may ABSTRACT A GLYOXYLATE SPECIFIC AMINOTRANSFERASE IN RAT LIVER PEROXISOMES BY Berlin Hsieh Ho Subcellular organelles from rat liver, rat kidney, spinach leaves, and pig liver were separated by isopycnic centrifugation. The distribution of the organelles in the gradient was determined using the following marker enzymes: catalase for microbodies, cytochrome c oxidase for mitochon- dria, NADH-cytochrome c reductase for microsomes, and acid phosphatase for lysosomes. Microbodies were obtained at densities ranging from 1.24 to 1.26 g/cm3. A glyoxylate Specific aminotransferase was found to be located exclusively in rat liver and kidney peroxi- somal fractions, but it was not found in spinach leaf and pig liver peroxisomes. This aminotransferase irreversibly catalyzed the conversion of glyoxylate to glycine with several amino acids. The activity of this aminotransferase for a leucine-glyoxylate reaction was 181 nmoles x min.1 x mg_1 peroxisomal protein, and 134 nmoles x min-l x mg-1 protein for a phenylalanine-glyoxylate reaction. Since it Berlin Hsieh Ho was most active with L-leucine, it was called a leucine- glyoxylate aminotransferase. The D isomers of the amino acids were inactive but not inhibitory. Glyoxylate served as the best amino group acceptor using leucine or phenyl- alanine as the amino donor. The activity with glyoxylate was 20 times higher than with other keto acids such as pyruvate, hydroxypyruvate, oxalacetate, and ketoglutarate. Activities with leucine, phenylalanine, and histidine as amono donors all peaked together during gel filtration and DEAE cellulose chromatography of the enzyme. The molecular weight of the enzyme was estimated to be about 72,000 daltons using gel filtration chromatography with Sephadex G-150. At a leucine and phenylalanine concen- tration of 16.7mm, the Km for glyoxylate was 0.5mM and 0.67mM respectively. At a fixed glyoxylate concentration of 2.08mM, the Km was 2.5mM for leucine and 2.8mM for phenylalanine. Substrate inhibition was observed at over 4mM glyoxylate. At a concentration of 16mM glyoxylate 49% of the enzyme activity was inhibited; however, no substrate inhibition was observed for the amino group donors. This glyoxylate specific aminotransferase from isolated peroxisomes was stable at 50°C for at least one hour. Treatment of the peroxisomes with Triton x-100 increased the enzyme activity by 80%, perhaps from solu- bilization of the particles. The leucine-glyoxylate Berlin Hsieh Ho aminotransferase was inhibited about 90% by both lmM p— chloromercuribenzoate and lmM phenylhydrazine. N-Ethyl- maleimide and isonicotinic acid hydrazide also inhibited the activity 30 to 50% at lmM concentration. No inhibition +, KCN, phosphate or oxalate. The enzyme was observed by Cu+ did not show any requirement for exogenous pyridoxal, pyridoxamine phosPhate, or pyridoxal phosphate, however addition of pyridoxal phosphate partially prevented enzyme inactivation during storage at -20°C. After breaking up the peroxisomes by treating with 0.01 M pyrophosphate, the suspension was separated into soluble, membrane and dense core material by a sucrose gradient. Both leucine-glyoxylate and phenylalanine- glyoxylate aminotransferase activities followed catalase distribution from the soluble matrix of the peroxisomes. The enzyme catalyzing both leucine-glyoxylate and phenylalanine-glyoxylate was thought to be one enzyme for the following reasons. Activities with combinations of amino acids in excess glyoxylate were not additive. Similar pH Optimum (pH 8.4) and heat inactivation curves were obtained. The two activities could not be separated by Sephadex and DEAE cellulose chromatography procedures. Both activities had similar distribution in sucrose density gradients and also in the soluble matrix of the peroxisomes. Both activities were affected similarly by inhibitors. Berlin Hsieh Ho Some activity toward an alanine-glyoxylate reaction was thought to be due to a degree of nonspecificity of this enzyme. Activity toward L-histidine-glyoxylate amino- transferase could not be separated from the activities with leucine and phenylalanine-glyoxylate aminotransferase by gel filtration and ion exchange column chromatography. How- ever, the Optimal pH for L-histidine-glyoxylate aminotrans- ferase activity was 6.2. Leucine-glyoxylate aminotransferase activity devel- oped postnatally and increased with the age until the rats were 40 days old, after which the activity was maintained at a high level. Activity of the aminotransferase was stimulated in male rats by a hypolipidemic drug, Clofibrate. In the Clofibrate fed rats, the specific activity of this aminotransferase increased 135% in the peroxisomal fraction, and total activity increased 50 to 100%. In two preliminary experiments, leucine-glyoxylate aminotransferase activity increased in rat livers as the protein content in the diets was increased from 3.4% to 55%. Rats fed with the 55% casein diet showed about twice as much leucine-glyoxylate aminotransferase activity com- pared to the activity found in rats fed with the low (3.4%) casein diet. In two preliminary experiments on the effect of starvation of the rats on the aminotransferase activity, a 60% to 70% decrease in enzymatic activity was observed within the first to second day. A GLYOXYLATE SPECIFIC AMINOTRANSFERASE IN RAT LIVER PEROXISOMES BY Berlin Hsieh Ho A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Biochemistry 1974 ACKNOWLEDGMENTS I wish to express my appreciation to Professor N. E. Tolbert for his guidance, patience, and support during this investigation and my graduate training. Also I wish to express my sincere thanks to my husband for his understanding, tolerance, and assistance. Without his encouragement, this dissertation would not have been possible. I thank Drs. L. L. Bieber, T. A. Helmrath, R. W. Luecke, and J. L. Fairley for their helpful suggestions on this manuscript. I would like to thank Dr. F. Ryan, Dr. R. Gee, and my fellow graduate students in this lab- oratory for the many interesting discussions. I eSpecially thank Mrs. Sarah Caldwell for her excellent technical assistance and her help in revising this dissertation. Support from a National Institute of Health and the Department of Biochemistry, Michigan State University is appreciated. ii TABLE OF CONTENTS LIST OF TABLES . C O O O C O C O O I LIST OF FIGURES . . . . . . . . . . . INTRODUCTION 0 O O O O O O O O O C 0 LITERATURE REVIEW . . . . . . . . . . Microbody . . . . . . . . . . . Occurrence and Distribution . . . . Metabolic Sequence and Function . . . L-a-Hydroxy Acid Oxidase . . . . . . Aminotransferases . . . Subc ellular Location of Glyoxylate Amino- transferase . . . . Hypolipidemic Drug and Microbody Prolifera- tion MATERIALS AN Isolat Isolation of Peroxisomes from Rats Fed with Diff Isolat Fracti Assays 1. 2. 3. 4. 5. 6. 7. 8. Glycin Protei Gel Fi DEAE C D METHODS O O I O I O O O 0 ion of Animal Peroxisomes . . . erent Diets . . . . . . 1on of Peroxisomes from Spinach . onation of Rat Liver Peroxisomes Catalase . . . . Alpha-Hydroxy Acid Oxidase . Urate Oxidase . . . . . . Cytochrome c Oxidase . . NADH-Cytochrome c Reductase Acid PhOSphatase . . . . . Leucine- Glyoxylate Aminotransferase Phenylalanine-Glyoxylate Aminotrans- ferase . . . . . . . . . . e Identification .“ . . . . . . n Assay . . . . . ltration by Sephadex G-150 . . . ellulose Chromatography . . . . iii Page vi viii H “\Dxlobtb .5 23 28 28 29 30 31 31 32 32 33 33 33 34 36 36 36 37 Page RESULTS . . . . . . . . . . . . . . 38 Gradient Separation of Particles . . . 41 Subcellular Distribution of the Glyoxylate Aminotransferase . . . . . . . . . 43 Aminotransferase Assay . . . . . . . 48 Substrate Specificities . . . . . . 56 Subcellular Distribution of L-Alanine- Glyoxylate Aminotransferase . . . . . 60 Glycine Product Identification . . . . 62 Stimulation of Peroxisomal Aminotransferase Activity by Triton X-100 . . . . . . 63 Choice of Buffers . . . . . 65 Heat Stability of Glyoxylate Aminotrans- ferase . . . . . . . . . . . 67 Inhibitor Studies . . . . . . 71 Evaluation of Pyridoxal Phosphate . . . . 73 Separation of Glyoxylate Aminotransferase Activities by Gel Filtration Chromatog- raphy . . . . . . . . . . . . 75 Molecular Weight Estimation . . . . . 83 Separation of the Peroxisomal Glyoxylate Aminotransferase Using Ion Exchange Chromatography . . . . . . . . . 83 PH Optima . . . . . . . . . . . . 86 Kinetic Parameters . . . . . . . . 87 Location of the Glyoxylate Aminotransferase within the Peroxisomes . . . . . . 92 Reversibility of the Aminotransferase Reaction . . . . . . . 93 Postnatal Development of Leucine-Glyoxylate Aminotransferase . . . . . . . . . 96 Effect of Starvation of Rats on Leucine- Glyoxylate Aminotransferase Activity . . 98 Effect of Diets on the Leucine-Glyoxylate Aminotransferase Activity . . . . 100 Proliferation of Peroxisomes by Clofibrate . 101 Effect of Sex on Peroxisomal Aminotrans- ferase and a-Hydroxy Acid Oxidase Activities . . . . . . . . . . . 105 Leucine-Glyoxylate Aminotransferase in Rat Kidney . . . . . . . 109 Absence of Leucine-Glyoxylate Aminotrans- ferase in Spinach Leaf Peroxisomes . . . 111 Preliminary Examination of the L-Histidine- Glyoxylate Aminotransferase Reaction in Rat Liver Peroxisomes . . . . . . . 113 iv Page SUMMARY AND DISCUSSION . . . . . . . . . 118 Subcellular Distribution . . . . . . . 119 Substrate Specificities . . . . . . . 120 Biochemical PrOperties . . . . . . . 123 PH Optimum and Kinetics . . . . . . 124 Stability and Inhibitor Studies . . . . 125 Physiological Studies . . . . . . . 128 Tissue Distribution . . . . . . . . 130 BIBLIOGRAPHY . . . . . . . . . . . . . 133 10. 11. 12. 13. 14. LIST OF TABLES Distribution of Enzymes in a Sucrose Density Gradient of Rat Liver Homogenate . . . . Glyoxylate Aminotransferase Activities in Organellar Fractions . . . . . . . . Specificity of Amino Group Donor for Gly- cine Formation by Rat Liver Peroxisomes . . Specificity of Amino Group Acceptor for Glyoxylate Aminotransferase . . . . . . Additive Tests with Peroxisomal Glyoxylate Aminotransferase . . . . . . . . . Distribution of Glyoxylate Aminotransferase Activities among Particles . . . . . . Glycine Product Identification . . . . . Triton X-100 Release of Leucine-Glyoxylate Aminotransferase from Rat Liver Peroxi- somes . . . . . . . . . . . . . Buffer Effects on Leucine-Glyoxylate and Phenylalanine-Glyoxylate Aminotransferase ActiVitY O O I O O O O O O O O 0 Heat Stability of the Peroxisomal Glyoxylate Aminotransferases . . . . . . . . . Effect of Inhibitors . . . . . . . . Effect of Pyridoxal Phosphate and Dialysis . Summary of Partial Enzyme Purification . . Postnatal DevelOpment of Leucine-Glyoxylate Aminotransferase . . . . . . . . . vi Page 42 47 57 59 60 61 63 65 66 68 72 74 76 97 Table Page 15. Starvation Effect on Leucine-Glyoxylate Aminotransferase Activity . . . . . . 99 16. Dietary Response of Leucine-Glyoxylate Aminotransferase . . . . . . . . . 102 17. Effect of Clofibrate Treatment on Peroxi- somal Leucine-Glyoxylate Aminotransferase . 104 18. Comparison of Leucine-Glyoxylate Amino- transferase Levels in Male and Female Rat Liver O O O O O O O O I O O O O 106 19. Comparison of Glycolate Oxidase Level in Male and Female Rat Liver . . . . . . 108 20. Distribution of Leucine-Glyoxylate Amino- transferase Among Particles from Rat Kidney O O O O O O O O O O O O O 110 21. Distribution of Leucine-Glyoxylate Amino- transferase in Spinach Leaves . . . . . 112 22. Distribution of L-Histidine-Glyoxylate Aminotransferase Reaction in Rat Liver . . 114 23. PH optimum for L-Histidine-Glyoxylate Amino- transferase Reaction . . . . . . . . 114 24. Heat Stability of the Histidine-Glyoxylate Aminotransferase . . . . . . . . . 115 vii Figure 1. 10. 11. LIST OF FIGURES Subcellular Distribution of Leucine-Gly- oxylate Aminotransferase . . . . . . . Distribution of Phenylalanine-Glyoxylate Aminotransferase in a Sucrose Density Gradient . . . . . . . . . . . . Standard Curves for the Colorimetric Forma- tion of a-Ketoisocaproate and Phenyl- pyruvate . . . . . . . . . . . . Colorimetric Assay for Aminotransferase . . Peroxisomal Glyoxylate Aminotransferase Radiochemical Assay . . . . . . . . Temperature Dependence of the Peroxisomal Glyoxylate Aminotransferase Activities . . Elution Profile of the Peroxisomal Gly- oxylate Aminotransferase by Gel Filtration Column Chromatography . . . . . . . . Estimation of the Molecular Weight of Leucine-Glyoxylate Aminotransferase from Rat Liver Peroxisome . . . . . . . . . Elution Profile of Peroxisomal Glyoxylate Aminotransferase by DEAE Cellulose Column Chromatography . . . . . . . . . . PH Optimum for the Glyoxylate Aminotrans- ferase from Rat Liver Peroxisomes . . . . Double Reciprocal Plots of Initial Velocity against substrate concentration for Leucine- Glyoxylate Aminotransferase and Pehnyl- alanine-Glyoxylate Aminotransferase . . . viii Page 39 44 50 52 54 69 78 81 84 88 88 Figure 12. 13. 14. 15. Double Reciprocal Plots of Initial Velocity vs Substrate Concentration for Leucine-Glyoxylate Aminotransferase . Double reciprocal Plot of Initial Velocity vs Substrate Concentration for Phenyl- alanine-Glyoxylate Aminotransferase . Location of Leucine-Glyoxylate Aminotrans- ferase and Phenylalanine-Glyoxylate Amino- transferase in the Soluble Matrix of Rat Liver Peroxisomes . . . . . . Gel Filtration Chromatography of L-Histi— dine-Glyoxylate Aminotransferase . ix Page 90 90 94 116 INTRODUCTION My total thesis research dealt with four problems concerning the enzymatic composition of peroxisomes from liver and kidney. Specific aminotransferases have been reported in spinach leaf peroxisomes (86), and germinating seed gly- oxysomes (98), but no investigation of aminotransferases in mamalian peroxisomes has yet been reported. In part one of this investigation a glyoxylate specific amino- transferase was shown to be located in peroxisomes of rat liver and kidney. This enzyme was partially isolated from the peroxisomes and its prOperties were elucidated. The results of this investigation constitute the total of this thesis report. Aminotransferases serve as metabolic links between the amino acids or proteins and the carbohydrate pools. An understnading of the number and characteristics of the peroxisomal aminotransferase may help to elucidate the function of peroxisomes in mammalian tissue. The second part of my Ph.D. program of study dealt with the substrate specificity of the peroxisomal enzyme, a-hydroxy“ acid oxidase from both the liver and kidney of rats and pigs. The a-hydroxy acid oxidase activity varied 1 with the type of tissue used and its sources. The peroxi- somal dehydroxy acid oxidase from rat or pig liver preferen- tially utilized glycolate and a-hydroxyisocaproate. The activities were shown to be due to a single liver peroxi- somal protein. Rat kidney peroxisomal a—hydroxy acid oxidase utilized a-hydroxyvalerate, a-hydroxycaproate, a-hydroxyisocaproate, and to some extent, longer chain a-hydroxy acids as substrates, but it did not catalyze the oxidation of glycolate or lactate. Peroxisomes from pig kidney contained both of these a-hydroxy acid oxidases, one for the short chain and the other for the long chain a-hydroxy acids, and both oxidized a-hydroxyisocaproate. These results have been published in reference 54. The third part of my research program was a survey for malate dehydrogenase in mammalian liver peroxisomes. Since plant peroxisomal malate dehydrogenase had been shown to be electrophoretically different than other isozymes in the mitochondria and cytosol, it was necessary to know whether this enzyme was present also in mammalian peroxi- somes. No malate dehydrogenase could be observed in mamma- lian liver or kidney peroxisomes. These results have been published in reference 31. The fourth part of my research program was concerned with the discovery of NAD: a-glycerol phosphate dehydroge- nase in animal peroxisomes. Because malate dehydrogenase, which might serve as a transport system for reducing equiva- lent NADH into and out of the peroxisomes in mammalian liver and kidney tissues was absent, it was necessary to search for other hydrogen transport systems in mammalian peroxi- somes. Significant amounts of NAD: a-glycerol phosphate dehydrogenase were found in many mammalian liver and kidney peroxisomes. This work was also published in reference 31. It is thought at this time that the NAD: a-glycerol phos- phate shuttle may be involved in transport into the animal peroxisomes. L ITE RATURE REVI EW Microbody Occurrence and Distribution The term microbody was first used by Rhodin (87) in 1954 to designate a special type of particle observed by electron micrOSCOpy in the tubule cells of mouse kidney. Later the term microbody was applied to all similar particles in other tissues. Microbodies are characterized by a single limiting membrane and by a fine granular matrix. In addi- tion, they often include a denser inner core or nucleoid which in the hepatic microbodies of many species shows a very regular, crystalloid or polytubular structure. In rat liver, this core is the site of location of urate oxi- dase. Generally speaking, microbody is a nonspecific, morphological term, currently in vogue for all organelles of this type. So far two different metabolic sequences in microbidies have resulted in two more specific designations, peroxisomes and glyoxysomes. Mammalian microbodies have been called peroxisomes by De Duve; they are the site of hydrogen peroxide formation and metabolism (19,20). As a result of biochemical studies, De Duve and his coworker concluded that liver and kidney peroxisomes represent in 4 general, flavin oxidases linked to catalase: RH + 0 .1 R + H 2 2 202 H20 + H20 + 1/2 02 2 The oxidases which have been reported in the peroxisomes are urate oxidase, D-amino acid oxidase and L-a-hydroxy acid oxidase. Hydrogen peroxide is formed by the flavin oxidases in the peroxisomes and is then immediately decom- posed by the high levels of catalase which is the most characteristic marker enzyme of peroxisomes. Microbody-like structure was first isolated from spinach leaf tissue by Tolbert's group (114). Leaf micro- bodies were named peroxisomes because they metabolically resembled peroxisomes found in liver and kidney. Later, all the higher plants leaf peroxisomes examined (47) were found to contain most of the enzymes for the glycolate pathway of metabolism from the photosynthetic carbon cycle (114). Leaf peroxisomal respiration is manifested by photorespiration. Frederick and Newcomb (28) and Vigil (123) have shown that a crystalline.core, the site of location of part of the catalase, could be observed in plant peroxisomes. Another plant organelles, which have the same physical characteristics as the peroxisomes but contain the enzymes of the glyoxylate cycle, was called the glyoxysomes by Beevers (6) to emphasize the functional role of the glyoxylate cycle in germinating fatty seeds. In 1969 Novikoff and Goldfischer (66,67) developed and modified an alkaline 3,3'-diaminobenzidine (DAB) stain- ing procedure based on catalase peroxidation of this dye for electron microscopic studies of "peroxisome-like" organelles. It is known that the alkaline DAB methods reveals the presence of catalase (65,66). Using this method, catalase rich particles, presumably peroxisomes, have been found in yeast (3), algae (117), two species of protozoa (4,59), animal tissues such as adrenal cortex (5), small intestine (16), even the striated muscle cells (36), and also other animal tissues (1,5,16,34,35,36,72,76,99). Attention was drawn by Novikoff to a Special type of per- oxisome which was generally smaller in size (about 0.15 to 0.25 n) and lacked a core sturcture, but which showed a positive reaction with DAB. Microperoxisome was proposed to designate this special smaller peroxisome by Novikoff's group (67). It is not yet clear whether micrOperoxisomes and peroxisomes vary in other properties than just their size, and whether they are to be considered as different particles. Despite tht tissue and size differences, all micro- bodies are surrounded by a single membrane which is pre- sumed to be highly permeable. The membrane of microbodies sometimes shown continuity with the smooth endoplasmic reticulum by means of a sleeve-like projection that is often contorted. It is thought that the microbodies proba- bly form as bulges that form at the terminal end of certain specialized areas of end0plasmic reticulum where they remain permanently attached by means of a stalk that is broken by the homogenization procedure. Alternatively, they may be formed in this manner and eventually bud off to become free in the cytoplasm. Water in microbodies exchange with the sucrose of the gradient used for isolation so that the whole particle will sediment to the specific density of 1.24 to 1.26 g x cm-3. Microbodies are now recognized as a ubiquitous subcellular particle in animal tissues (68), but they appear to vary quantitatively among different sources. Metabolic Sequences and Function The function and prOperty of leaf peroxisomes have been extensively studied (113,115,119). Tolbert's group (47,53) has established that some of the reactions of photorespiration are compartmentalized in the leaf peroxi- somes; however, three other cellular compartments also contain specific steps in these metabolic sequences. Gly- oxysomes have been shown by Beevers (6) to convert the fatty acids in the storage tissues of seedlings to C4 acids, which are then utilized for carbohydrate synthesis. In comparison to what is known about the metabolic sequences of plant microbodies, little is known about the metabolic pathways which occur in animal peroxisomes, and only postulations exist about the functional role of animal peroxisomes. Animal peroxisomes are thought to be involved in the detoxification of hydrogen peroxide. Because of the association of hydrogen peroxide-producing oxidases and catalase in hepatic microbodies, it was suggested that peroxisomal compartmentation serves as a protective role against an increased concentration of hydrogen peroxide in the cell (19,20). Reedy and Svoboda (79) have suggested that hepatic microbodies are not really unit organelles but merely electronopaque peroxisomal enzymes circulating in the endoplasmic reticulum channels. The presence of circulating enzymes in endoplasmic reticulum channels would be an effective way for rapid detoxification of hydrogen peroxide and serve as rapid transit shuttle sys- tems for transfer of electrons from one part of the cell to another. Microbodies are also thought to be involved in lipid metabolism. The functional role of a glycerol phos- phate dehydrogenase (31) and carnitine acetyltransferase (51) in animal peroxisomes is unknown but suggests an active metabolic system for fats. Based on the close con- tact of smooth ER, microperoxisomes and zymogen granules, Novikoff (65,67,69) suggests that microbodies may be involved in the metabolism of lipid. Speculation has also indicated that microbodies may be involved in steroid bio- synthesis and/or catabolism (67,77). This was supported by Reddy and Svoboda (77,80) who found peroxisomal like structures in the interstitial cells (Leydig cells) of testes of rats and mice. They pointed out that microbody catalase of testicular interstitial cells might regulate the intracellular cholesterol pool size and its utilization in androgen biosynthesis/or catabolism. L-Alpha-Hydroxy Acid Oxidase Microbodies from protoza, plants and animals con- tain an a-hydroxy acid oxidase (HAO), (19,115). The oxida- tion of L-a-hydroxy acids to the correSponding keto acids by the FMN containing a-hydroxy acid oxidase is one of these characteristic reactions of peroxisomes. Microbodies from plants metabolize large amounts of glycolate (a-hydroxy- acetate) during photorespiration but little or no lactate (15,111). Plant peroxisomal glycolate oxidase has been extensively studied (111,112). In Tetrahymena, this oxidase activity was more specific toward DL-a-hydroxybutyrate. In animal tissue this oxidase from rat liver most rapidly oxidizes glycolate and also lactate at a slower rate (48). One of the first reports of an a-hydroxy acid oxi- dation in animals was the oxidation of glycolic acid in 10 rat liver by Dohan in 1940 (21). In 1952 Kun partially purified and studied a soluble enzyme from rat liver which was responsible for the oxidation of glycolate to glyoxylate (48). This purified glycolate oxidase, like its plant counterpart, could oxidize L—lactate at about one third the rate of glycolate oxidation (49). In plants the HAO (glyco- late oxidase) was first reported to be in the cytosol. Later Tolbert's group demonstrated that glycolate oxidase has a distribution similar to catalase in a sucrose gradient and is considered exclusively a peroxisomal enzyme (47,111, 114). In the animal system, the HAO was at first reported by Nakano (62) to be associated with mitochondria. Later De Duve and coworkers, when they were able to separate peroxisomes from mitochondria, reported the L-a-hydroxy acid oxidases were associated with peroxisomes in rat liver and kidney tissues (19). Later Nakano gt_§1. (63) found that the enzyme preparation from rat liver also catalyzed the oxidation of L—a-hydroxyisocaproic acid. They concluded that L-a-hydroxyisocaproate oxidase was mainly located in the light mitochondria fraction, which would have been enriched with peroxisomes. The partially purified L-a- hydroxyisocaproate oxidase was similar to the liver gly- colate oxidase purified by Kun (49). It is a typical flav0protein, and FMN is required as the prosthetic group. More recently L-a-hydroxy acid oxidase and glycolate oxidase 11 were purified by Ushijima (122) separately from rat livers and found to be identical, judging by substrate specifi- cities, Km values and coenzyme (FMN), activation energy, inhibition rates by various reagents and pH optimum. The Km value for this liver enzyme was 2.4 x 10-4M for gly- colate and 1.26 x 10-3 M for L—a-hydroxyisocaproate. The significance of glycolate oxidase in the rat liver is not clear since its specific activity is low and the importance of glycolate and a-hydroxyisocaproate in any metabolic scheme is unknown. There is also an a-hydroxy acid oxidase in rat kidney tissue. Blanchard gt_al. (8) firstly purified an L-HAO from rat kidney and showed that it catalyzed the oxidation of some longer chain a—hydroxy acids (not glycolate) to the corresponding keto acids. This renal HAO was also a flav0protein in which the prosthetic group appeared to be FMN (62). Substrate specificity studies of this renal enzyme showed the enzyme to be most active with a-hydroxy— heptylate and a-hydroxyisocaproate, but not glycolate and lactate (62). Later this enzyme was reported by De Duve's group (19,20) to be present in the rat renal peroxisomes. The subcellular localization of this oxidase in rat kidney peroxisomes was confirmed by Nakano (64).: They concluded that the enzyme associated with peroxisomal fraction appeared to be the same as that in the supernatant, which could have come from peroxisomal rupture during homogenization. 12 Recently Shnitka“et;al. (101) using ferricyanide as an electron acceptor also demonstrated the L-HAO occurred in the peroxisomes of rat kidney in vivo. From all these studies, two HAO systems were established in rat peroxi- somes. One was specific for short chains such as gly- colate and L-lactate in rat liver, and the other for longer chains such as hydroxybutyrate and hydroxyvalerate in rat kidney. Isozymic forms from the rat kidney with dehydroxyburyrate were observed in 1965 by Allen and Beard (2). A glycolate oxidase was purified from pig liver by Schuman and Massey (100) in 1971. The amino acid composition and spectrum of the enzyme were studies exten- sively. The pig liver glycolate oxidase was similar to the one reported by Ushijima (122). The enzyme contained two moles of FMN per mole of protein with a molecular weight of 100,000 (100). No study has been reported con- cerning the subcellular localization and substrate specifi- city of this enzyme in pig liver. Two L-a-hydroxy acid oxidases were purified 500 to 1,000 fold from pig kidney by Robinson §E_31. in 1960 (44) and extensive studies in 1962 (90). Both short and long chain L-a-hydroxy acid oxidases were found to exist in hog renal cortex. The short chain L-HAO was most active for glycolate, but also had activity with L—a-hydroxyisocaproate and related short chain compounds. The long chain L-HAO 13 was specific for many aliphatic and aromatic acids larger than a-hydroxybutyrate and exhibited no reaction on gly- colate. The hydroxy acid oxidase from pig kidney were flavoproteins, in which the prosthetic group appeared to be FMN. The pH optimum was 7.7 for the short chain oxidase and 7.9 for the long chain oxidase (90). With reference to glycolate and lactate oxidation, the short chain oxidase was similar to the glycolate oxidase obtained from liver as described by Kun. Later Nakano gt_al, (96) demonstrated that both the short and long chain oxidase activities of rat kidney showed the same distribution among particulate fractions as did both D-amino acid oxidase and catalase (peroxisomal enzymes). They concluded that both oxidases were localized in hog renal peroxisomes. The two oxidases showed isozymic differences during polyacrylamide gel electrOphoresis. Aminotransferase Enzymatic transamination was first described in 1937 by Braustein and Kritsman (9), who observed that prepa— rations of pigeon breast muscle catalyzed the formation of glutamate from a-ketoglutarate and a large number of amino acids. At that time, that enzymatic transamination with metabolic significance was thought to be largely restricted to glutamate-asparate and alanine-glutamate reactions. Later investigations confirmed that transamination reactions with D—amino acid, m-amino acid, and aldehydes also occured 14 in nature (56). It is now known that all the natural amino acids participate in enzymatic transamination and that amino group transfer is an obligatory step in the biosynthe- sis and catabolism of many amino acids. The nomenclature of aminotransferases has varied over the years. The official designation of aminotrans- ferases is by the amino acid-keto acid reaction involved. This permits the use of two names because aminotransferese reactions are reversible. Asparate-a-ketoglutarate and alanine-a-ketoglutarate were the first two aminotransferase reactions studied extensively (70). Different transaminases specific for pyruvate and oxalacetate, were reported later by Green and Nocito (33). In addition to the reactions of a-ketoglutar- ate, pyruvate, and oxalacetate, a number of transamination reactions involving w—amino acids and aldehydes were demon- strated in 1954 (55). Transamination involving ornithine and glyoxylate was first reported in 1954 (55). In this reaction ornithine reacted equally well with glyoxylate and with pyruvate; glycine and glutamic-Y-semialdehyde formation was observed. Later various tissue and cell preparations were found to catalyze the transamination of glyoxylate with other amino acids to give glycine, i.e. in rat liver by Meister (55), in tobacco leaves by Wilson (126) and in cell free extracts of a Pseudomonas by Camp— bell (13). Wright (128) showed that glyoxylic acid could 15 be used as effectively as glycine for the growth of a Neurospora mutant which specifically required glycine. Among the keto acids in the body fluid in the silkworm, glyoxylate was present in the highest concentration. Fukuda and Hayashi (29) concluded that glycine which is 33.6% of the silkprotein, was synthesized by transamination from glyoxylate and amino acids in the silkgland-cell of the silkworm. The optimal pH of this silkworm glyoxylate aminotransferase was 9.4. All these observations suggest that transamination reactions involving glyoxylic acid were widespread and biologically important. Although formation of glutamate from glycine and d-ketoglutarate had been reported (29,55,126), the reaction proceeded poorly in this direction compared to the gluta- mate-glyoxylate reaction. The equilibrium of the glycine- glyoxylate transformation strongly favored the conversion of glyoxylate to glycine formation; studies on nonenzymatic transamination between glycine and a—ketoglutarate led to a similar conclusion (57). Weinhouse and Friedmann (126), using the intact rat, were able to demonstrate a conver- sion of glyoxylate to glycine but not of glycine to gly- oxylate. It was reported that the required concentration of glyoxylate in transamination equilibrium was relatively low (57) . One of the first reports of a glyoxylate aminotrans- ferase activity in plants was the glutamate-glyoxylate l6 aminotransferase found in tobacco leaves by Wison et al. (126). They measured this activity by incubating glycolate- 1-14 , glutamate, and pyridoxal-5'-phosphate with a leaf extract. The glycolate was oxidized to glyoxylate by gly- colate oxidase; the glyoxylate was then converted into gly- cine with glutamate serving as the amino donor. In crude homogenates of wheat leaves, King and Way- good (45) observed alanine, asparate, glutamate, and serine-glyoxylate aminotransferase activity as well as glutamate-pyruvate and asparate-a—ketoglutarate aminotrans— ferase activities. A partially purified serine-glyoxylate enzyme could not utilize D-serine, L-phosphoserine, gly— colate or glycoaldehyde. No pyridoxal phosphate activation could be observed. Using a more highly purified serine- glyoxylate aminotransferase, King and coworkers (10) showed in the studies of initial velocity, enzyme resolution and reconstitution, pyridoxal phosphate activation, and enzyme inhibitor studies that this serine-glyoxylate aminotrans- ferase reaction operated by the Ping Pong Bi Bi mechanism. The coenzyme pyridoxal phosphate was involved in the reaction; however, metal ions and sulfhydryl groups did not appear to be involved. Recently, Rehfeld and Tolbert (86) reported two different glyoxylate aminotransferase activities in Spinach leaf peroxisomes. One was a glutamate-glyoxylate amino- transferase and the other was a serine-glyoxylate'reaction. 17 Both enzymatic reactions were irreversible. For serine- glyoxylate aminotransferase the Km (glyoxylate) was 0.15 mM, and Km (serine) was 2.72mM, and the pH optimum was about 7.0. This enzyme was inhibited by D-serine and phosphate buffer. Both serine-glyoxylate and glutamate-glyoxylate aminotransferases nonspecifically with alanine. No require— ment for added pyridoxal phosphate was observed for this partially purified peroxisomal glyoxylate specific amino- transferase. A serine-glyoxylate aminotransferase has been puri- fied and studied in kidney beans by Smith (102,103). A 100- fold purified serine-glyoxylate aminotransferase, essentially free of other aminotransferases, co-purified with serine- pyruvate aminotransferase; both activities exhibited the Ping Pong Bi Bi reaction mechanism. Both the serine- glyoxylate and serine-pyruvate aminotransferase were inhibi- ted by 1ow concentrations of hydroxylamine; they were less sensitive to N-ethylmaleimide and p—chloromercuribenzoate. Serine-glyoxylate aminotransferase activity was inhibited by ammonium ion. The inhibition was linear competitive with serine and nonlinear non-competitive with glyoxylate. Several transminases have been discovered in animal catalyze irreversible conversions of glyoxylate to glycine. Nakada (61) partially purified and studied glutamate- glyoxylate aminotransferase from rat liver in 1964. He 18 found that only L-glutamic acid, and possibly L-glutamine at about a tenth of the rate of L—glutamate, could serve as the amino group donor to glyoxylate. The maximal activ- ity of the glutamate-glyoxylate aminotransferase was achieved at a pH between 7.2 to 7.4. The presence of a vitamin B6 derivative in the aminotransferase was suggested by the fact that 90% inhibition was observed at 20mM isonicotinic acid hydride. Strecker (105) purified an ornithine transaminase from rat liver in 1965. It had been reported earlier that this ornithine aminotransferase was active with ketoglutar- ate; however, Strecker's preparations from rat liver were found to be most active with pyruvate and glyoxylate. This enzyme activity was found to be irreversible in the direc- tion of glutamic-y-semialdehyde formation. Again the ornithine-pyruvate and ornithine-glyoxylate aminotransferase reactions were found to be the same enzyme protein. In 1966 Thompson and Richardson (109) partially purified an irreversible glutamate-glyoxylate aminotrans- ferase from human liver. Glutamate, glutamine, alanine, arginine, and methionine were all active as amino group donors. The Michaelis constant was 2 x 10-3 M for glutamate and glyoxylate, and the Optimal pH was 7.3. Glutamate- glyoxylate aminotransferase from human liver was similar to the transaminase isolated from rat liver by Nakada (61). 