Purification and characterization of VanXY C , a D , D -dipeptidase/ D , D -carboxypeptidase in vancomycin-resistant Enterococcus gallinarum BM4174 Adrian H. B. Podmore and Peter E. Reynolds Department of Biochemistry, University of Cambridge, UK VanXY C , a bifunctional enzyme from VanC-phenotype Enterococcus gallinarum BM4174 that catalyses D , D -pepti- dase and D , D -carboxypeptidase activities, was purified as the native protein, as a maltose-binding protein fusion and with an N-terminal tag containing six histidine residues. The kinetic parameters of His 6 –VanXY C were measured for a variety of precursors of peptidoglycan synthesis involved in resistance: for D -Ala- D -Ala, the K m was 3.6 m M and k cat , 2.5 s )1 ;forUDP-MurNAc-L-Ala- D -Glu-L-Lys- D -Ala- D - Ala (UDP-MurNAc-pentapeptide[Ala]), K m was 18.8 m M and k cat 6.2 s )1 ;for D -Ala- D -Ser, K m was 15.5 m M and k cat 0.35 s )1 .His 6 –VanXY C was inactive against the peptido- glycan precursor UDP-MurNAc- L -Ala- D -Glu-L-Lys- D - Ala- D -Ser (UDP-MurNAc-pentapeptide[Ser]). The rate of hydrolysis of the terminal D -Ala of UDP-MurNAc-penta- peptide[Ala] was inhibited 30% by 2 m MD -Ala- D -Ser or UDP-MurNAc-pentapeptide[Ser]. Therefore preferential hydrolysis of substrates terminating in D -Ala would occur during peptidoglycan synthesis in E. gallinarum BM4174, leaving precursors ending in D -Ser with a lower affinity for glycopeptides to be incorporated into peptidoglycan. Muta- tion of an aspartate residue (Asp59) of His-tagged VanXY C corresponding to Asp68 in VanX to Ser or Ala, resulted in a 50% increase and 73% decrease, respectively, of the specif- icity constant (k cat /K m )for D -Ala- D -Ala. This situation is in contrast to VanX in which mutation of Asp68fiAla pro- duced a greater than 200 000-fold decrease in the substrate specificity constant. This suggests that Asp59, unlike Asp68 in VanX, does not have a pivotal role in catalysis. Keywords: vancomycin resistance; D , D -dipeptidase; D , D - carboxypeptidase; Enterococcus gallinarum. Glycopeptide antibiotics are effective Gram-positive anti- bacterial agents that inhibit peptidoglycan synthesis by binding to cell wall precursors terminating in D -Ala- D -Ala [1]. VanA, VanB and VanD phenotypes have acquired resistance to glycopeptides, and synthesize D -Ala- D -lactate depsipeptides [2]. Three proteins are required for resistance: VanH (VanH B ,VanH D ) reduces pyruvate to D -lactate [3,4]; VanA (VanB, VanD) ligases catalyse synthesis of D -Ala- D - lactate [5] and VanX (VanX B ) D , D -dipeptidase inhibits the production of glycopeptide-susceptible precursors by hydrolysing D -Ala- D -Ala [6]. A fourth enzyme, VanY, is a D , D -carboxypeptidase that hydrolyses terminal D -Ala from UDP-MurNAc-pentapeptide [ D -Ala] if D -Ala- D -Ala hydro- lysis by VanX is incomplete [2] but is not necessary for VanA-type resistance [2]. Regulation of expression of the resistance genes is controlled by VanR (VanR B ,VanR D ) andVanS(VanS B ,VanS D ), a two-component regulatory system [7,8]. The VanA and VanB gene clusters are contained on transposons that are integrated either into self-transferable plasmids or the host chromosome [8,9]. VanC-type resistance is defined as intrinsic low-level resistance to vancomycin (2–32 lgÆmL )1 ), but not to teicoplanin [10]. It has been identified in Enterococcus gallinarum, E. casseliflavus and E. flavescens [10,11]. Resist- ance is based on the substitution of the terminal D -Ala in peptidoglycan precursors by D -Ser [12,13]. VanC phenotype glycopeptide resistance is mediated by VanC D -Ala: D -Ser ligase [14], VanXY C D , D -dipeptidase/ D , D -carboxypeptidase [15] and VanT C serine racemase [16]. These three proteins eliminate D -Ala-terminating peptidoglycan precursors and replace the terminal D -Ala- D -Ala with D -Ala- D -Ser. In VanA and VanB phenotypes, the elimination of precursors terminating in D -Ala requires two enzymes, VanX (VanX B ), astrict D , D -dipeptidase [6,17], and VanY (VanY B ), a strict membrane-bound D , D -carboxypeptidase [18]. Both of these enzymes are also active with substrates terminating in D -Ser. VanX is a metallo-protease: its catalytic, substrate-binding and Zn 2+ -binding sites have been characterized by a combination of kinetic [17], crystallographic [19] and site- directed mutagenic studies [20,21]. In this study the substrate specificity of purified VanXY C ,acytoplasmic bifunctional D , D -peptidase and D , D -carboxypeptidase, was characterized kinetically and an investigation initiated of the role of specific residues in determining the substrate selectivity. EXPERIMENTAL PROCEDURES Strains, plasmids and growth conditions Escherichia coli strains were grown in Luria–Bertani medium, and maintained on Luria–Bertani agar (1.5%), with the exception of E. coli JM83 [22] containing deriva- Correspondence to A. H. B. Podmore, R & D Lab, Bioproducts Laboratories (BPL), Dagger Lane, Elstree, WD6 3BX. Tel.: + 44 208 2582200, E-mail: adrian_podmore@hotmail.com Abbreviations: MBP, maltose-binding protein. (Received 28 January 2002, revised 11 April 2002, accepted 19 April 2002) Eur. J. Biochem. 