New facet of non-O1/non-O139 Vibrio cholerae hemolysin A: a competitive factor in the ecological niche

New facet of non-O1/non-O139 Vibrio cholerae hemolysin A: a competitive factor in the ecological... Abstract Different serogroups of Vibrio cholerae may inhabit the same ecological niche. However, serogroup O1/O139 strains are rarely isolated from their ecological sources. Quite plausibly, the non-O1/non-O139 vibrios and other bacterial species suppress growth of O1/O139 strains that share the same niche. Our bacterial inhibition assay data indicated that certain non-O1/non-O139 strains used a contact-dependent type VI secretion system (T6SS) to suppress growth of the O1 El Tor, N16961 pandemic strain. Comparative proteomics of the O1 and the suppressive non-O1/non-O139 strains co-cultured in a simulated natural aquatic microcosm showed that SecB and HlyD were upregulated in the latter. The HlyD-related effective factor was subsequently found to be hemolysin A (HlyA). However, not all hlyA-positive non-O1/non-O139 strains mediated growth suppression of the N16961 V. cholerae; only strains harboring intact cluster I HlyA could exert this activity. The key feature of the HlyA is located in the ricin-like lectin domain (β-trefoil) that plays an important role in target cell binding. In conclusion, the results of this study indicated that non-O1/non-O139 V. cholerae suppressed the growth of the O1 pandemic strain by using contact-dependent T6SS as well as by secreting the O1-detrimental hemolysin A during their co-persistence in the aquatic habitat. bacterial competition, hemolysin A, growth suppressive factor, ecological niche, non-O1/non-O139 Vibrio cholerae, O1/O139 Vibrio cholerae INTRODUCTION Vibrio cholerae bacteria are naturally found in aquatic ecosystems such as riverine, estuarine, marine and coastal waters (Colwell and Huq 1994, 2001; Faruque, Albert and Mekalanos 1998; Binsztein et al.2004; Constantin de Magny et al.2008). The bacterial strains are classified into more than 200 serogroups based on the surface O antigens (Shimada et al.1994). Only strains of O1 and O139 serogroups have inherent cholera epidemic/pandemic potential while those of non-O1/non-O139 serogroups cause merely sporadic infection such as gastroenteritis and/or septicemia. It has been observed that the isolation rate of the O1/O139 vibrios from native habitats is low compared with non-O1/non-O139 counterparts, even in cholera epidemic localities (West and Lee 1982; Khan et al.1984; Huq et al.1990; Minami et al.1991; Alam et al.2015). Several attributes have been used to explain this phenomenon. It has been shown that non-O1/non-O139 V. cholerae survive better in the aquatic environment (Colwell 1984). Also, the possibility exists of serogroup conversion between O1/O139 and non-O1/non-O139 V. cholerae (Colwell et al.1995; Montilla et al.1996). However, whether the non-O1/non-O139 V. cholerae could directly inhibit the growth of the O1/O139 epidemic/pandemic strains during their co-persistence in the environment remains unknown. Inhibition of growth and killing of heterologous bacteria by means of contact-dependent and/or contact-independent mechanisms have been used by various bacterial species to compete for niches and to thrive (Ruhe, Low and Hayes 2013). The classic onslaught that V. cholerae uses to kill sensitive strains and to acquire environmentally fit genes is by bacteriophages (Faruque and Mekalanos 2012). Integrating conjugative elements (ICEs), a self-transmissible genetic material found in some V. cholerae genomes, also transfer the genes that facilitate rapid adaptation of the bacteria to environmental changes as well as providing resistance of the host bacterium to antibiotics and heavy metals (Waldor, Tschape and Mekalanos 1996; Boltner et al.2002; Burrus, Marrero and Waldor 2006). Moreover, V. cholerae employ types III and VI secretion systems (T3SS and T6SS, respectively) as the weapons to kill other species such as Pseudomonas aeruginosa, Escherichia coli, and Dictyostelium discoideum (Dziejman et al.2005; Pukatzki et al.2006; Russell et al.2011; Zheng, Ho and Mekalanos 2011). Vibrio cholerae secrete bacteriocins and other toxins into the extracellular milieu to destroy their opponents (Wahba 1965). In this study, the mechanism that the non-O1/non-O139 vibrios used to compete with the O1 counterpart for their ecological habitat was investigated to better understand why the former are predominantly isolated from environmental samples. MATERIALS AND METHODS Bacterial strains, plasmid and culture conditions Four hundred and thirty isolates of non-O1/non-O139 V. cholerae collected during 2006–2009 from four environmental water sources in Thailand comprising Mae Klong River (MK, 73 isolates), Tha Chin River (TC, 195 isolates), Sanamchai Canal (SC, 97 isolates) and Bangkhunsri Canal (BK, 65 isolates) were used as the tested strains in a bacterial inhibition assay. Vibrio cholerae O1 El Tor strain N16961, O1 classical strain O395 and O139 serogroup were used as the indicator strains. The MK2-25 strain was used as a parental wild-type strain for mutant construction and as a tested strain in an in vitro competition assay. Escherichia coli strains DH5α λpir and SM10 λpir were used for cloning and mating, respectively. The pWM91 plasmid was used for mutant construction. Bacteria and plasmids used in this study are listed in Supplementary Table S1. All were obtained from stock cultures of the Department of Microbiology, Faculty of Public Health, Mahidol University. Luria–Bertani (LB) agar and broth were used to culture the bacteria at 37°C under shaking or static conditions. Streptomycin (100 μg mL−1), kanamycin (50 μg mL−1) and ampicillin (100 μg mL−1) were added to the medium when required. Bacterial inhibition assay Four hundred and thirty isolates of non-O1/non-O139 V. cholerae were tested for inhibitory activity against the O1 classical (strain O395) and El Tor (pandemic strain N16961) and O139 V. cholerae using a bacterial inhibition assay. Non-O1/non-O139 V. cholerae from stock cultures were streaked as a short line onto LB agar and incubated overnight at 37°C (45 isolates per plate). Meanwhile, indicator strains (O1 classical and El Tor, and O139 V. cholerae) were individually cultivated with shaking in LB broth at 37°C for 16 h. The overnight cultures were further diluted in 0.85% normal saline solution to a turbidity comparable to that of a 0.5 McFarland turbidity standard (∼108 cells mL−1), then thoroughly swabbed onto LB agar plates. Each of the tested strains was picked from the master plate, streaked in triplicate onto the lawn of the three indicator strains and incubated overnight at 37°C. A circumferential inhibition zone around the streaked line of the tested strain on the lawn of the indicator strain was observed and recorded in centimeters. Plaque formation assay Mid-logarithmic phase cultures of the non-O1/non-O139 V. cholerae that inhibited the indicator strain were treated with mitomycin C at a final concentration 200 ng mL−1. The cultures were incubated at 37°C for 4 h to induce the release of bacteriophage in the culture supernatant. Cell-free culture supernatants (CFSs) of the mitomycin C-treated tested strain cultures were collected by centrifugation, then filtered through a 0.2 μm membrane (Nalgene, USA) and stored at 4°C until used. In the meantime, mid-logarithmic phase cells (∼107 cells) of the indicator strains were harvested by centrifugation, mixed with 100 μL of the mitomycin C-treated CFS and incubated at 37°C for 1 h to allow the infection of bacteriophage. The mixtures were transferred to sterile tubes containing 4 mL of 0.35% soft LB agar, then poured and swirled in a circle onto LB agar plates. The plates were kept on the bench until the soft agar solidified, incubated at 37°C overnight and the plaque formation on the indicator's lawn was observed. Agar-well diffusion assay To investigate bacteriocin production, supernatants were collected by centrifugation from the late stationary phase cultures of the non-O1/non-O139 V. cholerae that inhibited the indicator strain, filtered through a 0.2 μm membrane (Nalgene, USA), and stored at 4°C until used. In the meantime, mid-logarithmic phase cells (∼107 cells) of the indicator strains were harvested by centrifugation, separately mixed with 4 mL of 0.35% soft agar, poured onto LB agar plates, swirled in a circle and then kept on the bench until the soft agar solidified. The solidified soft agar was drilled to give wells of 6 mm diameter. Each well was filled with 50 μL of an individual supernatant. The plate was placed at room temperature for 1 h, then incubated at 37°C overnight, and the clear zone around the well was observed. Detection of T3SS, T6SS and ICEs by PCR Overnight cultures of the non-O1/non-O139 V. cholerae that inhibited the indicator strain were heated to extract DNA and subjected to PCR. The reaction mixture contained 500 ng of DNA template, 1× ThermoPol reaction buffer (20 mM Tris–HCl pH 8.8, 10 mM (NH4)2SO4, 10 mM KCl, 2 mM MgSO4, 0.1% Triton X-100; New England Biolabs, USA), 200 μM of each dNTP, 200 nM of each primer and 1 unit of Taq DNA polymerase (New England Biolabs, USA), and was prepared in a final volume of 20 μL. The thermocycling program was optimized as follows: an initial denaturation at 95°C for 2 min, followed by 35 cycles of denaturation at 95°C for 15 s, annealing at 49–55°C for 30 s (Supplementary Table S2), and extension at 68°C for 45 s. After a final extension at 68°C for 5 min, PCR tubes were cooled to 4°C and the amplicon was electrophoresed and observed under UV light using the gel documentation system (BIS 303 PC, DNR Bio-Imaging Systems, Israel). Primers used to detect the representative genes of T3SS, T6SS and ICEs are listed in Supplementary Table S2. Preparation of laboratory microcosm The environmental microcosm was prepared for in vitro competition assay by collecting freshwater from the Chao Phraya River, Thailand. Five liters of freshwater was collected and sequentially filtered through 0.8 and 0.1 μm pore size membrane filters (Nalgene, USA) to remove debris and bacterial cells in both normal and dormant states, respectively. Physical characteristics of the river water such as pH, temperature and electrical conductivity were determined using a portable digital pH meter (pH Testr 2, Oakton Instruments, USA) and a conductivity/total dissolved solids/°C meter (Cyberscan CON II, Eutech Instruments, USA). This filtered river water was kept in a sterile bottle at ambient temperature as bacterial free-freshwater. In vitro competition assay Mid-logarithmic phase cells of the selected non-O1/non-O139 V. cholerae and N16961 strains were harvested by centrifugation. Cell pellets were washed three times with 1 mL of sterile river water (described above) and individually resuspended in 25 mL of sterile river water (325 μS conductivity, 30.6°C, pH 7.0), then pooled in a 1:1 ratio (non-O1/non-O139:N16961) in a sterile 250 mL Erlenmeyer flask and placed at ambient temperature. The growth profile of each strain was analyzed by the plate count technique using LB agar and LB agar containing 100 μg mL−1 streptomycin (the N16961 strain is streptomycin resistance). This differential count (colony forming units (CFU) mL−1) was used to calculate the competitive index by dividing the proportion of the tested strain/N16961 at the end to the proportion of the tested strain/N16961 at the start. Meanwhile, a mono-cultivation assay was also performed by cultivating each of the tested strains individually in 50 mL sterile river water to observe any defect in microcosm adaptation. All of the experiments were performed in triplicate and mean CFU mL−1 was used for calculation. Growth rate determination To monitor the growth rates of the tested strains in nutrient rich and nutrient deprived environments, late stationary phase culture of each strain was cultivated in 250 mL Erlenmeyer flasks containing 20 mL of high-nutrient (LB broth), average-nutrient (LB:sterile DW = 1:1) or low-nutrient medium (LB:sterile DW = 1:3) and incubated at 37°C with shaking. One milliliter of bacterial culture from each flask was pipetted into a plastic cuvette to measure the optical density at wavelength 600 nm (OD600nm) using a spectrophotometer (UV-160A, Shimadzu, Japan). Gene knockout In-frame deletions of secB, hlyD and hlyA were conducted by means of overlapped extension PCR and secB-based allelic exchange as previously described using pWM91 plasmid (Metcalf et al.1996; Zheng, Tung and Leung 2005). All primers used for constructing the mutant strains are listed in Supplementary Table S2. Effect of residual protein(s) in supernatant on growth pattern The N16961 strain was grown in the CFSs of the wild-type suppressive bacteria and mutants at 37°C with shaking. The growth pattern was observed by measuring the OD600nm using a spectrophotometer (UV-160A, Shimadzu, Japan). Detection and in silico analysis of hlyA Genomic DNAs of bacteria were extracted using the NucleoSpin® Tissue Kit (Macherey-Nagel, Germany) following the manufacturer’s instructions. All preparations were subjected to monoplex PCR to detect the presence of the hlyA gene using hlyA-specific primers (Supplementary Table S2). The PCR mixture (50 μL final volume) for hlyA amplification contained 1 μg of genomic DNA, 1× KAPA HiFi Fidelity buffer (Kapa Biosystems, USA), 300 μM of each dNTP, 250 nM of each primer (pTrcHis2-hlyA-F and pTrcHis2-hlyA-R) and 1 unit of KAPA HiFi DNA polymerase (Kapa Biosystems, USA). The thermocycling program was optimized as follows: an initial denaturation at 95°C for 2 min, followed by 35 cycles of denaturation at 98°C for 20 s, annealing at 65°C for 15 s and extension at 72°C for 30 s. After a