JB Accepts, published online ahead of print on 2 September 2014 J. Bacteriol. doi:10.1128/JB.01824-14 Copyright © 2014, American Society for Microbiology. All Rights Reserved.

1 2

Bicarbonate increases binding affinity of Vibrio cholerae ToxT to virulence gene promoters

3 4 5

Joshua J. Thomson and Jeffrey H. Withey#

6

Department of Immunology and Microbiology

7

Wayne State University School of Medicine, Detroit, MI 48201

8 9 10

# To whom correspondence should be addressed: Jeffrey H. Withey, Department of Immunology

11

and Microbiology, Wayne State University, 540 E. Canfield, Detroit, MI 48201, USA, Tel.:

12

(313) 577-1316; Fax: (313) 577-1155; E-mail: [email protected]

13 14 15 16

Running Title: Bicarbonate increases ToxT DNA binding affinity

17 18 19 20 21 22 23 24 1

25

ABSTRACT

26

The major Vibrio cholerae virulence gene transcription activator, ToxT, is responsible for

27

production of the diarrhea-inducing cholera toxin (CT), and the major colonization factor, toxin

28

co-regulated pilus (TCP). In addition to the two primary virulence factors mentioned, ToxT is

29

responsible for activation of accessory virulence genes such as aldA, tagA, acfA, acfD, tcpI, and

30

tarAB. ToxT activity is negatively modulated by bile and unsaturated fatty acids found in the

31

upper small intestine. Conversely, previous work identified another intestinal signal, bicarbonate,

32

which enhances the ability of ToxT to activate production of CT and TCP. The work presented

33

here further elucidates the mechanism for enhancement of ToxT activity by bicarbonate.

34

Bicarbonate was found to increase activation of ToxT-dependent accessory virulence promoters

35

in addition to those that produce CT and TCP. Bicarbonate is taken up into the V. cholerae cell,

36

where it positively affects ToxT activity by increasing DNA binding affinity for the virulence

37

gene promoters that ToxT activates, regardless of toxbox configuration. The increase in ToxT

38

binding affinity in the presence of bicarbonate explains the elevated level of virulence gene

39

transcription.

40 41 42 43 44 45 46 47

2

48

INTRODUCTION

49

Vibrio cholerae has two distinct phases in its life cycle. In the planktonic state in the

50

aquatic environment, V. cholerae represses expression of pathogenesis genes, while increasing

51

transcription of genes involved in environmental survival, such as those required for motility (1,

52

2). Upon entry into a human host, via consumption of contaminated water or food, the bacteria

53

encounter signals resulting in an inversion of their transcriptome profile. In this new, virulent

54

state, expression of genes involved in motility and environmental survival is repressed, while

55

expression of genes involved in host survival and pathogenesis is activated, resulting in the

56

diarrheal disease cholera (1, 2). Virulence is controlled by a cascade of positive transcription

57

regulators known as the ToxR regulon (3). Many of the signals required to initiate change in the

58

transcriptional profile have been identified in recent years (1, 2, 4).

59

The main signals that induce transcription of V. cholerae pathogenesis genes in the host

60

act on different steps in the virulence regulatory cascade. At the uppermost level of the cascade,

61

AphA and AphB activate transcription of tcpPH (5-7). AphB activity is enhanced by both low

62

pH and anaerobic conditions (8, 9), such as would be found as V. cholerae transitions through

63

the stomach and into the upper small intestine. The next level of the cascade includes the

64

aforementioned TcpP, along with the constitutively produced ToxR, both of which are integral

65

membrane proteins that work in tandem to activate toxT transcription (10). TcpP activity has

66

recently been shown to be enhanced by taurocholate (11), a bile salt, and interaction between

67

TcpP and ToxR is enhanced under oxygen-limiting conditions (12). The final level of the

68

cascade consists of ToxT, which directly activates production of the major virulence factors

69

cholera toxin (CT) and toxin co-regulated pilus (TCP) (13-16). ToxT also directly controls

70

transcription of other accessory virulence genes including aldA, tagA, acfA, acfD, tcpI, tarA, and

3

71

tarB (17-21). ToxT activity is reduced in the presence of the unsaturated fatty acid components

72

of bile (22, 23) and enhanced in the presence of bicarbonate (24), both of which are present in

73

the upper small intestine where V. cholerae preferentially colonizes.

74

ToxT is a 276 amino acid member of the AraC/XylS family of transcriptional regulators.

75

The ToxT C-terminal domain contains two helix-turn-helix motifs responsible for DNA binding

76

(25). ToxT activates virulence genes by binding to a 13bp degenerate DNA sequence called a

77

toxbox (19). All ToxT-dependent virulence genes have two toxboxes upstream of their

78

transcriptional start site in either direct or inverted repeat configurations, with the exception of

79

aldA, which has only a single toxbox (19). ToxT can bind toxboxes as a monomer (18, 19, 26),

80

but is presumed to dimerize to fully activate at least some virulence genes (27-31).

81

Activation of the ToxT-dependent virulence genes can be altered in the presence of the

82

positive and negative ToxT effectors. After egress from the stomach into the intestine, the

83

bacteria encounter high concentrations of both bile and bicarbonate. As mentioned above, the

84

unsaturated fatty acid components of bile are negative effectors of ToxT activity (22, 27, 29, 30).

85

The N-terminus of ToxT is involved in the response to these effectors, presumably by decreasing

86

ToxT dimerization (30). Bicarbonate is present in the upper small intestine to buffer stomach

87

acid and to also protect the epithelial layer (32). Bicarbonate enhances the ability of ToxT to

88

increase CT and TCP production (24). How bicarbonate enhances ToxT activity is not well

89

understood.

90

Here, we report that bicarbonate increases transcription activation of other ToxT-

91

dependent genes in addition to the genes encoding the two major virulence factors. We show that

92

bicarbonate enters the bacterium, where it could interact with ToxT in the cytoplasm.

