SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Nucleotide sequence accession numbers
  8. Conflict of interest
  9. References
  10. Supporting Information

The glycopeptide vancomycin was until recently considered a drug of last resort against Gram-positive bacteria. Increasing numbers of bacteria, however, are found to carry genes that confer resistance to this antibiotic. So far, 10 different vancomycin resistance clusters have been described. A chromosomal vancomycin resistance gene cluster was previously described for the anaerobic Desulfitobacterium hafniense Y51. We demonstrate that this gene cluster, characterized by its d-Ala-d-Lac ligase-encoding vanI gene, is present in all strains of D. hafniense, D. chlororespirans and some strains of Desulfosporosinus spp. This gene cluster was not found in vancomycin-sensitive Desulfitobacterium or Desulfosporosinus spp., and we show that this antibiotic resistance can be exploited as an intrinsic selection marker for Desulfitobacterium hafniense and D. chlororespirans. The gene cluster containing vanI is phylogenetically only distantly related with those described from soil and gut bacteria, but clusters instead with vancomycin resistance genes found within the phylum Actinobacteria that include several vancomycin-producing bacteria. It lacks a vanH homologue, encoding a D-lactate dehydrogenase, previously thought to always be present within vancomycin resistance gene clusters. The location of vanH outside the resistance gene cluster likely hinders horizontal gene transfer. Hence, the vancomycin resistance cluster in D. hafniense should be regarded a novel one that we here designated vanI after its unique d-Ala-d-Lac ligase.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Nucleotide sequence accession numbers
  8. Conflict of interest
  9. References
  10. Supporting Information

Antibiotic resistance is a two-edged sword. On one hand, it has proven extremely useful, as a convenient selection marker for biotechnological applications and a fast way to eliminate the susceptible fraction from enrichment cultures (Löffler et al., 2005; Schweizer, 2008). On the other hand, the spread of antibiotic resistance among bacteria can lead to multi-resistant pathogens, making their inhibition one of the most serious challenges in modern healthcare (Davies and Davies, 2010).

The glycopeptides vancomycin and teicoplanin are widely used antibiotics for treatment of infections with Gram-positive bacteria. They act by binding to the d-alanyl-d-alanine (d-Ala-d-Ala) terminus of intermediates in peptidoglycan formation, thereby inhibiting cell wall cross-linking (Courvalin, 2006). Since the late 1980s, vancomycin-resistant pathogens have emerged in hospitals worldwide. Vancomycin resistance results from the production of modified peptidoglycan precursors terminating in either d-alanyl-d-lactate (d-Ala-d-Lac) or d-alanyl-d-serine (d-Ala-d-Ser) to which vancomycin exhibits low binding affinities. Vancomycin resistance is classified into clusters based on the DNA sequence of the ligase gene vanA and its homologues that encode the key enzyme in the synthesis of d-Ala-d-Lac or d-Ala-d-Ser. Currently, 10 different types of vancomycin resistance gene clusters have been identified. d-Ala-d-Lac type resistance is associated with clusters vanA, vanB, vanD, vanF and vanM, whereas d-Ala-d-Ser resistance is linked to clusters vanC, vanE, vanG, vanL and vanN (Fraimow et al., 2005; Depardieu et al., 2007; Boyd et al., 2008; Xu et al., 2010; Lebreton et al., 2011). The vanA cluster also confers a high level of resistance to teicoplanin, whereas the other gene clusters of the d-Ala-d-Lac type give less or no resistance to teicoplanin (Kalan et al., 2009; Xu et al., 2010).

Three genes are considered essential for d-Ala-d-Lac-mediated vancomycin resistance: vanH, vanA and homologues, and vanX. These are found in three genes forming the vanHAX core that is conserved across pathogenic bacteria, glycopeptide antibiotic producers and other environmental bacteria (Fig. 1) (Marshall et al., 1998, Kalan et al., 2009 and the references cited therein). The vanH gene encodes a d-lactate dehydrogenase that converts pyruvate to d-lactate, vanA and homologues encode an ATP-dependent depsipeptide ligase involved in the synthesis of d-Ala-d-Lac, and vanX encodes a dipeptidase that cleaves d-Ala-d-Ala present in the cell, thereby enriching the cell wall in d-Ala-d-Lac. These key genes are found together with a flexible pool of accessory genes, such as those encoding D,D-carboxypeptidases and regulatory proteins (Fig. 1) (Depardieu et al., 2007; Hong et al., 2008).

figure

Figure 1. Alignment of vancomycin resistance gene clusters representing the currently known van gene clusters (see also Fig. 3). Arrows show the direction of transcription, while arrow size indicates gene size. Letters inside arrows denote gene names with the ‘van’ prefix removed. Empty arrows are hypothetical genes. Light grey arrows represent the d-Ala-d-Lac ligases giving name to the gene clusters. Numbers under arrows indicate gene size in bp, and numbers in italics above d-Ala-d-Lac ligases show percentage amino acid identity to VanI from Desulfitobacterium hafniense strain Y51. *VanM according to (Xu et al., 2010), VanM according to (Aminov and Mackie, 2007). Note that vanI previously was referred to as vanA by Kalan and coauthors (Kalan et al., 2009). Accession numbers are given in parentheses.

