aac accepts, published online ahead of print on 4 may...

1

A non-canonical vancomycin resistance cluster from Desulfitobacterium hafniense Y51 1

Lindsay Kalan1, Sara Ebert

2, Tom Kelly

3and Gerard D. Wright

1* 2

3

Michael G. DeGroote Institute for Infectious Disease Research, Department of Biochemistry and 4

Biomedical Sciences, McMaster University, Hamilton, Ontario, Canada1, Department of 5

Biological Sciences, University of Alberta, Edmonton , Alberta, Canada2, and Department of 6

Microbiology, St. Josephs’ Health Care, Hamilton, Ontario, Canada3. 7

8

*Corresponding author: Michael G. DeGroote Institute for Infectious Disease Research, 9

McMaster University, 1200 Main St W., Hamilton, Canada L8N 3Z5. Email: 10

[email protected]. 11

12

Running Title: New vancomycin resistance cluster in Desulfitobacterium 13

Copyright © 2009, American Society for Microbiology and/or the Listed Authors/Institutions. All Rights Reserved.Antimicrob. Agents Chemother. doi:10.1128/AAC.01408-08 AAC Accepts, published online ahead of print on 4 May 2009

on August 27, 2018 by guest

http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

2

ABSTRACT 14

The glycopeptide vancomycin is a drug of last resort for infection with Gram-positive organisms 15

and three genes are vital to resistance; vanH, vanA, and vanX. These genes are found in a 16

vanHAX cluster, which is conserved across pathogenic bacteria, glycopeptide antibiotic 17

producers, and other environmental bacteria. The genome sequence of anaerobic, Gram-18

positive, dehalogenating bacterium Desulfitobacterium hafniense Y51 revealed a predicted vanA 19

homolog however, it exists in a vanAWKmurFX cluster, unlike other vancomycin resistant 20

organisms. Using purified recombinant VanA from D. hafniense Y51 we determined its 21

substrate specificity and found it to have a 42-fold preference for D-Lactate over D-Alanine 22

confirming its activity as a D-Ala:D-Lac ligase and annotation as a VanA. Furthermore, we have 23

shown that D. hafniense Y51 is highly resistant to vancomycin where the minimum inhibitory 24

concentration for growth is 64 µg/mL. Finally, vanADh is expressed during growth in 25

vancomycin as demonstrated by RT-PCR. This finding represents a new glycopeptide antibiotic 26

resistance gene cluster and expands the genetic diversity of resistance to this important class of 27

antibiotic. 28

29

INTRODUCTION 30

The glycopeptide antibiotics such as vancomycin and teicoplanin continue to be frontline 31

therapies for the treatment of serious infections caused by Gram positive pathogens. These 32

antibiotics act on the outside of the cell by binding the terminal D-alanyl-D-alanine of nascent 33

peptidoglycan and precursors such as lipid II by a series of five hydrogen bonds (4, 17). 34

Clinical glycopeptide antibiotic resistance generally involves biosynthesis of 35

peptidoglycan terminating in D-Ala-D-X, where X is either Ser, whose side chain 36


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

3

hydroxymethyl group sterically interferes with the antibiotic-dipeptide interaction (20), or the α-37

hydroxy acid lactate (Lac) (4). High-level vancomycin resistance in vancomycin resistant 38

enterococci (VRE) and in glycopeptide producing bacteria is associated with the latter 39

substitution of D-Lac, resulting in a terminal depsipeptide (ester linkage) rather than a peptide 40

(amide linkage). Unlike D-Ala:D-Ser, D-Ala:D-Lac is isosteric with D-Ala:D-Ala, but the 41

substitution of the amide for an ester removes a critical hydrogen bond donor required for 42

optimal antibiotic interaction with peptidoglycan. This single change results in a 1000-fold 43

decrease in affinity of antibiotic for its target and clinical drug resistance (4). 44

