Fatty Acids - PDF Free Download (2024)

Fatty Acids Physiological and Behavioral Functions Edited by

David I. Mostofsky Shlomo Yehuda Norman Salem Jr.




AND 9 HEALTH Adrianne Bendich, Series Editor

Fatty Acids: Physiological and Behavioral Functions, edited by David I. Mostofsky, Shlomo Yehuda, and Norman Salem Jr., 2001 Nutrition and Health in Developing Countries, edited by Richard D. Semba and Martin W. Bloem, 2001 Preventive Nutrition: The Comprehensive Guide for Health Professionals, Second Edition, edited by Adrianne Bendich and Richard J. Deckelbaum, 2001 Nutritional Health: Strategies for Disease Prevention, edited by Ted Wilson and Norman J. Temple, 2001 Clinical Nutrition of the Essential Trace Elements and Minerals: The Guide for Health Professionals, edited by John D. Bogden and Leslie M. Klevey, 2000 Primary and Secondary Preventive Nutrition, edited by Adrianne Bendich and Richard J. Deckelbaum, 2000 The Management of Eating Disorders and Obesity, edited by David J. Goldstein, 1999 Vitamin D: Physiology, Molecular Biology, and Clinical Applications, edited by Michael F. Holick, 1999 Preventive Nutrition: The Comprehensive Guide for Health Professionals, edited by Adrianne Bendich and Richard J. Deckelbaum, 1997


DAVID I. MOSTOFSKY, PhD Department of Psychology, Boston University, Boston, MA

SHLOMO YEHUDA, PhD Psychopharmacology Laboratory, Bar Ilan University, Ramat Gan, Israel

NORMAN SALEM JR., PhD National Institute on Alcohol Abuse and Alcoholism, Rockville, MD



© 2001 Humana Press Inc. 999 Riverview Drive, Suite 208 Totowa, New Jersey 07512 www.humanapress.com All rights reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, microfilming, recording, or otherwise without written permission from the Publisher. All papers, comments, opinions, conclusions, or recommendations are those of the author(s), and do not necessarily reflect the views of the publisher.

Cover design by Patricia F. Cleary. Production Editor: Mark J. Breaugh. For additional copies, pricing for bulk purchases, and/or information about other Humana titles, contact Humana at the above address or at any of the following numbers: Tel.: 973-256-1699; Fax: 973-256-8341; E-mail: [emailprotected] or visit our website at http://humanapress.com This publication is printed on acid-free paper. ' ANSI Z39.48-1984 (American National Standards Institute) Permanence of Paper for Printed Library Materials. Photocopy Authorization Policy: Authorization to photocopy items for internal or personal use, or the internal or personal use of specific clients, is granted by Humana Press Inc., provided that the base fee of US $10.00 per copy, plus US $00.25 per page, is paid directly to the Copyright Clearance Center at 222 Rosewood Drive, Danvers, MA 01923. For those organizations that have been granted a photocopy license from the CCC, a separate system of payment has been arranged and is acceptable to Humana Press Inc. The fee code for users of the Transactional Reporting Service is: [0-89603-942-0/01 $10.00 + $00.25]. Printed in the United States of America. 10 9 8 7 6 5 4 3 2 1 Library of Congress Cataloging-in-Publication Data Fatty acids : physiological and behavioral functions / edited by David I. Mostofsky, Shlomo Yehuda, and Norman Salem. p. cm. -- (Nutrition and health) Includes bibliographical references and index. ISBN 0-89603-942-0 (alk. paper) 1. Essential fatty acids in human nutrition 2. Fatty acids--Metabolism. 3. Essential fatty acids--Psychological aspects. I. Mostofsky, David I. II. Yehuda, Shlomo. III. Salem, Norman. IV. Nutrition and health (Totowa, N.J.) QP752.E84 F38 2001 612.3'97--dc21 00-067293

SERIES EDITOR PAGE The Nutrition and Health series of books have, as an overriding mission, to provide health professionals with texts that are considered essential because each includes: (1) a synthesis of the state of the science, (2) timely, in-depth reviews by the leading researchers in their respective fields, (3) extensive, up-to-date fully annotated reference lists, (4) a detailed index, (5) relevant tables and figures, (6) identification of paradigm shifts and the consequences, (7) virtually no overlap of information between chapters, but targeted, inter-chapter referrals, (8) suggestions of areas for future research, and (9) balanced, data-driven answers to patient/health professionals questions that are based upon the totality of evidence rather than the findings of any single study. The series volumes are not the outcome of a symposium. Rather, each editor has the potential to examine a chosen area with a broad perspective, both in subject matter as well as in the choice of chapter authors. The international perspective, especially with regard to public health initiatives, is emphasized where appropriate. The editors, whose trainings are both research- and practice-oriented, have the opportunity to develop a primary objective for their book, define the scope and focus, and then invite the leading authorities from around the world to be part of their initiative. The authors are encouraged to provide an overview of the field, discuss their own research, and relate the research findings to potential human health consequences. Because each book is developed de novo, the chapters are coordinated such that the resulting volume imparts greater knowledge than the sum of the information contained in the individual chapters. Fatty Acids: Physiological and Behavioral Functions edited by David I. Mostofsky, Shlomo Yehuda, and Norman Salem, clearly exemplifies the goals of the Nutrition and Health series. In fact, this volume is surely ahead of the curve with regard to awareness of the importance of fatty acids in virtually every aspect of human health and disease prevention. Two fatty acids are considered essential nutrients for humans: linoleic and linolenic acids. Thus, there is no question about the importance of dietary intake of adequate levels of these essential nutrients. However, the story of the changes in our food supply and the consequences of consumption by us, as well as our farm animals, of foods that no longer have the same balance of these essential fatty acids had yet to be captured in one authoritative, up-to-date volume until now. Fatty Acids: Physiological and Behavioral Functions, edited by David I. Mostofsky, Shlomo Yehuda, and Norman Salem, has set the benchmark for providing the most critical data on fatty acids in the most accessible volume published to date. Understanding the metabolism of fatty acids and their roles in human health is certainly not simple and the terms used can often seem daunting; however, the editors and authors have focused on assisting those who are unfamiliar with this field in understanding the critical issues and important new research findings that can impact their fields of interest. Moreover, the two Forewords by the well-acknowledged leaders in the field, Drs. Ralph Holman and William Lands provide the historic perspective as well as a clear overview of the critical importance of fatty acid balance to human health. Emphasis is placed on the physiological role of the two essential fatty acids, linoleic acid (n-6) and linolenic acid (n-3) and their metabolites—arachidonic acid (n-6), docosahexaenoic acid (n-3), and eicosapentaenoic acid (n-3). The uninitiated reader is v


Series Editor Page

clearly guided through the, at first, complex terminology that often makes fatty acid biology seem foreboding. By making the chapters accessible to all readers, the editors have worked to broaden the base of professionals that can know the importance of fatty acids first-hand. All cells contain fatty acids in their membranes, from the outer cell membrane to the inner membranes including mitochondria, endoplasmic reticulum and nuclear membranes. Thus it is easy to understand that the physiological roles of these ubiquitous molecules are critical to the understanding of human health by many researchers in different disciplines such as cardiovascular function, brain and retinal function, immune responses, nephrology, respiratory function, etc. Moreover, by also including the novel findings of the critical value of fatty acids to brain structure and the functionings of the mind, the reader is made aware of this exciting new area of research. Once the reader learns that docosahexaenoic acid makes up about 50% of the fatty acids in the developing brain and retina, it becomes obvious why this major reference volume devotes one-quarter of the book to the behavioral consequences of fatty acid status. Fatty acids are also a source of energy for the body because these are sources of fat calories and the reader is guided through the dilemma that faces the neonate who has insufficient energy sources and must shift the balance of essential fatty acids needed for brain development to the more immediate need of energy for survival. The challenge is particularly relevant when a premature infant is not provided with sufficient fatty acid resources to continue the optimal development of its brain, retinas, and other vital organs, a process that requires high levels of long-chain fatty acids at that critical point in development. The provision of nutritional sources for premature infants has moved from the research bench to the political arena and has been a regulatory question for several years. These issues are touched upon in several of the chapters in this important volume. It is not generally recognized that certain fatty acids in cell membranes affect the electrical conductance through the cells. The consequences of lower-than-recommended levels of n-3 fatty acids in the food supply may be one important factor in the development of arrythmias in individuals with this type of cardiac tissue dysregulation. Again, the many roles of fatty acids in human physiology are critically reviewed in this volume and the newest research is highlighted with a focus on communicating the totality of the evidence and the current level of progress in these new areas of therapeutic roles for fatty acids. Drs. Mostofsky, Yehuda, and Salem have carefully chosen the very best researchers who can communicate the relevance of fatty acid biology to professionals who are not experts in this field. The authors have worked hard to make their information accessible to health professionals interested in public health, child health, nursing, pharmacy, psychology, as well as nutrition-related health professions. In conclusion, Fatty Acids: Physiological and Behavioral Functions provides health professionals in many areas of research and practice the most up-to-date, well-referenced, and easy-to-understand volume on the importance of fatty acids for optimal human health. This volume will serve the reader as the authoritative resource in this field for many years to come. Adrianne Bendich, PhD, FACN

FOREWORD The relationship of physiological and behavioral functions to dietary levels of t3 and t6 essential nutrients are now being investigated intensely, whereas talk of such relationships was almost heresy a few years ago. The climate was similar 70 years ago, for the community of nutritional scientists did not immediately accept the concept of essential fatty acids when George and Mildred Burr proposed it in 1930. They had found that elimination of fat in the diet induced a dermatitis in rats, and that dietary linoleic and linolenic acids could prevent or correct the dermatitis. Arild Hansen, who had done his thesis with George Burr at the University of Minnesota, pursued the possibility of the essentiality in human infants, and put his findings into the medical literature in the 1930s through the 1950s. Both linoleic acid and linolenic acid were effective suppressants of dermatitis, and thus came to be considered equivalent. At Texas A & M in the 1950s, my student Carl Widmer and I were the first to show that linoleic acid is the precursor of arachidonic acid, and that linolenic acid is the precursor of eicosapentaenoic and docosahexaenoic acids in the rat. Now, 70 years later, we are treating neurological diseases with t3 essential fatty acids. We have come a long way. Because linolenic acid is much more subject to autoxidation than is linoleic acid, nutritionists and industrial laboratories attempted to eliminate linolenic acid from food formulations to minimize rancidity during storage. Autoxidation was the great enemy for designers of stable foods, and a lifetime of effort was required to breed linolenic acid out of soybeans sufficiently to make soybean oil a more stable and convenient component of industrial products intended for human consumption. In the 1950s and 1960s, worldwide studies concluded that linoleate-rich dietary oils, such as corn oil and cottonseed oil, lowered the cholesterol level of human plasma, and therefore were advocated as preferred sources of polyunsaturated fatty acids. Hence, in the food industry, polyunsaturated began to mean two double bonds per molecule. More than two double bonds per molecule was associated with rapid unwanted rancidity. In the effort to minimize plasma cholesterol, food oils for humans became richer in linoleic acid (18:2t6) and lower in linolenic acid (18:3t3), but were also becoming deficient in essential t3 polyunsaturates normally found in high levels in brain and nerve lipids. Our studies in the 1960s revealed that with a constant dietary level of 18:3t3, increasing dietary 18:2t6 suppressed the metabolism of the 18:3t3 to its more highly unsaturated products. Our later studies of EFA profiles in human health and in disease, revealed that in immune-deficiency diseases, and in diseases with neurological manifestations, low levels of t3 polyunsaturated acids were found in plasma phospholipids. Our studies of several human populations revealed that Americans had the lowest levels of t3 polyunsaturated acids in their plasmas. We have come to believe that low t3 status is a feature of many diseases, and that the American public is chronically deficient in t3, in comparison with other national populations. We now realize that deficiencies of t3 essential fatty acids are pandemic, especially in modern industrialized societies, and that this is an underlying cause of many burgeoning neurological diseases. The United States of America probably is the current leader in t3 deficiencies. How can we reverse the trend? Our entire agricultural industry is currently dedicated to the production of crops and products low in t3 and high t6 vii



essential fatty acids. Most of our food animals are fed in feed lots, largely on corn, which contains little t3 fatty acids, but is rich in linoleic acid, a competitor for enzyme sites. Therefore, our major national sources of animal protein are now relatively t3-deficient. We cannot cure our t3 deficiency by eating t3-deficient meat. However, we can replace mammalian meat by fish and enhance our intake of t3 fatty acids. Present supplies of fish may not be sufficient to meet future demand. Current fish-farming practices must be modified to enhance the t3 content of the fish, for we cannot cure our t3 deficiencies with present-day corn-fed t3-deficient fish. Perhaps part of the solution to this national and worldwide problem, could be the insertion of genes for t3 synthesis into corn, rather than trying to shift to new crop species for farmers and their animals. This one effort to enhance the ratio of the t3/t6 in corn could solve the problems related to many of our farm animals. Another solution could be to revive the soybean strains that we had 50 years ago. At this stage of the game, one cannot predict what the solution will be, but a solution must be found to eliminate our current pandemic t3 deficiency. Ralph T. Holman, PhD

FOREWORD By the time that you read this foreword, essential fatty acids in your tissues will have already had profound effects on your body’s development and health. Knowing that all that’s past is prologue, the authors of this collection of reviews assembled knowledge from their past discoveries to set the stage for readers to anticipate another wave of discovery about essential fatty acids (EFA). For 70 years, a growing body of information illumined ways in which n-3 and n-6 fatty acids maintain health and also act in disease. The essential actions of these fatty acids in physiological and behavioral functions occur through three different modes of acyl chain interaction: specific lipid–protein actions in membrane function; specific lipid–protein actions inducing gene expression; and specific receptor-mediated eicosanoid signaling (see Figure 1). When any of these interactions is influenced differently by the n-3 or n-6 arrangement of double bonds in the essential acid, it produces an important consequence of daily food choice. Voluntary food choice is important in health maintenance because the relative abundance of n-3 and n-6 acids in each person’s tissues depend on the daily supply. To help people decide whether they wish to make different food choices in the future, the authors address some very complex processes that underlie simple terms like “seizure threshold,” “immune function,” “retinal function,” or “learning behavior.” Even now, there are uncertainties about the degree to which the three modes of EFA action in Figure 1 mediate these phenomena. Throughout the 23 chapters and 2078 citations in this book, the authors give information on physiological consequences of dietary EFA supply to help readers evaluate the impact of n-3 and n-6 acid supplies on health maintenance and disease prevention.

Fig. 1. Essential fatty acids in diets and disease.




THE “DRIFT” IN SUPPLIES OF EFA For all essential nutrients, a dietary supply is the sine qua non of their action. Chapter 1 points to the general change in intake of EFA over the past centuries that has produced adverse effects on tissue abundances with resultant undesired physiologic and public health outcomes. Negative health consequences of this apparently accidental drift in relative intakes are now seen for different populations. Appendix I of Chapter 1 gives readers clear recommendations for corrective levels of dietary intakes of n-3 and n-6 acids. Chronic inattention to the simple principle that dietary supplies affect tissue abundances of EFA has also led to nearly all experimental animal models in drug development to have excessive n-6 eicosanoid signals with little moderation by n-3 eicosanoid signals. Such polarized experimental models are useful in developing patented pharmaceuticals for treating the consequences of excessive n-6 abundances in tissues. However, measurements in animals fed such imbalanced diets may give little insight into an effective nutritional strategy for preventing the onset of the pathology in the first place. Readers may want to consider the relative abundance of the essential fatty acids in the various foods they routinely eat. To help identify palatable foods that can maintain relative tissue levels of n-3 and n-6 highly unsaturated fatty acids (HUFA) at whatever level desired, readers can use a convenient interactive software program, KIM (Keep It Managed), that is accessible through the website http://ods.od.nih.gov/eicosanoids. Knowledge about the different supplies of EFA in common food servings (over 9000 are listed) seems certain to affect future voluntary dietary choices of well-informed people.

Competition in Maintaining Tissue EFA The abundant dietary 18-carbon EFA compete vigorously for the limited space conserved for long-chain HUFA in tissue lipids. This competitive metabolism creates reciprocal changes between these acids in tissues. Also, when essential n-3 and n-6 supplies are limited, n-7 and n-9 HUFA accumulate in their place. Chapters 6 through 11 explore these competitions with a focus on healthy perinatal development. These chapters extend beyond the 40-yr-old quantitative evidence of hyperbolic competitive interactions between the n-3 and n-6 acids for elongation, desaturation, and incorporation into tissue lipids (Mohrhauer and Holman, 1963a,b). After reading these chapters, readers might enjoy re-examining that seldom-discussed evidence of competitive hyperbolic metabolic processes for both 18:2n-6 and 18:3n-3, which have midpoints near 0.1 percent of ingested calories (Mohrhauer and Holman, 1963a, b as confirmed by Lands, 1991). Clearly, the effective midpoint for maintaining tissue HUFA is far below the dietary supply now common in the United States. Bioequivalence is an important concept addressed in several chapters because the highly conserved tissue HUFA are formed from 18-carbon hom*ologs more abundant in the diet. The quantitative competitions among n-3 and n-6 HUFA in the liver, plasma, and visceral tissues parallel, but quantitatively differ from, those in the brain and retina where DHA dominates. Throughout this book, authors lead readers to recognize a special ability of brain and nervous tissue to sequester DHA, illustrating important brain/body differences in EFA dynamics. Tissue abundances of EFA maintained in response to dietary supplies are readily measured by gas chromatography, and extensive efforts were made



to predict plasma proportions produced from dietary intakes (e.g., Lands et al., 1992). As a result, the proportion of plasma phospholipid total HUFA that is n-6 HUFA has a predictable relationship to the various dietary EFA supplies. This biomarker of intake (% n-6 HUFA in total HUFA) also relates to the probable intensity of n-6 eicosanoid signaling when the tissue is stimulated. However, despite some progress in estimating tissue EFA, estimating their probable actions along all three modes in Figure 1 remains a major challenge for authors and readers alike. Preformed docosahexaenoate (DHA) has a 4- to 20-fold greater relative efficacy (or bioequivalence) over linolenic acid (LNA) as substrate for accumulating as brain DHA during perinatal development. Chapter 6 reviews the kinetic data from primates to strengthen the recommendation of including at least a modest supply of DHA in infant formulas to aid brain development. Discussion of the low levels of LNA in brain tissue may reflect the predominance of phosphoglycerides in brain lipids known to differ from triacylglycerols by accumulating linolenate but little LNA. We still have no explanation for how tissue triacylglycerols accumulate LNA while the metabolically related phosphoglycerides do not! Recent advances in evaluating the supply of DHA to the nervous system are noted in Chapter 7. The authors conclude that much ingested LNA is not available for synthesis of DHA, and that elongation and desaturation events in the liver must be accompanied by biosynthetic activity in brain and nervous tissue to maintain adequate DHA levels in those tissues. Radiolabeled long-chain fatty acids help quantify EFA incorporation rates, turnover, and half-lives and imaging brain phospholipid metabolism in vivo. Chapter 8 notes that the rapid entry of plasma non-esterified fatty acids into brain acyl-CoA pools may be adequate to meet any increased neuronal demands as long as plasma DHA levels are sufficient. Half-lives for turnover of some fatty acids in phosphatidyl inositol and phosphatidyl choline of brain were around 3 h (although that for DHA in phosphatidyl choline was 22 h). The results indicate that incorporation of arachidonate into brain lipids is stimulated by the muscarinic agent, arecoline, and diminished with chronic lithium treatment. Chapter 9 focuses attention on how oxidation and the reuse of acetyl-CoA units (carbon recycling) diverts LNA from its elongation and desaturation to long-chain n-3 HUFA. Carbon recycling is described as a process that decreases the bioavailability of the n-3 fatty acids needed for neural development. Most dietary LNA is completely oxidized for energy even in rapidly growing young animals, and less than 10% seems available for DHA synthesis and esterification in the suckling rat. A three- to four-fold higher oxidation of LNA in the early postnatal period makes carbon recycling more active in neonates than adults. Readers can find an interesting aspect of brain development in Chapter 10. Astrocytes may provide to neurons 22:6n-3 produced from the 20:5n-3 that had been made by cerebroendothelial cells from 18:3n-3 acquired from plasma. If multicellular transfers are needed to provide neuronal DHA, much more needs to be known about control of this intercellular transport. The authors illustrate ways in which interpreting the nonlinear responses of brain DHA to dietary abundances continues to challenge researchers studying DHA supply and conservation. Until those dynamics are better understood, the functional and behavioral consequences of these nonlinear responses seem certain to remain equally nonlinear and puzzling. Readers will find valuable quantitative insight in Chapter 11, which re-examines bioequivalence during the rapid regain of DHA in primate



brain phospholipids during recovery from a dietary n-3 fatty acid deficiency. The extensive results with rhesus monkeys indicate the existence of mechanisms in the brain to conserve HUFA and to manage rapid, reciprocal competitive changes in n-3 and n-6 HUFA. The authors describe analyses of brain and retina EFA that parallel (but quantitatively differ from) analyses of the easily obtained plasma and red cells, which have some quantitatively predictable relations to dietary supplies (Lands et al., 1992). The overall results give clear support that the concepts in the preceding chapters about the reversible, reciprocal changes as n-3 and n-6 HUFA compete for limited space are likely transferrable to human conditions.

Tissue Physiology as Proof of EFA Importance Vitamins and hormones were identified in the early 20th century by their impact on growth and physiology. Burr and Burr (1929, 1930) identified the n-3 and n-6 EFA when they restored normal physiology to young animals on fat-free diets. Poor overall growth, irregular ovulation, scaly skin, tail necrosis, renal degeneration, and water loss could then be better interpreted. In addition, subnormal testicular development was restored by either dietary EFA or by injected gonadotropin (Greenberg and Ershoff, 1951). Now we can ask which of the three modes in Figure 1 underlie EFA support of the needed pituitary hormone production. Similar questions can address the inadequate dermal integrity that led to greater water loss during EFA deficiency. When researchers used a water rationing protocol to study growth with EFA, a clear difference between n-3 and n-6 nutrients was seen (Thomassen, 1962) that was not apparent when the water supply and humidity were adequate (Burr et al., 1940). Overall, three general physiologic processes seem supported more effectively by n-6 than n-3 EFA: dermal integrity and water balance; renal function; parturition. Readers may see in Chapters 2, 22, and 23 tantalizing clues to mechanisms for those different physiologic outcomes. Eczema and watery stools were clear biomarkers of insufficient EFA for human infants, and they were 50% prevented by about 0.07 % calories as linoleate (Hansen et al., 1963). A later meta-analysis (Cuthbertson, 1976) noted that EFA symptoms in human infants are completely prevented by less than 0.5% of calories as linoleate. Such low thresholds are similar to the 0.3% calories of EFA that proved adequate for growing rats (Mohrhauer and Holman, 1963a,b), supporting the concept that EFA metabolism and physiology are quite similar in rats and humans (Lands et al., 1992). To help readers interpret body fluid homeostasis, Chapter 22 describes possible roles for n-3 EFA in ways that extend beyond past results with water balance. In addition, new information on mediators of energy homeostasis extends beyond past results on modulating growth hormones and cytokines. Whenever physiologic processes are differentially influenced by n-3 and n-6 EFA, then possible strategies for preventive nutrition can be developed. Chapter 23 illustrates ways in which EFA fit into our expanding awareness about the “information traffic” that integrates the body’s nervous system, immune system, and endocrine system to maintain health. The authors note that mechanisms for differential modulation of stress hormones or cytokines by n-3 and n-6 EFA may involve either membrane fluidity, oxidized eicosanoid signaling, or regulation of gene expression. Any successful approach to possibly preventing a disorder will need monitoring with biomarkers that reflect health imbalances prior to the full expression of the clinical



disorder that requires treatment. As current choices of traditional foods are now being extended with new “functional foods” to help maintain a balance of n-3 and n-6 EFA, we can expect more nutritional efforts to prevent or diminish the severity of some diseases. Clinicians familiar with the symptoms and methods of treating disorders must share with nutritionists and dietitians an interest in finding useful biomarkers that can properly assess the success of prevention efforts.

Biomarkers of Complex EFA-mediated Events Many chapters address specific processes that help readers evaluate biomarkers useful in health maintenance and disease prevention. For example, Chapter 3 explores ways that DHA suppresses the expression of VCAM-1, E-selectin, and ICAM-1 on the cell surface, and normal and neoplastic leukocytes exhibit diminished adhesion through the interactions of integrins and selectins. DHA also suppresses expression of major histocompatibility II molecules. In contrast, arachidonate increased adhesion of human blood leukocytes to endothelial cells. As in other chapters of this book, readers are challenged to discern which of the three modes of interaction in Figure 1 regulate the cellular events. In Chapter 4, the effects of essential fatty acids on shifting the voltage dependence of voltage-regulated ion channels are noted as affecting seizure thresholds in animals. The discussion of ion channels extends the important finding (Kang and Leaf, 1994) that many polyunsaturated fatty acid soaps can diminish arrhythmia, but arachidonate can also exacerbate it in a manner dependent on n-6 eicosanoid formation. Release of a mixture of HUFA from tissue phospholipids can give direct anti-arrythmogenic and indirect arrythmogenic actions (Li et al., 1997). The relative proportions of n-3 and n6 HUFA that had accumulated in the tissue prior to a physiologic challenge are clearly important to tissue responsiveness. Aging is accompanied by altered cytokine expression and release during a progressive shift in relative abundance of Th1, Th2, and CD5+ cells and decline in immune function, which Chapter 5 describes in detail. This chapter provides readers an opportunity to explore how the three different modes of EFA interactions in Figure 1 might participate in altering cytokine-mediated physiology in aging. This background may then be extended to the continually appearing reports on how EFA and their oxidized products regulate expression of genes for cytokines (e.g., Wallace et al., 2001) and metabolic mediators (e.g., Clarke, 2001). Chapter 14 examines disturbances of EFA metabolism during neural complications of diabetes to help readers interpret the reported decreases in arachidonoyl species of phospholipids. Eicosanoid imbalances were indicated for this condition when a cyclooxygenase inhibitor blocked the EFA-supported improvement in nerve conduction velocity and blood flow. Chapter 15 describes another disorder that affects EFA metabolism. It is linked to defects in protein import into peroxisomes by peroxins. The important role of peroxisome enzymes in providing DHA from its docosapentaenoate precursor gives a rationale for DHA therapy in these conditions, which is described in some detail. Chapter 16 extends beyond information in Chapter 4 as it describes diverse effects of fatty acids and ketones on neuronal excitability, exploring their implications for epilepsy and its treatment. The appearance of potentially toxic fatty acid ethyl esters as non-oxidative metabolites of ethanol is reviewed in Chapter 17.



Although the formation is not selective for EFA, these products represent an intriguing biomarker of alcohol exposure.

EFA in Vision, Learning, and Behavior An outstanding integrated view of DHA actions in vision and brightness discrimination, ranging from acyl chain packing to learning events, is in the combined information of Chapters 2, 12, and 13. Chapter 2 describes how the active form of the visual G- protein-coupled receptor, metarhodopsin II, is modulated by membrane phospholipid acyl chain packing (sometimes referred to as “fluidity”). An increased bilayer area per headgroup of the polyunsaturated phospholipid is associated with more favorable kinetic coupling of the signaling components and is also linked with an increased permeability to water. The authors noted that di-18:3-PC is five times as permeable to water as 18:0,18:1-PC and 18:0,22:6-PC is four times as permeable as 18:0,18:1-PC. Readers may imagine hundreds of other G-protein-mediated systems in which the principles developed for DHA actions with rhodopsin may be extended to signaling systems throughout the body. This concept is advanced further in Chapter 12, which extends beyond phospholipid interactions with rhodopsin to describe detailed kinetics of electroretinogram waveforms and how they are used to explore possible roles of omega-3 polyunsaturated fatty acids in photo pigment activation kinetics. Careful interpretation of results on electrophysiologic signals from photoreceptors led the authors to suggest that DHA is not essential for neural function, but is needed to avoid subtle neural anomalies and produce optimal function. The important action of retinal pigmented epithelium in recycling photoreceptor components with interreceptor retinoid binding protein also depends on DHA levels, providing an indirect means by which DHA abundance can affect vision. Thus, DHA deprivation provides a puzzling mixture of neural impairments perhaps due to altered receptoral mechanisms. Extending beyond previous information, Chapter 13 explores how performance in the brightness discrimination learning test is impaired by a relative n-3 EFA deficiency. Experimenters observed diet-dependent differences in behavior as a complex outcome of retinal function and conditioned appetitive behavior. In recovering full learning behavior, the experimenters showed an expected competition by high dietary linoleate, which decreased the efficacy of n-3 EFA supplements. Also, the observed decreased turnover of DHA-rich brain ethanolamine phospholipids during n-3-deficient conditions may functionally relate to decreased neurotrophin and synaptic vesicle densities in rat hippocampus, a brain area important to learning and memory. Thus, these three chapters on vision and learning introduce key concepts that may help interpret the behavioral and cognitive phenomena described in Chapters 18 through 21. Dramatic cross-national data in Chapter 18 associate seafood and n-3 EFA supply with psychiatric disorders including major depression, bipolar affective disorder, postpartum depression, hostility and homicide, and suicide. A more positive clinical efficacy of EFA over DHA raises the possibility that these disorders may have an imbalance in specific oxygenated eicosanoids that mediate receptor signaling or gene expression rather than having inadequate membrane DHA levels. Chapter 19 reviews possible disorders of phospholipid metabolism in schizophrenia, affective disorders, and neurodegenerative disorders, and Chapter 20 describes use of eicosapentaenoic acid as a potential new treatment for schizophrenia. Chapter 21 continues exploring the importance of DHA in optimal cognitive function by describing several rodent models that measure learning and motivation. The overall results support continued interest in providing some DHA to infants to ensure adequate



neurologic development. Readers of this book will find many reasons to re-examine future food choices for themselves and their families. William E. M. Lands, PhD

References Burr GO, Burr MM. A new deficiency disease produced by rigid exclusion of fat from the diet. J Biol Chem 1929; 82:345–367. Burr GO, Burr MM. On the nature and role of the fatty acids essential in nutrition. J Biol Chem 1930; 86:587–620. Burr GO, Brown JB, Kass JP, Lundberg WO. Comparative curative values of unsat urated fatty acids in fat deficiency. Proc Soc Exp Biol Med 1940; 44:242–244. Clarke SD. Polyunsaturated fatty acid regulation of gene transcription: A molecular mechanism to improve the metabolic syndrome. J Nutr 2001; 131:1129–1132. Cuthbertson WFJ. Essential fatty acid requirements in infancy. Am J Clin Nutr 1976; 29:559–568. Greenberg SM, Ershoff BH. Effects of chorionic gonadotropin on sex organs of male rats deficient in essential fatty acids. Proc Soc Exp Biol Med. 1951; 78:552–554. Hansen AE, Wiese HF, Boelsche AN, Haggard ME, Adam DJD, Davis H. Role of linoleic acid in infant nutrition. Clinical and chemical study of 428 infants fed on milk mixtures varying in kind and amount of fat. Pediatrics 1963; 31:171–192. Kang JX, Leaf A. Proc Nat Acad Sci 1994; 91(21):9886–9890 Lands WEM. Dose-response relationships for 3/ 6 effects. In Simoupoulos AP et al., eds. Health Effects of 3 Polyunsaturated Fatty Acids in Seafoods. World Review of Nutrition and Dietetics, Vol. 66, pp.177–194. Karger, Basel, 1991. Lands WEM, Libelt B, Morris A, Kramer NC, Prewitt TE, Bowen P, Schmeisser D, Davidson MH, and Burns JH. Maintenance of lower proportions of n-6 eicosanoid precursors in phospholipids of human plasma in response to added dietary n-3 fatty acids. Biochem Biophys Acta 1992; 1180:147–162. Li Y, Kang JX, Leaf A. Differential effects of various eicosanoids on the production or prevention of arrhythmias in cultured neonatal rat cardiac myocytes. Prostaglan dins 1997; 54(2):511–530 Mohrhauer H, Holman, RT. Effect of linolenic acid upon the metabolism of linoleic acid. J Nutr 1963a; 81:67–74. Mohrhauer H, Holman RT. The effect of dose level of essential fatty acids upon fatty acid composition of the rat liver. J Lipid Res 1963b; 4:151–159. Thomassen HJ. Essential fatty acids. Nature 1962; 194:973 Wallace FA, Miles EA, Evans C, Stock TE, Yaqoob P, Calder PC. Dietary fatty acids influence the production of Th1- but not Th2-type cytokines. J Leukocyte Biol 2001; 69:449–457.

PREFACE Among the major scientific research efforts of the recent period has been the recognition of the importance of the “essential fatty acids” (EFA). The profound effects of these special chemical entities, and equally profound effects of their deficit, are appreciated by a variety of disciplines, including (but not necessarily limited to) lipid biochemistry, physiology, nutrition, psychology, psychiatry, and, perhaps most intensely, by the neurosciences at large. Functions of the central nervous system, in particular, may be seriously compromised by deficits in the levels of these FA or the ratio (or balance) among major constituents. The role of the polyunsaturated fatty acids (PUFA) _-linolenic acid (LNA; omega-3; n-3; 18:3n-3) and linoleic acid (LA; omega-6; n-6; 18:2n-6) and their metabolites has generated the most exciting findings. Health and medical implications related to these FA extend to visual development in infants, cognitive and emotional development, immunological responses, and cardiovascular health. Several foci of interest are worth noting at this point; foci that are represented in the chapters that follow and that mirror the directions in the field of FA research. The experimental study of FA deficit has been characterized by investigations that utilize food deprivation or restrictions on nutritional intake, and by designs that have provided for dietary supplementation of the FA and/or their metabolites (especially DHA and its precursors EPA and LNA). Metabolic studies continue to address many of the unexplained complexities associated with the behavior performance observations in the laboratory. Among the questions of interest are: How do the EFAs get into the brain and other organs? What is the basis for the apparent selectivity of various organs, cells, and subcellular organelles for particular lipids and FA? Why is DHA (docosahexaenoic acid; 22:6n-3) concentrated in the brain? How can the adult brain maintain its DHA even when there is little support in the diet? How much can the metabolism of the precursors of DHA (e.g., LNA, EPA, etc.) support DHA composition in the brain in comparison to the incorporation of preformed DHA taken in the diet? In addition to their basic science value, these issues have practical implications for public health policy, such as the design of infant formulas. The studies of supplementation have drawn attention to peripheral effects, such as the beneficial consequences of DHA in reducing cardiovascular mortality, reduction of immune and inflammatory responses, and influences in the management of diabetes. Supplementation effects also continue to be studied in order to better delineate complex behavioral patterns, with some critical insight on aggression, as but one example, in human studies. Deprivation of n-3 in animal research has often been concentrated on the F2 offspring where demonstrable impairments in visual function and nonvisual cognitive behaviors have repeatedly been observed. Similar outcomes in human infants have been reported, with a pronounced increase in the frequency of randomized control trials being reported in the literature. Infant behavior appears to suffer quite seriously at the hands of nutritional deprivation, with some long-term followup studies suggesting that the early deficits appear to be maintained with functional loss in later years. The reader will soon discover that differences among outcome studies may be attributable, in part or in total, to variations in the test designs used to assess physiological or behavioral function. Often xvii



the attempts to describe complex cognitive and emotional behaviors by use of learning and performance paradigms require a liberal interpretation of the results to support such assessments, which may be open to question or dispute. Despite a number of known weaknesses, unexplained phenomena, and sorely needed pieces of information yet to be discovered, the present overview of activities in these areas allows one to justifiably conclude major advances in the chemistry and biochemistry of fatty acids have contributed to a considerable understanding about the metabolism and function of fatty acids and their impact on the physiology and behavior of whole organisms. The diversity of actions of fatty acids in many biological systems such as physiological, neurological, endocrinological, and immune begs for elucidation. The management of many chronic health issues will surely benefit from such knowledge in the near term. The purpose of Fatty Acids: Physiological and Behavioral Functions is to examine such a representative segment of the scientific aspects of this area, with topics ranging from molecular analyses to functional performance of physiological and cognitive behaviors. To assist the relative newcomer to the vocabulary of the field, we have provided a glossary at the end of the volume. Considerable additional helpful information is easily obtainable from many sources on the web, as even a brief search will indicate. We hope that Fatty Acids: Physiological and Behavioral Functions will facilitate a consolidation of understanding among the separate disciplinary specialists, and will excite other investigators to enter this arena, so that even more dramatic advances and developments in chemistry, behavior, and health management will be forthcoming. David I. Mostofsky, PhD Shlomo Yehuda, PhD Norman Salem JR., PhD

CONTENTS Series Editor Page .......................................................................................................... v Foreword by Ralph Holman, PhD ................................................................................. vii Foreword by William Lands, PhD .................................................................................. ix Preface ........................................................................................................................ xvii Contributors ............................................................................................................... xxiii Part I

Basic Mechanisms 1

Evolutionary Aspects of Diet: Essential Fatty Acids....................... 3 Artemis P. Simopoulos


Modulation of Receptor Signaling by Phospholipid Acyl Chain Composition ................................................................................ 23 Drake C. Mitchell and Burton J. Litman


Role of Docosahexaenoic Acid in Determining Membrane Structure and Function: Lessons Learned from Normal and Neoplastic Leukocytes ......................................................... 41 Laura J. Jenski and William Stillwell


Effects of Essential Fatty Acids on Voltage-Regulated Ionic Channels and Seizure Thresholds in Animals .................. 63 Robert A. Voskuyl and Martin Vreugdenhil Role of Dietary Fats and Exercise in Immune Functions and Aging .................................................................................... 79 Jaya T. Venkatraman and David Pendergast


Part II Phospholipid and Fatty Acid Composition and Metabolism 6

On the Relative Efficacy of _-Linolenic Acid and Preformed Docosahexaenoic Acid as Substrates for Tissue Docosahexaenoate During Perinatal Development ................................................... 99 Meng-Chuan Huang and J. Thomas Brenna


Recent Advances in the Supply of Docosahexaenoic Acid to the Nervous System .............................................................. 115 Robert J. Pawlosky and Norman Salem Jr.