Download figure to PowerPoint

Desulfitobacterium spp. include strictly anaerobic bacteria that have mainly been isolated from environments contaminated with halogenated organic compounds. Most strains of Desulfitobacterium spp. isolated to date are able to degrade one or more halogenated compounds by organohalide respiration (Villemur et al., 2006). One of the few exceptions is D. hafniense DP7, which was isolated from a human faecal sample (van de Pas et al., 2001). Previously, a vancomycin resistance gene cluster was identified in the tetrachloroethene (PCE) dehalogenating bacterium Desulfitobacterium hafniense Y51 (Kalan et al., 2009). The predicted vanA gene was not found in the usual context of three genes forming the vanHAX core, but rather was present in a cluster consisting of seven genes, referred to as vanXmurFvanKWARS (Fig. 1) (Kalan et al., 2009). This gene cluster does not contain a vanH gene, unlike clusters previously described for vancomycin-resistant bacteria. Functional expression of the genes in D. hafniense Y51 was confirmed by resistance to vancomycin, with a minimal inhibitory concentration (MIC) for growth of 64 μg/ml (Kalan et al., 2009).

In the present study, we investigated the distribution and phylogenetic affiliation of vancomycin resistance among Desulfitobacterium spp. of environmental and human origin. In addition, it was investigated whether vancomycin could be successfully used as a selective agent for isolation of Desulfitobacterium spp. from enrichment cultures.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Nucleotide sequence accession numbers
  8. Conflict of interest
  9. References
  10. Supporting Information

d-Ala-d-Lac ligases-in Desulfitobacterium spp

Chromosomal DNA was isolated from 13 Desulfitobacterium spp. strains, covering the majority of the currently described species of this genus. The DNA sequences of the two d-Ala-d-Lac ligase-encoding genes from D. hafniense Y51 (DSY3690) and D. hafniense DCB-2 (Dhaf_1664) were used to design primers for the detection of homologous d-Ala-d-Lac ligase genes in other Desulfitobacterium spp. strains. Amplicons of the expected size of 987 bp were obtained for nine of the 13 tested strains (Table 1). All nine polymerase chain reaction (PCR)-positive isolates were strains of D. hafniense, with the exception of D. chlororespirans, which is the closest relative of D. hafniense based on 16S rRNA gene similarity (Fig. 2). Sequence analysis of the obtained PCR products confirmed that they encode d-Ala-d-Lac ligases (data not shown).

Table 1. Desulfitobacterium spp. and Desulfosporosinus spp. strains used in this study
#StrainPresence of vanIVancomycin MIC [μg/ml]Teicoplanin MIC [μg/ml]
  1. a

    Received as a kind gift from Masatoshi Goto, Department of Bioscience and Biotechnology, Faculty of Agriculture, Kyushu University, Japan.

  2. b

    Received as a kind gift from Jan Dirk van Elsas, Department of Microbial Ecology, University of Groningen, The Netherlands.

  3. c

    DSM, Leibniz Institute DSMZ – German Collection of Microorganisms and Cell Cultures, Braunschweig, Germany.

  4. d

    Received as a kind gift from Nico Boon, Laboratory of Microbial Ecology and Technology, Ghent University, Belgium.

  5. e

    JGI gene ID.

  6. Accession numbers are given in parentheses.

(a) Desulfitobacterium spp. 
 1D. hafniense Y51aYes (AP008230.1)> 250> 50 < 75
 2D. hafniense TCE1bYes (HQ433581)> 250> 50 < 75
 3D. hafniense DP7 (DSMc 13498)Yes (HQ433580)> 150 < 250> 25 < 50
 4D. hafniense DCB-2 (DSMc 10664)Yes (CP001336.1)> 150 < 250> 25 < 50
 5D. hafniense PCE-S (DSMc 14645)Yes (HQ433585)> 250> 25 < 50
 6D. hafniense PCP-1 (DSMc 12420)Yes (HQ433583)> 150 < 250> 50 < 75
 7D. hafniense TCP-A (DSMc13557)Yes (HQ433582)> 250> 25 < 50
 8D. hafniense G2 (DSMc 16228)Yes (HQ433584)> 250> 50 < 75
 9D. chlororespirans (DSMc 11544)Yes (HQ433586)> 250> 50 < 75
10D. dehalogenans (DSMc 9161)No< 1< 1
11D. sp. PCE1 (DSMc 10344)No< 1< 1
12D. metallireducens (DSMc 15288)NoNDND
13D. dichloroeliminans DCA1dNoNDND
(b) Desulfosporosinus spp.
14D. youngia (DSMc 17734)Yes (2508507521)e> 10 < 25> 5 < 10
15D. orientis (DSMc 765)No< 1< 1
16D. meridiei (DSMc 13257)No< 1< 1
17D. acidiphilus (DSMc 22704)No< 1< 1
figure

Figure 2. 16S rRNA gene-based phylogenetic tree of Desulfitobacterium spp. and Desulfosporosinus spp. Sequences were aligned with muscle (Edgar, 2004), and a neighbour-joining tree was constructed and validated with 1000 bootstraps analysis using the mega5 software package (Tamura et al., 2011). Bootstrap values higher than 50% are given at corresponding nodes in the tree. The reference bar indicates 0.2% base substitutions per site. GenBank accession numbers of sequences used in this analysis are given in parentheses.

Download figure to PowerPoint

figure

Figure 3. Phylogenetic tree based on a 512 bp fragment of d-Ala-d-Lac ligase encoding genes representing the currently known Van gene clusters of the d-Ala-d-Lac type. Corresponding reference sequences (Aminov and Mackie, 2007; Xu et al., 2010), vanI sequences obtained in this study and sequences retrieved from genome data sets (Teo et al., 2011; Pester et al., 2012; Markowitz et al., 2012a) were aligned with muscle (Edgar, 2004), and a neighbour-joining tree was constructed and validated with 1000 bootstraps analysis using the mega5 software package (Tamura et al., 2011). *VanM according to (Xu et al., 2010), VanM according to (Aminov and Mackie, 2007). Bootstrap values higher than 50% are given at corresponding nodes in the tree. The reference bar indicates 5% base substitutions per site. GenBank accession numbers or JGI gene IDs of sequences used in this analysis are given in parentheses. Definition of van gene clusters was adapted from (Aminov and Mackie, 2007).