269, 2740–2746 (2002) Ó FEBS 2002 doi:10.1046/j.1432-1033.2002.02946.x tives of pMal-c2 (New England Biolabs), which was grown in TYG medium [1% (w/v) tryptone, 0.5% (w/v) yeast extract, 0.5% (w/v) NaCl, 0.2% (w/v) glucose, pH 7.2] containing 100 lgÆmL )1 ampicillin. Kanamycin (25 lgÆmL )1 ) was added to the medium for E. coli M15 [pREP4] (Qiagen) and ampicillin (100 lgÆmL )1 ) was added to the medium for E. coli M15[pREP4] containing deriva- tives of pQE-30 (Qiagen) and E. coli JM83 containing pAT704 [15]. DNA manipulations Digestion with restriction endonucleases, cloning, isolation of plasmid DNA, ligation and transformation were carried out using standard protocols [23]. Plasmid construction Plasmid pAT704 has been described previously [15]. Plas- mid pAP1 (encoding a maltose binding protein-VanXY C fusion protein) was constructed as follows; Pfu polymerase (Stratagene) was used to amplify vanXY C usingpAT704as template with primers A (5¢-GCTA GGTCTCAATGAAC ACATTACAATT-3¢)andB(5¢-TATG GAATTCTCATG CGAACTGCCTCA-3¢) that included BsaIandEcoRI restriction sites, respectively (underlined). The product was purified, digested with BsaI, treated with DNA poly- merase I large (Klenow) fragment, purified, digested with EcoRI and cloned in pMal-c2 under the control of the tac promoter. Plasmid pAP2 was constructed for production of VanXY C with an N-terminal tag of six histidine residues. Pfu polymerase was used to amplify vanXY C using pAT704 as template with primers C (5¢-CTCA GGATCCAACACA TTACAATTGATCAATA-3¢)andD(5¢-CACT AAGCTT TCATGCGAACTGCCTCAC-3¢) that included BamHI and HindIII restriction sites, respectively (underlined). The product was purified, digested with BamHI and HindIII and cloned in pQE30 under the control of the T5 promoter. Site-directed mutagenesis Mutants D59S and D59A were constructed from a pAP2 template by PCR mutagenesis using the Expand Long Template PCR System (Boehringer Mannheim) and sense/ antisense primer pairs D59S (CGTCTGGTA TCTGGGT AT/AATGTCCTTTTCTAGTCC), D59A (CGTCTG GTA GCTGGGTAT/AATGTCCTTTTCTAGTCC),P/W (AGTTATGAA TGGTGGCATTTTCG/GATACCGGT GATCTCTTG) and Q/V (GGAAAAAGAA GTGCGA CG/GTACGATACCCATCTACC), respectively (mis- match mutations are underlined). The purified PCR pro- ducts were ligated and used to transform E. coli M15[pREP4]. DNA sequence determination DNA sequencing on both strands was carried out by the dideoxynucleotide chain terminator method [24] using fluorescent cycle sequencing with dye-labelled terminators (ABI PrismTM Dye terminator Cycle Sequencing Ready Reaction Kit, PerkinElmer) on a 373 A automated DNA sequencer (PerkinElmer). Protein quantitation Protein concentration was determined by the method of Bradford with bovine serum albumin as standard [25]. Purification of VanXY C A culture of E. coli JM83 containing pAT704(vanXY C )was harvested after induction with 0.5 m M isopropyl thio-b- D - galactoside for 3 h. The bacteria were then washed and disrupted by sonication. After removal of cell debris, purification was attempted using ammonium sulphate fractionation (the majority of the enzyme was present in the 45–60% fraction), followed by ion-exchange chroma- tography on a MonoQ HR5/5 column (Pharmacia) and gel exclusion chromatography on a FPLC Superdex 75 H column. VanXY C eluted with an estimated mass of 42 kDa, approximately twice that of the monomer, suggesting that native VanXY C exists as a dimer. The activity of the partially purified material (Fig. 1A) was only 3% of that present in the original extract. In order to improve recovery and purity, the enzyme was produced as a fusion with the maltose binding protein. A culture of E. coli JM83 containing pAP1(pMal-c2[vanXY C ]) was incubated with 0.5 m M isopropyl thio-b- D -galactoside for 30 min to induce the fusion protein. The bacteria were harvested, washed, broken by sonication and the fusion protein purified by affinity chromatography on an amylose resin column. The yield, based on activity, was  50% and a single band (60 000 kDa) was observed by SDS/PAGE analysis (Fig. 1B). It proved impossible to release VanXY C from MBP-VanXY C by treatment with factor Xa. A second fusion protein (His 6 –VanXY C )wasalso purified by affinity chromatography after induction