Quantifying and Imaging Brain Phospholipid Metabolism In Vivo Using Radiolabeled Long Chain Fatty Acids ............. 125 Stanley I. Rapoport


Carbon Recycling: An Important Pathway in _-Linolenate Metabolism in Fetuses and Neonates ....................................... 145 Stephen C. Cunnane




Part III

DHA and CNS Development 10

Impact of Dietary Essential Fatty Acids on Neuronal Structure and Function ............................................................................. 159 M. Thomas Clandinin, R.A.R. Bowen, and Miyoung Suh


Dietary N-3 Fatty Acid Deficiency and its Reversibility: Effects upon Brain Phospholipids and the Turnover of Docosahexaenoic Acid in the Brain and Blood ................... 177 William E. Connor, Gregory J. Anderson, and Don S. Lin


The Role of Omega-3 Polyunsaturated Fatty Acids in Retinal Function ................................................................... 193 Algis J. Vingrys, James A. Armitage, Harrison S. Weisinger, B.V. Bui, Andrew J. Sinclair, and Richard S. Weisinger


Brightness–Discrimination Learning Behavior and Retinal Function Affected by Long-Term _-Linolenic Acid Deficiency in Rat ........................................ 219 Harumi Okuyama, Yoichi Fujii, and Atsushi Ikemoto

Part IV Pathology 14

Disturbances of Essential Fatty Acid Metabolism in Neural Complications of Diabetes ....................................... 239 Joseph Eichberg and Cristinel Mîinea


Docosahexaenoic Acid Therapy for Disorders of Peroxisome Biogenesis ........................................................ 257 Hugo W. Moser and Gerald V. Raymond


Effects of Fatty Acids and Ketones on Neuronal Excitability: Implications for Epilepsy and its Treatment ............................ 273 Carl E. Stafstrom


Fatty Acid Ethyl Esters: Toxic Nonoxidative Metabolites of Ethanol ............................. 291 Zbigniew M. Szczepiorkowski and Michael Laposata

Part V Psychiatry and Behavior 18

Omega-3 Fatty Acids and Psychiatric Disorders: Current Status of the Field ....................................................... 311 Joseph R. Hibbeln and Norman Salem Jr.


Disorders of Phospholipid Metabolism in Schizophrenia, Affective Disorders, and Neurodegenerative Disorders .......... 331 David F. Horrobin


Eicosapentaenoic Acid: A Potential New Treatment for Schizophrenia? ...................... 345 Malcolm Peet and Shaun Ryles


The Importance of DHA in Optimal Cognitive Function in Rodents ................................................................................. 357 Claus C. Becker and David J. Kyle




The Role of Omega-3 Polyunsaturated Fatty Acids in Body Fluid and Energy Homeostasis ................................... 377 Richard S. Weisinger, James A. Armitage, Peta Burns, Andrew J. Sinclair, Algis J. Vingrys, and Harrison S. Weisinger 23 PUFA: Mediators for the Nervous, Endocrine, and Immune Systems ................................................................. 403 Shlomo Yehuda, Sharon Rabinovitz, and David I. Mostofsky Glossary ...................................................................................................................... 421 Index ........................................................................................................................ 425

CONTRIBUTORS GREGORY J. ANDERSON • Oregon Health Sciences University, Portland, Oregon JAMES A. ARMITAGE • Department of Optometry and Vision Sciences, University of Melbourne, Melbourne, Victoria, Australia CLAUS C. BECKER • Martek Biosciences Corporation, Columbia, Maryland R.A.R. BOWEN • Nutrition and Metabolism Research Group, Department of Agricultural, Food and Nutritional Science and Department of Medicine, University of Alberta, Edmonton, Alberta, Canada J. THOMAS BRENNA • Division of Nutritional Sciences, Cornell University, Ithaca, New York B.V. BUI • Department of Optometry and Vision Sciences, University of Melbourne, Melbourne, Victoria, Australia PETA BURNS • Howard Florey Institute, University of Melbourne, Melbourne, Victoria, Australia M. THOMAS CLANDININ • Nutrition and Metabolism Research Group, Department of Agricultural, Food and Nutritional Science and Department of Medicine, University of Alberta, Edmonton, Alberta WILLIAM E. CONNOR • Oregon Health Sciences University, Portland, Oregon STEPHEN C. CUNNANE • Department of Nutritional Sciences, University of Toronto, Toronto, Ontario JOSEPH EICHBERG • Department of Biology and Biochemistry, University of Houston, Houston, Texas YOICHI FUJII • Faculty of Pharmaceutical Sciences, Nagoya City University, Nagoya, Japan JOSEPH R. HIBBELN • National Institute on Alcohol Abuse and Alcoholism, National Institute of Health, Rockville, Maryland DAVID F. HORROBIN • Laxdale LTD, Stirling, United Kingdom MENG-CHUAN HUANG • Division of Nutritional Sciences, Cornell University, Ithaca, New York ATSUSHI IKEMOTO • Faculty of Pharmaceutical Sciences, Nagoya City University, Nagoya, Japan LAURA J. JENSKI • Department of Biological Sciences, Marshall University, Huntington, West Virginia DAVID J. KYLE • Martek Biosciences Corporation, Columbia, Maryland MICHAEL LAPOSATA • Division of Laboratory Medicine, Massachusetts General Hospital, Boston, Massachusetts DON S. LIN • Oregon Health Sciences University, Portland, Oregon BURTON J. LITMAN • Section of Fluorescence Studies, Laboratory of Membrane Biochemistry and Biophysics, National Institute on Alcohol Abuse and Alcoholism, National Institute of Health, Rockville, Maryland CRISTINEL MÎINEA • Department of Biology and Biochemistry, University of Houston, Houston, Texas DRAKE C. MITCHELL • Section of Fluorescence Studies, Laboratory of Membrane Biochemistry and Biophysics, National Institute on Alcohol Abuse and Alcoholism, National Institute of Health, Rockville, Maryland xxiii



HUGO W. MOSER • Kennedy Krieger Institute, Baltimore, Maryland DAVID I. MOSTOFSKY • Department of Psychology, Boston University, Boston, Massachusetts HARUMI OKUYAMA • Faculty of Pharmaceutical Sciences, Nagoya City University, Nagoya, Japan ROBERT J. PAWLOSKY • FCL/Beltsville Human Nutrition Research Center, Beltsville, Maryland MALCOLM PEET • Department of Psychiatry, The Longley Centre, University of Sheffield, Sheffield, United Kingdom DAVID R. PENDERGAST • Department of Physiology and Biophysics, State University of New York at Buffalo, Buffalo, New York SHARON RABINOVITZ • Psychopharmacology Laboratory, Bar Ilan University, Ramat Gan, Israel STANLEY I. RAPOPORT • Brain Physiology and Metabolism Section, National Institute on Aging, National Institutes of Health, Bethesda, Maryland GERALD V. RAYMOND • Kennedy Krieger Institute, Baltimore, Maryland SHAUN RYLES •Department of Psychiatry, The Longley Centre, University of Sheffield, Sheffield, United Kingdom NORMAN SALEM JR. • National Institute on Alcohol Abuse and Alcoholism, National Institute of Health, Rockville, Maryland ARTEMIS P. SIMOPOULOS • The Center for Genetics, Nutrition and Health, Washington, DC ANDREW J. SINCLAIR • Department of Food Science, RMIT University, Melbourne, Victoria, Australia CARL E. STAFSTROM • Departments of Neurology and Pediatrics, University of Wisconsin, Madison, Wisconsin WILLIAM STILLWELL • Department of Biology, Indiana University, Purdue University Indianapolis, Indianapolis, Indiana MIYOUNG SUH • Nutrition and Metabolism Research Group, Department of Agricultural, Food and Nutritional Science and Department of Medicine, University of Alberta, Edmonton, Alberta, Canada ZBIGNIEW M. SZCZEPIORKOWSKI • Blood Transfusion Service, Massachusetts General Hospital, Boston, Massachusetts JAYA T. VENKATRAMAN • Department of Physical Therapy, Exercise and Nutrition Sciences, State University of New York at Buffalo, Buffalo, New York ALGIS J. VINGRYS • Department of Optometry and Vision Science, University of Melbourne, Melbourne, Victoria, Australia ROBERT A. VOSKUYL • Leiden/Amsterdam Center for Drug Research; Epilepsy Clinics Foundation of the Netherlands, Heemstede, The Netherlands MARTIN VREUGDENHILL • Division of Neuroscience, Department of Neurophysiology, Medical School, University of Birmingham, Birmingham, United Kingdom HARRISON S. WEISINGER • Department of Food Science, RMIT University, Melbourne, Victoria, Australia RICHARD S. WEISINGER • Howard Florey Institute, University of Melbourne, Melbourne, Victoria, Australia SHLOMO YEHUDA • Psychopharmacology Laboratory, Bar Ilan University, Ramat Gan, Israel

Chapter 1 / Evolutionary Aspects of Diet





Part I / Basic Mechanisms

Chapter 1 / Evolutionary Aspects of Diet



Evolutionary Aspects of Diet Essential Fatty Acids

Artemis P. Simopoulos 1. INTRODUCTION The interaction of genetics and environment, nature, and nurture is the foundation for all health and disease. This concept, based on molecular biology and genetics, was originally defined by Hippocrates. In the 5th century BC, Hippocrates stated the concept of positive health as follows: Positive health requires a knowledge of man’s primary constitution [which today we call genetics] and of the powers of various foods, both those natural to them and those resulting from human skill [today’s processed food]. But eating alone is not enough for health. There must also be exercise, of which the effects must likewise be known. The combination of these two things makes regimen, when proper attention is given to the season of the year, the changes of the winds, the age of the individual and the situation of his home. If there is any deficiency in food or exercise the body will fall sick.

In the last two decades, using the techniques of molecular biology, it has been shown that genetic factors determine susceptibility to disease and environmental factors determine which genetically susceptible individuals will be affected (Simopoulos and Childs, 1990; Simopoulos and Nestel, 1997; Simopoulos, 1999d). Nutrition is an environmental factor of major importance. Whereas major changes have taken place in our diet over the past 10,000 yr since the beginning of the Agricultural Revolution, our genes have not changed. The spontaneous mutation rate for nuclear DNA is estimated at 0.5% per million years. Therefore, over the past 10,000 yr there has been time for very little change in our genes, perhaps 0.005%. In fact, our genes today are very similar to the genes of our ancestors during the Paleolithic period 40,000 yr ago, at which time our genetic profile was established (Eaton and Konner, 1985). Genetically speaking, humans today live in a nutritional environment that differs from that for which our genetic constitution was selected. Studies on the evolutionary aspects of diet indicate that major changes have taken place in our diet, particularly in the type and amount of essential fatty acids (EFA) and in the antioxidant content of foods (Eaton and Konner, 1985; Simopoulos, 1991; Simopoulos, 1999a; Simopoulos, 1999c) (Table 1, Fig. 1). Using the tools of molecular biology and genetics, research is defining the mechanisms by which genes influence nutrient absorption, metabolism and excretion, taste perception, and degree of satiation, and the mechanisms by which nutrients influence gene expression. From: Fatty Acids: Physiological and Behavioral Functions Edited by: D. Mostofsky, S. Yehuda, and N. Salem Jr. © Humana Press Inc., Totowa, NJ



Part I / Basic Mechanisms

Table 1 Characteristics of Hunter–Gatherer and Western Diet and Lifestyles Characteristic

Hunter–Gatherer diet and lifestyle

Western diet and lifestyle

Physical activity level



Energy density Energy intake Protein Animal Vegetable Carbohydrate (rapidly absorbed) Fiber Fat Animal Vegetable Total long chain n-6 + n-3 Ratio n-6:n-3

Low Moderate High High Very low Low–moderate Slowly absorbed High Low Low very low High (2.3 g/d) Low (2.4)

High High Low–moderate Low–moderate Low–moderate Moderate Rapidly absorbed Low High High Moderate to high Low (0.2 g/d) High (12.0)

Vitamins (mg/d)

Paleolithic period

Current US intake

Riboflavin Folate Thiamin Ascorbate Carotene (retinol equivalent) Vitamin A (retinol equivalent) Vitamin E

6.49 0.357 3.91 604 5.56 (927) 17.2 (2870) 32.8

1.34–2.08 0.149–0.205 1.08–1.75 77–109 2.05–2.57 — 7.02–8.48 (1170–429) 7–10


Source: Modified from Simopoulos, 1999a.

Whereas evolutionary maladaptation leads to reproductive restriction (or differential fertility), the rapid changes in our diet, particularly the last 100 yr, are potent promoters of chronic diseases such as atherosclerosis, essential hypertension, obesity, diabetes, and many cancers. In addition to diet, sedentary lifestyles and exposure to noxious substances interact with genetically controlled biochemical processes, leading to chronic diseases. This chapter discusses evolutionary aspects of diet and the changes that have occurred in Western diets, due to the increase in omega-6 and decrease in omega-3 fatty acid intake from the large-scale production of vegetables oils, agribusiness, and modern agriculture, with emphasis on the balance of omega-6 : omega-3 fatty acids. The Appendix is a portion of the summary of The Workshop on the Essentiality of and Recommended Dietary Intakes (RDIs) for Omega-6 and Omega3 Fatty Acids, held at the National Institutes of Health (NIH) in Bethesda, MD, USA, April 7–9, 1999 (Simopoulos, et al., 1999).

Chapter 1 / Evolutionary Aspects of Diet


Fig. 1. Hypothetical scheme of fat, fatty acid (t-6, t-3, trans and total) intake (as percent of calories from fat) and intake of vitamins E and C (mg/d). Data were extrapolated from crosssectional analyses of contemporary hunter–gatherer populations and from longitudinal observations and their putative changes during the preceding 100 yr. (Simopoulos, 1999a).

2. EVOLUTIONARY ASPECTS OF DIET The foods that were commonly available to preagricultural humans (lean meat, fish, green leafy vegetables, fruits, nuts, berries, and honey) were the foods that shaped modern humans’ genetic nutritional requirements. Cereal grains as a staple food are a relatively recent addition to the human diet and represent a dramatic departure from those foods to which we are genetically programmed and adapted (Cordain, 1999; Simopoulos, 1995a; Simopoulos, 1999d). Cereals did not become a part of our food supply until very recently—10,000 yr ago—with the advent of the Agricultural Revolution. Prior to the Agricultural Revolution, humans ate an enormous variety of wild plants, whereas, today, about 17% of plant species provide 90% of the world’s food supply, with the greatest percentage contributed by cereal grains (Cordain, 1999; Simopoulos, 1995a; Simopoulos, 1999d). Three cereals, wheat, maize, and rice, together account for 75% of the world’s grain production. Human beings have become entirely dependent on cereal grains for the greater portion of their food supply. The nutritional implications of such a high grain consumption upon human health are enormous. And yet, for the 99.9% of mankind’s presence on this planet, humans never or rarely consumed cereal grains. It is only since the last 10,000 yr that humans consume cereals. Up to that time, humans were non-cerealeating hunter–gatherers since the emergence of hom*o erectus 1.7 million years ago. There is no evolutionary precedent in our species for grass seed consumption (Eaton and Konner, 1985). Therefore, there has been little time (DHA>eicosapentaenoic acid (EPA), increased neutrophil adherence in vitro concomitantly with upregulated expression of Mac-1 (but not the other integrins LFA-1 and CR4) (Bates et al., 1993). DHA decreased adhesion of human peripheral blood lymphocytes to activated endothelial cells and reduced endothelial VCAM-1 expression and L-selectin as well as LFA-1 expression on lymphocytes, but did not influence endothelial ICAM-1 or E-selectin expression (Khalfoun et al., 1996a). Some of the disparities reported for DHA and EPA relate to the stimulus used to induce adhesion molecules on cells (Weber et al., 1995). Both DHA and EPA are reported to reduce the production of mRNA for various adhesion molecules (Collie-Duguid & Wahle, 1996; Wahle & Rotondo, 1999), suggesting that DHA may affect adhesion through transcriptional control of adhesion proteins as well as through direct effects on membrane structure. CD8 and CD4, present on mature T-cell subsets, are not adhesion molecules per se but, rather, provide additional signals during antigen recognition, thereby enhancing the likelihood of T-cell activation. Oth et al. (Oth et al., 1990) observed that a fish-oil diet reduced CD4 expression on lymphoma cells (grown as ascites) without affecting expression of CD8, CD11a/CD18, and MHC I. The dietary fatty acids may have affected CD4 gene expression or modified the membrane so as to deform or mask the epitope detected by the anti-CD4 monoclonal antibody. With regard to the latter, we demonstrated that cells modified with DHA presented as a fatty acid in culture medium or in phospholipids fused into the plasma membrane of lymphocytes displayed altered expression of two of three CD8 epitopes (one epitope decreased, another increased, the last was unchanged) (Jenski et al., 1995a), arguing for a direct effect of DHA on CD8 conformation or the lateral distribution of CD8 and its interaction with other “masking” proteins perhaps in specific lipid microdomains.

9. DHA AND MAJOR HISTOCOMPATIBILITY COMPLEX PROTEINS Fatty acids are reported to affect the expression of MHC molecules. In some cases, the protein’s synthesis may be affected, whereas other experimental designs point to a direct effect of the lipid on plasma membrane-bound MHC. The function of MHC molecules is to bind peptides and present these peptides to specific T-cells, thereby clonally activating T-cells reactive to the peptide. The two key classes of MHC molecules functioning in this fashion are class I (MHC I) and class II (MHC II). MHC I molecules are present on all nucleated cells in the body and predominantly bind peptides generated by proteosomemediated cleavage of cytosolic proteins. The tissue distribution of MHC II is limited to B-cells, dendritic cells, and macrophages, although expression may be induced on various other cell types, primarily by cytokines. MHC II molecules bind peptides produced in endolysosomes by proteolytic cleavage of exogenously derived proteins. These two types of molecules share an overall general structure, that of four extracellular protein domains, of which the two distal to the plasma membrane directly participate in peptide binding, transmembrane region(s), and cytoplasmic tail. MHC I molecules have three

Chapter 3 / DHA and Leukocyte Membranes


extracellular domains provided by an _-chain and the fourth by an associated non-MHC protein, `2-microglobulin. Two MHC II polypeptides, the _-chain and the `-chain, each provide two extracellular domains. The expression of MHC I molecules may be modified by DHA. We have demonstrated altered expression of MHC I, CD8, and CD90 (Thy-1) on murine lymphocytes and leukemia cells enriched in DHA through diet or cell culture (Jenski et al., 1995a; Jenski et al., 1993). In addition, we have shown that MHC I expression can be modulated by the direct insertion of DHA-containing phospholipids into the plasma membrane of viable cells; because one MHC I epitope increased concurrently with the decrease of another epitope, the effect of DHA was not a global loss of the MHC protein (Pascale et al., 1993). This is an important observation, as it implies that the role of the DHA-containing phospholipids is to induce a physical change in MHC I that pre-exists at the cell surface, rather than exclusively through modification of MHC I biosynthesis. This conclusion was drawn more directly from the observation that more monoclonal antibodies against an MHC I conformation-dependent epitope bound to purified MHC I reconstituted into DHA-containing phosphatidylcholine liposomes than liposomes composed of other fatty acids (Jenski et al., 2000). Finally, catalytic hydrogenation of viable murine leukemia cells reduced the membrane content of linoleic acid (LA), AA, and DHA but did not affect the expression of MHC I, although another protein displayed increased surface expression (Benko et al., 1987). That plasma membrane structure is important for the expression of MHC I molecules is also suggested by experiments with other lipids. Cholesterol plays a role in modifying MHC surface expression, and, in general, cholesterol’s effects oppose those of phospholipids (phosphatidylcholine). Incubation of mouse splenocytes with cholesteryl hemisuccinate for 1–2 h in vitro decreased the expression of MHC I and increased lipid packing as monitored by DPH fluorescence, whereas a PC lipid mixture from hen egg yolk increased the expression of MHC I and fluidized the membrane (Muller et al., 1983). Cholesterol enrichment of cultured human B-lymphoblasts led to increased MHC I clustering (Bodnar et al., 1996), possibly as a result of conformational changes in the MHC protein. Bene et al. (1997) suggested that membrane depolarization may induce physical changes in the lipid bilayer and thereby produce conformational changes in human MHC I molecules. In general, DHA or mixtures of omega-3 fatty acids decrease the expression of MHC II molecules. As designed, most experiments detect changes in MHC II production rather than direct modification of plasma membrane structure. For example, human volunteers fed a fish-oil supplement displayed reduced expression of MHC II on peripheral blood monocytes before and after treatment with the cytokine a-interferon (Hughes et al., 1996a), and dietary fish oil decreased MHC II expression on rat lymphatic dendritic cells (Sanderson et al., 1997) and murine peritoneal exudate cells (primarily B-cells and macrophages) (Huang et al., 1992). The nature of the stimulus used to induce MHC II expression is an important consideration in evaluating DHA’s effects. Macrophages (thioglycollate-elicited peritoneal exudate cells) from n3-rich fish-oil-fed mice expressed more MHC II after a brief treatment with platelet-activating factor contrast than did macrophages from n6-rich safflower oil (Erickson et al., 1997), and similar results were obtained with EPA or DHA added in vitro. When EPA and DHA, combined in ratios commonly found in fish oil, were added to cultures of unstimulated human monocytes, MHC II expression was unaffected; however, these n3 combinations did inhibit MHC II


Part I / Basic Mechanisms

expression on a-interferon-stimulated monocytes (Hughes & Pinder, 1997). When these n3 fatty acids were added individually to unstimulated human monocyte cultures, EPA inhibited but DHA enhanced MHC II expression (both inhibited expression on stimulated monocytes) (Hughes et al., 1996b). In vitro, DHA inhibited MHC II expression on murine macrophages (thioglycollate-elicited peritoneal exudate cells) treated with a-interferon in a fashion not reversed with leukotrienes or 5-HETE, implying an action for DHA independent of the lipoxygenase pathway (Khair-El-Din et al., 1996). With regard to function, peptides were shown to bind with greater affinity to purified murine MHC II in the presence of PC, PS, PI, and cardiolipin (but not PE, sphingomyelin, or cholesterol), although the role of very long-chain polyunsaturated fatty acids (e.g., DHA) was not tested (Roof et al., 1990).

10. DHA AND CELL SIGNALING The influence of DHA on leukocyte signaling has focused on several membrane proteins, including surface receptors, ion channels, kinases, phosphatases, and phospholipases. Here, we examine three examples germane to leukocytes: the interleukin-2 receptor (IL2R), protein kinase C, and calcium channels (i.e., calcium mobilization).

10.1. IL2R Interleukin-2 signals through a trimeric plasma membrane receptor composed of _, `, and a polypeptide chains, thereby promoting cell cycling. T-Lymphocytes are a primary, although not an exclusive, target for IL2. Mitogen-stimulated peripheral blood lymphocytes from human volunteers consuming EPA and DHA ethyl esters displayed a reduced surface density of CD25, the IL2R _-chain induced by cell activation, implying reduced immunological responsiveness (Soyland et al., 1994). DHA acts, at least in part, at the level of transcription, dramatically decreasing IL2R mRNA in concanavalin A-stimulated murine splenic lymphocytes (Jolly et al., 1998). In this study, DHA- and EPAenriched diets were similar in their inhibitory effects on IL2R mRNA synthesis, showing approximately twice the inhibition produced by AA supplementation relative to the LA-rich safflower-oil control diet. DHA or EPA added to phytohemagglutinin-stimulated human peripheral blood mononuclear cells in culture-inhibited lymphocyte proliferation independently of oxidation and eicosanoid hormones, but flow cytometry suggested that these fatty acids, in this case, increased the surface expression of CD25 (Khalfoun et al., 1996b).

10.2. Protein Kinase C Protein kinase C, which exists in various isoforms, is activated when it translocates to the inner leaflet of the plasma membrane and interacts with the lipid bilayer including diacylglycerol (DAG). Omega-3 fatty acids induce or enhance PKC activation or translocation, often more so than other polyunsaturated fatty acids, and these actions have been reported in various cells including leukocytes. DHA, EPA, and AA induced the translocation of PKC_, PKC-`I, PKC-`II, and PKC-¡ isozymes to a particulate fraction in parallel with enhanced respiratory burst in stimulated macrophages (Huang et al., 1997). DHA also stimulated respiratory burst in neutrophils, however, in a fashion independent of PKC but involving calmodulin (Poulos et al., 1991). Under certain conditions, DHA displays extraordinary ability to activate PKC; Marignani et al. (Marignani et al.,

Chapter 3 / DHA and Leukocyte Membranes


1996) used an in vitro system of PKC in lipid vesicles to demonstrate that 18:0,22:6-snglycerol was more effective than 18:0,20:4-sn-glycerol, 18:0,20:5-sn-glycerol, or dioleoylglycerol in stimulating PKC activity.

10.3. Calcium Mobilization Within seconds after receiving a signal, many cells display increased levels of cytosolic free calcium. The sources of the cytosolic free calcium are intracellular stores, such as the endoplasmic reticulum, and the extracellular milieu. IP3, generated in concert with DAG from phosphatidylinositol 4,5-biphosphate (PIP2) cleavage by phospholipase C, stimulates mobilization of stored calcium. Depletion of these stores stimulates calcium influx from the extracellular milieu. Fatty acids affect calcium mobilization and are thus assumed to affect cell activation. Chow et al. (Chow et al., 1990) stimulated the human T leukemia cell line Jurkat with antibodies to the CD3 component of the T-cell antigen receptor in the presence of various unsaturated free fatty acids. DHA, as well as EPA, _-linolenic acid (ALA), AA, LA, and oleic acid, did not affect the initial rise in cytosolic free calcium but appeared to prevent the extracellular calcium influx required for sustained cytosolic-free-calcium levels. Fatty acids did not affect CD3 expression as measured by flow cytometry, nor involve PKC in their mode of action, and thus the initial interpretation was a direct effect of fatty acids on the receptor-operated calcium channels. It was shown subsequently that the free fatty acids inhibited sustained cytosolic free calcium levels in anti-CD3-stimulated Jurkat cells by increasing calcium extrusion, presumably through activation of the plasma membrane calcium ATPase (Breittmayer et al., 1993). In untreated Jurkat cells (i.e., without additional stimulation), free polyunsaturated fatty acids directly mobilized intracellular calcium pools (these pools were also sensitive to anti-CD3 and IP3) (Chow & Jondal, 1990); DHA appeared to be more effective than the other n3 (EPA, ALA) and n6 fatty acids (AA, LA). When various n3 and n6, but not n9, fatty acids were esterified into membrane phospholipids of Jurkat cells, the anti-CD3-induced rise in cytosolic free calcium was dampened, presumably because the influx of extracellular calcium, rather than release from intracellular stores, was impaired by the fatty acids (Chow et al., 1991).

11. CONCLUDING REMARKS Omega-3 fatty acids have long been recognized as beneficial foodstuff, but their actions are complex and thus poorly understood. There is now a resurgence of interest in omega-3 fatty acids, particularly DHA, among basic scientists and clinicians. DHA is being used in a variety of forms, from dietary supplements to novel drug conjugates, to benefit an extensive and varied series of conditions, including normal vision and neurologic development, heart disease, autoimmune disease, and cancer. The emerging literature suggests a diversity of mechanisms of action for DHA: modulation of eicosanoid hormone production, generation of free radicals, regulation of gene expression, and the fundamental actions of DHA on membrane structure and function. It is the current state of the art, however, that the physiological and even cellular processes affected by DHA are not clearly connected to a single action of DHA, and thus there is urgent need for the interdisciplinary research that will link the “what,” “how,” “when,” and “for whom” at the molecular, cellular, and organismal levels.


Part I / Basic Mechanisms

REFERENCES Ahkong QF, Fisher D, Tampion W, Lucy JA. Fusion of erythrocytes by fatty acids, esters, retinol, and alphatocopherol. Biochem J 1973; 136:147–155. Almeida PFF, Vaz WLC, Thompson TE. Lateral diffusion in the liquid phases of dimyristoylphosphatidylcholine/cholesterol lipid bilayers. A free volume analysis. Biochemistry 1992; 31:6739–6747. Anderson RE, Sperling L. Lipids of the ocular tissue. VII. Positional distribution of the fatty acids in the phospholipids of bovine rod outer segments. Arch Biochem Biophy 1971; 144:673–677. Applegate KR, Glomset JA. Computer-based modeling of the conformation and packing properties of docosahexaenoic acid. J Lipid Res 1991; 27:658–680. Arnold RS, Cornell RB. Lipid regulation of CTP : phosphocholine cytidyltransferase: electrostatic, hydrophobic and synergistic interactions of anionic phospholipids and diacylglycerol. Biochemistry 1996; 35:9917–9924. Bates EJ, Ferrante A, Harvey DP, Poulos A. Polyunsaturated fatty acids increase neutrophil adherence and integrin receptor expression. J Leukocyte Biol 1993; 53:420–426. Bene L, Szollosi J, Balazs M, Matyus L, Gaspar R, Ameloot M, et al. Major histocompatibility complex class I protein conformation altered by transmembrane potential changes. Cytometry 1997; 27:353–357. Benko S, Hilkmann H, Vigh L, van Blitterswijk WJ. Catalytic hydrogenation of fatty acyl chains in plasma membranes; effect on membrane lipid fluidity and expression of cell surface antigens. Biochim Biophys Acta 1987; 896:129–135. Bernsohn J, Stephanides LM. Aetiology of multiple sclerosis. Nature 1967; 215:821–823. Bodnar A, Jenei A, Bene L, Damjanovich S, Matko J. Modification of membrane cholesterol level affects expression and clustering of class I HLA molecules at the surface of JY human lymphoblasts. Immunol Lett 1996; 54:221–226. Bougnoux P. n-3 Polyunsaturated fatty acids and cancer. Curr Opin Clin Nutr Metab Care 1999; 2:121–126. Bougnoux P, Salem NJ, Lyons C, Hoffman T. Alterations in themembrane fatty acid composition of human lymphocytes and cultured transformed cells induced by interferon. Mol Immunol 1985; 22:1107–1113. Brand MD, Couture P, Hulbert AJ. Liposomes made from mammalian liver mitochondria are more polyunsaturated and leakier to proteins than those from reptiles. Comp Biochem Physiol 1994; 108:181–188. Breckenridge WC, Gombos G, Morgan IG. The lipid composition of adult rat brain synaptosomal membranes. Biochim Biophys Acta 1972; 266:695–707. Breittmayer J-P, Pelassy C, Cousin J-L, Bernard A, Aussel C. The inhibition by fatty acids of receptormediated calcium movements in Jurkat T-cells is due to increased calcium extrusion. J Biol Chem 1993; 268:20,812–20,817. Brown DA, London E. Structure of detergent-resistant membrane domains: Does phase separation occur in biological membranes? Biochem Biophys Res Commun 1997; 240:1–7. Brown DA, London E. Structure and origin of ordered lipid domains in biological membranes. J Membr Biol 1998; 164:103–114. Brown MF. Modulation of rhodopsin function by properties of the membrane bilayer. Chem Phys Lipids 1994; 73:159–180. Bruckdorfer KR, Sherry MK. The solubility of cholesterol and its exchange between membranes. Biochim Biophys Acta 1984; 769:187–196. Brzustowicz MR, Stillwell W, Wassall SR. Molecular organization of cholesterol in polyunsaturated membranes: a solid state 2H NMR investigation. FEBS Lett 1999; 451 197–202. Brzustowicz MR, Zerouga M, Cherezov V, Caffrey M, Stillwell W, Wassall SR. Solid state NMR and X-ray diffraction studies of cholesterol molecular organization in polyunsaturated phospholipid membranes. Biophys J 2000; 78:184A. Burns PC, Spector AA. Effects of lipids on cancer therapy. Nutr Rev 1990; 48:223–240. Calder PC. n-3 Polyunsaturated fatty acids and cytokine production in health and disease. Ann Nutr Metab 1997; 41:203–234. Chow SC, Ansotegui IJ, Jondal M. Inhibition of receptor-mediated calcium influx in T cells by unsaturated non-esterified fatty acids. Biochem J 1990; 267:727–732. Chow SC, Jondal M. Polyunsaturated free fatty acids stimulate an increase in cytosolic Ca2+ by mobilizing the inositol 1,4,5-trisphosphate-sensitive Ca2+ pool in T cells through a mechanism independent of phosphoinositide turnover. J Biol Chem 1990; 265:902–907.

Chapter 3 / DHA and Leukocyte Membranes


Chow SC, Sisfontes L, Jondal M, Bjorkhem I. Modification of membrane phospholipid fatty acyl composition in a leukemic T cell line: effects on receptor mediated intracellular Ca2+ increase. Biochim Biophys Acta 1991; 1092:358–366. Clamp AG, Ladha S, Clark DC, Grimble RF, Lund EK. The influence of dietary lipids on the composition and membrane fluidity of rat hepatocyte plasma membrane. Lipids 1997; 32:179–184. Collie-Duguid ESR, Wahle KWJ. Inhibitory effect of fish oil n-3 polyunsaturated fatty acids on the expression of endothelial cell adhesion molecules. Biochem Biophys Res Commun 1996; 220:969–974. Coolbear KP, Berde CB, Keough KMW. Gel to liquid-crystalline transitions of aqueous dispersions of polyunsaturated mixed-acid phosphatidylcholines. Biochemistry 1983; 22:1466–1473. Crespo-Armas A, Azavache V, Torres SH, Anchustegui B, Cordero, Z. Changes produced by experimental hypothyriodism in fibre type composition and mitochondrial properties of rat slow and fast twitch muscles. Acta Cientifica Venezolana 1994; 45:42–44. Crinier S, Morrot G, Neumann J-M, Devaux PF. Lateral diffusion of erythrocyte phospholipids in model membranes-comparison between inner and outer leaflet components. Eur Biophys J 1990; 18:33–41. Cullis PR, de Kruijff B. Lipid polymorphism and the functional roles of lipids in biological membranes. Biochim Biophys Acta 1979; 559:399–420. De Caterina R, Cybulsky MA, Clinton SK, Gimbrone MA Jr, Libby P. Omega-3 fatty acids and endothelial leukocyte adhesion molecules. Prostaglandins Leukotrienes Essential Fatty Acids 1995; 52:191–195. De Caterina R, Liao JK, Libby P. Fatty acid modulation of endothelial activation. Am J Clin Nutr 2000; 71:213S–223S. De Caterina R, Libby P. Control of endothelial leukocyte adhesion molecules by fatty acids. Lipids 1996; 31:S57–S63. Del Buono BJ, White SM, Williamson PL, Schlegel RA. Plasma membrane lipid organization an dthe adherence of differentiating lymphocytes to macrophages. J Cell Physiol 1989; 138:61–69. Demel RA, Geuts van Kessel WS. M., Van Deenen LL. M. The properties of polyunsaturated lecithins in monolayers and liposomes and the interactions of these lecithins with cholesterol. Biochim Biophys Acta 1972; 266:26–40. Demel RA, Jansen JWCM, Van Dijck PWM, Van Deenen LLM. The preferential interaction of cholesterol with different classes of phospholipids. Biochim Biophys Acta 1977; 465:1–10. Devaux PF. Static and dynamic lipid asymmetry in cell membranes. Biochemistry 1991; 30:1163–1173. Dratz EA, Deese AJ. The role of docosahexaenoic acid (22:6w3) in biological membranes: examples from photoreceptors and model membrane bilayers. In: Simopoulos AP, et al., eds. Health Effects of Polyunsaturated Fatty Acids in Sea Foods. Academic, New York, 1986, pp. 353–379. Dratz EA, Holte LL. The molecular spring model for the function of docosahexaenoic acid (22:6w-3) in biological membranes. In: Sinclair A, Gibson R, eds. Essential Fatty Acids and Eicosanoids. American Oil Chemists Society, Champaign, IL, 1992, pp. 122–127. Dratz EA, Van Breeman JF, Kamps KM, Keegstra W, Van Bruggen EF. Two-dimensional crystallizationof bovine rhodopsin. Biochim Biophys Acta 1985; 832:337–342. Dumaual AC, Jenski LJ, Stillwell W. Liquid crystalline/gel state phase separation in docosahexaenoic acidcontaining bilayers and monolayers. Biochim Biophys Acta 2000; 1463:395–406. Dusserre E, Pulcini T, Bourdillon, M-C, Ciavatti M, Berthezene F. w-3 Fatty acids in smooth muscle cell phospholipids increase membrane cholesterol efflux. Lipids 1995; 30:34–41. Dyerberg J, Jorgensen KA. Marine oils and thrombosis. Prog Lipid Res 1982; 21:255–269. Edidin M. Patches and fences: probing for plasma membrane domains. J Cell Sci 1993; 17(Supp):165–169. Ehringer W, Belcher D, Wassall S, Stillwell W. A comparison of the effects of linolenic (18:313) and docosahexaenoic (22:613) acids on phospholipid bilayers. Chem Phys Lipids 1990; 54:79–88. El Ayachi N, Begin M, Mercier D, Ells G, Oth D. Susceptibility of RDM4 lymphoma cells to LAK-mediated lysis is decreased in tumor bearers fed fish oil high fat regimen. Cancer Lett 1990; 49:217–224. Epand RM. Lipid polymorphism and protein-lipid interactions. Biochim Biophys Acta 1998; 1376:353–368. Erickson KL, Howard AD, Chakrabarti R, Hubbard NE. Alteration of platelet activating factor-induced macrophage tumoricidal response, IA expression, and signal transduction by n-3 fatty acids. Adv Exp Med Biol 1997; 407:371–378. Fontaine RN, Harris RA, Schroeder F. Aminophospholipid asymmetry in murine synaptosomal plasma membrane. J Neurochem 1980; 34:269–277. Fritsche KL, Johnston PV. Effect of dietary omega-3 fatty acids on cell-mediated cytotoxic activity on BALB/c mice. Nutr Res 1990; 10:577–588.


