MSU LIBRARIES v RETURNING MATERIALS: P1ace in book drop to remove this checkout from your record. FINES w111 be charged if book is returned after the date stamped be10w. 3-4 THE SEARCH FOR MITOCHONDRIAL CARNITINE OCTANOYL TRANSFERASE--AN INVESTIGATION OF CARNITINE ACYLTRANSFERASE ACTIVITIES IN BEEF HEART MITOCHONDRIA By Peter R. H. CTarke A DISSERTATION Submitted to Michigan State University in partial fulfiliment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Biochemistry 1981 Copyright by PETER R. H. CLARKE 1981 #1 ‘1 0" h «D ' 1 IL: (I. ABSTRACT THE SEARCH FOR MITOCHONDRIAL CARNITINE OCTANOYLTRANSFERASE--AN INVESTIGATION OF CARNITINE ACYLTRANSFERASE ACTIVITIES IN BEEF HEART MITOCHONDRIA By Peter R. H. Clarke The purpose of this study was to characterize the mitochondria] carnitine octanoyltransferase of beef heart. Carnitine acyltransferase activities were solubilized from isolated beef heart mitochondria using KCl and the non-ionic detergent, Triton X-lOO, at final concentrations of 1.! and 2%, respectively. Upon fractionation of the solubilized protein on Cibacron Blue Sepharose, two protein peaks with carnitine octanoyl- transferase were obtained. These two fractions accounted for all carnitine acyltransferase activity present in the original mitochondrial suspension. The first eluting peak was purified 400-fold by Sephadex 6-100 gel filtra- tion, CM-Sepharose ion exchange and hydroxylapatite chromatography to a single protein of greater than 95% purity. This carnitine acetyltrans- ferase (CAT) shows highest activity with acetyl and butyryl carnitine and coenzyme A esters. It has a subunit molecular weight of 62,600 daltons on SDS-polyacrylamide gel electrophoresis, a native molecular weight of 60,500 on Sephadex 6-200 gel filtration and an isoelectric pH of 8.20 on sucrose density gradient isoelectric focusing. (Iv ‘[‘ iv um .11- O U , fl 2 :n— I '. .‘ p.- “- Lug. f.\.\ a Peter R. H. Clarke The second peak of carnitine acyltransferase activity from Cibacron Blue Sepharose was purified l600-fold by fractionation on Sephadex G-lOO gel filtration, QAE-Sephadex ion exchange and hydroxylapatite chromatog- raphy to a single protein of greater than 95% purity. This enzyme was carnitine palmitoyltransferase (CPT). It is most active with decanoyl and lauryl ester substrates, has a subunit molecular weight of 67,000 daltons on SDS-PAGE, an isoelectric pH of 8.05 on sucrose density gradient isoelectric focusing and migrates as part of a detergent micelle of appar— ent molecular weight 510,000 on Sephadex 0-200 gel filtration. It is concluded that there are only two carnitine acyltransferase proteins present in beef heart mitochondria, one membrane-bound (CPT) and one membrane-associated (CAT). Each has significant activity toward hexanoyl, octanoyl and decanoyl carnitine and coenzyme A esters. The presence of a separate medium chain length-specific carnitine acyltrans- ferase in beef heart mitochondria is not confirmed by our results. The roles of micelles of the substrates octanoyl-, lauryl-, decanoyl-, myristoyl- and palmitoylcarnitine and of non-ionic detergents in the determination of the substrate specificity for the reverse reaction of a soluble purified beef heart mitochondrial carnitine palmitoyltransferase (CPT) are investigated. It is shown that the discontinuity in double reciprocal plots of substrate concentration versus reaction rate is attributable to the formation of substrate micelles at the critical micellar concentration (CMC) of the acylcarnitine. A lOO-fold increase in Km and a 6-fold increase in Vmax are obtained when reactions are carried out with micellar as opposed to monomeric concentrations of the Peter R. H. Clarke substrate myristoylcarnitine. With all substrates in the monomeric state, reactions performed in the absence of micelles of non-ionic deter- gent show that carnitine palmitoyltransferase is most specific for medium chain length substrates, with laurylcarnitine having the lowest Km (7.8 H!) and decanoylcarnitine the highest vmax (6.2 U/mg). The effect of the addition of micellar concentrations of nonionic detergent is to decrease the sharpness of the discontinuity seen at the substrate CMC in double reciprocal plots. At high detergent concentration, where the discontinuity is negligible, a substrate specificity pattern similar to that seen in detergent absence is observed. In the micellar environment, the enzyme shows up to 3-fold increases in Vmax for the vari- ous substrates and a consistent 5- to 6-fold increase in Km for octanoyl-, decanoyl-, lauryl- and myristoylcarnitine. Thus the kinetic parameters determined for carnitine palmitoyltransferase are shown to depend on the state of the substrates and on the enzyme's environment. DEDICATION To David and Maureen ii ACKNOWLEDGEMENTS The support of the Department of Biochemistry, its faculty and students, is gratefully acknowledged. The time and interest of the members of the thesis guidance committee--Drs. Richard Anderson, Jenny Bond, Richard Leuke and N. E. Tolbert--in following this research are greatly appreciated. Special thanks are due to Professor Loran Bieber, my Doctoral Thesis Advisor, whose consistent support and stimulating input have made this work rewarding and enjoyable. This work was sup- ported in part by Grant AM 18427 from the National Institutes of Health. iii TABLE OF CONTENTS Page LIST OF TABLES .................................................. vi LIST OF FIGURES ................................................. vii LIST OF ABBREVIATIONS ........................................... ix BACKGROUND ON THE CARNITINE ACYLTRANSFERASES .................... l Introduction ............................................... l Assay Methods .............................................. 4 Intracellular Localization ................................. 8 Carnitine Acetyltransferase (CAT) ..................... 8 Carnitine Palmitoyltransferase (CPT) .................. ll Properties of Carnitine Acyltransferases ................... 18 THE PROBLEM--CARNITINE OCTANOYLTRANSFERASE ...................... 27 EXPERIMENTAL PROCEDURES AND RESULTS ............................. 31 Materials .................................................. 31 Methods .................................................... 3l Mitochondrial Isolation ............................... 3l Solubilization of Mitochondria ........................ 32 Carnitine Acyltransferase Assays ...................... 32 Sephadex G-200 Chromatography ......................... 33 Comparison of Beef Liver and Heart Carnitine Acyl- transferase Activities ............................. 33 Purification of COT Activity-Containing Proteins from Beef Heart Mitochondria ....................... 33 CMC Determinations ................................... 35 Interaction of Lauryl-I4C-carnitine with Micelles of Tween-20 ........................................... 35 Other Methods ......................................... 36 Preliminary Investigations ................................. 37 Isolation and Purification of Mitochondrial Carnitine Octanoyltransferase Activities from Beef Heart .......... 61 Results ............................................... 61 Comparison of Carnitine Acyltransferase Activi- ties in Beef Liver and Heart Mitochondria ..... 6l iv Purification of Carnitine Octanoyltransferase (COT) Activity from Beef Heart Mitochondria... 64 Solubilization .............................. 64 Dialysis .................................... 65 Results of Column Chromatography ................. 67 Purification of the COT- and CAT-Containing Blue Sepharose Peak ................................ 67 Fractions 31-34 ............................. 67 Purification of the COT- and CPT-Containing Blue Sepharose Peak ................................ 70 Fractions 40-48 ............................. 70 Molecular Weight Determinations .................. 70 Isoelectric Point ................................ 70 Isoelectric Focusing--Approach to Equilibrium.... 75 Substrate Specificity ............................ 75 Amino Acid Analysis .............................. 83 Discussion ............................................ 83 Effect of Micelles on the Kinetics of Purified Beef Heart Mitochondrial Carnitine Palmitoyltransferase ............ 86 Results ............................................... 86 Discussion ............................................ 99 SUMMARY AND CONCLUSIONS ......................................... l03 APPENDIX A--COLLABORATIVE STUDIES ............................... 105 REFERENCES ...................................................... ll3 LIST OF TABLES TABLE I. Solubilization Experiments ................................ II. Abstract of COT Studies ................................... III. Purification Procedure .................................... IV. Purification Results ...................................... V. Synthesis of Acyl-L-Carnitines ............................ VI. Summary of the Purification of Carnitine Acyltransferase Proteins from Beef Heart Mitochondria ..................... VII. Relative Activities of Beef Heart Mitochondrial Carnitine Acyltransferases .......................................... VIII. The Amino Acid Compositions of Beef Heart Mitochondrial Carnitine Acyltransferases ................................ IX. Rates of Acyl Coenzyme A Formation from Acylcarnitines.... X. Effect of Tween-20 on Kms and Vmax Values for the Forma- tion of Acylcoenzyme A's from Acylcarnitine ............... vi Page 39 48 49 50 59 66 80 84 88 100 a' LIST OF FIGURES FIGURE l. Substrate specificity of carnitine acyltransferase in void volume of DEAE-cellulose experiment ....................... 2. Sephadex G-200 gel filtration chromatography of beef heart mitochondrial carnitine acyltransferase ................... 3. SDS polyacrylamide gel electrophoresis of purified COT.... 4. Subunit molecular weight of beef heart mitochondrial COT.. 5. Egbstrate specificity of purified beef heart mitochondrial 6. pH optimum of beef heart mitochondrial COT ................ 7. Substrate specificity profile of intact and partially l0. ll. l2. l3. T4. 15. fractionated beef heart and liver mitochondrial carnitine acyltransferases .......................................... . Purification of carnitine octanoyltransferase from beef heart mitochondria ........................................ . Chromatography of solubilized carnitine octanoyltransfer- ase from beef heart mitochondria .......................... Height determination of carnitine acyltransferases from beef heart mitochondria ................................... Isoelectric point determination of the carnitine acyl— transferases from beef heart mitochondria ................. Approach to isoelectric equilibrium of the CPT/COT enzyme. Comparison of carnitine acyltransferase activity of beef heart mitochondria with the purified transferase enzymes.. The relationship between the chain length of acylcarni- tines to the critical micellar concentration .............. The effect of Tween-20 on the kinetics of myristoyl-CoA formation from myristoylcarnitine ......................... vii Page 43 45 52 54 56 58 63 69 72 74 77 79 82 90 93 FIGURE Page l6. Association of laurylcarnitine monomers in micelles of Tween-20 .................................................. 96 l7. Double reciprocal plots of reaction velocity versus sub- strate concentration for C8, C10, C12 and C14 carnitine esters .................................................... 98 viii Bis Tris- CAT CM- CMC CoA CoASH COT CPT DEAE- DTNB DTT EDTA HAP HEPES Ki Km MOPS NAD pI POPOP PPO QAE- QO2 SOS-PAGE TNS Tris V max LIST OF ABBREVIATIONS l,3-bis(tris[hydroxymethyl]-methylamino)- Carnitine acetyltransferase Carboxymethylethyl- Critical micellar concentration Coenzyme A Reduced coenzyme A Carnitine octanoyltransferase Carnitine palmitoyltransferase Diethylaminoethyl- 5,5'-dithiobis-(2-nitrobenzoic acid) Dithiothreitol (Ethylenedinitrilo)-tetraacetic acid Hydroxylapatite N-Z-hydroxyethylpiperazine-N'-2-ethanesulfonic acid Inhibition constant Michaelis constant Morpholinopropanesulfonic acid Nicotine adenine dinucleotide Isoelectric pH l,4-bis(2-[4-methyl-5-phenyloxazolyl])benzene 2,5-diphenyloxazole Diethyl-(Z-hydroxypropyl)aminoethyl Respiratory quotient Sodium dodecyl sulfate polyacrylamide electrophoresis 6-p-toluidino-Z-naphthalenesulfonic acid Tris(hydroxymethyl)aminomethane Maximum velocity ix BACKGROUND ON THE CARNITINE ACYLTRANSFERASES Introduction Carnitine (y-trimethylamino-B-hydroxybutyrate) was first isolated in 1905 (l) from mammalian muscle but its significance in muscle and function in metabolism remained unknown for half a century thereafter. Despite its structural resemblance to choline, no role for carnitine as a neurotransmitter appeared possible (2). A compound, later identified ‘IEE. as carnitine (3), was discoverdd by Fraenkel gt_a1 in l948 (4) to be an essential nutrient for larva of the beetle Tenebrio molitor and was referred to subsequently in the literature as vitamin BT' Enzymatic acetylation of p-aminobenzoic acid was found by Friedman and Fraenkel in l955 (5) to be inhibited by carnitine and was apparently due to an enzyme in pigeon and sheep liver extracts which was able to acetylate carnitine by the reaction: 0-acetylcarnitine + coenzyme A === acetyl-00A + carnitine. Four years later Fritz (6) reported the carnitine-dependent stimula- tion of the oxidation of long chain fatty acids by particulate liver preparations. He reported little effect of carnitine on medium chain fatty acid degradation or on the oxidation of palmitoyl-CoA. This carni- tine independence of medium chain fatty acid oxidation has stood the test of time (7, 8). A role for carnitine in palmityl-CoA oxidation was sug- gested by the mitochondrial metabolism of palmitoyl-carnitine reported by Bremer (9) in 1962. Separate reports by Bremer (l0) and Fritz (ll) the l following year noted enzymatic synthesis of palmitoyl-carnitine and the finding of the following reversible reaction: palmitoyl-carnitine + CoA ==¥ palmitoyl-CoA + carnitine. Thus carnitine acyltransferases are defined to catalyze the follow- ing reactions: Forward Reaction acyl-CoA + carnitine + acyl-carnitine + CoASH Reverse Reaction acyl-carnitine + CoASH-+ acyl-6A + carnitine. 7? During this period (1962) Bremer (12) reported that mitochondria H from several rat tissues showed reversible acetylation of carnitine and Fritz (13) in 1963 reported the partial purification of carnitine acetyl— WI - firm X‘A.-:I‘!§l' -. ‘25... transferase (first called carnitine transacetylase (14)) from pig heart mitochondria. The roles of carnitine proposed by these two independent investigators (15, 16) were the same: a means of transporting activated long chain fatty acids across the GOA-impermeable (17) inner mitochondrial membrane to the site of e-oxidation and a means of transporting activated acetyl groups across the membrane. Thus long chain fatty acyl-CoA esters synthesized outside the matrix of the mitochondrion are transported into the mitochondrion by external transfer of acyl groups to form acyl- carnitines, movement of the acylcarnitine across the inner mitochondrial membrane, followed by regeneration of acyl-CoA by internal acyl transfer from carnitine. These roles for carnitine require at least three types of carnitine acyltransferase activity: a long chain transferase (a carnitine palmitoyl- transferase or CPT) accessible to the outer surface of the inner mito- chondrial membrane, a CPT accessible to the inner surface, and a carnitine acetyltransferase (or CAT) with access to the inner matrix of the mitochondrion where acetyl-CoA thioesters are formed from fatty acids, pyruvate and amino acids. The CAT partially purified by Fritz from pig heart did not react with long chain fatty acyl substrates, therefore at least two different proteins were necessary to account for the three activities. The simplest system would require only two enzymes, one for short and one for long chain fatty acyl groups, with each enzyme located h__ in the inner mitochondrial membrane with access to both surfaces, possibly acting as a transport factor as well for the carnitine esters. It is now believed (18) that the forward and reverse carnitine acyl- EAL _‘ transferase reactions involved in converting cytosolic acyl-CoA into mitochondrial CoA ester are not performed by one protein molecule; there is CPT available to the outer surface only and CPT accessible to the inner surface only, the two "separated" in function by a "translocase" protein (19, 20) which catalyses the one-for-one exchange of free carnitine or carnitine esters. The evidence to date (18, 21) is in favor of no extramitochondrial sites of CPT activity in the cell. A proposed role for CAT as an acetyl sink (22, 23), saving activated acetyl groups for fuel and allowing release of mitochondrial CoASH, requires only CAT activity inside the mitochondrion. A more extended role of acetylcarnitine to one like that of citrate, providing net export of acetyl groups for lipid synthesis and acetylation reactions without the ATP expense of the citrate system, would require at least one additional site of CAT activity, either on the external surface of the mitochondrial inner membrane or elsewhere in the cell. Bressler and Brendel in 1966 (24) reported that though their calcu- lations indicated that citrate is the major path for acetyl group movement r 0" .- an out of the pigeon muscle mitochondrion, carnitine showed a stimulation of the production of fatty acids and acetylsulfanilamide from labelled pyruvate without any effect on their production from citrate. In one report (25) comparing the carnitine and citrate transport systems, CAT is concluded to "be the most likely candidate for acetyl group transfer out of yeast mitochondria." me Looking for extramitochondrial carnitine acyltransferase activity A; in liver and kidney, Markwell et a1 (26) found, isolated and partially purified CAT proteins from peroxisomes and microsomes. (Recently, in the yeast Torulopsis bovina, Emaus (27) has reported CAT associated with the ‘5 nuclear fraction as well.) Thus though the role of carnitine in long chain fatty acid metabolism is presently thought to be solely associated with mitochondrial B-oxidation, its role in transfer of acetyl residues must be more diverse. Assay Methods Assay of carnitine acyltransferase activity is by no means standard- ized and has yielded results difficult and occasionally impossible to compare. In the forward direction, the synthesis of (140)-acy1carnitine (28) and acyl-(14C)-carnitine (29) from labelled substrates or the produc- tion of CoASH from acyl-CoA has been measured, the free CoA detected spectrophotometrically by reaction with a sulfhydryl reagent (30) such as DTNB (Ellman's reagent) or DPD (4, 4'-dipyridine disulfide) or measured fluorometrically (31) by a coupled assay with alpha keto glutarate dehydrogenase, producing reduced pyridine nucleotide. The reverse reaction is assayed as the liberation of (14C)-carnitine (32) from labelled substrate or by the detection of acyl-CoA by its characteristic absorption at 232 nm (33) or by the reaction of acyl-CoA with hydroxamate (11). Other assays involve monitoring changes in radioactive specific activity of substrate/product of the two reactions--the "isotope exchange" assay (10, 34)--coupling acyltransferase to fatty acid activation to F‘H measure acylcarnitine production from free fatty acid, and estimating - I transferase activities from flavoprotein reduction or oxygen consumption during mitochondrial B-oxidation of fatty acids, acyl-CoA's or acylcarni- tine esters. Very few investigators (29, 35) have reported carnitine F acyltransferase activities as measured by a variety of techniques for ‘ purposes of completeness or comparison; the usual practice is to perform the assay in one direction by one method. Some of the methods (3) do not allow a simple determination of back- ground acyl-CoA or possible acylcarnitine hydrolase activity. Another, the 232 assay used to measure the reverse reaction, requires few assump- tions, is simple and measures the concentration of reaction product directly without further manipulation and is therefore preferred in deter- mining kinetic characterization of purified transferases. The high absorp- tion of extraneous protein in crude preparations of enzyme precludes use of the 232 assay with intact mitochondria, tissue extracts or during the entire course of an enzyme purification procedure. Reduced CoA-trapping assays which measure the forward reaction are unaffected by extraneous protein but contain a high background absorbance in the presence of sig- nificant concentrations of sulfhydryl agent-reacting substances seen in crude tissue homogenates. These assays must contend with background Q! .;~ '1' up‘ On “1 I“: production of CoASH due to acyl-CoA hydrolases which can account for as much as 90% of the total activity, requiring large numbers of replica- tions and making