Plasmid pUM505 encodes a Toxin–Antitoxin system conferring plasmid stability and increased Pseudomonas aeruginosa virulence

Accepted Manuscript Plasmid pUM505 encodes a Toxin–Antitoxin system conferring plasmid stability and increased Pseudomonas aeruginosa virulence K.C. Hernández-Ramírez, V.M. Chávez-Jacobo, M.I. Valle-Maldonado, J.A. PatiñoMedina, S.P. Díaz-Pérez, I.E. Jácome-Galarza, R. Ortiz-Alvarado, V. Meza-Carmen, M.I. Ramírez-Díaz PII:

S0882-4010(17)30654-X

DOI:

10.1016/j.micpath.2017.09.060

Reference:

YMPAT 2506

To appear in:

Microbial Pathogenesis

Received Date: 6 June 2017 Revised Date:

20 September 2017

Accepted Date: 28 September 2017

Please cite this article as: Hernández-Ramírez KC, Chávez-Jacobo VM, Valle-Maldonado MI, PatiñoMedina JA, Díaz-Pérez SP, Jácome-Galarza IE, Ortiz-Alvarado R, Meza-Carmen V, RamírezDíaz MI, Plasmid pUM505 encodes a Toxin–Antitoxin system conferring plasmid stability and increased Pseudomonas aeruginosa virulence, Microbial Pathogenesis (2017), doi: 10.1016/ j.micpath.2017.09.060. 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.

ACCEPTED MANUSCRIPT

1

Plasmid pUM505 Encodes a Toxin–Antitoxin System Conferring Plasmid

2

Stability and Increased Pseudomonas aeruginosa Virulence

3 K.C. Hernández-Ramírez1, V.M. Chávez-Jacobo1, M.I. Valle-Maldonado1, J.A.

5

Patiño-Medina1, S.P. Díaz-Pérez1, I.E. Jácome-Galarza2, R. Ortiz-Alvarado3, V.

6

Meza-Carmen1 and M.I. Ramírez-Díaz1*

RI PT

4

SC

7 8

1

9

Morelia, Michoacán, México. 2Laboratorio Estatal de Salud Pública, Secretaría de

M AN U

Instituto de Investigaciones Químico-Biológicas, Universidad Michoacana,

10

Salud Michoacán, Morelia, México. 3Facultad de Químico-Farmacobiología,

11

Universidad Michoacana, Morelia, México.

12

*To whom correspondence should be addressed: Martha I. Ramírez-Díaz, Instituto

14

de Investigaciones Químico-Biológicas, Universidad Michoacana, Edificio B-3,

15

Ciudad Universitaria, 58030 Morelia, Mich., México. Phone/fax: (+52) (443) 326

16

5788; E-mail address: [email protected]

EP

TE D

13

17

19 20 21

For: Microbial Pathogenesis

AC C

18

September 2017

22

Key words: Pseudomonas, plasmid, virulence, toxin-antitoxin system, plasmid

23

stability.

24 1

ACCEPTED MANUSCRIPT

Abstract

26

Pseudomonas aeruginosa plasmid pUM505 possesses a pathogenicity island that

27

contains the pumAB genes that encode products with sequence similarity to Toxin–

28

Antitoxin (TA) modules. RT-PCR assays on the overlapping regions of the pumAB

29

genes generated a bicistronic messenger RNA, suggesting that they form an

30

operon. When the pumAB genes were cloned into the pJET vector, recombinant

31

plasmid pJET-pumAB was maintained under nonselective conditions in

32

Escherichia coli cells after six daily subcultures, whereas pJET without pumAB

33

genes was lost. These data indicate that pumAB genes confer post-segregational

34

plasmid stability. In addition, overexpression of the PumA protein in the E. coli

35

BL21 strain resulted in a significant growth inhibition, while BL21 co-expressing the

36

PumA and PumB proteins did not show growth inhibition. These results indicate

37

that pumAB genes encode a TA system where the PumB protein counters the toxic

38

effects of the PumA toxin. Furthermore, P. aeruginosa PAO1 transformants with

39

the pumA gene increased Caenorhabditis elegans and mouse mortality rate and

40

improved mouse organ invasion, effects neutralized by the PumB protein.

41

Moreover, purified recombinant His-PumA protein decreased the viability of C.

42

elegans, indicating that the PumA protein could acts as a toxin. These results

43

indicate that PumA has the potential to promoter the PAO1 virulence against C.

44

elegans and mice when is expressed in absence of PumB. This is the first

45

description, to our knowledge, of a plasmid-encoded TA system that confers

46

plasmid stability and encoded a toxin with the possible ability to increase the P.

47

aeruginosa virulence.

AC C

EP

TE D

M AN U

SC

RI PT

25

2

ACCEPTED MANUSCRIPT

1. Introduction

49

Pseudomonas aeruginosa is a versatile Gram-negative bacterium that is

50

ubiquitously distributed in different terrestrial, aquatic, animal, human, and plant

51

environments (Sousa and Pereira 2014). P. aeruginosa is one of the most

52

important opportunistic pathogens in humans. Patients with chronic diseases or

53

with compromised immune systems, such as immunodeficiency or burns, are at

54

greater risk of infection (Quick et al. 2014). In addition, P. aeruginosa uses a

55

shared subset of virulence factors to elicit disease in both plants and animals

56

(Rahme et al. 2000). The conjugative plasmid pUM505 is an IncI type of 123

57

kilobases (kb) in size, originally isolated from a P. aeruginosa clinical strain, which

58

confers resistance to chromate ions and inorganic mercury (Cervantes et al. 1990).

59

Sequence analysis revealed that pUM505 carries several adaptive genes,

60

including umuD, which functions as a transcriptional regulator of SOS genes (Díaz-

61

Magaña et al. 2015), as well as several genes related to virulence (Rodríguez-

62

Andrade et al. 2016). It has been suggested that pUM505 could provide

63

Pseudomonas strains with a wide variety of adaptive traits that can comprise

64

selective factors for the distribution and prevalence of this plasmid in diverse

65

environments (Ramírez-Díaz et al. 2011). The predicted products of Open Reading

66

Frames (ORF) 123 and 124 from pUM505 exhibit sequence similarity to Toxin–

67

Antitoxin (TA) systems (Ramírez-Díaz et al. 2011). TA systems are small genetic

68

modules found in bacterial mobile genetic elements, as well as in bacterial

69

chromosomes, which currently are group into the six types of TA systems (Leplae

70

et al. 2011; Page and Peti 2016). The toxins of all characterized bacterial TA

71

systems are proteins, while the antitoxins are either proteins or small RNA (sRNA)

AC C

EP

TE D

M AN U

SC

RI PT

48

3

ACCEPTED MANUSCRIPT

(Unterholznet et al. 2013). Antitoxins are more labile and more readily degraded

73

under stress conditions than toxins, allowing toxins to exert their toxic effects

74

(Yamaguchi et al. 2011). In TA systems, toxins are regarded as harmful molecules

75

to cells due to their cell-growth inhibitory effects, thereby conferring growth stasis in

76

stressful environments. The antitoxin molecules are usually binding partners of

77

toxins, either on a gene or on a protein level, and they regulate the activity of the

78

toxin (Park et al. 2013). Although TA systems were originally discovered in

79

plasmids, bacterial and archaeal genomes also contain diverse TA loci (Marakova

80

et al. 2009). Plasmid-encoded TA systems are important for plasmid stabilization

81

by a post-segregational killing mechanism that eliminates daughter cells that lost

82

the TA-encoding plasmid (Yamaguchi et al. 2011). In addition, plasmid TA systems

83

are important for the mediation of the exclusion of co-existing compatible plasmids

84

(Unterholzner et al. 2013). Chromosomally encoded TA systems typically compose

85

parts of genomic islands (Hayes and Van Melderen 2011), and it was recently

86

suggested that they comprise important elements in the virulence of bacterial

87

pathogens (Unterholzner et al. 2013; De la Cruz et al. 2015). The objective of this

88

work was to establish the function of the putative TA system from pUM505. We

89

report for the first time, to our knowledge, a plasmid-encoded TA system that

90

enhances plasmid prevalence in Pseudomonas cells whose toxin PumA has the

91

potential to increase the bacterial virulence.

SC

M AN U

TE D

EP

AC C

92

RI PT

72

93

2. Materials and Methods

94

2.1. Bacterial strains, culture conditions, and plasmids

4

ACCEPTED MANUSCRIPT

P. aeruginosa PAO1 (Li et al. 2007) and Escherichia coli JM101 (Yanisch-Perron

96

et al. 1985) are standard prototroph strains employed as hosts for recombinant

97

plasmids. E. coli BL21-CodonPlus® (DE3)-RP (Stratagene, San Diego, CA, USA)

98

was used for protein overexpression. PAO1-bearing plasmid pUM505 (HgR, CrR)

99

(Cervantes et al. 1990) was employed to evaluate the transcriptional expression of

RI PT

95

pumAB genes.

101

The culture media used were Nutrient Broth (NB) and Luria-Bertani broth (LB); for

102

solid media, 1.5% agar was added (Green and Sambrook 2012), or M9 salts

103

minimal medium (Sigma, St. Louis, MO, USA) supplemented with 20 mM glucose

104

and 0.1 mM CaCl2. When necessary, Ampicillin (100 µg ml–1) or Carbenicillin (400

105

µg ml–1) was added to agar plates. Cultures were routinely grown overnight at 37°C

106

with shaking, and growth was monitored as Optical Density (OD) at 600 nm

107

(OD600) with a spectrophotometer.