final extension at 72°C for 5 min, PCR tubes were cooled to 4°C and the amplicons were electrophoresed and observed under UV light using the gel documentation system (BIS 303 PC, DNR Bio-Imaging Systems, Israel). PCR products were cleaned up using a QIAquick PCR Purification Kit (Qiagen GmbH, Germany). The eluted PCR products were submitted to nucleotide sequencing (SolGent, South Korea). The sequences were submitted to the National Center for Biotechnology Information (NCBI) to obtain accession numbers. The hlyA sequences of the non-O1/non-O139 V. cholerae strains MK2-25, SC1-09, TC1-21, MK4-26, SC1-34, SC2-10 and TC2-57 were deposited in GenBank under accession numbers KU888666, KU888667, KU888668, KY369196, KY369197, KY369198 and KY369199, respectively. These sequences were then compared with those of other strains available in the NCBI GenBank database and a phylogenetic tree was constructed using Geneious version 7.1 (Kearse et al.2012). Their respective amino acid sequences were also compared. The PDB file and a ribbon structure of proteins were created using the SWISS-MODEL server (Arnold et al.2006) and CCP4mg software (McNicholas et al.2011), respectively. The ligand binding site within the β-trefoil domain was predicted using the I-TASSER server (Zhang 2008; Roy, Kucukural and Zhang 2010; Yang et al.2015; Yang and Zhang 2015). RESULTS AND DISCUSSION Non-O1/non-O139 V. cholerae inhibited growth of the O1 and O139 strains Because O1 and O139 V. cholerae are rarely isolated from the aquatic environments (West and Lee 1982; Khan et al.1984; Huq et al.1990; Minami et al.1991; Alam et al.2015), it was hypothesized that their growth might be restrained by the non-O1/non-O139 strains during their co-persistence. To validate this speculation, 430 isolates of non-O1/non-O139 V. cholerae obtained from two rivers and two canals in central Thailand during 2006–2009 were tested for their inhibitory activity on indicator O1 classical (O395) and El Tor (N16961) and O139 strains by bacterial inhibition assay. From the four sample collection sites, no O1/O139 was isolated, which conformed to the data reported previously from different geographical areas that O1 strains were rarely isolated from natural ecosystems (West and Lee 1982; Khan et al.1984; Huq et al.1990; Minami et al.1991; Alam et al.2015). From the growth inhibition assay, 22 non-O1/non-O139 isolates (5.1%) inhibited the growth of the indicators as shown by clear circumferences (inhibition halos) around the effectors’ streaked lines on the indicator lawns grown on agar. They comprised nine isolates, each from the TC and SC, four isolates from MK, and none from BK (Supplementary Table S3). Examples of the halos of growth inhibition are shown in Supplementary Fig. S1. The widths of the halos varied from 0.3 to 1.1 cm (Supplementary Table S3). Among the three indicators, O1 El Tor, N16961 (an isolate of the seventh cholera pandemic) was the most vulnerable to environmental non-O1/non-O139 inhibitory activity (Supplementary Table S3). A majority of the effective non-O1/non-O139 isolates (19/22 isolates; nos TC1-21, MK4-26, TC1-57, TC2-07, TC2-08, TC2-57, TC3-08, MK1-11, MK2-25, MK4-13, SC1-09, SC1-34, SC1-35, SC1-36, SC1-37, SC1-38, SC1-39, SC2-10 and SC2-38) inhibited O1 El Tor, strain N16961; three isolates (nos TC2-37, TC2-52 and TC2-59) inhibited O1 classical strain O395; and two (nos TC1-21 and MK4-26) were effective against all indicator strains. The inhibitory property of most of the tested strains except the TC1-21 and MK4-26 was against only one single indicator, indicating that intra-species inhibition is highly strain specific. Growth inhibition among the three indicators themselves was not observed (Supplementary Table S3). Usually the toxin-producing bacteria protect themselves from their own detrimental substance(s) using diverse mechanisms including immunity (e.g. producing immunity protein), resistance (e.g. modification of toxin receptor), tolerance (e.g. changing the translocation system) and production of anti-toxin proteins (Lazdunski 1988; Alonso, Vilchez and Rodriguez Lemoine 2000). For V. cholerae, non-O1/non-O139 V. cholerae strain V52 inhibited the growth of E. coli and Dictyostelium discoideum, which share the ecological niche, using T6SS as a competitive weapon (Pukatzki et al.2006; Zheng, Ho and Mekalanos 2011). Moreover, it protected itself by producing immunity proteins encoded by genes of the T6SS cassette (MacIntyre et al.2010; Dong et al.2013). T6SS is a contact-dependent inhibitory factor of non-O1/non-O139 V. cholerae Because the non-O1/non-O139 V. cholerae inhibited the O1 and O139 strains by a contact-dependent mode as observed in the previous experiment, several possible factors that might be used by the inhibitors for niche competition were investigated including bacteriophage (by plaque formation assay), bacteriocin production (by agar-well diffusion assay) and the presence of the T3SS, T6SS and SXT family of ICEs (by PCR). The results showed that none of the inhibitory non-O1/non-O139 strains released bacteriophage or produced bacteriocin to inhibit the growth of the indicators. Likewise, four representative genes of T3SS, i.e. vcsN2, vcsV2, vcsC2 and vspD, were undetected among the inhibitors. However, a gene coding for protein(s) of T6SS, known to be a weapon related to contact-dependent inhibition, was detected in all of the inhibitory isolates (Supplementary Table S4). Ten isolates (nos TC1-21, TC2-57, TC2-37, TC2-52, MK4-26, SC1-34, SC1-35, SC1-36, SC1-38 and TC2-07) were positive for hcp-2 and vgrG-2, which code for the T6SS apparatus. Four isolates (SC1-09, SC2-10, SC2-38 and TC2-08) were positive for hcp-2 only and eight isolates (MK1-11, TC2-59, MK2-25, MK4-13, SC1-37, SC1-39, TC1-57 and TC3-08) were positive for vgrG-2 only. Two isolates (nos TC1-21 and TC2-57) were positive for tseL and four isolates (TC2-37, TC2-52, MK1-11 and TC2-59) were positive for vgrG-3; both genes encode T6SS effectors. No isolate was positive for vasX, which encodes a dispensable factor for assembling the T6SS translocon complex. The SXT-R391 family of ICEs, comprising important factors for survival, was found previously in genomes of some V. cholerae isolates (Waldor, Tschape and Mekalanos 1996; Boltner et al.2002; Burrus, Marrero and Waldor 2006). The element was detected in only 3/22 effective non-O1/non-O139 strains (nos TC2-52, MK1-11 and MK4-26), suggesting that ICEs might not be the key inhibitory property of the effective non-O1/non-O1 vibrios. Non-O1/non-O139 V. cholerae suppress growth of O1 strain in an aquatic microcosm To study the non-O1/non-O139 V. cholerae-mediated growth suppression of the O1 pandemic strain in the natural aquatic niche, the bacterial competition assay was performed in a simulated environmental microcosm. Vibrio cholerae O1 El Tor, N16961 and four non-O1/non-O139 effectors, i.e. MK2-25, SC1-09, TC1-21 and MK1-11, which exerted varying degrees of N16961 growth suppression (widths of the inhibition halos were 0.9, 0.7, 0.5 and 0.3 cm, respectively), were selected for co-cultivation at ambient temperature. The daily growth profiles of the bacteria were monitored and compared. The results showed that MK2-25 and SC1-09 overtook N16961 growth in the microcosm (Fig. 1A. and B). The initial amounts of the effective MK2-25 and N16961 indicators were 7.5 × 107 and 5.8 × 107 CFU mL−1, respectively. After day 1 of co-cultivation, cultivable N16961 declined to 1.0 × 106 CFU mL−1 (58 fold), whereas MK2-25 increased to 1.1 × 108 CFU mL−1. The cultivable MK2-25 cell numbers ranged from ∼100 to 10 000 fold higher than that of N16961 throughout the co-cultivation period. At day 49, growth of N16961 was undetected while the live MK2-25 was 2.8 × 102 CFU mL−1; the live MK2-25 could be detected through day 163 (Fig. 1A). SC1-09 also suppressed growth of N16961 during the co-cultivation (Fig. 1B). The N16961 growth suppression was not observed during co-cultivation with the TC1-21 and MK1-11 strains. The numbers of cultivable TC1-21 cells were relatively similar to N16961 throughout the co-cultivation course (Fig. 1C), whereas the MK1-11 cells were fewer than those of N16961 (Fig. 1D). Figure 1. View largeDownload slide Co-cultivations of the pandemic O1 El Tor, N16961 strain with the non-O1/non-O139 strains MK2-25 (A), SC1-09 (B), TC1-21 (C) and MK1-11 (D) in a simulated aquatic microcosm. Mean ± standard deviation (SD) of colony forming units per milliliter (CFU mL−1) of individual strains are shown. Data are representative of two reproducible independent experiments Figure 1. View largeDownload slide Co-cultivations of the pandemic O1 El Tor, N16961 strain with the non-O1/non-O139 strains MK2-25 (A), SC1-09 (B), TC1-21 (C) and MK1-11 (D) in a simulated aquatic microcosm. Mean ± standard deviation (SD) of colony forming units per milliliter (CFU mL−1) of individual strains are shown. Data are representative of two reproducible independent experiments To confirm that the decline of living N16961 cells during co-cultivation with the effectors was due to suppressive factor(s) released from the latter and not from the defective adaptation of the former to the microcosm, the numbers of live N16961, MK2-25, SC1-09, TC1-21 and MK1-11 in mono-cultivations were monitored. No difference was observed in the cultivable cell numbers of N16961, MK2-25 and SC1-09 during mono-cultivations (Fig. 1A and B), implying that the N16961 growth retardation in the co-cultivations was not due to a bacterial defect in microcosm adaptation. The cultivable cell numbers of MK2-25 and SC1-09 during co-cultivation were slightly higher than those during the mono-cultivations (Fig. 1A and B). Quite possibly, they (the predators in the competitive microcosms) used by-products derived from the dead N16961 cells (the prey) to increase. This phenomenon has been described previously (Helling, Vargas and Adams 1987). The numbers of cultivable N16961 and MK1-11 cells did not differ during the mono- and co-cultivations (Fig. 1D) suggesting that the observed growth retardation of MK1-11 was not a result of a suppression factor released from N16961, but rather due to a defect in adapting to the new environment. Growth patterns of V. cholerae in nutrient-rich and nutrient-deprived environments It has been reported that the thriving of certain V. cholerae strains in an aquatic ecosystem varies depending on the ability of the bacteria to adapt to various environmental stresses, such as nutrient deprivation (Lutz et al.2013). In this study, the growth profiles of MK2-25, SC1-09, TC1-21 and N16961 were investigated under high and low nutritional status. The growth rate of N16961 did not differ from those of MK2-25, SC1-09 and TC1-21 in all concentrations of the LB broth used for cultivation (undiluted and diluted 1:2 and 1:4; Fig. 2A–C). This finding supports the above notion that growth retardation of N16961 during co-cultivation was the effect of suppressive factor(s) of the non-O1/non-O139 strain. Figure 2. View largeDownload slide Growth patterns of the mono-cultivated MK2-25, SC1-09, TC1-21 and N16961 strains in LB broth: (A) undiluted, (B) 1:2 diluted and (C) 1:4 diluted. The results are the means of two reproducible independent experiments. Figure 2. View largeDownload slide Growth patterns of the mono-cultivated MK2-25, SC1-09, TC1-21 and N16961 strains in LB broth: (A) undiluted, (B) 1:2 diluted and (C) 1:4 diluted. The results are the means of two reproducible independent experiments. SecB and HlyD endowed the non-O1/non-O139 V. cholerae with a growth suppressive property Under the simulated ecologic system, the ability of non-O1/non-O139 strains to suppress N16961 did not correlate with the presence of T6SS-specific genes (Fig. 1 and Supplementary Table S4). Strains SC1-09, SC2-38, TC2-08, TC2-59, MK2-25, MK4-13, TC1-57 and TC3-08 lack the essential T6SS component, yet they could suppress growth of N16961. Thus, T6SS should not be considered the sole factor used by the effective non-O1/non-O139 stains for niche competition with the O1 pandemic strain. Comparative proteomics data revealed that the MK2-25 cells produced SecB and HlyD when co-cultured with N16961. SecB (a protein-transporter protein) is a key component of the Sec system that plays a cooperative role with T5SS to deliver protein effectors and toxins to targets (Driessen, Fekkes and van der Wolk 1998; Benz and Meinhart 2014). HlyD (a membrane fusion protein) is a key component of T1SS that exports a hemolysin toxin (Lee et al.2012). Based on their established functions, SecB and HlyD proteins were likely candidates responsible for O1 suppressive activity of the non-O1/non-O139 effectors. To validate this speculation, secB and hlyD deletion mutants, i.e. MK2-25ΔsecB and MK2-25ΔhlyD, were constructed. MK2-25ΔsecB and MK2-25ΔhlyD were found to lose their parental suppressive properties on N16961 as determined by in vitro competitive growth assay. The numbers of the N16961 strain were 100- to 10 000-fold lower than the MK2-25 wild-type, but only 10- to 100-fold lower than MK2-25ΔsecB and MK2-25ΔhlyD (P < 0.05; Fig. 3A–C). Moreover, the values of the daily competitive index of MK2-25ΔsecB (Fig. 4B) and MK2-25ΔhlyD (Fig. 4C) decreased significantly when compared with the wild-type (P < 0.05) (Fig. 4A). On day 1 of the co-cultivations, the competitive index value of MK2-25 wild-type was 85.95, while those of MK2-25ΔsecB and MK2-25ΔhlyD were 49.53 and 35.17, respectively (P < 0.05; Supplementary Fig. S2). The results indicated that the non-O1/non-O139-mediated O1 growth suppression began within the first 24 h of the co-cultures. This finding is similar to the results of a related study (Pradhan, Mallick and Chowdhury 2013) showing that the N16961 strain was able to outcompete the O1 classical O395 strain within 24 h of the co-cultivation. Figure 3. View largeDownload slide Growth patterns (mean ± SD in CFU mL−1) of N16961 indicator and suppressors of wild-type MK2-25 (A), MK2-25ΔsecB (B), MK2-25ΔhlyD (C) and MK2-25ΔhlyA (D) in individual mono- and co-cultures. Data are representative of two reproducible independent experiments. Figure 3. View largeDownload slide Growth patterns (mean ± SD in CFU mL−1) of N16961 indicator and suppressors of wild-type MK2-25 (A), MK2-25ΔsecB (B), MK2-25ΔhlyD (C) and MK2-25ΔhlyA (D) in individual mono- and co-cultures. Data are representative of two reproducible independent experiments. Figure 4. View largeDownload slide Competitive indices (CIs) of N16961 strain and wild-type MK2-25 (A), MK2-25ΔsecB (B), MK2-25ΔhlyD (C) and MK2-25ΔhlyA (D) during 7 days of co-cultivation. The competitive index of suppressor was calculated by dividing the mean proportion of the suppressor/N16961 at the indicated time (day 0, 1, 2, ... 