f6 e. h R t E . Twev , T O a; 2 R 93 E; “a 0r If DWAYN' This is to certify that the thesis entitled AMINOTRANSFERASES IN PEROXISOMES AND DISTRIBUTION OF PEROXISOMAL ENZYMES AMONG LEAF CELLS presented by Dwayne Walter Rehfe 1d has been accepted towards fulfillment of the requirements for PhoDo dpgrpp in Biochemistry Major professor Date I I‘mémv 7 ¢7/ 0-7639 ABSTRACT r" ’ ' AMINOTRANSFERASES IN PEROXISOMES AND DISTRIBUTION OF PEROXISOMAL ENZYMES AMONG LEAF CELLS By Dwayne Walter Rehfeld Spinach leaf organelles were isolated by isopycnic centrifugation in sucrose gradients. The distribution of the organelles in the gradient was determined by measuring catalase for microbodies, cytochrome c oxidase for mito- chondria and chlorophyll for chloroplasts. Four amino- transferase activities were assayed by spectrophotometric and radiochemical procedures. Serinezglyoxylate amino- transferase was found to be located exclusively in the peroxisomes. Most of the alaninezglyoxylate aminotrans- ferase activity was also in the peroxisomes, but some activity which did not coincide with the mitochondria or broken chloroplast peaks was located at lower densities of sucrose. ASpartate aminotransferase activity was located in both the peroxisomes and mitochondria on the sucrose gradients. .Although the broken chloroplasts on the sucrose gradients did not contain aspartate amino- transferase, whole chloroplasts, isolated by differential centrifugation, did contain this aminotransferase. Dwayne Walter Rehfeld The activities of serinezglyoxylate, alanine: glyoxylate and g1utamate:glyoxylate aminotransferases from isolated peroxisomes were compared in various buffers at pH 7. Serinezglyoxylate aminotransferase was inhibited by phosphate buffer while the other two aminotransferase activities were unaffected. The phOSphate inhibition of serinezglyoxylate aminotransferase is probably not of physiological significance since it was only 34% at 10 mM phosphate. Serinezglyoxylate aminotransferase was also inhibited by D-serine. In the presence of D-serine, the l/velocity versus 1/[L-serine] plot was nonlinear. The amino acidzglyoxylate aminotransferases of isolated peroxisomes were separated by ion exchange chromatography on triethylaminoethyl-cellulose columns. Serinezglyoxylate aminotransferase was eluted as one peak, but no activity of alaninezglyoxylate aminotransferase or g1utamate:glyoxylate aminotransferase was detected. On isoelectric focusing columns, the serinezglyoxylate and g1utamate:glyoxylate aminotransferases were separated. The alaninezglyoxylate aminotransferase activity had coincident peaks with each of the other two glyoxylate aminotransferases. Serinezglyoxylate aminotransferase also catalyzed Dwayne Walter Rehfeld a serine:pyruvate aminotrasferase reaction. Both of these activities peaked together during ion exchange chrom- atography and isoelectric focusing. Also, both the serine: glyoxylate and the serine:pyruvate aminotransferase activ- ities were inhibited by D-serine and phosphate and both were equally sensitive to heat denaturation. At a serine concentration of 20 mM, the Km of the serine:g1yoxylate aminotransferase for glyoxylate was 0.15 mM while for pyruvate it was 2.82 mM. At a concentration of 1 mM glyoxylate the Km for L-serine was 2.72. The serine:g1yoxylate aminotransferase was not inhibited by adenosine mono- di- and triphosphate, sodium nitrate, potassium nitrite, O-phospho-L-serine, D-glycerate or 3-phOSphog1ycerate. Ammonium sulfate caused some inhibition. The serine:g1yoxylate, g1utamate:glyoxylate and alaninezglyoxylate anfinotransferase reactions were not reversible. A very small glycine-glyoxylate exchange was detected with isolated peroxisomes. Polyacrylamide disc gel electrophoresis of the aspartate aminotransferase from peroxisomes showed the presence of three isoenzymes. The chloroplasts and mitochondria each contained only one major band of Dwayne Walter Rehfeld aspartate aminotransferase activity. Both the chloroplast and the mitochondrial enzymes had the same electrophoretic characteristics as the fastest moving isoenzyme from the peroxisomes. The three peroxisomal isoenzymes were sep- arated both by ion exchange chromatography and by iso- electric focusing. None of the three aspartate amino- transferase isoenzymes appeared to coincide with the other aminotransferase activities of the peroxisomes. In the spinach leaf peroxisomes, the specific activities of the aminotransferases were: glutamate: glyoxylate, 2.40 umoles x min"1 x mg protein-1; serine: glyoxylate, 1.54; alanine:glyoxy1ate, 0.87; and aspartate: a-ketoglutarate 0.15. Peroxisomes and mitochondria were also isolated from rat liver and dog kidney tissues by isopycnic cen- trifugation in sucrose gradients. The mitochondria con- tained all of the particulate aspartate aminotransferase and the bulk of the particulate alanine:glyoxy1ate amino- transferase. However, the rat liver peroxisomes contained a significant amount of alanine:glyoxy1ate aminotransferase activity which could not be attributed to mitochondrial contamination. Serinezglyoxylate aminotransferase Dwayne Walter Rehfeld appeared to be located only in the liver peroxisomes, and this activity was inhibited by phosphate. Nearly all of the g1utamate:glyoxylate aminotransferase activity was in the soluble fraction. The dog kidney peroxisomes did not contain significant quantities of alanine:glyoxy1ate, g1utamate:glyoxylate, serine:g1yoxylate or aSpartate: d-ketoglutarate aminotransferase activities. The major portion of these activities were in the mitochondria and the supernatant. The distribution of the peroxisomal enzymes within the leaves of Spinach, wheat, corn and sugarcane plants was determined. The peroxisomal enzymes, glycolate oxidase, hydroxypyruvate reductase and catalase were found to be located in both the bundle sheath cells and the mesophyll cells of corn and sugarcane. The bundle sheath enzymes were not completely extracted from the leaf tissue even after two minutes of homogenization in a ‘Waring blendor whereas the mesophyll enzymes were extracted by this treatment. Breakage of the bundle sheath cells was accomplished with a mortar and pestle or a roller mill. The total activity of glycolate oxidase in corn and sugarcane, which do not lose carbon dioxide during photorespiration (without C02-photorespiration), was Dwayne Walter Rehfeld one-third to one-half of that found in spinach and wheat which do exhibit COZ-photorespiration. Both.Atriplex patula, with COZ-photorespiration and Atriplex.£2§g§, with- out COz-photorespiration had similar levels of glycolate oxidase. A differential grinding procedure, consisting of homogenization in a waring blendor followed by grinding in a mortar, was used to obtain extracts of the mesophyll cells and the bundle sheath cells. The kinetic character- istics of glycolate oxidase and hydroxypyruvate reductase from the two types of cells were similar. Glycolate oxidase had a Km (glycolate) of 0.5 mM and hydroxypyruvate reductase had a Km (hydroxypyruvate) of 77 uM. Glycolate oxidase from corn did not reduce p-nitroblue tetrazolium in a phenazine methosulfate-nitroblue tetrazolium system whereas Spinach glycolate oxidase readily reduced nitro- blue tetrazolium.in the same system. No isoenzymes of catalase or hydroxypyruvate reductase from corn leaf extracts were detected by electrophoresis. The function of the peroxisomal aminotransferases in the operation and possible control of peroxisomal Inetabolism via the glycolate pathway is discussed. In addition, the levels of the glycolate pathway enzymes in Dwayne Walter Rehfeld plant leaves is considered in relation to the process of photorespiration in plants. AMINOTRANSFERASES IN PEROXISOMES AND DISTRIBUTION OF PEROXISOMAL ENZYMES AMONG LEAF CELLS By Dwayne Walter Rehfeld A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Biochemistry 1971 Dedicated to my wife, Claudette and to our daughter, Shannon ii ACKNOWLEDGMENTS I thank Professor N. E. Tolbert for his guidance and encouragement during the course of my research and graduate training. My thanks to Mrs. Sandra Wardell and Mrs. Angelika Oeser Schnarrenberger for their technical assistance. I also thank R. Donaldson and G. Lorimer for their cooperation in some of the experiments reported in this thesis. For the many interesting and beneficial discus- sions, I thank my fellow graduate students and the post- doctoral students in the Department of Biochemistry. The encouragement of my parents and my wife's parents is greatly appreciated. I especially thank my wife for her love, faith and encouragement during my graduate training. Without her support, this thesis would not have been possible. The financial assistance from the National Insti- tute of Health and the Department of Biochemistry, ZMichigan State University is appreciated. iii TABLE OF CONTENTS LIST OF TABLES. LIST OF FIGURES . LIST OF ABBREVIATIONS . INTRODUCTION. LITERATURE REVIEW . Chapter I. Carbon Dioxide Fixation Pathways . Photorespiration . Glycolate Pathway. Peroxisomes. Aminotransferases. . . Early Studies of Aminotransferases in Plants . Glyoxylate Aminotransferases in Plants . Glyoxylate Aminotransferases From Sources Other Than Plants. Aspartate: a-Ketoglutarate Aminotransferase in Plants. Aspartate: a-Ketoglutarate Aminotransferase From Sources Other Than Plants . Subcellular Location of Aminotransferases. AMINOTRANSFERASES IN PEROXISOMES . Materials and Methods. Plants Animals. . Isolation of Spinach Leaf Peroxisomes. Isolation of Chloroplasts. . Isolation of Mammalian Peroxisomes . iv Page vii ix xi 12 13 15 17 18 21 24 26 28 33 35 35 35 35 39 40 Chapter Page Assays . . . . . . 41 Polyacrylamide Gel Electrophoresis . . . 46 Starch Gel . . . . . . 48 Staining Gels for ASpartate Aminotrans- ferase . . . . . . . . . . . 49 TEAE- Cellulose Columns . . . . . . . . . 50 Isoelectric Focusing . . . . . . . . . . 51 Results. . . . . . . . . . . . . . . . . . . 54 Serine: Glyoxylate Aminotransferase Assay. . . . . . . 54 Subcellular Location of Aminotransfer- ases in Spinach Leaves . . . . . . 60 Characteristics of the Peroxisomal Glyoxylate Aminotransferases . . . . . 67 Separation of Glyoxylate Aminotrans- ferase Activities Using Ion Exchange Chromatography . . . . 74 Separation of the Peroxisomal Glyoxylate Aminotransferases Using Isoelectric Focusing . . 78 Additional Studies of Serine:Glyoxy1ate Aminotransferase . . . . . . . . . 86 Reversibility of Glyoxylate Aminotransferases. . . . 95 Spinach Leaf Aspartate: a-Ketoglutarate Aminotransferase . . . . . 99 Summary of Spinach Peroxisomal Amino- transferases . . . . 111 Preliminary Studies on the Subcellular Location of Aminotransferases in Mammalian Tissues. . . . . . . . . . . 114 Discussion . . . . . . . . . . . . . . . . . 121 II. DISTRIBUTION OF PEROXISOMAL ENZYMES AMONG LEAF CELLS . . . . . . . . . . . . . . . . 136 Materials and Methods. . . . . . . . . . . . 137 Results. . . . . . . . . . . . . . . . . . . 138 Location of Glycolate Pathway Enzymes. . 138 Catalase . . . . . . . . . . . . . . . . 142 Chapter Page Hydroxypyruvate Reductase. . . . . . . . 142 Glycolate Oxidase. . . . . . . . . . . . 153 Discussion . . . . . . . . . . . . . . . . . 153 BIBLIOGRAPHY. . . . . . . . . . . . . . . . . . . . . 155 APPENDIX. . . . . . . . . . . . . . . . . . . . . . . 167 vi Table 10. 11. LIST OF TABLES Aspartate Aminotransferase Activity in Organelle Fractions. Buffer and pH Effects on Serine:Glyoxylate and Alanine:Glyoxy1ate Aminotransferase Activities D-Serine Inhibition of Glyoxylate Aminotransferases. . . Characteristics of Alanine:Glyoxy1ate Aminotransferase Peaks From a pH 5-8 Isoelectric Focusing Column. . . . . Comparison of Serine:Glyoxylate and Serine: Pyruvate Aminotransferase Activities . Effect of Various Compounds on Peroxisomal Serine:Glyoxylate Aminotransferase . Amino Acidzd-Keto Acid Exchange Reactions and Glycineza-Keto Acid Aminotransferase in Spinach Peroxisomes . Characteristics of Aspartate Aminotrans- ferase Peaks From a TEAE-Cellulose Column . . Specific Activities of Aminotransferase Activities in Isolated Spinach Leaf Peroxisomes. . . . . . . . . . . . . Distribution of Aminotransferase Activities Among Particles From.Rat Liver . . . . . The Effect of Phosphate and Triton X-100 on Rat Liver Peroxisomal Glyoxylate Aminotransferases. . . vii Page 66 68 70 85 94 96 98 109 113 115 118 Table Page 12. Distribution of Aminotransferase Activities Among Particles From Dog Kidney. . . . . . 120 13. Glycolate Oxidase Activity in Corn and Spinach Leaf Extracts. . . . . . . . . . . 147 14. Glycolate Oxidase Activity From Various Sources. . . . . . . . . . . . . . . . . . 152 viii '- no- 5. Figure 10. 11. 12. LIST OF FIGURES Isoelectric Focusing Apparatus . Serine:Glyoxylate Aminotransferase Spectro- photometric Assay. Serine:Glyoxylate Aminotransferase Radio- chemical Assay . . . . . . Distribution of Particulate Enzymes From Spinach Leaves on a Sucrose Gradient . D-Serine Inhibition of Serine:Glyoxylate Aminotransferase . Separation of Peroxisomal Enzymes by TEAE- Cellulose Column Chromatography. pH 3-10 Isoelectric Focusing of Peroxisomal Aminotransferases. pH 5-8 Isoelectric Focusing of Peroxisomal Aminotransferases. Reaction Kinetics for Serine:Glyoxylate Aminotransferase with Varying Glyoxylate Concentrations . . . . . . . . . Reaction Kinetics for SerinezPyruvate Amino- transferase Activity With Varying Pyruvate Concentrations . . . . . . . . . . . . Reaction Kinetics for Serine:Glyoxylate Amino- transferase With Varying L-Serine Concentrations . Aspartate Aminotransferase Activities on Polyacrylamide Gels. . . ix Page 53 57 59 63 73 76 80 83 88 90 93 101 Figure 13. 14. 15. 16. 17. Starch Gel Electrophoresis of Aspartate Aminotransferase Activities. Separation of the ASpartate Aminotransferase Isoenzymes by Ion Exchange Chromatography and Their Electrophoretic Patterns . . Proposed Sequence of Reactions Occuring During the Operation of the Glycolate Pathway. . . Reaction Kinetics for Hydroxypyruvate Reduc- tase and Glyoxylate Reductase from Meso- phy11 and Bundle Sheath Cells of Maize . Reaction Kinetics of Glycolate Oxidase From Mesophyll and Bundle Sheath Cells of Maize. . . . . . . . . . . . . Page 105 108 128 145 150 Bicine DCIP HEPES MES NBT PMS TEAE- Tris LIST OF ABBREVIATIONS N,N-Bis (2-hydroxyethyl) glycine 2,6 Dichloroindophenol N-2-Hydroxyethylpiperazine-N’-2- ethanesulfonic acid 2 (N-Morpholino) ethane sulfonic acid p-Nithoblue tetrazolium phenazine methosulfate triethylaminoethyl- tris (hydroxymethyl) aminomethane xi INTRODUCTION In the 1950's Calvin and coworkers (7, 26), elucidated the pathway of CO2 fixation during photosyn- thesis and found the first product of 002 fixation to be 3-P-g1ycerate. The 3-P-g1ycerate was reduced to glycer- aldehyde-B-phosphate, which was then converted to pentose and hexose, mono- and diphOSphate esters and finally the C02 acceptor molecule, ribulose-l,5-diphosphate was regenerated. This reductive pentose pathway is considered to be the only photosynthetic pathway for net CO2 fixation in higher plants and algae. During the past 5 years, other investigators such as Hatch and Slack (68, 70) have shown that many plants do not initially fix C02 into 3-P-glycerate, but that oxaloacetate, along with aspartate and malate are the first labeled products of 14C02 fix- ation. This system is called the C4-dicarboxylic acid pathway or for brevity the C4-pathway. This pathway appears to function for C02 trapping, storage and trans- port, but it does not result in net 002 fixation. Plants which possess a substantial amount of the 04-pathway 1 2 activity are often called C4-p1ants while plants with mainly the reductive pentose phosphate pathway are often called C3-p1ants. However it must be kept in mind that even in C4-plants, net COz-fixation occurs by the same process, namely the reductive photosynthetic carbon cycle or C3-pathway. During C02 fixation via the C3-pathway consid- erable quantities of glycolate are formed (152). This glycolate is one of the early products of C02 fixation and is thought to come from one of the diphosphate esters in the reductive pentose pathway (152). The metabolism of glycolate has been mainly studied by Tolbert and co- workers (72, 80, 126). Glycolate is rapidly converted to glycine and serine and can be further metabolized to glycerate and on to sugars. This sequence of reactions from glycolate to sugars is known as the glycolate path- way and this pathway may carry as high as 50% of the total CO fixed by the plant (176). Most of the enzymes of 2 the glycolate pathway are located in a subcellular or- ganelle called the peroxisome. The leaf peroxisomes which were first isolated from plants by Tolbert 25 El, (156) were similar to the rat liver peroxisomes previously described by de Duve and coworkers (42). The enzymes which have been found in the leaf peroxisomes include 3 catalase, glycolate oxidase, hydroxypyruvate reductase, malate dehydrogenase and at least four aminotransferase activities which are either associated with or are a part of the glycolate pathway (154). During the operation of the glycolate pathway in leaf tissue, 02 is consumed both during glycolate bio- synthesis in the chloroplast and during its oxidation, by glycolate oxidase in the peroxisomes. Carbon dioxide is released intzhe conversion of glycine to serine in the mitochondria. This overall uptake of 02 and formation of C02 in the light has been called photorespiration and the operation of the glycolate pathway will account for the total phenomenon (154). Photorespiration counteracts the net gain of photosynthesis in the plant and thus appears to be a wasteful process. It has been found that C4-plants, which appear to grow faster than C3-plants, do not exhibit photoreSpiration i.e., they do not lose C02 in the light (55). The C4-p1ants have also been reported to contain lower levels of the glycolate pathway enzymes and thus may not be expected to show as much photorespiration (120, 157). However, ijrkman and Gauhl (12) found that C4-plants contained bundle sheath cells which did not readily break during homogenization, but if they were broken, they contained the enzymes of the 4 C3-pathway of C02 fixation. Since the glycolate pathway and photorespiration were associated with the C3-pathway, one might expect the glycolate pathway enzymes to be also located in the bundle sheath cells of C4-plants. The research reported in this thesis is divided into two chapters. Chapter I deals with the aminotrans- ferase activities in peroxisomes of spinach leaves, dog kidney and rat liver. Of the four main aminotransferase activities in spinach leaf peroxisomes, aspartate amino- transferase and serine:pyruvate aminotransferase activities are catalyzed by separate proteins. A more difficult task has been to determine whether the other aminotrans- ferase activities are also catalyzed by these two enzymes or if other separate proteins are present. In this chap- ter, some of the characteristics of the spinach peroxisomal aminotransferases and some preliminary studies of the aminotransferase activities in mammalian peroxisomes are reported. An understanding of the number and character- istics of the peroxisomal aminotransferases has helped elucidate the function of peroxisomes (154). Aminotrans- ferases have important functions in the metabolism of cells, for they are involved in nitrogen metabolism and serve as a metabolic link between the amino acids or proteins and the carbohydrate pools. The second chapter in this thesis describes research on the distribution of the peroxisomal enzymes 5 among the leaf cells of C4- and C3-plants. Previous studies had indicated that low levels of the glycolate pathway enzymes were present in C4-plants (120, 157). How- ever, bundle sheath cells of the C4-plants were undoubtedly not broken in those studies so the results represented mesophyll cells and are invalid for the whole leaf. The purpose of the studies reported in this thesis was to determine if the bundle sheath cells of C4-p1ants contained the enzymes of the glycolate pathway and, if so, how did the total amount of activity of these enzymes in the C4- plants compare to the levels in the C3-p1ants. Since activity of the glycolate pathway is manifested by photo- respiration, the cellular distribution for enzymes of the glycolate pathway or peroxisomes in 04-p1ants should con- tribute to an understanding of the function of photores- piration and its regulation. Photorespiration appears to inhibit plant growth and a knowledge of the mechanism and control of it may be of great significance to agricultural production. It is hoped that this work describing both the aminotransferases in peroxisomes and the unique distri- bution of peroxisomal enzymes in the bundle sheath cells of 04-plants will help in the understanding of this sub- ject area of plant biochemistry and thereby contribute someday, to the improvement of world food production. 6 LITERATURE REVIEW CARBON DIOXIDE FIXATION PATHWAYS In the fixation of C02 by the reductive pentose phosphate pathway, 3-P-glycerate is the first product formed (7). The acceptor molecule is ribulose-l,5- diphosphate and the fixation is catalyzed by ribulose-l, 5-diphosphate carboxylase, sometimes also designated as carboxydismutase. Although oxaloacetate, malate and aspartate are the first products of C02 fixation in the C4-pathway, 3-P-glycerate is also labeled quite early in these C4-plants (68). It has been shown that the car- boxyl carbon of the 3-P-glycerate formed in these plants, was derived from the C-4 carbon of the dicarboxylic acids (81). Since the C4-plants appeared at first to contain low levels of ribulose-l,5-diphosphate carboxylase, Hatch and Slack (68) proposed a transcarboxylation reaction to form 3-P-g1ycerate from the C4 dicarboxylic acids. How- ever, Bj6rkman and Gauhl (12) and later Andrews and Hatch (3) found that the C -p1ants had levels of ribulose-l, 4 S-diphosphate carboxylase similar to those found in C3-plants. They also showed that most of the ribulose-l, 5-diphosphate carboxylase was located in the bundle sheath cells of the Ca-plants. Plants which possess the 7 04-pathway have well developed bundle sheath cells sur- rounding the vascular tissue (48, 94) and these bundle sheath cells contain chloroplasts. In contrast, the bundle sheath cells in C3-plants are few, small and with- out chloroplasts. Details of the two CO2 fixation reactions in C4-plants have been summarized (69) and so only a brief description of the pathway of CO2 fixation in Ca-plants will be given. Malate is formed in the mesophyll cells and is transported to the bundle sheath cells where it is decarboxylated by malic enzyme. The 002 is refixed by ribulose-1,5-diphosphate carboxylase, to form carboxyl labeled 3-P-g1ycerate. The pyruvate from.the malic enzyme reaction moves back to the mesophyll cell and is phosphorylated by pyruvate phosphate dikinase. The PEP can then serve as an acceptor for another molecule of C02. The net fixation of C02 occurs via the C3-pathway with the C4-cycle serving as the initial fixation reaction. Thus the 04-cyc1e may serve as a CO concentrating or trans- 2 port mechanism to the ribulose diphosphate carboxylase site. In C3-plants, carbonic anhydrase may function in a similar capacity to facilitate the transport of C02 to the ribulose diphosphate carboxylase. It has been found that Chlorella grown at high CO2 concentrations 8 could not readily fix C02 at lower C02 concentrations until after the level of carbonic anhydrase in the cells had increased (65). Likewise, inhibition of carbonic anhydrase inhibits photosynthesis (64). It should be noted that in plants with the C4-pathway of C02 fixation, both the meso- phyll cells and the bundle sheath cells are involved in the fixation of the C02. It is interesting that Rhoades and Carvalho in 1944 (131) suggested that the mesophyll cells of corn fixed the CO2 while the bundle sheath cells stored the carbohydrate. They observed that in variegated corn leaf tissue, the normal bundle sheath cells which were next to mutant mesophyll cells did not accumulate starch whereas the normal bundle sheath cells lying next to normal mesophyll cells did accumulate starch. M083 and Rasmussen (111) found that in short term 14CO2 fix- ation nearly all of the radioactivity in corn leaves was quickly transported into the bundle sheath cells. These and other experimenters have established that there is a rapid transport of photosynthetic products between the two types of cells (11, 121) but the mechanisms for these transport steps are not well elucidated. Each of the two cell types in 04-plants contain chloroplasts which are both morphologically and biochem- ically distinguishable. Laetsch and coworkers (92, 93, 94) 9 as well as other authors (15, 47) have shown that the bundle sheath chloroplasts are larger, contain more starch grains and in certain Species contain no grana whereas the mesophyll chloroplasts are smaller, do not contain many starch grains and always contain grana. The mesophyll chloroplasts contain PEP carboxylase, NADP- malate dehydrogenase, pyruvate phosphate dikinase, adeny- late kinase, pyrophosphatase and glycerate kinase (145, 146). The bundle sheath chloroplasts contain the enzymes of the C3-pathway, including ribulose diphosphate carboxy- lase and ribulose-S-phosphate kinase. They also contain malic enzyme (145, 146). Plants which do not possess the C4-pathway, i.e. C3-plants, do not have well developed bundle sheath cells which contain chloroplasts (47). Instead, all chloroplasts are of one type and have the characteristics of the bundle Sheath chloroplasts in the 04-plants. PHOTORESPIRATION PhotoreSpiration is defined as the uptake of 02 and the release of CO2 in the light. A distinguishing feature of C4-plants is that they do not release C02 in the light. However, they do consume 02 in the light and they also metabolize glycolate and glycine to CO2 (86, 178) 10 and thus they do possess photorespiration (53, 77, 78). Krotokov, Nelson and coworkers (54, 55, 160) extensively studied photorespiration by measuring a C02 burst when leaves were transferred from light to dark. The burst was interpreted to be due to photoreSpired 002, which normally would be refixed in the light. These authors observed that leaves of corn, a C4-p1ant, did not Show a C02 burst while soybean and tobacco which are C3-plants did release C02. Photorespiration has been shown to be a different process than dark or mitochondrial respiration. Krotkov and colleagues (55, 160) observed that at low 02 concentrations photorespiration was severely inhibited but the mitochondrial respiration was not affected. Photorespiration was also found to respond differently to C02 concentrations and temperature (63, 74, 75, 82). In the process of photorespiration, 02 is consumed and CO2 is released while in photosynthesis, 02 is formed and C02 is fixed. BjBrkman 25 a1, (13) vividly demon- strated the apparently wasteful process of photorespiration by growing beans and corn in an atmOSphere of 5% 02 where the rate of photorespiration would be very low. The beans grew twice as fast in the 5% 02 as compared to beans grown in air. However, the growth of corn was not affected by the 5% O2 atmOSphere. 