Microbiology Papers in Press. Published February 14, 2014 as doi:10.1099/mic.0.077057-0

1

Identification of Burkholderia multivorans ATCC 17616 genetic determinants for

2

fitness in soil by using signature-tagged mutagenesis

3 4

Yuji Nagata,* Junko Senbongi, Yoko Ishibashi, Rie Sudo, Masatoshi Miyakoshi,#

5

Yoshiyuki Ohtsubo, and Masataka Tsuda

6

Department of Environmental Life Sciences, Graduate School of Life Sciences, Tohoku

7

University, Sendai 980-8577, Japan

8 9

* Address correspondence to Yuji Nagata, [email protected]

10

#

11

University of Würzburg, Josef-Schneider-Strasse 2/Bau D15, 97080 Würzburg, Germany

Present address: Masatoshi Miyakoshi, Institute for Molecular Infection Biology,

12 13

Running title: Burkholderia genes important for fitness in soil

14 15

Keywords: Burkholderia; soil environment; signature-tagged mutagenesis (STM); fur.

16 17

Contents Category: Environmental and Evolutionary Microbiology

18 19

This manuscript consists of 23 text pages (4,409 words), 2 Tables, 3 Figures, and

20

Supplemental Materials (2 tables and 3 figures).

21 22

1

23

SUMMARY

24

To identify bacterial genetic determinants for fitness in a soil environment,

25

signature-tagged mutagenesis (STM) was applied to a soil bacterium, Burkholderia

26

multivorans ATCC 17616. This strain was randomly mutagenized by each of 36 different

27

signature-tagged plasposon, and 36 mutants with different tags were grouped as a set. A

28

total of 192 sets consisting of 6,912 independent mutants were each inoculated into soil

29

and incubated. Two-step STM screening based on quantitative real-time PCR of total

30

DNAs extracted from the resulting soil samples using the tag-specific primers led to the

31

selection of 39 mutant candidates that exhibited a reduction in relative competitive fitness

32

during incubation in the soil, and 32 plasposon-insertion sites were determined. Among

33

them, mutants having plasposon insertion in fur, deaD, or hrpA exhibited reduced fitness

34

during incubation in soil when compared with the control strain. The deficiency in the soil

35

fitness of the fur mutant was recovered by the introduction of the wild-type fur gene,

36

indicating that the fur gene is one of the genetic determinants for fitness in the soil.

37

2

38 39

INTRODUCTION Studies employing the pure cultivation of bacteria in laboratory media have helped

40

to reveal a number of ubiquitous principles of life in addition to the functions unique to

41

individual bacterial strains. However, it is now strongly suggested that the bacterial

42

activities in natural environments are quite different from those under the laboratory

43

conditions (van Veen et al., 1997; Rediers et al., 2005). In fact, we are now encountering

44

problems with the practical use of bacteria in complex natural environments (e.g.,

45

bioremediation and biological control of plant diseases), since the bacteria do not always

46

perform their expected functions that can be observed under the laboratory conditions (van

47

Veen et al., 1997; Ramos et al., 2011; Tyagi et al., 2011).

48

More information on the bacterial genetic determinants for fitness in the natural

49

environments will enable us to control the bacterial activities more effectively. For this

50

purpose, two types of genetic approaches, in vivo expression technology (IVET) and

51

signature-tagged mutagenesis (STM), have been developed to identify genes or loci that

52

specifically function under natural environments (e.g., surface and internal parts of plants

53

and animals and the plant rhizosphere) (Handfield and Levesque, 1999). IVET can identify

54

genes specifically expressed under a particular condition (Rediers et al., 2005), while STM

55

can identify the genes or loci important for growth and/or survival under a specific

56

condition (Mazurkiewicz et al., 2006). Many studies using STM employ a set of

57

transposons, each containing a tag signature with a unique oligonucleotide sequence. When

58

a set of mutants having insertions with differently tagged transposons is passed en masse

59

through a selected condition, individual mutants can be distinguished from the others by

60

their unique tags, thus allowing it possible to monitor changes in the relative abundance of

61

individual mutants during the selection procedure (Mazurkiewicz et al., 2006). In the 3

62

original STM, the tags of mutants were detected by Southern hybridization (Hensel et al.,

63

1995). A modified STM technique was later developed using quantitative real-time PCR

64

(qRT-PCR) to detect the different transposon mutants, with the advantage that the

65

qRT-PCR-based procedure can reduce the occurrence of false positives that often result

66

from cross-hybridization between different tags (Hunt et al., 2004). Although STM is a

67

powerful tool that has been widely applied to bacterial pathogens to define their various

68

virulence genes, its application to unravel the ecologically important and/or useful traits of

69

bacteria residing in the natural environments is still limited, and only a few studies using

70

hybridization-based STM have been reported—namely, studies identifying: Burkholderia

71

vietnamiensis G4 genes important for the rhizosphere colonization (O'Sullivan et al., 2007)

72

and Desulfovibrio desulfuricans and Shewanella oneidensis genes important for the fitness

73

in anoxic aquifer sediment (Groh et al., 2005, 2007; Luo et al., 2007).

74

B. multivorans ATCC 17616 belongs to β-proteobacteria and is a strain that was

75

isolated from a soil sample after anthranilate enrichment, and is capable of assimilating a

76

wide range of compounds (Stanier et al., 1966). Our previous application of an IVET

77

system to ATCC 17616 led to the successful identification of 116 loci that were expressed

78

in soil, but not in a minimal laboratory medium (Nishiyama et al., 2010). Our subsequent

79

analysis of one such locus, designated mls (B. multivorans ATCC 17616 loci induced in

80

soil environment), demonstrated that the expression of anthranilate dioxygenase genes,

81

which were the most repeatedly identified in the IVET screening and were drastically

82

induced in the soil (Nishiyama et al., 2010), plays a pivotal role in the proliferation in the

83

sterilized soil (Nishiyama et al., 2012).

84 85

In this study, we applied a qRT-PCR-based STM system to ATCC 17616, the same bacterial strain used in our previous IVET analysis (Nishiyama et al., 2010), in order to 4

86

identify the genetic determinants for fitness in soil. This is the first report in which both

87

IVET and STM systems were utilized with the same bacterial strain in the same natural

88

environment. Since the principles of the two systems are different from each other, use of

89

the two systems was expected to provide greater insight into the fitness of ATCC 17616 in

90

soil.

