SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

vanGCd, a cryptic gene cluster highly homologous to the vanG gene cluster of Enterococcus faecalis is largely spread in Clostridium difficile. Since emergence of vancomycin resistance would have dramatic clinical consequences, we have evaluated the capacity of the vanGCd cluster to confer resistance. We showed that expression of vanGCd is inducible by vancomycin and that VanGCd, VanXYCd and VanTCd are functional, exhibiting D-Ala : D-Ser ligase, D,D-dipeptidase and D-Ser racemase activities respectively. In other bacteria, these enzymes are sufficient to promote vancomycin resistance. Trans-complementation of C. difficile with the vanC resistance operon of Enterococcus gallinarum faintly impacted the MIC of vancomycin, but did not promote vancomycin resistance in C. difficile. Sublethal concentration of vancomycin led to production of UDP-MurNAc-pentapeptide[D-Ser], suggesting that the vanGCd gene cluster is able to modify the peptidoglycan precursors. Our results indicated amidation of UDP-MurNAc-tetrapeptide, UDP-MurNAc-pentapeptide[D-Ala] and UDP-MurNAc-pentapeptide[D-Ser]. This modification is passed on the mature peptidoglycan where a muropeptide Tetra-Tetra is amidated on the meso-diaminopimelic acid. Taken together, our results suggest that the vanGCd gene cluster is functional and is prevented from promoting vancomycin resistance in C. difficile.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Glycopeptide antibiotics, vancomycin and teicoplanin, are effective drugs against Gram-positive bacterial pathogens. Resistance to vancomycin was reported 30 years ago and has become a major public health problem (Arias and Murray, 2012). Vancomycin inhibits bacterial cell wall synthesis by binding the C-terminal dipeptide D-alanyl-D-alanine (D-Ala-D-Ala) of the muramyl-pentapeptide moiety of peptidoglycan precursors (Reynolds, 1989). Nine van gene clusters have been described that confer vancomycin resistance in enterococci by modifying the terminal D-Ala-D-Ala of peptidoglycan precursors. They are classified according to their ligases which synthesize D-Ala-D-Lac (VanA, VanB, VanD and VanM) or D-Ala-D-Ser (VanC, VanG, VanE, VanL and VanN) (Abadia Patino et al., 2002; Arias et al., 2003; Courvalin, 2006; Boyd et al., 2008; Xu et al., 2010; Lebreton et al., 2011).

All van gene clusters are organized in two modules. One includes resistance genes encoding enzymes responsible for modification of peptidoglycan precursors and for elimination of the natural D-Ala-D-Ala-ending cytoplasmic precursors and the dipeptide D-Ala-D-Ala. The other module regulates expression of vancomycin resistance. Regulatory and resistance modules may be organized in two (VanA, VanB, VanD, VanG, VanL, VanM) or a single operon (VanC, VanE, VanN) (Arthur et al., 1992; Casadewall and Courvalin, 1999; Arias et al., 2000a; Abadia Patino et al., 2002; Depardieu et al., 2003; Boyd et al., 2008; Xu et al., 2010; Lebreton et al., 2011). Resistance expression is controlled by a two-component regulatory system, composed of VanS, a membrane sensor kinase, and VanR, a cytoplasmic response regulator (for a review see Courvalin, 2006). The VanS-type sensor detects the presence of vancomycin which results in an ATP-dependent autophosphorylation of a histidine residue (Koteva et al., 2010). The phosphoryl group is then transferred to the VanR-type transcriptional regulator which induces expression of vancomycin resistance for the VanA-, VanB-, VanG-, VanE-, VanL- and VanM-type resistance or is responsible for constitutive expression for VanC-, VanD- and VanN-type (Arthur et al., 1992; Casadewall and Courvalin, 1999; Arias et al., 2000a; Abadia Patino et al., 2002; Depardieu et al., 2003; Boyd et al., 2008; Xu et al., 2010; Lebreton et al., 2011). In cases of two operons, VanR induces transcription of both the resistance and regulatory operons, except for VanG-type where transcription of the regulatory operon is constitutive (Depardieu et al., 2003).

Three enzymes are required for the synthesis of D-Ala-D-Ser ending precursors: (i) a racemase that converts L-Ser to D-Ser, (ii) a ligase that synthesizes D-Ala-D-Ser, and (iii) a bifunctional dipeptidase/carboxypeptidase that hydrolyses the dipeptide D-Ala-D-Ala synthesized by the host Ddl ligase and removes the C-terminal D-Ala residue of the natural peptidoglycan precursors (Reynolds and Courvalin, 2005). Such activities are carried on the vanC resistance operon in Enterococcus gallinarum and are encoded by (i) vanTc, (ii) vanC and (iii) vanXY respectively. However, the VanXY encoded by the vanG operon from Enterococcus faecalis lacks a D,D-carboxypeptidase activity (Depardieu et al., 2003). The D-Ala-D-Ser dipeptide is then incorporated into the peptidoglycan precursors via MurF which catalyses its ligation to UDP-MurNAc-tripeptide (Barreteau et al., 2008).

In 2006, the complete genome sequence of Clostridium difficile 630 revealed the presence of a van gene cluster designated ‘vanG-like’ (Sebaihia et al., 2006). This cluster displays the higher degree of identity with vanG of E. faecalis BM4518 and includes five open reading frames encoding putative proteins similar to VanR, VanS, VanG, VanXY and VanT (Sebaihia et al., 2006). C. difficile is responsible for 15 to 20% of antibiotic-related cases of diarrhoea and nearly all cases of pseudomembranous colitis (for a review see Blanckaert et al., 2008). The increasing incidence of C. difficile infections and the high rate (20%) of relapsing cases have a major impact on health care costs (Barbut and Petit, 2000). Metronidazole and vancomycin are the reference treatments of C. difficile infections (Surawicz and Alexander, 2011). In Europe, vancomycin is administrated in the relapsing cases or when C. difficile has a reduced susceptibility or is resistant to metronidazole (Tillotson and Tillotson, 2011). A distribution study showed that the vanG-like gene clusters are present in 85% of C. difficile clinical isolates (Ammam et al., 2012). However, to date, vancomycin resistance has not been reported in C. difficile. Moreover, no change in the global composition of peptidoglycan was observed in the presence of a sublethal concentration of vancomycin, suggesting that the vanG-like gene cluster is not functional (Peltier et al., 2011). Nevertheless, emergence of vancomycin resistance would have dramatic clinical consequences.

To distinguish the vanG-like gene cluster in C. difficile from the vanG-like gene cluster in other bacteria we renamed it vanGCd. The aim of this work was to evaluate the ability of vanGCd to confer vancomycin resistance.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

The vanGCd cluster is transcribed and its expression is inducible by vancomycin

Since C. difficile 630 is not resistant to vancomycin, transcription of the vanGCd gene cluster was first studied. This was performed by RT-PCR with the primers shown in Fig. 1A and Table S2. In the absence of vancomycin, only vanRCd and vanSCd were found to be transcribed (Fig. 1B). In the presence of vancomycin, the regulation and resistance operons were both and independently transcribed as shown by disruption of mRNA at the vanS/G intergenic region. A slight difference was observed in expression of vanRCd. Transcription of vanRCd and vanGCd was therefore quantified in the presence and in the absence of vancomycin by qRT-PCR (Fig. 1C and D). No significant differences were observed for vanRCd transcripts suggesting that transcription of the regulatory operon is constitutive (Fig. 1C). In contrast, transcription of vanGCd was only detected in the presence of vancomycin (0.5 or 1.5 mg l–1) (Fig. 1B and D). These results indicate that expression of the vanGCd resistance operon was inducible by vancomycin and may be due to a promoter located in the intergenic vanS/G region. For sake of simplicity, the entire intergenic vanS/G DNA was designated Pvan. We also examined vanGCd transcription when C. difficile 630 was exposed to 0.075 mg l–1 of teicoplanin (0.15 mg l–1 minimum inhibitory concentration, MIC), but no differences in expression were observed, suggesting that vanGCd is not induced by teicoplanin (data not shown). Since VanR is likely to be involved in transcription activation, transcription of vanGCd was analysed in a vanR mutant (using the 630Δerm vanR::erm strain). Since growth of the vanR mutant is affected by the presence of 1.5 mg l–1 of vancomycin, a concentration of 0.5 mg l–1 was used. As a control, vanGCd induction in the presence of 0.5 mg l–1 of vancomycin (Fig. 1E) was found to be similar to that of the parental strain in the presence of 1.5 mg l–1 of vancomycin. In contrast, similar amounts of vanGCd transcripts were observed in the vanR mutant in the absence or in the presence of 0.5 mg l–1 of vancomycin (Fig. 1F), suggesting that VanR is directly involved in the induction of vanGCd expression in the presence of vancomycin.

figure

Figure 1. The vanGCd is transcribed in the presence of vancomycin.

