucogenic amino acids cannot be synthesized from fatty acids) and is always energetically expensive (Fig. 6.3). Thus,
Mass-balance models
sloughed cells, hair, etc.
sloughed
Acid Pools
Pool
uric acid
cr urea
t
'Yz
3-phosphoglyceraldehyde
carrier-Hz
IATplt(gbmgenica'a')
&Toketone
t pyruvic acid
acetyl CoA
4a-ketoglutaric acid
bodies
YT
cycle
~ a 2 r - H4~ Aw+p,=m
lactic acid
t ,
4
carrier
Fig. 6.3. Carbon and nitrogen in organisms are found as components of macronutrients. This scheme outlines the potential interconversions among the primary macronutrients in an animal body. Although there is potential for considerable mixing of elements among the different macronutrient pools, mixing may be energetically expensive. Recall that in general catabolism generates ATP. Synthesis, however, requires both ATP and reducing equivalent (such as NADHP). For example, lipids can be synthesized from both carbohydrates and proteins but lipid synthesis entails a high cost (the synthesis of a single palmitate molecule from 8 Acetyl-CoA requires 7 ATPs and 14 NADPH). In a similar fashion, although dispensable amino acids can be synthesized from the carbon skeletons resulting from both carbohydrate and lipid metabolism, this process is ATP dependent. Furthermore, the addition of amino acids to a peptide chain requires ATP (Mathews et al. 2000). A corollary of this observation is that organisms should route dietary macronutrients.
animals should route macronutrients and the isotopes in them from diet into the same macronutrient types in their tissues. This phenomenon has been called isotopic or nutrient routing. Paleontologists have recognized the problems posed by isotopic routing for quite some time. For example, anthropologists and paleontologistshave traditionally used bone collagen, largely composed of protein, to analyze isotopic composition for dietary reconstruction.Collagen has two problems: (1)it contains 33%glycine, which is a relatively 13C-enriched amino acid, so collagen tends to be 13C-enrichedrelative to other tissues; and (2) collagen is largely composed of protein and the composition of body protein in omnivores often reflects the isotopic composition of dietary protein (Ambroseand
Physiological and ecological adaptations t o feeding in vertebrates
152
Norr, 1993). Recognition of the principle that in omnivores the isotopic composition of tissue protein often reflects that of dietary protein, and not that o€bulk diet, has led researchers to analysis of the carbonates in bone apatite (Tieszen and Fagre, 1993).These are synthesized from circulating bicarbonate derived from CO, and hence probably reflect the components of the diet that are catabolized for energy (Ambrose and Norr, 1993).Omnivorous animals feeding on diets with low protein content should allocate dietary protein for tissue maintenance and repair, rather than catabolize it for energy. Consequently apatite carbonates may underestimate the contribution of dietary protein (see below).
Macronutrient Concentration-dependent Mixing Models The notion that protein should be routed to protein can be formalized in mixing models that depend on the content of macronutrients (protein, carbohydrate, and lipid) in each diet component. For element X (carbon or nitrogen) in protein:
a
tissue protein
1I
=(
P[P,I + (1-p)[P,
I
)(=Pl~[pll +=p,(l-p)[p,l)
(12)
where [PI]and [P,] are the protein contents of diet components 1 and 2 respectively and 6X,,, and 6X,,, are the isotopic compositions of the protein in these components. Equation (12) assumes that the concentration of element X in the protein of components 1 and 2 is the same, which is a reasonable assumption. It also assumes that protein in both components is assimilated with equal efficiency. Differences in the elemental concentration and in assimilation efficiency between the protein contained in diet components can be easily incorporated into eqn (12) [see eqns (8)and (lo)]. Figure 6.4 illustrates the difference between the results of a concentrationdependent model and one that incorporates differences in protein content between diet components. The differences between the results of the two models increase as the disparity in protein composition between diets increases. Note the large errors a concentration-dependentmodel can engender if there is routing. Suppose that [P,] = 0.07 and 613C= -24 for the example depicted in Fig. 6.4.If there is protein routing this value represents a p of 0.25. The concentration-dependentmodel would estimate p as 0.75! A protein concentration-dependentmixing model for n-1 isotopes and n diets can be constructed as:
xtissue
protein
-
i=l
2piepi[pi1 i =l
Mass-balance m o d e l s
-28
1 0
I
I
I
I
I
153
I
I
I
I
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 P
1
Fig. 6.4. Mixing models that assume isotopic routing of dietary protein into tissue protein yield results different from those of concentration-dependent models. In this example, we applied a protein routing model [eqn (11)] to two diets with contrasting isotopic composition (PIC, = -27 and 613C2= -12). Diet 1 had a fraction of assimilated biomass equal to p and a protein content ([P,] on a dry mass basis) of 0.75. The protein content of diet 2 ([I-',]) varied from 0.07 (dashed line) to 0.75 (thick line). The values for P, in the remaining curves, from top to bottom, are 0.14, 0.28, and 0.54. The concentration-dependent model [from eqn (7)] assumes that [C,] = 57.7 and [C,] = 46.8 (i.e. the values of lean deer meat and corn respectively. The thick line represents both the concentration-dependent model and a protein routing model that assumes identical protein contents in diet components 1 and 2. Because the carbon contents of the two diet components are similar, the concentration-dependent model yields a linear mixing relationship.
where pi is the fraction of total biomass ingested comprised by component i, eI,ithe efficiency with which the protein contained in diet component i is assimilated, [Pi] its protein content, and 6X, the isotopic composition for element X of the protein in component i. For N and C, the matrix form of this linear system is
A,P=B,
(14)
where
r (6l3CI- B3CtisSue )epl[p,I
I AP =
(613C2- G3CtisSue )ep2[p21 (613C3- 83Ctissue )eP3[P3 I1
( g 5 ~- ~, 5 ~ , . s u e ) e p l ([ pg l 5~ ~-,6 1 5 ~ t i)ep2 s s u[p21 e ( 6 1 5 ~-,615~tis,e)e,[p, II
1
1
1
1
154
Physiological and ecological adaptations to feeding in vertebrates
r pI1
and
~01
P =I p, 1 , and B =I 0 1
Ip3J
111
The mixing model described by eqn (12) assumes that protein is always routed into protein. This assumption is likely to be correct if, when eaten alone, all diet components satisfy the animal's protein requirements. However, many interesting situations involve one dietary component that does not provide sufficient protein but which is abundant and provides energy in the form of carbohydrates and lipids. Good examples of these situations are fruit- and nectar-eatingbirds that satisfy most of their protein requirements with insects (Martinezdel Rio, 1994)and carnivorous mammals that ingest fruit and plants in addition to meat (Pritchard and Robbins, 1990).When one of the diets is protein deficient, at low ingestion levels of the protein-rich alternative there is probably significant routing of carbon from carbohydrates and, to a lesser extent from lipids, to protein. A mixing model that incorporates this added level of realism requires many assumptions and cannot be solved analytically for p. It is useful, however, to ascertain how wrong the results of our mixing models can be. We explored the potential effect of isotopic routing among macronutrients with an artificial situation. We assumed a large animal (65 kg) that consumes two diet components with contrasting isotopic and macronutrient compositions(Fig.6.5).The protein content of these two diets is such that only one of the diet components (component 1) has sufficient protein to g day-') if it ingests satisfy the animal's minimal protein requirements (PRmir, enough of it to satisfy its energy needs. The other diet component (component 2) is protein deficient. We assumed this animal to ingest enough mass of each diet component to maintain neutral energy balance. However, we also assumed that the animal protein intake was insufficient to satisfy its minimum protein requirements when it ate only component 2. Assuming the total protein consumption as equal to or higher than PRmin, we calculated the animal tissue protein 613Cusing eqn (11).However, were protein ingestion less than PR-, we assumed that the carbon needed to synthesize the protein deficit was derived from amination of carbon skeletons derived from carbohydratesin component 2 (the carbohydrate-rich, but protein-deficient component):
'
'"'tissue
protein
=
{
I
((pRmn - B ( p [ y 1 + ( 1 - p ) [ P , 1 ) ) 6 q if PRmmn > B(p[ql+(l- p)[P21) P R ~ ~ ~ and
Mass-balance models
155
-
813Ccat(non-protein) concentrationdependent mixing model perfect routing (protein to protein)
.........0 ........ imperfect routing
Fig. 6.5. A more realistic model of isotopic routing for two diet components differing in protein content. The protein content and isotopic composition of diet components 1 and 2 are identical to those in Fig. 6.4. The energy content of diet component 1was 25.1 kJ g-' and that of diet component 2.17 kJ g-' and we assumed that the animal ingested a mixture of diet components that satisfied its daily maintenance energy requirements (11,241 kJ day-'). We also assumed that the minimal protein requirements (PRmi,)equaled 77 grams day-' (these data correspond to a 65 kg human exercising moderately; Reeds and Becket, 1996). Finally, we assumed that diet component 1 contained 25% lipid and no carbohydrates and component 2 contained 89% carbohydrate and 4% lipid. The thick curve assumes concentration-dependent mixing. Note that assuming that some carbohydrate is routed to protein (curve with open circles, "imperfect routing") at low protein intakes does not yield results very different from those of a mixing curve that assumes that only protein is routed into protein. The curve labeled with diamonds assumes that the isotopic composition of the nutrients catabolized for energy reflects a mixture of the components that remain after protein has been allocated to satisfy minimal protein requirements. If animals route protein to protein and allocate carbohydrates and lipids to energy production, a concentration-dependent mixing model underestimates the contribution of the protein-rich component. Using a concentration dependent-model a value of 613C= -14%0in exhaled breath (or bone apatite) corresponds with a p - 0.12. However, if there is protein routing p = 0.19 (dotted lines).
As Fig. 6.5 shows, making this more realistic assumption yields results very similar to those obtained by assuming a simple protein-to-protein routing model (the maximal difference between the two models is 1.4%O ). Equation (15) assumes that only carbohydrate carbon from diet component 2 is incorporated into protein. An alternative is to assume that the carbon used to compensate the protein deficit is derived from aminating carbon skeletons from a mix of all macronutrients. This assumption yields almost identical results as eqn (15).
156
Physiological and ecological adaptations t o feeding in vertebrates
The critical assumption of eqn (15) is that carbon skeletons of carbohydrates can be aminated. Why would an animal with limited protein transfer an amino group from protein into carbohydrate.The reason lies in the inefficiency of metabolism. Even animals in negative nitrogen balance continue to catabolize protein. O'Brien et al. (2000,2002) have shown that hawkmoths feeding on sucrose-rich but nitrogen-free nectar, use nitrogen stored as larvae to synthesize dispensable amino acids that are incorporated into egg proteins. These authors found the carbon isotopic composition of nectar sugars in dispensable ("nonessential") amino acids, but not in indispensable ("essential"). We speculate that when nitrogen is limited, the ammonia resulting from inevitable protein turnover will be incorporated into carbon skeletons derived from carbohydratesto synthesize dispensable (nonessential) amino acids. Of course, essential amino acids are always derived exclusively from dietary protein. This hypothesis can be tested by analyzing the composition of specific amino acids (see O'Brien, 2002). Equations (12)and (15)can be used to predict the carbon isotopic composition of tissue protein. Assuming that protein is routed to protein, the isotopic composition of the mixture of macronutrients catabolized for energy (Fl3CCat) should reflect the isotopic composition of carbon contributed by the macronutrients that remain after allocation to protein. Estimating Fl3CCat is a simple exercise in accounting but because the resultant equation is long and awkward, it is not presented here. The value of Fl3CCat reflects the increased importance of the protein-deficient diet as a source of energy. If there is protein routing and a concentration-dependent mixing model is used to estimate diet composition from the 613Cof nonprotein tissue (e.g.breath CO, and bone apatite; Hatch et al., 2002), the fraction of the protein-rich component (p) will be underestimated (Fig.6.5). Several authors have pointed out that the carbon isotope ratio of breath CO, is a reliable indicator of the carbon isotope composition of bulk diet (Hatch et al., 2002 and references therein). The results depicted in Fig. 6.5 cast doubt on this assertion. Indeed, if diet components differ in protein content, breath CO, will have a carbon isotope will composition closer to bulk diet than the 613Cof body protein. Yet 6l3CCat have a value biased toward that of the protein-deficient component of the diet. This bias should increase with (a) the protein requirements of the animal and (b)the disparity in protein content between dietary components. The message of our admittedly simplistic routing models may be disappointing for ecologists eager to use stable isotopes to find out what their study animals eat. Our models suggest that isotopic ecology and nutritional ecology are inextricably linked. To understand incorporationof the isotopic signal of different diet components into animal tissues it appears that we must know not only the macronutrient content of these components,but also the efficiency with which these macronutrients are assimilated. Using concentration-dependent or macronutrient content-dependent models to reconstruct animal diets using stable isotopes cannot be done with
Mass-balance models
157
confidence until we validate the relative performance of these models in the laboratory. The models described in this section outline a research agenda. Although we believe that isotopic ecology can benefit from the adoption of more realistic mixing models, there will always be a place for the simple mixing models outlined in eqns (4) and (5).These models can be used to accurately describe the proportion of diferent diets incorporated into a given tissue. As emphasized above, these proportions may/may not represent the proportion in which these diets are ingested. Using any of the models described above is "correct" provided that in each case the assumptions of the models are identified and that the limitations of the inferences that can be derived from them are recognized.
Compound Specific Isotopic Analysis: I s I t a Way Out? The ability to measure the isotopic composition in specificbiochemical compounds (e.g. fatty acids, cholesterol, and amino acids) suggests an alternative to avoid the problem of nutrient routing in diet reconstruction (Hammer et al., 1998).The isotopic composition of indispensable ("essential") nutrients that the animal cannot synthesize must reflect the composition of the mix of these nutrients in the diet (O'Brien et al., 2002). There is enormous potential for use of the isotopic composition of individual essential nutrients to sort out an animal's dietary components.However, this new level of technological sophistication does not liberate us from mixing models. It is an easy exercise to modify eqn (12)to determine diet components from the isotopic composition of an essential nutrient (just substitute Pi for E, the concentration of the essential nutrient E in diet, and FX, for FX,,). he isotopic composition of an essential nutrient in animal tissues is the result of the concentration of this nutrient in each diet component and thus its isotopic composition in the diet in toto.
DYNAMICS OF ISOTOPIC INCORPORATION Mixing models assume equilibrium. They assume that an animal has ingested the diet components in a fixed combination long enough for the isotopic composition of its tissues to have reached a steady state. Because animals shift diets such is rarely the case. Indeed, stable isotopes have proven useful tools in documenting diet shifts in animals (Wolf and Martinez del Rio, 2000).The isotopic composition of a tissue is the result of the integration of isotopic inputs over some time in the past. Thus, using tissues with different turnovers will give information of the past diet of an animal over different time intervals. The turnover rate of tissue constituents governs the time window of isotopic incorporation (Tieszen et al., 1983). The dynamics of incorporation of an isotopic "signature" into a tissue depends on the rate at which the materials in the tissue turn over. Consider
158
Physiological and ecological adaptations t o feeding in vertebrates
Fig. 6.6. In a pool of element X of size 4,the rate at which the amount of heavy isotope changes equals the difference between the rate at which this isotope enters the pool (Ap,f,) minus the rate at which it exits the pool (A?"f,). A, is the size of the pool (in mols), ro and ri are the fractional rates of input and output of element X into the pool (with units equal to time-') respectively, f, the fraction of heavy isotope in the pool, and f,, the fraction of heavy isotope in diet. 1-(
df, A , dt
a tissue that contains A, mols of the element in question. We can envision A, as the pool of element X in a tissue. Then consider the amount (A, = f,AT) of the heavy isotope (13Cor 15N)in this pool. Let us call ri and ro (with units equal to time-l)the fractionalrates at which the element enters and leaves the pool respectively, and f,, and f, the fractions of the heavy isotope in the incoming materials (d for diet) and in the pool (b for body) respectively (Fig. 6.6).Then
and
Combining eqns (16)and (17),and because, we have.
1 dA = (5 - ro) A, dt
(-)-
1
Mass-balance models
159
Assuming that at time 0, f, = f,(O), and that fHd(t) is constant (f,,(t) = fHd), we may then integrate eqn (18)to yield:
Several studies have investigated the change in tissue isotopic composition after an animal has been subjected to two diets that differ markedly in isotopic composition (O'Brien et al., 2000 and references therein). These authors used coexponential functions of the form
to describe their data. Because the heavy isotope is usually rare (i.e. fH << 0.1),fH = RSamp, and. fH = RShnda, (1000 6X + 1).Substituting this expression for fH into eqn (19) leads to eqn (20). At steady state (i.e. ri = r, = r), the exponent r in eqns (19) and (20) represents the fractional turnover rate of element X in a given tissue. Figure 6.7 shows incorporation of the isotopic signal from diet into tissues with contrasting fractional turnover rates. Some tissues, such as liver and plasma proteins, have high turnover rates, and their isotopic composition reflects integration of recent inputs. In Japanese quail for example, carbon in liver has a half-life of approximately 3 days
V)
3 V) V)
.+ -
+
0 0 $! GO
-20
-22 -24
-26
0
50
100 150 Time in days
200
250
Fig. 6.7. Equations (19) and (20) can be used to investigate the carbon isotopic turnover of different tissues. The plot above was modified from data presented in Hobson and Clark (1992).In the example, an animal raised on a diet with 613C= -24%0is shifted to a diet with 613C= -19.5. Liver has a high turnover (r = 0.27 d-l) and its isotopic composition tracks diet changes. In contrast, tissue with low turnover such as collagen (r = 0.004 d-') reflect past diets rather than actual diet, even 100 days after a diet change. For simplicity, we assumed no isotopic discrimination between diet and tissues-which is not the case.
160
Physiological and ecological adaptations t o feeding in vertebrates
%of Saguaro fruit
6' 3~ (pectoral muscle)
days
% Saguaro fruit
Fig. 6.8. Tissues with different carbon turnover rates show different patterns of isotopic incorporation in response to changes in diet. Wolf and Martinez del Rio (2000) measured the percentage by dry mass of saguaro fruit (a CAM plant with a G13C = 13.1) in the crop of white-winged Doves (Zenaida asiafica). They also measured the carbon isotope composition of tissue with high and low carbon turnover (liver and pectoral muscle respectively). Note that 613C,iv,r"tracks" the percentage of saguaro in dove crops, whereas GITpectoralis fails to reflect changes in the isotopic composition of diet. Indeed, 613C,iver and percent and percent saguaro in diet do saguaro in diet corelate positively (b), whereas 613Cpectoralis not (c). Each point in this Figure is the mean from measurements of 6 to 10 individuals and bars are standard errors.
(i.e.r = 0.27; Hobson and Clark, 1992).Thus, 90% of the carbon found in the liver of these animals was incorporated at any moment over approximately 9 days. Other tissues exhibited low turnover and their isotopic composition reflected integration of inputs over a longer time period. In young quail, carbon in collagen had a half-life of 173days (r = 0.056, Hobson and Clark, 1992).Therefore 9O0/0 of the carbon in this tissue was incorporated over 575 days. The choice of tissue for a dietary reconstruction study depends on the question asked. Tissues with high nutrient turnover rates will track isotopic changes in diet closely (Fig. 6.8) whereas tissues with low nutrient turnover rates will integrate an isotopic signature from a large temporal window resulting in a smoother, less steep curve. If the question is to determine how animals track the availability of resources, a tissue with high turnover rate (i.e.plasma proteins and liver) must be used. Conversely, if determinationof the importance of various items in the diet of an animal over a long time period is the obyective, then a tissue with low turnover must be used.
Mass-balance models
161
POTENTIAL MECHANISMS BEHIND A MAGICAL NUMBER WHY DOES 615NINCREASE WITH TROPHIC LEVEL? The simple model for the dynamics of isotopic incorporation described in the previous section assumes that the isotopic composition of the element leaving the pool equals that of the pool. Modifying this assumption may hold the answer to a perplexing pattern. Many animals are enriched in 15N relative to their diet. In a classic paper DeNiro and Epstein (1981) documented an average 3.4 %o enrichment in 615Nvalue for whole-body samples over diet. This enrichment in "N is very useful because it provides ecologists with a tool for estimating the trophic position of an animal. Post (2002) summarized the method for estimating the trophic position from nitrogen isotope measurements in a single equation:
trophic position =
A+
(6' 'N
~ e c n n d a ~consumer y
-
615N
ba\e
)
(21)
A ,,
where his the trophic position of the organism used to estimate 615hTbase (i.e. h = 1for primary producers), 615Nis estimated by collecting and combusting the whole consumer, and Anis the enrichment in P5N.Typically, ecologists assume that n equals 3.4%O. The 3.4%0 enrichment per trophic level has acquired a magical status (see Eggers and Jones, 2000). Although there is significant evidence suggesting that there is a 15Nenrichment with trophic level, the magnitude of this enrichment is quite variable (it ranges from -1 %o to 6%; Peterson and Fry, 1987; Post, 2002). A variety of factors can determine variation in 15Nenrichment but most of them remain unexplored. 615Nenrichment varies among tissues within an individual (reviewed by Kelly, 1999),among individuals dependand among species ing on the C:N ratios of diet (Adams and Sterner, 2000)~ depending on diet type (vertebrate and invertebrate diets, Kelly, 1999).The following sections do not answer why there is a 3.4%0enrichment in 15N across each trophic level. Instead, we identify the measurements needed for answering this question. We also recognize the potential reasons why the 15Nenrichment across trophic levels should vary. Mass-Balance Model for Body Nitrogen Why is there a 15Nenrichment across trophic levels? It appears that answering this question in a mechanistic fashion is the key to understanding and then interpreting correctly the 15Nenrichment associated with increased trophic level. Surprisingly, the physiological mechanisms that determine this enrichment are poorly understood. It is believed that during catabolism, amino acids with amine groups containing 15Nare disproportionately retained relative to those with arnine groups containing 14N(Mackoand Epstep, 1984;Gaebler et al., 1966). The result is that excreted urinary nitrogen tends
162
Physiological and ecological adaptations t o feeding i n vertebrates Feces
Nitrogen pool Other
< Diet L--
7 '
15
h
'
T
NT
N=f
f15Nb
15Nb
qro >
6 1 5 ~
h\
L _ -
J
/
au'
f15N,
Urine
.,
Fig. 6.9. Mass-balance model for whole body nitrogen isotopic composition. The model assumes that the animal ingests food containing a fraction f,, of at a rate r,. The animal assimilates a fraction e of this food and the assimilation process has an apparent fractionation equal to a,*.The nitrogen in food enters the body nitrogen pool which in this pool. The animal voids contains N, moles of nitrogen. 15Nrepresents a fraction f,, nitrogen at a rate equal to ro. A fraction q of all voided nitrogen exits the animal through a We fractionated route (urine). The apparent fractionation of urinary nitrogen equals aU*. assume that fecal nitrogen (with a fraction 1-q of all nitrogen lost) is voided with no fractionation. Equations (22) and (23) summarize this model.
.,
to be isotopically light (15Ndepleted) relative to an animal's diet or tissues (Steele and Daniel, 1978,Minagawa and Wada, 1984). Figure 6.9 formalizes this observation in a mass-balance model. Substituting the corresponding terms in eqn (18)we obtain a mass-balance model for IN:
which simplifies to
The terms in eqn (23) are the fractional rates of nitrogen assimilation and excretion (riand rorespectively),the apparent fractionations associated respectively), the with assimilation and urine production (ai*and au* fraction of nitrogen excreted through a fractionated route (q), and the fractions of 15Nin diet and animal body (f,,., and, f,,, respectively).We call a,*and au*,apparent, because these values are quotients of fractions rather than of ratios (i.e. a,*=51' N'a"i"latd'), and because a variety of processes can f15 N*
Mass-balance models
163
lead to differences in the isotope composition between diet and assimilated nitrogen and between body and urinary nitrogen. Effect of Diet Varying in Protein Content on 15Nb0dy,iet Omnivores can ingest diets that vary enormously in nitrogen content. What effect might this variation have on their nitrogen isotope composition?Physiologists recognize two types of excreted nitrogen: endogenous urinary nitrogen (EUN)and metabolic fecal nitrogen (MFN).EUN is composed primarily of nitrogenous waste products (ammonia,uric acid, urea, and creatinine) whereas MFN is composed of nonabsorbed digestive enzymes, intestinal cellular debris, and undigested bacteria and mucus (Robbins, 1993).The parameter q in eqn (23)can be interpreted as:
EUN '= EUN+MFN During protein catabolism the amino group must be removed from the carbon skeletons of amino acids and the resultant nitrogen voided in urine (Fig. 6.3). It is widely believed that the nitrogen in urine is depleted in 15Nrelative to the pool of nitrogen in the body (i.e.q*< 1;Macko et al., 1986 and references therein). Thus the parameter q should be of crucial importance for the interpretation of an animal's nitrogen isotopic composition. The relationship between q and protein intake is not simple. If the animal is in negative protein/nitrogen balance and catabolizing body tissue, q probably decreases with protein intake. Conversely, if the animal is in neutral protein/nitrogen balance, q should increasewith protein intake (Jackson, 1998).Here we consider only the second option. The effect of negative nitrogen balance on an animal's nitrogen isotopic composition is examined in the next section. If the animal is not growing, is in neutral protein balance, and is consuming the same diet over time, then ri' = roand
df15
= 0. Thus,
dt
Under this assumption, equilibrium nitrogen isotopic compo>ition (Il5 ), increases with q in an accelerating fashion (i.e.).df~5 A
For values of aul,the relationship between f,,, by a straight line (Fig.6.10):
->O dq
and
d'flsNb
dq2
>Oifq< 1
and q is well approximated
164
Physiological and ecological adaptations t o feeding i n vertebrates
EUN = EUN t MFN
Fig. 6.10. If the animal is at neutral energy balance, the discrimination factor between body and diet (15N,o,y~,ie,) depends on q, the fraction of total nitrogen lost that is excreted as urine. The curves in this figure were generated using eqn (25). We assumed that 615Nbody = 10 and that there is no apparent fractionation during food assimilation ( a,*=0). Each curve represents a different apparent fractionation of urinated nitrogen (aU).
The data needed to test the predictions of eqn (25) are easy to obtain but remarkably scarce. There is very little data on the nitrogen isotopic composition of EUN or MFN (Steele and Daniel, 1978; Minagawa and Wada, 1984) and almost no data on the relationship between q and protein intake (reviewed for humans and rats by Jackson, 1998).One of the few studies that has varied protein intake and measured the isotopic composition of animal tissues shows a pattern consistent with the predictions of eqn (15; Fig. 6.11). It is reasonable to assume that protein intake (and hence q) is larger in carnivores than in herbivores. Hence we conjecture that 15N,0dy-di.t should be higher in carnivores than in herbivores. Although in general, carnivores have tissues enriched in 15Nover those of herbivores (Kelly, 1999), to our knowledge no study has compared 15Nb0,y,iet between herbivores and carnivores in a systematic fashion.
The Effect of Protein Quality on l5NbOdydi,. A nonintuitive corollary of the dependence of 15N,0,y-di,on q is that 15Nb0,s-diet must also depend on the quality of the ingested protem. The term "protein quality" refers to the ability of a dietary protein to provide the needs of an animal. Protein quality is dependent on the match between the amino acid
Ivlass-balance models
Percent nitrogen in diet in Fig. 6.22. The discrimination factor between diet and plasma proteins (15Nplasma yellow-rumped warblers (Dendroica coronata) varied linearly (y = 1.58 + 0.21 $, r2 = 0.94, regression on the means) with nitrogen intake. To vary nitrogen intake, birds were fed diets containing a homogeneous mixture of bananas (low-protein diet) and mealworms (highprotein diet). Points are means and bars are standard deviations of 6 individulas. Birds were fed on the controlled diets for 21 days (S. F. Pearson et al., unpubl. data).
composition of dietary protein and the amino acid needs of the animal (Robbins, 1993).A deficiency of one, or several essential amino acids will lead to greater protein requirement and ingestion and hence to higher catabolism of the amino acids not needed for protein synthesis (i.e. to an increase in EUN and hence in q). Thus, we predict that 15N,y,i, will decrease as protein quality increases.
Do Fasting Animals Get Heavier? Effect of Negative Nitrogen Balance on 615N Several authors have hypothesized that animals in negative nitrogen balance become progressively enriched in 15N(reviewed by Ben David et al., 1999).Negative nitrogen balance implies that r0>ri1. The simplest situation to model is one in which the animal is fasting (ri = 0). Because in fasting animals all nitrogen losses are urinary, and hence q = 1, eqn. (23) reduces to
and thus
166
Physiological and ecological adaptations t o feeding in vertebrates
Fraction of nitrogen pool remaining (N, (t)/N, (0)) Fig. 6.12. Fasting animals become enriched in 15Nrelative to their prefast nitrogen isotopic composition when their nitrogen pool decreases as a result of fasting. The magnitude of 15Nenrichment decreases as a power function of the fraction of the original nitrogen pool remaining. Curves were constructed using eqn (30) and several values for the apparent fractionation of urinated nitrogen (aU*).
In fasting animals
dN dt
2 = -ro(t)~,(t)
0'
and hence
'
Substituting eqn (29)into eqn (28) and after a bit of algebra, we have
Equation (30) resembles a Rayleigh distillation (Robinson,2001). It predicts that f,, will decrease as a power function of the fraction of the original nitrogen pool remaining (Fig. 6.12).To our knowledge, no study has examined the relationship between nitrogen isotope enrichment and protein loss. Schmidt et al. (1999)measured 615Nin fasting earthworms (Lumbricusfestivus)
,
Mass-balance models
167
and found no effect of fast length on nitrogen isotopic composition. However, the worms lost little mass (11%)and their C:N ratios deci ?ased during the 2-month fast, suggesting that protein was spared. This study emphasizes that care must be exercised when using 615Nin studies that aim to use nitrogen isotopiccomposition to measure body condition. Hobson et al. (1993) found that Ross' geese (Chen rossii) females lost 45% of their pectoral muscle mass and 60% of their liver mass while incubating eggs. In support of our prediction, these losses led to significant 15Nenrichment. The relationship between 615Nand nitrogen loss has been expected on intuitive grounds for quite some time (Hobsonet al., 1993),and has prompted wildlife biologists to speculate that nitrogen isotopic composition can be used as a tool to assess body condition (see Ben David et al., 1999 and references therein). Unfortunately the term "body condition" has a multitude of possible definitions, not all of which are compatible with the use of 615Nas an index of condition. This multiplicity of meanings is reflected in a plethora of body condition indices (reviewedby Hayes and Shonkwiler,2001). Body condition indices are numerical estimates believed to reflect health, nutritional status, and/or fat content (Hayes and Shonkwiler, 2001). 615Ncan be used as an index of condition only if loss in body condition is equated with nitrogen loss. As is the case with all condition indices, before 615Ncan be used as a condition index in the field, it must be calibrated in the laboratory (Weatherhead and Brown, 1996). An animal that is not fasting can be in negative nitrogen balance (i.e.ro> r,').Under these conditions, f,,, reaches an asymptotic equilibrium given by
(note that eqn (31)reduces to eqn (25)if ro= riq).Animals in negative nitrogen balance will always have more positive 615Nvalues than animals in neutral and positive nitrogen balance (Fig. 6.13).
Are Growing Animals Lighter? Effect of Growth on 615N Equation (31)and Fig. 6.13 suggest that fl,, decreases a%the ratio of nitrogen assimilation to nitrogen losses (ri1/r0) increases. Indeed, f15, tends asymptotically to qf,,,, as this ratio becomes large. These results suggest that growing animals in positive protein balance will have nitrogen isotopic compositions that are more depleted in 15Nthan those of animals in neutral or negative nitrogen balance. If the growth trajectory of the animal and hence the values of ri'(t)and ro(t)are known, then eqn ( 22) can be integrated and the ontogenetic change in 615Npredicted. Without this information we are left with a qualitative prediction: 15Nb0dy,i.t should decrease with growth rate (i.e.with ri1/r0).
168
-
Physiological and ecological adaptations t o feeding in vertebrates
Negative N balance
Positive N balance ('growth*)
Relative growth (ri/ro)
Fig. 6.13. 15N enrichment relative to assimilated nitrogen decreases sharply with relative growth estimated by the ratio of nitrogen assimilation to nitrogen output (ri'/ro).A value of rll/ro= 0 indicates steady state. Curves were constructed using eqn (31) and several values for the apparent fractionation of urinated nitrogen (aU).
As far as we know, this prediction has never been examined quantitatively. Several studies have reported variation (and lack of variation) in &15N with age. Although the results of these studies are not clear cut, they seem to support the predictions of our model. The first two studies reviewed here appear to contradict the model's prediction. On close assessment, however, the reasons for rejecting the hypothesis become weak. Minagawa and Wada (1984)reported invariant 615Nwith age, and hence presumably with age, in two species of clams. They concluded that there was no growth effect on nitrogen isotopic composition.However, although the clams differed almost fourfold in length, they maintained relatively constant fractional growth rates of the nitrogen pool (Fig.3 in Minagawa and Wada, 1984).Our model predicts that age per se should have no effect on 15Nbc,dy,iet. We should find ontogenetic changes in l5NbOdydiet only if age and the growth rate of the nitrogen pool are related. Arnbrose (2000)reported no age-specificeffects on l5NbOdyin rats but he measured the isotopic composition as a function of age only &rats older than 100 days. At this age, however, rats are not growing at high rates and hence the effect of growth on 15Nb,y,iet should be small. Severalstudies have compared 15Nb,y,ietbetween mammalian young feeding on milk and their mothers (reviewed by Jenkins et al., 2001). Because
Mass-balance models
169
young should be growing whereas mothers should have relatively constant masses, our model predicts higher 15N,0dy,i, in mothers than in suckling young. In concordance with this prediction, Jenkins et al. (2001)found that although mothers were significantly enriched in 15Nrelative to their diets (average 15Nb0dy-d,e, = 4.1 %o),fast growing nursing offspring were only very slightly enriched in 15Nrelative to milk (average 15Nb, d i e t = 1.9 %o).Given that the quality of milk protein for suckling young is likefy to be very high (see above: Efect of Protein Quality on '5Nb(,dy-d,,) this result may not be unexpected and the result of both growth and efficient use of dietary protein. In a widely cited study, Hobson et al. (1993) fed Japanese quail (Coturnixjaponica) on diets of identical nutrient and isotopic composition but at different rations. One ration allowed the quail to grow while the other sufficed to allow the birds to maintain mass but not to increase it. As predicted by the model, growing birds were depleted in 15Nrelative to the birds that maintained body mass.
Testing and Refining Nitrogen Isotopic Composition Massbalance Models Mathematical models play a variety of roles. They are bookkeeping constructs. They allow us to summarize what we know and to identify what we do not know. At best, they allow making crisp predictions and hence provide a research road map. All models make assumptions, many of which may be unrealistic. Contrasting a model's prediction with data allows distinguishing essential from inconsequential assumptions. Throughout this report we have attempted to identify the assumptions of our models and the experimental measurements needed to test them. Here we reiterate one assumption implicit in our models that may be critical but whose importance is not easy to assess without data. Our models are one-compartment models. We have assumed that assimilated nitrogen enters and then exits a single, well-mixed pool. In principle, testing our models requires measuring the nitrogen isotopic composition of whole body nitrogen. The single pool assumption is incorrect. As mentioned above, different tissues show characteristic turnovers of carbon and nitrogen, strongly indicating the existence of several element pools. Furthermore, the physiological details of nitrogen flux in animals hint at the existence of distinct, albeit interacting pools. Assimilated amino acids enter the liver where a fraction is deaminated for energy and a fraction passes through to be distributed to replace catabolized amino acids in tissues (Young and Marcini, 1990).Furthermore, different tissues have contrasting protein turnover rates (Johnsonet al., 2001), and even different proteins can have widely diverging half-lives (Dice, 1987).The picture becomes even more complicated if we recognize that many animals exhibit significant nitrogen recycling. Urea synthesized by the liver is used by bacteria in the gastrointestinal tract to manufacture protein that can then be assimilated by the host (Lapierre and
170
Physiological and ecological adaptations t o feeding in vertebrates
Lobley, 2001). The bacterial populations in the gastrointestinal tract and the rest of the body certainly represent two distinct but closely interrelated nitrogen pools. Do we need to incorporate the existence of many nitrogen pools and the complex, and to a large extent still unclear, fine details of protein metabolism in our model? To answer this question we advocate adopting an approach that Lewontin (2000)branded methodological reductionism. We begin by testing the models that invoke the simplest assumptionsand determine their performance by contrasting their predictions with the results of well-designed experiments. Discrepancies between data and predictions should then guide our decision to revise our assumptions and assess the need to include more mechanistic detail in a new generation of models. In essence, we advocate applying the cyclicalnature of the hypothetico-deductivemethod to animal isotopic ecology. Although we advocate the testing of single pool models, for now we also emphasize that many applications of isotopic ecology must recognize the existence of several nitrogen pools. An appealing feature of the use of stable isotopes to answer ecological questions is that the methodology of ten does not require destructive sampling. We can inform ecological questions by sampling renewable tissues (e.g.blood cells, plasma proteins, muscle samples, hair, and feathers). Using different tissues may yield quan titatively different answers, and interpreting these differences may require recognizing the physiological differences among tissues. Suppose for example, that we are interested in using 15Nenrichment as an indicator of protein loss. Protein catabolism during fasting does not occur at the same rate in all tissues (Cherel et al., 1988;Schwilch et al., 2002). In addition, the fraction of amino acids exported intact relative to those deaminated in situ by aminotransferases and dehydratases varies among tissues (Raju et al., 1993).The interaction between these two factors probably dictates the de gree of 15Nenrichment in a given tissue as a function of fasting time. We need better data on the behavior of different tissues in response to the variables we have identified as potential influences on 15Ncomposition (protein content in diet, protein loss, and growth). This observation does not invalidate the usefulness of single pool models. It emphasizes the need to use these models to guide research at the same time that we examine their assumptions. Results of experiments needed to test the predictions of onecompartment models can reveal the existence of several pools (Johnsonet a1 1999)and if need be multicompartment rrlodels readily constructed (Faddy and Jones, 1988).
Mass-balance models
CONCLUDING REMARKS: ECOLOGICAL PATTERN, SEARCH FOR MECHANISTIC EXPLANATIONS, AND ROLE OF PHYSIOLOGICAL ECOLOGISTS
Testing the mass-balance models described in this chapter requires conducting experimentsthat entail taking fairly simple physiological measurements. In essence, we need to construct detailed budgets of elements and isotopes for animals under a variety of conditions. Later we shall need to unravel the details of the biochemical processes that lead to differences in the isotopic composition of what animals eat and what they defecate and excrete. The concept of a budget is central to all animal physiological ecology (Kooijman, 2000) and thus the creation of a mechanistic and predictive framework for isotopic animal ecology falls naturally on the shoulders of physiological ecologists. We can use the familiar experimentalparadigms of physiological ecology to provide the mechanisms that explain important ecological patterns at remarkably broad temporal and spatial scales. Analysis of natural stable isotope ratios has created a methodological revolution in animal ecology. Stable isotope analyses have become firmly established as useful methods for animal ecologists and one of the central tools for the study of feeding ecology. So far animal ecologists have adopted a phenomenological approach and relied on the search for pattern in the distribution of stable isotope ratios. Because these patterns appear to be robust, the approach has been very useful. Without mechanisms to explain them, however, we cannot ascertain the generality of the patterns discovered nor the boundaries of their usefulness. As Levin (1992) argued, without a mechanism, prediction in ecology is hazardous. The models presented in this chapter attempt to start the quest for a mechanistic and hence predictive foundation for animal isotopic ecology. Acknowledgments
We thank Denise Dearing and Scott McWilliams for their thorough and constructive comments. Charlie Robbins pointed out the potential connection between A15N,,,p,ie, and the biological value of dietary protein. He also helped us clarify the meaning of wrong and right (as applied to models). We are especially thankful to Max Bastiani for illustrating a previous draft. REFERENCES Adams T. S. and Sterner R. W. 2000. The effect of dietary nitrogen content on trophic level 15N enrichment. Lirnnol. Oceanogr. 45: 601-607. Ambrose S. H. 2000. Controlled diet and climate experiments on nitrogen isotope ratios of rats. In: Biogeochemical approaches fo paleodietary analysis S. H . Ambrose and M. A. Katzenberg (eds.). Kluwer/Acad. Publ., Dordrecht, Netherlands, pp. 243-259. Ambrose S. H. and Norr L. 1993. Carbon isotopic evidence for routing of dietary protein to bone collagen, and whole diet to bone apatite carbonate: Purified diet growth experiments. In: Molecular Archaeology of Prehistoric Human Bone J. Lambert, and G. Groupe (eds.). Springer Verlag, Berlin, pp. 1-37.
172
Physiological and ecological adaptations t o feeding in vertebrates
Ball P. 2002. The Ingredients: a Guided Tour of the Elements. Oxford Univ. Press, Oxford, UK. Ben-David M. and D. M. Schell 2001. Mixing models in analyses of diet using multiple stable isotopes: a response. Oecologia 127: 180-184. Ben-David M. T. A., Hanley and Schell D. M. 1998. Fertilization of terrestrial vegetation by spawning Pacific salmon: the role of flooding and predator activity. Oikos 83: 47-55. Ben-David M., McColl C. J., R. Boonstra and Karels T. J. 1999. 15Nsignatures do not reflect body condition in Arctic ground squirrels. Can. J. Zool. 77: 1373-1378. Cerling J. M., Harris J. M., and McFadden B. J. 1998. Carbon isotopes, diets of North American equids, and the evolution of North American C4 grasslands. In: Stable Isotopes: Integration of Biological, Ecological, and Geochemical Processes H . Griffiths (eds.). Bios Sci. Publ. Oxford, UK, pp. 363-377. Cerling T. E. and Harris J. M. 1999. Carbon isotope fractionation between diet and bioapatite in ungulate mammals and implications for ecological and paleontological studies. Oecologia 120: 347-363. Cherel Y., Robin J.-P., and Le Maho Y. 1988. Physiology and biochemistry of long-term fasting in birds. Can. J. Zool. 66, 159-166. DeNiro M. J. and Epstein S. 1978. Influence of diet on the distribution of carbon isotopes in animals. Geochimica et Cosmochimica Acta 42: 495-506. DeNiro M. J. and Epstein S. 1981. Influence of diet on the distribution of nitrogen isotopes in animals. Geochimica et Cosmochimica Acta 45: 341-351. Dice J. F. 1987. Molecular determinants of protein half-lives in eukaryotic cells. FASEB J. 1: 349-357. Eggers T. and Jones T. H. 2000. You are what you eat.. .or are you? Trends in E c o l o ~ jand Evolution 15: 265-266. Faddy M. J. and Jones M. C. 1988. Fitting time-dependent multicompartrnent models: a case study. Biometries 44: 587-594. France R. L. 1995. Differentiation between littoral and pelagic food webs in lakes using stable isotopes. Limnology and Oceanography 40: 1310-1313. Gaebler 0.H., Vitti T. G., and R. Vukmirovich 1966. Isotope effects in metabolism of 14N and 15N from unlabeled dietary proteins. Can. J. Biochem. 44: 1249-1257. Griffiths H. 1998. Stable Isotopes: Integration of Biological, Ecological, and Geochemical Processes. Bios Sci. Publ., Oxford, UK. Hammer B. T., Fogel, M., and Hoering T. C. 1998. Stable carbon isotope ratios of fatty acids in seagrass and redhead ducks. Chemical Geology 152: 2941. Hatch K. A., Pinshow B., and Speakman J. 2002. The analysis of 13C/12Crations in exhaled CO,: Its advantages and potential application to field research to infer diet over time, and substrate metabolism in birds. Integ. Comp. Biol. 42: 21-33. Hayes J. P. and Shonkwiller J. S. 2001. Morphological indicators of body condition: useful or wishful thinking. In: Body composition analysis of animal: a handbook of nondestructive methods J. R. Speakman (ed.).Cambridge Univ. Press, Cambridge, UK. Hobson K. A. 1999. Tracing origins and migration of wildlife using stable isotopes: a review. Oecologia 120: 314-326. Hobson K. A. and Clark R. G. 1992. Assesing avian diets using stable isotopes I: Turnover of 13C in tissues. Condor 94: 181-188. Hobson K. A., Alsaukas R. T., and Clark R. G. 1993. Stable nitrogen isotope enrichment in avian tissues due to fasting and nutritional stress: Implications for isotopic analysis of diet. Condor 95: 388-394. Hobson K. A., Piatt J. F., and Pitocchelli J. 1994. Using stable isotopes to determine seabird trophic relationships. 1. Animal Ecol. 63: 786-798. Hoeffs J. 1997. Stable Isotope Geochernistrtj. Springer-Verlag, New York, NY. Jackson A. A. 1998. Salvage of urea-nitrogen in the large bowel: functional significance in metabolic control and adaptation. Biochem. Soc. Trans. 26: 231-236.
Mass-balance models
173
Jenkins S. G., Partridge S. T., Stephenson T. R., Farley S. D., and Robbins C. T. 2001 Nitrogen and carbon isotope fractionation between mothers, neonates, and nursing offspring. Oecologia 129: 336-341. Johnson H. A., Baldwin R. L., France J., and Calvert C. C. 1999. A model of whole-body protein turnover based on leucine kinetics in rodents. 1. Ntlfr. 129: 728-739. Johnson H. A., Baldwin R. L., Klasing K. C., and Calvert C. C. 2001. Impact of separating amino acids between plasma, extracellular and intracellular compartments on estimating protein synthesis in rodents. Amino Acids 20: 389400. Karasov W. H. 1990. Digestion n birds: chemical and physiological determinants and ecological implications. Sttidies in Avian Biology 13: 391415. Kelly J. F. 1999. Stable isotopes of carbon and nitrogen in the study of avian and mammalian trophic ecology. Can. J. Zool. 78: 1-27. Koch P. L., Fogel M. L., and Tuross N. 1994. Tracing the diets of fossil animals using stable isotopes. In: Stable isotopes in ecology and environmental science, K. Lathja and R. H. Michener (eds.). Blackwell Sci. Publ., Oxford, UK, pp. 63-92. Kooijman S. A. L. M. 2000. Dynamic Energy and Mass Budgets in Biological Systems. Cambridge Univ. Press, Cambridge, UK. Lajtha K. and Michener R. H. 1994. Stable Isotopes in Ecology atzd Environmental Science. Blackwell Sci. Publ., New York, NY. Lapierre H. and Lobley G. E. 2001. Nitrogen recycling in ruminants: a review. J. Dairy Sci. 84: E223-236. Levin S. A. 1992. The problem of pattern and scale in ecology. Ecology 73: 1943-1967. Lewontin R. 2000. It ain't necessarily so: the dream of the human genome and other illusions. New York Review Books, New York, NY. Macko S. A. and Estep M. L. F. 1984. Microbial alteration of stable nitrogen and carbon isotopic compositions of organic matter. Org. Geochem. 6: 787-790. Macko S. A., Fogel M. L., Engel M. H., and Hare P. E. 1986. kinetic fractionation of stable nitrogen isotopes during amino acid transamination. Geochim. Cosmochimica Acta 50: 2143-2146. Martinez del Rio C. 1994. Nutritional ecology in nectar- and fruit-eating volant vertebrates. In: Food and form and function of the mammalian digestive tract D. Chivers and P. Langer (eds.). Cambridge Univ. Press, Cambridge, UK, pp. 103-127. Mathews C. K., van Holde K. E., and Ahern K. K. 2000. Biochemistry. Benjamin Cummins, New York, NY. Minagawa M. and Wada E. 1984. Stepwise enrichment of 15N along food chains: further evidence and the relation between 6l9I and animal age. Geochim. Cosmochim. Acta 48: 1135-140. Mook W. G. 1986. I3C in atmospheric CO,. Netherlands J. Sea Res. 20: 211-223. O'Brien D. M., Schrag D. P., and Martinez del Rio C. 2000. Allocation to reproduction in a hawkmoth: a quantitative analysis using stable isotopes. Ecology 81: 2822-2831. O'Brien D. M., Fogel M. L., and Boggs C. L. 2002. Renewable and non-renewable resources: amino acid turnover and allocation to reproduction in lepidoptera. Proc. Natl. Acad. Sci. U S A 99: 4413-4418. Peterson B. J., and Fry B. 1987. Stable isotopes in ecosystem studies. Ann. Rev. Ecol. Syst. 18: 293-320. Phillips D. L. 2001. Mixing models in analyses of diet using multiple stable isotopes: a critique. Oecologia 127: 166-170. Phillips D. L, and P. L. Koch 2002. Incorporating concentration dependence in stable isotope mixing models. Oecologia 130:114-125. Post D. M. 2002. Using stable isotopes to estimate trophc position: models, methods, and assumption. Ecology 83: 703-718. Pritchard G. T. and Robbins C. T. 1990. Digestive and metabolic efficiencies of grizzly and black bears. Can J. Zool. 68: 1645-1651.
174
Physiological and ecological adaptations t o feeding i n vertebrates
Raju M. R. D., Vijayalakshmi K., and Karuna K. M. 1993. Influence of dietary protein on aminotransfrases and phosphatases in rat tissues. ]. Food Sci. Tech. 30: 183-186. Reeds P. J. and Beckett P. R. 1996. Protein and amino acids. In: Present knowledge in nutrition E. E. Ziegler and L. J. Filer (eds.). ILSI Press, Washington DC, pp. 67-86. Richardson S. M. and McSween H. Y. 1989. Geochemistry: pathways and processes. Prentice Hall, New York, NY. Robbins C. T. 1993. Wildlife Feeding and Nutrition. Acad. Press, New York, NY. Robbins C. T. Hilderbrand G . V., and Farley S. D. 2002. Incorporating concentration dependence in stable isotope mixing models: a response to Phillips and Koch. Oecologia 133: 10-13. Robinson D. 2001. 615N as an integrator of the nitrogen cycle. Trends Ecol. Evol. 16: 153-162. Schmidt O., Scrimgeour C. M., and Curry J. 1999. Carbon and nitrogen isotope ratios in body tissue and mucus of feeding and fasting earthworms (Lumbricusfestivus).Oecologia 118: 9-15. Schwilch R., Grattarola A., Spina F., and Jenni L. 2002. Protein loss during longdistance migratory flight in passerine birds: adaptation and constraint ]. Exp. Biol. 205: 687-695. Steele K. W. and Daniel R. M. 1978. Fractionation of nitrogen isotopes by animals: A further complication to the use of variation in the natural abundance of 15Nfor tracer studies. ]. Agric. Sci. 90: 7-9. Tieszen L. L., Boutton T. W., Tesdahl K. G., and Slade N. A. 1983. Fractionation and turnover of stable carbon isotopes in animal tissues: implications for 613C analysis of diet. Oecologia 57: 32-37. Tieszen L. L. and Fagre T. 1993. Effect of diet quality and composition on the isotopic composition of respiratory CO,, bone collagen, boapatite, and soft tissues. In: Molecular Archaeology of Prehistoric Human Bone J . Lambert, and G. Groupe (eds.). Springer Verlag, Berlin, pp. 123-135 Van Soest P. 1994. Nutritional Ecology of the Ruminant. Cornell Univ. Press, Ithaca, NY. Weatherhead P. J. and Brown G. P. 1996. Measurement versus estimation of condition in snakes. Can. ]. Zool. 74: 1617-1621. Wolf B. 0. and Martinez del Rio C. 2000. Use of saguaro fruit by White-winged doves: isotopic evidence of a tight ecological association. Oecologia 124: 536-543. Young V. R. and Marcini J. S. 1990. Mechanisms and nutritional significance of metabolic responses to altered intakes of protein and amino acids, with reference to nutritional adaptation in humans. Amer. 1. Clin. Nutr. 51: 270-289.
Structural Flexibility of the Digestive System of Tetrapods - Patterns and Processes at the Cellular and Tissue Level J. Matthias Starck University Munich (LMU), Department of Biology 11, Munich, Germany
SYNOPSIS This chapter addresses the question of how animals adjust their gastrointestinal system to fluctuating contitions in food availability and food quality. I begin with a brief summary of the main external ecological factors that affect food availability and food composition. Circannual seasonality provides a simple trigger for anticipation of diet switches. Changes in organ size and functions may be coupled to the endogenous h ~ r m o : , ~control ?l system and a variety of strategies has evolved to cope with such predictable fluctuations in food supply. However, unpredictable changes in food supply cannot be coupled to internal regulation systems and require immediate environment-organism interactions. This chapter is about such immediate responses of the organisms to unpredictable diet switches. The focus of the chapter is on the pattern of changes in organ size and function in birds, 'reptiles' and mammals and how tissues and cells drive the sometime considerable phenotypic changes. The chapter closes with a comparative analysis of patterns and processes underlying gastrointestinal flexibility and a reconstruction of its evolutionary history.
Seasonal changes in temperature, humidity, and daylength cause fluctuations in abundance of plants and animals that may be consumed as food by vertebrates. Such environmental changes shape most terrestrial and freshwater habitats and, to a lesser degree, marine habitats. Seasonal fluctuations
176
Physiological and ecological adaptations t o feeding in vertebrates
are extreme in polar regions, distinct in boreal and temperate regions, and also occur in subtropical and tropical habitats. The major patterns of seasonal fluctuations are predictable and organisms may use endogenous rhythms and environmental cues to adjust to changes in feeding conditions. Four strategies have evolved to cope with such seasonal fluctuations in food availability: (1)phenotype change and diet switching, (2)migration to alternate feeding grounds, (3)hibernation, and (4) fasting tolerance. In contrast to the seasonal changes of the environment,short-termchanges in food availabilityare unpredictable in timing and duration. Unpredictable fluctuations in food composition or food availability elicit responses from the organisms that are rapid, reversible, and repeatable to allow for phenotypic changes synchronized with the altered environment. The signals that initiate flexible changes of the phenotype must come from the environment and phenotypic change must be sufficiently fast to match environmental fluctuations. Presumably, responses to predictable and unpredictable fluctuations may be governed by different regulatory pathways, i.e. predictable fluctuations can be anticipated and thus integrated into the endogenous control system, while unpredictable fluctuations cannot be anticipated and need to be linked to an environmental signal. These types of phenotypic flexibilitymay be independent of each other. From an evolutionaryand physiological point of view, this distinction is important because organisms that undergo seasonal changes of their phenotype may lack the flexibility to adjust to short-term fluctuations, and vice versa. This appears to be an extreme and it is more lilely that both types of phenotypic flexibility are integrated in such a manner that only limited seasonal windows open during which organisms possess the ability to flexibly adjust their phenotype to fluctuating environmental conditions (seebelow). This review focuses on short-term responses of organisms to changes in food composition and food availability. Responses to diet switching and alternations of feeding and fasting at the level of tissues and cells are summarized without reference to genetic underpinning of flexible phenotypes. Birds, mammals, and ectotherm sauropsids are exemplified as little is known about amphibians and virtually nothing about fish. However, even such a sketchy picture allows development of some ideas about the evolutionary history of phenotypic flexibility if the data are analyzed in a phylogenetic framework. In that context, a key question is whether the flexibility of the gastrointestinal system and its underlying mechanisms evolved once or whether it evolved independently in the different clades of vertebrates. Many species fast when food is not available rather than switching diet. We consider fasting a physiologically normal response to food nonavailability characterized by depletion of fat stores. It is important to distinguish clearly between fasting and starvation. Time is not a valid criterion because some birds and some mammals tolerate long fasting periods, e.g. up to 120 days in emperor penguins (Aptenodytesforsteri, Le Maho et al., 1976,1977;
Flexibility of digestive system
177
Groscolas 1982,1986) or 4 to 5 months in polar bears (Oftedal et al. 1993). Some ectotherm sauropsids tolerate fasting periods of a year or even longer (Pope 1961;Greene 1997).Physiological criteria for fasting are metabolization of adipose tissue, associated with relatively high plasma levels of free fatty acids, as well as low levels of urea, uric acid, and creatine phosphate. When lipid reserves are depleted, animals utilize body protein at increasinglyhigher rates. Utilization of body protein may lead to starvation and may cause irreversible damage to the phenotype. It is associated with rapidly increasing levels of creatine phosphate, urea, and uric acid.
BIRDS: DIGESTIVE SYSTEMS BETWEEN DIET SWITCHING, HYPERPHAGIA, AND FASTING
Changes of gut length or gizzard size of birds in response to diet change are well documented and have recently been reviewed (Battley and Piersma, 2004; Starck, 1999,2003; Piersma and Lindstrom, 1997; Piersma and Drent, 2003). The first observations of flexible organ size date back to the 19t1'century when they fueled a discussion about Darwinian and Lamarckian principles of evolution (Hunter, 1839; Owen, 1861; Holmgren, 1872; Brandes, 1896; Fermi and Repetto, 1901; Babtik, 1903a,b, 1906).Today, the evolutionary principles of phenotypic flexibility are better known and it is widely accepted that phenotypic flexibility is an adaptive trait of organisms, even though fitness correlations are rare and the genetic principles debatable (Via et al., 1995; Schlichting and Pigliucci, 1998; Pigliucci, 2001; West-Eberhard, 2003). However, many studies are incomplete at two levels: (1)only a few reports distinguish clearly between seasonal (anticipated) adjustments and environmentally induced responses to changes in diet. Thus it is difficult, if not impossible, to estimate the relative contribution of each component. (2) the mechanistic basis, i.e. processes at the cellular and tissue level that allow for restructuring of the different parts of the gastrointestinal system, have only been investigated in a few studies.
InducedChangesvsSeasonalChanges The first observations of organ size plasticity date back to Hunter (1839)who maintained pigeons solely on a meat diet and induced a "raptor-stomach" in them. More than 100 years later, Spitzer (1972) observed considerable seasonal changes in gizzard size and small intestine length in bearded tits (Panurus biarmicus).A th-walled gizzard during summer and a thick muscular gizzard during winter were related to a seasonal change in diet from insects in summer to seeds during winter. Spitzer (1972) presented nonexperimental but notwithstanding convincing, evidence that morphometric changes of the gastrointestinal tract were elicited by fluctuations in food type. However, he was not able to distinguish induced from seasonal
Flexibility of digestive system
179
experimentally partitioned for endogenous regulation and external stimulus.These papers suggest a combined effect of endogenouscontrol (i.e. anticipatory adjustment of organ size to expected changes in environment and endogenous demand) and instantaneous responses to changes in environmental conditions; they do not estimate what the specific contributions were. Chapter 8 by Battley and Piersma (this volume) reviews such changes in greater detail. Bauchinger and Biebach (2001), Biebach and Bauchinger (2002), and Deerenberg et al. (2002)studied organ size changes in long-distancemigrating garden warblers (Sylvia borin). During spring migration they found considerable plasticity of virtually all organs including the skeletal musculature. During periods of extended flights without food consumption, e.g. trans-sahara migration, organ size declined. Internal organs increased during stopoverswhen the birds had an opportunity to feed and refuel. Size changes in the gastrointestinalsystem were in the range of 50% and about 2 days were required to reconstruct gut size during a stopover. Interestingly, the gonads continued to increase in size, independent of feeding condition (Bauchinger, 2002). Hume and Biebach (1996)exposed birds to periods of fasting and refeeding during spring migratory restlessness. They observed considerable variation of organ size, i.e. fasting birds had smaller guts than feeding birds. Of course, these birds were endogenously triggered for longdistance migration. Karasov et al. (2004) performed a similar experiment with blackcaps (Sylviaatricapilla) captured in the field during spring migratiori and then exposed to various fasting intervals. These birds also showed considerable flexibility of the gastrointestinal system. This study further provided some insights into cellular mechanisms of organ size change. When fasted for two days, the birds reduced the length of intestinal villi about 50Y0 by shedding part of their mucosa epithelium (Fig. 7.2). When refed, they reestablished a full-size and functional mucosal epithelium within two days. The functional interpretations of these studies are straightforward and suggest a mass reduction of unutilized systems during flight, thus saving costs of maintenance and total energeticexpenditure(Biebachand Bauchinger, 2002). They also suggest that internal organs, in particular flight and leg muscles, might be used as protein resource (Bauchingerand Biebach, 2001). However, it remains unclear whether such flexibility is restricted to periods of spring (and autumn) migration or whether they can be elicited throughout the year. An analysis of seasonal effects on flexibility is clearly required to partition between environment induced changes and endogenously controlled flexibility. Savory and Gentle (1976a,b)were the first to show that Japanese quail responded to diet switching as a possible seasonal effects by increasing different parts of their gastrointestinaltract. Only a few laboratory-controlled studies of the cellular changes that underlie changes in the mass of the gastrointestinal organs have been undertaken. After three days fasting, the
180
Physiological and ecological adaptations t o feeding in vertebrates
Fig. 7.2. Intestinal histology of blackcaps (Sylvia africapilla) differed among treatment groups. (A) Ad libitum fed bird; arrows indicate cell extrusion zone at tip of the villi. (B) Close-up of A; arrow indicates blood vessel, * indicates lymphatic spaces. (C) Fasted bird; disintegration of mucosal epithelium apparent at arrow tips. (D) Close-up of C; arrow shows area of disintegration. (E) Refed bird; normal histology has been reestablished. Extrusion zones reformed at tip of villi (arrow in E). Apical part of enterocytes loaded with vesicles (asterisk in F). All micrographs are of jejunum. For A, C, E scale bar 0.5 mm; for B, D, F scale bar 0.05 mm (from Karasov et al. 2004).
intestinal villi of young and adult white Leghorn chickens returned to normal length within one day of feeding (Yamauchi et al., 1996).In 3-day fasted chicks, significant increments in villus height were apparent already after 3 h of feeding (Shamoto and Yamauchi, 2000). Also, Karasov et al. (2004)
Flexibility of digestive system
I
I
I
I
I
I
0
2
4
6
8
I
L
10
12
Times in experiment (Week)
-P!
15%
30%
45%
60% 70%
% crude fiber in food
Fig. 7.3. Flexibility of the gizzard in Japanese quail (Coturnix japonica). (A) Up- and downregulation of gizzard size in response to repeated diet switches form standard food to a high fiber diet. (B) Exposed to increasing levels of high fiber contents in the diet, gizzard responsiveness reaches an upper limit at 45% crude fiber (from Starck, 1999).
observed that the initiation of the reconstitution of the mucosal epithelium might begin within four hours of feeding in blackcaps. Studies on Japanese quail (Starck and Kloss, 1995; Starck 1996, 1999b; Starck and Rahmaan, 2003) analyzed the scaled and reproducible response of the quail's gizzard and small intestine, i.e. up- and down-regulation of organ size, to changes in diet in a laboratory setup independent of season (Fig. 7.3). Responses were fast enough to accommodate a 25% size change in less than 2 days. The flexible changes after diet switching were fully reversible, i.e. organ size returned to original values and could be elicited repeatedly.
182
Physiological and ecological adaptations t o feeding in vertebrates
Phenotypic Responses of the Gastrointestinal System: Patterns and Processes a t the Level of Cells and Tissue Cellular and tissue-related changes underlying the remarkable size changes of the small intestine and gizzard in response to diet change have been studied in Japanese quail. Two mechanisms have been described: (1)changes in number of cells of an organ; i.e. hyperplasia for up-regulation and cellular dystrophy for down-regulation of organ size; (2) changes in cell size, i.e. hypertrophy for up-regulation and cellular hypotrophy for down-regulation. The size of the gizzard of Japanese quail changed by more than 100% when diet was switched from standard food to one containing a high percentage of nondigestible fiber (Starck, 199913; Fig. 7.3).These changes were rapid, reversible, and repeatable. Within 5 days after diet switching, the gizzard mass and length had doubled. This increase was exclusively based on hypertrophy (i.e.increase in cell size) of smooth muscle cells of the musculus crassus cranioventralis and m. crassus caudodorsalis (Starck and Rahmaan, 2003), while decreasing gizzard size during fasting could be attributed to smooth muscle cell hypotrophy. In no case was mitotic activity of satellite cells observed or any change in cell numbers, indicating that neither hyperplasia nor cellular dystrophy contributed to size changes in the gizzard. The small intestine of birds undergoes similar changes in size if diet is switched from standard food to low-quality food (see Chapter 8 by Battley and Piersma, this volume). Its length may increase by 30% within 1-2 weeks after diet switching. The functional relevance of increased gut length is a longer retention time for food and thus a potentially improved apparent digestive efficiency. However, this relationship has not been tested empirically in the context of flexible phenotypes. Another functionally important parameter is the absorptive surface of the mucosal epithelium. The absorptive surface increased by 50% when quail were fed a high fiber diet (40% nondigestible fiber). The cellular processes underlying those size changes are based on the cellular turnover that maintains the mucosal epithelium. The mechanism clearly differs from that observed in the gizzard. When mass is maintained, the mucosal epithelium of birds (and mammals) undergoes a permanent cellular turnover, i.e. cells proliferate from the intestinal crypts, migrating over the surface to the tip of the villi where they are shed into the lumen of the gut. Theoretically,the surface magnification of the small intestine can increase when the rate of cell proliferation increases or when the rate of shedding decreases. When the food of Japanese quail was switched from standard to a high fiber diet, cell proliferation rates of the intestinal crypts increased by 150% (Starck and Rahmann, 2003). The association of increasing mitotic activity in the intestinal crypts and an enlarged epithelial surface suggests that diet-switching triggered the increase in cell proliferation rate, resulting in an enlarged surface magnification of the small
Flexibility of digestive system
183
intestine. The regulatory mechanisms remain unknown. Also, contributions by other tissue compartments of the small intestine, i.e. the connective tissue, smooth muscle layer and all other associated tissues have largely been neglected. For birds, the mechanism of down-regulatingthe gastrointestinal system during fasting is more or less not known. A study of blackcaps in migratory restlessness at their stopover sites in Israel (Karasov et al., 2004) showed a surprisingly radical shedding of the mucosal epithelium of the small intestine when birds were fasted for 24 - 48 h. The distal half of the villi was shed into the lumen of the gut (Karasov et al., 2004). It is completely unclear at present to what degree the shed epithelium is recycled as a source of energy and protein. Although one should be careful with generalizations, these findings indicate an astonishing flexibility of the mucosal epithelium based on unexpected mechanisms.
CROCODILES Crocodiles are the closest living relatives of birds but their natural history differs strikingly.Crocodiles are ectothermic and carnivorous. When food is not available, they tolerate long fasting periods. Thus, the functional state of their gastrointestinal system switches from fasting to digesting. The internal anatomy of their gastrointestinalsystem, however, shows some similarities to that of birds. The muscular stomach is sac-like, resembling the stomach of carnivorousbirds and the arrangement of the musculature in the stomach
Time (days in experiment) Fig. 7.4. Size changes of the duodenum of Caiman latirostris after feeding. The arrow indicates feeding. The thickness of the mucosa was measured with ultrasonography on life caiman. Data are mean + SD from animals (from Starck et al., unpubl.)
184
Physiological and ecological adaptations t o feeding i n vertebrates
wall is similar to that of birds (Dieslich, 1939).The muscular stomach connects to the duodenum which, as in birds, is U-shaped. The small intestine has few external or internal histological differentiations except for a continuously decreasing length of villi and a continuous increase in number of goblet cells toward the posterior end of the gut. Typically, the small intestine of crocodiles has deep intestinal crypts, as in birds and mammals. In other "reptiles", e.g. snakes, intestinal crypts are not fully developed or may be completely lacking (Luppa, 1977). When crocodiles switch from fasting to feeding, the length of the villi doubles within 48 h (Fig.7.4).Doubling of villi length is possible because the configuration of the mucosal epithelium changes from a pseudostratified epithelium to a single-layered high prismatic epithelium. On all levels of investigation, from light microscopy to ultrastructure, the observed structural changes are identical to that described for a variety of snake species (see below). The functional epithelium is a single-layered high prismatic epithelium and does not differ from the equivalent functional epithelium in mammals, birds or crocodiles. In contrast, fasting cells are densely packed into a transitional or pseudostratified epithelium that rests on a folded basal membrane. Even on the ultrastructural level, we find large amounts of spare membrane as an adaptation for loading of epithelial cells with lipid droplets (Fig. 7.5).The configuration change from transitional to single-layered epithelium is supposedly driven by a hydraulic mechanism that involves increased blood flow volume and lymphatic volume being pumped into the villi as well as the enterocytes being loaded with lipid droplets. Thus, it seems safe to suggest that the same structures support the flexibility of mucosal thickness in crocodiles and snakes. A thin,"stratified" epithelium had already been observed by Reese (1915)in fasting (hibernating) alligators (Alligator mississippiensis) while feeding alligators had a much thicker mucosa. Taguchi (1920) published a detailed histological study of the complete gastrointestinal tract of several other crocodile species, but did not mention their feeding condition. However, his figures show a typical pseudostratified mucosal epithelium for the posterior part of the gut. Saint Girons (1976) refers to the small intestine of "reptiles" as stratified or pseudostratified, again without reference to the digestive status of the animals..Only when the feeding condition was experimentally controlled did it become obvious that epithelial configuration was associated with physiological condition. Digesting crocodiles always had a single-layered epithelium which presumably absorbs nutrients from the gut while fasting crocodiles always had a thick pseudostratified epithelium (Starck et al., 2003; Starck, Abe and Neto, unpubl. data).
Flexibility of digestive system
185
Fig. 7.5. Caiman latirostris, histological changes of the small intestinal mucosa epithelium; fasting animals (top), digesting animals (bottom). After feeding, the epithelium changes configuration from a pseudostratified to a single-layered high prismatic epithelium. Enterocytes of digesting caimans are loaded with lipid droplets (Starck, unpubl. material).
SQUAMATES
A whole suite of physiological, biochemical, and morphological parameters has been described for the postprandial response of snakes (Specific Dynamic Action = SDA; Benedict, 1932; Brody, 1945; Kleiber, 1975; Secor et al., 1994,2003; Secor and Diamond, 1995,1997,1998; Dorcas et al., 1997; Thompson and Withers, 1999;Jackson and Perry, 2000; Bedford and Christian, 2001; Starck and Beese, 2001,2002; Overgaard et al., 2002; Wang et al., 2003; Toledo et al., 2003; Zaidan and Beaupre, 2003). Size changes have been described for almost all organs (Secor et al., 1994;Starck and Beese, 2001,2002). However, size and cellular configuration of the absorptive surface of the intestine, i.e. the mucosal epithelium, is one of the key factors determining the uptake capacity of the gut. When a snake is fasting, the mucosal epithelium of the small intestine is a pseudostratified epitheliumwith the enterocytes densely packed in several layers. All enterocytes root on the (folded) basal membrane, however, thus forming a pseudostratified epithelium (Figs 7.6, 7.7). The microvilli of the enterocytes are short. Within 24 to 48
186
Physiological and ecological adaptations t o feeding in vertebrates
Time in experiment (days) Fig. 7.6. Up- and down-regulation of the thickness of the small intestinal mucosa of Burmese python (Python niolurus). Arrow heads indicate feeding events (from Starck and Beese, 2001).
hours after feeding, the absorptive surface of the small intestine increases 2 to 3 fold and the epithelial configuration changes from pseudostratified to single layered (Fig. 7.7). It was suggested that this configuration change of the epithelium is driven by a hydraulic pressure pump, i.e. blood and lymph being pumped into the villi, and by loading enterocytes with lipid droplets (Starck and Beese, 2001,2002; Starck, 2003). Even the enterocytes possess large amounts of spare membrane and can be inflated without biosynthesis of a new membrane. The cytological mechanism of organ size increase is surprisingly simple and no major energetic costs have been recognized to be associated with the increase of the mucosal surface because no new cells are produced (Starck and Beeses, 2001,2002;Overgaard et al., 2002; Starck, 2003). Secor (2003) estimated that only 5% of the overall energy expenditure that goes into SDA is associated with increase of the mucosal surface. The source of fuel for SDA has been a matter of discussion. Because of the necessary start-up costs of digestion, Secor and Diamond (1995,1998)suggested that snakes operate predominantly on the pay-before-pumping principle, i.e. snakes mobilize fuel from their body reserves to activate their gastrointestinal tract before absorbing prey (Secor and Diamond, 1995,1998; Secor and Nagy, 2000; Secor, 2001,2003). Given this principle, digestion of a meal becomes tremendously expensive (up to 30% of the prey) and snakes would need considerable energy savings to initiate digestion. We recently showed that most SDA is fueled from the prey directly and not from the snake tissue stores (Starck et al., 2004).In that stable isotope labeling study,
Flexibility of digestive system
187
Fig. 7.7. Burmese python: (A) Anatomical changes after feeding (upper panel) and during fasting (lower panel). B and C: histological micrographs of the mucosal epithelium in digesting conditions. D, E, and F: electronmicrographs of the epithelium (D, E), enterocytes loaded with lipid droplets, and (F) brush border of an enterocyte (from Starck and Beese, 2001).
fasting snakes fueled their maintenance metabolism from energy depots. Within 24 hours after feeding and certainly after initiating digestion they switched to fueling metabolism by catabolizing their prey. The isotopic signal from prey was detectable over a period of 2 weeks and was only slowly mixed with the snake's own tissue. Of course, there must be some start-up costs to initiate digestion but such costs may be lower than previously assumed and fully covered by fuel available from cytoplasmatic sources or blood plasma.
188
Physiological and ecological adaptations to feeding in vertebrates
Fig. 7.8. Burmese python in fasting condition. A and B are microphotographs of the transitional epithelium; C: electronmicrograph of the transitional epithelium showing that all enterocytes root on the basal membrane. D and E details of enterocytes showing spare membrane of "empty" enterocytes (from Starck and Beese, 2001).
The trigger that regulates the postprandial responses in snakes appears to be a combination of luminal nutrient signals from amino acids and peptides. Neither mechanical stimulation, e.g. extension of the stomach, nor luminal presence of glucose, lipids of pancreatic-bilary secretions, stimulated a postprandial response (Secoret al., 2002). However, increasingplasma concentrationsof gastrointestinal regulatory peptides (e.g. cholecystokinin, glucose-dependent insulin-trophic peptide, glucagons and neurotensin) immediately after feeding suggest that these peptides may play a role in regulating the python's digestive response (Secor et al., 2001).While feeding elicits a release of regulatory peptides from tissue to blood plasma, the intramural nervous system of the gut does not undergo detectable changes (see Holmgren and Holmberg, chapter 13 this volume).
Flexibility of digestive system
189
In summary, snakes show complex and manifold responses to feeding and fasting. Size changes of the mucosal surface are based on alternations of the cellular configurationof the mucosal epithelium. The supposed mechanism is completely different to that described for birds. The transitional epithelium allows for fast and energetically cheap increase of organ size and the prey is the source of fuel. The transitional epithelium has not only been described for the sit-and-wait foraging Burmese python (Python molurus; Starck and Beese, 2001; see Fig. 7.8), but also the ball python (Python regius; Starck and Wang, unpubl. data), garter snakes (Thamnophis sirtalis; Starck and Beese, 2002), rhombic eggeaters (Dasypeltis scabra, Starck unpubl. data), and rattlesnakes (Crotalusdurissus; Brito, unpubl. data; Crotalus viridis; Parker, unpubl. data). Thus it occurs as a feature of a variety of unrelated snake species with different feeding ecology. At the current state of knowledge, it is recognized as a character shared among snakes and does not exclusively relate to sit-and-wait foraging feeding strategy. The liver undergoes considerable size changes in snakes after feeding (Secor et al. 1994;Starck and Beese, 2002). The up-regulation of organ size is somewhat slower than in the intestine and mostly based on a temporary incorporation of lipid droplets into hepatocytes. Decrease in liver size is associated with decreasing size and number of lipid droplets in the hepatocytes (Starck and Beese, 2002).We detected no significant number of proliferating hepatocytes in digesting snakes.Thus, loading of cells with lipid droplets and possibly increased blood perfusion of the liver are the main contributions to liver size increment. From a physiological point of view, one would expect considerable changes in functional demands downstream organs such as the kidney depending on feeding or fasting. While nitrogen excretion does not present a problem during fasting, it increases tremendously during digestion. Secor and Diamond (1998) and Sector et al. (1994)reported a significant increase in kidney size but the functional and structural properties of the kidneys themselves remain to be studied. FROGS
Early observations on gastrointestinal plasticity in tadpoles date back to Babiik (1903a,b; 1906).By feeding tadpoles a purely vegetarian diet he elicited a gastrointestinalmorphology that differed from tadpoles raised solely on a meat diet. Later, anuran plasticity to adjust to fluctuating environmental conditionsbecame a model system for studying phenotypic plasticity as an evolutionary stable strategy (Newman, 1988a,b;Pfennig, 1992).As in sitand-wait foraging snakes, feeding and fasting imposes a major functional challenge to the gastrointestinal system of many anurans. For example, the horned frog (Ceratophrys cranwallii) is a typical sit-and-wait foraging frog. While fasting and waiting for prey, the horned frog lives half buried with
190
Physiological and ecological adaptations t o feeding i n vertebrates
only the snout peaking out of the soft ground. Prey is rare and unpredictable. We compared changes of organ size and tissue configuration in the horned frog with the African clawed frog (Xenopuslaevis), which is a typical suspension feeder that eats more or less continuously on small particles (Starck et al., unpubl. data). We exposed 12 individuals of both species to fasting periods of four weeks. After that fasting period half of the animals in each group were exposed to intensive feeding (up to 50% body mass in Ceratophrys).In both species, we found exactly the same changes of tissue configuration as described for snakes and crocodiles. In the fasting frogs, independent of their feeding ecology, the mucosal epithelium of the small intestine was a pseudostratified epithelium. Two days after feeding, the epithelium was single layered, and the cells were filled with lipid droplets. The structural changes were associated with equivalent size changes of the mucosal surface magnification.Although these are the only two species studied so far, the similar pattern of different frog species with different feeding ecology suggests that the pseudostratification of the epithelium and its ability to change cellular configuration are not related to the feeding ecology of frogs. A phylogenetically most parsimonious explanation for the similarity with snakes and crocodiles suggests that a transitional epithelium and the ability to tolerate long fasting periods is a shared character of ectothermic tetrapods. Thus the responses initially described for sit-and-wait foraging snakes are not a special adaptation of large constrictor snakes, but a feature that had already been evolved in the stem group of vertebrates.
Many mammals experience considerable fluctuations in food supply and food composition. Female polar bears tolerate a six-month-long winter fast during their gestation period, simultaneously enduring extreme cold and the elevated metabolic demands imposed by fetal growth. Male elephant seals do not feed during the intense mating and fighting activities during their reproductive season.Male caribou fast for several weeks of the rutting season, losing a considerable portion of their body mass. While foraging through their hundreds square kilometer territory, wolves tolerate fasts of several weeks during the arctic winter. All these mammals tolerate extreme climatic conditions while having to sustain additional energetic stress of temporary food deprivation, extreme physical activity, and reproduction. Unfortunately,none of the species is particularly suitable for an experimental study. Most are endangered, difficult to capture and handle in the wild, and potentially dangerous. Thus, almost all our knowledge on phenotypic flexibility of the gastrointestinal system in mammals is based on laboratory studies of rodents, and most studies have looked at diet switching or food intake changes in response to changing physiological demand; none has
Flexibility of digestive system
191
looked at long-term fasting except for hibernating animals (see Chapter 9 by Carey, this volume). Ideas about how mammals change the phenotype of their gastrointestinal system are dominated by our knowledge of the histology and pathology of the mammalian intestine. In the gut of all mammals, cell proliferation occurs in the intestinal crypts. Cell proliferation is continuous and newly formed cells migrate from the crypts over the surface of the intestinal villi to the tip of the villi. During that migration, they become functional and absorb nutrients. When the enterocytes reach the distal tip of a villus they become apoptotic and are shed into the intestinal lumen. Cell proliferation is balanced by an equal rate of cell loss at the tip of the villi (Madara and Trier, 1994; Raab et al., 1998).Cell proliferation follows a circadian rhythm inversely related to the activity of the gut, i.e. cell proliferation is high when the intestinal organs are inactive and low when activity is high (Ruby et al., 1973,1974;Scheving et al., 1974,1978,1983;Scheving, 1981;von Mayersbach, 1983;Sakata, 1987,1991).The turnover time (i.e. 100%replacement of cells in a tissue) has been estimated to range between 3 and 8 days (Lipkin, 1965; Scheving et al., 1974; Williamson and Chir, 1978; Steen, 1993; Madara and Trier, 1994;Dunel-Erb et al., 2001).The permanent migration of cells from the intestinal crypts to the tip of the villi allows for considerable flexibility of the mammalian intestine, but also incurs high permanent costs of maintaining an active intestine. The costs of maintaining the intestine in mammals have not been measured directly but estimates range between 20% and 30% of basal metabolic rate (BMR)and between 28% and 46% of whole body protein synthesis (Stevens and Hume, 1995). The high flexibility of the mammalian intestine allows for compensatory growth after partial resection of segments of the small intestine. Goss (1978) and Williamson and Chir (1978)reviewed experimentscarried out from the 1950s through the 1970s, in which parts of the mammalian intestine were resected, reversed, isolated, anastomosed or independently perfused to study the resultant effects of remaining parts of the intestine. Isolation of gut segments resulted in atrophy of unused segments, as characterized by a decrease of circumference, reduced segment length, and reduced villus height, while remaining segments exhibited enhanced cell proliferation activity and increased circumference, length, and villus height, to compensate for the lost segments. All effects were reversible, i.e. perfusion of an isolated segment with nutrients or mechanical stimulation counteracted atrophy. These data were confirmed by Hammond et al. (1994)who showed that mice possess considerable flexibility to adjust to a combination of peak energy demands. Under a combination of increasing energetic demands through exercise, temperature stress, and lactation, mice increased masses of intestine, liver and kidney but did not reach a limitation to assimilate energy to cope with the increased demands. Dry mass of the small intestine doubled under experimental conditions. Hammond et al. (1996)repeated studies reported by Goss
192
Physiological and ecological adaptations t o feeding in vertebrates
(1978)and showed that mice completely regenerated the small intestinewhen up to 50%was previously resected. Mice failed to survive resection of 70%. Cellular processes underlying the up- and down-regulation of the mammalian small intestine have been intensively investigated.Fasting or starvation of laboratory rats resulted in a decline of intestinal crypt cell mitotic rate and decreasing morphometric parameters of the small intestine (Brown et al., 1963;Steiner et al., 1968;Altrnann, 1972; Aldewachi et al., 1975; Eastwood, 1977). Growth of mucosal epithelium after refeeding was established by increased cell proliferation. Hypotheses about regulatory mechanisms include trophic regulatory peptides such as the epidermal growth factor (EGF) from stem cells in the crypts as well as polyamines contained in the food that trigger increasingrates of cell proliferation(Johnsonand McCormack, 1994). In laboratory rats, the mucosal surface was reestablished within a few days after refeeding (Boza et al., 1999; Dunel-Erb et al., 2001). Also in laboratory rats, which have a one-day enterocyte turnover (Karasov and Diamond, 1987), the villi returned to their normal length within a day after feeding was reinitiated after fasting (Buts et al., 1990;Hodin et al., 1994).Responses may be even more rapid. The first responses of the atrophied gut of fasted rats occurred as early as two hours after the first meal, when genes such as c-fos and c-jun, which represent the mitogenic response in many types of tissues, were first expressed in intestinal crypt cells (Hodin et al., 1994).Clarke (1975) claimed that as soon as two hours after refeeding, the villi of refed rats were taller than those of starved rats, contrarily Hodin et al. (1994)failed to find a significant increase in villus height even four hours after refeeding. Restoration of the mucosal surface was based on a combination of hyperplasia and hypertrophy.In domesticated pigs, rats, and calves, increase and decrease of the area of the mucosal surface were achieved by changing the balance between cell proliferation in the intestinal crypts and apoptosis of enterocytes at the tip of the villi (Raab et al., 1998; Boza et al., 1999; Claus et al., 2001; Mentschel et al., 2001). It should be pointed out that these observations refer to active mammals. Fasting and down-regulation of intestine size under conditions of hibernation have been studied by Carey (1990,1992),Carey and Cooke (1991)and Carey and Martin (1996).Because overall metabolism is down-regulated,the conditions of hibernation are quite different from those of an active mammal (see Chapter 9 by this volume). However, hibernators must up-regulate the small intestine after arousal from hibernation. Hume et al. (2002) studied seasonal changes of the gut morphometry and function in Alpine marmots. They described a precise adjustment of the marmot gut size and function to seasonal changes in food availability and changing internal demands (i.e. hibernation vs activity).Interestingly,arousal from hibernation, emergence from the den, and increased food intake were timely separated events and were reflected in highly significant differences in gut morphology and function. It was several weeks after emergence when food availability and food
Flexibility of digestive system
193
intake peaked during midsummer that the gastrointestinal tract almost doubled in size.