108

The pJET1.2/blunt vector (Thermo Fisher Scientific, Waltham, MA, USA) was

109

utilized to recover PCR-amplified DNA fragments as well as for cloning the pumAB

110

genes to assay plasmid stability. The recombinant plasmid pJET-orf131 contains

111

the orf131 gene from pUM505 (accession number AEQ93536) (Ramírez-Díaz et al.

112

2011), cloned in pJET1.2/blunt vector, which encodes a hypothetical protein of 65

113

amino acids (aa); this plasmid was employed as negative control in plasmid

114

stability tests. The pUCP20 Escherichia/Pseudomonas shuttle vector (West et al.

115

1994) was used for subcloning pum genes to determine their participation in

116

Pseudomonas virulence. pTrcHisC and pTrcHis2A vectors (Thermo Fisher

117

Scientific, Waltham, MA, USA) were utilized for the overexpression of Pum

118

proteins.

AC C

EP

TE D

M AN U

SC

100

5

ACCEPTED MANUSCRIPT

2.2. Genetic techniques and sequence analysis

120

Molecular genetics procedures were carried out by following standard methods

121

(Green and Sambrook 2012). Cloning processes were verified by DNA sequencing

122

conducted by Elim Biopharmaceuticals, Inc. (Hayward, CA, USA). Protein

123

sequence alignments were calculated with the CLUSTALW algorithm and BioEdit

124

Sequence Alignment Editor Software ver. 7.2.5.0. Sequence similarities were

125

searched in the protein database with the BLASTP program

126

(http://blast.ncbi.nlm.nih.gov/Blast.cgi). Probable promoter sequences were

127

searched with the Neural Network Promoter Prediction program

128

(http://www.fruitfly.org/seq_tools/promoter.html). Secondary protein structure was

129

determined using MINNOU (Membrane protein IdeNtificatioN withOUt explicit)

130

program (http://minnou.cchmc.org/).

M AN U

SC

RI PT

119

TE D

131

2.3. RNA isolation and RT-PCR analysis

133

Total RNA from bacterial cells grown in LB was isolated by utilizing the Tri reagent

134

(Molecular Research Center, Inc., Cincinnati, OH, USA). RNA was quantified by

135

spectrophotometric analysis at 260 nm, previously treated with RQ1 RNase-free

136

DNase (Promega, Madison, WI, USA). Reverse Transcription-PCR (RT-PCR) was

137

performed with total RNA samples and the Master AMP RT-PCR kit (Epicentre

138

Technologies, Madison, WI, USA) according to the manufacturer’s instructions.

139

Primers OD1 and OR1 (Table S1) were utilized to generate the 440-bp gene-

140

overlapping product of the pumAB genes (Fig. 1a). OR1 contains an XbaI

141

restriction enzyme site (underscored) for cloning proposes (see the cloning of the

142

Toxin–Antitoxin genes section). RT-PCR controls were treated with PCR Master

AC C

EP

132

6

ACCEPTED MANUSCRIPT

Mix (Promega #M7502) and with the primer pair previously described, using

144

genomic DNA templates (positive control) and total RNA in the absence of RT

145

(negative control). The size of RT-PCR products was assessed by agarose (1.5%)

146

gel electrophoresis employing the 1 kb-plus DNA ladder (Life Technologies,

147

Rockville, MD, USA) as a marker.

148

RI PT

143

2.4. Cloning of the pumAB genes

150

In order to amplify the pumAB genes, primers OD2 and OR2 (Table S1) were used,

151

employing DNA from pUM505 plasmid as template, and the fragment was

152

subsequently cloned into the pJET1.2/blunt vector. In order to amplify the pumA

153

and pumB genes independently, primers OD2 and OR1 and the OD3 and OR2 were

154

used, respectively (Table S1) (Fig. 1a). The amplified fragments were also purified

155

and cloned into the pJET1.2/blunt vector. Recombinant plasmids were transferred

156

by electroporation to E. coli JM101 and transformants were selected on LB agar

157

plates with Ampicillin.

158

For subcloning pumAB genes or pum individual genes into the pUCP20 vector,

159

recombinant plasmids from pJET derivatives were digested with restriction

160

enzymes BamHI and XbaI, and DNA fragments were ligated onto the

161

corresponding sites of pUCP20. Recombinant plasmids were transferred by

162

electroporation to P. aeruginosa PAO1 or E. coli JM101 and the transformants

163

were selected by culturing on LB agar with Carbenicillin or Ampicillin, respectively.

AC C

EP

TE D

M AN U

SC

149

164 165

2.5. Plasmid stability tests

7

ACCEPTED MANUSCRIPT

For stability tests, E. coli JM101 (pJET-pumAB) cultures were grown in NB without

167

Ampicillin selection at 37°C with shaking for 24 h. Serially diluted samples were

168

then plated on nutrient agar with or without Ampicillin. Plates were incubated for 24

169

h at 37°C and Colony-Forming Units (CFU) were measured for each condition. The

170

initial cultures were then diluted 1:50 in fresh medium without antibiotic, incubated

171

for 24 h, and the CFU were measured; the described procedure was repeated daily

172

for up to 6 days. As a negative control of plasmid stability, the same procedure was

173

followed using cultures of E. coli JM101 (pJET-orf131). To confirm the presence or

174

absence of plasmids pJET-pumAB or pJET-orf131, CFU were obtained from

175

nutrient agar plates with and without Ampicillin, and agarose gel electrophoresis of

176

plasmid DNA isolated by a mini-prep procedure was performed.

M AN U

SC

RI PT

166

177

2.6. Overexpression and purification of Toxin–Antitoxin proteins

179

The pumA gene coding region, as well as the DNA region from the pumA start

180

codon to the pumB stop codon (for co-expression of both PumA and PumB

181

proteins), were amplified by PCR using isolated pUM505 plasmid as the template;

182

the primers employed were Dir123pTrc and Rev123pTrc or ODpTrc2A123-4 and

183

ORpTrc2A123-4, respectively (Table S1). The amplified fragments were cloned

184

into the pJET1.2/blunt vector, and the recombinant plasmids were transferred by

185

electroporation to E. coli JM101 and transformants were selected on LB agar

186

plates with Ampicillin. The pumA gene coding region was obtained after digesting

187

the plasmid with the restriction enzymes PstI and HindIII, and the pumAB region

188

was obtained by enzyme digestion with BamHI and HindIII; both fragments were

189

purified and subcloned into their corresponding restriction enzyme sites of the

AC C

EP

TE D

178

8

ACCEPTED MANUSCRIPT

pTrcHisC or pTrcHis2A vectors, respectively. Recombinant plasmid pTrcHis-pumA

191

encodes the tagged His-PumA recombinant protein of ~16 kDa that possesses a

192

6X-His tag at the N-terminal end. Recombinant plasmid pTrcHis-pumAB encodes

193

the PumA protein of a theoretical 11.1 kDa in size and the tagged PumB-His

194

recombinant protein of ~17 kDa that possesses a 6X-His tag at the C-terminal end.

195

Fragment directionality was verified by sequencing with the pTrcHis Forward

196

primer (Thermo Fisher Scientific, Waltham, MA, USA). The recombinant plasmids

197

were transferred by electroporation to E. coli BL21-CodonPlus® (DE3)-RP, followed

198

by selection of transformants on LB agar plates with Ampicillin. Overnight cultures

199

of E. coli BL21 (pTrcHis-pumA) or E. coli BL21 (pTrcHis-pumAB) were diluted 1:50

200

into 250 ml of fresh NB medium and cultured with shaking at 37°C until OD600 =

201

0.5. Overexpression of His-PumA and PumA/PumB-His (in co-expression) proteins

202

was induced with 1 mM IsoPropyl-β-D-ThioGalactopyranoside (IPTG) and the

203

cultures were incubated for an additional 6 h. Cells were harvested by

204

centrifugation at 7,500 rpm for 30 min at 4°C; the pellets were suspended in

205

loading buffer [25 mM Tris (Tris [hydroxymethyl] amino methane)-HCl pH 6.8, 1%

206

SDS (Sodium Dodecyl Sulfate), 0.35 mM β-mercaptoethanol, 12.5% glycerol, and

207

0.005% bromophenol blue] and heated for 5 min at 95°C. Overproduction of the

208

His-PumA and PumB-His recombinant proteins were verified by SDS-PAGE using

209

a Tris-glycine buffer.

210

His-PumA and PumB-His proteins purification was performed as described by

211

Watanabe and Takada (2004). Cells were disrupted by sonic oscillation for

212

clarification and debris was removed by centrifugation. The supernatant was

AC C

EP

TE D

M AN U

SC

RI PT

190

9

ACCEPTED MANUSCRIPT

loaded onto a Nickel-NTA resin (Qiagen) packed into a column. The His-PumA and

214

PumB-His proteins were recovered by elution with 200 mM Imidazole as described

215

by Aranda et al. (2008). Purification was monitored by recombinant protein

216

detection by SDS-PAGE followed by staining with Coomassie blue.

217

For Western blot assays, a duplicate of the previously mentioned polyacrylamide

218

gel was transferred onto a nitrocellulose membrane (Hybond ECL). The membrane

219

was incubated 1 h at 37°C with HisProbe-HRP (Thermo, Rockford, IL, USA) diluted

220

at a 1:1,000 ratio, and the bound probe was visualized with the Super Signal West

221

Pico reagent (Thermo Fisher Scientific, Waltham, MA, USA).