A. Schematic representation of the vanGCd gene cluster. Arrows indicate ORF and sense of transcription. The primers used in RT-PCR are indicated by arrowheads; the size of the amplified regions is indicated in base pair.

B. Transcription analysis of the vanGCd gene cluster by RT-PCR. Co-transcription of vanRCd and vanSCd (vanR/S), vanSCd and vanGCd (vanS/G), vanGCd and vanXYCd (vanG/XY) and vanXYCd and vanTCd (vanXY/T) was tested in the absence (–) or in the presence (+) of vancomycin.

C and D. vanRCd is constitutively expressed and vanGCd expression is induced by vancomycin. vanRCd (C),vanGCd (D and E) and vanGCd in vanRCd mutant strain (F) transcripts were quantified by relative qRT-PCR. C. difficile was grown in TY or in TY containing 1.5 mg l–1 (D) or 0.5 mg l–1 (E and F) of vancomycin (TY + Vm). Values of vanR and vanG expression in the absence of vancomycin were arbitrarily fixed to 1. The mean values from 9 (C and D) or 3 (E and F) independent experiments are shown. Error bars represent standard deviations. vanGCd expression in the absence or in the presence of vancomycin was significantly different (P ≤ 0.0005). Of note, vanGCd expression was also induced in the presence of 0.5 mg l–1 of vancomycin (not shown). Induction in the presence of 0.5 mg l–1 of vancomycin was found to be similar to that of the parental strain in the presence of 1.5 mg l–1 of vancomycin using Wilcoxon rank sum test, P > 0.05; http://www.r-project.org (Bauer, 1972).

Download figure to PowerPoint

VanGcd, VanXYCd and VanTCd are functional in vitro

Functional analysis of VanGCd, VanXYCd and the C-terminal catalytic domain of VanTCd was performed on the purified proteins. Kinetic parameters determined for their respective substrates are shown in Table 1. VanGCd displayed high catalytic efficiency for D-Ala-D-Ser formation which was > 270-fold higher than that for D-Ala-D-Ala. This was due to a 200-fold higher affinity for D-Ser. VanXYCd was tested for its putative D,D-dipeptidase activity on D-Ala-D-Ala, and for its D,D-carboxypeptidase activity using UDP-MurNAc-pentapeptide[D-Ala] and UDP-MurNAc-pentapeptide[D-Ser] as substrates. The D,D-dipeptidase activity was of the same magnitude as that reported for VanXYC of E. gallinarum (Podmore and Reynolds, 2002). However, no D,D-carboxypeptidase activity was detected with either pentapeptide substrate. The VanTCd converted both L-Ser and L-Ala to D-Ser and D-Ala, respectively, but the alanine racemase activity appeared to be slightly higher than the serine racemase activity. Altogether, these results indicated that the resistance enzymes were functional. Since MurF is involved in the linkage of UDP-MurNAc-tripeptide to the natural dipeptide D-Ala-D-Ala or to the modified dipeptide D-Ala-D-Ser, we tested C. difficile MurF activity after purification. Optimal pH and Mg2+ concentration were determined using D-Ala-D-Ala as a substrate and found to be 7.8 and 40 mM respectively. The kinetic parameters for D-Ala-D-Ala and D-Ala-D-Ser are shown in Table 2. The Vmax values for both dipeptides were similar. In contrast, the Km value for D-Ala-D-Ser was 11-fold higher than that for D-Ala-D-Ala, suggesting that MurF binds preferentially to the D-Ala-D-Ala dipeptide.

Table 1. Kinetic parameters of resistance enzymes
EnzymeSubstrate/productKm1 (mM)Km2 (mM)kcat (min−1)kcat/Km (min−1 mM−1)
  1. a

    Not detected.

VanGCdD-Ala-D-Ala1.51402081.5
D-Ala-D-Ser1.50.7286408
VanXYCdD-Ala-D-Ala0.75 0.50.66
UDP-MurNAc-pentapeptide[D-Ala]  NDa 
UDP-MurNAc-pentapeptide[D-Ser]  NDa 
VanTCdL-Ser4.2 29871
L-Ala5.7 676119
L-Ser/L-Ala0.73 0.40.6
Table 2. Kinetic parameters of MurF
SubstrateKmVmaxVmax/Km
(μM)(μmol min−1 mg−1)(μmol min−1 mg−1 μM−1)
D-Ala-D-Ala43 ± 1124 ± 20.56 ± 0.15
D-Ala-D-Ser510 ± 19019 ± 20.037 ± 0.014

The vanGCd resistance operon allows Escherichia coli NR698 survival in the presence of vancomycin

Although vancomycin resistance in C. difficile was not observed, transcriptional and biochemical analyses indicated that the vanGCd gene cluster was expressed and functional. To test whether vanGCd could confer resistance in another host, it was introduced into the vancomycin-susceptible Escherichia coli NR698 (Ruiz et al., 2005). The vanC operon was used as a positive control. In the presence of 0.5 mg l–1 of vancomycin, E. coli NR698 did not grow (MIC = 0.25 mg l–1) whereas E. coli NR698 harbouring pTC27 (carrying the vanGCd operon) and pFA21 (carrying the vanC operon) (MICs = 0.5 mg l–1) exhibited a survival rate of 22.43 ± 5.79 and 45.6 ± 7.3 (%) respectively (Fig. 2). In the presence of a vancomycin concentration higher than 1 mg l–1, no growth was observed. This result suggests that the vanGCd operon affects susceptibility to vancomycin and is functional in vivo.

figure

Figure 2. The vanGCd operon allows survival in the presence of vancomycin.

A. The survival rate of E. coli NR698 harbouring control plasmid (pBlunt), vanGCd (pFA21) or vanC (pTC27) operons was determined as follows: the percentage of growth in the absence of vancomycin was fixed at 100%, and the relative survival rate in the presence of vancomycin (0.5 mg l–1) was determined. Means of three independent experiments are presented and standard derivations are indicated by error bars.

B. Similarly, the survival rate of the C. difficile 630Δerm vanG::erm strain and of the parental strain (control) were determined using 0, 0.5 or 1 mg l–1 of vancomycin.

C. Growth curves of C. difficile 630Δerm vanG::erm strain (squares) and of the parental strain (630E) (circles) are presented in TY medium in the absence of vancomycin (empty squares and empty circles respectively), 0.5 mg l–1 (crossed white squares and crossed white circles respectively), 1 mg l–1 (semi-filled white squares and semi-filled white circles respectively), 1.5 mg l–1 (black squares and black circles respectively) of vancomycin.

Download figure to PowerPoint

The vanC operon from E. gallinarum does not confer vancomycin resistance in C. difficile

To test whether lack of resistance to vancomycin in C. difficile was not due to a defective vanGCd resistance operon, the vanC operon from E. gallinarum, which was cloned downstream from the intergenic vanS/G region corresponding to the Pvan promoter, was introduced into C. difficile 630. The MIC of vancomycin against C. difficile 630 harbouring pFA27 (carrying the vanC resistance operon) or pTC49 (carrying Pvan) were determined (Table 3). The MIC of vancomycin for C. difficile 630 harbouring pTC49 (carrying Pvan) was ≤ 2 mg l–1 whereas that for the strain harbouring pFA27 (carrying the vanC resistance operon) ranged from 3 to 4 mg l–1 (Table 3). To further identify the gene responsible for this slight increase in MIC, vanC, vanXYC and vanTC were cloned separately under the control of Pvan. Only C. difficile 630 harbouring pTC57 (carrying vanXYC) had a vancomycin MIC of 3–4 mg l–1. The control experiment with pFA24 (carrying vanXYCd) showed that the MIC remained at 2 mg l–1. Because the main difference consists in the lack of D,D carboxypeptidase activity of VanXYCd compared with VanXYC, this result suggests that the D,D-carboxypeptidase activity of VanXYC, but not of VanXYCd, may contribute to vancomycin resistance.

Table 3. MICs of vancomycin against complemented C. difficile 630 strain
C. difficileVancomycin MIC (mg l−1)a; BHI
  1. a

    MICs average of 3 independent experiments determined by E-test.