Part I / Basic Mechanisms

Ghosh D, Williams MA, Tinoco J. The influence of lecithin structure on their monolayer behavior and interactions with cholesterol. Biochim Biophys Acta 1973; 291:351–362. Gibney MJ, Bolton-Smith C. The effect of a dietary supplement of n-3 polyunsaturated fat on platelet lipid composition, platelet function, and platelet plasma membrane fluidity in healthy volunteers. Br J Nutr 1988; 60:5–12. Giorgione J, Epand RM, Buda C, Farkas T. Role of phospholipids containing docosahexaenoyl chains in modulating the activity of protein kinase C. Proc Natl Acad Sci USA 1995; 92:9767–9770. Hasler CM, Trosko JE, Bennink MR. Incorporation of n-3 fatty acids into WB-F344 cell phospholipids inhibits gap junctional intercellular communication. Lipids 1991; 26:788–792. Henricks TH, Klompmakers AA, Daemen FJM, Bonting SL. Biochemical aspects of the visual process. XXXII. Movement of sodium ions through bilayers composed of retinal and rod outer segment lipids. Biochim Biophys Acta 1976; 433:271–281. Hernandez-Borrell J, Keough KMW. Heteroacid phosphatidylcholines with different amounts of unsauration respond differently to cholesterol. Biochim Biophys Acta 1993; 1153:277–282. Hinds A, Sanders TAB. The effect of increasing levels of dietary fish oil rich in eicosapentaenoic and docosahexaenoic acids on lymphocyte phospholipid fatty acid composition and cell-mediated immunity in the mouse. Br J Nutr 1993; 69:423–429. Holte LL, Koenig BW, Strey HH, Gawrisch KBJ. Structure and dynamics of the docosahexaenoic acid chain in bilayers studied by NMR and X-ray diffraction. Biophys J 1998; 74:A371. Holte LL, Senaka AP, Sunwell TM, Gawrisch K. 2H Nuclear magnetic resonance order parameter profiles suggest a change of molecular shape for phosphatidylcholines containing a polyunsaturated acyl chain. Biophys J 1995; 68:2396–2403. Huang H-W, Goldberg EM, Zidovetzki R. Ceramide induces structural defects into phosphatidylcholine bilayers and activates phospholipase A2. Biochem Biophys Res Commun 1996; 220:834–838. Huang S-C, Misfeldt ML, Fritsche KL. Dietary fat influences Ia antigen expression and immune cell populations in the murine peritoneum and spleen. J Nutr 1992; 122:1219–1231. Huang ZH, Hii CS. T., Rathjen DA, Poulos A, Murray AW, Ferrante A. n-6 and n-3 Polyunsaturated fatty acids stimulate translocation of protein kinase C_, -`I, -`II, and -¡ and enhance agonist-induced NADPH oxidase in macrophages. Biochem J 1997; 325:553–557. Huber LA, Xu QB, Jurgens G, Bock G, Buhler E, Gey KF, et al. Correlation of lymphocyte lipid composition, membrane microviscosity and mitogen response in the aged. Eur J Immunol 1991; 21:2761–2765. Hughes DA, Pinder AC. N-3 polyunsaturated fatty acids modulate the expression of functionally associated molecules on human monocytes and inhibit antigen presentation in vitro. Clin Exp Immunol 1997; 110:516–523. Hughes DA, Pinder AC, Piper Z, Johnson IT, Lund EK. Fish oil supplementation inhibits the expression of major histocompatibility complex class II molecules and adhesion molecules on human monocytes. Am J Clin Nutr 1996; 63:267–272. Hughes DA, Southon S, Pinder AC. (n-3) Polyunsaturated fatty acids modulate the expression of functionally associated molecules on human monocytes in vitro. J Nutr 1996; 126:603–610. Hui SW. The spatial distribution of cholesterol in membranes. In: Yeagle PL, ed. The Biology of Cholesterol. CRC, Boca Raton, FL, 1988, pp. 213–232. Hui SW, Stewart TP, Yeagle PL, Albert AD. Bilayer to non-bilayer transition in mixtures of phosphatidylethanolamine and phosphatidylcholine: implications for membrane properties. Arch Biochem Biophy 1981; 207:227–240. Hullin F, Bossant MJ, Salem NJ. Aminophospholipid molecular species asymmetry inthe human erythrocyte plasma membrane. Biochim Biophys Acta 1991; 1023:335–340. Huster D, Albert JJ, Arnold K, Gawrisch K. Water permeability of polyunsaturated lipid membranes measured by 17O NMR. Biophys J 1997; 73:856–864. Huster D, Arnold K, Gawrisch K. Influence of docosahexaenoic acid and cholesterol on lateral lipid organization in phospholipid mixtures. Biochemistry 1998; 37:17,299–17,308. James MJ, Gibson RA, Cleland LG. Dietary polyunsaturated fatty acids and inflammatory mediator production. Am J Clin Nutr 2000; 71:343S–348S. Jensen JW, Schutzbach JS. Modulation of dolichyl-phosphomannose synthase activity by changes in the lipid environment of the enzyme. Biochemistry 1988; 27:6315–6320. Jenski LJ, Bowker GM, Johnson MA, Ehringer WD, Fetterhoff T, Stillwell W. Docosahexaenoic acidinduced alteration of Thy-1 and CD-8 expression on murine splenocytes. Biochim Biophys Acta 1995; 1236:39–50.

Chapter 3 / DHA and Leukocyte Membranes


Jenski LJ, Nanda PK, Jiricko P, Stillwell W. Docosahexaenoic acid-containing phosphatidylcholine affects the binding of monoclonal antibodies to purified Kb reconstituted into liposomes. Biochim Biophys Acta, 2000; 1467(2):293–306. Jenski LJ, Sturdevant LK, Ehringer WD, Stillwell W. t-3 Fatty acids increase spontaneous release of cytosolic components from tumor cells. Lipids 1991; 26:353–358. Jenski LJ, Sturdevant LK, Ehringer WD, Stillwell W. Omega-3 fatty acid modification of membrane structure and function: I. Dietary manipulation of tumor cell susceptibility to cell and complement-mediated lysis. Nutr Cancer 1993; 19:135–146. Jenski LJ, Zerouga M, Stillwell W. t-3 Fatty acid-containing liposomes in cancer therapy. Proc Soc Exp Biol Med 1995; 210:227–233. Jolly CA, McMurray DN, Chapkin RS. Effect of dietary n-3 fatty acids on interleukin-2 and interleukin-2 receptor a expression in activated murine lymphocytes. Prostaglandins Leukotrienes Essential Fatty Acids 1998; 58:289–293. Kafawy O, Zerouga M, Stillwell W, Jenski LJ. Docosahexaenoic acid in phosphatidylcholine mediates cytotoxicity more effectively than other omega-3 and omega-6 fatty acids. Cancer Lett 1998; 132:23–29. Kang JX, Leaf A. Evidence that polyunsaturated fatty acids modify Na+ channels by directly binding to the channel protein. Proc Natl Acad Sci USA 1996; 93:3542–3546. Kariel N, Davidson E, Keough KM. W. Cholesterol does not remove the gel–liquid crystalline phase transition of phosphatidylcholines containing two polyenoic acyl chains. Biochim Biophys Acta 1991; 1062:70–76. Khair-El-Din TA, Sicher SC, Vazquez MA, Lu CY. Inhibition of macrophage nitric-oxide production and Ia-expression by docosahexaenoic acid, a constituent of fetal and neonatal serum. Am J Reprod Immunol 1996; 36:1–10. Khalfoun B, Thibault G, Bardos P, Lebranchu Y. Docosahexaenoic and eicosapentaenoic acids inhibit in vitro human lymphocyte–endothelial cell adhesion. Transplantation 1996; 62:1649–1657. Khalfoun B, Thibault G, Lacord M, Gruel Y, Bardos P, Lebranchu Y. Docosahexaenoic and eicosapentaenoic acids inhibit lymphoproliferative responses in vitro but not the expression of T cell surface activation markers. Scand J Immunol 1996; 43:248–256. Knapp HR, Hullin F, Salem NJ. Asymmetric incorporation of dietary n-3 fatty acids into membrane aminophospholipids of human erythrocytes. J Lipid Res 1994; 35:1283–1291. Koenig BW, Strey HH, Gawrisch K. Membrane lateral compressibility determined by NMR and x-ray diffraction: effect of acyl chain polyunsaturation. Biophys J 1997; 73:1954–1966. Lavoie C, Jolicoeur M, Paiement J. Accumulation of polyunsaturated free fatty acid coincident with the fusion of rough endoplasmic reticulum membranes. Biochim Biophys Acta 1991; 1070:274–278. Lewis BA, Engelman DM. Lipid bilayer thickness varies linearly with acyl chain length in fluid phosphatidylcholine vesicles. J Mol Biol 1983; 166:211–217. Li L, Zheng LX, Yang FY. Effect of propensity of hexagonal II phase formation on the activity of mitochondrial ubiquinol–cytochrome c reductase and H(+)-ATPase. Chem Phys Lipids 1995; 76:135–144. Maggio B, Diplock AT, Lucy JA. Interactions of tocopherols and ubiquinones with monolayers of phospholipids. Biochem J 1977; 161:111–121. Marignani PA, Epand RM, Sebaldt RJ. Acyl chain dependence of diacylglycerol activation of protein kinase C activity in vitro. Biochem Biophys Res Commun 1996; 225:469–473. Mayol V, Duran MJ, Gerbi A, Dignat-George F, Levy S, Sampol J, et al. Cholesterol and omega-3 fatty acids inhibit Na, K ATPase activity in human endothelial cells. Atherosclerosis 1999; 142:327–333. McCallum CD, Epand RM. Insulin receptor autophosphorylation and signaling is altered by modulation of membrane physical properties. Biochemistry 1995; 34:1815–1824. Meers P, Hong K, Papahadjopoulos D. Free fatty acid enhancement of cation-induced fusion of liposomes; symergism with synexin and other promoters of vesicle aggregation. Biochemistry 1988; 27:6784–6794. Menkes JH, Alter M, Steigleder GK, Weakley DR, Sung JH. A sex-linked recessive disorder with retardation of growth, peculiar hair and focal cerebral and cerebellar degeneration. Pediatrics 1962; 29:764–779. Mescher MF, Apgar JR. The plasma membrane “skeleton” of tumor and lymphoid cells: a role in cell lysis? Adv Exp Med Biol 1986; 184:387–400. Miljanich GP, Sklar LA, White DL, Dratz EA. Disaturated and dipolyunsaturated phospholipids in the bovine retinal rod outer segment disk membrane. Biochim Biophys Acta 1979; 552:294–306. Mitchell DC, Gawrisch K, Litman BJ, Salem NJ. Why is docosahexaenoic acid essential for nervous system function? Biochem Soc Trans 1998; 26:365–370. Mitchell DC, Litman BJ. Molecular order and dynamics in bilayers consisting of highly polyunsaturated phospholipids. Biophys J 1998; 74:879–891.


Part I / Basic Mechanisms

Morrot G, Cribier S, Devaux PF, Geldwerth D, Davoust J, Bureau JF, et al. Asymmetric lateral movement of phospholipids in the human erythrocyte membrane. Proc Natl Acad Sci USA 1986; 83:6863–6867. Mosior M, Golini ES, Epand RM. Chemical specificity and physical properties of the lipid bilayer in the regulation of protein kinase C by anionic phospholipids: evidence for the lack of a specific binding site for phosphatidylserine. Proc Natl Acad Sci USA 1996; 93:1907–1912. Muller CP, Stephany DA, Shinitzky M, Wunderlich JR. Changes in cell-surface expression of MHC and Thy1.2 determinants following treatment with lipid modulating agents. J Immunol 1983; 131:1356–1362. Nabekura J, Noguchi K, Witt MR, Nielsen M, Akaike N. Functional modulation of human recombinant gamma-aminobutyric acid type A receptor by docosahexaenoic acid. J Biol Chem 1998; 273:11,056– 11,061. Neill AR, Masters CJ. Metabolism of fatty acids by ovine spermatozoa. J Reprod Fertil 1973; 34:279–287. Neuringer N, Anderson GJ, Conner WE. The essentiality of n-3 fatty acids for the development and function of the retina and brain. Ann Rev Nutr 1988; 8:517–541. Niebylski CD, Litman BJ. Phase separation in bilayers containing disaturated and dipolyunsaturated acyl chains. Biophys J 1996; 70:A418. Niebylski CD, Littman BJ. Lateral phase heterogeneity in dipolyunsaturated and disaturated phospholipid bilayers monitored with novel pyrene-labeled phospholipids. Biophys J 1997; 72:A305. Niebylski CD, Salem NJ. A calorimetric investigation of a series of mixed-chain polyunsaturated phosphatidylcholines: effect of sn-2 chain length and degree of unsaturation. Biophys J 1994; 67:2387–2393. Oth D, Mercier G, Tremblay P, Therien HM, Begin ME, Ells G, et al. Modulation of CD4 expression on lymphoma cells transplanted to mice fed (n-3) polyunsaturated fatty acids. Biochim Biophys Acta 1990; 1027:47–52. Parks JS, Thuren TY, Schmitt JD. Inhibition of lecithin : cholesterol acyltransferase activity by synthetic phosphatidylcholine species containing eicosapentaenoic acid or docosahexaenoic acid in the sn-2 position. J Lipid Res 1992; 33:879–887. Pascale AW, Ehringer WD, Stillwell W, Sturdevant LK, Jenski LJ. Omega-3 fatty acid modification of membrane structure and function. II. Alteration by docosahexaenoic acid of tumor cell sensitivity to immune cytolysis. Nutr Cancer 1993; 19:147–158. Pepe S, Bogdanov K, Hallaq H, Spurgeon H, Leaf A, Lakatta E. w3 Polyunsaturated fatty acid modulates dihydropyridine effects on L-type Ca2+ channels, cytosolic Ca2+, and contraction in adult rat cardiac myocytes. Proc Natl Acad Sci USA 1994; 91:8832–8836. Poling JS, Karanian JW, Salem N Jr, Vicini S. Time- and voltage-dependent block of delayed rectifier potassium channels by docosahexaenoic acid. Mol Pharmacol 1995; 47:381–390. Popp-Snigders C, Shouten JA, Van Blitterswijck WJ, Van der Veen EA. Changes in membrane lipid composition of human erythrocytes after dietary supplementation of (n-3) polyunsaturated fatty acids. Biochim Biophys Acta 1986; 854:31–37. Popp-Snijders C, Schouten JA, Heine RJ, van der Meer J, van der Veen EA. Dietary supplementation of omega-3 polyunsaturated fatty acids improves insulin sensitivity in non-insulin-dependent diabetes. Diabetes Res 1987; 4:141–147. Poulos A, Robinson BS, Ferrante A, Harvey DP, Hardy SJ, Murray AW. Effect of 22–32 carbon n-3 polyunsaturated fatty acids on superoxide production in human neutrophils: synergism of docosahexaenoic acid with f-met-leu-phe and phorbol ester. Immunology 1991; 73:102–108. Robinson DR, Xu L-L, Knoell CT, Tateno S, Olesiak W. Modification of spleen phospholipid fatty acid composition by dietary fish oil and by n-3 fatty acid ethyl esters. J Lipid Res 1993; 34:1423–1434. Roof RW, Luescher IF, Unanue ER. Phospholipids enhance the binding of peptides to class II major histocompatibility molecules. Proc Natl Acad Sci USA 1990; 87:1735–1739. Salem N Jr, Niebylski CD. The nervous system has an absolute molecular species requirement for proper function. Mol Membr Biol 1995; 12:131–134. Salem NJ, Kim H-Y, Yergey JA. Docosahexaenoic acid: membrane function and metabolism. In Simopoulos AP, Kifer RR, and Martin RE, eds. Health Effects of Polyunsaturated Fatty Acids in Seafoods. Academic, New York, 1986, pp. 319–351. Sanderson P, Calder PC. Dietary fish oil appears to prevent the activation of phospholipase C-a in lymphocytes. Biochim Biophys Acta 1998; 1392:300–308. Sanderson P, MacPherson GG, Jenkins CH, Calder PC. Dietary fish oil diminshes the antigen presentation activity of rat dendritic cells. J Leukocyte Biol 1997; 62:771–777. Sankaram MB, Thompson TE. Interaction of cholesterol with various glycerophospholipids and sphingomyelin. Biochemistry 1990; 29:10,670–10,675.

Chapter 3 / DHA and Leukocyte Membranes


Schroeder F, Jefferson JR, Kier AB, Knittel J, Scallen TJ, Wood WG, et al. Membrane cholesterol dynamics: cholesterol domains and kinetic pools. Proc Soc Exp Biol Med 1991; 196:235–252. Schroeder RJ, Ahmed SN, Zhu Y, London E, Brown DA. Cholesterol and sphingolipid enhance the Triton X-100 insolubility of glycosylphosphatidylinositol-anchored proteins by promoting the formation of detergent-insoluble ordered membrane domains. J Biol Chem 1998; 273:1150–1157. Schwarz D, Kisselev P, Wessel R, Pisch S, Bornscheur U, Schmid RD. Possible involvement of nonbilayer lipids in the stimulation of the activity of cytochrome P450SCC (CYP11A1) and its propensity to induce vesicle aggregation. Chem Phys Lipids 1997; 85:91–99. Sellmayer A, Danesh U, Weber PC. Modulation of the expression of early genes by polyunsaturated fatty acids. Prostagladins Leukotrienes Essential Fatty Acids 1997; 57:353–357. Slater SJ, Kelley MB, Taddeo FJ, Ho C, Rubin E, Stubbs CD. The modulation of protein kinase C activity by membrane lipid bilayer structure. J Biol Chem 1994; 269:4866–4871. Slater SJ, Kelly MB, Yeager MD, Larkin J, Ho C, Stubbs CD. Polyunsaturation in cell membranes and lipid bilayers and its effects on membrane proteins. Lipids 1996; 31:S189–S192. Smaby JM, Momsen MM, Brockman HL, Brown RE. Phosphatidylcholine acyl chain unsaturation modulates the decrease in interfacial elasticity induced by cholesterol. Biophys J 1997; 73:1492–1505. Souvignet C, Pelosin JM, Daniel S, Chambaz EM, Ransac S, Verger R. Activation of protein kinase C in lipid monolayers. J Biol Chem 1991; 266:40–44. Soyland E, Lea T, Sandstad B, Drevon A. Dietary supplmentation with very long -chain n-3 fatty acids in man decreases expression of the interleukin-2 receptor (CD25) on mitogen-stimulated lymphocytes from patients with inflammatory diseases. Eur J Clin Invest 1994; 24:236–242. Stillwell W, Dallman T, Dumaual AC, Crump FT, Jenski LJ. Cholesterol versus _-tocopherol: effects on properties of bilayers made from heteroacid phosphatidylcholines. Biochemistry 1996; 35:13,353– 13,362. Stillwell W, Ehringer W, Jenski LJ. Docosahexaenoic acid increases permeability of lipid vesicles and tumor cells. Lipids 1993; 28:103–108. Stillwell W, Ehringer WD, Wassall SR. Interaction of a-tocopherol with fatty acids in membranes and ethanol. Biochim Biophys Acta 1992; 1105:237–244. Stillwell W, Jenski LJ, Crump FT, Ehringer W. Effect of docosahexaenoic acid on mouse mitochrondrial membrane properties. Lipids 1997; 32:497–506. Stillwell W, Jenski LJ, Zerouga M, Dumaual AC. Detection of lipid domains in docosahexaenoic acid-rich bilayers by acyl chain-specific FRET probes. Chem Phys Lipids 2000; 104:113–132. Stillwell W, Wassall SR, Dumaual AC, Ehringer W, Browning CW, Jenski LJ. Use of merocyanine (MC540) in quantifying lipid domains and packing in phospholipid vesicles and tumor cells. Biochim Biophys Acta 1993; 1146:136–144. Streicher-Scott J, Lapidus R, Sokolove PM. The reconstituted mitochondrial adenine nucleotide translocator: effects of lipid polymorphism. Arch Biochem Biophy 1994; 315:548–554. Stubbs CD, Smith AD. The modification of mammalian membrane polyunsaturated fatty acid composition in relation to membrane fluidity and function. Biochim Biophys Acta 1984; 779:89–137. Stubbs CD, Tsang WM, Belin J, Smith D, Johnson SM. Incubation of exogenous fatty acids with lymphocytes. Changes in fatty acid composition and effects on rotational relaxation time of 1,6-diphenyl-1,3,5hexatriene. Biochemistry 1980; 19:2756–2762. Suzuki H, Manabe S, Wada O, Crawford MA. Rapid incorporation of docosahexaenoic acid from dietary sources into brain microsomal, synaptosomal and mitochondrial membranes in adult mice. Int J Vitam Nutr Res 1997; 67:272–278. Talbot WA, Zheng L, Lentz BR. Acyl chain unsaturation and vesicle curvature alter outer leaflet packing and promote poly(ehtylene glycol)-mediated membrane fusion. Biochemistry 1997; 36:5827–5836. Tiwari RK, Venkatraman JT, Cinader B, Flory J, Wierzbicki A, Goh YK, et al. Influence of genotype on the phospholipid fatty acid composition of splenic T and B lymphocytes in MRL/MpJ-lpr/lpr mice. Immunol Lett 1988; 17:151–158. Treen M, Uauay RD, Jameson DM, Thomas VL, Hoffman RR. Effect of docosahexaenoic acid on membrane fluidity and function in intact cultured Y-79 retinoblastoma cells. Arch Biochem Biophy 1992; 294:564–570. Urquhart R, Chan RY, Li OT, Tilley L, Grieser F, Sawyer WH. Omega-6 and omega-3 fatty acids: monolayer packing and effects on bilayer permeability and cholesterol exchange. Biochem Int 1992; 26:831–841. Wahle KW, Rotondo D. Fatty acids and enothelial cell function: regulation of adhesion molecule and redox enzyme expression. Curr Opin Clin Nutr Metab Care 1999; 2:109–115.


Part I / Basic Mechanisms

Wahnon R, Cogan U, Mokady S. Dietary fish oil modulates the alkaline phosphatase activity and not the fluidity of rat hepatocyte plasma membrane. J Nutr Metab 1992; 29:279–288. Wassall SR, Yang McCabe RC, Ehringer WD, Stillwell, W. Effects of dietary fish oil on plasma high density lipoproteins: electron spin resonance and fluorescence polarization studies of ordering and dynamics. J Biol Chem 1992; 267:8168–8174. Weber C, Erl W, Pietsch A, Danesch U, Weber PC. Docosahexaenoic acid selectively attenuates induction of vascular cell adhesion molecule-1 and subsequent monocytic cell adhesion to human endothelial cells stimulated by tumor necrosis factor-alpha. Arteriosclerosis, Thromb Vascular Biol 1995; 15:622–628. Whiting LA, Harvey CC, Century B, Worwitt MK. Dietary alterations of fatty acids of erythrocytes and mitochondria of brain and liver. J Lipid Res 1961; 2:412–418. Wiegand RD, Anderson RE. Phospholipid molecular species of frog rod outer segment membranes. Exp Eye Res 1983; 37:159–173. Yaqoob P, Newsholme EA, Calder PC. Influence of cell culture conditions on diet-induced changes in lymphocyte fatty acid composition. Biochim Biophys Acta 1995; 1255:333–340. Yu J, Fischman DA, Steck TL. Selective solubilization of proteins and phospholipids from red blood cell membranes by nonionic detergents. Journal of Supramol Struct 1973; 3:233–248. Zerouga M, Stillwell W, Stone J, Powner A, Jenski LJ. Phospholipid class as a determinant in docosahexaenoic acid’s effect on tumor cell viability. Anticancer Res 1996; 16:2863–2868.

Chapter 4 / Actions on Na Channels and Seizures



Effects of Essential Fatty Acids on Voltage-Regulated Ionic Channels and Seizure Thresholds in Animals Robert A.Voskuyl and Martin Vreugdenhil

1. INTRODUCTION If polyunsaturated fatty acids (PUFAs) have an antiarrhythmic action on the heart, do they also have a suppressant action on other excitable cells, such as neuronal tissue? The answer to this question is important, because there are several conditions where lowering of neuronal excitability obviously has a beneficial effect and where there is still a need for new and better treatments. Neuropathic pain and neurological disorders such as epilepsy are two examples. The work of Leaf and co-workers (Leaf, Kang, Xiao, Billman & Voskuyl, 1999a; Leaf, Kang, Xiao, Billman & Voskuyl, 1999b) has convincingly demonstrated that the stabilizing effect of PUFAs on the electrical activity of isolated cardiac myocytes stems mainly from inhibition of voltage-regulated sodium and calcium currents. Some studies in neuronal preparations have pointed in the same direction (Park & Ahmed, 1992; Takenaka, Horie & Hori, 1987; Takenaka, Horie, Hori & Kawakami, 1988). In this respect, PUFAs bear a striking resemblance to local anesthetics and a number of antiepileptic drugs. Furthermore, one of the antiepileptic drugs that interact with sodium channels, valproic acid, is actually a short-chain fatty acid. This remarkable coincidence prompted us to investigate the possible antiepileptic action of a number of PUFAs on isolated hippocampal neurons and in an experimental epilepsy model in vivo. The second reason why PUFAs deserve interest in the context of epilepsy is the recent re-emergence of the ketogenic diet as an alternative for antiepileptic drug treatment. The diet has a high fat content and is low in carbohydrates. Although it does not seem to work in all patients, it has been remarkably successful in treating, in particular, the most difficult types of epilepsy in children. At present, it is unclear why and how the diet works, but it is not unreasonable to suppose that the composition of the diet and the different fat components should have an influence. This subject is discussed in detail in Chapter 17.

2. EPILEPSY, SYNDROMES, SEIZURES, AND ANTIEPILEPTIC DRUGS Epilepsy is one of the most common neurological disorders, with a prevalence of approximately 0.5–1% of the world population. It is not a hom*ogenous disease entity, but a collection of diverse syndromes. These are broadly divided in localization-related (focal, local, partial) epilepsies and generalized epilepsies, which involve the whole From: Fatty Acids: Physiological and Behavioral Functions Edited by: D. Mostofsky, S. Yehuda, and N. Salem Jr. © Humana Press Inc., Totowa, NJ



Part I / Basic Mechanisms

Fig. 1. Epileptiform discharges. Spontaneous epileptiform bursts recorded intracellularly in a CA1 pyramidal cell in a hippocampal slice of a young rat. The epileptiform activity was induced by perfusing the slice with 50 µM 4-aminopyridine. Most of the action potentials are clipped by the low sampling rate.

brain. However, the complete official classification of epileptic syndromes and epilepsies (1989) comprises almost 40 different types. Whatever the syndrome or type of epilepsy, they all have in common the sudden and unpredictable, repeated occurrence of epileptic seizures, which, again, may take a large variety of appearances. Epileptic seizures are characterized by abnormal, excessive synchronous discharges in neuronal cerebral networks, accompanied by behavioral manifestations, ranging from a hardly noticeable arrest of activity that lasts only a few seconds, to the dramatic tonic–clonic convulsions that may last several minutes. In the face of this bewildering complexity, it will be no surprise that different types of epilepsy and seizures ask for different drugs. Fortunately, in the majority of patients, seizures can be effectively controlled by one or more of the presently available drugs, allowing them a reasonable normal life. Nevertheless, there remains a considerable group of about 25% in which seizures cannot be suppressed by any drug or only at dosages that cause unacceptable side effects. Therefore, there is a continuing need for new drugs and new therapies. From a mechanistic point of view, the three major targets for antiepileptic drugs are enhancement of synaptic inhibition, depression of synaptic excitation, and containment of excessive action potential generation (e.g., Löscher, 1998; White, 1999). The importance of the latter is illustrated in Fig. 1, which shows a typical epileptiform burst of action potentials riding on top of a strong depolarization, induced in isolated brain tissue by one of the many available convulsants. As there is a remarkable similarity in how PUFAs decrease the excitability of cardiac myocytes and how local anesthetics, phenytoin, and other antiepileptic drugs do the same job for neurons, we will first describe the action potential mechanism at some length. Excellent, more detailed reviews on the molecular properties of sodium channels and actions of anticonvulsants have been published recently (Catterall, 1999; Ragsdale & Avoli, 1998; Taylor & Narasimhan, 1997).

3. VOLTAGE-REGULATED SODIUM CHANNELS Figure 2 shows how conformational changes of the sodium channels are involved in the generation of action potentials. Sodium channels are large glycoproteins that form a

Chapter 4 / Actions on Na Channels and Seizures


Fig. 2. Action potential generation. Opening and closing kinetics of sodium and potassium channels during an action potential. The curved arrows indicate the predominant conformation of the channels during the different phases of an action potential.

pore through the membrane. The opening and closing of this channel is regulated by the membrane potential. In neurons that are not active, the resting potential is about –70 mV (inside negative with respect to the outside), and at this potential, the large majority of the channels is in a nonconducting or closed conformation. This is indicated here schematically by a constriction in the pore. When the neuron is depolarized (e.g., by an excitatory postsynaptic potential), sensors in the sodium channel detect the voltage change and this is the signal for the channel to open. This allows sodium ions, for which the pore is selective, to rush into the cell and depolarize the cell further. As a result of this positive feedback mechanism, the membrane potential will change from –70 mV to approx +20 mV within a fraction of a millisecond. Depolarization not only causes opening or activation of the channel but, at a slightly slower time-scale, also closing or inactivation of the channel. The latter has been envisioned as a “hinged-lid” mechanism (the lid being an intracellular loop of the channel protein), where the lid physically plugs the pore. The inactivation of the channel blocks the influx of sodium, allowing the neuron to restore the resting potential, a process which is helped by opening of voltagesensitive potassium channels. At the time-scale of the action potential (i.e., about a millisecond), the potassium channels simply open upon depolarization and close when the membrane potential is restored, but more slowly than the sodium channels. It is important to realize that the conformational change from the “Closed” to the “Open” state is a reversible step, but the conversion from the “Open” to the “Inactivated” state is irreversible. Thus, upon continuing depolarization, all sodium channels eventually become trapped in the inactivated (nonconducting) state, from which they can only escape to the (also nonconducting) closed state by hyperpolarization (i.e., by restoration of the resting potential). This removal of inactivation is a relatively slow process and determines how quick a neuron can fire a new action potential. Properties of the activation and inactivation processes can be conveniently studied by clamping the membrane potential to


Part I / Basic Mechanisms

Fig. 3. Voltage-clamp protocols. Typical voltage-clamp protocols to assess the inactivation curve (A and B) and the recovery from inactivation (C and D). (A) The voltage protocol is shown in the top trace, the transient sodium current in the bottom trace (by convention an inward current is shown as a downward deflection). From rest, the membrane potential is first “stepped’’ to a conditioning potential (prepulse), and from there to a value near 0 mV for maximal activation of the sodium current. In this case, the prepulse is hyperpolarizing to remove sodium inactivation. (B) Inactivation curve (i.e., peak current as a function of the conditioning prepulse). The dashed line illustrates the typical leftward shift induced by phenytoin or carbamazepine. The dotted vertical line at about –75 mV illustrates that after a leftward shift, the maximal sodium current that can be achieved from that potential is reduced, and by that, the chance of generating an action potential. (C) In this protocol, all sodium channels are first brought into the inactivated state by activating them with a long depolarizing voltage step. After a recovery at the original resting potential for a certain time 6t, the sodium current is activated again. (D) Rate of recovery. The dashed line illustrates the delay of recovery by, for example, phenytoin.

different values in a stepwise fashion. The current needed to keep the potential at that level then provides information on the kinetics of the channel. Figure 3 shows some typical voltage-clamp protocols. It should also be mentioned that the sodium channel contains a number of intracellular phosphorylation sites. The degree of phosphorylation of these sites is under the control of cyclic AMP or protein kinase C. This provides further opportunities for modulation of sodium currents, which may modify high-frequency firing. Local anesthetics like lidocaine and anticonvulsants like phenytoin do not interfere with the process of activating sodium channels itself and thus leave the action-potentialgenerating mechanism intact. However, they cause a voltage- and use-dependent block of sodium current that reduces the number of channels available for action potential

Chapter 4 / Actions on Na Channels and Seizures


generation (Schwartz & Grigat, 1989; Taylor & Narasimhan, 1997; Vreugdenhil & Wadman, 1999), as do PUFAs in cardiac myocytes (Xiao, Kang, Morgan & Leaf, 1995; Xiao, Wright, Wang, Morgan & Leaf, 1998). The block is poor when the cell is activated from a strongly negative membrane potential, but it becomes progressively more effective when starting from more depolarized levels. This is illustrated by the typical leftward shift of the inactivation curve in Fig. 3B. Activating the neuron from a potential of about –75 mV (dotted vertical line), the peak sodium current is about 90% of the maximal achievable current, as indicated by the solid curve. In other words, at rest, 90% of the sodium channels are available for activation. If the curve is shifted to the left, as indicated by the dashed line, the peak sodium current will be only 50% of the maximum if activation starts from –75 mV. Only 50% of the channels are available for opening (i.e., 50% are in the inactivated state), and the chance of reaching the threshold for action potential generation is reduced. Furthermore, the block increases with repeated activation of the sodium channel. This can be explained by assuming preferential, but slow, binding to the inactivated channel. Binding is therefore incomplete during the short time window after a single activation, but will accumulate with repeated activation. Furthermore, most anticonvulsants also delay the recovery of inactivation. This combined action will keep more and more channels in the inactivated state and fewer channels are available for opening. The various anticonvulsants, local anesthetics, and PUFAs differ only in minor details in their actions. The voltage- and use-dependent block of sodium channels is a very desirable property for anticonvulsants, because it means that the block becomes effective only when the neuron is depolarized and firing at high frequency, which is the case during epileptic activity. This limitation of the frequency at which a neuron can fire action potentials is very characteristic and often the first indication of an effect on sodium channels (McLean & Macdonald, 1986a; McLean & Macdonald, 1986b). In cardiac myocytes, PUFAs have an analogous action (Leifert, McMurchie & Saint, 1999; Xiao et al., 1995), providing a low-pass filter for repetitive contraction, in effect preventing cardiac arrhythmias in rat (Hock, Beck, Bodine & Reibel, 1990) and man (de Logeril et al., 1994). Under normal conditions when no high-frequency firing is needed, the functioning of the channel is hardly changed. Because activation is not affected, the shape of the action potential is otherwise unaffected. Kang and Leaf provided evidence that PUFAs block sodium currents only in the free form (Kang & Leaf, 1994). Incorporated in the membrane, bound to albumin, or esterified, they are inactive. They also showed that the efficacy depends on the number of double bonds, the n-3 fatty acids docosahexaenoic acid (DHA) and eicosapentaenoic acid (EPA) being the most effective (Kang & Leaf, 1994). Finally, they showed that PUFAs have a similar action on calcium channels (Xiao, Gomez, Morgan, Lederer & Leaf, 1997), which behave in much the same way as sodium channels.

4. EFFECTS OF FATTY ACIDS ON SODIUM CURRENTS IN HIPPOCAMPAL NEURONS Reduction in sodium currents in squid giant axons (Takenaka et al., 1987; Takenaka et al., 1988) and dorsal root ganglion cells cultured in PUFA containing media (Park & Ahmed, 1992) has been reported earlier. We have tested whether PUFAs have a similar action on sodium currents in pyramidal neurons from the hippocampal CA1 area. Our


Part I / Basic Mechanisms

choice of neuron type was inspired by the crucial role this hippocampal output structure plays in temporal lobe epilepsy (Meier & Dudek, 1996). Neurons were acutely isolated from enzyme-treated CA1 tissue pieces, cut from the hippocampus removed from the brain of adult Wistar rats. Detailed methods are given in two earlier articles (Vreugdenhil et al., 1996; Vreugdenhil & Wadman, 1999). The neurons selected for recording had a truncated apical dendrite and allowed adequate voltage control in the whole-cell patch configuration. Voltage-dependent calcium currents and potassium currents were pharmacologically blocked.

