Interaction of the Histone-Like Nucleoid Structuring Protein and the General Stress Response Regulator RpoS at Vibrio cholerae Promoters That Regulate Motility and Hemagglutinin/Protease Expression

Abstract

The bacterium Vibrio cholerae colonizes the human small intestine and secretes cholera toxin (CT) to cause the rice-watery diarrhea characteristic of this illness. The ability of this pathogen to colonize the small bowel, express CT, and return to the aquatic environment is controlled by a complex network of regulatory proteins. Two global regulators that participate in this process are the histone-like nucleoid structuring protein (H-NS) and the general stress response regulator RpoS. In this study, we address the role of RpoS and H-NS in the coordinate regulation of motility and hemagglutinin (HA)/protease expression. In addition to initiating transcription of hapA encoding HA/protease, RpoS enhanced flrA and rpoN transcription to increase motility. In contrast, H-NS was found to bind to the flrA, rpoN, and hapA promoters and represses their expression. The strength of H-NS repression at the above-mentioned promoters was weaker for hapA, which exhibited the strongest RpoS dependency, suggesting that transcription initiation by RNA polymerase containing σ^S could be more resistant to H-NS repression. Occupancy of the flrA and hapA promoters by H-NS was demonstrated by chromatin immunoprecipitation (ChIP). We show that the expression of RpoS in the stationary phase significantly diminished H-NS promoter occupancy. Furthermore, RpoS enhanced the transcription of integration host factor (IHF), which positively affected the expression of flrA and rpoN by diminishing the occupancy of H-NS at these promoters. Altogether, we propose a model for RpoS regulation of motility gene expression that involves (i) attenuation of H-NS repression by IHF and (ii) RpoS-dependent transcription initiation resistant to H-NS.

INTRODUCTION

Cholera is an acute waterborne diarrheal disease caused by Vibrio cholerae of serogroups O1 and O139. This highly motile Gram-negative pathogen continues to be a major public health concern in areas of South Asia and Africa. Infecting Vibrio spp. that overcome the gastric acid barrier swim toward the intestinal mucosa and express two major virulence factors: the toxin coregulated pilus (TCP), required for intestinal colonization, and cholera toxin (CT), which is largely responsible for the profuse rice-watery diarrhea typical of this disease (16, 25). Later in infection, V. cholerae downregulates the expression of virulence factors and detaches to return to the environment (60). At this stage, the expression of motility and hemagglutinin (HA)/protease has been suggested to facilitate V. cholerae detachment from the intestinal mucosa (5, 17, 35, 47). Motility and HA/protease are positively regulated by the cyclic AMP (cAMP)-receptor protein (CRP), which acts by enhancing the quorum-sensing regulator HapR and the general stress response regulator RpoS (4, 29, 46). The expression of both phenotypes is diminished in response to an increase in the intracellular concentration of the second messenger cyclic diguanylate (c-di-GMP) (57).

The histone-like nucleoid structuring protein (H-NS) is a global regulator belonging to a family of small nucleoid-associated proteins that include the factor for inversion stimulation (FIS), the heat-unstable protein (HU), and integration host factor (IHF) (14, 15). Mutations that inactivate hns are highly pleiotropic and diminish bacterial growth, suggesting that H-NS influences a broad spectrum of physiological processes (1, 2, 23). H-NS consists of an N-terminal oligomerization domain connected by a flexible linker to a nucleic acid binding domain (2, 8, 14, 36). Both oligomerization and DNA binding are required for the biological activity of H-NS, which includes DNA condensation and the regulation of transcription (10, 50). In transcription regulation, H-NS has been shown to negatively affect gene expression by binding to promoters exhibiting AT-rich highly curved DNA regions that contain clusters of the more conserved 10-bp motif TCGATAAATT (28, 40, 55). In addition, H-NS can positively or negatively affect the expression of a broader spectrum of genes by acting indirectly or binding to mRNA to affect translation (7). A common theme in H-NS transcription regulation is the silencing of horizontally acquired genes (30, 34, 39). Consistent with this role, H-NS has been shown to silence virulence gene expression in V. cholerae by acting at different levels of the ToxR regulatory cascade, which includes the toxT, tcpA, and ctxA promoters (37). Further, V. cholerae hns mutants have been reported to exhibit diminished motility and intestinal colonization capacity (18, 27, 49, 53). There are numerous evidences indicating that repression by H-NS can be relieved in response to environmental stimuli that activate the expression of other regulators whose binding site overlaps with that of H-NS. For instance, transcriptional silencing of V. cholerae tcpA and ctxA promoters by H-NS is antagonized by the AraC-like transcriptional regulator ToxT and IHF (51, 52, 59).

The alternative sigma factor RpoS (σ^S) is a global regulator that controls the expression of more than 100 genes in response to environmental stresses (20). V. cholerae rpoS mutants are more sensitive to starvation, high osmolarity, and oxidative stresses, are less motile than the wild type (WT), and do not express HA/protease (35, 49, 58). We recently reported that RpoS diminishes the cellular concentration of c-di-GMP, an inhibitor of flagellar motility (57). Consistently, microarray studies have shown that rpoS mutants express reduced levels of multiple motility and chemotaxis genes, suggesting that rpoS could act at an early stage of the motility regulatory cascade (35). In a previous study, we addressed the role of H-NS in the regulation of V. cholerae RpoS and RpoS-dependent genes that affect motility and HA/protease production (49). We found that H-NS posttranscriptionally affects RpoS expression in a positive manner, which in turn enhances motility and HA/protease production (49). Accordingly, hns mutants were found to be significantly less motile than the wild type and secrete less azocasein activity to the culture medium (49). However, an ΔrpoS Δhns double mutant exhibited a slightly larger swarm diameter than the Δhns mutant, suggesting a more complex interplay between H-NS and RpoS in the regulation of motility. In addition, deletion of hns in an ΔrpoS background resulted in higher expression levels of flaA, flaC, and motX mRNA (49). These results suggested a model in which H-NS can positively affect motility by enhancing RpoS but could also function as a repressor in the absence of RpoS (49). In this study, we further examine the role of H-NS in transcription regulation of motility and HA/protease expression. To this end, we have analyzed the interaction between H-NS and RpoS at the flrA, rpoN, and hapA promoters. We show that H-NS binds to these promoters to repress transcription while RpoS acts to attenuate H-NS transcriptional silencing indirectly by enhancing the expression of IHF and directly by promoting transcription initiation resistant to H-NS.

MATERIALS AND METHODS

Strains and media.

The strains, plasmids, and oligonucleotide primers used in this study are listed and briefly described in Tables 1 and 2. Mutants and reporter strains were all derived from V. cholerae C7258 (El Tor biotype; Ogawa). V. cholerae strains were grown in tryptic soy broth (TSB) with agitation (225 rpm) at 37°C. For cloning purposes, Escherichia coli strains TOP10 (Invitrogen) and S17-1λpir (11) were grown in LB medium at 37°C. When necessary, the culture medium was supplemented with ampicillin (Amp; 100 μg/ml), kanamycin (Km; 25 μg/ml), rifampin (Rf; 150 μg/ml), polymyxin B (PolB; 100 units/ml), isopropyl-β-d-thiogalactopyranoside (IPTG; 20 μg/ml), or 5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside (X-Gal; 20 μg/ml).