reliable values difficult to obtain. Reliability of values reported and conclusions drawn about amount and locations of transferases is especially suspect when the method infers amount of acyltransferase activity from amount of product of a f“? subsequent reaction such as the reduction of flavoprotein during B—oxida- tion of acyl-CoA produced in intact mitochondria from external acylcarni- tine or such as the consumption of oxygen linked to carnitine-mediated transport of fatty acyl groups. As an example, in a study by Wood and E; Chang (36) of carnitine palmitoyltransferase (CPT) activity in rat liver and heart mitochondria, comparison of activity of intact organelles react- ing with external sustrates to that seen after detergent disruption gave for liver mitochondria a value of 0.54 for the ratio of CPT on the outer to that on the inner surface of the mitochondrial CoA barrier while the oxygen consumption of palmitoyl-CoA dependent on carnitine versus that of palmitoylcarnitine was 0.56, an excellent agreement confirming the notion of carnitine-dependent palmitoyl-CoA oxidation being a measure of outer CPT while palmitoylcarnitine oxidation measures inner CPT. Under identi- cal conditions, however, rat heart mitochondria showed the same value of 0.56 for 002: palmitoyl-CoA + carnitine/002: palmitoylcarnitine, but the ratio of outer CPT/inner CPT from the detergent study was 1.16. More dramatic are the results of Normann gt El (37) and of Bergstrom and Reitz (38). The former group found in guinea pig brown adipose tissue mitochondria that estimation of "outer" and "inner" CPT by monitoring acyl-CoA dehydrogenase flavoprotein redox level revealed that the apparent 1' OP 'IV ..', ‘1‘; (y initial rate of the inner CPT was 1000 times that of outer CPT. Bergstrom and Reitz found a similar result (a 450 fold difference) using this assay with rat liver mitochondria but also reported that assay of total CPT by the DTNB method before and after treatment of mitochondria with digitonin to remove outer CPT showed that the treatment removed 20-25% of total CPT with the remainder tightly bound and resistant to trypsin digestion (presumably corresponding to inner CPT protected by the intact inner mito- chondrial membrance). Comparison of tissue, animal, and laboratory differences in carni- tine acyltransferases reported is made difficult by the various ways in which the same assay is carried out. Various concentrations of albumin, whose ratio to fatty acid or acyl ester has a significant effect on fatty acid activation (39) and transfer (40), are included in the reaction mix- ture by different investigators. Some values reported are obtained from reactions utilizing dl-carnitine or dl-acylcarnitines. Studies on mito- chondrial CPT (41) and on pure CAT (42) have shown the enzymes to be specific for l-carnitine and its esters and inhibited by the d isomer with a Ki within the ranges employed for assay. The forward and reverse reactions have been performed at 25, 30, 35, 37 and 40° C. There is no agreement on the concentrations of substrates to be included in the incuba- tion. Ionic strength is not uniform and has been shown (43) to have a significant effect on long chain acyltransferase activity. Finally, detergent is present in some reactions while absent in others; various different detergents have been used, with the concentration of detergent in the final reaction mixture often not reported. eat-h 3 AF ".1 b . u— u P:~~ ' v- "u'. Intracellular Localization Carnitine Acetyltransferase (CAT) The localization of CAT in the mitochondrion has been studied by histochemistry/electron microscopy, by comparison of the mitochondrial oxidation of acetylcarnitine to that of acetyl-CoA + carnitine, and by membrane disruption using detergents, sonication, osmotic shock and freeze-thaw techniques. The electron microscopic method "observes" the electron dense product of reaction between uranyl acetate, potassium ferri- cyanide, and free CoASH released from acetyl-CoA in the presence of carnitine. Investigations of rat heart (44) and mouse skeletal muscle (45) have localized CAT by this technique to the space between the inner and outer mitochondrial membranes and to the outer surface of the inner mem- brane, respectively. None was seen by either group associated with the inner surface of the inner membrane or in the mitochondrial matrix. An opposite result is reported for isolated intact mitochondria of liver and mammary gland of goat, guinea pig and rat by measuring CAT activity before and after disruption of the mitochondria with detergent or freeze-thaw. In this study (46) on average only 7.5% of total CAT activ- ity was attributable to an outer CAT enzyme. Snoswell (47) has reported similar results for sheep liver, heart, skeletal muscle and kidney cortex where at least 90% of mitochondrial CAT was found to be latent. Comparing intact mitochondria to those treated with digitonin, Solberg (34) reported little or no outer short chain acyltransferase for mitochondria of rat and mouse liver, though the results with calf liver mitochondria were not as clear. She also noted that the 13-fold increase in mitochondrial CAT induced by clofibrate treatment was an increase in inner CAT only and not reflected in CAT activity on the outer surface of the mitochondria. Using digitonin to strip away the outer membrane of rat liver mito- chondria and to solubilize proteins from the outer surface of the inner membrane not integrally bound, Brdiczka gt al (48) noted the major part of CAT in the inner mitochondrial space while at least 25% of the total CAT was assigned to an outer compartment, one containing adenylate kinase, a marker for the space between the inner and outer mitochondrial membranes. After treatment with digitonin, these workers reported that the rate of oxidation of acetyl-CoA + carnitine was significantly decreased, confirm- ing that an outer CAT activity had been removed, one which when present is responsible for carnitine-dependent oxidation of acetyl-CoA. Likewise in blowfly flight muscle mitochondria separate investigators (49, 50) have reported the total absence of outer CAT based on a lack of oxidation of acetyl-CoA + carnitine by these organelles but a very high oxidation of acetylcarnitine. Harshaw (51) however found that though mitochondria from bovine fetal heart oxidized acetylcarnitine but not acetyl-CoA + carnitine, mitochondria isolated from calf heart were not as impaired, suggesting a deficiency of CAT outside the mitochondria only in early development. Similar results for rat heart mitochondria were reported by Tubbs and Chase (52) who found oxidation of acetyl-CoA + carnitine as well as of acetylcarnitine but of acetylcarnitine only if the mitochondria had been preincubated with bromoacetyl-CoA, a proposed irreversible inhibitor of CAT which is unable to cross the CoA barrier of intact mitochondria. They were able to inhibit both oxidations by using bromoacetylcarnitine. 10 They concluded that there are two pools of CAT, one inner and one outer, and that some preparations of mitochondria have lost the outer CAT during the isolation procedure. The association of even the inner CAT with the mitochondrial mem- brane is not a tight one; it is more "membrane-associated" than "membrane- bound". This was shown by Beenakkers and Klingenberg (53) who,by repeated extraction without membrane solubilization by the use of detergent or extensive sonication,were able to fully solubilize CAT from mitochondria of rat heart and locust flight muscle. Freeze-thawing mitochondria at appropriate ionic strength gave the same result for Barker gt 21 (46) with mitochondrial CAT from liver and mammary gland of goat, guinea pig and rat. In other organelles, Markwell gt_al (26) found rat liver peroxisomal CAT to be free in the organelle interior and released by its breakage while the liver microsomal CAT was membrane-associated but solubilized by 0.4 M KCl. These two CAT proteins were shown to have identical chromato- graphic, physical and kinetic properties; the difference in their solu- bilization may reflect the difference in the microsomal and peroxisomal membranes, the latter being a unilaminar, fragile structure. Thus the in _s_i_i_:_u evidence from electron microscopy and that from most studies with isolated mitochondria are clearly in conflict. More definitive proof of the existence or absence of an outer mito- chondrial CAT may be lacking due to a lack of theoretical necessity for its presence. Extra-mitochondrial sites of CAT activity in liver, kidney and heart that have been shown by Markwell and others may account for non- mitochondrial metabolism of acetylcarnitine. The presence of fatty acid ll synthetases for short and medium length fatty acids in the mitochondrial matrix and the permeability of the inner mitochondrial membrane for these groups would allow mitochondrial utilization of the shorter fatty acids without the carnitine-mediated system of transport. Carnitine Palmitoyltransferase (CPT) Though the inner mitochondrial membrane is permeable to free fatty acids, unlike the short and medium length fatty acids (54), long chain fatty acids are not activated to CoA esters within the inner compartment of the mitochondrion. Rather, long chain acetyl-CoA synthetases are found associated with microsomes (55) and the outer mitochondrial membrane (56). Therefore for carnitine to mediate the conversion of cytosolic long chain fatty acyl-CoA to mitochondrial acyl-CoA a long chain carnitine acyltrans- ferase must be present on the outer surface of the inner mitochondrial membrane or elsewhere in the cell. Careful intracellular distribution studies by Hoppel (35) and Markwell gt £1 (21) have shown carnitine palmitoyltransferase to be exclu- sively mitochondrial. The results of some early reports of CPT activity in microsomes as well as in mitochondria were later amended after more careful study to exclude a microsomal location (111, 112). The report of Fogle and Bieber (57) of CPT activity in rat heart microsomes has not been confimed or refuted; research into the properties of heart micro- somes is still very new. In 1965 Norum (58) presented data showing that less than 10% of total cellular CPT was extra-mitochondrial but that the amount of extra- mitochondrial CPT increased with diabetes, fasting or a high fat diet. PL. 12 More recently, Farrel gt 31 (59) have found that extra-mitochondrial CPT is observed in liver from mice treated with clofibrate and that this cytosolic CPT can constitute as much as 40% of the total cellular CPT. One possibility raised by these investigators to explain this extra- mitochondrial CPT is that it may represent a precursor form of mitochon- drial CPT which eventually contributes to the 3-5 fold increase in mito- chondrial activity reported by Kahonen, Markwell and others (60, 61, 62) after clofibrate treatment. A precedent for an active precursor form of a mitochondrial enzyme is the model presented recently by Kolattakuddy (63) for mitochondrial malonyl CoA decarboxylase. If it is agreed that the vast majority, if not all of cellular CPT is associated with the mitochondrion, how it is distributed with respect to the inner mitochondrial membrane is still a matter of dispute. A digitonin fractionation and sonication study of rat liver mitochondria by Hoppel gt a1 (35) showed between 15 and 30% of total CPT to correspond to outer CPT. Bergstrom and Reitz (38) have found 20-25% of rat liver mitochondrial CPT digitonin extractable, the rest tightly bound. Yates and Garland (64) also found 20% of total rat liver mitochondrial CPT to be outer CPT; this value was in agreement with the amount of CPT they observed solubilized from the mitochondria by sonication. Using the detergent Lubrol, Harano (65) found a 1:2 ratio of outer to inner CPT in rat liver mitochondria. A 1:2 ratio was seen using digitonin fractionation by Layzer EE.§1 (66) in mitochondria from normal and CPT deficient human sksletal muscle. Hood and Chang (36) using the detergent Triton X-100 reported a 1:2 ratio also for rat liver mitochondria but a 1:1 ratio for rat heart mitochondria assayed under identical 13 conditions. Bieber et 31 (67) reported approximately equal amounts of CPT on the two sides of the mitochondrial CoA barrier in livers of new- born, one-day and five-day piglets. A 1:1 ratio is also reported by Patten gt a1 (68) for mitochondria of normal human skeletal muscle. Based on the proposed selective sensitivity of the outer CPT for the inhibitor malonyl-CoA, McGarry £3.21 (69) concluded that one half of the total CPT is outside the rat liver mitochondrion; with sufficient malonyl- CoA, half of total CPT remained when palmitate oxidation, dependent on outer CPT, was totally inhibited. Finally, Swierczynski £3.21 (70) reported that mitochondria from human term placenta oxidized palmitoyl- CoA + carnitine at half the rate of palmitoylcarnitine, implying a 1:2 ratio of outer to inner CPT. Digitonin fractionation of mitochondrial compartments yields an underestimate for the proportion of total CPT on the outside of the inner membrane because the inner membrane itself begins to dissolve and leak interior marker enzymes before all of the "outer“ CPT is extracted. Thus a concentration of digitonin giving a fraction of total CPT clearly defined by outer markers such as adenylate kinease or monoamine oxidase and containing none of the soluble interior enzymes such as fumarase or glutamic dehydrogenase will not have removed all of the external CPT. The fraction of CPT outside the inner mitochondrial membrane is concluded to be between 20 and 50% of the total activity with liver mito- chondria and heart and skeletal mitochondria, respectively, possibly representing these extremes. As with CAT,which is proposed by Tubbs and Chase (52) to be easily lost during preparation, the lowest values of outer CPT may represent a partial loss of the outer enzyme during ¢""71‘ L:‘I" 'f‘.‘r 1r», .I’P. 'u l4 mitochondrial isolation. Changes in the outer CPT/inner CPT ratio have been observed during development and as a result of infectious or genetically acquired disease. In many animals a rise in mitochondrial fatty acid oxidative capacity is noted during suckling and attributed (71) to the lipid content of milk with lowest levels in late gestation immediately before birth and also a decline to adult levels after weaning. Augenfeld and Fritz (72) have shown that changes in total liver mitochondrial CPT parallel those of oxidative capacity. Increases in liver mitochondrial CPT have also been reported during fasting (73), diabetes (74), and high fat diets (75), conditions associated with an increased concentration of circulating tri- glycerides and fatty acids. Wolfe gt al (76) has reported that high-fat diet-associated increased mitochondrial CPT activity in neonatal pigs is not specific to the transferase enzyme but reflects a higher mg mitochondrial protein per g wet tissue weight seen with the treatment. Perinatal induction of enzymes necessary for fatty acid utilization has been shown by Aprille (77) to be relatively independent of dietary intake with identical increases in fatty acid oxidation being seen with rabbits nest-reared on mother's milk, formula-fed a diet containing 6% lipid or an equicaloric one containing no fat. In rabbits the rate of oxidation of octanoate and laurate was found to be equal to that of their carnitine esters. There was only a two-fold increase in octanoate and laurate oxidation dur- ing the first four days of life while four-fold increases were observed for oxidation of the corresponding carnitine esters and of glutamate- malate. This suggests a lag in the development of either an 15 intramitochondrial fatty acyl synthetase for these medium chain fatty acids or of external carnitine acyltransferase activity which may exert significant control on their metabolism. Similarly in the rat, based on carnitine-dependent CoASH release from palmitoyl-CoA in the presence and absence of detergent, Harano (65) reported adult levels of inner CPT but virtually absent outer CPT in the early gestation period. Outer CPT is seen to rise above adult levels after birth during suckling. Changes in palmitate oxidation, which is barely detectable in the fetus, are reported to parallel those of outer CPT activity. Tomec and Hoppel (78) reported palmitoyl-CoA + carnitine oxidation as 2-14% of palmitoylcarnitine oxidation in bovine fetal heart mitochondria Inn: did not attribute the difference to a deficiency of outer CPT at this stage of development, citing instead an abnormal CoA saturation curve of fetal heart mitochodrial transferase activity. Likewise Bieber gt a1 (67) presented evidence that CPT activity is equiv- alent on each side of the mitochondrial CoA barrier in newborn, one-day and five-day piglets despite the finding that though 24 hour old piglets had total CPT equal to that of adults (as measured by the isotope exchange assay), they oxidized palmitoyl-CoA at half the adult rate. Pace and Nannemacher (79) recently reported a decrease in outer CPT with no change in inner CPT in liver mitochondria of rats infected with Streptococcus pneumoniae. Decreased utilization of fatty acids caused by systemic carnitine deficiency may trigger the same cellular response as that seen with high lipid levels induced by fasting, fat-feeding or diabetes to give the abnormally high CPT activity reported by Boudin gt 31 (80) in liver, v11} :Tve A? {.3 '— ‘- .‘P; '1‘ b A’>.F — VJ.- ‘11:»: 41,. Il‘“ Cy~ C". (— O 16 muscle, myocardium and kidney epithelium of a woman who died of progres- sive muscle weakness due to carnitine deficiency. More recently, Scholte .gtqal (81) reported a case of a girl with systemic carnitine deficiency who died in acidosis whose inner CPT and palmitoyl-CoA synthetase levels were increased. A younger sister with decreased muscle and blood carni- tine was found to also have increased muscle inner CPT but to show normal outer CPT. The ratio of CPT outside to inside the mitochondrion may depend on the type of tissue, stage of development, concentration of carnitine, and possibly on the dietary or hormonal state. Hereditary deficiency of CPT activity therefore might be expected to present in various forms, depend- ing on the tissues and membrane locations affected. In the first reported case of carnitine palmitoyltransferase deficiency described in 1973 by DiMauro and DiMauro (82), the now-classic sign of CPT deficiency, recur- rent paroxysmal myoglobinuria, led to assay of CPT in muscle only, with the finding of very low (0-20% of normal) muscle CPT as measured by three CPT assay methods. A greater impairment of the isolated mitochondria to utilize palmitate than to oxidize palmitoylcarnitine in the presence of normal palmitoyl-CoA synthetase lead to a conclusion of a more severe defect of outer CPT than of inner CPT. Hostetler gt a1 (83) reported finding a complete lack of outer CPT with normal inner CPT in muscle mito- chondria of a patient with recurrent myoglobinuria. He also reported that the fasting plasma concentration of ketone bodies increased normally, an indication of normal liver CPT activity. Two additional patients (68, 84) studied with decreased muscle CPT but normal ketogenic capacity gave a conflicting pattern of CPT deficit: 17 normal outer CPT but deficient inner CPT in muscle mitochondria. One of these patients (84) was also tested for CPT in leukocytes with the same finding of normal outer CPT and decreased inner CPT. Leukocyte, platelet, fibroblast and skeletal muscle CPT were all decreased to 23-39% of normal levels in a patient recently described (66) presenting with recurrent paroxysmal myoglobinuria but normal ketones upon fasting. This patient appears to have a deficit of both inner and outer CPT activities: the same fraction of total CPT is extracted with digitonin from the patient's fibroblast mitochondria as from those of normals; the inner and outer CPT activities separated by detergent were each deficient to the same degree. Other CPT deficient individuals have been described, some with normal ketogenic function (85) and some with inadequate increase in ketones with fasting (86, 85), but with no attempt made to distinguish outer from inner CPT deficiency. That a liver CPT deficiency is responsible for the low fasting ketone production in the latter patients is shown by the observation (86) of a prompt ketonemia after ingestion of medium chain triglycerides. The prominent symptom with or without presumed liver CPT deficiency inferred from decreased ketogenesis is recurrent myoglobinuria. A dis- tinguishing symptom between the two may be greater muscle pain associated with the ketone insufficiency which was alleviated in one patient (85) by B-hydroxybutyrate. Liver biopsy for determination of carnitine biosynthetic enzymes in patients with systemic carnitine deficiency has been performed and shown no deficit of the enzymes catalyzing the conversion of trimethyllysine to carnitine (99). Post-mortem findings of CAT deficiency with normal rm ) Ill 18 CPT levels in liver and other tissues of a child who died of apparent neurologic and liver dysfunction led to a suggestion of a functional defect of acetyl-CoA utilization in brain mitochondria as a result of CAT deficiency. To date there have been no reports of liver biopsy to confirm the proposed hepatic CPT deficit in patients with recurrent paroxysmal myoglobinuria who are observed to have a diminished ketogenic response to fasting. One might also expect eventually to find an indi- vidual with deficient liver CTP and diminished ketogenic response but with normal muscle CPT activity. Properties of Carnitine Acyltransferases Substrate concentrations corresponding to half maximal activity in either reaction direction for purified pigeon breast muscle CAT are reported (87) to be independent of the concentration of the second sub- strate present, therefore a random order of addition of substrates is concluded. The catalysis by CAT of the formation of S-carboxymethyl-CoA- (-)-carnitine ester from CoASH and bromoacetyl-(-)—carnitine (88) sug- gests the formation of a ternary complex; it has been proposed (87) that the interconversion of such ternary complexes may be the rate limiting step for the reaction. Other CAT enzymes may not be as simple: acetyl- carnitine is reported by one group (114) to inhibit purified rat liver mitochondrial CAT, this inhibition is reversed by carnitine but not by acetyl-CoA; double reciprocal plots of forward reaction activity by yeast CAT (25) indicate an ordered addition of the substrates carnitine and acetyl-CoA. 19 The reaction mechanism of CPT may also be a simple random addition of substrates as concluded by Edwards (89) for outer CPT, but interpreta- tion of results reported is complicated by the hydrophobicity of the palmitoyl esters of carnitine and CoA and of free CoASH whose critical micellar concentrations have been observed to be 15 AU, 3-4 uM_and 30 pH, respectively. Unlike the yeast CAT, calf liver mitochondrial CPT purified by Kopec and Fritz (90) shows no effect of carnitine concentration on the Km for acyl-CoA. The other three substrates, however, palmitoyl-CoA, CoASH, and palmitoylcarnitine, do affect their respective second substrates by raising their Km's. Similar inhibition by palmitoyl-00A has been reported by numerous researchers. In 1967 Bremer and Norum (32) proposed that besides being a substrate for the enzyme, palmitoyl-CoA also acted as a competitive inhibitor for carnitine with a Ki (3 x 10'6) lower than its Km (10-5) as a substrate. They concluded in a separate study on the effect of deter- gents on CPT (91) that the major effect of detergents was to prevent palmitoyl-CoA from acting as a competitive inhibitor of carnitine and acylcarnitine, with less effect on its function as a substrate in the reaction. Other reactions controlling the rate of fatty acid oxidation and energy production for which palmitoyl-CoA has been proposed to be a significant physiological inhibitor are palmitoyl-CoA synthetase (39), the mitochondrial citrate transporter (92), and adenine nucleotide translocase (93). Changes in mitochondrial CPT activity with ionic strength-—a near linear rise in CPT with increasing ionic strength up to 0.06 Mf-was correlated by Wood (94) with a parallel linear increased association of 20 palmitoyl-CoA with mitochondria in this range of ionic strength. She found (43) that CPT released from the mitochondrion by digitonin was not affected by changes in ionic strength, showing that the hydrophobic environment provided by the detergent differs significantly from that provided by the outer surface of the mitochondrial membrane. Similarly she attributed (95) changes in kinetic properties of heart mitochondrial CPT resulting from chronic ischmia as follows: "As a result of ischemia, changes in the lipid components in the membrane containing carnitine CPT were postulated: alterations in the hydrophobic environment of the enzyme produce interference in the binding of palmitoyl-CoA to the second sub- strate site and may result in a decrease in the Km of the enzyme for carnitine." The nature of the interaction of another CoA ester, malonyl-CoA, with the outer mitochondrial CPT is also dependent upon the membrane association of the enzyme: McGarry and Foster (69) have reported that treatment of rat liver mitochondria with the detergent Tween-20 releases a malonyl-CoA insensitive CPT activity which they conclude is outer CPT which requires membrane association to be inhibited by malonyl-CoA. They have proposed (96) that malonyl-CoA and acetyl-CoA carboxylase may be largely responsible for control of fatty acid metabolism by the effect of the CoA ester on the outer CPT. Because malonyl-CoA is not a sub- strate for CAT and because the mitochondrial matrix contains a malonyl-CoA decarboxylase, the concentration of malonyl-CoA inside the mitochondria should not reflect that of the cyt0plasm, so that inner CPT is not regu- lated by the concentration of this intermediate of fatty acid synthesis. 21 McGarry and Foster report that inner CPT is insensitive to malonyl- CoA when the lipid environment of the membrane is undisturbed by detergent but the membrane's integrity is disrupted by osmotic shock, freeze-thawing, or sonication to allow the otherwise impermeable malonyl-CoA access to the inner CPT. But as with outer CPT, this enzyme also changes with Tween-20 treatment, becoming more sensitive to malonyl-00A, in contrast to the opposite effect of the detergent on the outer CPT. Fritz (41, 97) has reported that only membrane-bound CPT is inhibited by the d isomer of palmitoylcarnitine. Analogous to the ischemia-induced alteration of palmitoyl-CoA binding and outer CPT kinetics attributed by Wood to changes in the lipid environment of the inner mitochondrial membrane, starvation may influence the lipid composition and membrane environment of outer CPT to affect its interaction with malonyl-CoA. Such an influence would explain the recent finding of Cook_gt.al (98) that starvation increases the apparent Ki of rat mitochondrial CPT for malonyl-CoA. If changes in the membrane environment can change the kinetic properties of carnitine acyltransferase, perhaps the difference between the hydrophobic environment provided by the outer surface of the inner mitochondrial membrane and that provided by the inner surface could account for the difference in properties reported for outer and inner CPT's. (Perhaps tissue and species differences in mitochondrial CPT kinetics can be explained by differences in the lipid and protein composition of mito- chondria from different sources.) Acyl group specificity of carnitine palmitoyltransferases has been reviewed by Hoppel (18). He includes results received by personal com- munication from Edwards and Tubbs on the substrate specificity of inner 22 and outer CPT's. Portions of these data have appeared since in an abstract (89). The findings show that outer beef liver mitochondrial CPT (purified 850-fold to near homogeneity in the abstract) assayed in either direction has greatest activity with palmitoyl or myristoyl CoA or carnitine esters but has significant activity with nearly all even chain acyl substrates with values of 41 and 45% (relative to palmitoyltransferase) found for hexanoyl-CoA and hexanoylcarnitine, respectively. The inner CPT cited by Hoppel from the same data source, presumably the other activity found in the beef liver mitochondria, is much more specific for long chain acyl groups with relative activities of 15, 70 and 100 for octanoyl—, lauryl- and palmitoyl-CoA when assayed in the forward direction. Three years earlier in 1971, Tubbs with West and Chase (113) reported a similar separation of inner and outer CPT from beef liver mito- chrondria but with a different substrate specificity for the inner CPT when assayed for the reverse reaction with the 232 assay. In this direc- tion the inner CPT gave equal Vma values for palmitoyl- and oxtanoyl- x carnitines and activity 50% higher for laurylcarnitine. It is possible that the specificity of inner CPT from beef liver differs for CoA and carnitine esters, however it is more likely that the much greater Km's for medium chain acylcarnitine substrates reported in the earlier study explain this discrepancy. Aside from the differences in substrate specificity, beef liver inner and outer CPT were reported to differ in the measures required for their solubilization, in their interaction with bromopalmitoyl- CoA (it inhibited the isolated outer enzyme but was treated as a substrate by the inner CPT), their chromatographic properties during ion exchange, and by their isoelectric points. 23 Kopec and Fritz (97) also isolated two CPT enzymes from beef liver mitochondria in 1971 and they purified one of them, designated CPT I, to homogeneity. This enzyme was specific for palmitoyl-and myristoylcarni- tine with activity decreasing quickly below substrate carbon length of 14 carbons so that the relative activity of octanoylcarnitine was less than 2%. The protein fraction containing the other enzyme, CPT II, was even more specific for long chain fatty acylcarnitine with the rate for myristoyl transfer less than 20% that for palmitoyl-or stearoylcarnitine. Both CPT I and CPT II were more specific for long chain acyl substrates than the CPT's of West gt a1_(113) though all came from the same tissue source; they differed more in substrate specificity from those of West- .gt_al than the latter enzymes differed from each other. Assignment of CPT I and CPT II to sides of the inner mitochondrial membrane was not defined by their isolation procedure and was achieved by the production of antibodies against pure CPT I and their subsequent use to inhibit activity of soluble CPT I and CPT II and to inhibit CPT activ- ity on the beef liver mitochondrial surfaces. In this way CPT I was shown to be the outer enzyme and CPT II to be on the inside. It was found however that CPT II, defined kinetically and by immunological cross reactivity, could be generated from CPT I by treatment with denaturing agents such as urea and guanidium chloride. CPT II and I were originally obtained by elution from calcium phosphate gel in the absence and presence of the detergent Tween-20, respectively. Incubations for their assay did not contain additional detergent so that differences in the environments of the two enzymes during assay may account for their kinetic differences. 24 Similarly the outer and inner CPT activities isolated by West gt a1 were obtained from a lead acetate precipitate of the 20,000 g supernatant of a homogenate of frozen liver and from a butanol extract of the 20,000 g pellet, respectively. Again, the environments of the two enzymes may be sufficiently different to yield differing kinetic profiles, or one enzyme may be a partially denatured form of the other. Differences in chromatographic behavior and isoelectric point of these inner and outer enzymes may also be a function of the method of their isolation due to the presence of contaminating lipids. A relatively hydrophobic protein with adhering phospholipids would be expected to have a lowered measured pI than in their absence, with the value of the pI depending on the relative abundance of different phospholipids. CPT stud- ied by Nest 33 él_h39 major peaks of activity with pI's of pH 4.8 and 5.7. There is a tendency for mitochondrial proteins to have pI's near pH 8. Markwell (100) reported that CAT isolated from rat liver peroxisomes and microsomes and that of pigeon breast muscle had pI values of pH 8.3, 8.3, and 7.9 respectively. Members of the West group published a few years later a study (101) on the question of multiple forms of carnitine acetyltransferase which had been reported in the literature. They found that they were able to observe two forms of CAT in extracts of various animal tissues and attrib- uted these to free and membrane-associated CAT with membrane association responsible for differing apparent molecular weight and isoelectric point. They concluded that the two forms were freely interconvertible with similar kinetic properties and suggested the existence of only a single type of CAT. 25 Similar kinetic properties of soluble and particle-bound enzyme were reported by Bremer and Norum in 1967 (40) for rat liver mitochondrial CPT. More recently Bergstrom and Reitz (38) have reported the similar nature of inner and outer CPT from rat liver mitochondria. After separa- tion of outer and inner activities by digitonin the two enzyme fractions were subjected to identical purification procedures involving Tween-20 extraction, gel filtration and ammonium sulphate precipitation. Binding of each protein to detergent micelles is apparently responsi- ble for the 430,000 dalton molecular weight seen for each on gel filtra- tion. Because of the masking effect of the detergent micelle, no differ- ence in molecular weight such as that noted between the outer CPT purified by Edwards (89) (50,000) and CPT I of Kopec and Fritz (75,000) could be evaluated. Molecular weight estimates for CPT I given by the later group do not inspire confidence, however. From the near total exclusion of CPT I from a P-150 gel filtration column in the presence of the detergent Tween-20, they subsequently reported CPT I to be a dimer of molecular weight 150,000, a conclusion with little justification considering the expected association of the protein with detergent micelles and the inate unreliability of estimating molecular size of an excluded molecule. Though the enzymes isolated by Bergstrom and Reitz were not puri- fied sufficiently to allow monomer molecular weight estimate, their Km values for the substrates of forward and reverse reactions are nearly identical as are the relative rates of the two enzymes in the two direc- tions. From their results and the absence of a more definitive study of the physical and kinetic properties of CPT from inner and outer surfaces 26 of the inner mitochondrial membrane, the possibility cannot be excluded that the same protein may catalyse the transferase reaction at the two locations. THE PROBLEM--CARNITINE OCTANOYLTRANSFERASE It was probably Ephriam Racker who said, "One clean experiment is worth a thousand dirty calculations." Along this line, H. Solberg stated as a dedication/credo to her doctoral thesis on the "Acyl Group Specific- ity of Carnitine Acyltransferases" (102) the words,"Don't waste clean thoughts on crude enzymes." But she accompanied this thought with a second, "Life is too short for enzyme purification." Among the fruits of her labors in the middle ground between the frustrating extremes of the uncertainty of working with crude enzymes and the tedium of enzyme purifi- cation, was evidence for the existence of a third carnitine acyltrans- ferase, carnitine octanoyltransferase (COT). Carnitine palmitoyltransferase as purified to homogeneity by Kopec and Fritz from calf liver (97) was seen to be quite specific for the transfer of long chain fatty acyl residues with activity dropping dramatic- ally with decreasing substrate carbon length below myristoylcarnitine so that the octanoyltransferase activity of this enzyme was reported as less than 2% of that of the palmitoyltransferase. Carnitine acetyltransferase purified to homogeneity from pigeon breast muscle by Chase gt_gl (103) was reported to be correspondingly specific for short chain acyl residues with a precipitous drop in activity of the pure enzyme with substrates of more than four carbons. Solberg showed that a commercially available preparation of CAT (104) revealed a second peak of activity with maximal activity for acyl 27 28 groups between six and nine carbons, apparently containing a co-purifying contaminant carnitine acyltransferase specific for medium-chain acyl esters. She found evidence for the presence of this COT in substrate specificity profiles of carnitine acyltransferase activity in mitochondria of various rat tissues and reported that the optimum chain length of sub- strates for the inner pool of carnitine acyltransferase activity of rat, mouse and calf liver mitochondria (34) is seven carbons while that of the outer mitochondrial transferase of rat and mouse liver was nine or ten carbons. Thus the existence of a separate carnitine acyltransferase pro- tein specific for medium chain length fatty acid groups was proposed and its purification left to others. During the isolation of homogenous CPT from calf liver mitochondria, Kopec and Fritz (97) noted fractions differing in substrate specificity from the published CAT profile and from that of CPT when pure. One of these fractions not pursued further was an extract of calf heart mito- chondria with activity centered about a peak of the substrate octanoyl— carnitine. Markwell gt al (105) reported that rat liver mitochondria con- tained six times as much COT as CAT activity and that COT activity was present in amounts equivalent to CAT in peroxisomes and microsomes of rat liver. These organelles were devoid of CPT and the CAT subsequently purified from them (26) was clearly separated from COT activity. (Subsequent studies by Valkner and Bieber (106) have shown that in micro- somes, the location of the two enzymes is different, with CAT associated with both sides of the microsomal membrane and CDT on the cytosolic sur- face exclusively.) Attempts by Markwell gt 31 (26) to isolate and stabilize COT activity from these organelles were unsuccessful. 29 Medium chain fatty acids are produced by the mammary glands of some mammals such as the goat (107) but are otherwise not plentiful in the diets of the organisms in which COT has been detected. In fact, the provi- sion of medium chain fatty acids and triglycerides for individuals with genetically impaired transfer of long chain fatty acyl groups is an expensive therapy due to their relatively low natural abundance. It has seemed unlikely until recently that medium length fatty acids or acyl residues have any role in normal metabolism other than as short lived intermediates in fatty acid synthesis or B-oxidation. The recent possible exception involves B-oxidation discovered in liver peroxisomes (108) during which there is evidence that the breakdown of long chain fatty acyl-CoA to acetyl-CoA may be halted before its com- pletion (109) so that medium chain fatty acyl groups are the end product. If CoASH is to be recovered from the acetyl and medium chain acyl esters and if these groups are to be utilized elsewhere in the cell, a function for a mitochondrial COT enzyme can be inferred as well as those of peroxi- somal CAT and COT. This perosizome-associated role for mitochondrial COT had not been proposed when the present investigation was initiated, however, and many workers in the field (110) discounted the significance of COT activity and the existence of a separate medium chain acyltrans- ferase in mitochondria. A reasonable argument could be made for the evidence of the existence of mitochondrial COT being artifactual, consider- ing the uncertainty involved in trying to apply clean thoughts to crude enzyme preparations. It was decided that at the present stage of knowl- edge and inquiry into the roles of carnitine and the carnitine acyltrans- ferases, life was not too short for the enzyme purification required to 30 either confirm the existence of mitochondrial COT or to clarity why one appears to exist. EXPERIMENTAL PROCEDURES AND RESULTS Materials Coenzyme A and coenzyme A-esters were from P-L Biochemicals. Carnitine was a generous gift from Otsuka Pharmaceutical Co. and Sigma Tau farmaceutici. Ampholines and hydroxylaptite were from Bio-Rad. Sephadex and Sepharose chromatography media were from Pharmacia. Fluorescamine, Tween-20, Triton X-100 and 5,5'-dithio-bis-(2-nitrobenzoic acid) were from Sigma. DL-(methyl-14C)-carnitine hydrochloride was from Amersham. 2-p-Toluidinylnaphthalene—G-sulfonate (TNS) was from Eastman. Methods Mitochondrial Isolation Beef hearts were obtained from a local abbatoir. Immediately upon removal from the carcass, the heart was sliced and packed in ice for transport to the laboratory. All subsequent procedures were performed at 4°C. The ventricle was cleaned of all fat and connective tissue and cut into thin slices. Batches of 60 g of ventricle were homogenized in 500 ml of Sucrose Isolation Buffer (0.25 M sucrose, 5.0 mM HEPES, 0.25 mM EDTA, pH 7.7) for 45 seconds at high speed in a Haring blender. To insure more complete release of mitochondria, this was followed by a 30 second Polytron homogenization using the large probe in batches of 220 m1. Mitochondria were isolated from the homogenate by differential 31 cefitfl fU x 9 High spe + reagent Strokes 32 centrifugation employing 15 minute spins at 500,15,000, 500, 11,000 and 7000 x g in a Sorval RC-2 centrifuge using the 55-34 and the GSA rotors. High speed pellets were resuspended in Isotonic Sucrose Buffer (0.25 M reagent grade sucrose, 2.5 mM HEPES, 0.25 mm EDTA, pH 7.5) using two strokes of a loose fitting Potter-Elvejhm Teflon-glass homogenizer. The 7000 x g pellet was resuspended in a minimal volume of isotonic sucrose buffer, assayed for enzyme activities and protein, and stored at -80°C. Solubilization of Mitochondria Mitochondrial suspensions from five beef hearts were thawed, pooled and added to one half volume of 3.M KCl in 6% Triton X-100. The suspension was inverted for mixing and then treated batchwise with six passes of the Teflon-class homogenizer at 15°C. After centrifugation for 90 minutes at 89,000 x 9 (29,000 rpm) in the Type 30 rotor, a clear apricot-colored supernatant fluid containing solubilized mitochondrial protein was pipetted from between floating lipid and the flocculant surface of the mitochondrial pellet. Carnitine Acyltransferase Assays The forward reaction was assayed at 412 nm by the DTNB method of Bieber‘§t_al.