630 + pTC49 (pMTL84151ΩPvan)1.5–2
630 + pFA27 (pTC49ΩvanCXYCTC)3–4
630 + pTC56 (pTC49ΩvanC)2
630 + pTC57 (pTC49ΩvanXYC)3–4
630 + pTC58 (pTC49ΩvanTC)2
630 + pFA24 (pTC49ΩvanXYCd)2
630Δerm1.5
630Δerm vanG::erm0.75
630Δerm vanR::erm0.75
630Δerm vanG::erm+pTC490.75
630Δerm vanG::erm+pFA271
FM130.38–0.5
TL11430.5–0.75
30480.5–0.75
46020.25–0.38
CD30.75–1
ATCC 435980.38–0.5

The vanGCd operon slightly affected C. difficile vancomycin susceptibility

In order to test whether the vanGCd resistance operon confers a growth advantage in the presence of vancomycin, we constructed a inactivated vanGCd mutant using the clostron method (Heap et al., 2010) (see Experimental procedures). This mutant had a vancomycin MIC of 0.75 mg l–1. The survival rates of the parental strain (630Δerm) and of the vanGCd mutant (630Δerm vanG::erm) were compared in the presence of 0, 0.5 and 1 mg l–1 of vancomycin (Fig. 2B). The parental strain was able to survive up to 1 mg l–1 whereas 630Δerm vanG::erm only survived up to 0.5 mg l–1of vancomycin. Growth in liquid medium was also performed in the presence of various concentrations of vancomycin (Fig. 2C). Growth of the parental strain was not or poorly affected by the presence of vancomycin (up to 1.5 mg l–1) whereas 630Δerm vanG::erm was affected in the presence of 1 or 1.5 mg l–1 of vancomycin. The vanGCd resistance operon conferred therefore a growth advantage in the presence of low concentrations of vancomycin (up to 1.5 mg l–1). When the vanC resistance operon was introduced into strain 630Δerm vanG::erm the MIC of vancomycin raised to 1 mg l–1 (Table 3). Five strains lacking the vanGCd cluster (Ammam et al., 2012) have a MIC < 1 mg l–1 (Table 3). These results suggest that C. difficile strains devoid of the vanGCd cluster are slightly more susceptible to vancomycin.

C. difficile peptidoglycan precursors contain D-Ser

To detect if D-Ser was incorporated in peptidoglycan nucleotide precursors, these compounds were extracted and purified from the different strains [630 and 630 (vanC)] in the absence or in the presence of vancomycin (see Fig. 3A for HPLC profiles of extracts). The peptidoglycan precursors in the different peaks were identified by complementary analyses (amino acids and hexosamines content, mass spectrometry, and comparison with authentic standards, see Table S3). Only the UDP-MurNAc-tetrapeptides and UDP-MurNAc-pentapeptides, which are expected to be modified following vancomycin treatment, were analysed in details and their pool levels determined (Fig. 3B). In the absence of vancomycin, more than 80% were composed of UDP-MurNAc-pentapeptide[D-Ala] (peak 5 in Fig. 3A), the remaining consisting in UDP-MurNAc-tetrapeptide (peak 2). In the presence of a sublethal concentration of vancomycin, UDP-MurNAc-pentapeptide[D-Ala] relative quantity (40%) was significantly reduced and accumulation of a new precursor, further identified as UDP-MurNAc-pentapeptide[D-Ser] (peak 1), was observed. Interestingly, three other peptidoglycan precursors were detected in peaks 3, 4 and 6 (Fig. 3). Mass spectrometry showed that these compounds had molecular masses lower by one unit compared with UDP-MurNAc-pentapeptide[D-Ser], UDP-MurNAc-tetrapeptide and UDP-MurNAc-pentapeptide[D-Ala] respectively (Table S3). This suggests that these precursors were amidated either on the 2,6-diaminopimelic acid (A2pm) or on the glutamate (Glu) residue. Therefore, in the presence of vancomycin, the peptidoglycan precursor pool contained 49% of total (amidated + non-amidated) UDP-MurNAc-pentapeptides[D-Ala], 12% of UDP-MurNAc-tetrapeptides and 39% of UDP-MurNAc-pentapeptides[D-Ser]. In contrast, no UDP-MurNAc-pentapeptides[D-Ser] were detected neither in 630Δerm vanG::erm nor in 630Δerm vanR::erm in the presence of 0.5 mg l–1 of vancomycin whereas it represented 6% of the peptidoglycan precursors in the parental strain under the same conditions (data not shown). The ratio between D-Ala- and D-Ser-containing UDP-MurNAc-pentapeptides may account for the absence of vancomycin resistance in C. difficile 630. Since the vancomycin MIC of C. difficile 630 harbouring pFA27 (vanC) was increased to 3–4 mg l–1, the peptidoglycan precursors were analysed and found to be similar in composition to that of C. difficile 630 grown in the presence of vancomycin (Fig. 3). The ratio between the precursors was slightly different, and consistent with a higher incorporation of D-Ser in the peptidoglycan synthesis pathway: the pool of precursors was now composed of 17% of UDP-MurNAc-pentapeptides[D-Ala], 23% UDP-MurNAc-tetrapeptides and 60% of UDP-MurNAc-pentapeptides[D-Ser].

figure

Figure 3. vanGCd expression induces modifications of peptidoglycan precursors.

A. HPLC profiles of cytoplasmic peptidoglycan precursors from C. difficile 630 grown in the absence (a) or in the presence (b) of vancomycin and of C. difficile 630 harbouring pFA27 (vanC operon) in the presence of vancomycin (c). The positions of UDP-MurNAc-peptides, subsequently identified by mass spectrometry, are indicated by arrows: 1, UDP-MurNAc-pentapeptide[D-Ser]; 2, UDP-MurNAc-tetrapeptide; 3, UDP-MurNac-pentapeptide[D-Ser] (-1); 4, UDP-MurNAc-tetrapeptide (-1); 5, UDP-MurNAc-pentapeptide[D-Ala]; 6, UDP-MurNAc-pentapeptide[D-Ala] (-1).

B. Amounts (in nmol) of peptidoglycan precursors contained in 209 mg of cell dry weight. Percentage of each peptidoglycan precursor is presented between parentheses.

Download figure to PowerPoint

Peptidoglycan precursors amidation is passed on mature peptidoglycan

Mature peptidoglycan was described by Peltier et al. and none of the muropeptides were described to be amidated. Since 16% of the peptidoglycan precursors of C. difficile 630 appeared to be amidated after growth in the presence of vancomycin, such a modification was searched for in the mature peptidoglycan polymer (Fig. 4A). Peptidoglycan was purified from C. difficile 630 grown in the absence or in the presence of vancomycin and analysed by digestion by a muramidase and HPLC of the muropeptides. Three new peaks appeared in the presence of vancomycin. Compounds in peaks 1 and 3 were identified by mass spectrometry, with molecular masses by one dalton lower than those of the classical deacetylated Tetra (deAc) (observed mass [M + Na]+ 921.387; calculated [M + Na]+ 922.387) and deacetylated-Tetra-Tetra (deAc×2) (observed mass [M + Na]+ 1802.772; calculated [M + Na]+ 1803.774) fragments. This mass difference and structure modification most likely corresponds to that previously observed in the peptidoglycan precursors. To characterize this modification, a MS-MS fragmentation of the deacetylated Tetra-Tetra (deAc×2) fragment exhibiting a deficit of 1 dalton was performed (Fig. 4B). Amidation occurs most frequently either on the glutamate or the meso-diaminopimelic acid (A2pm) residues (Vollmer et al., 2008). The a1 and b1 ions clearly indicated that the mass deficit was not associated with the glutamate residues but rather with the A2pm residues. Fragments y1′ (m/z 921) and y2′ (m/z 992) indicated that the mass deficit was associated with the A2pm residue involved in the cross-link. If the A2pm had instead a free ε-amino group, y1′ and y2′ would have given the masses of m/z 922 and 993, respectively, masses that were never observed. The structure of the muropeptide is shown in Fig. 4B, being amidated on the A2pm residue. This result suggests that the reduction of mass observed in the peptidoglycan precursors also resulted from amidation of the same residue.

figure

Figure 4. Detection of amidated muropeptides in C. difficile growm in the presence of vancomycin.

A. HPLC analysis of muropeptides extracted from C. difficile 630 grown in the absence (a) or in the presence (b) of vancomycin. The numbers indicate the peaks difference under both conditions.

