Evolutional Origin of Bacterial Glutamate Racemase Title …repository.kulib.kyoto-u.ac.jp/dspace/bitstream/2433/... · 2012. 7. 12. · 46 feR Annual Report, Vol. 1, 1994 Evolutional

TitleEvolutional Origin of Bacterial Glutamate Racemase(MOLECULAR BIOFUNCTION - Molecular MicrobialScience)

Author(s) Soda, Kenji; Esaki, Nobuyoshi; Yoshimura, Tohru

Citation ICR annual report (1995), 1: 46-47

Issue Date 1995-03

URL http://hdl.handle.net/2433/65657

Right

Type Article

Textversion publisher

Kyoto University

46 feR Annual Report, Vol. 1, 1994

Evolutional Origin of Bacterial Glutamate Racemase

Kenji Soda, Nobuyoshi Esaki and Tohru Yoshimura

Glutamate racemase (EC 5.1.1.3), an enzyme of microbial origin, shows significant sequence similarity withmammalian myoglobins, in particular in the regions corresponding to the E and F helices, which constitute theheme binding pocket of myoglobins. Glutamate racemase binds tightly an equimolar amount of hemin leadingto loss of racemase activity. Although this enzyme shows sequence similarity with aspartate racemase, the latterdoes not bind hemin. Neither racemase has cofactors, but contain essential cysteine residues.

Keywords: Specific inhibition by hemin! E and F helices of myoglobin/ Essential cysteine residues

D-Glutamate is an essential component of peptido-glycans of bacterial cell walls, and is produced from L-glutamate by glutamate racemase (EC 5.1.1.3) or from a-ketoglutarate by D-amino acid aminotransferase (EC2.6.1.21) [1]. Most amino acid racemases, such asalanine racemase (EC 5.1.1.1), require pyridoxal 5'-phosphate (PLP) as a coenzyme, and the racemasereaction is facilitated by formation of internal andexternal Schiff base intermediates. In contrast, a fewother amino acid racemases, such as glutamate racemase[2] and aspartate racemase (EC 5.1.1.13) [3], areindependent of any cofactor, and contain no carbonylmoieties or metals. Their reaction mechanisms have notbeen elucidated. We have cloned the glutamateracemase gene from P. pentosaceus, expressed it in E. coliand purified the enzyme to homogeneity [4]. Thepurified enzyme contains no co-factors, but does haveessential cysteine residues.

Glutamate racemase showed considerable sequencesimilarity with aspartate racemase. Linear alignment oftheir sequences by introducing gaps to maximize identityrevealed an overall similarity of 14%. However,

sequence similarity in the internal region (69-192 of theglutamate racemase sequence) was much higher; 31 of124 residues being common. If the mutationally allowedsubstitutions for similar residues were taken into conside-ration, the similarity score increased to 68% in thisregion. In particular, the sequences around the twocysteine residues C4Cys and 184Cys of glufamateracemase) were highly similar to each other. Bothenzymes contain an essential cysteine residue as reportedpreviously, suggesting that either 74Cys or 184Cys, or bothplay an essential role in catalysis.

Glutamate racemase shows also a high sequencesimilarity with bovine myoglobin among various proteinsregistered in the National Biomedical ResearchFoundation and the Swissprot protein sequence data-banks. The analogous region between glutamateracemase and myoglobin occur mainly in the regionbetween 46Phe and 150Gly of bovine myoglobin whichcorresponds to the region from 92Yal to 183Gly ofglutamate racemase. Twenty-seven of the 92 residues ofglutamate racemase are common to the correspondingresidues of the myoglobin. The similarity score is 52%

BIOFUNCTIONAL MOLECULES -Molecular Microbial Science-

Associate Instructor:HIRASAWA, ToshikoStudents:TCHORZEWSKI, Marek (DC)LIU, liquan (DC)KUROKAWA, Yoichi (DC)JHEE, Kwang-hwan (DC)KISHIMOTO, Kazuhisa (DC)GUTTIERREZ,Aldo Francisco (DC)

CHOO, Dong-Won (DC)PARK, Chung (DC)LIU, Lidong (DC)AOKI, Tomoko (MC)

Instructor KITAMURA, Tae (MC)KURIHARA, Tatsuo MIHARA, Hisaaki (MC)

(D Eng) KURONO, Takeshi (MC)MIYAKE, Hitoki (MC)

InstructorYOSHIMURA, Tohru

(D Agr)

Associate ProfessorESAKI, Nobuyoshi

(D Agr)

ProfessorSODA, Kenji

(D Agr)

Scope of researchStructure and function of biocatalysts, in particular, pyridoxal enzymes and NAD enzymes are studied to elucidate thedynamic aspects of the fine mechanism for their catalysis in the light of recent advances in gene technology, proteinengineering and crystallography. In addition, the metabolism and biofunction of selenium and some other traceelements are also investigated. Development and application of new biomolecular functions of microorganisms arealso studied to open the door to new fields of biotechnology. For example, molecular structures and functions ofthermostable enzymes and their application are under investigation.

in this region, if the similar residues of permissiblemutational substitution are taken into account. Theamino acid sequences of myoglobins from various sourcesare highly conserved. The abalone myoglobin showshigh sequence similarity with human indoleamine 2,3-dioxy-genase, but not with other myoglobins. We foundno significant sequence similarity between the abalonemyoglobin and glutamate racemase. Similarity scoresbetween glutamate racemase and the other myoglobinswere: 21-27% identity in the range of the 92 amino acidresidues. Cyanobacterial myoglobin from Nostoccommune showed the lowest sequence similarity (21 %)with glutamate racemase. Significant sequence simi-larities were also found between glutamate racemase andother globin family proteins such as hemoglobins in thissame region. Bacterial hemoglobin from Vitreoscillashows the lowest sequence similarity with glutamateracemase among the various hemoglobins examined.

