Dominance of international ‘high-risk clones’ among Metallo- -β

Lactamase-Producing Pseudomonas aeruginosa in the United

Kingdom

Laura L. Wright1,2*, Jane F. Turton1, David M. Livermore1,2, Katie L. Hopkins1, Neil Woodford1

1Antimicrobial Resistance and Healthcare Associated Infections Reference Unit, Colindale,

Public Health England, London, NW9 5EQ, United Kingdom.

2Norwich Medical School, University of East Anglia, Norwich, Norfolk, NR4 7TJ, United

Kingdom.

Corresponding author:

Miss Laura L. Wright

Antimicrobial Resistance and Healthcare Associated Infections Reference Unit

Public Health England

61 Colindale Avenue

London NW9 5EQ

Email: laura.wright@phe.gov.uk

Tel: +44(0)2083276764

Running Title: MBL-producing P. aeruginosa in the UK

Keywords: VIM, ‘high-risk clone’, ST111, ST235, carbapenemase

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

SynopsisObjectives: Carbapenem-resistant isolates of Pseudomonas aeruginosa producing metallo-β-

lactamases (MBLs) are increasingly reported worldwide and often belong to particular ‘high-risk

clones’. This study aimed to characterise a comprehensive collection of MBL-producing P.

aeruginosa isolates referred to the UK national reference laboratory from UK laboratories over a 10-

year period.

Methods: Isolates were referred to the UK national reference laboratory between 2003 and 2012 for

investigation of resistance mechanisms and/or outbreaks. MBL genes were detected by PCR. Typing

was carried out by nine-locus variable number tandem repeat (VNTR) analysis and multi-locus

sequence typing (MLST).

Results: MBL-producing P. aeruginosa isolates were referred from 267 source patients and 89 UK

laboratories. The most common isolation sites were urine (24%), respiratory (18%), wounds (17%)

and blood (13%). VIM-type MBLs predominated (91% of all MBLs found) but a few IMP- and NDM-

type enzymes were also identified. Diverse VNTR types were seen, but 86% of isolates belonged to

six major complexes. MLST of representative isolates from each complex showed that they

corresponded to sequence types 111, 233, 235, 357, 654 and 773 respectively. Isolates belonging to

these complexes were received from between nine and 25 UK referring laboratories each.

Conclusions: The incidence of MBL-producing P. aeruginosa is increasing in the UK. The majority of

these isolates belong to several 'high-risk clones', which have been previously reported

internationally as host clones of MBLs.

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

IntroductionPseudomonas aeruginosa is a common opportunistic pathogen responsible for many hospital-

acquired infections. Carbapenems are important antibiotics for treatment of these infections, but

resistance is increasing worldwide. Although most carbapenem resistance in P. aeruginosa arises by

mutations that lead to the loss of porin OprD or (for meropenem, not imipenem) upregulation of

efflux pumps (e.g. MexAB-OprM),1 carbapenemases are also increasingly reported. Serine

carbapenemases of the KPC, GES and OXA families have occasionally been reported in this species,

with limited geographical scatter,2–4 but metallo-β-lactamases (MBLs), particularly VIM- and IMP-

types, are predominant, and have been reported globally.5

MBLs hydrolyse almost all β-lactam antibiotics, including carbapenems, and most MBL-producing P.

aeruginosa strains are resistant to other antibiotic classes, including fluoroquinolones and

aminoglycosides, often leaving polymyxins as the sole therapeutic options. The MBL genes often

reside on mobile genetic elements able to be transmitted between strains,5 but are commonly

associated with multi-resistant ‘high-risk clones’. These have been identified in several bacterial

species and, in the case of P. aeruginosa , the most commonly reported ‘high-risk clones’ belong to

sequence types (STs) 111, 235 and 175.6 Recent studies report major dissemination of an ST235

lineage with the VIM-2 MBL in Russia, Belarus and Kazakhstan,7 and of ST277 with the SPM-1 MBL in

Brazil.8

In recent years, the number of MBL-producing P. aeruginosa referred to the UK national reference

laboratory has steadily risen. We sought to determine the contribution of internationally-recognised

‘high-risk clones’ to this increase.

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

Materials and Methods

Bacterial isolatesThree-hundred and thirty-four MBL-positive isolates were identified amongst P. aeruginosa isolates

referred from UK hospital laboratories to Public Health England’s (PHE) Antimicrobial Resistance and

Healthcare Associated Infections Reference Unit (formerly the Health Protection Agency’s ARMRL

and LHCAI laboratories) between 2003 and 2012 for susceptibility testing, investigation of resistance

mechanisms, and/or strain typing. Isolates were identified as MBL-producers by PCR, either at the

time of referral (304 isolates), or by retrospective screening of isolates submitted from 2009-2012

that shared similar VNTR types with other MBL-producing isolates (30 isolates). Demographic and

isolation site data were obtained from the Laboratory Information Management System. Referring

laboratories were assigned codes to indicate the UK region and given a unique number within the

region, in the format “region_number” (e.g. North West_1).