7) by the mean proportion of the suppressor/N16961 at the start (day 0), while the competitive index of N16961 was obtained by dividing the mean proportion of the N16961/suppressor at the indicated time (day 0, 1, 2, ... 7) by the mean proportion of the N16961/suppressor at the start (day 0). Data are representative of two reproducible independent experiments. Significant difference between CIs of wild-type MK2-25 and each mutant were assessed using the Mann–Whitney U-test. Figure 4. View largeDownload slide Competitive indices (CIs) of N16961 strain and wild-type MK2-25 (A), MK2-25ΔsecB (B), MK2-25ΔhlyD (C) and MK2-25ΔhlyA (D) during 7 days of co-cultivation. The competitive index of suppressor was calculated by dividing the mean proportion of the suppressor/N16961 at the indicated time (day 0, 1, 2, ... 7) by the mean proportion of the suppressor/N16961 at the start (day 0), while the competitive index of N16961 was obtained by dividing the mean proportion of the N16961/suppressor at the indicated time (day 0, 1, 2, ... 7) by the mean proportion of the N16961/suppressor at the start (day 0). Data are representative of two reproducible independent experiments. Significant difference between CIs of wild-type MK2-25 and each mutant were assessed using the Mann–Whitney U-test. The secreted inhibitory factor of the non-O1/non-O139 effectors is HlyD dependent The SecB and HlyD themselves are not protein effectors; they are components of the protein secretion systems responsible for toxin secretion (Driessen, Fekkes and van der Wolk 1998; Lee et al.2012). Therefore, the protein(s) or toxin(s) translocated via these portals was of interest. Experiments were carried out to test whether the candidate toxin(s) is SecB or HlyD dependent by comparing the growth profiles of the N16961 strain in the cell-free CFSs of wild-type and secB- and hlyD-deletion mutants. It was found that the growth of N16961 (OD600nm) in the MK2-25ΔhlyD CFS was higher than in the wild-type CFS (P < 0.05) (Fig. 5). These results indicated that HlyD of the non-O1/non-O139 V. cholerae was principally involved in the transport of the N16961 growth suppressive factor(s). Figure 5. View largeDownload slide Growth patterns of N16961 during cultivation in cell-free culture supernatants (CFSs) of wild-type MK2-25, MK2-25ΔsecB, MK2-25ΔhlyD or MK2-25ΔhlyA mutants. Mean OD600nm of two independent experiments is shown. Statistical differences of OD600nm values between N16961 cultivated in wild-type CFS and in each of the mutant CFSs (ΔsecB, ΔhlyD, or ΔhlyA) at each time point were assessed by Student’s t-test. *P < 0.05 and **P < 0.01. Figure 5. View largeDownload slide Growth patterns of N16961 during cultivation in cell-free culture supernatants (CFSs) of wild-type MK2-25, MK2-25ΔsecB, MK2-25ΔhlyD or MK2-25ΔhlyA mutants. Mean OD600nm of two independent experiments is shown. Statistical differences of OD600nm values between N16961 cultivated in wild-type CFS and in each of the mutant CFSs (ΔsecB, ΔhlyD, or ΔhlyA) at each time point were assessed by Student’s t-test. *P < 0.05 and **P < 0.01. HlyA is the non-O1/non-O139 weapon for growth competition HlyD is a key component of T1SS. It forms a translocation apparatus with the inner membrane protein HlyB and the outer membrane channel protein TolC by acting as an adaptor protein that connects HlyB to TolC to provide a continuous transmembrane conduit for toxin export (Lee et al.2012). A well-known toxin secreted by this system is hemolysin A (HlyA; synonym V. cholerae cytolysin or VCC), which is a pore-forming cytotoxic and cytolytic toxin toward a wide spectrum of eukaryotic cells (Lee et al.2012). The effective MK2-25 with in-frame, non-polar deletion mutation of hlyA was tested for N16961 suppression. Like the ΔhlyD mutant, MK2-25ΔhlyA had reduced ability to suppress growth of N16961 compared with the wild-type (Fig. 3A and D; P < 0.05). Also, the competitive index value of MK2-25ΔhlyA reduced significantly compared with the wild-type (P < 0.05; Fig. 4A and D). Thus, HlyA produced by the non-O1/non-O139 V. cholerae was likely a growth suppressive factor during the co-cultivation with the N16961. For verification, growth of the N16961 strain in the CFS of wild-type and ΔhlyA mutant was monitored. As shown in Fig. 5, the OD600nm of N16961 grown in the ΔhlyA mutant CFS was higher than in the wild-type CFS from 3 to 8 h of culturing (P < 0.05). Thus, HlyA is the growth suppressive factor of the non-O1/non-O139 V. cholerae towards the O1 N16961. The growth rate of N16961 in the culture supernatant of the ΔhlyA mutant was higher than in the culture supernatants of wild-type, ΔsecB and ΔhlyD mutants (Fig. 5) indicating that HlyA is the secreted growth suppressive factor of the non-O1/non-O139 V. cholerae against the N16961 pandemic strain. The data demonstrated that hemolysin A of the non-O1/non-O139 V. cholerae strains not only plays a known critical role in cholera pathogenesis (Yamamoto et al.1984), but also exhibits a novel activity in competing for the ecological niche and thriving in ecosystems. Nucleotide sequences of effective and non-effective hemolysin A are different Vibrio cholerae HlyA belongs to a large class of pore-forming toxins encoded by a 2223 bp open reading frame that encodes an 82 kDa preHlyA protein (Rader and Murphy 1988). The toxin is secreted in water-soluble monomeric form as a proHlyA of 79 kDa. Each molecule of the proHlyA consists of a prodomain (the N-terminal 132 residues of 15 kDa, which is important for proper folding of the molecule but must be cleaved off to form mature toxins of 65 kDa with hemolytic activity), a core cytolytic domain containing a pre-stem (pore-forming loop), and two C-terminal domains with ricin/jacalin-like lectin domains called β-trefoil and β-prism for binding to the carbohydrate receptor on the target cell membrane (Yamamoto et al.1990; Nagamune et al.1996; Saha and Banerjee 1997). Once the toxin monomers bind to the receptor, self-oligomerization into a stable circular β-barrel amphipathic heptamer occurs followed by insertion of their stem loops into the membrane lipid bilayer to form a transmembrane pore. Most non-O1/non-O139 V. cholerae isolates carry hlyA genes (Yamamoto et al.1984). Nevertheless, only some strains could suppress growth of N16961 in the aquatic microcosm. To have some insight into the functional differences, HlyA-coding genes of the suppressive and non-suppressive non-O1/non-O139 strains were PCR amplified, sequenced and compared. The hlyA sequences of the suppressive MK2-25 and SC1-09 and the non-suppressive TC1-21 and MK4-26 strains were different. Based on the GenBank database, V. cholerae hlyA sequences can be classified in three clusters. The hlyA of non-O1/non-O139 suppressor strains belong to cluster I whereas those of the O1/O139 and non-O1/non-O139 non-suppressor strains belong to clusters II and III, respectively (Fig. 6). Analysis of the 35 V. cholerae hlyA sequences revealed 32 polymorphic loci (Supplementary data A). The hlyA sequences of MK2-25 and SC1-09 presented 99.0% identity and both were 98.5 to 99.2% identical to other hlyA sequences of the cluster I (Fig. 6). In contrast, the hlyA sequence of the TC1-21 strain showed similarity (98.3 to 98.8%) to O1 and O139 hlyA of cluster II (Fig. 6), which should explain why the TC1-21 could not suppress growth of the O1/O139 indicators because they should have similar (if not identical) immunological factor(s). Figure 6. View largeDownload slide Phylogenetic tree of partial hlyA nucleotide sequences of the non-O1/non-O139 V. cholerae strains MK2-25, SC1-09, TC1-21, MK4-26, SC1-34, SC2-10 and TC2-57 compared with other V. cholerae hlyA sequences from the GenBank database. The hlyA genes of non-O1/non-O139 suppressor strains belong to cluster I whereas those of O1/O139 and non-O1/non-O139 non-suppressor strains are in clusters II and III, respectively. The tree was generated using the UPGMA analysis in Geneious version 7.1. The scale bar indicates nucleotide substitutions per site. Figure 6. View largeDownload slide Phylogenetic tree of partial hlyA nucleotide sequences of the non-O1/non-O139 V. cholerae strains MK2-25, SC1-09, TC1-21, MK4-26, SC1-34, SC2-10 and TC2-57 compared with other V. cholerae hlyA sequences from the GenBank database. The hlyA genes of non-O1/non-O139 suppressor strains belong to cluster I whereas those of O1/O139 and non-O1/non-O139 non-suppressor strains are in clusters II and III, respectively. The tree was generated using the UPGMA analysis in Geneious version 7.1. The scale bar indicates nucleotide substitutions per site. Among the 32 single nucleotide polymorphisms of hlyA sequences, 26 loci were translated into the same amino acids. Interestingly, deduced amino acids from five polymorphic loci differed between the O1/O139 and non-O1/non-O139 clusters. One exceptionally important locus was at the nucleotide positions 1644–1646 (based on N16961 hlyA), which forms the ACC codon of threonine (Thr), in the hlyA of the suppressive MK2-25 and SC1-09 strains. At this locus, the strain TC1-21, which lacks the property of suppressing N16961, has the CCC codon for proline (Pro) (Supplementary data B). The codon at nt 1644–1646 (residue 548) is located in the β-trefoil domain important for receptor binding of monomeric HlyA. Changing the hydrophilic Thr to the hydrophobic Pro, a secondary structure breaker, interferes with the ligand binding site within the β-trefoil domain due to steric hindrance of the bulky prolyl ring (Piela, Nemethy and Scheraga 1987) as shown in Supplementary Fig. S3A–D. The consequence should be an inability of the toxin to bind to its receptor and oligomerize, which in effect is a failure to form a pore on the target membrane. The hlyA sequence of MK4-26 differed from the others and should be placed in cluster III (Fig. 6). It has one base deletion at position 942, leading to a frame shift and consequently formation of a TGA stop codon at positions 950–952, which leads to production of truncated HlyA (Supplementary data A). This should account for the inability of the MK4-26 strain to suppress the N16961 strain's growth upon co-cultivation. Rader and Murphy (1988) similarly reported that the hlyA sequence of V. cholerae O1 classical, 569B strain had an 11-bp deletion that interrupted the open reading frame, resulting in the production of truncated non-hemolytic HlyA (a 244-amino acid precursor form). The remaining 18 strains of non-O1/non-O139 V. cholerae that produced inhibition halos on the N16961 lawns were also tested for growth-suppressive property in the aquatic microcosm. Eight strains, i.e. MK4-13, SC1-38, SC2-38, TC1-57, TC2-07, TC2-08, TC2-59 and TC3-08, could suppress growth of N16961, while the other 10 isolates (MK1-11, SC1-34, SC1-35, SC1-36, SC1-37, SC1-39, SC2-10, TC2-37, TC2-52 and TC2-57) were refractory (Supplementary Table S4). It was found that hlyA sequences of the ineffective strains have a frame shift that should produce truncated HlyA, hence abolishing the toxin activity. The hlyA sequences of strains SC1-34, SC2-10 and TC2-57 are shown as representatives in Supplementary data A. In conclusion, the present study demonstrated that non-O1/non-O139 V. cholerae used T6SS to inhibit the N16961 strain in a contact-dependent manner and secreted HlyA to suppress the pandemic strain during co-persistence in the aquatic microcosms. Additional factors that may be involved in the intraspecies competition await further investigation. The proposed mechanism mediated by HlyA is described below. In the contact-independent mode, non-O1/non-O139 V. cholerae secretes HlyA after being activated by at least three regulators including the cholera toxin transcriptional activator (ToxR), hemolysin gene transcriptional activator (HlyU) and ferric uptake regulator (Fur) (Stoebner and Payne 1988; Williams and Manning 1991; DiRita 1994). HlyU and Fur induce HlyA, HlyB and HlyC expression (Stoebner and Payne 1988; Williams and Manning 1991; Ogierman et al.1997). ToxR induces only HlyB, which not only plays a critical role in protein export, but also acts as a chemotactic receptor in responding to nutrient concentration (Jeffery and Koshland 1993). HlyB regulates the production of the transmembrane pore or transporter that enables hemolysin secretion (Jeffery & Koshland 1993). After production, HlyA is translocated from the cytoplasm to the extracellular milieu using the conduit formed by HlyB, HlyD and TolC, which constitute the key components of T1SS (Lee et al.2012). In addition, HlyC, a lipase, is secreted by T2SS (Rosenau and Jaeger 2000) and hydrolyzes the N16961 outer membrane to facilitate HlyA penetration. The HlyA monomer interacts with the N16961 cell surface receptor using