«I; ~' 1:53. LIBRARY Michigan State Uilivc11.ity This is to certify that the thesis entitled EVIDENCE FOR THE FORMATION OF PYRIDOXIMINE-S-PHOSPHATE BY THE BIODEGRADATIVE L-THREONINE DEHYDRASE OF ESCHERICHIA COLI presented by John Frank Riebow has been accepted towards fulfillment of the requirements for M.S. degree in Blochemlstry / ”J M / flfaflyjk 1/, Major professor Q . Date ///é—r 22/773 M 07639 EVIDENCE FOR THE FORMATION OF PYRIDOXIMINE-S-PHOSPHATE BY THE BIODEGRADATIVE L-THREONINE DEHYDRASE OF ESCHERICHIA COLI BY John Frank Riebow A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Biochemistry 1978 ABSTRACT EVIDENCE FOR THE FORMATION OF PYRIDOXIMINE-S-PHOSPHATE BY THE BIODEGRADATIVE L-THREONINE DEHYDRASE OF ESCHERICHIA COLI BY John Frank Riebow Until recently, the a,8-elimination reactions of various amino acids, which are catalyzed by pyridoxal phosphate dependent enzymes, were considered to be irreversible. Snell and Watanabe have shown that the tryptophanase catalyzed a,B-elimination reaction is reversible (Watanabe, I., and Snell, E. E., Pro. Nat. Acad. Sci. USA 69, 1086 (1972). In addition, it was established that the mechanism by which the reverse reaction proceeded was different from the mechanism for the forward reaction. One intermediate for the reverse reaction was established as pyridoximine phosphate. This research project was undertaken to determine if the reversibility of the a,B-elimination reaction catalyzed by tryptophanase is unique to this enzyme; or, if the bio- degradative L-threonine dehydrase from E. coli, another pyridoxal phosphate dependent enzyme, would also catalyze the reverse of its reaction by a mechanism similar to that for tryptophanase. John Frank Riebow When threonine dehydrase was incubated in the presence of high levels of ammonium chloride, an intermediate was formed which has been indirectly shown to be pyridoximine phosphate. To my parents ii ACKNOWLEDGEMENTS The author wishes to extend his sincere appreciation to Dr. W. A. Wood for his encouragement and support through- out the course of this research. The many discussions and friendship of Mr. David Le Blond are also recognized. I would like to extend a special thanks to Mrs. Jane Fortman for her expertise in the typing of this manuscript. Finally, I would like to acknowledge some very dear friends of mine whose Christian fellowship and encouragement is forever appreciated: Greg Allen, Dick and Beverly Anderson, Gilbert and Jane Apps, Doris Bauer, Charley and Lydia Brown, Martha Burley, Ed and Pat Comandella, Helen Dorriell, Sheila Fairman, Al Getzen, Sandy Getzen, Jeff and Sally Harrold, Dave and Peggy Mingus, Herman and Henrietta Prether, Eric and Cheryl Satterlee, Don and Dorothy Snider, Minnie and Kim Worman. TABLE OF CONTENTS page INTRODUCTION........ ............................. ....... 1 LITERATURE REVIEW....................................... 3 Discovery of the Role of Pyridoxal Phosphate in Metabolism............. .......... ......... ....... 3 Pyridoxal Catalyzed Nonenzymatic Reactions............ 7 A Mechanism for Pyridoxal Catalyzed a,B-Elimination Reactions.............. ....... ....... ......... ...... l4 METHODS AND MATERIALS........ ...... ..................... 25 Bacteriological ..... . ......................... . ....... 25 Chemicals................ ....................... . ..... 25 Determinations and Procedures ......................... 26 Protein Determinations.... ....... ' .................... 26 Activity Assay for Threonine Dehydrase...... ...... .. 27 Activity Assay for Glutamate-Oxalacetate Transaminase ...................................... 28 Resolution of Glutamate-Oxalacetate Transaminase.... 33 Activation of apo-Glutamate-Oxalacetate Transaminase with Pyridoxamine-S-Phosphate........ 33 Removal of Ammonia and AMP from Threonine Dehydrase. 36 Purification of Threonine Dehydrase ................... 36 Crude Extract........... ...... . ..................... 43 AMP-Sepharose Affinity Chromatography ............... 43 Sephadex G-200 Column Chromatography ................ 44 iv page RESULTS.. ..... .............. ............................ 49 Evidence for the Enzymatic Formation of Pyridoximine Phosphate ....... .. ............. .. ...... 49 Effect of Ammonia Upon the Absorption Spectrum of Threonine Dehydrase... .............. .... 53 Absorption Spectra .................................. 54 Difference Spectra .................................. 54 DISCUSSION ...... ........ ................. .. ............. 62 BIBLIOGRAPHY ................... ...... .............. ..... 65 NO. 2 LIST OF TABLES Pyridoxal phosphate-dependent enzymes which catalyze a,8-elimination reactions of various amino aCids.I...O.IOOOOOOOOOOOOOO00.000.000.00... Formation of pyridoximine phosphate from threonine dehydrase and ammonia............................ vi page 18 52 LIST OF FIGURES No. page 1 Vitamin BS structures ............... ... ........... 5 2 The structure of salicylaldehyde derivatives related to pyridoxal........................... 10 3 Mechanism for the nonenzymatic a,B-elimination reaction of Metzler et a1. (18) of amino acids as catalyzed by pyridoxal ....... . ........ l6 4 Generalized mechanism for the a,8-elimination reaction of amino acids as catalyzed by pyridoxal phosphate-dependent enzymes .......... 20 5 Standard curves for the Lowry (H) and the fluorescamine (craa) assays for protein........ 30 6 Activity assay for glutamate-oxalacetate transaminase................ ................... 3l 7 Linearity of the glutamate-oxalacetate transaminase assay. ....... ........ ............. 32 8 Resolution of glutamate-oxalacetate transaminase (H) and the ability of pyridoxamine-S- phosphate to reactivate resolved transaminase ‘(o—O)oooo ooooooo coo-000.00.00.00. ...... 