(30). The 200 ul reaction volume contained 115 mM'Tris-HCl, 1.1 mM EDTA, 0.1% Triton X-100, 1.25 mM.1-(-)-carnitine, 250 A! DTNB and 100 HE acyl CoA at pH 8.0 and 25°C. The reverse reaction was assayed by the method of Srere gt 11 (33) in which CoA thioester bond formation is monitored at 232 nm. The 200 ul reaction volume contained 0.2 M Tris-HCl, 1 mM dithiothreitol (DTT), 0.5 mM EDTA, 0.1% Triton X-100, 60 uM_CoASH and 500 uM_acylcarnitine at pH 7.45 and 35°C. 33 Sephadex G-200 Chromatography Solubilized protein from beef heart mitochondria (10 ml) was chroma- tographed on a 90 cm x 1.8 cm column of Sephadex G-200 equilibrated with Isotonic Sucrose Buffer containing 1.0% Triton X-100 and l M_KCl; 4.0 m1 fractions were collected. Catalase and hemoglobin were similarly chromatographed as standards. Void and bed volumes were determined using Blue Dextran and K3Fe(CN)6, respectively. Comparison of Beef Liver and Heart Carnitine Acyltransferase Activities Heart mitochondria were isolated and solubilized as described above. Beef liver was treated in an identical manner with two exceptions: (l) the ratio of liver tissue to Sucrose Isolation Buffer used for homo- genization was 140 g: 420 ml and (2) the polytron homogenization/sonica- tion step was omitted. The solubilized protein solutions from liver and heart mitochondria were chromatographed separately on a 112 cm x 4.8 cm column of Sephadex G-100. Samples of 100 ml containing approximately 15 mg protein/ml were applied to the columns and 24 ml fractions were collected. Purification of COT Activity-Containing Proteins from Beef Heart Mitochondria Beef heart mitochondria were solubilized as described above. The mitochondrial protein solution was equilibrated with Blue Buffer (2% Triton X-100, 2.5 mM HEPES, 0.25 mM EDTA, 60 mM KCl, pH 7.5) by exhaustive dialysis and applied at a flow rate of 130 ml/hr to a 75 cm x 4.1 cm column of Cibacron Blue Sepharose 4B equilibrated with Blue Buffer. 34 Four bed volumes of buffer were then passed through the column to wash off unbound proteins before a 1500 ml linear gradient of 60-860 mM KCl in Blue Buffer was used to elute carnitine octanoyltransferase (COT) activity. Fractions eluting from the Blue Sepharose column containing both COT and carnitine acetyltransferase (CAT) activities (fractions 31-34) were pooled and passed through a 112 cm x 4.8 cm column of Sephadex G-100 equilibrated with CM Buffer (1% Triton X-100, 5.0 mM HEPES, 0.25 mM EDTA, 60 mM KCl, pH 7.3). Octanoyltransferase-containing fractions from the gel filtration column were pooled and applied at a flow rate of 80 m1/hr to a 29 cm x 4.1 cm column of CM-Sepharose CL-6B equilibrated with CM Buffer. The column was washed with four bed volumes of CM Buffer and then eluted with a 1000 ml linear gradient of CM Buffer containing 60-560 mM KCl. Fractions containing COT activity were pooled (117 ml), diluted with 117 ml of 20 mM KPO pH 6.6, and applied at a flow rate of 24 ml/hr 49 to a 13 cm x 2.2 cm column of hydroxylapatite previously equilibrated with HAP(-) Buffer (0.1% Triton X-100, 10 mM KPO 60 mM KCl, pH 6.8). 4, Four bed volumes of HAP(+) Buffer (identical to HAP(-) Buffer but with the addition of 0.25 mM EDTA) were used to wash the column before elution with a 200 ml linear gradient of HAP(+) Buffer containing 10-150 mM KP04, pH 6.8. Fractions having COT of a constant specific activity were pooled. Fractions eluting from the Blue Sepharose column with both COT and carnitine palmitoyltransferase (CPT) activities (fractions 40-48) were pooled and passed through a 112 cm x 4.8 cm column of Sephadex G-100 equilibrated with QAE Buffer (1.0% Triton X-100, 5.0 mM_Bis-Tris Propane Buffer, 0.25 mM_EDTA, 20 mM.KC1, pH 9.7). Carnitine octanoyltransferase- 35 containing fractions were pooled and applied at a flow rate of 30 ml/hr to a 48 cm x 4.1 cm column of QAE Sephadex 0-25-120 equilibrated with QAE Buffer. The COT activity which washed through the column was pooled, dialyzed against HAP(+) Buffer, and loaded at a flow rate of 24 ml/hr onto a 11.5 cm x 2.2 cm column of hydroxylapatite which was equilibrated with HAP(+) Buffer and eluted with 400 ml HAP(+) Buffer containing a 10-510 mM linear gradient of KP04, pH 6.8. Fractions containing COT of a constant specific activity were pooled. The stock enzyme was stored at 4°C at a concentration of 40 ug protein/ml in the presence of 0.025% (v/v) Triton X-100. (All assays contained enzyme at 2 ug/ml and Triton X-100 at 0.0013%. This is below the CMC of 0.012% determined for Triton X-100.) CMC Determinations Critical micellar concentrations of acylcarnitines and non-ionic detergents were measured by the fluorescence method of Horowitz (131) using TNS (2-p-Toluidinylnaphthalene-6-sulfonate). The 2.0 ml reaction volume contained 0.2 M Tris-HCl, 1 mM DDT, 0.5 mM EDTA, 11.0 mM_TNS, 60 uM_CoASH and varying concentrations of acylcarnitine or non-ionic detergent at 25°C and pH 7.70. Relative fluorescence measurements were made on an Aminco-Bowman Spectrofluorometer. Interaction of Laury1-14C-carnitine with Micelles of Tween-20 A 36 x 1.2 cm (I.D.) column of Sephadex G-100 was equilibrated at 25°C with a buffer solution (0.2 M Tris-HCl, 1 mM_DTT, 0.5 mM EDTA, 60 14 AU CoASH, pH 7.7) containing 700 A! C-lauryl-l-carnitine at 12,400 36 cpm/umole. After equilibration, a 2.0 ml aliquot of the equilibration solution, but with the addition of 0.4% (v/v) Tween-20, was applied to the column and 1.0 ml fractions were collected. For each fraction, the entire 1.0 m1 volume was added to 10.0 ml of scintillation cocktail (1 liter toluene, 1 liter Triton X-100, 4 g PPO and 100 mg Dimethyl POPOP) for radioactivity determination. An identical gel chromatographic pro- cedure was performed using non-labeled lauryl-l-carnitine and fractions were assayed for the presence of micelles by the method of Horowitz (131) to determine the elution volume of Tween-20. In both procedures, the lauryl carnitine concentration (700 A!) was below its CMC (1050 uM) while the non-ionic detergent Tween-20 concentration of 0.4% was in great excess of its CMC (0.0003%). Other Methods Carnitine esters were synthesized as described (120, 121). The synthesis of lauryl-14C-carnitine was identical except that a trace amount of DL-(methyl-14 C) carnitine hydrochloride was added to the unlabeled L-carnitine before the synthesis. Blue Sepharose 4B was synthesized by the method of Bohme gt al (124). Isoelectric focusing was performed in a sucrose density gradient after the method of Vesterberg (125). SDS-PAGE of purified proteins was performed by the method of Laemmli (122) employ- ing bovine serum albumin, catalase, fumarase, and chicken egg albumin as molecular weight standards. Catalase was assayed by the method of Boudhuin gt al (126). Protein was estimated with bovine serum albumin as standard according to the method of Lowry gt al (127) and, for solutions containing detergent, by the method of Bohlen gt_al (128). All chroma- tographic columns were treated with dimethyldichlorosilane. 37 Preliminary Investigations Solberg (104) found different carnitine acyltransferase specifici- ties in mitochondria from various organs of the rat. The investigation by Kopec and Fritz (90) of carnitine acyltransferase activities in calf tissue revealed extracts specific for octanoylcarnitine from the heart mitochondria. In deciding to pursue COT-containing enzymes in beef heart mitochondria we foresaw two significant limitations to the interpretation of the results: (1) Release of mitochondria from muscle, skeletal or cardiac, requires more severe treatment of the tissue than their release from most other organs such as liver, kidney, testis or brain. Longer homogenization is required to effectively break down the cells and a greater dilution of the homogenate is required to overcome the gel formed by broken down myosin in order to allow the mitochondria to sediment. It is possible that the harsh mitochondrial isolation procedure causes a relative enrichment of an inner COT due to a greater loss of CDT on the outer surface. (Tubbs and Chase (52) proposed that the outer CAT is more easily lost in mitochondrial isolation. The outer CPT of West gt al (113) was obtained in the 20,000 g supernatant of ox liver homogenized after being stored frozen. Yates and Garland (64) concluded that the CPT released by mild sonication of rat liver mitochondria was the outer activity.) (2) At least two species of mitochondria are recognized in heart muscle, intermyofibrillar and subsarcolemmar (115, 116). The former are more easily released by mechanical disruption of the tissue and the latter best obtained after sonication and/or limited treatment with a protease 38 such as nagarse. Nagarse treatment is reported (117) to preferentially attack proteins of the outer mitochondrial compartment, effectively sparing CPT but decreasing long chain fatty acyl synthetase activity. Because its effects on a possible outer COT were unknown, nagarse was not used in our mitochondrial isolation. Limited sonication, however, was employed as it allowed a near doubling of the yield of mitochondrial pro- tein and carnitine acyltransferase activities. Thus because of the tissue and methods required, our mitochondrial preparation would most likely be a mixture of interfibrillar and subsarcolemmar mitochondria. Beenakkers and Klingenberg (53) reported that rat heart CAT was completely solubilized by repeated extraction without the use of detergents or exhaustive sonication to disintegrate the inner mitochondrial membrane. Markwell gt a1 (26) found that CAT associated with the microsomal membrane could be released with 0.4 M_KCl. Reitz (118) saw that detergent was required to solubilize CPT from rat liver and heart mitochondria and that 98% of total CPT is released by treatment with a solution of KCl and Triton X-100. Using salts and detergent, therefore, the extraction of COT activity from beef heart mitochondria was compared to that of CAT and CPT. The results of the extraction experiments are shown in Table I. It is seen that as in other tissues, CAT is preferentially extracted with salts and CPT with detergent. COT is extracted by both measures, and in these mitochondria, with an approximate 70:30 CAT-like:CPT-like profile of membrane association. The most significant findings here are inferred from the ratios of COT to CAT or CPT in the extracts: (l) the ratio of COT:CAT in the salt extract is approximately equal to that value which we have observed in 39 mew: mosapo> quop mcFmemPo ocwuwccmo .0 van .m .< cw ummz .pr>Fuum mmmFoLu»; <00 qum vcaogmxuma com nmuumggou coma mm; Am x ooo.m~ an .:F5 mFV copummzwpcucmu Lmumm upapw “cmpmcgwazm tea. puppwa cp mpchs Ema xup>ruum mmmgmmmcmgupaum .Auv ooT x :OSFLF ammo. o 28 Amy _o¥ z ewe. o .Aa>onm \>V m. _ umuzppu wgmz FE\:wmpoLa ms mm do :owumcucwucou a pm gmmman mmoguam owcououw cw umumpomw mwgucocuouws exam; mmmm m.ou mmF om «.mp cc cum m.m Fm ompp “pauv Paouwspma m.m~ eem mop, m.oo mNNF emm F.m Nme mmew Abouv Paocapuo _.m_ Now mom_ m._m omm_ mpo m.m mum oucm Apwpaa amALGCmcmL» .oop-x copwch SF 8:5 umo> cw mmmcmemcmLu—zum mcwuvcgmu mo auwowmwowam mwmgumnzm .F mesmmm 43 [ too .I. E: I: l l l 12 O\ \O m (Tm/ulm/Salomu) asvaaasnval 1xov HMILINHVD 12 14 16 10 00A ESTER CHAIN LENGTH (carbons) {\l 44 .»Fm>wuumammg .cmgpxmu warn can wumcmAUwLme E:_MMmuoa mcwm: umcwscmpmv mcmz mmsapo> uwo> ucm mam .oopux mbumch RP vcm Fox :5 com m=L=_aS:ou Am.“ :8 .amocuzm A>\zv fim~.o ”aha“ :5 F .mmam: :5 opv mecan mmocaam cw cagaacmouasoccu ma: oop-x cepwcp NP\PU¥ z _ £832 uaNPFwnaFOW :meoaa _mLLu:o;uouLz .mmmgmmmcmgupaum mcwywccmo megncoguopwe “com; mmmn do mzamgmoumsogzu cowumgppw$ Pom oomuw xmumzamm .N mgzmwu 45 Ch Void Volume 90 ([m/utm/satomn) 30 - ~8 SSVHESSNVXL TADV ZNIIINXVD SO 40 FRACTION NUMBER Figure 2 46 It was decided to attempt to purify the detergent solubilized COT- containing protein to homogeneity in order to compare its properties and kinetics with those of previously reported carnitine acyltransferases. A number of chromatographic procedures were evaluated for fold purity and percent recovery of COT activity. Commercially available CoA-Sepharoses with the coenzyme attached at the -SH and the ribose positions (PL Bio- chemicals #s 5504 and 5491, respectively) had no selective affinity for COT under the conditions employed. Nor did carnitine when joined by ether linkage at the 8-0H moiety to a spacer arm and thence to Sepharose as synthesized from epoxy-activated Sepharose (Pharmacia) and l-carnitine in the laboratory. As mentioned above, all attempts at cation exchange were unsatisfac- tory. Anion exchange was pursued and revealed that as high as pH 10.2 using QAE-Sephadex and the buffer bis-tris propane, COT activity is unretarded with minimal loss of activity while most proteins do bind to the column. Another powerful technique with binding specificity and minimal loss was ersatz-affinity chromatography using Cibacron Blue- Sephadex (Bio-Rad). The enzyme apparently recognizes the dye ligand as a CoA-like nucleotide and is easily extracted by KCl at 300 mM. Expensive attempts to specifically elute COT from Blue-Sephadex using CoASH, NAD+ and other adenine nucleotides resulted in no improvement over KCl elution. Elution with a gradient of carnitine as a substitute for KCl also gave no improvement; the COT activity eluted at the same ionic strength with d,1-carnitine-HC1 or KCl as the salt used. Finally hydroxylapatite, used by Kopec and Fritz (90) to purify calf liver mito- chondrial CPT, bound beef heart mitochondrial COT and released it at 47 higher phosphate concentration with an acceptable recovery of activity. Using these methods--KCl/Triton X-100 quantitative solubilization of COT activity, gel filtration chromatography to separate free and micelle-bound acyltransferase proteins, and Blue-Sephadex, QAE-Sephadex and hydroxylapatite chromatography to purify the detergent-extracted COT activity--a COT-containing protein of greater than 90% purity was ob- tained and some of its properties were examined. Some of these findings are shown in Figures 3-6 and Tables II-IV, on pages 48-58. These data were originally prepared as a poster presentation (119). To summarize these results, a near homogeneous protein of monomer molecular weight approximately 67,000, pI 8.1 and pH optimum between pH 7.4 and 8.4 was isolated from beef heart mitochondria having optimum acyltransferase activity with the substrate nonanoyl CoA, greatest even-chain activity with decanoyl CoA, octanoyl activity greater than palmitoyl, and vanish- ingly small activity with substrates of less than six carbons in length. Thus it was shown that beef heart contains a carnitine acyltrans- ferase specific for medium chain length substrates as assayed in one direction. More work was required to synthesize the various acylcarnitine esters in order to assay the enzyme in the opposite direction, the reverse reaction, and thereby compare its specificity with those published for CPT by West et_al (113) and by Kopec and Fritz (90) using the 232 assay. The even carbon length esters of l-carnitine from acetyl to palmitoyl- carnitine were synthetized by the short and long chain methods of Bremer (120, 121); these results are summarized in Table V. Using these sub- strates, the relative activities of the beef heart enzyme using octanoyl decanoyl- , lauryl- , myristoyl- and palmi toylcarnitine were determined under 48 Table II. Abstract of COT Studies ABSTRACT #644 STUDIES ON CARNITINE OCTANOYL TRANSFERASE IN BEEF HEART AND LIVER MITOCHONDRIA P.R.H. Clarke and LL. Bieber, Biochemistry Dept., Michigan State University, East Lansing, MI 48824 Carnitine octanoyl transferase activity is present in two fractions separable by gel filtration chromatography of extracts from solubilized beef liver and heart mitochondria. The two peaks of activity represent the micellar form of a membrane protein referred to a carnitine octanoyl transferase (COT) and a soluble protein of molecular weight approxi- mately 60, 000 daltons emerging from the gel at the same position as the carnitine acetyl transferase (CAT) of the mitochondrial matrix. In heart mitochondrial extracts, 30 % of the total COT activity is found associated with the membrane protein while the remainder is found in the 60, 000 dalton peak. In liver mitochondria, where the highest carn- itine acyl transferase activities are for middle length chain fatty acyl CoA esters, fractionation of solubilized mitochondria reveals the bulk of C01 activity in the membrine protein fraction. The membrane COT protein has been purified from beef heart mitochondria and partially characterized. The enzyme has an isoelectric point of 8.3, similar to that reported for the CAT protein, and shows maximal activity with nonanoyl Coenzyme A as acyl donor. The purifica- tion and kinetics of the enzyme with respect to middle length chain fatty acyl CoA esters will be presented. (Supported in part by USPHS N. I. H. grant #AM 18427). 49 Table III. Purification Procedure PURIFICATION OF CARNITINE OCTANOYL TRANSFERASE FROM BEEF HEART MITOCHONDRIA Mitochondrial Isolation Beef hearts were obtained from a local abattoir, where they were sliced in half and packed in ice for transport to the laboratory. All sub- sequent procedures were performed at 4°C. Connective tissue was cut away from the ventricle which was then sliced into thin pieces, diluted 1 : 8 with Sucrose Buffer (0.25 M sucrose, 5 mM HEPES, 0.25 mM EDTA, pH 7.5) , homogenized for 45 seconds in a Waring blender at high speed, and finally treated for 30 seconds with a polytron homogenizer using the large blade/probe. Mitochondria were isolated from the homogenate by differential centrifugation employing spins at 650, 15,000, 650, 11,000 and 7000 x g. The final mitochondrial suspension, yielding approximately 1.2 g of mitochondrial protein per Kg cleaned ventricle, was assayed for carnitine octanoyl transferase activity and frozen at -80°C for later use. Solubilization Frozen mitochondrial suspensions from eight beef hearts were thawed, pooled and assayed for protein and enzyme activity. To two vol- umes of thawed mitochondrial suspension at 19 mg protein! ml was added one volume of a solubilization solution containing 6% Triton X-100 and 3 M KCI. The mitochondria solution was centrifuged for 90 minutes at 85,000 x g and the supernatant liquid retained as solubilized mitochon- drial protein. Purification ’ The mitochondrial protein solution was chromatographed in batches over Sephadex G-100 to separate carnitine acyl transferase activities into micellar and free protein populations. The micellar peak fractions of COT activity were pooled and dialyzed against HEPES Buffer (1% Triton X-100, 20 mM KCI, 0. 25 mM EDTA, 2.5 mM HEPES, pH 7. 5) and centrifuged to remove protein precipitated by the decrease in ionic strength. The super- natant containing COT activity was applied to a Ciba-cron blue Sepharose column to which the activity bound and was eluted by a 20 - 400 mM gradient of KCI in the HEPES Buffer. Peak fractions were pooled, dialyzed against a Bis-Tris Propane Buffer (1% Triton X-100, 20 mM KCI, 0. 25 mM EDTA, 5.0 mM Bis-Tris Pr0pane, pH 9.7)and applied to a column of QAE- Sephadex. COT activity did not bind to the column, but was washed through. Finally, the solution containing 001 activity was dialyzed against a Phosphate Buffer (1% Triton X-100, 20 mM KCI, 0.25 mM EDTA, 10 mM KlPhosphate, pH 6.8) and applied to a hydroxyl apatite column equilibrated with the same buffer m EDTA. A linear 10 - 400 mM KlPhosphate gradient eluted a peak of COT activity which is 700- fold enriched over that assayed in the original mitochondria and is approximately 90% pure by analysis of stained protein banding on SDS polyacrylamide gels. 50 Table IV. Purification Results Purification Results Purification Step Volume Units Protein Specific Fold Percent (ml) (umoleslmin) ng) Actwuty Recovery Thawed Mitochondria 628 079) 11,995 0.0899 1.0 100 335 f (0.0279H (31); Solubilized Mitochondria 1084 993 10,623 0.0935 1.04 92 Sephadex G-100 2625 294 6038 0.0487 1.75 88 20 mM KCI Dialysate 2380 281 3165 0.0888 3.19 84 Ciba-cron Blue Sepharose 1195 255 587 0.435 15.6 76 QAE - Sephadex 940 171 52.6 3. 