19 The human liver enzyme was inhibited by the p- chloromercuribenzoate while the enzyme from rat liver was not. This result suggested the presence of an essential sulfhydryl group in the enzyme from human liver. No inhi- bition was Observed at l x 10-3M isonicotinic acid hydra- zide. Later Thompson and Richardson (110) characterized and purified 900-fold a L-alanine-glyoxylate aminotrans- ferase from human liver. The enzyme catalyzed the trans- fer of the a-amino group of alanine to glyoxylate, forming glycine. This reaction was also irreversible. Pyridoxal phosphate was required for catalytic activity and enhanced the stability of the enzyme during purification and storage. In this purified enzyme from human liver, gly- oxylate was the only active amino group acceptor from alanine. The L—alanine-glyoxylate aminotransferase purified from rat liver (92,93) had different characteristics than the enzYme reported from human liver. NO enzyme activity was found in brain, heart skeletal muscle, or spleen of mouse, rat, pig, or rabbit. The highest activity was Observed in the liver; the kidney also had some activity. L—alanine-glyoxylate aminotransferase activity in rat liver was elevated during the neonatal period or after glucagon injection in the adult. This seems to parallel the develop— ment of hepatic gluconeogenesis in the rat. Glucagon treat- ment also increased hepatic gluconeogenesis in adult rats. 20 The data suggest that the enzyme might have a glucogenic role by initiating a process of glucose formation from glyoxylate via glycine. L-Alanine-hydroxypyruvate aminotransferase from rabbit liver, a key enzyme in the non-phosphorylated path- way for serine metabolism, has been purified and character- ized (26). The molecular weight Of the purified enzyme was estimated to be 41,000 daltons. By SDS gel electro- phoresis, the enzyme consists of a single polypeptide chain. Purified alanine-hydroxypyruvate aminotransferase catalyzed the transamination of glyoxylate as well as hydroxypyruvate with L-alanine as the preferred amino donor for both substrates. The two enzymatic activities could not be separated during purification, and a single enzyme which could catalyze the transamination of both glyoxylate and hydroxypyruvate was considered. Kinetic studies also demonstrated that the two a-keto acids are competitive substrates. The Km was 1.6mM for hydroxypyruvate and 9.3mM for glyoxylate. Substrate inhibition was observed for hydroxypyruvate at concentrations greater than 3.3mM and for glyoxylate at concentrations greater than 5.7mM. Again the L—alanine-glyoxylate aminotransferase was irreversible, but L-alanine-hydroxypyruvate was reversible. NO inhibi- tion was Observed with either p-chloromercuribenzoate or N—ethylmaleimide indicating that reactive sulfhydryl groups 21 were not involved in the catalysis. This hepatic L-alanine- hydroxypyruvate transaminase has been shown to be under dietary and hormonal control (14); its activity was increased by high protein diets (88% casein), glucagon, cortisone, and cyclic AMP. These Observed increases in the activity of hepatic L-alanine-hydroxypyruvate transaminase in rabbits were considered to be due to an altered rate of enzyme syn- thesis. Similar data were reported for serine-pyruvate aminotransferase (85) in rat liver. the enzyme activity increased at birth, was maintained throughout the suckling period, and increased markedly after glucagon or alloxan injection. Subcellular Location of Glyoxylate Aminotransferase Kisaki and Tolbert (46) first found glutamate- glyoxylate aminotransferase in spinach leaf peroxisomes. This peroxisomal aminotransferase could utilize D- or L- glutamate, L-alanine and L-serine as amino group donors. In any case glyoxylate was the superior amino group acceptor for the leaf peroxisomal aminotransferase. The availabil- ity of large amounts of glycolate from photosynthesis (47), the presence of an active glycolate oxidase for glyoxylate formation in the peroxisomes, and the irreversibility Of the glutamate-glyoxylate aminotransferase in spinach leaf peroxisomes indicated that the reaction proceeded primarily 22 toward the formation of glycine. This peroxisomal glutamate- glyoxylate aminotransferase showed a pH Optimum between 7.0 to 7.5 which was different from that of cytosol gluta- mate-glyoxylate aminotransferase. Isonicotinyl hydrazide was shown to inhibit the peroxisomal aminotransferase at high concentrations (46). Later Yamazaki and Tolbert (130) and Rehfeld and Tolbert (86) reported that serineeglyoxylate aminotransferase activity in the particulate fraction was localized entirely in the spinach leaf peroxisomes. This aminotransferase was most active for serine-glyoxylate which is considered as its in vivo substrate; however, it also catalyzed a serine-pyruvate reaction at one tenth the rate Of the serine-glyoxylate reaction and an alanine-gly- oxylate reaction nonspecifically. Several experiments indicated that the serine-glyoxylate and serine-pyruvate reactions were catalyzed by one enzyme. This spinach leaf peroxisomal serine-glyoxylate aminotransferase could be separated completely from the glutamate-glyoxylate enzyme by isoelectric focusing technique in a pH 3 to 10 grad- ient. Several glyoxylate aminotransferases with glutamate (124), serine and alanine (85) were also found in rat liver peroxisomes. No added pyridoxal phosphate was required for the peroxisomal glutamate-, or serine-glyoxylate amino- transferase activities as measured in the particulate 23 fractions. Both particulate glutamate-glyoxylate and serine-glyoxylate aminotransferases were localized in the peroxisomal fraction. The low activity observed in the mitochondrial fractions could be attributed to peroxisomal contamination. Although there was alanine-glyoxylate aminotransferase in the rat liver peroxisomes, most of it was located in the mitochondria (94). A11 peroxisomal glyoxylate aminotransferases are irreversible in the direction Of glycine formation. The alanine-glyoxylate aminotransferase in the rat liver cytosol and mitochondria are isoenzymes. The solubilized mitochondrial enzyme had a pH Optimum Of 8.6, an apparent Km of 0.24mM with respect to glyoxylate, and was inhibitied by glyoxylate at concen- trations above 5mM. The cytosol aminotransferase showed no distinct pH Optimum (7.0 to 9.0) and no inhibition by glyoxylate at high concentration. It is possible that this "soluble" form is due to broken peroxisomes. Hypolipidemic Drug and Microbody Proliferation Although considerable progress has been made in the understanding of the functional role of peroxisomes in plant cells (113,115,119), the importance Of these organelles in cellular metabolism in animal tissue remains unclear. Since microbodies have always been Observed in inti- mate spatial relationship with smooth endoplasmic reticulum 24 or lipid drOplets (67,72), it has been postulated that animal microbodies play a role in lipid metabolism which has been proven for glyoxysomes in germinating seeds (6). Because microbodies are present in interstitial cells of rat, mouse and guinea pig testis, Reddy and Svoboda (76, 78) suggested that microbodies might be related to the snythesis, storage or utilization of cholesterol or andro- gens. In order to elucidate the physiological role of animal microbodies, a hypolipidemic drug, d-p-chlorOPhenoxy- isobutyrate (CPIB, or Clofibrate) which is known to lower serum triglycerides and cholesterol in man and in experi- mental animals (37,106) has been found to be a valuable regulant. A rapid and sustained increase in the number of microbodies in liver and kidney cells can be induced in male rats by CPIB. The majority of the induced hepatic microbodies possessed the characteristic central core or nucleoid, though several anucleoid microbodies also were Observed. This phenomenon permits investigation of several aspects of microbody composition in a systematic fashion (106). The increase in the number of microbodies is associated with significant elevation in catalase activ- ity resulting from an enhanced rate of synthesis of this enzyme (74). Simultaneous administration of chloramphenicol and CPIB in male rats results in no increase in the micro- body numbers (106). The number of peroxisomes in rat liver 25 cells reverts to normal between two and three weeks after withdrawal Of CPIB. Extracts from rat tumor cells have been shown to inhibit 80% of the in vivo synthesis of catalase (40). It is believed that the depression of catalase activity in a tumor-bearing animal is due to a reduction in the rate of snythesis of catalase in these animals (75). In addition to depression of catalase activity, urate oxidase and D- amino acid oxidase activities are also reported to be reduced in the liver and kidney of tumor-bearing animals. When CPIB was given to these tumor-bearing animals, a significant increase in the number of microbodies in hepatic parenchymal cells was observed. This increase in microbody number is also associated with an elevation in the content and activity of catalase protein (75). These results indi- cate that, despite initially low levels of catalase in the liver Of a tumor-bearing animal, this enzyme can be induced effectively in the liver of these animals by administering CPIB. Recently, one or more double-walled tubular struc- tures, measuring 110 mu in diameter, were Observed in the hepatic microbodies of hypophysectomized rats, as well as in rats bearing liver tumors, following CPIB treatment (82). These tubules were oriented irregularly within the micro- bodies. Examination of several microbodies containing these tubules suggested that these structures may be formed 26 as a result of transformation of microbody matrix protein(s). Varying degrees of electron lucency in the microbody matrix and the size and shape of the channels containing these straight tubules are considered as further evidence to indicate that microbody proteins circulate within the endo- plasmic reticulum channels. There seems to be an interaction Of the sex hormones and CPIB on microbody formation. Reddy and Svoboda (107) Observed sex related differences in response to CPIB in adult rats. No increase in microbodies occurs in female rats following CPIB treatment; while in castrated females given testosterone prOpionate and CPIB a marked increase in the number of liver microbodies and catalase activity occurred. Recently another hypolipidemic drug, a CPIB derivative, 2-methyl-2[p-(l,2,3,4,-tetrahydrol-1 napthyl) phenony-prOpionic acid (Nafenopin) was reported to cause significant increases in the number of microbodies and catalase activity in livers of both female and male rats and mice (81,83). The increase in microbody population correlated with an increase in the amount of catalase protein. The absence of sex related differences in micro- body proliferation response in NafenOpin-treated rats and wild type mice is of particular significance. NafenOpin can, therefore, be used as an inducer of microbody pro- liferation and Of catalase synthesis in both sexes of rats 27 and mice. The serum glycerol-glycerides were markedly lowered in all the animals given NafenOpin; this paralleled the increase in liver catalase. All the effects of Nafeno- pin were fully reversed when the drug was withdrawn from the diet. The demonstration of microbody proliferation and catalase induction with hypolipidemic compounds, CPIB, nafenopin, and methyl clofenapate, suggests a possible but as yet unclarified relationship between microbodies and hypolipidemia. MATERIALS AND METHODS Isolation of Animal Peroxisomes Livers were obtained from 100-1609 young female or male Sprague Dawley rats, that had been starved overnight. The livers, 20 to 30 g total, were perfused in_§itu with a 8.5% (w/w) sucrose and 20mM glycylglycine buffer pH 7.5. Unless stated otherwise, all sucrose solutions, which were used for grinding media or gradients, were made in 20mM glycylglycine at pH 7.5. Livers were chOpped into small pieces with scalpels and homogenized in a Potter-Elvejhem homogenizer with a motor driven teflon pestle (73). All work was done at 4°C. The homogenate was filtered through 6 layers of coarse cheese cloth and centrifuged at 2709 for 10 min in a Sorvall RC-2B centrifuge to remove large tissue fragments and cell debris. The homogenate was then placed directly in the core Of a B-30 rotor for an Inter- national zonal centrifuge. The operation of the zonal centrifuge has been described in detail in references 2, 3, and 4. Separation of peroxisomes from the other cell con- stituents was achieved on a nonlinear sucrose gradient. The gradient was prepared from the following percent (w/W) sucrose solutions: 10 to 20 ml Of 5%, 30 ml each of 20%, 28 29 25%, 30%, 35%, 40%, 42%, 44%, 45%, 46%, 48%, 49%, and enough 56% sucrose to fill the rotor. The solutions were loaded into the zonal rotor from the edge, starting with the lease dense solution. Up to 100 ml of liver homogenate could be added to the rotor through the core orifice. After centrifugation at 10,000 rpm for 15 min, 20,000 rpm for 15 min, and finally 30,000 rpm for 3 hours, the rotor was unloaded by pumping water into the core and collecting 10 m1 fractions from the edge of the rotor. Pig livers or medula and cortex tissues from rat kidneys were minced and homogenized in the same manner as rat liver. The sucrose gradients were the same as the one used for the rat liver preparation. Isolation of Peroxisomes from Rats Fed Different Diets High and low protein diets were purchased from General Biochemicals Co. The high protein diet contained 55% casein, and the low protein diet contained 3.4% casein. The 55% casein diet contained (per Kg): vitamin mix 30 9, corn starch 3009, cotton seed oil 80 9, salt mix 40 g, and vitamin free casein 550 g. The 3.4% casein diet (per Kg) contained: vitamin free casein 34g, corn starch 815 g, and the rest Of the high protein diet contents. Diets were stored at 4°C except during feedings. Female Sprague- Dawley rats were fed either on the low or high protein 30 diet for 7 days. Livers from one day starved rats were perfused and homogenized as described above. The same sucrose density gradient was used. A hypolipidemic drug, p-chlorophenoxyisobuturate (CPIB) sodium salt, was a gift from Ayerst Laboratories Incorporated CO. in New York. Male Sprague-Dawley rats of 150 9 weight were fed with 0.2% CPIB in the ground diet for designated times. Livers from CPIB fed rats were per- fused, homogenized and centrifuged in the same way as normal rats. Isolation of Peroxisomes from Spinach Spinach was purchased at a local market. Approxi- mately 200 g of washed and deribbed spinach leaves were homogenized in a Waring blendor for 7 to 10 sec at low speed. The grinding medium was 30% (w/w) sucrose in 20mM glycylglycine buffer at pH 7.5. The spinach homogenate was squeezed through 8 layers Of cheese cloth and centrifuged at 300 g for 5 min to remove the cell debris and most of the whole chloroplasts. The supernatant was then placed directly in the center Of a B-30 rotor. The gradient was composed of the following sucrose solutions (w/w): 25 ml each of 20%, 25%, 30%, 35%, 40%, 42%, 44%, 45%, 46%, 47%, 48%, 49%, and 56% to fill the head. About 250 m1 of the spinach supernatant was put on the gradient. After centri- fugation at 30,000 rpm for 3 hours, 10 m1 fractions were collected. 31 Fractionation Of Rat Liver Peroxisomes A peroxisomal suspension from a zonal gradient was diluted with an equal volume of 0.01 M pyrophosphate buffer 1x>a final pH of about8.5. After being stored overnight at 4°C, this suspension Of broken peroxisomes was put on a sucrose gradient: 5 ml each of 56%, 49%, 47%, 45%, 42%, 38%, 30% sucrose (w/w). About 14 ml of the peroxisomal suspension was put on the gradient with a 7 ml overlay. This gradient was centrifuged for 3 hours at 23,000 rpm in the SW-25.2 rotor with the L-2 ultracentrifuge. Fractions of 2.5 ml were collected from the bottom of the tube. Assays All spectrophotometric assays were run with a Gil- ford recording spectorphotometer at 35°C. Cytochrome c (Type II from horseliver), NADH, NAD, and amino acids were Obtained from Sigma Chemical CO., St. Louis, MO. Unless stated otherwise, all amono acids were of the L form. 1. Catalase This peroxisomal marker enzyme was assayed spec- trophotometrically by the initial rate of disappearance of hydrogen peroxide at 240nm in 50mM phOSphate buffer at pH 7.0. The peroxide has an initial absorbance of 0.6 (7). A chnage in 1.00 A24o was equivalent to 83.25 nmoles of hydrogen peroxide in a 3 ml assay. 32 2. Alpha-Hydroxy Acid Oxidase The reduction of the dye, 2,6-dichlorophenolindo- phenol, at pH 8.7 was measured under anaerobic conditions at 600nm. The reaction mixture in a 3.0 m1 Thunberg cuvette 4 contained 0.66 M sodium pyrophosphate at pH 8.7, 1.5 x 10— M 2,6-dichlorophenolindOphenol, 1 x 10"4 M FMN, 0.8% Triton X-100, 300 ul of enzyme and water, and 4.2mM sodium gly- colate or l7mM of other a-hydroxy acids to initiate the reaction. Saturated aqueous solutions of the longer chain a-hydorxy acids were used when their solubility was less than 17mM. 3. Urate Oxidase The oxidation of uric acid in 0.15 M borate buffer at pH 9.25 was followed by the decrease in absorbance at 293nm. A change of 1.00 A293 was equivalent to 79.8 nmoles in the 1 ml assay (42). 4. Cytochrome c Oxidase (mitochondrial marker) After incubating an enzyme aliquot with 0.02% digito- nin for one minute, the rate of oxidation of cytochrome c was followed at 550nm in 80mM phosphate buffer pH 7.0. The oxidized form of cytochrome c had been reduced before use by adding a few crystals of sodium dithionite until the A550/ A565 ratio was greater than 6 but less than 12. A change 33 Of 1.00 A was equivalent to a change of 48.75 nmoles of 550 cytochrome c oxidized in the 1.0 m1 assay (104). §;_ NADH-Cytochrome c Reductase NADH-cytochrome c reductase was assayed by measuring the rate of cytochrome c reduction at 550nm in a microcuvette containing: 0.1 ml of 0.2 M phosphate at pH 7.5, 50 ul of oxidized cytochrome c (5 mg/ml), 5 ul Of lOmM KCN, and 70 ul of enzyme suspension plus water. The reaction was initiated with 50 ul of NADH (3 mg/ml). A change of 1.00 A was 550 equivalent to a change Of 12.8 nmoles of cytochrome c reduced in the 0.27 ml assay (22,52). 6. Acid Phosphatase p-NitrOphenyl phosphate was used as the substrate in the enzyme assay. After 10 min incubation at 35°C, sodium hydroxide was added to stop the reaction (11). A change of 1.00 A410 was equivalent to 11.8 nmoles of p- nitrophenolate formed in the 0.21 ml assay. 7. Leucine-Glyoxylate Aminotransferase A modified colorimetric procedure was used to assay the activity of leucine-glyoxylate aminotransferase (108). Twenty umoles L-leucine, 50 nmoles sodium pyrophosphate at pH 8.4, 0.1 umole pyridoxal phosphate, and 0.04% Triton X-100, then water and enzyme suspension were added to make 34 a final volume of 1.2 ml. The assay mixture was allowed to equilibrate for 5 min in a 40°C water bath. Reaction was initiated with 2.5 nmoles sodium glyoxylate and termi- nated 60 min later by the addition of 0.3 ml Of 50% TCA. After precipitation of the proteins, 0.5 ml of 0.3% 2,4- dinitrophenyl hydrazine-HCl was added. Hydrazone formation was allowed to proceed for 10 min at room temperature before the 2,4-dinitrophenyl hydrazone of a-ketoisocaproate was selectively extracted with 2.5 ml cyclohexane by vig- orous shaking for 20 sec. The phases were separated by centrifuging for 2 to 3 min in a clinical centrifuge, and 2 ml Of the upper phase were transferred into a graduate test tube. Care was taken not to transfer any of the insoluble hydrazone of glyoxylate at the interphase and on the wall of the tube. Shaking with 1.5 m1 of 10% sodium carbonate extracted the hydrazone from the cyclohexane. After centrifugation for l min, 1 ml of the reddish-brown carbonate phase was transferred to a test tube, and 1 ml Of 1.0 N NaOH was added. Absorbance at 440nm was measured 10 min later with a Gilford 300-N spectrOphotometer. A control sample was run each time with the same amount of boiled enzyme suspension. 