(55}; .7'88 5‘3 SUBSTRATE UTILIZATION DURING EXERCISE Muscular Efficiency When Fat is the Major Source of Energy By Kristian Balwin A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Food Science and Human Nutrition 1977 fie: ~19 n ;‘ ¥t .s A .s,\ x. ABSTRACT SUBSTRATE UTILIZATION DURING EXERCISE Muscular Efficiency When Fat is the Major Source of Energy By Kristian Balwin Previous investigators reported that muscular work efficiency decreases when fat is the major energy substrate. The decreased efficiency could result from dehydration caused by ketosis, or from glycogen depletion when high fat diets are fed. The purpose of this study was to investigate whether dehydration was a cause of impaired muscular work efficiency. Dehydration was found to be related to protein rather than fat intake. A low protein diet caused dehydration; but this was not observed when a higher percent of protein was fed, regardless of the nature of the protein and fat. Two series of exercise experiments were performed, and despite a similar ketosuria, a decreased work efficiency was found only in the latter series, in which the carbo- hydrate intake was lower and the duration of the experimental periods longer. The more severe carbohydrate restriction presumably can cause glycogen depletion, which may be responsible for the decreased efficiency. ACKNOWLEDGMENTS The author wishes to express his appreciation and gratitude to the following: Dr. Olaf Mickelsen for his excellent guidance and fine leadership Dr. W. Van Huss for his assistance and outstanding willingness to be helpful The subjects, William D. Hart and Rong-Ching Hsieh for enduring the diets Dr. Daniel Rosenfield, Worthington Foods for donating vegetarian meat substitutes Mrs. Thelma B. Peircy and the California Avocado Society for furnishing the avocadoes Ms. Cheri Lutz for her help in typing and proofreading the manuscript ii TABLE OF CONTENTS Page INTRODUCTION. 0 O O O O O O O O O O O O O O O O O O O O O O O O O 0 O O O O O O O O O O O O O O O I O O O O O O O O 1 REVIEW OF LITERATURE O I O I O O O O O O O O O O O O ..... I ..... O O O O O O O O O O O O O O O O O 3 PART ONE: SUBSTRATE UTILIZATION I. EXERCISE AND SUBSTRATE UTILIZATION......................... w A. ReStOOOOOOOOOOOOOOOOCOOOOICOOOOOOOOOOOOO0.0.00.0...O. B. Exercise of Short Duration and High Intensity........ C. Exercise of Short Duration and Low Intensity......... D. Exercise of Long Duration and Low Intensity.......... E. Exercise of Long Duration and High Intensity......... F. Summary.............................................. Uanb-l-‘J-‘w II. FAT AND EXERCISE........................................... 0\ A. Free Fatty Acids..................................... 6 1. Muscle Uptake.................................. 6 2. Splanchnic Uptake................... ....... .... 7 B. Triglycerides........................................ 8 C. Ketone Bodies........................................ 9 1. Purpose and site of ketone production.......... 9 2. Mechanism of ketone production................. 11 3. Control of ketone production................... 12 4. Muscle metabolism.............................. 13 5. Post-exercise ketosis.......................... 14 D. Summary.............................................. 14 III. CARBOHYDRATE METABOLISM.......... ..... ..................... 15 A. Lactate and Pyruvate...... ....... .................... 16 B. Glycogen............................................. 20 1. Liver glycogen................................. 20 a. Control of glucose production by the liver.................................... 24 2. Muscle glycogen................................ 24 3. Effect of diet........................... 24 b. Effect of exercise....................... 25 3. Glycogen in the central nervous system......... 26 C. Summary.............................................. 28 iii Page IV. PROTEIN METABOLISM......................................... 29 A. Effect of a Protein Diet............................. 32 B. Protein Intake and Serum Urea........................ 33 C. Starvation and Muscle Protein Metabolism............. 33 D. Effect of Fasting on Blood Uric Acid Level........... 35 E. The Alanine Cycle.................................... 36 l. The amino acid exchange in the postabsorptive state.......................................... 36 2. Alanine and gluconeogenesis.................... 37 3. The glucose-alanine cycle...................... 38 4. Purine nucleotide cycle........................ 39 5. Protein feeding................................ 4O 6. Starvation and the alanine cycle............... 40 7. Diabetes mellitus and the alanine cycle........ 42 8. Exercise and the alanine cycle................. 43 F. Summary.............................................. 45 V. HORMONAL CONTROL OF SUBSTRATE UTILIZATION.................. 45 A. The Insulin/Glucagon Ratio........................... 46 l. Glucose homeostasis............................ 46 2. Insulin/glucagon ratio and AA.................. 46 3. Insulin/glucagon ratio and FFA, ketone bodies.. 47 4. Effect of insulin/glucagon ratio on gluconeo- genic capacity................................. 49 B. Growth Hormone....................................... 50 C. Thyroid Hormone...................................... 51 D. Glucocorticoids...................................... 52 E. Hormone Regulation in Exercise....................... 53 l. Insulin/glucagon ratio during exercise......... 53 2. Aldostrone, cortisol and plasma renin activity. 56 F. Hormone Regulation During Starvation or Carbohydrate Restriction.......................................... 57 G. Summary.............................................. 58 PART TWO: EFFECTS OF LOW CARBOHYDRATE DIETS VI. EFFICIENCY AND HIGH FAT DIET.... ..... ...... ..... ... ...... .. 60 A. Substrate Efficiency................ ..... ............ 60 B. Endurance............................................ 63 C. Starvation and Work Performance...................... 65 D. Possible Explanation for Decreased Efficiency........ 66 E. Summary.............................................. 68 iv Page VII. EFFECT OF DEHYDRATION ON EXERCISE................. ....... .. 68 SlmaryOOOOIOOOOCOOOIOOO......OOOOOOOOOOOOOO ........ O 73 VIII. WATER BALANCE AND CARBOHYDRATE RESTRICTION ........... ...... 73 A. Weight Loss.......................................... 73 B. Reason for Weight Loss............................... 75 C. Water Loss During Glucose Deprivation................ 77 D. Possible Mechanisms for Altered Water Balance on Low Carbohydrate Diet.................................... 79 E. Summary................. ............ ................. 82 IX. LIMITING FACTORS DURING EXERCISE ..... .......... ....... ..... 83 A. Local Muscle Substrates.............................. 83 l. Triglycerides.................................. 83 2. Glycogen.......... ..... ........................ 83 3. Proteins....................................... 85 B. Liver Glycogen....................................... 87 C. Acid—base Balance.................................... 88 D. Other Humoral Factors................................ 92 1. Potassium...................................... 92 2. Other factors. ..... ............................ 93 E. Summary............ ........ ...... ....... ............. 93 EXPERMNTS O O O O O O O O O O O O O O I O O C O O O O I O O O O O O O O O O O O O O O O O O O O O O O O O O O O O I 95 PART ONE: EXERCISE EXPERIMENTS I. EXPERIMENT I. O O I O O O O O O O O O O O O O O O 0 O O O O O O O O O O O O O O O O O O O O O O O O O O O 95 A. Introduction......................................... 95 B. Procedures........................................... 96 l. Muscular efficiency............................ 96 2. Experimental outline........................... 96 3. Subjects....................................... 99 4. Diets.......................................... 100 5. Exercise....................................... 102 6. Measurements................................... 102 7. Equipment...................................... 103 8. Calculations and analysis...................... 104 C. Results.............................................. 104 1. Oxygen consumption and RQ during rest.......... 104 . RQ and oxygen consumption during exercise...... 104 . Blood data..................................... 108 . Body weight.................................... 108 J-‘UJN V. 5: Page 5. Water—balance data ............................. 113 6. Urinary ketone excretion ............. . ......... 113 II. EXPERIMENT III (EXERCISE) .................................. 116 A. Introduction ......................................... 116 B. Procedures.. ......................................... 117 C. Results ..... .... ..................................... 118 1. Standard work ........................... . ...... 118 2. Maximal oxygen consumption.... ..... . ....... .... 118 III. EVALUATION OF DATA ...................................... ... 120 A. Experiment I ......... .......... .......... . ..... . ..... 120 1. Resting oxygen consumption and R0 .............. 120 2. Exercise RQ........................... ........ . 121 3. Efficiency and oxygen consumption ....... ....... 122 4. Weight loss... ..... ............. ...... . ........ 125 5. Water-balance........ ........... ...... ....... .. 125 B. Experiment III (Exercise) ............................ 127 1. Standard work............. ............... ...... 127 2. Efficiency... ...... ...... .......... . ....... .... 130 3. Maximal oxygen consumption ............ . ........ 131 IV. SWY... ...... O 0000000000 ......OOOOOOOOOOOOOOOOO00.0.0... 132 PART TWO: EFFECTS OF THE DIETS v. EXPERIMENTS IIANDIII.0.0000000000000000...000000000oooooo 133 A. Introduction........................ ..... ............ 133 B. Procedures........................................... 134 1. Experimental design............................ 134 2. Diets.......................................... 135 3. Analysis....................................... 136 4. Experiment III................................. 138 VI. RESULTS FROM EXPERIMENT II................................. 140 A. General Observations................................. 140 B. Urine Analysis....................................... 140 C. Blood Analysis........................ ..... .......... 144 D. Weight and Water—balance............................. 145 vi «v- '-‘. ~01»? h... Inn-g ..v r“ n i” or] t; Page VII. RESULTS FROM EXPERIMENT IIIOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO 145 A. urine Analysj-S. O O O O O O O I O O O O O O O O O O O O O O O O O I O O O O O O 0 O O O O O 145 B. BlOOd AnaJ-YSiS O O O O O O O O O O O O O O I O O O O 0 I O O 0 O O O O O O O O O O O O O O O 150 C. Body Weight and Water-Balance........................ 150 VIII. EVAI‘UATION OF DATAOOOOOOOOOOOOOOOOOO. ..... ......OOOOOOCOOOO 150 A. Keton-SOOOOOOOOOOO................OOOOOOOOCCOOOIC.... 150 B. Urinary Nitrogen Excretion, BUN, and Blood Glucose... 156 C. Other Blood Constituents............................. 168 D. Metabolic Profiling of Organic Acids in the Urine.... 171 E. Cholesterol, Triglycerides and HMG................... 173 F. Body Weight Loss and Water—Balance................... 180 G. Dehydration...................... ...... .............. 198 IX. SUMMARY..................... ..... .......... ...... .......... 204 SUGGESTIONS FOR FURTHER RESEARCH ........... ..................... 205 BIBLIOGRAPHY.............. ......... ..... ...... .................. 209 APPENDICES A. Composition of the high carbohydrate control diet for Experiment 1.0.0.0000...O.........OOOOOOI....OOOOOOOOOOOOOO 242 B. Composition of the Fat I diet during Experiment I.......... 243 C. Composition of the Fat II diet during Experiment I......... 244 D. Composition of the Dressing used in the high fat diets dur- ing Experiment ICC...00............OOOOOOOOOOOO00.0.0000... 245 E. Composition of the high carbohydrate control diet for mperment IIOOOOOOOOOOOOOOOOOO.......OOOOOOOOOOIOOOO...... 246 F. Composition of the Fat I diet for Experiment II... ..... .... 247 G. Composition of the Fat II diet for Experiment II........... 248 H. Composition of the Dressing used in the Fat II diet during Experiment IIOOOOOOOOOOOOOOOOOOOO.......OOOOOOO0.00.0.00... 249 I. Composition of the high carbohydrate control diet for Experiment IIIOOOO......OOOOIO0.0.0.000.........OOOOOOOO... 250 J. Composition of the Fat I diet for Experiment III........... 251 vii a APPENDICES Composition of the Fat 11 diet for Experiment III.......... The content of EAA in the high carbohydrate control diet for Experiment 11, Calculated from the diet composition given inAppendix E00000...............OOOOOOOOCOOOOOOO.... The content of EAA in the Fat I diet for Experiment II, calculated from the diet composition given in Appendix F... The content of EAA in the Fat II diet for Experiment II, calculated from the diet composition given in Appendix G... viii Page 252 253 254 255 'a F1 LIST OF TABLES TABLE 1. 10. 11. 12. 13. 14. Experimental outline showing the sequence of the different diets, the kind and amount of fat and proteins, and the kind of exercise experiments performed..................... Weight, height, and birth-dates for the subjects........... Approximate composition of diets (percentage of calories) in Experiment IOOOOOOOOOOOOOOOO.............OOOOOOOOOOOI..0 R0 and oxygen consumption during rest in Experiment 1...... Pulse rate, RQ, and oxygen consumption during exercise, for KB, in Experiment I.0.00.00...O.......OOOOOOOOOOOOOOOOO.... Pulse rate, RQ, and oxygen consumption during exercise, for WH, in Experiment I........................................ Blood pH and pCO2 for KB in Experiment I................... Body weight and weight loss during exercise for KB in hperiment 1.0.0.0.....................OOOOOO00.00.000.000. Body weight and weight loss during exercise for WH in Experiment IOOOCOOOOOOOOO......OIOOOOOOO00.000.00.00....... Mean weight loss in kg (from Tables 8 and 9) during Experiment IOOCCOOOOOOOOOOOCO......OOOOOOCOOOOCCCOOOO...... pH and water-balance data for KB during Experiment I....... Mean water-balance and urinary pH for KB during Experiment IOOOOOOOOOOOOOOO.......OOOOOOOOOOOOOOOI.00...... Urinary ketone levels above control values during the Fat II regimen in Experiment IOOOOOOOOOOOOOO......OOOOOCOOO Approximate composition of the diets as percent of calories during Experiment IIIOOOOO0.0.0.0000.........OOOOOOOOOOOOOO ix Page 97 99 101 105 109 110 111 111 112 112 114 115 115 117 TABLE 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. Oxygen consumption, RQ and heart rate for KB during Standard walk in Experiment III.........OOOOCOOOOIOOOOO... Maximal oxygen consumption and heart rate for KB during Experiment IIIOOOOOOOOOOOOOO.......OOOOOOOOOOOC0.00.0.0... Approximate composition of diets as percent of calories in mperiment II...0.0.0.0....O............OOOOOOOOOOOOOOO... Composition of diets as percent of calories in kperment IIIOOOOOOOOOOOOOOOOOOOO......OOOOOOOOOOOIOOOOOO Urine pH and the excretion of ketones and nitrogen during Experiment II (5% of calories from protein). Each dietary PeriOd lasted five daYSOOOOO......OOOOOOOOOIOOOOO000...... Urinary excretion of organic acids during Experiment II... Arterialized pH and pCO2 as mm Hg in Experiment II........ Blood constituents during Experiment II (5% of calories from protein-8).00000000000000.0000oooooooooooooooooooooooo Water-balance and body weight data during Experiment II... The pH, ketones and nitrogen excretion in the urine during Experiment III (15% of calories from proteins)............ Blood constituents during Experiment III (15% of calories from protein8)ooooooooooooo0.00000000000000000000000000... Hemoglobin and hematocrit during Experiment III........... Water-balance and body weight data during Experiment 111.. Mean values of BUN, blood glucose, serum SGOT, and the nitrogen excretion in the urine during Experiments II and III.............OOOOOOOOOOOOOOOOOO......OOOOOOOOOOOOOOOOIO Serum calcium, phosphorous, bilirubin, and uric acid (in mg/dl).................................................... Serum cholesterol, urinary hydroxymethylglutarate, and germ triglycerideSOOOOOO...OOOOOOOOOOOOO0.00.0.0.0....... Water-balance and body weight loss during Experiment II-.. Page 119 120 136 139 141 143 144 146 147 148 151 153 154 157 169 174 181 TABLE 32. 33. 34. 35. 36. Water-balance and body weight loss during Experiment III... Total weight loss during each dietary period.. ......... .... Plasma albumin and total protein concentration at the end of the control and fat periods during Experiment II........ Mean plasma albumin and mean total protein concentration for the control and fat periods during Experiment 111...... Mean hematocrit data during Experiment III................. xi Page 181 182 199 199 200 (.5) ‘4 \(A) LIST OF FIGURES FIGURE 10. ll. 12. RQ's during exercise for KB and WH during Experiment I.... Nitrogen excretion in the urine, BUN, and blood glucose for KB during Experiment II............................... Nitrogen excretion in the urine, BUN, and blood glucose for KB during Experiment III.............................. Nitrogen excretion in the urine, BUN, and blood glucose forRHdurj-ng Experiment IIIOOOOI.....OOOOICOOOOOO00...... Serum cholesterol and hydroxymethylglutaric acid (HMG) forKBduring Experiment IIOOOOOOOOOOOCCOOCOOOOO.......... Plasma cholesterol and triglycerides for KB in Experiment IIIOOOOOOCOOOOOOOOOO0.0.0.0000... ..... 000...... Water-balance for KB during Experiment II................. Body weight, ketone excretion, and water—balance for KB during Experiment IIOOOOOOIOOOO......OOOOOOCOOOOO0.0.0.... Water-balance for KB during Experiment III................ Body weight and water-balance for KB during Experiment III Water-balance for RH during Experiment III................ Body weight and water-balance for RH during Experiment III xii Page 107 162 164 166 177 179 186 188 190 192 194 196 . . ”vb ‘5‘- ...F, he the f u.“ .351 U‘) 1; 1 INTRODUCTION When Saul became king over Israel, he was involved in many wars with the Philistines. On one such occasion he forbade his men to taste anything while they were pursuing the enemy. His son Jonathan, with his armorer, did not hear the command and finding some honey in the field, they tasted thereof. When Saul's men pointed out that they were forbidden to taste anything under oath, Jonathan replied that in this, Saul had commanded thoughtlessly because his men had become exhausted, whereas Jonathan and his armorer had become revived and their eyes were shining. (Holy Bible, I Samuel, chapter 14.) This is probably the first recorded experiment on the effect of nutrition upon physical performance: It appears that a little bit of sugar was able to significantly increase the endurance--an observation that is very relevant today. Krogh et a1. (1920) studied the relative efficiency of fat and carbohydrates as a muscular energy source in man and found that fat was about 10% less efficient. Marsh and Murlin (1928) reported similar results. On this basis, it was generally believed that fat was less efficient than carbohydrate as an energy substrate for muscular work. Christiansen and Hansen (1939) studied endurance on high fat and high carbohydrate diets in man; and found that endurance was less than half on a high fat diet as compared to a high carbohydrate diet. Bergstrom et al. (1967a) studied endurance on various diets and found, by taking needle biopsies from the exercising muscles, that the point of exhaustion correlated well with depletion of glycogen. Since fat is the major substrate for endurance activities, it is unlikely that it is less efficient than carbohydrate for muscular work. It was therefore decided to investigate the possible role of ketosis and dehydration in decreasing the efficiency. o p-n— P. L!» l ‘c H ...-o v.\. ———- 9253 as h: I" . in: tin: (Van :83 intex “5i as t‘: A. REVIEW OF LITERATURE PART ONE: SUBSTRATE UTILIZATION I. EXERCISE AND SUBSTRATE UTILIZATION When discussing the substrate used during exercise, it is very important to specify the type of exercise studied. Dynamic Muscular Endurance (DME) is the ability to perform under variable conditions, such as high intensity power exercise and endurance exercise. The DME can be subdivided into nine physiologically different levels, each supposedly with a different pattern of substrate utiliza— tion (Van Huss, 1977). At the two extremes we have very short duration, high intensity exercise (60 yards dash) and endurance-type exercise such as the marathon. A.'Rg§£ In the resting postabsorptive state, fat has been shown to contribute the majority of the energy (Havel, 1971), but ketone bodies may also contribute to a limited amount (Ziegler et a1., 1968). Glucose is also used to some extent and it is the only substrate for some tissues. In the postabsorptive state there is an increased re— lease of amino acids (AA), particularly alanine, from the muscle; but local oxidation has not been proven (Felig and wahren, 1970). r1 .=.'.€ 31’: 33:51. 'I'LEI'EEE H' 0‘ 'P'm _ ‘Nd'h c :m‘r, (- n.. (1' 'l I B. Exercise of Short Duration and High Intensity During this type of exercise the main substrate is preformed high energy phosphate; but some conversion of glycogen to lactate also appears to take place (Hultman, 1967). The oxygen consumption is very low. It is generally believed that the cytoplasmic ATP is compartment- alized, and that only a small fraction is in equilibrium with phospho- creatine (PC) (Bergstrom et al., 1971b). These authors also showed that the muscle level of PC is inversely related to the work load, whereas the ATP level is very little affected, except at very high intensity work. At exhaustive exercise, the PC level is reduced to less than 10%, whereas the ATP is reduced by about 40% of resting level. C. Exercise of Short Duration and Low Intensity Again, the phosphagens are an important source of energy; but it appears that glycolysis dominates as substrate, presumably because of the decreased rate of energy utilization, particularly as work time increases (Pernow and Wahren, 1962; and Jorfeldt, 1970). D. Exercise of Long Duration and Low Intensity As work time increases, the respiratory quotient (RQ) decreases from about 0.85 to about 0.70 (Havel et al., 1963). This indicates a shift to aerobic metabolism of free fatty acids (FFA). However, some muscle glycogen is still used, but this is primarily for aerobic oxida— tion (Hultman, 1967). As exercise continues, the fat supplies up to 90% of the substrate (nine hours work). At this time glycogen appears to be depleted, so the glucose is presumably supplied by gluconeogenesis. E. .1 k‘ne use subs: :Lsele fal cat degre auditions 1:1“ ‘MA ,.--.....an, L. -iition amt pr :, ‘ x8166 (: F. . 35 hue to u l E. Exercise of Long Duration and High Intensity When the intensity of exercise increases, the muscle begins to use substrate stored in the muscle cells. The glycogen content of the muscle falls rapidly, but lactate does not accumulate to any signifi- cant degree. It has been shown that duration of exercise under these conditions is well correlated with the initial glycogen content (Hultman, 1967). The lipid stores in the cell are apparently also used under such conditions, but they do not become depleted, and the amount used or amount present before exercise is not related to the amount of work per- formed (Froberg et al., 1971). F. Summary It is clear that the kind of substrate used for energy production depends very much upon the type of exercise performed. However, it is also influenced by the kind of muscles used for a particular task, which again, to some degree, is determined by the type of exercise performed. Muscles are generally classified as red oxidative and white glycolytic (Pette, 1971). In evaluating the effect of a particular substrate, it is desir— able to choose a type of exercise that primarily uses the substrate in question. At rest during the postabsorptive state, fat is the major sub- strate, as it is during exercise of long duration. During high intensity exercise at short duration, the phosphagens are the major substrates; but as the duration of the exercise increases, so does the importance of glycogen and blood glucose. During exercise of long duration and relatively h and may “395 l. .‘1 31' long (in! free fatty kitten (19 huur of ex ;rupurtio: 10112611! a“ m.‘ ' v41: “ 13;. .' ' " 1‘. “Q5 \ q.- sq ‘1 ‘1 3 d1 3“ relatively high intensity, glycogen becomes progressively more important and may under such circumstances be a limiting factor. II. FAT AND EXERCISE A. Free Fatty Acids 1. Muscle Uptake. Fat is the major energy substrate for exercise of long duration, and it is available to the muscle cell in three forms: free fatty acids (FFA), triglycerides (TG), and ketones. Hagenfeld and Wahren (1968a) studied the metabolism in the human forearm during one hour of exercise. They found that the muscular extraction of FFA was proportional to the arterial concentration; however, since the arterial concentration increases during exercise, the uptake also increases. The fractional uptake of the FFA was about 15% (arterial concentration 440 to 520). They also found that the muscle has some preference for oleic and linoleic acid as compared to palmitic acid; this is in con— trast to the exercising leg where Havel et a1. (1964) did not find any difference. The decreased preference for palmitic acid has been con- firmed by Nestel and Barter (1971), who found that the fractional turn- over of linoleic acid was greater than palmitic acid, and that less incorporation in plasma triglycerides took place. The uptake of FFA could account for about 50% of the fat oxidation, and about 60% of the FFA was oxidized, the remainder going primarily to B-hydroxybutyrate. They also demonstrated that the exercising forearm was able to oxidize glycerol. However, recently Hagenfeld (1975) concluded that there is no significant difference in muscle uptake of the individual FFA, whereas the splanchnic uptake varies with each individual FFA. iiage he uptake 72:: at he. FFA; and (a ilk: h u I" .1an W .n. *- '4. ‘4‘ r4- on .‘4' I‘ J rv (1') Hagenfeldt and Wahren (1972) found that at low work intensities, the uptake of FFA from the blood could account for all the fat oxidized, but at heavier work intensities, fat oxidation exceeded the uptake of FFA; and they suggested that fat stores in the muscles made up for the difference. They also found that during low work intensities, the fat was completely oxidized, whereas at higher intensities, only about 60-70% was oxidized to carbon dioxide, the rest of the label going to acetate, 3-hydroxybutyrate, 2-oxogluterate and citrate. 2. Splanchnic Uptake. When splanchnic FFA uptake was studied (Hagenfeldt and Wahren, 1973) by arterial and hepatic vein catheteriza- tion, arterial FFA, renal blood flow, and arterial - hepatic venous dif- ferences decreased during exercise, whereas lactate increased. Analysis showed that the decreased uptake is not solely due to lower arterial FFA level. They concluded that the splanchnic FFA uptake is reduced during exercise to allow greater FFA utilization by the muscle. Hagenfeldt and Wahren (1975b) have recently shown that arachidonic acid is handled differently than oleic acid. The fractional uptake dur- ing rest in both splanchnic area and in muscles is 50% higher; further- more, the turnover-rate is unaffected by exercise. Jones et a1. (1972) found that exercise under hypoxic conditions increased FFA metabolism as was the arterial (brachial) lactate con— centration. Hagenfeldt and Wahren (1973) found that the hepatic arterial lactate level correlates negatively with the FFA level and hepatic extraction. Hagenfeldt and Wahren (1975a) have studied turn- over of FFA during recovery from exercise by use of 14C labelled acid. The arterial concentration of FFA reached a maximum of twice the level- durin :uuclude th giasm pool Rage cf TEA inc saturation 53‘s. He Rupiah 3i the tut utilizati 3. Ce :uatribu: used by arterial :3 Me: below t‘ the eae he IEa ‘ I c wen 1“ E level during exercise six minutes after termination of exercise. They conclude that the postexercise peak is caused by release of FFA into the plasma pool; this in turn augments the removal of FFA. Hagenfeldt(1975) has shown that the splanchnic fractional uptake of FFA increases as the chain length decreases and as the degree of un- saturation increases; and the uptake varies greatly with the individual FFA's. Hagenfeldtand wahren (1973) concluded that the net splanchnic FFA uptake is reduced during exercise in order to permit redistribution of the turnover of FFA in the body, so that more FFA's are available for utilization by the muscle. B. Triglycerides Carlson et a1. (1970) showed that plasma triglyceride (in humans) contributes to the energy supply of the heart. Recently (1975) they have investigated whether triglyceride also contributes to the energy supply used by working muscle in humans. The problem here is to measure the arterial-venous difference with sufficient accuracy. They were not able to detect any significant differences in concentration; but differences below the standard error would still be able to account for up to 25% of the energy utilization by the muscle. This could be at least part of the reason why the FFA extraction cannot account for all the fat oxidized during strenuous exercise. Although fat appears to be the major substrate for muscle energy, feeding a high fat diet does not appear to increase performance. Krogh et a1. (1920) found early that the muscular efficiency decreased 10-12% in subjects fed a high fat diet, a result that was confirmed by Marsh and Eurlin (19' iecreased :3 a high and they a scores. 1 mi work aziiosis. C. l. 'eetne bo ated with E'3""€‘~‘er, i‘illiazsc L‘éartaut *7. Murlin (1928). Christensen and Hansen (1939) showed that the work time decreased by 2/3 when the subjects changed from a high carbohydrate diet to a high fat diet. Hultman and Bergstrom (1967) found similar results and they ascribed the decreased endurance to the depletion of glycogen stores. Henschel et a1. (1954) found a decreased ability to perform hard work during acute starvation, and ascribed this to dehydration and acidosis. C. Ketone Bodies 1. Purpose and site of ketonegproduction. The discovery of ketone bodies in the urine resulted in the belief that they were associ- ated with a pathological process or were products of abnormal metabolism. However, this view has given way to the physiological role of ketosis. Williamson (1971) suggests the following roles for ketones: They are important substitute fuels for the brain, and have an antilipolytic action. They can be considered normal fuels in situations such as starva- tion and exercise. The principal site of ketone body formation is the liver (Wieland, 1968). The liver in turn, is unable to use ketones because it lacks 3-oxoacid CoA—transferase (Williamson et al., 1971). However, there is some evidence that B—hydroxybutyrate is produced by the muscle during exercise (Hagenfeldtand wahren, 1968b), but it is probably less important than the liver production. The ketone bodies can be considered an easily soluble transport form of energy, and they are particularly useful for the brain, where they at least partially can substitute for glucose as substrate (Owen et al., (196?). Keto heart (Willie mutation < mu ’mo increases u: concentratic but :1 nus: decreases (I Keto tetug-enic c' I I Of {1131 Keith, 19. It fix”; its Q in With this kind “3 that “basis 10 (1967). Ketones have long been known to be the preferred fuel for the heart (Williamson and Krebs, 1961). After an overnight fast, the muscle consumption of ketones is low (in humans), but after three days of fast- ing with increased arterial concentration of ketones, the consumption increases dramatically. After twenty-four days of fasting, the arterial concentration of acetoacetate and 3—hydroxybutyrate is further increased; but the muscle utilization of these substrates remains unchanged or even decreases (Owen and Reichard, 1971b). Ketosis appears to have some clinical significance in that the ketogenic diet forms the basis for an apparently fairly successful treat- ment of many types of epilepsy that are otherwise not easily controlled (Keith, 1963; and Livingston, 1973). It is generally believed that the formation of ketone bodies is the basis for the success of this dietary regimen. It should be mentioned that besides the physiological ketosis and its clinical applications, there are cases of pathological ketosis in which ketone production appears to be out of control. Examples of this kind are the severe ketosis seen in the terminal stages of diabetes and that sometimes seen in lactating cows (Krebs, 1966). Pathological ke"insis is always associated with excessive gluconeogenesis and Krebs (1966) suggests that the use of oxaloacetate for gluconeogenesis de- presses the level of oxaloacetate to such a degree that the TCA cycle cannot handle the amount of acetyle CoA. However, during physiological ketosis a sufficient decrease of oxaloacetate has not been found. 2. gluconeoge is 53?. 1 fless the genesis is gluconeoge Xe; aim of o. triglycer; fasted ra thesize k. Perfusion :ricarbox: state, On. 50nd tha fasted st. amated ll 2. Mechanism of ketone production. Flatt (1972) found that gluconeogenesis has a permissive effect on ketogenesis. The link here is ATP. The conversion of fatty acid (FA) to ketones generates ATP, and unless the energy is used, the process comes to a standstill. Gluconeo- genesis is an energy utilizing process; therefore, when ATP is used for gluconeogenesis, ketogenesis is allowed to proceed to a higher degree. McGarry and Foster (1971a) investigated the ketogenesis by infu- sion of octanoic acid. This FA is not used directly for synthesis of triglycerides. With high concentrations of octanoate, the livers from fasted rats, unlike those from normal animals, could be induced to syn- thesize ketones at a rate approximately equal to diabetic rats. Perfusion studies with octanoate-l-l4C revealed that the activity of the tirrcharboxylic acid cycle, although modestly decreased in the ketotic State, only moderately influenced the rate of ketogenesis. They also folmd that the rate of ketogenesis from octanoic acid is higher in the fa~Sted state because of depressed lipogenesis: Normally this acid is elongated. McGarry and Foster (1971a) similarly found that decreased TCA- Q57°1e activity cannot account for the enhanced ketogenesis seen in fast- ing . They reported that the decreased triglyceride formation is suffi- c'fLe‘nt to account for the increase in ketogenesis seen during fasting. one point generally agreed upon is that a raised plasma level of FFA is a necessary and important prerequisite for ketogenesis; but McGarry and Foster (1972) emphasize that an increased uptake of FFA by the liver in itself is not sufficient to initiate maximal ketogenesis. If for ketoge higher lex and Pi-Su: lover leve rsaturaté he: the s for satura is apparex canceivabl Eugen-es: Th4 mantra is ccntro. N. III- M :1... in th. t3 ketone fiesterif be aireat. Mm 331.8 is l iliicatin! as mean: 331.3 i 39p} 12 If decreased reesterification of FFA in the liver is responsible for ketogenesis, it can explain why medium chain triglycerides give a higher level of ketones than corn oil (Tantibhedhyankul et al., 1967; and Pi-Sunyer et al., 1969). Unsaturated fat on the other hand, gives lower levels of triglyceride in blood, despite faster incorporation of unsaturated FA into triglyceride. Hagenfeldt et a1. (1972) have shown that the splanchnic fractional uptake is greater for unsaturated FFA than for saturated, and greater for short chain than for long chain FA, which is apparently due to faster esterification. Upon this basis, it is conceivable that the different FA should have different rates of ketogenesis. The liver uptake of an individual FFA is proportional to its concentration in plasma (Hagenfeldt et al., 1972), so presumably there is control of ketogenesis at two levels: First, the plasma concentra- tion is regulated, and secondly, there is regulation through the fate of FFA in the liver where the FFA can either be esterified, oxidized, or go to I(atone bodies. As mentioned earlier, it has been suggested that if reeSterification is depressed, as during the fasting state, the FA will be directed toward ketogenesis. McGarry and Foster (19713) have shown, however, that if the TCA Cycle is blocked, reesterification occurs even in the fasted state, 111dleating that the esterification process is fundamentally intact. This means that there must be some other means of control. 3. Control of ketone production. The catecolamines, particularly morePinepherine, are important effectors of lipolysis (Rossel and Ballard, 19.71), a lipolysi glutagon Sieberdo 11g chyl tin. l} presence that 733'; :issue c {197' ). ILEDBIIC inegat: Senior ; Of 3&2], Site-353 10m t} 1:531:95; 13 1971), and insulin is known to be a potent inhibitor of adipose tissue lipolysis (Cahill et al., 1966). Although controversial, it appears that glucagon also stimulates lipolysis in man (Gerich et al., 1976b). Bieberdorf et a1. (1970), increased plasma FFA in fasting rats by infus- ing chylomicrons and heparin, but this did not increase ketone produc- tion. When they infused insulin, ketone production decreased in the presence of sustained high plasma FFA. Pi-Sunyer et al. (1970) found that hyper—ketonemia had a feedback effect on lipolysis in adipose tissue of dogs. Similar results were obtained by Bj¢rntorp and Schersten (1970). Madison et a1. (1964) found that infusion of ketones in alloxan diabetic dogs caused hypoglycemia, decreased hepatic glucose output to 50% of control, and caused a greater than 50% fall in arterial FFA. They noted that these results are similar to those obtained by infusing insulin. Balasse and Neef (1975) demonstrated that the ketone bodies have a negative feedback effect on their own production in humans. However, Senior and Loridan (1968) failed to find any increase in insulin level of man, although they found a significant decrease in FFA, glycerol and glucCse upon infusion of 3-hydroxybutyrate. Grey et a1. (1975) also f°und that insulin did not mediate the negative feedback of ketones upon 1190137318 and ketogenesis in man, as it does in experimental animals. 4. Muscle metabolism. Hagenfeldt and Wahren (1968b) studied the metalbolism of ketone bodies in human subjects during rest and exercise With 3-hydroxybutyrate-4-14C as tracer. During rest, they found that the uptake of ketones by the forearm muscle was proportional to the am... 7 arterial tion. I? micati: 5. indicates exercise. ranted i :czditior glam PI 313219 u; OI Palmiz Eilanghn. usual a} l4 arterial concentration; but during exercise, there was no such correla- tion. 0n the contrary, they found a net release of 3-hydroxybutyrate, indicating intramuscular production of ketone bodies during exercise. 5. Post-exercise ketosis. The phenomena of post-exercise ketosis indicates the sharp raise in blood ketone level after the end of strenuous exercise. Johnson and Walton (1972) explain this phenomenon by the pro- nounced increase in FFA after the termination of exercise; under such condition (glycogen depletion), ketone production is proportional to plasma FFA. Rennie et a1. (1973) found that post-exercise ketosis is less in trained athletes than in untrained, even when they work at the same relative workload . Askew et al. (1975) studied this phenomenon in rats and found that training increased the capacity to oxidize ketone bodies; however, contrary to results in humans, they found higher blood ketone levels in trained than in untrained rats. This may be because the rats were exer- Cised to complete exhaustion, whereas the results in humans were at sub- mainDal workloads; or else it is possible that rats respond in a differ— ent Manner than humans, in terms of ketone regulation during exercise. D. Summary Organs differ greatly in their handling of fat substrates. The mUSCle uptake is proportional to the arterial concentration, which incerases during exercise. It appears that the fractional extraction of Palmitic acid is slightly lower than for the other FFA's. In the Splanchnic area, the FFA uptake decreases during exercise, and the frac- t101131 extraction decreases with an increase in chain length and the iggfee of 5 fit this ha aid is hat :1: area. is unaffeci Ketc irreases : Tie ketones in tissues 2;}: glucos mm the :zilizatior ketone bodi starvation. static meg} imletely 335 in the 15 degree of saturation. Triglycerides are used as substrate by the heart, but this has not been proven to be the case in the muscle. Arachadonic acid is handled in a special manner by both the muscle and the splanch- nic area. Its uptake is about 50% higher than other FFA's at rest, and is unaffected by exercise. Ketones are primarily produced by the liver, and the production increases in situations where the supply of carbohydrate is limited. The ketones function as a substitute substrate for glucose, particularly in tissues (such as in the central nervous system) which are dependent upon glucose. The ketone production increases gradually, reaching a maximum the third or fourth day of glucose starvation; but because the utilization of ketones by the muscle decreases, the blood level of ketone bodies may increase until the seventh to tenth day of carbohydrate starvation. It appears that ketone production is regulated by a homeo- Static mechanism. Though the regulation of ketone production is not completely understood, it is apparently regulated by the FFA concentra- tion in the blood and the insulin/glucagon ratio. Both ketones and FFA increase in the blood after exercise, presumably due to increased lipo— 1y818 , and ketone production is affected by training. III . QARBOHYDRATE METABOLISM The regulation of blood glucose is one of the most important homeostatic mechanisms, and glucoseican be used for energy substrate in “ON: tissues, although some, such as the heart, seem to prefer FFA. HOWEVer, other tissues such as the central nervous system, the renal medulla and the erythrocytes, normally use only glucose as a substrate ("Jeiss and Car' converted ea: be sto evidence t‘ extant 50' after itpo 197)). tissue). ' reSillthesi The to be {Gla r 0" “Us lltt' ltensity < 1 . tsacs )am {Rated to m” ConVer l6 (Weiss and L3ffler, 1970). Carbohydrate, besides being a direct source of energy, can be converted to fat and stored in the adipose tissue and, in small amounts, can be stored as glycogen (Hultman et al., 1974). While there is no evidence that fat can be converted to glucose, glycogen serves as an im- portant source of glucose in maintaining the blood glucose level; the other important gluconeogenic substrate being protein (Weiss and Lo'ffler, 1970) . Besides being oxidized to CO2 and water, glucose can be metab— olized to lactate (Hultman et al., 1974). The former process is far more efficient, but the latter has the advantage of being able to pro— ceed under anaerobic conditions. A. Lactate and Pyruvate Cori and Cori (1928) described the cyclic process of glucose metabolism to lactate in the peripheral tissue (particularly muscle t13811e). The lactate is then transported to the liver where it is res)"llthesized into glucose. The lactate formation in skeletal muscle during exercise appears to be related to the relative work intensity; below 502 of aerobic capa- city, little increase in blood lactate concentration is seen. At higher Sub‘maximal loads, blood lactate increases in proportion to the work intensity during the first minutes of exercise, and at close to maximal loads’ a constant production is observed. Either lactate formation is I:Qlated to hypoxia, or pyruvate is formed faster than it can be used and thus converted to lactate (Knuttgen, 1971). The lactate formed can go to the treat] an: the 4'.;. it ‘n ' “|‘lt TL- ‘ a: the 17 to the liver where it is resynthesized to glucose, or it can be used directly as an energy substrate in muscle tissue. It should be pointed out that not all the oxygen debt is due to lactate formation; the alactic debt appears to be due to the splitting of phosphagens. Karlsson (1971a) has shown that there exists a considerable gradient for lactate between the intracellular space and the blood; the exact value depends upon the intensity of the exercise. This has been confirmed by Hirche et a1. (1971) in rats. The work load applied was 50 to 60% of maximmn oxygen uptake and the maximal lactate concentration in the muscle was close to 29 mmoles/liter wet muscle; the corresponding blood lactate concentration was about 15 moles lactate/liter blood. Af ter extremely heavy bicycle exercise (exhaustion in 50 seconds), the Ini'='l-3£:L:mal muscle lactate concentration was 39 moles/liter wet muscle. This gave a lactate ratio of up to six, between inter- and extracellular la><—‘—tate. These experiments (Karlsson, 1971a) also indicate that the total lactate production at the time of exhaustion is constant, regard- leSS of workload, time to exhaustion, or muscle lactate concentration at time of exhaustion. A good correlation was also found between blood lactate and muscle lactate concentration. Karlsson (1971a) found that pyr:‘l-l-‘vate increases to only a minor degree, and suggests that oxygen def icit is involved in accumulation of lactate during heavy exercise- This may be mediated through the NAD/NADH ratio. Saltin and Karlson (1971a) found that training significantly decreased the muscle lactate production, even when their subjects worked at the same relative oxygen consumption. One explanation for this may be that a is the VC explain I 7-. I,- . 1::5335‘5‘ relativel this blor 2971) f: tion 0cm bleed). related 1 rrcducti; m 5k 7"? ate I 16165011 “Vane: a: fart}. H1 tan: Elie: 552 of 31.“ this a1 laxate a stain?! 