93

Furthermore, we have determined that the mechanism for bicarbonate-mediated enhancement of

4

94

ToxT activity is due to increased binding affinity of ToxT for the promoters it activates. This

95

increase in binding occurs at promoters of each ToxT-dependent gene, regardless of toxbox

96

configuration. This work establishes a direct mechanistic link between a signal from the host and

97

increased transcription of pathogenesis genes that occurs in the host.

98 99

MATERIALS AND METHODS

100

Bacterial strains and plasmids. Strains and plasmids used in this work are listed in

101

Table 1. All strains were maintained at -70°C in LB containing 20% glycerol. Wild-type V.

102

cholerae classical biotype strain O395 and an isogenic ∆toxT (15) with corresponding plasmids

103

were used for β-galactosidase assays. The tcpA::lacZ and ctxAB::lacZ promoter fusions were

104

constructed using pTL61t (19, 33) and contained 138bp and 76bp upstream of the transcriptional

105

start site, respectively. The pTL61t fusions aldA::lacZ and tagA::lacZ, contained 158bp and

106

92bp, respectively, upstream of their transcriptional start sites (17). Fusions acfA::lacZ and

107

acfD::lacZ were constructed in pTL61t (18) and contains 104bp and 99bp upstream of the

108

transcriptional start site, respectively. The tcpI::lacZ fusion contains 76bp upstream of the

109

translational start site (unpublished work). Strains and plasmids used in this study are listed in

110

Table 1. The antibiotic concentration for strains with the pTL61t plasmid was 100ug/mL

111

ampicillin and without plasmid were 100ug/mL streptomycin.

112

β-galactosidase assay. Bicarbonate virulence inducing conditions were as previously

113

described (24). Briefly, V. cholerae classical biotype strain O395 was grown overnight in LB at

114

37°C and subcultured 1:100 into AKI medium in the presence or absence of freshly prepared

115

0.3% sodium bicarbonate (36mM) for 4 hours. Cultures were grown statically for 4 hours at

116

37°C and analyzed. The β-galactosidase assay was performed as previously described (34).

5

117

H14CO3 uptake assay. V. cholerae classical biotype strain O395 was grown overnight in

118

LB at 37°C and subcultured 1:100 into AKI medium in the absence of bicarbonate for 3 hours.

119

At 3 hours, 1 μCi NaH14CO3 (54mCi/mmol) (Perkin-Elmer) was added for each milliliter of V.

120

cholerae subculture. Immediately, 1 mL of culture was centrifuged and supernatant was

121

discarded. The cell pellet was washed 3 times with 1 mL of AKI medium and re-centrifuged.

122

Cell pellets were resuspended in 100 uL AKI medium and added to 5 mL Scintillation cocktail

123

(Fisher Scientific). The same procedure was followed in 15 minute intervals for two hours. After

124

uptake, cpm was measured for each time point using an LS6000IC liquid scintillation counting

125

system (Beckman).

126

Cell Fractionation. V. cholerae strain O395 was grown overnight in LB at 37° C and

127

subcultured 1:100 into AKI medium in the absence of bicarbonate for 3 hours. At 3 hours, 1μCi

128

NaH14CO3 (54 mCi/mmol) (Perkin-Elmer) was added to one milliliter of subculture and

129

incubated at room temperature for one hour. The bacteria were harvested by centrifugation and

130

washed 3 times in 1X PBS. Bacteria were resuspended in a 1 mL solution of 20% sucrose, 20

131

mM Tris-HCl (pH 8.0), and 1 mM Na-EDTA. Cells were lysed by sonication (Branson Sonifier

132

450; 10 seconds on/10 seconds off; amplitude setting 60%; total time 7 min.). Bacteria were then

133

fractionated by centrifugation for 30 min at 10,000 x g to separate the periplasm/membranes and

134

cytoplasm. The supernatant was taken as the cytoplasmic portion and the pellet was resuspended

135

in PBS. The amount of cpm in each fraction was subsequently measured using an LS6000IC

136

liquid scintillation counting system (Beckman).

137

Protein Purification. Maltose binding protein (MBP)-ToxT and MBP-AraC purification

138

was performed as previously described (33). Briefly, fusion proteins were purified from

139

Escherichia coli strain BL21(DE3) with plasmid pMAL-c2e containing MBP-ToxT or MBP-

6

140

AraC. E. coli cells were subcultured until the optical density at 600 nm reached 0.5. Then, the

141

fusion protein was induced with the addition of 0.25 mM isopropyl-β-D-thiogalactopyranoside

142

(IPTG). Cells were lysed by French press and lysate was run over an amylose column (New

143

England Biolabs). Fractions containing MBP-ToxT were dialyzed against buffer containing 50

144

mM Na2HPO4 (pH 8.0), 10 mM Tris-HCl (pH 8.0), and 100 mM NaCl and then dialyzed again

145

against the same solution with 20% glycerol. The protein concentration was determined using a

146

Qubit 2.0 Fluorometer (Invitrogen).

147

Electrophoretic mobility shift assays (EMSA). EMSA was performed as previously

148

described (33). DNA probes were produced by PCR of pTL61T containing appropriate promoter

149

sequence with one unlabeled primer and one primer end labeled with γ-32P (Perkin-Elmer) by T4

150

polynucleotide kinase (New England BioLabs). The binding reactions were prepared to a final

151

volume of 30 μL containing: 10 μg/mL salmon sperm DNA, 10 mM Tris-acetate (pH 7.4), 1 mM

152

Potassium EDTA (pH 7.0), 100 mM KCl, 1 mM dithiothreitol (DTT), 0.3 mg/mL bovine serum

153

albumin (BSA) and 10% glycerol. Binding reactions contained various concentrations of purified

154

MBP-ToxT and all reactions had a constant concentration of labeled DNA probe. For assays

155

using MBP-AraC, binding reactions had a final concentration of 50 mM KCl, 5% glycerol, and

156

50 mM L-arabinose. Sodium bicarbonate, sodium biselenite, and sodium acetate were added to a

157

final concentration of 36 mM in binding reactions containing the respective molecules. The

158

binding reactions were incubated at 30° C for 30 minutes and immediately loaded into 6%

159

polyacrylamide gel and run at 4° C. Gels were dried and analyzed by autoradiography.