Table 1.

Strains and plasmids

Strain or plasmid	Description	Source or reference(s)
Strains
C7258	Wild type, El Tor biotype, Perú 1991	Clinical isolate
C7258HNS-FLAG	C7258 hns::hns-FLAG	This study
C7258ΔlacZ	C7258 lacZ deletion mutant	49
AJB80	C7258 ΔlacZ Δhns::Km	49
AJB50	C7258 ΔrpoS	29
AJB50HNS-FLAG	AJB50 hns::hns-FLAG	This study
AJB50ΔlacZ	C7258 ΔlacZ ΔrpoS	49
AJB81	C7258 ΔlacZ ΔrpoS Δhns::Km	49
C7258ΔlacZ RpoS-FLAG	C7258 ΔlacZ rpoS::pCVDRpoS-FLAG	This study
AJB80RpoS-FLAG	AJB80 rpoS::pCVDRpoS-FLAG	This study
HX120	C7258 ΔihfA	This study
HX121	C7258 ΔrpoS ΔihfA	This study
HX120HNS-FLAG	HX120 hns::hns-FLAG	This study
HX121HNS-FLAG	HX121 hns::hns-FLAG	This study
Plasmids
pHNS-FLAG	hns ORF cloned in pFLAG-CTC	Sigma-Aldrich
pTT3	rrnB T1T2 transcription terminator in pUC19	48
pTT3HNS-FLAG	hns-FLAG fusion cloned in pTT3	This study
pCVDHNS-FLAG	hns-FLAG-rrnB T1T2 cassette in pCVD442	This study
pKRZ1	Plasmid containing promoterless lacZ gene	44
pTT3flrA	390-bp flrA promoter region cloned in pTT3	This study
pFlrA-LacZ	850-bp fragment containing rrnB T1T2 and flrA promoter in pKRZ1	This study
pTT3flaA	450-bp DNA fragment carrying the flaA promoter in pTT3	This study
pFlaA-LacZ	890-bp DNA fragment containing rrnB T1T2 and flaA promoter cloned in pKRZ1	This study
pLacZ	Promoterless lacZ gene transferred from pKRZ1 to pBR322	This study
pTT4	Transcription terminator rrnB T1T2 reinserted in pUC19 as a KpnI-BamHI fragment	This study
pTT4RpoN	410-bp DNA fragment carrying the rpoN promoter in pTT4	This study
pRpoN-lacZ	DNA fragment containing rrnB T1T2 and rpoN promoter ligated to promoterless lacZ gene in pLacZ	This study
pHapLac11	Plasmid vector containing transcriptional hapA-lacZ fusions	46, 48
pTXB1	Expression vector for construction of in-frame fusions with the intein/chitin binding domain	New England BioLabs
pTXB1-HNS	hns ORF cloned in plasmid pTXB1	This study
pTXB1-RpoS	rpoS ORF cloned in plasmid pTXB1	This study
pCVDRpoS-FLAG	RpoS-FLAG fusion in suicide vector pCVD442	This study
pUCΔihfA	0.5-kb SacI-BamHI and 0.5-kb BamHI-SphI fragments flanking ihfA sequentially cloned in pUC19	This study
pUCΔihfA-Km	1.2-kb Kmr gene in BamHI site of pUCΔihfA	This study
pCVDΔihfA-Km	SacI-SphI ΔihfA::Km cassette in pCVD442	This study

Table 2.

Oligonucleotide primers

Primer name	Sequence (5′→3′)a
FlaAp-F	GAAGCATGCCCTTCAATGCCTTATGC
FlaAp-R	GCGAAGCTTGGTCATCGCCGACAC
FlrA-R1	GCTTCCGTGGTCGAGACTTGC
FlrA723	GCCGATGAGCATGCAGGTTAAACT
FlrA884	TCGGGAACACGTTTAAGCGGTAGT
FlrA-R2	GCCCAACAAGGCTACGAAACAG
?>FlrA-R3	TCCGAATTCACACTGCTCTCCTACA
FlrA-F41	GCCGAACCATAAAAATGATCCGCAA
FlrA-R42	CAGGCTAGTTAGTGGCTTCTATTCT
FlrAp-F	GAAGCATGCCTGATCCAAGGTAAG
FlrAp-R	GCGAAGCTTAGTGGCTTCTATTCTATT
HapA-F29	GGTGAATGTATTAAGCGTTGAACCG
HapA-R259	GGTAGGTGTCAAATTTTAAAGGCGC
HNS-F31	GATCGCATATGGTAATGTCGGAAATCACTAAG
HNS-R32	GATCGGCTCTTCAGCACAGAGCGAATTCTTCCAGAGA
IHFA11	AAAGGATCCGTGAGCGCCATAAAACTTCC
IHFA305	GAAGGATCCGTCGAAAAATAAGACCGAGC
IHFA504	GCCGAGCTCCGCAGCAGCGGTAGAGG
IHFA814	TAAGCATGCGGCGAGTCTCGTCACCG
IHFA-F	GATGCCAAGGATACGGTTGAGGTG
IHFA-R	CACGTCGAGCGGTAATAGGAATATC
IHFB-F	AGAAAGACTCTGTGCCGAACAAACG
IHFB-R	TTCCAACTCCACCTTATCGCCAGTC
RecA578	GTGCTGTGGATGTCATCGTTGTTG
RecA863	CCACCACTTCTTCGCCTTCTTTGA
RnnB-F	CGGGGTACCGATTTTCAGCCTGATAC
RnnB-R	CGCGGATCCTGGCTTGTAGATATGAC
RpoN596	CCTTTGGTGTGGCATCGCTTAATC
RpoN875	TGACCAGCCATCGACCTTGTTCTT
RpoN-F21	CCCGATAAAAGTCTTCATCCAGCA
RpoN-R22	GAATTCACGCAAGGTCAGGCATCT
RpoNp-R	CTAGTCTAGATTGGCCGCCAATTTTGAC
RpoNp-F	CGCGGATCCTTTGGGGACATGCAGATC
RpoS1020	CTTCTGCAGTTACTTGTCGTCATCG
RpoS-F11	CCCATATGAGTGTCAGCAATACCGTAACCAA
RpoS-R12	ACGCTCTTCAGCAGTTGTCGTATTCGACGTT
RpsM-F51	GCAACTGCGTGATGGTGTAGCTAA
RpsM-R52	GCTTGATCGGCTTACGCGGACC
TcpA-F1	GTTCATAATTTCGATCTCCACTCCG
TcpA-R2	GTTAACCACACAAAGTCACCTGCAA
VC1922-F61	TAGAAGGTTGACGAAACAAGCAATCA
VC1922-R62	GGTTCAACCACCATAGGTACGAGT
VC2520-F	ATGGGCACTTCACTGTTAGGT
VC2522-F	GGCGATGCTTTGCTGGATGAAC
VC2522-R	CGTTTGAGCCGATAATGTTACCG
VC2522-R1	CACCAATCATCATCGGCAGTTCTCG
VC2522-R2	GGCAATGTTTGAACCGAGGACG
VC2522-R3	AAAGAATTCACCATCATCTCGGGTGCTGAG
VC2523-F	ACTGACGGTGATTTACGGCG
VC2523-R	CAAGAGTTTACGCCCAAGTGC
VC2524-F	CATTACTGGCCGCCGCTCA
VC2524-R	TCCAGTTCGTTACGCGCTTG
VC2525-F	GGGAAATCACTGCACGCAGAG
VC2525-R	CTTTCATCGCTTGACCGTGTG
VC2527-F	CGTCGCAACCTTCTCGCAACT
VC2527-R	CCCTTTGCCATCACTATCGG
VC2528-F	CGCAAACTCTCGGTCGAAGAC
VC2528-R	GCCTTGGCTGACAATATACGC
VC2529-R	GGTTAGATTCCAATGCTTCCTG
VicH5	GGGAAGCTTGTAATGTCGGAAATCA
VicH7	GTTGCATGCATGTCGGAAATCACTA
VicH396	AGGAGATCTCAGAGCGAATTCTTCC
VicH567	GAGTTCCTGCACCTGCTTTTATATG

