L. I‘ I. V ,1. J .. 1 4‘. ~£ 'h M {mung-1 .— s ‘ .x x ‘ .& war. - :5 I? ' \ 1.1' "'1 A. i W; V1.5 :5 :w' w: I. ‘ : ‘5 g‘ufi'w-Ju ‘ a“ ‘3: " “‘4‘!" 3" ~ '1 . ‘.. "35»: v A' A x ¥g,t,g::.zh.\:§ . ~ 4:3“ .. 1‘. 3w 1 ., < v. _ 37“”:w '. '5' 34‘ ‘34-;qu 15$. "4 ‘I‘ v u n I! II‘:‘ , .1: 0 Jv .139 \d‘ *1- f: s 7‘5 ‘.'."."‘“..‘ \" d h ‘4’. I. . ’ n :1" in. '3' y.’ — . . K J" .4 - r." -7” , 7 ... u _ _ ,, . .- “Ha! ‘ ‘ - , ”1'4,” , -, . “' ,........!.. .. . V . ’I',\ny w .1 .,_ . .F . 4 . . . . .uu- w. .u. H . . . . .v, xr- ,_ .u - n». -.- . n. A’x'n ’-'. .. 1 ‘ -.... 4.. ”9 ,. . {1:13.31} :;.' ,3; “.0. m... .2 . . <3 --‘«'- ‘ 'Ihkrv -- a. vrluI/v- .1 o....u-..n. :p.‘v’via ; . -. .. . .- ~ 4.. wpvl¢m ...‘,., ,‘ ¢ ‘.. ’1‘ .7? . "' -..» ".5. * "r w. ' r » . - .. ... - ., . ,.,»...—._J ”H".-. . ‘.' "J . .3. -.-( V". , up”. . ,... . 'OrhvaI-v'. -!~-arvr -> .. -7.».-.. ' N-‘r --o. NEW RS SITY LIBRARI lllllllliltlllll‘llll’lllllllll llllll lllllll 3129300914527 lifillHi This is to certify that the dissertation entitled Characterizati in of a Medium/Long-Chain Garnitine Acyltransfcrase Associated with Rat Liver Endoplasmic Reticulum presented by Kathleen Lilly has been accepted towards fulfillment of the requirements for Ph.D. Biochemistry degree in Major professor Date Sept. 6, 1990 MS U i: an Affirmative Action/Equal Opportunity Institution 0-12771 f LIBRARY Mlchlgan State 1 University ————_ W PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or betore date due. DATE DUE DATE DUE DATE DUE T F—Wfi __J= II Ii - —_ _ _ MSU Is An Affirmative AetionlEqual Opportunity Institmion extrema“.- CHARACTERIZATION OF A MEDIUM/LONG-CHAIN CARNITINE ACYL'I'RANSFERASE ASSOCIATED WITH RAT LIVER ENDOPLASMIC RETICULUM By Kathleen Lilly A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Biochemistry 1990 ABSTRACT CHARACTERIZATION OF A MEDIUM/LONG-CHAIN CARNITINE ACYLTRANSFERASE ASSOCIATED WITH RAT LIVER ENDOPLASMIC RETICULUM By Kathleen Lilly Rat liver contains camitine acyltransferase activity with a medium to long acyl-chain length specificity located in both rough and smooth endoplasmic reticulum that is strongly inhibited by malonyl-CoA, although the existence of medium/long-chain camitine acyltransferase activity in the endoplasmic reticulum has been disputed. This thesis reports characterization of a microsomal medium/long-chain camitine acyltransferase that is membrane bound. It is designated microsomal camitine octanoyltransferase (COT) because it has a higher specific activity with medium-chain acyl-residues such as decanoyl-CoA than with palmitoyl- CoA. Microsomal camitine octanoyltransferase is not immunoprecipitated by antibody prepared against mitochondrial camitine palmitoyltransferase and it is only slightly immunoprecipitated by antibody prepared against peroxisomal camitine octanoyltransferase. This demonstrates it is antigenically distinct from either of the other liver camitine acyltransferases with medium to long acyl-chain length specificity. This characterization provides evidence that microsomal COT is distinct from mitochondrial and peroxisomal medium/long-chain camitine acyltransferases . The concentration of malonyl-CoA required for 50% inhibition is 5.3 11M; in the presence of 17 M decanoyl-CoA and 1.7 mM L-carnitine reSpectively. Microsomal camitine octanoyltransferase is also inhibited by etomoxiryl-CDA, with 0.6 uM etomoxiryl-CDA producing 50% inhibition. Although palmitoyl-CoA is a substrate at low concentrations, the enzyme is strongly inhibited by high concentrations of palmitoyl-CoA; 50% inhibition is produced by 11 uM palmitoyl-CDA. Microsomal COT is inhibited competitively with respect to L-camitine by DL-aminocarnitine; 0.5 mM DL-aminocamitine produces 50% inhibition. Microsomal COT follows Michaelis-Menten kinetics with Hill coefficients of 0.91 for decanoyl-CoA and 0.96 for L-camitine. The kinetics constants for microsomal COT show the K05 for L-carnitine is 0.42 mM and 1.9 M for decanoyl-CoA. The kinetic characteristics of microsomal COT can be used to distinguish it from mitochondrial CPT and peroxisomal COT. To my Grandmothers Mary Nicolls and Kathryn Lilly iv Acknowledgments The support and guidance of my thesis advisor, Dr. Loran Bieber, is gratefully acknowledged. I wish to thank the members of my guidance committee: Drs. Alan Davis, Jack Holland, Estelle McGroarty, and Shelagh Ferguson-Miller. Special thanks go to Drs. John Chimoskey and James Trosko. I wish to thank the past and present members of the laboratory for their help: Wieslawa Lysiak, Micheal Healy, Sue Drombroski, Janos Kemer, Sue Leek, Fabio Di Lisa, Andrea Cress, Chang Chung, Rena VenRenterghem, Gonghe Dai,and Lori Kurth. I thank my family especially my parents for their patient love and support. I am especially grateful to my husband, Paritosh, for his love and encouragment. Table of Contents Page List of Tables . X List of Figures . xi List of Abbreviations xiii Introduction 1 Methods and Materials 4 Isolation of Microsomes 4 Enzyme and Protein Assays 6 Immunoprecipitation 7 Labeleing of Liver Microsomes with [3Hl-Etomoxir 8 Determination of Kinetic Constants 9 Ammonium Sulfate Fractionation of Detergent Solubilized 11 Microsomes Column Chromatography 11 Materials 12 Chapter 1. Literature Review 14 Assay Methods for Camitine Acyltransferases 16 Distribution of Medium/long-chain Camitine Acyltransferase 19 Activity in Rat Liver vi Page Tissue Distribution of Microsomal Medium/long-Chain 22 Camitine Acyltransferase Activity Effect of Drug Treatment and Feeding/Fasting on 23 Medium/long-chain Camiime Acyltransferase Activity of Liver Microsomes Kinetics of Medium/long—chain Camitine Acyltransferase 24 Activity in Microsomes Solubility and Stability Characteristics of Microsomal 24 Medium/long—chain Camitine Acyltransferase Activity Palmitoyl—CDA Inhibition of Medium/long-chain Camitine 25 Acyltransferase Activity in Rat Liver Aminocamitine as Substrate and Inhibitor of Medium/long- 25 Chain Camitine Acyltransferase Activity in Rat Liver Malonyl-CoA Regulation of Medium/long-chain Camitine 27 Acyltransferase Actvitiy in Rat Liver Etomoxiryl-CoA Inhibition of Medium/long-chain Camitine 29 Acyltransferase Activity in Rat liver Chapter 2. Malonyl-CoA Inhibition of Medium/long-chain Camitine 32 Acyltransferase Activity of Rat Liver Microsomes Introduction 33 Results 35 Distribution of Malonyl-CoA Sensitive Medium] 35 long-chain Camitine Acyltransferase Activity in Rat Liver Malonyl-CoA and Etomoxiryl-CoA Inhibition of 37 Microsomal COT Palmitoyl-CoA Inhibition of Microsomal COT 46 Immunoprecipitation of Micrsomal COT with 49 Antiperoxisomal-COT and Antimitochondrial CPT vii Solubility and Stability of Microsomal COT [3H]-Etomoxir Labeling of Rat Liver Microsomal Proteins Discussion Localization of Malonyl-CoA Sensitive COT Chapter 3. Kinetic Characterization of Membrane Bound Microsomal COT Introduction Results Determination of Kinetic Constants Effect of pH on Microsomal COT Activity DL—Aminocamitine Inhibition of Microsomal COT Decanoyl—DL-Aminocarnitine and Palmitoyl- DL-Aminocamitine Inhibition of Microsomal COT Discussion Aminocamitine Inhibition Chapter 4. Attempted Purification of Microsomal COT Introduction Results Preparation of Microsomes Solubilization Ammonium Sulfate Fractionation Column Chromatography viii Page 53 53 62 62 68 69 71 71 76 76 76 85 86 9O 91 93 93 94 97 97 Page Restoration of COT Eluted from an Anion Exchange 103 Column Discussion 104 Chapter 5. Summary and Conclusions 107 Possible Fuctions of Microsomal COT 110 Future Research 112 List of References 114 List of Tables Chapter 1. Literature Review Table 1. Summary of Assay Methods for Camitine Acyltransferases Chapter 2. Malonyl-CoA Inhibition of Medium/long-chain Camitine Acyltrasnferase Activity of Rat Liver Microsomes Table 1. Distribution of COT Activity, Malonyl-CoA Sensitivity of COT, and Marker Enzymes in Fractions of a Rat Liver Homogenate Isolated by Differential Centrifugation Table 2. Malonyl-CoA Inhibition of COT Activity of Microsomes Isolated by Three Different Procedures Table 3. Detergent Solubilization of Microsomal COT Chapter 4. Attempted Purification of Microsomal COT Table 1. Effect of pH on the Solubility of Microsomal COT Chapter 5. Summary and Conclusions Table 1. Summary of the Properties of Rat Liver‘Mitochondrial CPT, Peroxisomal COT, and Microsomal COT Page 18 36 38 54 96 108 List of Figures Page Chapter 1. Literature Review . Figure 1. Structure of Etomoxiryl-CoA 31 Chapter 2. Malonyl-CoA Inhibition of Medium/long—chain Camitine Acyltransferase Activity of Rat Liver Microsomes Figure 1. Effect of Malonyl-CoA on the Production 40 of [1-"C1-Decanoylcamitine Figure 2. Malonyl-CoA and Etomoxiryl-CoA Inhibition of 42 Microsomal COT Figure 3. Effect of Etomoxiryl-CoA and Malonyl-CoA 44 on CHAPS Solubilized Microsomal COT Figure 4. Palmitoyl-CoA Inhibition of Microsomal COT 47 Figure 5. Immunoprecipitaion of Microsomal COT with 50 Antiperoxisomal-COT and Antimitochondrial-CPT Figure 6. Effect of Reduced. Glutathione on the Activity 55 and Malonyl-CoA Sensitivity of Microsomal COT after Freeze-Thawing Figure 7. HPLC Gel-Filtration of [3H]-Etomoxir Labeled 58 Microsomal Proteins Figure 8. SDS-PAGE of Microsomal Proteins After Incubation 60 with [3H]-Etomoxir xi Page Chapter 3. Kinetic Characterization of Membrane Bound Microsomal COT Figure 1. Velocity versus Decanoyl-CoA Concentration 72 Curve and Double-Reciprocal Plot for Microsomal COT Figure 2. Velocity versus L-Camitine Concentration 74 Curve and Double-Reciprocal Plot for Microsomal COT Figure 3. The pH Optimum of Microsomal COT 77 Figure 4. DL-Aminocamitine Inhibiton of Microsomal COT 79 Figure 5. Effect of DL—Aminocamitine of the Kinetic 81 Parameters of Microsomal COT Figure 6. Decanoyl-DL—Aminocamitine and Palmitoyl- 83 DL-Aminocamitine Inhibition of Microsomal COT Chapter 4. Attempted Purification of Microsomal COT Figure 1. Ammonium Sulfate Fractination of Detergent 98 Solubilized Microsomal COT ‘ Figure 2. HPLC Weak Anion Exchange Separation of 101 High PH - CHAPS Solubilized Microsomal COT xii CAT CHAPS CoA CoASH COT CPT CPT. ' l CPT O DTBP DTNB D'I'T EDTA Etomoxiryl—CDA HPLC Iso List of Abbreviations camitine acetyltransferase (acetyl-CoAz-L-carnitine O- acetyltransferase, EC 2.3.1.7) 3-[(3 -cholamidopropyl)dimethylammonio] l-propane sulfonate coenzyme A reduced coenzyme A camitine octanoyltransferase (octanoyl—CoAzL-camitine O- octanoyltransferase) camitine palmitoyltransferase (palmitoyl-CoAzIrcarnitine O- palmitoyltransferase, EC 2.3.1.21) form of camitine palmitoyltransferase in contact with the matrix of mitochondria malonyl-CoA sensitive form of camitine palmitoyltransferase in contact with the cytosol 4,4’-dithio-bispyridine 5,5’-dithio-bis(2—nitrobenzoic acid) dithiothreitol (ethylenediamine)—tetra-acetic acid (B877-38), (R)-2-[6-(4-chlorophenoxy)hexyl]-oxirane-2- carboxyl-coenzyme A ester high performance liquid chromatography the inhibitor concentration required for 50% inhibition under given assay conditions xiii IMV CPT Ki K05 Km munit OMV CPT PMSF SDS-PAGE TDGA-CoA Tris Tween-20 unit camitine palmitoyltransferase activity of inner mitochondrial membrane enriched vesicles inhibition constant Hill constant, K05 is equal to Km with Hill coefficient (n) equal to l Michaelis-Menten constant 1 nmole min‘ camitine palmitoyltransferase activity of outer mitochondrial membrane enriched vesicles phenylrnethylsulfonyl fluoride sodium dodecyl sulfate polyacrylamide gel electrophoresis 2-tetradecy1glycidyl-coenzyme A ester tris-(hydroxymethyl)aminomethane polyoxyethylene sorbitan monolaurate 1 mole min' maximum velocity xiv Introduction Camitine acyltransferases catalyze the reversible transfer of acyl-groups between L- carnitine and coenzyme A. Camitine acyltransferase activity with medium-chain and long- chain acyl-groups as substrate has been reported to be located in mitochondria, peroxisomes, and endoplasmic reticulum in rat liver, although the existence of medium/long-chain camitine acyltransferase aetivity in the endoplasmic reticulum has been disputed. This thesis reports the characterization of a camitine acyltransferase in smooth and rough endoplasmic reticulum of rat liver with medium-chain to long-chain acyl-group specificity that is completely inhibitable by malonyl-CoA. This medium/long-chain camitine acyltransferase is named microsomal camitine octanoyltransferase (COT). The characterization provides evidence that this enzyme is distinct from mitochondrial and peroxisomal enzymes. Chapter 1 is a review of literature about medium/long-chain camitine acyltransferase activity of rat liver microsomes. Microsomal medium/long—chain camitine acyltransferase activity is compared with the two other camitine acyltransferases in rat liver that use medium- chain and long-chain acyl-groups as substrate which are mitochondrial camitine palmitoyltransferase (CPT) and peroxisomal camitine octanoyltransferase (COT). The inhibition of microsomal COT, mitochondrial CPT, and peroxisomal COT by malonyl-CDA, palmitoyl-CoA, etomoxiryl-CoA, and aminocamitine is compared. 2 Chapter 2 presents data which show that microsomal COT is strongly inhibited by malonyl-CoA. Low contamination of the microsome fraction by mitochondria and peroxisomes is established using organelle marker enzymes. The inhibition of microsomal COT by palmitoyl-CoA and etomoxiryl-CoA is shown. The data show that malonyl-CDA sensitive microsomal COT is antigenically different than either mitochondrial CPT or peroxisomal COT. Chapter 3 presents data showing the kinetic characterization of membrane-bound microsomal COT. Kinetic constants K05 and Vm for decanoyl-CoA as the varied substrate and for L-camitine as the varied substrate are determined. The 150 for inhibition of microsomal COT by DL-aminocarnitine, decanoyl-DL-aminocamitine and palmitoyl-Dbaminocamitine and the effect of DL—aminocarnitine on the K05 and Vm with L—camitine as varied substrate are shown. The effect of pH on membrane bound microsomal COT activity is also presented. Chapter 4 is a summary of the attempted purification of microsomal COT. Purification of detergent solubilized COT was attempted by column chromatography. Gel filtration, anion exchange, dye affinity, and hydrophobic interaction chromatography were not successful in increasing the specific activity of the detergent solubilzed microsomal COT. Results of column chromatography are discussed. A prospectus for future purification is presented. Chapter 5 is a summary and conclusion of the microsomal COT data presented in Chapters 2, 3 and 4. Microsomal COT is compared to literature reports of mitochondrial CPT and peroxisomal COT to establish that microsomal COT is a distinct enzyme from 3 mitochondrial CPT or peroxisomal COT. The data show that there is more than one malonyl- CoA sensitive medium/long—chain camitine acyltransferase in rat liver. A prospectus for future research is presented. Materials and Methods Isolation of Microsomes Male, fed, Sprague-Dawely rats weighing 150-200 g were stunned lightly in C02 and decapitated. Livers were immediately collected, immersed in and coarsely minced in 0.25 M sucrose containing 25 rig/ml PMSF, 0.5 ug/ml Pepstatin A, and 0.05 rig/ml Leupeptin (homogenization solution). Livers were finely minced and rinsed with solution and homogenized at 4°C with 4 volumes of buffer using 4 passes of a loose-fitting Potter- Elvehjem homogenizer. Rough and smooth microsomes were prepared using differential centrifugation. The liver homogenate was centrifuged at 400 x g for 5 min followed by centrifugation of the 400 x g supernatant fluids at 6,000 x g for 10 min Cenuifugation of the 6,000 x g supernatant fluids at 10,000 x g for 10 min was followed by ultracentrifugation of the 10,000 x g supernatant fluids at 100,000 x g for 1 hour. The 6,000 x g pellet was resuspended in 0.25 M sucrose and the 100,000 x g pellet was resuspended in 10 mM potassium phosphate pH 7.5, 1 mM EDTA containing 20% glycerol. The microsomal membranes were stored frozen in aliquots at -20°C. . Rough and smooth microsomes were also prepared using CaCl2 precipitation (l). The liver homogenate was centrifuged at 12,000 x g for 15 min. and the supernatant fluid was filtered through a loose plug of glass wool and centrifuged again at 12,000 x g for 15 min. The supernatant fluid was diluted with 4 volumes of 12 mM mannitol, 1 mM EDTA, and 8 5 mM CaCl2 and kept on ice for 1 hour. The diluted supematant fluid was centrifuged at 1,500 x g for 15 min. and the microsomal pellet resuspended in 0.25 M mannitol, 25 mM MOPS pH 7.4, 1 mM EDTA and stored frozen in aliquots at -20°C. Rough microsomes were prepared using CsCl-sucrose density gradient centrifugation (2). The liver homogenate was centrifuged at 10,000 x g for 30 min. and