91 92

METHODS

93

Bacterial strains, plasmids, and culture conditions. The bacterial strains and plasmids

94

used in this study are listed in Table 1. Escherichia coli cells were grown at 37oC in Luria

95

Bertani (LB) broth (Maniatis et al., 1982) and B. multivorans cells at 30oC in 1/3LB (3.3 g

96

of Bacto tryptone, 1.7 g of yeast extracts, and 5 g of sodium chloride per liter). Antibiotics

97

were used at the final concentrations of 100 µg ml-1 for trimethoprim (Tp) and 20 µg ml-1

98

for tetracycline (Tc). The solid media were prepared by the addition of 1.5% agar.

99

DNA manipulations and DNA sequence analysis. Established methods were employed

100

for the preparation of plasmids and genomic DNAs, their digestion with restriction

101

endonucleases, ligation, and agarose gel electrophoresis, and the transformation of E. coli

102

cells (Sambrook et al., 1989). PCR for cloning was performed with KOD-Plus DNA

103

polymerase (TOYOBO, Osaka, Japan). The primers used are listed in Table S1. The

104

nucleotide sequences were determined using an ABI PRISM 3130xl sequencer and ABI

105

Prism Big Dye Terminator Kit (Applied Biosystems). The nucleotide and protein

106

sequences were analyzed using the Genetyx program version 13 (SDC Inc., Tokyo, Japan).

107

Homology searches were performed using the BLAST programs available at the National

108

Center for Biotechnology Information website (http://www.ncbi.nlm.nih.gov/BLAST/).

109

Some of the computational analyses, e.g., comparison of DNA sequences, design of PCR 5

110

primers, and drawing the gene organization figures, were performed by using the

111

GenomeMatcher applications (Ohtsubo et al., 2008), which are available at

112

http://www.ige.tohoku.ac.jp/joho/gmProject/gmhome.html.

113

Construction of signature-tagged plasposons and plasposon mutagenesis. A plasposon,

114

pTnMod-OTp', that carries pMB1-based replication machinery in a Tn5-derived

115

mini-transposon (Dennis and Zylstra, 1998) and 37 of the 41 tags designed by Hunt et al.

116

(2004) were used for the construction of our signature-tagged plasposons,

117

pTnMod-OTp’-tagn. Two oligonucleotides containing, respectively, the tag sequence

118

(Tagn) and its complementary sequence (Ctagn) (Table S1) were annealed and inserted

119

between the SalI and SfiI sites of pTnMod-OTp' to obtain pTnMod-OTp’-tagn. The

120

resultant tagged plasposons were essentially identical to those constructed by Hunt et al

121

(2004). Each of the 37 pTnMod-OTp’-tagn plasposons was introduced independently by

122

electroporation into the ATCC 17616 cells as described previously (Ohtsubo et al., 2006),

123

and the pTnMod-OTp’-tagn-mutagenized Tpr clones were selected on the 1/3LB agar

124

plates containing Tp.

125

Incubation of bacterial cells in soil. The soil sampled at the Ehime Research Institute of

126

Agriculture, Forestry, and Fisheries (formerly the Ehime Agricultural Experiment Station)

127

(Matsuyama, Japan) was sieved and used for the inoculation of B. multivorans cells.

128

Sterilization of soil was performed by use of an autoclave according to the protocol

129

described previously (Nishiyama et al., 2010). The physicochemical properties of the soil

130

before autoclaving have been described by Wang et al (2008). The 36 mutants of ATCC

131

17616, each of which contained a different tag, were pooled as a set, and 192 independent

132

sets consisting of 6,912 mutants were prepared. Cells from each set grown to the stationary

133

phase in 1/3LB containing Tp were harvested and washed three times with M9 medium 6

134

(Komatsu et al., 2003). Washed cells resuspended in M9 medium (adjusted to

135

approximately 106 cells in 300 µl) were inoculated into 2.7 g of a soil sample in a 50-ml

136

sterilized test tube. The water content was adjusted to 60% of the maximum water-holding

137

capacity. After incubation at 30oC for 14 days, DNA was directly extracted from the soil

138

by a Power Soil DNA Isolation Kit (MO BIO Laboratories Inc., Carlsbad, CA) and used as

139

the output pool. DNA from the cells just before the inoculation into the soil was extracted

140

by using a genomic DNA purification kit (Bio-Rad, Hercules, CA) and used as the input

141

pool. For the analysis of reconstructed mutants, each strain was singly inoculated into the

142

non-sterilized soil in a 50-ml sterilized test tube, and incubated at 30oC for 30 days, and

143

DNA was directly extracted from the soil.

144

qRT-PCR analysis. For the STM screening, the input pool and corresponding output pool

145

were analyzed by qRT-PCR using 36 primer pairs consisting of 36 unique forward primers

146

(Tagn) and a common reverse primer (STM-LC2) (Table S1). The reaction was performed

147

in a 15-µl solution containing 10 ng of template DNA, 5 µM of each primer, 0.75 µl of

148

dimethylsulfoxide and 7.5 µl of SYBR Premix ExTaq (TaKaRa Bio, Ohtsu, Japan) using a

149

DNA Engine Opticon 2 System (Bio-Rad). The cycling conditions consisted of an initial 1

150

min at 94oC; followed by 35 cycles at 94oC for 30 s, 60oC for 30 s, and 72oC for 15 s; and a

151

final 10 min at 72oC. The reaction mixture was then measured to determine the melting

152

temperature of amplified double-stranded DNA. A mutant, whose tag-specific amplicon

153

was detected at a similar PCR cycle number with the other different tagged mutants in the

154

input pool but at a much later PCR cycle number than the other different tagged mutants in

155

the output pool, was selected as a candidate whose relative abundance was drastically

156

reduced during the incubation. The exact cycle threshold (CT) value used for identifying a

157

clone with reduced abundance varied depending upon each independent trial. For assay of 7

158

the reconstructed mutants, the relative abundance of DNA from each mutant or

159

complemented strain relative to that from the control strain was estimated by qRT-PCR

160

using DNA extracted from each soil sample as the template. For the complementation

161

analysis, the common STM-5' primer (Table S1) was used instead of each Tagn primer.

162

Determination of plasposon-insertion sites. The DNA sequences that flanked the

163

genomic TnMod-OTp'-tagn insertion site were determined by direct sequencing using the

164

genomic DNA and pTnMod-OTpF or pTnMod-OTpR primers (Table S1) as described

165

previously (Nishiyama et al., 2010).