of E. coli M15[pREP4] containing pAP2(pQE30[vanXY C ]) for 30 min with 0.5 m M isopropyl thio-b- D -galactoside, followed by sonication of the washed bacterial suspension. The broken cell preparation was centrifuged at 43 000 g for 20 min to pellet the cellular debris. The supernatant was removed, and loaded onto a 4-mL Ni 2+ -nitrilotriacetic acid/agarose column. The column was washed with 200 mL 50 m M 1,3 bi[tris(hydroxymethyl)-methylamino] propane (pH 7.5), 30 m M imidazole, 300 m M NaCl, and His 6 – VanXY C was eluted with the same buffer in which the imidazole concentration had been increased to 250 m M . SDS/PAGE analysis showed a single band of  23 kDa (Fig. 1C), whereas gel filtration on FPLC Superdex 75 column resulted in a peak with an apparent molecular mass of 44 kDa. This method yielded 3–4 mg of His 6 –VanXY C from a culture volume of 1 L for enzyme characterization studies and was also used for purification of His 6 –VanXY C in studies of site-directed mutagenesis. Assay of D , D -dipeptidase and D , D -carboxypeptidase activity The method for assaying D -Ala- D -Ala dipeptidase activity of VanXY C was based on that of Reynolds et al.[6]. Twenty microliters of 150 m M 1,3 bi[tris(hydroxymethyl)- methylamino] propane (pH 7.5) containing substrate ( D -Ala- D -Ala, unless stated otherwise) was mixed with 10 lL of enzyme preparation and incubated at 37 °C. Samples were withdrawn at suitable time intervals and Ó FEBS 2002 Purification and characterization of VanXYC (Eur. J. Biochem. 269) 2741 assayed using the D -amino acid oxidase assay [6], except for kinetic studies when the modified cadmium-ninhydrin method [17] was used, with D -alanine or D -serine as standards. The sensitivity ranges of the D -amino acid oxidase and cadmium ninhydrin assays were 2–20 nmol D -alanine or 4–40 nmol D -serine, and 10–80 nmol D -alanine or D -serine, respectively. Effect of divalent cations and EDTA on D , D -dipeptidase and D , D -carboxypeptidase activity The effect of divalent cations on the actual D -amino acid oxidase assay was investigated as follows. Ten microliteres of 10 m M Me 2+ (Fe 2+ ,Cu 2+ ,Mn 2+ ,Co 2+ ,Zn 2+ ,Ni 2+ , Mg 2+ as the metal chlorides) in 150 m M 1,3 bi[tris (hydroxymethyl)-methylamino] propane (pH 7.5) was mixed with 10 lLofVanXY C , and incubated on ice for 15 min. D -Ala (20 nmol) was then added, to give a final volume of 30 lL. These samples were assayed using the D -amino acid oxidase assay. Mn 2+ ,Co 2+ and Cu 2+ caused a decrease in the absorbance at 460 nm (25% for Mn 2+ and Cu 2+ 37% for Co 2+ ), but this would not have masked a onefold or greater stimulation of D , D -dipeptidase activity. To study the effect of metal ions on activity of VanXY C ,10lLof10m M Me 2+ in 150 m M 1,3 bi[tris (hydroxymethyl)-methylamino] propane was mixed with 10 lLVanXY C and incubated on ice for 15 min. Next, 10 lLof10m MD -Ala- D -Ala was added, incubated at 37 °C for 30 min, and assayed using the D -amino acid oxidase assay. This experiment was repeated to determine the effect of divalent cations on D , D -carboxypeptidase activity, using Ni 2+ at concentrations of 1 and 5 m M ,and Zn 2+ at concentrations of 0.05 and 0.8 m M . The nucleo- tide peptide substrate UDP-MurNAc-pentapeptide[Ala] (3.3 m M )wasusedinsteadof D -Ala- D -Ala. EDTA did not affect the D -amino acid oxidase assay at a concentration of 0.05 m M and below. To test the effect of EDTA on enzyme activity VanXY C was mixed with various concentrations of EDTA (0.01, 0.05, 0.1, 1.0 and 5.0 m M ) and incubated on ice for 1 h. Samples were diluted with 50 m M 1,3 bi[tris(hydroxymethyl)-methylamino] propane (pH 7.5) to reduce the EDTA concentration to 0.05 m M . 10 lLEDTA-treatedVanXY C was mixed with 10 lLof 10 m M Me 2+ in 150 m M 1,3 bi[tris(hydroxymethyl)-meth- ylamino] propane and 10 lL10m MD -Ala- D -Ala, incuba- ted at 37 °C for 30 min, and assayed for D , D -dipeptidase activity using the D -amino acid oxidase assay. Kinetic analysis of VanXY C His 6 –VanXY C (2.25 · 10 )7 M) was incubated at 37 °C with various concentrations of D -Ala- D -Ala (2, 5, 10, 15, 20, 30, and 40 m M ) in 100 m M 1,3 bi[tris(hydroxymethyl)- methylamino] propane (pH 7.5). Samples (30 lL) were withdrawn at suitable time points and added to 750 lLof cadmium-ninhydrin stock solution; 70 lL of distilled water was added and incubated at 85 °C for 5 min. The absorb- ance was measured at 505 nm and quantified with free amino acid as standard. For determination of hydrolysis of D -Ala- D -Ser, His 6 –VanXY C was used at a concentration of 26 · 10 )7 M . Rates of hydrolysis of 10, 15, 20, 25, and 30 m MD -Ala- D -Ser were determined using five time points. The high A 505 at time 0 was attributable to D -Ala- D -Ser, Fig. 1. SDS/PAGE analysis. (A) Purification of VanXY C .Lane1, molecular mass standards (in kDa); Lane 2, partially pure VanXY C after gel filtration chromatography. (B) Purification of MBP-VanXY C . Lane 1, molecular mass standards; lane 2, cytoplasm containing MBP-VanXY C ;Lane3,MBP–VanXY C after amylose resin chroma- tography. (C) Purification of His 6 –VanXY C . Lane 1, molecular mass standards; lane 2, cytoplasm containing His 6 –VanXY C ;lane3His 6 – VanXY C after Ni 2+ -nitrilotriacetic acid/agarose chromatography. 