4.1. Voltage Dependence of Sodium Current Inactivation Is Affected To assess the voltage dependence of the inactivation process, we determined the voltage dependence of steady-state inactivation, using a double-pulse voltage protocol. During the first conditioning pulse of 0.5 s, the fraction of channels in the inactivated state was set by stepping to different potentials. The result was tested by a second step to –25 mV (voltage protocol is given as an inset in Fig. 4A). Figure 4A gives a typical series of current traces recorded using the double-pulse protocol. After a hyperpolarizing conditioning pulse, a maximal inward current was recorded that decayed exponentially. Following a depolarizing conditioning pulse, only a small fraction of the current was left. The normalized peak current amplitude as a function of the conditioning voltage was fit with a Boltzmann equation (Fig. 4B) that describes the voltage dependence with a potential of half-maximal inactivation (Vh; –60.6 ± 0.3 mV, n = 128) and a factor Vc proportional to the slope at Vh (5.6 ± 0.1 mV). After two measurements in virtually lipid-free control solution containing 1 mg of delipidated bovine serum albumin (BSA), the voltage dependence of steady-state inactivation was determined in the presence of fatty acids in the BSA-free perfusate. We first tested the effect of 16 µM of the PUFA cis4,7,10,13,16,19 docosahexaenoic acid (DHA: C22:6n-3). Sixteen micromolar DHA shifted the voltage dependence of steady-state inactivation by 13 mV to more hyperpolarized levels, with a small increase of Vc. We quantified the effect of different fatty acids on the shift in voltage dependence of steady-state inactivation. Figure 4C gives the negative shift in Vh (6Vh) for 16 µM of the PUFAs DHA, cis-5,8,11,14,17-eicosapentaenoic acid (EPA: C20:5n-3), and cis-9,12-octadecadienoic acid (linolenic acid; LA: C18:2n-6), the monounsaturated fatty acid cis-9-octadecaenoic acid (oleic acid; OA: C18:1n-9) and the unsaturated fatty acid hexadecanoic acid (palmitic acid; PA: C16:0) and, as a control, continued perfusion with BSA. There was a clear relationship between the amount of unsaturated bonds and the potency of the fatty acid to shift Vh, with the monounsaturated OA and the saturated PA not different from the BSA control. The same relationship was found in cardiac myocytes, where the potency was found to be related to membrane fluidity (Leifert et al., 1999). In correlation with the shift in the potential of half-maximal activation 6Vh, the slope factor Vc increased significantly for the PUFAs (0.8 ± 0.2 mV for DHA, 1.3 ± 0.2 mV for EPA, and 0.6 ± 0.2 mV for LA). As the prediction was that the refractory period would increase as a result of a slower recovery from inactivation, we tested the effect of DHA on the time-course of the recovery from inactivation. Using a double-pulse protocol, the current was first completely inactivated by a 20-ms conditioning pulse to –25 mV. After a recovery period of increasing duration at either –70 mV or –80 mV, the fraction of channels that were recovered from inactivation was assessed with a second pulse to –25 mV (Fig. 4D). The recovery from inactivation as a function of interval was fit by an exponential function with a time

Chapter 4 / Actions on Na Channels and Seizures


Fig. 4. PUFAs shift voltage dependence of sodium current inactivation. (A) Sodium current inactivation. A set of sodium currents recorded in an isolated CA1 neuron by a 10-ms depolarizing potential step to –25 mV, following different 0.5-s conditioning steps to potentials ranging from –140 mV to –35 mV. (B) Steady-state inactivation of the sodium current in the cell in Fig. 1A. Peak sodium current amplitude is normalized to the maximal current given as a function of conditioning step potential (V) in fat-free control solution (open symbols) and after perfusion with 16 µM DHA (filled symbols). Data are fitted with a Boltzmann equation of the form I/Imax= 1/(1 + e((V – Vh)/Vc)), indicating a shift in the potential of half-maximal inactivation Vh of 13 mV and an increase of the slope factor Vc by 1.1 mV. (C) Potency of different fatty acids. The shift in Vh induced by 16 µM of the PUFAs DHA (n = 14), EPA (n = 7), and LA (n = 7), the mono-unsaturated OA (n = 7), and the saturated PA (n = 4) are given and compared to the unspecific time-dependent shift with continued perfusion with BSA-containing control solution (n = 8). Error bars indicate ** *** SEM. Unpaired Student’s t-test significance is indicated as p < 0.01; p < 0.001. (D) Recovery from inactivation. The time-course of the recovery from inactivation was assessed using two 20-ms steps to –25 mV with an interval of varying duration. After complete inactivation by the first step, the sodium current was allowed to recover from inactivation during the interval. The current peak amplitude, normalized to the unconditioned amplitude, is given for the same cell as in Fig. 1A,B as a function of interval duration, for an interval potential of –80 mV (circles) and –70 mV (squares), in control solution (open symbols), and after perfusion with 16 µM DHA. Data are fit with a monoexponential function. The rate of recovery from inactivation is increased with more hyperpolarized interval potentials. DHA shifts this voltage dependence by about 10 mV in this neuron.


Part I / Basic Mechanisms

constant o of 11.6 ± 0.4 ms at –80 mV and of 21.4 ± 0.4 ms at –70 mV (n = 115). Sixteen micromolar DHA slowed down the rate of recovery from inactivation at –80 mV to the same level as at –70 mV in the control solution, indicating a shift in the voltage dependence of recovery from an inactivation of about 10 mV to more hyperpolarized levels. The increase in o at –80 mV induced by 16 µM of the fatty acids was 4.9 ± 0.9 ms for DHA, 2.7 ± 0.8 ms for EPA, 2.1 ± 0.9 ms for LA, –0.2 ± 0.8 ms for OA, and 0.1 ± 0.8 ms for PA. When compared to the small change with continued perfusion with BSA (–1.3 ± 0.7 ms), only the PUFAs DHA, EPA, and LA reduced the rate of recovery from inactivation significantly.

4.2. Sodium Current Activation Is Not Affected The voltage dependence of activation was assessed by determining the sodium conductance at different depolarizing steps, from –120 mV, and fitting the voltage– conductance relationship with a Boltzmann equation (see Vreugdenhil et al., 1996). The potential of half-maximal activation was –31.1 ± 0.4 mV (n = 86). At a concentration of 16 µM, not one of the tested PUFAs affected the voltage dependence of activation. DHA and EPA suppressed the sodium conductance significantly (by 35 ± 7% and 10 ± 6% respectively). This suppressive effect was absent at lower concentrations and was more pronounced in cardiac myocytes at higher concentrations (Leifert et al., 1999).

4.3. Concentration–Effect Relationship The PUFA-induced 6Vh was determined for different concentrations of DHA or EPA. The concentration–effect relationship for DHA (Fig. 5A) and EPA (Fig. 5B) was fit with a Hill equation. With a Hill coefficient of 2, the maximal effect was a shift of –11.2 ± 0.6 mV for DHA and –11.4 ± 0.2 mV for EPA. The concentration of half-maximal effect (EC50) was 2.1 ± 0.3 µM for DHA and 3.7 ± 0.2 µM for EPA. DHA and EPA are much more effective than the anticonvulsant carbamazepine. The 6Vh induced by 15 µM carbamazepine under identical conditions (6Vh= –4.3 mV) (Vreugdenhil & Wadman, 1999) can already be achieved by 1.6 µM DHA or 2.9 µM EPA.

5. EFFECTS OF PUFA ON CALCIUM CURRENTS In addition to an effect on sodium currents, many known anticonvulsant drugs reduce calcium currents as well (Elliott, 1990; Rogawski & Porter, 1990). PUFAs were shown to suppress calcium currents in cardiac myocytes (Xiao et al., 1997). In a parallel study, we tested the effect of DHA and EPA on voltage-dependent calcium currents in acutely isolated CA1 neurons and with similar voltage protocols to assess the voltage dependence of activation and steady-state inactivation of the high-voltage activated calcium current as for the sodium current. For details, see Vreugdenhil et al., 1996; Vreugdenhil & Wadman, 1994. As for the sodium current, the calcium current activation characteristics were not affected by PUFAs, but the calcium current inactivation accelerated and the voltage dependence of the steady-state inactivation shifted to more hyperpolarized levels. The concentration–effect relationship for DHA and EPA are shown in Fig. 5A, B. With a Hill coefficient of 2, the maximal effect of DHA was –6.7 ± 0.4 mV with an EC50 of 2.2 ± 0.4 µM. EPA was clearly less effective and had an EC50 > 15 µM. The monounsaturated fatty acid OA had no significant effect at 20 µM.

Chapter 4 / Actions on Na Channels and Seizures


Fig. 5. Concentration–effect relationships. (A) The concentration–effect relation ship of DHA. The shift in Vh is given as a function of DHA concentration for sodium currents (INa, squares) and calcium currents (ICa, circles) after subtraction of the unspecific time-dependent shift with continued perfusion with BSA-containing control solution. Data are fit with a Hill equation of the form h 6max/(1 + (c/EC50) ), where 6max is the maximal effect, c is the concentration of the drug, EC50 is the concentration of half maximal effect and h is the Hill coefficient. The EC50 was approx 2 µM for both the sodium current and the calcium current. (B) Concentration–effect relationship for EPA. The EC50 for the sodium current was lower than that for the calcium current.

6. EFFECTS OF FATTY ACIDS IN VIVO Yehuda and co-workers were the first to demonstrate that PUFAs can have an anticonvulsant effect in vivo (Yehuda, Carasso & Mostofsky, 1994). They administered a mixture of _-linolenic/linoleic acid in a ratio of 1 : 4 to rats for 3 wk and assessed the protection against acute convulsant doses of pentylenetetrazole (PTZ), repeated subconvulsive doses of PTZ (chemical kindling), in rats made epileptic by a FeCl3 injection in the amygdala and in rats made seizure-prone to acoustic stimulation by repeated injection with p-cresol. In all epilepsy models, the treatment either prevented the occurrence of seizures or increased the threshold for convulsions and diminished the severity and duration.


Part I / Basic Mechanisms

We investigated whether an anticonvulsant effect could be demonstrated after acute doses (Voskuyl, Vreugdenhil, Kang & Leaf, 1998). The cortical stimulation model was used for this purpose, as it allows frequent testing of the threshold for convulsive activity in individual animals and can, therefore, provide the time-course of effect in a single experiment (Voskuyl, Dingemanse & Danhof, 1989; Voskuyl, Hoogerkamp & Danhof, 1992). In this experimental epilepsy model, convulsive activity is induced by electrical stimulation of the motor area of the cortex. Because the bilateral, cortical electrodes are chronically implanted, the stimulation can take place via a cable in otherwise freely moving animals. A current pulse train is used that slowly increases in amplitude with each pulse. This results in a progressive pattern of behavioral, convulsive activity that can be seen during stimulation. Within this pattern two thresholds have been defined, the threshold for localized seizure activity (TLS) and, with continued stimulation, the threshold for generalized seizure activity (TGS). If stimulation is stopped before the TLS is reached (typically the start of clonic activity of the forelimbs), convulsive activity is immediately aborted. Crossing of the TGS is characterized by self-sustained seizure activity, continuing for 10–40 s after stimulation has stopped. “Postictal” (i.e., postseizure) threshold increases are avoided if the TGS is not crossed. Testing a great number of anticonvulsant drugs, it was found that drugs could increase the TLS, the difference between the TLS and TGS, or both (Della Paschoa, Hoogerkamp, Edelbroek, Voskuyl & Danhof, 2000; Hoogerkamp et al., 1996; Hoogerkamp, Vis, Danhof & Voskuyl, 1994). (It should be noted that if a drug increases the TLS, the TGS rises by necessity, because it is reached only when the stimulus intensity has passed the TLS.) Phenytoin and carbamazepine preferentially affect the TGS, but valproate increases the TLS primarily. Fresh emulsions of fatty acids made by mixing a small volume of concentrated fatty acid in alcohol, with 2 mL of physiological saline containing BSA and subsequent sonication, were administered via a previously implanted jugular vein canula. As pilot experiments had shown that bolus injections were ineffective, the emulsion was infused over a period of 30 min. We tested DHA, EPA, LA, OA, and vehicle. DHA and EPA moderately, but highly significantly, increased the TLS and TGS (Table 1). Both thresholds started to rise at the end of the infusion and reached a maximum after 7 h. The next day, the thresholds had partly returned to baseline. LA and OA had similar but smaller effects, but OA did not change the TLS. The vehicle had no effect at all. The time-courses of the threshold changes are illustrated in Fig. 6. Although it was less conspicuous than in cardiomyocytes and hippocampal neurons, the efficacy of the fatty acids appeared to correlate with the number of double bonds in these experiments as well. In comparison with phenytoin and carbamazepine, which induce changes in TGS of several hundred microampere, the effects of the fatty acids were rather small. However, a more conspicuous difference was the slow increase in threshold over a period of several hours. With intravenous administration of phenytoin, carbamazepine, or valproate, the maximal effect is reached almost immediately after injection, after which the threshold returns to baseline in 4–6 h. The pharmaco*kinetics were not investigated in detail, but the time-course of effect certainly did not follow the plasma concentration. Blood samples taken from some rats that were treated with DHA or EPA indicated that the plasma concentration was maximal at the end of the infusion and dropped to undetectable levels after 6 h (probably already after 3 h).

Chapter 4 / Actions on Na Channels and Seizures


Table 1 Increase in the Thresholds for Localized (TLS) and for Generalized Seizure Activity (TGS), 6 h After the Start of Infusion


TLS (µA)

TGS (µA)


77 ± 17 73 ± 13 49 ± 10 16 ± 7 15 ± 15

130 ± 19 125 ± 20 75 ± 11 54 ± 9 8 ± 11

7 7 9 9 6

Note: Values are expressed as mean ± sem. One-way analysis of variance indicated a highly significant difference between groups, both for the TLS and TGS (p < 0.001). Subsequent multiple range tests, using the leastsquares difference procedure, demonstrated that, with respect to the TLS, EPA, DHA, and LA differed from OA and control. With respect to the TGS, EPA and DHA differed from LA and OA and the latter two, in turn, from the control.

Fig. 6. Threshold increase in the cortical stimulation model. (A) Change in TLS induced by 30-min iv infusion of 40 µmol fatty acid, shown for DHA (n = 7), LA (n = 9), or vehicle (n = 6). The TLS is expressed as the increase with respect to the mean value during the preinfusion period. The infusion is indicated by black bar. Data are mean ± sem. (B) Change in TGS, measured simultaneously with the TLS in the same animals.

7. POTENTIAL ROLE OF PUFAS IN THE TREATMENT OF EPILEPSY In conclusion, in vitro studies on hippocampal neurons and in vivo studies in experimental epilepsy models have shown that PUFAs, especially DHA, in principle can have an anticonvulsant action. At low-micromolar concentrations, they exhibit the same effects as well-known antiepileptic drugs like carbamazepine, phenytoin, and valproic acid, in that they reduce sodium currents by shifting the voltage dependence of the inactivation in the hyperpolarizing direction and delaying the recovery of inactivation, without effect on activation. The manner in which the sodium current is reduced should increase the threshold for action potential generation and limit the firing rate, which has, indeed, been shown recently for CA1 and CA3 hippocampal neurons (Xiao & Li, 1999). The selective


Part I / Basic Mechanisms

effect appears to be positively correlated to the number of unsaturated bonds and to depend on the presence of a free carboxylic acid group. Furthermore, calcium currents are modified in the same way. This may be expected to reduce calcium-dependent neurotransmission and postsynaptic calcium influx in neurons. The combined action on sodium and calcium currents is characteristic for drugs effective against partial seizures (Elliott, 1990; Rogawski & Porter, 1990) and is already achieved by low-micromolar concentrations. Although it is tempting to attribute the anticonvulsant effect in vivo to the direct effects of PUFAs on sodium and calcium channels, there is, as yet, no proof that this is actually the case. The in vivo studies reveal some complicating factors. The two classical screening models for anticonvulsant action are the maximal electroshock (MES) test and the subcutaneous pentylenetetrazole (PTZ) test. The majority of the presently available anticonvulsants have been initially identified with one of these tests and often they show efficacy in only one of the tests. Phenytoin and carbamazepine, which presumably act primarily by suppressing sodium currents, are the two prototype compounds that show selective activity in the MES test, but not in the PTZ test. However, so far, the anticonvulsant action of PUFAs has not yet been demonstrated in the MES test, but only in the PTZ test. The selective effect of carbamazepine and phenytoin in the MES test is paralleled in the cortical stimulation model by a strong and immediate effect on the TGS, whereas the effect of PUFAs was moderate and reached a peak only after several hours. Furthermore, in the studies by Yehuda and co-workers, the fatty acids appeared to become effective only after the animals had been administered a fatty acid mixture for 3 wk. Although the authors did not discuss this point, it seems that this time-consuming design was chosen because acute doses were not effective. This raises the question of whether the administered PUFAs actually reach their targets and whether reduction of sodium and calcium currents indeed underlies the anticonvulsant action or that other mechanisms are involved. For example, a new class of “background” potassium channels has recently been discovered that is found only in the central nervous system and is exclusively activated by PUFAs (Fink et al., 1998). Activation of these channels would be indistinguishable from the aforementioned effects on Na+ channels. A role for these K+ channels in the protective effects of PUFAs against kainic acid-induced epileptic seizures has been suggested (Lauritzen et al., 2000). It is unlikely that PUFAs do not reach the brain, because they can pass the blood-brain barrier rapidly. Thus, it seems more likely that they are quickly stored somewhere, so that the free concentration in the extracellular space remains too low, at least initially, to interact with ion channels. From these stores, they can be released more slowly, accounting for a delayed effect. One possibility is that they are incorporated in neuronal membranes and later liberated by activity-dependent lipase (Dumuis, Sebben, Haynes, Pin & Bockaert, 1988). Alternatively, astroglial cells may buffer the rapid rise in fatty acid concentration and later release them at a much slower rate to the immediate vicinity of the neurons. The possibility of other mechanisms involved in the action of PUFAs should be considered. Some studies suggest that a reduction in sodium current may also be accomplished by activation of protein kinase C (Godoy & Cukierman, 1994; Linden & Routtenberg, 1989), possibly by modulating the degree of phosphorylation of the sodium channel (see Catterall, 1999). As an example of the complexity of correlating mechanisms observed at the molecular and cellular levels to the in vivo anticonvulsant action,

Chapter 4 / Actions on Na Channels and Seizures


Fig. 7. Synergistic action of PUFAs. (A) Voltage dependence of steady-state inactivation of the sodium current in the control solution (open symbols) after perfusion with 15 µM carbamazepine (CBZ, gray symbols) and after the addition of 0.5 µM EPA (black symbols). Data are fit with a Boltzmann equation. Like PUFAs, CBZ induces a shift in Vh and increases the slope factor Vc. The addition of 0.5 µM EPA induces an extra shift in Vh, but without affecting the Vc, whereas 0.5 µM EPA alone has no discernible effect on Vh. (B) The shift in Vh induced by addition of 0.5 µM EPA on top of 15 µM CBZ is given against the shift induced by 15 µM CBZ alone for CA1 neurons from healthy rats (open circles), for CA1 neurons from patients with pharmaco-resistant temporal lobe epilepsy (filled circles), and for neocortical neurons from epilepsy patients (squares). The potentiating effect of subthreshold concentrations of EPA is similar for all groups.

valproic acid can serve as an example. Although the typical effects on sodium channel inactivation and lowering of repetitive firing have been described in cultured neurons (McLean & Macdonald, 1986b; Van den Berg, Kok & Voskuyl, 1993), the expected changes in firing behavior could not be confirmed in hippocampal slices (Albus & Williamson, 1998). Thus, it remains an open question as to what extent the effects of valproate on sodium channels contribute to the antiepileptic action. Other proposed mechanisms are increased GABA-mediated inhibition via stimulation of GABA synthesis and release, reduction of the release of the proconvulsant amino acid a-hydroxybutyric acid, and attenuation of synaptic excitation mediated by N-methyl-D-aspartate (NMDA) receptors (see Löscher, 1999 for a review). DHA has also been reported to affect GABAmediated synaptic inhibition (Hamano, Nabekura, Nishikawa & Ogawa, 1996) and NMDA responses (Nishikawa, Kimura & Akaike, 1994). However, the actions were in the opposite direction and would be expected to counteract an anticonvulsant effect via sodium and calcium channels (this could, in part, be responsible for the moderate effect in vivo). It illustrates that much work still needs to be done to correlate molecular mechanisms of PUFAs to responses of the intact organism and differences between different members of this class of compounds should be considered.

8. SYNERGISTIC ACTION OF PUFA Nevertheless, there are some exciting avenues to explore, of which one is the already mentioned possible role of PUFAs in the ketogenic diet, as is covered in Chapter 17. Another aspect is the enhancement of the action of conventional drugs by nanomolar concentrations of PUFAs. This interesting phenomenon was observed at very low doses


Part I / Basic Mechanisms

of PUFAs. Figure 7A gives an example where 15 µM carbamazepine shifted the voltage dependence of the sodium current steady-state inactivation by 5.4 mV to more hyperpolarized levels. The addition of 0.5 µM EPA almost doubled this shift, without affecting the slope of the curve. When tested in isolation, this dose of EPA was without discernible effect (see Fig. 5B). This overadditive or synergistic effect was also found for DHA (not shown). In a similar fashion, subthreshold levels of the two-branched fatty acid valproic acid (100 µM) boosted the effect of 15 µM carbamazepine or 10 µM phenytoin on the Vh of sodium currents (M. Vreugdenhil, unpublished data). This synergistic action of valproic acid in vitro was confirmed in vivo in the cortical stimulation model for anticonvulsant action of drugs: subthreshold doses of valproic acid significantly boosted the anticonvulsant effect of phenytoin (Della Paschoa, Kruk, Hamstra, Voskuyl & Danhof, 1998). We determined the synergistic effect of 0.5 µM EPA on 15 µM carbamazepine in CA1 neurons from healthy rats, in CA1 neurons isolated from the hippocampus surgically removed from patients with intractable temporal lobe epilepsy, and in neocortical pyramidal neurons from the same patients (for details on methods, see Vreugdenhil & Wadman, 1999). Figure 7B shows a clear relation between the shift in the inactivation curve induced by 15 µM carbamazepine and the additional shift induced on top of 0.5 µM EPA. This suggests that the effect of a therapeutically relevant dose of carbamazepine can be boosted by subthreshold levels of PUFAs. This knowledge might be relevant for those epilepsy patients who need intolerably high doses of antiepileptic drugs to control their seizures. The shift in inactivation curve induced by 15 µM carbamazepine in CA1 neurons from the hippocampus of epileptic patients with hippocampal sclerosis (Vreugdenhil et al., 2000), as well as in CA1 neurons from chronic epileptic (kindled) rats (Vreugdenhil & Wadman, 1999), was smaller than that in CA1 cells from healthy rats (Fig. 7B). Instead of doubling the dose of carbamazepine, very low concentrations of PUFAs might provide the shift required for seizure control. The other way around, the success of treatment with antiepileptic drugs like carbamazepine might depend on PUFA levels in the brain, which can vary considerably, according to diet (de Logeril et al., 1994) and metabolism. This synergistic action may provide a basis for developing rational polytherapy of PUFAs and carbamazepine or phenytoin.

REFERENCES Commission on Classification and Terminology of the International League Against Epilepsy. Proposal for revised classification of epilepsies and epileptic syndromes. Epilepsia 1989;30:389–399. Albus H, Williamson R. Electrophysiologic analysis of the actions of valproate on pyramidal neurons in the rat hippocampal slice. Epilepsia 1998; 39(2):124–139. Catterall WA. Molecular properties of brain sodium channels: an important target for anticonvulsant drugs. In: Delgado-Escueta AV,Wilson WA, Olsen, RW, Porter RA, eds. Jasper’s Masic Mechanisms of the Epilepsies, 3rd ed. Advances in Neurology, Vol. 79. Lippincott Williams & Wilkins, Philadelphia, 1999, pp. 441–456. de Logeril M, Renaud S, Mamelle N, Salen P, Martin J-L, Monjaud I, et al. Mediterranean alpha-linolenic acid-rich diet in secondary prevention of coronary heart disease. Lancet 1994; 143:1454–1459. Della Paschoa OE, Hoogerkamp A, Edelbroek PM, Voskuyl RA, Danhof M. Pharmaco*kinetic-pharmacodynamic correlation of lamotrigine, flunarizine, loreclezole, CGP40116 and CGP39551 in the cortical stimulation model. Epilepsy Res 2000; 40:41–52. Della Paschoa OE, Kruk MR, Hamstra R, Voskuyl RA, Danhof M. Pharmacodynamic interaction between phenytoin and sodium valproate changes seizure thresholds and pattern. Br J Pharmacol 1998; 125(5):997–1004.

Chapter 4 / Actions on Na Channels and Seizures


Dumuis A, Sebben M, Haynes L, Pin JP, Bockaert J. NMDA receptors activate the arachidonic acid cascade in striatal neurons. Nature 1988; 336:68–70. Elliott P. Action of antiepileptic and anaesthetic drugs on Na- and Ca-spikes in mammalian non-myelinated axons. Eur J Pharmacol 1990; 175:155–163. Fink M, Lesage F, Duprat F, Heurteaux C, Reyes R, Fosset M, Lazdunski M. A neuronal two P domain K+ channel stimulated by arachidonic acid and polyunsaturated fatty acids. The EMBO Journal 1998; 17:3297–3308. Godoy CM, Cukierman S. Multiple effects of protein kinase C activators on Na+ currents in mouse neuroblastoma cells. J Membr Biol 1994; 140(2):101–110. Hamano H, Nabekura J, Nishikawa M, Ogawa T. Docosahexaenoic acid reduces GABA response in substantia nigra neuron of rat. J Neurophysiol 1996; 75:1264–1270. Hock CE, Beck LD, Bodine LC, Reibel DK. Influence of dietary n-3 fatty acids on myocardial ischemia and reperfusion. Am J Physiol 1990; 259:H1518–H1526. Hoogerkamp A, Arends RHGP, Bomers AM, Mandema JW, Voskuyl RA, Danhof M. Pharmaco*kinetic/ pharmacodynamic relationship of benzodiazepines in the direct cortical stimulation model of anticonvulsant effect. J Pharmacol Exp Ther 1996; 279(2):803–812. Hoogerkamp A, Vis PW, Danhof M, Voskuyl RA. Characterization of the pharmacodynamics of several antiepileptic drugs in a direct cortical stimulation model of anticonvulsant effect in the rat. J Pharmacol Exp Ther 1994; 269:521–528. Kang JX, Leaf A. Effects of long-chain polyunsaturated fatty acids on the contraction of neonatal rat cardiac myocytes. Proc Natl Acad Sci USA 1994; 91(21):9886–9890. Lauritzen I, Blondeau N. Heurteaux C, Widmann C, Romey G, Lazdunski M. Polyunsaturated fatty acids are potent neuroprotectors. The EMBO Journal 2000; 19:1784–1793. Leaf A, Kang JX, Xiao Y-F, Billman GE, Voskuyl RA. The antiarrhythmic and anticonvulsant effects of dietary N-3 fatty acids. J Membr Biol 1999; 172:1–11. Leaf A, Kang JX, Xiao Y-F, Billman GE, Voskuyl RA. Experimental studies on antiarrhythmic and antiseizure effects of polyunsaturated fatty acids in excitable tissue. J Nutr Biochem 1999; 10:440–448. Leifert WR, McMurchie EJ, Saint DA. Inhibition of cardiac sodium currents in adult rat myocytes by n-3 polyunsaturated fatty acids. J Physiol 1999; 520:671–679. Linden DJ, Routtenberg A. cis-Fatty acids, which activate protein kinase C, attenuate Na+ and Ca2+ currents in mouse neuroblastoma cells. J Physiol 1989; 419:95–119. Löscher W. New visions in the pharmacology of anticonvulsion. Eur J Pharmacol 1998; 342:1–13. Löscher W. Valproate: a reappraisal of its pharmacodynamic properties and mechanisms of action. Prog Neurobiol 1999; 58:31–59. McLean MJ, Macdonald RL. Carbamazepine and 10,11-epoxycarbamazepine produce use and voltage-dependent limitation of rapidly firing action potentials of mouse central neurons in cell culture. J Pharmacol Exp Ther 1986; 238:727–738. McLean MJ, Macdonald RL. Sodium valproate, but not ethosuximide, produces use- and voltage-dependent limitation of high frequency repetitive firing of action potentials of mouse central neurons in cell culture. J Pharmacol Exp Ther 1986; 237(3):1001–1011. Meier CL, Dudek FE. Spontaneous and stimulation-induced synchronized burst afterdischarges in the isolated CA1 of kainate-treated rats. J Neurophysiol 1996; 76:2231–2239. Nishikawa M, Kimura S, Akaike N. Facilitatory effect of docosahexaenoic acid on N-methyl-D-aspartate response in pyramidal neurones of rat cerebral cortex. J Physiol 1994; 475:83–93. Park CC, Ahmed Z. Alterations of plasma membrane fatty acid composition modify the kinetics of Na+ current in cultured rat diencephalic neurons. Brain Res 1992; 570:75–84. Ragsdale DS, Avoli M. Sodium channels as molecular targets for antiepileptic drugs. Brain Res Rev 1998; 26:16–28. Rogawski MA, Porter RJ. Antiepileptic drugs: pharmacological mechanisms and clinical efficacy with consideration of promising developmental stage compounds. Pharmacol Rev 1990; 42(3): 223–286. Schwartz J, Grigat G. Phenytoin and carbamazepine: potential- and frequency-dependent block of sodium currents in mammalian myelinated nerve fibres. Epilepsia 1989; 30:286–294. Takenaka T, Horie H, Hori H. Effects of fatty acids on membrane currents in the squid giant axon. J Membr Biol 1987; 95:113–120. Takenaka T, Horie H, Hori H, Kawakami T. Effects of arachidonic acid and the other long-chain fatty acids on the membrane currents in the squid giant axon. J Membr Biol 1988; 106:141–147.


Part I / Basic Mechanisms

Taylor CP, Narasimhan LS. Sodium channels and therapy of central nervous system diseases. Adv Pharmacol 1997; 39:47–98. Van den Berg RJ, Kok P, Voskuyl RA. Valproate and sodium currents in cultured hippocampal neurons. Exp Brain Res 1993; 93:279–287. Voskuyl RA, Dingemanse J, Danhof M. Determination of the threshold for convulsions by direct cortical stimulation. Epilepsy Res 1989; 3:120–129. Voskuyl RA, Hoogerkamp A, Danhof M. Properties of the convulsive threshold determined by direct cortical stimulation in rats. Epilepsy Res 1992; 12:111–120. Voskuyl RA, Vreugdenhil M, Kang JX, Leaf A. Anticonvulsant effect of polyunsaturated fatty acids in rats, using the cortical stimulation model. Eur J Pharmacol 1998; 341(2–3):145–152. Vreugdenhil M, Bruehl C, Voskuyl RA, Kang JX, Leaf A, Wadman WJ. Polyunsaturated fatty acids modulate sodium and calcium currents in CA1 neurons. Proc Natl Acad Sci USA 1996; 93:12,559–12,563. Vreugdenhil M, Wadman WJ. Kindling-induced long-lasting enhancement of calcium current in hippocampal CA1 area of the rat: relation to calcium-dependent inactivation. Neuroscience 1994; 59(1):105–114. Vreugdenhil M, Wadman WJ. Modulation of sodium currents in rat CA1 neurons by carbamazepine and valproate after kindling epileptogenesis. Epilepsia 1999; 40:1512–1522. White HS. Comparative anticonvulsant and mechanistic profile of the established and newer antiepileptic drugs. Epilepsia 1999; 40(Suppl. 5):S2–S10. Xiao YF, Gomez AM, Morgan JP, Lederer WJ, Leaf A. Suppression of voltage-gated L-type Ca2+ currents by polyunsaturated fatty acids in adult and neonatal rat ventricular myocytes. Proc Natl Acad Sci USA 1997; 94:4182–4187. Xiao YF, Kang JX, Morgan JP, Leaf A. Blocking effects of polyunsaturated fatty acids on Na+ channels of neonatal rat ventricular myocytes. Proc Natl Acad Sci USA 1995; 92(November), 11,000–11,004. Xiao YF, Li X. Polyunsaturated fatty acids modify mouse hippocampal neuronal excitability during excitotoxic or convulsant stimulation. Brain Res 1999; 846:112–121. Xiao, Y.-F., Wright SN, Wang GK, Morgan JO, Leaf A. N-3 fatty acids suppress voltage-gated Na+ currents in HEK293t cells transfected with the _-subunit of the human cardiac Na+ channel. Proc Natl Acad Sci USA 1998; 95:2680–2685. Yehuda S, Carasso RL, Mostofsky DI. Essential fatty acid preparation (SR-3) raises the seizure threshold in rats. Eur J Pharmacol 1994; 254:193–198.

Chapter 5 / Dietary Fats and Aging Immune System



Role of Dietary Fats and Exercise in Immune Functions and Aging Jaya T.Venkatraman and David Pendergast

1. INTRODUCTION Immunosenescence is a complex remodeling of the immune system that may contribute significantly to morbidity and mortality in the elderly. Much evidence suggests an association between immune function and well-being during aging and longevity. Despite several studies on the immune system in the elderly, little is known of the biological basis of immunosenescence in humans. Undoubtedly, some diseases to which the elderly are particularly susceptible, such as infections and autoimmune and neoplastic pathologies, include dysregulation of several immune functions in their pathogenesis. On the other hand, recent studies in healthy centenarians suggest that the immunological changes observed during aging are consistent with a reshaping, rather than a generalized deterioration, of the main immune functions. The infection rate and severity increase with aging as a result of decrease in immune function with aging. The primary changes of the aged host are in the T-lymphocytes, perhaps because of the involution of the thymus. Secondary changes (environmental) may be the result of changes in diet, drug intake, physical activity, and so forth or, alternatively, the result of underlying diseases (Wick and Grubeck-Loebenstein, 1997). There is a paradoxical increase in autoimmunity, and the responsiveness to exogenous antigens and tumors are reduced and may play a proinflammatory role in the development of many pathological conditions (Weyand et al., 1998). Recurrent stress, infections, and inflammation over the life-span play a significant role in producing age-associated changes in all systems. Genetic selection for infections has been implicated in coronary heart disease (McCann et al., 1998), early onset of Parkinson’s following influenza encephalitis, and Alzheimer’s disease (Stoessl, 1999). The restoration of immune functions of the aged individuals is possible and might be beneficial for them to cope with various diseases associated with aging (Hirokawa, 1997). Physiological thymic atrophy is controlled by both extrathymic and intrathymic factors and is not a totally irreversible process. The process of thymic atrophy might be explained by a further understanding of the relationship between the neuroendocrine and the immune systems (Hirokawa et al., 1994). Although the most obvious age-related structural alteration of the immune system occurs in the thymus, the role of thymic involution in immunosenescence is still not well understood. From: Fatty Acids: Physiological and Behavioral Functions Edited by: D. Mostofsky, S. Yehuda, and N. Salem Jr. © Humana Press Inc., Totowa, NJ



Part I / Basic Mechanisms

1.1. Immune Theory of Aging Original immunologic theory of aging suggests that aging in mammals is a selfdestructive process leading to a decline in immune response. The failure in immune homeostasis is associated with a consequent rise in autoimmunity (Walford, 1987). Agerelated phenomena such as increased prevalence of autoantibodies and monoclonal immunoglobulins reflect dysregulation of the senescent immune system rather than a simple decline in responsiveness. The age-related changes primarily occur in the T-celldependent immune system and are associated with increased susceptibility to infections and incidence of autoimmune phenomena in the elderly. One of the characteristics of all somatic cells is a finite life-span. Cells may proliferate until they reach a point, after which they can no longer produce daughter cells. This observation is central to the clonal exhaustion hypothesis, a mechanism cited to explain age-associated immune dysfunction. In this hypothesis, repeated division of lymphocytes leads to a replicative limit, after which the cells enter the senescent phase but are not lost from the pool of T-cells. Advancing age would then be associated with an increase in the number of T-cells that are unable to proliferate to a stimulus that induces a proliferative response in T-cells from younger individuals.