160

Binding curve analysis. Autoradiographs were analyzed using ImageJ software (NIH).

161

Percent of labeled DNA bound by protein was determined for each lane. Graphpad Prism 5

162

software

was

used

for

curve

7

fitting

to

the

equation

163

%Bound=Bmax*[Protein]^h/(Kd^h+ [Protein]^h) with Bmax constraint set to 100. The Kd for each

164

condition was calculated and significance was determined. Nonspecific binding was omitted due

165

to the excess of nonspecific salmon sperm DNA added to binding reactions.

166 167

RESULTS

168

Bicarbonate increases activation of ToxT-dependent promoters. Bicarbonate has

169

previously been shown to increase production of the major ToxT-dependent virulence factors,

170

TCP and CT (24); however, the effect of bicarbonate on other ToxT-activated promoters had not

171

been assessed. Earlier work had demonstrated that activation of genes aldA, tagA, acfA, acfD,

172

and tcpI are dependent on production of ToxT (17-19). We assessed transcriptional activity of

173

these ToxT-dependent promoters using plasmid-borne β-galactosidase promoter fusions in the

174

classical biotype V. cholerae strain O395 and an isogenic toxT deletion (15). The location and

175

orientation of the ToxT binding sites, or toxboxes, within each ToxT-dependent promoter is

176

shown (Fig. 1A,B). Fusion construct activity from ctxAB::lacZ, aldA::lacZ, tagA::lacZ,

177

acfA::lacZ, acfD::lacZ, and tcpI::lacZ was measured in the presence and absence of 36 mM

178

sodium bicarbonate to determine if bicarbonate enhanced the ability of ToxT to activate

179

transcription of these genes as it does with the major virulence genes tcpA and ctxAB. Culture

180

conditions used to assess activity were adopted from previous work with bicarbonate (24). Our

181

results confirm that bicarbonate activates each of these ToxT-dependent promoters in wild-type

182

O395, while having no effect on a ΔtoxT strain (Fig. 1C). Therefore, bicarbonate has a global

183

effect by enhancing ToxT activity at all ToxT-dependent promoters.

184

Radiolabeled bicarbonate is taken up by V. cholerae. Previous work (24) and our

185

findings (Fig. 1C) strongly suggested that the positive effect of bicarbonate on ToxT activity was

8

186

the result of a direct interaction, but this had not been confirmed. ToxT protein levels are roughly

187

equivalent in the presence or absence of bicarbonate (24), and therefore ToxT is thought to be in

188

a less active state in the absence of bicarbonate. The simplest mechanism for bicarbonate

189

activation of ToxT would be a direct interaction between bicarbonate and ToxT within the

190

bacterial cytosol. To determine whether such a direct interaction between bicarbonate and ToxT

191

is possible, we performed a NaHC14O3 uptake assay to measure bicarbonate import into the V.

192

cholerae cell. First, classical biotype V. cholerae was subcultured statically in AKI medium for

193

two hours in the absence of bicarbonate. Then, C14-labeled bicarbonate was added to the culture.

194

Aliquots were centrifuged at 15 minute intervals, washed with AKI medium three times and 14C-

195

radiolabel in the cell pellet was quantified using a scintillation counter.

196

Results of the bicarbonate uptake assay displayed a linear increase of radiolabel within

197

the cell pellet of the culture (Fig. 2). Linear regression of the dataset revealed an equation of the

198

best-fit line of y=27.46x+525, with an R2-value of 0.984. The results from the time course uptake

199

show that bicarbonate is entering the bacteria at a linear rate. Further work involving

200

fractionation of the V. cholerae cytosol from the periplasm/membranes revealed that 84.2% ±

201

3.3% of the pellet-associated radiolabeled bicarbonate was located in the cytosol. This finding

202

suggests that bicarbonate can interact directly with ToxT in the bacterial cytosol.

203

Bicarbonate increases ToxT equilibrium binding affinity to PtcpA. The simplest

204

mechanism for a direct interaction between bicarbonate and ToxT that results in increased

205

transcription activation by ToxT would be increased occupancy of toxboxes by ToxT, leading to

206

enhanced interactions between ToxT and RNA polymerase. In the murine pathogen Citrobacter

207

rodentium, an interaction between bicarbonate and another AraC/XylS protein family member,

208

RegA, leads to stabilization of RegA binding to the grlA promoter (35, 36). To determine

9

209

whether bicarbonate directly affects ToxT binding to its specific DNA sites, we characterized the

210

binding pattern of ToxT to virulence gene promoters using EMSA. The assays were designed to

211

establish an estimation of the equilibrium dissociation constant, Kd, thus quantifying the

212

interaction of ToxT with the promoter of ToxT-dependent virulence genes in the absence and

213

presence of bicarbonate. In regard to this work, Kd represents the concentration of ToxT needed

214

to bind 50% of the promoter DNA at equilibrium in the EMSA. Mathematically speaking, there

215

is an inverse relationship between Kd and binding affinity; thus meaning the lower the Kd, the

216

higher the binding affinity of ToxT to the DNA.