18 be that athletes use a larger proportion of the red muscle fibers. As the work load increases, more white fibers are activated, which can explain the increased production of lactate as the work load increases (Hermansen, 1971). The delay in adequate oxygen supply is often given as an explanation for lactate production. However, Carlson and Pernow (1961); and Keul et a1. (1967) have shown that it occurs regardless of relatively high oxygen tension in the venous blood from the muscle, but this blood comes from both active and nonactive muscles. Hermansen (1971) finds that at exhaustive work below ten minutes duration, exhaus- tiOn occurs at about the same blood lactate level (18 mmoles/liter blood). He also shows that the rate of lactate production is closely I'eILated to the work intensity, and that there is a pronounced lactate Production at work times as short as ten seconds. Jordfeldt (1971) studied the turnover of l4C—L(+)-lactate in human skeletal muscle during exercise and found a continuous uptake of lactate during exercise, most of which was oxidized to CO2 and other me tabolites. The fractional oxidation of lactate (7. of lactate take up c"-7"lilverted to C02) increased from 38% at ten minutes of exercise to 52% at forty minutes. Hermansen et a1. (1973) found that after a period of intermit- tent exercise, blood lactate decreased twice as fast when recovering at 652 of maximal oxygen uptake compared to receovery at rest. When exer— Q:Lsing at 65% of maximal consumption, there was a slight increase in lactate at five minutes; but from then on, lactate levels decreased Steadily. It is generally assumed (Rowell et al., 1966) that most of he lact :re rest :‘ze abov rscle d m the rise in he? for exercise Tinted iv; ~~Jate “Eleni 3". .mt' Mg‘ 8? c of x the Q 19 the lactate formed is removed by the liver; however, it is known that the resting skeletal muscle does use lactate (Jordfeldt, 1970), and the above results indicate an increased uptake of lactate by skeletal muscle during exercise. Minaire and Forichon (1973) investigated the lactate metabolism and the lactate-glucose interconversion during prolonged physical exer- cise in dogs, by using 14C-L-lactate and ll‘C-D-glucose as tracers. They found that the rate of lactate formation increased (doubled) during exercise, and that almost 75% of the lactate produced during exercise was immediately oxidized in the muscle. The glucose turnover rate also about doubled, but there was no change in the fractional interconversion rates of lactate to glucose and glucose to lactate during exercise. Thus, despite increased lactate production, there was no accumulation because of Oxidation in the muscle. They found that only 13% of the lactate re1n<>ved was converted to glucose. It then appears that there is an it1tj4nate balance between lactate formation and its oxidation in the SIgeletal muscle during exercise. It is a question whether at low levels of exercise, lactate is not formed to any important degree, or whether it is oxidized to 002 as fast as it is formed in the skeletal muscle. Ho"7ever, it is clear that during exercise close to the maximal oxygen uptake, and at supermaximal oxygen uptakes, lactate accumulates in the b:L'Dod (Hermansen, 1971). It can be calculated that during endurance type exercise, the energy derived from lactate formation is insignificant (less than 0.57. of the extra energy cost of the exercise) (Minaire and Forichon, 1973). Tee quest hit of e‘ mc'icati: sugplies :‘zis ATP stored e. Emmi, :ean val “Eight 0 m Nils in quest found 14 "Eight, ZnSCle a .‘j‘: S‘quct . stifiC ie: 20 Why is the lactate formed at all? One The question one must ask is: bit of evidence for an attractive theory is the work of Hoffman (1973), indicating that there is a special membrane compartment of ATP which + + supplies energy for one component of the Na /K - pump. The source of this ATP is anaerobic glycolysis. B. Glycogen Although glycogen is found in most tissues, the only important ones in terms of quantity are the muscle and liver (Hultman et al., 19 74). This does not preclude some functional importance of glycogen stored elsewhere, such as in the brain and nervous tissue (Janzo Ko 12 mi, 1974) . The highest content of glycogen is found in the liver, where the mean value is about 50 g/kg under normal conditions. With a liver Weight of 1.8 kg, the total amount of glycogen is about 100 g (Hultman and Nilsson, 1971) . The glycogen content of muscle varies depending upon the muscle in Question, but in the quadricepts femoris muscle, Hultman (1967) follnd 14 g/kg. If we assume that the muscle mass constitutes 40% of body we:Lght, a normal man will have a total of about 400 g of glycogen (in mtlacle and liver), but it varies with the metabolic condition of the Subj ect. l. Liver_glycogen. Earlier (Cahill, 1964; and Samols and Holdsworth, 1968), it was believed that the rate of gluconeogenesis was sufficiently high to maintain the blood glucose levels, and that the liver glycogen stores were only used in emergency situations. However, it has n: crease iI is also 1 :erbohyd' nearly all iyfrate . of about 133 g/kg tee, th‘ 21 it has now been shown that during starvation, there is a continuous de- crease in the glycogen stores of the liver due to glycogenolysis. This is also the case on a carbohydrate—free diet (Nilsson and Hultman, 1973; and Nilsson et al., 1973). The rate of decrease in liver glycogen varied, but on the average it was 0.3 mmoles glycosyl units/kg liver/ min- At such a rate the stores will be depleted after 24 hours of carbohydrate deprivation. Likewise, when the liver was depleted, or nearly depleted, by carbohydrate starvation or exercise, and a high carbo— hydrate diet was fed, the glycogen level increased to supernormal levels of about 500 mmoles glucosyl units/kg wet liver tissue (corresponding to 100 g/kg wet liver). When the subjects remained on a carbohydrate free diet , the glycogen level remained very low (Nilsson and Hultman, 1973). During hard, dynamic exercise, the liver glycogen stores are utilized, and may be depleted in exercise of long duration. When glyco— gen is broken down, water is liberated. The relationship has been determined directly by muscle biopsy. When 1 gm of glycogen is used, 0- 45 mEq. potassium and 2.7 g of water are released (Bergstro'm and Hultman, 1972). If the glycogen is metabolized aerobically, an addition- a1 0.6 g of water is formed. If an athlete uses 500 g of glycogen, he will liberate 1650 m1 of water, which will give a weight loss of more th-an 2 kg. It has been found that the glucose output from the liver in- ereases from 0.85 mmoles glycosyl units/min at rest up to 6 moles gly- Qosyl units/min at the end of a hard exercise period (Hultman, 1967). This increase is positively correlated both to work load and duration of exercise. aperiod iiet, the bad: case :e- glucog Liver was the low c the upta’e in; exerc gimose c Fe m 4 ,5; 1E 01.‘ firing se tributes T’t bleed gl 1 the kidne EiléSis 1 is about gar43515 c of thESe Hi the: glyc :7“. a ‘t slime ‘9; d»‘ . the li 22 exercise. In a study where the subjects performed heavy exercise after a period on a low carbohydrate diet and after a period on a normal mixed diet, the liver glucose production increased by a factor of three in both cases. After the mixed diet, the increase was almost entirely due to glucogenolysis, and the uptake of gluconeogenic substrate by the liver was unchanged despite a large increase in blood lactate. After the low carbohydrate diet, however, the liver glycogen was very low, and the uptake increased from 0.3 mmoles/min at rest to 2.14 mmoles/min dur- ing exercise, thus accounting for about 80% of the increase in the liver glucose output (Hultman and Nilsson, 1973). Felig and Wahren (1974), on the other hand, have shown that the muscle output of alanine increases from 50% during mild exercise to 500% during severe exercise, and they estimate that the gluconeogenic AA con- tributes from 10 to 15% of the liver output of glucose. The liver is the principal organ responsible for maintaining the bl°°d glucose level, although, under conditions of metabolic acidosis, the kidneys play a part in gluconeogenesis in connection with ammonia- getlesis (Steiner et al., 1968). In a resting man the glucose consumption 13 about 1 mol/day. The liver produces this glucose by either gluconeo- ge":lfiisis or glycogenolysis, and as mentioned earlier, the relative role of these two processes has been a matter of debate. Nilsson and Hultman (1973) found in the postabsorptive state that glycogenolysis produced about 0.54 mmoles glucose/min, WhiCh means the glycogen stores will be depleted after 24 hours. Since the output by the liver is about 0.85 mmoles glucose/min, gluconeogenesis must ' u: s-l Bbk J. n33: "re ad) . . 4| - 4 ..C 3 a .L .l C .nu 0 #5 v t s. s e . a .i.‘ v .. a .14 3. a: P. :5 RI. I «is :v .1... P. ~ 3‘ qP. 3.: {L 23 supply about 0.31 mmoles glucose/min, which was experimentally verified by liver vein catheterization. After a low carbohydrate diet, the splanchnic glucose production decreased to 0.30 mmoles/min, but in this case there was no glycogenolysis since the liver was depleted of glycogen. This means that when the glycogen stores are depleted, there is not a compensatory increase in gluconeogenesis in the resting state, whereas during exercise there is a several-fold increase in liver glucose output. When liver glucose output falls during carbohydrate starvation, the systemic glucose utilization must necessarily decrease and it is only after a time lapse of at least 24 hours that ketosis begins to develop (Bloom, 1967; and Hultman and Nilsson, 1975). Williamson et a1. (1971) found that the central nervous system (CNS) is able to use ketones al- ready in the fed state, and it is probably the ketone concentration in the blood that determines its utilization. Oxidation of ketones in the CNS therefore appears to substitute for the decreased glucose utiliza- tion. If exercise is performed on a low carbohydrate diet, low blood Sugar is occasionally seen (Bergstrom et al., 1967b) The gluconeogenic substrates are: lactate, pyruvate, AA (in pa-IS‘ticular, alanine), and glycerol. Glycerol comes from hydrolysis of fat to FFA, and AA from protein degradation. Since lactate and pyruvate are formed by glycolysis, no new glucose is made by this process. Felig and Wahren (1974) have estimated the relative importance of the gluconeogenic precursors in the normal postabsorptive state: 10—15% from pyruvate and lactate; alanine, 5—10%; other AA, 5%; and glycerol, 370- Glycogen normally contributes about 75%, but as fasting or glucose derivatic stile glufi iseéiate availabil: synthesis. up; 5: al., 1- 3236-51 und. liver its is limiti 3331’in sluiceort 24 deprivation extends beyond 24 hours, this becomes practically zero, while gluconeogenesis apparently remains unchanged (Nilsson and Hultman, 1973). a. Control of glucose production by the liver. Gluconeogenesis and glycogenolysis appear controlled mainly by hormones. There is an innnediate effect involving changes in enzyme activity and substrate availability, and a long-term effect associated with de novo enzyme synthesis. Epinepherine and glucagon stimulate glycogenolysis in the liver; this effect appears to be mediated through 3',5'-cyclic - AMP (Hultman 8t al., 1974). When gluconeogenic substrates such as alanine are in— fused under gluconeogenic conditions, the glucose production by the 11Ver immediately increases. This indicates that substrate availability is limiting the glucose output (Hultman and Nilsson, 1975). Glucagon Primarily increases the activity of the gluconeogenesis, while the glueocorticoids seem to stimulate de novo synthesis of gluconeogenic e‘3-4'5371nes. Epinepherine and growth hormone appear to stimulate lipolysis. Insulin opposes all these effects. 2 . Muscle glycogen. a. Effect of diet. The glycogen content of quadriceps femoris muscle at rest has a mean of 85 moles glycosyl units/kg and varies 51:01:: 60-120 mmoles/kg (Hultman, 1967). In the resting muscle no net glYeagen consumption can be measured, and it appears to rely solely “POI: FA oxidation (Andrews et al., 1956). Complete starvation or a carbohydrate-free diet therefore has only minor influence upon the gly- cogen stores of the muscle, provided no exercise is performed. Cb. 25 A slight increase in muscle glycogen was observed when the sub— ject changed from low to high carbohydrate diet. However, if the glyco- gen stores are first emptied by a period of exercise, a high carbohydrate diet will lead to rapid resynthesis of glycogen, whereas it will take up to nine days on a carbohydrate poor diet (Hultman and Bergstrom, 1967). If the muscle group is completely emptied of glycogen and the subjects are given a high carbohydrate diet, an. overshoot in the glycogen content can be obtained, but only in the muscles that were depleted. On a high carbohydrate diet resynthesis of the glycogen stores is complete within 24 hours; however, the glycogen content continues to increase to very high values for up to eight days (Ahlborg et al., 1967a). If a low carbohydrate diet is given, resynthesis takes about ten days. b. Effect of exercise. During muscle contraction, glycogenolysis and glycolysis increase in relation to the work load. The output seems to be regulated by the ATP concentration. During isometric contraction with maximal voluntary contraction force, only a fraction of the glyco- gen store is used; but during relatively heavy, dynamic exercise (about 857. of the subject's maximum oxygen uptake), the glycogen stores in the muscle are depleted and appear to be limiting for performance (Bergstrom at al., 1967b). At these work levels there is a linear relationship 13etween work time (at a given level) and the size of the muscle glycogen stores prior to exercise. The glycogen stores in the muscle cannot be used to maintain the blood glucose level since the muscle does not con- tain glucose-6-phosphatase; it therefore serves strictly as a local substrate for energy production (Hultman et al., 1974). 5:31:83 ‘ ...1‘. . 3.42.; 34 3"""5 Justus C158 12 1"! “rafts 5317:3115 26 3. Glycogen in the central nervous system. Hultman et al. (1974) states that the brain contains practically no glycogen and only a very small amount is found in fat tissue and kidney; however, very small amounts may play functionally important roles and such seems to be the case in CNS and adipose tissue. Shimizu and Kumamoto (1952), with the help of staining tech- niques, showed the presence of glycogen in neuropil—, neuroglial- and nerve cells in the brain of most mammalian species. They also showed that certain areas such as hypothalamus and area postrema, and area on the floor of the fourth ventricle, are particularly rich in glycogen. The gray matter always contains more glycogen than the white matter; glycogen is also found in some of the cell nuclei in the spinal cord. According to Koizumi (1974), glycogen granules have been demon- Strated with electronmicroscopy in nerve cell parikaryon, axons, axon terminals, and dendrites of specific regions in the CNS of various SPecies. Shimizu and Kumamoto (1952) demonstrated the presence of gly- cogen in the cytoplasm of neuroglial cells, and suggest that it plays an inlportant role in the mediation of nutrition from the blood to the nervous tissue proper. Koizumi (1974) states that psychotropic drugs, especially CNS depressant drugs such as chlorpromazine, are responsible for an increase in brain glycogen, and this indicates either an inhibition of the glyco- lYtic pathways or a depression of the functional activities in these areas. Kumamoto (1953) studied the effect of starvation on brain and liver glycogen in rabbits. He found that brain glycogen decreased even .1. a.— ...-v “:1.“ av: '58 4V; . tr: _ w~v -..e s~,. HZuIOI " ~ n; B" a,‘l 'lei. ‘ e U; :35 ;h3 27 faster than liver glycogen as fasting progressed, but the rate of de- crease varied in the different areas of the brain. Vrba (1954) found that brain glycogen decreased with physical exertion (1.5 hours of swiunning) in rats; and Jacoubek and Svorad (1959) found the lowest level of glyco— gen in rat brains after two to five hours of swimming. Insulin decreases the level of glycogen and free sugar in brains of normal cats and, to an even greater degree, in animals made incapable of secreting epinepherine (no lipolysis, (Kerr et al., 1937). Convulsive activity is accompanied by a decrease in brain glycogen, glucose, phos- phocreatine, adenosine triphosphate lactate and inorganic phosphate. There is a tremendous increase in glucose metabolism during the abnormal neuronal activity associated with an epileptic attack (Klein and Olsen, 1947). Ketone bodies or the associated acidosis are usually cited as the realson for the effectiveness of a ketogenic diet in treating epilepsy (Livingston, 1973). However, it seems reasonable to suggest that the depletion of brain glycogen stores, when the ketogenic diet is used, removes the energy source necessary for the tremendous neuronal activity asSociated with an epileptic attack. The neuromuscular system is often implicated as a cause of fatigue. Astrand and Rodahl (1970) discuss whether fatique to voluntarily mllsoular effort is located in the CNS or in the neuromuscular junction, but in either case, this may be related to the decreased function of the N8471?.- pump and the resulting ionic imbalance. Hoffman (1973) has shown some evidence for a specific membrane pool of ATP that is preferentially i Dub '1'. o a v .- u ..I- A tie est: a; o“: 3.. H c 2.x slit .Jt 28 used by one component of the Na+/K+—ATPase. Parker and Hoffman (1967) have shown that ADP in this pool is phosphorylated only through anaerobic glycolysis in the red blood cells (RBC). If this is also the case in other cells, it can explain why glycogen is so important for nerve and muscle function, even though fat or ketones can supply the major part of the energy. C. Summary Carbohydrate is found in the body as blood glucose and glycogen. During anaerobic metabolism glucose is converted to lactate, and the amount of lactate produced depends upon the intensity of the exercise. The lactate production is influenced by training, and at exercise of high intensity and short duration, the total lactate production appears to be correlated to the time to exhaustion. The resting skeletal muscle does not metabolize lactate, but lactate becomes a substrate for the muscle during exercise. Lactate production is increased parallel with the activation of the white muscle fibers, but most of this lactate is immedi- ately oxidized in the red muscle fibers, while the rest goes to the liver for gluconeogenesis. The two principal sites of glycogen storage are the liver and the muscle, and the pattern of utilization varies in these two organs. The liver glycogen is used in support of the blood glucose level and will be depleted in about 24 hours during carbohydrate starvation. If carbo— hYdl‘ate starvation is followed by a high carbohydrate diet, the glycogen 1"”61 increases above normal, but the liver glycogen is not repleted if a low carbohydrate diet is consumed. If exercise is performed when a low 29 carbohydrate or high carbohydrate diet is consumed, the liver glucose production will increase about three—fold in both cases; but in the former case, it is due to gluconeogenesis, and in the latter, to glyco- genolysis. The liver production of glucose is regulated by hormones. A low carbohydrate diet does not cause (or only slowly causes) 21 depletion of muscle glycogen, but it is depleted by exercise, and if ftillowed by a high carbohydrate diet, supernormal glycogen levels are fcnind. At work loads of about 85% of V0 max. the glycogen stores are 2 ccxrrelated with the time to exhaustion. Glycogen stores are also found 111 the CNS, which are depleted on a low carbohydrate diet, and may be a cause of the fatigue usually experienced when consuming such diets (Bloom and Azar, 1963). IV - PROTEIN METABOLISM Since the days of Leibig, many workers have supported the notion ttuat proteins are essential as energy substrates for work performance (Kraut and Lehman, 1948). However, it is now well-established that Luuier'most conditions, this is not the case. It is clear that when gly- cogen is present in sufficient supply, protein is not used for energy to anYappreciable degree; but recently there has been much interest in the rol£3 of amino acids during work of long duration (that is, more than one hour) (Keul et al., 1972). Consolazio et a1. (1975) recently studied the effect of heavy Physical activity and training for forty days on two levels of protein intakes (1.4 and 2.8 g/kg of body weight). Based on the high protein intake, it is not surprising that there was no difference in performance; vu‘ -.- a n‘ a: «i 30 Nitrogen balances were positive. They found an increased loss of urea in the sweat, but surprisingly, no compensatory decrease in urinary nitrogen excretion on the high protein diet. Rougier and Babin (1975) studied changes in blood and urine urea and uric acid in trained and untrained men during exercise of short dura- tion (running 3.5 km on a treadmill at a rate of 14 km/hour at 5% eleva- tion), and during long-lasting exercise (an average of 15 km) by running to exhaustion on the treadmill. During exercise of short duration, the blood urea increased 25% at the end of exercise and almost 507. one hour after exercise, but was back to normal the next morning. The pattern was similar during long-lasting exercise. Blood urea increased 407. at the end of exercise, about 55% after one hour, and was still 20% above normal the next day. The increase was higher in trained people. Similar changes were observed for uric acid. This increase in blood urea has been found by others (Refsum et al., 1973). Rougier and Babin (1975) tried to evalu- ate renal function immediately after exercise, one hour later, and the next day. They admit that their clearance values were not too reliable since the urine flow was often less than 2 ml/min. They found the great- est increase in urea excretion right after exercise of short duration aSSOCiated with a temporary increase in the urine flow; one hour later, urea excretion and urine flow were below resting values. During the night and the next day, urine flow gradually increased, but urea excretion incJ-‘Eased more than accounted for by the increased flow. Uric acid excretion increased immediately after exercise and continued to be above normal until the next day. During long-lasting 31 exercise, urea excretion decreased due to decreased urine flow. But after the exercise, urea excretion continued to increase above resting values as urine flow continued to increase toward resting levels. They also measured the hematocrit and showed that the moderately increased hematocrit could account for only an insignificant part of the increase in the blood urea level. The authors concluded that decreased urinary excretion could not account for the increase in blood urea, since some subjects maintained normal urine flow during exercise; furthermore, uric acid and urea were not retained to the same degree, and were not related to plasma creatinine levels. The authors did not measure sweat rates, but this has been studied by others (Cerny, 1973) who obtained similar (20% increase) re- sults for serum urea and uric acid when their subjects exercised for two hours on an ergometer. The calculated sweat urea excretion was 20 times higher, and sweat uric acid excretion 10 times higher than during rest, whereas the kidney creatinine and urea elimination rate decreased by 40% and 50% respectively during exercise. They also concluded that the increase in blood urea after two hours is not due to decreased renal excretion, but to increased protein catabolism. Similar results have been found by Kchatorian (1972); and Porzolt et a1. (1973). It is clear that blood urea nitrogen is elevated after prolonged exercise and it does not seem to be due to dehydration or impaired renal excretion. It has been suggested that the amino acids are used for energyinthe working muscle (Lowenstein, 1972). There is, however, no direct proof for this. But it is well-established that proteins play 32 a role in gluconeogenesis through the alanine cycle (Felig, 1975). A. Effect of a Protein Diet Tolstoi (1929) studied the effect of an exclusive meat diet on the blood constituents in two "normal" men for one year. Ketonuria was reported to occur daily, and blood uric acid concentration increased but returned to control levels after about three months on the meat diet. They had no control cholesterol values, but at the start of the meat diet, values as high as 800 mg/100 ml were reported, with an average of about 400 mg/100 ml. McClellan and Du Bois (1930) report results from the same men on the meat diet (15-25% of the calories from protein, 75-85% from fat, and 1—2% from carbohydrate). The subjects lost an average of two kg during the first week on the meat diet, which the authors explain as a shift in the water content of the body, and nausea and weakness were reported in one subject. Increased urine volume was reported when shifting from normal to carbohydrate free diet, and it was lowest when carbohydrate was again added. Ketone bodies in the urine remained relatively high (up to 12 g/day). One subject was given a glucose tolerance test, and showed glucose uria. Urea clearance test showed above normal excretion of urea after one year on the meat diet; but after a period on mixed diets follow— ing the meat diet, the urea clearance was below normal. No elevation in blood urea was reported, but large variations were observed, probably asSOciated with variations in protein intake. When the subjects switched to the high fat diet from the mixed diet, the blood ketone bodies in- m'Ieased until the fourth day, after which they remained constant in one p;ffl'VhL tint '3‘.‘ V :‘Vz'ba‘ a. wash. differ r—' - (11 O J4 33 subject, but decreased steadily in another subject; it was relatively low throughout in a third subject. Recently Heeley et al. (1975) studied the effect of a high protein diet in six young men for four days. The subjects lost an average of 1.4 kg during the four days, a significant increase in blood urea was found (from 4.65 to 8.13 mmoles/liter), and the urinary urea excretion increased from 23 to 39 g/day. B. Protein Intake and Serum Urea Addis et al. (1947) studied the relation between serum urea concen- tration and protein intake. The serum urea level was determined at three different protein intakes (0.5, 1.5 and 2.5 g/kg). They found consider— able individual variation, but the percent increase in serum urea level correlated closely with the increase in protein intake; and the variation in blood urea level for each level of protein intake was about 15%. Taylor et a1. (1974) showed that there is a correlation between serum urea and dietary nitrogen utilization; although, if the blood urea nitrogen is plotted as a function of protein intake, an equally good correlation is obtained, regardless of the source of proteins. However, when subjects were given 90-95% of their protein from wheat, their blood urea was only half that observed during an isonitrogenous control diet (Bolourchi et al., 1968). So it appears that under normal conditions, the serum urea level is proportional to the protein intake; but under certain conditions, other dietary factors appear to have an important influence. C. Starvation and Muscle Protein Metabolism During short periods of starvation, there is increased peripheral release of AA for gluconeogenesis (Pozefsky et al., 1976). The composition 34 and degree of weight loss was determined during different weight reduc— ing regimens: An 800 kcal mixed diet, an 800 kcal ketogenic diet, and total starvation. The respective average weight losses were: 278 g/day, 467 g/day, and 751 g/day. The average protein losses were: 9.5, 17.9, and 50.4 g/day, respectively. And the corresponding water losses were: 101.7, 285.3, and 457.2 g/day. There was no significant increase in ketones on the mixed diet, but on the ketogenic diet and during starva— tion, the urinary excretion of ketones seemed to reach a plateau on the seventh day of about 2 and 7 g/day, respectively. There was a signifi— cent increase in urinary ketone excretion on the third day only (Yang and Van Itallie, 1976). It is seen that the nitrogen excretion was only slightly higher on the ketogenic diet than on the mixed diet (same protein intake), but much higher during total starvation. Also, the subjects gained weight on the 1200 kcal post-experimental diet when it followed starvation or the 800 kcal ketogenic diet, whereas they continued to lose weight when it followed the 800 kcal mixed diet. This weight gain was due to water re- tention. They also found that the basic metabolic rate (BMR) was unaf— fected by the dietary regimen, although the volitional physical activity decreased 5% during and after the starvation period; and the energy cost of different activities was unaffected by the dietary regimen (as calcu— lated from a record of the subjects' activity). Yang and Van Itallie (1976) explain the increased weight loss as a shift in the water-balance during the different regimens. Looking at their data, however, it is clear that a substantial part of the weight 35 loss was due to glycogenolysis and protein catabolism (up to 500 g/day during the starvation period, including the associated water loss) (Olsson and Saltin, 1970). This interpretation is supported by the fact that the subjects gained weight on a 1200 kcal mixed diet when it follow— ed the 800 kcal ketogenic diet, but not when following the 800 kcal mixed diet. It appears obvious that after the ketogenic diet the liver and probably some muscle glycogen is depleted, which is probably not the case after the mixed diet. Blackburn and Flatt (1973) found that a daily protein intake of 0.6 to 1 g/day/kg of body weight (as the only food intake) was very similar to total fasting. The amount of urinary nitrogen is comparable, indicating that the rates of gluconeogenesis are similar. The ingestion of protein, however, will compensate for the degradation of AA and the nitrogen balance will be close to zero or even positive. The decrease in body mass is almost exclusively due to a temporary loss of water and decrease in body fat (Flatt and Blackburn, 1974). Infusion of AA without glucose does not prevent ketosis (Blackburn and Flatt, 1973), so fat is still mobilized to supply the energy need. D. Effect of Fasting_on Blood Uric Acid Level Lennox (1924) studied the effect of total fasting on the blood uric acid concentration, and found that it increased. In seventeen sub- jects starved for eight days or longer, the uric acid rose on the average, from 4 to 11 mg/100 ml, an increase of about 165%. Uric acid began to increase on the third or fourth day of starvation, and was maximal after about a week, being maintained to the end of starvation. 36 This increase in uric acid appeared to be due to decreased excretion. Lewis and Corley (1923) found similar results; the uric acid re- mained high when starvation was followed by a high fat diet. Harding (1927) observed that diets containing sufficient fat to produce ketosis also gave an increase in serum uric acid. Christofori and Duncan (1964) confirmed these results, and found that it was due to decreased AA and glucose reabsorption; these compounds normally compete with uric acid for reabsorption. Wilson et al. (1952) showed that a high protein diet increased uric acid excretion. Bonsnes and Dana (1946) suggested that there is competition between glucose and uric acid for reabsorption. Christofori and Duncan (1964) showed that glycine and glucose promptly increased uric acid excretion. E. The Alanine Cycle Felig and Wahren (1974) postulated a glucose-alanine cycle involv- ing peripheral synthesis of alanine by transamination of pyruvate. Protein metabolism is important for glucose homeostasis, since AA are released by the muscle and gut, and extracted by the liver. Alanine, which is rapidly converted to glucose in the liver, constitutes a large part of the interorgan flux, despite the fact that it accounts for less than 10% of the AA in the muscle. Glutamine also appears to play an important role. One important aspect of the alanine cycle is that it removes ammonia from the muscle in a nontoxic form. 1. The amino acid exchange in the postabsorptive state. The ob- servations on the net balances of AA across muscle, liver, gastrointes- tinal tract, kidney, and brain in normal man in the postabsorptive state 37 clearly demonstrate the key role of alanine and glutamine in the overall flux of AA between tissues (Felig, 1975). Free AA are released from muscle and gut, and are extracted by the liver. Alanine and glutamine account for more than 50% of the total alpha amino nitrogen released by the muscle. The gut also releases substantial amounts of alpha amino nitrogen, primarily as alanine. The liver is the major site of alanine uptake, where its extraction exceeds that of all other AA. Glutamine is primarily taken up by the kidney and gut, and used for alanine synthesis in the gut (Matsutaka et al., 1973) and for ammoniagenesis in the kidney (Cahill and Owen, 1970). The branched chain AA (valine, leucine, and isoleucine) are taken up by the brain rather than the liver (Felig et al., 1973). 2. Alanine and gluconeogenesis. Since alanine is the primary AA released by the muscle, it must be the most important AA for gluconeogene- sis, and indeed, gluconeogenesis has been demonstrated to increase propor- tionally to the availability of alanine (Hultman et al., 1974). For most AA, gluconeogenesis is saturated at three times their normal plasma AA concentration, but alanine does not reach saturation until about 25 times normal concentration. l4C—labelled alanine is promptly incorporated into glucose in the postabsorptive state and after prolonged fasting (Felig et al., 1970a). While all AA other than leucine are potentially gluconeo- genic (Krebs, 1964), alanine accounts for more than half of the AA used for hepatic gluconeogenesis. In the postabsorptive state, glycogenolysis accounts for about 75% of glucose formation, and alanine accounts for about 30% of the remainder (Felig, 1973). It is therefore mainly in 38 situations where glycogen is depleted, such as prolonged fasting, ketotic hypoglycemia and long-lasting exercise, that the alanine cycle can achieve functional importance. 3. The glucose-alanine cycle. Alanine accounts for more than 30% of the alpha amino nitrogen leaving the muscle, but muscle protein only contains 7-10% alanine (Odessey et al., 1975). Alanine must therefore be synthesized de novo in the muscle by transamination of pyruvate (Felig et al., 1970b). In the liver, alanine is then converted to glucose and urea. The amino-group, however, must come from some- where, and Felig and Wahren (1971a) suggest that the branched chain AA supply the amino group. Muscle has been demonstrated to be the site of oxidation of the branched chain AA (Manchester, 1965). There appears to be a linear relationship between pyruvate and alanine, which is not observed for any other AA, and 14C-labelled glucose indicates that 60% of the alanine released is derived from exogeneous glucose degradation products (Odessey et al., 1975). It is noteworthy that the branched chain AA account for only 10—22% of the alpha AA released from the fares arm muscle (Felig et al., 1970b), although these AA constitute about 20% of the muscle proteins (Odessey et al., 1975). It thus appears that muscle proteins not only participate in glucose homeostatis, but are also metabolized locally for energy. Also, physiological increments of branched chain AA increase the formation of alanine from 14C-labelled glucose (Odessey et al., 1975), and the release of amino groups from the branched chain AA can account for all the nitrogen recovered in alanine. Other AA added to the plasma fail to increase the alanine out- put. It has been suggested that the branched chain AA are released from 39 the liver to complement the glucose-alanine cycle (Odessey et al., 1975). Since lysine is not synthesized or catabolized in the muscle, its re- lease is a measure of muscle proteolysis, and since the muscle alanine content is similar to the lysine content, it can be estimated that 67% of the alanine released from the muscle is derived from glucose degrada- tion products (Felig, 1975). This has been confirmed with labelled glucose in rat diaphragm (Odessey et al., 1975). Felig (1975) also calculates that the alanine cycle operates at a rate of about 50% of the Cory cycle. 4. Purine nucleotide cycle. Besides functioning in glucose homeostasis, the alanine cycle also transports amino groups to the liver in a nontoxic form. The branched chain AA are preferentially oxidized by the muscle tissue, and this may be particularly important during exercise, since ammonia is produced by the contrasting muscle (Lowenstein, 1972). The ammonia production appears to be proportional to the amount of work. Lowenstein and Tornheim (1971) have proposed a purine-nucleo— tide cycle in which ammonia is released from AMP by the adenylate diaminase, and suggest that the ammonia has a regulatory effect on the energy-supply in muscle tissue. Tornheim and Lowenstein (1973) have demonstrated that the purine-nucleotide cycle and glycolysis are closely linked. The glucose-alanine cycle may also be important in terms of energy production. Conversion of glucose to alanine provides 8 mmoles of ATP compared to 2 mmoles ATP for lactate formation (Odessey et al., 1975). If one considers the oxidation of branched chain AA facilitated by alanine formation, 30-40 mmoles of ATP are formed per mol of AA oxidized (Krebs, 1964). 40 5. Protein feeding. When a protein meal is ingested, the plasma level of branched chain AA increases, while that of alanine decreases, regardless of the composition of protein (Frame, 1958; and Armstrand and Stave, 1973). Animal studies show that following protein ingestion, alanine predominates in portal flow, while glutamate and aspartate are absent. Uptake of alanine by the liver exceeds the gut output, indi- cating continual peripheral release in the absorptive state. Hepatic uptake of branched chain AA is low. The production of alanine by gut and muscle thus appears to take place in the absorptive--as well as postabsorptive state (Felig, 1975). There is evidence that this is also the case in man (Frame, 1958). Elwyn (1970) has suggested that plasma protein synthesized in the liver undergoes hydrolysis in the peripheral tissue, and thereby participates in the inter-organ AA trans- port. It must be remembered that the erythrocyte participates in the inter-organ transport of AA (Felig et al., 1973). 6. Starvation and the alanine cycle. Starvation can be divided into an early and a late phase. In the early phase, the response is primarily directed toward maintaining glucose homeostasis; but in the latter, toward preserving the muscle protein stores (Felig et al., 1969). Hultman and Nilsson (1971) have shown that the glycogen stores are rapidly depleted during fasting. Felig et al. (1969) state that there initially is an augmented hepatic uptake of glucose precursors, particu- larly alanine, to maintain blood glucose production. This appears to be different from feeding a carbohydrate poor diet, where Nilsson et a1. (1973) found no change in the hepatic uptake of alanine when the liver is 07.13 41 is depleted of glycogen. The cause of this difference is not known but on the carbohydrate diet, the caloric supply is adequate; during star— vation, it obviously is not. The increased gluconeogenesis seen during the first three days of starvation is due both to increased hepatic extraction of alanine (Felig et al., 1969) and to increased release of alanine from the muscle (Pozefsky et al., 1974). These changes in alanine metabolism appear to be related by the insulin/glucagon ratio. As starvation continues for weeks, there is a gradual decrease in protein catabolism as evidenced by a fall in the urinary nitrogen excretion (Owen et al., 1969). The decreased gluconeogenesis is made possible because the increased ketone production replaces glucose as an energy substrate in the brain (Owen et al., 1967). In this second phase of starvation, a general decrease in the plasma AA concentration is seen, and the alanine concentration in particular, decreases (Felig et al., 1969) because of decreased muscle output of alanine (Felig et al., 1970b). The hepatic alanine extraction is increased, if anything, and if alanine is given orally or intraven- ously, a prompt increase in hepatic glucose output is seen (Felig, 1972). The hepatic uptake of alanine is controlled by the insulin/glu— cagon ratio, and catecolamines may also play a role (Mallette et al., 1969). Glucagon has no effect on muscle release of alanine, and insulin has very little effect (Rudeman and Berger, 1974). Felig (1975) suggests that blood ketone level may have a feedback effect on muscle alanine release; however, on a low carbohydrate diet, alanine appears to continue being released despite elevated ketone bodies (Hultman and Nilsson, 3. 42 1975). In this situation, though, the alanine output was at all times much lower. The branched chain AA are rate-limiting essential precursors for protein synthesis, and in addition, they (especially leucine) promote synthesis and decrease degradation of proteins directly (Buse and Reid, 1975; and Fulks et al., 1975). When the carbocyclic analogs of the branched chain AA were fed to fasting man, a significant reduction was seen in nitrogen excretion (Sapir et al., 1974). As starvation prog— resses, ketone bodies increase in the blood, mainly due to decreased metabolism of the keto acids in the muscle (Owen and Reichard, 1971a). Garber et a1. (1974) has shown that ketoacid production by the splanch— nic bed is maximal by the third day of starvation. Fatty acids become the preferred substrate (Cahill, 1976), the oxidation of the branched chain AA decreases (Aoki et al., 1975), and blood levels increase. The increase in branched chain AA limits alanine output (Buse and Reid, 1975; and Fulks et al., 1975). The protein sparing effect of the AA is used clinically in starvation (Blackburn et al., 1973; and Hoover et al., 1975). 7. Diabetes mellitus and the alanine cycle. The diabetic environ- ment is somewhat similar to that of short term starvation, and probably even more comparable to that of a high fat diet. In short term starva- tion, there clearly is a caloric deficit; but in both diabetes mellitus and a high fat dietary regimen (less than 6% of calories from carbohy- drates), there is an adequate caloric supply, but a functional or real shortage of glucose. In both short-term starvation and diabetes mellitus, at: §. iflE 43 there is an increased branched chain AA concentration in the plasma, whereas alanine and other glycogenic AA decrease. Splanchnic uptake of gluconeogenic precursors (lactate, pyruvate, and alanine) increases 50-100% so that the precursors account for about 40% of the hepatic glu- cose output, compared to 15—20% in the normal postabsorptive state. This augmented glucose production is the result of a 2-3 fold increase in hepatic fractional extraction of the glucose precursors (Wahren et al., 1972). 8. Exercise and the alanine cycle. The effect of exercise on muscle AA metabolism varies with the duration and intensity of the exercise. During brief exercise (10-40 minutes), the only AA to have an increase in arterial concentration is alanine, and an output of this AA is observed from the exercising limb. The increased alanine output varies from 50-500%, depending upon the intensity of the exercise. The arterial alanine concentration increases 25-100%, and is directly propor— tional to the arterial pyruvate concentration (Felig and Wahren, 1971a). Stimulation of branched chain AA metabolism has been demonstrated during exercise (Turner and Manchester, 1973); and so has conversion of aspar- tate to oxaloacetate (Randle et al., 1971). This presumably partly accounts for the ammonia source for the increased alanine production. The alanine cycle thus provides a nontoxic mechanism for removing ammonia from the muscle. The splanchnic extraction of alanine during brief exercise is similar to the basal state; but since the hepatic glucose output increases three— to four—fold, its relative contribution to the glucose production decreases markedly. 44 During prolonged exercise (four hours), splanchnic release as well as hepatic extraction of alanine increases. As exercise continues beyond 40 minutes, splanchnic alanine extraction increases while peri- pheral release of alanine continues at elevated levels, resulting in a decreased arterial concentration of this AA. The overall uptake of glu— cose precursor increases about three—fold, so that gluconeogenesis accounts for about 45% of the hepatic glucose output. This situation is similar to that observed in diabetic patients (Wahren et al., 1972), and during short term fasting (Felig et al., 1972). The decreased insulin/glucagon ratio is probably responsible for the increase in gluconeogenesis (Ahlborg et al., 1974). During prolonged exercise, the branched chain AA become important as fuel for the exercising muscle (Ahlborg et al., 1974). A consistant uptake by the exercising leg is demonstrable for valine, leucine, and isoleucine; and there is a close relation between splanchnic production and muscle utilization of these AA during exercise. Glucagon appears to stimulate the release of branched chain AA from the liver (Mallette et al., 1969), while glucagon, epinephrine, and FA stimulate the oxida— tion of these AA by the muscle (Buse et al., 1973). During recovery, splanchnic extraction of gluconeogenic precursors increases 45-100% as a consequence of increased fractional extraction, and the total pre— cursor consumption accounts for about 45% of the hepatic glucose output. Uptake of alanine and release of glucose remain elevated during the recovery phase (Wahren et al., 1973). 