Virulence regulator AphB enhances toxR transcription in Vibrio cholerae

Background Vibrio cholerae is the causative agent of cholera. Extensive studies reveal that complicated regulatory cascades regulate expression of virulence genes, the products of which are required for V. cholerae to colonize and cause disease. In this study, we investigated the expression of the key virulence regulator ToxR under different conditions. Results We found that compared to that of wild type grown to stationary phase, the toxR expression was lower in an aphB mutant strain. AphB has been previously shown to be a key virulence regulator that is required to activate the expression of tcpP. When expressed constitutively, AphB is able to activate the toxR promoter. Furthermore, gel shift analysis indicates that AphB binds toxR promoter region directly. We also characterize the effect of AphB on the levels of the outer membrane porins OmpT and OmpU, which are known to be regulated by ToxR. Conclusions Our data indicate that V. cholerae possesses an additional regulatory loop that use AphB to activate the expression of two virulence regulators, ToxR and TcpP, which together control the expression of the master virulence regulator ToxT.


Background
The Gram-negative bacterium Vibrio cholerae is the etiologic agent of cholera. The ability of V. cholerae to colonize and cause disease in hosts requires production of a number of virulence factors during infection. The two major virulence determinants of V. cholerae are encoded by two separate genetic elements: cholera toxin (CT), which causes the diarrhea characteristic of cholera, and the toxin-coregulated pilus (TCP), which is essential for attachment and colonization of intestinal epithelia [1,2]. CT is encoded by the ctxAB genes on the lysogenic CTXÖ bacteriophage [3]. The genes required for TCP synthesis and the genes encoding the virulence transcriptional activators ToxT and TcpP are located on a 40-kb Vibrio pathogenicity island (VPI) [4]. Coordinate expression of V. cholerae virulence genes results from the activity of a cascading system of regulatory factors [5] (Fig. 1).
The primary direct transcriptional activator of V. cholerae virulence genes, including ctxAB and tcpA, is ToxT, a member of the AraC family of proteins [6]. The expression of ToxT is under the control of a complex regulatory pathway. The ToxR protein was identified as the first positive regulator of V. cholerae virulence genes [7]. ToxR activity requires the presence of another protein, ToxS, which is also localized to the inner membrane, but is thought to reside predominantly in the periplasm, where ToxR and ToxS are hypothesized to interact. ToxS serves as a mediator of ToxR function, perhaps by influencing its stability and/or capacity to dimerize [6]. To regulate expression of toxT, ToxR acts in conjunction with a second transcriptional activator, TcpP, which is also membrane-localized with a cytoplasmic DNA-binding and other periplasmic domains [8]. TcpP, like ToxR, requires the presence of a membrane-bound effector protein, TcpH, which interacts with TcpP [9]. Two activators encoded by unlinked genes, AphA and AphB, regulate the transcription of tcpPH. AphA is a dimer with an N-terminal winged-helix DNA binding domain that is structurally similar to those of MarR family transcriptional regulators [10]. AphA cannot activate transcription of tcpP alone, but requires interaction with the LysR-type regulator AphB that binds downstream of the AphA binding site [11].
The ToxR and ToxS regulatory proteins have long been considered to be at the root of the V. cholerae virulence regulon, called the ToxR regulon. The membrane localization of ToxR suggests that it may directly sense and respond to environmental signals such as temperature, osmolarity, and pH [12]. In addition to regulating the expression toxT, ToxR activates the transcription of ompU and represses the transcription of ompT, outer membrane porins important for V. cholerae virulence [13,14]. Microarray analysis indicates that ToxR regulates additional genes, including a large number of genes involved in cellular transport, energy metabolism, motility, and iron uptake [15]. It has been reported that levels of ToxR protein appear to remain constant under various in vitro conditions [16,17] and are modulated by the heat shock response [18].
To further investigate the relationship between toxR expression and other virulence regulators, we analyzed toxR transcription and ToxR protein levels in various virulence regulator mutants. We found that in addition to activating tcpP, AphB was required for full expression of ToxR in V. cholerae stationary growth phase. AphB regulated toxR directly as purified recombinant AphB binds to the toxR promoter. This study suggests that V. cholerae may use this additional layer of activation to turn on virulence factor production efficiently in optimal conditions.

Examination of toxR expression under different in vitro conditions using a transcriptional fusion reporter
ToxR is one of two proteins, along with TcpP, shown to activate the expression of ToxT, the master virulence activator in V. cholerae (Fig. 1). The expression of tcpP has been shown to be induced by AphA and AphB [11,19], while toxR has been thought to be constitutively expressed and only modulated by temperature [16,18]. To measure toxR expression, we placed the toxR promoter upstream of the luxCDABE operon on a plasmid [20] and transformed into wild type V. cholerae. We then grew the resulting cells at 37°C or 22°C. Expression of P toxR -luxCDABE was significantly increased at 22°C ( Fig. 2A), consistent with the previous report [18] that the expression of toxR is modulated by temperatures. Since the availability of oxygen concentrations is different during V. cholerae infection, we also examined the expression of toxR under varying oxygen concentrations (Fig. 2B). The lux expression was similar under each condition, suggesting that oxygen levels do not regulate toxR expression.

