L; /0/ '/ / ~17 ABSTRACT THE INFLUENCE'OF DIETARY POLYDNSATURATED AND SATURATED FATTY ACIDS 0N HEPATIC AND ADIPOSE TISSUE FATTY ACID SYNTHESIS IN THE MEAL-FED RAT By Steven Donald Clarke The primary objective of this research was to evaluate the contention that low dietary levels of polyunsaturated fatty acids spedifically depress the rates of liver and adipose tissue fatty acid synthesis and the activities of glucose-6-phosphate dehydro- genase (ECl.l.l.49), malic enzyme (ECl.l.l.40), citrate cleavage enzyme (EC4.l.3.8), acetyl-CoA carboxylase (E06.4.l.2) and fatty acid synthetase in rats trained to consume their daily food during a three hour period each day. Pure methyl esters of palmitate (C16:0). stearate (C18:0). oleate (C18zl)’ linoleate (618:2) and linolenate (C18:3) were employed as supplements to either a fat- free, high carbohydrate diet or a low fat, essential fatty acid adequate diet. Constant intakes of basal diet were generally main- tained among all treatments. 618:1’ C18:2 and (:18:3 were determined to have apparent digestibilities of 88, 87 and 89%, respectively, and these were supplemented to the basal diet as 3% of the amount of food allotted daily to each rat. However the poor digestibilities Steven Donald Clarke of C16:0 and 018:0 (40 and 35%) required that the level of supplemen- tation be raised to 7 and 8%, respectively. The first part of this investigation measured the response of hepatic and adipose fatty acid synthesis and the activities of lipo- genic enzymes to the supplementation for 7-10 days of various fatty acids to either fat-free, or low fat, high carbohydrate diets. Die- tary Cl8:2 and/or Cl8z3 resulted in a consistent and significant decline in the activities of hepatic glucose-6-phosphate dehydro- genase, malic enzyme, citrate cleavage enzyme, acetyl-CoA carboxyl- ase and fatty acid synthetase. -In contrast the additon of C16:0’ C18:0, or ClB:l to the basal diets had little suppressive action on the activities of these enzymes. Dietary polyunsaturated fatty acids were also associated with a 25% decline in the activities of hepatic pyruvate kinase (EC2.7.l.40) and glucokinase (EC2.7.l.2). Unlike ' the liver. adipose tissue glucose-6-phosphate dehydrogenase, malic enzyme and fatty acid synthetase activities remained unaffected by ’ the addition of either saturated or unsaturated fatty acids to the diets. The in vivo incorporation of 3H20 into liver fatty acids was significantly reduced by nearly 40% among all experiments involving dietary 018:2 and C18:3. Like the enzyme activities, adipose tissue fatty acid synthesis was not depressed by dietary fatty acids of any type. These effects of C18:2 and C18:3 on hepatic lipogenesis after 7-10 days were not due to differences among treatments in carbohy- drate intake or variations in absorbed fat, and were independent of essential fatty acid status. Steven Donald Clarke The second phase of research was directed at quantitating the concentration and composition of plasma free fatty acids, the concen- tration of hepatic long chain acyl-CDA esters and the hepatic ratio of lactate and pyruvate in rats fed the fat-free diet for five to seven days with and without 3% Cl8:2 or C18:3. The liver tissue sam- ples for determination of metabolites were collected by rapid freez- ing in situ using clamps precooled in liquid nitrogen. The concen- tration of these known effectors of hepatic fatty acid synthesis were not altered by dietary C18z2 or Cl8z3' The composition of plasma free fatty acids showed a fourfold rise in unesterified linoleate af- ter Cl8:2 feeding. If long chain acyl-CoA esters or free fatty acids play a role in regulating hepatic fatty acid synthesis, the action must be more dependent on the composition than the total concentra- tion of acyl esters. Because an animal very likely has reached a new steady-state after supplementing C18z2 or C18:3 for 7-l0 days, accurate detection of the initial point of inhibition exerted by Cl8:2 or C18:3 in rat liver may be difficult. Therefore the characterization of changes in activities of certain key hepatic lipogenic and glycolytic en— zymes as well as quantitative comparison to the rate of in vivo fat— ty acid synthesis during the attainment of a new steady-state was of particular importance. The irI vivo rate of Cz-unit incorporation into hepatic fatty acids, calculated from 3H20 incorporation, revealed that six to eight Steven Donald Clarke meals of 3% (:18:2 supplementation led to maximal depression in fatty. acid synthesis. Furthermore within a treatment group the in vitro activity of fatty acid synthetase and acetyl-CoA carboxylase was nearly identical to the in vivo rate of Cz-unit incorporation into hepatic fatty acids. Time sequence studies demonstrated that a min- imum of three meals containing C18:2 or C18z3 or about a 48 hour time span was essential before a significant decline in the rate of hepat- ic fatty acid synthesis was detectable. The consumption of four C18:2 containing meals resulted in a degree of depression in fatty acid synthesis that was very similar to that found after six meals of cl8:2' However the activity of fatty acid synthetase in vitro was not significantly reduced by three or four meals of Cl8:2 addi- tion, and it was well above the in vivo rate of Cz-unit incorpora- tion into fatty acids. Glucokinase activity also remained high during the transitional period and generally its activity was not reduced by C18:2. Low dietary levels of polyunsaturated fatty acids very ef- fectively reduce the rate of rat liver fatty acid synthesis as well as activities of many lipogenic enzymes. The initial point of in- hibition of hepatic fatty acid synthesis exerted by polyunsaturated fats would appear to be after glucose phosphorylation and prior to malonyl-CoA utilization. OELWHLVSNOAWOd THE INFLUENCE OF DIETARY POLUNSATURATED AND SATURATED-FATTY ACIDS ON HEPATIC AND ADIPOSE TISSUE FATTY ACID SYNTHESIS IN THE MEAL—FED RAT BY STEVEN DONALD CLARKE A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements~ for the degree of DOCTOR OF PHILOSOPHY Department of Food Science and HUman Nutrition 1976 ACKNOWLEDGEMENTS The author wishes to thank Dr. Dale Romsos, Associate Professor, for his guidance and support in the development of the research topic. In addition the suggestions and advice of Dr. G. A. Leveille, Chairman of the Department of Food Science, Dr. M. R. Bennink, Assistant Professor, and Dr. N. G. Bergen, Associate Pro- fessor, during the preparation of this manuscript, were greatly ap- preciated. My wife, Karen, and daughter, Kimberly, deserve special thanks for their unending patience, help and understanding during my graduate program. ii TABLE OF CONTENTS INTRODUCTION ......................... PART I REVIEW OF LITERATURE DE Novo FATTY_ACID SYNTHESIS . . . . . ............ LONG-TERM AND SHORT-TERM REGULATION OF RATES OF DE NOVO FATTY ACID SYNTHESIS. . ....... . ......... . . . REGULATION VIA FLUCTUATION IN ENZYME ACTIVITY ......... Glucokinase and hexokinase ..... ‘. . . ..- . . . . . . . Phosphofructokinase . . .................. Pyruvate kinase ...................... Pyruvate dehydrogenase. ............. . . . . . Acetyl-CoA carboxylase .......... . ........ Fatty acid synthetase . . . . ._ .............. Citrate cleavage enzyme . . ............... Glucose-G- -phosphate dehydrogenase and malic enzyme ..... REGULATORY ASPECTS OF LONG CHAIN ACYL-CoA ESTERS . . ..... THE ROLE OF REDOX STATE IN CONTROL OF FATTY ACID SYNTHESIS . . RELATIONSHIP AMONG SPECIES BETWEEN NUTRITIONAL STATE AND FATTY ACID SYNTHESIS. . . . .......... . .......... - Rats. . .......................... Mice . ..... . . ................ Page vii 39 39' 45 TABLE OF CONTENTS (cont'd.) Page Swine .............. ~ ............. 47 Chicken ........... . . . . . .......... 48 DE NOVO FAT SYNTHESIS AND TYPE OF DIETARY FAT. . . . . . . . . 50 PARAMETERS RELEVANT TO FAT METABOLISM AFFECTED BY DIETARY POLYUNSATURATED FATS ..................... 59 Hepatic fatty acid composition. . . . . . . . . . . . . . . 59 Desaturase activity .................... 62 Membranes . . . ............ , ........ - 65 Albumin binding of fatty acids. . ............. 68 Turnover of polyunsaturated fatty acids from adipose. . . . 70 Prostaglandins ..................... -. . 71 PART II DIFFERENTIAL EFFECTS OF DIETARY METHYL ESTERS 0F LONG CHAIN SATURATED AND POLYUNSATURATED‘ FATTY ACIDS ON RAT LIVER AND ADIPOSE TISSUE LIPOGENESIS INTRODUCTION . . .. ...................... 77 METHODS. . . ........ ~ ................ —. 80 General animal handling . . ................ 8O Fecal lipid extraction ........ - ..... . . . . . . 82 Enzyme assays ...................... . 82 In vitro fatty acid synthesis ....... . ....... 83 In vivo fatty acid synthesis ................ 83 Statistics. . . . . . ................ . 84 RESULTS. . .............. , ............ 84 Experiment 1.! ............ ' ........... 84 Experiment 2 ........... . ......... . . . 85 Experiment 3. ......... . . . ........... 88 Experiment 4. . . . . . . . . . .......... . . . . 91 DISCUSSION .......................... 93 iv TABLE OF CONTENTS (contid.) PART III SPECIFIC INHIBITION OF HEPATIC FATTY ACID SYNTHESIS EXERTED BY DIETARY LINOLEATE AND LINOLENATE IN ESSENTIAL FATTY ACID ADEQUATE RATS INTRODUCTION ......................... METHODS. . . . . . . . . . . . .~ ............... RESULTS AND DISCUSSION ................ . . . . Growth parameters . . . . . ................ Liver . . . . . . ........ _ ............. Adipose . .................... . . . . One meal 518:3 ......... . . . . . . . . . . . . . PART IV INFLUENCE OF DIETARY FATTY ACIDS ON LIVER AND ADIPOSE TISSUE LIPOGENESIS, AND ON LIVER METABOLITES IN MEAL-FED RATS‘ INTRODUCTION . . .................... . . METHODS. . . ........ . ........ ‘. . . .,. . . General animal handling . . . ............... Experiment I. . . ................ . . . . Experiment 2. . ., ..................... Experiment 3 ..... _ ................... Experiment-4. . .,. . . ._ ..... . ........ . . . RESULTSo‘oo'oooo'ooooocoo.ooooooooooo C13.0 vs. C13. .2 effects on lipogenesis. . . . ....... Metabolites . ........... . . . . ...... Hepatic glycolytic and lipogenic enzyme activities ..... DISCUSSION . . ........... . ............ Page 114 116 116 116 119 120 121 122 122 124 128 128 TABLE OF CONTENTS (cont'd.) Page PART V CHANGE IN HEPATIC FATTY ACID SYNTHESIS IN MEAL-FED RATS DURING SEVEN DAYS OF FEEDING POLYUNSATURATED FATTY ACIDS INTRODUCTION ..................... ,. . . . l37 METHODS ......... . .......... - ........ l39 General animal handling .................. ’ l39 In vivo fatty acid synthesis ....... ,. . . . ..... l39 Enzyme assays ................ - ....... I42 Experiments l-3 ..................... . l42 Experiment 4 ........ - ....... . ........ l43 Statistics ......................... 144 RESULTS ............ - ........ - ........ I44 DISCUSSION .......................... l54 PART VI SUMMARY AND CONCLUSIONS SUMMATIONS ............. - ............. l58 LITERATURE CITED ..... ~ ............... . . . I67 vi LIST OF TABLES Table PART II DIFFERENTIAL EFFECTS OF DIETARY METHYL ESTERS OF LONG CHAIN SATURATED AND POLYUNSATURATED FATTY ACIDS ON RAT LIVER AND ADIPOSE TISSUE LIPOGENESIS 1. Fat free basal diet composition ............ 2. Methylstearate vs. linolenate - Influence on lipogene» sis and lipogenic enzymes (Experiment 1) ....... 3. Methyl palmitate vs. linoleate - Influence on lipo- genesis and lipogenic enzymes (Experiment 2) ..... 4. Methyl oleate vs. linoleate - Influence on li o- genesis and lipogenic enzymes (Experiment 31. . .,. . 5. Influence of methyl palmitate, linoleate and linolenate on fatty acid synthesis and hepatic lipogenic enzymes (Experiment 4). . . . .,. .- ............. PART III SPECIFIC INHIBITION OF HEPATIC FATTY ACID SYNTHESIS EXERTED BY DIETARY LINOLEATE AND LINOLENATE IN ESSENTIAL FATTY. ACID ADEQUATE RATS l. Basal diet ingredients .......... -. . . . . . . 2. Efféct of c1$:Os Cigoi, or C13;2 supplementation to essential atty acid adequate diet (2.5% safflower oil) on rat liver lipogenesis . . . . ... . ..... vii Page 81 86 87 89- 92 102 105 LIST OF TABLES (cont'd.) Table* Page 3. Effect of C13:], C18.2 and C 3:3 supplements on liver lipogenesis in rats ed an essential fatty . aeid adequate diet (1% safflower oil) ........ l06 4. Influence of dietary C15;o, C13;1. Gig-2 or C18;3 on adipose tissue lipogenesis In essential fatty acid adequate rats . .-. . . .- ........ . . . . . . » 110 5. Effect of one meal containing C 8-3 on hepatic and adipose lipogenesis in essential fatty acid adequate rats. ._. ... . ... . . .-. ......... . . . . ll3 PART IV INFLUENCE OF DIETARY FATTY ACIDS 0N LIVER AND ADIPOSE TISSUE LIPOGENESIS, AND ON LIVER METABOLITES INT MEAL-FED RATS l. Fat-free basal diet composition . . . . . . . _ ..... ll7- 2.‘ Fatty acid synthesis in rats fed a fat-free diet.plus, C18:0 or 3% C18:2 (Experiment 1). . .- ..... .7. . 123 3. Liver long chain acyl-CoA concentration and Plasma FFA composition in rats fed a fat-free diet plus 3% Cigzz or C13:3 (Experiment 2) . . .~. . . .~. . . . . 126 4. Estimate of liver cytosolic redox in rats fed a fat- free diet plus 3% C18;2 or C18;3 (Experiment 3) . . . l27 5. Changes in rat liver glycolytic and lipogenic enzyme activities after seven days of C18;2 supplementa-~ 1 tion (Experiment 4) . ., ............. -. . 129 viii LIST OF TABLES (cont'd.) Table PART V CHANGE IN HEPATIC FATTY ACID SYNTHESIS IN MEAL-FED RATS DURING SEVEN DAYS OF FEEDING POLYUNSATURATED FATTY.ACIDS . .Fat-free basal diet composition ............ Effect of time after meal and dietary fat on rate of passage and fatty acid digestibility (Experiment 1). Effect of time and dietary fat on liver fatty acid synthesis after one meal containing fatty acids (Experiment I). . .,. . . .E. . ...... . . . . . Effect of two meals of fat-free diet containing stea- rate, linoleate or linolenate on rat liver fatty acid synthesis (Experiment 2) ...... . ..... Effect of three meals of fat- free diet containing stearate, linoleate, or linolenate on rat liver fatty acid synthesis (Experiment 2) . . . .E. . . . . . . . Change in liver fatty acid synthesis and associated enzyme activities following 3% linoleate addition to fat- free diet (Experiment 3) ........... Influence of one and two meals of fat-free diet on rat liver fatty acid synthesis after Six and seven meals of linoleate supplementation (Experiment 4) ..... ix Page 140 145 147 148 149 151 152 AcCBX ADP AMP ATP BHT C16:0 c16:1 C18:0 C18:1 c18:2 , C18:3 c20:3 c20:4 CCE C0 CoA CV FAS FF F6P LIST OF ABBREVIATIONS acetyl-COA carboxylase adenine diphosphate adenine monophosphate adenine triphosphate butylated hydroxytoluene palmitate palmitoleate stearate oleate linoleate linolenate eicosatrienoate arachidonate cyclic adenine monophosphate citrate cleavage enzyme carbon dioxide coenzyme A coefficient of variation fatty acid synthetase fat-free fructose-6-ph05phate LIST OF ABBREVIATIONS (cont'd.) FDP GEPD mg M9 min NAD NADH NADP NADPH PDH PFK PG PK ug uM fructose-l,6-diphosphate glucoggEE-phosphate dehydrogenase bicarbonate tritiated water intraperitoneal kilogram malic enzyme milligram magnesium minute millimolar nicotinamide adenine dinucleotide nicotinamide adenine dinucleotide, reduced nicotinamide adenine dinucleotide phosphate nicotinamide adenine dinucleotide phosphate, reduced pyruvate dehydrogenase phOSphofructokinase prostaglandins pyruvate kinase micrograms micromolar xi INTRODUCTION An elevation in blood triglycerides and cholesterol in humans has received wide acceptance as a warning Sign for cardiovascular prob- lems. One approach to controlling hypertriglyceridemia has been the dietary manipulation of reducing the proportion of saturated fat (e.g. butter, beef tallow, lard) and increasing the amount of polyunsaturated fat (e.g. corn oil, safflower oil) (l). The validity of this approach has not been universally accepted and the data are not all supportive (2-5). For example an isolcaloric exchange of safflower oil or corn oil for butter fat in diets of human subjects resulted in a signifi- cant drop in plasma cholesterol and triglycerides (3). On the other hand, subjects that ate a low fat diet displayed no reduction in blood low density lipoproteins when fed additional ethyl linoleate or stea- rate (6). Attempts have been made at attributing a mechanism of action to polyunsaturated fats by utilizing rats and mice as experimental models (7-10). Initial observations with high fat diets indicated polyunsaturated fats suppressed liver and adipose tissue fatty acid synthesis with greater efficacy than did saturated fats. Such an ef- fect in man, whose major site of de novo fatty acid synthesis is the liver, would have significant implications. A reduction in the amount of de novo fat synthesis in liver would potentially lead to less transport of triglyceride via blood to peripheral tissues and hence a lowered blood triglyceride concentration. Because of high dietary fat levels and because of variations in fatty acid composition of dietary fats, earlier studies were unable to explain the mechanism by which polyunsaturated fats affected rates of lipogenesis. One method adopted to avoid these difficulties was to supplement a fat-free diet of mice and rats with low levels of pure esters of individual fatty acids (7-l0). However specific differences in their action on liver and adipose lipogenesis attributed to indi- vidual fatty acids were overshadowed by the following oversights: a) variation among experiments and treatments in the type and amount of carbohydrate eaten by animals, b) differences among methyl esters in digestibility, c) the assumption that lipogenic enzyme activities re- flected rates of fatty acid synthesis, and d) failure to adequately examine the influence of dietary fatty acids on adipose tissue lipo- genesis. Therefore the primary objective of this research was to re- examine the contention that low levels of dietary polyunsaturated fat- ty acids specifically inhibit liver and adipose tissue fatty acid syn- thesis and associated enzymes in the meal-eating rat. Particular at- tention was given to possible differences in digestibility among the fatty acid methyl esters investigated and attention was also directed at correlating in vivo rates of fatty acid synthesis to in vitro lipo- genic enzyme activities. The second phase of experimentation inves- tigated parameters which could potentially explain the inhibition of fatty acid synthesis by dietary fatty acids, Specifically linoleate and linolenate. PART 1 REVIEW OF LITERATURE DE NOVO FATTY ACID SYNTHESIS De novo fatty acid synthesis involves a series of cytosolic reactions which unite eight acetate units to form the long chain fatty acid palmitate. The following equations depict the reactions: (l) citrate + ATP + COA + acetyl-COA + oxaloacetate + ADP + Pi (2) HOD; + ATP + acetyl-CoA : malonyl-COA + ADP + Pi (3) acetyl-COA + ACP - SH : acetyl-S-ACP + CoA (4) acetyl-S-ACP + 7 malonyl—COA + l4NADPH + 14 H+ . palmitoyl-COA + 7 C02 + 7 H20 + l4NADP The following discussion pertains to nonruminant animals. Cit- rate conversion to acetyl-COA plus oxaloacetate is catalyzed by citrate cleavage enzyme. Carboxylation of acetyl-COA is carried out by the bi- otin containing enzyme, acetyl-COA carboxylase. ACP refers to acyl carrier protein which is a fundamental constituent of fatty acid syn- thetase enzyme complex (ll). Fatty acid synthetase is the enzyme complex responsible for joining carbon-carbon bonds of acetate units originating from malonyl-COA, and in so doing oxidizes NADPH. The liberation of C02 in these reactions ensures irreversibility (ll, l2). NADPH is essential for fatty acid synthesis and is generated in most animals by flow of glucose-6-phosphate through the enzymes glucose-6-phosphate dehydrogenase and 6-phosphogluconate dehydrogenase of the hexose monophosphate Shunt. In addition