As a comparator group, all 209 P. aeruginosa isolates collected in 2011 as part of the British Society

for Antimicrobial Chemotherapy’s (BSAC) Bacteraemia Surveillance Programme from hospitals in the

UK and Irish Republic were also studied (http://www.bsacsurv.org/).

Analysis of MBL genesGenes encoding VIM-, IMP-, SPM-, GIM-, and SIM-type MBLs were sought by multiplex PCR, as

detailed by Ellington et al.9 Genes for NDM-type MBLs were sought by a single PCR, as previously

described.10 Sequencing of blaVIM and blaIMP MBL genes was carried out using previously-described

primers specific to MBL genes and class 1 integrons;11 sequencing of blaNDM genes was with primers

NDM-orfF (5’-ATGGAATTGCCCAATATTATG-3’) and NDM-orfR (5’-TCAGCGCAGCTTGTCGGCCA-3’).

TypingPulsed-Field Gel Electrophoresis (PFGE) was the routine typing and outbreak investigation method

used in the reference laboratory for P. aeruginosa between 2003 and 2009, and was carried out

using SpeI-digested genomic DNA.12 Nine-locus variable number tandem repeat (VNTR) analysis

became the routine method from 2009 onwards, and was performed as previously described. 12

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

Minimum spanning trees were produced using Bionumerics software v6.1 (Applied Maths, Sint-

Martens-Latem, Belgium). Multi-locus sequence typing (MLST) analysis was undertaken as described

by Curran et al;13 results were analysed using Bionumerics software, and sequence types (STs)

assigned using the P. aeruginosa MLST database (http://pubmlst.org/paeruginosa/).

Results

Patient demographicsThe 334 MBL-positive isolates were from a total of 267 patients, with VNTR profiles remaining

consistent when multiple isolates were received from the same patient. Accordingly, one isolate per

patient was selected for further study, leaving 267 non-duplicate-patient isolates from 89 UK

laboratories. Ages of the source patients ranged from 0-94, mean 54 yrs: proportions in age bands

were as follows: 0-2years, 4%; 2-15, 2%; 16-29, 6%; 30-44, 11%; 45-59, 22%; 60-74, 32% and >74,

12% . Predominant isolation sites were urine (24%), respiratory (18%), wounds (17%) blood (13) and

indwelling devices (7%) whereas few were from skin (3%) and faecal (3%) specimen lastly to

represent screeing rather than infection.. Sixty percent of patients were male, 35% female, and for

5% gender was not stated. blaVIM genes alone were detected in 243 isolates, blaIMP in 22, and blaNDM in

one; one further isolate had both blaVIM and blaNDM. Of the 89 referring laboratories, 83 submitted

isolates from fewer than ten patients over the ten-year period, with 71 submitting only one to two

isolates each. The remaining six laboratories, which submitted between 13 and 31 isolates each,

collectively accounted for 43% of all patients (116/267).

TypingAmong the 267 non-duplicate-patient isolates, 232 had VNTR data available, or generated in this

study, and a minimum spanning tree based on these data is shown in figure 1. The remaining 35

isolates were not available in the archive for retrospective VNTR typing, but 19 had previously-

determined PFGE profiles (not shown) identical to those of isolates with known VNTR types from a

suspected outbreak at the same hospital. These isolates were assumed to belong to the

89

90

91

92

93

9495

96

97

98

99

100

101

102

103

104

105

106

107

108

109

110

111

112

113

corresponding VNTR complex and are included in Table 1. The remaining 16/35 unavailable isolates

had no typing data, or belonged to PFGE types unique to their respective hospital; these latter

isolates, from 12 laboratories, are excluded from Table 1; all carried blaVIM.

Six VNTR complexes (designated A-F) accounted for 86% of the 251 isolates with VNTR data available

or inferred from PFGE data. Isolates belonging to these six complexes were referred from between

nine and 25 laboratories and, owing to their predominance,became the focus of further study.

Sixty-four representatives, covering the variation in VNTR profile within each complex, were selected

for MLST analysis. Isolates belonging to complexes A, B, C, D, E and F were found to belong to ST111,

ST235, ST233, ST654/ST964, ST357 and ST773, respectively (Table 1). Ten of 11 MLST-typed complex

D isolates belonged to ST654, but one belonged to ST964; this is a single locus variant (SLV) of ST654,

differing only in the acsA allele, where ST964 has allele 145 with a single C to T substitution

compared with allele 17 in ST654. Table 2 shows the distribution of each of the major complexes

among referring laboratories. The largest group of isolates was complex A (corresponding to ST111;

VNTR type 11,3,4,3,2,2,x,4,x, where x is variable), with 75 representatives. It included isolates from

25 laboratories spread across the UK, but with more than half of the isolates coming from outbreaks

at London_17 and Wales_1, referring 29 and 13 isolates, respectively, all with blaVIM. Most other

complex A isolates (28/33) also had blaVIM, but five, from two laboratories, had blaIMP. VNTR profiles

were highly conserved amongst all complex A isolates, with most only differing by repeat numbers

between six and eight at locus 61.