the specific ricin-like lectin domain (β-trefoil). After releasing the pro-domain, mature HlyA monomer self-assembles to a heptameric amphipathic β-barrel by circular oligomerization. Then the pre-stem loop within the core cytolysin domain of each protomer is inserted into the N16961 cell membrane and forms a pore leading to the target cell lysis (Olson and Gouaux 2005). In the contact-dependent mode of growth inhibition, the hemolysin-coregulated protein (Hcp) is produced under the regulation of HlyU similar to HlyA (Williams et al.1996). This toxic protein is injected into the N16961 cytoplasm directly by T6SS and causes the target’s death (Pukatzki et al.2006) (Fig. 7). The illustrated mechanisms allow the non-O1/non-O139 V. cholerae to suppress growth of O1 V. cholerae and hinder the recovery of the vegetative O1 when they are co-persistent in their natural aquatic habitats. Figure 7. View largeDownload slide Proposed mechanisms of contact-dependent and contact-independent growth suppression of pandemic O1, N16961 strain mediated by the non-O1/non-O139 strains. Solid lines indicate relationship based on existing data. ToxR and Fur are under environmental and iron regulation, respectively. Green lines indicate co-expression. Blue arrows indicate activation of protein expression. Dotted line indicates the predicted environmental stimuli. IM, inner membrane; OM, outer membrane; PP, periplasm. Figure 7. View largeDownload slide Proposed mechanisms of contact-dependent and contact-independent growth suppression of pandemic O1, N16961 strain mediated by the non-O1/non-O139 strains. Solid lines indicate relationship based on existing data. ToxR and Fur are under environmental and iron regulation, respectively. Green lines indicate co-expression. Blue arrows indicate activation of protein expression. Dotted line indicates the predicted environmental stimuli. IM, inner membrane; OM, outer membrane; PP, periplasm. SUPPLEMENTARY DATA Supplementary data are available at FEMSEC online. Acknowledgements This work was supported by the National Research University project of the Office of Commission on Higher Education, Ministry of Education, Thailand, through the Center of Biopharmaceutical Development and Innovative Therapy, Mahidol University. Author contributions: PD conceived the project, designed experiments, supervised PR and drafted the manuscript; PR did most of the experiments and prepared the figures; OR performed comparative proteomics; KS and WC helped PD. Conflict of Interest. None declared. REFERENCES Alam MT, Weppelmann TA, Longini Iet al.   Increased isolation frequency of toxigenic Vibrio cholerae O1 from environmental monitoring sites in Haiti. PLoS One  2015; 10: e0124098. Google Scholar CrossRef Search ADS PubMed  Alonso G, Vilchez G, Rodriguez Lemoine V. How bacteria protect themselves against channel-forming colicins. Int Microbiol  2000; 3: 81– 8. Google Scholar PubMed  Arnold K, Bordoli L, Kopp Jet al.   The SWISS-MODEL workspace: a web-based environment for protein structure homology modelling. Bioinformatics  2006; 22: 195– 201. Google Scholar CrossRef Search ADS PubMed  Benz J, Meinhart A. Antibacterial effector/immunity systems: it's just the tip of the iceberg. Curr Opin Microbiol  2014; 17: 1– 10. Google Scholar CrossRef Search ADS PubMed  Binsztein N, Costagliola MC, Pichel Met al.   Viable but nonculturable Vibrio cholerae O1 in the aquatic environment of Argentina. Appl Environ Microbiol  2004; 70: 7481– 6. Google Scholar CrossRef Search ADS PubMed  Boltner D, MacMahon C, Pembroke JTet al.   R391: a conjugative integrating mosaic comprised of phage, plasmid, and transposon elements. J Bacteriol  2002; 184: 5158– 69. Google Scholar CrossRef Search ADS PubMed  Burrus V, Marrero J, Waldor MK. The current ICE age: biology and evolution of SXT-related integrating conjugative elements. Plasmid  2006; 55: 173– 83. Google Scholar CrossRef Search ADS PubMed  Colwell RR Vibrios in the Environment . New York: Wiley, 1984. Colwell RR, Huq A. Viable but nonculturable V. cholerae. In: Wachsmuth K, Blake PA, Olsvik O (eds). Vibrio cholerae and Cholera: Molecular to Global Perspectives . Washington, DC: ASM Press, 1994, 116– 33. Colwell RR, Huq A. Marine ecosystems and cholera. Hydrobiologia  2001; 460: 141– 5. Google Scholar CrossRef Search ADS   Colwell RR, Huq A, Chowdhury MAet al.   Serogroup conversion of Vibrio cholerae. Can J Microbiol  1995; 41: 946– 50. Google Scholar CrossRef Search ADS PubMed  Constantin de Magny G, Murtugudde R, Sapiano MRet al.   Environmental signatures associated with cholera epidemics. Proc Natl Acad Sci U S A  2008; 105: 17676– 81. Google Scholar CrossRef Search ADS PubMed  DiRita VJ. Multiple regulatory systems in Vibrio cholerae pathogenesis. Trends Microbiol  1994; 2: 37– 8. Google Scholar CrossRef Search ADS PubMed  Dong TG, Ho BT, Yoder-Himes DRet al.   Identification of T6SS-dependent effector and immunity proteins by Tn-seq in Vibrio cholerae. Proc Natl Acad Sci U S A  2013; 110: 2623– 8. Google Scholar CrossRef Search ADS PubMed  Driessen AJ, Fekkes P, van der Wolk JP. The Sec system. Curr Opin Microbiol  1998; 1: 216– 22. Google Scholar CrossRef Search ADS PubMed  Dziejman M, Serruto D, Tam VCet al.   Genomic characterization of non-O1, non-O139 Vibrio cholerae reveals genes for a type III secretion system. Proc Natl Acad Sci U S A  2005; 102: 3465– 70. Google Scholar CrossRef Search ADS PubMed  Faruque SM, Albert MJ, Mekalanos JJ. Epidemiology, genetics, and ecology of toxigenic Vibrio cholerae. Microbiol Mol Biol Rev  1998; 62: 1301– 14. Google Scholar PubMed  Faruque SM, Mekalanos JJ. Phage-bacterial interactions in the evolution of toxigenic Vibrio cholerae. Virulence  2012; 3: 556– 65. Google Scholar CrossRef Search ADS PubMed  Helling RB, Vargas CN, Adams J. Evolution of Escherichia coli during growth in a constant environment. Genetics  1987; 116: 349– 58. Google Scholar PubMed  Huq A, Colwell RR, Rahman Ret al.   Detection of Vibrio cholerae O1 in the aquatic environment by fluorescent-monoclonal antibody and culture methods. Appl Environ Microbiol  1990; 56: 2370– 3. Google Scholar PubMed  Jeffery CJ, Koshland DEJr. Vibrio cholerae hlyB is a member of the chemotaxis receptor gene family. Protein Sci  1993; 2: 1532– 5. Google Scholar CrossRef Search ADS PubMed  Kearse M, Moir R, Wilson Aet al.   Geneious Basic: an integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics  2012; 28: 1647– 49. Google Scholar CrossRef Search ADS PubMed  Khan MU, Shahidullah MD, Haque MSet al.   Presence of vibrios in surface water and their relation with cholera in a community. Trop Geogr Med  1984; 36: 335– 40. Google Scholar PubMed  Lazdunski CJ. Pore-forming colicins: synthesis, extracellular release, mode of action, immunity. Biochimie  1988; 70: 1291– 6. Google Scholar CrossRef Search ADS PubMed  Lee M, Jun SY, Yoon BYet al.   Membrane fusion proteins of type I secretion system and tripartite efflux pumps share a binding motif for TolC in gram-negative bacteria. PLoS One  2012; 7: e40460. Google Scholar CrossRef Search ADS PubMed  Lutz C, Erken M, Noorian Pet al.   Environmental reservoirs and mechanisms of persistence of Vibrio cholerae. Front Microbiol  2013; 4: 375. Google Scholar CrossRef Search ADS PubMed  MacIntyre DL, Miyata ST, Kitaoka Met al.   The Vibrio cholerae type VI secretion system displays antimicrobial properties. Proc Natl Acad Sci U S A  2010; 107: 19520– 4. Google Scholar CrossRef Search ADS PubMed  McNicholas S, Potterton E, Wilson KSet al.   Presenting your structures: the CCP4mg molecular-graphics software. Acta Crystallogr D Biol Crystallogr  2011; 67: 386– 94. Google Scholar CrossRef Search ADS PubMed  Metcalf WW, Jiang W, Daniels LLet al.   Conditionally replicative and conjugative plasmids carrying lacZ alpha for cloning, mutagenesis, and allele replacement in bacteria. Plasmid  1996; 35: 1– 13. Google Scholar CrossRef Search ADS PubMed  Minami A, Hashimoto S, Abe Het al.   Cholera enterotoxin production in Vibrio cholerae O1 strains isolated from the environment and from humans in Japan. Appl Environ Microbiol  1991; 57: 2152– 7. Google Scholar PubMed  Montilla R, Chowdhury MA, Huq Aet al.   Serogroup conversion of Vibrio cholerae non-O1 to Vibrio cholerae O1: effect of growth state of cells, temperature, and salinity. Can J Microbiol  1996; 42: 87– 93. Google Scholar CrossRef Search ADS PubMed  Nagamune K, Yamamoto K, Naka Aet al.   In vitro proteolytic processing and activation of the recombinant precursor of El Tor cytolysin/hemolysin (pro-HlyA) of Vibrio cholerae by soluble hemagglutinin/protease of V. cholerae, trypsin, and other proteases. Infect Immun  1996; 64: 4655– 8. Google Scholar PubMed  Ogierman MA, Fallarino A, Riess Tet al.   Characterization of the Vibrio cholerae El Tor lipase operon lipAB and a protease gene downstream of the hly region. J Bacteriol  1997; 179: 7072– 80. Google Scholar CrossRef Search ADS PubMed  Olson R, Gouaux E. Crystal structure of the Vibrio cholerae cytolysin (VCC) pro-toxin and its assembly into a heptameric transmembrane pore. J Mol Biol  2005; 350: 997– 1016. Google Scholar CrossRef Search ADS PubMed  Piela L, Nemethy G, Scheraga HA. Proline-induced constraints in alpha-helices. Biopolymers  1987; 26: 1587– 600. Google Scholar CrossRef Search ADS PubMed  Pradhan S, Mallick SK, Chowdhury R. Vibrio cholerae classical biotype is converted to the viable non-culturable state when cultured with the El Tor biotype. PLoS One  2013; 8: e53504. Google Scholar CrossRef Search ADS PubMed  Pukatzki S, Ma AT, Sturtevant Det al.   Identification of a conserved bacterial protein secretion system in Vibrio cholerae using the Dictyostelium host model system. Proc Natl Acad Sci U S A  2006; 103: 1528– 33. Google Scholar CrossRef Search ADS PubMed  Rader AE, Murphy JR. Nucleotide sequences and comparison of the hemolysin determinants of Vibrio cholerae El Tor RV79(Hly+) and RV79(Hly−) and classical 569B(Hly−). Infect Immun  1988; 56: 1414– 9. Google Scholar PubMed  Rosenau F, Jaeger K. Bacterial lipases from Pseudomonas: regulation of gene expression and mechanisms of secretion. Biochimie  2000; 82: 1023– 32. Google Scholar CrossRef Search ADS PubMed  Roy A, Kucukural A, Zhang Y. I-TASSER: a unified platform for automated protein structure and function prediction. Nat Protoc  2010; 5: 725– 38. Google Scholar CrossRef Search ADS PubMed  Ruhe ZC, Low DA, Hayes CS. Bacterial contact-dependent growth inhibition. Trends Microbiol  2013; 21: 230– 7. Google Scholar CrossRef Search ADS PubMed  Russell AB, Hood RD, Bui NKet al.   Type VI secretion delivers bacteriolytic effectors to target cells. Nature  2011; 475: 343– 7. Google Scholar CrossRef Search ADS PubMed  Saha N, Banerjee KK. Carbohydrate-mediated regulation of interaction of Vibrio cholerae hemolysin with erythrocyte and phospholipid vesicle. J Biol Chem  1997; 272: 162– 7. Google Scholar CrossRef Search ADS PubMed  Shimada T, Arakawa E, Itoh Ket al.   Extended serotyping scheme for Vibrio cholerae. Curr Microbiol  1994; 28: 175– 8. Google Scholar CrossRef Search ADS   Stoebner JA, Payne SM. Iron-regulated hemolysin production and utilization of heme and hemoglobin by Vibrio cholerae. Infect Immun  1988; 56: 2891– 5. Google Scholar PubMed  Wahba AH. Vibriocine production in the cholera and El Tor vibrios. Bull WHO  1965; 33: 661– 4. Google Scholar PubMed  Waldor MK, Tschape H, Mekalanos JJ. A new type of conjugative transposon encodes resistance to sulfamethoxazole, trimethoprim, and streptomycin in Vibrio cholerae O139. J Bacteriol  1996; 178: 4157– 65. Google Scholar CrossRef Search ADS PubMed  West PA, Lee JV. Ecology of Vibrio species, including Vibrio cholerae, in natural waters in Kent, England. J Appl Bacteriol  1982; 52: 435– 48. Google Scholar CrossRef Search ADS PubMed  Williams SG, Manning PA. Transcription of the Vibrio cholerae haemolysin gene, hlyA, and cloning of a positive regulatory locus, hlyU. Mol Microbiol  1991; 5: 2031– 8. Google Scholar CrossRef Search ADS PubMed  Williams SG, Varcoe LT, Attridge SRet al.   Vibrio cholerae Hcp, a secreted protein coregulated with HlyA. Infect Immun  1996; 64: 283– 9. Google Scholar PubMed  Yamamoto K, Al-Omani M, Honda Tet al.   Non-O1 Vibrio cholerae hemolysin: purification, partial characterization, and immunological relatedness to El Tor hemolysin. Infect Immun  1984; 45: 192– 6. Google Scholar PubMed  Yamamoto K, Ichinose Y, Shinagawa Het al.   Two-step processing for activation of the cytolysin/hemolysin of Vibrio cholerae O1 biotype El Tor: nucleotide sequence of the structural gene (hlyA) and characterization of the processed products. Infect Immun  1990; 58: 4106– 16. Google Scholar PubMed  Yang J, Yan R, Roy Aet al.   The I-TASSER Suite: protein structure and function prediction. Nat Methods  2015; 12: 7– 8. Google Scholar CrossRef Search ADS PubMed  Yang J, Zhang Y. I-TASSER server: new development for protein structure and function predictions. Nucleic Acids Res  2015; 43: W174– 81. Google Scholar CrossRef Search ADS PubMed  Zhang Y. I-TASSER server for protein 3D structure prediction. BMC Bioinformatics  2008; 9: 40. Google Scholar CrossRef Search ADS PubMed  Zheng J, Ho B, Mekalanos JJ. Genetic analysis of anti-amoebae and anti-bacterial activities of the type VI secretion system in Vibrio cholerae. PLoS One  2011; 6: e23876. Google Scholar CrossRef Search ADS PubMed  Zheng J, Tung SL, Leung KY. Regulation of a type III and a putative secretion system in Edwardsiella tarda by EsrC is under the control of a two-component system, EsrA-EsrB. Infect Immun  2005; 73: 4127– 37. Google Scholar CrossRef Search ADS PubMed  © FEMS 2017. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png FEMS Microbiology Ecology Oxford University Press