217

To determine the Kd of the interaction between ToxT and PtcpA, we titrated purified MBP32

218

ToxT, in the presence or absence of 36 mM bicarbonate, with a constant concentration of

219

labeled PtcpA PCR product below the empirically estimated Kd. MBP-ToxT was used because the

220

independently folded MBP helps to solubilize ToxT and prevent aggregation during EMSA. The

221

percentage of bound DNA at each concentration of MBP-ToxT was calculated using

222

densitometry of the EMSA autoradiographs with ImageJ software. Curve fitting was performed

223

with

224

%Bound=Bmax*[Protein]^h/(Kd^h+ [Protein]^h). Bmax is the amount of bound DNA where the

225

curve plateaus, which we set to a constraint of 100%. This constraint represents all free DNA

226

being bound by ToxT. A representative EMSA for PtcpA is shown in Fig. 3A. Binding reactions

227

that correspond to lanes 1-7 were allowed to equilibrate in the absence of bicarbonate, while

228

binding reactions in lanes 8-14 included bicarbonate. The percentage of bound DNA was

229

calculated and curves were fit for both sets of reactions (Fig. 3B). With the addition of

230

bicarbonate, the concentration of MBP-ToxT to bind 50% of PtcpA was 145.3 nM, significantly

231

lower than the 287.9 nM Kd without bicarbonate. The lower Kd in the presence of bicarbonate

Graphpad

software

using

10

the

following

P

equation:

232

indicates a higher binding affinity for the tcpA promoter. Interestingly, the Hill coefficient of the

233

binding curves is greater than one and the curves take on a sigmoidal shape. This is consistent

234

with the possibility that there are multiple binding sites for ToxT binding and that there is a

235

degree of cooperativity in binding both sites. The presence of two toxboxes at PtcpA has been

236

previously shown (19) but further work would be needed for confirmation of cooperativity

237

between ToxT monomers. The increase in binding affinity in the presence of bicarbonate in an in

238

vitro EMSA indicates a proximal relationship between the bicarbonate ion and ToxT, and likely

239

a direct interaction.

240

The addition of bicarbonate to the binding reaction increases the pH from 7.4 to 8.6. To

241

confirm that the increase in binding affinity is due to the bicarbonate ion itself and not an

242

increase in pH, we performed an EMSA in which binding reactions at pH 7.4 (Fig. 3C, lanes 1-7)

243

and 8.6 (Fig. 3C, lanes 8-14) were compared in the absence of bicarbonate. Densitometric

244

analysis and nonlinear regression provided the binding curves and Kd values for the EMSA (Fig.

245

3D). The binding curves at the two pH values overlapped and the Kd values for each curve were

246

not statistically different. The Kd of these two conditions was lower than in Figure 3A because

247

the concentration of active ToxT protein varies among batches and decreases over time.

248

However, the overall response to the different binding conditions remains similar in different

249

purified protein batches.

250

Both ToxT and RegA, transcriptional regulators of the AraC/XylS family, exhibit

251

increased binding in the presence of bicarbonate (35, 36). We next analyzed whether another

252

family member, AraC, showed signs of stabilized promoter architecture in the presence of

253

bicarbonate to assess whether this was a common feature shared among the entire AraC family.

254

Our results indicate that the addition of bicarbonate inhibited binding of AraC to the araBAD

11

255

promoter (Fig. 3E,F). Therefore, not all AraC/XylS family members are positively modulated by

256

bicarbonate.

257

Other small effector molecules do not increase ToxT binding to PtcpA. The results

258

described above suggest that addition of bicarbonate to binding reactions increases the binding

259

affinity of ToxT to PtcpA. However, it is possible that this phenomenon is not limited to

260

bicarbonate specifically and could also be mediated by other small molecules similar to

261

bicarbonate. We next tested the binding response of ToxT to molecules with similar structure or

262

molecular weight to bicarbonate. Sodium biselenite has a similar structure to bicarbonate with a

263

different central atom. ToxT had a greatly reduced binding affinity in the presence of biselenite

264

(Fig. 4A,B). We also tested sodium acetate, which has similar molecular weight to sodium

265

bicarbonate. ToxT also showed decreased binding to PtcpA with the addition of this small

266

molecule (Fig. 4C,D). The increase in binding affinity of ToxT to PtcpA appears to be

267

bicarbonate-specific as other small molecules do not enhance ToxT binding or activity (data not

268

shown).

269

Bicarbonate increases binding to ToxT-dependent promoters regardless of toxbox

270

configuration. EMSA equilibrium binding experiments indicated that bicarbonate increases

271

ToxT binding affinity for PtcpA. However, there is significant diversity in the configuration and

272

spacing of toxboxes at different ToxT-activated promoters (19). ToxT directly controls

273

transcription of several other virulence genes besides tcpA, including ctxAB, aldA, tagA, acfA,

274

acfD, and tcpI (17-19). The tcpA promoter contains two toxboxes in a direct repeat

275

configuration. PctxAB, like PtcpA, is also configured as two direct repeat toxboxes (Fig. 1A).

276

However, the orientation of the toxboxes is in the opposite direction relative to the promoter at

277

PctxAB and the spacing between toxboxes differs between PtcpA and PctxAB (19, 33, 37). In contrast

12

278

to PtcpA and PctxAB, PtagA, PacfA, PacfD, and PtcpI each contain a pair of toxboxes in an inverted

279

repeat configuration and with variations in spacing between toxboxes (Fig. 1B). Finally, PaldA

280

contains a single toxbox (Fig. 1A). As described earlier, addition of bicarbonate to V. cholerae

281

cultures increases activity from all of these promoters (Fig. 1C).

282

To determine whether bicarbonate affects binding affinity of ToxT for each of these

283

unique promoters, we performed EMSAs and calculated Kd as previously described for PtcpA.

284

First, we tested the ability of bicarbonate to increase the binding affinity of ToxT to PctxAB, which

285

controls production of CT and has two direct repeat toxboxes. Figure 5A shows the

286

autoradiograph of ToxT binding to PctxAB in the absence and presence of bicarbonate.

