’ FAQ-gr ' - v‘ '_ » ‘_._._ .- -.‘ _ _ I I T 53‘ "is lit. . _1 I J" “ I i t. I! R _ y» ucma‘»h‘-l ‘. in . . v. ’ ‘ - 'L "'tn-Kib‘& 1" ‘hf‘e University n: -J¢Io'. -‘ .'—. 17-.- This is to certify that the thesis entitled CHANGES IN TISSUE AND BODY FLUID ACYLCARNITINES IN RESPONSE TO DIFFERENT PHYSIOLOGICAL STATES presented by KIM JOSEPH VALKNER has been accepted towards fulfillment of the requirements for ' / flé «mammmfly Major professor Date éVWéé /f/y 0-7 639 MSU LIBRARIES .—:—. RETURNING MATERIALS: PIace in back drOp to remove this checkout from your record. FINES will be charged if book is returned after the date stamped beIow. CHANGES IN TISSUE AND BODY FLUID ACYLCARNITINES IN RESPONSE TO DIFFERENT PHYSIOLOGICAL STATES By Kim Joseph Valkner A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Biochemistry 1981 ABSTRACT CHANGES IN TISSUE AND BODY FLUID ACYLCARNITINES IN RESPONSE TO DIFFERENT PHYSIOLOGICAL STATES t By I Kim Joseph Valkner Experimentally induced anoxia and hypoxia caused a preferential loss of short-chain acylcarnitines from heart tissue. With hypoxic perfused pig hearts a 24% decrease in free carnitine plus a 38% decrease in short-chain acylcarnitine was found. The acylcarnitines of hypoxic heart showed a 65% decrease in acetylcarnitine while propionyl-, isobutyryl-, and isovaleryl-carnitine increased. Alloxan diabetes elevated the levels of free and short-chain acyl-carnitine approximently 100 fold in sheep liver. Each acyl residue of acylcarnitine increased by about the same magnitude in diabetic sheep livers. In studies with humans, fasting and feeding did not affect the total carnitine concentration or the free-lshort-chain acyl-carnitine ratio in blood. The major acylcarnitine of human serum is acetyl with propionyl, butyryl, and valeryl accounting for most of the remainder. when comparing urine samples to serum samples, the ratio of free-lshort-chain acyl—carnitine was lower, and the acylcarnitines contained a lower proportion of acetyl, propionyl, and butyryl residues while isobutyryl and isovaleryl proportions increased. The urine of a patient with a metabolic dysfunction contained high levels of propionylcarnitine. A Fanconi syndrome patient showed high levels of free- and short-chain acyl-carnitine in the urine while the blood carnitine levels were well below controls. To Mary ii ACKNOWLEDGEMENTS I would like to express my gratitude to my advisor, Loran L. Bieber, for his guidance and encouragement in this research work. Special thanks also to to my colleagues and friends, Pat Sabourin, Ron Emaus, Peter Clarke, Carol Fiol, and Shawn Farrell, not only for their help and ideas but also for their friendships which made this work possible and the many hours in the laboratory pleasurable. I also want to express my greatest appreciation to my beloved wife, Mary, for her understanding, devotion, and support. iii TABLE OF CONTENTS LIST OF TABLESooooooooooooooooooooooooooooooooooooooooooooooooooooooVii LIST OF FIGURESOOOOOOOOOOOOOOOOOOOOOOO0.0.0.00.00.00.000.0.00.00000001x LIST OF ABBREVIATIONS..0....0...0.0.0....OOOOOOOOOOOOOOOOOOOOOOOOOOOOOx LITERATURE REVIEW OF THE ROLE OF CARNITINE IN INTERMEDIARY METABOLISM Carnitine and Its Role in Fatty Acid Metabolism..................1 Carnitine Acyltransferases.......................................2 Role of Carnitine in Branch-Chain Amino Acid Metabolism..........3 Carnitine and Myocardium Tissue..................................4 Effect of Ischemia on Carnitine and Other Metabolites........4 Modulation of the Myocardium AcetleoA/CoASH Ratio by Carnitine..................................................5 Role of Carnitine in Controlling Ketogenesis in Ruminants........6 Carnitine Levels in Humans.......................................7 Carnitine Concentration in Humans............................7 Carnitine and Carnitine Palmitoyltransferase Deficiences.....8 carnitine AssaySOOCIOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO0.0.0.010 iv STATEMENT OF PROBLEM OOOOOOOOOOOOOOO OOOOOOOOOOOOOO ..... 00.0.000000000012 EXPERIMENTAL PROCEDURE...............................................14 Materials.......................................................14 Methods.........................................................14 Sample Acid Extraction Procedure............................14 Carnitine Analysis..........................................15 Quantitation of Short-Chain Acylcarnitines..................16 Identification of the Acyl Residues.........................19 Synthesis of Acylcarnitines.................................19 Bio-Gel Calibration.........................................20 Rat Heart Preparations......................................21 Perfusion of Pig Hearts ................ ..... ............. ...21 RESULTS........... ....... ............................................25 Heart Data......................................................25 Affect of Anoxia on Short-Chain Acylcarnitines of Rat Heart.....................................................25 Affect of Hypoxia on the Carnitine Content of Perfused Pig Hearts................................................27 Sheep Liver Data................................................35 Affect of Alloxan Diabetes on the Carnitine Content in Sheep Liver...............................................35 Carnitine Concentration in Normal and Alloxan Diabetic Sheep LiverSOO0..00......0....00.0.0.0...OOOOOOOOOOOOOOO00......35 Quantitation of the Short Chain Acylcarnitines..............36 Human Studies.................................. ....... . ....... ..40 Sample Collection and Subject Information...................4D Short-Chain Acylcarnitines in the Serum and Urine from Fasted and Fed Humans.....................................41 Acylcarnitine Determination of Urine from Patient I.........43 Acylcarnitine Determination of Urine from Patient II........44 DISCUSSION OOOOOOOOOOOO OOOOOOOOOOOOOOOIOOOOOOOOOOOOOOOOOO ........ 0.00.51 SUMMARYOOOOOOO0.00......O...OOOOOOOOOOOOOOOOOOO0.0.0.000000000000000059 LIST OF REFERENCESOOOOOOOOOOOOOOOOOOOOO...OOOOOOOOOOOOOO0.0.00.00000061 vi LIST OF TABLES Effect of anoxia on carnitine levels in rat hearts..............31 Short-chain acylcarnitines of normoxic and anoxic rat heart and inCUbation mediaOOOOOOOOOOOOOOOOOOOOOOOOO....0.0.0.0.000....32 Carnitine levels in hypoxic and control perfused pig hearts.....33 Effect of hypoxia on the short-chain acylcarnitine levels of....34 perfused pig hearts Carnitine levels in normal and diabetic sheep livers............38 Acylcarnitines of normal and diabetic sheep liver...............39 Carnitine concentration in human serum and urine of fasted......45 and fed adults Urinary acylcarnitines content of fasted and fed adults.........46 Serum acylcarnitine content in humans...........................47 vii 10 11 12 Proportion of individual acylcarnitines in human serum and......48 urine Urinary acylcarnitine content of patient I......................49 Urinary acylcarnitine content of patient II.....................50 viii LIST OF FIGURES Figure 1 Elution profile of ADP, [1-146] octanoyl-, [1-14C] acetyl-carnitine, and [3H-methyTJDL-carnitine fra“ the Bio-Ge] P-Z calmnCOOOOOOOOOOOOOO...0.00.00.00.0000000024 The effect of anoxia on the rate of release of carnitine and short-chain acylcarnitines from rat hearts......................30 Gas chromatogram profiles of the acyl residues of acylcarnitine from normal and diabetic sheep livers.............37 Relation of carnitine to CoASH in the matrix of mitochondria....58 ix ADP CAT CoA . CoASH CPT DTNB GC NADH NEM TCA LIST OF ABBREVIATIONS Adenosine Diphosphate Carnitine acetyltransferase Coenzyme A Reduced coenzyme A Carnitine palmitoyltransferase 5,5'-dithiobis—(2-nitrobenzoic acid) Gas chromatogram Reduced nicotine adenine dinucleotide N-ethylmaleimide Tricarboxylic acid LITERATURE REVIEW Carnitine and Its Role in Fatty Acid Metabolism Carnitine (the betaine of Y-amino-e-hydroxybutyric acid) was isolated in 1905 by two independent groups (1,2) from mammalian muscle. I For forty years its function in metabolism remained unknown. Interest in carnitine increased in 1935 due to its structural similarities to acetylcholine, but no evidence was found to support its physiological role as a neurotransmitter (3,4). In 1948 Fraenkel and Blewett found a compound that was an essential nutrient for growth of the Tenebrio “991139; meal worm larva (5). That compound, which_was later identified as carnitine (6), was initially named "vitamin 87“ (7). Using the Tenebrio assay method (8) carnitine was found to be widely distributed in tissues of animals, plants, and microorganisms, with the highest levels in the muscles of both vertebrates and invertebrates (9). The possible function of carnitine in fatty acid metabolism was first observed by Friedman and Fraenkel in 1955 (10). They isolated an enzyme from pigeon liver which catalyzed the transfer of acetyl groups to carnitine. In that same year Fritz (11) showed that a factor in rat liver particulate preparations caused an increase in the oxidation of long chain fatty acids and the factor that causes this increase was found to be carnitine (12). In 1962 both Fritz and Bremer (13,14,15) demonstrated that carnitine catalyzes the mitochondrial oxidation of long-chain acyl-CoAs, and both proposed that activated long-chain fatty 1 2 acids are transported across the inner membrane of mitochondria to the matrix, which is the site of B-oxidation, via the conversion of carnitine to palmitoyl-carnitine. The reversible reaction: palmitoyl-CoA + t—carnitine u=r=-palmitoyl-t-carnitine + CoA is catalyzed by the enzyme carnitine palmitoyltransferase (CPT). These two investigators (13,15) proposed the following role for carnitine in intermediary metabolism of fatty acids. The enzyme carnitine acyltransferase transfers acyl groups from activated fatty acids to carnitine forming acylcarnitines by the reversible reaction: acyl CoA + (in-carnitine e (Id-acylcarnitine + CoA these acylcarnitines can now cross the mitochondrial acyl CoA impermeable inner membrane (16), where the above reaction is reversed by an internal carnitine acyltransferase to regenerate acyl CoAs inside the mitochondrial matrix. These acyl CoAs are then used as substrates for B-oxidation (17). Carnitine Acyltransferases As stated above Bremer (15) and Fritz (13) both reported enzymatic synthesis of palmitoyl-carnitine by the enzyme carnitine palmitoyltransferase (CPT). Carnitine palmitoyltransferase (CPT) was located on both the outer surface of the inner mitochondrial membrane, and the matrix surface. The role of CPT proposed by Bremer and Fritz is given above. In 1963 Fritz et al. (18) partially purified a carnitine acetyl transferase (CAT) from pig heart mitochondria. Carnitine acetyltransferase (CAT) reacts only with short chain fatty acyl substrates whereas CPT reacts only with long chain fatty acyls, therefore they must be 2 different proteins. The proposed role of CAT, 3 which is also located on both sides of the mitochondrial inner membrane (102,103), is to facilitate the transport of activated acetyl groups across the inner membrane. Thus CPT in mitochondria can be used to provide substrates for fatty acid oxidation while CAT provides acetyl CoA for the TCA cycle which is also in the matrix of mitochondria. If carnitine and the 2 carnitine acyltransferases (CAT and CPT) only function in metabolism to facilitate the transport of fatty acids to the mitochondrial matrix then the transferases should associate exclusively with mitochondria. Markwell et al. (19) demonstrated that both short-chain and mediumrchain carnitine acyltransferase activities are associated with microsomes, peroxisomes, and mitochondria in pig and rat liver. Markwell et al. (20) later partially purified carnitine acetyltransferase proteins from peroxisomes and microsomes, however no carnitine palmitoyltransferase was found in these organelles. Thus the role of