Nonstereospecific Transamination Catalyzed by Pyridoxal Phosphate-dependent Amino Acid Racemases of Broad Substrate Specificity
1998; Elsevier BV; Volume: 273; Issue: 7 Linguagem: Inglês
10.1074/jbc.273.7.4001
ISSN1083-351X
AutoresYoung Hee Lim, Tohru Yoshimura, Yoichi Kurokawa, Nobuyoshi Esaki, Kenji Soda,
Tópico(s)Drug Transport and Resistance Mechanisms
ResumoPyridoxal 5′-phosphate-dependent amino acid racemases of broad substrate specificity catalyze transamination as a side reaction. We studied the stereospecificities for hydrogen abstraction from C-4′ of the bound pyridoxamine 5′-phosphate during transamination from pyridoxamine 5′-phosphate to pyruvate catalyzed by three amino acid racemases of broad substrate specificity. When the enzymes were incubated with (4′S)- or (4′R)-[4′-3H]pyridoxamine 5′-phosphate in the presence of pyruvate, tritium was released into the solvent from both pyridoxamine 5′-phosphates. Thus, these enzymes abstract a hydrogen nonstereospecifically from C-4′ of the coenzyme in contrast to the other pyridoxal 5′-phosphate-dependent enzymes so far studied, which catalyze the stereospecific hydrogen removal. Amino acid racemase of broad substrate specificity from Pseudomonas putidaproduced d- and l-glutamate from α-ketoglutarate through the transamination withl-ornithine. Because glutamate does not serve as a substrate for racemization, the enzyme catalyzed the nonstereospecific overall transamination between l-ornithine and α-ketoglutarate. The cleavage and formation of the C–H bond at C-4′ of the coenzyme and C-2 of the substrate thus occurs nonstereospecifically on both sides of the plane of the coenzyme-substrate complex intermediate. Amino acid racemase of broad substrate specificity is the first example of a pyridoxal enzyme catalyzing nonstereospecific transamination. Pyridoxal 5′-phosphate-dependent amino acid racemases of broad substrate specificity catalyze transamination as a side reaction. We studied the stereospecificities for hydrogen abstraction from C-4′ of the bound pyridoxamine 5′-phosphate during transamination from pyridoxamine 5′-phosphate to pyruvate catalyzed by three amino acid racemases of broad substrate specificity. When the enzymes were incubated with (4′S)- or (4′R)-[4′-3H]pyridoxamine 5′-phosphate in the presence of pyruvate, tritium was released into the solvent from both pyridoxamine 5′-phosphates. Thus, these enzymes abstract a hydrogen nonstereospecifically from C-4′ of the coenzyme in contrast to the other pyridoxal 5′-phosphate-dependent enzymes so far studied, which catalyze the stereospecific hydrogen removal. Amino acid racemase of broad substrate specificity from Pseudomonas putidaproduced d- and l-glutamate from α-ketoglutarate through the transamination withl-ornithine. Because glutamate does not serve as a substrate for racemization, the enzyme catalyzed the nonstereospecific overall transamination between l-ornithine and α-ketoglutarate. The cleavage and formation of the C–H bond at C-4′ of the coenzyme and C-2 of the substrate thus occurs nonstereospecifically on both sides of the plane of the coenzyme-substrate complex intermediate. Amino acid racemase of broad substrate specificity is the first example of a pyridoxal enzyme catalyzing nonstereospecific transamination. Although enzymatic racemization of amino acid is apparently simple, consisting of a nonstereospecific rearrangement of the substrate α-hydrogen, several different types of amino acid racemases are found in microorganisms. Aspartate racemase (EC 5.1.1.13) (1Yamauchi T. Choi S.-Y. Okada H. Yohda M. Kumagai H. Esaki N. Soda K. J. Biol. Chem. 1992; 267: 18361-18364Abstract Full Text PDF PubMed Google Scholar) and glutamate racemase (EC 5.1.1.3) (2Nakajima N. Tanizawa K. Tanaka H. Soda K. Agric. Biol. Chem. 1986; 50: 2823-2830Crossref Scopus (3) Google Scholar, 3Gallo K.A. Knowles J.R. Biochemistry. 