53‘ ‘ nu VI A A 45 F. Summary When sufficient glycogen is present, proteins play only a minor role as energy substrate; however, when the glycogen stores are depleted (or nearly depleted), proteins appear to play a functionally important role as gluconeogenic substrate. The blood urea nitrogen (BUN) increases during exercise, and does not appear to be caused by decreased urinary excretion, but by protein catabolism. The BUN in— creases when the protein intake increases, but it also increases when a low carbohydrate diet is consumed, partly due to protein catabolism, or, to a direct affect of this dietary regimen on the kidney. The two primary AA released from the muscle are alanine and glu- tamine. Glutamine is converted to alanine in the gut, and alanine is extracted by the liver where it accounts for more than 50% of the total AA uptake. Glutamine is also used for ammoniagenesis in the kidney. The branched chain AA are primarily used by the muscle and in the brain. In the muscle, the branched chain AA are either used directly or trans— aminated to pyruvate to form alanine. The gluconeogenesis from alanine in the liver is controlled by the insulin/glucagon ratio, but the re— lease of alanine from the muscle appears to be controlled by the AA profile in the blood (increased level of branched chain AA). V. HORMONAL CONTROL OF SUBSTRATE UTILIZATION When man progresses from the absorptive state to the postabsorp- tive state, a change in substrate flux takes place that appears to be controlled by hormones. As fasting progresses, hormones further regu- late the substrate availability. During the absorptive state the SCI '0. t . CF‘A 643 «U and. I- A ”it. 46 hormonal environment favors storage of substrate, while in the postabsorp— tive state, we see substrate mobilization (Owen and Reichard, 1971a). A. The Insulin/Glucagon Ratio 1. Glucose homeostasis. The primary effect of the insulin/ glucagon ratio is to maintain glucose homeostasis. A high ratio favors the uptake of glucose by muscle tissue (liver and central nervous system not affected), and the conversion of glucose to fat in the liver. Similarly, when arterial glucose concentration falls, the insulin/ glucagon ratio decreases, which promotes gluconeogenesis in the liver from the appropriate substrates (Unger and Orci, 1976). 2. Insulin[glucagon ratio and AA. AA stimulates the release of both insulin and glucose (Pagliara et al., 1974; and Gerich et al., 1974a), but in general they are much better stimulators of glucagon secretion. The individual AA differ in their ability to affect the insulin/glucagon ratio. In man, arginine, lysine, and leucine are the most potent insulin stimulators (Fajans and Floyd, 1972), while arginine, glycine and alanine are most effective for glucagon secretion (Muller et al., 1975; and Wise et al., 1973). The branched chain AA (leucine, isoleucine and valine) have no effect on glucagon release (Rocha et al., 1972). In vitro studies indicate that AA stimulate insulin release some- what, but are potent stimulators of glucagon; however, in man the presence of glucose augments the insulin response to AA (Levin et al., 1971) while decreasing the glucagon response (Pagliare et al., 1974). AA, in conjunction with glucose, therefore play an important role in 47 regulating the insulin/glucagon ratio. The release of glucagon appears to prevent hypoglycemia otherwise caused by aminogenic insulin secre- tion (Unger et al., 1969). Insulin stimulates protein synthesis (Wool et al., 1972). This is partly due to increased transport of AA into the cell (Cahill et al., 1972b), but insulin also decreases AA catabolism (Fulks et al., 1975). 3. Insulin[glucagon ratio and FFA, ketone bodies. During glucose Starvation, hepatic glucose production increases. Initially, glyco- genolysis dominates; however, when glycogen is depleted, gluconeogenesis (primarily from AA) becomes the only source of hepatic glucose produc- tion. During prolonged starvation, the continued loss of protein is incompatible with life (Cahill et al., 1966). After several days of starvation, gluconeogenesis from AA decreases while ketone production increases and replaces glucose as the energy substrate for the central nervous system (Cahill et al., 1966). McGarry et a1. (1975) suggest that appropriate changes in the insulin/glucagon ratio are partly respon- sible for ketogenesis. The mechanism whereby glucagon induces ketogene- sis has not been elucidated, but it appears that a low level of glycogen and a high hepatic carnitine level are necessary. The hepatic ketogenic capacity does not increase until an hour after administration of gluca- gon; but unless insulin is simultaneously reduced, there is not suffi- cient lipolysis to sustain ketogenesis. In vitro FA and ketone bodies stimulate insulin release in the presence of non-stimulatory glucose concentrations (Hawkins et al., 1971). These agents also inhibit glucagon secretion in vitro (Edward 48 and Taylor, 1970). Ketone bodies stimulate insulin secretion in some species such as dog and rat (Madison et al., 1964; Hawkins et al., 1971; and Balasse et al., 1967), but in man and other species, this does not seem to be the case (Balasse and Ooms, 1968). Acute elevation of FFA has little effect on the plasma insulin level in fasting man (Balasse and Ooms, 1973; and Pelkowen et al., 1968); it does, however, increase the insulin response to subsequent glucose stimulation (Balasse and Ooms, 1973). Similar experiments show that lipids supress glucagon secretion (Andrews et al., 1975; and Gerich et al., 1974b); this action, unlike that of glucose, does not require insulin. It has been proposed that plasma FFA and ketones during starva- tion release enough insulin to prevent keto acidosis (Seyffert and Madison, 1967); and the inhibition of glucagon could be important, since it supposedly would decrease lipolysis and decreases ketone pro- duction by the liver (Gerich et al., 1964b). However, in man, FFA and ketones do not appear to influence the insulin/glucagon ratio (Balasse and Ooms, 1973). Insulin is known to be a potent inhibitor of lipolysis (Cahill et al., 1966). Glucagon on the other hand, is a powerful lipolytic agent (Steinberg et al., 1959), but there appear to be important species variation. Gerich et al. (1976b) recently showed that glucagon under certain circumstances can induce lipolysis in man; but others have not found lipolytic action (Marliss et al., 1970; and Lefebvre, 1972). However, glucagon is known to stimulate insulin secretion (Samols et al., 1966), which will counteract the lipolytic effect. 49 4._§§fect of insulin[glucagon ratio onjgluconeogenic capacity. It is well established that insulin lowers the AA content in the plasma of humans (Felig et al., 1969a; and Zinnerman et al., 1966). The de- crease is most pronounced for the branched chain AA, methionine, tyrosine, and phenylalanine; and the effect is due to inhibition of AA output from the muscle (Pozefsky et al., 1969), except for glutamate for which an uptake has been demonstrated (Aoki et al., 1973). Thus, the result is a decrease in plasma AA. In contrast, alanine concentration in the blood is not lowered by insulin, nor is muscle output inhibited; but when endogenous insulin secretion is stimulated by glucose, increased arterial alanine levels are observed in normal man (Felig et al., 1975). However, under appro- priate conditions, insulin stimulates incorporation of alanine into muscle protein (Manchester and Young, 1970). The seemingly anomalous behavior of alanine can be explained by the insulin-induced synthesis from pyruvate in the so-called alanine cycle (Felig et al., 1970b; and Felig, 1973). Insulin increases utilization of glucose by fat and muscle tissue, and inhibits its release from the liver (Madison, 1969). Studies of splanchnic AA balance after intravenous and oral glucose administration (to stimulate insulin secretion) indicate that the de- creased gluconeogenesis is due to an effect of insulin on the liver (Felig et al., 1975; and Felig and Wahren, 1971b). Insulin decreases alanine release from the muscle, since the arterial concentration of this AA is unchanged, or even increased, and 50 4C-alanine incorporation into glucose in the perfused rat liver is inhibited (Rudorff et al., 1970). Since some alanine continues to be taken up by the liver, it must be disposed of in a different way during hyperinsulinemia. The older studies of glucagon were performed with pharmacologi- cal levels rather than with physiological levels, which are about 100— 200 pg/ml (Unger, 1974). It is difficult to interpret studies with pharmacological levels of glucagon, since they result in insulin secre- tion (Samols et al., 1965). Physiological increments of glucagon during prolonged fasting resulted in about a 15% decrease in plasma AA; but there was no increase in urea excretion. So under these circumstances, there does not appear to be any catabolic or gluconeogenic effect (Marliss et al., 1970). Infusion of glucagon in high physiologic amounts has no effect on AA balance in the forearm in either the postabsorptive state or during short-term starvation (Pozefsky et al., 1974). Physiological levels of glucagon in arterial blood do not increase incorporation of alanine into glucose (Chiasson et al., 1974). B. Growth Hormone Evidence for the participation of growth hormone in human lipid and carbohydrate metabolism is based largely on experiments with pharmacological quantities of growth hormone, or with experiments in hypophysectomized individuals. In pharmacological doses, growth hormone decreases glucose utilization (Felig et al., 1971; and Fineberg and Merimee, 1974); it also induces lipolysis and ketosis (Raben and Hollenberg, 1959; and Felig et al., 1971). Using physiological levels an 51 however, similar results have not been reported (Gerich et al., 1976b). Hypophysectomy decreases ketosis and lipolysis as well as insulin requirements in diabetics (Pearson et al., 1960), but these effects may not be solely due to growth hormone (Harlan et al., 1963), and mobil- ization of FFA and ketosis occur despite growth hormone deficiency (Merimee et al., 1971). However, Gerich et al., (1976b) recently showed that under some conditions, physiological levels of growth hormone do augment lipolysis and ketonemia in man; but in the presence of physio— logical levels of insulin, these actions are normally not apparent. During exercise after an overnight fast, human subjects showed an increase in blood levels of growth hormone (Hunter et al., 1968). It is not known what activates this rise in growth hormone, but it has been suggested that it serves a protein preserving effect, and that it may be triggered by a change in the plasma AA profile; it is also observed in the long-term fasting state (Hunter, 1972). Alanine pre- sumably could play such a role, and Becker et a1. (1975) investigated the effect of alanine in protein calorie malnutrition in children, assuming that the low alanine seen in these patients caused the high levels of plasma growth hormone. They were surprised at seeing an in- crease instead of a drop, but this resultfflts the interpretation that alanine causes growth hormone secretion. C. Thyroid Hormone Triiodothyronine (T3) enhances catabolism of protein stores, which is indicated by a doubling of urea excretion during a fast when T3 was given, compared to the control period (Carter et al., 1975). Creatinine excretion increased six-to nine-fold, and increased plasma 52 levels of FFA as well as ketones were seen. The urinary excretion of ketones also increased. However, hyperthyroidism may simply increase the energy requirement. It has recently been shown that there is both an active and inactive form of triiodothyronine, and although the thyroxine level remains normal during starvation, the active triiodothyronine (3,5,3-T3) decreases strikingly, whereas the inactive triiodo (3',5',3- T3) increases (Portnoy et al., 1974; and Vagenakis et al., 1975). When refed a high carbohydrate diet, man increases the active triiodo- thyronine and decreases the inactive triiodothyronine. D. Glucocorticoids Glucocorticoids exert their effect over a period of time, prob- ably through de novo synthesis of gluconeogenic enzymes (Hultman et al., 1974). Animal studies show a protein catabolic effect in the muscle, and release of AA for gluconeogenesis in the liver when fasting animals are treated with glucocorticoids (Smith and Long, 1967). Similar data is not available for man; but hyperadrenocorticism shows increased plasma alanine, while other AA are unchanged (Wise et al., 1973). Hydrocortisone increased plasma alanine without change in other AA in experimental animals (Betheil et al., 1965). Corticosteroids also in- creased glucagon secretion (Marco et al., 1973), suggesting that the increased gluconeogenesis may be indirect, through the action of glucagon. After prolonged starvation, glucagon fails to increase protein catabol- ism (Owen and Cahill, 1973), but this may be due to elevated growth hormone levels. 230 be P) ”'I 53 E. Hormone Regulation in Exercise In evaluating the role of the various hormones during physical exercise, one must consider whether the work is of short duration or continuous, severe or mild, and whether the subjects are trained or untrained, normal or abnormal. The subjects' ages and sex are also of importance. Furthermore, prior nutritional history can influence such factors as glycogen stores, which may effect the hormonal response. 1. Insulin/glucagon ratio during exercise. Insulin inhibits both hepatic glycogenolysis and gluconeogenesis, but there appears to be a difference in the sensitivity of these two reactions toward insulin (Felig and Wahren, 1975). Glycogenolysis appears to be more sensitive than gluconeogenesis to small increments of insulin (Felig and Wahren, 1975). This has also been shown with high and low insulin doses during infusion of labelled alanine (Liljenquist et al., 1974). Serum insulin decreases during long as well as relatively short- term exercise (Wahren et al., 1971; and Ahlborg et al., 1974). Such a decrease in insulin is known to increase hepatic glucose output (Felig and Wahren, 1971b; and Wahren et al., 1971), and it will favor lipolysis (Cahill et al., 1966). During heavy work and very prolonged exercise, plasma glucagon concentration also increases, and may contribute to the increased fuel delivery (Ahlborg et al., 1971; and Felig et al., 1972). The insulin and glucagon responses to strenuous exercise are believed to be mediated by the sympathetic nervous system, and adrenergic blocking agents greatly reduce the exercise-enduced hyperglucagonemia (Harvey et al., 1974; and Luyckx and Lefebvre, 1972). However, recent evidence 54 suggests that changes in the insulin/glucagon ratio are not the sole determinant of hepatic glucose production during exercise. If plasma insulin and glucagon are maintained at basal levels by glucose infusion, hepatic glucose output still increases two to three times during exer— cise (Felig et al., 1974a). When hyperinsulinemia (more than 100 micra units/ml) is maintained by infusion of insulin, exercise still stimu- lates hepatic glucose output (Felig and Wahren, 1975). It is suggested that catecholamines which are released under these circumstances, in- fluence the hepatic glucose output during exercise. It is generally accepted that insulin secretion is blocked by catecholamines during exercise (Pruett, 1970); for that reason, during exercise of long duration, the enhanced glucose utilization must be attributed to a factor(s) other than insulin (Oseid and Hermansen, 1971). The glucose response to work of long duration depends upon both the intensity and duration of the work. Wahren et al., (1971) suggested that the blood insulin level decreases in order to preserve glucose for the brain; however, it is evident that glucose uptake in the muscle is stimulated despite decreased insulin levels (Metivier, 1973). During short-term, intermitant exercise, Hermansen et a1. (1970) found that both plasma immuno reactive insulin (IRI) and blood glucose increased (four-fold and two-fold, respectively). Similar increases were seen during glucose infusion, so under such conditions (bouts of maximal exercise), insulin secretion is not inhibited. Hansen (1971) compared the response of diabetics to nondiabetics during relatively short exercise (20 minutes). The diabetics all showed bl; 10: 1'10 pl eI' 55 an immediate, pronounced increase in growth hormone levels in the blood, but there was no change in the nondiabetic controls. During long-term exercise, Ericksson et a1. (1971) found a significant in- crease in plasma growth hormone. Hartog et a1. (1967) found no change in growth hormone until after 20 minutes; it then rose rapidly until 60 minutes, afterwhich it fell progressively. In these cases, growth hormone seemed not to be triggered by blood lactate plasma FFA or the plasma level of glucose. Buckler (1972) has shown that growth hormone secretion during exercise appears to depend upon cumulative effects. Hansen (1970) infers that there is a humoral factor released from the working muscle that triggers growth hormone release. Alanine presumably could serve such a role, and it deserves to be further investigated. Schalch (1967) suggests that growth hormone has an insulin—like effect (short-term exercise) and a lipolytic effect (long-term exercise). While the latter has been shown (Gerich et al., 1976b), there appears to be no direct evidence of the former, although there is much suggestive evidence. There is strong evidence to suggest that the corticosteroids play an important role in the energy supply during long-term exercise (Hazar et al., 1971). Others, however, have found a decrease of l7-hydroxy— corticosterone during exercise (Viru and Oks, 1972), which may be due to distribution of the hormone in a larger tissue volume (Cornil, 1965). Metivier et a1. (1973) found an increased secretion of growth hormone as work intensity increased; but though the results on cortisol are difficult to interpret, cortisol appears to be important during heavy exercise. 10'. 55 as EC DE ti C. (1") (I) 56 2. Aldostrone, cortisol and plasma renin activi_y, During pro- longed, heavy exercise, a somewhat variable increase in plasma renin activity (PRA) is seen (Bozovic et al., 1967), but during exercise of shorter duration, the rise in PRA is related to the work load as well as to plasma norepinephrine (Kotchen et al., 1971). Since it is well known that exercise decreases renal blood flow (Castenfors, 1967; and Grimby, 1965), it appears that exercise causes increased sympathetic nerve activity, leading to decreased renal blood flow, which in turn stimulates renin release. Sundsfjord et a1. (1973) found that long-lasting exercise (4.5 to 7 hours) increased aldosterone (26 to 134 pg/ml), PRA (0.20 to 3.30 ng aug. I/ml/h), and cortisol (12.6 to 34.9 g/100 ml); but they found a relatively poor correlation among these variables. Aldosterone secre— tion is controlled both by the renin system and ACTH. Since cortisol increased, ACTH must also have increased. Because angiotensin II inhi- bits cortisol production, cortisol could not have increased without stimulation from ACTH (Rayyis and Horton, 1971). Hepatic blood flow is decreased during exercise (Rowell et al., 1964); and since aldosterone is primarily cleared by the liver, plasma aldosterone will increase be— cause of decreased removal by the liver (hepatic extraction of aldoster- one is almost complete). This means that there are different factors acting simultaneously on the aldosterone level which can explain the great individual variation in this parameter. As serum potassium levels increased only slightly (from 4.38 to 4.71 meq/L), the direct stimulatory effect of serum potassium does not appear to be responsible for the increase in serum aldosterone levels. min c maxi: Howe‘ TracI infl the 03 C the both does and that SEC} 57 The 125I-o—iodohippurate clearance decreased from 538 g plasma/ min during rest to 343 g plasma/min during exercise (2.5 hours at 60% maximal oxygen uptake), and aldosterone clearance also decreased. However, equilibrium (isotope tracer) was not obtained during exercise. Tracer studies showed that duration of exercise as well as work load influences the cortisol level. A decrease in the latter was seen during the first hour, which may indicate increased distribution or metabolism followed by increased secretion, or decreased metabolic clearance rate on continued exercise. Sundsfjord et a1. (1973) have shown that during the first 60-90 minutes, the renin-angiotensin system is the primary stimulator of aldosterone (Metyldopa, a blocker of renin release reduced both PRA and PA during this time); but after this period, aldosterone does not increase further despite continued increase in PRA and ACTH, and decreased metabolic clearance. After about 90 minutes, it appears that the adrenal is unable to increase its aldosterone secretion; the secretion may even decrease. F. Hormone Regulation During Starvation or Carbohydrate Restriction During starvation or carbohydrate restriction, normal individuals show increased glucagon levels (50 to 100% above the postabsorptive concentration) (Muller et al., 1971), and a decreased insulin level. Thus, the insulin/glucagon ratio (molar) falls to about 0.4, the same as seen during severe exercise (Unger, 1971). This ratio maximizes the hepatic glucose production, although during short—term starvation, glucocorticoids also appear to be important for gluconeogenesis (Felig, 1975), and growth hormone seems to limit it during long—term starvation 58 (Hunter, 1972). Unger and Orci (1976) state that the low insulin/ glucagon ratio initially promotes glycogenolysis and when the glycogen stores are depleted, enhances gluconeogenesis and eventually ketone production. However, Nilsson et a1. (1973) found that there was no increase in gluconeogenosis on a carbohydrate free diet, despite depletion of glycogen, and that gluconeogenesis was the same as in the postabsorptive state. Presumably this can be a difference between total starvation and carbohydrate starvation. G. Summary While glucose stimulates insulin, AA primarily stimulate gluca— gon secretion; however, the presence of glucose decreases the effect of AA on glucagon while increasing the stimulatory effect of AA upon insulin secretion. Insulin promotes glucose utilization and stimulates protein synthesis, whereas glucagon promotes hepatic glucose output. Physiologically, it is the insulin/glucagon ratio that is important. A decreased insulin/glucagon ratio is a necessary factor in ketone production. Physiological levels of FFA and ketones do not appear to have any effect on the insulin/glucagon ratio in man. Glucagon alone, stimulates lipolysis in man; but it also stimulates insulin secretion, which counteracts this effect. Insulin inhibits lipolysis. Glucagon does not stimulate the output of most AA from the muscle. When the insulin level is low, growth hormone increases lipolysis and ketosis in man. Growth hormone secretion increases during prolonged starvation and exercise of long duration. T enhances protein catabol- 3 ism, but does not increase during starvation; however, there appears to 59 be a shift from the active to the inactive form, which is also seen on a high fat diet. Upon refeeding a high carbohydrate diet, this is reversed. Glucocorticoids also increase gluconeogenesis from proteins, but the effect may be indirect through secretion of glucagon. During exercise, serum insulin concentration decreases, which increases hepatic glucose output, first from glycogenolysis (most sen- sitive) and then, from gluconeogenesis; during heavy and prolonged exercise, serum glucagon also increases. Catecolamines released during exercise appear to have a direct stimulatory effect on the glucose out- put from the liver. Growth hormone is released during exercise, and besides a lipolytic action, it seems to have an insulin-like action. There are changes in the aldosterone secretion during exercise, but the significance of this does not seem clear. During both long-term and short-term starvation as well as severe exercise, the insulin/glucagon ratio falls to 0.4, which maximizes the hepatic glucose production. U) be IE dr We je, the the 1.5 PART TWO: EFFECTS OF LOW CARBOHYDRATE DIETS VI. EFFICIENCY AND HIGH FAT DIET A. Substrate Efficiency, Zuntz (1911) concluded that the working muscle can use equally well the major sources of energy (fat and carbohydrate). From the studies of Fletcher and Hopkins (1971), Krogh et a1. (1920) concluded that carbohydrate should be more efficient if the theories of Fletcher and Hopkins were correct, since according to their study, fat must first be transformed to compounds closely related to carbohydrates. For this reason, Krogh et al. (1920) decided to test the results of Zuntz (1911). The subjects were fed either a high fat diet or a high carbohy— drate diet. The composition of the diets is not given, but the major foods eaten are listed, and from them it can be concluded that they were diets very high in fat and carbohydrate, respectively. The sub- jects were fed each diet for four days; and they performed uncontrolled exercise during the first two days on the high fat diet, in order to deplete their glycogen stores. Two subjects started with the high carbohydrate diet and two with the high fat diet, in order to minimize the training effect. Both diets were low in protein. The work was performed for two hours on the bicycle ergometer in the postabsorptive state. Results were collected only for the last 1.5 hours. Before exercise started, the resting oxygen consumption was 60 61 determined for ten minutes, after twenty minutes of complete rest (sitting in a chair). The efficiency was calculated as the workload divided by the oxygen consumption during exercise minus the consumption during rest. The oxygen consumption was converted to the caloric equivalent based on the RQ. Their results showed that the muscular efficiency was about 10% higher on the high carbohydrate than on the high fat diet. They also found an increasing efficiency with training; and interestingly, they found (by comparing results during different times of the same exercise period) that the energy requirement increased as fatigue approached. That is, at exhaustion the efficiency was less than in the non-exhausted condition. This effect of fatigue showed itself much earlier when the high fat diet was fed (low RQ) than when the high carbohydrate diet was fed. The resting metabolism also depends upon diet, being lowest on a mixed diet, and increasing about 5% on a high fat diet and about 3% on a high carbohydrate diet. Both the net energy expenditure per unit of work and the ratio of fat to carbohydrate catabolized are linear functions of the RQ; so regardless of the amount of fat oxidized, Krogh et a1. (1920) concluded that 11% of the calories supplied by fat was wasted. When work starts, the respiratory quotient changes. If it is close to one, it decreases, and if it is very low, it increases. On this basis, it was suggested that when the RQ was below 0.8, carbohy— drate was formed from fat; and when the RQ was above 0.9, carbohydrate was converted to fat. 62 Marsh and Murlin (1928) made a similar study of the efficiency for one subject fed a high fat diet. The compositions of their diets were: high carbohydrate diet had 8% of calories from fat, 80% from carbohydrate, and 12% from protein; and the high fat diet had 80% from fat, 12% from carbohydrate, and 8% from protein. The high fat diet provided a slightly higher caloric content (2800 kcal/day compared to 2400 kcal/day). In these experiments, as in those of Krogh et a1. (1920), a bi- cycle ergometer was used for the exercise, but the work period was only eight minutes compared to two hours in Krogh's experiments, and it was done twice a day during the seven to eleven days each experimental period lasted. First they measured the efficiency when the normal diet was consumed, then followed a high carbohydrate dietary regimen and a high fat dietary period. The high carbohydrate and high fat period was repeated. They collected urine and calculated the nonprotein RQ. Furthermore, the urine was collected for one hour immediately before exercise (pre-work); and the urine collected during the work period in— cluded the recovery period and extended into the postdwork resting period. The average efficiencies were: 22.1% for the normal diet, 22.7% for the high carbohydrate diet, and 21.5% for the high fat diet. During the high fat dietary regimen, there was a progressive decrease in efficiency from the third or fourth day until the end of the period. This was not seen during the high carbohydrate dietary regimen. Extended to pure carbohydrate, they found fat to be 11-12% less efficient than carbohydrate. 63 They also found that the urinary nitrogen excretion decreased from the pre—work to the work period, with about 50—150 mg/hr; but not as much as when the high fat diet was consumed (about 100 mg/hr less). They concluded that this was due to a protein sparing effect of the glycogen mobilized during exercise. Less glycogen is available when a high fat diet is consumed (Nilsson et al., 1973). The urine excretion of nitrogen is also decreased during exercise (Rougier and Babin, 1975), whereas sweat excretion increases (Cerny, 1973), making an interpretation difficult. B. Endurance Christensen and Hansen (1939) compared the endurance during a high fat diet, normal diet, and a high carbohydrate diet. The composi— tion of these diets were: high carbohydrate, 4107 kcal—rprotein 5%, fat 3%, and carbohydrate 83%; high fat diet, 5448 kcal—~protein 2%, fat 95%, and carbohydrate 3%. These data are calculated from the diets listed for two specific days, and may not necessarily be the average (not given). As Krogh et al. (1920) had found, work decreased the high RQ, but increased the low RQ. The RQ decreased as work progressed, particu- larly on the high carbohydrate diet. Substrate utilization based on the RQ (not corrected for protein) showed that in the beginning, carbo- hydrate contributed about 75% of the energy, but the proportion contri- buted by fat increased to 70% at the end of exercise (that is, point of exhaustion). Christensen and Hansen (1939) also found a gradual increase in the oxygen uptake which became much steeper as the point of exhaustion 64 was approached; this was associated with a fall in blood glucose level. Calculation based on the RQ showed that at the time of exhaustion on the high carbohydrate diet, about 400 g of glucose had been used for energy compared to about 50 3 when the high fat diet was consumed. During the high fat diet, fat contributed up to 90% of the substrate, and the endurance was only about 1/3 of that seen during a high carbo- hydrate diet; there was a significant drop in blood glucose level at the time of exhaustion. They found that the oxygen uptake per minute fell some at very high RQ values, but otherwise, it was constant above a RQ of 0.80. When the RQ was between 0.80 and 0.74, there was a very steep increase in the oxygen utilization per minute. The oxygen uptake increased 10 to 15% as the RQ fell from 0.8 to 0.75. It must be pointed out that if the increase in oxygen consumption is due to lowered efficiency of fat as substrate, and if the RQ accurately reflects the kind of substrate used, one would expect to see a straight line relationship between the RQ and the oxygen uptake; but experimentally, it is more like an all or none type response. That is, as glycogen is depleted, the oxygen uptake/min increases or the efficiency decreases. Finally, it should be mentioned that the oxygen consumption was not related to ketonuria; that it was just as high during periods of low ketonuria as when keto- nuria was at a high level. In subjects fed a single meal of fat and protein, Goldsmith et a1. (1971) found a 10% increase in the oxygen consumption compared to the oxygen consumption in the postabsorptive state the same day; 65 whereas no increase was found after a carbohydrate meal. C. Starvation and Work Performance Henchel et al. (1954) studied the ability to perform hard work during acute starvation. During a 2.5-day fast and a 5-day fast, they found a weight loss of 4.5 kg and 5.5 kg, respectively. The work was walking on a treadmill at a 10% grade at 3.5 mph for four hours/day and three hours/day, respectively. There was no deterioration in work capacity the first day; but during work on the second day, blood glucose fell about 20% and the heart rate increased about 10%. The mechanical efficiency decreased about 6% during the 2.5-day fast. There was a definite increase in blood lactate and a 10-minute oxygen debt in the 5-day fast. Fitness generally deteriorated as the fast progressed. An increase in urine and blood acetone concentration in connection with the weight loss is taken to indicate that the acidoses causes dehydration; which in turn causes a loss in efficiency. However, the weight loss during the 5-day fast was not much higher than during the 2.5—day fast, despite the fact that blood ketones are known to increase progressively until the end of the second week of starvation (Cahill et al., 1966). Furthermore, a weight loss of this order can be partially explained by depletion of glucogen stores (Olsson and Saltin, 1970). The effects of consecutive fasting (five to six weeks apart) have been studied by Taylor et al. (1945). The subjects maintained their blood sugar levels better during work in the fifth period (second and third days of fasting) than in the first period. The urine acetone and the urinary nitrogen excretion were higher during the fifth fast 66 compared to the first period of fasting; however, the increased urinary nitrogen excretion was not consistent in all subjects, so increased gluconeogenesis is not the sole factor in the adaptation to fasting. D. Possible Explanation for Decreased Efficiency From the previous results, it is seen that the oxygen consumption is indeed increased when fat is the major substrate, but it does not appear due to a decreased deficiency from using fat as a substrate, since the efficiency does not decrease until the carbohydrate intake reaches a certain low level; and the work of Christensen and Hansen (1939) does not show any relation to ketosis. Hultman and Bergstrom (1967) indicate that the decreased efficiency is due to glucogen deple- tion. Krogh et al. (1920) suggest that the changes in R0 from rest to exercise are due to conversion of fat to glucose during a high fat di— etary regimen. There is no evidence for such a conversion, but gluco- neogenesis from protein probably does occur as well as glycogenolysis, which can alter the RQ. On the high carbohydrate diet, the decrease in the RQ during exercise is ascribed to lipogenesis, which presumably can occur, since on a high carbohydrate diet a large fraction of the glucose is converted to fat. Such interconversions, if they do occur during exercise, could significantly alter the energy requirement for a given task (Milligan, 1971). Using animal models, the in vivo energy yield from different substrates such as glucose and fat, has been shown to be proportional to their calorimetric values, indicating that the efficiency of utilization is equal, or, that the oxygen consumption of 67 fat or carbohydrate should correspond to the theoretically predicted consumption (Milligan, 1971). When the relation between energy intake and energy retention was studied in animals, both below and above maintenance levels, fat was found to be used more efficiently, resulting in a smaller increment of heat than does carbohydrate. Protein was used much less efficiently than was carbohydrate, this conclusion was true irrespective of species or feeding level. At below maintenance level, the energy must be entirely used to support body function, and therefore must be related to the efficiency of the particular substrate used. When the effi- ciency found calorimetrically was plotted against the efficiency calculated biochemically relative to glucose, a very good agreement was found except for ketone bodies, acetic acid, and butyric acid. The explanation given for this is that as a basis for the calculations, constant turnover was assumed; and that is not valid when butyric- and acetic acids are given. The animals developed acidosis and ketosis, blood sugar fell precipitously, and urinary nitrogen excretion increased. As a small amount of propionic acid prevented these changes, it seems likely that gluconeogenesis from proteins was increased, and resulted in the lower efficiency (Blaxter, 1971). There appears to be a relationship between the percentage of fat, carbohydrate and protein in the diet, and the efficiency of energy utilization in experimental animals. At low and high intakes, the ratio of fat to carbohydrate plays a role in the energy utilization. At high protein levels, a high fat/carbohydrate ratio increases the energy 68 utilization; but at low protein levels, a high fat/carbohydrate ratio depresses energy utilization. At intermediate protein levels, there was no effect of the fat/carbohydrate ratio (Hartsook and Hershberger, 1971). A similar relationship, if it holds in man, may be important, since most of the studies of efficiency have been done at low protein intakes. E. Summary There is not sufficient evidence to support the idea that fat is used less efficiently than carbohydrate as substrate; but the de- creased efficiency seen on a high fat diet is probably due to other factors. One explanation given above is that it is due to gluconeogene- sis, but other explanations will be considered. It is well established that a high fat diet can increase the oxygen consumption, and decrease the time to exhaustion. It has also been shown that work performance deteriorates during starvation. A high fat diet and starvation both cause increased blood levels of ketones and a depletion of the body stores of glycogen, and may cause dehydra- tion. The decreased efficiency generally ascribed to fat as substrate may therefore be due to a variety of factors, such as dehydration, acidosis, glycogen depletion, gluconeogenesis from protein, or a combi- nation of these. VII. EFFECT OF DEHYDRATION ON EXERCISE It has been suggested that ketosis produces dehydration, and that this is the reason for the poor exercise performance during 69 starvation (Henschel et al., 1954). Pitts et al. (1944) studied the effect of different dehydration levels during exercise on a treadmill. The subjects received water ad lib, water consumption equivalent to sweat loss, or no water during exercise. The performance as evaluated by rectal temperature was best when the water intake corresponded to replacement of the sweat loss. Apparently the subjects did not volun- tarily consume enough water. In studies with restricted caloric intake but adequate water consumption, maximal strength was not affected until the weight loss reached about 10% in both acute and chronic starvation (Keys et al., 1950; and Taylor et al., 1945). Bosco et a1 (1974) found a loss in endurance when dehydration was caused by starvation as well as by water deprivation, but there was also a significant decrease in performance of the control group; so other factors such as motivation must be involved. Saltin (1964b) studied dehydration caused by heat exposure, exercise, and both. Dehydration was continued until the subjects reached the same weight loss by each method. The performance was com- pared before and after dehydration at different work loads. Under no circumstances was the oxygen uptake (of the same.workload) affected by dehydration, but at the lower workloads, the heart rate increased (mean of 13 beats/min). However, at maximal load, there was no change in heart rate, but the time to exhaustion decreased markedly. Kozlowski and Saltin (1964b) studied the sweat loss and water distribution during thermal dehydration and exercise dehydration. 70 TBW (tritium space), ECF (insulin space), and PV (Evans blue) were determined before and after dehydration; sweat loss was also determined. The average weight loss was 4.1% or 3.1 kg, and the decrease in ECV was 1.4 and 0.2 liters respectively. The decrease in plasma volume was 0.7 and 0.1 liters respectively. There was no significant difference in the electrolyte loss, which was almost entirely eliminated through sweat. Na+ loss was higher during dehydration, whereas Kf loss was higher dur- ing exercise. There was no significant difference in the composition of sweat and urine during the two types of dehydration. The small decrease in extracellular fluid during exercise was surprising; however, this has also been found in prolonged heavy exercise (Astrand and Saltin, 1964). The energy use during dehydration and exercise was 500 (for the time required to loose the desired weight) and 3,100 calories respectively (the calory consumption was calculated from the respiratory quotient). In the case of exercise, this corresponds to an endogenous water produc- tion of 1.1 liters, but despite this water, there was a higher intra- cellular water loss during exercise dehydration. The increase in potassium in the plasma after exercise was not due to increased protein catabolism, but it might be due to glycogenolysis (Saltin, 1964a,b). It was found that a given sweat loss due to exercise dehydration is more detrimental to physical work capacity than the same sweat loss due to thermal dehydration; and exercise dehydration is primarily intra- cellular as compared to thermal dehydration, which is primarily extra- cellular (Saltin, 1964a and 1964b; and Astrand and Saltin, 1964). 71 Costill and Saltin (1974) found that during dehydration, the RBC shrink and thus, the venous hematocrit and the change in PV is under- estimated. Costill et al. (1974a) found that the percentage change in mean corpuscular volume (MCV) is accurately described by changes in the mean corpuscular hemoglobin concentration (MCHC). Taking change in the MCV into account by measuring MCHC, Costill et al. (1974a) found that during onset of exercise, PV decreased 12.2% (due to transcapillary fluid flux in the muscle); but during the final 110 minutes, PV only decreased an additional 3.6%. Thermal dehydration resulted in only a 3% decrease during the first 10 minutes of heat exposure, but a 15% decrease during the remaining period. The difference may be due to degradation of glycogen. During onset of exercise, muscle water in— creased 9%, but this disappeared after two hours of exercise. During the first 30 minutes of recovery after thermal dehydration, PV in- creased 7%, and the rectal temperature decreased. This suggests that during thermal dehydration, water enters the dermal tissue because of peripheral vasodilation and sweating, and then re-enters plasma during recovery. The decrease in PV was about the same in thermal and exercise dehydration in contrast to the results of Kozlowski and Saltin (1964). Serum protein concentration reflects changes in PV following thermal dehydration, but not after exercise dehydration (Costill and Sparks, 1973).. Costill and Saltin (1973) concluded that the decreased perform- ance during exercise dehydration compared to thermal dehydration is due to glycogen depletion. That ionic changes occur in the contracting muscle was shown by Hodgkin and Horowitz (1959). During exercise of relatively short 72 duration, potassium is lost from the muscle (Ahlborg et al., 1967b), but apparently not after about one hour of prolonged exercise (Haralambie, 1973a). An increase of extracellular fluid in the muscle has also been observed (Ahlborg et al., 1967b), which may explain the increased plasma protein and hematocrit seen during exercise (Bergstram et al., 1971). The change in muscle potassium has been suggested as a limiting factor in exercise (Streter, 1963). Bergstram and Hultman (1966) found a correlation between potassium and the glycogen content: 0.5 mEq potassium accumulates for each gram of glycogen deposited. This potassium is liberated from the muscle when glycogen is broken down (Bergstrom et al., 1967). Olsson and Saltin (1970) found an in- crease in total body water of about 2.5 kg during a high carbohydrate dietary regimen following work. Bergstram et al. (1971a) concluded that the water and electrolyte changes following exercise are not of such magnitude as to limit exercise. The changes in intracellular water and electrolytes seen during exercise can be explained largely by an accumulation of lactate and other metabolites in the muscle fibers (Karlsson, 1971a). There is a good correlation between lactic acid accumulation and the increase in intracellular water seen during exercise (BergstrSm et al., 1971a). As glycogen is depleted during long-lasting exercise, and fat supplies most of the energy, the lactate gradient probably disappears and this can explain the redistribution of PV seen in the beginning of exercise, but not later on during exercise (Costill and Fink, 1974). It appears that there is some difference between thermal- and exercise dehydration; 73 the latter gives a much more severe impairment of endurance, probably due to depletion of glycogen. Water loss as such, unless severe, does not particularly restrict exercise. Summary. A sweat loss of about 10% is detrimental to physical performance, but it makes a difference whether this sweat loss is obtained through exercise or thermal dehydration. Performance is most affected by exercise dehydration, and this is probably caused by a decrease in the glycogen stores. The performance decrease seen during starvation, may also be due to depletion of glycogen stores. It has been suggested that the ketosis of starvation or a high fat diet causes dehydration through a loss of sodium in the urine. VIII. WATER BALANCE AND CARBOHYDRATE RESTRICTION A. Weight Loss When obese patients were put alternately on 1,000 calories high fat, high carbohydrate, and high protein diets, they lost weight much more rapidly on the fat and protein diets than on the mixed diet, and they maintained weight for a few days while consuming the 1,000 calorie high carbohydrate diet when it followed the high fat or high protein reducing diet. The absorption of fat was above 90%, eliminating fecal loss as the explanation. The nitrogen balance was about the same on the high fat and high carbohydrate diets (that is, slightly negative on the average). During the high protein regimen, a definitely positive balance was observed at all times. The obese subjects showed resistance to hypoglycemia, severe ketonemia and acidosis while consuming the high 74 fat, low caloric diet, compared to nonobese subjects. The insensible water loss was higher during the high fat and high protein diets than during the high carbohydrate diet. Kekwick and Pawan (1957) explain their results as due to differences in metabolism between obese and non- obese subjects. Benoit et a1. (1965) compared the effect of starvation to a 1,000 calorie ketogenic diet and a 1,000 calorie mixed diet as control (half the subjects fasted first, and half consumed the ketogenic diet first). The mean weight loss during the lO-day periods was 9.6 kg, and 6.6 kg for the fast and ketogenic periods respectively. They found the negative nitrogen balance during the ketogenic period comparable to that of the mixed diet; and both much less (three times) than during total starvation. They also found the increase in plasma and urinary ketone bodies to be greater during the ketogenic diet than during starvation; and a negative potassium balance was seen only during the starvation period. A calculation of the composition of the weight loss for fasting gave 64.6% lean body tissue and 35.4% fat; and for the ketogenic, only 3% lean body tissue. The main point of Kekwick and Pawan's study (1957) is that the body water/body weight ratio remained unchanged during the different dietary periods, and that the difference in weight loss therefore, was due to metabolic alterations. However, a look at their water-balance data clearly indicates that their accuracy was not good enough to detect a shift in water-balance (Olesen and Quaade, 1960). 