222

M AN U

SC

RI PT

213

2.7. Bacterial growth assays

224

Liquid cultures of P. aeruginosa PAO1 or E. coli JM101 containing the pUC-pumA

225

recombinant plasmid were utilized to evaluate the effects of PumA protein

226

expression. Overnight cultures were diluted 1:50 in 250 ml of fresh NB medium and

227

growth was monitored at intervals by measuring OD600. Serially diluted samples

228

were then plated on nutrient agar with Ampicillin and incubated for 24 h at 37°C

229

and the CFU were measured.

230

In addition, the effects of His-PumA protein overexpression or the PumA and

231

PumB-His proteins co-expression in E. coli BL21 were assayed by measuring their

232

growth in liquid medium in the presence or absence of the IPTG inducer. Overnight

233

cultures of E. coli BL21 (pTrcHis-pumA) or E. coli BL21 (pTrcHis-pumAB) were

234

diluted 1:50 in 250 ml of fresh NB medium, and growth was monitored at intervals

235

by measuring OD600. At this point, PumA protein overexpression and PumAB

AC C

EP

TE D

223

10

ACCEPTED MANUSCRIPT

236

proteins co-expression were induced by adding 1 mM of IPTG, and bacterial

237

growth was monitored at intervals by measuring OD600.

238 2.8. RT-qPCR assays

240

For pumA gene expression measurement from the BL21 (pTrcHis-pumA) strain,

241

cultures were incubated for 6 h after their induction with 1 mM of IPTG, and cells

242

were harvested by centrifugation. Total RNA was isolated and quantified as

243

described in the RNA isolation method mentioned previously. Oligonucleotide

244

primers and hydrolysis probes for RT-qPCR of the pumA gene and the 16S rRNA

245

reference genes (listed in Table S1) were designed employing Biosearch

246

Technologies software

247

(https://www.biosearchtech.com/support/applications/realtimedesign-software) and

248

were purchased from Biosearch Technologies (Novato, CA, USA). Amplification of

249

the pumA gene and the 16S rRNA genes was performed using the 5′ exonuclease

250

probe RT-qPCR method. RT-qPCR was performed with total RNA samples (500

251

ng) and the SuperScript III Platinum One-Step RT-qPCR Reagent kit (Thermo

252

Fisher Scientific, Waltham, MA, USA) on the LightCycler 480 II System (Roche

253

Molecular Diagnostics, Pleasanton, CA, USA). Amplification signal curves were

254

analyzed at absorption wavelengths of 530 nm. Appropriate positive and non-

255

template controls were included in each test run. Relative expression of the pumA

256

gene was normalized with expression values obtained from 16S rRNA genes from

257

P. aeruginosa (accession number AE004091; region: 722096–723631) or E. coli

258

(accession number AJ605115). Estimation of relative gene expression was

AC C

EP

TE D

M AN U

SC

RI PT

239

11

ACCEPTED MANUSCRIPT

performed via the classical calibration dilution curve and slope calculation. A five-

260

fold dilution series (500–0.05 ng total RNA) was prepared and utilized as sample in

261

the RT-qPCR. Efficiency (E) was obtained from standard curves utilizing the

262

formula E = (10 −1/slope–1) × 100. Relative expression levels were determined with

263

the efficiency correction method, which takes into account amplification efficiencies

264

between target and reference genes (Pfaffl 2001).

SC

265

RI PT

259

2.9. Virulence assays

267

2.9.1. Caenorhabditis elegans killing assays

268

C. elegans Bristol N2 (Brenner 1974) worms were synchronized by hypochlorite

269

isolation of eggs from gravid adults, followed by hatching of eggs in S-basal

270

medium (Stiernagle 2006). L1 larvae were transferred onto NGM (Nematode

271

Growth Medium) (Stiernagle 2006) in plates seeded with the E. coli OP50 strain

272

previously grown on the plates as a food source and incubated at 20°C for 4–5

273

days until they reached the young adult phase. Worms were rinsed off the plates

274

and washed in S-basal medium. Cultures of P. aeruginosa strains were grown

275

overnight in M9 broth at 37°C, centrifuged, and the pellets were suspended in fresh

276

M9 broth and grown to an OD600 = 0.4. For each experiment, 20 worms were

277

dispensed into each well of a 24-well Costar plate (Corning, Inc., Corning, NY,

278

USA). The worms were incubated with 1 × 105 (±1 × 104) CFU of the bacterial

279

strain contained in a total volume of 1 ml, and the plates were incubated for 36 h at

280

20°C and scored for live worms at 6, 12, 24, and 36 h. P. aeruginosa PAO1

281

(pUCP20) and M9 medium were used as a positive and negative control,

282

respectively. A worm was considered dead when it no longer responded (moved)

AC C

EP

TE D

M AN U

266

12

ACCEPTED MANUSCRIPT

to a touch stimulus. Worms that died as a result of becoming adhered to the wall of

284

the plate were excluded from the analysis. For all assays, bacterial concentrations

285

were confirmed by dilution, plating, and counting. Two independent assays with

286

two replicates were conducted for each worm group. By supernatant effects assays

287

the cultures of P. aeruginosa strains were grown overnight in M9 broth at 37°C and

288

centrifuged to obtain the supernatant. The worms were incubated with 100 µl of the

289

bacterial strain supernatant in a total volume of 1 ml and the plates were incubated

290

for 36 h at 20°C and scored for live worms at 6, 12, 24, and 36 h. Two independent

291

assays with two replicates per experiment were conducted for each worm group.

292

PumA protein toxicity on C. elegans was determined by His-PumA purified protein

293

incubation. A total of 20 worms were dispensed into each well of 24-well plates

294

(Corning) and incubated with 10, 25, 50, or 100 µg ml–1 of His-PumA protein in a

295

volume of 100 µl, and the plates were incubated for 48 h at 20°C and scored for

296

live worms at 6, 12, 24, 36, and 48 h. C. elegans was incubated in M9 medium,

297

and M9 containing protein elution buffer with or without 0.2 M of Imidazole were

298

used as negative controls. Additionally, His-PumA protein toxicity to C. elegans

299

was analyzed after their denaturation by heating at 95°C for 2 h or by hydrolysis

300

utilizing 100 µg ml–1 of Pronase E (Sigma-Aldrich). Finally, the PumA toxicity effect

301

neutralized by PumB protein was determined by C. elegans incubation in presence

302

of His-PumA and PumB-His purified proteins in a ratio 1:1 and 1:2. Two

303

independent experiments with two replicates were performed for each worm group.

AC C

EP

TE D

M AN U

SC

RI PT

283

304 305

2.9.2. Mouse virulence experiments

13

ACCEPTED MANUSCRIPT

The protocol for mouse manipulation assays followed the recommendations for

307

animal manipulation issued by SAGARPA (Secretaría de Agricultura, Ganadería,

308

Desarrollo Rural, Pesca y Alimentación, México), 2001.

309

To assess the virulence and colonization ability of Pseudomonas strains, groups of

310

10, 8–10-week-old male BALB/c mice weighing 18–22 g were used. The virulence

311

assays were conducted as described by Rodríguez-Andrade et al. (2016) and

312

inoculated (100 µl) with as single dose administered by tail-vein injection with 3 ×

313

106 (±3 × 105) CFU. Mouse survival was monitored each day. Two independent

314

assays were conducted for each group.

315

To assess the colonization ability of P. aeruginosa PAO1 bearing pumAB genes,

316

the mice were anesthetized with sodium pentobarbital (24 mg/kg) and inoculated

317

via tail-vein administration with 5 × 106 (±5 × 105) CFU contained in a volume of

318

100 µl as described by Rodríguez-Andrade et al. (2016). At 48 h post-inoculation,

319

the mice were euthanized and samples were taken for analysis. Heart, lungs, and

320

kidneys were removed by dissection, weighed, and homogenized in 10 mM MgSO4

321

utilizing a mortar and pestle. Organ homogenates were serially diluted in 10 mM

322

MgSO4 and plated on LB agar for CFU determination. Three independent assays

323

were conducted for each mouse group. For all assays, bacterial inoculum

324

concentrations were confirmed by dilution, plating, and CFU determination.

SC

M AN U

TE D

EP

AC C

325

RI PT

306

326

2.10. Statistical analysis

327

Data analysis was carried out with GraphPad Prism 5 software. ANalysis Of

328

VAriance (ANOVA) employing the Tukey test was performed for all Pseudomonas

14

ACCEPTED MANUSCRIPT

329

assays. In addition, a Student t test was carried out for the analysis of the

330

expression assays. P <0.05 was considered statistically significant.