^aRestriction sites used for directional cloning are underlined.

Construction of mutants and reporter strains expressing H-NS and RpoS proteins tagged with the FLAG epitope.

To construct a strain containing a deletion and Km insertion in ihfA, encoding the IHF A subunit, DNA fragments flanking the ihfA locus (VC1222) were amplified using primer combinations IHFA504/IHFA11 and IHFA305/IHFA814 and an Advantage 2 PCR kit (BD Biosciences Clontech). The PCR products were sequentially cloned in pUC19 to yield pUCΔihfA. This plasmid was modified by insertion of a Km-resistant cassette from pUC4K (GenBank accession number X06404) to generate pUCΔihfA-Km, and the entire V. cholerae DNA harboring the Km-resistant gene replacing the ihfA open reading frame (ORF) was transferred to pCVD442 (13) to yield pCVDΔihfA-Km. Finally, the above-mentioned suicide vector was transferred to strains C7258 and AJB50, and mutants HX120 and HX121 (Table 1) were isolated by sucrose selection and confirmed by PCR and DNA sequencing. V. cholerae strains expressing H-NS tagged at its C terminus with the FLAG epitope were constructed to conduct chromatin immunoprecipitation (ChIP). To this end, the hns ORF (vicH, VC1130) lacking the stop codon was amplified from C7258 genomic DNA using primers VicH5 and VicH396. The amplification product was confirmed by DNA sequencing and cloned as a BglII-HindIII fragment in pFLAG-CTC (Sigma-Aldrich) in frame with the FLAG epitope to yield pHNS-FLAG. The H-NS–FLAG fusion was retrieved from pHNS-FLAG by PCR with primers VicH7 and RpoS1020, containing SphI and PstI overhangs, and cloned in pTT3 (48) upstream of the rrnB T₁T₂ transcription terminator to yield pTT3HNS-FLAG. The hns-FLAG-rrnB T₁T₂ cassette was transferred to the suicide vector pCVD442 to yield pCVDHNS-FLAG. Finally, pCVDHNS-FLAG was introduced in strains C7258, AJB50, HX120, and HX121 by conjugal transfer from S17-1λpir to generate strains C7258HNS-FLAG, AJB50HNS-FLAG, HX120HNS-FLAG, and HX121HNS-FLAG, respectively. Exconjugants were selected in LB agar containing Amp and PolB, and correct integration into the hns locus was confirmed by PCR using primer VicH567, which anneals to DNA sequences upstream of hns not present in pCVDHNS-FLAG, and RpoS1020, which anneals to DNA encoding the FLAG epitope. For strains expressing a chromosomally integrated rpoS-FLAG allele from native transcription/translation signals, the suicide vector pCVDRpoS-FLAG (57) was transferred by conjugation to C7258ΔlacZ and AJB80 (Table 1). Exconjugants C7258ΔlacZ RpoS-FLAG and AJB80 RpoS-FLAG were selected in LB agar containing Amp and PolB, and correct integration within the rpoS locus was confirmed by DNA sequencing.

RT-PCR.

V. cholerae cultures were treated with RNAprotect bacterial reagent (Qiagen), and total RNA was isolated using an RNeasy kit and RNase-free DNase set (Qiagen). Reverse transcription-PCR (RT-PCR) was performed with Superscript III reverse transcriptase (Invitrogen) according to the manufacturer's instructions. Quantitative RT-PCR (qRT-PCR) was conducted using an iScript two-step RT-PCR kit with SYBR green (Bio-Rad Laboratories). Relative expression values were calculated as 2^{(CtTarget − CtReference)}, where Ct is the fractional threshold cycle for the target gene and the reference is the recA mRNA. The following primer combinations were used: FlrA723/FlrA884 for flrA mRNA, IHFA-F/IHFA-R for ihfA mRNA, IHFB-F/IHFB-R for ihfB mRNA, RecA578/RecA863 for recA mRNA, and RpoN596/RpoN875 for rpoN mRNA.

Transcriptional start mapping of the flrA and rpoN promoters by 5′ RACE analysis.

To locate the rpoN promoter region, cDNA was synthesized using random hexamers and 20 ng of total RNA. Then, PCRs were carried out using primer combinations consisting of a forward primer annealing to each gene of the putative rpoN operon and a reverse primer annealing to the adjacent 3′ open reading frame. A control using total RNA as the template was run for each reaction to exclude chromosomal DNA contamination. Once the rpoN promoter region was identified, 5′ rapid amplification of cDNA ends (5′ RACE) was used to determine the transcription initiation site. Similarly, 5′ RACE was conducted to determine the transcription initiation site of flrA. To this end, first-strand cDNA synthesis was conducted using a second-generation 5′/3′ RACE kit (Roche Applied Sciences). Briefly, cDNA was synthesized from 5 μg of total RNA using primer FlrA-R1 for flrA and primer VC2522-R1 for rpoN as gene-specific primers. The cDNA was treated with RNase H and RNase T1, followed by purification with a High Pure PCR product purification kit (Roche Applied Sciences). Next, cDNA was incubated with terminal transferase for 30 min and the deoxyribosyladenine (dA)-tailed cDNA was amplified with FastStart Taq DNA polymerase (Roche Applied Sciences) according to the 5′/3′ RACE kit protocol and by using primers FlrA-R2 and VC2522-R2 for flrA and rpoN, respectively. Finally, nested PCRs were conducted using each amplified tailed cDNA as a substrate and primers FlrA-R3 and VC2522-R3 for flrA and rpoN, respectively. The nested PCR products were ligated into SalI- and EcoRI-digested pUC19, and 10 positive colonies were sequenced for each 5′ RACE experiment.

Construction of flrA-, flaA-, and rpoN-lacZ promoter fusions.