11 Another method of detecting photorespiration in plants is by measuring the C02 compensation point (54). The C02 compensation point is the concentration of CO2 at which the rate of photosynthesis equals photorespir- ation in a closed system. Plants which have the C3- pathway of CO fixation have C02 compensation points of 2 40-60 ppm CO to as high as 155 ppm C02 for leaves of 2 trees, while C4-p1ants have values of 0-5 ppm COZ. Moss and coworkers (29, 108) used the low compensation point of corn to test plants for the presence of photorespir- ation. When a C3-plant was placed in an illuminated closed chamber with corn, the C3-plant died in 2-5 days. The corn leaf lowered the CO2 concentration below the 002 compensation point of the C3-plant, and the C3-plant lost CO2 through photorespiration while the corn grew on this C02. This closed chamber system was used to test 2500 soybean varieties and a couple of thousand wheat varieties looking for a natural mutant which did not possess photo- respiration. The results were negative. ijrkman and coworkers (l4, l9) crossed a C3-plant and a C4-plant of the Atriplex family and observed that all of the F1 and F2 generations possessed photoreSpiration. Zelitch (178), Moss (110) and Goldsworth (63) 14 showed that recently fixed CO2 was released into 002 12 free air at a faster rate in the light than in the dark. They also observed the a-hydroxypyridinemethane sulfonic acid inhibited the release of 14C02 in the light. This sulfonic acid was known to be an inhibitor of glycolate oxidase (175, 177) and this result was consistent with the hypothesis that glycolate was the substrate of photo- respiration. GLYCOLATE PATHWAY Glycolate is one of the early products of photo- synthesis and is probably formed from one of the sugar diphosphate intermediates of the C3-pathway of C02 fix- ation (61, 152). The optimum conditions for the formation of glycolate are high light and a high [OZJIECOZJ ratio (125, 152). Although the mechanism of glycolate biosyn- thesis is not known, it has recently been shown that spinach leaves incorporate exogenously supplied 02 into the carboxyl group of glycine and serine which is derived from the carboxyl group of glycolate (Andrews, Lorimer and Tolbert, in manuscript). The reaction sequence of the glycolate pathway has been mainly elucidated by Tolbert and coworkers (72, 80, 126, 158). A Specific P-glycolate phOSphatase is located with the chloroplast and may be involved in the transport of glycolate out of the l3 chloroplast (132). The glycolate is oxidized to glyoxy- late by the FMN dependent glycolate oxidase in the peroxi- somes (181). The glyoxylate is converted to glycine by the irreversible g1utamate:glyoxylate aminotransferase (85) and then two molecules of glycine can be converted to one serine plus CO2 and ammonia. Hess and Tolbert (72) showed that in tobacco leaves the glycolate formed during short term 14C02 fixation was uniformly labeled, as were glycine and serine whereas 3-P-g1ycerate was carboxyl labeled. The serine is quite rapidly converted to glycerate and sugars (80, 126). The glycolate pathway is a gluconeogenic pathway, but 25% of the carbon is lost as C02 in the con- version of two glycines to one serine. This C02 loss has been postulated to be the source of the photoreSpired co2 (154). PEROXISOMES Tolbert g£_§1, (156) established that part of the glycolate pathway enzymes were located in a distinct organelle, called the peroxisome. The term peroxisome was used because the leaf organelle was similar to the perox- isomes which were isolated earlier from rat liver by de Duve and coworkers (41, 42). Peroxisomes have been isolated from the leaves of Spinach, wheat, Amaranthus, sunflower, Swiss l4 Chard, bean, pea and tobacco (157). Other sources are rat liver and kidney, chicken liver and kidney, frog liver and kidney, Tetrahymena, yeast (40) and sunflower cotyledons (Schnarrenberger, Oeser and Tolbert, in press). Peroxisomes have a single membrane, a finely gran- ularnatrix and sometimes contain a crystalline core (41, 57, 67). The enzyme composition of the peroxisomes varies among Species and among tissues within a Species. All peroxisomes contain catalase and most contain flavin oxi- dases (154). de Duve and coworkers (42) first isolated peroxisomes from rat liver by isopycnic centrifugation in sucrose gradients. The mammalian peroxisomes contained catalase, D-amino acid oxidase, L-amino acid oxidase, urate oxidase and a-hydroxyacid oxidase (9). The Spinach leaf peroxisomes were also isolated by glycolate oxidase (d- hydroxyacid oxidase) and NADH-hydroxypyruvate reductase (156). Frederick and Newcomb (56) and Vigil (167) have shown that the crystalline core, sometimes observed in plant peroxisomes, appears to be catalase. In addition to the enzymes mentioned above, Spinach leaf peroxisomes have been found to contain NAdealate dehydrogenase (173), NADP-isocitrate dehydrogenase (174), g1utamate:glyoxylate aminotransferase, alanine:glyoxy1ate aminotransferase, serine:g1yoxylate aminotransferase (85), serine:pyruvate 15 aminotransferase and aspartate:d-ketoglutarate amino- transferase (174). Another plant organelle which has the same physical characteristics of the peroxisomes but contains the enzymes of the glyoxylate cycle is called the glyoxysome. This organelle was first isolated from castor bean endosperm by Breidenbach and Beevers (22). The glyoxysomes which have also been isolated from the seedlings of several other fat storing plants (154) contain all of the enzymes needed for the conversion of fatty acids to succinate and they also contain catalase and glycolate oxidase (33, 34). The function of the glyoxysomes is to convert the fatty acids in the storage tissues of seedlings to carbo- hydrates. One of the functions of leaf peroxisomes is to form glycine which can then be converted to serine. The serine can go on to glycerate and then on to carbohydrates. Other possible functions of peroxisomes have been described in a recent review by Tolbert (154). AMINOTRANSFERASES The aminotransferase reaction was first observed in animal tissue in 1937 by Braunstein and Kritzmann (20), two Russian biochemists. In the early studies of amino- transferases, it was thought that only the transamination 16 reactions involving glutamate, aspartate and alanine were of physiological importance (106). AS the understanding of amino acid metabolism developed, however, it was recog- nized that most, if not all, naturally occurring amino acids were involved in physiologically important transam- ination reactions. Braunstein and Kritzmann preferred to call their enzymes aminopherases. The English writing authors, how- ever, preferred transaminases or aminotransferases and these are now the officially recognized names (76). The naming of aminotransferases has varied over the years. Some authors used the two amino acid designation, i.e. g1utamatezaSpartate aminotransferase while others have used the amino acidza-keto acid as in glutamatezoxaloacetate aminotransferase. One has to keep in mind that most, if not all, aminotransferase reactions are reversible and thus either designation seems adequate. The official designation (76) of naming aminotransferases is by the amino acidza-keto acid reaction involved which permits two names. However, if a-ketoglutarate is involved it should be in the name rather than glutamate. Aspartateza-keto- glutarate is accepted but glutamatezoxaloacetate is not acceptable. In this thesis, the selected designation is based upon the cited work by previous author(s). l7 EARLY STUDIES OF AMINOTRANSFERASES IN PLANTS The first aminotransferase reaction observed in plant tissue was reported by Virtanen and Laine in 1938 (168). They observed aspartate:pyruvate aminotransferase activity in peas. Other early investigators of amino- transferases in plants included Kritzmann, Adler and co- workers, and Cedrangalo and Carancante (100). Albaum and Cohen (1) observed very active aSpartate:a-ketoglutarate aminotransferase activity in oat seedlings. The pH optimum was 8.5. Rautanen (127) reported the presence of glutamate: pyruvate and aspartate:pyruvate aminotransferase activities in green plants. These two activities had pH optima of 6.9, both had temperature optimum of 41 C and both reactions were reversible. In the same tissue, he also observed aspartate:a-ketoglutarate and valine:a-ketoglutarate amino- transferase activities. Twenty plants plus wheat germ were reported by Leonard and Burris (99) to have an active glutamatezoxaloacetate aminotransferase. Generally the highest Specific activity, based upon nitrogen, was located in the roots of the plants. Wheat germ also contained alanineza-ketoglutarate aminotransferase activity. In none of this work was it known whether one general aminotransferase or many Specific enzymes were involved. 18 Wilson 35.31. (171) did an extensive study of transamination reactions in plants. Using lupine and barley, they showed that 17 amino acids could enzymatically donate their amino groups to a-ketoglutarate. Many of these activities were present in the particulate fractions of the plant extracts. Using corn radicles, pea, white and blue lupine, barley, oat and mung bean seedlings they detected transamination of alanine, aspartate, glycine, phenylal- anine, valine, leucine, methionine and histidine to a- ketoglutarate although not all plant sources exhibited all activities. Wilson 3; El. (171) also observed a glutamate: glyoxylate aminotransferase reaction in tobacco leaf juice. This reaction was dependent upon glutamate and pyridoxal-S- phosphate. The activities mentioned above described some of the earlier work with aminotransferases in plants. Many other activities have been observed and further information can be found in references 39, 71, 100 and 139. GLYOXYLATE AMINOTRANSFERASES IN PLANTS One of the first reports of a glyoxylate amino- transferase activity in plants was the g1utamate:glyoxylate aminotransferase found in tobacco leaves by'Wilson.gE‘§l. (171). They observed this activity by incubating l9 glycolate-l-14 C, glutamate and pyridoxal-S-phosphate with the leaf extract. The glycolate was very likely oxidized to glyoxylate by glycolate oxidase which was known to be present in tobacco leaves (31). The glyoxylate was then converted to glycine with glutamate serving as the amino donor. Cossins and Sinha (35) were the first to extensively study the glyoxylate aminotransferase activities in carrot tissue, corn cole0ptile, pea leaves and sunflower cotyledons. The amino donor Specificity was not great, being nearly equally active with glutamate and alanine and only slightly less active with serine and aspartate. The glutamate: glyoxylate aminotransferase activity in extracts of pea leaves had a broad pH Spectrum with good activity from pH 4.6 to 8.5. It was thought that the aspartate:glyoxylate and alanine:glyoxy1ate aminotransferase activities were reversible. These were relatively long term experiments and the percentage of glycine-14C converted to glyoxylate- 14Cwas 9.0% with oxaloacetate and 7.3% with pyruvate relative to a control value of 6.0% with no a-keto acid. In view of more recent data these results are probably not significant. In crude homogenates of wheat leaves, King and Waygood (83) observed alanine:, asPartatez, glutamate: and 20 serine:g1yoxylate aminotransferase activities as well as g1utamate:pyruvate and aSpartate:a-ketoglutarate amino- transferase activities. They partially purified the serine:g1yoxylate aminotransferase and found it had a pH optimum of 8.2. The substrate dependent Km values (fixed substrate at 13 mM) were 0.9 mM for serine and 0.25 mM for glyoxylate. The enzyme would not use D-serine, L- phosphoserine, glycolate or glycoaldehyde. No pyridoxal- 5-phosphate activation could be observed but phosphate activation was observed. In fact no serine:g1yoxylate aminotransferase activity was observed in plant extracts made in water or Tris buffer until phosphate was added to the assay. A more highly purified serine:g1yoxylate amino- transferase was obtained from oat leaves by King and coworkers (23). The Specificity of the enzyme varied according to the purification procedure used. One prep- aration which contained only a single protein band in disc gel electrophoresis showed serine:g1yoxylate, alanine: glyoxylate and g1utamate:glyoxylate aminotransferase activities. Using a different purification procedure, the alanine:glyoxy1ate and g1utamate:glyoxylate amino- transferase activities were separated from the serine: glyoxylate aminotransferase. The substrate independent 21 Km values for the serine:g1yoxylate aminotransferase were 2.88 mM for serine: and 0.508 mM for glyoxylate. The reaction kinetics showed that the mechanism of serine: glyoxylate aminotransferase was of the Ping Pong Bi Bi type of mechanism (45). Pyridoxal-S-phosphate activation was also observed with this enzyme. Kisaki and Tolbert (85) found several glyoxylate aminotransferase activities in the peroxisomes of spinach leaves. The most active, g1utamate:glyoxylate aminotransferase had a pH Optimum.of 7 and Km values for glutamate and glyoxylate of 3.6 mM and 4.4 mM, reSpectively. No pyridoxal-S-phosphate acti- vation was observed but isonicotinyl hydrazide inhibition was observed. GLYOXYLATE AMINOTRANSFERASES FROM SOURCES OTHER THAN PLANTS Braunstein and Kritzmann(20) who first discovered transaminase reactions reported glycinezd-ketoglutarate aminotransferase activity in some animal tissues. Cammarata and Cohen (27) found glycine:a-ketoglutarate activity in liver but none was observed in kidney. Ornithinezasparagine: and glutaminezglyoxylate amino- transferase activities were observed in rat liver by Meister and coworkers (105, 107). Silkworm larva have been found to contain glyoxylate aminotransferase activities with 22 alanine: glutamate, aspartate and cysteine serving as amino donors (60) while Pseudomanas aeruginosa has been reported to contain a glycine:a-ketoglutarate aminotransferase activity (6). In 1964 Nakada (114) partially purified glutamate: glyoxylate aminotransferase from rat liver. He found the aminotransferase to be irreversible and not active towards the D-isomers of the amino acids. Likewise the enzyme was not active with serine or aspartate but was active with glutamate)>alanine.>glutamine. The pH optimum.was 7.2 and the Km for glutamate was 4.6 mM and for glyoxylate it was 8.3 mM. A 91% inhibition was observed with 20 mM isonico- tinic hydrazide. Thompson and Richardson (150, 151) have charac- terized g1utamate:glyoxylate aminotransferase and alanine: glyoxylate aminotransferase from human liver. The characteristics of the human liver g1utamate:glyoxylate aminotransferase were similar to the rat liver enzyme. The pH optimum was 7.3 and the Km for both serine and glyoxylate was 2 mM. The relative rates of activity exhibited by the enzyme were glutamate-100, alanine-66, glutamine-39, methionine-l7 and arginine-lZ, while serine, valine, aspartate, histidine, phenylalanine, tyrosine and isoleucine showed no activity. Thompson and Richardson 23 (150) also found the g1utamate:glyoxylate aminotransferase to be irreversible. The human liver enzyme was not in- hibited by 1 mM isonicotinic hydrazide but 1 mM Cu'H' inhibited it 100%. The alanine:glyoxy1ate aminotransferase purified from human liver had different characteristics than the g1utamate:glyoxylate aminotransferase (151). The pH optimum was 8.4 and the Km.for alanine was 1 mM at a fixed concentration of 20 mM glyoxylate. At pH 8.4 the relative rates with various amino donors were alanine-100, serine-84, arginine-13 and DL tryptophan-4. Eighteen other amino acids including aspartate, glutamate and D-alanine had no activity. The a-keto acceptor reactions with alanine were limited to glyoxylate-100 and hydroxypyruvate-6. Again the reaction was not reversible. Cu++ at 1 mM inhibited the alanine:glyoxy1ate aminotransferase only 16%. The observed serine:g1yoxylate aminotransferase activity was attributed to the nonsPecificity of the alanine:glyoxy1ate aminotransferase since the two activities did not separate upon purification and the reaction rate with both alanine and serine present was not equal to the sum of the two individual rates. Vandor and Tolbert (160) found that g1utamate:glyoxylate aminotransferase activity was present in rat liver peroxisomes. 24 ASPARTATE:a-KETOGLUTARATE AMINOTRANSFERASE IN PLANTS The aspartate:a-ketoglutarate aminotransferase or as it is commonly called aspartate aminotransferase, appears to be the most frequently observed aminotrans- ferase. In spite of this, there are few reports of its characterization from plant tissue. Cruickshank and Isherwood (36) reported the glutamate:aspartate aminotrans- ferase in wheat had a pH optimum of 8-8.5 and they found it to be inhibited 40% by 1 mM AgNO3. Ellis and Davies (49) purified the enzyme from cauliflower florets and found that it did not react with v-hydroxyglutamate, y-methylene- glutamate, B-hydroxyaspartate, cysteate or cysteine- sulfinate. The pH optimum.was 7-8 and the Km values for glutamate, oxaloacetate, aspartate and d-ketoglutarate were 36mM, 0.08 mM, 7.2 mM and 0.66 mM reSpectively (37). They measured the specificity of the enzyme by observing amino acid inhibition of the aspartate + d-ketoglutarate reaction. No inhibition was found with 12 amino acids including DL-serine, glycine, D-alanine or L-alanine. Verjee (165) determined the following substrate independent Km values for the aSpartate:d-ketoglutarate aminotransferase from wheat germ: glutamate, 4.4 mM; oxaloacetate, 0.05 mM; aspartate, 0.8 mM and a-keto- glutarate, 0.5 mM. These values are lower, especially for 25 glutamate, than those reported by Davies and Ellis (37). The molecular weight of the wheat germ aspartateza-keto- glutarate aminotransferase was reported to be 75,000 I 5,000. The pH optimum was 8-8.5 and pyridoxal-S-phOSphate acti- vation was observed (166). The particulate and soluble fractions in germin- ating pea cotyledons contained aspartate:a-ketoglutarate aminotransferase (172). Wong and Cossins (172) reported the soluble enzyme had a pH optimum of 8 whereas the par- ticulate had a broad range with activity from 6.8 - 8.5. The soluble enzyme was not active with pyruvate, hydrox- pyruvate or glyoxylate in place of a-ketoglutarate. Like- wise, aspartate could not be replaced by D-aSpartate, aSparagine, serine, leucine, glutamine, y-aminobutyrate, alanine or glycine. The apparent Km values for the par- ticulate and soluble enzymes were Similar, although the same concentration of fixed substrate was not used in the assays for both of the enzymes. Characterization of aspartate:a-ketoglutarate amino- transferase from the particulate and soluble fractions of germinating pumpkin cotyledons has been reported by Splittstoesser (148). Both enzyme fractions had pH optimum.of 8. The soluble enzyme fraction was inhibited very strongly by hydroxylamine and to a lesser extent by 26 p-hydroxymercuribenzoate and sodium bisulfite. The most recent report of aSpartate:d-ketoglutarate aminotransferase activity in plants has been from Yamazaki and Tolbert (174). They studied the subcellular location of the enzyme and their work.will be reviewed in the section on the sub- cellular location of aminotransferases. ASPARTATE:a-KETOGLUTARATE AMINOTRANSFERASE FROM SOURCES OTHER THAN PLANTS The aspartate:d-ketoglutarate aminotransferase appears to be ubiquitous and thus only a few references will be cited. Green gt 31. (66) showed the pig heart aspartate:a-ketoglutarate aminotransferase to be quite Specific. Alanine, leucine, serine and methionine could not replace aspartate, and glutamine could not replace glutamate. In 1947 O'Kane and Gunsalus (119) separated glutamate:aspartate aminotransferase from glu- tamate:alanine aminotransferase, and proposed that the aspartate:a1anine aminotransferase activity in crude hmmogenates was an artifact caused by the presence of the two glutamate aminotransferases. Cammarata and Cohen (28) partially purified the glutamate:axaloacetate aminotransferase from pig heart muscle and found that sixteen amino acids including gly- cine and glutamine would not react with a-ketoglutarate. 27 Boyd (17) found that mammalian tissues had two isoenzymes of aspartate aminotransferase and that they had different pH curves and Km values. The soluble form had a pH optimum of 8.5 while the mitochondrial form was active from pH 5-9. The soluble enzyme had Km.values for d-ketoglutarate and aspartate of 0.2 mM and 2.1 mM respec- tively, whereas the mitochondrial enzyme had Km values of 1.0 liand 0.47 mM for a-ketoglutarate and aspartate reSpectively. Similar Km values have been reported by several authors (52, 118). Velick and Vavra (164) measured the Km values of all four substrates for the soluble pig heart enzyme and obtained the following results: Km (aspartate), 0.9 mM; Km.(d-ketoglutarate), 0.1 mM; Km (glutamate), 4 mM and Km (oxaloacetate), 0.04 mM. The kinetic characteristics for both the soluble and mitochondrial isoenzymes from beef liver were reported by Morino 25 31. (109). The Km values of glutamate and oxaloacetate were similar for both isoenzymes but instrument sensitivity may have been a limiting factor in their determinations. The observed Km values (mM) for glutamate, a-ketoglutarate, aspartate and oxaloacetate were for the soluble enzyme 20, 2.0, 0.4 and 0.02 and for the mitochondrial enzyme, 20, 0.3, 5.0 and 0.05 respectively. 28 The kinetic characteristics of aspartate:a- ketoglutarate aminotransferase appear to be affected by ions. Boyde (18) observed that as the concentration of phOSphate was increased, the Km of both aSpartate and a-ketoglutarate increased. This trend was observed with both isoenzymes but the affect varied between the iso- enzymes. Nisselbaum (117) reported that phOSphate and sulfate inhibit the rat liver mitochondrial aspartate aminotransferase but increased the activity of the soluble enzyme. Cl- increased the activity of the mitochondrial enzyme but had no affect on the soluble enzyme. Other authors have reported anion and cation affects on aSpartate aminotransferase (10, 161, 169). It has been reported that phOSphate and other anions interfere with the re- constitution of apoaspartate aminotransferase with pyridoxal- 5-phosphate (50, 143). SUBCELLULAR.LOCATION OF AMINOTRANSFERASES In 1960, Bone and Fowden (16) reported aSparate: a-ketoglutarate and alanine:a-ketoglutarate aminotrans- ferase activities in mung bean mitochondria. Particulate aSpartate:a-ketoglutarate has also been found in pumpkins (149) and pea cotyledons (172). Mukerji and Ting (112) reported that glutamate:axaloacetate aminotransferase was 29 located in nonaqueously prepared chloroplasts from cactus phylloclades. They also reported the aminotransferase was present in the mitochondria and chloroplasts prepared on sucrose gradients. However, NADH-malate dehydrogenase had the same distribution as glutamatezoxaloacetate and it has since been shown that chloroplasts do not contain NADH~malate dehydrogenase (133, 173) and so the presence of the aminotransferase in chloroplasts was not definitely established. Tolbert and Yamazaki (158) first reported that aspartate:a-ketoglutarate aminotransferase was located in the peroxisomes, mitochondria and chloroplasts of Spinach leaves. They also found that the peroxisomes contained three isoenzymes of the aminotransferase (174). It was thought that one of these isoenzymes was due to mitochon- drial contamination. COOper and Beevers (33) also found that aSpartate aminotransferase was located in both the glyoxysomes and the mitochondria of castor bean endosperm. Breidenbach (21) reported D-aspartate aminotransferase activity in glyoxysomes. Mathieu (104) reported glutamate: pyruvate and glutamatezoxaloacetate aminotransferase in nonaqueous prepared chloroplasts from.Kalanchoe. However, mitochondria and possibly other subcellular organelles are not well separated from chloroplasts by such a 30 procedure. Santarius and Stocking (137) also using non- aqueous isolation procedures found glutamate:oxaloacetate aminotransferase was located both inside and outside the chloroplasts. Yamazaki and Tolbert (174) found serine:pyruvate aminotransferase activity in spinach leaf peroxisomes. Earlier, Kisaki and Tolbert (85) had reported spinach peroxisomes contained several glyoxylate aminOtransferases with glutamate, alanine and serine being the best amino donors. The particulate serine:g1yoxylate aminotransferase from spinach leaves was first shown to be located only in the peroxisomes (129, this thesis). Serine:g1yoxylate, g1utamate:glyoxylate and aspartate:a—ketoglutarate amino- transferases have been found in the microbodies of castor bean endOSperm and sunflower cotyledons (Schnarrenberger, Oeser and Tolbert, in press). Alanine:glyoxylate and g1utamate:glyoxylate aminotransferase have been found in both the supernatant and particulate fractions of sun- flower cotyledons (35). The cationic and anionic isoenzymes of aSpartate aminotransferases from mammalian tissue were identified as soluble and mitochondrial forms by Boyd (17). These two isoenzymes have been found in many tissues and have been extensively studied as was described previously in 31 the section Aspartate:a-Ketoglutarate Aminotransferase From Sources Other Than Plants. de Duve and coworkers (8) and Muller (113) reported that aSpartate aminotransferase and alanine aminotransferase were not located in the per- oxisomes of rat liver but were located in the mitochondria. However, with the use of electron microscopy, Papadimitriou and VanDuijn (124) observed aspartate aminotransferase activity on the peroxisome membrane as well as in the mitochondria, and on the plasma membranes of many cells including erythrocytes and bacteria. Lee (96) found that some of the kidney aSpartate:a-ketoglutarate aminotrans- ferase was located in the mitochondria but the enzyme was concentrated in the subapical vesicles and suggested it may be involved in ammonia excretion. The mitochondrial aspartate aminotransferase has been reported to be located both in the matrix and on the mitochondrial membranes (103, 142). These differences in possible location of the aminotransferase may be related to the sticking of aSpartate aminotransferase to membranes under certain ionic conditions (130). Vandor and Tolbert (162) found g1utamate:glyoxylate aminotransferase activity in the peroxisomes of rat liver, but de Duve (40) has reported that it was not in the peroxisomes. It has recently been reported in an abstract that 32 most of the alanine:glyoxy1ate aminotransferase activity of rat liver was located in the mitochondria (147). In Neurospora, the alanine:glyoxy1ate aminotransferase was a soluble enzyme (32). Several excellent reviews covering some of the topics pertinent to this thesis have recently been pub- lished and are listed below. 