75 B. Reason for Weight Loss Olesen and Quaade (1960) studied the weight loss in obese women fed a 2,000 calories (25 g CHO) high fat diet, and found a remarkably high weight loss during the first 5—7 days; whereupon, the weight became stabilized for the remainder of the period (13 days). During the last 5-7 days when all the patients were fed a 1,250 calorie high carbohydrate diet, the weight remained constant even though caloric intake was below their measured basal metabolic rates. The constancy of weight must be due to water retention and possibly glycogen storage. The authors concluded that the subjects must have been dehydrated during the high fat diet, as indicated by the initially rapid weight loss. The only patient who was able to consume the 2,600 calories high fat diet (as used by Kekwick and Pawan, 1956) initially lost weight, but after one week, gained. Pilkington et a1. (1960) repeated the experiments of Kekwick and Pawan (1956), but for longer periods (24 days), and found no difference in the weight loss in the subjects when fed a 1,000 calorie high fat or high carbohydrate diet. However, body weight increased when the subjects switched from the high fat to the high carbohydrate diet, and decreased when the change was reversed. These deviations lasted about 10 days and amounted to approximately 2.5 kg; the rate of weight loss after the first 12 days was independent of diet. During the 85% high fat, 1,000 calorie diet, the patients complained about being tired and nauseated at the end of the study, but felt invigorated, yet hungry, after consum— ing the 91% carbohydrate diet. The variations in weight when changing 76 diets was found to be due not to alterations in salt content of the diets, but rather to variations in fluid balance. Werner (1955) found no alteration in weight when changing be- tween a high fat and a high carbohydrate diet, both providing 2875 calories, apart from minor fluctuation in salt and water-balance at the beginning of a dietary period. Russell (1962), though altering the proportions of fat, carbohy- drate, and protein, kept total calories, sodium and water constant. During the high carbohydrate diet, the subjects' weight loss was minimal, while sodium and water-balance were positive; however, during the high fat and protein regimen, they showed an increased weight loss, and the water and sodium balance became negative. Increased water excretion (urinary) does not change the tonicity of body fluids, since it is accompanied by Na+ excretion. It seems well established that the short-term increase in weight loss of subjects fed a high fat diet is due to a loss of body water and not to thermogenesis as claimed by Kekwick and Pawan (1957); however, Kasper et al. (1973) support their conclusion. They found that surfeit feeding of a high fat diet (corn and olive oil up to 6,800 calories per day) gave much less weight gain than expected from increased calorie intakes; and also a high weight loss on a diet high in fat but equal in calories compared to one high in carbohydrate. Recently, Yang and Van Itallie (1976) compared the effects of total fasting, an 800 kcal mixed diet, and an 800 kcal ketogenic diet during 10-day experimental periods. The energy-nitrogen method was 77 used to quantify the weight loss. The subjects' weight loss was greatest during starvation and intermediate during the ketogenic diet. The nitrogen excretion was a little higher during the ketogenic diet than the mixed diet, but much higher than both during total starvation. Although the percent of fat and water lost was very similar for the ketogenic diet and starvation, the protein loss was much higher during starvation. During the mixed diet, more fat and much less water was lost. The authors concluded that the energy value of body constituents lost during the different periods was almost independent of the dietary regimen, and that the differences in weight loss were almost entirely due to differences in water-balance. This, however, does not neces- sarily mean that a ketogenic diet causes dehydration. The fat stores appeared to be used primarily on the 800 kcal carbohydrate diet, and they are associated with far less water than are the protein and glyco- gen stores used during the ketogenic regimen. Yudkin and Carey (1960) gave an alternative explanation for the greater weight loss seen when a high fat diet is fed. They stated that carbohydrate makes the patient feel hungry, whereas fat gives a feeling of satiety; therefore, the patient eats less on a high fat diet. While this may be the case when patients choose their own food, it cannot explain the shift in water-balance seen in experiments where the caloric intake is controlled. C. Water Loss DuringgGlucose Dgprivation That water is lost during carbohydrate deprivation seems clear; but the mechanism whereby this occurs is not. During carbohydrate 78 starvation, glycogenolysis takes place, and this is associated with a significant water loss, since glycogen is stored together with water. When refeeding patients a high carbohydrate diet after depletion of glycogen stores, Olsson and Saltin (1970) found a 2.5 kg weight gain, most of which was probably water, since more than 3 g of water are liberated when 1 g of glycogen is metabolized (Bergstrom and Hultman, 1972). However, the muscle glycogen stores are only slowly depleted during a carbohydrate-poor dietary regimen of rest (Hultman and Berg- strom, 1967). So the amount of water released from glycogen depends upon physical activity and the composition of the previous diet; but at most, it can amount to about 2.5 kg (for a 70 kg man). Cahill (1976) states that during the gluconeogenic phase of starvation (first week or so), up to a pound of lean body mass may be lost per day. The loss of lean body mass necessarily gives a corresponding increase in urinary (or sweat) nitrogen loss. Yang and Van Itallie (1976) found an average protein loss in their subjects of 50 g/day for 10 days of starvation corresponding to a total loss of lean tissue, and glycogen of about 5.0 kg. The fat loss was given as 32.4%, corresponding to 2.4 kg when calculated on an energy deficit basis. Thus, during starvation the glycogen, lean tissue, and fat used can explain the total body weight loss (7.5 kg). Similar calculations made for their ketogenic diet indicate that the mean weight loss was 4.67 kg and the loss from protein, fat and glyco- gen (assuming 2.5 kg lost through glycogen depletion) was 4.73 kg. Considering the uncertainty of the glycogen loss, there is very good 79 agreement. If we look at the weight loss during the mixed diet, the actual weight loss was 2.78 kg for the 10 days, and assuming no glycogen loss (which is not necessarily true), the calculated loss is 2.15 kg, which is little more than a pound below the actual loss. This differ- ence can probably be explained by a difference in the size of the glycogen losses. These calculations show that under the conditions of this study (Yang and Van Itallie, 1976), the weight 1033 possibly can be explained without assuming any dehydration or shift in the water-balance; however, it is uncertain how much glycogen was actually lost in this experiment. The short-term weight loss for obese people during starvation is from 10 to 30 pounds, unless they have previously been on carbohydrate restricted diets, whereas for normal man, the initial weight loss is about four to six pounds. While the weight loss in normal persons may be due to glycogen, lean body mass, and fat used for energy, the enorm- ous weight loss reported in obese patients apparently is due to saline diuresis (Cahill, 1976). D. Possible Mechanisms for Altered Water Balance on Low Carbohydrate Diet Carbohydrate starvation is known to produce ketosis (Owen and Reichard, 1971a). Since the renal excretion of ketone bodies induces an obligatory loss of sodium (Gamble et al., 1923; and Sartorius et al., 1949), and since sodium is the main regulator of the extracellular volume, it is assumed to cause dehydration (Bloom, 1967). There are two major objections to this theory. First, the in- creased excretion of sodium should not lead to a loss of body water 80 unless the sodium intake is restricted. Actually, the weight loss dur— ing a salt free carbohydrate free dietary regimen is no greater than during the same diet containing salt (Bloom, 1967). Spark et a1. (1975) have shown that the administration of NaCl or mineralcorticoids has no effect on the weight loss during starvation. It has also been shown that there is a relationship between urinary sodium loss and ketone excretion (Hood et al., 1970). Secondly, the development of ketosis is gradual, reaching maxi- mal splanchnic production the third day of carbohydrate starvation (Garber et al., 1974), and blood levels continue to rise until the second week (Cahill et al., 1966). Sodium excretion, however, is high- est the first day of starvation, and then gradually decreases over the next three to four days (Bloom, 1967; and Gamble, 1923). During the first 24 hours, the sodium loss amounts to 50 to 250n1Eq/day, and then gradually decreases over the next 10 days to as low as l m Eq/day (Bloom and Mitchell, 1960). The decrease in sodium excretion is paralleled by a decrease in urinary nitrogen; the magnitude of this reaction depends on the level of protein ingestion prior to the fast (Gilder et al., 1967). Stinebaugh and Schloeder (1966a) tried to separate the effects of salt restriction and fasting by first feeding a low sodium diet (14 m Eq/day) until the sodium excretion stabilized at less than 20 m Eq/day. The urinary sodium excretion during fasting still increased, showing a peak the fourth day and decreasing to base level about the 10th to the 12th day. So sodium loss during fasting appears to be due partly 81 to salt withdrawal, and partly to the urinary loss of nitrogen—contain- ing compounds. In another study, Stinebaugh and Schloeder (1966b) tried to separate the effect of acidosis and fasting by feeding ammonium chloride. They first gave a low sodium diet as before, until urinary excretion was below 20 m Eq/day. Ammonium chloride was then added to produce acidosis similar in degree to that seen during total starvation; when sodium excretion fell to base level, fasting began. In this way they found 338 m Eq of sodium (lost before ammonium chloride was given) due to salt withdrawal, 190 m Eq acidosis (caused by ammonium chloride), and 153 m Eq due to the fast per se. However, if the ammonium chloride produces a mild sodium depletion and fasting produces a sodium surplus from tissue degradation, the kidneys would be expected to retain this surplus sodium during the fast. Veverbrants and Arky (1969), and Haag et a1. (1967) found that the administration of bicarbonate did not alter the sodium excretion during starvation; and Gamble (1947) stated that keto acids do not remove fixed base from the body during starva- tion. The same water loss is generally reported when a diet which is adequate calorically, but low in carbohydrate is fed; such a diet is generally far from restricted in salt. This suggests that if there is any water loss above that from tissue degradation, it appears to be due to carbohydrate deprivation. It has been demonstrated that the medullary protion of the kid- ney uses glucose for energy production. A large proportion of the energy expenditure is involved in sodium transport (Ullrich and Marsh, 82 1963). Wright et a1. (1963) have shown that the active sodium transport in the medullary part of the nephron is deficient during fasting. He suggests that there is not enough glucose available for sodium transport during fasting. The kidney is a major gluconeogenic organ, but it assumes this role gradually, after ketosis becomes severe (Owen et al., 1969). The glucose in the blood may not be available to the kidney during starvation when insulin is low, and the water loss during starva- tion appears to be dependent upon a decrease in the insulin/glucagon ratio (de Fronzo et al., 1975; and Spark et al., 1975). Hoffman (1973) has shown that there is a separate compartment of ATP in the cell wall which is used preferentially for one component of the sodium pump, and it appears that glycolysis is necessary to supply this ATP. Haag et a1. (1967) found that prevention of sodium loss during starvation by feeding carbohydrate is not dependent upon its effect on ketosis, as has also been found by Katz et al. (1968). (Sodium loss during fasting can be interrupted by feeding proteins without prevent- ing ketosis.) The osmotic effect of urea is not the reason for the sodium loss, since a high protein diet increases urea excretion but decreases sodium excretion. Presumably, proteins restore depleted muscle tissue, which retains sodium. E. Summary It is well demonstrated that the introduction of a carbohydrate restricting dietary regimen initially produces a greater than expected weight loss. This initial weight loss is primarily due to above normal water loss. Calculations based on results reported in the literature 83 show that the weight loss observed during starvation and ketogenic regimens can be explained by depletion of glycogen stores, lean tissue degradation, and loss of adipose tissue. Any possible water loss above that from tissue catabolism is in some unknown manner related to carbo- hydrate deprivation. IX. LIMITING FACTORS DURING EXERCISE A. Local Muscle Substrates l. Triglycerides. The local energy supplying substrates in the working muscle are ATP, phophorylcreatine (PC), glycogen, and trigly- cerides (Hultman and Bergstrfim, 1973). At 60% of the maximal oxygen uptake and with a work time of 90 minutes (mean), Froberg et a1. (1971) found a decrease in the triglyceride content of the working muscle (from 10.4 to 7.8 mmoles/gm wet muscle). Sixty-one percent of the energy was supplied by carbohydrates and 39% by fat. Of the 39% de- rived from oxidation of fat, 3/4 came from the triglycerides in the working muscle and 1/4 from the FFA in the blood. It appears that the FFA in the muscle serve a buffering function supplying FFA whenever the demand is higher than the supply from the blood. Fat has never been demonstrated to be a limiting factor during exercise (Hultman and Bergster, 1973). 2. Glycogen. The second largest potential energy store in the muscle is glycogen. It has been shown (Bergstrom et al., 1967b) that the work capacity is related to the glycogen stores at the beginning of exercise. At low work loads, glycogen was not limiting, since the 84 glycogen degradation was very slow and the stores are decreased very little (Saltin and Karlsson, 1971b). When the work level was 65 to 89% of the maximal oxygen uptake, glycogen was depleted at the time of exhaustion, and therefore appears to be a limiting factor. Above 89%, the muscle glycogen is not depleted at the time of exhaustion and therefore cannot be the limiting factor. At such high levels of exer- cise, muscle glycogen decreased rapidly at first, followed by a slower decrease; glucose-6-P and lactate increased markedly at first, then stabilized until exhaustion when a pronounced decrease was seen. Free glucose increased until the end of exercise when a decrease was seen (Bergstrom et al., 1971a). On this basis, Hultman and Bergstrom (1973) conclude that the rate of regenerating active phosphate is the limiting factor, and that the pronounced accumulation of lactate and therefore hydrogen ion, is a major factor in bringing about the decreased rate of glycolysis. Klausen et al. (1973) studied the effect of supermaximal exer- cise on glycogen utilization. Under such circumstances, part of the energy must come from anaerobic glycolysis. Bouts of supermaximal exer— cise were alternated with periods of submaximal exercise. In the begin— ning, blood and muscle lactate increased drastically, but as time went on blood and muscle lactate as well as glycogen decreased, and work time to exhaustion decreased as did the total oxygen uptake in the last bouts of exercise compared to the first. One explanation is that as muscle glycogen decreases, the muscle increasingly uses lactate as substrate, or the red oxidative fibers use the lactate formed by the white glyco- lytic fibers. Of course the liver extraction of lactate also increases 85 (Wahren et al., 1971). The decreased substrate availability in combina- tion with the increased intracellular lactate could limit the work done; however, these experiments seem to point to something besides muscle glycogen as a limiting factor. This could possibly be glycogen in the CNS. Essen et a1. (1973) have shown that lactate can be taken up by muscles even during severe exercise, and the magnitude of this uptake is related to arterial lactate as well as to intracellular lactate and glycogen concentration. 3. Proteins. Hultman and Bergstrom (1973) state that protein has never been shown to be used as an energy-producing substrate to any appreciable extent during exercise. It does, however, play some role. wahren et al. (1971) studied the splanchnic and leg exchange of substrates during four hours exercise at 30% of the subjects' maximal oxygen uptake. During such exercise, the glucose uptake by the leg at 40 minutes increased to ten times the basal value and to seventeen times basal value at 90 minutes. During the rest of the exercise, a slight decrease was seen, but after four hours exercise, the glucose uptake was still 12 times the basal value. Alanine was the only amino acid that showed a consistent release during exercise. The muscle output of alanine was slightly increased at 40 minutes, but had risen to three times the basal level after 240 minutes. Net uptake by the leg of the branched chain AA, serine, and citrulline was observed after 240 minutes of exercise. The uptake of oleic acid by the muscle was increased three-fold at 40 minutes of exercise, and it increased an additional 140% by the end of exercise. 86 The splanchnic glucose output increased two-fold at 40 minutes, and then remained constant until the end of exercise when a small decrease was observed. Splanchnic lactate uptake increased two-fold in the latter part of the exercise, and pyruvate and glycerol uptake increased three- to eleven-fold. Glycerol uptake was the result of a marked increase in the arterial concentration, but the increased lac- tate uptake was due to an increased fractional extraction. Alanine uptake increased 100% after 240 minutes as a result of augmented frac— tional extraction; an increased uptake was also seen for threonine, serine, proline and glycine. A significant splanchnic output of valine, isoleucine and leucine was observed after four hours of exercise. The splanchnic fractional uptake of oleic acid increased from 29 to 45%. These results show that at this level of work, glucose uptake by the muscle increases for the first three hours of exercise, but since fatty acid uptake also increases, the contribution of glucose as a substrate reaches a maximum at 90 minutes under these circumstances. After three hours of exercise, the fat contribution exceeded that of glucose. The combined uptake of glucose and FFA accounted for 65% of the energy utilization at 40 minutes, but at four hours, it had increased to 90%. Splanchnic glucose output changed little between 40 and 180 minutes, but a slow decrease was seen at the end of exercise. The total glucose output from the liver during the four-hour exercise was esti- mated to be 75 g; the maximal gluconeogenesis provided 15 to 20 g of the liver glucose output based on splanchnic uptake of lactate, pyruvate, 87 AA and glycerol. Since the average liver glycogen content in post- absorptive man is around 80 g, about 3/4 of the liver glycogen was mobilized during this exercise. On the basis of substrate balance, the hepatic gluconeogenesis increased from 25% in the postabsorptive state to 40% after four hours of exercise; and increased protein catab- olism is partly responsible for this. In many respects, the overall metabolic response to prolonged exercise is strikingly similar to that observed after three days of starvation. B. Liver Glycogen Hultman and Nilsson (1971), among others, have shown that hepatic glucose output increases during heavy exercise, and this in- crease is related to the work load and the duration of exercise. During the terminal part of the exercise, the glucose production increased markedly. At the end of heavy exercise, up to 50% of the glucose con- sumed by the muscle was released by the liver (Hultman, 1967). At that time, g1ucose-6-P was low in the muscle, which facilitates utilization of glucose from the liver, because g1ucose—6—P inhibits hexokinase (an enzyme for phosporylation of glucose). When strenuous exercise was performed after carbohydrate starva— tion (depletion of liver glycogen), a pronounced decrease in blood glu- cose occurred; and in some subjects, severe hypoglycemia was observed, which made them unable to continue the exercise (BergstrSm et al., 1967). This indicates that under certain conditions (low carbohydrate intake), the work capacity may be limited by the size of the liver glycogen store. 88 During three days of a carbohydrate-free dietary regimen, the blood sugar did not fall in the resting postabsorptive state; but the concentration of ketone bodies in the blood increased (Ffirst et al., 1971). Under such conditions, the CNS apparently switches to ketones as a substrate just as during starvation (Owen et al., 1967; and Owen et al., 1969). When alanine is infused in carbohydrate deprived sub- jects, there is an immediate increase in glucose production, showing that the low output of glucose is due to insufficient gluconeogenic substrate (Felig et al., 1969c; and Ffirst et al., 1971). When exercise was performed (90% of maximal V0 for 25 minutes) 2 after two days or a carbohydrate rich dietary regimen, a carbohydrate free diet, there were significant differences in liver metabolism. After the carbohydrate free dietary regimen ketogenesis was increased and the uptake of gluconeogenic substrates (alanine, lactate) by the liver was significantly increased, as was the output of glucose. But after a carbohydrate rich dietary regimen, there was no significant uptake of gluconeogenic substrate, and alanine concentration increased in arterial blood. There was no ketone production, but glycogenolysis was about 5 times higher than after the carbohydrate free dietary regi— men, where liver glycogen levels were very low (Hultman and Nilsson, 1971). C. Acid-base Balance The accumulation of lactic acid and other metabolites during exercise exerts an osmotic effect which may explain the increase in intracellular water observed in the muscle. As lactic acid increases 89 in the working muscle, there is an increase in the acidity of the cells, which is to a large degree buffered intracellularly. However, it appears that the sodium pump preferentially removes hydrogen ions, and this can explain the increase in sodium seen during exercise (Rooth, 1966; and Bergstrom et al., 19713). Lack of available energy in the form of ATP for transport of sodium out of the cell, or increased leakage could also explain these results (Bergstrom et al., 1971b). Hoffman (1973) has shown that there is a separate compartment of ATP preferentially used by one component of the pump, which appears to be derived from anaerobic glycolysis only. A watershift, per se, is apparently not a limiting factor, since it was most pronounced after five minutes, but work could be continued until depletion of glycogen (Bergstrom et al., 1971a). At high work intensities (above 89% of the maximal oxygen con— sumption), and in isometric work, there is evidence that inhibition of phosphofructokinase limits glycolysis (Bergstrom et al., 1971b). There is an inverse relationship between lactic acid concentration and the ratio of F-6-diphosphate/fructose-6—P concentration in the muscle; this could mean that the intracellular acidosis limits the availability of glycogen during this kind of exercise. At this level of work, the glycogen stores are not depleted (Bergstrom et al., 1971a). Hermansen and Osness (1972) have shown that during maximal con- tinuous, and intermittent exercise, the intracellular pH decreased sig— nificantly (from 6.92 to 6.41), while the capillary pH decreased from 7.42 to 6.93. They determined the intracellular pH by taking muscle biopsies and immediately freezing the samples in liquid nitrogen. 90 Muscle pH down to 6.32 and blood pH values as low as 6.80 were seen (capillary blood). The muscle pH increased rapidly immediately after exercise, whereas the blood pH continued to fall. With intermittent exercise, blood pH continued to fall in every bout of exercise, but muscle pH increased in the rest periods and fell to about the same value during each work period. These results agree well with those of Karlsson (1971a), who showed that lactate increases to about the same level at exhaustion. The intracellular pH has a pronounced effect on the metabolic reaction. Hill (1955) in in vitro studies, found that lactic acid formation in response to muscle stimulation ceased at a pH of about 6.3. This supports the theory that substrate availability (anaerobic glyco- lysis) limits exercise during maximal work. Relman (1972) has shown (in vitro at 370 C) that aerobic lactate production decreases about 60% when pH drops from 7.4 to 6.1, and anaerobic glycolysis falls about 65%. The main reason for this is the sensitivity of phosphofructokinase to ambient pH (the rate limiting enzyme in glycolysis). Relman (1972) also points out that diphosphofructose phosphatase (regulator of gluco- neogenesis) is activated by acidosis. Kemp (1969) has shown that a decrease in pH of 0.35 units decreased the maximal velocity of PFK (Phospho Fructo Kinase) from 110 units/mg to 75 units/mg. It appears that hepatic gluconeogenesis is relatively unaffected by the acid—base situation, whereas in the kidney, alkalosis inhibits and acidoses enhances gluconeogenesis (Kamm et al., 1969). 91 In evaluating the effect of changes in intracellular pH, it is important to take the intracellular compartmentation into account. The data of Carter (1972) indicates that there are marked differences in the pH of the various compartments. The pH values obtained by the muscle biopsy technique must give some kind of weighed average. Carter (1972) suggests that the cytoplasm is the primary compartment involved in discharging acidic metabolic products and in buffering acid or alkali loads added extracellularly. Apart from effecting the availability of substrate, hydrogen ions appear also to have a direct effect of decreas- ing the muscle contractility (Katz, 1970), and the increase in hydrogen ions reduces the binding capacity of troponin for calcium (Fuchs et al., 1970). In considering the possible effect that keto acids may have upon the availability of substrate and contractility, it appears unlikely that they can have any significant effect, since ketone bodies are mostly produced in the liver and therefore do not produce the same fall in intracellular pH as the locally produced lactate. However, there is some indication that acidosis and alkalosis produced by administering ammonium chloride and sodium bicarbonate respectively, do affect the work capacity (Denning et al., 1931). It was shown that work capacity decreased after the administration of ammonium chloride. Thus, the ability to neutralize lactic acid would be decreased during ketosis. Hirche et al. (1973) found that the muscle cell membrane is the main barrier that limits diffusion of lactic acid out of the muscle; and that an increase in bicarbonate concentration in the interstitial 92 fluid markedly increased the rate of lactate diffusion. It thus appears that a ketogenic diet not only decreases the glycogen stores, but also limits the degree to which glycogen can be used during maximal exercise, since it decreases the bicarbonate level in the blood. Staib et al. (1964) also found that NaH2C03 improved performance in runners; but Margaria et al. (1971) found that alkalosis induced by several different means, had no significant effect on performance during supermaximal exercise, nor did it affect lactate production. D. Other Humoral Factors The possible factors that limit work capacity and duration have received much attention, but it is generally agreed that the limiting factors may vary with the time of exercise, its duration and intensity as well as the training condition of the individual (Haralambie, 1973). However, it is very possible that different factors are limiting in different individuals. 1. Potassium. In short-term exercise, there is a relatively low potassium loss from the muscle (Hultman, 1967); but if the loss is calculated on the basis of dry weight, there is a loss of about 12% after 20 minutes of heavy exercise (Bergstrom et al., 19713) due to a shift of water into the cell. Potassium appears to be contained in the glycogen and released as the glycogen stores are used (Hultman, 1967). During long-lasting exercise, plasma potassium decreases (Haralambie, 1973a); this may be due to potassium lost in sweat, or to decreased use of glycogen as work time progresses. However, in 90 km cross-country ski racing, no important changes were seen in plasma potassium concen- tration (Refsum et al., 1973). 93 2. Other factors. Numerous other factors have been involved as a cause of fatigue: AMP, IMP, and urea are some of those most fre- quently mentioned (Haralambie, 1973b). The factors stressed here are those related to substrate utilization. Since a high-fat diet decreases muscular efficiency (Krogh et al., 1920) and endurance (Christensen and Hansen, 1939), it is possible that these phenomena are related. As some muscle units become exhausted, it is probable that less efficient motor units are recruited with a result— ant decrease in efficiency. Krogh et al. (1920) also observed that deficiency decreased most at the time of exhaustion. E. Summary Muscle glycogen stores appear to affect the time to exhaustion, when exercise is performed at 65-89% of the maximal oxygen uptake. At lower levels of exercise, muscle glycogen is only used very slowly and does not become depleted at time of exhaustion. At higher levels of exercise, muscle glycogen is not depleted either, and it appears that the intracellular acidosis limits the rate of glycolysis. The liver glucose output increases during exercise, partly due to increased gluconeogenesis and partly to increased glycogenolysis. The alanine released from the muScle during exercise plays an important role in gluconeogenesis. The utilization of liver glycogen depends upon both the degree and level of exercise. During long—lasting, severe exercise, it is depleted; and the liver glycogen may become a limiting factor under such circumstances. 94 During high work intensities, there is evidence that the in— crease in intracellular lactate concentration (which causes a decrease in pH) limits the energy supply to the working muscle, probably through inhibition of phosphofructokinase. The decreased intracellular pH may also directly affect the muscle contractility. Ketosis may decrease the rate of lactate diffusion out of the muscle, and thus indirectly be a limiting factor during certain types of exercise. EXPERIMENTS PART ONE: EXERCISE EXPERIMENTS I. EXPERIMENT I A. Introduction The plan for the present study was to determine whether the decreased efficiency, observed by Krogh et a1. (1920), and by Marsh and Murlin (1928) when a high fat diet was fed, was due to acidosis from the ketone production. The plan was to feed the subjects a diet high enough in fat to produce ketosis and then to measure the muscular efficiency during work on a bicycle ergometer. The high fat diet would then be repeated with the addition of enough bicarbonate to eliminate the effect of acidosis. By comparing the efficiency under these two conditions, it should be possible to see if the acidosis was responsible for the decreased efficiency observed during a high fat dietary regimen. It is well established that a high fat diet produces ketosis (Weis and L3ffler, 1970; Cahill, 1976; and Krebs, 1966); it is also well documented that a ketogenic diet produces an unexpectedly large weight loss due to a shift in the water-balance (Kekwick and Pawan, 1957; Olesen and Quaade, 1960; Pilkington et al., 1960; and Yang and Van Itallie, 1976). Since it is clear that dehydration impairs performance (Saltin, 1964b; Costill et al., 1974a; and Henschel et al., 1954), it 95 96 appeared likely that the ketogenic diet could cause dehydration, and the fatigue associated with dehydration could well be responsible for the decreased efficiency. B. Procedures l. Muscular efficiency. Muscular efficiency can be defined as work output divided by energy consumed during the work period, minus the energy expenditure during a similar period of rest. The energy output during work can be measured on a bicycle ergometer or treadmill, and the corresponding energy consumption can be derived from the RQ and the oxygen consumption during work, minus the oxygen consumption during rest. The oxygen consumption during rest is small relative to the work periods and can be disregarded when the efficiency for two dietary periods are compared. In this study, constant work and recovery periods are employed, and when comparing two such periods, the oxygen consumption is a measure of the efficiency, provided that correction is made for the higher caloric equivalent of 1 liter of oxygen when carbo- hydrate, rather than fat, is the substrate. In the following, it is therefore to be understood that a change in the oxygen consumption implies a change in the muscular work efficiency. 2. Experimental outline. A general outline of the present experiments is given in Table 1. In order to minimize the effect of training as observed by Krogh et a1. (1920), Experiment I began with a three-week period during which the subjects trained for 30 minutes daily on the bicycle ergometer, with work load and speed the same as that used 97 TABLE 1. Experimental outline showing the sequence of the different diets, the kind and amount of fat and proteins, and the kind of exercise experiments performed. Ex im t D 1 Di 2 Fat3 Protein4 5 per en ate at Kind Amt. Kind Amt. Exercise 3/31—4/18 Ad lib. Training 4/21-4/25 CHO I PUS 15 Soya 10 Ergometer 4/25-4/27 Ad lib. None 4/28-5/ 2 CHO II PUS 15 Soya 10 Ergometer I 5/ 2-5/ 4 Ad lib. None 5/ 5-5/ 9 Fat I PUS 75 Soya 10 Ergometer (1975) 5/ 9—5/11 Ad lib. None 5/12-5/16 CHO III PUS 15 Soya 10 Ergometer 5/16-5/18 Ad lib. None 5/19-5/23 Fat 11 PUS 85 Soya 6 Ergometer 5/23-5/25 Ad lib. None 5/26-5/30 CHO IV PUS 15 Soya 10 Ergometer 1/18—1/24 CHO I PUS 10 Soya 5 None II 1/25-1/30 Fat I Sat 89 Milk 5 None 1/30—2/ 7 CH0 II PUS 10 Soya 5 None (1976) 2/ 8-2/13 Fat II PUS 89 Soya 5 None 2/13-2/20 CH0 III PUS 10 Soya 5 None 4/ 4-4/17 CHO I PUS 25 Soya 15 Treadmill III 4/18-4/29 Fat I Sat 80 Meat 15 Treadmill 4/29-5/16 CH0 II PUS 25 Soya 15 Treadmill (1976) 5/17-5/28 Fat II PUS 80 Soya 15 Treadmill 1The dates overlap because blood samples were taken in the postabsorp- tive state at the end of each experimental period, the same morning the new dietary regimen was introduced. 2"Ad lib." indicates that the subjects were free to choose their own diet (Friday and Saturday); however, during the second and third ex- periments, they changed directly from one diet to the next. "CHO" in- dicates the high carbohydrate control diet to which the results for the high fat periods were compared. continued 98 TABLE l--continued 3"PUS" indicates that the fat came primarily from corn oil and avocadoes rich in PUS; "Sat" indicates that the fat came from either milk or meat. The amount of fat is given as percent of calories. 4The primary source of protein is indicated; however, other constituents of the diets also contributed some protein. The amount of protein is given as percent of calories. 5"Training"--the subjects trained for 30 minutes daily on the bicycle ergometer, but no expired air was collected. "Ergometer"--the subjects worked 10 minutes on the bicycle ergometer on the first, third, and fifth days of the experiment. (During the first "ergometer" period, the procedure was the same, but no results were collected.) "None"--no exercise was performed, and the normal activity of the sub- jects was kept as uniform as possible. "Treadmill"--a standard walk (5 minutes) and a maximal oxygen consump- tion experiment were carried out with one of the subjects (KB) on each Wednesday of the experimental periods. 99 in the following study periods. Each dietary period lasted five days, and on Friday and Saturday between periods, the subjects were allowed to choose their own food (except for alcoholic beverages). No data was collected during the first high carbohydrate control diet (CHO I), although the procedures were exactly the same as during the other con- trol periods (CHO II-IV). A second high carbohydrate control period was followed by the Fat I diet using about 75% of the calories from fat. After another control period (CHO III), the Fat II diet was introduced, using about 85% of calories from fat. Experiment I then terminated with a fourth control period (CHO IV). 3. Subjects. The subjects participating in these studies were William D. Hart (WH), Rong-Ching Hsieh (RH), and Kristian Balwin (KB). Weight (Kg), height (cm), and birth-dates are given in Table 2, the approximate energy needs are also shown. TABLE 2. Weight, height, and birth—dates for the subjects. Subject Weight Height Birth-date Energy Needs (Kg) (cm) (Kca1)1 WH 104 188 1945 2700 RH 61 175 1950 2300 KB 64 175 1935 2400 1Approximate energy consumption. 100 KB and RH are very similar in build and weight, both on the low side of the "desirable weight" standards (prepared by Metropolitan Life Insur- ance Company, 1960). KB is from Denmark and RH from Taiwan. WH is American, but of Scandinavian origin, heavier build, and about 50 1b. above the "desirable weight". WH had a history of developing high uric acid levels when consuming high fat diets, and for this reason did not participate in Experiment III. Therefore, RH became the second subject. For family reasons WH did not participate in CHO III and the first day of CHO IV in Experiment I. 4. Diets, The composition of the diets was calculated as shown in Appendices A, B, C, and D, and as given in Table 3. The aim of the high fat diet was to produce ketosis. The level of fat to accomplish this was not precisely known, so the first high fat diet contained 75% of the calories from fat, which was increased to 85% in the second high fat diet period. Proteins contributed 10% of the calories in the con- trol diet and the first high fat diet, but in the second high fat diet, the protein level was decreased to 6% of calories in order to make the diet more palatable. The composition of the diets was calculated based on information from the manufacturer where possible, and otherwise based on information in the Agriculture Handbook, No. 8 (Watt and Merrill, 1975). The subjects ate three different meals a day and received a little snack to take home. In order to decrease variation in composi- tion throughout the study, food was bought in one lot and refrigerated or frozen until used. The subjects ate the same food in the same amount every day. The food was prepared and eaten in the dietary kitchen of 101 TABLE 3. Approximate composition of diets (percentage of calories) in Experiment I. Diet Carbohydrate Fat Protein Control 75 15 10 Fat I 15 75 10 Fat II 9 85 6 the Food Science Building. The composition of the various diets is given in the above table. KB received a total of 2400 kcal and WH re- ceived 2700 kcal daily. The subjects also received one multivitamin and mineral tablet/day.1 The fat in the high fat diets was basically lUnicap, made by the Upjohn Company, Kalamazoo, Michigan 49001. The composition of these tablets were: Vitamins Vitamin A 1.5 mg Vitamin D 10 ug Thiamine mononitrate 10 mg Riboflavine 10 mg Ascorbic acid (Na—ascorbate) 300 mg Niacineamide 100 mg Pyridoxine Hydrochloride 2 mg Calcium Panthothenate 20 mg Cobalamine concentrate 4 ug Vitamin E 30 int. units Minerals Fe (as ferrous sulfate) 10 mg I2 (as potassium iodate) .15 mg Ca (carbonate) 50 mg Cu (sulfate) 1 mg Mu (sulfate) 1 mg Mg (sulfate) 6 mg K (sulfate) 5 mg 102 from corn oil and avocadoes, whereas the majority of the protein was soy-protein made by Worthington Company.1 In the control diet, soy- protein also supplied the larger share, but some protein came from potatoes and bread used in this diet. Most of the carbohydrate was in the form of fruits and vegetables. No coffee, tea, or alcohol was allowed. 5. Exercise. Exercise was performed on a bicycle ergometer at 70 rpm/min and constant load. Subjects arrived at the laboratory at 7:00 a.m. in the postabsorptive state and exercised for 10 minutes at about 50% of the maximal oxygen consumption (based on the maximal oxygen consumption as determined for KB in ExperimentIII); this exer- cise was followed by a lO-minute recovery period. Exercise was per- formed on Monday, Wednesday, and Friday of each experimental period. On Tuesday and Thursday, measurements were taken in the resting state only,at least 12 hours after the last meal, and after 20 minutes rest in the laboratory. The measurements were taken with the subjects sitting in a chair at ambient room temperatures. 6. Measurements. The water intake was recorded in a log book for water—balance calculations. Body weight measurements were taken nude in the morning after urination. An attempt was also made to measure the weight loss during exercise to get an idea of the sweat loss. Twenty-four-hour urine samples were collected for pH and ketone body measurements, and the urine was kept refrigerated under toluene lWorthington Foods, Worthington, Ohio. 103 during the day of collection. When the 24-hour collection was complete, pH and volume were measured immediately, and samples frozen. Friday, the last day of each experimental period, duplicate venous blood samples were taken before and immediately after the exercise period. The blood samples were taken anaerobically, placed on ice, and delivered to the Olin Health Center where they were analyzed for pH and pCOZ. The expired air was collected during the entire exercise periods. Each collection bag represented two minutes and the change of bags was done automatically when the operator pushed a button. The expired air was also collected for the first 10 minutes of the recovery period; two bags were used--one for the first three minutes, and one for the last seven minutes. Heart rate was measured for both work and recovery periods. 7. Equipment. The pH and pCO2 were measured on a Radiometer pHflmeter using glass electrodes. Weight was measured on a scale (Fairbanks FN-42) accurate to :;20 g. Exercise was performed on a modi- fied Schwin electromagnetic bicycle ergometer equipped with a 12-volt automobile alternator. The load was regulated by changing the resistance of the output and the same setting was used for both subjects. The load is independent of the pedaling speed. The expired air was collected through a low resistance OtiséMcKerrow respiratory valve and the modi— fied Douglas method was used in collecting the gas. The heart rate was counted on a Well's pulse rate meter with readout to a Gibson five channel recorder. 104 The oxygen content of the expired gas was analyzed on a Beckman paramagnetic analyzer (model E2); and the carbondioxide was analyzed on a Beckman infrared analyzer (model LB 15A). For calibration of the gas analyzers, the zero point was set with helium, and for the upscale values, atmospheric air and a standard gas were used. The composition of the standard gas corresponded to the expired gas (about 4.3% CO2 and 17.73% 02). The standard gas sample was calibrated with repeated Haldane gas analysis. 8. Calculations and analysis. The expired air was collected and immediately analyzed for CO and 02, the volume was measured and the 2 temperature recorded. The barometric pressure in the laboratory was also recorded at the time of analysis. The expired gas volume was then converted to STPD (O C, 760 mm Hg. dry.) conditions using the method of Consolazio et al. (1963). By using the measured concentrations of CO2 and 02, the RQ and the true oxygen consumption were determined from the monograms and formulas described by Consolazio et al. (1963). C. Results 1. Oxygen consumption and RQ during rest. Calculated RQ and oxy- gen consumption during rest are shown in Table 4. 2. RQ and oxygen consumption during exercise. The RQ always increased to a fairly steady level during the first three minutes or so of exercise, although a decrease was often observed at the end of the work period. The RQ, as expected, was lower during the high fat period than during the high carbohydrate period. The RQ's from two typical work periods are plotted in Figure l. A typical plot is shown for each period (fat and carbohydrate) for each subject. 105 TABLE 4. R0 and oxygen consumption during rest in Experiment I. Date Diet R03 02 Cone. ml/min3 R04 02 Cons. ml/min4 4/291 CHO II 0.76 270 0.78 310 5/ l2 CHO II 0.83 330 0.74 300 5/ 61 Fat I 0.77 260 0.81 290 5/ 82 Fat I 0.77 276 0.72 330 5/131 CHO III 0.84 271 5/152 CHO III 0.85 256 5/201 Fat II 0.75 340 0.78 320 5/222 Fat II 0.70 273 0.72 298 Mean CH05 0.82 282 0.76 305 Mean Fat6 0.75 287 0.76 310 1Second day of the dietary period. 