287

Densitometry and linear regression revealed that the Kd is reduced from 328.7 nM MBP-ToxT to

288

231.4 nM with bicarbonate, indicating an increase in binding affinity. Next, we investigated if

289

bicarbonate increased ToxT binding at the inverted repeat promoter of tagA. We used PtagA as an

290

example of inverted repeat toxbox configurations that also occur in the promoters of acfA, acfD,

291

and tcpI. As with the direct repeat promoters, bicarbonate increased the binding affinity of ToxT

292

to PtagA (Fig. 5C, D). Finally, we performed EMSAs to determine if bicarbonate increased

293

binding to the single toxbox promoter of aldA. The EMSA with binding reactions of MBP-ToxT

294

and PaldA in the absence and presence of bicarbonate is shown in Fig. 5E. The binding curve was

295

developed and revealed that bicarbonate decreased the Kd for ToxT binding to PaldA from 173.7

296

nM to 113.9 nM. The increase in binding affinity to each of the promoters with bicarbonate

297

agreed with the results of the analysis of each promoter in the β-galactosidase assays. Together,

298

these data demonstrate that bicarbonate increases binding at each promoter regardless of toxbox

299

orientation, including a single toxbox promoter, and reveals that the mechanism by which

13

300

bicarbonate enhances ToxT activity is most likely a change in protein conformation, resulting in

301

increased DNA binding affinity.

302 303

DISCUSSION

304

The regulatory network that controls production of the major virulence factors

305

responsible for disease symptoms of cholera culminates with the production of ToxT. ToxT is

306

the transcription regulator directly responsible for activation of tcpA and ctxAB transcription (13-

307

16). Previous studies have shown that bicarbonate, a signal located in the upper small intestine

308

where V. cholerae colonizes, enhances the ability of ToxT to activate these major virulence

309

genes (24). In addition to these genes, we have shown that bicarbonate increases transcription of

310

other ToxT-dependent virulence genes, aldA, tagA, acfA, acfD, and tcpI.

311

ToxT activity is modulated by different effector molecules. It is negatively regulated by

312

bile and its unsaturated fatty acid components (22, 23), as well as the small molecule virstatin

313

(28, 38). Bile is located at high concentration in the lumen of the upper small intestine and the

314

interaction between the unsaturated fatty acids of bile and ToxT are thought to decrease ToxT’s

315

ability to dimerize effectively and bind to DNA (27, 29, 30). Our findings have shown that

316

bicarbonate, which is also present at high concentration in the upper small intestine, has a

317

mechanism converse to bile. Our results indicate that bicarbonate-mediated activation of

318

virulence gene transcription occurs as a result of an increase in ToxT binding affinity to

319

virulence gene promoters. The increase in binding affinity in response to bicarbonate can be seen

320

at each ToxT-dependent promoter, regardless of toxbox configuration and positioning.

321

The binding curves generated from EMSAs performed on different ToxT-dependent

322

promoters were sigmoidal. Sigmoidal binding curves generally relate to cooperative binding to

14

323

multiple binding sites of a protein, similar to O2 binding by hemoglobin (39). Cooperativity

324

occurs when one bound molecule increases the binding affinity of subsequent molecules.

325

Positive cooperativity can also be seen in terms of transcription factors binding to promoters with

326

multiple binding sites. One example of positive cooperativity of transcription factors is seen with

327

the TtgR operator of Pseudomonas putida strain DOT-T1E (40). TtgR, which can form dimers in

328

solution, exhibits biphasic cooperative binding at this promoter (40). ToxT is thought to have

329

optimal activity when acting as a dimer (27-31). In contrast, it has also been shown that ToxT

330

can bind DNA as a monomer (18, 19, 26). The sigmoidal ToxT binding curve at two toxbox

331

promoters (Fig. 3B, 5B, 5D) suggests that a single ToxT monomer binds the DNA first,

332

coinciding with previous findings. Additionally, the possibility of positive cooperativity derived

333

from the sigmoidal binding curve and Hill coefficient implies that the binding of the first

334

monomer increases the binding affinity of the second monomer. This can be explained by an

335

increase in the incidence of association between ToxT monomers after one has already bound the

336

promoter, increasing the binding affinity of the second ToxT monomer. Interestingly, the ToxT

337

binding curve of the single toxbox promoter of aldA also exhibited a sigmoidal shape. Previous

338

studies have shown that this promoter contained a single toxbox that was necessary and

339

sufficient for full transcription activation (17); however, the ToxT binding curve at this promoter

340

opens the possibility that another, atypical and low affinity, ToxT binding site may be present in

341

the vicinity of the promoter. Further work is needed to determine the role of dimerization and

342

cooperativity at each of these promoters.

343

Potential mechanisms for the increased ToxT DNA binding affinity in response to

344

bicarbonate include a change in conformation of ToxT or increased frequency of ToxT

345

dimerization. Indirect mechanisms that increase ToxT binding can be ruled out as the increase in

15

346

binding was seen in binding experiments using purified components. Work on an AraC/XylS

347

family member of C. rodentium, RegA, revealed that bicarbonate increases RegA binding to

348

promoter regions (35, 36). RegA can act as a monomer or dimer, similar to ToxT, and it was

349

shown that bicarbonate did not increase the incidence of RegA dimerization (41). As a primary

350

assessment of ToxT dimerization in the presence of bicarbonate we analyzed the ability of

351

bicarbonate to increase ToxT binding to the single toxbox of PaldA, in which dimerization should

352

not be evident. However, as stated before, our findings revealed the possibility of more than one

353

toxbox in or near the promoter of aldA. Consequently, further investigation into the precise

354

mechanism of increased ToxT binding affinity in the presence of bicarbonate is needed.

355

Bicarbonate must first enter the bacterium to interact directly with ToxT. We report here

356

that bicarbonate does get taken up into the bacterial cytosol. However, further investigation into

357

the mechanism of uptake is needed. Previous work has shown that addition of the carbonic

358

anhydrase inhibitor, ethoxzolamide (EZA), decreases activation of PtcpA in the presence of

359

bicarbonate (24). V. cholerae contains genes encoding each of the three classes of bacterial

360

carbonic anhydrase. A triple carbonic anhydrase deletion did not reveal a decrease in H14CO3

361

uptake compared to WT O395 (data not shown). This suggests that the mechanism for uptake is

362

not completely reliant on carbonic anhydrase-dependent conversion of HCO3- to CO2.