the supernatant fluid made 15 mM in CsCl with the addition of l M CsCl. Ten ml of the 15 mM CsCl supernatant fluid was layered over 15 ml of 1.3 M sucrose, 15 mM CsCl in a 26.3 ml Beckman polycarbonate ultracentrifuge tube and centrifuged at 100,000 x g for 2 hours in a Beckman Ti70 rotor. The pellet containing rough microsomes was resuspended in 10 mM potassium phosphate pH 7.5, 1 mM EDTA and stored frozen in aliquots at -20°C. Detergent solubilization was done by making the microsomes 8 mM in CHAPS, incubating on ice for 1 hour, and storing frozen at -20°C for at least 12 hours. They were thawed at room temperature and centrifuged at 100,000 x g for 30 min. at 4°C in a Beckman Airfuge. Trichloroacetic acid extraction of microsomes was done by making the microsomes 30% (w/v) in trichloroacetic acid and then the sample was centrifuged 5 min. in an Eppendorf centrifuge. The pH of the supernatant fluids was adjusted to ~7 with the addition of potassium hydroxide. Extraction of microsomal lipids was done as described (3). Microsomes were extracted with CHAPS as described above and the 100,000 x g pellet was resuspended in 100 mM potassium phosphate pH 7.5. The pellet was weighed and homogenized with 17 volumes (ml/g) of chloroform/methanol (2/1, v/v). The sample was centrifuged at 1,000 x g for 15 min. and the supernatant fluids were filtered through a sintered glass funnel. The pellet was extracted again and the supernatant fluids combined. Enzyme and Protein Assays Microsomal COT activity was assayed spectrally (at room temperature) at 324 nm in 50 mM potassium phosphate, pH 7.5, 50 mM potassium chloride, 150 uM dithiopyridine, 17 M decanoyl-CoA, 1.7 mM L-carnitine (E324 = 19,600 M'1 em"). Values were corrected for camitine independent CoASH release (4). Malonyl-CoA inhibition was determined with 17 11M malonyl- CoA. Long-chain camitine acyltransferase activity was assayed with 17 “M palmitoyl-CoA. Microsomal COT was assayed spectrally unless otherwise indicated. For some studies as indicated, microsomal COT was also assayed (at room temperature) using a radiochemical, isotope forward assay (5) in 100 in with 17 W [1-“C1-decanoy1-CoA, 150 M dithiopyridine, 1.7 mM L-camitine in 50 mM potassium phosphate pH 7.5 and 50 mM KCl. The reaction was started with the addition of L-camitine and stopped by the addition of 400 in ice-cold methanol. The [l-“Cl-decanoylcamitine was separated from the unreacted [1-14C]-decanoyl-C0A (5). For Figures 1 and 4 in Chapter 2, the amount of [1-14C]- decanoylcarnitine was determined using a combined HPLC-Flo Scint-B—counter described previously (6). The buffer system was 5 mM butanesulfonic acid, 5 mM ammonium acetate, pH 3.4 (A) and methanol (B). The flow rate was 1.0 ml min'1 and the gradient was as follows: at zero time, 20% A, 80% B; at 10 min., 0% A, 100% B; at 30 min., 20% A, 80% B. The retention time of [1-“C]-decanoylcamitine was 8.3 min. The following marker enzymes were assayed: g1ucose-6-phosphatase, monoamine oxidase, urate oxidase, cytochrome c oxidase, N ADPH-cytochrome c reductase. Samples assayed for monoamine oxidase, urate oxidase, and cytochrome c oxidase were stored overnight at -70°C, thawed, and solubilized in 1% Tween 20. Glucose-6—phosphatase (EC 3.1.3.9) was assayed in a 0.5 ml volume at room temperature containing 14 mM glucose-6- 7 phosphate, 25 mM histidine, 1 mM EDTA pH 7.0 and the amount of inorganic phosphate measured at A660 using the Fiske-SubbaRow reagent with a potassium phosphate standard curve (10-100 nmole inorganic phosphate) (7). Urate oxidase (EC 1.7.3.3) was assayed spectrally at room temperature in 20 mM borate pH 9.5 with 40 11M uric acid and the decrease in A293 was measured (E293 = 12,600 M" cm") (8). Cytochrome c oxidase (EC 1.9.3.1) was assayed spectrally at room temperature in 200 mM potassium phosphate pH 6.0, 1 mM EDTA with 16 11M reduced cytochrome c and the decrease in A550 was measured (E550 = 19,600 M-1 cm") (9). NADPH Cytochrome c reductase (EC 1.6.2.4) was assayed spectrally at room temperature in 50 mM potassium phosphate pH 7.7, 0.1 mM EDTA with 36 11M cytochrome c (Sigma Type 111), 91 11M NADPH and the increase in A550 was measured (E550 = 21,000 M"1 cm'l) (10). Monoamine oxidase activity was determined using an end point assay by measuring the production of 4-hydroxy-quinoline. Each assay was incubated 90 min. at 37°C in a 1 ml volume containing 70 mM potassium phosphate pH 7.5, and 0.31 mM kyntn'amine dihydrobromide (11). The A330 was measured and the activity calculated as described in (12). Protein was determined by the modified Lowry method (13). Immunoprecipitation Antiperoxisomal-COT antibody (14), purified mouse liver peroxisomal COT (15), and anti-beef heart mitochondrial CPT IgG (16) prepared previously were used. The IgG fraction was purified from the anti-peroxisomal COT rabbit serum using a GammaBindTM G-PriePackTM cartridge from GENEX with the protocol supplied by the company. After elution from the GammaBindTM-G column, the IgG was stored frozen at -20° in 10 mM sodium phosphate, pH 7.0, 150 mM sodium chloride, at a protein concentration of 4.1 mg/ml. 8 The effect of antiperoxisomal-COT and antimitochondrial-CPT antibodies on microsomal COT was determined by immunoprecipitation. Samples were incubated at 4°C in 25 mM potassium phosphate, pH 7.5, 1 mM EDTA, 0.15 mM sodium chloride. Antiperoxisomal-COT IgG and the antimitochondrial-CPT IgG were added as indicated (see Chapter 2, Figure 5); varying amounts of bovine serum albumin were added to the antiperoxisomal-COT IgG samples to keep the amount of added protein constant. The samples were centrifuged at 10,000 x g for 10 nrin. and the supernatant fluids assayed immediately for COT. CPT in the 6,000 x g fractions was assayed with 50 1.1M decanoyl-CDA, 25 mM L- carnitine, 150 1.1M dithiopyridine in 50 mM potassium phosphate, 0.1% Triton X-100, pH 7.5. Labeling of L_iver Microsomes with 3H-Etomoxir A 540 pg sample of rat liver microsomes was incubated for 15 min. at room temperature with 5 mM ATP, 5 mM MgC12, 50 11M CoASH, 50 mM potassium phosphate, 50 mM potassium chloride, pH 7.5, and 5 uM 3H-etomoxir (specific radioactivity 40 Ci/mmole) in a final volume of 100 pl. After 15 min., the reaction was diluted with 1 volume of cold 50 mM potassium phosphate, 50 mM potassium chloride, pH 7.5, and centrifuged at 100,000 x g for 30 min. in a Beckman airfuge. The microsomal membrane pellet was rinsed three times with 200 ill of 50 mM potassium phosphate, 50 mM potassium chloride, pH 7.5, and the pellet was resuspended in 100 pl of electrophoresis sample buffer. For the pulse chase experiments, after the 15 minute incubation, etomoxiryl-CoA (unlabeled) was added to a final concentration of 50 11M and the reaction mixture was incubated for an additional 5 min. at room temperature. Microsome samples (50 111, 5.4 mg/ml)were extracted with 5.0 ml of cold hexanefisopropanol (3/2, v/v) by vortexing for 1.0 minute, followed by centrifugation for 15 9 min. at 3,000 x g. The pellet was extracted with 200 ill of water plus 5.0 ml of cold chloroform/methanol (3/2, v/v) and centrifuged, and the protein pellet was then dissolved in 300 pl 7% SDS, and aliquots were subjected to SDS-PAGE according to Laemmli (17). After electrophoresis, the gel was sliced into 2 mm slices, and the gel slices were treated with 0.5 ml of distilled water by shaking overnight and the radioactivity determined. [3Hl-Etomoxir labeled microsomal and mitochondrial proteins were separated isocratically on a Dupont G-250 gel filtration column ( 9.4 mm ID x 250 mm) in 100 mM Tris-Cl, pH 6.8, containing 10% glycerol and 0.5% SDS at a flow rate of 1 ml min‘l. The A280 was recorded using a Water’s Model 441 absorbance detector and Model 740 data module and 300 ill fractions were collected beginning at 5.6 rrrin. A 50 pl aliquot of each fraction was mixed with 10 ml of scintillation fluid (Safety Solve, RPI) and counted in a Packard Tricarb 1900 CA. Peaks containing radioactivity were pooled and a 50 pl aliquot extracted with 1.0 ml of chloroform/methanol (3/2, v/v). The radioactivity remaining after extraction and the radioactivity in the chloroform/methanol phase was determined with scintillation counting. Determination of Ki_netic ConsLtants Membrane bound rrricrosomal COT was assayed using a radiochemical, isotope forward assay with [1-‘4C1decanoy1.CoA (5). [1-14C1decanoy1-CoA was synthesized as described (18). Synthesis conditions were 2 mM [1-14C]decanoic acid, 20 mM ATP, 10 mM CoASH, 20 mM MgC12, 2 mM DTT, 100 mM Mops-NaOH pH 7.5, 0.1% Triton x-100 (w/v) and 2.5 units Acyl-CoA Synthetase in a 5 ml volume and the mixture was stirred at 35°C for 2 hours. The mixture was applied to a Prep-Sep C-18 extraction column ( Fisher Scientific, Fair Lawn, NJ) equilibrated in 100 mM MOPS-NaOH pH 7.5 and the column was washed with 1 ml of 50% 10 methanol (v/v) and the [1-14C1decanoyl-CoA was eluted with 20 ml of methanol. The solvent was evaporated to dryness using a rotary evaporator and the residue dissolved in 50 mM potassium phosphate pH 5.3. [1-‘4C1Decanoyl—CoA was purified using HPLC with a Water’s u—Bondapak C18 reverse phase column with an isocratic buffer of 50 mM potassium phosphate pH 5.3 and 32% acetonitrile at a flow rate of 1 ml min‘l. [1-14C]Decanoyl-C0A was detected by absorbance at 254 nm and the fractions containing radioactivity with the retention time of authentic decanoyl-CoA were collected and dried under vacuum using a rotary evaporator. The specific activity of the [1-“C]decanoyl-C0A was 19870 dpm/nmole. The radiochemical isotope forward assay of membrane bound microsomal COT was done in a 100 pl volume, at 30°C for 30 sec in 50 mM potassium phDSphate pH 7.5, 50 mM potassium chloride, and 150 M DTBP, and 2030 pg of nricrosomes. Substrate concentrations were 6 mM L-carnitine, 16 11M [1-14C]decanoyl-C0A or were varied as indicated. The assay was linear with 30 ug of microsomal protein up to 60 seconds; less than 15% of the decanoyl-CoA was consumed in 30 seconds. The reaction was stopped with the addition of 400 til ice cold methanol. [1-14C]decanoylcamitine was separated from unreacted [1-‘4C1decanoyl-CoA using a 0.5 x 3.0 cm DE—52 column equilibrated in H20 and the [1— 1“Cldccanoylcarrritine eluted with 1.0 ml of 80% methanol (v/v). The eluant was mixed with 10 ml of Safety-Solve cocktail and the radioactivity determined with a Packard 1900 CA liquid scintillation analyzer. A control, minus-camitine, assay was done at each [l-“Cldecanoyl-CDA concentration to correct for background. ll Ammonium Sulfate Fractionation of Detergent Solubilized Microsomal COT Membrane bound microsomal COT was made 1% in Tween-20, incubated on ice for 1 hour, and centrifuged at 100,000 x g for 60 min. at 4°C. Aliquots of the supernatant fluids containing 100 munits of COT were made 0-60% saturation in ammonium sulfate with the addition of saturated ammonium sulfate, pH 7.0. The samples were incubated on ice for 1 hour then centrifuged at 15,000 x g for 15 min. in an Eppendorf centrifuge. The samples were pipeted into a pasteur pipet plugged with glass wool and the supernatant collected. The pellet retained by the glass wool was eluted with 1 ml of 5 mM potassium phosphate, 1 mM EDTA, pH 7.5 containing 0.5% Tween-20. Column Chromatogr_apl_l_y HPLC gel filtration of solubilized COT was done at room temperature using a Dupont G-250 gel filtration column (9.4 mm ID x 250 mm) equilibrated in 200 mM ammonium acetate pH 8.0 containing 20% glycerol. The A280 was recorded using a Water’s Model 441 absorbance detector and Model 740 data module. The column was run isocratically at a flow rate of 1 ml min'1 and 1 ml fractions were collected and assayed for COT activity. Gel filtration of solubilized nricrosomal COT was also done using a 2.5 cm x 80 cm Biogel P100 column with a column volume of 400 ml. The column was run at 4°C at a flow rate of 1.1 ml min:1 and 4 ml fractions were collected and assayed for COT activity and the A280 determined. HPLC anion exchange chromatography was done using a Synchropak AX300 (4.1 mm x 250 mm) column with an average pore size of 300 A equilibrated in 10 mM potassium phosphate, pH 7.5, 2 mM CHAPS containing 20% glycerol (buffer A). The Am was recorded 12 using a Water’s Model 441 absorbance detector and model 740 data module. The column was washed at a flow rate of 0.5 ml min:1 with buffer A until the A280 was ~0. Proteins were eluted with a linear gradient to buffer A containing 1 M potassium chloride and 1 m1 fractions were collected and assayed for COT activity. Hydroxylapatite chromatography was done using a Bio-gel HTP column with a 6 m1 column volume equilibrated in 10 mM potassium phosphate, pH 7.5, 1 mM EDTA, 2 mM CHAPS, and 20% glycerol (buffer A). The column was washed until the A280 was ~0 and proteins were eluted with a linear gradient to buffer A containing 500 mM potassium phosphate, pH 7 .5. Fractions were collected and assayed for COT activity and the A280 determined. Materials Most chemicals, including acyl—CoA, CoASH, and [1-i4C]-decanoic acid sodium salt with a specific activity of 10.6 mCi/rnmole were purchased from Sigma Chemical Company (St. Louis MO). L-Camitine was a gift from Sigma-Tau (Rome, Italy). DL-Aminocamitine, decanoyl-DL-aminocamitine, and palmitoyl-DL-aminocamitine were a gift from Dr. Owen Griffith (Cornell University Medical School, New York, NY). Etomoxiryl-CoA (B877-38) and [3H1—etomoxir with a specific activity of 40 Ci/mmole were a gift from Byk Gulden (D-7750 Konstanz, Fed. Rep. of Germany). Pigeon breast muscle camitine acetyltranferase, Acyl-CoA synthetase, aus mikroorganismen, and CHAPS were from Boehringer Mannheim Biochemicals (Indianapolis, IN). DE 52, diethyl aminoethyl cellulose, was from Whatman (Hillsboro, OR). Safety-solve was from Research Products International Corp. (Mount Prospect, IL). Bio-Gel P100 and Bio-Gel HTP were from Bio-Rad (Richmond, CA). Blue Sepharose C1-6B was from 13 Pharrnacia (Piscataway, NJ). All other reagents were of analytical grade. Chapter 1 Literature Review Microsomal Medium\Long-Chain Camitine Acyltransferase Activity in Rat Liver Camitine acyltransferases catalyze the reversible transfer of acyl groups between L- carnitine and coenzyme A (19). They are classified according to their acyl-chain length specificity into short-chain and medium/long-chain camitine I acyltransferases. Short-chain camitine acyltransferases are commonly called camitine acetyltransferase (acetyl-CoAzL- camitine O-acetyltransferase, EC 2.3.1.7, CAT) while medium/long-chain transferases are further divided into camitine octanoyltransferase (octanoyl-CoA:L—carnitine O- octanoyltransferase, COT) and camitine palmitoyltransferase (palmitoyl-CoAzL-camitine O- palmitoyltransferase, EC 2.3.1.21, CPT). COT and CPT both catalyze the transfer of medium- chain and long-chain acyl groups. COT has a higher activity towards medium-chain and CPT (with physiological concentrations of camitine) towards long-chain acyl groups. CAT, COT, and CPT activities are located in mitochondria, peroxisomes and endoplasmic reticulum. The mitochondrial medium/long-chain camitine acyltransferase is named camitine palmitoyltransferase (CPT). CPT is a membrane bound, oligomeric enzyme that facilitates the transfer of long-chain acyl residues through the inner mitochondrial membrane for subsequent B-oxidation (5,19,20). CPTo is located outside the inner membrane of mitochondria and promotes the formation of acylcamitines using cytosolic acyl-CoAs. The camitine/acylcarnitine translocase acts to transfer acylcamitines into the matrix of mitochondria (21). CPTi is located 14 15 inside the inner membrane of mitochondria and promotes the formation of acyl-CoAs using acylcamitines. It has not been established if CP’I‘o and CPTi are the same enzyme with different locations and regulatory properties or if CPTo and CP’I‘i are two distinct enzymes. Results from some investigators have supported the view that CPTO and CPTi are distinct enzymes with CPT0 being more detergent labile than CPTi (22-24). Other investigators have supported the view that CP’I‘o and CPT} are the same protein (16). The topographical distribution of CP’I‘o and CPTi within the mitochondria has also not been established. The results of some investigators have shown that CPT} is located on the inner side of the inner mitochondrial membrane and CPTo on the outer side of the inner mitochondrial membrane (25- 28). The results of other investigators have suggested that while CPTi is on the inner side of the inner mitochondrial membrane, CP'I‘o is on the inner side of the outer mitochondrial membrane (11,29-32). ’ The peroxisomal medium/long-chain camitine acyltransferase is named camitine octanoyltransferase (COT) (14,33-35) and could function to shuttle