Influence of virulence regulatory proteins on toxR expression
To investigate molecular influences on toxR expression, we introduced the P toxR -lux construct into various strains of V. cholerae with mutations in virulence regulator genes. We also included a tcpA mutant because a previous study showed that TcpA, the major subunit of TCP pilin [2], affects cholera toxin gene expression in vivo but not in vitro [21]. We grew these strains at 37°C for 12 hours and measured luminescence (Fig. 3A). We found that ToxR and ToxS did not affect toxR expression, indicating that ToxR does not autoregulate. The expression of toxR in tcpPH, toxT, and tcpA mutants remained the same as that of wild type, but it was significantly decreased in aphA and aphB mutant strains (approximately 3-and 6-fold, respectively). Of note, toxR expression in wild type and aphA or aphB mutants remained similar in the early and logarithmic phases of growth (data not shown). We also examined toxR expression in wild type and various virulence regulatory mutants grown under the AKI condition [22], in which virulence genes are induced in El Tor strains of V. cholerae. We found that toxR expression was decreased in The ToxR regulon Figure 1 The ToxR regulon. AphA and AphB are known to activate tcpPH expression. TcpPH and ToxRS activate the expression of ToxT, which in turn activates the expression of the central virulence factors, cholera toxin (CT) and the toxin-coregulated pilus (TCP). ToxRS also upregulates OmpU and downregulates OmpT, which are outer membrane porins.
The expression of toxR in wild type under different conditions using a P toxR -luxCDABE transcriptional reporter both aphA and aphB mutants to a similar degree as those grown in LB medium (data not shown). These data suggest that AphA and AphB may be important factors in increasing toxR expression during V. cholerae stationary growth. These studies were confirmed by Western blot to examine ToxR protein levels ( Fig. 3B): compared to those of wild type and other mutant strains, ToxR protein levels were notably decreased in the aphA and aphB mutants. Interestingly, while toxR transcription was unchanged in toxS mutant (Fig. 3A), ToxR proteins were not detected in the absence of ToxS, suggesting that the ToxR effector ToxS may affect ToxR stability, at least in the stationary phase condition we tested. Beck et al. reported that loss of ToxS had no measurable negative effect on steady-state levels of the ToxR protein at the mid-log phase growth [9]. The decreased ToxR expression at stationary phase in a toxS mutant is the subject of another investigation.

AphB directly regulates toxR expression
Knowing that full expression of ToxR required both AphA and AphB, we sought to determine which was directly responsible for this effect. To this end, we placed aphA and aphB under control of an arabinose-inducible promoter and measured its effect on P toxR -luxCDABE transcription in E. coli. Overexpression of AphB, but not AphA, dramatically increased toxR transcription (Fig. 4A). We currently do not know why in V. cholerae, both AphA and AphB are required to fully activate toxR expression, while in E. coli, only AphB can induce P toxR -luxCDABE. One possibility is that in V. cholerae, the expression of aphB is dependent on AphA. However, we examined aphB expression in wild type and aphA mutant strains and did not detect any difference. Another possibility is that AphA may indirectly activate ToxR expression through an intermediate which is absent in E. coli, or that AphA is required to repress an inhibitor of AphB that is present in V. cholera but not in E. coli. AphA has been shown to regulate a number of other genes [23,24]. The activation of ToxR hinted at in this study may thus rely on the regulation of other members of the regulation cascade not yet elucidated. We further confirmed AphB regulation of toxR in V. cholerae using a chromosomal transcriptional toxR-lacZ fusion (Fig. 4B). We found that compared to that of wild type, toxR-lacZ expression was reduced in aphB mutants, while expression of aphB from a plasmid in this mutant restored toxR expression (Fig. 4B) and ToxR production (Fig. 4C).
To investigate whether AphB-mediated activation of toxR is direct or acts through another regulator present in E. coli, we purified AphB as an MBP (maltose-binding protein) fusion. Recombinant AphB is functional, as it could activate tcpP transcription in E. coli (data not shown). We then performed Electrophoretic Mobility Shift Assays (EMSA) using MBP-AphB and various lengths of toxR promoter DNA (Fig. 5A). Fig. 5B shows that purified MBP-AphB was able to shift the two large toxR promoter fragments. All of these mobility shifts could be inhibited by the addition of unlabeled specific DNA, indicating that the binding of AphB to these DNA sequences is specific (data not shown). AphB was unable to shift the shortest toxR promoter fragment containing the 130 base pairs closest to the toxR translational start site, suggesting that the AphB binding site is located between 130 and 450 base pairs upstream of the toxR gene. It has been reported that AphB binds and regulates tcpP and aphB promoter regions, and the AphB recognition sites in these promoters were identified [25]. We identified a similar putative AphB binding site in the toxR promoter region approximately 150 bp upstream of the toxR translational start (Fig. 5). Further studies are required to test whether AphB protein binds this putative recognition site. Consistent with the gel shift data, AphB could not induce toxR expression when the 130-bp fragment was fused with the luxCD-ABE reporter in E. coli (Fig. 5A). Taken together, these data suggest that AphB directly regulates toxR expression.