B. MS-MS identification and chemical structure assigned to peak 3 ([M + Na]+, m/z 1802), showing an amidated Tetra-Tetra (deAc × 2).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

We report that the vanGCd gene cluster in C. difficile genome is expressed and functional although the organism remains susceptible to vancomycin. Moreover, in the presence of sublethal concentrations of vancomycin, significant amounts of D-Ser-containing peptidoglycan precursors were produced. This modification resulted from the VanTCd and VanGCd activities. The UDP-MurNAc-pentapeptides[D-Ser] in E. coli (detected after peptidoglycan precursor purification; data not shown) together with VanXYCd dipeptidase activity allowed more than 20% survival of E. coli NR698 in the presence of 0.5 mg ml–1 of vancomycin as opposed to the control (Fig. 2A). The activities of VanGCd, VanTCd and VanXYCd have been reported to be sufficient to promote vancomycin resistance in other bacterial genera. The D-Ala-D-Ser specificity of VanGCd is in agreement with the identity level (67%) with the E. faecalis VanG ligase (Depardieu et al., 2003). The catalytic efficiency of VanTCd for L-Ala was c. twofold higher than that for L-Ser, in contrast to VanTC that has an alanine racemase activity 14% of the serine racemase (Arias et al., 2000b). Although the difference in substrate specificity of VanTCd was not significant, we hypothesize that it may reduce the synthesis of D-Ser and consequently that of peptidoglycan precursors ending in D-Ser. However, addition of D-Ser in the culture medium did not increase the vancomycin MIC (data not shown). Similarly to VanXYG (Depardieu et al., 2003), VanXYCd acted only as a dipeptidase. This result was confirmed by precursor analysis which showed that the basal D,D-carboxypeptidase activity (reflected by the percentage of UDP-MurNAc-tetrapeptide) did not increase in the presence of vancomycin. In contrast, when the vanC operon was expressed in C. difficile 630, the concentration of UDP-MurNAc-tetrapeptide increased. However, the vanC resistance operon did not confer vancomycin resistance either in the 630 C. difficile strain or in the 630Δerm vanG::erm mutant strain. A main difference between the vanGCd and vanC resistance operons resulted from the D,D-carboxypeptidase activity exhibited by VanXYC, but not by VanXYCd. This suggests that increased concentration of UDP-MurNAc-tetrapeptide may be attributed to VanXYC. We also tested whether VanXYCd could act against precursors ending in D-Ser, hence reducing their incorporation into the peptidoglycan biosynthesis pathway. However, such an activity was not detected using D-Ser-containing UDP-MurNAc-pentapeptide as a substrate (Table 1). Peltier et al. observed in the C. difficile mature peptidoglycan composition a high proportion of muropeptides Tetra due to activity of D,D-carboxypeptidases and hypothesized that it may be due to VanXYCd (Peltier et al., 2011). However, according to our results, D,D-carboxypeptidase activity could result from one or the two other chromosomal putative β-lactam-insensitive D,D-carboxypeptidase, but not by VanXYCd (Peltier et al., 2011). All proteins involved in vancomycin resistance are functional and putatively able to promote vancomycin resistance. Nevertheless, in the presence of low concentration of vancomycin (0.5 mg l–1), the concentration of UDP-MurNAc-pentapeptide[D-Ser] was only of 6%. Since transcription of vanGCd is similar in the presence of 0.5 mg l–1 of vancomycin as in the presence of a higher concentration (1.5 mg l–1), this suggests that the efficiency of VanGCd, VanXYCd and VanTCd to produce UDP-MurNAc-pentapeptide[D-Ser] is weak. This result is consistent with another experiment where the vanGCd or the vanC resistance operons cloned under the control of the aphA3 promoter and transferred into E. faecalis led to MICs of vancomycin of 2 (as the parental strain JH2.2) or 4 respectively (data not shown). This slight difference may be due to VanXYCd and VanXYC activities, but suggests that the vanGCd resistance operon encodes less efficient resistance proteins than the vanC cluster. We also investigated if vancomycin susceptibility could be due to a deficiency in the regulation of the vanGCd gene cluster. qRT-PCR demonstrated that the regulatory operon was constitutively expressed whereas the resistance operon was inducible by vancomycin. This result suggests that VanSCd senses properly vancomycin and that VanRCd is able to act as a transcriptional activator. Transcription of the vanGCd resistance operon was not inducible in a 630Δerm vanR::erm strain in the presence of 0.5 mg l–1 of vancomycin (Fig. 1F), whereas it was still induced in the same conditions in the parental strain (Fig. 1E). Our results show that VanRCd was involved in transcription of the vanGCd resistance operon, accordingly to the role of VanRCd in other Van-type resistances. Moreover, the whole regulation of the vanGCd gene cluster was similar to that of vanG from E. faecalis (Depardieu et al., 2003). Furthermore, the Pvan promoter was strong enough to drive the expression of an erythromycin resistance cassette and to confer resistance to erythromycin in C. difficile, even in the absence of vancomycin induction (data not shown). This indicates that transcription of the vanGCd resistance operon is effective. This result is consistent with another experiment where all attempts to select C. difficile 630 exhibiting increased MIC to vancomycin either by culturing in the presence of subconcentration of vancomycin (2–4 mg l–1) or by using Szybalski gradient (Szybalski and Bryson, 1952) were unsuccessful (data not shown). Noteworthy, the vanGCd resistance operon was not inducible by teicoplanin.

Vancomycin resistance is the result of a double mechanism, synthesis of peptidoglycan precursors ending in D-Ala-D-Ser or D-Ala-D-Lac and elimination of the natural precursors. D-Ala-D-Ser precursors lead to low level resistance compared with D-Ala-D-Lac. It is not clear if all the natural precursors should be eliminated or if only a certain amount of modified precursors is sufficient to promote vancomycin resistance. In VanB-type vancomycin resistance, a clear correlation between the ratio of UDP-MurNAc-pentapeptide[D-Ala] and UDP-MurNAc-pentadepsipeptide[D-Lac] pools and the vancomycin MIC was observed (Billot-Klein et al., 1994; Arthur et al., 1996); up to 46% of UDP-MurNAc-pentapeptide[D-Ala] was found in E. faecium displaying vancomycin MIC of 32 mg l–1. For VanG-type, 66% or 53% of UDP-MurNAc-pentapeptide[D-Ala] have been found in two strains which both had a vancomycin MIC of 16 μg ml–1 (Depardieu et al., 2003). Altogether, the results of these studies suggest that the decrease in concentration of the natural peptidoglycan precursor, concomitant with an increased concentration of the modified one, may be sufficient to promote vancomycin resistance. The minimal percentage of modified peptidoglycan precursor required for resistance may vary depending on the bacterial strain. In our study, 49% and 16% of UDP-MurNAc-pentapeptide[D-Ala] and amidated UDP-MurNAc-pentapeptide[D-Ala] were observed (Fig. 3), which correlated with MICs of vancomycin of 2 and 3–4 μg ml–1 respectively. The high concentration of UDP-MurNAc-pentapeptide[D-Ala] may explain why C. difficile is not resistant to vancomycin and may be due to a high activity of the host Ddl (Bisicchia et al., 2011). For this reason many vancomycin-resistant bacteria possess a mutated ddl gene (Depardieu et al., 2009; Perichon and Courvalin, 2012). Moreover, Bacillus subtilis harbouring VanB-type resistance genes is resistant to vancomycin only when ddl is mutated (Bisicchia et al., 2011). C. difficile possesses a single annotated Ddl with 69 % similarity to B. subtilis 168 Ddl (Bisicchia et al., 2011; Monot et al., 2011). Analysis of its primary structure showed that residues involved in the catalytic reaction are conserved (Tytgat et al., 2009). In addition, none of the amino acid changes found in impaired Ddls (Prevost et al., 2000; Gholizadeh et al., 2001) were detected. These features argue in favour of the fact that the chromosomal C. difficile Ddl is active. In order to test whether C. difficile Ddl influences the expression of vancomycin resistance, this enzyme was inhibited by D-cycloserine addition (up to 200 mg l–1; MIC = 240 mg l–1) (data not show). It is of interest to highlight that D-cycloserine also inhibits alanine racemase depriving the host of an essential residue within the resistance pathway (Lambert and Neuhaus, 1972). However, no change in the MIC for vancomycin was observed, suggesting that C. difficile Ddl was not responsible for blocking vancomycin resistance. The high concentration of UDP-MurNAc-pentapeptide[D-Ala] may also be the result of a weak incorporation of D-Ala-D-Ser in peptidoglycan precursors. The C. difficile MurF had a better affinity for D-Ala-D-Ala than for D-Ala-D-Ser (Table 2) which could limit to some extent the synthesis of UDP-MurNAc-pentapeptide[D-Ser]. Moreover, UDP-MurNAc-pentapeptide[D-Ser] may be poorly or not used by the penicillin-binding proteins (PBPs) and become toxic for C. difficile. However, C. difficile PBPs have not been studied and an alternative transpeptidase, which could catalyse the transpeptidation of the modified peptidoglycan precursors, may be encoded in the genome. As a similar concentration of UDP-MurNAc-pentapeptide[D-Ala] is also produced in E. faecalis possessing a vanG gene cluster, a second critical feature could be involved in C. difficile susceptibility to vancomycin. Indeed, we observed an important modification of peptidoglycan precursors also found in mature peptidoglycan. Since 16% of the peptidoglycan precursors appeared to be amidated (amidated UDP-MurNAc-pentapeptide[D-Ala], amidated UDP-MurNAc-tetrapeptide and amidated UDP-MurNAc-pentapeptide[D-Ser]) in the presence of vancomycin compared with 45% in C. difficile harbouring the vanC-resistance operon, this amidation may either reflect direct induction or a consequence of the presence of vancomycin (Fig. 3). In the last case, amidation may occur because either UDP-MurNAc-pentapeptides, lipid I or lipid II were accumulated. Moreover, Peltier et al. reported that approximately 75% of dimers contain 3-3 interpeptide bridges in the peptidoglycan of C. difficile mediated by the linkage of two meso-DAP residues. Amidation of meso-DAP residues may therefore affect PBPs activity involved in peptidoglycan transpeptidation, (Pratt, 2008), slowing down or totally blocking peptidoglycan biosynthesis.