New facet of non-O1/non-O139 Vibrio cholerae hemolysin A: a competitive factor in the ecological niche

Loading next page...
 
/lp/ou_press/new-facet-of-non-o1-non-o139-vibrio-cholerae-hemolysin-a-a-competitive-HboVTG0nX2
Publisher
Blackwell
Copyright
© FEMS 2017. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
0168-6496
eISSN
1574-6941
D.O.I.
10.1093/femsec/fix113
Publisher site
See Article on Publisher Site

Abstract

Abstract Different serogroups of Vibrio cholerae may inhabit the same ecological niche. However, serogroup O1/O139 strains are rarely isolated from their ecological sources. Quite plausibly, the non-O1/non-O139 vibrios and other bacterial species suppress growth of O1/O139 strains that share the same niche. Our bacterial inhibition assay data indicated that certain non-O1/non-O139 strains used a contact-dependent type VI secretion system (T6SS) to suppress growth of the O1 El Tor, N16961 pandemic strain. Comparative proteomics of the O1 and the suppressive non-O1/non-O139 strains co-cultured in a simulated natural aquatic microcosm showed that SecB and HlyD were upregulated in the latter. The HlyD-related effective factor was subsequently found to be hemolysin A (HlyA). However, not all hlyA-positive non-O1/non-O139 strains mediated growth suppression of the N16961 V. cholerae; only strains harboring intact cluster I HlyA could exert this activity. The key feature of the HlyA is located in the ricin-like lectin domain (β-trefoil) that plays an important role in target cell binding. In conclusion, the results of this study indicated that non-O1/non-O139 V. cholerae suppressed the growth of the O1 pandemic strain by using contact-dependent T6SS as well as by secreting the O1-detrimental hemolysin A during their co-persistence in the aquatic habitat. bacterial competition, hemolysin A, growth suppressive factor, ecological niche, non-O1/non-O139 Vibrio cholerae, O1/O139 Vibrio cholerae INTRODUCTION Vibrio cholerae bacteria are naturally found in aquatic ecosystems such as riverine, estuarine, marine and coastal waters (Colwell and Huq 1994, 2001; Faruque, Albert and Mekalanos 1998; Binsztein et al.2004; Constantin de Magny et al.2008). The bacterial strains are classified into more than 200 serogroups based on the surface O antigens (Shimada et al.1994). Only strains of O1 and O139 serogroups have inherent cholera epidemic/pandemic potential while those of non-O1/non-O139 serogroups cause merely sporadic infection such as gastroenteritis and/or septicemia. It has been observed that the isolation rate of the O1/O139 vibrios from native habitats is low compared with non-O1/non-O139 counterparts, even in cholera epidemic localities (West and Lee 1982; Khan et al.1984; Huq et al.1990; Minami et al.1991; Alam et al.2015). Several attributes have been used to explain this phenomenon. It has been shown that non-O1/non-O139 V. cholerae survive better in the aquatic environment (Colwell 1984). Also, the possibility exists of serogroup conversion between O1/O139 and non-O1/non-O139 V. cholerae (Colwell et al.1995; Montilla et al.1996). However, whether the non-O1/non-O139 V. cholerae could directly inhibit the growth of the O1/O139 epidemic/pandemic strains during their co-persistence in the environment remains unknown. Inhibition of growth and killing of heterologous bacteria by means of contact-dependent and/or contact-independent mechanisms have been used by various bacterial species to compete for niches and to thrive (Ruhe, Low and Hayes 2013). The classic onslaught that V. cholerae uses to kill sensitive strains and to acquire environmentally fit genes is by bacteriophages (Faruque and Mekalanos 2012). Integrating conjugative elements (ICEs), a self-transmissible genetic material found in some V. cholerae genomes, also transfer the genes that facilitate rapid adaptation of the bacteria to environmental changes as well as providing resistance of the host bacterium to antibiotics and heavy metals (Waldor, Tschape and Mekalanos 1996; Boltner et al.2002; Burrus, Marrero and Waldor 2006). Moreover, V. cholerae employ types III and VI secretion systems (T3SS and T6SS, respectively) as the weapons to kill other species such as Pseudomonas aeruginosa, Escherichia coli, and Dictyostelium discoideum (Dziejman et al.2005; Pukatzki et al.2006; Russell et al.2011; Zheng, Ho and Mekalanos 2011). Vibrio cholerae secrete bacteriocins and other toxins into the extracellular milieu to destroy their opponents (Wahba 1965). In this study, the mechanism that the non-O1/non-O139 vibrios used to compete with the O1 counterpart for their ecological habitat was investigated to better understand why the former are predominantly isolated from environmental samples. MATERIALS AND METHODS Bacterial strains, plasmid and culture conditions Four hundred and thirty isolates of non-O1/non-O139 V. cholerae collected during 2006–2009 from four environmental water sources in Thailand comprising Mae Klong River (MK, 73 isolates), Tha Chin River (TC, 195 isolates), Sanamchai Canal (SC, 97 isolates) and Bangkhunsri Canal (BK, 65 isolates) were used as the tested strains in a bacterial inhibition assay. Vibrio cholerae O1 El Tor strain N16961, O1 classical strain O395 and O139 serogroup were used as the indicator strains. The MK2-25 strain was used as a parental wild-type strain for mutant construction and as a tested strain in an in vitro competition assay. Escherichia coli strains DH5α λpir and SM10 λpir were used for cloning and mating, respectively. The pWM91 plasmid was used for mutant construction. Bacteria and plasmids used in this study are listed in Supplementary Table S1. All were obtained from stock cultures of the Department of Microbiology, Faculty of Public Health, Mahidol University. Luria–Bertani (LB) agar and broth were used to culture the bacteria at 37°C under shaking or static conditions. Streptomycin (100 μg mL−1), kanamycin (50 μg mL−1) and ampicillin (100 μg mL−1) were added to the medium when required. Bacterial inhibition assay Four hundred and thirty isolates of non-O1/non-O139 V. cholerae were tested for inhibitory activity against the O1 classical (strain O395) and El Tor (pandemic strain N16961) and O139 V. cholerae using a bacterial inhibition assay. Non-O1/non-O139 V. cholerae from stock cultures were streaked as a short line onto LB agar and incubated overnight at 37°C (45 isolates per plate). Meanwhile, indicator strains (O1 classical and El Tor, and O139 V. cholerae) were individually cultivated with shaking in LB broth at 37°C for 16 h. The overnight cultures were further diluted in 0.85% normal saline solution to a turbidity comparable to that of a 0.5 McFarland turbidity standard (∼108 cells mL−1), then thoroughly swabbed onto LB agar plates. Each of the tested strains was picked from the master plate, streaked in triplicate onto the lawn of the three indicator strains and incubated overnight at 37°C. A circumferential inhibition zone around the streaked line of the tested strain on the lawn of the indicator strain was observed and recorded in centimeters. Plaque formation assay Mid-logarithmic phase cultures of the non-O1/non-O139 V. cholerae that inhibited the indicator strain were treated with mitomycin C at a final concentration 200 ng mL−1. The cultures were incubated at 37°C for 4 h to induce the release of bacteriophage in the culture supernatant. Cell-free culture supernatants (CFSs) of the mitomycin C-treated tested strain cultures were collected by centrifugation, then filtered through a 0.2 μm membrane (Nalgene, USA) and stored at 4°C until used. In the meantime, mid-logarithmic phase cells (∼107 cells) of the indicator strains were harvested by centrifugation, mixed with 100 μL of the mitomycin C-treated CFS and incubated at 37°C for 1 h to allow the infection of bacteriophage. The mixtures were transferred to sterile tubes containing 4 mL of 0.35% soft LB agar, then poured and swirled in a circle onto LB agar plates. The plates were kept on the bench until the soft agar solidified, incubated at 37°C overnight and the plaque formation on the indicator's lawn was observed. Agar-well diffusion assay To investigate bacteriocin production, supernatants were collected by centrifugation from the late stationary phase cultures of the non-O1/non-O139 V. cholerae that inhibited the indicator strain, filtered through a 0.2 μm membrane (Nalgene, USA), and stored at 4°C until used. In the meantime, mid-logarithmic phase cells (∼107 cells) of the indicator strains were harvested by centrifugation, separately mixed with 4 mL of 0.35% soft agar, poured onto LB agar plates, swirled in a circle and then kept on the bench until the soft agar solidified. The solidified soft agar was drilled to give wells of 6 mm diameter. Each well was filled with 50 μL of an individual supernatant. The plate was placed at room temperature for 1 h, then incubated at 37°C overnight, and the clear zone around the well was observed. Detection of T3SS, T6SS and ICEs by PCR Overnight cultures of the non-O1/non-O139 V. cholerae that inhibited the indicator strain were heated to extract DNA and subjected to PCR. The reaction mixture contained 500 ng of DNA template, 1× ThermoPol reaction buffer (20 mM Tris–HCl pH 8.8, 10 mM (NH4)2SO4, 10 mM KCl, 2 mM MgSO4, 0.1% Triton X-100; New England Biolabs, USA), 200 μM of each dNTP, 200 nM of each primer and 1 unit of Taq DNA polymerase (New England Biolabs, USA), and was prepared in a final volume of 20 μL. The thermocycling program was optimized as follows: an initial denaturation at 95°C for 2 min, followed by 35 cycles of denaturation at 95°C for 15 s, annealing at 49–55°C for 30 s (Supplementary Table S2), and extension at 68°C for 45 s. After a final extension at 68°C for 5 min, PCR tubes were cooled to 4°C and the amplicon was electrophoresed and observed under UV light using the gel documentation system (BIS 303 PC, DNR Bio-Imaging Systems, Israel). Primers used to detect the representative genes of T3SS, T6SS and ICEs are listed in Supplementary Table S2. Preparation of laboratory microcosm The environmental microcosm was prepared for in vitro competition assay by collecting freshwater from the Chao Phraya River, Thailand. Five liters of freshwater was collected and sequentially filtered through 0.8 and 0.1 μm pore size membrane filters (Nalgene, USA) to remove debris and bacterial cells in both normal and dormant states, respectively. Physical characteristics of the river water such as pH, temperature and electrical conductivity were determined using a portable digital pH meter (pH Testr 2, Oakton Instruments, USA) and a conductivity/total dissolved solids/°C meter (Cyberscan CON II, Eutech Instruments, USA). This filtered river water was kept in a sterile bottle at ambient temperature as bacterial free-freshwater. In vitro competition assay Mid-logarithmic phase cells of the selected non-O1/non-O139 V. cholerae and N16961 strains were harvested by centrifugation. Cell pellets were washed three times with 1 mL of sterile river water (described above) and individually resuspended in 25 mL of sterile river water (325 μS conductivity, 30.6°C, pH 7.0), then pooled in a 1:1 ratio (non-O1/non-O139:N16961) in a sterile 250 mL Erlenmeyer flask and placed at ambient temperature. The growth profile of each strain was analyzed by the plate count technique using LB agar and LB agar containing 100 μg mL−1 streptomycin (the N16961 strain is streptomycin resistance). This differential count (colony forming units (CFU) mL−1) was used to calculate the competitive index by dividing the proportion of the tested strain/N16961 at the end to the proportion of the tested strain/N16961 at the start. Meanwhile, a mono-cultivation assay was also performed by cultivating each of the tested strains individually in 50 mL sterile river water to observe any defect in microcosm adaptation. All of the experiments were performed in triplicate and mean CFU mL−1 was used for calculation. Growth rate determination To monitor the growth rates of the tested strains in nutrient rich and nutrient deprived environments, late stationary phase culture of each strain was cultivated in 250 mL Erlenmeyer flasks containing 20 mL of high-nutrient (LB broth), average-nutrient (LB:sterile DW = 1:1) or low-nutrient medium (LB:sterile DW = 1:3) and incubated at 37°C with shaking. One milliliter of bacterial culture from each flask was pipetted into a plastic cuvette to measure the optical density at wavelength 600 nm (OD600nm) using a spectrophotometer (UV-160A, Shimadzu, Japan). Gene knockout In-frame deletions of secB, hlyD and hlyA were conducted by means of overlapped extension PCR and secB-based allelic exchange as previously described using pWM91 plasmid (Metcalf et al.1996; Zheng, Tung and Leung 2005). All primers used for constructing the mutant strains are listed in Supplementary Table S2. Effect of residual protein(s) in supernatant on growth pattern The N16961 strain was grown in the CFSs of the wild-type suppressive bacteria and mutants at 37°C with shaking. The growth pattern was observed by measuring the OD600nm using a spectrophotometer (UV-160A, Shimadzu, Japan). Detection and in silico analysis of hlyA Genomic DNAs of bacteria were extracted using the NucleoSpin® Tissue Kit (Macherey-Nagel, Germany) following the manufacturer’s instructions. All preparations were subjected to monoplex PCR to detect the presence of the hlyA gene using hlyA-specific primers (Supplementary Table S2). The PCR mixture (50 μL final volume) for hlyA amplification contained 1 μg of genomic DNA, 1× KAPA HiFi Fidelity buffer (Kapa Biosystems, USA), 300 μM of each dNTP, 250 nM of each primer (pTrcHis2-hlyA-F and pTrcHis2-hlyA-R) and 1 unit of KAPA HiFi DNA polymerase (Kapa Biosystems, USA). The thermocycling program was optimized as follows: an initial denaturation at 95°C for 2 min, followed by 35 cycles of denaturation at 98°C for 20 s, annealing at 65°C for 15 s and extension at 72°C for 30 s. After a