Conclusions The mammalian intestine is flexible and responds to changes in external and internal demands by up- or down-regulation of its morphological and functional parameters. Responsivenessof the intestine is based on a balance of cell proliferation in the intestinal crypts and cell loss at the tip of the villi. Fasting results in a cellular atrophy of the organ. Flexible responses are rapid but presumably energetically expensive.Costs arise because flexibility is based on production of new tissue and because up-regulation of the size of the intestine requires elevated levels of cell proliferation. An important, but open question is to resolve whether up- and down-regulation of the small intestine is based on changes in the rate of cell proliferation,rate of apoptosis, or a combination of both. Theoretically, increased rates of cell proliferation will result in additional energetic costs for up-regulation while a change in rate of apoptosis will incur no additional costs.
"Reptiles"
Anura
Rhynchocephala
Crocoditia
Marnrnalia
1
Fig. 7.9. Cladogram of the tetrapods. Black squares represent the occurrence of a transitional epithelium, and white squares indicate plasticity based on cellular turnover. A phylogenetically most parsimonious interpretation suggests that a transitional epithelium is a shared character of tetrapods among which mammals and birds have replaced it by cellular turnover. The phylogenetic distance of mammals and birds suggests an independent evolutionary origin of the turnover mechanism.
194
Physiological and ecological adaptations t o feeding in vertebrates
COMPARATIVE PERSPECTIVE Current data can be viewed in a comparative perspective to gain some insights into the evolutionary history of patterns and processes supporting phenotypic flexibility. A key question is whether the structural basis for upand down-regulating organ size evolved once or whether different clades of vertebrates invented plasticity independently. Several mechanisms have been described above that allow for rapid and reversible changes of organ size (and function); i.e. changes in cell number vs altered cell size. If the basic mechanisms of organ size change are plotted on a phylogram of the tetrapods (Fig. 7.9), a transitional epithelium is found in all ectothermic tetrapods but not in endothermic birds and mammals. Assuming that the most parsimonious solution is correct, the transitional epithelium must be considered a shared character for tetrapods which has been lost independently in mammals and birds. Mammals and birds change the size of the mucosal epithelium by means of cell proliferation and apoptosis or cell extrusion respectively. Because birds and mammals are not closely related we must consequently assume that the continuous turnover of mucosa epithelial cells evolved in parallel and independently in birds and mammals. The observation that mammals balance cell proliferation by apoptosis while birds extrude cells at the tip of the villi may support the view of independent evolution of mechanisms of organ turn-over in these two groups. However, this view needs experimental corroboration.
CONCLUSIONS AND OUTLOOK For future studies of phenotypic flexibility, several questions and perspectives arise from this review.
(1) We should integrate studies that include data frorr~biochemistry, physiology, histology, and morphology. We need to obtain data sets with correlated data from different "classical" fields to ascertain the concert of responses to feeding in digestive systems. (2) We miss an explicit evolutionary model to make testable predictions about when a species tends to be plastic and when it tends to be stable. Presumably, such a model must differ for ectothermic and endothermic tetrapods.
(3) Field studies in an explicit evolutionary (functional)context are needed to link detailed laboratory studies with the ecologically relevant field projects.
(4) Flexible changes of a phenotype may be based on completely different mechanisms, i.e. linked to environmental signals or endogenous clocks.
Flexibility of digestive system
195
Both systems may have even evolved independently. It is important to tease apart effects of season and effects due to direct responses to environment.
(5) What are the costs of flexibility? Basically nothing is known about the energetic costs of flexibility although costs of stable vs flexible phenotypes have been considered the driving force of phenotype evolution. (6) Hypotheses about costs of maintaining organs are central in our pre-
sumptive understanding of adaptive phenotypic flexibility. However, these hypotheses have never been tested. Although hampered with methodological difficulties we recognize an urgent need for experimental measurement of the energetic costs of organ maintenance during different phenotypic conditions. (7) The source of fuel for flexible responses is not known. However, looking at the source of fuel would be a central question that might also help in understanding possible constraints in plasticity.
(8) Most studies have focused on flexible responses of the gut to feeding, diet switching, and fasting. Downstream organs such as the kidneys and liver or patterns of blood flow have largely been neglected.
Al-Jaborae F.F. 1979. The influence of diet on the gut morphology of the starling (Sturntis vulgaris L. 1758). PhD diss. University Oxford, UK. Aldewachi H.S., Wright N.A., Appleton D.R., and Watson A.J. 1975. The effect of starvation and refeeding on cell population kinetics in the rat small bowel mucosa. J. Anat. 119: 105-121. Altmann G.G. 1972. Influence of starvation and refeeding on mucosal size and epithelian renewal in the rat small intestine. Arner. J. Anat. 133: 391400. Ankney C.D. 1977. Feeding and digestive organ size in breeding lesser snow geese. Auk 94: 275-282 Babak E. 1903a. Uber den EinfluB der Nahrung auf die Lange des Darmkanals, I. Biol. Zb. 13: 477-483. Babak E. 1903b. Uber den Einflug der Nahrung auf die Lange des Darmkanals, 11. Biol. Zb. 14/15: 519-521. Babak E. 1906. Experimentelle Untersuchungen iiber die Variabilitat der Verdauungsrohre. Arch. Entzoicklungsrnechanik d . Org. 21: 611-702. Battley P.F. and Piersma T. 2004. Adaptive interplay between feeding ecology and features of the digestive tract in birds. (This volume). Bauchinger U. 2002. Phenotypic flexibility of organs during long-distance migration in garden warblers (Sylvia borin): implications for migratory and reproductive periods. PhD thesis, TU, Munich. Bauchinger U. and Biebach H. 2001. Differential catabolism of muscle protein in garden warblers (Sylvia borin): flight and leg muscles act as protein source during Ion-gdistance migration. J. Cornp. Physiol. B 171:293-301. Bedford G.S. and Christian K.A. 2001. Metabolic response to feeding and fasting in the water python (Liasis fiscus). Austr. J. Zool. 49: 379-387. Benedikt F. G. 1932. The Physiology of Large Reptiles. Carnegie Inst., Washington Publ. 425, Washington DC.
196
Physiological and ecological adaptations t o feeding in vertebrates
Biebach H. And Bauchinger U. 2002. Energetic savings by organ adjustment during long migratory flights in garden warblers (Sylvia borin). In: Avian Migration P. Berthold and E. Gwinner (eds.). Springer Verlag, Heidelberg, pp. 269-280. Boza J.J., Moennoz D., Vuichoud J., Jarret A.R., et al. 1999. Food deprivation and refeeding influences growth, nutrient retention and functional recovery of rats. J. Nutr. 129: 1340-1346. Brandes G. 1896. iJber den vermeintlichen Einflui3 veranderter Ernahrung auf die Struktur des Vogelmagens. Biol. Zb. 16: 825-838. Breitenbach R.P., Nagra C.L., and Meyer R.K. 1963. Effect of limited food intake on cyclic annual changes in ring-necked pheasant hens. J. Wildl. Manag. 27: 24-36. Brody S. 1945. Bioenergetics and Growth. Reinhold Publ. Co., New York, NY. Brown H.O., Levine M.L., and Lipkin M. 1963. Inhibition of intestinal epithelial cell renewal and migration induced by starvation. Amer. J. Physiol. 205: 868-872. Buts J.P., Vivjerman V., Barudi C., De Keyser N., Maldague P., and Dive C. 1990. Refeeding after starvation in the rat: comparative effects of lipids, proteins and carbohydrates on jejunal and ileal mucosal adaptation. Eur. J. Clin. Invest. 20: 441-452. Carey H. V. 1990. Seasonal changes in mucosal structure and function in ground squirrel intestine. Am. J. Physiol. 259: R385-R392. Carey H. V. 1992. Effects of fasting and hibernation on ion secretion in ground squirrel intestine. A m . 1. Physiol. 263: R1202-R1208 Carey H. V. and Cooke 1991. Effect of hibernation and jejunal bypass on mucosal structure and function. A m . J. Physiol. 261: G37-G44. Carey H. V. and Martin S. L. 1996. Preservation of intestinal gene expression during hibernation. Am. 1. Physiol. 271: G805-G813. Clarke R.M. 1975. The time course of changes in mucosal architecture and epithelial cell production and cell shedding in the small intestine of the rat fed after fasting. J. Anatomy 120: 321-327. Claus R., Mentschel J., Blazey B., and Munz 0. 2001. Preliminary studies on epidermal growth factor (EGF) immunoreactivity in goblet cell of the small intestine by a speciesspecific antiserum in healthy piglets and piglets with diarrhoea. J. Anim. Feed Sci. 10: 309-315. Deerenberg C., Biebach H., and Bauchinger U. 2002. Spleen size variation during long distance migration in the garden Warbler (Sylvia borin). Avian Sci. 2: 217-226. Dekinga A., Dietz M.W., Koolhaas A. and Piersma T. 2001. Time course and reversibility of changes in the gizzards of red knots alternately eating and soft food. J. Exp. Biol. 204: 2167-2173. Dieslich H. 1939. Situs viscerum und groi3e GefaBe von Alligator mississippensis Daud. PhH Buschmann, Miinster. Dorcas M.E., Peterson C.R., and Flint M.E.T. 1997. The thermal biology of digestion in rubber boas (Charina bottae): physiology, behavior, and environmental constraints. Physiol. Zool. 70: 292-300. Drobney R.D. 1984. Effect of diet on visceral morphology of breeding wood duck. Auk 101: 93-98. Dubowy P.J. 1985. Seasonal organ dynamics in post-breeding male blue-winged teal and Northern shovelers. Comp. Biochem. Physiol. 82A: 899-906. Dunel-Erb S., Chevalier C., Laurent P., Bach A., Decrock F., and Le Maho Y. 2001. Restoration of the jejunal mucosa in rats refed after prolonged fasting. Comp. Biochem. Physiol. 129 A: 933-947. Eastwood G.L. 1977. Gastrointestinal epithelial renewal. Gastroenterology 72: 962-975. Fermi L. and Repetto R. 1901. ~ b e die r Einwirkung der Nahrungsweise auf die Entwicklung des Verdauungsapparates. Arch. Anat. Physiol. Abthlg. Suppl. pp. 84-97. Goss R. J. 1978. The Physiology of Growth. Academic Press, pp. 267-283. Greene H. W. 1997. Snakes - The evolution of mystery in Nature. Univ. Berkeley Press, Berkeley.
Flexibility of digestive system
197
Groscolas R. 1982. Changes in plasma lipids during breeding, moulting and starvation in male and female emperor penguins (Aptenodytes forsteri). Physiol. Zool. 55: 45-55. Groscolas R. 1986. Changes in body mass, body temperature and plasma fuel during the natural breeding fast in male and female emperor penguins, Aptenodytes forsteri. J. Comp. Physiol. B 156:521-527. Hammond K., Konarzewski M., Torres R. M., and Diamond J. 1994. Metabolic ceilings under a combination of peak evergy demands. Physiol. Zool. 67: 1479-1506. Hammond K., Lam M., Lloyd K., C. K. and Diamond J. 1996. Simultaneous manipulation of intestinal capacities a n d nutrient loads in mice. A m . J. Physiol. 271: G969-G979. Heitmeyer M.E. 1988. Changes in visceral morphology of wintering female mallards (Anas platyrhynchos). Can J. Zool. 66: 2015-2018. Hobaugh W.C. 1985. Body condition and nutrition of snow geese wintering in Southeastern Texas. J. Wildl. Managmt. 49: 1028-1037. Hodin R.A., Graham J.R., Meng S., and Upton M.P. 1994. Temporal pattern of rat small intestinal gene expression with. Amer. J. Physiol. 266: G89. HolmgrCn F. 1872. Om kottatande dufvor. Aftryck ur Uppsala Lakare.forensings Forhandlingar, Upsala 1872 (cited in Brandes, 1896). Hume I. D. and Biebach H. 1996. Digestive tract function in the long-distance migratory garden warbler, Sylvia borin. J. Comp. Physiol. B 166: 388-395. Hume I.D., Beiglbock C., Ruf T., Frey-Roos F., Bruns U., and Arnold W. 2002. Seasonal changes in morphology and function of the gastrointestinal tract of free-living alpine marmots (Marmota marmots). J. Comp. Physiol. B 172, 197-207. Hunter J. 1839. Ouevres complets. Traduit par fichelot, Paris 1839 Tome 1, p. 184 (cited in Brandes, 1896). Jackson K. and Perry G. 2000. Changes in intestinal morphology following feeding in the brown treesnake, Boiga irregularis. J. Herpetol. 34: 459462. Johnson and Mccormack S.A. 1994. Regulation of gastrointestinal mucosal growth. In: Physiology of the Gastrointestinal Tract. Johnson L.R. (ed.), 3rd edition, Raven Press, New York, pp. 611-641. Karasov W.H. and Diamond J.M. 1987. Adaptation of intestinal nutrient transport. In: Physiology of the Gastrointestinal Tract. L. R. Johnson, (ed.). Raven Press, New York, NY, pp. 1489-1497. Karasov W.H., Pinshow B., Starck J.M., and Afik D. 2004. Anatomical and histological changes in the alimentary tract of migrating blackcaps (Sylvia atricapilla): a comparison among fed, fasted, food-restricted and refed birds (Sylvia atricapilla). Physiol. Biochem. Zool. 77: 149-160. Kleiber M. 1975. The Fire of Life. Krieger, New York, NY. Le Maho Y., Delclitte P,, and Chatonnet J. 1976. Thermoregulation in fasting emperor penguins under natural conditions. Amer. J. Physiol. 231: 913-922. Le Maho Y., Delclitte P., and Groscolas R. 1977. Body temperature regulation of the emperor penguin (Aptenodytes forsteri) during physiological fasting. In: Adaptations within the Antarctic ecosystems G.A. Llano, (ed.). Smithsonian Institution, Washington DC, pp. 501-509. Lipkin M. 1965 Cell replication in the gastrointestinal tract of man. Gastroenterology 48: 616-624. Luppa H. 1977. Histology of the digestive tract. In: Biology of the Reptilia C. Cans and T.S. Parsons, (eds.). Acad. Press, New York, NY, pp. 225-313. Madara J.L. and Trier J.S. 1994. The functional morphology of the mucosa of the small intestine. In: Physiology of the Gastrointestinal Tract. L.R. Johnson (ed.). Raven Press, New York. NY, pp. 1577-1622. Mayersbach H. 1983. An overview of the chronobiology of cellular morphology. In: Biological Rhythms and Medicine: Cellular, Metabolic and Pharmacological Aspects. A. Reinberg A. and Smolensky M.H. (eds.). Springer-Verlag, Berlin, pp. 47-78.
198
Physiological and ecological adaptations t o feeding in vertebrates
Mentschel J., Leiser R., Miilling C., Pfarrer C., and Claus R. 2001. Butyric acid stimulates rumen mucosa development in the calf mainly by a reduction of apoptosis. Arch. Anim. Nutr. 55: 85-102. Moorman T.E., Baldassarre G.A., and Richard D.M. 1992. Carcass mass, composition and gut morphology dynamics of mottled ducks in fall and winter in Louisiana. Condor 94: 407-417. Mortensen A., Nordoy E.S., and Blix A.S. 1985. Seasonal changes in body composition of the Norwegian Rock ptarmigan Lagopus mutus. Ornis Scand. 16: 25-28. Newman R.A. 1988a. Genetic variation for larval anuran (Scaphiopus couchii) development time in an uncertain environment. Evolution 42: 763-773. Newman R.A. 198813. Adaptive plasticity in development of Scaphiopus couchii tadpoles in desert ponds. Evolution 42: 774-783. Oftedal O.T., Alt G.L., Widdowson E.M., and Kakubasz M.R. 1993. Nutrition and growth of suckling bears during their mother's winter fast. British J. Nutr. 70: 59-79. Overgaard J., Andersen J.B. and Wang T. 2002. The effects of fasting duration on the metabolic response to feeding in Python molttrus: an evaluation of the energetic costs associated with gastrointestinal growth and up-regulation. Physiol. Biochem. Zool. 75: 360-368. Owen R. 1861. Essays and Observations on Natural History, London, UK. Pfennig D.W. 1992. Polyphenism in spadefoot toad tadpoles as a locally adjusted evolutionary stable strategy. Evolution 46: 1408-1420. Piersma T. and Lindstrom A. 1997. Rapid, reversible changes in organ size as acomponent in adaptive behaviour. Trrnds Ecol. Evol. 12: 134-138. Piersma T. and Drent J. 2003. Phentypic fledxibility and the evolution of organismal design. Trends Ecol. Evol. 18: 228-233. Pigliucci M. 2001. Phenotypic Plasticity. Beyond Nature and Nurture. Johns Hopkins Univ. Press, Baltimore, MD. Pope C.H. 1961. The Giant Snakes. A.A. Knopf, New York. Raab S., Leiser R., Kemmer H., and Claus R. 1998. Effects of energy and purines in the diet on proliferation, differentiation, and apoptosis in the small intestine of the pig. Metabolism 47: 1105-1111. Reese A.M. 1915. The Alligator and its Allies. G.P. Putnamfs Sons, The Knickerbocker Press New York, NY. Ruby J.R., Scheving L.E., Gray, S.B., and White K. 1973. Circadian rhythm of nuclear DNA in adult rat liver. Exp. Cell. Res. 76: 136-142. Ruby J.R., Scheving L.E., Gray S.B., and White K. 1974. Demonstration of a circadian rhythm of nuclear DNA in liver. In: Chronobiology. L.E., Scheving F. Halberg, and J.E. Pauly (eds.). Georg Thieme Publ., Stuttgart, pp. 33-37. Saint Giron H. 1976. Comparative histology of endocrine glands, nasal cavities and digestive tract of anguiomorph lizards. In: Morphology and Biology of Reptiles A. D. Bellaris and C.B. Cox,(eds.). Linnean Soc. Symp. Series 3. London, UK, pp. 205-216. Sakata T. 1987. Stimulatory effect of short chain fatty acid on epithelial cell proliferation in the rat intestine: a possible explanation for trophic effects of fermentable fibre, gut microbes , and lumenal trophic factors. Brit. J. Nutr. 58: 95-103. Sakata T. 1991. Stimulatory effects of SCFA on epithelial cell proliferation in the intestine. In: Hindgut '91. T. Sakata and R.L. Snipes (eds.). Senshu Univ. Press, Tokyo, pp. 71-90. Savory C.J. and Gentle M.J. 1976a. Effects of dietary dilution with fibre on the food intake and gut dimensions of Japanese quail. Brit. Poultry Sci. 17: 561-570. Savory C.J. and Gentle M.J. 197613. Changes in food intake and gut size of Japanese quail in response to manipulation of dietary fiber content. Br. Poultry Sci. 17: 571-580. Scheving L.E. 1981. Circadian rhythms in cell proliferation: Their importance when investigating the basic mechanism or normal versus abnormal growth. In: Biological Rhythms in Structure and Function. Von Mayersbach, H . L.E. Scheving, and J.E. Pauly (eds.). Alan R. Liss. Inc., New York, NY, pp. 39-79.
Flexibility of digestive system
199
Scheving L.E., von Mayersbach H., and Pauly J.E. 1974. An overview of chronopharmacology (a general review). J. Eur. Toxicol. 7: 203-227. Scheving L.E., Burns E.R., Pauly J.E., and Tsai T.H. 1978. Circadian variation in cell division of the mouse alimentary tract, bone marrow, and corneal epithelium. Anat. Rec. 191: 479-485. Scheving L.E., Pauly J.E., Tsai T.H. and Scheving L.A. 1983. Chronobiology of cell proliferation. Implications for cancer chemotherapy. In: Biological Rhythms and Medicine: Cellular, Metabolic and Pharmacological Aspects. A. Reinberg and Smolensky, M.H. (eds.). Springer-Verlag, Berlin, pp. 79-130. Schlichting C.D. and Pigliucci M. 1998. Phenotype Evolution: A Reaction Norm Perspective. Sinauer, Sunderland, MA. Secor S.M. 2001. The source and sense of the metabolic response to feeding. A m . Zool. 41: 1583-1584. Secor S.M. 2003. Gastric function and its contribution to the postprandial metabolic response of the Burmese python Python rnolurus. J. Exp. Biol. 206: 1621-1630. Secor S.M. and Diamond J. 1995. Adaptive responses to feeding in Burmese Pythons: pay before pumping. J. Exp. Biol. 198: 1313-1325. Secor S.M. and Diamond J. 1997. Determinants of postfeeding metabolic response in Burmese python, Python rnolurus. Physiol. Zool. 70: 202-212. Secor S.M. and Diamond J. 1998. A vertebrate model of extreme physiological regulation. Nature 395: 659-662. Secor S.M. and Nagy T. 2000. Postprandial response of plasma lipids and the hormone leptin in pythons. Amer. Zool. 203: 1205. Secor S.M., Stein E.D., and Diamond J. 1994. Rapid up-regulation of snake intestine in response to feeding: a new model of intestinal adaptation. Amer. J. Physiol. 266: G695-G705. Secor S.M., Fehsenfeld D., Diamond J., and Adrian T.E. 2001. Responses of python gastrointestinal regulatory peptides to feeding. Proc. Natl. Acad. Sci. USA, 98: 13637-13642. Secor S.M., Lane J.S., Whang E.E., Ashley A.W., and Diamond J. 2002. Lumenal nutrient signals for intestinal adaptation in pythons. Anzer. J. Physiol. 283: 1298-1309. Shamoto K. and Yamauchi K. 2000. Recovery responses of chick intestinal villus morphology to different refeeding procedures. Poultry Sci. 79: 718-723. Spitzer G. 1972. Jahreszeitliche Aspekte der Biologie der Bartmeise (Panurus biarrnicus). J. Orn. 113: 241-275. Starck J.M. 1996. Phenotypic plasticity, cellular dynamics, and epithelial turnover of the intestine of Japanese quail (Coturnix coturnix japonica). J. Zool. Lond. 238: 53-79. Starck J.M. 1999a. Structural flexibility of the gastro-intestinal tract of vertebrates. Implications for evolutionary morphology. Zool. A n z . 238: 87-101. Starck J.M. 199913. Phenotypic flexibility of the avian gizzard: Rapid, reversible and repeated changes of organ size in response to changes in dietary fibre content. J. Exp. Biol. 202: 3171-3179. Starck J.M. (2003) Shaping up: how vertebrates adjust their digestive system to changing environmental conditions. Animal Biol. 53: 245-257. Starck J.M. and Beese K. 2001: Structural flexibility of the intestine of Burmese python in response to feeding. J. Exp. Biol. 204: 325-335. Starck J.M. and Beese K. 2002. Structural flexibility of the small intestine and liver of garter snakes in response to feeding and fasting. J. Exp. Biol. 205: 1377-1388. Starck J.M. and Kloss E. 1995. Structural responses of Japanese quail intestine to different diets. Dtsch. Tierarztl. Wochenschr. 102: 146-149. Starck J.M., and Rahmaan G.H. 2003. Phenotypic flexibility of structure and function of the digestive system of Japanese quail. J. Exp. Biol. 206: 1887-1897. Starck, J.M., Abe A.S., and Neto A. 2002. Morphological and physiological responses to feeding in caiman. Integrative and Comparative Biology 42: 1317-1318.
200
Physiological and ecological adaptations t o feeding i n vertebrates
Starck J.M., Moser P., Werner W., and Linke G. 2004. Pythons metabolize prey to fuel the response to feeding. Proc. Roy. Soc. London, B. 271: 903-908. Steen R. 1993. A Conspiracy of Cells: The Basic Science of Cancer. Plenum Press, New York, NY. Steiner M., Bourges H.R., Freedman L.S., and Gray S.J. 1968. Effect of starvation on the tissue composition of the small intestine in the rat. Amer. J. Physiol. 213: 75-77. Stevens C. E. and Hume I.D. 1995. Comparative Physiology of the Vertebrate Digestive System. 2nded. Cambridge University Press, Cambridge, UK. Tagouchi H. 1920. Beitrage zur Kenntnis iiber die feinere Struktur der Eingeweideorgane der Krokodile. Mitt. Med. Fak. Kaiserl. Univ. Tokyo, 25: 119-188. Thompson G.G. and Withers P.C. 1999. Effect of sloughing and digestion on metabolic rate in the Australian carpet python, Morelia spilota impricata. Austral. J. Zool. 47: 605-610. Toledo L.F., Abe A.S., and Andrade D.V. 2003. Temperature and meal size effects on the postprandial metabolisms and energetics in a boid snake. Physiol. Biochem. Zool. 76: 240-246. Via S., Gomulkiewicz R., de Jong G., Scheiner S.M., Schlichting C.D., and Tienderen P.H.V. 1995. Adaptive phenotypic plasticity: consensus and controversy. Trends Ecol. Evol. 10: 212-216. Wang T., Zaar M., Arvedsen S., Vedel-Smith C. and Overgaard J. 2003. Effects of temperature on the metabolic response to feeding in Python molurus. Comp. Biochem. Physiol. A . 133: 519-527. West-Eberhard M.J. 2003. Developmental Plasticity and Evolution. Oxford University Press. New York, NY. Williamson C.N. and Chir M.B. 1978. Intestinal adaptation I. New Engl. J. Med. 298: 1393-1402. Yamauchi K., Kamisoyama H., and Isshiki Y. 1996. Effects of fasting and refeeding on structures of the intestinal villi and epithelial cells in White Leghorn hens. Brit. Poultry Sci. 37: 909-921. Zaidan F. 111, and Beaupre S.J. 2003. Effects of body mass, meal size, fast length, and temperature on specific dynamic action in the timber rattlesnake (Crotalus horridus). Physiol. Biochem. Zool. 76: 447-458.
Adaptive Interplay Between Feeding Ecology and Features of the DigestiveTract in Birds Phil F. Battley3 and Theunis Piersrna'~~ 'Royal Netherlands Institute for Sea Research (NIOZ), Dept. Marine Ecology and Evolution, Den Burg, Texel, The Netherlands *Universityof Groningen, Dept. Animal Ecology, Centre for Ecological and Evolutionary Studies, Haren, The Netherlands 'university of Otayo, Department of Mathematics and Statistics, Dunedin, New Zealand
SYNOPSIS Different foods require different digestive processing. This simple fact has led to a wide adaptive variety in digestive systems. This chapter considers the relationship between the size of three digestive organs (stomach, intestine, and liver) and food intake, diet type (e.g. degree of softness/hardness in the case of carnivores and ,fiber-content in the case of herbivores), and seasonal factors. Our arguments are based on comparisons of diet-organ size associations within ecologically and phylogenetically uniform groups and on analyses of organ size changes within species and individuals. The phenotypic flexibility of the digestive organs is considerable; changes of 10-1 00% in organ size in response to changing conditions are found in a wide range of bird groups. Experimental work has shown changes in gizzard size of 50-100°/0 in just 5-7 days. Ease of breakdown of food (a function of the strength of protective hard parts or the amount of fibrous ingredients) and total food intake determine the size of the stomach (and more specifically, its muscular part, the gizzard). Food intake also influences the size of the small intestine and the liver. Among waders (Suborder Charadrii), taxa consuming and crushing hard-shelled mollusks also have a heavier intestine (but not liver) than taxa eating softer foods. Much work remains to be done investigating the fine-tuning of capacities among the different digestive organs, the precise trade-offs involved in organ size changes, and the genetic basis of organ size and flexibility.
202
Physiological and ecological adaptations t o feeding in vertebrates
While the structural characteristics of birds and mammals are essentially fixed in size once maturity has been reached, internal organs can vary substantially in size and mass throughout the annual cycle (Jehl, 1997; Piersma and Drent, 2003; Piersma and Lindstrom, 1997). Many species experience changes in habitat, diet, energy demand or climate during the year, all of which may influence digestive morphology.Digestive organs can be large and metabolically highly active and therefore energeticallyexpensive. Thus there will be energetic and mass-related costs associated with having a large digestive system.A large gut may provide the ability to use foods of low quality that are difficult to handle mechanically or chemically. The size of the digestive tract should therefore reflect the balance between the costs of maintaining the system, and the benefits of having organs of that size. Because the internal and external demands for individuals or populations may vary over time, internal flexibility rather than constancy may be the norm in many bird species. Here the following questions are asked: (1)How widespread are digestive organ changes in different bird groups? (2) Are these (endogenously controlled) anticipatory changes related to upcoming needs, or responses to variation in external conditions? (3)What are the roles of diet in determining the size of the digestive organs? We base our analyses on two main approaches, comparisons among higher taxa, and comparisons within species and individuals. We focus on the role of ecological factors, especially diet, in determining organ size. Our analyses involve three main sources of data: (1) changes in organ size in various taxa (especiallygalliformes and waterfowl)published over the past 30 years; (2)body composition and organ masses of waders (or shorebirds, Order Charadriiformes) dissected at the Royal Netherlands Institute for Sea Research (T.Piersma and P.F. Battley, unpubl. data);and (3)recent noninvasive studies that illustrate the reversible morphologicalflexibility of the digestive tract within individuals. The digestive system of birds can be divided into several main components (Table 8.1). In any one taxon, each of these digestive organs will have morphological characteristics that functionally reflect the evolutionary history and contemporary ecology of the taxon (Stevens and Hume, 1995). We do not deal here with these morphological features per se, but instead take them as given and examine overall size patterns, under the assumption that size (or mass) indicates functional capacity. The relationship between size and function is complex and must also involve adjustments at lower levels of organ structure (tissue structures, cell metabolism; see Chapter 7 by Starck, this volume). Increasing costs associated with organ size will cause the net benefit of increasing organ size to be a decreasing benefit function rather than a linear function. Nevertheless, at the level of organizationstudied
Adaptive interplay
203
Table 8.1 Subdivision of the major components of the digestive system in birds (from Klasing, 1998) Organ
Oesophagus
Alternative name /components
+ crop
Proventriculus Gizzard Small intestine Ceca Rectum Cloaca Liver Pancreas
Primary functions
Food storage, movement of food toward proventriculus Glandular stomach Gastric secretion Muscular stomach Crushing or grinding food, mixing food with gastric secretions Duodenum, jejunum, Enzymatic digestion, absorption of digestive ileum end products Microbial fermentation, water and nitrogen absorption, immunosurveillance Colon, large intestine Electrolyte, water and nutrient absorption Storage and excretion of urine and faeces Metabolism of absorbed nutrients, production of bile acids and bile salts Secretion of digestive enzymes
here (entire organ mass) we assume that organ size roughly equates to organ function. The alimentary tract (primarily the gizzard, small intestine or gut to the rectum) and the liver to a lesser extent have been widely studied irl birds and other animals. Accordingly, our review is restricted to those three organs. Lack of data rather than lack of interest precluded making equivalent studies of other organs involved more in intermediary metabolism.
Before assessing variation in organ size, it is worth considering some general factors that may set upper bounds to the size of the digestive system in birds.
Space It is obvious that there is a physical limit to the volume available for organs within the abdominal cavity of a bird. Fat deposits in the abdomen may further reduce the available space for the digestive tract. For example, migratory shorebirds deposit a thick layer of fat around the outer side of the gizzard, covering the intestines and rectum, and pressing up against the posterior end of the skeleton. In great knots (Calidris tenuirostris) about to depart on migration from Australia, fat in the abdominal cavity (95% of which was in discrete deposits) made up 38%of the total abdominal tissue mass (P.F.Battley, unpubl. data). At such times the profile of the abdomen
204
Physiological and ecological adaptations t o feeding in vertebrates
changes, with fat birds showing a "bulge" behind the legs (Owen, 1981; Wiersma and Piersma, 1995).Conflicts for space between fuel stores and organs may limit the maximum size of the digestive organs. Mass Internal organ mass may directly affect flying performance of the birds. The morphology and physiology of most birds favors minimal body mass. Because the costs of flight increase with body mass (Kvist et al. 2001) and maneuverability may be impaired at heavier masses (e.g.Kullberg et al. 1996; Metcalfe and Ure, 1995),minimizing digestive organ mass may be an important consideration. In a comparative study of raptors, Hilton et al. (1999) showed that "pursuing" species (that require fast and maneuverable flight) had shorter guts than "searching" species, presumably reflecting the pursuers' need to minimize mass. The shorter retention time of these species would reduce the period of time they had food in their guts, also keeping body mass low, though at a cost of lowered digestive efficiency. Sedinger (1997) showed that grouse have ceca 4-5 times longer than waterfowl feeding on the same diet, and suggested that the small ceca in waterfowl reflect a balance between the costs of flight for waterfowl and the benefits of the ceca for nutrient balance. Minimizing mass during flight has been proposed as the explanation for why some shorebirds seem to reduce the mass of their digestive organs shortly before departure on migratory flights (Piersma, 1998; Piersma et al. 199913; Piersma et al. 1993). In these cases, reducing the size of the digestive organs would also increase the space available for fat deposition. Energy Turnover Even in animals at rest, mass-specific maintenance costs of the intestine and liver are greater than those of muscle and adipose tissue (Blaxter, 1989; see chapter 2 by Hume, this volume; Krebs, 1950; Scott and Evans, 1992). Reductions in metabolically active tissues therefore dramatically reduce basal energy requirements (Battley et al. 2001a; Konarzewski and Diamond, 1995).
CHANGING ORGAN SIZES I N GALLIFORMES, PASSERINES, WADERS, AND WATERFOWL Organ size and mass changes are widespread and substantial in wild and captive birds. Figure 8.1 summarizes, from published literature, the magnitude of changes in three major gastrointestinal components, the gizzard, gut or small intestine, and liver. Data are separated into four taxonomic groups: galliformes (primarily grouse and quail), passerines, waders (shorebirds), and waterfowl. Changes in organ mass of 20-80% were common across a wide range of species; length changes of the intestine were lower, generally 10-20%.
Adaptive interplay gizzard mass
150
liver mass 6
4
4
20
90
100
gut length 7
6
15
Fig. 8.1. Absolute change (either increase or decrease) in digestive organs of birds. Data represent maximum percent changes recorded in studies for age- or sex-classes of birds, due to factors including diet type or quality, migration, breeding, and food intake. Measurements from wild and captive birds are included. The gut category represents data for both the entire gut (including ceca and rectum) and the small intestine only. Sample sizes for each taxonomic group are given above the box or points. Boxes represent the 25-75 percentiles (divided by the median), whiskers represent the loth and 90th percentiles, and outliers are shown as dots. For groups with only a few values, individual data points are plotted.
Interspecific Comparisons Surprisingly few interspecific comparisons of digestive organ size in birds have been made (Table 8.2) and the comprehensivenessof the comparisons varied greatly (samplesizes varied from 3 to 154 species). Substantial variation in organ size between species was attributed to differences in diet even though a direct link was not always established. A dietary explanation was suggested when variation in organ mass could not be fully accounted for by differences in body mass. Differences in gizzard and intestine mass were attributed to diet. For the gizzard, hardness of the diet is implicated, while for the intestine ease of energy assimilationof the diet is suggested (DuBowy, 1985; Richardson and Wooller, 1986). Intraspecific Comparisons Numerous studies have compared organ sizes in groups of birds collected during different periods of the year, or under different ecological or energetic conditions. For galliformes, the literature includes a good mixture
Gizzard
Organ
20
Passeriformes
18
5
Anseriformes
Anseriformes
Charadriiformes 19
6
10
Ayfhya spp.
Mass
Species Mass
Species Mass
Species Mass
Diet
Length
Diet
Measure
Barnadius, Glossopsitta, Neophema, Melopsittacus, Plafycercus, Trichoglossus Acanfhorhynchus, Acanfhiza, Anfhochaera, Colluricincla, Eopsalfria, Lichmera, Malurus, Meliphaga, Pachycephala, Pardalotus, Phylidonyris, Sericornis, Smicornis, Zosterops Arenaria, Calidris, Charadrius, Haemafopus, Limosa, Numenius, Tringa Anas, Ayfhya, Branfa, Bucephala, Clangula, Lophodyfes, Mergus, Somaferia
Factor
Species Length
Genera
Bonasa, Dendragapus, Lagopus, Tefrao
No. of spp.
Varied
Varied
Varied
Varied
Varied
Varied
Diet?
Diet
Diet
Diet
Diet
Diet
Response Reason
Differences between some species remained after body mass was accounted for.
contd.
6
5
4
3
2
1
Source
Species feeding on hard prey had gizzards much heavier than the (allometric) average. Comparison of feeding guilds - carnivore, omnivore, and herbivore. Differences in organ mass between species over and above differences due to body mass.
Nectar-feeding lorikeets had smaller gizzards than nonnectarivorous parrots. Nectarivores had smaller gizzards than insectivores of similar size, apart from two species of nectarivores that include larger insects in their diets.
Spruce grouse feeding on pine and spruce needles had especially large gizzards.
Explanation
Table 10.2. Interspecific comparisons of digestive organ size and dietary factors in wild birds
Psittaciformes
Galliformes
Taxonomic group
-
a
V,
n,
5
2rn
ID
<
-.3
3 lo
7D" ID
o
V,
-
3 o
3 a, c.
a, a
a,
E
lo. n
0
2
n, n
a
3
a,
-
n a,
2 0 S.
".
-<
Or
o N
Clangula, Histrionicus, Melanitta, Sonlateria
5
Anseriformes
20
3
Galliformes
Galliformes
Passeriformes 3 4
Measure
Diet
Varied
Length
Length
Varied
Varied
Diet
Diet
Diet
Diet
2 <
rn
a > -
contd.
u
8
-<
c L11
0 7
rr
3
9 = -. '
8
7
Source
Length of intestine reflects 10 diet of the three species willow ptarmigan has shortest intestine, eats willow twigs and buds. Rock and whitetailed ptarmigan consume more birch, or have a more varied diet, and have longer intestines.
Grouse ("browsers") have longer intestines for their mass than seed-eating galliformes.
Insectivores, frugivores and granivores separable on gizzard and intestine morphology.
Scaled against body mass, largest gizzards in species with high proportions of hard prey.
Response Reason Explanation
Structure Varied
Species Mass
Factor
Alectoris, Bonasa, Callipepla, Dief Canachifes, Centrocercus, Colinus, Cyrtonyx, Dendragapus, Lagopus, Lophortyx, Meleagris, Oreortyx, Pedioecetes, Perdix, Phasian us, Tetrao, Tympanuchus Diet Lagopus
34 genera of New World passerines
Genera
No. of SPP-
Taxonomic group
+ intestine
Intestine
Gizzard
Organ
Table 20.2 contd.
20
Passeriformes
Anseriformes 5
Anseriformes 5
Anseriformes 5
No. of SPP.
Taxonomic Group
Clangula, Histrionicus, Melanitta, Sonlateria
Clangula, Histrionicus, Melanitta, Somateria
Acanthorhynchus, Acan thiza, An thochaera, Colluricincla, Eopsaltria, Lichmera, Malurus, Me1iphaga, Pachycephala, Pardalotus, Phylidonyris, Sericornis, Smicornis, Zosterops Aythya spp.
Genera
Mass
Measure
Species Mass
Species Mass
Species Mass
Diet
Factor
None
Varied
Varied
Varied
Diet?
Diet?
Diet
Response Reason
Scaled against body mass, no difference among species.
Differences among some species remained after body mass was accounted for. Scaled against body mass, longest in common eider and shortest in harlequin duck and black scoter.
7
7
6
3
Source
Table 10.2 contd.
Insectivores had longer intestines for their mass than nectarivores, reflecting the more easily assimilated food (nectar).
Explanation
'Moss (1983); 2Richardson and Woller (1990); 3Richardson and Woller (1986); *Piersma et al. (1993); 5Barnes and Thomas (1987); 6Kehoe and Ankney (1985); 7Goudie and Ryan (1991); 8Ricklefs (1996); 9Leopold (1953); l0Moss (1974).
Liver
Organ
(D
V)
rCl n (D
(D
a 7
<
-.
n -. 3 a
$
V)
0 3
%
z
nl
n
nl
2
o a o -.
n
(D
n
3
nl
E!
g. n
g
s.
K
N 0 03
Adaptive interplay
209
of comparisons between categories of wild birds and captive experiments. Studies of passerines have tested for organ changes due to diet, season, breeding, migration, fasting and energy turnover. Field studies of waders have focused on changes associated with migratory preparation and longdistance flight; captive studies have investigated the influence of diet hardness on gizzard size. In the following summaries, organ changes in response to changing factors (e.g. season, diet quality) are identified, and the proposed reasons are given. Because different organs may respond differently to the same factor, data are summarized by organ rather than by study. As the biological questions addressed in these studies tend to be similar within bird groups, we review the intraspecificliterature in four taxonomic groups: the galliformes, passerines, waders, and waterfowl. By far the largest body of information has been compiled for waterfowl, for which seasonal changes in organ size have been studied in many taxa. These studies are sometimes difficult to interpret because more than one environmental or biological change occured during the period. Nevertheless, combining results from these four groups gives a good overview of the nature of organ changes across diverse taxa (Table 8.3). Gizzard Food or energy intake was the most common factor affecting gizzard size in the studies surveyed. Changes in intake resulted from changing food quality, increased energy demand during migratory fueling or experimental work level, and changes in feeding activity during breeding. Any factor that causes a change in volume of food being processed will apparentlyresult in a change in gizzard size. The primary role of the gizzard is to mechanically grind or crush food. It is not surprising that the nature of the diet also influenced gizzard size in all four bird groups. The common dietary correlate with changes in gizzard size was fiber content of the food. Captive experiments demonstrated that birds on high-fiber diets have larger gizzards than those on low-fiber diets (Kehoe et al., 1988;Miller, 1975; Savory and Gentle, 1976;Starck, 1999),but a shift to a high-fiber diet is also generally associated with an increase in food intake to compensate for the lower digestiveefficiency of a high-fiber diet (LiukkonenAnttila et al., 1999; Miller, 1975; Savory and Gentle, 1976). Studies of wild birds demonstrated that fiber alone cannot explain some of the changes observed in gizzard size. In female mallards (Anasplatyrhynchos),gizzards were generally large when fiber was high, but there was also an increase before migration, even though fiber was low at that time (Heitmeyer,1998). In spur-winged geese (Plectropterusgambiensis),gizzards increased dramatically during molt when the geese eat water plants high in fiber (Halse, 1985). This increase was only partly directly due to increased fiber (gizzards increased proportionately more than other digestive organs), as gizzard mass
210
Physiological a n d ecological adaptations t o feeding i n vertebrates
returned to original levels or lower after molt when dietary fiber content was higher (but intake presumably lower). The hardness of the prey also directly affected gizzard size, at least in red knots (Calidris canutus) that crush shellfish whole in their gizzards (e.g. captive birds fed on soft prey had much smaller gizzards than wild, shellfisheating counterparts: Piersma et al., 1993)and in canvasbacks (Aythyaafiinis) feeding on hard pondweed seeds during molt (Thompson and Drobney, 1994).Experiments with captive quail (Coturnixcoturnix) demonstrated that changes in grinding requirements (through varying amounts of nondigestible fiber in the diet) led to dramatic, reversible changes in gizzard size that were considered to result primarily from diet composition rather than food intake (Starck, 1999). In passerines, an increase in the proportion of plant food in the diet also resulted in increased gizzard mass (Ankney and Scott, 1988; Spitzer, 1972) but whether the supposed lower digestibility of plant food affected gizzards via changes in intake or grinding requirements was not clear. In waders, staging red knots that fly several thousand kilometers typically build up the size of the digestive tract (including the gizzard) as they fuel up (Piersma, 2002; Piersma et al. 199910). In contrast, in a short-hop migrant, the western sandpiper (Calidris mauri), gizzard enlargement was found only during migration periods and not during fueling in the tropics that preceded it, suggesting that the increases were in response to feeding en route during migration rather than fueling beforehand (Guglielmo and Willians, 2003).Whether this reflects a general pattern in short-hop migrants (that need to be able to refuel rapidly during short stopovers) in contrast to long-haul migrants, is not known. Decreases in gizzard mass were also recorded for nondietary reasons. In waders, a decrease in gizzard size just prior to migration has been documented in red knots and bar-tailed godwits (Limosa lapponica), and been interpreted as a strategic loss to minimize body mass for long migratory flights (Piersma, 1998; Piersma et al., 1999b; Piersma et al., 1993).During long-distance flight and fasting in captivity, digestive organs of great knots decreased in lean mass, reflecting unavoidable protein turnover and disuse atrophy (Battley et al., 2001b; Battley et al., 2000). Equivalent patterns were found in migratory passerines, with even more rapid changes than in the knots (Biebach, 1998;Hume and Biebach, 1996;Karasov and Pinshow, 1998). Passerines after long flights also had smaller gizzards than birds after shorter flights (Schwilchet al., 2002). Such decreases in gizzard mass can affect the immediate fueling capacity on arrival (Hume and Biebach, 1996; Karasov and Pinshow, 2000; Klaassen and Biebach, 1994).In waterfowl, factors associated with breeding were identified (egg formation: Mann and Sedinger, 1993) or implicated (postbreeding increase: Austin and Fredrickson, 1987) in gizzard reductions. A rapid decrease in gizzard size during breeding in female common eiders (Somateria mollissima),when birds feed very little if
Adaptive interplay
211
at all, could reflect protein demand rather than "functional" atrophy (Korschgen,1977). Intestine As with the gizzard, general diet type was associated with differences in intestine size (Table 8.3),particularly between galliform populations that fed on foods of differing digestibilities. Scottish red grouse (Lagopus lagopus scoticus)feeding on heather (Calluna vulgaris), high in fiber and tannins, had the longest guts of any population studied, while rock and white-tailed ptarmigan (Lagopus mutus and Lagopus leucurus) populations feeding on willow (Salix, which is highly digestible) had unusually short intestines (Moss, 1983). Likewise, black grouse (Tetrao tetrix) in Scotland feeding on heather had longer guts than Russian birds feeding on birch (Betula) catkins and twigs. Because high fiber results in low digestibility,lengthening of the intestine with high fiber levels allows for more food to be processed (higher intake) more efficiently. Moss (1989)showed that increasing the proportion of heather in the diet of captive red grouse resulted in increased food intake and intestine length. Concomitantly the heather in the diet was digested increasingly well (from a partial digestibility of 9.4% to the average for wild birds of 46%). He suggested that at high fiber levels, bulk limits food intake and the intestine (and perhaps more importantly the ceca) enlarges to improve digestion. For capercaillie (Tetrao urogallus),Russian birds had longer intestines than Scottishbirds, despite similar diets, suggesting that the colder and shorter days in Russia resulted in higher food intake (Moss, 1983). Although diet quality/fiber content and food intake interact strongly, other studies demonstrate that the intestine per se can respond to intake rather than fiber. Indeed, the factor most frequently identified behind changes in intestine length or mass was food intake. Rufous-collared sparrows (Zonotrichia capensis)had longer intestines in winter due to increased digesta volume (Novoa et al., 1996).Yet at the same time the diet shifted from seeds to high-energy insects, a change that was expected to lead to intestinal shortening. In waterfowl, changes in intestine mass were mostly ascribed to changes in intake, though increased intestine masses in birds on a low-quality or low-lipid diet probably reflected a need for increased digestive efficiency rather than changed food intake (Ankney and Afton, 1988;Thompson and Drobney, 1994).Food intake may also have been responsible for the increased intestinal size attributed to fiber (high-fiber diet mallards recorded by Kehoe et al., 1988). In migrant passerines and waders, the intestine may show hypertrophy during fueling. Like the gizzard, the intestine responds to periods without food intake, whether during long-distance flight or during a sedentary experimental fasting period, with mass loss. For red knots, there is evidence of strategicmass reductionsin the intestine before long-distance flights (Piersma et al., 1999b, also see below).
1
1
2 (1)
2 (1)
5 (1) 2
1
2 (1)
7
7
3
1 (1)
2
1
3
3 4 3
2
1
1
1
2 1 (1)
Diet Metabolism Intake Fibre Qua- Hard- Type Fat Protein Stra- Flight Fasting Breed lity ness tegic ingmass loss
Galliformes 1 (2) Passeriformes 5 (1) Choradriiforrnes 4 Anseriformes 6 (2)
Gizzard Galliformes Passeriforrnes Choradriiforrnes Anseriformes Intestine
Organ
1
3
3
1
1-4
Sources
18
6
contd.
21, 22, 25-31, 34-37, 43, 44
15-19, 56
11 4, 46-52 11 5-7, 10-13, 53-55
4
Diges- No No No. tive expla-change of effi- nation studies ciency
Table 8.3. Changes in three major components of the avian digestive system. Reasons given for any changes in the mass (or length for some intestinal studies) of the organ are summarized by taxonomic group. Numbers refer to the number of times that a change in organ size was attributed to that cause. Numbers in parentheses represent possible causes. Number of studies is the number reviewed for each organ/group. Because some studies comprised multiple comparisons or findings, the total number of causes documented may be higher than the number of studies.
-
-
-
-
-
1 1
-
(1)
4 (3)
1 (1)
(1)
3
1 1
2 1 (2)
Fat Protein Stra- Flight Fasting Breed tegic ingmass loss
Metabolism
-
-
-
-
-
Diges- No No No. Sources tive expla-change of effi- nation studies ciency
-
-
Table 8.3 contd.
'Liukkonen-Anttila et al. (1999); 2Moss (1989); 3Savory and Gentle (1976); 4Starck (1999); 5Al-Dabbagh et al. (1987); 6Ankney and Scott (1988); %iebach 1998); 'Brugger (1991); 9Dykstra and Karasov (1992); 'OGeluso and Hayes (1999); llHume and Biebach (1996); '2Schwilch et al. (2002); 13Spitzer(1972); 14Walsbergand Thompson (1990); 15Battleyet al. (2001b); I6Battley et al. (2000); '7Guglielmo and Williams (2003); "LandysCiannelli et al. (2003); I9Piersma et al. 1999b); 20Ankney(1977); 21Ankneyand Afton (1988); 22Austinand Fredrickson (1987); 23Drobney(1984); 2 4 d ~ B o(1985); ~ y 25Gauthieret al. (1984); 26Gauthieret al. (1992a); 27Gauthieret al. (1992b); 28Halse(1984); 29Halse(1985); 30Heitmeyer(1998); 31Hoba~gh (1985); 32Kehoeet al. (1988); 33Klaassen(1999); 34Korschgen(1977); 35Mainguyand Thomas (1985); 36Mannand Sedinger (1993); 37Miller(1975); 3sMiller(1986); 39Moormanet al. (1992); 40Paulus(1982); 41Reineckeet al. (1982); 42Smithand Sheeley (1993); 43Thompsonand Drobney (1994); "Tome (1984); %%yte and Bolen (1985); 46Fennaand Boag (1974); 47Leopold(1953); 4sMillanet al. (2001); 49Moss(1974); 50Moss (1983); 51Pulliainen (1976); 52Starckand Kloss (1995); 5 9 a v i s (1961); 54Karasovand Pinshow (1998); 55Novoaet al. (1996); 56Piersma (2002); 57Mortensenet al. (1983); 5sPiersmaand Gill (1998).
-
Diet Intake Fibre Qua- Hard- Type lity ness
Galliformes (1) Passeriformes 1 Charodriiformes 4 Anseriformes 4 (1) - -
Liver
Organ
214
Physiological and ecological adaptations t o feeding in vertebrates bivalves
l6 14
41
b4
V
10
3 , 3
12:
8
C,
m
6;
4
8
11
12
arthropods 33
mussels 14
-
n
m
periwinkles - mussels 15
t
bivalves 30
6 .1'
e+ se Pe
6
+ $ $
-
2 arrival
O Wadden Sea late winter
early
mid
Iceland northward migration
late
Canada Iceland Wadden Sea early southward mou lting breeding migration early winter
Fig. 8.2. Changes in the stomach mass of red knots of the subspecies islandica through the annual cycle, with indications of diet type and general activity. Data for the Wadden Sea are from T. Piersma (unpublished), for Iceland from Piersma et al. (1999b) and for Arctic Canada (northern Ellesmere Island) from R.I.G. Morrison, N.C. Davidson & T. Piersma (unpubl. data). Sample sizes are given above each box.
Liver Although livers show as much capacity for change as the gizzard and the intestine (Fig.8.1), liver size changes have attracted less attention. As with gizzard and intestine masses, increases in liver mass were associated with increased energy intake, while decreases were associated with flight and fasting (Table 8.3). In red knots and other very long-distance migrant waders, there is evidence that livers show strategic mass reductions before takeoff (Piersma, 1998;Piersma and Gill, 1998; Piersma et al., 199913). In addition, livers may undergo considerable temporary mass increases before and during the egg-layingperiod (Ankney and Afton, 1988; Drobney, 1984).It has been suggested that these variations correlated with changes in the intensity of lipid storage and protein metabolism (Ankney and Scott, 1988; Mortensen et al., 1983), mechanisms that would also explain changes in liver size during simultaneous molt in waterfowl (Austin and Fredrickson, 1987). In summary, the gizzard generally changes size with changes in fiber content (energy dilution of the diet), diet hardness (texture;this can also be a factor in fiber content), and flight considerations. The small intestine generally changes size with fiber content and/or substantial changes in food
Adaptive interplay
215
intake. The liver changes size with dietary dilution, food intake, and flight mass considerations (and undoubtedly also diet secondary chemistry,though this was not reviewed here). Thus, knowing the generality of these changes, we should be able to predict how changes in organ size will occur in birds in new situations.
FLEXIBILITY OF THE DIGESTIVE TRACT: DIET-RELATED VARIATIONS W I T H I N RED KNOTS Correlationsbetween digestive organ size changes and food type and intake illustrate the variety of functional interactions that can be found in birds. It is probable that the same range of interactions occur within species or even individuals through an annual cycle. We examine this possibility on the basis of the variable masses of the stomach (i.e.gizzard plus proventriculus, which is small in most waders: Piersma et al., 1993) of the red knot. This molluskivore species has a relatively large stomach (seeTable 8.2 and below) and spends most of the year in seashore habitats, but breeds much farther north on high arctic tundra where it eats spiders and arthropods rather than bivalves and gastropods (Piersmaet al., 2003). Red knots enable us to examine the effects of diet type, food requirement,and long-distanceflight on digestive organ size within a species. The red knots that breed in northern Greenland and northeast Canadian Arctic and spend the nonbreeding season in northwestern Europe (subspecies islandica), show a twofold variation in stomach mass over the course of a year (Fig. 8.2). In late winter on the nonbreeding grounds, the stomach has an average mass of more than 10 g, while on the tundra breeding grounds the stomach weighs only about 5 g. These data provide a basis for evaluating the roles of diet type, food intake and possible weight savings during longdistance flights. The stark difference in stomach size during breeding and overwintering cannot be a result of differences in energy turnover, as the overall energy expenditure of islandica knots on the tundra breeding grounds and the coasts of Europe in midwinter is remarkably similar (Piersma,2002; Wiersma and Piersma, 1994). The differencein stomach mass must instead reflect the contrast in diet type between the two periods; the spider and arthropod diet on the tundra requires only half the stomach mass of the hard-shelled mollusk diet in late winter. Increases in stomach mass of red knots between early and late winter in the Wadden Sea are probably due to increases in energy expenditure (Wiersma and Piersma, 1994)and therefore higher food intake on the molluskan diet. We have experimentally confirmed that stomach size does affect intake rate in knots eating hard-shelled prey. It is possible to manipulate the stomach size of captive knots by feeding them soft food pellets rather than their natural hard-shelled prey (Dietz et al., 1999a; Piersma et al., 1993). The gizzard of
216
Physiological and ecological adaptations t o feeding in vertebrates
long-term captive birds is roughly a quarter to half the mass of wild birds'. By using ultrasonography to non-invasively estimate stomach size and mass (Dietz et al., 1999a; Dietz et al., 1999b; Starck et al., 2001), the relationship between stomach size and intake rate could be studied. In an experimental situation,hungry red knots with reduced stomachshad significantly lower intake rates than birds with large stomachs (van Gils et al., 2003). In fact, the maximal sustained intake rate of hard-shelled prey (includingtime spent in digestive pauses) was directly proportional to stomach size. In contrast, when feeding on soft bivalve flesh (shells removed) there was no difference in intake rate between the groups. Changes in stomach mass of islandica knots during their stopover in Iceland (Fig. 8.2; Piersma et al. 1999b) illustrate the loss of stomach mass before and/or during a 2,500 km flight from the southern North Sea basin to Iceland (note low stomach masses at arrival) and the build-up of stomach mass during the fueling phase when food intake is necessarily high. The subsequent decline in stomach mass as birds approach departure condition for another 2,000-3,000 km flight from Iceland to the tundra breeding grounds may be an endogenously controlled mass reduction before flight or an adaptive preparation to the impending diet change on the breeding grounds (Piersmaet al. 1999a). A diet shft during staging from periwinkles to mussels supplemented with chironomid larvae (Alerstam et al. 1992) could either reflect or induce such a change. The stomach mass of red knots staging in Iceland during southward migration is low and may still reflect adjustments to a diet of soft foods on the tundra. Small stomachs are likewise found in red knots of the canutus subspecies arriving in the Dutch Wadden Sea from the Siberian tundra (A. Dekinga et al. unpubl. data). During their spring stopover in Delaware Bay, red knots of the rufa subspecies totally rely on the small eggs of horseshoe crabs (Limulus polyphemus) which are abundant on some beaches (Tsipoura and Burger, 1999). Horseshoe crab eggs require a grinding rather than a crushing action of the stomach, and small stones are found in the stomachs of red knots from Delaware Bay, presumably to assist with grinding. Patterns of change in stomach mass are remarkably similar in Iceland and Delaware Bay despite considerable differences in prey type and processing style (Piersma et al. 1999a). An initial increase in size at both sites is followed by a decrease in mass. Stomach masses peak at about 8 g when refueling rates are highest. The time course of changes in stomach mass has been confirmed by experimental studies on red knots that demonstrate halvings and doublingsof stomach mass over 5-7 days if the diet shifts are drastic (Dekinga et al. 2001). Recent field studies have confirmed the tight relationships between diet and stomach mass (estimated from ultrasonography) of radio-tagged red knots in the Wadden Sea (Piersma et al., 1999a). In August 1997estimated stomach mass correlated with the extent to which the flocks in which
217
Adaptive interplay 2 L)
,
stomach
intestine
liver
I I
I
II
h
h
05
-0 5 I8
14
26
22
ZO
I 8
I?
22
26
30
22
18
I4
26
30
(log) body mass (g) 10
a
a
b
c
ab
a
a
a
a
h
h
r/l
=s
04
0 2 -
00
I -
diet type
Fig. 8.3. Allometry of the fresh masses of stomach, intestine (the entire but emptied intestinal tract) and liver in 41 species of wader from four different families, assigned to four different diet "hardness" categories. The lower three panels show the (log) mean organ masses adjusted for differences in (log) body mass for the four diet categories. Diet categories sharing the same letter do not differ statistically. Covariance analyses were carried out in SYSTAT 10, and, incidentally, found no evidence for organ mass differences at the family level.
radio-tagged birds were found fed on cockles (Cerastoderma edule)rather than other bivalve prey that are easier to swallow and to crack. No other phenotypic characteristic of the radio-tagged individuals (sex, age, bill length, wing length, tarsus length, body mass) correlated with these differences in diet. Similarly,in August-September 1998red knots fed at two sites, one dominated by rather soft, small shorecrabs (Carcinusmaenas), and another where cockles were abundant. Radio-tagged individuals recorded feeding only at the shorecrab site had significantly smaller stomachs than individuals found only at the cockle site. Individuals feeding at both sites had intermediate stomach sizes (Piersmaet al., 1999a).
FEEDING ECOLOGY AND DIGESTIVE TRACT: COMPARISONS AMONG WADERS Our work on red knots has demonstrated that this species, depending on ecological context, can have stomachs that are either very small (in captive birds feeding on soft food pellets and in breeding birds on the tundra) or very
218
Physiological and ecological adaptations t o feeding in vertebrates
large (in wintering birds needing lots of hard-shelled mollusks to maintain a favorable energy balance). It would be very hard to pick a stomach mass 'typicalf of red knots for comparative analyses (cf. Harvey and Pagel, 1991) and, given the above summaries of studies on other birds (Fig. 8.1), this is likely to be true for most birds with seasonally variable diets. For a comparative analysis of organ size characteristics in such an ecologically and phylogenetically homogeneous group as the waders it seems much more relevant to categorize the diet of the birds than to worry about phylogenetic relationships. We therefore performed an allometric comparative analysis (Ricklefs and Starck, 1996) of the size of stomach, intestine, and liver in 41 wader species belonging to four Charadriiform families: oystercatchers Haematopodidae, avocets Recurvirostridae, plovers Charadriidae, and sandpipers and allies Scolopacidae. The diet of each species was categorized on the basis of stomach and intestinal content and on data from the literature (Table 8.4).For species with large body size differences between the sexes (ruff Philomachus pzrgnax, Eurasian curlew Nzrmenius avquata, and bar-tailed godwit Limosa lapponica), males and females were handled as separate entities, bringing the effective sample size to 44 taxa. We then assigned scores to the diets, ranging from 1- 4; these are 1 = only soft prey items such as polychaete worms, or extracted shell flesh as in oystercatchers, 2 = a mix of soft prey and soft-cased arthropods, 3 = mainly arthropods, and 4 = predominantly hard-shelled molluscan prey). We tested the prediction from this review and earlier studies (Piersma et al. 1999a; Piersma et al. 1993) that the stomach mass increases with diet hardness. In view of the interspecific correlation between stomach and intestine size shown from a smaller database (Piersmaet al. 1999a),we predicted that a harder diet should also lead to a larger intestine. For stomach mass, increasing diet hardness clearly led to a heavier stomach (Fig.8.3)while no differences were found between the two soft prey categories (1 and 2). As with red knot stomachs from the tundra and coastal wintering areas (Fig. 8.2), the stomach mass (adjusted for body mass) for wader species eating hard-shelled prey was more than double that of species eating soft prey. The intestine of molluskivore species (category4) was larger than that of taxa of intermediate diet categories (2 and 3) but not of species with the softest diet. For species that ingest mollusks whole, crush the prey in their stomach, and evacuate the crushed shells via the intestinal tract, heavier intestines may reflect the need to withstand the wear and tear of hard and sharp fragments during passage. As expected, there were no differencesin the relative size of the liver between the different diet categories.
- -
-
-
-
Common Name
-
Scolopacidae
NW Australia NW Australia
5 2 4
2 1 4
Charadrius semipalmatus Charadrius alexandrinus Charadrius leschenaultii Charadrius veredus Lymnocryptes minimus Gallinago media
semipalmated plover
greater sand-plover Oriental plover
great snipe
jack snipe
Kentish plover
Charadrius hiaticula
ringed plover
Netherlands Norway
-
terrestrial insects (mainly beetles), snails and seeds annelids, adult and larval insects earthworms, gastropods, adult and larval insects
crustaceans including crabs, mollusks
2
1
3
3
2
2
2
2
1
Code
contd.
polychaetes, gastropods crustaceans, polychaetes and mollusks Mauritania
polychaetes
6
Pluzlialis squatarola
Canada Guinea-Bissau, Mauritania, Netherlands Mauritania, Morocco, Wadden Sea Canada Morocco,
grey plover
-
terrestrial insects polychaetes, mollusks
bivalve flesh, polychaetes small worms + crustaceans
Predominant Diet
-
4 7
1
Netherlands Netherlands, Germany
Localities
earthworms, terrestrial insects earthworms, terrestrial insects
northern lapwing Vanellus vanellus Eurasian golden-plover Pluvialis apricaria American golden-plover Pluvialis dominica
Charadriidae
4 3
n
Netherlands 107 Netherlands
avocet
Haematopus ostralegus Recurviros tra avosetta
scientific ~ a m e
-
Recurvirostridae
Haematopodidae Eurasian oystercatcher
-
Family
Table 8.4. Source details and diets of shorebirds used in analyses of digestive organ size in relation to body mass and diet. Species are ordered taxonomically. Specimens from Canada are breeding birds from Churchill, Manitoba (R.E. Ricklefs, unpub. data). All other specimens were analyzed at the Royal Netherlands Institute of Sea Research. Diet code is a categorical scale from soft (1) to hard (4).
16
I-'
ru
Family 1
3
44 3 6
5 4 4 2 6
5 56 40
Gallinago gallinago Limnodromus griseus Limosa limosa Limosa haemastica Limosa Iapponica Numenius minutus Numenius phaeopus
Numenius arquata Tringa totanus Tringa flavipes Tringa terek Arenaria interpres Aphriza virgata Calidris tenuirostris Cnlidris canutus
short-billed dowitcher black-tailed godwit Hudsonian godwit bar-tailed godwit little curlew whimbrel
Eurasian curlew redshank lesser yellowlegs terek sandpiper ruddy turnstone surfbird great knot red knot
3 3
n
Scientific Name
Common Name common snipe
Localities
Netherlands, Germany
NW Australia
Wadden Sea, Mauritania Canada NW Australia Morocco, Mauritania Alaska
Mauritania, Netherlands, Canada Netherlands
NW Australia
Canada Netherlands Canada Netherlands
Europe
barnacles, bivalves, periwinkles and limpets bivalves, gastropods, crustaceans, annelids, sea cucumbers bivalves, gastropods, some crustaceans
larval and adult insects crabs crustaceans, annelids, mollusks
mollusks, crabs, polychaetes, earthworms crustaceans
crabs, terrestrial arthropods, earthworms
contd.
4
4
4
2
3
3
3
3
3
3
2
2
Code larval (+adult) insects, earthworms, small crustaceans larval, pupal and adult insects adult and larval insects larval insects polychaetes, bivalves, crustaceans, Tipulid larvae, earthworms insects, spiders, vegetable matter
Predominant Diet
Table 8.4 contd.
Adaptive interplay W d h l d T l ' T l ' h l
222
Physiological and ecological adaptations t o feeding i n vertebrates
GENERAL DISCUSSION
The three digestive organs reviewed here are all equally flexible in size in the wild and in captivity. While the reasons behind the size changes overlap, interpretation of size changes of the gizzard (digestive preparations), the intestine (digestion and absorption), and the liver (various metabolic processes related to digestion)becomes more uncertain the further one "climbs the digestive chain". This may reflect the diversity of physiological functions of each of the organs (i.e. low in gizzard, high in liver), on the basis of which we would expect size variations in gizzards to be most easily accounted for by characteristics of diet and food intake characteristics. However, it may also reflect our ignorance about these functions, and the relationships between functional capacities and size, of the most complicated and understudied organ, the liver (e.g.Whittow, 2000). Expanding the view to the entire digestive system (Table 8.1),do we expect strong correlation between size and capacity of these organs, which vary in complexity and range of function?In other words, is there consistent coadjustment between different parts of the digestive tract (Weibel, 2000)? Such studies would call for a much better understanding of the specific functions of different digestive organs, especial1 the intestine, the liver and accessory organs such as the pancreas and thc rceca. Detailed studies on organ function are also required to assess the other side of the equation, i.e. the costs associated with having organs of a certain size.Organs are clearly not maintained at maximal size and capacity throughout the year, but must reflect the costs and benefits in different ecological contexts. For example, what are the yields relative to the costs (in energetic terms) of temporary reductions in organ size and capacity before and during migration flights with different lengths and risk factors (Piersma, 1998; Weber and Hedenstrom, 2001)? This calls for studies of the trade-offs in functional morphology, leading to the field "behavioral ecology" of organ systems (Piersma and Lindstrom, 1997). Piersma et al. (1993)provided a heuristic scheme for understanding the influences of exogenous and endogenous factors on gizzard mass (Fig. 8.4). In general, some components of this model were supported in the next ten years. The model suggests that the mass of the gizzard is influenced by direct (endurance)training and by atrophy through disuse if the ingested volume of food decreases and/or food gets softer or less fibrous, and additionallyby factors such as protein reallocation. There is strong evidence for exogenous effects but the presence of a direct influence of an endogenous circamual oscillator on gizzard mass remains to be demonstrated (note that Dietz et al. 1999b, failed to find such an effect in red knots). Likewise, the extent to which proteins stored in the gizzard may be strategically shifted to other organs during the final phase of a migratory stopover, remains to be properly documented as well.
Adaptive interplay
1
MIGRATORY NUTRIENT DEMANDS
REALLOCA1-ION OF PROTEIN STORES
223
~
Fig. 8.4. Scheme outlining the causal feedback loops between gizzard mass, diet type and other external or internal modifiers. The valve in the upper right corner of the scheme indicates that there may be a minimum gizzard mass that determines whether or not hard-shelled or fibrous rich food can actually be ingested. Unmodified from Piersma et al. (1993).
Levels of adaptive modification other than the size and capacity of the digestive organs exist (Perez-Barberia et al., 2001). For example, gizzard linings may become more keratinized and the thickness of the intestinal mucosa may increase as prey become harder and their fragments sharper. We know that some waders replace their entire gizzard linings periodically (Meeuwset al., 1985)but the functional context and consequences for feeding are not yet clear. In addition to structural modifications, there may be chemical adaptations required to feed on parts of plants, including fruits, that are not readily digestible and may carry obnoxious chemical compounds (also see Levey and Martinez del Rio, 2001). An important area not explored in this chapter is whether there are genetic limits to the enormous phenotypic flexibility shown by digestive organs. How much additive genetic variation is there in organ size and/or flexibility for natural (or artificial) selection to work on? As an example, in a study of differences in growth, energetics,organ size and capacity in junglefowl (Gallus gallus) and their highly selected relatives, domestic broiler chickens (Jackson and Diamond, 1996),it was shown that selection for big bodies, rapid growth,
224
Physiological and ecological adaptations t o feeding in vertebrates
and long and heavy intestines has also led to small brains and light bones. This implies (1)that there is genetic variation for selection to work on and (2) that trade-offs exist not just within, but between different organ systems in the body. Complete understanding of variation in organ dimensions will eventually require studies of these trade-offs. Ackno wledgrnents
Over the past few years, our work on shorebirds has been supported by a PIONIER-grant of the Netherlands Organisation for Scientific Research (NWO)to T.P. and a doctoral scholarship from Griffith University, Brisbane to P.F.B. We are very grateful to the ornithologists that made shorebird carcasses available and the many people, especially Anne Dekinga and Sander Holthuijsen, who assisted during carcass analyses in the laboratory. We also thank R.E. Ricklefs, R.I.G. Morrison, N.C. Davidson, and R.E. Gill, Jr. for providing additional material on shorebird body composition. Sue Moore and two anonymousreferees made very helpful comments on the manuscript.
Al-Dabbagh K., Jiad J. H., and Waheed I. N. 1987. The influence of diet on the intestine length of the White-cheeked Bulbul. Ornis Scand. 18: 150-152. Alerstam T., Gudmundsson G . A., and Johannesson K. 1992. Resources for longdistance migration: intertidal exploitation of Littorina and Mytilus by knots Calidris canutus in Iceland. Oikos 65: 179-189. Ankney C. D. 1977. The use of nutrient reserves by breeding male lesser snow geese Chen caerulescens caerulescens. Can. J. Zool. 55: 1984-1987. Ankney C. D. and Afton A. D. 1988. Bioenergetics of breeding Northern Shovelers: diet, nutrient reserves, clutch size, and incubation. Condor 90: 459472. Ankney C. D. and Scott D. M. 1988. Size of digestive organs in breeding Brown-headed Cowbirds, Molothrus ater, relative to diet. Can. J. Zool. 66: 1254-1257. Austin J. E. and Fredrickson L. H. 1987. Body and organ mass and body composition of postbreeding female Lesser Scaup. A u k 104: 694-699. Barnes G. G. and Thomas V. G . 1987. Digestive organ morphology, diet, and guild structure of North American Anatidae. Can. J. Zool. 65: 1812-1817. Battley P. F., Dekinga A., Dietz M. W., Piersma T., Tang S. and Hulsman K. 2001a. Basal metabolic rate declines during long-distance migratory flight in Great Knots. Condor 103: 838-845. Battley P. F., Dietz M. W., Piersma T., Dekinga A., Tang S. and Hulsman K. 2001b. Is long-distance bird flight equivalent to a high-energy fast? Body composition changes in freely migrating and captive fasting Great Knots. Physiol. Biochem. Zool. 94: 435-449. Battley P. F., Piersma T., Dietz M. W., Tang S., Dekinga A., and Hulsman, K. 2000. Empirical evidence for differential organ reductions during trans-oceanic bird flight. Proc. Ro. Soc. London, B 267: 191-195. Biebach H. 1998. Phenotypic organ flexibility in Garden Warblers Sylvia borin during longdistance migration. J. A v i a n Biol. 29: 529-535. Blaxter K. 1989. E n e r g y m e t a b o l i s m i n a n i m a l s a n d m a n . Cambridge Univ. Press, Cambridge, UK. Brugger K. E. 1991. Anatomical adaptation of the gut to diet in Red-winged Blackbirds (Agelaius guttata). A u k 108: 562-567. Davis J. 1961. Some seasonal changes in morphology of the Rufous-sided Towhee. Condor 63: 313-321.
Adaptive interplay
225
Dekinga A., Dietz M. W., Koolhaas A., and Piersma T. 2001. Time course and reversibility of changes in the gizzards of red knots alternately eating hard and soft food. J. Exp. Biol. 204: 2167-2173. Dietz M. W., Dekinga A., Piersma T. and Verhulst S. 1999a. Estimating organ size in small migrating shorebirds with ultrasonography: An intercalibration exercise. Physiol. Biochem. Zool. 72: 28-37. Dietz M. W., Piersma T. and Dekinga A. 1999b. Body-building without power training: endogenously regulated pectoral muscle hypertrophy in confined shorebirds. J. Exp. Biol. 202: 2831-2837. Drobney R. D. 1984. Effect of diet on visceral morphology of breeding Wood Ducks. Auk 101: 93-98. DuBowy P. J. 1985. Seasonal organ dynamics in post-breeding male blue-winged teals and northern shovelers. Comp. Biochem. Physiol. 82A: 899-906. Dykstra C. R. and Karasov W. H. 1992. Changes in gut structure and function of house wrens (Troglodytes aedon) in response to increased energy demand. Physiol. Zool. 65: 422-442. Fenna L. and Boag D . A. 1974. Adaptive significance of the caeca in Japanese quail and spruce grouse (Galliformes). Can. J. Zool. 52: 1577-1584. Gauthier G., BCdard J., Huot J. and BCdard Y. 1984. Protein reserves during staging in greater snow geese. Condor 86: 210-212. Gauthier G., ChioniPre L. and Savard J.-P. L. 1992a. Nutrient reserves of wintering American Black Ducks in the St. Lawrence estuary, Quebec. Canadian Wildlife Service Progress Notes 202. Gauthier G., Giroux J.-F. and BCdard J. 199213. Dynamics of fat and protein reserves during winter and spring migration in greater snow geese. Can. 1. Zool. 70: 2077-2087. Geluso K. and Hayes J. P. 1999. Effects of dietary quality on basal metabolic rate and internal morphology of European starlings (Sturnus vulgaris). Physiol. Biochem. Zool. 72: 189-197. Goudie R. I. and Ryan P. C. 1991. Diets and morphology of digestive organs of five species of sea ducks wintering in Newfoundland. J. Yamashina Inst. Om. 22: 1-8. Guglielmo C. G. and Willams T. C. (2003) Phenotypic flexibility of body composition in relation to migratory state, age, and sex in the western sandpiper. Physiol. Biochem. Zool. 76: 84-98. Halse S. A. 1984. Diet, body condition, and gut size of Egyptaon geese. J. Wildl. Managmt. 48: 569-573. Halse S. A. 1985. Diet and size of the digestive organs of Spur-winged Geese. Wildfowl 36: 129-134. Harvey P. H. and Page1 M. S. 1991. The Comparative Method in Evolutionary Biology. Oxford Univ. Press, Oxford, UK. Heitmeyer M. E. 1998. Changes in the visceral morphology of wintering female mallards Anas platyrhynchos. Can. J. Zool. 66: 2015-2018. Hilton G. M., Houston D . C., Barton N. W. H. and Furness R. W. 1999. Digestion strategies of meat- and fish-eating birds. In: 22nd Int. Omith. Cong. N. J. Adams and R. J. Slotow (eds.). BirdLife South Africa, Johannesburg, pp. 2184-2197. Hobaugh W. C. 1985. Body condition and nutrition of snow geese wintering in southeastern Texas. J. Wildl. Managmt. 49: 1028-1037. Hume I. D. and Biebach H. 1996. Digestive tract function in the long-distance migratory garden warbler Sylvia borin. 1. Comp. Physiol. B 166: 388-395. Hume I. D . 2004 Concepts of digestive efficiency. In : Physiological and Ecological Adaptations to Feeding in Vertebrates. Science Publ. Enfield, NH. USA. pp 43-58 Jackson S. and Diamond J. 1996. Metabolic and digestive responses to artificial selection in chickens. Evolution 50: 1638-1650. Jehl J. R. 1997. Cyclical changes in body composition in the annual cycle and migration of the Eared Grebe Podiceps nigricollis. J. Avian Biol. 28: 132-142.
226
Physiological and ecological adaptations t o feeding i n vertebrates
Karasov W. H. and Pinshow B. 1998. Changes in lean mass and in organs of nutrient assimilation in a long-distance passerine migrant at a spring-time stopover site. Physiol. ZOO^. 71: 435-448. Karasov W. H. and Pinshow B. 2000. Test for physiological limitation to nutrient assimilation in a long-distance passerine migrant at a springtime stopover site. Physiol. Biochem. Zool. 73: 335-343. Kehoe F. P. and Ankney C. D. 1985. Variation in digestive organ size among five species of diving ducks (Aythya spp.). Can. 1. Zool. 63: 2339-2342. Kehoe F. P., Ankney C. D., and Alisauskas R. T. 1988. Effects of dietary fiber and diet diversity on digestive organs of captive mallards (Anas platyrhynchos). Can. /. Zool. 66: 1597-1602. Klaassen M. 1999. Physiological flexibility and its impact on energy metabolism and foraging behaviour in birds. In Herbivores: Between Plants and Predators, H . Olff, V. K. Brown, and R. H. Drent (eds.). Blackwell Science, Oxford, pp. 427445. Klaassen M. and Biebach H. 1994. Energetics of fattening and starvation in the longdistance migratory garden warbler, Sylvia borin, during the migratory phase. I. Comp. Physiol. B 164: 362-371. Klasing K. C. 1998. Comparative Avian Nutrition. CAB Int., Wallingford, UK. Konarzewski M. and Diamond J. 1995. Evolution of Basal Metabolic Rate and organ masses in laboratory mice. Evolution 49: 1239-1248. Korschgen C. E. 1977. Breeding stress of female eiders in Maine. /. Wild. Managlnt 41: 360-373. Krebs H. A. 1950. Body size and tissue respiration. Biochinrica et Biophysica Acta 4: 249-269. Kullberg C., Fransson T., and Jacobsson S. 1996. Impaired predator evasion in fat blackcaps (Sylvia atricupillu). Proc. Roy. Soc. London, B 263: 1671-1675. Kvist A., Lindstrom A., Green M., Piersma T., and Visser G . H. 2001. Carrying large fuel loads during sustained bird flight is cheaper than expected. Nature 413: 730-732. Landys-Ciannelli M. M., Piersma T., and Jukema J. 2003. Strategic size changes of internal organs and muscle tissue in the bar-tailed godwit during fat storage on a spring stopover site. Func. Ecol. 17: 151-159. Leopold A. S. 1953. Intestinal morphology of gallinaceous birds in relation to food habits. J. Wild. Managmt 17: 197-203. Levey D. J. and Martinez del Rio C. 2001. It takes guts and more to eat fruit: lessons from avian nutritional ecology. Auk 118: 819-831. Liukkonen-Anttila T., Putaala A., and Hissa R. 1999. Does shifting from a commercial to a natural diet affect the nutritional status of hand-reared Grey Partridges Perdix perdix? Wildl. Biol. 5: 147-156. Mainguy S. K. and Thomas V. G. 1985. Comparisons of body reserve buildup and use in several groups of Canada geese. Can. J. Zool. 63: 1765-1772. Mann F. E. and Sedinger J. S. 1993. Nutrient-reserve dynamics and control of clutch size in Northern Pintails breeding in Alaska. Auk 110: 264-278. Meeuws H., van Dooren J. and Voet H. 1985. Whimbrels Numenius phaeopus ejecting the lining membrane of their gizzard. Wielewaal 51: 305-309. Metcalfe N. B. and Ure S. E. 1995. Diurnal variation in flight performance and hence predation risk in small birds. Proc. Roy. Soc. London, B 261: 395400. Millin J., Gortizar C., and Villafuerte R. 2001. Marked differences in the splanchometry of farm-bred and wild red-legged partridges (Alectoris rufa L.). Poultry Sci. 80: 972-975. Miller M. R. 1975. Gut morphology of mallards in relation to diet quality. J. Wild Managmt 39: 168-173. Miller M. R. 1986. Northern pintail body condition during wet and dry winters in the Sacramento Valley, California. I. Wild. Managmt. 50: 189-198. Moorman T. E., Bladassarre G. A., and Richard D. M. 1992. Carcass mass, composition and gut morphology dynamics of Mottled Ducks in fall and winter in Louisiana. Condor 94: 407-417.
Adaptive interplay
227
Mortensen A., Unander S., Kolstad M., and Blix A. S. 1983. Seasonal changes in body composition and crop content of Spitzbergen Ptarmigan Lagopus mutus hyperboreus. Ornis Scand. 14: 144-148. Moss R. 1974. Winter diets, gut lengths, and interspecific competition in Alaskan ptarmigan. Auk 91: 737-746. Moss R. 1983. Gut size, body weight, and digestion of winter foods by grouse and ptarmigan. Condor 85: 185-193. Moss R. 1989. Gut size and the digestion of fibrous diets by Tetraonid birds. J. Exp. Zool. 3: 61-65. Novoa F. F., Veloso C., Lopez-Calleja M. V., and Bozinovic F. 1996. Seasonal changes in diet, digestive morphology and digestive efficiency in the Rufous-collared Sparrow Zonotrichia capensis in central Chile. Condor 98: 873-876. Owen M. 1981. Abdominal profile - a condition index for wild geese in the field. J. Wild. Managmt. 45: 227-230. Paulus S. L. 1982. Gut morphology of gadwalls in Lousiana in winter. J. Wild. Managmt. 46: 483-489. Perez-Barberia F. J., Gordon I. J., and Illius A. W. 2001. Phylogenetic analysis of stomach adaptation in digestive strategies in African ruminants. Oecologia 129: 498-508. Piersma T. 1998. Phenotypic flexibility during migration: optimization of organ size contingent on the risks and rewards of fueling and flight? J. Avian Biol. 29: 511-520. Piersma T. 2002. Energetic bottlenecks and other design constraints in avian annual cycles. Integ. Comp. Bio!. 42: 51-67. Piersma T. and Lindstrom A. 1997. Rapid reversible changes in organ size as a component of adaptive behavior. Trends Ecol. Evol. 12: 134-138. Piersma T. and Gill R. E. 1998. Guts don't fly: small digestive organs in obese Bar-tailed Godwits. Auk 115: 196-203. Piersma T. and Drent J. 2003. Phenotypic flexibility and the evolution of organismal design. Trends Ecol. Evol. 15: 18 : 228-233. Piersma T., Koolhaas A., and Dekinga A. 1993. Interactions between stomach structure and diet choice in shorebirds. Auk 110: 552-564. Piersma T., Dietz M. W., Dekinga A., Nebel S., 1999a. Reversible size-changes in stomachs of shorebirds: When, to what extent, and why? Acta Ornithologica 34: 175181. Piersma T., Gudmundsson G. A., and Lilliendahl K. 1999b. Rapid changes in the size of different functional organ and muscle groups during refueling in a long-distance migrating shorebird. Physiol. Biochem. Zool. 72: 405-415. Piersma T., Rogers D. I., Gonzalez P. M., Zwarts L., 2003. Fuel storage rates before northward flights in red knots world-wide: hitting an ecological constraint in tropical intertidal environments? In: Birds of Two Worlds, P. P. Marra and R. Greenberg (eds.). Smithsonian Institution Press, Washington, DC. Pulliainen E. 1976. Small intestine and caeca lengths in the willow grouse (Lagopus lagopus) in Finnish Lapland. Ann. Zool. Fennici 13, 195-199. Reinecke K. J., Stone T. L., and Owen R. B. 1982. Seasonal carcass composition and energy balance of female Black Ducks in Maine. Condor 84: 420426. Richardson K. C. and Wooller R. D. 1986. The structures of the gastrointestinal tracts of honeyeaters and other small birds in relation to their diets. Austral. J. Zool. 34: 119124. Richardson K. C. and Wooller R. D. 1990. Adaptations of the alimentary tracts of some Australian lorikeets to a diet of pollen and nectar. Aust. J. Zool. 38, 581-586. Ricklefs R. E. 1996. Morphometry of the digestive tracts of some passerine birds. Condor 98: 279-292. Ricklefs R. E. and Starck J. M. 1996. Applications of phylogenetically independent contrasts: a mixed progress report. Oikos 77: 167-172. Savory C. J. and Gentle M. J. 1976. Effects of dietary dilution with fiber on the food intake and gut dimensions of Japanese quail. Brit. Poultry Sci. 17: 561-570.