2742 A. H. B. Podmore and P. E. Reynolds (Eur. J. Biochem. 269) Ó FEBS 2002 determined using a D -Ala- D -Ser standard curve. For determination of the rate of hydrolysis of UDP-MurNAc- pentapeptide[Ala], His 6 –VanXY C was used at a concentra- tion of 2.25 · 10 )7 M . The rates of hydrolysis of 2, 5, 10, 15, 20 and 30 m M UDP-MurNAc-pentapeptide[Ala] were determined using six time points. For studies of hydrolysis of UDP-MurNAc-pentapeptide[Ser], His 6 –VanXY C was used at a final concentration of 16 · 10 )7 M .Therates of hydrolysis of 2, 5, 10, 15, and 20 m M UDP- MurNAc-pentapeptide[Ser] were determined using five time points. To confirm the degree of hydrolysis of the nucleotide- peptide substances, measurements were also carried out using HPLC by measuring the decrease in the amount of substrate and increase in the amount of product (UDP- MurNAc-tetrapeptide). His 6 –VanXY C (3.2 · 10 )7 M )was incubated with nucleotide-peptide substrates in 100 m M 1,3 bi[tris(hydroxymethyl)-methylamino] propane (pH 7.5). Samples were withdrawn at 0 and 30 min, heated at 90 °C for 5 min, and analysed by HPLC following the method of Reynolds et al. [13]. To determine whether D -Ala was cleaved from UDP-MurNAc-tetrapeptide (2 m M ), the increase of UDP-MurNAc-L-Ala- D -Glu- L -Lys (UDP-Mur- NAc-tripeptide) was also measured. No conversion of UDP-MurNAc-tetrapeptide to UDP-MurNAc-tripeptide was observed. The hydrolysis of 2 m M UDP-MurNAc- pentapeptide[Ala] was also measured in the presence of either UDP-MurNAc-pentapeptide[Ser] or D -Ala- D -Ser at final concentrations of 2 m M . RESULTS Purification of VanXY C The vanXY C gene was expressed in E. coli JM83 (pAP1) and conventional purification of native VanXY C attempted as described in Materials and methods. The purification procedure did not give good separation of VanXY C from other proteins and the activity was spread over many fractions during ion-exchange chromatography. This meth- od yielded 0.2 mg of VanXY C together with contaminating proteins from a culture volume of 1 L. Therefore, the procedure was changed in order to synthesize and purify VanXY C as a maltose-binding protein (MBP) fusion in E. coli. The MBP–VanXY C fusion was purified to homo- geneity using amylose affinity chromatography in yields up to 3 mgÆL )1 . MBP–VanXY C was kinetically characterized, but it was not possible to remove MBP using factor Xa, and compare the activity of the fusion protein with that of VanXY C in the absence of MBP. This problem was caused by the stringent steric requirements for factor Xa cleavage that have been reported previously for this expression system [27–29]. In order to investigate whether MBP might be interfering with activity, an alternative purification strategy was used. VanXY C was expressed with a smaller N-terminal tag of six histidine residues (His 6 –VanXY C )in E. coli M15 (pAP2) and purified to homogeneity using nickel affinity chromatography in yields up to 4 mgÆL )1 . Kinetic analysis of His 6 –VanXY C revealed that its activity was of the same order of magnitude, within the limits of the assay, as that of MBP–VanXY C . Studies with VanX using MBP attached or cut off did not affect the kinetic data [20,21]. We therefore assumed (as the tag could not be removed easily) that the smaller His tag was unlikely to influence activity significantly. The predicted molecular mass of His 6 –VanXY C and VanXY C , 23.6 kDa and 22.3 kDa, respectively, was consistent with the molecular mass estimated by SDS/PAGE analysis. Gel permeation chromatography of the native protein indicated a mobility consistent with a mass of approximately 42–44 kDa. This suggests that VanXY C exists as a dimer in its native form. Effect of divalent cations and EDTA on D , D -dipeptidase and D , D -carboxypeptidase activity VanX copurified with near stoichiometric amounts of Zn 2+ [20]. The Zn 2+ binding residues of VanX were identified using site-directed mutagenesis [20], and later confirmed by analysis of the crystal structure [19]. Comparison of the active site of VanXY C with VanX- and VanY-type enzymes indicated that all of these enzymes