1993; 32: 3981-3990Crossref PubMed Scopus (106) Google Scholar) are independent of cofactors, and one or more cysteinyl residues play an important role in the abstraction of the α-hydrogen from the substrate. Phenylalanine racemase (EC 5.1.1.11), which is involved in gramicidin S synthesis, utilizes ATP (4Kanda M. Hori K. Kurotsu T. Miura S. Saito Y. J. Biochem. (Tokyo). 1989; 105: 653-659Crossref PubMed Scopus (13) Google Scholar). Alanine racemase (EC 5.1.1.1) in several microorganisms (5Yoshimura T. Soda K. Fukui T. Soda K. Molecular Aspects of Enzyme Catalysis. Kodansha, Tokyo1994: 147-163Crossref Scopus (5) Google Scholar), arginine racemase (EC 5.1.1.9) from Pseudomonas graveolens (6Yorifuji T. Ogata K. Soda K. J. Biol. Chem. 1971; 246: 5085-5092Abstract Full Text PDF PubMed Google Scholar), and amino acid racemases of broad substrate specificity (EC 5.1.1.10) of Aeromonas punctata (7Inagaki K. Tanizawa K. Tanaka H. Soda K. Agric. Biol. Chem. 1987; 51: 173-180Google Scholar), Pseudomonas striata (8Soda K. Osumi T. Methods Enzymol. 1971; 17B: 629-636Crossref Scopus (32) Google Scholar), and Pseudomonas putida (9Lim Y.-H. Yokoigawa K. Esaki N. Soda K. J. Bacteriol. 1994; 175: 4213-4217Crossref Google Scholar) all depend on pyridoxal 5′-phosphate (PLP) 1The abbreviations used are: PLP, pyridoxal 5′-phosphate; AspAT, aspartate aminotransferase; BCAT, branched chainl-amino acid aminotransferase; d-AAT,d-amino acid aminotransferase; PMP, pyridoxamine 5′-phosphate; Tricine, N-tris(hydroxymethyl)methylglycine; HPLC, high performance; DAP, meso-α,ε-diaminopimelate.. Faraci and Walsh (10Faraci W.S. Walsh C.T. Biochemistry. 1988; 27: 3267-3276Crossref PubMed Scopus (52) Google Scholar) proposed a mechanism for alanine racemase that is probably similar to that of other PLP-dependent amino acid racemases. The reaction is initiated by transaldimination. In this step, PLP bound with the active-site lysyl residue through an internal Schiff base (Scheme FSI A) reacts with a substrate to form an external Schiff base (B). The subsequent α-hydrogen abstraction results in the formation of a resonance-stable anionic intermediate (C). If the reprotonation occurs at C-2 of the substrate moiety on the opposite face of the planar intermediate to that where the proton abstraction occurs, an antipodal aldimine is formed (D). The aldimine complex is subsequently hydrolyzed to form isomerized amino acid and regenerates the bound PLP (E). The random return of hydrogen to the anionic intermediate is a characteristic of enzymatic racemization among various pyridoxal enzyme reactions. In aminotransferase reactions, the abstracted hydrogen is stereospecifically transferred to C-4′ of the cofactor, and a ketimine intermediate is formed (11Martinez-Carrion M. Hubert E. Iriarte A. Mattingly J.R. Zito S.W. Christen P. Metzler D.E. Transaminases. John Wiley & Sons, Inc., New York1985: 308-316Google Scholar). The pyridoxamine 5′-phosphate (PMP) form of the enzyme and a keto acid are produced by hydrolysis of the ketimine intermediate. Amino acid racemases are reported to catalyze the transamination as a side reaction (12Yorifuji T. Misono H. Soda K. J. Biol. Chem. 1971; 246: 5093-5101Abstract Full Text PDF PubMed Google Scholar). The transamination catalyzed by amino acid racemases can be attained through a sequence either A → B → F (or G) → H or A → B → C → F (or G) → H (Scheme FSI). An equivalent route can be delineated for the antipode: E → D → F (or G) → H or E → D → C → F (or G) → H. In transamination, mutual hydrogen transfer between the substrate and C-4′ of the cofactor occurs. In all previous studies of transaminations catalyzed by aminotransferases such as l-aspartate aminotransferase (AspAT) (13Besmer P. Arigoni D. Chimia. 