228
Physiological and ecological adaptations t o feeding in vertebrates
Schwilch R., Grattarola A., Spina F., and Jenni L. 2002. Protein loss during longdistance migratory flight in passerine birds: adaptation and constraint. J. Exp. Biol. 205: 687-695. Scott I. and Evans P. R. 1992. The metabolic output of avian (Sfurnus vulgaris, Calidris alpina) adipose tissue, liver and skeletal muscle: implications for BMR/body mass relationships. Comy. Biochem. Physiol. 103A: 329-332. Sedinger J. S. 1997. Adaptations to and consequences of an herbivorous diet in grouse and waterfowl. Condor 99: 314-326. Smith L. M. and Sheeley D. G. 1993. Factors affecting condition of northern pintails wintering in the southern high plains. J. Wildl. Managmt 57: 62-71. Spitzer G. 1972. Jahreszeitliche Aspekte der Biologie der Bartmeise (Panurus biarmicus). J. Ornifhol. 113: 241-275. Starck J. M. 1999. Phenotypic flexibility of the avian gizzard: rapid, reversible and repeated changes of organ size in response to changes in dietary fiber content. J. Exy. Biol. 202: 3171-3179. Starck J. M. and Kloss E. 1995. Structural responses of Japanese quail intestine to different diets. Deufsche tierarztliche Wochenschrift 102: 146-150. Starck J. M., Dietz M. W., and Piersma T. 2001. The assessment of body composition and other parameters by ultrasound scanning. In Body Composition Analysis of Animals. A Handbook of Non-destructive Methods, J. R. Speakman (ed.). Cambridge Univ. Press, Cambridge, UK, pp. 188-210. Stevens C. E. and Hume I. D. 1995. Comparative physiology of the vertebrate digestive system, Cambridge Univ. Press, Cambrige, UK (2nded.). Thompson J. E. and Drobney R. D. 1994. Nutritional implications of moult in male canvasbacks: variation in nutrient reserves and digestive tract morphology. Condor 98: 512-526. Tome M. W. 1984. Changes in nutrient reserves and organ size of female Ruddy Ducks in Manitoba. Auk 101: 830-837. Tsipoura N. and Burger J. 1999. Shorebird diet during spring migration stopover on Delaware Bay. Condor 101: 635-644. van Gils S. A., Piersma T., Dekinga A. and Dietz M. W. 2003. Cost-benefit analysis of mollusc-eating in a shorebird 11. Optimizing gizzard size in the face of seasonal demands. J. Exy. Biol. 206: 3369-3380. Walsberg G. E. and Thompson C. W. 1990. Annual changes in gizzard size and function in a frugivorous bird. Condor 92: 794-795. Weber T. P. and Hedenstrom A. P. 2001. Long-distance migrants as a model system of structural and physiological plasticity. Evol. Ecol. Res. 3: 255-271. Weibel E. R. 2000. Symmorphosis: On Form and Function in Shaping Life. Harvard Univ. Press, Cambridge, MA (USA). Whittow G. C. 2000. Sturkie's avian physiology. Acad. Press, San Diego, CA. Whyte R. J. and Bolen E. G. 1985. Variation in mallard digestive organs during winter. J. Wild. Managmt 49: 1037-1040. Wiersma P. and Piersma T. 1994. Effects of microhabitat, flocking, climate and migratory goal on energy expenditure in the annual cycle of Red Knots. Condor 96: 257-279. Wiersma P. and Piersma T. 1995. Scoring abdominal profiles to characterize migratory cohorts of shorebirds: an example with Red Knots. J. Field Omith. 66: 88-98.0
Gastrointestinal Responses to Fasting in Mammals: Lessons from Hibernators Hannah V. Carey University of Wisconsin, School of Veterinary Medicine, Department of Comparative Biosciences, Madison, WI, USA
SYNOPSIS Mammalian hibernators undergo dramatic seasonal changes in food intake and use of the gastrointestinal tract. For most hibernators, feeding ceases during the winter months and metabolism is fueled primarily by fat stores. As expected, the absence of food intake leads to significant atrophy of the intestinal mucosa.Yet, the overall structure of the epithelial layer remains intact with well-defined crypts and villi, little change in microvillus dimensions and only modest effects on the specific activity of brush border enzymes. Intestinal transport function is markedly reduced during periods of torpor, but when measured at 37OC rates of mass-specific absorption and secretion are similar to or even greater than those in fed, summer animals. The hibernatory fast is associated with several indicators of stress to the intestinal mucosa, likely due to the nutritional impact of extended fasting and the profound changes in splanchnic blood flow associated with torpor-arousal cycles. In response, hibernators appear to utilize a variety of protective mechanisms that maintain gut integrity and insure the retention of adequate digestive function upon emergence from hibernation in the spring.
INTRODUCTION The high cost of endothermy in mammals requires continual input of oxidizable substrates to support ATP synthesis and heat generation. Most mammals meet this need by the daily intake of food or feeding at semiregular intervals. Some mammalian species, however, display one-time or periodic
230
Physiological and ecological adaptations t o feeding i n vertebrates
intervals when feeding is suspended in favor of an alternative strategy that relies on stored lipids as the primary metabolic fuel. Perhaps the best examples of this strategy are the heterothermic mammals that periodically forego the maintenance of a constant, high body temperature (T,). These species undergo short- or long-term fasting concomitant with the use of daily or seasonal torpor to reduce energy demands. Examples include several species of desert rodents and hamsters that enter into and arouse from torpor on a daily basis and the seasonal hibernators that undergo multiday bouts of torpor and can fast for up to 8 months per year, such as marmots and ground squirrels. In contrast to heterotherms, other mammals undergo extended periods of fasting yet remain essentially euthermic, accompanied by only moderate or no effects on basal metabolic rate. These species include some ungulates during rut (Miquelle,1990),polar bears during the ice-free period (Atkinsonet al., 1996),and northern elephant seal pups that undergo a 2-3 month postweaning fast (Ortiz et al., 2001). This chapter begins with a brief overview of the effects of fasting on the mammalian gastrointestinal tract, focusing on responses in species that do not normally fast for more than 1-2 days. The remainder of the chapter describes current knowledge of the effects of hibernation on gastrointestinal structure and function in mammals. Included are gastrointestinal responses closely associated with the extended period of fasting, as well as other effects that derive from nonnutritional aspects of the hibernating phenotype.
GASTROINTESTINAL RESPONSES TO FASTING I N NONHIBERNATING MAMMALS The gastrointestinal tract is the first organ system directly affected by changes in nutrient intake and displays the most rapid and extensive responses to nutrient deprivation, both structurally and functionally. These changes can significantly affect digestion and absorption when feeding resumes due to loss of functional tissue as well as changes in the expression or activities of specific proteins, such as hydrolytic enzymes, solute transporters, and ion channels. Fasting appears to have both positive and negative effects on intestinal nutrient and ion transport. Functional responses to fasting may help alleviate structural changes that reduce total intestinal absorptive capacity. Nevertheless, ion transport responses to fasting may exacerbate the effects of enteric pathogens and other secretory agents. The effects of short-term fasting on the mammalian small intestine were recently reviewed (Ferraris and Carey, 2000) and hence just briefly summarized here.
Morphological Changes The mammalian intestine is respons:iblefor up to 25% of whole-body oxygen consumption (Cant et al., 1996) and is one of the most highly proliferative
Hibernation in mammals
231
tissues in the body. Hence it is an expensive organ to maintain in terms of both energy allocation and utilization of body components such as protein. Perhaps not surprisingly, gastrointestinal tissue shows considerable phenotypic flexibility (Piersma and Lindstrom, 1997); indeed, of all body tissues the gut shows the most dramatic response to reductions in food intake. Fasting for more than 1-2 days leads to a remodeling of the gut that reduces intestinal length and mass (in particular the mucosal compartment), while still preserving the basic crypt-villus architecture of the intestinal epithelium (Beck and Dinda, 1973; Ferraris and Carey, 2000). Thus, fasted animals continue to function with a smaller, yet still functional digestive tract. In a proximate sense, this effect is closely associated with the loss of lumenal nutrition and not the metabolic consequencesof nutrient deprivation per se. This is evident from studies demonstrating that total parenteral nutrition, which provides calories and nutrients to the body only through the systemic circulation, also induces intestinal mucosal atrophy (Dahly et al., 2002; Peterson et al., 1996). The general role of lumenal contents in regulating gastrointestinal growth is the subject of several excellent reviews (e.g. Goodlad et al., 1988; Johnson and McCormack, 1994). Fasting generally results in a decrease in mucosal surface area in the small intestine due largely to changes at the villus level. Fasting for as little as 24 h can reduce villus height, particularly observable in young animals, but the effect is commonly observed in all age groups for fasts of 72 h or longer. Villus height decreases with fasting due to significant reduction in number of cells along the crypt-villus axis. This results from reductions in cell proliferation and migration rates (Goodlad et al., 1988;Holt et al., 1986) as well as from increases in rates of cell loss (Boza et al., 1999).Intestinal epithelial cells (enterocytes) contain finger-like extensions of the plasma membrane known as microvilli. These greatly increase the membrane surface area and thus the density of hydrolytic enzymes, nutrient transporters, and other proteins involved in the digestiveprocess. Fasting has been reported to have a variety of effects on microvillus dimensions. For example, fasting in rats for 2-6 days produces longer and more slender microvilli and an increase in microvillus number per unit area, which results in an increase in microvillus membrane surface area (Gupta and Waheed, 1992;Waheed and Gupta, 1997). In contrast, in hamsters fasted for 2 days a decrease in microvillus height was observed (Misch et al., 1980), while a detailed stereologtcalstudy of rats either fed or fasted for 2 days revealed no significant effects on microvillus length, diameter, or density (Mayhew, 1990). These differencesin the effects of fasting on microvillus dimensions could be due to differences in location of the intestinal sections along the length of the gut, in the position of enterocytesalong the crypt-villus axis, or in the measurement techniques employed. In general, however, there is little evidence to support a consistent and significant effect of fasting on microvillus dimensions as is typically observed for villi.
232
Physiological and ecological adaptations t o feeding in vertebrates
Functional Changes Fasting has significant effects on intestinal nutrient absorption. Because fasting drastically reduces the lumenal concentrations of solutes, it not only decreases intestinal mucosal mass but also removes a known signal for upregulation of many solute transporters in well-fed animals (Ferraris and Diamond, 1989). Paradoxically, many studies have reported increases in brush border nutrient transport per mg intestine or per mg protein in fasted animals (Ferraris and Carey, 2000). Despite this apparently compensatory effect of fasting on the function of individual enterocytes, increases in massspecific nutrient transport may be insufficient to offset the marked decreases in mucosal mass during prolonged fasting in terms of whole animal nutrient absorption. The mechanisms responsible for the reported increases in massspecific nutrient transport during fasting are still unclear but could include changes in enterocyte membrane composition and fluidity (Gupta and Waheed, 1992),changes in brush-border membrane potential, and changes in gene expression of specific transport proteins (Ferraris and Carey, 2000). Some of these effects may be mediated by sensory and effector functions of neurohumoral reflexes throughout the gut that are activated by the change in food intake. Fasting also has marked effects on intestinal ion transport. Fasting for as little as 2 days has been shown to induce a shift in basal ion transport in the small intestine from a neutral or net absorptive flux to a more secretory state. This effect is most commonly reported as elevated basal short-circuitcurrent (Isc) of jejunal or ileal tissues (Carey et al., 1994), which reflects active ion transport (Darmon et al., 1996;Young and Levin, 1990b). The ionic basis for elevated Isc after 48-72 h of fasting has been identified in piglets and rats as an increase in net Cl- and in some cases, HCO; secretion (Carey and Cooke, 1989; Young and Levin, 1990b). The effect of fasting on ion transport in isolated tissues is typically paralleled by changes in fluid accumulation in vivo. Net anion secretion provides the electrical driving force for passive Na' movement into the lumen and both ions then provide the osmotic drive for fluid secretion. Thus, basal fluid absorption in ileal segments of rats is reversed to net secretion after 2 days of fasting (Young and Levin, 1992). Fasting also enhances ion transport responses to a broad range of intestinal secretagogues including agonists of cyclic nucleotide-dependent pathways (CAMPand cGMP) as well as those that stimulate intracellular calcium mobilization (Ferraris and Carey, 2000; Young and Levin, 1990a). Many of these are endogenous mediators of secretory pathways that regulate normal intestinal responses to feeding, immune defense, and other processes, which suggests that fasting has the potential to induce a hypersecretory state in an otherwisehealthy gut with no other underlying disease. Although intestinal hypersecretion is more commonly observed in animals subjected to fasting, nutrient deprivation can also enhance responses to agents that stimulate
Hibernation in mammals
233
intestinal ion absorption. For example, fasting causes a greater increase in Isc when Na+-couplednutrients, such as D-glucose, are added to solutions bathing the mucosal surface of tissues mounted in Ussing chambers (Carey and Cooke, 1989;Young and Levin, 1990a). The absence of food intake increases the movement of macromolecules across the intestinal epithelium in adult rats, infant rabbits and infant mice, and in children. The fasting-induced enhancement of macromolecular absorption appears to be due to increased endocytic uptake across the brush border membrane (Ferraris and Carey, 2000), which may represent an adaptive response to maximize protein absorption during starvation or malnutrition. On the other hand, it may also increase the likelihood of antigenic stimulation of the mucosal immune system and hypersensitivity responses, particularly in animals that do not normally experience fastinginduced changes in intestinal permeability on a regular basis. In summary, fasting leads to structural and functional changes in the mammalian intestine that appear to reflect both pathophysiological effects as well as compensatory adaptive changes that may serve to increase survival during periods of food scarcity. Much of our knowledge of the gastrointestinal response to fasting in mammals is based on studies using species that do not regularly experience extended periods without food intake during their lifetime. Hibernators represent an excellent example of mammals that shift each year from very active use of the gut to long periods of disuse. What can we learn from their experiences?
GASTROIN'TESTINAL RESPONSES TO FASTING I N MAMMALIAN HIBERNATORS From a nutritional perspective, mammals that hibernate can be grouped into those species that subsist primarily on fat stores during the h:ibernation season, and those that rely on cached food for their sole fuel source or as a supplement to their fat stores. Fat-storing hibernators, which include such species as marmots, ground squirrels, dormice and bears, essentially eliminate food ingestion during the hibernation season and instead rely primarily on the products of lipid hydrolysis (fatty acids and glycerol) obtained from white adipose tissue as their primary fuel source. Fat-storing hibernators are, therefore, in a fasted state for periods much longer than most other mammals ever experience. In contrast, food-storing hibernators such as chipmunks and hamsters are fasted only during torpor bouts because they consume cached food during periodic arousals to euthermy (Humphries et al., 2003). In some primarily fat-storing species (e.g.ground squirrels) stored food may be consumed by males late in the hibernation season during periodic arousals to support testis growth before aboveground emergence (Barneset al., 1986; Kenagy, 1989; Michener, 1998). Most of what we currently know
234
Physiological and ecological adaptations t o feeding in vertebrates
Nov
/;:.
/ Dee /;,'.,*
,
...............; Gut j i otrop hy i
Jan Fig. 9.1. Annual cycle of feeding and fasting in fat-storing hibernators such as ground squirrels or marmots. A typical hibernation season is indicated by shaded months, and is highly dependent on species, age-sex class, and locality. Dotted and solid lines indicate times when gut mass undergoes atrophy or growth, respectively. Changes in gut mass generally follow the annual cycle of food intake in hibernators.
about changes in gastrointestinal function in hibernators is from the fatstoring species. Thus, future studies that examine gastrointestinalresponses in food-storinghibernators will be a valuable contribution to our knowledge of the diversity of mammalian feeding patterns and the plasticity with which the gut responds to feeding and fasting.
Morphological Changes in Hibernators Hibernation has a marked effect on the structure of the gastrointestinal tract in fat-storing species (Carey, 1992; Hume et al., 2002). The annual cycle of food intake in these hibernators is paralleled by changes in small intestinal mass: intestinal mass and protein content increase significantly in the spring from a low during the hibernation season to the highest in midsummer, then decrease in the fall shortly before hibernation (Hume et al., 2002; Carey, 1990; Fig. 9.1).In 13-lined ground squirrels (Sperrnophilus tridecemlineatus), jejunal mucosal mass and protein content per cm are reduced by about 5075% from their peak values in midsummer (Carey, 1990; Carey and Sills,
Hibernation in mammals
235
1992).Mucosal atrophy in the hibernating gut is readily apparent from the reduction in villus length and total mucosal surface area (Carey, 1990, 1992). Interestingly, a 3-day fast in summer 13-linedground squirrels significantly reduced body mass, yet had little effect on mucosal mass or villus/crypt dimensions (Carey, 1992). These observations suggest that mammalian hibernators may be relatively more resistant to short-term fasting than are nonhibernating species, because in the latter a fast of this duration significantlyreduced mucosal mass and villus height (Baker and Campbell, 1989; Young and Levin, 1990a). Changes in gastrointestinal tissue dimensions and enterocyte proliferative activity in Alpine marmots (Marmota marmota) also reflect changes in food intake from hibernation through the active season (Hume et al., 2002). Marmots were obtained from the field shortly after emergencefrom hibernation in April, during the early part of the active season (April-May),in mid summer (July),and just before reentry into hibernation (mid-late September). The earliest measurements were obtained within the first 3 days after emergence, prior to the start of feeding. Lengths of stomach, small intestine, cecum, and proximal and distal colon were lowest immediately after emergence and increased as the active season progressed. The length of the distal colon decreased from July to September. Fresh tissue mass of these segments generally followed the same seasonal pattern, except that the decrease in distal colon length was not paralled by a decrease in mass from July to September, and the small intestine decreased in mass over this period. Mass of the small intestine increased 259% from the end of hibernation to midsummer, an increase similar to that observed for the change in mucosal mass per cm of jejunal tissues of 13-lined ground squirrels from hibernation to spring (Carey, 1990). Mucosal thickness of the stomach showed minimal change, but in the duodenum and ileum mucosal thickness increased significantlybeginning 20-30 days after emergence. Mucosal thickness of cecum and proximal colon also increased as the active season progressed but distal colon thickness remained essentially constant throughout the season. Despite the marked atrophy at the villus level, microvillus dimensions in ground squirrels are remarkably stable during the hibernation season with no change in microvillus height and a modest increase in microvillus density (Carey and Sills, 1996).Moesin, a member of the ezrin-radixin-moesin(ERM) familyof membrane cross-linkers, is expressed in the brush-border membrane of intestinal epithelial cells in torpid ground squirrels but not in summeractive squirrels nor in euthermic hibernators during periodic arousals (Gorhamet al., 1998).Moesin and the closely-relatedprotein, ezrin, provide structural and functional connections between the plasma membrane and the underlying actin cytoskeleton. In adult mammals, including 13-lined ground squirrels of all activity states, ezrin is found constitutively in the brush-border membrane of enterocytes.However, in torpid squirrels moesin
236
Physiological and ecological adaptations t o feeding in vertebrates
colocalizes with ezrin (Gorham et al., 1998). Although moesin has not been previously detected in small intestinal epithelial cells of adult mammals, in the rat small intestine moesin mRNA levels increase during embryonic development and then drop off after birth (Barila et al., 1995). In contrast, ezrin mRNA levels continue to increase after birth until the villus epithelium has become morphologically differentiated (Barila et al., 1995). This suggests the intriguing possibility that expression of moesin during torpor reflects a transient return to the fetal condition, consistent with the extended period of inactivity of the digestive tract during this time. Because ERM proteins are involved in the formation of membrane protrusions such as microvilli, induction of moesin during hibernation may play a role in the maintenance of intestinal microvilli (Carey and Sills, 1996) and the enhanced rates of active glucose uptake that occur during hibernation (Carey and Sills, 1992, 1996;see below). The dramatic reduction in intestinal mass and in particular the mucosal mass during hibernation likely provides substantial energy conservation during the hibernation season, because gastrointestinal tissue is expensive to maintain in terms of energy and protein balance (Cant et al., 1996). The proximate cause of hibernation-induced mucosal atrophy is the well-known effects of lumenal contents and, to a lesser extent, feeding-associated neurohumoral cues on gastrointestinal mucosal growth in all mammals (Johnson and McCormack, 1994).The important influence of lurnenal contents on changes in mucosal growth during the annual cycle is underscored by studies that compared mucosal structure in ground squirrels subjected to intestinal bypass procedures and either allowed to hibernate or remain active for a similar period of time (8-9 weeks). Mucosal mass and epithelial dimensions in bypassed segments of active and hibernating squirrels were reduced to a similar extent compared with values in both groups of squirrels that underwent sham bypass procedures (Carey and Cooke, 1991). Thus, lumenal nutrients exert a profound influence on mucosal growth regardless of whether animals are hibernating or not. In addition to the fasting-induced reduction in mucosal growth during the hibernation season, intestinal cell proliferation is further reduced by the depressed metabolism and low T, that occur during torpor (Fig. 9.2). This effect is consistent with the profound reduction in transcription and translation while hibernators are in the torpid state (Frerichs et al., 1998; Knight et al., 2000; Van Breukelen and Martin, 2001). DNA synthesis in small intestinal epithelial cells continues during torpor at about 4% of rates in active animals (Adelstein et al., 1967; Kruman et al., 1988);however, mitotic activity ceases during deep torpor (Adelstein et al., 1967; Carey and Martin, 1996; Dubinin et al., 1995; Kruman et al., 1988; Kruman, 1992; Mayer and Bernick, 1958;Soria-Milla,1986). Enterocytesprogress from the G, to S phases of the cell cycle during torpor but are apparently blocked in G, or late S (Adelsteinet al., 1967;Kruman et al., 1988). It is not clear whether the block
Hibernation in mammals
(I_
Spring -+Fall (fed)
Cells sloughed into lumen In 3-5 days
/
Hibernation (fasted) Cells sloughed into lumen
Cells retained on vill~5 days
T
Cells mlgrate
arrested I
237
\
Torpor
Interbout Arousal
T
-I
Low T,, MR
H~gh T, MR
Fig. 9.2. Effect of hibernation on enterocyte proliferation in the small intestine. Enterocyte proliferation is affected by feeding status as well as metabolic/thermal state.
in the cell cycle during torpor is due solely to low Tbs,or whether the G, arrest is a regulated response that minimizes errors in cell replication that might occur at low or variable temperatures (Kruman,1992). Enterocyteproliferation resumes rapidly once hibernatorsbegin feeding postemergence. For example, in alpine marmots the percentage of proliferating cells (mitotic index) in crypt regions of the duodenum and ileum is about 40% shortly after emergence and increases to 60% in July; in the ileum, the mitotic index then decreases from July to September, most likely due to reduction in food intake as animals prepare for hibernation (Hume et al., 2002). Rates of enterocyte proliferation during interbout arousals for most, but not all (Suomalainen and Oja, 1967) hibernators are similar to or exceed those recorded in the summer-active state (Adelstein et al., 1967; Dubinin et al., 1995).This is somewhat surprising because the absence of food intake during the hibernation season would be expected to reduce enterocyte proliferation regardless of T,, due to the strong link between food intake and mucosal growth. On the other hand, high rates of cell proliferation during interbout arousals are consistent with the observation that protein synthesis on a whole body level is higher shortly after ground squirrels have aroused from torpor compared with squirrels that had been active for 1-2 days (Zhegunov et al., 1988). Corresponding to changes in enterocyte proliferation is a drastic reduction during torpor in the migration of enterocytes from the proliferative zone in the crypts to the villus tips (Fig.9.2); this process resumes shortly after T,begins to rise during a periodic arousal (Carey and Martin, 1996). However, the duration of typical arousal periods (< 24 h) prevents most cells from reaching the villus tips, and cells are arrested in the position they held when torpor resumed. Full turnover of villi can take as long as 3 weeks in a hibernating squirrel compared with the typical turnover time of 3-5 days for nonhibernators (includingsummer squirrels).Interbout arousals therefore provide a euthermic period when new enterocytes are born and begin migration along the crypt-villus axis, which is necessary for their
238
Physiological and ecological adaptations t o feeding in vertebrates
maturation into one of the four functional cell types within the epithelium (absorptive epithelial cells, mucus-secreting goblet cells, enteroendocrine cells, and in the crypt, Paneth cells).Thus, enterocytes of hibernators represent an intriguing natural model of cellular senescence. Although young in the physiological sense (i.e. time spent at euthermic temperatures), these cells are chronologically much older than the typical enterocyte of a nonhibernator or enterocytes of ground squirrels during the active season. Both cell senescence and fasting probably contribute to recent observations that rates of enterocyte cell death (apoptosis)are increased in the small intestine during the hibernation season in ground squirrels. The percent of villus cells undergoing apoptosis, as indicated by TUNEL staining, increases as the hibernation season progresses, and expression of several pro- and antiapoptotic signaling pathways is greater in hibernating than in active animals (Fleck and Carey, 2003). Torpor appears to have similar effects on mucosal growth in food-storing hibernators as in fat-storing species. However, the presence of food in the lumen during periodic arousals appears to blunt the reduction in intestinal mucosal mass during the hibernation season and thus helps to maintain the gut close to its normal (fed) size. Small intestinal mucosal mass and protein content of European hamsters (Cricetuscricetus)was reduced after 12 days of torpor compared with active animals before or after the hibernation season, but mucosal mass in the hibernators returned to near-normal levels 48 h after arousal when feeding had resumed (Galluser et al., 1988).In garden dormice (Eliomys quercinus) which fed during periodic arousals, small intestinal mucosal thickness was increased by about 10% after a 30-day hibernation period, an effect suggested to be due to reduced blood flow to mucosal regions leading to edema in the intestinal wall (Soria-Milla, 1986). Epithelial cell height increased by 24%, which may be due to the longer residence time of cells on villi, and the mitotic index in the crypts was reduced by about 50% (Soria-Milla, 1986). Unfortunately,we lack detailed information on the response of the gut to winter dormancy in bears, which undergo metabolic depression in winter at T,s that are only mildly reduced from summer-activevalues (32-34°C minimal T,s during torpor). Bears remain in a fasted state during the 3-7 month hibernation season, which would be expected to reduce intestinal mucosal mass compared with levels in actively feeding bears. It has been suggested that the small intestine does not undergo atrophy during winter dormancy in black bears (Ursus americanus; Jones and Zollman, 1997)although data to support this are not available. Hibernation has a marked effect on the exocrine pancreas. Pancreatic protein content is decreased about 50% in hibernating bats and goldenmantled ground squirrels compared with active values (Bauman et al., 1987; Bauman, 1990). Since the vast majority of pancreatic tissue is made up of acinar cells and other supporting structures involved in digestive enzyme
Hibernation in mammals
239
and fluid secretion, this represents a considerable decrease in total pancreatic exocrine capacity. Ultrastructural observations have been made in the pancreas of hibernating hazel dormice (Muscardinus avellanarius),a species which stores little or no food during winter. Acinar cell zymogen granules (that contain digestive enzymes) are reduced in size during hibernation but rapidly regain their original size upon each arousal even in the absence of food intake (Malatesta et al., 1998).Levels of a-amylase protein accumulate in the rough endoplasmic reticulum and in zymogen granules as the hibernation season progresses (Malatesta et al., 1998). a-amylase specific activity and protein expression are reduced by 40-50% in hibernating 13lined ground squirrels compared with values in summer squirrels (BalslevClausen et al., 2003). This effect is consistent with the strong influences of fasting and diet composition on regulating pancreatic amylase levels in nonhibernating species.However, hibernators may retain pancreatic amylase levels during the winter fast to a greater degree than is possible in nonhibernators because maintenance of laboratory rats on a nitrogen-free diet decreases pancreatic amylase levels by as much as 72% (Brannon, 1990).
Functional Changes in Hibernators This section summarizes current understanding of how hibernation and the associated fasting period affect gastrointestinal function. Where information is available, pertinent details concerning experimental conditions are provided. This is relevant particularly for studies that compare the effects of hibernation on digestive function in animals sampled when in the torpid vs active state. Under these conditions, it is important to differentiate between processes measured at the low temperatures of torpor and at euthermic temperatures (i.e. -37"C), since Q,, effects alone can result in reduced tissue function whether or not an animal exhibits the hibernating phenotype. Thus, suppression of a function in a hibernating animal or its tissues when measured at low T, (e.g. intestinal solute transport) is not necessarily a consequence of hibernation per se. It is sometimes useful to ask whether a particular digestive process is still functional in hibernators at the low T,s of torpor, and in that case studying animals within the hibernation season at both euthermic and torpid Tbsis appropriate.Similarly,to determine whether the hibernating phenotype has altered a digestive process relative to animals in the active season, the animals or their tissues are best studied at the same temperature, most commonly at euthermic T,. Bile production Seasonal changes in composition of gall bladder bile have been reported in black bears. There were no differences in total bile acid concentrationsof 8 active and 14 dormant bears. However, the concentration of the bile acid tauroursodeoxycholate decreased and levels of calcium, magnesium, and copper increased in dormant bears compared with their active counterparts
240
Physiological and ecological adaptations t o feeding in vertebrates
('Jones and Zollman, 1997). Cholesterol content in bear bile increased nearly 6-fold during dormancy and the bile was more viscous, possibly due to increased mucin content. Gastrointestinal Motility Intestinal smooth muscle also loses mass during the hibernation season (Tooleet al., 1999),although the degree of muscle atrophy appears to be less than that typically observed for the mucosa. Motility patterns, either spontaneous or induced, are absent in duodenal segments harvested from torpid 13-lined ground squirrels when incubated in vitro at 4°C (Weekley and Harlow, 1986). However, tissues rewarmed to 37°C display intrinsic motor function and respond to addition of agonists such as cholecystokinin and angiotensin 11. This may be due to Q, effects on smooth muscle and neurohumoral activity, as duodenal tissues from active squirrels cooled from 37°C to 4°C also showed little spontaneous or agonist-induced motility (Weekley and Harlow, 1986). There are differences in the pattern of distribution of myenteric neurons in the intestine of active and hibernating golden hamsters (Mesocricetus auratus). Numbers of myenteric neurons immunoreactive for serotonin are reduced during hibernation whereas those containing tyrosine hydroxylase, substance P (SP), calcitonin gene-related peptide (CGRP), and vasoactive intestinal polypeptide are increased (Toole et al., 1999). The numbers of SP and CGRP neurons are also increased in myenteric ganglia in the proximal stomach during hibernation (Shochina et al., 1997). The functional significance of these changes has yet to be identified. However, alterations in neurochemical coding of myenteric neurons during the hibernation season may occur as a protective mechanism to help maintain the integrity of mucosal and muscular tissues when the normal stirnulatory effect of lumenal contents is reduced (Toole et al., 1999). Gastric Function Mucous secretion is increased in the stomach of Arctic ground squirrels (Spermophilusparryii) during torpor. Twenty-four h after return to euthermy induced by forced arousal, mucus content decreased to levels in active squirrels prior to hibernation (Mayer and Bernick, 1958). Staining of parietal (acid-secreting) and chief (pepsin-secreting) cells was less evident during hibernation and returned to normal after arousal. Functional correlates of these changes are evident in a different sciurid, the woodchuck (Marmota monax). Secretion of gastric juice continued at a basal rate in hibernating woodchucks (Friedman and Armour, 1936);the secretion was of relatively low pH (3.54.1) but devoid of pepsin. Intestinal Brush Border Enzymes Specific activities of the membrane-bound hydrolytic enzymes sucrase, isomaltaseand intestinal alkaline phosphatase are not affected by hibernation
Hibernation in mammals
241
in 13-lined ground squirrels; this constancy of expression is also evident at the protein level for sucrase-isomaltase and at the mRNA levels for sucraseisomaltase and intestinal alkaline phosphatase (Carey and Martin, 1996). Amino oligopeptidase specific activity is reduced during hibernation by about 20% although this relatively modest change was not apparent at the protein or mRNA levels (Carey and Martin, 1996).The effects of hibernation on digestive enzyme activity in food-storing hamsters are also relatively modest and appear to vary depending on the gut segment. Specific activities of sucrase and isomaltase were reduced after 12 d of torpor in the jejunum but not the ileum and activities of these enzymes returned to normal summer values after 40 h of feeding during a euthermic interval (Galluseret al., 1988). In contrast, lactase activity increased after 12 d of torpor and declined after the 48 h euthermic period (Galluser et al., 1988). Compared to values in summer hamsters, amino oligopeptidase activity was unchanged in the jejunum after a torpor bout but was enhanced in the ileum. Values declined to active levels after 48 h of refeeding (Galluser et al., 1988). Although based on only two species, these results suggest that specific activities of intestinal brush border enzymes show only modest changes from summer-active to hibernating states. This is somewhat surprising, given the dramatic reduction in lumenal substrates which are known to stimulate enzyme expression in the intestine of nonhibernating species. Indeed, it is likely that the differences between food- and fat-storing hibernators in response of brush-border enzymes to hibernation may be explained by the periodic ingestion of food by the former group during interbout arousals. Intestinal Nutrient and Electrolyte Transport It was first reported in the 1960s that rates of intestinal glucose absorption (normalized to tissue mass) were greater in hibernating ground squirrels than in summer squirrels, when tissues were studied in vitro at 37OC (Musacchia and Westhoff, 1964). Results of more recent studies indicated that net transepithelial absorption of Na+and the nonmetabolizable glucose analog, 3-0-methylglucose are essentially zero when jejunal tissues of hibernators are bathed in solutions similar to the T, of deep torpor (7OC), which is consistent with thermal effects of cooling on transport kinetics. However, transport rates increased rapidly when tissues were rewarmed and at 37OC were similar to those in summer animals (Carey, 1990; 1992; Carey and Cooke, 1992).Indeed, when transport rates were normalized to mucosal mass or surface area, absorption was actually greater in the hibernators than in summer squirrels, similar to the results of Musacchia and Westhoff (Musacchia and Westhoff, 1964). This remarkable ability to enhance nutrient transport at the level of individual enterocytes has been confirmed with other in vitro techniques, including short-circuit current (Isc) changes induced by mucosal addition of Na+-couplednutrients (Carey 1990), use of everted jejunal sleeves which measure absorption rates across the
242
Physiological and ecological adaptations t o feeding in vertebrates
brush border membrane (Carey and Sills, 1992),and glucose transport into isolated brush border membrane vesicles (Carey and Sills, 1996). The latter experiments demonstrated that enhanced sugar uptake in hibernators is evident even in isolated brush border membranes that contain the Na+-glucose transporter (SGLT1)but are dissociated from the rest of the enterocyte. In addition to sugar, amino acid absorption is also increased in hibernator tiss~~es on a mass- or area-specific basis (Carey, 1990; Carey and Sills, 1992). Examination of the kinetics of proline absorption in everted jejunal sleeves revealed that the Na+-dependency of proline uptake in tissues from hibernators (49%)is significantly greater than in summer-active squirrels, in which only 15% of total proline uptake is linked to the Na+electrochemical gradient across the brush border membrane (Carey and Sills, 1992). This suggests that hibernators maintain a greater capacity for concentrative amino acid uptake than do active squirrels, an effect that may facilitate the absorption of amino acids present at low concentrations in the intestinal lumen due to breakdown from sloughed epithelial cells. Alternatively, the shift in Na+-dependence of amino acid transport may allow for high rates of amino acid uptake shortly after emergence from hibernation in the spring, when food availability and quality as well as mass of absorptive tissue are relatively low. Studies that compare intestinal nutrient absorption in vivo in hibernating and active animals are needed to help put the in vitro results into a wholeanimal perspective. For example, it is important to determine whether the enhanced mass-specific nutrient transport rates measured in the hibernating intestine in vitro would be effective in intact animals when the substantial atrophy of the intestinal mucosa is accounted for. Some insights can be gleaned from studies by Musacchia and Bramante (1967) who measured glucose absorption in vivo in active and hibernating ground squirrels using an isolated perfused loop procedure. The T,s of active squirrels during the procedure was 36-38°C; hibernator T,s were initially 4-5°C and then rose gradually during the experiment, ultimately reaching 14-18°C. Glucose absorption from 5 cm segments of jejunum over a 30 min period was about 50% less in the hibernators when values were normalized to dry mass of tissue. Because transport rates were corrected for the difference in mucosal mass between active and hibernating squirrels, the reduction in sugar absorption in the hibernators was likely due to thermal effects on transport. However, 4 h after arousing from torpor, rates of glucose absorption in hibernators were close to, but did not exceed rates in active squirrels. The disparities between in vitro and in vivo measurementsof intestinal absorptive capacity likely involve the other factors that contribute to whole-body absorptive function, including hemodynamics and actions of local paracrine agents. Nonetheless, these results confirm the general conclusion that active nutrient absorption is preserved in long-term fasted hibernators. It is also clear from the study of Musacchia and Bramante (1967)that at T,s from -5 to
Hibernation i n m a m m a l s
243
18°Csignificant nutrient absorption occurs in the gut of a fasted hibernator. This absorption is partially due to active transport, because addition of phloridzin (which inhibits SGLT1) significantly reduced absorption rates compared to untreated segments. Maintenance of active transport function during hibernation provides strong evidence that despite overall mucosal atrophy, cellular proteins required for intestinal epithelial function are well preserved in hibernating mammals. Indeed, protein and mRNA levels of SGLTl are not altered by hibernation (Carey and Martin, 1996)and mRNA and specific activity of the basolateral enzyme Na+,K+-ATPase,which drives Na+-coupledsolute uptake into enterocytes, are also unaffected by hibernation (Careyand Martin, 1996). Observations that hibernation leads to enhanced absorptive function at least in vitro are consistent with the important role of lumenal nutrients in regulating intestinal function. In active ground squirrels subject to jejunal bypass procedures, rates of sugar and amino acid absorption normalized to mucosal mass or surface area are greater than in tissues proximal to the bypass that are still intact, or in tissues from squirrels subject to sham surgeries (Carey and Cooke, 1991). Interestingly, the transport rates in bypassed segments of active squirrels are similar to those in bypassed segments from hibernators (tissues were tested at 37"C),which suggests that the enhanced rates of mass-specific nutrient absorption in hibernating animals are closely associated with the absence of lumenal contents during the hibernation season. For hibernators as well as fasted animals, enhanced absorptive function of the intestinal epithelium may be an adaptive response to offset the reduced mucosal mass that occurs when lumenal nutrients are removed. The precise mechanisms that mediate the enhanced transport function are not entirely clear but may involve changes in brush border membrane composition and fluidity, changes in membrane potential, or other aspects of enterocyte physiology (Carey, 1995; Ferraris and Carey, 2000). Whether these effects occur in intestines of food-storing hibernators awaits further studies. In addition to absorptive function, active secretion of electrolytes is also preserved during hibernation. Although small intestinal tissues exhibit minimal basal or stimulated secretory activity when tissues are incubated at the T, of deep torpor (e.g.7"C),when warmed to 37°C basal rates of active ion transport are similar in tissues from active and hibernating squirrels (Carey, 1992). Basal tissue conductance, which includes both passive and active routes of ion transport across the epithelium, is increased in hibernating squirrelsby about 50%. The change in chloride secretion induced by electrical stimulation of enteric nerves or by addition of secretory agonists (carbachol or serotonin) is significantly greater in tissues of hibernators compared with active squirrels when rates are normalized to tissue mass. This again is likely related to the absence of lumenal contents during hibernation, because a 3-day fast in active ground squirrels also enhances intestinal electrolyte
244
Physiological and ecological adaptations t o feeding in vertebrates
secretion (Carey, 1992).Furthermore, in actively feeding squirrels segments of intestine that are bypassed from the normal flow of nutrients also show enhanced secretory activity (Carey and Cooke, 1992). Urea Recycling There is evidence to suggest that hibernating mammals can recycle urea during the hibernation season as a means of conserving protein stores during the winter fast (Barboza et al., 1997;Nelson, 1978;Steffen et al., 1980;Wright et al., 1999). Because microbial urease activity is required for urea hydrolysis in mammals, maintenance of a functional pool of gut flora that includes ureolytic bacteria during the hibernation season would be required. Only a few studies have addressed the effect of hibernation on the resident flora of the gut. In alpine marmots significant microbial activity as indicated by concentrations of short-chainfatty acids (SCFA)in the cecum is evident when animals first emerge from hibernation in the spring (Hume et al., 2002). Thus, microbial fermentation in the hindgut appears to continue at a basal level during hibernation. Interestingly, the total SCFApool size declined between July and September,which may reflect the gradual cessation of food intake as marmots prepare for the hibernation season. In 13-lined ground squirrels there is some reduction in numbers of viable bacteria in the cecum during hibernation (hibernators were without food for 6 or 42 days) but overall the cecal microflora remained remarkably stable (Barnes, 1970). Urea recycling during hibernation may be more efficient in bears who maintain a higher T, than in the smaller, deep hibernators such as ground squirrels and marmots; urea hydrolysis and synthesis of new amino acids from urea nitrogen may be restricted in the smaller hibernators to periodic arousals when appropriate enzyme activity is maximal. Mucosal Immune System A striking feature of the gut during hibernation is the marked increase in lamina propria and intraepithelial lymphocytes (IEL)in the mucosa (Carey et al., 2000a; Fichtelius and Jaroslow,1969;Shivatcheva and Hadjioloff, 1987). This change occurs in the absence of any increase in polymorphonuclear leukocytes or evidence of inflammation. The increase in lamina propria lymphocytes and IEL, which is observed in both torpid and interbout euthermic squirrels during the hibernation season, contrasts with the reduction in lymphocyte numbers within the specialized Peyer 's patches during hibernation (Shivatcheva and Hadjioloff, 1987). The torpid state is characterized by systemic leukopenia (Saito et al., 1978; Terien et al., 2001) and it has been suggested that circulating leukocytes are sequestered in the gut (and spleen) during torpor and are returned to the systemic circulation during arousals. However, because IEL are not known to cycle into and out of the intestinal epithelium, it is possible that intestinal immune defense is specifically increased during the hibernation season.The specific phenotypes of IEL that populate the intestinal epithelium during torpor are not yet known,
Hibernation in mammals
245
but a sigruficantproportion of IEL in rodents express the y6 surface phenotype. This subset is thought to recognize self-antigens and may function in the repair of stressed or damaged enterocytes (Boismenu and Havran, 1994; Groh et al., 1998).Thus, increase in the IEL population during hibernation may play a protective role to maintain epithelial function during the extended winter fast. This is suggested by observations that within a few days after feeding resumes in spring, IEL numbers return to summer values. In contrast, in recently aroused squirrels that have not yet fed IEL numbers are similar to those during hibernation (Fleck and Carey, pers. obs.). Gastrointestinal Response to Stress during Hibernation Fasting is intimately associated with the hibernating phenotype. The acquisition and ingestion of food and the state of torpor are for the most part mutually exclusive, and eliminating the energetic costs associated with foraging makes an important contribution to energy conservationduring the hibernation season. Yet, as described in this section, the hibernatory fast can impose nutritional stress to the gut and other organs. In addition, the profound physiologic changes associated with torpor-arousal cycles have the potential to compromise gastrointestinalfunction after spring emergence. Fasting is a viable strategy for mammalian hibernatorsbecause energetic demands during torpor and inte:rbouteuthermy are met by oxidation of fatty acids released from white adipose tissue, which spares body protein and plasma glucose. Thus, new dietary input to sustain energy demands is not necessary during hibernation. However, fasting also eliminates dietary sources of protective molecules that are not directly involved in meeting energy demands, such as antioxidants. Indeed, there is growing evidence to suggest that hibernators are at risk for oxidative stress during the hibernation season, which may be due both to processes that increase levels of reactive oxygen species (ROS)and loss of dietary antioxidants and their precursors during the winter months. Lipid peroxidation is one indicator of oxidative stress that is elevated in several tissues of hibernators (Harlow and Frank, 2001), including plasma (Chauhan et al., 2002). Conjugated diene concentrations in the intestinal mucosa are significantly greater in 13-lined ground squirrels during the hibernation season compared with levels in summer-active squirrels (Carey et al., 2000b). Within the hibernation season, these lipid peroxide metabolites are greater in early torpor (within -24 h of entering a torpor bout) than in squirrels arousing from torpor (Careyet al., 2000b). Intestinal oxidative stress is also evident during hibernation from changes in glutathione redox balance in mucosal tissues. Glutathione (GSH) is the major thiol-disulfide redox buffer within cells and in conjunctionwith GSH redox cycle enzymes, plays a key role in the detoxification of endogenous and exogenous ROS including lipid peroxides (Aw, 1999). Cellular oxidative stress is often manifested as a shift in the ratio of glutathione from its reduced (GSH)to oxidized (GSSG)forms, with lower ratios reflecting a more oxidized
246
Physiological and ecological adaptations t o feeding in vertebrates
state. The GSH:GSSG ratio in intestinal mucosa of hibernating 13-lined ground squirrels is 5-fold lower than in summer animals, an effect due primarily to elevated GSSG concentrations during hibernation. The change in intestinal GSH redox balance during hibernation does not appear to be due to differences in T,, but rather to other events associated with hibernation such as fasting or metabolic depression. For example, GSSG concentration and the GSH:GSSG ratio in euthermichibernators during an interbout arousal are very different from summer squirrels at the same T, -36 OC (Carey et al., 2003b). Fasting during the hibernation season could lead to GSH redox imbalance because the intestine relies heavily on dietary or biliary-derived GSH and GSH precursors for maintenance of mucosal GSH pools (Aw, 1994). Furthermore, enterocytes require exogenous glucose for NADPH production to maintain function of the GSH redox cycle and this may be limited by decreased availability of lumenal glucose in fasted hibernators (Aw and Rhoads, 1994). Physiological changes associated with hibernation apart from fasting also increase the potential for oxidative stress to the gut and other tissues. For example, the gastrointestinal tract is one of the most poorly perfused organ systems during torpor (-5% of normal rates) and one of the last to receive normal flow upon arousal (Bullard and Funkhouser, 1962). The depressed metabolic rates and low T,s of torpor certainly provide protection against ischemia-reperfusion injury by reducing cellular oxygen demands concomitant with reduced perfusion rates, and by reducing rates of reactions that lead to ROS production. However, it is possible that hibernators are still at risk for ischemic damage during entrance into and arousal from torpor, when oxygen and nutrient delivery to sensitive tissues such as the intestinal mucosa may not be adequately matched to metabolic demand. Other evidence that the gut is at risk for physiological stress during hibernation is the 2-3 fold increase in intestinal protein-ubiquitin conjugates during entrance and deep torpor compared with levels in the active season (Van Breukelen and Carey, 2002). Ubiquitin is a highly conserved 14 kD polypeptide that when conjugated to cellular proteins tags them for degradationby the 26s proteasome. An increase in the abundance of proteins conjugated to ubiquitin has been used as an indicator of the extent of protein damage induced by stress. Interestingly, ubiquitin-conjugate levels in arousing or interbout arousal squirrels are similar to those in summer squirrels (Van Breukelen and Carey, 2002). Thus, increased rates of protein damage associated with entrance into a metabolically depressed state may lead to accumulation of ubiquitylated proteins as animals enter torpor. Hibernation is associated with changes in levels of stress proteins in the intestine, including HSP70 and GRP75 (Carey et al., 1999, 2000b, 2003b). Both are known to be induced by oxidative stress associated with ischemic events in nonhibernating species. Another stressactivated protein whose expression is altered by hibernation is the
Hibernation in mammals
247
transcription factor nuclear factor-KB(NF-KB).Nuclear translocation of NFKBis greater in intestinal mucosa from hibernating ground squirrels than in summer animals (Carey et al., 2000b). Within the hibernation season, activation of NF-KBin the gut is evident as animals enter torpor, remains high throughout a torpor bout, and is lowest in hibernators arousing from torpor (Carey et al., 2000b). This pattern is not common to all tissues because brown adipose tissue displays little evidence of basal NF-KBactivation in either active or hibernating squirrels (Carey et al., 2000b). In cells of nonhibernating species, NF-KB activation is induced or enhanced by prooxidative shifts in the GSH redox state, particularly when GSSG concentrationis elevated (Droge, 2002). Thus, increased activation of NF-KB in the gut during hibernation is consistent with higher levels of GSSG during this time (Carey et al., 2003b). Although NF-KBactivation in nonhibernators is associated with a variety of gastrointestinal diseases such as ischemiareperfusion injury and inflammatorybowel disease, whether its target genes contribute to disease pathology or exert protective effects varies with the particular condition and the timing of NF-KBexpression. Because hibernation does not impart overt damage to the gut, it is likely that NF-KBactivation plays a protectiverole during hibernation. Many of the gene products induced by NF-KBare involved in antioxidant defense, recruitment of immune cells to specific sites via production of cytokines and chemokines, and regulation of cell proliferation and apoptosis. All of these pathways may be important for the hibernating phenotype, at least for the gut. As described above, lamina propria and intraepithelial lymphocytenumbers are increased in the mucosa during hibernation. In addition, enterocyte apoptosis increase during hibernation (Careyet al., 2000a; Fleck and Carey, 2003), which suggests that activation of NF-KBtarget genes may play antiapoptotic roles and thereby promote enterocyte survival during this time.
ECOLOGICAL AND PHYSIOLOGICAL CONSEQUENCES OF FASTING I N MAMMAUAN HIBERNATOR§
Although quantitative measurements are not yet available, it is likely that fasting in fat-storing hibernators provides significant energetic savings during the winter months when energy conservation is paramount. Because the primary fuel source switches from ingested nutrients in the active season to stored lipids during hibernation, maintenance of gastrointestinal tissue at the normal fed level is superfluous and would be energetically wasteful. Yet, despite the presumed energetic benefits of fasting for fat-storinghibernators, sole reliance on stored lipids for energy during the winter months may carry with it physiological and ecological disadvantages not encountered by species that rely primarily on stored food for fuel (Humphries et al., 2002).
248
Physiological and ecological adaptations t o feeding in vertebrates
Deposition of adequate fat stores to support hibernation requires time during the active season for food ingestion, assimilation, and conversion to lipids, and may increase the risk of predation due to reduced mobility. Restoration of gut mass after the hibernation season likely involves a significant fraction of the energy consumed in the days and weeks following spring emergence, and may also utilize any lipid stores present after the hibernation season has ended. Because these resources could be spent on other processes such as growth of lean body tissue (particularly for yearlings), reproduction and territorial defense, the cost of fasting for a fat-storing hibernator could be significant.Furthermore, as outlined above, there is growing evidence that the physiologcalconsequences of eliminatingfood intake for extended periods may be risky for hibernators. Such considerations may help explain the diversity among hibernating species in fat- vs food-storage as mechanisms to support energetic costs during the hibernation season (Humphries et al., 2003).However, reliable estimates of the energetic savings accrued by reducing gut mass during hibernation as well as the costs of gut restoration in the spring are needed to fully evaluate these relationships. Humphries et al. (2001)reported a positive correlation between amount of time eastern chipmunks spent torpid and the dry matter digestibility of food consumed during the hibernation season. This was interpreted as evidence that in contrast to fat-storinghibernators, food-storingspecies may maintain some level of digestive function during bouts of torpor to increase the assimilation of food consumed during periodic arousals (Humphries et al., 2001). The minimal Tbs during torpor in the chipmunks studied by Humphries (Humphries et al., 2001) averaged 17-19°C (despite a constant ambient temperature of 7°C). It is therefore possible that significant nutrient digestion and absorption will only occur at relatively high Tbsduring torpor, i.e. at Tbsabove those that are typical of hibernators in deep torpor (0-5°C). In-vitro absorption of glucose and sodium by intestinal tissues of a fat-storing hibernator (ground squirrel) was as high as 40% of euthermic values at an intermediate temperature of 25°C (Carey, 1990). Furthermore, as described above, active glucose absorption can occur at significant rates in ground squirrels at Tbs between -5 and 18°C (Musacchia and Bramante, 1967). Because pancreatic lipase retains 1/3 of its maximal activity for lipolysis at Tbsas low as O°C, it is conceivable that enzymatic hydrolysis of food can occur during intermediate T,s within the hibernation season if pancreatic secretion takes place. Because mammals typically produce pancreatic digestive enzymes well in excess of the substrates available for hydrolysis, even a 50% reduction in enzyme production during hibernation may not impose significant costs to the hibernator. Thus, both food- and fat-storing hibernators may be capable of digestion and absorption if nutrients are present in the gut at intermediate Tbsduring entrance or arousals from torpor or during shallow torpor bouts. The possibility that torpor and digestion are not incompatible awaits confirmation from studies that measure the range of
Hibernation i n mammals
249
T,s over which food is digested and absorbed in food- and fat-storing hibernators. The degree of torpor expressionby food-storinghibernators may be limited by rates of ingestion and/or gastric emptying during euthermic intervals. Regardless of whether enzymatic digestion of food and nutrient absorption continue during torpor bouts, food ingestion is not compatible with torpor. Thus, the length of euthermic intervals must be sufficient to allow ingestion of enough energy to sustain energetic costs accrued during the next torporarousal cycle as well as to maintain a functional gastrointestinal tract (Humphries et al., 2001). An analysis of available literature on food habits and hibernation patterns revealed that species that rely on hoarded food were primarily those with granivorousdiets; of the species examined, four of five granivorous groups are food-storing hibernators, and all eleven nongranivorous groups are fat-storers (Humphrieset al., 2003). Interestingly, body size does not appear to be a factor in discrimination of food- vs fatstoring hibernators.
CONCLUSIONS Direct comparisonsbetween hibernating and nonhibernating mammals in the gastrointestinalresponse to fasting are not easy to make because ethical issues limit the length of experimentally imposed fasts in species that consume food on a regular basis. Furthermore, in those species that do undergo extended fasts for periods of more than several days but do not undergo metabolic depression (e.g. polar bears, ungulates during rut), comparable measurements such as changes in gastrointestinal tissue mass, histology and function are for the most part not available. However, based on the existing studies it is clear that the gastrointestinalresponse to fasting in mammalian hibernators shares several similarities with fasting in nonhibernators. For example, tissue mass and length of most intestinal segments is significantly reduced during hibernation from peak values in midsummer. Mucosal tissue, particularly in the small intestine, undergoes marked atrophy with significant reductions in villus height and crypt depth. In contrast,both fasting and hibernation have lesser impacts on enterocyte microvillus dimensions.The enterocytes that remain or are born during the hibernation season appear to be functionally intact, and transport nutrients and electrolytes (when tested at euthermic T,) at rates similar to or greater than during the fed state, depending on the basis of normalization of transport rates. Similarly, mass-specific upregulation of nutrient and electrolyte transport is also observed in nonhibernators fasted for several days, although it is difficult to assess whether intestinal epithelial function would be as well preserved in nonhibernating species fasted for several months as it is in hibernators. In some cases, however, it does appear that
Gut associated immune system
261
microbes. The damaged epithelial cells quickly begin apoptosis and new epithelial cells migrate over the defective cells, thus resealing the mucosa and reestablishing the barrier (Blikslager and Roberts, 1997; Feil et al., 1989). If physical or chemical damage causes the loss of patches of epithelial cells, repair results from a combination of villi shortening and migration of remaining enterocytes over the basement membrane to close the gap. This process, called epithelial restitution, can be a primary driving force of intestinal morphology (Feil et al., 1989).Chronic damage results in high rates of cell replacement and triggers high rates of cell division in the crypts, which leads to expanded crypt area and depth. Villi may be blunted when cell replacement does not keep pace with the loss of damaged cells (Cuvelier et al., 2001; Ferreira et al., 1990; Heitman et al., 1980). The peristaltic action of the intestines flushes digesta, mucus, and bacteria through the GI tract. The flow of digesta is usually greatest in the esophagus followed by the small intestine. Following meals, the small intestine receives a steady input of digesta from the stomach and has a sufficiently small diameter to create a continuous distal movement of contents. The intestinal tissue has a layer on its surface that is not disturbed by peristalsis (Strocchi and Levitt, 1991).This unstirred layer comprises mucus entrapped in the glycocalyx and provides a stable but viscous ecosystem for microflora adapted to this environment. Some species of commensal microflora are able to use mucus as a nutrient source.
Secretions Acidification in the stomach serves as an effective defense. The degree of acidification varies considerably among animal species but in many a pH of 3 or less occurs when food is present (Klasing, 1998; Morton, 1979; Stevens and Hume, 1995).Few species of bacteria can survive such a low pH for an extended period of time, although Helicobacter species are exceptions (Lee, 1999).However, many species of bacteria can transit through the acid stomach, especially if they are encapsulated or associated within food particles, and establish residence in the intestines where a neutral pH is maintained. Mucus comprises a family of glycoproteins called mucins. The highly glycosylated forms, such as MUC5AC and MUC6, form a viscous gel extending from the glycocalyx of the epithelial cells and protects them by reducing the shear forces of the lumenal stream of digesta (de Bolos et al., 2001). Secretory IgA (sIgA) adheres to mucin and, by binding and trapping bacteria, prevents their attachment to epithelial cells (Belley et al., 1999).Mucus continually flows posteriorly, sweeping out bacteria and antigenic debris. Mucin production increases markedly as the result of physical or microbial insult to the epithelium (Kelly et al., 1994).The importance of the mucous layer is illustrated by the observation that mucin knockout mice suffer from inflammatory bowl disease (Van der Sluis et al., 2002). Diet can affect the number of goblet cells and their rate of mucin secretion as well as the rate of degradation of mucin and its loss in the GI lumen (Lien
262
Physiological and ecological adaptations t o feeding in vertebrates
et al., 2001). In general, goblet cell numbers increase with dryness and coarseness of diet (Deprez et al., 1987; Yang et al., 2001a, b). Given the nutritional cost of mucin production (Nyachoti et al., 1997),dietary factors that increase mucus production and loss are usually interpreted as negative dietary attributes. However, the increased mucin in the lower GI tract induced by short chain fatty acids resulting from fermentation of fiber has been interpreted as a protective factor because it results in lower rates of bacterial translocation across the epithelium (Buddington et al., 2002a,b.) Paneth cells, a type of epithelial cell present in low frequency in the small intestine, produce defensins (Yang et al., 2002). Defensins are small cationic peptides that have antibiotic-likeeffects on a spectrum of bacteria, fungi, and viruses. Paneth cells also secrete the bactericidal protein lysozyme during a local immune response. Penetration of pathogens into the lamina propria of the GI tract triggers macrophages, mast cells, and eosinophils to release cytokines, bioactive amines, and eicosanoids (Bailey et al., 2000; Simmons et al., 2001; Yun et al., 2000). These regulatory factors act on the epithelium to: (1)stimulate goblet cells to increase mucous secretion; (2) stimulate paneth cells to release defensins and lysozyme; (3) induce enterocytes to down-regulate nutrient absorption and secrete chloride; (4) dilate capillaries so that increased blood flow to the epithelium delivers leukocytes and facilitates leakage of water and plasma into the lumen; and (5)stimulate contraction of smooth muscle to increase gut motility. This coordinated response provides a chemical and physical defense against lumenal pathogens and is effective at sweeping lumenal bacteria out of the GI tract. The normal digestive secretions of the GI tract are important components of the mucosal defense system. Acidification by HC1 in the stomach and enzymatic digestion throughout the tract denatures and hydrolyzes bacterial and food antigens to nonantigenic nutrients (e.g. amino acids, glucose).Even with these defenses, there is a continuous low level of translocation of bacteria and food antigens through the intestinal epithelium (Duffy, 2000; Gardner, 1988). Essentially all of the bacteria and food antigens that escape the intestine's exclusion mechanisms are intercepted and processed by its extensive immunosurveillance system (Par, 2000; Rueda and Gil, 2000).
Mucosal Immune System The gut-associated lymphatic tissue is the largest part of the immune system and consists of two primary components: the large organized lymphoid follicles (e.g.tonsils and Peyer patches) and the widely distributed leukocytes in the lamina propria and within the epithelium of the mucosa. Like all homeostatic systems, the immune system has an afferent arm that senses the presence of a potential problem and an efferent arm responsible for eliminating it. The afferent processes in the GI tract occur predominantly in the organized collections of lymphoid follicles, while the efferent processes
Gut associated immune system
263
are mediated mostly by the leukocytes scattered along the lamina propria and within epithelium. The organized follicles are located directly within the epithelial layer and contain germinal centers that resemble those of lymph nodes. In the intestines, the largest collections of follicles are called Peyer patches. Effectorcells include phagocytes, granulocytes,and lymphocytes. Macrophages and dendritic cells are the primary phagocytes and they reside mostly in the lamina propria where they intercept bacteria and macromolecules that penetrate the epithelium. Mast cells and eosinophils are the primary granulocytes and are also found in the lamina propria. Effector lymphocytes are found in both the lamina propria and the epithelium, although the types of lymphocytes in each of these locations are very different. The intraepithelial lymphocytes comprise around 2-30% of the normal epithelial cells of the intestines, with highest proportions in the small intestine and lowest in the large intestine. They are located above the basement membrane toward the basal end of the enterocyte,but may occasionallybe found toward the apical end. Most intraepithelial lymphocytes in the villi are CD8+T-lymphocytes; B-lymphocytes are rare and more often found in the crypts. Intraepitheliallymphocytes that encounter antigen can rapidly migrate through pores in the basement membrane and return to the lamina propria where they initiate immune responses (Mayer,2000; Nagler-Anderson, 2000; Rothkoetter et al., 1999).The majority of the lymphocytes in the lamina propria of the intestines are CD4+T-helper lymphocytesand B lymphocytes. Functionally the aggregates of organized follicles, which serve as the afferentcomponent of the GALT, are mostly located at transitions between areas of the GI tract that are highly populated by microbes and those that are more sterile (Fig. 10.1).Microflora and food antigens in the lumen of the GI tract are sampled by the M cells and passed on to the organized lymphoid follicles located on their basal surface. Those lymphocytes that recognize antigen proliferate in the germinal centers, mature into effector cells, and then migrate to the epithelia throughout the GI tract. These activated lymphocytes are especially enriched in areas where microfloral numbers are kept low, such as the proximal small intestine.
Variability in the GALT There are major anatomical differences in the layout of the GALT across species. Phylogenetic differences in the GALT across the vertebrate classes can explain some of the variability when comparing across divergent taxa (Zapata and Cooper, 1990), but tremendous variability remains between closely related species as well. Although detailed interspecificcomparisons of the GALT have not been made, some generalizationscan be surmised from published reports on the number and size of the organized follicles in a variety of mammalian species (Table 10.1). The general pattern is one of organized follicles (the affecter arm) located mostly around the pharynx and in the jejunum and ileum of the small intestine (i.e.Peyer patches)and effector
264
Physiological and ecological adaptations t o feeding in vertebrates
IE and LP \eS;/
Fig. 10.1. Typical locations of organized lymphoid follicles and densities of lumen microflora and intraepithelial (IE) and lamina propria (LP) lymphocytes in the gastrointestinal tract.
Carnivora
Carnivora
Primate
Rodentia Rodentia
Cat
Dog
Human
Rat Mouse
Pig
Artiodactyla
Mongolian Rodentia Gerbil Rabbit Rodentia
Order
Species
carnivore
Dietary
Predominant in stomach and jejunum; omnivore some in proximal colon and cecum
Predominant in jejunum and ileum; omnivore some in proximal colon and distal cecum Evenly distributed along the herbivore small intestine Predominant in jejunum and ileum; herbivore some in proximal colon and appendix
Predominant in jejunum and ileum; omnivore some in proximal colon and appendix
Predominant in stomach, duodenum carnivoreand jejunum; some in stomach, cecum, omnivore and large intestine Predominant in jejunum, ileum and omnivore appendix; few in duodenum, colon
Predominant in duodenum and jejunum; some in stomach, cecum, and large intestine
Areas of organized follicles1
Also have large continuous patch in cecum (cecal tonsils) where B-cell diversification occurs Also have large continuous patches in ileum where B-cell diversification occurs
Few PPs in duodenum; B-cell diversification does not occur in GALT
B-cell diversification does not occur in GALT
Also have large continuous patches in ileum when B-cell diversification occurs Very little organized follicles in large intestine; B-cell diversification does not occur in GALT
3-4 times more epithelial lymphocytes than other domestic species
Comments
Table 10.1. Localization of the GALT in the intestines of mammals
contd.
(Pabst and Rothkoetter, 1999; Pescovitz, 1998; Schummer et al., 1979)
(Mage, 1998; Stepankova et al., 1980; Yamasaki, 1971)
(Cornes, 1965; Kroese, 1998) (Defresne, 1998; Komazawa et al., 1991; Poskitt et al., 1984 (Komazawa et al., 1991)
(Cornes, 1965)
(Felsburg, 1998; Kolbjornsen et al., 1994; Schummer et al., 1979)
(Lutz, 1998; Schummer et al., 1979; Waly et al., 2001)
References
3
in
z
-<
V)
m
1 c
2
cr
P
herbivore
folivore
Perissodactyla Predominant in jejunum and ileum; some in cecum at apex and in proximal colon Diprotodontia Follicles in terminal half of the small intestine and a large follicle at the cecal-colon junction
Diprotodontia Follicles along the entire small intestine folivore
Horse
Koala
Ringtail possum
(Hannant, 1998; Schummer et al., 1979)
(Butler, 1981; Goddeeris, 1998; Liebler et al., 1988; Morrison, 1986; Parsons et al., 1991)
(Griebel, 1998; Josefsen and Landsverk, 1996; Schummer et al., 1979)
References
-
-
-
-
-
-
-
-
- -
'Follicles involved in development of the B-cell repertoire (bursa1 equivalent) are included, when known.
-
-
Brushtail Diprotodontia Follicles along the entire small intestine, folivore-omnivore (Hemsley et al., 1996a, b) especially in the duodenum possum DasyuroSeveral large patches in the insectivorous The total number of follicles in (Poskitt et al., 1984) Dusky small intestine the small intestine is two-times Marsupial morphia greater in this species than in Mouse domestic eutherian mice.
(Hemsley et al., 1996a, b)
Devoid of organized follicles in (Hemsley et al., 1996a, b) most of the cecum and large intestine
Also have large continuous patches in ileum where B-cell diversification occurs
herbivore
Predominant in jejunum; scattered patches in duodenum and ileum; smaller follicles in proximal cecum, proximal large intestine, and terminal rectum
Artiodactyla
Cow
Comments Also have large continuous patches in ileum where B-cell diversification occurs
Predominant in jejunum; scattered patches in duodenum and ileum; smaller follicles in proximal cecum, proximal large intestine, and terminal rectum
Artiodactyla
Lamb
Dietary strategy herbivore
Areas of organized follicles1
Order
Species
Table 10.1 contd.
Gut associated i m m u n e system
-..-..-..
Intraepithelial lymphocytes (celldm) Bacteria (Iog/g lumena1 content&)
Fig. 20.2. The relationship between numbers of intraepithelial lymphocytes (Vervelde and Jeurissen, 1993) and the population of bacteria in the lumen (Barnes et al., 1972) of various sections of the GI tract of chickens.
cells enriched in the intestine, especially the duodenum and jejunum (Fig. 10.1).In some species, organized follicles may also occur in the esophagus, stomach, ceca, and large intestine. Follicles in the ceca and large intestine are usually most prominent in areas near their connections with the ileum. The presence of specialized areas within the GI tract for microbial fermentation seems to be an important determinant of the location and quantity of GALT. Among domestic species, the pharynx-associated lymphoid tissues are proportionally larger and more developed in ruminants than nonruminants (Schummer et al., 1979).Presumably, this is caused by the high microbial content of the rumen, coupled with rumination. While all species of birds and mammals that have been examined have Peyer patches, these vary in size and location along the intestine (Table 10.1).Some of this variability is due to the use of specific regions of intestinal lymphoid tissue for diversificationof the repertoire of B-lymphocytes. For example,birds utilize their bursa, rabbits their cecal tonsils, and pigs, dogs, and ruminants use Peyer patches in their ileum as a primary lymphoid tissue for B-lymphocyte development (Landsverk et al., 1991).The remaining aggregates of follicles are involved in local defense and are greater in number and size in domestic herbivores than in omnivores. Fermentation-type ceca and large intestines typically have collections of follicles involved in managing the resident microflora (Table 10.1).
268
Physiological and ecological adaptations t o feeding i n vertebrates
It is striking that the distribution of the microflora within the GI tract differs from the distribution of the immune system along the tract (Fig.10.2). Areas with robust microflora, such as the rumen, ceca, and large intestine have relatively lower frequency of intraepithelial and lamina propria leukocytes and fewer aggregates of lymphoid follicles than does the small intestine (Josefsen and Landsverk, 1996;Pabst and Rothkoetter, 1999;Suzuki et al., 2002; Waly et al., 2001). In particular, the rumen is virtually devoid of immunoglobulin secreting cells (Josefsenand Landsverk, 1996; Sato et al., 1990),indicating that there is little management of microflora via the immune system in the lumen of this fermentation organ. The proximal small intestine has the highest concentrations of effector leukocytes in its epithelium, but a very low population of microbes. For example, in pigs, chickens, and mice there are almost ten times more lymphocytes per mm of epithelium in the duodenum than the cecum. Presumably the immune system's antibody and cellular responses to microflora in the duodenum suppress microbial numbers. This observation is supported by the fact that immunosuppressive drugs permit the overgrowth of microflora from the lower intestine into the more proximal intestine, resulting in malabsorption and diarrhea (Alverdy et al., 2000; Jones et al., 2001; Tomar, 2001). Interestingly, the amount of lymphoid tissue associated with foregut fermentation is less than that associated with hind gut fermentation. The rumen of the cow, for example, has few intraepitheliallymphocytes and little organized lymphoid follicles in the epithelium relative to the ceca (Butler, 1981;Morrison, 1986;Schummer et al., 1979).This may be because the water and nutrient absorptive responsibilities of the rumen are less than the ceca and the rumen can function with a keratinized stratified epithelium. This occlusive epithelium does not require extensive support by the immune system relative to the absorptive single cell epithelium of the hind gut. Faunivores frequently consume prey weakened by infectious disease and might be expected to have robust immune defenses relative to species that consume plant materials that are mostly pathogen free. It is interesting to note that domestic cats have more than twice as many intraepithelial lymphocytes (79/100 epithelial cells) in their small intestine (Waly et al., 2001) as do omnivorous pigs, mice, chickens, and humans (German et al., 1999; Rothkoetter et al., 1991; Suzuki et al., 2002; Vervelde and Jeurissen, 1993).However, insectivorous marsupial dusky mice have twice as many lymphoid follicles per unit of small intestine than omnivorous domestic mice (Poskittet al., 1984)and insect prey presumably do not possess microbes that are potential pathogens for mice. Currently it is not possible to separate the independent effects of phylogeny, dietary composition, and pathogens that contaminate the diet. Multiple comparisons between faunivores and phylogenetically matched species that specialize in the consumption of readily digestible plant components (i.e.frugivores, granivores, nectarivores) would provide a more defensible analysis but data are not yet available.
Gut associated immune system
269
The morphological features of the GI tract exhibit considerable plasticity between individuals of a species and within an individual throughout the year. The structural and cellular dynamics of the GALT may well exceed that of any other component of this organ system. For example, mice whose intestines have a normal conventional microflora have about 7 times more intraepithelial lymphocytes in their intestines than germ-free mice (Suzuki et al., 2002).The Peyer patches are similarly smaller in cellularity and size in germ-free animals, especially if they are fed antigen-free diets (Coates and Gustafsson, 1984;Honjo et al., 1993;Pabst and Rothkoetter, 1999;Stepankova et al., 1980).In the event of a pathogen challenge to the mucosa, the nurrlber of leukocytes in the affected area increases several hundredfold. Even in animals with normal microfloral populations, the numbers and types of lymphocytes in the mucosa are responsive to diet in omnivorous rodents, pigs, chicks, and humans (Cunningham-Rundlesand Lin, 1998;Field et al., 2002; Klasing, 1999).Surprisingly, leukocyte populations of the rurnen, which is mostly protected by structural attributes, are also influenced by diet (Josefsenand Landsverk, 1997).Though some of the diet-induced changes in populations of leukocytes are secondary to shifts in microbial populations or their fermentation products, some are directly induced by the molecular properties of dietary constituents.
Functional Responses of the Intestinal Immune System A crucial enigma of the GALT is how it can respond vigorously and effectively to antigens associated with true pathogens while ignoring or being tolerant to a constant deluge of antigens associated with the diet and commensal microflora. Antigens in the lumen of the GI tract are continually being sampled by membranous epithelial (M) cells, dendritic cells, and by the enterocytes themselves. M cells are specialized affecter cells that are interspersed among enterocytesunderlying the Peyer patches of a wide variety of avian and mammalian species (Jeurissenet al., 1999;Kitagawa et al., 2000; Pastoret, 1998;Zapata and Cooper, 1990).They transport antigens that they phagocytize to their basal membranes and onto underlying macrophages, lymphocytes, and dendritic cells. Dendritic cells are also located throughout the intestinal epithelium. They reside under the enterocytes but extend processes between enterocytes and into the intestinal lumen where they can sample the antigenic milieu (Nagler-Anderson, 2001). Apparently those antigens and microbes that are obtained through active sampling usually induce a tolerizing immune response characterized by IgA secretion and the differentiation of effector-memory T-lymphocytes. Those antigens and microbes that invade the epithelium causing damage, as well as those that have very specific adjuvant properties and engage Toll receptors, induce a more aggressive immune response characterized by proliferation of activated effector T-lymphocytes that invoke a local inflammatory response (Elson et al., 2001; Husband, 2002; Matzinger, 2002; Nagler-Anderson, 2000; Simmons
270
Physiological and ecological adaptations t o feeding in vertebrates
et al., 2001). Thus, it appears that the immune sytem uses the presence or absence of damaged tissue to decide whether it will mount a benign response aimed at clearing the antigen with minimal damage to the integrity of the epithelium or an aggressive response aimed at killing pathogens or infected cells but also damaging healthy tissue (i.e. immunopathology). IgA responses in the GALT B-lymphocytes that are stimulated by antigen in the Peyer patches mature into IgA secreting plasma cells due to the specific cytokine environment within the follicles. These cells travel to the GI mucosa and populate the lamina propria especially in the crypts, ducts, and glands. Their sIgA is transported across enterocytes and into the GI lumen where much of it binds to the glycocalyx and mucus, preventing IgA loss as digesta flows through. This intestinal IgA is strategically located to intercept food antigens that have escaped digestion in the intestinal lumen and to prevent the attachment of bacteria to enterocytes (Garside and Mowat, 2001). Food proteins bound by IgA have increased susceptibility to proteolysis by digestive enzymes. Additionally, IgAin the lamina propria can bind food proteins that translocate through the epithelium. The IgA-antigen complex is then taken up by cells expressing polymeric-Ig receptors including intestinal epithelial cells, hepatocytes (e.g.rodents), and bile duct epithelium (e.g.humans) and cleared from the body without initiating a systemic immune response (Brown et al., 1984; Garside and Mowat, 2001; Mayer, 2000). IgA is unique among the immunoglobulins in that its structure precludes it from fixing complement and activating inflammatory cells. Thus, IgA prevents dietary hypersensitivities because it intercepts antigens before they can activate more aggressive components of the immune system, such as those mediated by IgG, IgE, and T-lymphocytes. The importance of copious IgA secretion along the intestines can be appreciated by the fact that about 80% of all antibody-producing cells of the body are located in the GI mucosa (Brandtzaeg et al., 1989). Cell-mediated responses Cytokines regulate most aspects of an immune response, controlling the type, vigor, and duration of the response. The Peyer patches are rich in the cytokine transforming growth factor$ (TGFP),which drives antigen-activated lymphocytes to differentiate into phenotypes that mediate a condition best described as "controlled inflammation". CD-8 T-lymphocytes, which are predominant effector cells that kill pathogen-infected host cells, are influenced by TGFP to differentiate into a memory phenotype. After leaving the Peyer patches, these cells migrate along the epithelium and await the appearance of newly infected host cells.TGFP also influences the predominant regulatory lymphocyte, CD-4 T-lymphocytes, to assume a unique regulatory phenotype (Trl) that, upon contact with antigens, secrete cytokines that induce a hyporesponsive condition in effector leukocytes. This combination of
Gut associated immune system
271
regulatory and effector lymphocytes orchestrates the replacement of infected enterocytes without causing inflammation. Importantly,the T- lymphocyte does not kill the enterocyte but causes it to be pushed into the lumen and replaced, preserving the integrity of the epithelium (Lambolez and Rocha, 2001). In most species, many of the intraepithelial T-lymphocytes express a y6 T-cell receptor. This phenotype endows the T-cell to recognize unprocessed bacterial antigens and to recognize pathogens directly without the aid of antigen presenting cells. Their unique bacterial recogrution capabilitiespermit these intraepithelial lymphocytes to serve as sentinels of the GALT and to orchestrate cellular responses directed to bacteria without being distracted by food proteins (Boismenu and Havran, 1997; Ferrick et al., 1996). The result of these unique IgA and T-lymphocyte-mediated immune responses is oral tolerance. It is interesting that the immune system recognizes food antigens and commensal microflora, but its response is directed to their clearance from the mucosa without invoking cell-mediated responses that could damage the epithelial barrier. However, inappropriate responses to nonpathogenic antigens can occur. Some food macromolecules are particularly resistant to digestion and contain antigens that have molecular patterns that stimulate phagocytes and lymphocytes to orchestrate a Thl response, which causes inflammation and hypersensitivity (Elson et al., 2001).For example, plant lectins such as phytohemagglutinins from legumes trigger activation of intraepithelial T-lymphocytes, infiltration of additional lymphocytes, and changes in villi morphology. Lectins stimulate crypt hypertrophy, increased crypt cell production, and an even greater increase in rate of death of epithelial cells at the villi tip. This combination of cellular events leads to shortening of villi, increased crypt depth or proliferation, impaired absorption, and sometimes diarrhea (Ayyagari et al., 1993;Banwell et al., 1993; Ferreira et al., 1990; Radberg et al., 2001; Sjolander et al., 1986). Similar to the case with pathogens, food components that cause physical damage to the epithelium leading to cellular necrosis induce the release of intracellular molecules (e.g.heat shock proteins) that have adjuvant activity and mediate a vigorous immune response. Presumably, these immunostimulatory systems of plants have evolved as defenses against their herbivores.
FUTURE PERSPECTIVES Recent developments have detailed the mechanisms used by lymphocytes and phagocytes to distinguish invasive pathogens from background antigens of commensal microflora and food. Now that the functional attributes of the various GALT components are understood, we are poised for new questions that broaden our perspective on GI defenses. Virtually all of the studies of
272
Physiological and ecological adaptations t o feeding in vertebrates
mechanisms have used mouse or human cells. Although research with domestic cattle, pigs, and chickens provides an additional perspective, little beyond anatomical descriptions is known about other species. Given that the morphology and physiology of other systemsof the GI tract diverge greatly depending upon dietary strategy, research on the functional organization of the GALT of faunivorous, folivorous, granivorous, frugivorous, and nectarivorousspecies would be illuminating. The trade-off between the cost of immune defenses and productive processes such as growth and reproduction is a defining axis of an animal's life history. The GALT is the largest component of the immune system and should be a focal point for studies examining the role of immunity in life history. Determining the relationships between dietary strategies, commensal microflora, and immunological strategies would certainly provide exciting perspectives in animal biology. REFERENCES Alverdy J., Holbrook C., Rocha F., S e i d e n L., e t al. 2000. Gut-derived sepsis occurs when the right pathogen with the right virulence genes meets the right host: evidence for in vivo virulence expression in Pseudomonas aeruginosa. Ann. Surg. 232: 480489. Alverdy J. and S t e m E. 1998. Effect of irnmunonutrition on virulence strategies in bacteria. Nutrition 14: 580-584. Ayyagari R., Raghunath M., a n d Narasinga Rao B. S . 1993. Early effects and the passible mechanism of the effect of Concanavalin A (Con A) and Phaseolus vulgaris lectin (PHA-P) on intestinal absorption of calcium and sucrose. Plant Foods Hum. Nutr. 43: 63-70. Bailey M., Lamb C. E., Whiting C. V. and Bland P. W. 2000. Secretion of regulatory cytakines by T cells from the intestinal lamina propria. Res. Vet. Sci. 68: 17-28. Banwell J. G., Howard R., Kabir I., Adrian T. E., Diamond R. H., and Abramowsky C. 1993. Small intestinal growth caused by feeding red kidney bean phytohemagglutinin lectin to rats. Gastroenterology 104: 1669-1677. Bsmes E. M., Mead G. C., Barnum D . A. and Harry E. G . 1972. The intestinal flora of the chicken in the period 2 to 6 weeks of age, with particular reference to the anaerobic bacteria. Br. Poult. Sci. 13: 311-326. Belley A,, Keller K., Goettke M. and Chadee K. 1999. Intestinal mucins in colonization and host defense against pathogens. Amer. J. Trop. Med. Hyg. 60: 10-15. Blaut M., Collins M . D., Welling G. W., Dore J., van Loo J., and d e Vos W. 2002. Molecular biological methods for studying the gut microbiota: the EU human gut flora project. Brit. J. Nutr. 87 Suppl 2: 203-211. Blikslager A. T. a n d Roberts M . C. 1997. Mechanisms of intestinal mucosal repair. J A V M A 211: 1437-1441. Boismenu R. and Havran W. L. 1997. An innate view of gamma-delta T cells. Cur. Opin. Immunol. 9: 57-63. Bosi P. 2000. Modulation of immune response and barrier function in the piglet gut by dietary means. Asian-Austr. J. Anin. Sci. 13: 278-293. Brandtzaeg P., Halstensen T. S., Kett K., Krajci P., e t al. 1989. Immunobiology and immunopathology of human gut mucosa: humoral immunity and intraepithelial lymphocytes. Gastroen ter. 97: 1562-1584. Brown T. A., Russell M. W., and Mestecky J. 1984. Elimination of intestinally absorbed antigen into the bile by IgA. J. Immunol. 132: 780-2.