peroxisomal B-oxidation- shortened acyl chains out of the peroxisome (36). Peroxisomal COT is easily solubilized by freeze\thaw treatment of peroxisomes (15,34,37). Peroxisomal COT is a soluble enzyme located in the matrix of peroxisomes (38,39). Recently, it has been reported that the malonyl- CoA sensitivity and ratio of decanoyltransferase activity to palmitoyltransferase activity of peroxisomal COT is altered when COT is solubilized from the peroxisome (40). The endoplasmic reticulum (ER) also contains a medium/long—chain camitine acyltransferase (37,39,41). It has a higher activity with decanoyl-CDA compared to palmitoyl- CoA; herein this enzyme is referred to as microsomal COT. Characterization of microsomal COT has been hindered by the instability of the enzyme and the difficulty in solubilizing it . .Ew‘iiigwaw 16 from the membrane (37,39). Microsomal COT is tightly associated with the outer side of the endoplasmic reticulum membrane (42). The function of microsomal COT is not known. Microsomes contain camitine acyltransferase activity both with octanoyl-CoA as substrate (COT) and with acetyl-CoA as substrate (CAT) with approximately equal initial velocities (37). The initial velocity with both octanoyl-CoA and acetyl-CoA present is the sum of that obtained with each substrate alone (37). Microsomal CAT has different solubility characteristics, regulatory characteristics and substrate affinities than microsomal COT, and microsomal CAT has been purified free of medium\long-chain transferase activity. On the basis of these observations it is concluded that microsomal CAT is a different enzyme than microsomal COT (37,39). Microsomes are sealed vesicles derived from the disruption of the ER which maintain the same cytoplasmic side out sidedness as does the ER; they vary in size, density, and surface charge (43). The ER of rat liver contains 19% of total cellular protein, 48% of total cellular phospholipid, and 58% of total cellular RNA (43). The microsomal membrane is composed of 60-70% protein and 30-40% phospholipid (43). The predominant microsomal membrane phospholipid is phosphatidylcholine (55%) followed by phosphatidylethanolarninc (20-25 %), phosphatidylserine (5-10%), phosphatidylinositol (540%), and sphingomyelin (4-7 %) (43). The microsomal membrane contains at least 38 polypeptides (44). The microsomal membrane is permeable to uncharged molecules with MW < 600 daltons; it is impermeable to charged molecules > 90 daltons and to macromolecules (43). l7 Assay Methods for Camitine Acyltransferases Reports of microsomal camitine acyltransferase activity in the literature have been complicated by the use of different assay methods which can give varying measurements of camitine acyltransferase activity. The reaction is represented below. Acyl-CoA + L-carnitine H Acylcarnitine + CoASH In the forward direction the enzymes can be assayed spectrally by directly measuring the disappearance of acyl-CoA by the decrease in absorbance at 232 nm (45,46) or indirectly by measuring the appearance of free CoASH using a thiol-trapping agent like DTNB, (5,5 ’-dithio- bis(2-nitrobenzoic acid), (4) or DTBP, (4,4’-dithio-bispyridine), (47). In the reverse direcdon the production of acyl-CoA can be measured using hydroxylamine and quantitating the acylhydroxamate formed (48). Camitine acyltransferases can be assayed radiochemically in the forward or reverse direction using a radioactively labelled camitine or acyl group and separating labelled product and substrate (49,50). The forward radiochemical assay can be done using fatty acid as substrate and coupling it with fatty acid synthetase, ATP-Mg2+,and CoASH to generate acyl-CoA (51). Camitine acyltransferases can also be assayed using an isotope exchange method (46); in the forward direcan the incorporation of radiolabeled camitine into radiolabeled acylcamitine is measured. These assay methods are summarized in Table 1. 18 Table 1. Summary of Assay Methods for Camitine Acyltransferases. Acyl-CoA + L-carnitine H Acylcarnitine + CoASH Spectrophotometric Continuous Rate Assays Forward direction only: Measure CoASH with sulflrydryl trapping reagent such as DTNB (3412:13300 Mlem") (4) or DTBP (E324=19,600 Wm") (47). Need to correct for camitine independent CoASH release. DTNB and DTBP may inhibit reactive thiol groups in the enzyme essential for catalysis. Forward or reverse direction: Measure the disappearance (forward) or appearance (reverse) of acyl-CoA directly by the A232 of the thioester bond (45 ,46). Need to correct for camitine independent CoASH release. Usually used for nearly homogeneous enzymes because of high background A232 and low extinction coefficient (E232: 4500 M‘lcm'l). Reversibility of reaction hinders initial rate measurements. End Point Assays Forward or Reverse direction: Use radioactively labeled camitine or acyl-group. Need to separate labeled product from labeled substrate (49,50). Need to ensure initial rate measurement. Forward direction only: Can be used with free fatty acid as substrate coupled to fatty acid synthetase to generate acyl-CoA (51). Reverse direction only: Use hydroxylamine to measure the production of acyl-hydroxymate from acyl-CoA (48). Isotopic Exchange Assays Both forward and reverse direction: Measure the rate of incorporation of radioactivity from camitine fraction into the acylcamitine faction or vice versa (46). Cannot be used for initial rate measurements since run at or near equilibrium. Rate underestimated in the presence of acyl-CoA hydrolase. l9 Mbufion of Mediumllong-chain Camitine AcfitLansferaLse Activity in Rat Liver Microsomal long-chain camitine acyltransferase (CPT) activity was first reported in 1962 by Bremer, 9;. al. who showed the production of [14C1-palmitoylcamitine from [14C]- carnitine, palmitate, ATP-Mg+, and CoASH in rat liver microsomes and mitochondria using an endpoint isotope forward assay (52). The microsomal CPT activity was estimated to be 70% of the mitochondrial CPT activity. In 1967 though, Nonrm and Bremer reported that CPT activity was exclusively localized in liver mitochondria (53). The mitochondrial CPT activity accounted for 65% and the microsomal activity for only 8% of the total rat liver homogenate CPT activity measured using an exchange assay in which the incorporation of radioactivity into [14C1-palmitoylcamitine was measured from unlabeled palrnitoylcarnitine, CoASH, and [“Cl- camitine. They proposed that the microsomal fracan isolated in 1962 had been contaminated by mitochondria leading to the putative microsomal CPT activity. An alternative explanation is that the different assay conditions used in 1962 and 1967 gave different estimates of CPT activity. The exchange assay used in 1967 (53) can underestimate CPT activity in the presence of palmitoyl-CoA hydrolase (4) and microsomes contain higher levels of palmitoyl-CDA hydrolase than do mitochondria (54,55). Van T01 and Hulsmann in 1969 also reported a dual localization of CPT in both mitochondria and microsomes and that measurement of microsomal CPT activity is dependent on the assay method used (51). They assayed CPT using both an exchange assay and an isotope forward assay measuring the production of [3H ]-pa1mitoylcarnitine from [3H]-camitine, palmitate, MgClz, ATP, and CoASH. They found that ~50% of the total homogenate CPT activity was in the mitochondria when either assay was used. The percentage of total homogenate CPT activity in the microsomal fraction was dependent on how the CPT was 20 assayed; the exchange assay gave 12% and the isotope forward assay gave 28% as the percentage of total homogenate CPT in the microsomal fraction. Hoppel and Tomec in 1972 reported the intracellular distribution of CPT in rat liver using different assay methods (27). Using an isotope forward end point assay, 12% of the total homogenate CPT activity was located in the microsome fraction. Using an isotope reverse end point assay, it was only 7% and using a hydroxamate assay, it was 5%. The mitochondrial CPT activity, in contrast, was 50% of the total homogenate CPT activity using the isotope forward assay, 66% using the isotope reverse assay, and 71% using the hydroxamate assay. In 1973, the subcellular distribution of long-chain camitine acyltransferase (CPT) activity in rat liver was studied by Markwell, e_t. a_l. (41). Microsomal CPT activity assayed spectrally using DTNB with 37.5 11M palmitoyl-CoA was ~10% of the mitochondrial CPT activity. Microsomal CPT activity can be underestimated by this spectral assay because assay conditions included high concentrations of palmitoyl—CoA and the presence of 0.1% Triton X- 100 both of which inhibit the microsomal enzyme. Also, it was reported that a high background rate for palmitoyl-CoA hydrolase (carrritine independent CoASH release) was present in the microsomal fraction which could have masked low CPT rates. In rat liver, microsomal palmitoyl-CoA hydrolase has a higher specific activity than does the mitochondrial palmitoyl-CoA hydrolase (54,55). The distribution of medium—chain camitine acyltransferase (COT) activity in rat liver microsomes was also reported in 1973 by Markwell 9; g (41). The intracellular distribution of carnitine acyltransferase activity using octanoyl-CoA as substrate (COT activity) showed COT was 20% peroxisomal, 52% mitochondrial, and 28% microsomal. COT activity was assayed with DTNB and 100 11M octanoyl-CoA. The intracellular distribution was determined 21 using fractions isolated by isopynic sucrose density gradient centrifugation. Further studies by Markwell, e_t. 5131., showed COT was present in a microsome fraction prepared using differential and zonal centrifugation that was free of marker enzymes for golgi and plasma membranes (37). COT activity was found in both rough and smooth microsomes of rat liver, and the specific activity 0f COT was two fold higher in smooth microsomes than in rough microsomes (37). Valkner and Bieber in 1972 reported that microsomal COT activity is located exclusively with the cytoplasmic face of the endoplasmic reticulum (42). In 1974, Van Tol reported that CPT was located in both mitochondria and microsomes of rat liver, and that different assay conditions give varying measurements of microsomal CPT activity (56). The presence of enzymes and substrates for fatty acid activation and an ATP regenerating system (complete assay) were necessary for maximum CPT activity. A simplified assay procedure using palmitoyl-CoA as substrate in an isotope forward assay gave (50% of the microsomal CPT activity as did the complete assay procedure. When the complete assay was used the microsomal CPT activity was estimated to be ~35% of the mitochondrial CPT activity. Kahonen in 1976 showed the intracellular distribution of CPT in rat liver fractions isolated by isopynic sucrose density gradient centrifugation (57). CPT was assayed spectrally. Microsomes contained 11% and the mitochondria contained 81% of the CPT activity of the total homogenate. The microsome fraction, though, contained high levels of acyl-CoA hydrolase activity. Kahonen also reported COT activity in microsomes (57). The subcellular distribution of COT was done using isopynic sucrose density gradient fractionation. COT was assayed spectrally using DTNB in an assay that contained 0.1% Triton X-100 and 100 LLM octanoyl-CoA. The distribution of COT was 9% peroxisomal, 77% mitochondrial, 22 6% microsomal, and 9% soluble. Literature reports of medium-chain camitine acyltrans ferase (COT) and long-chain camitine acyltransferase (CPT) activity in the endoplasmic reticulum have been contradictory (27,37,39,41,42,51-53,56,57). The use of different assay methods has contributed to the problem. For instance, the use of palmitoyl-CoA as substrate to assay medium/long—chain camitine acyltransferase activity underestimates the contribution of the microsomal enzyme. Contamination of organelle fractions of liver homogenates is well documented (58,59) and could also contribute to contradictory reports of the distribution of COT and CPT activity in rat liver. For example, electron microscopic studies have shown association between rough ER and mitochondria and putative rolrgh-ERVnitochondria complexes have been isolated (60). It has also been shown that in rat liver there is a subfraction of ER associated with mitochondria that sediments at 10,000 x g (61). It was proposed that this mitochondrial-ER fraction is involved in lipid transfer (61). Indeed, reports that CPTo is located in outer mitochondrial membrane vesicles could result from contamination of outer membrane preparations with ER. Outer membrane preparations can contain as much as 14% ER (58). Tissue Distribution of Microsomal Medium\long-c_h§in Camitine Acyltransferase Activity The subcellular distribution studies of COT discussed above used rat liver. These studies showed that COT/CPT activity in rat liver is located in mitochondria, microsomes, and peroxisomes (37,39,4l,51). In 1978, Fogle and Bieber reported the presence of COT and CPT in rat heart nricrosomes (62). Microsomes from rat heart were prepared by differential centrifugation and CaCl¢ precipitation. The distribution of COT in heart was 10.5% microsomal and 32.6% mitochondrial of the total COT activity of the homogenate. The 23 distribution of CPT in heart was 13.2% microsomal and 29.6% mitochondrial of the total CPT activity of the' homogenate. Marker enzyme data showed the microsomes were <10% contaminated by mitochondria. The ratio of C10/C16 activity of rat heart microsomes is 1.75. Effect of Dru_g Trealnentjpd Feeding\£asting on Medium\Long-chain Camitine Acyltra_nsfera_se Activity of Liver Microsomes Markwell gt g reported in 1977 that treatment of rats with clofibrate (ethyl-p- chlorophenoxy isobutyrate), a hypolipidemic drug, increases the specific activity of COT in microsomes from male, fed rats isolated by isopynic sucrose gradient centrifugation from 8.0 to 20.9 nmole min'1 mg’1 protein (63). Treatment with phenobarbital, a proliferator of smooth ER, did not change the specific activity of microsomal COT (63). Kahonen in 1976 reported the effect of clofibrate treatment on the specific activity and percent distribution of microsomal COT and CPT in a membrane fraction in rats (57). The specific activity of microsomal COT in clofibrate treated rats was increased 4.5 fold compared to normal while the percentage microsomal COT in the membrane fraction compared to total homogenate COT was 5.8% in normal rats and 5.0% in clofibrate treated rats. The specific activity of nricrosomal CPT increased 2.5 fold in clofibrate treated rats compared to normal rats. The percentage microsomal CPT in the membrane fraction compared to total homogenate CPT was 10.6% in normal and 5.1% in clofibrate treated. Camitine acyltransferase activity was assayed spectrally in the forward direction with octanoyl-CoA or palmitoyl-CoA as substrate using DTNB. Van Tol in 1974 reported the effect of fasting on microsomal and mitochondrial CPT activity. Microsomal CPT activity from 48 hour fasted rats increased approximately 2 fold 24 from 6.4 in fed to 14.0 nmole min'1 mg’1 protein in fasted rats (56). In contrast, mitochondrial CPT increased only 1.1 fold from 16.7 in fed to 18.7 nmole min-l mg’1 protein in fasted rats (56). A recent report in abstract form by Ramsay showed camitine acyltransferase activity with both decanoyl-CoA and palmitoyl—CoA as substrate increased in gradient purified microsomes from fasted as compared to fed rats (64). giggly of Meditgmlaong-Chain Camitine Acyltraan_f(:;_ase Activity in Microsomes Van Tol in 1974 (56) reported that the K05 for the microsomal medium\long-chain camitine acyltransferase is 7.1p.M for palmitoyl—CoA and 0.18mM for L-camitine. Assays were done using a radiochemical isotope forward assay with [3H]-camitine. The K05 of the microsomal medium\long-chain camitine acyltransferase for palmitoyl-CoA and for L-carnitine were not significantly altered with fasting (56). Solubility and Stability Chgcteristics of Microsomal Medigm\long-chain and Short-chain Cm Acyltransferase Activity COT and CAT of rat liver microsomes are membrane bound and require detergent for solubilization (39). Freeze\thaw treatment of microsomes solubilizes <10% of COT (39). Microsomal COT can be completely solubilized in 1% Triton X-100, 0.4M KCl but the activity is not stable (39). Conditions such as 0.4M KCL, or 1% Triton X-100 which solubilized and stabilized CAT caused a complete loss of COT activity (37). Microsomal CAT can be selectively solubilized be treatment of microsomes with 0.3M sucrose in 0.1M pyrophosphate pH 7.5 which solubilizes 80-90% of CAT activity while 81-83% of microsomal COT remains with the pellet fraction (39). 