363

Signals that V. cholerae encounters during passage to the intestine result in the

364

production of ToxT (4), but due to the presence of a high concentration of bile and unsaturated

365

fatty acids in the lumen of the intestine, ToxT is initially in an inactive form (22, 23, 27, 30).

366

Bicarbonate is also present in the lumen where bile is present and could potentially be activating

367

some ToxT protein. The bacteria in the lumen still express flagella and undergo positive

368

chemotaxis towards mucin proteins (1, 42). As the bacteria swim closer to the intestinal

16

369

epithelium, the concentration of bile decreases as the large molecules that comprise bile cannot

370

enter the mucus layer. The concentration of bicarbonate increases as the bacteria get closer to the

371

epithelium due to direct secretion of bicarbonate from these cells (32), but more importantly, the

372

ratio of bicarbonate to UFAs increases. ToxT protein becomes active as it reaches the high

373

bicarbonate/low UFA breakpoint and active ToxT increases transcription of genes involved in

374

virulence, such as CT and TCP (24). In addition to these major virulence factors, bicarbonate

375

activates accessory virulence factors important in disease progression. Our findings here reveal

376

that the mechanism for bicarbonate-mediated enhancement of ToxT activity is due to an increase

377

in ToxT binding affinity to virulence gene promoters. This provides a direct relationship between

378

an intestinal signal and increased expression of genes required for pathogenesis.

379 380

ACKNOWLEDGEMENTS

381

The authors thank the members of the Withey and Neely laboratories for helpful

382

discussions. This work was supported by P.H.S. grants K22AI071011 and R56AI093622 (to

383

J.H.W.) and by Wayne State University funds.

384 385 386

REFERENCES

387 388 389 390 391 392 393 394 395

1. 2. 3. 4.

Butler SM, Camilli A. 2005. Going against the grain: chemotaxis and infection in Vibrio cholerae. Nat Rev Microbiol 3:611-620. Krukonis ES, DiRita VJ. 2003. From motility to virulence: Sensing and responding to environmental signals in Vibrio cholerae. Curr Opin Microbiol 6:186-190. Peterson KM, Mekalanos JJ. 1988. Characterization of the Vibrio cholerae ToxR regulon: identification of novel genes involved in intestinal colonization. Infect Immun 56:2822-2829. Matson JS, Withey JH, DiRita VJ. 2007. Regulatory networks controlling Vibrio cholerae virulence gene expression. Infect Immun 75:5542-5549.

17

396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441

5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21.

Kovacikova G, Skorupski K. 2001. Overlapping binding sites for the virulence gene regulators AphA, AphB and cAMP-CRP at the Vibrio cholerae tcpPH promoter. Mol Microbiol 41:393-407. Kovacikova G, Lin W, Skorupski K. 2004. Vibrio cholerae AphA uses a novel mechanism for virulence gene activation that involves interaction with the LysR-type regulator AphB at the tcpPH promoter. Mol Microbiol 53:129-142. Kovacikova G, Skorupski K. 1999. A Vibrio cholerae LysR homolog, AphB, cooperates with AphA at the tcpPH promoter to activate expression of the ToxR virulence cascade. J Bacteriol 181:4250-4256. Behari J, Stagon L, Calderwood SB. 2001. pepA, a gene mediating pH regulation of virulence genes in Vibrio cholerae. J Bacteriol 183:178-188. Kovacikova G, Lin W, Skorupski K. 2010. The LysR-type virulence activator AphB regulates the expression of genes in Vibrio cholerae in response to low pH and anaerobiosis. J Bacteriol 192:4181-4191. Krukonis ES, Yu RR, Dirita VJ. 2000. The Vibrio cholerae ToxR/TcpP/ToxT virulence cascade: distinct roles for two membrane-localized transcriptional activators on a single promoter. Mol Microbiol 38:67-84. Yang M, Liu Z, Hughes C, Stern AM, Wang H, Zhong Z, Kan B, Fenical W, Zhu J. 2013. Bile salt-induced intermolecular disulfide bond formation activates Vibrio cholerae virulence. Proc Natl Acad Sci U S A 110:2348-2353. Fan F, Liu Z, Jabeen N, Birdwell LD, Zhu J, Kan B. 2014. Enhanced Interaction of Vibrio cholerae Virulence Regulators TcpP and ToxR under Oxygen-Limiting Conditions. Infect Immun 82:1676-1682. Hulbert RR, Taylor RK. 2002. Mechanism of ToxT-dependent transcriptional activation at the Vibrio cholerae tcpA promoter. J Bacteriol 184:5533-5544. Yu RR, DiRita VJ. 1999. Analysis of an autoregulatory loop controlling ToxT, cholera toxin, and toxin-coregulated pilus production in Vibrio cholerae. J Bacteriol 181:25842592. Champion GA, Neely MN, Brennan MA, DiRita VJ. 1997. A branch in the ToxR regulatory cascade of Vibrio cholerae revealed by characterization of toxT mutant strains. Mol Microbiol 23:323-331. DiRita VJ, Parsot C, Jander G, Mekalanos JJ. 1991. Regulatory cascade controls virulence in Vibrio cholerae. Proc Natl Acad Sci U S A 88:5403-5407. Withey JH, Dirita VJ. 2005. Vibrio cholerae ToxT independently activates the divergently transcribed aldA and tagA genes. J Bacteriol 187:7890-7900. Withey JH, DiRita VJ. 2005. Activation of both acfA and acfD transcription by Vibrio cholerae ToxT requires binding to two centrally located DNA sites in an inverted repeat conformation. Mol Microbiol 56:1062-1077. Withey JH, DiRita VJ. 2006. The toxbox: specific DNA sequence requirements for activation of Vibrio cholerae virulence genes by ToxT. Mol Microbiol 59:1779-1789. Bradley ES, Bodi K, Ismail AM, Camilli A. 2011. A genome-wide approach to discovery of small RNAs involved in regulation of virulence in Vibrio cholerae. PLoS pathogens 7:e1002126. Richard AL, Withey JH, Beyhan S, Yildiz F, DiRita VJ. 2010. The Vibrio cholerae virulence regulatory cascade controls glucose uptake through activation of TarA, a small regulatory RNA. Mol Microbiol 78:1171-1181.