Complex B (ST235; 13,3,6,4,5,1,x,2,x) was the secondlargest group, comprising 52 isolates referred

from 25 different laboratories. There were potential outbreaks at sites London_13 and Scotland_2,

referring eight and six isolates, all with blaVIM; these groups were received over seven months and

two years respectively. Another laboratory (London_7) also submitted isolates carrying blaVIM from

eight patients over an eight-year period. The remaining 30 Complex B isolates were from 22

laboratories, each submitting one to three isolates; 24 isolates had blaVIM, and six had blaIMP.

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

Twenty-six isolates belonged to complex C (ST233; 12,3,4,5,3,1,x,x,x). These all carried blaVIM and

were from 16 laboratories, which submitted one to four representatives each. Nineteen isolates

from 11 laboratories belonged to complex D (ST654; 11,3,2,15,3,1,x,3,x); seven of these were

submitted from South East_6 and all had blaVIM and shared an identical VNTR profile, whilst one or

two isolates were referred from each of the remaining ten laboratories; ten isolates had blaVIM, one

had blaIMP, and one blaNDM. Most of the 30 complex E isolates (ST357; 13,2,1,5,2,3,6,5,x), all with

blaVIM, were from an outbreak at North West_15, with 22 isolates referred over a nine-year period;

12 of these 22 were submitted in 2007. Each of the remaining eight complex E isolates was referred

from a different laboratory. Finally, 13 isolates, all with blaVIM, belonging to complex F (ST773;

12,4,6,5,3,1,10,x,x) were received from 11 laboratories in diverse areas of the UK.

The diversity of carbamenemes gene sequences was investigated by testing representatives of the

VNTR variationwithin the six major complexes. Irrespective of VNTR variation all blaVIM-–positive

representatives of complexes A, C, D, E and F (STs 111, 233, 654, 357 and 773) had blaVIM-2, whereas

the eight isolates selected from complex B (ST235) variously had blaVIM-1 (2 isolates), blaVIM-2 (4),

blaVIM-4 (1) and blaVIM-6 (1). blaIMP-1 (3 isolates), blaIMP-7 (1), blaIMP-10 (1) and blaIMP-13 (2) alleles were seen

amongst seven representatives of the 22 blaIMP positive isolates variously belonging to complexes

XXXXXXXXXXXX,. Both blaNDM positive isolates, one belonging to complex D, the other a ‘diverse’

strain, had the blaNDM-1 allele.

The 14% of isolates (n=36) that did not belong to any of these six major complexes were diverse by

VNTR, with 26 different VNTR profiles represented. They were referred from 24 laboratories across

the UK, each submitting one to three isolates each; 26 had blaVIM, 10 had blaIMP and one had both

blaVIM and blaNDM genes.

Regional distribution of MBL-positive isolates The UK distribution of the 267 non-duplicate-patient isolates is shown in figure 2. London accounted

for 47% of these organisms, with MBL-positive isolates referred from 28 London-region laboratories;

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161162

163

all six major complexes (A-F) were represented and nine other VNTR types seen. Four London

laboratories were amongst the six sites that submitted more than ten MBL-positive isolates.

London_17 submitted 29 complex A (ST111) isolates over six years, with source patients on multiple

wards. London_13 had separate outbreaks of MBL-positive isolates belonging to complexes B and C

(STs 235 and 233), with eight and four representatives each, respectively. Each “outbreak” lasted

one to two years separated by a six-year gap. London_7 and London_12, referred 19 and 15 isolates,

respectively, over seven to eight years, with isolates belonging to diverse VNTR complexes.

Greater Manchester accounted for 12% of isolates (31/267). Isolates belonging to complex E (ST357)

predominated and included all 22 from an outbreak at North West_13, although isolates belonging

to complexes A, B and D (STs 111, 235 and 654) also were seen. In Sussex, Surrey and Kent half of

the 18 isolates received belonged to complex D (ST654); seven of these 18 were from South East_6;

the remaining nine belonged to complexes B and C (STs 235 and 233) or had other VNTR profiles.

Wales accounted for 6% of isolates, mostly (13/15) from an outbreak of ST111 P. aeruginosa at

Wales_1 although one ST111 isolate was from another laboratory, and one ST773 isolate also was

referred from the region. Finally, the 12 (4%) isolates submitted from Scotland belonged to

complexes A, B, C and F (STs 111, 235, 233 and 773). Fewer than ten isolates were referred from

each of the remaining UK regions.

Typing of comparator P. aeruginosa isolatesThe comparator set, comprising P. aeruginosa isolates collected as part of the BSAC Bacteraemia

Resistance Surveillance Programme in 2011 showed far greater diversity than the MBL producers,

with 136 different VNTR profiles represented amongst the 209 isolates (figure 3). Although 4·8% (10

isolates) were imipenem resistant, none produced MBLs. Two isolates shared VNTR profiles with

complex B (ST235) and one shared a profile with complex C (ST233); the remaining 206 isolates did

not share VNTR profiles with any of the major complexes identified amongst the MBL-producing

isolates.

