Evolution of IncHI1 plasmids: Two distinct lineages

Accepted Manuscript Evolution of IncHI1 Plasmids: Two Distinct Lineages Amy K. Cain, Ruth M. Hall PII: DOI: Reference:

S0147-619X(13)00033-4 http://dx.doi.org/10.1016/j.plasmid.2013.03.005 YPLAS 2159

To appear in:

Plasmid

Received Date: Accepted Date:

16 February 2013 27 March 2013

Please cite this article as: Cain, A.K., Hall, R.M., Evolution of IncHI1 Plasmids: Two Distinct Lineages, Plasmid (2013), doi: http://dx.doi.org/10.1016/j.plasmid.2013.03.005

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Evolution of IncHI1 Plasmids: Two Distinct Lineages

1 2

Amy K. Cain1 and Ruth M. Hall

3 4 5

School of Molecular Bioscience, The University of Sydney, NSW, 2006, Australia

6 7 8

Corresponding author.

9

Ruth M. Hall

10

Mailing address: School of Molecular Bioscience,

11

Molecular Bioscience Building G08,

12

The University of Sydney, NSW 2006, Australia.

13

Phone: 61-2-9351-3465.

14

Fax: 61-2-9351-4571.

15

E-mail: [email protected]

16 17

1

18

Pathogen Genomics

19

Wellcome Trust Sanger Institute

20

Hinxton, Cambridge, U.K.

Present address:

21 22

Abbreviation: MARR Multiple antibiotic resistance region

23

Abstract

24

The IncHI1 plasmid pSRC27-H from Salmonella enterica serovar Typhimurium carries a region

25

containing several genes that confer resistance to different antibiotics, and this resistance region

26

is in the same position as related resistance regions in a group of sequenced IncHI1 plasmids

27

from various sources that includes pHCM1. Four further additional segments are found in

28

pHCM1 relative to another IncHI1 plasmid, R27. Using PCR or DNA sequencing to detect the

29

presence or absence of each of these additional segments in the same position in the IncHI1

30

backbone, plasmid pSRC27-H was found to include them. However, in one case the additional

31

segment was smaller in pSRC27-H, lacking a transposon carrying a second resistance region in

32

pHCM1. The sequences of IncHI1 plasmids, pO111_1 and pMAK1, were also examined and

33

found to share the same or closely related additional segments. The structure of the additional

34

material in pHCM1, pO111_1 and pMAK1 was examined, and potential novel transposons were

35

identified. These additional segments define an IncHI1 lineage (pHCM1, pO111_1, pMAK1,

36

pSRC27-H) which we designated type 2 to distinguish it from type 1 (R27, pAKU_1, pP-stx-12).

37

A segment from the Escherichia coli genome and an adjacent copy of IS1 in pHCM1 was defined

38

by comparison to pO111_1 and pMAK1, which lack it. pSRC27-H also lacks it. This structure is

39

present in the same position in R27 and type 1 plasmids, but in the opposite orientation, and

40

appears to have been incorporated via IS1-mediated transposition. The PCRs developed provide a

41

simple means of distinguishing type 1 and type 2 IncHI1 plasmids based on the presence or

42

absence of variable regions.

43

44

Key words

45

IncHI1 plasmids

46

Plasmid typing

47

Resistance plasmids

48

Plasmid evolution

49

Cryptic transposon

50

1. Introduction

51

Tracking plasmids that carry antibiotic resistance genes in Gram-negative bacteria is complicated

52

by the fact that there are so many plasmid types, and many variations occur within each type.

53

Furthermore, it is well established that antibiotic resistance, which increasingly limits antibiotic

54

therapy, often arises by incorporation of transposons or other mobile elements carrying antibiotic

55

resistance genes into plasmids. This process has the potential to further increase diversity within

56

groups of related plasmids. However, recently, the locations of transposons and other features

57

have been found to delineate specific lineages within plasmid groups, for example in the IncP1α

58

(Haines et al., 2007; Pinyon and Hall, 2011), IncW (Revilla et al., 2008) and IncHI2 (Cain and

59

Hall, 2012b) plasmid families.