Several reports in the literature state that various resistance genes were found to be cryptic in their original host. For instances, various resistance mechanisms were found to be cryptic in their original bacterial host. For example cfiA from Bacteroides fragilis, oxa-23 from Acinetobacter radioresistens, aac(6′)-Iy from Samonella enterica and tetX from Bacteroides fragilis are silent resistance determinants (Speer et al., 1991; Podglajen et al., 1994; Magnet et al., 1999; Poirel et al., 2008). In the majority of cases the absence of resistance was due to lack of transcription. IS elements are able to bring promoter sequences leading for example, to imipenem resistance by expression of cfiA and oxa-23 in Bacteroides fragilis and Acinetobacter baumannii respectively. The aac(6′)-Iy gene confers aminoglycoside resistance when under the control of the nmp porin promoter following a large chromosomal deletion which leads to a transcriptional fusion. Intriguingly, the tetX determinant encoding tetracycline modification does confer resistance in Bacteroides fragilis but only in E. coli when the bacteria are grown under aerobic conditions.

In summary, our study shows that: (i) C. difficile is able to produce peptidoglycan precursors ending in D-Ala-D-Ser in the presence of vancomycin; (ii) C. difficile vanGCd mutant did not produce any of the modified peptidoglycan precursor; (iii) C. difficile vanGCd mutant exhibited a lower MIC compared with the 630 strain; (iv) survival rates of E. coli NR698 carrying vanGCd or vanC were similar; (v) D-Ala-D-Ser peptidoglycan precursors were detected in E. coli carrying the vanGCd resistance operon; (vi) the vanC resistance operon was not able to promote resistance when expressed into C. difficile or in the vanGCd mutant. This indicates that the regulation and resistance proteins encoded by the vanGCd gene cluster are functional. Lack of resistance of C. difficile may be due to several independent or associated factors, such as the weak activity of resistance proteins, a preferential incorporation of D-Ala-D-Ala dipeptides into peptidoglycan precursors by MurF, or a preferential use of lipid II ending with D-Ala-D-Ala instead of D-Ala-D-Ser or amidated lipid II by PBPs. These factors, combined with a gene cluster that confers a low level vancomycin resistance, may prevent the emergence of vancomycin resistance based on a D-Ala/D-Ser modification of peptidoglycan precursors in C. difficile. However, the vanGCd cluster is fully functional and largely distributed in C. difficile strains. The reasons why approximately 85% of C. difficile strains harboured this vancomycin resistance cluster (Ammam et al., 2012), why it has been acquired or lost and how it spreads remain unclear. As described by Ammam et al., integration of the vanGCd cluster occurred in a 19 bp inverted repeat and was flanked by a 17 bp direct repeat but cannot be assigned to a mobile element or a transposition event. The cluster is located upstream from CD630_16290 gene of unknown function and the presence of such an intergenic region (of 270 bp) is uncommon with an upregulation. Moreover, since this cluster is inducible, we could predict that it has a low cost for C. difficile fitness, according to the results of Foucault et al. with the vanB operon in enterococci (Foucault et al., 2010). The vanGCd mutant or strains lacking the vanGCd cluster had a vancomycin MIC < 1 mg l–1, which is lower than that for the 630 strain. The vanGCd cluster may therefore bring a slight advantage to C. difficile in the presence of very low concentration of vancomycin (≤ 2 mg l–1).

In conclusion, vanGCd was inducible by vancomycin and encoded functional proteins. It conferred a minor susceptibility change, but it cannot be excluded that genome modification(s) would allow vanGCd mediated vancomycin resistance to emerge. Therefore, to rule out the emergence of resistance in C. difficile, vancomycin should be confined to colitis relapses following the failure by metronidazole therapy.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Bacterial strains and growth conditions

Escherichia coli strains were grown at 37°C in LB (Luria Broth, Difco Laboratories) (Miller, 1972). E. coli TOP10, BL21 and TG1 were used as hosts for plasmid constructions or proteins production. E. coli HB101 (pRK24) was used for mating experiments. The following antibiotics (Eurobio) were added to the culture medium: kanamycin (40 mg l–1), ampicillin (100 mg l–1), chloramphenicol (25 mg l–1), or vancomycin (0.5 mg l–1, DAKOTA®Pharm). C. difficile strains (Table S1) were grown in an anaerobic chamber at 37°C in Brain Heart Infusion (BHI) or in Tryptone Yeast extract (TY) medium (Difco Laboratories). The following antibiotics were added to the culture medium: thiamphenicol (15 mg l–1, MP Biomedical), supplement (composed of 250 mg L-1 of D-cycloserine and of 8 mg ml–1 of cefoxitin Oxoid), erythromycin (10 mg l–1) and vancomycin (1.5 mg l–1). The MICs of vancomycin were determined by E-test (bioMérieux) on BHI agar in three independent experiments. The MICs were found to be nearly identical when determined on Muller–Hinton medium (data not shown).

DNA manipulation

Plasmid extraction, endonuclease digestion, ligation and agarose gel electrophoresis were carried out as described. (Maniatis et al., 1982). Polymerase chain reaction (PCR) for cloning was carried out with high fidelity Phusion DNA polymerase (Finnzymes) and for screening with Red'y'Gold Mix (Eurogentec). Restriction enzymes were used according to the manufacturer's instructions (Biolabs). All cloning steps were verified using restriction map and sequencing.

The intergenic region between vanSCd and vanGCd was amplified by PCR using primer pair (TC66/TC67). The 292 bp PCR product was digested with SacII/NdeI and cloned into similarly digested pMTL84151 plasmid (Heap et al., 2009) to generate pTC49. Primer pairs containing NdeI and BglII (TC53 and TC54; TC62 and TC63) or FauI and BglII (TC64 and TC65) restriction sites were used to amplify the vanTC, vanXYC and vanC genes respectively. The PCR products were cloned into pCR-Blunt vector (Invitrogen), sequenced and sub-cloned into pTC49 previously digested using NdeI and BglII giving rise to pTC56, pTC57, and pTC58 plasmids respectively. The vanCXYCTC genes were amplified from E. gallinarum BM4174 (Dutka-Malen et al., 1992) DNA using primer pair (TC54/TC62) and cloned into pCR-Blunt to give pFA21which was digested using NdeI/BglII and the insert was sub-cloned into similarly digested pTC49 to generate pFA27. The vanXYCd gene was amplified using primers FA35 and FA36 cloned into pBlunt TOPO vector and sub-cloned into pTC49 using NdeI/BamHI to give pFA24. All plasmids were transferred from E. coli HB101 (pRK24) to C. difficile 630 by heterogramic conjugation as described (Bouillaut et al., 2011).

A 4078 bp DNA fragment including the vanGCd, vanXYCd and vanTCd genes amplified using C. difficile 630 genomic DNA and TC32 and TC33 primers (Table S2) was cloned into pCR-Blunt giving rise to pTC27.