25 Lalmitoyl-CDA Inhibition of Medium\long-chain Camitine Acyltransferase Activity in Rat HEEL N 0 studies have reported palmitoyl-CDA inhibition of microsomal COT, but studies have shown that varying estimates of microsomal COT activity with palmitoyl-CoA as substrate are obtained with different assay methods. The assay methods have varied in the concentration of palmitoyl-CoA, and the presence of bovine serum albumin. Indirectly these studies show that high concentrations of palmitoyl-CoA as substrate could underestimate microsomal COT activity. For example when palmitoyl-CoA is generated in the assay using palmitate, CoASH, ATP-MG+, and fatty acyl-CoA syntlretase the COT activity is higher than when the same preparation is assayed using palmitoyl-CoA directly as substrate (56). CPT activity of outer mitochondrial membrane vesicles is inhibited by palmitoyl-CoA (29). The CPT activity versus palmitoyl-CoA concentration curve shows inhibition of CPT activity above 15 “M palmitoyl-CoA (29). The inhibition was not reversed with addition of bovine serum albumin or with lowering the palmitoyl-CoA concentration (29). It was also shown that the CPT activity of inner mitochondrial membrane vesicles was not inhibited by concentrations of palmitoyl-CoA up to lSOuM (29). This agrees with previous studies which showed that CPT activity versus palmitoyl-CoA concentration curves for CPTo of intact rat liver mitochondria are sigmoid (32,65,66). Aminocamitine as Substrate andhlhibitor of Med_i;rm\long—chain Carmitine Acyltrainsferase Activity in Rat Liver DL-Aminocarnitine (3-amino-4—trimethylaminobutyric acid) inhibition of mitochondrial camitine palmitoyltransferase (CPT) was first reported by Jenkins and Griffith in 1985 (67). 26 DL-Aminocamitine and acetyl-DL-aminocarnitine both inhibited rat liver mitochondrial CPT with 511M DL-aminocarnitine inhibiting 64% and 511M acetyl-DL-aminocamitine inhibiting 15% of total CPT activity of Triton X-100 treated mitochondria (67). CPT was assayed spectrally in the forward direction using DTNB (67). In 1986, Jenkins and Griffith reported that decanoyl-Dbaminocamitine and palmitoyl-DL—aminocamitine also inhibited mitochondrial CPT (68). CPT was assayed spectrally in Triton X-100 treated mitochondria and 511M decanoyl-Dbaminocamitine inhibited 75% and palmitoyl-DL-aminocamitine 99% of total CPT activity (68). DL-Aminocamitine was not acylated by palmitoyl-CoA by rat liver mitochondrial CPT (67). The inhibition of mitochondrial CP'I‘o by L-aminocarnitine was reported in 1989 by Kanamaru and Okazaki (69). Acetyl-L-aminocamitine, propionyl-L-aminocamitine, and butryl- L-arninocarnitine were isolated from a culture filtrate of Emericella quadrilineata IFO 5 859 in the process of screening for long-chain fatty acid oxidation inhibitors. L-Aminocarnitine was named emeriamine. The ISO for L-aminocarnitine for CPTo of intact rat liver mitochondria assayed using a radiochemical, isotope forward assay with L-[3H]camitine was 62.5 M (see Figure 6 of ref. 69). The 150 for palnfitoyl-baminocamitine was 2.2pM. Unlike the report by Jenkins and Griffith in 1985 (67), Kanamaru and Okazaki (69) report that palmitoyl-L— aminocamitine is formed by liver mitochondria from L-aminocamitine and palmitoyl-CoA The two medium\long-chain camitine acyltransferases of mitochondria (CPTo and CPTi) and the medium\long-chain camitine acyltransferase of peroxisomes (COT) have been compared in their inhibition by L-aminocamitine and their ability to use L-aminocamitine as substrate using a radiochemical isotope forward assay with [1-14C]pa1mitoyl-CoA or [1- l4Cidecanoyl-CoA as substrate (70). The 150 for L-aminocarnitine of CPT0 of outer 27 mitochondrial membrane vesicles is approximately 250uM while the 150 for L-aminocamitine of CPTi of inner mitochondrial membrane vesicles and purified mitochondrial CPT is 2511M. The 150 for L-aminocamitine of medium\long-chain camitine acyltransferase activity of intact peroxisomes is approximately ZSOILM (70) and purified peroxisomal COT is not inhibited by 2mM L-aminocamitine. L-Aminocamitine was not acylated with palmitoyl-CoA as cosubstrate by mitochondrial CPT}, mitochondrial CP'I‘o or by peroxisomal COT in agreement with Jenkins and Griffith (67). L—Aminocamitine was acylated, though, with octanoyl-CoA as cosubstrate by CPT0 of outer mitochondrial membrane vesicles; the rate with 20mM L-arninocamitine was 34% of the rate with 5mM L—camitine for CPT0 of outer mitochondrial membrane vesicles. CPT} of inner membrane vesicles and purified mitochondrial CPT used L-aminocarnitine with octanoyl-CoA as cosubstrate at 20% of the rate with L-camitine. M_alonyl-COA Regul_ation of Medium\longflain Camitine Acyluwe Actiflty in RM Malonyl-CDA, the first committed intermediate in fatty acid synthesis acts in the coordinate regulation of fatty acid synthesis and degradation via inhibition of the outer form of carnitine palmitolytransferase (CPl‘o) of mitochondria. Malonyl-CDA inhibition of mitochondrial CPTo was first reported by Mchy 9; £1 in 1978 (71,72). Malonyl-CoA regulation of mitochondrial CPTo allows inhibition of fatty acid oxidation via CPTo when fatty acid synthesis is occurring preventing a futile cycle of synthesis and oxidation (30). Malonyl- CoA does not inhibit the inner mitochondrial CPT (CPTi) (71,72). It has not been established if malonyl-CoA sensitive CPTo and malonyl-CoA insensitive CPT} are the same protein or if they are two different proteins ( 19). It has been proposed that CPTo and CPTi are different proteins (22-24,70,71,72) and that CPTo is destroyed by detergent solubilization (22,23,73). 28 Recently it has been reported that malonyl-CoA sensitive CPT can be solubilized in octylglucoside (16). Bergsetlr e_t a_l have shown that rat liver mitochondrial CPT solubilized from the membrane can be separated from a malonyl—CoA binding protein(s) and have proposed that CPT is sensitive to malonyl-CoA inhibition through association with a regulatory protein (74). Zammit e_t _a_l (75) have reported different molecular weights for malonyl-CoA sensitive CPTo and the inner CPTi and Zammit e_t a_l (76) have reported different molecular weights for CPT‘O and malonyl-CoA binding. Malonyl-CoA inhibition of mitochondrial CPTo has been extensively studied. In intact mitochondria from fed rats the Ki for malonyl-CoA inhibition of CPT0 is 1.511M (77). The Ki for malonyl-CoA inhibition increases to 5.00M in mitochondria from 42 hour fasted rats (77). Experimental conditions which cause substrate depletion, malonyl-CoA depletion, or high acyl- CoA concentrations can influence the determination of malonyl-CoA sensitivity (77). The medium\long-chain camitine acyltransferase (COT) activity of intact rat liver peroxisomes isolated by centrifugation through iso-osmotic Nycodenz solution is inhibitable by malonyl-CoA (40). The concentration of malonyl-CoA required to inhibit 50% of COT activity of intact peroxisomes is 2.211M (40). The ratio of C10\C16 activity for COT of intact peroxisomes is 2.1 (40). The COT activity of intact peroxisomes is approximately 20% of the COT activity of the liver homogenate (40). The specific activity of COT of intact peroxisomes assayed with an isotope forward assay using L-3H-camitine is 4.9 i 0.43 nmole min'l mg'1 with palmitoyl-CoA as substrate and 10.5 i 1.3 nmole min-1 mg1 with decanoyl-CoA as substrate (40). Seventy percent of the COT activity of intact peroxisomes can be released by sonication but is only 20% inhibited by 10p.M malonyl—CoA compared with 90% inhibition of intact peroxisomal COT by 1011M malonyl-CoA (40). 29 The effect of malonyl-CoA on the COT activity of rrricrosomal medium\long-chain canritine acyltransferase activity has not been reported. It has been shown though that the COT activity of microsomes assayed ‘with palmitoyl-CoA as substrate is increased with fasting (56). The concentration of malonyl-CoA in rat liver is decreased in the fasted state. The malonyl-CoA content of liver from fed rats is 7.5 nmole g‘1 wet weight and from 24 hour fasted rats it is 1.7 nmole g‘1 wet weight (78). Etomoxgy‘ l-CoA Inhibition of Meditlm\long-chain Camitine Acyltransferase Activitm Rat 1335; The regulation of medium\long-chain camitine acyltransferase activity by malonyl—CoA has been studied using epoxy containing fatty acid derivatives such as TDGA (2-tetraglycidic acid) and etomoxir (ethyl-2-[6-(-chlorophenoxy)hexyl]oxirane-2-carboxy1ate) (22,23 ,79-8 1). These derivatives are activated in vivo via conjugation to coenzyme A. The structure of etomoxiryl-CDA is shown in Fig. 1. Etomoxiryl-CoA is also called B827-33. Etomoxiryl-CoA is a specific inhibitor of mitochondrial Cl’l‘0 (82). Etomoxiryl-CDA has been proposed to be an active-site directed, irreversible inhibitor of mitochondrial CPTo (22,23); TDGA-CoA has been characterized as an active-site directed, irreversible inhibitor of CP'l‘0 (79). In rat liver mitochondria 3H-etomoxir forms a covalent adduct to an approximately 94,000 MW protein while in rat skeletal muscle mitochondria 3H-etomoxir forms a covalent adduct to an approximately 86,000 MW protein. It has been proposed that these labeled proteins are CPTo (22.23). Lopaschuck §_t_ _a_l_ have used etomoxir in studying heart function in fatty acid perfused ischemic rat hearts (83,84). It was proposed that etomoxir can protect hearts from fatty acid 30 induced ischemic injury by inhibiting CPTo and decreasing myocardial long-chain acylcamitine levels. Lopaschuck e_t g1 showed that etomoxir protects hearts from fatty acid induced ischemic injury independent of changes in long-chain acylcamitine and long-chain acyl-CoA (83). They concluded that the protective effect of etomoxir could have resulted from a stimulation of glucose oxidation and that etomoxir at micromolar concentrations can inhibit both CP’I‘o and the inner mitochondrial CPTi. (84). 31 N N S to ‘3‘“ 9 9 ' 87% inhibitable by malonyl-CoA. Nine out of the 25 preparations isolated by differential centrifugation are 100% inhibited by malonyl-CoA. In these assays microsomal COT was assayed spectrally using a rate forward assay. Since microsomes contain high acyl-CoA hydrolase activity which contributes to large blank values due to camitine independent CoASH release, a series of experiments was done to confirm that the COT activity measured spectrally by the rate forward assay is producing the expected product, 38 Table 2. Malonyl-CoA Inhibition of COT Activity of Microsomes Isolated by Three Different Procedures COT was assayed spectrally with 17 uM decanoyl-CoA, 1.7 mM L-carnitine, 150 M DTBP in 50 mM potassium phosphate, pH 7.5. Malonyl—CoA was 17 1.1M. Data presented as mean i SEM. n = number of preparations assayed. Isolation COT Activity %Inhibition Procedure n munit/mg protein by Malonyl- CoA Differential 25 6.2 :1: 2.7 87 :I: 11.7 Centrifugation CsCl-Sucrose 3 ’ 3.7 i 0.3 98 i 1.3 Gradient CaCl.z 3 4.8 i 0.04 82 :l: 5.3 Precipitation 39 decanoylcamitine, and that decanoylcamitine formation is inhibited by malonyl—COA. Microsomes were incubated with [1-“C1decanoy1-CoA and L—camitine and the [1- 14C]—decanoylcamitine formed was separated by HPLC and the dpm in the [1-14C]-decanoyl- camitine peak determined. Figure 1A demonstrates the production of [l-“Cl-decanoylcamitine by microsomes representing a COT activity of 5.7 nmole min'1 mg‘1 protein and figure 1B shows inhibition of [1-“C]-decanoylcamitine production by malonyl—CoA In this experiment, 100% inhibition of [1-“c1-decanoy1camitine production occurred. The same microsomal preparation was also assayed spectrally using identical conditions with unlabeled decanoyl- CoA; the cor activity assayed spectrally was 4.6 nmole rrtin'1 rng'l protein with 92% inhibition by malonyl-CoA. The effect of the concentration of malonyl-CoA and etomoxiryl-CoA on membrane bound microsomal COT activity is shown in Figure 2. When microsomal COT is assayed spectrally with 17 pM decanoyl-CoA and 1.7 mM L—camitine, the concentration of malonyl- CoA required to inhibit 50% of microsomal COT is 5.3 i 0.43 ttM; no preincubation of microsomal COT with malonyl-CoA is required. When microsomal COT is assayed either spectrally or radiochemically, the concentration of etomoxiryl—COA required for 50% inhibition of microsomal COT is 0.58 :t 0.17 1.1M; membrane bound microsomal COT was incubated with etomoxiryl-CoA for two minutes prior to assay. The effect of the concentration of malonyl—CoA on detergent solubilized microsomal COT activity is shown in Figure 3. 'Microsomal COT was solubilized with the zwitterionic detergent CHAPS (3-[(3-Cholamid0propyl)-dimethylammonio]-1-propane sulfonate). The specific activity of membrane bound microsomal COT is 7.9 i 1.1 nmole min'1 mg‘1 . The specific activity of the CHAPS-solubilized microsomal COT is 4.7 a: 0.6 nmole min‘l mg'1 . 40 Figure 1. Effect of Malonyl-CoA on the Production of [l-“Cl-Decanoylcarnitine. The production of [11‘Cl-decanoylcamitine by microsomal bound COT was detemrined with a radiochemical assay using [1-“C]-decanoyl-COA as described in Materials and Methods. Panel A shows the HPLC separation of the [1-‘4c1decanoy1camitine produced; the cor activity was 5.73 nmole min"l mg”1 protein. Panel B shows the effect of the addition of 17 11M malonyl- CoA to the assay. 41 AmmSEEV 08: c2223. 9 O. m m. 0. m p 111 — . d< _ \< ”LN 44: <8 3865.... .m. 6:80 .< Figure 1. urdo 42 Figure 2. Malonyl-CoA and Etomoxiryl-CoA Inhibition of Microsomal COT. Microsomes were prepared by differential centrifugation. COT was assayed spectrally as described in Material_s and Methods in the presence of the concentrations of malonyl-CoA indicated The values were corrected for camitine independent release of CoASH at each malonyl-CoA concentration (n=3). The concentration of malonyl-CoA required to inhibit 50% of the COT activity is 5.3 :t 0.43 ttM. cor was assayed radiochemically using [1-“CJ-decanoyl-COA as described in Materials and Methods in the presence of the concentrations of etomoxiryl-COA indicated (n=2). Microsomes were preincubated with etomoxiryl-COA and [1—14C]-decanoyl- CoA for 2 minutes. The concentration of etomoxiryl-CoA required to inhibit 50% of the COT activity is 0.58 i 0.17 M. COT was also assayed spectrally in the presence of the indicated concentrations of etomoxiryl-CoA and the 150 was identical to the ISO from the radiochemical assay, within experimental error. Initial COT activity (100%) was 10.9 + 2. 9 nmole min l'rng protein for malonyl-CoA and 7. 9 + 0.80 nmole min’1 mg protein for etomoxiryl-CoA. The data are presented as the mean + SEM. Figt 43 Figure 2. '00 o Malonyl CoA O Etomoxir CoA % COT Activity 01 O 1 W- - 5 IO l5 pm Inhibitor 44 Figure 3. Effect of Etomoxiryl-CoA and Malonyl-CoA on CHAPS-Solubilized Microsomal COT. CHAPS-solubilized microsomes were prepared and COT was assayed spectrally as described in Mategls and Methods. COT was determined in the presence of the indicated concentrations of etomoxiryl-CoA with a 2 minute preincubation (closed circles) or malonyl-CoA with no preincubation (open circles). COT activity was corrected for camitine independent CoASH release at each inhibitor concentration. The concentration of etomoxiryl- CoA required for 50% inhibition of COT activity is 1.5 M. The initial (100%) COT activity is 4.7 i 0.6 nmole rnin'l mg‘1 protein. The data are plotted as the mean i SEM (n=2). 45 .21 5:228:00 8:95: on new . 8. ole ' L 0 ID meav .LOO % 00. Figure 3. 46 COT was assayed spectrally with no preincubation for malonyl-CoA and with a 2 minute preincubation for etomoxiryl-CoA. The upper curve (0) shows the effect of malonyl-CoA. CHAPS-solubilized microsomal COT is only inhibited 11% by ZOOuM malonyl-CoA. The lower curve (o) shows the effect of etomoxiryl-CoA. The concentration of etomoxiryl-CoA required to inhibit 50% of CHAPS-solubilized microsomal COT activity is 1.5 uM. CHAPS solubilized microsomal COT is inhibited 11% by 200 uM malonyl-CoA and 78% by 14 1.1M etomoxiryl-CoA. When both 200 pM malonyl-CoA and 14 [1M etomoxiryl-CoA are added to the assay, CHAPS solubilized microsomal COT is inhibited 67%, less than the inhibition with only etomoxiryl-COA. The inhibition of microsomal COT by etomoxiryl-CoA is reversible. When membrane bound microsomal COT is incubated with 40 11M etomoxiryl-CoA for 1 hour, the COT activity is completely inhibited. When these etomoxiryl-CoA inhibited microsomes are solubilized in CHAPS and the CHAPS-solubilized supernatant passed over a Biogel P6 desalting column, the COT activity is completely restored. Two experiments gave an average recovery of 98%. Experiments were tried to reverse etomoxiryl-CoA inhibition of membrane bound microsomal COT with washing, but microsomal COT activity was not stable to multiple washing steps. filmitoyl-COA Inhibition of Microsomal COT Membrane bound microsomal COT was assayed spectrally with decanoyl-CoA as substrate in the presence of increasing concentrations of palmitoyl-CoA. The data are shown in Figure 4. The concentration of palmitoyl-CoA required to inhibit 50% of microsomal COT activity is 10.9 i 0.46 M. Palmitoyl-CoA is also a substrate of microsomal COT at lower concentrations. The ratio of decanoyltransferase to palmitoyltransferase activity of microsomes 47 Figure 4. Palmitoyl-CoA Inhibition of Microsomal COT. Microsomes were prepared using differential centrifugation and COT was assayed specrrally as described in the Materials and Methods. Palmitoyl-CoA was added to the assay at the concentrations indicated. The values were corrected for camitine independent CoASH release at each palmitoyl-CoA concentration. The initial COT activity (100%) was 12.9 r. 2.4 nmole rnin-l mg’1 protein. The data are plotted as the mean of three experiments 1‘ SEM. The concentration of palmitoyl-CoA required to inhibit 50% of the COT activity was 10.9 i 0.46 “M. The effect of palmitoyl-CoA on microsomal COT was also determined using a radiochemical assay as described in Materials and Methods. The %[1-“C]-decanoylcamitine formation in the presence of 17 “M palmitoyl- CoA is shown by the x (n=2); initial cor activity (100%) was 14.6 :t 3.5 nmole min" mg-1 protein. Figure 4. 