1969; 23: 190Google Scholar), l-alanine aminotransferase (14Floss H.G. Vederas J.C. Tamm Ch Stereochemistry. Elsevier Biomedical Press, Amsterdam1982: 161-199Google Scholar), dialkylamino acid aminotransferase (15Bailey G.B. Kusamrarn T. Vuttivej K. Fed. Proc. 1970; 29: 857Google Scholar), pyridoxamine:pyruvate aminotransferase (16Ayling J.E. Dunathan H.C. Snell E.E. Biochemistry. 1968; 7: 4532-4542Crossref PubMed Scopus (47) Google Scholar), d-amino acid aminotransferase (d-AAT) (17Yoshimura T. Nishimura K. Ito J. Esaki N. Kagamiyama H. Manning J.M. Soda K. J. Am. Chem. Soc. 1993; 115: 3897-3900Crossref Scopus (61) Google Scholar), and branched chain l-amino acid aminotransferase (BCAT) (17Yoshimura T. Nishimura K. Ito J. Esaki N. Kagamiyama H. Manning J.M. Soda K. J. Am. Chem. Soc. 1993; 115: 3897-3900Crossref Scopus (61) Google Scholar) as well as other pyridoxal enzymes such asl-serine hydroxymethyltransferase (18Voet J.G. Hindenlang D.M. Blanck T.J.J. Ulevitch R.J. Kallen R.G. Dunathan H.C. J. Biol. Chem. 1973; 248: 841-842Abstract Full Text PDF PubMed Google Scholar),l-tryptophan synthase (19Dunathan H.C. Voet J.G. Proc. Natl. Acad. Sci. U. S. A. 1974; 71: 3888-3891Crossref PubMed Scopus (92) Google Scholar), l-glutamate decarboxylase (20Sukhareva B.S. Dunathan H.C. Braunstein A.E. FEBS Lett. 1971; 15: 241-244Crossref PubMed Scopus (22) Google Scholar), and l-aspartate β-decarboxylase (21Chang C.C. Laghai A. O'Leary M.H. Floss H.G. J. Biol. Chem. 1982; 257: 3564-3569Abstract Full Text PDF PubMed Google Scholar), the hydrogen transfer between substrate and cofactor occurs strictly stereospecifically on the si or re face of the plane of the anionic intermediate. However, if the transamination catalyzed by amino acid racemases proceeds as depicted in Scheme FSI, the hydrogen transfer should occur nonstereospecifically on both faces of the planar intermediate. We were therefore interested in the stereospecificity of the hydrogen transfer during transamination catalyzed by amino acid racemases as a side reaction. We provide here the first evidence that hydrogen removal from C-4′ of PMP occurs randomly on both faces of the substrate-cofactor imine plane during half transamination catalyzed by the amino acid racemases. We also show that the enzyme catalyzes nonstereospecific overall transamination between l-ornithine and α-ketoglutarate as well. Amino acid racemases with low substrate specificity from P. putida (9Lim Y.-H. Yokoigawa K. Esaki N. Soda K. J. Bacteriol. 1994; 175: 4213-4217Crossref Google Scholar), P. striata (8Soda K. Osumi T. Methods Enzymol. 1971; 17B: 629-636Crossref Scopus (32) Google Scholar), and A. punctata (7Inagaki K. Tanizawa K. Tanaka H. Soda K. Agric. Biol. Chem. 1987; 51: 173-180Google Scholar) were purified as described previously. BCAT of Escherichia coli K-12 (22Kuramitsu S. Ogawa H. Kagamiyama H. J. Biochem. (Tokyo). 1985; 97: 993-999Crossref PubMed Scopus (56) Google Scholar) was kindly supplied by Professor H. Kagamiyama and Dr. K. Inoue of Osaka Medical College, Takatsuki, Japan. AspAT from pig heart was obtained from Boehringer Mannheim (Germany), 3H2O (3.7 GBq/g) was from DuPont. The other chemicals were of analytical grade. The activity of the amino acid racemases was determined by measurement of a change in optical rotation at 365 nm with a Perkin-Elmer 241 polarimeter and also by a coupled assay procedure with d- or l-amino acid oxidase (9Lim Y.-H. Yokoigawa K. Esaki N. Soda K. J. Bacteriol. 