final extension at 72°C for 5 min, PCR tubes were cooled to 4°C and the amplicons were electrophoresed and observed under UV light using the gel documentation system (BIS 303 PC, DNR Bio-Imaging Systems, Israel). PCR products were cleaned up using a QIAquick PCR Purification Kit (Qiagen GmbH, Germany). The eluted PCR products were submitted to nucleotide sequencing (SolGent, South Korea). The sequences were submitted to the National Center for Biotechnology Information (NCBI) to obtain accession numbers. The hlyA sequences of the non-O1/non-O139 V. cholerae strains MK2-25, SC1-09, TC1-21, MK4-26, SC1-34, SC2-10 and TC2-57 were deposited in GenBank under accession numbers KU888666, KU888667, KU888668, KY369196, KY369197, KY369198 and KY369199, respectively. These sequences were then compared with those of other strains available in the NCBI GenBank database and a phylogenetic tree was constructed using Geneious version 7.1 (Kearse et al.2012). Their respective amino acid sequences were also compared. The PDB file and a ribbon structure of proteins were created using the SWISS-MODEL server (Arnold et al.2006) and CCP4mg software (McNicholas et al.2011), respectively. The ligand binding site within the β-trefoil domain was predicted using the I-TASSER server (Zhang 2008; Roy, Kucukural and Zhang 2010; Yang et al.2015; Yang and Zhang 2015). RESULTS AND DISCUSSION Non-O1/non-O139 V. cholerae inhibited growth of the O1 and O139 strains Because O1 and O139 V. cholerae are rarely isolated from the aquatic environments (West and Lee 1982; Khan et al.1984; Huq et al.1990; Minami et al.1991; Alam et al.2015), it was hypothesized that their growth might be restrained by the non-O1/non-O139 strains during their co-persistence. To validate this speculation, 430 isolates of non-O1/non-O139 V. cholerae obtained from two rivers and two canals in central Thailand during 2006–2009 were tested for their inhibitory activity on indicator O1 classical (O395) and El Tor (N16961) and O139 strains by bacterial inhibition assay. From the four sample collection sites, no O1/O139 was isolated, which conformed to the data reported previously from different geographical areas that O1 strains were rarely isolated from natural ecosystems (West and Lee 1982; Khan et al.1984; Huq et al.1990; Minami et al.1991; Alam et al.2015). From the growth inhibition assay, 22 non-O1/non-O139 isolates (5.1%) inhibited the growth of the indicators as shown by clear circumferences (inhibition halos) around the effectors’ streaked lines on the indicator lawns grown on agar. They comprised nine isolates, each from the TC and SC, four isolates from MK, and none from BK (Supplementary Table S3). Examples of the halos of growth inhibition are shown in Supplementary Fig. S1. The widths of the halos varied from 0.3 to 1.1 cm (Supplementary Table S3). Among the three indicators, O1 El Tor, N16961 (an isolate of the seventh cholera pandemic) was the most vulnerable to environmental non-O1/non-O139 inhibitory activity (Supplementary Table S3). A majority of the effective non-O1/non-O139 isolates (19/22 isolates; nos TC1-21, MK4-26, TC1-57, TC2-07, TC2-08, TC2-57, TC3-08, MK1-11, MK2-25, MK4-13, SC1-09, SC1-34, SC1-35, SC1-36, SC1-37, SC1-38, SC1-39, SC2-10 and SC2-38) inhibited O1 El Tor, strain N16961; three isolates (nos TC2-37, TC2-52 and TC2-59) inhibited O1 classical strain O395; and two (nos TC1-21 and MK4-26) were effective against all indicator strains. The inhibitory property of most of the tested strains except the TC1-21 and MK4-26 was against only one single indicator, indicating that intra-species inhibition is highly strain specific. Growth inhibition among the three indicators themselves was not observed (Supplementary Table S3). Usually the toxin-producing bacteria protect themselves from their own detrimental substance(s) using diverse mechanisms including immunity (e.g. producing immunity protein), resistance (e.g. modification of toxin receptor), tolerance (e.g. changing the translocation system) and production of anti-toxin proteins (Lazdunski 1988; Alonso, Vilchez and Rodriguez Lemoine 2000). For V. cholerae, non-O1/non-O139 V. cholerae strain V52 inhibited the growth of E. coli and Dictyostelium discoideum, which share the ecological niche, using T6SS as a competitive weapon (Pukatzki et al.2006; Zheng, Ho and Mekalanos 2011). Moreover, it protected itself by producing immunity proteins encoded by genes of the T6SS cassette (MacIntyre et al.2010; Dong et al.2013). T6SS is a contact-dependent inhibitory factor of non-O1/non-O139 V. cholerae Because the non-O1/non-O139 V. cholerae inhibited the O1 and O139 strains by a contact-dependent mode as observed in the previous experiment, several possible factors that might be used by the inhibitors for niche competition were investigated including bacteriophage (by plaque formation assay), bacteriocin production (by agar-well diffusion assay) and the presence of the T3SS, T6SS and SXT family of ICEs (by PCR). The results showed that none of the inhibitory non-O1/non-O139 strains released bacteriophage or produced bacteriocin to inhibit the growth of the indicators. Likewise, four representative genes of T3SS, i.e. vcsN2, vcsV2, vcsC2 and vspD, were undetected among the inhibitors. However, a gene coding for protein(s) of T6SS, known to be a weapon related to contact-dependent inhibition, was detected in all of the inhibitory isolates (Supplementary Table S4). Ten isolates (nos TC1-21, TC2-57, TC2-37, TC2-52, MK4-26, SC1-34, SC1-35, SC1-36, SC1-38 and TC2-07) were positive for hcp-2 and vgrG-2, which code for the T6SS apparatus. Four isolates (SC1-09, SC2-10, SC2-38 and TC2-08) were positive for hcp-2 only and eight isolates (MK1-11, TC2-59, MK2-25, MK4-13, SC1-37, SC1-39, TC1-57 and TC3-08) were positive for vgrG-2 only. Two isolates (nos TC1-21 and TC2-57) were positive for tseL and four isolates (TC2-37, TC2-52, MK1-11 and TC2-59) were positive for vgrG-3; both genes encode T6SS effectors. No isolate was positive for vasX, which encodes a dispensable factor for assembling the T6SS translocon complex. The SXT-R391 family of ICEs, comprising important factors for survival, was found previously in genomes of some V. cholerae isolates (Waldor, Tschape and Mekalanos 1996; Boltner et al.2002; Burrus, Marrero and Waldor 2006). The element was detected in only 3/22 effective non-O1/non-O139 strains (nos TC2-52, MK1-11 and MK4-26), suggesting that ICEs might not be the key inhibitory property of the effective non-O1/non-O1 vibrios. Non-O1/non-O139 V. cholerae suppress growth of O1 strain in an aquatic microcosm To study the non-O1/non-O139 V. cholerae-mediated growth suppression of the O1 pandemic strain in the natural aquatic niche, the bacterial competition assay was performed in a simulated environmental microcosm. Vibrio cholerae O1 El Tor, N16961 and four non-O1/non-O139 effectors, i.e. MK2-25, SC1-09, TC1-21 and MK1-11, which exerted varying degrees of N16961 growth suppression (widths of the inhibition halos were 0.9, 0.7, 0.5 and 0.3 cm, respectively), were selected for co-cultivation at ambient temperature. The daily growth profiles of the bacteria were monitored and compared. The results showed that MK2-25 and SC1-09 overtook N16961 growth in the microcosm (Fig. 1A. and B). The initial amounts of the effective MK2-25 and N16961 indicators were 7.5 × 107 and 5.8 × 107 CFU mL−1, respectively. After day 1 of co-cultivation, cultivable N16961 declined to 1.0 × 106 CFU mL−1 (58 fold), whereas MK2-25 increased to 1.1 × 108 CFU mL−1. The cultivable MK2-25 cell numbers ranged from ∼100 to 10 000 fold higher than that of N16961 throughout the co-cultivation period. At day 49, growth of N16961 was undetected while the live MK2-25 was 2.8 × 102 CFU mL−1; the live MK2-25 could be detected through day 163 (Fig. 1A). SC1-09 also suppressed growth of N16961 during the co-cultivation (Fig. 1B). The N16961 growth suppression was not observed during co-cultivation with the TC1-21 and MK1-11 strains. The numbers of cultivable TC1-21 cells were relatively similar to N16961 throughout the co-cultivation course (Fig. 1C), whereas the MK1-11 cells were fewer than those of N16961 (Fig. 1D). Figure 1. View largeDownload slide Co-cultivations of the pandemic O1 El Tor, N16961 strain with the non-O1/non-O139 strains MK2-25 (A), SC1-09 (B), TC1-21 (C) and MK1-11 (D) in a simulated aquatic microcosm. Mean ± standard deviation (SD) of colony forming units per milliliter (CFU mL−1) of individual strains are shown. Data are representative of two reproducible independent experiments Figure 1. View largeDownload slide Co-cultivations of the pandemic O1 El Tor, N16961 strain with the non-O1/non-O139 strains MK2-25 (A), SC1-09 (B), TC1-21 (C) and MK1-11 (D) in a simulated aquatic microcosm. Mean ± standard deviation (SD) of colony forming units per milliliter (CFU mL−1) of individual strains are shown. Data are representative of two reproducible independent experiments To confirm that the decline of living N16961 cells during co-cultivation with the effectors was due to suppressive factor(s) released from the latter and not from the defective adaptation of the former to the microcosm, the numbers of live N16961, MK2-25, SC1-09, TC1-21 and MK1-11 in mono-cultivations were monitored. No difference was observed in the cultivable cell numbers of N16961, MK2-25 and SC1-09 during mono-cultivations (Fig. 1A and B), implying that the N16961 growth retardation in the co-cultivations was not due to a bacterial defect in microcosm adaptation. The cultivable cell numbers of MK2-25 and SC1-09 during co-cultivation were slightly higher than those during the mono-cultivations (Fig. 1A and B). Quite possibly, they (the predators in the competitive microcosms) used by-products derived from the dead N16961 cells (the prey) to increase. This phenomenon has been described previously (Helling, Vargas and Adams 1987). The numbers of cultivable N16961 and MK1-11 cells did not differ during the mono- and co-cultivations (Fig. 1D) suggesting that the observed growth retardation of MK1-11 was not a result of a suppression factor released from N16961, but rather due to a defect in adapting to the new environment. Growth patterns of V. cholerae in nutrient-rich and nutrient-deprived environments It has been reported that the thriving of certain V. cholerae strains in an aquatic ecosystem varies depending on the ability of the bacteria to adapt to various environmental stresses, such as nutrient deprivation (Lutz et al.2013). In this study, the growth profiles of MK2-25, SC1-09, TC1-21 and N16961 were investigated under high and low nutritional status. The growth rate of N16961 did not differ from those of MK2-25, SC1-09 and TC1-21 in all concentrations of the LB broth used for cultivation (undiluted and diluted 1:2 and 1:4; Fig. 2A–C). This finding supports the above notion that growth retardation of N16961 during co-cultivation was the effect of suppressive factor(s) of the non-O1/non-O139 strain. Figure 2. View largeDownload slide Growth patterns of the mono-cultivated MK2-25, SC1-09, TC1-21 and N16961 strains in LB broth: (A) undiluted, (B) 1:2 diluted and (C) 1:4 diluted. The results are the means of two reproducible independent experiments. Figure 2. View largeDownload slide Growth patterns of the mono-cultivated MK2-25, SC1-09, TC1-21 and N16961 strains in LB broth: (A) undiluted, (B) 1:2 diluted and (C) 1:4 diluted. The results are the means of two reproducible independent experiments. SecB and HlyD endowed the non-O1/non-O139 V. cholerae with a growth suppressive property Under the simulated ecologic system, the ability of non-O1/non-O139 strains to suppress N16961 did not correlate with the presence of T6SS-specific genes (Fig. 1 and Supplementary Table S4). Strains SC1-09, SC2-38, TC2-08, TC2-59, MK2-25, MK4-13, TC1-57 and TC3-08 lack the essential T6SS component, yet they could suppress growth of N16961. Thus, T6SS should not be considered the sole factor used by the effective non-O1/non-O139 stains for niche competition with the O1 pandemic strain. Comparative proteomics data revealed that the MK2-25 cells produced SecB and HlyD when co-cultured with N16961. SecB (a protein-transporter protein) is a key component of the Sec system that plays a cooperative role with T5SS to deliver protein effectors and toxins to targets (Driessen, Fekkes and van der Wolk 1998; Benz and Meinhart 2014). HlyD (a membrane fusion protein) is a key component of T1SS that exports a hemolysin toxin (Lee et al.2012). Based on their established functions, SecB and HlyD proteins were likely candidates responsible for O1 suppressive activity of the non-O1/non-O139 effectors. To validate this speculation, secB and hlyD deletion mutants, i.e. MK2-25ΔsecB and MK2-25ΔhlyD, were constructed. MK2-25ΔsecB and MK2-25ΔhlyD were found to lose their parental suppressive properties on N16961 as determined by in vitro competitive growth assay. The numbers of the N16961 strain were 100- to 10 000-fold lower than the MK2-25 wild-type, but only 10- to 100-fold lower than MK2-25ΔsecB and MK2-25ΔhlyD (P < 0.05; Fig. 3A–C). Moreover, the values of the daily competitive index of MK2-25ΔsecB (Fig. 4B) and MK2-25ΔhlyD (Fig. 4C) decreased significantly when compared with the wild-type (P < 0.05) (Fig. 4A). On day 1 of the co-cultivations, the competitive index value of MK2-25 wild-type was 85.95, while those of MK2-25ΔsecB and MK2-25ΔhlyD were 49.53 and 35.17, respectively (P < 0.05; Supplementary Fig. S2). The results indicated that the non-O1/non-O139-mediated O1 growth suppression began within the first 24 h of the co-cultures. This finding is similar to the results of a related study (Pradhan, Mallick and Chowdhury 2013) showing that the N16961 strain was able to outcompete the O1 classical O395 strain within 24 h of the co-cultivation. Figure 3. View largeDownload slide Growth patterns (mean ± SD in CFU mL−1) of N16961 indicator and suppressors of wild-type MK2-25 (A), MK2-25ΔsecB (B), MK2-25ΔhlyD (C) and MK2-25ΔhlyA (D) in individual mono- and co-cultures. Data are representative of two reproducible independent experiments. Figure 3. View largeDownload slide Growth patterns (mean ± SD in CFU mL−1) of N16961 indicator and suppressors of wild-type MK2-25 (A), MK2-25ΔsecB (B), MK2-25ΔhlyD (C) and MK2-25ΔhlyA (D) in individual mono- and co-cultures. Data are representative of two reproducible independent experiments. Figure 4. View largeDownload slide Competitive indices (CIs) of N16961 strain and wild-type MK2-25 (A), MK2-25ΔsecB (B), MK2-25ΔhlyD (C) and MK2-25ΔhlyA (D) during 7 days of co-cultivation. The competitive index of suppressor was calculated by dividing the mean proportion of the suppressor/N16961 at the indicated time (day 0, 1, 2, ... 