Download figure to PowerPoint

Phylogenetic analysis

Previously, a phylogenetic analysis was performed using a data set of concatenated open reading frames (ORFs) comprising vanH, vanA and its homologues, and vanX (Aminov and Mackie, 2007). This analysis showed a very early branching event, leading to the formation of two large clusters, one of which consists of genes associated with vancomycin-resistant enterobacteria and other vancomycin-resistant gut and soil bacteria (cluster I). The second cluster consists of members of the Actinobacteria, some of which have been shown to produce vancomycin (cluster II). Furthermore, this phylogenetic analysis suggested that vanHAX and homologous gene clusters behave as a single evolutionary unit (Aminov and Mackie, 2007).

We constructed a phylogenetic tree using the nucleotide sequences of 512 bp fragments corresponding to part of the d-Ala-d-Lac ligase genes obtained in this study. This tree showed essentially the same clustering as when using the sequences of the entire vanHAX and homologous gene clusters (data not shown), indicating that the sequence fragments obtained in this study are sufficient for robust phylogenetic analysis. We then included in the phylogenetic tree the 512 bp sequence fragments obtained from Desulfitobacterium spp., closely related Desulfosporosinus spp., and sequences from the newly described vanM cluster (Abicht et al., 2011; Teo et al., 2011), resulting in a tree containing representative sequences from all currently described vancomycin resistance clusters of the d-Ala-d-Lac type (Fig. 3). The phylogenetic tree shows two main and deep-branching clusters. Cluster I consists of vanA, vanM*, vanF, vanB and vanD, while cluster II consists of vanM, vanN, and a separate subcluster comprising all d-Ala-d-Lac ligase-encoding sequences obtained from Desulfitobacterium spp. and Desulfosporosinus spp. (Fig. 3). Although the sequences obtained from glycopeptide producing Actinobacteria, Desulfitobacterium spp. and Desulfosporosinus spp. belong to cluster II, it can be assumed that desulfitobacteria do not produce vancomycin, as no vancomycin synthesis pathway was identified in the two published genomes of Desulfitobacterium hafniense DCB-2 and Y51 (Nonaka et al., 2006; Kim et al., 2012), nor in the available draft genome sequence of strain DP7 (AFZX00000000.1).

Sequences obtained from Desulfitobacterium spp. and Desulfosporosinus spp. were found to be more similar to those of the d-Ala-d-Lac ligases belonging to cluster II than to any of the sequences from cluster I (Supporting Information Table S1). Based on these results, it is clear that the d-Ala-d-Lac ligase gene found in Desulfitobacterium spp. and Desulfosporosinus spp. does not belong to the vanA cluster, but instead is part of a distinct vancomycin resistance cluster, which we propose to name vanI after its unique d-Ala-d-Lac ligase.

Presence of vanI outside the Desulfitobacterium genus

BlastP analysis against all sequenced bacterial genomes identified vanI homologues in the genomes of both Desulfosporosinus youngiae JW/Y JL-B18 and Desulfosporosinus sp. OT (Fig. 3), but not in Desulfosporosinus orientis Singapore I, D. meridiei S10 or D. acidiphilus SJ4 (Abicht et al., 2011; Pester et al., 2012; Markowitz et al., 2012a). We did not detect vanI-like genes in any of the metagenomes publicly available in the JGI IMG/M repository (as of 4 February 2013), demonstrating that vanI genes are not widespread in the environment or the human gut.

Architecture of the vanI gene cluster

Inspection of the genomic sequences of Desulfitobacterium spp. and Desulfosporosinus spp. revealed the vanXImurFIvanKIWIIRISI (vanXmurFvanKWARS in Kalan et al., 2009) gene cluster to be conserved (Fig. 1). However, it should be noted that vanKI, encoding the enzyme adding the branched amino acid(s) to the stem pentapeptide of peptidoglycan precursors (Hong et al., 2005), was only found in Desulfitobacterium hafniense Y51 and Desulfosporosinus sp. OT. An interesting feature of the vanI gene cluster is the location of the gene encoding the d-Lac dehydrogenase (vanH) outside the vancomycin resistance gene cluster. Usually, vanH genes are part of the three genes vanHAX or homologues that form the core of vancomycin resistance gene clusters (Hong et al., 2008). Previously, four genes have been proposed as potential vanH-encoding genes (DSY0996, DSY1673, DSY3442 and DSY4020) in the genome of D. hafniense Y51 (Kalan et al., 2009). The homologues of these four genes were also found in the genomes of D. hafniense DCB-2 (Dhaf_1344, Dhaf_1980, Dhaf_2079 and Dhaf_2820) and D. hafniense DP7 (data not shown). The complete genomes of D. hafniense Y51 and DCB-2 (Nonaka et al., 2006; Kim et al., 2012) were aligned and inspected for signs of horizontal transfer of the vancomycin resistance gene cluster. No genes encoding integrases or transposases, nor large deviations in dinucleotide frequency or GC content, were found in the vicinity of the vancomycin resistance gene clusters, suggesting that they are not located on mobile genomic islands.