18

442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485

22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36.

37. 38.

Chatterjee A, Dutta PK, Chowdhury R. 2007. Effect of fatty acids and cholesterol present in bile on expression of virulence factors and motility of Vibrio cholerae. Infect Immun 75:1946-1953. Schuhmacher DA, Klose KE. 1999. Environmental signals modulate ToxT-dependent virulence factor expression in Vibrio cholerae. J Bacteriol 181:1508-1514. Abuaita BH, Withey JH. 2009. Bicarbonate Induces Vibrio cholerae virulence gene expression by enhancing ToxT activity. Infect Immun 77:4111-4120. Higgins DE, Nazareno E, DiRita VJ. 1992. The virulence gene activator ToxT from Vibrio cholerae is a member of the AraC family of transcriptional activators. J Bacteriol 174:6974-6980. Bellair M, Withey JH. 2008. Flexibility of Vibrio cholerae ToxT in transcription activation of genes having altered promoter spacing. J Bacteriol 190:7925-7931. Prouty MG, Osorio CR, Klose KE. 2005. Characterization of functional domains of the Vibrio cholerae virulence regulator ToxT. Mol Microbiol 58:1143-1156. Shakhnovich EA, Hung DT, Pierson E, Lee K, Mekalanos JJ. 2007. Virstatin inhibits dimerization of the transcriptional activator ToxT. Proc Natl Acad Sci U S A 104:23722377. Lowden MJ, Skorupski K, Pellegrini M, Chiorazzo MG, Taylor RK, Kull FJ. 2010. Structure of Vibrio cholerae ToxT reveals a mechanism for fatty acid regulation of virulence genes. Proc Natl Acad Sci U S A 107:2860-2865. Childers BM, Cao X, Weber GG, Demeler B, Hart PJ, Klose KE. 2011. N-terminal residues of the Vibrio cholerae virulence regulatory protein ToxT involved in dimerization and modulation by fatty acids. J Biol Chem 286:28644-28655. Childers BM, Weber GG, Prouty MG, Castaneda MM, Peng F, Klose KE. 2007. Identification of residues critical for the function of the Vibrio cholerae virulence regulator ToxT by scanning alanine mutagenesis. J Mol Biol 367:1413-1430. Hogan DL, Ainsworth MA, Isenberg JI. 1994. Review article: gastroduodenal bicarbonate secretion. Aliment Pharmacol Ther 8:475-488. Dittmer JB, Withey JH. 2012. Identification and characterization of the functional toxboxes in the Vibrio cholerae cholera toxin promoter. J Bacteriol 194:5255-5263. Miller JH. 1972. Experiments in molecular genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Tauschek M, Yang J, Hocking D, Azzopardi K, Tan A, Hart E, Praszkier J, RobinsBrowne RM. 2010. Transcriptional analysis of the grlRA virulence operon from Citrobacter rodentium. J Bacteriol 192:3722-3734. Yang J, Hart E, Tauschek M, Price GD, Hartland EL, Strugnell RA, RobinsBrowne RM. 2008. Bicarbonate-mediated transcriptional activation of divergent operons by the virulence regulatory protein, RegA, from Citrobacter rodentium. Mol Microbiol 68:314-327. Yu RR, DiRita VJ. 2002. Regulation of gene expression in Vibrio cholerae by ToxT involves both antirepression and RNA polymerase stimulation. Mol Microbiol 43:119134. Hung DT, Shakhnovich EA, Pierson E, Mekalanos JJ. 2005. Small-molecule inhibitor of Vibrio cholerae virulence and intestinal colonization. Science 310:670-674.

19

486 487 488 489 490 491 492 493 494 495 496 497 498 499

39. 40.

41. 42.

Goutelle S, Maurin M, Rougier F, Barbaut X, Bourguignon L, Ducher M, Maire P. 2008. The Hill equation: a review of its capabilities in pharmacological modelling. Fundam Clin Pharmacol 22:633-648. Krell T, Teran W, Mayorga OL, Rivas G, Jimenez M, Daniels C, Molina-Henares AJ, Martinez-Bueno M, Gallegos MT, Ramos JL. 2007. Optimization of the palindromic order of the TtgR operator enhances binding cooperativity. J Mol Biol 369:1188-1199. Yang J, Dogovski C, Hocking D, Tauschek M, Perugini M, Robins-Browne RM. 2009. Bicarbonate-mediated stimulation of RegA, the global virulence regulator from Citrobacter rodentium. J Mol Biol 394:591-599. Allweiss B, Dostal J, Carey KE, Edwards TF, Freter R. 1977. The role of chemotaxis in the ecology of bacterial pathogens of mucosal surfaces. Nature 266:448-450.

500 501 502 503

FIGURE LEGENDS

504

FIGURE 1. Transcriptional effects of exogenous bicarbonate on ToxT-dependent

505

virulence gene promoters. (A) Orientation and positioning of direct repeat toxbox virulence gene

506

promoters PtcpA and PctxAB, and single toxbox promoter, PaldA. (B) Orientation and positioning of

507

inverted repeat toxbox virulence gene promoters, PtagA, PacfA, PacfD, and PtcpI. Toxbox represents

508

known ToxT binding sites. (C) β-galactosidase activity produced from plasmid-borne virulence

509

gene promoter fusion constructs in classical strain O395 and isogenic ∆toxT strain. Cultures were

510

grown in AKI media in the absence or presence of 36 mM NaHCO3. Statistical significance

511

determined using Student’s t-test (*, P < 0.00025). Error bars represent +/- standard deviation.