0.00000 35 9 Activation of apo-glutamate-oxalacetate trans- aminase by pyridoxamine-S-phosphate (H) . Also included is a Lineweaver Burk plot (CF[enzyme-PLP-intermediate]3 [enzyme-PLP-a,B-unsaturated 1 amino acid] PLP-enzyme + d,B-unsaturated amino acid . l a-keto acid + NH ¢> a-imino acid 3 H20. Experimental evidence in support of this mechanism has been provided from both isotope incorporation and spectral experi- ments. Snell and his associates have recently demonstrated that the a,B-elimination reaction, as catalyzed by tryptophanase, proceeds by way of a mechanism different from the one shown above in the terminal steps of the reaction. This new alternative pathway involves the tautomerism of the a-aminoacrylate pyridoxal phosphate Schiff base (enamine) to the corresponding ketamine, which is then hydrated to the carbinolamine. As a result of an elimination, free pyruvate is formed along with enzyme bound pyridoximine phosphate. Hydrolysis of the latter enzyme intermediate results in the regeneration of the pyridoxal phosphate form of the enzyme with the release of ammonia. As a result of this work by Snell's group, there is now a need to deter- mine if this new alternative mechanism is unique to trypto- phanase, or if there are other pyridoxal phosphate enzymes which function by a similar mechanism. The work, presented here, is an attempt to establish whether or not an appreciable amount of enzyme bound pyridoximine phosphate is formed as a result of the inter- action between high levels of ammonia and threonine dehydrase. The enzyme directed formation of pyridoximine phosphate, under the above conditions, would be evidence in support of the mechanism proposed by Snell, and that this mechanism is not unique to tryptophanase. LITERATURE REVIEW Since the discovery of vitamin B during the 1930's, 6 a great deal of research effort has been directed toward establishing its involvement in metabolism and elucidating its function at the molecular level. Several detailed reviews on various aspects of vitamin B have appeared 6 over the past twenty years (1-4). These reviews examine in detail, the many reaction types in which vitamin B6 plays a role in catalysis. The following general discussion is intended to summarize one of the classes of reactions for which vitamin B6 is known to play a functional role, the a,B-elimination reaction. Discovery of the Role of Pyridoxal Phosphate in Metabolism The involvement of vitamin 36 (see figure 1) in protein metabolism became evident during the mid 1940's. Earlier work by Gyorgy had demonstrated that a nutritional factor, termed vitamin 36' would cure a specific dermatitis of young rats which were fed a vitamin-deficient diet. In 1936, Birch and Gyorgy partially characterized vitamin B6 chemically. Crystalline pyridoxol from various natural sources was prepared in several laboratories (5-7) during 1938, and was subsequently synthesized in other laboratories (8). .Hoxocflumm nufl3 wanmmcmnonmucfl com: me mcflxocfluhm Emmy can .waucmuuso .mm :wEmufi> £ua3 hanmmcmnoumucfl coma coon .ummm may Ca .mmn mcfixocfiuhm Bump was .om cflamufl> mo mEHOm msoflum> on“ mum mum: cmumflq .mmusuosnum mm awfimufl> .H Gunman H mhsmwm Euzamoi -nsoEonoEtE 3238an -n 1882;; 2/ o»: 2/ on: Nznoaowzo \ oz- «Imoaofo \ or £260 96 $55302}. .3825 3.82:5 z/ o»: z/ o»: z/ o»: EEOC? EEOC? :ofo \ oz N:zNIo o: IONS Following the isolation of crystalline pyridoxol, several nutritional studies during the early 1940's established a requirement for pyridoxol in protein metabo- lism. Experiments involving vitamin B6-deficient animals revealed that protein metabolism was impaired under these conditions. It was also shown that high protein diets caused an increase in the rate of appearance of vitamin deficiency in these animals. The toxic effects which are caused by high intake of certain amino acids (serine, glycine, and methionine in particular) in vitamin BG-deficient animals could be controlled by administering pyridoxol to these animals. In addition to these mammalian systems, the biosynthesis of alanine, threonine, lysine, and other amino acids was shown to be inhibited in certain strains of bacteria (lactic acid bacteria and CZostridium welchii) when grown on vitamin 36-deficient media [see (2) and references therein]. During this same period of time, Snell (9) established that other forms related to pyridoxol, which were obtainable from natural sources, were much more effective in stimulating the growth of lactic acid bacteria. Subsequently, Snell demonstrated that these pyridoxol derivatives were pyridoxal and pyridoxamine. In addition, it was found that pyridoxal and pyridoxamine account for the majority of the vitamin 36 which is found in natural substances. Additional work by Snell (10) demonstrated that pyridoxamine could be formed by a nonenzymatic reaction between pyridoxal and various amino acids. This led to the prediction that the interconversion between pyridoxal and pyridoxamine might occur in viva during transamination between various amino acids as is shown below. amino acid + pyridoxal keto acid 1 keto acidl + pyridoxamine + pyridoxamine 2 . . amino acid2 + pyridoxal ‘ amino acid + keto acid amino acid + keto acid 2 2 l 1 Concurrent work in other laboratories on various enzymes associated with amino acid metabolism was beginning to suggest a functional role for vitamin BG. Various studies on bacterial systems had revealed that vitamin B6 was necessary for the decarboxylation of various amino acids (11). In addition, it was demonstrated that the actual form of vitamin B required for decarboxylation 6 was pyridoxal. Further work on purified apo-enzymes of amino acid metabolism, in particular enzymes involved in transamination reactions, provided direct evidence that pyridoxal-S-phosphate was the active form of the coenzyme (12-14). Pyridoxal Catalyzed Nonenzymatic Reactions As mentioned above, pyridoxal will catalyze nonenzymatic transamination reactions. Metzler and Snell were able to take advantage of this system to evaluate the importance of each substituent of pyridoxal towards the rate of ketoglutarate formation from glutamate [see (1) and references therein]. Figure 2 gives some of the model compounds which Metzler and Snell employed in their studies. The results of these studies indicate that the hetero- cyclic nitrogen atom, the phenolic hydroxyl group at position 3, and the formyl group at position 4 are all essential for nonenzymatic catalysis. The methyl group at position 2 may contribute to the catalytic efficiency of pyridoxal, through inductive effects, but its presence is not an absolute requirement for catalysis. The hydroxy- methyl group at position 5 is not required for nonenzymatic catalysis; although, it is essential for the functioning of pyridoxal biologically since it is this group which is phospharylated to give pyridoxal-S-phosphate. While evaluating the importance of the hydroxymethyl group, Metzler et a1. (15) found that either the absence of this group, or its phosphorylated form, was actually more efficient catalytically than was pyridoxal. They suggested that these results indicated that, under the experimental conditions used, the aldehyde group of pyridoxal is in equilibrium with its hemiacetal form. Hemiacetal formation, in effect, lowers the concentration of free aldehyde necessary for catalysis. OH ./ H—C= H"C\o /// HO CHZOH ‘5‘ HO -7 CH2 \ _. 33c H3C '\\\ Figure 2. The structure of salicylaldehyde derivatives related to pyridoxol. The various derivatives of salicylaldehyde shown were used by Metzler and Snell to establish the structure-function relationship of the various substituents of pyridoxal in nonenzymatic catalysis. 