To construct an flrA-lacZ promoter fusion, we amplified a 390-bp fragment containing the flrA promoter (as defined by 5′ RACE analysis) with primers FlrAp-F and FlrAp-R. For the flaA-lacZ promoter fusion, a 450-bp fragment containing the flaA promoter (26) was amplified with primers FlaAp-F and FlaAp-R. In both cases, the promoter fragments were inserted downstream of the rrnB T₁T₂ transcription terminator in plasmid pTT3 to generate pTT3flrA and pTT3flaA, respectively. Then, the terminator-promoter fragment was inserted upstream of a promoterless lacZ gene in plasmid pKRZ1 (44) to generate pFlrA-LacZ and pFlaA-LacZ. To construct an rpoN-lacZ promoter fusion, we first constructed plasmid pLacZ by ligating a 4.4-kb BamHI-PstI fragment containing the lacZ gene from pKRZ1 to a 3.2-kb BamHI-PstI fragment from pBR322. The rrnB T₁T₂ transcription terminator was amplified from pTT3 (48) with primers RrnB-F and RnnB-R and reinserted in pUC19 to generate pTT4. Next, a 410-bp fragment containing the rpoN promoter region was amplified with primers RpoNp-F and RpoNp-R and inserted in pTT4 to generate pTT4RpoN. Finally, the terminator-promoter fragment was inserted upstream of the promoterless lacZ gene in plasmid pLacZ to generate pRpoN-lacZ. The construction of plasmid pHapLac11 containing a hapA-lacZ promoter fusion has been described previously (46, 48). The resulting plasmids, pFlrA-LacZ, pFlaA-LacZ, pRpoN-lacZ, and pHapLac11, were introduced into the C7258ΔlacZ (WT), AJB50ΔlacZ (ΔrpoS), AJB80 (Δhns), and AJB81 (ΔrpoS Δhns) strains, respectively, by electroporation (31).

Purification of H-NS and RpoS and preparation of polyclonal antibodies.

To purify H-NS, the hns open reading frame (ORF) was amplified from C7258 genomic DNA using primers HNS-F31 and HNS-R32. The amplified fragment was digested with restriction enzymes NdeI and SapI and ligated into similarly digested pTXB-1 (Table 1) to generate pTXB1-HNS. The H-NS–intein fusion was confirmed by DNA sequencing using T7 universal and Mxe intein reverse primers from New England BioLabs. H-NS was then expressed from the T7 promoter in E. coli ER2566 and purified using an Impact kit (New England BioLabs). Briefly, E. coli ER2566 containing plasmid pTXB1-HNS was grown in shaken flasks (220 rpm) containing LB supplemented with Amp at 37°C to an optical density at 600 nm (OD₆₀₀) of 0.3 to 0.4. At this point, the expression of H-NS was induced with IPTG (0.4 mM) and the culture was incubated for 3 h at 29°C. The cells were collected by centrifugation, resuspended in 20 mM Tris-HCl, pH 8.0, 0.5 M NaCl, and 1 mM EDTA, and disrupted by sonication. The cell debris was removed by centrifugation, and H-NS was purified by following the Impact kit protocol. H-NS-containing fractions were combined and dialyzed against 20 mM Tris-HCl, pH 8.0, 1 mM EDTA, and 5 mM sodium citrate at 4°C. The purity of recombinant H-NS determined by SDS-PAGE was found to be higher than 90%. For purification of RpoS, primers RpoS-F11 and RpoS-R12 were used to amplify the rpoS ORF. The amplification product was subcloned in plasmid pTXB1 to yield pTXB1-RpoS, and the construct was confirmed by DNA sequencing as described above. Finally, RpoS was expressed and purified as described for H-NS. Custom polyclonal antibody production was conducted at SouthernBiotech (Birmingham, AL). Protein concentrations were determined by the Bradford method, using a Bio-Rad protein assay kit according to the manufacturer's instructions.

EMSA.

Electrophoresis mobility shift assays (EMSA) were conducted using a second-generation digoxigenin (DIG) gel shift kit (Roche Applied Sciences). Briefly, the reaction mixtures consisted of 20 mM HEPES, pH 7.6, 1 mM EDTA, 10 mM (NH₄)₂SO₄, 1 mM dithiothreitol (DTT), 0.2% Tween 20, 30 mM KCl, 25 ng of calf thymus DNA (Sigma-Aldrich), 3 fmol of DIG-labeled target DNA, and 12 to 48 ng of pure H-NS in a total volume of 20 μl. The binding reaction mixtures were incubated for 20 min at 30°C. Finally, protein-DNA complexes were separated by electrophoresis in 5% Tris-borate-EDTA (TBE) polyacrylamide gels and transferred to nylon membranes, and DNA was visualized using an anti-DIG Fab fragment-AP conjugate, followed by chemiluminescence detection. The following primer combinations were used to amplify the promoter regions under study: FlrA-F41 and FlrA-R42 for flrA, HapA-F29 and HapA-R259 for hapA, RpoN-F21and RpoN-F22 for rpoN, and TcpA-F1 and TcpA-R2 for tcpA. The promoter of VC1922, an ortholog of Salmonella enterica serovar Typhimurium STM1033 not regulated by H-NS (34), was amplified using primers VC1922-F61 and VC1922-F62.

ChIP.

For H-NS promoter occupancy, strains expressing H-NS-FLAG were grown to the stationary phase in TSB medium (16 h). Then, 40 ml of culture was sequentially treated with Rf (20 min, 37°C), 1% formaldehyde (cross-linking, 10 min, 30°C), and 227 mM glycine (30 min, 4°C). Cells were collected by centrifugation, washed twice with phosphate-buffered saline (PBS) supplemented with protease inhibitor cocktail (PIC) and phenylmethylsulfonyl fluoride (PMSF; Roche Applied Science), and divided into aliquots equivalent to 1/(culture OD₆₀₀ reading) ml, and the cell pellets were maintained at −80°C if not processed immediately. Next, the cells were lysed by suspending the frozen pellets in 500 μl of 10 mM Tris-HCl, pH 8.0, 50 mM NaCl containing 20 ng/μl of RNase A, and 10⁵ kU of Ready-Lyse lysozyme (Epicentre Biotechnologies), followed by a 30-min incubation at 37°C. One volume of double-strength immunoprecipitation (IP) buffer (200 mM Tris-HCl, pH 7.5, 600 mM NaCl, 4% Triton X-100) containing PIC and PMSF was added to each lysate, and DNA was broken down to a range of 150 to 1,000 bp by sonication. The cell debris was removed by centrifugation, and the lysate was diluted 10-fold in IP buffer. At this stage, a 10-μl input sample was saved as a reference and PCR efficacy control. Protein-DNA complexes were immunoprecipitated by overnight incubation at 4°C with 8 μg of anti-FLAG M2 monoclonal antibody (Sigma-Aldrich) or 8 μg of an unrelated anti-Xpress monoclonal antibody (Invitrogen) for a mock ChIP. The antibody-protein-DNA complexes were pulled down with salmon sperm DNA-treated protein A/G agarose beads (Imgenex, San Diego, CA) for 1 h at 4°C. The beads were washed twice with 100 mM Tris-HCl, pH 7.5, 250 mM LiCl, and 2% Triton X-100, collected in Spin-X centrifuge tube filters (Costar), and washed three times with IP buffer containing 600 mM NaCl, IP buffer, and TE buffer (10 mM Tris-HCl, pH 8.0, 1 mM EDTA). The immunoprecipitated complexes were eluted from the beads by incubation at 65°C for 30 min in TE buffer containing 1% SDS. After reversal of cross-linking (4 h, 65°C), proteins were removed by incubation with 20 μg of proteinase K (1 h, 45°C). Then, immunoprecipitated DNA was purified using a MiniElute PCR purification kit (Qiagen). At least three ChIP assays were conducted for each strain.