25 117 51 Hyd roxyl Apatite 92 116 5.93 19.6 702 35 :l.’ Numbers in parentheses are calculated values for the micellar COT protein correcting for COT activity associated with the free protein peak separated by gel filtration (see Figure 1). The values for fold purification and percent recovery for the Sephadex G-100 and subsequent steps are based on the calculated values. 51 Figure 3. SDS polyacrylamide gel electrophoresis of purified COT. Approximately 5 ug of purified beef heart mitochondrial was subjected to SDS-PAGE according to the method of Laemmli (122). Protein was stained with Coumassic Blue and its density in the slab gel was moni- tored using a Zeiss spectrophotometer. Sharp peaks at the extreme left and right of the tracing represent the top of the gel and the dye front (Bromephenol Blue), respectively. 52 SCAN OF PURIFIED COT Oil SDS-PAGE M Figure 3 53 .A<>ov cwsan~m>o cm; new Azauv mmmcmsam .Apouv mmwcwmmcmcupzpmum mcwpwccmo mpomss “memes commwa .Aon mvcmnzmpm :wmpoca mg» we mcowumcmme mg» saw: umcmaaou m? Am mesmwu mmmv m_mmco;aocpom_m pom mowsmpzcuwapoa mom m:_g:c Foo umwmwgza mo cowumcmws m>wpmpmc as» .Hou Fawcvcosuogws “Lam; moan we usmwmz cmpzumFoE pwczazm .e mczmwm 54 e mgamwm zoapmmaaz u>Hh¢4mx and“ woum, on»: mmuo we“: can: anus an.pv w . / ... $5 a w. . E: e ... w 1 1.? a 5 e I .9 .25 r 0 5m 1 r.’ m. 3 r.’ d d d J u d - MD 0 mom =o_~wpum mmmcmwmcmgppzum mcvuwcgmo do mmapm> mzocm Atacama mZFQ asp >5 vmzmmmmv cowgummc ucmzLom as» Low boo umwmwczn mo mpwmoga Auwowewumqm mpmcumnsm och .Hoo mecuconuouwe “Lam; moon wwwmvgsa mo zuwuwmwumam mumcumnam .m mgzmwu 56 cu mu vu Nu m mgzmwu :hozmwuz a are Zoommcu hIIIIIIAVlhIAYIIIAYIPIAYIlIAYIhIAVIIIAVIhILVIIILVILILVIIIAYIPIAWHHHA*. r... oo-o° 00'? 333N681 'IAOU 00’9 it B 8 oo' 36 39 + d u q q d d mm§wmm=<§ #2250 m_._.:_..m5 SEE—E. “E BEBE KGB—”mam mpéhmmam 57 mstcm och .Ps\cwwpoga as o.m upwpmswxocaam m? cowpmcucmocou .—E\cve\_oE: cw ummmmcaxm m? ap_>wuom wmmcmemcmgppzocmuuo mcwpwcgmo .Hou mecvcosoopws “Lam; moan do Ezswuao :a .o mgamwu 58 c mgzmwu .Lm oc.~u ov.m~ ow.m ou.m cm.m oo.m atuh om.w cN.@ - p p b b ow. a: 8: 22.53 2.: em a + a: 2: 3% e a .4 ._ .. 2.... 2: $0: + - fi 4 q u l- d J- - ch :m_u_m=¢ no :=z_h¢o :m Table V. Synthesis of Acyl-L-carnitines. 59 L-Carnitine % Free Specific Ester L-Carnitine Rotation Acetyl 0.95 -7.0 Butyryl 0.97 -7.0 Hexanoyl 0.57 -7.0 Octanoyl 0.49 -7.0 Decanoyl 0.48 -6.8 Lauryl 0.87 -5.0 Myristoyl 0.93 -4.2 Palmitoyl l.55 -4.0 Acyl-du-carnitines were synthesized by published methods (l20, l21) and judged to be approximately 99% pure by TLC analysis (120). Percent con- tamination by free L-carnitine was determined by radioactive carnitine assay before and after alkaline hydrolysis. ments were made using a Zeiss polarimeter. Specific rotation measure- 60 the conditions employed by Kopec and Fritz with their calf liver mito- chondrial CPT. The results were nearly identical. The relative rates with these substrates reported by Kopec and Fritz (90) were 2, 26, 3l, 99 and lOO, respectively; our values were 8, 38, 38, 79 and 100. It was con- cluded that the beef heart mitochondrial enzyme corresponds to the CPT I isolated from calf liver mitochondria. At this point it was decided to isolate and identify ALL carnitine acyltransferase proteins from beef heart mitochondria (and to fractionate free and micellar protein from beef liver mitochondria) for the following purposes: (I) to obtain more of the protein previously isolated in order to investigate and explain the apparent differences in substrate speci- ficity of the enzyme for the forward and reverse reactions, (2) to deter- mine whether the free protein fraction seen during gel filtration of solubilized mitochondria indeed contains but one carnitine acyltransferase enzyme, CAT, whose activity with octanoyl CoA totally accounts for COT activity in this fraction, (3) to determine whether a separate CPT enzyme had been overlooked or lost in the previous isolation of COT activity from the micellar protein fraction, and (4) to determine whether, if only two transferase enzymes exist in beef heart mitochondria, is this also true for beef liver mitochondria, based on the substrate specificity of heart and liver mitochondrial free and micelle-bound transferase activi- ties separated by gel filtration. The following section presents the results of this investigation, essentially as submitted for publication (l23) under the title of the heading below. Though the free protein frac- tion was early not considered a source of a separate medium chain 61 acyltransferase due to its high CAT activity thought to account for the activity with octanoyl-CoA, the procedures described below for the purifi- cation of transferase protein from this fraction were developed in parallel with those of the enzyme in the micellar fraction from gel fil- tration of the solubilized mitochondrial protein. Isolation and Purification of Mitochondrial Carnitine Octanoyltransferase Activities from Beef Heart Results Comparison of Carnitine Acyltransferase Activities in Beef Liver and Heart Mitochondria Substrate specificity profiles of carnitine acyltransferase activi- ties in liver and heart mitochondrial are presented in Figure 7. Figure 7, A and E show the profiles of intact liver and heart mitochondria. To ascertain whether or not these profiles reflect the activity of one enzyme or more than one enzyme, solubilized liver and heart mitochondria were chromatographed separately on Sephadex G-lOO. Fractions were assayed for the distribution of carnitine octanoyltransferase activity (B and F). Substrate specificity profiles of the two COT containing peaks separated by gel filtration are pictured in C and D and in G and H for liver and heart, respectively. For each tissue, gel filtration of solubilized mitochondria gave two protein peaks containing transferase activities. Virtually all carnitine acyltransferase activity was released from mito- chondria during solubilization (data not shown, but see results from purification of heart mitochondrial transferase activities in Table VI). 62 Figure 7. Substrate specificity profile of intact and partially fractionated beef heart and liver mitochondrial carnitine acyltransferases. Carnitine acyltransferase activities were determined by the DTNB assay described in the Methods. Activities are given as nmol x min-1 at 25°C. The substrate specificity profiles of intact liver mitochondria are given in A and intact heart mitochondria in E. Frames B and F show the transferase activities determined with octanoyl CoA of the fractions obtained when liver and heart mitochondria were solubilized in 2% Triton X-l00 containing 1.M KCl and chromatographed on a l12 cm x 4.8 cm Sephadex G-lOO column as described in Methods. The micellar peak (the first peak) from Figure 7B and 7F were analyzed for substrate specificity; these profiles are shown in frames C and G, respectively. The substrate specificity of the free protein fraction (the second peak) are given in D and H. 63 UVER A. IN TACT MITOCHONDRIA 0 IO I4 CARBON CHAIN LENGTH ACYL TRANSFERASE AWL TRANSFERASE ACTIVITY an o FRACTION NUMBER 4045 50566065 7075 HEART l . . . . . . 1 r I E. INTACT MITOCHONDRIA 8 45 I I I I l I ESEPHADEX G-IOO CHM CF SOLUBILIZED DRIAL - PROTEIN 45 50 55 60 65 70 FRACTION NUMBER :mo . . v . I . v . mo . - r E c. MICELLAR E e. MICELLAR 2gizoo- FRACTION 5 I50- FRACTION ‘ Iso- byzo« . g l20 g so- g so . e .0. . I‘s‘ “° E E é’j g V 'é"b 15 w' k a g 2 4 s a m I2 H B I CARBON CHAIN LENGTH CARBON CHAIN LENGTH > 240 1' I I I I v T t 5"- I r I I v I : QFREEI*DTBN 5i ftfiEI munsm Eaco- FR wa glflr annum IGO‘ IZO‘ - III III 2 I20 < 90. . eo- EIar ~ 40 i g SOJ .— " - . . . t3 .1 E's—Y c ., : ., , , , g V a 4 s 3 IO I2 n l6 l8 5 2 4 s a n n w m a CARBON CHAIN LENGTH CARBON CHAIN LENGTH Figure 7 64 There was no appreciable loss of any of the transferase activities as a result of gel filtration. The first peak of carnitine octanoyltransferase activity to emerge from each Sephadex column contains enzyme or enzyme aggregates of appar- ent molecular weight greater than 300,000 daltons. Based on the analysis by other investigators (l29) of the gel filtration behavior of detergent- soluble proteins in the presence of non-ionic detergent, we conclude that this peak represents predominately proteins associated with micelles of Triton X-l00. For solubilized mitochondrial protein from both tissues, the first peak contains all of the long chain and some of the medium length carnitine acyltransferase activities. All of the short chain transferase activities and the remainder of the medium chain activities eluted with an apparent molecular weight of 60,000 daltons. The sub- strate specificity profiles of micelle-associated activities solubilized from mitochondria of beef liver and heart are similar (see Figures 7, C and G) as are the profiles of the activities of the 60,000 dalton peak eluted (Figure 7, D and H). Purification of Carnitine Octanoyltransferase (COT) Activity from Beef Heart Mitochondria Solubilization. Preliminary studies (data not shown) indicated that beef heart mitochondrial carnitine acetyltransferase (CAT) activity is mostly soluble. More than 90% is released when the mitochondria were disrupted by freeze-thaw or sonication in the presence of 0.5 to 2.0 M KCl. Once the mitochondria were disrupted, 60 mM,KCl was required to keep proteins containing carnitine acyltransferase activity in solution. In contrast, carnitine palmitoyltransferase is poorly solubilized by 65 treatment with KCl requiring instead Triton X-lOO (l.0-2.0%) to release greater than 90% of the CPT into the 90.000 x g supernatant fluid. A significant fraction of total COT activity is found in both supernatant fluid and pellet with either the salt or the detergent extraction pro- cedure. Therefore, in order to maximize the release of COT activities by one extraction procedure, thawed mitochondria were solubilized in 2% Triton X-lOO in the presence of l M_KCl. The results in Table VI show that this combined detergent/salt extraction solubilized 79% of the mito- chondrial protein and 75, 78 and 90% of CAT, COT and CPT activities, respectively. Dialysis. Dialysis of the solubilized protein solution preparatory to Blue Sepharose chromatography lowers the effective KCl concentration to 60 mM. At this concentration approximately half of all solubilized protein precipitated out of solution, including a fraction of the carni- tine acyltransferase activities. Our preliminary experiments (data not shown) indicated that in dilute (< 1 mg protein/ml) solution, carnitine acyltransferase activity from solubilized beef heart mitochondria is not appreciably precipitated at a KCl concentration of 60 mM. There is measurable precipitation of CAT and to a lesser extent of COT at 50 mM KCl, with near total precipitation of CAT, a lesser fraction of total COT and insignificant loss from the supernatant fluid of CPT at 20 mM KCl. The precipitation of significant CAT and COT activities during this procedure may be due to a reduction of the actual concentration of KCl inside the dialysis bag because of charge contributions of the protein. 6E5 .m egaaww we u ecu w .m mpeeee ea meceemeggeu emw< .me_u_>wuee wee ecu woo seen mcwcweucee Ameloewv mcewueegw emegezeem ezpm :e eesgewgee mesaeeuege cewuaewwwgan meueeweeuuu .m egeaww we a use 0 .m mpeeee ea meeeemoggee emp< .mewe_>wuee wwuee xge>eeeg ,Imwea u_wwemam ecaecea me_== Awauv emegewmceguPAequpea xuw>wuee age>eeeg lapewlewwweeem «smegma mews: IINHouv emegmwmcegupaeceuuo xuw>wuue age>eeeg epewluwwweeem aceecme muwez ~H eweew BC Ta 67 The protein pelleted after dialysis contains a third of the total CAT activity, a smaller fraction of COT and negligible CPT activity (see Table VI). The ratio of COT to CAT activity assayed in the pellet is less than 0.5. The dialysis supernatant fluid which was loaded directly onto the Blue Sepharose column contains 67, 72 and 96%, respectively, of the initially solubilized CAT, COT and CPT activities. Results of Column Chromatography Greater than 95% of COT activity present in the soluble heart mito- chondrial protein fraction after dialysis against 60 mM_KCl is retained on Blue Sepharose. This activity elutes as two well-separated peaks after administration of a linear KCl gradient (see Figure 8, A). CAT and CPT activities are completely separated; each co-elutes with one of the COT peaks. Acyl-CoA hydrolase activities are partially removed during the dialysis and are nearly totally absent in the Blue Sepharose peaks of carnitine acyltransferase activity (data not shown). Both CAT and CPT are purified about 20-fold by this step. Purification of the COT- and CAT-Containing Blue Sepharose Peak Fractions 3l-34. A further 20-fold purification of the CAT/COT activity is effected by CM-Sepharose ion exchange chromatography (Figure 8, B). Chromatography of the peak from CM-Sepharose on hydroxyl- apatite (HAP) yields nearly superimposable peaks of protein and COT activity (see Figure 8, C). Fractions from the hydroxylapatite column of constant specific activity were pooled and a sample was analyzed for purity by SDS-polyacrylamide gel electrOphoresis. This procedure 68 Figure 8. Purification of carnitine octanoyltransferase from beef heart mitochondria. In A, transferase was solubilized as described for Figure 7E and the solution was exhaustively dialyzed and applied to a 75 cm x 4.l cm column Cibacron Blue Sepharose 48 column equilibrated with the Blue Buffer described in the methods. The sample was applied and the column was washed with 4 bed volumes Blue Buffer and then eluted with a 60-860 mM KCl linear gradient in Blue Buffer. The numbers in parentheses represent carnitine acetyltransferase activity (the diamond shaped points in the figure) and the other numbers repre- sent carnitine octanoyltransferase (the open circles) and carnitine palmitoyl- transferase activity (the open squares). The solid thin line represents the protein. The KCl gradient is indicated by the dots; the units mS represent millisemens conductivity. In B, the first transferase peak, fractions 3l-34 of Figure 8A, were pooled and passed over a Sephadex G-lOO column to equilibrate the protein with the CM Buffer described in the methods. The protein in CM buffer was applied to a 29 cm x 4.l cm column of CM-Sepharose CL-6B and washed with 4 bed volumes of CM Buffer. The activity was then eluted by 60-560 mM linear gradient of KCl in CM Buffer. The symbols are identical to those described in Figure 8A. The open circles represent carnitine octanoyltransferase activity. Fractions 37-4l of Figure BB were pooled and diluted with an equal volume of 20 mM KZHPO4, pH 6.6. The solution was applied to a 13 cm x 2.2 cm hydroxyl- apatite which was equilibrated with the HAP (-) Buffer described in the methods. The column was washed with 4 bed volumes of HAP (+) Buffer and then eluted with a 10-510 mM’KHPO4, pH 6.8 linear gradient. The symbols are identical to those described in Figure 8B. Fraction l7-l9 (the peak ones)of Figure 8C were pooled and analyzed for purity by SDS polyacrylamide gel electrophoresis as by the method of Laemmli, see Figure 80. T represents the top of the gel and 8 represents the bottom of the gel. The DYE was bromphenol blue. In E, fractions 40-48 of Figure 8A were pooled and passed through a Sephadex G-lOO column equilibrated with the QAE Buffer described in the methods. The protein solution was then passed through a 48 cm x 4.l cm column of QAE Sephadex Q-ZS-l20 equilibrated with QAE Buffer. The open circles represents carnitine octanyltransferase activity. Fractions 23-29 of Figure 8E were dialyzed exhaustively against several changes of HAP (+) Buffer and then applied to a ll.5 cm x 2.2 cm column of hydroxyl- apatite and processes as described in Figure 8C. In this panel, F, the open circles represent carnitine octanyltransferase, but the solid line represents protein. For the peak fractions protein and transferase activity were super- imposable. In G, the peak fractions (12-l4 of F) were combined and subject to SDS polyacrylamide electrophoresis as described for Figure 80. The data in Figure 8 are summarized in Table VI where * represents purification described in Panels B, C and D and ** represents the purification shown in panels E, and G. 69 26 IO FRACTION NUMBER 8 FRACTION NUMBER .? 2.5: '5 L5 ‘3 ‘” x020) A” .. E '1': o 6 -- LO L5.S E (8.0) z 180 z ..... I.O E 460 2 0.5 O -40 gem) 0.5 E :1 é 120 V °o ab 0 I FRACTION NUMBER 3 7: : 25'}; 1.. P U) L. E LO" B 1505 E EO-B" E .20 x ." 0-8 s ." 3 f5 " I 15‘16’ “'5 E 0.6 ,,. loo ; 4Q E g X E - 30 x 0.4 - d '0 '— ,,on m c> % .. ‘SOE'ZO €02. .5 E E 0.2 0'. IO E . 3 3 o - . - o l 3 o . . J o 20 so 40 a: 25 so FRACTION NUMBER FRACTION NUMBER .. T: w '8' -l20 x E I00 x ...... . g g 75 g ‘ 80 z ‘80 o z IiJ «so 5° I5 . -40§ :3 25 g : 1 : ° m- o o I ° B 0.6 D 06' G e ' B E c .3 0.4L ‘0 04 It": 'r DYE UN) ' r M d 0.2 ~ Jr ‘5 02 . T 1 0' L C5 ‘ H 0.0.. 1 4 4 1 1 1 Q 1 1 F- LI 0 2 4 6 8 IO 0 2 4 6 8 IO MIGRATION DISTANCE (cm) MIGRATION DISTANCE (cm) Figure 8 1 1 l 5 3 g 3...... l mS 70 revealed the presence of one major protein band and two discernible con- taminating proteins of much lesser abundance. The densitometer scan of the gel stained for protein (Figure 8, D) indicates that the purified CAT/COT protein is approximately 95% pure. Purification of the COT- and CPT-Containing Blue Sepharose Peak Fractions 40-48. A two-fold purification was achieved during Sephadex G-lOO chromatography of the pooled CPT/COT fractions from the Blue Sepharose column. A further nine-fold purification was obtained by passing the protein solution through a column of QAE-Sephadex at pH 9.7 (see Figure 8, E). Hydroxylapatite chromatography of the protein effluent from the QAE-Sephadex column revealed a sharp, symmetrical peak of con- stant specific activity (Figure 8, F). The CPT/COT protein is estimated to be greater than 95% pure by SDS-polyacrylamide gel electrophoresis (see Figure 8, G). Molecular Weight Determinations Comparison of the elution of CAT/COT and CPT/COT activities after Sephadex G-200 chromatography with those of catalase and hemoglobin indi- cate apparent molecular weights for the two transferases of 60,500 and 5l0,000 daltons, respectively (see Figure 9). Subunit molecular weight determination by SDS-PAGE using bovine serum albumin, catalase, fumarase and ovalbumin as standards yields values of 62,600 and 67,000, respec- tively, for beef heart mitochondrial CAT and CPT (Figure l0). Isoelectric Point Aliquots of the two purified proteins were combined and subjected to isoelectric focusing in a sucrose density gradient. The results shown 7I Chromatography of solubilized carnitine octanoyltransferase Figure 9. from beef heart mitochondria. The solid circles represent carnitine octanoyltransferase activity measured by the DTNB assay. The thin solid line is protein and the arrows indicate the peak positions of molecular weight markers. (no 3 min" 1 all") COT Activity 72 .050 .070 .l00 .l25 I .025 nu Como“ "cannula - \ 20 30 40 50 FRACTION NUMBER Figure 9 60 K3FO‘CN)‘ I7 W |.0 70 PROTEIN (mg x ml") 73 Figure 10. Weight determination of carnitine acyltransferases from beef heart mitochondria. The peak fractions from Figure 8C and Figure 8F were subjected to the polyacrylamide gel (10% gel) electrophoresis as described by Laemmli (l22). CPT represents the CPT/COT protein and CAT represents the CAT/COT protein described in the results. BSA is bovine serine albumin. 74 CPT BSA 0" 70.000 \1 Cota|0u\ 60.000 MOLECULAR WEIGHT 50,000 Fumorau .- \ 0.4 0.5 0.6 RELATIVE MOBILITY Figure l0 75 in Figure ll give pI values of 8.20 for the CAT protein and 8.05 for CPT of mitochondria from beef heart. Isoelectric Focusing-figpproach to Equilibrium In order to investigate the possibility that the micelle-associated COT and CPT activities are associated with two distinct proteins which may have co-purified during our purification procedure, the purified CPT/COT protein was subjected to isoelectric focusing on a pH 3-l0 gradient and assayed before the protein(s)' migration had reached equilib- rium. Similar but different proteins which have indistinguishable pI's may show different patterns in approaching isoelectric equilibrium. The results presented in Figure l2 show that the two activities are superimposable, suggesting that both activities reside in a single enzyme. Substrateg§pecificity The substrate specificities of the two carnitine acyltransferases purified from beef heart mitochondria are presented in Table VII. Included are results obtained for commercial pigeon breast muscle mito- chondrial CAT (Sigma) assayed under identical conditions. For purposes of comparison, literature values for a preparation of purified carnitine palmitoyltransferase (90) are listed, showing relative rates of the reverse reaction. The CPT/COT enzyme has high medium chain transferase activity in the forward direction in contrast to the reverse reaction. In Figure 13 the substrate specificities of the purified carnitine acyltransferase proteins (lower panel) are compared with that obtained initially for the thawed suspension of beef heart mitochondria. The CAT and CPT enzyme profiles have been normalized to equalize the heights of 76 Figure ll. Isoelectric point determination of the carnitine acyl- transferases from beef heart mitochondria. Aliquots of the pooled purified proteins described for Figure l0 were subjected to isoelectric focusing, see Methods. 