10 HO Salicylaldehyde HO _ , , 4-Nltrosollcylaldehyde H0 N02 . . . I e-Nltrosollcylaldehyde HO I OZN 3-Nltrosohcyloldehyde l - HO . OZN N02 3,5-Dmmosollcylaldehyde Figure 2 11 It has been known for some time that reactions between amines and aldehydes lead to the formation of imine (Schiff's base) compounds. By analogy, the yellow color, which appears when various amino acids are mixed with pyridoxal (or pyridoxal phosphate), can be attributed to imine for- mation between the amino group of the amino acid and the free aldehyde group of pyridoxal. These imine compounds have a characteristic absorption maximum in the region from about 340 to 440 nm. The exact location of the absorption maximum is dependent upon the particular amino acid and the pH of the solution. Experimental evidence obtained by Metzler (16) and Christensen (17) demonstrated that, under physiological conditions, a significant amount of pyridoximines are formed during the reaction between pyridoxal and various amino acids. The existence of such pyridoximines had been postulated by several investigators prior to the above publications. As a result of imine formation between an amino acid and pyridoxal, a structure is formed in which there is a strong withdrawal of electrons from the a-carbon atom of 12 the amino acid toward the heterocyclic nitrogen atom; thus weakening the various bonds to the a-carbon atom. Support for this concept of charge delocalization, as a result of imine formation, was provided by the demon- stration that 4-nitrosalicylaldehyde and 6-nitrosalicyl- aldehyde were effective catalysts for nonenzymatic trans- amination. For these compounds, electron withdrawal from the a-carbon atom (for the corresponding imines to the ones shown above), through resonance structures, can be demon- strated. The compounds 5-nitrosalicylaldehyde and 3,5- dinitrosalicylaldehyde will not catalyze nonenzymatic transamination of amino acids. For these compounds, the position of the nitro group(s) does not allow for electron delocalization through resonance structures. These results indicated that the positioning of the strongly electro- negative nitro group dictated the catalytic efficiency of various nitrosalicylaldehyde compounds (see figure 2 for structures). The phenolic group of pyridoxal is important due to the fact that, through hydrogen bonding, it can stabilize the imine which is formed between an amino acid and the aldehyde group of pyridoxal. It has been known for some time that the rate of nonenzymatic transamination of amino acids, catalyzed by pyridoxal, was greatly enhanced by various metal ions. This rate enhancement has been ex- plained in terms of chelated ring structures which involve the phenolic group of pyridoxal. The chelated ring 13 structures add stability to the imine structure and also force this entire system of double bonds into a planar T T O “rm—‘36 H N~ ,0 C/ \D?’ (5 HOCH2 CH3 N configuration that allows for maximal electron withdrawal from the a-carbon atom by the heterocyclic nitrogen atom. These various studies on the nonenzymatic transamina- tion by pyridoxal, and structurally related compounds, allowed Metzler et al. (18) to propose a general mechanism for pyridoxal catalysis in 1954. This mechanism could account for all of the known reactions of amino acids in which pyridoxal (or pyridoxal phosphate) was a catalyst. These reactions included racemization, transamination, decarboxylation, a,8-e1imination, B,y-elimination, and B-substitution reactions. All of these reactions proceed through a common imine intermediate and are subsequently the result of specific bond cleavage of the imine inter- mediate. Just prior to this time, Braunstein and Shamyakin (19) proposed a very similar mechanism for enzymatic reactions in which pyridoxal phosphate is a coenzyme. Their mechanism was based on theoretical considerations only. 14 A Mechanism for Pyridoxal Catalyzed a,B-Elimination Reactions In the following discussion, only the mechanism for a,8-elimination will be considered. This mechanism is shown in figure 3 for a nonenzyme catalyzed reaction. Initially pyridoxal and a-amino acid combine to form an imine structure (structure I in figure 3), which is stabilized by the presence of a metal ion (M). As a result of the strong withdrawal of electrons from the a-carbon of the amino acid toward the heterocyclic nitrogen atom, the proton on the a-carbon atom dissociates to give structure II (figure 3). Rearrangement of this conjugated double bond system results in the release of the R‘ group, as an anion, from the B-carbon atom of the amino acid. The resulting a,B-unsaturated imine (structure III in figure 3) then decomposes, in aqueous media, through an a,B-unsaturated amino acid, to yield the corresponding B-keto acid and free ammonia. Several pyridoxal phosphate-dependent enzymes are known which catalyze a,B-elimination reactions of amino acids by a similar mechanism (19). There are some impor- tant differences between nonenzymatic and the correspon- ding enzymatic reactions. Enzymatic reactions, for which pyridoxal phosphate is a coenzyme, proceed at a rate much greater than for the corresponding nonenzymatic reaction. The enzymatic reaction is highly specific, each enzyme catalyzes one reaction. In nonenzymatic systems, several reactions occur simultaneously with 15 .meocfluwm an cmnaamumo mm mcfiom OCHEM mo Away .Nc um “mango: mo cowuommu GOwumcfleflamlm.c oeumewucmcoc can MOM Emflcmsomz .m Guzman 16 m musmflm mIz Iz 1.1 M .. a .3. a @ooo-w-w-m cocoonmé lll. oooo-o.o-m O I E .10 _2/ m . I Y 2.2 (0-: / _ O. 2 I n 2 . . . /, :o o .... D Hoe . z _ _ foo: Os :0 _ l ..o :o \ £8: .2 1: Aw :Z\.I mBV\\\ 0:0 _ __ ”w H )«IZ _.._ cs0 .. 0! _Im . z zoooeei O \ NIUOI .2. to]: o. .z I w..a-w-m Os _ _ 17 little specificity. This substrate specificity and rate enhancement, present in enzymatic systems, can only be attributed to the protein moiety. Table I lists some enzymes which catalyze a,B-elimination reactions. The currently accepted mechanism for pyridoxal phosphate- mediated a,B-elimination reactions is shown in figure 4 (structures I, II, III, IV, V, and VI). The individual steps of this mechanism will be discussed in terms of the experimental evidence which supports their involvement. Recently, Snell and his associates (20) presented evidence for an alternative mechanism for the terminal steps of the reaction. This alternative mechanism is also shown in figure 4 (structures I, II, III, IV, V, VII, and VIII). For all enzymes, in which pyridoxal phosphate is a coenzyme, the pyridoxal phosphate group is covalently bound, as an imine, to a lysyl residue of the enzyme (structure I in figure 4). This imine can be reduced with sodium borohydride and the resulting N6-pyridoxyl- lysine residue can be isolated following proteolytic diges- tion of the enzyme (21). In addition, as a result of the imine, all pyridoxal phosphate-dependent enzymes exhibit a characteristic absorption spectra in the region of 410 to 430 nm (21-25). The appearance of an absorption peak in this region is consistent with the theoretically predicted location for the imine form of pyridoxal phosphate. Structure II represents the binding of substrate to the enzyme (formation of a Michaelis complex). In general, 18 Table l. Pyridoxal phosphate-dependent enzymes which catalyze a,B-elimination reactions of various amino acids. enzyme R __£E__. L-Threonine Dehydrase CH3 0H D-Serine Dehydrase H OH Tryptophanase H N indole B-Tyrosinase H OH phenol Cystiene Sulfhydrase H SH 19 .HHH> can .HH> .