1994; 175: 4213-4217Crossref Google Scholar). Protein concentrations were determined by dye staining with a Bio-Rad protein assay reagent with bovine serum albumin as a standard. Apo-amino acid racemases were prepared as described previously (7Inagaki K. Tanizawa K. Tanaka H. Soda K. Agric. Biol. Chem. 1987; 51: 173-180Google Scholar). An appropriate amount of enzyme (1–3 mg) was dialyzed against 10 mm potassium phosphate buffer (pH 8.0) containing 30 mm hydroxylamine and 0.1% 2-mercaptoethanol for 24 h and then dialyzed against 10 mm potassium phosphate buffer (pH 8.0) at 4 °C overnight. The apoenzyme formed was determined by measurement of the activity of the enzyme in the presence or absence of 10 mmPLP. Apo-enzymes recovered about 80% of their initial activities on addition of PLP. Apo-AspAT, apo-d-AAT, and apo-BCAT were prepared as described previously (17Yoshimura T. Nishimura K. Ito J. Esaki N. Kagamiyama H. Manning J.M. Soda K. J. Am. Chem. Soc. 1993; 115: 3897-3900Crossref Scopus (61) Google Scholar). Spectrophotometric measurements were made with a Shimadzu UV-visible recording spectrophotometer UV-260 with a 1.0-cm light path at 25 °C. (4′S)- and (4′R)-[4′-3H]PMP were prepared by incubation of the randomly labeled [4′-3H]PMP with apo-BCAT and apo-AspAT, respectively, as described previously (17Yoshimura T. Nishimura K. Ito J. Esaki N. Kagamiyama H. Manning J.M. Soda K. J. Am. Chem. Soc. 1993; 115: 3897-3900Crossref Scopus (61) Google Scholar). The specific radioactivities of (4′S)- and (4′R)-[4′3H]PMP prepared were 1.54 × 106 and 1.35 × 106 dpm/μmol, respectively. Stereospecificities for labeling were confirmed by measurement of tritium liberation from both PMPs catalyzed by apo-BCAT and apo-AspAT in the presence of α-ketoglutarate as described previously (17Yoshimura T. Nishimura K. Ito J. Esaki N. Kagamiyama H. Manning J.M. Soda K. J. Am. Chem. Soc. 1993; 115: 3897-3900Crossref Scopus (61) Google Scholar). The reaction mixture (100 μl) contained 10 μmol of potassium phosphate buffer (pH 8.0), 5 nmol of sodium pyruvate, 1 nmol of (4′S)- or (4′R)-[4′-3H]PMP, and 5 nmol of each apo-amino acid racemase. The reaction was carried out at 30 °C for 3 h and terminated by the addition of 100 μl of 1 m HCl. The mixture was immediately frozen in liquid nitrogen and dried with a Speed Vac concentrator. The residue was dissolved with 200 μl of H2O and subjected to a radioactivity assay. The tritium released from PMP was expressed as volatile radioactivity, which was estimated by subtraction of the radioactivity finally remaining from the radioactivity initially added to the reaction mixture. The reaction mixture (100 μl) with apo-d-AAT and apo-AspAT contained 10 μmol of Tris-HCl buffer (pH 8.0), 10 nmol of sodium α-ketoglutarate, 1.0 nmol of (4′S)- or (4′R)-[4′-3H]PMP, and 168 μg of apo-AspAT or 182 μg of apo-d-AAT, respectively. The reaction was carried out at 30 °C for 15 min and terminated by the addition of 100 μl of 1 m HCl. Other conditions were the same as those of the reactions with amino acid racemases. Radioactivity was determined with a Packard Tri-Carb scintillation spectrometer with Clear-solI (Nacalai Tesque, Japan) as a scintillator. The reaction mixture (1 ml) consisted of 10 μmol ofl-ornithine, 10 μmol of α-ketoglutarate, 0.05 μmol of PLP, 50 μmol of Tricine buffer (pH 8.5), and 20 μg of enzyme. The reaction was carried out at 37 °C for 2 h and terminated by boiling. After centrifugation, amino acids in the supernatant solution were derivatized to diastereomers with Marfey's reagent (23Marfey P. Carlsberg Res. Commun. 