Three genes are essential for D-Ala:D-Lac-mediated vancomycin resistance in 45

enterococci (VRE): vanH, vanA and vanX. vanH encodes a pyruvate dehydrogenase that 46

converts pyruvate to D-Lac. vanA encodes an ATP-dependent depsipeptide ligase that catalyzes 47

the synthesis of D-Ala:D-Lac. vanX encodes a dipeptidase that cleaves existing D-Ala:D-Ala in 48

the cell ensuring a cell wall enriched in D-Ala:D-Lac (1, 13). The phenotype is named after the 49

ligase VanA conferring high level resistance to both vancomycin and teicoplanin while 50

organisms displaying the VanB, D or F phenotype harbor D-Ala:D-Lac in their cell walls but 51

have varying levels of induction to different glycopeptide antibiotics. The vanHAX cluster (Fig. 52

1) is conserved in pathogens (VRE), in glycopeptide producing actinomycetes (17, 19), and can 53

also be found embedded in the genomes of environmental bacteria that do not produce 54

glycopeptides such as Streptomyces coelicolor (12), Paenibacillus sp. (8-10), and Frankia sp. 55

(Kalan and Wright unpublished). Furthermore, in a survey of ~500 soil bacteria 1% were shown 56

to be vancomycin resistant, and all resistant strains save 1 contained a vanHAX cluster as 57

determined by PCR (6). 58


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

4

Regardless of origin (VRE, environmental bacterium or glycopeptide producer) these 59

genes in high level vancomycin resitant organisms are always found in a vanHAX cluster (Fig. 60

1). Despite the ubiquity of this cluster, there are a number of different glycopeptide antibiotic 61

resistance phenotypes in VRE and other bacteria (15). Given the conservation of the vanHAX 62

cluster and of the associated biochemical mechanism, differences in glycopeptide resistance 63

phenotype are attributed to variable regulation of expression of these genes by a two-component 64

regulatory system comprised of VanS, a His kinase, and VanR, a response regulator (2) (Fig. 1). 65

A scan of recently sequenced genomes identified a vanA homolog in the genome of 66

Desulfitobacterium hafniense Y51, a strictly anaerobic, Gram positive rod with low G+C content 67

isolated from a site in Japan contaminated with halogenated organic compounds (18, 21). 68

Surprisingly, this homolog was not found in a vanHAX cluster but with a nearby vanX and a 69

vanW homolog, a protein of unknown function associated in VRE with the VanB phenotype (Fig 70

1). A similar arrangement was also found in the unpublished genome of D. hafniense DCB-2 71

(genome.jgi-psf.org/draft_microbes/desha/desha.home.html). 72

Here we report the biochemical characterization of this putative VanA as an active D-73

Ala:D-Lac ligase, part of a non-canonical vancomycin resistance gene cluster in the genome of 74

D. hafniense Y51. 75

76

MATERIALS AND METHODS 77

Cloning, expression and purification of D. hafniense Y51 VanA and Ddl homologs: 78

Desulfitobacterium hafniense Y51 strain and genomic DNA were generously provided by Dr. 79

Masatoshi Goto and Dr. Kensuke Furukawa (Kyushu University, Fukuoka, Japan). Cultures 80

were maintained in m-ISA medium consisting of 1% tryptone peptone, 0.35% sodium lactate, 81


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

5

0.05% Na2SO3, 0.2% MgSO4·7H2O, 0.05% iron (III) ammonium, and 0.001% resazurin, pH 7.2 82

and incubated at 30ºC. 83

Cloning of the putative D-Ala:D-Lac and D-Ala:D-Ala ligase gene from D. hafniense 84

Y51 was achieved by PCR amplification of the genes DSY3690 and DSY1579 respectively from 85

genomic DNA. Primers were designed to amplify 1.1 kB fragments from D. hafniese Y51 86

genomic DNA. Primers had attB sites engineered within to facilitate cloning using the 87

Gateway® technology (Invitrogen, Carlsbad, CA). The specific primers used for DSY3690 are : 88