48 IOO 75 Eh. 2 a. r—‘ -8pr ~ 5|.000 uolonyl 00A 1 1 0.2 0.4 0.6 0.8 l.0 Relative mobility DISCUSSION Localization of Malonyl-CoA Sensitive COT The data presented herein show that rat liver contains more than one malonyl-CoA sensitive medium-chain/long-chain camitine acyltransferase. In addition to the malonyl-CoA sensitive CPTo associated with mitochondria, marker enzyme distribution studies show that there is a malonyl-CoA sensitive COT associated with microsomes. This enzyme has been tentatively designated a medium-chain transferase (COT) because both microsomes and a partially purified preparation show a higher activity with decanoyl—CoA than with palmitoyl- CoA as substrate. Recent studies indicate the COT activity of intact rat liver peroxisomes is also inhibitable by malonyl-CoA (40). If so, all of the camitine acyltransferases of liver that exhibit medium-chain and long—chain camitine acyltransferase activity and which are in contact with cytosolic pools of acyl-CoAs can be inhibited by malonyl-CoA. Surprisingly, the activity in the 100,000 x g pellet, the fraction enriched in microsomes, is more inhibited by 17 11M malonyl-CoA than the activity in the 6,000 x g fraction which contains primarily CPT0 and peroxisomal COT. For the experiment shown in Table 1, COT in the microsomal enriched fraction was 77% inhibited by malonyl-CoA, while the mitochondrial enriched fraction (6,000 x g pellet) was only 37% inhibited. This degree of inhibition is less than that shown for purer mitochondrial preparations (42); it seems likely that the low percent inhibition is due to the presence of solubilized peroxisomal COT and also to damaged mitochondria, thereby exposing CPTi (CPT ~11). The medium-chain and long-chain 62 63 camitine acyltransferase activities of microsomes are both inhibited by malonyl-CoA and palmitoyl-CoA, suggesting these activities result from a single enzyme. Microsomal CAT activity is not inhibited by malonyl-CoA, which suggests it is due to a different enzyme as has been proposed (39). The potential contribution of mitochondria and peroxisomes to the microsome enriched fractions used for these studies has been determined. The specific activity of the malonyl-CoA sensitive microsomal COT (see Table II) is comparable to the specific activity of malonyl-CoA sensitive COT in intact density gradient purified peroxisomes (9.32 munits/mg) (40), while the percent recovery of urate oxidase in the microsome enriched fraction is < 10% (see Table 11). Thus, the COT in our microsome preparations could not result entirely from contamination of the 100,000 x g pellet with peroxisomes. This conclusion is confirmed by the finding that only 17% of the COT activity of microsomes is irnmunoprecipitated by anti-peroxisomal COT using conditions that completely inhibit peroxisomal COT (see Fig. 5). Recent reports indicate CP'I‘o is associated with the outer mitochondrial membrane (1 1,2932). Contamination of microsomes by outer membrane fragments seems plausible, so the percent recovery in the microsome enriched fraction of both cytochrome c oxidase, an inner mitochondrial membrane marker, and monoamine oxidase, an outer mitochondrial membrane marker, was determined. They both represent < 3.0% of the total marker activity (see Table I). Anti-CPT that inactivates both CPT0 and CPT} of mitochondria (16), and that reacts to a single peptide on Western blots of rat liver mitochondria (see Fig. 5 of ref. 89) had no detectable effect on microsomal COT (see Fig. 5B). Thus, the microsome enriched fractions used for these studies were not significantly contaminated with either peroxisomes, or mitochondrial inner or outer membranes. The microsomal COT has properties similar to the malonyl-CoA sensitive CPT that 64 occurs in preparations enriched with outer mitochondrial membranes of liver. These include a tight association with the membrane and a high degree of malonyl-CoA sensitivity; some of our rough and smooth endoplasmic reticulum preparations were completely inhibited by 17 11M malonyl-CoA, such as the preparation shown on Fig. 1. The malonyl-CoA sensitivity of outer mitochondrial membranes and of our preparations is quite labile, sensitive to some detergents, and the 150’s are low micromolar. . The microsomal COT is quite stable in CHAPS and Tween-20, but the activity is rapidly lost in Triton X-100 (data not shown). However, the titration curves for malonyl-CoA inhibition of COT are different for liver microsomes (see Fig. 2) than those shown for rat liver mitochondria (65,66,71). Abrupt breaks in the titration curve for mitochondria were not found. Microsomal COT is strongly inhibited by concentrations of palmitoyl~CoA > 11 11M. The inhibition curve is almost identical to the one reported for inhibition of the CPT activity of the outer membrane enriched preparations reported by Murthy and Pande (29); compare Figure 3 to Figure 2 of Murthy and Pande (29). In contrast, the velocity versus [palmitoyl-CoA] curves for CPTo of intact rat liver mitochondria appear sigmoid (see Fig. 1 of Saggerson et al. (66), Fig. 1 of Cook et al. (32),'and Fig. 2 of Grantham and Zammit (65). The inhibition of microsomal COT by increasing concentrations of palmitoyl-CoA provides a method for differentiating CPTo of rat liver mitochondria from microsomal COT. Etomoxiryl-CoA inhibits CPT0 of intact rat liver mitochondria with an 150 value of approximately 3 nm (22). The 150 of membrane bound microsomal COT for etomoxiryl-CoA is approximately 600 nm (see Figure 2). There is a ZOO-fold difference in the etomoxiryl-CoA 150 of membrane bound microsomal COT and mitochondrial CPTO. Octyl glucoside solubilized CPT activity in outer mitochondrial membrane enriched vesicles (OMV CPT) is inhibited 41% 65 by 0.2 M etomoxiryl-CoA (88). The ISO of detergent solubilized microsomal COT is approximately 1.5 11M (see Figure 3). The concentration of etomoxiryl-CoA required to inhibit OMV CPT and microsomal COT is similar. [am-Etomoxir also labels proteins with similar molecular weights in outer mitochondrial membrane enriched vesicles and in microsomes. In microsomes [3H]-etomoxir forms a covalent adduct to a major protein with a molecular weight of approximately 51-57,000 daltons and to a minor protein with a molecular weight of 87 ,000 daltons (see Figure 8). In octyl glucoside solubilized OMV CPT eluted from a hydroxyapatite column [3H]-etomoxir forms a covalent adduct to a major protein with a molecular weight of approximately 90,000 daltons and to a minor protein with a molecular weight of approximately 45,000 daltons (88). The microsomal COT has properties similar to those reported for the malonyl-CoA sensitive medium-chain camitine acyltransferase of intact peroxisomes. In addition to comparable specific activities, the concentration of malonyl-CoA required to inhibit 50% of COT activity is 5.3 :l: 0.43 11M in microsomes (see Fig. 2) and is 2.2 1.1M in intact peroxisomes (40). Although both enzymes show a higher Vm with C10-CoA than C16-CoA, the ratio of C10\C16 activity for the microsomal COT is approximately 10 and the ratio for the medium- chain transferase of intact peroxisomes is 2.1 (40). The data in Table I show that the COT activity of microsomes represents at least 15% of the total COT activity of the liver homogenate. The COT activity of intact peroxisomes is approximately 20% of the COT activity of the liver homogenate (40). While the microsomal COT and the peroxisomal COT show similar substrate specificities and kinetic properties, the data in Fig. 5 clearly show that the microsomal COT is antigenically different than purified peroxisomal COT. The solubilization characteristics of 66 microsomal-bound COT and the COT activity of intact peroxisomes are strikingly different. Sonication releases at least 70% of the COT activity of intact peroxisomes (40). Previously, it was shown that freeze/thaw treatment of peroxisomes completely releases peroxisomal COT (34,39), while freeze/thaw treatment of microsomes releases < 10% of microsomal COT activity (39). While peroxisomal COT is a soluble, matrix enzyme, microsomal COT is firmly membrane-bound, requiring detergents for solubilization. The mechanism of etomoxiryl-CoA and malonyl-CoA inhibition of microsomal COT seems to be different. Etomoxiryl-CoA inhibition is time dependent while malonyl-CoA inhibition does not require preincubation. Membrane bound rrricrosomal COT is inhibited by malonyl-CoA and etomoxiryl-CoA (see Figure 2) while detergent (CHAPS) solubilized microsomal COT retains inhibition by etomoxiryl-CoA but is not inhibited by up to 200 tLM malonyl-CoA (see Figure 3)., It has been proposed that malonyl-CoA is a reversible inhibitor of CPT and etomoxiryl-CoA is an irreversible, active site directed inhibitor of CPT (22,23) , but the etomoxiryl-CoA inhibition of microsomal COT is reversed with detergent solubilization and desalting the solubilized supernatant. It has also been reported that the etomoxiryl-CoA inhibition of rat liver mitochondrial CPT can be reversed with dialysis (90). The reversibility of etomoxiryl-CoA inhibition of microsomal COT by detergent solubilization could indicate that a regulatory protein is separated from a catalytic protein. There is no direct evidence that the [3H]-etomoxir labeled proteins are microsomal COT or that they are involved in the etomoxiryl-CoA inhibition of microsomal COT. The reversal of etomoxiryl-CoA inhibition of rrricrosomal COT with detergent solubilization suggests that etomoxiryl-CoA is not forming a covalent adduct with microsomal COT and that the [3H]-etomoxir labeled proteins are not microsomal COT. The incorporation of label into /. 67 the approximately 87,000 dalton molecular weight microsomal protein is decreased by preincubation with malonyl-CoA suggesting that this protein could be involved in the malonyl- CoA inhibition of microsomal COT. The labeling in the major microsomal protein with a molecular weight of approximately 51-57,000 daltons is not decreased by preincubation and is decreased by a chase incubation with unlabeled etomoxiryl-CoA indicating it may have other functions unrelated to microsomal COT activity. Chapter 3. Kinetic Characterization of Membrane Bound Microsomal COT Rat liver end0plasmic reticulum contains malonyl-CoA sensitive medium/long-chain camitine acyltransferase activity (microsomal COT). The kinetic constants for microsomal COT determined using a radiochemical assay show the K05 for L-carnitine is 0.42 :l: 0.04 mM and the K05 for decanoyl-CoA is 1.9 :t 0.1 uM. Microsomal COT exhibits Michaelis-Menten kinetics; Hill coefficients are 0.91 i 0.03 for decanoyl-CoA as varied substrate and 0.96 :I: 0.12 for L-carnitine as varied substrate. Microsomal COT is inhibited by DL-aminocarnitine. The concentration of DL-arninocamitine required for 50% inhibition of malonyl—CoA sensitive COT is 0.5 :l: 0.12 mM. DL-Aminocarnitine inhibits microsomal COT competitively with respect to L—carnitine with a Ki of 40 ttM. Decanoyl-DL-aminocamitine and palmitoyl-DL- arninocamitine also inhibit microsomal COT. The concentrations of decanoyl-DL- aminocamitine and palmitoyl-DL—aminocamitine required for 50% inhibition of malonyl-CoA sensitive microsomal COT are 6.8 :l: 1.1 11M and 4.3 :t 1.0 M, respectively. The results show that the DL-aminocarnitine inhibition of malonyl-CoA sensitive medium/long—chain camitine acyltransferase activity of rat liver endoplasmic reticulum is similar to the L-aminocarnitine inhibition of the malonyl-CoA sensitive medium/long-chain camitine acyltransferase activity of outer mitochondrial membrane enriched vesicles of rat liver. The kinetic characteriStics of microsomal COT can be used to distinguish it from mitochondrial CPI“ and peroxisomal COT. 68 Introduction Medium/long-chain camitine acyltransferase activity in rat liver is associated with mitochondria, peroxisomes, and endoplasmic reticulum. The medium/long-chain camitine acyltransferase activity associated with rat liver endoplasmic reticulum (microsomal COT) is strongly inhibited by malonyl-CoA (see Chapter 2), like the well characterized inhibition of the outer mitochondrial medium/long-chain camitine acyltransferase (mitochondrial CPTO). (30,71,72,85). Medium/long-chain camitine acyltransferase activity is also located on the inner side of the inner mitochondrial membrane (mitochondrial CPl‘i); mitochondrial CPTi is not sensitive to inhibition by malonyl-CoA (71,72). It is not known if CPTo and CPT} are the same protein with different locations and regulatory properties or if they are distinct proteins (19). Mitochondrial CPTo has been reported to be located both on the outer side of the inner mitochondrial membrane (25-28) and on the inner side of the outer mitochondrial membrane (11,29-32). Peroxisomal medium/long-chain camitine acyltransferase activity is associated with the matrix (peroxisomal COT) (14,33-35). The ldnetic characteristics of membrane bound microsomal COT were determined for comparison with the known kinetic properties of mitochondrial CP’I‘o and CPTi and with peroxisomal COT. Mitochondrial enriched fractions of rat liver homogenates prepared using differential centrifugation can be contaminated by significant quantities of peroxisomes and endoplasmic reticulum (35,58,59,61). The kinetic properties of membrane bound microsomal COT could be useful in determining the fraction of medium/long-chain camitine acyltransferase 69 activity associated with rat liver mitochondria that is due to microsomal contamination. Aminocamitine (3-amino-4-trimethylaminobutyric acid) is a noncovalent inhibitor of camitine acyltransferases (67-70). Aminocamitine has been proposed as a antiketogenic agent which could be useful in the treatment of ketoacidosis associated with diabetes mellitus (67- 69). The aminocamitine inhibition of both mitochondrial and peroxisomal medium/long-chain camitine acyltransferase activities in rat liver has been studied (67-70) but the aminocamitine inhibition of rat liver microsomal medium/long-chain camitine acyltransferase activity has not been characterized. Herein the DL-aminocarnitine inhibition of microsomal COT and the inhibition of microsomal COT by decanoyl-DL—aminocarnitine and palmitoyl-DL- aminocamitine were determined. The DL-aminocamitine inhibition of microsomal COT is similar to the L-aminocamitine inhibition reported for the CPTo activity of outer mitochondrial membrane enriched vesicles. 7O Results Determination of Kinetic Constants Mitochondrial CPT, microsomal COT, and peroxisomal COT contribute to the cytosolic production of medium-chain and long-chain acylcamitines in liver. The kinetic properties of purified mitochondrial CPT and purified peroxisomal COT have been studied. The kinetic properties of membrane-bound microsomal COT were determined for comparison. A velocity versus decanoyl-CoA concentration plot for membrane bound microsomal COT is shown in Figure 1A and a velocity versus L-camitine concentration plot is shown in Figure 2A. The velocity versus substrate concentration plots for microsomal COT are hyperbolic; Hill coefficients determined from the slope of a Hill plot are 0.91 i: 0.03 for decanoyl—CoA as varied substrate and 0.96 :t 0.12 for L-carnitine as varied substrate. The double reciprocal plot with decanoyl-CoA as the varied substrate is shown in Figure 1B and for L-camitine as the varied substrate in Figure 2B. The double-reciprocal plots were linear indicating microsomal COT follows Michaelis-Menten kinetics. Kinetic constants calculated from the equation for the line show K05 for decanoyl-CoA is 1.9 r. 0.1 HM and vmam is 16.3 a 1.1 nmole min'l mg’1 protein while the K05 for L-carnitine is 0.42 i 0.04 mM and the Vmax is 10.7 i 0.8 l nmole min' mg]. 