60

IncHI1 plasmids are known to play a role in the acquisition of antibiotic resistance in

61

Salmonella enterica serovars Typhi and Paratyphi A which cause life threatening infections in

62

much of the world (Holt et al., 2011; Phan et al., 2009; Wain et al., 2003). A comparison of three

63

sequenced IncHI1 plasmids, R27, recovered from Salmonella serovar Typhimurium (Sherburne

64

et al., 2000), pHCM1, recovered from Salmonella serovar Typhi (Wain et al., 2003), and

65

pAKU_1, recovered from serovar Paratyphi A (Holt et al., 2007), revealed that they share a

66

common backbone of 164.4 kb that varies only slightly (0.01-0.03% of bases differ) (Holt et al.,

67

2007). Further information on these plasmids is listed in Table 1. However, a number of DNA

68

segments that are present in pHCM1 but not in R27 or pAKU_1 (A-E in Figure 1) have been

69

identified (Holt et al., 2007; Phan et al., 2009; Wain et al., 2003) and these may reflect two

70

separate evolutionary pathways that involved either loss or gain of these segments.

71 72

Transposons carrying antibiotic resistance genes have been incorporated at different positions in the backbone of the three IncHI1 plasmids shown in Figure 1. Though some

73

resistance genes or transposons are shared, they have been acquired via different events and may

74

also have different contexts. For example, both R27 and pAKU_1 have independently acquired

75

the transposon Tn10 carrying the tet(B) tetracycline resistance determinant (Figure 1), whereas

76

pHCM1 includes only part of Tn10 as part of the multiple antibiotic resistance region (MARR) in

77

yet another position. Plasmid pHCM1 shares additional resistance genes with pAKU_1, and both

78

carry a large insertion that includes several antibiotic resistance genes designated MARR in

79

Figure 1. In pAKU_1, the MARR is a transposon that is bounded by directly-oriented copies of

80

IS1 and derived from Tn2670. However, two consecutive inversion events resulting from

81

recombination between regions of identity have subsequently split this MARR into two parts

82

(Holt et al., 2007). The precursor of pAKU_1 with the inversions notionally reversed to show the

83

original insertion point of the transposon is shown in Figure 1. The pHCM1 MARR also includes

84

a region of 23.9 kb that is bounded by directly-oriented copies of IS1 and derived from Tn2670,

85

and this transposon is located adjacent to the partial Tn10. pAKU_1 also carries the strA-strB

86

streptomycin resistance gene pair in a region identified as a copy of the transposon Tn5393a

87

(Cain and Hall, 2011), whereas the strA-strB gene pair is also found within the pHCM1 MARR

88

in the transposon Tn6029B (Cain et al., 2010). Finally, in pHCM1, a second transposon related to

89

Tn1696 and carrying two further antibiotic resistance genes is found at a separate location (Cain

90

et al., 2010; Partridge et al., 2001). Hence, the resistance gene complement and their context and

91

location is highly variable in the IncHI1 family.

92

However, in one case the location of the MARR does appear to be conserved. The MARR

93

from an IncHI1 plasmid pSRC27-H, isolated from a S. enterica serovar Typhimurium of equine

94

origin, was shown recently to be related to the pHCM1 MARR, despite significant differences in

95

the resistance gene content (Cain and Hall, 2012a). In the same study, the MARR of two

96

additional completely sequenced IncHI1 plasmids, pMAK1 and pO111_1 (see Table 1), were

97

analysed and found to be further relatives. These MARR were all found in the same location after

98

an inversion in pHCM1, that has occurred after the incorporation of a second copy of IS10 into

99

pHCM1 (Figure 2a), was reversed (Cain and Hall, 2012a). These MARR are part of region A in

100 101

Figure 1. Here, we have further analysed region A and used PCR to look in pSRC27-H for the

102

additional regions, B-E (in Figure 1) that distinguish plasmid pHCM1 from R27 and pAKU_1.

103

The sequences of IncHI1 plasmids pMAK1 and pO111_1 (Table 1) that were reported without

104

analysis of their structures were also searched for B-E. The structure of these additional features

105

and two further potential mobile elements was also examined.

106 107

2. Materials and Methods

108

2.1 Plasmids

109

Plasmids examined are listed in Table 1. pSRC27-H, an IncHI1 plasmid conferring resistance to

110

several antibiotics was described previously (Cain and Hall, 2012a), and the sequence of the

111

MARR and flanking segments is found in GenBank accession number HQ840942. Plasmid R27

112

(see Table 1) was used as a control in all PCR experiments.