carnitine and CPT in long-chain fatty acid metabolism is solely associated with mitochondrial B-oxidation. Carnitine and CAT system role in transfering acyl residues to other organelles and its role in other forms of intermediary metabolism must be more diverse (34). . Role of Carnitine in Branched-Chain Amino Acid Metabolism In 1970 Solberg and Bremmer (21) showed that mouse and rat tissues produce [3HJ-labeled branched-chain acylcarnitines from branched-chain ketoacids and [methyl-3HJ-z-carnitine. Bieber et al. (22-27) have demonstrated the presence of branch-chain acylcarnitines and also branched-chain carnitine acyltransferase activities in both rat and beef tissues demonstrating a possible involvement of carnitine in amino acid metabolism. Addition of carnitine stimulates the oxidation of 4 branched-chain amino acids in both rat skeletal muscle and heart (28) but not in liver (29). Carnitine also stimulates the oxidation of branched-chain ketoacids in rat liver, heart kidney and muscle (30,31). The stimulation of branched-chain ketoacid dehydrogenase activity by carnitine results from removing the inhibitory levels of the branched- chain acyl-CoAs (32) by forming the branched-chain acylcarnitine via carnitine branched-chain acyltransferase activity (30,31,33). Therefore, carnitine stimulates branched-chain amino acid metabolism by forming acylcarnitines from the branched-chain acyl-CoA groups. This will increase the CoASH/acyl CoASH ratio inside the mitochondria and this should activate the branched-chain keto acid dehydrogenase (34). Carnitine and Myocardium Tissue Effect of Ischemia on Carnitine and Other Metabolites Investigations by others have demonstrated that induction of myocardial ischemia or anoxia causes a decrease in tissue levels of both free carnitine and acetylcarnitine and increases the amounts of some other acylcarnitines (35,36,50,51). The extent of these changes is -somewhat dependent on the time of and severity of oxygen deficiency. Long-chain acylcoenzyme A esters also accumulate during ischemia (37) and it has been proposed that such esters can reduce mitochondrial energy production (35,37,38,39). Evidence has also been presented that carnitine protects ischemic myocardium (40). Such studies indicate that the levels of carnitine and acylcarnitines are affected by hypoxia and anoxia and that changes in the levels can affect heart viability. It is well established that carnitine has an obligatory role in the oxidation of long-chain fatty acids in both normoxic and ischemic conditions 5 (41,42). Studies with humans have shown that tissue deficiencies of carnitine or carnitine palmitoyltransferase can cause fatal cardiomyopathies (43,44). Another possible function of carnitine in the heart besides acting as a transport system for fatty acids into the mitochondria is discussed in the next section. Modulation of thegflyocardium Acetyl CoA/CoASH Ratio by Carnitine Pearson and Tubbs (45) first suggested that carnitine may be involved in regulating the acetyl CoA/CoASH ratio, with acetylcarnitine acting as an acetyl CoA buffer. This was supported by the finding that the ratio of acetylcarnitine to acetyl CoA is high in heart (>20) and by the observation that the reactants of the carnitine acetyltransferase system remain near equilibriun during different metabolic states (46). Neely et al. (46-48) suggested the following function of carnitine and carnitine acetyltransferase in regulating the ratio. Acetyl CoA produced in the matrix of mitochondria, for example by B-oxidation, can be either oxidized or transported out of the matrix by forming acetylcarnitine via carnitine acetyltransferase. The acetylcarnitine can then be converted back to acetyl CoA which can be used as a substrate in other parts of the cell. Therefore carnitine and carnitine acetyltransferase provide a way to remove “excess" intramitochondrial acetyl CoA which is generated during ”excess" fatty acid uptake and oxidation. The regeneration of acetyl CoA outside of the matrix can then inhibit fatty acid activation by lowering the amount of extramitochondrial CoASH (49). This process in reversed during the start of work with an increase in citrate synthetase which lowers the levels of intramitochondrial acetyl CoA. This causes carnitine 6 acetyltransferase to form acetylcarnitine which is transported to the matrix to replace acetyl CoA. Formation of acetylcarnitine will also stimulate fatty acid activation via the regeneration of extra- mitochondrial CoASH (48). Several groups have shown that excess acetyl CoA in isolated heart mitochondria can also inhibit the myocardial nucleotide translocase system (35-38) and Pande and Blanchaer (35) found that the addition of carnitine could reverse this acetyl CoA induced inhibition. Excess acetyl CoA can affect many other sites of metabolism inside the mitochondria, thus carnitine with carnitine acetyltransferase may provide a pathway for removing excess acetyl residues generated inside the mitochondrial matrix. Role of Carnitine in Controlling Ketogenesis in Ruminants The ratio of free CoA to acetyl-CoA is a major factor in control of hepatic ketogenesis. This ratio in normal sheep and cow liver ranges from 0.2 to 1.0 (52,53), which is much lower than the ratio observed in ketotic rats and guinea pigs (range of 2:1 to 9:1). In starved and alloxan-diabetic sheep livers, Snoswell (54) observed a 4-7 fold increase in carnitine compared to normals, whereas the concentration of carnitine remained constant in similarly treated rat livers. Snoswell (55) stated that these differences between sheep and rat livers are due to carnitine and the carnitine acetyltransferase system exerting an "acetyl buffer" function in sheep liver. The rate that ketone body concentration increased in hepatic venous blood of sheep was found to be inversely related to carnitine concentration following alloxan treatment. Thus carnitine and carnitine acetyltransferase can transfer 7 acetyl groups from the matrix of mitochondria, where ketogenesis occurs, to the cytoplasm. This will remove acetyl CoA as a substrate for ketogenesis, and provides a mechanism for regulating hepatic ketogenesis. Carnitine in Humans Carnitine Concentrations In Humans In 1978, Mitchell (56,57) reviewed and tabulated the published levels of carnitine in various body tissues and fluids of normal and diseased humans. These articles provide an excellent source for obtain- ing normal total carnitine levels in human, although Mitchell noted it is difficult to obtain accurate normal values. This is due to several factors: 1) No criterion for age, sex, diet was used by most investiga- tors. 2) No standardized method was used for determining the carnitine concentration. Some of the methods used are described in the carnitine assay section. 3) Many of the carnitine determinations of normal humans tissues were samples from patients who were hospitalized for diseases other than the symptom under study. More recent papers on the total carnitine concentration in human serum and the amount of carnitine in urine extraction have tried to address some of these factors. Tanphaichitr et al. (58) measured plasma and urinary carnitine concen- trations in Thai adults living in either a city or village. The village dwellers had lower total carnitine concentrations in both plasma and urine which they attributed to a lower carnitine intake with their diet. So diet may have some affects on carnitine levels. When investigating other factors that may affect carnitine concentrations in humans there are many reports with conflicting data. Carrier and Berthillier (59) showed that the amount of total carnitine excreted in urine increases with age, while the serum levels of carnitine remain constant in both children and adults. Previously, Cederblad (60) reported that there are significant differences in plasma carnitine concentrations between the sexes and various age groups. Another example of conflicting values is found in carnitine levels during fasting. Maebashi et al. (61) reported an increase in the amount of urinary carnitine excreted during a 6 day fast, while Frohlich et al. (62) showed a decrease in serum and urinary free carnitine levels, but acetylcarnitine increased during a 24-36 hour fast. Recently Hoppel and Genuth (63) measured carnitine levels of normal and obese humans during a fast. There was a rapid increase in the amount of short- and long-chain acylcarnitine in plasma with a delayed decrease in free carnitine observed in both groups. Urinary short-chain acylcarnitines increased parallel to plasma levels while free carnitine levels decreased and then increased slightly. Even though there are conflicting results, the human provides one of the best models for observing other functions for carnitine because metabolic dysfunctions due to carnitine deficiency or CPT deficiency have been reported. Some of these cases are listed below. Carnitine and Carnitine Palmitoyltransferase Deficiencies As mentioned earlier, Mitchell (56,57) reviewed the data of some of the cases reported prior to 1978. Briefly, in 1973 Engel and Angelini (64) first described a lipid storage myopathy which was related to a deficiency of serum and muscular carnitine. Since then several cases have been reported (65-74). These studies showed that at least two types of carnitine deficiency syndromes exist, which can be classified 9 as predominantly myopathic or systemic (104). The systemic syndrome is characterized by a lower than normal content of carnitine in the blood and, presumably, other organs. The other syndrome is a myopathy where serum concentrations are normal but carnitine transport across cell membrane in various tissues such as skeletal muscle, cardiac muscle, Schwann's cell, and leukocytes is impaired, resulting in decreased carnitine content in these tissues. Possible causes of these defficiences included impaired carnitine biosynthesis; excessive carnitine loss from body fluids; impaired active transport of carnitine into cells; excessive release of carnitine from cells, and excess catabolism (104). Carroll et al. (75) have reported a patient with carnitine deficiency that showed biochemical features of both types of syndromes. Glasgow et al. (76) have reported a case where systemic carnitine deficiency stimulates recurrent Reye's like syndrome. Chapoy et al. (77) reported that systemic carnitine deficiency is a cause of recurrent Reye's-like syndrome, treatable by administering oral carnitine. Pola et al. (78) gave oral carnitine in therapy to hyperlipidemic patients .which lowered their serum cholesterol and triglyceride levels. The first report of carnitine palmitoyltransferase (CPT) deficiency was described by DiMauro and DiMauro (79), in a patient with recurrent paroxysmal myoglobinuria (a classic sign of CPT deficiency) where CPT activity in muscles was < 20% of normal levels. This was apparently due to a lack of the inner CPT activity. Hostetler et al. (80) reported a case of no CPT outer, but nonmal CPT inner in muscle mitochondria. Two other cases (81,82) of decreased muscle CPT inner activity have been reported. Layzer et al. (83) reported a case where both CPT inner and 10 outer activities were deficient. DiDonato et al. (84) showed that cultured fibroblasts from a patient deficient in muscle CPT also contained reduced CPT activity and this supports the concept that CPT deficiency may be a systemic rather than just a muscular condition. Ionasescu et al. (85) have linked a combined partial deficiency of muscle carnitine and CPT activity with an autosomal dominant inheritance. These cases demonstrate the various malfunctions that can occur if carnitine content in blood and tissues are lower than normal, and they provide information about other functions of carnitine. Carnitine Assays The first assay technique for determining the amount of carnitine used was the Tenebrio molitor bioassay (9). Growth of larvae consuming a diet