Detection of immunoprecipitated DNA.

Immunoprecipitated DNA was qualitatively detected by PCR and agarose gel electrophoresis using the primer combinations to amplify the promoter fragments described for the EMSA. Real-time quantitative PCR (qPCR) was used to quantitate promoter occupancy by H-NS. To this end, PCR was conducted using iTaq SYBR green supermix with ROX (Bio-Rad). The quantity of immunoprecipitated DNA was calculated as the percentage of the DNA present in the input sample using the formula IP = 2^{(Ctinput − CtIP)}, where Ct is the fractional threshold cycle of the input and IP samples. The relative IP was calculated by standardizing the IP of each sample by the IP of the corresponding mock ChIP. Finally, a region within the rpsM ORF to which H-NS does not bind acting as a transcriptional repressor was amplified with primers RpsM-F51 and RpsM-R52 and used as a negative control.

Western blot detection of H-NS and RpoS.

For detection of native H-NS and RpoS proteins, serum exhibiting high anti-H-NS and anti-RpoS titers was first preabsorbed with crude extracts of V. cholerae Δhns and ΔrpoS mutants, respectively. To this end, cells were collected by centrifugation at 4°C and resuspended in Tris-buffered saline containing 0.1% Tween 20 (TBST) supplemented with PIC. The cells were disrupted by sonication, and the debris was removed by centrifugation. Then, a 1:1,000 dilution of anti-H-NS or anti-RpoS serum was incubated overnight at 4°C with crude extracts of strain AJB80 (Δhns) or AJB50 (ΔrpoS) containing 2 mg/ml protein, respectively. For Western blot analysis, a volume of cells corresponding to 1.0 OD₆₀₀ units was centrifuged and the pellet was resuspended in 0.1 ml of Laemmli's sample buffer (Bio-Rad Laboratories). The cell suspension was placed in a boiling water bath for 10 min, and the cell debris was removed by centrifugation. Proteins were separated using Criterion Precast 10% gels (Bio-Rad) and transferred to polyvinylidene difluoride (PVDF) membranes. The expression of H-NS and RpoS was determined using the corresponding preabsorbed antiserum and horseradish peroxidase (HRP)-conjugated anti-rabbit IgG (Thermo Fisher Scientific, Rockford, IL). A similar procedure was used to detect RpoS-FLAG and H-NS-FLAG but by using monoclonal anti-FLAG M2-peroxidase (Sigma-Aldrich). Membranes were developed using a BM bioluminescence Western blotting kit (Roche Applied Science). To estimate the concentration of H-NS and RpoS in cell lysates, known quantities of each protein were analyzed by Western blotting. Then, a standard curve was constructed for each regulator by plotting band intensities determined using TotalLab Quant software (TotalLab Ltd., Newcastle upon Tyne, United Kingdom) versus concentration.

Enzyme assays.

β-Galactosidase activity was measured as described by Miller (33) using the substrate o-nitrophenyl-β-d-galactopyranoside. Specific activities are given in Miller units [1,000 × (OD₄₂₀/t × v × OD₆₀₀)], where t is the reaction time and v is the volume of enzyme extract per reaction.

TEM.

For transmission electron microscopy (TEM), the cells were pelleted and fixed by reconstitution in 2.5% glutaraldehyde sodium cacodylate buffer, pH 7.3. Samples were adhered to a carbon-coated grid and stained with 1% uranyl acetate before microscopy.

RESULTS

Deletion of rpoS partially suppresses the slow-growth phenotype of the hns mutant.

In our previous study, we observed that an ΔrpoS Δhns double mutant exhibited a slightly larger swarm diameter than the Δhns single mutant (49). Since differences in growth rate could affect the swarm diameter observed in semisolid agar, we examined the growth patterns of the WT, ΔrpoS, Δhns, and ΔrpoS Δhns strains. As shown in Fig. 1A, the Δhns mutant exhibited diminished growth compared to the WT and ΔrpoS strains. Interestingly, deletion of rpoS in the hns mutant led to a partial suppression of its slow-growth phenotype. Transmission electron microscopy showed an abundance of elongated cells in the Δhns mutant (Fig. 1B). However, the presence of elongated cells was not suppressed by deletion of rpoS (Fig. 1B). It is noteworthy that both the Δhns and ΔrpoS Δhns mutants were flagellated but still appeared similarly depressed for motility by hanging-drop bright-field microscopy examination.

Fig 1.

Growth pattern and morphology of V. cholerae Δhns mutants. (A) Growth. C7258ΔlacZ (WT; □), AJB50ΔlacZ (ΔrpoS; ■), AJB80 (Δhns; ▲), and AJB81 (ΔrpoS Δhns; △) strains were grown with agitation in TSB medium at 37°C, and growth was monitored by measuring the OD₆₀₀. (B) Transmission electron microscopy. Morphology of C7258ΔlacZ, AJB50ΔlacZ, AJB80, and AJB81 strains grown to the stationary phase in TSB medium.

The finding that deletion of rpoS partially suppressed the slow-growth phenotype of an Δhns mutant suggested that, as observed in E. coli, H-NS acts as a repressor of many RpoS-dependent genes (3). Thus, we used specific H-NS and RpoS antisera to measure the expression of these regulators at different stages of the bacterial growth curve. As shown in Fig. 2, H-NS can be detected in both exponentially growing and stationary-phase cultures, while RpoS was induced in the late logarithmic phase (OD₆₀₀ ≥ 2). Thus, we decided to use cells in the stationary phase (expressing both proteins) for subsequent studies on the interaction between these global regulators.

Fig 2.

Detection of H-NS (A) and RpoS (B) expression levels. The V. cholerae C7258ΔlacZ strain was grown in TSB medium, and samples were withdrawn for analysis in the log phase (OD₆₀₀, 0.5) (lane 1), late log phase (OD₆₀₀, 2.0) (lane 2), and stationary phase (4 h after the culture reached an OD₆₀₀ of 2.0) (lane 3). H-NS and RpoS protein concentrations are shown below the bands.

Regulation of the flrA, rpoN, hapA, and flaA promoters by RpoS and H-NS.