The primer pairs CdLBasF and CdLXhoR, CdLXYBsaF and CdXYXhoR, and CdTBsaF and CdTXhoR were used to amplify the vanGCd, vanXYCd genes and the C-terminal cytosolic domain of vanTCd, respectively, from C. difficile 630 using Pfu DNA polymerase (Fermentas) which were cloned into pCR-Blunt and sub-cloned into pET28a(+) (Novagen) previously digested using NcoI and XhoI, giving rise to pAT916, pAT917, and pAT918 respectively. The murF gene was amplified with FA37 and FA38 primers and cloned it into pMG25 (Kruse et al., 2003) using BamHI and EcoRI giving rise to pFA30.

The vanR and vanG clostron mutants (630Δerm vanR::erm and 630Δerm vanG::erm) were constructed as described by Heap et al. (Heap et al., 2010). Briefly, each clostron were amplified (see primers in Table S2) and cloned into pCR-Blunt vector (Invitrogen). Clostrons were then sub-cloned into pMTL007C-E5 and sequenced giving rise to pTC97 and pTC98, respectively for vanG and vanR clostrons. After conjugation in 630Δerm, mutants were screened using PCR. Introns were respectively introduced in base pair position 490 and 45 for vanG and vanR respectively.

Transcription analysis

Total RNA from C. difficile 630 grown at OD600 = 0.8 in TY or TY containing 1.5 mg l–1 vancomycin was extracted as follows. After centrifugation, culture pellets were suspended in RNApro™ solution and RNA extracted using the FastRNA Pro Blue Kit (MP Biomedicals). Absence of DNA was controlled by a real-time PCR (as described below) using FA15 and FA16 primers (Table S2). The RNA quality was determined using Bioanalyser Agilent 2100 and RNA 6000 Nano Reagents (Agilent).

For reverse transcription (final volume, 20 μl), 1 μg of total RNA was heated at 65°C for 5 min and cooled at 4°C. cDNA was synthesized for 1 h at 50°C in the presence of 240 ng of random primer (Invitrogen) and 200 U Superscript III Reverse Transcriptase (Invitrogen) which was then inactivated by incubation at 70°C for 15 min. An aliquot of 10 ng of cDNA was amplified by PCR using 20 pmol of each primer. Negative and positive controls were performed using, respectively, 10 ng DNA-free RNA and 10 ng of C. difficile DNA.

Primer 3 software (Rozen and Skaletsky, 2000) was used to design primers FA15, FA16, FA26, FA27, FA28, and FA29 (Table S2) used for real-time quantitative RT-PCR [qRT-PCR SYBR Green method (Roche)]. Each reaction was carried out in quadruplicate on a CFX (Bio-Rad) with the following programme: 45 cycles at 95°C for 10 s, 60°C for 5 s and 72°C for 10 s; a denaturing cycle at 95°C for 10 min. The internal reference gene was dnaF (Antunes et al., 2011). Relative expression levels of vanGCd and vanRCd genes were measured using Pfaffl mathematical model for relative quantification (Pfaffl, 2001). Comparisons between conditions were performed using the Student's paired t-test. Differences were considered significant at P ≤ 0.05.

Survival assay

E. coli and C. difficile were grown either in LB or BHI to OD = 1 at 600 nm. Dilutions up to obtain approximately 100 colony-forming units (CFU) were plated on medium without or with 0.5, 1 or 1.5 mg l–1 of vancomycin. Two plates were used for each dilution and experiment was performed at least three times independently. Survival rates were calculated by comparison of CFUs from plates with and without vancomycin.

Protein purification and enzymatic studies

VanGCd, VanXYCd, and VanTCd proteins were produced from E. coli BL21 (DE3)-pREP4 harbouring the corresponding plasmid. The C-terminal His6-tagged proteins were purified by nickel affinity chromatography as described for the purification of VanG from E. faecalis (Weber et al., 2009). The fractions containing the His6-tagged proteins were pooled, dialysed overnight against 50 mM HEPES, pH 7.5, 150 mM KCl, 20% glycerol (v/v), concentrated to 5–10 mg ml–1 and stored at –80°C. The purity of the proteins was over 98%. N-terminal sequences were determined by automated Edman degradation and the molecular masses were confirmed by surface-enhanced laser desorption/ionization time-of-flight (SELDI-TOF) mass spectrometry.

The initial rates of hydrolysis were determined with an Uvikon UV931 spectrophotometer (Kontron Instruments, Saint-Quentin-en-Yvelines, France) at 340 nm where the production of ADP is coupled to the oxidation of NADH through pyruvate kinase and L-lactate dehydrogenase (Daub et al., 1988). The steady-state kinetic parameters of vanGCd were determined by using the equations described previously (Neuhaus, 1962). Km1 for the N-terminal amino-acid, Km2 for the C-terminal amino-acid, and kcat were determined. To obtain the kcat and Km2 values, the enzyme was incubated in the presence of increasing concentrations of D-Ala (lowest concentration, 20 mM). Km1 was calculated from the activity measured in the presence of low concentrations of D-Ala (0.5–3 mM). The D-Ala : D-Ser ligase activity was measured in the presence of 10 mM D-Ala and increasing concentrations of D-Ser and was corrected for the background of D-Ala : D-Ala ligase activity.

The method for assaying D,D-dipeptidase and D,D-carboxypeptidase activities of VanXYCd was based on the amino-acid oxidase-lactate dehydrogenase coupled assay (Badet et al., 1984). The reaction was carried out in 0.5 ml containing 50 mM HEPES, pH 8.0, 0.2 mM NADH, 0.05 μg of lactate dehydrogenase (LDH), 0.15 μg of D-aminoacid oxidase (D-AAO), 500 units of catalase at 37°C, 0.5 μg of VanXYCd and various concentrations of D-Ala-D-Ala, D-Ala-D-Ser, UDP-MurNAc-pentapeptide[D-Ala], and UDP-MurNAc-pentapeptide[D-Ser]. The steady-state kinetic parameters were determined using the Hanes-Woolf plot (Henderson, 1992).

The method for assaying Ala- and Ser-racemase activities of VanTCd was the same as that for VanXYCd. VanTCd converts L-Ala and L-Ser into D-Ala and D-Ser respectively. D-AAO converts D-Ala and D-Ser into pyruvate and 3-hydroxy-pyruvate, respectively, which are reduced by LDH. The consumption of NADH, which reflects the formation of lactate and 2,3-dihydroxy-propane, respectively, was followed at 340 nm. The steady-state kinetic parameters were determined using the Hanes-Woolf plot (Henderson, 1992).

Recombinant protein MurF was purified using a BD Talon cobalt affinity resin (Clontech Laboratories) according to the protocol provided by the manufacturer. Elution fractions containing MurF were pooled, dialysed overnight against 100 mM Tris-HCl and 150 mM KCl (pH 8.6), and stored at –80°C after addition of equal volume of 100% glycerol. The MurF activity assay measures the formation of UDP-MurNAc-L-Ala-γ-D-Glu-meso-A2pm-D-Ala-D-[14C]Ala in a mixture (40 μl, final volume) containing 100 mM Tris-HCl, pH 7.8, 40 mM MgCl2, 5 mM ATP, 200 μM UDP-MurNAc-L-Ala-γ-D-Glu-meso-A2pm, 10–400 μM D-Ala-D-[14C]Ala (400 Bq) and enzyme (20 μl of an appropriate dilution in 20 mM phosphate buffer, pH 7.2, 1 mM dithiothreitol). After 30 min at 37°C, the reaction was stopped by the addition of glacial acetic acid (8 μl) followed by lyophilization. Radioactive substrate and product were separated by thin-layer chromatography on silica gel plates LK6D (Whatman, Maidstone, UK) using 1-propanol/ammonium hydroxide/water (6:3:1; v/v) as the mobile phase, and the radioactive spots were located and quantified with a radioactivity scanner (Rita Star, Raytest Isotopenmeßgeräte GmbH, Straubenhardt, Germany). For the determination of the kinetic parameters of D-Ala-D-Ser, a slightly different assay was used in which the formation of UDP-MurNAc-l-[14C]Ala-γ-D-Glu-meso-A2pm-D-Ala-D-Ser was measured. Mixtures were similar as above except that they contained 200 μM UDP-MurNAc-L-[14C]Ala-γ-D-Glu-meso-A2pm (400 Bq) and 0.1–4 mM D-Ala-D-Ser. Radioactive substrate and product were separated on a Nucleosil C18 5 μm column (150 × 4.6 mm; W. R. Grace S. A., Epernon, France) using 50 mM ammonium formate, pH 4.2, at a flow rate of 0.6 ml min–1. Radioactivity was detected with an LB 506 C-1 HPLC radioactivity monitor (Berthold France, Thoiry, France) using the Quicksafe Flow 2 scintillator (Zinsser Analytic, Maidenhead, UK) at 0.6 ml min–1. Quantification was performed with the Radiostar software (Berthold France). In all cases, the enzyme concentration was chosen so that substrate consumption was < 20%, the linearity being ensured within this interval even at the lowest substrate concentration. Data were fitted to the equation v = VmaxS/(Km + S) by the Levenberg–Marquardt method (Press et al., 1986) and values ± standard deviations at 95% of confidence were calculated. The MDFitt software developed by M. Desmadril (UMR 8619 CNRS, Orsay, France) was used for this purpose.