71 72 Figure 1. Velocity versus Decanoyl-CoA Concentration Curve and Double-Reciprocal Plot for Microsomal COT. Microsomal COT activity was determined in the presence of increasing decanoyl-CoA concentrations. Velocity versus decanoyl-CoA concentration plot is shown in Panel A. Double-reciprocal plot is shown in Panel B; the line was fitted by least- squares regression with correlation coefficient (r) = 0.989. Kinetic constants calculated from the equation for the line show K05 = 1.9 :t 0.1 11M and V"mm = 16.3 :1: 1.1 nmole min'l mg’1 protein. Data plotted as mean of three experiments. 73 Figure 1. 1C)“ E IO '1' .E E a) 5 '5 E C l l T IO 20 [Deconoyl-CoA] ,ttM __ B . . CE" 0.3 T .E E 0.2 .S.’ E r: O.| [ \>. I J LO 2.0 VI-peconoyl-COA] FM 74 Figure 2. Velocity versus L-Carnitine Concentration Curve and Double-Reciprocal Plot for Microsomal COT. Microsomal COT activity was determined in the presence of increasing L-carnitine concentrations. Velocity versus decanoyl-CoA concentration plot is shown in Panel A. Double-reciprocal plot is shown in Panel B; the line was fitted by least— squares regression with correlation coefficient (r) = 0.985. Kinetic constants calculated from the equation for the line show K0.5 = 0.42 i 0.04 mM and Vmam = 10.8 :1: 0.8 nmole min’1 mg’1 protein. Data plotted as mean of two experiments. 75 . _. A Flgure 2. |Cfi C E '0 _ . .‘s E 0 '5 5 E C l l 4 a [L- Camitine] mM '0 0.3 _. B E T .E E 0.2 2 E c 0.1 \>.. — l I l/|:L - Camitine] mM Effect of pH on Microsomal COT Activity The effect of pH on microsomal COT activity is shown in Figure 3. The pH optimum for membrane bound microsomal COT assayed in the forward direction is 8.5. DL-Aminocamitine Inhibition of Microsomal COT The effect of DL-aminocamitine concentration on microsomal COT activity is shown in Figure 4. Membrane bound microsomal COT is completely inhibited by DL—aminocarnitine. The concentration of Dir-aminocamitine required to inhibit 50% of microsomal COT is 0.5 i 0.12 mM. The effect of DDaminocarnitine on the kinetic parameters of microsomal COT with L-camitine as variable substrate is shown in Figure 5. DL-Aminocarnitine is a competitive inhibitor with respect to L—camitine. The Ki for DL—arninocamitine is 40 M calculated from a replot of Km app versus [DL-aminocarnitine]. DL-aminocarnitine is a substrate of microsomal COT with decanoyl-CoA as cosubstrate. Microsomal COT activity with 10 mM DL- aminocamitine is 0.12 nmole min'1 mg"1 protein. Decgnoyl-DL-Aminocamitine and Pglrnitoyl-DIIAmrniarnitine Inhibition of MicrosomJal COT The effect of decanoyl-DL-aminocarnitine and- palmitoyl-DL-aminocamitine concentration of microsomal COT activity is shown in Figure 6. Decanoyl-DL-aminocamitine and palmitoyl-DDaminocamitine inhibit microsomal COT. The concentrations of decanoyl- DL—aminocarnitine and palmitoyl-DL-aminocarnitine required to inhibit 50% of microsomal COT activity are 6.8 i- 1.1 11M and 4.3 i 1.0 11M, respectively. we it 77 Figure 3. The pH Optimum of Microsomal COT. Microsomal COT was assayed with 6 mM L-carnitine and 16 uM [l—“C]-decanoyl-C0A in 50 mM potassium phosphate at the pH values indicated. The pH optimum for membrane bound microsomal COT assayed in the forward direction is 8.5. Data plotted as the mean of two experiments i SEM. 78 Figure 3. _ IO- 5 .9: TEE 29:: IO 79 Figure 4. DL-Aminocarnitine Inhibition of Microsomal COT. Microsomal COT activity was determined in the presence of increasing concentrations of DL—aminocamitine as indicated. The concentration of DL-aminocarnitine required to inhibit 50% of microsomal COT is 0.5 :t 0.12 mM. Data plotted as the mean of two experiments i SEM. 80 Figure 4. IO- OI nmole min"I mg" l l IO 20 DL - Aminocamitine mM " HF. '"u-a‘ n1- 81 Figure 5. Effect of DL-Aminocarnitine on the Kinetic Parameters of Microsomal COT. Microsomal COT was assayed with increasing concentrations of L-carnitine in the presence of the indicated concentrations of DL—aminocamitine. A double-reciprocal plot shows DL- aminocamitine is a competitive inhibitor of microsomal COT with respect to L—carnitine. A replot of K058“, versus [BL-aminocamitine] gives a K, = 40 11M. Data plotted as the mean of two experiments. Figure 5. l . /v, nmole mln"'mtg"l 0.6 0.3 82 A 0.2 mM DL- Aminocamitine o 0.! mM DL-Aminocomitine 0 Control ‘ O H |.0 2.0 '/[L-,Comiiine],mM 83 Figure 6. Decanoyl-DL-Aminocarnitine and Palmitoyl-DL-Aminocarnitine Inhibition of Microsomal COT. Microsomal COT activity was determined in the presence of increasing concentrations of decanoyl-DL-aminocamitine and palmitoyl-Dlraminocamitine as indicated The concentrations of decanoyl-DL—aminocamitine and palmitoyl-Dbaminocarnitine required to inhibit 50% of microsomal COT are 6.8 i 1.1 1.1M and 4.3 a 1.0 M, respectively. Data } plotted as the mean of two experiments i SEM. Figure 6. nmole min" mg IO U! 84 O Decency! - DL-Aminocarnitine O Palmitoyl - DL-Aminocarnitlne O o - IN 0 K9 l 1 4| l 5 l0 “ 25 [Inhibitor] pM Discussion Rat liver contains medium/long-chain camitine acyltransferase activity located in microsomes, peroxisomes, and mitochondria (19,41). Microsomal COT, like purified peroxisomal COT, exhibits Michaelis-Menten kinetics. Velocity versus substrate plots for microsomal COT are hyperbolic with a hill coefficients of 1 (see Figures 1 and 2). Hill coefficients for purified mouse liver peroxisomal COT for acyl-CoA as variable substrate are 1-1.2 (14). Purified rat liver peroxisomal COT also exhibits Michaelis-Menten kinetics (34). In contrast, purified beef heart mitochondrial CPT and membrane bound CPT of intact rat heart mitochondria show allosteric kinetics with sigmoid velocity versus substrate plots indicating cooperative binding of acyl-CoA and L-camitine (5,20). Purified rat liver mitochondrial CPT also shows complex, biphasic kinetics (34). Microsomal COT can be distinguished from mitochondrial CPT since microsomal COT does not show cooperative substrate binding. Although it is difficult to compare kinetic constants determined using different assay conditions, the K05 for decanoyl-CoA of microsomal COT is similar to values reported for purified mouse liver peroxisomal COT and membrane bound rat heart and liver mitochondrial CPT. The K05 of purified peroxisomal COT for L-camitine with decanoyl-CoA as cosubstrate, however is ~7-8 fold less than the K05 of membrane bound microsomal COT and membrane bound mitochondrial CPT with decanoyl-CoA as cosubstrate. . The K05 of microsomal COT for decanoyl-CoA is 1.9 11M and 0.42 mM for L-carnitine (see Figures 1 and 2) determined using a rate forward, spectral assay at pH 7.5 using DTBP as a thiol trapping agent. The K05 85 86 of purified mouse liver peroxisomal COT for decanoyl-CoA is 2.2 M and 55 uM for L- camitine determined with a rate forward, spectral assay at pH 8.0 using DTNB (14). The K05 of membrane bound rat heart mitochondrial CPT for decanoyl-CoA is 3 uM and 0.2-0.7 rnM for L—carnitine determined with a rate forward, spectral assay at pH 8.0 using DTNB (5). The K05 of membrane bound rat liver mitochondrial CPTo for palmitoyl-CoA is 27 11M and 0.16 mM for L-camitine determined with an isotope forward assay using [I‘m-camitine (91). The K05 of mitochondrial CPT and peroxisomal COT for L-carnitine varies with acyl-CoA chain length (14,20). The K05 of microsomal COT for L-carnitine was only determined with decanoyl-CoA as cosubstrate. Microsomal COT can be distinguished from peroxisomal COT by its higher K05 for L-camitine with decanoyl-CoA as cosubstrate. The pH optimum for membrane bound microsomal COT is different than the reported pH optimum values for purified mitochondrial CPT and peroxisomal COT. The pH optimum for microsomal COT assayed in the forward direction is 8.5 (see Figure 3). The pH optimum for purified mouse liver peroxisomal COT assayed in the reverse direction is 8.0 (14). Purified beef heart mitochondrial CPT in octylglucoside assayed in the forward direction has a pH optimum at pH 7.0 (92). Membrane bound microsomal COT can be distinguished from peroxisomal COT and mitochondrial CPT by its higher pH optimum. Aflnogmflfine Inhibition The data presented herein shown that while microsomal COT is less sensitive to inhibition by DL-arninocarnitine than mitochondrial CPT activity, the DL-aminocamitine inhibition of microsomal COT is similar to the L-aminocamitine inhibition reported for CPT activity of outer mitochondrial membrane enriched vesicles (OMV CPT). Microsomal COT 87 is less sensitive to DL—aminocarnitine inhibition than the CPT activity of Triton X-100 treated mitochondria and to the L-aminocarnitine inhibition reported for mitochondrial CPT. The kinetic properties of microsomal COT are also different than the kinetic properties reported for mitochondrial CPT and peroxisomal COT. Microsomal COT is inhibited by DL- Aminocarnitine with an 150 value of 0.5 mM (see Figure 4) and OMV CPT is inhibited by L- aminocamitine with an 150 value of 0.25 mM (70). Microsomal COT uses Dbanrinocamitine as substrate with decanoyl-CoA as cosubstrate although presumably only L-aminocarnitine is acylated. OMV CPT uses L-aminocarnitine as substrate with octanoyl—CoA as cosubstrate but not with palmitoyl-CoA as cosubstrate (70). Microsomal COT, though, is inhibited by high concentrations of palmitoyl-CoA (see Figure 4 of Chapter 2). The DL-aminocarnitine inhibition of microsomal COT is also similar to the L-aminocamitine inhibition of the medium/long-chain camitine acyltransferase activity associated with intact, gradient ptnified peroxisomes (70). The DL-aminocarnitine inhibition of microsomal COT differs from the aminocamitine inhibition of CPT activity of detergent treated mitochondria, intact mitochondria, inner mitochondrial membrane enriched vesicles (IMV CPT) and purified mitochondrial CPT. DL- Aminocarnitine at a concentration of 5 11M inhibits 64% of the CPT activity of rat liver mitochondria treated with Triton X—100(67). Decanoyl-DL-Aminocarnitine and palmitoyl-DL- aminocamitine at 5 uM inhibit 75 % and 99%, respectively, of the CPT activity of Triton X- 100 treated mitochondria (68). The 150 for inhibition of CPTo of intact rat liver mitochondria is 62.5 M by L-aminocarnitine and 2.2 M by palmitoyl-Iraminocamitine (69). The 150 for inhibition of IMV CPT activity and for purified mitochondrial CPI‘ is 25 M (70). In contrast, microsomal COT is less sensitive to aminocamitine inhibition. The 150 for DL—arninocamitine 88 inhibition of microsomal COT is 0.5 mM (see Figure 4) and the ISO for decanoyl-DL- aminocamitine and palmitoyl-DL-aminocamitine inhibition of microsomal are 7 and 4 11M, respectively (see Figure 6). Microsomal COT also differs from mitochondrial CPT in the ability to use aminocamitine as substrate. Microsomal COT like the OMV CPT activity uses aminocamitine as substrate with decanoyl-CoA as cosubstrate for microsomal COT and octanoyl-CoA as cosubstrate for OMV CPT (70). Triton X-100 treated mitochondria do not use aminocamitine as substrate with palmitoyl-CoA as cosubstrate (67). It has been reported that intact rat liver mitochondria use aminocamitine as substrate with palmitoyl-CoA as cosubstrate (69). Rat liver mitochondria, though, can be significantly contaminated by microsomes (61), so the enzyme responsible for the formation of the palmitoyl-Iraminocarnitine could be microsomal. Pmified mitochondrial CPT, IMV CPT and purified peroxisomal COT do not use aminocamitine as substrate with either octanoyl-CoA or palmitoyl-CoA as cosubstrate (70). The sensitivity of liver medium/long-chain camitine acyltransferase activity to aminocamitine inhibition has been proposed as a method for distinguishing mitochondrial CPT activity from peroxisomal COT activity (70). Aminocamitine inhibition of microsomal COT activity can also be used to distinguish medium/long-chain camitine acyltransferase activity of microsomal origin from mitochondrial CPT and peroxisomal COT. The DL-aminocamitine inhibition of microsomal COT is similar, though, to the L-aminocamitine inhibition reported for the CPT activity of outer mitochondrial membrane enriched vesicles and the COT activity associated with intact peroxisomes (70); further research is needed to demonstrate if the medium/long—chain camitine acyltransferase activity located in the endoplasmic reticulum, outer mitochondrial membrane enriched vesicles, and intact peroxisomes is due to one enzyme or if it is due to multiple enzymes. 89 Chapter 4. Attempted Purification of Microsomal COT Rat liver endoplasmic reticulum contains a medium/long-chain camitine acyltransferase referred to herein as microsomal camitine octanoyltransferase (COT). Microsomal COT activity is tightly associated with the microsomal membrane and requires detergents or high pH conditions for solubilization. Greater than 90% of microsomal COT can be solubilized with the zwitterionic detergent, CHAPS. Microsomal COT can also be solubilized at pH>10.5. Purification of CHAPS solubilized COT using column chromatography was not successful Gel filtration, anion exchange, hydmphobic interaction, and dye affinity chromatography were used. CHAPS solubilized COT eluted from an anion exchange column with a lower specific activity than the initial sample. Reconstitution of the low activity COT eluted from the anion exchange column with phospholipids did not lead to recovery of activity. 90 Introduction Rat liver contains at least three camitine acyltransferases with medium/long-chain acyl- group specificity: mitochondrial camitine palmitoyltransferase (CPT), peroxisomal camitine octanoyltransferase (COT), and microsomal camitine octanoyltransferase (COT) (19,41). The role of mitochondrial CPT in the transport of long-chain acyl-CoA through the inner membrane of mitochondria for subsequent B-oxidation is well established (19,85). It has been proposed that peroxisomal COT is involved in the transport of peroxisomal B-oxidation chain-shortened acyl-CoA from the peroxisome to the mitochondria (36). The function of microsomal COT is not known. Mitochondrial CPT and peroxisomal COT in rat liver have been purified to homogeneity (34). Microsomal COT is tightly associated with the outer surface of the microsomal membrane (42). Characterization of microsomal COT has been hindered by the instability of the detergent solubilized enzyme (37,39), with the result that the enzyme has not yet been purified. Mitochondrial CPT is also membrane bound and is located both outside (CPTO) and inside (CPTi) the inner mitochondrial membrane (19). It is not known if CPTo and CPTi are different proteins or if they are the same protein with different locations and regulatory properties. Only one protein with medium/long-chain camitine acyltransferase activity from liver mitochondria has been purified to homogeneity (19). The purified liver mitochondrial CPT is oligomeric with a native molecular weight of 280,000 - 320,000 estimated by gel filtration chromatography and a subunit molecular weight of ~70,000 estimated by SDS-PAGE 91 92 (34). Purified CPT exhibits complex allosteric kinetics with c00perative binding of acyl-CoA and camitine (5,34). Although it is thought that purified CPT and CPTi are the same protein, it is not known if CPTo and purified CPT are the same protein. Unlike purified CPT, CPTo is inhibitable by malonyl-CoA (19,30). It has been reported that CPTo is a different protein than either purified CPT or CPTi and that it is destroyed by detergent solubilization (22-24). Peroxisomal COT is soluble and is located in the matrix of peroxisomes (39). Peroxisomal COT has been purified to homogeneity from rat liver with a subunit molecular weight of 66,000 daltons estimated by SDS-PAGE (34); peroxisomal COT purified from mouse liver has a native and subunit molecular weight of 60,000 daltons estimated by gel filtration chromatography and SDS-PAGE (14). Peroxisomal COT is not inhibited by malonyl- CoA (34). The data presented herein describe an attempted purification of microsomal COT. The characteristics of purified microsomal COT could be compared to the known properties of purified mitochondrial CPT 'or purified peroxisomal COT. Although microsomal COT was not purified to homogeneity, the attempted purification steps can be. used to gain insight into the solubility characteristics and stability of the solubilized enzyme. Column chromatography and ammonium sulfate fractionation purification attempts of solubilized microsomal COT are reported. Results Preparation of Microsomes Microsomes were prepared by differential centrifugation of a rat liver homogenate in the presence of the protease inhibitor PMSF. The average yield of