'Microbodies-Peroxisomes and Glyoxysomes (154); Photosynthetic COZ-Fixation Pathways (69); The Peroxisomes: a New Cytoplasmic Organelle (41); Photorespiration (78) and Nitrogen Metabolism of Amino ‘Acids (136). CHAPTER I AMINOTRANSFERASES IN PEROXISOMES Leaf peroxisomes were first isolated by Tolbert gt a}, (156) and were found to contain most of the enzymes associated with the glycolate pathway (154, 156). In the conversion of glycolate to glycerate, two transamination reactions occur: glyoxylate is converted to glycine and the serine is converted to hydroxpyruvate. Four major amino- transferase activities have been found in spinach leaf peroxisomes. Kisaki and Tolbert (85) found glutamate: glyoxylate and alanine:glyoxy1ate aminotransferase activ- ities. The aminotransferase activities with glutamate: glyoxylate and alanine:glyoxy1ate were suggested to be from one enzyme since mixed substrate assays were not additive. Aspartate aminotransferase activity was also detected by Kisaki and Tolbert (85) but they thought it was due to the nonspecificity of the g1utamate:glyoxylate aminotransferase. Yamazaki and Tolbert (174) found spinach peroxisomes contained a serine:pyruvate aminotransferase activity. These authors also concluded that the peroxisomal aspartate 33 34 aminotransferase activity could not be accounted for by mitochondrial contamination and it could not be due to the nonSpecificity of the g1utamate:glyoxylate aminotransferase since the specific activity of the aspartate aminotrans- ferase was higher than the g1utamate:glyoxylate amino- transferase. With the use of isoelectric focusing, Yamazaki and Tolbert (174) Showed that serine:pyruvate aminotransferase in peroxisomes was a different protein than the aspartate aminotransferase. They also showed that after electrophoresis, the aspartate aminotransferase activity from peroxisomes was present as three isoenzyme bands on the polyacrylamide gels. However, one of these may have been due to mitochondrial contamination. 0f the four major aminotransferase activities reported in peroxisomes, it was shown that serine:pyruvate and aspartate:a-ketoglutarate aminotransferase were dif- ferent proteins. Whether these two proteins also exhibited g1utamate:glyoxylate and alanine:glyoxy1ate aminotransferase activities was not known. Also it was not known if the three isoenzymes of aspartate:a-keto- glutarate aminotransferase were Specific for aspartate and a-ketoglutarate or if other aminotransferases such as the serine:pyruvate aminotransferase were exhibiting some activity towards aspartate and a-ketoglutarate. The purpose of this study was to determine how 35 many different enzymes were involved in the peroxisomal aminotransferase activities and to determine some of the characteristics of the enzymes. MATERIAL AND METHODS PLANTS Spinach was either purchased at a local market or grown in growth chambers at 20 C and a light period of 11 hr. Spinach was grown in a soil-peat moss mixture and watered with Hoagland's solution at least three times a week. ANIMALS Female Spraque-Dawley rats, weighing 200-300 g, were purchased commercially and fed Purina Laboratory chow and watered ag_libitum. Kidneys from dogs (Species un- known) were furnished by the staff in the Department of Human Development. From an anesthetized animal, the kid- ney was perfused with saline solution and immediately used. ISOLATION OF SPINACH LEAF PEROXISOMES Small Zonal Rotor Procedure Peroxisomes and mitochondria were obtained by three procedures. The first procedure was used for inter- mediate size preparations of subcellular organelles and involved the use of the International Zonal Centrifuge. washed and deribbed spinach leaves weighing 250 g were 36 homogenized in a Waring blendor for 7-10 sec at high Speed. The grinding medium.was 30% (w/w) sucrose in 20 mM glycyl- glycine at pH 7.5. Unless stated otherwise, all sucrose solutions which were used for grinding media or gradients were made in 20 mM glycylglycine at pH 7.5 and the per- centages of sucrose are based upon w/w. The Spinach homogenate was squeezed through 8 layers of cheesecloth. Cellular debris and most whole chloroplasts were removed by centrifugation at 650 g for 5 min. The homogenate was then placed directly in the center of a B-30 rotor. The operation of the zonal centrifuge will not be discussed here but further information can be found in references 2, 46, 163. The gradient which was found to result in the best separation of subcellular organelles from Spinach leaves was camposed of the following sucrose solutions: 50 ml of 56% sucrose, 20 ml each of 53%, 52.5%, 52%, 51%, 50%, 49%, 44%, 43%, 42%, 41%, 40%, and 10 ml each of 35%, 30%, and 25% (46). Starting with the least dense sucrose solution, the solutions were loaded into the zonal rotor from the edge. After the gradient was in the rotor, the remaining volume was filled with 56% sucrose. The small zonal rotor (B-30) would hold approximately 560 m1. Up to 260 ml of spinach homogenate was added to the rotor through the core orifices. Centrifugation at 30,000 rpm for 2 hr separated the mitochondria and peroxisomes. The 37 rotor was unloaded by pumping water into the core and collecting 10 ml fractions from the edge of the rotor. The small zonal rotor was also used for processing up to 1 kg of Spinach. The Spinach was homogenized in 250 g batches and applied to the zonal rotor in the following manner. The homogenate from the first 250 g of Spinach leaves was placed in the rotor as previously described. Instead of centrifuging at 30,000 rpm for 2 hr, the rotor was spun at 30,000 rpm for 15 min to move the particles just into the gradient. The rotor was Slowed to 3,000 rpm and 56% sucrose was used to push the supernatant out through the core. The next batch of 250 g of spinach was loaded through the core and the process repeated. After the last spinach homogenate was added, the 2 hr centrifu- gation was run to move the organelles to their isopycnic point. Large Zonal Rotor Procedure The second procedure was the same as the first except that the large zonal rotor (B-29) with a volume of 1500 ml was used. The gradient was generally the same except that 50 ml of each sucrose solution was used. The 250 g batches of Spinach tissue were prepared in the same manner. Spinach extracts equal to 500 g of spinach leaves 38 could be placed in the rotor at one time. By using the process of applying one spinach homogenate, centrifuging a short time, removing the super- natant and applying the second homogenate, etc., 4 kg of spinach.were processed in one experiment. Simulated Zonal Procedure The third procedure was performed in the SW 25.2 swinging bucket rotor with a Beckman L-2 Ultracentrifuge and was similar to that described by Tolbert (153). The gradient was composed of 2 m1 of each of the same sucrose solutions that were used in the small zonal rotor pro- cedure. The homogenate from 20 g of spinach tissue was applied to the gradient and centrifuged for 2.5 hr at 25,000 rpm. Fractions of 1 ml were collected by puncturing the bottom of the tube and draining. It was found that a better yield of peroxisomes was obtained from 20 g of spinach tissue by grinding the tissue with a mortar and pestle rather than a mechanical blender. The tissue was first placed in a beaker containing 30% sucrose grinding media and cut into very small pieces with a scissors. The tissue was kept immersed in the grinding media during the cutting. The finely cut tissue in the grinding media was then placed in a mortar. A small 39 amount of sand was added and then the tissue was ground until no pieces of tissue remained. The homogenate was squeezed through 4 layers of cheesecloth, centrifuged at 650 g for 5 min to remove cell debris and most of the whole chloroplasts, and then placed on the sucrose grad- ient. The mortar and pestle procedure resulted in a little higher percentage recovery of peroxisomal enzymes in par- ticulate form and about twice the yield that was obtained from a Sorvall Omni-mixer. ISOLATION OF CHLOROPLASTS Most of the intact chloroplasts were pelleted by centrifuging the Spinach homogenate at 650 g for 5 min. This pellet was resuspended in the grinding medium (30% sucrose) and layered over 15 ml of 30% sucrose in a cen- trifuge tube. The chloroplasts were pelleted through this sucrose by centrifugation at 270 g for 10 min while the smaller particles remained in the supernatant. The pelleted chloroplasts were resuspended in buffer and either used directly or were centrifuged at 100,000 g for 1 hr to remove most of the chlorophyll and then the nearly color- less supernatant was used. 40 ISOLATION OF MAMMALIAN PEROXISOMES Female rats were injected with Triton WR 1339 (Ruger Chemical Co., New Jersey) 3.5 days before sacri- ficing. The rats were decapitated and the livers were perfused with either water or 7% sucrose solution before removing. Livers from 3-10 rats were immersed in 7% sucrose at 0 C, minced with either a scissor or a razor blade and homogenized in a motor drived Potter-Elvehem homogenizer for only one stroke down and back up. The homogenate was filtered through 4 layers of cheesecloth. Debris from the homogenate was removed by either centri- fuging at 270 g for 5 min or by the differential centrifu- gation procedure of de Duve 93 31, (43). The homogenate was applied to the small zonal rotor (B-30) filled with the following sucrose gradient: 40 ml, 56%; 12 ml each of 49%, 48%, 47.5%, 47%, 46.5%, 46%, 45%, 40%, 38%, 36%, 33%, 30%, 25%, 20% and 15%. Centrifugation was at 30-35,000 rpm for only 35 min. Fractions of 5 ml each were collected from the edge of the rotor. Medula and cortex tissue from a dog kidney was minced with either a meat grinder, mortar and pestle or a tissue grinder-press. Homogenization in 25% sucrose was achieved with a motor drived Potter-Elvehem homogenizer. The extract was filtered through 4 layers of cheesecloth, 41 and centrifuged for 5 min at 480 g. The supernatant was either placed directly in the zonal rotor or was first centrifuged at 39,000 g for 30 min and the pellet, resus- pended in grinding media, was placed in the zonal rotor. The velocity of the zonal rotor was increased in incre- ments of 10,000 rpm for 15 min, 20,000 rpm for 15 min and finally held at 30,000 rpm for 2 hr. The sucrose gradient was similar to the one used for the rat liver preparation. Fractions were collected from the edge of the rotor. ASSAYS All spectrophotometric assays were conducted with a Gilford recording Spectrophotometer at 25 C. The radio- chemical assays on plant tissue were at 25 C while the mammalian tissues were assayed at 36 C. The pH of all reagents was adjusted to the pH of the assay. Unless Stated otherwise, all amino acids were of the L form. A unit of enzyme activity is defined as that amount of enzyme which will form 1 umole of product in 1 min. Serine:Glyoxylate Aminotransferase (EC 2.6.1.-) The spectrophotometric procedure of measuring the activity of serine:g1yoxylate aminotransferase was a modification of the linked enzyme assay described by Brock EEH21° (23). The hydroxypyruvate formed in the 42 transamination reaction was enzymatically reduced by hydroxy- pyruvate reductase and the oxidation of NADH was measured at 340 mu. The assay mixture contained 0.7 m1 of 0.1 M HEPES at pH 7, 0.04 ml of 4.2 mM NADH, 0.02 ml of 5 mM pyridoxal-S-phOSphate, 0.05 ml of 20 mM sodium.g1yoxylate, 0.05 units of crystalline Spinach glyoxylate reductase (Sigma), 0.05 ml of 0.4 M L-serine and 0.12 ml of enzyme plus water. The addition of NADH was used to start the measurement of the endogenous rate and the transamination reaction was started with L-serine. An extinction coe- ficient of 6.2 x 103 cm'1 XM"1 for NADH was used. All rates were corrected for the endogenous rate of glyoxylate reduction by the hydroxypyruvate reductase. The serine:pyruvate aminotransferase activity was measured using the same protocol as in the serine:g1yoxy- late aminotransferase assay with the following changes: 0.02 ml of 4.2 mM NADH, 0.03 ml of 0.1 M sodium pyruvate and 0.16 ml of enzyme plus water. The serine:pyruvate aminotransferase procedure is a modification of the assay used by Yamazaki and Tolbert (174). Alanine:Glyoxy1ate Aminotransferase (EC 2.6.1.12) The alanine:glyoxy1ate aminotransferase activity was measured in a manner similar to the serine:g1yoxylate 43 aminotransferase and was described by Brock £3 31. (23). Lactate dehydrogenase was used to reduce the pyruvate and the oxidation of NADH was followed at 340 mu. The protocol for the reaction mixture was the same as for the serine: glyoxylate aminotransferase except that 0.03 units of lactate dehydrogenase (Sigma) and 0.05 ml of 0.4 M L-alanine were used. Glutamate:Glyoxylate Aminotransferase (EC 2.6.1.4) The g1utamate:glyoxylate aminotransferase activity was measured by the method of Kisaki and Tolbert (85). The assay mixture contained 0.7 ml of 0.1 M phosphate or cacadylate at pH 7, 0.02 ml of 5 mM pyridoxal-S-phosphate, 0.08 ml of 0.4 M L-glutamate and 0.180 ml of enzyme plus water. The reaction was started with the addition of 0.02 14C. The reaction mixtures ml of 0.25 M glyoxylate-U- were kept in ice until the glyoxylate was added and then they were placed in a water bath of 25 or 36 C. The reaction mixture reached 25 C within 60 sec. The reaction was terminated by placing the test tube in boiling water for 3 min. The reaction mixture was cooled and placed on a Dowex-l acetate column (0.6 x 2 cm) to remove excess glyoxylate-U-14C. The effluent was collected directly in a glass scintillation vial. The reaction tube and 44 Dowex column were washed twice with 1 ml of water and the column was then blown dry. Control columns showed 99.2% of the glyoxylate-U-14C stayed on the column and 100% of added glycine-14C was eluted from the column. After adding 17 ml of scintillator fluid [50 mg POPOP (phenyl-oxazolylphenyl-oxazo1ylphenyl), 4 g PPO (2,5-diphenyloxazole), 1000 m1 toluene, 1000 m1 Triton X-100] to each vial, the samples were counted in a Packard Scintillation Counter“ Quenching was observed by the channel ratio method. The same volume of glyox- ylate-U-14C that was used in each assay, was added directly to a counting vial containing 3 ml of water. This standard had the same degree of quenching as the reaction mixtures and was used to calculate the specific activity of the glyoxylate-U-IAC. The specific activity of glyoxylate-U- 140 varied between experiments. Other Radiochemical Aminotransferase Assays Serine:g1yoxylate aminotransferase and alanine: glyoxylate aminotransferase were also assayed by a radio- chemical procedure. The same protocol as in the g1uta- mate:glyoxylate aminotransferase assay was used except that L-serine or L-alanine was substituted for the L- glutamate. The radiochemical assay was also conducted using 45 14C labeled amino acids. In this case, the reaction was started with the amino acid. The reaction was stopped by the addition of 0.35 ml of 10% trichloroacetic acid. The reaction mixtures were placed on Dowex 50-H+ columns to remove the excess amino acid and the column washed twice with 1 ml of 10 mM HCl. The rest of the procedure was the same as previously described. In‘a control experiment 100% of added g1yoxylate-U-14C was recovered in the effluent. Exchange reactions between an amino acid and its corresponding d-keto acid were conducted using the same basic radiochemical protocol of glutamate: glyoxylate aminotransferase. The assay mix contained a 14C labeled amino acid and 8 mM of the corresponding a-keto acid. The 14C labeled final concentration of 0.8 mM a-keto acids were eluted frmm the Dowex 50-H+ column as described above. Aspartate:d-Ketoglutarate Aminotransferase (EC 2.6.1.1) Aspartate aminotransferase activity was measured as described by Yamazaki and Tolbert (174). The oxalo- acetate formed in the transamination reaction was reduced by NADH and malate dehydrogenase, and the change in con- centration of NADH was measured at 340 mu. The assay mixture contained 0.7 ml of 0.1 M HEPES at pH 7, 0.03 ml 46 of 0.1 M a-ketoglutarate, 0.02 ml of 5 mM pyridoxal-S- phOSphate, 0.02 ml of 4.2 mM NADH, 2 units of malate dehydrogenase (Sigma), 0.16 ml of enzyme plus water and 0.05 ml of 0.4 M L-aspartate. The reaction was started with the L-aspartate. Other Assays Catalase was measured by following the decrease in the absorbance of H202 at 240 mu (102). An extinction coefficient of 42 cm.1 x M.1 for H202 was used. Cyto- chrome c oxidase activity was measured as described by Tolbert gt 31. (156) and an extinction coefficient of 21.1 x 103 cm'1 x M"1 for the reduced cytochrome c was used. The cytochrome c was reduced by sodium dithionite until the ASSO/AS65 ratio was equal to 6-10. Malate dehydro- genase and hydroxypyruvate reductase were assayed by measuring the oxidation of NADH (156). Chlorophyll was determined by Arnon's procedure (4). Protein was assayed by the procedure of Lowry 2; 31. (101) with crystalline bovine serum albumin as the standard protein. POLYACRYLAMIDE GEL ELECTROPHORESIS Three stock solutions for the preparation of polyacrylamide gels were prepared according to the pro- cedure of Davis (38). Solution A.consisted of 18.3 g of 47 Tris, 0.11 ml of TEMED, 24 m1 of l N HCl and water to 50 ml (pH 8.9); solution B was composed of 14 g of acrylamide, 0.367 g of BIS and water to 50 ml; solution C was 0.07 g of ammonium persulfate dissolved in 50 m1 of water. Stock solution A and B were stored for up to one month but solution C was prepared freshly once a week. The gels were prepared by mixing 1 part solution A, 2 parts solution B, 1 part water and 4 parts solution C. The gel mixture was placed in glass tubes to a height of 7 cm and then a layer of distilled water was carefully layered on top of the gel. The gels were set in 30-45 min and then they were placed in the cold room for at least 1 hr prior to adding the sample. Up to 0.1 ml samples, containing at least 10% (w/w) sucrose, were layered onto the top of the gels. Samples directly from the sucrose gradients gave the same electrophoretic results as samples which were treated by osmotic shock or deter- gent to break the organelles prior to electrophoresis. In the case of the samples from the TEAE-cellulose columns, solid sucrose was added to the sample before it was applied to the gel. The electrode buffer solution was cmmposed of 3 g Tris, 14.4 g glycine and water to 5 1 (pH 8.3). Electrophoresis was generally run for 2 hr at 2-5 ma per gel. Tracking dye (Bromo Phenol blue) was 48 either mixed with one sample or placed on a separate gel which had no enzyme sample. STARCH GEL A method similar to that of Fine and Costello (51) was used to prepare the starch gel. The starch mixture, consisting of 36 g hydrolyzed starch.(Connaught Medical Research Lab., Toronto, Canada) and 300 ml of 5 mM Tris-38 mM glycine at pH 8.3 was stirred continuously and heated until it reached 70-80 C. The hot mixture was quickly poured into a 2 1 suction flask which was sitting in hot tap water and a vacuum was applied until very few bubbles remained. The gel was then poured into a Plexiglas form. A mold for sample wells was placed on the gel. The gel was left at room temperature until firm (about 30 min), covered with Saran Wrap and placed in the cold room. Gels were left in the cold room.l—12 hr before use. Approximately 0.1 m1 of sample was placed in each sample well. The electrode buffer was the same as that used in the gel. Sponge wicks were placed on top of each end of the gel to form the bridge to the buffer tanks. Electro- phoresis was conducted at 400-500 volts, for 10-16 hr. 49 STAINING GELS FOR ASPARTATE AMINOTRANSFERASE A Stain Specific for oxaloacetate has been reported (5) and was used to stain electrophoretic gels for aspar- tate aminotransferase (24). The staining solution was a modification of the one used by Yamazaki and Tolbert (174) and consisted of 0.05 ml of 0.4 M L-aspartate, 0.05 ml of 0.1 M d-ketoglutarate, 0.02 ml of 5 mM pyridoxal-S- phosphate, 0.8 ml 0.1 M TES at pH 7 and 0.08 ml of water containing 1 mg of Fast Violet B. The dye was added to the rest of the staining solution just before applying the staining solution to the gels. The staining solution is not sensitive to light, however, it did turn reddish- brown with time, and it would not stain the enzyme if it was mixed too early before use. The polyacrylamide gels were immersed in staining solution in test tubes. Starch gels were sliced in half, and a piece of filter paper was placed on the eXposed surface to absorb some of the mois- ture. The first filter paper was removed and a second filter paper which had been soaked with the staining solution was placed on the gel surface. The purple Stains in both types of gels were fixed by treatment with 5% acetic acid. 50 TEAE-CELLULOSE COLUMNS TEAE-cellulose was prepared by washing with 0.5 N KOH until the effluent was colorless, rinsing with water until neutral, washing with 0.5 N HCl until washes were colorless and again rinsing with water until neutral. The TEAE-cellulose was mixed with 5 mM Tris-H01 at pH 8.3 and poured into a column. The column was equilibrated overnight with 5 mM Tris-HCl at pH 8.3. The sample which was applied to the TEAE-cellulose column was a high Speed supernatant of broken peroxisomes. Spinach leaf peroxi- somes isolated on sucrose gradients, were broken by dilution or dialysis and the broken peroxisomes were centrifuged at 100,000 g for 1 hr to remove the membranes. The peroxisomal membranes were used by R. Donaldson for other assays and their removal was probably not necessary for my work. The supernatant (150-220 ml) was applied to the TEAE-cellulose column. The column was washed with approximately two void volumes to remove the glycolate oxidase. The aminotransferases were eluted with a 200 ml linear gradient of 0 to 0.3 M KCl in 5 mM Tris-HCl at pH 8.3. 