Proteins analogous to bovine myoglobin in primarystructure were also searched by means of the same data-banks. The sequence similarity is dependent on the kindof proteins and their sources: myoglobins from othersources, 38-85%; a and p-chains of mammalianhemoglobins, 21-31 %; Vitreoscilla hemoglobin, 24%; N.commune myoglobin, 16%; glutamate racemase, 26% (inthe range between 46Phe and 150Gly of bovinemyoglobin). Bovine myoglobin shows higher sequencesimilarity with glutamate racemase than prokaryoticmyoglobin and hemoglobin. Aspartate racemase wasalso analogous to bovine myoglobin in the region from102Ile to 196Gly corresponding to that from 46Phe to 150Gly

of bovine myoglobin: 14 residues were common betweenthe two proteins. However, this sequence similarity wasmuch lower than that found between glutamate racemaseand bovine myoglobin.

The analogous range (residue numbers, 46-150) ofbovine myoglobin contains the regions corresponding toE and F helices, which constitute the heme bindingpocket. E7 of the E helix of bovine myoglobin, 64His,which is essential in binding molecular oxygen, is replacedby GIn in the bacterial myoglobin and the bacterialhemoglobin. An analogous GIn occurs as HOGln inglutamate racemase. Moreover, 68Yal of Ell, which ishighly conserved among globin family proteins, is alsoconserved as 114Yal. Accordingly, we examined theinteraction of glutamate racemase and aspartate racemasewith hemin. When the enzymes were assayed in thepresence of various concentrations of hemin, onlyglutamate racemase was inhibited by hemin. Theinhibition was concentration-dependent. A plot ofreciprocal of glutamate racemase activity against heminconcentrations showed that hemin produced a mixed typeinhibition. The K i value for hemin was estimated to beabout 3.7 mM from these data. When glutamateracemase was incubated with hemin at variousconcentrations, a stoichiometric complex was formed andisolated by gel filtration. However, no appreciableamount of hemin was bound with aspartate racemaseunder the same conditions. The complex of glutamate

47

racemase with hemin was reduced with dithionite. TheESR spectrum of the oxidized form resembled that ofhemoglobin under the same conditions. Thus,glutamate racemase resembles hemoglobins in having ahemin binding pocket, in which two nitrogen atoms ofsome amino acid residues are probably ligated to iron inthe coordination complex with hemin. Hemin inhibitsglutamate racemase either by binding near the active siteor at some other site where the binding causes aconformational change of the active site.

Proline racemase, 4-hydroxyproline epimerase anddiaminopimelate epimerase contain an essential cysteineresidue, and show sequence similarity with each other inthe moiety around the cysteine residues. These enzymeshave been proposed to evolve from a common ancestralprotein. Glutamate racemase as well as aspartateracemase also contains an essential cysteine residue, butshows no sequence similarity to these three enzymes.However, a high sequence similarity in the regions of twocysteine residues occurs between glutamate racemase andaspartate racemase. It is suggested that glutamateracemase and aspartate racemase have derived from acommon evolutionary origin which is different from thecommon ancestor for proline racemase, 4-hydroxyprolineepimerase and diaminopimelate epimerase.

The high sequence similarity between glutamateracemase and the globin family proteins, in particularmyoglobins, and formation of its inactive equimolarcomplex with hemin, suggest that the enzyme may bederived from the evolutionary origin of globin familyproteins. Aspartate racemase also may have evolvedfrom the common ancestral protein, but its structure mayhave been altered more extensively than· glutamateracemase by divergence. Lactic acid bacteria may havebeen producing glutamate racemase and aspartateracemase, namely globin family-like proteins, whichdiverged from an ancestral globin protein after the abilityto synthesize hemin was lost. Alternatively, lactic acidbacteria inherently never produced hemin, and acquiredfrom other organisms the gene for the globin familyproteins, which then diverged to glutamate racemase andaspartate racemase. Whatever may be the case,glutamate racemase is the first proven microbial enzymethat is structurally similar to globin family proteins and tostoichiometrically bind hemin to form a catalyticallyinactive complex.

References1. Yoshimura T, Ashiuchi M, Esaki N, Kobatake C,

Choi S and Soda K, 1. Bioi. Chem., 268,24242-24246(1993).

2. Choi S, Esaki N, Yoshimura T and Soda K, J.Biochem., 112, 139-142 (1992).

3. Yamauchi T, Choi S, Okada H, Yohda M, Kumagai. H, Esaki N and Soda K, J. Bioi. Chem., 267,

18361-18364 (1992).4. Choi S, Esaki N, Yoshimura T and Soda K, Protein

Expr. Purif., 2, 90-93 (1991).

icrar1994 54icrar1994 55