to hexose ShUnt pro- duction of NADPH, reducing equivalents can result from oxaloacetate in cytosol vja reduction by NADH to malate and subsequent malate decar- boxylation to pyruvate catalyzed by malic enzyme with the generation of NADPH. In chickens this is the primary source of NADPH (ll, 13). Citrate originates from the mitochondria via condensation of mitochondrial oxaloacetate and acetyl-COA. Citrate efflux from the mitochondria is under carrier mediated control and could be a key point in regulating rates of fatty acid synthesis. In de novo fatty acid synthesis mitochondrial acetyl-COA primarily is derived from pyruvate decarboxylation through the regulatory enzyme complex, py- ruvate dehydrogenase (14). The dietary precursor of pyruvate will depend on the type of .carbohydrate being consumed by an individual or animal. The immedi- ate control step of pyruvate production is at the level of the allo- steric enzyme pyruvate kinase which catalyzes dephosphorylation of phosphoenolpyruvate to pyruvate (15). If the dietary carbohydrate is glucose then two steps in glycolysis (a) glucose phosphorylation by glucokinase and (b) fructose 1, 6-diphosphate synthesis by phospho- fructokinase are potential points for regulating glucose degradation, glycolytic flow, and hence pyruvate production (15). Fructose and glycerol are sources of dietary carbohydrate which by-pass these two regulatory steps and result in different effects on fatty acid syn- thesis in liver and adipose tissue (15, 16). Clearly several locations exist for modulation of glycolytic flux and de novo fatty acid synthesis. Control of these points is under hormonal and metabolite mediation and likely involves a combination of both factors. In the following sections several steps' of glycolysis and lipogenesis will be reviewed with particular refer- ence to regulatory significance. LONG-TERM AND SHORT-TERM REGULATION OF RATES OF DE NOVO FATTY ACID SYNTHESIS' Control of fatty acid synthesis can be via long-term regu- lation. Long-term control would be effectors or regulation which requires many hours or days for effectiveness, whereas short-term control is a response by the system in seconds or minutes to both positive and negative modulators. Long-term control of fatty acid synthesis is best charac- terized by the time required to observe alterations in tissue con-~ tent of catalytic enzyme machinery. Generally lipogenic enzymeS' are considered to be in ample concentrations and increase in amount in response to greater substrate flux (l7, 18). However key enzymes, such as glucokinase, phOSphofructokinase, acetyl-COA carboxylase and fatty acid synthetase respond to dietary and hormonal manipulations- in a way which suggests that the level of the enzyme could be a reg- ulator of fatty acid synthesis. For example under conditions of high carbOhydrate intakes by rats the maximal rate of fatty acid synthesis may be restricted to the level of fatty acid synthetase enzyme in liver (19). Antibodies have been prepared against several lipogenic en- zymes of rat, mouse, and chicken liver and/or adipose tissue (l3, 20-24). Quantitative precipitin analyses indicate that changes in tissue en- zymeactivity reflect Shifts in enzyme protein content. The fluctuation in enzyme protein content appears to be a4 product of differential rates of enzyme synthesis and not varia- tion in catalytic efficiency or precursor protein transformation (20-24). Protein synthesis can be controlled at transcription, trans— lation or both. Those factors which have been proposed to precip— itate modifications in rates of enzyme synthesis include variations in hormonal relationships, cAMP and metabolite concentrations (25, 26). Based on the following data, Nepokroeff et a1. (27) have pro- posed that variation in the activity of.lipogenic enzymes which occurs with dietary manipulation is a product of coordinated regulation of synthesis of glucokinase, glucose-6-phosphate dehydrogenase, malic en- zyme, citrate cleavage enzyme, acetyl-COA carboxylase and fatty acid synthetase. Streptozotocin-diabetic rats fasted for 48 hours and re- fed a high carbohydrate diet displayed no elevation above fasting levels in lipogenic enzyme activities, whereas those refed animals administered insulin had linear increases for 48-60 hours in lipo- genic enzyme activities (27). The mode of action of refeeding car- bohydrate in normal rats was suggested to be via stimulation of in- sulin release (28). In comparison to insulin, injecting glucagon to normal, refed rats every 12 hours for 48 hours prevented the rise in lipogenic enzyme activity by 50% with the most notable effect on glucokinase which was only 14% of normal (27). To attribute these ef- fects of glucagon or insulin specifically to control of coordinated synthesis of the enzymes involved without considering substrate flux or metabolite changes may oversimplify the issue. Glucokinase may regulate the rate limiting step in hepatic glucose utilization. Thus effects on glucokinase activity, may be the only specific effect of insulin or glucagon. The resulting activity of other lipogenic en- zymes then reflects reduced metabolite flow or concentrations. ThiS- is substantiated by the observation that fatty acid synthesis, acetyl- CoA carboxylase activity and fatty acid synthetase activity were main- tained at normal rates in diabetic rats fed fructose (25). From the foregoing discussion it is clear that the exact role the metabolite and hormonal factors play in enzyme protein synthesis regulation is undefined at this time and likely involves a combina-' tion of-interrelated factors. Short-term regulation of fatty acid synthesis involves al- tering the rate of fatty acid synthesis without changing the total content of liver and/or adipose tissue glycolytic and lipogenic ma- chinery. Such control mechanisms permit the organism to maintain metabolic homeostasis during intermittent periods of energy con- sumption and deprivation. The nature of-Short-term regulators has been linked to changes in the concentration of several metabolites such as ATP, cAMP, long chain acyl—COA esters, fructose—6-phosphate, NADH/NAD ratios (19, 29-32). The quantitative importance of short-term regulators is typi- fied in meal-eating rats. Rates of liver and adipose tissue fatty acid synthesis peak very rapidly after beginning a meal, begin to de- cline within Six hours post-meal initiation, and are negligible just prior to a meal (33, 34). These fluctuationS'are also seen in malonyl- CoA concentration changes in meal-eating rats (19). In contrast to the diurnal variation in fatty acid synthesis, total in vitro activi- ties of acetyl-COA carboxylase, fatty acid synthetase, glucokinase etc., remain essentially constant throughout the 24 hour period (35). A number of locations for short-term regulation have been proposed: 1) reduced uptake and/or phosphorylation of glucose, (2) less pyruvate production and/or decarboxylation, (3) suppressed cit- rate efflux from mitochondria and hence reduced substrate for fatty acid synthesis, and (4) less amount of active acetyl-COA carboxyl- ase. Although variousmetabolites may act as the immediate agent, fluctuations in their concentration frequently are hormonally medi- ated; and as with control of enzyme protein synthesis, short-term ef- fectors probably involve a combination of hormone and metabolite changes. REGULATION VIA FLUCTUATION IN ENZYME ACTIVITY Glucokinase and hexokinase. The utilization of circulating glucose for the production of pyruvate (via glycolysis) and ultimately as a substrate for de novo fatty acid synthesis initially requires 10 phosphorylation to glucose-6-phosphate. This reaction is catalyzed by hexokinase which exists in four isozyme forms. Control of intracellu- lar glucose utilization at this point is a key regulatory step (15, 37). Three of the hexokinases (Types I to III) are low km enzymes located in most tissuesincluding the liver. Type IV hexokinase called glucokinase is exclusively found in the liver and is the enzyme of ma- jor importance in phosphorylating intracellular hepatic glucose. Glu- cokinase has a high km for glucose (lOmM vs. O.lmM for Types I-III), is not inhibited by glucose-6-phOSphate and varies in activity with changes in carbohydrate intake (36) in the rat, hamster, mouse, hu- man and dog. Ruminants and chickens have very little glucokinase ac- tivity in the liver (37, 15). Since the liver is freely permeable to glucose, several investigators believe that the phosphorylation of glucose catalyzed by glucokinase is the rate limiting step in overall hepatic glucose utilization (37, 38). After consuming a carbohydrate meal the concentration of glu- cose in portal blood increases to concentrations exceeding lO-12 mM (36). At this level of glucose the low km hexokinase would be satu- rated and has been estimated to have a maximum catalytic activity of 0.1 umoles/min/g liver, 25°. In contrast glucokinase will not be at near maximal velocity and has an estimated maximum catalytic activity of 1.5 umoles/min/g liver, 25°. In addition, a build up of glucose- 6-phOSphate in the liver would inhibit hexokinase but have no effect on glucokinase (39). Thus these two properties of the enzyme cata- lyzing the possible rate limiting step of glucose metabolism in liver 11 are compatible with the anatomical location and elevated rate of gly- colysis, glycogen synthesis and lipogenesis following carbohydrate consumption (36). The level of glucokinase in the liver is dependent on both the amount of dietary glucose as well as circulating insulin (36). Rats fed a carbohydrate-free high fat diet possessed very low gluco- kinase activities. The oral administration of a glucose solution to the animals caused a tremendous increase in glucokinase activity within three hours. However oral administration of glucose plus anti-insu1in serum treatment prevented the rise in glucokinase ac- tivity (36). In addition, insulin administration alone to diabetic rats could not enhance glucokinase activity (40). The action of in- sulin appeared to be Specific for glucokinase since normal levels of phOSphofructokinase and pyruvate kinase could be maintained in dia-' betic rats by feeding a fructose-glycerol diet (37). Fasting also led to a rapid decline in glucokinase activity in rats, mice and ham- sters but not guinea pigs (41). Upon refeeding of a high carbohydrate diet glucokinase activity increased, with the greatest effect result- ing from dietary glucose (42). From changes in enzyme activity during fasting, the half-life of glucokinase has been estimated at 32 hours. Evidence with protein synthesis and transcription inhibitors has indicated that an elevation in glucokinase activity is a product of enhanced enzyme synthesis (37). In addition immunological titrations revealed that the amount of 12 enzyme activity in vitro in rats of various nutritional states re- flected the amount of active enzyme protein (37). Glucokinase activity is not only altered by the level of di- etary fat but also responds differently to the composition of dietary fat (43, 44). Replacing 15% of the glucose in a high carbohydrate diet of gerbels with safflower oil was associated with a Significant decline in glucokinase activity. However, the addition of 15% coco- nut oil caused no change in glucokinase activity relative to the high carbohydrate-fed group. The reason for more efficient inhibition of glucokinase by safflower oil-containing diet was unknown but this ob- servation was in accord with observation in mice and rats that poly- unsaturated fats inhibit hepatic lipogenic enzymes with greater effi- cacy than saturated fats (7-10). Inhibition at the first step of glu- cose utilization in liver would be an efficient mechanism by which to inhibit lipogenesis resulting from dietary triglyceride. The reasons for differential response to saturated and unsaturated fatty acids at this time can only be speculative and require further elucidation. Glucokinase activity in vitro has been shown to be inhibited by added free fatty acids, acetyl-COA and phosphoenolpyruvate (45). The negative effect of free fatty acids could be prevented by the ad- dition of glucose. These observations were interpreted as a control mechanism in vivo for glucokinase during the transition of the liver glycolysis to gluconeogenesis. However the inhibition of glucokinase by free fatty acids was time and dose dependent which supports the contention that free fatty acids and their CoA derivatives exert a 13 non-specific detergent effect on most enzymes except acetyl-COA car- boxylase (29). Similarly the inhibitory effect of acetyl-COA and phosphoenolpyruvate was concentration dependent and non-competitive with glucose (45). More recent alternatives for the regulation of glucose uptake and release in liver during the transition from gly- colysis to gluconeogenesis have been (a) substrate cycling catalyzed by glucokinase and glucose-6-phosphatase and (b) compartmentation of glucokinase and glucose-6-phosphatase (36). Unlike the liver, adi- pose tissue is not freely permeable to glucose (36). Therefore in order for adipose tissue to utilize glucose for de novo fatty acid synthesis, glucose must first pass through the adipocyte plasma mem- brane. The entry of glucose into fat cells is greatly facilitated by insulin and may be the rate limiting step in adipose tissue glu- cose utilization (36). Under such conditions adipose tissue hexo- kinase activity may play less of a key role in determining rates of glucose utilization than does glucokinase activity in the liver. Nevertheless adipose hexokinase is an adaptive enzyme, responsive to nutritional manipulation and may still serve an important function in regulating the rate of substrate flow through glycolySis and ul- timately lipogenesis (46). Hexokinase in rat adipose tissue is the low km isozyme and the variety which predominates appears to be age dependent (15). Fasting and alloxan-diabetes in the rat was associated with a de- crease in hexokinase activity, while refeeding a high carbohydrate diet or injecting insulin resulted in an increase in hexokinase 14 activity (15). Human adipose tissue hexokinase activity responds to fasting and refeeding in a manner similar to rats, but diabetic hu- mans did not display a reduction in hexokinase activity nor did the activity increase upon insulin injection (47). 14C-histidine From the use of protein synthesis inhibitors and incorporation into hexokinase, the elevation in activity appears to involve enzyme protein synthesis. The stimulus for synthesis includes both glucose and insulin (15, 48). A Short-term regulatory mechanism for hexokinase activity may exist in the ratio of soluble vs. membrane bound enzyme (49, 51). A very large portion of hexokinase activity has been identified as associated with mitochondria (50). In chick brain evidence has indicated that conditions which lower ATP availa- bility (e.g. anoxia) increased the amount of mitochondrially bound hexokinase whiCh increased the k1 for glucose-6-phosphate five-fold and reduced the km for ATP seven-fold (51). In vitro bound hexon kinase was solubilized from the mitochondria and reduced in activity by added glucose-G—phosphate (50). Binding of hexokinase to mito- chondria purportedly stabilizes the enzyme and places it in closer proximity to ATP generating machinery (50, 52). Even though in rats fed a high carbohydrate diet the amount of bound hexokinase was greater than in fasted rats, the exact physiological role soluble and bound enzymes play in glucose metabolism of adipose tissue re- quires further elucidation (53). Phosghofructokinase. ~Phosphofructokinase (PFK) exists in i- sozyme form among tissues with the L-type in liver, M-type in muscle, 15 and a M-L variety in adipose tissue. Purified PFK (generally M-type) is inhibited by ATP, citrate and creatine; and stimulated by fructose- l,6-diphosphate (FDP), AMP, cAMP and Pi (30, 54). In addition to these metabolite effectors of PFK, a phosphorylated form of PFK has been isolated from mouse skeletal muscle (55). The active form of PFK has been proposed to be a phosphorylated tetramer which dissoci- ates to inactive protomers upon dephosphorylation by a magnesium de- pendent phosphatase (56). The reactivation rate by ATP of the in- active protomers in rat liver was greatly reduced by only Six hours of fasting but the rate was restored to normal within six hours of refeeding following a 24 hour fast (56). The physiological signify icance of these modifiers varies with tissue and the effect of the modifiers is not always consistent with metabolite concentrations and physiological conditions. For example, injection of glucagon into portal vein of rats increased liver cAMP levels three fold but PFK activity was depressed by 50% within four minutes (30). Insulin on the other hand increased PFK activity 50% with no change in cAMP concentration (30). Three days of fasting reduced rates of glycolySis in rat liver while the cAMP concentration dramatically rose (57). These observations are not consistent with cAMP actiVation Of puri- fied PFK but are in accord with known effects of fasting on rates of glycolysis (57). If ATP inhibits PFK then a low ATP/ADP ratio would favor PFK activation and promote gylcolysis. However physiological states such as fasting and diabetes result in low rates of hepatic glycolysis'and a low ATP/ADP ratio (31). Because the ratio of 16 ATP/ADP and the absolute content of ATP in adipose tissue changed very little in fed, fasted-refed or fasted rats, Ballard (32) has questioned the significance of the ATP/ADP ratio as a control mech- anism of adipose tissue fatty acid synthesis. In support of this- the concentration of adenine nucleotides per milligram wet tissue or per milligram DNA in rat epididymal fat pads incubated in vitro with albumin, insulin or epinephrine did not vary from pads of normal rats incubated with no additions (58). Attributing control of glycolytic flux and in particular PFK activity to one ratio or metabolite may be an oversimplistic approach (54). The oscillatory behavior of glycolytic flux in rat muscle ex- tracts indicated that PFK activity is not only dependent on ATP/ADP but greatly affected by FDP concentration plus the concentration of AMP and fructose-6-phosphate (F6P) (54). Several workers have sug- gested-that a better indicator of glycolytic rate may be the ratio of F6P or FDP to ATP (38, 54, 59). This approach is supported by in vitro metabolite data in fat pads isolated from normal and diabetic rats (58). The addition of insulin to the media increased the uptake ofglucose, the concentration of glucose-6-phosphate (GGP) and pre- sumably F6P, and the level of glycerol-phosphate which collectively~ indicated an accelerated glycolytic flow. The inclusion of epi- nephrine with insulin reduced GGP content of the fat pads by 50% and glycerol-phosphate content by several fold. Under both condi-- tions ATP/ADP ratios did not significantly change, although citrate was elevated (58). These parameter variations are in accord with 17 insulin's ability to promote lipogenesis and epinephrine's effects in fasting. The role of citrate as an inhibitor of PFK and hence glyco- lysis is sometimes inconsistent with the part it plays in promoting lipogenesis in liver and adipose tissue (15). Little difference in the hepatic content of citrate existed between starved, diabetic, fasted-refed or insulin-injected diabetic rats (19, 38, 60). In ad- ipose tissue both fasting and epinephrine treatment (in vitro)-re- sulted in a significant increase in rat adipose tissue citrate levels (32, 58), which agrees with its inhibitory effect on glycolysis and PFK activity but disagrees with its stimulatory role for acetyl-COA carboxylase. Similarly resting muscle preferentially utilizes cir- culating free fatty acids which leads to a build-up of muscle citrate and-thus potentially could Slow glucose utilization via glycolysis by PFK inhibition (61). Recently Ramadoss et al.