> .>H .HHH .HH .H mmusuosuum .Aomv Hamcm can mmmcmumz cmmomonm coon no: mmmcmsmoumhnu How >m3zumm m>HumcumuHm G< .H> can .> .>H .HHH .HH .H mousuosuum an czosm ma Emflcmsoms acumcmm cmumooom haucmuuso one .mmEhnsm unmocmmmclmumnmmonm meocfluhm ma Umumamumo mm modem ocean mo cowuommu coflumawsflamlm.d on» How Emwcmsome coufiamumcmw .v musmflm ’1 (I ”nunuuuuu, ull mmmnu,’ “HHIHHuuu Illlllllllllt; ‘5‘ {I \iltlllllctollr, I \ f .0: g“ I C) H -c- R' Hz-C-H coon N 20 \\\lllHllll!ll/ \ \“HIIHIHIHI/ \ (I “nunnnuuu I I “unuunuu I \ummu \ \ l 2:: f“ 2-2 9 \\|||lll|'g..."l L H “HHHHIHH \\ton llllllllll‘l \ (,_.L_. Lvs *dH H8 HC)’, Hm I \I 'I'l'l,’ \ CHZO ® ‘\\ \EF 4 I CH3\N m_I ’5 +239 \a /Z I L I I 8 0 <3 ~..__ 8 0 I C“- U I fifl-‘ZZ (;)=:2: I :z N I 9 o m I a: I C) CD 7 9 CD1=CJ l 6! I 9 a: + M v I 2! Figure 4 21 the formation of this type of a complex can be considered a necessary prerequisite for catalysis. Evidence that a Schiff's base is formed between the enzyme bound pyridoxal phosphate moiety and the substrate for the enzyme (structure III) comes from absorption and circular dichroism spectra for various pyridoxal phosphate enzymes in the presence of their respective substrates (4). The use of various inhibitors (both competitive and noncompetitive) has been quite helpful in these types of studies. Evidence for structure IV is largely the result of spectral observations of pyridoxal phosphate enzymes in the presence of their substrates. When the enzyme is incubated with its substrate, there is a shift in the absorption maximum from around 410 to 430 nm. to near 500 nm. (the exact location of these peaks is dependent upon the particular enzyme). The position and the shape of the latter absorption maximum is more characteristic of quinonoid structures (structure IV in figure 4) than for aromatic structures. These longer wavelength peaks are transient and disappear as the substrate is consumed in the reaction (see reference 4). The existence of structure V has been established on the basis of an observed absorption difference peak in the region of 450 to 470 nm. Structures such as V, are theoretically predicted to absorb in this vicinity. Also, these peaks are dependent upon the presence of substrate and disappear as the substrate is removed. 22 The conversion of structure V to structure I with the corresponding enamine remaining bound to the enzyme is supported by the work of Flavin and Slaughter (26). These investigators were able to demonstrate that the addition of N-ethylmaleimide to reactive enamine inter- mediates, similar to structure V, results in the stereo- specific formation of the product a-keto-B-[3'-(N-ethyl- 2',3'-dioxopyrrolidyl)] butyrate. When the threonine dehydrase reaction is run in deuterium oxide (D20), deuterium is placed on the B-carbon atom of the product of a-keto- butyrate. In addition, if sodium borohydride is added to the reaction mixture, D,L-a-aminobutyrate can be isolated. These two experiments taken together indicate that free a-aminocrotonate is released from the enzyme and that it is spontaneously converted to a a-ketobutyrate and ammonia via a-iminobutyrate Wood and Phillips (21) provided some additional infor- mation concerning this mechanism. From isotope incorpor- ation experiments, it was demonstrated that deuterium and 180 from the aqueous medium are back incorporated into threonine by threonine dehydrase. These results indicated that the reactions leading from structure II through structure V are freely reversible and that, at least for threonine dehydrase, the rate limiting step in the mechanism occurred somewhere after structure V. As mentioned above, Snell and his associates (20) have established an alternative pathway for the terminal steps 23 of the mechanism shown in figure 4 (structures I, II, III, IV, V, VII, and VIII). This alternative mechanism became evident from experi- mental data for the enzymatic synthesis of tryptophan from indole, pyruvate, and ammonia as catalyzed by tryptophanase. From a series of kinetic experiments, Watana-e and Snell (20) were able to develop a kinetic mechanism for the reverse of the tryptophanase reaction. This mechanism was shown to be consistent with the ordered Ter-Uni mechanism of Cleland (27). This mechanism indicated that pyruvate was the second substrate to bind to the enzyme with indole or ammonia being the first substrate to bind. It was not possible, from the kinetic data alone, to demonstrate whether indole or ammonia was the first substrate to bind to the enzyme. Holotryptophanase has a characteristic absorption maximum at 337 nm. In the presence of l M ammonium chloride, this peak is intensified considerably and is shifted to near 420 nm. Free pyridoxal phosphate also shows a similar red shift in the presence of high levels of ammonia (from about 385 to 405 nm.)(20). Reduction of the incubation mixture of pyridoxal phosphate and ammonium chloride, with sodium borohydride, resulted in the formation of pyridoxamine phosphate. From these spectral studies, Snell and his associates concluded that the shift in the absorption maximum of tryptophanase to 420 nm., in the presence of ammonium chloride, was the result of the formation of an enzyme-bound pyridoximine phosphate complex. The results 24 of these spectral studies, along with the kinetic experi- ments allowed Snell's group to conclude that in the for- mation of tryptophan, by tryptophanase, ammonia was the first substrate to bind to the enzyme, then pyruvate, and finally indole. METHODS AND MATERIALS Bacteriological An isoleucine requiring mutant of Escherichia coli (ATCC 9739, designated LA-9) was maintained on a nutrient agar slant at 4°C. Large quantities of the organism were prepared by transferring cells from the nutrient slant into 10 ml. of nutrient broth. The organism was allowed to grow for 12 hours at 37°C., with mechanical shaking. The incubating solution was then