51 ISOELECTRIC FOCUSING G. N. Godson (62) described a procedure for running isoelectric focusing in Small columns and a modified column, most of which was constructed by G. Lorimer, was used in the present studies (Figure l). The column was a 5 m1 pipet with the tip cut off. Polyacrylamide gel was used to form a plug in the bottom of the column. Layered onto the polyacrylamide plug was 1 ml of cathode solution (5% ethanolamine in water) containing 35% (w/w) sucrose. A linear gradient of 10-30 (w/w) sucrose containing up to 5 mg of peroxisomal protein and 1.14% of either pH 3-10 or pH 5-8 Ampholine (LKB Produkter.AB) was placed in the column. Anode solution (5% H3PO4 in water) was added to fill the column. All of the weight of the solution was on the polyacrylamide plug, and it may be advantageous to make a slight constriction at the cut off end of the pipet so the polyacrylamide plug does not slip out of the column. A 250 m1 graduated cylinder filled with cathode solution was placed around the column. If the cylinder is placed around the column before the column is filled, the polyacrylamide plug will tend to move up into the column. The cathode solution was stirred to help dissipate the heat from the column. Platinum electrodes were immersed in the two electrode solutions. Unless stated otherwise, 52 Figure l Isoelectric Focusing Apparatus A 5 ml pipet with the tip cut off served as the column. A polyacrylamide plug kept the sucrose gradient in the column. The one electrode solution was layered on top of the sucrose gradient. The other electrode solution was in the 250 m1 graduated cyliner but a portion of it was also placed on top of the polyacrylamide plug. The solu- tion in the graduated cylinder was stirred to help dissi- pate the heat. 53 Platinum Wire 5ml Pipet —— 250 ml Graduate Cylinder —— Sucrose Gradient " -— Polyacrylamide Plug £——Stirring Bar 54 the isoelectric focusing was run at 500 volts, at (5 ma for 9 hr in the cold room. To drain the column, a capil- lary tubing was carefully forced through the polyacrylamide plug and approximately 0.1 ml fractions were collected. RESULTS The aminotransferase activities of peroxisomes isolated from three different biological sources have been examined. Most of the studies were done with spinach 5 leaf peroxisomes but other sources of peroxisomes were checked to determine whether all microbodies contained the aminotransferases. The four aminotransferase activ- ities to be considered are: g1utamate:glyoxylate, serine: glyoxylate, alanine:glyoxy1ate and aspartate:a- ketoglutarate. SERINE:GLYOXYLATE AMINOTRANSFERASE ASSAY In the serine:g1yoxylate aminotransferase reaction, hydroxypyruvate and glycine are the products. The activity was assayed Spectrophotometrically by linking the reaction with hydroxypyruvate reductase which, however, also re- duces glyoxylate. The glyoxylate concentration was kept at 1 mM which.was low enough that the endogenous rate of the glyoxylate reduction by hydroxypyruvate reductase was low and could be subtracted. This concentration of 55 glyoxylate was, however, higher than the Km (glyoxylate) value of 0.15 mM for the aminotransferase. The NADH- hydroxypyruvate reductase from spinach has a Km for glyoxylate of 50 mM and a Km for hydroxypyruvate of 0.05 mM (89). The difference in these Km values permitted hydroxypyruvate reductase to be used for measuring the serine:g1yoxylate aminotransferase activity. The assay was linear with both time and enzyme concentration (Figure 2). The reaction was always started with L-serine so the endogenous rate of glyoxylate reduction could be measured. The reaction was dependent upon enzyme, L- serine and glyoxylate. Serine:g1yoxylate aminotransferase activity was also measured by a radiochemical procedure. Measuring the formation of glycine-14C from glyoxylate-U-IAC or the formation of hydroxypyruvate-14C from serine-U-14C gave similar results. The reaction was dependent upon enzyme, L-serine and glyoxylate. The reaction was linear with protein and appeared to be linear with time although the line did not go through zero (Figure 3). The reason for this is not known. Since the reaction was linear with protein up to the formation of 1.2 umoles of glycine-14C, substrate concentrations were probably not limiting. Glyoxylate is quite easily converted to glycine 56 Figure 2 Serine:Glyoxylate Aminotransferase Spectrophotometric Assay Spinach leaf peroxisomes, isolated on a sucrose gradient, were used. The assay mixture contained 70 umoles of HEPES at pH 7, 0.17 umoles of NADH, 0.1 umoles of pyridoxal-S-phosphate, l umole of glyoxylate, 0.05 units of glyoxylate reductase, 20 umoles of L-serine and enzyme in a total volume of 1 ml. l 4:. C? 57 A34o l l 1 (III— I 2 3 4 TIME (MIN) NMOLES X MIN m on o o I F 5 l l J l l 20 4O 60 80 IOO pl ENZYME 58 Figure 3 Serine:Glyoxylate Aminotransferase Radiochemical Assay Isolated spinach leaf peroxisomes were used for the assays. The reaction mixture contained 70 umoles of either phosPhate or cacadylate at pH 7, 0.1 umoles of pyridoxal-S-phosEhate, 32 umoles of L-serine, 5 umoles of glyoxylate-U- 4C and enzyme in a total volume of 1 ml. 59 I I I I 30 20 IWME “MW“ IO _ 5 A O _ ?m 0 O mmaoza 4wuo< CH muw>fiuo< mcofiuumum . meowuumuh huw>wuu< aowuomum umdeonoasu umdeouoanu CH >uw>fluo< CH huw>auo< ofimwoomm mHHmamwuo maoaz emxoum Hmfiupcosoouwz HmEomeoumm .cowumadoamo mamamm m pom uxau mam .maowuomum oHHmcmwuo oSu mo some aw mahuco Hoxume some mo uqsoem can cons ummmn oua3.sowumcwswucoo mo mawmuamouad osH uwxouma .ommamumo "mHoS mmHHmameo onu pom mumxumz .mummHmouoasu .Haxsa0H0H50 mmwupconoouwa .mmmpwxo o oaousoouhu mmoEom .uamwvmnw amouosm m %n moaamamwuo porno are new cowumwnmwuucmo Hmfiuaapommwt he poumaomw muoB mummamouoaso maos3la£H meowuumnm oaaaemwuo cw muw>wuu< ammummmcmuuoawe< oumuummm< .H magma 67 Thus aSpartate:a-ketoglutarate aminotransferase is located in all three subcellular organelles while only the peroxisomes contain serine:g1yoxylate, alanine:glyoxy1ate and g1utamate:glyoxylate aminotransferase activities. It is not known whether all of the in_yixg activity of these aminotransferases is located in the organelles or if the cytosol contains some soluble activities. CHARACTERISTICS OF THE PEROXISOMAL GLYOXYLATE AMINOTRANSFERASES Buffer and pH Effects Both serine:g1yoxylate and alanine:glyoxy1ate amino- transferases had maximum activities around pH 7 under the assay conditions described (Table 2). The observed rate of the serine:g1yoxylate aminotransferase activity was nearly the same at pH 7 in HEPES, cacodylate or Bicine buffer. The activity was also nearly maximum in Bicine at pH 8 but declined at pH 6 in cacodylate. Phosphate buffer severely inhibited the serine:g1yoxylate amino- transferase activity. However, alanine:glyoxy1ate amino- transferase activity was not affected by phosphate buffer and this can be a test to distinguish between the two enzymes. The alanine:glyoxy1ate aminotransferase had nearly the same activity in HEPES, cacadylate, phosphate and Bicine buffers at pH 7. The rate was about 50% less 68 Table 2. Buffer and pH Effects on Serine:Glyoxylate and Alanine:Glyoxy1ate Aminotransferase Activity Peroxisomes were isolated using the Small zonal rotor and broken by dilution with buffer. A 30-60% (NH4) SO precipitate of the broken peroxisomes was resuSpen ed and passed through a Sephadex G-25 column. This protein fraction.was used for the assays. Activity units are nmoles x min" x ml' . Serine:Glyoxylate Alanine:Glyoxy1ate Aminotransferase Aminotransferase Buffer . . . . . . Act1v1ty Relative Act1v1ty Relative Rate Rate HEPES pH 7 1290 100 645 100 Cacodylate pH 6 645 50 354 55 pH 7 1130 88 677 105 Phosphate pH 6 129 10 516 80 pH 7 258 20 580 90 Bicine pH 7 1220 95 774 120 pH 8 1130 88 354 55 69 at pH 6 in cacadylate and at pH 8 in Bicine. The glu- tamatezglyoxylate aminotransferase activity in the spinach leaf peroxisomes had the same activity in cacadylate and phOSphate buffers at pH 7 (data not shown). Kisaki and Talbert (85) reported that the g1utamate:glyoxylate amino- transferase in spinach leaf peroxisomes had a pH Optimum of 7 and this is the same pH optimum reported for both the rat liver and human liver enzymes (114, 150). King and Waygood (83), using phoSphate buffer, reported a pH optimum of 8.2 for serine:g1yoxylate aminotransferase from wheat leaves. Of the three glyoxylate aminotransferases assayed in spinach peroxisomes, only serine:g1yoxylate amino- transferase was affected by phosPhate. This difference in the effect of phOSphate suggests the serine:g1yoxylate aminotransferase reaction may be catalyzed by a different protein than the g1utamate:glyoxylate and alanine:glyoxy- 1ate aminotransferase reactions. D-Serine Inhibition D-Serine at 40 mM was found to inhibit the serine: glyoxylate aminotransferase 85%, whereas the alanine: glyoxylate aminotransferase was inhibited only 35% (Table 3). However, D-alanine had little affect on the 70 Table 3. D-Serine Inhibition of Glyoxylate Aminotransferases Spinach leaf peroxisomes were isolated on a sucrose gradient and assayed by the Spectrophotometric procedure. Specific activity is in nmoles xmin'1 x mg protein" . The D-amino acids were at a final concen- tration of 40 mM. ' Addition to Serine:Glyoxylate Alanine:Glyoxy1ate Assay Aminotransferase Aminotransferase Specific Percent Specific Percent Activity Inhibition Activity Inhibition None 630 - 510 - D-Serine 97 85 330 35 D-Alanine 530 16 390 23 71 alanine:glyoxy1ate or on the serine:g1yoxylate amino- transferase activities. The D-serine inhibition of serine:g1yoxylate aminotransferase did not appear to be purely competitive or noncompetitive (Figure 5). The Km of L-serine appeared to be affected by the D-serine. The l/velocity versus l/[L-serine] plots at various con- centrations of D-serine were nonlinear (Figure 5). These kinetics are characteristic of allosteric enzymes. How- ever, since the aminotransferase reaction probably occurs by the Ping Pong Bi Bi mechanism, the D-Serine may interact differently with the various enzyme-coenzyme- substrate complexes which occur in the aminotransferase reaction and thus diSplay the complex kinetics. The enzyme may not have an allosteric Site. The D-serine inhibition was probably not caused by some contaminant in the D-Serine. Sigma quality control reports that lot 108B-l860 of D-serine which was used in the assay had only one ninhydrin positive spot or one iodine positive Spot after thin layer chromatography in three different solvent systems. The experimentally determined nitrogen content was 13.54% while 13.33% is the theoretical value. Heavy metals were less than 10 ppm. To the authors knowledge this is the first report of D-amino acid inhibi- tion of an L-amino acid aminotransferase. ASpartate 72 Figure 5 D-Serine Inhibition of Serine:Glyoxylate Aminotransferase Spinach peroxisomes were isolated on a sucrose grad- ient. The spectrophotometric procedure was used and the glyoxylate concentration was at 1 mM. The D-serine was added to the assay mix before the enzyme. Incubation of the enzyme with D-serine for two minutes gave the same results as with no incubation period. The concentration of D-serine was either 0,4,8 or 20 mM. A. Velocity versus Concentration of L-Serine B. Lineweaver-Burk Plot at Various Concentrations of D-Serine 73 D-Serine 0 mM v (nmoles x min") L-Serine (mM) 20mM 9 "6" f q 8 01M ’x 4 mM '2 : ,1 , ' I I 0.4 D-Serine 0 mM \ If5‘ 2'.o I/L- Serin'e (mM" I 1 0.5 I.O l 2.5 74 aminotransferase from spinach peroxisomes was not effected by D-aspartate (data not shown) and as shown above, D- alanine had little effect on the alanine:glyoxy1ate amino- transferase. Kisaki and Talbert (85) reported that in spinach peroxisomes D-glutamate was more active than L- glutamate in the transamination of glyoxylate. SEPARATION OF GLYOXYLATE AMINOTRANSFERASE ACTIVITIES USING ION EXCHANGE CHROMATOGRAPHY The serine:g1yoxylate aminotransferase was in- hibited by D-serine and phosphate whereas the alanine: glyoxylate and g1utamate:glyoxylate aminotransferase activities were not inhibited and these differences suggested that separate proteins may catalyze these reactions. Broken peroxisomes from spinach leaf tissue were placed on a TEAE-cellulose column and eluted with a linear gradient of 0 to 0.3 M KC1. The profile of the elution is shown in Figure 6. Serine:g1yoxylate amino- transferase eluted at a KCl concentration of approximately 0.2 M. Only one peak of activity was observed but the recovery was only 50%. This peak had coincident activity with either glyoxylate or pyruvate as the amino acceptor and the data shown is that for serine:pyruvate aminotrans- ferase activity. It will be shown in subsequent results that serine:g1yoxylate and serine:pyruvate aminotransferase 75 .Hanwfin maEHu OH maaumawxouaam an paso3 muH>Huom ammummmcmuuoawam aumamxozawuacwuam asp new hua>auom ammuammaMHuoafiew aum>bnhd "acwuam How one ao>Hw maDHm> 65H .Hooououa ammuammamuuoaaam auw>an%duaawuom use he pawmmmm mm3 ammuammcmuuocwsw aumakxohaw "aafluam mnfi .ammuoseau aum>Dnhmhonpxs .«u.u.n.i< use mandamumo . a. . . . . < mamwcawoupmsap madame m Darrin mammuammamuuoawam aumahxokawuaafluam .OIIIIIC mammHaMmeHuoaflEm oumuumamm .Qlllllo .uaowpmuw HUM_Z m.o on o m nuH3Vpou5Ha anaB mashuca 05H .aafiaoo amOHDHHmUumoEaH auaB mocmuneaa can use xoosm owuaEmo %n saxoun aua3 uaaflomuw amouodm a so paumHOmH .manmeouam msamnwoumEouso quHoo amoHsHHaoum¢MH an maahnam HmEOmeouam mo aowumummam e was»; 76 (Ml-U X._U!w x sa|oww) esalomg V----v 0, N ”Z N .1 O I l I “Jul at '_Ul|u x seioum) osaueboapxuea slalom H N m a I I I (i—I‘” x '_l.llu.l x SOlOlLIII) ssaisnpea ewandkxoapKH v-._' 8 8 8 I I I °A " O... ‘ Q‘II ~~fl\ - O-o-«.~.~. 563:... ___ < A . ----q I O a ° I . Q g l l l I l l l 8 8 8 8 8 8 8 N N (um xl_ugu.l x smowu) :iaaaimioiex-v-zeiauadsv O-——o 1 1 1 1 l 1 o o o o o 0 <3 9 6 ca v N ('JUJ x l_uyui x SOIOUJU) eiaiflxonig :suues H I40- 0.3— 0.2- .1. o‘ (w) it»: —- Fraction Number 77 activities are very likely catalyzed by the same enzyme. The column fractions were also assayed for glutamate: glyoxylate aminotransferase and alanine:glyoxylate amino- transferase but no activity was detected. Washing the column with 1.0 M KCl still did not elute the glutamate: glyoxylate aminotransferase activity. However, the glu- tamate:g1yoxy1ate aminotransferase was active in an aliquot of the original solution which had been applied to the column. These results suggest that peroxisomes contain a serine:g1yoxylate aminotransferase which does not catalyze the g1utamate:glyoxylate or alanine:glyoxylate aminotrans- ferase reactions. Brock £3 £1. (23) also reported that upon partial purification of serine:g1yoxylate aminotrans- ferase from oat leaves, the g1utamate:glyoxylate and alanine:glyoxylate aminotransferase activities were sep- arated from the serine:g1yoxylate aminotransferase. The spinach peroxisomal catalase peak coincided exactly with the serine:g1yoxylate aminotransferase (Figure 6). Other peroxisomal proteins such as malate dehydrogenase and hydroxypyruvate reductase were partially separated from the serine:g1yoxylate aminotransferase. The aspartate aminotransferase data will be discussed later. However, it should be noted that the serine:g1yoxylate amino- transferase peak does not correspond to any of the 78 aspartate aminotransferase peaks of activity. SEPARATION OF THE PEROXISOMAL GLYOXYLATE AMINOTRANSFERASES USING ISOELECTRIC FOCUSING An isoelectric focusing column with a total volume of 5 ml was used to separate the aminotransferases. The distribution of the enzymes in the pH 3-10 gradient showed only one peak of serine:g1yoxylate aminotransferase (Figure 7). The recovery of the serine:g1yoxylate amino- transferase activity from the column was 92%. One peak of g1utamate:glyoxylate aminotransferase and two peaks of alanine:glyoxy1ate aminotransferase were also observed on the column. The serine:g1yoxylate aminotransferase was separated from the g1utamate:glyoxylate aminotransferase but both enzyme peaks had an alanine:glyoxylate amino- transferase activity. The pI values of the serine: glyoxylate aminotransferase and the g1utamate:glyoxylate aminotransferase were 6.7 and 5.8 reSpectively. In the peak of the serine:g1yoxylate aminotrans- ferase, the serine:g1yoxylate and alanine:glyoxylate amino- transferase activities were not additive and likewise in the g1utamate:glyoxylate aminotransferase peak the gluta- matezglyoxylate and alanine:glyoxylate activities were not additive (data not shown). This indicates that the alanine:glyoxylate aminotransferase activity is due to 79 Figure 7 pH 3-10 Isoelectric Focusing of Peroxisomal Aminotransferases Peroxisomes were isolated on a sucrose gradient, broken by osmotic shock, centrifuged to remove the traces of chlorophyll and placed on a pH 3-10 ampholyte isoelectric focusing column. The column was run at 500 volts for 9 hr. Enzyme activities are in arbitrary units per m1. Values in parenthesis for aspartate:a-ketoglutarate aminotransferase are relative values of the peak fractions. Specific activ- ities of the peroxisomal enzymes are given in Table 9. Arbitrary Units Serine: '\'\p'l: _ Glyoxylate __ Alanine: Glyoxylate _ Glutamate: Glyoxylate " Serine: _. Pyruvate (IOOIIL Aspaerttgte:fi_ 1%}45) (28) (29) _ glutarate L Fraction Number 81 nonspecificity of the other two aminotransferases. The spinach peroxisomal serine:g1yoxylate and g1utamate:glyoxylate aminotransferase reactions were catalyzed by two separate proteins. Separate enzymes for these two aminotransferase activities appear to also be present in rat liver, since partially purified glutamate: glyoxylate aminotransferase did not have any activity with serine (114). However, the rat liver enzyme did react with alanine just as the spinach peroxisomal enzyme appears to do. Human liver g1utamate:glyoxylate amino- transferase was also not active with serine but again did react with alanine (150). Human liver also contains an alanine:glyoxylate aminotransferase which does not react with glutamate but does have activity with serine (151). Since an the pH 3-10 isoelectric focusing column the serine:g1yoxylate and g1utamate:glyoxylate aminotrans- ferase peaks were relatively close, an isoelectric fo- cusing column of pH 5-8 was used to obtain a better reso- lution of these two enzymes. The activity in the column fractions was low but a general separation of the two aminotransferases was achieved (Figure 8). The alanine: glyoxylate aminotransferase activity was smeared through- out the column with a peak corresponding with the g1uta- mate:glyoxylate aminotransferase and a slight peak 82 .ammuommamuuocwfim aumaxonkuaCHumm . D....;u use ammuomwcmuquHEm mumaxx0%aw ”mumEmOSHw .4 uuuuu 4 mammuammcmuuoawew aumahxokawuoaacmam .OIlllJo .HL NN Hmcowuwppm am you auasu eachuaHmE one muao> com on H: m mo powuad m um>o mammouocH mmB SEDHoo win we m£u no ammuao> can umnu uaaoxm N muswwm CH ca>ww mm dawn atu mmB anatmaoua ash mommuammamuuoawa< HmEomeouom mo wcflmsuom owuuooaoomH mum ma w madman 83 O. ..mmEzz cozooi . Ono-‘jl lb... mania: ... L........ 85:24 . ..nuloIo as Q ~ ~ \nlwhmwn . _..\.I. 1223830 .mctom .. i s .. \aieieiebie a a «. .\.\. \IW—u o\.\ 32.3830 w In “30:63.0 1 siiun KJoJuqJV 84 corresponding with the serine:g1yoxylate aminotransferase. The enzymes were probably not completely focused since the peaks were very broad and the p1 of g1utamate:glyoxylate aminotransferase was 4.8 in this column as compared to 5.8 in the pH 3-10 column. Studies were conducted on the two peaks of alanine: glyoxylate aminotransferase to determine if they had the same characteristics (Table 4). The alanine:glyoxylate aminotransferase activity in peak 2 (the serine:g1yoxylate aminotransferase peak) was most active at pH 8 in borate buffer and it was least active in phosphate buffer at pH 7. The alanine:glyoxylate aminotransferase activity in peak 1 had about the same activity under all assay con- ditions. These different ratios of activity are consis- tent with the idea that the two peaks of alanine:glyoxy- late aminotransferase are different proteins. The question remains whether the alanine:glyoxy1ate aminotransferase activities are separate enzymes or whether these reactions are nonspecific activities of the g1utamate:glyoxylate and serine:g1yoxylate aminotrans- ferases. Although the results presented here are con- sistant with the alanine:glyoxy1ate aminotransferase activity being due to the other two aminotransferases, no definite conclusion can be drawn until further experiments 85 Table 4. Characteristics of Alanine:Glyoxy1ate Amino- transferase Peaks From a pH 5-8 Isoelectric Focusing Column Fractions were assayed by the Spectrophotometric procedure. Peak 1 contained g1utamate:glyoxylate amino- transferase and peak 2 contained serine:g1yoxylate aminotransferase (Figure 8). Buffer Alanine:Glyoxy1ate Aminotransferase Peak 1 Peak 2 - -l umoles x min 1 x ml Cacodylate pH 7 1.15 0.44 Borate pH 8 1.11 1.52 PhoSphate pH 7 1.25 0.30 86 are conducted. Preferably, future experiments would involve purification of the aminotransferases to homo- geneity and then definite characteristics could be established. ADDITIONAL STUDIES OF SERINEzGLYOXYLATE AMINOTRANSFERASE The serine:g1yoxylate aminotransferase also catalyzed a serine:pyruvate aminotransferase activity. Yamazaki and Talbert (174) had reported earlier that a serine:pyruvate aminotransferase was located in Spinach leaf peroxisomes. Osmond and Harris (122) found that Atriplex and sorghum contained a serine:g1yoxylate amino- transferase activity. The extracts of these plants also catalyzed a serine:pyruvate aminotransferase reaction but at a slower rate than the serine:g1yoxylate aminotrans- ferase reaction. Sallach and coworkers (30, 170) have reported that the green leaves of several plants contain alanine:hydroxypyruvate aminotransferase activity. In spinach peroxisomes, the enzyme was more active with glyoxylate than pyruvate and thus the enzyme has been designated as serine:g1yoxylate aminotransferase. At a fixed serine concentration of 20 mM, the Km for glyoxylate was 0.15 mM while for pyruvate it was 2.82 mM (Figures 9 and 10). These Km values were determined using the 87 Figure 9 Reaction Kinetics for Serine:Glyoxylate Aminotransferase With Varying Glyoxylate Concentrations Isolated spinach leaf peroxisomes and the spectra- photometric assay procedure were used. The serine concen- tration was 20 mM for all assays. .-| 'V' (umoles x min I s/v (mM/pmoles xminql 2.0 I.6 |.2 0.8 0.4 1 1 l l 0.2 0.4 0.6 0.8 Glyoxylate (mM) |.O Km = O.I5 mM I 1 1 l 0.2 0.4 0.6 0.8 Glyoxylate (mM) 1 LG 89 Figure 10 Reaction Kinetics for SerinezPyruvate Aminotransferase Activity With Varying Pyruvate Concentration Isolated spinach leaf peroxisomes were assayed by the spectrophotometric procedure. The serine concen- tration was held constant at 20 mM. v (nmoles x min') s/v (mM/nmoles x min 'I 90 1 1 i 1 1 | 2 3 4 5 Pyruvate (mMI O.IO ’ oasi- 0.06- Km = 2.82 mM 0.04 . J 1 L J 1 l 2 3 4 5 Pyruvate (mM) 91 spectrophotometric assay. Even though the linking enzyme in the assay (hydroxypyruvate reductase) acts upon the the glyoxylate, the glyoxylate concentration did not change significantly. In fact at the low glyoxylate concentra- tions, no endogenous rate was observed because of the high Km that hydroxypyruvate reductase has for glyoxylate (89). The Km for serine was 2.72 mM at a fixed concentration of 1 mM glyoxylate (Figure 11). It appears to be a general characteristic of aminotransferases that the Km for the amino acid is higher than that of the a-keto acid. Several experiments were performed to try and Show that the serine:g1yoxylate and serine:pyruvate aminotrans- ferase activities were due to one enzyme. The two activ- ities peaked together on TEAE-cellulose columns and on a pH 3-10 isoelectric focusing column (Figure 7). The recoveries from the isoelectric focusing column were 92% for the serine:g1yoxylate aminotransferase and 103% for the serine:pyruvate aminotransferase activity. Bath enzyme activities were inhibited by phosphate and D-serine and both activities were equally sensitive to heat denatur- ation of the protein (Table 5). The serine:pyruvate amino- transferase activity was not inhibited by the D-serine to the same extent as was the serine:g1yoxylate aminotrans- ferase activity. This may indicate that D-Serine also 92 Figure 11 Reaction Kinetics for Serine:Glyoxylate Aminotransferase With Varying L-Serine Concentrations Spinach leaf peroxisomes were isolated on a sucrose gradient and assayed by the spectrophotometric procedure. The glyoxylate concentration was 1 mM for all assays. 93 8 I4 IO L—Serine 6 A — i _ . 4. 2rd 2 I. O O O O. 0 ...:E x $353 .>. Km = 2.72 mM I4 l8 IO L- Serine (mM) Imfl law 5 5 3 2 l $.56 x $353.25 >\m 94 Table 5. Comparison of Serine:Glyoxylate and Serine: Pyruvate Aminotransferase Activities Peroxisomes which were isolated by isopycnic centri- fugation in different experiments were used for the various treatments. All assays were by the spectrophotometric procedure. The D- and L-serine were each at 20 mM. Treatment Serine:Glyoxylate Serine:Pyruvate Ratio Aminotransferase Aminotransferase nmoles xmin"1 xml'1 Heated at 50 C 0 min 955 90 10.6 5 439 46 9.5 15 310 31 10.0 25 284 28 10.1 35 232 26 9.0 Buffer HEPES 419 53 8.0 Phosphate 97 0 - Serine L-Serine 232 22 10.6 D-Serine 0 0 - D- + L-Serine 64 17 3.8 95 affects the binding of the d-keto acid to the serine: glyoxylate aminotransferase. As shown previously (Figure 5), the D-serine appeared to affect the binding of the L-serine. While these experiments do not prove that only one enzyme is catalyzing both the serine: pyruvate and serine:g1yoxylate aminotransferase reactions, they do strongly suggest that only one enzyme is involved. The serine:g1yoxylate aminotransferase was checked for inhibition by other compounds which may be involved in the possible regulation of the enzyme (Table 6). ATP, ADP, AMP, O-phospho-L-serine, 3-P-glycerate and D-glycerate did not Significantly inhibit the enzyme. The phosphate inhibition is probably not of physiological importance since at 10 mM the inhibition was only 34%. Because a portion of the glycolate pathway involves nitrogen metab- olism the effect of nitrate, nitrite and ammonium ions was checked. Ammonium sulfate at 12 mM was inhibitory but as with phosphate it is probably not of physiological importance. In fact it may have been the sulfate that was causing the inhibition. REVERSIBILITY OF GLYOXYLATE AMINOTRANSFERASES Some reports have shown an aminotransferase reaction using glycine as the amino donor (27, 35, 171). 96 Table 6. Effect of Various Compounds on Peroxisomal 'SerinezGlyoxylate Aminotransferase Isolated peroxisomes were assayed by the spectra- photometric procedure. Unless state otherwise, all compounds were at a final concentration of 10 mM. Compound Relative Rate ATP ADP AMP O-Phospho-L—Serine 3-P-Glycerate D-Glycerate NaNO 3 KNOZ (NH4)ZSO4 12 mM K3PO4 K3PO4 70 mM D-Serine 20 mM 100 71 86 83 76 79 62 77 77 34 66 23 31 . . v . .u _ . p . 