(62) demonstrated that free fatty acids inactivated rabbit muscle PFK. This-effect could be lessened by prior incubation of the enzyme with F6P, AMP,_FDP or cAMP.A Al- bumin would partially protect the enzyme if added before the free fatty acids. However once the enzyme was inactivated by palmitate or oleate, albumin was unable to reverse the inhibition. Because of the high concentrations of palmitate and oleate utilized and the irreversible nature of the inhibition, the effect probably_was a nonspecific detergent action of the free fatty acids rather than a Specific regulatory mechanism (29). 18 In addition to modulation of PFK activity by several differ- ent metabolites the actual amount of enzyme can vary with nutritional conditions. PFK activity in rat liver and adipose tissue was de- pressed by fasting and alloxan diabetes (15, 63). Refeeding or in- sulin injection resulted in the activity of PFK returning to normal levels (15, 64). The effect of insulin may be secondary to its ac— tion on glucose uptake and phosphorylation because PFK activity could be maintained at high levels in the liver of diabetic rats fed a fructose-glycerol diet (37). Pyruvate kinase. Pyruvate kinase (PK) catalyzes the forma- tion of pyruvate from phosphoenolpyruvate with an equilibrium cone stant greatly in favor of pyruvate formation (36, 65). The in vitro activity of PK suggests it catalyzes a reaction at near equilibrium rates whereas the mass-action ratio indicates the reaction is not nearequilibrium.(36). The allosteric properties of PK offer an explanation for this discrepancy and contribute to the regulatory role PK plays in glucose metabolism via glycolysis (63). PK is sharply inhibited by ATP and alanine while the inhibition is rap- idly reversed by FDP (66). During gluconeogenesis inhibition of PK activity is desireable to avoid futile dephosphorylation of phos- phoenolpyruvate. Thus in light of alanine's major contribution as a gluconeogenic precursor, its inhibition of PK activity appears reasonable.) Similarly activation of PK by FDP would facilitate flow of glucose through glycolysis and enhance pyruvate production (36, 15, 38); However the function of ATP is unclear Since in rat 19 liver its concentration increases during high rates oflipogenesis and glycolysis which is inconsistent with ATP-inhibition of phospho- fructokinase and PK (38).* As previously cited the ratio of FDP/ATP‘ may be an important, fundamental factor in controlling PK activity (38). Like phOSphofructokinase, PK does appear to be adaptative in rat liver but does not reSpOnd to dietary manipulation in chick liv- er and only minimal changes occur in rat adipose (15). Fasted and alloxan-diabetic rats displayed lowered hepatic PK activities while refeeding or insulin injection elevated PK activity above normal. The increase in hepatic PK was prevented by first administering ac- tinomycin or ethionine which suggested the increase in enzyme activ- ity under these conditions represented de novo enzyme synthesis (45). As with phOSphofructokinase the effect of insulin may be secondary~ to its primary effect on glucose phosphorylation in liver because PK activity could be maintained above normal in diabetic rats-fed fruc- tose (37). Adipose tissue PK was not reduced in fasting rats or by treating them with alloxan (64). The enzyme did slightly increase in refed rats fed a high carbohydrate diet and in meal-trained rats (57. 68). ‘ Pyruvate dehydrggenase. In rats, mice, chickens, pigs and presumably man the converSion of pyruvate to acetyl-CoA in liver and/or adipose tissue is fundamental to producing substrate for fat- ty acid synthesis.. This reaction is catalyzed by pyruvate dehydrogenase 20 (PDH), an intramitochondrial enzyme, which is very important in modi- fying rates of pyruvate decarboxylation. PDH is a nonadaptive enzyme whose activity is controlled by shifts between phosphorylated (inactive) and nonphOSphorylated (active) forms of the enzyme (69, 14). Phosphorylation of PDH is governed by a kinase requiring Mg-ATP and which is independent of cAMP (14, 15). The inactivation of PDH in isolated mitochondria of rat heart was prevented by high concentrations of pyruvate (70). Presumably pyruvate inhibits kinase activity (71). Dietary regimens which-are associated with high rates of glycolysis and lipogenesis would favor pyruvate production and enhance active amounts of PDH. Rats in a fed state have been Shown to have one-sixth active PDH in liverrand two-thirds dephosphorylated formin adipose tissue (72).. A large proportion of active hepatic PDH has been found in rats-re- fed a high carbohydrate diet, injected with insulin or perfused with fructose (73-75). Similarly adipose tissue PDH was activated by ad- ding glucose, fructose or pyruvate to an in vitro incubation media and by insulin in the absence of sabstrate (76). The action of in- sulin in adipose has been proposed to be via an enhancement of PDH- phosphatase activity. Such an effect of insulin would potentially override the negative influence of elevated ATP/ADP ratio in liver and adipose tissue of rats-during high rates of lipogenesis (58, 77). Dephosphorylation of PDH is-catalyzed by a Mg++ - dependent phosphatase (15). The phosphatase has a Km for Mg++ which is well above the.total cell Mg++ concentration. Therefore its activity is 21 sensitive to changes in Mg++ levels (15). An increase in intramito- chondrial ATP concentration would lead to a greater proportion of Mg- ATP complex, reduce the availability of Mg++ and thereby lower phos- phatase activity (15, 54). A reciprocal relationship between mito- chondrial ATP levels and the proportion of active PDH has been demon- strated in rat liver mitochondria (14). Free fatty acids and/or their CoA esters have been prOposed to control pyruvate utilization by reg- ulating the proportion of active PDH (14, 73, 78, 79). Injection of oleate promoted the amount of phosphorylated rat liver PDH; addition of oleate or octanoate to an in vitro media depressed the proportion Of rat adipose tissue active PDH (73, 78). The reduced quantity of active PDH in liver and adipose elicited by free fatty acids was suggested to be due to the inhibition of mitochondrial adenine nucle- otide translocase activity by long chain acyl-COA esters (14, 80) which led to an increase in mitochondrial ATP concentration. In addition to ATP pyruvate dehydrogenase activity is affected by several metabolities: pyruvate, NADH, acetyl-COA and ADP (79, 8l-83). The effects of these metabolites are exerted on the inactivating kinase portion of the PDH-complex (79, 83). The PDH-kinase has been shown to be activated by NADH and acetyl-COA and inhibited by pyruvate and ADP (79). Recent investigations with isolated rat liver (81) and heart (79) mitochondria incubated with octanoate or palmitoylcarnitine have demonstrated an increase in the amount of NADH and acetyl-COA with a concomittant decrease in pyruvate decarboxylation and active PDH (79, 81). Because these changes were independent of changes in ATP 22 concentration or ATP/ADP, the inhibitory action of fatty acids on the activity of PDH may be more related to increases in NADH/NAD and acetyl-CoA/COA ratios-from B-oxidation than to elevated ATP/ADP ra- tio resulting from translocase inhibition (79). This is also con- sistent with a lower NADH/NAD ratio during high rates of hepatic li- pogenesis even though ATP/ADP ratio increases (36). AcetyJ-COA carboxylase. Dietary and hormonal manipulations which increase or decrease rates of liver and adipose tissue fatty acid synthesis are also associated with parallel changes in acetyl- CoA carboxylase (AcCBX) activity (13, 25, 84-87). Hepatic AcCBX activity is reduced in rats, mice and chicks by fasting, high fat diet and alloxan-diabetes (25, 84-86) and elevated in activity by refeeding high carbohydrate diet, feeding a fat-free diet, and feed- ing fructose (13, 25, 84-86). Quantitative antibody precipitin anal- yses have indicated that these changes reflect alterations in enzyme protein content which is accomplished by differential rates of en- zyme synthesis without changes in rates of degradation of AcCBX (84). Glucocorticoid administration to adult rats depressed rat adipose tissue AcCBX content without affecting rat liver AcCBX. Glu- cagon injection during refeeding of a high carbohydrate diet pre- vented the rise in rat liver AcCBX by 50% (87). Similarly insulin injection to diabetic rats-elevated hepatic and adipose AcCBX (25). The effects of these hormones have been proposed to be prior to al- dolase because fructose feeding to diabetic rats maintained normal rates of liver fatty acid synthesis and AcCBX (25). 23 Although the level of AcCBX probably has a governing effect on rates of fatty acid synthesis under prolonged nutritional situa- tions, its half-life of 40-50 hours does not explain the normal di- urnal short term changes which occur in rates of fatty acid syn- thesis. The key role which AcCBX plays in short term or fine reg- ulation of fatty acid synthesis can be appreciated from the al- losteric properties of the enzyme (15). From electron microscopy and density-gradient centrifugation purified AcCBX has been shown to exist in an inactive protomer which can be converted to an ac- tive polymer (ll). Polymer formation of purified AcCBX is enhanced by citrate while depolymerization is promoted by long chain fatty acyl-CoA derivatives (88, 89). This mechanism has been confirmed in rat liver and adipose tissue and avian liver (90, 91). The Ki of AcCBX for fatty acid CoA derivatives varies with the type of fat- ty acid and appears to be lowest for saturated fatty acyl-COA es» ters (92). The physiological significance of citrate's influence on AcCBX activity is open to question. The.citrate content of liver (whole tissue analysis) in rats fed a high fat diet was two to three times higher than nibbling, meal-fed, or refed rats and yet the rate of fatty acid synthesis in the fat-fed rats was very low (19). Similarly fat pads of fasted rats contained more citrate per gram of wet tissue than tissue from fed or refed rats (32). These observations are consistent with citrate's negative effect 24 on glycolysis via inhibiting phOSphofructokinase but inconsistent with its activator role with AcCBX. One difficulty with total tis- sue analysis is an inability to differentiate metabolite compart- ments which could alter the local effective concentration of a me- tabolite. Thus total tissue citrate may not change but the concen- tration in various micropools could fluctuate. Several observations support a reversible, specific inac- tivation of AcCBX by long chain fatty acyl-COA compounds and sup- port the contention AcCBX is a key step in short-term regulation of fatty acid synthesis. Many enzymes are inhibited by CoA deriv- atives of fatty acids (93, 94) but these effects were shown to be nonspecific and irreversible, and likely the result of a detergent action of the fatty acids. Goodridge (29) clearly demonstrated that palmitoyl-COA in the presence of high amounts of albumin in- hibited both AcCBX and 14C-citrate incorporation into fatty acids. This inhibition could be reversed by increasing the albumin con- centration of the system. Following intubation of corn oil to rats previously fed a fat free diet, a two fold increase in he- patic fatty acyl-CoA concentration and a 100% reduction in hepatic in vitro fatty acid synthesis occured within two hours (95, 96). Since this time period was too Short to attribute the change in fatty acid synthesis to a reduced enzyme protein concentration, the alternative explanation could be an inactivation of some en- zyme in lipogenic pathway, presumably AcCBX (24, 95, 96). 25 Data for the effect of acyl-CoA esters in adipose tissue are more variable. The in vitro exposure of rat fat pads to insulin plus glucose doubled the amount of active AcCBX without influencing total AcCBX activity. Insulin alone had no effect (89). Conversely the inclusion of epinephrine with insulin and glucose caused a 50% drop in polymeric form of AcCBX (88). A negative correlation between AcCBX activity and adipose tissue fatty acyl-COA concentrations was observed but this correlation was inconsistent (88). For example, the long chain fatty acyl-CoA level in adipose tissue of rats fasted for 36 hours was Significantly below the concentration in fed ani- mals, but during this time total adipose AcCBX changed very little (88).. These results are in contrast to data derived from rat liver (97) where the long chain acyl-COA content in liver of fasted rats (18-48 hours) was twice normal. The inconsistency in data between liver and adipose tissue may be related to the high triglyceride content of adipose tissue which must be extracted before analysis for CoA derivatives (88). Fatty acid synthetase, -For several years the reaction cat- alyzed by acetyl-COA carboxylase was considered the rate limiting step in lipogenesis (98). However following improvements in acetyl- CoA carboxylase assay system, data now indicate that either carbox- ylase or fatty acid synthetase (FAS) catalyzed reactions may be rate limiting depending upon the nutritional state of the animal (19, 70). 26 Unlike acetyl-CoA carboxylase or pyruvate dehydrogenase, al- losteric regulation of FAS probably is insignificant in the Short term control of fatty acid synthesis(35). Early work with purified pigeon liver and rat liver FAS indicated inhibition of the enzyme by palmitoyl-COA (99, 100). However Goodridge (29) clearly demonstrated that palmitoyl-COA inhibition of enzymes other than acetyl-CoA car- boxylase was due to an irreversible nonspecific detergent effect. Pigeon liver FAS was reported to be stimulated by phosphorylated sug- ars-(e.g. FDP) and inhibited by malonyl-COA (101).. However the phys- iological significance of these observations remains to be defined because the concentrations of phosphorylated SHgars used were well above the physiological level (102). Further research has been un~ able to confirm the earlier work of Wakil and associates (101, 102). FAS activity-varies with nutritional and hormonal conditions (86—88). Feeding a high carbohydrate fat-free diet to rats, mice and chicks resulted in a relatively high liver and/or adipose tissue FAS activity (7-10, 103). Addition of safflower oil, corn oil, lino- leate or linolenate to the fat-free diet precipitated a rapid and marked decline in FAS activity in rat, mouse and chick liver (7-10, 103). Fasting for 24448 hours also greatly reduced rat liver FAS activity while refeeding led to a large increase in FAS activity (20). Alloxan-diabetic rats possessed very low liver and adipose tissue FAS activities but treatment with insulin or fructose feed- ing restored hepatic FAS activity to near normal (25).! The injection 27 of glucagon to rats during a refeeding period greatly prevented the expected rise in hepatic FAS activity as well as several other li- pogenic enzymes (27). However this preventative effect may well have been secondary to the near total lack of increase in gluco- kinase activity (27). Rat adipose tissue FAS activity was reduced by glucocorticoid administration but liver FAS was unaffected (87). Clearly FAS responds to different dietary and hormonal manipula- tions. Whether these effects are primary to the agent or secondary to some other change (e.g. glucokinase vs. FAS activity) remains to be ascertained. Changes in FAS activity have been demonstrated not to be the product of alterations in catalytic efficiency but reflect ac- tual variations in enzyme protein content (20, 25). Control of enzyme protein content occurs by altered rates of synthesis and/or degradation. Under most conditions-the level of FAS activity in a tissue is varied by differential rates of synthesis (20, 26). The rate of degradation of FAS has been estimated in nonfasted rats and a half-life of 47-75 hours has been reported (20, 26). In the absence of definitive data for allosteric effectors of FAS, Guynn et al.(19) have concluded that in rats-fed a high- carbohydrate diet.the rate of hepatic fatty acid synthesis depends on the quantity of FAS. This conclusion was based on the observa- tion that animals fed the carbohydrate diet has a build-up of liver malonyl-COA content. Such control of hepatic fatty acid synthesis 28 would be long-term in nature and does not preclude-shOrt-term or fine control through allosteric changes in glycolytic enzymes. pyruvate dehydrogenase or acetyl-COA carboxylase. Citrate_cleavgge enzyme. In nonruminant animals citrate cleav- age enzyme (CCE) catalysis of citrate hydrolysis to oxaloacetate and acetyl-COA has been considered a major mechanism in the generation of cytosolic acetyl-COA for de novo fatty acid synthesis. Like several other lipogenic enzymes CCE activity in liver and adipose tissue of rats appears to be adaptive. Because an increase in CCE activity was associated with greater rates of enzyme synthesis in rat liver, Gibson et al. (104) concluded that differential rates of enzyme syn- thesis govern the level of CCE activity. CCE activity was shown to be reduced in rat and chick liver, and rat and pig adipose tissue by fasting, fat-feeding and alloxan diabetes (105-109). Conversely feeding a high carbohydrate diet promoted CCE synthesis and increased its activity in rat liVer and adipose (105, 106). Current data have not revealed short-term effectors for CCE and generally its activity is sufficiently high so that it would not be a rate limiting reaction. However Yeh and Leveille (llO) sug- gested that in fasted chicks a limited availability of free COA be- cause of the rapid rise in hepatic long chain acyl-COA esters could limit the activity of CCE. Such a hypothesis has not been proposed for other species. Glucose-G-Dhosphate dehydrogenase and malic enzyme,~ De novo fatty acid synthesis-requires reducing equivalents in the form of NADPH. 