Vancomycin and teicoplanin susceptibility

In total, 15 strains belonging to three different species of Desulfitobacterium and four species of Desulfosporosinus were tested for growth on plates containing different concentrations of vancomycin or teicoplanin. All strains for which a vanI gene was detected are resistant to vancomycin and teicoplanin (Table 1). Strains of Desulfitobacterium spp. showed high level of resistance with MIC values > 150 μg/ml vancomycin and > 25 ug/ml teicoplanin (Table 1). The two strains that tested negative for vanI genes, namely D. dehalogenans and the closely related strain PCE1, are highly susceptible to both vancomycin and teicoplanin, with MIC values < 1 μg/ml. The same was the case for the tested members of the Desulfosporosinus genus. Desulfosporosinus youngia, whose genome contains a vanI gene cluster, is resistant to vancomycin and teicoplanin, although their MICs were approximately tenfold lower than those observed for Desulfitobacterium spp. The three tested Desulfosporosinus spp. strains lacking a vanI gene are all susceptible to both vancomycin and teicoplanin (Table 1).

Vancomycin as selective agent

Previous experiments with cultures of the PCE-dehalogenating strain D. hafniense TCE1, received from collaborators as well as from Deutsche Sammlung von Mikroorganismen und Zellkulturen (DSMZ), led us to suspect that the cultures contained a microbial contaminant. Partial 16S rRNA gene sequencing confirmed that the D. hafniense TCE1 culture was contaminated with a Sedimentibacter sp. (97% 16S rRNA gene sequence identity to Sedimentibacter sp. C7, AY766466.1) (data not shown). Sedimentibacter spp. are gram-positive, members of the Firmicutes, and as a consequence likely susceptible to vancomycin (Breitenstein et al., 2002). We, therefore, decided to test vancomycin as a potential selective agent for the elimination of the Sedimentibacter contaminant.

A contaminated D. hafniense TCE1 culture was first grown in liquid medium with 20 mM lactate and 10 mM PCE as electron donor and acceptor, respectively, in order to preselect for the dechlorinating Desulfitobacterium. Subsequently, an early exponential growth phase culture was plated on medium containing 40 mM pyruvate and supplemented with 50 μg/ml vancomycin. After 1 week, only one colony type was observed. Two colonies were picked and grown in liquid medium containing 20 mM lactate and 10 mM PCE, supplemented with 50 μg/ml vancomycin and subsequently transferred to liquid medium containing 40 mM pyruvate and no vancomycin. This resulted in a pure culture of D. hafniense TCE1 without the presence of any Sedimentibacter, as confirmed by screening of a general bacterial 16S rRNA gene clone library with primers specific for Desulfitobacterium spp.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Nucleotide sequence accession numbers
  8. Conflict of interest
  9. References
  10. Supporting Information

In this contribution, the distribution of vancomycin resistance among Desulfitobacterium spp. and Desulfosporosinus spp. is described. Nine of the 13 tested strains of Desulfitobacterium spp. display high levels of vancomycin resistance, whereas one out of four strains of Desulfosporosinus spp. was found moderately vancomycin-resistant (Table 1). Previously, a vancomycin MIC value of 64 μg/ml was reported for D. hafniense Y51 (Kalan et al., 2009), which is considerably lower than the values obtained in this study (Table 1). A possible explanation for the higher MIC values could be the differences in the plating medium, and a protocol between this study and the study of Kalan and colleagues (Kalan et al., 2009). It was, for instance, demonstrated that the growth on plates of Desulfitobacterium dehalogenans was much improved when 0.8% of highly purified agar was used instead of 1.5% agar of standard purity (Smidt et al., 1999).

Phylogenetic analysis of predicted d-Ala-d-Lac ligase encoding genes showed that those obtained from Desulfitobacterium spp. together with Desulfosporosinus spp. belong to cluster II where they form a distinct branch, next to genes retrieved from glycopeptide-producing members of the phylum Actinobacteria (Fig. 3). It has often been speculated that antibiotic resistance, including vancomycin resistance, was developed as a mechanism to prevent suicide by antibiotic-producing bacteria. According to this hypothesis, resistance genes then spread to soil bacteria before, finally, via an intermediate host bacterium, ending up in enterobacteria and other gut bacteria (Benveniste and Davies, 1973; Nonaka et al., 2006; D'Costa et al., 2007; Forsberg et al., 2012). The phylogenetic analysis of Aminov and Mackie (2007) suggested an early diversification between cluster I (soil and gut bacteria) and cluster II (glycopeptide-producing Actinobacteria), with no later exchange of genes between these (Aminov and Mackie, 2007). The vancomycin resistance genes found in Desulfitobacterium spp. and Desulfosporosinus spp. form a distinct group in cluster II, indicating that they have been acquired from vancomycin producers at a later stage than those belonging to cluster I (Fig. 3). Therefore, we propose that the vancomycin resistance gene cluster found in Desulfitobacterium spp. and Desulfosporosinus spp. should be considered as a novel cluster, designated vanI rather than a variant of the vanA type as initially suggested for D. hafniense Y51 (Kalan et al., 2009).

The basic structure of the vanXImurFIvanKIWIIRISI gene cluster was found to be conserved in the published genomes of D. hafniense strain Y51, DCB-2, Desulfosporosinus sp. OT and Desulfosporosinus youngiae (Fig. 1) (Nonaka et al., 2006; Abicht et al., 2011; Kim et al., 2012; Pester et al., 2012). The only exceptions to this conserved structure found so far are that murFI was not found in Desulfosporosinus youngiae, and vanKI is only present in D. hafniense Y51 and Desulfosporosinus sp. OT (Fig. 1). A characteristic feature of the VanI vancomycin resistance gene cluster is the absence of a gene encoding the d-Lac dehydrogenase (VanH). This is the first vancomycin resistance gene cluster, where the vanH gene is located on the chromosome outside the resistance cluster.