1984; 49: 591-596Crossref Scopus (1397) Google Scholar) and analyzed by HPLC. When the amino acid racemase of broad substrate specificity from P. putida was incubated with l-alanine, the absorption at 420 nm derived from the internal Schiff base decreased, with a concomitant increase in the absorption at 330 nm (Fig.1 a). This suggests that the coenzyme form of the enzyme was converted from PLP to PMP during the racemization. The reversal from PMP to PLP also occurred on the basis of a spectral shift. The addition of pyruvate to the PMP form of enzyme, which was prepared by incubation of PMP with the apo-enzyme, led to a decrease in absorbance at 330 nm and an increase in that at 420 nm (Fig. 1 b). Similar spectral shifts were observed in the reactions of the PMP form of the amino acid racemases from P. striata and A. punctata with pyruvate (data not shown). The results indicate that these amino acid racemases catalyze the transamination as a side reaction. Amino acid racemase of P. putida catalyzes the overall transamination between ornithine and pyruvate. The specific activity for the transamination was 0.26 units/mg (∼10.7 min−1). The rate of transamination was lower than that of the racemization of ornithine by a factor of 1.1 × 104. However, the rate of enzymatic transamination was at least several orders of magnitude higher than that of the nonenzymatic transamination, which was lower than the minimum value for accurate determination, ∼4.0 × 10−6 min−1. When the PMP form of an enzyme is converted to the PLP form by transamination with an amino acceptor (keto acid), one of the two hydrogens at C-4′ of PMP is usually transferred stereospecifically to Cα of the keto acid (24Dunathan H.C. Adv. Enzymol. Relat. Areas Mol. Biol. 1971; 35: 79-134PubMed Google Scholar). We studied the stereospecificity of amino acid racemases for hydrogen abstraction from C-4′ of PMP using the method described previously (17Yoshimura T. Nishimura K. Ito J. Esaki N. Kagamiyama H. Manning J.M. Soda K. J. Am. Chem. Soc. 1993; 115: 3897-3900Crossref Scopus (61) Google Scholar). Stereospecificity is determined by measurement of the radioactivity of 3H released from the PMPs, which are stereospecifically tritiated at C-4′. Each 5 nmol of apo-amino acid racemase was incubated with 1 nmol of (4′S)- or (4′R)-[4′-3H]PMP and 5 nmol of sodium pyruvate. We deduce that PMP was completely converted to PLP, because the PMP form of the amino acid racemase from P. putidarecovered 100% of the activity theoretically expected. As shown in Table I, tritium was released equally from both (4′S)- and (4′R)-[4′-3H]PMPs in the presence of amino acid racemases. The amount of tritium released from each PMP was about 50% that which initially existed. The control experiment was done withd-AAT and AspAT. As reported previously (17Yoshimura T. Nishimura K. Ito J. Esaki N. Kagamiyama H. Manning J.M. Soda K. J. Am. Chem. Soc. 1993; 115: 3897-3900Crossref Scopus (61) Google Scholar),d-AAT and AspAT catalyzed the stereospecific removal of tritium from (4′R)- or (4′S)-[4′3H]PMP, respectively, as shown in Table I. These results confirm the stereospecific tritium labeling of both PMPs. No tritium was released from each PMP in the absence of enzyme. Thus, the amino acid racemases catalyze the non-stereospecific abstraction of hydrogen from C-4′ of PMP. They are the first class of pyridoxal enzyme catalyzing the hydrogen removal on both sides of the plane of a substrate-cofactor complex during transamination.Table IRelease of 3H from [4′-3H]PMPs by amino acid racemases4′-S-[4′-3H]PMP1-aThe initial radioactivity in the reaction mixture was 1,540 dpm.3H released1-bVolatile radioactivity.4′-R-[4′-3H]PMP1-cThe initial radioactivity in the reaction mixture was 1,340 dpm.3H released1-bVolatile radioactivity.dpm%1-dRatio of the released radioactivity to that initially added in the reaction mixture.dpm%1-dRatio of the released radioactivity to that initially added in the reaction mixture.apo-AAR1-eAmino acid racemase from P. putida.6664365549apo-AAR1-fAmino acid racemase from P. striata.6684371553apo-AAR1-gAmino acid racemase from A. punctata.6754462747apo-d-AAT362.386064apo-AspAT12097800Without enzyme00001-a The initial radioactivity in the reaction mixture was 1,540 dpm.1-b Volatile radioactivity.1-c