5’- GGGGACAAGTTTGTACAAAAAAGCAGGCTTAATGGATCGGTTGAAAATCGCA-3’, 89

5’-GGGGACCACTTTGTACAAGAAAGCTGGGTCCCCATCCTATCGGCA-3’ 90

For DSY1579 the specific primer sequences are:: 5’-91

GGGGACAAGTTTGTACAAAAAAGCAGGCTYYCATATGATGACACGGCAAAAGATTA92

TTATTC-3’, 5’-93

GGGACCACTTTGTACAAGAAAGCTGGGTYAAGCTTCTAACGGGATATTTTCCGGG-3’ 94

The resulting PCR products were inserted into the pDEST17 (Invitrogen, Carlsbad, CA) 95

destination vector containing a 6xHis tag for ease of downstream purification. Gene integrity 96

was confirmed by DNA sequencing. Plasmids were propagated in Escherichia coli TOP10’ cells 97

and subsequently used to transform E. coli BL21 (DE3) Rosetta (Novagen, Darmstadt, Germany) 98

cells for high level protein expression under control of the T7 promoter. For VanAAo, VanADh, 99

and DdlDh overexpression, cells were propagated in 1 litre of Luria-Bertani broth until the optical 100

density at 600 nm reached 0.6. Protein expression was induced by addition of isopropyl β-D-1-101

thiogalactopyranoside to a final concentration of 1 mM and incubation of cultures at 16°C for 18 102

hours. Cells were harvested and washed in 0.85 % (w/v) NaCl before resuspension in 10 mL 103


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

6

lysis buffer (50 mM NaH2PO4; 300 mM NaCl; 10 mM imidazole; 1 mM phenylmethanesulfonyl 104

fluoride; 1 mM DNase; pH 8.0). Cells were lysed by four passes through a French pressure cell 105

at 1250 psi and cell debris removed by centrifugation at 27 000 x g for 30 min. The supernatant 106

was collected and purified enzyme was obtained using Nickel-NTA immobilized metal affinity 107

chromatography (Qiagen, Valencia, CA). Fractions containing the purified enzyme were pooled 108

and dialysed into 50 mM HEPES; 5 mM MgCl2; 1 mM EDTA; pH 8.0 at 4°C. 109

E. coli BL21(DE3) harboring pET28bvanAAo from the vancomycin producer 110

Amycolatopsis orientalis C329.2, was prepared as previously described (17) and purified as a 111

6xHis tagged protein as described above. E. coli W3110 harboring pTB2 for expression of the 112

D-Ala:D-Ala ligase DdlB was previously reported (24) . 113

Ddl assays: For qualitative determination of VanADh and DdlDh substrate specificity; initial 114

enzymatic characterization was carried out using the pyruvate kinase/lactate dehydrogenase 115

coupled assay to monitor ADP formation (22). Amino and hydroxy acid substrate specificity 116

was determined by thin-layer-chromatography and using radiolabelled substrates. Due to cost of 117

[U-14

C]-D-Alanine, [U-14

C]-L-Alanine was isomerized to a racemic mixture of [14

C]-L/D-118

Alanine with one unit of Bacillus stearothermophilus alanine racemase (SigmaAldrich). Ligase 119

reactions contained 50 mM HEPES pH 7.5; 10 mM MgCl2; 40 mM KCl; 6 mM ATP; 2 µM 120

enzyme; 0.1 µCi [U-14

C]-L/D-Ala; 1 mM D-Ala and 10 mM D-X substrate. Reactions were 121

quenched with 50% methanol and applied onto a PEI- cellulose TLC plate (SigmaAldrich). The 122

plates were developed in 12:3:5 butanol:acetic acid: water, dried overnight and exposed to a 123

phosphor-storage imaging screen. The screens were imaged using a Typhoon™ variable mode 124

imager and relative radioactive intensity was quantified using ImageQuant 5.2 software. 125


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

7

Michaelis-Menten kinetics were determined by using the software program GraFit 126

version 4.0.21 (Erithacus software) and initial rates were determined using the non-linear least 127

squares method and equation 1 (14). 128

v = (kcat/Et)[S]/(KM + [S]) [1] 129

RNA preparation: Cultures of D. hafniense Y51 were grown in m-ISA media (ref) for 24 hours 130

and diluted 1/100 into m-ISA containing 0 µg/mL or 125 µg/mL vancomycin and incubated at 131

30ºC for 24 or 48 hours. Cultures were harvested by centrifugation for 15min at 3000 x g and 132

the pellets resuspended in 1mL of RNAprotect bacterial reagent (Qiagen, Valenica, CA). After 133

subsequent centrifugation, total RNA was extracted using the RNeasy Mini Kit (Qiagen, 134