Desulfitobacteria have been found both in the environment and in the human gut (Villemur et al., 2006). Hence, it is tempting to speculate that they potentially could act as a shuttle for vancomycin resistance between the environment and the gut microbiota. However, no indications were found in the available Desulfitobacterium spp. genomes that the vancomycin resistance genes are or have been mobile. The location of vanH homologues elsewhere in the genome, rather than in the vancomycin resistance cluster, likely acts as a barrier against horizontal transfer of this phenotype.

The genera Desulfitobacterium and Desulforosporosinus are phylogenetically closely related, both belonging to the family Peptococcaceae (Fig. 2) (Spring and Rosenzweig, 2006). It seems likely that the vanI gene cluster was acquired by a common ancestor of Desulfitobacterium spp. and Desulforosporosinus spp. based on the very high sequence similarity between their vanI sequences, and the characteristic location of vanH homologues outside the resistance gene cluster (Figs 1 and 3). It is interesting to note that Streptomyces coelicolor vanM is the d-Ala-d-Lac ligase encoding homologue showing highest identity to vanI (Supporting Information Table S1). This is also the only other vancomycin resistance gene cluster in which vanK has been found (Hong et al., 2004; 2008). A Streptomyces coelicolor ΔvanK mutant was shown to be unable to grow in the presence of vancomycin (Hong et al., 2005). In contrast, vanK is not essential for vancomycin resistance in Desulfitobacterium spp., as both Desulfitobacterium hafniense DP7 and DCB-2 are resistant to vancomycin, despite not encoding a vanKI. Inspecting the genomes of Desulfitobacterium hafniense DCB-2 and DP7 confirmed that vanK unlike vanH is not located elsewhere in the genomes. Streptomyces spp. do not share a recent common ancestor with Desulfitobacterium spp. or Desulfosporosinus spp.; thus, it is plausible that the vanI gene cluster was acquired horizontally by a common ancestor of Desulfitobacterium spp. and Desulfosporosinus spp. from Streptomyces coelicolor or its ancestor. This would also offer a plausible explanation for the presence of vanKI in Desulfitobacterium hafniense Y51 and Desulfosporosinus sp. OT as a redundant reminiscence that has been lost from most vanI-containing strains due to lack of selection pressure.

It was investigated whether vancomycin could be used as a selective agent for purifying Desulfitobacterium spp. from enrichment cultures containing other, susceptible Gram-positive populations, such as Sedimentibacter spp., which have been repeatedly co-enriched with organohalide-respiring species of Desulfitobacterium and the closely related genus Dehalobacter (Van Doesburg et al., 2005; Bedard et al., 2006; Breitenstein et al., 2007; Sánchez-Andrea et al., 2012). A D. hafniense TCE1 culture, contaminated with a Sedimentibacter sp., was successfully purified using vancomycin as a selection agent. Searching public available meta/genome databases showed that vanI could not be found in any of the screened metagenomes but only in the genomes of the members of both D. hafniense or chlororespirans and the closely related Desulfosporosinus spp., supporting the hypothesis that vanI, unlike other vancomycin resistance clusters found in the gut microbiota, is not subject to frequent horizontal gene transfers. The use of vancomycin or teicoplanin may, thus, help in future isolation of new Desulfitobacterium or Desulfosporosinus spp. strains and may also be used as a convenient selection marker for genetic manipulations.

In conclusion, we have shown that D. hafniense, D. chlororespirans and Desulfosporosinus spp. contain a novel chromosomal gene cluster, designated vanI, conferring medium to high levels of vancomycin resistance, and medium level teicoplanin resistance. This feature can be both a useful intrinsic selection marker or a potential risk if mobilized into clinically relevant strains.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Nucleotide sequence accession numbers
  8. Conflict of interest
  9. References
  10. Supporting Information

Strains and culturing conditions

The bacterial strains used in this study are listed in Table 1. Escherichia coli XL1-blue was grown in lysogeny broth at 37°C. Desulfosporosinus spp. were grown in DSMZ medium DSM 641 supplemented with 50 mM pyruvate for growth of D. meridei, or 12.5 mM pyruvate and 6.25 mM malate for growth of D. youngia or D. orientis.

Desulfitobacterium spp. were grown as 20 ml cultures in 120 ml serum bottles under an atmosphere of 100% N2. Desulfitobacterium dichloroeliminans DCA1 was grown as described previously (De Wildeman et al., 2003). Desulfitobacterium metallireducens was grown in the medium recommended by the DSMZ (DSM 838). All other strains of Desulfitobacterium spp. were cultivated in basal medium as described previously (Neumann et al., 1994), supplemented with 1 g/l yeast extract and with additional modifications from the original medium as follows.

After autoclaving, the medium was supplemented with vitamins and trace elements from anaerobic filter sterilized (0.22 μm pore size) stocks giving final concentrations per litre medium of 1.34 μM EDTA; 10.06 μM FeCl2●4H2O; 0.51 μM MnCl2●4H2O; 0.8 μM CoCl2●6H2O; 0.51 μM ZnCl2; 0.02 μM CuCl2; 0.04 μM AlCl3●6H2O; H3BO3; 0.15 μM Na2MoO4; 0.1 μM NiCl2●6H2O, 0.75 mM CaCl2●2H2O; 0.5 mM MgCl2●6H2O; 3 μg Na2SeO3; 9 μg Na2WO4●2H2O; 0.04 μl concentrated HCl; 50 μg biotin; 250 μg P-aminobenzoate; 50 μg pantothenate; 20 μg folic acid; 50 μg lipoic acid; 100 μg pyridoxine; 550 μg nicotinamide; 100 μg thiamine HCl; 50 μg riboflavin; 50 μg cyanocobalamine; and finally reducing buffer solutions to a final concentration of 1 mM Na2S●9H2O, 5.6 mM NH4HCO3 and 44.4 mM NaHCO3.