2. IMMUNE SYSTEM CHANGES WITH AGING 2.1. Immune Function 2.1.1. AGING AND IMMUNE CELL SUBSETS The innate immunity is preserved over the life-span. Immune changes, dysregulations, mainly affect acquired immunity and lead to a gradual increase in T-helper (Th) 2/Thelper 1 cells. This change is the result initially of decreased thymic function, and later of accumulative antigen pressure over the life-span. T-Cell subset distribution in peripheral lymphocytes changes with age to a higher memory/naive cell ratio, accompanied by changes in cytokine production patterns (Fernandes and Venkatraman, 1993; Fernandes et al., 1997; Thoman and Weigle, 1989; Venkatraman and Fernandes, 1994; Venkatraman et al., 1994). Proliferative capacity of T-cells declines with age (Venkatraman & Fernandes, 1997; Makinodan, 1998; Makinodan et al., 1987). The composition of T-cell subsets in the periphery gradually change with age, resulting in the alteration of T-cell functions in the elderly. There appears to be a macrophage–lymphocyte disequilibrium in aged persons (Lesourd, 1999). There is a shift in the equilibrium of peripheral T- and B-lymphocyte subsets, T helper 1 subset (Th1) to Th2 and CD5– to CD5+ cells with aging. T-Cell responses are more affected than B-cell responses that may result in T-cellmediated dysregulation of antibody responses and low affinity and self-reactive antibodies (Doria and Frasca 1997). T-Cell activity may be restricted to immune surveillance of neoplastic transformation. Even the B-cells exhibit intrinsic defects and natural-killer (NK) cells have a profound loss of activity. Aging leads to replacement of virgin cells by memory cells and to the accumulation of cells with defects in signal transduction. The aging immune system is characterized by a progressive decline in the responsiveness to exogenous antigens and tumors in combination with a paradoxical increase in autoimmunity. From a clinical viewpoint, deficiencies in antibody responses to exogenous antigens, such as vaccines, have a major impact and may reflect intrinsic B-cell defects or altered performance of helper T-cells. Aging is associated with the emergence

Chapter 5 / Dietary Fats and Aging Immune System


of an unusual CD4 T-cell subset characterized by the loss of CD28 expression (Weyand et al., 1998). CD28 is the major costimulatory molecule required to complement signaling through the antigen receptor for complete T-cell activation. CD4+ CD28– T-cells are long-lived, typically undergo clonal expansion in vivo, and react to autoantigens in vitro. Despite the deficiency of CD28, these unusual T-cells remain functionally active and produce high concentrations of interferon-a (IFN-a) and interleukin-2 (IL-2). The loss of CD28 expression is correlated with a lack of CD40 ligand expression, rendering these CD4 T-cells incapable of promoting B-cell differentiation and immunoglobulin secretion, thus aberrations in immune responsiveness (Weyand et al., 1998). Aging-related accumulation of CD4+ CD28– T-cells should result in an immune compartment skewed toward autoreactive responses and away from the generation of high-affinity B-cell responses against exogenous antigens. We propose that the emergence of CD28-deficient CD4 T-cells in the elderly can partially explain age-specific aberrations in immune responsiveness. In a cohort of apparently healthy and well-nourished elderly women, total T (CD3+), T-helper (CD4+), or T-cytotoxic (CD8+) cell number, NK cell number, cytotoxicity, phagocytosis, and subsequent oxidative burst were similar to values in young women. However, they had lower T-cell proliferation responses and significantly reduced response to phytohemagglutinin (Krause et al., 1999). T-cells show the largest agerelated differences in distribution and function with aging with thymus involution as the apparent underlying cause (Shinkai et al., 1998). 2.1.2. AGING AND CYTOKINES Complex remodeling of cytokine production is likely to be a characteristic of immunosenescence. Cytokines are produced by various immunocompetent cells in response to appropriate stimuli, are able to mediate many immune functions, orchestrate the immune system continuously, and act as molecular signals between immunocompetent cells. An excessive or insufficient production of cytokines may contribute to infectious, immunological, and inflammatory diseases. They act via specific receptor sites on cytokine-secreting cells (autocrine action) or immediately adjacent immune cells (paracrine action). The production of IL-2 is decreased, with a decrease of total T-cell count, and often with changes in T-cell subsets and proliferative response to mitogens during aging. T-cell-dependent functions are most dramatically compromised, which is most likely a result of age-related involution of the thymus gland. Defects in T-cell proliferative capacity/responsiveness, IL-2 production and receptor expression, signal transduction, and cytotoxicity are frequently cited problems associated with immunosenescence (Cinader et al., 1993). As a consequence of their differentiated state, memory T-cells can express higher levels and a greater variety of lymphokines than naive T-cells; the former more efficiently generate and regulate humoral and cell-mediated immune responses to recall antigens than naive T-cells (Ernst et al., 1990). Transforming growth factor-` (TGF-`) for instance, is known to have immunosuppressive properties. A balance between proflammatory and anti-inflammatory cytokines derived from Th1 or Th2 cells are very essential for the maintenance of a sound immune system as they stimulate growth, differentiation and functional development of immune cells. Following a primary encounter with an antigen, naive T-cells express early-response genes, including those that encode the T-cell growth factor, IL-2 and IL-2 receptor chains. IL-2 functions through high-affinity receptors to drive activated cells to proliferate and differentiate into effector T-cells that mediate primary humoral or cell-mediated


Part I / Basic Mechanisms

immune responses (Ernst et al., 1995). In the process, a large number of differentiated, antigen-specific T-cells (memory cells) is generated. The memory cells are in the differentiated state and can express cytokines that naive T-cells cannot. The naive-to-memory cell conversion is accompanied by quantitative changes in the expressed levels of several membrane molecules that can alter cellular function (Ernst et al., 1995). The age-dependent increase in memory T-cells, which are well differentiated and unable to mutate their TCR genes somatically, could result in a peripheral pool of blood cells that respond well to previously encountered antigens but cannot recognize and respond to new antigens. The increased pool of memory T-cells could represent a greater risk for dysregulation by perhaps overproducing certain cytokines, which may suppress appropriate immune responses or amplify inappropriate ones (Ernst et al., 1995). Lymphocytes of the elderly show decreased proliferation after induction with mitogens, decreased release of IL-2, soluble IL-2R, IFN-a, and other Th1 cytokines and increase in Th2 cytokines such as IL-4 and IL-10, suggesting a dysregulation of the Th1/Th2 system in the elderly (Rink et al., 1998). Cytokines and their antagonists are significant factors in host responses to infections and inflammatory stimuli and the elderly have higher levels of cytokine antagonists IL-1RA and sTNF-R, and neopterin (produced by activated macrophages and monocytes) in their plasma and IL-2 production by PBMN cells is decreased (Catania et al., 1997). 2.1.3. NEUROENDOCRINE SYSTEM Numerous interactions exist among the nervous, endocrine, and immune systems and are mediated by neurotransmitters, hormones, and cytokines. Aging is associated with enhanced responsiveness of the T-cell compartment and alterations in temporal architecture of the neuroendocrine-immune system (Mazzoccoli et al., 1997). There are relevant integrations between pituitary–thyroid hormones and immune factors favoring the development and maintenance of both thymic and peripheral immune efficiency. The GHinsulin-like growth factor (IGF)-I axis is dysregulated in aging, in catabolic states, and in critical illness. The pineal gland seems to regulate, via circadian secretion of melatonin, all basic hormonal functions and immunity (Pierpaoli, 1998). Studies with in vivo models have shown that this fundamental role of the pineal gland decays during aging. Melatonin is a ubiquitous molecule and can be found in a large variety of cells and tissues. Binding sites and “receptors” have been identified in many tissues and cells of the neuroendocrine and immune system. 2.1.4. APOPTOSIS AND AGING The progressive decline in immune response and increased incidence of autoimmune phenomenoa might be the result of modified cellular mechanisms affecting the immune system in the course of aging. The apoptotic deletion of activated T-cells has been proposed as the key mechanism to maintain T-cell homeostasis, and in this respect, CD95 (Fas antigen) seems to play a major role in this course of events (Potestio et al., 1998). Immune senescence is a sum of dysregulations of the immune system and its interaction with other systems. Two predominent features of immune senescence are altered T-cell phenotype and reduced T-cell response (Mountz et al., 1997). Cell lines derived from human premature aging diseases have a higher sensitivity to Fas-mediated apoptosis. Data indicate that, in vivo, there is a gradual decrease in apoptosis with aging, resulting in the accumulation of senescent T-cells with increased DNA damage. Lymphocytes

Chapter 5 / Dietary Fats and Aging Immune System


from elderly have an increased expression of CD95 and enhanced apoptosis (Potestio et al., 1998). Recent evidence suggests that apoptotic deletion of activated mature lymphocytes is an essential physiological process implicated in both the regulation of the immune response and the control of the overall number of immunocompetent cells. During the course of aging, numerous alterations of these signaling pathways may shift the balance toward cell death (Phelouzat et al., 1996). 2.1.5. AGING AND TRANSMEMBRANE SIGNALING Aging is also associated with alterations in transmembrane signaling and decreased GTPase activity in lymphocytes and granulocytes. A defective protein kinase C (PKC)_ translocation and a decreased tyrosine kinase activity following TCR or CD3 stimulation in T-cells from the elderly might also contribute to alteration in immune responses in the elderly. The activation of other protein kinases after T-cell activation with PHA (p56 lck, JAK kinases) was also affected by aging, suggesting a slower activation and a lower degree of phosphorylation. The src family of protein tyrosine kinase p56 lck is expressed in T-cells, NK cells, and lymphoid cell lines (Guidi et al., 1998). P56 co-precipitates with CD4 and CD8 molecules, CD2 and CD28, and the `-chain of IL-2 receptors. These kinases have a crucial role in T-cell activation. The protein expression and degree of phosphorylation are reduced in elderly subjects compared to adult controls, suggesting that aging may lead to a drastic impairment in the ability of PBL to be efficiently activated by mitogens or costimuli (both proliferation and IL-2 production) in the elderly (Guidi et al., 1998). Early events associated with transmembrane signaling, such as PKC translocation, IP3 generation, and Ca2+ mobilization, are impaired by aging, whereas PLC-a1 activity seems to be unaffected. This suggests that the signaling steps upstream of PLC-a1 activation, namely thyrosine phosphorylation, may be defective in the elderly. 2.1.6. AGING AND TELOMERASES Normal human cells undergo a finite number of cell divisions and ultimately enter a nondividing state called replicate senescence. Telomerase shortening has been proposed as the molecular clock that triggers senescence. A causal relationship seems to exist between telomerase shortening and in vitro cellular senescence (Bodnar et al., 1998). The reduction of proliferative capacity of cells from elderly donors and patients with premature aging syndromes and the accumulation in vivo of senescent cells with altered patterns of gene expression implicate cellular senescence in aging and age-related pathologies. A common finding in different cells suffering replicative senescence is the shortening of the telomers (Solana and Pawelec, 1998). Telomers are essential genetic elements that stabilize chromosome ends. The synthesis of these elements is controlled by telomerase, an enzyme that synthesizes new telomeric DNA, thus compensating for the loss that occurs as a result of the “end-replication problem” inherent in DNA replication. Some tumor-suppressor genes (p21, p53, Rb) have been reported to accumulate in senescent cells. Escape from senescence is greatly enhanced in cells that have lost Rb or p16, and both alterations behave redundantly with respect to each other. 2.1.7. AGING AND THYMUS Thymic regrowth and reactivation of thymic endocrine activity may occur in older animals by different endocrinological or nutritional manipulations than in young animals. Intrathymic transplantation of pineal gland or treatment with melatonin, implantation of a growth hormone (GH) secreting tumor cell line or treatment with exogenous


Part I / Basic Mechanisms

GH, exogenous luteinizing hormone-releasing hormone (LH-RH), exogenous thyroxine or triiodothyronine, and nutritional interventions such as arginine or zinc supplementation have shown potential (Fabris et al., 1997). These data strongly suggest that thymic involution is a phenomenon secondary to age-related alterations in neuroendocrine– thymus interactions and it is the disruption of such interactions in old age that is responsible for age-associated dysfunction. The effect of GH, thyroid hormones, and LH-RH may be the result of the presence of the specific hormone receptors on thymic epithelial cells supposed to produce thymic peptides. Melatonin or other pineal factors may also act through specific receptors, but experimental evidence is still lacking. The role of zinc, whose turnover is usually reduced in old age, may range from reactivation of zincdependent enzymes, required for both cell proliferation and apoptosis, to the reactivation of thymulin, a zinc-dependent thymic hormone. Recent preliminary data obtained both in animal and human studies suggest that the endocrinological manipulations are capable of restoring thymic activity in old age and may act also by normalizing the altered zinc pool (Fabris et al., 1997). A significant age-related decrease in RAG-1 and RAG-2 expression has been reported in thymocytes of mice aged-12 mo and over compared to young mice (Yehuda et al., 1998). Cells derived from immature thymocytes of the old donors fail to express RAG-1 and RAG-2, the bone-marrow-derived cells did not exhibit this trend, and there was no difference in V` rearrangement of the TCR. T-cell progenitors have a potential to give rise to T-cells with TCR rearrangements, and the expression is determined by the thymic stroma.

2.2. Exercise and Aging Exercise poses a stress on the immune system; however, moderate chronic exercise may enhance positive immune functions. Severe or intense long-term exercise may compromise the immune system by suppressing concentrations of lymphocytes, naturalkiller cell activity, lymphocyte proliferation, and secretory IgA. During the period of immune impairment (“open window”), microbial agents may invade the host and cause infections. The changes in immune functions, lower exercise capacity, and nutritional factors in the elderly may lead to a greater and longer “open window” period, increased risk of infections which may contribute to morbidity and mortality. The specific impact of intrinsic aging has not yet been clearly dissociated from genetic traits and age-related differences in nutritional status, habitual physical activity, and exposure to psychological stressors (Rumyantsev 1998). Age-related reduction in muscle is a direct cause of the age-related decrease in muscle strength. There has been a growing belief that an appropriate regular dose of endurance exercise might slow the age-related decline in immune function by enhancing or suppressing various immunological stimuli. A decline in lean body mass, referred to as sarcopenia, and an accompanying increase in fat are known to occur during aging. The consequences of these physiological changes may include decreased physical activity, altered energy metabolism, and impaired resistance to infection. The mechanisms behind these age-related events remain unknown, but they may include changes in some of the humoral and cytokine mediators that seem to regulate body composition. Age-related loss in skeletal muscle mass has been and is a direct cause of the age-related decrease in muscle strength. There is an average loss of 10% or more per decade in maximal aerobic power with aging (Kasch et al., 1999; Ceddia et al 1999). The reduced V˙O2max is associated with a reduced proportion of the cardiac output going to exercising skeletal muscle and a reduced

Chapter 5 / Dietary Fats and Aging Immune System


cardiac output reserve (Beere et al., 1999). The elderly have higher ventilation during exercise both below and above the anaerobic threshold, which is related to higher lactic acid levels (Prioux et al., 2000). The anaerobic power of the elderly is inevitability reduced in both legs and arms (Marsh et al 1999). There is a reduction in the responsiveness of the sympathoadrenergic to exercise with aging, resulting in a reduced secretion of adrenaline for the medulla (Zouhal et al., 1999). Thus, a given activity or exercise would impose a relatively greater stress to the elderly.

2.3. Exercise and Immune Function Aging leads to a diminution of resting immune function, increasing the risk of infection, tumor development, and autoimmune diseases (Shephard and Shek, 1995). The production of IL-2 is decreased, sometimes with a decrease of total T-cell count, and often with changes in T-cell subsets and proliferative responses to mitogens. However, NK cell activity remains unchanged. In theory, moderate exercise training should help to reverse the adverse effects of aging upon the immune system. However, there have been relatively few studies comparing the immune responses of young and older individuals to acute exercise and to training. A single bout of moderate exercise seems to be well tolerated by the elderly. The NK cell response is as much as in younger individuals, but perhaps because of a low initial proliferative capacity, older subjects show less stimulation of lymphocyte proliferation by moderate activity and less suppression with exhausting exercise. Perhaps because resting immune function is less than in the young, moderate training programs seem to stimulate immune function to a greater extent in the elderly than in young subjects (Shephard et al., 1994). The proliferative response of the T-cells is enhanced in aging rodents, whereas in young animals, it is suppressed. Moreover, the resting NK cell activity of elderly human subjects seems to be increased by training. Nevertheless, the therapeutic use of exercise must be cautious in the elderly, because aging also enhances susceptibility to overtraining. Aging affects the muscle precursor cells (satellite cells or myobloasts) and their regeneration after exercise (Grounds 1998). Aging may also affect proliferation and fusion of myoblasts in response to injury; signaling molecules that stimulate satellite cells with aging; host environment, inflammatory cells, growth factors and their receptors, and the extracellular matrix. There is a reduction in growth hormone, total and free testosterone, and cortisol, both at baseline and after exercise. The decreased anabolic effects on muscles may explain the loss of muscle mass and strength with aging (Hakkinen et al., 1998). The more primitive components of immune defense, including natural killer cells, phagocytes, acute-phase proteins, and regulatory cytokines, may be altered in elderly. Several investigators have found that the proportion of neutrophils will be increased in the circulation following physical exercise. An increase in neutrophils has been correlated with an increase in plasma cortisol. The number of circulating monocytes and NK cells change as a result of exercise. Nonspecific immunity may be further stressed by maximal physical exertion that may contribute to an increased susceptibility of the elite sports person to infection. At rest, the total number of lymphocytes and CD4+ and CD8+ T-cell subset populations was lower in the elderly than the young (Mazzeo et al., 1998). There is significant exercise-induced leukocytosis, primarily lymphocytosis and neutrophilia, in both young and elderly; however, the magnitude in the elderly is reduced (Ceddia et al., 1999). The elderly have higher memory and lower CD4 and CD8 T-cells than the young. Acute


Part I / Basic Mechanisms

maximal exercise increases CD8+ and CD4+ in the elderly and young. The aged recruited fewer numbers of CD4+ naive and transitional cells than the young (Ceddia et al., 1999). Proliferative responses are lower in the elderly than the young. Proliferative response does not increase above rest in elderly subjects as it does in young subjects. Middleaged subjects had a blunted growth hormone response to exercise than the young subjects, and this was not altered by training (Zaccaria et al., 1999) and which could not be explained by augmentation of somatostatin tone and/or diminished GH-release hormone (Marcell et al., 1999). The immune system is loosely linked to the neuroendorcine system, which responds to stress. Increased cortisol, catecholamines, glucocorticoids, adrenaline, `-endorphin, growth hormones, other stress hormones, and insulin are decreased during aging. There is an increase in soluble IL-2 receptors (sIL-2R), soluble intercellular adhesion molecule-1 (sICAM-1), soluble TNF-` receptors (sTNF-R), and neopterin with exercise. These insoluble matter increases suggest that immune activation may be involved in the pathogenesis of impaired immune function after exercise. The natural-killer cell response to a single bout of exercise is normal in older individuals, but immediately after exercise, the elderly subjects manifest less suppression of phytohemagglutinin (PHA)-induced lymphocyte proliferation than younger individuals (Shinkai et al 1998). Strenuous exercise seems to induce a more sustained post-exercise suppression of cellular immunity in the elderly than in the young. The magnitude of the increase in cortisol is substantially lower in trained subjects who exercise at the same level or intensity as untrained subjects. Habitual physical activity enhances NK cell activity through mitogenesis and reduced cytokine production (Shinkai et al., 1998).

2.4. Aging, Exercise, and Antioxidant Defense System The oxidant/antioxidant balance is an important determinant of immune cell function, including maintaining integrity and functionality of membrane lipids, cellular proteins, nucleic acids, and for control of signal transduction and gene expression in immune cells. Enzymatic and nonenzymatic antioxidants play a vital role in protecting tissues from excessive oxidative damage during exercise. Antioxidant enzymes play an important role in defending the cells against free-radical-mediated oxidative damage. In rats, hepatic and myocardial antioxidant enzymes are declined at older age, whereas activity of glutathione-related enzymes in the liver and mitochondrial enzymes in the heart are increased significantly. Skeletal muscle antioxidant enzymes are uniformly elevated during aging. The generation of oxygen free radicals and other reactive oxygen species may be the underlying mechanism for exercise-induced oxidative damage, but a causal relationship remains to be established. Depletion of each of the antioxidant systems increases the vulnerability of various tissues and cellular components to reactive oxygen species. As acute strenuous exercise and chronic exercise training increase the requirement for various antioxidants, it is conceivable that dietary supplementation of specific antioxidants would be beneficial. Older subjects may be more susceptible to oxidative stress and may benefit from the antioxidant protection provided by vitamin E. During severe oxidative stress such as strenuous exercise, the enzymatic and nonenzymatic antioxidant systems of skeletal muscle are not able to cope with the massive free-radical formation, which results in an increase in lipid peroxidation. Vitamin E decreases exercise-induced lipid peroxidation. The exercise may increase superoxide anion generation in the heart, and the increase in the activity of superoxide dismutase (SOD) in skeletal muscle may be

Chapter 5 / Dietary Fats and Aging Immune System


indirect evidence for exercise-induced superoxide formation. Therefore administration of SOD may prevent exercise-induced oxidative stress. An acute bout of exercise can increase the activity of certain antioxidant enzymes in various tissues. The mechanism of this activation is unclear. Exercise training has little effect on hepatic or myocardial enzyme systems, but can cause adaptive responses in skeletal muscle antioxidant enzymes, particularly glutathione peroxidase. These findings suggest that both aging and exercise may impose an oxidative stress to the body. Optimal levels of antioxidants are required for maintenance of the immune response across all age groups. The genetic material of our cells is susceptible to damage by a wide variety of extrinsic and intrinsic entities. The amount of genetic material damage accumulated in vivo will depend on an individual’s ability to defend against and/or repair DNA damage (Barnett and Barnett, 1998). T-cells in vivo have been shown to accumulate DNA damage and mutations over time, occurring mainly in naive and memory phenotypes. Dietary supplementation with thiolic antioxidants prevents mitochondrial degeneration in the aged and preserves immune function in aged mice. Treatment with _-tocopherol and antioxidants eliminated abnormal nuclear factor kappa B (NFgB) activity, reduced peroxide levels in membrane lipids, and corrected the dysregulated expression of cytokines. This suggested that abnormal activation of NFgB may be due to the activation of intermediate kappa B (IgB) kinase by excesses of reactive oxygen species and that it may be responsible for the dysregulated expression of certain cytokines observed in aging.

3. THERAPEUTIC APPROACHES Aging is associated with the decline in multiple areas of immune function, but, to date, no single mechanism has emerged as responsible for all of the observed changes. It is being increasingly acknowledged that autoimmune processes play a proinflammatory role in the development of many pathological conditions, such as atherosclerosis. It is likely that the mechanisms underlying age-related changes in immunity are multifactorial, with both genetic and environmental factors playing a significant role (Burns et al., 1997). The progression of and recovery from infectious diseases depends, at least in part, on immune responses and nutritional status in the elderly. Nutrition therapy may improve the immune responses of elderly people, particularly those with total-, protein-, fat-, or micronutrient-energy balances on prescribed low protein or fat diets (Lesourd, 1997). Micronutrient supplementation improves the immune status in the aged individuals. Episodes of disease in the aged leads to the depletion of the body’s nutritional stores and causes protein-energy malnutrition, undernutrition associated with immunodeficiency, and, finally cachexia (Lesourd, 1999). It has been suggested that interventions include grafting cells and tissues, dietary manipulation, free-radical scavengers or antioxidants, thymic peptides, endocrine manipulation, and physical exercise (Hirokawa, 1997). There are a number of factors that can modify immunity. These include age, genetic, metabolic, environmental, anatomical, physiological, and microbial factors.

3.1. Nutrition Intervention Undernutrition is a common factor underlying many pathological conditions in the elderly and a common symptom in over 50% of hospitalized elderly patients. Macronutrient and micronutrient deficits are though to be partially responsible for the depressed


Part I / Basic Mechanisms

immune system in the elderly, even in the apparently healthy independent-living elderly (Lesourd, 1999). Protein-energy malnutrition is associated with decreased lymphocyte proliferation, reduced cytokine release, and decreased antibody responses to vaccines. Deficits in zinc, selenium, and vitamin B6 in the elderly compound the depression of the immune system. The elderly do not have an adequate supply of fats for proliferation of lymphocyte and macrophage populations, antibody production, and hepatic synthesis of acute-phase proteins. Fat oxidation is reduced in the elderly at rest, during exercise, and in response to meal ingestion, as a result, in part, to the loss of fat-free mass, decreased intrinsic fat oxidation in muscle, and hormonal factors. These changes in fat oxidation are only partially corrected with training (Calles-Escandon and Poehlman, 1997). Lipids are known to have modulatory effects on the cellular immune system at the biochemical and molecular levels, including the production and expression of cytokins. Dietary t-6 lipids (linoleic acid) generally increase the levels of proinflammatory cytokines and inflammatory prostaglandins (PGs), while t-3 lipids (eicosapentaenic acid [EPA] and docosahexaenoic acid [DHA]) may decrease the levels of these cytokines and inflammatory PGs. Destruction of polyunsaturated fatty acids (PUFAs) is an important reaction to the synthesize the very long-chain fatty acids that are necessary for membrane function and formation of eicosanoids. Reduced fat intake may modulate lymphoid cell subsets, suppression of pokeweed mitogen, CD4+/CD8+ ratio, proliferative response to mitogens, cytokine production, and so forth, thus increasing negative or inflammatory effects of exercise. Dietary lipids have an effect on cytokine balance (IL-2 and TGF-`) and immune function. Diets that are low in protein, zinc, selenium, vitamin B6, and fat may collectively depress immune function. This type of diet may be associated with either a low-caloricintake diet or low-fat, low-meat-products diet. Zinc is an essential trace element for many biological functions, including immune functions. Indeed, zinc is required for the biological activity of a thymic hormone, called thymulin in its zinc-bound form, and is important for the maturation and differentiation of T-cells. With advancing age, zinc, thymic functions, and peripheral immune efficiency show a progressive decline. Supplementing zinc in old age restores immune efficiency. Nutrients have a profound effect upon the production and actions of cytokines. Protein-energy malnutrition, dietary t-3 polyunsaturated fatty acids, and vitamin E suppress the production of specific cytokines. The synthesis of acute-phase proteins and glutathione is dependent on the adequacy of dietary sulfur-containing amino acids. The consequences of the modulatory effects of previous and concurrent nutrient intake on cytokine biology are the depletion of resources and damage to the host, which ranges from mild and temporary to severe, chronic, or lethal. It has been suggested that high pharmacological doses of zinc or vitamin E may improve immune function in the elderly (Lesourd, 1997). Vitamin E supplementation has been show to improve some aspects of immune function in aging animal and humans. Vitamin E supplementation in old mice exhibited a high significant reduction in lung viral titer, which is associated with antioxidant effects (Han and Meydani, 1999). Conjugated linoleic acid (CLA) has been suggested to have immuno-enhancing properties. Aged mice that consumed diets containing CLA (1 g CLA /100 g for 8 wk) (Hayek et al., 1999) significantly increased all CLA isomers measured in hepatic neutral lipids and phospholipids. Old mice fed 1 g CLA/100 g had significantly higher proliferative response to Con A, whereas CLA had no effect on NK cell activity, prostaglandin E2 (PGE2) production

Chapter 5 / Dietary Fats and Aging Immune System


or delayed-type hypersensitivity (DTH) in young or old mice. The observation that altering lipid substrates effects immunosuppressive eicosanoid production in the macrophages has generated many questions regarding the importance of specific PUFAs upon host immune response (Chaet et al., 1994). Fish oils are rich in the long-chain t-3 polyunsaturated fatty acids (PUFAs), eicosapentanoic (20:5t-3) and docosahexaenoic (22:6t-3) acids. Linseed oil and green plant tissues are rich in the precursor fatty acid, _-linolenic acid (18:3t-3). Most vegetable oils are rich in the t-6 PUFA linoleic acid (18:2t-6), the precursor of arachidonic acid (20:4t-6). Arachidonic-acid-derived eicosanoids such as prostaglandin E2 are proinflammatory and regulate the functions of cells of the immune system. Consumption of fish oils leads to the replacement of arachidonic acid in cell membranes, diminishes lymphocyte proliferation, T-cell-mediated cytotoxicity, NK cell activity, macrophagemediated cytotoxicity, monocyte and neutrophil chemotaxis, major histocompatibility class II expression, and antigen presentation, production of proinflammatory cytokines (IL-1, IL-6, TNF), and adhesion molecule expression. Studies on animal models indicate that fish oil reduces acute and chronic inflammatory responses, improves survival to endotoxin, and prolongs the survival of grafted organs in models of autoimmunity. Feeding fish oil reduces cell-mediated immune responses. Fish-oil supplementation may be clinically useful in acute and chronic inflammatory conditions and following transplantation. t-3 PUFAs may exert their effects by modulating signal transduction and/or gene expression within inflammatory and immune cells (Calder, 1998). Dietary long-chain (t-3) fatty acids from fish oil and low-intensity exercise have been reported, independently, to inhibit tumor growth in rats and is perhaps related to alterations in immune function. Individually, but not in combination, long-chain (t-3) fatty acids and low-intensity exercise may be advantageous by augmenting cell-mediated immune function and NK cell cytotoxicity in healthy rats (Robinson and Field, 1998). Lipid sources could alter the development of autoimmune diseases and the life-span of short-lived mice (Fernandes, 1995; Venkatraman and Fernandes, 1993). The decline in autoimmune disease found with fish oil (FO) is linked to the decrease in t-6 PUFAs, such as 18:2,20:4, that serve as precursors for proinflammatory prostaglandins of the E series. FO is also reported to decrease the levels of c-myc and c-Ha-ras oncogenes expression in the spleens of autoimmune mice and increase the levels of TGF-`1 (Chandrasekar et al., 1995; Fernandes et al., 1994; Venkatraman and Chu, 1999; Venkatraman and Chu, 1999a). The increase of TGF-`1 in the spleens may be anti-inflammatory and immunosuppressive (Fernandes et al., 1994). t-3 lipids are reported to reduce the activation of macrophages, expression of class II antigens, production of IL-1, TNF mRNA, LTB4 and TXB2 levels, and platelet aggregation. It is possible that t-3 lipids would maintain normal immune function by preventing the loss of IL-2 and IFN-a and lowering TGF` levels in T- and B-cells. Several age-associated diseases, particularly autoimmune diseases with a viral etiology, appear to be exacerbated by high-fat diets with a large proportion of vegetable oils high in t-6 fatty acids. t-3 lipids containing FO supplemented with adequate levels of vitamin E enhanced the activity and mRNA levels of hepatic antioxidant enzymes in autoimmune mice (Venkatraman et al., 1994a). These oils could increase autoimmune disease by increasing free-radical formation and decreasing the levels of antioxidant enzymes, thus further decreasing immune function by inhibiting the development of anti-inflammatory cytokines such as IL-2 and TGF-`. In contrast, t-3 lipids could protect against autoimmunity by enhancing TGF-` mRNA levels and


Part I / Basic Mechanisms

preventing an increase in oncogene expression (Fernandes etal, 1994; Venkatraman and Chu, 1999). Immune cell reactivity is directly influenced by the enrichment of lymphoid cell membranes with particular fatty acids (Venkatraman and Fernandes, 1994). Specifically, membrane enrichment in linoleic acid was associated with immunosuppression. Immunosuppression may occur through subsequent increase in the content of membrane arachidonic acid (a precursor for PGE2) which results from high 18:2t6 supplementation. PGs of the E2 series suppress blastogenesis, lymphokine production, and cytotoxicity. In aging ad libitum (AL)-fed rats, increased membrane levels of 20:4t6 may increase precursors for PGE2 and thereby immune responses. It has been indicated that aging can alter the fatty acid composition of immune cells, and this change may alter the fluidity of microenvironment in the plasma membranes. Membrane receptors can be modulated by lipid composition, microenvironment, and physicochemical characteristics of membranes, which can be altered by different dietary treatments. The percent composition of 18:2, 20:4, 22:4, and 22:5 were found to be significantly affected by age in the nonadherent spleen cells for total lipids, phosphatidyl choline, and phosphatidyl ethanolamine fractions of aged AL-fed rats. Subcellular membrane fatty acids changes and membrane rigidity may result in impairment of immune regulation along with increasing free-radical production and/or other immunosuppressive products such as PGs and certain cytokines such as IL-4, IL-5, IL-6, and IL-10 and lowering levels of IL-2 and IFN-a production. Diets containing fish oil are known to extend the life-span and inhibit free-radical formation by modulating fatty acid composition of microsomal and mitochondrial membranes of livers and spleens of animals of various ages (Byun et al., 1995; Fernandes et al., 1990; Laganiere et al., 1990). Dietary t-3 lipids induce apoptosis and apoptosis mediators in splenic lymphocytes of mice and Fas apoptotic gene expression (Fernandes et al., 1995; Fernandes et al., 1998; Reddy Avula et al., 1999). Although supplementation with t-3 fatty acids did not significantly alter the humoral immune response to keyhole limpet hemocyanin (KLH) in geriatric beagles (Wander et al., 1997), it significantly suppressed the cell-mediated immune response based on results of a delayed-type hypersensitivity (DTH) skin test. After consumption of the 1.4:1 diet, stimulated mononuclear cells produced 52% less PGE2 than those from dogs fed the 31:1 diet.

3.2. Exercise Training Few studies have followed the impact of long-term training on the immune systems of elderly people. Given that a number of age-related changes occur in many systems (e.g., neuroendocrine) known to alter immune function both at rest and during exercise, it would be of value to learn the extent to which both acute and chronic exercise influence immune function in the elderly. In older humans, aerobic exercise training lowers the heart rate at rest, reduces levels of the heart rate and plasma catecholamines at the same absolute submaximal workload, and improves left ventricular performance during peak exercise, but it does not reduce, and may even increase, basal sympathetic nerve activity. With age, there is a slow but significant reduction in muscle mass and ability to perform certain physical activities. This may be the result of changes with the age of muscle composition and protein turnover, as well as the decrease of trophic influence in neural control of muscles of the elderly. Maintenance of aerobic fitness in the older group of exercisers partially prevented the age-associated decline in platelet protein kinase C activity and in stimuli-induced enzyme redistribution.

Chapter 5 / Dietary Fats and Aging Immune System


Endurance training improves the physical capacity and body composition of elderly persons; however this effect is lost after 1 yr, unless the program is continued (Morio 2000; Tsuji et al., 2000; Kasch et al., 1999; Woods et al., 1999). V˙O2max and its trainability includes a significant genetic component (Bouchard et al., 1999). Exercise training can increase cardiac output and skeletal muscle blood flow and thus V˙O2max (Beere et al., 1999). Exercise training increases oxidative metabolism, thus decreasing anaerobic metabolism, which leads to lower ventilation (Prioux et al., 2000). Exercise training does not reverse the reduced responsiveness of the adrenal medulla to sympathetic nervous activity with aging, thus lower adrenaline (Zouhal et al., 1999). The balance between pro-oxidants and antioxidants is altered with aging (Ji et al., 1998). Exercise increases the disturbances of intracellular pro-oxidant–antioxidant homeostasis that poses a serious stress threat to the cellular antioxidant defense system (Ji et al., 1998). The elderly demonstrate a less resilient leukocytosis and a different lymphoproliferative response following acute maximal exercise (Ceddia et al., 1999). Insulin-like growth factor-I (IGF-1) declines with age, as does aerobic power, but they are not related (Haydar et al., 2000). Acute exercise may stress the immune system, whereas chronic exercise training may enhance it. Exercise-induced changes in the immune system include the release of inflammatory mediators, the activation of various white blood cells, and the complement and induction of acute-phase proteins (Venkatraman et al., 2000; Venkatraman et al., 2000a). There are signs of immunosuppression, such as decreased T- and B-cell function or impaired cytotoxic or phagoycytic activity. Because of the reduced exercise capacity, the relative stress of exercise on the immune system may be exaggerated, leading to vulnerability to acute and chronic inflammation and reduced post-exercise recovery. Increasing total caloric intake by 25% to match energy expenditure and the dietary fat intake to 32% in athletes appears to reverse the negative effects on immune function and lipoprotein levels reported on a low-fat diet (Pendergast et al., 1997; Venkatraman et al., 1997). Increasing the dietary fat intake of athletes to 42%, while maintaining caloric intake equal to expenditure, does not negatively affect immune competency or blood lipoproteins, but it improves endurance exercise performance at 60–80% of V˙O2max in runners (Pendergast et al., 1997; Venkatraman et al., 1997). Both exercise and a high-fat diet increase the level of IL-2 and lowered the level of IL-6, suggesting that it is possible to modulate the level of specific proinflammatory cytokines through increased fat intake, thus offsetting the proinflammatory effects of exercise (Venkatraman et al., 1997; Venkatraman and Pendergast, 1998). Aging is characterized by shrinking of muscle fibers and protein loss from these muscle fibers. Bone is also lost, and matrix and mineral levels are also lost equally. The predominant breakdown of synthesis is probably the fundamental cause of both muscle and bone loss. Little can be done to prevent this by dietary means, but physical activity is of vital importance in helping to maintain the integrity of both muscle and bone. Resistance training is an effective means of preserving or increasing skeletal muscle mass and functional status in the elderly. In addition, resistance training has been demonstrated to increase energy requirements, protein retention, bone mass, and levels of physical activity in the healthy elderly as well as the very old and frail. The influence of 4 wk of anaerobic training program with 30-min sessions of weight lifting per week in middleaged, moderately trained men (40–50 yr) was studied, and significant increases of the mean arm muscle force by 7% was found (Weiss et al., 1995). Highly conditioned elderly women retained greater lymphocyte proliferative response to PHA, IL-2 production, and


Part I / Basic Mechanisms

greater NK cell activity than sedentary peers, and two groups had similar leukocyte and lymphocyte subset compositions. These observations indicate higher per cell functional activities in the well-trained group, suggesting that regular habitual physical activity may help to encounter the age-related decline in the potential of the T-cells (Nieman et al., 1993). IFN-a and IL-4 production were also higher in the elderly runners. Although the mechanism by which exercise training enhances the immune function is unclear, one possibility might be a persisting input of IL-1 from macrophages active in repairing muscle microtraumata. A single bout of moderate exercise seems to be well tolerated by the elderly. The NK cell response is found to be equal to that of younger individuals. However, older subjects show less stimulation of lymphocyte proliferation by moderate activity and also show less suppression with exhausting exercise. This depressed response may occur because of the depressed resting immune function or as the elderly have markedly less activity than the young, and thus a moderate training program seems to stimulate immune function to a greater extent in older than in young subjects. During exercise, the release of catecholamines and cortisol may be suppressive to NK activity, whereas the release of IL-1, IFN, and `-endorphin may be stimulating. Although NK activity in the healthy elderly may be similar to the young at baseline, they would show a diminution of their response to acute stressors (i.e., exercise or IL-2), as has been commonly observed in other systems with aging. The preserved responsiveness to NK cells in the older groups is even more remarkable because catecholamine secretion is increased with exercise in the elderly, compared to the young, which would presumably mediate more suppressive influences on NK activity with age.