Gut associated i m m u n e system
273
Buddington K. K., Donahoo J. B., and Buddington R. K. 2002a. Dietary oligofructose and inulin protect mice from enteric and systemic pathogens and tumor inducers. J. Nutr. 132: 472-477. Buddington R. K., Kelly-Quagliana K., Buddington K. K., and Kimura Y. 2002b. Nondigestible oligosaccharides and defense functions: Lessons learned from animal models. Brit. J. Nutr. 87: S231-S239. Butler J. E. 1981. The Ruminant Immune System. Plenum Press, New York, NY. Coates M. E. 1980. The gut-microflora and growth. In: Growth in Animals T. L. J. Lawrence (ed.). Butterworths, London, pp. 179-191. Coates M. E. and Gustafsson B. E. 1984. The Germ-free Animal in Biomedical Research, Laboratory Animals Ltd., London. Cooper E. L. 1976. Comparative Immunology. Prentice-Hall. Englewood Cliffs, N.J.: Cornes J. S. 1965. Peyer's patches in the human gut. Proc. R. Soc. Med. 58: 716-722. Cummings J. H. and Macfarlane G. T. 1997. Role of intestinal bacteria in nutrient metabolism. Clin. Nutr. 16: 3-11. Cunningham-Rundles S. and Lin D. H. 1998. Nutrition and the immune system of the gut. Nutrition 14: 573-579. Cuvelier C., Demetter P., Mielants H., Veys E. M., and De Vos M. 2001. Interpretation of ileal biopsies: Morphological features in normal and diseased mucosa. Histopath. 38: 1-12. de Bolos C., Real F. X. and Lopez-Ferrer A. 2001. Regulation of mucin and glycoconjugate expression: From normal epithelium to gastric tumors. Front. Biosci. 6: 1256-1263. Defresne M. P. 1998. The Mouse Model. In: Handbook of Vertebrate immunology, P, Pastoret P. Griebel H. Bazin and A. Govaerts (eds.). Academic Press, San Diego, CA, pp. 563-618. Deprez P., Deroose P., Van den Hende C., Muylle E., and Oyaert W. 1987. Liquid versus dry feeding in weaned piglets: The influence on small intestinal morphology. J. Vet. Med. Ser. B 34: 254-259. Drasar B. S. and Barrow P. A. 1985. Intestinal Microbiology. Van Nostrand Reinhold, Wokingham, Berkshire, (UK). Duchmann R., Schmitt E., Knolle P., Meyer zum Buschenfelde K. H., and Neurath M. 1996. Tolerance towards resident intestinal flora in mice is abrogated in experimental colitis and restored by treatment with interleukin-10 or antibodies to interleukin-12. Eur. J. Immunol. 26: 934-938. Duffy L. C. 2000. Interactions mediating bacterial translocation in the immature intestine. 1. Nutr. 130: 432s-436s. Durmic Z., Pethick D. W., Pluske J. R., and Hampson D. J. 1998. Changes in bacterial populations in the colon of pigs fed different sources of dietary fibre, and the development of swine dysentery after experimental infection. J. Appl. Microbial 85: 574-582. Elson C. O., Cong Y., Iqbal N., and Weaver C. T. 2001. Immuno-bacterial homeostasis in the gut: new insights into an old enigma. Semin. Immunol. 13: 187-194. Ewing W. N. and Cole D. J. A. 1994. The Living Gut : A n Introduction to Micro-organisms in Nutrition. Dungannon, Co. Tyrone, Context, N. Ireland. Feil W., Lacy E. R., Wong Y. M., Burger D., 1989. Rapid epithelial restitution of human and rabbit colonic mucosa. Gastroenterology 97: 685-701. Felsburg P. J. 1998. Immunology of the Dog. In: Handbook of Vertebrate Immunology, P. Pastoret, P. Griebel, H. Bazin and A. Govaerts (eds.). Academic Press, San Diego, CA, pp. 261-289. Ferreira R. C., Forsyth L. E., Richman P. I., Wells C., Spencer J., and MacDonald T. T. 1990. Changes in the rate of crypt epithelial cell proliferation and mucosal morphology induced by a T-cell-mediated response in human small intestine. Gastroenterology 98: 1255-1263. Ferrick D. A., Braun R. K., Lepper H. D., and Schrenzel M. D. 1996. Gamma-delta T cells in bacterial infections. Res. Immunol. 147: 532-541.
274
Physiological and ecological adaptations t o feeding in vertebrates
Field C. J., Johnson I. R., and Schley P. D. 2002.Nutrients and their role in host resistance to infection. 1. Leukoc. Biol. 71: 16-32. Gardner M. L. 1988. Gastrointestinal absorption of intact proteins. Ann. Rev. Nutr. 8: 329-350. Garside P. and Mowat A. M. 2001. Oral tolerance. Sem. Immunol. 13: 177-185. German A. J., Hall E. J., and Day M. J. 1999.Analysis of leucocyte subsets in the canine intestine. J. Comp. Pathol. 120: 129-145. Gil A. and Rueda R. 2000. Modulation of intestinal microflora by specific dietary components. Microb. Ecol. Health Dis. 12: 31-39. Goddeeris B. 1998. Immunology of Cattle. In: Handbook of Vertebrate Immunology, P. Pastoret, P. Griebel, H. Bazin, and A. Govaerts (eds.). Academic Press, San Diego, CA, pp. 439-484. Griebel P. J. 1998.Sheep Immunology and Goat Peculiarities. In: Handbook of Vertebrate Immunology P. Pastoret, P. Griebel, H. Bazin, and A. Govaerts (eds.). Academic Press, San Diego, CA, pp. 485-554. Hannant D. 1998. Immunology of Horses and Donkeys. In: Handbook of Vertebrate Immunology P. Pastoret, P. Griebel, H. Bazin, and A. Govaerts (eds.). Academic Press, San Diego, CA, pp. 343-372. Heitman D. W., Pool T. B. and Cameron I. L. 1980.Changes in proliferation and surface morphology in the rat ileum in response to total parenteral nutrition. J. Anat. 130: 603-615. Hemsley S. W., Canfield P. J., and Husband A. J. 1996a.The distribution of organised lymphoid tissue in the alimentary tracts of koalas (Phaseolarctoscinereus) and possums (Trichosurus vulpecula and Pseudocheirus peregrinus). J. Anat. 188: 269-78. Hemsley S. W., Canfield P. J . and Husband A. J. 1996b. Histological and immunohistological investigation of alimentary tract lymphoid tissue in the koala (Phascolarctos cinereus), brushtail possum (Trichosurus vulpecula) and ringtail possum (Pseudocheirus peregrinus). J. Anat. 188: 279-288. Honjo K., Hagiwara T., Itoh K., Takahashi E. and Hirota Y. 1993. Immunohistochemical analysis of tissue distribution of B and T cells in germfree and conventional chickens. I. Vet. Med. Sci. 55: 1031-1034. Hooper L. V., Wong M. H., Thelin A., Hansson L., Falk P. G., and Gordon J. 1. 2001. Molecular analysis of commensal host-microbial relationships in the intestine. Science 291: 881-884. Husband A. J. 2002. Mucosal memory: Maintenance and recruitment. Vet. Immunol. Immunopath. 87: 131-136. Husby S. (2000).Normal immune responses to ingested foods. J. Pediatr. Gastroenterol. Nutr. 30: S13-Sl9. Jeurissen S. H., Wagenaar F. and Janse E. M. 1999. Further characterization of M cells in gut-associated lymphoid tissues of the chicken. Poult. Sci. 78, 965-972. Johnson I. T. 2001.New food components and gastrointestinal health. Proc. Nutr. Soc. 60: 481-488. Jones P. H., Roe J. M., and Miller B. G. 2001. Effects of stressors on immune parameters and on the faecal shedding of enterotoxigenic Escherichia coli in piglets following experimental inoculation. Res. Vet. Sci. 70: 9-17. Josefsen T. D. and Landsverk T. 1996. T cell subsets and Langerhans cells in the forestomach mucosa of adult sheep and sheep foetuses. Vet. Immunol. lmmunopath. 51: 101-111. Josefsen T. D. and Landsverk T. 1997. Increase in gammadelta T cells in the ruminal mucosa of reindeer calves (Rangifer tarandus tarandus L.) induced by baled grass silage. Vet. Immunol. Immunopath. 60: 197-202. Kelly D., Begbie R. and King T. P. 1994.Nutritional influences on interactions between bacteria and the small intestinal mucosa. Nutr. Res. Rev. 7: 233-257. Kitagawa H., Shiraishi S., Imagawa T., and Uehara M. 2000. Ultrastructural characteristics and lectin-binding properties of M cells in the follicle-associated epithelium of chicken caecal tonsils. J. Anat. 197: 607-616.
Gut associated immune system
275
Klasing K. C. 1998. Comparative Avian Nutrition. USA: Cab Int. Wallingford, Oxon, UK; Klasing K. C. 1999. Interactions between nutrition and immunity. In: Nutrition and Immunology., M. Gershwin (ed.). Humana Press, pp. 363-373. Kolbjornsen O., Press C. M., Moore P. F., and Landsverk T. 1994. Lymphoid follicles in the gastric mucosa of dogs. Distribution and lymphocyte phenotypes. Vet. Immunol. Immunopath. 40: 299-312. Komazawa, S., Iki, H., Ohmura, M., Tsutsui, S., and Fujiwara, K. (1991). Comparative studies on distribution and fine morphology of the intestinal Peyerfs patches in Mongolian gerbils (Meriones unguiculatus) and mice. J. Vet. Med. Sci. 53: 899-904. Kroese F. G . 1998. Immunology of the Rat. In: Handbook of vertebrate Immunology. P. Pastoret, P. Griebel, H. Bazin, and A. Govaerts (eds.). Acad. Press, San Diego, CAI pp. 137-222. Lambolez F. and Rocha B. 2001. Immunology. A molecular gut reaction. Science 294: 1848-1489. Landsverk T., Halleraker M., Aleksandersen M., McClure S., Hein W. and Nicander L. 1991. The intestinal habitat for organized lymphoid tissues in ruminants; comparative aspects of structure, function and development. Vet. Immunol. Immunopathol. 28: 1-16. Lee A. 1999. Helicobacter pylori: opportunistic member of the normal microflora or agent of communicable disease? In Medical Importance of the Normal Microfora, G. W. Tannock (ed.). Kluwer, Dordrecht, Pull., Netherlands, pp. 295-311. Liebler E. M., Pohlenz J. F., and Woode G . N. 1988. Gut-associated lymphoid tissue in the large intestine of calves. I. Distribution and histology. Vet. Pathol. 25: 503-508. Lien K. A., Sauer W. C., and He J. M. 2001. Dietary influences on the secretion into and degradation of mucin in the digestive tract of monogastric animals and humans. J. Anim. Feed Sci. 10: 223-245. Luckey T. D. 1972. Introduction to intestinal microecology. Amer. J . Clin. Nutr. 25: 1292-1294. Lutz H. 1998. Immunology of the Cat. In: Handbook of Vertebrate Immunology, P. Pastoret, P. Griebel, H. Bazin, and A. Govaerts (eds.). Acad. Press, San Diego, CA, pp. 290-336. MacFarland S., Cummings J. H. and MacFarlane G . T. 1999. Bacterial colonisation of surfaces in the large intestine. In: Colonic Microbiota, G. R. Gibson and M. B. Roberfoid eds.). Kluwer, Publ. Dordrecht, Netherlands, pp. 71-87. Mackie R. I., Sghir A. and Gaskins H. R. 1999. Developmental microbial ecology of the neonatal gastrointestinal tract. Amer. J. Clin. Nutr. 69: 1035-1045. Mage R. G . 1998. Immunology of Lagomorphs. In: Handbook of Vertebrate Immunology, P. Pastoret, P. Griebel, H. Bazin, and A. Govaerts (eds.). Acad. Press, San Diego, CAI pp. 223-260. Manning M. J. and Turner R. J. 1976. Comparative immunobiology. Wiley & Sons, New York, NY. Marchalonis J. J. 1976. Comparative immunology. Wiley & Sons, New York, NY. Matzinger P. 2002. The danger model: A renewed sense of self. Science 296: 301-305. Mayer L. 2000. Mucosal immunity and gastrointestinal antigen processing. J. Pediatr. Gastroenterol. Nu tr. 30: 4-12. Morrison W. I. 1986. The Ruminant Immune System in Health and Disease. Cambridge Univ. Press, New York. Morton J. E. 1979. Guts: the form and function of the digestive system. Univ. Park Press, Baltimore, MA. Nagler-Anderson C. 2000. Tolerance and immunity in the intestinal immune system. Crit. Rev. Immunol. 20: 103-120. Nagler-Anderson C. 2001. Man the barrier! Strategic defences in the intestinal mucosa. Nut. Rev. Immunol. 1: 59-67. Neish A. S. 2002. The gut microflora and intestinal epithelial cells: A continuing dialogue. Microbes Infect. 4: 309-317.
276
Physiological and ecological adaptations t o feeding in vertebrates
Nusrat A., Sitaraman S. V., and Neish A. 2001. Interaction of bacteria and bacterial toxins with intestinal epithelial cells. Curr. Gastroenferol Rept. 3: 392-398. Nusrat A., Turner J. R., and Madara J. L. 2000. Molecular physiology and pathophysiology of tight junctions. IV. Regulation of tight junctions by extracellular stimuli: nutrients, cytokines, and immune cells. Amer. J. Physiol. 279: G851-G857. Nyachoti C. M., De Lange C. F. M., McBride B. W., and Schulze H. 1997. Significance of endogenous gut nitrogen losses in the nutrition of growing pigs: A review. Can. J. Anen. Sci. 77: 149-163. Pabst R. and Rothkoetter H. J. 1999. Postnatal development of lymphocyte subsets in different compartments of the small intestine of piglets. Vet. lrnmunol. Immunopath. 72: 167-173. Par A. 2000. Gastrointestinal tract as a part of immune defence. Acta Physiologica Hungar. 87: 291-304. Parsons K. R., Bland A. P. and Hall G. A. 1991. Follicle associated epithelium of the gut associated lymphoid tissue of cattle. Vet. Pathol. 28: 22-29. Pastoret P.-P. 1998. Handbook of Vertebrate Immunology. Acad. Press, San Diego, CA. Pescovitz M. D. 1998. Immunology of the Pig. In: Handbook of Vertebrate Immunology, P. Pastoret, P. Griebel, H. Bazin, and A. Govaerts (eds.). Acad. Press, San Diego, CAI pp. 373-420. Pluske J. R., Siba P. M., Pethick D. W., Durmic Z., Mullan B. P., and Hampson D. J. 1996. The incidence of swine dysentery in pigs can be reduced by feeding diets that limit the amount of fermentable substrate entering the large intestine. I. Nutr. 126: 2920-2933. Pluske J . R., Durmic Z., Pethick D. W., Mullan B. P., and Hampson D . J. 1998. Confirmation of the role of rapidly fermentable carbohydrates in the expression of swine dysentery in pigs after experimental infection. 1. Nutr. 128: 1737-1744. Poskitt D. C., Duffey K., Barnett J., Kimpton W. G., and Muller H. K. 1984. The gutassociated lymphoid system of two species of Australian marsupial mice, Antechinus swainsonii and Antechinus stuartii. Distribution, frequency and structure of Peyer's patches and lymphoid follicles in the small and large intestine. Austr. J. Exper. Biol. Med. Sci. 62: 81-88. Radberg K., Biernat M., Linderoth A., Zabielski R., Pierzynowski S. G., and Westrom B. R. 2001. Enteral exposure to crude red kidney bean lectin induces maturation of the gut in suckling pigs. J. Anim. Sci. 79: 2669-2678. Rothkotter H. J., Ulbrich H. and Pabst R. 1991. The postnatal development of gut lamina propria lymphocytes: number, proliferation, and T and B cell subsets in conventional and germ-free pigs. Pediatr. Res. 29: 237-242. Rothkoetter H. J., Pabst R. and Bailey M. 1999. Lymphocyte migration in the intestinal mucosa: Entry, transit and emigration of lymphoid cells and the influence of antigen. Vet. Immunol. Immunopath. 72: 157-165. Rowland I. R. 1999. Toxicological implications of the normal microflora. In: Medical Importance of the Normal Microflora. G. W . Tannock (ed.). Kluwer, Publ., Dordrecht, netherlands, pp. 295-311. Rueda R. and Gil A. 2000. Influence of dietary compounds on intestinal immunity. Microbial Ecol. Health Dis. 12: 146-156. Sato S., Ogimoto K. and Nakai Y. 1990. Comparative distribution of immunoglobulincontaining cells in stomach, intestine and associated lymph nodes of cattle. Jap. J. Vet. Sci. 52: 63-70. Savage D. C. 1977. Microbial ecology of the gastrointestinal tract. Ann. Rev. Microbial. 31: 107-133. Schiffrin E. J. and Blum S. 2002. Interactions between the microbiota and the intestinal mucosa. ELK J. Clin. Nutr. 56: 60-64. Schummer A., Nickel R. and Sack W. 0. 1979. The viscera of the domestic mammals. Springer-Verlag, New York.
Gut associated immune system
277
Simmons C. P., Clare S. and Dougan G. 2001. Understanding mucosal responsiveness: lessons from enteric bacterial pathogens. Semin. Irnmunol. 13: 201-209. Sjolander A., Magnusson K. E. and Latkovic S. 1986. Morphological changes of rat small intestine after short-time exposure to concanavalin A or wheat germ agglutinin. Cell Struct. Funct. 11: 285-293. Stepankova R., Kovaru F. and Kruml J. 1980. Lymphatic tissue of the intestinal tract of germfree and conventional rabbits. Folia Microbial. 25: 491495. Stevens C. E. and Hume I. D. 1995. Comparative Physiology of the Vertebrate Digestive System. Cambridge Univ. Press, New York. Strocchi A. and Levitt M. D. 1991. A reappraisal of the magnitude and implications of the intestinal unstirred layer. Gastroenterology 101: 843-847. Suzuki H., Jeong K. I., Itoh K. and Doi K. 2002. Regional variations in the distributions of small intestinal intraepithelial lymphocytes in germ-free and specific pathogen-free mice. Exper. Molec. Pathol. 72: 230-235. Tannock G. W. 1999. Analysis of the intestinal microflora: A renaissance. Antonie van Leeuwenhoek 76: 265-278. Tomar B. S. 2001. Intestinal infections. Indian J. Ped. 68: 8-19. Turner R. J. 1994. Immunology : a comparative approach. J . Wiley, & Sons, Chichester, NY. Van der Sluis M., Makkink M. K., Sutmuller M., Buller H. A., 2002. MUC2 mucin knockout mice are more susceptible to dextran sodium sulfate-induced colitis. Gastroenterology 122: 386-392. Van Soest P. J. 1994. Nutritional ecology of the ruminant. Comstock Publ., Ithaca, NY. Vervelde L. and Jeurissen S. H. 1993. Postnatal development of intra-epithelial leukocytes in the chicken digestive tract: phenotypical characterization in situ. Cell Tissue Res. 274: 295-301. Waly N., Gruffydd-Jones T. J., Stokes C. R., and Day M. J. 2001. The distribution of leucocyte subsets in the small intestine of healthy cats. J. Comp. Pathol. 124: 172-182. Yamasaki F. 1971. Comparative anatomical studies on the lymphoid apparata at the ileocaecal region in Lagomorpha. Okajimas Folia Anat. Jpn. 47: 407425. Yang D., Biragyn A., Kwak L. W., and Oppenheim J. J. 2002. Mammalian defensins in immunity: More than just microbicidal. Trends Immunol. 23: 291-296. Yang J. S., Jung H. J., Xuan Z. N., Kim J. H., 2001a. Effects of feeding and processing methods of diets on performance, morphological changes in the small intestine and nutrient digestibility in growing-finishing pigs. Asian-Austr. J. Anim. Sci. 14: 1450-1459. Yang J. S., Lee J. H., KO T. G., Kim, T. B., et al. 2001b. Effects of wet feeding of processed diets on performance, morphological changes in the small intestine and nutrient digestibility in weaned pigs. Asian-Austr. J. Anem. Sci. 14: 1308-1315. Yun C. H., Lillehoj H. S. and Lillehoj E. P. 2000. Intestinal immune responses to coccidiosis. Develop. Comp. Immunol. 24: 303-324. Zapata A. G. and Cooper E. L. 1990. The Immune System: Comparative Histophysiology. Jiwiley & Sons, Chichester, NY.
Effects of Digestion on the Respiratory and Cardiovascular Physiology of Amphibians and Reptiles Tobias Wangl, Johnnie B. Andersenl and JamesW. Hicks2 ' University Aarhus, Department of Zoophysiology, Aarhus, Denmark University California, Department of Ecology and Evolutionary Biology, Irvine, CA, USA
SYNOPSIS Many carnivorous amphibians and reptiles ingest very large meals at infrequent intervals. Within several hours following ingestion of these meals, a variety of physiological and metabolic processes occur, including secretion of digestive juices to the stomach and intestines, nutrient transport, protein synthesis and, at least in some cases, marked and rapid growth of the gastrointestinal tract. The digestive process results in two major physiological challenges that are metabolic and acid-base in origin. Digestion leads to elevated oxygen consumption rates (specific dynamic action or SDA); rates that can approach or even exceed the values achieved during maximal activity. However, while maximal activity can only be sustained for minutes, the postprandial increases in oxygen demands can last for many days. In addition to the metabolic challenges, digestion is also associated with a net acid secretion into the stomach lumen that causes an increase in plasma HCO-, concentration (the "alkaline tide"). The physiological challenges associated with digestion in ectotherms provide novel models for investigatingthe "design" principlesand physiologicalmechanisms that underlie the cardiopulmonary responses to elevated oxygen consumption, and the mechanisms of acid-base regulation. Here, we review the cardiorespiratory consequences of SDA in amphibians and reptiles and describe how studies on postprandial exercise may enhance further understanding of the physiological and structural constraints that limit the ability of reptiles and amphibians to sustain high metabolic rates.
280
Physiological and ecological adaptations t o feeding in vertebrates
Intermittent feeding, wherein prolonged periods of fasting are interspersed with occasional ingestion of very large meals, is a characteristic of many carnivorous ectothermic vertebrates. For example, field studies show that some snakes feed only a few times per year and often ingest prey items more than 25% their own body weight (Greene,1983,1997).In some extreme cases the size of the ingested prey may even exceed the weight of the predator. The period following ingestion is characterized by a variety of physiological and metabolic processes including secretion of digestive juices to the stomach and intestine, nutrient transport, protein synthesis and, at least in some cases, by a marked and rapid hypertrophy of the gastrointestinal (GI)system (Secor et al., 1994; Secor and Diamond, 1995; Secor, 2001; Starck and Beese, 2001; Starck, 2004).Initiation and up-regulation of these digestive processes result in major metabolic and acid-base-related physiological challenges. In reptiles and amphibians, the metabolic demands associated with digestion lead to elevated oxygen consumption rates, i.e. rates that can approach or even exceed the values achieved during maximal exercise. However, in contrast to exercise, which usually lasts only minutes, the postprandial increments in oxygen demand can last for several days. In addition, ingestion of a large meal requires secretionof large amounts of H+ion into the stomach to initiate the digestive process. The resultant reciprocal movement of HCO; into the plasma, induses a large alkaline tide, which poses a significant acid-base challenge. Thus, digestion in ectothermic vertebrates places simultaneous physiological demands on the GI and cardiopulmonary (CP)systems. The postprandial metabolic and acid-base challenges that occur in ectotherms provide novel models for investigating the "design" principles and physiological mechanisms that underlie the CP responses and mechanisms of acid-base regulation. This review provides a broad summary of the currently existing data on CP responses to feeding in amphibians and reptiles. It is important to note that only a few species have been studied so far. Thus, the general patterns that appear to exist with regard to the CP responses to feeding may be challenged when more species are studied. The cardiorespiratory consequences of digestion were previously reviewed by Wang et al. (2001a),but a number of studies have been published within the past few years that allow for a more comprehensive description.
CHANGES I N METABOLISM DURING DIGESTION The observation that metabolic rate, typically measured as oxygen uptake or heat production, increases after feeding dates back to the 17thcentury when Lavoisier and coworkers started to describe metabolism in humans and other
Effects of digestion
281
animals (e.g. Fulton and Wilson, 1930).The increased rate of oxygen uptake (vo,)following feeding is commonly referred to as the "specific dynamic action of food", abbreviated SDA (Rubner, 1902; see Kleiber, 1961),postprandial calorigenesis (McCue et al., 2004a, b), or the heat increment of feeding (Kleiber, 1961).In its strictest sense, SDA includes only the metabolic costs involved with digestion, absorption and utilization of food, whereas the apparent SDA, measured as the change in metabolic rate throughout the postprandial period, also includes other costs associated with feeding, such as prey handling (e.g. Cruz-Neto et al., 1999,2001)and GI hypertrophy (e.g. Secor, 2001). The apparent SDA has been described for a number of invertebrates and vertebrates (see references in Jobling, 1981;Burggren et al., 1993;Hawkins et al., 1997; Hicks et al., 2000; Whiteley et al., 2001; Andrade et al., 2004b). The largest factorial changes in metabolism following feeding occur in reptiles and the response is particularly pronounced in intermittently feeding snakes. This was first established by Benedict (1932),who showed that heat production and gas exchange of large constricting snakes may increase ten times following a meal, and that this high metabolic rate was sustained for many days. Subsequent and much more detailed studies confirmed these observations and it was reported that oxygen uptake of digesting pythons may be more than forty times higher than fasting values (Secor and Diamond, 1995). Although the large factorial increment is related, at least in part, directly to the low metabolic rate of fasting, resting snakes, it is impressive that oxygen uptake during SDA may exceed 20 ml min-I kg-]for an extended period of time, e.g. 24 h (Secor and Diamond, 1995).This value is similar to the resting metabolic rates of equivalent size mammals and the maximum oxygen uptake measured in monitor lizards running on a treadmill. These lizards are often considered "the athletes" among reptiles, but are capable of sustaining high rates of oxygen consumption for only relatively short periods of time (between 15 and 60 min; e.g. Mitchell et al., 1981; Wang et al., 1997a). Oxygen uptakes at rest, digestion, and exercise of several species of amphibians and reptiles are presented in Fig. 11.1. In the snakes Python and Crotalus, maximal oxygen uptake during digestion exceeds oxygen uptake during muscular exercise. In contrast to exercise, which normally involves a large proportion of anaerobic metabolism (e.g.Bennett, 1994;Gleeson, 1996; Andersen and Wang, 2003), the plasma lactate levels remain low during SDA and the increased metabolism met, therefore, entirely by aerobic metabolism (Overgaard et al., 1999; Busk et al., 2000a, b; Andersen and Wang, 2003). Although the factorial SDA response is indeed remarkable for snakes, the maximum rates of oxygen uptake during SDA are nevertheless not larger than the resting oxygen uptake for similar-size mammals (note that the yaxis for the rat in Fig. 11.1is different). Many reptiles have rather low metabolic rates and remain inactive over long periods, so little energy is spent on locomotion. Thus, the energetic costs
282
Physiological and ecological adaptations t o feeding in vertebrates Rattus
Rana (22OC) 25
100 1
Bufo (25OC) 25 1
1,
Q,
Y
(21
Python (30%)
Crotalus (30%)
Alligator (30%)
Varanus (35OC)
Fig. 11.1. Comparison of oxygen uptake at rest, following feeding and during exercise for selected species discussed in this review. Experiments on the different species have been performed at the preferred body temperatures of the respective animals, and are based on different food rations, indicated above the grey bars, relative to body weight. Open bars indicate oxygen uptake for resting animals, grey bars indicate oxygen uptake during digestion and filled bars represent oxygen uptake during muscular exercise. Note that the y-axis for Rattus is compressed compared to the other species. Sources: Rattus (Gonzalez et al., 1998); Bufo (Andersen and Wang, 2003); Rana (Busk et al., 2000a; Hillman, 1987); Python (Secor and Diamond, 1997; Secor et al., 2000); Crotalus (Andrade et al., 1997); Alligator (Busk et al., 2000b; Farmer and Carrier, 2000); Varanus (Hicks et al., 2000; Mitchell et al., 1981).
associated with digestion may constitute a very significant portion of the overall energy budget. When estimating the annual energy budget of a population of cottonmouth,McCue and Lillywhite (2002)concluded that approximately one-third of the energy consumed is used for digestion (Fig. 11.2). Laboratory studies support this estimate. Thus, when expressing the total amount of energy used for digestion relative to the energy content of the food, normally called the SDA coefficient, it seems that most species of snakes spend between 15 and 30% of a meal on its digestion (e.g. Overgaard et al., 2002; Secor, 2001; Toledo et al., 2003; Zaidan and Beaupre, 2003). The causes for the large metabolic response that characterize digesting snakes and other reptiles have remained elusive although numerous studies have addressed this issue over the past decade. Because digestion correlates with a pronounced increment in intestinal mass and an up-regulation of the capacity of the intestinalbrush-border transport proteins, it was originally speculated that up-regulation of GI organs was expensive and accounted for a significant portion of the SDA response (e.g. Secor and
283
Effects of digestion
Jan
Feb
Mar
Apr
May Jun
Jul
Aug
Sep
Oct
Nov
Dec
Mean Temperature "C / Month
Fig. 11.2. An estimation of the annual energy budget for a nonreproductive cottonmouth snake (Agkistrodon piscivorus) with a body mass of 500g). This particular population of snakes inhabits a peninsula in Northern Florida where they can feed on fish that are dropped by colonial nesting birds, which may explain their low energy allocation activity (McCue and Lillywhite, 2002).
Diamond, 1998).More recent studies, however, indicate that the increased intestinal mass following feeding occurs by simple swelling of the individual enterocytes (Starckand Beese, 2001),which seems to take place at relatively low energetic cost (Overgaard et al., 2002). It was recently suggested that gastric acid secretion may account for up to 50% of the SDA response in Python (Secor, 2003) but inhibition of gastric acid secretion does not reduce the metabolic response to digestion (Andrade et al., 2004a). The role of protein synthesis should, however, not be neglected. Already during Benedict's (1932)early studies, it was found that fat only elicited very small metabolic changes compared to the changes elicited by ingestion of protein. However, while a number of contemporary studies reached conflicting conclusions regarding the role of protein synthesis and protein degradation (e.g.Lusk, 1931;Wilhelmj, 1935;Borsook, 1936;see Kleiber, 1961),recent and very comprehensive studies by McCue et al., (2004) confirmed the pivotal role of dietary protein. These observations are consistent with the general view that increased protein synthesis makes up for a large proportion of the SDA in most animals (Coulsonand Hernandez, 1979;Houlihan, 1991;Brown and Cameron, 1991a, b).
284
Physiological and ecological adaptations t o feeding in vertebrates
EFFECTS OF DIGESTION ON ACID-BASE STATUS In contrast to exercise, which is most often associated with acidosis, the postprandial period is characterized by alkalization of the plasma. These changes may be particularly critical during digeztion when metabolic demands are high. Thus, if transferred to the intracellular space, alkalinization may interfere with cellular metabolism through the direct effects of pH on enzyme activities, and if transferred to the red blood cells, it may increase blood oxygen affinity.
Alkaline tide and its causes The act of feeding and the presence of food in the stomach lead to secretion of acid into the stomach lumen (e.g. Hersey and Sachs, 1995; Niv and Fraser, 2002), affecting the acid-base status of extra- and intracellular body fluids. Gastric acid secretion is accomplished by the highly specialized parietal cell, illustrated in Fig. 11. 3, which possesses an apical H+-K+-ATPare that exchanges cytosolic H+for lumenal K+(e.g. Hersey and Sachs, 1995). K+, however, diffuses back into the stomach lumen in exchange for C1-, so the overall result becomes a net transfer of HC1 into the stomach lumen. As a consequence of the associated increase in strong ion difference (SID; the millimolar differences between strong cations and anions; Stewart, 1983), HCO, increases in the parietal cells and this rise is transferred to plasma and other body fluids. As a result, there is normally a tight correlation between reduction in plasma C1-concentration and increment in plasma HCO; (Fig. 11.3B).The increased strong ion difference and rise of plasma HCOi concentrationis termed the "alkaline tide" but, as discussed in greater detail below, does not necessarily involve an increase in arterial pH. Digestion is also associated with a base transfer to the intestine by pancreatic secretion, which serves to restore a more neutral pH of the chyme as it leaves the stomach. Thus, in animals that eat frequently and hence ingestion and passage of food between the stomach and the intestine are relatively continuous, the simultaneous secretion of acid in the stomach and gastric and pancreatic base excretion cancel each other out; so hence the acid-base status of the body is not affected. So, it is primarily animals that eat infrequently, which experience large alkaline tides, consequent to a temporal mismatch between gastric acid secretion and pancreatic base secretion. The extent to which acid-base balance is affected by digestion varies considerably among the different groups of vertebrates and among specieswithin a given group. Some of these differences are likely to be related to differences in meal size and frequency of feeding but other factors may contribute. In mammals, which generally eat relatively small meals, the alkaline tide is relatively small and plasma HCO; concentration rarely increases by more than a few mrnol (Rune, 1965;Rune and Lassen, 1968; Langbroek et al., 1990; Johnson et al., 1995; Niv and Fraser, 2002). This is likely to represent the
Effects of digestion
285
Time (h) Fig. 11.3. Schematic drawing of the parietal cell and the active proton excretion to the stomach lumen causing that "alkaline tide" manifested as increased plasma HCO-, concentration (A). Changes in plasma C1- and HCO-, concentrations after feeding in Rana catesbeiana (B). The insert shows the relationship between the changes in the plasma concentrations (based on Busk et al., 2000a).
temporal overlap between gastric and pancreatic secretion, but a rapid and that the kidneys may compensate by increasingbase output in the urine (e.g. Vaziri et al., 1980). Compared to mammals, most reptiles and amphibians eat much larger meals relative to their own body mass and it is likely that these larger meals require more acid to be excreted. Furthermore, compared to mammals, food normally remains in the stomach much longer before entering the intestine. Most of the gastric acid secretion is likely to take place before pancreatic base
286
Physiological and ecological adaptations t o feeding in vertebrates
output is raised, therefore. In the past few years, the alkaline tide has been characterized for numerous species of reptiles (snakes, lizards, and crocodilians) and anuran amphibians (frogs and toads) that were equipped with chronic catheters, so blood samples could be taken with minimal disturbance (Overgaard et al., 1999;Busk et al., 2000a, b; Wang et al., 2001; Andersen and Wang, 2003).In all species, plasma HCO, concentrationrose within 612h and reached maximum at 24-48 h into the postprandial period. Plasma HCO-, concentration rarely increased by more than 15 mmol l-I and some species exhibit smaller changes. Intracellular acid-base following feeding has only been studied in Rana (Busk et al., 2000a). The responses observed in all species where blood samples have been obtained from chronically implanted cathetersare in striking contrast to the very large alkaline tide reported by Coulson et al. (1950) on alligators in which blood samples were obtained through cardiac puncture. Coulson et al. (1950) reported that plasma HCO-, concentration increased by as much as 70 mmol 1-l,and that blood pH increased by 0.4 units after ingestion of approximately 5% meat relative to the alligator body mass. Although it remains uncertain how stress associated with blood sampling should cause such a profound metabolic alkalosis, it is noteworthy that a subsequent study on the same species equipped with a chronically implanted catheter found that plasma [HCO-,I increased by approximately 13 mmol 1-l,and never exceeded 45 mmol l-'(Busk et al., 2000b). Arterial blood gases, at least to our knowledge, have not been determined during the postprandial period in fish. However, given that the physicochemical properties of water make blood PCO, relatively insensitive to ventilatory adjustments in aquatic animals, this subject deserves attention. It is likely that the large capacity for branchial ion exchange is fully sufficient to counteract any changes in strong ion difference of the plasma that occurs as a consequence of gastric acid secretion.In this context, it would be interesting to measure transbranchial ion exchange during the postprandial period and likewise interesting to study fish that inhabit ion-poor waters in which ionic regulation may be compromised.
Ventilatory Compensation of the Alkaline Tide Arterial acid-base status of arterial plasma from the frog Rana catesbeiana is presented as Davenport diagrams in Fig. 11.4; similar qualitative and temporal data have been obtained on other amphibians and reptiles (see Wang et al., 2001a, b, c).All species studied so far exhibited a rise in arterial PCO, (PaCO,) that dampened the increase in pH. Thus, in all air-breathing vertebrates, the postprandial period is characterized by a metabolic alkalosis the alkaline tide-which is, at least in part, compensated by respiratory acidosis. The rise in PaCO, is caused by a relative hypoventilationwhere the effective lung ventilation does not rise proportional to the increase in metabolic production of CO, (Glass et al., 1979; Hicks et al., 2000; Secor et al.,
Effects of digestion
287
2000). It is nevertheless important to note that overall ventilation does increase during the postprandial period. In Varanus, increased ventilation is predominantly accomplished through a doubling of tidal volume, whereas breathing frequency increases only slightly (Hicks et al., 2000). In Python, increased ventilation is caused by an up to sixfold increase in breathing frequency (Secor et al., 2000) while tidal volume actually decreases within the initial 24 h after feeding. The relative hypoventilationmay reflect overall changes in ventilatory control, or as originally speculated by Milsom (1988), it is possible that this reduction reflects a mechanical constraint imposed by the prey on the lungs and/or the respiratory muscles. The ventilatory regulation of PaCO, is more difficult to interpret in amphibians because a large proportion of the CO, is excreted over the skin (e.g.Feder and Burggren, 1985; Wang et al., 2001). Nevertheless, direct measurements on Bufo showed a reduction in VE/VCO, following peptone injection in the stomach (Wang et al., 1995), which is consistent with the postprandial increase of PaCO, in amphibians (Busk et al., 2000a; Andersen and Wang, 2003). In addition to the ventilatory compensation of arterial pH, the alkaline tide may be compensated by transepithelial fluxes of alkaline equivalents
Arterial pH
Fig. 11.4. Davenport diagram with calculated PCO, isoclines (grey lines), showing changes in plasma acid-base status in the frog Rann cntesbeinna before and during digestion of a rodent meal (approximate 10% of body weight). Numbers by the data points indicate the time (h) after feeding (Busk et al., 2000a).
288
Physiological and ecological adaptations t o feeding i n vertebrates
over the kidney. Although the role of the reptilian kidney in acid-base balance is not well known (reviewedby Jackson, 1986),the few existing studies indicate that the kidneys play only a minor role following acid-base disturbances during long-term hypercapnia or bicarbonate infusion (Silver and Jackson, 1985,1986; Glass and Heisler, 1986).Therefore, over the short term, as during the initial alkaline tide, it is not known whether the kidney regulates acid-basebalance as observed in mammals (Jackson,1986).In amphibians, the kidney, bladder, and skin serve as important sites for the transfer of acid-base equivalents (Toewsand Boutilier, 1986).Consistent with this view, the total transepithelial transfer of acid-base equivalents over kidney, bladder, and skin change from a net acid secretion during fasting to a net base secretion following feeding (Busk et al., 2000a). Although the data were not statistically significant, the change in base excretion may contribute to as much as 50% of the total base excess (Busk et al., 2000a). A number of studies have shown that the kidneys of reptiles and amphibians are relatively poor at compensatingfor metabolic acid-base disturbances. Thus, plasma bicarbonate remains elevated for days after direct infusion of bicarbonate into the blood stream (Jackson,1969; Andersen et al., 2003).It is not clear why amphibians and reptiles do not use transepithelial ion gas exchange (i.e. kidney and skin in the case of amphibians) to compensate for the alkaline tide. However, given the alkaline tide may be the most commonly occurring situation wherein animals experience a metabolic alkalosis, it is possible that their slow compensation of an experimentallyinduced metabolic alkalosis represents an adaptation that temporarily maintains excess bicarbonate, so it can be readily utilized for subsequent excretion by pancreatic secretion. While it seems plausible that the respiratory compensation of pH during digestion serves a homeostatic function by preventing disturbances of acidbase balance to protect enzyme function and metabolic processes, the underlying regulation of acid-basebalance is not well understood. The ventilatory regulation of acid-base status in reptiles is governed by central chemoreceptors located within the medulla oblongata and peripheral chemoreceptors located in the vasculature. The central chemoreceptors are separated from the blood by the blood brain barrier and are bathed in the poorly buffered cerebrospinal fluid. The blood brain barrier is rather impermeable to ions, while CO, crosses relatively easily, so that the central chemoreceptors become more responsive to PaCO, compared to arterial pH. It may therefore appear surprising that all animals regulated pH so tightly by allowing PaCO, to increase during the postprandial period. It could even be argued that the postprandialrise in PaCO, merely reflects an ineffective ventilatory compensation to the increased metabolic rate that fortuitously acts to regulate pHa. To study whether pHa or PCO, constitute the regulated variable during digestion, we have used pharmacological inhibition of gastric acid secretion in the snake Boa and the toad Bufo. Gastric acid secretion can be inhibited by
Effects of digestion
289
the specific proton-pump inhibitor omeprazole that binds to cysteine groups of the Ht, Kt ATPase and inhibits its ability to secrete acid (Fellenius et al., 1981; Sachs et al., 1995; Huang and Hunt, 2001). In both Bufo and Boa, omeprazole greatly diminished-the postprandial rise in plasma HCO;, and the postprandial rise in PaCO, was abolished in spite of an increased metabolism (Fig. 11.5;Andersen et al., 2003, Andrade et al., 2004a).These data strongly suggest that the relative hypoventilation normally observed during digestion represents an actual regulation of pHa. Nevertheless, future studies should investigate the specific roles of the various chemoreceptors that control ventilation during the elevated metabolic rate during digestion.
PARADOXICAL BEHAVIOR OF ARTERIAL BLOOD GASES DURING SDA
A consistent finding in postprandial toads, lizards, snakes, and alligators, is the respiratory acidosis that partially compensatesfor the postprandial metabolic alkalosis (Busk et al, 2000a, b; Overgaard et al., 1999; Hartzler et al., 2003). This relative hypoventilation occurs despite increases in tissue oxygen demands; demands that approach or even exceed the values measured during maximal activity. Although the relative hypoventilation compensates for the metabolic alkalosis, minimizing changes in blood pH, the respiratory response correspondingly leads to reductions in lung PO, that may impair the transfer of oxygen to the blood. Paradoxically, most studies reveal that the postprandial rise of PaCO, is accompanied by significant increases in arterial PO,, a pattern of bloodgases not predicted by standard respiratory gas equations (Fig. 11.6).This apparent paradox may result from the unique cardiac morphology of amphibians and reptiles, a morphology that allows for regulation of venous admixture, or cardiac shunt. The undivided ventricle of the amphibian and reptilian heart makes cardiac shunts possible (Shelton, 1985; Hicks, 1998). These are typically defined by their direction, either as right-to-left (R-L)or left-to-right (L-R).An RL shunt represents bypass of the pulmonary circulation and recirculation of systemic venous blood (oxygen poor) back into the systemic arterial circulation. The direction and magnitude of cardiac shunts is largely determined by the vascular resistance of the pulmonary and systemic circulations with the regulation of these vascular circuits resulting from changes in autonomic tone and the release of neurohumoral factors (Hicks, 1998;Wang et al., 2001~). Changes in autonomic tone are partially generated within the central nervous system but stimulation of pulmonary stretch receptors and vascular chemoreceptors may also be involved (Wang et al. 199%). The varying magnitude of cardiac shunts has a direct and pronounced effect on oxygen transport. In the presence of R-L shunt, oxygen poor venous blood bypasses the lungs and consequently cannot become oxygenated.
290
Physiological and ecological adaptations t o feeding in vertebrates Boa constrictor
Bufo marinus
0.2
0 i
untreated control omeprazole treated
.
Time after feeding (hours)
Fig. 11.5. Changes in acid-base status following feeding in control snakes and toads (closed symbols) and in animals that was treated with the specific proton-pump inhibitor omeprazole (open symbols). From Andersen and Wang (2003) and Andrade et al. (2004a).
Shunted systemic venous blood decreases arterial oxygen saturation, resulting in arterial hypoxemia and reduction in tissue oxygen delivery. In contrast, an L-R shunt represents recirculation of pulmonary venous blood (oxygen rich) into the pulmonary circulation. Thus regulation of cardiac shunts allows an alternative mechanism for changing blood gases independent of respiratory control, a level of blood gas regulation not possible in mammals and birds.
Effects of digestion Alligator mississippiensis
140
0 Lung PO,
120
Arterial PO, -
Arterial PCO,
0
20
40
60
80
Time (h) Fig. 11.6. The paradoxical behaviour of arterial blood and lung gases during digestion in the American alligator (Alligator mississippiensis) (data from Busk et al., 2000b). In addition to arterial PC02, this figure presents direct measurements of arterial PO2 and calculated values for lung PO2 that have been estimated on basis of gas exchange and arterial PC02 using the lung gas equation (see for example Overgaard et al., 1999).
Several studies in amphibians and reptiles strongly suggest that the average magnitude and direction of the cardiac shunts correlates with overall oxygen demands (Wang and Hicks, 1996,2002). When oxygen demands are relatively low, e.g. during resting conditions, arterial blood oxygen levels are generally lower than pulmonary venous levels, indicating a prevailing R-L shunt. In contrast, elevation of metabolic rate, e.g. activity and temperature, appears to diminish the amount of mixing within the ventricle,
292
Physiological and ecological adaptations t o feeding in vertebrates
directly increasing arterial oxygen levels (Hedrick et al., 1999; Wang et al., 2001~). The paradoxical behavior of arterial PO, and PCO, measured during SDA may be explained by changes in cardiac shunting, i.e. amphibians and reptiles reduce the size of the R-L shunt in response to a postprandial increase in oxygen demands. Given this capacity for using the cardiovascular system to regulate blood oxygen levels, these animals effectively respond to both the metabolic and acid-base related challenges associated with digestion. Thus the flexibility afforded by control of cardiac shunting allows these animals to control pH through a respiratory acidosis, while simultaneously raising blood oxygen levels to meet the higher oxygen demands. It should be noted that shunting patterns during digestion have not been quantified and remain a fruitful area for investigation.
CARDIOVASCULARADJUSTMENTS TO ELEVATED METABOLIC RATE The postprandial increase in metabolism and the subsequent transport of absorbed nutrients from the intestines requires augmented blood flows. Increased blood flow to the gastrointestinal system during digestion can be accomplished through redistribution from other organs but often an overall increase in total systemic blood flow is also required. In addition to increasing and redistributing systemic blood flow, changes in cardiac shunt patterns can also enhance the oxygen concentration of arterial blood.
Postprandial Intestinal Hyperaemia It is primarily the gastrointestinal and digestive organs that increase metabolic rate during digestion. The increased blood flow to these organs can thereforebe accomplished through a redistr:ibutionof blood flows from other organs. Redistribution of blood flows only occurs when the overall changes are small relative to total cardiac output, so that the reduction of blood flows to other organs is not injurious for their function. Postprandial up-regulation of intestinal blood flow has been demonstrated in fish, reptiles and mammals. In resting fish, blood flow to the GI tract constitutes 2040% of cardiac output, which in response to feeding increases by 60-20O0/0 (e.g. Axelsson et al., 1989; Axelsson and Fritsche, 1991; Axelsson et al., 2002). In mammals, resting is somewhat lower than in fish: 20-25% of (Chou, 1983). Postprandial intestinal hyperemia depends on the size and composition of the meal. In mammals gastrointestinal blood flow typically doubles for 2-3 hours after feeding (Mathesonet al., 2000).A feeding-induced redistribution of blood flows has also been shown in Crocodylus where blood flow in the celiac artery increases by 187% following feeding (Axelsson et al., 1991).
Effects of digestion
293
The increased gastrointestinal blood flow during digestion is potentially controlled by 1) neural mediators (sympathetic and parasympathetic tone and neuropeptides), 2) circulating humoral mediators (catecholamines,angiotensin 11, vasopressin, serotonin, bradykinin), 3) circulating paracrine and autocrine mediators (such as endothelin-1, NO, prostaglandins) and 4) metabolic vasodilators (PO,, PCO,, pH and other metabolites; Matheson et al., 2000). In fish and Crocodylus, there is a well-established a-adrenergic component to the humoral and/or nervous regulation of gastrointestinal blood flow, but other mechanisms certainly contribute to the prolonged increase in gastrointestinal blood flows (Axelssonet al., 2000; Axelsson et al., 1991). The additional components may include cholinergic, and nonadrenergic-noncholinergic (NANC) regulatory peptides. Immunohistochemical studies demonstrate that the GI organs of amphibians and reptiles contain several regulatory peptides and physiological studies show that many of these have hemodynamic effects (e.g.Jensen et al., 1991;Karila et al., 1995;Holmgren, 1995;Kagstrom et al., 1998; reviewed by Morris, 1989; Morris and Nilsson, 1994).Given the very large increases in overall blood flows and heart rate in amphibians and reptiles, especially in Python and Boa, there are most certainly alterations in the distribution of blood flows following feeding but experimental evidence is scarce.
Increased Heart Rate and Cardiac Output During Digestion The act of feeding often involves killing prey by crushing or constriction. Such behavior is often associated with high activity levels accompanied by profound increases in heart rate and blood flow. Examples of heart rate and other hemodynamic variables during eating are shown for the lizard Varanus exanthemata and the snake Boa constrictor in Fig. 11.7. While killing, food handling and ingestion are associated with maximal heart rates in both these species, the cardiovascular status normally returning to normal resting values within a few hours. Then, as digestive processes commence and metabolic rate rises, heart rate and blood flows increase more gradually and remain elevated. An elevated heart rate during digestion has been documented in amphibians and reptiles (Dumsday, 1990;Wang et al., 1995,2000; Hicks et al., 2000; Secor et al., 2000; Andersen and Wang, 2003; Zaar et al., 2004). Increased heart rate leads to an increased overall cardiac output, as stroke volume is either unaffected (Hicks et al., 2000) or increased (Secor et al., 2000). The postprandial increase in systemic blood flow often does not match the factorial increase in oxygen uptake, but the arterial 0, concentration remains constant (Hicks et al., 2000; Wang et al., 2001a; Andersen and Wang, 2003). Digestion must consequently be associated with increased oxygen extraction and reduced mixed venous oxygen levels, but venous blood gases during digestion for reptiles or amphibians have yet to be described. As noted above, a reduction in the prevailing R-L shunt would also increase oxygen extraction (Hicks and Wang, 1996). In mammals,
294
Physiological and ecological adaptations t o feeding in vertebrates
6
200
I
150 -
8
Python molurus
-
I00 -
m ",
strikes & constricts
swallowing
-
0 60 -
I I
I
I1
I
h 7
."
-&E
40 20
-
2 min
0 80 Varanus exanthematicus
h
v
'C .-
E -
60
Sw*prey
-
-Eg
40-
6
0-
20-,
*
---*
-,
Time
Fig. 11.7. Cardiovascular changes during prey capture, subduing, and ingestion in the Burmese python (Python molurus) and the Savannah monitor lizard (Varanus exanthematicus). In both cases, the animals had been instrumented for measurements of blood flows or blood pressures several days before they were presented with a food item. In both cases, heart rate reaches its maximal values which is similar to that recorded during muscular exercise (Wang et al., 2001; Hicks et al., 2000).
postprandial cardiac output is increased due to an increased heart rate and stroke volume (e.g.Fronek and Fronek, 1970; Kelbaek et al., 1987;Waaler et al., 1990) and the magnitude of postprandial tachycardia is related to the size and composition of the meal (Waaleret al., 1991).This is consistent with data from fish in which a marked postprandial increase in cardiac output occurs (e.g. Axelsson et al., 1989; Axelsson and Fritsche, 1991; Axelsson et al., 2002) The hearts of reptiles and amphibiansare innervated by both cholinergic (inhibitory) and adrenergic (excitatory) nerves that run together in the vagosympathetic trunk. Autonomic regulation of the heart was recently studied during rest, digestion, and exercise in Boa (Wang et al., 2001~).In Boa, release of vagal tone is responsible for the postprandial tachycardia, with the adrenergic tone remaining low during digestion (Wang et al., 2001~). Furthermore, heart rate following total autonomic blockade (i.e.after injection of propranolol and atropine) was significantly higher during digestion
Effects of digestion
295
than during rest, and it was suggested that this increase could be attributed to NANC factors acting directly on the heart (Wang et al., 2001~). Pointing to a similar mechanism in mammals, Kelbaek et al. (1987) reported an increased heart rate during digestion in humans that could not be blocked by atropine and that was not associated with increased levels of circulating catecholamines.
HEMATOLOGICAL RESPONSES AND CHANGES I N BLOOD OXYGEN AFFINITY Oxygen delivery can be increased by elevatingthe capacity for blood to transport oxygen and the affinity by which oxygen is bound to the blood may be altered to facilitate unloading in the tissues. In the amphibians Bufo and Rana, blood hemoglobin concentration and hematocrit increase by 30-60% after feeding (Wang et al., 1995, Busk et al., 2000a). Some of this response increase may be caused by contractionof the spleen. However, it is also likely to reflect a hemoconcentrationcaused by fluid shifts from the extracellular to the intracellular space, and/or from the vascular system to the lymph (see Pinder and Smits, 1993; Malvin et al., 1995).The pattern observed in amphibians is not general for all vertebrates and there is no increase in hematocrit followingfeeding in reptiles (Overgaard et al., 1999;Busk et al., 2000a, b) or humans, (e.g.Ehrly and Jung, 1973).Nevertheless, in dogs hematocrit increase as much as 10% followingfeeding and removal of the spleen virtually abolished this response (Kurata et al., 1993). Blood oxygen affinity is affected by acid-base status but the effects of digestion on blood oxygen binding properties have not been studied in much detail. If the alkaline tide is transmitted to the red cells through the anion exchanger, alkalosis will increase blood oxygen affinity through the Bohr effect;it has been suggested that respiratory pH compensation during the postprandial period acts, at least in part, to diminish the increase in blood oxygen affinity(Busk et al., 2000a). Crocodilian hemoglobin is unique in the ability of HCO, to bind, thereby, lowering oxygen affinity (Bauer and Jelkmann, 1977; Jensen et al., 1998).It has been suggested that this feature counteracts the increase in blood oxygen affinity associated with the alkaline tide to maintain constant blood oxygen affinity during SDA (Weber and White, 1986)and this notion was supported by determination of blood oxygen affinity from fasting and digesting alligators (Busk et al., 2000b).In other vertebrates in which the hemoglobin is not endowed with bicarbonatesensitive hemoglobin, it would be possible to maintain blood oxygen affinity by altering the levels of allosteric modulators, such as ATP and other organic phosphates, within the red cells. This possibility has only been studied in Python. In Python, wherein blood pH only increases very slightly during
296
Physiological and ecological adaptations t o feeding in vertebrates
digestion, Overgaard and Wang (2002)predicted that in vivo blood oxygen affinitywould increase slightly during digestion. This increase correlated with a decrease in organic phosphates within the red cells and cell swelling. This result is surprising because lung PO, appears to remain high during digestion and the increased blood oxygen affinity would, everything else being equal, impair unloading in the tissues. It is possible, however, that a decrease in pulmonary transit time due to increased pulmonary blood flow renders pulmonary gas exchange much more diffusion-limited than at rest; the increased affinity acts to overcome this limitation. Nevertheless, it is still not clear why Alligator and Python exhibit opposite responses; measurements in future of the diffusive resistances for oxygen in the lungs and the systemic circulation may provide an explanation for this phenomenon.
INTERACTION BETWEEN DIGESTION AND OTHER PHYSIOLOGICAL STATES Organisms often undertake many different kinds of activities simultaneously and consequently their physiological design must accommodate the concomitant demands of these activities (Alsop and Wood, 1997; Bennett and Hicks, 2001).Along these lines, animals in nature may be active while eating and digesting. For example, the white-throated monitor lizard, Varanus albigularis,has been tracked walking an average of 1.5-2 km per day during the feeding and growth season (Phillips, 1995).Consequently the overall "design" of the oxygen transport system may have to accommodate postprandial exercise. As discussed throughout this chapter, in many infrequently feedingreptiles and amphibians exercise and digestion independently increase rates of oxygen consumption (vo,), such that the metabolic scopes of these two activities are similar (Benedict,1932;Secor and Diamond, 1995,1997;Secor et al., 2000; Andersen et al., 2003).In Python molurus, for example, postprandial VO2values are equal to those of snakes engaging in maximal sustainable exercise (Secor et al., 2000). Consequently, when animals undergo the challenge of postprandial exercise, the cardiopulmonarysystem responds to this challenge either by according priority to one activity (i.e. exercise prioritized over digestion) or by sharing common support systems (i.e. both digestion and exercise are supported equally and simultaneously;Bennett and Hicks, 2001). The postprandial exercise response in amphibians and reptiles provides an ideal organismal model system for investigating how cardiopulmonary design accomplishes the challenges of simultaneousmetabolic demands in various organs and tissues (Bennett and Hicks, 2001). In a recent study Andersen and Wang (2003)presented data on gas exchange, heart rate, and blood pressure during digestion and forced activity
Effects of digestion
297
of fasting and postprandial toads (Bufo marinus). Maximal VO, during activity was not influencedby digestive state. This response is consistent with the idea that all metabolism associated with digestion is curtailed during activity (exerciseis prioritized over digestion).But it is possible that allocation of energy to digestion continues unaltered, leaving less aerobic scope for activity (Andersen and Wang, 2003). The study could not distinguish between the two possibilities.The increased bicarbonate buffer capacity of postprandial toads tended to dampen reduction in pHa immediately following activity, but there were no marked effects on the rate of recovery of arterial acidbase parameters. Hence the alkaline tide, with increased bicarbonate levels did, not alter the acid-base disturbance during and following activity in toads (Andersenand Wang, 2003).Maximal heart rate and blood pressure in Bufo was higher when exercise was combined with digestion, which is consistent with data on Python (Andersen and Wang, 2003; Secor et al., 2000). The pattern of oxygen transport during fasting exercise and postprandial exercise was recently investigated in python, Python morulus, and monitor lizard, Varanus exanthematicus (Secor et al., 2000; Bennett and Hicks, 2001). The latter study specifically investigated whether monitor lizards exhibited priority: oxygen delivery directed primarily to digestion (curtailing locomotor performance) or directed to exercise (curtailing digestion). The study alternatively tested the hypothesis that the cardiopulmonary system may exhibit additivity, in which the two processes are supported equally. Interestingly, the monitor lizard exhibited a 25% increase in V02max(Fig. 11.8)supporting the hypothesis that the cardiopulmonary response to postprandial exercise is additive (Bennett and Hicks, 2001). This conclusion follows similar observations on exercising postprandial pythons (Secor et al., 2000). Interestingly, maximal heart rate in exercising postprandial pythons exceeds that of either resting postprandial or exercising fasting animals (Secor et al., 2000). Wang et al. (2001b) suggested that this additional increase in heart rate may be explained by noncholinergic, nonadrenergic factors, possibly a peptide released from GI organs during digestion that act directly on the heart. These studies suggest that, at least in reptiles, excess oxygen transport capacity in the cardiopulmonarysystem exists that are not accessed by maximal muscle activity. It is possible that in these animals considerable structural and physiological plasticity exists during the postprandial state. This includes up-regulation of transport capacities within the intestine and, in some instances, an increase in organ size (Secor and Diamond, 1995,1997). Consequently, during the early phases of the postprandial period (within 24 h) there may be significant remodeling of the cardiopulmonarysystem, such that its transport capacity is increased. The underlying mechanisticbasis for this excess capacity in reptiles remains to be investigated.
298
Physiological and ecological adaptations t o feeding in vertebrates
Varanus exanthematicus
U Fasting 4 Post-prandial
-
4
REST
0,5
1,O
Treadmill Speed
REST
0,5
1,o
13
Treadmill speed (km/h) Fig. 11.8. Rates of oxygen consumption at rest and during treadmill exercise in fasting and postprandial Savannah monitor lizards (Varanus exanthematicus). Note that the increased oxygen uptake due to digestion is additive to that elicited by exercise. The insert shows the change in (postprandial fasting) as a function of treadmill speed (Bennett and Hicks, 2001).
GENERAL CONCLUSIONSAND SOME PROSPECTIVES Physiological responses to food intake are very large in reptiles and amphibians and digestion is associated with marked changes in acid-base status and cardiorespiratorycompensations. In some species, the maximal rates of oxygen uptake are similar during exercise and digestion.However, the physiological nature of activity and digestion are very different; activity predominantly involves an increased metabolism of skeletal muscle whereas digestion is associated with an increased metabolism of the gastrointestinal system. Activity is predominantly fueled by anaerobic metabolism and associated with acidosis while digestion is aerobic and characterized by alkalosis. In most cases, activity can only be endured for short periods (less than lh) whereas increased metabolism lasts for several days during digestion. These large differences make it possible to compare and contrast the physiological and structural limits to sustainabilityof high metabolic rates in reptiles.
Effects of digestion
Acknowledgements
This study was supported by The Danish Research Council and The National Science Foundation.
Alsop D.H. and Wood C.M. 1997. The interactive effects of feeding and exercise on oxygen consumption, swimming performance and protein usage in juvenile rainbow trout (Oncorhynchus mykiss). J. Exp. Biol. 200: 2337-2346. Andersen J.B. and Wang T 2003. Cardiorespiratory effects of forced activity and digestion in toads. Physiol. Biochem. Zool. 76: 459-470. Andersen J.B., Andrade D. and Wang T. 2003. Effects of pharmacological inhibition of gastric acid secretion on arterial acid-base status during digestion in the toad Bufo marinus. Comp. Biochem. Physiol. 135A: 425-433. Andrade D.V., Cruz-Neto A.P. Abe A.S. 1997. Meal size and specific dynamic action in the rattlesnake Crotalus durissus (Serpentes: Viperidae). Herpetologica 53: 485-493. Andrade D.V., Toledo L.F., Abe A.S. and Wang T. 2004a. Ventilatory compensation of the alkaline tide in the snake Boa constrictor. J. Exp. Biol. 207: 1379-1385. Andrade D.V., Cruz-Neto A.P., Abe A.S. and Wang T. 2004b. Specific dynamic action in ectothermic vertebrates: A general review of the determinants of postprandial metabolic response in fishes, amphibians, and reptiles (this volume). Axelsson M. and Fritsche R. 1991. Effects of exercise, hypoxia and feeding o n the gastrointestinal blood flow in the Atlantic cod, Gadus morhua. J. Exp. Biol. 158: 181-198. Axelsson M., Driedzic W. R., Farrell A. P. and Nilsson S. 1989. Regulation of cardiac output and gut blood flow in the sea raven, Hemiptripterus americanus. Fish Physiol. Biochem. 6: 315-326. Axelsson M., Altimiras J. and Claireaux G. 2002. Post-prandial blood flow to the gastrointestinal tract is not compromised during hypoxia in the sea bass, Dicentrarchus labrax. J. Exp. Biol. 205: 2891-2896. Axelsson M., Thorarensen H., Nilsson S. and Farrell A.P. 2000. Gastrointestinal blood flow in the red Irish lord, Hmilepidotus hemilepidotus: Long term effects of feeding and adrenergic control. J. Comp. Physiol. B 170: 145-152. Axelsson M., Fritsche R., Holmgren S., Grove D. J. and Nilsson S. 1991. Gut blood flow in the estuarine crocodile, Crocodylus porosus. Acta Physiol. Scand. 142: 509-516. Bauer C. and Jelkmann W. 1977. Carbon dioxide governs the oxygen affinity of crocodile blood. Nature 269: 825-827. Benedict F. G. 1932. The Physiology of Large Reptiles with Special Reference to the Heat Production of Snakes, Tortoises, Lizards and Alligators. Carnegie Inst. Publ., Washington, DC. Bennett A. F. 1994. Exercise performance in reptiles. Adv. Vet. Sci. Comp. Med. 38B: 113-138. Bennett A. F. and Hicks J. W. 2001. Postprandial exercise: Prioritization or additivity of the metabolic responses? J. Exp. Biol. 204: 2127-2132. Borsook H. 1936. The specific dynamic action of protein and amino acids in animals. Biol. Rev. 11: 147-180. Brown C. R. Cameron J. N. 1991a. The induction of specific dynamic action in channel catfish by infusion of essential amino-acids. Physiol. Zool. 64: 276-297. Brown C. R. Cameron J. N. 1991b. The relationship between specific dynamic action (SDA) and protein- synthesis rates in the channel catfish. Physiol. Zool. 64: 298-309. Burggren W. W., Moreira G . S. and Santos M. F. 1993. Specific dynamic action and the metabolism of the brachyuran land crabs Ocypode quadrata, Goniopsis cruentata and Cardisoma guanhumi. J. Exp. Mar. Biol. Ecol. 169: 117-130.
300
Physiological and ecological adaptations to feeding in vertebrates
Busk M., Jensen F.B. and Wang T. 2000a. The effects of feeding on blood gases, acid-base parameters and selected metabolites in the bullfrog Rana catesbeiana. Amer. J. Physiol. 278: R185-R195. Busk M., Overgaard J., Hicks J.W., Bennett A.F. and Wang T. 2000b. Effects of feeding on arterial blood gases in the American alligator, Alligator mississippiensis. J. Exp. Biol. 203: 3117-3124. Chou C.C. 1983. Splanchnic and overall cardiovascular hemodynamics during eating and digestion. Fed Proc. 42: 1658-1661. Coulson R.A. and Hernandez T. 1979. Increase in metabolic rate of the alligator fed proteins or amino acids. J. Nutr. 109: 538-550. Coulson R.A., Hernandez T. and Dessauer H.C. 1950. Alkaline tide in alligators. Soc. Exp. Biol. Med. 74: 866-869. Cruz-Neto A.P., Andrade D.V. and Abe A.S. 1999. Energetic cost of predation: Aerobic metabolism during prey ingestion by juvenile rattlesnakes, Crotalus durissus. J. Herp. 33: 229-234. Cruz-Neto A.P., Andrade D.V. and Abe A.S. 2001. Energetic and physiological correlates of prey handling and ingestion in lizards and snakes. Comp. Biochem. Physiol. 128A: 513-531. Dumsday B. 1990. Resting heart rate of the toad Bufo marinus: a long-term study of individual differences and environmental influences. Physiol. Zool. 63: 420431. Ehrly A.M. and Jung G. 1973. Circadian rhythm of human blood viscosity. Biorheology 10: 577-583. Farmer C. and Carrier D.R. 2000. Ventilation and gas exchange during treadmill locomotion in the American alligator (Alligator mississippiensis). J. Exper. Biol. 203: 1671-1678. Feder M.E. and Burggren W.W. 1985. Cutaneous gas exchange in vertebrates: Designs, patterns, control and implications. Biol. Rev. 60: 1 4 5 . Fellenius E., Berglindh T. and Sachs G. 1981. Substituted benzimidazoles inhibit gastric acid secretion by blocking (H+,K+)ATPase. Nature 290: 159-161. Fronek K. and Fronek A. 1970. Combined effect of exercise and digestion on hemodynamics in conscious dogs. Amer. J. Physiol. 218: 555-559. Fulton J.F., and Wilson L.G. 1930. Selected Readings in the History of Physiology. Charles C Thomas Publ. Chicago, USA. Glass M.L. and Heisler N. 1986. The effects of hypercapnia on the arterial acid-base status in the Tegu lizard, Tupinambis nigropunctatus (Spix). J. Exp. Biol. 122: 13-24. Glass M.L., Wood S.C., Hoyt R.W. and Johansen K. 1979. Chemical control of breathing in the lizard, Varanus exanthematicus. Comp. Biochem. Physiol. 62A: 999-1003. Gleeson T.T. 1996. Post-exercise lactate metabolism: A comparative review of sites, pathways, and regulation. Annu. Rev. Physiol. 58: 564-581. Gonzalez N.C., Clancy R.L., Moue Y. and Richalet J.P. 1998. Increasing maximal heart rate increases maximal 0, uptake in rats acclimatized to simulated altitude. J. Appl. Physiol. 84: 164-168. Greene H.W. 1983. Dietary correlates of the origin and radiation of snakes. Amer. Zool. 23: 431-441. Greene H.W. 1997. Diet and feeding. In: Snakes: The Evolution of Mystery in Nature. Univ. California Press, Berkeley, CA. Hartzler L. K., Munns S. L., Bennett A. F. and Hicks J.W. 2003. pH regulation in ectothermic vertebrates: digestion and exercise-induced changes. FASEB J. 17(4-5): Abstract no. 87.2. Hawkins P.A. J., Butler P. J., Woakes A. J. and Gabrielsen, G.W. 1997. Heat increment of feeding in Brunnich's guillemot Uria lomvia. J. Exp. Biol. 200: 1757-1763. Hedrick M.S., Palioca W.B. and Hillman S.S. 1999. Effects of temperature and physical activity on blood flow shunts and intracardiac mixing in the toad Bufo marinus. Physiol. Biochem. Zool. 72: 509-5-19. Hersey S. J. and Sachs G. 1995. Gastric-acid secretion. Physiol. Rev. 75: 155-189.
Effects of digestion
301
Hicks J.W. 1998. Cardiac shunting in reptiles: mechanisms, regulation and physiological function. In: Biology of the Reptilia, vol. G., The Visceral Organs. C. Gans and A.S. Gaunt (eds.). Soc. Study Amphibians Reptiles, Ithaca, NY, pp. 425-483. Hicks J.W. and Wang T., 1996. Functional role of cardiac shunts in reptiles. J. Exp. Zool. 275: 204-216. Hicks J.W., Wang T. and Bennett A.F. 2000. Patterns of cardiovascular and ventilatory response to elevated metabolic states in the lizard, Varanus exanthematicus. J. Exp. Biol. 203: 2437-2445. Hillman, S.S. 1987. Dehydrational effects on cardiovascular and metabolic capacity in two amphibians. Physiol. 2001. 60: 608-613 Holmgren S. 1995. Neuropeptide control of the cardiovascular system in fish and reptiles. Braz. J. Med. Biol. Res. 28: 1207-1216. Houlihan D. F. 1991. Protein turnover in ectotherms and its relationships to energetics. Adv. Comp. Physiol. Biochem. 7: 1-43. Huang J-Q. and Hunt R.H. 2001. Pharmacological and pharmacodynamic essentials of H,-receptor antagonists and proton pump inhibitors for the practising physician. Best Prac. Res. Clin. Gastroenterol. 15: 355-370. Jackson D.C. 1969. The response of the body fluids of the turtle to imposed acid- base disturbances. Comp. Biochem. Physiol. 29: 1105-1110. Jackson D.C. 1986. Acid base regulation in reptiles. In: Acid-base Regulation in Animals. N . Heisler, (ed.). Elsevier Sci. Publ. Amsterdam, Netherlands, pp. 235-263. Jensen J., Axelsson M. and Holmgren S. 1991. Effects of substance P and vasoactive intestinal polypeptide on gastrointestinal blood flow in the Atlantic cod Gadus morhua, J. Exp. Biol. 156: 361-373. Jensen F.B., Wang T., Jones D.R. and Brahm J. 1998. Carbon dioxide transport in alligator blood and its erythrocyte permeability to anions and water. Amer. J. Physiol. 274: R661-R671. Jobling M. 1981. The influences of feeding on the metabolic rate of fishes: a short review. J. Fish Biol. 18: 385-400. Johnson C.D., Mole D.R. and Pestridge A. 1995. Postprandial alkaline tide: Does it exist? Digestion. 56: 100-106. Kagstrom J., Olsson C., Axelsson M. and Franklin C. E. 1998. Peptidergic control of gastrointestinal blood flow in the estuarine crocodile, Crocodylus porosus. Amer. J. Physiol. 43: R1740-R1750. Karila P., Axelsson M., Franklin C. E. Fritsche R. et al. 1995. Neuropeptide immunoreactivity and co-existence in cardiovascular nerves and autonomic ganglia of the estuarine crocodile, Crocodylus porosus, and cardiovascular effects of neuropeptides. Reg. Pept. 58: 25-39. Kelbaek H., Munck O., Christensen N.J. and Godtfredsen J. 1987. Autonomic nervous control of postprandial hemodynamic changes at rest and upright exercise. J. Appl. Physiol. 63: 1862-1865. Kleiber M. 1961. The Fire of Life: A n Introduction to Animal Energetics. John Wiley and Sons, New York, NY. Kurata M., Nakamura H., Baba A., Asano T., et al., 1993. Postprandial change in canine blood viscosity. Comp. Biochem. Physiol. 105A: 587-592. Langbroek A.J., Nijmeijer A., Rispens P. and Zijlstra W.G. 1990. Pitfalls in acid/base experiments with conscious dogs. Pfliigers Arch. 417: 157-160. Lusk G. 1931. The specific dynamic action. J. Nutr. 3: 519-530. Malvin G., Macias S., Sanchez M., Dasalla R., Park A. Duran M. 1995. Lymphatic regulation of hematocrit during hypoxia in the toad Bufo woodhousei. Amer. J. Physiol. 269: R814-R821. Matheson P.J., Wilson M.A. and Garrison R.N. 2000. Regulation of intestinal blood flow. J. Surg. Res. 93: 182-196. McCue M.D. and Lillywhite H.B. 2002. Oxygen consumption and the energetics of island-dwelling Florida cottonmouth snakes. Physiol. Biochem. Zool. 75: 165 -178.
302
Physiological and ecological adaptations t o feeding in vertebrates
McCue M.D., Bennett A.F., and Hicks J.W. 2005. The effect of meal composition on Specific Dynamic Action in Burmese Pythons (Python molurus). Physiol. Biochem. Zool. 78 (in press). Milsom W.K. 1988. Control of arhythmic breathing in aerial breathers. Can. J. Zool. 66: 99-108. Mitchell G. S., Gleeson T. T. and Bennett A. F. 1981. Pulmonary oxygen transport during activity in lizards. Respir. Physiol. 43: 365-375. Morris J.L. 1989. The cardiovascular system. In: The Comparative Physiology of Regulatory Peptides, S. Holmgren, (ed.). Chapman and Hall: London, pp. 272-307. Morris J.L. Nilsson S. 1994. The circulatory system. In: Comparative Physiology and Evolution of the Autonomic Nervous System, S. Nilsson S. Homgren, (eds.). Harwood Acad. and Publ. Chur, Switzerland, pp. 193-256. Niv Y. and Fraser G.M. 2002. The alkaline tide phenomenon. J. Clin. Gastro. 35(1): 5-8. Overgaard J. Wang T. 2002. Oxygen binding properties of whole blood, before and after feeding, in the snake Python molurus. J. Exp. Biol. 205: 3327-3334. Overgaard J., Andersen J.B. and Wang T. 2002. The effects of fasting duration on the metabolic response to feeding in Python molurus: An evaluation of the energetic costs associated with gastrointestinal growth and upregulation. Physiol. Biochem. Zool. 75: 360-368. Overgaard J., Busk M., Hicks J.W., Jensen F.B. and Wang T. 1999. Acid-base status and arterial oxygen transport following feeding in the snake Python molurus. Comp. Biochem. Physiol. 124A: 361-367. Phillips J. A. 1995. Movement patterns and density of Varanus albigularis. J. Herpetol. 29: 407-416. Pinder A. W. and Smits A.W. 1993. Mechanisms of acute hemoconcentration in bullfrogs in response to hypoxemia. Amer. J. Physiol. 33: R687-R695. Rubner M. 1902. Die Gesetze des Energieverbrauchs bei der Ernahrung. Frank Dauticke, Publ. Leipzig, Germany. Rune S.J. 1965. The metabolic alkalosis following aspiration of gastric acid secretion. Scand. J. Clin. Lab. Invest. 17: 305-310. Rune S.J. and Lassen N.A. 1968. Diurnal variations in the acid-base balance of blood. Scand. J. Clin. Lab. Invest. 22:151-156. Sachs G., Shin J.M., Briving C., Wallmark B. and Hersey S. 1995. The pharmacology of the gastric acid pump: the H+, K+ ATPase. Annu. Rev. Pharmacol. Toxicol. 35: 277-305. Secor S.M. 2001. Regulation of digestive performance: An adaptive response. Comp. Biochem. Physiol. 128A: 563-575. Secor S.M. 2003. Gastric function and its contribution to the postprandial metabolic response of the Burmese python Python molurus. J. Exp. Biol. 206: 1621-1630. Secor S.M. and Diamond J. 1995. Adaptive responses to feeding in Burmese pythons: pay before pumping. J. Exp. Biol. 198: 1313-1325. Secor S.M. and Diamond J. 1997. Determinants of the postfeeding metabolic response of Burmese pythons, Python molurus. Physiol. Zool. 70: 202-212. Secor S.M. and Diamond J. 1998. A vertebrate model of extreme physiological regulation. Nature. 395: 659-662. Secor S.M., Stein E.D. and Diamond J. 1994. Rapid upregulation of snake intestine in response to feeding: A new model of intestinal adaptation. Amer. J. Physiol. 29: G695-G705. Secor S.M., Hicks J.W. and Bennett A.F. 2000. Ventilatory and cardiovascular responses of pythons (Python molurus) to exercise and digestion. J. Exp. Biol. 203: 2447-2454. Shelton G. 1985. Functional and evolutionary significance of cardiovascular shunts in the amphibia. In: Cardiovascular S h u n t s . Johansen and W.W. Burggren (eds.). Munksgaard, Denmark. Silver R.B. and Jackson D.C. 1985. Ventilatory and acid-base responses to long-term hypercapnia in the freshwater turtle, Chrysemys picta bellii. J. Exp. Biol. 114: 661-672.
Effects of digestion
303
Silver R.B. and Jackson D.C. 1986. Ionic responses with no renal responses to chronic hypercapnia in the freshwater turtle, Chrysemys picta bellii. Amer. J. Physiol. 20: R1228-R1234. Starck J.M. 2004. Structural flexibility of the digestive system of tetrapods- patterns and processes at the cellular and tissue level (this volume). Starck J.M. and Beese K. 2001. Structural flexibility of the intestine of Burmese python in response to feeding. J. Exp. Biol. 204: 325-335. Stewart P.A. 1983. Modern quantitative acid-base chemistry. Can. J. Physiol. Pharm. 61: 1442-1461. Toews D. P. and Boutilier R. G. 1986. In: Acid-base Regulation in Animals. N. Heisler, (ed.). Elsevier Sci. Publ., Amsterdam, Netherlands, pp. 265-308. Vaziri N.D., Byrne C., Ryan G. and Wilson A. 1980. Preservation of urinary postprandial alkaline tide despite inhibition of gastric acid secretion. Amer. J. Gastroenterol. 74: 328-331. Waaler B.A., Eriksen M. and Janbu T. 1990. The effect of a meal on cardiac-output in man at rest and during moderate exercise. Acta Phys. Scand. 140: 167-173. Waaler B.A., Eriksen M. and Toska K. 1991. The effect of meal size on postprandial increase in cardiac output. Acta Phys. Scand. 142: 33-39. Wang T. and Hicks J.W. 1996. The interaction of pulmonary ventilation and the right-left shunt on arterial oxygen levels. J. Exp. Biol. 199: 2121-2129. Wang T. and Hicks J.W. 2002. An integrative model to predict maximum oxygen uptake of animals with central vascular shunts. Zoology 105: 45-53. Wang T., Burggren W.W. and Nobrega E. 1995. Metabolic, ventilatory, and acid-base responses associated with specific dynamic action in the toad Bufo marinus. Physiol. Zool. 68: 192-205. Wang T., Carrier D. R. and Hicks J. W. 1997a. Ventilation and gas exchange in lizards during treadmill exercise. J. Exp. Biol. 200: 2629-2639. Wang T., Krosniunas E. Hicks J.W. 1997b. The role of cardiac shunts in the regulation of arterial blood gases. Amer. Zool. 37: 12-22. Wang T., Busk M. and Overgaard J. 2001a. The respiratory consequences of feeding in amphibians and reptiles. Comp. Biochem. Physiol. 128A: 533-547. Wang T., Axelsson M., Jensen J. a n d Conlon J.M. 2000. Cardiovascular actions of python bradykinin and substance P in the anesthetized python, Python regius. Amer. J. Physiol. 279: R531-R538. Wang T., Taylor E.W., Andrade D. and Abe A.S. 2001b. Autonomic control of heart rate during forced activity and digestion in the snake Boa constrictor. J. Exp. Biol. 204: 3553-3560. Wang T., Warburton S.J., Abe A.S. amd Taylor E.W. 2001c. Vagal control of heart rate and cardiac shunts in reptiles: Relation to metabolic state. Exper. Physiol. 86: 777-786. Weber R.E. and White F.N. 1986. Oxygen binding in alligator blood related to temperature, diving, and "alkaline tide". Amer. J. Physiol. 20: R901-R908. Whiteley N.M., Robertson R.F., Meagor J., El Haj A.J. and Taylor E.W. 2001. Protein synthesis and specific dynamic action in crustaceans: effects of temperature. Comp. Biochem. Physiol. 128A: 593-604. Wilhelmj C.M. 1935. The specific dynamic action of food. Physiol. Rev. 15: 202-230. Zaar M., Larsen E. and Wang T. 2004. Hysterises of heart rate and heat exchange of fasting and postprandial savannah monitor lizards (Vasanus exanthematicus). Comp. Biochem. Physiol. 137A: 675-682. Zaidan F. and Beaupre S.J. 2003. Effects of body mass, meal size, fast length, and temperature on specific dynamic action in the timber rattlesnake (Crotalus horridus). Physiol. Biochem. Zool. 76: 447-458.
Specific Dynamic Action in Ectothermic Vertebrates: A Review of the Determinants of Postprandial Metabolic Response in Fishes, Amphibians, and Reptiles Denis V. Andradel, Ariovaldo P. Cruz-Netol, Augusto S. Abel and Tobias Wang2 'Universidade Estadual Paulista, Departament de Zoologia, Rio Claro, SP, Brasil. 2University of Aarhus, Dept. Zoophysiology, Denmark
SYNOPSIS This chapter summarizes current knowledge regarding the postprandial metabolic response in fishes, amphibians, and reptiles. Even though these "lower" vertebrates represent phylogenetically distinct groups, they are all characterized by being ectothermic, which has implications for vitually all aspects of their biology. As some of their morphological, physiological and ecological characteristicsare directly influenced by this trait, it is possible to draw broad generalizations for some of these features. An attempt is made here to bring together one such characteristic, the postprandial increase in metabolism (herein referred to as specific dynamic action, SDA). As the nomenclature associated with SDA differs appreciably among studies, an attempt to standardize these diverse nomenclatures comes first. Next the proximate factors that affect the SDA response and its consequences for the energy budget of ecothermic vertebrates are reviewed. Lastly the mechanistic causes of SDA are briefly described.