Cultures were routinely grown fermentatively on 40 mM pyruvate at the temperatures recommended by the DSMZ, unless stated otherwise. Pyruvate, lactate and PCE were added from sterile anaerobic stocks to final concentrations of 40, 20 and 10 mM, respectively, when indicated.

Antibiotic susceptibility testing

All plating of Desulfitobacterium spp. and Desulfosporosinus spp. was performed in an anaerobic glove box containing an atmosphere of 96% N2 and 4% H2. Strains were streaked with cotton swaps, on plates prepared as described previously (Smidt et al., 1999), with a modification using Difco™ agar noble (Becton Dickinson, Le Pont de Claix, France) for solidification. The medium contained 40 mM pyruvate, 6 mM malate and 10 mM SO4 as carbon and energy source. Plates were prepared with various concentrations of vancomycin or teicoplanin (0, 1, 5, 10, 25, 50, 75, 150 and 250 μg/ml) and incubated anaerobically at 30°C.

DNA isolation

Cells were harvested from 5 ml cultures by spinning at 6700 × g for 10 min. DNA was extracted from cell pellets using the Bio101 FASTDNA isolation kit according to manufacturers' protocols (MP Biomedicals, Solon, OH, USA).

PCR amplification, cloning and sequencing of vanI genes

All PCR reactions were performed using the GoTaq DNA Polymerase Kit (Promega, Leiden, The Netherlands) in a total reaction volume of 30 μl, with 0.5 μl DNA extract or cell lysate as template. PCR products were analysed on a 1% agarose gel using gene ruler 1 kb or 100 bp plus ladders (Fermentas, St. Leon-Rot, Germany).

A 987 bp fragment of the vanI gene was amplified using primers K46F/K47R (Supporting Information Table S2) and the following temperature programme: 94°C for 120 s, followed by 35 cycles of 94°C for 30 s, 56°C for 30 s and 72°C for 50 s, and a final elongation step at 72°C for 6 min. Resulting amplicons were purified prior to sequencing using the Zymo DNA Clean & Concentrator Kit (Zymo Research, San Francisco, CA, USA), sequenced in one direction using primer K46 and initially analysed using nucleotide blast (Altschul et al., 1990). An overlapping fragment of 512 bp was obtained for all amplicons, and used for phylogenetic analysis.

Purity of Desulfitobacterium hafniense TCE1

Nearly full-length 16S rRNA gene fragments were amplified from D. hafniense TCE1 using primers 27F/1492R (Supporting Information Table S2) and the following temperature programme: 94°C for 120 s, followed by 35 cycles of 94°C for 30 s, 52°C for 40 s and 72°C for 90 s, ending with a final elongation step of 72°C for 5 min.

PCR products were purified as described above; cloning of 16S rRNA genes was done using the cloning vector pGEMT-Easy (Promega) and E. coli XL1-Blue as host following the manufacturer's instructions. Full-length inserts were amplified from transformant cells, lysed by heating of 10 μl o/n culture in 90 μl TE (10 mM tris-HCl, 1 mM EDTA, pH 7.5) at 97°C for 10 min. PCR amplification using primers T7 and SP6, targeting the flanking region of the pGEMT-Easy cloning site (Supporting Information Table S2), was done as described above, using the following temperature programme: 94°C for 120 s, followed by 35 cycles of 94°C for 30 s, 52°C for 40 s and 72°C for 90 s, ending with a final elongation step of 72°C for 5 min.

Clones were analysed for the presence of Desulfitobacterium spp. derived 16S rRNA gene sequences using genus-specific primers DSB406F/619R (Supporting Information Table S2), and the following temperature programme: 94°C for 4 min, followed by 35 cycles of 94°C for 30 s, 58°C for 20 s, 72°C for 35 s, and a final elongation step of 72°C for 10 min. Inserts from three randomly selected clones that did not give a product with Desulfitobacterium specific primers were partially sequenced and analysed using blast (Altschul et al., 1990).

DNA sequence analysis

Alignments of DNA sequences were done using muscle (Edgar, 2004), and trees constructed using the neighbour-joining method and tested by 1000 bootstraps using the mega5 software package (Tamura et al., 2011). Sequences obtained in this study were first aligned with those used in the study of Aminov and Mackie (2007) and one sequence representing the recently described cluster vanM (Xu et al., 2010). Finally, sequences obtained from genome databases were added (Markowitz et al., 2012a). Identity matrixes were generated using MatGAT with default settings (Campanella et al., 2003). Full genome alignment and analysis was done using the Artemis comparison tool, version 10 (Carver et al., 2005).