Valencia, CA) following the enzymatic lysis and proteinase K digestion of bacteria (protocol 4) 135

before purification of total RNA (protocol 7). An on column DNAse I digestion was performed 136

in addition to a post elution DNAse I digestion and additional column purification. RNA was 137

quantified using a NanoDrop ND-1000 spectrophotometer and checked for genomic DNA 138

contamination by PCR analysis. 139

Expression analysis: Reverse transcription was carried out with 0.3 µg of DNAse I treated RNA 140

and 200 ng of random hexamers to generate cDNA using the Superscript III RT kit (Invitrogen, 141

Carlsbad, CA). The samples were incubated at 25ºC for 5 minutes prior to 50ºC for 45 minutes 142

followed by inactivation at 70ºC for 15 minutes. 1µL of the RT reaction was used for real-time 143

PCR with SYBR green in a SmartCycler system (Cepheid, Sunnyvale, CA). The following 144

primers were designed to amplify a 293 bp vanADh product and a 270 bp ddlDh product in a 25µL 145

reaction volume. vanA Dh primers are 5’-TTCTTCTTGGCGGCATACTT-3’ and 5’-146

ATCTGGTGTTTCCCGTTCTG-3’ while ddlDh primers are 5’-147

GTGAAGAACGGGGAAAATCA-3’ and 5’-CAATCCGGAGAACATGAGGT-3’. The results 148


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

8

were normalized by the 2-∆∆C

T method (16) to 16S rRNA expression using the primers 5’-149

AGGCCTTCGGGTTGTAAAGT- 3’ and 5’-ATACCCAGTTTCCGATGCAG-3’ to amplify a 150

237 bp product. Products were analyzed on a 1% agarose gel to confirm a single product was 151

amplified. 152

Antibiotic susceptibility testing : D. hafniense Y51 susceptibility tests for vancomycin and 153

teicoplanin using disc-diffusion assays were performed according to CLSI methods (5).. In 154

addition, the minimum inhibitory concentration (MIC) for vancomycin was determined using 155

Etest gradient strips (Dalvagen, Solna, Sweden). The organism was grown on purchased 156

Brucella agar with 5% sheep blood, vitamin K and hemin supplements (PML Microbiologicals, 157

Wilsonville, OR,) and the manufacturer’s protocol was followed for MIC determination. 158

159

RESULTS & DISCUSSION 160

Expression and characterization of VanA and Ddl homologues from D. hafniense 161

Y51. BLAST search of the D. hafniense Y51 genome revealed two predicted D-Ala-D-X ligase 162

genes. The first, DSY1579, is a predicted D-Ala:D-Ala ligase (DdlDh) while the second, 163

DSY3690, is homologous to VanA-like D-Ala:D-Lac ligases (VanADh) (Table 1). The ddlDh 164

gene is located proximal to other genes predicted to encode peptidoglycan assembly proteins 165

(e.g. DD-carboxypeptidase). Conversly, vanADh is clustered with murF, vanX, vanK, and vanW 166

homologues and a predicted vanSR two component regulatory system (Fig 1). We hypothesized 167

that this novel vanAWKmurFX cluster may encode an alternative glycopeptide resistance cluster 168

rather than the canonical vanHAX found in VRE, glycopeptide producers, and other 169

environmental bacteria. As noted above, VanX is a D-Ala:D-Ala hydrolase required to eliminate 170

constitutively produced D-Ala:D-Ala. VanK (11, 12) is a FemX protein that catalyzes the cross 171


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

9

linking reaction of peptidoglycan terminating in D-Ala:D-Lac (where native FemX enzymes will 172

not), while MurF adds the D-Ala:D-Lac depsipeptide to the growing chain (7). VanW on the 173

other hand is a protein of unknown function associated with VanB phenotype resistance clusters 174

(Fig. 1). Bateman et al. hypothesized that VanW may be important in localizing other 175

vancomycin resistant proteins to unlinked peptidoglycan based on the presence of a G5 domain 176

in the C-terminus, which may bind to N-acetylglucosamine residues (3). 177

In order to experimentally verify the predicted biochemical activities of the D. hafniense 178