332

3. Results and Discussion

333

RI PT

331

3.1. The pumAB Genes of Plasmid pUM505

335

Owing to their similarity with previously characterized proteins, ORF 123 and 124

336

from the pathogenicity island of pUM505 were reported to encode a probable TA

337

system (Ramírez-Díaz et al. 2011). These ORF were renamed as pumA and pumB

338

genes for the pUM505 Toxin–Antitoxin system. PumA is a predicted protein of 99

339

aa (GenBank accession number AEQ93530) that was previously reported to

340

exhibit 55% sequence identity with the RelE protein (toxic component) of the TA

341

system from Klebsiella pneumoniae (ZP_06016753) (Ramírez-Díaz et al. 2011). A

342

recent BLAST analysis showed that the pumA gene encodes a protein that is 73%

343

identical to an uncharacterized addiction module killer protein from Pseudomonas

344

putida (NCBI Reference Sequence WP_011005862). Additionally, an amino acid

345

sequence alignment with proteins of the RelE toxin superfamily with known

346

biological function was conducted. The alignment revealed that the PumA protein

347

possesses only 22% identity to MjRelE from Methanococcus jannaschii, 18% to

348

YoeB from E. coli, and 14% to PhRelE from Pyrococcus horikoshii (Fig. S1,

349

Supplementary Material). The low identity of PumA with RelE-like proteins is not

350

surprising, in that it has been reported that amino acid sequences of RelE-like

351

proteins are quite divergent, even those originating from the same or closely

352

related bacterial species (Goeders et al. 2013). In general, RelE-like sequences

AC C

EP

TE D

M AN U

SC

334

15

ACCEPTED MANUSCRIPT

present 24–47% similarity among themselves and 15–28% identity with the

354

canonical RelE protein from E. coli K-12 (Francuski and Saenger 2009). Although

355

highly divergent at the primary sequence level, the PumA protein contains a

356

secondary structure predicted by the MINNOU program that is characteristic of

357

toxins of the RelE superfamily, constituted by five-stranded β-sheet flanked on one

358

side by α-helices α1 and α2 and, on the other side, by α-helix α3 (Francuski and

359

Saenger 2009) (Fig. S1, Supplementary Material). It has been reported that,

360

despite very low conservation at the amino acid sequence level, RelE-like proteins

361

share secondary and tertiary structures with similar folding (Goeders et al. 2013).

362

Additionally, PumA contains conserved-type residues (Val4, Val41, Val44, Ile53,

363

Val61, and Val81) and highly-conserved residues (Asp40, Arg52, His54, and

364

Arg60) that are found in the MjRelE and YoeB proteins (Fig. S1, Supplementary

365

Material). The mutation of Arg62 in MjRelE, which corresponds to Arg65 of

366

PhRelE, renders a toxin inactive (Francuski and Saenger 2009); the Arg62 residue

367

located in MjRelE corresponds to Arg60 in the PumA sequence. This analysis

368

suggested that PumA could belong to the RelE toxin family.

369

The pumB gene encodes a putative transcriptional regulator of 96 aa (GenBank

370

accession number AEQ93531) with 92% identity to the antitoxin HTH protein

371

(Helix-Turn-Helix xenobiotic-resistant family-like proteins) from Pseudomonas

372

alcaligenes (AAD40346) (Ramírez-Díaz et al. 2011). The PumB protein contains a

373

predicted HTH DNA-binding domain (Fig. S2, Supplementary Material) that is

374

characteristic of the Xre family of antitoxins, and this domain has often been found

375

in characterized antitoxins (Yamaguchi et al. 2011). In type II TA systems, the

AC C

EP

TE D

M AN U

SC

RI PT

353

16

ACCEPTED MANUSCRIPT

antitoxin is generally a DNA-binding protein that binds to the operator of the operon

377

to repress its own transcription. In the majority of cases, the HTH motif is located at

378

the N-terminal domain of the antitoxins (Chan et al. 2015). Examples of the

379

existence of the HTH motif in solved structures of antitoxins include PezA (Khoo et

380

al. 2007), as well as HigA (Schurek et al. 2014). However, the HTH motif of PumB

381

was located at the C-terminal region of the protein (Fig. S2, Supplementary

382

Material). A similar position of the HTH motif was found in the functional antitoxin

383

MqsA of E. coli (Brown and Shaw 2003). This analysis suggests that the PumB

384

protein could function as an HTH-type antitoxin protein.

385

As mentioned previously, the pumA gene encodes a putative RelE-type toxin, while

386

the pumB gene encodes a putative HTH-type antitoxin, and it is suggested that

387

these genes encode a hybrid TA system. In general, it has been assumed that

388

each toxin family is associated with a specific antitoxin family; for instance, RelE

389

proteins are frequently associated with RelB antitoxins (Grady and Hayes 2003).

390

However, many hybrid systems exist in which a TA locus contains a toxin of one

391

class and an antitoxin of another class (Anantharaman and Aravind 2003).

392

Together, these results suggest that pumAB genes from plasmid pUM505 encode

393

a putative TA hybrid system.

SC

M AN U

TE D

EP

AC C

394

RI PT

376

395

3.2. pumAB Genes Form an Operon

396

The pumA and pumB genes exhibited a genetic arrangement where pumA, which

397

encodes a putative toxin, was located upstream of pumB, which encodes an

398

antitoxin (Fig. 1a). In the majority of TA systems, antitoxin genes are located

399

upstream of their cognate toxin genes, suggesting that antitoxins appear to 17

ACCEPTED MANUSCRIPT

possess an advantage over their cognate toxins with regard to their production

401

(Yamaguchi et al. 2011). However, there are exceptional cases in which toxin

402

genes are located upstream of their cognate antitoxin genes, suggesting that these

403

differences in gene order might exert a subtle effect on the toxin-to-antitoxin ratio in

404

the cell, which in turn regulates toxin function (Yamaguchi et al. 2011). Additionally,

405

sequence analysis revealed that pumA and pumB genes contain a putative σ70

406

promoter that showed at the non-typical –35 to –10 box in the 5’ region of pumA

407

(Fig. 1a); no promoter region was identified for pumB, suggesting that these two

408

genes constitute an operon that could be expressed from the pumA promoter. In

409

type II TA systems, toxin and antitoxin genes are commonly co-transcribed,

410

forming operons (De la Cruz et al. 2013), suggesting an evolutionary advantage of

411

the close vicinity of coding regions. Additionally, pumAB genes have overlapped

412

coding regions; the stop codon of pumA and the initiation codon of pumB share

413

four nucleotides (Fig. 1a). Overlapping of coding sequences is a conserved feature

414

of paired TA systems (Leplae et al. 2011).

415

To investigate the existence of a possible transcriptional linkage of the pumAB

416

genes, RT-PCR analyses were performed. Primers were designed to amplify

417

complementary DNA (cDNA) synthesized from a transcript spanning the

418

overlapping pumAB region (Fig. 1a). When total RNA from exponential-phase

419

cultures of the P. aeruginosa PAO1 (pUM505) strain was probed, a sole ~440-bp

420

DNA fragment, covering the overlapping region, was detected (Fig. 1b, lane 2). No

421

corresponding fragment was detected in the negative control (Fig. 1b, lane 1). A

422

less intense DNA band was observed with RNA obtained from stationary-phase

AC C

EP

TE D

M AN U

SC

RI PT

400

18

ACCEPTED MANUSCRIPT

cultures (Fig. 1b, lane 3), which is in agreement with previous reports of TA

424

systems displaying higher expression levels in the exponential growth phase (Nolle

425

et al. 2013). These data demonstrate that the pumAB genes are co-transcribed into

426

a bicistronic mRNA, and strongly suggest that they form an operon.

427

RI PT

423

3.3. The pumAB Genes Confer Plasmid Stability

429

Since the function of plasmid-encoded TA systems is related to plasmid

430

stabilization, we investigated whether pumAB genes are able to promote plasmid

431

maintenance. Because pUM505 additionally to possess the putative TA system

432

contains several genes probably related with the partition, segregation,

433

maintenance, and stabilization of plasmids, as well as parA, parB, xerD, xerC, and

434

ssb, respectively, (Ramírez-Díaz et al. 2011), in order to test this, pumAB genes

435

were first cloned into the Ampicillin (Ap)-resistance pJET vector and the

436

recombinant plasmid pJET-pumAB obtained was transferred to the E. coli JM101

437

strain. Plasmid segregation was tested by measuring CFU in E. coli JM101 (pJET-

438

pumAB) and JM101 (pJET-orf131) strains after 24-h consecutive subcultures in

439

non-selective NB medium; samples were withdrawn each day, diluted, and plated

440

on nutrient agar with or without Ap, as described under Methods. CFU of pJET-

441

orf131-containing cells decreased in each subculture on the Ap plates from day 3

442

(corresponding to 216 generations) and, after day 6 (corresponding to 432

443

generations), these were undetectable (Fig. 2). In contrast, CFU counts in

444

transformants bearing the pJET-pumAB plasmid remained without significant

445

changes throughout the assay. Agarose gel electrophoresis of mini-prep plasmid

446

DNA isolations from the subcultures confirmed the absence of plasmid bands from

AC C

EP

TE D

M AN U

SC

428

19

ACCEPTED MANUSCRIPT

pJET-orf131 clones, as well as the presence of recombinant plasmid in clones from

448

pJET-pumAB transformants (Fig. S3, Supplementary Material). These data

449

indicate that the presence of pumAB increases the stability of the pJET plasmid

450

and suggest that they do so by means of a post-segregation killing mechanism

451

involving a TA system, probably composed of the PumA toxin and its PumB

452

cognate antitoxin.

RI PT

447

SC

453

3.4. Overexpression of the pumA Gene Inhibits E. coli Growth

455

To investigate whether a functional toxin is encoded by pumA of pUM505, this

456

gene, including its putative promoter, was subcloned into the pUCP20 shuttle

457

vector; recombinant plasmid pUC-pumA was transferred to the P. aeruginosa

458

PAO1 strain, and growth kinetics in liquid cultures was measured. The presence of

459

the pUC-pumA plasmid did not cause growth inhibition in PAO1 cells (Fig. S4a,

460

Supplemental Material). This effect was not expected because the pumA gene was

461

identified in a plasmid from a Pseudomonas strain and, as was shown, is

462

expressed in this strain together the pumB gene as an operon (Fig. 1b). However,

463

an explanation to this result may be attributed to the fact that the pumA gene is not

464

expressed in PAO1 under its own promoter into the pUC20 vector. However,

465

another possibility could be that the product encoded by the pumA gene was less

466

toxic to PAO1 than to E. coli, bearing in mind that the function of pumAB genes in

467

plasmid stability was determined in E. coli. To discern between these possibilities,

468

pumA gene expression was determined by RT-qPCR assays from cultures of P.