1‘ .v _ o . _ . . o .— _ . . .. . . .D _- C » . . v i A . xi...“ v—L—‘A & 97 The authors who observed this activity have generally used crude homogenates and reaction times of 1 hr. There are some reports that glycine:a-keto acid aminotransferase activity could not be detected using partially purified enzymes (23, 150, 151). In spinach leaf peroxisomes, Kisaki and Talbert (85) did not detect the reverse reaction of g1utamate:glyoxylate aminotransferase. In the present studies with isolated peroxisomes, the reverse reactions for g1utamate:glyoxylate, serine:g1yoxylate and alanine: glyoxylate aminotransferases could not be detected (Table 7 - Lower part). The reason for the apparent lack of the reverse reaction of amino acid:glyoxylate aminotrans- ferase may be because the equilibrium lies so far towards glycine formation that the reverse reaction can not be detected. If the glyoxylate aminotransferases function by a Ping Pong Bi Bi mechanism, as reports suggest (23), then one would expect to observe an isotope exchange between glyoxylate and glycine. Using whole peroxisomes, some glyoxylate-glycine exchange was observed (Table 7 - Upper part) but this exchange was very small. Kisaki and Talbert (85) also reported a little glyoxylate-glycine exchange. Whether all of the peroxisomal glyoxylate aminotransferases catalyzed this exchange is not known. Thompson and Richardson (151) reported that purified 98 Table 7. Amino Acid:a-Keto Acid Exchange Reactions and Glycine:a-Keto Acid Aminotransferase in Spinach Peroxisomes Peroxisomes were isolated by isopynic centrifuga- tion. For the exchange reactions, radioactive amino acid and the corresponding unlabeled d-keto acid (alanine- C + pyruvate) were used in the radiochemical assay procedure. The formation of radioactive a-keto acid was measured. In the g1ycine:a-keto acid aminotransferase rfiaction, the transamination reaction between glycine- 1 C and a-keto acid was measured by the radiochemical procedure. Exchange Reactions 14C Amino Acid a-Keto Acid 14C a-Keta Acid nmoles x 30 min-1 Alanine Pyruvate 1300 Glutamate a-Ketoglutarate 2000 ASpartate Oxaloacetate 81 Serine Hydroxypyruvate 1510 Glycine Glyoxylate 25 Glycineza-Keto Acid Aminotransferase . l4 a-Keto Ac1d C Glyoxylate nmoles x 30 min-1 Hydroxypyruvate 0 Pyruvate 0 Oxaloacetate 0 a-Ketoglutarate 0 99 alanine:glyoxylate aminotransferase from human liver did not catalyze a glyoxylate-glycine exchange. SPINACH LEAF ASPARTATE:a-KETOGLUTARATE AMINOTRANSFERASE Yamazaki and Talbert (174) reported this enzyme to be present in chloroplasts, mitochondria and peroxi- somes and that the peroxisomes contained 3 isoenzymes. Additional Studies on the electrophoretic properties of the aSpartate aminotransferases in spinach leaf fractions is reported here. Both chloroplasts and mitochondria had a single protein band that had the same electrophoretic mobility (Figure 12). Mixtures of chloroplastic and mitochondrial fractions likewise, resulted in only one band of enzyme activity. The whole chloroplasts (i.e. the pellet of an initial low speed centrifugation of spinach homogenate) had a band that moved with the dye front and stained in the assay in the absence of aspartate. If the chloroplasts were resuSpended and centrifuged through a 30% (w/w) sucrose layer, the nonenzymatic band was not detected. The nature of this nonenzymatic band is unknown, but the report by Yamazaki and Talbert (174) that there might be another chloroplastic isoenzyme of aspartate aminotransferase is in error. Crude chloro- plast fractions sometimes contained one of the peroxisomal liliillill I ‘ 100 ll‘ ' | l l I II III ‘I. I): .aSmmHu mama use mo cowumNHCaonos wcHHDU Hamman mmaa mean: %a paumdapd mmB mausoonon paumhucaoaoo .amouonm Nom mo Mahma m :wnousu cowuomuw umdeouoanu aHon3 notcmamsmah can wcwwamwuuaao he pmaamuao uaaama asp mm3 coauomum ummamouoano aHo£3 poemmB anH .w one an aumcaonon somawmm wafiwsmwhucmo Baum uaHHaa onu mmB mummadouoHno aHo£3 pmHHaan aofiuomum aSH .ucawpmuw amouoam m :o pmumaomw auaz mwupaosoouwe tam manmeond any mama auHEmfikuomNHom co maflufl>fluo< ammuommcmuuocwe< aumuHMQm< NH muswua 101 auqaawoaam auauu vouuuuaaocoo moaanwxouam aunaamouoanu + QOEOQHNOHOQ «uneconooflz ewuvconoouqz + munaamouoano ILV_ a.“ uvcoauouaz + noeaaaxauom use: unoum «by luv uauéucoaoa numuaaauaanu 325 e383 euoeanouaano macs: 102 isoenzymes, but this was not consistently seen. Two slower moving bands of aspartate:a-ketoglu- tarate aminotransferase on the polyacrylamide gels were distinctly peroxisomal (Figure 12). A third fast moving enzyme band in the peroxisomes had the same mobility as the enzyme in the mitochondria and chloroplasts. Mix- tures of the samples showed that the fast moving band in all three organelles was the same enzyme. The peroxi- somal bands were not observed in dilute leaf homogenates, but in concentrated homogenates the two peroxisomal bands were detectable. It should be noted that in whole leaf homogenates of spinach leaves there were only the three isoenzyme bands and every band corresponded to one of the particulate isoenzymes. In other words, there was no specific soluble isoenzyme for the cytosol. It has been reported many times that mammalian tissues contain a soluble and a mitochondrial form of aspartate amino- transferase (136). The Spinach leaf isoenzymes of aspartate amino- transferase were also examined by starch gel electro- phoresis (Figure 13). A single isoenzyme in the chloro- plastic and mitochondrial fractions again had identical mobility. The peroxisomal activity was in two bands: a dark band of low mobility and a trace of activity located 103 at the same R.F as the mitochondrial chloroplastic iso- enzyme (Figure 13). This data on the peroxisomal forms of the enzyme was thus considerably different from the results with polyacrylamide gel separation but support the idea that the peroxisomes do not contain the same enzyme as the mitochondria and chloroplast. To determine if any of the three peroxisomal aspartate aminotransferase isoenzyme bands were due to some nonspecificity of the glyoxylate aminotransferases, the proteins in isolated peroxisomes were separated on a TEAE-cellulose column. Upon elution with a linear KCl gradient, three peaks of aspartate aminotransferase were observed (Figure 6). None of the three peaks coincided with the serine:g1yoxylate aminotransferase. As mentioned previously, the g1utamate:glyoxylate and alanine:glyoxy- 1ate aminotransferase activities could not be detected on TEAE-cellulose columns. In a fraction where the aspartate aminotransferase and serine:g1yoxylate aminotransferase overlapped, assays containing all four substrates were the sum of the individual aminotransferase activities (data not shown). These results strongly suggest the aspartate aminotransferase isoenzymes in the peroxisomes are not the result of nonspecificity of the glyoxylate aminotransferases. 104 Figure 13 Starch Gel Electrophoresis of ASpartate Aminotransferase Activities The whole chloroplasts were prepared by differential centrifugation while the peroxisome and mitochondria were isolated by isopynic centrifugation in sucrose gradients. Both the gel and the electrode tanks contained 5 mM Tris- glycine at pH 8.3. Electrophoresis was at 400 volts for 10 hr. 105 (_) origin (+I whale chloroplasts mitochondria g peroxi somes a E 106 Each of the three peaks of aspartate aminotrans- ferase from a TEAE-cellulose column was subjected to polyacrylamide electrophoresis. Each peak contained a different isoenzyme (Figure 14). The peak, which eluted first off of the TEAE-cellulose column, was called peak 1 and contained the Slowest moving electrophoretic band, labelled band 1. Peak 2 contained the central electro- phoretic band, and peak 3 the fastest moving band. There was some cross contamination, as could be expected by the TEAE-cellulose elution pattern. The electrophoretic band 1 from the TEAE-cellulose columns was not one distinct band, but it appeared to be 2 or 3 closely spaced bands. This phenomena was observed only on samples from the TEAE columns. Removal of the KCl in peak 1 by dialysis did not change the electrophoretic pattern. In some earlier experiemnts, the aSpartate amino- transferase activity in the mitochondrial fraction of a sucrose gradient appeared to be inhibited by phOSphate buffer more than the activity in the peroxisomal fraction. Therefore assays on the three peaks of aspartate amino- transferase from the TEAE-cellulose column were run in different buffers (Table 8). The activity in peak 3 was inhibited by phosphate buffer Similarly to the mitochon- drial form of the enzyme. AS mentioned previously, peak 3 107 Figure 14 Separation of the Aspartate Aminotransferase Isoenzymes by Ian Exchange Chromatography and Their Electrophoretic Patterns The peroxisomal aspartate aminotransferase activity was eluted from a TEAE-cellulose column with a linear KCl gradient as in Figure 6. An aliquot from each peak was then run on polyacrylamide gel electrophoresis and stained for aspartate aminotransferase activity. nmoles x min"I x ml"I 108 IOO— Peak I 2 3 A ear " . A ‘1’ 0.3 20i— Fraction Number Polyacrylamide Electrophoresis KCI (MI 109 Table 8. Characteristics of Aspartate Aminotransferase Peaks From a TEAE-Cellulose Column The fractions from a TEAE-cellulose column (Figure 14) were assayed by the Spectrophotometric procedure. Peak Buffer l 2 3 Relative Rates MES pH 7.4 100 100 100 MES pH 6.7 80 84 41 PhoSphate pH 7.1 86 97 68 110 contained the fastest moving isoenzyme, which on the poly- acrylamide gels had the same mobility as the form of the enzyme in the mitochondria. This isoenzyme was also more inhibited by MES buffer slightly b610W'the pH optimum than the other two distinctly peroxisomal isoenzymes. The separation of the 3 peroxisomal aSpartate aminotransferase activities was also achieved on a pH 3-10 isoelectric focusing column (Figure 7). It should be noted that none of the aspartate:a-ketoglutarate aminotransferase peaks correspond exactly to any of the glyoxylate aminotransferases. These data confirm the results of the TEAE-cellulose column, that the aspartate aminotransferase peaks were not caused by any of the three glyoxylate aminotransferases. Boiled controls of the aspartate aminotransferase peaks had no activity and the recovery from the column was 100%. Yamazaki and Talbert (174) reported finding only one peak of aspartate amino- transferase on an isoelectric focusing column. No explanation for this difference in the results is known. It is not clear whether peroxisomes contain two isoenzymes or three isoenzymes of aspartate aminotrans- ferase. The fastest moving isoenzyme band in the peroxisomes has the same characteristics as the mito- chondria-chloroplast isoenzyme. However, this does not 111 mean the fast moving band is not a peroxisomal enzyme. Based upon the intensity of the stain in the polyacryl- amide gels, the fastest moving band in the peroxisomes seemed to comprise 40-50% of the total activity. It is estimated that, of the total peroxisomal aspartate amino- transferase activity, only 15% was due to contamination by the other organelles (Table 1). Thus the fastest moving band may also be a peroxisomal enzyme. In the starch gel electrophoresis, the aspartate aminotrans- ferase activity in the peroxisomes did not form three bands, but was present as one major band of low mobility and a little activity corresponding to the mitochondria- chloroplast isoenzyme. This data suggests that most of the aspartate aminotransferase activity in the peroxi- somes was not separated by starch gel electrophoresis, but that a small amount of contamination by other organ- elles was present and moved separately. SUMMARY OF SPINACH PEROXISOMAL AMINOTRANSFERASES All aminotransferase activities with glyoxylate were located exclusively in the peroxisomes, where the glyoxylate is formed. These reactions were irreversible under the experimental conditions employed. Spinach peroxisomes contain a serine:g1yoxylate aminotransferase 112 which appears to have some activity as a serine:pyruvate aminotransferase and possibly also some alanine:glyoxy1ate aminotransferase activity. A g1utamate:glyoxylate aminotransferase is a separate peroxisomal protein and this enzyme appears to have alanine:glyoxylate amino- transferase activity. The specific activities of the peroxisomal aminotransferases Show g1utamate:glyoxylate to be the most active (Table 9). The descending order of the remaining Specific activities are serine:g1yoxy- late aminotransferase, alanine:glyoxylate aminotrans- ferase, aspartate:a-ketoglutarate aminotransferase and the activity for the serine:pyruvate aminotransferase reaction. Kisaki and Talbert (85) reported that in Spinach leaf peroxisomes the most active aminotransferase was g1utamate:glyoxylate aminotransferase, followed by alanine:glyoxylate and serine:g1yoxylate aminotransferase activities. Probably the reason these authors observed less serine:g1yoxylate aminotransferase activity is that they used phOSphate buffer and DL-serine in their assays and as it has been shown, both phOSphate and D-serine inhibit serine:g1yoxylate aminotransferase. Spinach peroxisomes also contain an aSpartate: a-ketoglutarate aminotransferase which forms 3 isoenzyme 113 Table 9. Specific Activities of Aminotransferase Activities in Isolated Spinach Leaf Peroxisomes . -1 -1 Enzyme umoles x min xnmg protein Glutamate:Glyoxylate 2.40 Serine:Glyoxylate 1.54 Alanine:Glyoxy1ate 0.87 Aspartate:d-Ketoglutarate 0.15 Serine:Pyruvate 0.03 114 bands on polyacrylamide gels. This result is in agreement with the work of Yamazaki and Talbert (174). PRELIMINARY STUDIES ON THE SUBCELLULAR.LOCATION OF AMINOTRANSFERASES IN MAMMALIAN TISSUES Rat Liver Mammalian tissues are reported to contain a mito- chondrial and a soluble form of aspartate aminotrans- ferase. de Duve and coworkers (8) and Miller (113) looked for aspartate aminotransferase in rat liver peroxi- somes but concluded it was not a peroxisomal enzyme. Like- wise, de Duve (40) reported that rat liver peroxismmes did not contain g1utamate:glyoxylate aminotransferase. However, Vandor and Talbert (162) did observe glutamate: glyoxylate aminotransferase activity in the rat liver peroxisomes. The subcellular location of alanine:glyoxy- 1ate aminotransferase in rat liver was recently reported in an abstract by Snell 25.31. (147), who concluded that a large portion of this aminotransferase activity was located in the mitochondria. In a group effort in the laboratory, rat liver peroxisomes were isolated in sucrose gradients using a procedure similar to that described by de Duve 25.31. (43). I assayed fractions for aminotransferase activities (Table 10). All glyoxylate aminotransferase assays were carried 115 .AH manna oomv cowuomum Hmwupaosoouwa asp mo coaumafiamucoo Hmaomwxouam ou pauanwuuum an emu nofin3 cowuomnm Hmanuaonoouwa use aw muw>wuom ammuammamuuoawam can mo ammucaoham may one masam>n .AH aHAaH mamv aowuomum HQEOmeoHam can we dowumcaemuaoo Hmwupaosoouwa ou pausnfluuum an coo sown3 cowuomum HmEomeouaa can a“ muw>auom ammuommamuuoawem oau mo owmuaoouaa mnu one maaam>a m.o ONH oqw.m oom.oa can ouwumuaawouaxud "aumuummm< on m.o me mm mm oumahxohaw "magnum - - mmH o a mumaaxosuo "oumemuaau OH 0 SNH mom HNH mumaaxoaao ”mafiama< N N HIHB x Huawa x mmaoac aowuomum meowuomum Hm.uuaosoouwz HmEomeonom SH muw>wuo< aw >ufl>wuo¢ ucmumchamsm mwhpaosoouflz mmaomwxouam ammummmamuuoafla< anaemwxopom Hmwupaosoouwz .ounpaooum uwuuofiouosm nouuoamm on“ Na pakmmmm mm3 nowne ammuammamuuoaHEm aumnmuSHwouaxud"oumuumamm Ham umooxo endpoooua HmuHEmnoowpmn can an paammmm anaB mawuw>auom ammummmcwuuosHem HH< .Houou Hmaon a fig unawpmuw amouusm a no paumaomw aua3 moaaoamwno Hoasaaoonam Ha>flq Dam Baum moHoHuHmm waoe< onuH>Huo< ommHmMmamuuoaH8< mo aowusnwuumwn .OH oHQMH 116 out using the radiochemical procedure because the mammalian tissues had a high endogenous rate in the spectrophoto- metric procedure. Aspartate:a-ketoglutarate aminotransferase was located in the mitochondria and all of the activity that was in the peroxisomal fraction could be attributed to mitochondrial contamination (Table 10). The rat liver per- oxisomes differ in this reSpect from the peroxisomes of Tetrahymena pyriformis and Spinach leaves as well as the glyoxysomes of castor bean endOSperm, all of which do con- tain aspartate aminotransferase (40). The rat liver peroxisomes contained a serine: glyoxylate aminotransferase activity. The mitochondria also contained some serine:g1yoxylate aminotransferase activity but most of it appeared to be from peroxisomal contamination. Most of the alanine:glyoxylate aminotrans- ferase activity was located in the mitochondria as pre- viously reported by Snell £5 31. (147). However, the peroxisomes also contained a Significant amount of this activity. Mitochondrial contamination of the peroxisomes could account for only 6% of this peroxisomal alanine: glyoxylate aminotransferase activity. The serine:g1yoxylate, alanine:glyoxy1ate and aspartate:a-ketoglutarate aminotransferase activities were 117 also present in the supernatant fraction. It has often been reported that mammalian tissues have a distinct soluble aspartate aminotransferase isoenzyme (l7) and apparently this isoenzyme is not a peroxisomal enzyme. Snell 25 al. (147) reported that the soluble alanine: glyoxylate aminotransferase had a different Km value and pH curve than the mitochondrial enzyme. It is possible that this "soluble form" is actually due to broken perox- isomes. Likewise the soluble serine:g1yoxylate amino- transferase may be due to the breakage of peroxisomes. Most of the g1utamate:glyoxylate aminotransferase activity was located in the supernatant fraction from the rat liver homogenate. A very Small amount of activity observed in the peroxisomes is probably not of significance. Since in spinach leaves the serine:g1yoxylate aminotransferase was inhibited by phosPhate, the rat liver glyoxylate aminotransferases were also tested for phos- phate inhibition (Table 11). As in spinach leaves, the serine:g1yoxylate aminotransferase in rat liver peroxi- somes was more sensitive to phosphate inhibition than was the alanine:glyoxylate aminotransferase activity. All three glyoxylate aminotransferases, particularly glutamate: glyoxylate aminotransferase, were inhibited by Triton X-100 treatment. In the case of the very low levels of gluta- mate:g1yoxylate aminotransferase, the phOSphate appeared to protect the activity against the Triton X-100 inhibition. 118 Table 11. The Effect of PhOSphate and Triton X-100 on Rat Liver Peroxisomal Glyoxylate Aminotransferases Rat liver peroxisomes were isolated on a sucrose gradient and assayed by the radiochemical procedure. Buffers were at pH 7. Buffer Serine: Alanine: Glutamate: Glyoxylate Glyoxylate Glyoxylate Relative Rates Cacodylate 100 100 100 Cacodylate + Triton X-100 73 84 O PhoSphate + Triton X-100 39 85 73 119 Dog Kidney Aminotransferase activity of isolated kidney per- oxisomes has not been reported. Dog kidney peroxisomes were isolated in a sucrose gradient in a zonal rotor and assayed for four aminotransferases listed in Table 12. The major portion of all four aminotransferases activities was located in the mitochondrial fraction. The peroxi- somes appeared to contain small amounts of all of the aminotransferase activities except for g1utamate:glyoxylate aminotransferase which was not detected in the peroxisomes. Mitochondrial contamination accounted for 71% of the per- oxisomal aSpartate aminotransferase activity and therefore this enzyme may not be in dog kidney peroxisomes. This 71% contamination was calculated on the basis of the total activity of aspartate aminotransferase observed in the mitochondrial fraction with Triton X-100 present in the assays. Triton X-100 treatment of the mitochondria was found to release considerable latent aSpartate amino- transferase activity. The glyoxylate aminotransferase assays did not contain Triton X-100 as it was inhibitory to these reactions in the rat liver peroxisomes. The glyoxylate aminotransferase activities observed in the peroxisomes are probably near the total activity present in the organelles since peroxisomes did not exhibit much II.|II|III II III I 120 .AH manna mamv coauomum HmwnpaosoouHB oau mo aowumawewuaoo HmBomeouam ou wouanauuum on sea noH£3 cowuomum Hmfiupaonooufie eta a“ >uw>wuom ammuommcmuuoawam asp mo ammuaaonaa one one maDHm> .AH oHan oomvnaowuomum HmBomeouam one mo cofiumawamuaoo HmwupaosoouHB ou wouanwuuuw an coo nowaB cowuomum HmBomwxopaa one ca Nufi>auom anamommamuuocwam are mo awwucaouam can one moaam>a N Hm ma mmm do oumhmunawoquud "aumuumdm< m cm ON mud em aumamxo%au "ocfiuam - . NmH «mm o mumHNonHo "aumBmHDHU N mm om NeNH em mumuexosao "aawcma< A N HuHB.x Huch.x maHoBG naofiuomum meowuomum Hmwuuconooufiz HmBOmeouom :« mua>fluo< aw huw>wuo< HmBomNNOMom Hmwuucosoouwz uamumauomsm mwuoaoaoouwz moBowfixouom amenaMmamHuoaHB¢ .aunuaooua uwhuaB nouosdouuooam can we vehemmm mm3 noH£3 ammuommamuuocfiam oumumusawouoxnd"mumuummmm Ham unmoxa ouapoooud HmofiBanoowpmu onu kn toxemmm oua3 mowuw>wuom ammuammcmuuoawem HH< .Houou Hanan a BN ucowpmuw amouosm a co paumaomw muo3 moHHocmwuo umHSHHaonnm hocpHM_won Baum maaowuumm wcoB¢ mmauw>auo< mwmnommcmuuoch< mo Gowuanwuumwa .NH oHan 121 latency with aspartate aminotransferase. Since mitochondria did exhibit latency, the true glyoxylate aminotransferase activities in the mitochondria may be 2-3 times higher than reported in Table 12. If this would be the case, then nearly all of the observed activity in the peroxisomal fraction would be attributable to mitochondrial contamin- ation. Data from the Triton X-100 treatment of mitochondria may be very difficult to interpret if both latency and inhibition of the enzymes occur. The amount of the aminotransferase activities in the kidney peroxisomal fraction was very small and it is likely that these activities are insignificant. Lee (96) observed in electron micrographs of kidney that the aspartate aminotransferase was concentrated in the sub- apical vesicles, which may be physically similar to peroxisomes. DISCUSSION Spinach leaf peroxisomes were found to contain a serine:g1yoxylate aminotransferase, an alanine:glyoxy1ate aminotransferase and an aspartate:d-ketoglutarate amino- transferase was located only in the peroxisomes. This distribution agrees with Yamazaki and Talbert (174) who measured the enzyme by its serine:pyruvate aminotransferase 122 reaction. Recent studies have also shown serine:g1yoxy- late and g1utamate:glyoxylate aminotransferase activities are present in the microbodies of castor bean endOSperm and sunflower cotyledons (Schnarrenberger, Oeser and Talbert, in press). Cooper and Beevers (33) had reported earlier that castor bean glyoxysomes did not contain g1utamate:glyoxylate aminotransferase. In the C4-plant, Atriplex spongioso, Osmond and Harris (7) found the serine:g1yoxylate aminotransferase is mainly located in the bundle sheadicells. The significance of this obser- vation will become evident in the next chapter of this thesis. The spinach serine:g1yoxylate aminotransferase was inhibited by phoSphate. The nature of this inhibition is not known, but Severin and Dixon (143) have reported phOSphate interferes with the recombination of pyridoxal- 5-phosphate with apoaspartate aminotransferase. However, the serine:g1yoxylate aminotransferase activity in iso- lated Spinach peroxisomes did not Show a requirement for exogenous pyridoxal—S-phosphate. The significance of phosphate inhibition of the serine:g1yoxylate aminotrans- ferase activity is unknown. Peroxisomal metabolism does not utilize phosphorylated substrates or form ATP or phOSphorylated products. Thus phosphates might be important 123 regulants upon peroxisomal activity. The kinetic studies of the D-serine inhibition of serine:g1yoxylate aminotransferase indicate the D-serine affects the binding of the L-serine. In the presence of D-serine, the Lineweaver-Burk plot, with L-serine as the variable substrate, was nonlinear. These results are similar to those reported for malic enzyme from E. coli which showed substrate cooperativity only in the presence of an allosteric inhibitor of the enzyme (138). Aspar- tate aminotransferase from rat liver has recently been shown to be inhibited by DL-glyceraldehyde-3-phosphate (90, 91). However, the Lineweaver-Burk plots at various concentrations of DL-glyceraldehyde-3-phosphate were linear. The serine:g1yoxylate aminotransferase was not active with D-serine. D-amino acid aminotransferases have been found in microorganisms (136). Breidenbach reported the glyoxysomes in castor bean endosperm contained D- aspartate:a-ketoglutarate aminotransferase activity (21). D-aspartate did not significantly inhibit the peroxisomal aSpartate aminotransferase nor did D-alanine inhibit the alanine:glyoxylate aminotransferase activity. Glutamate: glyoxylate aminotransferase has also been reported not to be inhibited by D-glutamate (79). The serine:g1yoxylate inhibition by phosphate may 124 not be of physiological significance because mM concen- trations were required. However, phosphate concentration in plant cells is often quite substantial. However, another compound or compounds of similar nature may be found which would be the physiological regulator of this enzyme. It may be that the D configuration and/or phosphate are im- portant for binding to the enzyme, and if so a phOSphory- lated carbohydrate or amino acid may be the ig_yigg.regu- lator of this transaminase. The aspartate:a-ketoglutarate aminotransferase activity was found to be present in the chloroplasts, mitochondria and peroxisomes, as Yamazaki and Talbert (174) had previously reported. The chloroplastic and mitochondrial enzymes had the same electrophoretic charac- teristic and thus appear to be the same enzyme. One could speculate that the enzyme is coded by the same nuclear gene. The peroxisomes had an isoenzyme which had the characteristics of the chloroplastic-mitochondria1 enzyme. In addition, the peroxisomes had two other isoenzymes of aspartate aminotransferase. None of the three isoenzymes of aspartate aminotransferase activity were due to the nonSpecific activity of the glyoxylate aminotransferases present in the peroxisomes. However, other aminotrans- ferases may be present in the peroxisomes and if they had 125 some activity towards aspartate and a-ketoglutarate, they could explain the reason for the three isoenzymes of aspartate aminotransferase. Another possible explanation for the observed isoenzymes is that different amounts of another protein such as malate dehydrogenase are stuck to the aSpartate aminotransferase. It has been reported that partially purified aspartate aminotransferase from cottonseeds had three bands of activity on electrophoretic gels and the bands also had malate dehydrogenase activity (50). Likewise aspartate aminotransferase and malate dehydrogenase from Neurospora tended to stick together (88). Malate dehydrogenase sticking to aSpartate aminotrans- ferase is probably not the reason for the observed peroxi- somal isoenzymes in spinach. Even though the malate dehydrogenase smeared and appeared to stick to the other proteins during ion exchange chromatography, its elution pattern did not follow the aSpartate aminotransferase peaks. Also, mitochondria contain malate dehydrogenase and yet only one band of aspartate aminotransferase was observed in the polyacrylamide gels. Yamazaki and Talbert (174) reported that of the aminotransferases present in Spinach peroxisomes, aspar- tate aminotransferase had the highest Specific activity while g1utamate:glyoxylate aminotransferase had the lowest 126 specific activity. The present findings give an opposite conclusion. A portion of the higher aspartate aminotrans- ferase activity previously observed in the peroxisomes may have been due to more mitochondrial contamination. Kisaki and Talbert (85) observed that the g1utamate:glyoxylate aminotransferase activity in spinach peroxisomes was five times more active than the glutamate:oxaloacetate amino- transferase activity. Several authors (1, 16, 37) have reported that the glutamate plus oxaloacetate reaction is 2-3 times faster than the reverse reaction, but other authors (33) report aspartate plus a-ketoglutarate is 2-5 times faster. In either case, the aspartate:a-keto- glutarate aminotransferase does not appear to be more active than the g1utamate:glyoxylate aminotransferase in spinach leaf peroxisomes. The two Spinach peroxisomal aminotransferases with the highest Specific activities are postulated to function in the main pathway of glycolate metabolism as depicted in Figure 15. The numbers in the figure refer to the reactions discussed in this text. In reaction 1 glycolate, which is synthesized in the chloroplast, is oxidized in the peroxisome to glyoxylate by glycolate oxidase. Reaction 2 is the formation of glycine by glutamate: glyoxylate aminotransferase. This formation of glycine 127 .uxau osu SH pommaomwp one noHQB mcowuommu ou Human muanBsc oaH .maBomeouad osu mo onwmuso unoao moBHH panama mo meowuomon oHH£3 mmBomeouom mama baa aw Haooo omega pwaom CH mBoHuomom %m3£umm oumaooxau can mo cowumuaao asp magnum wafiuusooo meowuomam mo aoaaawam venomoum ma muswnm 128 anon—nomad. I \ \I/I F\ de. 1/ oumnmuaawouaxud \ .