plus the test material was compared to control larvae growing on the diet plus a known amount of carnitine. The problems of all bioassays are inherent in this method, which would affect the determination. Friedman (86) developed a chemical method which involves esterification of carnitine and colorimetric determination, however the -determination was non-specific due to interfering quaternary amines. Marquis and Fritz (87) developed an enzymatic assay using carnitine acetyl transferase: acetyl CoA + carnitine ‘:4=- acetyl-z-carnitine + CoASH The CoASH produced is measured spectrophotometrically by reaction with DTNB. Pearson and Tubbes (45) coupled this reaction to another enzyme reaction and measured citrate and NADH. In 1972 Cederblad and Lindstedt (88) increased the sensitivity down to 20-30 picomoles by using [1-14CJ-acetyl CoA and excess acetyl CoA. 11 Other methods developed recently, include a microbiological method using Torulopsis bovina yeast (89) and a gas chromatographic method (90). Three papers have been published (91,92,93) which have refined the Cederblad and Lindstedt method. These three appear to be the most sensitive methods because they overcome the problem of nonlinearity by using a compound to trap CoASH formed instead of using excess acetyl-CoA, which makes the assay very expensive. These three methods it also use CoASH trapping agents which prevent reversibility of the reaction and which do not inhibit the carnitine acetyltransferase enzyme. | STATEMENT OF PROBLEM As stated earlier, the possibility that carnitine has other roles in intermediary metabolism, besides shuttling activated long-chain fatty acids into the matrix of mitochondria for B-oxidation has been proposed. One role proposed by several investigators is that carnitine and the carnitine acetyltransferase enzyme of the mitochondria modulate the acetyl CoA/CoASH ratio in the matrix of mitochondria by shuttling excess acetyl residues back out of the matrix in the form of acetyl carnitine (42,45,47,54). Recently Bieber, et al. (22.24.34) have suggested that the role of carnitine and the carnitine acyltransferases in regulating the ratio should be expanded to include other short-chain and medium-chain acyl CoAs instead of just acetyl CoA. This would include branched-chain acyl residues, which arise during oxidative degradation of some amino acids. This suggestion correlates with Van Hinsberghs (30) proposed role of carnitine in removing excess branched-chain acyl CoAs which inhibit the keto acid dehydrogenase and regenerates CoASH. Reports that there are both short-chain and medium-chain carnitine acyl- transferase activities associated with three organelles, while the long- chain carnitine acyltransferase has so far been found only present in mitochondria, suggest that carnitine may be involved in shuttling short- chain and medium-chain acylcarnitines from one organelle to another. The goal of my thesis project was to find if quantification of acylcarnitines provide information about the levels of the products 12 13 formed and the substrates used by carnitine acyltransferases during various metabolic states. This was investigated by determining changes in the amounts and/or proportion of each specific acyl residue, of acyl carnitine from various tissues, in response to different physiological states. These short-chain acylcarnitines should be the products of the carnitine acyltransferases, which generate the acylcarnitine to maintain the equilibrium with the substrates; acyl CoA. Changes in the proportion of the acyl residues of acylcarnitine may provide some insight of the metabolic pathways that are occurring in the tissue and to the role of carnitine in regulating the acyl CoA/CoASH ratio. Normal and hypoxic myocardium tissue, normal and alloxan diabetic sheep livers, and also human serum and urine were investigated for carnitine levels and the composition of their water-soluble acyl carnitines. MATERIALS AND METHODS Materials Carnitine was a generous gift from 0tsuka Pharmaceutical Co. Acetyl-CoA, [1-14C] acetyl CoA were from P-L Biochemicals. DL-(methyl-3HJ-carnitine was from Amersham. The Bio-Gel P-2 and Dowex resins were from Bio-Rad. NEM (N-ethylmaleimide), ADP, carnitine acetyltransferase and other compounds were from Sigma. The GC packing material was from Supelco and the fatty acids were from J.T. Baker. Methods Sample Acid Extraction Procedure Pig and rat hearts: The frozen tissues were homogenized in 6% HCl04 (1:6 wt/v) at 4°C. The homogenate was centrifuged at 12,000 x g for 15 minutes and the supernatant fluid collected. The volume of the fluid was recorded, an aliquot was removed for carnitine analysis, and the remaining vol we was used for quantitation of the acylcarnitines. Rat Heart Media: The same procedure was used as above except the ratio of HCl04 to media was 1:3 (v/v). Human serum: This also used the same procedure except that the ratio of HClD4 to the sample was 1:2 (v/v). 14 15 Human Urine: The urine was initially concentrated 10 fold before adding 6% HClO4 at a ratio of 1:2 (v/v). The supernatant fluid was adjusted to pH 6.0 and stored at -80°. Carnitine Analysis The method used for carnitine determination is a modified procedure of Cederblad-Lindstedt (88), Parvin-Pande (91) and Bieber-Lewin (94). The assays are based on the reaction: t—carnitine + acetyl CoAI‘=-=>-acetyl-t-carnitine + CoASH The method was used for quantitation of the amounts of free carnitine and water soluble 0-acylcarnitines. Aliquots from the perchloric acid extracts, as described in the "Sample Acid Extraction Procedure“ section, were neutralized to pH 6.6 with KOH and the precipitated KClO4 was removed by centrifugation at 12,000 x g for 10 minutes. The supernatant fluid was used to determine the amount of free carnitine. 2N KOH was used to bring an aliquot of the pH 6.6 supernatant to 0.2N KOH. These samples were heated to 40°C for 30 minutes to saponify the acyl residues of the acylcarnitines. The .fluid was again neutralized to pH 6.6 with HCl04 and centrifuged as previously. Aliquots from each of the two preparations were assayed for carnitine, the former gives the amount of free carnitine and the latter gives the total carnitine (free carnitine plus water soluble short-chain acylcarnitines). The reaction solution for assaying the carnitine contains in a 0.2 mL volume: 20 pmoles potassium phOSphate buffer at pH 7.6, 288 pmoles (1-14C) acetyl coenzyme A (0.0167 ucr), so ug NEM, soo pmoles acetyl-CoA, and the sample. The range of the assay was 10-200 16 pmoles of t-carnitine. The reaction was initiated by adding 25 ml of carnitine acetyltransferase (0.400 units/assay) and incubated for 30 minutes at 25°C. The reaction was stopped by loading 200 pl onto a 2 x 25 mm Dowex-1 X8 Cl' form (100-200 mesh) column. The effluent and a 1 mL water wash of the col mm were collected in a scintillation vial to which 10 ml of scintillation cocktail (4 g of 2,5-diphenyloxazole, 100mg of 1,4-bis [2-(4-methyl-5-phenyloxazolyl)] benzene in 1 liter of toluene and 1 liter of Triton X-100) was added. The amount of t-carnitine per sample was determined by comparing the amount of [1-14C] acetylcarnitine formed during the reaction to the amount formed in an experimentally determined carnitine standard curve. Quantitation of Short-Chain Acylcarnitines The method described below provides a method for identifying and quantitating the short-chain acylcarnitines, and removing all other short-chain acyl residues which may be present in the sample. The method used to separate and quantitate the water soluble acylcarnitines is essentially that described previously (94,95) with several modifications that will be elaborated in the multistep procedure described below. Procedure: 1) The perchloric acid extract, from the “Sample acid extraction procedure" section, is mixed with an internal standard of either crotonylcarnitine or valerylcarnitine (20-30 ug). Valerylcarnitine was the preferred standard because the crotonylcarnitine is unstable in the aqueous phase over an extended period of time, but was used in samples where valerylcarnitine was previously detected. 17 2) The solution was neutralized to pH 6.6 with 2N KOH and the neutralized solution was placed on ice for 30 minutes to precipitate the KCl04. The sample was then centrifuged at 12,000 x g for 15 minutes. The supernatant was decanted and saved. The pellet was resuspended in 15-20 mL of HClO4 and was recentrifuged as above. The supernatant fluid was combined with the previous supernatant fluid. 3) This step was used only for human urine samples and entails a 1:2 dilution of the supernatant from the previous step with 95% ethanol. The solution was mixed and then set on ice for 30 minutes before centrifugation at 12,000 x g for 15 minutes. The supernatant fluid was collected and the pellet was washed with absolute ethanol and was recentrifuged as above. This supernatant fluid was combined with the first supernatant fluid. This step helps to remove some of the salts. 4) The combined fluids from either step 3 or’4 were concentrated to about 10 mL using a vacuum rotary evaporator. Several drops of octanol were added to urine extracts to prevent foaming during evaporation. Nine volumes of isopropanol were then added to the remaining fluid and it was thoroughly mixed. After standing on ice for 30 minutes the samples were centrifuged at 12,000 x g for 20 minutes and the supernatant was collected. It was then concentrated to approximately 3 mL, being sure to remove all ethanol and isopropanol. 5) To the sample were added a 400 molecular weight marker, 10 ul of 0.2 M ADP, and 0.5 ml of the elution buffer, 0.1 mM KH2P04 at pH 6.6. The sample was then loaded onto a Bio-Gel P-2 column. This 2.5 cm X 45 cm column was used to separate low molecular weight and high molecular weight molecules from carnitine and the water soluble 18 acylcarnitines. The calibration of this column is described in the Bio-Gel Calibration section. The sample was eluted with the phosphate buffer and the 400-200 molecular weight effluent was collected. 6) The effluent was passed through a 2.5 cm X 13 cm Dowex 1-X8 HC03 form (100-200 mesh) column. The total effluent and the 1.5 bed volune water wash of the col min were collected. The effluent may be alkaline after passing through the column, so 1.0 ml of 1N HCl was added to the collecting flask prior to the elution. After all of the elutate was collected the pH was adjusted to 6.0, and the volume of the effluent was concentrated to 10 ml with the rotary evaporator. 7) The sample was adjusted to pH 2.0 with 1N HCl and was loaded onto a Dowex 50-X8 H form (100-200) 1.0 cm X 12 cm column. The effluent was discarded and the column was washed with 2 bed volumes of water. The acylcarnitines were eluted with 1.0 N NH40H: 95% ethanol (8:2 v/v), and the effluent was collected when the pH became alkaline. Three bed volumes of the effluent, approximately 40 ml, were collected and 2.0 ml of 2N KOH was added, mixed, and incubated for 30 minutes at 40°C to saponify the 0-acyl derivatives of carnitine. 