4. SUMMARY AND CONCLUSION It would seem that modulation of lipid intake and exercise training may mediate and reverse some of the age-associated depression in the immune system. As human aging generally is accompanied by a reduction in the level of physical activity and impaired responsiveness of the immune system, there may be a relationship between these concurrent changes and the increased incidence of, and mortality from, cancer, autoimmune disorders, and chronic infectious diseases with age. The studies on the immune system of centenarians suggest that the immune system in this population may not be declining with age, instead it is being constantly remodeled and reshaped as required. Appropriate fat and protein intake may improve immune function. Lipids modulate immune function by several factors and mediators. Both the quantity and type of dietary lipids are known to have modulatory effects on the cellular immune system at biochemical and molecular levels. The mechanisms by which lipids modulate the immune function may involve several factors, including the production and expression of cytokines. Dietary t-6 lipids generally increase the levels of proinflammatory cytokines and inflammatory PGs, whereas t-3 lipids may decrease the levels of these cytokines and inflammatory PGs. Moderate exercise in the elderly may further help them to lead a disease-free life by preserving their immune system. Further cross-sectional and longitudinal studies are required that would examine the type, intensity, and total quantity of exercise needed to optimize the immune function and the way in which this optimum dose may also change with aging. It is important that the dose of physical activity needed to optimize the immune function be defined more clearly at various points during the aging process, and

Chapter 5 / Dietary Fats and Aging Immune System


that its preventive efficacy must be explored extensively. Despite the relative normality of the response to moderate exercise in the elderly, they have a reduced tolerance of free radicals and greater vulnerability to micro-injuries and an acute-phase response from overexertion. It might thus be anticipated that they would more easily reach the point where natural immunity is suppressed and vulnerability to infection is enhanced. Older adults may have to adopt a more cautious approach and follow a moderate exercise regimen along with a nutritionally well-balanced diet. Training in theory has a number of actions which could help to reverse the impact of aging upon the immune system, including a direct modulation of sympathetic activity in the neurohypophysis, a reduction of cross-linkages and a diminution of free-radical formation.

REFERENCES Barnett YA, Barnett CR. DNA damage and mutation: contributors to the age-related alterations in T cellmediated immune responses? Mech Ageing Dev 1998; 102:165–175. Beere PA, Russell SD, Morey MC, Kitzman DW, Higginbotham MB. Aerobic exercise training can reverse age-related peripheral circulatory changes in health older men. Circulation 1999; 100:1085–1094. Bodnar, AG, Ouellette M, Frolkis, M, Holt SE, Chiu CP, Morin GB, et al. Extension of life span by introduction of telomerase into normal human cells. Science 1998; 279:349–352. Bouchard C, An P, Rice T, Skinner JS, Wilmore JH, Gagnon J, et al. Familial aggregation of VO2(max) response to exercise training: results from the HERITAGE Family study. J Appl Physiol 1999; 87:1003–1008. Burns EA, Goodwin JS. Immunodeficiency of aging. Drugs Aging 1997; 11:374–397. Byun DS, Venkatraman JT, Yu BP, Fernandes, G. Modulation of antioxidant activities and immune response by food restriction in aging Fischer-344 rats. Aging: Clin Exp Res 1995; 7:40–48. Calder PC. Immunoregulatory and anti-inflammatory effects of n-3 polyunsaturated fatty acids. Braz J Med Biol Res 1998; 31:467–490. Calles-Escandon J, Poehlman ET. Aging, fat oxidation and exercise. Aging 1997; 9:57–63. Catania A, Airaghi L, Motta P, Manfredi MG, Annoni G, Pettenati C, et al. Cytokine antagonists in aged subjects and their relation with cellular immunity. J Gerontol: Biol Sci 1997; 52A:B93–B97. Ceddia MA, Price EA, Kohlmeier CK, Evans JK, Lu Q, McAuley E, et al. Differential leukocytosis and lymphocyte mitogenic response to acute maximal exercise in the young and old. Med Sci Sports Exerc 1999; 31:829–836. Chaet MS, Garcia VF, Arya G, Ziegler MM. Dietary fish oil enhances macrophage production of nitric oxide. J Surg Res 1994; 57:65–68. Chandrasekar B, Troyer DA, Venkatraman JT, Fernandes G. Dietary t-3 lipids delay the onset and progression of autoimmune lupus nephritis by inhibiting transforming growth factor ` mRNA and protein expression. J Autoimmun 1995; 8:381–393. Cinader B, Doria G, Facchini A, Kurashima C, Thorbecke GJ, Weksler ME. Aging and the immune system: Workshop #4 of 8th International Congress of Immunology, Budapest, Hungary. Aging: Immunol Infec Dis 1993; 4:47–53. Doria G, Frasca D. Genes, immunity, and senescence: looking for a link. Immunol Rev 1997; 160:159–170. Ernst DN, Hobbs MV, Torbett BE, Glasebrook AL, Rehse MA, McQuitty DN, et al. Differences in the subset composition of CD4+ T cell population from young and old mice. Aging: Immunol Infec Dis 1990; 2:105–109. Ernst DN, Weigle O, Hobbs MV. Aging and lymphokine gene expression by T cell subsets. Nutr Rev 1995; 53:S18–S26. Fabris N, Mocchegiani E, Provinciali M. Plasticity of neuroendocrine–thymus interactions during aging. Exp Gerontol 1997; 32:415–429. Fernandes, G. Effects of calorie restriction and omega-3 fatty acids on autoimmunity and aging. Nutr Rev 1995; 53:S72–S79. Fernandes A, Venkatraman JT, Tomar V, Fernandes G. Effect of treadmill exercise and food restriction on immunity and endocrine hormone levels in rats. FASEB J 1990; 4:751A. Fernandes G, Venkatraman JT. Effect of food restriction on immunoregulation and aging. In R. R. Watson (Ed.), Handbook of Nutrition in the Aged. 2nd ed. CRC, Boca Raton, FL, 1993, pp. 331–346.


Part I / Basic Mechanisms

Fernandes G, Bysani C, Venkatraman JT, Tomar V, Zhao W. Increased TGF` and decreased oncogene expression by t-3 lipids in the spleen delays autoimmune disease in B/W mice. J Immunol 1994; 152:5979–5987. Fernandes G, Chandrasekar B, Mountz JD, Zhao W. Modulation of Fas apoptotic gene expression in spleens of B/W mice by the source of dietary lipids with and without calorie restriction. FASEB J 1995; 9:4559. Fernandes G, Troyer DA, Jolly CA. The effects of dietary lipids on gene expression and apoptosis. Proc Nutr Soc 1998; 57:543–550. Fernandes G, Venkatraman JT, Turturro A, Attwood VG, Hart RW. Effect of food restriction on life span and immune functions in long-lived Fisher-344 × Brown Norway F1 rats. J Clin Immunol 1997; 17:85–95. Grounds MD. Age-associated changes in the response of skeletal muscle cells to exercise and regeneration. Ann NY Acad Sci 1998; 854:78–91. Guidi, L.,Tricerri A, Frasca D, Vangeli M, Errani AR, Bartoloni C. Psychoneuroimmunology and aging. Gerontology 1998; 44:247–261. Hakkinen K, Pakarinen A, Newton RU, Kraemer WJ. Acute hormone responses to heavy resistance lower and upper extremity exercise in young versus old men. Eur J Appl Physiol Occup Physiol 1998; 77:312–319. Han SN, Meydani SN. Vitamin E and infectious diseases in the aged. Proc Nutr Soc 1999; 58:697–705. Haydar ZR, Blackman MR, Tobin JD, Wright JG, Fleg JL. The relationship between aerobic exercise capacity and circulating IGF-1 levels in healthy men and women. J Am Geriatr Soc 2000, 48:139–145. Hayek MG, Han SN, Wu D, Watkins BA, Meydani M, Dorsey JL, et al. Dietary conjugated linoleic acid influences the immune response of young and old C57BL/6NCrlBR mice. J Nutr 1999; 129:32–38. Hirokawa K. Reversing and restoring immune functions. Mech Ageing Dev 1997; 93:119–124. Hirokawa K, Utsuyama M, Kasai M, Kurashima C, Ishijima S, Zeng YX. Understanding the mechanism of the age-change of thymic function to promote T cell differentiation. Immunol Lett 1994; 40:269–277. Ji LL, Leeuwenburgh C, Leichtweis S, Gore M, Hollander J, Bejm J. Oxidative stress and aging. Role of exercise and its influences on antioxidant systems. Ann NY Acad Sci 1998; 854:102–117. Kasch FW, Boyer JL, Schmidt PK, Wells RH, Wallace JP, Verity LS, et al. Aging of the cardiovascular system during 33 years of aerobic exercise. Age Ageing 1999; 28:531–536. Krause D, Mastro AM, Handte G, Smiciklas-Wright H, Miles MP, Ahluwalia, N. Immune function did not decline with aging in apparently healthy, well-nourished women. Mech Ageing Dev 1999; 112:43–57. Laganiere S, Yu BP, Fernandes G. Studies on membrane lipid peroxidation in omega-3 fatty acid-fed autoimmune mice: effect of vitamin E supplementation. In: Bendich A, Phillips M, Tangerdy RB, eds. Adv Exp Med Biol, Vol. 262, Plenum Press, New York, 1990, pp. 95–103. Lesourd BM. Nutrition and immune in the elderly: modification of immune response with nutritional treatments. Am J Clin Nutr 1997; 66:478S–484S. Lesourd BM. Immune response during disease and recovery in the elderly. Proc Nutr Soc 1999; 58:85–98. Makinodan T. Studies on the influence of age on immune response to understand the biology of immunosenescence. Exp Gerontol 1998; 33:27–38. Makinodan T, Chang MP, Norman DC, Li SC. Vulnerability of T cell lineage to aging. In: Goidl E, ed. Aging and Immune Response: Cellular and Humoral Aspects. Marcel Dekker, New York, 1987, pp. 27–39. Marcell TJ, Wiswell RA, Hawkins SA, Tarpenning KM. Age-related blunting of growth hormone secretion during exercise may not be due to increased somatosatatin tone. Metab: Clin Exp 1999; 48:665–670. Marsh GD, Paterson DH, Govindasamy D, Cunningham DA. Anaerobic power of the arms and legs of young and older men. Exp Physiol 1999; 84:589–597. Mazzeo RS, Rajkumar C, Rolland J, Blaher B, Jennings G, Eshler M. Immune response to a single bout of exercise in young and elderly subjects. Mech Ageing Dev 1998; 100:121–132. Mazzoccoli G, Correra M, Bianco G, De Cata A, Balzanelli M, Giuliani A, et al. Age-related changes of neuroendocrine-immune interactions in healthy humans. J Biol Regul Homeostat Agents 1997; 11:143–147. McCann SM, Licinio J, Wong ML, Yu WH, Karanth S, Rettorri V. The nitric oxide hypothesis of aging. Exp Gerontol 1998; 33:813–326. Morio B, Barra V, Ritz P, Fellmann N, Bonny JM, Beaufrere B, et al. Benefit of endurance training in elderly people over a short period is reversible. Eur J Appl Physiol Occup Physiol 2000; 81:329–336. Mountz JD, Wu J, Zhou T, Hsu H-C. Cell death and longevity: implications of Fas-mediated apoptosis in T-cell senescence. Immunol Rev 1997; 160:19–30. Nieman DC, Henson DA, Gusewitch G, Warren BJ, Dotson RC, Butterworth DE, Nehlsen-Cannarella SL. Physical activity and immune function in elderly women. Med Sci Sports Exerc 1993; 25:823–831. Pendergast DR, Horvath PJ, Leddy JJ, Venkatraman JT. The role of dietary fat on performance metabolism and health. Am J Sports Med 1997; 24:S53–S58.

Chapter 5 / Dietary Fats and Aging Immune System


Phelouzat MA, Arbogast A, Laforge T, Quadri RA, Proust JJ. Excessive apoptosis of mature T lymphocytes is a characteristic feature of human immune senescence. Mech Ageing Dev 1996; 88:25–38. Pierpaoli W. Neuroimmunomodulation of aging. A program in the pineal gland. Ann NY Acad Sci 1998; 840:491–497. Potestio M, Caruso C, Gervasi F, Scialabba G, D’Anna C, Di Lorenzo G, et al. Apoptosis and ageing. Mech Ageing Dev 1998; 102:221–237. Prioux J, Ramonatxo M, Hayot M, Mucci P, Prefaut C. Effects of ageing on the ventilatory response and lactate kinetics during incremental exercise in man. Eur J Appl Physiol Occup Physiol 2000; 81:100–107. Reddy Avula CP, Zaman AK, Lawrence R, Fernandes G. Induction of apoptosis and apoptotic mediators in Balb/c splenic lymphocytes by dietary n-3 and n-6 fatty acids. Lipids 1999; 34:921–927. Rink L, Cakman I, Kirchner H. Altered cytokine production in the elderly. Mech Ageing Dev 1998; 102:199–209. Robinson LE, Field CJ. Dietary long-chain (n-3) fatty acids facilitate immune cell activation in sedentary, but not exercise-trained rats. J Nutr 1998; 128:498–504. Rumyantsev SN. Constitutional and non-specific immunity to infection. Rev Sci Tech 1998; 17:26–42. Shephard PJ, Rhind S, Shek PN. Exercise and the immune system. Natural killer cells, interleukins and related responses. Sports Med 1994; 18:340–369. Shephard RJ, Shek PN. Exercise, aging and immune function. Intl J Sports Med 1995; 16:1–6. Shinkai S, Konishi M, Shephard RJ. Aging and immune response to exercise. Can J Physiol Pharmacol 1998; 76:562–572. Solana R, Pawelec G. Molecular and cellular basis of immunosenescence. Mech Ageing Dev 1998; 102:115–129. Stoessl AJ. Etiology of Parkinson’s disease. Can J Neuro Sci 1999; 26:S5–S12. Thoman ML, Weigle WO. The cellular and sub-cellular bases of immunosenescene. Adv Immunol 1989; 46:221–261. Tsuji I, Tamagawa A, Nagatomi R, Irie N, Ohkubo T, Saito M, et al. Randomized controlled trial of exercise training for older people (sendai Silver Center Trial; SSCT): study design and primary outcome. J Epidemiol 2000; 10:55–64. Venkatraman JT, Chu W. Effects of dietary t-3 and t-6 lipids and vitamin E and on proliferative response, lymphoid cell subsets, production of cytokines by spleen cells and splenic protein levels for cytokines and oncogenes in MRL/MpJ-lpr mice. J Nutr Biochem 1999; 10:582–597. Venkatraman JT, Chu W. Effects of dietary t-3 and t-6 lipids and vitamin E on serum cytokines, lipid mediators and anti-DNA antibodies in a mouse model for rheumatoid arthritis. J Am Coll Nutr 1999; 18:602–612. Venkatraman JT, Fernandes G. Modulation of immune function during aging by dietary restriction. In: Yu BP, ed. Modulation of Aging Processes by Dietary Restriction. CRC, Boca Raton, 1994, pp. 193–219. Venkatraman JT, Fernandes G. Exercise, immunity and aging. Aging: Clin Exp Res 1997; 9:42–55 Venkatraman JT, Attwood VG, Turturro A, Hart RW, Fernandes G. Maintenance of virgin T-cells and immune functions by food restriction during aging in long-lived B6D2F1 female mice. Aging: Immunol Infect Dis 1994; 5:13–25. Venkatraman JT, Bysani C, Kim JD, Fernandes G. Effect of n-3 and n-6 fatty acids on activities and expression of hepatic antioxidant enzymes in autoimmune-prone NZB/NZWF1 mice. Lipids 1994; 29:561–568. Venkatraman JT, Fernandes G. t-3 lipids in health and disease. Nutr Res 1993; 13:19S–45S. Venkatraman JT, Pendergast DR. Effects of the level of dietary fat intake and endurance exercise on plasma cytokines in runners. Med Sci Sports Exerc 1998; 30:1198–1204. Venkatraman JT, Horvath PJ, Pendergast DR. Lipids and exercise immunology. In: Nieman DC, Pedersen B, eds. Nutrition and Exercise Immunology. CRC, Boca Raton, 2000, pp. 43–74. Venkatraman JT, Leddy JJ, Pendergast DR. Dietary fats and immune status in athletes: clinical implications. Med Sci Sports Exerc 2000; 32:S1–S7. Venkatraman JT, Rowland JA, DeNardin E, Horvath PJ, Pendergast DR. Influence of the level of dietary lipid intake and maximal exercise on immune status in runners. Med Sci Sports Exerc 1997; 29:333–344. Walford RL. MHC regulation of aging: an extension of the immunologic theory of aging. In: Warner HR, Butler RN, Sprott RL, Schneider EL, eds. Modern Biological Theories of Aging. Raven, New York, 1987, pp. 243–255. Wander RC, Hall JA, Gradin JL, Du SH, Jewell DE. The ratio of dietary (n-6) to (n-3) fatty acids influences immune system function, eicosanoid metabolism, lipid peroxidation and vitamin E status in aged dogs. J Nutr. 1997; 127:1198–205.


Part I / Basic Mechanisms

Weiss C, Kinscherf R, Roth S, Friedmann B, Fischbach T, Reus J, et al. Lymphocyte subpopulations and concentrations of soluble CD8 and CD4 antigen after anaerobic training. Intl J Sports Med 1995; 16:117–121. Weyand CM, Brandes JC, Schmidt D, Fulbright JW, Goronzy JJ. Functional properties of CD4+ CD28– T-cells in the aging immune system. Mech Ageing Dev 1998; 102:131–147. Wick G, Grubeck-Loebenstein B. The aging immune system: primary and secondary alterations of immune reactivity in the elderly. Exp Gerontol 1997; 32:401–13. Woods JA, Ceddia MA, Wolters BW, Evans JK, Lu Q, McAuley E. Effects of 6 months moderate aerobic exercise training on immune function in the elderly. Mech Ageing Dev 1999; 109:1–19. Yehuda AB, Friedman G, Wirtheim E, Abel L, Globerson A. Checkpoints in thymocytopoiesis in aging: expression of the recombination activating genes RAG-1 and RAG-2. Mech Ageing Dev 1998; 102:239–247. Zaccaria M, Varnier M, Piazza P, Noventa D, Ermolao A. Blunted growth hormone response to maximal exercise in middle-aged versus young subjects and no effect of endurance training. J Clin Endocrinol Metab 1999; 84:2303–2307. Zouhal H, Gratas-Delamarche A, Rannou F, Granier P, Bentue-Ferrer D, Delamarche P. Between 21 and 34 years of age, aging alters the cateholamine responses to supramaximal exercise in endurance trained athletes. Intl J Sports Med 1999; 20:343–348.

Chapter 6 / _-LNA and DHA During Perinatal Development





Part II / Phospholipid and Fatty Acid

Chapter 6 / _-LNA and DHA During Perinatal Development



On the Relative Efficacy of _-Linolenic Acid and Preformed Docosahexaenoic Acid as Substrates for Tissue Docosahexaenoate During Perinatal Development Meng-Chuan Huang and J. Thomas Brenna

1. PERINATAL N-3 AND N-6 FATTY ACID METABOLISM AND NEURAL DEVELOPMENT The polyunsaturated fatty acids (PUFA), _-linolenic acid (LNA, 18:3n-3) and linoleic acid (LA, 18:2n-6) are indispensable dietary components in mammals. It is thought that the primary metabolic role of LA and LNA are as essential precursors for conversion to their respective long-chain metabolites. Of these long-chain PUFA (LCP, C>20) docosahexaenoic acid (DHA, 22:6n-3) and arachidonic acid (AA, 20:4n-6), have received the most attention, because of their distinct functional roles in fetal and infant health. They are present in human breast milk at small but significant amounts, ranging from 0.1% to 0.6% and 0.5% to 1%, respectively (Jensen, 1996). DHA and AA are the most abundant n-3 and n-6 LCP found in the lipids of humans and are concentrated in the brain and central nervous system (CNS), which is second only to adipose tissue in containing the greatest percentage of lipid. The brain is comprised of 50–60% lipids (dry wt) serving mostly structural roles, as opposed to storage of energy of triacylglycerol in adipose tissues. A related organ, the retina, is also rich in DHA, which comprises approximately 50% of phospholipid acyl groups in retinal photoreceptor membranes (Fliesler & Anderson, 1983; Sastry, 1985). The functional importance of DHA is related to its functional significance in supporting perinatal neurological and visual development. Depletion of DHA in the brains of monkeys during fetal and early postnatal development results in permanent deficits of visual function despite later biochemical repletion of brain and retinal DHA (Neuringer, Connor, Lin, Barsted, & Luck, 1986; Neuringer, Connor, Petten, & Barstad, 1984). AA is a structural component of membrane phospholipids and is known to exert its biological actions by serving as the precursor of eicosanoids, which exercise ubiquitous physiological influences. Docosahexaenoic acid and AA can be obtained either directly from the diet or synthesized from their C18 precursors LNA or LA, respectively, via a series of equivalent From: Fatty Acids: Physiological and Behavioral Functions Edited by: D. Mostofsky, S. Yehuda, and N. Salem Jr. © Humana Press Inc., Totowa, NJ



Part II / Phospholipid and Fatty Acid

desaturation and elongation reactions. The synthetic pathway of n-3 and n-6 fatty acids have long been known to interact (Mohrhauser & Holman, 1963); the dietary level of one series of fatty acids can interfere with the metabolism of the other (Lands, 1995). For instance, a dietary excess of n-6 fatty acids can reduce the tissue concentration of n-3 fatty acids, leading to concerns about deficits of n-3 LCP. Similarly, an excess of n-3 fatty acids can hinder conversion of LA to n-6 LCP. Thus, moderate ratios of dietary n-3 and n-6 C18 precursors are required to facilitate balanced LCP biosynthesis for fastgrowing nervous tissue to achieve optimal function, especially when dietary preformed DHA and AA are absent. The roles of n-3 and n-6 LCP in infant nutrition have been extensively reviewed by several authors (Connor, Neuringer, & Reisbick, 1992; Innis, 1991; Jensen, 1996), and it is agreed that DHA is essential to perinatal neurological development. In the early 1970s, Benolken et al. reported that the amplitude of electroretinogram a-waves was lower in rats fed a fat-free diet than in those raised on a fat sufficient diet (Benolken, Anderson, & Wheeler, 1973). The key components causing electroretinogram changes were later identified as dietary n-3 fatty acids (Wheeler & Benolken, 1975). Further, the association between n-3 fatty acid deficiency and subsequent impaired neural outcomes was confirmed in rodents by a number of investigators (Bourre, Durand, Pascal, & Youyou, 1989; Yamamoto, Saitoh, Moriuchi, Nomura, & Okuyama, 1987). Strong evidence that formula-fed human infants may be at risk of n-3 fatty acid or DHA deficiency arises from infant rhesus monkey studies. Neuringer et al. first reported that dietary n-3 fatty acid deprivation depressed retinal and other visual functions in primates. Infant monkeys fed an n-3 fatty-acid-depleted diet (with 18:2n-6/18:3n-3 ratio = 255) had impaired visual acuity development as assessed by electroretinography and the preferential looking method compared to those fed n-3-adequate diets (Neuringer et al., 1986; Neuringer et al., 1984). This functional deficit was ascribed to DHA deficiency because the neural deficits were positively correlated with loss of DHA from the cerebral cortex and retina. In a related human study, Farquharson et al. reported on the cerebral cortex fatty acid composition of term and preterm infants expired as a result of sudden infant death syndrome. They showed that formula-fed infants had poorer DHA levels than breast-fed infants, with the average brain DHA 22–23% lower for formula-fed relative to breast-fed (Farquharson, co*ckburn, Patrick, Jamieson, & Logan, 1992; Farquharson et al., 1995). Until 10 yr ago, most commercial infant formulas had high LA/LNA ratios, ranging from 10 to 50 (Jensen, Hagerty, & McMahon, 1978) and did not contain preformed LCP. Infants fed such formulas would have to derive all their LCP from C18 precursors and might be expected to have impaired conversion of LNA to DHA because of high dietary LA. These findings sparked numerous clinical trials in preterm/term infants to establish PUFA compositions for optimal neural development during infancy (Crawford, Costeloe, Ghebremeskel, & Phylactos, 1998; Heird, Prager, & Anderson, 1997). These studies generally show that preterm infant visual and cognitive function improves with addition of LCP, particularly DHA, to formula, whereas functional data in term infants sometimes shows improvement and sometimes not, leading to a consensus that both DHA and AA should be added to preterm formula, and although no universal consensus has emerged for term formula as yet. In most of these studies, changes in fatty acids in blood compartments (plasma, red cells) were considered as indices of the respective fatty acid status in neural tissues such as brain and retinas, because ethical constraints limit sampling to

Chapter 6 / _-LNA and DHA During Perinatal Development


blood even though the target organs are the brain and retina. However, caution is warranted because data derived from animal studies (Hrboticky, Mackinnon, & Innis, 1990; Pawlosky, Ward, & Salem, 1996; Rioux, Innis, Dyer, & MacKinnon, 1997; Su et al., 1999a) and human studies (Makrides, Neumann, Byard, Simmer, & Gibson, 1994) show that blood-compartment fatty acid profiles are sensitive to short-term dietary changes and do not necessarily mirror that in neural tissues. In addition, practical constraints limit the number of dietary treatments that can be investigated. The net outcome of all these trials is confidence in the safety and efficacy of LCP supplementation in a limited number of dietary PUFA levels. In clinical studies of growing human infants, it is impossible to determine the relationship of dietary PUFA levels and tissue n-3/n-6 fatty acid accretion, information required to establish optimal dietary LCP levels, because of the inability to sample target tissue. For this, animal studies are required.

2. CONSIDERATIONS FOR ANIMAL STUDIES It has been noted that the timing of brain growth varies among species and that studies aimed at investigating brain development must make observations relevant to the period when the brain is growing. The period of brain development during which brain growth rate exceeds that of other organs is known as the brain growth spurt (Dobbing & Sands, 1979). Animals born relatively mature, such as the guinea pig, called “precocial,” generally grow their brains in utero and, consequently, have their highest demands for DHA prenatally. Animals born immature such as rats (“altricial”), grow their brains postnatally. The human brain growth spurt is rare in that it is perinatal. The infant brain accumulates DHA from the beginning of the third trimester in utero and continues up to the first 24 mo of early neonatal growth (Martinez, Ballabriga, & Gil-Gibernau, 1988). It is this period when human neural development is most dependent on adequate supplies of nutrients. Equally important to brain nutrition is the relative brain-weight percentage. For nonprimates, the percentage of total body weight represented by the brain is less than 5% at birth (Brody, 1945). For instance, the rat brain is about 2% of body weight at birth. In contrast, the brain weight of rhesus monkeys, baboons, and other primates is greater than 10% at birth, whereas the brain weight of term human infants is about 14% at birth. Because nutrient requirements are best expressed in terms of a fraction of dietary energy, it stands to reason that requirements for an animal supporting the growth of a brain of 10% body weight are more applicable to humans than an animal with a brain of 2% body weight. The desaturase activities required for conversion of LNA and LA to DHA and AA, respectively, have been detected in perinatal animals and humans. 6-6 and 6-5 desaturase activities have been found in prenatal animal tissues, including rat liver and brain (Bourre, Piciotti, & Dumont, 1990). In humans, expression of these enzyme activities were found in neonates (Poisson et al., 1993) and in fetuses as early as 17–18 wk of gestation and are stable throughout the third trimester (Chambaz et al., 1985; Rodriguez et al., 1998). Using stable-isotope-tracer methodology, human term and preterm infants were shown, in vivo, to be capable of converting the C18 precursor to the long-chain metabolites, DHA and AA, subsequent to oral administration of 13C-labeled LNA and LA, respectively (Carnielli et al., 1996; Salem, Wegher, Mena, & Uauy, 1996; Sauerwald et al., 1997). In these and subsequent studies, a wide range of variability was observed in the appearance of labeled


Part II / Phospholipid and Fatty Acid

Fig. 1. Graphical representation of study by Anderson and co-workers. After laying hens were fed an n-3-deficient diet for 2 mo, their chicks were fed one of three diets for three wk: control and n-3-deficient supplemented with equivalent amounts of LNA (+LNA) or DHA (+DHA). Brain DHA levels (plotted on the ordinate) for the +LNA are 25% of those for the +DHA group. (Based on data from Anderson, Connor, & Corliss, 1990).

LCP in the plasma, indicating a wide range of DHA synthetic competence (Uauy, Mena, Wegher, Nieto, & Salem, 2000). All evidence supports the ability of human fetuses/ infants to convert LNA to DHA in vivo. However, none of these studies addresses the quantitative efficacy of this process because, as with all clinical studies, sampling from blood compartments cannot be used to reliably compute the adequacy of LCP biosynthesis and incorporation for neural tissues.

3. DIETARY LNA AND DHA AS SUBSTRATES FOR BRAIN AND RETINA DHA: COMPOSITIONAL STUDIES In nontracer experiments, relative contribution of dietary LNA and DHA to CNS DHA accretion have been examined in various species, including chicks, rat pups, newborn piglets, and guinea pigs. Anderson and colleagues studied the relative efficacy of LNA and DHA in restoring neural DHA levels in newly hatched chicks, as presented in Fig. 1 (Anderson, Connor, & Corliss, 1990). Laying hens were fed a n-3-deficient diet for 2 mo, and their hatched chicks were then fed a control diet or n-3-deficient diets supplemented with LNA (+LNA) or DHA (+DHA) at 3.6 wt% (0.44% kcal) for 3 wk. After 3 wk, the DHA group showed brain DHA levels similar to that of controls (12.3% vs 8.3%), whereas dietary LNA alone brought the level to only 25% of the controls. Similar results were observed for retinal DHA accretion. It was concluded that dietary DHA exerts a fourfold greater potency compared to LNA as a substrates for brain and retina DHA accretion. Based on this finding, Arbuckle et al. postulated that neonatal pigs fed formula with an LNA concentration four times greater than that used in the chick study (1.7% kcal vs 0.44% kcal) would support retinal DHA at levels similar to a sow-fed group (Arbuckle & Innis, 1992). Four test formulas were used: high LNA (1.7% kcal), low LNA (0.3% kcal), low LNA plus fish oil (C20 + C22 = 0.4% kcal) with n-3 LCP concentration used comparable to those used in human clinical trials (Carlson, Rhoides, Rao, & Goldgar, 1987; Uauy, Birch, Birch, Tyson, & Hoffman, 1990) and sow-reared with milk containing 1.1 wt% as LNA and 0.1 wt% as DHA. In retina phosphatidylethanolamine and in synaptic membranes, piglets fed the high-LNA diets, but not the low-LNA group, exhibited DHA comparable to that in the sow-reared group and the low-LNA plus fish oil

Chapter 6 / _-LNA and DHA During Perinatal Development


group. It was thus concluded that dietary LNA is 24% as effective as C20 and C22 n-3 fatty acids as a source of membrane DHA, yielding, again, a bioequivalence of about 4 : 1. Current LCP-free infant formulas contain an LA/LNA ratio of between 7 and 10. One strategy to induce greater conversion of LNA to DHA is to reduce this ratio substantially, thereby reducing competition for LCP biosynthesis from LA. Woods and co-workers tested this hypothesis in an artificial feeding study in rat pups (Woods, Ward, & Salem, 1996). DHA accretion in the brain and retina was compared among artificially reared rat pups fed diets containing LA/LNA ratios of 10 : 1, 1 : 1, and 1 : 12 (LNA = 3%, 15%, and 25% of total fatty acids, respectively) and compared to dam-reared rat pups. They found that brain DHA was comparable to that of dam-reared pups consuming milk containing 1.1% DHA only in the 1 : 12 group LA/LNA ratio. Retinal DHA accretion, however, was still 10% lower in this treatment compared to the dam-reared group. They conclude that efficacy of DHA accretion from dietary DHA is at least 20-fold greater than LNA. It is notable that in this study, as in all dietary fatty acid studies, the LCP content of reference-group diets is critical in assessing adequate tissue levels. Dietary LCP are known to produce higher tissue LCP levels than their corresponding C18 precursors, possibly because the storage of dietary LCP in their nascent state is energetically more efficient than extensive metabolism prior to storage. By the same reasoning, dietary LNA and LA may be converted to LCP only to the extent that LCP is required, with increasing levels of dietary LNA and LA shunted to storage or other metabolic fates. Thus, higher levels of LCP in reference groups consuming LCP may reflect excess rather than active metabolic requirements. In the present study, the estimated conversion ratio depends on the dietary fatty acid composition of the dams’ feed, which contained 6 wt% of n-3 LCP, and induced relatively high milk LCP concentrations, and there is no evidence of any functional significance for these higher DHA levels. In a human term infant study, Jensen et al. investigated the effects of formulas with differing LA/LNA ratios on plasma phospholipid fatty acids and transient visual evoked responses up to 240 d of life (Jensen et al., 1997). Infants who consumed formula for 4 mo with an LA/LNA ratio of 4.8 had higher plasma phospholipid DHA but lower AA compared to those who consumed formulas with LA/LNA ratios of 9.7 or higher, showed no improvement in visual responses, and were significantly lower in body weight than the highest LA/LNA group (44 : 1). From these data, it was concluded that low LA/LNA ratios may not be an appropriate means to normalize plasma fatty acid content because of adverse effects on growth, and that there was no major improvement in the functional outcome measured. Very recently, Abedin et al. reported a study on weanling guinea pigs in which LNA and LA levels were studied along with LCP treatments. Brain and retina DHA levels were analyzed in 15-wk-old guinea pigs after 12 wk of feeding with one of five diets: low LNA (“Lo LNA”), medium LNA (“Med LNA”), high LNA (LNA: 7%; “Hi LNA”), and two diets constructed to resemble DHA and AA concentration comparable to human breast milk (Lo LCP: 1%, 0.6% DHA) or three times more human breast milk (Hi LCP), as shown in Fig. 2 (Abedin, Lien, Vingrys, & Sinclair, 1999). It was found that brain and retina DHA levels were similar between animals fed the high-LNA diet and animals fed Lo LCP after 12 wk of feeding. This indicates a straightforward bioequivalence between dietary LNA and DHA of about 10 : 1 for brain and retina phospholipid DHA accretion. It should be noted that the weanling guinea pigs are well past their brain growth spurt,


Part II / Phospholipid and Fatty Acid

Fig. 2. Graphical representation of study by Abedin and co-workers. The lower panel presents the dietary fatty acid concentrations fed for 12- to 15-wk-old guinea pigs. The Lo LCP diet is intermediate in LCP content for human breast milks; Hi LCP has threefold higher LCP concentrations. The upper panel shows the outcome variable, DHA concentration in brain or retina phosphatidylethanolamine (PE). Brain and retina DHA were similar between animals fed the Hi LNA diet or Lo LCP. (Based on data from Abedin, Lien, Vingrys, & Sinclair, 1999).

which may explain why no differences in brain or retina DHA were found in the Lo LNA and Med LNA diet groups. Even when there are acute demands for brain DHA, LNA appears to serve as a relatively poor substrate for its synthesis and accretion. Bourre and colleagues (Bourre et al., 1989; Bourre, Youyou, Durand, & Pascal, 1987) found that complete recovery of neural DHA took 2–3 mo in rats initially fed an n-3-deficient diet that were refed with LNA as the sole n-3 fatty acid. These compositional studies all confirm that dietary DHA is considerably more effective than LNA for DHA accretion.

4. DIETARY LNA AND DHA AS SUBSTRATES FOR BRAIN AND RETINA DHA: TRACER STUDIES Data generated from compositional studies cannot distinguish endogenous and exogenous sources of n-3 fatty acids or the metabolic origin of precursors. Therefore, precise conversion ratios can be best established with tracer studies. Early studies were performed with radiotracers in rodents, were mostly short term in nature, and established important aspects of basic physiological handling of LNA and DHA in target organs. A series of recent and ongoing studies were conducted in our laboratory have established quantitative aspects of brain DHA accretion in primates. We first review selected studies in rodents, then discuss our primate results in detail.