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181182

183

184

185

186

187

188

Discussion

Most (86%) MBL-producing P. aeruginosa isolates referred to the UK reference laboratory between

2003 and 2012 belonged to six VNTR complexes, corresponding to internationally recognised ‘high-

risk clones’, STs 111, 235, 233, 357, 654 and 773, and most had VIM-type MBLs. These findings are in

striking contrast to the general population structure of P. aeruginosa, where a recent UK study14

shows considerable diversity, with overlap between environmental isolates and those causing

clinical infection, albeit with a few prevalent clusters in diverse locations. The major complexes seen

here were not prevalent amongst MBL-negative isolates in this previous study, nor were they

prevalent in our comparator collection of (largely susceptible) P. aeruginosa isolates from the BSAC

Bacteraemia Resistance Surveillance Programme. Rather, these ‘high-risk clones’ seem to represent

a distinct subset of P. aeruginosa lineages, which may be successful precisely due to a particularly

strong ability to acquire and/or maintain resistance genes compared with the general P. aeruginosa

population. Notably, the reference laboratory has also received MBL-negative representatives

corresponding to these six major complexes, including ST235 isolates carrying blaGES-5 and a ST773

isolate with blaOXA-181, both genes encoding non-metallo-carbapenemases (unpublished data; J.

Turton, K. Hopkins). Moreover, in three of these complexes, we encountered isolates that variously

had either blaVIM or blaIMP genes, and in one complex (XXX), various different blaVIM alleles.

Each of the six main VNTR complexes has been reported internationally as a host for MBLs. Complex

A (corresponding to ST111) was the largest group in our collection, with most isolates carrying blaVIM-

2. VNTR profiles within this complex were highly similar; most differing only at locus 61, which is the

most variable of the nine VNTR loci.12 ST111 is frequently recorded amongst MBL-producing P.

aeruginosa, with ST111 isolates reported to be of serotype O12, which has been recognised as a

common host of multiresistance in Europe since the 1980s.15,16 ST111 was the predominant ST type

189

190191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

seen in a nationwide study of MBL-producing P. aeruginosa in The Netherlands in 2010-11,16 and

was responsible for outbreaks affecting hospitals in central Greece,17 and Italy.18 In all these settings,

as in the UK, the ST111 isolates generally carried blaVIM-2. Smaller numbers of ST111 isolates with

blaVIM-2 have been reported in other European countries including Sweden,11 Croatia,19, Spain20 and

Germany,21 whilst isolates with this ST have been associated with blaVIM-4 in Hungary,22 blaVIM-1 in

Greece23 and blaIMP-13 in France.24 Outside Europe, an isolate of ST111 from a patient in Colombia was

recently found to carry both blaVIM-2 and blaKPC-2.25

The second-most-prevalent complex (B) in our collection corresponded to ST235. Isolates in this

complex had more diverse VNTR profiles than those in complex A, and were more widely

disseminated among referring laboratories, with no more than eight from any single site. Again,

most had blaVIM, but with diverse blaVIM alleles identified. Together with the VNTR diversity, this

carbapenemase diversity suggests multiple imports and/or acquisitions of blaVIM by ST235 P.

aeruginosa. This variation is in contrast to the ST235 clone with blaVIM-2 that is widespread across

Russia, Belarus and Kazakhstan.7 ST235 was the most commonly identified MBL-producing P.

aeruginosa type in a study of five Mediterranean countries,26 and ST235 isolates with blaVIM-2 genes

have also been reported in Spain,20 Croatia,19 Germany,21 and Greece,23 blaVIM-4 in Hungary,22

Norway,11 and Belgium,27 and blaVIM-13 in Spain.28 In Asia, outbreaks of ST235 clones with blaIMP-6 genes

have occurred in Japan,29 and South Korea,30 with blaVIM-positive isolates with this ST also seen in

Thailand, Malaysia, Sri Lanka and Korea.31 There are single reports of ST235 isolates carrying blaNDM-1

and blaSPM-1 genes in France,32 and Brazil,8 respectively.

The other four major complexes (C-F) also correspond to ‘high-risk clones,’ previously reported as

hosts for MBLs. ST233 (complex C) isolates with blaVIM-2 have been found in Norway (imported from

Ghana),11 Japan,33 South Africa,34 and carrying a blaIMP-1 variant in Singapore.35 We likewise

consistently saw blaVIM-2 in complex C isolates. ST654 (complex D) has been reported in Sweden

(imported from Tunisia) carrying blaVIM-2 genes,11 in Singapore carrying blaIMP-1 and blaIMP-26,35 and in

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

Argentina carrying a KPC carbapenemase.4 We predominantly saw blaVIM-2 in ST654 isolates, but one

isolate had blaNDM-1. The sole ST964 isolate (also complex D) had blaIMP-1, as also reported in this ST in

Singapore.35 We saw only blaVIM-2 in representatives of complex E (ST357), but this ST has previously

been reported with blaVIM-2 or blaIMP-7 genes in a hospital in the Czech Republic,36 whilst a few ST357

isolates with blaIMP-7 were reported in Poland.37 ST773 (complex F) has been recently reported among

Indian isolates carrying blaVIM-2.31 Interestingly, complex F isolates tested here also all had blaVIM-2 and

four isolates were from patients who had recently travelled to India; travel history for the other nine

patients was not available. The remaining VNTR profiles, accounting for 14% of MBL producers, were

diverse and probably represent separate acquisitions of blaVIM and blaIMP genes.