29 In rat liver and adipose tissue the production of NADPH results from the flow of G6P through the hexose monophOSphate shunt and via the conversion of malate to pyruvate. In contrast to the rat, chick liver has low hexose Shunt activity and depends heavily on malate- pyruvate cycle. The two key enzymes involved are glucose-6-phosphate dehydrogenase (G6PD) for the hexose shunt and malic enzyme (ME) for the malate utilization (ll, 15). At one time the activities of G6PD and ME and the subsequent utilization of NADPH were considered a de- termining force in driving fatty acid synthesis (111). However fur- ther studies have demonstrated that the activities of GGPD and ME can change independent of fatty acid synthesis rates and probably respond to total tissue NADPH demands (112-114). In adipose tissue undergoing reasonable rates of-fatty acid synthesis, McLean (115) found that G6PD activity was not function- ing at maximal capacity. The tissue G6PD activity could be stimu- lated by addition of acetate as a fatty acid precursor or by addi- tion of phenazine methosulphate, an electron acceptor (115). GGP is not only substrate for G6PD but can also be metabo- lized via glycolysis or utilized for glycogen syntheSis. The de- termining factor in flow of G6P through GGPD does not appear to be substrate limitation because the hepatic concentration of GGP is three times higher than GGPD Km (15). Taketa and Wantanabe (21) have isolated monomer, dimer and tetramer forms of GGPD from rat liver cytosol. NADP stimulated dimer formation and the dimer exhib- ited sigmoidal kinetics for NADP. They proposed that the 30 physiologically active form of G6PD was the dimer, and that as NADPH levels rose the structural NADP component of the dimer was replaced with NADPH thereby inactivating the enzyme (21). Thus high rates of NADPH utilization would promote activation of G6PD and flow of G6P through the hexose Shunt (21). Such utilization occurs in rats with high carbohydrate intakes and low polyunsaturated fat diets (112, 113). Both G6PD and ME activities fluctuate with dietary and nor- monal alterations. Fasted and alloxan-diabetic rats displayed re- duced hepatic and adipose tissue G6PD and ME activities (15, 116, 117). Upon refeeding fasted rats a high glucose diet both enzymes rebound in activity to the point of ”overshooting" the levels found in unfasted rats maintained on a similar diet (111, 114, 118). An apparent greater increase in G6PD and ME resulted from a second fasting-refeeding period (119). Although both enzymes were reduced in liver of rats fed high fat diets (113), diets high in polyunsat- urated fatty acids precipitated the reduction with far greater ef- ficacy (112, 113). Goodridge (120) was unable to observe a decrease in liver ME in chicks fed diets containing 10 or 15% corn oil. In contrast Yet et al. clearly demonstrated chicks did respond to in- creased amounts of dietary fat by lowering ME activity (107). In addition to response to dietary fat, G6PD activity was shown to be very sensitive to the amount of glucose consumed by rats each day (121). This response may be due to greater availability of glucose per se, a metabolite of glucose or the product of enhanced 31 insulin release. Insulin injections markedly elevated G6PD activity in rat liver but this effect was overshadowed by an increased food intake caused by insulin (116, 122). The type of dietary carbohy- drate (e.g. glucose, fructose) may also influence the activity of the enzymes and may lead to independent responses between rat adi- pose tissue and liver (15). For example, in conjunction with ac- celerated rates of liver fatty acid synthesis with fructose and sucrose feeding was an increase in G6PD activity above that seen in rats fed glucose (123, 124). However under these conditions adi- pose tissue G6PD activity was unchanged (124). G6PD and ME do respond differently to dietary protein (125). AS previously stated refeeding a high carbohydrate diet to fasted rats caused an "overshoot" in the activities of both enzymes. How- ever if only carbohydrate is refed G6PD activity does not increase (114) whereas ME activity did increase even on extremely low protein diets (126). G6PD and ME both appear to fulfill the following criteria established by Weber and associates (63) to determine if in vitro enzyme activity increases represent an actual rise in enzyme amount and not catalytic efficiency: a) blockage of rise in activity by protein synthesis inhibitors; b) increased incorporation of amino acids into purified enzyme protein, and c) immunological evidence for elevated amounts of purified enzyme. Puromycin injection prevented the rise in G6PD activity upon refeeding rats a high carbohydrate diet (111). Similarly, injection 32 of 8-azaguanine blocked the overshoot in G6PD activity, but did not prevent G6PD from returning to prefasted normal level (127). Since 8-azaguanine inhibits transcription, the G6PD overshoot presumably involves additional RNA synthesis. Recently studies using antibodies against rat liver G6PD revealed that the enzyme from rats fed a standard pellet diet or high carbohydrate diet reacted identically to antibody precipita- tion (118). Therefore the higher activity in rats fed the high carbohydrate diet was considered a difference in concentration and not a change in enzyme catalytic efficiency (118). From the rate 14C-amino acids into immunoprecipitated liver of incorporation of G6PD, Garcia and Holten (128) demonstrated the elevation in G6PD upon refeeding fasted rats was the product of enhanced enzyme syn- thesis. Injection of glucagon during the refeeding period dampened the increase in G6PD activity by reducing its synthesis but without influencing the rate of degradation (128). Similarly quantitative immunoprecipitation of ME in rats re- fed a high carbohydrate diet demonstrated that the higher enzyme ac~ 14 tivity was associated with greater C—leucine incorporation into the enzyme protein (117, 125). Thyroxine treatment which enhances 14C-leucine incor- ME activity in rats fed chow, also accelerated poration into ME protein (125). Quantitatively Murphy and Walker (117) demonstrated that thyroxine treatment led to a four fold rise in ME synthesis and a Six fold rise in Specific activity. In neonatal 33 chicks fed a commercial diet the rate of ME synthesis rose 54 times while local activity increased 63 fold (13). Both ME and G6PD appear to increase their activities primar- ily by accelerated rates of enzyme protein synthesis. The G6PD and ME activities of liver and adipose tissue do not determine rates of fatty acid synthesis but most likely change in activity in response to NADPH demands. The modulators of G6PD and ME synthesis have been proposed to include insulin, unsaturated fatty acids, glucose and cAMP (15, 113, 114, 118, 128). However Characterization of the re- pressors and stimulators remains to be ascertained. REGULATORY ASPECTS OF LONG CHAIN ACYL-COA ESTERS Rats and chickens that were fasted or fasted and refed a fat diet displayed very low rates of fatty acid synthesis and Signifi- cant elevations in total hepatic long chain acyl-COA thioester con- centrations (19, 31, 97, 100, 129). In addition to the influence of acyl-COA thioesters exert on acetyl-COA carboxylase activity dis- cussed previously, these derivatives have been implicated in nega- tively affecting two mitochondrial transport systems: a) ATP/ADP translocase and b) citrate efflux (ll, 36). Adenine nucleotide translocase is an inner mitochondrial en- zyme which catalyzes exclusively a molecule-for-molecule exchange of ADP and ATP between the cytosol and mitochondria (11). In this 34 way the energy state of the two compartments remain interconnected. The translocation of adenine nucleotides across the inner mitochon- drial membrane can be inhibited by atractylate, bongkrekic acid, and long chain fatty acyl-COA esters (ll, 80). The acyl-COA inhib- itory action cannot be mimicked by free fatty acids or carnitine 14C-ADP by isolated rat and guinea pig liver derivatives. Uptake of mitochondria was inhibited by COA derivatives of myristic, palmitic, stearic and oleic acids. Linoleyl-COA also inhibited translocation but with considerably less effectiveness (130). Guinea pig mito- chondria. Octanoyl-COA was without effect with mitochondria of ei- ther species. Kinetic analyses of ADP uptake indicated palmitoyla COA inhibition was reversible and competitive in both rat and guinea pigs. Mitochondria isolated by differential centrifugation from the livers of alloxan~diabetic rats or monkeys, and of hibernating 32 ground squirrels showed very low capacities for P-ATP exchange 14C-ADP translocation. These low rates were reversed by adding and 5mM D, L-carnitine or 15 mg albumin which apparently facilitated re- moval of the fatty acyl-COA esters by forming carnitine derivatives (131). 14C-ADP translocation was inhibited 92% by 3mM fatty acyl-COA esters but in all cases 5mM carnitine addition readily reversed the inhibition (131). This was taken as evidence that the Site of acyl- CoA inhibition is at the inner mitochondrial membrane and the inhi- bition is reversible (131). 35 Recently translocase activity was quantitated in ischemic and non-ischemic regions of dog heart (132). Within 15 minutes ischemic areas had Significantly less translocase activity than non-ischemic regions. Concomittantly there was a two-fold increase in total tis- sue acyl-COA concentration in ischemic hearts. Translocase activity continued to sharply decline at 30 and 60 minutes while acyl-COA con- tent rose linearly. Mitochondria isolated from livers of fasted or alloxan~ diabetic rats contained twice the amount of acid insoluble COA lev- els relative to carbOhydrate-fed animals (133). Whether these COA derivatives are intramitochondrial, Specifically bound to mitochon- drial sites, or nonspecifically attached during mitochondrial iso- lation has yet to be established. In addition to having elevated acyl-COA contents fasted or fasted-refed (fat diet) rats had con- siderably greater intramitochondrial ATP/ADP and lower NAD/NADH ra- tios than animals fed a carbohydrate diet (31). These changes are consistent with the inhibitory effect long chain fatty acids appear to exert on pyruvate metabolism. In conjunction with adenine nucleotide translocase control by fatty acyl-COAS they also appear to affect citrate or tricarboxylic acid transport from the mitochondria (133). Citrate efflux from mitochondria in nonruminant animals is essential for the transfer of mitochondrial acetyl-units into the cytosol for utilization in de novo fatty acid synthesis (11). Cit- rate exit from the mitochondria supposedly is via a specific carrier 36 system which functions by exchange diffusion. This means the exit of one citrate molecule is accompanied by the uptake of a tricarboxylic or dicarboxylic acid, e.g. malate (134, 135). Palmitoyl-COA greatly 14C~citrate with unlabelled inhibited the exchange of mitochondrial media citrate (136). This effect could be prevented by including al- bumin in the system, was partially reversed by the addition of carni- tine and was competitive with citrate (136). The exchange of malate with inorganic phOSphate (media) was also inhibited by palmitoyl-COA but the Ki was three times greater for all dicarboxylic acid systems than tricarboxylic systems (136). Using the citrate exchange technique, Cheema-Dhadli and Halperin (133) found that mitochondria isolated from diabetic or fasted rats had a Km for citrate transport which was two-fold greater than mitochondria from fed rats.. Interestingly Vmax for citrate transport did not differ among the various states. Con- comittantly the fasted and diabetic rats had twice the amount of mitochondrial acid-insoluble COA content (133). The inhibition of adenine nucleotide translocase and cit- rate transporter by long chain fatty acyl-COA derivatives is com- patible with known in vivo changes in total tissue acyl~CoA content which occur during fasting, fat-feeding and diabetes. In addition these changes are associated with low rates of de novo fatty acid synthesis-and consistent with control of pyruvate oxidation at the level of pyruvate dehydrogenase (15, 82). 37 Yeh and Leveille (129) proposed that free fatty acid acyla- tion with free COA can limit the amount of free available COA for citrate cleavage enzyme function and thereby limit rates of lipogenr esiSa This contention is supported by the knowledge that rat liver fatty acid activating enzyme has a lower Km for COA than does cit- rate cleavage enzyme (129, 137). Increased plasma free fatty acid levels are found in feeding high fat diets to chicks and rats (100, 129) and in short—term fasting in chicks (110). The elevated plasma free fatty acid levels were accompanied by a large increase in he- patic long chain acyl-COA derivatives and significantly less free COA (100, 129). The quantitative significance of each mechanism is unknown but obviously long chain fatty acyl-COA compounds play a widespread role in controlling de novo fatty acid synthesis. THE ROLE OF REDOX STATE IN CONTROL OF FATTY ACID SYNTHESIS The rate at which reducing equivalents are utilized during de novo fatty acid synthesis in liver and adipose tissue may be an impor- tant regulator of the rate of lipogenesis (138). Diabetic, fasted and fasted-fat refed rats-display low rates of hepatic and adipose fatty acid synthesis and high lactate/pyruvate cytosolic ratios; whereas rats: fed a high carbohydrate diet or fasted-carbohydrate refed animals have 38 lower lactate/pyruvate cytosolic ratios, rapid glycolytic flux and high rates of fatty acid synthesis (31, 46). Adipose tissue taken from fasted rats displayed high lactate/ pyruvate ratios and low rates of glucose conversion to fatty acids (46). The addition of acetate reduced the concentration of lactate and led to accelereated rates of lipogenesis. The implication is that fasting does not prevent acetate utilization for fatty acid syn- thesis in rat adipose tissue but rather exerts a greater influence on glucose degradation to acetate. The elevated lactate/pyruvate ratio presumably represents an increased cytosolic NADH/NAD ratio which would slow glycolytic flux markedly at the level of glyceraldehyde phosphate dehydrogenase (46). A rise in cytosolic NADH generally is associated with a rise in mitrochondrial NADH (31). Thus the amount of active pyruvate dehydrogenase would be lessened and pyruvate con— version to acetate would be affected (79). In the liver of fasted rats the level of acyl-COA esters is increased along with the rate of B-oxidation (100, 139). The net effect in the liver of fasted rats is a more reduced cellular state and blockage of glucose conversion to acetate (31, 46). Control of fatty acid synthesis through changes in redox state of liver was demonstrated in rats fed diets containing buta- nediol (140). Liver slices of animals fed butanediol had depressed rates of glucose conversion to fatty acids, but acetate incorporation was unimpaired (140). In conjunction with lower fatty acid synthe- sis rates, rats fed butanediol had markedly elevated cytOplasmic 39 NADH/NAD ratios in the liver (141). Purportedly a high NADH/NAD ratio slowed flux through glyceraldehyde-3-phosphate dehydrogenase because of inadequate NAD supplies (140). Cytosolic NADH/NAD and NADPH/NADP pools are closely integrated and vary in concert (31, 142). Thus as the cytosolic NADH level in- creases, then the proportion of NADPH rises. The amount of NADPH has extreme implications with control of pentose shunt activity. Glucose- 6-phosphate dehydrogenase requires NADP for activity-(116). Thus dur- ing high rates of lipogenesis more NADPH iS oxidized (31) and hexose shunt flux is accelerated (36). Conversely enhanced fatty acid de- gradation and increased NADH-NADPH levels in liver will reduce both hexose shunt flow (31) and glycolytic rates with the net effect of re- duction in fatty acid synthesis. RELATIONSHIP AMONG SPECIES BETWEEN NUTRITIONAL STATE AND FATTY ACID SYNTHESIS Rats, The rat synthesizes fatty acids both in the liver and adipose tissue. The proportion that each tissue contributes to over- all lipogenesis has not definitively been ascertained and may vary with the type of diet (123) and pattern of dietary consumption (143). In the non meal-eating animal both organs can generally be considered to contribute equally to overall fatty acid synthesis. Masaro and associates (144) demonstrated several years ago that liver slices from rats consuming a carbohydrate-free diet 40 14C-glucose into fatty displayed very little capacity to incorporate acids. Hepatic incorporation of glucose and acetate into fatty acids both in vitro and in vivo was negatively correlated to the content of dietary fat. As little as 2.5% fat effectively depressed liver lipo- genesis rates (145). This relationship existed in the presence of constant carbohydrate intake and irrespective of the degree of satu- ration of the dietary fat (145). Like the liver, in vitro lipogenesis in rat epididymal adi~ pose tissue, as quantitated by differences in CO2 release, was mark- edly lowered by diets containing 48% of calories as fat (146). Adi- pose tissue slices taken from rats that ate a diet with 14% fat 14C-glucose (corn oil, lard or coconut oil) incorporated 50% less into fatty acids than animals fed a 2% corn oil diet (147). The re~ duced rates of hepatic and adipose fatty acid synthesis were paral- leled by similar reductions in lipogenic enzymes (147, 148). A distinction exists between long chain triglycerides and medium chain triglycerides as to their influence on rates of lipo- genesis (147). Rats that consumed diets containing 14% fat mostly as lard, corn oil, or coconut oil all displayed greatly reduced rates of hepatic fatty acid synthesis (147) whereas comparable lev- els of medium chain triglycerides had no depressive action on lipo- genesis in comparison to a low fat basal diet. The inability of medium chain triglycerides to exert an effect on lipogenesis was attributed to the mode of absorption and rapid uptake and oxida- tion by liver (147). 41 Fasting reduces the rate of both liver and adipose tissue fat- ty acid synthesis in a manner similar to feeding diets high in fat (8, 31, 32). A few days of food deprivation greatly lessened the a! mount of 14C-acetate, «glucose or -fructose recovered in hepatic fat- ty acids (149). A 48 hour fast reduced the rate of incorporation of 14C-acetate into liver fatty acids by 90% and lowered immensely the rate of 14C-pyruvate incorporation into adipose tissue fatty acids (8). Fasting for 72 hours nearly abolished liver fatty acid synthe- tase activity, and resulted in glucose-Gaphosphate dehydrogenase and malic enzyme activities which were only one-third and one-half re- spectively that of the fed state (8, 148, 149). The rate of-lipogenesis was rapidly restored by refeeding, and in many instances "overshot" the original rates (8, 111, 113). Within five hours-after initiation of refeeding a fat~free-diet to rats fasted 48 hours, the rate of adipose fatty acid synthesis had increased 20-fold and by 24 hours adipose lipogenesis had plateaued at a rate lOO-fold higher than the fasting rate (8). The extent of the rise upon refeeding depended on the type of diet refed (8). For example refeeding a fat-free, high carbohydrate diet for 40 hours re- sulted in a 90-fold rise over fasting in the in vitro rate of hepatic fatty acid synthesis from acetate (8) while realimentation with a labéchow diet caused only a 15-fold increase in lipogenesis (8). The variation in rates can be explained on the basis of type of carbohy- drate and level of fat in the two diets. Lipogenic enzymes also un- derwent large increases in activity in the liver. However activity 42 changes lagged behind the rise in liver fatty acid synthesis rates by eight hours which suggested the level of enzyme activity at least initially was not limiting (150). The maximal rate of fatty acid synthesis attainable under refeeding conditions may however be re- stricted by the level of enzyme in particular fatty acid synthetase (19). A Training rats to consume their daily food during a two or three hour period each day is a form of fasting and refeeding which precipitates a number of metabolic changes (l43). Meal-trained rats (250-275 9) will consume about 80% as much food as nibbling counter- parts and yet gain weight at the same rate (143). Thus these animals appear to be more efficient in body weight gain and energy gain (151). The explanation for greater efficiency of food utilization under these conditions remains unclear but apparently is not related to reduced basal metabolic rate (143). One major change in meal-fed rats is a hypertrophy of stom- ach and small intestine. Because of the enlarged small intestine, the total absorptive capacity for glucose is markedly increased, but per gram of intestine no difference exists between nibblers and meal- eaters (143).. Clearly certain metabolic shifts must occur to handle the large, rapid influx of nutrients from the gut. Meal-trained rats display a greater ability than nibblers to clear blood glucose (152, 153) following an oral or intraperitoneal glucose load. This im- proved clearance capability is the product of higher circulating 43 insulin levels both in fed and fasted states and of greater tissue sensitivity to insulin (152). A comparison of liver, adipose and muscle tissues indicates that adipose tissue undergoes the greatest adaptive metabolic change in meal-eating rats (143). In the meal-fed rat liver metabolism of glucose does not appear to differ drastically from the nibbling animal (68, 154). The liver of both the meal-eating rat and nibbling rat fasted 22 hours and refed two hours contain about the same amount of glycogen. Synthesis rates