& \.\ auouoomonmxo o oumemuSHw A..\ \xi..=mnDnNQ mancmam no no onwama oumnmuanwouaxse oumBMuan w /\ .2. a. ma Wm H” -V mannmm A / mum>fin§mhxonfihs m mumnmonnaw IV ... \ 38.3w T m Senbsbw AllnI 338.3» I a N . oum>anh ouncmam \’ moo R / ontmuanw YA.V aumnmunawouoxnd oumBMuan $2 + mfivaez u - Mn, \ m Q42 l Ilt\ mozaiiv mmzl A v A onwaausaw 129 can also occur with alanine serving as the amino donor (reaction 3). This alanine:glyoxylate aminotransferase reaction is probably also catalyzed by the glutamate: glyoxylate aminotransferase. The regeneration of alanine can occur with glutamate serving as the amino donor (reaction 4). The g1utamate:glyoxylate aminotransferase may catalyze this g1utamate:pyruvate aminotransferase reaction since mixed substrate studies indicate they were the same enzyme (85). Thus the g1utamate:glyoxylate amino- transferase functions in the conversion of glyoxylate to glycine whether glutamate or alanine serves as the amino donor. The regeneration of glutamate from.a-ketoglutarate in reaction 2 and 4 will be discussed later. The glycine to serine conversion apparently occurs in the mitochondria. Bruin (25) and Kisaki and Talbert (85) have reported the mitochondrial fraction of sucrose gradients could catalyze the glycine to serine conversion. Recently, Kisaki g£_§l, (84, 87) further investigated the mitochondria as the subcellular location of this activity. The serine enters the peroxisome and is con- verted to hydroxypyruvate by serine:g1yoxylate aminotrans- ferase (reaction 6). This conversion was previously postulated to occur with pyruvate serving as the amino acceptor (158). The serine:g1yoxylate aminotransferase 130 reaction eliminates the necessity of alanine and pyruvate being intermediates and as such permits the pathway to operate more independently. The serine:g1yoxylate amino- transferase had a Km (glyoxylate) of 0.15 mM while the g1utamate:glyoxylate aminotransferase reportedly had a Km (glyoxylate) of 4.4 mM (85). These values would suggest that any serine present in the peroxisome would prefer- entially be used as the amino donor for glyoxylate and thus keep the glyoxylate pathway functioning to form glycerate. If the glutamate was used preferentially, serine would tend to accumulate. The pool size of serine may be controlled by regulators acting upon the serine: glyoxylate aminotransferase. The formation of glycerate from hydroxypyruvate (reaction 7) is catalyzed by hydroxypyruvate reductase. Reducing power is shuttled into the peroxisome via malate dehydrogenase and the oxaloacetate product is converted to aSpartate by the aspartate aminotransferases of the peroxisome (reactions 8 and 9). The aspartate could then move or diffuse to the chloroplast, where it would be converted to oxaloacetate by the chloroplast aspartate aminotransferase (reaction 10). The chloroplast NADP-malate dehydrogenase would reduce the oxaloacetate to malate (reaction 11) to com— plete a cycle for the transport of reducing capacity into 131 the peroxisomes. It should be noted that no net nitrogen or ammonia source is needed for this shuttle if the glu- tamate and a-ketoglutarate move between chloroplast and peroxisome. The regeneration of glutamate for reactions 2 and 4 is shown as occurring in two possible ways. The a- ketoglutarate could undergo reductive amination by NADP- glutamate dehydrogenase which is known to be in the chloroplast (reaction 12) (97). This pathway would be a means for the plant to transfer and utilize the ammonia formed from the reduction of N03. However, no net input of ammonia is needed to operate the glycolate pathway if a second method of forming glutamate were used (reaction 12). In this latter scheme, the ammonia released in the conversion of glycine to serine is used to form glutamate. Since the glycine to serine conversion appears to occur in the mitochondria, the mitochondrial NAB-glutamate dehydro- genase would be expected to catalyze this reaction. How- ever, it is interesting to note that Leech and Kirk (97) have suggested that the mitochondrial NAB-glutamate dehydrogenase functions in the direction of oxidation of glutamate and that it is the NADP-glutamate dehydrogenase in the chloroplasts which synthesizes glutamate. Although the glycolate pathway normally is 132 considered to function for the conversion of glycolate to glycerate, a portion of the pathway may be reversed. The peroxisomal glycerate-hydroxypyruvate interconversion is an anaerobic reversible metabolic pathway, but in the oxidative direction it occurs readily only at high pH (135). The conversion of hydroxypyruvate to serine (reac- tion 13) may occur with glutamate as the amino donor. Kisaki and Talbert (85) observed significant rates for glutamate:hydroxypyruvate aminotransferase. The peroxi- somal serine:g1yoxylate aminotransferase may also catalyze the hydroxypyruvate to serine conversion, but probably not with glycine as the amino donor. The serine:pyruvate aminotransferase activity of this enzyme is very low and unless the reverse alanine:hydroxypyruvate aminotransferase reaction is considerably faster, the serine:g1yoxylate aminotransferase probably would not significantly con- tribute in the conversion of hydroxypyruvate to serine. The g1utamate:glyoxylate aminotransferase is postulated to function in the formation of glycine which can then be converted to serine. The glycine and serine can be used for porphyrin and protein synthesis or used to form C1 moieties for nucleic acid and cell wall biosynthesis in the leaf. The g1utamate:glyoxylate amino- transferase was found to greatly increase in activity in 133 response to light in germinating sunflower cotyledons suggesting it may be important for the formation of gly- cine during the greening of leaf tissue (Schnarrenberger, Oeser and Talbert, in press). If the leaf tissue over produces glycine and serine, as during photoreSpiration, the serine:g1yoxylate aminotransferase can direct the carbon compounds to gly- cerate and back into sugars. Since this aminotransferase catalyzes the first step in the conversion of serine to glycerate it may well be under metabolic regulation. The inhibition by D-serine indicates this serine:g1yoxylate aminotransferase is an atypical aminotransferase and warrants further study. The function of the peroxisomal aSpartateza-keto- glutarate aminotransferase is postulated to be in the malate-aspartate shuttle of reducing power into the organelle. This function for aspartate aminotransferase has been proposed previously (95, 98). The peroxisomal isoenzymes of aspartate aminotransferase may have different kinetic characteristics which would favor one direction of the reaction. The peroxisomal malate dehydrogenase was found to be different than the mitochondrial enzyme and was postulated to function in the direction of malate oxidation (173). 134 The aminotransferase content of the mammalian peroxisomes was different than that in spinach leaf perox- isomes. The rat liver peroxisomes did not contain aSpar- tate:a-ketoglutarate aminotransferase nor significant levels of g1utamate:glyoxylate aminotransferase. However, they did appear to contain serine:g1yoxylate and alanine: glyoxylate aminotransferase activities. Rowsell and coworkers (134) have suggested that alanine:glyoxylate aminotransferase in rat liver is directly correlated with the gluconeogenic capabilities of the liver since it in- creased in activity with glucagon treatment. The alanine: glyoxylate aminotransferase is located in both the mito- chondria and peroxisomes of rat liver. Because of pre- vious Speculation on the gluconeogenic function of peroxisomes (40, 154), it is tempting to speculate that the increased alanine:glyoxy1ate aminotransferase in liver occurred in the peroxisomes. Peroxisomes seem to have a correlation with gluconeogenesis in that they are found in significant amounts only in those tissues capable of gluconeogenesis. It should be pointed out that anyone isolating and purifying alanine:glyoxylate aminotransferase from rat liver by the classical enzyme isolation procedures would likely end up with the mitochon- drial enzyme Since it has the most activity. 135 The diversification of the enzyme composition of peroxisomes is evident with the dog kidney peroxisomes. These microbodies contained very low amounts of those aminotransferases which were assayed. In the kidney, the mitochondria contained most of the aminotransferase activities. In the present studies the kidney cortex and medulla tissues were not separated before isolation of the peroxisomes. It may very well be the two types of tissues would contain peroxisomes with different enzyme compositions. The enzymatic composition of microbodies is known to vary depending upon the source of the organelles (40). This variation probably reflects the different functions of the microbodies. The glyoxysomes in castor bean endo- sperm and sunflower cotyledons function to form succinate from fatty acids. Numerous functions for peroxisomes in spinach leaves have been described (154), but glycine and serine synthesis from glycolate is certainly well estab- lished. The function of liver and kidney peroxisomes is not known. The liver peroxisomes do not contain aspartate aminotransferase or malate dehydrogenase and if these enzymes function in the shuttling of reducing power, as has been postulated, then liver peroxisomes may not be associated with the oxidation-reduction of NAD. CHAPTER II DISTRIBUTION OF PEROXISOMAL ENZYMES AMONG LEAF CELLS The plants which fix 002 only by the reductive pentose phOSphate cycle are called C3-plants. These plants exhibit photorespiration, that is, they take up 02 and form CO2 in the light. It has been suggested that the glycolate pathway is the source of the photoreSpired COZ. In contrast to the C3-plants, the C4-p1ants do not lose C02 by photorespiration. The 04-p1ants contain PEP- carboxylase and initially fix C02 into oxaloacetate. In the earlier studies of C4-plants, the RuDP carboxylase activity was reported to be lower than that found in the C3-plants and could not account for the in_yi!3 photo- synthetic rates observed in the C4-p1ants. Since the glycolate pathway has always been closely associated with the C3-cycle, it was also not surprising to find low levels of the glycolate pathway enzymes in C4-plants. It was thus suggested that low levels of activity for the glycolate pathway may be the reason C4-plants did not photoreSpire (120). 136 137 ngrkman and Gauhl (12) and Andrews and Hatch (3) found that exhaustive grinding of the tissues of C4-plants resulted in recoveries of much higher values of RuDP carboxylase than previously reported. Bjérkman showed that "gentle" grinding broke mesophyll cells but did not release the RuDP carboxylase. After very harsh grinding to break the bundle sheath cells, most of the RuDP carboxy- lase activity was solubilized. This was interpreted to mean that RuDP carboxylase was located in the bundle Sheath cells. The mesophyll cells, which were broken easily, contained most of the PEP-carboxylase. Thus it appeared that in C4-plants the mesophyll cells contained the C4- cycle enzymes while the bundle sheath cells contained the C3-cycle enzymes. If the glycolate pathway were associated with the C3-cycle within C4-plants, then the glycolate pathway enzymes would also be expected to be located in the bundle sheath cells. Thus the purpose of this study was to determine the cellular location and some of the characteristics of the glycolate pathway enzymes in C4-plants. MATERIALS AND METHODS ‘Most of the materials and methods used in these studies have been published and can be found in the 138 reprint in Appendix A. The preparation of the starch gel for starch gel electrophoresis was the same as that described in Chapter I. The same buffer was used in both the starch gel and the electrode tanks. For catalase, the buffer was 6 mM Tris-50 mM HEPES at pH 6.8. The buffer system for hydroxypyruvate reductase was 10 mM Tris-HCl at pH 8.3, while for glycolate oxidase the buffer was 12 mM Tris-3 mM citrate at pH 7.0. RESULTS LOCATION OF GLYCOLATE PATHWAY ENZYMES The major portion of this work has been published (129) and a reprint is attached in the appendix. Thus the results will only be summarized here. BjBrkman and Gauhl (12) used a differential grinding technique to establish that the bundle sheath cells of C4-plants contained most of the RuDP carboxylase, whereas the mesophyll cells contained most of the PEP carboxylase. The differential grinding procedure was used to determine the distribution of some of the enzymes of the glycolate pathway in 03- and CA-plants. With corn leaf tissue, only 60% of the glycolate oxidase and 40% of the P-glycolate phosphatase were extracted after 2 min of homogenization in 139 a Waring blendor (Figure l) in Appendix A. The rest of these activities were obtained by harsh grinding of the bundle Sheath cells with a roller mill. 3-P-Glycerate phOSphatase was completely extracted by the 2 min of homogenization and thus was used as a marker enzyme for mesophyll cells. The peroxisomal enzymes were readily solubilized from spinach and wheat by homogenization in.a Waring blendor (Table 1) in Appendix.A. This was to be expected since these plants do not contain bundle sheath cells. In corn and sugarcane, a 30 sec homogenization solubilized 30-50% of the glycolate oxidase and hydroxypyruvate re- ductase activities. The remaining 50-70% of these activ- ities were solubilized by grinding with the roller mill. Since 30% of the 3-P-g1ycerate phosphatase activity was in the extract from the roller mill, some mesophyll cells were probably contributing to the activities observed in the extract from the roller. The fact that the peroxisomal enzymes; catalase, hydroxypyruvate reductase and glycolate oxidase; varied in their distribution may indicate that the peroxisomes in the two types of cells have different enzyme compositions. The varied distribution could also be an artifact caused by the differential solubilization of the peroxisomal enzymes. Fredrick and Newcomb (58) 140 found the peroxisomes in the bundle sheath cells of corn and sudan grass stained darker for catalase activity than did the mesophyll cell peroxisomes. They also reported that each bundle sheath cell of corn contained three to four times as many microbodies as did the mesophyll cell. The activity of glycolate oxidase found in corn and sugarcane was two to three times higher than previously reported (157), but these C -p1ants still contained about 4 half the activity found in the C3-p1ants, spinach and wheat. However, the lack of CO -photorespiration in C4- 2 plants can not be attributed to the lack of glycolate pathway enzymes. This is particularly evident in the Atriplex. Atriplex ragga, a C4-p1ant contained nearly the same amount of glycolate oxidase as Atriplex patula a C3-p1ant. Further support of the location of the glycolate pathway enzymes was sought in variegated corn mutants. If only one cell type (bundle sheath or mesophyll) was affected in the mutation, and if the chlorophyll-less cells would contain lower levels of glycolate oxidase, as in etiolated tissue, then assays of the mutants should reveal in which cell the glycolate oxidase was located. Micro- scopic observations on many variegated mutants of corn were made but none was found in which only one type of cell 141 contained all of the chlorophyll. Likewise the glycolate oxidase distribution was relatively constant among all of the mutants. Thus no additional support for the location of the glycolate pathway enzymes in the C4-plants was obtained by studies with the available mutants. The peroxisomal enzyme distributions suggested that both the mesophyll and the bundle sheath cells may contain the glycolate pathway enzymes. This idea is supported by the observations of Fredrick and Newcomb (58) and other authors (73, 92) that microbodies are present in both types of cells. If these enzymes are present in both cell types, they may be present as iso- enzymes and have different characteristics. Therefore studies were conducted to determine if catalase, hydroxy- pyruvate reductase or glycolate oxidase activities in extracts of the mesophyll cells as obtained by homogeni- zation in a Waring blendor, were distinguishable from the activities in extracts of the bundle sheath cells as obtained by the grinding in the roller Inill. CATALASE The catalase activity in corn appeared to be present in both bundle sheath and mesophyll cells. Starch gel electrophoresis was used to measure the electrophoretic 142 pattern of catalase from whole leaf extract, mesophyll cell extract and bundle sheath extract. Each sample contained one anionic band of catalase activity and the enzyme had the same mobility in all three samples. Like- wise, only one band was observed in mixtures of the three samples. There have been reports of catalase isoenzymes in plants (141, 144). Scandalios (141) has shown that in corn endosperm tissue some catalase isoenzymes are formed because the parent plants had different molecular forms of catalase. If parent X has catalase composed of AAAA and parent Y has a catalase of BBBB, then the F1 will have five isoenzymes produced by the combinations of the two types of subunits A and B. In one report by Scandalios (140) two catalase isoenzymes were present in the leaf tissue of an inbred line of corn. In the hybrid corn leaf tissues used in my studies, isoenzymes were not observed. HYDROXYPYRUVATE REDUCTASE The kinetic characteristics of the hydroxypyruvate reductase activities in the two types of cells from corn leaves were studied. The Km for hydroxypyruvate in the mesophyll cells and in the bundle sheath cells was 73 uM and 81 uM respectively. These values are not considered .. III-Ill" Ii III It'll 143 to be significantly different (Figure 16). Hydroxypyru- vate reductase from Spinach leaves also is known to reduce glyoxylate to glycolate (89, 159). The glyoxylate reduc- tase activities in the two types of cells from corn leaves also exhibited similar kinetics. The Km for glyoxylate was 1.0 mM and 0.4 mM for the mesophyll cell and bundle sheath cell respectively (Figure 16). Spinach leaf hydroxypyruvate reductase has a Km for hydroxypyruvate of 50 uM which is similar to that observed for the enzyme from corn leaves (89, 159). However, the glyoxylate reductase from corn leaves had a 25-50 fold lower Km for glyoxylate than that reported for the enzyme from spinach leaves (89, 159). This difference appears significant. Starch gel electrophoresis was used to search for isoenzymes of hydroxypyruvate reductase. The bundle sheath cells contained a small cationic band of enzyme activity and a heavy anionic band. The mesophyll cell extract had only a small anionic band at the same Rf as from the bundle sheath cells. The mesophyll may have also contained the cationic form but the activity was too low to detect. Since nearly the same activity from each extract was applied to the gel, the reason for the weaker staining of the mesophyll extract is not known. 144 Figure 16 Reaction Kinetics for Hydroxypyruvate Reductase and Glyoxylate Reductase from Mesophyll and Bundle Sheath Cells of Maize Mesophyll extracts were prepared by homogenization of corn leaf tissue in a Waring blendor. The residue from the homogenization was ground further in a roller mill to obtain the bundle sheath cell extracts. Both reductases were assayed by measuring the oxidation of NADH. s/v (mM/nmoles x min") s/v (mM/nmales x min") ,0 i“ ,o i Q N I 145 Mesophyll x Km= 73 uM ’/ Bundle Sheath Km= 8| 11M I I 1 l 0.5 I.O I.5 2.0 Hydroxypyruvate (mM) N Ofl‘ 61' m 5 l 01 Mesophyll ,’ Km= |.O mM ,’ Bundle Sheath Km = 0.4 mM 1 l l 1 IO IS 20 25 Glyoxylate (mM) 1 5 146 GLYCOLATE OXIDASE Extracts of corn leaf tissue were electrophoresed and stained for glycolate oxidase activity. Both starch gels and polyacrylamide gels were used, but no activity was ever observed. Glycolate oxidase from Spinach leaves moved into both types of gels and Stained very readily. In the staining reaction for glycolate oxidase, phenezine methylosulfate (PMS) supposedly links the transfer of electrons from the reduced FMN of glycolate oxidase to nitroblue tetrazolium (NBT), which in the reduced state is blue and less soluble. The necessity for PMS in the staining solution was not determined. Since the corn gly- colate oxidase could not be detected after electrophoresis, the staining reaction was studied in the Spectrophoto- meter to determine if glycolate oxidase activity could be observed with PMS and NBT. The spinach enzyme gave a measurable rate of NBT reduction, but the corn enzyme did not (Table 13). Something in the corn enzyme preparation may have been binding or destroying the PMS, but this seems unlikely since the rate of spinach glycolate oxidase was unchanged in the presence of corn extract. If the PMS concentration was increased 2.5 fold, a very slight reaction was observed under anaerobic conditions. The Spinach glycolate oxidase activity was considerably higher under 147 Table 13. Glycolate Oxidase Activity in Corn and Spinach Leaf Extracts The leaf extracts were assayed for glycolate oxidase by the standard DCIP linked assay as previously published (Appendix.A). The PMS-NBT assay mixture con- sisted of 46 umoles and 0.25 umoles of FMN. All assays contained 200 ug of PMS except where noted. The reaction was started with the addition of 25 umoles of glycolate. Total assay volume was 2.5 ml and all assays were anaer- obic except for the last one. Plant Extract DCPIP Assay PMS-NBT Assay Relative Rate Spinach 100 100 Corn 109 0 Spinach + Corn - 100 Corn (500 ug PMS) - 11 Spinach (aerobic) - 19 148 anaerobic conditions as compared to aerobic conditions and so it is very unlikely that the corn enzyme could be detected under aerobic conditions. The necessity of PMS in the glycolate oxidase stain was not determined. Frigerio and Harbury (59) reported spinach glycolate oxidase did not link with one electron acceptors and so the spinach enzyme was very likely reacting directly with NBT as other flavoproteins are known to do (44). This failure of glycolate oxidase to reduce NBT is not limited to C4-plants such as corn. Glycolate oxidase from wheat, a C3-plant, also could not be detected on electrophoretic gels (E. B. Nelson, personal communication). The algae enzyme (discussed below) appeared to link to PMS (116), but it was not stained by the PMS-NBT system (E. B. Nelson, personal communication). Extracts of corn leaves, after differential grinding to obtain the mesophyll cells and bundle sheath cells, were assayed by the DCIP assay to determine the kinetics of glycolate oxidase. The Km for glycolate was nearly the same in the two types of cell homogenates (Figure 17). The Km value of 0.5 mM for glycolate oxidase from corn is similar to that reported for the enzyme from Spinach (154). Nelson and Talbert (116) found that the glycolate 149 .mHUQ ma canuospan man wannammoB he aa%mmmm mo3 ommpnxo oumHaoham use .uomnuxm HHoo gunmen oaaane men Gnmuno on HHnB noHHon a an naannam pcdonw was conumnnaowoBan can Bonn oaanmon aaH .noaaaan wannmz a an mammnu mama anao mo BonummnaawoBos he aonmmonm onaB anomnuxa Hahamomaz buns: no .28 fimmam «Scam was 3.3882 Banm ammanxo aumaaomau ma monuoanm aanuommm S onawnm 150 A525 323:0 m. o. m 17 a _o .28 9025. foo o cam ...m _u ind .26 m6 "Ex :EQomoz .o (|_ugu1 x semwwww) .m/s 151 oxidase from algae did not link to O2 and was therefore called glycolate dehydrogenase. The algae enzyme could not be stained by the PMS-NBT system, it was strongly inhibited by CN' and it would oxidize D-lactate but not L-lactate. In contrast the spinach glycolate oxidase was not significantly affected by CN- and it oxidized L-lactate but not D-lactate (116). The characteristics of the corn glycolate oxidase were determined to see if it was related to the algae enzyme since as already mentioned, neither the algae enzyme nor the corn enzyme could be stained with the PMS-NBT system. The corn enzyme oxidized L-lactate but not D-lactate (Table 14). Likewise glycolate oxidase from sugarcane had the char- acteristics of the spinach enzyme. Cyanide did not appreciably affect the activity of the corn enzyme (data not shown). The above results on catalase, hydroxypyruvate reductase and glycolate oxidase suggest that the mesophyll cells and the bundle sheath cells in C4-p1ants do not contain distinct isoenzymes of these enzymes. The only significant difference found in the enzymes between C3- and Ca-plants was in a low Km (glyoxylate) for the glyoxylate reductase activity. 