contained the same Zn 2+ binding motif. VanX homologues present in the cyanobac- terium Synechocystis strain PCC6803, and the glycopeptide antibiotic producer Streptomyces toyocaensis had also been purified as MBP-fusions and both of these enzymes copurified with near stoichiometric quantities of Zn 2+ [21]. EDTA has been shown to abolish VanY activity [18], but not VanX activity [17]. The addition of a low concentration of Zn 2+ to EDTA-inactivated VanY resulted in the recovery of activity. Replacement of Zn 2+ in VanX by the direct addition of various divalent metal cations to the purified protein affected the D -Ala- D -Ala dipeptidase activity in some instances. When added to VanX at their predetermined optimum stimulatory concentration, Zn 2+ , Fe 2+ ,Co 2+ and Ni 2+ increased the k cat by sixfold to 168- fold [17,19,30]. All the divalent metals tested inhibited VanXY C D , D -dipeptidase activity. Mg 2+ inhibited activity the least, followed by Ni 2+ ,andthenZn 2+ (Table 1). Co 2+ ,Mn 2+ ,Cu 2+ ,andFe 2+ caused total inhibition of D , D -dipeptidase activity at 1.3 and 3.3 m M .Ni 2+ at concentrations of 1 and 5 m M resulted in 10% and 75% inhibition of D , D -carboxypeptidase activity. Zn 2+ at con- centrations of 0.8 and 0.05 m M inhibited D , D -carboxypept- idase activity by 50 and 20%, respectively. Therefore, VanXY C , like VanY [18], was not stimulated by the direct addition of metal ions. EDTA at concentrations between Table 1. Effect of divalent cations on D , D -dipeptidase activity. Cations Concentration of cation (m M ) Activity (nmol D -Ala- D -Ala hydrolysed per min) a – – 0.13 Mg 2+ 1.3 0.11 Mg 2+ 3.3 0.10 Mg 2+ (enzyme omitted) 3.3 0 Ni 2+ 1.3 0.11 Ni 2+ 3.3 0 Ni 2+ (enzyme omitted) 3.3 0 Zn 2+ 1.3 0.06 Zn 2+ 3.3 0 Zn 2+ (enzyme omitted) 3.3 0 a Determined using the D , D -dipeptidase/ D -amino acid oxidase assay with VanXY C . Ó FEBS 2002 Purification and characterization of VanXYC (Eur. J. Biochem. 269) 2743 0.01 and 5.0 m M did not affect the D , D -dipeptidase activity of VanXY C , suggesting that the Zn 2+ molecule, if essential for activity, is tightly bound in the active site. Kinetic analysis of VanXY C The modified cadmium-ninhydrin method [17] alters sample conditions such that ninhydrin preferentially binds to free amino acids in the presence of peptides. The high A 505 values at time zero were attributable to D -Ala- D -Ala, D -Ala- D -Ser or UDP-MurNAc-pentapeptide[Ala] determined from the relevant standard curves. The kinetic parameters for His 6 – VanXY C acting as a D , D -peptidase and D , D -carboxypepti- dase are given in Table 2 and the data from which these were derived are plotted as Fig. 2A–C. K m and k cat were determined by fitting the experimental data obtained to the equation V max ¼ v +[(v/s)ÆK m ], using the direct linear plot [31] and by plotting s/v against s using the equation s/v ¼ K m /V +(1/V)Æs. This assay is crude, being two-part rather than continuous, and repetitive measurement of the K m value with the same enzyme preparation resulted in values that were as much as twofold different. Hydrolysis of terminal D -Ser from UDP-MurNAc- L - Ala- D -Glu- L -Lys- D -Ala- D -Ser (UDP-MurNAc-pentapep- tide[Ser]) was not detected. The K m values for hydrolysis of D -Ala- D -Ser and UDP- MurNAc-pentapeptide[Ala] were similar. Consequently the extent of interference of precursors terminating in D -Ser on the rate of hydrolysis of UDP-MurNAc-pentapeptide[Ala] was measured using HPLC. The rate of removal of the terminal D -Ala of 2 m M UDP-MurNAc-pentapeptide[Ala] was shown by HPLC to be inhibited 30% by the presence of either 2 m MD -Ala- D -Ser or 2 m M UDP-MurNAc-penta- peptide[Ser]. Site-directed mutagenesis of vanXY C Mutated VanXY C fusion proteins containing an N-terminal tag of six histidine residues were purified using the method described for His 6 –VanXY C . The yields for mutants D59S and D59A were similar to that obtained for His 6 –VanXY C . No band corresponding to His 6 –VanXY C was identified in the fractions eluted from the affinity column during purification of His 6 –VanXY C harbouring P154W or Q67V mutations, and no D , D -dipeptidase activity was detected. The kinetic parameters for D59S and D59A resultant proteins acting as a D , D -peptidase and D , D - carboxypeptidase are given in Table 3. It was not possible to determine the kinetic parameters of D59A for hydrolysis of UDP-MurNAc-pentapeptide[Ala]. However, its activity was comparable with that of D59S and His 6 –VanXY C .The rates of hydrolysis of D -Ala- D -Ala and UDP-MurNAc- pentapeptide[Ala] by His 6 –VanXY C are lower than those estimated previously because the enzyme had lost activity during storage