113 114

2.2 Plasmid analysis

115

Plasmid DNA was extracted using an alkaline lysis miniprep method optimized for the extraction

116

of large plasmids as described previously (Cain and Hall, 2012b; Cain et al., 2010). Polymerase

117

chain reactions (PCR) were performed as previously described (Cain et al., 2010) using primers

118

designed using GenBank accession numbers AL513383 and AF250878 and listed in Tables 2 and

119

3. Primers were designed to enable all annealing steps to be carried out at 60oC. Briefly, PCR

120

reactions contained 200 µM of each dNTP, 1 U Taq Polymerase and 50 pmol of primer in 25 µ l

121

of 1 x PCR reaction buffer (supplied with Taq polymerase, New England BioLabs). Reactions

122

were run in an Eppendorf Thermocycler (Eppendorf) with an initial incubation at 95°C for 5 min,

123

followed by 35 cycles of 95°C for 30 sec, 60°C for 30 sec, and 72°C for 1-3 mins, then a final

124

incubation at 72°C for 5 min. DNA fragments were resolved by gel electrophoresis on 1% (w/v)

125

agarose gel, stained with ethidium bromide (5 µg/ml) and visualised using a GelDoc1000 image

126

analysis station (BioRad, Hercules, CA). Amplicons were identified using molecular weight

127

standards and identities confirmed either by digestion with restriction enzymes, as per the

128

manufacturer’s instructions (New England BioLabs) or by DNA sequencing.

129 130

2.3 Sequencing and Sequence Analysis

131

Sequencing using purified PCR products as substrates and sequence analysis was performed as

132

described previously (Cain and Hall, 2011; Cain et al., 2010). Bioinformatic analysis was

133

undertaken with all plasmids for which sequences were available in the GenBank non redundant

134

DNA database (accession numbers are listed in Table 1). A further IncHI1 plasmid that has been

135

reported to be prevalent in recent S. Typhi isolates (Holt et al., 2011) was not analysed because

136

the sequence is not available in the GenBank non redundant DNA database.

137 138

3. Results

139

3.1 Region A and the context of the MARR

140

The region designated A in Figure 1 includes the MARR. However, there is a difference between

141

pO111_1, pMAK1 and pSRC27-H on the one hand and pHCM1 on the other caused by an

142

inversion in pHCM1 (Figure 2a) that alters the orientation but not the general structure of the

143

region. A set of PCRs that establish this location for MARR in the more common configuration

144

for this group and the configuration in pHCM1 are shown in Table 2.

145

A more detailed analysis of the sequences adjacent to the MARR revealed differences

146

between pHCM1 and R27 that have not been noted previously (Figure 2b). The MARR and a

147

region of 4 kb in pHCM1 (149,592-153,534), pO111_1 and pMAK1 found adjacent to the left

148

hand IS1 of the MARR (IS1a in Figure 2) is not found in R27 or pAKU_1. The MARR and this

149

4kb are replaced in R27 and pAKU_1 by a different 7.5 kb segment (bp 223,127-230,610 in R27)

150

carrying citrate utilization genes, citA and citB (Figure 2b). Both regions encode a potential

151

transposase (tnp1 and tnp2 in Figure 3b), and Tnp1 is 72% identical to Tnp2. However, inverted

152

repeats marking potential transposon boundaries were not found in either case.

153

Part of the 4 kb segment was present in the sequenced region from pSRC27-H (GenBank

154

accession number HQ840942). PCRs using a primer RH1272 (see Figure 2b and Table 3) in the

155

sequence shared by all the IncHI1 plasmids with RH866 in the 4 kb segment produced a 1.1 kb

156

amplicon from pSRC27-H DNA, indicating that the 4 kb segment was complete. In contrast, R27

157

which carries the 7.5 kb segment produced a 1.9 kb amplicon using primer RH1272 with RH867

158

(see Figure 2b and Table 3).

159 160

3.2. Segment B in pSRC27-H, pO111_1 and pMAK1

161

The additional region B shown in Figure 1 which is known to be present in pHCM1 but not in

162

R27 or pAKU_1, was also examined in more detail. The structure of region B is shown in Figure

163

3a. The 14.7 kb segment defined by comparison between pHCM1 and R27 sequences and

164

previously classed as region B in pHCM1 (Phan et al., 2009) was found to be comprised of two

165

parts, only one of which (bp 113,599-114,496 and 127,197-128,349) was found in the pO111_1

166

and pMAK1 sequences. In pHCM1, a derivative of transposon Tn1696 carrying the dfrA14

167

trimethoprim resistance gene cassette and a Tn1696/Tn5036 mercury resistance module (see

168

(Cain et al., 2010; Partridge et al., 2001) for the structure) has been inserted into this segment and

169

is surrounded by a 5 bp duplication. The core segment of 2,051 bp is flanked by a 3 bp

170

duplication but inverted repeats were not found at the boundaries. It contains 2 open reading

171

frames (ORF) of unknown function. For pSRC27-H, PCR using primers RH857 and RH858

172

yielded an amplicon of 2.3 kb (Table 3), indicating that only the core segment was present. R27

173

yielded a smaller product indicating the absence of region B.