of liver glycogen may differ between nibblers and meal-eaters depending upon the initial level of hepatic glycogen (143, 154). Rats adapted to meal-eating do not show changes in hepatic activities of glucokinase, pyruvate kinase, a-glycero- phosphate dehydrogenase or acetyl-COA carboxylase (68) relative to their nibbling counterparts. Following the consumption of~a meal, the rates of liver fatty acid synthesis rapidly rise such that within two hours the rates are five times pre-meal values (33, 34). Although fatty_acid synthesis rates, immediately after a meal, are several fold higher in livers of meal-trained rats than in nibbl- ing animals fasted 22 hours and refed two hours, the rate of lipo- genesis in meal-fed rat liver does not reach that of the fed nib- bler until five hours after meal initiation (111). The rate of hepatic fatty acid synthesis in meal-eating rats gradually re- turns to pre-meal values over the subsequent 22 hour period of food deprivation (33, 34). ”'10 \L’AS fl ' 44 In muscle of meal-trained rats the activity of hexokinase is about 20% above that in nibblers (68). This probably is related to the higher circulating insulin levels found in meal-eaters (152). Consistent with higher hexokinase activity in muscle, was a greater accumulation of glycogen upon realimentation by meal-eaters relative to fasted (22 hours) nibblers refed for two hours (154). Although hexokinase activity was elevated in meal-eating rats, pyruvate kin- ase and a-glycerophosphate dehydrogenase activities were unchanged by meal pattern (68). During adaptation to the meal-eating regimen, the rate of adipose tissue fatty acid synthesis after a meal increased over that found in nibblers. (155). While the rate of lipogenesis was accel- erating, the activity of several lipogenic enzymes was decreasing and reached a minimum at five days (155). This suggests the ini- tial rise in lipogenesis rates in adipose tissue of meal-eating rats is not restricted by level of enzyme machinery. When rats were fully adapted to a meal-eating program, adipose tissue hexo- kinase, and acetyl-COA carboxylase activities were four and ten times greater than in nibbling rats (68). In addition pyruvate kinase, pyruvate carboxylase and a-glycerophosphate dehydrogenase activities were significantly elevated in meal-fed rats (88). In accordance with large increases in glycolytic and lipogenic enq zyme activities in adipose tissue, the rates of fatty acid synthe- sis in adipose tissue both before and after a meal were well above those found in nibbling rats (143, 154). This large rise in 45 lipogenesis in adipose tissue correlates well with better glucose tol- erance and higher circulating insulin levels in meal-eating rats (152, 153), Glycogen stores in adipose greatly increased above those found in nibbling animals after only five days of-adaptation to meal-eating and reached a point in fully adapted animals of being several fold higher (after a meal) than in nibbling rats (143, 154, 155). However prior to the meal glycogen concentrations in adipose were very simi- lar to those found in fed or fasted nibbling rats (154). The Signir ficance of large glycogen reserves may lie in a need for the produc~ tion of a-glycerophOSphate for triglyceride synthesis after comple- tion of the daily eating period (156). The rate of fatty acid synthesis in adipose tissue of meal- trained rats~was shown to be inhibited by dietary fat, but the adin pose tissue may be less sensitive to dietary fat control than in the nibbling rat (157). Increasing the level of dietary fat from 10 to 20% resulted in an 80% decline in the rate of fatty acid synthesis in adipose tissue of nibbling rats, but only a 50% decline in meal~ eating animals (157). A level of 30% dietary fat was needed to de- press the rate of fatty acid synthesis in meal-fed rats to the same level as achieved with 20% fat in nibbling rats (157). In conjunc- tion with reduced rates of fatty acid synthesis were depressed ac- tivities for adipose glucose-6-phosphate dehydrogenase and malic en- zyme (1'57) . Mice, -Like rats. the primary sites of fatty acid synthesis in mice are liver and adipose tissue. These organs also tend to vary 46 the rates of lipogenesis in response to dietary manipulations in a 14C-acetate recovered in fatty manner similar to rats.9 The percent acids of liver slices obtained from mice fed a diet containing 15% corn oil was only 30% that recovered from mice fed a fatefree diet (158). Similarly feeding mice a diet with 15% safflower oil or triolein for five days almost completely abolished in vitro hepatic fatty acid synthesis (159). Simultaneously fat feeding was associ- ated with a highly significant reduction in the activities of he- patic fatty acid synthetase, citrate cleavage enzyme, and glucose-6- phOSphate dehydrogenase (7, 159). The influence of dietary fat on adipose lipogenesis rates in mice is less well defined. The in vitro rate of 14C-pyruvate cone version into mouse adipose tissue fatty acids was not altered by the inclusion of 4% coconut oil in a fat-free diet (7), but the addition of 2% linoleate dramatically dropped the in vitro rate of fatty acid 14C-glucose synthesis (7). By determining the amount of dietary U incorporated into epididymal fat pads of mice, Jansen et al. (160) found little difference in the rate of lipogenesis by mice fed 1 or 5% corn oil; but 10 and 20% dietary fat reduced the conversion ofe glucose to fatty acids by 50%. In comparison to the liver, their data indicated that de novo fatty acid synthesis in adipose tissue of mice was less sensitive to dietary fat inhibition. These experi— ments are subject to criticism because of the long time period (60 minutes) between administration of the 14C-glucose and the removal 47 of fat pads, and for not correcting for differences in the specific radioactivity of the glucose pool (33, 160). Mice that are on an intermittent fasting (24 hours): refeed- ing (48 hours) schedule show no change in body composition relative to ad libitum fed mice (161). However relative rates of fatty acid synthesis in adipose and liver tissue varied considerably. 0n the day of fasting liver Slices had only 6% of the capacity of ad libi- tum fed mice to synthesize fatty acids: upon refeeding this capacity increased to 400% on the first day and 247% on the second day. The adipose tissue on the day of fast had 374% the capacity of the con~ trols and this rose to 927 and 1100% on days one and two of refeeding (161). Hepatic malic enzyme activity remained unchanged on days of fasting and feedings, while adipose malic enzyme significantly ine creased upon refeeding (161).. Swing, The primary site of lipogenesis in the pig is the adi- pose tissue (162). When the levels of dietary corn oil were 1, 4, 7, 10 and 13%, the rate of fatty acid synthesis from glucose and the ace tivities of malic enzyme and citrate cleavage enzyme in pig adipose tissue were depressed in a linear fashion (109, 163). The reduction in rates of lipogenesis was minimal between 1 and 4% dietary fat, while the greatest decline resulted with the increase from 10 to 13% corn oil. Feeding weanling pigs isocaloric diets containing various' amounts of corn oil caused a Significant depression in malic enzyme and citrate cleavage enzyme activities as well as marked reduction 48 in adipose fatty acid synthesis rates (108, 163). In both studies the drop in enzyme activities was less pronounced than the decline in rate of fatty acid synthesis (108, 109, 164). The adipose tissue of pigs responded to fasting (2-7 days) with a depressed rate of fatty acid synthesis in adipose tissue, and a concomittant drop in the activity of glucose-6-phOSphate dehydro- genase, malic enzyme, citrate cleavage enzyme and acetyl-COA carbox- ylase (162). Refeeding a low fat diet restored the in vitro rate of lipogenesis and assOciated enzymes except for the activity of cit— rate cleavage enzyme. In contrast refeeding a high protein diet or high fat diet prevented the restoration of lipogenesis by nearly 50% (162). As noted with feeding fat-containing diets, the response of lipogenic enzymes to fasting and refeeding was less dramatic than the obServed changes in fatty acid synthesis rate. This may suggest that under such conditions initially short-term regulators are more impor- tant in controlling rates of lipogenesis and that enzyme levels are adequate and change in level as a secondary response to diet. Chicken. Unlike the pig, chickens synthesize most of their fatty acids in the liver (164). Some conflict has existed as to whether or not dietary fat lowers hepatic lipogenesis (120, 165). Goodridge (120) fed up to 15% corn oil to growing chicks and found no reduction in the rate of hepatic fatty acid synthesis. However Yeh et al. demonstrated both in vitro and in vivo that growing chicks fed diets containing 10 or 20% corn oil displayed markedly depressed rates of lipogenesis (107). Furthermore feeding a high carbohydrate T‘L‘ _‘, I. . :fi 9‘ 49 diet to chicks previously fed a high fat diet resulted in a four-fold increase_in hepatic malic enzyme activity and a two-fold increase in citrate cleavage enzyme activity plus a slight increase in hexose monophosphate shunt dehydrogenases (166). This further substantiated the adaptive nature of fatty acid synthesis. Young growing chicks adapted to a meal-feeding regimen were unable to consume sufficient food to grow at a rate comparable to nibbling counterparts (167). However meal-feeding older chickens (900 9) did not result in a reduced body weight gain relative to nibbling animals but the meal-eating chickens did eat signifiCantly less food (168). In the meal—trained rat adipose tissue becomes the major site of lipogenesis and this is reflected in elevated lipogenic en- zyme activities and lipogenic capacity (143). However in the meal- trained chicken the activities of hepatic malic enzyme and fatty acid synthetase were not elevated above those in nibbling chicks (143). Just prior to the meal the rate of liver fatty acid synthe- sis was very 1ow and increased 50-fold within one hour after the meal (143). This great change in lipogenesis without an elevation in enzyme activity reflects transient change in substrate availa- bility and indicates the level of enzyme is not a rate determining factor. 50 DE NOVO FAT SYNTHESIS AND TYPE OF DIETARY FAT De novo fatty acid synthesis is inhibited in several species by the addition of fat to a diet (15, 145, 157-160, 163, 166). Evi- dance has been accumulating from rat and mouse studies which indie cates that polyunsaturated fats are more effective than saturated fats in depressing hepatic and adipose lipogenesis (7-10). Reiser and associates found that when 30% of the carbohy- drate in a fat-free diet of rats was replaced with various triglyc- erides, rat liver fatty acid synthesis rates were differentially inhibited by the fat (169). In relation to the fatefree diet, tributyrin, tricaproin and tricaprylin had little effect on the in vivo rate of hepatic fatty acid synthesis while tricaprin, trilaurin, trimyristin and tripalmitin caused a three to five fold decline in the amount of 14C-acetate recovered in fatty acids. Although lard and palmitoyl-diolein suppressed fatty acid synthesis nearly 13— fold, the percent-14C-recovered in fatty acids was still twice that observed with trilinolein or safflower oil containing diets (169). The high linoleate diets resulted in negligible rates of hepatic fatty acid synthesis (169). After five days of supplementing a high carbohydrate with 15% tripalmitin or triolein, the amount of 14C-acetate incorporated by rat liver slices was less (not statistically signifiCant) than slices from a group fed a fat-free diet. However a 15% safflower oil diet reduced the in vitro rate of fat synthesis by 65% (159). Concomittantly the activities of hepatic fatty acid synthetase, 51 citrate cleavage enzyme, malic enzyme, and glucose-6—phosphate dehy- drogenase were consistently lowest in rats fed the safflower oil diet (159). Unlike the rat, 15% tripalmitin, triolein, or safflower oil were all effective in significantly depressing the rate of acetate incorporation into mouse liver slices. However safflower oil invar- iably precipitated the lowest lipogenic enzyme activities (159). Consistent with these effects of safflower oil in rats at relatively high fat intakes were the observations of Wiegand et al. (148) who found a negative linear relationship between the activity of rat liver fatty acid synthetase and the quantity of dietary saf- flower oil (ranges 2.5-15%). In contrast cocoa butter was without effect on fatty acid synthetase activity until a dietary level Of 15% was reached.‘ At this point the reduction in synthetase activ- ity was comparable to the 2.5% safflower oil diet (148). In opposition to the previous cited work, Tepperman and Tepperman reported that 10% dietary corn oil and hydrogenated coco- nut oil were equally effective in depressing in vitro liver fatty acid synthesis (113). In addition when vegetable oil, hydrogenated vegetable oil, lard or corn oil was added to a high carbohydrate diet at a level of 15%, all fat sources precipitated a marked re— duction in rat liver fatty acid synthesis relative to a fatwfree diet. Lard and corn oil appeared to have the most efficient de- pressive action (145). Although the rates of fatty acid synthesis- in these studies were altered equally by saturated or unsaturated dietary fat, saturated fat diets were assOciated with Significantly 52 higher hepatic hexose monophosphate dehydrogenase activities and gen- erally malic enzyme activity (112, 113, 147). The activities of he- patic glucose-6-phOSphate dehydrogenase and malic enzyme have been negatively correlated to the intake of-linoleate and linolenate (112). The reason(s) for maintenance Of high glucose-6~phosphate dehydro- genase activity in the presence of high saturated fat diets has~been attributed to a requirement of NADPH in desaturation and elongation for the synthesis of polyunsaturated acids (170). A recent study had indicated that a high safflower oil- containing diet was more effective in significantly reducing rat liver-fatty acid synthesis than comparable levels of tallow (171). However tallow had a greater inhibiting influence on adipose lipo- genesis (171). In accordance with the apparent site shift in fat synthesis was a large rise in plasma triglyceride levels and a 50% decline in the activities of adipose fatty acid synthetase and acetyl-CoA carboxylase in rats-fed tallow (171); Data for chicks indicated no inhibitory advantage of dietary polyunsaturated or saturated fat on hepatic fatty acid synthesis or associated en~ zymes (171). On the other hand tallow-fed pigs displayed signin ficantly higher rates of adipose fatty acid synthesis and fatty acid synthetase activity than pigs receiving the safflower oil diet which agreed with the effect of tallow on rat adipose tissue (171). There seems to be some confusion at high dietary fat levels as to the specific action of polyunsaturated fat on affecting 53 hepatic fatty acid synthesis. In part these discrepancies may be at- tributed to the very high levels of dietary fat, variations in lipid digestibility (103), and Species variation (171). Considerable evidence in mice and rats fed low levels (3%) of pure methyl esters of fatty acids has been accumulating which sug- gests that only polyunsaturated fatty acids are capable of lowering hepatic lipogenic enzyme activities and rates of fatty acid synthe- sis (7-10). Switching mice from a chow diet to a high carbohydrate, fat-free diet resulted in a greatly enhanced liver fatty acid syn- thetase activity (lO-fold higher at 12 days and 20-fold greater at 25 days). The addition of 4% corn oil or 2% methyl linoleate at 18 days immediately reduced fatty acid synthetase activity such that within three days the activity was only three to five times above values of the animals fed chow diet (7). In contrast the inclusion of 4% hydrogenated coconut oil or 2% palmitate or oleate did not alter hepatic fatty acid synthetase activity (7). Methyl linolenate and arachidonate also possess the ability to Specifically depress hepatic fatty acid synthesis when added to a high carbohydrate, fat-free diet (8-10). The removal of essen~ tial fatty acids from a high sucrose diet resulted in a significant increase in rat liver glucose-6-phosphate dehydrogenase in only four days, and by seven days both glucose-6-phosphate dehydrogenase and fatty acid synthetase were elevated to near maximum (172). The oral administration (100 mg) of methyl linoleate, linolenate or arachi- donate caused a significant drop in both enzymes within two days and 54 within four days all activities were comparable to essential fatty acid adequate control rats (172). Similarly linoleate intubation to rats for three days appeared to be more effective than palmitate, myristate, or oleate in depressing fatty acid synthetase or citrate cleavage enzyme (9). All dietary methyl esters examined except ole- ate reduced the activity of citrate cleavage enzyme, fatty acid syn- thetase and acetyl-COA carboxylase (9). Linolenate and arachidonate had the greatest depressive influence on all hepatic enzyme activi- ties (9, 172). Interestingly palmitoleic acid resulted in enzyme activity patterns identical to dietary methyl linoleate (9). Gozukara et al. (121) proposed that the only mechanism by which polyunsaturated fats altered lipogenesis in rat liver was via a reduction in carbohydrate intake. However when rats were pair- fed a fat-free diet and subsequently intubated with oleate or lino- lenate precipitated a decline in hepatic glucose-6-phosphate dehy- drogenase, fatty acid synthetase, and citrate cleavage enzyme (10). Gozukara et al. (121) based their conclusions only on changes in glucose-6-phosphate dehydrogenase activity which may not always re— flect the response of other lipogenic enzymes (112, 170). The ability of polyunsaturated fatty acids to control adipose tissue lipogenesis of mice and rats is less well defined than in liv- er. At high dietary levels both safflower oil and tallow reduced rat adipose tissue fatty acid synthesis but safflower oil permitted sig- nificantly higher rates of lipogenesis than did tallow (171). A sim- ilar phenomenon was found in swine adipose tissue (171). In contrast 55 mice maintained on a high sucrose plus 2% hydrogenated coconut oil for 18 days had rates of pyruvate incorporation into fatty acids by epididymal fat pad slices which were two to three times greater than mice fed a chow diet (7). Replacing the coconut oil with linoleate immediately led to a sharp decline in lipogenesis which was compar- able to the rats fed chow within two days (7). Oleate also lowered fatty acid synthesis rates but at a much slower rate, while palmi- tate addition had no effect (7). Adipocytes isolated from essential fatty acid deficient rats 14C-glucose per milligram triglyceride incorporated 11 times more than did cells from adequate animals. Adding 5% hydrogenated coco- nut oil to the deficient diet for seven days did not lower the a- f 14C per milligram of triglyceride, but adding 2.3% methyl mount 0 linoleate reduced the rate of lipogenesis to normal (173). Adi- pocytes isolated from rats maintained for five weeks on a diet with 10% hydrogenated coconut oil incorporated six times more glucose into fatty acids (per milligram protein) than did adipocytes from rats fed a 10% safflower oil diet. Adipocytes from rats fed hy- drogenated coconut oil or a fat-free diet for four months still incorporated more glucose into fatty acids per unit weight of lip- id than did adipocytes from safflower oil diet controls. The de- gree of difference between coconut oil and safflower oil groups was much less than at five weeks. When the data were expressed as glucose incorporated per milligram protein, no statistical, differences