7) by the mean proportion of the suppressor/N16961 at the start (day 0), while the competitive index of N16961 was obtained by dividing the mean proportion of the N16961/suppressor at the indicated time (day 0, 1, 2, ... 7) by the mean proportion of the N16961/suppressor at the start (day 0). Data are representative of two reproducible independent experiments. Significant difference between CIs of wild-type MK2-25 and each mutant were assessed using the Mann–Whitney U-test. Figure 4. View largeDownload slide Competitive indices (CIs) of N16961 strain and wild-type MK2-25 (A), MK2-25ΔsecB (B), MK2-25ΔhlyD (C) and MK2-25ΔhlyA (D) during 7 days of co-cultivation. The competitive index of suppressor was calculated by dividing the mean proportion of the suppressor/N16961 at the indicated time (day 0, 1, 2, ... 7) by the mean proportion of the suppressor/N16961 at the start (day 0), while the competitive index of N16961 was obtained by dividing the mean proportion of the N16961/suppressor at the indicated time (day 0, 1, 2, ... 7) by the mean proportion of the N16961/suppressor at the start (day 0). Data are representative of two reproducible independent experiments. Significant difference between CIs of wild-type MK2-25 and each mutant were assessed using the Mann–Whitney U-test. The secreted inhibitory factor of the non-O1/non-O139 effectors is HlyD dependent The SecB and HlyD themselves are not protein effectors; they are components of the protein secretion systems responsible for toxin secretion (Driessen, Fekkes and van der Wolk 1998; Lee et al.2012). Therefore, the protein(s) or toxin(s) translocated via these portals was of interest. Experiments were carried out to test whether the candidate toxin(s) is SecB or HlyD dependent by comparing the growth profiles of the N16961 strain in the cell-free CFSs of wild-type and secB- and hlyD-deletion mutants. It was found that the growth of N16961 (OD600nm) in the MK2-25ΔhlyD CFS was higher than in the wild-type CFS (P < 0.05) (Fig. 5). These results indicated that HlyD of the non-O1/non-O139 V. cholerae was principally involved in the transport of the N16961 growth suppressive factor(s). Figure 5. View largeDownload slide Growth patterns of N16961 during cultivation in cell-free culture supernatants (CFSs) of wild-type MK2-25, MK2-25ΔsecB, MK2-25ΔhlyD or MK2-25ΔhlyA mutants. Mean OD600nm of two independent experiments is shown. Statistical differences of OD600nm values between N16961 cultivated in wild-type CFS and in each of the mutant CFSs (ΔsecB, ΔhlyD, or ΔhlyA) at each time point were assessed by Student’s t-test. *P < 0.05 and **P < 0.01. Figure 5. View largeDownload slide Growth patterns of N16961 during cultivation in cell-free culture supernatants (CFSs) of wild-type MK2-25, MK2-25ΔsecB, MK2-25ΔhlyD or MK2-25ΔhlyA mutants. Mean OD600nm of two independent experiments is shown. Statistical differences of OD600nm values between N16961 cultivated in wild-type CFS and in each of the mutant CFSs (ΔsecB, ΔhlyD, or ΔhlyA) at each time point were assessed by Student’s t-test. *P < 0.05 and **P < 0.01. HlyA is the non-O1/non-O139 weapon for growth competition HlyD is a key component of T1SS. It forms a translocation apparatus with the inner membrane protein HlyB and the outer membrane channel protein TolC by acting as an adaptor protein that connects HlyB to TolC to provide a continuous transmembrane conduit for toxin export (Lee et al.2012). A well-known toxin secreted by this system is hemolysin A (HlyA; synonym V. cholerae cytolysin or VCC), which is a pore-forming cytotoxic and cytolytic toxin toward a wide spectrum of eukaryotic cells (Lee et al.2012). The effective MK2-25 with in-frame, non-polar deletion mutation of hlyA was tested for N16961 suppression. Like the ΔhlyD mutant, MK2-25ΔhlyA had reduced ability to suppress growth of N16961 compared with the wild-type (Fig. 3A and D; P < 0.05). Also, the competitive index value of MK2-25ΔhlyA reduced significantly compared with the wild-type (P < 0.05; Fig. 4A and D). Thus, HlyA produced by the non-O1/non-O139 V. cholerae was likely a growth suppressive factor during the co-cultivation with the N16961. For verification, growth of the N16961 strain in the CFS of wild-type and ΔhlyA mutant was monitored. As shown in Fig. 5, the OD600nm of N16961 grown in the ΔhlyA mutant CFS was higher than in the wild-type CFS from 3 to 8 h of culturing (P < 0.05). Thus, HlyA is the growth suppressive factor of the non-O1/non-O139 V. cholerae towards the O1 N16961. The growth rate of N16961 in the culture supernatant of the ΔhlyA mutant was higher than in the culture supernatants of wild-type, ΔsecB and ΔhlyD mutants (Fig. 5) indicating that HlyA is the secreted growth suppressive factor of the non-O1/non-O139 V. cholerae against the N16961 pandemic strain. The data demonstrated that hemolysin A of the non-O1/non-O139 V. cholerae strains not only plays a known critical role in cholera pathogenesis (Yamamoto et al.1984), but also exhibits a novel activity in competing for the ecological niche and thriving in ecosystems. Nucleotide sequences of effective and non-effective hemolysin A are different Vibrio cholerae HlyA belongs to a large class of pore-forming toxins encoded by a 2223 bp open reading frame that encodes an 82 kDa preHlyA protein (Rader and Murphy 1988). The toxin is secreted in water-soluble monomeric form as a proHlyA of 79 kDa. Each molecule of the proHlyA consists of a prodomain (the N-terminal 132 residues of 15 kDa, which is important for proper folding of the molecule but must be cleaved off to form mature toxins of 65 kDa with hemolytic activity), a core cytolytic domain containing a pre-stem (pore-forming loop), and two C-terminal domains with ricin/jacalin-like lectin domains called β-trefoil and β-prism for binding to the carbohydrate receptor on the target cell membrane (Yamamoto et al.1990; Nagamune et al.1996; Saha and Banerjee 1997). Once the toxin monomers bind to the receptor, self-oligomerization into a stable circular β-barrel amphipathic heptamer occurs followed by insertion of their stem loops into the membrane lipid bilayer to form a transmembrane pore. Most non-O1/non-O139 V. cholerae isolates carry hlyA genes (Yamamoto et al.1984). Nevertheless, only some strains could suppress growth of N16961 in the aquatic microcosm. To have some insight into the functional differences, HlyA-coding genes of the suppressive and non-suppressive non-O1/non-O139 strains were PCR amplified, sequenced and compared. The hlyA sequences of the suppressive MK2-25 and SC1-09 and the non-suppressive TC1-21 and MK4-26 strains were different. Based on the GenBank database, V. cholerae hlyA sequences can be classified in three clusters. The hlyA of non-O1/non-O139 suppressor strains belong to cluster I whereas those of the O1/O139 and non-O1/non-O139 non-suppressor strains belong to clusters II and III, respectively (Fig. 6). Analysis of the 35 V. cholerae hlyA sequences revealed 32 polymorphic loci (Supplementary data A). The hlyA sequences of MK2-25 and SC1-09 presented 99.0% identity and both were 98.5 to 99.2% identical to other hlyA sequences of the cluster I (Fig. 6). In contrast, the hlyA sequence of the TC1-21 strain showed similarity (98.3 to 98.8%) to O1 and O139 hlyA of cluster II (Fig. 6), which should explain why the TC1-21 could not suppress growth of the O1/O139 indicators because they should have similar (if not identical) immunological factor(s). Figure 6. View largeDownload slide Phylogenetic tree of partial hlyA nucleotide sequences of the non-O1/non-O139 V. cholerae strains MK2-25, SC1-09, TC1-21, MK4-26, SC1-34, SC2-10 and TC2-57 compared with other V. cholerae hlyA sequences from the GenBank database. The hlyA genes of non-O1/non-O139 suppressor strains belong to cluster I whereas those of O1/O139 and non-O1/non-O139 non-suppressor strains are in clusters II and III, respectively. The tree was generated using the UPGMA analysis in Geneious version 7.1. The scale bar indicates nucleotide substitutions per site. Figure 6. View largeDownload slide Phylogenetic tree of partial hlyA nucleotide sequences of the non-O1/non-O139 V. cholerae strains MK2-25, SC1-09, TC1-21, MK4-26, SC1-34, SC2-10 and TC2-57 compared with other V. cholerae hlyA sequences from the GenBank database. The hlyA genes of non-O1/non-O139 suppressor strains belong to cluster I whereas those of O1/O139 and non-O1/non-O139 non-suppressor strains are in clusters II and III, respectively. The tree was generated using the UPGMA analysis in Geneious version 7.1. The scale bar indicates nucleotide substitutions per site. Among the 32 single nucleotide polymorphisms of hlyA sequences, 26 loci were translated into the same amino acids. Interestingly, deduced amino acids from five polymorphic loci differed between the O1/O139 and non-O1/non-O139 clusters. One exceptionally important locus was at the nucleotide positions 1644–1646 (based on N16961 hlyA), which forms the ACC codon of threonine (Thr), in the hlyA of the suppressive MK2-25 and SC1-09 strains. At this locus, the strain TC1-21, which lacks the property of suppressing N16961, has the CCC codon for proline (Pro) (Supplementary data B). The codon at nt 1644–1646 (residue 548) is located in the β-trefoil domain important for receptor binding of monomeric HlyA. Changing the hydrophilic Thr to the hydrophobic Pro, a secondary structure breaker, interferes with the ligand binding site within the β-trefoil domain due to steric hindrance of the bulky prolyl ring (Piela, Nemethy and Scheraga 1987) as shown in Supplementary Fig. S3A–D. The consequence should be an inability of the toxin to bind to its receptor and oligomerize, which in effect is a failure to form a pore on the target membrane. The hlyA sequence of MK4-26 differed from the others and should be placed in cluster III (Fig. 6). It has one base deletion at position 942, leading to a frame shift and consequently formation of a TGA stop codon at positions 950–952, which leads to production of truncated HlyA (Supplementary data A). This should account for the inability of the MK4-26 strain to suppress the N16961 strain's growth upon co-cultivation. Rader and Murphy (1988) similarly reported that the hlyA sequence of V. cholerae O1 classical, 569B strain had an 11-bp deletion that interrupted the open reading frame, resulting in the production of truncated non-hemolytic HlyA (a 244-amino acid precursor form). The remaining 18 strains of non-O1/non-O139 V. cholerae that produced inhibition halos on the N16961 lawns were also tested for growth-suppressive property in the aquatic microcosm. Eight strains, i.e. MK4-13, SC1-38, SC2-38, TC1-57, TC2-07, TC2-08, TC2-59 and TC3-08, could suppress growth of N16961, while the other 10 isolates (MK1-11, SC1-34, SC1-35, SC1-36, SC1-37, SC1-39, SC2-10, TC2-37, TC2-52 and TC2-57) were refractory (Supplementary Table S4). It was found that hlyA sequences of the ineffective strains have a frame shift that should produce truncated HlyA, hence abolishing the toxin activity. The hlyA sequences of strains SC1-34, SC2-10 and TC2-57 are shown as representatives in Supplementary data A. In conclusion, the present study demonstrated that non-O1/non-O139 V. cholerae used T6SS to inhibit the N16961 strain in a contact-dependent manner and secreted HlyA to suppress the pandemic strain during co-persistence in the aquatic microcosms. Additional factors that may be involved in the intraspecies competition await further investigation. The proposed mechanism mediated by HlyA is described below. In the contact-independent mode, non-O1/non-O139 V. cholerae secretes HlyA after being activated by at least three regulators including the cholera toxin transcriptional activator (ToxR), hemolysin gene transcriptional activator (HlyU) and ferric uptake regulator (Fur) (Stoebner and Payne 1988; Williams and Manning 1991; DiRita 1994). HlyU and Fur induce HlyA, HlyB and HlyC expression (Stoebner and Payne 1988; Williams and Manning 1991; Ogierman et al.1997). ToxR induces only HlyB, which not only plays a critical role in protein export, but also acts as a chemotactic receptor in responding to nutrient concentration (Jeffery and Koshland 1993). HlyB regulates the production of the transmembrane pore or transporter that enables hemolysin secretion (Jeffery & Koshland 1993). After production, HlyA is translocated from the cytoplasm to the extracellular milieu using the conduit formed by HlyB, HlyD and TolC, which constitute the key components of T1SS (Lee et al.2012). In addition, HlyC, a lipase, is secreted by T2SS (Rosenau and Jaeger 2000) and hydrolyzes the N16961 outer membrane to facilitate HlyA penetration. The HlyA monomer interacts with the N16961 cell surface receptor using the specific ricin-like lectin domain (β-trefoil). After releasing the pro-domain, mature HlyA monomer self-assembles to a heptameric amphipathic β-barrel by circular oligomerization. Then the pre-stem loop within the core cytolysin domain of each protomer is inserted into the N16961 cell membrane and forms a pore leading to the target cell lysis (Olson and Gouaux 2005). In the contact-dependent mode of growth inhibition, the hemolysin-coregulated