166

Reconstruction of the sun, fur, tig, deaD, and hrpA mutants of ATCC 17616. Genomic

167

DNAs prepared from the original STM mutants with plasposon insertions in the sun, fur,

168

tig, deaD, and hrpA genes (2F9-22, 2B6-12, 2G3-7, 2C9-19 and 1A5-6, respectively) were

169

digested with HindIII (sun) or BamHI (fur, tig, deaD, and hrpA). The digested DNA

170

fragments were self-ligated, and the self-replicating Tpr plasmids were recovered by

171

transformation of E. coli DH5α. The plasmids thus recovered were digested with HindIII

172

(sun) or BamHI (fur, tig, deaD, and hrpA), dephosphorylated, and introduced into the

173

wild-type ATCC 17616 strain by electroporation. The double crossover-mediated

174

homologous recombination of the wild-type sun, fur, tig, deaD, and hrpA genes by the

175

corresponding genes with TnMod-OTp'-tagn insertions in the Tpr transformants was

176

confirmed by PCR, and the reconstructed disruptants were designated ATCC

177

17616sun::Tn, ATCC 17616fur::Tn, ATCC 17616tig::Tn, ATCC 17616deaD::Tn, and

178

ATCC 17616hrpA::Tn, respectively.

179

Construction of the complemented strains of the fur and hrpA mutations. For the

180

integration of a gene into the ATCC 17616 genome, plasmid pYO103 was constructed by

181

modifying pEX18Tc (Hoang et al., 1998), which has the sacB gene for counter selection 8

182

(see Fig. S1 for details). pYO103 can be used so that any DNA fragment in conjunction

183

with the gfp-lacZ cassette (Ohtsubo et al., 2012) is integrated at a specific position [at bp

184

coordinates between 980,094 and 980,095 on chromosome 1 (Chr 1)] in the ATCC 17616

185

genome through double crossover-mediated homologous recombination by using the

186

sacB-based counter-selective procedure (Schweizer, 1992). Each of PCR-amplified DNA

187

fragments containing the ATCC 17616 fur (535,061-535,560 on Chr 1) and hrpA

188

(2,298,654-2,302,842 on Chr 1) genes with their own promoters was cloned into

189

multiple-cloning sites of pYO103, and the resultant plasmids, pYO103-fur and

190

pYO103-hrpA, were introduced into ATCC 17616fur::Tn and ATCC 17616hrpA::Tn,

191

respectively, by triparental mating using HB101(pRK2013) as a helper. Use of the

192

sacB-based counter selection and subsequent PCR-based confirmation of the genomic

193

integration of the fur–gfp–lacZ and hrpA–gfp–lacZ clusters led to the generation of the

194

integrant strains, ATCC 17616fur::Tn(chr1::fur) and ATCC 17616hrpA::Tn(chr1::hrpA),

195

respectively, with concomitant loss of the remaining portion of pYO3.

196

Phenotype analysis of the reconstructed mutants. To assess the antibiotic resistance of

197

the mutants, the cells at log phase cultured in 1/3LB were diluted, and cell suspensions

198

containing 102-107 colony-forming unit (c.f.u.) ml-1 were spotted on 1/3LB agar plates

199

containing the one of the following antibiotics at the final concentration: 100 µg ml-1 for

200

ampicillin, 10 µg ml-1 for kanamycin, 20 µg ml-1 for polymyxin B, 5 µg ml-1 for

201

chloramphenicol (Cm), 10 µg ml-1 for Tc, 30 µg ml-1 for gentamycin (Gm), 30 µg ml-1 for

202

thiamphenicol, 30 µg ml-1 for florfenicol, and 100 µg ml-1 for fusaric acid. The plates were

203

incubated at 30oC for 2 days. The bacterial growth in 1/3LB at low temperature (15oC) was

204

monitored by measuring OD660 values of cultures using a TVS062CA biophotorecorder

205

(Advantec, Tokyo, Japan). 9

206 207

RESULTS

208

Construction of a signature-tagged mutant library. The original versions of the 37

209

signature-tagged plasposons, pTnMod-OTp’-tagn (n=1~41 except for 13, 21, 24, and 30),

210

constructed by Hunt et al. were successfully used for the identification of B. cenocepacia

211

genes that were required for the survival in rat lungs (Hunt et al., 2004). We constructed

212

essentially identical 37 pTnMod-OTp’-tagn plasposons that exhibited only small

213

differences from those created by Hunt et al. (2004) with respect to the insertion site of the

214

tags. Mutagenesis of ATCC 17616 cells by each of our 36 tagged plasposons, except for

215

pTnMod-OTp’-tag36, generated a number of the Tpr mutants. The 36 mutants of ATCC

216

17616, each of which contained a different tag, were pooled as a set, and 192 independent

217

sets consisting of 6,912 mutants were prepared.

218

Screening of mutants deficient in fitness in a soil environment. Each of the 192 sets of

219

mutant cells was inoculated into sterilized and non-sterilized soils. In this initial STM

220

screening, both types of soil were used in order to distinguish the effects of the other

221

microorganisms co-residing in non-sterilized soil. As a pilot screening, a few sets among

222

the 192 sets of the mutant library were incubated for 7, 14, and 30 days, and qRT-PCR

223

analysis was conducted to measure the relative abundance of DNA from each mutant in the

224

inoculum. The 7-day incubated soil sample revealed no difference in the relative

225

abundances of all the strains between the inoculant (input) and soil-incubated (output)

226

samples, while the 14-day incubated soil samples revealed differences in some mutant

227

strains, and such differences were similarly observed at the 30-day incubated soil samples

228

(data not shown). On the basis of this result, the subsequent massive screening of all sets of

229

the mutant library was conducted using 14-day incubation period. By this screening, 50 10

230

and 250 plasposon-inserted mutants were selected from sterilized and non-sterilized soils,

231

respectively, as the candidates that were reduced in their relative abundances after

232

incubation in the soil. Among the candidates, 15 mutants were commonly selected from

233

both sterilized and non-sterilized soils, and a total of 285 (= 50+250-15) mutants were

234

selected as candidates by this initial screening. To eliminate potential false-positive

235

mutants and to identify the mutants exhibiting a clearer phenotype for competitive fitness

236

in the soil, the 285 candidate mutants were redundantly rearranged into 21 sets, each

237

containing 36 mutants with different tags, on the basis that the largest number of 21

238

candidate mutants with tag20 were selected (Table S2). Since non-sterilized soil represents

239

a natural environment more so than sterilized soil, our second screening was performed

240

only in the non-sterilized soil, and 39 mutants were selected as those most likely to be

241

deficient in competitive fitness in the soil. An attempt was made to determine the

242

plasposon-insertion sites in the 39 mutants by the direct sequencing method, and the

243

insertion sites in 32 mutants were successfully determined (Table 2). Among the 32

244

mutants, five (2F12-20, 2F9-20, 2H10-20, 1C1-19, and 2C8-23) and two (1A5-6 and

245

2E12-12) had the plasposon insertions into intergenic regions and the hrpA gene,

246

respectively, and thus a total of 26 genes were identified as candidate genes for the

247

competitive fitness in the soil (Table 2). These genes were classified into various COG

248

categories (http://www.ncbi.nlm.nih.gov/COG/), and no auxotrophic genes were identified

249

in our screening, although such mutants were not excluded from our STM mutant library.