512

FIGURE 2. Radiolabeled bicarbonate is taken up by V. cholerae. Classical strain O395

513

cells were grown in AKI media in the absence of bicarbonate for 2 hours. NaH14CO3 was added

514

at time 0 and uptake was analyzed every 15 minutes by liquid scintillation counting. NaH14CO3

20

515

uptake was measured as CPM, reflecting the amount of radioactive 14C found within V. cholerae

516

culture. Error bars represent +/- standard deviation of three separate experiments.

517

FIGURE 3. Bicarbonate increases binding affinity of MBP-ToxT to PtcpA. MBP-ToxT

518

binding to PtcpA was analyzed using EMSA. Autoradiograph of EMSAs presented are

519

representative of three or more independent experiments. (A) Binding reactions between MBP-

520

ToxT and PtcpA in lanes 1-7 were conducted in the absence of NaHCO3. Lanes 8-14 were

521

incubated in the presence of 36 mM NaHCO3. Lanes 1 and 8 contained PtcpA DNA in the absence

522

of MBP-ToxT. Subsequent lanes contained a titration of MBP-ToxT with concentrations labeled

523

in the figure. (B) Binding curve for the autoradiograph shown in (A). Densitometry of

524

autoradiograph was performed with ImageJ software. Circles represent percent PtcpA bound by

525

MBP-ToxT in the absence of bicarbonate. Solid line corresponds to the binding curve for MBP-

526

ToxT to PtcpA determined by the equation %Bound=Bmax*[Protein]^h/(Kd^h+ [Protein]^h) with

527

Bmax constraint set to 100 using Graphpad Prism 5 software. Squares and dashed line represent

528

percent bound and binding curve, respectively, in the presence of bicarbonate. Kd for each

529

condition is inset and significant difference between the best-fit values of each data set is

530

denoted by * (P < 0.00025). (C) Autoradiograph of EMSA showing titration of MBP-ToxT

531

bound to PtcpA at pH 7.4 (Lanes 1-7) and pH 8.6 (Lanes 8-14). (D) Binding curves of each pH

532

condition with Kd. (E) Autoradiograph of EMSA showing titration of MBP-AraC bound to PBAD

533

in the absence of NaHCO3 (Lanes 1-7) and presence of 36 mM NaHCO3 (Lanes 8-14). (F)

534

Binding curves of MBP-AraC in the absence and presence of 36 mM NaHCO3.

535

FIGURE 4. Small molecules, biselenite and acetate, do not enhance MBP-ToxT binding

536

to PtcpA. Autoradiograph of EMSA showing binding reactions of MBP-ToxT to PtcpA with

537

different small molecules similar to NaHCO3. (A) Binding reactions in the absence and presence

21

538

of 36 mM NaHSeO3 with corresponding binding curves (B). (C) Binding reactions in the

539

absence and presence of 36 mM NaC2H3O2 along with corresponding binding curves (D).

540

FIGURE 5. Bicarbonate increases ToxT binding affinity to promoters having all known

541

toxbox configurations. Each autoradiograph is a representative or three or more independent

542

experiments. Binding reactions in lanes 1-7 in all autoradiographs were conducted in the absence

543

of NaHCO3. Binding reactions in lanes 8-14 in all autoradiographs were conducted in the

544

presence of 36 mM NaHCO3. All binding curves were generated using Graphpad Prism 5

545

software using the equation %Bound=Bmax*[Protein]^h/(Kd^h+ [Protein]^h) with Bmax constraint

546

set to 100. (A) Autoradiograph of EMSA showing MBP-ToxT binding to PctxAB, in which two

547

toxboxes are oriented as direct repeats. Representative binding curves shown in (B). (C)

548

Autoradiograph of EMSA showing MBP-ToxT binding to PtagA, in which two toxboxes are

549

oriented as inverted repeat. (D) Binding curves corresponding to MBP-ToxT binding to PtagA. (E)

550

Autoradiograph of EMSA showing MBP-ToxT binding to single toxbox promoter, PaldA. (F)

551

Binding curves of PaldA EMSA.

552 553 554

22

555 556

Table 1. Strains and plasmids used in this study. Strain or Plasmid Strains JW 467 JW 1560 JW 9 JW 150 JW 18 JW 441 JW 87 JW 97 JW 86 JW 89 JW 99 JW 515 JW 1809 JW 167 JW 168 JW 165 JW 166 JW 169

Description Escherichia coli BL21(DE3) JW 467 + pJW 407 (MBP-ToxT) Vibrio cholerae classical strain O395 O395 ∆toxt pJW 54 + JW 9 pJW 211 + JW 9 pJW 82 + JW 9 pJW 89 + JW 9 pJW 81 + JW 9 pJW 84 + JW 9 pJW 91 + JW 9 pJW 54 + JW 150 pJW 211 + JW 150 pJW 82 + JW 150 pJW 89 + JW 150 pJW 81 + JW 150 pJW 84 + JW 150 pJW 91 + JW 150

Lab Collection Lab Collection Lab Collection Lab Collection (19) (33) (17) (17) (18) (18) Lab Collection Lab Collection (33) Lab Collection Lab Collection Lab Collection Lab Collection Lab Collection

Plasmids pJW 407 pJW 54 pJW 211 pJW 82 pJW 89 pJW 81 pJW 84 pJW 91

pMAL-c2e-ToxT pTL61t containing 138bp upstream of tcpA pTL61t containing 76bp upstream of ctxAB pTL61t containing 158bp upstream of aldA pTL61t containing 92bp upstream of tagA pTL61t containing 104bp upstream of acfA pTL61t containing 99bp upstream of acfD pTL61t containing 76bp upstream of tcpI

(33) (19) (33) (17) (17) (18) (18) Lab Collection

557

23

Source

A.

B.

C.

6000

CPM

4000

2000

0

0

50

100

Minutes

150

200

A.

B.

C.

D.

E.

F.

A.

B.

C.

D.

A.

B.

C.

D.

E.

F.