existed among dietary treatments-in adipocyte 56 lipogenesis rates. The adipocytes isolated from either the rats of fat-free or hydrogenated coconut oil did tend to have higher rates than controls. The younger essential fatty acid deficient rats, regardless of the mode of data expression had a Significantly in- creased rate of lipid synthesis (174). Essential fatty acid defi- cient rats have smaller body weights and presumably smaller fat cells. Therefore an admitted restriction of the previous two stud- ies is that the rate of glucose conversion to fatty acids may dif— fer because of differences in adipocyte size, number and triglyc- eride content (175). Thus an erroneous conclusion may result if data are expressed as glucose converted fatty acids per unit weight of triglyceride. Obviously less triglyceride per cell will result in a high ratio. In comparing rats which differed in body weight by 100 grams, a product of restricted feeding, Hubbard and Matthew (176) discov- ered that the lighter animals had twice the number of adipocytes per 100 milligrams adipose tissue and as much as three times the number of adipocytes per gram triglyceride. Therefore the data of Du and Kruger (173) are subject to erroneous conclusions because expressing 14C-glucose incorporation into fatty acids per unit triglyceride means that each unit of triglyceride from essential fatty acid defi- cient.ratsmay represent three times the number of adipocytes. Ide- ally rates of fatty acid synthesis in adipocytes should be reported on a per cell basis. Recognizing this problem, Demeyer et al. (174) expressed their data both as per unit triglyceride and per unit 57 cellular protein and found no significant differences in older rats among fat-free, hydrogenated coconut oil, or safflower oil treat- ments., However the absolute values still showed fat-free and hydro- genated coconut oil treatments having rates of lipogenesis consid- erably greater than safflower oil-fed rats. Du and Kruger (173) still maintained that the essential fatty acid deficient rats continued to have a sevenfold greater rate of fatty acid synthesis based on pro- tein content of adipocytes; Contrary to in vivo data, fatty acid synthesis in chick hep- atocytes (177), rat hepatocytes (178); and skin fibroblasts (179) is most effectively inhibited by stearate and palmitate and least influ- enced by linoleate and arachidonate.l The degree of inhibition of incorporation of tritium from water into rat hepatocyte fatty .acids elicited by 50011”, fatty acids bound to 1% albumin after 60 minutes followed a pattern of stearate >oleate >1inoleate >palmitate >myristate (178). In chick hepatocytes 500 uM palmitate in 2% albumin had little effect on C14 -acetate incor- poration into fatty acids, whereas stearate inhibited fat synthesis- by 84% (177). Skin fibroblasts exposed for 0.5 minutes to 5 uM al~ bumin bound-fatty acids followed by a 10 minute incubation with 14C-acetate displayed a marked reduction in lipogenesis rate (179). Stearate and palmitate inhibited fatty acid synthesis in these cells by 67 and 48% respectively., Linoleate and arachidonate were less effective, inhibiting synthesis by 26 and 30% respectively (179). 58 The changes in lipogenesis did not correlate with alterations in cit- rate or acid-insoluble COA concentrations. A lack of correlation with these parameters has also been cited in adipose tissue (88). The short time of exposure of the various types of cells to the albumin bound-fatty acids indicates a very Short term type reg- ulation of-carbon flux and_not a reduction in total enzyme concene tration.‘ In support of the cell culture data, Numa and associates (92) reported that purified rat liver acetyl-COA carboxylase has a lower K1 for stearoyl-COA (0.33 uM) than for either palmitoyl-COA (0.91 DM) or oleyl—COA (0.67 pM).~ The marked effectiveness of stearate in prohibiting lipo- genesis was suggested to be the result of a reduced rate of stea- rate utilization or an enhanced rate of linoleate oxidation by the liver (177, 178). Chick hepatocytes reeesterified albumin-bound palmitate to triglyceride several times faster than stearate, and oxidized palmitate to CO2 at twice the rate (177). The conflict between cell cultures and in vivo experiments (7-10) in relation to those fatty acids which are most inhibitory to lipogenesis is perplexing. Such conflicting evidence strengthens the need for finding a mechanistic explanation as to why polyunsat- urated fatty acids appear to Specifically reduce hepatic lipogenesis in essentialfatty acid deprived animals. Furthermore, assessment of polyunsaturates' influence on fatty acid synthesis in essential fatty acid adequate animals becomes imperative. 59 The role of essential fatty acids in affecting lipogenesis in adipose tissue appears to be less well defined and more ambiguous than the data for liver tissue. 'In both organs, particularly adipose, one must be aware that a-nonphysiological state of essential fatty acid deficiency is frequently used as a baseline. Thus results become more difficult to interpret. PARAMETERS RELEVANT TO FAT METABOLISM AFFECTED BY DIETARY POLYUNSATURATED FATS Hepatic fatty_acid ccmposition.- Essential fatty acids are fatty acids which cannot be synthesized by the animal and thus are dietary requirements for maximal growth and prevention of skin der- matitis (11, 180). They are considered to include linoleate (C18;2)’ a-linolenate (C1833) and arachidonate (C2034). Linolenate purport- edly promoted growth in rats when added to a,fat~free;diet.but was' not as effective as linoleate or arachidonate. Linolenate did not prevent dermatitis in rats (180). Maximal growth and minimal dermal scores were obtained with arachidonate (180). However linoleate can be converted via desaturation and elongation to arachidonate (180). Thus linoleate will meet the arachidonate requirement if provided in sufficient amounts (l-2% dietary energy).‘ Rats and mice deprived of essential fatty acids undergo many changes in hepatic fatty acid composition. Most notable is an accum- ulation of eicosatrienoic (5, 8, 11-C20;3), palmitoleic (9'C1631) and 60 and oleic (9'61851) acids (7-9, 181). After only five days of feed- ing mice an essential fatty acid free diet, the linoleate content of the liver fell from 20% to 5% of the total hepatic fatty acids, by 19 days the linoleate content of the liver was less than 1% of all fatty acids. In addition the palmitoleic and oleic acid content of these livers had increased threefold while stearate and arachidonate content fell drastically (7). Comparable, but less dramatic changes occured in mouse adipose tissue (7). As the level of linoleate, linolenate and arachidonate in- creased from 0 to 3.75% of dietary calories, the percentage of he- patic eicosatrienoic acid linearly declined (180). Arachidonate was the most effective acid in this respect (183). Increasing the dietary level of linoleate, linolenate or arachidonate resulted in elevated hepatic levels of the respective acids and their deriv- atives (180). Rats which were previously maintained on a chow diet and then fasted for 48 hours displayed a fourfold reduction in total hepatic fatty acid content and a fivefold drop in total hepatic linoleate content (8). Upon refeeding a high carbohydrate (lino- leate-free) diet, the percent hepatic linoleate declined dramat- ically. Part of this decline can be attributed to dilution by the tremendous increase in total hepatic saturated fatty acids which resulted from refeeding (8, 182). At 10 and 23 hours post-feeding the total hepatic content of linoleate was about 35% and 50% lower than the fasting level and within 48 hours of realimentation with 61 the fat-free diet, hepatic linoleate content was near zero (8). In contrast adipose tissue displayed very little change in fatty acid composition during the 48 hours refeeding period., Bartley and Abraham (159) observed that mice and rats fed diets containing 15% safflower oil had a hepatic linoleate content several times greater than animals fed fat-free diets, or tripal- F—_ mitin or tiolein containing diets. In addition the livers of mice and rats fed tripalmitin, triolein or fat-free diets had elevated amounts of palmitoleic and oleic acids (149).. The increased pro~ portion of palmitoleic and oleic acids is believed to represent de-novo synthesis (8). Depletion of liver linoleate appears to be associated with a large-increase in hepatic fatty acid synthesis when mice and rats are fed a high carbohydrate.diet.‘ On the other hand supplementation of a high carbohydrate diet with pure methyl linoleate returns he- patic fatty acid synthesis and linoleate levels towards normal (7, 10). However this does not prove cause and effect relationship. High hepatic linoleate concentrations in mice and rats fed a 15% safflower oil diet were associated with very low rates of lipogen- esis (159), but 15% tripalmitin and triolein diets also significantly depressed (relative to a fat-free regimen) hepatic rates of fatty acid synthesis in mice while the linoleate and arachidonate levels (per gram liver) were comparable to mice fed a fat-free diet (159). This was not the case in rats where only safflower oil precipitated a significant reduction in hepatic lipogenesis (159). Therefore 62 although high linoleate concentrations in the liver may be affiliated with reduced rates of lipogenesis, this is not always true, (depending upon species) and in fact the two parameters may function independently. Desaturase activity. The elevation in eicosatrienoic (C20:3), palmitoleic (C16:1), and oleic (C18:1) acid concentrations, which ac- company low dietary intakes of essential fatty acids, may be an at- tempt to maintain vital functions (e.g.-membrane structure) requir- ing unsaturated fatty acids (170, 183). The de novo synthesis of these unsaturated fatty acids requires the enzyme, fatty acid desat-' urase, and a fatty acid elongation system (11). Three different desaturase enzymes have been proposed: a) the delta-9 desaturase removes hydrogen from carbons 9, 10 of pal- mitate and stearate to yield palmitoleate (9-C16:])and oleate (9-C18:]0; b) delta-6 desaturase removes hydrogenfrom carbons 6, 7 of oleate (9-C18:1), linoleate (9, 12'C18:2)’ and linolenate (9, 12, 15-C18:3) to form the eicosatrienoate precursor 6, 9’Cl8:2’ the arachidonate precursor 6, 9, 12'c18z3’ and the acid 6, 9, 12, 15'C18z4’ respectively; and c) delta-5 desaturase introduces a double bond be- tween carbons 5, 6 of C20 acids to yield eicosatrienoate and arach- idonate. The substrates of delta-6 desaturase are reported to be competitive with one another but the enzyme has greatest affinity fOr linolenate. .Generally hepatic 6-desaturation exceeds 9-desat- uration (170). Therefore delta-9 desaturase has been proposed to be the rate limiting reaction in in vivo production of eicosatrienoate from oleate (186). The desaturase system is a microsomal oxygenase 63 of liver and adipose tissue which requires NADPH as a coreductant (11), and uses the acyl-CoA derivative not the free acid as substrate. Mammalian and bird desaturases cannot remove hydrogen atoms from the sixth or third carbon atom from the methyl end of fatty acid. Thus animals-fed linoleate-free diets can only synthesize 9-C16:1, 9- C18:1, 6, 9'Cl8:2 and 5, 8, 11’C20:3 polyenoic acids this would ac- count for the abnormal fatty acid composition of liver and adipose tissue. Wahle (187) has examined the differences in activity of delta-9 desaturase among the sheep, rat and chicken. Microsomes . from sheep liver had very low capacity to convert stearate to ole- ate while the chicken had very high delta-9 desaturase capabilities. The rat liver microsomes desaturated stearate at about half the rate of chicken liver microsomes., In contrast rat and chicken adi- pose (perinephric) microsomes had similar desaturation rates, while sheep perinephric adipose tissue had about twice the capacity of either rat or chicken. Sheep subcutaneous fat possessed the great- est desaturase activity. Data from sheep liver and adipose mi- crosomes indicate that the delta-9 desaturase prefers endogenous fatty acids and not exogenous long chain fatty acids (187). The delta-6 and delta-9 desaturase systems apparently are adaptable and vary independently in activity with different dietary reg- imens (1970). Depriving rats of.a chow diet for 24 hours nearly abolished delta-9 desaturation but had little effect on delta-6 desaturation. Refeeding resulted in a marked rise in both 64 desaturase systems with delta—9-having the most dramatic rise (170). Refeeding an all glucose (48 hours) diet induced delta-9 but not delta-6 desaturation; refeeding 95% casein diet stimulated delta-6 desaturation tremendously with no effect on delta-9 desaturation (170). Sheep mammary gland delta-9 desaturase varied with lacta- tion (demands. Inmediately-post-partum the activity was high but dropped precipitously as weaning approached (187). A fat-free diet precipitated a many fold increase in delta-9 desaturation-in mice» and rats (170, 188) relative to a chow diet. ~Concomittant with' 3..— this was a shift in hepatic fatty acid composition in favor of pal- mitoleic and oleic acids as previously discussed. Inclusion of 15% triolein or tristearin partially (50%) prevented the tremendous rise in delta-9 desaturase activity. However 15% safflower oil added to a fat-free diet nearly abolished delta-9 desaturation (188). Utili- zing a 20% fat diet and various combinations ofhydrogenated coconut oil and safflower oil, delta—9 desaturation in rat liver was found to linearly decrease as the dietary content of safflower oil in- creased (170). Thus as the intake of linoleate increases the syn- thesis of eicosatrienoate is lessened because flux through the rate limiting delta-9 desaturation step is reduced (170, 188). With the use of protein synthesis inhibitors Inkpen et al. (170) demonstrated that the adaptive rise in desaturase activity upon refeeding required protein synthesis» Furthermore Oshino and Sato (189) in studies utilizing cycloheximide and actinomycin D haw enz. [p 65 have stated that desaturase activity is controlled by the level of enzyme protein.) Cytochrome b5, which is an integral part of the microsomal desaturase system, was stimulated by stearyl-COA in rats fed a fat- free diet to yield a greater rate of oxidation. Similarly liver microsomes from rats fed safflower oil did not have an accelerated rate of cytochrome b5 oxidation (204). .Since hepatic cytochrome b5 levels between the two treatments were similar, the influence of safflower oil on desaturase activity may partially be related to the lower rate of D5 oxidation (188). Interestingly, among the species studied the primary lipo- genic tissues have the greatest desaturase activity. Also in non- ruminants those dietary and hormonal factors (e.g. insulin) which influence rates of lipogenesis similarly affect delta-9 desatura- ,tion rates (170, 188) which is consistent with Wahle's (187) cona' clusion that desaturase systems preferentially use endogenous fat- ty»acids. Wahle (187) has proposed some type of link between fatty. acid synthetase-and desaturase activities. Most likely such a link reflects the quantity of endogenous substrate for desaturatiOn, although the possibility of desaturase end-product stimulation can- not~be eliminated. Membranes. ~Phospholipids are a fundamental_part of eukaryote organelle membranes (190)-and in essential fatty acid adequate ani- mals the phospholipids contain a large percentage of linoleate and 66 arachidonate (8, 180). In the absence of dietary linoleate, arachi- donate or linolenate; polyenoic acids can be synthesized from oleate and utilized for phospholipid synthesis. If oleate derived polyenoic acids are utilized, the membrane lipid composition shifts to a greater proportion of palmitoleate, oleate, and eicosatrienoate (8). rm: Mice and rats fed a fat-free diet for five days or fasted for i 48 hours and refed a fat-free diet for 48 hours were found to greatly increase the relative amount of hepatic palmitoleate and-oleate while the quantity of linoleate and arachidonate fell to a negligible amount a, (7, 8, 159). The lipid composition of liver nuclei, mitochondria and microsomal membranes displayed a pattern similar to whole tissUe (8). Although less dramatic than in liver, animals fed low essential fat- ty acid diets also have abnormal fatty acid compositions in heart, adipose and total carcass (7). Aberrations in membrane fatty acid composition occur long before visible essential fatty acid deficiency symptoms (8) . As previously discussed, in association with lipid composi- tional changes in rat liver membranes was an enhanced rate of fatty acid synthesis-(8) and an elevated delta-9 desaturase activity (170). Whether the accelerated rate of hepatic fatty acid synthe- sis associated with low linoleate containing diets (7~lO) is a consequence of membrane compositional alterations or simply that the two events are independent phenomenons remains to be ascertained. Long term studies with rats fed diets low in essential fat-' ty acids have indicated that certain membrane enzymes and mitochondrial 67 transport may undergo changes in kinetic behavior (191, 192). Alter- ations in membrane function, transport or enzyme properties (e.g. ATPase) potentially could affect the overall energy metabolism of an animal and contribute to the varying effects dietary fats exert on lipogenesis. Plasma membranes of erythrocytes of rats fed diets varying in content of polyunsaturated fat contained the same amount of cholesterol and total phospholipid but the degree of unsaturation or- in the membrane was lower in those rats fed diets low in essential fatty acids (192). Such a change influences membrane fluidity which Flifi'Am -' t purportedly alters the inetic properties of (Na+, K+)-ATPase (192). In addition, variations in membrane lipid composition have been sug- gested to alter hormone receptor conformation, thereby affecting hor- mone interaction withcell membranes (174). However adipocytes iso- lated from essential fatty acid adequate and deficient rats Showed the same relative increase in the rate of carbon dioxide production and fatty acid synthesis with added insulin (173). Hepatic mitochondria isolated from rats fed a fat-free diet were reported.to be in a swollen state (193). In contrast Haeffner- and Privett (191) found that as the level of arachidonate in a rat diet increased, the rate of mitochondrial anion and cation trans- location was enhanced when measured by swelling agents such as glutathione,;phosphate,-and ammonium chloride under alkali condi- tions. The activities of liver mitochondrial glutamate dehydrogen- ase, B-hydroxybutyrate dehydrogenase, and cytochrome C oxidase were 68 significantly higher in rats-fed archidonate rich diets rather than hydrogenated coconut oil (191). Similarly mitochondrial ATPase ac- tivity in rats fed diets high in polyunsaturated fat was higher than in animals fed comparable diets with saturated fat (191, 194). Plasma membrane (Na+, K+)-ATPase allosteric properties have been reported to be affected by membrane lipid composition and de- gree of saturation (192). Total (Na+, K+)-ATPase activity of rat erythrocytes, heart, kidney and brain microsomes was not altered by membrane fluidity changes but the regulatory mechanisms (e.g. Hill coefficient) were affected (192). The influence these membrane alterations exert on the energy. balance of an animal remain to be ascertained. Since mitochondrial properties have been most consistently altered by diets varying in degree of unsaturated fat (191-194), the function of pyruvate dehy- - drogenase, adenine translocase and/or citrate transport may contri- bute to the control of fatty acid synthesis by various dietary fats. Albumin binding of fatty.acids. -P1asma free fatty acids are generally bound to albumin (195). At physiological concentrations of free fatty acids, this binding involves hydrophobic and electro- .static interactions without gross conformational changes in albumin (195). Albumin possesses various binding sites for free fatty acids and the acids appear to compete for the sites with each acid dif- fering in affinity for albumin (196). The order of binding strength for long chain fatty acids reportedly is oleate >stearate >1inoleate 69 >palmitate >myristate for human albumin (196). In support of this the perfused rat liver was shown to take up linoleate more readily than stearate (197). Hepatic uptake of individual free fatty acids in humans varies with the type of acid (198, 199) such that the order of fractional splanchnic uptake per minute has been reported to be palmitoleate >1inolenate >1inoleate >arachidonate >oleate >palmitate >stearate. The fractional uptake by liver for oleate in male humans determined by isotope infusion was 80% greater for oleate than stearate (200). Similarly the fractional turnover rate for arachidonate in liver was determined to be 50% higher than that of oleate (199). The greater fractional turnover rates in the splanchnic area of human subjects for unsaturated acids agrees well with their af- finity for albumin (196). The physiological significance of polyun- saturated fatty acid uptake by the liver very likely depends on their plasma concentration. However the higher values are consistent with the in vivo inhibition of de novo hepatic fatty acid synthesis in rats and mice (7-10). Although it is tempting to prOpose that polyunsaturated fatty acids (e.g.