2Fourth day of the dietary period. :Results for KB. Results for WH. .Mean of all CHO periods. O‘UI Mean of all Fat periods. FIGURE 1. 106 RQ's during exercise for KB and WH during Experiment I. (The RQ's are plotted as a function of time for each subject for a typical work period during a high fat and a high carbohydrate regimen. The value plotted for zero time is the resting RQ measured on a different day.) 107 RHQ° 0 High carbohydrate diet 0 High fat diet 1 For KB 100 ‘ i .8 W - _ Ltime in 6 1 1 ‘ l r min 2 4 6 8 10 12 l4 16 18 20 1 Exercise _L Recovery ' r I l RéQ. For WH 1'0 1 M ’0\/ .8 ‘time in .6 l. - A l 1 J L A A ‘ _r min 2'4 6 8101214161820 FIGURE 1 108 Since apart from the first four minutes of work, the R0 was fairly stable, the mean of the next three 2 min periods has been calcu- lated. The total oxygen consumption during work and recovery in each dietary period has been calculated, and is given as work and recovery 02. Since the oxygen consumption during the first four minutes usually was less than the rest of the work period, the work oxygen is divided into Initial 02 (first four minutes) and Main 02 (last six minutes). In order to minimize daily variations, the mean oxygen con- sumption is calculated for each dietary period. The results are shown in Table 5 for KB and in Table 6 for WH. The mean R0 for each dietary period corresponding to the main oxygen consumption is also calculated and shown in Tables 5 and 6. The maximal pulse rates are likewise given. 3. Blood data. On the last day of most dietary periods, two venous blood samples were taken before and after exercise for measure- ment of pCO and pH. The mean values of two determinations for KB are 2 given in Table 7. 4. Body weight. The body weight was measured throughout the study (nude), but on the days of exercise, it was recorded before and after exercise. The results are shown in Table 8 for KB and in Table 9 for WH. The average weight losses during exercise for the fat and carbohydrate periods were calculated as was the total weight loss dur— ing the fat periods. The results are shown in Table 10. 109 TABLE 5. Pulse rate, R0, and oxygen consumption during exercise, for KB, in Experiment I. Date Diet PR1 R02 Total3 Work4 Rec. Int. Main7 4/28 CHO II 148 .86 30.90 25.51 5.39 9.84 15.67 4/30 CHO II 148 .86 29.76 24.12 5.64 9.26 14.86 5/ 2 CHO II 152 .87 30.38 24.15 6.23 9.18 14.97 Mean CHO II 149 .86 30.35 24.59 5.75 9.43 15.17 5/ 5 Fat I 148 .83 31.54 25.20 6.34 9.58 15.62 5/ 7 Fat I 147 .83 29.30 23.51 5.79 8.89 14.62 5/ 9 Fat I 156 .82 32.60 25.18 7.42 9.47 15.71 Mean Fat I 150 .83 31.15 24.63 6.52 9.31 15.32 5/12 CHO III 144 .91 29.55 24.00 5.55 8.88 15.12 5/14 CH0 111 144 .98 27.61 22.34 5.27 8.33 14.01 5/16 CHO III 147 .88 30.20 24.40 5.80 8.83 15.57 Mean CHO III 145 .92 29.12 23.58 5.54 8.68 14.90 5/19 Fat II 152 .87 30.11 24.46 5.65 9.39 15.07 5/21 Fat II 144 .89 28.78 22.72 6.06 8.50 14.22 5/23 Fat 11 159 .81 29.34 22.59 6.74 9.13 13.46 Mean Fat 11 152 .86 29.41 23.26 6.15 9.01 14.25 5/26 CHO IV 147 .89 28.33 22.50 5.83 8.46 14.04 5/28 CHO IV 145 .87 29.26 23.79 5.47 8.72 15.07 5/30 CHO IV 150 .89 29.45 23.72 5.73 9.22 14.50 Mean CHO IV 147 .88 29.01 23.34. 5.68 8.80 14.54 1Pulse rate (beats/min), at the end of exercise. 2The mean RQ of the last six minutes of exercise. 3Total oxygen consumption (liter/20 min). O‘Ul-L‘ Initial oxygen consumption during the first four minutes (liter/4 min). \I Oxygen consumption during work (liter/10 min). Oxygen consumption during recovery (liter/10 min). Main oxygen consumption during the last six minutes (liter/6 min). 110 TABLE 6. Pulse rate, R0, and oxygen consumption during exercise, for WH, in Experiment I. Date Diet PRl RQ2 Total Work4 Rec. Int. Main7 4/28 CHO II 148 .95 31.10 24.81 6.29 8.76 16.05 4/30 CHO II 140 .85 31.27 24.78 6.49 8.24 16.54 5/ 2 CH0 II 144 .92 33.42 25.07 8.35 9.27 15.80 Mean CHO II 144 .91 31.93 24.89 7.04 8.76 16.13 5/ 5 Fat I 148 .84 32.67 25.92 6.75 9.24 16.68 5/ 7 Fat I 150 .86 29.67 23.54 6.13 9.15 14.39 5/ 9 Fat I 159 .86 36.28 29.03 7.25 11.16 16.87 Mean Fat I 152 .85 32.87 26.16 6.71 9.86 16.31 5/198 Fat 11 147 .90 31.48 25.21 6.27 8.96 16.25 5/21 Fat II 162 .89 30.25 24.92 5.33 9.14 15.78 5/23 Fat II 171 .84 33.11 26.21 6.90 9.55 16.66 Mean Fat II 160 .88 31.61 25.45 6.17 9.22 16.23 5/289 CHO IV 144 .96 29.77 23.88 5.89 8.72 15.16 5/30 CHO IV 156 .94 29.50 23.47 6.03 9.01 14.46 Mean CHO IV 150 .95 29.64 23.68 5.96 8.87 14.81 lPulse rate (beats/min). 2The mean RQ of the last six minutes of exercise. 3Total oxygen consumption (liter/20 min). 4 5 6 7 Oxygen consumption during work (liter/10 min). Oxygen consumption during recovery (liter/10 min). Initial oxygen consumption during first 4 min (L/4 min). Main oxygen consumption during the last 6 min (L/6 min). 8 9 WE did not participate in the previous control period. WH did not participate the first day of the period. 111 TABLE 7. Blood pH and pCO2 for KB in Experiment I. Date Diet le sz ApH3 pC024 pCOZS ApCO2 5/ 9 Fat 7.34 7.30 0.04 55.3 51.2 4.1 5/16 CH0 7.29 7.28 0.01 59.6 42.9 16.7 5/23 Fat 7.36 7.32 0.04 51.5 47.8 3.7 1 . pH before exercise. 2pH after exercise. 3Decrease in pH during exercise. 4pCO2 before exercise. 5pCO after exercise. 6 2 Decrease in pCO during exercise. 2 TABLE 8. Body weight and weight loss during exercise for KB in Experiment I (all values in kg). Date Diet W11 W22 AW3 4/28 CHO II 65.68 65.62 0.04 4/30 CH0 II 65.50 65.40 0.10 5/ 2 CH0 II 65.54 65.32 0.22 5/ 5 Fat I 66.06 65.88 0.18 5/ 7 Fat I 65.76 65.64 0.12 5/ 9 Fat I 64.82 64.13 0.69 5/12 CHO 111 65.54 64.90 0.64 5/14 CHO III 65.44 65.32 0.12 5/16 CHO III 65.56 65.35 0.11 5/19 Fat II 65.14 64.40 0.74 5/21 Fat II 64.96 64.76 0.20 5/23 Fat II 64.00 63.74 0.26 1Body weight before exercise. 2Body weight after exercise. 3Body weight loss during exercise. 112 TABLE 9. Body weight and weight loss during exercise for WH in Experiment I (all values in kg). Date Diet W11 W21 AWl 4/28 CHO II 99.64 99.34 0.30 4/30 CHO II 99.66 99.22 0.44 5/ 2 CHO II 99.18 98.88 0.30 5/ 5 Fat I 99.78 99.62 0.16 5/ 7 Fat I 98.66 98.18 0.48 5/ 9 Fat I 97.66 97.42 0.24 5/19 Fat 112 99.66 99.30 0.36 5/21 Fat 11 97.68 97.36 0.32 5/23 Fat 11 97.14 96.92 0.22 lFor explanation of abbreviations see Table 8. 2WH did not participate in the second control period because of a funeral. TABLE 10. .Mean weight loss in kg (from Tables 8 and 9) during Experiment 1. Diet Total weight loss Average exercise loss CHO II 0.001 0.462 0.211 0.352 Fat 1 1.241 2.122 0.331 0.292 Fat 11 1.141 2.522 0.401 0.302 1Results for KB. 2Results for WH. 113 5. Water-balance data. Water-balance is used to indicate the difference between water intake and urinary output. Tea, coffee, and soft drinks were not permitted. The daily urine volume was recorded, and the total water intake calculated from the record of drinking water consumed, plus the water contained in the food. The metabolic water is ignored; although on a weight basis, fat produces more water, on a caloric equivalent basis, carbohydrate produces slightly more water. The water-balance represents the water lost by perspiration and other losses. These data and the urinary pH are shown in Table 11 for KB, and the mean values in Table 12. 6. Urinary ketone excretion. The subjects' urinary ketone excre- tion was evaluated with Keto-Stix (Ames) based on the purple color developed when the ketone bodies react with nitroprusside. The ketone level is graded as "small", "medium", and "high". During the Fat I regimen (75% of calories from fat), the tests were negative, except for an occasional, very weak reaction toward the end of the 5—day period. During the Fat 11 regimen (85% of calories from fat), a trace of ketones was found in the urine on the second day, which increased to medium on the last day. The results for Fat 11 regimen are tabulated in Table 13. There was no significant difference between the subjects. 114 TABLE 11. pH and water-balance data for KB during Experiment I. Date Diet Water intake1 Urine Water-balance2 pH3 4/28 CHO II 2338 1000 1338 7.00 4/29 CH0 II 2588 900 1688 7.05 4/30 CHO II 2838 690 2148 6.67 5/1 CH0 II 2088 900 1188 6.95 5/2 CH0 II 2388 850 1488 7.09 5/5 Fat I 2296 1319 906 6.95 5/6 Fat I 2046 1250 796 6.86 5/7 Fat I 1796 950 850 6.58 5/8 Fat I 2046 890 546 6.66 5/12 CHO 111 2338 920 1418 6.96 5/13 CHO III 2588 810 1778 7.20 5/14 CHO III 2838 890 1948 6.83 5/15 CHO III 2838 530 2308 6.50 5/19 Fat 11 1428 490 938 6.07 5/20 Fat 11 2428 540 1888 6.10 5/21 Fat II 2428 580 1848 6.61 5/22 Fat II 2928 1170 1758 6.25 1Water intake consists of the water contained in the food and the water or other fluid intake. 2Water-balance is calculated as the water intake minus the urine volume = water lost in stool, through perspiration, and as insensible water loss. 3pH of 24-hour urine collection. 115 TABLE 12. Mean water-balance and urinary pH for KB during Experiment I. Diet Mean water-balance1 Mean urine pH CHO II 1570 6.95 Fat I 775 6.76 CHO III 1868 6.87 Fat II 1608 6.26 lMean water-balance in ml, calculated as the difference between water intake and urine. TABLE 13. Urinary ketone levels above control values during the Fat II regimen in Experiment I. Ketone level1 Date 5/19 0 5/20 trace 5/21 small 5/22 small-to-medium 5/23 medium lGraded from 0 (control value) to high (severe ketosis). 116 II. EXPERIMENT III (EXERCISE) A. Introduction As can be seen from the RQ's in the previous experiment (Tables 5 and 6), there was not much difference between the substrate utilized during exercise when the fat or carbohydrate diet was consumed. This was no doubt due to the utilization of muscle glycogen. In subject KB, however, a significant decrease was seen in the R0 in the last experi- ment during the Fat II regimen in Experiment I; this probably indicates a functionally important decrease in muscle glycogen. Also, the exer- cise RQ's during the Fat II period generally were higher than those for the Fat I regimen, despite the fact that Fat II contained more fat (85% vs. 75%). This may well have been due to increased glycogen stores at the initiation of the Fat II diet. Partial depletion of the glycogen during the Fat I regimen might have caused supercompensation of the glycogen stores when the high carbohydrate diet (CHO III) was fed. There are two possible ways to accomplish a higher utilization of fat. First of all, one can increase the level and particularly the duration of the exercise to deplete the glycogen stores, or the glycogen stores can be depleted prior to the experiment. The latter approach is undesirable because it causes undue hardship on the subjects, as will be discussed later. More importantly, perhaps, depletion of muscle glycogen stores may well be the factor that causes the decreased effici- ency when a high fat diet is fed. At low work levels fat is the major substrate (Havel,197l), and muscle glycogen is not the limiting factor (Saltin and Karlsson, 1971). It therefore seems possible to increase 117 the utilization of fat by decreasing the work load. Another exercise experiment was carried out with the subject KB at a lower exercise level. In this series, a treadmill was used instead of the bicycle ergometer, but otherwise, the methodology was the same except for shorter work and recovery periods. The composition of the diets is given in Table 14 as percent of calories, and the calculations are shown in Appendices E, F, G, and H. TABLE 14. Approximate composition of the diets as percent of calories during Experiment III. Diet Protein Carbohydrate Fat High CHO 15 60 25 High Fat 15 5 80 In contrast to the bicycle experiment, which was done in the postabsorptive state, the treadmill experiment was done in the absorp— tive state (1-3 hours after the meal); this appears to favor utilization of the substrate fed (fat on a high fat diet). The speed and grade of the treadmill could be varied, and in addition to determining the maxi- mal oxygen consumption, a standard walk experiment was performed. B. Procedures The experimental outline can be seen from Table I (page 97). The standard work was done by walking on the treadmill at 3 1/2 mph for five minutes, followed by a recovery period of five minutes. Because 118 of the relatively short periods, important variations in oxygen consump— tion can be expected when the two diets were fed, due to differences in lactate production and rate of recovery. Treadmill work in general, is less precise than work on the bicycle ergometer. No training period preceded these experiments, but the subject KB was in good physical con— dition. The maximal oxygen consumption was determined immediately after the recovery period for the standard run, which served as a warm-up period. Determination of the latter involved a standard run on the treadmill on the level for five minutes at 6 mph; the speed was then in- creased to 7.5 mph and maintained for two minutes. The expired gas collection started at the end of the standard run and continued until exhaustion (bag changed every 30 seconds). The grade was increased 2.5% each minute until the subject became exhausted. Air collection and analysis were done in the same way as during Experiment I. C. Results 1. Standard work. The mean R0 and oxygen consumption (during work, recovery, and total), as well as the maximal heart rate for the different dietary periods are shown in Table 15. The average oxygen consumption for the work period during the high fat diet was 7.58 liters (per 5 minutes), compared to 6.46 liters (per 5 minutes) for the high carbohydrate diet. For the total oxygen consumption, the corresponding values were 10.08 and 8.96, respectively. 2. Maximal oxygen consumption. The maximal oxygen consumption, average RQ, and maximal heart rate (beats/min) are given in Table 16. 119 TABLE 15. Oxygen consumption, R0 and heart rate for KB during standard walk in Experiment III. Date Diet Work 021 Recovery1 Total 02 R0 HR2 4/14 CHO I 6.95 2.71 9.66 0.86 120 4/273 Fat I 7.63 2.41 10.03 0.70 121 5/ 5 CHO II 6.28 2.80 9.08 0.87 115 5/12 CHO II 6.16 1.98 8.13 0.95 105 5/19 Fat 11 7.05 3.06 10.11 0.62 110 5/26 Fat II 8.06 2.05 10.11 0.64 110 Mean4 Fat 7.58 2.51 10.08 0.65 114 Mean5 CHO 6.46 2.50 8.96 0.89 113 lLiter/5 min. 2Mean heart rate for one minute at the end of exercise. 3The measurements on 4/28 were discarded because of error in measuring the volume of expired air. 4Mean of all the fat periods. 5Mean of all the carbohydrate periods. 120 Table 16. Maximal oxygen consumption and heart rate for KB during Experiment III. Date Diet Max. VO 1 R0 Max. HR2 2 4/14 CHO I 4.28 0.92 194 4/283 Fat 1 4.48 0.90 190 5/124 CHO II 3.96 1.19 185 5/19 Fat 11 4.48 0.84 184 5/26 Fat II 4.32 0.78 187 lLiter/min. 2Heart rate at end of exercise (beats/min). :No determination on 4/21. Data not available for 5/5 due to faulty gas collection. III. EVALUATION OF DATA A. Experiment I 1. Resting oxygen consumption and RQ. There is a fairly great variation in the resting oxygen consumption as seen from Table 4 (page 105); the two main reasons for this are probably variation in the temperature of the room where the measurements were taken, and the degree to which the subjects were relaxed on the different days. The means are shown in Table 4, as are the corresponding mean RQ's. The inter individual variation seen in this table is partly explained by the difference in body mass (on a weight basis, the value is less for WH). The resting oxygen consumption does not appear to depend upon the dietary regimen. Although the average RQ for KB is lower when the 121 high fat diets were eaten, there is no difference for WH. This is not surprising, since in the resting postabsorptive state, the primary sub- strate is fat, regardless of diet. The last determination of the resting RQ was made on the next to the last day of each dietary period; this R0 was 0.70 for KB and 0.72 for WH, which is close to pure fat metabolism. Only KB responded with an increased carbohydrate metabolism in the postabsorptive state during the high carbohydrate diet. The low resting RQ's agree well with the results obtained using radioactive tracers (Havel, 1971). From such studies, it appears that glucose contributes no more than 20% of the substrate; this corresponds to a RQ of 0.76 (Swift and Fisher, 1964). It is interesting to notice that in KB, the high carbohydrate diet caused an increase in the utilization of CHO in the resting state (about 50% at a RQ of 0.85). In recent studies, Felig et a1. (1975) showed by catheterization techniques, that the liver is the primary site of disposal of an oral glucose load; and some of the glucose retained by the liver is undoubtedly released later for peripheral utilization. 2. Exercise RQ. The exercise RQ's are not consistently lower during the high fat dietary regimen compared to the high carbohydrate diet; but on the 1aSt day of exercise during the high fat period, the RQ begins to decrease (Tables 5 and 6, pages 109 and 110). During exercise of this length (10 min), glucose plays an important role as substrate (Havel, 1971); but during the high fat period, the liver glycogen will be depleted (Nilsson and Hultman, 1973), whereas one would not expect the muscle glycogen to be exhausted (Hultman and 122 Nilsson, 1975). However, the relatively lower RQ on the last day of exercise indicates that even with a relatively low activity level, the muscle glycogen is significantly decreased. The work RQ's obtained during the high fat period are considerably higher than those of Christensen and Hansen (1939); but their data were based on exercise of long duration, and they saw a continual decrease in the RQ's with time. For KB, the RQ's are only slightly higher than those observed by Marsh and Murlin (1928), who used a work time of 20 minutes. Based on the RQ, fat contributed (for KB) from 60-63% of the calories on the last day of the high fat dietary periods (Swift and Fisher, 1964). This shows the importance of the muscle glycogen during exercise. During the high carbohydrate periods, fat contributed on the average, about 35% of the calories during exercise; but in some experiments, as little as 6% of the calories came from fat. In both KB and WH the RQ appears to be lower during the Fat I period than during the Fat II period (Tables 5 and 6). This may be due to carbohydrate supercompensation during the CH0 III period between the two high fat periods (Hultman and Nilsson, 1975). 3. Efficiency and oxygen consumption. Krogh et al. (1920) calculated the muscular work efficiency by subtracting the resting oxygen consumption from the oxygen consumption during work (converted to the caloric equivalent). However, since the variation in the oxygen con- sumption during work is greater than the resting oxygen consumption, the latter is just as good a measure of efficiency. The oxygen consumption varies only slightly with the substrate used, so on a normal diet, no 123 corrections need be made. However, the caloric value of 1 liter of oxygen is 7.7% higher when pure carbohydrate rather than fat is the substrate (Swift and Fisher, 1964). If we assume that the caloric requirement for a given task is independent of the substrate used, this means that the oxygen consumption will decrease 7.15% when sub- jects change from pure fat to pure carbohydrate as the energy substrate. Looking at Table 5, it is seen that on 5/14 (CHO III), fat contributed about 6% of calories, and on 5/23 (Fat 11), the contribution was about 61%, an increase of 55%. The oxygen consumption increased about 6% compared with an expected increase of 4.3% based on the caloric equiv- alents; this difference is within experimental error. If we look at the total oxygen consumption for KB, we can clearly see the effect of training at the beginning of Experiment I (4-5%) between the first two carbohydrate diets (CHO II and CH0 III); but thereafter, very little, if any effect is seen. This agrees with the observations of Krogh et a1. (1920). Looking at the total oxygen consumption for KB, there is very little increase in the oxygen consumption during the high fat dietary regimen; the increase is 1—3%, which is to be expected because of the lower caloric value of oxygen when fat supplies the energy. This difference practically disappears when we look at the oxygen consump- tion for either the total work period or the last six minutes. During recovery period, the oxygen consumption is about 10% higher during the high fat periods, and it is about 4% higher during the first four minutes of exercise. These data suggest that the substrate has no effect on the 124 oxygen consumption when caloric equivalents are considered; however, there appears to be a trend toward less anaerobic glycolysis during the high fat periods. The venous pH (Table 7, page 111) surprisingly, was higher (0.04) during the high fat periods, but after exercise, the difference was insignificant (0.02) because of less decrease during the exercise. The pre-exercise p00 is somewhat higher during the high carbohydrate 2 periods, but it becomes much less after exercise. The decrease was about four times greater during the high carbohydrate periods, which also indi- cates a higher lactate production and/or a decreased buffering capacity. Because of variation in analysis from day to day, the differences in pH and pCO between pre— and post-exercise values during the diets are 2 probably more meaningful than the absolute values. The urinary ketone body excretion increased progressively (Table 13, page 115); but apparently, the blood is well buffered, since the venous pH did not appear to decrease (see Table 7, page 111). As will be described later, in Experiment II the arterial blood pH was about 0.03 units lower during the high fat periods. Sutton et al. (1976) has found that a higher pH (alkalosis) favors anaerobic glycoly— sis, which is supported by the increase in'anaerobic glycolysis seen during the high carbohydrate period. The slow development of ketosis agrees well with the results of Garber et a1. (1974); and Bloom (1967). A decrease in the urinary pH is alSo seen when fat diets are consumed (Table 11, page 114). 125 4. Weight loss. There is no difference in the mean body weight loss during exercise (Table 10, page 112) during the fat and carbohydrate periods for WH; but for KB, the fat periods appear to be associated with a higher weight loss during exercise. Both Krogh et a1. (1920), and Marsh and Murlin (1928) report perfuse sweating during exercise when high fat diets are consumed, and Kekwick and Pawan (1957) report higher insensible water loss during a high fat dietary regimen. The total weight 1053 (Table 10) in these experiments was no greater than can be adequately eXplained by glycogen and lean tissue degradation. The lean tissue will be discussed later, but Olsson and Saltin (1970) have shown that 2.4 kg can be gained in weight during glycogen repletion in average young men. The RQ decreased at the end of the high fat periods, indicating a large depletion of muscle glycogen in addition to liver glycogen. 5. Water-balance. Since a high fat diet is reported to produce excessive weight loss by a shift in the water-balance, it was monitored. The total water intake plus metabolic water minus the urine volume, should be equal to sweat loss and insensible water loss, assuming no change in body water and stool water. The water loss is not easy to measure; but if all the factors influencing water-balance (stool volume, insensible evaporation, and sweat) are unaltered, the water—balance should be the same during the carbohydrate and fat diets, and any change repre- sents the extra water loss due to an effect of high fat diets. The en— vironmental temperature and the relative humidity may, however, change-- which can be expected to alter the insensible water loss; and there is 126 some evidence that the insensible water loss is higher on high fat diets. The voluntary activity of the subjects may also vary during the dietary periods; there are reports that a high fat diet decreased activity (Livingston, 1972), and stool volume is affected by diet. However, if the water-balance is unaffected by the diet, the difference in water- balance between the carbohydrate diets and fat diets should correspond to the weight loss during the high fat dietary period (no weight loss during the carbohydrate period), and thus be a measure of the cumulative effect of dehydration and lean tissue catabolism. From Table 12 (page 115), it is seen that for KB there was a nega- tive mean daily water-balance of 797 g, or a total negative water-balance of almost 4 kg in five days, compared to an actual weight loss of 1.24 kg. This was undoubtedly partly due to differences in weather conditions and differences in stool water. The average water-balance increased 300 m1 during the CH0 III period (compared to CHO 11); if this occurred dur- ing the Fat II period it can account for most of the discrepancy. During the Fat II period, the weight loss was 1.14 kg and the difference between the mean water-balance amounts to 1.30 kg (calculated as the difference between the mean water-balance during the previous control period and the high fat period, and multiplied by the five days the experiment lasted). The metabolic water is ignored in the above calculations, since on a caloric basis, fat and carbohydrate produce nearly the same amount of water. Metabolism of 1 g of carbohydrate produces 0.6 g of water, whereas metabolism of l g of fat produces 1.1 g of water. Since the energy requirement of the subject was unchanged, only 4/9 g of fat can 127 substitute for l g of carbohydrate on a caloric basis; under such cir- cumstances, the metabolic water production would be 0.49 g of water per 4 kcal from fat. Since the latter value is not very different from the metabolic water produced by a calorically equivalent amount of car- bohydrate (0.6 g per 4 kcal), the difference in metabolic water produced appears negligible in relation to the other factors. B. Experiment III (Exercise) 1. Standard work. As already pointed out, Experiment I showed more anaerobic oxygen consumption during the high carbohydrate diet. Because of the shorter work and recovery periods (five instead of ten minutes) in this experiment, significant differences in oxygen consump- tion can be expected. It is seen from Table 15 (page 119) that the oxy— gen consumption showed much more variability when the subject ran on the treadmill than occurred when the subjects worked on the bicycle ergom— eter in Experiment I. This was true even when the same diet was consumed. A small variation in the speed of the treadmill has an important effect on the oxygen consumption, whereas the Pedaling speed on the bicycle ergometer has no effect on the work performed. The mean oxygen consump- tions on the different diets are compared in Table 15, but these experi- ments are for one subject (KB) only, and with just one or two experiments for each dietary period. Despite large intra-individual variations, the mean heart rates were equal for the two diets. Because of the low work load, fat was the major substrate during the high fat diets, but a surprisingly high per- centage (64%) was from carbohydrate during the high carbohydrate diet. 128 These experiments therefore, show that under appropriate conditions (exercise of short duration and low intensity), it is possible to promote substrate utilization in the desired direction. The RQ's need further comment. First, during all the high fat diets, they were very low; but particularly so for Fat II. Such low R0 was also observed by Krogh et a1. (1920), but values below 0.69 have generally been ascribed to experimental errors. In the Fat I diet, the fat was primarily from butter and meat, which can partly explain the higher R0 in this experiment, since butterfat has a minimum R0 of 0.72, rather than 0.707 for mixed fats. In the Fat II experiment, corn oil was the primary fat used. A possible explanation for the low RQ's observed during exercise, during the high fat dietary regimen, is hypoventilation; but since the hydrogen ions associated with ketosis stimulate the respiratory center, that explanation is not a very likely one. Furthermore, hypoventilation would probably not produce consistently low RQ's on the high fat diets. Hawley et a1. (1933) found many studies with consistently low RQ's (that is, below 0.69), comparable to those found when their sub- jects consumed a high fat diet. They concluded that it could not be due to error, but consistently occurs in some subjects. Such low quotients are also normally found in diabetics. Production of ketone bodies rather than complete oxidation of fat would decrease the RQ's, but for this to become important, ketone bodies would have to accumulate in the blood. If the ketone bodies are produced in the liver and oxidized in the muscle, they have no effect on the RQ. The urinary ketone body 129 excretion in this experiment was not much different from the excretion in Experiment I, where no such low RQ's were observed. Hawley et al. (1933) described experiments in which consistently low RQ's (average 0.65) were found in their subjects from 1/2 to three hours after being fed a high protein, high fat meal; furthermore, deliberate underventilation did not change the results. They could find no relation between the degree of ketosis or ketonuria, and the low RQ's. They concluded that the low RQ's were due to the formation of glucose from proteins and stored as glycogen. Formation of glucose from protein could explain our results ob- tained in the absorptive state when a high protein, high fat diet was fed. Although the RQ's appear depressed in both fat periods, it is particularly so in the last. As will be discussed later, both BUN and 24-hour urinary nitrogen excretion increased significantly during'both fat periods; but particularly during the Fat II diet, which supports the idea of gluconeogenesis taking place from proteins. At higher exercise levels, as during the maximal oxygen consumption determination that followed, the use of muscle glycogen plays a dominating role and no low RQ's were seen. According to Hultman and Nilsson (1975), glycogen is not stored in either liver or muscle when a high fat diet is fed (at least, only very slowly). But the results of Felig and Wahren (l97la) show that even mild exercise stimulates gluconeogenesis from proteins, which.may give the basis for the accumulation of glucose, and possibly explain the low RQ's. 130 2. Efficiency. From Table 15 (page 119), it is seen that there was no change in the oxygen consumption during recovery; but the mean oxygen consumption in the work period was 17% higher during the high fat periods than during the high carbohydrate periods. When the oxygen consumption is adjusted for the lower caloric value of one liter of oxygen when fat is the substrate (4.6%, assuming pure fat oxidation dur- ing the high fat periods), the increase in oxygen consumption is about 12%. This is comparable to the value found by Krogh et al. (1920) and Marsh and Murlin (1928). However, as has just been pointed out, the low RQ's indicate that gluconeogenesis was increased during the exercise period, and this may be the reason for the increased oxygen consumption during exercise in the high fat periods. Ketosis was similar during the two exercise experiments; but only in the treadmill exercise (Experiment III) was there a decreased effici- ency when fat was the major substrate. This indicates that ketosis was not a factor in causing the decreased efficiency. However, these experi- ments were carried out on one subject only, and with only two experi- ments during each dietary period. When one also considers the greater uncertainty of determining the exact workload on a treadmill (compared to a bicycle ergometer), these results need further confirmation. Krogh et a1. (1920) found the standard error for the oxygen determinations to be 2% when done on the same day. However, if this is done on a treadmill the error must be considerably larger because varia— tions in the actual speed affect the oxygen consumption (the pedaling speed does not). 131 3. Maximal oxygen consumption. The RQ's during the determination of the maximal oxygen consumption (Table 16, page 120) are all very high, which is to be expected in high intensity work of short duration, and for the high fat dietary periods. The high RQ's emphasize the impor- tance of muscle glycogen during exercise; however, the RQ during the second week of the Fat II diet (5/26) was somewhat lower, probably indi- cating a decrease in the glycogen stores. During the experiment on 5/12, which was the second week of the CH0 II period, consistently high RQ's were observed, the highest being 1.24. This, of course could be explained by hyperventilation, but since the high RQ's were seen throughout the entire period, and the subject was well acquainted with the procedures, there is no reason to believe that this was the case. Krogh et a1. (1920) also occasionally observed such high RQ's, but considered them to be due to hyperventilation. However, another explanation relates to the fact that the subject was in the absorptive state, and during a high carbohydrate diet, this causes fat synthesis from glucose (Owen and Reichard, 1971a). RQ's greater than 1 have also been found by Hatch et a1. (1955) in subjects fed the Kempher rice diet. Such high RQ‘s can best be explained by fat synthesis, which gives a theoretical RQ of 8.0 (Swift and Fisher, 1964). This experiment shows that there is no difference between the maximal oxygen consumption during the high fat period, and the high carbohydrate periods. During this kind of work, it is likely that the muscle's ability to utilize oxygen is the limiting factor during exer— cise (Pernow and Saltin, 1971). If this is the case, the maximal oxygen 132 consumption is a measure of the functioning of the TCA-cycle. And these data then show that the function of the tricarbocylic acid cycle (TCA-cycle) is not decreased during a high fat diet as claimed by Krebs (1964) to be responsible for ketogenesis. This agrees with the hypothe- sis that ketogenesis is caused by an increased plasma FFA in connection with a decrease in the insulin/glucagon ratio (Cahill, 1976). IV. SUMMARY Experiment I did not show any decrease in efficiency when fat was the major substrate; whereas a reduction was seen during Experiment III. Since the increase in ketone body production was similar during the two experiments, this indicates that ketosis is not the cause of the de— creased efficiency observed during high fat dietary regimens. During Experiment III, fat supplied a much higher fraction of the substrate than during Experiment I, probably due to the lower level of exercise, and possibly because the exercise was performed in the absorp- tive state. The longer duration of the high fat dietary regimen, and presumably a lower muscle glycogen level, may be a factor in lowering the work efficiency in Experiment III. The oxygen consumption was sig- nificantly increased during the work period of the second week, as com— pared to the first week; and the muscle glycogen was probably decreased to low values in the second week. The increased gluconeogenesis could also presumably be a factor in the decreased efficiency. These conclu- sions are limited by the relatively few experiments with one subject in Experiment III; but they demonstrate that during low work loads of short duration, fat becomes the major energy substrate. 133 PART TWO: EFFECTS OF THE DIETS V. EXPERIMENTS II AND III A. Introduction As no decrease in efficiency was seen during Experiment I, and less than expected ketosis developed, it appeared likely that the high fat diets (high in polyunsaturated fats and containing soy-protein only) were less ketogenic than conventional high fat diets. Sinclair (1964) postulated that linolenic acid has antiketogenic pr0perties, and this theory appears to be supported by the results of Tantibhedhyangkul et a1. (1967), who did not find any ketosis after ingestion of corn oil in the postabsorptive state compared to MCT (medium chain triglycerides). On this basis it was decided to compare the ketogenic properties of a high fat diet based on dairy cream as used by Christensen (1939), which did produce ketosis, with our high fat diet based on corn oil and soy-protein. The protein content was kept low (5% of calories) in Experiment II. In Experiment III the protein content was increased to 15% of calories, while the carbohydrate content was unaltered in order to study the effect of protein intake on ketogenesis, gluconeogenesis, and weight loss. Proteins are generally considered antiketogenic (Keit, 1963); however, Worthington and Taylor (1974) found that ketosis increased on a 1200 kcal high protein diet compared to a 1200 kcal mixed diet; and a pemmican diet produced a high degree of ketosis (Mark et al., 1944). Recently Wahren et a1. (1976) compared the effect of ingestion of a 134 protein (beef) meal in diabetic and normal postabsorptive men. In diabetic, but not in normal men, gluconeogenesis increased, showing that under appropriate conditions (low insulin/glucagon ratio), proteins promote gluconeogenesis. In the diabetics (but not in the control group), protein feeding caused a pronounced increase in ketogenesis. The purpose of Experiment III was to evaluate the effect of protein intake on gluconeogenesis, ketogenesis, and dehydration. B. Procedures 1. Experimental design. The experimental outline can be seen in Table 1 (page 97). Each dietary period lasted five days (beginning Sunday morning and ending Friday morning). But the diet was continued Friday and Saturday until the next diet was introduced Sunday morning (there were no ad lib. periods). The study began with a high carbohydrate control period (CHO I) followed by the Fat I dietary regimen (based on dairy cream). After the CHO II period, the Fat 11 regimen (based on corn oil and soy-protein) was introduced. The study terminated with a final control period (CHO III). The 24-hour urine was collected daily, and the subjects were weighted (nude) every morning. The urine was checked for ketones and pH, and the volume recorded. A sample was frozen for analysis of total nitrogen. The subjects recorded their own water intake for calculation of water-balance. A blood sample was taken the last day of each period (at least 12 hours after the last meal), and the second day of the high fat period, for analysis of blood constituents. The subjects continued with their usual physical activity during the 135 study, and no exercise was carried out except for the standard walk and the maximal oxygen determination with KB (twice during each experimental period in Experiment III). 2. Diggg, The calculations of the diets are shown in Appendices E to H, and the approximate composition shown in Table 17. The high carbohydrate control diet (CHO I) was basically the same as the one used during the exercise study (Experiment I). It consisted primarily of fruit, vegetables, cereals, and soy-protein products (supplied free of charge by Worthington Foods, Inc.1). The protein content was kept quite low in Experiment II(about 30 g/day) as compared to Experiment III (about 90 g/day). Three different meals were fed each day, but the same menu was repeated throughout each period. The same high carbohydrate control diet was used for all the CHO periods in both experiments. The Fat I regimen was based on heavy whipping cream (42% fat), the protein content was adjusted with 96% soya protein (from Fearn), and fruit canned in water was added. The cream was whipped, and the other ingredients mixed in a blender before being added to the cream. Sufficient mixture was made for the entire period and frozen. The mix- ture was fed three times a day for the duration of the period (Fat I). During the Fat II period, breakfast consisted of soya protein (Fearn), apples, avocado, lemon, and corn oil, mixed in appr0priate pro- portions and eaten with a small amount of granola. Lunch and supper consisted of a few carrots and a piece of soyameat (Worthington, chicken lWorthington Foods, Inc., 900 Proprietors Rd., Worthington, Ohio 43085. 136 style) with a dressing made from avocado (Fuertel), corn oil, and pecans. The same menu was repeated every day throughout the period. The composi— tion of the diets is given in Table 17. The caloric intake was adjusted to maintain weight during the first control period (CH0 1). In Experiment 11 the protein intake was made very low (about 30 g/day) in order to compare the results with those from the higher (90 g/day) protein intake of Experiment III. The daily intake of EAA was calculated and compared with the requirement in Appendices L to N. According to these calculations, the total sulfur—containing AA are low— est, but sufficient, relative to requirement. TABLE 17. Approximate composition of diets as percent of calories in Experiment II. Diet Carbohydrate Fat Protein Control 85 10 5 1 Fat I 6 89 5 Fat 112 6 89 5 1Cream. Corn oil. 3. Analysis. The urine was collected in polyethylene bottles which were kept sealed and in a refrigerator except during collection; pH and volume were determined immediately after the last voiding 1Supplied free by the CALIFORNIA AVOCADO SOCIETY, P. O. Box 4816, Saticoy, California 93003. 137 (in the morning), and a large and a small sample frozen for later use. The urine pH was measured on a Fisher pH/ion meter (model 420) using glass electrodes. The urinary nitrogen was determined using the semi- micro Kjeldahl method. Potassium sulfate was used to increase the boil- ing point and a 10% copper/sulfate solution was used as catalyst. One-half m1 of urine was added to the digestion flask with an Eppendorf pipet (No. 3130). The digest was transferred quantitatively to a 50 m1 volumetric flask and analyzed on a technicon autoanalyzer (No. 2) using the color reaction with sodium nitroprusside, the color being read at 660 nm. The relative concentration of acidic compounds in the urine when the subjects ate the various diets was determined by the automated metabolic profiling method (Sweeley et al., 1974). This was done with the LKB 9000 gas chromatographdmass spectrometer in interphase with a digital PDP 8/c Computer using the MSSMET program as described by Gates (1977). TrOpic acid was used as internal standard and analysis was done the last two days of each experimental period.1 Blood samples were taken in the postabsorptive state at the Labora- tory of Clinical Medicine, Lansing, Michigan, where the serum analyses (automated SMA 12 or SMA 18) were performed using standard procedures (Hycel).2 The following blood constituents were determined: serum bili- rubin, glucose, urea nitrogen, cholesterol, albumin, total protein, uric acid, calcium, phosphorus, and serum glutamic oxaloacetic transaminase 1These analyses were carried out by Stephen Carl Gates, in the laboratory of Professor C. C. Sweeley, of the Biochemistry Department, MSU. 21974 Manual. HYCEL, INC., P. O. Box 36329, Houston, Texas. 138 (SGOT). The blood was taken on the last day of each dietary period and also the second day of the high fat periods. On the last day of each experimental period, an arterialized blood sample was obtained by immersing the hand in warm water, making a cut in the end of the thumb with a stylet, and filling capillary tubes with blood. The tubes were sealed and used for analysis of pH and pCO2 on the radiometer (Copenhagen) pH-meter. Water intakes were recorded by the subjects for calculation of water-balance as described under the exercise study. The body weight was determined in the same way as during the exercise study. 4. Experiment III. This study was similar to Experiment 11, except that no metabolic profiling of organic acids by gas chromatog- raphydmass spectrometry was carried out on the urine. Sodium, potassium, chloride, and triglycerides were determined in the blood the last day of each dietary period in addition to the other blood parameters. Urine was collected each day except for weekends, but during CHO I, only during the last week, since the caloric intake was slightly modified after the first week. The dietary periods were expanded to 14 days for the high carbo- hydrate control periods, and to 11 days for the high fat study periods; there was no final control period in this study. The first high fat diet (Fat 1) consisted primarily of meat, butter, and some cheese and eggs. A few carrots were added. The meat was bought in one lot, cut in suitable pieces,and frozen. 0n the second high fat diet (Fat II), soy— protein (from Worthington), which contains very little carbohydrate, was 139 used; the fat came from corn oil and avocadoes (Fuerte). Mayonaise (with polyunsaturated fats) and some cottage cheese were also used. The control diet also contained soy-protein. The composition of the diets is given in Table 18. (The contribution of carbohydrate, fat, and protein was the same in the two high fat diets.) The calculations are shown in Appendices I, J and K. TABLE 18. Composition of diets as percent of calories in Experiment III. Diet Carbohydrate Fat Protein Control 60 25 15 High Fat 5 80 15 The following variables were determined with the Coulter Counter (Model S):1 RBC count, million per cubic mm Hemoglobin (HGB), gms per 100 m1 Hematocrit (HCT), vol. percent Mean corpuscular volume (MCV): cubic 0 Mean corpuscular hemoglobin (MCH): pug Mean corpuscular hemoglobin concentration (MCHC): percent The subject WH, having kidney problems, did not participate. KB and RH completed this series. A third subject, JG, started this experiment, but found the dietary regimen too unacceptable, and did not complete the study. 1These analyses were made by the Laboratory of Clinical Medicine, Lansing,‘Michigan. 