protein (Hcp) is produced under the regulation of HlyU similar to HlyA (Williams et al.1996). This toxic protein is injected into the N16961 cytoplasm directly by T6SS and causes the target’s death (Pukatzki et al.2006) (Fig. 7). The illustrated mechanisms allow the non-O1/non-O139 V. cholerae to suppress growth of O1 V. cholerae and hinder the recovery of the vegetative O1 when they are co-persistent in their natural aquatic habitats. Figure 7. View largeDownload slide Proposed mechanisms of contact-dependent and contact-independent growth suppression of pandemic O1, N16961 strain mediated by the non-O1/non-O139 strains. Solid lines indicate relationship based on existing data. ToxR and Fur are under environmental and iron regulation, respectively. Green lines indicate co-expression. Blue arrows indicate activation of protein expression. Dotted line indicates the predicted environmental stimuli. IM, inner membrane; OM, outer membrane; PP, periplasm. Figure 7. View largeDownload slide Proposed mechanisms of contact-dependent and contact-independent growth suppression of pandemic O1, N16961 strain mediated by the non-O1/non-O139 strains. Solid lines indicate relationship based on existing data. ToxR and Fur are under environmental and iron regulation, respectively. Green lines indicate co-expression. Blue arrows indicate activation of protein expression. Dotted line indicates the predicted environmental stimuli. IM, inner membrane; OM, outer membrane; PP, periplasm. SUPPLEMENTARY DATA Supplementary data are available at FEMSEC online. Acknowledgements This work was supported by the National Research University project of the Office of Commission on Higher Education, Ministry of Education, Thailand, through the Center of Biopharmaceutical Development and Innovative Therapy, Mahidol University. Author contributions: PD conceived the project, designed experiments, supervised PR and drafted the manuscript; PR did most of the experiments and prepared the figures; OR performed comparative proteomics; KS and WC helped PD. Conflict of Interest. None declared. REFERENCES Alam MT, Weppelmann TA, Longini Iet al.   Increased isolation frequency of toxigenic Vibrio cholerae O1 from environmental monitoring sites in Haiti. PLoS One  2015; 10: e0124098. Google Scholar CrossRef Search ADS PubMed  Alonso G, Vilchez G, Rodriguez Lemoine V. How bacteria protect themselves against channel-forming colicins. Int Microbiol  2000; 3: 81– 8. Google Scholar PubMed  Arnold K, Bordoli L, Kopp Jet al.   The SWISS-MODEL workspace: a web-based environment for protein structure homology modelling. Bioinformatics  2006; 22: 195– 201. Google Scholar CrossRef Search ADS PubMed  Benz J, Meinhart A. Antibacterial effector/immunity systems: it's just the tip of the iceberg. Curr Opin Microbiol  2014; 17: 1– 10. Google Scholar CrossRef Search ADS PubMed  Binsztein N, Costagliola MC, Pichel Met al.   Viable but nonculturable Vibrio cholerae O1 in the aquatic environment of Argentina. Appl Environ Microbiol  2004; 70: 7481– 6. Google Scholar CrossRef Search ADS PubMed  Boltner D, MacMahon C, Pembroke JTet al.   R391: a conjugative integrating mosaic comprised of phage, plasmid, and transposon elements. J Bacteriol  2002; 184: 5158– 69. Google Scholar CrossRef Search ADS PubMed  Burrus V, Marrero J, Waldor MK. The current ICE age: biology and evolution of SXT-related integrating conjugative elements. Plasmid  2006; 55: 173– 83. Google Scholar CrossRef Search ADS PubMed  Colwell RR Vibrios in the Environment . New York: Wiley, 1984. Colwell RR, Huq A. Viable but nonculturable V. cholerae. In: Wachsmuth K, Blake PA, Olsvik O (eds). Vibrio cholerae and Cholera: Molecular to Global Perspectives . Washington, DC: ASM Press, 1994, 116– 33. Colwell RR, Huq A. Marine ecosystems and cholera. Hydrobiologia  2001; 460: 141– 5. Google Scholar CrossRef Search ADS   Colwell RR, Huq A, Chowdhury MAet al.   Serogroup conversion of Vibrio cholerae. Can J Microbiol  1995; 41: 946– 50. Google Scholar CrossRef Search ADS PubMed  Constantin de Magny G, Murtugudde R, Sapiano MRet al.   Environmental signatures associated with cholera epidemics. Proc Natl Acad Sci U S A  2008; 105: 17676– 81. Google Scholar CrossRef Search ADS PubMed  DiRita VJ. Multiple regulatory systems in Vibrio cholerae pathogenesis. Trends Microbiol  1994; 2: 37– 8. Google Scholar CrossRef Search ADS PubMed  Dong TG, Ho BT, Yoder-Himes DRet al.   Identification of T6SS-dependent effector and immunity proteins by Tn-seq in Vibrio cholerae. Proc Natl Acad Sci U S A  2013; 110: 2623– 8. Google Scholar CrossRef Search ADS PubMed  Driessen AJ, Fekkes P, van der Wolk JP. The Sec system. Curr Opin Microbiol  1998; 1: 216– 22. Google Scholar CrossRef Search ADS PubMed  Dziejman M, Serruto D, Tam VCet al.   Genomic characterization of non-O1, non-O139 Vibrio cholerae reveals genes for a type III secretion system. Proc Natl Acad Sci U S A  2005; 102: 3465– 70. Google Scholar CrossRef Search ADS PubMed  Faruque SM, Albert MJ, Mekalanos JJ. Epidemiology, genetics, and ecology of toxigenic Vibrio cholerae. Microbiol Mol Biol Rev  1998; 62: 1301– 14. Google Scholar PubMed  Faruque SM, Mekalanos JJ. Phage-bacterial interactions in the evolution of toxigenic Vibrio cholerae. Virulence  2012; 3: 556– 65. Google Scholar CrossRef Search ADS PubMed  Helling RB, Vargas CN, Adams J. Evolution of Escherichia coli during growth in a constant environment. Genetics  1987; 116: 349– 58. Google Scholar PubMed  Huq A, Colwell RR, Rahman Ret al.   Detection of Vibrio cholerae O1 in the aquatic environment by fluorescent-monoclonal antibody and culture methods. Appl Environ Microbiol  1990; 56: 2370– 3. Google Scholar PubMed  Jeffery CJ, Koshland DEJr. Vibrio cholerae hlyB is a member of the chemotaxis receptor gene family. Protein Sci  1993; 2: 1532– 5. Google Scholar CrossRef Search ADS PubMed  Kearse M, Moir R, Wilson Aet al.   Geneious Basic: an integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics  2012; 28: 1647– 49. Google Scholar CrossRef Search ADS PubMed  Khan MU, Shahidullah MD, Haque MSet al.   Presence of vibrios in surface water and their relation with cholera in a community. Trop Geogr Med  1984; 36: 335– 40. Google Scholar PubMed  Lazdunski CJ. Pore-forming colicins: synthesis, extracellular release, mode of action, immunity. Biochimie  1988; 70: 1291– 6. Google Scholar CrossRef Search ADS PubMed  Lee M, Jun SY, Yoon BYet al.   Membrane fusion proteins of type I secretion system and tripartite efflux pumps share a binding motif for TolC in gram-negative bacteria. PLoS One  2012; 7: e40460. Google Scholar CrossRef Search ADS PubMed  Lutz C, Erken M, Noorian Pet al.   Environmental reservoirs and mechanisms of persistence of Vibrio cholerae. Front Microbiol  2013; 4: 375. Google Scholar CrossRef Search ADS PubMed  MacIntyre DL, Miyata ST, Kitaoka Met al.   The Vibrio cholerae type VI secretion system displays antimicrobial properties. Proc Natl Acad Sci U S A  2010; 107: 19520– 4. Google Scholar CrossRef Search ADS PubMed  McNicholas S, Potterton E, Wilson KSet al.   Presenting your structures: the CCP4mg molecular-graphics software. Acta Crystallogr D Biol Crystallogr  2011; 67: 386– 94. Google Scholar CrossRef Search ADS PubMed  Metcalf WW, Jiang W, Daniels LLet al.   Conditionally replicative and conjugative plasmids carrying lacZ alpha for cloning, mutagenesis, and allele replacement in bacteria. Plasmid  1996; 35: 1– 13. Google Scholar CrossRef Search ADS PubMed  Minami A, Hashimoto S, Abe Het al.   Cholera enterotoxin production in Vibrio cholerae O1 strains isolated from the environment and from humans in Japan. Appl Environ Microbiol  1991; 57: 2152– 7. Google Scholar PubMed  Montilla R, Chowdhury MA, Huq Aet al.   Serogroup conversion of Vibrio cholerae non-O1 to Vibrio cholerae O1: effect of growth state of cells, temperature, and salinity. Can J Microbiol  1996; 42: 87– 93. Google Scholar CrossRef Search ADS PubMed  Nagamune K, Yamamoto K, Naka Aet al.   In vitro proteolytic processing and activation of the recombinant precursor of El Tor cytolysin/hemolysin (pro-HlyA) of Vibrio cholerae by soluble hemagglutinin/protease of V. cholerae, trypsin, and other proteases. Infect Immun  1996; 64: 4655– 8. Google Scholar PubMed  Ogierman MA, Fallarino A, Riess Tet al.   Characterization of the Vibrio cholerae El Tor lipase operon lipAB and a protease gene downstream of the hly region. J Bacteriol  1997; 179: 7072– 80. Google Scholar CrossRef Search ADS PubMed  Olson R, Gouaux E. Crystal structure of the Vibrio cholerae cytolysin (VCC) pro-toxin and its assembly into a heptameric transmembrane pore. J Mol Biol  2005; 350: 997– 1016. Google Scholar CrossRef Search ADS PubMed  Piela L, Nemethy G, Scheraga HA. Proline-induced constraints in alpha-helices. Biopolymers  1987; 26: 1587– 600. Google Scholar CrossRef Search ADS PubMed  Pradhan S, Mallick SK, Chowdhury R. Vibrio cholerae classical biotype is converted to the viable non-culturable state when cultured with the El Tor biotype. PLoS One  2013; 8: e53504. Google Scholar CrossRef Search ADS PubMed  Pukatzki S, Ma AT, Sturtevant Det al.   Identification of a conserved bacterial protein secretion system in Vibrio cholerae using the Dictyostelium host model system. Proc Natl Acad Sci U S A  2006; 103: 1528– 33. Google Scholar CrossRef Search ADS PubMed  Rader AE, Murphy JR. Nucleotide sequences and comparison of the hemolysin determinants of Vibrio cholerae El Tor RV79(Hly+) and RV79(Hly−) and classical 569B(Hly−). Infect Immun  1988; 56: 1414– 9. Google Scholar PubMed  Rosenau F, Jaeger K. Bacterial lipases from Pseudomonas: regulation of gene expression and mechanisms of secretion. Biochimie  2000; 82: 1023– 32. Google Scholar CrossRef Search ADS PubMed  Roy A, Kucukural A, Zhang Y. I-TASSER: a unified platform for automated protein structure and function prediction. Nat Protoc  2010; 5: 725– 38. Google Scholar CrossRef Search ADS PubMed  Ruhe ZC, Low DA, Hayes CS. Bacterial contact-dependent growth inhibition. Trends Microbiol  2013; 21: 230– 7. Google Scholar CrossRef Search ADS PubMed  Russell AB, Hood RD, Bui NKet al.   Type VI secretion delivers bacteriolytic effectors to target cells. Nature  2011; 475: 343– 7. Google Scholar CrossRef Search ADS PubMed  Saha N, Banerjee KK. Carbohydrate-mediated regulation of interaction of Vibrio cholerae hemolysin with erythrocyte and phospholipid vesicle. J Biol Chem  1997; 272: 162– 7. Google Scholar CrossRef Search ADS PubMed  Shimada T, Arakawa E, Itoh Ket al.   Extended serotyping scheme for Vibrio cholerae. Curr Microbiol  1994; 28: 175– 8. Google Scholar CrossRef Search ADS   Stoebner JA, Payne SM. Iron-regulated hemolysin production and utilization of heme and hemoglobin by Vibrio cholerae. Infect Immun  1988; 56: 2891– 5. Google Scholar PubMed  Wahba AH. Vibriocine production in the cholera and El Tor vibrios. Bull WHO  1965; 33: 661– 4. Google Scholar PubMed  Waldor MK, Tschape H, Mekalanos JJ. A new type of conjugative transposon encodes resistance to sulfamethoxazole, trimethoprim, and streptomycin in Vibrio cholerae O139. J Bacteriol  1996; 178: 4157– 65. Google Scholar CrossRef Search ADS PubMed  West PA, Lee JV. Ecology of Vibrio species, including Vibrio cholerae, in natural waters in Kent, England. J Appl Bacteriol  1982; 52: 435– 48. Google Scholar CrossRef Search ADS PubMed  Williams SG, Manning PA. Transcription of the Vibrio cholerae haemolysin gene, hlyA, and cloning of a positive regulatory locus, hlyU. Mol Microbiol  1991; 5: 2031– 8. Google Scholar CrossRef Search ADS PubMed  Williams SG, Varcoe LT, Attridge SRet al.   Vibrio cholerae Hcp, a secreted protein coregulated with HlyA. Infect Immun  1996; 64: 283– 9. Google Scholar PubMed  Yamamoto K, Al-Omani M, Honda Tet al.   Non-O1 Vibrio cholerae hemolysin: purification, partial characterization, and immunological relatedness to El Tor hemolysin. Infect Immun  1984; 45: 192– 6. Google Scholar PubMed  Yamamoto K, Ichinose Y, Shinagawa Het al.   Two-step processing for activation of the cytolysin/hemolysin of Vibrio cholerae O1 biotype El Tor: nucleotide sequence of the structural gene (hlyA) and characterization of the processed products. Infect Immun  1990; 58: 4106– 16. Google Scholar PubMed  Yang J, Yan R, Roy Aet al.   The I-TASSER Suite: protein structure and function prediction. Nat Methods  2015; 12: 7– 8. Google Scholar CrossRef Search ADS PubMed  Yang J, Zhang Y. I-TASSER server: new development for protein structure and function predictions. Nucleic Acids Res  2015; 43: W174– 81. Google Scholar CrossRef Search ADS PubMed  Zhang Y. I-TASSER server for protein 3D structure prediction. BMC Bioinformatics  2008; 9: 40. Google Scholar CrossRef Search ADS PubMed  Zheng J, Ho B, Mekalanos JJ. Genetic analysis of anti-amoebae and anti-bacterial activities of the type VI secretion system in Vibrio cholerae. PLoS One  2011; 6: e23876. Google Scholar CrossRef Search ADS PubMed  Zheng J, Tung SL, Leung KY. Regulation of a type III and a putative secretion system in Edwardsiella tarda by EsrC is under the control of a two-component system, EsrA-EsrB. Infect Immun  2005; 73: 4127– 37. Google Scholar CrossRef Search ADS PubMed  © FEMS 2017. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com

Journal

FEMS Microbiology EcologyOxford University Press

Published: Jan 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

Monthly Plan

  • Read unlimited articles
  • Personalized recommendations
  • No expiration
  • Print 20 pages per month
  • 20% off on PDF purchases
  • Organize your research
  • Get updates on your journals and topic searches

$49/month

Start Free Trial

14-day Free Trial

Best Deal — 39% off

Annual Plan

  • All the features of the Professional Plan, but for 39% off!
  • Billed annually
  • No expiration
  • For the normal price of 10 articles elsewhere, you get one full year of unlimited access to articles.

$588

$360/year

billed annually
Start Free Trial

14-day Free Trial