~1inoleate) can be extracted by the liver more readily than saturated acids and potentially shift the composition of hepat- ic long chain acyl-COA in favor of polyunsaturated acids, not all data agree with the previous results. Perfused rat and dog livers were reported to extract all acids at similar rates (201, 202). Soler-Argilaga et al. (197) proposed that methodology in these studies 70 may not have been adequately sensitive to detect differences in dogS‘ (201) but the conflicting rat data could not be resolved (202). Hagenfeldt and Wahren have reported that much of the high rate of arachidonate fractional turnover could partially be due to exchange of the radioactive label with plasma and endothelial cells (199). Therefore at this point methodological problems make clear interpre- tation difficult. Turnover of polyunsaturated fatty acids from adipose. Unsat- urated fatty acids may not only be extracted by the liver more read- ily but in addition there is evidence accumulating which suggests polyunsaturated fats are mobilized (turnover) more rapidly than sat- urated fats (203). When rats were fed a diet with 30% safflower oil for eight weeks and then fasted for 72 hours, they lost a greater percentage of dry weight and-two to three times more fat (most as linoleate) than animals under comparable conditions fed a lard diet. During the fast, blood glucose levels fell precipitiously and to a much lower level for the safflower oil group than the lard treatment (203). Similarly rats that consumed the safflower.oil diet had the greatest rise in plasma ketone concentrations. These data sug- gest linoleate was mobilized and oxidized more rapidly than satu- rated fatty acids. Rats fed for six weeks either a 20% lard or corn oil diet. had similar weight gains but considerably-different fat pad in vitro 71 lipolysis rates. The corn oil—fed animals had a rate of fat pad lipo- lysis which was 50% greater than the rats cOnsuming lard (204). Contrary to these previous data, DePury and Collins (205) found that essential fatty acid deficient rats possessed nearly twice the hepatic triglyceride content as essential fatty acid adequate rats. In addition the deficient animals had significantly higher a“; serum free fatty acid concentrations. The animals in question had) consumed the deficient diet for several weeks and had visible signs of dermatitis. Their body weights were substantially lower than L. adequate rats. Difficulty arises in examining fat metabolism para~ meters in-animals with such body weight and physiological differ- ences (176). In summary there exists evidence that animals adapted to high linoleate diets have greater rates of lipolysis. In addition lino, leate may be bound with the least affinity to albumin, and extracted by the liver at a greater-rate than saturated fatty acids. Prostaglandins, Prostaglandins (PG) are varied in structure and are synthesized in vivo from essential fatty acids, most notably-' arachidonate (206, 207)., These compounds are purported to have hormone-like actions and/or may be secondary intracellular messengers related in some fashion to cAMP production (208). PGE] and PGE2 are the prostaglandins studied most extensively in liver and adipose tis- sue as possible effectors of carbohydrate and lipid metabolism. The dependence of prostaglandins on essential fatty acids permits speculation as to prostaglandin involvement in fat synthesis 72 as affected by the level of dietary essential fatty acid (205, 10). The picture that emerges as to the influence of—PGE1 and PGE2 on hepatic cAMP and ultimately carbohydrate and lipid metabolism is quite confusing. Lemberg and coworkers (209) found an elevated per- fusate glucose concentration with PGE1 perfusion, which was compar- able to norepinephrine infusion PGE1 and norepinephrine did not have an additive mechanism. Of possible importance was the observation that perfusion per se caused a marked release of glucose which pre- sents the feasibility of cell damage. One minute after intraportal injection into rats of 40 mg of-PGE1 and PGE2, a 60% increase in hepatic cAMP occured (208). The response to PGE1 and PGE2 relative to tissue cAMP level was slower and much less extensive than an injection of 0.025 mg of glucagon. The addition of 20 pM PGE1 to an in vitro hepatic adenylate cyclase assay doubled the rate of cAMP production (208). However, intrapor- tal injection of PGE1 and PGE2 simultaneously with glucagon greatly prevented the rise in liver cAMP after only 15 seconds., Inhibition by PGE1 was dependent on glucagon concentration and appeared to be competitive. Since PGE1 had no inhibitory action on phosphodiester- ase the possibility exists that glucagon and prostaglandins compete for the same binding sites. In addition PGE1 prohibited glucagon's inactivation of glucogen synthetase (208). Of interest would have been insulin plus PGE.l injections and their influence on carbohy- drate metabolism in the liver. 73 Relative to insulin and protaglandins, the intravenous infu- sion of PGE] and PGE2 (10 ug/min) lowered basal serum insulin and greatly depressed insulin response to intravenous glucose treatment. Such action was similar to the mechanism of epinephrine on insulin release, but considerably less extensive (210). The response of in- at: sulin release to prostaglandins has been found to be quite variable and depends on Species, state of animals, etc. (210). One particu- lar problem with prostaglandin injections are changes in blood pres- sure and blood flow. This makes interpretation of long term (minutes) i' studies difficult. DeRubertis and associates (208) suggested that PGE1 and PGE2 may function as feedback regulators on controlling glu- cagon's action on the liver. A similar idea has been proposed for adipose tissue (211). Contrary to the previous reports, infusion Of PGE.l (100 ug/hr) did not significantly change hepatic cAMP during a 60 minute time period. In addition PGE1 perfusion had no effect on perfusate free fatty acid, cAMP or glucose concentrations. PGE1 was also without effect when administered to fasted rats. Simultaneous perfusion of epinephrine and PGE1 following a 15-minute epinephrine treatment had no significant influence on hepatic cAMP levels although there was a trend downward (212).' Levine concluded that if PGE1 influences he- patic lipogenesis or gluconeogenesis, it does so independently of cAMP activity. A rise in hepatic PGE1 and/or PGE2 with increased dietary linoleate could theoretically elevate the liver concentra-' tion of cAMP and negatively influence the action of insulin (208, 212). 74 These possibilities are consistent with the inhibitory role of lino- leate and arachidonate on liver lipogenesis, but do not explain the similar inhibitory action of linolenate which is not a precursor for prostaglandins (7-10). If PGE] and/or PGE2 alter lipogenesis in the liver, the exact role and mechanism has yet to be elucidated. The role of prostaglandins in adipose tissue appears to be more definitive. PGE1 and PGE2 are synthesized in adipose tissue from cis-8,11,14-eicosatrienoic acid and arachidonate respectively (213, 214). Both of these acids are derivatives of linoleic acid. The fat cell PG-synthetase appears to be membrane-associated (214). Therefore Dalton and Hope (213) presented the possibility that mem- brane phOSpholipids were the source of the C20:3 and C20z4 precur- sors of prostaglandins. AS supportive evidence the analyses of fat cell lipids revealed most of the c20:3 and Cé054 acids were asso- ciated with phospholipid with only small amounts found in neutral lipid. These data are expressed as percent of total fatty acid and do not represent absolute amounts. In adipose tissue PGE1 and PGE2 have been proposed to atten- uate hormonally induced cAMP production (211, 213). Elevated adi- pose content of cAMP precipitated by catechOlamines, theOphylline, thyroxine, etc. leads to activation of hormone-sensitive lipase via a phosphorylation mechanism requiring a protein kinase (12). The result is an accelerated rate of lipolysis with free fatty acid and glycerol release into peripheral blood. To prevent exessive rates of lipolysis attenuation of-this system would seem reasonable. 75 Whether or not prostaglandins in vivo Specifically regulate adipose response to various hormonal factors via control of cAMP levels, remains to be conclusively demonstrated. ~Dalton and Hope (215) could barely detect prostaglandins in freshly isolated fat cells of~rats: This was well below the concentration of 140 pM re- quired to prevent theophylline-induced cAMP accumulatiOn (213). If prostaglandins-are realistic regulators then the site of control (adenyl cyclase vs. phOSphodiesterase) becomes a question. Current- ly prostaglandins are thought to exert regulation on adenyl cyclase. However a prostaglandin effect on adenyl cyclase in cell homogenates has not-been-demonstrated (215). Because linoleate and arachidonate are immediate precursors of PGE1 and PGE2 and because these prostaglandins appear to be in- volved in preventing cAMP accumulation, several workers have speca ulated that low dietary intakes of essential fatty acids may alter lipolysis and lipogenesis rates (10, 174, 205, 214). Essential fatty acid deficient rats (12 weeks) had substantially greater serum free fatty acid levels than adequate animals (205).’ Fat pads from deficient rats released much less PGE.I and PGE2 into an incubation media than did rats in an adequate state (214). The basal rate of in vitro glycerol release per gram of fat pad tissue was 50% greater for essential fatty acid deficient rats. Purportedly this was the: product of elevated adipose cAMP due to a lack of PGE1 and/or PGEZ. Adding PGE1 to the media caused a Slight and comparable decline in glycerol release rates for both adequate and inadequate states. 76 Epinephrine inclusion led to a 23% and 50% increase in lipolysis in adequate and deficient animals respectively. In both conditions in-, clusion of PGE1 with epinephrine prevented the stimulus (214). These studies-are complicated by the fact that essential fatty acid defi- cient animals are smaller and have smaller adipocytes. Therefore, per gram of tissue, the smaller rats likely have more adipocytes and greater rates of lipolysis per gram tissue (176). When glycerol re— lease (1ipolysis) from adipocytes was expressed per milligram protein, adequate and inadequate rats showed no significant differences (174). The data were quite variable but based on mean values adequate ani- mals actually had 50% higher rate of glycerol release into media. Fain (211) has found that when lipolysis rates were expressed on a per cell basis, differences between essential fatty acid adequate and inadequate animals did not exist. The role of PGE in attenuating lipolysis and the requirement of linoleate for prostaglandin synthesis does not fit with the obser- vations that rats fed diets containing safflower oil mobilize more lipid and generate more ketones during fasting or upon norepinephrine treatment than rats fed diets with lard (203). However if diets high in linoleate do promote adipose tissue prostaglandin synthesis and reduce rates of lipolysis, this may be in accord with the observations that pigs and rats fed diets high in safflower oil have higher rates. of adipose lipogenesis than animals fed high lard diets (171). Fur- ther work is required to clarify the role of PGE] andPGE2 in he- patic and adipose lipogenesis. PART II DIFFERENTIAL EFFECTS OF DIETARY METHYL ESTERS OF LONG CHAIN SATURATED AND POLYUNSATURATED FATTY ACIDS ON RAT LIVER AND ADIPOSE TISSUE LIPOGENESIS INTRODUCTION Many individuals’with hyperlipoproteinemia appear to respond to an increased prOportion of dietary fat as polyunsaturated fat by displaying reduced blood triglyceride levels (1). The mechanism of polyunsaturated fatty acids still has not been elucidated. However, a specific inhibition Of fatty acid synthesis in human liver, the primary site of de novo fatty acid synthesis, exerted by polunsatu- rated fatty acids would offer a potential explanation for lowered blood triglyceride concentrations. Rats-and mice have been the animal models most commonly used to study the mode of action of dietary fat on liver and adipose tis- sue lipogenesis (7-9, 121, 148, 172). Early work indicated that high fat diets of either predominately saturated or unsaturated fat- ty acid composition could precipitate marked depreSsions in rat and mouse liver and adipose tissue fatty acid synthesis (95, 169). Re- cently a differential response to high polyunsaturated fatty acids or saturated fat diets was reported to occur in rat adipose and liver lipogenesis (171). That is the rates of fatty acid synthesis in the liver were more depressed by polyunsaturated fatty acids while in adipose tissue the tallow diet was associated with lower rates of fatty acid-synthesis. 77 78 Using high fat diets of mixed amounts of saturation presents problems in interpretation because the mode of action of specific fat- ty acids cannot be differentiated. By utilizing low dietary levels of pure~fatty acid esters one can attribute specific mechanisms of action to particular fatty acids. With this approach Allmann and Gibson (7) reported that sup- plementing a fat-free diet with 2% linoleate precipitated a rapid fall in mouse liver fatty acid synthetase activity. However, the inclusion of palmitate or oleate in the diet had no inhibitory in-9 fluence. Rats also respond quickly to small amounts of dietary polyunsaturated fat by displaying significant reductions in liver fatty acid synthetase, acetyl—COA carboxylase, citrate cleavage en- zyme and glucose-6-ph09phate dehydrogenase. Again, palmitate and oleate had no suppressive action (9, 10, 172). The activities of lipogenic enzymes do not always reflect rates of fatty acid synthesis of a tissue (113). Yet little attempt has been made to correlate changes in liver lipogenic enzyme activié ties precipitated by polyunsaturated fatty acids to rates of hepatic fatty acid synthesis. In mice a small amount of dietary linoleate did lead to a decline in fatty acid synthesis in a liver supernatant preparation (7), whereas similar amounts of oleate and palmitate had no depressive effect. Similar comparisons with various fatty acids at low dietary intakes have not been made in rats. Even though adipose tissue represents 50-70% of the total body de novo fatty acid synthesis in rats (154), its response to 79 various fatty acid methyl esters has not been significantly investi- gated. Supplementing a fat-free diet with linoleate or oleate waS' associated with a reduction in the in vitro rate of adipose tissue lipogenesis of mice. While linoleate exhibited the greatest inhibi- tory effect, palmitate had no depressive action (7). Only isolated adipocyte data exists for the effect of low levels of specific die- tary fatty acids on lipogenesis rates in rat adipose tissue...A1- though there is some suggestion of an inhibitory effect of linoleate,v the data are‘susceptible to a problem of fat cell size differences be- tween essential fatty acid adequate and deficient rats(l73, 174, 176). Studies comparing methyl esters of saturated and unsaturated acids as well as the respective triglycerides have unfortunately over- looked differences in digestibility as a factor in explaining differ- ential effects on lipogenesis (7-10). Additionally, cell culture and isolated hepatocyte (177-179) data indicate stearate is the most po- tent inhibitor of fatty acid synthesis and yet no in vivo data is available for this acid.. Since little infOrmation is-available on rat liver and adipose tissue fatty acid synthesis rates, we have quantitated both in vitro and in vivo rates of fatty acid synthesis in rat liver and adipose tissue and at the same time examined the activities of various lipogenic enzymes as affected by palmitate, stearate, oleate, linoleate-and linolenate. Since variations in absorption may be.a factor in explaining responses to various fat- tay acids, the apparent digestibilities of all methyl esters were determined.- 80 METHODS General animal handling, ‘A high carbohydrate: fat-free basal diet (FF-diet) (Table l) was supplemented with methyl esters of pal- mitate (C16:0)’ stearate (€18zo)’ oleate (C18:1)’ linoleate (C18:2) or linolenate (Cl8i3)‘ In eXperiments 1-3 esters of 99% purity were added as 3% of the daily FF-diet consumption. In experiment 4, C16:0 was increased to 7% of the daily FF-diet intake in order to compensate for its poor digestibility. To avoid rancidity the esters were mixed into the diet daily. The animals were housed in individual stainless steel cages and rats had free access to water. Eight male Sprague-Dawley rats (1009) per treatment were adapted to a three-hour per day meal-eating regimen (900-1200 hours). This protocol was adopted to facilitate control of carbohydrate in- take among treatments. During the adaptation phase all rats received the basal diet plus 2-3% safflower oil. This minimized the unphysio—‘ logical state of essential fatty acid deprivation. Following adapta- tion to meal-feeding (8-10 days), all animals were meal-fed the FF- diet for seven days. On the eighth day animals were matched for body weight and food intake and assigned to blocks adopting the random- ized complete block design described by Steel and Torrie (217). Each block contained three or four animals depending on the number of treatments in the experiment and each experiment utilized eight blocks of-animals. Each block of animals was allotted 85% of the average amount of fat-free diet consumed by those animals during the previous 81 TABLE 1 Fat free basal diet composition —: ~ —_ —_ Ingredient Parts Carbohydrate1 72.0 Casein 20.0 Nonnutritive fiber2 3.0 D, L-methionine 0.3 Choline chloride 0.3 Vitamin mix3 0.4 Mineral mix4 4.0 100.0 1Glucose was utilized except in experiment 2 in which sucrose replaced glucose. 2Solka-floc. Brown Company, Berlin, New Hampshire. 3Vitamin mixture was that de- scribed by Yeh & Leveille (216). 4Rat mineral mix #4164. Teklad Test Diets, 2826 Latham Drive, Madison, Wisconsin.. 