Ongoing research indicates a variety of VIM-containing class 1 integron structures amongst the MBL

producers. Complexes A, D, E and F (STs 111, 654, 773 and 357) typically have a single predominant

blaVIM-2 carrying integron, whereas a different (but converved) blaVIM-2-carrying integron was seen in

all complex C isolates and some belonging to complex B (STs 233 and 235) (Wright et al.,

unpublished data).

Although comprehensive epidemiological data are lacking, VIM-type MBLs generally are the

predominant carbapenemases seen in P. aeruginosa in Europe, with only sporadic isolations, and/or

local spread of strains producing IMP- or NDM- types. The 22 isolates harbouring blaIMP genes

belonged to diverse VNTR types, with diverse blaIMP alleles and no known epidemiological links

between most isolates. Just two blaNDM-1-positive isolates were found, both with different VNTR

types. These IMP- and NDM-MBL-producing isolates may have been imported from outside the UK,

or acquired locally, with local spread at a small number of sites. Unfortunately, data on patient travel

were not available for most isolates, but single isolates carrying blaNDM and blaIMP were from patients

who had travelled to India and Pakistan, respectively.

MBL-positive P. aeruginosa were referred from around half the hospital laboratories in the UK, with

all six major complexes found in multiple UK regions. Referral of suspect isolates is not mandatory,

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

and some likely MBL producers were no longer viable in our archives, so the numbers reported here

under-estimate the true incidence of MBL-producing P. aeruginosa. They were rare at most referring

sites, but a few sites did have persistent problems with single clones. These include an outbreak of

ST111 isolates at London_17 and of ST654 isolates at South East_6, both associated with

contamination of the waste-water networks,38 and an outbreak of ST357 P. aeruginosa at North

West_15 where the strain, which also produced a VEB-1a extended-spectrum-β-lactamase, was

thought to have been imported via a patient transferred from an Indian hospital, but to have

acquired the VIM-MBL locally in the UK.39 In contrast, two laboratories (London_7 and London_12)

referred MBL-positive isolates of diverse types over seven and eight year periods, implying that MBL-

producing P. aeruginosa had been introduced repeatedly.

Since these ‘high-risk clones’ are reported amongst MBL-producing P. aeruginosa worldwide, it is

important not to assume that UK isolates with the same VNTR or ST profile are directly related

unless this view is supported by evidence of epidemiological links between affected patients; it is

just as likely that cases could represent repeated imports of the same clone from different sources.

Although, MBL-producing P. aeruginosa have so far caused outbreaks at few UK hospitals, those that

have occurred have been linked to environmental reservoirs within the hospitals, including waste-

water networks, as well as to patient-to-patient cross-infection.

AcknowledgmentsWe thank staff at the Antimicrobial Resistance and Healthcare Associated Infections Reference Unit

for carrying out PFGE and some of the VNTR typing and MBL detection, performed as part of the

reference service. We are grateful to the UK hospital laboratories for submitting isolates to us and to

the British Society for Antimicrobial Chemotherapy for allowing us to access isolates from their

Bacteraemia Surveillance Programme. Part of this work was presented at the 53rd ICAAC, 2013,

Denver, USA (Abstract: C2-1597).

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

FundingThis work was supported by Public Heath England through a competitively awarded PhD

studentship.

Transparency DeclarationsN.W. and D.L. have received research grants and/or fees from various pharmaceutical companies.

D.L. also holds shares in several pharmaceutical companies. None of these poses a conflict of

interest with this work. All other authors; none to declare.

References

1. Breidenstein EBM, de la Fuente-Nunez C, Hancock REW. Pseudomonas aeruginosa: all roads lead to resistance. Trends Microbiol 2011; 19: 419–26.

2. Sevillano E, Gallego L, García-Lobo JM. First detection of the OXA-40 carbapenemase in P. aeruginosa isolates, located on a plasmid also found in A. baumannii. Pathol Biol (Paris) 2009; 57: 493–5.

3. Viedma E, Juan C, Acosta J, et al. Nosocomial spread of colistin-only-sensitive sequence type 235 Pseudomonas aeruginosa isolates producing the extended-spectrum beta-lactamases GES-1 and GES-5 in Spain. Antimicrob Agents Chemother 2009; 53: 4930–3.

4. Pasteran F, Faccone D, Gomez S, et al. Detection of an international multiresistant clone belonging to sequence type 654 involved in the dissemination of KPC-producing Pseudomonas aeruginosa in Argentina. J Antimicrob Chemother 2012; 67: 1291–3.

5. Cornaglia G, Giamarellou H, Rossolini GM. Metallo-β-lactamases: a last frontier for β-lactams? Lancet Infect Dis 2011; 11: 381–93.

6. Woodford N, Turton JF, Livermore DM. Multiresistant Gram-negative bacteria: the role of high-risk clones in the dissemination of antibiotic resistance. FEMS Microbiol Rev 2011; 35: 736–55.