microsomes was 10.1 i 1.5 mg l g wet weight of liver (n=5); The specific activity of COT was ~6 nmole rnin'l mg’1 protein (see Table 2 of Chapter 2) giving a yield of microsomal COT activity of ~60 munit / g wet weight of liver. An attempt was made to scale-up the preparation of microsomes by using CaCl2 precipitation of rat livers which were stored frozen at -70°C. The rat livers were collected and stored in 0.25 M Mannitol, 1 mM EDTA, pH 7.4, and 25 [lg/ml PMSF. A large scale preparation of microsomes was done using CaClz precipitation of the post-mitochondrial supernatant obtained from the frozen livers. An unusually high yield of COT activity (383 munits COT I g wet weight of liver) was obtained. The specific activity of COT was 20.7 nmole rnin'1 mg’1 protein which was also high compared to a typical specific activity of ~5 nmole rnin'l mg'1 protein obtained from fresh rat livers (see Table 2 of Chapter 2). Marker enzymes were assayed and the specific activity of the microsomal marker enzyme, NADPH cytochrome c reductase was 17.6 nmole min'1 mg'1 protein which was low compared to a typical specific activity of ~50 nmole min'l mg1 protein from fresh rat liver (see Table 1 of Chapter 2). The specific activity of the marker for the outer mitochondrial membrane, monoamine oxidase, was 6.3 nmole rrlin’1 mg‘1 protein which is higher than the typical value 93 94 of ~0.7 nmole min‘1 mg'1 protein from fresh rat livers (see Table 1 of Chapter 2). CaCl2 precipitated microsomes obtained from frozen rat livers contain high amounts of monoamine oxidase indicating they are significantly contaminated by mitochondria. Solubilization Microsomal COT is firmly membrane bound and is not solubilized from the membrane by freeze/thawing or sonication (37,39). Microsomal COT can be solubilized from the microsomal membrane using either detergent or high pH conditions. The effect of different detergents on cor activity is shown in Table 3 of Chapter 2. The optimum conditions for detergent solubilization were to make the microsomal membranes 8 mM in CHAPS and incubating on ice for one hour followed by freezing at -70°C for at least 12 hours. The microsomes containing CHAPS were thawed at room temperature and centrifuged at 100,000 x g for 30—60 minutes at 5°C. For a representative experiment, the specific activity of the membrane bound microsomal COT was 7.9 :1: 1.1 nmole min" mg" protein while the specific activity of the CHAPS-solubilized supernatant was 4.7 :l: 0.6 nmole min'1 mg'1 protein (n=2). CHAPS-solubilized COT is stable both at 4°C and to freezing at -20°C. Microsomal COT can also be solubilized from the membrane using high pH conditions. The optimum conditions for high pH solubilization were to make the microsomes pH 10.5 with the addition of NH4OH followed by three cycles of freezing in a dry ice/acetone bath and thawing at room temperature. The frozen-thawed microsomes were made 1 mM in dithiothreitol, flushed with N2, and sonicated 4 times at power setting 4 for 30 seconds using a Model W-220F Heat Systems - Ultrasonics, Inc. sonicator. The sonicated microsomes were centrifuged at 100,000 x g for 30-60 minutes at 5°C. For a representative experiment the 9S extent of solubilization was 78%; of an initial 141.6 munits of membrane bound microsomal COT there were 110.2 munits of COT in the 100,000 x g supernatant fluids and there were 25.3 munits of COT in the 100,000 x g pellet. For this experiment the specific activity of the high pH solubilized COT was 14.1 nmole min'1 mg‘1 protein. The high pH solubilized microsomal proteins re-aggregated when the pH was lowered and were pelleted when the sample was centrifuged at 100,000 x g. The effect of pH on the solubility of the high pH solubilized microsomes is shown ill Table 1. The solubilized microsomes were adjusted to the pH values indicated and incubated at 4°C for 12 hours. The samples were again centrifuged at 100,000 x g and the COT activity in the supernatant fluid and pellet was determined. COT activity remained soluble after the 12 hour incubation at pH 9.7, and pH 8.5, while 81% of the COT activity had re-aggregated at pH 6.6. Microsomal COT can also be solubilized from the microsomal membrane by a combination of high pH and detergent solubilization techniques. CHAPS was included during the high pH solubilization to optimize the solubility of the high pH solubilized enzyme. Optimum conditions were to make the microsomes 2 mM in CHAPS and pH 10.5 with the addition of NH4OH and to incubate on ice for at least 1 hour. The high pH - CHAPS microsomes were subjected to three cycles of freezing in a dry ice/ acetone bath and thawing at room temperature followed by centrifugation at 100,000 x g for 30-60 minutes. Sonication was not used. For a representative experiment the specific aetivity of the membrane bound microsomal COT was 11.4 nmole min1 mg‘1 protein while the specific activity of the high pH - CHAPS solubilized supernatant was 8.4 nmole min'1 mg’1 protein. 96 Table 1. Effect of pH on the Solubility of Microsomal COT pH of extracted COT 9.7 8.5 6.6 Activity of extracted 10.72 15.32 16.84 COT nmole min“1 The extracted COT was incubated at 4°C for 12 horns then centrifuged at 100,000 x g. Activity of Soluble COT 11.48 14.56 2.28 (100,000 x g Supernatant) nmole rrlin'l Activity of Re-aggregated 0.76 2.3 9.94 COT (100,000 x g Pellet) nmole min’l 97 Ammonium Sulfate antionation Ammonium sulfate fractionation of microsomal COT solubilized in 1% Tween 20 is shown in Figure 1. The optimum concentration for ammonium sulfate fractionation was 40%. The 40% ammonium sulfate precipitated protein does not pellet with centrifugation at 10,000 x g. The 40% ammonium sulfate supernatant and pellet were separated by filtering through a column of glass wool and the precipitated protein was eluted with 5 mM potassium phosphate, pH 7.5, 1 mM EDTA containing 0.5% Tween 20. The COT activity of the resuspended 40% ammonium sulfate precipitate was not soluble and pelleted with centrifugation at 100,000 x g. Conditions were not found to solubilize stable COT activity from the resuspended 40% ammonium sulfate precipitate. Column Chromatogr_aphy A. Purification of high pH-CHAPS solubilized microsomal COT was attempted using gel filtration chromatography. Analytical scale HPLC gel filtration chromatography using a Dupont G-250 sieving column showed that solubilized microsomal COT eluted with a molecular weight ~60,000. The column was equilibrated in 200 mM ammonium acetate pH 8.0 containing 20% glycerol. The solubilized microsomal COT eluted with a retention time of 11.1 min. while the retention time of bovine serum albumin which has a molecular weight of 66,000 daltons was 10.3 min. The recovery of COT eluted from the column was 3.8%; 6.9 munits were loaded and 0.26 munits were recovered. B. Purification of microsomal COT using gel filtration was scaled up for use as a preparative chromatographic step. A 2.5 cm x 80 cm Bio gel P100 column was prepared and equilibrated in 50 mM potassium phosphate, pH 7.5 with 20% glycerol. The column was calibrated with the following molecular weight standards: Ribonuclease A, Chymotrypsinogen 98 Figure l. Ammonium Sulfate Fractionation of Detergent Solubilized Microsomal COT. Aliquots of 1% Tween-20 solubilized COT containing 100 munits of COT activity were made the percentage of ammonium sulfate indicated and the supernatant and pellet were collected and assayed spectrally for COT as described in Materials and Methods. 99 Figure 1. metnm rcchrCEat c...“ E09. 34.0on At 0.0% .5 t. t t C om. EEOCCCQDm -_. 0.0% cu 0.0? siluntu 10:) 100 A, Ovalbumin and, Bovine Serum Albumin; Blue Dextran was used to determine the void volume. High pH-CHAPS solubilized microsomal COT eluted from the column in the void volume with a 10.4% recovery of initial activity indicating the solubilized COT had re- aggregated during the chromatography. The P100 column was equilibrated in 50 mM Tris-Cl pH 8.5, 1 mM EDTA, 4 mM CHAPS, and 1 mM DTT‘ to attempt to optimize conditions for COT stability and solubility. The high pH-CHAPS solubilized COT again eluted with the void volume with a 37% recovery again indicating the COT had re-aggregated. Conditions were not found to allow preparative scale gel filtration chromatography of the solubilized enzyme. C. An HPLC weak anion exchange separation of the high pH - CHAPS solubilized microsomal COT was attempted. A Synchropak AX 300 column was equilibrated in 10 mM potassium phosphate pH 7.5, 2 mM CHAPS containing 20% glycerol (buffer A). The solubilized COT was loaded and the column washed with buffer A until the A280 was ~0. The COT was eluted with a linear gradient of buffer A containing 1 M KCl. A representative column profile is shown in Figure 2. For this experiment 17.2 munits of COT with a specific activity of 10.8 nmole min" mg" protein were loaded onto the column The recovery of COT eluted from the column was 2.7 munits (15.7%); the specific activity of the COT eluted from the column was 7.05 nmole min'1 mg‘1 protein. COT eluted from the column was soluble and did not pellet with centrifugation at 100,000 x g. Fractions containing COT activity were separated on a 10% SDS polyacrylamide gel and silver stained. SDS- PAGE showed the presence of at least 27 polypeptides with a molecular weight range of ~24,000 - 100,000 daltons. N 0 conditions were found using anion exchange chromatography which resulted in an increase in the specific activity of microsomal COT. W‘ I ”Lil‘- ‘» a ., it -. i, ». . 3- ~. 9 2...”... a... 101 Figure 2. HPLC Weak Anion Exchange Separation of High pH - CHAPS Solubilized Microsomal COT. A Synchropak AX 300 column was equilibrated in 10 mM potassium phosphate pH, 7.5, 2 mM CHAPS, and 20% glycerol (buffer A) at a flow rate of 0.5 ml min". COT (17.2 munits) was loaded isocratically. The column washed with buffer A until the A280 was ~0 and proteins eluted with a linear gradient of buffer A containing 1 M potassium chloride beginning at 18 minutes as indicated. One ml fractions were collected and assayed ‘ spectrally for COT activity as indicated. N 102 “m" Pugul aloulu 10:) t0 Figure 2. .56 06:. cozcflcm cm on O¢ Om ON 0. oqll. o‘clo'c'.Tq'dl'o old!‘ '0'. C'Ol. O t A \ \ \ m \. memo \ o' . __._—O .- .—.-.-.- 6x 5:10 \ I""""‘\ 1 N66 082v 103 D. Additional column chromatography techniques which were attempted included: dye affinity, hydroxylapatite, and hydrophobic interaction column separations. Cibacron blue sepharose and orange sepharose columns were used but the yield of COT activity was low. Hydroxylapatite chromatography in one experiment gave a 17.6% recovery of COT activity. For this experiment a Biogel-HTP column was used and of the 930 munits of COT loaded on the column, 164 munits of COT were eluted with a linear potassium phosphate gradient. COT activity was not stable during hydr0phobic interaction chromatography; phenyl-sepharose and octyl-sepharose columns were used. Prism—ration of COT Eluted from an_Anion Exchallge Column Restoration of microsomal COT eluted from the anion exchange column was attempted to increase specific activity. The specific activity of the microsomal COT eluted from the column was 7.05 nmole min'1 mg'1 protein. Addition of 0.2% Tween-20 to the assay mix did not change the specific activity of the COT. Chloroform/methanol extracted microsomal phospholipids added to the assay also did not change the specific activity. Addition of 0.001% asolectin to the assay gave a specific activity of 7.7 nmole min’1 mg’1 protein. A neutralized acid extract of microsomes was also added to the assay and the specific activity of the COT was decreased to 2.8 nmole min" mg" protein. Addition of 1 mM ATP and 1 mM MgCl2 to the assay also decreased the specific activity of COT to 5.6 nmole min" mg" protein. No conditions were found that increased the specific activity of the COT containing fractions eluted from the anion exchange column. Discussion The purification of microsomal COT was attempted from a microsomal membrane pellet prepared by differential centrifugation of a post mitochondrial supernatant obtained from fresh rat livers. The preparation of microsomes using CaC12 precipitation of a post mitochondrial supernatant obtained from livers which were stored frozen at -70°C resulted in a microsomal fraction that contained high amounts of monoamine oxidase activity. Monoamine oxidase is a marker enzyme for the outer mitochondrial membrane indicating the fraction was significantly contaminated by the outer membrane of mitochondria. Since there have been reports that in liver mitochondrial medium/long-chain camitine acyltransferase activity is located in the outer membrane (1 129-32), these microsomal preparations were not used for further purification. Microsomal COT is tightly associated with the microsomal membrane and is not solubilized by freeze-thaw treatment of microsomes (39). Microsomal COT can be solubilized by treatment of the microsomes with 1% Triton X-100 containing 0.4 M potassium chloride but the solubilized COT activity was not stable (39). The data reported herein show that microsomal COT can be solubilized using detergent or high pH conditions; 20% glycerol was present during solubilization. Osmolytes such as glycerol have been shown to act as protein stabilizers during detergent solubilization (93). Optimum detergent solubilization is achieved using the zwitterionic detergent, CHAPS. CHAPS - solubilized microsomal COT activity is stable but purification attempts of detergent solubilized COT activity were not successful. 104 105 Microsomal COT is also solubilized from the membrane using high - pH conditions (pH 10.5) although the enzyme re-aggregates when the pH is lowered (see Table l). A recent report in abstract form describes solubilization of the three most hydrophobic polypeptides of the b6f complex of b cytochromes using pH > 10.5 and proposed that electrostatic forces could be important is stabilizing this complex in the membrane (94). This suggests that microsomal COT is solubilized from the membrane at pH 10.5 due to disruption of electrostatic forces. A combination of high pH and detergent conditions also solubilized microsomal COT from the membrane although the enzyme still re-aggregated when the pH was lowered for gel filtration chromatography. In contrast, the high pH - CHAPS solubilized microsomal COT eluted from an anion exchange column was soluble (see Figure 2). Microsomal COT eluted from the anion exchange column with a lower specific activity than the initial high - pH solubilized sample (compare 7.05 nmole min”1 ing'l protein with 10.8 nmole min'1 mg'1 protein). Restoration of COT activity eluted from the anion exchange column using asolectin or microsomal phospholipids was not successful. The purification attempts of microsomal COT reported herein were not successful in purifying the enzyme to homogeneity. One explanation of the difficulty of solubilizing microsomal COT from the membrane and loss of COT activity with anion exchange chromatography is that microsomal COT may require association with another protein for maximum activity and stability. Future purification of microsomal COT could include attempts to restore the low specific activity COT fractions eluted from the anion exchange column with other protein containing fraction separated by the column to yield a high activity, stable enzyme complex. Another purification attempt could focus on restoring the malonyl-COA sensitivity of 106 solubilized microsomal COT. Membrane bound microsomal COT is inhibited by malonyl-COA while detergent solubilized microsomal COT is not inhibited by up to 200 ltM malonyl-COA (see Figure 3 of Chapter 2). Membrane bound microsomal COT may be sensitive to inhibition by malonyl-COA through association with a putative malonyl-COA binding regulatory protein. Such a putative regulatory protein has been proposed to be involved in the malonyl-COA inhibition of mitochondrial CPT (74) although direct evidence is lacking (19). If the putative malonyl—COA binding regulatory protein is purified from mitochondria it could be used to try and reconstitute the fractions with low COT activity eluted from the anion exchange column to yield a high activity, malonyl—COA sensitive enzyme. Chapter 5. Summary and Conclusions Rat liver contains at least three camitine acyltransferases with medium-chain to long- chain acyl-group specificity located in mitochondria, peroxisomes, and endoplasmic reticulum. Medium/long-chain camitine acyltransferase activity located in mitochondria with access to cytosolic acyl-CoA’s (mitochondrial CPTO) and medium/long-chain camitine acyltransferase activity located in the endoplasmic reticulum are regulated through inhibition by malonyl-COA. Medium/long-chain camitine acyltransferase activity associated with the matrix of peroxisomes is not inhibitable by malonyl-COA. Malonyl-COA,‘ an intermediate in fatty acid synthesis, acts to inhibit fatty acid oxidation via inhibition of cytosolic medium/long-chain acylcamitine production preventing a futile cycle of synthesis and oxidation. Table 1 summarizes a comparison of previously reported characteristics of medium/long-chain camitine acyltransferase activity located in mitochondria and peroxisomes with data given in this thesis characterizing medium/long-chain camitine acyltransferase activity located in the endoplasmic reticulum. The data show that microsomal COT has different characteristics than either mitochondrial CPT or peroxisomal COT although there are also similarities between these enzymes. Microsomal COT, peroxisomal COT and mitochondrial CPT all transfer medium- chain and long-chain acyl-groups between acyl-COA and L—camitine. Microsomal COT is antigenically different than either mitochondrial CPT or peroxisomal COT. Microsomal COT 107 108 BEE see 852% Eamuez eases 0:55:82 21 : u a: 5:52 05.5802 02 02 65568.4 55 <00 -585 :53 H 1 : Sm c852 £2222 as .8555 52 a: 2: 8m 5 8.2555 52 9.: a; 65 oz $3 530.3% 53.5%: 388 ES... mg; n a :5 $8 58025562052 as 8555 so: Eu eo§52om as 8.555 sec mac :5 $3 3 J etc as 21 85 5 8.5555 see 950 >5: 21 5 A 5 8555 Lao >20 9.: oz 6: a; Eu e5 0.50 $9 055.53 55 £6.52. a? 325 28553 358880 anemone £3an0 85555 $0-385: 85555 60.5255 9004580258555. .3 Busmamooaocsfifim Suzaeeeeeoegé . .3 eBfiEEBaocsEEH 855588550 omcomccaa 855808.20 0505M 55 mm 5 beats... 