250

Reconstruction of disruption strains of five genes. Among the 26 candidate genes, five

251

genes (sun, fur, tig, hrpA, and deaD; Fig. 1) were further analyzed, mainly because mutants

252

carrying insertion mutations in these genes were reproducibly selected in the second STM

253

screening (Table 2). To exclude the possibility that another unexpected and distantly 11

254

located mutation caused the phenotype in each of the original mutants, each of the five

255

wild-type genes in ATCC 17616 was disrupted using the TnMod-OTp’-tagn-containing

256

genomic fragments from the original STM mutant as described in Methods. When

257

compared with the wild-type strain, the five reconstructed mutants (ATCC 17616sun::Tn,

258

ATCC 17616fur::Tn, ATCC 17616tig::Tn, ATCC 17616deaD::Tn, and ATCC

259

17616hrpA::Tn) showed no apparent growth defects in 1/3LB (data not shown). Each

260

reconstructed mutant was singly inoculated into the non-sterilized soil and incubated at

261

30oC, and the incubation period was prolonged on the basis of a pilot experiment (Fig. S2).

262

The amount of the singly inoculated control strain 2C4-34 did not change after the 20- to

263

40-day incubation (Fig. S2a). On the other hand, the relative abundance of ATCC

264

17616hrpA::Tn to 2C4-34 did not change after the 20-day incubation, but was significantly

265

reduced after the 30- to 40-day incubation (Fig. S2b). Such longer incubation was indeed

266

required because the number of inoculated reconstructed mutant cells of each type was

267

larger than that of the corresponding type of mutant cells used in the two-step STM

268

screening. The relative abundances of ATCC 17616hrpA::Tn, ATCC 17616fur::Tn, and

269

ATCC 17616deaD::Tn after the 30-day incubation in the soil, as compared with that of the

270

control strain, 2C4-34, were significantly reduced in at least one of three independent

271

measurements (Fig. 2a), and thus their fitness deficiencies in the soil could be attributed to

272

the plasposon insertions or unknown mutations located very close to the insertion sites. On

273

the other hand, the two other reconstructed mutants, ATCC 17616sun::Tn and ATCC

274

17616tig::Tn, did not show any obvious fitness deficiencies in our three independent

275

measurements (Fig. 2a). Although the results for ATCC 17616fur::Tn and ATCC

276

17616deaD::Tn varied in the first three measurements (Fig. 2a), nine additional

277

independent measurements supported the fitness deficiency of ATCC 17616fur::Tn in the 12

278

soil (Fig. 2b: see below).

279

Complementation of the reconstructed mutants. The wild-type fur and hrpA genes of

280

ATCC 17616 were successfully introduced into ATCC 17616(fur::Tn) and ATCC

281

17616(hrpA::Tn) genomes, respectively, to obtain ATCC 17616fur::Tn(chr1::fur) and

282

ATCC 17616hrpA::Tn(chr1::hrpA). Since the deaD gene could not be cloned in E. coli, the

283

subsequent complementation analysis was not conducted. The ATCC 17616 deaD gene

284

may have a toxic or lethal effect on E. coli cells. Our observation that ATCC

285

17616fur::Tn(chr1::fur) recovered fitness (Fig. 2b) led to the conclusion that the fur gene is

286

one of the genetic determinants for fitness in the soil. On the other hand, ATCC

287

17616hrpA::Tn(chr1::hrpA) still showed a fitness deficiency in the soil (Fig. 2b),

288

suggesting that the hrpA gene itself is not important for fitness in the soil, and that the

289

plasposon insertion may have created a polar effect on gene expression producing an

290

altered phenotype.

291

Altered phenotypes of the reconstructed mutants. Sensitivities of the reconstructed

292

mutants towards nine antibiotics, ampicillin, kanamycin, polymyxin B, Cm, Tc, Gm,

293

thiamphenicol, florfenicol, and fusaric acid, were assayed. ATCC 17616hrpA::Tn and

294

ATCC 17616fur::Tn were more sensitive to Cm (Fig. 3a) and to Gm and Tc (Fig. 3b),

295

respectively, than the wild-type strain, and these sensitivities were complemented by the

296

introduction of the corresponding wild-type genes (Fig. 3ab). Furthermore, ATCC

297

17616deaD::Tn showed an obvious growth defect at 15oC (Fig. 3d). These results

298

suggested that the fur, hrpA, and deaD genes are involved in these phenotypes.

299 300 301

DISCUSSION In this study, we conducted STM screening of the mutant library of a bacterial strain, 13

302

B. multivorans ATCC 17616, in order to identify genetic determinants for fitness in soil.

303

Our STM screening using approximately 7,000 independent mutants indeed supported this

304

approach for identifying the genes important for survival in the soil. We were able to

305

successfully select 32 candidate mutants after the two-step STM screening (Table 2), and

306

three (fur, deaD, and hrpA) out of the five mutants indeed exhibited fitness defects in the

307

non-sterilized soil at least once in our repeated experiments (Fig. 2a). There is no apparent

308

preference in the COG categories for the candidate genes of B. multivorans that were

309

required for the soil fitness, and such genes are mainly distributed on Chr 1, which is

310

considered to be the main chromosome carrying most of the essential genes in B.

311

multivorans (Nagata et al., 2005). This distribution contrasts with that of the 116 mls, 52%

312

of which are located on Chr 2 (Nishiyama et al., 2010). Almost none of the candidate

313

genes for fitness in the soil were located within nor in the vicinity of the 116 mls, probably

314

because the genes that were identified by the IVET screening to be induced in the soil are

315

not necessarily important for survival even in the same environment, and the genes

316

identified by the STM screening are not always inducible. However, the number of

317

mutants obtained through STM screening may have been insufficient for such a

318

comparison, and a larger scale STM screen may improve the chance of isolating some of

319

the 116 mls genes. The only exception at this stage was a repeatedly identified locus, mls54

320

(carrying the putative aco operon for acetoin metabolism), which was closely associated

321

with a plasposon-inserted gene (acoR) in 2H3-29 (Table 2), whose wild-type gene is

322

postulated to encode a transcriptional activator for the aco operon (Fig. S3).