8) The sample was concentrated to 2 mL using the rotary evaporator and the pH adjusted to 8.5 with 1N H2504. The sample was then completely dried either by evaporation or by lyophilization. The evaporation procedure was used for samples that were extracted the same day and the lyophilization procedure was used if the sample was to be stored and extracted at a latter time. 9) The dry K+ salts of the volatile fatty acids were resuspended and converted to the free acids by adding a small volume of 25% meta- phosphoric acid (250 g/l of 36% HP03). The volume of acid used to 19 resuspend the fatty acids depended on the initial concentration of acylcarnitines present in the samples. The minimum range required for reproducible results was 3-5 nmoles of fatty acids/pl. Any excess salts in the resuspension were removed by centrifugation at 500 x g for 5 minutes to prevent loading of excess non-volatiles on to GC column packing material. Identification of the Agyl Residues Gas chromatography was the method used for the quantitation and identification of short-chain fatty acid residues that resulted from saponification of the acylcarnitines. The fatty acids were separated on a 2mm i.d. x 6 foot GC column, packed with 15% SP-1220 1% H3P04 on 100/120 Chromosorb W AW (Supelco). Due to nonvolatile compounds in the sample preparation a short 6 inch pre-column packed with phosphoric acid treated glass wool was used to collect these components before they contaminated the analytical column. The temperature program for the separation of the fatty acids started at 115° for 5 minutes and then increased to 140°C at a rate of 4°/minute. To calculate the amount of ‘each acyl residue present in a sample, a separate standard solution was prepared which contained a known amount of each fatty acid. The response of the amount of fatty acid/unit of area detected by the flame ionization detector for the standard was then compared to the area obtained for each fatty acid in the sample. Synthesis of the Acylcarnitines Crotonyl-, valeryl-, and octanoyl-carnitine were synthesized as described by Bohmer and Bremer (96). [1-14C] acetylcarnitine was 20 synthesized enzymatically with [1-14C] acetate, CoASH, acetyl CoA synthetase enzyme, t-carnitine, NEM, and carnitine acetyltransferase. The initial reaction volume of 1.0 mL contained 100 mM sodium phosphate buffer, pH 7.6; 1 mg acetyl CoA sythetase; 35 nM ATP; 0.35 nM CoASH; 13,5 mM MgCl2; 0.35 mM [1-14C] Na acetate; 0.35 mM z-carnitine. These components were mixed and stood at room temperature for 3 hours before adding 4.3 mg NEM and 2 units of carnitine acetyltransferase. After 1 hour at room temperature the reaction was tenminated by placing in a boiling water bath for 3 minutes. Denatured protein was removed by centrifugation. [1-14C1Acetyl-t-carnitine was separated from [1-14C] acetate or [1-14C] acetyl CoA by passing the solution through a 5 x 35 um col mm of Dowex 1-X8 (chloride form, 100-20 mesh). The effluent was tested for any contaminating acetate and the remaining effluent was adjusted to pH 6.0 and was stored at -80°. Bio-Gel Calibration The P-2 Bio-gel column was calibrated for the molecular weight range of 400-200. The effluent in this range would contain all of the water soluble short-chain acylcarnitines, and would exclude larger and smaller molecular weight materials. ADP was used as the 400 molecular weight marker in each sample because the elution pattern is similar to that of the largest water soluble acylcarnitines; [1-14c] octanoylcarnitine (95). [14C]Acylcarnitine was used as the lower limit marker. The elution profile is shown in Fig 1. The UV absorbance at 254 nm was monitored until ADP began to elute and 1.8 ml fractions were then collected and 0.1 ml aliquots were counted for the 21 radiolabels. The volume required to elute ADP to the last of the [14CJ-acetylcarnitine was then used for all of the samples. The elimination of low molecular weight cations and anions would prevent these components from interfering with remaining ion-exchange columns. Rat Heart Preparations Prior to sacrificing, 10 mL of modified Krebs/Henseleit bicarbonate buffer containing five millimolar a-hydroxybutyrate was prepared for each heart by equilibrating with either the control gas mixture 95% air:5% C02 or the anoxic gas mixture 95% N2:5% C02 in a 50 mL Erlenmeyer flask fitted with an airtight neoprene cap. Two size 20 hypodermic needles were inserted in the cap and the longer one extended into the buffer through which the buffer was flushed with gas. Hearts from either fasted or fed 200 9 male Sprague/Dawley strain of rats were removed rapidly, split and immersed in the incubation buffer. They were incubated in the presence or in the absence of s-hydroxybutyric acid for 30 minutes at 30°C with the gas mixture continuously bubbling into the buffer. The incubation was stopped by freeze clamping the heart tissue and lyophilizing the buffer. Perfusion of Piggflearts This procedure was performed by S. Ely and G. Scott of the Physiology Department at Michigan State University. Poland-China pigs (30-40 kg) were anesthetized with sodium thiamylal (5 mg/lb, i.v.) and maintained with nitrous oxide and supplemental doses of sodium thiamylal. The animals were incubated and ventilated by a positive 22 pressure respirator (Harvard model 613) on room air with supplemental oxygen, at a 5 cm H20 end expiratory pressure. Volume and rate of ventilation were adjusted to maintain arterial P02, Pco2 and pH within the normal physiological range. Blood anticoagulation was achieved by the intravenous administration of sodium heparin (600 U/kg plus 250 U/kg per hour, i.v.). Esophageal temperature was monitored (Yellow Springs) and maintained at 37°C with a heating pad. All blood pressures were continuously monitored with Statham (low volume displacement) pressure transducers and recorded via inputs into a direct writing oscillograph (Hewlett-Packard). Lead II of the ECG was monitored for determination of heart rate and detection of arrhythmias. The left femoral artery and vein were cannulated (PE 240) for monitoring arterial pressure and intravenous infusions, respectively. The chest was opened by median sternotomy and the pericardium incised and sutured to the inside of the chest wall to form a cradle. An extracorporeal coronary perfusion system was constructed by withdrawing venous blood from the cannulated femoral vein and pumping it (Masterflex roller pump) into the pulmonary artery of an isolated lung obtained from a small pig. Pulmonary venous blood flowed into a large bore cannula tied into the preserved left atrium, passed through a water bath heated at 39°C and was delivered (Holter roller pump) at constant pressure (100 mm Hg) to the cannulated right coronary artery. Pulmonary venous pressure in the isolated lung was monitored from a catheter in the left atrium and maintained at 5 mm Hg with the use of a feedback controller system (Leeds-Northrup) which controlled the speed of the pulmonary arterial inflow pump. The extracorporeal perfusion circuit was primed with cross-matched blood obtained from the donor lung animal. Coronary 23 perfusion pressure was monitored from the coronary cannula just proximal (3 cm) to its entry into the vessel, and held constant by a second feedback control system. Coronary blood flow was determined from the coronary pump speed which was continuously recorded on an oscillograph. The cyclic variations seen in the coronary blood flow recordings represent oscillations in the servo-feedback control system. At the end of each experiment, the pump was calibrated by timed collections of blood and there was a linear relationship of pump speed to pump flow over the range of flows encountered. The lack of interarterial pump interarterial coronary anastamoses is well documented in the pig (Circulation 1:10-27, 1950; Circulation 27:717-721, 1963). Thus, it is safe to conclude that changes in the gas content of the blood perfusing the right ventricle would not affect gas tensions in the left ventricle. Ventilation of the isolated perfused lung with various gas mixtures produced changes in blood gas tensions of the right coronary perfusate without affecting systemic gas tensions. Normoxia control was produced with 20% 02, 5% C02, 75% N2. Hypoxia was produced by ventilating the isolated lung with 5% C02, 95% N2. Blood gases were measured on a radiometer blood gas analysis (Radiometer-Copenhagen). Figure 1. 24 ADP3 5- . Q‘A 04 8 J E? X4- _l A E ‘Q’ “ 2 o. O 2- . 0 3O 40 50 Fraction Number Elutlon profile of ADP, [1-14C octanoyl- [1-1C]acetyl-carnitine, and [H {Tsthle Dl-carnitine from the Bio-Gel P- 2 1iolumn.[ C]0ctanoylcarni£ine (2. 0 x 104 cpm), [1-1 C] acetylcarnitine (2.5 x 10 cpm), [3H- methyl] DL-carnitine (2.5 x 104 cpm), and 0.01 mL of 0.2M ADP were combined with 3 mL of the 0.1 mM P04 elution buffer. This was applied to the Bio-Gel W? column and the eluate was collected in 1. 7 mL fractions, and 0.1 mL aliquots were counted for radioactivity. ADP was monitored continuously by measuring the absorbance at 254 nm. RESULTS As described in the introduction myocardium anoxia causes a change in the levels of tissue carnitine. We performed two types of experiments with hearts to test the effect of anoxia and hypoxia on the retention of and changes in the levels of carnitine and its derivatives. The media was analyzed to determine if heart loses short-chain acylcarnitines and to determine if specific acyl derivatives are lost or whether the loss reflects the general short-chain acylcarnitine content of the cardiac tissue. Affect of Anoxia on Short-Chain Acylcarnitines of Rat Heart. A series of experiments were performed in which hearts from fed and 48 hour fasted rats were incubated in the presence of air/C02 or in the presence of N2/C02 with and without a-hydroxybutyric acid. The -results of these experiments are shown in Table I. 1) Acylcarnitines and carnitine present in the buffer were derived solely from the heart tissue. Fasting decreased the ratio of free carnitine to short-chain acylcarnitines in both heart and incubation media. 2) After 30 minutes the incubation buffer contained a lower ratio of free carnitine to short-chain acylcarnitines than the tissue. Thus, rat hearts incubated with either normoxic conditions and or anoxic conditions appear to preferentially lose short-chain acylcarnitines or selectively retain free carnitine. 3) Within the experimental conditions used, there was 25 26 little difference between the loss of carnitine and acylcarnitines from normoxic tissue and anoxic tissue. It should be noted that in these ex- periments the hearts were not perfused, but rather were rapidly removed from the animal, split open and incubated in the bicarbonate buffer. Since the loss of short-chain acylcarnitines plus free carnitine was less than 10% of the total carnitine content of heart and our anoxic conditions did not increase the amount of acylcarnitines lost, the time course for loss of free carnitine and short-chain acylcarnitines was done to determine if the loss of carnitine was a continuous process. The results of such an experiment are shown in Figure 2. The data show that the release of both free carnitine and short-chain acylcarnitines was linear for 30 minutes and, as with the data in Table I, anoxia did not enhance the release of carnitine or acylcarnitines at the levels reported by Shug gt al. (35,36,50,51). The composition of the individual short-chain acylcarnitines from normoxic and anoxic experiments was determined because the rat heart appeared to selectively release short-chain acylcarnitines into the buffer. In addition, fasted animals released proportionately larger ‘amounts of short-chain acylcarnitines. These data are shown in Table II for the tissue and buffer from 10 combined individual experiments. The samples were combined because inordinately large amounts of time are required for doing each individual acylcarnitine analysis and because of the low concentration of acylcarnitines in individual samples. The data show that with both control (normoxic) and anoxic tissue, the short-chain acylcarnitines lost to the buffer reflect the short-chain acylcarnitine content of the tissue. However, the relative distribution of specific acylcarnitines was different even though the major short 27 chain acylcarnitine was acetylcarnitine in all samples. During normoxic conditions, the heart and buffer contained relatively large amounts of propionyl- and butyryl-carnitine, both of which might be derived from s-oxidation of long-chain fatty acids. In contrast under anoxic conditions, the amount of propionyl carnitine in both tissue and media decreased and the isobutyryl- and caproyl-carnitine amount increased. Affect of Hypoxia on the Carnitine Concentration of Perfused P1g_ Hearts- Although