469

aeruginosa PAO1 containing the recombinant pUC-pumA plasmid. The results

470

showed that the pumA gene was expressed in the P. aeruginosa PAO1 strain (Fig.

AC C

EP

TE D

M AN U

454

20

ACCEPTED MANUSCRIPT

S4c, Supplemental Material). The latter suggested that this gene is functional in

472

this strain. To determine whether the product of the pumA gene affects the growth

473

of E. coli cells, recombinant plasmid pUC-pumA was transferred to E. coli JM101

474

and growth kinetics in liquid cultures were analyzed. The results demonstrated no

475

differences in the growth of E. coli cells through the presence of the pUC-pumA

476

plasmid (Fig. S4b, Supplemental Material). Additionally, the results of expression

477

by RT-qPCR analysis showed that the pumA gene was expressed in this strain

478

(Fig. S4c, Supplemental Material). These results noted that when PumA is

479

expressed in physiological levels, it does not affect cell growth, suggesting that a

480

high expression level is required. In Mycobacterium tuberculosis, a correlation

481

between the expression levels of the VapC toxin from TA systems and its ability to

482

confer a toxic phenotype has been reported (Ahidjo et al. 2011). To determine

483

whether the overexpression of PumA affects bacterial growth, the pumA gene was

484

subcloned into the pTrcHisC expression vector under the control of an IPTG-

485

inducible promoter, and recombinant plasmid pTrcHis-pumA was transferred to the

486

E. coli BL21 strain. Growth kinetics with or without IPTG induction was then

487

measured in BL21 derivatives. BL21 cells containing the vector alone, with IPTG,

488

exhibited no harmful effects on growth. In contrast, IPTG-induced BL21

489

transformants containing recombinant plasmid pTrcHis-pumA displayed clear

490

growth inhibition as compared with non-induced transformants (Fig. 3a), and these

491

results were consistent with a decrease in the CFU count (Fig. S5, Supplemental

492

Material). Additionally, pumA gene expression levels with or without IPTG induction

493

were determined, and the results showed an increase of about 70 times of the

494

pumA transcript with respect to the non-induced condition (Fig. 3b), indicating that

AC C

EP

TE D

M AN U

SC

RI PT

471

21

ACCEPTED MANUSCRIPT

overexpression of PumA affects E. coli growth. Several reports demonstrated that

496

overexpression of the toxins of TA systems produced cell growth inhibition; for

497

example, in E. coli K-12 (MG1655), RelE, MazF, YeeV, and YkfI toxins inhibited

498

bacterial growth when they were ectopically overexpressed (Pedersen et al. 2002,

499

Brown et al. 2013). Additionally, has been reported that some toxins caused cell

500

growth inhibition when they were heterologously and ectopically overexpressed in

501

E. coli BL21, as well as the VapC-3 and Tox28 toxins from Acidithiobacillus

502

ferrooxidans, YoeB from Streptococcus suis, and YoeB from Staphylococcus

503

equorum (Bustamante et al. 2014; Zheng et al. 2015; Nolle et al. 2015). Therefore,

504

our results indicated that, like others toxins from TA systems, a high expression

505

level of the pumA gene is required to affect bacterial growth, suggesting that PumA

506

functions as a toxin in E. coli. Since many genes can be toxic to cells when they

507

are overexpressed, it was important to determine whether cell growth inhibition by

508

PumA is neutralized by expressing its cognate antitoxin. We compared the growth

509

of E. coli BL21, in which the pumA was expressed either individually or together

510

with its cognate pumB, under the control of the same inducible promoter, to

511

determine whether pumB allowed cell growth in the presence of the inducer. The

512

pumAB gene pair was cloned into the pTrcHis2A vector and transferred to E. coli

513

BL21, as described in Methods. When co-expression of PumA and PumB-His

514

proteins was induced by IPTG, BL21 cells did not exhibit growth inhibition in

515

comparison with the strain with individual His-PumA overexpression (Fig. 3a).

516

These results were consistent with the CFU count (Fig. S5, Supplemental

517

Material), suggesting that PumB protein could neutralize the growth inhibitory effect

518

of PumA. Overexpression of recombinant His-PumA and PumB-His proteins was

AC C

EP

TE D

M AN U

SC

RI PT

495

22

ACCEPTED MANUSCRIPT

verified by SDS-PAGE and by Western blot analysis (Fig. S6, Supplementary

520

Material). In type II TA systems, the major role of the antitoxin is to neutralize the

521

toxic function of its cognate toxin mediated by the toxin-binding domain

522

(Yamaguchi et al. 2011). The majority of the TA systems from E. coli and

523

Mycobaterium smegmatis tested experimentally showed that the toxin inhibited

524

bacterial growth, indicating an effect that is neutralized by its cognate antitoxin

525

(Sala et al. 2014). These results indicate that PumA acts as a toxic protein in E.

526

coli and that its effects are neutralized by the presence of PumB protein, which

527

supports the notion that these proteins form a functional TA system in

528

Pseudomonas.

529

M AN U

SC

RI PT

519

3.5. Expression of the pumA Gene has the potential to Increases P.

531

aeruginosa Virulence

532

It has been reported that some strains of P. aeruginosa, such as the highly virulent

533

PA14 strain, can kill the C. elegans nematode, utilizing virulence factors that are

534

also required for pathogenesis in mammalian or plant hosts (Tan et al. 1999). In

535

addition, chromosomally encoded TA systems may be related to the virulence of

536

bacterial pathogens (Unterholzner et al. 2013; De la Cruz et al. 2013). To

537

determine whether the pumAB TA system from pUM505 participates in P.

538

aeruginosa virulence, assays to test their ability to kill C. elegans were conducted.

539

Since the pUM505 contains several genes that are related with the virulence of P.

540

aeruginosa (Rodríguez-Andrade et al. 2016), pumAB genes were cloned, either

541

paired or individually, in the pUCP20 vector, and recombinant plasmids pUC-

542

pumAB, pUC-pumA and pUC-pumB were transferred to P. aeruginosa PAO1. The

AC C

EP

TE D

530

23

ACCEPTED MANUSCRIPT

results revealed that transformants containing the pumAB genes displayed similar

544

viability levels as compared to the vector-only PAO1control strain (~60% of worms

545

killed in 36 h) (Fig. 4a), suggested that in the conditions carried out the PumAB TA

546

system is not involved in the virulence of P. aeruginosa. However, interestingly

547

transformants with the pumA gene in the same conditions assayed increased

548

PAO1 ability to kill worms at ~95% at 36-h post-incubation as compared to the

549

control strain, whereas transformants with plasmid constructions containing pumB

550

displayed similar viability levels as compared to the control strain (~60% of worms

551

killed in 36 h) (Fig. 4a). In order to rule out the possibility that the virulence effects

552

observed may be due to differences in the growth rate of the transformants

553

analyzed, growth assays were performed and the results exhibited no harmful

554

effects on growth among the transformants (Fig. S7, Supplementary Material).

555

These data indicate that the pumA gene product improves the ability of PAO1 to kill

556

C. elegans, this effect is neutralized by the pumB gene product. Additionally, the

557

effect of the supernatant from Pseudomonas strains on the C. elegans viability was

558

assayed and the results showed that the supernatant from P. aeruginosa contained

559

the pumA gene was able to kill the ~75% of worms in 36 h of incubation; whereas

560

P. aeruginosa strains contains the pumB or pumAB genes showed similar killer

561

(~25%) than the P. aeruginosa control strain (Fig. S8, Supplementary Material),

562

suggested that the toxin could be secreted to the media growth by the bacteria or

563

release from bacterial lysis to generate toxic effects on the bacterial host. In

564

Rickettsia containing vapBC TA systems has been suggested that the toxic effect

565

of on its host could be related to the release of toxin in host cell cytoplasm from a

566

TA complex that is secreted or released following rickettsial death (Audoly et al.

AC C

EP

TE D

M AN U

SC

RI PT

543

24

ACCEPTED MANUSCRIPT

2011). In addition, comparative genomic studies have shown that the presence of

568

TA systems is significantly associated with the pathogenicity of bacteria (De la

569

Cruz et al. 2013). M. tuberculosis contains at least 30 functional TA systems,

570

whereas its non-pathogenic counterpart, M. smegmatis, has three functional TA

571

systems (Sala et al. 2014). Furthermore, TA modules have been shown to be

572

involved in the virulence of Salmonella and Leptospira interrogans (De la Cruz et

573

al. 2013; Komi et al. 2015). These results suggest that a toxin from a plasmid-

574

encoded TA system might contribute to increasing the virulence of its host.

575

Additionally, to determine whether the PumA protein is able to kill C. elegans,

576

worms were incubated with increasing concentrations of purified His-PumA protein.

577

The results showed that 10 µg ml–1 of His-PumA protein was able to kill ~55% of

578

worms in 36 h, and that this effect increased at ~70% with 100 µg ml–1 of

579

recombinant protein (Fig. 4b). These harmful effects were not observed when the

580

worms were incubated only with the buffers used for protein elution in these assays

581

(Fig. S9, Supplementary Material). Similar results were previously reported to the

582

VapC toxin from the VapBC TA system of Rickettsia spp, this protein was toxic to

583

mammalian cells when injected into the cytoplasm as a free toxin (Audoly et al.