323

The deficiency in the soil fitness of the fur mutant was recovered by the introduction

324

of the wild-type fur gene, indicating that the fur gene is one of the genetic determinants for

325

fitness in the soil (Fig. 2b). The fur product, a ferric uptake regulator, is a global 14

326

transcriptional regulator for the iron regulon in many Gram-negative bacterial species

327

(Faulkner and Helmann, 2011). As has been noted in other phylogenetically distant

328

bacteria, the Fur protein in ATCC 17616 plays pleiotropic roles in the maintenance of iron

329

homeostasis, removal and/or resistance to reactive oxygen and nitrogen species (ROS and

330

RNS, respectively), and energy metabolism of carbon and energy sources (Yuhara et al.,

331

2008). We have also indicated that there is a higher level of ferric ion in the ATCC 17616

332

fur mutant, leading to the generation of a larger amount of very toxic ROS, i.e., hydroxyl

333

radicals, by the Fenton reaction (Kimura et al., 2012). The fur mutant additionally showed

334

higher sensitivities to Tc and Gm than the wild-type strain (Fig. 3b). Therefore, the fur

335

mutant is likely to be easily attacked by ROS, RNS, and the antibiotics produced by

336

unidentified microorganisms co-residing in the non-sterilized soil. Under laboratory

337

conditions, Fur also positively regulates the expression of the andA operon

338

(andAcAdAbAa) for anthranilate catabolism in ATCC 17616 (Nishiyama et al., 2012),

339

whose expression was increased dramatically in the type of soil the same as used in this

340

study (Nishiyama et al., 2010). The proliferation ability of the andAc mutant in the

341

sterilized soil was very low compared with that of the co-incubated wild-type strain

342

(Nishiyama et al., 2012). Considering all of these observations together, the fitness defect

343

in the soil of the fur mutant may be due to its increased sensitivity to ROS, RNS, and

344

antibiotics, as well as loss of the positive regulation of the expression of various genes that

345

are directly or indirectly mediated by Fur.

346

In this study, although the complementation analysis of the deaD mutant could not

347

be conducted and the fitness defect of hrpA mutation could not be complemented by the

348

wild-type hrpA gene itself (Fig. 2b), both genes deserve to be analyzed in more detail,

349

because (i) they both encode putatively similar DExD/H-box RNA helicases, (ii) ATCC 15

350

17616deaD::Tn exhibited, as has been shown in the deaD mutants in other bacterial

351

species, cold sensitivity (Fig. 3d), (iii) two hrpA mutants were independently identified in

352

the STM screening (Table 2 and Fig. 1d), (iv) ATCC 17616hrpA::Tn was more sensitive to

353

Cm than the wild-type strain, and the sensitivity was decreased by supply of the wild-type

354

hrpA gene (Fig. 3a). In the complementation analysis of the hrpA mutant, the cloned

355

genomic region for the complementation might have been lacking the cis-acting element

356

specifically necessary for proper expression of the hrpA gene for fitness in the soil.

357

Although the general functions of DExD/H-box RNA helicases remain obscure in

358

prokaryotes, such enzymes are suggested to be involved in a wide variety of RNA-related

359

functions in an energy-dependent manner, including roles in global gene regulation, in

360

some bacterial strains (Koo et al., 2004; Cordin et al., 2006; Iost and Dreyfus, 2006;

361

Salman-Dilgimen et al., 2011; Tamura et al., 2012; Redder and Linder, 2012). HrpA and

362

DeaD of ATCC 17616 might also be involved in the RNA-related functions for fitness in

363

the soil.

364 365

ACKNOWLEDGEMENTS

366

This work was supported by Grants-in-Aid from the Ministry of Education, Culture, Sports,

367

Science and Technology, Japan.

368 369

REFERENCES

370

Cordin, O., Banroques, J., Tanner, N. K. & Linder, P. (2006). The DEAD-box protein

371

family of RNA helicases. Gene 367, 17-37.

16

372

Dennis, J. J. & Zylstra, G. J. (1998). Plasposons: modular self-cloning minitransposon

373

derivatives for rapid genetic analysis of gram-negative bacterial genomes. Appl Environ Microbiol

374

64, 2710-2715.

375

Faulkner, M. J. & Helmann, J. D. (2011). Peroxide stress elicits adaptive changes in bacterial

376

metal ion homeostasis. Antioxid Redox Signal 15, 175-189.

377

Figurski, D. H. & Helinski, D. R. (1979). Replication of an origin-containing derivative of

378

plasmid RK2 dependent on a plasmid function provided in trans. Proc Natl Acad Sci USA 76,

379

1648-1652.

380

Groh, J. L., Luo, Q., Ballard, J. D. & Krumholz, L. R. (2005). A method adapting

381

microarray technology for signature-tagged mutagenesis of Desulfovibrio desulfuricans G20 and

382

Shewanella oneidensis MR-1 in anaerobic sediment survival experiments. Appl Environ Microbiol

383

71, 7064-7074.

384

Groh, J. L., Luo, Q., Ballard, J. D. & Krumholz, L. R. (2007). Genes that enhance the

385

ecological fitness of Shewanella oneidensis MR-1 in sediments reveal the value of antibiotic

386

resistance. Appl Environ Microbiol 73, 492-498.

387

Handfield, M. & Levesque, R. C. (1999). Strategies for isolation of in vivo expressed genes

388

from bacteria. FEMS Microbiol Rev 23, 69-91.

389

Hensel, M., Shea, J. E., Gleeson, C., Jones, M. D., Dalton, E. & Holden, D. W. (1995).

390

Simultaneous identification of bacterial virulence genes by negative selection. Science 269,

391

400-403.

392

Hoang, T. T., Karkhoff-Schweizer, R. R., Kutchma, A. J. & Schweizer, H. P. (1998). A

393

broad-host-range Flp-FRT recombination system for site-specific excision of

394

chromosomally-located DNA sequences: application for isolation of unmarked Pseudomonas

395

aeruginosa mutants. Gene 212, 77-86. 17

396

Hunt, T. A., Kooi, C., Sokol, P. A. & Valvano, M. A. (2004). Identification of Burkholderia

397

cenocepacia genes required for bacterial survival in vivo. Infect Immun 72, 4010-4022.