Microarray studies have shown that deletion of rpoS results in reduced expression of motility and chemotaxis genes belonging to the class II, III, and IV hierarchies (35). Therefore, we chose to examine the role of RpoS and H-NS in the transcription of the upstream rpoN and flrA regulator genes. To specifically address the effect of RpoS and H-NS on the rpoN and flrA promoters, we determined the transcription initiation sites of both genes and constructed lacZ promoter fusions.

The rpoN gene is predicted to be part of a large operon, suggesting that it might lack its own promoter (26). By using RT-PCR and different primer combinations, we found that the rpoN promoter is located in the intergenic region between VC2520 and VC2522 (data not shown). We then used 5′ RACE analysis to determine the transcription initiation site of rpoN and its cognate σ⁵⁴-dependent activator flrA. Sequencing of the 5′ RACE 500-bp nested PCR product amplified using primer VC2522-R3 indicated that rpoN transcription is initiated at a cysteine located 277 bp upstream of the start codon of VC2522. Further, sequencing of the 5′ RACE 250-bp nested PCR product amplified using primer FlrA-R3 indicated that flrA transcription is initiated at a cysteine located 146 bp upstream of the flrA start codon. Based on this information, we constructed lacZ fusions consisting of the rpoN and flrA promoters linked to a promoterless lacZ gene. A similar fusion consisting of the RpoS-dependent hapA promoter linked to a promoterless lacZ gene has been described previously (48). Expression of the above-noted lacZ fusions were examined in the WT, ΔrpoS, Δhns, and ΔrpoS Δhns genetic backgrounds. As shown in Fig. 3, transcription of rpoN, flrA, and hapA was diminished in the ΔrpoS mutant while deletion of hns enhanced their transcription. However, flrA and rpoN differed from hapA in that the first two promoters were less dependent on RpoS but more strongly repressed by H-NS. The expression of flrA and rpoN in the absence of both H-NS and RpoS reflects the level of unrepressed transcription initiation by RNA polymerase (RNAP) containing σ⁷⁰, which was negligible in the case of hapA. We also examined the effect of the regulation on the expression of flaA, a gene belonging to the downstream class III transcription hierarchy encoding the major V. cholerae flagellin (Fig. 3). As predicted, regulation of flaA in stationary-phase cultures followed the same pattern as that of its upstream regulators RpoN and FlrA.

Fig 3.

Expression levels of flrA-, rpoN-, flaA-, and hapA-lacZ promoter fusions in V. cholerae ΔrpoS and Δhns mutants. C7258ΔlacZ (WT), AJB50ΔlacZ (ΔrpoS), AJB80 (Δhns), and AJB81 (ΔrpoS Δhns) strains containing flrA-, rpoN-, flaA-, and hapA-lacZ promoter fusions were grown to the stationary phase in TSB at 37°C. Detection of β-galactosidase activity was measured as described in Materials and Methods and expressed in Miller units. Each value is the mean for six independent cultures. The error bars indicate standard deviations (*, significantly different from that of the wild type [P < 0.01 by a one-tailed t test]). The spans of the promoter fragments used relative to the start codon were as follows: positions −536 to −142 for flrA, positions −685 to −273 relative to VC2522 for rpoN, positions −407 to +43 for flaA, and positions −410 to +3 for hapA.

H-NS binds to the flrA, rpoN, and hapA promoters in vitro.

Our results suggested that H-NS negatively affects the transcription of rpoN, flrA, and hapA. Therefore, we decided to examine whether pure H-NS can bind to these promoters. H-NS is known to bind DNA with a more relaxed sequence specificity than that of other regulators that have more-stringent binding requirements. Thus, we used the tcpA promoter, a gene silenced by H-NS (37), and the region upstream of VC1922 as positive and negative controls, respectively. VC1922 is an ortholog of Salmonella STM1033 located in a chromosomal region not occupied by H-NS, as described previously in a ChIP-on-chip study (34). We used qRT-PCR to confirm that this gene is also not affected by H-NS in V. cholerae (data not shown). As shown in Fig. 4, H-NS was found to bind to the tcpA, flrA, rpoN, and hapA promoters while very little binding to the VC1922 promoter could be detected.

Fig 4.

Binding of H-NS to the tcpA, flrA, rpoN, and hapA promoters. DIG-labeled DNA fragments encoding the tcpA (A), flrA (B), rpoN (C), hapA (D), and VC1922 (E) promoters were incubated with pure H-NS protein. In panels A to D, labeled DNAs were incubated with 0 (lanes 1), 12 (lanes 2), 18 (lanes 3), and 24 (lanes 4) ng of H-NS and with 24 ng of H-NS and a 62.5-fold excess of unlabeled competitor DNA (lanes 5). In panel E, labeled DNA was incubated with 0 (lane 1), 12 (lane 2), and 24 (lane 3) ng of H-NS and with 24 ng of H-NS and a 62.5-fold excess of unlabeled competitor DNA (lane 4). The spans of the promoter fragments used relative to the start codon were as follows: positions −306 to −83 for tcpA, positions −356 to −136 for flrA, positions −447 to −229 relative to VC2522 for rpoN, positions −272 to −19 for hapA, and positions −113 to +43 for VC1922.

RpoS negatively affects H-NS promoter occupancy.

To determine H-NS occupancy at the flrA, rpoN, and hapA promoters at the cellular level, we constructed strains C7258HNS-FLAG and AJB50HNS-FLAG (Table 1) and conducted ChIP. In these experiments, we used two negative controls: (i) a DNA sequence located within an ORF of the housekeeping gene rpsM and (ii) the promoter region of VC1922, to which H-NS failed to exhibit significant binding affinity in the EMSA. The rationale for the rpsM ORF control was that though H-NS could bind to this sequence, such binding would not be relevant to transcription regulation and would provide a cutoff for distinguishing H-NS binding as a repressor versus nucleoid organization. In the ChIP analysis whose results are shown in Fig. 5, the tcpA promoter exhibited the highest H-NS occupancy, followed by flrA and hapA. The relative IP values obtained for VC1922 and rpsM were not considered indicative of H-NS transcription repression activity. According to these criteria, H-NS occupancy at the rpoN promoter was not significant under the experimental conditions used. Interestingly, deletion of rpoS resulted in higher H-NS promoter occupancy at the tcpA, flrA, and hapA promoters (Fig. 5B). Western blot analysis did not reveal differences in H-NS–FLAG expression between the wild-type and ΔrpoS strains, indicating that the observed increase in promoter occupancy is not due to an increase in H-NS–FLAG expression in the ΔrpoS mutant (Fig. 5C).

Fig 5.

ChIP analysis of H-NS binding to different promoters. (A) Agarose gel electrophoresis of the PCR products obtained after ChIP using anti-FLAG and anti-Xpress (mock reaction) monoclonal antibodies. (B) Quantification of H-NS occupancy at the corresponding promoter in strains C7258HNS-FLAG (WT) and AJB50HNS-FLAG (ΔrpoS) by qPCR. Each value represents the mean for three experiments, and error bars indicate standard deviations (*, significantly different from that of the wild type [P < 0.05]; **, P < 0.01). (C) Detection of H-NS–FLAG protein expression in C7258HNS-FLAG (WT) and AJB50HNS-FLAG (ΔrpoS). In all experiments, V. cholerae strains were grown in TSB medium at 37°C to the stationary phase.