4.1. Neonatal Conversion in Rodents Several studies have examined the question of the uptake of unsaturates in the developing brain (Anderson & Connor, 1988; Dhopeshwarkar, Subramanian, & Mead, 1971; Hassma & Crawford, 1976; Sinclair, 1975; Sinclair & Crawford, 1972), with delivery of

Chapter 6 / _-LNA and DHA During Perinatal Development


the radioisotope either orally or by injection. Among them, only two have addressed the relative efficacy of LNA and DHA as precursors for brain DHA accumulation in newborns. It is well established that rats lay down most of their brain DHA postnatally, ending around 30 days of life (Sinclair & Crawford, 1972). In order to isolate the ability of the brain, apart from the liver, as a synthetic site for DHA from LNA, Anderson and co-workers (Anderson & Connor, 1988) compared brain uptake of DHA and LNA by injecting the 5 µCi of each labeled fatty acid intravenously in suckling rats that had been functionally hepatectomized to minimize liver contributions. At 30 min postdose, it was found that brain lipid radioactivity steadily increased with increasing degree of unsaturation (16:0LA>LNA>>>18 : 1. In whole human perfused placenta, the order of selectively for transfer from the maternal to the fetal circulation was reported as DHA>LNA>LA>18 : 1>AA (Haggarty et al., 1997). Other perfusion studies indicate that the human placenta preferentially incorporates PUFA into phospholipids on the fetal side, thereby eliminating the possibility of back transport (Kuhn & Crawford, 1986). Overall, these studies consistently show that the placenta assists in essential fatty acid transport to the fetus. In our maternal-dose study, DHA* plateaued at 15–35 d postdose in both DHA* and LNA* dose groups. This observation prompted us to choose 2 wk postdose as the time for collection of tissues in the neonatal study. By that time, DHA* accretion would have stabilized and would reflect overall, integrated levels of brain DHA (Su et al., 1999a). We computed that a minimum of 92% of the LNA*-derived DHA* was present in the brain at 2 wk postdose, indicating that, at most, a modest 4% per week of brain DHA turnovers during the dose period. In addition to being a strong confirmation of the slow turnover hypothesis, this estimate also shows that the measurement reflects the actual integrated bioequivalence. The term “bioequivalence,” as used here, is chosen to imply a relative efficacy in accretion between two sources of brain DHA, in analogy to the use of the term in reference, for instance, to retinol and `-carotene. The crucial clinical issue for infant formulations is to establish the amount of DHA to be added to LCP-free formulas as a precursor for neonate brain development. In our neonate study, the commercial formula contained 1.8% of calories as LNA, and the only dietary DHA that these animals consumed was from the dose. Thus, the bioequivalence of 7 : 1 applies directly to the addition of small amounts of DHA to formula, meaning that the addition of DHA at 0.26% of calories may provide an equal amount of brain DHA as the entire 1.8% calories as LNA. Factors driving the addition of less DHA include possible interference with AA metabolism, the possibility of contaminants added incidentally in DHA oils, and expense. The potency of DHA relative to LNA suggests that the addition of amounts as small as 0.1% of calories would support brain growth, a figure similar to the lowest levels of DHA found in human breast milk. Finally, we note that the purely biochemical nature of our studies to date cannot establish whether LNA can completely substitute for DHA. Studies in human preterms suggest that it cannot, whereas those in term infants remain controversial (Cunnane, Francescutti, Brenna, & Crawford, 2000).

5. DIETARY LNA METABOLISM OTHER THAN FOR DHA Many studies including our own show that very little LNA is detected in neural tissues, including the brain, retina, and retinal pigment epithelium (RPE). It has long been known that these organs are very low in LNA, despite measurements showing that LNA traverses the blood-brain barrier (Edmond, Higa, Korsak, Bergner, & Lee, 1998). One possible interpretation of the ineffectiveness of LNA as a precursor for brain DHA is the major diversion of LNA to other metabolic pathways. Studies investigating metabolic routes of LNA* administered to animals suggest that very little LNA is sequestered in the brain for making DHA.


Part II / Phospholipid and Fatty Acid

In a feto-maternal rhesus monkey experiment conducted in our laboratory, Scheaff Greiner et al. found that only 0.24% of LNA* appeared in fetal brain DHA after a LNA* dose to the mother, with recycling of carbon from LNA into saturates and monosaturates being the predominant metabolic pathway (Sheaff Greiner et al., 1996). Studies in artificially reared rats show that all brain palmitate is made in the brain from acetate derived from PUFA when the route of consumption is oral (Marbois, Ajie, Korsak, Sensharma, & Edmond, 1992). Other small-animal studies show that 0.1% and 0.02% of labeledLNA (14C-LNA) is found in the brains of guinea pigs (Fu & Sinclair, 2000a) and developing rats, the latter showing preference for LNA use for cholesterol and palmitate of 16 times and 30 times, respectively, over DHA (Menard, Goodman, Corso, Brenna, & Cunnane, 1998). Preferential labeling of palmitate was also found in brains of suckling rats fed orally (Dhopeshwarkar et al., 1971; Sinclair, 1975) or injected intraperitoneally with 14C-LNA (Dhopeshwarkar et al., 1971). Total `-oxidation represents a substantial source of LNA loss from the body. Leyton et al. (Leyton, Drury, & Crawford, 1987) reported that the rate of 14C-LNA oxidation measured as CO2 recovered in reference to administered dose was the fastest among unsaturates, including LA, 18:3n-6, and AA, and at a rate similar to 18 : 1 and 12 : 0, serving as the most efficient energy substrate. A recent tracer study has identified the skin as a previously unidentified route of major loss for LNA in the guinea pig. The 46% of dose found in the nonesterified fatty acid fraction of skin and fur lipids suggests that this use may, in part, account for the poor conversion efficiency to DHA and awaits confirmation by measurements in humans (Fu & Sinclair, 2000b). These data clearly identify significant metabolic roles for LNA other than as the precursor of DHA. Based on the experimental results from many laboratories, we can conclude that LNA may have essential roles other than as a DHA precursor and that consumption of LCP-free diets in the perinatal period puts infants at risk of inadequate DHA for the rapid-growing neural tissue.

6. CONCLUSION Data from compositional studies and tracer studies unequivocally show that DHA is a better substrate for brain and retina DHA accretion compared to LNA. In nontracer studies, the conversion ratio between LNA and DHA to neural DHA reported was over a range of 4–20, depending on outcome variable. In radiotracer studies, DHA* is incorporated into neural tissues at a faster rate compared to the labeled C18 precursor. Although, it has been shown that human infants are capable of converting LNA to DHA in vivo, the relative efficacy of LNA and DHA as the substrate for neural DHA cannot be obtained quantitatively from blood compartments. Nonhuman primate studies have shown that fetal brain DHA accretion reached plateau levels following intravenous doses of either 13C-LNA or 13C-DHA to the pregnant mother or fetus, suggesting brain DHA turnover rate during perinatal is slow and bioequivalence of LNA- or DHA-derived DHA can be computed only after the DHA plateau is reached. Brain bioequivalence obtained from nonhuman primates using stable-isotope methodology is 7 : 1 or 8 : 1 when administered directly to the developing animal, and 20 : 1 when administered to the pregnant female. These data suggest that the addition of modest amounts of DHA to infant formula, as low as 0.1% of calories, should measurably improve DHA status. No definitive evidence yet exists for an essential role for LNA. However, several metabolic roles for LNA have been

Chapter 6 / _-LNA and DHA During Perinatal Development


identified, for which LNA is shunted at much higher rates that to DHA synthesis. The weight of evidence suggests that at least a modest supply of DHA in formula would improve the health of preterm and term infants.

REFERENCES Abedin L, Lien EL, Vingrys AJ, Sinclair AJ. The effects of dietary alpha-linolenic acid compared with docosahexaenoic acid on brain, retina, liver, and heart in the guinea pig. Lipids 1999; 34(5): 475–482. Al MD, Hornstra G, van der Schouw YT, Bulstra-Ramakers MT, Huisjes HJ. Biochemical EFA status of mothers and their neonates after normal pregnancy. Early Hum Dev 1990; 24(3):239–248. Anderson G, Connor WE. Uptake of fatty acids by developing rat brain. Lipids 1988; 23:286–290. Anderson GJ, Connor WE, Corliss JD. Docosahexaenoic acid is the preferred dietary n-3 fatty-acid for the development of the brain and retina. Pediatr Res 1990; 27(1):89–97. Arbuckle LD, Innis SM. Docosahexaenoic acid in developing brain and retina of piglets fed high or low alpha-linolenate formula with and without fish oil. Lipids 1992; 27(2):89–93. Benolken RM, Anderson RE, Wheeler TG. Membrane fatty acids associated with the electrical response in visual excitation. Science 1973; 182:1253–1254. Bourre JM, Durand G, Pascal G, Youyou A. Brain-cell and tissue recovery in rats made deficient in N-3 fattyacids by alteration of dietary-fat. J Nutr 1989; 119(1):15–22. Bourre JM, Piciotti M, Dumont O. Delta-6 desaturase in brain and liver during development and aging. Lipids 1990; 25:354–356. Bourre JM, Youyou A, Durand G, Pascal G. Slow recovery of the fatty acid composition of sciatic nerve in rats fed a diet initially low in n-3 fatty acids. Lipids 1987; 22(7):535–538. Brenna JT. High-precision gas isotope ratio mass spectrometry: recent advances in instrumentation and biomedical applications. Acc Chem Res 1994; 27:340–346. Brody, S. Bioenergetics and Growth, with Special Reference to the Efficiency Complex in Domestic Animals. Reinhold, New York, 1945. Campbell FM, Gordon MJ, Dutta-Roy AK. Preferential uptake of long chain polyunsaturated fatty acids by isolated human placental membranes. Mol Cell Biochem 1996; 155(1):77–83. Carlson SE, Rhoides PG, Rao VS, Goldgar DE. Effect of fish oil supplementation on the n-3 fatty acid content of red blood cell membranes in preterm infants. Pediatr Res 1987; 21:507–510. Carnielli VP, Wattimena DJ. L., Luijendijk IH. T., Boerlage A, Degenhart HJ, Sauer PJJ. The very low birth weight premature infant is capable of synthesizing aracidonic and docosahexaenoic acids from linoleic and linolenic acids. Pediatr Res 1996; 40(1):169–174. Chambaz J, Ravel D, Manier M-C, Pepin D, Mulliez N, Bereziat G. Essential fatty acids interconversion in the human fetal liver. Biol Neonate 1985; 47:136–140. Connor WE, Neuringer M, Reisbick S. Essential fatty acids: the importance of n-3 fatty acids in the retina and brain. Nutr Rev 1992; 50(4):21–29. Crawford MA, Costeloe K, Ghebremeskel K, Phylactos A. The inadequacy of the essential fatty acid content of present preterm feeds [Erratum: Eur J Pediatr 1998; 157(2):160]. Eur J Pediatr 1998; 157(Suppl 1):S23–S27. Cunnane SC, Francescutti V, Brenna JT, Crawford MA. Breast-fed infants achieve a higher rate of brain and whole body docosahexaenoate accumulation than formula-fed infants not consuming dietary docosahexaenoate. Lipids 2000; 35(1):105–111. Dhopeshwarkar GA, Subramanian C, Mead JF. Fatty acid uptake by the brain. V. Incorporation of (I-14C)linolenic acid into adult rat brain. Biochim Biophy Acta 1971; 239(2):162–167. Dobbing J, Sands J. Comparative aspects of the brain growth. Early Hum Dev 1979; 3(1):79–83. Dutta-Roy AK. Transport mechanisms for long-chain polyunsaturated fatty acids in the human placenta. Am J Clin Nutr 2000; 71(Suppl 1):315S-322S. Edmond J, Higa TA, Korsak RA, Bergner EA, Lee WN. Fatty acid transport and utilization for the developing brain. J Neurochem 1998; 70(3):1227–1234. Farquharson J, co*ckburn F, Patrick WA, Jamieson EC, Logan RW. Infant cerebral cortex phospholipid fattyacid coposition and diet. Lancet 1992; 340:810–813. Farquharson J, Jamieson EC, Abbasi KA, Patrick WJA, Logan RW, co*ckburn F. Effect of diet on the fatty acid compostition of the major phospholipids of infant cerebral cortex. Arch Dis Child 1995; 72:198–203.


Part II / Phospholipid and Fatty Acid

Fliesler SJ, Anderson RE. Chemisty and metabolism of lipids in the vertebrate retina. Prog Lipid Res 1983; 22:79–131. Fu Z, Sinclair AJ. Increased alpha-linolenic acid intake increases tissue alpha-linolenic acid content and apparent oxidation with little effect on tissue docosahexaenoic acid in the guinea pig. Lipids 2000; 35(4):395–400. Fu Z, Sinclair AJ. Novel pathway of metabolism of alpha-linolenic acid in the guinea pig. Pediatr Res 2000; 47(3):414–417. Giussani DA, Farber DM, Jenkins SL, Yen A, Winter JA, Tame JD, Net al. Opposing effects of androgen and estrogen on pituitary–adrenal function in nonpregnant primates. Biol Reprod 2000; 62(5):1445–1451. Goodman KJ, Brenna JT. High sensitivity tracer detection using high precision gas chromatography combustion isotope ratio mass spectrometry and highly enriched [U-13C]-labeled precursors. Anal Chem 1992; 64(10), 1088–1095. Greiner RC, Winter J, Nathanielsz PW, Brenna JT. Brain docosahexaenoate accretion in fetal baboons: bioequivalence of dietary alpha-linolenic and docosahexaenoic acids. Pediatr Res 1997; 42(6):826–834. Haggarty P, Page K, Abramovich DR, Ashton J, Brown D. Long-chain polyunsaturated fatty acid transport across the perfused human placenta. Placenta 1997; 18(8):635–642. Hassma AG, Crawford MA. The differential incorporation of labelled linoleic, gamma-linolenic, dihom*ogamma-linolenic and arachidonic acids into the developing rat brain. J Neurochem 1976; 27(4):967–968. Heird WC, Prager TC, Anderson RE. Docosahexaenoic acid and the development and function of the infant retina. Curr Opin Lipidol 1997; 8(1):12–16. Hrboticky N, Mackinnon MJ, Innis SM. Effect of a vegetable oil formula rich in linoleic-acid on tissue fattyacid accretion in the brain, liver, plasma, and erythrocytes of infant piglets. Am J Clin Nutr 1990; 51(2):173–182. Innis SM. Essential fatty acids in growth and development. Prog Lipid Res 1991; 30(1):39–103. Jensen CL, Prager TC, Fraley JK, Chen H, Anderson RE, Heird WC. Effect of dietary linoleic/alpha-linolenic acid ratio on growth and visual function of term infants. J Pediatr 1997; 131(2):200–209. Jensen RG. The lipids in human milk. Prog Lipid Res 1996; 35(1):53–92. Jensen RG, Hagerty MM, McMahon KE. Lipids of human milk and infant formulas: a review. Am J Clin Nutr 1978; 31(6):990–1016. Knipp GT, Audus KL, Soares MJ. Nutrient transport across the placenta. Adv Drug Deliv Rev 1999; 38(1):41–58. Kuhn DC, Crawford MC. Placental essential fatty acid transport and prostaglandin synthesis. Prog Lipid Res 1986; 25:345–353. Lands WE. Long-term fat intake and biomarkers. Am J Clin Nutr 1995; 61(3 Suppl), 721S–725S. Leyton J, Drury PJ, Crawford MA. In vivo incorporation of labeled fatty acids in rat liver lipids after oral administration. Lipids 1987; 22(8):553–558. Ma Y, Lockwood CJ, Bunim AL, Giussani DA, Nathanielsz PW, Guller S. Cell type-specific regulation of fetal fibronectin expression in amnion: conservation of glucocorticoid responsiveness in human and nonhuman primates. Biol Reprod 2000; 62(6):1812–1817. Makrides M, Neumann MA, Byard RW, Simmer K, Gibson RA. Fatty acid composition of brain, retina, and erythrocytes in breast- and formula-fed infants. Am J Clin Nutr 1994; 60(2):189–194. Marbois BN, Ajie HO, Korsak RA, Sensharma DK, Edmond J. The origin of palmitic acid in brain of the developing rat. Lipids 1992; 27(8):587–592. Martinez M, Ballabriga A, Gil-Gibernau JJ. Lipids of the developing human retina: I. Total fatty acids, plasmologens, and fatty acid compostition of ethanolamine and choline phosphoglycerides. J Neurosci Res 1988; 20:484–490. Menard CR, Goodman KJ, Corso TN, Brenna JT, Cunnane SC. Recycling of carbon into lipids synthesized de novo is a quantitatively important pathway of alpha-[U-13C]linolenate utilization in the developing rat brain. J Neurochem 1998; 71(5):2151–2158. Mohrhauser H, Holman RT. The effect of dose level of essential fatty acids upon fatty acid composition of the rat liver. J Lipid Res 1963; 6:494–497. Naval J, Calvo M, Laborda J, Dubouch P, Frain M, Sala-Trepat JM, et al. Expression of mRNAs for alphafetoprotein (AFP) and albumin and incorporation of AFP and docosahexaenoic acid in baboon fetuses. J Biochem (Tokyo) 1992; 111(5):649–654. Neuringer M, Connor W, Lin D, Barsted L, Luck S. Biochemical and functional effects of prenatal and postnatal omega 3 fatty acid deficiency on retina and brain in rhesus monkeys. Proc Natl Acad of Sci USA 1986; 83:4021–4025.

Chapter 6 / _-LNA and DHA During Perinatal Development


Neuringer M, Connor WE, Petten CV, Barstad L. Dietary omega-3-fatty acid deficiency and visual loss in infant rhesus monkeys. J Clin Invest 1984; 73:272–276. Pawlosky RJ, Ward G, Salem N Jr. Essential fatty acid uptake and metabolism in the developing rodent brain. Lipids 1996; 31(Suppl):S103–S107. Poisson JP, Dupuy RP, Sarda P, Descomps B, Narce M, Rieu D, et al. Evidence that liver-microsomes of human neonates desaturate essential fatty-acids. Biochim Biophy Acta 1993; 1167(2):109–113. Rioux FM, Innis SM, Dyer R, MacKinnon M. Diet-induced changes in liver and bile but not brain fatty acids can be predicted from differences in plasma phospholipid fatty acids in formula- and milk-fed piglets. J Nutr 1997; 127(2):370–377. Rodriguez A, Sarda P, Nessmann C, Boulot P, Leger CL, Descomps, B. Delta6- and delta5-desaturase activities in the human fetal liver: kinetic aspects. J Lipid Res 1998; 39(9):1825–1832. Ruyle M, Connor WE, Anderson GJ, Lowensohn RI. Placental-transfer of essential fatty-acids in humans— venous arterial difference for docosahexaenoic acid in fetal umbilical erythrocytes. Proc Natl Acad Sci USA 1990; 87(20):7902–7906. Salem JN, Wegher B, Mena P, Uauy R. Arachidonic and docosahexaenoic acids are biosynthesized from their 18-carbon precursors in human infants. Proc Natl Acad Of Sci USA 1996; 93:49–54. Sastry PS. Lipids of nervous tissue: composition and metabolism. Prog Lipid Res 1985; 24:69–176. Sauerwald TU, Hachey DL, Jensen CL, Chen H, Anderson RE, Heird WC. Intermediates in endogenous synthesis of C22:6 omega 3 and C20:4 omega 6 by term and preterm infants. Pediatr Res 1997; 41(2):183–187. Sheaff Greiner RC, Zhang Q, Goodman KJ, Giussani DA, Nathanielsz PW, Brenna JT. Linoleate, alphalinolenate, and docosahexaenoate recycling into saturated and monounsaturated fatty acids is a major pathway in pregnant or lactating adults and fetal or infant rhesus monkeys. J Lipid Res 1996; 37(12):2675–2686. Sinclair AJ. Incorporation of radioactive polyunsaturated fatty acids into liver and brian of developing rat. Lipids 1975; 10(3):175–184. Sinclair AJ, Crawford MA. The incorporation of linolenic acid and docosahexaenoic acid into liver and brain lipids of developing rats. Fed Exp Biol Soc Lett 1972; 26:127–129. Stammers J, Stephenson T, Colley J, Hull D. Effect on placental transfer of exogenous lipid administered to the pregnant rabbit. Pediatr Res 1995; 38(6):1026–1031. Su HM, Bernardo L, Mirmiran M, Ma XH, Corso TN, Nathanielsz PW, et al. Bioequivalence of dietary alphalinolenic and docosahexaenoic acids as sources of docosahexaenoate accretion in brain and associated organs of neonatal baboons. Pediatr Res 1999; 45(1):87–93. Su HM, Bernardo L, Mirmiran M, Ma XH, Nathanielsz PW, Brenna JT. Dietary 18:3n-3 and 22:6n-3 as sources of 22:6n-3 accretion in neonatal baboon brain and associated organs. Lipids 1999; 34(Suppl):S347–S350. Su H-M, Corso TN, Nathanielsz PW, Brenna JT. n-3 Fatty acid accretion after a 13C-18:3n-3 dose to fetal baboons. Third Congress of the International Society for the Study of Fatty Acids and Lipids, Lyon, 1998. Tame JD, Winter JA, Li C, Jenkins S, Giussani DA, Nathanielsz PW. Fetal growth in the baboon during the second half of pregnancy. J Med Primatol 1998; 27(5):234–239. Uauy R, Mena P, Wegher B, Nieto S, Salem N Jr. Long chain polyunsaturated fatty acid formation in neonates: effect of gestational age and intrauterine growth. Pediatr Res 2000; 47(1):127–135. Uauy RD, Birch DG, Birch EE, Tyson JE, Hoffman DR. Effect of dietary omega-3–fatty-acids on retinal function of very-low-birth-weight neonates. Pediatr Res 1990; 28(5):485–492. Wheeler TG, Benolken RM. Visual membranes: specificity of fatty acid precursors of the electrical response to illumination. Science 1975; 188(4195):1312–1314. Woods J, Ward G, Salem NJ. Is docosahexaenoic acid necessary in infant formula? Evaluation of high linolenate diets in the neonatal rat. Pediatr Res 1996; 40:687–694. Yamamoto N, Saitoh M, Moriuchi A, Nomura M, Okuyama H. Effect of dietary alpha-linolenate/linoleate balance on brain lipid compostions and learning ability of rats. J Lipid Res 1987; 28:144–151.

Chapter 7 / Supply of DHA to the CNS



Recent Advances in the Supply of Docosahexaenoic Acid to the Nervous System Robert J. Pawlosky and Norman Salem Jr.

1. DHA IN THE MEMBRANE ENVIRONMENT OF THE NEURON Docosahexaenoic acid (DHA, 22:6n-3) is the most abundant polyunsaturated fatty acid (PUFA) acylated to the aminophospholipids phosphatidylethanolamine (PE) and phosphatidylserine (PS) in membranes of neurons within the central nervous system (CNS) (Naughton, 1981; Salem et al., 1986). It can occur in concentrations exceeding 30-mol% of the fatty acids (Salem, 1989). The high enrichment of DHA in synaptosomes is especially striking and suggests that DHA has unique properties that are required for optimal neuronal function. This concentration in the CNS is even more remarkable when one considers that sources of n-3 fatty acids are disproportionately limited in the terrestrial food chain compared to the much more abundant n-6 fatty acids. Although the precise function of DHA in the membrane has yet to be determined, the biophysical properties of membranes that are enriched with this fatty acid appear to be optimized for signal transduction through neuronal pathways. An example illustrating this apparent optimization is the visual transduction process through the photoreceptor cells and neurons within the retina. DHA is more highly enriched in the outer segment disks of rod and cone cells than in any compartment of any other cell type (Salem, 1989). The high concentration of DHA in these membranes relative to other fatty acids appears to be important for facilitating the formation of the photon-initiated transition state of rhodopsin (Litman & Mitchell, 1996). Considering that DHA may impart unique properties to membrane domains nuclear magnetic resonance (NMR) investigations have been used to determine changes in the order parameters of the carbon atoms of the fatty acid acylated at the sn-1 position of phospholipids in relation to the degree of unsaturation of the fatty acid at the sn-2 position (Holte et al., 1995; Holte et al., 1996; Mitchell & Litman, 1998). The results have suggested that phospholipids with fatty acids having a high degree of unsaturation acylated at the sn-2 position can give rise to highly transitional molecular shapes, which may impart unique packing properties within the matrix of the neuronal membrane. It is reasonable to infer from the biophysical data that rapid alterations in the protein conformation of rhodopsin may require a membrane environment capable of transitioning through highly alterable dynamic states (Holte et al., 1996). The high concentration of From: Fatty Acids: Physiological and Behavioral Functions Edited by: D. Mostofsky, S. Yehuda, and N. Salem Jr. © Humana Press Inc., Totowa, NJ



Part II / Phospholipid and Fatty Acid

DHA in the rod outer segments may impart unique membrane characteristics adapted to facilitating rhodopsin activation and association with G-proteins. With regard to the physiological consequences of the fatty acid composition of disk membranes, several studies have documented the effects that a low intake of dietary n-3 fatty acids has on the electroretinograms (ERG) of young animals that may be directly attributed to the low content of DHA in the disks (Neuringer et al., 1986; Weisinger et al., 1996; Pawlosky et. al., 1997). In most species (Neuringer et al., 1986; Connor & Neuringer, 1988; Pawlosky et. al., 1997), the n-6 fatty acid hom*olog of DHA, DPAn-6 (22:5n6), increases in concentration in the disks of animals fed the low n-3 fatty acid diet. The preponderance of evidence from these studies demonstrate convincingly that low amounts of DHA in the retina result in a diminished rod cell response to light stimulus as observed in the altered voltage potentials in the electroretinograms. It is theoretically plausible, then, to presume that “enhanced” signal transduction through other neuronal pathways that are responsive to ligand-mediated G-protein activation (analogous to lightactivated rhodopsin stimulation) may be related to the concentration of DHA in synaptic membranes. Conversely, it may be proposed that dietary deficiencies, disease states, genetic abnormalities, or environmental conditions that diminish the supply of DHA to neurons will adversely affect cell signaling in the central nervous system.

2. GENETIC AND DIETARY FACTORS WHICH INFLUENCE N-3 FATTY ACID METABOLISM Because of the inability to synthesize n-3 fatty acids de novo, all animals require these fatty acids in their diet to meet their demand for maintaining a high concentration of DHA in the brain. Although little direct evidence exists in any species concerning the quantitative conversion of n-3 fatty acid precursors to DHA, it has been estimated based on rodent studies that an n-3 fatty acid intake of 0.5% of energy as _-linolenic acid (LNA) is needed in order to maintain an adequate level of DHA in the brain (Bourre et al., 1989). However, it must be recognized that the ability to biosynthesize DHA from LNA or other n-3 fatty acids varies among different animal species (Rivers et al., 1975; Hassam et al., 1977; Sinclair et al., 1979; Clandinin et al., 1985; Scott & Bazan, 1989: Salem & Pawlosky, 1994; Pawlosky et al., 1994: Fu & Sinclair, 2000). Moreover, the composition of fat in the diet has a significant influence on the liver production of long-chain PUFAs (Salem & Pawlosky, 1994; Pawlosky et al., 1994). For instance, it was observed that when nonhuman primates were fed a diet that contained relatively low levels of longchain PUFAs (where eicosapentaeonic acid [EPA] and DHA were present at a level of 0.54% and 0.64% of the total dietary fat, respectively) the formation of labeled-DHA from labeled-LNA was inhibited in the liver (Pawlosky & Salem, 1993). However, both arachidonic acid (AA) (from labeled-LA) and docosapentaenoic acid (DPAn-3) were synthesized in the liver and detected in the blood of the same animals on this diet. When animals were then placed on a diet devoid of long-chain PUFAs, the synthesis of DHA was observed in the liver, and labeled-DHA was detected in the blood after 3 wk. This strongly suggests that the conversion of DPAn-3 to DHA in the liver is partly controlled by the concentration of DHA in the diet. It is interesting to theorize whether the regulation of the biosynthesis of DHA from DPAn-3 is maintained at the level of transcription of a 6-6 desaturase which is needed to catalyze the conversion of 24:5n3 to 24:6n-3 (Marzo et al., 1996). If so, this form of regulation would have the advantage of selectively

Chapter 7 / Supply of DHA to the CNS


controlling DHA production, yet it would not necessarily inhibit the synthesis of other long-chain PUFAs (e.g., arachidonic acid). Because of the genetic and dietary factors, which are capable of controlling and influencing the production of long-chain PUFAs, different species have developed various independent strategies for obtaining DHA. In the cat family, for instance, preformed DHA in the diet appears to be necessary to maintain a high concentration of DHA in the CNS (Pawlosky et al., 1997). The need for preformed DHA in the diet is probably caused by a low PUFA biosynthetic capability of this species (Hassam, 1977; Rivers, 1975; Sinclair, 1979) as well as an inherent inability to produce DHA in the feline liver (Pawlosky et al., 1994). However, there is increasing evidence that suggests that the production of DHA from LNA may be a highly inefficient process in other species, as well (Menard et al., 1998; Su et al., 1999). Nevertheless, it appears that the majority of species (other than members of the cat family) have some capacity to biosynthesize DHA from LNA in their livers. Although, the liver has long been recognized as an important site of PUFA biosynthesis (Buzzi et al., 1997; Clandinin et al., 1985; Pawlosky et al., 1992; Schenck et al, 1996), a number of animal studies in various species have shown that long-chain PUFAs (in particular, 22:6n-3) can be synthesized by different components of the nervous system (Dhopeswarkar et al., 1974; Clandinin et al., 1985; Delton-Vandenbrouke et. al, 1997; Chen et. al, 1999; Moore, 1993; Moore et al., 1991; Pawlosky et al., 1994; Pawlosky et al., 1996; Protstein, 1996). The cells of the nervous system, like other cells of the body, take up DHA and other n-3 fatty acids from lipoproteins that are carried in the blood. There is evidence that the preferred form of DHA for uptake into the brain is as a lysophospholipid rather than as a free fatty acid (Bernoud et al., 1999). This route may offer an efficient transfer of DHA into the neuron for phospholipid synthesis and membrane biogenesis. Although felines are the only species in which it has been demonstrated that the entire brain accretion of biosynthesized DHA is the result of production that occurs within the CNS (Pawlosky et al., 1994) other species carry on similar intra-CNS processes to obtain at least part of their DHA (Pawlosky et al., 1996). Figure 1 depicts a representation of the feline strategy for the accretion of DHA in brain. From dietary sources, LNA or EPA is taken up into the liver where the fatty acids are converted into DPAn-3. DPAn-3 is released from the liver and carried on lipoproteins to the CNS, where it is converted to DHA. There is similar evidence from other species that have shown that brain cells (Moore et al., 1991) or cells isolated from the cerebral vasculature (DeltonVandenbroucke et al., 1997) are capable of biosynthesizing DHA from n-3 fatty acid precursors. Several investigators have described plausible mechanisms for the production of DHA in the CNS. Moore and co-workers described the biosynthesis of DHA and transport of fatty acids through microcapillary cerebral endothelial cells, astrocytes, and neurons (Moore et al., 1991; Moore, 1993). In this model, microcapillary endothelial cells produce DPAn-3 from LNA, which is turned over to the astrocytes to be synthesized into DHA. The astrocytes then release DHA, which is taken up by neurons. In contrast, Delton-Vandenbroucke and co-workers found that cerebral vascular cells produced appreciable amounts of labeled-DHA from DPAn-3 (Delton-Vandenbroucke et al., 1997). They theorize that cerebral endothelial cells will convert circulating DPAn-3 into DHA, which is then taken up into the brain. These reports suggest that in the CNS, unlike the liver in which biosynthesis of DHA from LNA takes place entirely within the hepatocyte,


Part II / Phospholipid and Fatty Acid

Fig. 1. A schematic representation of the biosynthesis and accretion of DHA into the CNS of felines. Dietary n-3 fatty acids (LNA) are taken up into the liver and synthesized into DPAn-3 (22:5n-3). DPAn-3 is released from the liver and carried on lipoproteins in the blood to the nervous system. The synthesis of DHA is completed in the brain from DPAn-3.

more than a single cell type (either an astrocyte or endothelial cell) may act in a synergistic fashion to contribute to the synthesis and accretion of DHA.

3. DHA BIOSYNTHESIS AND EARLY DEVELOPMENT The accumulation of DHA in the brain is especially important during brain growth periods (Green & Yavin, 1998), and although no known systematic study has been undertaken which attempts to compare rates of DHA biosynthesis to the development of the CNS or brain growth in any species, there is evidence to suggest that the capacity to synthesize DHA may be correlated with early brain development in some species (Rodriguez et al., 1998; Pawlosky et. al., 1996; Salem et al., 1996; Greiner et al., 1997; Su et al., 1999). Using stable isotopically labeled fatty acids and mass spectrometry, the biosynthesis of DHA has been demonstrated in both human infants (Salem et al., 1996; Carnielli et al., 1996; Sauerwald et al., 1996) and in fetal baboons (Greiner et al., 1997; Su et al., 1999). An example that illustrates the inherent capacity to synthesize long-chain PUFAs during the early developmental period was provided by felines (Pawlosky & Salem, 1996). Adult felines do not actively synthesize long-chain PUFAs from the 18-carbon precursors (Sinclair et al., 1979; Pawlosky et al., 1994). However, juvenile felines before weaning were capable of synthesizing labeled long-chain PUFAs in their livers, which could then be detected in the blood using mass spectrometry. When the mothers were given an oral dose of labeled-18-carbon essential fatty acids, neither the maternal blood nor milk contained any of the biosynthesized long-chain PUFAs. Apparently, the demands of pregnancy and lactation did not provide sufficient stimulus to activate the biosynthetic pathway in adult cats. As the young animals developed (and begin accepting meat in their

Chapter 7 / Supply of DHA to the CNS


diet), the ability to synthesize long-chain PUFAs diminished. This active capacity to synthesize long-chain PUFAs in juvenile felines is intriguing and suggests that development may be a significant factor, which triggers an enhanced biosynthesis of long-chain PUFAs during nervous system formation. Notably, the capacity to synthesize DHA in immature felines may not be sufficient to meet their brain requirement for DHA. It was observed that animals reared on any of several corn-oil-based diets had very low amounts of either DPAn-3 or DHA in their brains at 8 wk of age (Pawlosky et al., 1997). It may be inferred from this study that throughout feline development, the maternal diet must contain some amount of preformed DHA (EPA or DPAn-3 may partially substitute for DHA) for adequate accumulation of brain DHA.

4. SUPPLYING DHA TO THE BRAIN: RODENT AND FELINE MODELS OF DHA SYNTHESIS Although much of our understanding concerning the regulation of DHA biosynthesis has been ascertained from studies in rodents, it does not appear that either rats or mice are the most appropriate model for understanding n-3 metabolism in humans. Rodents seem to be more adept at synthesizing long-chain PUFAs than either felines or rhesus monkeys and are less influenced by dietary alterations (Salem & Pawlosky, 1994b). Rodents maintain a capacity to synthesize DHA in their livers (Scott & Bazan, 1989) as well as in the CNS (Pawlosky et al., 1996b) and have more active desaturases than several other species when measured in vitro (Willis, 1981). Using stable isotopically labeled substrates, Pawlosky and Salem demonstrated that DHA precursors could be taken up into the brain of developing rats and mice at a time when the brain is rapidly growing (Pawlosky et al, 1996b). The uptake of labeled-DPAn-3 into the brain appeared to be appreciable, as there was a ratio of approximately 1 : 5 of labeled-DPAn-3 to that of labeledDHA in whole-brain preparations. Both of these labeled-fatty acids were synthesized from labeled-LNA in the liver and carried in the blood to the brain. Gradually, there was a disappearance of labeled-DPAn-3 from the brain. Over the same period (about 7 d), the labeled-DHA continued to increase. The quantitative importance of the brain biosynthetic pathway for supplying part of the CNS with DHA during development is unknown, but based on differential uptake of labeled-LNA into the rapidly growing cerebellum compared to that of the frontal cortex region, it is clear that these precursors are indeed taken into the brain parenchyma during CNS development. Early studies in domestic felines demonstrated that cats had a low 6-6 desaturase activity, which severely limited their capacity to synthesize arachidonic acid from linoleic acid (Hassam et al., 1977; Sinclair et al., 1979). Owing to a low desaturase activity, it may be assumed that synthesis of long-chain n-3 PUFAs arising from LNA would also be very limited. It was later shown that a low essential fatty acid diet could stimulate the synthesis of both long-chain n-6 and n-3 PUFAs via a 6-6 desaturase (Pawlosky et al., 1994). In the liver, the route for the biosynthesis of DHA from LNA is initiated on smooth endoplasmic reticulum. Through a series of alternating enzymatic processes that desaturate and elongate LNA, DPAn-3 is produced. It is believed that DPAn-3 is then elongated to 24:5n3 and desaturated (by a 6-6 desaturase) to 24:6n-3 (Luthria et al., 1996). This fatty acid is transferred to peroxisomes, where it is partially oxidized to form DHA, which is then reincorporated into phospholipids in the microsomal membranes. Felines provide an interesting model for studying the accretion of DHA in the CNS because they lack the


Part II / Phospholipid and Fatty Acid

capacity to produce any DPAn-6 or DHA in their livers. The ability to synthesize DPAn-3 but not DHA in their livers indicates that the second desaturation by 6-6 desaturase may not be active. Consequently, in order to maintain a high level of DHA in the brain, felines must carry out the conversion of DPAn-3 to DHA within the CNS. The 6-6 desaturase in the brain may thus be specific for catalyzing the conversion of 24:5n3 to 24:6n-3. The compartmental separation of the different desaturation steps lends support to the suggestion that distinct forms of 6-6 desaturase are needed in the production of DHA (Marzo et al., 1996).