The initial radioactivity in the reaction mixture was 1,340 dpm.1-d Ratio of the released radioactivity to that initially added in the reaction mixture.1-e Amino acid racemase from P. putida.1-f Amino acid racemase from P. striata.1-g Amino acid racemase from A. punctata. Open table in a new tab If the hydrogen is introduced nonspecifically to C-2 of the keto acid moiety of the anionic intermediate on both sides of the planar intermediate during the half reaction of transamination, racemic amino acid is formed from the keto acid (Scheme FSI; H → F (or G) → C → B → A or H → F (or G) → C → D → E). We studied the stereochemistry of glutamate formed from α-ketoglutarate by transamination with l-ornithine catalyzed by the amino acid racemase of P. putida. After the reaction, the products were derivatized to diastereomers with Marfey's reagent (23Marfey P. Carlsberg Res. Commun. 1984; 49: 591-596Crossref Scopus (1397) Google Scholar) and subjected to HPLC. As shown in Fig. 2, both enantiomers of glutamate and ornithine were found. The amino acid racemase from P. putida catalyzes the racemization of ornithine, but glutamate is inert as a substrate for the racemase reaction (9Lim Y.-H. Yokoigawa K. Esaki N. Soda K. J. Bacteriol. 1994; 175: 4213-4217Crossref Google Scholar). Thus, both enantiomers of glutamate were directly formed by transamination, not by racemization of one enantiomer produced through transamination. Amino acid racemase from P. putidacatalyzes the nonstereospecific overall transamination. We here show that nonstereospecific hydrogen abstraction from C-4′ of PMP occurs in the transamination between PMP and pyruvate catalyzed by amino acid racemases with broad substrate specificity. The enzyme from P. putida also catalyzes nonstereospecific overall transamination between ornithine and α-ketoglutarate. These results indicate that the cleavage and formation of the C–H bonds at C-4′ of the coenzyme occur on both sides of the plane of the anionic intermediate. This model is compatible with the proposed mechanism for racemization. The removal and return of α-hydrogen of substrate occur on both sides of the plane (Scheme FSI). It may be interesting to examine whether the removal of substrate α-hydrogen and the introduction of coenzyme C-4′ hydrogen occur on the same side of the plane. The examination will be facilitated by analysis of the configuration of3H-labeled PMP formed in 3H2O by transamination with either d or l enantiomer of an appropriate substrate. However, the amino acid used for this purpose needs to be inert as a substrate for racemization; otherwise, PMP will be also derived from the antipode of the substrate enantiomer added initially. Glutamate might be the only candidate substrate for this purpose; however, this is not the case. Glutamate is inert as an amino donor in transamination, 2Lim, Y.-H., Yoshimura, T., Kurokawa, Y., Esaki, N., and Soda, K, unpublished results.although α-ketoglutarate serves as an amino acceptor as described above. Two mechanisms have been proposed for enzymatic racemization: a two-base mechanism and a one-base mechanism (25Cardinale G.J. Abeles R.H. Biochemistry. 1968; 7: 3979-3987Crossref PubMed Scopus (156) Google Scholar, 26Soda K. Tanaka H. Tanizawa K. Dorphin D. Pyridoxal Phosphate: Chemical, Biochemical, and Medical Aspects. 1B. John Wiley & Sons, Inc., New York1986: 223-251Google Scholar). In the two-base mechanism, two different bases participate in the catalysis; one abstracts the hydrogen from a substrate, and the other returns a hydrogen to the deprotonated intermediate. Glutamate racemase (27Choi S.Y. Esaki N. Yoshimura T. Soda K. J. Biochem. (Tokyo). 1992; 112: 139-142Crossref PubMed Scopus (47) Google Scholar, 28Gallo K.A. Tanner M.E. Knowles J.R. Biochemistry. 