7. Edelstein MV, Skleenova EN, Shevchenko O V, et al. Spread of extensively resistant VIM-2-positive ST235 Pseudomonas aeruginosa in Belarus, Kazakhstan, and Russia: a longitudinal epidemiological and clinical study. Lancet Infect Dis 2013; 13: 1–10.

8. Silva FM, Carmo MS, Silbert S, et al. SPM-1-producing Pseudomonas aeruginosa: analysis of the ancestor relationship using multilocus sequence typing, pulsed-field gel electrophoresis, and automated ribotyping. Microb drug Resist 2011; 17: 215–20.

9. Ellington MJ, Kistler J, Livermore DM, et al. Multiplex PCR for rapid detection of genes encoding acquired metallo-β-lactamases. J Antimicrob Chemother 2007; 59: 321–2.

288

289

290

291

292

293

294

295

296297

298299300

301302303

304305306

307308

309310

311312313

314315316

317318

10. Mushtaq S, Irfan S, Sarma JB, et al. Phylogenetic diversity of Escherichia coli strains producing NDM-type carbapenemases. J Antimicrob Chemother 2011; 66: 2002–5.

11. Samuelsen O, Toleman MA, Sundsfjord A, et al. Molecular epidemiology of metallo-beta-lactamase-producing Pseudomonas aeruginosa isolates from Norway and Sweden shows import of international clones and local clonal expansion. Antimicrob Agents Chemother 2010; 54: 346–52.

12. Turton JF, Turton SE, Yearwood L, et al. Evaluation of a nine-locus variable-number tandem-repeat scheme for typing of Pseudomonas aeruginosa. Clin Microbiol Infect 2010; 16: 1111–6.

13. Curran B, Jonas D, Grundmann H, et al. Development of a Multilocus Sequence Typing Scheme for the Opportunistic Pathogen Pseudomonas aeruginosa Development of a Multilocus Sequence Typing Scheme for the Opportunistic Pathogen Pseudomonas aeruginosa. J Clin Microbiol 2004; 42: 5644–9.

14. Martin K, Baddal B, Mustafa N, et al. Clusters of genetically similar isolates of Pseudomonas aeruginosa from multiple hospitals in the UK. J Med Microbiol 2013; 62: 988–1000.

15. Pitt TL, Livermore DM, Pitcher D, et al. Multiresistant serotype O12 Pseudomonas aeruginosa: evidence for a common strain in Europe. Epidemiol Infect 1989; 103: 565–76.

16. Van der Bij AK, Van der Zwan D, Peirano G, et al. Metallo-beta-lactamase-producing Pseudomonas aeruginosa in the Netherlands: the nationwide emergence of a single sequence type. Clin Microbiol Infect 2012; 18: E369–E372.

17. Liakopoulos A, Mavroidi A, Katsifas EA, et al. Carbapenemase-producing Pseudomonas aeruginosa from central Greece: molecular epidemiology and genetic analysis of class I integrons. BMC Infect Dis 2013; 13: 505.

18. Edalucci E, Spinelli R, Dolzani L, et al. Acquisition of different carbapenem resistance mechanisms by an epidemic clonal lineage of Pseudomonas aeruginosa. Clin Microbiol Infect 2008; 14: 88–90.

19. Sardelic S, Bedenic B, Colinon-Dupuich C, et al. Infrequent finding of metallo-β-lactamase VIM-2 in carbapenem-resistant Pseudomonas aeruginosa strains from Croatia. Antimicrob Agents Chemother 2012; 56: 2746–9.

20. Viedma E, Estepa V, Juan C, et al. Comparison of Local Features from Two Spanish Hospitals Reveals Common and Specific Traits at Multiple Levels of the Molecular Epidemiology of Metallo-β-Lactamase-Producing Pseudomonas spp. Antimicrob Agents Chemother 2014; 58: 4992.

21. Hentschke M, Goritzka V, Campos CB, et al. Emergence of carbapenemases in Gram-negative bacteria in Hamburg, Germany. Diagn Microbiol Infect Dis 2011; 71: 312–5.

22. Libisch B, Watine J, Balogh B, et al. Molecular typing indicates an important role for two international clonal complexes in dissemination of VIM-producing Pseudomonas aeruginosa clinical isolates in Hungary. Res Microbiol 2008; 159: 162–8.

23. Koutsogiannou M, Drougka E, Liakopoulos A, et al. Spread of Multidrug-Resistant Pseudomonas aeruginosa Clones in a University Hospital. J Clin Microbiol 2013; 51: 665–8.

319320

321322323

324325

326327328329

330331

332333

334335336

337338339

340341

342343344

345346347

348349

350351352

353354

24. Fournier D, Jeannot K, Robert-Nicoud M, et al. Spread of the blaIMP-13 gene in French Pseudomonas aeruginosa through sequence types ST621, ST308 and ST111. Int J Antimicrob Agents 2012; 40: 562–3.

25. Correa A, Montealegre MC, Mojica MF, et al. First report of a Pseudomonas aeruginosa isolate coharboring KPC and VIM carbapenemases. Antimicrob Agents Chemother 2012; 56: 5422–3.