the studies with rat hearts show that the hearts selectively lost short-chain acylcarnitines under both normoxic and anoxic conditions, anoxia did not appear to enhance the loss of carnitine or short-chain acylcarnitines. These were non-perfused samples and the total carnitine loss was relatively small, less than 10% in 30 minutes. Thus another heart model was tested. This model was in .sitg perfused pig heart in which the left ventricle served as the control and the right ventricle was the hypoxic treated sample. In these experiments, the right ventricle was made hypoxic for 10 minutes by ventilating the isolated lung with a gas mixture containing 5%, 95% N2. This ventilation mixture dropped the P02 in the coronary perfusate from 144 to 12 mm Hg with little change in pH (7.27-7.34). After sacrificing the animal, the tissue samples were removed and freeze clamped and subsequently analyzed for acylcarnitines. Representative samples were also taken from non-perfused pig hearts, both the left and right ventricle, to serve as controls for the free carnitine and short-chain acylcarnitine analyses. The carnitine content of the blood before and after passage through the heart was not analyzed because of 28 the difficulty in accurately measuring the small changes in large fluid vol unes. The results of these experiments are summarized in Table III and they show that perfusion caused a reduction of or loss of short-chain acylcarnitines compared to non-perfused ventricles. There was a 37% de- crease in short-chain acylcarnitines and a 24% decrease in free carni- tine; compare the non-perfused heart values with the left ventricle values for the perfused heart [note the left ventricles were perfused with normoxic conditions]. When the heart was perfused with hypoxic conditions, the loss of carnitine, especially short-chain acylcarni- tines, was greater than when the tissues were perfused under normoxic conditions. Compare 173 nmoles/g short-chain acylcarnitine in the left ventricle and 108 nmoles/g in the right ventricle. The value of short- chain acylcarnitines in non-perfused heart was 268 t 43 nmoles/g, this was reduced to 108 t 9 nmoles/g by hypoxic perfusion, a 60% reduction in short-chain acylcarnitines. Similar, but not as large of a reduction in free carnitine was found. In contrast, a small increase in the total amount of long-chain acylcarnitine was obtained by hypoxic perfusion. Since the heart lost large amounts of short-chain acylcarnitines during hypoxic perfusion, the short-chain acylcarnitines of perfused control ventricles, left ventricle, and perfused hypoxic ventricles, right ventricle, were determined. These data are given in Table IV and show that hypoxia caused a major decrease in acetylcarnitine (approximately 65%) and an increase in propionyl-,isobutyryl-, butyryl- and valeryl-carnitine. The large decrease in acetylcarnitine could indicate a decrease in B—oxidation due to hypoxia or a selective loss of acetylcarnitine to the blood, or both. Figure 2. 29 The effect of anoxia on the rate of release of carnitine and short-chain acylcarnitines from rat hearts. The incubations contained 5 mM s-hydroxybutyric acid. Hearts from fed animals were used and analyses were performed as described for Table I. *Free carnitine (air-CO;). irFree carnitine (NZ-C02 .Free + Short-chain acylcarnitines (air-C0§). (bFree + Short-chain acylcarnitines (NZ-C02 SBlnNIW 30' NMOLES CARNITINE IN MEDIA/G TISSUE Oh (I) O 31 rati5l; .. u . O. .saummm Amouuwzv u.xo=o oz» a» soumxw mouvc*uv .ocucou mg» m=_L~QEOu ma ~o.ov o=Fc> a umou u PFou orb. .acoe.c xaxuuc .vouao_v:. mews: tom: no: ouueausnxxocuzgiu :5 m "Layman as» ou:. map—ans; aaozceucoo an oc:ux.2 mom ago—cqocnno as» :9.3 ohm an mouscve on Lou vuuonsucp was u__nm ago: macaw: on» om.o m.H « m ~. H a a. AN. we.“ "N « can se « osm . «ou-nz -.o m.~ A o a. a u H. an o¢.~ mm a man gm « mom i u E u mm.o ~.¢ « m N. n w w. on. o~.~ fin « owe Hm A mom + u He.c 5.: a m N. m u e. an o~.~ ow « mac m~ « “mm + wouue_o vogue» Ego; we mm.~ ~.~ « o m.¢ « n.ow¢ ~m.~ Nu « mmw sc « Hos + Noeluz -.— ~.~ « N ~.n « ~.~¢ mm.~ m_ « flaw cm a who + wou-ewa mung voumouicoz enemas c. mac—u.ccou_>u< uc—u_ceau “new: :— mocpuvccaupxu< warn—cane uuocnuanuwocvuriu covu.ucou c_ozu “Loam =_ogu ugogm once =.ogu aeogm cvogu acogm mock poucmsvconxu c» «we; on omen opuoa Loupaa ou=_ vomoupoc uxwopos: o_»oa mama.» cvlmwwo—oe: macaw: an: e? m_u>og u:.u.=eau co nexo=< yo uuowuu ~ o—aoh He: Bu TITTICLOUR— 32 Table II Short-Chain Acylcarnitines of Normoxic and Anoxic Rat Heart and Incubation Buffer. Sample Acyl Control Anoxic Analyzed Carnitine nmoleslg; nmoles/g Heart Tissue acetyl 150.6 183.2 propionyl 50.7 10.4 isobutyryl 19.4 80.5 butyryl 67.7 29.5 isovaleryl 22.2 18.0 valeryl ref. ref. tiglyl 14.8 10.2 caproyl 29.5 40.0 sum of individual 5 . 371.5 Total Short Chain 440 460 Buffer acetyl 32.6 12.0 propionyl 11.7 3.6 isobutyryl 0.3 9.1 butyryl 10.4 6.0 isovaleryl 1.2 2.9 valeryl ref. re . tiglyl 1.1 3.0 caproyl 3.5 8.0 sum of individual 6078' m Total Short Chain 68.0 54.0 The hearts and buffer from 10 experiments described in Table I were analyzed for short-chain acyl carnitines and free carnitine. The short-chain acylcarnitines were isolated and the volatile fatty acids determined by gas chromatography as described in methods (7). Hearts from 10 fasted animals were -used and the incubation buffer contained 5 mM a-hydroxybutyrate. ('1 "sale See: z‘rt ‘ienti :‘t lentr‘ 3:31 Term 3?". ‘ientr‘ rat Vent: :‘T‘ Ventr' int Vent I‘irerfuses tan Va] u Sid. Dev. F311 Ulffe \ Iii-”USN c F" Moved Q: "IDOXI c "I’Wgh a fused t J; ventricl 33 Table III Carnitine Levels in Hypoxic and Control Perfused Pig Hearts Nonperfused Hearts Perfused Hearts nmol eng nmol esjl Short Chain Short Chain Tissue Sample Free Carnitine Acylcarnitines Free Carnitine Acylcarnitines 1. Right Ventricle 248 253 213 122 2. Left Ventricle 246 225 239 168 Right Ventricle 247 253 160 105+ 3. Left Ventricle* --- --- 185 182 Right Ventricle 248 345 164 135+ 4. Left Ventricle 208 274 255 170 Right Ventricle 181 242 159 103+ 5. Left Ventricle* --- --- 278 171 Right Ventricle 388 246 207 102+ Nonperfused Perfused Mean Value 272 268 Control 239 173 Std. Dev. :72 :43 :40. 1 6 Perfused Hypoxic 181 108 :27 x 9 Percent Difference (Perfused Control: Hypoxic) -24.4 -37.4 (Nonperfused Control: Perfused Control) -12.2 -35.6 Nonperfused control hearts were obtained from air ventillated pigs whose blood had been removed for the perfusion studies. For perfused hearts the right ventricle was made hypoxic for 10 minutes by circulating the blood from the right leg and pumping it through a donor lung which contained 0% 02 (P02 150 to P02 20). The blood then passed through the pigs lungs to obtain normal P02 and then passed through the left ventricle. The pig was sacrificed by stopping the blood flow for 2 minutes and tissue samples were then removed and freeze clamped and the carnitine levels determined. The values in nmol/g tissue for long chain acylcarnitines were: 41.6 t 4.7 for nonperfused tissue, 61.6 t 2.1 for the perfused control and 72.5 t 2.1 for the perfused hypoxic samples; n=4. *Left ventricle not collected for analyses. +Wilcoxon's Signed Rank Test (p < .05). ESTISTIF.‘ 34 Table IV Effect of Hypoxia on the Short-Chain Acylcarnitine Levels of Perfused Pig Hearts Sample Acylcarnitine nmoleszg 1) Right Ventricle acetyl 57.1 (hypoxic) propionyl 13.8 isobutyryl 11.0 butyryl 11.4 isovaleryl 3.3 valeryl 11.2 caproyl 10.4 Sun 118.2 *Total Short-Chain 108 2) Left Ventricle acetyl 153.1 (normoxic) propionyl 3.9 isobutyryl 1.5 butyryl 6.8 isovaleryl 1.8 valeryl 1.4 caproyl 1.2 Sum 169.7 *Total Short-Chain 173 Equal amounts of tissue from 6 left ventricles of control samples and right ventricles of 6 perfused hearts described in Table III were pooled into separate groups and analyzed for short-chain acylcarnitines. The samples were pooled due to the lack of enough tissue and the inordinate amount of time required to analyze each sample. *Determined by independent acyl-carnitine analyses. 35 Affect of Alloxan Diabetes on Carnitine Content in Sheep Livers This was a joint project with Dr. A.N. Snoswell, from Waite Agricultural Research Institute, Adelaide, Australia. He provided lyophilized perchloric acid extracts from both normal and alloxan diabetic sheep livers. The neutralized perchloric acid extracts were resuspended in water and were then analyzed for the acylcarnitines as described in the methods. Examples of the gas chromatogram profiles of the acyl residues of acycarnitine from normal (animal no. 2) and diabetic (animal no. 5) sheep livers are shown in figure 3. The various labeled peaks correspond to the peaks obtained during a gas chromatographic run of known fatty acids. Valerylcarnitine was used as the internal standard because priminary studies contained only trace amounts of this acyl residue in both normal and diabetic livers (data not shown). Carnitine Concentration in Normal and Alloxan Diabetic Sheep Livers The free-, short-chain acyl-, and total water soluble-carnitine levels in the livers are shown in Table V. There was a significant increase in the carnitine concentration of alloxan diabetic livers, free carnitine increased 2-9 fold while the short-chain acylcarnitine concentration increased 10 to 150 fold when compared to the normals. a communication with Dr. Snoswell, he described animal no. 6 as being (gnly moderately diabetic due to the lack of fatty material in the liver. But even this animal had increased levels of carnitine in the liver. the onset of alloxan diabetes leads to increased levels of carnitine in sheep liver. 36 Quantitation of the Short Chain Acylcarnitines The amount of each identified acyl residue of the acylcarnitines from both normal and diabetic livers are given in Table VI. In both metabolic states acetylcarnitine was the predominant acyl derivative, 77% in normals and 71% in diabetic, however the absolute amounts of the acetyl carnitine in diabetic livers are increased by at least 40 fold. All of the other acyl residues are also increased many fold in the diabetic livers, although their percent distribution of the total acylcarnitines were similar to the acyl distribution in normal livers. These data illustrate two major points. One is that the amounts of acylcarnitine can change in response to the metabolic state of the animal, in this case alloxan diabetes. Second is that the tissue contains several different acylcarnitines including propionyl-, isobutyryl-, and isovaleryl-, which could be products of oxidative degradation of branch-chain amino acids. In a personal communication with Dr. A.N. Snoswell he has informed me that the specific carnitine acetyl-, and isobutyryl-transferase have increased activity significantly in the diabetic state, however the medium-chain and palmitoyl transferase activities in both metabolic states are similar. The means of the transferase activitys of acetyl, isobutyryl, octanoyl, and palmitoyl were 3.2, 1.6, 1.4, and 2.0 nmol/min/mg protein in controls and 6.3, 3.4, 2.0, and 1.0 nmol/min/mg protein respectively in alloxan diabetic sheep. Figure 3. 37 g ‘mm I 5) -‘ .4 ‘4- '. 'C‘ a] [II I I [f Hr: fiF Gas chromatogram profiles of the acyl residues of acylcarnitines from normal and diabetic sheep livers. Acyl residues of the acylcarnitines from alloxan diabetic and normal sheep livers were isolated and injected into a GC as described in the methods. The profile of known fatty acids is presented in the first figure (1) above. The letter after each peak corresponds.tp these fatty acids; A:acetic; B:propionic; C:isobutyric; Dzbutyric; E:isovaleric; F:valeric; G:tiglic; H:captoic. Profile (2) is from alloxan diabetic liver number 5 and profile (3) is from control liver number 2. 38 so» umxmmmu cog» mm: uooguxm ugh .xpm:o*>mca umnPLomwu avenues we» mcpm: mc_upceuu .m.