8. Phenylalanine-Glyoxylate Aminotransferase The phenylalanineéglyoxylate aminotransferase activ- ity was measured using the same protocol as in the leucine- 35 glyoxylate aminotransferase except that 20 nmoles of L- phenylalanine was used instead of L-leucine. 9. Radiochemical Amino- transferase Assays Leucine-glyoxylate, phenylalanine-glyoxylate, and other amino acid-glyoxylate aminotransferases were assayed using glyoxylate-14C as described by Rehfeld and Tolbert (86). The assay mixture contained 0.5 m1 of 0.1 M perphos- phate buffer at pH 8.4, 10 ul of lmM pyridoxal-5'-phos— phate, 20 ul Of 5% Triton X-100, 200 ul of 0.1 M L-amino acid and 400 pl of enzyme plus water. The reaction mixtures were preincubated at 40°C for 5 to 10 min. Reactions were started by the addition of so pl of 0.05 M glyoxylate-U-14C, incubated for 60 min,.and then terminated by placing the test tube in a boiling water for 5 min. Denatured proteins and pyrophosphate were precipitated by 50% alcohol. The supernatant was cooled and placed on a Dowex-l-acetate 14C. The column (0.6 x 3 cm) to remove excess glyoxylate-U- Dowex column was washed three times with 1 m1 of water. Only 1 ml of the collected effluent was transferred into a glass scintillation vial. After adding 10 m1 of scintilla- tor fluid [60 g naphthalene, 4 g PPO (2,5-diphenyloxazole), 0.2 g POPOP (phenyl-oxazolylphenyl-oxazolylphenyl), 100 ml methanol (absolute), 20 m1 ethylene glycol, and p-dioxane to make 1 liter], the samples were counted in a Packard scintillation counter. Quenching was Observed by the 36 channel ratio method. The specific activity of glyoxylate- U-14C varied between experiments. The radiochemical assay was also conducted using l4C-U-labelled amino acid. In this case, the reaction was stopped by the addition of 0.35 ml of a 10% TCA solution. The reaction mixtures were placed on Dowex-50-H+ columns to remove excess amino acid and the column washed three times with 1 ml of 10mM HCl. The rest of the procedure was the same as previously described for 14C-U-glyoxylate assay. Glycine Identification Thin layer plates, coated with silica gel F254, 0.25 mm thick were prewashed in butanol:acetic acid:water (60:20: 20, w/w/w). The plates were dried by heating to 80°C for about 60 min. Aliquots of the standard amino acids along with the radioactive samples eluted from the Dowex-l- acetate columns were applied to the plate, about 1.5 cm from the edge. Following ninhydrin development, the amino acids were possible to identify by comparison with the standards. Protein Assay The Lowry procedure was used to estimate the protein content of each enzyme samples (32,50). Gel Filtration by Sephadex G-150 Sephadex G-150 (Pharmacia) was suspended in either 20mM glycylglycine at pH 7.8 or 10mM phosphate at pH 7.4 37 with 1mM B—mercaptoethanol, 1mM pyridoxal phosphate and 1mM EDTA, and allowed to swell overnight on a steam water bath. The Sephadex G—150 column (1.5 X 90 cm) was eluted ' with the same swelling buffer. DEAE Cellulose Column After precycling and degassing, DEAE cellulose (Whatman DE 52) was equillibrated with two changes of the elution buffer overnight. A heat-concentrated peroxisomal sample was applied to the DEAE cellulose column (2.5 x 30 cm). The aminotransferases were eluted with a 600 ml linear gradient of 10mM to 200mM phosphate buffer at pH 7.8. RESULT Microbodies differ enzymatically from mitochondria. In order to separate the microbodies and mitochondria, suc- rose density gradients were made with minimal density changes over a wide part Of the gradient in the fractions between 1.15 and 1.25 g x cc-l as shown in Figure 1. This resulted in a steeper gradient at lower densities where organelles not being investigated were entrapped. The marker enzymes used to detect microbodies were catalase, L-a- hydroxyacid axidase, and urate oxidase. The marker enzyme for mitochondria was cytochrome c oxidase. Acid phosphatase was the marker enzyme used to detect lysosomes. Most of the peroxisomes were well separated in the sucrose gradient from the mitochondria and lysosomes using the centrifugational procedure (Figure 1). Most of the acid phosphatase was in a broad peak in less dense sucrose than the mitochondria. A small distinct peak of phosphatase activity was located in denser sucrose peaking around 1.21 g x cc-l. Sufficient resolution was Obtained, so that the distribution pattern for an enzyme could be assigned to one Of the other parti- cle. The absence of peroxisomal enzymes was especially 38 39 Figure l.--Subcellular Distribution of Leucine-Glyoxylate Aminotransferase. The total homogenate from about 25 grams of rat livers, after centrifugation at 270 g for 10 min, was placed on an isopycnic sucrose gradient and centrifuged for 3 hours at 30,000 rpm. Leucine- glyoxylate aminotransferase was assayed colori- metrically. 40 32 x .22 x 8.653 +..9 m 178ersz a “I I6 il.08 \CYTOCHROME \c OXIDASE 290$ .92 x _-z__2 .1 $40.22 6|? 5 m 5 O 5 n u a - GLYOXYLATE AMINOTRANSFERASE LEUCINE - _ r _ I 2 ...__2 x .22 x 8.62.). .To 298.. 70.2 x _-z_s_ x 8.622 AT? Or 5 O 5 2 L: x _.z_s_ x 84022 on? 200 300 400 500 600 GRADIENT VOLUME (ML) IOO 41 apparent in the main lysosomal fraction in less dense sucrose. These results were Obtained without the injec- tion of Triton W-1339 before aminal sacrifice. Gradient Separation of Particles The detection, isolation, and partial characteriza- tion Of liver peroxisomes has been accomplished. The detailed methods have been published (116,118). The zonal type rotor was chosen to purify the liver peroxisomes, becasue Of its large capacity (the B-30 rotor contains 560 ml). Several hundred ml Of tissue homogenate could be processed at one time, and it was not necessary to pellet the particles. Usually the low speed supernatant of a homogenate from several rat livers (a total Of approximately 25 g) was loaded directly on the B-30 rotor. Details are given in the materials and methods section. A typical representative isopycnic sucrose density gradient for rat liver homogenate is presented in Figure 1. Generally, about 50% of the catalase activity in liver homo- genates was located in the peroxisomal fraction, and about 40% was in the soluble fraction (Table 1). Cytochrome c oxidase in the gradients of rat livers was mainly (77% or more) found in the mitochondria. The catalase peak was found at a density of about 1.25 g x cc.1 and the cytochrome c oxidase peak was found at a density of 1.20 g x cc-l. The amount of protein in the peroxisomal fractions was very low 42 m mm m «H Hma om Ommummmsmuuocfiam mumameMHOIOOHOsOA o o comb mu mud m OmmOon O maounooumu Goa x 5.0 mm Goa x ~.o ma moa x m.H em mnmamumu HH ucmflomuw m em v Hm «ma me mmmummmsmuuosHE¢ mumamxomawlmcflcmamamcmnm v em e ma Goa me Ommnmmmsmuuosflam mumamx0>amlmswosmq o o ooom ms mm m ommOon O oEOHnoouwo OOH x m.o hm Goa x m.o m Goa x ma Hm mmmamumo H ucmwpmuw aflououm Hume samuoum Hume cwmuoum Hume x Hucwfi w x HIGHS w x HIGHS w x mOHOEs x meOEc x mOHOEc moshucm ucmumcnmmsm manoconoouwz m0§Omaxouom H0>HA umm .mumcmeEom mo ucmwpmnw waflmcwo wmouosm O OH moshusm no sowasnanumfloll.a Handy 43 so that the specific activity of its enzymes were very high. The peroxisomal peak fractions which were free of cytochrome c oxidase and low in acid phosphataSe were used to characterize the properties of the glyoxylate specific aminotransferase. Subcellular Distribution of the Glyoxylate Aminotransferase Reports have indicated that many aminotransferases are associated with microbodies in plant tissues (46,86,130). Little was known about the subcellular localization of the glyoxylate specific aminotransferases in mammalian tissues. Most of my studies were done with rat liver peroxisomes, but other sources Of peroxisomes were checked to determine whether all microbodies contained the aminotransferase. The glyoxylate aminotransferase found in rat liver peroxisomes used L-leucine and L-phenylalanine as best substrates (the substrate specificity will be discussed later). The distribution of L-leucine-glyoxylate amino- transferase and L-phenylalanine-glyoxylate aminotransferase on the gradients of rat liver homogenate, shown in Figure 1 and Figure 2, coincided with the peroxisomal marker-cata- lase. A distinct peak of transaminase activity was associa- ted with the peroxisomes. The percent distribution of this glyoxylate aminotransferase was very similar to that for catalase (Table 1). Fahimi has reiterated recently that 44 Figure 2.--Distribution of Phenylalanine-Glyoxylate Aminotransferase in a Sucrose Density Gra- dient. The total homogenate from about 25 grams of rat livers, after low speed centrifugation at 270 g for 10 min, was placed on a sucrose gradient and centrifuged for 3 hours. Phenyla- lanine-glyoxylate aminotransferase was assayed colorimetrically. 45 Ls. x _-z_s_ x $.62: ¢.9 8 4. 2 . . _ Z_w._.Omn_ T05. x_.2=2 x mMJOEZ .010. O :OO 8:528 m -l0 -IO 50 00 -|290 ~ |.|45 ‘ IOOO ' 6 CYTOCHROME c OX l DASE I'llll'lllu.‘ PHENYLALAINE-GLYOXYLATE AMINOTRANSFERASE GRADIENT VOLUME (ML) _ _ _ p _ _ 2 nlu 8 6 4 2 w _ O O D L: x .22 x 8.622 on? Ls. x _.z_s_ x $4022 I 290%. To: 1.2.: x 8.522 .7? 46 all the catalase was essentially located in the peroxisomes (25). He indicated that prolonged rinsing of glutaraldehyde fixed tissue in buffer would lead to the diffusion of cata- lase from peroxisomes into the cytosol and consequently to the artifactual staining of the catalase in cytosol. Using the assumptions that all the soluble catalase was from peroxisomal breakage, and that the solubilization factor for all the peroxisomal enzymes was similar, about 90% to 100% of the total leucine-glyoxylate aminotransferase and phenylalanine-glyoxylate aminotransferase would be in the peroxisomes. The low activity associated with mitochondrial fraction varied among experiments. Sometimes no peak Of activity could be detected in the mitochondrial fractions (Figure l). The variable amount of the glyoxylate amino- transferase activity in the mitochondrial fraction might be due to broken peroxisomes attached to mitochondria. The specific activity of the two aminotransferases in the peroxi- somes was about 30 fold greater than in the mitochondria (Table 2) where the activity was probably derived from con- taminating peroxisomes. The percentage of the total leucine- glyoxylate and phenylalanine-glyoxylate aminotransferase activities in each of the organelle fractions attributed to contamination by other organelles was calculated on the basis of the percent of activity Of the marker enzymes in each of the organelle fraction. An example of this calcu- lation: 47 TABLE 2.--Glyoxylate Aminotransferase Activities in Organellar Fractions. Specific Activity Organellar Fraction 1 . - -1 . nmoles x m1n x mg protein Leucine-Glyoxylate Aminotransferase Peroxisomes 181 Mitochondria 6 Phenylalanine-Glyoxylate Aminotransferase Peroxisomes 134 Mitochondria 4 Note: The particles were isolated from two different sucrose density gradients. Markers for the organelles were: Catalase for peroxisomes; Cytochrome c Oxidase for mitochondria. catalase in mitochondria transaminase in peroxisome x catalase in peroxisomes transaminase in mitochondria x 100 =,% of the total transaminase activity in mitochondria attributable to peroxisomes. Less than 1% of the total leucine-glyoxylate aminotransferase of phenylalanine-gly- oxylate aminotransferase activities in the peroxisomes could be attributed to mitochondrial contamination. How- ever, most of the aminotransferase activities in the mito— chondrial fractions were due to contamination by peroxisomes. 48 Thus it can be concluded that there is (are) glyoxylate specific aminotransferase(s) which used leucine and phenyl- alanine as substrates, and this (these) aminotransferase(s) is (are) located mainly in the rat liver peroxisomes. Aminotransferase Assay The products of the lecuine-glyoxylate aminotrans- ferase reaction are ketoisocaproate and glycine. Usually the measurement of transaminase activity is linked to a NADH or NADPH reductase which uses the keto acid product specifically. Since no commercial preparation of ketoisoca- proate dehydrogenase was available, this assay method could not be used. In 1966, Taylor (108) developed a technique to measure the aminotransferase colorimetrically by the formation of a-ketoisocaproate-hydrazone. The solubility difference of the two keto acid hydrazones in a cyclohexane- water mixture permitted the selective removal of the ketoisocaproate from the reaction mixture. Excess 0.3% 2,4-dinitrophenylhyrazine-HCl reacted with both the remain- ing glyoxylate and the product ketoisocaproate. However, only the phenylhydrazone of ketoisocaproate was selectively dissolved into the solvent cyclohexane. The hydrazone was extracted from the cyclohexane with a 20% sodium carbonate solution. An equal volume of 1 N NaOH was added into the carbonate solution for 10 min to allow the development of a red color for the hydrazone which was measured at 440nm. 49 Under this assay condition, 1.0 umole/1.2 m1 of a-ketoisoca- proate gives an absorbance of 2.13 (Figure 3). In the phenylalanine-glyoxylate aminotransferase reaction the assay procedure was the same as for leucine except for the substitution of phenylalanine as amino group donor. The phenylhydrazone of phenylpyruvate could also be dissolved into cyclohexane and then measured colorimetrically. In this case, 1.0 umole/l.2 ml Of phenylpyruvate gives an absorbance of 1.82 (Figure 3). This cyclohexane procedure can also be used to measure transamination from leucine or phenylalanine to pyruvate, ketoglutarate. hydroxypyruvate, and oxalacetate. The enzyme assay was linear with both time and enzyme con- centration (Figure 4). The glyoxylate aminotransferase activities can also be measured using a radiochemical procedure involving the formation of glycine-14C from g1yoxylate-U-14C. The reac— tion was linear with the incubation time and was dependent on the enzyme concentration (Figure 5). Boiled controls were done with different concentrations of protein, but the amount of glycine formation was almost constant (Figure 5). Since these data were consistent with a previous report (60) that glyoxylate could easily be converted into glycine by non-enzymatic reaction, a boiled enzyme control was used as a blank in every experiment. 50 Figure 3.--Standard Curves for the Colorimetric Formation of d-Ketoisocaproate and Phenylpyruvate. Each assay mixture contained from 0 to 0.8 nmoles a-ketoisocaproate or phenylpyruvate, 50 nmoles sodium perphosphate at pH 8.4, 9.1 umole pyridoxal—phosphate, and 0.04% Triton X—100 in a final volume of 1.2 ml. Mixtures were incubated for 10 min at room temperature before a—ketoisocaproate or phenylpyruvate determina- tions were made as described under materials and methods. 51 L6 - a-KETOISOCAPROATE 52 Figure 4.--Colorimetric Assay for Aminotransferase. Rat liver peroxisomes, isolated on a sucrose gradient were used. x x Leucine-glyoxylate aminotransferase reaction. Phenylalnine-glyoxylate aminotrans- ferase reaction. O O D440 iv “4:. 53 J J I 40 80 I20 _ TIME (MIN) 1 Q) 02 ENZYME (ML) 54 Figure 5.--Peroxisoma1 Glyoxylate Aminotransferase Radiochemical Assay. Isolated peroxisomal fractions from a sucrose density gradient were used. Leucine-glyoxylate aminotrans- ferase. --A ----- A-- Boiled Enzyme as blank. 55 1I6000 2 O. 3 I2000 >- t 3 8000 f— U 4 4000 TIME (MIN) 1 6000— 2 CL 8 4000- x >' x I: x 2 2000/ S b—A——A——A—-A-———A ————— <[ I , l 200 400 ENZYME (UL) 56 Substrate Specificities In leaf peroxisomes there are three different amino- transferases, glutamate-glyoxylate, serine-glyoxylate, and an asparate-ketoglutarate (86). Separation of these three aminotransferase proteins was achieved by isoelectric focus- ing. In aminal tissue, alanine-glyoxylate aminotransferase was first reported to be associated with mitochondria by Rowsell (95), but later a small percentage Of this enzyme was found to be associated with peroxisomes of rat liver (85). A glutamateeglyoxylate aminotransferase reaction was also Observed in the rat liver_peroxisomes by Vandor and Tolbert (124). In order to evaluate the reactivity of the rat liver peroxisomal leucine-glyoxylate aminotransferase, substrate specificities were tested by the assay conditions described in the methods section. Transaminase activity was measured between various amino acids and glyoxylate in the peroxisomal fractions from rat livers after separation by sucrose density gradient centrifugation. The most active transamination reaction was Observed between L-leucine, L-phenylalanine, or L—alanine with glyoxylate at pH 8.4 (Table 3). Little or no significant activity could be Observed between D-leucine and D-phenylalanine were not inhibitory to L—leucine or L-phenylalanine transamination activity. All these reactions involving_glyoxylate were essentially irreversible. There was no isotope exchange TABLE 3.--Specificity of Amino Group Donor for Glycine Formation by Rat Liver Peroxisomes. Amino Acid CPM figiztive L-Leucine 6400 100 L-Phenylalanine 6300 98 L-Alanine 5700 90 L-Histidine 4800 75 L-Proline 450 7 L-Tryptophan 260 4 L-Serine 1800 28 L—Glutamate 0 0 L-Threonine 400 6 L-Cysteine 300 4 L-Tyrosine 0 0 L-Aspartate 950 15 L-Asparagine 5300 82 L-Lysine 460 7 L-Arginine 190 3 L-Glycine 90 l D-Leucine 0 0 DL-Ornithine 400 6 L-Glutamine 3500 54 Note: Assay mixture contained 20 nmoles L- or D-amino acid or 40 nmoles of the DL form. corrected for boiled enzyme controls. Rate S were 58 with g1yoxy1ate-U-14C when glycine was used as the amino donor. L-Histidine was about 75% as active as L-leucine. L-Glutamate and L-serine, which had previously been used as amino donors in rat liver peroxisomes at pH 7.0 (124), were either not active under the assay conditions or, in the case of serine, had considerably less activity than some Of the other amino acids. Significant activity with the amides, asparagine and glutamine, were also noted. Since an L-alanine-glyoxylate aminotransferase from rabbit liver (26) was reported to utilize hydroxypyruvate equally well as glyoxylate as the amino acceptor, specifici- ties among different keto acids with either L-leucine or L-pheynlalanine as the amino group donor were tested (Table 4). The best amino group acceptor for the rat liver peroxi- somal transaminase was always glyoxylate. All other keto acids tested, e.g., a-ketoglutarate, hydroxypyruvate, pyruvate, and oxalacetate, were nearly inactive with the rat liver peroxisomes. Though the most active transamination activities were Observed with the three substrates, L-leucine, L- phenylalanine, and L—alanine, activity with any two sub- strates together such as leucine and phenylalanine, leucine and alanine, alanine and phenylalanine was not additive (Table 5) using either the colorimetric assay or radio- chemical assay. The results suggested that only one 59 TABLE 4.--Specificity of Amino Group Acceptor for Gly- oxylate Aminotransferase. Amino Group Acceptor Relative Rate L-Leucine Glyoxylate 100 Pyruvate 6 Hydroxypyruvate 6 Oxalacetate 3 Ketoglutarate 0 L-Phenylalanine Glyoxylate .100 Pyruvate 8 Hydroxypyruvate 8 Oxalacetate 4 Ketoglutarate 1 60 TABLE 5.