77 IIZ 0' R E 03 L3" U N . N o v OT c L 12A v. .. o . n v. n T I l L c ‘ ‘ ' ~ P P h o 3.0 0.0 «.0 O .72.. u TEE u .225: umdzwum2(¢h 4>U( uz..—._z¢wpum m>wumpmg om mmsx~=m houthu ms“ Lot oop um “mm mm; wmmgmmmcmgppaouwspmn mo Auw>wuum mnp new FOU\Pw¢um ms» .uw mgzmwu seem mcowpumgm xmma asp mew HOU\hmo 0;» van ow mesa?“ sogw mcowpumgm xmma mg» mpcmmmggmg zpm>wuum hoo\pdmm< mmcm>mm xwmm< vcngou admm< mmgm>mm xdmmmivcmwcou mzuwmmm mmwgmwmcmLHP»Opwspwm mcwuwcgmu mmmgwwmcmLuFApmo< mcwumcgmu qu< mmmmgmmmcmcppxu< mcwpwcgmu Pm_gu=osuopwz “Lam: mmmm mo mmwuw>wuu< m>wpmpmm .mmmmgmwmcmgupzuq mcwpwcgmu megucoguouwz exam: mmmm mo mmw9w>wpu< m>wpmpmm .HH> mpnm» 81 Figure 13. Comparison of carnitine acy1transferase activity of beef heart mitochondria with the purified transferase enzymes. See the section on substrate specificity in the Resu1ts for detai1s of this experiment. A. whoie mitochondria B. purified proteins ($0! C4) CARNITINE ACYL TRANSFERASE ACTIVITY 82 A Mitochondria O 0 C2 CO co C. CIO 612 C14 C38 B \\V car/cat CPT/001’ c2 04 CG CI Clo CI2 CM C16 CARBON CHAIN LENGTH Figure 13 83 C-2 and C-16 in upper and 1ower portions of the figure. The cross- hatching indicates over1apping of medium chain activities. Amino Acid AnaTysis Beef heart mitochondria1 CAT and CPT were subjected to acid hydro1ysis for 24 hours and the amino acids were measured using a Beckman amino acid ana1yzer. The amino acid compositions of the beef heart transferases are shown in Tab1e VIII. Discussion Ge] fi1tration of the detergent-so1ubi1ized mitochondria separates short and Tong chain acy1transferase activities, with each fraction contributing COT activity. In beef 1iver, the majority of COT activity chromatographs with carnitine pa1mitoy1transferase (CPT) whi1e in heart, in which carnitine acety1transferase (CAT) is much more abundant re1ative to 1ong chain transferase activity, the majority of COT is found with the short chain transferase. The subsequent iso1ation of on1y two carni- tine acy1transferase proteins from beef heart mitochondria, one corre- sponding to each of the free and mice11ar peaks seen with ge1 fi1tration, together with the simi1arity of substrate specificity for the correspond- ing heart and 1iver free and mice11ar mitochondria1 protein fractions 1eads to the conc1usion that beef 1iver mitochondria most 1ike1y a1so contain but two carnitine acy1transferase enzymes. The two proteins obtained from beef heart mitochondria can account for near1y a11 COT activity initia11y detected in the intact mitochondria. Carnitine acy1transferase activities which were precipitated during dia1ysis of so1ubi1ized mitochondria1 protein preparatory to co1umn 84 Tab1e VIII. The Amino Acid Compositions of Beef Heart Mitochondria1 Carnitine Acy1transferases.* Amino Acid CAT/COT CPT/COT Asx 8.99 - 7.72 Thr 4.70 4.49 Ser 6.97 6.33 G1x 11.36 9.46 Pro 6.23 5.95 G1y 5.04 5.26 A1a 8.65 9.76 Va1 6.71 5.03 Met 3.13 2.53 I1e 4.40 5.00 Leu 10.77 11.52 Tyr 4.83 4.89 Phe 5.09 8.01 His 2.74 2.40 Lys 5.67 5.70 Arg 4.74 5.94 Cys nd nd Trp nd nd * Va1ues for each amino acid are mo1e percent of tota1 amino acid de- tected after 24-hr hydro1ysis; nd = not determined. The CAT/COT is from the poo1ed peak fractions from Figure 8C and the CPT/COT is from the poo1ed peak fractions from Figure 8F. 85 chromatography were not further purified. These precipitated proteins contained 33% of the initia11y so1ubi1ized CAT, 28% of COT and 4% of tota1 CPT activities. It is possib1e that this fraction contains a COT protein different from those fina11y purified from the dia1ysis super- natant f1uid. However, based on the COT activities present in the pure CAT/COT and CPT/COT proteins, more than ha1f (17 of the 28%) of tota1 COT found in this fraction can be accounted for by the CAT/COT and CPT/COT proteins which are partia11y precipitated by this procedure. Though the remaining 11% of COT cou1d represent a separate acy1transferase protein, it shou1d be noted that the precipitate contained a11 of the acy1-CoA hydroIase activities (data not shown) so that the b1anks measured in this fraction are very high and the error for the trans- ferase activities reported in the precipitate is corresponding1y 1arge. The two COT-containing proteins are separated by cqumn chroma- tography on Cibacron B1ue Sepharose and are further purified to near homogeneity by CM-Sepharose and QAE-Sephadex chromatography, respectiver, for the acecy1 and pa1mitoy1transferase containing enzymes (see Figure 8, A). The two proteins have simi1ar isoe1ectric points and amino acid compositions, though the CPT/COT protein is s1ight1y richer in hydrophobic amino acid residues (see Tab1e VIII) as expected from its affinity for the detergent Triton X-100. The native moIecu1ar weight (60,600) and pI (8.2) of the CAT/COT protein are simi1ar to those reported by Markwe11 (26) (MW = 59,000, p1 = 8.3) for carnitine acety1transferases partia11y purified from rat 1iver peroxisomes and microsomes. A compari- son of the re1ative activities of the purified CAT/COT protein using acy1carnitine and acy1-CoA substrates with those of commercia1 purified 86 pigeon breast musc1e CAT (see Tab1e VII) 1eads us to conc1ude that this beef heart mitochondria1 protein is a carnitine acety1transferase (EC 2.3.1.7) and not a nove1 COT enzyme. Carnitine pa1mitoy1transferase has been purified to homogeneity from ca1f 1iver mitochondria by Kopec and Fritz (90) and the re1ative rates of the reverse reaction using acy1carnitine substrates are reported therein. A comparison of these data with those obtained for the beef heart mitochondria1 CPT/COT protein assayed under identica1 conditions shows a striking1y simi1ar pattern (see Tab1e VII). For the forward reaction, the purified CPT/COT protein gives a marked1y different sub- strate specificity profi1e with a preference for medium chain 1ength substrates. Based on the simi1arity of the substrate specificity of the CPT/COT protein to that of CPT purified by other investigators, it is conc1uded that the beef heart mitochondria1 CPT/COT protein is a carni- tine pa1mitoy1transferase (EC 2.3.1.21), a1so not a nove1 COT enzyme. An investigation of the apparent difference between substrate specificity of purified beef heart mitochondria1 CPT for the forward and reverse directions was next undertaken. The fo11owing section contains the resu1ts of this study as submitted for pub1ication (130) under the tit1e of the heading be1ow. Effect of Mice11es on the Kinetics of Purified Beef Heart Mitochondria1 Carnitine Pa1mitoy1transferase Resu1ts In the previous paper (123) we reported for purified beef heart mitochondria1 carnitine pa1mitoy1transferase (CPT) rates of the reverse 87 reaction: acy1carnitine + CoA + carnitine + acy1-CoA using 500 pg acy1carnitines simi1ar to the re1ative rates described by Kopec and Fritz (90) for CPT purified from ca1f 1iver mitochondria. During the investiga- tion of the kinetics of the reaction with various carnitine ester sub- strates, biphasic kinetic phenomena in the reactions with 1aury1- and myristoy1carnitine were observed (these data not shown, but see Figure 15) suggesting that at 1ower substrate concentrations, a different pro- fi1e of substrate specificity wou1d be seen. The initia1 rates of reac- tion with different substrates at two concentrations were investigated. The data in Tab1e IX give the re1ative rates of reaction for C6-C16 acy1carnitines at substrate concentrations of 500 and 50 HM acy1carnitine. It is observed that in decreasing the substrate concentration from 500 ME to 50 ufl_acy1carnitine, the highest activity shifts from pa1mitoy1— carnitine to 1aury1carnitine. The greatest decreases in abso1ute activity are associated with the 1onger acy1carnitine esters, myristoyI- and pa1mitoy1carnitine, suggesting a significant1y higher Km for these sub- strates. To determine whether the bisphasic kinetics observed with 1aury1- and myristoy1carnitine and the apparent shift to higher Km va1ues for myristoy1- and pa1mitoy1carnitine were attributab1e to a shift from mono- meric to mice11ar forms of the substrate, the critica1 mice11ar concentra- tions (CMC's) of the acy1carnitines were determined and these va1ues compared to the substrate concentrations at which points of discontinuity occur in doub1e reciproca1 p1ots for 1aury1- and myristoy1carnitine acy1- transferase activity. CMC's determined for the various acy1carnitines are shown in Figure 14, indicated by the so1id circ1es. As the acy1 chain 88 Tab1e IX. Rates of Acy1 Coenzyme A Formation from Acy1carnitines Carnitine _?00 HM _1 f? H! -1 Ester nm01 x min x protein . % nm01 x min x protein % Hexanoy1 I 1800 8 920 23 Octanoy1 E 2070 12 1110 27 1 Decanoy1 E 6810 40 3510 87 Laury1 E 6920 41 4050 100 Myristoy1 ' 13210 78 3040 75 Pa1mitoy1 17000 100 3420 85 Assay conditions are described in the Methods except that the detergent was 0.04% Triton X-100. The enzyme protein concentration was 2.0 pg/m1. The % represents the percent of the greatest rate. 89 Figure 14. The re1ationship between the chain 1ength of acy1carnitines to the critica1 mice11ar concentration. The soTid circ1es represent the critica1 concentrations (CMCs) of the even chained acy1carnitines (CMC determinations for acy1carnitines were performed as described in the Methods). The open circ1es represent points of discontinuity on Lineweaver—Burk p1ots of 1aury1-and myristoy1- carnitine acy1transferase activities. The enzymatic reactions were performed (as described in the Methods) at a Triton X-100 concentration of 0.0025% and an enzyme protein concentration of 2.0 pg/m1. 90 IO0,000 g 8 1,0001- IOO‘ Log Concentration (p. molor) ' 10 f2 12 1E CHAIN LENGTH Figure 14 91 1ength decreases from 16 to 8 carbons by successive two carbon units, the CMC concentration rises corresponding1y 10-fo1d, yie1ding a 1inear p1ot of 109 [acy1carnitine] versus carbon chain 1ength. As is shown by the open circ1es in Figure 14, discontinuities in Lineweaver-Burk p1ots of 1aury1-and myristoy1carnitine reactions occur at the same concentra- tions as their respective CMC's. This indicates that a phase shift of the substrate from monomer to mice11e is responsib1e for the kinetic findings. To investigate the effect of detergent mice11es on this phase shift, kinetics measurements using 1aury1-and myristoy1carnitine in the presence of varying concentrations of non-ionic detergents were examined. The data in Tab1e IX are from reactions performed in the presence of 0.04% Triton X-100. Above this concentration the background absorbance at 232 nm of Triton X-100 significant1y interferes with activity determina— tions. Therefore, in order to study the effect of higher detergent concentrations, the non-ionic detergent Tween-20 was used. With either detergent, the effect of increasing the detergent concentration in the assay was to decrease the sharpness of the discontinuity in doub1e reciproca1 p1ots and to 1ower s1ight1y the substrate concentration at which the break occurs. These data for myristoy1carnitine are summarized in Figure 15, where it is seen that at 0.4% (v/v) Tween-20 (the 1imiting concentration due its background absorbance) the break in the Lineweaver- Burk p10t is virtua11y absent. The data points indicated by so1id circ1es represent assays done in the absence of detergent mice11es whi1e those indicated by X's were performed in the presence of 0.4% Tween-20 (CMC = 0.0003%). Identica1 data were obtained using 1aury1carnitine as 92 F .cowpmcpcmocoo ucmmcmumu as“ cw mmmcmcu new: mcauuo zuwacwpcoo -mNu as“ gown: um cowumcpcmucou mpmgumnzm mg“ cw mmcmso esp mucmmmcaw; ocwp :mxogn mg» me x 2 Lo :_muoga a: emu muscwe Lea umELoe .xuwuo~m> . .umpmuwucw mews; vow: mm: o~1cmmzh m¢.o new Faxm: o.N mm: cowumcpcwucoo cwmuoga och .mcwuwccmupxopmwcxs soce cowumELow <00 onpmngs eo muwpmcmx exp :0 omucwmzh mo pummem use .mp mean?» 93 mp mgzmwu 9...): 053500300 .310 _ E 009.0N ON 52; $10 +/ e .5923 o: 89.0. o u Nd 1 0.0 ..l=, 94 substrate except that the break occurs around a substrate concentration of 1000 ufl. The smooth transition of enzyme activity with substrate concentra- tions from be10w to above the CMC for myristoy1carnitine seen in Figure 15 in the presence of 0.4% Tween-20 imp1ies the presence of mixed mice11es in which the acy1carnitineznon-ionic detergent ratio gradua11y rises as the substrate concentration is increased. That acy1carnitines F, wi11 form mixed mice11es be10w their CMC in the presence of mice11es of Tween-20 is shown in Figure 16 which shows that radioactive1y 1abe1ed 1aury1carnitine associates with mice11es of Tween-20 at 0.4% (v/v) as observed during ge1 fi1tration. As the mice11es of Tween-20 pass through the co1umn, monomers of 1aury1carnitine preferentia11y associate with the mice11e so that there is an excess of radioactive 1aury1carnitine where the mice11es e1ute, 1eaving a corresponding deficit of 1abe1ed 1aury1carnitine in its wake. In order to determine the substrate specificity of the so1ubi1ized beef heart mitochondria1 CPT with a11 substrates in the monomeric state and to compare these va1ues to those obtained with the enzyme in a mice11u1ar environment, kinetic studies were performed with a11 substrate concentrations be10w their mice11ar concentrations. Activities for pa1mityocarnitine monomers cou1d not be accurate1y determined due to the 1imited sensitivity of the assay at these concentrations (i.e., be10w 10 0M). Doub1e reciproca1 p1ots of carnitine acy1transferase activity versus concentration of octanoy1-, decanoy1-, 1aury1- and myristoy1carnitine are shown in Figure 17 where the so1id 1ine denoted E indicates reaction in the presence of 0.4% Tween-20 and the broken 1ine denoted E indicates 95 Figure 16. Association of 1aury1carnitine monomers in mice11es of Tween-20. Radioactive 1abe1ed 1aury1 (d1) carnitine was synthesized as described (120, 131). Ge1 fi1tration chromatography was performed as described in the Methods. The number above the dashed 1ine indicates cpm of 14C- 1aury1carnitine bound by mice11es of Tween-20. The number be10w the 1ine indicates the dep1etion of 1aury1carnitine. The 1ower curve represents the presence of mice11es as assayed by the f1uorescence method of Horowitz (131) described in the Methods. The sca1e indi- cates units of re1ative f1uorescence. CPM 96 Y 20,000‘ I5,000‘ V (0,000“ 5,000" ~11- <1- .11. -VI-w ‘ - IO 15 20 25 FRACTION NUMBER Figure 16 3'0 j—_ .__ . L.— .._.'!'I£_‘ 97 .Auv ON-=amzN A>\>v Ne.o to mucmmaca asp mpmuwucw mx mucwmcmumu uwcowuco: mo moppmows we moqunw mg» cw umsgoegma mammmm pcwmwgawc m umxcme mcwp esp use mmpucwu uNFom .»Pm>wuomammg .Ps\m: o.N cam Nmmoo.o mew: :wmuocm.msx~cm ccm oopux copwch we mcowamcucmocou och .muozpmz ms» cw umnwcummu mm teammmm mew: mosz~cm .mgmumw wcwpwcgmuuepu ucm mpg .opu .mu Lee cowpmcpcmucoo mumcpmnzm mzmcw> apwuo~m> cowuuwmg we muopa Fmoocawuwc mpazoo .NP weaved Essa...» I Sow . 000.8. coonoo. 000MB 0 oonw.31 000.00. 000.00 000.3 1 000. o o. 1 Y °_ \\ \ up mg=m_m Weigh—0010.0“— 080» 08.8 08.0. 0 80.6.- 80.8- 08.0»- p ‘1- 1 99. its absence. The kinetic parameters Vmax and Km obtained from these p1ots are presented in Tab1e X. In the absence of detergent, it is seen that the beef heart mito- chondria1 CPT has its 1owest Km for 1aury1carnitine, fo11owed by myristoy1-and octanoy1carnitine. These resu1ts showing a specificity for decanoy1 and 1aury1 esters are in accord with those described for the forward reaction with beef heart mitochondria1 CPT in the previous paper (123) and to those for ox 1iver mitochondria1 CPT reported by West gt 31 (113) for the reverse reaction. The contributions of acy1- carnitine/detergent mixed mice11es to the K.m and Vmax va1ues obtained for reactions performed in the presence of 0.4% Tween-20 may be signifi- cant and have not been ana1yzed further in this study. However, the va1ues given in Tab1e X for reactions done in the presence of detergent are simi1ar to those obtained in its absence: 1aury1carnitine shows the 1owest Km’ and again the reaction with decanoy1carnitine has the highest V x’ though under these conditions, myristoy1carnitine is equa11y as ma active as a substrate. Discussion Kinetic measurements with purified beef heart mitochondria1 carni- tine pa1mitoy1transferase (CPT), revea1ed biphasic Lineweaver-Burk p1ots for the reverse carnitine acy1transferase reaction with 1aury1- and myristoy1carnitine. These resu1ts 1ed us to compare the substrate specificity of beef heart mitochondria1 CPT at different substrate concen- trations and to investigate the ro1es of substrate and detergent mice11es on the enzymatic activity. We are ab1e to resoIve the apparent 100 00.0 _.0 W 0000 00 0000 0_ 00000000: 00.N 0.0 w 0000 00 000N 0.N .00000 00.0 0.0 0000, 000 00N0 00 00000000 mm.P 0.m cop“ mpor come mwp P0000000 any xms> n<~ Ex 0000000 me\~0s: 0000000 msxpos: 0000000 Amv xme> Amy Ex xme> A: V Ex x05> A: 0 Ex mcwpwcgmupzu< 0N10003p 00.0 .m 000000000 oz .< x05 mcwpwcgmopau< soc; m.< 05>~0000000< $0 000005000 0;» L0» waspm> > 000 max :0 om1cmmzh $0 pum$$m .x mpnwh 101 discrepancy between significant1y different specificity profi1es for the reverse carnitine acy1transferase reaction reported for CPT by other investigators by the data presented in Tab1e IX. At a substrate concen- tration of 500 uM_acy1carnitine, our data for beef heart mitochondria1 CPT agreed with those of Kopec and Fritz (90) who reported re1ative rates using ca1f 1iver mitochondria1 CPT of 2, 31 and 100% for octanoy1-, 1aury1- and pa1mitoy1transferase activities, respective1y. At 50 ME acy1- carnitine, the profi1e of substrate specificity shifts to one simi1ar to that described by Nest gt_al. (113) for ox 1iver CPT where they report re1ative rates of 93, 152 and 100% for the same substrates. Our va1ue for octanoy1carnitine re1ative to pa1mitoy1carnitine, 27%, is 1ower than that reported by the 1atter investigations (93%). The va1ues for ox 1iver CPT, however, are 1isted as re1ative maxima1 ve1ocities. The 1ow re1ative octanoy1 activity at 50 RM acy1carnitine is exp1ained by its high Km re1ative to the 1onger chain 1ength acy1carnitines (see data in Tab1e X). From the data depicted in Figure 15, Km va1ues of 14, 84 and 2000 HE myristoy1carnitine are observed, depending on the presence or absence of detergent and on the substrate concentration range chosen. In theory, any intermediate va1ue between 14 and 2000 0M can be obtained by choosing the appropriate concentration of detergent, with va1ues between 14 and 55 0M seen for reactions measured be10w the CMC of myristoy1carnitine and va1ues between 85 and 200 0M possib1e for the mice11ar range of sub- strate concentration. Under these conditions, the Vmax wi11 correspond- ing1y range from 3.0 to 20 U/mg, these being the respective maxima1 ve1ocities extrapo1ated for the monomer and mice11ar portions of the curve 102 obtained in the absence of detergent mice11es. To separate the effect of a mice11ar environment on the enzyme from the mice11ar effects of substrate, submice11ar concentrations of the acy1carnitines were used to determine the kinetic parameters Km and Vmax which are presented in Tab1e X. As seen in the co1umn at the right in Tab1e X, Vmax for octanoy1-and decanoy1carnitine increases by 65% when the enzyme is in a mice11ar environment whi1e the increase for 1aury1-and myristoy1carnitine is approximate1y 3-f01d. The concentrationsof substrates emp10yed in this experiment are just be10w the CMC's of 1aury1-and myristoy1- carnitine but far be10w (by one to two orders of magnitude) those for the octanoy1 and decanoy1 esters so that contributions to Vmax from the forma- tion of mixed mice11es of substrate with detergent wou1d be more signifi- cant for the Ionger chained acy1carnitines and may account for the differ- ence in Vmax ratios. The ratios in Tab1e X of Km for acy1carnitine in the presence and absence of detergent mice11es, however, revea1 a remark- ab1e consistency. That is, the addition of a mice11ar environment reduces the enzyme's apparent affinity for acy1carnitines by five to six-fo1d, regard1ess of the acy1 chain 1ength of the substrate investigated. In either the free state in so1ution or in a mice11ar environment, the enzyme has its greatest affinity for 1aury1carnitine and greatest activity for the reverse reaction with decanoy1carnitine, giving a substrate specificity pattern simi1ar to that reported in the previous paper (123) for the forward reaction using acy1-CoA substrates. SUMMARY AND CONCLUSIONS In summary, two carnitine acy1transferase enzymes have been purified from beef heart mitochondria. Comparison of their substrate specificities with those of enzymes purified by other investigators shows them to be 3" carnitine acety1transferase (CAT) and carnitine pa1mitoy1transferase "ffl£0 (CPT). Partia1 purification of transferase activities from beef 1iver mitochondria a1so reVea1s two protein fractions whose substrate specificity profi1es are very simi1ar if not identica1 to those of the heart mitochon- dria1 enzymes. The pure beef heart carnitine acy1transferase enzymes have simi1ar amino acid compositions, subunit moIecuIar weights, and iso- e1ectric pH's but differ marked1y in their membrane association. The CAT is seen to be membrane-associated and re1eased by sa1t extraction, whi1e the CPT is found to be membrane-bound, requiring membrane disintegration for its so1ubiTization. Carnitine octanoy1transferase (COT) activity is present in signifi- cant amounts in both of the enzymes purified from beef heart mitochondria and in both of the enzyme fractions separated from beef 1iver mitochondria. The two proteins purified from beef heart mitochondria satisfactori1y account for a11 of the COT activity detected in the mitochondria initia11y iso1ated. In beef heart mitochondria, the greater portion of the tota1 COT activity resides with the CAT enzyme, whi1e in beef 1iver the majority is associated with the CPT enzyme. This distribution of COT activity between CAT and CPT enzymes para11e1s the re1ative abundance of CAT and CPT in beef heart and 1iver mitochondria. 