174 175 176

3.3. Segments C, D and E in pSRC27-H, pO111_1 and pMAK1 Region C consists of 2 short segments (885 and 1,418 bp) located 193 bp apart at 7,962-

177

8,846 and 9,040-10,457 in pHCM1. However, no feature that would indicate whether they are

178

insertions or deletions was found. Searches revealed that region C was present in pO111_1 and

179

pMAK1. It was also detected in pSRC27-H but not in R27, using primers RH1278 and RH616

180

listed in Table 3.

181

Segment D, located at bp 77,104-81,204 in pHCM1, is 4,101 bp long and is located in the

182

trh2 transfer region between trhN and trhI (Figure 3b). It is bounded by imperfect inverted

183

repeats of 27 bp (24/27 matches) and flanked by an 8 bp duplication, suggesting that it is a

184

transposon, here designated TnD. This conclusion is supported by the fact that a region that is

185

99.8% identical to segment D is found in a different context, but also flanked by an 8 bp

186

duplication in the IncHI2 plasmids, R478, pK29, and pEC-IMP (GenBank accession numbers

187

BX664015, EF382672, EU855787). TnD includes 3 open reading frames (ORF), two of

188

unknown function. However, the product of ORF2 is related to DNA helicases and may be

189

involved in movement. TnD was found in the same location in pO111_1 and pMAK1, but in

190

pO111_1, it is interrupted by an insertion sequence, IS629, at the position indicated by the

191

vertical arrow in Figure 3b. TnD was also shown to be present in pSRC27-H using PCR primers

192

listed in Table 3, while R27 yielded the shorter product indicating its absence.

193

A short segment (bp 141,604-142,292 in pHCM1) located 289 bp to the right of the IS10

194

end of the MARR (as shown in Figure 2b), previously designated E (Phan et al., 2009) (see

195

Figure 1), was found in pO111_1, pMAK1 and in the sequenced region from pSRC27-H

196

(GenBank accession number HQ840942) but not in R27 or pAKU_1. pAKU_1 has a further 3.7

197

kb deletion on the left. Segment E thus appears to be characteristic of pHCM1-related plasmids.

198 199

3.4 A fragment of the E. coli chromosome in IncHI1 plasmids

200

A further difference was identified in the original three sequenced IncHI1 plasmids (Holt et al.,

201

2007). Its location is shown by the position of an IS1 close to the left end in Figure 1. In this case,

202

a segment that is present in both R27 and pAKU_1 is the same position in pHCM1 but in the

203

opposite orientation (Figure 4). This region was not found in the sequences of pO111_1 or

204

pMAK1, allowing a region made up of a single copy of IS1 (boxed in Figure 1) and an adjacent

205

segment of 1,697 bp to be defined. The adjacent segment exhibits 99.1-99.7% identity to over 40

206

Escherichia coli genomes. The corA gene has been shown to be important in Mg2+ transport

207

(Daniels et al., 1992). The IS and corA gene were found to be flanked by a duplication of 8 bp

208

(Figure 4). An 8 bp duplication is expected for IS1, and this indicates that IS1 is likely to have

209

mobilized DNA adjacent it from the E. coli chromosome. The segment designated IS1-corA in

210

Table 3 was shown to be missing from pSRC27-H using primers RH1279 and RH1280 (Table 3).

211

Thus, this insertion does not seem to be characteristic of plasmids in the pHCM1 group and its

212

presence in pHCM1 remains to be to explained.

213 214

3.5 Fragment of an F plasmid in IncHI1 plasmids

215

The segment marked FIA in Figure 1 represents a transposon consisting of 1644 bp that is 83%

216

identical to the FIA replication region of plasmid F (Saul et al., 1988) flanked by directly-

217

oriented copies of IS1. This transposon is flanked by an 8 bp duplication, but it appears to have

218

been acquired prior to the separation of the pHCM1 and R27 lineages. However, it is not found in

219

the pMAK1 sequence where only a single IS1 remained. Using primers RH1281 and RH1282

220

(Table 3) which flank this transposon, the F-like segment was shown to be present in pSRC27-H.