306
Physiological and ecological adaptations t o feeding in vertebrates
All animals must eat to match energy expenditure associated with resting metabolism and activities. Procuration and subdual of suitable prey items and their subsequent ingestion and digestion are therefore fundamental for all animals and it seems obvious that a cost-effective digestive system would increase fitness and have been favored through evolution. Given that the demands exerted by digestion have a number of biological consequences,it is not surprising that the mechanisms of feeding and digestion and their physiological implications remain the subject of numerous studies. Hence, for example, digestion exerts profound effects on behavior (e.g. Huey, 1982; Dorcas et al., 1997; Blouin-Demers and Weatherhead, 2001) and triggers rapid and dramatic morphological changes (e.g. Piersma and Lindstrom, 1997;Starck, 1999a,b;Konarzewski and Starck, 2000; Starck and Beese, 2001). Among the major physiological effects, metabolism increases during digestion. This postprandial increment of metabolic rate, is often referred to as the specific dynamic action of food and abbreviated SDA (e.g. Kleiber, 1961).The processes underlying the SDA response involve mechanical and biochemical degradation along with de novo synthesis. The SDA response is particularly profound in ectothermic vertebrates because of their low basal metabolic rate and because they often ingest meals that are very large compared to their own body mass. Ectothermic animals have accordingly proven convenient for studies on the mechanisms that underlie the SDA response. Nevertheless, even though the metabolic stimulationof food has been studied. since the time of Lavoisier, many aspects remain uncertain and there has been considerable controversy about the respective metabolic costs of intestinal growth, protein synthesis, gastric acid secretion, etc. (Secor, 2001, 2003; Starck and Beese, 2001, 2002; Starck et al. 2004). Given that the metabolic costs associated with feeding and digestion are an integral part of organismal function and permeate virtually all aspects of an organism's physiology and ecology, it is important to understand these mechanisms. SDA accounts for a considerableportion of the energy budget in ectothermic animals. The SDA coefficient,with few upper or lower exceptions, usually accounts for 10%to 20% of the energy obtained from the meal (Jobling, 1981; Pandian, 1987; Secor, 2001). In snakes, two studies have projected the contribution of the postprandial metabolic response to the whole energy budget of free-ranging animals. Secor and Nagy (1994) estimated that 43% and 19%of field metabolic rate of Crotalus cerastes and Masticophisflagellum respectively is due to the costs associated with SDA during the active season. In the cottonmouth, Agkistrodon piscivorus, McCue and Lillywhite (2002)estimated that SDA accounts for almost 40% of the annual energy budget. Considering all the other energy-consuming activities that animals engage in during their life cycle and the "sustained" cost of maintenance of the
SDA in ectothermic vertebrates
307
physiological systems, the amount of energy allocated for digestion per se is truly amazing. The high cost also stresses the importance SDA may have for the energetic and ecological relationships of ectothermic vertebrates. Perhaps this pattern arises from the "low energy" approach of living typical of ectothermic vertebrates, which allows them to maintain low levels of metabolic activity (Pough, 1983). A review is given here of the metabolic consequences of feeding in ectothermic vertebrates. Ectothermic vertebrates rely on behavioral adjustments to control their body temperature and their metabolic rates are an order of magnitude lower than that of birds and mammals. CHARACTERIZATION OF METABOLIC RESPONSETO FEEDING: SPECIFIC DYNAMIC ACTION
Rubner (1902)coined the German term "spezifisch dynamischer Effekt" to describe the increased metabolism during digestion, which was later somewhat inappropriately translated to "specific dynamic action" (SDA). Originally, SDA only included up to postabsorptive processes while "apparent SDA" included the entire postprandial response (Beamish, 1974). Given the difficulties of separatingSDA and apparent SDA, SDA has become the most commonly used term to describe the metabolic increment after feeding. The metabolic response to feeding is commonly measured as the increase in the rate of oxygen consumption, although some studies have also characterized changes in heat production by calorimetry or measured the rate of CO, excretion.When only the calorigenic effect is measured, SDA is sometimesreferred to as "heat increment" (Kleiber, 1961).While CO, excretion is often easier to measure than oxygen uptake, its usage to adequately estimate energy expenditure requires assumptions regarding a respiratory quotient. As discussed below, the SDA response differs among and within species according to meal size and composition, temperature and other factors. However, in general, the SDA response is characterized by a relatively rapid metabolic increment followed by a slower return to the fasting level (See Fig. 12.1).As summarized in Fig. 12.1and Table 12.1, the SDA response can be characterized with regard to the factorial increase in metabolism, the maximal rate of metabolism attained during the SDA, or lastly as the integrated SDA response, which represent the total energetic costs of digestion relative to energeticcontent of the meal. While each of these expressions may be adequate to characterize a given facet of the SDA response, it should be emphasized that some expression can be misleading. For example, the factorial increase in oxygen uptake can be very large if the preceding fasting is low. In that case, usage of the factorial increase may serve to overemphasizethe effects of metabolic stimulation by feeding compared to expressing the maximal rate of oxygen consumption or the integrated SDA response.
Physiological and ecological adaptations t o feeding in vertebrates
308
Y a,
a +-
n 3
c
a, 0)
X
6
Time
Fig. 12.1. Schematic representation of the metabolic response to feeding and the calculation of the SDA response (SDA - Specific Dynamic Action). Standard metabolic rate is illustrated in dark grey, while the increase in oxygen consumption caused by digestion (SDA) is shown by light grey.
The SDA response may include some of the metabolic costs incurred during the preingestive handling of the meal (foraging, capture, ingestion, mastication, etc.).Jn Boa, constriction and ingestion of the prey are largely anaerobic (Canjani et. al., 2003) and pythons experience a large increase in metabolism after prey constriction even if the prey is not ingested (Secor and Diamond, 1997).Thus, some of the initial cost of digestion, at least in constricting snakes,may include an oxygen debt generated before ingestion (see CruzNeto et al., 2001). FACTORS AFFECTING POSTPRANDIAL METABOLISM Most studies on the postprandial stimulation of metabolism in fishes, amphibians, and reptiles have merely characterized metabolic rate after feeding,while fewer studies have explicitly focused on the factors that affect and/or determine the SDA response. Furthermore, while some determinants, such as temperature and meal size, have consistent and largely intuitive effects, the actual causes of SDA remain an area with conflicting conclusions and unresolved issues. Within each phylogenetic group, different species are separated to various degrees and present features may, at least in part, reflect historical constraints or persistence of ancestral traits, rather than adaptation toward a present-day situation (Harvey and Pagel, 1993).Aware of this possibility, we adopted a mechanistic/causal approach in this review, rather than an evolutionary historical perspective hindered by lack of data. What contributesto the cost of digestion?The causal processes for SDA response are complex but can be grouped into 1)catabolic processes, which include enzymatic and mechanical digestion, intestinal transport, peristalsis,
SDA in ectothermic vertebrates
309
Table 12.1. Definition of the metabolic change during the SDA response SDA duration
The period between ingestion and the return of metabolism to fasting level.
Peak rate
Maximal rate of oxygen consumption during SDA.
Time to Peak
The period between ingestion and Peak Rate.
Factorial Scope during Digestion : The relative increase in metabolism during digestion, calculated as Peak Rate divided by Fasting Rate. Mean SDA Rate
The mean rate measured over the duration of the SDA.
SDA Total Cost
The total energy expended during the SDA.
SDA Net Cost
The S D A Tofal Cost discounted for maintenance cost (usually estimated from the Fasting Rate).
Meal Energy Content
The caloric content of the meal.
Assimilation Efficiency
The amount of energy that is assimilated from the meal (usually expressed in relative terms).
Energy Assimilated from the Meal
The energy actually assimilated from the ingested meal. Calculated from the Meal Energy Content and the Assimilation Efficiency.
S DA coefficient
The S D A Net Cost expressed relative to Energy Assimilated from the Meal. If assimilation efficiency is not determined, the SDA coefficient can be calculated relative to Meal Energy Content.
Fasting rate
The rate of the parameter examined (0,uptake, heat production, etc.) of post-absorptive animals. Ideally, this should be measured in animals fasting sufficiently long to avoid influence from a previous meal. Fasting animals should, nevertheless, be healthy and in good condition.
Meal Energy Content
The caloric content of meal.
absorption and nutrient storage, and 2) anabolic processes, including the costs of amino acid deamination, synthesis of excretory products, increased synthesis and deposition of proteins (Jobling, 1981,1983; Peck, 1998; Peck and Veal, 2001; McCue et al., 2005).
Meal Size Effects Meal mass is probably the singlemost influential determinant of the SDA response. In general, increased meal size is accompanied by a proportional rise in SDA response, which is manifested as higher maximal rates of oxygen consumption and larger factorial increments. Increased meal size also prolongs SDA response and maximal rates of oxygen consumption tend to occur later with larger meals. These effects of meal size are well characterized in fishes (e.g.Beamish, 1974;Jobling and Davies, 1980;Soofiani and Hawkins, 1982;Billerbeck et al., 2000).For example, the SDA scope of the SDA increases linearly with meal size in juvenile cod and the Atlantic silverside (Gadus morhua and Menidia menidia respectively; Soofiani and Hawkins, 1982;
310
Physiological and ecological adaptations t o feeding in vertebrates
Billerbeck et al., 2000), while the duration of the SDA curve increases linearly with meal size in largemouth bass (Micropterussalmoides; Beamish, 1974). An increased SDA response with elevated meal size has also been documented in reptiles and some amphibians (e.g. Andrade et al., 1997; Secor and Diamond, 1997;Secor and Faulkner, 2002).In snakes, able to ingest very large prey, there seems to be no plateau for the metabolic response to feeding in response to meal size increment. For example, in pythons, boas, and rattlesnakes fed a broad range of meal mass, the pattern of metabolic change during digestion was always characterized by an initial rapid increase that peaked 1or 2 days after meal ingestion, followed by the slower return to fasting metabolism level in the course of days (e.g. Andrade et al., 1997; Secor and Diamond, 1997; McCue and Lillywhite, 2002; Toledo et al., 2003).Maximal 0, uptake increased progressively with meal mass and in an extreme case of Python ingesting 10O0/0 of its own body mass (Secor and Diamond, 1995),has been reported to increase 44 times above fasting levels. Smaller, yet still impressive factorial increments between 5-to 15-fold are more commonly reported (Andrade er al., 1997; Secor, 2001; McCue and Lillywhite, 2002).As in fishes, the peak in the postprandial metabolic response is delayed as meal size increases, which together with the prolonged SDA response, may, at least in part, reflect the fact that snakes ingest their prey without mastication. As would seem intuitive the costs associated with digestion increase with meal size, but because the energetic content of the food also increases proportionally, the SDA coefficient of pythons, boas, and rattlesnakes tends to be rather unaffected by meal size (e.g Andrade et al., 1997; Secor and Diamond, 1997; Toledo et al., 2003). However, in some species, such as the cottonmouth, the SDA coefficient increases with meal size indicating that large meals are proportionally more energeticallyexpensive to digest (McCue and Lillywhite, 2002).In fishes, the effects of meal size on the SDA coefficient are more variable and somewhat controversial. It has been argued that meal size and the SDA coefficient are inversely related in Antarctic fishes (Peck, 1998).For example, in the Antarctic plunderfish, Harpagifer antarcticus, the SDA coefficient varied from 9 to 55% as meal size decreased (Boyce and Clarke, 1997).On the other hand, Soofiani and Hawkins (1982)and Billerbeck et al. (2000)found that the SDA coefficient increased with meal size in cods and the Atlantic silverside (Gadus morhua and Menidia menidia respectively). Finally, there was no effect of meal size on the SDAcoefficientin largemouth bass, Micropterus salmoides (Bearnish, 1974).It is possible that these differences reflect variation in feeding ecology, different functionaland/or physiological constraints or adaptations to specific environments. However, how these factors interact to affect the costs of digestion remains an open question for future studies.
SDA in ectothermic vertebrates
Composition of Diet Secor and Phillips (1997)compared the SDA of Varanus albigularis fed young rats, hard-boiled eggs, ground turkey, and snail mixture. Peak VO, and the integrated SDA response was largest with turkey, snail or rat compared to hard-boiled eggs, but when the SDA response was expressed relative to caloric content of the meal, there were no differences among meal types (Secor and Phillips, 1997). For sharks, Ferry-Graham and Gibb (2001) attributed differences in SDA response between squid and fish diets (Simsand Davies, 1994)to differences in caloric content. The finding that the SDA response increases proportional to energy content is consistent with the increased SDA response with meal size of a given food item. However, to establish whether some components of the food elicits larger metabolic effects than others, several older and recent studies have manipulated food composition or have given artificial diets (amino acids solutions, gelatinepeptone, or others).Early studies on reptiles showed that protein-rich meals elicit larger metabolic changes than diets composed of fat or carbohydrates (e.g. Benedict, 1932) and these findings were later verified. Thus, force-feeding snakes with fat elicits almost no metabolic response (Benedict,1932;McCue et al., 2005);there are some indications that the SDA response is affected by the amount of fat in the diet in the salt water crocodile, Crocodylus porosus (Garnett, 1988).Carbohydrates,in general, fail to stimulatemetabolism. Coulson and Hernandez (1983)force-fed alligators with vegetable proteins and noticed that it would appear virtually unchanged in the feces after a few days, indicating that alligators are incapable of digesting this material. Using a stomach tube, they also infused ground rice, wheat flour, corn meal, and potato flour and subsequently followed the changes in carbohydrate levels in plasma. Glucose was the only carbohydrate that was elevated after feeding and it was concluded that crocodilians, as true carnivores with no requirement for dietary carbohydrate, are unable to digest polysaccharides. In the turtle Kinixys spekii, Hailey (1998)found that the cost of digestion varied with diet from 16%,21°/0, and 30%of the absorbed energy when fed on fungi, leaves, and millipedes respectively. When expressed relative to protein content, the costs of digestion were strikingly similar (Hailey, 1998).In fishes, the SDA response and its duration also increased with elevated protein content in the diet (Joblingand Davies, 1980; Somanath et al., 2000). There is ample evidence in ectotherrnic vertebrates that anabolic processes, and in particular protein metabolism, constitute a dominating factor in SDA response (Ashworth, 1969; Coulson and Hernandez, 1979). In alligators and catfishes, amino acid levels are increased shortly after feeding (Herbert and Coulson, 1976; Brown and Cameron, 1991a)and, if fasting animals are infused with a mixture of amino acids, metabolism will increase to levels comparable to those observed in fed individuals (Coulsonand Hernandez,
312
Physiological and ecological adaptations t o feeding in vertebrates
1979; Brown and Cameron, 1991b).In catfish, it has even been shown that inhibition of protein synthesis completely abolishes the metabolic response to feeding (Brown and Cameron, 1991a,b; see also McCue et al., 2005 for similar data in snakes). The metabolic and energetic implications of intraspecific differences in food types clearly need more investigation since widespread dietary variation occurs in natural populations and ontogenetically. Species that are omnivorous and generalists may offer suitable models to explore the effects of food type on the postprandial response by characterizing the effects of prey items that differ in composition or digestibility. The effect of prey composition and energetic content on energetic return is an ecologically important relationship and should be taken into account when assessing energy flux from prey to predators. The many facets included in this simple question could be explored in future studies using the natural variation in diet, determined by geographical differencesor ontogenetic changes.
Effects of Venom Many snakes use venom to kill the prey and since the venom often contains digestiveenzymes,it has been speculated that the venom may aid degradation of the prey and reduce the ensuing costs of digestion.The effects are largely unexplored, however. Thomas and Pough (1979)observed that digestion in some species of snakes was accelerated if the prey were envenomated. Rodriguez-Robles and Thomas (1992) likewise showed that the venom of the Puerto Rican racer, Alsophis portoricensis, also accelerated digestion. M. McCue and S. P. Brito (unpubl.data),working independently,recently showed that envenomation increased the postprandial peak in metabolism, while both duration and the SDA coefficient reduced. This response holds for species possessing proteolytic and miotoxic venom and is more pronounced at low temperatures (M. McCue and S. P. Brito, unpubl. data). Clearly, the physiological and biochemical reasons for these effects deserve to be studied in more detail. Scaling Effects Body mass affects metabolic rate and although this scaling effect has long been known and demonstrated to be universal, it remains one of the complex issues in comparative physiology (e.g.Somero, and Childress, 1980; Heusner, 1982; Withers, 1992; Coulson, 1997; Gillooly et al., 2001). In ectotherms, metabolism normally scales by the 0.8 power of body mass (Hemmingsen, 1969; Gillooly et al., 2001), but there are considerable differences (e.g. Thompson and Withers, 1994; Chappell and Ellis, 1987). The scaling effects persist when metabolic rate is increased during exercise or digestion but the scaling exponents may be state dependent. In Burmese pythons, for example, SMR scaled with a power of 0.68, while maximal oxygen uptake during digestion scaled with a power of 0.9 (Secor and
SDA in ectothermic vertebrates
313
Diamond, 1997). Also, in sharks (Cephaloscyllum ventriosum), there is a tendency toward higher maximal rates of oxygen consumption during digestion in large vs small individuals (Ferry-Grahamand Gibb, 2001) and a similar pattern was observed in largemouth bass (Beamish,1974).However, body mass did not affect maximal rates of oxygen consumption during SDA response in dogfish or Antarctic plunderfish, although the SDA responses were prolonged in large animals (Sims and Davies, 1994; Boyce and Clarke, 1997).In turtles and the Atlantic silverside, the duration of SDA response decreased with body mass (Sievertet al., 1988; Billerbeck et al., 2000). Although body mass often affects the timing and metabolic profile of the SDA response, the SDA coefficient is generally independent of body mass. This has been shown over a very large range of body mass in pythons (Secor and Diamond, 1997), toads (Secor and Faulkner, 2002), and several species of fish (Beamish, 1974; Boyce and Clark, 1997; Billerbeck et al., 2000). In the rattlesnake (Crotalus horridus), however, the SDA coefficient appears to increase with body mass (Zaidan and Beaupre, 2003). This would mean that large snakes spend more energy on the digestion of a given prey than smaller snakes. The biological and physiological reasons for this pattern were not studied.Nevertheless,in general, the energeticcosts associated with digestion are independent of body mass. As discussed earlier in this section, mass specific metabolic rate is smaller in large compared to small animals. Therefore, if the absolute change in metabolism associated with digestion of a given absolute meal size is independent of body mass, large animals would be expected to have a larger factorial increase in postprandial metabolism. This may also explain why maximal rates of postprandial oxygen consumption scaled with a higher component than resting metabolic rate, although kinetic factors may have to be considered. A similar reasoning has been used to explain the differences in scaling factors between rest and exercise metabolism in amphibians and reptiles (Garland, 1984; Pough and Andrews, 1984; Garland and Else, 1987; Gatten et al., 1992).
Feeding Frequency Effects Many vertebrates experience long fasting periods that can be caused by seasonal fluctuations in food abundance or by rare encounter with suitable prey. Seasonal fluctuation in food abundance is well known in temperate areas where many endothermic and ectothermic vertebrates survive the cold periods in dormancy without eating for many months (e.g. Carey, 1993; McWilliams and Karasov, 2001; Hume et al., 2002).In tropical areas, seasonal variations in prey abundance are often linked with rainfall and many vertebrates enter aestivation during dry periods in which they fast until rainfall resumes. Many tropical and subtropical reptiles, especially larger snakes, appear to undergo long periods of fasting that last for many months even in the absence of seasonal variations (Greene, 1997; Shine et al., 1998). In snakes, the long fasting periods are partially compensated by an impressive
314
Physiological and ecological adaptations t o feeding in vertebrates
ability to ingest very large meals. Digestion of large meals after a long fasting period exerts considerable demands on the flexibility of the gastrointestinal function. The gastrointestinal (GI)organs are normally considered metabolically very active and it is generally believed that these organs contribute significantly to resting metabolism. Thus it seems advantageous to reduce the energetic costs of maintaining the GI organs during fasting periods, as long as it is possible to regain their digestive functions immediately after ingestion of prey. A number of recent studies on snakes have documented large and very rapid changes in intestinal mass and brush border transport rates after feeding (e.g. Secor and Diamond, 1995, 1998, 2000; Starck and Beese, 2001, 2002). The phenotypic flexibility of the GI system of snakes provides an interesting example of adaptation to an intermittent feeding regime (Piersma and Lindstrom, 1997;Secor and Diamond, 1998). Given the large structural and functional changes in GI morphology that occur soon after ingestion, it was logical to propose that this growth accounts for a large portion of the SDA response (e.g.Secor and Diamond, 1995,1997). To investigate this possibility, Secor and Diamond (2000)compared metabolic and morphologcalchanges during the postprandial period of four frequently feeding species of snakes with four species that feed infrequently. They concluded that infrequent feeders digest slower but have a more pronounced metabolic response, which reflects the larger functional and morphological gastrointestinal changes. Thus it was concluded that the relatively high SDA coefficient of infrequently feeding species reflects energetic costs associated with rebuilding form and function. However, other digestive processes, such as secretion, production of enzymes and acid, and up-regulation of brush-border transporters could also contribute (see also Secor, 2003). To test the contribution of intestinal remodeling in pythons, Overgaard et al. (2002)studied the effects of the previous fasting duration on SDA response. Upon feeding, intestinal mass and function remain elevated for many days. Thus, if intestinal remodeling is energetically expensive, then a second meal, ingested while intestinal function is still elevated, should elicit a smaller SDA response. Overgaard et al. (2002)showed that the SDA coefficient does not change with a fasting durationbetween 60 to 3 days, and it was concluded that intestinal growth does not constitute a major contributor to SDA response. In a similar experiment, Iglesias et al. (2003)found that the SDA coefficient of the skink, Eulamprus quoyii, feeding frequently (8.8%)did not differ from that of lizards fed with the same amount of food infrequently (9.4%). These findings are consistent with the proposal that intestinal expansion is structurally simple and energetically cheap (Starck and Beese, 2001). Indeed, Secor (2003) estimated that gastrointestinalup-regulation contributeswith only 5% of python SDA overall cost. On the other hand, the production of HC1 (and enzymes)and protein synthesiswere the processes that responded
SDA in ectothermic vertebrates
315
for the largest portion of the postprandial metabolic response of pythons (55'' and 26% respectively; Secor, 2003). A small contribution of intestinal growth was also suggested for turtles (Hailey, 1998). Ideally, size changes in GI organs should be measured concomitant with metabolic rate to test for the existence of a causal relationship between rnorphological change and energetic expenditure.Isolated evaluation of the metabolic changes experienced by the different tissues/organs involved in digestion is also a promising approach.
Temperature Effects Temperature affects virtually all physiological processes of ectothermic vertebrates and the postprandial metabolic response is no exception. Many studies have addressed behavioral selection of higher body temperatures during digestion, i.e., "the postprandial thermophilic response", but there continues to be considerable controversy about the magnitude and functional importance of this response (e.g. Huey 1982; Dorcas et al. 1997; Sievert and Andreadis 1999; Peterson et al., 1993; Regal 1966; McGinnis and Moore 1969; Touzeau and Sievert, 1993; Witters and Sievert, 2001). This may, at least in part, reflect that some studies have investigatedrelatively few animals and that laboratory studies may yield resdlts that differ from studies in natural settings. For example, the colubrid snake, Elaphe obsolete, exhibited a postprandial thermophilic response in laboratory but not in the field (BlouinDemers and Weatherhead, 2001).Nevertheless, as documented in many other species of ectothermic vertebrates, body temperature is less variable during digestion of free-ranging Elaphe (see also Dorcas et al., 1997). This is accomplished through altered behavior as digesting Elaphe were more likely to be found at the edges rather than within the forest compared to fasting individuals (Blouin-Demers and Weatherhead, 2001).Thus, due to complex interactions with the environment, more field studies are highly desirable. There are plenty of reports regarding upper and lower limits outside which animals are unable to complete digestion (Dorcas et al., 1997; Du et al., 2000; Wang et al., 2003). The ultimate consequences of failure to attain adequate body temperatures during digestion include regurgitation and/or death. On the other hand, the benefits in selecting warmer temperatures during digestion include an elevated rate of digestion and/or enhanced efficiency of digestion (Stevenson et al., 1985; Lillywhite, 1987; Hailey and Davies, 1987; Reinert, 1993;Sievert and Andreadis, 1999; Wang et al., 2003). An optimal rate of digestion has been modeled as the ratio between enzyme breakdown and rate of intestinal absorption (Logan et al., 2002). In accordance with this model, the rate of gastrointestinal motility, secretion, and absorption increase with elevated temperature (e.g. Dandrifosse, 1974; Skoczylas, 1978; Stevenson et al., 1985;Lillywhite, 1987;Hailey and Davies, 1987; Reinert, 1993; Sievert and Andreadis, 1999; Wang et al., 2003) and some digestive enzymes have maximal activity at rather high temperatures
316
Physiological and ecological adaptations t o feeding i n vertebrates
in lizards (Licht, 1964).As examples of the direct effects of temperature, gut motility increases with temperature in vivo in the lizards Varanus and Ctenosaura (Mackay, 1968) as well as in Caiman (Diefenbach, 1975a).Also, secretion rate of gastric acid increases with temperature in Caiman (Diefenbach, 1975a,b)and similar effects have been documented in the snake Natrix, which also increase the secretion of digestive enzymes at high temperatures (Skoczylas, 1970a,b).Temperature effects on the dynamics of intestinal transporters have not been investigated. In almost all studies to date, elevated temperature led to shortening of SDA response while the maximal rates of oxygen consumption were higher (Hailey and Davies, 1987;Wang et al., 2003; Toledo et al., 2003). However, in the plaice (Pleuronectesplatessa),maximal oxygen consumption following a meal does not change with temperature (Joblingand Davies, 1980).As the higher rate of oxygen consumption is balanced by a shorter duration of SDA response, the energy expenditure associated with digestion, the SDA coefficient, does not change markedly with temperature (Wanget al., 2003).Thus, in Python the SDA coefficientwas virtually unaffected between 20 and 35°C (Wang et al., 2003) while it decreased somewhat when temperature was increased from 25 to 30°C in Boa (Toledo et al., 2003). On the other hand, Powell et al. (1999)noted a trend toward increasing SDA coefficient with higher body temperatures in the horned frog, Ceratophrys cranwelli. In the Atlantic silverside, Menidia menidia, Billerbeck et al. (2000) found that the fish spent 23% more energy when digesting at 28°C vs 17°C. The postprandial peak in metabolism and its magnitude are usually delayed at low temperatures (Hailey and Davies, 1987; Wang et al., 2003; Toledo et al., 2003). However, as prefeeding metabolism is also affected by temperature, the SDA factorial scope is, in general, not affected by temperature (Jobling and Davies, 1980;Toledo et al., 2003).However, if the metabolic machinery is working close to maximal capacity, the factorial scope may be reduced. For example, in Boa the factorial scope was temperature independent during digestion of small meals while the metabolic increment following large meals was larger at 25°C vs to 30°C (Toledo et al., 2003). This may indicate that metabolic demand imposed by the digestion of exceedingly large meals (40%) in combination with elevated body temperature (30°C) approached the maximal oxygen transport capacity of the Boa's cardiorespiratory system (Toledo et al., 2003). Animals that ingest large meals, such as many snakes, may have their locomotor and defensive abilities temporarily impaired during digestion (Garland and Arnold, 1983; Ford and Shuttlesworth, 1986). In that case, a faster digestion at elevated temperature may serve to reduce predation risks and allow for better mobility (Pauly and Benard, 2002). In addition, a faster rate of digestion may enable larger food intake, which would enhance body condition, facilitate growth and, possibly, increase fitness (e.g. Du et al., 2000).
SDA in ectothermic vertebrates
INTERACTION OF SDA AND OTHER METABOLICALLY DEMANDING ACTIVITIES Under natural conditions, animals often need to perform multiple tasks at the same time. When the metabolic demands arising from different activities, such as exercise, digestion, and thermoregulation, occur simultaneously, it may be important to exert priorities so that one activity can be emphasized at the expense of an other activity. However, in some cases the demands placed by the concurrent activities may be additive so that both tasks can be performed simultaneously. Digestion and exercise are likely to occur at the same time. Jobling (1981) observed that maximal oxygen uptake during swimming was greater in postprandial fish than in fasting animals and that the difference increased with meal size. He therefore suggested that postprandial metabolism is limited by cellular metabolism rather than the capacity of the cardiorespiratorysystem to transport oxygen. Nevertheless, this may not be a general pattern as some species of fish do not exhibit this additive effect between exercise and digestion and other fishes appear to attain maximal rates of oxygen consumption even after small meals (Vahl and Davenport, 1979; Boyce and Clarke, 1997). In the largemouth bass, Micropterus salmoides, exercise did not affect the SDA coefficient or SDA duration in fishes forced to swim at 1.4 to 2.5 BL"' after eating 4% of their body mass (Beamish, 1974). In toads, maximal oxygen uptake was not affected by digestive state (Andersen and Wang, 2003).The interaction between exercise and digestion has also been studied in pythons and varanid lizards (Secor et al., 2000; Bennett and Hicks, 2001). In both species the metabolic response to exercise and digestion were virtually additive so that maximal metabolic rates during exercise were larger for postprandial animals than for fasting animals. Interestingly, when digesting pythons exercised there was a marked increase in total ventilation but not in cardiac output. In Varanus, however, total ventilation during exercise was not significantly affected by digestion. The additivity of the responses also indicates that the cardiorespiratory system is unlikely to constrain metabolic rate during exercise of fasting lizards or snakes. It is possible that oxygen diffusion and/or aerobic capacity of muscles are the limiting factors (see Secor et al., 2000). Although the examples from pythons and Varanus may suggest that metabolic demand associated to SDA and exercise always occurs additively, it is possible that other proximal factors, e.g. prey size, temperature, may affect this interaction.In the colubrid snake Natrix maura, there is an indication that exercise and digestive metabolic demands are not additive (Hailey and Davies, 1987). Finally, the interaction between SDA and other metabolic demands may vary according to the nature of these other demands and/or with SDA duration and magnitude. For example, in tegu lizards, Tupinambis merianae, which exhibit a marked cycle of circadian variation in metabolism, the interaction between the metabolic demands
318
Physiological and ecological adaptations t o feeding in vertebrates
arisen from this circadian changes with that arisen from meal digestion was found to vary as a function of digestion duration. On the two immediate days following feeding, priority was given to digestion and circadian changes in metabolism were abolished. On the third and subsequent days postprandial, the metabolic demands were clearly added to each other and circadian changes in metabolism were resumed. This response seems to be a regulated response of the animal, which becomes less active after food ingestion,rather than an inability of the respiratory system to support simultaneous demands at the first two days (W. Klein, unpublished data). The conceptual framework proposed by Bennett and Hicks (2001)to describe the interaction of conflicting metabolic demands includes the two most probable modes in which such interaction may occur in ectotherm vertebrates, i. e. priorization and additivity. Nonetheless, an alternative not considered by Bennett and Hicks (2001), and usually not considered for ectotherms,is "substitution", meaning that the heat generated by one activity is transferred to another. The classical example would be use of SDA or exerciseheat for therrnoregulationby endotherrns that might prevent the use of extra energy for this purpose (Costa and Kooyman, 1984).For ectotherms, it is generally assumed that the heat generated during digestion is insufficient for thermal regulation (Hailey and Davies, 1987),However, increased body temperature speeds digestion and may result in ecological and energetic benefits. Is it possible that metabolically derived heat during SDA bears the same effects on the digestive process as those caused by postprandial thermophilic response? In a recent study, Tattersall et al. (2004)documented a significant increase in body temperature in the South American rattlesnake, Crotalus durissus, due to thermogenesis. They also found that heat generation during rattlesnake digestion was meal size dependent, and that the amount of heat generated may be ecologicallyrelevant for improvement of the snake's digestion (see also Benedict, 1932; Van Mierop and Barnard, 1976;Marcellini and Peters, 1982).Bennett et al. (2000) found that metabolic rate of Varanus increased three-four times during the digestive process but that body temperature increased less than 1°C. Since many activities other than exercise take place while the animals are digesting (pregnancy,thermoregulation, etc.), there is a clear need to study these interactions and interprets their consequences in an ecologically relevant context.
Considerableadvances have been made in recent years regarding the physiological, morphological, and endocrine consequencesof feeding in ectothermic vertebrates (Busk et al., 2000; Holrriberg et al., 2003; Wang et al., 2001a, b; Overgaard and Wang, 2002; Starck and Beese, 2001, 2002; Conlon et al.,
SDA in ectothermic vertebrates
319
1997a, b; White et al., 1999; Secor et al., 2000,2001). However, it seems that further investigations of the metabolic and energetic consequenceswould be interesting.Many aspects regarding the metabolic consequences of feeding are relatively well known for fishes but very little information is available for amphibians (e.g. Wang et al., 1995; Powell et. al., 1999; Busk et al., 2000). Within reptiles, most attention has been placed on snakes to the neglect of other groups. Matrix filling experiments would thus be very welcome to facilitate correlations of SDA responses with habitats, feeding habits, and taxonomic/phylogenetic groupings. At a more mechanistic level, it is desirable to investigate neural and humoral control mechanisms underlying activation of the postprandial metabolic response. Acknowledgments
Denis V. Andrade was supported by a Jovem Pesquisador .FAPESP grant. Augusto S. Abe was supported by a CNPq grants (523728/95-6). Tobias Wang was supported by The Danish Research Council. We are grateful to Sirnone P. Brito and Marshall McCue for kindly allowing us access to unpublished material. REFERENCES Andersen J. B. and Wang T. 2003. Cardio-respiratory consequences of forced activity and digestion in toads. Physiol. Biochem. Zool. 76: 459-470. Andrade D. V., Cruz-Neto A. P., and Abe A. S. 1997. Meal size and specific dynamic action in the rattlesnake Crotalus durissus (Serpentes: Viperidae). Herpetologica 53: 485-493. Ashworth A. 1969. Metabolic rates during recovery from protein-calorie malnutrition: The need for a new concept of Specific Dynamic Action. Nature 223: 407-409. Beamish F. W. H. 1974. Apparent specific dynamic action of largemouth bass, Micropterus salmoides. J. Fish. Res. Bd Can. 31: 1763-1769. Beck D. D. 1996. Effects of feeding on body temperatures of rattlesnakes: a field experiment. Physiol. Zool. 69: 1442-1455. Benedict F. G. 1932. T h e Physiology of Large Reptiles w i t h Special Reference to the Heat Production of Snakes, Tortoises, Lizards, and Alligators. Carnegie Inst. Publ., Washington, DC. Bennett A. F. and Hicks J. W. 2001. Postprandial exercise: priorization or addivity of the metabolic responses? J. Exp. Biol. 204: 2127-2132. Bennett A. F., Hicks J. W., and Cullum A. J. 2000. An experimental test of the thermoregulator~hypothesis for the evolution of endothermy. Evolution 54: 1768-1773. Billerbeck J.M., Schultz E. T. and Conover D.O. 2000. Adaptive variation in energy acquisition and allocation among latitudinal populations of the Atlantic silverside. Oecologia, 122: 210-219. Blouin-Demers G. and Weatherhead P.J., 2001. An experimental test of the link between foraging, habitat selection and thermoregulation in black rat snakes, Elaphe obsoleta obsoleta. J. A n i m . Ecol. 70: 1006-1013. Boyce S. J. and Clarke A. 1997. Effect of body size and ration on specific dynamic action in the Antarctic plunderfish, Harpagifer antarcticus Nybelin 1947. Physiol. Zool. 70: 679-690. Brown C.R. and Cameron J.N. 1991a. The relationship between specific dynamic action (SDA) and protein synthesis rates in the channel catfish. Physiol. Zool. 64: 298-309.
320
Physiological and ecological adaptations t o feeding i n vertebrates
Brown C.R. and Cameron J.N. 1991b. The induction of specific dynamic action in channel catfish by infusion of essencial amino acids. Physiol. Zool. 64: 276-297. Busk M., Jensen F. B and Wang T. 2000. Effects of feeding on metabolism, gas transport, and acid-base balance in the bullfrog Rana catesbeiana. Am. J. Physiol. 278: R185-R195. Canjani C., Andrade D. V., Cruz-Neto A., and Abe A. S. 2003. Aerobic metabolism during predation by a boid snake. Comp. Biochem. Physiol. 133A: 487-198. Carey H. V. 1989. Seasonal variation in intestinal transport in ground squirrels. In: Living in the Cold, vol. II. A. Malan and B. Canguilhem (eds.). John Libbey Eurotext, London, pp. 225-233. Carey H. V. 1993. Regulation of gut structure and function in hibernators. In: Life in the Cold: Ecological, Physiological and Molecular Mechanisms. C. Carey, G. L. Florant, B. A. Wunder, and B. Horwitz (eds.). Westview Press, Boulder, CO (USA), yp. 155-165. Chappell M.A. and Ellis T.M. 1987. Resting metabolic rates in boid snakes: Allometric relationships and temperature effects. J. Comp. Physiol. 157: 227-235. Conlon J. M., Adrian T. E., and Secor S. M. 1997a. Tachykinins (substance P, neurokinin A and neuropeptide g) and neurotensin from the intestine of the Burmese python, Python molurus. Peptides 18: 1505-1510. Conlon J. M., Secor S. M., Adrian T. E., Mynarcik D . C., and Whittaker J. 1997b. Purification and characterization of islet hormones (insulin, glucagons, pancreatic polypetide and somatostatin) from the Burmese python, Python molurus. Regul. Peptides 71: 191-198. Costa D . P. and Kooyman G. L. 1984. Contribution of specific dynamic action to heat balance and thermoregulation in the sea otter, Enhydra lutris. Physiol. Zool. 57: 199-203. Coulson R. A. 1997. W h y David Was a Threat to Goliath: Engineering Limitations to Enzyme Kinefics in Vivo. Vantage Press, New York, NY. Coulson R. A. and Hernandez, T. 1971. Catabolic effects of cyclohexamide in the living reptile. Comp. Biochem. Physiol. 408: 741-749. Coulson R. A. and Hernandez T. 1979. Increase in metabolic rate of the alligator fed proteins or amino acids. J. Nutr. 109: 538-550. Coulson R.A. and Hernandez T. 1983. Alligator metabolism: studies on chemical reactions in vivo. Comp. Biochem. Physiol. 74: 1-182. Cowles E. R. and Bogert C. M. 1944. Thermophilic response following feeding in certain reptiles. Copeia 1944: 588-590. Cruz-Neto A. P., Andrade D. V., and Abe A. S. 2001. Energetic and physiological consequences of feeding in amphibians and reptiles. Comp. Biochem. Physiol. 128A: 515-533. Dandrifosse G. 1974. Digestion in reptiles. In: Amphibia and Reptilia. M. Florkin and B. Scheer (eds.). Acad. Press, New York, NY, vol. 9, pp. 249-276. Diefenbach C. 0. 1975a. Gastric function in Caiman crocodilus (Crocodylia: Reptilia). I. Rate of gastric digestion and gastric motility as a function of temperature. Comp. Biochem. Physiol. 51A: 259-265. Diefenbach C. 0. 1975b. Gastric function in Caiman crocodilus (Crocodylia: Reptilia). IT. Effects of temperature on pH and proteolysis. Comp. Biochem. Physiol. 51A: 267-274. Dorcas M. E., Peterson C. R. and. Flint M. E. T. 1997. The thermal biology of digestion in rubber boas (Charina bottae): physiology, behavior and environmental constraints. Physiol. Zool. 70: 292-300. Du W. G. Yan S. J. and Ji X. 2000. Selected body temperature, thermal tolerance and thermal performance in adult blue-tailed skinks, Eumeces elegans. J. Therm. Biol. 25: 197-202. Ferry-Graham L.A. and Gibb A. C. 2001. Comparison of fasting and postfeeding metabolic rates in a sedentary shark, Cephaloscyllium ventriosum.Copeia 2001: 1108-1113. Ford N. B. and Shuttlesworth G. A. 1986. Effects of variation in food intake on the locomotory performance of juvenile garter snakes. Copeia 1986: 999-1001. Garland T. 1984. Physiological correlates of locomotory performance in a lizard: an allometric approach. A m . J. Physiol. 247: R806-R815.
SDA in ectothermic vertebrates
321
Garland T. and Arnold S. J. 1983. Effects of a full stomach on locomotory performance of juvenile garter snakes (Thamnophis sirtalis). Copeia 1983: 1092-1096. Garland T. and Else P.L. 1987. Seasonal, sexual, and individual variation in endurance and activity metabolism in lizards. Amer. J. Physiol. 252: R439-R449. Garnett S. 1988. Digestion, assimilation and metabolism of captive estuarine crocodiles, Crocodylus porosus. Comp. Biochem. Physiol. 90A: 23-29. Gatten R. E., Miller K., and Full R. J. 1992. Energetics at rest and during locomotion. In: Environmental Physiology of the Amphibians M . E. Feder and W. W. Burggren (eds.). Univ. Chicago Press, Chicago, IL (USA), pp. 314-377. Gillooly J.F., Brown J.H., West G.B., Savage V.M., and Charnov E.L. 2001. Effects of size and temperature on metabolic rate. Science 293: 2248-2251. Greene H.W. 1997. Snakes: The Evolution of Mystery i n Nature. Univ. California Press, Berkeley, CA. Greenwald 0.E. and Kanter, M. E. 1979. The effects of temperature and behavioral thermoregulation on digestive efficiency and rate in corn snakes (Elaphe guttata guttata). Physiol. Zool. 52: 398-408. Hailey A. 1998. The specific dynamic action of the omnivorous tortoise Kinixys spekii in relation to diet, feeding pattern, and gut passage. Physiol. Zool. 71: 57-66. Hailey A. and Davies P. M. C. 1987. Digestion, specific dynamic action, and ecological energetics of Natrix maura. Herpetol. J. 1: 159-166. Harvey P. H. and Page1 M. D. 1993. The Comparative Method in Evolutionary Biology. Oxford Univ. Press, Oxford, UK, 239 pp. Hemmingsen, A. M. (1969). Energy metabolism as related to body size and respiratory surfaces, and its evolution. Rept. Steno. Mem. Hosp. Nordisk Insulinlaboratorium 9: 1-110. Herbert, T. and Coulson, R. A. (1976). Plasma amino acids in reptiles after feeding protein or amino acids, and after injecting amino acids. J. Nutr. 106, 1097-1101. Heusner A.A. (1982). Energy metabolism and body size. I. Is the 0.75 mass exponent of Kleiber's equation a statistical artifact? Respir. Physiol. 48: 1-12. Holmberg A., Kaim J., Persson A., Jensen J., Wang T., and Holmgren S. 2003. Effects of digestive status on the reptilian gut. Comp. Biochem. Physiol. 133A: 499-518. Huey R.B. 1982. Temperature, physiology, and the ecology of reptiles. In. Biology of the Reptilia, Physiology C , Physiological Ecology. C. Gans and F.H. Pough (eds.). Acad. Press, Ithaca, NY, pp. 25-74. Hume I. D., Bieglbock C., Ruf T., Frey-Ross F., Bruns U. and Arnold W. 2002. Seasonal changes in morphology and function of the gastrointestinal tract of free-living alpine marmots (Marmota marmota). J. Comp. Physiol. B 172: 197-207. Iglesias S., Thompson M. B., and Seebacher F. 2003. Energetic cost of a meal in a frequent feeding lizard. Comp. Biochem. Physiol. 135: 377-382. Jobling M. 1981. The influences feeding on the metabolic rate of fishes: a short review. J. Fish Biol. 18: 385-400. Jobling M. 1983. Towards an explanation of specific dynamic action (SDA). J. Fish. Biol. 23: 549-555. Jobling M. and Davies P. S. 1980. Effects of feeding on metabolic rate, and the specific dynamic action in plaice, Pleuronectes platessa L. J. Fish Biol. 16: 629-638. Kleiber M. 1961. The Fire of Life: A n Introduction to Animal Energetics. Wiley & Sons, New York, NY. Konarzewski M. and Starck J.M. 2000. Effects of food shortage and oversupply on energy utilization, histology, and function of the gut in nestling song thrushes (Turdus philomelos). Physiol. Biochem. Zool. 73: 416-427. Lang J. W. 1979. Thermophilic response of the American alligator and the American crocodile to feeding. Copeia 1979: 48-59. Licht P. 1964. The temperature dependence of myosin adenosine triphosphate and alkaline phosphatase in lizards. Comp. Biochem. Physiol. 12: 331-341.
322
Physiological and ecological adaptations t o feeding in vertebrates
Lillywhite H. B. 1987. Temperature, energetics, and physiological ecology. In: Snakes: Ecology and Evolutionary Biology. R. A. Siegel, J. T. Collins and S. S. Novak (eds.). New York, McGraw-Hill, NY, pp. 422477. Logan J. D., Joern A., and Wolesensky W. 2002. Location, time, and temperature dependence of digestion in simple animal tracts. J. Theor. Biol. 216: 5-18. Mackay R.S. 1968. Observations on the peristaltic activity versus temperature and circadian rhythms in undisturbed Varanus flavescens and Ctenosaura pectininata. Copeia, 1968: 252-259. Marcellini D.L. and Peters A. 1982. Preliminary observations on endogenous heat production after feeding in Python molurus. J. Herpetol. 16: 92-95. McCue M. D. and Lillywhite H. B. 2002. Oxygen consumption and the energetics of island-dwelling Florida cottonmouth snakes. Physiol. Biochem. Zool. 75: 165-178. McCue M. D., Bennett A. F and Hicks, J. W. 2005. The effects of meal composition on specific dynamic action in Burmese pythons (Python molurus). Physiol. Biochem. Zool. 78 (in press). McGinnis S. M. and Moore R. G. 1969. Thermoregulation in the boa constrictor Boa constrictor. Herpetologica 25: 3845. McWilliams S . R. and Karasov W. H. 2001. Phenotypic flexibility in digestive system structure and function in migratory birds and its ecological significance. Comp. Biochem. Physiol. 128A: 577-593. Overgaard J., and Wang T. 2002. Increased blood oxygen affinity during digestion in the snake Python molurus. J. Exp. Biol. 208: 3327-3334. Overgaard J., Andersen J. B., and Wang T. 2002. The effects of fasting duration on the metabolic response to feeding in Python molurus: an evaluation of the energetic costs associated with gastrointestinal growth and upregulation. Physiol. Biochem. Zool. 75: 360-368. Overgaard J., Busk M., Hicks J. W., Jensen F. B. and Wang T. 1999. Respiratory consequences of feeding in the snake Python molurus. Comp. Biochem. Physiol. 124: 359-365. Pandian T.J. 1987. Fish. In: Animal Energetics T. J. Pandian and F. J. Vemberg (eds.). Acad. Press, San Diego, C.A., pp. 358465. Pauly G.B. and Benard M.F. 2002. Crotalus viridis oreganos (Northern Pacific Rattlesnake). Costs of feeding. Herpetol. Rev. 33: 56-57. Peck L. S. 1998. Feeding, metabolism and metabolic scope in Antarctic ectotherms. In: Cold Ocean Physiology. H . 0 . Portner and R. C. Playle (eds.). Cambridge Univ. Press, Cambridge, UK, pp. 365-390. Peck L. S. and Veal R. 2001. Feeding, metabolism and growth in the Antarctic limpet, Nacella concinna (Strebel 1908). Marine Biology 138: 553-560. Peterson C.R., Gibson A.R., and Dorcas M.E. 1993. Snake thermal ecology: The causes and consequences of body-temperature variation. In: Snakes: Ecology and Behavior. R.A. Seigel and J.T. Collins (eds.). McGraw-Hill, New York, NY. pp. 241-314. Piersma T. and Lindstrom A. 1997. Rapid reversible changes in organ size as a component of adaptive behavior. Trends Ecol. Evol. 12: 134-138. Pough F.H. 1983. Amphibians and reptiles as low-energy systems. In: Behavioral Energetics: T h e Cost of S u r v i v a l i n Vertebrates. W.P. Aspey. and S.I.Lustick S.I. (eds.). Co1umbus:Ohio State Univ. Press, Columbus, OH (USA), pp. 141-188. Pough F. H. and Andrews R. M. 1984. Individual and sibling group variation in metabolism of lizards: the aerobic capacity model for the origin of endothermy. Comp. Biochem. Physiol. 79A: 415-419. Powell M. K., Mansfield-Jones J. and Gatten Jr., R. E. 1999. Specific dynamic effect in the horned frog Ceratophys cranwelli. Copeia 1999: 710-717. Regal P. J. 1966. Thermophilic response following feeding in certain reptiles. Copeia 1966: 588-590. Reinert H. K. 1993. Habitat selection in snakes. In: Snakes: Ecology and Behavior. R. A. Siegel and J. T. Collins (eds.). McGraw-Hill Book Co., New York, NY, pp. 201-240.
SDA in ectothermic vertebrates
323
Rodriguez-Robles J. A., and Thomas R. 1992. Venom function in the puertorican racer, Alsophis portoricensis (Serpentes: Colubridae). Copeia. 1992: 62-68. Rubner M. 1902. Die Gesetze des Energieverbrauchs bei der Erniihrung. Deuticke Leipzig, Germany, 363 pp. Secor S. M. 1995. Digestive response to the first meal in hatchling Burmese pythons (Python molurus). Copeia 1995(4): 947-954. Secor S. M. 2001. Regulation of digestive performance: a proposed adaptive response. Comp. Biochem. Physiol. 128: 565-577. Secor S. M. 2003. Gastric function and its contribution to the post-prandial metabolic response of the Burmese python Python molurus. J. Exp. Biol. 206: 1621-1630. Secor S. M. and Nagy K. A. 1994. Bioenergetic correlates of foraging mode for the snakes Crotalus cerastes and Masticophis flagellum. Ecology. 75: 1600-1614. Secor S. M. and Diamond J. 1995. Adaptive responses to feeding in Burmese pythons: Pay before pumping. J. Exp. Biol. 198: 1313-1325. Secor S. M. and Diamond J. 1997. Determinants of the postfeeding metabolic response of Burmese pythons, Python molurus. Physiol. Zool. 70: 202-212. Secor S. M. and Phillips J. A. 1997. Specific dynamic action of a large carnivorous lizard, Varanus albigularis. Comp. Biochem. Physiol. 117A: 515-522. Secor S. M. and Diamond J . 1998. A vertebrate model of extreme physiological regulation. Nature 395: 659-662. Secor S. M. and Diamond J. 2000. Evolution of regulatory responses to feeding in snakes. Physiol. Biochem. Zool. 73: 123-141. Secor S. M. and Faulkner A. C. 2002. Effects of meal size, meal type, body temperature, and body size on the specific dynamic action of the marine toad, Bufo marinus. Physiol. Biochem. Zool. 75: 557-571. Secor S. M., Hicks J . W. and Bennett A. F. 2000. Ventilatory and cardiovascular responses of a python (Python molurus) to exercise and digestion. J. Exp. Biol. 203: 2447-2454. Secor S. M., Fehsenfeld D., Diamond J. and Adrian T. E. 2001. Responses of python gastrointestinal regulatory peptides to feeding. Proc. Natl. Acad. Sci. U S A 98: 13637-13642. Secor S. M., Whang E. E., Lane J. S., Ashley S. and Diamond J. 2000. Luminal and systemic signals trigger intestinal adaptation in the juvenile python. A m . J. Physiol. 279: G1177-G1187. Shine R., Harlow P. S., Keogh J. S., and Boeadi. 1998. The allometry of life-history traits: insights from a study of giant snakes (Python reticulatus). J. Zool. Lond. 244: 405414. Sievert L. M. 1989. Postprandial temperature selection in Crotaphytus collaris. Copeia 1989: 987-993. Sievert L. M. and Andreadis P. 1999. Specific dynamic action and postprandial thermophily in juvenile northern water snakes, Nerodia sipedon. J. Therm. Biol. 24: 51-55. Sievert L. M., Sievert G. A., and Cupp jr., P. V. 1988. Metabolic rate of feeding and fasting juvenile midland painted turtles, Chrysemys picta marginata. Comp. Biochem. Physiol., 90A: 157-159. Sims D. W. and Davies D.J. 1994. Does specific dynamic action (SDA) regulate return of appetite in the lesser spotted dogfish, Scyliorhius canicula? J. Fish Biol. 45: 341-348. Skoczylas R. 1970a. Influence of temperature on gastric digestion in the grass snake, Natrix natrix L. Comp. Biochem. Physiol. 33: 793-803. Skoczylas R. 1970b. Salivary and gastric juice secretion in the grass snake, Natrix natrix L. Cornp. Biochem. Physiol. 35: 885-903. Skoczylas R. 1978. Physiology of the digestive tract. In: Biology of the Reptilia, vol. 8 Gans, C., Gans, K.A.8 (eds.). Acad. Press, (NY), Pp. 589-717. Somanath B., Palavesam A., Lazarus S., and Ayyapan M. 2000. Influence of nutrient source on specific dynamic action of pearl spot, Etroplus suratensis (Bloch). Naga 23: 15-17.
324
Physiological and ecological adaptations t o feeding in vertebrates
Somero G.N. and Childress J.J. 1980. A violation of the metabolism-size scaling paradigm: Activities of glycolytic enzymes in muscle increase in larger-size fish. Physiol. ZOO^. 53: 322-337. Soofiani N. M. and Hawkins A. D. 1982. Energetic costs at different levels of feeding in juvenile cod, Gadus morhua L. J. Fish. Biol. 21: 577-592. Starck J.M. 1999a. Structural flexibility of the gastro-intestinal tract of vertebrates implications for evolutionary morphology. 2001.Anz. 238: 87-101. Starck J.M. 1999b. Phenotypic flexibility of the avian gizzard: rapid, reversible and repeated changes of organ size in response to changes in dietary fibre content. J. Exp. Biol. 202: 3171-3179. Starck J. M. and Beese K. (2001). Structural flexibility of the intestine of burmese python in response to feeding. I. Exp. Biol. 204: 325-335. Starck J. M., and Beese K. 2002. Structural flexibility of the small intestine and liver of garter snakes in response to feeding and fasting. 1. Exp. Biol. 205: 1377-1388. Starck J. M., Moser P, Werner R. and Linke P. 2004. Pythons metabolize prey to fuel the response to feeding. Proc. Roy. Soc. London B 271: 903-908. Stevenson R.D., Peterson C.R., and Tsuji J.S. 1985. The thermal dependence of locomotion, tongue flicking, digestion, and oxygen consumption in the wandering garter snake. Physiol. Zool. 58: 46-57. Taigen T.L. 1983. Activity metabolism of anuran amphibians: implications for the origin of endothermy. Amer. Nat. 121: 94-109. Tattersall G., Milsom W. K., Abe A. S., Brito S. P., and Andrade D. V. 2004. The thermogenesis of digestion in rattlesnakes. J. Exp. Biol. 207: 579-585. Thomas R. G. and Pough E H. 1979. The effect of rattlesnake venom on digestion of prey. Toxicon 17: 221-228. Thompson G.G., and Withers RC. 1994. Standard metabolic rates of two small australian varanid lizards (Varanus caudolineatus and V. acanthurus). Herpetologica 50: 494-502. Toledo L. E, Abe A. S., and Andrade D. V. 2003. Temperature and meal mass effects on the post-prandial metabolism and energetics in a boid snake. Physiol. Biochem. Zool. 76: 240-246. Touzeau T, and Sievert L.M. 1993. Postprandial thermophily in rough green snakes (Opheodrys aestivus). Copeia 1993: 1174-1176. Vahl 0. and Davenport J. 1979. Apparent specific dynamic action of food in the fish Blennius pholis. Mar. Ecol. Prog. Ser. 1: 109-113. Van Mierop L.H.S., and Barnard S. M. 1976. Thermoregulation in a brooding female Python molurus bivittatus. Copeia, 1976: 398401. Wang T., Burggren W. W., and Nobrega E. 1995. Metabolic, ventilatory, and acid-base responses associated with specific dynamic action in the toad Bufo marinus. Physiol. ZOO^. 68: 192-205. Wang T., Busk M. and Overgaard J. 2001a. The respiratory consequences of feeding in amphibians and reptiles. Comp. Biochem. Physiol. 128: 535-549. Wang T., Taylor E. W., Andrade D. V., and Abe A. S. 2001b. Autonomic control of heart rate during forced activity and digestion in the snake Boa constrictor. J. Exp. Biol. 204: 3553-3560. Wang T., Zaar M., Arvedsen S., Vedel-Smith C., and Overgaard J. 2003. Effects of temperature on the metabolic response to feeding in Python molurus. Comp. Biochem. Physiol. 133A: 519-527. White A. M., Secor S. M. and Conlon J. M. 1999. Insulin and proglucagon-derived peptides from the horned-frog, Ceratophrys ornata (Anura: Leptodactylidae). General and Comparative Endocrinology 115: 143-154. Withers !F C. 1992. Comparative Animal Physiology. Saunders College Publ., Orlando, FL, 950 pp. Witters L.R. and Sievert L. 2001. Feeding causes thermophily in the Woodhouse's toad (Bufo woodhousii). 1. Therm. Biol. 26: 205-208. Zaidan E and Beaupre S. J. 2003. Effects of body mass, meal size, fast length, and temperature on specific dynamic action in the timber rattlesnake (Crotalus horridus). Physiol. Biochem. Zool. 76: 447-458.
Control of Gut Motility and Secretion in Fasting and Fed Nonmammalian Vertebrates Susanne Holmgren and Anna Holmberg Goteborg University, Department of Zoophysiology, Gijteborg, Sweden
SYNOPSIS Feeding induces several patterns of activity in the gut to process the food in an optimal manner, including increased motility and secretion of acid, enzymes, and mucus. Control of these processes is exerted mainly by the autonomic nervous system and local hormones. This chapter summarizes current knowledge on the differences and similarities between vertebrate groups in the neuronal and hormonal activation of motility and secretion on feeding. There are striking similarities in the anatomy of the gut wall and the enteric nervous system among vertebrates. Extrinsic nerves to the gut include parasympathetic and sympathetic pathways. Interstitial cells of Cajal initiate muscle activity and mediate nervous control. Hormones and neurotransmitters affecting gut motility and secretion are numerous; special emphasis is given here to cholecystokinin, gastrinreleasing peptide (and bombesin), and ghrelin, which additionally affect satiety or hunger. Particular attention is paid to what is known to date of the integrated control of gut motility and secretion. In the interdigestivestage, there is a low level of basic secretion and propagating contractions sweep the gut at regular intervals. Feeding increases secretion of gastric acid from a basal level and motility patterns such as gastric receptive relaxation, gastric emptying, and peristalsis are initiated. Changes in innervation around onset of feeding and during prolonged fasting are also discussed.
326
Physiological and ecological adaptations t o feeding in vertebrates
Feeding induces several patterns of activity in the gut to process the food in an optimal manner. This includes increased motility and increased secretion of acid, enzymes and mucus. In the interdigestive stage, other patterns of motility and a low basal level of secretion prevail. In the wild, several species of vertebrates experience periods of prolonged fasting due to seasonal variations in food abundance or simply intermittent detection/capture of prey. This puts further demands on the functions of the gut in order to balance the cost of an upregulated effective digestion versus energy conservation achieved by down-regulation of mechanisms during fasting periods. Control of gut motility and secretion is exerted mainly by the autonomic nervous system (ANS) and by local hormones (released from/in the gut wall). This chapter summarizes current knowledge about these mechanisms in nonmammalian species, using mammals as a reference. The aim was identificationof the differences and similaritiesamong vertebrate groups in neuronal and hormonal activation of motility and secretion on feeding. A summary of current knowledge of the development of the enteric nervous system (ENS) around the onset of feeding, and on the consequences of fasting and refeeding on the control systems of the gut is also included. To keep the number of references reasonable, the reader is referred sometimes to comprehensive recent reviews, particularly for studies of mammals. For detailed overview reviews the reader is referred to e.g. Kunze and Furness (1999; motility, mammals), Olsson and Holmgren (2001; motility, comparative), Lindstrom et al. (2001; secretion, general), and Jonsson (1994; secretion, comparative).
The gastrointestinal (GI)tract of vertebrates in general comprises an esophagus for transport of food from mouth to stomach, an acid-secreting stomach, an intestine where the main part of food digestion and absorption takes place, and a rectum where wastes are stored. Variations in this general plan occur. Some of these variations are clearly related to animal differences in feeding behaviors. For example, in certain species which capture prey that are very large compared to their own size, such as many reptiles, the esophagus is used for storage and even initial digestion of food. Some species, e.g. cyprinids (carp) and labrids (wrasses), lack a stomach; instead the anterior intestine secretes the corresponding "gastric" enzymes (but not gastric acid) (e.g.Koelz, 1992).Variations may also be related to type of food. Thus herbivores in general have a longer intestine than carnivores and, at least in birds, changes in diet composition with season may even affect the length of the intestine (e.g. Gentle and Savory, 1975).
Control of gut motility
mucosa
L
submucosa
+
circular
-
a!
..Dm
I..I I.. I
+ submucosal plexus
I
+ myenteric plexus
Fig. 13.1. A schematic and simplified drawing of the location of the enteric nervous system in the gut wall. In most vertebrate guts, two dominating ganglionated plexuses are found. The submucous plexus is situated within the submucosa and the neurons innervate mainly (solid line) vessels and glands in the inner part of the gut wall (i.e. the mucosa and submucosa). The myenteric plexus is located between the two muscle layers (i.e. circular and longitudinal). Myenteric neurons are mainly innervating the muscle layers.
Although the general plan may thus vary among animal groups due to specializations in feeding behavior and processing, the overall architecture of the gut wall is amazingly consistent among vertebrates, and from the oral to the anal end of the GI canal in a single individual (e-g. Nilsson, 1983; Furness and Costa, 1987; Olsson and Holmgren, 2001). From the lumenal side, there is a mucosa, a submucosa, a circular muscle layer, a longitudinal muscle layer, and outermost a serosal epithelial layer (Fig. 13.1).However, the thickness of the individual layers may vary considerably among species and with feeding status (see Chapter 7 by J.M. Starck and Chapter 9 by H. Carey, this volume). The mucosa is more or less folded and lined toward the lumen by a layer of epithelial cells, including a varying number of secretory cells, absorbing cells, and hormone-secreting endocrine/paracrine cells depending on the part of the gut. Secretory cells may also cluster in glands protruding into the submucosa. A thin smooth muscle layer, the muscularis mucosa, often separates the mucosa from the submucosa. The submucosa consists of connective tissue, glands, and small vessels. Except for the esophagus, where striated muscle forms at least the anterior (upper)two-thirds of the outer muscle layers, the outer muscle coat of the gut consists of smooth muscle. The muscle comprises an inner, thick layer of circularly orientated muscle cells and an outer, thinner longitudinal layer of cells. The individual smooth muscle cells are usually long, thin, and spindle shaped. Gap junctions often connect the muscle cells electrically to each
328
Physiological and ecological adaptations t o feeding i n vertebrates
other. These connections allow changes in membrane potential to spread through the muscles layer, which therefore function as an electrical syncytium (Gabella and Blundell, 1981).
CONTROL SYSTEMS OF THE GUT
The quickest and most direct control of the gut functions is performed by the autonomic nervous system (ANS),primarily by its enteric division, Hormones released from cells in the gut mucosal epithelium have both paracrine effects (i.e. local effects in the direct vicinity of the secreting cell) and endocrine effects (on other parts of the gut). In addition, nerve reflexes and hormones reaching the central nervous system (CNS)may have further effects on the gut through hormonal release from the pituitary or nervous pathways back to the gut. The gut is also influenced by circulatinghormones secreted from endocrine glands such as the pancreas and the adrenal glands, but these effects are not reported here.
Autonomic Innervation of the Gut Visceral functions such as gut motility and secretion are controlled by the ANS (Langley1898; Fig. 13.2).From studies in mammals, the ANS has been Control of gut motility
/
A
neuronal control (ANS)
GUT I.
vagus ( X )
hormonal control
circulating and gut hormones.
1I - ;.*""..q !
splanchnic;e=*(e--*-= nerves I I
parasympathetic neurons
pelvic nerves
sympatheric neurons
i CNS
enteric neurons
ICC cells smooth muscle cells
Fig. 13.2. Schematic drawing of the control systems of the gut. Neuronal (autonomic nervous system) and hormonal components control gut functions like gut motility and secretion. The gut is innervated by extrinsic (parasympathetic and sympathetic neurons) and intrinsic neurons (enteric neurons). Generally, the extrinsic neurons do not innervate the smooth muscle cells directly but synapse on enteric neurons. The enteric neurons, in turn, synapse directly on muscle fibers or on interstitial cells of Cajal (ICCs). Both circulating and local gut hormones control gut functions, directly or via an action on enteric nerves. ANS, autonomic nervous system; CNS, central nervous system; ENS, enteric nervous system.
Control of gut motility
329
divided into parasympathetic (cranial-sacraloutflow),sympathetic (thoraciclumbar or splanchnic outflow), and enteric sections (Langley, 1921). The ENS is the intrinsic nervous system of the gut, comprisingnetworks of ganglia and nerve fibers within the gut wall. Parasympathetic pathways have preganglionic neurons with long axons connecting to short postganglionic neurons in ganglia nearby or on/in the effector organ. Indeed, in the gut innervation, the postganglionic parasympathetic neuron is usually also an enteric neuron, i.e. situated within the enteric plexuses. The parasympathetic fibers to the gut run in the vagus nerve (cranial outflow) and pelvic nerves (sacral outflow).Sympatheticpathways usually have short preganglionic neurons, synapsingwith long postganglionicneurons in the paravertebral ganglia of the sympathetic chain or in large prevertebral ganglia associated with the origins of the celiac and mesenteric arteries. The preganglionic sympathetic pathways to the gut run in the splanchnic nerves to the prevertebral ganglia and from there the postganglionic fibers follow the main vessels to the gut wall. In nonmammalian species, the distinction between parasympathetic nerves and sympathetic nerves is not always clear. The vagus is usually considered a parasympathetic nerve but in all groups except reptiles, there are reports of sympathetic fibers entering the vagus from the sympathetic chains, thus forming a vagosympathetic trunk. In the pelvic region, a mixture of parasympathetic and sympathetic fibers is seen in the pelvic nerves in most nonmammalian species, posing difficulties in distinguishing between sacral parasympathetic (pelvic)and posterior sympathetic outflows. Therefore a modified classification is often used in nonmammalian species, referring to cranial autonomic outflow (mainly parasympathetic) and spinal autonomic (mixed sympathetic and parasympathetic) outflow (Nilsson, 1983). The vagus also carries sensory fibers from the anterior gut to the CNS; indeed, the majority of negve fibers in the vagi arise from sensory neurons located in the nodose ganglia. These fibers register distension of the gut or various components in the chyme such as amino acids, fatty acids, and glucose (Nilsson, 1983; Goyal and Hirano, 1996).One study in a toad species concludes that three different types of mechanoreceptors.are present in the gut wall, inducing activity in vagal fibers (Niijima, 1967).Sensory neurons from the posterior parts of the gut follow spinal nerves to the spinal cord; these fibers are thought to primarily register noxious stimuli, causing sensations of pain (Nilsson, 1983; Goyal and Hirano, 1996). Rather than innervating the muscle fibers or secretory cells directly, extrinsic autonomic neurons often synapse on enteric neurons (Furness and Costa, 1987),producing a widespread effect. Vagus In cyclostomes, the intestinalbranches of the vagus are fused into the ramus intestinalis impar, which.runs along the dorsal surface of the gut and
330
Physiological and ecological adaptations t o feeding in vertebrates
probably also innervates it (Nicol, 1952; Johnels, 1956; Pick, 1970). In all other vertebrates, the visceral branch of each vagus innervates the gut separately. Generally one or several smaller ramifications from the main visceral branch innervate the esophagus and others go to the stomach and upper intestine. Both striated and smooth muscles are innervated by vagal ramifications. Posterior extension varies with species. In some mammals the vagal fibers may reach as far as the proximal colon. In most teleost species, the vagus only innervates the upper part of the gut, i.e. esophagus and stomach (see Nilsson, 1983)but this may vary. In some stomachless fish such as Tinca tinca, the vagus innervates the entire gut (Ohnesorge and Rehberg, 1963), while in other stomachless species such as the flatfish Rhombosolea tapirina and Ammotretis rostrata, the vagus only innervates the striated muscle of the esophagus (Grove and Campbell, 1979). In contrast to mammals (Botar et al., 1950), microganglia and solitary ganglion cells are present in the rami of the vagus running to the esophagus and gut in fish (Huber, 1900; Ray, 1950).Similar, but often somewhat larger ganglia are also found along the esophago-gastric rami of the amphibian vagus (Gibbinset al., 1987).Some different neurotransmitters are present in the ganglion cells. Galanin-likeand tachykinin-likepeptides have been found in the Atlantic cod (Jensen et al., 1993; Karila et al., 1993),and the ganglion cells projecting to the toad stomach are noncholinergic, vasoactive intestinal polypeptide (VIP)-containingneurons, presumably providing an inhibitory innervation to the gut. It has been speculated from their unipolar appearance that the neurons in the amphibian vagus are true postganglionic vagal neurons, rather than displaced enteric neurons which most often are multipolar in amphibians (Gibbins, 1994).Contrarily, in the Atlantic cod most enteric neurons are unipolar (Karila et al., 1993) and thus the unipolar vagal neurons may be either displaced enteric neurons or postganglionic parasympathetic. Vagal motor neurons to the gut include both stirnulatorycholinergicpathways and inhibitory pathways, commonly VIP-containing,in all vertebrates except cyclostomes and elasmobranchs (see Nilsson, 1983).In cyclostomes, stimulation of the vagus produces at most a weak relaxation, which is probably neither adrenergic nor cholinergic (Patterson and Fair, 1933; Fange, 1948). In elasmobranchs, stimulation of vagal fibers produces inhibition only (Campbell,1975).In general, the major effect of vagal stimulation in all vertebrates is an influence on the ENS and the resultant effect may be excitatory or inhibitory depending on the species or the current feeding status of the gut. Spinal autonomic system innervating the gut There are large variations in the anatomy of the spinal autonomic system in different nonmammalian vertebrates, once the preganglionic fibers have left the spinal cord. Usually, segmental paravertebral ganglia are present, which
Control of g u t motility
331
Fig. 13.3. Microphotographs illustrating the myenteric plexus in Rana femporaria visualised by immunohistochemistry. A: Nerve fibers and cell bodies (arrow) showing nitric oxide synthase (NOS) -like immunoreactivity in the distal intestine. B: Nerve fibers stained by antibodies against substance P in the myenteric plexus of the cardiac stomach. Calibration bars: 100 pm. '
332
Physiological and ecological adaptations t o feeding in vertebrates
are interconnected to different degrees to so-called sympathetic chains in most species. Fusions of ganglia in the chains are common. Most postganglionic [sic!]fibers innervating the gut run in two or more splanchnic nerves, which innervate both the stomach and intestine in most species. Whether the posterior splanchnic nerve in fish is in fact homologous to the pelvic nerve in tetrapods has been discussed. Tetrapods have a pelvic nerve and a pelvic plexus in the sacral region, while the presence of distinct pelvic ganglia and pelvic nerves in fish could not be established with certitude (Nilsson, 1983; Gibbins, 1994). The spinal fibers may be adrenergic, cholinergic or nonadrenergic, noncholinergic (NANC)with no evolutionarytrend and the effects of nerve stimulation may consequently vary between species. A remarkable feature of the spinal autonomic innervation of the gut occurs in birds. A ganglionated nerve, Remak's nerve, runs along the gut from the pelvic ganglion through the mesentery of the intestine to the celiac ganglion. Most axons project orally but some project anally or laterally to supply the gut directly.Four types of ganglion cells with distinct morphology, distribution and projection patterns are described in chicken, and a difference in neural control of the small and large intestine is suggested (Lunam and Smith, 1996). Distension of the chicken jejunum activates intestinofugal cholinergicfibers projecting to ganglion cells in Remak's nerve, which in turn may inhibit both motility and secretion oral to the point of distension (Smith and Lunam, 1998).Both adrenergic and nonadrenergic neurons are present in Remak's nerve (Young, 1990); nonadrenergic neurons produce a noncholinergic contraction of the hind gut of chicken (Bartlet and Hassan, 1971;Kanazawa et al., 1980; Komori and Ohashi, 1982)possibly at least in part, by release of a tachykinin (Komoriet al., 1986).Acetylcholine released on stimulation of Remak's nerve has a prejunctional inhibitory effect on the release of the NANC transmitter (Komoriand Ohashi, 1984). Enteric nervous system The ENS is a nervous system intrinsic to the gut, which is found in all vertebrates. The nerve cells originate from a population of neural crest cells, which is distinct from those forming the rest of the ANS (LeLievre and Le Douarin, 1975). The ENS of mammals contains a large number of nerve cells, estimated to equal approximately the number of nerve cells of the spinal cord (Furness and Costa, 1987),and comprising sensory neurons, interneurons and motor neurons. Organization of the ENS is amazingly similar between vertebrate species and groups (Fig. 13.1).A ganglionated nerve plexus, the myenteric (or Auerbach's) plexus (Fig. 13.3),is located between the longitudinal and circular muscle layers. The submucosa contains a second more or less well-defined ganglionated nerve plexus, the submucous (or Meissner 's) plexus. The muscle layers are innervated by mostly aganglionic plexuses. The innermost part of the circular muscle is often more heavily innervated and this region is known as the deep muscular plexus. Bundles of nerve
Control o f g u t motility
333
fibers connect the ganglia to a network in the plexuses and traverse all the layers of the gut. The mucosa is innervated by sensory and secretomotor neurons with cell bodies in either of the plexuses (e.g.Nilsson, 1983;Furness and Costa, 1987; Gibbins 1994). The nurrlber and organization of neurons in the ENS vary among vertebrate groups but differencesmay also occur in individual species. In general, there appear to be larger ganglia formations in the plexuses and more distinct networks of nerve fiber bundles in amniote tetrapods (reptiles,birds, mammals) than in amphibians and fish. For example, a ganglionated myenteric plexus and a submucosalplexus are present all along the gut in chelonians (turtles), crocodiles, and the python (Hukuhara et al., 1976; Timmermans et al., 1991;Jensen and Holmgren, 1994;Holmberg et al., 2003). In cyclostomes, fish, and amphibians, the ENS seems to be less structured. Nerve cell bodies are usually scattered along the nerve bundles rather than clustered in ganglia in the nexuses of the nerve net, and existing ganglia are small (microganglia) and scarce. Few or no nerve cell bodies are found in the submucosa (Wong and Tan, 1978;Ezeasor, 1979;Gibbins, 1994). However, the density and complexity of the enteric innervation in these groups should not be underestimated. For example, the total density of nerve cell bodies in the myenteric plexus of the Atlantic cod has been calculated to be of the same magnitude as in small mammals (Gabella,1987,mammals; Olsson and Karila, 1995, Atlantic cod). Interstitial Cells of Cajal Interstitial cells of Cajal (ICCs),first described by S. R.Y. Cajal in the late 19th century, may initiate "spontaneous" rhythmic activity of the gut smooth muscle and may also mediate most of the control exerted by enteric nerves on the smooth muscle. The cells, spindle shaped or with several processes, are now known to be of nonneural, mesenchymal origin. They are coupled to each other and to smooth muscle cells, mainly via gap junctions, and receive input from entericnerve cells. ICCs express a protooncogene, c-Kit, which is a transcell membrane tyrosine kinase receptor. The c-Kit is often used as a marker in immunohistochemicalstudies to determine the presence of ICCs, and the development and functions of ICCs have been studied after genetic and pharmacological manipulation of c-Kit signaling (Horowitzet al., 1999; Young, 1999; Huizinga, 2001). In mammals, ICCs have been observed in most regions and layers of the gastrointestinaltract, although most abundantly in the circular muscle layer and myenteric plexus. Two morpholo~callyand functionally different types of ICCs have been described in the mammalian stomach: 1)ICC-IMs, which are intramuscular and may be mediators between enteric nerves and smooth muscle, and 2) ICC-MY, associated with the myenteric plexus and acting as pacemakers for rhythmic activity (Daniel,2001; Huizinga, 2001; Mitsui and Komuoro, 2002).
334
Physiological and ecological adaptations t o feeding in vertebrates
The mesenchymal origm of ICCs was first demonstrated in chicken (Lecoin et al., 1996). h birds (turkey, chicken),ICCs are frequent in the ileum, but are also present in the duodenum, cecum, and rectum. As in mammals, the ICCs are mainly located close to the myenteric plexus, and it is proposed that they are involved in generation of peristalsis (Reynhout and Duke, 1999).In the amphibian stomach, in contrast to mammals, the ICCs are mainly found in the longitudinal muscle layer (and to some extent on the border to the myenteric plexus). Removal of the longitudinal muscle abolishes spontaneous electrical activity (slow waves) in the remaining stomach wall, while the slow waves remain after removal of the circular muscle (Prosser, 1995).The significance of the alternative location in amphibians compared to mammals has not been investigated. In the lizard, Podarcis hispanica, ICCs are frequent in particular in the circular muscle interstice and in the myenteric plexus in the stomach and intestine respectively (Martinez-Ciriano et al., 2000; Junquera et al., 2001).
Gut Hormones and Neurotransmitters Affecting Motility and Secretion Control of gut motility and secretion is enormously complex in that so many gut hormones and neurotransmitters are involved and may act, interact with, and counteract each other. The chemical identity of hormones and transmitters comprises amines, amino acids, and peptides. The amines are small molecules such as acetylcholine, adrenaline, noradrenaline, serotonin (5hydroxytryptamine), and histamine. Of these, acetylcholine and probably noradrenaline are synthesized in nerves only, while the others are expressed in both endocrine cells (to be released as hormones) and gut neurons. A large number of the transmitters / hormones are peptides, often expressed in identical or similar forms in both nerves and endocrine cells. Detailed summaries of reports on the occurrence and effects of individual neurotransmitters in nonmammalian species were made by Jensen and Holmgren (1994) and Jonsson (1994). Innervation of the reptilian gut was recently reviewed by Holmberg et al. (2003).Several reviews deal with aspects on the evolution of vertebrate neuropeptides, e.g. tachykinins, the neuropeptide Y (NPY)family, and the VIP/PACAP (pituitary adenylate cyclase-activating peptide) family (Hoyle, 1998;Cerda-Reverter and Larhammar, 2000; Holmgren and Jensen, 2001), and cholecystokinin (CCK)and gastrin (Johnsen, 1998).Control of gut motility in different nonmammalian vertebrate groups by tachykinins, the VIP family, and nitric oxide, and the transmission via receptors for acetylcholine, tachykinins, serotonin, VIP/PACAP and nitric oxide is summarized by Olsson and Holmgren (2001). In this review, we concentrate on a few other peptide groups that have widespread effects on the performance of the gut and also affect food intake. Some gut hormones/neurotransrnitters, released in response to food or feeding, induce satiety in addition to their effects on gut motility and secretion.
Control of gut motility
335
Cholecystokinin and gastrin-releasingpeptide (GRP)are two such peptides, released when food is present in the gut. The effect on the central feeding system is either hormonal, or through an effect on vagal efferent endings in the gut wall. Ghrelin, on the other hand, is orexigenic,i.e. it stimulates food intake. Control of food intake in non-mammalian vertebrates by peripheral regulatory peptides and the CNS was recently reviewed by Jensen (2001). Cholecystokinin The presence of food in the duodenum causes the release of CCK from endocrine cells in the gut wall. The stimulant varies with species. Proteins and/ or amino acids on the one hand, and fat and/or fatty acids on the other hand are usually involved. Cholecystokinin may also be released from gut neurons or neurons in the CNS. In mammals, CCK induces satiety by a peripheral action on vagus nerve endings and by a central action in the hypothalamus. Cholecystokinin also has several other actions, including gallbladder contraction, secretion of pancreatic enzymes, stimulation of gastric and intestinal motility and inhibition of gastric emptying-all processes that assist in digestion of food in the intestine before more food is supplied from the stomach (e.g.Liddle, 1997).The known effects of exogenous CCK on the gut in nonmammalian vertebrates are summarized in Table 13.1. It is clear that in these species also the effects are many and various. However, most of the studies are performed using exogenous CCK and it has been seriously questioned, at least in birds, whether high enough plasma levels are reached from endogenous release of CCK to elicit the corresponding effectsby a humoral mechanism (Furuse, 1999).On the other hand, CCK/gastrin-like material occurs in myenteric gut neurons in several fislt species (Holmgren and Nilsson, 1983; Bjenning and Holmgren, 1988; Nilsson and Holmgren 1992)) and high exogenous concentrations may actually be needed to mimic the effect of endogenous release from such neurons. The effects on gut motility seem to depend on species and type of tissue. Most pharmacological studies in nonmammals have been performed in chicken, using exogenous CCK, mostly CCK8s. The results suggest an intriguing variety in the mode of action of CCK on different tissues. Cholecystokinin injected in vivo causes a relaxation of the stomach and the colon. The effect on the stomach is mediated by the vagus and by release of nitric oxide (Martinez et al., 1993;Rodriguez-Sinovas et al., 1994).At the same time, the duodenum increases its activity, probably by a direct effect of CCK on the smooth muscle (Martinez et al., 1993).The migrating myoelectric complex pattern changes from "unfed" to "fed" and may be totally disrupted after exposure to high CCK concentrations (Martinez et al., 1995b) and intralumenal pressure of the ceca is increased by direct action on the circular muscle and a mixture of direct and indirect action via serotonergic neurons on the longitudinal muscle (Martin et al. 1995).In the longitudinal smooth muscle of the ileum, CCK causes contraction in part by direct stimulation of
336
Physiological and ecological adaptations! t o feeding in vertebrates
Table 13.1. Effects of cholecystokinin on the non-mammalian gut
Stomach motility
Species
Effect
Reference
Raja sp. (elasmobranch)" Gadus morhua (teleost) Oncorhynchus mykiss (teleost) Necturus maculosus (amphibian)" Chicken and turkey
+ LM +
Andrews and Young (1988a) Jonsson et al. (1987) Olsson et al. (1999)
Intestine/ Squalus acanthias (elasmobr.) rectum motility
Raja sp. (elasmobranch)* Necturus maculosus (amphibian)" Turkey duodenum Chicken
Gastric Oncorhynchus mykiss (teleost) emptying GallRaja sp. (elasmobranch) bladder Oncorhynchus mykiss (teleost) motility Salmo salar (teleost) Lepomis macrochirus (teleost)
Fundulus heteroclitus (teleost) Amia calva (holostean) Rana catesbeiana (amphibian) Gadus morhua (teleost) Rana catesbeiana (amphibian)
Gastric acid secretion Pancreatic Oncorhynchus mykiss (teleost) Secretion
Salmo salar (teleosts)
+,+ CM
+
+ + LM
-
+ +
+ + + + + +
Holmgren et al. (1985) Savory et al. (1981); Martinez et al. (1993) Aldman et al. (1989) Andrews and Young (1988a) Holmgren et al. (1985) Savory et al. (1981) Martin et al. (1994, 1995); Martinez et al. (1993, 1995b); Rodriguez-Sinovas et al. (1994, 1996) Olsson et al. (1999) Andrews and Young (1988b) Aldman and Holmgren (1987); Aldman et al. (1992) Einarsson et al. (1997) Rajjo et al. (1988) Rajjo et al. (1988) Rajjo et al. (1988) Nielsen et al. (1998) Holstein (1982) Nielsen et al. (1998)
+, Eilertson et al. (1996) SOM release + TRY, Einarsson et al. (1997) CH-TRY
Legend: +, stimulation; -,inhibition; CH-TRY, chymotrypsin; CM, circular muscle; LM, longitudinal muscle; SOM, somatostatin; TRY, trypsin; * experiments performed'with pentagastrin which is the common C-terminal to gastrin and CCK.
the muscle, in part by activation of tachykinin neurons and serotonergic neurons. The receptors are similar to the mammalian CCK-A type (Martinet al., 1994).On the circular muscle from the ileum, CCK acts via purinergic neurons, which in turn act on nitric oxide-producing cells and VIP-neurons, which inhibit the muscle cells (Martin et al., 1998).