--Additive Tests with Peroxisomal Glyoxylate Aminotransferase. Amino donor cpm by isotope assay L-leucine 23,500 L-phenylalanine 28,000 L-alanine 21,500 L-leucine + L-phenylalanine 23,500 L-leucine + L-alanine 26,400 L-phenylalanine + L—alanine 28,800 Amino donor nmoles/min, ml by colorimetric assay L-leucine 26 L-leucine + D-leucine 26 L-phenylalanine 22 L-phenylalanine + D-phenylalanine 20 L—leucine + L-phenylalanine 26 aminotransferase was catalyzing all 3 reactions. Amino- transferase activities with amides or serine were not further investigated. Subcellular Distribution of L-Alanine-Glyoxylate Aminotransferase Subcellular organelles were prepared from rat livers by our sucrose density centrifugation procedure. As indicated in Table 6, leucineéglyoxylate and phenylalanine- glyoxylate aminotransferase activities were recovered to a greater extent in peroxisomes than in mitochondria. This 61 .uxwu may as szonm mnw3 cowumswsmucoo psmonmm mnu mo mcoflumasoamo .NHHOOHEmnOOHpmH ommmmmm mums mowufi>fiuom Ommnmmmcwuuosaem mane ”muoz on mp Ha v.o mm mumamxomawImcfloomA me me an 4.0 mm mumasxosao ImaflsmamHmcmsm mm 4 mas m.m mm mumstosHo Imcwsmad cowumcwfimucoo cowumcwsducoo MIOH x Emu HMEOmAxOHOm MIOH x Emu HMHHOGOSOODHZ MIOH x emu mmmuwmmcmuuocfias usuumsuwmsm mauoconoouwz mmEOmwaHmm .mmaoauumm macaw mmwua>fluod Ommummmsmnuocflad mumameMHw mo sofiusnfluumaoII.m mnmda 62 is consistent with the data shown in Table 1. However, a smaller percentage of the alanine-glyoxylate aminotrans- ferase activity was in the peroxisomes than in the mito- chondrial fraction. The data suggested that a major per- centage of the alanine-glyoxylate aminotransferase activity was in the mitochondria although some may be in the peroxi- somes. It is not clear whether the small amount of alanine- glyoxylate aminotransferase in the peroxisomes is signifi- cant. Its presence may be accounted for by the lack of amino donor specificity of an aminotransferase whose primary donor is other than alanine. Snell et. al. (18,32), Rowsell (95), and Refheld (85) have also concluded that most particu- late alanine-glyoxylate aminotransferase activity was associa- ted with the mitochondria in rat liver. Glycine Product Identification To follow the course Of glycine formation, amino- transferase experiments with l4C-U-glyoxylate were stopped in ice-cold 50% (v/v) ethanol. Controls contained boiled enzyme. After centrifugation, the supernatant was applied to a Dowex-l-acetate column (0.6 x 3 cm) to absorb excess glyoxylate but not the amino acids. The columns were washed with 3 ml distilled water. The effluent was evaporated to dryness at room temperature and redissolved in 0.6 ml of 0.02 M glycine. Samples Of 2 01 were develOped in n-butanol/ acetic acid/water (60:20:20, w/w/w). The standard amino 63 acids were located with ninhydrin, and spots Opposite stand- ard amino acids were collected and suspended in 10 ml Bray's solution. The samples were then counted in a Packard scin- tillation counter (Table 7). Only traces of radioactivity were found in the spot corresponding to leucine. Radio- activity was found at the spot for glycine. The boiled control sample has no labelled glycine. Similar results were Obtained in 4 experiments. TABLE 7.--Glycine Product Identification. Compound Rf Radioact1v1ty cpm Leucine 0.45 -- Phenylalanine 0.48 -- Glycine 0.154 -- Leucine-Glyoxylate Reaction Mixture 0.44 22 0.135 909 Phenylalanine-Glyoxylate Reaction Mixture 0.48 27 0.135 897 Boiled Control with Leucine 0.44 17 0.13 59 Note: Reaction samples were chromatographed on silica gel thin layer plates with n-butanol/acetic acid/ water (3:1:1). Stimulation of Peroxisomal Amino- transferase Activity by Triton X-100 Except for the aminotransferase assay (129), 0.15% Triton x-100 has been routinely used in our other assay 64 \ media to disrupt the peroxisomal membrane (116,118). Since both spinach leaf and rat liver peroxisomal aminotrans- ferases for glutamate-glyoxylate, alanine-glyoxylate, and serine-glyoxylate had been found to be inhibited by 0.01% Triton x-100 (85,86), it was necessary to investigate how Triton X-100 affected the other glyoxylate aminotransferase reactions in rat liver peroxisomes (Table 8). When 0.04% to 0.16% Triton X-100 was included in the reaction mixture, there was a 80% increase in detectable leucine-glyoxylate aminotransferase activity in the peroxisomal fraction as compared to the assay without this detergent. This 80% increase in activity of leucineéglyoxylate aminotransferase by the presence of a non-ionic detergent, Triton x-1oo, though different from other aminotransferase, , was con- sistent with the 30% increase in glycolate oxidase, malate dehydrogenase and NADéglyoxylate reductase activities (129). This Triton stimulation effect on lecuine-glyoxylate amino- transferase was different from the effect on alanine-gly- oxylate aminotransferase in rat liver peroxisomes; the latter was inhibited by Triton X-100 (85). The data suggest that rat liver leucineéglyoxylate aminotransferase is located inside the peroxisomes which must be broken up by the detergent before assaying. 65 TABLE 8.--Triton X-100 Release Of Leucine-Glyoxylate Amino- transferase from Rat Liver Peroxisomes. Triton X-100 Concentration Enzyme ACthltY % nmoles x min-1 x ml-1 0.00 28 0.004 44 0.04 50 0.08 48 0.16 49 Note: Similar results were obtained in 3 experiments. Choice of Buffers It has been reported that serineeglyoxylate amino- transferase in both spinach leaf and rat liver peroxisomes was inhibited by phosphate buffer (60). However, leucine- glyoxylate aminotransferase and phenylalanine-glyoxylate aminotransferase was not inhibited by phosphate (Table 9). In fact, they were stimulated about 40% by phOSphate or pyrophosphate buffer as compared to glycylglycine buffer. This stimulation may be related to the fact that 0.01 M. pyrophosphate is used to rupture peroxisomes (23), and in doing so would release the bound aminotransferase. Glycylglycine or phosphate buffer at pH 7.8 has been used in subsequent dialysis studies because no loss of activity was Observed after dialysis overnight in either of the buffers. With perphosphate buffer at pH 8.4 or 66 TABLE 9.--Buffer Effects on Leucine-Glyoxylate and Phenyl- alanine-Glyoxylate Aminotransferase Activity. Leucine-Glyoxylate Aminotransferase Buffer nmoles x min‘l x ml'1 relative rate Glycylglycine pH 7.7 8.3 39 Glycylglycine pH 8.6 14.5 68 Phosphate pH 7.8 20.2 96 Pyrophosphate pH 7.7 18.2 86 Pyrophosphate pH 8.6 21.2 100 Phenylalanie—Glyoxylate Buffer Aminotransferase nmoles x min"l x ml'1 relative rate Glycylglycine pH 7.9 6.4 41 Glycylglycine pH 8.5 9.4 61 Phosphate pH 7.8 15.1 97 Pyrophosphate pH 7.7 14.6 94 Perphosphate pH 8.6 15.7 100 Note: Rat liver peroxisomes isolated in a sucrose gradient were dialyzed overnight against 0.02 M glycylglycine buffer at pH 7.5 before assaying in the 0.1 M designated buffer. 67 Tris-citrate buffer at pH 8.3, 80% of the original activity was lost after overnight dialysis. Heat Stability of Glyoxylate Aminotransferase Nakada (61) has reported that glutamate-glyoxylate aminotransferase from rat liver could be heated to 60°C for 5 min without loss of activity. Tompson and Richardson (109) partially purified this aminotransferase from human liver by heating the preparation at 55°C for 10 min. A similar heat stability was Observed for the alanine-gly- oxylate aminotransferase from human liver (110) and in fact a 20% increase in the activity of this aminotrans- ferase was found by treatment at 60°C for 10 min with 5 uM exogenous pyridoxal phosphate (110). However, a different case has been found in plant tissue. A 50% inhibition of serine-glyoxylate aminotransferase from spinach leaf peroxisomes was observed by heating at 50°C for only 5 min (86). The leucineéglyoxylate, phenylalanine-glyoxylate, and alanine-glyoxylate aminotransferase reactions from rat liver peroxisomes were also found to be relatively heat stable as are other mammalian liver aminotransferases. All three of these activities were stable for 60 min at 50°C (Table 10). In addition, all three activities had the same inactivation profiles at different temperatures (Figure 6). The reaction rate increased about 10 to 20% 68 .mmumnquOm mo mDGOOEm can .Oafl» cowumnsosw mo cpmcma .mshncm mo unsosc assoc cm cufi3 nonsmmofi mum3 ammufl>wuom Ommummmcmuuocflam Had no mumsmmocm ZEcH umcflmom ucoficuo>o Ocuwamac muo3 mwfiomfixouwm cmumHOmH wumamxomauImsacde mumame>HwImcflcmamahcwcm mumamxomawImcfloomq comm comp coon cc comm coon cows co comm coco ccHh cm comm cccb comb ca comm comb coon c Emu Emu Emu Ommuommcmuuocflfid mmmuommsmuuocflad mwmummmcmuuocflam Acfiec Uoom um mafia ucmsummua .mommumwmsmuuocafid mumaaxOMHo HMEOMonumm mcu mo muwaflnmum ummmII.cH mqmda 69 Figure 6.--Temperature Dependence of the Peroxisomal Gly- oxylate Aminotransferase Activities. Isolated peroxisomes from rat livers were heated at indicated temperature for 10 min, then cooled immediately to room temperature in an ice bath before the reaction was started. Leucine-Glyoxylate Aminotrans- x x ferase ----A----A---— Alanine-Glyoxylate Aminotrans- ferase 0 go Phenylalanine-Glyoxylate Amino- transferase ACTIVITY (CPM) 4000 3000 2000 IOOO 70 l L l L 20 4O 60 80 I00 TEMPERATURE (°C) 71 between 23°C and 50°C after which point it sharply declined. The significance of the increase in activity between 230 to 500 is not understood. Inhibitor Studies Various compounds known to react with the active site of several aminotransferases were tested as inhibitors of the leucine-glyoxylate and phenylalanineéglyoxylate aminotransferase. All inhibitors were preincubated with the enzyme for 20 min at 370 before adding the substrate to initiate the reaction. p—Chloromercuribenzoate at 4M, hydroxylamine at 10-3 10- M were reported to inhibit alanine-glyoxylate aminotransferase (110) and glutamate- glyoxylate aminotransferase from human liver (109) and from rat livers (61). Similar results were obtained in the present studies for the rat liver peroxisomal enzyme. p-Chloromercuribenzoate and glyoxal were the most potent inhibitors. They inhibited the enzyme activity 70% at 0.1mM (Table 11). Hydroxylamine and isonicotinic acid hydrazide at 10-3 M also inhibited these glyoxylate amino- transferase activities. Cu++, KCN, oxalate, and oxamate did not inhibit the enzyme activity at all. Glyoxal, which has been shown to react preferentially with lysine and arginine residues (30), inhibited alanine-hydroxy- pyruvate transaminase from rabbit liver (26) at 50 uM 4 concentration. This compound at 10- M concentration 72 TABLE 11.--Effect of Inhibitors. Leucine- Phenylalanine— Inhibitor Concentration S$§::{::::_ Ag;{::{::§:_ ferase ferase M Relative Activity None 100 100 p-Chloromercuri- _3 benzoate l 10 0 0 N-Ethylmaleimide 1 10"3 75 -- Cu++ .1 10'3 90 32 Hydroxylamine 1 10'3 33 48 KCN 1 10"3 98 120 Oxalate 1 10‘3 94 128 Isonicotinic _3 a01d hydraz1de l 10 50 50 Urate l 10-3 95 -- NH4+ 1 10"3 68 -— Glyoxal 1 10"4 37 -- Phenylhydrazine l 10-3 10 -- Note: Inhibitors were preincubated at 37°C for 20 min.with rat liver peroxisomes from a sucrose gradient. 73 inhibited 68% of the leucine-glyoxylate aminotransferase. Ammonium ion at 1mM, inhibited 90% of the serine-glyoxylate aminotransferase activity in both spinach leaf peroxisomes (85) and kidney beans (103). Ammonium ion at 1mM only inhibited leucine-glyoxylate aminotransferase from liver peroxisomes 33%. This effect may not have physiological significance. All the numbers shown in Table 11 were the average from two experiments. Evaluation of Pyridoxal Phosphate The addition of pyridoxal analogues (pYridoxal-5'- phosphate, pyridoxal, pyridoxamine, and pyridoxamine-5'- phosphate) to liver peroxisomal preparations did not affect the rate of leucine-glyoxylate aminotransferase activity (data not shown) in most experiments. Prolonged dialysis, with or without pyridoxal phosphate, of the isolated rat liver peroxisomes did not inhibit leucine-glyoxylate amino- transferase (Table 12). When the enzyme was extensively dialyzed without 100 uM pyridoxal phosphate, about 25% of the activity was lost when the enzyme was stored at -20°C for three weeks as compared to the enzyme dialyzed with 100 uM pyridoxal phosphate and stored at -200C for three weeks. Heating the enzyme preparation at 50°C for 60 min without exogenous pyridoxal phosphate produced no loss of activity (Table 9). All of these results were inconclusive in demonstrating the presence of pyridoxal phosphate in 74 mucommummu Hogans comm mammamwc Hound consumms mm3 hufl>fluom .musmafiummxm 03¢ mo momnm>m may .UooNI um mmmuoum “mums Hound mxmms woman can .omms mos .mumnmmonm meocfluhm 28H.c HoocUHs HO nuwz .GOfiumnmmem Hmeomwxonmm Hw>HH umn oo~>HMflo "muoz mm mm oumcmmocm Hoxoofinwm usonufls mmmuoum can mHmMHMHn Hound NHH mad mumcmmocm meoowumm cufi3 wmmuoum can mdmhamma Hound cad mHH mumcmmonm meocflnmm escapes nansanao “mums «Ha mad mumcmmocm anxoeahsd sues nansaman “mama ccH cod mammamao OHOmom w w Ommummmcmuuoaflsm wwmummmsmnuosfls< uswsummue mumamxomHOI mcfiqflnmamcmnm mumahxomaoImswonmq .nflmsamao can mumsdnonm meocanmm mo uommmmuu.ma mamas 75 the enzyme. However, the carbonyl-binding reagent, phenyl- hydrazine inhibited the leucine-glyoxylate aminotransferase activity 90% at 1mM concentration (Table 11). Also the observed inhibition by hydroxylamine and isonicotinic acid hydrazide suggested the presence Of pyridoxal phosphate as an integral part of leucine-glyoxylate aminotransferase active site. Nevertheless because the aminotransferase cofactor is pyridoxal phosphate, it was routinely added during purification steps along with a sulfhydryl reagent. Separation of Glyoxylate Amino- transferase Activities by Gel Filtration Chromatography Several attempts were made to establish whether one or more aminotransferases were responsible for the glyoxy- ' late reaction with different amino acid donors. One of them was to try to separate the proteins by sephadex column chromatography according to their molecular weight differ- ences. Peroxisomes, isolated by isopycnic sucrose gradient centrifugation from rat livers, were dialyzed overnight against either 10mM potassium phosphate or 20mM glycyl- glycine buffers pH 7.8 containing 1mM EDTA, 1mM B-mercapto- ethanol and 100 uM pyridoxal-5'-phosphate. After dialysis, no loss Of activity occurred; in fact, a 20% increase was always noted after removal of the sucrose (Table 13). The 76 .hpfi>fiuom mmmummmcmnuoswsm mumamxomHmImcHOst coma comma mum mumc whose; NH m.eom Hmm maoauonuu omauo xwtmndmm ea m.m~ was banana emunuucmoaoo canned ma m.m~ New madsnn tmunmm «a v.c~ mac mmEOmeOHmm Omumamao NH >.mH cam msofluomum HmEOmwxoumm ccH m.m Hmvv mumcmmoeom cflmuoum aum>oomm w on HI huw>fluod owmwommm awe x mmaoes awe x mOHOEs huw>auod Hmuoa mmum coaumOHMHHsm t.sowumowmwusm Oaxaca ammuumm mo mHmEEsmII.mH wands 77 preparation was then heated at 50°C for 30 min, and immedi- ately cooled to 4°C in an ice bath. The heat-treated sample was concentrated by ultrafiltration (Diaflo ultrafiltration cell, Amicon Corp.) through a PM-30 filter which has a general retention of molecular weights over 30,000. About 85% of the activity remained in the cell. After treating with 0.01% Triton x-1oo for 5 min at 0°C, the concentrated sample was placed on a Sephadex G-150 column which had been previously equilibrated with the dialysis buffer contain- ing 10-4M pyridoxal phosphate. The enzyme was chromato- graphed with the same buffer at a flow rate Of 8-10 ml per hour. Fractions of 3 ml were collected (Figure 7). About 80 to 90% of the total activity was recovered from the sephadex column. The addition of Triton x—1oo was necessary; otherwise, part of the aminotransferase activity would be found in the void volume from the column, as if the enzyme was still bound to large particulate fractions. All three glyoxylate enzyme activities with leucine, phenylalanine, and alanine coincided exactly with one another in fractions from the Sephadex G-150 column chromatography, although the enzyme was separated from the major protein peak (Fig- ure 7) . The ratio. between any two aminotransferase activi- ties was not changed significantly. After separation from Sephadex column chromatography, almost a 100-fold purifi- cation was achieved (Table 13). This probably was due to 78 Figure 7.--Elution Profile Of the Peroxisomal Glyoxylate Aminotransferase by Gel Filtration Column Chromatography. Rat liver peroxisomes, isolated on a sucrose density gradient, were dialyzed against 10mM phosphate buffer pH 7.8 containing 1mM EDTA, 1mM B-mercaptoethanol and 0.1mM pyridoxal phosphate, then heated to 50°C for 30 min, and immediately cooled in an ice bath. The heat-treated peroxisomal sample was broken by 0.01% Triton X-100 and placed on a G-150 sephadex column which had been previously equillibrated with the dialysis buffer. The enzymes were also eluted by the same buffer. All these aminotransferase reactions were assayed radiochemically. x x 280nm O.D. reading for proteins. A————A Leucine-Glyoxylate Aminotransferase. o————o Phenylalanine-Glyoxylate Aminotransferase. ------- Alanine-Glyoxylate Aminotransferase. :2ro >C>PU< O 1 m 11 5 m m . % XI X X o (IIILIIIeIIIleIIIII:IIsI.II o r'l'."' X” m m 9 ' ' 'l' X B 7 "'e-o' 66m M m.» U - 2m N x a I m N X/ I T .. C b m . F 6 A 2 Om N o o 117 78 Figure 7.--E1ution Profile of the Peroxisomal Glyoxylate Aminotransferase by Gel Filtration Column Chromatography. Rat liver peroxisomes, isolated on a sucrose density gradient, were dialyzed against 10mM phosphate buffer pH 7.8 containing 1mM EDTA, 1mM B-mercaptoethanol and 0.1mM pyridoxal phosphate, then heated to 500C for 30 min, and immediately cooled in an ice bath. The heat-treated peroxisomal sample was broken by 0.01% Triton X-100 and placed on a G-150 sephadex column which had been previously equillibrated with the dialysis buffer. The enzymes were also eluted by the same buffer. All these aminotransferase reactions were assayed radiochemically. x x 280nm O.D. reading for proteins. A——-—A Leucine-Glyoxylate Aminotransferase. o————o Phenylalanine-Glyoxylate Aminotransferase. ------- Alanine-Glyoxylate Aminotransferase. 79 - I5OO canoe >E>PO< FRACTION NUMBER 80 the fact that the bulk of the peroxisomal protein, as catalase, had been coagulated by the heat treatment. Of particular importance, however, was the fact that after Sephadex G-150 fractionation the enzyme was very unstable. Although 80 to 90% Of the total aminotransferase activities put on the Sephadex G-150 column were recovered, the enzyme activities were lost actively and could not be stored or further handled after separation from the bulk of the pro- teins. Several attempts were made to maintain this enzyme activity. Addition of pyridoxal-5'-phosphate into the elution buffer on Sephadex G-150 column was unsuccessful. Dithiothreitol at 2mM, 50% sucrose, 10mM B-mercaptoethanol, or 1mM L-leucine was also added into the Sephadex fractions which contained the highest leucine-glyoxylate aminotrans- ferase activity, but none of these stabilized the enzyme. When fractions containing the most enzyme activity from Sephadex columns were pooled and concentrated by Diaflo ultrafiltration cell, additional activity was lost. There- fOre, in the subsequent ion exchange separation, heat- treated peroxisomal enzyme preparations were put on the columns instead of fractions from the Sephadex columns, because not enough activity could be maintained for cellu- lose ion exchange chromatography after the Sephadex steps. Polyacrylamide gel electrOphoresis of the 100-fold purified enzyme from Sephadex G-150 column showed many protein bands (data not shown). 