Check for vanI homologues in (meta)genomes

All available (as of 4 February 2013; 1236 metagenomes covering a wide range of environments) metagenomes on the JGI IMG/M server (Markowitz et al., 2012b) were screened for the presence of vanI homologues. We performed an initial BlastP using the strictest possible cut-off, E < e-50, against the entire img/m database using VanI (dsy_3690) from Desulfitobacterium hafniense Y51 as query. We obtained 1338 sequences with homology to VanI, created a local blast database containing the contigs harbouring the corresponding DNA sequences, and performed a tblastx analysis against all contigs, using the D. hafniense Y51 vanI gene sequence as query. This was done in order to correct for genes that potentially have not been incorporated correctly into the BlastP database, due to for example frameshift mutations or assembly errors. The eight full-length d-Ala-d-Lac ligase gene homologues with the best hits to vanI (E ≤ 3.00E-135) were aligned and checked for phylogenetic affiliation with vanI.

vanI homologues in sequenced bacterial genomes outside the Desulfitobacterium genus

All available genomes (as of 19 May 2014) at the JGI/IMG server (Markowitz et al., 2012a) were screened for the presence of vanI homologues. Initially, as for metagenome mining, BlastP search was done using D. hafniense Y51 VanI (dsy3690) sequence as query against all available genomes. This was followed by manual check of sequences giving strong (E = 0) hits to VanI. The vanI sequence identified in Desulfosporosinus sp. OT was found to be N-terminally truncated. A closer inspection of the genome sequence showed that this is due to either a frameshift mutation or sequencing error; by combining two reading frames, we were able to assemble a 1032 bp fragment. The corresponding predicted amino acid sequence showed 93.3% identity to the vanI homologue from Desulfosporosinus youngiae JW/Y JL-B18 (Supporting Information Table S1).

Nucleotide sequence accession numbers

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Nucleotide sequence accession numbers
  8. Conflict of interest
  9. References
  10. Supporting Information