Y51 predicted D-Ala-D-X ligase genes, each was overexpressed as His6-tag fusions in E. coli, 179

and purified by immobilized metal affinity chromatography by standard methods yielding pure 180

proteins (Fig. 2). 181

Substrate specificity of recombinant D-Ala-D-X ligases: The amino acid specificity of 182

VanADh and DdlDh was qualitatively determined using [U-14

C]-D-Ala and unlabled D-amino and 183

D-hydroxy acid substrates followed by separation of the products by thin-layer chromatography 184

on a PEI-cellulose plate. DdlB, the D-Ala:D-Ala ligase from E. coli and VanAAo, the D-Ala:D-185

Lac ligase from the vancomycin producer A. orientalis C329.2, were used as positive controls. 186

DdlB is only able to produce D-Ala:D-Ala, while VanAAo catalyzes D-Ala:D-Lac synthesis. 187

Although the amino acid sequence identity is only 34%, DdlDh exhibits activity similar to that of 188

E. coli DdlB, producing only D-Ala:D-Ala even in the presence of excess (10-fold) D-Lac. In 189

contrast, the conserved amino acid identy between VanADh and VanAAo is 67% and VanADh 190

shows substrate specificity similar to VanAAo, producing only D-Ala:D-Lac (Table 1, Fig.3). 191

The functions of the two D-D ligases in D. hafniense Y51 clearly show different substrate 192

specificities indicating that VanADh is able to produce D-Ala:D-Lac, an observation consistent 193

with our hypothesis that vanAWmurFKX may encode an alternate glycopeptide resistance cluster. 194


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

10

Steady state kinetics of D-Ala:D-Ala/Lac Ligases: We performed steady state kinetic analyses 195

to quantify the enzyme activity of VanADh. Purified ligases were kinetically characterized for 196

utilization of D-Ala, D-Lac, and ATP by monitoring the change in absorbance at 340 nm with 197

the coupled pyruvate kinase-lactate dehydrogenase continuous ADP release assay. The results 198

indicate (Table 2) that D-Lac is the preferred substrate for VanADh where the KM is 0.53 mM 199

compared to 22 mM for D-Ala. Furthermore, the catalytic efficiency (kcat/KM) is 42-fold higher 200

for D-Lac under the same conditions. DdlDh was unable to use D-Lac as a substrate (10 mM) and 201

the kcat/KM for D-Ala is comparable to the kcat/KM for D-Lac utilization by VanADh (3.1x103 and 202

1.1x103 M

-1s

-1 respectively). 203

The enzymes bind two molecules of D-amino/hydroxy acids in two distinct binding sites 204

(24). The first is always D-Ala thus we are measuring the kinetic parameters for the second 205

binding site using the coupled assay. Although the active site discriminates which hydroxy 206

amino acid it prefers, in the case of VanADh some D-Ala:D-Ala is concurrently being formed in 207

addition to D-Ala:D-Lac. Unfortunately, it is difficult to resolve the rate and the effect on D-208

Ala:D-Lac formation using the spectrophotometric coupled assay. The markedly large increase 209

in kcat/KM is a good indicator of the discrimination for D-Lac in the active site of the enzyme; 210

however, this assay is unable to quantify directly the formation of D-Ala:D-Lac. Therefore, a 211

direct TLC assay involving radiolabelled substrates was employed. Using this approach, the KM 212

for D-Lactate is 0.8 mM while the kcat is 7.6 x 10-2

s -1

and the kcat/KM is 1.7 x 102

M-1

s-1

(Table 2). 213

D. hafniense Y51 is vancomycin resistant: D. hafniense Y51 was found to be resistant to 214

vancomycin as indicated by no inhibition of growth around a 5 µg vancomycin paper disc while 215

only intermediately resistant to teicoplanin indicated by a 40 mm zone of inhibition around a 30 216


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

11

µg disc. The minimum inhibitory concentration for vancomycin was determined using Etest 217

gradient strips and found to be 64 µg/mL. 218

vanADh expression levels during growth in vancomycin: The induction of vancomycin 219

resistance by vanADh was examined by real-time PCR. Relative RNA levels were normalized to 220