= mg» on ugoqmcmeu o» eopca umeponnexp gas» can uo~vpoeu=wc we: uuacuxo one .v.ua uvLo—zucma no ;u_: vouuueuxw mew: mocpupccau ovum on» one umqsopuumNowcu ma; ca>_~ some mo wom+a Em m a use uo>osmc mew: meo>wp awoem upumnupc cuxo_~m m can Fascoz m mmm mNNN comm o¢~ mmH mug AEmLm\mm—oEev maneuvccmu mpaapom upon —cuou .Loscp as» =_ _awtaaae afloat Lo cue. 0;» an acne cowpucpsgmuou a mu: m_;u .mpaswcu.cm;uo ogu mm ovumnwvv apmcm>mm me no: mo: Fmepcu mvshw mafi mwm cam cam“ oo- ooe oH omN mg meg NH mmfi hmEmLm\mmpoecv Asmemxmmposcv mm:_uwccmopzum :_o;onucosm mcwupccou out» mcm>P4 ammzm ovumnmwo was puscoz cw m—m>a4 weep—econ > spank so m c upaana_o mposcoz 39 .ucmvcopm Pmceoucr we» we wow: mmsimc_uvceuupxua_a> .> «Fan» soc» opnsam mc*::eewe an» Soc» vocpeuno mew: mocwu_ccuopaum mpnspom vane on» .muozaas on» :P umnecommu mo Euemouusoezu men a ween: umcweeouov mew: mm:*u_ceuupaoo as» m.~ m.~ ~.m m.~m om.oi mm.o m~.o ~¢.o Hm.o —aocamu ~.o o.~ m.m H.~H «H.o m~.o mecca Hm.o wH.o .apuwp H.~ m.H w.m «.mo NH.o -.o mo.o H¢.o m~.o Facm~m>om~ H.m w.m H.Hm mam ~.m ~N.~ mm.o w~.o mm.~ pxgauam H.~ ~.~ m.~ o.¢m No.0 NH.o ~o.o mH.o o~.o pacauanomfi e.mH o.m~ ~.oH mwm m.o~ m.H mm.” we.” oo.m Fxcoraoca o.H~ «OH mam ecca m.m~ m.mH w.HH m.mH m.oH FAuou< a fig. 2 a s. m Ema 2 Ne :_ :55 Am\~oe :v Au\PoE:v mcwupccmu—zw< ovuonuro _mscoz Lm>vm ammzm mo maneuvccmo—au< H> mpamh HUMAN STUDIES The human may be one of the best systems for demonstrating the various functions of carnitine because of metabolic dysfunctions due to carnitine deficiency or deficiency in carnitine palmitoyltransferase activity as described in the introduction. Urine and serun*were analyzed for the total amount of free carnitine and short-chain acylcarnitine content in six normal adults and in two children with metabolic problems. Sample Collection and Subject Information Six adult controls (29 1 9 years old) were fasted overnight for 12 hours prior to collecting 10 mL of whole red blood and total morning urine. The urines were frozen at -80°C and stored for carnitine analysis at a later time as described in methods. The whole red blood samples were centrifuged to obtain serum, which was also stored at -80°C for analysis at a later time. The subjects were fed a high protein breakfast and lunch and both blood and urine were collected 6 hours after the initial morning collection. Patient I had metabolic dysfunctions of unknown origin, with persistent vomiting and hyperactivity resulting in dehydration and acidosis. Dr. P. Maur, at Children's Hospital, Cincinnati, Ohio, sent a sample of lyophilized urine collected during a 24 hour fast because he had observed high amounts of an unusual acid present in the urine. He tested serum 40 41 carnitine levels and found that free carnitine was very low, but the short-chain acylcarnitines were much higher than normal (personal communication). Patient II was a child with Fanconis syndrome (a disease process involving defective absorption of specific ions by the kidneys). Ron Emaus in Dr. Bieber's laboratory at Michigan State University had found low levels of total carnitine in the blood (9 umolar compared to the normal 40-50 umolar). A 24 hour urine sample was collected during the time that blood samples were drawn. Short-Chain Acylcarnitines in the Serum and Urine from Fasted and Fed M3115. Six adults (29 1 9 years old) were fasted overnight for 12 hours prior to collecting a 10 mL aliquot of whole blood and the total overnight urine. The blood was clotted and was centrifuged to obtain serum. Serun and urine samples were again collected 6 hours later which was also 1 hour after a high protein lunch. Aliquots of the samples were analyzed for total water soluble carnitine content (free carnitine and short-chain acylcarnitines), results shown in Table VII. The data show: 1) Fasting does not affect the ratio of free carnitine to short-chain acylcarnitines in either serum and urine. 2) The concentration of total (free and short-chain acyl) carnitine in urine increases significantly in the fasted samples, while the serum total carnitine concentration remained constant. This may be due to the difference in time allowed for accumulation of carnitine in the urine between the fasted and fed samples. 3) The ratio of free carnitine to short-chain acylcarnitine decreases significantly when compared to the sam ind fre uri ur' ac; da‘ di sa sh fr ac de St pc it Hi Vi III 01 ac 42 same ratio in serum. This occurred in both fasted and fed samples indicating the kidneys are preferentially or selectively reabsorbing free carnitine while the short-chain acylcarnitines are retained in the urine. The composition of the individual short-chain acylcarnitine from urines of both fasted and fed individuals were determined because the acylcarnitine appear to be selectively retained in the urine. These data are shown in Table VIII. The data show that the relative distribution of some of specific acylcarnitines differ between the samples from fasted and fed individuals even though the major short-chain acylcarnitine was acetylcarnitine in all samples. The urine from fed individuals contained a higher portion of propionyl-, isobutyryl-, and isovaleryl-carnitine while the portion of acetylcarnitine decreased. Increases in these acylcarnitines could be derived from branch chain amino acids, reflecting increases in the steady state level of the respective acyl coenzyme A levels and, possibly, increases in the contribution of the metabolic pathways leading to these compounds. The unidentified acylcarnitines were made up of many minor unidentified acyl groups, valerylcarnitine, and 3 peaks containing about 5% of total acylcarnitines. Valerylcarnitine was not be reported because it was used as the internal standard. In 2 separate urine samples in which crotonylcarnitine standard was used, valerylcarnitine represented 3% of the total. The composition of the individual short-chain acylcarnitines from human serum was also determined to ascertain if the acyl composition of serum is similar to that in urine. Results of the total short-chain acylcarnitine concentration analysis and the acyl composition are given 43 in Table IX. The major acylcarnitine was acetylcarnitine, but relatively large amounts of propionyl- and butyryl carnitine, both of which might be derived from B-oxidation of long-chain fatty acids, are also present. The mean values of the proportion of individual acylcarnitines in serum and urine are shown in Table X. The data show that the percent of acetyl-, propionyl-, and butyryl-carnitine in serum is almost double that in urine, while the percent of isobutyryl-, and isovaleryl-carnitine is much higher in urine than in serum. The data indicate that not only is free carnitine reabsorbed into the blood (Table VII) but it appears that acetyl-, propionyl-, and butyryl-carnitine are also selectively reabsorbed, while isovaleryl-, isobutyryl-, and caproyl-carnitine are retained in the urine. Agylcarnitine Determination of Urine from Patient I Results of the total amount of free and short-chain acyl carnitine and acylcarnitine composition present in a 24 hour urine collection of the patient suffering from dehydration and acidosis arising from some unknown metabolic problem are given in Table XI. The short-chain acylcarnitine fraction of the patient contained a large proportion of propionylcarnitine. In controls propionylcarnitine is about 5% of the total short-chain acylcarnitines which in the patient propionylcarnitine represents greater than 80% of the total short-chain acylcarnitines. This indicates a possible build-up of propionyl CoA in this individual. 44 Acylcarnitine Determination of Urine from Patient II The urine and blood of a person with Fanconi syndrome was collected during a 24 hour fast. The patient had low blood levels of total carnitine but the urine contained large amounts of total carnitine. The short-chain acylcarnitine composition is interesting (Table XII), because of the increased proportion of butyrylcarnitine and decreased proportion of acetylcarnitine compared to control values. The increased proportion of butyrylcarnitine indicates a build-up of butyryl CoA, which should be generated during B-oxidation of long-chain fatty acids. Increased butyryl CoA would be generated by incomplete B-oxidation which may account for the decreased proportion of acetylcarnitine. 45 gene mmpaEmm were: ecu sscom upon: mo DE oH sage uocwcuno mg: ssewm on» ma mm ca mm Hm mm um ms mm ms Hm om mace R .mmc.p_ceuo—aum cvogo ugogm + mcvupcemu mute n ocwupccoo oppapom Laue: _muou mucomocame Fences O coo 0‘? com 6'? #¢ SON Cd) so Nd’ com o ”H . on o _»u< ccaeu Steam ”.me ¢.¢¢ m.~e H.m¢ ~.o¢ m.¢e m.mm o.wm o.o¢ o.e¢ o.¢m m.am ¢.om ~.Am H.5N H.mm n.4m ¢.wm a.a~ o.mm o.¢m ~.me c.5m ¢.H¢ mung .papoh Ta 855 Ezcmm wave: vac escmm cogs: cw covumgucmocou ocwuwcemu me av mm Hm mm mm mN ea Hm um mm mm mac; n mm «Ha moa Nmm amH HNN Hm mwa o¢~ mwm fim owH mem ccagu-oao;m «a No” ca: SNN em mSN NoH Smm em” mew mew owe SN mm um HNN emw sac «mm can on as ecu awe «mum .Paam» T8 355 a:_L= mopae< eaa ecu assume to H~> mpnmh Nmm mam me «ea com omc cum mmH com Cum coo m¢m Aggy 25 P2, .mcogums we» :_ uoapeomuv wczuuuocn vaponapowuoe uwuosa~cm on» anew: um:_sgwuou mew: mcopuecucmocoo mcwupccmu use .mmpasem vegan» mg» Luumm mesa; m use Fume cwmuoca saw; a coupe Lao; mco venom—poo mum: m~a=u_>wvcw no» use eaoaaa own uwummu vmu pounce emu empmaa use eaumaa emu nuance .msvu dawn on» xpmuuevxocnam um vapomppoo mew: mmpasmm wave: on» use .uoopn .umwm wurccm>o Lao; NH a Levee umcruuno mew: mmpqsmm voumwm as» on me a“ m: N§ fie mpqemm 46 .uo.n «zone a. o=_u.:coo—aLo—~> yoga uu:o;m uwpa2om acoucmqocc~ .paau.>.uc_ usam as» soc» case» «so: no» a. opqeom van vegan» H. u—ae~m .~> a_no» c. vocweLouuu mmc_u.ccoupxuu Lo pesos» page» as» o» cocoaeoo voguue no xn v~:.ouao pesoEQ acmmoeamc —auo mo acouewa use .vemucaum poccauc— us» no oc_u_ccoupALu—os a:.m= .mvoguoe a. van.LUmou no wove—acoaa vac uo_c.u=mu’ new: mmevupceeupxuo as» ._> o_nah sage uo=.~unc new: o=.u_:eou—auo Go mopoe pauou on» .Auoc.uuno opasem vegan» Luueo mesa; my .oos cvouoea zu.g a sauce use: use ago one» Lao; Na a cause «apnea o soc» vac—ouno mew: mac—umeuxo meet: page» 3:84: 9:3 can: c.3922“? 3.2.8 3mm 2mm: 5...: m.o~ o.~ - o.m N." ~.m~ ~.m~ c.mm ~.om o o.» o.~ . ~.a o.~ ".mfi w.~ w.em ~.om m o.m m.o o.o o.o~ c.~ ~.m~ v.9 H.~m n.om e ~.H~ e.~ m.o 0.5 m.e o.m~ o.m ~.m¢ ~.o~ m m.n~ ~.¢ n.~ ~.o~ m.~ o.m~ ~.m ~.¢¢ o.m¢ ~ e.m m.m ~.~ m.u o.~ o.m~ 5.9 ~.em «.mc H. can coccemue< .m 3min 3nd flouuml .2an Sum Eng Sum Emma: clad o.o a.~ - c.w ~.o o.e~ m.¢ ¢.om m.nn o e.m ~.~ o.o m.e m.m o.mH H.o ~.oo e.mm m ~.- w.c e.o m.- ~.~ o.w~ m.m m.me m.oo c ~.- o." m.~ h.e H.~ ~.¢H ~.e m.~m m.m~ m ~.o~ ~.~ ".H w.m ¢.v m.o~ ~.~ «.me m.~m m A.NH e.~ m.o m.~ N." n.o~ ~.¢ c.~o “.mo fl‘ pounce acpccoz .< vu.».ucouvca paocauu —>—m¢u Pucopo>omp pacausa paexuznom. —»co.aoca paumuo mc.u.:eoopau< «pesom o:.u.ceou—»uo ceeguiucogm page» we ozv.moe pace a mup=u< van ecu Gounod we acoucou oc.u_cccupzu< xumcega -~> open» :Pozo-ucogm 33:3. :3» 47 .compvuz .c_mcoom—: .>—:= .coccog m.coo x: vovv>acn opasom mm—oacv>.ucv N soc» um—ooa we: eseomn .-p»ua =.~gu-ucogm + omen mucamoeaoe oc_u_ccnu pauopu .ucovcuum puceouc— .u.u on» no: oc.u_:cau pacouocu .vozuoe vo—onup u.uasx~co as» us coc.=couov oc.u_ccuopxua page» as» o» .u.c as» an vw=.eLmuov .xoo some Lo unsoEe on“ mcpceaeou an voc.ouno no: maqumL page zone mucomocqmc peace we acaueoa one .mcogume on» c? uoa.cumoc mo oc.u.:eaopxuo co mucoeoqeou page oxu can co.uocucoocco oc_u.ccau so» vmnxpoco cog» «so: meseum use .co_uom:e?eucoo an vac—p vane—amoou as» see» vmc.euno mo: sseom as» use mapsuo to» a sac» Age any c:ocv any: moPanm poops o—ogx to ... no - a." n a." to .+. no 3 “mm gum 3 .+. 92 3 .. mi .2! o.H u m.~ c.~ w.a a." o.m~ m.mm ~.¢ o.om am o." . ¢.m sm.o m.c ~.~ .ou o.w~ w.m m.eo c m.~ u ~.v N.“ H.@ o.~ m.- o.- m.¢ m.m¢ m m.o o ¢.m o.o ~.m —.~ o.- ~.m~ ¢.m w.mc N 5.: n w.m m.o c.m ¢.