152 Table 14. Glycolate Oxidase Activity From Various Sources Spinach and sugarcane samples were whole leaf homogenates while the two corn extracts were prepared by the differential grinding procedure. The algae extract was prepared by E. B. Nelson. Substrate Extract Spinach Sugarcane Corn Algae Bundle Mesophyll Sheath nmoles x min'1 x ml-1 Glycolate 213 25 27 12 18.3 D-Lactate 0 0 0 0 18.8 L-Lactate 163 25 21 8 0 153 DISCUSSION By differential grinding of corn leaf tissue it has been shown that the enzymes of the glycolate pathway were preferentially located in the bundle sheath cells but that some activity was also very likely in the mesa- phyll cells. Even though the two types of cells appear to have specialized functions, no isoenzymes or kinetic differences for glycolate oxidase, hydroxypyruvate reduc- tase or catalase were found between the two cell types. The corn did have an NADH-glyoxylate reductase activity with a lower Km than is found in spinach leaves. Whether this activity is catalyzed by the NADH-hydroxypyruvate reductase, as it is in Spinach, or whether a separate enzyme is present in corn is not known. An NADPH-glyoxylate reductase with a low Km for glyoxylate is present in spinach chloroplsts but this enzyme is not very reactive with NADH (159, 180). Spinach glycolate oxidase reduced NBT in a PMS- NBT system. Since the necessity of PMS was not established, the glycolate oxidase reduction of NBT may have been direct rather than linked through the PMS. The corn, wheat and algae enzymes would not stain on electrophoretic gels with a PMS-NBT staining solution. Dixon (44) in his report on the characteristics of flavoproteins mentions only one 154 enzyme which exhibits different acceptor Specificities depending upon the source of the enzyme. Xanthine oxidase from chicken liver and pigeon liver will not react with O 2 whereas the mammalian enzyme does react with O Glyco- 2. late oxidase appears to be another example where acceptor specificity depends upon the source of the enzyme. The glycolate oxidase from all plants and algae studied so far will reduce DCIP. The algae enzyme will also reduce PMS but not 02 nor several other acceptors (116). The corn and wheat enzymes will reduce O2 (155) but appear not to reduce a PMS-NBT system. Further discussion about the distribution of the glycolate pathway enzymes in the bundle sheath and mesophyll cells in C3- and C4-plants has been published (129, Appendix A). BIBLIOGRAPHY 10. 11. 12. BIBLIOGRAPHY Albaum, H. G. and P. P. Cohen. 1943. J. Biol. Chem. 149:19-27. Anderson, N. G., D. A. Waters, W. D. Fisher, G. B. Cline, C. E. Nunley, L. H. Elrod and C. T. Rankin, Jr. 1967. Anal. Biochem. 21:235-252. Andrews, T. T. and M. D. Hatch. 1971. Phytochem. 10:9-15. Arnon, D. I. 1949. Plant Physiol. 24:1-15. Babson, A. L., P. O. Shapiro, P. A. R. Williams and G. E. Phillips. 1962. Clin. Chim, Acta. 7:199-205. Bachrach, U. 1957. Biochem. J. 66:559-561. Bassham, J. A., A. A. Benson, D. Kay, Z. Harris, A. T. Wilson and M. Calvin. 1954. J. Am. Chem. Soc. 76:1760-1770. Baudhuin, P., H. Beaufay, Y. Rahman-Li, 0. Z. Sellinger, R. Wattiaux, P. Jacques and C. de Duve. 1964. Biochem. J. 92:179-205. Baudhuin, P., M. Miller, B. Poole and C. de Duve. 1965. Biochem. Biophys. Res. Comm. 20:53-59. Bergami, M., G. Marina and V. Scardi. 1968. Biochem. J. 110:471-473. Berry, J. A., W. J. S. Downton and E. B. Tregunna. 1970. Can. J. Bot. 48:777-786. Bjarkman, o. and E. Gauhl. 1969. Planta. 88:197-203. 155 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 156 Bjorkman, 0., W. M. Hiesey, M. Nobs, F. Nicholson and R. W. Hart. 1967. Carnegie Inst. Washington Yearb. 66:228-232. BjBrkman, 0., R. W. Pearcy and M. A. Nobs. 1970. Carnegie Inst. Washington Yearb. 69:640-648. Black, Jr., C. C. and H. H. Mollenhauer. 1971. Plant Physiol. 47:15-23. Bone, D. H. and L. Fowden. 1960. J. Exp. Bot. 11:104-115. Boyd, J. W. 1961. Biochem. J. 81:434-441. Boyde, T. R. C. 1968. Biochem. J. 106:581-586. Boynton, J. E., M. A. Nobs, 0. Bjorkman and R. W. Pearcy. 1970. Carnegie Inst. Washington Yearb. 69:629-632. Braunstein, A. E. and M. G. Kritzmann. 1937. Nature. 140:503-504. Breidenbach, R. W. 1969. Ann. N. Y. Acad. Sci. 168:342-347. Breidenbach, R. W. and H. Beevers. 1967. Biochem. Biophys. Res. Comm. 27:462-469. Brock, B. L. W., D. A. Wilkinson and J. King. 1970. Can. J. Biochem. 48:486-492. Brubaker, J. L., M. D. Upadhya, Y. Makinen and T. MacDonald. 1968. Physiol. Plant. 21:930-940. Bruin, W. J. 1969. Ph. D. Thesis. Michigan State University. Calvin, M. and J. A. Bassham. 1962. The Photosynthe- sis of Carbon Compounds. W. A. Benjamin, Inc., New York. p. 127. Cammarata, P. S. and P. P. Cohen. 1950. J. Biol. Chem. 187:439-452. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 157 Cammarata, P. S. and P. P. Cohen. 1951. J. Biol. Chem. 193:53-62. Cannell, R. Q., W. A. Brun and D. N. Moss. 1969. Crop Sci. 9:840-841. Cheung, G. P., I. Y. Rosenblum and H. J. Sallach. 1968. Plant Physiol. 43:1813-1820. Clagett, C. 0., N. E. Tolbert and R. H. Burris. 1949. J. Biol. Chem. 178:977-987. Combepine, G. 1969. C. R. Acad. Sci., Paris, Ser. D. 268:2694-2696. [Cited from Chem. Abst. 71:78285 (1969)]. Cooper, T. G. and H. Beevers. 1969. J. Biol. Chem. 244:3507-3513. Cooper, T. G. and H. Beevers. 1969. J. Biol. Chem. 244:3514-3520. Cossins, E. A. and S. K. Sinha. 1965. Can. J. Biochem. 43:495-506. Cruickshank, D. H. and F. A. Isherwood. 1958. Biochem. J. 69:189-195. Davies, D. D. and R. J. Ellis. 1961. Biochem. J. 78:623-630. Davis, B. J. 1964. Ann. N. Y. Acad. Sci. 121, 404. Davis, D. D. 1968. Recent Aspects of Nitrogen Metabolism in Plants. Ed. E. J. Hewitt and C. V. Cutting. Academic Press, N. Y. de Duve, C. 1969. Ann. N. Y. Acad. Sci. 168:369- 381. de Duve, C. 1969. Proc. Roy. Soc. B. 173:71-83. de Duve, C. and P. Baudhuin. 1966. Physiol. Rev. 46:323-357. ‘ 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 158 de Duve, C., B. C. Pressman, R. Gianetto, R. Wattiaux and F. Appelmans. 1955. Biochem. J. 60:604- 617. Dixon, M. 1971. Biochim. Biophys. Acta. 226: 269-284. Dixon, M. and E. C. Webb. 1964. Enzymes. Academic Press, N. Y. p. 950. Donaldson, R. 1971. Ph. D. Thesis, Michigan State University. Downton, W. J. S., T. Bisalputra and E. B. Tregunna. 1969. Can. J. Bot. 47:915-919. Edwards, G. E. and C. C. Black, Jr. 1971. Plant Physiol. 47:149-156. Ellis, R. J. and D. D. Davies. 1961. Biochem. J. 78:615-623. Fasella, P., F. Bossa, C. Thrano and A. R. Fanelli. 1966. Enzymoligia. 30:198-205. Fine, 1. H. and L. A. Costella 1963. Methods of Enzymology. 6:958-972. Fleisher, G. A., C. S. Potter and K. G. Wakim. 1960. Proc. Soc. Exp. Biol. Med. 103:229- 231. Fock, H., H. Schaub, W. Hilgenberg and K. Eg1e. 1969. Planta. 86:77-83. Forrester, M. L., G. Krotkov and C. D. Nelson. 1966. Plant Physiol. 41:422-427. Forrester, M. L., G. Krotkov and C. D. Nelson. 1966. Plant Physiol. 41:428-431. Frederick, S. E. and E. H. Newcomb. 1969. J. Cell Biology. 43:343-353. Frederick, S. E. and E. H. Newcomb. 1969. Science. 163:1353-1355. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 159 Frederick, S. E. and E. H. Newcomb. 1971. Planta. 96:152-174. Frigerio, N. A. and H. A. Harbury. 1958. J. Biol. Chem. 231:135-157. Fukuda, T. and T. Hayashi. 1958. J. Biochem. (Japan) 45:469-474. Gibbs, M. 1969. Ann. N. Y. Acad. Sci. 168:356-368. Godson, G. N. 1970. Anal. Biochem. 35:66-76. Goldsworthy, A. 1966. Phytochem. 5:1013-1019. Graham, D. and M. L. Reed. 1971. Nature New Biol. 231:81-83. Graham, D. and C. P. Whittingham. 1968. Z. Pflan- zenphysiol. 58:418-427. Green, D. E., L. F. Leboir and V. Nocito. 1945. J. Biol. Chem. 161:559-582. Gruber, P. J., R. N. Trelease, W. M. Becker and E. H. Newcomb. 1970. Planta. 93:269-288. Hatch, M. D. and C. R. Slack. 1966. Biochem. J. 101:103-111. Hatch, M. D. and C. R. Slack. 1970. Ann. Rev. Plant Physiol. 21:141-162. Hatch, M. D., C. R. Slack and H. S. Johnson. 1967. Biochem. J. 102:417-422. Herbst, R. M. 1944. Advances in Enzymology. 4: 75-97. Hess, J. L. and N. E. Tolbert. 1966. J. Biol. Chem. 241:5705-5711. Hilliard, J. H., V. E. Gracen and S. H. West. 1971. Planta. 97:93-105. Hofstra, G. and J. D. Hesketh. 1969. Planta. 85:228-237. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 160 Holmgren, P. and P. G. Jarvis. 1967. Physiol. Plant. 20:1045-1051. International Union of Biochemistry. 1965. Enzyme Nomenclature. Elsevier Publ. Co., N. Y. p. 219. Jackson, W. A. and R. J. Volk. 1969. Nature. 222:269-271. Jackson, W. A. and R. J. Volk. 1970. Ann. Rev. Plant Physiol. 21:385-432. Jenkins, W. T., D. A. thantis and I. W. Sizer. 1959. J. Biol. Chem. 234:51-57. Jimenez, E., R. L. Baldwin, N. E. Tolbert and W. A. Wood. 1962. Arch. Biochem. Biophys. 98: 172-175. Johnson, H. S. and M. D. Hatch. 1969. Biochem. J. 114:127-134. Jolliffe, P. A. and E. B. Tregunna. 1968. Plant Physiol. 43:902-906. King, J. and E. R. Waygood, 1968. Can. J. Biochem. 46:771-779. Kisaki, T., A. Imai and N. E. Tolbert. 1971. Plant and Cell Physiol. 12:267-273. Kisaki, T. and N. E. Tolbert. 1969. Plant Physiol. 44:242-250. Kisaki, T. and N. E. Tolbert. 1970. Plant Cell Physiol. 11:247-258. Kisaki, T., N. Yoshida and A. Imai. 1971. Plant Cell Physiol. 12:275-288. Kitto, G. B., M. E. Kottke, L. H. Bertland, W. H. Murphey and N. 0. Kaplan. 1967. Arch. Biochem. Biophys. 121:224-232. Kohn, L. D. and W. A. Warren. 1970. J. Biol. Chem. 245:3831-3839. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 161 Kopelocich, L., L. Sweetman and J. S. Nisselbaum. 1970. J. Biol. Chem. 245:2011-2017. Kopelovich, L., L. Sweetman and J. S. Nisselbaum. 1971. Eur. J. Biochem. 20:351-362. Laetsch, W. M. 1968. Amer. J. Bot. 55:875-883. Laetsch, W. M. and I. Price. 1969. Amer. J. Bot. 56:77-87. Laetsch, W. M., D. A. Stetler and.A. J. Vlitos. 1966. Z. Pflanzenphysiol. 54:472-474. Lardy, H. A., V. Paetkau and P. Walter. 1965. Proc. Nat'l. Acad. Sci. 53:1410-1415. Lee, S. H. 1970. J. Cell Biol. 45:644-649. Leech, R. M. and P. R. Kirk. 1968. Biochem. Biophys. Res. Comm. 32:685-690. Lehninger, A. L. 1970. Biochemistry. Worth Publishers, Inc., N. Y. p. 833. Leonard, M. J. K. and R. H. Burris. 1947. J. Biol. Chem. 170:701-709. Loomis, W. D. and P. K. Stumpf. 1958. Handbuch der Pflanzenphysiologie. Ed. W. Ruhland. 8:249-261. Lowry, H. 0., N. J. Rosebrough, A. L. Farr and R. J. Randall. 1951. J. Biol. Chem. 193:265-275. Luck, H. 1965. 'Methods of Enzymatic Analysis. Ed. H. U. Bergmeyer. Academic Press, N. Y. p. 886. Mahler, H. R. and E. H. Cordes. 1966. Biological Chemistry. Harper and Row Publ., N. Y. p. 872. Mathieu, Y. 1966. Pysiol. Vegetale. 4:299-316. [Cited from Chem. Abst. 66:489 (1967)]. Meister, A. 1954. J. Biol. Chem. 206:587-596. 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 119. 120. 121. 162 Meister, A. 1962. The Enzymes. Ed. P. D. Boyer, H. Lardy, and K. Myrback. 6:193-217. Meister, A., H. A. Sober, S. V. Tice and P. E. Fraser. 1952. J. Biol. Chem. 197:319-330. Menz, K. M., D. N. Mess, R. Q. Cannell and W. A. Brun. 1969. Crop Sci. 9:692-694. Morino, Y., H. Itoh and H. Wada. 1963. Biochem. Biophys. Res. Comm. 1963. 13:348-352. Moss, D. N. 1968. Crop Sci. 8:71-76. Moss, D. N. and H. P. Rasmussen. 1969. Plant Physiol. 44:1063-1068. Mukerji, S. K. and I. P. Ting. 1968. Phytochem. 7:903-911. Mfiller, M. 1969. Ann. N. Y. Acad. Sci. 168:292- 301. Nakada, H. I. 1964. J. Biol. Chem. 239:468-471. Nakada, H. I. and S. Weinhouse. 1953. J. Biol. Chem. 204:831-836. Nelson, E. B. and N. E. Tolbert. 1970. Arch. Biochem. Biophys. 141:102-110. Nisselbaum, J. S. 1968. Anal. Biochem. 23:173- 181. Nisselbaum, J. S., and C. Bodansky. 1964. J. Biol. Chem. 239:4232-4236. O'Kane, D. E. and I. C. Gunsalus. 1947. J. Biol. Chem. 170:425-432. Osmond, C. B. 1969. Biochim. Biophys. Acta. 172:144-149. Osmond, C. B. 1971. Aust. J. Biol. Sci. 24: 159-163. 122. 123. 124. 125. 126. 127. 128. 129. 130. 131. 132. 133. 134. 135. 136. 137. 163 Osmond, C. B. and B. Harris. 1971. Biochim. Biophys. Acta. 234:270-282. Osmond, C. B., J. H. Troughton and D. J. Goodchild. 1969. Z. Pflanzenphysiol. 61:218-237. Papadimitriou, J. M. and P. Van Duijn. 1970. J. Cell Biol. 47:71-83. Poskuta, J. 1969. Physiol. Plant. 22:76-85. Rabson, R., N. E. Tolbert and P. C. Kearney. 1962. Arch. Biochem. Biophys. 98:154-163. Rautanen, N. 1946. J. Biol. Chem. 163:687-688. Rehfeld, D. W. and N. E. Tolbert. 1971. Plant Physiol. 47S:28. Rehfeld, D. W., D. D. Randall and N. E. Tolbert. 1970. Can. J. Bot. 48:1219-1229. Rendon, A. and A. W. Waksman. 1971. Biochem. Biophys. Res. Comm. 42:1214-1219. Rhoades, M. M. and A. Carvalho. 1944. Bull. Torrey Bot. Club. 71:335-346. Richardson, K. E. and N. E. Tolbert. 1961. J. Biol. Chem. 236:1285-1290. Rocha, V. and I. P. Ting. 1970. Arch. Biochem. Biophys. 140:398-407. Rowsell, E. V., K. Snell, J. A. Carnie and A. H. Al-Tai. 1969. Biochem. J. 115:1071-1073. Sallach, H. J. 1966. Methods of Enzymol. 9:221- 228. Sallach, H. J. and L. A. Fahien. 1969. Metabolic Pathways. Ed. D. M. Greenberg. 3:1-94. Santarius, K. A. and C. R. Stocking. 1969. Z. Naturforsch. B. 24:1170-1179. 138. 139. 140. 141. 142. 143. 144. 145. 146. 147. 148. 149. 150. 151. 152. 153. 164 Sanwal B. D. and R. Smando. 1969. J. Biol. Chem. 244:1824-1830. Sanwal, B. D., M. W. Zink and G. Din. 1964. Moderne Methoden der Pflanzenanalyse. 7:361- 391. Scandalios, J. 1964. J. Heredity. 55:281-285. Scandalios, J. 1969. Biochem. Genetics. 3:37-79. Schnaitman, C. and J. W. Greenawalt. 1968. J. Cell Biol. 38:158-175. Severin, E. S. and H. B. F. Dixon. 1968. Biochem. J. 110:19P. Shannon, L. M. 1968. Ann. Rev. Plant Physiol. 19:187-210. Slack, C. R. 1969. Phytochem. 8:1387-1391. Slack, C. R., M. D. Hatch and D. J. Goodchild. 1969. Biochem. J. 114:489-500. Snell, K., E. V. Rowsell and J. A. Carnie. 1970. Biochem. J. 119:64P. Splittstoesser, W. E. 1970. Plant Cell Physiol. 11:579-588. Splittstoesser, W. E. and S. A. Stewart. 1970. Physiol. Plant. 23:1119-1126. Thompson, J. S. and K. E. Richardson. 1966. Arch. Biochem. Biophys. 117:599-603. Thompson, J. S. and K. E. Richardson. 1967. J. Biol. Chem. 242:3614-3619. Tolbert, N. E. 1963. Nat. Acad. Sci. Nat. Res. Coun. Publ. 1145:648-662. Tolbert, N. E. 1971. Methods in Enzymology 23:665-682. 154. 155. 156. 157. 158. 159. 160. 161. 162. 163. 164. 165. 166. 167. 168. 165 Tolbert, N. E. 1971. Ann. Rev. Plant Physiol. 22:45-74. Tolbert, N. E. and M. S. Cohan. 1953. J. Biol. Chem. 204:639-648. Tolbert, N. E., A. Oeser, T. Kisaki, R. H. Hageman and R. K. Yamazaki. 1968. J. Biol. Chem. 243:5179-5184. Tolbert, N. E., A. Oeser, R. K. Yamazaki, R. H. Hageman and T. Kisaki. 1969. Plant Physiol. 44:135-147. Tolbert, N. E. and R. K. Yamazaki. 1969. Ann. N. Y. Acad. Sci. 168:325-341. Tolbert, N. E., R. K. Yamazaki and A. Oeser. 1970. J. Biol. Chem. 245:5129-5136. Tregunna, E. B., G. Krotkov and C. D. Nelson. 1966. Physiol. Plant. 19:723-733. Turano, C., P. Fasella and A. Giartosio. 1962. Biochim. Biophys. Acta. 58:255-261. Vandor, S. L. and N. E. Tolbert. 1970. Biochim. Biophys. Acta. 215:449-455. Vasconcelos, A., M. Pollack, L. R. Mbndiola, H. P. Hoffmann, D. H. Brown and C. A. Price. 1971. Plant Physiol. 47:217-221. Velick, S. F. and J. Vavra. 1962. J. Biol. Chem. 237:2109-2122. Verjee, Z. H. M. 1969. Enzymologia 37:110-118. Verjee, Z. H. M. and D. F. Evered. 1969. Biochim. Biophys. Acta 185:103-110. Vigil, E. L. 1970. J. Cell Biol. 46:435-454. Virtaner, A. I. and T. Laine. 1938. Nature. 141:748-749. 169. 170. 171. 172. 173. 174. 175. 176. 177. 178. 179. 180. 181. 166 Wada, A. and Y. Morino. 1964. Vitamins & Hormones. 22:411-444. Willis, J. E. and H. J. Sallach. 1963. Phytochem. 2:23-28. Wilson, D. G., K. W. King and R. H. Burris. 1954. J. Biol. Chem. 208:863-874. Wong, K. E. and E. A. Cossins. 1969. Phytochem. 8:1327-1338. Yamazaki, R. K. and N. E. Tolbert. 1969. Biochim. Biophys. Acta 178:11-20. Yamazaki, R. K. and N. E. Tolbert. 1970. J. Biol. Chem. 245:5137-5144. Zelitch, I. 1958. J. Biol. Chem. 233:1299-1303. Zelitch, I. 1959. J. Biol. Chem. 234:3077-3081. Zelitch, I. 1964. Ann. Rev. Plant Physiol. 15:121-142. Zelitch, I. 1966. Plant Physiol. 41:1623-1631. Zelitch, I. 1968. Plant Physiol. 43:1829-1837. Zelitch, I. and A. M. Gotto. 1962. Biochem. J. 84:541-546. Zelitch, I. and S. Ochoa. 1953. J. Biol. Chem. 201:707-718. APPENDIX APPENDIX A D. W. Rehfeld, D. D. Randall and N. E. Tolbert. 1970. Enzymes of the Glycolate Pathway in Plants Without C0 -Photorespiration. Can. J. Bot. 48:1219-12 6. 167 1 II: ‘I‘ i 1 ‘li Enzymes of the glycolate pathway in plants without Cog-photorespirationl D. W. REHFELD, D. D. RANDALL, AND N. E. T OLBERT Department of Biochemistry, Michigan State University, East Lansing, Michigan Received October 6, 1969 Dedicated to the memory of the late Dr. G. P. Krotkov and the late Dr. C. D. Nelson REHFELD, D. W., D. D. RANDALL, and N. E. TOLBERT. 1970. Enzymes of the glycolate pathway in plants without COz-photorespiration. Can. J. Bot. 48: 1219—1226. Extracts, mainly from mesophyll cells, were obtained by grinding cells in a Waring Blendor; then extracts of parenchyma sheath cells were obtained by exhaustive grinding of the blender residue in a roller mill or mortar with sand. The specific activities of P-glycolate phosphatase, glycolate oxidase, catalase and reduced nicotinamide adenine dinucleotide- (NADH-) hydroxypyruvate reductase were fourfold higher in extracts of the parenchyma sheath cells than in the mesophyll cells from corn, sugar- cane, and Atriplex rosea. P-Glycerate phosphatase was mainly located in the mesophyll cells. The total activity of glycolate oxidase in plants without COz-photorespiration averaged about one-third that found in other plants on a wet-weight basis. Glycolate oxidase activity in Atriplex rosea, without C02- photorespiration, was about the same as in Atriplex patula, with COz-photorespiration. It is concluded that enzymes for glycolate metabolism are present in all leaves in substantial amounts and are located in both cell types, although a higher specific activity is in the parenchyma sheath cells. Thus it is proposed that photorespiration occurs in all plants, but that C02 evolution from glycolate metabolism is not manifested in plants which have high levels of activity for the C4-dicarboxylic acid cycle of C02 fixation. 1219 Introduction From the extensive studies initiated by Krot- kov, Nelson, and co-workers (7, 8, 23), plants have been classified according to whether they do or do not exhibit photorespiration. These measurements were based upon a light-depend- ent C02 evolution either as a gush when the lights are turned off (7, 8, 23), as l“€02 evolution from newly formed photosynthetic products (9, 26), or upon a C02 compensation point. Photorespiration, as measured by C02 release, may not be manifested in plants with high levels of phosphoenolpyruvate (PEP) carboxy- lase, because they are efficiently refixing the C02 released. Photorespiration appears to occur in these types of plants when assessed by 02 exchange (6, 11), by metabolism of glycolate- 14C and glycine-14C (13, 17, 25), by addition of hydroxysulfonates to show glycolate accumula- tion during photosynthesis (24) and, as shown in this paper, by the presence of the enzymes involved in glycolate metabolism or photorespira- tion. Consequently the term “without photo- respiration” for corn and similar plants is mis- leading, and in this paper it has been modified to “without Gog-photorespiration” to designate the measurement involved. 1Supported in part by NSF grant GB 4154 and pub- lished as journal article No. 4876 of the Michigan Agricultural Experiment Station. Though there are many differences between plants with and without COz-photorespiration, for this paper the pertinent properties common to plants with COz-photorespiration are (a) COz-fixation by the reductive pentose phos- phate pathway, (b) the absence of well-developed parenchyma sheath cells (2, 4), and (c) high levels of enzymes for glycolate metabolism (16, 22). Plants without cog-photorespiration are characterized by (a) initial COz-fixation by the C.,-dicarboxylic acid cycle, (6) presence of well- developed parenchyma sheath cells or green veins, and (c) low levels of glycolate oxidase and ribulosediphosphate (RuDP) carboxylase. The metabolism of glycolate appears to be directly related to photorespiration (9, 15, 24, 25). In photorespiration, light is necessary for photosynthetic glycolate biosynthesis in the chloroplast, 02 uptake occurs during glycolate oxidation to glyoxylate by glycolate oxidase in the peroxisomes (21, 22), and C02 release occurs in the conversion of two glycines to one serine (13). Peroxisomes were isolated from some plants with COz-photorespiration, but low re- coveries of the particles were reported for plants without COz-photorespiration. A similar relation- ship was observed when we were comparing plants upon the basis of glycolate oxidase (16, 22). The magnitude of photorespiration might also be related to the amount of glycolate biosyn- thesis. The only enzyme, so far related to this 1220 process, is a specific P-glycolate phosphatase, and, indeed, lower levels of it were found in some plants without Cog-photorespiration (18). Alternatively, plants without cog-photorespira- tion usually have high levels of a 3-P-glycerate phosphatase, which has similar but not identical properties to the P—glycolate phosphatase (D. D. Randall and N. E. Tolbert, unpublished). Slack and Hatch (20) showed that plants without COz-photorespiration have high levels of PEP carboxylase and low levels of RuDP carboxylase, which is located in the parenchyma sheath cells (19). However, Bjorkman and Gauhl (3) found higher levels of RuDP carboxylase in plants without cog—photorespiration, because the tissue was ground sufficiently to rupture most of the parenchyma sheath cells. Since high levels of glycolate oxidase are present in plants with the reductive pentose phosphate pathway, en- zymes involved in glycolate metabolism also may be located in parenchyma sheath cells of plants without Cog-photorespiration and may have been underestimated because of inadequate grinding. Whereas the Waring Blendor breaks mainly mesophyll cells, grinding by a roller mill or mortar and pestle with sand is necessary to rupture the parenchyma sheath cells. En- zyme assays on extracts from sequential grinding may be used, though not quantitatively, to dif- ferentiate between the activities in the two cell types. We have measured three peroxisomal en- zymes, glycolate oxidase, catalase, and NADH- hydroxypyruvate reductase, involved in glycolate metabolism. Assays on the two phospha- tases,one hydrolyzingP-glycolate and one hydro- lyzing P-glycerate, were run to possibly locate the site of glycolate biosynthesis. These activities are compared with the distribution of RuDP carboxylase and PEP carboxylase between the two types of cells. Materials and Methods Plants and Extraction Plants with COz-photorespiration were spinach, Spin- ach oleracea L., varieties unknown; wheat, Tritieum vulgare L., variety Genesee; and Atriplex patula. Plants without COz-photorespiration were com, Zea mays L., variety Michigan 500 unless otherwise specified; sugar- cane, Saccharum, variety CL 41-223; and Atriplex rosea. Seeds of variegated corn mutants were furnished by Dr. Robert Lambert, Maize Genetic Collection, University of lllinois, Urbana. Dr. Elmer C. Rossman, Crop and Soil Science, Michigan State University, provided the CANADIAN JOURNAL OF BOTANY. VOL. 48. 