probably due to aggregation of the enzyme. DISCUSSION The substrate specificity constant (k cat /K m ) for hydrolysis of D -Ala- D -Ser was 24-fold lower than for D -Ala- D -Ala, the result of a 3.8-fold increase in K m and a 6.3-fold decrease in k cat .Wuet al. [17] determined that the substrate specificity constant of VanX for D -Ala- D -Ser was only sevenfold lower than for D -Ala- D -Ala, the result of a 2.8-fold increase in K m and a 2.6-fold decrease in k cat (the rate of hydrolysis of D -Ala- D -Ser was estimated using DL -Ala- DL -Ser with the assumptions that a quarter of the racemic mixture is D -Ala- D -Ser and the other three isomers have no inhibition effect Table 2. Kinetic parameters for His 6 –VanXY C . ND, hydrolysis not detected. Substrate K m (m M ) k cat (s )1 ) k cat/ K m (m M )1 Æs )1 ) D -Ala- D -Ala 4.0 2.2 0.55 D -Ala- D -Ser 15.5 0.35 0.02 UDP-MurNAc-pentapeptide[Ala] 17.0 5.9 0.35 UDP-MurNAc-pentapeptide[Ser] ND ND ND Fig. 2. Initial velocity/substrate concentration vs. substrate plots of His 6 –VanXY C for (A) D -Ala- D -Ala; (B) D -Ala- D -Ser; (C) UDP-Mur- NAc-pentapeptide[Ala]. 2744 A. H. B. Podmore and P. E. Reynolds (Eur. J. Biochem. 269) Ó FEBS 2002 on VanX) [17]. The likely effect on hydrolysis of terminal D -Ala from UDP-MurNAc-pentapeptide[Ala] in the pres- ence of D -Ala- D -Ser was not clear from an examination of their K m values, being 17 m M and 15 m M , respectively, as determined using the cadmium-ninhydrin method. HPLC analysis of the rate of UDP-MurNAc-pentapeptide[Ala] hydrolysis in the presence of D -Ser-terminating precursors was carried out. The presence of 2 m MD -Ala- D -Ser or 2 m M UDP-MurNAc-pentapeptide[Ser] caused a 30% reduction intherateofhydrolysisof2m M UDP-MurNAc-penta- peptide[Ala]. These data showed that His 6 –VanXY C selec- tively hydrolysed D -Ala-terminating precursors in the presence of D -Ser-terminating precursors. The role of VanX in VanA-type resistance is to hydrolyse preferentially D -Ala- D -Ala but not D -Ala- D -Lactate: consequently it has specificity for dipeptides. As a result, VanX hydrolyses D -Ala- D -Ser relatively rapidly in addition to D -Ala- D -Ala. However, in the VanC phenotype, VanXY C must specific- ally hydrolyse D -Ala- D -Ala with minimal activity against D -Ala- D -Ser, a very different type of specificity. VanXY C , VanX-type and VanY-type enzymes contain most of the active site residues identified in VanX. VanXY C has 39% identity and 74% similarity to VanY in an overlap of 158 amino acids, and low amino-acid identity to VanX, except for a stretch of 22 amino acids that constitute most of theactivesite.However,VanXY C hydrolyses D , D -dipep- tides such as D -Ala- D -Ala, whereas VanY is inactive against this substrate. The small active site cavity of VanX, deduced from crystallographic studies, only allows access of dipep- tides [19]. Therefore, VanXY C and VanY-type enzymes presumably have a less restrictive active site to accommo- date larger substrates such as UDP-MurNAc-pentapep- tide[Ala]. However VanY, unlike VanXY C , will not hydrolyse D -Ala- D -Ala. His 6 –VanXY C has an almost threefold higher k cat for UDP-MurNAc-pentapeptide[Ala] than for D -Ala- D -Ala, which suggests that the active site can accommodate the larger substrate more easily than dipep- tides. Also, VanXY C and VanY showed activity against UDP-MurNAc-pentadepsipeptide, albeit at a much reduced level when compared to UDP-MurNAc-pentapep- tide[Ala] [18]. VanX did not hydrolyse the depsipeptide [6,17]. These data suggest that VanXY C has evolved from an ancestor of the VanY-type enzymes to contain both D , D - dipeptidase and D , D -carboxypeptidase activities, but it remains unclear why VanXY C but not VanY can hydrolyse D -Ala- D -Ala [18]. To investigate the role of specific residues in this selectivity some amino acids presumed to be involved in binding or catalysis were targeted for site-directed mutagenesis. Asp68 is part of the Arg71-Asp68-Tyr35 hydrogen- bonding triad that is believed to orientate Arg71 for transition state stabilization in VanX [19]. This traid is present in all VanX-type enzymes, but Asp68 and Tyr35 equivalents are absent in VanY-type enzymes. VanY-type enzymes contain a conserved serine residue that has a corresponding position to Asp68 in VanX and a glutamine residue (Gln143) that is postulated to function as Asp68 in VanX [32]. However, VanXY C contains both an Asp68 equivalent (Asp59) and a VanY-type Gln143 equivalent (Gln67). Consequently, both of these residues were mutated and the corresponding proteins purified as His-tagged proteins to determine which of these residues may position the arginine (Arg62) for transition-state