398

Iost, I. & Dreyfus, M. (2006). DEAD-box RNA helicases in Escherichia coli. Nucleic Acids Res

399

34, 4189-4197.

400

Kimura, A., Yuhara, S., Ohtsubo, Y., Nagata, Y. & Tsuda, M. (2012). Suppression of

401

pleiotropic phenotypes of a Burkholderia multivorans fur mutant by oxyR mutation. Microbiology

402

158, 1284-1293.

403

Komatsu, H., Imura, Y., Ohori, A., Nagata, Y. & Tsuda, M. (2003). Distribution and

404

organization of auxotrophic genes on the multichromosomal genome of Burkholderia multivorans

405

ATCC 17616. J Bacteriol 185, 3333-3343.

406

Koo, J. T., Choe, J. & Moseley, S. L. (2004). HrpA, a DEAH-box RNA helicase, is involved

407

in mRNA processing of a fimbrial operon in Escherichia coli. Mol Microbiol 52, 1813-1826.

408

Luo, Q., Groh, J. L., Ballard, J. D. & Krumholz, L. R. (2007). Identification of genes that

409

confer sediment fitness to Desulfovibrio desulfuricans G20. Appl Environ Microbiol 73,

410

6305-6312.

411

Maniatis, T., Fritsch, E. F. & Sambrook, J. (1982). Molecular cloning : a laboratory

412

manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, N Y.

413

Mazurkiewicz, P., Tang, C. M., Boone, C. & Holden, D. W. (2006). Signature-tagged

414

mutagenesis: barcoding mutants for genome-wide screens. Nat Rev Genet 7, 929-939.

415

Nagata, Y., Matsuda, M., Komatsu, H., Imura, Y., Sawada, H., Ohtsubo, Y. & Tsuda,

416

M. (2005). Organization and localization of the dnaA and dnaK gene regions on the

417

multichromosomal genome of Burkholderia multivorans ATCC 17616. J Biosci Bioeng 99,

418

603-610.

18

419

Nishiyama, E., Ohtsubo, Y., Nagata, Y. & Tsuda, M. (2010). Identification of

420

Burkholderia multivorans ATCC 17616 genes induced in soil environment by in vivo expression

421

technology. Environ Microbiol 12, 2539-2558.

422

Nishiyama, E., Ohtsubo, Y., Yamamoto, Y., Nagata, Y. & Tsuda, M. (2012). Pivotal role

423

of anthranilate dioxygenase genes in the adaptation of Burkholderia multivorans ATCC 17616 in

424

soil. FEMS Microbiol Lett 330, 46-55.

425

O'Sullivan, L. A., Weightman, A. J., Jones, T. H., Marchbank, A. M., Tiedje, J. M. &

426

Mahenthiralingam, E. (2007). Identifying the genetic basis of ecologically and

427

biotechnologically useful functions of the bacterium Burkholderia vietnamiensis. Environ

428

Microbiol 9, 1017-1034.

429

Ohtsubo, Y., Goto, H., Nagata, Y., Kudo, T. & Tsuda, M. (2006). Identification of a

430

response regulator gene for catabolite control from a PCB-degrading beta-proteobacteria,

431

Acidovorax sp. KKS102. Mol Microbiol 60, 1563-1575.

432

Ohtsubo, Y., Ikeda-Ohtsubo, W., Nagata, Y. & Tsuda, M. (2008). GenomeMatcher: a

433

graphical user interface for DNA sequence comparison. BMC bioinformatics 9, 376.

434

Ohtsubo, Y., Ishibashi, Y., Naganawa, H., Hirokawa, S., Atobe, S., Nagata, Y. &

435

Tsuda, M. (2012). Conjugal transfer of polychlorinated biphenyl/biphenyl degradation genes in

436

Acidovorax sp. strain KKS102, which are located on an integrative and conjugative element. J

437

Bacteriol 194, 4237-4248.

438

Ramos, J. L., Marques, S., van Dillewijn, P. & other authors (2011). Laboratory research

439

aimed at closing the gaps in microbial bioremediation. Trends Biotechnol 29, 641-647.

440

Redder, P. & Linder, P. (2012). DEAD-box RNA helicases in gram-positive RNA decay.

441

Methods Enzymol 511, 369-383.

19

442

Rediers, H., Rainey, P. B., Vanderleyden, J. & De Mot, R. (2005). Unraveling the secret

443

lives of bacteria: use of in vivo expression technology and differential fluorescence induction

444

promoter traps as tools for exploring niche-specific gene expression. Microbiol Mol Biol Rev 69,

445

217-261.

446

Salman-Dilgimen, A., Hardy, P. O., Dresser, A. R. & Chaconas, G. (2011). HrpA, a

447

DEAH-box RNA helicase, is involved in global gene regulation in the Lyme disease spirochete.

448

PLoS One 6, e22168.

449

Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular cloning: a laboratory manual,

450

2nd edn. Cold Spring Harbor, N.Y., USA: Cold Spring Harbor laboratory Press.

451

Schweizer, H. P. (1992). Allelic exchange in Pseudomonas aeruginosa using novel ColE1-type

452

vectors and a family of cassettes containing a portable oriT and the counter-selectable Bacillus

453

subtilis sacB marker. Mol Microbiol 6, 1195-1204.

454

Stanier, R. Y., Palleroni, N. J. & Doudoroff, M. (1966). The aerobic pseudomonads: a

455

taxonomic study. J Gen Microbiol 43, 159-271.

456

Tamura, M., Kers, J. A. & Cohen, S. N. (2012). Second-site suppression of RNase E

457

essentiality by mutation of the deaD RNA helicase in Escherichia coli. J Bacteriol 194,

458

1919-1926.

459

Tyagi, M., da Fonseca, M. M. & de Carvalho, C. C. (2011). Bioaugmentation and

460

biostimulation strategies to improve the effectiveness of bioremediation processes. Biodegradation

461

22, 231-241.

462

van Veen, J., van Overbeek, L. & van Elsas, J. (1997). Fate and activity of

463

microorganisms introduced into soil. Mirobiol Mol Biol Rev 61, 121-135.

464

Wang, Y., Shimodaira, J., Miyasaka, T., Morimoto, S., Oomori, T., Ogawa, N.,

465

Fukuda, M. & Fujii, T. (2008). Detection of bphAa gene expression of Rhodococcus sp. strain 20

466

RHA1 in soil using a new method of RNA preparation from soil. Biosci Biotechnol Biochem 72,

467

694-701.