The effects of AphB on ToxR-regulated genes
In addition to regulation of toxT, ToxR has been previously shown to alter the porin levels in V. cholerae by activating expression of ompU and repressing ompT [26,27]. Since we showed that AphB affects ToxR levels, we hypothesized that AphB might thus indirectly modulate the expression of ompU and ompT as well. We performed SDS-PAGE on total protein extracts of wild type V. cholerae as well as toxR and aphB mutants. As expected, the toxR strain had significantly lower OmpU and higher OmpT levels than in the wild-type strain. Interestingly, the aphB mutant strain produced slightly higher levels of OmpT than wild type, though OmpU levels did not seem to change (Fig. 6A). In addition, Provenzano et al. showed that ToxR-dependent modulation of outer membrane proteins enhances V. cholerae resistance to antimicrobial compounds such as bile salts and sodium dodecyl sulfate (SDS) [28]. We confirmed that the toxR mutant strain had a reduced minimum bactericidal concentration (MBC) of SDS compared to wild type strains, but AphB did not affect the MBC of SDS (Fig. 6A). Thus, AphB may only subtly modulate outer membrane porin expression through its effect on toxR expression. This may be another downstream effect of AphB on the virulence capabilities of V. cholerae in addition to its better characterized influences on ToxT levels. Moreover, as both ToxR and TcpP are required to activate toxT expression and AphB is required to activate tcpP expression (Fig. 1) [19,29], we tested whether AphB effects on toxR expression affect toxT expression under the AKI virulence induction condition [22]. As expected, toxT expression in aphB mutants was significantly reduced as compared to that of wild type (Fig. 6B), however, bypassing the AphB regulation of tcpP by constitutively expressing tcpPH (pBAD-tcpPH induced AphB binds to the toxR promoter region to regulate toxR gene expression Figure 5 AphB binds to the toxR promoter region to regulate toxR gene expression. (A) Three different lengths of toxR promoter regions used in (B) were PCR amplified and cloned into pBBRlux containing a transcriptional lux reporter. The level of Lux induction with pBAD-aphB compared to pBAD24 in E. coli in the presence of 0.01% arabinose is given in the table. Alignment of putative AphB binding sites in tcpP, aphB, and toxR promoter region is given. (B) Gel shift assays using purified MBP-AphB and DNA containing various lengths of the regulatory regions of the toxR promoter. Protein concentrations used in the gel shift assay (shown as shaded triangles) were 0, 20, 40, 80 ng/reaction (5 μl).
with 0.01% arabinose) restored toxT expression in aphB mutants. These data suggest that AphB modulation of toxR expression has minor effects on virulence gene expression as compared to that of AphB regulation of tcpP under the condition we tested.

Conclusion
The ToxR regulon is the classic virulence gene regulation pathway in V. cholerae. In this pathway, AphA and AphB activate tcpP transcriptional expression directly by binding to different promoter regions of tcpP. ToxR and TcpP cooperate in turn by binding different sites of the toxT promoter to activate transcription, leading to the production of the virulence factors TCP and CT. However, the full ToxR regulon is more complex than previously thought. In this paper, we showed that AphA and AphB are also necessary for full ToxR production at the stationary phase. Furthermore, we demonstrated that AphB is sufficient for toxR transcriptional activation in the heterogenic host E. coli through binding of the toxR promoter region. Thus, the effect of AphB on ToxR levels propagates further in the transcription cascade, increasing the transcription of a key gene in V. cholerae pathogenesis, toxT. We have therefore identified another factor responsible for altering end product levels in the V. cholerae virulence axis. Since AphB is at the top of a virulence cascade with multiple end pathways, it appears now that AphB is a central factor in switching the cell from an environmental state to a virulent one. Since it activates ToxR in addition to TcpP, and further influences porin expression, AphB is a divergence point at which nonlinearity is introduced into the V. cholerae virulence pathway. Eukaryotic cells have extremely complex networks of protein and DNA interactions leading to precise control of protein expression levels. Having a more complex network of transcriptional activation and repression in the V. cholerae virulence cascade could enable the bacterial cell to fine-tune its expression levels to optimize its ability to colonize the intestine and spread to other hosts.