RpoS enhances the expression of IHF.

Gene expression profiling of an ΔrpoS mutant of strain C7258 grown to the stationary phase revealed that the expression of IHF was downregulated in the mutant (unpublished results). Since IHF has been shown to enhance tcpA expression by binding to its promoter at a site that overlaps the H-NS binding site (52), we hypothesized that the expression of RpoS could diminish H-NS promoter occupancy at the tcpA, flrA, and hapA promoters by inducing the expression of IHF. Thus, we used qRT-PCR to determine the expression of ihfA and ihfB genes, encoding the IHF A and B subunits, respectively. As shown in Table 3, ihfA and ihfB were expressed in exponentially growing and late logarithmic cells at low levels but increased significantly in stationary-phase cells. The expression levels of ihfA and ihfB were reduced 8.2- and 4.2-fold in the ΔrpoS mutant, respectively. Lower expression levels of ihfA and ihfB were also observed in the Δhns mutant, as expected from H-NS positively affecting the expression of RpoS (see Fig. 8) (49).

Table 3.

Expression of IHF subunits in V. cholerae ΔrpoS and Δhns mutants^a

Strain	Relative expression of ihfA at:			Relative expression of ihfB at:
	OD600 of 0.5	OD600 of 2.0	Stationary phase	OD600 of 0.5	OD600 of 2.0	Stationary phase
C7258ΔlacZ (WT)	0.75 ± 0.14	1.50 ± 0.28	78.77 ± 9.51	0.36 ± 0.03	0.71 ± 0.12	55.39 ± 5.12
AJB50ΔlacZ (ΔrpoS)	0.85 ± 0.10	1.52 ± 0.18	9.59* ± 1.39	0.48 ± 0.02	0.86 ± 0.09	13.34* ± 0.97
AJB80 (Δhns)	1.48 ± 0.41	2.08 ± 0.61	39.52* ± 2.85	0.49 ± 0.09	0.83 ± 0.11	8.70* ± 1.84
AJB81 (ΔrpoS Δhns)	1.05 ± 0.13	1.92 ± 0.50	6.16* ± 0.97	0.47 ± 0.06	0.87 ± 0.04	4.05* ± 1.84

^aEach value represents the mean for at least three independent cultures ± the standard deviation.

^*, significantly different from that of the wild type [P < 0.01].

Fig 8.

Regulation of RpoS (σ^S) expression by H-NS. C7258ΔlacZ RpoS-FLAG (WT) and AJB80RpoS-FLAG (Δhns::km) strains were grown with agitation in TSB medium at 37°C to an OD₆₀₀ of 2.0 (time zero). Then, samples were taken at 1-h intervals to determine the expression of RpoS-FLAG, as described in Materials and Methods.

Integration host factor positively affects the expression of flrA and rpoN.

Analysis of the flrA and rpoN promoters using Virtual Footprint software (http://www.prodoric.de/vfp/vfp_promoter.php) showed the presence of IHF binding sites with scores higher than those found at the tcpA promoter, which is enhanced by this regulator (52). Thus, we introduced an ihfA deletion in strains C7258 and AJB50 (ΔrpoS) and used qRT-PCR to assess the effect of RpoS and IHF on the expression of flrA and rpoN. As shown in Fig. 6, significantly less flrA and rpoN mRNA was detected in the rpoS and ihfA mutants. For both genes, the ΔrpoS ΔihfA double mutant exhibited the lowest level of expression. To examine the role of IHF in H-NS occupancy at the tcpA, flrA, and rpoN promoters, the allele encoding H-NS–FLAG was also introduced in strains containing ΔihfA and ΔrpoS ΔihfA mutations. As shown in Fig. 7, higher promoter occupancies at the tcpA and flrA promoters were found in the ihfA mutant than in the wild type. The ΔrpoS ΔihfA double mutant exhibited the highest H-NS occupancies for tcpA, flrA, and rpoN, suggesting the occurrence of more than one mechanism involved in this process.

Fig 6.

Role of integration host factor in the expression of rpoN and flrA. C7258 (WT), AJB50 (ΔrpoS), HX120 (ΔihfA), and HX121 (ΔrpoS ΔihfA) strains were grown with agitation in TSB medium at 37°C until the stationary phase, and the relative expression levels of flrA and rpoN were determined by qRT-PCR as described in Materials and Methods. Each value represents the mean for three experiments, and error bars indicate standard deviations (*, significantly different from that of the wild type [P < 0.01]).

Fig 7.

Effect of IHF on H-NS promoter occupancies. Strains C7258HNS-FLAG, HX120HNS-FLAG, and HX121HNS-FLAG were grown to the stationary phase in TSB medium at 37°C, and H-NS promoter occupancies were determined by ChIP and qPCR as described in Materials and Methods. Each value represents the mean for three experiments, and error bars indicate standard deviations (**, P < 0.01).

H-NS can negatively affect its occupancy of target promoters by enhancing RpoS.

The data noted above revealed a regulatory pathway by which the expression of RpoS in the stationary phase could diminish H-NS occupancy of target promoters at which σ^S could initiate transcription (Fig. 5). In a previous study, we showed that H-NS positively affected the expression of rpoS by stabilizing its mRNA (49). To demonstrate that H-NS enhances RpoS protein expression, we compared the production levels of a chromosomally integrated rpoS-FLAG allele from its native transcription and translation signals in WT and Δhns strains. As shown in Fig. 8, the Δhns mutant expressed reduced RpoS-FLAG expression compared to the wild-type strain.

DISCUSSION

The general stress response regulator RpoS and the nucleoid structuring protein H-NS are global regulators known to affect intestinal colonization, virulence gene expression, and mucosal escape in the cholera bacterium (18, 27, 32, 35, 49). Here, we focus on the role of these proteins in the transcription of motility regulators rpoN and flrA as well as hapA, encoding HA/protease. Like other Gram-negative bacteria, V. cholerae hns mutants exhibit a slow-growth phenotype. It has been suggested that the inadequate overexpression of numerous cellular proteins and defective chromosome replication contribute to slow growth of hns mutants (1, 20, 24). In E. coli, RpoS and entire sets of RpoS-dependent genes are repressed by H-NS, and it has been shown that deletion of hns leads to deleterious overexpression of these genes in the exponential phase (3, 6, 19, 20). Our finding that deletion of rpoS partially suppressed the slow-growth phenotype of a V. cholerae hns mutant is consistent with the presence of an important set of genes in the cholera bacterium's genome under dual regulation by RpoS and H-NS. The fact that deletion of rpoS did not fully suppress the Δhns slow-growth phenotype indicates that other cellular processes are compromised in the hns mutant. For instance, we observed that V. cholerae hns mutant cells appear elongated compared to wild-type cells, and their morphology was not suppressed by deletion of rpoS. This result is consistent with data showing that defective chromosome replication in hns mutants is independent of RpoS (1). Microscopic examination revealed that the Δhns and ΔrpoS Δhns mutants are both flagellated and similarly depressed for swimming. Thus, we suggest that the increase in swarm diameter observed in the ΔrpoS Δhns double mutant compared to that of its Δhns precursor (49) is a consequence of partial suppression of the Δhns slow-growth phenotype.