1970 field corn. Spinach and corn were field grown and the other plants were raised in a greenhouse. Mature leaves were used except for wheat leaves which were harvested when about 12cm high. All leaves were extracted im- mediately after harvest except spinach, which was stored for as long as 1 week at 4°. Leaves were washed with water and blotted dry, and the midrib removed from Spinach, corn, and sugarcane leaves. The remaining steps were performed in a cold room at 4° or in ice baths. Extracts for all plants were obtained by homogenizing 20 g of leaf tissue in 100 ml of grinding medium with a Waring Blendor at high speed for 30 8, except for spinach, which was blended for only 10 s. The grinding medium consisted of 0.02 M glycylglycine, pI-l 7.5. Each homoge- nate was squeezed through eight layers of cheesecloth. The residue remaining in the cheesecloth was then run repeatedly through a serrated roller mill for further crushing until the residue, squeezed on the roller, was nearly colorless. About 100 m1 of grinding medium was used to wash and remove the extract from the rollers of the mill. This mixture was also squeezed through eight layers of cheesecloth. Each extract was adjusted to pH 7.5 and volumes recorded. The procedure for extraction of the Atriplex leaves differed by using 5 g of leaf tissue in 100 ml of grinding medium. After the blender homog- enate was squeezed through cheesecloth, the residue was ground exhaustively with a mortar and pestle with sand and extracted with buffer. Neither whole leaves nor the Waring Blendor residue of A. rosea could be ex- tracted with the roller mill because of the gelatinous nature of these crushed leaves. The roller mill was developed for isolating mito- chondria (12) by John D. Jones (present address: Food Research Institute, Central Experimental Farms, Ottawa, Ontario) to whom we are grateful for a copy of the design. Tissue was crushed between two diagonally ser- rated, finely knurled, stainless steel rollers that are appressed. , Enzyme Assays Glycolate oxidase (EC. 1.1.3.1) was assayed anaerobi- cally at pH 8.7 with the dye, 2,6-dichlorophenolindo- phenol (DCPIP), as electron acceptor (22). All values are corrected to saturating levels of DCPIP (N. E. Tolbert, submitted to Methods of Enzymology), which involves a multiplication factor of 3.95. Because the 0.0. of the dye concentration which saturates the enzyme is too high for the spectrophotometer, Vm must be obtained by extrapolation from a standard plot of enzyme activity versus usable dye concentrations. NAD-hydroxypyruvate reductase (EC. 1.1.1.29) was assayed by following the oxidation of NADH at 340 mu (22). Glyoxylate was used as a substrate because it is more stable and less expensive than hydroxypyruvate. This enzyme from spinach reduces hydroxypyruvate 4 to 5 times faster than glyoxylate and the Km with hydroxy- pyruvate is about lOO-fold less than with glyoxylate. How- ever, values reported in this paper are with glyoxylate and are not corrected to activity with hydroxypyruvate. Triton X-100 was routinely added to both the glycolate oxidase and the reductase assays since it was used in assays on peroxisomes to overcome latency caused by the particle. This use of Triton X-100 may not have REHFELD ET AL: GLYCOLATE PATHWAY been necessary since the present grinding procedures without sucrose should not have preserved many of the peroxisomes. Catalase (EC. 1.11.1.6) was measured spectrophoto- metrically by following the decrease in H202 concentra- tion (21). Triton X-100 was not added in this assay since there is no latency in this assay with peroxisomes. P-Glycolate phosphate (EC. 3.1.3.18) and P-glycerate phosphatase were assayed for 10 min at 30° with 5 micromoles of substrate in 20 mM sodium cacodylate, pH 6.3. For the P-glycolate phosphatase assay, the reaction mixture contained 1 mM MgCl2. Reactions were stopped upon the addition of 10% trichloroacetic acid, the precipitate removed by centrifugation, and the re- leased inorganic phosphate measured (5). Protein was determined by the Lowry procedure (14) and chlorophyll by Amon’s procedure ( 1). Activities are expressed as nanomoles of substrate changed per minute. Specific activities are based on grams fresh weight, milligrams protein, or milligrams chlorOphyll as in- dicated. Results Release of Enzymes from Corn and Sugarcane Leaves Two plants without C02-photorespiration, corn and sugarcane, have been compared with two plants with cog-photorespiration, spinach and wheat. Spinach represents a leaf quickly 1221 homogenized in a Waring Blendor and wheat, like corn and sugarcane, a leaf that is difficult to grind completely with the blender. Wheat plants with cog-photorespiration do not have well- developed parenchyma sheath cells. For spinach and wheat the Waring Blendor solubilized most of each enzyme, which is in mesophyll cells (Table I). The extraction of 3-P-glycerate phos- phatase from wheat leaves was not complete by the blender for reasons unknown. After the leaves of corn and sugarcane were ground for 30 s by the Waring Blendor, about one-third of the total glycolate oxidase was solubilized, about half of the P-glycolate phos- phatase and NADH-hydroxypyruvate reductase, and even more than half of the catalase was re- leased (Table I). Microscopic examination of the residue revealed many intact parenchyma sheath cells attached to vein cells but few intact mesophyll cells. The actual percentage of par- enchyma sheath cells broken by the blender is unknown but the value must be significant. Most of the remaining parenchyma sheath cells were crushed by vigorous grinding in the roller mill and the rest of these enzymes released. The TABLE I Enzyme activities in leaf extracts after differential grinding‘ With C02-photorespiration Without C02-photorespiration Spinach Wheat Corn Sugarcane Blender Mill Blender Mill Blender Mill Blender Mill Glycolate oxidase % total 98 2 100 0 31 69 31 69 S.A., protein 142 47 47 — 8 43 8 20 S.A., chlorophyll 4 657 2192 1335 — 158 2070 142 869 NADH-hydroxypyruvate reductase % total 98 2 97 3 39 61 48 51 S.A., protein 140 70 49 15 10 38 29 48 S.A., chlorophyll 4 586 3111 1362 l 500 218 1801 739 2176 Catalase % total 97 3 97 3 53 47 64 36 S.A. X 10‘3, protein 180 110 180 60 20 40 20 20 S.A. X 10‘3, chlorophyll 6 060 4750 4880 6 210 440 2090 460 700 P-Glycolate phosphatase % total 99 1 92 '8 47 53 45 55 S.A., protein 435 202 336 195 47 148 73 139 S.A., chlorophyll 14 350 9100 9300 19 350 1055 7290 1865 6230 P-Glycerate phosphatase % total 96 4 74 26 72 28 57 44 S.A., protein 210 186 39 99 22 22 122 150 S.A., chlorophyll 6 900 8300 1070 9 750 486 1015 3130 6850 ‘Specific activity (S.A.) is expressed as nmoles min‘1 mg‘l. 1222 percentage distribution of these enzymes between the two homogenates from the sequential grind- ing procedure is indicative of their cellular location, but it is not quantitative. In the dif- ferential grinding procedure, as a first approxi- mation, the enzymes released by the Waring Blendor will be referred to as in mesophyll cells and those released by a subsequent roller mill or mortar and sand extraction will be referred to as in parenchyma sheath cells. After prolonged grinding of up to 2 min on the blender many of the parenchyma sheath cells still remained un- broken as observed microscopically, and the total solubilization of glycolate oxidase or P- glycolate phosphatase was far from complete (Fig. l). The phosphatase for 3-P-glycerate is, however, almost completely solubilized by the Waring Blendor. From these results it appears that the enzymes for glycolate metabolism are located in both cell types, while the 3-P-glycerate phosphatase is located almost exclusively in the mesophyll cells. From similar arguments it has been proposed that PEP carboxylase and the C.,-dicarboxylic acid cycle are located in the mesophyll cells, while RuDP carboxylase and the reductive pentose phosphate pathway for C02 fixation are mainly localized in the paren- chyma sheath cells (3). T l l l l i: g 100 — 3—PGA pnospmns: o l: n 3: :: : E 5 80— ‘ l! g g E / :: 3 ‘f B GLYCOLATE oxmnse o" g 3 5' E 60* ./0/ I: g :3 a. 8 ° 0 " g 5\ B 40_ /|"—. ii 3 § o\° .flG'LYCOLATE ii 2‘ § 20 PHOSPHATASE ll '1 \ :/ :: § 1 l 4L 1' § 1 _ 120" L 10 so so so H ”Roller Mill Waring Blendor (See) Flo. 1. Rate of enzyme release by Waring Blendor. Ditlerent leaf samples of field-grown corn, variety A509 X M81334, were ground in the Waring Blendor for indicated length of time. After homogenization for 120 s, the residue from that sample was further crushed by the roller mill an the percentage of the total enzymatic activity released is indicated on the right side. 3—P- Glycerate (3-PGA) phosphatase, 0; glycolate oxidase. O and open column; P-glycolate phosphatase, I and closed column. CANADIAN JOURNAL OF BOTANY. VOL. 48. 1970 Breakage of the surviving parenchyma sheath cells by the roller mill released glycolate oxidase, catalase, NADH-hydroxypyruvate reductase, and P-glycolate phosphatase of a much higher specific activity on a protein or chlorophyll basis than obtained by the Waring Blendor rupture of the mesophyll cells. These higher specific activ- ities in the roller mill fractions suggest a higher concentration of the enzymes for glycolate me- tabolism in parenchyma sheath cells. The percentage distribution of catalase in the two homogenates did not follow that for glyco- late oxidase (Table I). From sugarcane leaves more catalase was in the Waring Blendor frac- tion than in the roller mill fraction and for corn about an equal distribution was found. On the basis of specific activities the roller mill fraction was still the highest. Also the percentage dis- tribution of NADH-hydroxypyruvate reductase and P-glycolate phosphatase was more nearly equal between two cell fractions than was found for glycolate oxidase. Investigation with Corn Mutants Ten variegated corn mutants were grown in the field, mature leaves from plants before tas- seling were subjected to the differential grinding procedure, and the two homogenates examined for activity of glycolate oxidase and the two phosphatases (Table II). The distribution be- tween the two cell types after differential grind- ing and the total level of enzyme activity for glycolate metabolism varied little from that found with normal green plants. Microscopic examination of the leaves did not show a pre— ponderance of green cells of one type. Thus the use of mutant corn varieties failed to delineate in which cell types these enzymes were located. There is the possibility that glycolate metabolism would occur in chlorophyll-less mutant cells ad- jacent to green cells, and thus the distribution would be the same as found in normal corn. As mentioned above, the distribution of the phos- phatases in the normal plants seemed to be dis- tinctly different between the two cell types. In all of the mutants the distribution of these two phosphatases between the two homogenates also did not vary radically from that found in normal corn plants. It was not possible to find a positive correlation for the minor variations shown in Table II with the type of striation seen visually in the leaf. REHFELD ET AL.: GLYCOLATE PATHWAY Level of Glycolate Oxidase in Plants with and without COz-Photorespiration The specific activities of glycolate oxidase varied several-fold depending upon the type, variety, age, and nutrition of the plant and also upon the grinding procedures. Higher values are obtained with older leaves and after more vigorous grinding. The specific activity values obtained for this oxidase are summarized in Table III. Current values are the sum of the activities of the Waring Blendor extract plus the subsequent roller mill extraction, or they were obtained by grinding whole leaves directly in the roller mill. Both procedures gave similar results. The specific activities from spinach are maxi- mum, and nearly two-fold higher than average values previously reported. This variation can be 1223 attributed to a great deal of variation experienced in the plant material. The higher values for wheat previously reported were due to the use of older tissue. In the present experiments young seed- lings about 12 cm high were used. The level of glycolate oxidase, on the basis of fresh weight, in leaves of plants without C02- photorespiration generally ranged from 20 to 50% and averaged one-third of that found in plants with C02-photorespiration. 0n the basis of protein specific activity, plants without C02- photorespiration had only 10 to 20% of the activity of spinach or wheat leaves. If one com- pares the two Atriplex species, A. rosea without cog—photorespiration contained nearly as much glycolate oxidase as A. patula. However, a pat- tern of lower activity of glycolate oxidase in TABLE II Variation in total (nmoles min‘1 g‘1 wet wt.) and blender-extracted' (% of total) enzyme activities in corn mutants P-Glycerate phosphatase Glycolate oxidase P-Glycolate phosphatase Extracted Extracted Extracted Mutants? Total by blender, % Total by blender, % Total by blender, % Lineate 802 38 2960 45 1020 60 Striate-l 604 24 3070 32 1200 58 Striate-Z (waxy) 585 29 2995 33 1178 57 Striate-2 750 25 3140 31 1405 52 Fine stripe-1 367 29 2940 29 1005 63 Japonica-l 719 35 3520 33 925 60 Iojap 332 22 3010 26 1028 53 ‘After 30 s of Waring Blendor treatment. TThe corn mutants were from the University of Illinois, Maize Genetic Collection. The collection numbers were as follows: lineate. 67-1915-1e; striate-l, 67-771-8/771-2; striate-2 (waxy), 65-346-26) and 66-1795-4/-6; striate-2. 63-3432—66 and 63-3463-7$; fine stripe-1, 66457-70; Japonica-l, 63-3783-76; Iojap, 65-322-2/321-2 and 65-322-7/321-6. TABLE III Total glycolate oxidase activity in plant extracts,‘ nmoles min“1 Currently obtained Previously reported’r g'l ms'1 2‘1 mg'1 fresh wt. protein fresh wt. protein With COZ-photorespiration Spinach 4159 134 2030 83 Wheat 1825 63 3219 115 Atriplex patula 802 12 402i — Without C02-photorespiration Corn (old) 932 16 — —— Corn (young) 237 12 118 4 Sugarcane 561 12 190 12 Atriplex rosea 640 ——- 300i —— Amaranthus — —— ‘All values are corrected to Vm with saturating amount of DCPlP. tValues from Tolbert et al. (21) recalculated as mentioned in first footnote. :0. Bjorkman. personal communication. l 224 CANADIAN JOURNAL OF BOTANY. VOL. 48, 1970 TABLE IV Enzymatic activities from Atriplex varieties of extracts prepared by differential grinding nmoles min-l % mg‘1 protein . P-Glycolate 3-P-Glycerate RuDP PEP Sequence of Glycolate oxndasc phosphatase phosphatase car- car- extraction boxylase, boxylase, procedures A. rosea A. patula A. rosea A. patula A. rosea A. patula A. rosea‘ A. rosea‘ First: 30 s 41 96 31 85 69 96 116 580 Waring Blendor Followed by: 59 4 69 15 31 4 310 2 mortar with sand ‘Values from Biorkman and Gauhl ( 3) 8228: gave specific activity after an unspecifiedtime of Waring Blendor homogenization and after grinding of the cells tn a mortar with glass plants without Cog-photorespiration is gener- ally evident. Nevertheless, the activity of glyco- late oxidase is very substantial in all plants, and the absence of Cog-photorespiration cannot be attributed to very low levels of this enzyme. For corn leaves, the activity of glycolate oxidase after thorough grinding was eight times greater on a weight basis than reported earlier and the specific activity on a protein basis was four times greater. A portion of this difference can be attributed to the previous use of young corn seedlings rather than more mature leaves. How- ever, the use of the roller mill even on young corn leaves extracted twice as much glycolate oxidase as previously reported. In sugarcane leaves the roller mill extraction triples the level of measurable glycolate oxidase activity on a weight basis although on a protein basis no change was observed. Difl'eremial Extraction of Enzymes from Atriplex The solubilization of the enzymes in A. patula, the variety with COz-photorespiration, followed the pattern found for spinach and wheat leaves. Most of the activity was released after a 30-s homogenization in the Waring Blendor (Table IV). In contrast, A. rosea is without COz-res- piration and has well-developed parenchyma sheath cells. Less than half of the glycolate oxidase and P-glycolate phosphatase activity in A. rosea was released by use of the Waring Blendor. Grinding by mortar with sand was necessary to break the parenchyma sheath cells and this procedure released a large portion of these enzymes with a high specific activity. These results were similar to those found by Bjorkman and Gauhl (3) for RuDP carboxylase. It is con- cluded that these enzymes in A. rosea, as in corn and sugarcane leaves, are located mainly in the parenchyma sheath cells. However, a portion of the enzymes associated with glycolate metabo- lism are located in both types of cells, although a lower percentage and a lower specific activity is to be found in the mesophyll cells. Discmdon The differential grinding procedure of a War- ing Blendor to break mesophyll cells and a roller mill or mortar with sand to break parenchyma sheath cells is not quantitative and the results can only be indicative of enzyme distribution between these two cell types. Results are ex- pressed as percentage distribution as well as specific activities on a protein and a chlorophyll basis. Absolute quantitation of the enzyme activ- ities cannot be achieved with any of these units. The observed percentage distribution of enzyme activity is related to the number of each cell type broken under the given treatment. Varia- tion in extractable chlorophyll and protein in the cell types would change the enzyme specific activity, whereas the total amount of enzyme activity in each cell type could be nearly the same. Previous citation to location of the car- boxylases in a particular cell type was based upon protein specific activity (3), but the dis- tribution may not be nearly as complete in one cell type as indicated. By sequential grinding procedures (3) or iso- lation of different types of chloroplasts (19) the concept has developed that the mesophyll cells contain the C4-dicarboxylic acid cycle and that the parenchyma sheath cells contain most of the REHFELD ET AL: GLYCOLATE PATHWAY reductive pentose phosphate cycle for C02- fixation. Since glycolate biosynthesis and metab- olism had always been associated with the re- ductive pentose phosphate cycle in plants with photorespiration, it was expected that the en- zymes for the glycolate pathway would be in the parenchyma sheath cells. Indeed, in corn and sugarcane leaves at least half or more of the activity of these enzymes was in the fraction of cells ground by the roller mill or mortar with sand. However, a substantial part (about one- third to one-half) of the activities were first re- leased upon grinding the mesophyll cells by Waring Blendor. It is tentatively concluded that the glycolate pathway of metabolism associated with photosynthesis is in both cell types. Visually there are several times (three to five) more mesophyll cells than parenchyma sheath cells in corn leaves, yet about two-thirds of the glyco- late oxidase activity was found in the paren- chyma sheath cells. Thus the level of glycolate oxidase per parenchyma sheath cell could be as much as lO-fold greater than in a mesophyll cell. This assumption is consistent with the higher specific activities of these enzymes in the extracts after grinding exhaustively with the roller mill. The distribution of each enzyme activity as- sociated with glycolate metabolism was not the same between the two cell types in corn or ‘ sugarcane leaves (Table I). Three of these en- zymes, glycolate oxidase, NADH-hydroxypyru- vate reductase, and catalase were located to- gether in peroxisomes from spinach leaves (21). However, from corn or sugarcane leaves 31% of the glycolate oxidase was released by grinding the tissues for 303 with the Waring Blendor, whereas 49 or 39% of the reductase and 53 or 63% of the catalase was released. The results suggest the possibility of peroxisomes of dif- ferent enzymatic composition or soluble forms of the enzymes in the two cell types. One objective of this paper was to examine the level of activity of enzymes for glycolate metab- olism in plants without Cog-photorespiration. These values varied widely, but on an average there was about a third as much enzymatic activity in corn leaves on a total fresh weight basis as in spinach or wheat leaves. Nearly the same level of glycolate oxidase was found in the two Atriplex varieties. The absence of C02- photorespiration in plants can not be attributed to very low levels of enzymatic activities asso- ..__.‘-._._ ._ l 225 ciated with glycolate metabolism, but is more likely due to very efficient refixation of the C02. The reduced level of enzyme activity for gly- colate metabolism may well reflect a lower level of overall photorespiration in these plants. Complete grinding of corn and sugarcane leaves is an extremely difficult undertaking. Care- ful and prolonged use of the roller mill for com- plete cell breakage has been used as another procedure besides exhaustive grinding by mortar with sand. The isolation and investigation of peroxisomes from plants without COz-photo- respiration has not yet been accomplished. The drastic procedures necessary to break paren- chyma sheath cells are believed to break also the peroxisomes, which are very fragile. We are thus left with the dilemma of how to break parenchyma sheath cells from this type of plant without also breaking the peroxisomes. 1. ARNON, D. I. 1949. Cu enzymes in isolated chloro- plasts—polyphenoloxidase in Beta vulgaris. Plant Physiol. 24: 1—15. 2. BISALPUTRA, T., W. J. S. DOWNTON, and E. B. TREGUNNA. 1969. The distribution and ultrastructure of chloroplasts in leaves differing in photosynthetic carbon metabolism 1. Wheat, Sorghum and Aristida (Gramineae). Can. J. Bot. 47 : 15-21. 3. BJORKMAN, 0., and E. GAUHL. 1969. Carboxydis- mutase activity in plants with and without B-car- boxylation photosynthesis. Planta. 88: 197—203. 4. DOWNTON, W. J. S., T. BISALPUTRA, and E. B. TREGUNNA. 1969. The distribution and ultrastructure of chloroplasts in leaves differing in photosynthetic carbon metabolism II. Atriplex rosea and Atriplex hastata (Chenopodiaoeae). Can. J. Bot. 47: 915-919. 5. Ftsxa, C. H., and Y. SUBBAROW. 1925. The color- imetric determination of phosphorous. J. Biol. Chem. 66: 375-400. 6. Focx, H., H. SCHAUB, W. HILGENBERG, and K. Eats. 1969. Uber den einfiuss niedriger und hoher 02-partialdrucke auf den Sauerstoff und Kohlendio— xidumsatz von Amaranthus und Phaseolus wahrend der Lichtphase. Planta. 86: 77—83. 7. FORRESTER, M. L., G. Kaorxov, and C. D. NELSON. 1966. Effect of oxygen in photosynthesis, photo- respiration. and respiration in detached leaves I. Soybean. Plant Physiol. 41: 422—427. 8. FORRESTER, M. L., G. KROTKOV, and C. D. NELSON. 1966. Effect of oxygen in photosynthesis, photo- respiration, and respiration in detached leaves II. Corn and other monocotyledons. Plant Physiol. 41: 428—431. 9. GOLDSWORTHY, A. 1966. Experiments on the origin of C02 released by tobacco leaf segments in the light. Phytochemistry, 5: 1013-1019. 10. HATCH, M. D., C. R. SLACK, and H. S. JOHNSON. 1967. Further studies on a new pathway of photo- synthetic carbon dioxide fixation in sugarcane and its occurrence in other plant species. Biochem. J. 102: 417—422. ll.’ JACKSON, W. A., and R. J. VOLK. 1969. Oxygen up- ta7ke by illuminated maize leaves. Nature, 222: 269- 2 1. l 226 12. JONES, J. D., and A. C. HULME. 1961. Preparation of 13. 14. 15. l6. 17. 18. I9. mitochondria from the peel of apples. Nature, 191: 370—372. KISAKI, T., and N. E. TOLBERT. 1970. Glycine as a substrate for photorespiration. Plant Cell Physiol. In press. Lowav, O. H., N. J. ROSEBROUGH, A. L. FARR, and R. J. RANDALL. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193: 265—275. Moss, D. N. 1968. Photorespiration and glycolate metabolism in tobacco leaves. Crop Sci. 8: 71-76. OSMOND, C. B. 1969. B- carboxylation photosynthesis and photorespiration in higher plants. Biochim. Biophys. Acta, 172: 144—149. RABSON, R., N. E. TOLEERT, and P. C. KEARNEY. 1962. Formation of serine and glyceric acid by the glycolate pathway. Arch. Biochem. Biophys. 98: 154- 163. RANDALL, D. D., and N. E. TOLBERT. 1968. Sugar- cane 3-phosphoglycerate phosphatase. Plant Physiol. 43: S—39. SLACK, C. R. 1969. Localization of certain photo- synthetic enzymes in mesophyll and parenchyma sheath chloroplasts of maize and Amaranthuspalmeri. Phytochemistry, 8: 1387—1 391. 20. 23. 24. 25. 26. CANADIAN JOURNAL OF BOTANY. VOL. 48, I970 SLACK, C. R., and M. D. HATCH. 1967. Compar- ative studies on the activity of carboxylases and other enzymes in relation to the new pathway of photo- synthetic carbon dioxide fixation in tropical grasses. Biochem. J. 103: 660—665. . TOLBERT, N. E., A. OESER, T. KISAKI, R. H. HAGB- MAN, and R. K. YAMAZAKI. 1968. Peroxisomes from spinach leaves containing enzymes related to glyco- late metabolism. J. Biol. Chem. 243: 5179-5184. . TOLEERr, N. E., A. OESER, R. K. YAMAZAKI, R. H. HAGEMAN, and T. KISAKI. 1969. A survey of plants for leaf peroxisomes. Plant Physiol. 44: 135447. TREGUNNA, E. B., G. Kaorxov, and C. D. NELSON. 1966. Effect of oxygen on the rates of photorespira- tion in detached tobacco leaves. Physiol. Plant. 19: 723—733. ZEerCH, I. 1958. The role of glycolic acid oxidase in the respiration of leaves. J. Biol. Chem. 233: 1299- 1303. ZELITCH, I. 1966. Increased rate of net photosyn- thetic carbon dioxide uptake caused by the inhibition of glycolate oxidase. Plant Physiol. 41: 1623-1631. ZEUTCH, I. 1968. Investigation on photorespiration wit3h] a sensitive l4C-assay. Plant Physiol. 43: 1829— 18 .