221 222

4. Discussion

223

4.1 A second type of IncHI1 plasmid

224

This study defines a group of IncHI1 plasmids (pHCM1, pO111_1, pMAK1 and pSRC27-H)

225

that are most closely related to pHCM1. This grouping was predicted previously based on the

226

location and structure of the MARR (Cain and Hall, 2012a). Region A, as originally defined was

227

found to be composed of distinct parts, the MARR and a segment adjacent to IS1a that is not

228

found in R27 and relatives (Table 1). Here, the four plasmids were shown to share the 4 kb

229

segment adjacent to the MARR, which is part of region A. The additional segments B-E

230

previously found to be present in pHCM1 when it was compared to R27 were also present in

231

pO111_1, pMAK1 and pSRC27-H. Hence, these plasmids form a second lineage of IncHI1

232

plasmids, here named type 2 IncHI1 plasmids, that is characterized by carriage of these 5

233

features. However, segment B, as originally defined in pHCM1, also consists of two parts only

234

one of which, core B, is found in the other type 2 plasmids. Core B is interrupted by a transposon

235

carrying an integron in pHCM1, and segment D (TnD) is interrupted by an IS in pO111_1

236

indicating that evolution is ongoing.

237

The extension of the features found in pHCM1 to other type 2 plasmids should simplify

238

future analysis and annotation of IncHI1 plasmid sequences. For example, pP-stx-12 (Table 1)

239

does not contain any of the 5 regions and hence belongs to type 1. Only part of the IncHI1

240

backbone is present in a recently reported IncHI1 plasmid sequence (Dolejska et al., 2013), but

241

we found (data not shown) that some of the type 2-specific features are present, indicating that

242

the segment is derived from a type 2 genome.

243 244

4.2 Mobile elements

245

As plasmids evolve it is to be expected that they will acquire mobile elements. However, only

246

one of the characteristic differences between type 1 and type 2 IncHI1 plasmids, namely segment

247

TnD, can be predicted confidently to be an insertion in type 2. TnD has terminal inverted repeats

248

and has generated of a duplication of the target sequence, both features that are associated with

249

mobile genetic elements. Its mobility is confirmed by that finding that it is also found in IncHI2

250

plasmids also surrounded by a target site duplication. However, the origins of segments B, C and

251

E are less clear and they could represent insertions in type 2 or have been lost from type 1.

252

The insertion of IS1 and the associated corA-containing segment from an E. coli

253

chromosome into the type 1 backbone has also occurred via an IS-mediated transpositional

254

mechanism as evidenced by the 8 bp duplication. The presence of this translocated structure

255

appears to be characteristic of type 1, with its presence in pHCM1 an exception. Moreover, the

256

fact that this structure is in the identical location but in the inverse orientation in pHCM1, is

257

difficult to explain.

258 259

4.3 Features of type 1 plasmids

260

As further sequences of IncHI1 plasmids become available, it will be of interest to see if the IS1-

261

corA segment continues to be largely confined to type 1 plasmids. The type 1 plasmids also some

262

share specific features. Region A is substituted by a different segment carrying citrate utilization

263

genes the presence of which can be detected by PCR, and which were previously shown to confer

264

the ability to grow on citrate (Sherburne et al., 2000). An additional 534 bp found in R27 and

265

pAKU_1 but not in pHCM1 (Holt et al., 2007) may also serve as a useful feature distinguishing

266

the type 1 from type 2 as we did not find it in the pO111_1 or pMAK1 sequences. Whether it is

267

in pSRC27-H was not tested.

268 269

4.4 Distinguishing type 1 and type 2 plasmids

270

A plasmid multilocus sequence typing (pMLST) scheme was developed for IncHI1 plasmids

271

(Phan et al., 2009) and to date 8 PST (plasmid sequence types have been identified (Holt et al.,

272

2011). pHCM1, pO111_1 and pMAK1 all PST1 but R27 and pAKU_1 are PST5 and PST7,

273

respectively. Single nucleotide polymorphisms have also been used to group IncHI1 plasmids.