In V. cholerae, RpoS activates hapA and enhances the expression of motility genes belonging to the class II, III, and IV hierarchies (35, 46, 57). We examined the role of RpoS and H-NS in the expression of rpoN and flrA, which control the expression of the class II, III, and IV hierarchy motility genes (42). In this study, we demonstrate that RpoS can enhance motility by positively affecting the expression of rpoN and flrA while H-NS acts as a repressor of these genes. These results suggest that RNAP containing σ^S can contribute to the transcription of flrA and rpoN in the stationary phase. Repression by H-NS was stronger at the flrA and rpoN promoters than at the hapA promoter. The difference between flrA, rpoN, and hapA is that while the first two promoters are transcribed by RNAP containing σ⁷⁰ in exponentially growing cultures, the expression of hapA is tightly RpoS dependent and restricted to the stationary phase. The behavior of these promoters appears to be in agreement with studies suggesting that H-NS exhibits higher selectivity for inhibition of transcription initiated by RNAP containing σ⁷⁰ (19–21, 54). Thus, we propose that participation of RNAP-σ^S in the transcription of flrA and rpoN in the stationary phase could render these promoters more resistant to repression by H-NS. As expected, a similar interplay between RpoS and H-NS was observed in the regulation of the downstream gene flaA, encoding the major V. cholerae flagellin. The observed regulation raises the question of why the hns mutant is less motile in spite of being flagellated and having elevated flaA expression. One possibility is that the altered cell morphology of this mutant could indirectly hamper flagellum rotation and/or bacterial swimming speed. A second explanation, suggested by analogy to E. coli, is that H-NS could participate directly in torque generation by interacting with the switch complex proteins FliG and MotA (12). In this case, the lack of H-NS leads to a paralyzed flagellum.

Gel retardation assays indicated that H-NS can bind the flrA, rpoN, and hapA promoters with an affinity comparable to that of tcpA, a promoter known to be silenced by H-NS (37). H-NS is known to exhibit a broad specificity window for DNA binding that could be affected by the presence of other regulators. Thus, we conducted ChIP to determine H-NS occupancy at the tcpA, flrA, rpoN, and hapA promoters, using the promoter region of VC1922 and a sequence within the rpsM ORF as negative controls. According to our data, the highest H-NS occupancy was found for tcpA, followed by flrA and hapA. Very little H-NS occupancy was found at the rpoN promoter, suggesting that, at the time cells were collected for ChIP, other regulatory factors could prevent H-NS binding to this promoter. Deletion of rpoS significantly enhanced H-NS occupancy at the tcpA, flrA, and hapA promoters. The effect of RpoS on H-NS occupancy suggests a second manner by which the expression of σ^S can partially counteract H-NS repression. Sigma S has less affinity for the core polymerase than σ⁷⁰, with which it needs to compete for binding before it can access promoters (9). In addition, RNAP containing either σ⁷⁰ or σ^S does not compete with H-NS for binding to promoter DNAs (45). Thus, we suggest that RpoS could diminish H-NS occupancy indirectly by inducing the expression of other trans-acting transcriptional regulators. IHF is known to alleviate H-NS silencing of S. enterica hilA (43), E. coli csgD (38), Shigella flexneri vir genes (41), and the bacteriophage Mu early promoter (56). In V. cholerae, IHF positively affects tcpA expression by binding to its promoter at a position that overlaps the H-NS binding site (52). Here, we show that the expression of RpoS in the stationary phase led to a significant increase in IHF expression. Moreover, we found that, similar to tcpA (52), IHF enhanced the expression of flrA and rpoN. The finding that deletion of both ihfA and rpoS diminished the expression of flrA and rpoN in an additive manner is consistent with the occurrence of more than one mechanism affecting the expression of these genes in the double mutant. Furthermore, H-NS occupancies at the tcpA and flrA promoters were significantly enhanced in the ihfA mutant compared to those in the wild-type strain.

It has been suggested that the open initiation complex formed with the promoter in RNAP-σ⁷⁰ differs in architecture from that in RNAP-σ³. The complex formed by RNAP-σ⁷⁰ facilitates lateral oligomerization of H-NS by cooperative recruitment of H-NS molecules, leading to the formation of a repression loop (45). In contrast, the RNAP-σ^S open initiation complex does not promote H-NS oligomerization and exhibits H-NS-resistant transcription initiation (45). Consistent with this mechanism, the elevated promoter occupancies found in the ΔrpoS ΔihfA double mutant could result from the absence of IHF, required to displace H-NS from its binding site, and the formation of only RNAP-σ⁷⁰ open initiation complexes that favor H-NS oligomerization along DNA. In addition, deletion of rpoS could lead to enhanced activity of σ⁷⁰ (19–21), favoring a higher association between H-NS and promoter DNA. Also, our finding that H-NS exerts weaker repression at the hapA promoter is in agreement with the above-mentioned model for H-NS promoter selectivity.

Consistent with our previous finding that H-NS enhances rpoS mRNA stability (49), here we show that H-NS positively affected the expression of RpoS at the protein level. This regulation could generate a negative regulatory loop in which elevated RpoS expression could diminish H-NS occupancy at target promoters that could be transcribed using RNAP-σ^S by activating the expression of IHF or other trans-acting regulatory factors.

Based on our results, we propose a model for the interplay between RpoS and H-NS at V. cholerae promoters that control motility (Fig. 9). Briefly, H-NS binds to the flrA and rpoN promoters to repress transcription. RpoS is expressed in the stationary phase, and RNAP-σ^S contributes to the transcription of flrA and rpoN, rendering these promoters more resistant to H-NS repression (Fig. 9, I). In addition, the expression of RpoS can indirectly attenuate H-NS-mediated repression of flrA and rpoN by inducing the expression of IHF (Fig. 9, II). IHF acts by displacing H-NS from its binding site to enhance transcription of flrA and rpoN, as described for the tcpA promoter (52). Our data provide an explanation for RpoS regulation of motility and mucosal escape (35). Moreover, RpoS was shown to be required for intestinal colonization in an El Tor biotype strain (32). Our results also suggest that RpoS could favor colonization by diminishing H-NS repression of motility and tcpA expression, which are required to establish infection (22, 47).

Fig 9.

Model for RpoS regulation of motility in V. cholerae. The expression of RpoS in the stationary phase enhances rpoN and flrA transcription by two parallel mechanisms. In mechanism I, RNAP-σ^S can initiate transcription at the rpoN and flrA promoters. Transcription initiation by RNAP-σ^S is more resistant to H-NS repression. In mechanism II, RpoS diminishes H-NS occupancy at the rpoN and flrA promoters indirectly by enhancing the expression of IHF, which could compete with H-NS for binding to DNA.