Peptidoglycan precursors and muropeptide analysis

Strains were grown in TY with or without vancomycin to the exponential phase (A600 = 0.9). A 1 l culture (corresponding to 209 mg) was rapidly chilled and harvested at 4°C. Analysis of the peptidoglycan precursors was performed as described previously (Mengin-Lecreulx et al., 1982). MALDI-TOF mass spectra were recorded on a Perseptive Voyager-DE STR instrument (Applied Biosystems, Carlsbad, CA) in the reflectron mode with delayed extraction. The samples were prepared as follows. One microlitre of a 10 mg ml–1 solution of 2,5-dihydroxybenzoic acid in 0.1 M citric acid was deposited on the plate and mixed with 0.5 or 1 μl of compound at 20 pmol ml–1 in water. After evaporation, desorption and ionization were obtained by pulses from a 337 nm nitrogen laser. Spectra were recorded in the negative mode at an acceleration voltage of –20 kV and an extraction delay time of 200 nanoseconds. A mixture of UDP-MurNAc, UDP-MurNAc-dipeptide and UDP-MurNAc-pentapeptide was used as an external calibrant.

Peptidoglycan was extracted as described by Candela and Fouet (Candela & Fouet, 2005), without the acetone step, from 200 ml stationary phase cultures grown in the absence or the presence of vancomycin. Peptidoglycan was digested with mutanolysin (10 mg ml–1) and reduced with sodium borohydride. Muropeptides were then separated on a reverse-phase Hypersil ODS18 column (250 × 4.6 mm, 3 μm particle size) with a gradient of methanol (0 to 15%) in sodium phosphate buffer (pH 4.3 to 5) for 120 min at 1 ml min–1. Detection of peaks was performed at 206 nm. The collected muropeptides were desalted on the same column but in 0.05% trifluoroacetic acid and a 0 to 25% acetonitrile gradient. MALDI-TOF mass spectra and MS-MS analysis of peptidoglycan muropeptides were performed using a 4800 MALDI-TOF/TOF (AB-MDS-SCIEX) (Proteopole at Pasteur Institute).

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Thanks to K. Gerdes and N. Minton for providing pMG25 and pMTL84151, respectively, J.C. Marvaud for pAT918 construction and M. Monot and C. Deloménie for helpful discussions. F.A. was funded by MNRT. This work was in part supported by an ERC starting Grant to IGB (PGNfromSHAPEtoVIR n°202283) and by an ANR Grant (VANORAMA; ANR-07-MIME-026-02).

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Abadia Patino, L., Courvalin, P., and Perichon, B. (2002) vanE gene cluster of vancomycin-resistant Enterococcus faecalis BM4405. J Bacteriol 184: 64576464.
  • Ammam, F., Marvaud, J.C., and Lambert, T. (2012) Distribution of the vanG-like gene cluster in Clostridium difficile clinical isolates. Can J Microbiol 58: 547551.
  • Antunes, A., Martin-Verstraete, I., and Dupuy, B. (2011) CcpA-mediated repression of Clostridium difficile toxin gene expression. Mol Microbiol 79: 882899.
  • Arias, C.A., and Murray, B.E. (2012) The rise of the Enterococcus: beyond vancomycin resistance. Nat Rev Microbiol 10: 266278.
  • Arias, C.A., Courvalin, P., and Reynolds, P.E. (2000a) vanC cluster of vancomycin-resistant Enterococcus gallinarum BM4174. Antimicrob Agents Chemother 44: 16601666.
  • Arias, C.A., Weisner, J., Blackburn, J.M., and Reynolds, P.E. (2000b) Serine and alanine racemase activities of VanT: a protein necessary for vancomycin resistance in Enterococcus gallinarum BM4174. Microbiology 146 (Part 7): 17271734.
  • Arias, C.A., Pena, J., Panesso, D., and Reynolds, P. (2003) Role of the transmembrane domain of the VanT serine racemase in resistance to vancomycin in Enterococcus gallinarum BM4174. J Antimicrob Chemother 51: 557564.
  • Arthur, M., Molinas, C., and Courvalin, P. (1992) The VanS-VanR two-component regulatory system controls synthesis of depsipeptide peptidoglycan precursors in Enterococcus faecium BM4147. J Bacteriol 174: 25822591.
  • Arthur, M., Depardieu, F., Reynolds, P., and Courvalin, P. (1996) Quantitative analysis of the metabolism of soluble cytoplasmic peptidoglycan precursors of glycopeptide-resistant enterococci. Mol Microbiol 21: 3344.
  • Badet, B., Roise, D., and Walsh, C.T. (1984) Inactivation of the dadB Salmonella typhimurium alanine racemase by D and L isomers of beta-substituted alanines: kinetics, stoichiometry, active site peptide sequencing, and reaction mechanism. Biochemistry 23: 51885194.
  • Barbut, F., and Petit, J.C. (2000) [Epidemiology, risk factors and prevention of Clostridium difficile nosocomial infections]. Pathol Biol (Paris) 48: 745755.
  • Barreteau, H., Kovač, A., Boniface, A., Sova, M., Gobec, S., and Blanot, D. (2008) Cytoplasmic steps of peptidoglycan biosynthesis. FEMS Microbiol Rev 32: 168207.
  • Bauer, D. (1972) Constructing confidence sets using rank statistics. J Am Stat Assoc 67: 687690.
  • Billot-Klein, D., Gutmann, L., Sable, S., Guittet, E., and 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, and Enterococcus gallinarum. J Bacteriol 176: 23982405.
  • Bisicchia, P., Bui, N.K., Aldridge, C., Vollmer, W., and Devine, K.M. (2011) Acquisition of VanB-type vancomycin resistance by Bacillus subtilis: the impact on gene expression, cell wall composition and morphology. Mol Microbiol 81: 157178.
  • Blanckaert, K., Coignard, B., Grandbastien, B., Astagneau, P., and Barbut, F. (2008) [Update on Clostridium difficile infections]. Rev Med Interne 29: 209214.
  • Bouillaut, L., McBride, S.M., and Sorg, J.A. (2011) Genetic manipulation of Clostridium difficile. Curr Protoc Microbiol Chapter 9: Unit 9A 2.
  • Boyd, D.A., Willey, B.M., Fawcett, D., Gillani, N., and Mulvey, M.R. (2008) Molecular characterization of Enterococcus faecalis N06-0364 with low-level vancomycin resistance harboring a novel D-Ala-D-Ser gene cluster, vanL. Antimicrob Agents Chemother 52: 26672672.
  • Candela, T., and Fouet, A. (2005) Bacillus anthracis CapD, belonging to the gamma-glutamyltranspeptidase family, is required for the covalent anchoring of capsule to peptidoglycan. Mol Microbiol 57: 717726.
  • Casadewall, B., and Courvalin, P. (1999) Characterization of the vanD glycopeptide resistance gene cluster from Enterococcus faecium BM4339. J Bacteriol 181: 36443648.
  • Courvalin, P. (2006) Vancomycin resistance in gram-positive cocci. Clin Infect Dis 42 (Suppl. 1): S25S34.
  • Daub, E., Zawadzke, L.E., Botstein, D., and Walsh, C.T. (1988) Isolation, cloning, and sequencing of the Salmonella typhimurium ddlA gene with purification and characterization of its product, D-alanine:D-alanine ligase (ADP forming). Biochemistry 27: 37013708.
  • Depardieu, F., Bonora, M.G., Reynolds, P.E., and Courvalin, P. (2003) The vanG glycopeptide resistance operon from Enterococcus faecalis revisited. Mol Microbiol 50: 931948.
  • Depardieu, F., Foucault, M.L., Bell, J., Dubouix, A., Guibert, M., Lavigne, J.P., et al. (2009) New combinations of mutations in VanD-Type vancomycin-resistant Enterococcus faecium, Enterococcus faecalis, and Enterococcus avium strains. Antimicrob Agents Chemother 53: 19521963.
  • Dutka-Malen, S., Molinas, C., Arthur, M., and 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: 5358.
  • Foucault, M.L., Depardieu, F., Courvalin, P., and Grillot-Courvalin, C. (2010) Inducible expression eliminates the fitness cost of vancomycin resistance in enterococci. Proc Natl Acad Sci USA 107: 1696416969.
  • Gholizadeh, Y., Prevost, M., Van Bambeke, F., Casadewall, B., Tulkens, P.M., and Courvalin, P. (2001) Sequencing of the ddl gene and modeling of the mutated D-alanine:D-alanine ligase in glycopeptide-dependent strains of Enterococcus faecium. Protein Sci 10: 836844.
  • Heap, J.T., Pennington, O.J., Cartman, S.T., and Minton, N.P. (2009) A modular system for Clostridium shuttle plasmids. J Microbiol Methods 78: 7985.
  • Heap, J.T., Kuehne, S.A., Ehsaan, M., Cartman, S.T., Cooksley, C.M., Scott, J.C., and Minton, N.P. (2010) The ClosTron: mutagenesis in Clostridium refined and streamlined. J Microbiol Methods 80: 4955.
  • Henderson, P. (1992) Statistical Analysis of Enzymes Kinetic Data. Oxford: IRL Press.
  • Koteva, K., Hong, H.J., Wang, X.D., Nazi, I., Hughes, D., Naldrett, M.J., et al. (2010) A vancomycin photoprobe identifies the histidine kinase VanSsc as a vancomycin receptor. Nat Chem Biol 6: 327329.
  • Kruse, T., Moller-Jensen, J., Lobner-Olesen, A., and Gerdes, K. (2003) Dysfunctional MreB inhibits chromosome segregation in Escherichia coli. EMBO J 22: 52835292.
  • Lambert, M.P., and Neuhaus, F.C. (1972) Mechanism of D-cycloserine action: alanine racemase from Escherichia coli W. J Bacteriol 110: 978987.
  • Lebreton, F., Depardieu, F., Bourdon, N., Fines-Guyon, M., Berger, P., Camiade, S., et al. (2011) D-Ala-D-Ser VanN-type transferable vancomycin resistance in Enterococcus faecium. Antimicrob Agents Chemother 55: 46064612.
  • Magnet, S., Courvalin, P., and Lambert, T. (1999) Activation of the cryptic aac(6′)-Iy aminoglycoside resistance gene of Salmonella by a chromosomal deletion generating a transcriptional fusion. J Bacteriol 181: 66506655.
  • Maniatis, T., Fritsch, E.F., and Sambrook, J. (1982) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press.
  • Mengin-Lecreulx, D., Flouret, B., and van Heijenoort, J. (1982) Cytoplasmic steps of peptidoglycan synthesis in Escherichia coli. J Bacteriol 151: 11091117.
  • Miller, J.H. (1972) Experiments in Molecular Genetics. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.
  • Monot, M., Boursaux-Eude, C., Thibonnier, M., Vallenet, D., Moszer, I., Medigue, C., et al. (2011) Reannotation of the genome sequence of Clostridium difficile strain 630. J Med Microbiol 60: 11931199.
  • Neuhaus, F.C. (1962) The enzymatic synthesis of D-alanyl-D-alanine. II. Kinetic studies on D-alanyl-D-alanine synthetase. J Biol Chem 237: 31283135.
  • Peltier, J., Courtin, P., Meouche, I.E., Lemee, L., Chapot-Chartier, M.P., and Pons, J.L. (2011) Clostridium difficile has an original peptidoglycan structure with a high level of N-acetylglucosamine deacetylation and mainly 3-3 cross-links. J Biol Chem 286: 2905329062.
  • Perichon, B., and Courvalin, P. (2012) Staphylococcus aureus VRSA-11B is a constitutive vancomycin-resistant mutant of vancomycin-dependent VRSA-11A. Antimicrob Agents Chemother 9: 46934696.
  • Pfaffl, M.W. (2001) A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res 29: e45.
  • Podglajen, I., Breuil, J., and Collatz, E. (1994) Insertion of a novel DNA sequence, 1S1186, upstream of the silent carbapenemase gene cfiA, promotes expression of carbapenem resistance in clinical isolates of Bacteroides fragilis. Mol Microbiol 12: 105114.
  • Podmore, A.H., and Reynolds, P.E. (2002) Purification and characterization of VanXY(C), a D,D-dipeptidase/D,D-carboxypeptidase in vancomycin-resistant Enterococcus gallinarum BM4174. Eur J Biochem 269: 27402746.
  • Poirel, L., Figueiredo, S., Cattoir, V., Carattoli, A., and Nordmann, P. (2008) Acinetobacter radioresistens as a silent source of carbapenem resistance for Acinetobacter spp. Antimicrob Agents Chemother 52: 12521256.
  • Pratt, R.F. (2008) Substrate specificity of bacterial DD-peptidases (penicillin-binding proteins). Cell Mol Life Sci 65: 21382155.
  • Press, W.H., Flannery, B.P., Teukolsky, S.A., and Vetterling, W.T. (1986) Numerical Recipes: The Art of Scientific Computing. Cambridge: Cambridge University Press.
  • Prevost, M., Van Belle, D., Tulkens, P.M., Courvalin, P., and Van Bambeke, F. (2000) Modeling of Enterococcus faecalis D-alanine:D-alanine ligase: structure-based study of the active site in the wild-type enzyme and in glycopeptide-dependent mutants. J Mol Microbiol Biotechnol 2: 321330.
  • Reynolds, P.E. (1989) Structure, biochemistry and mechanism of action of glycopeptide antibiotics. Eur J Clin Microbiol Infect Dis 8: 943950.
  • Reynolds, P.E., and Courvalin, P. (2005) Vancomycin resistance in enterococci due to synthesis of precursors terminating in D-alanyl-D-serine. Antimicrob Agents Chemother 49: 2125.
  • Rozen, S., and Skaletsky, H. (2000) Primer3 on the WWW for general users and for biologist programmers. Methods Mol Biol 132: 365386.
  • Ruiz, N., Falcone, B., Kahne, D., and Silhavy, T.J. (2005) Chemical conditionality: a genetic strategy to probe organelle assembly. Cell 121: 307317.
  • Sebaihia, M., Wren, B.W., Mullany, P., Fairweather, N.F., Minton, N., Stabler, R., et al. (2006) The multidrug-resistant human pathogen Clostridium difficile has a highly mobile, mosaic genome. Nat Genet 38: 779786.
  • Speer, B.S., Bedzyk, L., and Salyers, A.A. (1991) Evidence that a novel tetracycline resistance gene found on two Bacteroides transposons encodes an NADP-requiring oxidoreductase. J Bacteriol 173: 176183.
  • Surawicz, C.M., and Alexander, J. (2011) Treatment of refractory and recurrent Clostridium difficile infection. Nat Rev Gastroenterol Hepatol 8: 330339.
  • Szybalski, W., and Bryson, V. (1952) Genetic studies on microbial cross resistance to toxic agents. I. Cross resistance of Escherichia coli to fifteen antibiotics. J Bacteriol 64: 489499.
  • Tillotson, G.S., and Tillotson, J. (2011) Clostridium difficile – a moving target. F1000 Med Rep 3: 6.
  • Tytgat, I., Colacino, E., Tulkens, P.M., Poupaert, J.H., Prevost, M., and Van Bambeke, F. (2009) DD-ligases as a potential target for antibiotics: past, present and future. Curr Med Chem 16: 25662580.
  • Vollmer, W., Blanot, D., and de Pedro, M.A. (2008) Peptidoglycan structure and architecture. FEMS Microbiol Rev 32: 149167.
  • Weber, P., Meziane-Cherif, D., Haouz, A., Saul, F.A., and Courvalin, P. (2009) Crystallization and preliminary X-ray analysis of a D-Ala:D-Ser ligase associated with VanG-type vancomycin resistance. Acta Crystallogr Sect F Struct Biol Cryst Commun 65: 10241026.
  • Xu, X., Lin, D., Yan, G., Ye, X., Wu, S., Guo, Y., et al. (2010) vanM, a new glycopeptide resistance gene cluster found in Enterococcus faecium. Antimicrob Agents Chemother 54: 46434647.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information
FilenameFormatSizeDescription
mmi12299-sup-0001-si.pdf162KSupporting Information

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.