140 VI. RESULTS FROM EXPERIMENT II A. General Observations The subjects generally felt well when consuming the high carbo- hydrate diet. The stool volume was large and transit time short dur- ing the high carbohydrate diet (daily bowel movements); whereas during the high Fat I dietary regimen (dairy cream), the stool was very small and bowel movements very infrequent. During the second high fat diet (Fat II), bowel movement was regular, but the stool volume was small. The subjects generally felt tired or had headache while consuming the high fat diet. On the second day of the first high fat diet, KB was very weak and could hardly walk; but when the blood was checked the next morning at the Laboratory of Clinical Medicine for ketone bodies (by the nitroprusside reaction), none was detected. Apparently, the condition was due to low blood glucose levels. The following morning when the blood samples were taken, the subject felt better. The blood glucose was elevated, probably due to an overshoot in glucose production. B. Urine Analysis Urinary pH, total ketone bodies, and total nitrogen excretion are reported in Table 19. For the last two days of each experimental period, a metabolic profile for organic acids was made on the urine. The amount of the different compounds is expressed relative to the internal standard (trOpic acid). In general, the profiles are very similar for the same subjects on the same diet, and there is good agreement in the general trend between the subjects, although the profiles show a pattern 141 TABLE 19. Urine pH and the excretion of ketones and nitrogen during Experiment 11 (5% of calories from protein). Each dietary period lasted five days. Data for KB Data for WH Date Diet pH Ket.l Total N pH Ket.1 Total N2 1/22 CHO I 7.00 4.86 5.86 5.55 1/23 CHO I 7.17 5.05 7.03 8.45 Mean CHO I 7.09 0 4.96 6.45 0 7.00 1/25 Fat I 7.23 0 5.83 1/26 Fat I 5.94 L-M 6.92 5.15 L 9.77 1/27 Fat I 5.62 L—M 10.57 5.28 M 8.34 1/28 Fat I 5.55 L-M 9.77 5.26 M—H 11.34 1/29 Fat I 5.54 M 7.50 5.21 M 10.58 Mean Fat I 5.98 8.12 5.22 10.01 2/ 5 CHO II 7.40 5.71 6.70 6.14 2/ 6 CH0 II 7.10 4.41 6.93 8.82 Mean CHO II 7.25 0 5.06 6.82 0 7.48 2/ 8 Fat II 6.43 0 11.55 5.97 0 2/10 Fat II 5.37 M 8.80 5.12 M 11.89 2/11 Fat II 5.75 M 7.65 4.82 M 11.64 2/12 Fat 11 5.26 M-H 6.24 4.84 M 10.96 Mean Fat 11 5.85 8.69 5.27 11.75 2/18 CHO III 5.97 3.91 6.96 6.80 2/19 CHO III 7.93 3.81 6.92 Mean CHO III 6.95 0 3.86 6.96 0 6.86 1Excretion of ketones in the urine in relation to control levels (zero) graded from low to high (L,M,H). 2Nitrogen excretion in the urine (g/day). 142 characteristic of the individual. Since the three carbohydrate periods are very similar, the average value for these periods is used as a reference to which the mean of the two determinations for each fat diet can be compared. These results are given in Table 20, listing the following compounds for which analysis was made: LA = Lactic acid B-OH—Bu = B-hydroxybutyric acid GL = Glycerol P = Inorganic phosphorous Gu = Glutaric acid Ad = Adipic acid HMG = B-hydroxy-Bdmethylglutaric acid VMA = Vanillyl Mandelic acid Hx = Hexuronic acid Hp = Hippuric acid The metabolic profiling was done according to the method of Gates (1977), which can detect many metabolic excretion products; but in this study, particular attention was given to compounds of special interest in evaluating the effects of the high fat diets, and to com- pounds that show unusually large variations. Because there are large individual variations in the metabolic profile for most compounds, these data are most useful on an intra-individual basis. 143 TABLE 20. Urinary excretion of organic acids during Experiment 11. (The figures express the peak area relative to that of tropic acid, after the method of metabolic profiling [Gates 1977].) Compoundsl Control2 Fat I2 Fat 112 Control3 Fat I3 Fat II3 LA .68 .23 .64 1.38 .56 .96 B—OH—Bu 2.41 8.90 9.38 1.83 8.90 8.12 GL .05 .37 .20 .06 .29 .16 P .89 .70 .82 1.04 .46 .25 Cu .73 .70 .60 .66 .08 .17 Ad .16 2.57 .88 .07 .46 .10 HMG 2.00 .38 .25 .32 .08 .10 Ci .30 .33 .52 .28 .11 .20 VMA 5.21 .35 .72 1.12 .27 .26 Hx 2.38 2.00 1.57 1.01 .41 .30 Hp 10.00 2.75 6.88 6.48 .39 1.26 1For explanation of these abbreviations, see page 142. 2Values for KB. 3Values for WH. 144 C. Blood Analysis On the last day of each period, an arterialized blood sample was taken and analyzed for pH and pCOZ. The results are given in Table 21. (The pCO2 data are not complete.) TABLE 21. Arterialized pH and pCO2 as mm Hg in Experiment II. 1 Diet le p0021 sz pCOz2 Control 7.43 7.40 Fat I 7.41 7.37 Control 7.45 34.4 7.42 38.0 Fat II 7.41 34.9 7.37 33.8 Control 7.43 7.40 37.0 1Values for KB. 2Values for WH. The results of the blood analysis are shown in Table 22. The abbreviations, units, and adult normal ranges are given below: Chol = Cholesterol, 150-300 mg/dl' Ca = Calcium, 8.5—10.5 mg/dl P = Phosphorus, 2.5-4.5 mg/dl Bili = Bilirubin, 0.2-1.5 mg/dl 1These pH values were measured by Gary R. Hunter in the Labora- tory of Professor Richard R. Heisey, Department of Physiology, MSU. 145 Alb = Albumin, 3.5-5.0 g/dl TP = Total protein, 6.0-8.0 g/dl UA = Uric acid, up to 8.5 mg/dl (for males) BUN = Blood urea nitrogen, 6-23 mg/dl Glu = Glucose, 70—115 mg/dl SGOT = Serum glutamic oxaloacetic transaminase, 2-45 u/L Na = Sodium, 135-145 mEq/L K = Potassium, 3.3-4.6 mEq/L Cl Chloride, 96-107 mEq/L TG Triglycerides, 50-155 mg/dl The cholesterol increase both on the saturated and unsaturated high fat diets, while total serum proteins, indicate some dehydration as will be discussed in the section evaluation of data. D. Weight and Water-balance The body weight, total water intake (TWI)(including water in the food), urine volume (UV), and water-balance (WB) (calculated as TWI—UV), are recorded in Table 23. The urine volume shows great variation during the high fat regimens, which may be due to variation in glycogen and lean tissue catebolism. VII. RESULTS FROM EXPERIMENT III A. Urine Analysis The pH was measured in the urine as before, and total nitrogen and ketone bodies determined. The results are given in Table 24. 146 .m3 How muasmom e .mM mom muHomoMm .HHo :uoo scum umMIIHH umm .aoouo mufioo scum quIIH uMhm .mqaueqfl momma mom .msoaumfi>ounno mo soauoooamxo pomH mm om m.HH n.o ~.n w.q o. e.m m.m CNN Houuaou om\~ we mm m.oa «.ma w.n w.q m. o.m w.m «om HH umm mH\N mm mm «.mH N.m m.n w.q n. m.m m.m NmH HH umm OH\N mo NOH w.m 5.5 o.n o.e w. m.~ m.oa «ma Houuaoo o \N mm NOH a.ma a.mH w.m o.m m. ¢.m m.0H OHN H use om\H we mm m.oa q.m «.5 o.q w. m.~ w.m mom Houuaoo «mN\H mm mm o.oH o.m N.o H.q m. ~.N m.m mmH Houuoou 0N\H oe on n.NH m.n w.c m.q w. m.~ m.0H mum HH mom MH\N on no H.5H ¢.o m.o m.e w. m.m N.0H mam HH umm 0H\N mm mm m.m N.m «.0 m.q m. m.~ m.m «ma Houucoo o \N no we m.mH m.m m.m a.q m. w.~ m.oH omm H pom om\H mm NHH o.wa 0.0 m.o o.q o. H.m m.m mud H umm c~\H mm mm N.oH m.m m.o m.q o. e.H ~.oa Hma Houuoou mm~\H Houm 3H0 23m oHnnm Ho GOHumammeo nomH o.mn o.HOH w.m H.H¢H HH umm wN\n m.mw m.NOH m.q q.QMH HH omu ¢H\m m.on m.m0H m.q m.qu H udh 1VmN\m N.wm m.NOH o.¢ m.NqH HH umm mm\m N.mn q.NOH H.q «.HQH HH omo ¢H\m ©.Nw N.¢OH N.q n.¢qH H mom mmm\q UH Ho M oz NuoHH moon oooaaoooouumm mamae 153 AmHomomuoo\w11v :oHuouusooaoo GHQOHwoeon uoHoomomuou sooz Auaoouonv cOHuouuooocoo aHAOHwoaon uoHaomsmuou cmoz AoHoosmuoo\: oHnsov osdHo> uoHaumsmuoo one: va :HAOHmoao: Ase 0Hn50\s0HHHHav assoc HHoo voOHm ooMH oooao one swam uoHv Houuaoo .ououomsonnmu anm QU‘QNQ Ausoouomv uHuooumaom NM mm on om as a.ma oH.n mm an em on m.ma an.a m~\n mm on mm as a.ea mo.m mm om mm mm m.mH as.o am\m mm on am He «.3H 65.6 mm aN mm 03 a.mH 65.4 maa\n mm on om Ne 6.3H em.o am mm mm as H.4H mm.o oa\m am on as so a.ma oo.m am mm mm on m.mH an.o a \m am am om mo m.eH ma.a on on mm as o.nH oa.o an \m mm on mm as a.mH wo.m mm ow mm as m.aH Ho.o o~\o mm om on so a.ma mo.m mm mm em «a n.3H no.8 m~\a mm om om m3 a.eH oa.m mm mm mm as n.4a 6H.m o~\e mm mm mm as H.3H ma.e mm mm mm as a.ma mm.o oH\e mm as em as m.ea mo.m mm mm om No a.ma om.e so \8 - m m use: mo: «>02 meow mom 6mm oozes mes: e>oz meom News Home a a mm wow moon HM How soon .HHH usoaHummxm wcHnow uHuooumaos mam anonoaom .oN mqm<8 154 TABLE 27. Water-balance and body weight data during Experiment III. Data for KB Data for RH Date Dietl TWI2 0V3 KB4 Weight5 TWI2 UV3 WB4 Weight5 4/11 CH0 I 2346 830 1516 63.6 2828 1430 1398 60.7 4/12 CHO I 2176 940 1236 64.1 3169 2100 1069 60.5 4/13 CHO I 2630 990 1640 63.6 3254 2100 1154 60.5 4/14 CHO 1 2517 875 1642 63.6 2800 1380 1420 60.5 4/15 CHO I 2517 950 1567 64.3 2828 1010 1818 60.5 4/18 Fat I 1434 950 484 64.3 2553 2095 458 61.1 4/19 Fat I 1944 1050 894 64.1 2071 1635 436 60.5 4/20 Fat I 2341 1550 791 63.4 1788 1350 438 60.5 4/21 Fat 1 1263 1330 133 62.7 2213 1120 1093 59.5 4/22 Fat I 2227 1040 1187 62.3 2185 2000 185 59.5 4/25 Fat I 1037 730 307 61.4 1930 1620 310 59.1 4/26 Fat I 1377 700 677 61.4 1759 1735 24 59.1 4/27 Fat I 1490 1190 300 61.4 2241 1200 1041 59.1 4/28 Fat 1 1604 950 654 61.4 1986 1570 416 59.1 4/29 CH0 II 2687 770 1917 61.4 3282 1360 1922 58.6 4/30 'CHO 11 2233 780 1453 62.3 2800 1230 1570 59.5 5/ 1 CHO 11 2460 740 1720 63.2 2885 1455 1430 59.5 5/ 2 CH0 II 1893 1195 698 63.6 3282 2330 952 59.8 5/ 3 CH0 II 2403 950 1453 63.6 2857 1665 1202 59.8 5/ 4 CH0 II 2233 1005 1228 64.1 3084 1960 1124 60.5 5/ 5 CH0 II 2233 1030 1203 64.1 2687 960 1727 60.5 5/ 6 CH0 II 2120 1020 1100 64.1 2573 1170 1403 60.5 5/ 9 CH0 II 2006 830 1176 64.5 3254 1450 1804 60.9 5/10 CHO II 2573 850 1723 64.5 2715 1080 1635 60.5 5/12 CH0 II 2630 940 1690 64.5 3310 1630 1680 60.7 5/13 CHO II 2516 860 1710 64.1 2176 1090 1086 60.7 5/17 Fat II 2176 1650 517 64.5 2451 1040 1411 60.2 5/18 Fat 11 2167 1130 1037 64.1 2734 1819 915 60.0 5/19 Fat 11 1770 1220 550 63.6 3310 2045 1265 59.5 5/20 Fat II 1714 940 774 63.2 2933 1020 1913 59.5 5/23 Fat II 1997 930 1067 62.6 2593 1350 1242 59.5 5/24 Fat II 1997 1000 997 63.6 3010 1520 1498 58.9 5/25 Fat II 1770 1100 670 63.2 2479 1420 1079 58.9 5/26 Fat II 1884 855 1029 62.7 3160 1610 1550 58.6 5/27 Fat II 1657 860 797 62.7 3500 1540 1960 58.6 1 CH0 (I and II) is high carbohydrate control diet, Fat I is the high fat diet based on butter and meatl, and Fat II is based on corn oil and avocados. The total water intake includes the water contained in the food. 24-hr urine volume (ml). TWI-UV. Body weight of the subjects in kg. Water-balance = (£wa 155 The data in Table 19 (page 141) show that increased ketone levels (above resting levels) began to appear in the urine the second day of the high fat diet, and steadily increased throughout the period, al- though some indication of stabilization of the ketone level is seen. There does not appear to be any difference between the two types of fat diets or between the subjects. From the data in Table 20 (page 143), there is seen a difference in the control level of ketones (the mean of the three control periods) between the subjects; but not much difference in the final level of B-hydroxybutyric acid during the high fat periods. For KB, the relative value increased from 2.41 to 9.38, about a four-fold increase. For WH, the relative value increased from 1.83 to 8.12, about a 4.5-fold increase. WH, however, reached the highest value of 8.9 on the fourth day of the dietary period; whereas, KB reached the highest value on the last (5th) day of the period. When we look at the ketone production as determined by the nitro— prusside reaction during Experiment III (with high levels of protein, Table 24 page 148), there again is no important difference in ketone production between the two diets or between the subjects. Large day-to- day variations are seen, which may be due to variation in the activity level of the subjects. There is a tendency toward a lower level of ketosis in this study compared to Experiment II, but the day—to-day variation makes it difficult to compare the results. The fat level in Experiment III was lower than in Experiment 11, but that was probably not important, since the carbohydrate intake was the same and it is the 156 deficiency of carbohydrate rather than an increase in fat intake that causes ketone production (Weis and Laffler, 1970). The higher intake of proteins can, however, be expected to decrease ketosis, as has been reported by Bell et a1. (1969). The gradual increase in ketone production seen in these experi- ments is in good agreement with the data of Bloom (1967); Bell et al. (1969); and Cahill (1976). It seems reasonable to conclude that there is no difference in the degree of ketosis produced by the two types of fat in these diets. Although ketosis is primarily determined by the level of carbohydrate in the diet, the level of protein may have a slight effect. B. Urinary Nitrogen Excretion, BUN, and Blood Glucose The effect of the different dietary regimens can be seen from Tables 19, 22, and 24 (pages 141, 143, and 148, respectively). The mean urinary nitrogen excretions (MUN) for the different dietary periods are given in Table 28; also included for comparison are mean BUN (MBUN), mean blood glucose (MGlu), and mean SGOT (MSGOT). It is clear that the mean values given in Table 28 do not reflect the variation throughout a period, as can be verified by comparing the data in that table with those in Tables 19, 22, and 24. For example, blood glucose tends to be above control values immediately after the high fat diet is introduced, when both glycogenolysis and gluconeogenesis are actively supporting the blood glucose level; but when the liver glycogen is depleted, and gluconeogenesis alone supports the blood glu- cose level, very low levels are often seen (Hultman et al., 1974). 157 TABLE 28.1 Mean values of BUN, blood glucose, serum SGOT, and the nitrogen excretion in the urine during Experiments II and III. Diet MUN3 MBUN4 MGlu5 MSGOT6 MUN3 MBUN4 Mglu5 MSGOT6 Experiment II (5% of calories from protein) Subject2 §§_ EH. CHO I 4.96 10.2 88 29 7.00 10.3 92 48 Fat I 8.12 18.8 90. 52 10.01 13.9 102 93 CHO II 5.06 9.3 88 32 7.48 9.8 102 43 Fat II 8.69 14.9 69 40 11.75 12.0 84 50 CHO III 3.86 10.0 77 25 6.86 11.5 86 33 Experiment III (15% of calories from protein) Subject 2 KB 311 Control 11.01 15.0 81.7 21.8 10.24 10.3 89.4 21.8 Fat 1 13.73 18.2 68.4 31.7 14.62 11.5 83.8 26.3 Control 10.51 12.6 82.8 20.9 10.79 11.3 93.1 23.9 Fat II 16.67 22.9 74.8 28.7 15.10 14.1 84.4 27.0 lThe caloric intakes were: KB, 2400 kcal; WH, 2700 kcal; and RH, 2300 kcal. Weight of subjects: KB = 140 lbs; WH = 230 lbs; RH = 135 lbs. The mean nitrogen excretion for each dietary period (g/day). The mean BUN for each dietary period (mg/d1). The mean blood glucose concentration for each dietary period (mg/d1). O‘U'lkLON The mean SGOT level for each dietary period (u/L). 158 Likewise, systematic changes are seen in the urinary nitrogen excretion during the high fat periods, which are probably due to metabolic (hormonal) adaptation to the prolonged exposure to a low protein diet as described by Cahill (1976). It should be emphasized that the protein intake is constant throughout each experiment. Despite such shortcomings, the mean values are useful in evalu- ating the results. The urinary nitrogen excretion during the control period for Experiment II is higher than expected on the basis of the protein intake, and may indicate a negative nitrogen balance. However, only two measurements were taken for the control period, so the values may be too high. The urinary nitrogen excretion for the high protein diet (Experiment III) agrees well with the protein intake. The urinary nitrogen excretion during the fat periods increased about 4 g compared to the control, regardless of whether the nitrogen intake was 5% or 15% of the calories (Table 28). This means that about 25 g of endogenous protein were used for gluconeogenesis. The corre- sponding BUN was elevated (particularly in KB) during Experiment 11 (up to 100%), but only increased to a minor extent in the other subjects (Tables 22 and 25, pages 146 and 151, respectively). The BUN increased to about 20 mg/dl when KB consumed the high fat diet, regardless of the protein intake. The only exception to this was the Fat 11 diet in Experiment II (Table 22), where the BUN was almost back to normal the last day of the period. The reason for this may be found in the steadily declining urinary nitrogen excretion during the period, as seen from Table 19 (page 141). At that time the subjects had had a low protein 159 intake for four weeks; and KB was probably beginning to adjust to it. The long-term protein sparing effect may be comparable to that dis- cussed by Cahill (1976). When the mean BUN's for the two protein levels are compared for KB during the control periods, it is seen that a tripling of the protein intake increased the BUN value about one-third, which is in good agree- ment with the results of Addis et al. (1947) who reported much higher absolute values. The difference may be due to the fact that their blood samples were taken about four hours after lunch; whereas, our results were secured with blood drawn after an overnight fast. Generally, the blood glucose level fell during the high fat diets; but as long as liver glycogen was available for glycogenolysis, blood glucose levels appeared to be elevated. When the liver glycogen is depleted, the blood glucose level falls, often to hypoglycemic levels, indicating that gluconeogenesis cannot keep up with glucose utilization. As the subjects continued to consume the low carbohydrate diet (high fat), ketone production increased, allowing the blood glucose levels to return toward normal (because of its glucose sparing effect). This probably explains the improved well-being reported by the subjects after the first few days of the high fat diets. These results agree well with the findings of Hultman et a1. (1974); and Cahill (1976). The subjective feelings of weakness and fatigue were always most severe on the second or third day, corresponding to the low blood glucose levels before ketone production became maximum. 160 According to White et al. (1973a), the amino-transferases increase during fasting to facilitate gluconeogenesis from amino acids (AA). As seen from Table 28 the values of SGOT in the blood when the subjects were consuming the high fat diets increased, compared to the control periods. This again indicates that gluconeogenesis from AA takes place during carbohydrate deprivation, but it can also be an artifact due to hemolysis of RBC (which was occasionally seen during the high fat studies). The relationship between BUN, urinary nitrogen excretion, and blood glucose is plotted for KB in Figure 2 for Experiment 11 (low pro- tain intake), and in Figure 3 for Experiment III (high protein intake). The data for RH during Experiment III are plotted in Figure 4. Looking at Figures 2 and 3, we see a fairly good positive corre— lation between the urinary nitrogen excretion and the BUN; however, when the high carbohydrate control diets were consumed, the BUN levels were lower than would be expected from the urinary nitrogen excretion, indi- cating that factors other than nitrogen excretion affect the BUN level. Similar results are seen in Figure 4 for RH in Experiment III, although the BUN's were not quite so high during the high fat periods. A decrease in BUN when wheat protein was fed, was observed by Bolourchi et al. (1968). In most cases the first blood sample was taken 48 hours (or more) after introduction of the high fat diet. At that time, all the liver glycogen should have been depleted. According to Hultman et al. (1974) no compensatory increase in gluconeogenesis takes place under resting 161 A.maHououa Boum moHuono «0 Mn Sues mooomoo>m was HHo ouoo EOHH .HH umm masons muHmo Boom .H ummv .HH uooaHuomxm wcHHov HM now omooon wOOHn one .ZDH .moHus oSu :H GOHuouoxo sowouqu .N MMDOHH 1152 4* 00 ten r8H .oHH vomH «H r mu ammo CH oEHu .sHo mammHm nV11 Ho\wa .oao ‘ hoo\w z: N.HHM=HUHKH ad 23m Hoxhzou HH Hm was HHo Shoo aouH .HH umm momma cam Houuon aouH .H ummv .HHH ucoBHuomxH wsHuow HM pom mmooon vooHn one .ZDH .oaHuo osu aH GOHumnoxo cowouqu .m HMDUHH 164 .1 m EGHH VQHH YONH anvxwa 3H0 .eH .HH va amousHo mamon 22H HH Hm one HHo :uoo aoum .HH mom momma one umuuon aouH .H ummv .HHH unmeHuoaxH onuow HM pom mmooon HOOHH cam .ZDH .ocHuo onu GH GOHumHoxo dmmouqu .q HMDUHH 1156 too vooa ‘OaH a‘ muzeouam mm fiw NH ma AOMHZOU xvIIIIIIIIIII:A\\\\\xv1111 H Hx n» «a I 4 o \\ 2 . «a J «A qoxazou . 0H 2.}... 23 4 f Hu\_ 2: 167 conditions, which explains the low blood glucose level seen at this time. As carbohydrate deprivation continues, the increased ketone pro- duction decreases the utilization of glucose, enabling gluconeogenesis to maintain the blood glucose level (Cahill, 1976; and Hultman et al., 1974). During the Fat 11 regimen (polyunsaturated fat and soy-protein) in Experiment III, no such adjustment is seen. Both the urinary ketone bodies and the urinary nitrogen continued at high or increasing levels, while blood glucose remained low or even decreased. The reason for this is not known. During the Fat I period in Experiment II, the first blood sample was taken 24 hours after the dietary period began, which explains the high blood glucose (112 mg/dl) seen in KB (Figure 2). Apparently, the combined effect of gluconeogenesis and glycogenolysis produced an over— shoot in the blood glucose level. The blood glucose level during the high fat diets was sufficiently low, so that if challenged with exercise, the blood glucose level could severely limit Performance. However, under such circumstances, gluconeogenesis from protein is further in- creased (Felig, 1975). In most cases, gluconeogenesis (as measured by the increase in urinary nitrogen excretion and BUN) reached a peak early in the high fat dietary periods and then gradually decreased toward the end (Figures 2, 3, and 4). This agrees well with the results of Pozefsky et al. (1976), who found a decreased serum alanine, and decreased alanine release from the forearm muscle during a 60-hour fast compared to a 24- hour fast. The decreased release of gluconeogenic AA during prolonged 168 fasting is believed to be due to increased levels of branched chain AA (Cahill, 1976). For some reason, such a decrease in gluconeogenesis was not seen during the Fat II regimen (corn oil, soy-protein) in Experiment III; whereas, it was observed on the same type of fat diet with a low protein intake (Experiment II). Though the reason for this is not known, it could, presumably, be due to either an altered AA pro- file caused by the source of protein (soy-protein), or an altered hormonal response. The linear regression lines and correlation coefficients (r) between UN and BUN, and between UN and Glu (in plasma) during Experiment II, were calculated for KB; for Experiment II: 0.71 BUN 1.72 UN + 3.24; r Glu -4.78 UN + 111.0; r = -0.51 For KB during Experiment III: BUN = 1.16 UN + 1.81; r 0.75 Glu = -1.20 UN + 92.7; r = -0.50 C. Other Blood Constituents The mean values for calcium, inorganic phosphate, bilirubin and uric acid during the different dietary periods are given in Table 29. There were no important differences in blood calcium level, which is not surprising in view of the strict homeostatic regulation of this ion. During Experiment II, there was a tendency for the inorganic phosphate concentration to be higher during the high fat diets. However, in Experiment III, this was the case only when the protein came from meat. As seen from Tableifl.(page 144), the pH of arterialized blood decreased 169 TABLE 29. Serum calcium, phosphorous, bilirubin, and uric acid (in mg/dl). Diet Serum Calcium Phosphorous Bilirubin Uric Acid Experiment II Control 10.21 9.82 1.61 2.92 .6 .8 91 8.42 Fat I4 10.1 10.3 3.0 3.4 .7 .9 .1 15.9 Control 9.9 10.3 2.5 2.9 .5 .8 2 7.7 Fat 115 10.4 9.9 3.1 3.3 .8 .7 .2 10.3 Control 9.9 9.9 2.2 3.6 .5 .6 0 6.7 Experiment III Control 9.71 9.53 2.81 3.43 .8 .8 .31 5.13 Fat 16 9.9 9.5 3.2 3.8 .7 .7 .7 6.1 Control 9.5 9.3 2.8 3.5 .6 .7 .8 4.4 Fat 115 9.4 9.2 2.8 3.3 .7 .8 2 4.9 1Data for KB. 2Data for WH. 3Data for RH. 4Fat from dairy cream. UI Fat from 0‘ Fat from corn oil and avocado. butter and meat. 170 during the high fat diets, and the urinary pH was definitely lower (Tables 19 and 24, pages 141 and 148, respectively). Acidosis has been reported to increase calcium loss in the urine (Farquarson et al., 1931; and Lemann et al., 1967), and a high protein diet reportedly gives a negative calcium balance (Anand and Linkswiler, 1974). Furthermore, it has been shown that Eskimos have a relative bone deficiency in calcium (Mazess and Mather, 1973). So apparently, in Experiment II, the in- creased bone mobilization due to acidosis raised the blood level slightly, but during Experiment III, the phosphorus in the meat appeared to be the dominating factor. In Experiment II, bilirubin was definitely increased when KB consumed the high fat diets, but there appeared to be no difference in Experiment III. This probably indicates a decreased hemoglobin degradation during the high carbohydrate diets in Experiment II; but despite occasional lysed RBC,during the high fat diets in Experiment III, there did not appear to be an increase in bilirubin above control. Uric acid showed a significant increase in Experiment II, both when fat came from dairy cream and corn oil; but during Experiment III, there was little, if any, increase in uric acid when the subjects con- sumed the Fat II diet based on corn Oil. The increase in serum uric acid seen in Experiment III when fat came from.meat and butter, was less than the increase seen during the high fat, low protein diets (Experiment II). The level of serum uric acid in WH was very high for both control and high fat diets, indicating that this subject had a problem with purine metabolism. For this reason, WH did not participate in Experiment III. 171 Increased blood uric acid during starvation was reported by Lennox (1924). Christofori and Duncan (1964) concluded that during fasting or consumption of high fat diets, decreased levels of glucose and/or AA in the glomerular filtrate increased uric acid reabsorption in the renal tubules because of decreased competition from these com- pounds. If this indeed is the case, it can explain the results seen in this study. D. Metabolic Profiling of Organic Acids in the Urine When interpreting the excretion of organic acids in the urine, it is important to consider the different sources for an altered excre- tory pattern. When the diet is altered, certain compounds present or not present in a particular diet, will affect metabolites present in the urine. Furthermore, drastic alterations in the diet are likely to affect the intestinal microflora as reported by Reddy et a1. (1975), which can greatly alter the metabolites available for urinary excretion. In addition, the composition of the food consumed can greatly alter the metabolic setting, so that products such as ketoacids are produced in altered amounts; and the secretion of hormones (catecolamines) causes an alteration in the excretion of hormonal degradation products. Finally, alteration can be expected due to individual variation in the metabolic pathways and physiological responses. From Table 20 (page 143), a basic difference is easily seen between KB and WH when they consumed the same foods in the same amounts. KB had a relatively low lactic acid excretion, but high B—hydroxybutyric acid excretion when consuming the control diets; whereas, the opposite 172 was true for WH. This indicates that WH had a higher reliance on carbo- hydrate metabolism than KB, which can also be seen in Figure 1 (page 107) where the RQ's for WH were higher during exercise. There was no differ- ence in B-hydroxybutyrate production during the high fat diets. Train- ing is known to increase reliance on fat metabolism, and on the basis of the RQ values, KB appears to have been in better physical condition. WH, being 6'3" tall and weighing 230 lbs compared to KB's 5'7" and 140 lbs., is about 50 lbs overweight; and it is possible that high carbohy- drate substrate utilization is caused by obesity. To my knowledge, such a relationship has not been investigated. Glycerol excretion was very similar in both subjects, and in- creased about six-fold during the high fat periods. This should be ex- pected, since fat was the principal substrate during these periods. Phosphate excretion showed no change in KB and even decreased in WH dur— ing the high fat dietary periods, despite the elevated plasma levels during that regimen. This seems to indicate an altered renal clearance. Glutaric acid, one of the end products of lysine metabolism, showed no change in KB, but decreased in WH when he consumed high fat diets, probably indicating an alteration in the intestinal microflora (White et al., 1973c). Adipic acid increased tremendously during the Fat I diet (dairy cream) in Experiment II, which probably was due to w—oxidation of the short chain fatty acids found in butterfat (White et al., 1973b). Citric acid showed no variation in either subjects, which is interesting, since it has been implicated in acid—base balance (Tischler 173 et al., 1970). Vanillylmandelic acid (VMA) in the urine was much higher during the control period for both subjects. Being an end product of catecolamine degradation, it could indicate a decreased excretion of catecolamines during the high fat dietary periods; however, the elevated excretion when the control diets were fed, may have resulted from the fact that the bananas in that diet increased the excretion of VMA (Zilva and Pannall, 1975). Variation in hexuronic acid is apparently due to variation in bacterial action and dietary constituents (White et al., 1973). The high excretion of hippuric acid during the CHO periods was probably due to the high content of benzoic acid in the control diet (White et al., 1973). E. Cholesterol, Triglycerides and HMG During most high fat periods, the serum cholesterol level in- creased and then decreased again during the following control period, as seen from Tables 22 and 25 (pages 146 and 151, respectively). The serum cholesterol from the last day of the period probably best represents the effect of the particular dietary regimen. For this reason, the serum cholesterol level (mg/d1) from the last analysis is given in Table 30. Also included is the relative level of hydroxymethylglutarate (HMC) in the urine (the average value for the last two days) from Experiment II, and the serum triglyceride concentration (mg/d1) for the last day of each dietary period in Experiment III. As expected, cholesterol increased when the fat came from satu- rated fat; but it was surprising that the serum cholesterol level also increased when the fat came from corn oil, despite the absence of any 174 TABLE 30. Serum cholesterol, urinary hydroxymethylglutarate, and serum triglycerides. Experiment II 1 2 Date Diet Subject Chol HMG 1/23 CH0 1: KB 181 1.97 1/30 Fat I KB 230 .38 2/ 6 CH0 116 KB 184 4.00 2/13 Fat II KB 225 .25 2/20 CHO III KB 195 1.05 1/23 CHO 1: us 205 .32 1/30 Fat I WH 210 .08 2/ 6 CHO II6 WH 194 .47 2/13 Fat II WH 204 .10 2/20 CHO III WH 220 .22 Experiment III 1 3 Date Diet Subject Chol TG 4/16 CH0 1; KB 212 4/29 Fat I KB 292 82.6 5/14 CHO II6 KB 234 73.2 5/28 Fat II KB 186 98.2 4/16 CH0 1; RH 156 4/29 Fat I RH 214 76.3 5/14 CHO II6 RH 159 83.5 5/28 Fat II RH 144 73.0 Serum cholesterol (mg/d1) from the last day of the dietary period. Hydroxymethylglutarate (relative to tr0pic acid) in the urine. Triglycerides (mg/d1). High carbohydrate control diet. Fat from dairy cream. Fat from corn oil and avocadoes. \IO‘U‘I-l-‘UNH Fat from butter and meat. 175 cholesterol in the diet. However, this was seen only in Experiment II. Anderson et a1. (1971), found no effect of dietary protein on serum cholesterol, provided that the minimum requirement was met. Intakes below the minimum requirement usually reduce plasma cholesterol, as seen in kwashiorkor and semi—starvation (Keys et al., 1950). In Experiment III, a decrease in plasma cholesterol occurred when the subjects were fed the corn oil soy—protein diet. Estimation of cholesterol by the methods routinely used in clinical laboratories is usually not very reliable (White head at al., 1973b); however, the Hycel method used in these experiments has a standard deviation of 2.3 over a lO-day period, and our results appear very consistent. The serum cholesterol levels in Experiment II seem to be negative- ly correlated with the urinary excretion (and therefore, presumably the blood level) of hydroxymethylglutaric acid (HMG) (Table 30). It is generally believed that HMGA-CoA reductase is the controlling enzyme in cholesterol synthesis; and this enzyme is competitively inhibited by HMG (Beg and Lupien, 1972). HMG-CoA is converted to HMG by the enzyme hydrolase in an essentially irreversible reaction. It appears that diet affects the level of HMG, which in turn seems to control cholesterol synthesis. This hypothesis needs further investigation. The serum cholesterol and relative urinary HMG concentration for KB during Experi- ment II are plotted in Figure 5; serum cholesterol and triglycerides during Experiment III are plotted in Figure 6. The linear regression lines and correlation coefficients for cholesterol and HMG, were calculated for KB and WH during Experiment II. FIGURE 5 . 176 Serum cholesterol and hydroxymethylglutaric acid (HMG) for KB during Experiment II. (The control diet was a high carbohydrate diet. During Fat I the fat came primarily from dairy cream and during Fat II, primarily from corn oil and avocadoes. The concentration of HMG is relative to the internal standard.) 177 Chol ns/lOO m11 250- Cholesterol 2251 200‘ CL._‘ CONTROL FAT I CONTROL FAT II CONTROL 170‘ HMG (Relatlve) L '4 \ , 3 L2 \ HMG P]. a? 25 1, 1, 1} £5 1? ‘_ Date '— FIGURE 5 178 A.mmovmoo>m was HHo Chou Eouw zaflumeflum .HH new wcausw new Emouu hpamp Scum >HHumaHun memo wow .uoHc Houuaoo mumuvhnonumo swan ocu mmumofivafi omov wzu H nah wcfiuzo .HHH ucoefiuomxm CH mx How mowfiuoozawfiuu cam Houmummaoso mammam .o mmDOHm 1179 o mmson m~ mm mm nN Hm ma NH ma ma ad a A m n Hun o~ RN nw nN aw ma an ma ma Ha mus UHNQ d 4 d1 1 1 a m J u A U 4 q q q 1— d 4 d 1 d 4* ‘4 dw q d OQH _ _ . _ . oom _ VF J _ _ _ _ 1 on“ _ _ _ 00 fl _ _ _ .. con . . _ om .. _ _ .04. _ _ . .2. _ . _ :2 28 E _ o5 _ . Ha ooa\ma own ONL .u.H _ 180 For KB: Chol -11.8 HMG + 221.1; I -0.79 For WH: Chol -0.57. -33.1 HMG + 214.5; r The poor correlation for WH can be expected, since he showed relatively little variation in either variable as the diet was altered. The serum triglyceride levels during the high protein study were not related to the cholesterol level; and for KB they decreased when the control diet was consumed, but for RH, they increased. Glueck et al. (1969) found that the average rise in TG was 100 mg/ml in normal people when they changed from a diet containing 40% of the calories from carbo- hydrate to one with 80% of the calories from carbohydrates. However, people who subsist on starchy foods do not appear to have elevated TG levels (Florey et al., 1973). Antonis and Bersohn (1961) found that when white Europeans changed from a European to an African diet (based on corn), the plasma TG rose and cholesterol fell; however, after eight months on the African diet, the TG-levels had returned to normal. Since KB habitually consumes a high carbohydrate diet, no increase in plasma TG would be expected. In RH there was not much variation either, but at least the change was in the anticipated direction. It should be pointed out that all results are well within the normal range. F. Body Weight Loss and Water—Balance Using the data in Table 24 (page 148% the average water-balance (WB) from the three control (CHO) periods and from each of the high fat 181 periods was calculated. The results for Experiment 11 are given in Table 31, and in Table 32 for Experiment III. TABLE 31. Water-balance and body weight loss during Experiment 11. Data for KB Data for WH Diet Water-Balancel Weight Loss2 Water-Balancel Weight Loss2 Control3 1283 0.0 2142 0.0 Fat I 478 3.2 1226 5.5 Fat II 520 2.0 925 5.5 lRelative water-balance in g/24 hrs. 2Weight loss in kg for the entire dietary period. 3Mean for the three CHO periods. TABLE 32. Water-balance and body weight loss during Experiment III. Data for KB Data for RH Diet Water-Balancel Weight Loss2 Water-Balancel Weight Loss2 Control3 1435 0.0 1457 0.0 Fat I 603 3.0 489 2.0 Fat II 826 1.8 1423 1.6 lRelative water-balance in g/24 hrs. 2Weight loss in kg during the entire dietary period. 3Mean for the two CHO periods. In Table 33 the actual weight loss (AWl) is compared with the weight loss based on the water-balance (AWZ), and the maximal loss 182 TABLE 33. Total weight loss during each dietary period. Study Diet AWl AWZ AW3 Subject Fat I 3.2 4.0 3.0 KB 4 Fat II 2.0 3.8 3.1 KB Experiment II Fat I 5.4 4.6 4.2 WH Fat II 5.4 6.1 4.4 WH Fat I 3.0 9.2 3.5 KB 5 Fat II 1.8 6.7 4.5 KB Experiment III - — Fat I 2.0 10.6 3.9 RH Fat II 1.6 0.3 4.1 RH 1The observed weight loss in kg for the entire period. 2The weight loss calculated from the water-balance data. 3Weight loss calculated on the basis of estimated glycogen degradation and calculated lean tissue catabolism (see text for explanation). 4Each period lasted five days. 5Each period lasted eleven days. 183 calculated from the estimated (max.) glycogen loss and the lean tissue loss (AWB). AW was calculated from the average water-balance during 2 the control period minus the average water-balance during each of the fat periods multiplied by the number of days the subjects were on the particular diet. AW3 was calculated from the loss due to glycogen depletion, as found by Olsson and Saltin (1970) and the lean tissue loss due to protein catabolism estimated from the difference between the mean nitrogen excretion during the fat periods and the mean nitrogen excre- tion during the control periods. The value was converted to protein, multiplying by 6.25, and to lean tissue mass, multiplying the latter by 5. This figure was then multiplied by the number of days the subject consumed the diet. From Table 33 it is seen that the weight loss calculated from the water-balance results is not very reliable. The two primary reasons for this are fluctuation in the actual water loss (sweat and insensible perspiration), and variation in stool moisture content. From Table 27 (page 154), it is seen that the actual water loss varied from about 1100 (ignoring one very low value of about 700) to about 1700. Even if the actual water loss was unaffected by the diet, but the mean water loss is in error by 300 ml/day, it would give an accumulated error of more than 3 kg for an ll-day period. The stool volumes on the high carbohydrate diets were usually 500 g or more, compared to almost no stools during the saturated fat diet (Fat 1) though some subjects had occasional diarrhea); and a small but consistent stool during the unsaturated fat diet (Fat II). Assuming 80% water in the stool, this could account for 184 a difference of about 400 g water/day, or about 4.5 g for 11 days. Taking these variations into consideration, the inconsistency between Awl and AW2 can be explained. Water intake and urine output are plotted in Figure 7 and in Figure 9 for KB during Experiments II and III, respectively; and in Figure 11 for RH during Experiment III. Generally, the difference became much less during the high fat diet, indicating that endogenous water was available. Figures 8, 10, and 12 show the total water-balance, and the body weight. In Experiment II, the water-balance reached the lowest value at the beginning of the study, and then gradually increased toward the control value; but during Experiment 111, there is no con- sistent change throughout the study. This indicates first of all, that there was no correlation between the degree of water loss and urinary excretion of ketones, since ketone excretion was at control level at the beginning of the period, but in- creased toward the end. The large water loss on the first day was probably due in part to liberation of the water associated with liver glycogen (about 350 ml), which is depleted within the first 24 hours (Hultman et al., 1974; and Bergster and Hultman, 1972). In Experiment III, the water—balance was followed during the transition from the high fat to high carbohydrate diet, and as seen from Figures 10 and 12 and Table 27 (page 154), there was an increased water retention the first few days, indicating resynthesis of glycogen. Secondly, the more constant water—balance during the high fat periods in Experiment III may indicate a glycogen sparing effect of the FIGURE 7. 185 Water-balance for KB during Experiment II. (CHO indicates the high carbohydrate control period. During Fat I the fat came primarily from dairy cream and during Fat II, primarily from corn oil and avocado. Total water intake includes both drinking water and water contained in the food. Urine is the 24-hour urine volume.) 186 2500, 2000. Total H O 1500. Incaki 1000,. ‘ 4—0 Urine 500. cuo ' PAT | cuo ' FAT I CH0 1 i | | A ' cl 1 A l; L r L n l J L I; A J L ‘L Date 1-22 2‘ 26 28 30 2-1 3 5 7 9 11 13 15 17 19 FIGURE 7 187 FIGURE 8. Body weight, ketone excretion, and water-balance for KB during Experiment II. (CHO indicates the high carbohy- drate control periods, and Fat, the high fat dietary periods. Body weight is the cumulative weight. Ketones indicates serum concentration above control values (determined by the nitroprusside reaction). Water-balance is the difference between the total water intake and the 24-hour urine volume.) 188 I ' I ' Weight 33) 0.11—.0 I I | .140 I \ ' o N Weight I I *135 I I I I I I I I I I I >130 I I I I Ketones ' I ' . High I I I I Med I I I I "Low o-o—o- I . .0—0—0 0 I I I I I I | I I I l I 111 1120 AI CEO I FAT I CHO I FAT . c110 I I I I 1500 - ' ' ! | O I I I I I I 1000 " I '\O-O H20 balance I I I 500 L I I I I 1 Al 1 l 1‘ . l L A 11 L j; I] L 1 1 Is. Date 1-22 24 26 28 30 2-1 3 5 7 9 11 13 15 17 19 FIGURE 8 A.ma:Ho> mcfiud “doslqm mnu ma Guam: .poom MSu Ga pm Icfimuaou umumB paw “mums wcflxawuv nuon mowsaunw umum3 Hmuos .ocmoo>m mam HHo cuou Eouw hawumafium .HH umm wafiuaw cam umuusn was puma aoum mafiumswum mama umm mxu .H umm wafiusa .pOHuma Houuaou mumuwmsonumo swan mnu mmumowpcw omuv .HHH unmaflummxm waflHSp mM How muamamnlumumz 189 .a MMDOHh 190 sump m MMDOHm ow “N mm mm AN aa aa ma na 33 o a n HIn oN RN mN nN HN ma “a na na HHIo A 4 a I u a I WI 4 A < 1 I1 3 4 M— U u u w a _W q 1 . _ _ . . . . . con _ . _ _ News: . . . . . oooa O _ ._ . coma %£ aoooe . . 1 ooo~ . _ i . . u oonu _ . . _ . . . coon .5; 0:0 H mean: on» was mxmucfi HmumB Hmuou can demands moamumwmaw ecu mH ousmamnlnmumz paw mpasom GH ma unwam3 mwom .ovmoo>m was Hfio once no woman wm3 HH umm cam .umoa was umuunn co momma mm3 H umm .poaumm Houuaoo mumnvmnonumo swag mfiu mwumofiwafi omov .HHH uaoawuomxm wcflusp mm Mom moamHmnIHmum3 was unwflmz hpom .OH mMDuHm 192 CH mMDUHm mu ma n~ mm "N aa Na na ma HA o n n HIn «N 5N am an AN Qu Na n4 MA MHIQ “I1 n q . 4 q «d u a . 4 q j u 4. 4 q . q 4 .4 . . _ _ _ — . _ fl . can _ _ oucmamn . ofil _ . . . 1.82 O .- — I . . _ . cone _ _ . . _ _ ._ . _ . 1.82 N _ — _ o m a; one a: one 1. . _ . 1 mg I _ _ uzwao3 _ ocaw _ _ _ a mafia _ _ . 3 3303 A: new?) 193 a.mE=Ho> mafia: usoslqm msu mH moan: .poom onu CH pmcHMuaoo nouns was “mums wcfixafiuw :uon mopdaoafi mxmucfi umums HmuoH .opmoo>m was HHo auoo no woman mdB HH uwm mafi£3 .ude was Houusn :0 pumps mos H umm .pofiuoa Houuaoo oumupmnonumo swag was moumUficcH omov .HHH unmafiummxm wafiunw mm Mom ooamfimnlumumz .HH MMDOHm 194 was HH NMDDHm mN RN mm mm Hm mm NH ma ma Ha m n n n film ow KN nN MN HN ad NH we mu HHIQ W - u I q - uI— q dI - d a 1 1 < u— 1 q 4 q I 41 q 1 . _ _ . z scan: msu was mxmuafl umum3 HouOu ozu som3uwn monouwHHHp mzu wH 60:6HmnIuoum3 paw .mpasoa GH mH unwfim3 Hpom .opmoo>m was HHo Shoe no woman was HH umm muses was “wanna no woman mos H umm .pOHuom Houucoo mumuvhsonumo nwfiz mnu mmumonGH omov .HHH ucweHuomxm wcHusw mm mom moamHmnIHmumB can usmHms hwom .NH MMDUHH 196 oumo mNH omH mMHT am mm mm MN HN oH NH mH MH HH a n n Q H” Him aN 5N nN nN AN ad NH “A Md Hale L a . . 4 I . ._ 4 q q a a I a d d .qfi - a q . q H. a q _ . _ con . _ OOOH Oond uocmHmn +‘ nous: . own HI . _ . _ r E . one at . 28 . _ 238301016; _ T 197 protein compared to Experiment II. Additional support for this concept comes from the decreased rate of weight loss during the high fat diets in Experiment III (Table 33, page 182). That the decreased weight loss during the high protein diet (Experiment III) was due largely to a glycogen sparing effect rather than a protein sparing effect, is indi- cated by a similar increase in urinary nitrogen excretion during the high fat diets in Experiments II and III. The caloric content and physical activities were unaltered during these studies. If high dietary protein levels indeed have a glycogen sparing effect, this could be important for endurance performance. From Figure 8 (page 188), it is seen that the rate of weight loss was highest in the first part of the study periods, when the water— balance was lowest; however, the rate of weight loss was also highest in the first part of the high fat periods during Experiment III, but in this case, there was no increase in the water-balance. From Table 33 (page 182), it is seen that the weight loss calculated from the water— balance data is grossly in error, probably due to the accumulation of errors during the longer periods of this study. When the actual weight loss is compared with.the weight loss based on glycogen and lean tissue mass, it is seen from Table 33 that in Experiment III, the calculated weight loss (AW3) was always higher than the actual weight loss (AWl), indicating the possibility of a glycogen sparing effect during a high protein diet. Hultman and Nilsson (1975) have shown that in subjects fed a low carbohydrate diet, the liver glycogen is not replenished as it is in some animal species, indicating 198 a lower gluconeogenic capability in man. In the resting state, muscle glycogen is only slowly depleted during starvation (about 50% in four days) according to Hultman and Nilsson (1975). The rate of muscle depletion may very well depend upon the level of protein intake, since the branched chain amino acids apparently are oxidized for energy util- ization in the muscle; and the AA profile appears to control the release of alanine from the muscle (Felig, 1975; and Cahill, 1976). During Experiment II, it is apparent that the weight loss calcu- lated from glycogen and lean tissue mass catabolism (AW3) was less than the actual weight loss (AWl) in most cases. This could possibly be explained by glycogen super-compensation during the high carbohydrate period (Bergstr3m and Hultman, 1966). However, considering the rela- tively short dietary periods, it is not very likely that the muscle glycogen was depleted (Hultman and Nilsson, 1975), indicating that part of the weight loss had a different origin. G. Dehydration The plasma albumin (Alb) and total protein (TF) concentration on the last day of each experimental period are listed in Table 34. The average value for the three control periods and the average from the two fat diets are used. It is seen that for KB, the plasma albumin in- creased 92 on the high fat diet, while the total protein increased 11%. For WH, the respective values are 4% and 10%. These results seem to indicate a 10% decrease in the extracellular fluid volume (EFV). During the Experiment III, more blood samples were taken and the dietary periods were longer; for this reason, the mean albumin (Alb) and 199 TABLE 34. Plasma albumin and total protein concentration at the end of the control and fat periods during Experiment II. Data for KB Data for WH . 