Control of g u t motility
337
There are strong indications that CCK released in the gut induces satiety in nonmammalian species as well as in mammals. Intraperitoneal injections of CCK in the goldfish Carassius auratus, chicken and the white-crowned sparrow Zonotrichia leucophrys gambelii suppressed food intake in all these species (Himick and Peter, 199413; Savory and Gentle, 1980; Covasa and Forbes, 1994; Rodriguez-Sinovas et al., 1997a; Richardson et al., 1993).The exact mechanisms for this have not been clearly established but seem to differ in at least the two bird species. In Python molurus, there was a 25-fold increase in plasma level of CCK after feeding and a corresponding decrease in tissue level in the intestine (Secor et al., 2001); it is, not yet known, however, whether this affects the reptile's feeling of hunger or satiety. Gastrin-releasing peptide Gastrin-releasing peptide is released in the gut when food is ingested. Gastrin-releasing peptide and/or the closely related peptide bombesin (commonly used in pharmacological studies) stimulated gut secretion in all animals tested (Table 13.2).It also increased blood flow to the gut in tested species (Table 13.2).It stimulated gastrointestinal motility in most instances with some notable exceptions: the circular muscle from the pigeon proventriculus and the Necturus stomach and intestine (Table 13.2).The corresponding longitudinal muscle in these species was stimulated, which makes it likely that GRI' is involved in control of peristalsis or other rhythmic contractions in these species. One characteristicof the descending reflex in peristalsis is the simultaneous relaxation of the circular muscle and contraction of the longitudinal muscle. Several studies suggest that GRP/bombesin induces satiety in nonmammalian vertebrates as well as in mammals (Merali et al., 1999, mammals). In carp, Cyprinus carpio, injections of bombesin delayed the onset of feeding after a previous meal, and in the goldfish, bombesin injections suppressed food intake (Beach et al., 1988; Himick and Peter, 1994a).Intravenous injections of bombesin in turkey and chicken similarly decreased food intake (Savory and Hodgkiss, 1984; Savory, 1987; Denbow, 1994).However, the mechanisms of action are not clear and it has been proposed both from studies in mammals (McCoy and Avery, 1990;Lee et al., 1994)and in chicken (Savory, 1987)that the effect, especially after high concentrations of bombesin, is one of aversion rather than satiety. Ghrelin Ghrelin is a 28 amino acid polypeptide, with similaritiesto the peptide m o t h . In mammals, ghrelin is synthesized in type A-like endocrine cells throughout the gut. The number of cells is highest in the stomach and decreases anally. It is a GH (growth hormone) secretagogueand acts through the receptor GHS-R (growth hormone secretagogue receptor), which until the recent discovery of ghrelin was considered an orphan receptor (i.e. a receptor with an unknown ligand). GHS-Rs are found in the stomach, intestine, and brain regions associated with feeding control (in particular the arcuate
338
Physiological and ecological adaptations to feeding in vertebrates
Table 13.2. Effects of GRP and bombesin on the non-mammalian gut Species
Effect
Reference
Proventri- Pigeon - CM Gascoigne et al. (1988) culus + LM Gascoigne et al. (1988) muscle + Holmgren and Jonsson (1988); Stomach Gadus nrorhua (teleost) Thorndyke and Holmgren (1990) sm. muscle + Holmgren, (1983) Oncorhynchus mykiss (teleost) Necturus maculosus (amphibian) - CM Holmgren et al. (1985) + LM Holmgren et al. (1985) + Lundin et al. (1984) Intestine/ Squalus acanthias (elasmobr.) rectum motility Gadus morhua (teleost) - (weak) Jensen and Holmgren (1985) Holmgren and Jonsson (1988) Necturus maculosus (amphibian) -/+ CM Holmgren et a1. .(1985) 0 / + LM Holmgren et al. (1985 + Kim et al. (2001) Xenopus Iaevis (amphibian) + Holmgren et al. (1989) Cainran crocodylus (reptile) + Falconieri-Erspamer et al. (1988) Tortoise (reptile) Chicken + Erspamer et al. (1972) Falconieri-Erspamer et al. (1988) + Holstein and Humphrey (1980) Gastric Gadus morhua (teleost) + Ayalon et al. (1981) Rana ca tesbeiana (amphibian) acid + Linari et al. (1975) secretion Chicken + Campbell et al. (1994) Turkey Rana catesbeiana (amphibian) + (eso- Shirakawa and Hirschowitz Pepsin/ pepsinogen phagus) (1985) Hirschowitz et al. (1990) secretion + Campbell et al. (1991) Pancreatic Turkey secretion Legend:
+, stimulation; -,inhibition; CM, circular muscle ; LM, longitudinal muscle
nucleus of the hypothalamus).Ghrelin,besides stimulatingGH release from the anterior pituitary, stimulated food intake and reduced fat utilization, leading to an increase in weight. The levels of ghrelin in stomach mucosa were high in the fasting stage but declined rapidly after feeding; there was also a rapid decline in the plasma level of ghrelin after food intake (Kojima et al., 1999; Ariyasu et al., 2001; Wang et al., 2002). Ghrelin also produced direct effects on the mammalian gut. Addition of ghrelin increased gut motility and gastric acid secretion in rats (Masuda et al., 2000). It is not known whether the predominant effect of ghrelin on the gut is direct or via a central pathway (vagus),but it is suggested that it does not involve the release of GH (Masuda et al., 2000). Since ghrelin is so recently discovered, little is known of its presence and function in nonmammalian vertebrates. As in mammals, ghrelin is present in bullfrog gut and in particular the stomach wall. Although the bullfrog form is modified compared to mammalian ghrelin and did not induce secretion of G H in rats, it was effective in releasing both GH and prolactin from
Control of gut motility
339
bullfrog pituitary cells (Kaiya et al., 2001). A few intensely ghrelin reactive cells in the stomach mucosa were found using immunohistochemistryin the frog Rana esculenta (Galaset al., 2002). In chicken, ghrelin mRNA is predominantly expressed in the proventriculus, which may be compared to the fundic part of the mammalian stomach (Kaiya et al., 2002).Interestingly,injection of ghrelin into the cerebral ventricles inhibited feeding in a dose-dependent manner, an effect opposite to that in mammals (Furuse et al., 2001). Studies using a ghrelin receptor-specific ligand in a fish species suggest that fish also express a ghrelin-like peptide (Shepherdet al., 2000)but it has yet to be ascertained whether it is synthesized in the gut.
Changes in Innervation between (prolonged) Fasting and Fed Stages in Nonmammalian Species As mentioned earlier, various animals, e.g. a number of reptile species, experience long periods of fasting. This prolonged period of fasting has consequences for the digestive tract. In reptiles for instance the structure, level of gut enzymes, and activity of transporters of the small intestine are affected by the digestive status (seeChapter 7by J.M. Starck, this volume). But does a prolonged period of fasting influence the neuronal and hormonal control systems of the gut? Until recently, the only studies to address this issue were done on mammals. In hibernating golden hamsters, the total number of neurons in the gut were unaffected by the digestive status, while a selective upand down-regulation of their contents of some neurotransmitters were observed (Shochina et al., 1997;Toole et al., 1999).To our knowledge, the only nonrnammalian species in which the effects of digestive status on innervation and motility have been studied is the Burmese python, Python molurus. As in mammals, the total number of gut neurons seemed to be unaffected in pythons that had been starved for a minimum of three weeks compared to fed snakes. There was a tendency, however, toward a denser innervationby neurons expressing tachykinins (Fig. 13.4),galanin, PACAP, and nitric oxide synthase (NOS)in fasting snakes (Holmberg et al., 2003). This could be a consequence of the decreased volume of surrounding tissues that is seen in the small intestine during fasting (see chapter 7 by J.M.Starck, this volume). An increased level of transmitters (and therefore an increased immunoreaction) during fasting might also be explained as a storage effect since the snake is not feeding and presumably therefore not activating and releasing the transmitter from a number of nerve pathways. Secor et. al. (2001) measured a concurrent decrease in gut tissue concentration of certain regulatory gut peptides (CCK, gastric inhibitory polypeptide, GIP and neurotensin) with a corresponding increase in plasma concentrations one day after feeding in the Burmese python. The potential up-regulation during fasting might be a protective mechanism to maintain the integrity of the gut in the absence of food, but this needs to be further investigated.In mammals, muscle activity, e.g. in the form of migrating motor complexes (MMC)
340
Physiological and ecological adaptations t o feeding in vertebrates
Fig. 13.4. Tachykinin-like immunoreactivity in neurons in the muscle layers of the large intestine of fed (A) and fasting (B) Python molurus, respectively. The figures indicate the tendency of an apparently denser innervation by tachykinin-immunoreactive nerves in fasting compared to fed specimen. CM, circular muscle layer; LM, longitudinal muscle layer; MEP, myenteric plexus. Calibration bars: 100 pm. Reproduced with permission from Holmberg et al. (2003).
persists during fasting (see below). The basic spontaneous motility (in in vitro preparations) likewise persisted during fasting in Python molurus (Holmberget al., 2003).Furthermore, the effect of two excitatory neuropeptides (substanceP, Fig. 13.5and galanin) and of bradykinin on the motility of the small intestinewas independent of the digestive state in Python molurus (Holmberg et al., 2003). These findings suggest that the intestine remains functional and that motility persists during fasting, while awaiting the next feeding opportunity.
loo
Substance P
1 A
tasted fed
!
T
log M Fig. 13.5. Dose-response curves showing the effects of substance P on longitudinal preparations from the proximal intestine of fasted (3 weeks) and fed (48h) Python molurus. The effect of substance P seems to be independent of digestive state. Reproduced with permission from Holmberg et al. (2003).
342
Physiological and ecological adaptations t o feeding in vertebrates
detected before the onset of feeding, while other transmitters appear later (Maake et al., 2001). To our knowledge, there is no information available on the effects on motility of these transmittersbefore the onset of feeding. Thus it remains to be studied whether these transmitters are released and are able to regulate/ modulate motility before the onset of feeding. Another possibility is that these transmitters act as trophic factors at this stage. It has been suggested for mammals that transmitters such as VIP and PACAP exhibit a trophic effecton the nervous system (Waschek, 1995; Waschek et al., 1998).
CONTROL OF GUT MOTILITY
Gut smooth muscle cells are coordinated in a number of motility patterns aimed to process and transport food in an optimal way. One prerequisite for this is the electrical coupling between the smooth muscle cells; gap junctions between the cells allow depolarizations and hence contraction waves (or inhibitions) to spread orally, anally (aborally) or circumferentially over the gut. Short reflexes within the ENS of the gut wall, or long reflexes involving pathways to and from the CNS control and sustain these contraction/inhibition waves. Motility During Fasting Gut smooth muscle may be active although there is no food present to be transported. So called "spontaneous" contractions usually occur in in-vitro preparations of smooth muscle. Similar single local contractions may occur in the gut in vivo in the interdigestivestate but more often MMCs are formed, which sweep along the gut orally or aborally. Spontaneous contractions Rhythmic muscle activity of the gut is due to regular slow waves, i.e.fluctuations in membrane potential of the muscle cell, which allow action potentials to occur in bursts at regular intervals (when depolarization exceeds the threshold value). ICCs and, at least in some species, also enteric and extrinsic nerves are involved in the initiation and modulation of such slow waves and the subsequent "spontaneous" rhythmic contractions of the mammalian gut (Sanders, 1996; Daniel, 2001). For example, nitric oxide reduces the amplitude of rhythmic contractions, presumably by an action via ICCs (Cayabyab et al., 1997; Keef et al., 1997).Tetrodotoxin (TTX),which selectively blocks nerve transmission, enhances the contractile activity in the canine gut, implying the presence of an inhibitory nervous tonus (Cayabyab et al., 1997). In the stomach of a frog (Rana pipiens) and a toad (Xenopzls laevis), rhythmic depolarizations and subsequent contractions proved dependent on the
Control of gut motility
343
presence of ICCs (Prosser, 1995).In Xenopus, these contractions might not have involved enteric nerves. Although contractions caused by electrical stimulation of stomach preparations were abolished by TTX (Johansson et al., 2002), spontaneous contractions of these preparations were not affected by TTX (A. Johansson,pers. comm.).Contrarily, in the Atlantic cod intestine, TTX as well as atropine and the 5-HT receptor antagonist methysergide, reduced or abolished the contractile activity, indicating a dependency on cholinergic and serotonergic neurons (Karila and Holmgren, 1995; Olsson and Holmgren, 2000).Antagonists of nitric oxide formation increased spontaneous activity in both Xenopus and Atlantic cod indicating that, as in most mammals, nitric oxide is tonically released and counteracts excitatory influences on the gut (Karila and Holmgren, 1995; Olsson and Holmgren, 2000; Olsson, 2002). Migrating motor complexes (MMCs) In mammals, depolarization waves sweep over the entire gut in the interdigestive state or during fasting. These migrating myoelectric complexes initiate contraction waves (migrating motor complexes, both abbreviated MMCs), whose function may be to keep the gut clean of debris from food, secreted mucus and dead epithelial cells, and to control bacterial growth. The MMCs are divided into four phases. Phase I is no activity, phase I1 irregular single contractions, phase I11 intensive rhythmic contractions, and phase IV the transitional phase to phase I. MMCs may occur spontaneously, by input from ICCs and the ENS, or by the release of gut hormones such as motilin, and may be modulated by the vagus nerve (Sarna, 1985; Tanaka et al., 2001). In birds, two types of migrating motility patterns are suggested. Rhythmic oscillating complexes (ROCs)comprise high-speed, anally propagating bursts of spike potentials (SPBs)that subsequently change into orally propagating SPBs, both presumably initiating rapid contraction waves recycling the intestinal contents. ROCs occur only in fasted avian species (Clench and Mathias, 1992; Jimenez et al., 1994; Rodriguez-Sinovas et al., 199%). The presence of MMCs in birds has also been reported (Clench et al., 1989; Schlamp et al., 1991; Jimenez et al., 1994).In chicken and turkey, these are predominantly phase 111-type (although of different character in the two species) and occur in the intestine during both fasting and fed states. Inhibition of turkey MMCs on feeding is independent of the sight of food, presence of food in the mouth and esophagus or of extrinsic nerves; in other words, inhibition is due to local stimuli of the gut wall (Mueller et al., 1990;Schlamp et al., 1991).In chicken, the MMCs were slowed down or abolished by CCK or intralumenal fat (e.g. Martinez et al., 1995a,b). In teleost fish also, propagating contractions have been suggested as analogous to the MMC phase IIIlike activity observed in in vitro stomach and intestinal preparations from fasted teleosts (Karila and Holmgren, 1995; Olsson et al., 1999).
344
Physiological and ecological adaptations t o feeding in vertebrates
Consequences of Feeding on Gut Motility The intake of food initiates a number of well-coordinated motor activities in the gut that mix and propel the food forward at an optimal pace. The presence of food in the back of the mouth initiates swallowing and peristaltic waves in the esophagus. Distension of the esophagus and of the stomach wall on arrival of food initiates gastric receptive relaxation or accommodation to provide more space for the arriving food. This is followed by contractions in the stomach wall, both local mixing contractions and peristaltic waves propagating toward the pyloric sphincter. Gastric emptying is controlled by the coordinated action of relaxation in the pyloric sphincter and the aforesaid contractions in the posterior part of the stomach propelling boluses of food (chyme)into the intestine. Peristalsis in the intestine moves the chyme forward at an appropriate rate. The mechanisms controlling these motor of patterns in mammals have been reviewed by Furness and Costa (1987);Kunze and Furness (1999);Olsson and Holmgren (2001). Although there are several studies dealing with effects of individualneurotransmitters and hormones on isolated preparations of gut smooth muscle from nonmammalian species (e.g. Tables 13.1 and 13.2, for CCK and GRP respectively), studies dealing with the integrated function of these signal substances in the integrated reflexes stimulated by feeding to transport the food along the gut are scarce indeed. What little is known is summarized below. Reception and miking of food in the stomach In mammals, distension of the esophagus and stomach wall initiates gastric receptive relaxation and gastric accommodation respectively through vagovagal reflexes, i.e. both the sensory branch and the motor of branch of the reflex run in the vagus (Abrahamsson,1973).The final effect is release of the inhibitory transmitters VIP and /or nitric oxide, which causes relaxation of the stomach wall. Similarly, in both rainbow trout and Atlantic cod, distension of the stomach causes reflex relaxation of the stomach wall, which in both species,is mediated by enteric neurons independent of extrinsic nerves (Grove and Holmgren, 1992a,b). Gastric distension also induces phasic contractions, which grind, mix and propagate the food. This involves both excitatory and inhibitory enteric reflexes. The excitatory reflexes are predominantly cholinergic but involve at least 10 other neuron types with different transmitter combinations (Schemann et al., 1995; Hennig et al., 1997).In rainbow trout, contractions initiated by stomach distension are blocked by a corr~binationof cholinergic and serotonergicantagonists. The serotonergic reflex is reduced by somatostatin (from local endocrine cells),while the cholinergic pathway is inhibited by local VIP neurons (Grove and Holmgren, 1992a).Cholinergic excitation may be common among vertebrates but both histological and pharmacological evidence suggest that it is absent in the Atlantic cod (Grove and Holmgren, 1992b).
Control of gut motility
345
Gastric emptying and intestinal transit After acidification (usually) and initial digestion in the stomach, the food is delivered in boluses to the intestine by the integrated action of peristaltic contractionsin the stomach and relaxations of the pyloric sphincter. The rate of gastric emptying and subsequent transit of food through the intestine is influenced by numerous factors including frequency of feeding, amount of food in the stomach, and composition of the contents in stomach and intestine (solid or liquid, pH and osmolarity, carbohydrates, fat, protein, etc.). Once food enters the intestine, further emptying as well as upper gastrointestinal motility in general, is slowed down (the duodenal brake, the intestinal brake). This is mainly dependent on components of the food (in particular fat) stimulating release of duodenal CCK and intestinal peptide YY (PYY) and gastrin from the stomach mucosa. Also, a number of transmitters and other hormones such as acetylcholine, calcitonin gene-related peptide (CGRP),galanin, nitric oxide, noradrenaline, motilin, serotonin, tachykinins, and VIP have been implicated (Liddle et al., 1986; Torsoli and Severi, 1993; Daniel et al., 1994; Orihata and Sarna, 1994; Ohtani et al., 2001). Similarly, in chicken an infusion of fat into the ileum delayed gastric emptying and intestinal transit (Martinez et al., 1995a). In rainbow trout, injections of CCK, in vivo, delayed gastric emptying, which suggests the important role of CCK in this control in fish also (Olsson et al., 1999). Peristalsis of the intestine Distension of the gut wall by a bolus of food elicits reflexes leading to peristalsis. Ascending reflexes lead to contraction of the circular muscle (and relaxation of the longitudinal muscle) orad to the bolus. Simultaneous descending reflexes cause relaxation of the circular muscle and contraction of the longitudinal muscle. This leads to widening and shortening of the gut anal to the bolus, which facilitates movement of the bolus in this direction. The types and projections of the numerous neurons involved in peristalsis in mammals have been reviewed in detail by e.g. Furness and Costa (1987) and Kunze and Furness (1999). Very few comprehensive studies on these complex mechanisms have been performed in nonmammalian vertebrates. Some studies combining myotomies with immunohistochemistry demonstrate the presence of acetylcholine, serotonin, and/or substdnceP or related tachykinins in ascending pathways in the Atlantic cod intestine (Karila et al., 1998).VIP/PACAP, galanin, and NOS containing nerves are predominantly descending, i.e. projecting anally (Olsson and Karila, 1995; Karila and Holmgren, 1997; Fig. 13.7). In the lizard, Podarcis s. sicula, an accumulation of galanin-like imrnunoreactivity on the oral side of a cut of the myenteric plexus suggested that galanin is involved in descending pathways in this species also (Lamanna et al., 1999). Furthermore, electrical stimulation of the intestine from cod elicited a contraction oral to the stimulation site, involving cholinergic and serotonergic
346
Physiological and ecological adaptations t o feeding in vertebrates
by passing food ECs (containing 5-HT?)
sensory neuron*
MEP LM
motor neurons
ANAL
interneurons
motor neurons
ORAL
* location Unknown
Fig. 13.7. A hypothetical model of the peristaltic reflex in the Atlantic cod, based on pharmacological and immunohistochemical data, including myotomies. The neurotransmitters ACh, 5-HT and tachykinins are part of the ascending reflex causing a contraction of the circular muscle layer oral to the food bolus. NO, VIP, galanin are involved in the descending pathway causing a relaxation of the circular muscle fibers anal to the bolus. The location of the sensory neurons is hypothetical. Redrawn from Karila, 1997. ACh, acetylcholine; CM, circular muscle layer; EC, endocrine cell; Gal, galanin; LM, longitudinal muscle layer; M, mucosa; MEP, myenteric plexus; NO, nitric oxide; 5-HT, serotonin; SM, submucosa; Tach, tachykinins; VIP, vasoactive intestinal polypeptide; +, stimulation; -, inhibition; *, location not known.
mechanisms. A simultaneous relaxation anal to the stimulation site involves nitric oxide production and serotonergic pathways (Karila and Holmgren, 1995; Fig. 13.7). None of these findings contradict what is reported in mammals. In birds, there are several anatomical specializations, such as the crop (part of the esophagus) and gizzard (muscular stomach),and the presence of two intestinal ceca, which lead to specific patterns of motility in the upper part of the intestine and in these ceca (Hodgkiss, 1986; Reynhout and Duke, 1999). From studies in turkey, it is suggested that a neurogenic pacemaker located between the gizzard and the glandular stomach initiates contractions in the stomach parts and in the duodenum (Chaplin and Duke, 1988, 1990). In addition to "normal" (anterograd) peristalsis, there are regular refluxes of intestinal contents toward and into the gizzard in many birds. In the chicken ceca, the presence of food increased the number of propagating action potentials (and presumably subsequent contractions) but reduced
Control of gut motility
347
their speed of propagation. Interestingly,there was no relationship between activity in the ceca and that in the ileum (Clench and Mathias, 1996).
CONTROL OF SECRETION The different mechanisms for secretion of gastric acid in different vertebrate groups are compared here. Gastric acid secretion is not the only type of secretion, by far performed by the gut, but it is almost the only secretory mechanism in the gut for which comparative studies of nonmammalian vertebrates have been performed to some extent. We shall also shortly report what little is known of the secretion of alkali and pepsinogen from the gut wall. Mucus is secreted from single mucosal cells all along the gut in most species,but to our knowledge there are no studies of the mechanismsbehind this in nonmammalian species.Endocrine secretion from mucosal hormoneproducing cells affects gut performance, as mentioned above regarding motility patterns, and below in the control of secretion. Furthermore, accessory glands such as the salivary glands, liver, and pancreas release an exocrine secretion into the gut lumen through their glandular ducts. Comparative studies of pancreatic secretion have been performed to some extent but are not reviewed here. Studies on gut secretion (includingthe pancreas) and its control in nonmamrnalian vertebrateshave been carefully reviewed by Srnit (1968) and Jonsson (1994).
Gastric Acid Secretion In nonmammalian vertebrates, gastric acid is usually secreted from so- called oxynticopepticcells gathered in tubular glands in the mucosa of the cardiac (anterior)part of the stomach. These cells secrete both gastric acid and pepsinogen (Bishop and Odense, 1966; Mattisson and Holstein, 1980; Ezeasor, 1981;Garrido et al., 1993;Gallego-Huidobro and Pastor, 1996). In mammals, separate cells secrete gastric acid (parietal cells) and pepsinogen (chief cells) and these cells normally occupy distinct areas of the stomach mucosa (Helander, 1981;Koelz, 1992).The oxynticopeptic cell might not be present in all nonmammalian species and differences between even closely related species may occur. Thus, in one elasmobranch species, Hexanchus griseus, secretion of gastric acid and pepsinogen occurred in different cells (Michelangeli et al., 1988), but not in another elasmobranch, Halaelurus chilensis, in which the gastric glands contain one form of secretory cells only, similar in ultrastructure to the oxynticopeptic cells (Rebolledoand Vial, 1979). A recent study in white flounder using in-situ hybridizationwith RNA probes for two pepsinogen genes and one proton pump gene (indicatingthe ability to secrete acid) confirmed the presence of oxynticopeptic cells expressing both pepsinogen and proton pumps in fish. Interestingly, the proton pump in addition is expressed on mucous cells suggesting that these cells also secrete acid (Gawlicka et al., 2001).
348
Physiological and ecological adaptations t o feeding i n vertebrates
Species variations are also evident in extension of acid-secretingmucosa; e.g. in the Atlantic cod, Gadus morhua, a distinct border occurs between the esophagus and the acid-secreting mucosa of the stomach (Bishop and Odense, 1966),while in the rainbow trout, the mucosa of the distal part of the esophagus contains gastric type glands (Ezeasor, 1984).Not only stomachless fish such as cyprinids (carp) and labrids (wrasses),but also monotremes, which have a stomach, lack an acid-secreting mucosa (e.g.Koelz, 1992). Basal levels of gastric acid secretion Most vertebrates appear to have a basal secretion of gastric acid, i.e. there is a low level of pH-dependent secretion even when the stomach is not digesting. This functions to maintain a low pH in the resting stomach. A pH of 0.8 - 2.0 has been measured in species representing sharks, rays, crocodiles, lizards, tortoises, birds, and mammals (Sullivan, 1905-1906; Dobreff, 1927; Smit, 1968). The secretion rate (from isolated stomach mucosal preparations) is in the same order of m a p t u d e for the few different species that have been studied, although they represent four of the major vertebrate groups (Table 13.3).The small differences there are may be related to different feeding habits. In the Atlantic cod in vivo basal gastric acid secretion is almost abolished after vagotomy (Holstein and Cederberg, 1980).It has likewise been concluded that basic secretion is under vagal control in chicken (Burhol, 197313). Feeding and gastric acid Secretion Food intake, in general, increases gastric acid secretion. Already in 1905, this was observed in an elasmobranch species (Sullivan, 1905-1906). Control of acid secretion associated with food intake comprises three phases in mammals: 1)the cephalic phase, triggered by visual, olfactory, auditory stimuli and even the anticipation of food, acting via the vagus nerve; 2) the gastric phase, triggered by distension of the esophagus and stomach wall, by components in the food, or by a high pH in the stomach; and 3) the intestinal phase, initiated by the entry of food into the duodenum. It has been shown in ducks that the sight of food increases gastric acid secretion, and trained ducks also secrete gastric acid on a signal in anticipation of food (Smit, 1968).Thus a cephalic phase in the control of acid secretion exists at least in birds. In fish and amphibians,distension of the stomach wall (part of the "gastric phase") initiates secretion (Smit, 1968). Table 13.3. Basal gastric acid secretion rate from isolated stomach mucosa Animal Species Secretion Rate Reference Group (nEqH+cm-*min-') Elasmobranch Teleost Amphibian Amphibian Mammal
Squalus acanfhias Gadus morhua Rana esculenfa Bufo viridis Guinea pig
26 6.0 10 10 ca. 20.3
+_
+
+ + +
3.3 0.6 5 5 1.7
Hogben (1967) Bomgren and Jonsson (1996) Negri and Erspamer (1973) Negri and Erspamer (1973) Visvanathan (1992)
Control of gut motility
Stimulation of acid secretion Several agents are involved in control of the oxynticopeptic cells. As in mammals, histamine, gastrin, acetylcholine,and somatostatin play central roles in the control of acid secretion,but several other neurotransmitters and hormones have additional effects (mammals: Dockray, 1999; Lindstrom et al., 2001; amphibians, Fig. 13.8; teleosts, Fig. 13.9).Histamine and acetylcholine stimulated secretion of gastric acid in the absolute majority of investigated elasmobranchs, teleost; amphibians, reptiles, and birds (e.g. Smit, 1968; Burhol and Hirschowitz, 1970; Ruoff and Sewing, 1972; Holstein, 1976; Ruiz and Michelangeli, 1984). Histamine, synthesized and stored in enterochromaffin cells situated close to the oxynticopepticcells, seems to act as the final common pathway for all extrinsic input to the parietal cell in mammals, although several agents may have additional direct effects on the parietal cells. Histamine stimulated secretion in all species investigated, representing all the major vertebrate groups (exceptcyclostomes). The effect of histamine is mediated by H2-receptors on parietal cells in mammals. In cod, H2-receptor antagonists blocked or reduced acid secretion
m...........
H+ H+
I --(
inhibition
nerve ending
Fig, 13.8. Schematic model of the regulation of gastric acid secretion in amphibians, based on available data. In addition, bombesin and caerulein stimulate gastric acid secretion, but the site of action is not clear. ACh, acetylcholine; Adr, adrenalin; HC, hydrocortisone. Redrawn from Bomgren (2001).
350
Physiological and ecological adaptations t o feeding in vertebrates
PACAP VIP
\r somatostatin
,',
;
?
(
'....'.
Oxynticopeptic cell
............
inhibition hormonal effect
-( nerveending Fig. 13.9. Schematic model of the regulation of gastric acid secretion in teleosts, based on available data. In addition, bombesin stimulates, CCK and caerulein inhibit, and 5-HT and tachykinins have mixed effects on gastric acid secretion, but the site of action is not clear. ACh, acetylcholine; Adr, adrenalin; HC, hydrocortisone. PACAP, pituitary adenylate cyclase-activating peptide; VIP, vasoactive intestinal polypeptide. Redrawn from Bomgren (2001).
(Holstein, 1976; Bomgren and Jonsson, 1996).Some mammalian H2-receptor agonists stimulated secretion while others were ineffective, as are HI agonists (Holstein,1986).Also, the receptors in bullfrog appear to be H2-like (Lin et al., 1986), suggesting that the mechanism of action of histamine on acid secretion is well conserved among vertebrates. Acetylcholine, presumably released from the vagus nerve in vivo and acting on muscarinic receptors, likewise has a general stimulatory effect on acid secretion in vertebrates.Atropine blocked the effect of cholinergic drugs in elasmobranchs and teleosts (Smit, 1968; Holstein, 1977),as did histamine H, antagonist metiamide in the Atlantic cod, suggesting an effect of acetylcholine via histamine release (Holstein, 1976).Ruiz and Michelangeli (1984) demonstrated that acetylcholine acts both directly on the oxynticopeptic cell and indirectly via histamine release in the bullfrog. Gastrin was probably the first identified secretagoguefor gastric acid and there are reports of a stimulatory action on gastric acid secretion in birds
Control of gut motility
351
(Dimaline and Lee, 1990)and amphibians (Davidson et al., 1966),while the effect in the Atlantic cod, was contrarily inhibitory (Holstein,1982).It is now known that the stimulatory effect of gastrin in mammals is mainly by an action on histamine cells (Lindstrom et al., 2001). Both indirect (via histamine) and direct stimulation of the oxynticopeptic cells are suggested in amphibians (Ekblad, 1985; Ruiz and Michelangeli, 1986). GRP and/or the closely related peptide bombesin may stimulate acid secretion directly in some species, such as Rana catesbeiana (Ayalon et al., 1981), via release of gastrin in e.g. the chicken (Linari et al., 1975) or by inhibition of VIP in e.g. the Atlantic cod (Holstein and Humphrey, 1980).
Pepsin/Pepsinogen Secretion Feeding induces an increased secretion of pepsinogen from a basic interdigestive level. In an acidic milieu, inactive pepsinogen is immediately converted to active digestive enzyme pepsin. In nonmammalian vertebrates possessing a stomach, pepsinogen is stored together with gastric acid in the oxynticopeptic cells in the stomach mucosa (Smit, 1968; Helander, 1981). Pepsinogen-immunoreactive cells have been found in species from all the major nonmammalian groups (Yasugi, 1987; Yasugi et al., 1988). Cells secreting only pepsinogen occur additionally in the mucosa of the esophagus in amphibian species (Simpson et al., 1980). Using "pure" pepsinogen cells from the esophagus of the frog Rana catesbeiana, it was found that pepsinogen secretion is stimulated by e.g. cholinergic drugs and bombesin, but not by pentagastrin or histamine (Simpson et al., 1980; Shirakawa and Hirschowitz, 1985).Induced secretion is inhibited by somatostatin (Fonget al., 1991).In the Atlantic cod in vivo tachykmins and serotonin have strong stimulatory effects while acetylcholine and histamine have weak effects only (Holstein and Cederberg, 1984,1986). It appears that gastric acid and pepsinogen may be separately released from the oxynticopeptic cells due to different sensitivities to the controlling agents. For example, in chicken pentagastrin stimulated pepsin release more strongly than release of gastric acid (Burhol, 1973a),and in the Atlantic cod, tachykinins showed strong stimulatory effects on pepsinogen secretion but only weak effects on acid secretion, while histamine and carbachol had the opposite effect (Holstein and Cederberg, 1984,1986). Cholinergic stimulation may be general to all vertebrates and presumably reflects vagal control. The influence of spinal innervation and/or circulating catecholamines appears to vary. Enteric nerves and local peptidergic nerves are also important for control. Secretion of Bicarbonate Bicarbonate is secreted in the stomach mucosa as part of the protective mucous-bicarbonate barrier, and in the intestine to neutralize the acid chyme arriving from the stomach, thereby optimizing the effect of secreted enzymes.
352
Physiological and ecological adaptations t o feeding i n vertebrates
Secretion is stimulated by the presence of acid in the stomach and intestine respectively. In experiments in amphibian species (Rana catesbeiana, Necturus maculosus) it was found that bicarbonate secretion was blocked by atropine, suggesting a cholinergic mechanism. Secretion was stimulated by glucagon in both stomach and intestine but gastric inhibitory peptide (GIP) and noradrenaline were inhibitory on stomach secretion and excitatory on the intestine. CCK stimulated secretion from the gastric mucosa but, in contrast to the mammalian intestine, the amphibian intestinal secretion was unaffected by CCK (Flemstrom and Garner, 1980; Flemstrom et al., 1982). CONCLUDING REMARKS
It appears that many features are common among vertebrates in the control of gut motility and secretion. Comprehensive studies in fish are few and information pertaining to amphibians and reptiles even more scant, but what information there is seems to support the concept of a strong conservation during evolution.In contrast, with the several unique anatomical as well as physiological specializationsamongbirds, there are also distinct patterns of motility and differences in control of secretion. It remains to be elucidated whether such variations from the general (or mammalian) plan also occur among other vertebrate groups. Clearly, many more comparative integrated studies on the control of gut motility and secretion,and the consequencesof feeding on this control are needed. REFERENCES Abrahamsson H. 1973. Studies on the inhibitory nervous control of gastric motility. Acta. Physiol. Scand. Suppl. 390: 1-38. Aldman G. and Holmgren S. 1987. Control of gallbladder motility in the rainbow trout, Salmo gairdneri. Fish Physiol. Biochem. 4: 143-155. Aldman G., Grove D.J. and Holmgren S. 1992. Duodenal acidification and intraarterial injection of CCK8 increase gallbladder motility in the rainbow trout, Oncorhynchus mykiss. Gen. Comp. Endocrinol. 86: 20-25. Aldman G., Jonsson A.C., Jensen J. and Holmgren S. 1989. Gastrin/CCK-like peptides in the spiny dogfish, Squalus acanthias; concentrations and actions in the gut. Comp. Biochem. Physiol. 92C: 103-109. Andrews P.L.R. and Young J.Z. 1988a. The effect of peptides on the motility of the stomach, intestine and rectum in the skate (Raja). Comp. Biochem. Physiol. 89C: 343-348. Andrews P.L.R. and Young J.Z. 198813. A pharmacological study of the control of motility in the gallbladder of the skate (Raja). Comp. Biochem. Physiol. 89C: 349-354. Ariyasu H., Takaya K., Tagami T., and Ogawa Y. 2001. Stomach is a major source of circulating ghrelin, and feeding state determines plasma ghrelin-like immunoreactivity levels in humans. J. Clin. Endocrin. Metab. 86: 4753-4758. Ayalon A., Yazigi R., Devitt P.G., Rayford P.L. and Thompson J.C. 1981. Direct effect of bornbesin on isolated gastric mucosa. Biochem. Biophys. Res. Commun. 99: 1390-1397.
Control of gut motility
353
Bartlet A.L. and Hassan T. 1971. Contraction of chicken rectum to nerve stimulation after blockade of sympathetic and parasympathetic transmission. Quart. J. Exp. Physiol. 56: 178-183. Beach M.A., McVean A., Roberts M.G. and Thorndyke M.C. 1988. The effects of bombesin on the feeding of fish. Neurosci. Lett. 32: 46. Bisgrove B.W., Raible D. W., Walter V., Eisen J. S. and Grunwald D. J. 1997. Expression of c-ret in the Zebrafish embryo: Potential roles in motoneuronal development. J. Neurobiol. 33: 749-768. Bishop C. and Odense P.H. 1966. Morphology of the digestive tract of the Atlantic cod, Gadus morhua. J. Fish. Res. Bd. Canada 23: 10. Bjenning C. and Holmgren S. 1988. Neuropeptides in the fish gut; a study of evolutionary trends. Histochemistry 88: 155-163. Bomgren P. 2001 Gastrin and histamine in amphibians and fish. PhD thesis, Goteborg University, Sweden. Bomgren P. and Jonsson A.-C. 1996. Basal, H2-receptor stimulated and pH-dependent gastric acid secretion from an isolated stomach mucosa preparation of the Atlantic cod, Gadus morhua, studied using a modified pH-static titration method. Fish Physiol. Biochem. 15: 275-285. Botar J., Afra D., Moritz P., Schiffman H. and Scholz M. 1950. Die Nervenzellen und Ganglien des N. vagus. Acta Anat. 10: 284-314. Burhol P.G. 1973a. Gastric secretory relationship between H+and pepsin in fistula chickens. Scand. J. Gastroenter. 8: 283-288. Burhol P.G. 1973b. Stimulation by 2-deoxy-D-glucose and inhibition by insulin of gastric secretion in fistula chickens. Scand. J. Gastroenter. 8: 761-764. Burhol P.G. and Hirschowitz B.I. 1970. Single subcutaneous doses of histamine and pentagastrin in gastric fistula chickens. Amer. J. Physiol. 218: 1671-1675. Burns A.J. and Le Douarin N.M. 2001. Enteric nervous system development: Analysis of the selective developmental potentialities of vagal and sacral neural crest cells using quail-chick chimeras. Anat. Rec. 262: 16-28. Campbell G. 1975. Inhibitory vagal innervation of the stomach in fish. Comp. Biochem. Physiol. 50C: 169-170. Campbell B., Garner A., Dimaline R. and Dockray G.J. 1991. Hormonal control of avian pancreas by gastrin-releasing peptide from the proventriculus. Amer. J. Physiol. 261: G16-G21. Campbell B.J., Garner A., Dockray G.J., Hughes J. and Dimaline R. 1994. The mechanism of action of gastrin releasing peptide (GRP) in stimulating avian gastric acid. Regulatory Peptides 49: 249-255. Cayabyab F.S., Jimenez M., Vergara P., De Bruin H. and Daniel E.E. 1997. Influence of nitric oxide and vasoactive intestinal peptide on the spontaneous and triggered electrical and mechanical activities of the canine ileum. Can. I. Physiol. 75: 383-397. Cerda-Reverter J.M. and Larhammar D. 2000. Neuropeptide Y family of peptides: structiure, anatomical expression, function and molecular evolution. Biochem. Cell. Biol. 78: 371-392. Chaplin S.B. and Duke G.E. 1988. Effect of denervation on initiation and coordination of gastroduodenal motility in turkeys. Amer. J. Physiol. 255: GI-G6. Chaplin S.B. and Duke G.E. 1990. Effect of denervation of the myenteric plexus on gastroduodenal motility in turkeys. Amer. J. Physiol. 259: G481-G489. Clench M.H. and Mathias J.R. 1992. A complex avian intestinal motility response to fasting. Amer. J. Physiol. 262: G498-G504. Clench M.H. and Mathias J.R. 1996. Myoelectric activity of the cecum in fed and fasted domestic fowl (Gallus sp.). Comp. Biochem. Physiol. 115A: 253-257. Clench M.H., Pifieiro-Carrero V.M. and Mathias J.R. 1989. Migrating myoelectric complex demonstrated in four avian species. Amer. J. Physiol. 256: G5984603.
354
Physiological and ecological adaptations t o feeding i n vertebrates
Covasa M. and Forbes J.M. 1994. Exogenous cholecystokinin octapeptide in broiler chickens: Satiety, conditioned colour aversion, and vagal mediation. Physiol. Behav. 56: 39-49. Daniel E.E. 2001. Physiology and pathophysiology of the interstitial cell of Cajal: from bench to bedside. 111. Interaction of interstitial cells of Cajal with neuromediators: an interim assessment. Amer. J. Physiol. 281: G1329-GI332 Daniel E.E., Tougas G., Allescher H.D., Vergara P. and Fox-Threlkeld J.A. 1994. Mediators and enteric nerve pathways controlling gastric emptying. Dig. Dis. Sci. 39 (12 Suppl.): S63-S68. Davidson W.D., Lemmi C.A.E. and Thompson J.C. 1966. Action of gastrin on the isolated gastric mucosa of the bullfrog. Proc. Soc. Exp. Biol. 121: 545-547. Denbow D.M. 1994. Peripheral regulation of food intake i n poultry. J. Nutr. 124: 1349s-1354s. Dimaline R. and Lee C.M. 1990. Chicken gastrin: a member of the gastrin/CCK family with novel structure-activity relationships. Amer. J. Physiol. 259: G882-G888. Dobreff M. 1927.Experimentelle Studien iiber die vergleichende Physiologie der Verdauung. Arch. Ges. Physiol. 217: 221-234. Dockray G.J. 1999. Topical review. Gastrin and gastric epithelial physiology. J. Physiol. 518: 315-324. Eilertson C.D., Carneiro N.M., Kittilson J.D., Comley C. and Sheridan M.A. 1996. Cholecystokinin, neuropeptide Y and galanin modulate the release of pancreatic somatostatin-25 and somatostatin-14 in vitro. Regul. Pept. 63: 105-112. Einarsson S., Davies P.S. and Talbot C. 1997. Effect of endogenous cholecystokinin on the discharge of the gallbladder and the secretion of trypsin and chymotrypsin from the pancreas of the Atlantic salmon, Salmo salar L. Comp. Biochem. Physiol. 117C: 63-67. Ekblad E., Wahlestedt C., Hikanson R., Sundler F., Watanabe T. and Wada H. 1985. Is histamine a neurotransmitter in the gut? Evidence from histidine decarboxylase immunocytochemistry. Acta Physiol. Scand. 123: 225-227 Epperlein H-H., Krotoski D., Halfter W. and Frey A. 1990. Origin and distribution of enteric neurons in Xenopus laevis. Anat. Embryol. 182: 53-67. Epstein M.L., Hudis J. and Dahl J.L. 1983. The development of peptidergic neurons in the foregut of the chick. J. Neurosci. 3: 2431-2447. Erspamer V., Falconieri-Erspamer G., Inselvini M. and Negri L. 1972. Occurrence of bombesin and alytesin in extracts of the skin of three European discoglossid frogs and pharmacological actions of bombesin on extravascular smooth muscle. Brit. J. Phar. 45: 333-348. Ezeasor D . N . 1979.Ultrastructural observations on the submucous plexus of the large intestine of the rainbow trout (Salmo gairdneri, Rich.). Z. Mikrosk. Anat. Forsch. 93: 803-812. Ezeasor D.N. 1981. The fine structure of the gastric epithelium on the rainbow trout, Salmo gairdneri, Richardson. J. Fish Biol. 19: 611-627. Ezeasor D.N. 1984. Light and electron microscopic studies on the oesophageal epithelium of the rainbow trout, Salmo gairdneri. Anat. Anz. 155: 71-83. Falconieri-Erspamer G., Severini C., Erspamer V., Melchiorri P., Delle Fave G. and Nakajima T. 1988. Parallel bioassay of 27 bombesin-like peptides on 9 smooth muscle preparations. Structure-activity relationships and bombesin receptor subtypes. Regul. Pept. 21: 1-11. Fange R. 1948. Effect of drugs on the intestine of a vertebrate without sympathetic nervous system. Arkiv Zool. 40A: 1-9. Flemstrom G. and Garner A. 1980. Stimulation of gastric acid and bicarbonate secretions by calcium in guinea pig stomach and amphibian isolated mucosa. Acta Physiol. Scand. 110: 419-426. Flemstrom G., Heylings J.R. and Garner A. 1982. Gastric and duodenal HCO-, transport in vitro: effects of hormones and local transmitters. Amer. J. Physiol. 242: GlOO-GllO.
Control of gut motility
355
Fong J.C., Hong S.-J. and Wang F.-F. 1991. Somatostatin inhibits pepsinogen secretion via a cyclic AMP-independent pathway. Regul. Pept. 31: 351-360. Furness J.B. and Costa M. 1987. The Enteric Nervous System. Churchill Livingstone, Edinburgh. Furuse M. 1999. Release and endogenous actions of the gastrin/cholecystokinin (CCK) family in the chicken. J. Exp. Zool. 283: 448454. Furuse M., Tachibana T., Ohgushi A., Ando R., Yoshimatsu T. and Denbow D.M. 2001. Intracerebroventricular injection of ghrelin and growth hormone releasing factor inhibits food intake in neonatal chicks. Neurosci. Lett. 301: 123-126. Gabella G. 1987. The number of neurons in the small intestine of mice, guinea-pigs and sheep. Neuroscience 22: 737-752. Gabella T. and Blundell D. 1981. Gap junctions of the muscle of the small and large intestine. Cell Tiss. Res. 219: 469488. Galas L., Chartrel N., Kojima M., Kangawa K. and Vaudry H. 2002. Immunohistochemical localization and biochemical characterization of ghrelin in the brain and stomach of the frog Rana esculenta. J. Comp. Neurol. 450: 34-44. Gallego-Huidobro J. and Pastor L.M. 1996. Histology of the mucosa of the oesophagogastric junction and the stomach in adult Rana perezi. J. Anat. 188: 439444. Garrido M.V.O., Torres M.I.N. and Equisoain M.A.A. 1993. Histological, histochemical, and ultrastructural analysis of the gastric mucosa in Oncorhynchus mykiss. Aquaculture 115: 121-132. Gascoigne L., Allen B., Thorndyke M.C. and Bevis P.J.R. 1988. Bombesin 1-14 causes relaxation of circular muscle in the pigeon proventriculus in vitro. Regul. Pept. 22: 405. Gawlicka A., Leggiadro C.T., Gallant J.W. and Douglas S.E. 2001. Cellular expression of the pepsinogen and the gastric proton pump genes in the stomach of winter flounder as determined by in situ hybridization. J. Fish Biol. 58: 529-536. Gentle M.J. and Savory C.J. 1975. The effect of dietary dilution on the intestinal anatomy of the Japanese quail (Coturnix coturnix japonica). Res. Vet. Sci. 19: 284-287. Gibbins I. 1994. Comparative anatomy and evolution of the autonomic nervous system. In: The Autonomic Nervous System. G. Burnstock (ed.). Comparative Physiology and Evolution of the Autonomic Nervous System. S. Nilsson and S. Holmgren (Vol. eds.). Harwood Acad. Publ., Chur, Switzerland, pp. 1-67. Gibbins I.L., Campbell G., Morris J. L., Nilsson S. and Murphy R. 1987. Pathwayspecific connections between peptide-containing preganglionic a n d postganglionic neurons in the vagus nerve of the toad (Bufo marinus). J. Autonom. Nerv. Syst. 20: 43-55. Goyal R.K. and Hirano I. 1996. Mechanisms of disease: The enteric nervous system. New Engl. J. of Medicine. 334: 1106-1115. Grove D.J. and Campbell G. 1979. The role of extrinsic and intrinsic nerves in the coordination of gut motility in the stomachless flatfish Rhombosolea tapirina and Ammotretis rostrata. Comp. Biochem. Physiol. 63C: 143-159. Grove D.J. and Holmgren S. 1992a. Intrinsic mechanisms controlling cardiac stomach volume of the rainbow trout (Oncorhynchus mykiss) following gastric distension. J. Exp. Biol. 163: 33-48. Grove D.J. and Holmgren S. 1992b. Mechanisms controlling stomach volume of the Atlantic cod (Gadus morhua) following gastric distension. J. Exp. Biol. 163: 49-63. Helander H.F. 1981. The cells of the gastric mucosa. Int. Rev. Cytol. 70: 217-289. Hennig G.W., Brookes S.J.H. and Costa M. 1997. Excitatory and inhibitory motor reflexes in the isolated guinea-pig stomach. J. Physiol. 501: 197-212. Himick B.A. and Peter R.E. 1994a. Bombesin acts to suppress feeding behavior and alter serum growth hormone in goldfish. Physiol. and Beh. 55: 65-72. Himick B.A. and Peter R.E. 1994b. CCK/gastrin-like immunoreactivity in brain and gut, and CCK suppression of feeding in goldfish. Amer. J. Physiol. 267: R841-R851.
356
Physiological and ecological adaptations t o feeding in vertebrates
Hirschowitz B.I., Uemura N., Matsumoto H. and Dickinson K.E. J. 1990. The role of Ca, in the time-dependent pepsinogen secretion of frog oesophageal peptic glands stimulated by bombesin. Acta Physiol. Scand. 140: 401412. Hodgkiss J.P. 1986. Intrinsic reflexes underlying peristalsis in the small intestine of the domestic fowl. J. Physiol. 380: 311-328. Hogben C.A.M. 1967. Response of the isolated dogfish gastric mucosa to histamine. Proc. Soc. Exper. Biol. Med. 124: 890-893. Holmberg A., Hagg U., Fritsche R. and Holmgren S. 2001. Occurrence of neurotrophin receptors and transmitters in the developing Xenopus gut. Cell Tiss. Res. 306: 3547. Holmberg A., Kaim J., Persson A., Jensen J., Wang T. and Holmgren S. 2003. Effects of digestive status on the reptilian gut. Comp. Biochem. Physiol., 133A: 499-518. Holmgren S. 1983. The effects of putative non-adrenergic, non-cholinergic autonomic transmitters on isolated strips from the stomach of the rainbow trout, Salmo gairdneri. Comp. Biochem. Physiol. 74C: 229-238. Holmgren S. and Nilsson S. 1983. Bombesin-, gastrin/CCK-, 5-hydroxy-tryptamine, neurotensin-, somatostatin- and VIP-like irnmunoreactivity and catecholamine fluorescence in the gut of the elasmobranch, Squalus acanthias. Cell Tiss. Res. 234: 595-618. Holmgren S. and Jonsson A. C. 1988. Occurrence and effects on motility of bombesin related peptides in the gastrointestinal tract of the Atlantic cod, Gadus morhua. Comp. Biochem. Physiol. 89C: 249-256. Holmgren S. and Jensen J. 2001. Evolution of vertebrate neuropeptides. Brain Res. Bulletin 55: 723-735. Holmgren S., Jensen J., Jonsson A.C., Lundin K. and Nilsson S. 1985. Neuropeptides in the gastrointestinal canal of Necturus maculosus. Cell Tiss. Res. 241: 565-580. Holmgren S., Axelsson M., Jensen J., Aldman G., Sundell K. and Jonsson A.C. 1989. Bombesin-like immunoreactivity and the effect of bombesin in the gut, circulatory system and lung of the caiman, Caiman crocodylus crocodylus, and the crocodile, Crocodylus porosus. Exper. Biol. 48: 261-271. Holstein B. 1976. Effect of the H2-receptor antagonist metiamide on carbachol- and histamine-induced gastric acid secretion in the Atla~ticcod, Gadus morhua. Acta Physiol. Scand. 97: 189-195. Holstein B. 1977. Effect of atropine and SC-15396 on stimulated gastric acid secretion in the Atlantic cod, Gadus morhua. Acta Physiol. Scand. 101: 185-93. Holstein B. 1982. Inhibition of gastric acid secretion in the Atlantic cod, Gadus morhua, by sulphated and desulphated gastrin, caerulein, and CCK-octapeptide. Acta Physiol. Scand. 114: 453-459. Holstein B. 1986. Characterization with agonists of the histamine receptors mediating stimulation of gastric acid secretion in the Atlantic cod, Gadus morhua. Agents and Actions 19: 4247. Holstein B. and Cederberg C. 1980. Effect of vagotomy and glucose administration on gastric acid secretion in the Atlantic cod, Gadus morhua. Acta Physiol. Scand. 109: 37-44. Holstein B. and Humphrey C.S. 1980. Stimulation of gastric acid secretion and suppression of VIP-like immunoreactivity by bombesin in the atlantic codfish, Gadus morhua. Acta Physiol. Scand. 109: 217-223. Holstein B. and Cederberg C. 1984. Effects of 5-HT on basal and stimulated secretions of acid and pepsin and on gastric volume outflow in the in vivo gastrically and intestinally perfused cod, Gadus morhua. Agents and Actions 15: 290-305. Holstein B. and Cederberg C. 1986. Effects of tachykinins on gastric acid and pepsin secretion and on gastric outflow in the Atlantic cod, Gadus morhua. Amer. J. Physiol. 250: G309-G315. Horowitz B., Ward S.M. and Sanders K.M. 1999. Cellular and molecular basis for electrical rhythmicity in gastrointestinal muscles. Annu. Rev. Physiol. 61: 1943. Hoyle C.H.V. 1998. Neuropeptide families: evolutionary perspectives. Regul. Pept. 73: 1-33.
Control of gut motility
357
Huber G.C. 1900. A contribution on the minute anatomy of the sympathetic ganglia of the different classes of vertebrates. J. Morphol. 16: 27-86. Huizinga J.D. 2001. Physiology and pathophysiology of the interstitial cells of Cajal: from bench to bedside. 11. Gastric motility: lessons from mutant mice on slow waves and innervation. Amer. J. Physiol. 281: G1129-G1134. Hukuhara T., Naitoh T., Ochi K. and Nakai-Kameyama H. 1976. Role of the intramural ganglion cells played in the motility of the large intestine of the tortoise, Geoclemys reevesii. Jpn. J. Smooth Muscle Res. 12: 121-130. Jensen J. 2001. Regulatory peptides and control of food intake in non-mammalian vertebrates. Comp. Biochem. Physiol. 128A: 471479. Jensen J. and Holmgren S. 1985. Neurotransmitters in the intestine of the Atlantic cod, Gadus morhua. Comp. Biochem. Physiol. 82C: 81-89. Jensen J. and Holmgren S. 1994. The gastrointestinal canal. In: The Autonomic Nervous System. G. Burnstock (ed.). Comparative Physiology and Evolution of the Autonomic Nervous System. S. Nilsson and S. Holmgren vol. Harwood Acad. Publ., Chur, Switzerland, pp. 119-167. Jensen J., Karila P., Jonsson A.C., Aldman G. and Holmgren S. 1993.' Effects of substance P and distribution of substance P-like immunoreactivity in nerves supplying the stomach of the cod Gadus morhua. Fish Physiol. Biochem. 12: 237-247. Jimenez M., Martinez V, Rodriguez-Membrilla A, Rodriguez-Sinovas A., Gonalons E., and Vergara P. 1994. Rhythmic oscillating complex: characterization, induction, and relationship to MMC in chickens. Amer. J. Physiol. 266: G585-(2595. Johansson A., Holmgren S. and Conlon J.M. 2002. The primary structures and myotropic activities of two tachykinins isolated from the African clawed frog, Xenopus laevis. Peptides 108: 113-121. Johnels A.G. 1956. On the peripheral autonomic nervous system of the trunk region of Lampetra planeri. Acta Zool. (Stockholm) 39: 9-12. Johnsen A.H. 1998. Phylogeny of the cholecystokinin/gastrin family. Frontiers Neuroendocri. 19: 17-33. Jonsson A.C. 1994. Glands. In: The Autonomic Nervous System. G. Burnstock (ed.), Comparative Physiology and Evolution of the Autonomic Nervous System. S. Nilsson and S. Holmgren (vol. eds.). Harwqod Acad. Publ., Chur, Switzerland, pp. 169-192. Jonsson A. C., Holmgren S. and Holstein B. 1987. Gastrin/CCK-like immunoreactivity in endocrine cells and nerves in the gastrointestinal tract of the cod, Gadus morhua, and the effect of peptides of the gastrin/CCK family on cod gastrointestinal smooth muscle. Gen. Comp. Endocri. 66: 190-202. Junquera C., Martinez-Ciriano, C., Castiella, T., Serrano, P., et al., 2001. Enteric plexus and interstitial cells of Cajal: interrelationship in the stomach of Podarcis hispanica (Reptilia). An ultrastructural study. Histol. Histopathol. 16: 869-881. Kaiya H., Kojima M., Hosoda H., Koda A. 2001. Bullfrog ghrelin is modified by noctanoic acid at its thirds threonine residue. J. Biol. Chem. 276: 40441-40448. Kaiya H., Van der Geyten S., Kojima M., Hosoda H. 2002. Chicken ghrelin: purification, cDNA cloning, and biological activity. Endocrinology 143: 3454-3463. Kanazawa T., Ohashi H. and Takewaki T. 1980. Evidence that cell bodies of noncholinergic, excitatory neurons which supply the smooth muscle of the chicken rectum are located in the ganglia of Remak's nerve. Brit. 1. Pharmacol. 71: 519-524. Karila P. 1997. Nervous control of gastrointestinal motility in the Atlantic cod, Gadus morhua: autonomic pathways and enteric reflexes. PhD Thesis, Goteborg University, Sweden. Karila P. and Holmgren S. 1995. Enteric reflexes and nitric oxide in the fish intestine. J. Exp. Biol. 198: 2405-2411. Karila P. and Holmgren S. 1997. Anally projecting neurons exhibiting immunoreactivity to galanin, nitric oxide synthase and vasoactive intestinal peptide, detected by confocal laser scanning microscopy, in the intestine of the Atlantic cod, Gadus morhua. Cell Tiss. Res. 20: 525-533.
358
Physiological and ecological adaptations t o feeding in vertebrates
Karila P., Jonsson A.C., Jensen J. and Holmgren S. 1993. Galanin-like immunoreactivity in extrinsic and intrinsic nerves to the gut of the Atlantic cod, Gadus morhua, and the effect of galanin on the smooth muscle of the gut. Cell Tiss. Res. 271: 537-544. Karila P., Shahbazi F., Jensen J. and Holmgren S. 1998. Projections and actions of tachykininergic, cholinergic, and serotonergic neurones in the intestine of the Atlantic cod. Cell Tiss. Res. 291: 403413. Keef K.D., Murray D.C., Sanders K.M. and Smith T.K. 1997. Basal release of nitric oxide induces an oscillatory motor pattern in canine colon. J. Physiol. 499: 773-786. Kim J.B., Johansson A., Holmgren S. and Conlon J.M. 2001. Gastrin-releasing peptides from Xenopus laevis: purification, characterization, and myotropic activity. Amer. J. Physiol. 281: R902-R908. Koelz H.R. 1992. Gastric acid in vertebrates. Scand. J. Gastroenter. 27: 2-6. Kojima M., Hosoda H., Date Y., Nakazato M., Matsuo H. and Kangawa K. 1999. Ghrelin is a growth-hormone-releasing acylated peptide from the stomach. Nature 402: 556-660. Komori S. and Ohashi H. 1982. Some characteristics of transmission from non-adrenergic, non-cholinergic excitatory nerves to the smooth muscle of the chicken. J. Autonom. Nerv. Syst. 6: 199-210. Komori S. and Ohashi H. 1984. Presynaptic, muscarinic inhibition of non-adrenergic, non-cholinergic neuromuscular transmission in the chicken rectum. Brit. J. Pharm. 82: 73-84. Komori S., Matsuo K., Kanamaru Y. and Ohashi H. 1986. Smooth muscle excitatory substances from Remak's nerve of the chicken and a comparison of their pharmacological and chemical properties with substance P. Jpn. J. Pharm. 40: 1-11. Kunze W.A.A. and Furness J.B. 1999. The enteric nervous system and regulation of intestinal motility. Annu. Rev. Physiol. 61: 117-142. Lamanna C., Assisi L., Lucini C. and Botte V. 1999. Galanin-containing-neurons in the gastrointestinal tract of the lizard Podarcis s. sicula, as components of anally projecting nerve pathway. Neurosci. Lett. 268: 93-96. Langley J.N. 1898. On the union of the cranial autonomic (visceral) fibers with the nerve cells of the superior cervical ganglion. J. Physiol. (Lond.) 208: 240-270. Langley J.N. 1921. The Autonomic Nervous System, part I. Heffer Publ., Cambridge, UK. Le Lievre C.S. and Le Douarin N.M. 1975. Mesenchymal derivatives of the neural crest: analysis of chimaeric quail and chick embryos. J . Embryol. Exper. Morphol. 34: 125-154. Lecoin L., Gabella G. and Le Douarin N. 1996. Origin of the c-kit-positive interstitial cells in the avian bowel. Development 122: 725-733. Lee M.C., Schiffman S.S. and Pappas T.N. 1994. Role of neuropeptides in the regulation of feeding behavior: a review of cholecystokinin,bombesin, neuropeptide Y, and galanin. Neurosci. Biobehav. Rev. 18: 313-323. Liddle R.A. 1997. Cholecystokinin cells. Annu. Rev. Physiol. 59: 221-242. Liddle R.A., Morita E.T., Conrad C.K. and Williams J.A. 1986. Regulation of gastric emptying in humans by cholecystokinin. J. Clin. Invest. 77: 992-996. Lin T.M., Evans D. C., Warrick M.W. and Pioch R.P. 1986. Actions of nizatidine, a selective histamine H-2 receptor antagonist on gastric acid secretion in dogs, rats and frogs, Rana catesbeiana. J. Pharm. Exper. Ther. 239: 406410. Linari G., Baldieri M., Angelucci L. 1975. The action of bombesin on gastric secretion of the chicken. Eur. J. Pharm. 34: 143-150. Lindstrom E., Chen D., NorlCn P., Andersson K. and Hakanson R. 2001. Control of gastric acid secretion: the gastrin-ECL cell-parietal cell axis. Comp. Biochem. Physiol. 128A: 505-514. Lunam C.A. and Smith T.K. 1996. Morphology and projections of neurons in Remak's nerve of the domestic fowl revealed by intracellular injection of biotin. Cell Tiss. Res. 284: 215-222.
Control of g u t motility
359
Lundin K., Holmgren S. and Nilsson S. 1984. Peptidergic functions in the dogfish rectum. Acta Physiol. Scand. 121: 46A. Maake C., Kaufmann C. and Reinecke M. 2001. Ontogeny of neurohormonal peptides, serotonin and nitric oxide synthase in the gastrointestinal neuroendocrine system of the axolotl (Ambystoma mexicanum): an immunohistochemical analysis. Gen. Comp. Endocrinol. 121: 74-83. Martin M.T., Goiialons E. and FernPndez E. 1998. Contribution of inhibitory neurotransmitters to the CCK induced relaxation of the circular muscle of avian ileum. Life Sci. 62: 937-946. Martin M.T., FernPndez E., FernPndez A.G. and Goiialons E. 1994. Mechanisms mediating the effects of cholecystokinin on avian small intestine longitudinal smooth muscle. Regul. Pept. 51: 91-99. Martin M.T., Ferndndez E., Rodriguez-Sinovas A. and Goiialons E. 1995. Effects of cholecystokinin on chicken cecal motility: mechanisms involved. Life Sci. 56: 601-610. Martinez V., Jimenez M., Gonalons E. and Vergara P. 1993. Mechanism of action of CCK in avian gastroduodenal motility-evidence for nitric oxide involvement. Amer. J. Physiol. 265: G842-G850 Martinez V., JimCnez M., GoAalons E. and Vergara P. 1995a. Intraluminal lipids modulate avian gastrointestinal motility. Amer. J. Physiol. 269: R445-R452. Martinez V., JimCnez M., Goiialons E. and Vergara P. 1995b. Modulation of migrating myoelectric complexes by cholecystokinin and gastrin in the gastrointestinal tract of chickens. Poultry. Sci. 74: 563-576. Martinez-Ciriano C., Junquera C., Castiella T., Gomez-Barrena E., Aisa J. and Blasco J. 2000. Intrinsic innervation in the intestine of the lizard Podarcis hispanica. Histol. Histopath. 15: 1093-1105. Masuda Y., Tanaka T., Inomata N., Ohnuma N., et al. 2000. Ghrelin stimulates gastric acid secretion and motility in rats. Biochem. Biophys. Res. Commun. 276: 905-908. Mattisson A. and Holstein B. 1980. The ultrastructure of the gastric glands and its relation to induced secretory activity of cod, Gadus morhua (Day). Acta Physiol. Scand. 9: 51-59. McCoy J.G. and Avery D.D. 1990. Bombesin: potential integrative peptide for feeding and satiety. Peptides 11: 595-607. Merali Z., McIntosh J. and Anisman H. 1999. Role of bombesin-related peptides in the control of food intake. Neuropeptides 33: 376-386. Michelangeli F., Ruiz M.C., Dominguez M.G. and Parthe V. 1988. Mammalianlike differentiation of gastric cells in the shark Hexanchus griseus. Cell Tiss. Res. 251: 225-227. Mitsui R. and Komoro T. 2002. Direct and indirect innervation of smooth muscle cells of rat stomach, with special reference to interstitial cells of Cajal. Cell Tiss. Res. 309: 219-227. Mueller L.R., Duke G.E. and Evanson O.A. 1990. Investigations of the migrating motor complex in domestic turkeys. Amer. J. Physiol. 259: G329-4333. Negri L. and Erspamer V. 1973. Action of caerulein and caerulein-like peptides on "short-circuit current" and acid secretion in the isolated gastric mucosa of amphibians. Naunyn Schmiedebergs Arch. Pharm. 277: 401-412. Newgreen D. and Young H.M. 2002. Enteric nervous system: Development and development disturbances, Part 2. Pediatr. Dev. Pathol. 5: 329-49. Nicol J. 1952. Autonomic nervous system in lower chordates. Biol. Rev. 27: 1 4 9 . Nielsen K.G, Bomgren P., Holmgren S. a n d Johnsen A.H. 1998. Gastrin and cholecystokinin of the bullfrog, Rana catesbeiana, have distinct effects upon gallbladder motility and gastric acid secretion in vitro. Gen. Comp. Endocrin. 112: 247-254. Nieuwkoop P.D. and Faber J. 1994. Normal Table of Xenopus laevis (Daudin). Garland Publ. Inc. New York, NY.
360
Physiological and ecological adaptations t o feeding in vertebrates
Niijima A. 1967. Afferent impulses in the gastric and oesophageal branch of the vagal nerve of toad. Physiol. Behav. 2: 1 4 . Nilsson S. 1983. Autonomic Nerve Function in the Vertebrates. Springer-Verlag, Berlin. Nilsson S. and Holmgren S. 1992. Autonomic nerve function and cardiovascular control in lungfish. In: Physiological Adaptations in Vertebrates. S.C. Wood, R.E. Weber, A.R. Hargens and R.W. Millard (eds.). Marcel Dekker Inc., New York, NY, pp. 377-395. Ohnesorge F.K. and Rehberg M. 1963. Der Darm der Schleie (Tinca vulgaris) als pharmakologisches Versuchsobject. Naunyn Schmiedebergs Arch. Exper. Path. Pharm. 246: 81-82. Ohtani N., Sasaki I., Naito H., Shibata C. and Matsuno S. 2001. Mediators for fatinduced ileal brake are different between stomach and proximal small intestine in conscious dogs. J. GI Surg. 5: 377-382. Olsson C. 2002. Distribution and effects of PACAP, VIP, nitric oxide and GABA in the gut of the African clawed frog Xenopus laevis. J. Exp. Biol. 205: 1123-1134. Olsson C. and Karila P. 1995. Coexistence of NADPH-diaphorase and vasoactive intestinal polypeptide in the enteric nervous system of the Atlantic cod (Gadus morhua) and the spiny dogfish (Squalus acanthias). Cell Tiss. Res. 280: 297-305. Olsson C. and Holmgren S. 2000. PACAP and nitric oxide inhibit contractions in proximal intestine of the Atlantic cod, Gadus morhua. J. Exp. Biol. 203: 575-583. Olsson C. and Holmgren S. 2001. The control of gut motility. Comp. Biochem. Physiol. 128 A: 481-503. Olsson C., Aldman G., Larsson A. and Holmgren S. 1999. Cholecystokinin affects gastric emptying and stomach motility in the rainbow trout Oncorhynchus mykiss. J. Exp. Biol. 202: 161-170. Orihata M. and Sarna S.K. 1994. Inhibition of nitric oxide synthase delays gastric emptying of solid meals. J. Pharm. Exper. Ther. 271: 660-670. Patterson T.L. and Fair E. 1933. The action of the vagus on the stomach-intestine in the hagfish. Comparative studies. VIII. J. Cell. Comp. Physiol. 3: 113-119. Pick J. 1970. The Autonomic Nervous System: Morphological, Comparative, Clinical and Surgical Aspects. J.B. Lippincott, Philadelphia, PA. Prosser C.L. 1995. Rhythmic electrical and mechanical activity in stomach of toad and frog. Amer. J. Physiol. 269: G386-G395. Rajjo I.M., Vigna S.R. and Crim J.W. 1988. Actions of cholecystokinin-related peptides on the gallbladder of bony fishes in vitro. Comp. Biochem. Physiol. 90C:267-273. Ray D.L. 1950. The peripheral nervous system of Lampanychtus leucosparsus. J. Morph. 87: 61-178. Rebolledo I.M. and Vial J.D. 1979. Fine structure of the oxynticopeptic cell in the gastric glands of an elasmobranch species (Halaelurus chilensis). Anat. Rec. 193: 805-822. Reinecke M., Muller C. and Segner H. 1997. An immunohistochemical analysis of the ontogeny, distribution and coexistence of 12 regulatory peptides and serotonin in endocrine cells and nerve fibers of the digestive tract of the turbot, Scophthalmus maximus (Teleostei). Anat. Embry. (Berl.). 195: 87-101. Reynhout J.K. and Duke G.E. 1999. Identification of interstitial cells of Cajal in the digestive tract of turkeys (Meleagris gallopavo). J. Exp. Zool. 283: 426-440. Richardson R.D., Boswell T., Weatherford S.C., Wingfield J.C. and Woods S.C. 1993. Cholecystokinin octapeptide decreases food intake in white-crowned sparrows. Amer. J. Physiol. 264: R852-R856. Rodriguez-Sinovas A., Fernandez E. and Gonalons E. 1996. Central and NO-mediated mechanisms are involved in the inhibitory effects of CCK on the chicken cecorectal area. Life Sci. 58: 1869-1882. Rodriguez-Sinovas A., Martin M.T., Fernandez E. and Gonalons E. 1994. Cecocolonic motility in the chicken. Effects of cholecystokinin. Life Sci. 55: 1743-1755. Rodriguez-Sinovas A., Fernandez E., Manteca X., Fernandez A.G. and Gonalons E. 1997a. CCK is involved in both peripheral and central mechanisms controlling food intake in chickens. Amer. J. Physiol. 272: R334-R340. '
Control of gut motility
361
Rodriguez-Sinovas A., Jimenez M., De Qlercq P., Peters T.L. and Vargara P. 1997b. Rhythmic oscillating complexes in gastrointestinal tract of chickens: a role for motilin. Amer. J. Physiol. 272: G916-G922. Ruiz M.C. and Michelangeli F. 1984. Evidence for a direct action of acetylcholine on the gastric oxyntic cell of the amphibian. Amer. J. Physiol. 246: G16-G25. Ruiz M.C. and Michelangeli F. 1986. Stimulation of oxyntic and histaminergic cells in the gastric mucosa by gastrin C-terminal tetrapeptide. Amer. J. Physiol. 251: G529-G537. Ruoff H.J. and Sewing K.-F. 1972. Inhibition of gastric acid secretion in chickens by atropine, insulin and 2-deoxy-D-glucose. Naunym Schmiedeb Arch. Pharm. 19: 296-300. Sadaghiani B. and Vielkind J. R. 1990. Distribution and migration pathways of HNK1-immunoreactive neural crest cells in teleost fish embryos. Development 110: 197-209. Sanders K.M. 1996. A case for interstitial cells of Cajal as pacemaker and mediators of neurotransmission in the gastrointestinal tract. Gastroenter. 111: 492-515. Sarna S.K. 1985. Cyclic motor activity: Migrating motor complexi Gastroenter. 89: 894-913. Savory C.J. 1987. An alternative explanation for apparent satiating properties of peripherally administered bombesin and cholecystokinin in domestic fowls. Physiol. Behav. 39: 191-202. Savory C.J. and Gentle M.J. 1980. Intra-venous injections of cholecystokinin and caerulein suppress food intake in domestic fowls. Experientia (Basel) 36: 1191-1192. Savory C.J. and Hodgkiss J.P. 1984. Influence of vagotomy in domestic fowls on feeding activity, food passage, digestibility and satiety effects of two peptides. Physiol. Behav. 33: 937-944. Savory C.J., Duke G.E. and Bertoy R.W. 1981. Influence of intravenous injections of cholecystokinin on gastro-intestinal motility in turkeys and domestic fowls. Comp. Biochem. Physiol. 70A: 179-190. Schemann M., Schaaf C. and Mader M. 1995. Neurochemical coding of enteric neurons in the guinea pig stomach. J. Comp. Neurol. 353: 161-178. Schlamp J., Duke G.E. and Evanson O.A. 1991. The relationship between feed intake and frequency of the migrating motor complex in domestic turkeys: is there a cephalic phase? Poultry. Sci. 70: 1758-1762. Secor S.M., Fehsenfeld D., Diamond J. and Adrian T.E. 2001. Responses of python gastrointestinal regulatory peptides to feeding. Proc. Natl. Acad. Sci. USA. 98:1363713642. Shepherd B.S., Eckert S.M., Parhar I.S., and Vijayan M.M. 2000. The hexapeptide KP102 (D-Ala-D-B-Nal-Ala-Trp-D-Phe-Lys-NH2) stimulates growth hormone release in a cichlid fish (Orechromis mossambicus). J. Endocrin. 167: R7-R10. Shirakawa T. and Hirschowitz B.I. 1985. Interaction between stimuli and their antagonists on frog, Rana catesbeiana, esophageal peptic glands. Amer. J. Physiol. 249: G668G673. Shochina M., Belai A., Toole L., Knight G. and Burnstock G. 1997. Neurochemical coding in the myenteric plexus of the upper gastrointestinal tract of hibernating hamsters. lnt. J. Dev. Neurosci. 15: 353-362. Simpson L., Goldenberg D. and Hirschowitz B.I. 1980. Pepsinogen secretion by the frog esophagus in vitro. Amer. J. Physiol. 238: G79-G84. Smit H. 1968. Gastric secretion in the lower vertebrates and birds. In: Handbook of Physiology, sect. 6: Alimentary Canal, vol. V: Bile, Digestion, Ruminal Physiology. C.F. Code (ed.). Amer. Physiol. Soc. pp. 2791-2805. Smith T.K. and Lunam C.A. 1998. Electrical characteristics and responses to jejunal distension of neurons in Remak's juxta-jejunal ganglia of the domestic fowl. J. Physiol. 510: 563-575.
362
Physiological and ecological adaptations t o feeding in vertebrates
Sullivan M.X. 1905-6. The physiology of the digestive tract of elasmobranchs. Amer. J. Physiol. 15: 42-45. Tanaka T., VanKlompenberg L.H. and Sarr M.G. 2001. Selective role of vagal and nonvagal innervation in initiation and coordination of gastric and small bowel patterns of interdigestive and postprandial motility. J. GI Surg. 5: 418433. Thorndyke M.C. and Holmgren S. 1990. Bombesin potentiates the effect of acetylcholine on isolated strips of fish stomach. Regul. Pept. 30: 125-135. Timmermans J.P., Scheuermann D.W., Gabriel R., Adriaensen D., Fekete E. and De Groodt-Lasseel M.H.A. 1991. The innervation of the gastrointestinal tract of a chelonian reptile, Pseudemys scripta elegans. I . Structure and topography of the enteric nerve plexuses using neuron-specific enolase immunohistochemistry. Histochem. 95: 397-402. Toole L., Belai A., Shochina M. and Burnstock G. 1999. The effects of hibernation on the myenteric plexus of the golden hamster small and large intestine. Cell Tiss. Res. 296: 479-487. Torsoli A. and Severi C. 1993. The neuroendocrine control of gastrointestinal motor activity. J. Physiol. 87: 367-374. Villani L. 1999. Developmental pattern of NADPH-diaphorase activity in the peripheral nervous system of the cichlid fish Tilapia mariae. Eur. J. Histochem. 43: 301-310. Visvanathan R. 1992. Effect of luminal acidification on guinea pig gastric mucosa. Dig. Dis. Sci. 37: 1600-1605. Wang G., Lee H.-M., Englander E. and Greeley Jr., G.H. 2002. Ghrelin - not just another stomach hormone. Regul. Pept. 105: 75-81. Waschek J.A. 1995. Vasoactive intestinal peptide: an important trophic factor and developmental regulator? Devl. Neurosci. 17: 1-7. Waschek J.A., Casillas R.A., Nguyen T.B., DiCicco-Bloom E.M., Carpenter E.M. and Rodriguez W.I. 1998. Neural tube expression of pituitary adenylate cyclase-activating peptide (PACAP) and receptor: potential role in patterning and neurogenesis. Proc. Natl. Acad. Sci. U S A . 4: 9602-9607. Wong, W.C. and Tan, C.K. (1978). Fine structure of the myenteric and submucous plexuses in the stomach of a coral fish, Chelmon rostratus Cuvier. J. Anat. 126, 291-301. Yasugi S. 1987. Pepsinogen-like immunoreactivity among vertebrates: Occurrence of common antigenicity to an anti-chicken pepsinogen antiserum in stomach gland cells of vertebrates. Comp. Biochem. Physiol. 86B: 675-680. Yasugi S., Matsunaga T. and Mizuno T. 1988. Presence of pepsinogens immunoreactive to anti-embryonic chicken pepsinogen antiserum in fish stomachs: Possible ancestor molecules of chymosin of higher vertebrates. Comp. Biochem. Physiol. 91A: 565-569. Young H.M. 1990. The ultrastructure of the intestinal nerve of Remak in the domestic fowl. Cell Tiss. Res. 260: 601-616. Young H.M. 1999. Embryological origin of interstitial cells of Cajal. Microsc. Res. Technique 47: 303-308. Young H.M. and Newgreen D. 2001. Enteric neural crest-derived cells: origin, identification, migration and differentiation. Anat. Rec. 262: 1-15.
Effects of Dietary Fatty Acids on the Physiology of Environmental Adaptation in Fish David J. McKenzie Centre de Recherche sur les ~ c o s ~ s t & mMarins es et Aquacoles de L'Houmeau, CNRS-Ifremer, L'Houmeau, France
SYNOPSIS Evidence that the physiology of environmental adaptation in fish is influenced significantly by the spectrum of fatty acids (FA) consumed in the diet is reviewed. Although the studies in this field are limited, they nonetheless reveal that dietary FA intake and consequent tissue FA phenotype have a profound influence on metabolic rate, tolerance of hypoxia, cardiac performance, exercise performance, and osmoregulatory ability of fish. They also influence tolerance of a range of environmental stressors in fish larvae. The studies are largely descriptive with little understanding as yet of the mechanisms by which the phenomena occur. Effects on metabolic rate, tolerance of hypoxia, and cardiac performance may be due to changes in mitochondria1 membrane FA composition and consequent function; effects on exercise performance may be linked to preferences in use of FA as substrates for aerobic metabolism, while effects on osmoregulation are presumably due to changes in gill membrane FA composition and function, including eicosanoid formation. These effects translate across levels of organismal organization to influence the physiology of the whole animal and there can be direct relationships between tissue content of particular FAs and traits of metabolism or performance. To date, all studies have employed artificial diets and were aimed at improving conditions and productivity of fish in aquaculture. Diversities in tissue FA composition within natural fish populations are, however, wider than those elicited in these laboratory studies. If such natural diversity in phenotype is associated with a similar diversity in physiological traits, then this would imply that dietary FA intake can make a significant contribution to phenotypic plasticity, and may have a profound effect on the physiology of environmental adaptation within wild fish populations.
364
Physiological and ecological adaptations t o feeding in vertebrates
Diet is the primary interface for chemical interaction between animals and their environment, and nutrition provides all the structural materials and energy required for expression of the genotype, the phenotype. The interaction between animals and their environment through their diet stretches back throughout evolutionary history and has become extremely complex. This complexity is revealed by the large number of "essential" nutrients that animals require in their food and the syndromes associated with deficiency in these compounds have been the focus of much research. There is also, however, a growing body of evidence to indicate that the relative intake of specific nutrients can have a profound effect on the phenotype and physiology of well-nourished animals (Crawford and Marsh, 1989;.Artset al., 2001; McKenzie, 2001; Winberg et al., 2001). One major nutrient group of particular interest in this regard is the fats, in particular the fatty acids (FA).The FAs have three primary roles in animals: as fuels catabolizedby P-oxidation, as an integral part of phosphoglycerides, the structural units of biological membranes, and as precursors for the formation of eicosanoids, a complex cascade of paracrine hormones derived from specific FAs. Eicosanoids exert a variety of effects on cell and tissue metabolism and function and are generally produced under stressful situations (Lands, 1991; Rowley et al., 1995; Sargent et al., 1999).The structure of FAs is relatively simple,namely a carbon backbone of varying length stretched between a methyl group and a carboxyl group. In saturated FAs (SFA),all carbons in the chain are completely hydrogenated; monounsaturated FAs (MUFA) contain one double bond between adjacent carbons, polyunsaturated FAs (PUFA) contain multiple methylene-interrupted double bonds. There are two recognized methods of FA nomenclature. In the most commonly used nomenclature linoleic acid, for example, is gven as 18:2n-6. The nurrlber before the colon gives the number of carbon atoms in the molecule (18in this example);the number after the colon reports the number of double bonds (2 for linoleic acid), while n-6 defines the carbon at which the first double bond occurs, counting from the terminal methyl group. This nomenclature, where carbons are counted from the methyl terminus, is also known as the omega nomenclature (linoleicacid being 18:20.16).The delta nomenclature, on the other hand, describes the structure of FAs by counting carbons starting from the other end of the molecule, the terminal carboxyl group. This nomenclature is not generally used to describe FAs themselves, but rather their biosynthesis and transformation. Fatty acid desaturases, enzymes which insert double bonds into the carbon backbone, are known as delta-desaturases according to their site of action on the backbone. Thus a delta-12 desaturase inserts a double bond in 18:ln-9 to transform it into 18:2n-6.
Dietary fatty acids in fish
365
The SFA, and most MUFA, can be synthezised de novo by animals. However, the PLTFA of the n-3 and n-6 groups, which have important biological roles in animals, cannot be synthezised de novo because animals lack the delta-12 and delta-15 desaturases required to insert a double bond at the n3 and n-6 position of the carbon backbone. Therefore, linoleic acid and alinolenic acid (18:3n-3) are nutritionally essential and must be obtained through the diet from plants. In actual fact, within the animal these FA are not essential per se, but rather are the precursors for the formation, by elongation and desaturation reactions, of three long chain and highly unsaturated fatty acids (HUFA)which are essential for normal growth and development. These are docosahexanoicacid (DHA,22:6n-3);eicosapentanoic acid (EPA, 20:5n-3) and arachidonic acid (20:4n-6). Futhermore, in all marine fish species studied to date, these HUFA are themselves essential; the fish are unable to create them by elongation and desaturation of their 18Cprecursors (Sargent et al., 1999).It is believed that marine fish may have lost (or never developed) the ability to synthesize essential HUFA because they are readily available in the marine food web (Sargent et al., 1999). Indeed, although the FA composition of animal tissues is a product of dietary intake and endogenous metabolism, diet actually plays the major role in determining the fatty acid composition in the tissues of well-nourished vertebrates.This is because the spectrum of FAs consumed in the diet accumulates in tissue lipid stores and some are utilized for insertion into phosphoglycerides.As a result, the FAs formed in primary producers that accumulate through the food web are markers of nutritional history. Interestingly, the FAs found within foodwebs can differ very significantly, which is particularly true for the essential HUFA (Sargent and Whittle, 1981. Crawford et al., 1999). Those of the n-3 series (n-3HUFA),namely EPA and DHA, are typical of aquatic, in particular temperate marine food webs, where they accumulate from algal primary producers (Sargent and Whittle, 1981). They are much rarer in terrestrial food webs, where primary producers mostly create shorter (18C)PUFA, primarily of the n-6 series (Crawfordet al. 1999). Fatty acid metabolism in vertebrates is extremely complex. Because FAs differ in properties according to their chain length and degree of saturation, the dietary intake has the potential to influence the use of fats as fuels, the physical and biological properties of membranes, and the characteristics of eicosanoid production. Among the FAs, SFA and MUFA appear to be preferred over PUFA as fuels for p-oxidation (Henderson and Sargent, 1985; Side11and Driedzic, 1985)and the essential HUFA are not energy substrates but rather have a generalized role in maintaining the structure and integrity of cell membranes, and EPA and AA have specific roles as precursors for the formation of eicosanoids (Sargent et al., 1999).The eicosanoids produced from AA and EPA differ in their biological effects and there is competitive inhibition between these two HUFA in eicosanoid formation, and between their eicosanoid products, so it is the relative intake of these essential HUFA that determines the final biological effect (Sargent et al., 1999).