26. Maatallah M, Cheriaa J, Backhrouf A, et al. Population structure of Pseudomonas aeruginosa from five Mediterranean countries: evidence for frequent recombination and epidemic occurrence of CC235. PLoS One 2011; 6: e25617.

27. Glupczynski Y, Bogaerts P, Deplano A, et al. Detection and characterization of class A extended-spectrum-beta-lactamase-producing Pseudomonas aeruginosa isolates in Belgian hospitals. J Antimicrob Chemother 2010; 65: 866–71.

28. Juan C, Beceiro A, Gutiérrez O, et al. Characterization of the new metallo-beta-lactamase VIM-13 and its integron-borne gene from a Pseudomonas aeruginosa clinical isolate in Spain. Antimicrob Agents Chemother 2008; 52: 3589–96.

29. Kitao T, Tada T, Tanaka M, et al. Emergence of a novel multidrug-resistant Pseudomonas aeruginosa strain producing IMP-type metallo-β-lactamases and AAC(6’)-Iae in Japan. Int J Antimicrob Agents 2012; 39: 518–21.

30. Seok Y, Bae IK, Jeong SH, et al. Dissemination of IMP-6 metallo-β-lactamase-producing Pseudomonas aeruginosa sequence type 235 in Korea. J Antimicrob Chemother 2011; 66: 2791–6.

31. Kim MJ, Bae IK, Jeong SH, et al. Dissemination of metallo-β-lactamase-producing Pseudomonas aeruginosa of sequence type 235 in Asian countries. J Antimicrob Chemother 2013; 68: 2820–4.

32. Janvier F, Jeannot K, Tessé S, et al. Molecular characterization of blaNDM-1 in a sequence type 235 Pseudomonas aeruginosa isolate from France. Antimicrob Agents Chemother 2013; 57: 3408–11.

33. Tsutsui A, Suzuki S, Yamane K, et al. Genotypes and infection sites in an outbreak of multidrug-resistant Pseudomonas aeruginosa. J Hosp Infect 2011; 78: 317–22.

34. Jacobson RK, Minenza N, Nicol M, et al. VIM-2 metallo-β-lactamase-producing Pseudomonas aeruginosa causing an outbreak in South Africa. J Antimicrob Chemother 2012; 67: 1797–8.

35. Koh TH, Khoo CT, Tan TT, et al. Multilocus sequence types of carbapenem-resistant Pseudomonas aeruginosa in Singapore carrying metallo-beta-lactamase genes, including the novel bla(IMP-26) gene. J Clin Microbiol 2010; 48: 2563–4.

36. Papagiannitsis CC, Studentova V, Ruzicka F, et al. Molecular characterization of metallo-β-lactamase-producing Pseudomonas aeruginosa in a Czech hospital (2009-2011). J Med Microbiol 2013; 62: 945–7.

37. Hrabák J, Cervená D, Izdebski R, et al. Regional spread of Pseudomonas aeruginosa ST357 producing IMP-7 metallo-β-lactamase in Central Europe. J Clin Microbiol 2011; 49: 474–5.

355356357

358359

360361362

363364365

366367368

369370371

372373

374375

376377

378379

380381

382383384

385386387

388389

38. Breathnach AS, Cubbon MD, Karunaharan RN, et al. Multidrug-resistant Pseudomonas aeruginosa outbreaks in two hospitals: association with contaminated hospital waste-water systems. J Hosp Infect 2012; 82: 19–24.

39. Woodford N, Zhang J, Kaufmann ME, et al. Detection of Pseudomonas aeruginosa isolates producing VEB-type extended-spectrum beta-lactamases in the United Kingdom. J Antimicrob Chemother 2008; 62: 1265–8.

390391392

393394395

Table 1: Typing data for the six main VNTR complexes identified (n=251)

VNTR Complex

VNTR type a No. of different VNTR profiles

MLST type(s) (no. of isolates tested)

No. of isolatesb

No. of submitting laboratories

MBLs detected (no. of isolates)

A 11,3,4,3,2,2,x,4,x 6 ST111 (11) 75 25 VIM (70)

IMP (5)

B 13,3,6,4,5,1,x,2,x 16 ST235 (18) 52 25 VIM (46)

IMP (6)

C 12,3,4,5,3,1,x,2,x 11 ST233 (10) 26 16 VIM (26)

D 11,3,2,15,3,1,x,3,x 6 ST654 (10), ST964 (1)

19 11 VIM (17)

IMP (1)

NDM (1)

E 13,2,1,5,2,3,6,x,x 7 ST357 (9) 30 9 VIM (30)

F 12,4,6,5,3,1,10,x,x 3 ST773 (5) 13 11 VIM (13)

Others Diverse 26 Not done 36 25 VIM (25)

IMP (10)

VIM and NDM (1)

ax represents loci where the repeat number varies between isolates within a complex

bone isolate per patient was included; these numbers include four isolates (complex B), 14 isolates

(complex E) and one isolate (complex F) where the MBL-positive organisms were no longer available

in the archive for VNTR analysis, but which were previously found to share a PFGE profile, and are

from the same hospital outbreak as other isolates in the respective complex. Isolates were also

received from an additional 39 patients at London_17 with a PFGE profile corresponding to complex

A. These are not included here as they had not been screened for MBL genes and were no longer

available in our archives.