1 2 l 2 Diet Alb TP Alb TP Control3 4.4 6.4 4.7 7.1 9414 4.9 7.1 4.9 7.8 Serum albumin (g/dl). Total protein in serum (g/dl). Average for the last day of the three CHO periods. L‘LDNI-I Average for the last day of the two fat periods. the mean of the total protein concentration in the blood were calculated for the fat periods and for the control periods (from Table 25, page 151), and the results are shown in Table 35. No systematic change is seen in Table 25, and the data in Table 35 shows no indication of dehydration during the high fat periods of Experiment III. TABLE 35. Mean plasma albumin and mean total protein concentration for the control and fat periods during Experiment III. Data for KB Data for RH 1 2 1 2 Diet Alb TP Alb TP Contr013 4.0 6.4 4.2 6.9 FAT4 4.0 6.5 4.3 7.1 Serum albumin (g/dl). Total proteins in serum (g/dl). Average for the two CHO periods. L‘UDNI-I Average for the fat periods. 200 Sodium, potassium, and chloride concentration (Table 25) were determined in the blood the last day of each dietary period, and there was a tendency for these values to be higher on the saturated fat meat diet; but in view of the few observations, this is probably not important. The hematocrit (HTC) and hemoglobin data for Experiment III are given in Table 36. No consistent variation between the different periods was seen. Costill et al. (1974a) have shown that the mean corpus— cular volume (MCV) decreases because of dehydration (increased osmolality of the blood), and that this shrinkage is accurately described by the mean corpuscular hemoglobin concentration (MCHC). Consequently, dehydra— tion as measured by changes in the HTC will be erroneous unless correc— tions are made for changes in the MCV. In the present study, no system— atic changes were observed in either of these variables, as seen from Table 26 (page 153). The mean values for the fat periods and for the control periods are given in Table 36; and as seen, there is no differ— ence, indicating again, that no dehydration was detected during the high fat period of Experiment III. TABLE 36. Mean hematocrit data during Experiment III. Data for KB Data for RH Diet HTC1 MCV1 MCHCl HTCl MCVl MCHCl Control 41 85 29 43 86 34 FAT 41 84 29 43 86 35 1For explanations of abbreviations, see page 139. 201 From changes in plasma proteins, it seems that a dehydration affecting the EPV occurred during Experiment II, whereas none was observed during Experiment III. Unfortunately, no hematocrit data were available for the former period. Gamble et al. (1923) showed that water is lost from the body during fasting, and concluded that it is due to destruction of protoplasm and reduction of tissue glycogen. Bloom and Azar (1963) observed a prompt weight loss associated with a negative sodium balance during starvation and when carbohydrate free diets were fed, regardless of the amount of salt consumed by the subjects. In the present study, it is possible that the sodium intake was limited during Experiment II. These diets were semi—synthetic mixtures to which no sodium was added, nor generally used by the subjects. This is in contrast to the more natural diets used in the Experiment III, where the subjects made liberal use of salt to improve the taste. This prob— ably explains the high sodium level seen in the blood during Experiment III (Table 25, page 151). However, if we accept the data of Bloom and Azar (1963), we can exclude differences in sodium intakes as a cause for the possible dehydration during the fat periods of Experiment II. This is in contrast to the results of Maagée (1968); and Veverbrants and Arky (1969). Bloom et al. (1966) found that carbohydrate deficiency during total fasting caused a decrease in extracellular fluid volume (including blood volume), as determined from serum protein and hematocrit. In contrast to the results of Bloom and Azar (1963), the weight loss in the present study appeared to depend on the protein intake; but the 202 diets employed by Bloom and Azar were carbohydrate free, which could make a difference. Katz et al. (1968) found an increased salt excretion during starvation, which was not abolished by giving supplementary salt comparable to that contained in a normal diet. When the salt intake was kept constant, either carbohydrate or protein plus fat prevented the salt loss; however, carbohydrate feeding also abolished the ketosis, whereas protein feeding did not. These results support our finding that the amount of proteins in the high fat diets determined whether or not they caused dehydration; and that dehydration is not caused by ketosis. One possible explanation for these results is that carbohydrate deprivation causes depletion of the glycogen stores in the kidney, so that the supply of glucose is insufficient to operate the sodium- potassium pump at a sufficient rate to prevent sodium loss. Hoffman (1973) has shown that there is a special compartment of ATP that is preferentially used by one component of the ATP'ase. In the red blood cell, this ATP is derived from anaerobic glycolysis (Parker and Hoffman, 1967). The saline diuresis appears to be related to the decrease in the insulin/glucagon ratio (DeFronzo et al., 1975; and Spark et al., 1975). It may be that the low insulin/glucagon ratio prevents sufficient glu- cose from entering the renal cells, and this may limit Na reabsorption. As carbohydrate deprivation proceeds, kidney gluconeogenesis may supply glucose internally, and this gluconeogenesis may be accelerated by a high protein intake. A similar explanation has been suggested by Wright et al. (1963). 203 Cizek et a1. (1977), from experiments with rabbits, suggest that the sodium loss is due to decreased bicarbonate in the blood; but that is difficult to understand because the mechanism commonly used to explain urine acidification is through exchange of sodium with hydrogen ions. These hydrogen ions are then buffered by bicarbonate ions or HP04—-, with the effect of excreting H - and reabsorbing HCO3 (as 2P04 C02). If these buffers, in very severe acidosis, become depleted, ammonia is used as a buffer (Gottschalk and Lassiter, 1974). Cizek et al. (1977) studied the effect of water deprivation on the sodium loss dur- ing starvation and found it to be independent of water intake; but if rabbits function in a similar manner, an increased urine volume would be expected, regardless of water intake (endogenous water). There is no reason to expect that water deprivation would abolish the loss of sodium. No sign of dehydration was found (by the methods employed) during Experiment III. Weight loss of the subjects is adequately explained by catabolism of glycogen and lean tissue mass. This was shown by Garrow (1974), who recalculated the results of many studies in the literature in view of the present knowledge about body glycogen stores and their associated water. He concluded that the weight loss during weight reduc- tion could be adequately explained on the basis of glycogen depletion lean tissue, and fat tissue catabolism. During Experiment II, there was some evidence of a decrease in the extracellular body fluid (about 10%). Assuming 15 kg of water, this could mean a weight loss of 1.5 kg for KB. Looking at the weight losses in Table 33 (page 182), it would mean that in order to explain the total weight loss (AWl), one need only assume 204 that about one-half of the glycogen stores were depleted, which probably could be expected during the five days of carbohydrate starvation. IX. SUMMARY In conclusion, it may be said that there was some evidence of dehydration during the low protein, high fat periods (Experiment II), but not during Experiment III, at least not of a magnitude that would have any influence on physical performance. The dehydration in Experi- ment II was presumably related to sodium excretion caused by carbohy- drate deprivation; the exact mechanism of this is unknown but it seems to depend upon the protein intake. With adequate protein intake, there was no evidence of dehydration, and the weight losses observed are adequately explained by glycogen and lean tissue catabolism. The ketosis seen during the high fat periods was not severe, was independent of the nature of fat and protein, and only slightly dependent upon the amount of protein eaten. The level of ketosis was comparable to that observed in most studies of this nature. The serum cholesterol level during the low protein study (Experiment II) did not appear to be affected by the nature of the fat ingested, but was found to be corre— lated to the amount of hydroxydmethylglutaric acid excreted in the urine. SUGGESTIONS FOR FURTHER RESEARCH It is clear from the literature that starvation causes a water loss not easily accounted for. As has been confirmed by the present studies, a somewhat similar loss was observed in subjects consuming a low carbohydrate diet. The negative water-balance is caused by carbo- hydrate deprivation, but the exact mechanism whereby this occurs is not apparent. The present studies indicate that dehydration is related to the level of protein intake (in addition to the carbohydrate content of the diet), since dehydration was observed only when the intake of both carbohydrate and protein was very low. Many explanations for the body weight loss seen during a low carbohydrate regimen are given. Some of the more plausible are: Catabolism of glycogen and lean body tissue, dehydration, ketosis, and sodium loss. Because it is impossible to introduce a carbohydrate free dietary regimen without causing a decrease in glycogen stores and lean tissue mass, it is difficult to determine the factor(s) that is (are) responsible for the weight loss. It appears that dehydration can be caused by either salt restric- tion (starvation), or a diet low in carbohydrate and protein. To separate these two possible factors, the following experiment is proposed: Subjects should be fed a normal control diet with a 205 206 sufficiently high but pre-determined salt intake. When their metabolism has stabilized, a high fat dietary regimen should be introduced. Both carbohydrate and protein intake must be kept low (5% of calories or less), with the same amount of salt as during the control period. The salt balance should be determined, and changes in the various water compartments monitored with tracer dilution techniques. When the study period is completed, it should be repeated without any (or very low) salt intakes. Similar experiments can be performed with higher levels of carbo- hydrate and protein (with and without salt restriction), raising first one and then the other to 15 or 20% of calories. With this experimen- tal design, it should be possible to distinguish the effect of a salt restricted diet from that of a diet low in carbohydrate and protein. The effect of glycogen and lean tissue catabolism on the body weight loss can be studied in the following way: Body weight and nitro- gen balance are determined for subjects fed a normal control diet. At the end of this period, muscle- and liver-biopsies can be taken to determine the glycogen content, and the total muscle mass estimated by total body potassium determinations. Total body water can be determined with tritium labelled water, and the plasma volume with 1131 marked albumin. The glycogen stores should then be depleted (as far as possible) by having the subjects exercise to exhaustion during an endurance type performance, such as running on a treadmill at about 75% of the maximal oxygen consumption. Determination of the water spaces, muscle and liver 207 glycogen, and body weight measurements should be recalculated. After the subjects have been allowed to stabilize for about three days while consuming a low carbohydrate, high protein diet, a nitrogen balance should be carried out and muscle- and liver-glycogen, body weight measure- ment and body water determinations made as before. Finally, the subjects are fed a high carbohydrate diet for four days and all measurements repeated. The protein intake should be kept constant and relatively high throughout all the dietary periods, and nitrogen balance and body weight measurements carried out daily. From this experiment (with about ten, physically uniform sub— jects), the contribution of glycogen stores, lean tissue, and water to the weight loss observed during a high fat, low carbohydrate dietary regimen, could be determined. If the biopsy technique is not feasible in man, a suitable animal model might be selected. The relationship between substrate (fat or carbohydrate) and the muscular work efficiency can best be studied in an experiment with short work periods (lo-20 minutes) and low work intensities(about 30% of the maximal oxygen consumption). The present experiments have shown that under such conditions, the substrate utilization is easily influ- enced by the dietary regimen; and the glycogen stores are not depleted. It therefore appears possible to separate the effect of using fat as substrate for muscle energy, from other effects of feeding a high fat diet, such as ketosis, dehydration, and glycogen depletion. This type of experiment should be done in the post—absorptive state to eliminate any possible effects of the subjects being in the absorptive state. 208 It is preferable to have subjects of uniform age, weight, and particu- larly, physical condition. During Experiment II (low protein), a high content of polyun- saturated fats in the diet appeared to increase the serum cholesterol level. Since this increase was not seen during Experiment III (high protein), it could not have been caused by the high fat intake, but the increased muscle catabolism during the high fat diet could possibly be a factor. However, a fairly good negative correlation (r = -0.79) was found with the urinary excretion of hydroxydmethylglutaric acid. This possibility needs further study. BIBLIOGRAPHY BIBLIOGRAPHY Addis, T., E. Barrett, L. J. P00, and D. W. Yuen. 1947. The relation between the serum urea concentration and the protein consumption of normal individuals. J. Clin. Invest. 26:869-874. Ahlborg, B., J. BergstrSm, J. Brohult, L.-G. Ekelund, and E. Hultman. 1967a. Human muscle glycogen content and capacity for prolonged exercise after different diets. F6rsvarsmedicin. 3:85-100. Ahlborg, B., J. Bergstram, L.-G. Ekelund, and E. Hultman. 1967b. Muscle glycogen and muscle electrolytes during prolonged physical exercise. Acta Physiol. Scand. 70:129-142. Ahlborg, B., P. Felig, L. Hagenfeldt, R. Hendler, and J. Wahren. 1974. Substrate turnover during prolonged exercise in man. Splanchnic and leg metabolism of glucose, free fatty acids and amino acids. J. Clin. Invest. 53:1080-1090. Ahrens, E. H. Jr., J. Hirsch, A. Oetta, J. W. Farquahar, and Y. Stein. 1961. Carbohydrate-induced and fat—induced lipemia. Trans. Ass. Am. Physns. 74:134-146. Anand, C. R. and H. M. Linkswiler. 1974. Effect of protein intake on calcium balance of young men given 500 mg calcium daily. J. Nutr. 104:695-700. Anderson, J. T., F. Grande, and A. Keys. 1971. Effect on man's serum lipids of two proteins with different amino acid composition. Am. J. Clin. Nutr. 24:524-530. Andres, R., G. Cader, and K. Zierler. 1956. The quantitatively minor role of carbohydrate in oxidative metabolism by skeletal muscle in intact man in the basal state. Measurements of oxygen and glucose uptake and carbon dioxide and lactate production in the forearm. J. Clin. Invest. 35:671-682. Andrews, S. S., S. A. Lopez, and W. G. Blackard. 1975. Effect of lipids on glucagon secretion in man. Metabolism. 24:35-44. Antonis, A. and I. Bersohn. 1961. The influence of diet on serum-tri- glycerides in South Africa white and Bantu prisoners. Lancet. 1:3_9o 209 210 Aoki, T. T., W. A. Muller, M. F. Brennan, and G. F. Cahill Jr. 1973. Blood cell and plasma amino acid levels across forearm muscle during a protein meal. Diabetes. 22:768-775. Aoki, T. T., C. J. Toews, A. A. Rossini, N. B. Ruderman, and G. F. Cahill Jr. 1975. Glucogenic substrate levels in fasting man. Advances in Enzyme Regulation. 13:329-336. Armstrong, M. D. and U. Stave. 1973. A study of plasma free amino acid levels. I. Study of factors affecting validity of amino acid analyses. Metabolism. 22:549-560. Askew, E. W., G. L. Dohm, and R. L. Huston. 1975. Fatty acid and ketone body metabolism in the rat: Response to diet and exer— cise. J. Nutr. 105:1422—1432. Rstrand, P. and K. Rodahl. 1970. Textbook of WOrk Physiology. McGraw—Hill Book Company, New York. p. 87. astrand, P. and B. Saltin. 1964. Plasma and red cell volume after prolonged severe exercise. J. Appl. Physiol.. 19:829-832. Astrand, P., I. Hallback, R. Hedman, and B. Saltin. 1963. Blood lac- tates after prolonged severe exercise. J. Appl. Physiol. 18:619-622. Balasse, E. O. and M. A. Neef. 1975. Inhibition of ketogenesis by ketone bodies in fasting humans. Metabolism. 24:999-1007. Balasse, E. O. and H. A. Oooms. 1968. Changes in the concentration of glucose, free fatty acids, insulin and ketone bodies in the blood during sodium B-hydroxybutyrate infusions in man. Diabetologia. 4:133—135. Balasse, E. O. and H. A. Oooms. 1973. Role of plasma free fatty acids in the control in insulin secretion in man. Diabetologia. 9:145—151. Balasse, E. 0., E. Couturier, and J. Franckson. 1967. Influence of sodium B-hydroxybutyrate on glucose and free fatty acid metab- olism in normal dogs. Diabetologia. 3:488—493. Becker, D. J., B. L. Pimstone, S. Kronheim, and E. Weinkove. 1975. Effect of alanine insusions on growth hormone, insulin, and glu- cose in protein-calorie malnutrition. Metabolism. 24:953—958. Beg, Z. H. and P. J. Lipien. 1972. Ig_vitro and ig_vivo inhibition of hepatic cholesterol synthesis by 3-hydroxy-3Hmethyglutaric acid. 211 Bell, J. D., S. Margen, and D. H. Calloway. 1969. Ketosis, weight loss, uric acid, and nitrogen balance in obese women fed single nutrients at low caloric levels. 'Metabolism. 18:193-208. Benoit, F. L., R. L. Martin, and R. H. Watten. 1965. Changes in body composition during weight reduction in obesity. Ann. Int. Med. 63:604-612. Bergstram, J. and E. Hultman. 1966. Muscle glycogen synthesis after exercise: An enhancing factor localized to the muscle cell in man. Nature (Lond.). 210:309-310. Bergstrfim, J. and E. Hultman. 1967. Synthesis of muscle glycogen in man after glucose and fructose infusion. Acta Medica Scand. 182:93-107. Bergster, J. and E. Hultman. 1972. Nutrition for maximal sports per- formance. JAMA. 221:999-1006. Bergster, J., V. Beroniade, E. Hultman, and A. E. Roch—Norlund. 1967a. Relation between glycogen and electrolyte metabolism in human muscle. In: F. Kruck, ed., Symposium Uber Transport und Function Intra-cellularer Elektrolyte. Urban und Schwarzenberg, Mfichen. pp. 108- Bergstrfim, J., G. Guarnieri, and E. Hultman. 1971a. Carbohydrate metabolism and electrolyte changes in human muscle tissue during heavy work. J. Appl. Physiol. 30:122-125. Bergster, J., G. Guarnieri, and E. Hultman. 1973. Changes in muscle water and electrolytes during exercise. In: J. Keul, ed., Limiting Factors of Physical Performance. Georgn Thiene, Stuttgart. pp. 173—178. Bergster, J., R. C. Harris, E. Hultman, and L.-O NordesjB. 1971b. Energy rich phosphagens in dynamic and static work. Adv. Exp. Med. Biol. 11:341-355. Bergstrfim, J., L. Hermansen, E. Hultman, and B. Saltin. 1967b. Diet, muscle glycogen and physical performance. Acta Physiol. Scand. 71:140-150. Bergstram, J., E. Hultman, and A. E. Roch-Norlund. 1972. Muscle glyco- gen synthetase in normal subjects: Basal values, effect of gly- cogen depletion by exercise and of a carbohydrate-rich diet following exercise. Scand. J. Clin. Lab. Invest. 29:231-236. Bergstram, J., E. Hultman, and B. Saltin. 1973. Muscle glycogen con- sumption during cross-country skiing (the Vasa ski race). Int. Z. angew. Physiol. 31:71-75. 212 Betheil, J. J., M. Feigelson, and P. Feigelson. 1965. The differential effects of glucocorticoid on tissue and plasma amino acid levels. Biochim. Biophys. Acta. 104:92-97. Bieberdorf, F. A., S. S. Chernick, and R. O. Scow. 1970. Effect of insulin and acute diabetes on plasma FFA and ketone bodies in the fasting rat. J. Clin. Invest. 49:1685-1693. Bjarntorp, P. and T. Schersten. 1967. Effect of B-hydroxybutyrate on lipid mobilization. Am. J. Physiol. 212:683—687. Blackburn, G. L., J. P. Flatt, G. H. A. Clowes Jr., T. F. O'Donnell, and T. E. Hensle. 1973. Protein sparing therapy during periods of starvation with sepsis or trauma. Ann. Surg. 177:588-594. Blaxter, K. L. 1971. Methods of measuring the energy metabolism of animals and interpretation of results obtained. Federation Proc. 30:1436—1443. Bloom, W. L. 1962. Inhibition of salt excretion by carbohydrate. Arch. Intern. Med. 109:26—32. Bloom, W. L. 1967. Carbohydrates and Water balance. Amer. J. Clin. Nutr. 20:157-162. Bloom, W. L. and G. J. Azar. 1963. Similarities of carbohydrate deficiency and fasting. Arch. Intern. Med. 112:333-343. Bloom, W. L. and W. Mitchell. 1960. Salt excretion of fasting patients. Arch. Intern. Med. 106:321-326. Bloom, W. L., G. J. Azar, and E. E. Smith Jr. 1966. Changes in heart size and plasma volume during fasting. ‘Metabolism. 15:409-413. Bolourchi, S., J. S. Feurig, and O. Michelsen. 1968. Wheat flour, blood urea concentrations, and urea metabolism in adult human subjects. Am. J. Clin. Nutr. 21:836—843. Bonsnes, R. W. and E. S. Dana. 1946. On the increased uric acid clear? ance following intravenous infusion of hypertonic glucose solu« tions. J. Clin. Invest. 25:386-388. Bosco, J. S., J. E. Greenleaf, E. M. Bernauer, and D. H. Card. 1974. Effects of acute dehydration and starvation on muscular strength and endurance. Acta Physiol. Pol. 25:410—421. Bozovic, L., J. Castenfors, and M. Piscator. 1967. Effect of prolonged, heavy exercise on urinary protein excretion and plasma renin activity. Acta Physiol. Scand. 70:143—146. 213 Buckler, J. M. H. 1972. Exercise as a screening test for growth hor- mone release. Acta Endo. 69:219—229. Buse, M. G. and S. S. Reid. 1975. Leucine. A possible regulator of protein turnover in muscle. J. Clin. Invest. 56:1250-1261. Buse, M. G., J. F. Biggers, C. Crier, and J. F. Buse. 1973. The effect of epinephrin, glucagon, and the nutritional state on the oxida- tion of granched chain amino acids and pyruvate by isolated hearts and diaphragms of the rat. J. Biol. Chem, 248:697-706. Cahill, G. F. Jr. 1964. Some observations on hypoglycemia in man. Advan. Enz. Reg. 2:137-148. Cahill, G. F. Jr. 1976. Starvation in man. Clinics in Endocrinology and Metabolism. 5:397-415. Cahill, G. F. Jr. and 0. E. Owen. 1970. The role of the kidney in the regulation of protein metabolism. In: H. N. Munro and J. B. Allison, eds., Mammalian Protein Metabolism, Vol. IV. Academic press, New York. pp. 559-581. Cahill, G. F. Jr., T. T. Aoki, M. F. Brennan, and W. A. Mfiller. 1972a: Insulin and muscle amino acid balance. Proc. Nutr. Soc., 31:233-238. Cahill, G. F. Jr., T. T. Aoki, and E. B. Marliss. 1972b. Insulin and muscle protein. In: D. F. Steiner and N. Freinkel, eds., Handbook of Physiology, Endocrine Pancreas. Am. Physiol. Soc., Washington DC. pp. 563-578. Cahill, G. F. Jr., M. G. Herrera, A. P. Morgan, J. S. Soeldner, J. Steinke, P. L. Levy, G. A. Reichard Jr., and D. M. Kipnis. 1966. Hormone-fuel interrelationships during fasting. J. Clin. Invest. 45:1751-1769. Campbell, J. and C. H. Best. 1956. Symposium on Diabetes: Physiologic aspects of ketosis. Metabolism. 5:95-113. Carlson, L. A. and B. Pernow. 1961. Studies on the peripheral circula— tion and metabolism in man. I. Oxygen utilization and lactate- pyruvate formation in the legs at rest and during exercise in healthy subjects. Acta Physiol. Scand. 52:328-342. Carlson, L. A., L. Kaijser, and B. W. Lassers. 1970. Myocardial metab— olism of plasma triglycerides in man. J. molec. cell. Cardiol. 1:467-475. 214 Carter, N. W. 1972. Intracellular pH. Kidney International. 1:341-346. Carter, W. J., K. M. Shakir, S. Hodges, F. H. Faas, and J. O. Wynn. 1975. Effect of thyroid hormone on metabolic adaptation to fasting. Metabolism. 24:1177-1183. Castenfors, J. 1967. Renal clearances and urinary sodium and potassium excretion during supine exercise in normal subjects. Acta Physiol. Scand. 70:207-214. Cerny, F. 1973. Protein metabolism during two hour ergometer exer- cise. In: H. Howals and J. R. Poortmans, eds., Metabolic Adaptation to Prolonged Physical Exercise. Birkhauser Verlag, Basel. pp. 232—237. Chiasson, J. L., J. Cook, J. E. Liljenquist, and W. W. Lacy. 1974. Glucagon stimulation of gluconeogenesis from alanine in the intact dog. Am. J. Physiol. 227:19—23. Christensen, E. H. and O. Hansen. 1939. Arbeitsfahigkeit und Ernahrung. Scand Arch. Physiol. 81:160-171. Christofori, F. C. and G. G. Duncan. 1964. Uric acid excretion in obese subjects during periods of total fasting. Metabolism. 13:303-311. Cizek, L. J., S. Simchon, and M. R. Nocenti. 1977. Effects of fasting on plasma volume and fluid and sodium exchanges in male rabbits (39658). Proc. Soc. Exp. Biol. Med. 154:299-303. Consolazio, C. F., H. L. Johnson, R. A. Nelson, J. G. Dramise, and J. H. Skala. 1975. Protein metabolism during intensive physical training in the young adult. Am. J. Clin. Nutr. 28:29—35. Consolazio, C. F., R. E. Johnson, and L. J. Pecora. 1963. Physiological Measurements of Metabolic Functions in Man. 'McGraw—Hill Book Company, New York. pp. 5-12. Consolazio, C. F., L. O. Matoush, H. L. Johnson, H. J. Krywicki, G. J. Isaac, and N. F. Witt. 1968. Metabolic aspects of calorie re— striction: Hypohydration effects on body weight and blood para- meters. Amer. J. Clin. Nutr. 21:793-802. Cori, C. F. and G. T. Cori. 1928, The mechanism of epinephrine action. II. The influence of epinephrine and insulin on the carbohydrate metabolism of rats in the postabsorptive state. J. Biol. Chem. 79:321-341. Cornil, A. (De Coster). 1965. Effect of muscular exercise on the plasma level of cortisol in man. Acta End. 48:163-168. 215 Costill, D. L. and W. J. Fink. 1974. Plasma volume changes following exercise and thermal dehydration. J. Appl. Physiol. 37:521-525. Costill, D. L. and B. Saltin. 1973. Muscle glycogen and electrolytes following exercise and thermal dehydration. In: H. Howald and J. R. Poortmans, eds., Metabolic Adaptation to Prolonged Physi- cal Exercise. Birkhauser Verlag, Basel. pp. 352-360. Costill, D. L. and B. Saltin. 1974. Changes in the ratio of venous to body hematocrit following dehydration. J. Appl. Physiol. 36:608—610. Costill, D. L. and K. E. Sparks. 1973. Rapid fluid replacement follow- ing thermal dehydration. J. Appl. Physiol. 34:299-303. Costill, D. L., L. Branam, D. Eddy, and W. Fink. 1974a. Alterations in red cell volume following exercise and dehydration. J. Appl. Physiol. 37:912-916. Costill, D. L., P. D. Gollnick, E. D. Jansson, B. Saltin, and E. M. Stein. 1973. Glycogen depletion pattern in human muscle fibers during distance running. Acta Physiol. Scand. 89:374-383. Costill, D. L., E. Jansson, P. D. Gollnick, and B. Saltin. 1974b. Glycogen utilization in leg muscles of men during level and up- hill running. Acta Physiol. Scand. 91:475-481. Cremer, J. 1971. Incorporation of label from D- -hydroxy[14C ]butyrate and [3-140 ]acetoacetate into amino acids in rat brain i§_vivo. Biochem. J. 122:135-138. DeFronzo, R. A., C. A. Cooke, R. Andres, G. R. Faloona, and P. J. Davis. 1975. The effect of insulin on the renal handling of sodium, potassium, calcium and phosphate in man. J. Clin. Invest. 55:845—855. Dennig, H., J. H. Talbott, H. T. Edwards, and D. B. Dill. 1931. Effect of acidosis and alkalosis upon capacity for work. J. Clin. Invest. 9:601—613. Deuel, H. J. 1957. The lipids: Their chemistry and biochemistry. In: Biochemistry, Vol. III. Interscience Publishers, Inc., New York. pp. 120-209. Edwards, J. C. and K. W. Taylor. 1970. Fatty acids and the release of glucagon from isolated guinea-pig islets of Langerhans incubated ix_vitro. Biochem. Biophys. Acta. 215:310-315. 216 Edwards, R. H. T., L.-O NordesjB, D. Koh, R. C. Harris, and E. Hultman. 1971. Isometric exercise-—factors influencing endurance and fatigue. Advan. Exp. Med. Biol. 11:357-360. Elwyn, D. H. 1970. The role of the liver in regulation of amino acid and protein metabolism. In: H. N. Munro, ed., Mammalian Pro- tein Metabolism, Vol. IV. Academic Press, New York. pp. 523- 558. Ericksson, B., B. Persson, and J. Thorell. 1971. The effects of re- peated prolonged exercise on plasma growth hormone, insulin, glucose, free fatty acids, glycerol, lactate and hydroxybutyric acid in 13 year old boys and in adults. Acta Paediatr. Scand. (Suppl.) 217:142-146. Essen, B., B. Pernow, P. D. Gollnick, and B. Saltin. 1973. Muscle glycogen content and lactate uptake in exercising muscles. In: H. Howald and J. R. Poortmans, ed., Metabolic Adaptation to Prolonged Physical Exercise. Birkhauser Verlag, Basel. pp. 130-134. Fajans, S. S. and J. C. Floyd Jr. 1972. Stimulation of islet cell secretion by nutrients and by gastrointestinal hormones released during ingestion. In: R. O. Greep and E. B. Astwood, eds., Handbook of Physiology, Endocrinology, Sec. 7, Vol. I. Am. Physiol. Soc., Washington DC. pp. 473-494. Farquharson, R. F., W. T. Salter, D. M. Tibbetts, and J. C. Aub. 1931. Studies of calcium and phosphorus metabolism. XII. The effect of the ingestion of acid-producing substances. J. Clin. Invest. 10:221-249. Felig, P. 1972. Interaction of insulin and amino acid metabolism in the regulation of gluconeogenesis. Isr. J. Med. Sci. 8:262-270. Felig, P. 1973. The glucose-alanine cycle. Metabolism. 22:179-207. Felig, P. 1975. Amino acid metabolism in man. Annual Review of Biochem. 44:933-955. Felig, P. and J. Wahren. 1970. Evidence for a glucose—alanine cycle: Amino acid metabolism during muscular exercise. J. Clin. Invest. 49:28a. Felig, P. and J. Wahren. 1971a. Amino acid metabolism in exercising man. J. Clin. Invest. 50:2703—2714. Felig. P. and J. Wahren. 1971b. Influence of endogenous insulin secre- tion on splanchnic glucose and amino acid metabolism in man. J. Clin. Invest. 50:1702-1711. Felig, Felig, Felig, Felig, Felig, Felig, Felig, Felig, Felig, Felig, Felig, Felig, Felig, 217 P. and J. Wahren. 1974. Protein turnover and amino acid metab— olism in the regulation of gluconeogenesis. Federation Proc. 33:1092-1097. P. and J. Wahren. 1975a. Muscle nitrogen repletion after pro- tein ingestion: Key role of branched chain amino acids and evidence of a new defect in diabetes. Clin. Res. 23:420a. P. and J. wahren. 1975b. The liver as site of insulin and g1u- cagon action in normal, diabetic and obese humans. Isr. J. Med. Sci. 11:528-539. P., E. Marliss, and G. F. Cahill Jr. 1969a. Plasma amino acid levels and insulin secretion in obesity. N. Engl. J. Med. 281:811-816. P., E. Marliss, and G. F. Cahill Jr. 1971. Metabolic response to human growth hormone during prolonged starvation. J. Clin. Invest. 50:411-421. - P., E. Marliss, O. E. Owen, and G. F. Cahill Jr. 1969b. Blood glucose and gluconeogenesis in man. Arch. Intern. Med. 123: 293-298. P., E. Marliss, O. E. Owen, and G. F. Cahill Jr. 1969c. Role of substrate in the regulation of hepatic gluconeogenesis in fasting man. Advanc. Enzyme Reg. 7:41-46.' P., E. Marliss, T. Pozefsky, and G. F. Cahill Jr. 1970a. Amino acid metabolism in the regulation of gluconeogenesis in man. Am. J. Clin. Nutr. 23:986-992. P., O. E. Owen, J. Wahren, and G. F. Cahill Jr. 1969d. Amino acid metabolism during prolonged starvation. J. Clin. Invest. 48:584-593. P., T. Pozefsky, E. Marliss, and G. F. Cahill Jr. 1970b. Alanine: Key role in gluconeogenesis. Science. 167:1003-1004. P., J. Wahren, and G. Ahlborg. 1973a. Uptake of individual amino acids by the himan brain. Proc. Soc. Exp. Bio. Med. 142:230-231. P., J. Wahren, and R. Hendler. 1974a. Sensitivity of hepatic gluco-regulatory mechanisms in exercising man. In: Proc. 56 Ann. Meet. Endocrine Soc. Atlanta. J. B. Lippincott Company, Philadelphia. pp. P., J. Wahren, and R. Hendler. 1975. Influence of oral glucose ingestion on splanchnic glucose and gluconeogenic substrate metabolism in man. Diabetes. 24:468-475. 218 Felig, P., J. Wahren, R. Hendler, and G. Ahlborg. 1972. Plasma gluca- gon levels in exercising man. N. Engl. J. Med. 287:184-185. Felig, P., J. Wahren, R. Hendler, and T. Brunoin. 1974b. Splanchnic glucose and amino acid metabolism in obesity. J. Clin. Invest. 53:582-590. Felig, P., J. Wahren, and L. Raf. 1973b. Evidence of inter—organ amino-acid transport by blood cells in humans. Proc. Nat. Acad. Sci. USA. 70:1775-1779. Fineberg, S. E. and T. J. Merimee. 1974. Acute metabolic effects of human growth hormone. Diabetes. 23:409-504. Flatt, J. P. 1972. On the maximal possible rate of ketogenesis. Diabetes. 21:50-53. Flatt, J. and G. L. Blackburn. 1974. The metabolic fuel regulatory system: Implications for protein—sparing therapies during caloric deprivation and disease. Amer. J. Clin. Nutr. 27:175-187. Fletcher, W. M. and F. G. Hopkins. 1917. Croonian Lecture: The re- spiratory process in muscle and the nature of muscular motion. Proc. Roy. Soc. Lond., Ser. B. 89:444—467. Florey, C. Du V., H. McDonald, W. E. Miall, and R. D. G. Milner. 1973. Serum lipids and their relation to blood glucose and cardio- vascular measurements in a rural population of Jamaican adults. Frame, E. G. 1958. The levels of individual free amino acids in the plasma of normal man at various intervals after a high—protein meal. J. Clin. Invest. 37:1710-1723. Friedman, M. and R. H. Rosenman. 1969. Association of specific overt behavior pattern with blood and cardiovascular findings. J. Am. Fraberg, S. O., L. A. Carlson, and L.-G. Ekelund. 1971. Local lipid Stores and exercise. Advan. Exp. Med. Biol. 11:307-313. Fuchs, F., Y. Reddy, and F. N. Briggs. 1970. The interaction of cations with the calcium-binding site of troponin. Biochim. Biophys. Acta, 221:407-409. Fulks, R. M., F. Li, and A. Goldberg. 1975. Effects of insulin, glu— cose, and amino acids on protein turnover in rate diaphragm. J. Biolog. Chem. 250:290-298. 219 Ffirst, P., G. Guarnieri, and E. Hultman. 1971. The effect of the admin- istration of L-tryptophan on synthesis of urea and gluconeo- genesis in man. Scand. J. Clin. Lab. Invest. 27:183-191. Gamble, J. L. 1947. Physiological information gained from studies on the life raft ration. Harvey Lectures. 42:247-273. Gamble, J. L., G. S. Ross, and F. F. Tisdall. 1923. The metabolism of fixed base during fasting. J. Biolog. Chem. 57:633-695. Garber, A. J., P. H. Menzel, G. Boden, and O. E. Owen. 1974. Hepatic ketogenesis and gluconeogenesis in humans. J. Clin. Invest. 54:981-989. Garrow, J. S. 1974. Energy Balance and Obesity in Man. North-Holland Publishing Company, Amsterdam. pp. 39-46 and 192-210. Gates, 8. C. 1977. Automated metabolic profiling of organic compounds in human urine by gas chromatographydmass spectrometry. Dissertation, Department of Biochemistry, Michigan State Univer- sity. Gerich, J. E., M. A. Charles, and G. M. Grodsky. 1974a. Characteriza- tion of the effects of arginine and glucose on glucagon and insulin release from the ig_vitro perfused rat pancreas. J. Clin. Invest. 54:833-841. Gerich, J. E., M. A. Charles, and G. M. Grodsky. 1976a. Regulation of pancreatic insulin and glucagon secretion. Annual Review of Physiol. 38:353-388. Gerich, J. E., M. Langlois, V. Schneider, J. H. Karam, and C. Noacco. 1974b. Effects of alterations of plasma free fatty acid levels on pancreatic glucagon secretion in man. J. Clin. Invest. 53:1284-1289. Gerich, J. E., M. Lorenzi, D. M. Bier, E. Tsalikian, V. Schneider, J. H. Karam, and P. H. Forsham. 1976b. Effects of physiologic levels of glucagon and growth hormone on human carbohydrate and lipid metabolism. J. Clin. Invest. 57:875—884. Gilder, H., G. N. Cornell, W. R. Grafe, J. R. Macfarlane, J. W. Asaph, W. T. Stubenbord, G. M. Watkins, J. R. Rees, and B. Thorbjarnar- son. 1967. Components of weight loss in obese patients sub- jected to prolonged starvation. J. Appl. Physiol. 23:304-310. Glueck, C. J., R. I. Levy, and D. S. Fredrickson. 1969. Immunoreactive Insulin, Glucose Tolerance and Carbohydrate Inducibility in Types II, III, IV, and V Hyperlipoproteinemia. Diabetes 18: 739-747. RC660D52. 220 Goldsmith, R., G. S. Riddell, I. F. Russel, M. M. Sawyer, and R. B. Smith. 1971. The energy cost of a task after the ingestion of protein, fat or carbohydrate. Proc. Nutr. Soc. Lond. 30:4A-5A. Gollnick, P. D., R. B. Armstrong, W. L. Sembrowich, R. E. Shepherd, and B. Saltin. 1973. Glycogen depletion pattern in human skeletal muscle fibers after heavy exercise. J. Appl. Physiol. 34:615- 618. Gollnick, P. D., J. Karlsson, K. Piehl, and B. Saltin. 1974a. Selective glycogen depletion in skeletal muscle fibers of man following sustained contractions. J. Physiol. 241:59—67. Gollnick, P. D., K. Piehl, and B. Saltin. 1974b. Selective glycogen depletion pattern in human muscle fibers after exercise of vary- ing pedalling rates. J. Physiol. 241:45-57. Gollnick, P. D., K. Piehl, C. W. Saubert IV, R. B. Armstrong, and B. Saltin. 1972. Diet, exercise, and glycogen changes in human muscle fibers. J. Appl. Physiol. 33:421—425. Gottschalk, C. W. and W. E. Lassiter. 1974. Mechanisms of urine forma- tion. In: V. B. Mountcastle, ed., Medical Physiology, Vol. II, 13th Edition. C. V. Mosby Company, Saint Louis. pp. 1065-1105. Grey, N. J., I. Karl, and D. M. Kipnis. 1975. Physiologic mechanisms in the development of starvation ketosis in man. Diabetes. 24:10-16. Grimby, G. 1965. Renal clearnaces during prolonged supine exercise at different loads. J. Appl. Physiol. 20:1294—1298. Haag, B. L., M; M. Reidenberg, C. R. Shuman, and B. J. Channick. 1967. Aldosterone, l7-hydroxycorticosteroid, 17-ketosteroid, and fluid and electrolyte responses to starvation and selective refeeding. Amer. J. Med. Sci. 254:652-658. Hagenfeldt, L. 1975. Turnover of individual free fatty acids in man. Federation Proc. 34:2246-2249. Hagenfeldt, L. and J. Wahren. 1968a. Human forearm muscle metabolism during exercise. 11. Uptake, release and oxidation of indivi- dual FFA and glycerol. Scand. J. Clin. Lab. Invest. 21:263-276. Hagenfeldt, L. and J. Wahren. 1968b. Human forearm muscle metabolism during exercise. III. Uptake, release and oxidation of B—hydroxybutyrate/acetoacetate ratio. Scand. J. Clin. Lab. Invest. 21:314-320. 221 Hagenfeldt, L. and J. Wahren. 1972. Human forearm muscle metabolism during exercise. VII. FFA uptake and oxidation at different work intensities. Scand. J. Clin. Lab. Invest. 30:429-436. Hagenfeldt, L. and J. wahren. 1973. Effect of exercise on splanchnic exchange of free fatty acids. Metabolism. 22:815-820. Hagenfeldt, L. and J. Wahren. 1975a. Turnover of free fatty acids dur- ing recovery from exercise. J. Appl. Physiol. 39:247-250. Hagenfeldt, L. and J. Wahren. 1975b. Turnover of plasma-free arachi- donic oleic acids in resting and exercising human subjects. Metabolism. 24:799-806. Hagenfeldt, L., J. Wahren, B. Pernow, and L. Raf. 1972. Uptake of individual free fatty acids by skeletal muscle and liver in man. J. Clin. Invest. 51:2324-2330. Hansen, A. P. 1970. Abnormal serum growth hormone response to exercise in juvenile diabetics. J. Clin. Invest. 49:1467-1478. Hansen, A. P. 1971. Effect of intravenous glucose infusion on the exercise-induced serum growth hormone rise in normals and juvenile diabetics. Scand. J. Clin. Invest. 28:207-212. Haralambie, G. 19733. Changes in electrolytes and trace elements dur- ing long-lasting exercise. In: H. Howald and J. R. POortmans, eds., Metabolic Adaptation to Prolonged Physical Exercise. Birkhauser Verlag, Basel. pp. 340-351. Haralambie, G. 1973b. Importance of humoral changes to physical performance. In: J. Keul, ed., Limiting Factors of Physical Performance. Georgn Thiene, Stuttgart. pp. 189-200. Harding, V. J., K. D. Allin, and B. A. Eagles. 1927. Influence of fat and carbohydrate diets upon the level of blood uric acid. J. Biol. Chem. 74:631-643. Harlan, W. R., J. Laszlo, M. D. Bogdonoff, and E. H. Estes Jr. 1963. Alterations in free fatty acid metabolism in endocrine disorders. I. Effect of thyroid hormone. J. Clin. Endocrinol. Metab. 23:33-40. Harris, R. C., E. Hultman, and L.-O. Nordesj3. 1974. Glycogen, glyco- lytic intermediates and high-energy phosphates determined in biopsy samples of musculus quadriceps femoris of man at rest. Methods and variance of values. Scand. J. Clin. Lab. Invest. 33:109-120. 222 Hartog, M., R. J. Havel, G. Copinschi, J. M. Earll, and B. C. Richie. 1967. The relationship between changes in serum levels of growth hormone and mobilization of fat during exercise in man. Quart. J. Exp. Physiol. 52:86-96. Hartsook, E. W. and T. V. Hershberger. 1971. Interactions of major nutrients in whole-animal energy metabolism. Federation Proc. 30:1466-1473. Harvey, W. D., G. R. Faloona, and R. H. Unger. 1974. The effect of adrenergic blockade on exercise-induced hyperglucagonemia. Endocrinology. 94:1254-1258. Hatch, F. T., L. A. Liese, and F. E. Kendall. 1955. Effects of restriction of dietary fat and cholesterol upon serum lipids and lipoproteins in patients with hypertension. Amer. J. Med. 19: 48-600 Havel, R. J. 1971. Influence of intensity and duration of exercise on supply and use of fuels. Advanc. Exp. Med. Biol. 11:315-325. Havel, R. J., L. A. Carlson, L.-G Ekelund, and A. Holmgren. 1964. Turnover rate and oxidation of different fatty acids in man dur- ing exercise. J. Appl. Physiol. 19:613-618. Havel, R. J., A. Naimark, and C. F. Borchgrevink. 1963. Turnover rate and oxidation of free fatty acids of blood plasma in man duri 4 exercise: Studies during continuous infusion of palmitate-l-C . J. Clin. Invest. 42:1054-1063. Hawkins, R. A., K. G. M. M. Alberti, C. R. S. Houghton, D. H. Williamson, and H. A. Krebs. 1971. The effect of acetoacetate on plasma insulin concentration. Biochem. J. 125:541-544. Hawley, E. E., C. W. Johnson, and J. R. Murlin. 1933. The possibility of gluconeogenesis from fat. II. The effect of high fat diets on the respiratory metabolism and ketosis of man. J. Nutr. 6:523-557. Heeley, D. M., I. M. Sharman, and D. F. Cooper. 1975. Variations in the composition of blood and urine following the ingestion of a high-protein diet. Proc. Nutr. Soc. 34:69A. Henschel, A., H. L. Taylor, and A. Keys. 1954. Performance capacity in acute starvation with hard work. J. Appl. Physiol. 6:624— 633. Hermansen, L. 1971. Lactate production during exercise. Advanc. Exp. 223 Hermansen, L. and J. Osnes. 1972. Blood and muscle pH after maximal exercise in man. J. Appl. Physiol. 32:304-308. Hermansen, L. and B. Saltin. 1967. Blood lactate concentration during exercise at acute exposure to altitude. In: Exercise at Altitude. Excerpta Medica Foundation, Amsterdam. pp. 48- Hermansen, L., S. Maehlum, E. D. R. Pruett, O. Vagge, H. Waldmm, and T. Wessel-Aas. 1973. Lactate removal at rest and during exer- cise. In: H. Howald and J. R. Poortmans, eds., Metabolic Adaptation to Prolonged Physical Exercise. Birkhauser Verlag, Basel. pp. 101-105. Hermansen, L., E. D. R. Pruett, J. B. Osnes, and F. A. Giere. 1970. Blood glucose and plasma insulin in response to maximal exercise and glucose infusion. J. Appl. Physiol. 29:13-16. Herrera, M. G., D. Kamm, N. Ruderman, and G. F. Cahill Jr. 1966. Non-hormonal factors in the control of gluconeogenesis. Advan. Enzyme Reg. 4:225-235. Hill, A. V. 1955-56. The influence of the external medium on the internal pH of muscle. Proc. Roy. Soc. Lond., Ser. B. 144:1—22. Hirche, H., H. D. Langohr, and U. wacker. 1973. Lactic acid accumula- tion in working skeletal muscle. In: J. Keul, ed., Limiting Factors of Physical Performance. Georgn Thiene, Stuttgart. pp. Hodgkin, A. L. and P. Horowitz. 1959. Movements of Na and K in single muscle fibers. J. Physiol. (Lond.). 145:405-432. Hoffman, J. F. 1973. Molecular aspects of the Na+, K+ -Pump in red blood cells. In: M. Nakao and L. Packer, eds., Organization of Energy-transducing Membranes. University of Tokyo Press, Tokyo. pp. 9-21. Hood, C. E. A., J. M. Goodhart, R. F. Fletcher, J. Gloster, P. V. Bergrand, and A. C. Crooke. 1970. Observations on obese patients eating isocaloric reducing diets with varying propor- tions of carbohydrate. Br. J. Nutr. 24:39-44. Hoover, H. C. Jr., J. P. Grant, C. Gorschboth, and A. J. Ketcham. 1975. Nitrogen-sparing intravenous fluids in postoperative patients. N. Engl. J. Med. 293:172-175. Hubbard, J. L. 1973. The effect of exercise on lactate metabolism. J. Physiol., Lond. 231-1-18. 224 Hultman, E. 1967. Studies on muscle metabolism of glycogen and active phosphate in man with special reference to exercise and diet. Scand. J. Clin. Lab. Invest. 19:1-63. (Suppl. 94:7) Muscle glycogen synthesis in rela- Hultman, E. and J. Bergster. 1967. Acta Medica Scand. tion to diet studied in normal subjects. 182:109-117. Hultman, E. and J. Bergstram. 1973. Local energy-supplying substrates as limiting factors in different types of leg muscle work in J. Keul, ed., Limiting Factors of Physical normal man. In: 113-125. Performance. Georgn Thiene, Stuttgart. pp. 1971. Liver glycogen in man. Effect Hultman, E. and L. H. Nilsson. Advan. Exp. Med. Biol. of different diets and muscular exercise. 11:143-151. Liver glycogen as a glucose- In: J. Keul, ed., Limiting Georgn Thiene, Stuttgart. Hultman, E. and L. Nilsson. 1973. supplying source during exercise. Factors of Physical Performance. pp. 179-189. 1975. Factors influencing carbohydrate Hultman, E. and L. H. Nilsson. metabolism in man. Nutr. Metabol. 1974. Normal carbohydrate Hultman, E., J. BergstrSm, and L. H. Nilsson. Acta Anas- metabolism and carbohydrate metabolism in trauma. teciol Scand. 55:28-49. (Suppl.) Hultman, E., J. Bergstrom, and A. E. Roch-Norlund. 1971. Glycogen storage in human skeletal muscle. Advanc. Exp. Med. Biol. 11:273-288. Hunter, W. M. 1972. Secretion of human growth hormone. Proc. Nutr. 1968. Proceedings of Hunter, W. M., W. M. Rigal, and M. Y. Sukkar. Excerpta the international symposium, Milan, in Growth Hormone. Medica Foundation, Amsterdam. pp. 408- Johnson, Jr. E., R. Passmore, and F. Sargent, II. 1961. Multiple factors in experimental human ketosis. Archives of Internal. Med. 107:111-118. The effect of exercise upon Johnson, R. H. and J. L. Walton. 