366
Physiological and ecological adaptations t o feeding in vertebrates
Recently, some nutritionists have proposed that the availability of particular FAs in the diet, notably the n-3 HUFA, was a driving force in vertebrate evolution (Crawford and Marsh, 1989;Crawford et al., 1999). In vertebrate neuronal and retinal tissues, DHA can comprise up to 50%of the fatty acids in membrane phospholipids, where it is essential for their biological activity.Formation of this n-3HUFA by elongation and desaturation of linolenic acid is very slow in vertebrates; yet provision of adequate DHA in the diet of human neonates is conditionally essential for optimal development of visual acuity and intellectual capacity (Crawford et al., 1999; Broadhurst et al. 2002). It has been suggested, therefore, that evolution of the large brain characteristicof Homo sapiens would only have been permissible in populations with a high consumption of marine or lacustrine food resources, and could not have occurred within the terrestrial food webs of the African savanna (Broadhurst et al., 1998;Crawford et al., 1999; Broadhurst et al., 2002). There is also compelling evidence from studies on mammals to indicate that dietary FA intake, and consequent tissue FA phenotype, have multifaceted effects on the physiology and performance of individuals and on the relative morbidity of populations. Clinical and epidemiological studies have established that dietary FA have profound effects on human health, with major causes of death in Western societies such as degenerative cardiovascular diseases, strokes, and cancer, being linked to aspects of dietary FA intake (Bang and Dyerberg, 1985; Dolecek, 1992; Holman, 1998; Connor, 2000; Stillwell, 2000). Specifically,diets containing too many saturated fatty acids (SFA) and/or PUFA of the n-6 series have negative effects on human health, whereas diets rich in n-3 HUFA are beneficial. Therefore, a healthy human diet should be low in SFA and have a high n-3/n-6 ratio (British Nutrition Task Force, 1992). This requirement of humans for "aquatic" nutrients is an interesting correlate of the colonization of land during vertebrate evolutionary history. The question that arises is how such nutrients, characteristic of aquatic food webs, might influence the phenotype, and consequent physiology, of fish. This chapter reviews the existing information on how dietary FAintake, and consequent tissue FA phenotype, can exert profound effects on the bioenergetics, physiological performance, and environmental tolerance of fish. Although the number of studies is limited, the results are compelling. All of the studies were performed to investigate the impact of diet on fish reared in aquaculture and although many were aimed at identifying potentially beneficial effects of n-3 HUFA, these were not the only FAs that proved to have marked effectson fish physiology and performance. The research is largely descriptiveinnature with little understanding as yet of the underlying mechanisms. This review concludes with some speculations regarding potential mechanisms for the diverse effects of the FAs, followed by consideration of the ecological implications of dietary FA intake for natural fish populations.
Dietary fatty acids in fish
367
EFFECTS OF TISSUE FA PHENOTYPE ON RESPIRATORY METABOLISM AND TOLERANCE OF HYPOXIA
Among the first beneficial effects of n-3 HUFA to be described in mammals was that they influence heart function and improve resistance to cardiac and cerebral ischaemia (Hornstra, 1989; Hock et al., 1990; Charnock et al., 1992; Ellis et al., 1992; Paulson et al., 1992). An important element of ischemia is that it causes local tissue hypoxia (Ford, 2002). Unlike most airbreathing vertebrates, water-breathing fish often experience periods of hypoxia as a result of the low solubility of 0,in water, which can be a common stress in intensive aquaculture. This led us to investigatewhether n-3 HUFA levels in cultured fish diets influenced tolerance of hypoxia. An initial study (Randall et al., 1992) compared responses to a gradual reduction in water 0,partial pressure (PO,) by Adriatic sturgeon (Acipenser naccarii) fed for six months with either a normal commercial diet or the same diet supplemented with n-3 HUFA (as 80g kg-' dry weight of menhaden, Brevoortia tyrranus, oil) and vitamin E (as 500 mg kg-l tocopherol acetate), When the sturgeon were allowed to consume 0, in a closed system, those fed the supplemented diet showed less severe reductions in blood oxygen levels, and increases in plasma lactate concentrations than did those fed the control diet, and also appeared to be less agitated. This study demonstrated that dietary lipids could have a measurable influence on the respiratory physiology of fish and on their tolerance for hypoxia. Subsequent studies compared two experimental diets that were isonitrogenous and isocaloric, differing only in FA and n-3 HUFA composition of their dietary lipids. The diets were chosen to represent the opposing ends of a theoretical health spectrum with n-3 HUFA at one extreme and SFA at the other. Two economically important species of farmed fish were studied: a chondrost, the Adriatic sturgeon, and a teleost, the European eel (Anguilla anguilla). Fish were maintained for extended periods (6 months to 1year) on a commercial diet supplemented either with menhaden (Brevoortia tyrannus) oil, a marine fish oil, or with coconut (Cocos nucifera) oil, a terrestrial vegetable oil. The menhaden oil (MO) diet was rich in n-3 HUFA such as EPA and DHA, the coconut oil (CO)diet rich in SFA such as myristate (14:O)and stearate (18:O; McKenzie et al., 1995a, 2000). No differences in growth relatable to the kind of oil supplement given were detected (McKenzie et al., 1994, 2000). The diets lead to consistent differences in animal phenotype, however, in terms of their tissue FA composition. For both sturgeon and eel, the subjects fed MO had higher levels of EPA and DHA in liver, muscle, and heart total lipids than did those fed CO, which had higher levels of myristate and stearate (Agnisola et al., 1996; McKenzie et al., 1997,2000). The primary impact of hypoxia is that it limits the ability of fish to provide oxygen for aerobic metabolism. Aerobic metabolism is typically
368
Physiological and ecological adaptations t o feeding i n vertebrates
measured as oxygen uptake (MO,), which is stoichiometrically related to energy consumption. Preliminary measurements of MO, in sturgeon revealed that fish fed the diet enriched in n-3 HUFA had a reduced metabolic rate (McKenzie et al., 1995a),so this feature was explored further. An automated respirometry system (Steffensen,1989; McKenzie et al., 1995a)was employed to investigatethe effects of the diets on patterns of aerobic metabolism and standard metabolic rate (SMR),at the level of minimum energy requirements for organismal maintenance. Measurements of MO, were collected on resting postabsorptive sturgeon and eel once every 10min over an 8 h period, and a frequency distribution of instantaneous MO, was described (McKenzie et al., 1997,2000). Dietary fatty acid composition had marked effects on patterns of MO,, as shown in Fig. 14.1. In both, sturgeon and eel, the MO phenotype consumed oxygen in a narrow range of low rates, while the CO phenotype consumed it in a wide range of more elevated rates. As a result, the MO phenotype had a significantly lower mean MO, than the CO phenotype (McKenzie et al., 1997,2000).The lowest rates of MO, measured under these circumstances can be taken as an estimate of SMR, which was significantly lower in both sturgeon and eel fed n-3 HUFA rather than SFA (McKenzie,2001). The close parallel in the manner by which dietary n-3 HUFA versus SFA influenced Mo, in sturgeon and eel indicates that dietary fats have similar effects on metabolic rate in very different taxonomic groups (McKenzie,2001). Interestingly, the relative dietary intakes of SFA and PUFA can also have effectson metabolic rate in terrestrial vertebrates. In herbivorous rodent hibernators, increased dietary PUFA intake had no effect on metabolic rate during arousal but caused more profound declines of metabolic rate in hibernation, and longer bouts of torpor (Florant, 1998). In desert iguanas (Dipsosaurusdorsalis), increasing body temperature from 30°C to 40°C had no effect on the metabolic rate of animals fed diets rich in PUFA, but doubled the metabolic rate of iguanas fed a diet rich in SFA (Simandle et al., 2001). Furthermore, in a direct parallel with the effects on metabolic rate described in fish, Pepe and MacLennan (2002)found that the 0, consumption of isolated rat hearts, working in vitro, was significantly higher in animals previously fed a diet rich in SFA compared to hearts isolated from animals fed a diet rich in n-3 HUFA. Unfortunately, these authors did not measure the metabolic rates of the rats fed the two diets. A series of experimentswere performed to investigatethe extent to which the differences in metabolic rate between MO and CO phenotypes of sturgeon and eel might influence their relative tolerance of hypoxia. For the sturgeon there was clear evidence that the CO phenotype, with elevated metabolic rate, was less tolerant of hypoxia than the MO phenotype. Figure 14.2shows the effects of exposure to moderate hypoxia (PO, = 6.6 kPa) for 3 h, followed by 3 h normoxic recovery, on MO, as measured every 10 min with the automated respirometry system (McKenzieet al., 1995a).The MO
Dietary fatty acids in fish
1n
CO mean =
-
rng Kg.1 h-I
m e a n = 1 8 6 - 1 7 m g Kg-' h'' 25
B -
MO mean = 4 7 - 9 m g Kg-'h"
25
t
m e a n = 85
- 13 m g Kg.'
h.'
-
7
-
Rates of 0, uptake (mg Kg-' h-')
Fig. 14.1. Frequency distribution of rates of oxygen uptake in (A) Adriatic sturgeon (Acipenser naccarii) and ( B ) European eels (Anguilla anguilla) fed diets supplemented with menhaden oil or coconut oil, as measured every 10 min for 8h in normoxia. For sturgeon, n = approx. 330 observations on 7 fish from each group; for eels n = approx. 660 observations on 18 animals from each group. MO, menhaden oil phenotype, CO, coconut oil phenotype. Each graph also reports the mean (-c 1 SEM) value of normoxic 0, uptake for that phenotype (n = 7 for sturgeon, 18 for eels). In both species, mean normoxic 0, uptake of the MO phenotype was significantly lower than the CO phenotype. Modified from McKenzie et al. (1997, 2000).
phenotype showed no effects of hypoxia on M02whereas the CO phenotype was unable to maintain MO, at normoxic levels and as a result exhibited a significanthypoxic depression of aerobic metabolism. Furthermore, as can be seen in Fig. 14.2, the CO phenotype also consumed more oxygen than the MO phenotype during 3 h recovery in normoxia, evidence of a larger "oxygen debt". Differences in ventilatory responses to hypoxia were also found between the two phenotypes (McKenzie et al., 1997). Exposure to three consecutive levels of hypoxia (30min at p02= 11,6,and 4.5 kPa) did not affect ventilation significantly in the MO phenotype but elicited hyperventilation in the CO L
370
Physiological and ecological adaptations t o feeding i n vertebrates
50
5
CON
HYPOXIA
RECOVERY
Fig. 14.2. Mean (+ SE) oxygen consumption (MO,) in normoxia (CON), 3h hypoxia (PO, = 6.6 kPa) and 3h recovery to normoxia in Adriatic sturgeon fed diets supplemented with menhaden oil or coconut oil. n = 9 in both cases. Simple line = water PO,. Area between dotted lines = period when water became hypoxic. MO, menhaden oil phenotype; CO, coconut oil phenotype. In hypoxia, the MO phenotype was able to regulate MO, at the same routine rate as measured in normoxia, whereas the CO phenotype was not, and exhibited a significant reduction in 0, from its routine normoxic rates. Modified from McKenzie et al. (1995a).
phenotype. This hyperventilation by the CO phenotype probably reflected greater stimulation of vascular oxygen receptors sensitive to rates of 0, delivery by the blood (Burlesonet al., 1992; McKenzie et al., 1995b),because the CO phenotype exhibited a much more profound decline in blood total 0, content than the MO phenotype during hypoxia (McKenzie et al., 1997). Therefore, hyperventilation in the CO phenotype indicates that for any particular degree of hypoxia the sturgeon experienced greater problems with maintenance of tissue oxygen delivery, presumably because of their elevated metabolic rate. Indeed, despite the adaptive hyperventilatory responses, the CO phenotype was unable to maintain oxygen uptake at an hypoxic pO, of 6.6 kPa (see above). By the same token, the absence of hyperventilation in the
Dietary fatty acids in fish
371
Fig. 14.3. Mean (+ SE) swimming speed due to spontaneous activity in normoxia and hypoxia (PO, = 10.8 kPa) in Adriatic sturgeon fed the diets supplemented with menhaden oil or coconut oil. n = 6. Area between dotted lines = period when water became hypoxic. MO, menhaden oil phenotype, CO, coconut oil phenotype; BL, body lengths. In hypoxia, the MO phenotype was able to maintain spontaneous swimming activity at the same speed as measured in normoxia, whereas the CO phenotype was not, and exhibited a significant reduction in swimming speed relative to its spontaneous normoxic values. Modified from McKenzie et al. (1995a).
MO phenotype, at least at the 3 levels of hypoxia tested, presumably indicates that the sturgeon were able to meet their lower tissue oxygen demands without an adaptive ventilatory response. This result is consistent with their measured ability to regulate metabolic rate at an hypoxic PO, of 6.6 kPa (see above). There is also behavioral evidence that the CO phenotype experienced problems with maintenance of tissue oxygen delivery in hypoxia. In fish, reductions in spontaneous activity in hypoxia are considered to be a response necessary for reducing energy expenditure (Nilsson et al., 1993; Schurmann and Steffensen, 1994) and so, conversely, the ability of fish to maintain spontaneous activity may be considered a sensitive indicator of their capacity to tolerate hypoxia. Spontaneous locomotory activity of the two phenotypes was measured in normoxia and hypoxia (McKenzie et al. 1995a),using a computerized imaging technique (Schurmannand Steffensen,
372
Physiological and ecological adaptations t o feeding in vertebrates
1994).As shown in Fig. 14.3, the two phenotypes were spontaneouslyactive in normoxia, with similar levels of swimming activity. upon exposure for 3h to an hypoxic PO, of 10.8 kPa, the CO phenotype exhibited a significanthypoxic depression of spontaneous locomotor activity, an effect not observed in the MO conspecifics (McKenzieet al., 1995a). This, then, is another clear indication that the CO phenotype was more sensitive to hypoxia because the fish were obliged to enact energy-saving adaptations that were not observed in the MO sturgeon. In eel, the marked differences in metabolic rate between the two phenotypes were not associated with the profound differencesin tolerance of hypoxia observed in sturgeon (McKenzieet al., 2000). The eel, however, has an exceptional tolerance of hypoxia and hypoxemia (van Waarde, 1983; McKenzie et al., 2002,2003), and possesses cardiorespiratory and percutaneous gas-exchange systems able to regulate tissue oxygen supply and metabolic rate despite extreme reductions in blood oxygen content (McKenzie et al., 2002,2003).Therefore, the relationship between metabolic rate and tolerance of hypoxia depends on a species' innate capacity to regulate metabolism in hypoxia. The effects of diet on phenotype did not transcend the genotypic characteristics of hypoxia tolerance in the eel (McKenzie, 2001).
ow ever,
DIFFERENT LEVEL OF ORGANISMAL ORGANIZATION: EFFECTS OF TISSUE FATTY ACID PHENOTYPE ON PERFORMANCE OF THE HEART It is well established in mammals that dietary n-3 HUFA levels have significant beneficial effects on cardiac physiology (Hornstra 1989;Connor 2000). h particular, elevated dietary and tissue n-3 HUFA levels improve resistance to, and recovery from, reduced perfusion of the heart, i.e. cardiac ischemia (Hocket al., 1990;Charnock et al., 1992).An important element of ischemia is the associated tissue hypoxia. Bell et al. (1991,1993)found that diets containing sunflower oil, with a very low n-3/n-6 ratio, caused development of cardiac lesions in Atlantic salmon (Salmosalar),which were linked in turn to mortal susceptibility to a live transport-induced stress syndrome. With this in mind, cardiac performance in the sturgeon and eel MO and CO phenotypes was investigated, using an in-vitro saline-perfused isolated spontaneouslyworking heart preparation (Agnisola et al., 1996). Studies in sturgeon revealed that hearts isolated from the MO phenotype had a greater ability to perform work than those from the CO phenotype. That is, measurement of the amount of work that the hearts were able to do (power output, PO) revealed that the MO phenotype hearts had a lower basal power output under physiological conditions of input and output pressures (Fig. 14.4),but a greater in-vitro scope for cardiac work (almost twice the maximal to basal PO ratio relative to the CO phenotype; Agnisola
Dietary fatty acids in fish
Fig 24.4. Mean (* SE) in vitro basal cardiac power output (A), and maximum cardiac power output following an acute change from oxygenated (hatched bars) to aerated (open bars) perfusate (B), in spontaneously working hearts isolated from Adriatic sturgeon fed diets supplemented with menhaden oil or coconut oil. n = 5. MO, menhaden oil phenotype, CO, coconut oil phenotype. Basal cardiac power output of the MO phenotype was lower than the CO phenotype (A). The MO phenotype was able to maintain maximum power output when perfusate oxygen levels were reduced (change from oxygenated to aerated saline) whereas the CO phenotype suffered a significant decline in maximum power output (B). Modified from Agnisola et al. (1996).
et al., 1996).Sensitivity of the hearts to reduced oxygen supply was assessed in sturgeon hearts working at maximum PO, by instigating an acute reduction in perfusate oxygen content. This delivered a hypoxic stimulus to the heart (Agnisola et al., 1996). As shown in Fig. 14.4, in hearts from the CO phenotype reduction in oxygen supply caused a significant (circa 40%) decrease in the maximal in-vitro PO, but had no effect on hearts from the MO phenotype (Agnisola et al., 1996). This is an unusual apparent parallel of the effects observed in the whole animal, wherein the CO phenotype showed hypoxic depressions of oxygen consumption and spontaneous activity that were not observed in their MO conspecifics (McKenzie et al., 1995a). It seems reasonable to speculate therefore, that the different sensitivities to reduced 0, supply in the isolated hearts might be linked to differences in tissue metabolic oxygen demand between the two groups. Indeed, Pepe and MacLennan (2002) reported that hearts isolated from rats fed diets rich in n-3 HUFA had a lower 0, consumption
374
Physiological and ecological adaptations t o feeding i n vertebrates
than those from rats fed SFA, and found that the reduced metabolic rate of n-3 HUFA hearts was linked to improved mechanical efficiency (i.e.work done per mole 0, consumed) and a significantly increased tolerance of ischemia. In the eel, no effects of dietary oil on sensitivityto reduced oxygen supply were observed in isolated hearts working in vitro (C. Agnisola, D.J.McKenzie, unpubl. obs.). There were no differences in spontaneous heart rate or PO, and an acute change from perfusion with a saline solution equilibrated with pure oxygen to perfusion with a saline solution equilibrated with air had little effect on PO in either group (C. Agnisola, D.J.McKenzie, unpubl. obs.). Thus these results are also consistent in some respects with those obtained in the whole animal inasmuch as they revealed that differences in tissue oxygen demand between the two dietary groups were not sufficient to influence the eel heart's inherent tolerance of hypoxia (Davie et al., 1992). EFFECTS OF TISSUE F A T W ACID PHENOTYPE ON SWIMMING PERFORMANCE
Studies on the sturgeon revealed that tissue FA phenotype influenced sensitivity to reduced oxygen supply (hypoxia) with a phenotype rich in n-3 HUFA apparently more tolerant, both in terms of the whole animal (McKenzie et al. 1995a, 1997) and an isolated organ, the heart (Agnisola et al., 1996). Studies on the heart also revealed that a phenotype rich in n-3 HUFA might have an increased capacity for myocardial work (Agnisolaet al., 1996). This stimulated our interest in the potential effects of n-3 HUFA on the capacity of fish to respond to increased oxygen demand and perform sustained exercise. Sustained aerobic exercise (swimming) by fish requires the integration of molecular, biochemical, and physiological systems working at different levels of organismal organization and hence is considered an integrated index of organismal performance (Randall and Brauner, 1991).The anticipation was that n-3 HUFA would prove beneficial. An exploratory study was therefore undertaken to establish the effects on Atlantic salmon (Salmo salar) growth, tissue FA phenotype, and consequent exercise performance, of diets containing different proportions of n-3 HUFA (Dosanjhet al., 1998;McKenzie et al., 1998).Menhaden oil (MO) was used as a source of marine lipids in mixtures with canola oil (CAO). CAO is derived from specific strains of oilseed rape (Brassica napus and B. campestris), and is rich in 18C unsaturates such as oleic (18:l n-9), linoleic and a-linolenic acids. The choice of canola oil was prompted by the need of the aquaculture community to find an alternative to fish oils as lipid sources in salmon feeds, since fish oil supplies were expected to diminish in future (Tacon, 1996; Pauly et al., 1998) and canola oil provides essential n-6 and n3 PUFA. Dietary proportions of MO and CAO in the four diets were as shown in Table 14.1.
Dietary fatty acids in fish
375
Table 14.1. Dietary menhaden oil (MO) versus canola oil (CAO) ratios used to investigate effects of FA phenotype on swimming performance in Atlantic salmon (Dosai~jhet al., 1998; McKenzie et al., 1998).
-
DIET
% MO versus CAO
100% MO 75% MO; 25% CAO 62% MO; 38% CAO
50% MO; 50°/o CAO
Although the salmon grew equally well on all four diets, they created four significantly different tissue FA phenotypes (Dosanjhet al., 1998).The differences in phenotype were not, however, a consequence of consistent differences in tissue n-3 HUFA composition but rather in tissue 18C unsaturate composition. A direct relationship was established between content of the 18C unsaturates in CAO and content of these FA in the salmon tissues, with phenotypes exhibiting a linear increase in oleic and linoleic acids from diets 1to 4 (Dosanjh et al., 1998). These differences in FA phenotype were linked to significant differences in exercise performance, a graded increase in the salmon's maximum as defined by Brett, 1964) from sustainable aerobic swimming speed (UCri, phenotypes 1to 4, and the phenotype created by the diet wherein 10O0/0of dietary lipids furnished as MO had a significantly lower Ucritthan the phenotype created by the diet in which lipids were furnished as 50% MO and 50% CAO (McKenzie et al., 1998).As a consequence, remarkable positive linear relationships were revealed between Ucritand muscle levels of oleic and linoleic acids derived from dietary CAO (Fig. 14.5).This study was the first to investigate the influence of a graded change in dietary and/or muscle fatty acid composition on a physiological variable such as Ucritand the linear nature of the relationship between the two is striking. The results also indicate quite clearly that FAs other than n-3 HUFA have the ability to influence the physiology of fish (McKenzie, 2001). There were no significant differences in the relationship between swimming speed and MO, among the groups, nor in the maximum oxygen consumption achieved (McKenzie et al., 1998).Thus, the difference in Ucri, could not be ascribed to differences in the maximum allocation of 0,to power swimming among the phenotypes (McKenzieet al., 1998).Therefore, phenotype 4 was either able to achieve a greater swimming speed for essentially the same energetic cost as phenotype 1, or the two had a greater capacity for anaerobic muscle work at the top end of their performance range. Because linoleic and oleic were the only FAs that showed a direct linear relationship between their content in muscle lipids and Ucnt(Fig. 14.5)the data strongly suggest that the metabolism of these 18 carbon unsaturates was responsible for the differences in exercise performance of the salmon.
376
Physiological and ecological adaptations t o feeding i n vertebrates
EFFECTS OF DIETARY F A l T Y ACIDS O N OSMOREGULATORY CAPACITY There are marked differences in the FAs typically found in freshwater versus marine food webs. Freshwater fish have tissue lipids that are relatively low in PUFA, whereas marine fish are rich in HUFA (Henderson and Tocher, 1987). Thus for euryhaline fish, particularly species whose life history comprises migrations between marine and freshwater habitats, there will be a transition in dietary FA intake and tissue FA phenotype as they move from one environment to the other. Migration of juvenile salmonids into sea water is preceded by profound changes in morphology, physiology, and behavior, under endocrine and neuroendocrine control, which are known collectively as smoltification (Hoar, 1976). There is evidence that this "preadaptation" to the marine environment includes modifications to the FA composition of salmonid tissues, with an increase in the content of n-3 HUFA (Sheridan et al., 1985;Li and Yamada, 1992). In the salmon farming industry, freshwater parr are typically fed diets in which the lipid is provided as a northern hemisphere oil, such that they consume a "marine" rather than a "freshwater" spectrum of FA (Bell et al., 1997). Wild Atlantic salmon smolts have more elevated tissue levels of AA and n-6 HUFA than their hatchery-raised conspecifics; it has been suggested that this is linked to improved smolting success in wild fish (Ackman and Takeuchi, 1986). These observations led Bell et al. (1997) and Tocher et al. (2000)to investigate whether there were changes in FA desaturation/elongation reactions during smolting, how these were affected by dietary FA intake, and how environmental and dietary modifications of FA metabolism and tissue phenotype might influence subsequent tolerance of seawater exposure. Bell et al. (1997)and Tocher et al. (2000) compared salmon fed a diet in which lipids were provided as fish oil (FO), rich in long-chain n-3 HUFA, against salmon fed a diet in which lipids were provided as vegetable oils (VO) containing the nutritionally essential C18 PUFA. In both studies, the onset of smoltification was associated with a significant increase in the activity of elongation/desaturation enzymes in the liver, but this increase was significantly less in the fish fed FO (Bell et al., 1997; Tocher et al., 2000). In all animals, smoltification was associated with a large increase in AA and DHA levels in the phosphoglycerides of the liver and gills but in the salmon fed VO, the increased fatty acyl desaturation and elongation led to a higher AA/ DHA ratio. When exposed to a seawater challenge (acute exposure to seawater for 24 h), the fish fed VO regulated plasma chloride better than those fed FO, indicating that AA plays an important role in hypoosmoregulation in salmonids (Bell et al., 1997; Tocher et al., 2000). These studies reveal that FA metabolism is important in determining 0smoregulatory ability in salmonids and that this can be significantly influenced by dietary FA intake. Tocher et al. (2000) concluded that provision of
Dietary fatty acids in fish
A
Dietary 18:2 17-6
-
7
1.4 0
0
a, V)
m .-
I 1
I I
3
6
9
12
18:2 n - 6 content (% total fatty acids)
A
I>
1 I
2.2
B Muscle 18:1 n-9
I I
I I
18:1 n-9 content (% total fatty acids)
Fig 14.5. Least squares linear regression analysis of the relationship between content of 18:Zn-6 (A) and 18:l n-9 (B), in muscle and dietary lipids, and maximum sustainable swimming speed (Ucri,),in Atlantic salmon (Salmo salar) fed one of four diets where the dietary lipid was composed of differing proportions of menhaden oil and canola oil. Open symbols, relation of mean (+ SE) muscle % fatty acid (FA) content versus mean (k SE) Ucrit; closed symbols, mean dietary content versus mean (+ SE) U',,. N = 9 for muscle FA analyses, between 6 and seven for Ucrit,3 for dietary FA analyses. The group with the highest dietary and muscle content of these two FAs had a significantly higher Ucr, than the group with the lowest levels. Modified from McKenzie et al. (1998).
dietary lipid as fish oil to smolts appears inappropriate because the consequent elevation of tissue n-3 HUFA levels inhibits desaturation and elongation of tissue PLTFA, notably of linoleic acid to AA, thereby having a negative impact on osmoregulatory ability and seawater tolerance. The role of dietary FA and tissue FA metabolism in osmoregulation by other euryhaline fish groups remains to be described.
378
Physiological and ecological adaptations t o feeding in vertebrates
EFFECTS OF DIETARY FATTY ACIDS O N ENVIRONMENTAL TOLERANCE OF FISH LARVAE The levels of essential HUFA in broodstock diets can have a significanteffect on fecundity, fertilization rates and egg quality in fish, with AA apparently having a role in fecundity and fertilizationbut DHA influencing subsequent egg quality (Izquierdo et al., 2001; Sargent et al., 1999). There is much evidence to indicate that, as may be the case for all vertebrates, provision of adequate DHA to fish larvae is essential for correct development of central nervous system function and vision (Sargent et al., 1999) and can, thereby, have an impact on larval survival (Shields et al., 1999). In the larvae of marine species such as seabass (Dicentrarchus labrax), herring (Clupea haringus), and halibut (Hippoglossushippoglossus),a diet relatively deficient in DHA (i.e.with a low DHA/EPA ratio) leads to poor accumulation of this FA in eye lipids, impaired retinal development, and impaired visual performance (Bell et al., 1995, 1996; Shields et al., 1999). In halibut, inadequate provisioi~of DHA also impairs metamorphosis (completeeye migration and dorsal pigmentation; Shields et al., 1999). In yellow tail (Seriolaquinqueradiata) inadequate DHA provision has negative effects on retinal and neural development that impair the development of effective schooling behavior (Masuda et al., 1998; Ishizaki et al., 2001). There is also recent evidence, however, to indicate that dietary FA intake and consequent tissue FA phenotype can influence resistance to environmental stressors in fish larvae and juveniles. Czesny et al. (1999)fed juvenile walleye (Stizostedion vitreum),a freshwater fish, with diets containing different EPA/DHA ratios and found that survival of a salinity challenge (acute transfer to water at a salinity of 25%0)correlated negatively with the EPA/ DHAratio. Tago et al. (1999)compared the stress tolerance of Japanese flounder (Paralichthys olivaceus) larvae fed on diets with differing quantities of DHA or EPA. These authors found that the larvae which accumulated the most tissue DHA from their diet exhibited the lowest mortality rates during exposure to progessive increases in water temperature or to progressive hypoxia. Subsequently,Logue et al. (2000)performed a similar experiment on Dover sole (Solea solea)juveniles raised from larvae fed with diets either n-3 HUFA enriched or n-3 HUFA deficient. They found that the diet deficient in n-3 HUFA increased mortality during acute exposure to high temperature, low temperature with low salinity, and also acute hypoxia. Koven et al. (2001)provided gilthead seabream (Sparusaurata) larvae with diets enriched in AA and DHA, with different AA/DHA ratios, and found that survival following handling stress (transfer between aquaria) was significantly improved in animals fed the diet with the highest AA/DHA ratio. It is not known whether the improved tolerance to hypoxia in Japanese flounder and Dover sole larvae fed DHA-rich diets (Tagoet al., 1999;Logue et al., 2000) is
Dietary fatty acids in fish
379
associated with reduced metabolic rate and therefore occurs by the same mechanism as the improved tolerance of hypoxia described by McKenzie et al. (1995a, 1997) and Agnisola et al. (1996)in Adriatic sturgeon. MECHANISMS UNDERLYING THE PHYSIOLOGICAL EFFECTS OF FATTY ACIDS
A vast biomedical literature is availableon the diverse effects of dietary fatty acids on such human afflictions as heart disease, respiratory diseases, cancer, rheumatoid arthritis,autoimmune diseases, alcoholism, blindness, schizophrenia, depression, malaria, and multiple sclerosis (Holman, 1998;Connor, 2000; Stillwell, 2000). The many effects of FA and n-3 HUFA on mammalian and human cardiovascular function have received a great deal of attention from the biomedical community but extensive research efforts notwithstanding, the complexity of vertebrate fatty acid metabolism is such that underlying mechanisms have yet to be understood fully (Fitzgerald, 1996; Harris, 1996;Pepe and MacLennan, 2002). It is only possible to speculate about the mechanisms responsible for the effects of fatty acids on the physiology of fish, although this is an absolutely essential area for future research. Given the diversity of effects attributed to FAs in the biomedical literature, such speculation could be taken in many directions. Hence what follows is not an exhaustive analysis of potential mechanisms based on mammalian literature, but simply a few salient observations related to the phenomena described in fish. The positive linear relationship between the UCr,of salmon and their muscle content of oleic and linoleic acid may be a consequence of the use of these FAs as fuels for aerobic muscular work. Aerobic exercise is primarily fueled by FA oxidation in fish (Hochachkaand Somero, 1984)and there is invitro evidence indicating that mitochondria from the liver, heart, and red muscle exhibit differences in their ability to oxidize various FA, with oleic and linoleic acids being preferred substrates (Henderson and Sargent, 1985; Sidell and Driedzic, 1985; Egginton, 1986,1996; Sidell et al., 1995). Differences in the availability of these FA in the lipid stores of salmon red muscle (and heart) might translate into differences in the efficiency of aerobic ATP production by mitochondria during exercise,with an increased efficiency of mitochondria1 FA oxidation in exercising myocytes that contain more elevated levels of oleic and linoleic acid. This might in turn allow the salmon to use aerobic red muscle more efficiently and thereby postpone recruitment of white muscle and achieve greater swimming speeds before exhaustion (McKenzie et al., 1998; McKenzie, 2001). It might seem plausible that substrate preferences for FA oxidation also caused differences in metabolic rate observed in sturgeon and eel fed n-3 HUFA versus SFA. It seems unlikely,
380
Physiological and ecological adaptations t o feeding in vertebrates
however, that the reduced metabolic rate of fish fed n-3 HUFA was a consequence of impaired use of these FAs as fuels, because this should have been associated with reduced growth (McKenzie,2001). Modulation of membrane properties through changes in the FA composition of phosphoglycerides may well be responsible for many of the observed effects of dietary FAs on fish physiology, because such modulation may influence lipid-protein interactions, membrane protein function, and membrane-dependent signaling and metabolism (Pepe and MacLennan, 2002). As a consequence, it can modify cellular processes such as ion flux, respiratory electron transport, carrier-mediated transport, membrane-bound enzyme activity, receptor function, intracellular-lipid-based second messengers, and eicosanoid synthesis (Hulbert and Else 1999, Pepe and MacLennan, 2002). Pepe et al. (1999) and Pepe and MacLennan (2002) suggested that the reduced metabolic rate and improved tolerance of ischemia in hearts isolated from rats fed a diet rich in n-3 HUFA versus SFA were due to changes in the FA composition of mitochondrial membranes, with consequent changes in mitochondrial function. Pepe et al. (1999) found that mitochondria with high n-3 HUFA content had low oxygen consumption, particularly during uncoupled respiration, and suggested that this was because n-3 HUFA in the inner mitochondrialmembrane contributed to thermodynamic efficiency by reducing proton leak. The same authors also found that mitochondria rich in n-3 HUFA had increased levels of cardiolipin, a phosphoglyceride containingDHA and found exclusively in mitochondrial membranes, where it plays a fundamental role in creating the correct degree of fluidity for efficient function of the electron transport chain. Pepe and MacLennan (2002) also found that regulation of mitochondrial Ca2+levels after ischemia was improved in hearts rich in n-3HUFA, presumably as a consequence of more efficient control of ion movements. Similar effects of n-3 HLJFA on mitochondrial membrane composition and consequent function in fish may translate across levels of organismal organization and lead to reductions in wholeanimal metabolic rate as observed in sturgeon and eel (McKenzieet al., 1995a, 1997,2000),and also to improved tolerance of hypoxia in sturgeon -both the whole animal (McKenzie et al., 1995a, 1997) and the isolated heart (Agnisola et al., 1996). It is important to investigate to what extent the individual n-3 HUFA, DHA versus EPA, are responsible for the observed physiological phenomena. In terms of modulation of membrane properties, it is DHA that has the most pervasive effects (Brown, 1994; Salem et al., 2001; Stillwell, 2000). The n-6 HUFA AA is also an important component of reactive membranes and Tocher et al. (2000)suggested that increased gill phosphoglycerideAA content in salmon may increase their hypoosmoregulatoryability and seawater tolerance through effects on the activity of membrane-bound ion-transporting enzymes (Spector and Yorek, 1985;Gerbi et al., 1994). Arachidonic acid is also a precursor for the formation of eicosanoids, a cascade of paracrine
Dietary fatty acids in fish
381
hormones formed by the enzymes cyclooxygenase and lipoxygenase,which oxidize particular C20 HUFA following their cleavage from membrane phosphoglycerides through the action of phospholipase A, (Lands, 1991). Eicosanoids have a wide array of biological effects, activating many metabolic processes within the cell (Lands, 1991).Tocher et al. (2000) suggested that the effects of AA on seawater tolerance in salmon may also be linked to the production of eicosanoids, which are known to influence gill ion transport in fish (van Praag et al., 1987). Eicosapentenoic acid is also a precursor for the formation of eicosanoids and exhibits competitive inhibition with AA for the enzymes cyclooxygenase and lipoxygenase (Sargent et al., 1999).The eicosanoids formed from EPA are less reactive than those from AA, a fact linked in mammalian literature to the protective effects of n-3 HUFA against a number of inflammatory responses (Lands, 1991; Fitzgerald, 1996). It is conceivable that differencesin constitutiveeicosanoid formationmight lead to different degrees of metabolic "activation" in the tissues and organs of fish fed n-3 HUFAversus SFA and thereby contribute to differencesin wholeanimal metabolic rate (McKenzie,2001). Indeed, it is probable that at least some physiological phenomena are the result of a number of effects occurring simultaneously.
ECOLOGICAL :tMPLICATIONS OF DIETARY FATTY ACID AVAILABILITY FOR NATURAL FISH POPULATIONS
As stated earlier, all the above studies were performed to improve the culture conditions and productivity of farmed fish. It is worthwhle, however, considering how differences in dietary FA availability,and consequent tissue FA phenotype, might influence the ecological performance of wild fish populations. The ecologcal implications of feeding have generally been considered solely in bioenergetic terms, diet quantity, whereby energy acquisition will influence fitness by determining the proportion of energy that can be retained for somatic growth and reproduction rather than consumed to meet the costs of living. There are two immediately obvious ways in which diet quality, in terms of FA availability and intake, might influence the ecology of fish populations. In marine fish, the essential HUFA (EPA, DHA, and AA) are obtained exclusively through the diet, i.e. across trophic levels from primary producers (Sargent and Whittle, 1981; Sargent et al., 1999). Juvenile fish feed near the base of the food web and experimental trophic chains have established that their essential HUFA composition reflects very closely that of primary producers and can vary over a matter of days (St. John and Lund, 1996; St. John et al., 2001). Food webs based on dinoflagellates, which are rich in DHA (Volkmann et al., 1989), create fish rich in DHA; food webs based on diatoms, rich in EPA (Volkmann et al., 1989),create fish rich in EPA (St. John
382
Physiological and ecological adaptations t o feeding in vertebrates
and Lund, 1996; St. John et al., 2001). Many coastal areas are nurseries for juvenile fish but also experience seasonal algal blooms that cause profound variations in the temporal and spatial availability of FA at the base of the food web (Galois et al., 1996).Links between foodweb FA availability and FA composition of wild populations of juvenile fish are not known but could have important implications for larval development and tolerance of stress, in exactly the same manner as observed in laboratory studies (Bell et al., 1995; Masuda et al., 1998; Czesny et al., 1999;Shields et al., 1999;Tago et al., 1999;Logue et al., 2000; Koven et al., 2001; Ishizaki et al., 2001). Availability of FA in the food web may also influence the physiological performance and environmental tolerance of postlarval fish. Large adult fish respond to changes in the FA composition of their food web within three weeks (Kirsch et al., 1998).Within wild fish populations there are wider diversities in tissue FA phenotype, including EPA, DHA, and oleic acid (Jangaard et al., 1967; Hskanson, 1989; Klungs~yret al., 1989; Kuusipalo and Kakela, 2000; Joernsen and Grahl-Neilsen, 2001; Budge et al., 2002; Iverson et al., 2002), than those associated with the differencesin metabolic rate, hypoxia tolerance, and exercise performance described by McKenzie et al. (1995a, 1997, 1998). These specific physiological traits have all been proposed to contribute to fitnessby determining the ecological niches that a fish can occupy successfully (Prosser, 1950). Standard metabolic rate is a fundamental determinant of a fish's energetic relationship with its environment, defining its minimum energetic requirements for maintenance, and appears to have a direct bearing on fish life-historystrategies (Metcalfeand Monaghan, 2001; Wikelski and Ricklefs, 2001). Dissolved oxygen is an abiotic factor that has played an important role in vertebrate evolution, tolerance of hypoxia has determined the ability of fish to colonize particular habitats and their competitive abilities within them (Randall et al., 1981). Hypoxia is now a growing problem in coastal ecosystems worldwide, affecting thousands of square kilometers in the oceans surrounding Europe, North and South America, Africa, India, Southeast Asia, China, and Japan (Diaz and Rosenberg, 1995).The capacity for sustained aerobic exercise has long been considered influential in the ability of a fish to meet its "ecological aspirations" within a given environment (Brett, 1964; Randall and Brauner, 1991). Its capacity for osmoregulation in the face of salinity changes can likewise significantlyaffect fish survival (Brauner et al., 1994). A further implication in the reports that natural fish populations exhibit a large diversity in their tissue FA phenotypes must be recognized (Jangaard et al., 1967;Hiikanson, 1989; Klungs~yret al., 1989; Kuusipalo and Kakela, 2000; Joernsen and Grahl-Neilsen, 2001; Budge et al., 2002; Iverson et al., 2002). If such diversity in phenotype is indeed associated with a diversity in physiological traits similar to that observed in laboratory studies (Bell et al., 1997;McKenzie et al., 1995a, 1997,1998;Tocher et al., 2000), this implies that
Dietary fatty acids in fish
383
dietary FA intake can make a significant contribution to adaptive phenotypic plasticity within fish populations. It has long been assumed that natural selection would act upon adaptive phenotypic plasticity, i.e. upon diversity in traits of physiological adaptation to the enviroment, to elicit evolution within a species (Prosser, 1950; Spicer and Gaston, 1999). Effects of diet upon phenotypic plasticity would decouple genetic variability from phenotypic variability and thereby modulate the strength, direction, and genetic consequencesof natural selection. The extent to which natural variations in dietary FA availability and fish FA phenotype contribute to adaptive phenotypic plasticity and ecological performance are interesting areas for future study. Marine food webs are the most important global source of n-3 HUFA but marine fish stocks are declining worldwide; so harvesting at many levels of the food web, literally "fishing down" the web, has come into practice (Pauly et al., 1998). There is thus good reason to investigate how the transfer of valuable n-3 HUFA through marine food webs influences the phenotype and fitness of animals at different trophic levels. REFERENCES Ackman R.G. and Takeuchi T. 1986. Comparison of fatty acids and lipids of smolting hatchery-fed and wild Atlantic salmon Salrno salar. Lipids 21: 117-120. Agnisola C., McKenzie D.J., Taylor E.W., Bolis C.L., and Tota B. 1996. Cardiac performance in relation to oxygen supply varies with dietary lipid composition in sturgeon. Arner. 1. Physiol. 271: R417-425. Arts M.T., Ackman R.G. and Holub B.J. 2001. "Essential fatty acids" in aquatic ecosystems: a crucial link between diet and human health and evolution. Can. 1. Fish. Aquat. Sci. 58: 122-137. Bang H.O. and Dyerberg J. 1985. Fish consumption and mortality from coronary heart disease. N e w Eng. 1. Med. 313: 822-823. Bell J.G., McVicar A.H., Park M. and Sargent J.R. 1991. High dietary linoleic acid affects fatty acid compositions of individual phospholipids from tissues of Atlantic salmon (Salrno salar): association with stress susceptibility and cardiac lesion. 1. Nutr. 121: 1163-1172. Bell J.G., Dick J.R., McVicar A.H., Sargent J.R. and Thomson K.D. 1993. Dietary sunflower, linseed and fish oils affect phospholipid fatty acid composition, development of cardiac lesions, phospholipase activity and eicosanoid production in Atlantic salmon (Salrno salar). Prostaglandins, Leukotrienes Essent. Fatty Acids 49: 665-673. Bell J.G., Tocher D.R., Farndale B.M., Cox D.I., McKinney R.W. and Sargent J.R. 1997. The effect of dietary lipid on polyunsaturated fatty acid metabolism in Atlantic salmon undergoing parr-smolt transformation. Lipids 32: 515-525. Bell M.V., Dick J.R., Thrush M. and Navarro J.C. 1996. Decreased 20:4n-6/20:5n-3 ratio in sperm from cultured sea bass, Dicentrarchus labrax, broodstock compared to wild fish. Aquaculture 144: 189-199. Bell M.V., Batty R.S., Dick J.R., Fretwell K., Navarro J.C. and Sargent J.R. 1995. Dietary deficiency of docosahexaenoic acid impairs vision at low light intensities in juvenile herring (Clupea harengus, L.). Lipids 30: 443-449. Brauner C.J., Iwama G.K. and Randall D.J. 1994. The effect of short-duration seawater exposure on the swimming performance of wild and hatchery-reared juvenile coho salmon (Oncorhynchus kisutch) during smoltification. Can. 1. Fish. Aquat. Sci. 51: 2188-2194.
384
Physiological and ecological adaptations t o feeding in vertebrates
Brett J.R. 1964. The respiratory metabolism and swimming performance of young sockeye salmon. J. Fish. Res. Board. Can. 21: 1183-1226. British Nutrition Task Force. 1992. Unsaturated Fatty Acids. Nutritional and Physiological Signvicance. Chapman and Hall, London, UK. Broadhurst C.L., Cunnane S.C. and Crawford M.A. 1998. Rift Valley lake fish and shellfish provided brain-specific nutrition for early Homo species. Brit. J. Nutr. 79: 3-21. Broadhurst C.L., Wang Y., Crawford M.A., Cunnane S.C., Parkington J.E. and Schmidt W.F. 2002. Brain-specific lipids from marine, lacustrine, or terrestrial food resources: potential impact on early African Homo species. Comp. Biochern. Physiol. 131B: 653-673. Brown M.F. 1994. Modulation of rhodopsin function by properties of the membrane bilayer. Chem. Phys. Lipids 73: 159-180. Budge S.M., Iverson S.J., Bowen W.D. and Ackman R.G. 2002. Among- and withinspecies variability in fatty acid signatures of marine fish and invertebrates on the Scotian Shelf, Georges Bank and southern Gulf of St. Lawrence. Can. J. Fish. Aquat. Sci. 59: 886-898. Burleson M.L., Smatresk N.J. and Milsom W.K. 1992. Afferent inputs associated with cardioventilatory control in fish. In: Fish Physiology, The Cardiovascular System. W.S. Hoar, D.J. Randall, and A.P. Farrell (eds.). Acad. Press, New York, NY, pp. 390426. Charnock J.S., McLennan P.L. and Abeywardena M.Y. 1992. Dietary modulation of lipid metabolism and mechanical performance of the heart. Mol. Cell. Biochern. 116: 19-25. Connor W.E. 2000. Importance of n-3 fatty acids in health and disease. Arner. J. Clin. Nutr. 71: 171s-175s. Crawford M. and Marsh D. 1989. The Driving Force: Food, Evolution, and the Future. Harper and Row, New York, NY. Crawford M.A., Bloom M., Broadhurst C.L., Schmidt W.F., et al. 1999. Evidence for the unique function of docosahexaenoic acid during the evolution of the modem hominid brain. Lipids 34: S39-S47. Czesny S., Kolkovski. S., Dabrowski K. and Culver D. 1999. Growth, survival and quality of juvenile walleye Stizostedion vitreurn as influenced by n-3 HUFA enriched Arternia naupli. Aquaculture 178: 103-115. Davie P.S., Farrell A.P. and Franklin C.F. 1992. Cardiac performance of an isolated eel heart: Effects of hypoxia and responses to coronary artery perfusion. J. Exp. Zool. 262: 113-121. Diaz J.R. and Rosenberg R. 1995. Marine benthic hypoxia: a review. Its ecological effects and the behavioural responses of benthic macrofauna. Oceanogr. Mar. Biol. Ann. Rev. 33: 245-303. Dolecek T.A. 1992. Epidemiological evidence of relationships between dietary polyunsaturated fatty acids and mortality in the multiple risk factor intervention trial. Proc. Soc. Exp. Biol. Med. 200: 177-182. Dosanjh B., Higgs D.A., McKenzie D.J., Deacon G. and Randall D.J. 1998. Influence of blends of menhaden oil and canola oil on the performance, muscle lipid composition and thyroidal status of Atlantic salmon (Salrno salar) in seawater. Fish Physiol. Biochern. 19: 123-134. Egginton S. 1986. Metamorphosis of the American eel Anguilla rostrata Le Seur: 1. Changes in metabolism of skeletal muscle. J. Exp. Zool. 237: 173-184. Egginton S. 1996. Effect of temperature on optimal substrate for B-oxidation. J. Fish Biol. 49: 753-758. Ellis E.F., Police R.J., Dodson L.Y. McKinney J.S. and Holt S.A. 1992. Effect of dietary n-3 fatty acids on cerebral microcirculation. Arner. J. Physiol. 262: H1379-1386. Fitzgerald G.A. 1996. Omega-3 fatty acids and vascular function. Omega3 News 9: 1-3. Florant G.L. 1998. Lipid metabolism in hibernators: The importance of essential fatty acids. Arner. Zool. 38: 331-340. Ford D.A. 2002. Alterations in myocardial lipid metabolism during myocardial ischemia and reperfusion. Prog. Lipid. Res. 41: 6-26.
Dietary fatty acids in fish
385
Galois R., Richard P. and Fricourt B. 1996. Seasonal variations in suspended particulate matter in the Marennes-Oleron Bay, France, using lipids as biomarkers. Estuar. Coast Shelf Sci. 43: 335-357. Gerbi A., Zerouga M., Debray M., Durand G., Chanez C. and Bourre J.-M. 1994. Effect of fish oil diet on fatty acid composition of phospholipids of brain membranes and on kinetic properties of Na+,K+-ATPaseisoenzymes of weaned and adult rats. J. Neurochem. 62: 1560-1569. Hiikanson J.L. 1989. Conditon of larval anchovy (Engraulis mordax) in the Southern California Bight, as measured through lipid analysis. Mar. Biol. 102: 153-159. Harris W.S. 1996. n-3 fatty acids and lipoproteins: comparison of results from human and animal studies. Lipids 31: 243-252. Henderson R.J. and Sargent J.R. 1985. Chain-length specificities of mitochondria1 and peroxisomal B-oxidation of fatty acids in livers of rainbow trout (Salmo gairdneri). Comp. Biochem. Physiol. 82B: 79-85. Henderson R.J. and Tocher D.R. 1987. The lipid composition and biochemistry of freshwater fish. Prog. Lipid Res. 26: 281-347. Hoar W.S. 1976. Smolt transformation: Evolution, behaviour and physiology. J. Fish. Res. Board Can. 33: 1234-1252. Hochachka P.W. and Somero G.N. 1984. Biochemical Adaptation. Princeton Univ. Press, Princeton, NJ. Hock C.E., Beck L.D., Bodine R.C. and Reibel D.K. 1990. Influence of dietary fatty acids on myocardial ischemia and reperfusion. Amer. 1. Physiol. 259: H1518-1526. Holman R.T. 1998. The slow discovery of the importance of 0 3 essential fatty acids in human health. J. N u f r . 128: 4275-433s. Hornstra G., 1989. The significance of fish and fish-oil enriched food for prevention and therapy of ischaemic heart disease. In: T h e Role of Fats i n H u m a n Nutrition. A.J. Vergroesen and M. Crawford (eds.). Acad. Press, London, UK, pp. 151-236. Hulbert A.J. and Else P.L. 1999. Membranes as possible pacemakers of metabolism. J. Theor. Biol. 199: 257-274. Ishizaki Y., Masuda R., Uematsu K., Shimizu K., Arimoto M. and Takeuchi T. 2001. The effect of dietary docosahexaenoic acid on schooling behaviour and brain development in larval yellowtail. J. Fish Biol. 58: 1691-1703. Iverson S.J., Frost K.J. and Lang S.L.C. 2002. Fat content and fatty acid composition of forage fish and invertebrates in Prince William Sound, Alaska: factors contributing to among- and within species variability. Mar. Ecol. Prog. Series 241: 161-181. Izquierdo M.S., Fernandez-Palacios H. and Tacon A.G.J. 2001. Effect of broodstock nutrition on reproductive performance of fish. Aquaculture 197: 2542. Jangaard P.M., Ackman R.G. and Sipos J.C. 1967. Seasonal changes in fatty acid composition of cod liver, flesh, roe and milt lipids. J. Fish. Res. Bd. Canada 24: 613-627. Joensen H. and Grahl-Neilsen 0. 2001. The redfish species Sebastes viviparus, Sebastes marinus and Sebastes mentella have different compositions of their tissue fatty acids. Comp. Biochem. Physiol. 129B: 73-85. Kirsch P.E., Iverson S.J., Bowen W.D., Kerr S.R. and Ackman R.G. 1998. Dietary effects on the fatty acid signature of whole Atlantic cod (Gadus morhua). Can. J. Fish. Aquat. Sci. 55: 1378-1386. Klungsnryr J., Tilseth S., Wilhelmsen S., Falk-Petersen S. and Sargent J.R. 1989. Fatty acid composition as an indicator of food intake on cod larvae Gadus morhua from Lofoten, Northern Norway. Mar. Biol. 102: 183-188. Koven W., Barr Y., Lutzky S., Ben-Atia I., 2001. The effect of dietary arachidonic acid (20:4 n-6) on growth, survival and resistance to handling stress in gilthead seabream (Sparus aurata) larvae. Aquaculture 193: 107-122. Kuusipalo L. and Kakela R. 2000. Muscle fatty acids as indicators of niche and habitat in Malawian cichlids. Limnol. Oceanogr. 45: 996-1000. Lands W.E.M. 1991. Biosynthesis of prostaglandins. A n n . Rev. N u t r . 11: 41-60.
386
Physiological a n d ecological adaptations t o feeding i n vertebrates
Li H.-0. and Yamada J. 1992. Changes of the fatty acid composition in smolts of masu salmon (Oncorhynchus masou) associated with desmoltification and seawater transfer. Comp. Biochem. Physiol. 103A: 221-226. Logue J.A., Howell B.R., Bell J.G. and Cossins A.R. 2000. Dietary n-3 long-chain polyunsaturated fatty acid deprivation, tissue lipid composition, ex-vivo prostaglandin production, and stress tolerance in juvenile Dover sole (Solea solea L.). Lipids 35: 745-755. Masuda R., Takeuchi T. and Tsukamoto K. 1998. Critical involvement of dietary docosahexaenoic acid in the ontogeny of schooling behaviour in the yellowtail. J. Fish Biol. 53: 471-484. McKenzie D.J. 2001. Effects of dietary fatty acids on the respiratory and cardiovascular physiology of fish. Comp. Biochem. Physiol. 128A: 607-621. McKenzie D.J., Taylor E.W., Bronzi P. and Bolis L. 1995b. Aspects of cardioventilatory control in the Adriatic sturgeon (Acipenser naccarii). Respir. Physiol. 100: 44-52. McKenzie D.J., Taylor E.W., Dalla Valle A.Z. and Steffensen J.F. 2002. Tolerance of acute hypercapnic acidosis by the European eel (Anguilla anguilla). J. Comp. Physiol. 172B: 347-354. McKenzie D.J., Higgs D.A., Dosanjh B., Deacon G. and Randall D.J. 1998. Dietary lipid composition influences swimming performance in Atlantic salmon (Salmo salar) in seawater. Fish Physiol. Biochem. 19: 111-122. McKenzie D.J., Piccolella M., Dalla Valle A.Z., Taylor E.W. and Steffensen J.F. 2003. Tolerance of chronic hypercapnia by the European eel (Anguilla anguilla). J. Exp. Biol. 206: 1717-1726. McKenzie D.J., Piraccini G., Taylor E.W., Steffensen J.F., Bronzi P. and Bolis L. 1994. Effects of dietary lipids on responses to stress in fish. In: High Performance Fish D.D. MacKinlay (ed.). Amer. Fish. Soc., Vancouver, WA, pp. 4046. McKenzie D.J., Piraccini G., Steffensen J.F., Bolis C.L., Bronzi P. and Taylor E.W. 1995a. Effects of diet on spontaneous locomotor activity and oxygen consumption in the Adriatic sturgeon (Acipenser naccarii). Fish Physiol. Biochem. 14: 341-355. McKenzie D.J., Piraccini G., Piccolella M., Steffensen J.F., Bolis C.L. and Taylor E.W. 2000. Effects of dietary fatty acid composition on metabolic rate and responses to hypoxia in the European eel, Anguilla anguilla. Fish Physiol. Biochem. 22: 281-296. McKenzie D.J., Piraccini G., Papini N., and Galli C. 1997. Oxygen consumption and ventilatory reflex responses are influenced by dietary lipids in sturgeon. Fish Physiol. Biochem. 16: 365-379. Metcalfe N.B., and Monaghan, P. 2001. Compensation for a bad start: Grow now, pay later? Trends Ecol. Evol. 16: 254-260. Nilsson G.E., Rosen P. and Johansson D. 1993. Anoxic depression of spontaneous locomotor activity in crucian carp quantified by a computerised imaging technique. J. Exp. Biol. 108: 153-162. Paulson D.J., Smith J.M., Zhao J. and Bowman J. 1992. Effects of dietary fish oil on myocardial ischaemic/reperfusion injury of Wistar Kyoto and stroke-prone spontaneously hypertensive rats. Metabolism 41: 533-539. Pauly D., Christensen V. and Dalsgaard J. 1998. Fishing down marine food webs. Science 279: 860-863. Pepe S. and MacLennan P.L. 2002. Cardiac membrane fatty acid composition modulates myocardial oxygen consumption and postischemic recovery of contractile function. Circulation 105: 2303-2308. Pepe S., Tsuchiya N., Lakatta E.G. and Hansford R.G. 1999. PUFA and aging modulate cardiac mitochondria1 membrane lipid composition and Ca2+activation of PDH. Amer. J. Physiol. 276: H149-H158. Prosser C.L., (ed.). 1950. Comparative Animal Physiology. W.B. Saunder, Philadephia, PA (1" ed.). Randall D.J. and Brauner C.L. 1991. Effects of environmental factors on exercise in fish. J. Exp. Biol. 160: 113-126.
Dietary fatty acids i n fish
387
Randall D.J., Burggren W.W., Farrell A.P. and Haswell M.S. 1981. The Evolution of AirBreathing in Vertebrates. Cambridge Univ. Press, Cambridge, UK. Randall D.J., McKenzie D.J., Abrami G., and Bondiolotti G.P. 1992. Effects of diet on responses to hypoxia in the sturgeon (Acipenser naccarii). J. Exp. Biol. 170: 113-125. Rowley A.F., Knight J. and Lloyd-Evans P. 1995. Eicosanoids and their role in immune modulation in fish. Fish Shellfish lmmunol. 5: 549-567. Salem N., Litman B., Kim H.Y. and Gawrisch K. 2001. Mechanisms of action of docosahexaenoic acid in the nervous system. Lipids 36: 945-959. Sargent J.R. and Whittle K.J. 1981. Lipids and hydrocarbons in the marine food web. In: Analysis of Marine Ecosystems. A.R. Longhurst (ed.). Acad. Press, London, UK, pp. 491-533. Sargent J. Bell G., McEvoy L., Tocher D. and Estevez A. 1999. Recent developments in the essential fatty acid nutrition of fish. Aquaculture 177: 191-199. Schurmann H.H. and Steffensen J.F. 1994. Spontaneous swimming activity of Atlantic cod, Gadus morhua L., exposed to graded hypoxia at three temperatures. J. Exp. Biol. 197: 129-142. Sheridan M.A., Allen W.V. and Kerstetter T.H. 1985. Changes in the.fatty acid composition of steelhead trout, Salmo gairdneri Richardson, associated with parr-smolt transformation. Comp. Biochem. Physiol. 80B: 671-676. Shields R.J., Bell R.G., Luizi F.S., Gara B., Bromage N.R. and Sargent J.R. 1999. Natural copepods are superior to enriched Artemia nauplii as feed for halibut larvae (Hippoglossus hippoglossus) in terms of survival, pigmentation and retinal morphology: Relation to dietary essential fatty acids. I. Nutr. 129: 1186-1194. Sidell B.D. and Driedzic W.R. 1985. Relationship between cardiac energy metabolism and cardiac work demand in fishes. In: Circulation, Respiration and Metabolism. R. Gilles R (ed.). Springer-Verlag, Berlin, pp. 381401. Sidell B.D., Crockett E.L. and Driedzic W.R. 1995. Antartic fish tissues preferentially catabolize monenoic fatty acids. 1. Exp. Zool. 271: 73-81. Simandle E.T., Espinoza R.E., Nussear K.E. and Tracy C.R. 2001. Lizards, lipids and dietary links to animal function. Physiol. Biochem. Zool. 74: 625-640. Spector A.A. and Yorek M.A. 1985. Membrane lipid composition and cellular function. 1. Lipid Res. 26: 1015-1035. Spicer J.I. and Gaston K.J. 1999. Physiological Diversity and Its Ecological Implications. Blackwell Science, Oxford, UK. St. John M.A. and Lund T. 1996. Lipid biomarkers: linking the utilisation of frontal plankton biomass to enhanced condition of juvenile North Sea cod. Mar. Ecol. Prog. Series 131: 75-85. St. John M.A., Clemmesen C., Lund T. and Koster T. 2001. Diatom production in the marine environment: implications for fish growth and condition. lCES J. Mar. Sci. 58: 1106-1113. Steffensen J.F. 1989. Some errors in the respirometry of water breathers: how to avoid and correct for them. Fish Physiol. Biochem. 6: 49-59. Stillwell W. 2000. Docosahexaenoic acid and membrane lipid domains. Curr. Org. Chem. 4: 1169-1183. Tacon A.G.J., 1996. Feeding tomorrow's fish. World Aquacult. 27: 20-32. Tago A., Yamamoto Y., Teshima S. and Kanazawa A. 1999. Effects of 1,2-di-20:5phosphatidylcholine (PC) and 1,2-di-22:6-PC on growth and stress tolerance of Japanese flounder (Paralichthys olivaceus) larvae. Aquaculture 179: 231-239. Tocher D.R., Bell J.G., Dick J.R., Henderson R.J., 2000. Polyunsaturated fatty acid metabolism in Atlantic salmon (Salmo salar) undergoing parr-smolt transformation and the effects of dietary linseed and rapeseed oils. Fish Physiol. Biochem. 23: 59-73. Van Praag D., Farber S.J., Minkin E. and Primor N. 1987. Production of eicosanoids by the killifish gill and opercular epithelia and their effect on active transport of ions. Gen. Comp. Endocrin. 67: 50-57.
388
Physiological and ecological adaptations to feeding in vertebrates
van Waarde A*, van den Thillart G. and Kesbeke F. 1983. Anaerobic energy metabolism of the European eel, Anguilla anguilla L. J. Comp. Physiol. 149B: 469-475. Volkman J.K., Jeffrey S.W. and Nichols P.D. 1989. Fatty acid and lipid composition of 10 species of microalgae used in mariculture. J. Exp. Mar. Biol. Ecol. 128: 219-240. Wikelski M. and Ricklefs R.E. 2001. The physiology of life histories. Trends Ecol. Evol. 16: 479-481. Winberg S., Overli, 0. and Lepage, 0. 2001. Suppression of aggression in rainbow trout (Oncorhynchus mykiss) by dietary L-tryptophan. J. Exp. Biol. 204: 3867-3876.
Aspects of Protein and Amino Acid Digestion and Utilization by Marine Fish Larvae Ivar Rannestad' and Luis E.C. Concei@02 University Bergen, Department of Biology, Bergen, Norway Universidade Algarve, Centro De Ciencias do mar, Campus de Gembelos, Faro, Portugal
SYNOPSIS In larval marine fish, the sum of protein deposition, turnover, and catabolism necessary for rapid growth dictates a high amino acid (AA) requirement. Fish larvae seem to have control over AA catabolism comparable to that of juvenile fish and use dispensable AA preferentiallyto indispensableAA as energy substrates. In the few species analyzed, AA has been estimated to account for 60-95% of total substrate oxidation of larvae in the first weeks after onset of exogenous feeding. The alimentary canal is vital in ensuring a supply of dietary AA to the growing larval tissues. The majority of marine fish larvae, including those targeted for aquaculture, hatch from small, mostly pelagic eggs and their digestive tracts are still developing at the onset of exogenous feeding. A fully developed digestive system, including a functional (acid-producing) stomach, is present only later. Although the larial gut is not adult-like at the onset of exogenous feeding, it is capable of supporting high growth rates, provided suitable feed is available. Presently available data suggest that free amino acids (FAA) are rapidly absorbed from the gut into the larval tissues, with minimal evacuated losses to the water. There seems to be lower digestive absorption rates and assimilation of protein than FAA.
390
Physiological and ecological adaptations to feeding in vertebrates
This chapter deals with aspects of digestion and utilization of dietary nitrogen by marine pelagic fish larvae which have a true larval stage with special features that are later replaced or lost during metamorphosis. These species represent the majority of marine fish and contrast with the minority of marine fish and also many freshwater species, including salmonids, which hatch at a very advanced stage of development.In marine species the term "egg stage" is usually employed until hatching, "yolk-sac stage" after hatching, when there is still yolk remaining, "larval stage" until metamorphosis, and "juvenile stage" after metamorphosis, until the young fish have reached the reproductive "adult stage" (Kendallet al., 1984).Normally exogenous feeding begins shortly before the yolk reserves are exhausted; at this time feeding is not essential but the functional status of the larvae has developed sufficiently to permit exogenous feeding. For Atlantic halibut, a cold-water marine species currently targeted for aquaculture in northern Europe and northern America, claims of both early (190 degree day posthatching, "DPH) and late (260-290 "DPH) first-feeding have been put forward (Pittman et al., 1990; Lein and Holmefjord, 1992;Harboe, and Mangor-Jensen, 1998).Arguments in favor of early onset of exogenous feeding for this species are based on the characterization of morphological and histological development, including presence of zymogen granules in the pancreas (Luizi et al., 1999) and a functional jaw apparals (Kjorsvik and Reiersen, 1992).On the other hand, late onset of exogenous feeding (close to the end of yolk resorption, 260-290 "DPH),which is the common commercial rearing practice today, is based on empirical constraints in the rearing systems, including hygiene and more rapid initiation of prey ingestion (Harboe and Mangor-Jensen, 1998). The capacity for digestion and absorption of proteins and other complex nutrients in larval marine fishes is still a matter of discussion. Some authors claim the larval digestive tract is capable of complete degradation of food items, since analyses of larvae have shown high activity levels of certain digestive enzymes (Munilla-Moran and Stark, 1989; Watanabe and Kiron, 1994; Oozeki and Bailey, 1995; Gawlicka et al., 2000), including the key enzyme trypsin (Pedersen et al., 1987,1990; Ueberschar, 1988; Hjelmeland, 1995; Kurokawa and Suzuki, 1996).Others have underlined and discussed the contribution of the prey's own enzymes and its autolysis during the passage through the gut lumen as possible factors in triggering the larval digestive process (Walford and Lam, 1993; Watanabe and Kiron, 1994; Kurokawa et al., 1998).This chapter intends to clarify the in-vivo performance of the digestive system in marine fish larvae, in particular that which concerns proteins and amino acids. The data are related to available information on the amino acids requirements and metabolism in the early life stages of marine fish.
Digestion in marine fish larves
DIGESTIVE FUNCTION: STATUS AND CHANGE I N CAPACITY W I T H ONTOGENY
The sum of protein deposition, turnover, and catabolism, together with the rapid growth rate of fish larvae, dictate a high AA requirement. When the yolk source of AA has been exhausted, the digestive tract becomes critical in ensuring a steady supply of dietary AA to the metabolic pathways of the growing larval tissues. Before the ingested food is available for anabolic and catabolic processes, it must be digested and absorbed from the gut. Digestive capacity seems to increase with larval age (Rust, 1995; Day et al., 1997). This is probably associated with the lower capacity of the digestive tract of most fish larvae to hydrolyze food at first feeding (Govoniet al., 1986).There is also variation in the speed of development of the digestive tract between species. In the African catfish, the digestive tract seems to be functionally complete at about 200 degree days after fertilization and 24 mg wet weight (Verreth et al., 1992).In turbot, a functional stomach only appears at the end of metamorphosis (Segner et al., 1994),which normally occurs around 30 days after fertilization and 180mg wet weight. Furthermore, in fish larvae, digestive efficiency seems to decrease at high food intakes due to higher passage rates (Govoni et al., 1986; Day et al., 1997).The importance of factors such as type of food and temperature on digestibility remains to be established in fish larvae. The Stomach An important question is to what extent the lack of a stomach in first-feeding larvae affects overall digestive capacity and imposes constraints on amino acid absorption and, in a more practical perspective, what are the resultant implications for first-feed formulation in aquaculture. One of the primary roles of the stomach is the breakdown of large food particles, allowing regulated release of partly digested food (chyme) to the midgut, thereby maintaining an optimal proportion of food to digestive enzymes in the midgut (Joblingand Hjelmeland, 1992).The stomach thus acts as a short-term food store. The presumptive stomach of fish larvae, before the onset of gastric secretion, can serve to some extent as a storage compartment, with functional esophagal and pyloric valves (Rsnnestad et al., 2000~).At the onset of exogenous feeding, there is no pyloric valve. However, there are functional valves in the esophagus and two located caudally in the digestive tract: the first is located between the midgut and hindgut, and the second closes the anus. Visual observations on early feeding larvae (Y. Barr, Akvaforsk, Norway, pers.comrn.) indicate that food is stored in the midgut, where it is physically processed and digested by pancreatic enzymes, bile, and smooth muscular contractions. Thus the valves between the midgut and the hindgut may regulate the amount of food transferred to the hindgut for final digestion and absorption. Such control would be analogous to the manner in which the
392
Physiological and ecological adaptations t o feeding in vertebrates
pyloric valve regulates the release of chyme into the midgut once the stomach has developed. Once in the hindgut, food particles, including proteins, are thought to be taken up by endocytosis, followed by intracellular digestion. However, the quantitative importance of hindgut processing is not clear (see below) and the contribution of each gut-segment in digesting and absorbing food has yet to be clarified. The ontogeny of gastric pepsin secretion and HC1 production during the early stages of development in most marine fish larvae has been established by histological observations and pH measurements, and recently also by reverse transcription-polymerasechain reaction (RT-PCR).All the histological data collected to date demonstratethat most marine fish larvae do not possess HC1-producingcells (gastric glandular acini) at the first feeding stage (Tanaka, 1973; Tanaka et al., 1996). pH has been measured by many techniques, including microelectrodes and indicator strips using dissected gut sections (Mahr et al., 1983;Buddington, 1985;Walford and Lam, 1993;Hoehne-Reitan et al., 2001), and immersion or tube-feeding of pH indicators in vivo (Verreth et al., 1992; Bengtson et al., 1993; Rust et al., 1993; Rust 1995; Rannestad et al., 2000~). None of these studies could demonstrate acid pH at first feeding. This is also supported by an RT-PCR study of Winter flounder, which demonstrated that the expression of proton pumping genes occurred at 20 days post hatch and coincided with the appearance of gastric glands (Douglas et al., 1999).A gradual fall in lumenal pH, starting at the onset of metamorphosis, suggests a low but increasing capacity for HC1 production and secretion into the stomach lumen. At the end of metamorphosis, the pH in the stomach of Asian seabass, Japanese flounder, and turbot was lower than 4 (Walford and Lam, 1993; Rannestad et al., 2000c; Hoehne-Reitan et al., 2001). Gastric pepsin-like activity seems to be very low at hatching (Cahu and Zambonino Infante, 2001). In developing Japanese flounder, it increased slowly from 10 units (assessed as pg product produced 30-' min-' ind") in early larvae to about 30 units at the climax of metamorphosis, to more than 150 units in recently settled larvae, and in excess of 300 units in completely metamorphosed fish (Alvarez et al., 1999).In Winter flounder, sequential expressions of two forms of pepsinogens have been shown (Douglas et al., 1999). Pepsinogen IIa expression occurred one week before the onset of metamorphosis, while the expression of pepsinogen IIb coincided with proton pump, gastric glands, and measured pepsin activity (Douglas et al., 1999). Using gastric-specific RNA probes in the same species, in-situ hybridization demonstrated that gastric gland cells contain transcripts (mRNA) of pepsinogen IIa and IIb as well as proton pump genes (Gawlickaet al., 2001). This suggests synchronized activation of the mechanisms responsible for the production and secretion of pepsin and HC1 in gland cells, which would ensure HC1-mediated pepsin activation from its zymogen form, as well as
Digestion i n marine fish larves
393
establishing the optimum pH range for pepsin activity. However, the physiological role of an earlier expression of pepsinogen IIa (Douglaset al., 1999)remains to be clarified. It is also interesting to note that Gawlicka et al. (2001) were able to demonstrate the presence of mRNA for alpha-proton pump in mucous cells, suggesting that cells other than gastric glands are capable of producing HC1. Due to the absence of HC1- and pepsin-secreting cells in fish larvae at the onset of exogenous feeding, there is no preparatory acid denaturation of ingested proteins. Cahu and Zambonino Infante (2001) suggested that the lack of a stomach does not hamper enzymaticprotein digestion in fish larvae, since this is ensured by the pancreatic and intestinal enzymes. However, native proteins present a smaller surface area than denatured proteins for alkaline enzymatic attack. In the stomachless Carassius auratus gibelo, Jany (1976) found that the hydrolysis of intact protein was about 120h that of denatured protein. Studies comparing the digestion of native and denaturated proteins in fish larvae are still needed. However, preliminary studies on Atlantic halibut (Tonheim et al., 2004) as well as more elaborate studies on zebra fish, striped bass, and walleye (Rust, 1995) suggest lower rates of digestibility and assimilation of protein in pregastric larvae than in postmetamorphic fish.
The Midgut The rnidgut appears to be alkaline throughout development (Walford and Lam, 1993; Rsmestad et al., 2000~).This alkaline condition has two likely sources in pregastric marine fish larvae: it may be due to drinking sea water (pH about 8.2),which occurs from hatching (Mangor-Jensenand Adoff, 1987; Tytler et al., 1993),to pancreatic or bile secretions of alkaline fluid (i.e.rich in HCO;) into the gut lumen area, or to a combination of both. At present, no firm conclusions can be drawn regarding the stage at which the alkaline secretion starts, since pH values in the presumptive stomach, midgut, and sea water are all within similar alkaline range from first feeding. Immunohistochemical analysis for trypsin-like enzymes has demonstrated secretion into the gut lumen from first feeding (Kurokawaand Suzuki, 1996) and it is likely that the production and secretion of pancreatic HCO, are related to this activity. It is clear that in postgastric marine fish larvae, the high pH of the midgut is closely regulated. Rsnnestad et al. (2000~) demonstrated an immediate rise in pH as indicator solution passed from the stomach to the midgut of Japaneseflounder. Levels of proteolytic pancreatic enzymes were low when marine fish larvae examined to date commenced exogenous feeding but rose as metamorphosis approached (Hjelmeland, 1995; Gawlicka et al., 2000; Cahu and Zambonino Infante, 2001). This is in contrast to the initially high cytosolic peptidase activity which decreased as the larvae developed, concurrent with rising levels of brush-border enzymes. These topics were
394
Physiological and ecological adaptations t o feeding in vertebrates
recently reviewed by Cahu and Zambonino Infante (2001) and Kolkovski (2001). The Hindgut Fish larvae have been seen to absorb large molecules, including proteins, by endocytosis and intracellular digestion in the mucosal cells of the hindgut (Iwai and Tanaka, 1968; Watanabe, 1984; Kurokawa et al., 1996; Luizi et al., 1999). On the basis of these observations it has been suggested that low levels of intestinal proteolytic activity may be compensated for by hindgut protein endocytosis. However, in smelt it took 10 to 24 hrs for the protein taken up by hindgut pinocytosis to be hydrolyzed intracellularly (Watanabe, 1984).Such processing rates are too low to satisfy the high metabolic and anabolic demands of rapidly growing fish larvae. McLean et al. (1999)suggested that a primary role of macromolecule absorption inthe fish hindgut may be antigen sampling, as found in mammals. It is clear that more studies are required in this area.
AVAILABILITY OF AMINO ACIDS AND PROTEIN I N LIVE PREY
Live feeds remain essential for meeting the nutritional requirements at initiation of exogenous feeding of the larvae in most marine fish production facilities (Shields, 2001). The AA in live feed (phyto-and zooplankton) are mainly present in the form of various proteins. However, marine zooplankton contain a pool of FAA that comprises up to 10-20 % of their total AA content (Fyhn et al., 1995;van der Meeren et al., 2001).Alarge FAA pool is typical of most marine invertebrates, in which a very high intracellular FAA concentration contributes to osmoregulation in the saline environment (Yancey et al., 1982).This contrasts with typically low FAA concentrationsin freshwater invertebrates, including zooplankton, in which the low environmentalionic concentrations represent an opposite osmotic challenge. AFAApool of 2-5% of the total AA is typically found in freshwatercopepods and branchiopods (Dabrowskiand Rusiecki, 1983).Thenutrient composition of the natural prey available to marine and freshwater fish larvae thus appears to be quite different, depending on their respective habitats. It has been suggested that fish larvae are evolutionarily adapted to the nutritional composition of the natural prey in their particular environments (Fyhn, 1990; Fyhn et al., 1999) and that this limits the characteristics of the diet which they are able to digest and absorb. The FAA may be particularly important in the initial stages of exogenous feeding, when the digestive system is still largely undeveloped (see above). Although this hypothesis was proposed more than a decade ago (Fyhn, 1989) it still awaits proper experimentaltesting.
Digestion in marine fish larves
395
Marine phytoplankton also contain a pool of FAA (e.g. Fyhn et al., 1993). For some fish species it is common practice to add live microalgae ("green water" technique) to the rearing systems since this may enhance production (van der Meeren, 1991;Naas et al., 1992; Reitan et al., 1994; Lazo et al., 2000). It has also been shown that phytoplankton are actively filtered and ingested by the fish larvae for a short period before they start consuming zooplankton. Rapid release of the FAA during digestion may play a sigruficantphysiological role during early feeding. Brine shrimp (Artemia spp.), by far the most important live planktonic prey organism in aquaculture, is of freshwater origin although adapted to a highly saline medium. The FAA content of Artemia varies both with species and population, normally from 4-6% of total AA, and is markedly lower than that of marine copepods (Naess et al., 1995;Helland et al., 2000; van der Meeren et al., 2001). Studies of the post-mortem lysis of planktonic prey organisms under simulated digestive conditions have shown that small peptides (MW < 1500 Da), rather than FAA, are the major end-products of autolysis in Brachionus (Hjelmeland et al., 1993).These data are supported by studies showing that FAAcontent does not increase postmortem in Calanus (Fyhnet al., 1993)or Artemia (I. Holmefjord, pers. comm.).It should be noted that dietary peptides originating from protein hydrolysates have been shown to promote the development of brush-border enzyme expression and gut maturation in larval fish (Zambonino Infante et al., 1997; Cahu and Zambonino Infante, 1995a, 2001). Some attempts have been made to increase the content of low molecular weight water-soluble nutrients in live prey for marine fish. Liposomes have been tested as a means of enriching Artemia nauplii (Hontoria et al., 1994; Ozkizilcik and Chu, 1994) and also as a component of formulated feeds (Koven et al., 1999).Recent experiments on a laboratory scale have shown that the content of free methionine can be increased by up to 60 times in Artemia nauplii using a liposome technique (Tonheim et al., 2000). Once enriched, the FAA can be retained in Artemia for an extended period. In the Tonheim et al. (2000)study, 80%of the free methionine was still present eight hours after incubation in clean sea water (13°C).Furthermore, this study showed that the nature of the liposomes can affect FAA enrichment;higher levels of methionine in Artemia nauplii could be obtained by enrichment with liposomes made from pure egg yolk lecithin (99% phosphatidyl choline) than from crude egg yolk lecithin. These results suggest that Artemia nauplii can be efficiently enriched with specific FAA for the purpose of feeding fish larvae. A follow-upstudy was unable to demonstrateany sigruficant effect of incremental dietary free methionine in Artemia on growth of gilthead sea bream larvae (Lacuisse et al., 2002). This shows that an increase in dietary free methionine, the likely first limiting IAA in the Artemia FAApool, will not improve the growth of sea bream larvae. It remains to be clarified whether growth and survival might be significantlyimproved if liposomes included
398
Physiological
and ecological adaptations t o feeding
in
vertebrates
Videocamera Dissecting - microscope
B
Syringe
Well with incubation water C0,-trap
Fig. 15.1. An in vivo method for controlled tube-feeding in fish larvae. A: the experimental set up. The larva is placed in a water droplet under the dissecting microscope. B: Metabolically produced 14C0, is entrapped through aeration and manipulation of the pH of the incubation water. This allows the fraction evacuated from the gut (or excreted elsewhere) to be distinguished from 14C0, .originating from catabolism of the absorbed nutrients (adapted from hnnestad et al., 2001a). C: Early stages of different fish larvae being tubefed tests solutions.
and TCA solubles which include FAA (Fig. 15.4;Rojas-Garcia and bnnestad, 2003a). Data revealed a gradual increase in amount of tracer (14C-lysine) observed in the TCA-precipitated fraction in all compartmentsincluding the gut (Fig.15.4).This reflects incorporationof lysine and other AA in the cellular proteins of the digestive tract tissues. Using the data collected in the study, protein synthesis in the gut tissues was calculated to be about 20% of the protein synthesis in the whole body (Rojas-Garcia and bnnestad, 2003a). The high proportion of gut protein synthesis agrees with findings in other vertebrates, in which the gastrointestinal tract represents a substantial fraction of the protein turnover of the whole body due to active proliferation and secretion of the cells lining the lumen (McBride and Kelly, 1990). Observations of lower digestive absorption rates of protein than FAA (Rmnestad et al., 2000a; Barr et al., 2001, Rojas-Garcia and Rmnestad, 2003a) need to be verified with nonrnethylated proteins (Rojas-Garciaand Rmnestad,
Digestion in marine fish larves
I
Evacuated (Unabsorbed AA)
399
I
Fig. 15.2. Compartmental analysis of post-prandial handling of amino acids. The simplistic model is based on distribution of the tracer after tube-feeding a single pulse of 14Clabelled amino acids solution in combination with the experimental setup described in Fig. 15.2, sampling at various times, dissection of the gut and scintillation counting.
Test solution: U-I4C algal protein
"1
In incubation water
.. .. .. .. .... .... .... .... .... .... .... .... .... .... .... .... .. .. .. .. .. .. .. .. ........................................ .. .. .. .. .. .. .. .... .... .................................. ... .. ..... ..... ..... .... ... ... ................... ..... .... .. .. .. .... .... .... .. ... .... .... .... .. .... .... .. . ... ...:.
in larval body
TCA precipitates incl Protein
TCA solubles incl FAA-pool
... ... ... ... ... ... ... ... ... ...... ................................................... ... .... ... .... ... .... ... .... ....... ....... ....... ....... ....... ....... ....... ....... ....... ....... .. .. .. .. .. .. .. .... .... .... .... .... .... .... .... .... .... .. ... .. ... .. ... .. ... .. ..... ..... ..... ..... ..... ..... ..... ..... ..... ..... .. ... .. ... .. ... .. ..................................................... ............................. .. .. .. .. .. .. .. .. .. .. .. . . . ... .. ... .. ... .. ... .. ... .. ..... ..... ..... ..... ......................... .. .. .. .. .. .. .. .... .... .... .... .... .... .... .... .... .... .. .. ... .... ... .... ... .... ....... ....... ....... ....... ....................................:. .. .. .. .. .. .... .... .... .... .... .... .... .... .... .... ......... .. . .. . .. . .. ... ... ... ... ... ... ... ... . . .. .
Catabolised C02
Unabsorbed Evacuated
Fig. 15.3. Compartmental distribution of tracer in Atlantic halibut, Hippoglossus hippoglossus, post-larvae (46 days post-first feeding) tube-fed a 14C-labelled algal protein (adapted from Rlannestad et al., 2001a).
400
Physiological and ecological adaptations t o feeding in vertebrates Test solutions: Mixture of FAA added 14C labelled FAA
Lysine
Arginine
Glutamate
Alanine
Fig. 15.4. Postprandial handling (8h post-tube-feeding) of dietary free amino acids in post-larval Senegalese sole, Solea senegalensis, aged 32 days after hatching. The diet was administered as a single pulse (36 nL; 43.1 mmol/L) of a dissolved mixture of crystalline amino acids (AA). In the four separate treatments the diet contained L [U 14C]tracer for two indispensable AA, lysine and arginine or two dispensable AA, glutamate and alanine (adapted from Rannestad et al., 2001a).
2003b), but such findings are in line with previous data for juvenile and adult fish (Atlanticcod, Berge et al., 1994; Atlantic salmon, Espe et al., 1993, 1999; rainbow trout, Yamada et al., 1981) and other vertebrates, including man (Metgeset al., 2000).A recent study in premetamorphotic Atlantic halibut using a native and hydrolyzed 14C labeled salmon serum protein also supports these findings. However, in pigs with an identical amino acid composition. Deutz et al. (1996) found similar uptake rates for FAA, moderately hydrolyzed protein, and intact protein. The amino acid composition, concentration of protein, pH and ionic characteristics of dissolving medium are some of the important abiotic factors that can affect the intestinal digestion of dietary proteins. Nevertheless, the molecular complexity of the dietary nitrogen is clearly a key issue in larval nutrition.