274

(Holt et al., 2011). Sequencing for pMLST schemes can be expensive and complete sequences of

275

plasmids cannot always be obtained in the high numbers needed for epidemiological

276

investigations and simpler tools are needed for that purpose. The set of PCR assays developed

277

here to detect variable regions using the difference in size of PCR amplicons to distinguish the

278

two branches of the IncHI1 plasmid family should help to simplify the classification of IncHI1s

279

into these two groups, facilitating studies of the epidemiology of IncHI1 plasmids. Variable

280

region typing and determining the location of antibiotic resistance transposons, in particular the

281

MARR of the type 2 plasmids (Table 3), adds to the methods being developed to type related

282

plasmids such as restriction mapping, plasmid multilocus sequence typing (pMLST) and long

283

range PCRs or hybridization to detect specific regions in the backbone that have all been applied

284

to the IncHI1 family (Phan et al., 2009).

285 286

5. Conclusions

287

We have defined two types of IncHI1 plasmid based on the presence or absence of specific

288

regions, and developed a simple method for determining if they are present. This approach to

289

plasmid typing of using PCR to track insertions and or deletions that are characteristic of specific

290

lineages within a closely related group of plasmids provides a simple tool that has the potential to

291

simplify and enhance epidemiological investigations of plasmids belonging to commonly

292

encountered plasmid types.

293 294

Funding

295

A.K.C. was supported by a University of Sydney Postgraduate Award and R.M.H. was partly

296

supported by NHMRC grant 358713. The project was partly supported by NHMRC project grant

297

402584.

298 299

Transparency declarations

300

None to declare

301

302

References

303

Cain, A. K., Hall, R. M., 2011. Transposon Tn5393e carrying an aphA1-containing transposon

304

upstream of strAB does not confer resistance to streptomycin. Microb. Drug Resist. 17,

305

389-394.

306 307 308 309 310

Cain, A. K., Hall, R. M., 2012a. Evolution of a multiple antibiotic resistance region in IncHI1 plasmids: reshaping resistance regions in situ. J. Antimicrob. Chemother. 67, 2848-2853. Cain, A. K., Hall, R. M., 2012b. Evolution of IncHI2 plasmids via acquisition of transposons carrying antibiotic resistance determinants. J. Antimicrob. Chemother. 67, 1121-1127. Cain, A. K., et al., 2010. Transposons related to Tn1696 in IncHI2 plasmids in multiply antibiotic

311

resistant Salmonella enterica serovar Typhimurium from Australian animals Microb.

312

Drug Resist. 16, 197-202.

313 314 315

Daniels, D. L., et al., 1992. Analysis of the Escherichia coli genome: DNA sequence of the region from 84.5 to 86.5 minutes. Science. 257, 771-778. Dolejska, M., et al., 2013. Complete sequencing of an IncHI1 plasmid encoding the

316

carbapenemase NDM-1, the ArmA 16S RNA methylase and a resistance-nodulation-cell

317

division/multidrug efflux pump. J. Antimicrob. Chemother. 68, 34-39.

318 319 320 321

Haines, A. S., et al., 2007. Sequence of plasmid pBS228 and reconstruction of the IncP-1alpha phylogeny. Plasmid. 58, 76-83. Holt, K. E., et al., 2011. Emergence of a globally dominant IncHI1 plasmid type associated with multiple drug resistant typhoid. PLoS Negl. Trop. Dis. 5, e1245.

322

Holt, K. E., et al., 2007. Multidrug-resistant Salmonella enterica serovar paratyphi A harbors

323

IncHI1 plasmids similar to those found in serovar typhi. J. Bacteriol. 189, 4257-4264.

324 325

Partridge, S. R., et al., 2001. Family of class 1 integrons related to In4 from Tn1696. Antimicrob Agents Chemother. 45, 3014-3020.

326

Phan, M. D., et al., 2009. Variation in Salmonella enterica serovar typhi IncHI1 plasmids during

327

the global spread of resistant typhoid fever. Antimicrob. Agents Chemother. 53, 716-727.

328 329 330 331 332 333

Pinyon, J. L., Hall, R. M., 2011. Evolution of IncP-1α plasmids by acquisition of antibiotic and mercuric ion resistance transposons. Microb. Drug Resist. . 17, 339-343. Revilla, C., et al., 2008. Different pathways to acquiring resistance genes illustrated by the recent evolution of IncW plasmids. Antimicrob. Agents Chemother. 52, 1472-1480. Saul, D., et al., 1988. A replication region of the IncHI plasmid, R27, is highly homologous with the RepFIA replicon of F. Mol. Microbiol. 2, 219-225.