81 .camuonm was no 0E5H0> coauoao on» ma m> .wEsHO>. cao> may we o> .msaao> can anuop 0:» ma u> “coaumswm wcp Eoum omumasoamo was M .hHHmOHHumEHHOHOO OOMMmmm wok muw>wuom mmwuom Imcmuuocflsm mumathMHmImcHOsmq .ucmflm3 HMHDOOHOE czocx mo mason Iona umcuo on cwummfioo mos GEOHOO cmHIw xwomcmmm nosoucu muaaflnoe m>flumamu mus com ma magma CH confluommc mm cmwmausm mm3 mshncm one .mEomfixouwm Hm>flq pom Eoum wmmummmsmuu Iocflfid mumamxomeImcHOsmA mo unmflmz MMHDOOHOS map mo sOHumEHummII.m whomwm 82 x m. m. / _ _ O w20m100h>o wmrmo 4010044 mmfluom mmmummmsmuuosafid mumamxomHUImsflosmA XIIIIIIxI .cwwuoum How mowcmmn .o.o Escmm II.III.II .z~.c Op 2 Hc.c Bonn mumcmmocm mo ucmwcmum HMOGHH m cuss uwmmsn on» cufl3 Oman cousam mums mmemucm was .Hmmmsn mammamflc meow on» Spas Omumunflaaflovm MHmOOfl>OHm soon to: coac3 cEsHOo OmOHsHHOOIm¢mo o co Omomam mos cam ccHIx confine wac.c an OONHHfinoHom mos Manson amEOmaxOHmm cmumouulummn one .nvmn OOH so as OOHOOO mawsmwcmssfl com .cafi cc How Oocm on cmumoc cmcu can .wpmsmmosm meocwumm SEH.c cam HocmcumoummouOEIm 28H mcwcamucoo m.> mm Hmmmsn mumcmmocm SEcH umcflmmm commamflo mums Dawsomum mufimcwc wmouosm o co cmumHOmH .mOEOmeoumm HO>HH pom .wsmmumoumeouco sEsHOU mmoasaamo mama ma mmmummmcmnuosafid mumamxomaw HmEOmwxoumm mo maamoum GOHDSHHII.m madman 85 W[31VHdSOHd]——- so 09300 NIBlOtId +-+ n N. I T l 1 l N ,rOl x was so ””00 80 FRACTION NUMBER 86 two possibilities; (1) Most of the alanineeglyoxylate aminotransferase reaction was probably catalyzed by a dif- ferent protein than the other aminotransferase reactions. (2) The activity Of alanineéglyoxylate aminotransferase may be due to the nonspecificity of transaminase in the peroxisomal fractions. More evidence is needed to distin- guish between these possibilities. From the chromatog- raphy results with Sephadex G-150 (Figure 7) and DEAE cellulose (Figure 9), it is concluded that leucine-gly- oxylate and phenylalanine-glyoxylate aminotransferase activities are probably catalyzed by the same enzyme. PH Optima The peroxisomal leucine-glyoxylate aminotrans- ferase and phenylalanine-glyoxylate aminotransferase had similar maximum activity around pH 8.4 in pyro— phosphate (Figure 10). The activities in phosphate buffer were greater between pH 7 to 8 than in perphosphate for reasons unknown. Both transaminase activities were lower in glycylglycine buffer than in perphosphate buffer (Table 9). Alanine-glyoxylate aminotransferase had a simi- lar pH activity profile but the Optimum was about 8.6. The pH Optimum for alanine-glyoxylate aminotransferase was the same as reported in rat liver mitochondria by Rowsell (94). 87 Kinetic Parameters Double reciprocal plots of the initial velocity versus the concentration of the amino donor for a series Of fixed concentrations of glyoxylate are presented in Figure 11. The assays were performed colorimetrically as described in the methods section, except that lower sub- strate concentrations were used. These double reciprocal plots were linear and parallel indicating that the reaction followed the ping-pong mechanism, i.e. the first product was formed prior to reaction with the second substrate. The Km for glyoxylate was 0.5mM at a fixed concentration of 16.7mM leucine (Figure 12), and 0.76mM with 16.7mM phenylalanine (Figure 13). At a fixed glyoxylate concen- tration of 2.08mM, the Km for leucine was 2.5mM, and the Km for phenylalanine was 2.8mM (Figures 12 and 13). Thus the Km values with these two amino acids are similar. The concentration independent constants can be derived by secondary plots from the Michaelis-Menton equa- tion. In Equation 1, a and b are substrate concentrations, ka and Kb are Michaelis constant, and V'is the maximum velocity at a given enzyme concentration. By replotting the vertical intercepts versus the reciprocal of the Ka Kb 1 + —— +‘—— ° ' ° ° - ° - - - - - - a b 1 <|<' II 88 .xamaauommnmu sauce x He.a cam ZvIOH x mm.o mo chHumuucwocoo mumexomHm nuHs mmmnmwmcmuuocHEm ODMH>XO>HmImcHOOOH How mum m can 0 .0 mo>uso .xHo>Huowmmou OHMH>xO>Hm SMIOH x mo.m cam EvIOH x m.m LDH3 mmmummmcmuuosHEO OHMH>xO>Hm IOCHGMHmHmcmnm Mom mum m was d mm>uoo .OmmnwwmcmuuocHE¢ mummeome IOchmHmecmzm com wmmuommcmuuocHE¢ oumexoxHuIOCHosoq now coHumHu Icmocoo mumuumndm umchmm wuHOOHm> HnHuHaH mo whose anoondHomm manaooII.HH masses Emu mm .ummwsn mumsmmonm Iouhm CH >HH>HuOm HE\CHE\mOHOEc mm .uwmmsn oumcmmonm Iouwm :H >DH>HuOm as}? \mOHOEc mm .ummmon mumnmmosm cH qu>Huom OIIIIIO .mHHMOHEwso IOHOMH cwusmmme was >uH>Huom OHMHhxomeIOOHcOHm cam .mmuscoo Iona HmOHquEHHOHOO uh cousmmms mums mOHuH>HuOm wmmuwmmcmquCHEm mumexO>HmImchmHMH>cm£m cam ODOH>XO>HmIOcHODOA .Omms mums mucchmuo muHmcoc mmOHOSM unmumm IMHO Scum moEOmeouom um>HH pom .meOmeoumm um>HA umm EON“ mmmummmcmuuocHE¢ munflsxosao we» now sasHudo mmuu.oa enemas 89 .JW ",,NIW X SBWMN/ I SSVUBJSNVUIONIWV 3NIOOB'I IO I/[AMINO ACID] mM 5 O DIN 1“,le x SB‘TOWN/ I BSVBBJSNVUIONIWV 3NlNV‘IV1AN3Hd WdD PHENYLALANINE- GLYOXYLATE /’-\ I I" ALANINE- §§§ GLYOXYLATE GLYOXYLATE LEUCINE - 1 I Figure 10 20> . 8 3 '_1W x |_NIW x sa'IowN .jIw x I,NMI x snow Figure 11 90 .ommuwmmcmquCHsd OHMwaowHoncHOsmq .OmmummmcmquCHE¢ mumexowHw ImchmHmecmcm you GOHumuucmocou Mom :OHumuucmocOU oumuum munuunham ma suaooam> HanHaH Ihsm ma suHooH0> HmHuHaH mo uOHm HmooumHomm OHQOOQII.mH musmHm mo muon HmooumHomm OHASOQII.NH musmHm 91 MH musmHm 22Hu2_z<._<._>zurd<_ o. m H O. m H d .2 E who "Ex L a) 0. mm. NH musmHm .25 $2.83.: N _ O II4( q n A m m 1 mo. 3 S X m N. XI . w .25 USfixofioS .o.. m a e o n 8. A m w 0 HI 3 S x o.. m NP x w .IP 3. (,le x |,NIIII X SB‘IOWN) All (.31w x ,,NIw x SEI'IOINN) A/l 92 changing fixed substrate concentrations, -l/Ka and -l/Kb can be read from the abscissa and l/V is the intercept on the ordinate. The Km values obtained in this way were similar to those reported in Figures 15 and 16. Substrate inhibition was observed for this peroxi- somal glyoxylate aminotransferase at high concentrations of glyoxylate. The activity with either amino acid was severely inhibited by increasing the glyoxylate concentra- tion over 4mM (Figures 12 and 13). A 49% inhibition was Observed at leM concentration Of glyoxylate. NO substrate inhibition was observed at high concentrations of either leucine of phenylalanine. Location of Glyoxylate Aminotransferase within the Peroxisomes To rupture isolated rat liver peroxisomes (22,23,51), the fraction with 50% sucrose from the sucrose gradient was diluted 1:1 with 0.01 M pyrophosphate. The final pH was about 8.5. After storing overnight at 4°C, the sus- pension Of broken peroxisomes was separated on a sucrose step-gradient which was designed to separate the broken peroxisomes into the soluble matrix, membrane, and the dense core material (22,23). Fractions Of 2.5 ml were collected from the bottom of the 50 ml tubes. The amino- transferase activity with either phenylalanine or leucine was in the solubilized tOp fraction along with catalase 93 (Figure 14). Urate oxidase which is known to be a compo- nent of the peroxisomal core, sedimented into the denser sucrose portion; it was accompanied cytochrome c reductase located in limiting membrane Of the peroxisome. Therefore, in rat liver peroxisomes, the glyoxylate specific amino- transferase which utilizes either leucine or phenylalanine is located in the soluble matrix of the organelle. Reversibility of the Aminotransferase Reaction The reaction was run rat liver peroxisomal prepara- tions with 14C-U-glycine and a-ketoisocaproate of l4C-U— glycine and phenylpyruvate as starting substrates. Reactions were stopped by adding tri’chl‘oroacetic acid. After centrifu- gation, the supernatant was run through a Dowex-50-H+ column to absorb excess l4C-glycine. Each column was washed with 3 m1 0.01N HCl to elute any 14C-glyoxylate, 14C in a Packard and the effluent was counted for scintillation counter. NO detectable activity in the reVerse direction was Observed even though the incuba- tion was carried out overnight. The amount of 14C in the acidic fraction from the resin was the same in both enzymatic and non-enzymatic reactions. Several investiga- tors have reported similar results for other glyoxylate aminotransferases in animals (26,92-94) and in plants (46, 124,130). All these enzymes were very active in the for- ward direction, but no transamination was Observed in the 94 Figure l4.--Location of Leucine-Glyoxylate Aminotrans- ferase and Phenylalanine-Glyoxylate Aminotrans- ferase in the Soluble Matrix of Rat Liver Peroxisomes. Seven ml of a rat liver peroxisomal suspension, isolated on a sucrose gradient, was diluted with 7 m1 of 0.01 M pyrophosphate. The final pH was about 8.5. After being stored overnight at 4°C, the suspension of broken peroxisomes was put on a sucrose gradient of 5 ml each of 56%, 49%, 47%, 45%, 42%, 38%, and 30% sucrose (w/w). The grad- ient was centrifuged for 3 hours at 23,000 rpm in the SW-25.2 rotor. Fractions of 2.5 ml were collected from the bottom of the tube. ion, with was I 4°C, .t on III, grad- were 95 F‘- Lfi‘"! ‘20 I I... T I.22’- t-L'LDIO. S] ,1 'I ' '- “l6 x 0 1 Z x |.|8- L1 5 (9 I v ‘1 “|2 x C . m (7, |.I4" L L3 2 PHENYLALANINE- O 8 GLYOXYLATE -8 E : ,,O_ AMINOTRANSFERASE I -4 106» T_, T 2 '20“ -I.2 3' X _ _x 'g CATALASE Z 2 x 80“ CYTOCHROME c “.8 E U, REDUCTASE m “J Lu J a O 40- '.4 2 E 2 Z l' I JJJ I P_——"'r-P 1“: I00) LEUCINE- 42° GLYOXYLATE _ T {T AMINOTRANS- '3: '2" 80' I FERASE 45 x r" —.- x URATE | g T OXIDASE J 2 Z S 60' I. .2 x x I a E I a _ 2 I 40" rJ 8 z I ' I ' I J 20- I _ g 4 {'1 J' 'g ' L_-J'" L-l" ”J I l—‘ ~‘5'LH1_L IO 20 FRACTION NUMBER 96 reverse direction with glycine and the apprOpriate a-keto acid as substrates. Postnatal Development of Leucine- Glyoxylate Aminotransferase Recognizable changes in peroxisomal number, morph- ology, and enzyme composition have been detected in the tissues Of various animals during development (24). Fetal mice contain about 1/10 of the catalase activity in adult mice (38). Fetal rat liver peroxisomes contained no detec- table core, and the matrix appeared to be low in electron density. The number of peroxisomes in rat liver tissue increased with age; 5- to 10-day Old rats contained less peroxisomes in the liver than those of 40-day Old rats (120). For developmental studies on peroxisomal leucine- glyoxylate aminotransferase, livers from 3 to 12 nonstarved rats of differnet age were used (Table 14). After removal of the cell debris at low speed centrifugation, the total organelles were pelleted at 2,700 g for 30 min and then resuspended separately in 20mM glycylglycine pH 7.5 and 8.5% sucrose. Leucine-glyoxylate aminotransferase activity was assayed both in the organell rich fraction and in the supernatant fraction. Two days after birth, the leucine- glyoxylate aminotransferase activity was so low that it could not be accurately measured. After 15 days the 97 .mumbHH mo 0 mm now on amono CH poms mums mums O>Hosu on mouse nouoz men mam II- II- as mmm mmm o.m h.mm ow mwm NMH m.o m.m Hm omH cc N.m v.H mH I OHnmuOOUOOIcOG I c.c H.N m mammHu HIm x HIcHE x mOHosc wommHu HIm x HIsHE x mmHOEE HOHHOQ usmumcummsm HOHHmm uswumcummsm 0mm MMMMWWmom mmmummmcmuuocHE< mmMHmumo OHOmeowHwImcHOSOH .OmmnmmmsmuuocHad ODMmeOMHwImcHOOOH mo ucwEmOHm>mn HmumcumomlI.vH wands 98 activity increased dramatically and reached maximum level when the animals were about 40 days Old, after which this high level was maintained. This slow development of a peroxisomal aminotransferase activity during growth is similar to that reported by Tsukada (120) for urate oxidase and catalase. Effect of Starvation of Rats on Leucine- Glyoxylate Aminotransferase Activity In most of the studies, the rats were routinely starved overnight in order to lower the glycogen level (27) which allowed better particle isolation. However, the leucine-glyoxylate aminotransferase was most active in livers of nonstarved rats. The activity decreased a dra- matic 50% after one day of starvation when the data is expressed on the basis of specific activity or total activity (Table 15). These decreases occurred in the first day of starvation when the liver protein content had not decreased although a slow reduction in body weight and liver weight had occurred. After the initial first day drop in this aminotransferase activity, the activity remained low through the fourth day of starvation but increased markedly on the fifth day, beyond which the experiments could not be run because of animal deaths. An explanation for the increase on the last day could be termi- nal protein catabolism for survival which led to the 99 .mucmEHummxm Ozu cH omsHmuno mums muHsme HMHHEHm .mHunmc HHOO Obosmn on sHE cH you c chm um oomBMHnucwo was mumsmmosoa was .Omouosm wm.m cuHs m.h mm mcHoaHmHmomHm 28cm CH OONHcmmosoa mums mums %OH3maImsmmumm OHMEOH Om>nmum Eonm mHO>HH «muoz om om mob m.m ocH m HN mm mmheH N.h mHH o mH mm cmweH H.m mNH m mm mo NomeH c.m mMH N Hm Ho omm.H m.0H omH H on mm mcc.H m.~H omH o wwwmmoww In? new... 1%. mm are muH>Huom muH>Huom OHHHommm Hmuoa mabaaon annoy «0 uanmz «0 panmz aoHun>nnum .huH>Hu0« mmmummmsmnuocHad mummeoanImcHost so uommmm cOHum>umumII.mH mHmHH mo macho memImucmsa .mucmEHHmmxw moan» sH OmchunO .chHumummwum HmEOmeoumm comm you com: .nsne a mom AbbmHms stop a case mums masom Op .QHH to com mums ucmucoo sHmmmo mcHeHuwg fiHs mumHo ”muoz c.m cH ccH How mm c.m c.m cmH cmw mm m.m m.h mmH com c.m chuoum HImE x GHODOHQ HIOE x chuOHm HImE x HO>HH H m x HIcHE x mOHoss HIsHE x mmHOEc HIcHE x MOHoec HIsHE x MOHOEG Achmmo my . .< . DOHQ unopmcummsm MHuccocooqu mmEOmeoumm cadencesom .mmmuommsmnuocHE4 OUMmeomeIOGHOsmH mo Oncommmm humumHnII.cH fiance 103 both enzyme activities are increased in the isolated peroxi- somal fractions from rat liver by in yiyg Clofibrate treat- ment (unpublished data from Tolbert's group). Consequently, the level of the aminotransferase and glycolate activities in the peroxisomal frantions from rats fed a regular diet supplemented with 0.2% Clofibrate in the ground diet was examined. The Clofibrate treatments were similar to those reported in the literature (37,74,106), except that the sodium salt rather than the methyl ester was used. The methyl ester of Clofibrate is reported to be more readily absorbed by the intestine, but similar peroxisomal prolif- eration is achieved with the sodium salt. In the male rats fed with the sodium salt of Clofibrate for 11 days there was a proliferation of peroxi- somes as Observed by electron microscopy (unpublished data from Tolbert's group). These data confirmed previous reports (37,74,106). The total amount of catalase activity in the Clofibrate treated animals increased 50%, total gly- colate oxidase activity increased 50%, and total leucine- glyoxylate aminotransferase increased 70% (Table 17). Sometimes one fold increase in the total amount Of the above three enzyme activities was Observed (data not pub- lished). Reddy and Svoboda have always reported one fold increase of peroxisomal numbers in Clofibrate treated male rate (74,82). The 50% increase in enzyme activity Observed in our studies could be due to the poorer absorption Of the 104 .mucmEHHmmxw mossy CH Umchuno OHO3 muHsmmu unHHsHm .emnmlmumz num>HH no 6 mm snobs .nsnt HH no“ umHt munubHHoHo e~.o m csz .nHH on com mums wHHmHuHcH o cmH #5090 much MOHsmoImsmmumm mHmE me "msoz 44 m NH m mm cm c.m mmmummmcmuuocHE< OHMmeO>HUImsHOOOH 44 4 NH m mm 44 4.4 mnncon munHoosHO o o as mmom mH ooH II mnncon o esohaoouso c4 ccH x c.c m ccH x m.c mm ccH x c.m mcH x c.m mmMHmumU mumHQHmOHU so mumm co m cH N hH c.m >.H mmmnmumcmuuocHfid OHMHhxowHwImcHost mm m NH a Hm ea a.m mnncon munaoosao c c on mmnH cH «M II mmmpon Q Osonnoouhu mm ccH x c.c c ccH x 4.c mH ccH x c.m mcH x c.m mmMHmumU mumu Houusoo sHmuOHm 0E w sHmuOHm me w sHmuoum m5 ckuOHm me x \cHE\meOEc \cHE\mmHOEc \GHE\meOEs \cHE\meOEc mofihusm ucmumcummsm MHHOcocooqu mmsOMonumm mumsmmOEom .mmmuwumcmuu IocHE« OHMHaxOMHOImcHosOH HmEOmeonmm so usmfiumona OHMHQHHOHU mo uommmmII.hH wands 105 sodium salt. The peroxisomal fraction from the Clofibrate treated animals had a higher specific activities (at least two fold) for catalase, glycolate oxidase and leucine-gly- oxylate aminotransferase. The increases in specific activi- ties were higher than the increases in total amount of these enzymes. This could be caused by cleaner peroxisomal fractions, which are always contaminated by a few mito- chondria and significant amounts of endoplasmic reticulum. In the experiments reported in Table 14, a 10% higher level of peroxisomal enzymes were recovered in the peroxisomal fraction on the sucrose gradient from the Clofibrate treated animals. It is not certain whether this is experimental variation, because the particles are so easily ruptured, or the Clofibrate might have increased the stability of the peroxisomes for isolation. Effect of Sex on Peroxisomal Amino- transferase and a—Hydroxy Acid Oxidase Activities The total activity of liver peroxisomal leucine- glyoxylate aminotransferase revealed a sex difference (Table 18), in male and female rats of approxiamtely 200 g in body weight. Leucine-glyoxylate aminotransferase was at least twice as active in the adult female liver than in male, when expressed on the basis of nmoles x min-1 x g 1 liver tissue. In contrast, higher activity was Observed for catalase in male rats than in female rats. A higher 106 cc cm nm em mNH N4 0H0: cm c4H Hm cmH mom mm OHmEOm msmmHu HIm msmmHu HIm x mommHu HIm x mamme HIm x m HIcHE x mOHOEc w HIsHE x mcHOEc HIGHE x OHOEc HIcHE x mOHOEE Homumcummsm mmEOmeOHmm wumcwoosom Hmuoa wwmcmmoeom Hmuoa xwm wmmummmcmuuocHss mummeoanImsHoswq OOMHmumo .HO>HH umm OHmeh can mHmS cH MHm>mH mmmummmsmuuocHem OHMmeomHOImcHOsmH mo cOMHHmmEOUII.mH Manda 107 percent recovery in the peroxisomal fractions was observed in female rat liver. It has been reported that glycolic acid oxidase activity is 40% higher in male adult rat livers than in female adult rat livers (89). Since it is now known that glycolate oxidase is a peroxisomal enzyme (54), the sex difference in glycolate oxidase activity in the subcellular fractions was studies (Table 19). Not only the glycolate oxidase specific activity but also the catalase specific activity was 3 to 4 fold higher in liver homogenate from male rats than in female rats liver homogenate. The spe- cific activity of catalase and glycolate oxidase in the surviving peroxisomal fractions was also somewhat higher in male rats, though the total