16S rRNA expression and it was determined that vanADh expression is increased 181 and 256 221

fold after 24 and 48 hours incubation in vancomycin respectively. Furthermore, ddlDh RNA 222

levels remained constant regardless of growth in the presence or absence of vancomycin. The 223

fold change of ddlDh upon addition of vancomycin was only 1.19 and 1.23 after 24 and 48 hours 224

growth respectively (Fig. 4). 225

Summary and conclusions. We have identified a glycopeptide antibiotic resistance cluster with 226

novel gene organization in the dehalorespiring organism D. hafniense Y51. This cluster includes 227

predicted vanA and vanX genes as well as additional auxiliary genes predicted to be required for 228

inducible glycopeptide resistance. Biochemical analysis of recombinant VanADh confirms that it 229

is a functional D-Ala:D-Lac ligase. Although, a D-lactate dehydrogenase (vanH) was not part of 230

this new cluster, four annotated D-isomer specific 2-hydroxyacid dehydrogenase genes 231

(DSY0996, DSY1673, DSY3442, and DSY4020) are present in the genome of D. hafniense Y51. 232

The organism exhibits high level resistance to vancomycin even though these dehydrogenases 233

have less than 30% identity to VanH, and therefore at least one of these gene products can 234

provide the requisite D-Lac substrate for VanADh. 235

This new vancomycin resistance cluster expands our understanding of the genetic 236

diversity that encompasses the vancomycin resistome, which includes all the genes in pathogenic 237

and environmental bacteria that can give rise to glycopeptide resistance (23). The presence of 238

this new vancomycin resistance cluster gives rise to a number of unanswered questions 239


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

12

including: why do D. hafniense harbor these genes (is it antibiotic resistance or some 240

physiological advantage for D-Ala:D-Lac terminating cell walls?); is this gene cluster more wide 241

spread in other organisms?; and can this new cluster be mobilized into pathogenic bacteria like 242

vanHAX? While the answers to these questions require further study, what is clear is that 243

continued microbial genome sequencing is revealing the remarkable depth of the antibiotic 244

resistome within microbial populations across the globe. 245

246

ACKNOWLEDGEMENTS 247

We are grateful for the assistance of Dr Julia Foght, University of Alberta. This work was 248

supported by Grant MT-13536 and a graduate student award from the Canadian Institutes of 249

Health Research and a Canada Research Chair to GDW. 250


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

13

REFENCES 251

1. Arthur, M., and P. Courvalin. 1993. Genetics and mechanisms of glycopeptide 252

resistance in enterococci. Antimicrob Agents Chemother 37:1563-71. 253

2. Arthur, M., C. Molinas, and P. Courvalin. 1992. The VanS-VanR two-component 254

regulatory system controls synthesis of depsipeptide peptidoglycan precursors in 255

Enterococcus faecium BM4147. J Bacteriol 174:2582-91. 256

3. Bateman, A., M. T. Holden, and C. Yeats. 2005. The G5 domain: a potential N-257

acetylglucosamine recognition domain involved in biofilm formation. Bioinformatics 258

21:1301-3. 259

4. Bugg, T. D., G. D. Wright, S. Dutka-Malen, M. Arthur, P. Courvalin, and C. T. 260

Walsh. 1991. Molecular basis for vancomycin resistance in Enterococcus faecium 261

BM4147: biosynthesis of a depsipeptide peptidoglycan precursor by vancomycin 262

resistance proteins VanH and VanA. Biochemistry 30:10408-15. 263

5. CLSI. 2007. Methods for Antimicrobial Susceptibility Testing of Anaerobic Bacteria, 7 264

ed. Approved standard M11-A7 ed, vol. 27, Wayne, PA. 265

6. D'Costa, V. M., K. M. McGrann, D. W. Hughes, and G. D. Wright. 2006. Sampling 266

the antibiotic resistome. Science 311:374-7. 267

7. El Zoeiby, A., F. Sanschagrin, and R. C. Levesque. 2003. Structure and function of the 268

Mur enzymes: development of novel inhibitors. Mol Microbiol 47:1-12. 269

8. Guardabassi, L., and Y. Agerso. 2006. Genes homologous to glycopeptide resistance 270

vanA are widespread in soil microbial communities. FEMS Microbiol Lett 259:221-5. 271