The obtained partial vanI D-Ala-D-Lac ligase gene sequences are available in The National Center for Biotechnology Information (NCBI) database under accession numbers HQ433580 to HQ433586.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Nucleotide sequence accession numbers
  8. Conflict of interest
  9. References
  10. Supporting Information
  • Abicht, H.K., Mancini, S., Karnachuk, O.V., and Solioz, M. (2011) Genome sequence of Desulfosporosinus sp. OT, an acidophilic sulfate-reducing bacterium from copper mining waste in Norilsk, Northern Siberia. J Bacteriol 193: 61046105.
  • Altschul, S., Gish, W., Miller, W., Myers, E., and Lipman, D. (1990) Basic alignment search tools. J Mol Biol 215: 403410.
  • Aminov, R.I., and Mackie, R.I. (2007) Evolution and ecology of antibiotic resistance genes. FEMS Microbiol Lett 271: 147161.
  • Bedard, D.L., Bailey, J.J., Reiss, B.L., and Jerzak, G.V. (2006) Development and characterization of stable sediment-free anaerobic bacterial enrichment cultures that dechlorinate aroclor 1260. Appl Environ Microbiol 72: 24602470.
  • Benveniste, R., and Davies, J. (1973) Aminoglycoside antibiotic inactivating enzymes in actinomycetes similar to those present in clinical isolates of antibiotic resistant bacteria. P. Natl. Acad. Sci. USA 70: 22762280.
  • 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.
  • Breitenstein, A., Andreesen, J.R., and Lechner, U. (2007) Analysis of an anaerobic chemostat population stably dechlorinating 2,4,6-trichlorophenol. Eng Life Sci 7: 380387.
  • Breitenstein, A., Wiegel, J., Haertig, C., Weiss, N., Andreesen, J.R., and Lechner, U. (2002) Reclassification of Clostridium hydroxybenzoicum as Sedimentibacter hydroxybenzoicus gen. nov., comb. nov., and description of Sedimentibacter saalensis sp. nov. Int J Syst Evol Microbiol 52: 801807.
  • Campanella, J.J., Bitincka, L., and Smalley, J. (2003) MatGAT: an application that generates similarity/identity matrices using protein or DNA sequences. BMC Bioinformatics 4: 29.
  • Carver, T.J., Rutherford, K.M., Berriman, M., Rajandream, M.A., Barrell, B.G., and Parkhill, J. (2005) ACT: the Artemis comparison tool. Bioinformatics 21: 34223423.
  • Courvalin, P. (2006) Vancomycin resistance in gram-positive cocci. Clin Infect Dis 42: S25S34.
  • Davies, J., and Davies, D. (2010) Origins and evolution of antibiotic resistance. Microbiol Mol Biol Rev 74: 417433.
  • D'Costa, V.M., Griffiths, E., and Wright, G.D. (2007) Expanding the soil antibiotic resistome: exploring environmental diversity. Curr Opin Microbiol 10: 481489.
  • De Wildeman, S., Diekert, G., Van Langenhove, H., and Verstraete, W. (2003) Stereoselective microbial dehalorespiration with vicinal dichlorinated alkanes. Appl Environ Microbiol 69: 56435647.
  • Depardieu, F., Podglajen, I., Leclercq, R., Collatz, E., and Courvalin, P. (2007) Modes and modulations of antibiotic resistance gene expression. Clin Microbiol Rev 20: 79114.
  • Edgar, R.C. (2004) MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res 32: 17921797.
  • Forsberg, K.J., Reyes, A., Wang, B., Selleck, E.M., Sommer, M.O.A., and Dantas, G. (2012) The shared antibiotic resistome of soil bacteria and human pathogens. Science 337: 11071111.
  • Fraimow, H., Knob, C., Herrero, I.A., and Patel, R. (2005) Putative VanRS-like two-component regulatory system associated with the inducible glycopeptide resistance cluster of Paenibacillus popilliae. Antimicrob Agents Chemother 49: 26252633.
  • Hong, H.J., Hutchings, M.I., and Buttner, M.J. (2008) Vancomycin resistance VanS/VanR two-component systems. Adv Exp Med Biol 631: 200213.
  • Hong, H.J., Hutchings, M.I., Hill, L.M., and Buttner, M.J. (2005) The role of the novel fem protein VanK in vancomycin resistance in Streptomyces coelicolor. J Biol Chem 280: 1305513061.
  • Hong, H.J., Hutchings, M.I., Neu, J.M., Wright, G.D., Paget, M.S.B., and Buttner, M.J. (2004) Characterization of an inducible vancomycin resistance system in Streptomyces coelicolor reveals a novel gene (vanK) required for drug resistance. Mol Microbiol 52: 11071121.
  • Kalan, L., Ebert, S., Kelly, T., and Wright, G.D. (2009) Noncanonical vancomycin resistance cluster from Desulfitobacterium hafniense Y51. Antimicrob Agents Chemother 53: 28412845.
  • Kim, S.-H., Harzman, C., Davis, J., Hutcheson, R., Broderick, J., Marsh, T., and Tiedje, J. (2012) Genome sequence of Desulfitobacterium hafniense DCB-2, a gram-positive anaerobe capable of dehalogenation and metal reduction. BMC Microbiol 12: 21.
  • 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.
  • Löffler, F.E., Sanford, R.A., and Ritalahti, K.M. (2005) Enrichment, cultivation, and detection of reductively dechlorinating bacteria. Methods Enzymol 397: 77111.
  • Markowitz, V.M., Chen, I.M.A., Palaniappan, K., Chu, K., Szeto, E., Grechkin, Y., et al. (2012a) IMG: the integrated microbial genomes database and comparative analysis system. Nucleic Acids Res 40: D115D122.
  • Markowitz, V.M., Chen, I.M.A., Chu, K., Szeto, E., Palaniappan, K., Grechkin, Y., et al. (2012b) IMG/M: the integrated metagenome data management and comparative analysis system. Nucleic Acids Res 40: D123D129.
  • Marshall, C.G., Lessard, I.A.D., Park, I.S., and Wright, G.D. (1998) Glycopeptide antibiotic resistance genes in glycopeptide-producing organisms. Antimicrob Agents Chemother 42: 22152220.
  • Neumann, A., Scholz-Muramatsu, H., and Diekert, G. (1994) Tetrachloroethene metabolism of Dehalospirillum multivorans. Arch Microbiol 162: 295301.
  • Nonaka, H., Keresztes, G., Shinoda, Y., Ikenaga, Y., Abe, M., Naito, K., et al. (2006) Complete genome sequence of the dehalorespiring bacterium Desulfitobacterium hafniense Y51 and comparison with Dehalococcoides ethenogenes 195. J Bacteriol 188: 22622274.
  • van de Pas, B.A., Harmsen, H.J.M., Raangs, G.C., de Vos, W.M., Schraa, G., and Stams, A.J.M. (2001) A Desulfitobacterium strain isolated from human feces that does not dechlorinate chloroethenes or chlorophenols. Arch Microbiol 175: 389394.
  • Pester, M., Brambilla, E., Alazard, D., Rattei, T., Weinmaier, T., Han, J., et al. (2012) Complete genome sequences of Desulfosporosinus orientis DSM765T, Desulfosporosinus youngiae DSM17734T, Desulfosporosinus meridiei DSM13257T, and Desulfosporosinus acidiphilus DSM22704T. J Bacteriol 194: 63006301.
  • Sánchez-Andrea, I., Rojas-Ojeda, P., Amils, R., and Sanz, J.L. (2012) Screening of anaerobic activities in sediments of an acidic environment: Tinto river. Extremophiles 16: 111.
  • Schweizer, H.P. (2008) Bacterial genetics: past achievements, present state of the field, and future challenges. Biotechniques 44: 633641.
  • Smidt, H., Song, D.L., van der Oost, J., and de Vos, W.M. (1999) Random transposition by Tn916 in Desulfitobacterium dehalogenans allows for isolation and characterization of halorespiration-deficient mutants. J Bacteriol 181: 68826888.
  • Spring, S., and Rosenzweig, F. (2006) The genera Desulfitobacterium and Desulfosporosinus. In The Prokaryotes. Rosenberg, E., Stackebrandt, E., Thompson, F., Lory, S., and De Long, E. (eds). Berlin-Heidelberg, Germany: Springer-Verlag, pp. 771786.
  • Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M., and Kumar, S. (2011) MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol Biol Evol 28: 27312739.
  • Teo, J.W.P., Krishnan, P., Jureen, R., and Lin, R.T.P. (2011) Detection of an unusual van genotype in a vancomycin-resistant Enterococcus faecium hospital isolate. J Clin Microbiol 49: 42974298.
  • Van Doesburg, W., Van Eekert, M.H.A., Middeldorp, P.J.M., Balk, M., Schraa, G., and Stams, A.J.M. (2005) Reductive dechlorination of β-hexachlorocyclohexane (β-HCH) by a Dehalobacter species in coculture with a Sedimentibacter sp. FEMS Microbiol Ecol 54: 8795.
  • Villemur, R., Lanthier, M., Beaudet, R., and Lepine, F. (2006) The Desulfitobacterium genus. FEMS Microbiol Rev 30: 706733.
  • 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. Nucleotide sequence accession numbers
  8. Conflict of interest
  9. References
  10. Supporting Information
FilenameFormatSizeDescription
mbt212139-sup-0001-si.xls44K

Table S1. Identity matrix showing percentage amino acid identity of d-Ala-d-Lac ligases representing the currently known van gene clusters of the d-Ala-d-Lac type. The sequences used here are translated, using the standard bacterial genetic code, from the full-length genes used in Fig. 3. Accession numbers are given in parentheses. The identity matrix were generated using MatGAT as described under experimental procedures. Colour codes highlight the different vancomycin resistance clusters. Shades of brown indicates representatives of cluster II, whereas blue and purple indicates representatives of cluster I.

mbt212139-sup-0002-si.doc37K

Table S2. Primers used in this study.

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.