transferred to a flask con- taining 1 Z. of complex medium [2% (w/v) N-Z amine NAK (an enzymatic digest of casein), 1% (w/v) yeast extract, and 0.5% (w/v) potassium phosphate (dibasic)]. The or- ganism was allowed to grow at 37°C. for 12 hours (without shaking). Following this, the organism was transferred to 100 f. of the same complex medium and were grown for 4 hours at 37°C. The cells were then harvested and stored at -20°C. until needed for preparation of crude extract. Chemicals Glutamate-oxalacetate transaminase (holoenzyme, GOT) and malate dehydrogenase were purchased from Boehinger Mannheim Biochemicals. Lactate dehydrogenase (beef heart, 25 26 type III), D,L-dithiothreitol, pyridoxol-S-phosphate, pyridoxamine-S-phosphate, and adenosine—5'-monophosphate were purchased from Sigma Chemical Company. Nicotinamide adenosine dinucleotide reduced disodium salt (NADH) was obtained from P-L Biochemicals, Inc. L-Glutamic acid and a-ketoglutarate sodium salt (both A grade) were purchased from Calbiochem. L-Aspartic acid and yeast extract were obtained from General Biochemicals. N-Z Amine NAK was obtained from Sheffield Chemical Co. Nutrient agar and nutrient broth were purchased from Difco Laboratories. All other chemicals were purchased from commercial sup— pliers. Reagent or analytical grades were always used. Determinations and Procedures Protein Determinations In most instances, protein concentration was estimated by the method of Lowry et al. (28). The protein samples (approximately 100 ug) and bovine serum albumin standards (20 to 200 ug) were diluted to 0.5 ml. with water. These samples were then further diluted by the addition of 5.0 m1. of a solution containing 0.01% (w/v) copper sulfate, 0.02% (w/v) sodium potassium tartrate, and 3% (w/v) sodium carbonate in 0.1 N sodium hydroxide. After standing for 20 minutes at room temperature, 0.5 ml. of a 1 N phenol solution was added to each sample. These samples were incubated for 10 minutes at room temperature and then the absorbance at 750 nm. was measured for each sample. This method was always used to quantitate protein levels. 27 The fluorescamine assay for proteins (29) was used to monitor protein present in column effluents. Under these conditions, protein bands from columns were determined in a relative manner and no attempt was made to quantitate the protein. Aliquots (50 ufl) from various column frac- tions were diluted with 1.25 ml. of sodium borate buffer (0.2 M, pH 9.0). Fluorescamine reagent (0.35 ml. of 0.30 mg/ml. fluorescamine in acetonitrile) was added to each sample with concurrent mixing with a Vortex mixer. The sample with the greatest fluorescence was used to adjust the fluorimeter to 90% of full scale deflection and the fluorescence of all other samples were measured relative to this most fluorescent sample. Figure 5 shows a typical standard curve for the Lowry protein assay and demonstrates that the fluorescamine assay is linear with respect to protein concentration. Activity AssayAfor Threonine Dehydrase Threonine dehydrase activity was measured by the lactate dehydrogenase-coupled spectrophotometric method of Dunne et a1. (30). All assays were performed in the presence of 75 mM potassium phosphate buffer, pH 8.0, 20 mM L-threonine, 5 mM dithiothreitol, 5 mM AMP, 0.4 mM NADH, and 26 units of lactate dehydrogenase as the coupling enzyme. The final volume of the assay was 0.20 ml. 28 L-Threonine =:a-Ketoglutarate + ammonia threonine dehydrase + NADH + H -\\ lactate dehydrogenase NAD+ (j .9 a-Hydroxybutyrate Activity Assay for Glutamate-Oxalacetate Transaminase The enzymatic activity of glutamate-oxalacetate trans- aminase was measured spectrophotometrically using an enzyme- coupled assay as shown in figure 6 (31). The velocity of the transaminase was calculated from the continuous decrease in absorbance at 340 nm. as a result of NADH utilization during the course of the assay. All assays were performed in quartz microcuvettes (0.5 ml. capacity) with a 1.0 cm. light path. A beckman Du monochromator equipped with a Gilford Model 2000 Automatic Recording Spectrophotometer was used for measuring the change in absorbance with time. All enzymatic assays were performed at 28°C. Each cuvette contained 92 mM potassium phosphate buffer (pH 7.4), 184 mM aspartate, 20.2 mM a-ketoglutarate, 0.51 mM NADH, and 1.9 units of malate dehydrogenase. The reaction was initiated by the addition of 5 pi. of the transaminase. The final volume in the cuvette was 0.25 ml. Figure 7 demonstrates that, under the conditions used, the assay is linear with respect to the transaminase concen- tration. 29 .cwmuoum How mammmm :YQ mcflsmommuosam 93 can AIL C3304 may you mm>uso pumps—mum .m musmwm 30 eoueoselom] amulet] LO 0 L0 IGO I20 80 N to N O T l l l l 4 “3. S". N O O O Luuog/j eouoqiosqv Protein,ug Figure 5 31 mmmcflfimmcmuu mumumomameImumEmusHm How momma hufl>fluo¢ .o musmflm ESE nzma me<3Hommu mum>auommu ou mumnmmonmlmlwcHmeocfiuhm mo xuflaflnm on» cam. ACOV mmmcasmmcmuu mumumomameImumEmusam mo counuaaommm .m muzmwm Percent Full Activity 35 lOO IL SO l\) O"! l O I Number of Ammonium Sulfate Precipitations Nt- OJ ' ' Figure 8 36 was then measured, as described earlier. The ability of pyridoxamine-S-phosphate to activate apo-GOT is shown in figures 9 and 10. Under these activation conditions, it was possible to reactivate apo-GOT to 100% of its original activity. The Lineweaver-Burke plot, shown in figure 9, was used to determine the Ka of apo-GOT for pyridoxamine phosphate. Under the activation conditions described above, a Ka value of approximately 20 uM was obtained for pyri- doxamine phosphate. A Ka value of 1.0 uM has been reported by O'Kane and Gunsalus (33). Removal of Ammonia and AMP from Threonine Dehydrase Ammonia and AMP were removed from threonine dehydrase by column chromatography with Sephadex G-25 (fine) according to the method of Rabinowitz (34). The column (20 x 0.7 cm.) was poured and equilibrated with potassium phosphate buffer (0.1 M, pH 8.0) containing 1 mM dithiothreitol. The enzyme sample (0.25 ml.) was passed through the column with the same buffer. Figure 11 shows a typical elution profile for this column. Purification of Threonine Dehydrase A mutant strain of Escherichia coli (LA-9) deficient in biosynthetic L-threonine dehydrase, was grown as described in the methods section. Biodegradative L-threonine dehydrase was purified from this strain of E. coZi according to an unpublished method of Le Blond and Wood (35). 