peroxisomes recovered as a percent yield was much lower in male rats than in females. Consequently, in the soluble fraction from male rat livers the specific activity of these enzymes was ten fold greater than that from female rats. Presumably, the soluble cata- lase was due to the breakage Of the more active or more numerous peroxisomes in the male. These data indicated that the peroxisomes from the male rat appeared much more fragile as only about 20% of the peroxisomes survived in the particle gradient frac- tion, whereas about 54 to 66% Of the peroxisomes from female rats seemed to survive during the isolation proced- ure. Because of this phenomenon, De Duve's group (19,20) 108 .mucmEHHomxm o3» sH omsHmvno oum3 muHsmOH HMHHEHm huHmsmo mmouosm an Omumummmm mums mums xmm umnuHm EOHH moEOMHxOHOm .pschmnm HO>HH .OOHHHR mums Am com mHoumEonummmc mpmu OHmEmm HO OHME “mcHHm "Opoz mo c.H Hm c.m c.m mmmcon oumHOOMHO mu 5mm mH mva chm OmMHmpmo OHmz hH H.c cc H.m c.o Ommcon mucHoome mm mm 4m Nmm ccH mmmHmme mHmEOm m AsHmuoum ms w sHouOHm ms Achuonm ms \sHE\mmHOE:c \cHfi\mOHOEsc \GHE\mmHOEsc mumcmmoson Oaxaca xmm usmumcuwmsm McEOwaoumm Hmuoa .HO>HH umm OHmEmm can OHM: cH Hm>mH mmmcon oumHoowHw mo sOmHHmmEOUII.mH mqmma 109 have advocated the use of female rats for peroxisomal research rather than males. Leucine-Glyoxylate Aminotransferase in Rat Kidney Rat kidney peroxisomes were isolated on a sucrose density gradient in a similar manner to that used for liver tissue. Leucine-glyoxylate aminotransferase was also present in the rat kidney peroxisomes (Table 20). About 35% of the catalase in kidney homogenates was in the peroxi- somal fraction, 12% was smeared throughout the mitochondrial fraction, and about 37% was in the soluble fraction. This lower recovery of rat kidney peroxisomes on the gradient has previously been observed (31,54) and is indicative of the harder grinding necessary to break up the kidney. The distribution of leucine-glyoxylate aminotransferase was very similar to the catalase distribution, except that there was one more distinct peak fraction of this amino- transferase activity on the gradient between catalase and cytochrome c oxidase at a sucrose density 1.215 g x cc-l. About 22% Of the total leucine-glyoxylate aminotransferase activity was in the peroxisomal fraction, 17% was distribu- ted through the mitochondria, and 39% was in the soluble part. Mitochondrial contamination only accounted for 1.7% of the peroxisomal activity. The highest specific activity was Observed in the peroxisomal fractions. Therefore, the 110 00H x NH.o m.H HNNH mcH x mo.o m.cH mmmummmcmuuocHsd Ackuonm ms \QHE\mmHOEcv GHOHOHQ ma \sHE\mmHOEsc ucmumcummsm MHHOGOBOOHHS OHOmeomeImcHosmH com mmmcon o msonnooumo ccH x 4m mmMHmumU AsHmuoum m8 \GHE\mOHOEsc mfihucm mmEOmeOHmm .mmscHM pom Eoum mmHOHunmm m:oE¢ mmmnmmmsmuuocHfim ODOHaxomHUIOsHOOOH mo coHuanHMMHQII.cm mqmma 111 leucine-glyoxylate aminotransferase should be considered as a peroxisomal enzyme in kidney tissue. The total amount of leucine-glyoxylate aminotransferase in rat kidney was low in comparison to that in rat liver. Since rat kidney contains a high amount of leucine-ketoglutarate aminotransferase activity in mitochondria and cytosol (127), the significance of a small amount of peroxisomal leucine- glyoxylate aminotransferase is not clear. Absence of Leucine-Glyoxylate Amino- transferase in Spinach Leaf Peroxisomes Leucine-glyoxylate aminotransferase assays were also run on sucrose gradient fractions of the separated particles from spinach leaves (Table 21). The major por- tion of the leucine-glyoxylate aminotransferase was in the supernatant fraction at the tOp Of the gradient, and only 3.8% of the total activity was found associated with the peroxisomal fraction. However, from the marker enzymes, 80% of this activity in the peroxisomal area could be attributed to mitochondrial contamination. Therefore, significant amounts of leucine-glyoxylate aminotransferase may not be present in spinach leaf peroxisomes. This result is another example of the different enzymatic compo- sition between plant and animal microbodies. 112 TABLE 21.--Distribution of Leucine-Glyoxylate Aminotrans- ferase in Spinach Leaves. Peroxisomes Supernatant Enzyme nmoles x min-1 % nmoles x min-1 % x ml-1 x ml"1 5 5 Catalase 7.3 x 10 33 2.6 x 10 58 Leucine-Glyoxylate Aminotransferase 0.9 3.8 7.7 91 113 Preliminary Examination of the L-Histidine—Glyoxylate Amino- transferase Reaction in Rat Liver Peroxisomes An aminotransferase reaction between L-histidine and_glyoxylate has not been reported. Rat liver peroxi- somes were isolated in a sucrose density gradient and assayed for L-histidine-glyoxylate transaminase activity. The histidine-glyoxylate aminotransferase was measured radiochemically by the same procedures as used for leucine- glyoxylate aminotransferase. The gradient distribution profile for an enzymatic histidine-glyoxylate transamina- tion reaction was similar to that for catalase and the data are analyzed in Table 22. In this run a second peak of catalase activity coincided with the cytochrome c oxidase peak, so that histidine-glyoxylate transamination activity was also observed in the mitochondrial fractions. Mito- chondrial contamination only accounted for 2.8% of the total peroxisomal activity of this aminotransferase while peroxisomal contamination, as catalase, in the mitochondrial fractions accounted for 74% of the histidine-glyoxylate reaction. Therefore, this aminotransferase activity seems to be primarily located in the rat liver peroxisomes. The cytosol activity could be accounted for by the break- age Of the peroxisomes during homogenization. The pH Optimum for the peroxisomal L-histidine- glyoxylate aminotransferase is shown in Table 23. The pH 114 TABLE 22.--Distribution of L-Histidine-Glyoxylate Amino- transferase Reaction in Rat Liver. Histidine- Cytochrome c glyoxlate Organelle ( mgizgigii/ml) Oxidase 'aminotrans- “ (nmoles/min/ml) ferase (0pm) Peroxisomes 1500 46 1401 Mitochondria 1500 2245 1927 Supernatant 4500 0 7730 TABLE 23.--PH Optimum for L-Histidine-Glyoxylate Amino- transferase Reaction. PH Value Net CPM 5.56 2736 5.82 3425 6.16 3568 6.96 3328 7.20 2581 7.48 2141 7.69 2581 7.94 2249 8.27 1906 8.47 1719 8.68 1355 8.82 1244 Note: Isolated rat liver peroxisomes were used. enzyme was run as the blank control. results were obtained in 2 experiments. Boiled Similar 115 optimum for this transaminase was around 6.2 which was com- pletely different from that of leucine-glyoxylate amino- transferase at pH 8.4. This result indicates that probably different proteins were involved in the transamination reaction of histidine glyoxylate and leucine-glyoxylate. This histidine-glyoxylate aminotransferase was also rela- tively heat stable, as was leucine-glyoxylate aminotrans- ferase (Table 24). Both activities were stable at 50°C for 60 min. Attempts to separate the histidine-glyoxylate from the other aminotransferases in rat liver peroxisomes by Sephadex G-150 gel filtration and DEAE cellulose chroma- tography were unsuccessful. On those Chromatographic runs, histidine-glyoxylate activity coincided with the peak of peroxisomal leucine-glyoxylate aminotransferase activity (Figures 9 and 15). TABLE 24.--Heat Stability of the Histidine-Glyoxylate Aminotransferase.* Treatment Time at 50°C (min) Net CPM 0 3783 10 4023 20 3430 40 4083 60 3871 *Similar results were Obtained in three experiments. 116 Figure 15.--Gel Filtration Chromatography of L-Histidine- Glyoxylate Aminotransferase. Rat liver peroxisomes were isolated, dialyzed and heated in the same procedures as in Figure 7. After being cooled in an ice bath, the heat-treated sample was broken by 0.01% Tri- ton X-100 and placed on a G-150 sephadex column. The glyoxylate aminotransferase reactions were assayed radiochemically. A ----- A 280nm OD reading for protein. o—————o histidine-glyoxylate aminotransferase activity. x leucine-glyoxylate aminotransferase activity. 117 SEQ >._._>_._.U< m FRACTION NUMBER SUMMARY AND DISCUSSION One characteristic of peroxisomes or microbodies is that reactions involving glyoxylate formation and utilization, including its transamination to glycine are located in this organelle (46,86,98,102). In microbodies from plants these aminotransferases are serine-glyoxylate, glutamate-glyoxylate, and alanine-glyoxylate. Aminotrans- ferase activities for glyoxylate with L-leucine and L- phenylalanine and probably other amino acids in rat liver peroxisomal fractions have not been reported. This gly— oxylate specific aminotransferase had higher activity in rat liver tissue than in rat kidney tissue. The enzyme is different from the general branched chain amino acids aminotransferase previously reported in liver (108); it utilizes only leucine among the three branched chain amino acids. Since the enzyme is more active with L-leucine than the other amino acids, the combination, leucine-gly- oxylate aminotransferase, will be used to represent this enzyme. The function of a rather specific aminotransferase in mammalian peroxisomes is not understood. The specificity and location in peroxisomes of the enzyme may be a necessary 118 119 compartmentation for all reactions involving glyoxylate, because this reactive aldehyde is an excellent substrate for lactate dehydrogenase in the cytoplasm, and glyoxylate will nonenzymatically react with amino group on proteins, particularly amides. However, the specificity of the liver enzyme toward L-leucine and L-phenylalanine cannot be explained, and the fate of the keto acids that are formed from leucine and phenylalanine is not known. Especially the keto acid product phenylpyruvate does not appar to be metabolically useful, although there may be some, as yet unidentified function for this enzyme. Subcellular Distribution About 50 to 60% of the total leucine-glyoxylate or phenylalanine-glyoxylate aminotransferase was recovered in the peroxisomal fraction (Table 1). If this activity were corrected for peroxisomal breakage during homogena- tion by the factor obtained from catalase recovery in the isolated particle (25), the percentage recovery in the peroxisomal fraction could be as high as 90 to 100% of the total activity of the homogenate. The specific activity of leucine-glyoxylate aminotransferase in the peroxisomal fractions from isopycnic sucrose gradients was about 180 nmoles x min.1 x mg-1 of peroxisomal protein, and about 130 nmoles x min-1 x mg-l Of peroxisomal protein for phenylalanine-glyoxylate aminotransferase. There was at 120 least a 25 to 30 fold higher specific activity in the per- oxisomes than in the supernatant. These results eliminate any nonspecific binding of this enzyme from cytosol and microsomes to the peroxisomal fraction during isolation. Therefore, this leucine-glyoxylate aminotransferase should be considered as a peroxisomal enzyme. The subcellular location of this glyoxylate specific aminotransferase in rat liver peroxisomes is consistent with other glyoxylate aminotransferases which were found to be exclusively in leaf peroxisomes (46,130) and seed glyoxysomes (98). Within the peroxisomes this aminotransferase is located in the soluble matrix because it had the same solubilization prOp- erties as the matrix marker, catalase, upon mild rupture Of the peroxisomes by pyrophosphate buffer. Substrate Specificities The rat liver peroxisomal leucine-glyoxylate amino- transferase was specific for glyoxylate as the amino group acceptor and for the L-isomer Of the amino donor. The leucine-glyoxylate aminotransferase was inactive toward oxalacetate and ketoglutarate. Though the partially iso- lated peroxisomal enzyme also catalyzed a pyruvate to alanine transamination, it was 17 times more active with glyoxylate than with pyruvate. The low activity with pyruvate could be attributed to some degree of low speci- ficity or to a contaminating alanine aminotransferase. 121 Similar substrate specificities were observed in human liver alanine-glyoxylate aminotransferase (110) and serine- glyoxylate aminotransferase (85). The best naturally occurring amino group donor for the glyoxylate aminotrans- ferase at pH 8.4 in rat liver peroxisomes was L-leucine, although L-phenylalanine and L-alanine were nearly as good (Table 3). Some peroxisomal activity was also Observed with L-histidine, L-asparagine and L-glutamine, but this could be due to other enzymes. Transamination Of the amides with glyoxylate have not been investigated further. The D form of the amino acids, such as D-leucine and D-phenyl- alanine were not active with this transaminase. Also these D amino acids were not inhibitory for the L-amino acid transamination reactions. Of particular note was the absence of transamination reaction in the rat liver peroxi- somes for glutamate or serine with glyoxylate, which have been reported to be most active in the leaf peroxisomes (86,130). The rate Observed with alanine could probably be attributed to the nonspecificity Of the leucine-glyoxylate aminotransferase and a different enzyme. The absence of additive aminotransferase activities with leucine plus alanine, or phenylalanine plus alanine to glyoxylate, a similar heat stability at 50°C, and a similar heat inacti- vation profile at high temperature, all indicated that 122 probably one enzyme existed for catalyzing these three reactions. However, the Triton x-100 and phosphate stimu- lation Of the leucine-glyoxylate but inhibition Of the alanine-glyoxylate reaction in the peroxisomal fractions, and the noncoincident peaks activities on DEAE cellulose columns indicated that probably different proteins were involved. Specificity studies will have to await success- ful purification Of the enzyme. Similar results have been obtained for the leaf peroxisomal preparations (86), in which the activity for alanine-glyoxylate aminotransferase in the leaf peroxisomal fraction was due to the non- specificity Of both serine-glyoxylate and glutamate-glyoxy- late aminotransferases. 14 14 The product C-glycine, which formed from C- glyoxylate, was identified by silica gel thin layer chroma- tography. It was the only product formed which had radio- activity and the recovery was about 90%. One common prop- erty Of all the transaminases that utilize glyoxylate is the physiological irreversibility of the reaction. In the present studies with rat liver peroxisomal enzyme, no reaction could be demonstrated when 14 C-glycine and the appropriate a-keto acids were utilized as substrates. One hypothesis is that glyoxylate exists as a monohydrate, which has a AF (~3000 calories) of hydration. This high AF prevents the occurrence of the reverse reaction from glycine to glyoxylate. 123 Biochemical PrOperties Rat liver peroxisomal leucine-glyoxylate amino- transferase was not inhibited by either Triton X-100 or phOSphate. This was in contrast to the 20 to 30% inhibition by 0.01% Triton X-100 of most leaf peroxisomal aminotrans- ferases (86,130). Glycolate oxidase (47), and carnitine acetyltransferase activities (51) in liver peroxisomes are also similarly stimulated by Triton X-100. These phenomena are attributed to the change of permeability of the peroxi- somal membrane. Phosphate, which has been reported to severely inhibit aminotransferases in leaf peroxisomes, had no effect on the activity of leucine-glyoxylate amino- transferase in rat liver peroxisomes. On the contrary, this liver enzyme had higher activity in perphosphate and phosphate buffer. It had been thought that no phosphate esters were utilized by peroxisomes and that maybe phos- phate itself was a regulant for peroxisomal activity (85). Since the discovery of a-glycerol phosphate dehydrogenase in mammalian peroxisomes (31), the hypothesis that phOSphate esters were not peroxisomal substrates became invalid, and likewise the absence of phosphate inhibition of this amino- transferase indicated that there was probably no phosphate effect on mammalian peroxisomes. Several aminotransferases that utilize glyoxylate as a substrate have been purified from mammalian livers 124 and plant leaves. These include alanine-glyoxylate from human liver (109) and rabbit liver (26), glutamate-gly- oxylate from human liver (109) and rat liver (61), and serine-glyoxylate, glutamate-glyoxylate and aspartate aminotransferases (130) from spinach leaf peroxisomes. Studies on the pyridoxal phosphate cofactor of spinach leaf peroxisomal aminotransferases closely resembled that Of leucine-glyoxylate aminotransferase from rat liver peroxisomes. The presence of a possible requirement for pyridoxal phosphate was supported by inhibition with 1mM hydroxylamine and isonicotinic acid hydrozide which are known to bind the cofactor. However, leucine-glyoxylate aminotransferase did not show a requirement for exogenous pyridoxal-5'-phosphate, pyridoxal, or pyridoxamine phos- phate. Extensive dialysis of this transaminase in phos- phate or glycylglycine buffer without pyridoxal phosphate showed no decrease in catalytic activity. However, storing the dialyzed enzyme at -20°C without exogenous pyridoxal phosphate resulted in a 30% loss in activity in three weeks when compared with samples stored with exogenous cofactor in the buffer. Therefore, it is probable that pyridoxal phosphate is the cofactor, but it must be very tightly bound. PH Optimum and Kinetics A pH Optimum at 8.4 to 8.5 for the rat liver peroxi- somal aminotransferase was reasonably distinct, with 125 approximately 50% of the Optimal rate remaining at pH 6.5. This was consistent with a similarly distinct pH optimum reported for glutamate-glyoxylate aminotransferase from rat liver (61) and in contrast to the broad Optimal pH Observed for glutamate-glyoxylate aminotransferase from human liver (109). PH Optima between 8.2 to 8.7 are con- sidered to be a characteristic of peroxisomal enzymes. The parallel character of the double reciprocal plots Of the initial velocity of the enzyme reaction indi- cated a typical Ping-Pong Bi Bi mechanism at pH 8.4. This is similar to the mechanism suggested for serine-gly- oxylate aminotransferase from oat leaves (10), kidney beans (102). Since the Ping-Pong mechanism is usually a reversi— ble process, one would expect to Observe an isotope ex- change between glyoxylate and glycine. However, glyoxylate aminotransferases are known for their irreversibility in the direction Of glycine formation (46,92,94). Indeed, an insignificant exchange between glyoxylate and glycine was Observed with isolated rat liver peroxisomes (Table 3). This leucine-glyoxylate aminotransferase has a lower Km for glyoxylate (0.5mM) than for leucine (2.5mM) and phenylalanine (2.8mM). Stability and Inhibitor Studies Since the glyoxylate aminotransferase showed almost equal activity toward leucine and phenylalanine, heat 126 stability and inhibitors were used to study whether one or two enzymes were involved. NO difference in activity was observed when heating at 50°C for one hour. All four amino- transferase activities (leucine, phenylalanine, alanine, and histidine-glyoxylate) were stable at 50°C. Even the same heat