1993; 32: 3991-3997Crossref PubMed Scopus (99) Google Scholar), aspartate racemase (1Yamauchi T. Choi S.-Y. Okada H. Yohda M. Kumagai H. Esaki N. Soda K. J. Biol. Chem. 1992; 267: 18361-18364Abstract Full Text PDF PubMed Google Scholar), proline racemase (25Cardinale G.J. Abeles R.H. Biochemistry. 1968; 7: 3979-3987Crossref PubMed Scopus (156) Google Scholar, 30Runick G. Abeles R.H. Biochemistry. 1975; 14: 4515-4522Crossref PubMed Scopus (100) Google Scholar, 31Albery J. Knowles J. Biochemistry. 1986; 25: 2572-2577Crossref PubMed Scopus (83) Google Scholar), and diaminopimelate epimerase (32Wiseman J.S. Nichols J.S. J. Biol. Chem. 1984; 259: 8907-8914Abstract Full Text PDF PubMed Google Scholar), which do not require cofactors, catalyze racemization by this mechanism. In the one-base mechanism, a single amino acid residue abstracts the α-hydrogen from a substrate and nonstereospecifically returns it to the anionic intermediate. A swinging door motion has been proposed by Henderson and Johnston (33Henderson L.L. Johnston R.B. Biochem. Biophys. Res. Commun. 1976; 68: 793-798Crossref PubMed Scopus (27) Google Scholar) as a model to fit the one-base mechanism; the plane of the substrate·PLP complex acts like a swinging door in order that the base can be located on both faces of the plane. If only a single base is involved, one can expect that the α-hydrogen derived from the substrate will be retained at the α-position of the product (25Cardinale G.J. Abeles R.H. Biochemistry. 1968; 7: 3979-3987Crossref PubMed Scopus (156) Google Scholar). Such an internal retention of the α-hydrogen was verified in the reactions catalyzed by two PLP-dependent racemases: amino acid racemase with low substrate specificity from P. striata(34Shen S. Floss H.G. Kumagai H. Yamada H. Esaki N. Soda K. Wasserman S.A. Walsh C. J. Chem. Soc. Chem. Commun. 1983; : 82-83Crossref Scopus (25) Google Scholar) and α-amino-ε-caprolactam racemase from Achromobacter obae (35Ahmed S.A. Esaki N. Tanaka H. Soda K. Biochemistry. 1986; 25: 385-388Crossref PubMed Scopus (44) Google Scholar). Thus, it was supposed that these reactions proceed through a single-base mechanism (34Shen S. Floss H.G. Kumagai H. Yamada H. Esaki N. Soda K. Wasserman S.A. Walsh C. J. Chem. Soc. Chem. Commun. 1983; : 82-83Crossref Scopus (25) Google Scholar, 35Ahmed S.A. Esaki N. Tanaka H. Soda K. Biochemistry. 1986; 25: 385-388Crossref PubMed Scopus (44) Google Scholar). However, Shostak and Schirch (36Shostak K. Schirch V. Biochemistry. 1988; 27: 8007-8014Crossref PubMed Scopus (49) Google Scholar) argued in their studies on the mechanism of alanine racemization catalyzed by serine hydroxymethyltransferase that the internal retention of the substrate-derived α-hydrogen could also allow a two-base mechanism by assuming a hydrogen shuttle between the bases as proposed for the aconitase reaction; the latter enzyme is known to have a network of at least five interchangeable protons at the active site that only exchange slowly with the solvent (37Kuo D.J. Rose I.A. Biochemistry. 1987; 26: 7589-7596Crossref PubMed Scopus (16) Google Scholar). Although the reaction catalyzed by the amino acid racemase from P. striata was accompanied by clear internal retention of the α-hydrogen, a two-base mechanism has been proposed on the basis of the complete disagreement between the substrate enantiomers examined for the relative rates of deuterium incorporation from 2H2O into separate enantiomers (38Reynolds K. Martin J. Shen S.-J. Esaki N. Soda K. Floss H.G. J. Basic Microbiol. 1991; 31: 177-188Crossref PubMed Scopus (6) Google Scholar). Our findings shown here indicate that the catalytic base(s) responsible for α-hydrogen abstraction and addition are situated on both faces of the plane of the substrate-cofactor complex. Our enzyme is assumed to be closely homologous to the P. striata enzyme, because the latter strain is now classified into the same group as P. putida. If this is the case, our enzyme would also use two bases for catalysis. Recently, Shaw et al. (39Shaw J.S. Petsko G.A. Ringe D. Biochemistry. 