103 104 The substrate specificity of purified beef heart mitochondria1 CPT is shown to be dependent upon the presence of mice11es of substrate and of non-ionic detergent. By varying the concentration of substrates and detergent, the differing specificity profi1es reported by other investiga- tors for preparations of purified CPT can be dup1icated. With a11 sub- strates in the monomeric state, in the presence or absence of detergent mice11es, the preferred substrates for the reverse reaction of beef heart mitochondria) CPT are medium chain 1ength fatty acy1carnitines. Medium chain acy1-CoA esters are 1ikewise the preferred substrates in the forward direction. It is conc1uded that no third carnitine acy1transferase enzyme, in addition to CAT and CPT, is present in beef heart mitochondria. Rather beef heart mitochondria1 CPT is itse1f specific for medium chain fatty acy1 transfer when assayed in its pure form or in iso1ated mitochondria. The presence of on1y two carnitine acy1transferases and a simi1ar speci- ficity of CPT for medium chain fatty acy1 transfer is proposed for beef 1iver mitochondria, based on simi1ar fractionation of transferase activi- ties from beef heart and 1iver mitochondria. Because the rate of reaction of beef heart mitochondria1 CPT depends upon the form, monomeric or mice11ar, of the substrate, and because the form itse1f depends upon the substrate's chain 1ength, concentration and hydrophobic environment, the "physio1ogica1 substrate specificity" of CPT in vivg_might be expected to vary with the re1ative abundance of the various chain 1ength fatty acy1 groups. Fina11y, as judged by the 1imited criteria for purity emp10yed in this study--subunit mo1ecu1ar weight, isoe1ectric pH, approach to J. 105 isoe1ectric equi1ibrium, and protein/activity superimposition during chromatography--beef heart mitochondria are found to contain one form of CAT and one form of CPT, supporting the conc1usion that the inner and outer forms of mitochondria) CPT may not be isoenzymes but are the same protein whose kinetic characteristics may be seen to differ under the appropriate assay conditions due to the different membrane and physio- 1ogica1 environments of the two surfaces of the inner mitochondria1 membrane. APPENDIX COLLABORATIVE STUDIES APPENDIX COLLABORATIVE STUDIES Three investigations performed in co11aboration with other workers in Dr. Bieber's 1aboratory and with members of Dr. To1bert's group have been prepared for pub1ication. Copies of two of these are contained in the fo110wing pages (132, 133). The third (134) is in the press. 106 107 Tue Jovauat or Biowcicu Cal-turn Vol, 232. .N'o 22. Issue of November 25. pp. 7930—7931. 1977 Printed .11 L S A Quantitation of Water-soluble Acy1carnitines and Carnitine Acyltransferases in Rat Tissues* (Received for publication. April 1. 1977) Y. R. C1101. P. .1. Focus. P. R. H. Cuuusa. AND L. L. Bianca From the Department ofBiochemistry, Michigan State University, East Lansing, Michigan 48824 The water-soluble acylcarnitines isolated from rat heart. skeletal muscle. liver. and testis have been characterized. The following acyl residues derived from the acylcarnitine fraction were found: acetyl. propionyl. isobutyryl. butyryl. a-methylbutyryl. isovaleryl. tiglyl. caproyl. B-methylcro- tonyl and methacrylyl. The amounts of these acylcarnitines in heart. liver. testis and skeletal muscle from fed rats were determined. Acetylcarnitine was the most abundant acyl- carnitine: however. appreciable quantities of propionyl-. isobutyryl-. isovaleryl-. and tiglyl-carnitine were found. The levels of carnitine octanyltransferse. carnitine acetyltrans- ferase and carnitine palmityltransferase activities were de~ termined in several tissues. In addition. carnitine isovaler- yltransferase and isobutyryltransferase activities were mea- sured in heart. skeletal muscle. liver. testis and kidney. In all instances the specific activity of isobutyryltransferase was similar to the specific activity of acetyltransferase. The results are consistent with the proposal that carnitine is involved in the catabolism of branched-chain amino acids. Carnitine acetyltransferase and carnitine octanyltransfer- ase activities are associated with peroxisomes and microsomes as well as with mitochondria in rat liver (1-5). This multior- ganelle distribution of the octanyl- and acetyltransferase activities indicates that carnitine has roles other than the well established one of translocating long chain acyl residues across the acyloCoA barrier of mitochondria (6-9). ‘ The methods. Results. and References of this paper (including Figs. 1 and 2 and Tables 1 and II) are presented in miniprint at the end of this paper. Full size photocopies are available from the Journal of Biological Chemistry, 9650 Rockville Pike. Bethesda. Md. 20014. Request Document No. 77311-469. cite authortsi. and include a check or money order for 81.20 per set of photoc0pies. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "adiertisemen!" in awordance with 18 U.S.C. Section 1734 solely to indicate this fact. As part of an effort to determine whether carnitine has additional roles in intermediary metabolism. we recently reported the occurrence of 4-carbon and 5-carbon acyl esters of carnitine in beef heart 110-12). Herein. we show that these 4-carbon and 5£arbon acyl esters of carnitine occur in rat heart. liver, skeletal muscle. and testis and that these tissues also contain carnitine octanyltransferase as well as carnitine isovaleryl- and isobutyryltransferase activities. DISCUSSION The volatile fatty acids associated with carnitine in rat muscle. liver. heart. and testis are qualitatively similar to those found in beef heart. The presence of branched chain 4- carbon and 5-carbon acyl derivatives of carnitine in rat mus- cle. liver. heart. and testis is consistent with the previous suggestion that carnitine is involved in branched chain amino acid catabolism 110-12). The presence of large amounts of acetylcarnitine in rat heart. in beef heart (12). in piglet heart (>600 nmol/g of tissueil indicates that acetylcarnitine may function as an immediately available supply of acetyl units that could serve as an energy source during the initial phases of increased energy demands. The occurrence of carnitine octanyltransferase activity in all of the tissues tested indicates a general role in metabolism for this enzyme activity. The finding that the levels of cami- tine isobutyryltransferase activity in heart. muscle. kidney. testis. and liver are similar to the levels of carnitine acetyl- transferase activity while the carnitine isovaleryltransferase activity was much lower is surprising. It could mean that the isobutyryltransferase and acetyltransferase activities are due primarily to the same enzyme. namely. carnitine acetyltrans- ferase. while the isovaleryltransferase activity might be at- tributable to a different enzyme. This is in contrast to the commercial preparation of carnitine acetyltransferase in which the isobutyryl- and isovaleryl transferase activities were much lower than the acetyltransferase activity. 'Y. R. Choi. P. J. Fogle. P. R. H. Clarke. and L. L. Bieber. unpublished data. 7930 “I! II. “M- at qw- fit.“ to rug-nu - was“. if ‘I'Oi I.- V th.’ i hi... I - mousse--1 It. 1'. WW 0 ‘ 'b “- -.9‘10 cum-Im- "can. as not at- ". nae-a a -.u- '|- II on. 1u ' ")1 I e! clue-wu- uI- cow‘- 1'- a- u s- sass-ant ant-s assess". GM '1' *uat a 1t .| us‘ I. t. 9' lua'ts'I't‘I. "Grim-two an I. dev— a‘ outs-mugs .91 - hug-art 1 it- lrsmmmmo‘ I can't ”sow. 'wrn.~1r‘t -'~ 1 is 1e. I Il_~ w-" alt-slave I‘..'I..OW.’13M'p'('.-°.- aneu'lclefl Q- r on scale ---s «m M... “1.0!.“ Io'ee't'elte .10 In. s-ls an flatn- .- vU-sqn’um ’ I I"- at 1-!— ”no. 1. a- "we mun-1 -n m-- M 1m ns-Iw- .'- -ts 0' "I. .- «I! “die .0110 ‘~ W a. Ian-.rp-I" udfllmw»l|l(li‘ on a tee ‘sw as “to "1 nuns. 'h senses-m sues-1“ 'W I sumac-tum! Q's at: can «was «In-.0“) .w 'Isfl 10.n- -- It“. G‘m01~~~~1~ tfldtll.-' Lev "not-1 -M'DN “.r'l at «11- s‘ris '9an I. ‘Q'-- I‘ .‘D o" I II a' ”I t 1 cit-Io "0'. ‘ ‘I‘ ' sub. outs-oi -3 ’II'vs so. thori. ' II C 1:" g - — .r-1tne_‘u~1I- as pump... was ‘ I! fill-la a I use! 1.0! 9 equal I “Irvin. ‘0 lum'tflvfit)nbvtrt‘s "I— -.v-.IvsieI-eayI~—I II. a! “a, O. t. cos. no tot one II II) .‘g-qs amt-s no -11...- Ctr-1l'vot-stt‘. hear-I d I I 'n— s-1I um. It. 01.!- ‘0‘-— u‘mnma'nm-t-ivsn “in-" .IU'OIV.NWII~‘. “TI hut-I haven't-us .09"- lot .v .11 “he seal-I . '9 e-ovl- btnl. '-I - van-0'1“ a, - O "0"!” - “hurrah-s or unit-- M-" in II ~ .w M “totals-.1, .00 .1. than .yxunum u- w 1a 'II 1 1 e i g ' a 'hpes III—s" mole M'IW noun-u 1| can on a ”are. mm C'II- 001'. t-Ia ' I mu O-ID- ‘ Iv- r“ un- .. Ian I! u- t Inn l-II- ' ‘b anon on - '. 'Iooalrvr .- “I. «10- ‘ous-Il In! wtfl I- pets. .‘I"‘. 11 as ‘ .1“ .1 ‘ 1. 4:_1.‘~.'t..t.x 11’)”: I L‘s/«-14 “we? 11. .111 W .fififi "ti—‘1'; ‘ . m“ : ' :IOUI “‘0 :o. 0- 1| hhm-I‘h-ut‘u thsot‘rul-nrulhiw- Clea-QI-aus-e.~t-1 «i.enca Hm'ua°wu.t°nmu.:'1NVK.I'DIPV e-YQIII _v~*~lwm1x,t- rack mU’K.I' ~*‘WIW.."~IW 'I-IOKI'O ’im 10 htiniumumnH-uos.~ Mu-Wh "vo- 1eIIQII‘e-Iala9-e'1."u ‘0' equatrurucu 'v 108 Short Chain Acylcarnitines a! ”(an On urin- ins. tun-n. sua- I”. 1 mm tra- ketv' fl 7 . ' I . - I m I II C ‘0" II ~01 9. '00 3.01 6.. I. I-HI I I - 3 '1 l 4 I '3 1.. - 8.. fl.) “'0‘” 'm G n - II I. I .0 'm ”Fifi ‘C-O’ I‘vll I,.fi .1 “I l—tnl' ‘ D t l' ll I I D ‘0] I OJ 1‘ I "Q." ~— O.” ' OJ LI I 0.. I.) w ". OHIO. ll! I) On loan. sad—es coon-vs t. can. Oars-Malta tea-s. uIII ~‘sv. 'vIeI-Ies.uamvsnst-I1I--t~u as..- t- sI—n Dates- was .0. eaves use 2.- ssr'IuMh-a .1.- I one- I on- at 0 us s—m-s nines “m-IU-I-‘ea-hmmaflnus ”Health-OI ‘- WI Isla "I ton-Iota one “than an. t 'l 'l I 1 'n M v' we Ion-I I-I-e-re 011-": stylus-1th. Ova-r1 wanna-Q-snrmm‘ru: must—14‘- mvr I'm-~10- cau- ' 8"? :1..- spouse'- “1 at ans-9v cm "finl‘W‘II‘. Errand-Q. noes 'rl um"- u 11.1. town-In 'bula blah I he" on" I- - lusts we a-Iau‘ .asie an 11 i I 1 Is I *2_T‘:w‘-'~ "to" Is thou-e- at-“ m t :a .w-H-e n. — 'o'tva- -00- In! ~I arena-'1'." Hun-ti a4 u~'I‘ 11m vase .nI-o' .5— I1- aIIIutI-u‘e-na e- uwutvae -_ mu tea-o'er"- usv 0 "our - I. tune. w. v. - 'hl e‘ t- I!‘ M‘I'l-a .Cy - Imus I0 II sent-us an I'C-I'”.1 the w ‘aI1oe y a «was nan-n ltl’v‘t-Ula'a ms 1,. a. nu- nun-H- Itw erase 'e‘stwa'. to” haul an actual M' '0. III a- a Lou- .IIIO‘ “10‘ um”. ' ..I.O.t.ll a «1 'wav sI- .IIa-OI IUD“.- v'e-e ll'i'l'OI “. s1. - I WI "U' W'I'. urn-narra- est-v1 u N I. n e' we a. ash-N; \- "an“ 'I - I1.- ask-.- I I ‘0 “I 0t. t. “Nels all at". . ts. less) .100. ~01”. III!" .~ 11" :aflttl- umetl'a-a'aui Itu-I'. - '1‘ :1 with. ' UW~1I-I:.‘ van tall-cusses trees .I :1 to v "0' IO— - ”It In I; res-ones ovum-a ’iseat cuss-neat. use-I1 can 0' m-n- unwind-0's. utmvtp 'h been a! (rune “elm- =.m°:.. .uqllmn u t. ”a" at .n .- no 4 W l‘. alumnus-um .I unu-«aut' (bun-1. s- omn- .‘ostglsmrm m an n‘t I- ayttw'au ITN'I‘O- 010 'I ht 1. I- ent-cu . our - '0. .10 - 'eta! -‘I I I have 1t. “we - II as.” ' - 0. his 0. cats ~ I. seer-es ~10. "It. I -110 - I. an -. I-llulb nu fit. “u unu- t-s m .blv'tm- starvin- 7931 Y“ I: bum-a 1.0-rm - was. tau-loan“ In“ use I U. hula. have d that d M cam-ll L '8'” "IV 31' 3:1 111 ii: 1:) mm 1:11-1:91: ugu 101;” Iiigu 'WI' ‘I:I.‘ .01.. (IO... I’ll-OJ lllzll nun-n1 ‘igl. megs! saga.) ugu tog-i 1.6:..3 ‘13:.0 30:3! I':l0 "Due Ina-nineteen 0“:- eases-s. new 01e- I'e-II um: ale-m1- «an I , ‘1‘. Ice '01 as Molas- - h I. s W no: us .I as :s - ”Vt-.0 1- xn't'n uri- l‘li?"|- 0101:"- I V Is!- a- ') I -o. Iago-usu'u-qJ'eI-u-yraw-id'1Iwnwmlt ’0 '0‘. Inn-mluI-oos‘m-Imo-e ‘0'.- .1'a-1‘» VI— a- ".11.- "1-! 0‘“. ct: Inn. - Np ate-o1 U s- .0013 “'C- h1"'\‘fl ml‘ m'hl‘. 412—ens” e’ {3 usage I ‘ II o I’m-Ina Isolates-alre- out. ~ heal— eI-I'Ja-s'm vslm IV . --s~ry|vacs'm QI 1‘ .0 1| v I- anon-teach"- "1*“ hues-ut- an IM'- - .0. 00"" .- Int-I '- ans-NW..- ale-m we -~ ”I eI't 0' Miss a... Q‘s-ICONI Mt. MID-"Q” I- 'II' ass I l OI. - Que.- tnnam e0 -10- m rimill. .m. [0.m.a.s."dn.ll 1"": a. is Qua-HI sash-ant. ”Iteus- wall-I. -‘.0s.wm.s.lum.1tttn1mww 0.0..1. dbl-whit HIV'._l—&.. .. nos . ' 'fll use, ‘1'. ”'u it ' -1 at ..~ - .1) . . NJ b .z-u "we. I on his I 0 a- h. g. 1".” 10.10:“: "ire-III .os : fl. D's-n.- 1.1,. tuners-1.11m”. use [.0 I. mJ '.~’|w.l. $.5-".‘.I-.th"’l’d an. 3.. HI '1“. w. . s u 1.1. LI 1'0!” we. and Ir- tb y.” al- For. .1" n1.- ’9'. 2. In a- Club. ' in n. . .a.‘ . was nah-11,04 HIN- Due: 01...- 'uae .uau ”mus-fl. ”U. I. 1 ._l_. D | (Mia-1a- J. m. 'I’Il 109 Communication Studies on the Oxidation of Isobutyrylcarnitine by Beef and Rat Liver Mitochondria“ (Received for publication. April 2. 1979. and in revised fonn. April 27. 1979) Young R. Choi. Peter R. H. Clarke. and boron L. Bieber From the Department of Biochemistry, Michigan State University. East Lansing. Michigan 48824 Mitochondria from beef 1iver oxidize isobutyrylcar- nitine at approximately 50% the rate of succinate in the presence of rotenone. However, the oxidation rate of lsobutyryl coenzyme A in the presence of I(-)ocarnitine is very low and can be negligible in both rat and beef liver mitochondria. The limited stimulation of isobu- tyryl-CoA oxidation by l(-)-carnitine appears to be due to inhibition of isobutyrylcarnitine translocation rather than lack of formation of isobutyrylcarnitine. This conclusion is supported by the fact that: l) isobu- tyrylcarnitine oxidation is inhibited by l(-)-carnitine; 2) some oxidation of isobutyryl-CoA is obtained when a low concentration (50 pm) of l(-)-carnitine is used; and 3) under conditions of high isobutyryl-coenzyme A and I(-)-carnitine concentrations (1 mat), isobutyryl- carnitine is produced in near theoretical amounts by these rat liver mitochondria. Other studies demon- strated that less than 25% of the carnitine isobutyryl- transferase activity of beef liver mitochondria and rat liver mitochondria is located on the cytosol side of the acylcoenzyme A barrier of these mitochondria. Beef heart (I). as well as testis. skeletal muscle. liver. and heart from rats (2). contain 4ocarbon and 5-carbon branched chain acylcarnitines. These tissues also contain considerable quantities of carnitine isobutyryltransferase and carnitine iso- valeryltransferase activity (2). The occurrence of 4.carbon and 5-carbon branched chain acylcarnitines and branched chain acyltransferase activity in these tissues indicates that carnitine may be involved in the metabolism of the branched chain amino acids. valine. leucine. and isoleucine. Since the oxida- tion of the carbon skeletons of these amino acids occurs partly or entirely in mitochondria, a role for carnitine in shuttling the branched chain acyl residues of cytosolic acyl-CoAs across the acyl-CoA barrier of mitochondria was previously proposed (1). In this paper, we show that beef and rat liver mitochondria readily oxidize exogenous isobutyrylcarnitine. but oxidize much less isobutyryl-CoA in the presence of carnitine. KATERIAH AN D MODS Mitochondria were isolated from rat liver and bovine liver essen- tially as described previously (3). Beef livers were obtained from a ' This work we supported in part by Grant AM 18427 from the National Institutes of Health. This is paper 8653 from the Michigan Agricultural Experiment Station. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement" in accord- ance with 18 U.S.C. Section 1784 solely to indicate this fact. Tu Joraruu. or Blowctcu. Cluster" Vol. 254. No. 13. lane of July 10. pp was). 1979 Printed in USA. local abatoir at the time of slaughter; the liver was sliced into thin strips and put in ice-cooled 0.25 M sucrose buffer for transportation to the laboratory. With succinate as substrate. ADP stimulated the respiration with rat liver mitochondria 2- to 5-fold; and with beef liver mitochondria. the increase was 1.5- to 3-fold. Respiration studies were conducted essentially as described in Ref. 3 except that mitochondria (0.1 to 1.3 ml) were added to a buffered KCI solution to make the final concentration in 4.0 ml as follows: 89 mm KCl. 45 mm Tris-KCI. 3.0 mat MgClz. 6.0 mm P.. pH 7.5. at a temperature of 33°C. Carnitine acyltransferase assays were performed as described elsewhere (2). The relative distribution of carnitine isobutyryltransferase on the cytosol side and the matrix side of the acyl-CoA barrier of beef liver mitochondria and rat liver mitochondria was determined as previously described for the distribution of carnitine palmityltransferase; see Table II of Ref. 3. In order to investigate the production of isobutyrylcarnitine by intact mitochondria given a high concentration (1 mat) of t-(-)- carnitine and of isobutyryl-CoA, tracer amounts of til-[’charnitine were included in incubations identical to those used in the respirations studies. At varying times after the initiation of the reaction by addition of the substrates. aliquots were removed and immediately boiled for 3 min. cooled. and subsequently centrifuged. Tritium-la- beled carnitine and isobutyrylcarnitine were then separated from the supernatant liquid by the system of Bohmer and Bremer (4) and quantitated by liquid scintillation counting of the scraped silica areas (scintillation mixture: 4 g of 2.5-diphenyloxazole and 100 mg of 1.4- bia(2-(4-methyl-5-phenyloxazolyl)lbenzene in 1 liter of toluene with 1 liter of Triton X-lm). When the dI-[JHkamitine preparation used was chromatographed separately or in the presence of supernatant liquid from a boiled mitochondria solution. a contaminant accounting for 1.0% of the total radioactivity was found to co-migrate with the isobutyrylcarnitine standard. Counts measured in silica scraped from the isobutyrylcarnitine area of the thin layer chromatography plate were corrected for this contaminant. Proteins were determined by a modification of the method of Lowry (described in Ref. 5). Isobutyr- ylcarnitine. the I-(-)-isomer was synthesized as described previously for synthesis of crotonylcarnitine (6). Carnitine was the generous gift of Otsulta Pharmeutical. Tokyo. Japan. Male rats. 150 to 200 g. were used in this study. During the long term fasting. 8 days. the animals had free access to water. RESULTS The purpose of this study was to determine if mitochondria from rat liver or beef liver oxidize isobutyryl-CoA or isobutyr- ylcarnitine. or both. As shown in Fig. la. beef liver mitochon- dria oxidize isobutyrylcarnitine. This oxidation is rotenone- insensitive but cyanide-sensitive. However. these same mito- chondria do not oxidize hobutyryl-COA in the presence of [- (-)-carnitine nearly as rapidly, this observation was also made for rat liver mitochondria; see Fig. lb. These experiments were repeated several times with beef liver and rat liver mitochondria and in all instances, the stimulation of oxidation of isobutyryl-CoA by added carnitine was very slight or not detectable, see Fig. lb. The concentrations of isobutyryl-CoA were varied over 2 orders of magnitude with essentially no change. These mitochondria were still capable of B oxidation and the carnitine palmityltransferase was still active under the experimental conditions since addition of palmityl-CoA caused a significant increase in the respiration rate in the presence of l-(-)-carnitine. see Fig. lb. When the experiments shown in Fig. lb were repeated, but 50 uM carnitine was added rather than 900 an carnitine, a small stimulation of oxygen consumption was obtained which increased when isobutyryl- carnitine was added (data not shown). This indicated that the added carnitine might be inhibiting the oxidation of isobutyr- ylcamitine. 5580 'l'lO Oxidation of Isobutyrylcamitine 5581 t' seas a. was.“ a myrytcarni- tinetAlend'nobutyryl-CoAtleyliva mitochondria. The numbers below the betmsindiesu- thenanogramatomsof ensign/minim atoms of protein. In A. 7.1mofbeel'livermitnchondrialproteii werensedinafinalvohrmeoftOmLIn B.5.2mgotratlivermitocbondrialpo- teinfromb-dey-iastedanimalswereuasd inafinalvolumeot4nml.lntbewr curve. 1.3 mu carnitineandinloraer.3.0 mu carnitine. ADP - 0.5 ml; ibcA - 'uobutyryl—COA; pcA - palmityl-CoA; sure - suecinate. 0.3 mil; hen - K‘ cyanide.6mit.'l’henunrbersundernenth arethenenep‘amatemofflysen/min/ n3 atoms of protein. ibc. bobutyrylear- I tracings" aanepsmatomsoteaysen/min/nsatemsot‘pmtein. lnA.6.6mgolmitochondrielprotainine4.0-mlflnalvohimewero usdlnB. DJudmitochondrialpmteinintnmlwer-emsd Diflerentpnparetionsofmitochasdriewsreusedinidandfl. Hue -mlteebendrla;mt-retenens. The pouible inhibition of isobutyrylcarnitine oxidation by l-(-l-carnitine was tested. A eeries of experiments were per- formed which showed that H -)-carnitine, the natural isomer. inhibited the oxidation of isobutyrylcarnitine in both rat liver and beef liver mitochondria. High concentrations of carnitine gave greater inhibition than low concentrations. A typical experiment is shown in Fig. 2a. In contrast, palmityicarnitine oxidation was not inhibited by carnitine; see Fig. 2b. The oxidation of isobutyrylearnitine by rat liver mitochondria from fastedandfedratswascornpared.'l‘heeeresultsarenot shown. In some preparations. mitochondria from fasted ani- mals showed greater oxidation rates than mitochondria from fed animals, but the increase was not large. The very limited oxidation of iaobutyryl-CoA in the pres- ence of I-(-)-carnitine could also indicate that these mito- chondria contidn small amounts of isobutyryltransterase on the cytosol side of the acyl coenzyme barrier. This was tested by determining the per cent or relative distribution of carni- tine isobutyryltransferase on each side of the acyl-00A barrier of both beef liver and rat liver mitochondria. As shown in Table 1. less than 25% of the total mitochondrial transferase activity was detected on the cytosol side of the coenzyme A barrier. These experiments were repeated several times with .entially the same results. in contrast, the distribution of carnitine palmityltransl’erase activity on each side of the acyl coenzyme A barrier was about 1:3 for bovine liver mitochon- dria and near 1:1 for rat liver mitochondria. A similar distri- bution was found previously for rat liver mitochondria (3). In these experiments. glutamic dehydrogenase was used as a marker for mitochondrial breakage and conditions were ad- justed to prevent absorbance changes due to swelling or shrinking of mitochondria The controls. all components pres- ent except 5.53dithiobist2-nitrobensoic acid) showed negligi- ble abeor'oance changes. In the presence of high concentrations (l ml) of l-(--)- carnitine and ieobutyryl-CoA. the “outer" mitochondrial can-ferase is able to produce iaobutyrylcarnitine under the conditions used in the respiration etudiea as demonstrated by the data presented in Fig.3 .The curves containing the solid symbols indicate a near equilibration of the transferase prod- ucts by 30 min and show further that the inn-(erase activity of intact miwhondria '19 significantly less than that of mito- chondria which have been dienrpted with Triton X-100. In an attempt to obtain a more. linear initial rate of isobutyrylcar- nitine producrion and tc quantitate the relative outer and total transfeme activities under these conditions. dilutions of mitochondrial protein of 5-. lcr, and Nick! were made before lnitiatim the incubation reactions with labeled substrate. The data {or e lO-t'old dilution are presented by the open symbols 111 5582 Oxidation of Icobutyrylcamitine suns I Reloriuenmounlofcornia’ne' ‘ ' ,‘ m.—.'.‘..'. W Lu- INA...“ 1 - t , Bedllw Mll- Pad PM he I days In!" Total ectr'v- i- m ol Advhy it scam- } ol‘l'otal X-lm Total shen- ol Total 3- ah-cs d It! My Timx' Salt-ml u“ 1mm” Senora! not/mill. possu- nmol/mill” m ulna/1M ClB'l' 31.3 2.9 0.3 11 0.40 m 0.2 13 154 Cl’l' 32.8 3.3 20.3 8.2 4.3 53.4 10.4 3.3 31.7 GDH 5!!) Nu. dotactod 0 “mtmmmmmchmmmmu-ummmammmmanamum. H O L .- U! L 5 VI A 1 Conversion 0' dl-carnlllne to laobnlynleovnltlne butyrylcarnitl'neliven equalmolarinitialeoncsnuatiouof acyl-00A and l-(-)-oarnitina. Th'm theoretical maximum is doeelyapproximatadinsomsoltheincubstionswhichhave beenperlormed.However.mfireactionsyieldmaximsbe~ tween lbsndMasinFi‘fidueprohablytofl-Jpresencsof hydrolases.thsscyl-CoAhydrolasm.drersmov-lot‘oneo! thambnamandthsubsequentshilfin‘oflhsequilibrium Discussion “Ismailninl’is. ldsmonm'atathstbedlivsrmitochon- h- _lal. tlismtsofnrcdnateoxidstiom'l‘hsrotenoneimenaitivityot thhoxidstioniscons'ltantwiththeflavoprotoin-linkedde- hydrocenaseinvolvedinthsfirstuspintheoxidationol iohtyryl-CoAT‘hedatap'veninTableldmwtbatthsse thanontbsmatrixn'ds,theratioo(activitiss'naboutl:5. 1 t 0‘ 20 ~00.th YI-o ll I‘IIIOI ”3 LL .4: l' l ' l L l-(-)-carnitrne by rat liver mitochondria. Incubation Ice Identical tothose ormsdin the mud-swab theesuption that H— —)-carnitine (1 mill. hobutyryl-CoA (1 ml), and til-{'lflcarnitine ReactionmixtuL'CmelThssymbobars hum-ms. milochondrianJSTriumX-lmzwng/mlolprotain. memo—Quito Hint-ct 2.10 chondfia+0.1$1‘rihnX-1lll.031m¢/mlolMD—U.intact olprotein; ----- ,lsestsquar-le'rI-‘on lineforT-Ztowmm. ian 3. Anslysisot'thaleutsquarssrs‘r-‘onlinesforthe anenimatethstintact isfound in the mitochondrial prepare. has been disruptod by the amount found with the M’dithiobiam-nitrobsnaoic acid) amsyshowninstleLl-lowevmttoflofthsdutamic dehydrogenasewasfotmdinthesolubleh'actionatthesndo! theexperhnennlndicatin‘some d ' 1"— nobotyryltranaferasefoundini‘ig. 3. The equilibrium constantfcrtheresctlon: scylcarnitine+ CoASl'lucarmunev-ecleoAhaabssnrspoI-tsdtobeos b rssu\l|awuv Bys:suse d-(+)-carnitinsisnotanrbshatefortheacetylhsns~ ferase(8).onscanpsdictthatatsquilibfium.23$o{lhe tritium-labeled dl-carnitins will be converted to labeled iso- rsqrectfully. 1‘hafacttbstapprntimatdylfiot'thetsobutyryltrsnder- —"“"‘,u ' isapparentiy located A I'J an al. tin A barrieriscona'ntent with the limited oxidation olieobutyryLCoA in the presence investigated by Runny and Tubbs (10). To our knowledge. arch an exchange system has not been demonstrated for liver ' ondria. The lack of stimulation of oxygen consumption by added carnitine in the presence of isobutyryl-CoA and the inhibition of isobutyrylcarnidne oxidation by added carnitine could be caused by inhibitionofthefnrmationotisobutyryl- CoAinthu .matris. Wouldoccmiftheconcentrationolcarnitinein the matrix were elevated to a level whereby the equilibrium of the reaction: isobutyrylcarnitine + CoASH es arnltine + hobutyryl-CoA precluded formation of nbstrate amounts of isobutyryl-CoA. Thisseemsunllkely because: carnitinedidnotinhibitpalmi tylcarnitine oxidation and accumulation of carnitine by liver mitochondria has not been red The data imply 112 Oxidation of Isobutyrylcamitine that mitochonm'is would oxidise little hobutyryl-CoA in con. tact with the cytosolic side of the inner membrane of mito- chondria when comiderable carnitine is present. In other studies. data to be published elsewhere, we have found that carnitine isobutyryltrsnaferase is also amociated with peroxio somes; thin. isobutyrylearniu'ne might be formed in other cellular compartments as well as in mitochondria. The complete aubcellular distribution of the enzymes in- volved in valine metabolism. a source of isobutyryl coenzyme A. is not known. Considerable quantities of the a-keto acid dehydrogenase that oxidizes the keto acid derived from vaiine are amociated with mitochondria (11), but po-ibly on the cytosol side of the inner membrane (12). The activation of abortchainfatxyacidscanoccurinthe matrixofbeefliver mitochondria (13). Thus, some isobutyryl-CoA could be formed directly from the free acids in the matrix of mitochon- dria; however. the need for large amounts of isobutyryltr'ans- ferase in this compartment is not apparent. One possibility 'u that carnitine acyltransferases such as carnitine acetyltrans- ferase, carnitine isobutyryltr'ansferase. and carnitine octanyl- ti-anst‘eraseactivitiesareneededtoemirethstmitochondria and other cellular compartments always have adequate amounts of free CoASH. This could be accomplished by ensuring that mitochondria and other cellular compartments always have achquate amounts of free CoASH. This could be accomplished by maintaining a relatively constant ratio of free CoASH to acyLCoA. Since the equilibrium constant for the carnitine acyltransferases is near unity and the amount of carnitinegreatlyexceedsthetotslamountofcoenzymeAin mammaliantiesireathepresenceofabroadspecmimof carnitine acyltransferase activity and free carnitine could an- able the cell to maintain adequate levels of free reduced coenzymeAIncertsinmetsbohcainnfiomtheahoi-tchain acylcarnitinesmightbesxportedfromthematrixofmito— chondriainsxchanget‘orfieecarnitinewhicheouldbeusedto maintain the CoASH/acyl-CoA ratio as recently suggested for 5583 skeletal mucle (14). The acylcarnitinescouldnrbeequently beutilisedastheretioofCoASH/scyl-CoAiircreaseaSucha gsneralrolewouldstilibecomistentwiththespecifichmco tions for carnitine such as tramlocsting long chain acyl r.- duesacromtheinnermembraneofmitochondrisandforthe rolesproposedinhrsnchsdchsinaminoacidmetabol'nn. REFERENCES 1. Bieber.LL.andChoi.Y.(1977)Proc.NorLAcod.Sci 11.8.4. 74, 2795-2798 2. Choi. Y.. Fogle. P.. Clarke. R. and Bieber. 1.. L (1977) J. Biol. Chem. 252. 7930-7931 3. Bieber.LL.MarkwelLM.K..Blair.M..andl-lelmrath.T.A. (1973) Biochim. Biophys. Am 836. 145-154 4. Bohmer. J.. and Bremer. J. (1968) Biochim. Biophys. Acto 1B. 5. HarkwelLM. AK. lines. 3.. Bieber. LLandTolbertN. E (1978)Ariol.8ioclian.87. 206-210 6. Choi.Y.. andBieber. LL (1977)AnaLBrocliaii.73. 03-418 7. m1. 8.. Schultz.S. anndSrere. P. A. (1963)J.Br'oLCIieiii. “2509-2517 8.1’ritt..1.B ..andSchultx.S.lL(1965)JBioLOran.M2183- 2192 9. Panda. S. V..andParvin.R.(1976)JBioLCAan.367.6683—6591 10. Benny. R R, uid'hrbbe. P. K (1976) Eur. J. Biochem. 60.299- 303 11. Shinneck. P. L. and Harper. A. P. (1976) Biochim. Biophys. Acto 437. 477% 12. Johnson. W.A..andConnelly.J.1.. (1972)Br’ocliaar'soylo. 1967- 1973 13. Grooty. P. H. F... Scholte. H. K. and Hit-hon. W. C. (1976) Adv. Lt'pr'dRes. 14. 75-126 14. vanHarsbsrgltV.W..Veerkamp.J.H..Enelen.P.M.J..and Gum w. J. (1978) Biochan. that to. 115—124 15. Markwell. M.A.. Bieber.1..L.andTolbert.N.E.(1977) Biodieis. Phamaaol. 6. 1697-1702 16. Lassrow. P. 8.. and Deduve. C. (1975) Proc. Natl. Acnd. Sci. U. S. A. 73. 2043-2046 17. Markwell.M.A. LMcGroarty.1-1J..Bieber.1.1..and‘l‘olbsn. ME (1973)J.Bt'ol. MMW 18. team, P. M. (1978) J. Biol. Chem. 263. 1522-1528 REFERENCES (”VG 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. REFERENCES . Gu1ewitsch, w. and R. Krimberg, Z. Physio1. Chem. 35 326 (1905). . Strack, E. and Foster1ing, K., Arch. Expt1. Patho1. Pharmako1. 185 612 (1937). . Carter, H. E., Bhattacharyya, P. K., Weidman, K. R. and Fraenke1, 6., Arch. Biochem. Biophys. 38 405 (1952). . Fraenke1, G., B1ewett, M. and Co1es, N., Nature 161 981 (1948). . Friedman, S. and Fraenke1, G., Arch. Biochem. Biophxs._§9 491 (1955). Fritz, I. B., Amer. J. Phx5101. 197 297 (1959). Warshaw, J. B., Deve1opmenta1 Bio1qu.g§ 537 (1972). Oram, J. F., Bennetsch, S. L. and Nee1y, J. R., J. Bio1. Chem. 248 5299 (1973). Bremer, J., Nature 196 993 (1962). Bremer, J., J. Bio1. Chem. 238 2774 (1963). Fritz, I. B. and Yue, K. T. N., J. Lipid Res. 5 279 (1963). Bremer, J., J. Bio1. Chem. 237 2228 (1962). Fritz, I. B., Schu1tz, S. S. and Srere, P. A., J. Bio1. Chem. 238 2509 (1963). Fritz, I. B. and Schu1tz, 5., Fed. Proc. 22 637 (1963). Fritz, I. B., Advan. Lipid Res. l_285 (1963). Bremer, J., Biochem. J. 89 25p (1963). K1ingenberg, M. and Pfaff, E., Symp. Reg. Metab. in Mite, Bari, E1sevier Amsterdam (1965). Hoope1, C. L., The Enzymes of Bio1ogica1 Membranes, vo1ume 2, Ed. A. Martonosi, P1enum, New York, p.119 (1976). Ramsay, R. R., Biochem. Soc. Trans. 6 72 (1978). 113 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 114 Pande, S. V., Proc. Nat1. Acad. Sci. (USA) 12 883 (1975). Markwe11, M. A. K., McGroarty, E. J., Bieber, L. L. and To1bert, N. E., J. Bio1. Chem. 248 3426 (1973). Pearson, D. J. and Tubbs, J. K., Biochem. J. 105 953 (1967). Snoswe11, A. M. and Henderson, G. 0., Biochem. J. 119 59 (1970). Bress1er, R. and Brende1, K., J. Bio1. Chem. 247 4092 (1966). Koh1haw, G. B. and Tan-Ni1son, A., J. Bacterio1ggy 129 1159 (1977). Markwe11, M. A. K., ToTbert, N. E. and Bieber, L. L., Arch. Biochem. Biophys. 176 479 (1976). Emaus, R., persona1 communication (1981). Hitte1s, B. and Hochstein, P., J. Bio1. Chem. 242 126 (1967). Hitte1s, B. and B1um, J. J., Biochim. Biophys. Acta 152 220 (1968). 8iebeg, L. L., Abraham, T. and He1mrath, T., Ana1. Biochem. 50 509 1972 . Yates, 0. w. and Garland, P. 8., Biochem. J. 119 547 (1970). Bremer, J. and Norum, K. R., J. Bio1. Chem. 242 1744 (1967). Srere, P. A., Seubert, w. and Lynen, F., Biochim. Biophys. Acta_§§ 313 (1959). So1berg, H. E., Biochim. Biophys. Acta 360 101 (1974). Hoppe1, C. L. and Tomec, R. J., J. Bio1. Chem. 247 832 (1972). Wood, J. M. and Chang, C. H., Fed. Proc. §4_673 (1975). Normann, P., Ingebretsen, 0. C. and F1atmark, T., Viochim. Biophxs. Acta 591 286 (1978). 8ergs§rom, J. D. and Reitz, R. C., Arch. Biochem. Biophys. 204 71 1980 . {de11-Henger, J., Oram, J. F. and Nee1y, J. R., Fed. Proc. 34 480 1975). Bremer, J. and Norum, K. R., Eur. J. Biochem. 1 427 (1967). Fritz, I. B. and Marquis, N. R., Proc. Nat1. Acad. Sci. (USA) 54 1226 (1965). 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 115 Fritz, I. B. and Schu1tz, S. K., J. Bio1. Chem. 240 2188 (1965). Wood, J. M., Biochemistry 12 5268 (1973). Higgins, J. A. and Barrnett, R. J., J. Ce11 Sci. 6 29 (1970). Makita, T., Kiwaki, S. and Sandborn, E. R., Histochemistry J..§ 335 1973 . Barker, P. J., Fincham, N. J. and Hardwick, D. C., Biochem. J. 110 739 (1968). Snoswe11, A. M. and Koudakjian, P. P., Biochem. J. 127 133 (1972). Brdiczka, 0., Gerbitz, K. and Pette, 0., Eur. J. Biochem. ll_234 1969 . Chi1dress, C. C., Sacktor, B. and Traynor, D. R., J. Bio1. Chem. 242 L 754 (1967). Banks, S. M. and Chappe11, J. B., FEBS Letters 59 230 (1975). Harshaw, J. B., Biochim. Biophys. Acta 223 409 (1970). Tubbs, P. K. and Chase, J. F. A., Biochem. J. 100 48p (1966). Beenakkers, A. M. T. and K1ingenberg, M., Biochim. Biophys. Acta .84 205 (1965). Aas, M., Biochim. Biophys. Acta 231 32 (1971). Martin, P. A., Temp1e, N. J. and Connock, M. J., Eur. J. Ce11 Bio1. _19 3 (1979). Rossi, C. R. and Gibson, 0. M., J. Bio1. Chem. 239 1694 (1964). Fog1e, P. J. and Bieber, L. L., Int. J. Biochem. g 761 (1978). Norum, K. R., Biochim. Biophys. Acta 98 562 (1965). Farre11, S. 0. and Bieber, L. L., submitted for pub1ication (1981). Kohonen, M. T., Biochim. Biophys. Acta 428 690 (1976). Markwe11, M. A. K., Bieber, L. L. and To1bert, N. E., Biochem. Pharmaco1. 26 1697 (1977). Patterson, J. E., Biochim. Biophxs. Acta 306 1 (1973). Ko1attakuddy, P., persona1 communication. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 116 Yates, 0. w. and Gar1and, P. B., Biochem. Biophys. Res. Commun. 23 460 (1966). Harano, Y., Kowa1, J. and Mi11er, M., Fed. Proc. 31 863 (1972). anzeg, R. B., Have1, R. J. and McI1roy, M. B., Neuro1ogyI§g 627 1980 . Bieber, L. L., Markwe11, M. A. K., B1air, M. and He1mrath, T. A., Biochim. Biophys. Acta 326 145 (1973). Patten, B. M., Wood, J. M., Harati, Y., Hefferan, P. and Howe11, R. R., Amer. J. Med. 61 167 (1979). McGarry, J. 0., Leatherman, G. F. and Foster, 0. H., J. Bio1. Chem. ‘1 .253 4128 (1978). Swierczynski, J., Seis1owski, P. and A1eksandrowicz, 2., Biochem. _ .Mgg.‘l§ 55 (1976). 'i ‘ Hahn, P. and Seccombe, D., Carnitine Biosynthesis, Metabo1ism and Functions Ed. R. Frenke1 Academic Press, New York p. 177 (1980). Augenfer, J. and Fritz, I. 8., Can. J. Biochem. 48 288 (1970). Hahn, P. and Ska1a, J. P., Can. J. Phys. Pharm. 59 355 (1981). Harano, Y., Kowa1, J., Yamazaki, R., Lavine, L. and Mi11er, M., Arch. Biochem. Biophys. 153 426 (1972). Ishii, H., Fukumori, N., Horie, S., and Suga, T., Biochim. Biophys. Acta 611 1 (1980). No1fe, R. G., Maxwe11, C. V. and Ne1son, E. C., J. Nutr. 108 1621 (1978). Apri11e, J. R., J. Ce11 Bio1. 67 11A (1975). TomecS R. J. and Hoppe1, C. L., Arch. Biochem. Biophys. 170 716 1975 . Pace, J. G. and Nannemacher, R. N., Fed. Proc. 32 1727 (1980). Boudin, G., Miko1, J., Gui11ard, A. and Enge1, A. G., J. Neuro1. Sci, 30 313 (1976). Scho1te, H. R., Meijer, A. E. F. H., van Nijngaarden, G. K. and Leenders, K. L., J. Neuro1. Sci. flg_87 (1979). DiMauro, S. and DiManro, P. M. M., Science 182 929 (1973). Hostet1er, K. Y., Hoppe1, C. L., Romine, J. S., Sipe, J. C., Gross, S. and Higginbottom, P., C1in. Res._§§ 125A (1977). .-I- 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 117 Scho1te, H. R., Jennikens, P. G. I. and Bonvy, J. J. B. J., J. Neuro1. Sci. 49 39 (1979). Reza, M. J., Kar, N. C., Pearson, C. M. and Karp, R. A. P.,_Ann. Intern. Med. 88 610 (1978). Bank, N. J., DiManro, S., Boni11a, E., Capuzzi, D. M. and Row1and, L. P., N. Eng]. J. Med. 292 443 (1975). Chase, J. F. A. and Tubbs, P. K., Biochem. J. gg 32 (1966). Chase, J. F. A. and Tubbs, P. K., Biochem. J. 111 225 (1969). Edwards, M. R., Proc. Anst. Biochem. Soc. 1g 18 (1977). Kopec, B. and Fritz, I. B., Can. J. Biochem. 49 941 (1971). Bremer, J. and Norum, K. R., J. Bio1. Chem. 242 1749 (1967). Cheema-Dhad1i, 5., Robinson, B. H. and Ha1perin, M. L., Can. J. Biochem. 54 561 (1976). Be11, F. P., Atherosc1erosis 37 21 (1980). Hood, J. M. and Merri11, J. M., Fed. Proc. 32 363 (1973). Wood, J. M., Sordah1, L. A., Lewis, R. M. and Schwartz, A., Circ. Res. 32 340 (1973). McGarry, J. D. and Foster, 0. N., Diabetes 29 236 (1980). Kopec, B. and Fritz, I. B., Can. J. Biochem. 49 941 (1971). Cook, 5. A., Otto, D. A. and Corne11, N. w., Biochem. J. 192 955 (1980). Rebouche, C. J. and Enge1, A. G., Fed. Proc. 39 1045 (1980). Markwe11, M. A. K., Ph.D. thesis, Department of Biochemistry, MSU 1974. Edwards, Y. H., Chase, J. F. A., Edwards, M. R. and Tubbs, P. K. Eur. J. Biochem. 4§_209 (1974). So1berg, H. E., Acy1 Group_§pecificity_of Carnitine Acy1transferases, Ph.D. Thesis Institute of C1inicaTBiochemistry, University of 0s1o (1974). Chase, J. F. A., Pearson, D. J. and Tubbs, P. K., Biochim. Biophys. Acta 26 162 (1955). 104. ‘105. 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. 118 So1berg, H. E., FEBS Letters 12_134 (1971). Markwe11, M. A. K., McGroarty, E. J., Bieber, L. L. and To1bert, N. E., Fed. Proc. §g_671 (1973). Va1kner, K. and Bieber, L. L., Fed. Proc. 49 1643 (1981). Grunnet, I. and Knudsen, J., Biochem. Biophys. Res. Commun. 80 745 (1978). Lazarow, P. B., J. Bio1. Chem. 253 1522 (1978). 0smundsen, H., Neat, C. E. and Norum, K. R., FEBS Letters 22 292 1979 . C1arke, P. R. H., persona1 observation from 1979 Meeting of the American Society of Bio1ogica1 Chemists in Da11as, Texas. Norum, K. R., Acta Physio1. Scand. 66 172 (1966). Norum, K. R. and Bremer, J., J. Bio1. Chem. 242 407 (1967). West, 0. N., Chase, J. F. A. and Tubbs, P. K., Biochem. Biophys. Res. Commun. 42 912 (1971). Mitta1, B. and Kurup, C. K. R., Biochim. Biophys. Acta 619 90 (1980). Pa1mer, J. N., Tand1er, B. and Hoppe1, C. L., J. Bio1. Chem. 252 8731 (1977). Nood,)J. M., Ho1kowicz, P. E. and Chu, A., Circ. Supp1._§§ II-115 1978 . Pande, S. V. and B1anchaer, M. C., Biochim. Biophys. Acta 202 43 1970). Reitz, R. C., Fed. Proc. 36 331 (1977). C1arke, P. R. H. and Bieber, L. L., Fed. Proc._§§ 665 (1979). Bohmer, T. and Bremer, J., Biochim. Biophys. Acta 152 559 (1968). Bremer, J., Biochem. Prep. 12 69 (1968). Laemm1i, U. K., Nature 227 680 (1970). C1arke, P. R. H. and Bieber, L. L., J. Bio1. Chem. 256 9861 (1981). Bohme, H. J., Kopperschager, G., Schu1z, J. and Hofman, E. J. Chromatography 69 209 (1972). 119 125. Vesterberg, 0., Methods in Entymo1ogy XXII 389 (1971). 126. Boudhuin, P., Beaufay, H., Rahman-Li, Y., Se11inger, 0. Z., Nattiaux, R., Jacques, P. and de Duve, C., Biochem. J. g; 179 1964 . 127. Lowrey, 0. H., Rosebrou h, N. J., Farr, A. L. and Randa11, R. J., J. Bio1. Chem. 193 265 11951). 128. Boh1en, P., Stein, 5., Dairman, N. and Undenfriend, 5., Arch. Biochem. Biophys. 155 213 (1973). 129. Springer, T. A., Mann, 0. L., DeFranco, A. L. and Strominger, A. L., J. Bio1. Chem. 252 4682 (1977). 130. C1arke, P. R. H. and Bieber, L. L., J. 8101. Chem. 256 9869 (1981). 131. Horowitz, P., J. Co11oid Interface Sci. 61_197 (1977). 132. Choi, Y. R., Fof1e, P. J., C1arke, P. R. H. and Bieber, L. L., J. Bio1. Chem. 252 7930 (1977). 133. Choi, Y. R., C1arke, P. R. H. and Bieber, L. L., J. Bio1. Chem. 2§4_5580 (1979). 134. Krah1ing, J., C1arke, P. R. H., Va1kner, K., To1bert, N. E. and Bieber, L. L., Archives of Biochem. Biophys. (in press).