396

397

398

399

400

401

402

403

404

405

Table 2: Geographic and temporal distribution of MBL-producing P. aeruginosa belonging to the six

major complexes (215 isolates) among referring laboratories

VNTR Complex

Major Contributors (≥ 5 referred isolates) Minor Contributors

Referring laboratorya

Number of isolatesb

Time period over which isolates referred

MBL genes detected

A (n=75) London_17 c 29 (39%) 80 months all with blaVIM 2 laboratories with 4 representatives each

Wales_1 13 (17%) 18 months all with blaVIM 2 laboratories with 3 representatives each

19 laboratories with 1 representative each

B (n=52) London_7 8 (15%) 87 months all with blaVIM 1 laboratory with 3 representatives each

London_13 8 (15%) 7 months all with blaVIM 6 laboratories with 2 representatives each

Scotland_2 6 (12%) 27 months all with blaVIM 15 laboratories with 1 representative each

C (n=26) London_7 5 (19%) 27 months all with blaVIM 3 laboratories with 2 representatives each

London_13 4 (15%) 18 months all with blaVIM 11 laboratories with 1 representative each

D (n=19) South East_6 7 (37%) 36 months all with blaVIM 2 laboratories with 2 representatives each

8 laboratories with 1 representative each

E (n=30) North West_15 22 (73%) 103 months all with blaVIM 8 laboratories with 1 representative each

F (n=13) No major contributors 2 laboratories with 2 representatives each

9 laboratories with 1 representative each

areferring laboratories are coded in the format ‘UK region_number’.

bone isolate per patient is included; these numbers include four isolates (complex B), 14 isolates

(complex E) and one isolate (complex F) that were not available in the archive for VNTR analysis, but

406

407

408

409

410

which shared a PFGE profile and are from the same hospital outbreak as other isolates in the

complex.

c Isolates were also received from an additional 39 patients at London_17 with a PFGE type

corresponding to complex A. These are not included here as they had not been screened for MBL

production and were no longer available in our archives.

Figure 1: Minimum spanning tree, based on clustering at the first eight VNTR loci for MBL-positive P.

aeruginosa, with one isolate included per patient (n=232). The six main complexes A-F are labelled.

Coloured segments of the circles indicate laboratories that submitted three or more isolates, white

segments represent laboratories submitting one or two isolates. The diameter of the circle is relative

to the number of isolates with that VNTR profile. Shading indicates complexes. Thick solid lines

represent single locus variants; thin solid lines and dotted lines represent multi-locus variants.

Figure 2: Geographical sources of isolates in the UK, and distribution of the six main complexes (A-F)

in each of the five regions referring more than ten isolates. This figure appears in colour in the online

version of JAC and in black and white in the print version of JAC.

Figure 3: Minimum spanning tree based on clustering at the first eight VNTR loci for P. aeruginosa

isolates from the BSAC Bacteraemia Resistance Surveillance Programme (n=209). The diameter of

the circle is relative to the number of isolates with that VNTR profile. Shading indicates complexes.

Thick solid lines represent single locus variants while thin solid lines and dotted lines represent multi-

locus variants. Isolates corresponding to VNTR complexes B and C (STs 235 and 233) are shown in

black circles. Isolates corresponding to previously reported clones as described by Martin et al.14 are

indicated and had the following VNTR profiles; Clusters A (8,3,4,5,2,3,5,2,x), D (10,3,5,5,4,1,3,x,x,), E

(11,4,5,2,2,1,x,2,x) and H (12,5,1,5,2,2,x,x,x), Clone C (11,4,5,2,2,1,x,2,x), and PA14

411

412

413

414

415

416

417

418

419

420

421

422

423

424

425

426

427

428

429

430

431

432

433

434

(12,2,1,5,5,2,x,5,x). Two novel clusters are indicated with VNTR profiles (12,5,5,5,4,3,7,6,x) and

(12,8,2,2,4,3,5,1,x) respectively.

435

436

437

Figure 2438

439

Figure 3

1

2

2

2

3

4

2

3

6

3

3

7

5

5

2

5

6

28

1

2

2

2

3

5

5

6

7

7

7

9

9

12

15

18

31

126

0 50 100 150

Devon, Cornwall and Somerset

Avon, Gloucestershire and Wiltshire

North East

Northern Ireland

Bedfordshire, Hertfordshire andNorthamptonshire

Hampshire, Isle of Wight and Dorset

Lincolnshire, Leicestershire,Nottinghamshire and Derbyshire

Thames Valley

Cheshire and Merseyside

Cumbria and Lancashire

Norfolk, Suffolk, Cambridgeshire and Essex

West Midlands

Yorkshire and Humber

Scotland

Wales

Sussex, Surrey and Kent

Greater Manchester

London

Isolates

Referring hospitals

440

441

Figure 4

PA14 (ST253)

Cluster H

Clone C (ST17)

Complex B (ST235)

Complex C (ST233)

Cluster A (ST27) Cluster D

(ST395)Cluster E

Novel cluster 1

Novel cluster 2

442

443