~ m.o~ o.m~ ¢.e ~.wm fl paocavu —»#m_u pawn—u> fluuopa>omp Fauxuan ~Nuau=nomp pummwnocn pnwuon ocwuwccuu—»Ua ocwuvceuu wmmEom :Fczuuueocm opnuo» mac—accou—Auo :_ogo-ueogm Fouou we a Lo—oES- «cogs: :. acuucou ocru.ceau~»u< ssgmm x— «pooh - 48 Table X Proportion of Individual Acylcarnitines in Human Serum and Urine Acycarnitine In Urine .Sgggp :q acetyl 44.7 1 9.9 73.5 1 3.1 ' 3 propionyl 8.7 1 2.6 12.0 1 1.4 isobutyryl 18.1 1 4.7 2.2 1 0.7 _ butyryl 2.9 1 0.9 6.3 1 2.1 ,V isovaleryl 10.6 1 5.0 0.7 1 0.4 valeryl 3.0 1 1.1 3.9 1 1.0 tiglyl 1.0 1 1.0 0.0 1 0.0 caproyl 10.6 1 1.6 .0.8 1 0.4 Tabulation of the mean values of the % of each acylcarnitine in serum and urine were obtained from Tables VIII and IX. 49 .A__~> mpaupv mppzvm topmow xwm we were: men soc» umcwoupo mu=_m> poeucou .muosums c. umnPLommv we mmcruvzenoPAUn cemeuuucogm on» we unseen FAum Lo covuvmoaeou ecu cowuucucmucoo wcwupccou com vm~apocm mo: Am.mouvum new covmeuA:mvv msmpaoca uwponmuoe cue: ucmpumq a sage awe» Lao; em a mcvczu venom—poo wave: ~.HH m.~H in- vm_w.pcmupcs w.~ H. mfio. mcwuvccuoupxocaou ~.c so. woo. a=_ucetau-_a_muu ~.~ m. ace. mcwupccoUIPALopu>omv m.~ co. Nooo. mcpupceouu_»uauan m.e~ m.o umo. mcwuwceuoupxcausnomw m.m o.~m ~¢.mH unpuvccmu-Paco_noca ¢.mm m.e emu. occupeemuupxumuu -u- ii- m.- Auuuv _auop ooH ooH m.m~ mcpupccmu Pau< .u.m ii- Tu- o.HH mc_uvccau mack mcpuwceuolpaum nape—ecmo—aua cpogoupeozm mesa; <~\mmFoESx acmcoasoo cemzouueozm co a peace co m upau< Poseoz pocucou H acovumm mo acmucou o:+u_:cmu_»o< Apogee: Hx mpnm» 5C) .maaEm com :5 vov.>ocn no: o:.uwccuu—»ua c_o=uiueozm tea «as» acaoea on» ma sumo .A__g> o_nuhv mu_:vo vegan» x.m mo oc.c= an» ace» noe¢ouao mum: mos—u: mu—suo poeucau .mvozums c? non.eumou no umseoueoa new: mocwu.:cau—»uo we» we covu~u_u.uca=o .osoeuczw «reeves; zu.: vP—gu a no one» em a oc—Lav uuuuoppou no: cowuocuxo agacpea .auou on» 3.1: ~.- ea.e.u=ae.== m.~ m.e omm.¢ -Faozaau ~.o ~.~ oe:.~ -_»Pm.u A.“ o.m omm.¢ -Fxgaraso._ n.~ o.w~ :am.¢H -_»L:S=n e.¢~ o.m mos.” -_»g»S=SOm. n.m e.~ mac.» -Fsco.aOLa e.mm ”.mm Ago.e~ -.»»au. mme.wa oe.u.=aau _»o< .u.m o6: $592.30 00....— ocz—Eou—huo .u.m to m ocvupEuuphou 52319.93 «.50... ¢~\mu—95i acmcoasou 2:2 .252. .33 .3 a .oaueou __ acovuoa Lo acuucou oc_u.ccuupxo< xuacwca :x 032. DISCUSSION Effect of Hypoxia on Heart Acylcarnitines The data presented herein show that when rat hearts are incubated under anoxic conditions, the tissue preferentially loses short-chain acylcarnitines. This loss of short-chain acylcarnitines and also free carnitine is linear for at least 30 minutes. In our experiments we did not observe a preferential loss of carnitine from cardiac tissue due to anoxia, in contrast to others, but it should be noted that we did not use tissues perfused with blood or media for the rat heart experiments. However, when pig hearts were perfused with hypoxic conditions, a 37% loss of short-chain acylcarnitines occurred compared to the control perfused ventricle and 70% loss occurred when compared to non-perfused left ventricles. These data are in good agreement with the observations of others who have observed a loss of tissue carnitine during hypoxia and ischemia (39,40,50,51). Our results also show that during hypoxia the long-chain acylcarnitines in tissues of perfused pig heart are increased, presumably due to inhibition of B-oxidation of long-chain fatty acids. In contrast, there was a marked decrease in acetylcarnitine of hypoxic ventricles, again, presumably due to decreased flux of long-chain fatty . acids through s-oxidation and, consequently, decreased formation of acetleoA (the precursor of acetylcarnitine). 51 52 Hypoxia caused an increase in the amounts of propionyl-, isobutyryl-, butyryl-, and valeryl-carnitine. Increases in propionyl-, isobutyryl-, and isovaleryl-carnitine should reflect increases in the steady state level of the respective acyl coenzyme A levels and, possibly, increases in the contribution of the metabolic pathways leading to such compounds. These acylcarnitines could be derived from branched chain amino acids (22). Similarily, the increases in butyrylcarnitine and caproylcarnitine could indicate increases in the steady state level of the intermediate products of B-oxidation. The rat heart data are more difficult to interpret since anoxia caused a decrease in the tissue propionylcarnitine but an increase in the isobutyrylcarnitine and lesser changes in the other acylcarnitines. The decrease in the amount of propionylcarnitine in rat heart should reflect a decrease in the steady state level of propionylcoenzyme A. Propionylcoenzyme A is derived from the carbon skeletons of isoleucine, valine and from odd chain fatty acids; hence the decrease could represent decreased isoleucine, valine metabolism and/or decreased oxidation of odd chain fatty acids. This seems reasonable since anoxia should completely inhibit a-oxidation of fatty acids. Alternatively, the data are also consistent with inhibition of the pathways for metabolism of the aliphatic short chain coenzyme A derivatives and their subsequent conversions to acylcarnitines. Thus, the increased levels of isobutyrylcarnitine observed during limited or no oxygen could reflect a blockage of the enzymatic step for - the conversion of isobutyrylcoenzyme A to isobutenylcoenzyme A, a flavin linked dehydrogenase. Carnitine could then be acting as a sink for isobutyryl units at the level of isobutyrylcoenzyme A. 53 Acylcarnitines of Human Urine and Serum The data show that the proportion of the total water soluble carnitines represented by free carnitine remains constant in serum and urine when comparing fasted to fed levels, however the percent free carnitine is significantly different between urine and serum for both fasted and fed individuals. This indicates that the kidneys may be retaining short-chain acylcarnitines in the urine while preferentially ‘ fit. an, reabsorbing free carnitine. This may provide the human with a mechanism of removing some acyl residues from the body via the acylcarnitines, which will neutralize the fatty acids and may also allow for the removal a of "excess" carnitine, since carnitine is not metabolized in animals and is only removed from the body by urine excretion. Therefore carnitine can act as a carrier which binds and neutralizes the fatty acyl residue for excretion in urine. Identification of the acylcarnitines of urine showed that acetylcarnitine is the major component, however its proportion was less in urine samples of fed human, while propionyl-, isobutyryl-, and isovaleryl-carnitine were increased slightly. As mentioned earlier, increases in these acylcarnitines could reflect increased levels of branched-chain amino acid metabolites. One would expect increases in Branched-chain amino acid oxidation after a high protein meal, which could generate the acylcarnitines found in the urine. Human serum contains a high proportion of acetylcarnitine, and propionyl-, valeryl-, and butyryl-carnitine represent almost all of the remaining acyl carnitines. It is interesting that the proportion of the acylcarnitines in urine and serum of fed adults give different results: as mentioned above serum contains high proportions of acetyl-, 54 propionyl-, butyryl-, and valeryl-carnitine, which are almost double the relative proportion of these acyl residues in the urine. However, the urine contains isobutyryl-, and isovaleryl-carnitines, in proportion 10 times greater than these observed in serum. Thus the kidneys not only selectively reabsorb free carnitine and excrete acylcarnitines, but they also preferentially reabsorb acetyl-, propionyl-, butyryl-, and valeryl- carnitine while they excrete isobutyryl-, and isovaleryl-carnitine in the urine. This is related to the data that have shown (101) that the translocase which moves isobutyrylcarnitine into the matrix of mitochon- dria has a low affinity (high Km) for isobutyrylcarnitine and conse- quently uptake of carnitine and/or some other acylcarnitines are prefer- red when all are present. Thus the kidneys may have a lower affinity for reabsorbing isobutyryl-, and isovaleryl-carnitine. Other studies have shown (24) that isobutyrylcarnitine is readily oxidized by beef liver and rat liver mitochondria in the absence of free carnitine but when free carnitine is present the oxidation is inhibited, presunably, due to the competition of free carnitine for isobutyrylcarnitine at the translocase step. The increases in isobutyrylcarnitine during anoxia and hypoxia are consistent if one assumes that isobutyrylcarnitine is in equilibrium with isobutyryl- coenzyme A and that carnitine may be a sink for disposing of excess short-chain acyl residues at level of acylcoenzyme A. The short chain acylcarnitine composition of the urine from patient I supports the statement that carnitine removes excess acyl residues because the proportion of propionylcarnitine in urine was greater than 80% of the total acylcarnitines while in controls propionylcarnitine is about 5% of the total acylcarnitines. Thus carnitine may be a sink for 55 excess propionyl CoA which will restore the acyl CoA/CoASH ratio. Apparently the increase of propionylcarnitine indicates a block in the conversion of propionyl CoA to methylmalonyl CoA by the biotin linked carboxylase enzyme, or in the vitamin 812 linked mutase that converts L-methylmalonyl CoA to succinyl CoA. The urine from patient II (Fanconi syndrome) contain large amounts of free carnitine and short-chain acylcarnitines (40% of total water soluble carnitine). The urine contained each of the acyl residues that are present in acylcarnitines from normal human urine. Since the blood had very low levels of total carnitine, it was assummed that the patient had a deficiency in carnitine uptake which resulted in a muscle deficiency. However the presence of acylated carnitines indicates that the carnitine had been intracellular in some tissues prior to excretion in the urine. It seems the kidneys were not reabsorbing carnitine, thus the muscle carnitine deficiency was due to a kidney defect, which would indicate a systemic deficiency instead of a myopathic deficiency. Diabetic Sheep Livers The effect of alloxan diabetes on sheep liver acylcarnitines sup- ports the idea that the levels of acylcarnitine reflect the steady state levels of the respective acyl CoA levels. The levels of each acylcarni- tine increased many fold in alloxan diabetic liver compared to normal livers, however the relative proportion of the total acylcarnitines represented by each acylcarnitine remains constant between the two metabolic states. For example, acetyl- and propionyl-carnitine were 76.5 and 10.5% of total acylcarnitines in normal livers and these acyls were 71.0 and 13.6% of total acylcarnitine in diabetic livers. 