584

2011). Additionally, we determined the toxic effect of PumA on C. elegans after its

585

denaturalization. The results demonstrated that worm death levels comprised up to

586

~35 or ~45% when PumA was denaturized by heat or by protease E treatment,

587

respectively, as compared to non-denaturized protein at the same concentration

588

that was able to kill the ~70% of worms (Fig. 4c). The denaturized protein

589

decreased in lethality toward C. elegans, suggesting that the toxic effect was not

590

provided by the components of the reaction mix, since a similar worm death was

AC C

EP

TE D

M AN U

SC

RI PT

567

25

ACCEPTED MANUSCRIPT

observed between treatments and also suggest that correct structure of the protein

592

is required for its function. Finally, the neutralization of the toxicity of PumA on C.

593

elegans by the PumB protein was analyzed, and the results showed that when the

594

worms were incubated simultaneously with PumA and PumB proteins in a 1:2

595

(toxin:antitoxin) radio, the toxic effect of PumA decrease 50%; whereas, in a 1:1

596

radio no changes in the toxicity was observed (Fig. 4d), supporting the role of

597

neutralization of PumB and suggested a molar relationship between PumA and

598

PumB proteins. The ratio of the toxin:antitoxin of TA systems had been described

599

is important to determine the stoichiometry of the TA protein complex as well as to

600

regulate the transcription of the TA systems, this phenomenon was termed as

601

conditional cooperativity that reflects that excess toxin often perturbed operator

602

binding by TA complexes (Hayes and Kędzierska 2014). An example of this

603

cooperativity are the CcdAB system from F plasmid of E. coli, where both modules

604

form a (CcdA)2–(CcdB)2 complex which binds DNA to regulate the repression

605

state of the ccd operon, and the addition of CcdB extra to protein-DNA complex

606

lead an operon derepression (Afif et al. 2001). Together, these results indicated

607

that the PumA protein affects the worm viability and suggest that this protein in

608

absence of PumB possess the potential to function as a toxin. It had been

609

described that under non-stress conditions, the toxicity of RelE toxin from RelEB

610

system of E. coli is neutralized by the RelB antitoxin by forming a tight non-toxic

611

RelBE complex; but under environmental stress conditions, the antitoxin is

612

degraded by cellular proteinases and RelE is activated (Christensen et al. 2001). In

613

Streptomyces cattleya DSM 46488 it was described that the relBE2sca operon

614

transcription was induced under osmotic stress, along with the ClpP proteinase

AC C

EP

TE D

M AN U

SC

RI PT

591

26

ACCEPTED MANUSCRIPT

genes, and subsequent in vivo analysis showed that the antitoxin was degraded by

616

ClpP (Li et al. 2016). This reports suggests that in some bacterial growth conditions

617

is possible that the decrease in the antitoxins levels allow that the toxin perform its

618

toxic effects.

619

To test if pUM505 pumA gene may also has the ability to increase the virulence in

620

an animal model, a total of 3 x 106 (±3 x 105) CFU of PAO1-derivative strains were

621

individually inoculated by intravenous (i.v.) administration, and mouse survival was

622

monitored each day. It was found that mice infected with P. aeruginosa PAO1

623

(pUC-pumA) led to 20% lethality at day 3 post-inoculation, and reached 60%

624

lethality at day 12 compared with mice inoculated with P. aeruginosa (pUCP20) or

625

with P. aeruginosa (pUC-pumB), which both maintained 100% survival until day 12

626

post-inoculation (Fig. 5a). This result indicates that the toxin encode by the pumA

627

gene increase the pathogenicity of P. aeruginosa in a mammalian host.

628

Additionally, the effect of the pumA on mouse colonization was determined with the

629

previously mentioned PAO1-derivative strains, as described under Methods. The

630

presence of the pumA gene significantly increased PAO1 CFU in mouse heart

631

(two-fold), lung (six-fold), and kidney (six-fold) as compared with those inoculated

632

with transformants with the plasmid containing the pumB gene, which yielded CFU

633

counts similar to those of the empty-vector strain (Fig. 5b). To determine whether

634

the expression of the PumB antitoxin could neutralize the virulence effects supplied

635

by the presence of the pumA gene, mice were inoculated with P. aeruginosa

636

transformants containing pumAB genes. The results showed that the CFU count

637

was similar to that of the empty-vector strain or to that of the plasmid containing the

638

pumB gene (Fig. 5b). These data indicate that PumA expressed alone possess the

AC C

EP

TE D

M AN U

SC

RI PT

615

27

ACCEPTED MANUSCRIPT

ability to increases PAO1 invasiveness into mouse organs, an event neutralized by

640

the pumB product. Several chromosomal TA systems have been shown to cause

641

increased virulence in their bacterial hosts against animals, including TA systems

642

from E. coli (Norton and Mulvey 2012), Salmonella typhimurium (De la Cruz et al.

643

2013), M. tuberculosis (Sala et al. 2014), L. interrogans (Komi et al. 2015)

644

however, this relationship has not been reported for elements of plasmid-encoded

645

TA systems. These results further confirmed that pumAB genes encode a TA

646

system, comprising the PumA toxin and its cognate PumB antitoxin, being the toxin

647

the element with capacity to generate toxic effects. To our knowledge, this is the

648

first toxin from a plasmid-encoded TA system from P. aeruginosa that has

649

demonstrated participation in virulence. The increase of PAO1 virulence and the

650

invasion by PumA toward the organisms employed in this work (invertebrates and

651

mammals) suggest that the virulence mechanism involved may be conserved in

652

these evolutionarily divergent hosts. The mechanism by which the PumA toxin

653

exerts its toxic actions is currently under study.

654

In conclusion, pUM505 contains pumAB that encodes a TA system conferring

655

plasmid stability and encodes a toxin with potential to increasing P. aeruginosa

656

virulence in animal models.

SC

M AN U

TE D

EP

AC C

657

RI PT

639

658

4. Acknowledgments

659

We thank M.P. Chávez-Moctezuma and R.I. Reyes-Gallegos for their valuable

660

technical support in mouse tissue invasion assays.

661 662

5. Funding source 28

ACCEPTED MANUSCRIPT

This work was supported by grants from Coordinación de la Investigación

664

Científica (UMSNH; 2.6, 2.35) and Consejo Nacional de Ciencia y Tecnología

665

(CONACYT; 181747, 167071). KCH-R, VMC-J, MIV-M, JA P-M and SPD-P were

666

supported by postgraduate fellowships from CONACYT.

667 6. Conflicts of Interest

669

The authors declare no conflict of interest.

SC

668

RI PT

663

AC C

EP

TE D

M AN U

670

29

ACCEPTED MANUSCRIPT

671

Supplementary Table

672

Table S1. Primers and probes used in this study.

673 Supplementary Figures

675

Figure S1. Multiple sequence alignment of RelE homologues.

676

Figure S2. Multiple sequence alignment of HTH-type antitoxin homologues.

677

Figure S3. Isolation of plasmids from E. coli JM101 transformants.

678

Figure S4. Effects of pumA gene expression on bacterial growth.

679

Figure S5. Viability of E. coli cells by overexpression of pumA or pumAB co-

680

expression.

681

Figure S6. Overexpression and detection of recombinant His-PumA and PumB-His

682

proteins.

683

Figure S7. Growth kinetics of P. aeruginosa strains.

684

Figure S8. Effects of supernanats of P. aeruginosa (pUC-pumA) on the

685

Caenorhabditis elegans survival.

686

Figure S9. Effects of buffers employed for protein elution on the Caenorhabditis

687

elegans survival.

689

SC

M AN U

TE D

EP

AC C

688

RI PT

674

30

ACCEPTED MANUSCRIPT

References

691

Afif, H., Allali, N., Couturier, M., Van Melderen, L. 2001. The ratio between CcdA

692

and CcdB modulates the transcriptional repression of the ccd poison-antidote

693

system. Mol Microbiol 41, 73-82.

RI PT

690

694

Ahidjo, B.A., Kuhnert, D., McKenzie, J.L., Machowski, E.E., Gordhan, B.G., Arcus,

696

V., Abrahams, G.L., Mizrahi, V. 2011. VapC toxins from Mycobacterium

697

tuberculosis are ribonucleases that differentially inhibit growth and are neutralized

698

by cognate VapB antitoxins. PLOS ONE 6, e21738.

699

M AN U

SC

695

Anantharaman, V., Aravind, L. 2003. New connections in the prokaryotic toxin-

701

antitoxin network: Relationship with the eukaryotic nonsense-mediated RNA decay

702

system. Genome Biol 4, R81.

703

TE D

700

Aranda, J., Garrido, M.E., Cortés, P., Llagostera, M., Barbé, J. 2008. Analysis of

705

the protective capacity of three Streptococcus suis proteins induced under divalent-

706

cation-limited conditions. Infect Immun 76, 1590-1598.

AC C

707

EP

704

708

Audoly, G., Vincentelli, R., Edouard, S., Georgiades, K., Mediannikov, O.,

709

Gimenez, G., Socolovschi, C., Mège, J.L., Cambillau, C., Raoult, D. 2011 Effect of

710

rickettsial toxin VapC on its eukaryotic host. PLOS ONE 6, e26528.

711 712

Brenner, S. 1974. The genetics of Caenorhabditis elegans. Genetics 77, 71-94.

713 31

ACCEPTED MANUSCRIPT

714

Brown, B.L., Lord, D.M., Grigoriu, S., Peti, W., Page, R. 2013. The Escherichia coli

715

toxin MqsR destabilizes the transcriptional repression complex formed between the

716

antitoxin MqsA and the mqsRA operon promoter. J Biol Chem 288, 1286-1294.