Bacterial strains, plasmids and media
All experiments were performed with El Tor Vibrio cholerae C6706 [30] or Escherichia coli DH5α, which were grown in LB with relevant antibiotics at 37°C, except where noted. V. cholerae virulence genes were induced in vitro (the AKI condition) as previously described [22]. Briefly, 3 ml of AKI medium was inoculated with 0.5 μl of overnight culture and incubated for 4 hrs at 37°C without agitation. 1 ml of culture was transferred to a fresh tube and incubated with shaking for a further 4 hrs at 37°C. P toxR -luxCDABE fusion plasmid was constructed by polymerase chain reaction (PCR) amplifying the toxR promoter regions, ranging from 450 bp, 300 bp, to 130 bp, respectively, and cloning them into the pBBRlux vector [20]. P toxT -luxCDABE plasmid was constructed by cloning toxT promoter regions into the pBBRlux vector. The chromosomal toxR-lacZ transcriptional fusion was constructed by cloning the 5' toxR region into the suicide vector pVIK112, which also contains a promoterless lacZ gene [31]. The resulting plasmid was then integrated into the The influence of AphB on V. cholerae outer membrane com-position, SDS resistance, and toxT expression Figure 6 The influence of AphB on V. cholerae outer membrane composition, SDS resistance, and toxT expression. (A). Analysis of outer membrane preparations of V. cholerae derivatives. SDS-PAGE gel stained with Coomassie blue. OmpT and OmpU are indicated at the right. The minimum bactericidal concentration (MBC) of SDS is listed below the SDS-PAGE gel. (B). Wild type or aphB mutant containing a P toxT -luxCDABE reporter plasmid with or without pBAD-tcpPH were grown under the AKI condition. 0.01% arabinose was added to induce P BAD -tcpPH. Lux expression (blue bars) was measured and normalized against toxT expression in wild type. The results are the average of three experiments ± SD. chromosomes of V. cholerae lacZstrains by homologous recombination to create a single-copy toxR-lacZ and an intact copy of toxR. P BAD -controlled aphA and aphB plasmids were constructed by cloning aphA and aphB coding sequences into the pBAD24 vector [32]. pBAD-tcpPH plasmid construct was described in [8]. In-frame deletions of toxR, toxS, tcpP, tcpA, toxT, aphA, and aphB were either described previously [15] or constructed by cloning the regions flanking target genes into the suicide vector pWM91 containing a sacB counter-selectable marker [33]. The resulting plasmids were introduced into V. cholerae by conjugation and deletion mutants were selected for double homologous recombination events.

Lux activity assays
Bacteria were grown at 37°C or 22°C under conditions indicated. At different time points, cultures were withdrawn and luminescence was measured by using a Bio-Tek Synergy HT spectrophotometer. Lux expression is calculated as light units/OD 600 .

Western blotting and SDS-PAGE electrophoresis
Whole-cell lysates were prepared from bacteria overnight cultures in LB conditions at 37°C and samples were normalized to the amount of total protein as assayed by the Biorad protein assay (Biorad). The isolation of outer membrane (OM) proteins from V. cholerae was performed using the method described by Miller and Mekalanos [34]. Whole-cell lysates or OM preparations were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) on a 10% polyacrylamide gel and stained with Coomassie brilliant blue for visualization. SDS-PAGE gels were transferred to nitrocellulose membrane for Western blot analysis using polyclonal rabbit anti-ToxR antibody.
Gel retardation assays MBP-AphB protein was purified through amylose columns according to the manufacturer's instructions (New England Biolabs). PCR products of the different lengths of toxR promoter regions were digested with EcoRI and endlabeled using [α-32 P]dATP and the Klenow fragment of DNA polymerase I. Binding reactions contained 0.1 ng of DNA and MBP-AphB proteins in a buffer consisting of 10 mM Tris-HCl (pH 7.9), 1 mM EDTA, 1 mM dithiothreitol, 60 mM KCl, and 30 mg of calf thymus DNA/ml. After 20 minutes of incubation at 25°C, samples were size-fractionated using 5% polyacrylamide gels in 1× TAE buffer (40 mM Tris-acetate, 2 mM EDTA; pH 8.5). The radioactivity of free DNA and AphB-DNA complexes was visualized by using a Typhoon 9410 PhosphorImager (Molecular Dynamics).

Authors' contributions
XX, AS, ZL, BK, and JZ designed research; XX, AS, and ZL performed research; XX, AS, and JZ analyzed data, XX, AS, ZL, BK, and JZ wrote the paper. All authors read and approved the final manuscript.