1972. Quart. acetoacetate metabolism in athletes and non-athletes. J. Exp. Physiol. 57:73-79. Johnson, R. H., J. L. Walton, H. A. Krebs, and D. H. Williamson. 1969. Post-exercise ketosis. Lancet. 2:1383-1385. 225 Jokoubek, B. and D. Svorad. 1959. Uber die Veranderungen in der Konzentration des Gehirnglykogens bei kBrperlicher Belastung. - Pffigers Arch. ges. Physiol. 268:444-448. Jones, N. L., D. G. Robertson, J. W. Kane, and R. A. Hart. 1972. Effect of hypoxia on free fatty acid metabolism during exercise. J. Appl. Physiol. 33:733-738. Jorfeldt, L. 1970. Metabolism of L(+)-lactate in human skeletal muscle during exercise. Acta Physiol. Scand. 78:1—67. (Suppl. 338) - Jorfeldt, L. 1971. Turnover of l4C-L(+)—-lactate in human skeletal muscle during exercise. Advanc. Exp. Med. Biol. 11:409-417. Jorfeldt, L. and J. Wahren. 1971. Leg blood flow during exercise in man. Clin. Sci. 41:459-473. Kachadorian, W. A. 1972. The effects of activity on renal function. In: Fitness and Exercise, Proc. CIC Symposium. The Athletic Institute, Chicago. pp. 97-116. Kamm, D. E. and G. F. Cahill Jr. 1969. Effect of acid-base status on renal and hepatic gluconeogenesis in diabetes and fasting. Kark, R. M., R. E. Johnson, and J. S. Lewis. 1945. Defects of pemmican as an emergency ration for infantry troops. war Medicine. 7:345-352. Karlsson, J. 19713. Lactate and phosphagen concentrations in working muscle of man. Acta Physiol. Scand. 81:1-72. (Suppl. 358) Karlsoon, J. 1971b. Muscle ATP, CP, and lactate in submaximal and 'maximal exercise. Advan. Exp. Med. Biol. 11:383-393. Karlsson, J., L.-O. NordesjB, and B. Saltin. 1974. Muscle glycogen utilization during exercise after physical training. Acta Physiol. Scand. 90:210-217. Kasper, H., H. Thiel, and M. Ehl. 1973. Response of body weight to a low carbohydrate, high fat diet in normal and obese subjects. Amer. J. Clin. Nutr. 26:197-204. Katz, A. I., D. R. Hollingworth, and F. H. Epstein. 1968. Influence of carbohydrate and protein on sodium excretion during fasting and refeeding. J. Lab. Clin. Med. 72:93-104. Katz, A. M. 1970. Contractile proteins of the heart. Physiol. Rev. 50:63-158. 226 Keith, H. M. 1963. Ketogenic diet in the treatment of epilepsy. In: Convulsive Disorders in Children. Little & Brown, Boston. Kekwick, A. and G. L. S. Pawan. 1956. 'Caloric intake in relation to body weight changes in the obese. Lancet. 2:155-161. Kekwick, A. and G. L. S. Pawan. 1957. Metabolic study in human obesity with isocaloric diets high in fat, protein or carbohydrate. Metabolism. 6:447-460. Kemp, R. G. 1969. Allosteric properties of muscle phosphofructokinase. II. Kinetics of native and thiol-modified enzyme. Biochem. 8:4490-4496. Kerr, S. E., C. W. Hampel, and M. Ghantus. 1937. The carbohydrate metabolism of brain. IV. Brain glycogen, free sugar, and lactic acid as affected by insulin in normal and adrenal- inactivated cats, and by epinephrine in normal rabbits. J. Biol. Chem. 119:405-421. Keul, J., E. Doll, and D. Keppler. 1967. The substrate supply of the human skeletal muscle at rest, during and after work. Experientia. 23:1-6. Keul, J., E. Doll, and D. Keppler. 1972. Energy Metabolism of Human Muscle. S. Karger, Basel. pp. 174-202. Keys, A., J. Brozek, A. Henschel, O. Mickelsen, and H. L. Taylor. 1950. The Biology of Human Starvation. University of Minnesota Press, Minneapolis. 1385 pp. Kibler, R. F., W. J. Taylor, and J. D. Myers. 1964. Effect of glucagon on net splanchnic balances of glucose, amino acid nitrogen, urea, ketones, and oxygen in man. J. Clin. Invest. 43:904-915. Klausen, K., K. Piehl, and B. Saltin. 1973. Muscle glycogen stores and capacity for anaerobic work. In: H. Howald and J. R. Poortmans, eds., Metabolic Adaptation to Prolonged Physical Exercise. Birkhauser Verlag, Basel. pp. 127-129. Klein, J. R. and N. S. Olsen. 1947. Effect of convulsive activity upon the concentration of brain glucose, glycogen, lactate, and phosphates. J. Biol. Chem. 167:747-756. Knuttgen, H. G. 1971. Lactate and oxygen debt: An introduction. Advanc. Exp. Med. Biol. 11:361-369. 227 Koizumi, J. 1974. Glycogen in the central nervous system. Prog. Histochem. Cytochem. 6:1—35. Kotchen, T. A., L. H. Hartley, T. W. Rice, E. H. Mougey, L. G. Jones, and J. W. Mason. 1971. Renin, norepinephrine, and peinephrine responses to graded exercise. J. Appl. Physiol. 31:178-184. Kozlowski, S. and B. Saltin. 1964. Effect of sweat loss on body fluids. J. Appl. Physiol. 19:1119-1124. Kraut, H. and G. Lehmann. 1948. Der Eiweissbedark des Schwerarbeiters. I. Physiologisches und funktionelles Eiwissminimum. Biochem. Krebs, H. A. 1964. The metabolic fate of amino acids. In: H. N. Munro and J. B. Allison, eds., Mammalian Protein Metabolism, Vol. 1. Academic Press, New York. pp. 125-176. Krebs, H. A. 1966. The regulation of the release of ketone bodies by the liver. Advan. Enzy. Reg. 4:339-353. Krogh, A., J. Lindhard, G. Liljestrand, and K. G. Andresen. 1920. The relative value of fat and carbohydrate as sources of muscular energy. Biochem. J. 14:290-363. Krzywicki, H. J., C. F. Cpnsolazio, H. L. Johnson, and N. F. Witt. 1972. Metabolic aspects of caloric restriction (420 kcal): Body composition changes. Amer. J. Clin. Nutr. 25:67-73. Kumamoto, T. 1953. Histochemical studies on the changes of brain glycogen caused by starvation. Osaka Daigaku Igaku Zasshi. Landau, R. L. and K. Lugibihl. 1969. Effect of glucagon on concentra- tion of several free amino acids in plasma. Metabolism. ' 18:264—276. Lefebvre, P. 1972. Glucagon and lipid metabolism. In: P. Lefebvre and R. H. Unger, eds., Glucagon: Molecular Physiology, Clinical and Therapeutic Implication. Pergamon Press, Elmsford, New York. pp. 109-121. Lehmann, J., J. R. Litzow, and E. J. Lennon. 1967. Studies of the mechanism by which chronic metabolic acidosis augments urinary calcium excretion in man. J. Clin. Invest. 46:1318-1328. Lennox, W. G. 1924. Increase of uric acid in the blood during pro- longed starvation. JAMA. 82:602-604. 228 Levin, S., J. Karam, S. Hane, G. Grodsky, and P. Forsham. 1971. Enhancement of arginine-induced insulin secretion in man by prior administration of glucose. Diabetes. 20:171-176. Lewis, H. B. and R. C. Corley. 1923. Studies in uric acid metabolism. III. The influence of fats and carbohydrates on the endogenous uric acid elimination. J. Biol. Chem. 55:373-384. Liljenquist, J. E., J. L. Chasson, R. E. Finger, and W. W. Lacy. 1974. Differential effect of insulin on glycogenolysis and gluconeo- genesis. (Abstract) Diabetes. 23:349. Livingston, S. J. 1972. Dietary treatment of epilepsy. In: Compre- hensive Management of Epilepsy. C. C. Thomas, Springfield, Illinois. pp. 378-405. Lowenstein, J. M. 1972. Ammonia production in muscle and other tissues: The purine nucleotide cycle. Physiol. Rev. 52:382-414. Lowenstein, J. M. and K. Tornheim. 1971. Ammonia production in muscle; the purine nucleotide cycle. Science. 171:397-400. Luyckx, A. S. and P. J. Lefebvre. 1972. Role of catecholamines in exercise-induced glucagon secretion in rats. Diabetes. 32:334. (Suppl. 1) Maagde, H. 1968. Changes in blood volume during absolute fasting with and without sodium chloride administration. Metabolism. 17: 133-138. Madison, L. L. 1969. Role of insulin in the hepatic handling of glucose. Arch. Int. Med. 123:284-292. Madison, L. L., D. Mebane, R. H. Unger, and A. Lochner. 1964. The hypoglycemic action of ketones. II. Evidence for a stimulatory feedback of ketones on the pancreatic beta cells. J. Clin. Invest. 43:408-415. Mallette, L. E., J. H. Exton, and C. R. Park. 1969a. Control of gluco- neogenesis from amino acids in the perfused rat liver. J. Biol. Chem. 244:5713-5723. Mallette, L. E., J. H. Exton, and C. R. Park. 1969b. Effects of gluca— gon on amino acid transport and utilization in the perfused rat Manchester, K. L. 1965. Oxidation of amino acids by isolated rat dia— phragm and the influence of insulin. Biochim. Biophys. Acta. 100:295-298. 229 Manchester, K. L. and F. G. Young. 1970. The control by insulin of amino acid accumulation in muscle. Biochem. J. 117:457-465. Marco, J., C. Calle, D. Roman, M. Diaz-Fierros, M. Villanueva, and I. Valverde. 1973. Hyperglucagonism induced by glucocorticoid treatment in man. N. Engl. J. Med. 228:128-131. Margaria, R., P. Aghemo, and G. Sassi. 1971. Effect of alkalosis on performance and lactate formation in supramaximal exercise. Int. 2. Angew. Physiol. 29:217-223. Marliss, E. B., T. T. Aoki, R. H. Unger, J. S. Soeldner, and G. F. Cahill Jr. 1970. Glucagon levels and metabolic effects in fasting man. J. Clin. Invest. 49:2256-2270. Marsh, M. E. and J. R. Murlin. 1928. Muscular efficiency on high carbo- hydrate and high fat diets. J. Nutr. 1:105-137. Matsutaka, H., T. Aikawa, H. Yamamoto, and E. Ishikawa. 1973. Gluco- neogenesis and amino acid metabolism. III. Uptake of glutamine and output of alanine and ammonia by non-hepatic splanchnic organs of fasted rats and their metabolic significance. J. Biochem. Tokyo. 74:1019-1029. Mazess, R. B. and W. Mather. 1973. Bone mineral content of North Alaskan Estimos. Amer. J. Clin. Nutr. 27:916-925. McClellan, W. S. and E. F. DuBois. 1930. Clinical Calorimetry. XIV. Prolonged meat diets with a study of kidney function and ketosis. J. Biol. Chem. 87:651-668. McGarry, J. D. and D. W. Foster. 1971a. Regulation of ketogenesis from octanoic acid: Role of the tricarboxylic acid cycle and fatty acid synthesis. J. Biol. Chem. 216:1149—1159. McGarry, J. D. and D. W. Foster. 1971b. Regulation of ketogenesis from oleic acid and the influence of antiketogenic agents. J. Biol. Chen. 246:6247—6253. McGarry, J. D. and D. W. Foster. 1972. Regulation of ketogenesis and clinical aspects of the ketotic state. Metabolism. 21:471-489. McGarry, J. D., C. Robles-valdes, and D. W. Foster. 1975a. Role of carnitine in hepatic ketogenesis. Proc. Natl. Acad. Sci. US. 72:4385-4388. McGarry, J. D., P. Wright, and D. Foster. 1975b. Hormonal control of ketogenesis: Rapid activation of kepatic ketogenic capacity in fed rats by anti-insulin serum and glucagon. J. Clin. Invest. 55:1202—1209. 230 Merimee, T. J., P. Felig, E. Marliss, S. E. Fineberg, and G. F. Cahill Jr. 1971. Glucose and lipid homeostasis in the absence of human growth hormone. J. Clin. Invest. 50:575-582. Metivier, G. 1973. The effects of long-lasting physical exercise and training on hormonal regulation. In: H. Howald and J. R. Poortmans, eds., Metabolic Adaptation to Prolonged Physical Exercise. Birkhauser Verlag, Basel. pp. 276-292. Milligan, L. P. 1971. Energetic efficiency and metabolic transforma- tions. Federation Proc. 30:1454—1465. Minaire, Y. and J. Forichon. 1973. Lactate metabolism and glucose lactate conversion in prolonged physical exercise. In: H. Howald and J. R. Poortmans, eds., Metabolic Adaptation to Prolonged Physical Exercise. Birkhauser Verlag, Basel, pp. 106-112. Mfiller, W. A., T. T. Aoki, and G. F. Cahill. 1975. Effect of alanine and glycine on glucagon secretion in postabsorptive and fasting obese man. J. Clin. Endocrinol. Metab. 40:418-425. Mfiller, W. A., G. R. Faloona, and R. H. Unger. 1971. The influence of antecedent diet upon glucagon and insulin secretion. N. Engl. J. Med. 285:1450—1454. Nazar, K., Z. Brzezinska, J. Lyszczara, and A. Danielewicz-Kotowicz. 1971. Sympathetic control of the utilization of energy sub- strates during long-term exercise in dogs. Arch. Intern. Physiol. Bioch. 79:873-879. Nestel, P. J. and P. Barter. 1971. Metabolism of palmitic and lino- leic acids in man: Differences in turnover and conversion to glycerides. Clin. Sci. 40:345-350. Nestel, P. J., K. F. Carroll, and N. Havenstein. 1970. Plasma trigly— ceride response to carbohydrates, fats and caloric intake. Metabolism. 19:1-18. Nilsson, L. H:son. 1973. Liver glycogen content in man in the post- absorptive state. Scand. J. Clin. Lab. Invest. 32:317-323. Nilsson, L. H:son, and E. Hultman. 1973. Liver glycogen in man--the effect of total starvation or a carbohydrate—poor diet followed by carbohydrate refeeding. Scand. J. Clin. Lab. Invest. 32:325-330. Nilsson, L. H:son., P. Ffirst, and E. Hultman. 1973. Carbohydrate metabolism of the liver in normal man under varying dietary conditions. Scand. J. Clin. Lab. Invest. 32:331-337. 231 Odessey, R., E. Khairallah, and A. L. Goldberg. 1975. Origin and possible significance of alanine production by skeletal muscle. J. Biol. Chem. 249:7623-7629. Olesen, E. S. and F. Quaade. 1960. Fatty foods and obesity. Lancet. 2:1048-1051. Olsson, A. G., B. Eklund, L. Kaijser, and L. A. Carlson. 1975. Extraction of endogenous plasma triglycerides by the working human forearm muscle in the fasting state. Scand. J. Clin. Lab. Invest. 35:231-236. Olsson, K.-E. and B. Saltin. 1970. Variation in total body water with muscle glycogen in man. Acta Physiol. Scand. 80:11-18. Oseid, S. and L. Hermansen. 1971. Hormonal and metabolic changes dur- ing and after prolonged muscular work in pre-pubertal boys. Acta Paediat. Scand. 217:147-153. (Suppl.) Owen, O. E. and G. F. Cahill Jr. 1973. Metabolic effects of exogenous glucocorticoids in fasted man. J. Clin. Invest. 52:2596-2605. Owen, O. E. and G. A. Reichard Jr. 1971a. Fuels consumed by man: The interplay between carbohydrates and fatty acids. Progr. Biochem. Pharmacol. 6:177-213. Owen, O. E. and G. A. Reichard. 1971b. Human forearm metabolism during progressive starvation. J. Clin. Invest. 50:1538-1545. Owen, O. E., P. Felig, A. P. Morgan, J. Wahren, and G. F. Cahill Jr. 1969. Liver and kidney metabolism during prolonged starvation. J. Clin. Invest. 48:574-583. Owen, O. E., A. P. Morgan, H. G. Kemp, J. M. Sullivan, M. G. Herrera, and G. F. Cahill Jr. 1967. Brain metabolism during fasting. J. Clin. Invest. 46:1589-1595. Pagliara, A. S., I. E. Karl, D. C. DeVivo, R. D. Feigin, and D. M. Kipnis. 1972. Hypoalaninemia: A concomitant of ketotic hypo— glycemia. J. Clin. Invest. 51:1440—1449. Pagliara, A. S., S. N. Stillings, B. Hover, D. M. Martin, and F. Matschinsky. 1974. Glucose modulation of amino acid-induced glucagon and insulin release-in the isolated perf sed rat pancreas. J. Clin. Invest. 54:819-832. Parker, J. C. and J. F. Hoffman. 1967. The role of membrane phospho- glycerate kinase in the control of glycolytic rate by active cation transport in human red blood cells. J. Gen. Physiol. 50:893-916. 232 Pearson, O. H., J. M. Dominguez, E. Greenberg, A. Pazianos, and B. S. Ray. 1960. Diabetogenic and hypoglucemic effects of human growth hormone. Trans. Assoc. Am. Physicians Phila. 73:217-226. Pelkonen, R., T. Miettinen, M. Taskinen, and E. Nikkila. 1968. Effect of acute elevation of plasma glycerol, triglyceride and FFA levels on glucose utilization and plasma insulin. Diabetes. Pernow, B. and B. Saltin, eds. 1971. Discussion. Advanc. Exp. Med. Biol. 11:548-551. Pernow, B. and J. Wahren. 1962. Lactate and pyruvate formation and oxygen utilization in the human forearm muscles during work of high intensity and varying duration. Acta. Physiol. Scand. 56:267-285. Pette, Dirk. 1971. Metabolic differentiation of distinct muscle types at the level of enzymatic organization. Advan. Exp. Med. Biol. 11:33—49. Pilkington, T. R. E., H. Gainsborough, V. M. Rosenoer, and M. Carey. 1960. Diet and weight—reduction in the obese. Lancet. Pi—Sunyer, F. X., R. G. Campbell, and S. A. Hashim. 1970. Experimental— ly induced hyperketonemia and insulin secretion in the dog. Metabolism. 19:263-270. Pi-Sunyer, F. X., S. A. Hashim, R. G. Campbell, and T. B. Van Itallie. 1967. Induced hyperketonemia on insulin secretion in man. Amer. J. Clin. Nutr. 20:365. Pi-Sunyer, F. X., S. A. Hashim, and T. B. Van Itallie. 1969. Insulin and ketone responses to ingestion of medium and long-chain tri- glycerides in man. Diabetes. 18:96-100. Pitts, G. C., R. E. Johnson, and F. C. Consolazio. 1944. Work in the heat as affected by intake of water, salt and glucose. Amer. J. Poortmans, J. R. 1973. Effects of long lasting physical exercise and training on protein metabolism. In: H. Howald and J. R. Poortmans, eds., Metabolic Adaptation to Prolonged Physical Exercise. Birkhfiuser Verlag, Basel. pp. 212-228. Portnay, G. I., J. T. O'Brian, J. Bush, A. G. Vagenakis, F. Azizi, R. A. Arky, S. H. Ingbar, and L. E. Braverman. 1974. The effect of starvation on the concentration and binding of thyroxine and triiodothyroxine in serum and the response to TRH. J. Clin. Endocrinology and Metab. 39:191-194. 233 Porzolt, F., D. Wagner, and K. H. Bichler. 1973. Das Serumkreatinin und die Nierenfunktion water k5rperlicher Belastung. Sportarzt. u. Sportmed. 24:27—30. Pozefsky, T., P. Felig, J. Tobin, J. S. Soeldner, and G. F. Cahill Jr. 1969. Amino acid balance across tissues of the forearm in post- absorptive man. Effects of insulin at two dose levels. J. Clin. Invest. 48:2273-2282. ' Pozefsky, T., R. G. Tancredi, R. T. Moxley, J. Dupre, and J. Tobin. 1974. Forearm tissue metabolism in post-absorptive and 60-h- fasted man: Studies with glucagon. J. Clin. Invest. 53:61a. Pozefsky, T., R. G. Tancredi, R. T. Moxley, J. Dupre, and J. D. Tobin. 1976. Effects of brief starvation on muscle amino acid metabol- ism in nonobese man. J. Clin. Invest. 57:444-449. Pruett, E. D. R. 1970. Molbilization during and after prolonged severe muscular work in men. J. Appl. Physiol. 29:809-815. Pruett, E. D. R. and S. Osid. 1970. Effect of exercise on glucose and insulin response to glucose infusion. Scand. J. Clin. Invest. 26:277-285. Raben, M. S. and C. H. Hollenberg. 1959. Effects of growth hormone on plasma free fatty acids. J. Clin. Invest. 38:484-488. Randle, P. J., P. J. England, and R. M. Denton. 1971. Control of the tricarboxylate cycle and its interactions with glycolysis during acetate utilization in rat heart. Biochem. J. 117:677-695. Rayyis, S. S. and R. Horton. 1971. Effect of angiotensis II on adrenal and pituitary function in man. J. Clin. Endocr. 32:539-546. Reddy, B. S., J. H. Weisburger, and E. L. wynder. 1975. Effects of high risk and low risk diets for colon carcinogenesis on fecal microflora and steroids in man. J. Nutr. 105:878-884. Refsum, H. E., B. Tveit, H. D. Meen, and S. B. Stramme. 1973. Serum electrolyte, fluid and acid-base balance afrer prolonged heavy exercise at low environmental temperature. Scand. J. Clin. Lab. Relman, A. S. 1972. Metabolic consequences of acid—base disorders. Kidney International. 1:347-359. Rennie, M. J., S. Jennett, and R. H. Johnson. 1974. The metabolic ef— fects of strenuous exercise: A comparison between untrained subjects and racing cyclists. Quart. J. Exp. Physiol. 59:201- 212. 234 Rocha, D., G. R. Faloona, and R. H. Unger. 1972. Glucagon-stimulating activity of twenty amino acids in dogs. J. Clin. Invest. 51: 2346-2351. Rooth, G. 1966. Acid-base electrolyte and water changes during tissue hypoxia in rats. Clin. Sci. 30:417-424. Rosell, S. and K. Ballard. 1971. Adrenergic neuro-humoral control of lipolysis in adipose tissue. Advan. Exp. Med. Biol. 11:111-117. Rossini, A. A., T. T. Aoki, O. P. Ganda, J. S. Soeldner, and G. F. Cahill Jr. 1975. Alanine-induced amino acid interrelationships. Metabolism. 24:1185-1192. Rougier, G. and J. P. Babin. 1975. A blood and urine study of heavy muscular work on ureic and uric metabolism in man. J. Sports Med. 15:212—222. Rowell, L. B., J. R. Blackmon, and R. A. Bruce. 1964. Indocyanine green clearance and estimated hepatic blood flow during mild to maximal exercise in upright man. J. Clin. Invest. 43:1677—1690. Rowell, L. B., K. K. Kraning, II, T. 0. Evans, J. W. Kennedy, J. R. Blackmon, and F. Kusumi. 1966. Splanchnic removal of lactate and pyruvate during prolonged exercise in man. J. Appl. Physiol. 21:1773-1783. Ruderman, N. B. and M. Berger. 1974. The formation of glutamine and alanine in skeletal muscle. J. Biol. Chem. 249:5500-5506. Rudorff, K.-H., G. Albrecht, and W. Staib. 1970. Effect of insulin and proinsulin on the metabolism of alanine in the rat liver. Horm. Metab. Res. 2:49-51. Russell, G. F. M. 1962. The effect of diets of different composition on weight loss, water and sodium balance in obese patients. Clin.Sci. 22:269-277. Saltin, B. 1964a. Aerobic and anaerobic work capacity after dehydra— tion. J. Appl. Physiol. 19:1114-1118. Saltin, B. 1964b. Circulatory response to submaximal and maximal exercise after thermal dehydration. J. Appl. Physiol. 19:1125- 1132. Saltin, B. 1973. Adaptive changes in carbohydrate metabolism with exer— cise. In: H. Howald and J. R. Poortmans, eds., Metabolic Adapta- tion to Prolonged Physical Exercise. Birkhauser Verlag, Basel. pp. 94—100. 235 Saltin, B. and B. Essen. 1971. Muscle glycogen, lactate, ATP, and CP in intermittent exercise. Advan. Exp. Med. Biol. 11:419-424. Saltin, B. and J. Karlsson. 1971a. Muscle ATP, CP, and lactate during exercise after physical conditioning. Advan. Exp. Med. Biol. 11:395-399. Saltin, B. and J. Karlsson. 1971b. Muscle glycogen utilization during work of different intensities. Advanc. Exp. Med. Biol. 11:289- 299. Samols, E. and D. Holdsworth. 1968. Disturbances in carbohydrate metabolism: Liver disease. In: F. Dickens, P. J. Randle, and W. J. Whelan, eds., Carbohydrate Metabolism and Its Disorders, Vol. II. Academic Press, New York. pp. 289-336. Samols, E., G. Marri, and V. Marks. 1965. Promotion of insulin secre- tion by glucagon. Lancet. 2:415-417. Samols, E., G. Marri, and V. Marks. 1966. Interrelationship of gluca- gon, insulin and glucose: The insulinogenic effect of glucagon. Diabetes. 15:855-866. Sapir, D. G., O. E. Owen, T. Pozefsky, and M. Walser. 1975. Nitrogen sparing induced by a mixture of essential amino acids given chiefly as their ketoanalygues during prolonged starvation in obese subjects. J. Clin. Invest. 54:974-980. Sartorius, O. W., J. C. Roemmelt, and R. F. Pitts. 1949. The renal regulation of acid—base balance in man. IV. The nature of the renal compensations in ammonium chloride acidosis. J. Clin. Invest. 28:423-439. Schalch, D. S. 1967. The influence of physical stress and exercise on growth hormone and insulin secretion in man. J. Lab. Clin. Med. 69:256-269. Senior, B. and L. Loridan. 1968. Direct regulatory effect of ketones on lipolysis and on glucose concentrations in man. Nature. 219:83-84. Seyffert, W. A. Jr. and L. L. Madison. 1967. Physiologic effects of metabolic fuels on carbohydrate metabolism. 1. Acute effect of elevation of plasma free fatty acids on hepatic glucose output, peripheral glucose utilization, serum insulin, and plasma gluca- gon levels. Diabetes. 16:765-776. Shimizu, N. and T. Kumamoto. 1952. Histochemical studies on the glyco- gen of the mammalian brain. Anat. Rec. 114:479-497. 236 Sinclair, H. M. 1964. Carbohydrates and fats. In: G. H. Beaton and E. W. McHenry, eds., Nutrition, Vol. 1. Academic Press, New York. pp. 41-114. Smith, 0. K. and C. N. H. Long. 1967. Effect of cortisol on the plasma amino nitrogen of eviscerated adrenalectomized-diabetic rats. Endocrinology. 80:561-566. Spark, R. F., R. A. Arky, P. R. Boulter, C. D. Saudek, and J. T. O'Brien. 1975. Renin, aldosterone and glucagon in the natriuresis of fasting. N. Engl. J. Med. 292:1335-1340. Sréter, F. A. 1963. Distribution of water, sodium, and potassium in resting and stimulated mammalian muscle. Canad. J. Biochem. 41:1035-1045. Staib, A., K. Feller, and K. Andreas. 1964. Beobachtungen Uber pharma- kologishe Beeinflussungmaglichkeiten des Blut-pH. Med. und Sport. 4:181—182. Steinberg, D. M., H. Shafrir, and M. Vaughan. 1959. Direct effect of glucagon on release of esterified fatty acids (J. F. A.) from adipose tissue. Clin. Res. 7:250. Steiner, A. L., A. D. Goodman, and D. H. Treble. 1968. Effect of meta- bolic acidosis on renal gluconeogenesis ig_vivo. Amer. J. Stinebaugh, F. J. and F. X. Schloeder. 1966a. Studies on the natri- uresis of fasting. I. Effect of pre-fast intake. Metabolism. 15:828-837. Stinebaugh, B. J. and F. X. Schloeder. 1966b. Studies on the natri- uresis of fasting. II. Relationship to acidosis. Metabolism. Sundsfjord, J. A., S. B. Stromme, and A. Aakvaag. 1973. Plasma aldo- sterone (PA), plasma renin activity (PRA), and cortisol (PF) during exercise. In: H. Howald and J. R. Poortmans, eds., Metabolis Adaptation to Prolonged Physical Exercise. Birkhfiuser Verlay, Basel. pp. 308-314. Sutton, D. C., N. L. Jones, and C. T. Toewa. 1976. Growth hormone secretion in acid-base alteration at rest and during exercise. Clin. Sci. Mol. Med. 50:241-247. Sweeley, C. C., N. D. Young, J. F. Holland, and S. C. Gates. 1974. Rapid computerized identification of compounds in complex bio- logical mixtures by gas chromatography-mass spectrometry. 237 Swift, R. W. and K. H. Fisher. 1964. Energy metabolism. In: G. H. Beaton and E. W. McHenry, eds., Nutrition, Vol. I. Academic Press, New York. pp. 181-260. Tantibhedhyangkul, P. S. A. Hashim, and T. B. VanItallie. 1967. Effects of ingestion of long—chain and medium-chain triglycerides on glucose tolerance in man. Diabetes. 16:796-799. Taylor, H. L., J. Brozek, A. Henschel, O. Mickelsen, and A. Keys. 1945. The effect of successive fasts on the ability of men to withstand fasting during hard work. Amer. J. Physiol. 143:148-155. Taylor, H. L., A. Henschel, and A. Keys. 1954. Some effects of acute starvation with hard work on body weight, body fluids and metab- olism. J. Appl. Physiol. 6:613-623. Taylor, Y. S. M., N. S. Scrimshaw, and V. R. Young. 1974. The relation- ship between serum urea levels and dietary nitrogen utilization in young men. Br. J. Nutr. 32:407-411. Tischler, V., 0. Pavkovcekova, P. Beno, J. Jacina, I. Lahita, and D. Kahanec. 1970. Der Zitronensauremetabolismus der Niere im Sauglingsalter bei der Applikation von NH4C1. Kinderarztliche Praxis. 6:241-246. Tolstoi, E. 1929. The effect of an exclusive meat diet on the chemical constituents of the blood. J. Biol. Chem. 83:753-758. Tornheim, K. and J. M. Lowenstein. 1973. The purine nucleotide cycle. III. Oscillations in metabolite concentrations during the operation of the cycle in muscle extracts. J. Biol. Chem. 248:2670-2677. Turner, L. V. and K. L. Manchester. 1973. Influence of denervation of the free amino acids of the rat diaphragm. Biochim. Biophys. Acta. 320:352-356. Ullrich, K. J. and D. J. Marsh. 1963. Kidney and water and electrolyte metabolism. Ann. Rev. Physiol. 25:91-142. Unger, R. H. 1971. Glucagon physiology and path0physiology: Abnormal alpha cell function as a characteristic of human diabetes mel- litus. N. Engl. J. Med. 235:443-449. Unger, R. H. 1974. Alpha- and beta-cell interrelationships in health and disease. Metabolism. 23:581-593. Unger, R. H. 1975. Pancreas as a regulator of metabolism. MTP Inter— national Review of Science. Endocrine Physiology. Physiology Series One. 5:179-204. 238 Unger, R. H. and L. Orci. 1976. Physiology and patholphysiology of glucagon. Physiol. Rev. 56:778—826. Unger, R. H., A. Ohneda, E. Aguilar-Parada, and A. M. Eisentraut. 1969. The role of aminogenci glucagon secretion in blood glucose homeostasis. J. Clin. Invest. 48:810-822. Vagenakis, A. G., A. Burger, G. I. Portnay, M. Rudolph, J. T. O'Brian, F. Azizi, R. A. Arky, P. Nirod, S. H. Ingbar, and L. E. Braverman. 1975. Diversion of peripheral thyroxine metabolism from activating to inactivating pathways during complete fasting. J. Clin. Endocrinology and Metabolism. 41:191-194. Van Huss, W. D. 1977. Effect of specific exercise regimen. Pending publication in Osteopathic Annals. Veverbrants, E. and R. A. Arky. 1969. Effects of fasting and refeeding. I. Studies on sodium, potassium, and water excretion on a con- stant electrolyte and fluid intake. J. Clin. Endocr. 29:55-62. Viru, A. and M. Oks. 1972. Effect of physical exertion on the blood level of bound and free corticoids. End. Exp. 66:227-230. Vrba, R. 1954. Beitrag zum Studium des Gehirn—metabolismus im Zusammen— hang mit kSrperlicher Anstrengung. I. Glykogen und Glykolyse. - Physiol. bohemoslov. 3:122-134. Wahren, J., P. Felig, G. Ahlborg, and L. Jorfeldt. 1971. Glucose metab- olism during leg exercise in man. J. Clin. Invest. 50:2715-2725. Wahren, J., P. Felig, E. Cerasi, and R. Luft. 1972. Splanchnic and, peripheral glucose and amino acid metabolism in diabetes mellitus. J. Clin. Invest. 51:1870-1878. Wahren, J., P. Felig, and L. Hagenfeldt. 1976. Effect of protein ingestion on splanchnic and leg metabolism in normal man and in patients with diabetes mellitus. J. Clin. Invest. 57:987-999. Wahren, J., P. Felig, L. Hagenfeldt, R. Hendler, and G. Ahlborg. 1973a. Splanchnic and leg metabolism of glucose, free fatty acids and amino acids during prolonged exercise in man. In: H. Howald and J. R. Poortmans, eds., Metabolic Adaptation to Prolonged Physical Exercise. Birkhauser Verlag, Basel. pp. 144-153. Wahren, J., P. Felig, R. J. Havel, L. Jorfeldt, B. Pernow, and B. Saltin. 1973. Amino acid metabolism in McArdle's syndrom. N. Engl. J. Med. 288:774-777. 239 Wahren, J., P. Felig, R. Hendler, and G. Ahlborg. 1973b. Glucose and amino acid metabolism during recovery after exercise. J. Appl. Physiol. 34:838-845. Wahren, J., L. Hagenfeldt, and P. Felig. 1975. Splanchnic and leg exchange of glucose, amino acids, and free fatty acids during exercise in diabetes mellitus. J. Clin. Invest. 55:1303-1314. Watt, B. K. and A. L. Merrill. 1975. Composition of Foods. Agricul- ture Handbook, No. 8. U. S. Department of Agriculture, Washington, DC. 188 pp. Weinsier, R. L. 1971. Fasting-—a review with emphasis on the electro- lytes. Amer. J. Med. 50:233-240. Werner, S. C. 1955. Comparison between weight reduction on a high- calorie, high-fat diet and on an isocaloric regimen high in carbohydrate. N. Engl. J. Med. 252:661—665. Weiss, L. and G. LBffler. 1970. Interrelationship between adipose tissue and liver: Gluconeogenesis and ketogenesis. Hormone. Met. Res. Suppl:196-203. White, A., P. Handler, and E. L. Smith. 1973a. Principles of Biochem- istry, 5th Edition. McGraw-Hill, New York. p. 453. White, A., P. Handler, and E. L. Smith. 1973b. Principles of Biochem- istry, 5th Edition. McGraw—Hill, New York. pp. 555-556. White, A., P. Handler, and E. L. Smith. 1973c. Principles of Biochem- istry, 5th Edition. McGraw-Hill, New York. p. 992. Whitehead, R. G., W. A. Coward, and P. G. Lunn. 1973a. Serum-albumin concentration and the onset of Kwashiorkor. Lancet. 1:63-66 Whitehead, T. P., D. M. Browning, and A. Gregory. 1973b. A comparative survey of the results of analyses of blood serum in clinical chemistry laboratories in the United Kingdom. J. Clin. Path. 26:435-445. Wieland, O. 1968. Ketogenesis and its regulation. Advanc. Metabolic Disorders. 3:1—47. Williams, M. H. 1976. Nitritional Aspects of Human Physical and Athletic Performance. Charles C. Thomas, Springfield, Illinois. pp. 27-30. Williamson, D. H. 1971. Physiological ketosis, or why ketone bodies? Postgraduate Medical J. Supp1:371-375. 240 Williamson, D. H., M. W. Bates, M. A. Page, and H. A. Krebs. 1971. Activities of enzymes involved in acetoacetate utilization in adult mammalian tissues. Biochem. J. 121:41-47. Williamson, J. R. and H. A. Krebs. 1961. Acetoacetate as fuel of respiration in the perfused rat heart. Biochem. J. 80:540-547. Willms, B., M. B3ttcher, v. Wolters, N. Sakamoto, and H. D. SESling. 1969. Relationship between fat and ketone body metabolism in obese and nonobese diabetics and nondiabetics during norepine- phrine infusion. Diabetologia. 5:88-96. Wilson, D., C. Bishop, and J. H. Talbott. 1952. A factorial experiment to test the various types of diets on uric acid excretion of normal human subjects. J. Appl. Physiol. 4:560—565. Wise, J. K., R. Hendler, and P. Felig. 1973a. Evaluation of alpha-cell function by infusion of alanine in normal, diabetic and obese subjects. N. Engl. J. Med. 288:487-490. Wise, J. K., R. Hendler, and P. Felig. 1973b. Influence of glucocorti- coids on glucagon secretion and plasma amino acid concentrations in man. J. Clin. Invest. 52:2774-2782. Wool, I. G., J. J. Castles, D. P. Leader, and A. Fox. 1972. Insulin and the function of muscle ribosomes. In: D. F. Steiner and N. Freinkel, eds., Handbook of Physiology, Sec. 7, Vol. I, Endocrine Pancreas. Amer. Physiol. Soc., Washington, D.C. pp. 385-394. Worthington, B. S. and L. E. Taylor. 1974. Balanced low-calorie vs. high-protein-low-carbohy-rate reducing diets. J. of the Amer. Dietetic Assoc. 64:47-55. Wright, H. K., D. S. Gann, and K. Albertsen. 1963. Effect of glucose on sodium excretion and renal concentrating ability after starva- tion in man. Metabolism. 12:804-811. Yang, M. and T. B. Van Itallie. 1976. Composition of weight lost dur- ing short-term weight reduction: Metabolic responses of obese subjects to starvation and low-calorie ketogenic and nonketogenic diets. J. Clin. Invest. 58:722-730. Yudkin, J. and M. Carey. 1960. The -reatment of obesity by the "high-fat" diet. Lancet. 2:939-941. Zierler, K. L., A. Maseri, G. Klassen, D. Rabinowitz, and J. Burgess. 1968. Muscle metabolism during exercise in man. Trans. Ass. 241 Zilva, J. F. and P. R. Pannall, 1975. Clinical Chemistry. In: Diagnosis and Treatment, 2nd Edition. Year Book Medical Pub- lishers, Chicago. pp. 441-443. Zinnerman, H. H., F. Q. Nuttall, and F. C. Goetz. 1966. Effect of endogenous insulin on human amino acid metabolism. Diabetes. 15:5-8. Zuntz, N. 1911. Betrachtung fiber die Beziehungen Zwischen Nahrstoffen und Leisbungen des K5rpers. Oppenheimer's Handb. Biochemie. 4:826-881. APPENDICES 242 APPENDIX A Composition of the high carbohydrate control diet for Experiment I (approx. 10% of calories from protein and 75% from carbohydrates). Ingredients Weight2 Cal.3 Water2 Protein CHO2 Fat2 Soyamel4 50 247 2 9 28 12 Strawberries 100 37 90 l 10 1 Banana 100 85 76 l 22 0 Orange 100 49 86 l 13 0 Lemon 50 13 45 0 5 0 Grapefruit 150 66 132 l 18 0 Granolas so 230 3 6 34 9 Breakfast Total 600 727 436 19 130 22 Mixed Vegetables6 481 .149 438 5 37 0 Baked Potatoes 150 140 113 4 33 0 Sweet Potatoes6 241 274 171 2 67 0 Fruit Cocktail 411 200 329 2 78 0 Soyameat7 100 232 110 18 11 13 Dinner Total 1383 1095 1161 31 226 13 Vegetable Soup6 305 201 259 4 37 6 Apple 100 60 84 0 16 1 Bread 107 236 46 12 47 2 Cake 50 150 18 2 26 4 Honey 50 165 8 0 41 0 Jam 50 l6 l6 0 53 0 Supper Total 662 815 431 18 200 13 TOTAL 2645 2637 2028 68 556 48 1The total caloric intake was adjusted to the subjects' individual re- quirements by varying the amount of breakfast. The same diet was used for all CHO periods. 2Weight in grams. 3Calories in kcal. 4Fortified Soyamel, Worthington Foods, Worthington, Ohio. SKellogg Company, Battle Creek, Michigan. 6Food Club, Meijer, Inc., Grand Rapids, Michigan. 7Soyameat, Salisbury Steak style, Worthington Foods. 243 APPENDIX B Composition of the Fat I diet during Experiment 1. (approx. 10% of calories from protein and 15% from carbohydrates.) Ingredients Weight2 Cal.3 Water2 Protein2 CHO2 Fat2 Granola4 5 25 124 O 3 16 5 Soya Protein 20 76 0 19 O 0 Orange 50 25 43 0 6 0 Fresh Fruit 100 40 90 0 5 0 Corn Oil 60 540 0 0 O 60 Avocadog 100 171 74 2 4 17 Breakfast Total 355 976 207 24 31 82 Baked Potatoes 50 47 38 l 10 0 Soyameat6 100 212 62 17 10 11 Dressing7 8 150 736 56 3 8 83 Fruit Cocktail 50 38 18 0 10 0 Whipped Topping 25 65 15 0 5 5 Dinner Total 375 1098 189 21 43 99 Bread 54 118 23 6 24 1 Peanut Butter 25 145 0 7 5 13 Cream Cheese 100 353 54 7 4 35 Dressing7 50 246 19 1 3 28 Supper Total 229 862 96 21 36 77 TOTAL 959 2911 '497 66 108 256 1The total caloric intake was adjusted to the subjects' individual re- quirements by varying the amount of greakfast. N Weight in grams. Calories in kcal. u: Kellogg Company, Battle Creek, Michigan. Soya Protein, Fearn Soya Foods, Melrose Park, Illinois. Soyameat, Wham style, Worthington Foods, Worthington, Ohio. For composition of the dressing, see Appendix D. Food Club, Meijer, Inc., Grand Rapids, Michigan. \oooxloxmb Fuerte variety (Calavo), California. 244 APPENDIX C Composition of the Fat II diet during Experiment I (approx. 6% of calories from protein and 9% from carbohydrates). Ingredients Weight2 Cal3 Water2 Protein2 CH02 Fat2 Granola4 5 25 124 o 3 16 5 Soya Protein 15 56 0 14 0 0 Lemon 50 13 45 0 5 0 Fresh Fruit 100 40 90 0 5 0 Corn Oil 100 884 O O O 100 Avocado6 100 171 74 2 4 17 Breakfast Total 390 1287 209 19 30 122 Baked Potatoes 50 47 38 1 10 0 Soyameat7 100 206 70 9 3 18 Dressing8 150 736 56 3 8 83 Dinner Total 300 990 164 13 21 101 Amer. Process Cheese9 50 194 22 ll 2 16 Heavy Whipping Cream 130 458 74 3 4 49 Fruit Cocktail 50 36 41 0 10 0 Supper Total 230 688 137 l4 16 65 TOTAL 920 2965 510 46 67 288 1The total caloric intake was adjusted to the subjects' individual re— quirements by varying the amount of breakfast. N Weight in grams. Calories in kcal. Kellogg Company, Battle Creek, Michigan Fearn Soya Foods, Melrose Park, Illinois. Fuerte variety (Calavo), California. Soyameat, Chicken style, Worthington Foods, Worthington, Ohio. For composition of the dressing, see Appendix D. Food Club, Meijer, Inc., Grand Rapids, Michigan. \OGJNO\UIJ-\w 245 APPENDIX D Composition of the Dressing used in the high fat diets during Experiment I. Ingredient Weightl Cal.2 Water1 Proteinl CHOl Fat1 Avocad03 100 171 74 2 4 17 Soyamel 50 235 3 11 23 ll Corn Oil 300 2532 0 0 0 300 Vinegar 50 7 47 0 3 0 Water 100 0 100 0 0 0 TOTAL 600 2945 224 13 30 328 Dressing5 100 491 37 2 5 55 1Weight in grams. 2Calories in kcal. 3Fuerte variety (Calavo), California. 4Fortified Soyamel, Worthington Foods, Worthington, Ohio. 5Composition per 100 g dressing. 246 APPENDIX E Composition of the high carbohydrate control diet for Experiment 11 (approx. 5% of calories from protein and 85% from carbohydrates).1 Ingredients Weight2 Cal.3 Water2 Protein2 CHO2 Fat2 Orange 200 98 172 2 24 0 Grapefruit 100 40 88 l 10 0 Lemon 50 14 45 l 4 0 Apple 260 156 218 1 36 2 Banana 300 255 227 3 65 l Dried Apple 185 509 44 2 127 3 Soya Protein 3 12 O 3 0 0 Breakfast Total 1098 1084 794 13 266 6 Corn Oil 10 88 0 0 0 10 Sweet Potatoes 250 210 190 4 55 1 Minute Rice5 50 194 35 4 41 0 Baked Potatoes 200 186 150 5 42 0 Jelly 150 450 44 O 112 0 Dinner Total 660 1128 419 13 250 11 Vegetable Soup5 305 201 259 4 37 6 Apple 200 120 168 0 30 1 Orange 100 49 86 1 12 0 Bread 26 59 9 3 12 O Margarine 10 72 2 O 0 8 Jam 24 72 6 0 l8 0 Supper Total 665 573 530 8 109 15 TOTAL 2423 2785 1743 34 625 32 1The total caloric intake was adjusted to the subjects' individual re- quirements by varying the amount of breakfast. Weight in grams. Calories in kcal. Fearn Soya Foods, Melrose Park, Illinois. Food Club, Meijer, Inc., Grand Rapids, Michigan. {.11wa 247 APPENDIX F Composition of the Fat I diet for Experiment II (approx. 5% of calories from protein and 6% from carbohydrates).1 Ingredients Weight2 Cal.3 Water2 Protein2 CHO2 Fat2 Heavy Whipping Cream 250 945 140 5 6 100 Soya Protein4 6 24 0 6 0 0 Water-canned Fruit 125 39 114 1 9 0 TOTAL 381 254 1008 12 15 100 1 The same mixture was eaten three times a day, and portions were adjusted to the subjects' individual caloric requirements. 2Weight in grams. 3Calories in kcal. 4Fearn Soya Foods, Melrose Park, Illinois. 248 APPENDIX G Composition of the Fat II diet for Experiment 11 (approx. 5% of calories from protein and 6% from carbohyerates). Ingredients Weight2 Cal.3 Water2 Protein2 CHO2 Fat2 Avocado4 200 342 148 4 8 34 Corn Oil 70 619 0 O 0 70 Lemon 5 50 14 45 l 4 O Soya Protein 8 32 0 8_ 0 0 Breakfast Total 328 1007 193 13 12 104 Canned Carrots6 50 16 47 O 4 0 Dressing7 241 839 133 6 10 89 Soyameat8 70 144 49 6 2 13 Dinner Total9 361 999 229 12 16 102 TOTAL 1050 3005 651 37 44 308 1The total caloric intake was adjusted to the subjects' individual re— quirements by varying the amount of breakfast. 2Weight in grams. 3Calories in kcal. 4Fuerte variety (Calavo), California. Fearn Soya Foods, Melrose Park, Illinois. Food Club, Meijer, Inc., Grand Rapids, Michigan. For composition of the dressing, see Appendix H. Soyameat, Chicken style, Worthington Foods, Worthington, Ohio. \OmNO‘UI This meal was used for both dinner and supper. 249 APPENDIX H Composition of the Dressing used in the Fat II diet during Experiment II. Ingredients Weight1 Cal.2 Water1 Protein1 CHO1 Fat1 Avocad03 150 257 111 3 6 26 Almonds 25 150 l 5 4 l4 Soya Protein4 1 4 0 l 0 0 Lemon 50 13 45 0 5 0 Corn Oil 100 884 O O O 100 Water 50 0 50 0 O 0 TOTAL 376 1308 207 9 15 140 Dressings 100 348 55 2.4 4 37 1Weight in grams. 2Calories in kcal. 3Fuerte variety (Calavo), California. 4Fearn Soya Foods, Melrose Park, Illinois. 5Composition per 100 g dressing. 250 APPENDIX 1 Composition of the high carbohydrate control diet for Experiment III (approx. 15% of calories from protein and 60% from carbohydrates).1 Ingredients Weight2 Cal.3 Water2 Protein2 CHO2 Fat2 Orange 100 51 85 l 12 0 Apple 100 60 84 0 l4 1 Lemon 50 13 45 1 4 0 Grapefruit 200 82 176 l 21 0 Banana 4 100 85 76 1 22 0 Granola 50 230 3 6 34 9 Soyamel 6 50 247 2 9 27 13 Soya Protein 15 56 0 l4 0 0 Breakfast Total 665 824 471 ' 33e.e_. 134 fi23 Baked Potatoes 200 186 150 5 41 0 Soyameat7 70 162 39 13 8 9 Soybeans 100 150 66 14 14 4 Canned Pears 400 272 328 0 72 0 Cool Whiplo 50 154 28 o 11 11 Dinner Total 820 924 611 32 146 24 Vegetable Soup9 305 200 259 4 36 6 Bread 57 140 14 6 27 2 Peanut Butter 30 174 1 8 5 15 Soyameatll 70 148 43 12 7 8 Apple 200 120 168 0 28 2 Orange 12 100 51 85 l 12 O Waffer Snack 50 150 18 2 26 4 Supper Total 812 983 588 33 141 37 TOTAL 2297 2731 1670 98 421 84 1The total caloric intake was adjusted to the subjects' individual re- quirements by varying the amount of breakfast. N Weight in grams. 3Calories in kcal. 4Kellogg Company, Battle Creek, Mi. Fortified Soyamel, Worthington Foods, Worthington, Ohio. Fearn Soya Foods, Melrose Park, Illinois. Soyameat, Salisbury Steak style, Worthington. Soybeans, Boston Bean style, Loma Linda Foods, Riverside, CA. Food Club, Meijer, Inc., Grand Rapids, Michigan. 10General Foods, White Plains, New York. 11Soyameat, Wham (ham style slices), Worthington. «>00onan 12Nutty Bars, Little Debbie, McKee Baking Company, Collegedale, Tenn. 251 APPENDIX J Composition of the Fat I diet for Experiment III (approx. 15% of calories from protein and 5% from carbohydrates).1 Ingredients Weight2 Cal.3 Water2 Protein2 CHO2 Fat2 Eggs 4 100 216 68 13 1 l7 Amer. Process Cheese 57 210 40 12 2 18 Sour Cream4 225 468 162 8 8 47 Breakfast Total 382 894 270 33 11 82 Canned Carrots4 65 20 60 0 5 0 Butter 45 324 14 1 1 37 Meat5 4 80 190 44 28 0 8 Sour Cream 113 234 81 4 4 24 Dinner Total6 303 768 199 33 10 69 TOTAL 988 2430 668 99 31 220 1The total caloric intake was adjusted to the subjects' individual re- quirements by varying the amount of breakfast. N Weight in grams. Calories in kcal. Food Club, Meijer, Inc., Grand Rapids, Michigan. Beef bottom round (lean and trimmed). O‘UI-I-‘UJ This meal was used for both dinner and supper. 252 APPENDIX K Composition of the Fat II diet for Experiment III (approx. 15% of calor- ies from protein and 5% from carbohydrates).1 Ingredients Weight2 Cal.3 Water2 Protein2 CHO2 Fat2 Avocado4 50 84 37 1 2 8 Soyameat 225 446 153 29 7 36 Almonds 10 60 0 2 2 5 Mayonaise 40 286 9 0 0 31 Breakfast Total 325 876 199 32 ll 80 Soyamea 7 340 700 240 30 10 60 Avocado 50 84 37 1 2 8 Mayonaise 15 109 3 0 0 12 Dinner Total 405 893 280 31 12 80 TOTAL 1135 2662 759 94 35 240 1The total caloric intake was adjusted to the subjects' individual re- quirements by varying the amount of breakfast. 2Weight in grams. 3Calories in kcal. 4Fuerte variety (Calavo), California. Chicken style slices, Worthington Foods, Worthington, Ohio. Kraft Foods, Chicago, Illinois. Chicken style pieces, Worthington. (”NONU‘I This meal was used for both dinner and supper. 253 APPENDIX L The content of EAA in the high carbohydrate control diet for Experiment II, calculated from the diet composition given in Appendix E.1 (All values are in g.) Protein Source N Ile. Leu. Lys. TSAA TAAA Thr. Thp. Val. Apple 0.48 0.11 0.19 0.18 0.06 0.12 0.11 0.028 0.12 Soya Protein 0.53 0.11 0.14 0.17 0.20 0.23 0.11 0.039 0.14 Banana 0.48 0.09 0.12 0.12 0.14 0.20 0.10 0.035 0.12 Sweet Potato 0.64 0.15 0.22 0.14 0.11 0.25 0.15 0.068 0.31 Potato 0.80 0.19 0.30 0.23 0.09 0.34 0.19 0.082 0.26 Rice 0.67 0.16 0.35 0.16 0.14 0.36 0.16 0.052 0.23 Vegetable Soup4 0.64 0.14 0.25 0.20 0.08 0.26 0.15 0.047 0.19 Citrus Fruit 0.80 0.14 0.14 0.26 0.14 0.29 0.08 0.035 0.19 Bread 0.53 0.12 0.20 0.08 0.12 0.24 0.10 0.036 0.14 TOTAL 5.57 1.12 1.91 1.54 1.08 2.29 1.15 0.42 1.70 Requirements 6.37 0.70 1.10 0.80 1.01 1.10. 0.50 0.25 0.80 1 \J‘I-l-‘UJN The AA composition is based on the FAO Based on the mean of the major constituents. Total aromatic AA (Phenylalanine and Tyrosine). tables (Italy, 1970). Total S-containing AA (Methionine and Cystine). From Table 6, p. 40, column 8 of "Improvement of protein nutriture" (NAS, Washington, DC. , 1974). These figures are for a 70 kg man and are the highest values required for any subject in the experiment. 254 APPENDIX M The content of EAA in the Fat I diet for Experiment II, calculated from the diet composition given in Appendix F.1 Protein Source N Ile. Leu. Lys. TSAA2 TAAA3 Thr. Thp. Val. Cream 0.78 0.31 0.61 0.35 0.16 0.64 0.22 0.071 0.36 Soya Protein 1.01 0.28 0.49 0.40 0.16 0.51 0.24 0.081 0.30 Canned Pears 0.16 0.03 0.04 0.03 0.02 0.06 0.03 0.009 0.04 Sum 1.95 0.62 1.14 0.78 0.34 1.21 0.49 0.161 0.70 TOTAL4 5.85 1.86 3.42 2.34 1.02 3.63 1.47 0.48 2.10 Requirements 6.37 0.70 1.10 0.80 1.01 1.10 0.50 0.25 0.80 1The AA composition is based on the FAO tables (Italy, 1970). 2Total S-containing AA (Methionine and Cystine). 3Total aromatic AA (Phenylalanine and Tyrosine). 4 The total intake per day is calculated by multiplying the sum by three, since identical meals were fed three times a day. 5From Table 6, p. 40, column 8 of "Improvement of protein nutriture" These figures are for a 70 kg man, and are the highest values required for any subject in the experiment. (NAS, Washington, DC, 1974). APPENDIX N 255 The content of EAA in the Fat II diet for Experiment from the diet composition given in Appendix C. II, calculated 2 3 Protein Source N Ile. Leu. Lys. TSAA TAAA Thr. Thp. Val. Avocado 1.37 0.29 0.47 0.37 0.34 0.50 1.12 0.170 0.40 Lemon 0.20 0.04 0.03 0.07 0.03 0.07 0.02 0.009 0.05 Soya Protein 1.56 0.44 0.76 0.62 0.25 0.79 0.38 0.125 0.47 Almonds 1.24 0.22 0.38 0.31 0.19 0.42 0.22 0.087 0.28 Soyameat 2.10 0.60 1.02 0.84 0.34 1.06 0.51 0.168 0.63 TOTAL 6.47 1.59 2.66 2.21 1.15 2.84 2.25 0.56 1.83 Requirement4 6.37 0.70 1.10 0.80 1.01 1.10 0.50 0.25 0.80 1The AA composition is based on the FAO tables (Italy, 1970). 2Total S-containing AA (Methionine and Cystine). 3Total aromatic AA (Phenylalanine and Tyrosine). 4 From Table 6, p. 40, column 8 of "Improvement of protein nutriture" These figures are for a 70 kg man, and are the highest values required for any subject in the experiment. (NAS, Washington, DC, 1974).