RI PT

717 718

Brown, J.M., Shaw, K.J. 2003. A novel family of Escherichia coli toxin-antitoxin

719

gene pairs. J Bacteriol 185, 6600-6608.

SC

720

Bustamante, P., Tello, M., Orellana, O. 2014. Toxin-Antitoxin Systems in the

722

Mobile Genome of Acidithiobacillus ferrooxidans. PLoS One 9, e112226.

M AN U

721

723

Cervantes, C., Ohtake, H., Chu, L., Misra, T.K., Silver, S. 1990. Cloning, nucleotide

725

sequence and expression of the chromate resistance determinant of Pseudomonas

726

aeruginosa plasmid pUM505. J Bacteriol 172, 287-291.

TE D

724

727

Chan, W.T., Balsa, D., Espinosa, M. 2015. One cannot rule them all: Are bacterial

729

toxins-antitoxins druggable? FEMS Microbiol Rev 39, 522-540.

730

EP

728

Christensen, S.K., Mikkelsen, M., Pedersen, K., Gerdes, K. 2001. RelE, a global

732

inhibitor of translation, is activated during nutritional stress. Proc Natl Acad Sci U S

733

A 98, 14328-14333.

734

AC C

731

735

De la Cruz. M.A., Zhao, W., Farenc, C., Giménez, G., Raoult, D., Cambillau, C.,

736

Gorvel, J.P., Méresse, S. 2013. A toxin-antitoxin module of Salmonella promotes

737

virulence in mice. PLOS Pathog 9 (12):e1003827. 32

ACCEPTED MANUSCRIPT

Díaz-Magaña, A., Alva-Murillo, N., Chávez-Moctezuma, M.P., López-Meza, J.E.,

739

Ramírez-Díaz, M.I., Cervantes, C. 2015. A plasmid-encoded UmuD homologue

740

regulates expression of Pseudomonas aeruginosa SOS genes. Microbiology 161,

741

1516-1523.

RI PT

738

742

Francuski, D., Saenger, W. 2009. Crystal structure of the antitoxin-toxin protein

744

complex RelB-RelE from Methanococcus jannaschii. J Mol Biol 393, 898-908.

SC

743

745

Goeders, N., Drèze, P.L., Van Melderen, L. 2013. Relaxed cleavage specificity

747

within the RelE toxin family. J Bacteriol 195, 2541–2549.

M AN U

746

748

Grady, R., Hayes, F. 2003. Axe-Txe, a broad-spectrum proteic toxin antitoxin

750

system specified by a multidrug-resistant, clinical isolate of Enterococcus faecium.

751

Mol Microbiol 47, 1419-1432.

752

TE D

749

Green, M.R., Sambrook, J. 2012. Molecular Cloning: a Laboratory Manual, 4th ed.

754

Cold Spring Harbor, NY, USA: Cold Spring Harbor Laboratory.

AC C

755

EP

753

756

Hayes, F., Kędzierska, B. 2014. Regulating toxin-antitoxin expression: controlled

757

detonation of intracellular molecular timebombs. Toxins (Basel) 6, 337-358.

758 759

Hayes, F., Van Melderen, L. 2011. Toxins-antitoxins: diversity, evolution and

760

function. Crit Rev Biochem Mol Biol 46, 386-408.

761 33

ACCEPTED MANUSCRIPT

Khoo, S.K., Loll, B., Chan, W.T., Shoeman, R.L., Ngoo, L., Yeo, C.C., et al. 2007.

763

Molecular and structural characterization of the PezAT chromosomal toxin-antitoxin

764

system of the human pathogen Streptococcus pneumoniae. J Biol Chem 282,

765

19606-19618.

RI PT

762

766

Komi, K.K., Ge, Y.M., Xin, X.Y., Ojcius, D.M., Sun, D., Hu, W.L., Zhao, X., Lin, X.,

768

Yan, J. 2015. ChpK and MazF of the toxin-antitoxin modules are involved in the

769

virulence of Leptospira interrogans during infection. Microbes Infect 17, 34-47.

SC

767

M AN U

770 771

Leplae, R., Geeraerts, D., Hallez, R., Guglielmini, J., Drèze, P., Van Melderen, L.

772

2011. Diversity of bacterial type II toxin-antitoxin systems: a comprehensive search

773

and functional analysis of novel families. Nucleic Acids Res 39, 5513-5525.

TE D

774 775

Li, L.L., Malone, J.E., Iglewski, B.H. 2007. Regulation of the Pseudomonas

776

aeruginosa quorum-sensing regulator VqsR. J Bacteriol 189, 4367-4374.

EP

777

Li, P., Tai, C., Deng, Z., Gan, J., Oggioni, M.R., Ou, H.Y. 2016 Identification and

779

characterization of chromosomal relBE toxin-antitoxin locus in Streptomyces

780

cattleya DSM46488. Sci Rep doi:10.1038/srep32047.

781

AC C

778

782

Nolle, N., Schuster, C.F., Bertram, R. 2013. Two paralogous yefM-yoeB loci from

783

Staphylococcus equorum encode functional toxin-antitoxin systems. Microbiology

784

159, 1575-1585.

785 34

ACCEPTED MANUSCRIPT

Norton, J.P., Mulvey, M.A. 2012. Toxin-antitoxin systems are important for niche-

787

specific colonization and stress resistance of uropathogenic Escherichia coli. PLOS

788

Pathog 8, e1002954.

789

RI PT

786

790

Page, R., Peti, W. 2016. Toxin-antitoxin systems in bacterial growth arrest and

791

persistence. Nat Chem Biol 12, 208-214.

SC

792

Park, S.J., Son, W.S., Lee, B.J. 2013. Structural overview of toxin-antitoxin

794

systems in infectious bacteria: a target for developing antimicrobial agents.

795

Biochim Biophys Acta 1834, 1155-1167.

M AN U

793

796

Pedersen, K., Christensen, S.K., Gerdes, K. 2002. Rapid induction and reversal of

798

a bacteriostatic condition by controlled expression of toxins and antitoxins. Mol

799

Microbiol. 45, 501-510.

TE D

797

800

Pfaffl, M.W. 2001. A new mathematical model for relative quantification in real-time

802

RT-PCR. Nucleic Acids Res 29, 2000–2007

AC C

803

EP

801

804

Quick, J., Cumley, N., Wearn, C.M., Niebel, M., Constantinidou, C., Thomas, C.M.,

805

Pallen, M.J., Moiemen, N.S., Bamford, A., Oppenheim, B., Loman, N.J. 2014.

806

Seeking the source of Pseudomonas aeruginosa infections in a recently opened

807

hospital: an observational study using whole-genome sequencing. BMJ Open 4

808

(11):e006278.

809 35

ACCEPTED MANUSCRIPT

Rahme, L.G., Ausubel, F.M., Cao, H., Drenkard, E., Goumnerov, B.C., Lau, G.W.,

811

Mahajan-Miklos, S., Plotnikova, J., Tan, M.W., Tsongalis, J., Walendziewicz, C.L.,

812

Tompkins, R.G. 2000. Plants and animals share functionally common bacterial

813

virulence factors. Proc Natl Acad Sci U S A 97, 8815-8821.

814

RI PT

810

Ramírez-Díaz, M.I., Díaz-Magaña, A., Meza-Carmen, V., Johnstone, L., Cervantes,

816

C., Rensing, C. 2011. Nucleotide sequence of Pseudomonas aeruginosa

817

conjugative plasmid pUM505 containing virulence and heavy-metal resistance

818

genes. Plasmid 66, 7-18.

M AN U

819

SC

815

Rodríguez-Andrade, E., Hernández-Ramírez, K.C., Díaz-Peréz, S.P., Díaz-

821

Magaña, A., Chávez-Moctezuma, M.P., Meza-Carmen ,V., Ortíz-Alvarado, R.,

822

Cervantes, C., Ramírez-Díaz, M.I. 2016. Genes from pUM505 plasmid contribute

823

to Pseudomonas aeruginosa virulence. Antonie Van Leeuwenhoek. 109, 389-396.

824

TE D

820

Sala, A., Bordes, P., Genevaux, P. 2014. Multiple Toxin-Antitoxin systems in

826

Mycobacterium tuberculosis. Toxins 6, 1002-1020.

AC C

827

EP

825

828

Schureck, M.A., Maehigashi, T., Miles, S.J., Marquez, J., Cho, S.E., Erdman, R., et

829

al. 2014. Structure of the Proteus vulgaris HigB-(HigA)2-HigB toxin-antitoxin

830

complex. J. Biol. Chem. 289, 1060–1070.

831 832

Sousa, A.M., Pereira, M.O. 2014. Pseudomonas aeruginosa diversification during

833

infection development in cystic fibrosis lungs- a review. Pathogens 18, 680-703. 36

ACCEPTED MANUSCRIPT

834 835

Stiernagle, T. 2006. Maintenance of C. elegans. Worm book 11, 1-11.

836 Tan, M.W., Mahajan-Miklos, S., Ausubel, F.M. 1999. Killing of Caenorhabditis

838

elegans by Pseudomonas aeruginosa used to model mammalian bacterial

839

pathogenesis. Proc Natl Acad Sci U S A 96, 715-720.

SC

840

RI PT

837

Unterholzner, S.J., Poppenberger, B., Rozhon, W. 2013. Toxin-antitoxin systems:

842

Biology, identification, and application. Mob Genet Elements 1:3:e26219.

M AN U

841

843

Watanabe, S., Takada, Y. 2004. Amino acid residues involved in cold adaptation of

845

isocitrate lyase from a psychrophilic bacterium, Colwellia maris. Microbiology 150,

846

3393-3403.