468

Yuhara, S., Komatsu, H., Goto, H., Ohtsubo, Y., Nagata, Y. & Tsuda, M. (2008).

469

Pleiotropic roles of iron-responsive transcriptional regulator Fur in Burkholderia multivorans.

470

Microbiology 154, 1763-1774.

21

471

FIGURE LEGENDS

472 473

Fig. 1. Organizations of the sun, fur, tig, hrpA, and deaD genes and their flanking regions

474

of ATCC 17616. Transposon-insertion sites of 2F9-22 (a), 2B6-12 (b), 2G3-7 (c), 1A5-6

475

and 2E12-12 (d), and 2C9-19 (e) are marked with triangles. See Table 2 for detailed

476

information on each locus.

477 478

Fig. 2. Survival of reconstructed sun, fur, tig, deaD, and hrpA mutants and their derivatives

479

in the soil. Each mutant was grown in 1/3LB, singly inoculated into the non-sterilized soil,

480

and incubated at 30oC for 30 days. DNA sample directly extracted from the soil after the

481

incubation was used to estimate the relative abundance of each mutant to the control strain

482

(2C4-34 and 1G1-31 in panels a and b, respectively) by qRT-PCR using the DNA as the

483

template. a. The averages and standard deviations of three independent measurements (#1,

484

#2, and #3) are shown. Since the standard deviations for the fur and deaD mutants are

485

relatively large, values of three measurement are also shown at the bottom. N.D., not

486

detectable. b. The averages of nine other independent measurements and the standard

487

deviations are shown. Statistical analysis was performed using the paired Student t test: *,

488

P < 0.005.

489 490

Fig. 3. Altered phenotypes of the reconstructed mutants. Cells at log phase cultured in

491

1/3LB were diluted, and cell suspensions containing 102-107 c.f.u. ml-1 were spotted on

492

1/3LB agar plates containing 5 µg/ml of Cm (a), 30 µg/ml of Gm (b), and 10 µg/ml of Tc

493

(b) and incubated at 30oC for 2 days. Growth of the mutants in 1/3LB at 30oC (c) and 15oC

494

(d) was analyzed by measuring the OD660 values of cultures. More than three independent 22

495

measurements showed the same tendency, and the representative results are shown.

23

Table 1. Bacterial strains and plasmids used in this study Strain or plasmid Escherichia coli DH5α HB101 Burkholderia multivorans ATCC 17616 ATCC 17616sun::Tn ATCC 17616fur::Tn ATCC 17616tig::Tn ATCC 17616deaD::Tn ATCC 17616hrpA::Tn 2C4-34 1G1-32 ATCC 17616fur::Tn(chr::fur) ATCC 17616hrpA::Tn(chr::hrpA) Plasmids pRK2013 pTnMod-OTp' pTnMod-OTp'-tagn pTnMod-OKm' pEX18Tc pKLZG pYO103

Relevant characteristics F-, Φ80dlacZΔM15, Δ(lacZYA-argF)U169, deoR, recA1, endA1, hsdR17(rK-, mK+), phoA, supE44, λ-, thi-1, gyrA96, relA1 supE44, Δ(mcrC-mrr), recA13, ara-14, proA2, lacY1, galK2, rpsL20, xyl-5, mtl-1, leuB6, thi-1 Soil isolate; genomovar II of the Bcc Tpr; ATCC 17616sun::TnMod-Tp'-tag22; Sun: 16S rRNA methyltransferase Tpr; ATCC 17616fur::TnMod-Tp'-:tag12; Fur: Ferric uptake regulator Tpr; ATCC 17616tig::TnMod-Tp'-tag7; Tig: Trigger factor Tpr; ATCC 17616deaD::TnMod-Tp'-tag19; DeaD: DEAD-box RNA helicase Tpr; ATCC 17616hrpA::TnMod-Tp'-tag12; HrpA: DEAH-box RNA helicase Tpr; ATCC 17616potA::TnMod-Tp'-tag34; control strain Tpr; ATCC 17616astD::TnMod-Tp'-tag32; control strain ATCC 17616 fur::Tn(chr 1::fur) ATCC 17616 hrpA::Tn(chr 1::hrpA) ColE1::RK2 Tra+, Kmr pMB1 replicon, Tn5 inverted repeat, Tpr pTnMod-OTp' with tagn pMB1 replicon, Tn5 inverted repeat, Tpr Tcr sacB oriT; suicide vector for gene replacement carrying the pUC18-derived multiple cloning sites pKLZ-X derivative; carries gfp gene from pGreen TIR in front of the lacZ gene pEX18Tc-based plasmid for integration into ATCC 17616 chromosome 1 of a gene in conjunction with pKLZ-G-derived gfp and lacZ genes Tcr; pYO103 carrying fur Tcr; pYO103 carrying hrpA

pYO103-fur pYO103-hrpA a American Type Culture Collection b The tags were intoduced into pTnMod-OTp' in this stuy, but the resultant tagged plasposons are essentially identical to those constructed by Hunt et al (2004).

Source or reference Sambrook et al., 1989 Maniatis et al., 1982 ATCCa This study This study This study This study This study This study This study This study This study Figurski et al., 1979 Dennis and Zylstra, 1998 This studyb Dennis and Zylstra, 1998 Hoang et al., 1998 Ohtsubo et al., 2012 This study (see Fig. S1) This study This study