37 .mumc mEmm wnu mo AOIOV ...on mxncmlumtrmmzmcflq m mun cocoaoca Omam . AIL oumnmmonmlm imcHmeocHqu an mmmcHEmmcmuu mumumomHuinmmeMDSHmlomm mo coaum>fluo< .m musmflm Velocity (units/ml) 38 (Velocity)4 x IOBmI-units' O 0 too 200 300 400 500 Pyridoxamine phosphate concentration xlO8 M 0 IO 20 30 4O 50 (Pyridoxamine phosphate concentration)1 x lO'4L.iter-mole'I Figure 9 39 .mumnmmonmlmlmcflaaxocfinhm nu«3 mmmcflfiamcmuu mumuoomamxoamuosmucamlomm mo coaum>wuoa may now m>uso cnmccmum .OH apnoea 40 4:. m 00 ES 0 o o o _ I I I - I O l l l l [\D Q I 1 Specific Activity (units-min'-mg) ' ‘ O '40 80 l20 ISO 200 Pyridoxamine phosphate concentration xlO7M Figure 10 41 II. $3233 maficomucu Scum AOIOV mam can canoe—Em m>o€mu on com: maz sown», GEDHOO mmtw xmcmnmmm on» How mdflmoum :oflunam mnu m3onm musmflm mane .HH Gunman 42 (wad) KIIAIIonpuoo :EONQ LmnEsz 2282.1 mm“ Ha musmfim @_ N. E23 [82?»sz mm” mww .qw rd Omdr nXUI 1 flow; I .Iomtx mXume mxwm _ _ Aum_ ngvm Aummw nvme mXuQ (Iw/nI) Mwwv esmpfiuea 43 Crude Extract Frozen E. coli cells were suspended in sonication buffer [containing 1.0 mM ATP, 0.1 mM ADP, 0.1 mM NADP+, 0.2 mM NAD+, and 0.4 ml. phenylmethylsulfonyl fluoride (25 mg/ml. in butanol)], on the basis of 1.0 m1. of buffer per 1.0 g of cells. The cells were maintained at a temperature below 5°C. at all times during the fractionation process. The cells were disrupted by 15 3-minute pulses at 80 to 85 watts using a Branson Sonifier-Cell Disrupter. Following sonication, the suspension was centrifuged at 20,000 rpm for 90 minutes to remove cell debris. AMP-Sepharose Affinity Chromatography The supernatant from the previous step was applied directly to an AMP-Sepharose column (9.0 x 2.5 cm.). The AMP ligand was attached through its N6 nitrogen atom of the adenosine ring to a hexane arm of Sepharose-4B (P-L 'Laboratories, Sepharose-4B-hexane-[N6]AMP, type 2). The dolumn was washed by the sequential application of potassium phosphate buffers (0.1 M, pH 6.8) containing 5 mM dithio- threitol, and the following additional compounds: (a) 1 mM ATP, 0.01 mM ADP, 0.1 mM NADP+, and 0.2 mM NAD+, 80 ml., (b) 0.4 M potassium chloride, 40 ml., and finally (c) potassium phosphate buffer containing 5 mM dithiothreitol. This final wash was continued until the absorbance at 280 nm became constant. 44 The enzyme was then eluted from the column by means of a 0 to 100 mM AMP continuous gradient. The gradient was prepared using potassium phosphate buffer (0.1 M, pH 6.8) containing 5 mM dithiothreitol. A typical elution profile is shown in figure 12. The fractions which contained the greatest yellow color were pooled. The enzyme was precipitated by the addition of 2.15 times the volume of the pooled fraction of a-monium sulfate (3.75 M, pH 7.0) containing 5 mM dithio- threitol, 5 mM AMP, and 1 mM EDTA. The precipitate was collected by centrifugation at 20,000 rpm for 1 hour. The pellet was resuspended in approximately 3 m1. of potassium phosphate buffer (1.0 M, pH 8.0) containing 1 mM dithiothreitol, and 5 mM AMP. Sephadex G-200 COlumn Chromatography The enzyme solution from the previous step was applied to a Sephadex G-200 column (90 x 2.5 cm.) which had been equilibrated with the same buffer. The activity was eluted by ascending chromatography with the same buffer. Figure 13 shows the elution profile for this column. The fractions from the G-200 column with the most intense yellow color were pooled. The enzyme was pre— cipitated by the addition of ammonium sulfate (as above) and centrifuged at 20,000 rpm for 1 hour. The pellet was resuspended in a minimal volume of potassium phosphate buffer (0.1 M, pH 8.0) containing 1 mM dithiothreitol, and 5 mM AMP. The enzyme solution was stored at -20°C. 45 . AIL hufltfiuom mmmuminmw maficomusu can AIL cosmommHOSHm 959.20% .GEdHoo amoumnmmmlmz¢ so Scum ammuchsmo unaccounu How mawmoum coausam .NH musmflm 46 SOUSOSSJOllL-j eAIIplea LO 0 L0 l\ In N O I l I 20 ISO— I00- 5 O O (\J 250L (Iw/n'l) Wiley esmpfiuea eUIuoeJLIl 6 IO 1“ E N v L... B 5531 :30) 22 an -.-I Ch .9. +— 0 C L. LL. 47 .111 3:: . . .uoa when a c a 85500 oomlo xmcmnmmm M WOWGAGOounu can AIL co m mmmuphcmc acacomunu wwwmwuosHm m>aumHmm . Hflmoum swaps . Hm .ma musmflm 48 eoueoseionld eAIIolea LO l\ 8 (“xi T T l 0 LO 6 o . . O N l I ll 1 l i O O O O O 0 LO 0 LO O In N N "" ‘— (lLlJ/fl'I) KIIAIIov 8801;14qu euluoelui Figure 13 E N L.“ G) .12 E :3 , Z C: .9. 4— C) C3 L... LL. RESULTS It was the intent of this research project to try to establish whether or not an appreciable amount of pyridox— imine phosphate was formed on threonine dehydrase in the presence of high levels of ammonia. The experimental approach used here was two-fold: (1) An attempt was made to trap, by sodium borohydride reduction, any pyridoximine phosphate which may be formed when threonine dehydrase was incubated with high levels of ammonia. The reduction product so formed would be pyridoxamine phosphate as a result of borohydride reduction of pyridoximine phosphate, as determined by its ability to activate resolved glutamate-oxalacetate transaminase; (2) Since the dehydrase has a characteristic absorption maximum near 410 nm., and near that of pyridoximine phosphate, an attempt was made to detect the formation of pyridoximine phosphate by difference spectroscopy. Evidence for the Enzymatic Formation of Pyridoximine Phosphate Free ammonia and AMP were removed from the threonine de- hydrase by chromatography on Sephadex G-25, as described in the Methods section. This enzyme was then immediately used in the various experiments. In a typical experiment, 49 50 threonine dehydrase (7.5 units) was incubated with 1.5 M ammonium chloride for 30 minutes at 4°C. Sodium borohydride was then added to a final concentration of 0.1 mg/ml, and the solution allowed to stand for an additional 30 minutes at 4°C. The final volume of the reaction mixture in all experiments was 0.1 ml. The solution was then heated to 90° to 100°C. for 15 minutes to destroy excess unreacted borohydride and to denature totally the threonine dehydrase. The denatured protein was removed by centrifugation and the supernatant was assayed for its ability to activate resolved glutamate-oxalacetate transaminase. Two control experiments were also performed. In the first, threonine dehydrase was treated with sodium boro- hydride and incubated for 30 minutes at 4°C. In the second control, the incubation mixture was made 1.5 M in ammonium chloride, following sodium borohydride reduction, and allowed to stand for an additional 30 minutes at 4°C. Both controls were then treated in the same manner, as were the previous samples, to destroy excess borohydride and to denature the protein from the samples. All experiments were performed both in the presence and absence of AMP. The results of these experiments are listed in Table 2. All experiments were performed in triplicate. The results presented here do indicate that when threonine dehydrase is incubated in the presence of ammonia, there is an appre- ciable amount of pyridoximine phosphate formed above that formed in the control. 51 A.COHumfl>mc oumccmum map A wuanmmozm mcflEaxocfluhm 2: mo commmumxm mum muacmmu one .28 m mmz .cmms cans .m2¢ mo :ofiumuucmo Icoo one .HE H.o mmz anon some ca mEsHo> Hacflm was .wcfiuoHno eswcoaem z m.H can .mcflucmsouon Eswcom HE\mE H.o .mmmuohnmc mcwcomunu mo mafia: m.h .BBQ 25 H .o.m Ia .Hmmmsn muosmmonm Bafimmouom SE ms cmcfiaucoo manna Hafiv .MHGOEEM 0cm ammucmnmc mcwcocunu Scum mumsmmonm mcflfiflxoofiuhm mo coaumEHOh .m manna 52 oe.v AH.m om vmmmz om Hoamz mo.AHAm.m He.m so.aaa~.m em.e om vmmmz om Aoamz mm.m a~.a om ammmz om Hoamz Ha.A mm.o om Hoemz om ammmz oe.oamm.a ma.m GA.oham.o mm.o om Hoemz om aemmz em.a Am.o om Hovmz om ammmz me.a mm.o om --- om vemmz sm.oaem.a mm.o o~.oamm.o mm.o om It- on ammmz aa.A mo.a on It- om ammmz own“ a man. can“ a «2:1 lease mafia :oAuAeea lease 62A» eoAuAeea mmmum>< om azumm mmaum>< Om Izuwm :oflumnsocH oncomm cowumnsocH umuflm m2a+ czar N manna 53 Effect of Ammonia Upon the Absorption Spectrum of Threonine Dehydrase Two approaches were used to evaluate the effect of ammonia upon the absorption spectrum of threonine dehydrase. First, the actual absorption spectrum (where the reference cuvette contained buffer only) was examined. The second approach involved an examination of the effect of ammonia upon the difference spectrum of threonine dehydrase (here both the reference and the sample cuvettes contained enzyme). The latter approach would allow for the resolution of an absorption peak, due to an interaction between threonine dehydrase and ammonia, which might occur at a wavelength very near that for the enzyme and thus be obscured in the absorption band for the enzyme. Free ammonia and AMP were removed from threonine dehydrase by chromatography on Sephades G-25, as described in the Methods section. In a typical experiment, threonine dehydrase (1.25 mg/ml) was incubated with ammonium chloride (3 M) for 30 minutes at 4°C. In addition, two control experiments were performed. The first control, threonine dehydrase was reduced by preincubation with sodium boro- hydride (30 ug/ml) for 30 minutes at 4°C. before being incubated with ammonium chloride (3 M0. In the second control, threonine dehydrase was incubated in the presence' of a-monium chloride (3 M) for 30 minutes at 4°C., and then sodium borohydride (30 ug/ml) was added and the solution incubated for an additional 30 minutes at 4°C. 54 Following the final incubation period, the various samples were transferred into cuvettes and the two types of spectra were recorded. All experiments were performed both in the presence and absence of AMP. For the two control experiments, sodium borohydride reduced-threonine dehydrase was used in the reference cuvette. Absorption Spectra The effect of ammonia upon the absorption spectrum of threonine dehydrase, both in the presence and absence of AMP, is shown in figure 14. The results here indicate that in the presence of high levels of ammonia, the absorption spectrum of threonine dehydrase becomes a very broad intense band between 350 and 440 nm. AMP has no effect upon the absorption spectrum. These spectra do not show any specific absorption phenomena in that there are no sharp, well defined peaks. Difference Spectra The conditions used here were very similar to those for determining the absorption spectra except that the absorption due to threonine dehydrase was concealed by placing enzyme in the reference as well as the sample cuvette. As a result of this, absorption maxima now occurring would be due to the interaction between ammonia and the enzyme only. The difference spectra are shown in figures 15 and 16, for the presence and absence of AMP, respectively. Figure 14. 55 The effect of ammonia upon the absorption spectrum of threonine dehydrase. (Curve 1 represents the absorption spectrum of threonine dehydrase. Curve 2 represents the absorption spectrum of sodium borohydride-reduced threonine dehydrase. Curve 3 represents the absorption spectrum of threonine dehydrase in the presence of ammonium chloride. The sample cuvette contained potassium phosphate buffer, pH 8.0, 1 mM DTT, and 1.25 mg/ml threonine dehydrase. When used, the final concentration of AMP was 5 mM, sodium borohydride 0.03 mg/ml, and ammonium chloride 3 M. The contents of the reference cuvette were identical except that enzyme was omitted. Absorbance 56 0.08 Threonine Dehydrase- +NH3 0.07 - 0.06 0.05 0.04 003' 0'02- ’/Threonine \ \ Dehydrase I ‘ OOl - +NH3+BH4 \“W “ ‘_“‘“ 1“ . v' Q 360 380 4K3 440 470 500 530 Wavelength (nm) Figure 14 Figure 15. 57 The effect of ammonium ion upon the difference spectra of threonine dehydrase in the presence of AMP. (All experiments were performed in 75 mM potassium phosphate buffer, pH 8.0, containing 5 mM AMP and 1 mM DTT. The concen- tration of threonine dehydrase was 1.25 mg/ml. When used, the concentration of ammonium chloride was 3 M and sodium borohydride was 0.03 mg/ml. The final volume for all experi- ments was 0.3 m1. For all spectra shown here, compartment 2 of the reference cuvette con- tained phosphate buffer and ammonium chloride, while compartment 2 of the sample cuvette contained phosphate buffer only. (A) The effect of ammonia upon the difference spectrum of threonine dehydrase. Compartment 1 of the reference cuvette contained phosphate buffer and threonine dehydrase. The sample cuvette contained phosphate buffer, threonine dehydrase, and ammonium chloride. (B) The effect of ammonia upon the difference spectrum of sodium borohydride-reduced threonine dehydrase. Com- partment l of the reference cuvette contained phosphate buffer, threonine dehydrase, and sodium borohydride. Compartment 1 of the sample cuvette contained phosphate buffer, threonine dehydrase, sodium borohydride, and ammonium chloride. In this spectrum, the threonine dehydrase was reduced with sodium borohydride prior to incubation with ammonium chloride. (C) The cuvettes contained the same solutions as in (B) except that the threonine dehydrase in the sample cuvette was incubated with ammonium chloride prior to reduction by sodium borohydride.) The dashed lines indicate the scanning of the same cuvettes 10 minutes after the first scanning. Absorbance 58 0.02 0.0! 0.02 00! 