847

TE D

844

West, S.E., Schweizer, H.P., Dall, C., Sample, A.K., Runyen-Janecky, L.J. 1994.

849

Construction of improved Escherichia-Pseudomonas shuttle vectors derived from

850

pUC18/19 and sequence of the region required for their replication in

851

Pseudomonas aeruginosa. Gene 148, 81-86.

AC C

852

EP

848

853

Yamaguchi, Y., Park, J.H., Inouye, M. 2011. Toxin-antitoxin systems in bacteria

854

and archaea. Annu Rev Genet 45, 61-79.

855

37

ACCEPTED MANUSCRIPT

Yanisch-Perron, C., Vieira, J., Messing, J. 1985. Improved M13 phage cloning

857

vectors and host strains: nucleotide sequences of the M13mp18 and pUC19

858

vectors. Gene 33, 103-119.

859

RI PT

856

Zheng, C., Xu, J., Ren, S., Li, J., Xia, M., Chen, H., Bei, W. 2015. Identification and

861

characterization of the chromosomal yefM-yoeB toxin-antitoxin system of

862

Streptococcus suis. Sci Rep. 14, 13125.

SC

860

AC C

EP

TE D

M AN U

863

38

ACCEPTED MANUSCRIPT

864

Figure Legends

865 Figure 1. Schematic representation and transcriptional analysis of pumAB

867

genes from pUM505. (a) The coding regions of the pumAB genes are depicted by

868

large, gray arrows indicating transcription direction; size of predicted polypeptides

869

(aa) is also shown. Black arrows indicate positions of the oligonucleotides

870

described in the text. Location and nucleotide sequence of the –10 and –35

871

regions (below) of the putative Promoter (P) are shown. The nucleotide sequences

872

at the boundaries of the coding regions of pumA and pumB genes are depicted in

873

the upper part. The start codon of pumB (bold) and the stop codon of pumA

874

(underscored) are highlighted. (b) RT-PCR assays were carried out as described

875

under Methods, and amplified fragments were separated in agarose gels. These

876

assays used complementary DNA (cDNA) from exponential phase (lane 2) or

877

stationary-phase (lane 3) cultures of P. aeruginosa PAO1 (pUM505) utilizing

878

primers OD1 and OR1 for the pumAB intergenic region. M, Molecular size markers;

879

PCR was performed with total RNA in the absence of RT (negative control, lane 1)

880

and with genomic DNA templates (positive control, lane 4). An arrow to the right

881

signals the position of the 0.44-kb amplification band.

SC

M AN U

TE D

EP

AC C

882

RI PT

866

883

Figure 2. Plasmid stability conferred by pumAB genes from pUM505. Cultures

884

were grown in Nutrient Broth at 37°C with shaking, and consecutive 24-h

885

subcultures were performed without antibiotic selection. Serial dilutions from each

886

subculture were plated on Nutrient agar medium with Ampicillin, and CFU were

39

ACCEPTED MANUSCRIPT

counted as described under Methods. Symbols: (○ ○), E. coli JM101 (pJET-orf131)

888

as a negative control; (●), E. coli JM101 (pJET-pumAB). Data shown are the

889

means of three independent assays in duplicate with Standard Error (SE) bars

890

shown.

RI PT

887

891

Figure 3. Overexpression of pumA inhibits bacterial growth and the PumB

893

protein neutralized this effect. (a) Cultures were grown at 37°C with shaking in

894

Nutrient Broth, and growth was monitored at intervals as Optical Density (OD) at

895

600 nm (OD600) with a spectrophotometer. Symbols: (♦), E. coli BL21 (pTrcHisC)

896

with 1 mM IPTG; (●) or (○ ○), E. coli BL21 (pTrcHis-pumA) with or without 1 mM

897

IPTG, respectively; (■) and (□), E. coli BL21 (pTrcHis-pumAB) with or without 1

898

mM IPTG, respectively. The data shown are means from the duplicates of three

899

independent assays with Standard Error (SE) bars shown. (b) Expression levels of

900

pumA from cultures of E. coli BL21 (pTrcHis-pumA) strain with or without 1 mM

901

IPTG determined by RT-qPCR as described under Methods. Values represent the

902

means of two independent determinations in duplicate with Standard Error (SE)

903

bars shown, normalized with respect to transcription of the 16S gene from E. coli.

904

Statistical analysis was conducted with Student t test to compare the expression,

905

and significant differences (p <0.0001) are indicated by asterisks (***).

M AN U

TE D

EP

AC C

906

SC

892

907

Figure 4. The product of pumA gene kills Caenorhabditis elegans. (a) Effects

908

of P. aeruginosa transformants with pumAB genes on C. elegans viability. Groups

909

of 20 worms were incubated with 1 × 105 (± 1 × 104) Colony-Forming Units (CFU) 40

ACCEPTED MANUSCRIPT

of the indicated strains, as described under Methods. Mixtures were incubated at

911

20°C and scored for live worms at the indicated times. Symbols: (◊), M9 medium; (

912

●), P. aeruginosa PAO1 pUCP20; (♦), P. aeruginosa PAO1 (pUC-pumAB); (▲), P.

913

aeruginosa PAO1 (pUC-pumA), and (▼), P. aeruginosa PAO1 (pUC-pumB). (b)

914

Effects of the PumA protein on C. elegans viability. Groups of 20 worms were

915

incubated with purified His-PumA as described under Methods. Mixtures were

916

incubated at 20°C and scored for live worms at the indicated times. Symbols: (○),

917

M9 medium with 0.2 M Imidazole; (▼), (♦), ( ), and (●), M9 medium with 10, 25,

918

50, or 100 µg ml–1 of His-PumA, respectively. (c) Effects of previously denaturized

919

His-PumA on C. elegans viability. Symbols: (◊), M9 medium; (○), M9 medium with

920

0.2 M Imidazole; (●), M9 medium with 100 µg ml–1 of His-PumA; (▲), and (♦), M9

921

medium with 100 µg ml–1 of His-PumA denaturized by Pronase E or by heat,

922

respectively. (d) Neutralization of the PumA protein effects on C. elegans viability

923

by PumB. Groups of 20 worms were incubated with purified His-PumA with or

924

without PumB as described under Methods. Mixtures were incubated at 20°C and

925

scored for live worms at the indicated times. Symbols: (○), M9 medium with 0.2 M

926

Imidazole; (●), M9 medium with 5 µg ml–1 of His-PumA, (▲) and (▼), M9 medium

927

with 5 µg ml–1 of His-PumA with 1:1 or 1:2 (His- PumA:PumB-His) ratio,

928

respectively. The data shown are means from the duplicates of two independent

929

assays.

AC C

EP

TE D

M AN U

SC

RI PT

910

930 931

Figure 5. Mouse virulence by Pseudomonas aeruginosa strains. (a) Bacterial

932

virulence in mice was determined performing inoculations with a total of 3 × 106 (±3 41

ACCEPTED MANUSCRIPT

× 105) Colony-Forming Units (CFU) by intravenous (i.v.) administration. Mouse

934

survival was monitored daily for 12 days. Symbols, mice inoculated with: (○), P.

935

aeruginosa PAO1 (pUCP20); ( ) P. aeruginosa PAO1 (pUC-pumB), and (●), P.

936

aeruginosa PAO1 (pUC-pumA). Plots show the mean data pulled from two

937

independent experiments expressed as percentages of mice surviving post-

938

infection in each experimental group. (b) Mouse invasivity by P. aeruginosa strains.

939

Groups of 10, 8–10-week-old male BALB/c mice were infected with 5 × 106 (±5 ×

940

105) CFU of the indicated strains, as described under Methods. At 48 h post-

941

inoculation, heart, lungs, and kidneys were removed by dissection and their

942

homogenates were serially diluted and plated on Luria Broth agar for determination

943

of CFU counts. Control corresponds to the PAO1 pUCP20 strain. Values are the

944

means of three independent experiments conducted for each mouse group. One-

945

way ANalysis Of VAriance (ANOVA) with Tukey post-hoc test was used to

946

compare the strains, and significant differences (p <0.05) are indicated with the

947

letters a and b.

SC

M AN U

TE D

EP AC C

948

RI PT

933

42

AC C

EP

Figure 1.

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

10

RI PT

6

SC

4

2

0 1

2

M AN U

CFU (x 108 ml-1)

8

3

4

5

AC C

EP

Figure 2.

TE D

Subcultures (days)

6

AC C EP TE D B

L2 1

(p T

0.01 0 2 4

Figure 3.

Time (h)

rc H

is -p

um

A

)+

)

IP TG

A

6

(p T

RI PT

5

um

10

L2 1

1

B

b)

is -p

a)

rc H

0.1

Relative e xpression (x 10-3)

SC 0

M AN U

O.D. (600 nm)

ACCEPTED MANUSCRIPT

15 ***

ACCEPTED MANUSCRIPT

b)

75

75

50

50

25

25

SC

100

% Survival

100

RI PT

a)

0

0 12

24

0

36

12

24

36

12

24

36

M AN U

0

Time (h)

c)

d)

100

100

75

TE D

% Survival

75

50

0

12

AC C

0

EP

25

Figure 4.

24

50

25

0 36

0

Time (h)

AC C

Figure 5.

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

Highlights: pUM505 contains the pumAB genes that encode a non conventional TA system. The PumAB TA system plasmid-encoded confers plasmid stability.

RI PT

The PumA protein has the potential to increase the virulence of Pseudomonas aeruginosa.

AC C

EP

TE D

M AN U

SC

PumB protein neutralizes the negative effects of PumA protein.