1997; 36: 1329-1342Crossref PubMed Scopus (257) Google Scholar) clarified the three-dimensional structure of alanine racemase from Bacillus stearothermophilus. They suggested that Tyr-265 and Lys-39, the PLP binding lysine, serve as the bases. Sawada et al. have presented kinetic evidence to show that the alanine racemase reaction follows a two-base mechanism (40Sawada S. Tanaka Y. Hayashi S. Ryu M. Hasegawa T. Yamamoto Y. Esaki N. Soda K. Takahashi S. Biosci. Biotechnol. Biochem. 1994; 58: 807-811Crossref Scopus (10) Google Scholar). The stereospecificity for the hydrogen transfer reflects the structure of the active site of pyridoxal enzymes, especially the topographical relationship between the catalytic base for the hydrogen transfer and the bound coenzyme. Therefore, stereospecificity has been discussed in relation to the molecular evolution of the pyridoxal enzymes (17Yoshimura T. Nishimura K. Ito J. Esaki N. Kagamiyama H. Manning J.M. Soda K. J. Am. Chem. Soc. 1993; 115: 3897-3900Crossref Scopus (61) Google Scholar, 18Voet J.G. Hindenlang D.M. Blanck T.J.J. Ulevitch R.J. Kallen R.G. Dunathan H.C. J. Biol. Chem. 1973; 248: 841-842Abstract Full Text PDF PubMed Google Scholar,24Dunathan H.C. Adv. Enzymol. Relat. Areas Mol. Biol. 1971; 35: 79-134PubMed Google Scholar). No whole primary structure or three-dimensional structure of amino acid racemases with low substrate specificity has been determined. However, the primary structures of alanine racemases, which are homologous to the amino acid racemases in the sequences around the lysine residue binding PLP (41Tanizawa K. Ohshima A. Scheidegger A. Inagaki K. Tanaka H. Soda K. Biochemistry. 1988; 27: 1311-1316Crossref PubMed Scopus (61) Google Scholar), show few similarities to those of other PLP enzymes (42Mehta P.K. Christen P. Eur. J. Biochem. 1993; 211: 373-376Crossref PubMed Scopus (48) Google Scholar). Therefore, amino acid racemase with low substrate specificity probably belongs to the same family of proteins as alanine racemase: a unique family containing only alanine racemase, mammalian ornithine decarboxylase, and meso-α,ε-diaminopimelate (DAP) decarboxylase with little similarity to other PLP enzymes (43Grishin N.V. Phillips M.A. Goldsmith E.J. Protein Sci. 1995; 4: 1291-1304Crossref PubMed Scopus (343) Google Scholar). We have shown that the decarboxylation of DAP catalyzed by DAP decarboxylase proceeds through inversion of the configuration of the α-carbon of DAP (29Asada Y. Tanizawa K. Sawada S. Suzuki T. Misono H. Soda K. Biochemistry. 1981; 20: 6881-6886Crossref PubMed Scopus (49) Google Scholar). This indicates that the enzyme function must be conducted on both sides of the plane of the substrate·PLP complex so as to decarboxylate on one side and to introduce a proton on the other side. The DAP decarboxylase reaction is homologous to the racemase reaction in this respect. It may be interesting to examine whether DAP decarboxylase and ornithine decarboxylase catalyze the removal of tritium nonstereospecifically from C-4′ of both (4′S)- and (4′R)-[4′-3H]PMP in the same manner as amino acid racemase with low substrate specificity. Our preliminary results have shown that alanine racemase from B. stearothermophilusalso catalyzes the nonstereospecific removal of tritium from both [4′-3H]PMPs. Whatever the results of the DAP decarboxylase and ornithine decarboxylase reactions may be, it is interesting to note the clear relationship between the stereospecificity of hydrogen transfer and families of PLP enzymes.
Referência(s)