334

Sherburne, C. K., et al., 2000. The complete DNA sequence and analysis of R27, a large IncHI

335

plasmid from Salmonella typhi that is temperature sensitive for transfer. Nucleic Acids

336

Res. 28, 2177-2186.

337

Wain, J., et al., 2003. Molecular analysis of IncHI1 antimicrobial resistance plasmids from

338

Salmonella serovar Typhi strains associated with typhoid fever. Antimicrob Agents

339

Chemother. 47, 2732-2739.

340

341

Figure legends

342

Figure 1. Schematic alignment of IncHI1 plasmids showing locations of resistance regions and

343

other key points of difference. Drawn to scale based on sequences under the GenBank accession

344

numbers AL513383 for pHCM1, AF250878 for R27 and AM412236 for pAKU_1 but to

345

facilitate alignment the inversions that have occurred in pAKU_1 have been reversed. The

346

location of the three replication initiation regions, HIA, HIB and FIA are indicated above. The

347

location of the segment in the par partition locus that is used in plasmid based replicon typing is

348

indicated by an asterisk, and the additional regions A to E in pHCM1 identified previously (Phan

349

et al., 2009) are marked above. The vertical arrow indicates the position of an additional 434 bp

350

segment present in R27 and pAKU_1 identified previously (Holt et al., 2007). The locations of

351

transposons and multiple antibiotic resistance regions (MARR) are also indicated above. Copies

352

of IS1 are shown as open boxes. Genes or regions involved in replication repA, partition, par,

353

conjugal transfer, trh1 and trh2, and htd are indicated.

354 355 356

Figure 2. Schematic showing the detail of the region surrounding the type 2 MARR.

357

(a) Origin of the inversion in pHCM1. The area surrounding the MARR in pO111_1, pMAK1

358

and pSRC27-H is shown in the top line and the area surrounding the MARR in pHCM1 in the

359

bottom line. An intermediate configuration with an additional IS10 is shown in between. Black

360

flags represent the 9 bp repeats generated by the IS10 insertion. The open boxes represent IS with

361

the IS number inside and the orientation indicated be an arrow below. Only the tetA and tetR

362

genes are shown via horizontal arrows below and the asterisk indicates the position of a 356 bp

363

deletion in pSRC27-H. The parts of the MARRs contained between the 2 IS1s is not shown. The

364

hatched lines represent the part of the IncHI1 backbone located to the left of the MARR in

365

pSRC27-H and the thick line represents the backbone on the right.

366

(b) Comparison of the region surrounding the pO111_1, pMAK1 and pSRC27-H MARRs

367

compared the equivalent region in R27 and pAKU_1. Symbols are as in (a) except that only the

368

IS1a and IS10 at the outer ends of the MARR are shown, and the stippled line represents the

369

region unique to R27 and pAKU_1. The location of region E is shown via the deletion in R27

370

and pAKU_1. An additional deletion in pAKU_1 is also shown. Diagnostic PCR amplicons are

371

shown below with primers indicated by short vertical lines at each with the primer name and

372

orientation, indicated by an arrowhead. The predicted size is below.

373 374

Figure 3. The structures of regions B and D in type 2 IncHI1 plasmids. The IncHI1 backbone is

375

shown by dashed lines and additional regions by solid lines. Nucleotide sequences of direct

376

repeats flanking additional segments are shown above, and open reading frames are indicated by

377

horizontal arrows below. In (a), the Tn1696-like transposon is represented schematically without

378

detail above. In b), the 27 bp imperfect inverted repeats that bound the insertion are shown as

379

thick vertical lines, and the vertical arrow indicates where a copy of IS629 has inserted in ORF1

380

in pO111_1. PCR primers and amplicons, and the amplicon size, are shown below. Primers are

381

shown as short vertical lines with the primer name and orientation, indicated by an arrowhead,

382

beside it. Drawn to scale using GenBank accession nos AL513383, AF250878, AM412236,

383

AB366440, and AP010961 for pHCM1, R27, pAKU_1, pMAK1 and pO111_1, respectively.

384 385

Figure 4. The structures of the regions containing a segment of the E. coli chromosome. IS1 is

386

shown as a box with the orientation of the reading frames indicated by an arrow below. The

387

vertical arrow marks the position of a 33 bp deletion in IS1 in pHCM1. The nucleotide sequence

388

of direct repeats flanking the additional segments are shown above. Drawn to scale using

389

GenBank accession nos AL513383, AF250878, AM412236, AB366440, and AP010961 for

390

pHCM1, R27, pAKU_1, pMAK1 and pO111_1, respectively.