stabilization in VanXY C . Mutation of Asp59 to Ser or Ala in His 6 – VanXY C resulted in a 50% increase and 73% decrease, respectively, of the substrate specificity constant (k cat /K m ) for D -Ala- D -Ala. The mutation D59S caused a 1.3 fold increase of the substrate specificity constant (k cat /K m )for hydrolysis of UDP-MurNAc-pentapeptide[Ala]. The His- tagged enzyme with the D59A mutation had similar activity to the enzyme with the D59S mutation against UDP- MurNAc-pentapeptide[Ala]. The effect of mutating this aspartate residue is markedly different for VanX, where mutation to Ala caused a 268-fold decrease in k cat and a 750-fold increase in K m . These results suggest that Asp59 in VanXY C is unlikely to be involved in stabilizing Arg62. Mutation of Gln67 to Val resulted in a complete absence of His-tagged mutant protein. The same situation resulted when a conserved Pro (corresponding to Trp182 in VanX) was mutated to Trp. This Pro residue is conserved in all VanY-type enzymes (EPWH motif), except the VanY homologue from Streptococcus mutans, but Trp is present at this position in all VanX-type enzymes (EWWH motif). The reason for the lack of expression of these mutant His-tagged VanXY C enzymes is unknown. ACKNOWLEDGEMENTS This work was carried out under the tenure of a BBSRC studentship to A.H.B.P.WethankC.HillandJ.Lester,CambridgeCentrefor Molecular Recognition for synthesis of oligonucleotides and automated DNA sequencing, respectively. REFERENCES 1. Reynolds, P.E. (1989) Structure, biochemistry and mechanism of action of glycopeptide antibiotics. Eur. J. Clin. Microbiol. Infec. Dis. 8, 943–950. Table 3. Kinetic parameters for His 6 –VanXY C mutant products obtained by site-directed mutagenesis. ND, not determined. Enzyme Substrate K m (m M ) k cat (s )1 ) k cat/ K m (m M )1 Æs )1 ) VanXY C a D -Ala- D -Ala 1.5 0.5 0.33 D59S D -Ala- D -Ala 2.5 1.4 0.56 D59A D -Ala- D -Ala 9.0 0.8 0.09 VanXY C a UDP-MurNAc-pentapeptide[Ala] 9.5 0.8 0.08 D59S UDP-MurNAc-pentapeptide[Ala] 1.5 23 15 D59A UDP-MurNAc-pentapeptide[Ala] ND ND ND a His 6 –VanXY C . Ó FEBS 2002 Purification and characterization of VanXYC (Eur. J. Biochem. 269) 2745 2. Arthur, M., Reynolds, P. & Courvalin, P. (1996) Glycopeptide resistance in enterococci. Trends Microbiol. 4, 401–407. 3. Arthur, M., Molinas, C., Dutka-Malen, S. & Courvalin, P. (1991) Structural relationship between the vancomycin-resistance protein VanH and 2-hydroxycarboxylic acid dehydrogenases. Gene 103, 133–134. 4. Bugg, T.D., Wright, G.D., Dutka-Malen, S., Arthur, M., Courvalin, P. & Walsh, C.T. (1991) Molecular basis for vanco- mycin resistance in Enterococcus faecium BM4147: biosynthesis of a depsipeptide peptidoglycan precursor by vancomycin resistance proteins VanH and VanA. Biochemistry 30, 10408–10415. 5. Bugg, T.D.H., Dutka-Malen, S., Arthur, M., Courvalin, P. & Walsh, C.T. (1991a) Identification of vancomycin resistance pro- tein VanA as a D -alanine: D -alanine ligase of altered substrate specificity. Biochemistry 30, 2017–2021. 6. Reynolds, P.E., Depardieu, F., Dutka-Malen, S. & Courvalin, P. (1994) Glycopeptide resistance mediated by enterococcal trans- poson Tn1546 requires production of VanX for hydrolysis of D -alanyl- D -alanine. Mol. Microbiol. 13, 1065–1070. 7. Arthur, M., Molinas, C. & Courvalin, P. (1992) The VanS-VanR two-component regulatory system controls synthesis of depsi- peptide peptidoglycan precursors in Enterococcus faecalis BM4174. J. Bacteriol. 174, 2582–2591. 8. Evers, S. & Courvalin, P. (1996) Regulation of vanB-type vancomycin resistance gene-expression by the VanS (B)-VanR (B) 2-component regulatory system in Enterococcus faecalis V583. J. Bacteriol. 178, 1302–1309. 9. Quintiliani, R. & Courvalin, P. (1994) Conjugal transfer of the vancomycin resistance determinant VanB between enterococci involves the movement of large genetic elements from chromo- some to chromosome. FEMS Microbiol. Lett. 119, 359–363. 10. Dutka-Malen, S., Molinas, C., Arthur, M. & Courvalin, P. (1992) Sequence of the vanC gene of Enterococcus gallinarum BM4174 encoding a D -alanine: D -alanine ligase-related protein necessary for vancomycin resistance. Gene 112, 53–58. 11. Navarro, F. & Courvalin, P. (1994) Analysis of genes encoding D -alanine: D -alanine ligase-related enzymes in Enterococcus casseliflavus and Enterococcus flavescens. Antimicrob. Ag. Chemother. 38, 1788–1793. 12. Billot-Klein, D., Gutmann, L., Sable, S., Guittet, E. & Van Heijenoort, J. (1994) Modification of peptidoglycan precursors is a common feature of the low-level vancomycin-resistant VanB-type Enterococcus D366 and of the naturally glycopeptide-resistant species Lactobacillus casei, Pediococcus pentosaceus, Leuconostoc mesenteroides,andEnterococcus gallinarum. J. Bacteriol. 