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.