391

Figure 1

Figure 2

Figure 3

Figure 4

392

Table 1

393

IncHI1 plasmids

394 Plasmid

Organism

Year

Country

GenBank accession number

Size bp

Antibiotic resistance genes

Type 1 R27 pAKU_1

S. Typhimurium S. Paratyphi A

1961 2002

UK Pakistan

AF250878 AM412236

180,461 212,711

pP-stx-12

S. Typhi

NKa

India

CP003279

181,431

tetA(B) blaTEM, catA1, dfrA7, strA-strB, sul1, sul2, tetA(B) tetA(B)

Type 2 pHCM1

S. Typhi

1993

Vietnam

AL513383

218,160

pO111_1

E. coli

2001

Japan

AP010961

204,604

pMAK1

S. Choleraesuis

NKa

NKa

AB366440

208,409

pSRC27H

S. Typhimurium

1999

Australia

HQ840942

NDb

395

a

Not known

396

b

Only the sequence of the MARR and flanking regions is available

397 398

blaTEM, catA1, dfrA14, strA-strB, sul1, sul2, tetA(B) aphA1b, blaTEM, catA1, strA-strB, sul2, tetA(B) aadA2, blaTEM, catA1, dfrA12, mphA, mrx, strAstrB, sul1, sul2, tetA(B) aacC2, aadA2, aphA1b, blaTEM, catA1, dfrA12, strAB, sul2, tetA(B)

399

Table 2

400

Primers targeting the location of the MARRa in type 2 IncHI1 plasmids Linkage LHS

b

Primer RH624 RH621 RHS as in pSRC27-H RH627 RH612 RHS as in pHCM1 RH628 RH612 extra IS10c RH628 RH1262

Sequence CAGAGGGCAACAGCGAAGTG GCGAATTGAGCGGCATAACC CGATCATGCTCAATTGCATAC TGCTGAGCACTGAGAGATCC CTGTTCATGCCAGCGATAAG TGCTGAGCACTGAGAGATCC CTGTTCATGCCAGCGATAAG TGACCAGGTCCCCCATAAAC

Size (bp) 466 1,593 1,602 1,814

401

a

MARR extends from IS1a to IS10 (see Figure 2).

402

b

RHS (right hand side) and LHS (left hand side) are as indicated for pSRC27-H in Figure 2, and

403

PCRs span these IS..

404 405

c

the extra IS10 is on the right as pHCM1 is shown in Figure 2a

406

Table 3

407

Primers targeting the locations of IncHI1 variable regions Backbone variable a regions

Primer

Sequence (5'-3')

A1b

RH1272 RH867 RH1272 RH866 RH857 RH858 RH1278 RH616 RH1276 RH1277 RH1279

ACCAGCAGACTGGCGAACAT TCCATCTTGGCGTGGTTATG ACCAGCAGACTGGCGAACAT TGTGGCACAATGGATTACCC CCTGTTAGTCACCGGCTTGC CCCGGCAATTGACAAACAC AATGGTCCGGGCAATTCTG TCCGCAATAATATGCTCTGG CGCTCAGGCCATTAAAGAGG TGATAGTCCAAAGTTCACCATGC CGCCAAGGAGTCGCTTAATG

A2b Bc C c

D

IS1-corA

f

IS1-FIA-IS1

RH1280 AGCGCCTGACTGAACAGGAG RH1281 TTCCGCTCCATGGTGGTAAC RH1282 CATGAGCTCCGGGTCACTTC

408

a

409

b

Primer positions are shown in Figure 2.

410

c

Primer positions are shown in Figure 3.

411

d

Much longer in pHCM1.

412

e

Longer in pO111_1.

413

f

Primer positions are shown in Figure 4.

regions A-D and FIA are indicated in Figure 1.

Size (bp) with insertion 1,909

Size (bp) insertion without found in insertion NA type 1

1,088

NA

type 2

2,354

300

type 2d

3,027

1,004

type 2

4,378e

269

type 2

3,120

679

type 1 and pHCM1

3,354

942

type 1 and 2

414 415 416 417 418 419 420 421

* Two groups of IncHI1 plasmids were distinguished by sequences present only in type 2 * Only 1 of the definitive sequences appeared to be a transposon * PCR assays to detect the presence of definitive sequences were developed * An IncHI1 plasmid from S. Typhimurium was demonstrated to belong to type 2