Digestion in marine fish larves
AMINO ACID UTILIZATION AND REQUIREMENT
Fish larvae have very high relative growth rates compared to subsequent life stages (Karnler,1992;ConceiqZo, 1997;Otterlei et al., 1999).Growth in fish is primarily due to muscle protein deposition (Houlihan et al., 1995a; Carter and Houlihan, 2001) and it therefore follows that the flow of AA from food to the growingbiomass must be high. In addition to protein deposition, several studies have demonstrated that AA are a major fuel during the early life stages of severalmarine teleost species (Rernnestadand Fyhn, 1993;ConceiqZo et al., 1993; Finn et al., 1995a,b; Sivaloganathan et al., 1998;Parra et al., 1999; bnnestad et al., 1999; Finn et al., 2002). In fish, the same 10 AA are regarded as indispensable (IAA) as in other animals: arginine, histidine, leucine, isoleucine, valine, threonine, lysine, methionine, phenylalanine, and tryptophan (Wilson 1989). Tyrosine and cysteine are conditionally indispensable. In addition to this specific IAA requirement, fish also have a nonspecific requirement for amino groups, which can be met either by IAA or by dispensable AA (DAA). The common DAA are: glutamate, glutamine, aspartate, asparape, serine, alanine, glycine, proline, (cysteineand tyrosine). Total amino acid requirements in fish vary from species to species as a function of feeding habits. They can also differ within the same species, depending on several factors (Wilson, 1989):(1)size and age, with younger fish normally having higher requirements; (2) water temperature in some species, with the requirement increasing with temperature; (3) protein to energy balance; (4)dietary AA profile; and (5)digestibility of the diet. Estimates of the total AA requirements of juvenile fish are normally between 30 and 55% of the total diet (Wilson, 1989).Nonetheless, little is known about the AA requirements of larval fish and how these change during ontogeny. Compared to older fish, larvae have a higher total AA requirement (Dabrowski, 1986). Fiogb6 and Kestemont (1995) for instance, found that goldfish larvae have much higher IAA requirements (g AA.g-'protein)than juvenile or adult fish. Although most fish require a higher dietary protein concentration than other vertebrates, their dietary protein requirementsfor growth or maintenance are generally not higher. At growth rates normally observed, both protein intake (g protein ingested.g-'body weightday-') and protein retention efficiency (g protein retained.g" protein ingested)are comparable in fish and other vertebrates (Bowen, 1987). Thus, fish should not be regarded as displaying poor protein utilization, although they use a high proportion of their protein for energy purposes. Fish have a lower maintenance energy requirement than higher vertebrates, due to their poikilo- (exo-) thermic nature. Dietary AA are mostly absorbed as freeAA (FAA)and tissue concentrations are kept within narrow limits (Houlihan et al., 1995b).Tissue concentration
402
Physiological and ecological adaptations t o feeding in vertebrates
of FAA is low at the start of exogenous feeding but increases asymptotically with growth (R.N. Finn, Univ. Bergen, Norway, pers. comrn.).Throughout the larval phase of turbot larvae, the whole body total FAA pool ranged from 1to 4% larval dry weight (Concei$io, 1997).The large FAA pool present in all marine pelagic fish eggs analyzed to date is located in the yolk compartment and not in the cytoplasm (bnnestad et al., 1993; Finn, 1994).This FAA pool may account for up to 60% of the total AA (Thorsen, 1995; Rannestad et al., 1999) and has multiple functions for the developing embryo, including swelling prior to ovulation, buoyancy, and further is an important substrate for energy and protein synthesis during the endogenous feeding stages (Fyhn,1993; Wright and Fyhn, 2001). In the tissues, FAA are the currency of AA metabolism (Fig.15.6).Absorbed dietary AA may be used for the synthesis of protein. AA which are not polymerized into proteins can be used for energy production (catabolized), transaminated into other AA, used in gluconeogenesis or lipogenesis, or used in the synthesis of other nitrogen-containingmolecules (e.g. purines, pyrimidines or hormones). Furthermore, there is a dynamic relationship between the FAA and the protein pools, as protein is in continuous turnover. Absorption of individual AA in the gut depends on different transport systems (Jiirss and Bastrop, 1995) and may proceed at different rates. Variations in the rates of absorption of individual AA may lead to transitory AA imbalances and thus to an increase in AA catabolism. Different enzymes are involved in the transamination and catabolism of AA (Cowey and Walton, 1989;Jiirss and Bastrop, 1995),allowing for the differential use of individual AA in these processes. Amino acid utilization also depends on the dietary AA profile. Maximum utilization efficiency will depend on the digestibility and the rates of absorption of each AA, on the AA profile of proteins being synthesized, and on the use of individual AA for energy or other purposes. The ideal AA profile may differ between species, and change within species, depending on environmentalconditions (e.g.temperature), age, and physiological state. Amino acids are used for the synthesis of a number of nonproteinic Ncontaining molecules. However, losses of AA through these pathways in vertebrates are generally regarded as of little quantitative significance, especially for the IAA (Simon, 1989).AA are precursors in the synthesis of many other important biomolecules involved in the growth process, e.g. purines and pyrimidines of the nucleic acids, the porphyrine nucleus of hemoglobin and cytochromes, phosphocreatine, as well as peptide- or amino acid-derived hormones (see Bender, 1985and Stryer, 1995for details). In fish, AA are reported to be the best precursors for lipid and carbohydrate synthesis (Nagai and Ikeda, 1972,1973).The a-keto acids resulting from AA deamination can be used for this purpose through gluconeogenesisand/or lipogenesis. Whether a given AA can be used as a precursor for carbohydrates
Digestion in marine fish larves
403
Amino Acids in Feed
Pyrimidins Hormones etc
Transamination Protein synthesis Protein Protein degradation
A
4
Gluconeogenesis, Lipidogenesis
a-ketoacids
b 4
Lipids Carbohydrates
Catabolism
Fig. 15.5. Overview of the processes involved in amino acid metabolism.
(glucogenic)and/or lipids (ketogenic)depends on its carbon backbone (see Stryer, 1995for details). Gluconeogenesisis believed to be a minor pathway of AA metabolism in fish and is only sigruficantin fish larvae after the onset of exogenousfeeding (vanWaarde, 1988).Of the gluconeogenicAA, alanine is the most important in fish (Walton, 1985). This suggests that lactate recycling is an important function of gluconeogenesis.In addition, gluconeogenesis is also involved in the synthesis of glucose from dietary AA, supply of glucose to some tissues, furniture of metabolic intermediates for metabolic pathways such as ribose and deoxyribose synthesis. DAA can be synthesized de novo from a-keto acids or through transamination and other reactions from both IAA and DAA. Glutamate has a pivotal role, being involved in the synthesis of most of the other DAA, either as a precursor or as an amino group donor (see Bender, 1985; Stryer, 1995). Cysteine and tyrosine are regarded as special cases since, although they are DAA, each can only be synthesized from a single IAA, methionine and phenylalanine respectively. Synthesis of DAA from glucose has been demonstrated in fish, although it is not known whether this de novo synthesis of AA is of quantitative significance (Cowey and Walton, 1989).
404
Physiological and ecological adaptations t o feeding in vertebrates
Transaminases and other AA-converting enzymes have been found in juvenile and adult fish (for review see Cowey and Walton, 1989),but their importance in the AA flux is largely not known. They can be important in improving protein utilization as they may compensate for DAA imbalances in dietary protein. In African catfish larvae, transaminase activities changed both with development and with type of diet (Segner and Verreth, 1995). Amino acid catabolism involves removal of the amino group (deamination).After deamination, the resultant a-keto acids can be oxidized to carbon dioxide and water via the tricarboxylic acid (TCA) cycle, but can also be used in lipid or carbohydrate synthesis.The pathways through which the carbon backbones of individual AA enter the TCA cycle are complex and sometimes multiple (for details see Cowey and Walton, 1989; Stryer, 1995). As pointed out above, larval fish may have an even higher AA catabolism than older stages. The high AA usage for energy purposes in fish larvae has been attributed to a reduced catabolic adaptability, related to their strictly carnivorous nature (Dabrowski, 1986).Even so, Senegalese sole postlarvae (Rernnestad et al., 2001a), herring larvae (Conceil50 et al., 2002a) and early metamorphosing Atlantic halibut larvae (Applebaum and b m e s t a d , 2003) all use DAA preferentially to IAA as energy substrates. Therefore, fish seem to have the capacity to spare their IAA at the expense of DAA from very early stages of development. These results suggest that fish larvae and postlarvae have some control over AA catabolism comparable to that of juvenile fish (Cowey and Sargent, 1979;Kim et al., 1992)and other animals (e.g.Tanaka et al., 1995; Heger et al., 1998; Roth et al., 1999). In some species of adult fish maintenance requirements are almost completely met by AA (Brett and Zala, 1975)and AA may contribute more than 40% of the energy expenditure required for routine activity (van Waarde, 1983). Amino acids are also an important source of energy in eggs and in yolk-sac larvae (e.g.Dabrowski et al., 1984; Fyhn and Serigstad, 1987; Fyhn, 1989; Rernnestad et al., 1992a,b, 1993; Rernnestad and Fyhn, 1993; Verreth et al., 1995; Finn et al., 1995,1996).
AMINO ACID CATABOLISM AND FUEL ALLOCATION
Few studies have examined the importance of AA catabolism in exogenously feeding larvae but existing data based on comparison of the molar rates of ammonia excretion and oxygen consumption indicate that they continue to be a major fuel at this stage of development. With lipids as the assumed cosubstrate, amino acids have been estimated to account for 70-95% of total substrate oxidation of larvae in the first 3 4 weeks post hatch. Beyond this age (corresponding to a standard length of 7 mm), reliance on amino acids as fuel begins to decline but even in juveniles of 40-60 mm SL, amino acids still represent the dominant source of carbon fueling metabolism (Finn et al., 2002).
Digestion i n marine fish larves
405
Findings on fuel allocation during energy dissipation are based on flux (respiration and excretion) studies and in the endogenous feeding stages they are also supported by combined compositional (content)measurement studies (e.g. Rsnnestad et al., 1992a,b, 1994; RQnnestad and Fyhn, 1993; Finn et al., 1995a,b, 1996).Based on the flux studies it is usual to calculate the nitrogen quotient (NQ), which is the molar ratio of the nitrogenous end products ammonia and urea to the oxygen consumption (Gnaiger, 1983). Using the NQ it is possible to estimate the contribution of AA fuel to energy dissipation (Gnaiger, 1983; Finn, 1994; Lauff and Wood, 1996). The stoichiometric calculation is based on a state of fully aerobic metabolism, a condition which has been demonstrated in turbot larvae (Finnet al., 1995~). Depending on the specific AA utilized, NQ values of 0.23-0.25reflect a rate of energy dissipation which is fully based on AA utilization (Finn et al., 1995a,b, 1996). At lower NQ values, the contribution of AA to energy expendituredepends on the degree of participation of lipid or carbohydrate fuels. Identification of a particular fuel can be further refined by simultaneous measurements of CO, flux. While NQ values are a powerful tool in studying fish larvae energy resource partitioning, calculation of the true NQ value requires detailed flux analysis of all the quantitativenitrogenous end products (Wright and Fyhn, 2001; Terjesen, 2001; Terjesen et al., 2002). USE OF DIETARY PROTEINS, PEPTIDES, AND FREE AMINO ACIDS
In stomachlessstages, in particular, fish larvae seem to have a limited capacity for digesting proteins. However, these lirmtationsmay be related to the nature and complexity of the dietary proteins. Srivastava et al. (2002) performed invitro studies on digestion and demonstrated that while some protein species are easily digested, other proteins display a strong resistance to proteolysis under conditions similar to those in first-feeding Japanese flounder. Fish meal-based microdiets have been shown to have a higher proportion of larger molecular weight proteins compared to Artemia nauplii (Garcia-Ortega, 1999). This may help to explain the traditional difficulties found in weaning marine fish larvae with fish meal-based diets. Partial replacement of native protein by di- and tripeptides in diets for larval sea bass resulted in improved development and survival in European sea bass (Zambonino Infante et al., 1997).The best results in this study were found in sea bass larvae fed at a 20% peptide substitution level. According to Zambonino Infante et al. (1997),the improved larval performance observed in the peptide-fed groups was related to enhanced proteolytic capacity of the pancreas and to the earlier development of intestinal digestion. Moderate, but not high levels of protein hydrolysates have been shown to improve growth and increase survival in young stages of different fish species
406
Physiological and ecological adaptations t o feeding i n vertebrates
(Zambonino Infante et al., 1997;Carvalho et al., 1997; Day et al., 1997;Cahu et al., 1999; Hamre et al., 2001). Adult fish fed diets containing high levels of FAA supplements have lower growth and poorer assimilation efficiency than fish fed diets containing intact protein of identical AA composition (Espe and Lied, 1994;Berge et al., 1994; Zarate et al., 1999).These differenceshave been explained by reduced feed intake (Ng et al., 1996) or differences in digestion and absorption rates of the constituent AA (Yamada et al., 1981). However, poor results with inclusion of low molecular weight nitrogen sources in larval fish diets may also be caused by high rates of leaching of small peptides and FAA from the microdiets (Lopez-Alvarado et al., 1994), producing a diet of reduced nutritional value and possibly unbalanced at the time of ingestion. Recently, protein substitution levels as high as 50% hydrolysate (containing di- and tripeptides) have been reported to greatly reduce the incidence of larval malformations in sea bass (Cahu et al., 2001). In piglets, higher intestinal absorption rates of lysine and glycine have been demonstrated in animals fed dipeptides containing these AA compared to those fed the same AA in a free form (Liet al., 1999).It was suggested that this might be due to a special transport mechanism for dipeptides (Li et al., 1999). Further studies are needed to elucidate the absorption kinetics and role of peptides in marine fish larval nutrition. A potential pitfall in using dietary FAA is overloading the metabolic systemsby AA flux from the digestivetract. Studiesby Li et al. (1999)suggest that metabolic overloading could be an even larger problem when using small peptides. Since the FAA pool in feeding fish larvae is small and kept within narrow limits (Houlihan et al., 199513; ConceiqZio, 1997), absorbed dietary AA are either used for protein synthesis or are processed otherwise. AA can be channeled into energy production, gluconeogenesisor lipogenesis (Bergeet al., 1994),or may even be lost as intact molecules through the urine or gills. Urinary loss is not a major route of AA excretion in older fish (Ng et al., 1996).In larval fish it is difficult to distinguish experimentally between FAA loss via urine, the glls, or by evacuation.However, as mentioned above, recent data (hnnestad et al., 2001a, b; Conceiqiio et al., 2002a; Applebaum and Rnmnestad, 2003) suggest that only a small proportion of the FAA supplied to fish larvae as a single pulse by tube-feeding is lost in intact form. High FAA availability has also been demonstrated in juvenile rainbow trout (Rodehutscord et al., 2000), in which free L-lysine was totally absorbed from the intestine. This does not support the suggestion of Kolkovski (2001)that a rapid flow of short peptides and FAA through the gut wall cannot be handled in terms of FAA absorption and that, as a result, most of these metabolites will be flushed out of the digestive system.Whether Merences in FAAloading to the gut, or in other experimental procedures, may explain these differences remains to be shown.
Digestion in marine fish larves
Test solution: FAA Added [U]
I4c lysine
GUT
100
80
s a>
--
60
40
0
a
a
20
3
0
+
LIVER
u
CARCASS
2
4
6
8
Time after tubefeeding (h) Fig. 15.6. Postprandial handling of a tube-fed a mix of dissolved FAA (based on the composition of Bovine Serum Albumin) with added I4C-labeled lysine in Atlantic halibut (Hippoglossus hippoglossus) at 59-60 days post-first feeding (adapted from Rojas-Garcia and Rernnestad, 2003a).
The benefits of using a CO, trap when studying evacuation of intact FAA was demonstrated in the studies of Rlannestad et al. (2001a,b; Fig. 15.6). Taking alanine and glutamate as examples, these authors showed that 45% and 68% of the total labeled FAA tube-fed to Senegalese sole were recovered in the water and that virtually all (95%)of this was due to released CO,, i.e. catabolism (bnnestad et al., 2001b).This emphasizesthe need for separating evacuation and excretion in studies of digestive absorption efficiency. At the
408
Physiological and ecological adaptations t o feeding in vertebrates
same time, the high catabolism of AA demonstrated in this study, 41% and 65% of total fed AA (Rannestad et al., 2001b), supports the earlier estimates of AA catabolism based on apparent NQ. Further studies with increasing concentrations of FAA should be performed in order to determine the limit above which the FAA begins to be evacuated. The challenge in optimizing the intestinal performance of fish larvae in relation to dietary AA is probably to find the ideal balance between the different molecular forms of AA. Furthermore, improvement of dietary AA utilization will also depend on defining the ideal dietary IAA profile at different larval stages (Conceiqsoet al., 2003).Finally, despite the relatively immature digestive system of marine fish larvae, it should be stressed that they have one of the highest potential growth rates among vertebrates. Acknowledgments
Supported by Research Council of Norway projects 141990/120,138382/ 140,115876/122, FCT (Portugal) grant - SFRH/BPD/7149/2001, and project POCTI/ 1999/CVT/34608 (FCT,Portugal and FEDER, European Union).This is publication #I32 from the University of Bergen Fish Larval Locus. REFERENCES Alvarez M.d.C., Perez R., Seikai T., Takahashi Y. and Tanaka M. 1999. Ontogenetic development of the digestive enzyme activities under different feeding regimes during the early life stages in Japanese flounder, Paralichthys olivaceus. J. Fish Biol. in press. Applebaum S.L and Rsnnestad I. 2004. Absorption, assimilation and catabolism of individual free amino acids by late larval Atlantic halibut (Hippoglossus hippoglossus). Aquaculture 230: 313-322. Barr Y., Rojas-Garcia C.R. and Rsnnestad I. 2001. The digestion capacity of protein and amino acids in the Atlantic halibut (Hippoglossus hippoglossus). In: Larvi 2001. C.I. Hendry, G. van Stappen, W. Willie, and P. Sorgeloos (eds.). Eur. Aquacult. Soc., Oostende, Belgium, Sp. Publ. no. 30, pp. 50-53. Bender D.A. 1985. Amino Acid Metabolism. John Wiley & Sons, New York, NY. Bengtson D.A., Borms D.N., Leibovitz H.E. and Simpson K.L. 1993. Studies on structure and function of the digestive system of Medidia beryllina (Pisces, Atherinidae). In: Physiology and Biochemistry of Fish Larval Development. B.T. Walther and H.J. Fyhn (ed.). Univ. Bergen, Bergen Press, Bergen, Norway, pp. 199-208. Berge G.E., Lied E. and Espe M. 1994. Absorption and incorporation of dietary free and protein bound (U14C)-lysinein Atlantic cod (Gadus morhua ). Comp. Biochem. Physiol. 109A: 681-688. Bowen S.H. 1987. Dietary protein requirement of fishes - a reassessment. Can. J. Fish. Aquat. Sci. 44: 1995-2001. Brett J.R. and Zala C.A. 1975. Daily pattern of nitrogen excretion and oxygen consumption of sockeye salmon (Oncorhynchus nerka) under controlled conditions. J. Fish. Res. Bd Can. 32: 2479-2486. Buddington R.D. 1985. Digestive secretions of lake sturgeon, Acipenser fulvescens, during early development. J. Fish Biol. 26: 715-723. Cahu C.L. and Zambonino Infante J.L. 1995a. Effect of the molecular form of dietary nitrogen supply in sea bass larvae: Response of pancreatic enzymes and intestinal peptidases. Fish Physiol. Biochem. 14: 209-214.
Digestion in marine fish larves
409
Cahu C.L. and Zambonino Infante J.L. 1995b. Maturation of the pancreatic and intestinal digestive functions in sea bass (Dicentrarchus labrax): effect of weaning with different protein sources. Fish. Physiol. Biochem. 14: 431-437. Cahu C.L. and Zambonino Infante J.L. 2001. Substitution of live food by formulated diets in marine fish larvae. Aquaculture 200: 161-180. Cahu C.L., Zambonino Infante J.L. and Takeuchi T. 2001. Nutrients affecting quality in marine larval fish development. In: Larvi 2001. C.I. Hendry, G. van Stappen, W. Willie, and P. Sorgeloos (ed.). Oostende, Belgium: Eur. Aquacult. Soc., Sp. Publ. no. 30, pp. 94-95. Cahu C.L., Zambonino Infante J.L., Quazuguel P. and LeGall M. M. 1999. Protein hydrolysate vs. fish meal in compound diets for 10-day-old sea bass Dicentrarchus labrax larvae. Aquaculture 171: 109-119. Carter C.G. and Houlihan D.F. 2001. Protein synthesis. In: Fish Physiology, vol. XX P.A. Wright and P.M. Anderson (eds.). Acad. Press, New York, NY, pp. 178-252. Carvalho A.P., Escaffre A.-M., OlivaTeles A. and Bergot P. 1997. First feeding of common carp larvae on diets with high levels of protein hydrolysates. Aquacult. Int. 5: 361-367. Chitty N. 1981. Behavioural observations of feeding larvae of bay anchovy, Anchoa mitchilli, and bigeye anchovy, Anchoa lamprotaenia. ~ a ~ P-V-. p . Reun. Cons. Int. Explor. Mer. 178: 320-321. ConceiqHo L.E.C. 1997. Growth in early life stages of fishes: an explanatory model. PhD Thesis, Wageningen Agric.Univ. Wageningen, Netherlands. ConceiqHo L.E.C., Rennestad I. and Tonheim S.K. 2002a. Metabolic budgets for lysine and glutamate in unfed herring (Clupea harengus) larvae. Aquaculture 206: 305-312. ConceiqZo L.E.C., Grasdalen H. and Rennestad I. 2003. Amino acid requirements of fish larvae and post-larvae: new tools and recent findings. Aquaculture. 227: 221-232. ConceicHo L.E.C., Verreth J.A.J., Scheltema T. and Machiels M. 1993. A simulation model for the metabolism of yolk sac larvae of the African catfish Clarias gariepinus. Aquacult. Fish. Manage. 24: 297-309. ConceiqHo L.E.C., Skjermo J., Skjik-Braek G. and Verreth J.A.J. 2002b. Effects of an immunostimulating alginate on protein turnover of turbot (Scophthalmus maximus L.) larvae. Fish Physiol. Biochem. 24: 207-212. Cowey C.B. and Sargent J.R. 1979. Nutrition. In: Fish Physiology, vol. VIII. W.S. Hoar, D.J. Randall and J.R. Brett (eds.). Acad. Press, New York, NY, pp. 1-70. Cowey C.B. and Walton M.J. 1989. Intermediary metabolism. In: Fish Nutrition, J.E. Halver (ed.). Acad. Press, New York, NY, pp. 259-329. Dabrowski K.R. 1986. Ontogenetical aspects of nutritional requirements in fish. Comp. Biochem. Physiol. 85A: 639-655. Dabrowski K. and Rusiecki M. 1983. Content of total and free amino acids in zooplanktonic food of fish larvae. Aquaculture 30: 3142. Dabrowski K., Kaushik S.J. and Luquet P. 1984. Metabolic utilization of body stores during the early life of whitefish Coregonus lavaretus L. j. Fish Biol. 24: 721-729. Day O.J., Howell B. R. and Jones D.A. 1997. The effect of dietary hydrolysed fish protein concentrate on the survival and growth of juvenile Dover sole, Solea solea (L.), during and after weaning. Aquacult. Res. 28: 911-921. Deutz N.E.P., Welters C.F.M. and Soeters P.B. 1996. Intragastric bolus feeding of meals containing elementary, partially hydrolyzed or intact protein causes comparable changes in interorgan substrate flux in the pig. Clin. Nutr. 15: 119-128. Douglas S.E., Gawlicka A., Mandla S. and Gallant J.W. 1999. Ontogeny of the stomach in Winter flounder: characterization and expression of the pepsinogen and proton pump genes and determination of pepsin activity. J. Fish. Biol. 55: 897-915. Espe M. and Lied E. 1994. Do Atlantic salmon (Salmo salar) utilize mixtures of free amino acids to the same extent as intact protein sources for muscle protein synthesis? Comp. Biochem. Physiol. 107A: 249-254. Espe M., Lied E. and Torrissen K.R. 1993. Changes in plasma and muscle free amino acids in Atlantic salmon (Salmo salar) during absorption of diets containing different amounts of hydrolysed cod muscle protein. Comp. Biochem. Physiol. 105A: 555-562.
410
Physiological and ecological adaptations t o feeding in vertebrates
Espe M., Sveier H., Hsgsy I. and Lied E. 1999. Nutrient absorption and growth of Atlantic salmon (Salmo salar L.) fed fish protein concentrate. Aquaculture 174: 119-137. Fernindez-Diaz C. and Yiifera M. 1997. Detecting growth in gilthead seabream, Sparus aurata L.; larvae fed microcapsules. Aquaculture 153: 193-102. Finn R.N. 1994. Physiological energetics of developing marine fish embryos and larvae. PhD. thesis, Univ. of Bergen, Bergen, Norway. Finn R.N., Rsnnestad I. and Fyhn H.J. 1995b. Respiration, nitrogen and energy metabolism of developing yolk-sac larvae of Atlantic halibut (Hippoglossus hippoglossus L.). Comp. Biochem. Physiol. 111A: 647-671. Finn R.N., Widdows J. and Fyhn H.J. 1995c. Calorespirometry of developing embryos and yolk-sac larvae of turbot (Scophthalmus maximus). Mar. Biol. 122: 157-163. Finn R.N., Fyhn H.J., Henderson R.J. and Evjen M.S. 1995a. Physiological energetics of developing embryos and yolk-sac larvae of Atlantic cod (Gadus morhua); I. Respiration and nitrogen metabolism. Mar. Biol. 124: 355-369. Finn R.N., Fyhn H.J., Henderson R.J. and Evjen M.S. 1996. The sequence of catabolic substrate oxidation and enthalpy balance of developing embryos and yolk-sac larvae of turbot (Scophthalmus maximus L.). Comp. Biochem. Physiol. 115A: 133-155. Finn R.N., Rsnnestad I., van der Meeren T. and Fyhn H.J. 2002. Effect of light and temperature on the metabolic scaling and utilisation of amino acid fuels in Atlantic cod (Gadus morhua). Mar. Ecol. Progr. Ser. 243 : 217-234. FiogbC E.D. and Kestemont P. 1995. An assessment of the protein and amino acid requirements in goldfish (Carassius auratus) larvae. J. Appl. Ichthyol. 11: 282-289. Fyhn H.J. 1989. First feeding of marine fish larvae: Are free amino acids the source of energy?. Aquaculture 80: 111-120. Fyhn H.J. 1990. Energy productin in marine fish larvae with emphasis on free amino acids as a potential fuel. In: Comparative Physiology. Animal Nutrition and Transport Processes. I. Nutrition in Wild and Domestic Animals, vol. 5. J. Mellinger (ed.). Karger, Basel, pp. 176-192. Fyhn H.J. 1993. Multiple functions of free amino acids during embryogenesis in marine fishes. In: Physiology and Biochemistry of Fish Larval Development. B.T. Walther and H.J. Fyhn (eds.). Univ. Bergen, Bergen, Norway, pp. 299-308. Fyhn H.J. and Serigstad B. 1987. Free amino acids as energy substrate in developing eggs and larvae of the cod Gadus nzorhua. Mar. Biol. 96: 335-341. Fyhn H.J., Rsnnestad I. and Berg L. 1995. Variation in free and proteinic amino acids of marine copepods during the spring bloom. In: Larvi. 95, I? Lavens, E. Jaspers and I. Roelants (eds.), European Aquacul. Soc., Gent, Belgium, Sp. Publ. 24, pp. 321-324. Fyhn H.J., Finn R.N., Reith M. and Norberg B. 1999. Yolk protein hydrolysis and oocyte free amino acids as key features in the adaptive evolution of teleost fishes to seawater. Sarsia 84: 451456. Fyhn H.J., Finn R.N., Helland S., Rsnnestad I. and Lsmsland E. 1993. Nutritional value of phyto- and zooplankton as live food for marine fish larvae. In: Fish Farming Technology. H. Reinertsen, L.A. Dahle, L. Jargensen. and K. Tvinnerheim (eds.).Balkema Publ., Rotterdam, Netherlands, pp. 121-126. Garcia-Ortega A. 1999. Nutritional value of decapsulated cysts of Artemia and their use as protein source in experimental microdiets for fish larvae. PhD thesis, Agric. Univ. Wageningen, Netherlands. Gawlicka A., Leggiadro C.T., Gallant J.W. and Douglas S.E. 2001. Cellular expression of the pepsinogen and gastric proton pump genes in the stomach of Winter flounder as determined by in situ hybridization. J. Fish. Biol. 58: 529-536. Gawlicka A., Parent B., Horn M.H., Ross N., Opstad I. and Torrissen 0. 2000. Activity of digestive enzymes in yolk-sac larvae of Atlantic halibut (Hippoglossus hippoglossus): indication of readiness for first feeding. Aquaculture 184: 303-314. Gnaiger E. 1983. Calculation of energetic and biochemical equivalents of respiratory oxygen consumption. In: Polarographic Oxygen Sensors. E. Gnaiger and H. Forstner (eds.). Springer-Verlag, Berlin.
Digestion in marine fish larves
4.11
Govoni J. J., Boehlert G. W. and Watanabe Y. 1986. The physiology of digestion in fish larvae. En. Biol. Fish. 16: 59-77. Hamre K., Naess T., Espe M., Holm J.C. and Lie 0. 2001. A formulated diet for Atlantic halibut (Hippoglossus hippoglossus, L.) larvae. Aquacult. Nutr. 7: 123-132. Harboe T. and Mangor-Jensen A. 1998. Time of first feeding of Atlantic halibut, Hippoglossus hippoglossus L., larvae. Aquacult. Res. 29: 913-918. Heger J., Mengesha S. and Vodehnal D. 1998. Effect of essential: total nitrogen ratio on protein utilization in the growing pig. Brit. J. Nutr. 80: 537-544. Helland S., Triantaphyllidis G. V., Fyhn H. J., Evjen M. S., Lavens P. and Sorgeloos P. 2000. Modulation of the free amino acid pool and protein content in populations of the brine shrimp Artemia spp. Mar. Biol. 137: 1005-1016. Hjelmeland K. 1995. Trypsin in fish. PhD thesis, Univ. of Tromss, Tromss, Norway. Hjelmeland K., Ugelstad I. and Olsen Y. 1993. Proteolytic activity and post mortem autolysis in prey for marine fish larvae. In: Physiology and Biochemistry of Fish Larval Development. B.T. Walther and H.J. Fyhn (eds.). Univ. Bergen, Bergen, Norway, pp. 229-232. Hoehne-Reitan K., Kjersvik E. and Reitan K.I. 2001. Development of the pH in the intestinal tract of larval turbot. Mar. Biol. 139: 1159-1164. Hontoria F., Crowe J.H., Crowe L.M. and Amat F. 1994. Potential use of liposomes in larviculture as a delivery system through Artemia nauplii. Aquaculture. 127: 255-264. Houlihan D.F., Carter C.G. and McCarthy I.D. 1995a. Protein synthesis in fish. In: Fish Molecular Biology and Biochemistry, Vol. 4. P. Hochachka and T. Mommsen (eds.). Elsevier Press, Amsterdam, Netherlands, pp. 191-220. Houlihan D.F., McCarthy I.D., Carter C.G. and Martin F. 1995b. Protein turnover and amino acid flux in fish larvae. ICES Mar. Sci. Symp. 201: 87-99. Iwai T. and Tanaka M. 1968. The comparative study of digestive tract of teleost larvae111. Epithelial cells in the posterior gut of halfbeak larvae. Bull. Jap. Soc. Sci. Fish. 34: 44-48. Jany K.D. 1976. Studies on the digestive enzymes of the stomachless bonefish Carassius auratus gibe10 (Bloch). Endopeptidases. Comp. Biochem. Physiol. 53B: 31-38. Jobling M., Hjelmeland K. 1992. Ernaering og metabolisme. In: Fiskens Fysiologi K. Dsving and E. Reimers (eds.). John Grieg Forlag, Bergen, Norway, pp. 234-276. Jiirss K. and Bastrop R. 1995. Amino acid metabolism in fish. In: Biochemistry and Molecular Biology of Fishes, Vol. 4. P. Hochachka and T. Mommsen (eds.). Elsevier Press, Amsterdam, Netherlands, pp. 159-189. Kamler E. 1992. Early Life History of Fish: A n Energetic Approach. Chapman & Hall, London. Kendall A. W., Ahlstrom E. H. and Moser H. G. 1984. Early life history stages of fishes and their characters. In: Ontogeny and Systematics of Fishes. H.G. Moser, W.J. Richards, D.M. Cohen, M.P. Fahay, A.W. Kendall, and S.L. Richardson (eds.).Amer. Soc. Ichthyol. Herpetol, pp. 11-22. Kim K., Grimshaw T.W., Kayes T.B. and Amundson C.H. 1992. Effect of fasting or feeding diets containing different levels of protein or amino acids on the activities of the liver amino acid-degrading enzymes and amino acid oxidation in rainbow trout (Oncorhynchus mykiss). Aquaculture 107: 89-105. Kjersvik E. and Reiersen L. 1992. Histomorphology of the early yolk-sac larvae of the Atlantic halibut (Hippoglossus hippoglossus)-an indication of the timing of functionality. J. Fish Biol. 41: 1-19. Kolkovski S. 2001. Digestive enzymes in fish larvae and juveniles - implications and applications to formulated diets. Aquaculture 200: 181-200. Koven W., Kolkovski S., Hadas E. and Gamsiz K.T.A. 2001. Advances in the development of microdiets for gilthead seabream, Sparus aurata: a review. Aquaculture 194: 107-121. Koven W., Barr Y., Hadas E., BenAtia I., et al. 1999. The potential of liposomes as a nutrient supplement in first-feeding marine fish larvae. Aquacult. Nutr. 5: 251-256.
412
Physiological and ecological adaptations t o feeding in vertebrates
Kurokawa T. Shiraishi M. and Suzuki T. 1998. Quantification of exogenous protease derived from zooplankton in the intestine of Japanese sardine (Sardinops melanostictus) larvae. Aquaculture 161: 491-499. Kurokawa T. and Suzuki T. 1996. Formation of the diffuse pancreas and the development of digestive enzyme synthesis in larvae of the Japanese flounder Paralichthys olivaceus. Aquaculture 141: 267-276. Kurokawa T., Tanaka H., Kagawa H. and Otha H. 1996. Absorption of protein molecules by the rectal cells in eel larvae Anguilla japonica. Fish. Sci. 62: 832-833. Lacuisse M., Koven W., Rsnnestad I., Lutzky S., et al., 2002. The use of liposomes to enrich Artemia nauplii with the limiting EAA methionine in order to increase growth and protein retention in gilthead seabream (Sparus aurata) larvae. In Tenth Int. Symp. Nutr. Feeding Fish. Rhodes, Greece, June 2002. Lauff R.F. and Wood C.M. 1996. Respiratory gas exchange, nitrogenous waste excretion, and fuel usage during aerobic swimming in juvenile rainbow trout. J. Comp. Physiol. 166B: 501-509. Lazo J.P., Dinis M.T., Holt J.G., Faulk C. and Arnold C.R. 2000. Co-feeding microparticulate diets with algae: towards eliminating the need 'of zooplankton at first feeding in red drum (Sciaenops ocellatus). Aquaculture 188: 339-351. Lein I. and Holmefjord I. 1992. Age at first feeding of Atlantic halibut larvae. Aquaculture 105: 157-164. Li D.F., Zhao X.H., Yang T.B., Johnson E.W. and Thacker P.A. 1999. A comparison of the intestinal absorption of amino acids in piglets when provided in free form or as a dipeptide. Asian-Australasian J. Anim. Sci. 12: 939-943. Lopez-Alvarado J., Langdon C.J., Teshima S.I. and Kanazawa A. 1994. Effects of coating and encapsulation of crystalline amino acids on leaching in larval feed. Aquaculture 122: 335-346. Luizi F.S., Gara B., Shields R.J. and Bromage N.R. 1999. Further description of the development of the digestive organs in Atlantic halibut (Hippoglossus hippoglossus) larvae, with notes on differential absorption of copepod and Artemia prey. Aquaculture 176: 101-116. Mahr K., Graubner R. Hofer R., and Moser H. 1983. Histological and physiological development of the stomach in Coregonus sp. Polskie Arch. Hydrobiol. 98:344-356. Mangor-Jensen A. and Adoff G.R. 1987. Drinking activity of the newly hatched larvae of cod Gadus morhua L. Fish Physio1.Biochem. 3: 99-103. McBride B.W. and Kelly J.M. 1990. Energy cost of absorption and metabolism in the ruminant gastrointestinal tract and liver a review. J. Anim. Sci. 68:2997-3010. McLean E., Ronsholdt B., Sten C. and Najamuddin J.F. 1999. Gastrointestinal delivery of peptide and protein drugs to aquacultured teleosts. Aquaculture 177: 231-247. Metges C.C., ElKhoury A.E., Selvaraj A.B., Tsay R.H. et al. 2000. Kinetics of L-[I-C13lleucine when ingested with free amino acids, unlabeled or intrinsically labeled casein. Amer. J. Physiol. 278: E1000-E1009. Munilla-Moran R. and Stark J.R. 1989. Protein digestion in early turbot larvae, Scophthalmus maximus (L.). Aquaculture 81: 315-327. Murakami K. and Etlinger J.D. 1987. Degradation of proteins with blocked amino groups by cytoplasmaic proteases. Biochem. Biophys. Res. Comm. 146: 1249-1255. Naas K.E., Naess T. and Harboe T. 1992. Enhanced feeding of halibut larvae (Hippoglossus hippoglossus L.) in green water. Aquaculttire 105: 143-156. Naess T., Germain-Henry M. and Naas K.E. 1995. First feeding of Atlantic halibut (Hippoglossus hippoglossus) using different combinations of Artemia and wild plankton. Aquaculture 130: 235-250. Nagai M. and Ikeda S. 1972. Carbohydrate metabolism in fish. 111. Effects of dietary composition om metabolism of glu~ose-U-~~C and glutamate-U-14C in carp. Bull. Jpn. Soc. Scie. Fish. 38: 137-143. Nagai M. and Ikeda S. 1973. Carbohydrate metabolism in fish. IV. Effects of dietary composition on metabolism of acetate-U-14C and alanine-U-14Cin carp. Bull. Jpn. Soc. Scie. Fish. 39: 633-643.
Digestion in marine fish larves
413
Ng W.K., Hung S.S.O. and Herold M.A. 1996. Poor utilization of dietary free amino acids by white sturgeon. Fish Physiol. Biochem. 15: 131-142. 0 i e G., Makridis P., Reitan K.I. and Olsen Y. 1997. Protein and carbon utilization in rotifers (Brachionus plicatilis) in first feeding turbot larvae (Scophthalmus maximus). Aquaculture 153: 103-122. Oozeki Y. and Bailey K.M. 1995. Ontogenetic development of digestive enzyme activities in larval walleye pollock, Theragra chalcogramma. Mar. Biol. 122: 177-186. Otterlei E., Nyhammer G., Folkvord A. and Stefansson S.O. 1999. Temperature- and size-dependent growth of larval and early juvenile Atlantic cod (Gadus morhua): A comparative study of Norwegian coastal cod and northeast Arctic cod. Can. J. Fish. Aqua. Sci. 56: 2099-2111. Ozkizilcik S. and Chu F.E. 1994. Uptake and metabolism of liposomes by Artemia nauplii. Aquaculture. 128: 131-141. Parra G., Rennestad I. and Yufera M. 1999. Energy metabolism in developing eggs and unfed larvae of Solea senegalensis. J. Fish Biol. 55: 205-214. Pedersen B.H., Nilssen E.M. and Hjelmeland K. 1987. Variations in the content of trypsin and trypsinogen in larval herring (Clupea harengus) digesting copepod nauplii. Mar. Biol. 94: 171-181. Pedersen B.H., Ugelstad I. and Hjelmeland K. 1990. Effects of a transitory, low food supply in the early life of larval herring (Clupea harengus ) on mortality, growth and digestive capacity. Mar. Biol. 107: 61-66. Pittman K., Skiftesvik A.B. and Berge L. 1990. Morphological and behavioural development of halibut, Hippoglossus hippoglossus (L.) larvae. J. Fish Biol. 37: 455-472. Reitan K.I., Bolla S. and Olsen Y. 1994. A study of the mechanism of algal uptake in yolk-sac larvae of Atlantic halibut (Hippoglossus hippoglossus L.). J. Fish Biol. 44: 303-310. Rodehutscord M., Borchert F., Gregus Z., Pack, M. and Pfeffer E. 2000. Availability and utilisation of free lysine in rainbow trout (Oncorhynchus mykiss), 1. Effect of dietary crude protein level. Aquaculture 187: 163-176. Rojas-Garcia C.R. and Rannestad I. 2002. Cholecystokinin and tryptic activity in the gut and body of developing Atlantic halibut (Hippoglossus hippoglossus): evidence for participation in the regulation of protein digestion. J. Fish Biol. 61: 973-986. Rojas-Garcia C.R. and Rennestad I. 2003a. Assimilation of dietary free amino acids, peptides and protein in post-larval Atlantic halibut (Hippoglossus hippoglossus). Mar. Biol. 142: 801-808. Rojas-Garcia C.R. and Rennestad I. 2003b. Comment on the article "Assimilation of dietary free amino acids, peptides and protein in post-larval Atlantic halibut (Hippoglossus hippoglossus)". Mar. Biol. 142: 799. Rsnnestad I. and Fyhn H.J. 1993. Metabolic aspects of free amino acids in developing marine fish eggs and larvae. Rev. Fish. Sci. 1: 239-259. Rannestad I. and Naas K.E. 1993. Oxygen consumption and ammonia excretion in larval Atlantic halibut (Hippoglossus hippoglossus) at first feeding: a first step towards an energetic model. In: Physiology and Biochemistry of Fish Larval Development. B.T. Walther and H.J. Fyhn (eds.). Univ. Bergen Press, Bergen Norway, pp. 279-284. Rannestad I., Fyhn H.J. and Gravningen K. 1992b. The importance of free amino acids to the energy metabolism of eggs and larvae of turbot (Scophthalmus maximus). Mar. Biol. 114: 517-525. Rsnnestad I., Groot E.P. and Fyhn H.J. 1993. Compartmental distribution of free amino acids and protein in developing yolk-sac larvae of Atlantic halibut (Hippoglossus hippoglossus). Mar. Biol. 116: 349-354. Rennestad I., Thorsen A. and Finn R.N. 1999. Fish larval nutrition: Recent advances in amino acid metabolism. Aquaculture 177: 201-216. Rennestad I., PerCz Dominguez R. and Tanaka M. 2000b. Ontogeny of digestive tract functionality in Japanese flounder, Paralichthys olivaceus, studied by in vivo
414
Physiological and ecological adaptations t o feeding in vertebrates
microinjection: pH and assimilation of free amino acids. Fish Physiol. Biochem. 22: 225-235. Rannestad I., Rojas-Garcia C.R. and Skadal J. 2000c. Retrograde peristalsis, a possible mechanism for filling the pyloric cecae? J. Fish Biol. 56: 216-218. Rannestad I., Finn R.N., Groot E.P. and Fyhn H.J. 1992a. Utilization of free amino acids related to energy metabolism of developing eggs and larvae of lemon sole Microstomus kitt reared in the laboratory. Mar. Ecol. Progr. Ser. 88: 195-205. Rannestad I., ConceiqZo L.E.C., AragZo C. and Dinis M.T. 2000a. Free amino acids are absorbed faster and assimilated more efficiently than protein in postlarval Senegal sole (Solea senegalensis). J. Nutr. 130: 2809-2812. Rannestad I., Conceiqso L.E.C., Araggo C. and Dinis M.T. 2001a. Assimilation and catabolism of dispensable and indispensable free amino acids in post-larval Senegal sole (Solea senegalensis). Cornp. Biochem. Physiol. 130C: 461466. Rannestad I., Rojas-Garcia C.R., Tonheim S.K. and Conceiqfo L.E.C. 2001b. In vivo studies of digestion and nutrient assimilation in marine fish larvae. Aquaculture 201: 161-175. Rennestad I., Koven W.M., Tandler A., Hare1 M. and Fyhn H.J. 1994. Energy metabolism during development of eggs and larvae of gilthead sea bream (Sparus aurata). Mar. Biol. 120: 187-196. Roth F.X., Gotterbarm G.G., Windisch W. and Kirchgessner M. 1999. Influence of dietary level of dispensable amino acids on nitrogen balance and whole-body protein turnover in growing pigs. J. Anim. Physiol. Animal Nutr. 81: 232-238. Rust M.B. 1995. Quantitative aspects of nutrient assimilation in six species of fish larvae. PhD, School of Fisheries, Univ. Washington, Seattle, WA, USA. Rust M.B., Hardy R.W. and Stickney R.R. 1993. A new method for force-feeding larval fish. Aquaculture 116: 341-352. Segner H. and Verreth J. 1995. Metabolic enzyme activities in larvae of the African catfish, Clarias gariepinus: Changes in relation to age and nutrition. Fish Physiol. Biochem. 14: 385-398. Segner H., Storch V., Reinecke M., Kloas W. and Hanke W. 1994. The development of functional digestive and metabolic organs in turbot, Scophthalmus Maximus. Mar Biol. 119: 471-486. Shields R.J. 2001. Larviculture of marine finfish in Europe. Aquaculture 200: 55-88. Simon 0 . 1989. Metabolism of protein and amino acids. In: Protein Metabolism in Farm Animals. H.-D. Bock, B.O. Eggum, A.G. Low, 0 . Simon, and T. Zebrowska (eds.). Oxford Univ. Press, Oxford, UK, pp. 273-366. Sivaloganathan B., Walford J., Ip Y.K. and Lam T.J. 1998. Free amino acids and energy metabolism in eggs and larvae of seabass, Lutes calcarifer. Mar. Biol. 131: 695-702. Smith S. 1947. Studies on the development of the rainbow trout (Salmo irideus). I. The heat production and nitrogenous excretion. J. Exp. Biol. 23: 357-377. Srivastava A. S., Kurokawa T. and Suzuki T. 2002. mRNA expression of pancreatic enzyme precursors and estimation of protein digestibility in first feeding larvae of the Japanese flounder, Paralichthys olivaceus. Comp. Biochem. Physiol. 132A: 629-635. Stryer L. 1995. Biochemistry. Freeman Publ., New York, NY. Tanaka H., Shibata K., Mori M. and Ogura M. 1995. Metabolism of essential amino acids in growing rats at graded levels of soybean protein isolate. J. Nutr. Sci. Vitaminol. 41: 433-443. Tanaka M. 1973. Studies on the structure and function of the digestive system of teleost larvae. PhD thesis, Dep. of Fish. Fac. of Agriculture, Kyoto University, Japan. Tanaka M., Kawai S., Seikai T. and Burke J.S. 1996. Development of the digestive organ system in Japanese flounder in relation to metamorphosis and settlement. Mar. Fresh. Behav. Physiol. 28: 19-31. Tejesen B.F. 2001. Aspects of nitrogen metabolism and excretion during ontogeny of the Atlantic halibut (Hippoglossus hippoglossus) and the African catfish (Clarias gariepinus). PhD thesis, Univ. Bergen, Bergen, Norway.
Digestion in marine fish larves
415
Terjesen B.F., Finn R.N., Norberg B. and Rennestad I. 2002. Kinetics and fates of ammonia, urea, and uric acid during oocyte maturation and ontogeny of Atlantic halibut (Hippoglossus hippoglossus). Comp. Biochem. Physiol. 131A: 443-455. Thorsen A. 1995. Oogenesis in marine bony fishes: physiological mechanisms of oocyte hydration and egg buoyancy. PhD, University of Bergen, Bergen, Norway. Tonheim S.K., Koven W.M. and Rennestad I. 2000. Enrichment of Arternia with free methionine. Aquaculture. 190: 223-235. Tonheim, S. K, M. Espe, A. J. Raae, M.J. Darias and I. Rennestad. 2004. In vivo incorporation of [U]-14C-amino acids : An alternative protein labelling procedure for use in examining larval digestive physiology. Aquaculture. 235: 553-567. Tytler P., Bell M.V. and Robinson J. 1993. The ontogeny of osmoregulation in marine fish, with particular reference to the effects of changes in salinity and temperature. In: Physiology and Biochemistry of Fish Larval Development. B.T. Walther and H.J. Fyhn (eds.). Univ. Bergen, Bergen Norway, pp. 249-258. Ueberschar B. 1988. Determination of the nutritional condition of individual marine fish larvae by analysing their proteolytic enzyme activities with a highly sensitive fluorescence technique. Meeresforsch. 32: 144-154. van der Meeren T. 1991. Algae as first food for cod larvae, Gadus morhua L.: filter feeding or ingestion by accident? J. Fish Biol. 39: 225-237. van der Meeren T., Fyhn H.J., Pickova J. and Hamre K. 2001. Biochemical composition of copepods: Seasonal variation in lagoon-reared zooplankton. In: Larvi 2001 C.I. Hendry, G. van Stappen, W. Willie, and P. Sorgeloos (eds.). Eur. Aquacult. Soc., Oostende, Belgium, Sp. Publ. no. 30, pp. 614-615. van Waarde A. 1983. Aerobic and anaerobic ammonia production by fish. Comp. Biochem. Physiol. 74B: 675-684. van Waarde A. 1988. Biochemistiy of non-protein nitrogenous compounds in fish including the use of amino acids for anaerobic energy production. Comp. Biochem. Physiol. 91B: 207-228. Verreth J., Polat A*, van Herwaarden H., ConceiqZo L. and Huisman E.A. 1995. A comparison of methods to study energy resource partitioning in early life stages of fish, with special reference to the African catfish, Clarias gariepinus Burchell. ICES Mar. Sci. Symp. 201: 57-63. Verreth J.A.J., Torreele E., Spazier E., van der Sluiszen A., Rombout J.H.W.M. and Booms R. 1992. The development of a functional digestive system in the African Catfish Clarias gariepinus (Burchell). J. World Aquacult. Soc. 23: 286-298. Walford J. and Lam T.J. 1993. Development of digestive tract and proteolytic enzyme activity in seabass (Lutes calcarfer). Aquaculture 109: 187-205. Walton M.J. 1985. Aspects of amino acid metabolism in teleost fish. In: Nutrition and Feeding in Fish. C.B. Cowey, A.M. Mackie, and J.G. Bell (eds.). Acad. Press, London, UK, pp. 47-67. Watanabe T. and Kiron V. 1994. Prospects in larval fish dietetics. Aquaculture 124: 223251. Watanabe Y. 1984. Morphological and functional changes in rectal epithelium cells of pond smelt during post embryonic development. Bull. Jpn. Soc. Sci. Fish. 50: 805-814. Werner R.G. and Blaxter J.H.S. 1980. Growth and survival of larval herring (Clupea haerengus) in relation to prey density. Can. J. Fish. Aqua. Sci. 37: 3545. Wilson R.P. 1989. Amino acids and proteins. In: Fish Nutrition. J.E. Halver (eds.). Acad. Press, New York, NY, pp. 111-151. Wright P.A. and Fyhn H.J. 2001. Nitrogen excretion and metabolism in early life stages. In: Fish Physiology, vol. XX. P.A. Wright and P.M. Anderson (eds.). Acad. Press, New York, NY, pp. 178-252. Wright P.A., Felskie A. and Anderson P.M. 1995. Induction of ornithine-urea cycle enzymes and nitrogen metabolism and excretion in rainbow trout (Oncorhynchusmykiss) during early life stages. J. Exp. Biol. 198: 127-135.
416
Physiological and ecological adaptations to feeding in vertebrates
Yamada S., Simpson K.L., Tanaka Y. and Katayama T. 1981. Plasma amino acid changes in rainbow trout (Salmo gairdneri) force-fed casein and a corresponding amino acid mixture. Bull. Jpn. Soc. Sci. Fish. 47: 1035-1040. Yancey P.H., Clark M.E., Hand S.C., Bowlus R.D. and Somero G.N. 1982. Living with water stress: evolution of osmolyte systems. Science 217: 1214-1222. Zambonino Infante J.L., Cahu C.L. and Peres A. 1997. Partial substitution of di- and tripeptides for native proteins in sea bass diet improves Dicentrarchus labrax larval development. J. Nutr. 127: 608-614. Zarate D.D., Love11 R.T. and Payne M. 1999. Effects of feeding frequency and rate of stomach evacuation on utilization of dietary free and protein-bound lysine for growth by channel catfish lctalurus punctatus. Aquacult. Nutr. 5: 17-22.
INDEX Atrophy 4, 2, 88, 191, 193, 210, 211, 222, A-linolenic acid 365, 374 Acid-base status of arterial plasma 286 Active transport 67, 1, 02, 114, 115, 116, 118, 121, 123, 124, 126, 127, 129, 135, 140, 243, 387
Actomyosin 118 Adaptive modulation hypothesis 116, 121, 137
Adductor mandibulae 7 Adductor muscles 5, 33 Adriatic sturgeon (Acipenser naccarii) 367, 369, 373, 379, 386
Agkistrodon piscivorus 283, 306 Alkaline tide 279, 280, 284, 285, 286, 287, 288, 295, 297, 299, 300-303
Autonomic nervous system 302, 325, 326, 328, 355, 357, 358, 359, 360
B B-lymphocytes 263, 267, 270 Bar-tailed godwit 210, 218, 220, 226, 227 Bats 17, 34, 238 Beak 6, 18, 21, 22, 27, 411 Bears 150, 173, 177, 190, 198, 230, 233, 238, 239, 244, 249, 250, 251, 254, 318
Bidirectional flow 14, 35, 37 Biting 15, 21, 31, 33, 74, 78, 91, 357, 375 Bivalve 91, 100, 106, 215, 216, 217, 218, 220, 221
AlkaIoids 27, 257 Allies scolopacidae 218 Alligator mississippiensis 32, 33, 184, 286, 289, 291, 295, 299, 300, 311, 319
Allometry 81, 92, 93, 206, 218, 320 Alsophis portoricensis 312, 322 Ambystoma 33, 35, 36, 39, 341, 359 American robins (Turdus migratorius) 105 Ammotretis rostrata 330, 355 Amylase 96, 239, 250 Antinutrients 257 Antioxidants 245 Apoptosis 192, 193, 194, 197, 198, 238, 247, 252, 261
Apparent digestibility 47 Apfenodytes forsteri 176, 197 Arachidonic acid 365, 380, 385 Arctic ground squirrels (Spermophilus parryii) 240, 253 Assimilation efficiency 56, 71, 73, 76, 147, 149, 152, 309, 397, 406
Blackcaps 104, 105, 110, 179, 180, 181, 183, 197, 226
Bladder bile 239 Boa constrictor 290, 293, 299, 303, 322, 324 Bolus 1, 10, 12, 19, 20, 21, 22, 23, 24, 25, 28, 30, 344, 345, 346, 409
Bovids 17 Bramble-wake model 11 Branchiomeric musculature 3 Broad-tailed hummingbirds (Selasphorus platycercus) 74, 100 Brush-border membrane 110, 114, 115, 119, 125, 131, 133, 134, 137, 138, 232, 235, 254
Buccal cavity 3, 4, 10 Bufo 282, 287, 288, 289, 290, 295, 296, 297, 299, 300, 301, 303, 323, 324, 348, 355
Bufo marinus 290, 296, 299, 300, 303, 323, 324, 355
Burmese pythons 199, 302, 312, 322, 323 C
Atlantic salmon 354, 372, 374, 375, 376, 377f 3831 384f 386, 387, 400,
229, 231, 234, 235, 236, 238, 240, 242, 243, 249
409t
410
Caecilians 17, 26, 37 Caiman 183, 185, 199, 316, 320, 338, 356
418
Physiological and ecological adaptations t o feeding i n vertebrates
Capillarity 17 Capture 1, 5, 8, 9, 11, 12, 14, 15, 16, 17, 19,
Compensatory feeding 73, 74, 75, 76, 78,
21, 24, 32, 33, 34, 35, 36, 37, 38, 39, 40, 96, 179, 190, 294, 308, 326
Compensatory suction 13 Concentration-dependent mixing models
Capture/subjugation 1, 11, 12 Carbohydrate 45, 49, 59, 72, 82, 84, 104, 108, 121, 131, 140, 311,
113, 123, 132, 151, 345,
114, 124, 133, 152, 402,
115, 125, 134, 154, 404,
116, 126, 135, 155, 405,
117, 127, 136, 156, 413,
118, 129, 137, 196, 424,
119, 130, 139, 276,
Cardiac performance 363, 372, 383, 384 Cardiolivin 380 Carrier-mediated absorption 67 Carrier-mediated transport 113, 114, 117, 120, 122, 132, 380
Catfishes 311 Ceca 50, 52, 53, 64, 83, 102, 203, 204, 205, 211, 222, 235, 244, 255, 257, 265, 266, 267, 268, 334, 335, 346, 34% 353; 359, 414
Cell division 199, 261 Cell loss 191, 193, 231 Cell membranes 117, 122, 365 Cell proliferation 104, 182, 191, 192, 193, 194, 198,c 199, 231, 236, 237, 247, 250, 252, 273
80, 86
147, 152
Cost-benefit curve 70
Coturnix japonica 110, 111, 169, 199, 355 Cranial kinesis 4, 16, 31, 33, 34, 37 Crocodiles 4, 15, 17, 20, 21, 24, 27, 32, 38, 183, 184, 190, 286, 311, 321, 333, 348
Crocodylus 32, 292, 293, 299, 301, 311, 321, 338, 356
Crop 17, 21, 26, 27, 28, 40, 89, 60, 188, 203, 221, 226, 262, 263, 269, 310, 317, 319, 331, 341, 346, 412
Crossflow filtration 24, 38
Crotalus cerastes 306, 323 Crotalus durissus 189r 299, 300, 318, 319 Crotalus horridus 200, 303, 313, 324 Crotalus viridis 189, 322 Ctenosaura 39, 316, 322 Cuticle 29 Cyclooxygenase 381 Cytokines 247, 262, 270, 271, 272, 276
Cellular dystrophy 182 Cellular hypotrophy 182
Defenses 45, 250, 255, 259, 260, 262, 268,
Cephaloscyllum ventriosum 312 Ceratophrys cranwallii 189, 316, 322
Defensins 262, 277 Deglutition 24 Dendritic cells 263, 269 Desmognath 23, 26, 39 Diatoms 381 Diet hardness 209, 214, 218 Diet microflora 256 Diet preferences 67 Diet quality 57, 58, 85, 87, 89, 90, 111, 174,
Chameleons 16, 34, 40 Charadriiformes 202, 212, 218 Chemical digestion 1, 22, 28 Chemical reactors 45, 58, 59, 60-62, 6567, 73-75, 80-85
Chemoreception 17
Chen rossii 167 Chewing 19, 21, 22, 27, 28, 89, 91, 100 Claudins 116, 140 Cold exposure 74, 80 Colon 46, 50, 52, 53, 54, 56, 83, 136, 203, 235, 253, 258, 259, 265, 266, 272, 273, 275, 283, 330, 335, 341, 358, 360, 364, 366, 382
Comminution 6, 22, 51
271, 272, 424
209, 211, 226, 381
Diffusive absorption 67 Digestibility 43, 44, 46, 47, 49, 51, 54, 55, 56, 63, 70, 82, 210, 211, 248, 277, 312, 361, 391, 393, 396, 397, 401, 402, 414
Digestion coefficient 46, 47, 55 Digestive constraints 66, 67, 72, 78, 80, 82, 85, 87, 88, 89, 91, 93, 95, 97, 99, 101, 103, 105, 107, 108, 109, 111, 424
Index Digestive efficiency 31, 43, 44, 45, 46, 47, 49, 50, 51, 52, 53, 54, 55, 56, 57, 59, 71, 72, 93, 94, 97, 105, 106, 108, 182, 204, 209, 211, 225, 227, 256, 321, 391, 396, 421, 424 Digestive enzymes 28, 50, 51, 93, 96, 104, 105, 163, 203, 239, 248, 270, 312, 315, 316, 390, 391, 411, 412 Digestive function 59, 61, 62, 63, 65, 67, 69, 71, 73, 75, 77, 79, 81, 83, 85, 229, 239, 248, 250, 314, 391, 409, 424 Digestive spare capacity 80 Dinoflagellates 381 Discrimination 136, 145, 146, 159, 164, 165, 249 Djungarian hamsters 92 Docosahexanoic acid 365 Dormice (Eliomys quercinus) 238 Dover sole (Solea solea) 378 Ducks 105, 109, 172, 198, 225, 226, 227, 228, 348
E Eicosanoids 262, 364, 365, 380, 381, 387 Elaphe obsolete 315 Elasmobranchs 13, 22, 33, 41, 330, 349, 350, 362 Electrolytes 73, 137, 243, 249 Energy assimilation 71, 72, 74, 75, 76, 79, 205 Energy substrates 365, 389, 404 Enteric nervous system 325, 326, 327, 328, 332, 335, 353, 355, 358, 359, 360 Enteric neurons 240, 327, 328, 329, 330, 341, 344, 354, .-61 Enterochromaffin cells 349 Enterocyte 104, 109, 111, 117, 122, 130, 132, 180, 184, 185, 186, 187, 188, 191, 192, 231, 232, 235, 236, 237, 238, 241, 242, 243, 245, 246, 247, 249, 252, 258, 260, 261, 262, 263, 269, 270, 271, 283 Enterocyte life span 104 Enterocyte turnover 104, 192 Environmental stressors 363, 378 Eosinophils 262, 263 Epithelial barrier 16, 122, 271
419 Epithelial configuration 184, 186 Epithelium 9, 134, 136, 138, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 192, 193, 194, 199, 231, 233, 236, 238, 243, 244, 252, 255, 256, 258, 259, 260, 261, 262, 263, 268, 269, 270, 271, 273, 275, 276, 328, 354, 415 Equilibrium fractionation 145 Esophageal papillae 13, 40 Esophagus 3, 10, 11, 19, 21, 23, 24, 25, 28, 203, 260, 261, 267, 326, 327, 330, 343, 344, 346, 348, 351, 361, 391 Eulamprus quoyii 314 Eurasian curlew 218, 220 European eel (Anguilla anguilla) 367, 386 European starlings (Sturnis vulgaris) 105 Eutamias amoenus 54 Everted intestinal sleeve method 127
F Fasting tolerance 176 Fat stores 176, 229, 233, 248 Fecal dry matter 46 Feeding bout 10, 11, 12 Feeding cycle 11 Feeding ecology 36, 57, 85, 110, 171, 189, 190, 195, 201, 217, 424, 310 Feeding mode 10, 27, 36 Feeding stages 1, 10, 11, 12, 402, 405 Filter feeding 12, 415 Fish larvae 363, 378, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 401, 403, 404, 405, 406, 408, 409, 410, 411, 412, 414, 415, 425 Fish-spearing 12 Fluctuations in food availability 176 Food webs 142, 172, 365, 366, 376, 381, 383, 386 Foraging strategy 10 Fractionation 144, 145, 146, 162, 164, 166, 168, 172, 173, 174 Frogs 15, 16, 17, 20, 27, 32, 37, 40, 189, 190, 286, 302, 354, 358 Fructose transport 115, 135, 137, 139, 140 Frugivory 22, 66, 69, 72, 84, 127, 228, 272
420
Physiological and ecological adaptations to feeding in vertebrates
Gadus morhua 299, 301, 309, 310, 323, 336, 338, 348, 353, 355, 356, 357, 358, 359, 360, 385, 387, 408, 410, 412, 413, 415 Galliformes 202, 204, 205, 206, 207, 209, 212, 213, 225 Gape cycle 11, 12 Gastric acid secretion 283, 284, 285, 286, 288, 299, 300, 302, 303, 306, 338, 347, 348, 349, 350, 353, 356, 358, 359, 361 Gastric mill 1, 29, 30 Gastrin 325, 334, 335, 336, 337, 345, 349, 350, 351, 352, 353, 354, 355, 356, 357, 358, 359, 361 Gastrointestinal blood flow 292, 293, 299, 301 Gastrointestinal defenses 259 Gastroliths 27, 39 Gavials 15 Ghrelin 325, 335, 337, 338, 339, 352, 355, 357, 358, 359, 362 Gilthead seabream (Sparus aurata) 378, 385, 4 12 Giraffes 17 Gizzard 1, 27, 28, 29, 30, 93, 94, 100, 104, 105, 109, 111, 177, 181, 182, 196, 199,
Glossophaga longirostris 96 GLUT2 102, 103, 110, 114, 115, 119, 121, 125, 131, 132, 133, 134, 135, 137, 138 GLUT5 115, 119, 135, 137, 139, 140 Glutathione 245, 250, 252 Golden hamsters (Mesocricetus auratus) 240 Granivores 29, 91, 100, 107, 207, 257, 269 Granulocytes 263 Great knots 203, 210, 224 Grinding plates 29 Grit 27, 29, 33, 117, 123, 229, 240, 250, 255, 270, 271, 339, 365 Ground squirrels 172, 230, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 250, 251, 252, 253, 254, 320 Grouse 57, 104, 111, 204, 206, 207, 211,
Gut motility 63, 262, 315, 325, 326, 328, 329, 331, 333, 334, 335, 337, 339, 341, 342, 343, 344, 345, 347, 351, 352, 353, 355, 357, 359, 360, 425 Gut volume 61, 72, 75, 99
327, 338, 349, 361,
Gut-associated lymphoid tissue (GALT) 255, 256 Gymnophione amphibians 17
Haematopodidae 218 Halibut (Hippoglossus hippoglossus) 378, 408, 411, 412, 413, 414, 415 Hamsters (Cricetus cricetus) 238 Handling stress 378, 385 Harpagifer antarcticus 310, 319 Heart rate 293, 294, 295, 296, 297, 300, 303, 324, 374 Helicobacter 261, 275 Herbivores 7, 28, 29, 30, 46, 50, 53, 57, 83, 91, 109, 164, 201, 226, 256, 257, 267, 271, 326 Herring (Clupea haringus) 378 Heterodonty 22 Hexose absorption 59, 69, 70, 79 Hibernation 92, 176, 192, 196, 229, 231, 233, 234, 235, 236, 237, 238, 240, 241, 242, 243, 244, 245, 246, 248, 249, 250, 251, 252, 253, 254, 362, 368 Hoatzin 28, 34 Homology 11, 27 Horseshoe crabs 216 Hummingbirds 17, 67, 69, 70, 71, 72, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 91, 100, 101, 110, 122, 127, 128, 129, 130, 136, 137, 139, 140, 141 Hydrostatic elongation of the tongue 10, 15 Hyobranchial apparatus 1, 5, 9, 12, 13, 14, 15, 16 Hyoid apparatus 7, 8, 9, 13, 34, 39, 41 Hyolingual apparatus 5, 9, 20 Hyolingual transport 10, 20 Hyperplasia 182, 192, 252
Index Hypertrophy 182, 192, 211, 225, 271, 280, 28 1 Hypobranchial muscles 7 Hypoxia 299, 301, 363, 367, 368, 369, 370, 371, 372, 374, 378, 379, 380, 382, 384, 386, 387
Iguana 16, 21, 368 Immune system 233, 244, 2 56, 257, 259, 260, 261, 262, 263, 265, 267, 268, 269, 270, 271, 272, 273, 275, 277 Inertial suction 9, 13 Inertial transport 20 Ingestion 1, 11, 12, 14, 15, 16, 17, 19, 20, 21, 28, 41, 49, 50, 60, 61, 63, 66, 72, 75, 81, 82, 85, 118, 128, 138, 154, 165, 233, 241, 245, 248, 249, 279, 280, 283, 284, 286, 293, 294, 300, 306, 308, 309, 310, 314, 318, 390, 396, 406, 415 Insectivory 22 Intake response 74, 77, 78, 80, 85, 111 Interstitial cells of Cajal 325, 328, 333, 354, 357, 359, 360, 361, 362 Intestinal absorption 67, 72, 113, 118, 121, 132, 135, 136, 139, 253, 272, 274, 315, 406, 412, 424 Intestinal blood flow 292, 293, 299, 301 Intestinal crypts 182, 184, 191, 192, 193 Intestinal epithelia 114, 115, 117, 122, 126, 134, 135, 137, 138, 196, 231, 233, 235, 236, 243, 244, 249, 250, 252, 253, 258, 260, 262, 269, 270, 276 Intraepithelial lymphocytes 244, 263, 267, 268, 269, 271, 272, 277 Intraoral transport 19, 20, 23 Isolated heart 373, 374, 380 Isotopic composition 142, 143, 144, 145, 146, 147, 148, 150, 151, 152, 153, 155, 156, 157, 159, 160, 161, 162, 163, 164, 166, 167, 168, 169, 171, 173, 174 Isotopic routing 146, 149, 151, 153, 154, 155 Isotopic signature 142, 160
Japanese flounder (Paralichfhys olivaceus) 378, 387
421 Japanese quail 104, 110, 111, 159, 169, 179, 181, 182, 198, 199, 225, 227, 228, 355 Jaw muscles 30 Jaw prehension 12, 14, 15, 16, 17
K Kidneys 189, 195, 285, 288 Kinetic fractionation 145, 173 Kinetic skulls 16 Kingfishers 21 Kinixys spekii 311, 321 Knots (Calidris canutus) 100, 210
L Lactation 88, 92, 98, 99, 105, 106, 109, 116, 191, 254 Large intestine 56, 94, 138, 203, 254, 255, 263, 265, 266, 267, 268, 275, 276, 332, 340, 354, 355, 357, 362 Lepidosaurian reptiles 16 Linear mixing models 146, 147, 149 Lingual ingestion 17 Lingual prehension 15, 16 Linoleic acid 364, 365, 375, 377, 379, 383 Lipid storage 214 Lipoxygenase 381 Lips 17, 20, 23, 146, 147, 148, 150, 173, 174, 296, 302, 311, 323 Long-distance migratory species 52
Lumbricus festivus 166, 174 Lymphocytes 244, 252, 255, 256, 257, 259, 263, 264, 265, 267, 268, 269, 270, 271, 272, 276, 277 Lysozyme 262
M M cells 192, 263, 269, 274, 275, 328, 415 Macrostomatan snakes 23 Marginal teeth 4, 15, 22, 23, 26 Marmofa marmofa 52, 54, 57, 192, 230, 233, 234, 235, 237, 244, 252, 321 Masseter 7 Mast cells 262, 263 Mastication 1, 6, 7, 22, 24, 30, 32, 34, 39, 43, 50, 51, 57, 308, 310
422
Physiological and ecological adaptations t o feeding i n vertebrates
Masticophis flagellum 306, 323 Maximum metabolizable energy 92, 93 Mean retention time 43, 50, 57, 68, 95, 97, 106 Menidia menidia 309, 310, 316 Metabolic fecal dry matter 46 Metabolic fecal losses 48 Metabolic fecal nitrogen 46, 47, 163 Metabolic rate 44, 75, 81, 84, 85, 88, 93, 109, 111, 127, 139, 191, 200, 224, 225, 226, 230, 246, 279, 280, 281, 288, 289, 291, 292, 293, 298, 300, 301, 306, 307, 308, 312, 313, 315, 317, 318, 319, 320, 321, 323, 324, 363, 368, 370, 371, 372, 374, 379, 380, 381, 382, 386 Metabolizability 43, 44, 47, 48, 52 Microbes 49, 50, 58, 198, 255, 256, 257, 261, 263, 268, 269, 276, 424 Microflora 244, 255, 256, 257, 258, 259, 260, 261, 263, 264, 268, 269, 271, 272, 273, 274, 275, 276, 277 Microphagous fishes 26 Micropterus salmoides 310, 317, 319 Microtus townsendii 54 Migrating motor complexes 339, 343 Mixing models 141, 143, 146, 147, 149, 150, 152, 153, 154, 157, 172, 173, 174 Mollusks 23, 201, 218, 220, 221 Molt 21, 209, 210, 214, 376, 377, 383, 385, 386, 387 Monotremes 14, 22, 34, 348 Mucosal defense system 262 , Mucosal epithelium 134, 179, 180, 181, 182, 183, 184, 185, 187, 189, 190, 192, 194, 328 Mus musculus 98, 105, 109 Muscardinus avellanarius 239 Muscular hydrostats 9 Myrmecophages 17 Mysticete whales 12
Net assimilation rate 71 Neuropeptides 293, 301, 334, 340, 353, 356, 358, 359 Neurotransmitters 325, 330, 334, 339, 341, 344, 346, 349, 357, 359 Nonadrenergic-noncholinergic (NANC) 293 Nonmediated passive uptake 114 Nutrient absorption 50, 55, 70, 83, 84, 93, 101, 102, 103, 104, 106, 108, 109, 110, 112, 119, 130, 132, 133, 134, 136, 137, 203, 232, 242, 243, 249, 258, 260, 262, 410, 421 Nutrient assimilation 25, 59, 60, 110, 225, 226, 396, 414 Nutrient uptake 54, 55, '60, 86, 90, 94, 109, 112, 113, 121, 131, 140
N Natrix maura 317, 321 Nectarivory 17, 66, 74, 80, 96, 110, 127, 137, 206, 272 Neonate mammals 14
Paracellular permeability 113, 125, 130, 132, 134, 138 Parasites 256, 257 Parasympathetic 293, 325, 328, 329, 330, 353
0 Occlusion 22 Okapi 17 Opossums 20 Optimal digestion 45, 46, 69, 71, 72, 83, 85 Oral cavity 12, 19, 21, 24 Osmoregulation 80, 134, 136, 363, 376, 377, 382, 394, 415 Owls 22, 361 Oxidative stress 245, 246, 250, 251, 252 Oxygen uptake 280, 281, 282, 293, 298, 303, 307, 312, 317, 368, 369, 370 Oystercatchers 218
P Palatal teeth 20, 28 Palate 4, 20, 22, 23, 24, 25, 30 Pancreas 94, 203, 222, 238, 239, 251, 253, 328, 347, 353, 354, 390, 405, 412 Pancreatic digestive enzymes 248 Paneth cells 262 Panurus biarmicus 177, 199 Paracellular flux 113, 117, 118, 119, 121, 123, 125, 132, 133
Index Parietal cell 284, 285, 347, 349, 358 Parrots 19, 21, 22, 206 Passage rates 1, 30, 51, 391 Passerines 72, 79, 80, 81, 204, 207, 208, 209, 210, 211, 213 Passive absorption 69, 79, 82, 102, 103, 108, 133, 135, 123, 124, 125, 126, 127, 129, 134, 135, 136, 139 Passive permeability 61, 126 Pattern generators 11 Pepsin / pepsinogen 351 Peristalsis 23, 24, 25, 256, 261, 308, 325, 334, 337, 344, 345, 346, 356, 414 Peyer patches 257, 262, 263, 267, 269, 270 Pharmacokinetics 129, 139, 140 Pharyngeal jaws 8, 19, 23, 24, 26, 36 Pharyngeal musculature 24 Pharyngeal skeleton 3, 4, 7, 8, 9, 12, 13 Pharyngognathy 8, 26 Pharynx 1, 3, 4, 5, 7, 8, 13, 14, 15, 19, 20, 22, 23, 24, 30, 39, 41, 267, 275 Phascolarctos cinereus 51, 57, 274 Phenotypic flexibility 89, 90, 93, 94, 111, 176, 177, 190, 194, 195, 199, 201, 223, 225, 227, 228, 231, 250, 314, 322, 324 Phloretin 133 Phloridzin 119, 122, 133, 243 Phospholipase 381, 383 Phytohemagglutinins 257, 271 Pinnipeds 14, 19 Piscivory 22 Pleuronectes platessa 316, 321 Plovers Charadriidae 218
Podarcis s. sicula 345, 358 Polyunsaturated Fatty acids (PUFA) 364 Postprandial response 185, 188, 199, 307, 312 Postprandial stimulation 308 Prairie voles (Microtus ochrogaster) 99 Prey capture 9, 16, 32, 33, 34, 35, 36, 37, 38, 39, 40 Prey manipulation 8, 34 Prey size 27, 28, 35, 317 Primary producers 161, 365, 381 Primates 12, 17, 21
423 Properistalsis 24 Protein metabolism 170, 214, 311, 414 Protein reallocation 222 Protein synthesis 165, 173, 191, 237, 252, 254, 279, 280, 283, 303, 306, 312, 314, 319, 398, 402, 406, 409, 411 Proton leak 380 Proventriculus 28, 29, 203, 215, 337, 339, 353, 355 Pseudocheirus peregrinus 51, 274 Pseudostratified epithelium 184, 185, 186, 190 Puncture-crushing 21, 22 Pylorus 27 Python molurus 189, 198, 199, 200, 294, 296, 297, 302, 320, 322, 323, 324, 337, 339, 340, 341 Python regius 189, 303
R
Raja sp. 336 Ram feeders 13 Rana catesbeiana 285, 286, 287, 300, 320, 336, 338, 351, 352, 358, 359, 361 Rana pipiens 31, 342 Raptors 17, 22, 51, 56, 204 Rattlesnakes 12, 28, 189, 300, 310, 319, 324 Regurgitation 28, 315 Reserve capacity 89 Retention times 50, 57, 71, 94, 133 Rhamphotheca 17, 22 Rhinoceroses 17 Rhombosolea tapirina 330, 355 Rhythmic oscillating complexes 343, 361 Rodents 17, 21, 54, 56, 57, 82, 108, 173, 190, 230, 245, 269, 270 57, 218, 221, 277 Rumenoreticulum 100 Ruminant herbivores 91 S Safety margin 55, 58, 78, 89, 93, 112 Salamanders 13, 14, 15, 16, 23, 26, 32, 35, 36, 39, 41 Salinity challenge 378
424
Physiological and ecological adaptations to feeding in vertebrates
Saliva 6, 21, 22, 28, 30, 323, 347 Salmo salar 336, 354, 372, 374, 377, 383, 384, 386, 387, 409, 410 Sandpipers 218 Sauropsids 176, 177 Scleroglossan lizards 17 Scolecophidian snakes 20 SDA 84, 110, 137, 174, 185, 186, 279, 281, 282, 283, 289, 292, 293, 295, 299, 300, 305, 306, 307, 308, 309, 310, 311, 312, 313, 314, 315, 316, 317, 318, 319, 321, 323, 409 Seabass (Dicentrarchus labrax) 378 SGLTl 114, 115, 119, 122, 123, 124, 125, 129, 131, 133, 242, 243 SGLT2 114, 137 Shorebirds 20, 38, 92, 93, 111, 202, 203, 204, 217, 218, 224, 225, 227, 228 Shrikes 21 Sieving 24, 130 Simple diffusive absorption 67 Sodium fluorescein 132, 13 Soft palate 24 Solvent drag 113, 117, 118, 124, 125, 130, 133, 134, 139 Spare capacity 78, 80, 87, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 101, 102, 103, 111, 126 Specific dynamic action 185, 200, 279, 281, 299, 301, 302, 303, 305, 306, 307, 308, 319, 320, 321, 322, 323, 324, 424 Spermophilus columbianus 54 Spiders 215, 220 Spinal autonomic system 330 Splanchnocranium 7 Squalus acanfhias 41, 336, 338, 348, 352, 356, 360 Stable isotopes 141, 142, 143, 144, 145, 146, 156, 157, 170, 172, 173 Standard metabolic rate (SMR) 368 Stomach mass 214, 215, 216, 218 Stratified epithelium 184, 185, 260 Stress tolerance 378, 386, 387 Sfurnus vulgaris 111, 138, 178, 195, 225, 228 Sucrase 54, 61, 68, 69, 75, 76, 84, 96, 98, 100, 240, 241
Sucrose 54, 59, 61, 68, 69, 70, 71, 74, 75, 76, 77, 78, 79, 80, 81, 84, 96, 100, 101, 115, 126, 156, 272 Sucrose hydrolysis 59, 61, 68, 69, 70, 75, 78, 79 Suction feeding 5, 9, 12, 13, 14, 15, 27, 31, 32, 34, 37, 41 Surface tension transport 20 Suspension feeding 3, 7, 12, 13, 14, 22, 24, 26, 35, 37, 38, 42, 190 Sustained aerobic exercise 374, 382 Swallowing 1, 5, 9, 11, 19, 20, 21, 22, 23, 24, 25, 26, 34, 37, 40, 51, 344 Swimming performance 299, 375, 383, 384, 386 Sylvia atricapilla 179, 180, 197, 226 Sylvia borin 52, 56, 57, 179, 195, 196, 197, 224, 225, 226 Sympathetic nervous system 293, 294, 325, 328, 329, 330, 332, 353, 354, 357 T T-lymphocytes 259, 263, 269, 270, 271 Teeth 3, 4, 6, 8, 15, 18, 20, 21, 22, 23, 26, 28, 30, 32, 33, 51, 145 Teleosts 8, 23, 24, 36, 336, 343, 349, 350, 4 12 Temporalis 7 Tenrecs 20 Thamnophis sirtalis 189, 320 Throughput time 61, 68, 70, 76 Tight-junction 116, 117, 118, 121, 122, 123, 125, 130, 134 Tilapia 37, 38, 341, 362 Tinca tinca 330 Tongue 1, 5, 9, 10, 11, 14, 15, 16, 17, 19, 20, 21, 22, 24, 25, 30, 32, 33, 35, 36, 37, 38, 39, 40, 84, 250, 324 Tooth reduction 6 Torpor 75, 101, 229, 230, 233, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 253, 368 Transepithelial impedances 118 Transitional epithelium 188, 189, 190, 193, 194 Tree shrews 20 True digestibility 47
Index Trunk 17, 20, 23, 35, 39, 294, 329, 357 Tuatara 16, 28, 34 Tupinambis merianae 317 Turdus migratorius 54, 105 Turnover time 52, 104, 106, 108, 191, 237 Turtles 6, 10, 13, 14, 17, 19, 21, 22, 24, 27, 31, 33, 34, 37, 39, 40, 313, 315, 323, 333
u Unidirectional flow 14 Urea 163, 19-69, 172, 177, 244, 251, 253, 254, 405, 415, 416 Urinary losses 48 Ursus americanus 238, 251, 253
v Vagus 329, 330, 335, 338, 343, 344, 348, 350, 353, 355, 360 Varanus 39, 282, 287, 293, 294, 296, 297, 298, 300, 301, 302, 311, 316, 317, 318, 322, 323, 324 Varanus albigularis 296, 302, 311, 323 Varanus exanthematicus 293, 294, 297, 298, 300, 301 Venom 12, 28, 40, 312, 322, 324 Ventilation 36, 82, 286, 287, 289, 300, 303, 317, 369, 370 Ventriculus 28, 29, 203, 215, 337, 339, 353, 355
425
Villus 55, 139, 180, 191, 192, 199, 229, 231, 235, 236, 237, 238, 249, 253 Visceral arches 3, 7 W Waders 201, 202, 204, 209, 210, 211, 213, 214, 215, 217, 218, 223 Walleye (Stizostedion vitreum) 378 Walruses 14 Water absorption 72, 85, 117, 122, 124 Water-balance 72 Waterfowl 19, 202, 204, 208, 209, 210, 211, 213, 214, 228 White-throated sparrows 93, 94, 95 Woodpeckers 17 X
Xenopus laevis 131, 190, 338, 341-343, 356, 357, 358, 359, 360
354,
Y
Yellow tail (Seriola quinqueradiata) 378 Yellow-eyed Juncos (Junco phaeonotus) 106 Yellow-rumped warblers 82, 96, 97, 110, 127, 134, 165
z Zonotrichia albicollis 34, 93