Table 2. 32 transposon-insertion sites of mutants selected by our second STM screening strain 1st screening soil frequencya replicon insertion siteb gene predicted function COG category 2F9-22 both 100% (2/2) Chr1 117300 sun 16S rRNA methyltransferase J (Translation, ribosomal structure and biogenesis) 2B6-12 non-sterilized 100% (2/2) Chr1 535385 fur ferric uprake regulator P (Inorganic ion transport and metabolism) 2G3-7 non-sterilized 100% (2/2) Chr1 2043673 tig trigger factor O (Posttranslational modification, protein turnover, chaperones) 1A5-6 sterilized 100% (3/3) Chr1 2300695 hrpA DEAH-box RNA helicase L (DNA replication, recombination and repair) 2E12-12 non-sterilized 100% (2/2) Chr1 2300745 hrpA DEAH-box RNA helicase L (DNA replication, recombination and repair) α-amylase-like protein 2G12-6 non-sterilized 100% (3/3) Chr3 728830 no name c G (Carbohydrate transport and metabolism) 2H3-29 non-sterilized 100% (1/1) Chr1 1949194 acoR sigma54-dependent DNA-binding transcriptional regulator Q (Secondary metabolites biosynthesis, transport and catabolism) 2C9-19 non-sterilized 100% (1/1) Chr1 2171180 deaD DEAD-box RNA helicase LK (Transcription) J 2F12-20 non-sterilized 100% (1/1) Chr1 2443228 intergenic 2E1-20 non-sterilized 100% (1/1) Chr1 2657876 no name mechanosensitive ion channel M (Cell envelope biogenesis, outer membrane) 2F9-20 non-sterilized 100% (1/1) Chr1 2975142 intergenic 2H10-20 non-sterilized 100% (1/1) Chr2 1765329 intergenic 1H10-10 sterilized 67% (2/3) Chr1 21100 no name RND efflux system outer membrane lipoprotein MN (Cell motility and secretion) 2D10-38 sterilized 67% (2/3) Chr1 2773000 acrR acrAB operon repressor K (Transcription) 1E6-12 sterilized 50% (1/2) Chr1 18722 gspL general secretion pathway protein L no hit 1B8-9 sterilized 50% (1/2) Chr1 1905456 hisZ ATP phosphoribosyltransferase regulatory subunit E (Amino acid transport and metabolism) 1C1-19 non-sterilized 50% (1/2) Chr1 2213291 intergenic 2F8-10 non-sterilized 50% (1/2) Chr1 2743765 xylB xylulokinase G (Carbohydrate transport and metabolism) 2D2-41 non-sterilized 50% (1/2) Chr1 2811791 no name OmpR family two-component system sensor histidine kinase T (Signal transduction mechanisms) 2G1-40 non-sterilized 50% (1/2) Chr1 2910139 no name FAD/FMN-containing dehydrogenase C (Energy production and conversion) 2G6-33 non-sterilized 50% (1/2) Chr1 2974524 no name putative transcriptional regulator no hit 2F4-27 non-sterilized 50% (1/2) Chr2 547430 ophA2 phthalate dioxygenase oxygenase PR (General function prediction only) 2E4-31 both 50% (1/2) Chr2 2263100 hit Histidine triad(Hit)-like nucleotide-binding protein F (Nucleotide transport and metabolism) GR 2C9-29 non-sterilized 50% (1/2) Chr2 293577 no name putative protein-S-isoprenylcysteine methyltransferase no hit 1C5-7 non-sterilized 33% (1/3) Chr1 1296743 mltD membrane-bound lytic murein transglycosylase D M (Cell envelope biogenesis, outer membrane) 2C4-14 non-sterilized 25% (1/4) Chr1 627299 glcF glycolate oxidase iron-sulfur subunit C (Energy production and conversion) 2H7-8 non-sterilized 25% (1/4) Chr1 3092765 no name conserved hypothetical protein no hit 2B7-14 non-sterilized 20% (1/5) Chr1 2761798 adhC S-(hydoroxymethyl)glutathione dehydrogenase C (Energy production and conversion) 2F4-35 non-sterilized 20% (1/5) Chr1 3009298 no name putative ABC transport system ATP-binding protein EPRG 2F7-4 non-sterilized 19% (4/21) Chr1 1990859 no name conserved hypothetical protein no hit 2C8-23 non-sterilized 17% (1/6) Chr2 486172 intergenic 2G7-28 non-sterilized 10% (1/10) Chr1 1270944 araJ arabinose polymer transporter G (Carbohydrate transport and metabolism) a in the second STM screening (See Supplemental Table S2) b positions and locus tag names used are on the basis of our deposited sequences [AP009385, AP009386 , and AP009387 for Chr1, Chr2, and Chr3, respectively, in DDBJ/EMBL/GenBank databases (29)] c no gene name associated with this open reading frame in the annotated ATCC 17616 genome

locus tagb BMULJ_00110 BMULJ_00500 BMULJ_01897 BMULJ_02139 BMULJ_02139 BMULJ_06005 BMULJ_01812 BMULJ_02017

locusb 116031..117434 complement (535072..535500) complement (2042902..2044248) complement (2298662..2302798) complement (2298662..2302798) 728555..731950 1949097..1951013 2171016..2172533

BMULJ_02473

complement (2657579..2658967)

BMULJ_00019 BMULJ_01627 BMULJ_00016 BMULJ_01770

complement (20375..21958) 2772679..2773317 17600..19006 complement (1904551..1905600)

BMULJ_02549 BMULJ_02611 BMULJ_02697 BMULJ_02754 BMULJ_03712 BMULJ_05199 BMULJ_03468 BMULJ_01211 BMULJ_00584 BMULJ_02864 BMULJ_02567 BMULJ_02782 BMULJ_01853

2742592..2744073 2811572..2813119 complement (2906814..2910848) 2974520..2974867 complement (546783..548114) complement (2263046..2263465) complement (292953..294236) 1295060..1296646 complement (626795..628021) 3092315..3093604 complement (2761318..2762424) complement (3009151..3009945) 1990916..1991413

BMULJ_01183

complement (1270047..1271252)

1A5-6 2E12-12

Fig. 1. Nagata et al.

Relative amounts of specific PCR products

a 1.2 1 0.8 0.6 0.4 0.2 0

#1 #2 #3

control (2C4-34)

sun::Tn

1 1 1

0.94 0.86 0.88

fur::Tn N.D. 0.4 0.86

tig::Tn

deaD::Tn hrpA::Tn

1.05 0.96 0.87

N.D. 0.93 0.8

N.D. N.D. N.D.

Relative amounts of specific PCR products

b 1.4 1.2 1 0.8

*

0.6

*

0.4

*

0.2 0

control (1G1-32)

fur::Tn

fur::Tn Chr::fur

hrpA::Tn

Fig. 2. Nagata et al.

hrpA::Tn Chr::hrpA

a

1/3LB

1/3LB + Cm 5 µg/ml

Wild type hrpA::Tn hrpA::Tn Chr::hrpA [c.f.u. ml-1]

b

107

105

103

1/3LB

107

105

103

1/3LB + Gm 30 µg/ml

1/3LB + Tc 10 µg/ml

107

107

Wild type fur::Tn fur::Tn Chr::fur [c.f.u. ml-1]

107

105

c

103

105

d

30oC 2

105

103

15oC

1

1.5

1

●　wild type △　hrpA::Tn □　deaD::Tn

0.5

0

Growth (OD660)

Growth (OD660)

103

0.8

●　wild type △　hrpA::Tn □　deaD::Tn

0.6 0.4 0.2 0

0

12

18

24

0

Time (h)

16

24 Time (h)

Fig. 3. Nagata et al.

48

72