9. Guardabassi, L., H. Christensen, H. Hasman, and A. Dalsgaard. 2004. Members of 272

the genera Paenibacillus and Rhodococcus harbor genes homologous to enterococcal 273

glycopeptide resistance genes vanA and vanB. Antimicrob Agents Chemother 48:4915-8. 274

10. Guardabassi, L., B. Perichon, J. van Heijenoort, D. Blanot, and P. Courvalin. 2005. 275

Glycopeptide resistance vanA operons in Paenibacillus strains isolated from soil. 276

Antimicrob Agents Chemother 49:4227-33. 277

11. Hong, H. J., M. I. Hutchings, L. M. Hill, and M. J. Buttner. 2005. The role of the 278

novel Fem protein VanK in vancomycin resistance in Streptomyces coelicolor. J Biol 279

Chem 280:13055-61. 280

12. Hong, H. J., M. I. Hutchings, J. M. Neu, G. D. Wright, M. S. Paget, and M. J. 281

Buttner. 2004. Characterization of an inducible vancomycin resistance system in 282

Streptomyces coelicolor reveals a novel gene (vanK) required for drug resistance. Mol 283

Microbiol 52:1107-21. 284

13. Kahne, D., C. Leimkuhler, W. Lu, and C. Walsh. 2005. Glycopeptide and 285

lipoglycopeptide antibiotics. Chem Rev 105:425-48. 286

14. Leatherbarrow, R. J. 1992. GraFit, 4.0.21 ed. Erithacus Software Ltd. 287

15. Levine, D. P. 2006. Vancomycin: a history. Clin Infect Dis 42 Suppl 1:S5-12. 288

16. Livak, K. J., and T. D. Schmittgen. 2001. Analysis of relative gene expression data 289

using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25:402-290

8. 291

17. Marshall, C. G., I. A. Lessard, I. Park, and G. D. Wright. 1998. Glycopeptide 292

antibiotic resistance genes in glycopeptide-producing organisms. Antimicrob Agents 293

Chemother 42:2215-20. 294


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

14

18. Nonaka, H., G. Keresztes, Y. Shinoda, Y. Ikenaga, M. Abe, K. Naito, K. Inatomi, K. 295

Furukawa, M. Inui, and H. Yukawa. 2006. Complete genome sequence of the 296

dehalorespiring bacterium Desulfitobacterium hafniense Y51 and comparison with 297

Dehalococcoides ethenogenes 195. J Bacteriol 188:2262-74. 298

19. Pootoolal, J., M. G. Thomas, C. G. Marshall, J. M. Neu, B. K. Hubbard, C. T. 299

Walsh, and G. D. Wright. 2002. Assembling the glycopeptide antibiotic scaffold: The 300

biosynthesis of A47934 from Streptomyces toyocaensis NRRL15009. Proc Natl Acad Sci 301

U S A 99:8962-7. 302

20. Reynolds, P. E., H. A. Snaith, A. J. Maguire, S. Dutka-Malen, and P. Courvalin. 303

1994. Analysis of peptidoglycan precursors in vancomycin-resistant Enterococcus 304

gallinarum BM4174. Biochem J 301 ( Pt 1):5-8. 305

21. Villemur, R., M. Lanthier, R. Beaudet, and F. Lepine. 2006. The Desulfitobacterium 306

genus. FEMS Microbiol Rev 30:706-33. 307

22. Wampler, D. E., and E. W. Westhead. 1968. Two aspartokinases from Escherichia 308

coli. Nature of the inhibition and molecular changes accompanying reversible 309

inactivation. Biochemistry 7:1661-71. 310

23. Wright, G. D. 2007. The antibiotic resistome: the nexus of chemical and genetic 311

diversity. Nat Rev Microbiol 5:175-86. 312

24. Zawadzke, L. E., T. D. Bugg, and C. T. Walsh. 1991. Existence of two D-alanine:D-313

alanine ligases in Escherichia coli: cloning and sequencing of the ddlA gene and 314

purification and characterization of the DdlA and DdlB enzymes. Biochemistry 30:1673-315