56 As stated earlier a proposed function for carnitine is to modulate the CoASH/acyl CoA ratio (24,34) by formation of acylcarnitine from the acyl CoA derivatives via the various carnitine acyltransferases that occur in tissues (23). Thus carnitine could buffer the CoASH/acyl CoA ratio, thereby insuring the availability of CoASH for oxidative metabolism in the matrix of mitochondria. CoASH is required at this site for the conversion of: pyruvate to acetyl CoA, long fatty acids to acetyl CoA, a-ketoglutarate to succinyl CoA, and branched-chain a—ketoacids (valine, leucine, and isoleucine) to branched-chain acyl CoAs. Figure 4 depicts the relationship of carnitine to the acyl CoA pool and the regeneration of CoASH in the matrix of mitochondria. Reactions 1 (branched-chain a-ketoacid dehydrogenases), 2 (a-ketoglutarate dehydrogenase), 3 (carnitine acyltransferases), 4 (B-oxidation enzymes), 5 (pyruvate dehydrogenase) all require CoASH, and all of the products except for reaction 2 are in equilibrium with carnitine. Thus generating acylcarnitines via the carnitine acyltransferases liberates CoASH in the matrix of mitochondria, as long as carnitine is available. If carnitine levels are decreased then the "acyl buffer" effect is diminished, which would reduce flux through reactions 1-5 and some reaction (1 and 5) would be modified due to allosteric control of these enzymes by the CoASH/acyl CoA ratio (99,100). Both enzymes are highly regulated and respond to allosteric modulators such as the CoASH/acyl CoA ratio. During anoxic and hypoxic conditions, cardiac tissue selectively retains free carnitine apparently at the expense of short-chain acylcarnitines which are lost from the tissue. This is consistent with the suggestion that retention of free carnitine would enable the 57 remaining carnitine to buffer the CoASH to acylcoenzyme A ratio, thereby insuring the availability of CoASH for oxidative metabolism in the matrix of mitochondria. Such a role for carnitine would be consistent with the observed affects of free carnitine on ischemia (24,97) even as it may relate to pyruvate oxidation (43). The reduction in the amounts of short-chain coenzyme A derivatives would also be consistent with the inhibition of adenine translocase by acylcoenzyme A derivatives during ischemia. The reduction of the translocase inhibition by carnitine would be via formation of acylcarnitines. One could speculate that patient I in Table XI was severely acidotic because of inadequate availability of CoASH and an abnormally high ratio of acleoA/CoASH due to the large amounts of propionyl CoA. 58 RELATION OF CARNITINE TO CoASH IN THE MATRIX OF MITOCHONDRIA / BCA-CAFNITIIA / BCACoA CoASH BCKA I “'KG 2 , PYRUVATE Co SH SuccCoA 5 4 ACETYLCoA LCA-CoA ACETYILARNITINE LCA-CARNITINE K CoASH ) K CoASH 3/ IM.) ACETYLCARNITINE‘Z \LCA CARNITINE SUMMARY Studies with anoxic and Hypoxic hearts The investigations show that anoxia results in preferential loss of short-chain acylcarnitines from rat hearts. The loss was linear for 30 minutes. Hypoxic perfused pig hearts also preferentially lost short-chain acylcarnitines. Hypoxia caused a 70% decrease in the total carnitine concentration compared to non-perfused hearts. Analysis of the carnitine of hypoxic heart tissue showed increased long-chain acylcarnitines and decreased acetylcarnitine, which indicates a block in B-oxidation of fatty acids. Hypoxia also increased the amount of branched-chain acylcarnitines, which could be derived from amino acid oxidation. Studies with alloxan diabetic sheep Alloxan diabetes results in very high levels of free- and short-chain acyl-carnitines in sheep livers when compared to normals. Analysis of theacylcarnitines showed that each of the acyl residues were about equally elevated in the alloxan treated sheep livers. These acylcarnitines should give an indication of the steady state levels of the respective acyl CoAs. 59 60 Studies with Humans Serum samples from fed or fasted adults showed no change in either the total carnitine concentration or in the ratio of free carnitine to short-chain acylcarnitine. The major acyl residue of the acylcarnitines was acetyl with propionyl, butyryl, and valeryl accounting for the remaining residues. When comparing urine samples to serum samples, it was found that the urine samples contained a lower ratio of free Ea carnitine/acylcarnitine ratio and the acylcarnitines contained a lower It proportion of acetyl, propionyl, and butyryl residues while isobutyryl and isovaleryl proportions increased. Analysis of the acylcarnitines in a urine from patient I showed very high levels of propionyl carnitine suggesting that carnitine could be used as a sink for excess short-chain acyl residues that are generated in the mitochondrial matrix. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. REFERENCES Gulewitich, v.5. and R. Krimberg, z. Physiol. Chem. 3; 325, 1905. Kutscher, F.Z., Untersuch Nabr. u Genussm, lg_528, 1905. Strack, E., Wordehoff, P. and H. Schwaneberg, Z. Physiol. Chem. g§§_ 183, 1936. Strack, E. and K. Forsterling, Naunyn-Schmudeberg's Arch. Exptl. Path. u Pharmakol. 185 612, 1937. Fraenkel, G. and M. Blewett, Biochem. J. gl_469, 1947. Carter, H.E., Bhattacharyya, P.K., Weidman, K.R. and G. Fraenkel, Arch. Biochem. Biophys. 38 405, 1952. Fraenkel, G., Blewett, M. and M. Coles, Physiol. Zool. 23 92, 1950. Fraenkel, 6., Biol. Bull. 194 359, 1953. Fraenkel, G. and S. Friedman, Vitamins and Hormones 1§_73, 1957. Friedman, S. and G. Fraenkel, Arch. Biochem. Biophys. §2_491, 1955. Fritz, I.B., Acta Physiol. Skand. §4_367, 1955. Fritz, I.B., Am. J. Physiol. lg; 297, 1962. Fritz, I.B. and K.T.N. Yue, J. Lipid Res. fl_279, 1963. Bremer, J., J. Biol. Chem. g§Z_3628, 1962. Bremer, J., J. Biol. Chem. 2§§_2774, 1963. Klingenberg, M. and E. Pfaff, Symp. Reg. Metab. in Mito, Bari Elsevier, Amsterdam (1965). Fritz, I.B. and N.R. Marquis, Proc. Nat. Acad. Sci. §4_1226, 1965. Fritz, I.B., Schultz, 5.5. and P.A. Srere, J. Biol. Chem. 238 2509, 1963. Markwell, M.A.K., McGroarty, E.J., Bieber, L.L. and N.E. Tolbert, J. Biol. Chem. 24§_3426, 1973. 61 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 62 Markwell, M.A.K. Tolbert, N.E. and L.L. Bieber, Arch. Biochem. Biophys. l1§_479, 1976. Solberg, H.E. and J. Bremer, Biochem. Biophys. Acta 222 372, 1970. Bieber, L.L. and Y.R. Choi, Proc. Natl. Acad. Sci. 14 2795, 1977. Choi, Y.R., Fogle, P.J., Clarke, P.R.H. and L.L. Bieber, J. Biol. Chem. 252 7930, 1977. Choi, Y.R., Clarke, P.R.H. and L.L. Bieber, J. Biol. Chem. g§g_ 5580, 1979. Choi, Y.R., Fogle, P.J. and L.L. Bieber, J. Natr. lgg_155, 1979. Fogle, P.J. and L.L. Bieber, Biochem. Med. 22 119, 1979. Bieber, L.L., Sabourin, P.J. Fogle, P.J, Valkner, K.J. and R. Lutrick, In Carnitine Biosynthesis, Metabolism, and Functions, Frenkel, R.A. and 0.0. McGarry, eds., Academic Press, New York, 1980. VanHinsbergh, V.W.M., Veerkamp, J.H. Engeleu, P.J.M. and W.J. Ghijsen, Biochem. Med. gp_115, 1978. Paul, H.S. and S.A. Adibi, Am. J. Physiol. 2§fi_E494, 1978. Van Hinsbergh, V.W.M., Veerkamp, J.H. and J.H.G. Cordewener, Int. J. Biochem. l2_559, 1980. May, H.E. Aftring, R.P. and M.G. Buse, J. Biol. Chem. 255 8394, 1980. Parker, P.J. and P.J. Randle, Biochem. J. lZl_751, 1978. Bremer, J. and E.J. Davis, Biochem. BiOphys. Acta 528 269, 1978. Bieber, L.L. Emaus, R.K. Valkner, K.J. and S. Farrell, in print. Pande, S.V. and M.C. Blanchaer, J. Biol. Chem. 24§_402, 1971. McLean, P., Gumaa, K.A. and A.L. Greenbaum, FEBS Lett. ll_345, 1971. Shug, A.L., Shrago, E., Bittar, N. Folts, J.D. and J.R. Koke, Am. J. Physiol. 228 689, 1975. Shug, A.L., Lerner, E., Elson, C. and E. Shrago, Biochem. Biophys. Res. Comm. 4; 557, 1971. Shug, A.L., Texas Reports on Biology and Medicine §g_409, 1979. DiDonato, S., Rimoldi, M., Morse, A., Bertagnoglis, B. and G. Uziel, Neurol. 29 1578, 1978. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 63 Neely, J.R. and H.E. Morgan, Ann. Rev. Physiol. §§_413, 1974. Opie, L.H., Am. Heart J. 21 375, 1979. DiPalma, J.R., Ritchie, D.M. and R.F. McMitchell, Arch. Int'l. Pharm. Ther. g;1_245, 1975. Hart, Z.H., Chang, C.H., DiMauro, S., Farooki, Q. and R. Ayayr, Neurol. 28 147, 1978. 4 Pearson, D.J. and P.K. Tubbs, Biochem. J. lQ§_953, 1967. Oram, J.F., Bennetch, S.L. and J.R. Neely, J. Biol. Chem. 248 5299, 1*? 1973. ' Oram, J.F., Wenger, J.I. and J.R. Neely, J. Biol. Chem. 2§Q_73, 1975. Hochachka, P.W., Neely, J.R. and W.R. Driedzic, Fed. Proc. §§_2009, . p 1977. ij deJong, J.W. and w.c. Hulsmann, Biochem. Biophys. Acta lgz_127, 1970. Shug, A.L., Thomsen, J.D., Folts, J.D., Bittar, N., Klein, M.I., Koke, J.R. and P.J. Huth, Arch. Biochem. Biophys. 1§Z_25, 1978. Shug, A.L., Hayes, 8., Huth, P.J., Thomsen, J.H., Bittar, N., Hall, P.V. and R.H. Demling, In Carnitine Biosynthesis, Metabolism, and Function. R.A. Frenkel and J.D. McGarry, eds., Academic Press, New York, 1980. Snoswell, A.M. and G.D. Henderson, Biochem. J. ;;g_59, 1970. Baird, G.D, Heitzman, R.J. and A.M. Snoswell, Eur. J. Biochem. gg_ 704, 1972. Snoswell, A.M. and P.P. Koundakjian, Biochem. J. 1g1_133, 1972. Snoswell, A.M. and P.P. Koundakjian, Proc. Aust. Biochem. Soc. 6 36, 1973. Mitchell, M.E., Am. J. Clin. Nutr._§1 481, 1978. Mitchell, M.E., Am. J. Clin. Nutr. §1_645, 1978. Tanphaichitr, V., Lerdvuthisopan, N., Dhanamitta, S. and H.P. Broquist, Am. J. Clin. Nutr. 33.876, 1980. Carrier, H.N. and G. Berthillier, Muscle and Nerve 3 326, 1980. Cedarblad, 6., Clin. Chim. Acta 61 207, 1976. Maebashi, M., Kawamura, N., Sato, M. and K. Yoshinago, J. Lab. Clin. Med. 81 760, 1976. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 64 Frohlich, J., Seccombe, D.H., Hahn, P., Dodek, P. and I. Hynie, Metabolism 27 555, 1978. Hoppel, C.L. and S.M. Genuth, Am. J. Physiol. g§§_E409, 1980. Engel, A.G. and C. Angelini, Science lzg_899, 1973. Markesbery, W.R., McQuillen, M.P., Procopis, P.G., Harrison, A.R. and A.G. Engel, Arch. Neurol. §l_320, 1974. ' VanDyke, D.H., Griggs, R.C., Markesbery, W.R. and S. DiMauro, Neurol. 25 154, 1975. Karpati, G., Carpenter, 5., Engel, A.G., Watters, G., Allen, J., Rothman, S., Klassen, G. and 0.A. Mamer, Neurol. g§_15, 1975. Smyth, D.P.L., Lake, B.D., McDermot, J. and J. Wilson. Lancet 1 1198, 1975. Angelini, C., Pieroban, 5., Luke, S. and F. Cantarutti, Neurol. g§_ 633, 1976. Boudin, G., Mikol. J., Guillard, A. and A.G. Engel, J. Neurol. Sci. §Q_313, 1976. : Isaacs, H., Heffron, J.J.A., Badenhorst, M. and A. Pickering, J. Neurol. Neuros. Psych. §2_1114, 1976. Cornelio, F., DiDonato, 5., Pelucchetti, D., Bizzi, A., Bertagnolio, B., D'Angelo A. and U. Wiesmann, J. Neurol. Neuros. Psych. 49_170, 1977. Engel, A.G., Banker, 8.0. and R.M. Eiben, J. Neurol. Neuros. Psych. 49 313, 1977. Scarlato, G., Pellegrin, G., Cerri, C., Meola, G. and A. Veicsteinas, J. Can. Sci. Neurol. §_205, 1978. Carroll, J.E., Brooke, M.H, DeVio, D.C., Shumate, J.B., Kratz, R., Ringel, S.P. and J.H. Hagberg, Neurol. 39 618, 1980. Glasgow, A.M., Eng, G. and A.G. Engel, J. Pediatrics 2§_889, 1980. Chapoy, P.R., Angelini, C., Brown, W.J., Stiff, J.E., Shug, A.L. and S.D. Cederbaum, N. Engl. J. Med. §Q§_1389, 1980. Pola, P., Savi, L., Grilli, M., Flore, R. and M. Serricchio, Curr. Ther. Res. 21_208, 1980. DiMauro, S. and P.M.M. DiMauro, Science 1§g_929, 1973. Hostetler, K.Y., Hoppel, C.L., Romine, J.S., Sipe, J.C., Gross, S. and P. Higginbottom, Clin. Res. g§_125A, 1977. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 65 Scholte, H.R, Jennikens, P.G.I. and J.J.B.J. Bonvy, J. Neurol. Sci. 49_39, 1979. Patten, B.M., Wood, J.H., Harati, Y., Hefferan, P. and R.R. Howell, Am. J. Med. §Z_167, 1979. Layzer, R.B., Havel, R.J. and M.B. McIlroy, Neurology §Q_627, 1980. DiDonato, S., Cornelio, F., Pacini, L., Peluchetti, 0., Rimoldi, M and S. Spreafico, Ann. Neurol. 4_465, 1978. Ionasescu, V., Hug, G. and C. Hoppel, J. Neurol. Neuros. Psych. 43. 679, 1980. Freidman, 5., Arch. Biochem. Biophys. Z§_24, 1958. Marquis, N.R. and I.B. Fritz, J. Lipid Res. §_184, 1964. Cederblad, G. and S. Lindstedt, Clin. Chim. Acta §Z_235, 1972. Travassos, L.R. and C.O. Sales, Anal. Biochem. §§_485, 1974. Lewin, LM., Pershin, A and B. Sklary, Anal. Biochem. 68 531, 1975. Parvin, R. and S.V. Pande, Anal. Biochem. 19_190, 1977. McGarry, J.D. and o.w. Foster, .J. Lipid Res. y_ 277, 1975. Pace, J.A., Wannemacher Jr., R.W. and H.A. Neufeld, Clin. Chem. _2_4_ 32, 1978. Bieber, L.L. and L.M. Lewin, Methods in Enzymology, 12_276, 1981. Choi, Y.R. and L.L. Bieber, Anal. Biochem. 12_413, 1977. Bohmer, T. and J. Bremer, Biochim. Biophys. Acta 152 559, 1968. Folts, J.D., Shug, A.L., Koke, J.R. and N. Bittar, Amer. J. Cardiol. 4; 1209, 1978, Idell-Wenger, J.A. and L.W. Grotyohann, J. Biol. Chem. 2§§_5597, 1981. Williamson, J.R., Walajtys-Rode, E. and K.E. Coll, J. Biol. Chem. 254 11511, 1979. Waymack, P.P., DeBuysere, M.S. and M.S. Olson, J. Biol. Chem. 255 9773, 1980. VanHinsbergh, V.W.H., Veerkamp, J.H. and J.G.E.M. Zuurveld, FEBS Lett. gg_1oo, 1978. 66 102. Brdiczka, 0., Gerbitz, K. and D. Pette, European J. Biochem. I; 234, 1969. 103. Beenakkers, A.M.T. and P.T. Henderson, European J. Biochem. 1 187, 1967. 104. Engel, A.G., In Carnitine Biosynthesis, Metabolism, and Functions, R.A. Frendel and J.D. McGarry, eds., Academic Press, New York, 1980. -