0.02 0.0l L L L l l l 360 380 4l0 440 470 500 530 0 Wavelength (nm) Figure 15 59 Figure 16. The effect of ammonium ion on the difference spectrum of threonine dehydrase in the absence of AMP. The experimental conditions are the same as described in figure 15 except AMP was omitted from all cuvettes. Absorbance 60 0.0 l 0045 0.03 0.02 0.0 l 0 0.0l L l l L O _ L 360 380 410 440 470 500 Wavelength (nm) Figure 16 61 The difference spectra, like the absorption spectra, are difficult to interpret. The broadness of the absorption bands would again tend to indicate some type of nonspecific interaction between threonine dehydrase and ammonia. DISCUSSION The goal of this research was to determine if the incubation of threonine dehydrase in the presence of high levels of ammonia would lead to the formation of pyridox- imine phosphate. The experimental approach used was two- fold. First, an attempt was made to trap any pyridoximine phosphate which might be formed during the incubation by reduction with sodium borohydride, to pyridoxamine phos- phate. Following denaturation and removal of the threonine dehydrase from the incubation mixture, this solution was assayed for its ability to activate resolved transaminase. The second experimental approach involved determining if high levels of ammonia would effect the absorption spectra of threonine dehydrase. Difference spectra were also ex- amined in an attempt to resolve any absorption, due to an interaction between ammonia and threonine dehydrase, which might occur at a wavelength near the absorption band char- acteristic of the enzyme and thus not be directly observable. The results, shown in table 2, indicate that there is a significant amount of pyridoximine phosphate formed when threonine dehydrase is incubated with high levels of ammonia. When these results are expressed in terms of the fraction of total pyridoxal phosphate present, on a molar basis, 62 63 which was converted to pyridoxamine phosphate, a value of approximately 17% is obtained. This value takes into a-count the fact that the threonine dehydrase used was about 60% pure. The spectral experiments, which were performed, do not provide any additional evidence for the formation of pyridoximine phosphate during the incubation of threonine dehydrase with ammonia. The spectra, which were obtained, are not easily interpretable. The general appearance of the absorption bands are very broad and do not indicate any specific absorption maxima. The result of this research indicates that enzyme- bound pyridoximine phosphate formation is not unique to the tryptophanase reaction mechanism. There is now a need to further evaluate the involvement of pyridoximine phosphate in the reaction mechanism for other pyridoxal phosphate- dependent enzymes. In particular, it is now of interest to determine if pyridoximine phosphate might be a common intermediate for all enzyme catalyzed a,B-elimination reactions. The following discussion describes some future experiments which should allow a more detailed evaluation of the role of pyridoximine phosphate in pyridoxal phosphate- mediated enzymatic catalysis. For threonine dehydrase, it is now necessary to more accurately establish the amount of pyridoximine phosphate which is generated when the enzyme is incubated with ammonia. There is a need to establish the time course 64 for the formation of this intermediate. More importantly, it is necessary to demonstrate directly, by purification, that the material which activates apo-transaminase is actually pyridoxamine phosphate. Finally, an attempt should be made to trap pyridoximine phosphate during the conversion of threonine to a-ketobutyrate by threonine dehydrase and also to determine if the presence of a-keto- butyrate decreases the accumulation of pyridoximine phosphate. The same experimental approaches, which are described above for threonine dehydrase, should be used to determine if pyridoximine phosphate is formed in the reaction mechanism for other pyridoxal phosphate-dependent enzymes. These include tryptophanase, B-tyrosinase, D-serine dehydrase, various amino acid decarboxylases, and phosphorylase. BIBLIOGRAPHY l. 13. 14. BIBLIOGRAPHY Snell, E. E., Vitamins and Hormones 16, 77-124 (1958). Braunstein, A. E., in P. D. Boyer, H. Lardy, and K. Myrback (Editors), The Enzymes, Second Edition, Vol. II, Academic Press, Inc., New York, 1960, pages 113-184. Snell, E. E., and Di Mari, S. J., in P. D. Boyer, H. Lardy, and K. Myrback (Editors), The Enzymes, Third Edition, Vol. II, Academic Press, Inc., New York, 1972, pages 335-370. Davis, L., and Metzler, D. E., in P. D. Boyer, H. Lardy, and K. Myrback (Editors), The Enzymes, Third Edition, Vol. VII, Academic Press, Inc., 'New York, 1972, pages 33-74. Gyorgy, P., J. Am. Chem. Soc. 60, 983 (1938). Keresstesy, J. C., and J. R. Stevens, J. Am. Chem. Soc. 60, 1267 (1938). Lepkovsky, S., Science 87, 169 (1938). Harris, S. A., and Folkers, K., J. Am. Chem. Soc. 61, 1245, 3307 (1939). Snell, E. E., J. Biol. Chem. 154, 313 (1944). Snell, E. E., J. Biol. Chem. 157, 491 (1945). Baddiley, J., and Gale, E. F., Nature 155, 727 (1945). Lichstein, H. C., Gunsalus, I. C., and Umbreit, W. W., J. Biol. Chem. 161, 311 (1945). Green, D. E., Leloir, L. F., and Noeito, V., J. Biol. Chem. 161, 559 (1945). Schlenk, F., and Fisher, A., Arch. Biochem. 8, 337 (1945). 65 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 66 Metzler, D. E., and Snell, E. E., J. Am. Chem. Soc. 77, 2431 (1955). Metzler, D. E., J. Am. Chem- Soc. 79, 485 (1957). Christensen, H. N., J. Am. Chem. Soc. 80, 99 (1958). Metzler, D. E., Ikawa, M., and Snell, E. E., J. Am. Chem. Soc. 76, 648 (1954). Braunstein, A. E., and Shemyakin, M. M., Biokhimiya 18, 393 (1953). Chem. Abstr. 48, 46030 (1954). Watanabe, I., and Snell, E. E., Pro. Nat. Acad. Sci. USA 69, 1086 (1972). Phillips, A. T., and Wood, W. A., J. Biol. Chem. 240, 4703 (1965). Niderman, H. A., Rabinowitz, K. W., and Wood, W. A., Biochem. Biophys. Res. Commun. 39, 951 (1969). Shizuta, Y., Nakazawa, A., Takushige, M., and Hayaishi, 0., J. Biol. Chem. 244, 1-83 (1969) Snell, E. E., Advances in Enzymology 42, 287 (1976). Kiumagai, H., Yamada, H., Matsui, H., Ohkishi, H., and Ogata, K., J. Biol. Chem. 245, 1773 (1970). Flavin, M., and Slaughter, C., J. Biol. Chem. 244, 1434 (1969). Cleland, W. W., in P. D. Boyer, H. Lardy, and K. Myrback (Editors), The Enzymes, Third Edition, Vol. II, Academic Press, Inc., New York, 1972, pages 13-18.~ Lowry, O. H., Roseborough, N. J., Farr, A. L., and Randall, R. J., J. Biol. Chem. 193, 265 (1951). Bohlen, P., Stein, 8., Dairman, W., and Undendriend, 8., Arch. Biochem. Biophys. 155, 213 (1973). Dunne, C. P., Gerlt, J. A., Rabinowitz, K. W., and Wood, W. A., J. Biol. Chem. 248, 8189 (1973). Bergmeyer, H. U., and Bent, E., in H. U. Bergmeyer (Editor), Methods of Enzymatic Analysis, Second Edition, Vol. 2, Academic Press, Inc., New York, 1974, pages 727-735. 32. 33. 34. 35. 67 Wada, H., and Snell, E. E., J. Biol. Chem. 237, (1962). 127 O'Kane, D. F., and Gunsalus, I. C., J. Biol. Chem. 170, 425 (1947). Rabinowitz, K. W., Ph.D. Thesis, Michigan State University, 1970. Le Blond, D. J., and Wood, W. A., unpublished experiments.