82 seven days. Increases in food allotted were permitted when all ani- mals in each block completed all the diet in the three hour period. Controlling carbohydrate consumption would hopefully eliminate the criticism that dietary fat depresses lipogenesis by reducing carbo- hydrate intake (121). After seven days of supplementation of the FF-diet with the respective esters, all animals were killed one hour following their last meal and liver and adipose tissue removed. Fecallipid extraction. During the seven day period of es- ter supplementation the animals were transferred to metabolic cages and feces were collected in order to determine the apparent digesti- bility of the methyl esters. Although this procedure does not per~ mit exact calculation of the amount of absorption of each specific acid, it does permit an estimate of the apparent degree of digestion of the respective esters. Duplicate 1.0g samples of ground, dried feces were suspended in 10.0 ml 1.0 N HCl and extracted with two 15 ml volumes of chloroformzmethanol (2:1). The chloroform phase was removed and dried in pre-weighed aluminum pans. The quantity of lipid was determined gravimetrically. Engyme assays. .Immediately upon removal, liver and adipose tissue were homogenized in cold KCl (0.15 M), MgCl2 (1.0 mM) and n-acetyl-cysteine (10 mM) buffer, pH 7.6. Following centrifugation at 100,000 x g for 40 minutes, the supernatant was used for quantir tation of enzyme activities. Fatty acid synthetase activity was deternfined by following the rate of NADPH oxidation (218). Glucose- 6-phosphate dehydrogenase (EC 1.1.1.49) and NADP-malic enzyme 83 (EC 1.1.1.40) activities were quantitated by the rate of NADP reduc- tion (107, 219). Protein content of the supernatant fraction was quantitated by the method of Lowry et al. (220). In vitro fatty_acid synthesis. The rate of fatty acid synthe- sis in experiments 1 and 2 was determined by incubating 100-200 mg liver slices and pieces of adipose tissue in 3.0 ml Krebs-Ringer buf- fer (37°) containing 0.10 units porcine insulin per ml and lOOmM glu- 14 cose for liver and 10mM glucose for adipose tissue (154). U- C glu- cose was added at a concentration of 0.1 pCi per ml buffer. In ex- periment 3 fatty acid synthesis was quantitated in liver slices and adipose tissue pieces using a double labelled design. 3H20 incor- poration avoids possible differences in Specific activity of fatty acid precursor pools which may result from the dietary treatments. Therefore, liver slices were incubated in Krebs-Ringer buffer (37°) containing 50 pCi 3H20 and 0.03 pCi U-14C glucose per ml buffer. Adi- pose tissue was incubated in Krebs-Ringer buffer (37°) containing 50 uCi 3H20 and 0.01 uCi u-‘4C glucose. After two hours the tissue Slices were removed and saponified. Following extraction of nonsaponifiable compounds with 3-5 ml wash- ings of petroleum ether, the alcoholic-KOH phase was acidified with HCl and the fatty acids extracted with 3-5 ml washings of petroleum ether. The extracted fatty acids were counted in scintillation fluid. In vivo fattyLacid synthesis.--In experiment 4 the in vivo rate of fatty acid synthesis was ascertained by determining the amount of 3H20 incorporated into liver and adipose tissue fatty acids. Each 84 rat was injected intraperitoneally with 1.5 mCi 3 H20 in 0.5 ml phys- iological saline. The animals were killed 10 minutes post-injection. Following killing, liver and adipose tissue were rapidly removed and weighed. Epldidymal fat samples (200-300 mg) were deposited directly into 30% KOH for saponification. Livers were homogenized in an equal volume of water and 0.5 ml aliquots were removed for saponification. Following extraction of fatty acids, the amount of 3 H‘in fatty acids- was quantitated by liquid scintillation counting. The scintillation fluid contained 4.09 scintillant dissolved in 230 ml absolute ethanol L...» and toluene to one liter. Plasam free fatty acids were extracted and quantitated according to the procedure described by K0 and Royer (221). Statistics. aAll data were statistically evaluated by means of analysis of variance for randomized complete-block design. Treatment differences were ascertained using Tukey's t-test procedure (217). RESULTS Experiment 1.1 After three days of supplementing C18:3 to the basal diet, food consumption became depressed. This adverse effect on appetite was attributed to rapid lipid peroxidation of residual ester in food cups which reduced diet palitability.- Precautions were taken to minimize rancidity and subsequently food intake rapidly im- proved. Additon of C or Cl8:3 to the fat-free diet at a level of 18:0 3% of the daily food intake had no influence on liver or epididymal (I) 85 fat pad weights nor on total weight gain (Table 2). In comparison to rats pair-fed the FF-diet or FF+3% C18:0 diet, dietary C18z3 precip- itated a significant drop in hepatic fatty acid synthetase and glucose- 6-phosphate dehydrogenase activities while having no effect on malic enzyme activity. In contrast to C18'3’ supplementation with Cl8-O actually elevated hepatic fatty acid synthetase, glucose-6-phosphate “‘7 dehydrogenase and malic enzyme activities over those observed in rats. fed the FF-diet (Table 2). In association with changes in hepatic lipogenic enzyme activities rats fed Cl8z3 had a tremendous reduction . 1*“ in U-14C glucose incorporation into fatty acids by liver slices. Al- though these data substantiate earlier conclusions (7-9) that polyun- saturated fatty acids specifically inhibit hepatic fatty acid synthe- sis, interpretations become less conclusive after consideration is given to the very poor digestibility of Cl8:0 (35%) relative to C18z3 (89%) (Table 2).- Unlike the liver, lipogenic enzyme activities and the rates of fatty acid synthesis in rat epididymal adipose tissue were unaf- fected by dietary methyl ester supplementation (Table 2). Experiment 2. Rats fed the FF or FF+C16:0 diets differed very little in final body weights, weight gain, liver weights or epi- didymal fat pad weight (Table 3). However, rats fed the PRC-18:2 diet tended to have greater weight gains and heavier fat pads. After considering the great difference in apparent digestibility (Table 3) between C16°O (40%) and C 3 (87%), these parameters seem to be 18: 86 TABLE 2 Methyl stearate vs. linolenate - Influence on lipogenesis and lipogenic enzymes (Experiment 1) m Treatment 5 '2’! Parameter Basal +Cl8z3 +C18zO Body wgt., g 169 172 171 Wgt. gain, 9 13 15 15 Daily food intake, 9 11.8 11.5 11.8 Ester digestibility, % --- 89 35 Liver wgt., g 6.7 :_0.2 6.7 :_0.3 6.9 :_O.2 Epididymal fat wgt., g 1.1 :.0.1 1.1 1 0.1 1.1 :_0.1 r-— Enzyme Activities2 Liver: FAS3 11111b 10.6a 1511C 4 b a c G6PD 97 112 37 1 3 121 11 ME4 15 112b 14 1 53 21 1 5b Adipose: FAS3 38 1 5a 39 1 5a 42 1 4a cm“ 135 110a 109 11761 118 116a ME4 147 1 24"il 168 1 236‘ 191 119a Fatty Acid Synthesis5 Livers 300 1 16b 151 1 24‘“ 381 1 42b Adipose6 3071 1 467“1 2765 1 4813 4052 1 947a 1Those values with different superscript letters are significantly different (P < 0.05). 2 Mean i_SEM n=4. 3 1 Nanomoles NADPH oxidized min' mg'1 protein at 37°. FAS = fatty acid synthetase. ‘ l l Nanomoles NADP reduced min“ mg" protein at 25°. G6PD = glucose- 6-phosphate dehydrogenase, ME = malic enzyme. 5Mean 1 SEN n=8. 6Nanomoles U-14C-glucose incorporated into fatty acids per 100 mg wet tissue per 2 hrs. at 37°. 4 1'11 87 TABLE 3 Methyl palmitate vs. linoleate - Influence on lipogenesis and lipogenic enzymes (Experiment 2) m Treatment Parameter Basal +018;2 +216:0 Body wgt., g 200 1 5 208 1’. 5 199 1 4 Wgt. gain, 9 18 + 1 26 + 2 19 + 1 Daily f00d intake, 9 13.2 13.2 13.2" Ester digestibility, % ~-- 87 4O Liver wgt., g 9.3 :_O.2 9.0 1_0.2 9.4 i 0.2 Epididymal fat wgt., g 1.3 1.0.1 1.5 :_0.1 1.3 :_0.1 Enzyme Activities2 Liver: FAS3 30 1 6‘” 18 1 4al 24 1 63 cm“ 107 112b 67 1 5a 123 114b NE4 53 1 8b 32 1 4a 57 1 7b Adipose: FAS3 41 1 7a 41 1 6a 40 1 4al cm“ 209 1 41a 182 1 34‘“ 190 1 39al ME4 218 1 293 197 1143 216 1 20a Fatty Acid synthesis2 Livers 159 1 29a 121 1 23a 131 114al . Adiposes 2081 1 209al 1876 1 1603 2132 1 244a 1Values with different superscript letters are significantly dif- ferent (P < 0.05). 2Mean 1 SEM n=8. 3Nanomoles NADPH oxidized min'1 mg' acid synthetase. 4Nanomoles NADP reduced min"1 mg"1 protein at 25°. G6PD a glucose-G-phOSphate dehydrogenase, ME = malic enzyme. 5Nanomoles U-14C-glucose incorporated into fatty acids per 100 mg wet tissue per 2 hrs. at 37°.' 1 protein at 37°. FAS = fatty fa 88 reasonable products of differences in energy intake and/or essential fatty acid status. In this experiment sucrose replaced glucose as the source of carbohydrate. Therefore, the activities of the hepatic lipogenic enzymes are higher than in experiment 1 (123). Even though rats fed the FF+C18:2 diet had hepatic fatty acid synthetase activity 40% be- low that of the basal group, neither dietary Cl8:2 nor 016:0 signif- cantly altered liver fatty acid synthetase activity (Table 3) accord- ing to Tukey's t-test analysis (217). Hepatic activities of glucose- 6-phosphate dehydrogenase and malic enzyme were significantly de- pressed when the FF+C]8:2 diet was fed but not when the FF-diet was supplemented with Cl6:0 (Table 3). Rates of in vitro hepatic fatty acid synthesis were Slightly lower in animals fed methyl esters (Table 3), but like fatty acid synthetase activity these differences were not Significantly reduced by dietary C Like 618-0’ methyl 18:2' C16°0 was very poorly digested (40%) and may partially explain the lack of influence on the hepatic activities of lipogenic enzymes. As in experiment 1, 018.2 or C supplementation to a fat- 16:0 free diet had no influence on adipose fatty acid synthetase, glucose- 6-phosphate dehydrogenase or malic enzyme activities. Similarly, in vitro U-14C glucose incorporation into adipose tissue fatty acids was not impaired by dietary esters of C18:2 or C16:0 (Table 3). Experiment 3. In order to investigate the digestibility of of C18,] and its role in controlling lipogenesis, and to re-examine the effect of C18'2 on fatty acid synthesis, these esters were 89 TABLE 4 Methyl oleate vs. linoleate - Influence on lipogenesis and lipogenic enzymes (Experiment 3) Treatments1 Basal +°l8z2 +C18:1 Body wgt., g 191 :_5 198 :.4 196 :_5 Wgt. gain, g 28 :_2 34 :_0.8 30 :.2 Daily food intake, g 13.6 13.8 13.6 Ester digestibility, % --- 62 87 b 88 b Liver wgt., g 7.4 1 0.2 7.6 10.2 7.4 10.2 Epididymal fat wgt., g 1.8 :_0.1 2.1 :_0.1 2.0 :_0.1 Plasma FFA, ueq/L 541 :_44 552 :_38 580 :_56 Enzyme Activities Liver: FAS3 16 12c 1011b 1411c G6PD4 142 1 15c 86 1 7b 142 1 8c ME4 36 1 2c 25 1 3b 37 1 3C Adipose: FAS3 32 1 2bb 31 1 3bb 33 1 3b G6PD4 140 113 136 111 124 1 5b ME4 221 113b 221 110b 209 1 7b In vitro Fatty Acid Synthesis Liver: b b u-l4C-gTucose5 183 1 22C 135 117 159 119 ’C 3H206 1642 1 180c 1375 1136b 1539 1 174m Adipose: b b b u-l4C-g1ucose5 1808 1 138 1821 1 112 1600 1 148 3H206 11,584 1 1032b 12,278 1 933b 9877 1 863b 1Mean :_SEM, n=8. 2Values with different superscript letters are significantly different (P < 0.05). 3 Nanomoles NADPH oxidized min'1 mg"1 protein at 37°. FAS = fatty acid synthetase. 1 Nanomoles NADP reduced min'1 mg" protein at 25°. G6PD = glucose-6- phosphate dehydrogenase, ME = malic enzyme. Nanomoles U-14C—glucose incorporated into fatty acids per 100mg wet tissue per 2 hrs. at 37°. Dmg3H-incorporated into fatty acids per 100mg wet tissue per 2 hrs at 37 . 90 supplemented for seven days to a fat-free basal diet in a manner sim- ilar to experiments 1 and 2. Digestibility of C18:l was comparable to 018:2 (Table 4). In addition °l8:2 and °18:l supplementation resulted in similar epididymal fat pad weights which were slightly heavier than in basal rats. Weight gain was the greatest for rats fed °l8:2 (Table 4). Dietary methyl ester supplementation did not significantly alter liver weights (Table 4). Both dietary C18zl and °l8:2 lowered the activity of hepatic fatty acid synthetase but only the effect of Cl8:2 attained statisti- cal significance (Table 4). As observed in experiments 1 and 2, only the essential fatty acid, °18:2’ significantly depressed glucose-6- phosphate dehydrogenase or malic enzyme activity in the liver. Unlike experiment 2, dietary~C18z2 significantly impaired the rate of u-‘4C glucose and 3H20 incorporation into hepatic fatty acids. Supplementa- tion of the basal diet with °18:1 yielded intermediate rates of fatty acid synthesis by liver slices based on either U-14C glucose or 3H20 incorporation (Table 4). Both U-14 3 C glucose and H20 produced com- parable results indicating that the fatty acid precursor specific ac- tivity in the liver slices was not influenced by dietary treatment. AS in previous experiments, rates of fatty acid synthesis and activities of lipogenic enzymes in epididymal adipose tissue were not influenced by the source of dietary fatty acid (Table 4). An elevation in plasma free fatty acids(FFA) may be associ- ated with a reduced rate of hepatic lipogenesis (110). Therefore, blood samples were obtained at the time of sacrifice in order to 91 determine if differences in plasma free fatty acid concentrations could potentially explain the marked inhibition of °l8:2 on hepatic lipogenesis. However, plasma FFA did not differ in concentration among the treatments (Table 4). The lack of difference in total concentration of the plasma FFA does not necessarily preclude the possibility of an increase in concentration of liver tissue long chain free fatty acids or their COA derivatives (129). Experiment 4. The first three experiments indicated two points which required further clarification: a) the lack of influ- ence of °16:0 on hepatic,1ipogenesis might be partially explained by its poor digestibility (Table 3); (b) C appeared to inhibit 18:3 hepatic fatty acid synthesis and associated enzymes more effectively than °18'2’ however these esters were not compared within the same experiment. Therefore, experiment 4 was conducted to determine if, after correction for low digestibility, dietary C could impair 16:0 hepatic lipogenesis. In additon dietary °18:2 and °18:3 were com- pared for influence on hepatic fatty acid synthesis rate. The amount of each ester absorbed daily was not significantly different among fatty acid types (Table 5). Weight gain did not dif- fer significantly among dietary treatments, but the rats fed fatty acids tended to gain slightly more weight than the control animals (Table 5)which was in accord with the higher energy intakes and/or adequate essential fatty acid status of these animals. Clearly the inclusion of 3% °18:2 or °l8'3 in a fat-free diet markedly and sig- nificantly depressed the activities of hepatic fatty acid synthetase, 92 .mmuzcws oP cmq mammwu Eecm cma mkum xupme one? umpecoacoucv sawumcu anm _ .msx~cm ovpms u m: .mmecmmocuacmu muegamocaimummouspm u camw .omw we :wmuoca as ewe uwusumc moppmw u m mmogpm - .Nue 5% H :85 EN H 25.2 15mm 8; are H eNN. : eNmNm H 23: _ meeeflz numm + eoN._ npmm + mem.F oump.u mnm.e ummmtu m_m.¢ mcm>v4 ”memmspczm upon zuuem o>w> :H eNm N am m om on m we em m Nm em: 1H 3 no H we .63 H N: on H N2 «88 np,+ o n_ + m o__+ ~_ mop + mp m mwgue mstcm owpmnmz ~.o.H m.~ ~.o.H P.“ m.o.H m.n P.o.H n.m m ..pm3 cm>w4 e H 3N N H 8N e H ON... .1. me 68636 2.38 58 N.o H 3: 2 H 92 N5 H 9: N.o H 2.: a - 8.8.5 88 58 N H 2 N H 2 PH ON PH 2 a .58 .32 N H o: m H E N H N: N H e: a :3: been :2: ,mnmpo am+ Numpu Nm+ oquQ N~+ meem Fmpcwsummch Ac pcoepcwnxuv mmez~=o owcmmoawr orange; new mvmmzpcam upon sauce :6 mpmcmpocpp wee mpempocwp .mueuwspea pxsums co mucospmcm m m4mWF m sea mceue xuuee cps? nmwmcoacoocw : Emu ”mu: .2mm.H new: m .mex~:m cepms u use .mmecmmocuxgmn muenamozaioimmouapm u meow .omm we Flewmuoca me Picws vacant; ea mmozhm .pmwu mmcmuuee e we :owumucmEmFaaam memm 2? AmNNV.Fm um excepo x; cwcvecmumo .mn: .2mm.H cams use mmwuw>wpom msxucm use mcmumsmcma spzocww eoem m seem emNmP m Nme.oF eNNm m mm_.op e__e_ m PNNN ememeeeeae-__ pm; co pro cmzcwmmmm am.mv pawn mumzamvm neon xgpee Fewucmmmm op cowueucmsmpaaam u-—p co .mpu .o"o_u mo pummwu N m4m

5F m sea muwue zupee ops? empecoacoucw 2 San m 1:55 emuzumc ao mmochm .8": .zum.H eeezN .Nue .zmm.H eeez .mmmcmmocuxzmu muecamosaimimmouapm u anew .omm um picwmuoca as P 1:55 umNVowxo zao5uu< oszucm eN.o.H N.N e.em.o H e.N e.em.c.H _.N eN.o H m.e 8 ..883 cases 8.8 N.8 8.8 e.e 8 .exeee5 eeec 55588 eeN m N am N 8% m N 1m NN 8 .58 .83 :38 em + em. eN+ Fe. e.eN + emp 8N + omp 8 ..883 seem Nm N_e am+ NN N50 xm+ 55 N58 um+ .__emee e56< Nepal seesaw: pro cmzopemem upv pope mumacmve cmwm anew; mwr commw so use mums =5,m5mm=mmoawp sm>5~ co mucmempaqsm -.m—u use ..m u .P.mpu co uummcm m u4m wmoshm .Nue .2mm.H ewe: m muegamocaimimmoosz u meow ..omm um Picwmaoca me .mmepmzucxm neon acute u mFm>wpumammc N use F ucmswcmaxu :5 Pro cmzopwmem NF use um.~ umcweucou pm_u Pememp ememp.w mmm.m -1- me.H cm am.H mm_ mumpu+ emmmN.H omN m -1- mm.” mm am.” Pop ~.wpo+ enwm~.H eon.m -1- mm.H em a.e~.H omp Fumpu+ meoom + Nom.~— 11. mm + mm mm + om, meem "N acmewcmaxm eaim FNN.NN a: m 8N, 1m om eNm E N859. eomxm.w mo¢.mm ee_.H esp em.” om em.” Fmp anwu+ emmqm.w Num.m~ 68F.H wmp 85.” Pm em.H Pmp onoFu+ mooom + mmm.mp esp + mmp em + mm emu + va Femem up ucoswcquu em_meeeesm-FuceoFFchFm 8cm “cmchmgxm :8 :ngF3 mpaFcomcmqam ucmcmFFFu ssz mw=Fe> omosF v m .N pcmchmaxm :F on: 8:8 F pcoEchaxm cF Fuc mews: sum.“ cmmzN .me>Fpomammc N use F pcwEFcuaxm :F FFo cmZoFFFem NF use nm.~ emcFeucou umFu Femem F I I I I cm F893 9881’: eOONF + omNN 88888 + N88.N_ 88N8N + 888.NF 88888 + 8N8.8F eses :8 8888588 I I I I 88 3658884... eNmN + N888 eNNN + N8N8 e888_ + 8888 8F58F + FNNN e>5> =8 858888: I- I- 888 H N5 eNN H 88 8e :88; 85 2.8553 88.8 H E eN8 H 8.8 eN8 H 8.8 eN8 H 8.8 8 :8: ts: 8.8 8.8 NS 8.2 8 .335 85 8:8 1 H 2 1H NN & H 2 65 H N 8 .58 .83 :38 88H of 8N H 82 88 H 83 8N H 8: 8 :88: 888 NmumFu Nm+.r . -meem iNmumFo Nm+ TFemmm FNucmchmaxm Fm acmchmaxm TIT mums 88838868 uFuezuumF Fmecmmmm cF mmecmmanF mmanne can uFueqm; co manu 8chFepcou Fame mco Fo uuwcmm m m4m euepaoepaoeee ease eoaneoFeu mmeem>< .Lmumsoeea xcm co pumemm acoEpemep “seawmwcmwm oz .opnc “sum + meow: p igm>__ m .oom no Ficmmuoca as 1.... En Ficwe o>w> cw copesoaeoocw uep go In to am e m .»_:o pave Fumes mmeeiuemm —i:ws awesome ¢oe> :H Nu“ NN N.“ «N _.H mm N.“ ew eoaeeexooe_u Nqnee WOMEN Nonmg Nenmg era: nea>w4 FIXGU GE _em mam Pam ---- .meoeeoooa ewoa enema I l .1 Lane as m_ + woe mm + ooe mm + ape, ---- .oxaoe_ weoe enema 8.0 + m.m_ 5.0 + e.mp 8.0 + o.e_ e.o + m.~P a .Noxaoea eooe »_eae womeo Rea «wepo em+ d e_o em+ . ea toooEaeaa _eao< eased xgeooeo AN ucoswswaxmv mammsucxm uwoe sage; em>pp an; ac muecmpoc_F .eo «peopocwp .muogempm newcvmucoo amen omseiuew to mpems oz» co pummmm e m4m mmoiN .Figm>wp m picws muwue xuuem one? o>w> cw umpecoacoucw oep co :m we Ease .on we Ficwmpoga as FicE umeUPxo zma< .opuc “sum.“ cmwzF ammm.u wem_ aeom.u Moe. amen.“ meow aeNN.H a_mp oeeoooa-ua_ -_ eon.“ mme mew.“ om“ ewe.“ eeofi ewe.“ eeop omzm ”omrmmsucxmiw> :H am a new m em “EN m 3 our“. 5 momflofiim 23 .58 P + om N + mm P + KN N + mm eomaeexoo=_e mdnmg Noumg fio+9m Naneg 0.3% ugw>w4 mam cam _mm --- awe can we .meoeeomee eeoe stoma em.“ emm NF.“ mme we.“ mmoe --- aae see me .oxaeea e_oe steam em.o.u m.~_ eo.o.u ~.e_ am.o.u F.m_ ace.o.u e.m_ a .Noxeeee eooe xeweo mumru flm+ Nuwru Nm+ oumpu flw+ mu memEmme Pewo< stead heaters Am pcoewemaxmv .mvmoguczm twee apnea su>w~ um; co muecmpoepp so .mpempocwp .maegomum unwaveucou newt owcmiuew.mo mpeme mecca eo uowmem m m4m mmonp ._icvmpoca as m m news new: newcaimu mmFosocmze news o>w> cw muwue xuuem o“ voucm>cou muwcsimu mmposocez .omm pm —i:wmpoea as “com umpicwmuoca me p F icws tomato; ¢o_4 «N H «2 2 H we 2 H .3; 1. an to. me .935 Boa been. 3 H 3: as H 3; .2 H e: 2 H 3: a £8.35 .88 3.25 aw H mm eNH 8 am H 8 mm H 2 . a .58 .3: :38 S H SN e H 8N a H mom a H 2: a :3: as :2: N ape m oa-F ae-~ ca-m eoeoEeeaa . 3:8 A 3:: muse am we 11 1 TTP : m P z fie acmeremaxuv . cowuoucwempmnzm mnempocvp to mpems cm>mm new me Lance mwmmsucam upon zuuem em>_P we; so pave vaccine» mo mpems o3» use use we mucmschu n u4m