176, 2398–2405. 13. Reynolds, P.E., Snaith, H.A., Maguire, A.J., Dutka-Malen, S. & Courvalin, P. (1994) Analysis of peptidoglycan precursors in vancomycin-resistant Enterococcus gallinarum BM4174. Biochem. J. 301, 5–8. 14. Park, I.S., Chung-Hung, L. & Walsh, C.T. (1997) Bacterial re- sistance to vancomycin: overproduction, purification and char- acterization of VanC-2 from Enterococcus casseliflavus as a D-Ala: D-Ser ligase. Proc. Natl Acad. Sci. USA 94, 10040–10044. 15. Reynolds, P.E., Arias, C.A. & Courvalin, C. (1999) Gene vanXY C encodes D, D -dipeptidase (VanX) and D , D -carboxypeptidase (VanY) activities in vancomycin-resistant Enterococcus gallinarum BM4174. Mol. Microbiol. 34, 341–349. 16. Arias, C.A., Martin-Martinez, M., Blundell, T.L., Arthur, M., Courvalin, P. & Reynolds, P.E. (1999) Characterization and modelling of VanT: a novel, membrane-bound, serine racemase from vancomycin-resistant Enterococcus gallinarum BM4174. Mol. Microbiol. 31, 1653–1664. 17. Wu, Z., Wright, G.D. & Walsh, C.T. (1995) Overexpression, purification, and characterization of VanX, a D , D -dipeptidase which is essential for vancomycin resistance in Enterococcus faecium BM4147. Biochemistry 34, 2455–2463. 18. Arthur, M., Depardieu, F., Cabanie, L., Reynolds, P. & Courvalin, P. (1998) Requirement of the VanY and VanX D , D -peptidases for glycopeptide resistance in enterococci. Mol. Microbiol. 30, 819–830. 19. Bussiere, D.E., Pratt, S.D., Katz, L., Severin, J.M., Holzman, T. & Park, C.H. (1998) The structure of VanX reveals a novel amino- dipeptidase involved in mediating transposon-based vancomycin resistance. Mol. Cell 2, 75–84. 20. McCafferty, D.G., Lessard, I.A.D. & Walsh, C.T. (1997) Muta- tional analysis of potential zinc-binding residues in the active site of the enterococcal D -Ala- D -Ala dipeptidase VanX. Biochemistry 36, 10498–10505. 21. Lessard, I.A.D., Pratt, S.D., McCafferty, D.G., Bussiere, D.E., Hutchins,C.,Wanner,B.L.,Katz,L.&Walsh,C.T.(1998) Homologs of the vancomycin resistance D -Ala- D -Ala dipeptidase VanX in Streptomyces toyocaensis, Escherichia coli and Synecho- cystis: attributes of catalytic efficiency, stereoselectivity and regu- lation with implications for function. Chem. Biol. 5, 489–504. 22. Yanisch-Perron, C., Vieira, J. & Messing, J. (1985) Improved M13 phage cloning vectors and host strains: nucleotide sequences of the M13mp18 and pUC18 vectors. Gene 33, 103–119. 23. Sambrook, J., Fritsch, E.F. & Maniatis, T. (1989) Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring. Harbour Laboratory Press, Cold Spring Harbour, NY. 24. Sanger, F., Nicklen, S. & Coulson, A. (1977) DNA sequencing with chain-terminating inhibitors. Proc. Natl Acad. Sci. USA 74, 5463–5467. 25. Bradford, M.M. (1976) A rapid and sensitive method for rapid quantification of microgram quantities of protein utilising a principle of protein-dye binding. Anal. Biochem. 72, 248–254. 26. Messer, J. & Reynolds, P.E. (1992) Modified peptidoglycan pre- cursors produced by glycopeptide-resistant enterococci. FEMS Microbiol. Lett. 94, 195–200. 27. Riggs, P.D. (1994) Expression and purification of maltose-binding protein fusions. In Current protocols in Molecular Biology, John Wiley & Sons, NY. 28. Liger, D., Masson, A., Blanot, D., Van Heijenoort, J. & Parquet, C. (1995) Over-production, purification, and properties of the uridine-diphosphate-N-acetylmuramyl: L -alanine ligase from Escherichia coli. Eur. J. Biochem. 230, 80–87. 29. Pryor, K.D. & Leiting, B. (1997) High-level expression of soluble protein in Escherichia coli using a His 6 -tag and maltose-binding protein double-affinity fusion system. Prot. Expr. Purif. 10, 309–319. 30. Brandt, J.J., Chatwood, L.L. & Crowder, M.W. (1998) VanX, a metalloenzyme conferring high-level vancomycin resistance. J. Inorg. Biochem. 74,80. 31. Eisenthal, R. & Cornish-Bowden, A. (1974) A new graphical procedure for estimating enzyme kinetic parameters. Biochem. J. 139, 715–720. 32. Lessard, I.A.D. & Walsh, C.T. (1999) Mutational analysis of active-site residues of the enterococcal D -Ala- D -Ala dipeptidase VanX and comparison with Escherichia coli D -Ala- D -Ala ligase and D -Ala- D -Ala carboxypeptidase VanY. Chem. Biol. 6,177– 187. 2746 A. H. B. Podmore and P. E. Reynolds (Eur. J. Biochem. 269) Ó FEBS 2002 . amplify vanXY C usingpAT704as template with primers A (5¢-GCTA GGTCTCAATGAAC ACATTACAATT-3¢)andB(5¢-TATG GAATTCTCATG CGAACTGCCTCA-3¢) that included BsaIandEcoRI restriction. Purification and characterization of VanXY C , a D , D -dipeptidase/ D , D -carboxypeptidase in vancomycin-resistant Enterococcus gallinarum BM4174 Adrian