5. ACCRETION OF DHA INTO THE CNS OF YOUNG BABOONS The synthesis and accretion of DHA in the CNS has been studied in fetal and neonatal baboons (Greiner et al., 1997; Su et al, 1999). Unlike LA, which may be extensively esterified to membrane phospholipids, LNA is present only in small amounts in membrane complex lipids. Presumably most of the available plasma LNA would be used for synthesis of long-chain PUFAs. However, when 4-wk-old baboons were given oral doses of either 13C-labeled LNA or 13C-labeled-DHA, only about 0.2% of labeled-LNA was found in the brain as 13C-labeled DHA after 2 wk (Su et al., 1999). Such low utilization of LNA for brain accretion of DHA suggests a highly inefficient process in the biosynthesis of DHA during a period of rapid brain growth. Utilization of labeled-DHA for brain accretion of DHA was nearly seven times more efficient than that of LNA. The amount of labeled-DHA in the retina was between 12-fold and 15-fold greater in animals receiving labeled-DHA compared to those receiving the labeled-LNA. Part of the inefficient use of LNA for DHA synthesis may be explained by the apparent recycling of the carbon atoms of LNA into other pathways. Cunnane and co-workers demonstrated that a large proportion of LNA is partially oxidized and returned to the acetate pool for synthesis into nonessential fatty acids and cholesterol in neonatal rats (Menard et al, 1998). Also, Fu and Sinclair recently found that much of the radiolabeled LNA fed to young guinea pigs was presumed oxidized (39%) or found as labeled-LNA in the fur and skin (46%) (Fu & Sinclair, 2000). In humans, a high oxidation rate of LNA has also been observed when it was found that as much as 20% of the 13C-lableled LNA was oxidized and expired as CO2 within the first 12 h after receiving an oral dose (Vermunt et al., 2000). The inefficient conversion of LNA to DHA and the large loss and recycling of LNA into other metabolic pools is puzzling and requires further investigation to be fully explained.

6. CONSERVATION OF DHA IN THE CNS Conservation of DHA in the CNS may be an especially important determinant in the maintenance of DHA in the brain and retina. Brenna and co-workers estimated that the DHA turnover in the brain of neonatal baboons was low, only 4% per week based on stable isotope analysis (Brenna et al., 1999). The epithelial cells that line the retina and form a barrier between the photoreceptor cells and the blood appear to be specialized for the uptake of n-3 fatty acids (Rodriguez de Turco et al., 1991; Wang & Anderson, 1993). In the retina, a series of isotope studies have demonstrated that DHA may be conserved through a recycling process that involves the transfer of DHA between the pigment epithelial tissue and the photoreceptor cells (Stinson et al., 1991; Gordon et al., 1992; Rodriguez et al., 1999). It appears that DHA can be recycled to the rod cells from sloughed off outer disk membranes after phagocytosis by the retinal pigment epithelial cells (Gor-

Chapter 7 / Supply of DHA to the CNS


don et al., 1992). This process appears effective in maintaining the high concentration of DHA in photoreceptors as the disks undergo constant membrane biogenesis and renewal.

7. SUMMARY The development of a fully quantitative procedure to determine mechanisms for accretion of DHA in the brain of any species is needed. Species differences, aged-related determinants, and diet appear to be the major factors controlling and influencing the biosynthesis and accretion of DHA in the brain. The use of stable-isotope-labeled substrates in animal and human studies has provided a powerful tool for determining the metabolic fate of the fatty acids in the synthesis of DHA (Brenna, 1994; Brenna et al., 1997, Pawlosky et al., 1992; Sauerwald et al., 1997). These studies have revealed important new information in the area of metabolism of fatty acids in general (Salem et al., 1996; Salem et al., 1999; Greiner et al., 1997; Su et al, 1999) and specifically within the brain (Pawlosky et al., 1994; Menard et al., 1998; Pawlosky et al., 1996). It appears from several lines of evidence in animals that much of the ingested LNA is not available for synthesis of DHA. It may be recycled into cholesterol or nonessential fatty acids, lost to oxidation (Menard et al., 1998) or taken up by other tissue compartments (Fu & Sinclair, 2000). In studying in vivo metabolism in humans, it appears that the biosynthesis of DHA in the liver is similar to other species in that most of the available LNA is not converted to DHA (Salem et al., 1999; Vermunt et al., 2000). The complexity in determining the in vivo metabolism of essential fatty acids and the influences of diet, genetics, age, gender, and disease requires the use of more sophisticated analytical tools to comprehend the various interactions of these parameters. Mathematical approaches utilizing physiologic compartmental models are now available to researchers to be used for fuller descriptive analysis. Such modeling programs can assess the complex interactions of the kinetics of metabolism (isotope data), dietary conditions (specific dietary intake values), tissue steady-state determinants (the homeostasis within compartments), and population statistics (subject variability) through a diverse set of experimental and clinical conditions. Mathematical approaches can be employed to describe quantitative differences among animal species, the effects of development and aging on EFA metabolism, and alterations in dietary habits in a individual or throughout a given population. It is expected that as such new approaches are applied to questions of brain accretion of DHA, a well-formed biochemical understanding and better nutritional guidelines will become available, leading to the adoption of sound nutritional policies. The use of such mathematical approaches to describe EFA metabolism in humans is in the near future. It is expected that one of the immediate applications of these new quantitative approaches will be the application to questions concerning the adequacy of infant formulas in supplying both long-chain n-6 and n-3 PUFAs to the brain.

REFERENCES Bernoud N, Fenart L, Moliere P, Dehouck MP, Lagarde M, Cecchelli R, et al. Preferential transfer of 2-docosahexaenoyl-1-lysophosphatidylcholine through an in vitro blood-brain barrier over unesterified docosahexaenoic acid. J Neurochem 1999; 72(1):338–345. Bourre JM, Durand G, Pascal G, Youyou A. Brain cell and tissue recovery in rats made deficient in n-3 fatty acids by alterations of dietary fat. J Neurochem 1989; 119:12–22. Brenna JT. Use of stable isotopes to study fatty acid and lipoprotein metabolism in man. Prostaglandins Leukotrienes Essential Fatty Acids 1997; 57(4&5):467–472.


Part II / Phospholipid and Fatty Acid

Brenna JT. High-precision gas isotope ratio mass spectrometry: recent advances in instrumentation and biomedical applications. Arch Chem Res 1994; 27:340–346. Buzzi M, Henderson RJ, Sargent JR. Biosynthesis of docosahexaenoic acid in trout hepatocytes proceeds via 24-carbon intermediates. Biochem Physiol 1997; 116B(2):263–267. Carnielli VP, Wattimena DJ, Luijendijk IH, Boerlage A, Degenhart HJ, Sauer JP. The very low birth weight premature infant is capable of synthesizing arachidonic and docosahexaenoic acids from linoleic and linolenic acids. Pediatr Res 1996; 40(1):169–174. Chen H, Ray J, Scarpino V, Acland GM, Aguirre GD, Anderson RE. Synthesis and release of docosahexaenoic acid by the RPE cells of prcd-affected dogs. Invest Ophthalmol Vision Sc 1999; 40:2418–2422. Clandinin MT, Wong K, Hacker RR. Synthesis of chain elongation–desaturation products of linoleic acid by liver and brain microsomes during development of the pig. Biochem J 1985; 226:305–309. Connor WE, Neuringer M. The effects of n-3 fatty acid deficiency and repletion upon the fatty acid composition and function of the brain and retina. In: Karnovsky ML, Leaf A, Bolis LC, eds. Biological Membranes: Aberrations in Membrane Structure and Function. Alan R Liss, New York, 1988, pp. 275–294. Delton-Vandenbrouke I, Grammas P, Anderson RE. Polyunsaturated fatty acid metabolism in retinal and cerebral microvascular endothelial cells. J Lipid Res 1997; 38:147–159. Dhopeshwarkar GA, Subranamanian C. Metabolism of linolenic acid in developing brain: I. Incorporation of radioactivity from 1-14C-linolenic acid into brain fatty acids. Lipids 1974; 10:238–241. Fu Z, Sinclair AJ. Novel pathway of the metabolism of alpha-linolenic acid in the Guinea pig. Pediatr Res 2000; 47(3):414–417. Gordon WC, Rodriguez de Turco EB, Bazan NG. Retinal pigment epithelial cells play a central role in the conservation of docosahexaenoic acid by photoreceptor cells after shedding and phagocytosis. Curr Eye Res 1992; 11(1):73–83. Green P, Yavin E. Mechanisms of docosahexaenoic acid accretion in the fetal brain. J Neurol Res 1998; 52:129–136. Greiner RC, Winter J, Nathanielsz PW, Brenna JT. Brain docosahexaenoate accretion in fetal baboons: Bioequivalence of dietary alpha-linolenic and docosahexaenoic acids. Pediatr Res 1997; 42(6):826–834. Hassam AG, Rivers JPW, Crawford MA. The failure of the cat to desaturate linoleic acid: its nutrient implications. Br J Nutr 1977; 39:227–231. Holte LL, Peter SA, Sinnwell TM, Gawrisch K. 2H nuclear magnetic resonance order parameter profiles suggest a change of molecular shape for phosphatidylcholines containing a polyunsaturated acyl chain. Biophys J 1995; 68(6):2396–2403. Holte LL, Separovic F, Gawrisch K. Nuclear magnetic resonance investigation of hydrocarbon chain packing in bilayers of polyunsaturated phospholipids. Lipids 1996; 31:S199–S203 Litman J, Mitchell DC. A role for phospholipid polyunsaturation in modulating membrane protein function. Lipids 1996; 31S:S193–S197. Luthria DL, Mohammed BS, Sprecher WH. Regulation of the biosynthesis of 4,5,10,13,16,19-docosahexaenoic acid. J Biol Chem 1996; 271(27):16,020–16,025. Marzo I, Alva AA, Pineiro A, Naval J. Biosynthesis of docosahexaenoic acid in human cells: evidence that two different delta-6 desaturase activities may exist. Biochim Biophys Acta 1996; 1301:263–272. Menard CR, Goodman KJ, Corso TN, Brenna JT, Cunnane SC. Recycling of carbon into lipids synthesized de novo is a quantitatively important pathway of alpha-[U-13C] linolenate utilization in the developing rat brain. J Neurochem 1998; 71:2151–2158. Mitchell DC, Litman BJ. Molecular order and dynamics in bilayers consisting of highly polyunsaturated phospholipids. Biophys J 1998; 74:879–891. Moore SA. Cerebral endothelium and astrocytes cooperate in supplying docosahexaenoic acid to neurons. Adv Exp Med Biol 1993; 331:229–233. Moore SA, Yoder E, Murphy S, Dutton GR, Spector AA. Astrocytes not neurons produce docosahexaenoic acid (22:6n-3) and arachidonic acid (20:4n-6). J Neurochem 1991; 56:518–524. Naughton JM. Supply of polyenoic fatty acids to the mammalian brain: the ease of conversion of the short chain essential fatty acids to their longer chain polyunsaturated metabolites in liver, brain, placenta and blood. Intl J Biochem 1980; 13:21–32. Neuringer M, Connor WE, Lin DS, Barstad L, Luck S. Biochemical and functional effects of prenatal and postnatal n-3 fatty acid deficiency on retina and brain in rhesus monkeys. Proceedings of the National. Academy of Science USA 1986; 83:4021–4025. Pawlosky RJ, Sprecher HW, Salem N Jr. High sensitivity negative ion GC–MS method for the detection of desaturated and chain elongated products of deuterium-labeled linoleic and linolenic acids. J Lipid Res 1992; 33:1711–1717.

Chapter 7 / Supply of DHA to the CNS


Pawlosky RJ, Salem N Jr. The metabolism of essential fatty acids in mammals. In: Sinclair A, Gibson R, eds. The 3rd International Conference on Eicosanoids and Essential Fatty Acids. American Oil Chemists’ Society, Champaign, IL, 1993, pp. 26–30. Pawlosky R, Barnes A, Salem N Jr. Essential fatty acid metabolism in the feline: relationship between liver and brain production of long-chain polyunsaturated fatty acids. J Lipid Res 1994; 35(11), 2032–2040. Pawlosky R, Denkins Y, Ward G, Salem N Jr. Retinal and brain accretion of long-chain polyunsaturated fatty acids in developing felines: the effects of corn oil-based maternal diets. Am J Clin Nutr 1997; 65:465–472. Pawlosky R, Salem N Jr. Is dietary arachidonic acid necessary for feline reproduction? J Nutr 1996; 126:1081S–1085S. Pawlosky RJ, Ward G, Salem N Jr. Essential fatty acid uptake and metabolism in the developing rodent brain. Lipids 1996; 31S:S103–S107. Protstein NP, Pennacchiotti GL, Sprecher H, Aveldano MI. Active synthesis of C24:5n-3 fatty acid in retina. Biochem J 1996; 316:859–864. Rivers JPW, Sinclair AJ, Crawford MA. Inability of the cat to desaturate essential fatty acids. Nature 1975; 258:171–173. Rodriguez A, Sarda P, Nessmann C, Boulot P, Poisson J-P, Leger CL, et al. Fatty acid desaturase activities and polyunsaturated fatty acid composition in human liver between the seventeenth and thirty-sixth gestational weeks. Am J Obstetr Gynecol 1998; 179(4):1063–1070. Rodriguez de Turco EB, Gordon WC, Bazan NG. Rapid and selective uptake, metabolism and cellular distribution of docosahexaenoic acid among rod and cone photoreceptor cells in the frog retina. J Neurosci 1991; 111(1):3667–3678. Rodriguez de Turco EB, Parkins N, Ershov AV, Bazan NG. Selective retinal pigment epithelial cell lipid metabolism and remodeling conserves photoreceptor docosahexaenoic acid following phagocytosis. J Neurosci Res 1999; 57:479–486. Salem N Jr, Kim H-Y, Yergey JA. Docosahexaenoic acid: membrane function and metabolism. In: Simopoulos AP, Kifer RR, Martin R, eds. The Health Effects of Polyunsaturates in Seafoods. Academic, New York, 1986, pp. 263–317. Salem N Jr, Omega-3 fatty acids: molecular and biochemical aspects. In: Spiller GA, Scala J, eds. New Protective Roles for Selected Nutrients Alan R. Liss, New York, 1989, pp. 109–228. Salem N Jr, Pawlosky RJ. Health Policy Aspects of Lipid Nutrition and Early Development. In: Galli C, Simopoulos AP, Tremoli E, eds. Fatty Acids and Lipids: Biological Aspects. World Review of Nutrition and Diet Vol. 75. Karger, Basel, 1994, pp. 46–51. Salem N Jr, Pawlosky RJ. Arachidonate and docosahexaenoate biosynthesis in various species and compartments in vivo. In: Galli C, Simopoulos AP, Tremoli E, eds. Fatty Acids and Lipids: Biological Aspects World. Review of Nutrition and Diet Vol. 75. Karger, Basel, 1994, pp. 114–119. Salem N Jr, Wegher B, Mena P, Uauy R. Arachidonic and docosahexaenoic acids are biosynthesized from their 18-carbon precursors in human infants. Proc Natl Acad of Sci USA 1996; 93:49–54. Salem N Jr, Pawlosky R, Wegher B, Hibbeln J. In vivo conversion of linoleic acid to arachidonic acid in human adults. Prostaglandins Leukotrienes Essential Fatty Acids 1999; 60(5–6):407–410. Sauerwald TU, Hachey DL, Jensen CL, Chen H, Anderson RE, Heird WC. Intermediates in endogenous synthesis of C22:6n-3 and C20:4n6 by term and preterm infants. Pediatr Res 1991; 41(2):183–187. Schenck PA, Rakoff H, Emken EA. Delta-8 desaturation in vivo of deuterated eicosatrienoic acid by mouse liver. Lipids 1996; 31(6):593–600. Scott BL, Bazan NG. Membrane docosahexaenoate is supplied to the developing brain by the liver. Proc Natl Acad Sci USA 1989; 86:2903–2907. Sinclair AJ, McLean JG, Monger EA. Metabolism of linoleic acid in the cat. Lipids 1979; 14:932–936. Stinson AM, Wiegand RD, Anderson RE. Recycling of docosahexaenoic acid in rat retinas during n-3 fatty acid deficiency J Lipid Res 1991; 32:2009–2017. Su H-M, Bernardo L, Mirmiran M, Ma X-H, Nathanielsz PW, Brenna JT. Dietary 18:3n3 and 22:6n-3 as sources of 22:6n-3 accretion in neonatal baboon brain and associated organs. Lipids 1999; 34(3):S347–S350. Vermunt SH, Mensink RP, Simonis MM, Hornstra G. Effects of dietary alpha-linolenic acid on the conversion and oxidation of 13C-alpha-linolenic acid. Lipids 2000; 35(2):137–142. Wang N, Anderson RE. Synthesis of docosahexaenoic acid by the retina and retinal pigment epithelium. Biochemistry 1993; 32:13,703–13,709. Weisinger HS, Vingrys AJ, Sinclair AJ. Effect of dietary n-3 deficiency on the electroretinogram in the guinea pig. Ann Nutr Metab 1996; 40(2):91–98. Willis AL. Unanswered questions in EFA and PG research. Prog Lipid Res 1981; 20:839–850.

Chapter 8 / In Vivo Brain Phospholipid Kinetics



Quantifying and Imaging Brain Phospholipid Metabolism In Vivo Using Radiolabeled Long Chain Fatty Acids Stanley I. Rapoport

1. INTRODUCTION Phospholipids are major components of neuronal and glial membranes and participate in membrane remodeling and signal transduction (Axelrod, Burch, & Jelsema, 1988; Fisher & Agranoff, 1987; Porcellati, Goracci, & Arienti, 1983; Stephenson et al., 1994). Many of their functions involve the release of the essential polyunsaturated fatty acids (FAs), arachidonate (20:4n-6) and docosahexaenoate (22:6n-3) in signaling processes. Docosahexaenoate can modulate membrane fluidity and neuronal recovery following injury and participate in signal transduction and synaptic plasticity, whereas arachidonate and its bioactive metabolites (eicosanoids, leukotrienes, and monohydroxyeicosatetraenoic acids) are important second messengers (Axelrod, 1995; Horrocks & Yeo, 1999; Wolfe & Horrocks, 1994). However, in pathological conditions, such as inflammation, ischemia, and trauma, large quantities of FAs are liberated from phospholipids and contribute to cell dysfunction or death (Bazán & Rodriguez de Turco, 1980; Rabin et al., 1997). Abnormal brain phospholipid metabolism also occurs in essential FA deficiency (Bourre et al., 1989; Contreras et al., 1999b), Alzheimer disease (Farooqui, Rapoport, & Horrocks, 1997a; Ginsberg, Rafique, Xuereb, Rapoport, & Gershfeld, 1995; Pettegrew, Moossy, Withers, McKeag, & Panchalingam, 1988), chronic alcohol exposure (Pawlosky & Salem Jr., 1995), and possibly in human depression and bipolar disorder (Hibbeln, 1998; Stoll et al., 1999). For these many reasons, it would be of interest to quantify and image in vivo FA kinetics in brain phospholipids, in animals and in humans. Our laboratory has elaborated a method and model to do this (Rapoport, In press; Rapoport et al., 1997; Robinson et al., 1992).

2. FATTY ACID MODEL 2.1. Experimental Method The FA method involves the intravenous injection or infusion of a radiolabeled albumin-bound FA, then measuring plasma and brain radioactivities. In awake rodents, the From: Fatty Acids: Physiological and Behavioral Functions Edited by: D. Mostofsky, S. Yehuda, and N. Salem Jr. © Humana Press Inc., Totowa, NJ



Part II / Phospholipid and Fatty Acid

brain can be prepared for biochemical analysis or quantitative autoradiography using a [3H]FA or [14C]FA. In humans or higher primates, positron emission tomography (PET) can be used to noninvasively image regional incorporation of [11C]FA (Arai et al., 1995; Chang et al., 1997a; Rapoport, Chang, Connolly, Carson, & Eckelman, 1999; Rapoport et al., 2000; Robinson et al., 1992). To determine regional incorporation coefficients k* (Eq. 3) in awake rats, following tracer injection labeled and cold FA concentrations in arterial plasma are measured until the animal is killed after 10–15 min. After removing the brain, frozen sections are cut on a cryostat and prepared for quantitative autoradiography using radioactive standards. For this, [3H]FAs are preferred to minimize nonvolatile aqueous background radioactivity because of `-oxidation. Values for k* are calculated by dividing net brain radioactivity by integrated plasma radioactivity. To determine FA turnover rates and half-lives in brain phospholipids, the labeled FA is infused intravenously at a programmed rate for 5 or 10 min to establish constant plasma radioactivity. Arterial blood is withdrawn at regular intervals to monitor plasma concentrations of labeled and unlabeled FAs. After infusion, the animal is rapidly anesthetized and its brain is subjected to focused-beam microwave irradiation to stop metabolism. One half is used to quantify acyl-CoA, the other to quantify phospholipids and the distribution of the tracer by established analytical methods. A correction is made for radioactivity in blood (Chang et al., 1997a; Chang, Bell, Purdon, Chikhale, & Grange, 1999; Deutsch, Grange, Rapoport, & Purdon, 1994; Deutsch, Rapoport, & Purdon, 1996; Grange et al., 1995; Rapoport, In press; Washizaki, Smith, Rapoport, & Purdon, 1994).

2.2. Metabolic Pathways Figure 1 illustrates the diffusional and metabolic pathways that can be taken by a FA entering the brain from blood. In blood, the FA may exist as the unesterified species in plasma, esterified within phospholipids, cholesterol, and triglycerides of lipoproteins, or covalently bound within erythrocytes and platelets (Bazán, 1990; Staufenbiel, 1988). Unesterified FAs in plasma with chain length ) 22 carbons are largely bound to circulating albumin (> 99% binding), whereas longer-chain unesterified FAs are preferentially bound to circulating lipoproteins (Shafrir, Gatt, & Khasis, 1965; Wosilait & SolerArgilaga, 1975). The demonstration of lipoprotein receptors on the luminal surface of the cerebrovascular endothelium suggested that they mediate brain uptake of FAs esterified within lipoproteins (Dehouck et al., 1997; Méresse, Delbart, Fruchart, & Cecchelli, 1989). However, in awake adult rats, such receptor-mediated entry is unimportant and its flux term, J2 (see Fig. 2), can be neglected. Thus, circulating FAs cross the blood-brain barrier essentially in only the unbound unesterified form, after being hydrolyzed from lipoproteins by lipoprotein lipase within blood or at the cerebral capillary bed (Fig. 1) (Brecher & Kuan, 1979; Purdon, Arai, & Rapoport, 1997; Spector, In press). The process involves simple diffusion and not additionally, as in the heart, facilitated diffusion by a translocase (a translocase is not found in brain) (Glatz, In press; Luiken et al., 1999). Incorporation from plasma into brain phospholipids is proportional to the FA plasma concentration and is determined by brain metabolic demand. This is because the Km for acyl-CoA synthetase (the enzyme which converts the FA to acyl-CoA) is 180 µmol, twice the total brain unesterified FA concentration (Deutsch, Rapoport, & Purdon, 1997; Rabin et al., 1997; Rapoport & Spector, Submitted; Sigiura et al., 1995).

Chapter 8 / In Vivo Brain Phospholipid Kinetics


Fig. 1. Brain metabolic pathways for fatty acids. Thick arrows indicate major pathways followed rapidly by a FA after its entry into brain from plasma, and the enzymes involved in the pathways. Thin arrows indicate alternative pathways that are followed to a lesser extent within the 5- to 20-min experimental period of the FA method, the major one of concern from the acyl-CoA pool involving `-oxidation within mitochondria (see text).

About 5% of unesterified FA is extracted as blood passes through the rat brain, and extraction is independent of blood flow (Chang et al., 1997a; Pardridge & Mietus, 1980; Robinson et al., 1992; Yamazaki, DeGeorge, Bell, & Rapoport, 1994). After entering the brain, the FA rapidly and preferentially follows the pathway indicated by the thick arrows in Fig. 1. From the FA pool, it is activated to acyl-CoA (FA-CoA) by an acyl-CoA synthetase (sem*nkovich, 1997; Watkins, 1997), then enzymatically incorporated into stable lipids (predominantly phospholipids) by an acyltransferase (Yamash*ta, Sugiura, & Waku, 1997). At steady state, the rate of incorporation into phospholipids equals the rate of release back into the unesterified FA pool via the brain free FA pool, to complete a “cycle.” Release is catalyzed by a phospholipase A1 (PLA1) acting at the stereospecifically numbered (sn)-1 site for saturated FAs, or by a phospholipase A2 (PLA2) acting at the sn-2 site for polyunsaturated FAs (Dennis, 1994; Pete, Ross, & Exton, 1994). Thin arrows in Figure 1 indicate alternative pathways that may be taken by a FA once within the brain. Those that are quantitatively significant within the few minutes of our pulse-labeling studies are the conversion of polyunsaturated FAs to bioactive metabolites (see Sec. I), and `-oxidation within mitochondria. The extent of `-oxidation, thus the rate of formation of nonvolatile aqueous metabolites, depends on the FA and how it is labeled. The saturated [1-14C]palmitate tracer is not ideally suited to estimate brain incorporation of palmitate into phospholipids with quantitative autoradiography, because about half of the tracer is converted via `-oxidation to background aqueous nonvolatile labeled metabolites, mainly glutamate and aspartate (Miller, Gnaedinger, & Rapoport, 1987; Noronha, Larson, & Rapoport, 1989). Using [9,10-3H]palmitate can overcome this limi-


Part II / Phospholipid and Fatty Acid

Fig. 2. Model for relation between net FA flux from final precursor pool (acyl-CoA) into brain lipids (JFA) and the net flux rates from plasma unesterified FA (J1) and from plasma esterified FA (J2) into acyl-CoA. J2 is negligible in the adult brain.

tation, as 75% of the tritium is converted to [3H]H2O during `-oxidation (Greville & Tubbs, 1968), and can be removed during drying. For PET scanning with [1-11C]palmitate, `-oxidation of carbon-labeled palmitate can be reduced by pretreatment with an inhibitor of its entry into mitochondria, methylpalmoxirate (methyl-2-tetradecylglycidate) (Chang et al., 1998; Chang, Wakabayashi, & Bell, 1994; Rapoport, 1999; Rapoport et al., 1999; Tutwiler, Ho, & Mohrbacher, 1981). Unlike saturated FAs, carbon-labeled polyunsaturated arachidonate and docosahexaenoate can be used with autoradiography or PET scanning without the inhibitor, because they are minimally oxidized (Osmundsen, Cervenka, & Bremer, 1982); only 15% of their label is found in the brain nonvolatile aqueous compartment 20 min after an intravenous injection. [3H]Arachidonate or [3H]docosahexaenoate produce only 10% nonvolatile aqueous background activity.

2.3. Brain Compartments The complex representation of Fig. 1 can be simplified to Fig. 2, which identifies three compartments that have to be assessed experimentally to apply the FA model. These compartments are (1) plasma unesterified FA, (2) the precursor brain FA-CoA pool, and (3) the “stable” brain phospholipid compartment (Rapoport et al., 1997; Robinson et al., 1992). Fluxes between them—J1, J2, J3, JFA—are defined in the legend to Fig. 2. The simplified Fig. 2 can be used because: (1) the half-life of the FA tracer in plasma is less than 1 min, (2) FA uptake into brain from blood is independent of cerebral blood flow over a wide range of unlabeled FA plasma concentrations (Chang et al., 1997a; Yamazaki et al., 1994), (3) labeled FA in plasma rapidly equilibrates (1 min or less) with label in FA-CoA, the precursor pool for FA incorporation into brain phospholipids, and (4) the FA tracer in brain is rapidly incorporated into brain phospholipids (80–90% within 1 min) (Rapoport et al., 1997; Robinson et al., 1992).

Chapter 8 / In Vivo Brain Phospholipid Kinetics


Rapid entry of a FA from plasma into the brain FA-CoA pool allows increased neuronal demand for the FA to be easily met by the large reservoir of unesterified FA in plasma. One to two minutes after a step elevation in plasma [9,10-3H]palmitate or labeled arachidonate, specific activity of the respective brain FA-CoA pool has reached a steady state (Grange et al., 1995; Washizaki et al., 1994). At this time, the ratio of FA-CoA specific activity to plasma FA specific activity (h in Eq. 4) is 0.02–0.04, attesting to marked dilution of plasma-derived FA-CoA by FA released from brain phospholipids (Fig. 2).

2.4. Operational Equations We have derived and validated operational equations to quantify FA fluxes from plasma into the brain FA-CoA pool and from the FA-CoA pool into individual brain phospholipids, and turnover rates and half-lives within these phospholipids (Rapoport et al., 1997; Robinson et al., 1992). The incorporation rate of a FA radiotracer from plasma into a stable brain lipid compartment i is given by, dt/dc*br,i = k*ic*pl


where k*i (mL/s/g, or s–1) is the incorporation coefficient, c*pl is the plasma concentration of labeled unesterified FA, and c*br,i is the brain concentration of label in i. Integration of Eq. 1 to time T of sampling gives k*i , T

k*i = c*br,i /





For whole brain or for a brain region in which analysis is by quantitative autoradiography or PET, we have an overall k* equal to T

k* =

⌺i k* = c* (T ) / 兰c* i





where c*br(T ) equals the brain radioactivity at time T. Dilution factor h represents the extent to which brain FA-CoA is derived from unesterified FA in plasma (J1), compared with esterified plasma FA (J2) and FA produced by recycling plus de novo synthesis within brain (J3) [de novo synthesis via phosphatidic acid is less than 1% of the recycling contribution (Murphy, 1998)]. As J2 2.5 log cd.s/m2), although isolating the R1 is not easily achieved. One of the reasons that the ERP is not studied often is that it is hard to isolate reliably. ERP isolation requires fast sampling, bright-light stimuli, nonmetallic electrodes, and the removal of artifactual responses (melanin response). As the ERP commences immediately after light stimulation (no delay), it requires high sampling rates for proper isolation (>20 kHz; Fig. 5). A small R2 can sometimes be found superimposed on the leading edge of the a-wave using slower sampling rates (2–4 kHz), but these do not give reliable nor repeatable signals because of undersampling. If metallic electrodes are used they will give rise to a photovoltaic artifact (electron release from metal molecules by light photons). This can interfere with the leading edge of the ERP (R1) and can only be overcome by using nonmetallic electrodes (Dawson & Galloway, 1991). A metallic electrode with high sampling rates can be used to isolate the R2 (Fig. 5), as the photovoltaic artifact has a very fast time-course. In many cases, this approach may be adequate, but it needs to be mentioned that the ratio R1/R2 is known to provide useful additional information (Dawson & Galloway, 1991). The problem is that the extraction of the R1 is further complicated by the temporal characteristics of the light source and the melanin response. Any pigment molecule can interact with light to generate a potential, as does melanin within the eye. This produces a fast electronegative potential ( teff


In Eq. 1, rmax (µamp) is the magnitude of the saturating (maximal) photocurrent. The constant, teff (s), is the delay that arises from a number of small delays during the transduction cascade, but mostly those inherent to the recording system. The amplification, A (per isomerization per second), is the product of numerous amplifications occurring during phototransduction. Lamb and Pugh (1992) demonstrated that the model accurately described the current recorded from single rods in response to a wide range of stimulus intensities. They also noted that as intensity increased, the values of teff and A decreased (Lamb & Pugh Jr, 1992), a finding confirmed by others in single-cell and massed retinal recordings (Breton et al., 1995; Cobbs & Pugh, 1987; Lyubarski & Pugh, 1996). This is believed to arise from the saturation of PDE in the hydrolysis of cGMP (Breton et al., 1994; Lamb & Pugh Jr, 1992). A modification of the delayed Gaussian model of Lamb and Pugh (1992) (Eq. 1), has been applied to the leading edge of the ERG (i.e., the fast PIII) (Hood & Birch, 1990a) as given by PIII(i,t)  [1 – exp {–iS (t – td)2}] Rmax


In Eq. 2, the amplitude of the PIII (µV) represents the sum of the individual rod responses and is a function of both flash intensity, i (cd.s/m2) and time, t (s) after the onset of the flash. S (m2.cd–1.s–3) is a sensitivity parameter that scales i, Rmax is the saturated PIII response (µV) and td is the delay (s). This model provides good fits to the raw waveform when fitted as an ensemble (i.e., to a group of intensities; Fig. 7) at low to medium light levels (Hood & Birch, 1990b) or when the responses to individual intensities are fitted


Part III / DHA and CNS Development

Fig. 7. Representative ensemble fitting (lines) to raw ERG data (symbols) over three exposures (shown to right) for one n-3-sufficient (unfilled symbols, thin line) and one n-3-deficient (filled symbols, thick line) rat. These data are typical of the respective diet groups and show a reduced R and lowered sensitivity (S) in deficient animals. In this figure, the sufficient animal shows a max marked R2 response in the first 3 ms of the waveform (compare with Fig. 5), which frustrates isolation of the fast PIII.

separately (Cideciyan & Jacobson, 1996). We usually maximize the likelihood of our fit by minimizing the root-mean-square error term over an ensemble of intensities using the solver module of a Microsoft Excel spreadsheet [Fig. 7 (Bui & Vingrys, 2000)]. The maximum intensity for an ensemble fit must be chosen to show response saturation (maximal a-wave slope). More complex formulas may be applied to fit better rod or cone responses by allowing for membrane capacitance (Cideciyan & Jacobson, 1996). How-

Chapter 12 / Omega-3 Fatty Acids in Vision


ever, we prefer the less complicated model for rod responses because of the ill-defined aspects of membrane capacitance. When performing fits, only S and Rmax should be floated because the other parameters (delay, capacitance, etc.) are usually fixed in physiological terms. If these variables are floated, then their interdependency may yield erroneous and physiologically irrelevant outcomes that would require single-cell studies for confirmation. In our laboratory, we usually derive td (and capacitance) by floating these variables in our normal controls and then refit all experimental data using the mean of this value. Sometimes, at high intensities, we find that an R2 can complicate fast P3 extraction and we allow for this by ignoring the initial 3-ms of negativity (see Fig. 7). Changes in the parameters Rmax and S can be interpreted as distinct functional alterations to either the photoreceptor or the phototransduction cascade (Birch et al., 1995; Holopigian et al., 1997; Hood & Birch, 1994a; Vingrys et al., 1998; Weisinger et al., 1999). Rmax describes the saturated rod response and reflects the change in current flow about the outer segments following light stimulation. This saturated response has been shown to be related to the total number of cationic channels and reflects outer segment area or length (Baylor et al., 1979; Breton et al., 1994). As such, Rmax is reduced when outer segment morphology is altered or when there is a reduction in the number of receptors, as may occur in some disease states (Hood & Birch, 1994b). However, in developing retina, Rmax has also been shown to be related to the level of active rhodopsin available in the rod outer segments (Fulton, Hansen and Findl, 1995b) implying that the concentration of unbleached rhodopsin may be an important factor in determining the maximal rod response. The sensitivity parameter, S, can be considered as reflecting a change in flash energy. Structural factors such as rod-packing density, receptor alignment, and pigment content will all affect S. Likewise, the efficiency of the rhodopsin– transducin–phosphodiesterase cascade will affect sensitivity and will manifest as an altered slope (S) at a fixed Rmax. Figure 7 shows that dietary DHA deprivation, in rats, acts to reduce both Rmax and sensitivity of the fast PIII, which is consistent with our previous reports in guinea pigs (Vingrys et al., 1998; Weisinger et al., 1999). The Rmax loss is not great (25–50%), and in guinea pigs, it shows a dose dependency that becomes manifest when tissue DHA levels drop below 75% of their normal levels (

Fatty Acids - PDF Free Download (2024)
Top Articles
Latest Posts
Article information

Author: Foster Heidenreich CPA

Last Updated:

Views: 5974

Rating: 4.6 / 5 (76 voted)

Reviews: 91% of readers found this page helpful

Author information

Name: Foster Heidenreich CPA

Birthday: 1995-01-14

Address: 55021 Usha Garden, North Larisa, DE 19209

Phone: +6812240846623

Job: Corporate Healthcare Strategist

Hobby: Singing, Listening to music, Rafting, LARPing, Gardening, Quilting, Rappelling

Introduction: My name is Foster Heidenreich CPA, I am a delightful, quaint, glorious, quaint, faithful, enchanting, fine person who loves writing and wants to share my knowledge and understanding with you.