inactivation profiles above 60°C were Observed. The slight 10% stimulation of activity by a 50°C treatment was a reproducible result. Most mammalian aminotrans- ferases (109,110) are similarly heat-stable at 50°C to 60°C, but Spinach aminotransferases are stable only to about 350C to 40°C. The heat stability of the liver peroxisomal enzyme was used to advantage in the purification scheme. The leucine-glyoxylate aminotransferase was inhibi- 4M, suggesting an ted by p-chloromercuribenzoate at 10- essential sulfhydryl group. The presence of a tightly bound pyridoxal phosphate in this enzyme is suggested by the hydroxylamine and isonicotinic acid hydrazide inhibi- 3M. The absence of sodium cyanide and Cu++ tions at 10- inhibition suggests that no divalent cation was required for Optimal enzyme activity. The inhibition Observed by glyoxal suggests the possible presence of lysine and arginine_groups (30) at the active site of the enzyme and probably to serve as the linkage between the enzyme mole- cule and the cofactor pyridoxal phosphate. By several criteria one enzyme was judged to be catalyzing the transamination reaction for both L-leucine 127 and L-phenylalanine. Rates with combinations of amino acids with excess glyoxylate were not additive. Similar pH activity curves and heat inactivation curves were ob- tained. The two activities could not be separated by Sephadex and DEAE cellulose chromatography procedures. Both activities had a Similar distribution in the peroxi- somal fractions and in the soluble matrix of the peroxi- somes. Both activities were affected Similarly by inhibi- tors. The liver peroxisomal aminotransferase did have lower activity for a number of other amino acids. The Significant rate for the histidine—glyoxylate transamina- tion could not be separated from the activity with leucine and phenylalanine by either Sephadex or DEAE cellulose, but the reaction with histidine did have a lower pH optimum around 6.2. A histidine-pyruvate aminotransferase has been reported in liver mitochondria (58); the relation- ship of this peroxisomal glyoxylate aminotransferase to that histidine-pyruvate aminotransferase iS not clear. Although one enzyme which catalyzes all the glyoxylate aminotransferase reactions in the rat liver peroxisomes is favored, the possibility Of the presence of another protein for histidine-glyoxylate in rat liver peroxisomes cannot be ruled out. 128 Physiological Studies It has generally been accepted that the level of aminotransferases can be varied with protein intake or hormones (12,26,41). Most investigations have been concen- trated on the ketoglutarate aminotransferases, and little is known about glyoxylate aminotransferase in this respect. In my studies, it was found that a 50% decrease in the Specific and total activities Of leucine-glyoxylate aminotransferase occurred subsequent to fasting for one or two days. This was in contrast to a leucine-ketoglutarate aminotransferase reported in rat liver and kidney which Showed no change in Specific and total activities after one day starvation (12). The Significance of a decrease in activity is probably related to the lack of availabil- ity of amino acids during fasting. The effects Of various dietary treatments on leucine-glyoxylate aminotransferase were studied in order to understand the physiological role of this rat liver peroxisomal enzyme. The total activity of leucine-glyoxylate aminotransferase increased as the protein content in the diet increased, but the Specific activity in the isolated peroxisomes Showed a Slight de- crease. An increase or decrease in the availability of amino acids may again explain the increase or decrease in total activity of leucine-glyoxylate aminotransferase when the rats were fed with a 55% casein or 3.4% casein 129 diet respectively. The slight increase in specific activ- ity of leucine-glyoxylate aminotransferase in low protein diet rats is probably due to the cleaner peroxisomal isola- tion. The decrease Of Specific activity in peroxisomes from rats on the high protein diet may be due to either more contamination in the isolated peroxisomes or due to increased amount of proteins in the peroxisomes. The total catalase activity also increased 3 fold in the high protein diet rats as compared to those on a low protein diet (data not shown). Dietary uptake is suggested as a research tool to distinguish whether one or more than one enzyme is involved in the transamination of leucine, phenylalanine, histidine etc. with glyoxylate. If each amino acid induced only the corresponding transaminase activity, this evidence would favor separate enzyme proteins. In developmental studies with rat age, increase in leucine-glyoxylate aminotransferase followed increases in catalase. The total activity of the glyoxylate amino- transferase in the organelle fractions developed dramati- cally between 2 and 40 days of the age and then remained constant in Older rats. This development is Similar to other reports (38,120) that microbodies develOp postnatally in animal tissues. Similar stimulation of the total activities by Clofibrate of leucine-glyoxylate aminotransferase, catalase, 130 and a-hydroxy acid oxidase (glycolate oxidase) in rat liver peroxisomes was Observed. Since leucine-glyoxylate amino- transferase and glycolate oxidase activities were increased along with catalase, it can be postulated that peroxisomes in the Clofibrate treated rats might contain the same enzy- matic composition as the untreated rats. Tissue Distribution Contrary to the substantial amount of leucine-gly- oxylate aminotransferase (140 to 180 nmoles x min.1 x mg-1 peroxisomal protein) found in rat liver, very low activity (20 nmoles x min.1 x 1119-1 peroxisomal protein) was observed in rat kidney. No detectable activity was found in pig liver (data not Shown) and Spinach leaf peroxisomes. The difference in distribution of this aminotransferase is - comparable with another peroxisomal enzyme, a-hydroxy acid oxidase, which iS Specific for glycolate in rat liver peroxi- somes, but specific for d-hydroxycaprylate or longer chain a-hydroxy acids in rat kidney peroxisomes. Rat liver a- hydroxy acid oxidase iS isozymic to that in rat kidney. Similarly, differences exist in the substrate specificity of carnitine acetyltransferase (51) in liver and kidney peroxisomes. All these results reflect the varying enzyme composition of the microbodies in different tissues. The significance of the low activity of leucine- glyoxylate aminotransferase in rat kidney peroxisomes and the absence of activity in pig liver peroxisomes is not 131 clear. One possibility iS that different substrate Spe- cificities for the amino group donor may exist and were missed in the preliminary survey in the rat kidney peroxi- Somes. Since high amounts Of leucine-ketoglutarate amino- transferase exist in rat kidney mitochondria, there is probably no need for a leucine-glyoxylate aminotransferase in the rat kidney peroxisomes. The other possibility may be that the a-hydroxy acid oxidase of the rat kidney does not oxidize glycolate (54) and therefore does not form glyoxylate. Consequently there is no requirement for this glyoxylate Specific aminotransferase in rat kidney. Rowsell (91) in 1955 found that L-leucine, L-phenyl- alanine, L-histidine, or DL-methionine was transaminated with pyruvate in rat liver, mouse liver, and pigeon liver. There was little activity for L-leucine, L-phenylalanine or DL-methionine with pyruvate in rat kidney and none in pig liver. It is now established that many previously reported pyruvate aminotransferases actually utilize gly- oxylate better. Thus, Rowsell's old Observation may be similar to my study Of a glyoxylate Specific aminotrans- ferase in peroxisomes. If true, this specificglyoxylate aminotransferase could also exist in mouse liver and pigeon liver peroxisomes. Several functions for this leucine-glyoxylate amino- transferase in rat liver peroxisomes can be prOposed. 132 Most likely, it is a mechanism to safely remove glyoxylate rather than oxidize it to oxalate by the a-hydroxy acid oxidase (88). At the same time this provides for the bio— synthesis Of glycine in rat liver. The enzyme utilizes L- leucine, L-phenylalanine, L-alanine, and probably other . amino acids in the direction of glycine formation. The non- specificity toward the amino group donor of this glyoxylate aminotransferase can lead to a large amount of glycine formation in the peroxisomes. Another function for the peroxisomal glyoxylate aminotransferase may be to serve as an enzyme for leucine degradation in the rat liver. Transamination of the branched chain amino acids occurs chiefly outside the liver; trans- aminases for leucine, isoleucine, and valine in the rat are most active in heart and kidney, and lowest in liver (40). However the corresponding keto acid decarboxylase was found to have the highest activity in liver tissue (17,127). The rationale for this organ division between kidney and liver Of this keto acid decarboxylase has been a puzzle. The discovery of liver peroxisomal leucine- glyoxylate aminotransferase probably will be integrated into this puzzle by being a mechanism for the keto acid formation in the same tissue with the keto acid decarboxy— lase. BIBLIOGRAPHY 133 10. 11. 12. 13. 14. 15. BIBLIOGRAPHY Ahlabo, I., and T. Barnard. 1971 J. Histochem. Cytochem. 19: 670. Allen, J. M., and M. E. Beard. 1965 Science 149: 1507. AverS, C. J., and M. Federman. 1968 J. Cell Biol. 37: 555. Baudhuin, P., M. Muller, B. Poole, and C. De Duve- 1965 Biochem. BiophyS. Res. Comm. 20: 53. Beard, M. E. 1972 J. Histochem. Cytochem. 20: 173. Beevers, H. 1969 Ann. N. Y. Acad. Sci. 168: 313. Beers, R. F., Jr., and I. W. Sizer. 1952 J. Bio. Chem. 195: 133. Blanchard, M., D. E. Green, V. Nocito-Carroll, and S. Ratner. 1946 J. Biol. Chem. 162: 137. Braunstein, A. E., and M. G. Dritzman. 1937 Nature 140: 503. Brock, B. L. W., D. A. Wilkinson, and J. King. 1970 Can. J. Biochem. 48: 486. Burch, H. B. 1957 Methods in Enzymology. S. P. Colowick and N. 0. Kaplan. Academic Press, N. Y. III, 947- Buse, M. G., F. Biggers, C. Drier, and J. F. Buse. 1973 J. Biol. Chem. 248: 697. Campbell, L. L. 1956 J. Bacteriol. 71: 81. Cheung, G. P., J. P. CotrOpia, and H. J. Sallach. 1969 Arch. Biochem. BiOphys. 129: 672. Clagett, C. O., N. E. Tolbert, and R. H. Burris. 1949 J. Biol. Chem. 178: 977. 134 135 16. Connock, M. J., and P. R. Kirk. 1970 Histochem. J. 1970 2: 371. 17. Dancis, J., J. Hutzler, and M. Levitz. 1961 Biochem. BiOphyS. Acta. 52: 60. 18. David, B. J. 1964 Ann. N. Y. Acad. Sci. 121: 404. 19. De Duve, C. 1966 Physiol. Rev. 26: 323. 20. De Duve, C. 1969 Proc. R. Soc. 173: 71. 21. Dohan, J. S. 1940 J. Biol. Chem. 135: 793. 22. Donaldson, R. P. 1971 Ph.D. dissertation. Michigan State University. 23. Donaldson, R. P., N. E. Tolbert, and C. Schnarren- berger 1972 Arch. Biochem. Biophys. 152: 199. 24. Essner, E. J. Histochem. Cytochem. 1970 18: 80. 25. Fahimi, H. D. 1973 J. Histochem. Cytochem. 21: 999. 26. Feld, R. D., and H. J. Sallach. 1973 Arch. Biochem. Biophys. 59: 757. 27. Freedland, R. A. 1967 J. Nutrition 91: 489. 28. Frederick, S. E., and E. H. Newcomb. 1969 J. Cell Biol. 43: 343. 29. Fukuda, T., and T. Hayashi. 1958 J. Biochem. 45: 469. 30. Fukui, T., A. Kamogawa, and Z. Nikuni. 1970 J. Bio- chem. 67: 211. 31. Gee, R., E. McGroarty, B. Hsieh, D. Wied, and N. E. Tolbert. 1974 Arch. Biochem. Biophys. 161: 211. 32. Gerhardt, B., and H. Beevers. 1968 Anal. Biochem. 24: 337. 33. Green, D. E., L. F. Leloir, and V. Nocito. 1945 J. Biol. Chem. 161: 559. 34. Hamor, T., and E. T. Garside. 1973 Comp. Biochem. Physiol. 453: 147. III I Ivlv Ill ill I I lllllll lll ill] Ii 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 136 Hand, A. R. 1973 J. Histochem. Cytochem. 21: 131. Hand, A. R. 1974 J. Histochem. Cytochem. 22: 207. Hess, R., W. Riess, and W. Stéubli. 1967 Progr. Biochem. Pharmacol. 2: 325. Holmes, R. S. 1971 Nature, New Biol. 232: 218. Hruban, Z., and M. Recheigl Jr. 1969 Microbodies and Related Particles. Int. Rev. Cytol. Suppl. I: 173 New York Academic Press. Ichihara, A., and E. Koyama. 1966 J. Biochem. 59: 160. Ichihara, A., H. Takahashi, K. Aki, and A. Shirai. 1967 Biochem. Biophys. Res. Comm. 26: 674. Kalckar, H. M. 1947 VJ. Biol. Chem. 167: 429. Kashiwagi, K., T. Tobe, and T. Higashi. 1971 J. Biochem. 70: 785. Keay, L., J. C. Robinson, R. Molinari, and I. W. Sizer. 1960 Fed. Proc. 19: 26. King, J., and E. R. Waywood. 1968. Can. J. Biochem. 46: 771. Kisaki, T., and N. E. Tolbert. 1969 Plant Physiol. 44: 242. Kisaki, T., A. Imai, and N. E. Tolbert. 1971 P1. Cell Physiol. Tokyo 12: 267. Kun, E. 1952 J. Biol. Chem. 194: 603. Kun, E., J. M. Dechary, and H. C. Pitot. 1954 J. Biol. Chem. 210: 269. Lowry, O. H., N. J. Roserough, A. L. Farr, and R. J. Randall. 1951 J. Biol. Chem. 193: 265. Markwell, M. A. K., E. J. McGroarty, L. L. Bieber, and N. E. Tolbert. 1973 J. Biol. Chem. 248: 3426. Martin, E. M., and R. K. Morton. 1956 Biochem. J. 62: 696. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 137 McGroarty, E., and N. E. Tolbert. 1973 J. Histochem. Cytochem. 21: 949. McGroarty, E., B. Hsieh, D. Wied, R. Gee, and N. E. Tolbert. 1974 Arch. Biochem. Biophys. 161: 194. Meister, A. 1954 J. Biol. Chem. 206: 587. Meister, A. 1962 The Enzymes. Ed. P. D. Boyer, H. Lardy, and K. Myrback. Vol. 6: 193. Metzler, D. E., J. Olivard, and E. E. Snell. 1954 J. Am. Chem. Soc. 76: 644. Morris, M. L., Sc-C. Lee, and A. E. Harper. 1973 J. Biol. Chem. 248: 1459. Mfiller, M., and K. M. Moller. 1967 J. Protozool. 14 (Suppl.): 11. Nakada, H. 1., and S. Weinhouse. 1953 J. Biol. Chem. 204: 831. Nakada, H. I. 1964 J. Biol. Chem. 239: 468. Nakano, M., and T. S. Danowski. 1966 J. Biol. Chem. 241: 2075. Nakano, M., Y. Ushijima, Y. Tsutsumi, and H. Asami. 1968 Biochem. Biophys. Acta. 167: 9. Nakano, M., M. Saga, and Y. Tsutsumi. 1969 Biochem. BiOphys. Acta. 185: 19. Novikoff, A. B., and S. Goldfischer. 1968 J. Histo- chem. Cytochem. 16: 507. Novikoff, A. B., and S. Goldfischer. 1969 J. Histo- chem. Cytochem. 17: 675. Novikoff, A. B., P. M. Noviloff, C. Davis, and N. Quintana. 1972 J. Histochem. Cytochem. 20: 1006. Novikoff, A. B., and P. M. Novikoff. 1973 J. Histo- chem. Cytochem. 21: 737. Novikoff, A. B., and P. M. Novikoff. 1973 ‘J. Histo- chem. Cytochem. 21: 963. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. O'Kane, E. D., and I. C. Ornstein, L. Petrik, P. Poole, Reddy, Reddy, Reddy, Reddy, Reddy, Reddy, Reddy, Reddy, Reddy, Reddy, Reddy, Rehfeld, D. W. Rehfeld, D. W., and N. E. Tolbert. Rhodin, J. 138 Gunsalus. 1947 J. Biol. Chem. 170: 425. 1964 Ann. N. Y. Acad, Sci. 121: 321. 1971 J. Histochem. Cytochem. 19: 939. B. H. 1968 Ph.D. dissertation. The Rocke- feller University. 1971 Bio- 318. J. K., M. Chiga, and D. Svoboda. chem. Biophys. Res. Comm. 43: 1971 Am. J. Pathol. J. K., and D. Svoboda. 37: 99. 1972 Lab. Invest. 26: J., and D. Svoboda. 657. Svoboda. 140. J., and D. chem. 20: 1972 J. Histochem. Cyto- D. Svoboda. 1972 J. Histochem. 20: 793. J.‘K., and Cytochem. J., and D. Svoboda. 1973 Am. J. Pathol. 70: 421. Histochem. J., and D. 1973 J. Cytochem. Svoboda, 21: 967. J., D. Svoboda, and D. Azarnoff. 1973 Biochem. Bipphys. Res. Comm. 52: 537. 1973 Virchows. 83. J., and D. Svoboda. Arch. Abt. B. Zellpath. 14: J. K., D. Prasad. L. Azarnoff, D. J. Svoboda, and J. D. 1974 J. Cell Biol. 61: 344. J., D. J. Nat. Cancer Inst. 51: Svoboda, D. Azarnoff, and R. Dawar 1973 891. 1971 Ph.D. dissertation. State University. Michigan 1972 J. Biol. Chem. 247: 4803. 1954 Aktiebolagel Godvil Stockholm. lation of ultrastructural organization and function in normal and experimentally changed mouse kidney cells. Corre- 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 139 Richadson, K. E., and N. E. Tolbert. 1961 J. Biol. Chem. 236: 1280. Richardson, K. E. 1964 Endocrinology 74: 128. Robinson, J. C., L. Keay, R. Molinari, and I. W. Sizer. 1962 J. Biol. Chem. 237: 2001. Rowsell, E. V. 1955 Biochem. J. 64: 246. Rowsell, E. V., K. Snell, J. A. Carnie, and A. H. Al-Tai. 1969 Biochem. J. 115: 1071. Rowsell, E. V., J. A. Carnie, and K. Snell. 1969 Biochem. J. 112: 7p. Rowsell, E. V., K. Snell, J. A. Carnie, and K. V. Rowsell. 1972 Biochem. J. 127: 155. Rowsell, E. V., A. H. Al-Tai and J. A. Carnie. 1973 Biochem. J. 134: 349. Saga, M., Y. Tsutsumi, and M. Nakano. 1969 Biochem. Biophys. Acta. 184: 213. Schepartz, B. 1973 Regulation of Amino Acid Meta- bolism in Mammals. Chapter 4, p. 51-119. W. B. Saunder Co. Schnarrenberger, C., A. Oeser, and N. E. Tolbert. 1971 Plant Physiol. 48: 566. Schneeberger, E. E. 1972 J. Histochem. Cytochem. 20: 180. Schuman, M., and V. Massey. 1971 Biochem. Biophys. Acta. 227: 500. Shnitka, T. K., and G. G. Talibi. 1971 Histochemi. 27: 137. Smith, I. K. 1973 Biochem. Biophys. Acta. 321: 156. Smith, I. K. Inhibition of Serine-Glyoxylate Amino- ' transferase by Ammonium Ions (Personal communication). Smith, 2. 1955 Methods of Biochem. Analysis 2: 427. 105. 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 119. 140 Strecker, H. J. 1965 J. Biol. Chem. 240: 1225. Svoboda, D., H. Grady, and D. Azarnoff. 1967 J. Cell Biol. 35: 127. Svoboda, D., D. Azarnoff, and J. Reddy. 1969 J. Cell Biol. 40: 734. Taylor, R. T., and W. T. Jenkins. 1966 J. Biol. Chem. 241: 4391. Thompson, J. S., and K. E. Richardson. 1966 Biochem. Biophys. Acta. 117: 599. Thompson, J. S., and K. E. Richardson. 1967 J. Biol. Chem. 242: 3614. Tolbert, N. E., C. O. Clagett, and R. H. Burris. 1949 J. Biol. Chem. 181: 905. Tolbert, N. E., and R. H. Burris. 1950 J. Biol. Chem. 186: 791. Tolbert, N. E., 1962 Photosynthetic Mechanisms in Green Plants. National Science Foundation- National Research Council Publication 1145, p. 648. Tolbert, N. E., A. Oeser, T. Kisaki, R. H. Hageman, and R. K. Yamazaki. 1968 J. Biol. Chem. 243: 5179. Tolbert, N. E. 1971 Ann. Rev. Plant. Physiol. 22: 45. Tolbert, N. E. 1971 Methods in Enzymology (San Pietro, Ed.), Vol. 23, Part A, p. 665 Academic Press, New York. Tolbert, N. E. 1972 Photorespiration by Algae. Ed. W. D. P. Stewart. Oxford: Blackwell Scientific Publications Ltd. Tolbert, N. E. 1973 Methods in Enzymology (Fleisher, S., Packer, L., and Estabrook, R. W.), Vol. 31, Academic Press, New York. Tolbert, N. E. 1973 Compartmentation and Control in Microbodies. Symposia of the Society for Exp. Biology No. XXVII 215-239. 120. 121. 122. 123. 124. 125. 126. 127. 128. 129. 130. 141 Tsukada, H., Y. Mochizuki, and T. Konishi. 1968 J. Cell Biol. 37: 231. Tubbs, P. K., and G. D. Greville. 1959 Biochem. Biophys. Acta. 34: 290. Ushijima, Y. 1973 Arch. Biochem. BiOphys. 155: 361. Vigil, E. L. 1970 J. Cell. Biol. 46: 435. Vandor, S. L., and N. E. Tolbert. 1970 Biochem. BiOphys. Acta. 215: 449. Weinhouse, S., and B. Friedmann. 1951 J. Biol. Chem. 191: 707. Wilson, D. G., K. W. King, and R. H. Burris. 1954 J. Biol. Chem. 208: 863. Wohlhueter, R. M., and A. E. Harper. 1970 J. Biol. Chem. 245: 2391. Wright, B. E. 1951 ARch. Biochem. Biophys. 31: 332. Yamazaki, R. K. 1969 Ph.D. dissertation. Michigan State University. Yamazaki, R. K., and N. E. Tolbert. 1970 J. Biol. Chem. 245: 5137. '11i1ll‘l: