Glutamate Racemase Is an Endogenous DNA Gyrase Inhibitor
2002; Elsevier BV; Volume: 277; Issue: 42 Linguagem: Inglês
10.1074/jbc.c200253200
ISSN1083-351X
AutoresMakoto Ashiuchi, Eriko Kuwana, Takashi Yamamoto, Kazuya Komatsu, Kenji Soda, Haruo Misono,
Tópico(s)Biopolymer Synthesis and Applications
ResumoAlmost all bacteria possess glutamate racemase to synthesize d-glutamate as an essential component of peptidoglycans in the cell walls. The enforced production of glutamate racemase, however, resulted in suppression of cell proliferation. In the Escherichia coli JM109/pGR3 clone, the overproducer of glutamate racemase, the copy number (i.e. replication efficiency) of plasmid DNA declined dramatically, whereas the E. coli WM335 mutant that is defective in the gene of glutamate racemase showed little genetic competency. The comparatively low and high activities for DNA supercoiling were contained in the E. coli JM109/pGR3 and WM335 cells, respectively. Furthermore, we found that the DNA gyrase ofE. coli was modulated by the glutamate racemase of E. coli in the presence of UDP-N-acetylmuramyl-l-alanine, which is a peptidoglycan precursor and functions as an absolute activator for the racemase. This is the first finding of the enzyme protein participating in both d-amino acid metabolism and DNA processing. Almost all bacteria possess glutamate racemase to synthesize d-glutamate as an essential component of peptidoglycans in the cell walls. The enforced production of glutamate racemase, however, resulted in suppression of cell proliferation. In the Escherichia coli JM109/pGR3 clone, the overproducer of glutamate racemase, the copy number (i.e. replication efficiency) of plasmid DNA declined dramatically, whereas the E. coli WM335 mutant that is defective in the gene of glutamate racemase showed little genetic competency. The comparatively low and high activities for DNA supercoiling were contained in the E. coli JM109/pGR3 and WM335 cells, respectively. Furthermore, we found that the DNA gyrase ofE. coli was modulated by the glutamate racemase of E. coli in the presence of UDP-N-acetylmuramyl-l-alanine, which is a peptidoglycan precursor and functions as an absolute activator for the racemase. This is the first finding of the enzyme protein participating in both d-amino acid metabolism and DNA processing. UDP-N-acetylmuramyl-l-alanine fast protein liquid chromatography pyridoxal 5′-phosphate isopropyl-1-thio-β-d-galactopyranoside Bacterial cell walls contain d-amino acids as essential components of peptidoglycans (alternatively, mureins).d-Glutamate is introduced into peptidoglycan through its addition to UDP-N-acetylmuramyl-l-alanine (UDP-MurNAc-l-Ala),1a peptidoglycan precursor, by UDP-MurNAc-l-Ala:d-glutamate ligase (EC6.3.2.9). Glutamate racemase (EC 5.1.1.3) catalyzes the racemization of glutamate (1Ashiuchi M. Yoshimura T. Kitamura T. Kawata Y. Nagai J. Gorlatov S. Esaki N. Soda K. J. Biochem. (Tokyo). 1995; 117: 495-498Crossref PubMed Scopus (24) Google Scholar). The genes encoding glutamate racemase are ubiquitously inherited in bacteria, and the Escherichia coli WM335 mutant, in which the enzyme gene was disrupted, requiredd-glutamate for growth (2Doublet P. van Heijenoort J. Mengin-Lecreulx D. J. Bacteriol. 1992; 174: 5772-5779Crossref PubMed Google Scholar). These findings indicate that glutamate racemase provides d-glutamate for peptidoglycan synthesis. Glutamate racemase and its gene hereafter are designated MurI and murI, respectively. The activity of glutamate racemase, however, usually cannot be detected in the cells of bacteria, except in a part of lactobacilli and bacilli (3Ashiuchi M. Tani K. Soda K. Misono H. J. Biochem. (Tokyo). 1998; 123: 1156-1163Crossref PubMed Scopus (87) Google Scholar). Doublet et al. (4Doublet P. van Heijenoort J. Mengin-Lecreulx D. Microb. Drug Resist. 1996; 2: 43-49Crossref PubMed Scopus (18) Google Scholar) showed that MurI of E. coli catalyzed the glutamate racemization only in the presence of UDP-MurNAc-l-Ala and concluded that this peptidoglycan precursor is an absolute activator of the MurI enzyme. On the other hand, the fact that no ribosome-binding sequence is found upstream of the open reading frames in the murI genes of various bacteria and the initial codon of the open reading frames often includes the substitution of the usual ATG for the unusual TTG or GTG (5Yoshimura T. Ashiuchi M. Esaki N. Kobatake C. Choi S.-Y. Soda K. J. Biol. Chem. 1993; 268: 24242-24246Abstract Full Text PDF PubMed Google Scholar, 6Ashiuchi M. Soda K. Misono H. Biosci. Biotechnol. Biochem. 1999; 63: 792-798Crossref PubMed Scopus (35) Google Scholar) suggests that the murI gene is translated on a specific mechanism that is still unidentified. It is generally assumed that MurI is strictly controlled so as to operate only during peptidoglycan synthesis (eventually during cell division). In E. coli clones, the enforced production of MurIs resulted in characteristic changes, such as aberration in nucleoid separation (7Balikó G. Venetianer P. J. Bacteriol. 1993; 175: 6571-6577Crossref PubMed Google Scholar) and suppression of cell proliferation (6Ashiuchi M. Soda K. Misono H. Biosci. Biotechnol. Biochem. 1999; 63: 792-798Crossref PubMed Scopus (35) Google Scholar, 7Balikó G. Venetianer P. J. Bacteriol. 1993; 175: 6571-6577Crossref PubMed Google Scholar), indicating that the attenuation of murI gene expression and the regulation of MurI production are physiologically significant. In the absence ofd-glutamate, the cells of the E. coli WM335 mutant formed filament elongates (2Doublet P. van Heijenoort J. Mengin-Lecreulx D. J. Bacteriol. 1992; 174: 5772-5779Crossref PubMed Google Scholar), as did the mutants defective in the genes involved in cell division. These observations imply that MurI plays a role in cell homeostasis other than in thed-glutamate supply. However, there have been surprisingly few examples focusing on the multifunctionality of MurI until now. In this study, the in vivo and in vitro effects of MurI on some DNA processing in E. coli were investigated. Here we report the novel function of glutamate racemase, i.e. the modulation of DNA gyrase activity. Supercoiled pBR322, calf thymus DNA topoisomerase I, and isopropyl-β-d-thiogalactopyranoside (IPTG) were purchased from TaKaRa Shuzo. Relaxed pBR322 was prepared from supercoiled pBR322 with the DNA topoisomerase I by the method of Ferro and Olivera (8Ferro A.M. Olivera B.M. J. Biol. Chem. 1984; 259: 547-554Abstract Full Text PDF PubMed Google Scholar). UDP-MurNAc-l-Ala was prepared from the cell extract of E. coli WM335 according to the method described previously (6Ashiuchi M. Soda K. Misono H. Biosci. Biotechnol. Biochem. 1999; 63: 792-798Crossref PubMed Scopus (35) Google Scholar). A novobiocin-Sepharose resin was prepared by the method of Nakanishi et al. (9Nakanishi A. Oshida T. Matsushita T. Imajoh-Ohmi S. Ohnuki T. J. Biol. Chem. 1998; 273: 1933-1938Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar). All other chemicals were of analytical grade. E. coli JM109 was purchased from TaKaRa Shuzo. E. coli WM335 mutant was a kind gift from Prof. Dr. W. Messer of the Max Planck Institute for Molecular Genetics. Vector plasmids pKK223-3 and pTrc99A, both of which contain an ampicillin (Amp) resistance gene, were obtained from Amersham Biosciences, and pACYC184 that carries a tetracycline (Tet) resistance gene was from Nippon Gene. Both plasmids pGR2 and pGR3 were constructed according to the strategy described previously (5Yoshimura T. Ashiuchi M. Esaki N. Kobatake C. Choi S.-Y. Soda K. J. Biol. Chem. 1993; 268: 24242-24246Abstract Full Text PDF PubMed Google Scholar) and used for analysis of the function of glutamate racemase. For the production of E. coli DNA gyrase (GyrAB), plasmid pGYRA, a pTrc99A having the gyrA gene, and plasmid pGYRB, a pTrc99A having the gyrB gene, were designed by the method of Mizuuchi et al. (10Mizuuchi K. Mizuuchi M. O'Dea M.H. Gellert M. J. Biol. Chem. 1984; 259: 9199-9201Abstract Full Text PDF PubMed Google Scholar). To obtain E. coliDNA toposiomerase IV (ParCE), plasmid pPARC, a pTrc99A having theperC gene, and plasmid pPARE, a pTrc99A having theparE gene, were prepared by the method of Kato et al. (11Kato J. Suzuki H. Ikeda H. J. Biol. Chem. 1992; 267: 25676-25684Abstract Full Text PDF PubMed Google Scholar). E. coli cells were transformed with the vector plasmid DNA by the CaCl2 method (12Sambrook J. Fritsch E.F. Maniatis T. Molecular Cloning: A Laboratory Manual. 2nd Ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1989Google Scholar) or the TSS method (13Chung C.T. Niemela S.L. Miller R.H Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 2172-2175Crossref PubMed Scopus (1162) Google Scholar) and grown on the LB media containing appropriate antibiotics (1Ashiuchi M. Yoshimura T. Kitamura T. Kawata Y. Nagai J. Gorlatov S. Esaki N. Soda K. J. Biochem. (Tokyo). 1995; 117: 495-498Crossref PubMed Scopus (24) Google Scholar, 12Sambrook J. Fritsch E.F. Maniatis T. Molecular Cloning: A Laboratory Manual. 2nd Ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1989Google Scholar). Plasmid copy numbers were determined densitometrically with a Digital Science EDAS 120 LE system (Invitrogen) by the method of Projan et al. (14Projan S.J. Carleton S. Novick R.P. Plasmid. 1983; 9: 183-190Crossref Scopus (216) Google Scholar). Glutamate racemase was assayed by the method described previously (3Ashiuchi M. Tani K. Soda K. Misono H. J. Biochem. (Tokyo). 1998; 123: 1156-1163Crossref PubMed Scopus (87) Google Scholar). One unit of MurI was defined as the amount of enzyme that catalyzed the formation of 1 μmol ofl-glutamate per min. DNA gyrase was assayed as follows. The reaction mixture (20 μl) comprised 2 μmol of Tris-HCl (pH 8.0), 1.4 μmol of KCl, 0.2 μmol of MgCl2, 0.1 μmol of ATP, 0.1 μmol of spermidine hydrochloride, 0.1 μmol of dithiothreitol, 2 nmol of EDTA, 0.2 mg of glycerol, 0.5 μg of relaxed pBR322, and enzyme. The enzyme was replaced with water in a blank. The reaction was essentially performed at 37 °C for 1 h and terminated by the phenol-chloroform extraction (12Sambrook J. Fritsch E.F. Maniatis T. Molecular Cloning: A Laboratory Manual. 2nd Ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1989Google Scholar). The reaction mixture was electrophoresed in 1% (w/v) agarose gel (14 × 14 cm) submerged in TPE buffer (90 mm Tris-phosphate, 2 mm EDTA, pH 8.0) at 4 V/cm at 25 °C for 2.5 h. Gel was stained with ethidium bromide (1 μg/ml). The activity was estimated from both an increase in the density of bands corresponding to the supercoiled DNA thus formed (10Mizuuchi K. Mizuuchi M. O'Dea M.H. Gellert M. J. Biol. Chem. 1984; 259: 9199-9201Abstract Full Text PDF PubMed Google Scholar) and a decrease in that of the relaxed DNA substrate, which were quantitated with the Digital Science EDAS 120 LE system. One unit of GyrAB was defined as the amount of enzyme that converted one-half of relaxed pBR322 into the supercoils per 30 min. DNA topoisomerases I (TopA) and IV (ParCE) were assayed in the absence (15Bhaduri T. Bagui T.K. Sikder D. Nagaraja V. J. Biol. Chem. 1998; 273: 13925-13932Abstract Full Text Full Text PDF PubMed Scopus (60) Google Scholar) and the presence of ATP (11Kato J. Suzuki H. Ikeda H. J. Biol. Chem. 1992; 267: 25676-25684Abstract Full Text PDF PubMed Google Scholar), respectively. One unit of both topoisomerases was defined as the amount of enzyme required to fully relax 0.5 μg of supercoiled pBR322 per h. E. coli MurI was purified by the method described previously (1Ashiuchi M. Yoshimura T. Kitamura T. Kawata Y. Nagai J. Gorlatov S. Esaki N. Soda K. J. Biochem. (Tokyo). 1995; 117: 495-498Crossref PubMed Scopus (24) Google Scholar). The specific activity was 9.6 units mg−1. Protein concentrations were determined spectrophotometrically using its theoretical molar extinction (ε278 = 22,280 m−1cm−1) (1Ashiuchi M. Yoshimura T. Kitamura T. Kawata Y. Nagai J. Gorlatov S. Esaki N. Soda K. J. Biochem. (Tokyo). 1995; 117: 495-498Crossref PubMed Scopus (24) Google Scholar, 4Doublet P. van Heijenoort J. Mengin-Lecreulx D. Microb. Drug Resist. 1996; 2: 43-49Crossref PubMed Scopus (18) Google Scholar). Since overproduction of DNA gyrase into an active form, i.e.a GyrA2B2 heterotetramer (molecular mass, 373,767 Da), results in strong inhibition of cell growth in bacteria,E. coli GyrAB was prepared by reconstitution of the A subunit (GyrA, 96,975 Da) and the B subunit (GyrB, 89,893 Da) (10Mizuuchi K. Mizuuchi M. O'Dea M.H. Gellert M. J. Biol. Chem. 1984; 259: 9199-9201Abstract Full Text PDF PubMed Google Scholar). Cell lyses with lysozyme and Brij-58, removal of the DNA extracted with streptomycin, and precipitation of proteins with ammonium sulfate were conducted according to the method of Nakanishi et al. (9Nakanishi A. Oshida T. Matsushita T. Imajoh-Ohmi S. Ohnuki T. J. Biol. Chem. 1998; 273: 1933-1938Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar). The GyrA preparation from E. coli JM109/pGYRA cells (110 mg of protein) was applied to a Bio-Rad Bio-Logic FPLC system equipped with a Bio-Scale DEAE anion-exchange column (volume, 20 ml; Bio-Rad) (3Ashiuchi M. Tani K. Soda K. Misono H. J. Biochem. (Tokyo). 1998; 123: 1156-1163Crossref PubMed Scopus (87) Google Scholar); the fraction eluted with 0.4 m NaCl contained only the GyrA protein. The GyrB preparation from E. coli JM109/pGYRB cells (88 mg of protein) was subjected to affinity chromatography on a novobiocin-Sepharose column (1.5 × 7 cm; volume, 12.5 ml) (9Nakanishi A. Oshida T. Matsushita T. Imajoh-Ohmi S. Ohnuki T. J. Biol. Chem. 1998; 273: 1933-1938Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar); the fraction eluted with 2 m KCl plus 5 m urea contained only the GyrB protein. The enzyme was reconstituted by incubation of the purified GyrA and GyrB at an equimolar ratio at 25 °C for 30 min in the presence of 0.1 mm ATP. The specific activity was 3.8 × 104 units mg−1. Protein concentrations were determined using its theoretical molar extinction (ε278 = 248,200m−1 cm−1) (10Mizuuchi K. Mizuuchi M. O'Dea M.H. Gellert M. J. Biol. Chem. 1984; 259: 9199-9201Abstract Full Text PDF PubMed Google Scholar). E. coli ParCE was purified by the method of Kato et al. (11Kato J. Suzuki H. Ikeda H. J. Biol. Chem. 1992; 267: 25676-25684Abstract Full Text PDF PubMed Google Scholar). The specific activity was 3.3 × 103units mg−1. Protein concentrations were determined using its theoretical molar extinction (ε278 = 229,920m−1 cm−1) (11Kato J. Suzuki H. Ikeda H. J. Biol. Chem. 1992; 267: 25676-25684Abstract Full Text PDF PubMed Google Scholar). To examine whether the production of MurI affects DNA processing in E. coli, we first assayed the plasmid copy numbers in several E. coliclones. In this experiment, both plasmid pGR2, a pKK223-3 having themurI gene isolated from the E. coli chromosome, and plasmid pGR3, a pKK223-3 having the designed murI gene in which a typical ribosome-binding sequence was introduced (5Yoshimura T. Ashiuchi M. Esaki N. Kobatake C. Choi S.-Y. Soda K. J. Biol. Chem. 1993; 268: 24242-24246Abstract Full Text PDF PubMed Google Scholar), were used. The plasmid pKK223-3, the multiplication of which is entirely dependent on the replication machineries of the host cells, i.e. the θ mechanism (16Inoue N. Uchida H. J. Bacteriol. 1991; 173: 1208-1214Crossref PubMed Google Scholar), was used as a control (usually over 20 copies). As shown in Fig. 1, the enforced production of MurI by the use of the plasmid pGR3 resulted in a dramatic decrease in the copy numbers in both E. coliJM109 and WM335 clone cells (one to several copies). The result also showed that a decreased plasmid copy number in the E. coliWM335 mutant was restored by complementation of the nativemurI gene with the plasmid pGR2. A decline in negative superhelicity of the plasmid in the MurI overproducer, E. coli JM109/pGR3, but not of those in E. coli JM109/pKK223–3 and JM109/pGR2 was observed by the use of chloroquine (data not shown), and this was consistent with what Balikó and Venetianer (7Balikó G. Venetianer P. J. Bacteriol. 1993; 175: 6571-6577Crossref PubMed Google Scholar) had proposed. It thus seems likely that the replication of ColE1-type plasmids such as pKK223-3 is difficult to complete properly in E. coli cells exhibiting aberration of the production of MurI due to deficiency in the controls of DNA superhelical density (11Kato J. Suzuki H. Ikeda H. J. Biol. Chem. 1992; 267: 25676-25684Abstract Full Text PDF PubMed Google Scholar). MurI would also influence the replication of chromosome, since the overexpression and disruption of themurI gene allowed the appearance of abnormal cell growth in E. coli (2Doublet P. van Heijenoort J. Mengin-Lecreulx D. J. Bacteriol. 1992; 174: 5772-5779Crossref PubMed Google Scholar, 6Ashiuchi M. Soda K. Misono H. Biosci. Biotechnol. Biochem. 1999; 63: 792-798Crossref PubMed Scopus (35) Google Scholar, 7Balikó G. Venetianer P. J. Bacteriol. 1993; 175: 6571-6577Crossref PubMed Google Scholar). While examining characteristics of the E. coli WM335 mutant, we found that its genetic competence declined dramatically compared with that of E. coli JM109, which has an intact murI gene, and, further, that such a decline in genetic competence of the mutant was recovered by the complementation of the murI gene with the plasmid pGR2 (Table I). Chandler and Smith (17Chandlar M.S. Smith R.A. Gene (Amst.). 1996; 169: 25-31Crossref PubMed Scopus (17) Google Scholar) recently reported that, in bacteria, TopA catalyzing the reverse of the DNA gyrase reaction was essential for the development of genetic competence. The relationship between the development of the genetic competence and the intracellular activity for DNA supercoiling was demonstrated. Our results indicated that MurI affects the intracellular activity for DNA supercoiling.Table ICompetency of E. coli WM335 mutant defective in the glutamate racemase geneE. colirecipientsVectorsaFifteen ng of the DNA was used for the transformation experiment.CompetencyCaCl2bAn E. colitransformation method with CaCl2; Sambrook et al.(12).TSScAn E. colitransformation method with polyethylene glycol; Chung et al.(13).JM109pKK223–37.8 × 1031.2 × 105pACYC1844.0 × 1039.2 × 104WM335pKK223–307.8pACYC18405.4JM109/pGR2pACYC1842.2 × 1033.4 × 104WM335/pGR2pACYC1840.8 × 1026.6 × 102E. coli JM109 transformants with pKK223–3 or pACYC184,E. coli WM335 transformants with pKK223–3 or pACYC184, and both E. coli JM109/pGR2 and WM335/pGR2 transformants with pACYC184 were grown at 37 °C for 24 h on an LB plate (12Sambrook J. Fritsch E.F. Maniatis T. Molecular Cloning: A Laboratory Manual. 2nd Ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1989Google Scholar) containing Amp (50 μg/ml) or Tet (12.5 μg/ml), a d-glutamate-LB plate (2Doublet P. van Heijenoort J. Mengin-Lecreulx D. J. Bacteriol. 1992; 174: 5772-5779Crossref PubMed Google Scholar) containing Amp or Tet, and an LB plate containing Amp, Tet, and 0.1 mm IPTG, respectively. Competency of each E. coli recipient was assessed on the basis of the number of colonies being formed. Data are representative of an average of eight independent experiments.a Fifteen ng of the DNA was used for the transformation experiment.b An E. colitransformation method with CaCl2; Sambrook et al.(12Sambrook J. Fritsch E.F. Maniatis T. Molecular Cloning: A Laboratory Manual. 2nd Ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1989Google Scholar).c An E. colitransformation method with polyethylene glycol; Chung et al.(13Chung C.T. Niemela S.L. Miller R.H Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 2172-2175Crossref PubMed Scopus (1162) Google Scholar). Open table in a new tab E. coli JM109 transformants with pKK223–3 or pACYC184,E. coli WM335 transformants with pKK223–3 or pACYC184, and both E. coli JM109/pGR2 and WM335/pGR2 transformants with pACYC184 were grown at 37 °C for 24 h on an LB plate (12Sambrook J. Fritsch E.F. Maniatis T. Molecular Cloning: A Laboratory Manual. 2nd Ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1989Google Scholar) containing Amp (50 μg/ml) or Tet (12.5 μg/ml), a d-glutamate-LB plate (2Doublet P. van Heijenoort J. Mengin-Lecreulx D. J. Bacteriol. 1992; 174: 5772-5779Crossref PubMed Google Scholar) containing Amp or Tet, and an LB plate containing Amp, Tet, and 0.1 mm IPTG, respectively. Competency of each E. coli recipient was assessed on the basis of the number of colonies being formed. Data are representative of an average of eight independent experiments. The intracellular activity for DNA supercoiling was practically determined in the balance of the activities of TopA and GyrAB (18Snoep J.L. van der Weijden C.C. Andersen H.W. Westerhoff H.V. Jensen P.R. Eur. J. Biochem. 2002; 269: 1662-1669Crossref PubMed Scopus (91) Google Scholar), and net DNA supercoiling activity of bacterial cells can be determined by using the DNA gyrase assay system because the cell extracts essentially contain both the activities of GyrAB, which introduces negative supercoils into DNAs, and of TopA, which unwinds the supercoiled DNAs. In this study, comparatively low and high DNA supercoiling activities were observed in the cells of E. coli JM109/pGR3 and WM335, respectively (Fig. 2). The supercoiling activity ofE. coli JM109/pGR3 cells was one-tenth that of E. coli JM109/pKK223-3 cells; the intracellular activities ofE. coli WM335 and E. coli WM335/pGR3 were 2-fold higher and 4-fold less than that of E. coli WM335/pGR2, respectively. In contrast, there was apparently little difference in the activities of TopA among the E. coli clones used: JM109/pKK223-3, 31 ± 7 (units mg−1); JM109/pGR2, 34 ± 12; JM109/pGR3, 39 ± 14; WM335/pKK223-3, 28 ± 13; WM335/pGR2, 30 ± 5; WM335/pGR3, 33 ± 6. It was thus suggested that MurI affects the intracellular activity of GyrAB inE. coli. To clarify if MurI modulates GyrAB, we followed a change in the apparent activity of GyrAB in the coexistence of various concentrations of MurI. Bovine serum albumin, instead of E. coli MurI, was added to the DNA gyrase assay system as a negative control (Fig.3, open diamonds). As shown in Fig. 3 A, GyrAB was inhibited by E. coli MurI in the presence of UDP-MurNAc-l-Ala (closed circles) but not both enantiomers of glutamate (open triangles and open squares). Fig. 3 A also reveals that GyrAB probably forms a 1:1 molar stoichiometric inactive complex with MurI activated by UDP-MurNAc-l-Ala. The apparent inhibition constant K i(app) of MurI against GyrAB was estimated to be 6.8 nm (Fig.3 B) On the other hand, E. coli MurI (0.5 nm) kept the activity of glutamate racemase even in the presence of a 250-fold molar excess of GyrAB. ParCE, a GyrAB paralogue, participates in the chromosome partitioning (11Kato J. Suzuki H. Ikeda H. J. Biol. Chem. 1992; 267: 25676-25684Abstract Full Text PDF PubMed Google Scholar) that occurs after the DNA replication in cell division. The enzyme (20 nm), however, remained active when coexisting with abundant MurI (125 nm). These suggest that DNA gyrase is the target of glutamate racemase in E. coli. DNA gyrase (alternatively, bacterial DNA topoisomerase II) is a pivotal enzyme for various types of DNA processing, including DNA replication and gene expression, and the strict modulation of the intracellular activity is indispensable for proper cell division (10Mizuuchi K. Mizuuchi M. O'Dea M.H. Gellert M. J. Biol. Chem. 1984; 259: 9199-9201Abstract Full Text PDF PubMed Google Scholar). The first endogenous DNA gyrase modulator, designated as GyrI (9Nakanishi A. Oshida T. Matsushita T. Imajoh-Ohmi S. Ohnuki T. J. Biol. Chem. 1998; 273: 1933-1938Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar), was identified from E. coli; it was identical to the SbmC protein, which had been previously characterized as a member of the DNA repair system. Because glutamate racemase, however, has been classified into the group of d-amino acid-metabolic enzymes but not into that of DNA-processing proteins, such as the GyrI protein, and does not have any consensus sequences that can be observed in the DNA-binding motifs (3Ashiuchi M. Tani K. Soda K. Misono H. J. Biochem. (Tokyo). 1998; 123: 1156-1163Crossref PubMed Scopus (87) Google Scholar, 5Yoshimura T. Ashiuchi M. Esaki N. Kobatake C. Choi S.-Y. Soda K. J. Biol. Chem. 1993; 268: 24242-24246Abstract Full Text PDF PubMed Google Scholar, 6Ashiuchi M. Soda K. Misono H. Biosci. Biotechnol. Biochem. 1999; 63: 792-798Crossref PubMed Scopus (35) Google Scholar), the discovery of the novel function of glutamate racemase as the endogenous inhibitor for DNA gyrase was quite unexpected. Nevertheless, the existence of such multifunctional glutamate racemases would be advantageous in the peculiar processes seen only during bacterial cell division, in which the synthesis of the peptidoglycan molecules (as an essential constituent of the septum and cell walls) follows DNA processing (e.g. the decatenation of daughter DNAs catalyzed by DNA gyrase) (11Kato J. Suzuki H. Ikeda H. J. Biol. Chem. 1992; 267: 25676-25684Abstract Full Text PDF PubMed Google Scholar), for example. In bacterial glutamate racemases, E. coli MurI is functionally particular; it is led into an active form by UDP-MurNAc-l-Ala (4Doublet P. van Heijenoort J. Mengin-Lecreulx D. Microb. Drug Resist. 1996; 2: 43-49Crossref PubMed Scopus (18) Google Scholar). E. coli MurI thus operates only during the peptidoglycan synthesis and changes again into an inactive form with exhaustion of UDP-MurNAc-l-Ala after peptidoglycan synthesis. As shown in Fig. 3 A, UDP-MurNAc-l-Ala was essential for the inhibition of GyrAB by E. coli MurI, suggesting that the function of DNA gyrase is suppressed by glutamate racemase at the early stage of the septation in E. coli cells. It seems likely that this step is important to avoid the appearances of abnormal DNA replication and decatenation due to excessive DNA gyrase activity. We have identified the glutamate racemase isozyme of Bacillus subtilis, YrpC (lately B. subtilis MurI), which was not modulated by UDP-MurNAc-l-Ala (6Ashiuchi M. Soda K. Misono H. Biosci. Biotechnol. Biochem. 1999; 63: 792-798Crossref PubMed Scopus (35) Google Scholar). Unlike E. coli MurI,B. subtilis MurI, whether coexisting with UDP-MurNAc-l-Ala or not, inhibited DNA gyrase. 2M. Ashiuchi, E. Kuwana, K. Komatsu, K. Soda, and H. Misono, manuscript in preparation. Although Doubletet al. (4Doublet P. van Heijenoort J. Mengin-Lecreulx D. Microb. Drug Resist. 1996; 2: 43-49Crossref PubMed Scopus (18) Google Scholar) previously mentioned a toxicity ofd-glutamate produced by glutamate racemase on the metabolism and growth of bacterial cells, our observations show that an active form of glutamate racemase itself but not its reaction product,i.e. d-glutamate, is required for the inhibition of DNA gyrase, a most crucial enzyme in bacterial cell homeostasis. Unlike usual amino acid racemases, glutamate racemase contains no coenzymes such as PLP (3Ashiuchi M. Tani K. Soda K. Misono H. J. Biochem. (Tokyo). 1998; 123: 1156-1163Crossref PubMed Scopus (87) Google Scholar). There is no sequence homology between glutamate racemase and usual amino acid racemases (6Ashiuchi M. Soda K. Misono H. Biosci. Biotechnol. Biochem. 1999; 63: 792-798Crossref PubMed Scopus (35) Google Scholar). The catalytic efficiency of glutamate racemase is exceedingly low compared with those of PLP-dependent amino acid racemases (5Yoshimura T. Ashiuchi M. Esaki N. Kobatake C. Choi S.-Y. Soda K. J. Biol. Chem. 1993; 268: 24242-24246Abstract Full Text PDF PubMed Google Scholar). Interestingly, the orthologues of glutamate racemase have been found from various organisms, including peptidoglycan-less organisms such as archaea, plants, and humans (19Strausberg, R. (2000) The Human Tumor Gene Index for Cancer Genome Anatomy Project (CGAP) at National Cancer Institute(www.ncbi.nlm. nih. gov/ncicgap).Google Scholar). Most of these proteins, however, have not been characterized yet because of their loss of function as glutamate racemase. It was recently reported that the glutamate racemase orthologue-encoding gene was expressed to excess in human carcinoma (19Strausberg, R. (2000) The Human Tumor Gene Index for Cancer Genome Anatomy Project (CGAP) at National Cancer Institute(www.ncbi.nlm. nih. gov/ncicgap).Google Scholar), in which abnormal DNA replication frequently took place during cell division. On the other hand, among other PLP-independent amino acid racemases, the proline racemase from Trypanosoma cruziwas found to exhibit the function as an immune B-cell mitogen (20Reina-San-Martin B. Degrave W. Rougeot C. Cosson A. Chamond N. Cordeiro-Da-Silva A. Arala-Chaves M. Coutinho A. Minoprio P. Nat. Med. 2000; 6: 890-897Crossref PubMed Scopus (134) Google Scholar). Some PLP-independent amino acid racemases including glutamate racemase have possibly evolved from ancient proteins that had not been originally involved in amino acid metabolism. It remains to be investigated whether glutamate racemase-like proteins substantially modulate DNA topoisomerases, as their detailed structural analyses could provide insights into the molecular evolution ofd-amino acid-metabolic enzymes, deepen the understanding of the mechanism of cell division, and result in the design of novel pharmaceuticals.
Referência(s)