2.0 “28 so 5883 5.55: E 5283 325 a» oz a; $6 + etc 285 ceases: eSO 558222 you 585655 «etc .35 5522852 FOO EEOmOLUmE Ufla .FOU EEOmmXOLQQ .FQU Etta—0:063: 50>: an“ uO wflthOhn— up: ac hLNEEDM .H 3an 109 8:0 21 Na n on: 5:5 10.00.3555: .3 03355:: 00 00800580: 080:: 55 0080008 .0358 EU 0:000 085502 0 02050:, 8:058 085805 0:5 :0 .9568 .EU a 00:050., 005050 08.5808 :030 00 .3568 .EU .. a. 0:0 .m .N 500850 5 Sim 880 008065 005080 8055 0 05205005: 00 x588 0...: 55> 83:00 E .50 00 9:0": a 600:8 05 555 88:00 5 .EU 0.0 9:00 0.65:8 (00-30082 _ 05:88:??49 2: 00m u a: 0:000 085602 51 E u s: 008200 21 :0 .t. a: 0:000 085802 as 25 N B B555 52 $3 8:025 00.9 005:: .21 g u a: at 513 u a: .50 8:58 a: 21 2 u a: .050 2.: as 21 02 u a: Leo >20 55 e a: oaeeooe .50 05.5 52: 8:5: as 2a m u a: 0.50 8055.: 655885555 8:55.: 80-583520 «HOD 88000552 .80 55852.0: .50 .380 55228:: 085.80 .5 2.5:. 110 like mitochondrial CPT is membrane bound and unlike peroxisomal COT which is soluble. Microsomal COT like peroxisomal COT shows Michaelis-Menten kinetics while mitochondrial CPT shows complex allosteric kinetics. Membrane bound microsomal COT like membrane bound mitochondrial CPT is inhibited by malonyl-CoA while peroxisomal COT is not inhibited by malonyl-CoA. Microsomal COT, both membrane bound and soluble, and peroxisomal COT are inhibited by micromolar concentrations of etomoxiryl-CoA while membrane bound mitochondrial CPI‘ is inhibited by nanamolar concentrations of etomoxiryl-CoA. Microsomal COT is less sensitive to aminocamitine inhibition than mitochondrial CPT while peroxisomal COT is not inhibited by aminocamitine. Microsomal COT has similar properties to the CPT activity of outer mitochondrial membrane enriched vesicles (OMV CPT) and the CPT activity associated with intact peroxisomes. These properties include: inhibition by malonyl-CoA and aminocamitine. Rat liver mitochondria prepared by differential centrifugation can be significantly contaminated by both peroxisomes and mitochondria (35,58,59,61). Further studies are needed to clarify the relationship between microsomal COT, OMV CPT and the CPT associated with intact peroxisomes to determine if they are due to a single enzyme activity or to separate enzymes with similar properties. Possible Function_s:of Microsomal COT Although the function of the malonyl-CoA sensitive COT of microsomes is not established, the strong inhibition by low amounts of malonyl-CoA indicates it is subject to short-term metabolite regulation in a manner that would reduce or prevent long-chain and medium-chain acylcamitine formation in the fed state. The condensing enzyme(s) of rat liver 111 involved in microsomal fatty acid elongation are located on the cytosolic surface of the endoplasmic reticulum (97). This enzyme can catalyze the condensation of palrnitoyl-CoA with malonyl-CoA, the initial step in the microsomal chain elongation system. The strong inhibition of microsomal COT by palmitoyl-CoA and malonyl-CoA suggests that microsomal acylcamitine formation is inhibited under metabolic conditions that promote fatty acid elongation. CPT, purified from beef heart mitochondria exhibits a log relationship between the acyl-CoA chain length and the K05 for camitine, indicating that at physiological, non- saturating concentrations of L-camitine, it has the capacity to kinetically select for long-chain acyl—CoA derivatives (20). Although it is well-established that medium-chain fatty acids can be activated in the mitochondrial matrix, the fraction of medium-chain fatty acids activated in the cytosolic compartment compared to the matrix compartment in vivo is not known. Due to the acyl-CoA impermeable barrier, camitine is required for the mitochondrial B—oxidation of cytosolic medium-chain acyl-CoAs; thus, regulation of their conversion to acylcamitines in the cytosolic compartment should be expected. Microsomal COT can convert cytosolic medium-chain acyl-CoAs to acylcamitines which subsequently enter mitochondria for [3- oxidation. This could permit B—oxidation of both cytosolic long-chain acyl-CoAs and medium- chain acyl-CoAs when mixtures of cytosolic acyl-CoAs are present, especially in the fasted state. Alternatively, medium-chain camitine acyltransferase activity in the microsomes may function in the detoxification of acyl residues as camitine conjugates that can be eliminated in the urine of humans. Valproic acid therapy and pivampicillin therapy cause the excretion of both valproylcarnitine and pivaloylcamitine, respectively (98,99). Similarly, several human 112 disease states promote urinary excretion of specific acylcamitines (100,101). The camitine acyltransferase(s) responsible for the formation of these acylcamitines has not been determined. Since many detoxification systems are located in the endoplasmic reticulum of liver, this may prove to be a function of the microsomal COT. Future Research It is important to establish if microsomal COT is a separate enzyme than either mitochondrial CPT or peroxisomal COT. The purification of microsomal COT to homogeneity could allow studies to be done to determine the relationship between these enzymes. Mitochondrial CPT (24) and peroxisomal COT (102) have been cloned and the cDN A sequenced to determine the deduced amino acid sequence. N-Terminal amino acid sequence obtained from the purified enzyme or deduced amino acid sequence obtained from a cDNA for microsomal COT could be used to unequivocally establish is microsomal COT is a separate enzyme. The mechanism of malonyl-CoA inhibition of medium/long-chain camitine acyltransferase activity is not known (19). It is important to determine if the camitine acyltransferase itself binds malonyl-CoA or if the camitine acyluansferase associates with a regulatory protein which binds malonyl-CoA. The purification of microsomal COT could allow studies to be done to determine the mechanism of malonyl-CoA inhibition of microsomal COT. For example it could be possible to purify malonyl-CoA sensitive microsomal COT and determine if it is a single protein or a complex of proteins. Reconstitution of the purified microsomal COT protein(s) into phospholipid vesicles may be required since microsomal COT is malonyl-CoA sensitive when it is membrane bound. It has been shown that the sensitivity 113 of solubilized mitochondrial CPTo to malonyl—CoA inhibition is enhanced by reconstitution into asolectin liposomes (29). 10. '11. 12. 13. 14. 15. 16. List of References Albro, P.W., Corbett, J.T., and Schroeder, 1.1.. (1987) Lipids 22 (10), 751-756. Dallner, G. (1974) Methods in Enz. 31, 196-198. Radin, NS. (1969) Methods in Enz. 14, 245-254. Bieber, 1.1., Abraham, r., and Helrnrath, r. (1972) Anal. Biochem. 50, 509-518. Fiol, C.J., Kemer, J., and Bieber, LL. (1987) Biochim. Biophys. Acta 916, 482-492. Lysiak, W., Toth, P.P., Suelter, C.H., and Bieber, LL. (1986) J. Biol. Chem. 261, 13698-13703. Aronson, N.N.,Jr., and Touster, O. (1974) Methods in Enz. 31, 90-102. Bergmeyer, H.U. (1974) Methods of Enzymatic Analysis (2nd Ed.) Vol. 1, p. 518, Academic Press, Inc. New York. Yonetani, T. (1967) Methods in Enz. 10, 332-335. Masters, B.S.S., Williams, C.H.,Jr., and Kamin, H. (1967) Methods in Enz. 10, 565-573. ' Murthy, M.S.R. and Pande, S.V. (1987) Proc. Natl. Acad. Sci. USA 84, 378-382. Catravas, G.N., and McHale, CG. (1974) Radiat. Res. 58, 462-469. Markwell, M.A.K., Haas, S.M., Tolbert, N.E., and Bieber, LL. (1981) Methods in Enz. 72D, 296-303. Farrell, S.O., F101, C.J., Reddy, J.K., and Bieber, LL. (1984) J. Biol. Chem. 259, 13089-13095. Farrell, 8.0. and Bieber, LL. (1983) Arch. Biochem. Biophys. 222, 123-132. Kemer, 1., and Bieber, L. (1990) Biochemistry 29, 4326-4334. 114 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 29. 30. 31. 32. 33. 34. 35. 115 Laemmli, UK (1970) Nature 227, 680-685. Taylor, D.C., Weber, N., Hogge, L.R., and Underhill, E.W. (1990) Anal. Biochem. 184, 311-316. Bieber, LL. (1988) Ann. Rev. Biochem. 57, 261-283. Fiol, CJ. and Bieber, LL. (1984) J. Biol. Chem. 259, 13084—13088. Pande, S.V. (1975) Proc. Nat. Acad. Sci. USA 72, 883-887. Declercq, P.B., Falck, J.R., Kuwajima, M., Tyminski, 11., Foster, D.W., and McGarry, JD. (1987) J.Biol.Chem. 262, 9812-9821. Woeltje, K.F., Esser, V., Weis, B.C., Cox, W.F., Schroeder, J .G., Lio, S.-T., Foster, D.W., and McGarry, JD. (1990) J. Biol. Chem. 265, 10714-10719. Woeltje, K.F., Esser, V., Weis, B.C., Sen, A., Cox, W.F., McPhaul, M.J., Slaughter, C.A., Foster, D.W., and McGarry, JD. (1990) J. Biol. Chem. 265, 10720-10725. Clarke, P.R.H. and Bieber, LL. (1981) J.-Biol. Chem. 256, 9861-9868. Bergstrom, J .D. and Reitz, RC. (1980) Arch. Biochem. Biophys. 204, 71-79. Hoppel, CL. and Tomec, RJ. (1972) J. Biol. Chem. 247, 832-841. Brosnan, J.T., Kopec, B., and Fritz, LB. (1973) J. Biol. Chem. 248, 4075-4082. Murthy, M.S.R., and Pande, S.V. (1987) Biochem. J. 248, 727-733. McGarry, J.D., Woeltje, K.F., Kuwajima, M., and Foster, D.W. (1989) Diabetes/Metabolism Rev. 5, 271-284. Kolodziej, MP. and Zammit, V.A. (1990) Biochem. J. 267, 85-90. Cook, G.A., Khan, B., and Heimberg, M. (1988) Biochem. Biophys. Res. Commun. 150, 1077-1082. Bieber, L.L., Krahling, J.B., Clarke, P.R.H., Valkner, K.J., and Tolbert, NH. (1981) Arch. Biochem. Biophys. 211, 599-604. Miyazawa, S., Ozasa, H., Osumi, T., and Hashirnoto, T. (1983) J. Biochem. 94, 529-542. Ramsay, RR. (1988) Biochem. J. 249, 239-245. 36. 37. 38. 39. 40. 41. 42. 43. 45. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 116 Bieber, L.L.,Emaus, R., Valkner, K., and Fanell, S. (1982) Fed. Proc. 41, 2858- 2862. Markwell, M.A.K and Bieber, LL. (1976) Arch. Biochem. Biophys. 172, 502-509. Bieber, LL. and Farrell, S. (1983) The Enzymes 5, 627-644. Markwell, M.A.K, Tolbert, N.E., and Bieber, LL. (1976) Arch. Biochem. Biophys. 176, 479-488. Derrick, J.P. and Ramsay, RR. (1989) Biochem. J. 262, 801-806. Markwell, M.A.K., McGroarty, E.J., Bieber, LL. and Tolbert, NE. (1973) J. Biol. Chem. 248, 3426-3432. Valkner, K1. and Bieber, LL. (1982) Biochim. Biophys. Acta 689, 73-79. DePierre, J.W. and Emster, L. (1977) Ann. Rev. Biochem. 46, 201-262. Bailey, DJ., Murray, R.K., and Rolleston, ES. (1974) Can. J. Biochem. 52, 1003- 1012. Srere, P.A., Suebert, W., and Lynen, F. (1959) Biochim. Biophys. Acta 33, 313- 319. Norum, KR. (1964) Biochim. Biophys. Acta 89, 95-108. Grassetti, DR. and Murray Jr., J.F. (1967) Arch. Biochem. Biphys. 119, 41-49. Crabtree, B. and Newsholrne, EA. (1972) Biochem. J. 130, 697-705. Bremer, J. (1967) J. Biol. Chem. 242, 1744—1748. Bremer, J. (1981) Biochim. Biophys. Acta 665, 628-631. Van To], A. and Hulsman, WC. (1969) Biochim. Biophys. Acta 189, 342353. Bremer, J. (1962) J. Biol. Chem. 237, 2228-2231. Norum, K.R. and Bremer, J. (1967) J. Biol. Chem. 242, 407-411. Berge, R.K. and Farstad, M. (1979) Eur. J. Biochem. 95, 89-97. Dixon, A., Osterloh, J., and Becker, C. (1990) J. of Pharm. Sci. 79, 103-105. Van To], A. (1974) Biochim. Biophys. Acta 357, 14-23. 57. 58. 59. 60. 61. 62. 63. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 117 Kahonen, M.T. (1976) Biochim. Biophys. Acta 428, 690-701. Hovius, R., Lambrechets, H., Nicolay, K., and de Kruiff, B. (1990) Biochim. Bi0phys. Acta 1021, 217-226. Healy, M.J., Kemer, J., and Bieber, LL. (1988) Biochem. J. 249, 231-237. Pickett, CB., Rosenstein, N.R., Jeter, R.L., Montisano, D., and Cascarono, J. in Biochemistry, Biophysics and Regulation of Cytochorme P-450 (1.}. Gustafsson gt a_l_ eds). PP 441-446, Elsevier/North-Holland Biomedical Press (1980). Vance, J.E. (1990) J. Biol. Chem. 265, 7248-7256. Fogle, RI. and Bieber, LL. (1978) Int. J. Biochem. 9, 761-765. Markwell, M.A.K., Bieber, L.L., and Tolbert, NE. (1977) Biochem. Pharmacology 26, 1697-1702. Ramsay, RR. (1990) FASEB J. 4,A803, #3108. Grantham, BD. and Zammit, V.A. (1988).Biochem. J. 249, 409-414. Saggerson, E.D., Carpenter, C.A., and Tselentis, BS. (1982) Biochem. J. 208, 667- 672. Jenldns, D.L. and Griffith, O.W. (1985) J. Biol. Chem. 260, 14748-14755. Jenkins, D.L. and Griffith, O.W. (1986) Proc. Natl. Acad. Sci USA 83, 290-294. Kanamaru, T. and Okazaki, H. in Novel Microbial Products for Medicine and Agriculture (Demain,A.L. e_t al_ eds), pp 135-144, Elsevier/North Holland Biomedical Press (1989). Murthy, M.S.R., Ramsay, RR, and Pande, S.V. (1990) Biochem. J. 267, 273-276. McGarry, J.D., Leatherman, GR, and Foster, D.W. (1978) J. Biol. Chem. 253, 4128-4136. McGarry, JD. and Foster, D.W. (1979) J. BioL Chem. 254, 8163-8168. Woeltje, K.F., Kuwajima, M., Foster, D.W., and McGarry, JD. (1987) J. Biol. Chem. 262, 9822-9827. Bergseth, S., Lund, H., and Bremer, J. (1986) Trans. Biochem. Soc. 14, 671-672. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 91. 92. 118 Zammit, V.A., Corstorphine, C.G., and Kelliher, MG. (1988) Biochem. J. 250, 415- 420. Zammit, V.A., Corstorphine, C.G., and Kolodziej, MP. (1989) Biochem. J. 263, 89- 95. McGarry, JD. and Foster, D.W. (1981) Biochem. J. 200, 217-223. McGarry, J.D., Mills, S.E., Long, C.S., and Foster, D.W. (1983) Biochem J. 214, 21-28. Kiorpes, T.C., Hoerr, D., Ho, W., Weaner, LE, Inman, M.G., and Tutwiler, GP. (1984) J. Biol. Chem. 259, 9750-9755. Declercq, P.E., Venincasa, M.D., Mills, S.E., Foster, D.W., and McGarry, JD. (1985) J. Biol. Chem. 260, 12516-12522. Tutwiler, G.F., Kirsh, T., Mohrbacher, M., and Ho, W. (1978) Metabolism 27, 1539-1556. Eistetter, K. and Wolf, H.P.O. (1986) Drugs of the Future 11, 1034-1036. Lopaschuk, G.D., Wall, S.R., Olley, P.M., and Davies, NJ. (1988) Circ. Res. 63, 1036-1043. Lopaschuk, G.D., McNeil, G.F., and McVeigh, J .J. (1989) Molec. Cell. Biochem. 88, 175-179. Bremer, J. (1983) Physiol. Rev. 63, 1420—1480. Tutwiler, G.F., Brentzel, H.J., and Kiorpes, T.C. (1985) Proc. Soc. Exp. Biol. Med. 178, 288-296. Wolf, H.P.O. and Engel, D.W. (1985) Eur. J. Biochem. 146, 359-363. Murthy, M.S.R. and Pande, S.V. (1990) Biochem. J. 268, 599-604. Lilly, K., Bugaisky, G.E., Umeda, P.K., and Bieber, LL. (1990) Arch. Biochem. Biophys. 280, 167-174. Nusing,R. Thesis, School of Biology, University of Konstanz, July 1985. Brady, P.S. Dunker, A.K., and Brady, L]. (1987) Biochem. J. 241, 751-757. Fiol, CI. and Bieber, LL. (1988) Lipids 23, 120-125. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 119 Maloney, RC. and Arnbudkar, S.V. (1989) Arch. Biochem. Biophys. 269, 1-10. Cramer, W.A., Furbacher, P.N., Szczepaniak, A, and Tae, G.-S. (1990) Abstract #6, at "Control of Charge Transfer in Cytochrome and Chlorophyll Complexes", Satellite Meeting of X Intemat. Congress of Biophysics, Montreal, Canada, August 4-7. Chung, C. (1990) Personal Communication. Lilly, K. (1990) Unpublished data. Osei, P., Suneja, S.K., Laguna, J.C., Nagi, M.N., Cook, L., Prasad, M.R., and Cinti, D.L. (1989) J. Biol. Chem. 264, 6844-6849. Vickers, 8., Duncan, C.A.H., White, S.D., Ramjit, H.G., Smith, J.L., et al. (1985) Xenobiotica 15, 453—458. Melegh, B., Kemer, J., and Bieber, LL. (1987) Biochem. Pharm. 36, 3405-3409. Roe, C.R., Millington, D.S., and Maltby, D.A. (1986) in Clinical Aspects of Human Camitine Deficiency (Borum,P.R., Ed.), pp. 97-107, Pergamon Press, New York. Schmidt-Sommerfeld, E., Penn, D., Kemer, J ., and Bieber, LL. (1989) Clin. Chirn. Acta 181, 231-238. Chatterjee, B., Song, C.S., Kim, J .M., and Roy, AK. (1988) Biochemistry 27, 9000-9006. HICHI 1111111111110