82. 316

317


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

15

Figure Legends 318

Fig 1: Organization of vancomycin resistance cassettes in glycopeptide producing (Actinoplanes 319

teichomyceticus,and Streptomyces toyocaensis), pathogenic (Vancomycin Resistant Enterococci 320

Van A and B) and environmental bacteria (Paenibacillus apiarius PA-B2B, Streptomyces 321

coelicolor, Desulfitobacterium hafniense Y51 and DCB-2). 322

Fig 2. Purification of recombinant D. halfniense D-Ala-D-X-ligases. Ni-NTA purified fractions 323

of A) VanADh and B) DdlD were separated on a 4-12 % SDS-polyacrylamide gel and stained 324

with Coomassie Blue. 325

Fig 3. Substrate specificity of D-Ala:D-X ligases. Substrate specificity was examined using [U-326

14C]-D-Ala and either D-Ala or D-Lac. The products of each reaction were separated by TLC 327

and exposed to a phospor-storage screen. All reaction contain 0.1 µCi [U-14

C]-D-Ala and a) 10 328

mM D-Ala; b) 1 mM D-Ala, 10 mM D-Lac or c) 10 mM D-Lac. VanADh exhibits activity 329

similar to the D-Ala:D-Lac ligase VanAAo from Amycolatopsis orientalis while DdlDh shows 330

activity similar to the E. coli D-Ala:D-Ala ligase DdlB. The negative control (-) has no enzyme. 331

Fig 4. Change in expression levels of vanADh (striped bars) and ddlDh (shaded bars) after growth 332

in 125 µg/mL vancomycin. Relative RNA levels are normalized to 16S rRNA levels after 24 333

and 48 hours of growth. 334


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

16

Table 1: Comparison of the percent identity between DdlDh and VanADh with various Ddls: 335

DdlB (E. coli); DdlBc (B. cereus); DdlCb (C. botulinum A str. ATCC 3502); DdlStm ( S. 336

typhimurium); VanASt (S. toyocaensis); VanAAt (A. teichomyceticus); VanAAo (A. orientalis 337

C329.2); VRE VanA (E. faecium); VanASt ( S. toyocaensis) and VanA,B,D,F ( Enterococcal 338

clinical phenotypes). 339

340

Enzyme DdlDh VanADh VanXDh VanWDh

DdlB 34 31

DdlBc 31 33

DdlCb 31 34

DdlStm 35 31

VanA 35 64

VanB 37 66

VanD 35 61

VanF 36 64

VanASt 35 74

VanAAt 36 70

VanAAo 36 67

VanX 66

VanXB 65

VanXD 66

VanXF 69

VanXAt 74

VanXSt 74

VanXAo 71

VanWG 54

VanWB 51

341

342

Table 2. Kinetic characterization of D. hafniense Y51 D-Ala-D- Lac and D-Ala:D-Ala

ligase using the pyruvate kinase/lactate dehydrogenase assay or [14

C]-D-Ala radioactive

assay

Enzyme Product KM (mM) kcat (s-1

) kcat/KM (M-1

s-1)

VanADh D-Ala:D-Ala 22 ± 2.6 0.58 ± 0.03 2.6 x 101

D-Ala:D-Lac 0.53 ± 0.06 0.61 ± 0.02 1.1 x 103

ATP 0.013 ±0.001 0.65 ± 0.01 4.7 x 104

[14

C]-D-Ala:D-Lac 0.80 ± 0.08 0.0076 ± 0.003 1.7 x 102

DdlDh D-Ala:D-Ala 0.26 ± 0.05 0.80 ± 0.05 3.1 x 103

ATP 0.018 ± 0.002 0.68 ± 0.02 3.8 x 104


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

17

Figure 1 343

344

345

346


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

18

70

55

40

kDa A B

Figure 2 347

348

349

350

351

352

353

354

355


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

19

Figure 3 356

357

358

359

360

361

362

363

364

365

366

367

368

369

370

371

372

373


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

20

1

10

100

1000

24 48

Time (hrs)

Rela

tive R

NA

levels

Figure 4 374

376


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

aac accepts, published online ahead of print on 4 may...

Documents