Insight into the resilience and susceptibility of marine bacteria to T6SS attack by Vibrio cholerae and Vibrio coralliilyticus

The type VI secretion system (T6SS) is a nanomachine capable of killing adjacent microbial cells in a contact-dependent manner. Due to limited studies, relatively little is known about the range of marine bacteria that are susceptible to T6SS attack. Here, 15 diverse marine bacterial isolates from the phyla Bacteroidetes and Ɣ-Proteobacteria were challenged against the marine bacterium and human pathogen, Vibrio cholerae, which has a well described T6SS. V. cholerae killed several of the tested Ɣ-Proteobacteria, including members of the orders Vibrionales, Alteromonadales, Oceanospirillales, and Pseudomonadales. In contrast, V. cholerae co-existed with multiple Bacteroidetes and Ɣ-Proteobacteria isolates, but was killed by Vibrio coralliilyticus. Follow-up experiments revealed that five V. coralliilyticus strains, including known coral and shellfish pathogens survived the T6SS challenge and killed V. cholerae. By using predicted protein comparisons and mutagenesis, we conclude that V. coralliilyticus protected itself in the challenge by using its own T6SS to kill V. cholerae. This study provides valuable insight into the resilience and susceptibility of marine bacteria to the V. cholerae T6SS, and provides the first evidence for a functional T6SS in V. coralliilyticus, both of which have implications for human and ocean health.


Introduction
Bacterial-bacterial antagonism plays a major role in shaping bacterial community structure and function [1][2][3][4][5]. Early studies investigating marine bacterial-bacterial antagonism predominantly focused on the production and release of antibiotics by predatory bacteria as a means to inhibit their preys' growth [6][7][8]. While these findings demonstrated that select marine bacteria were capable of killing other bacteria, it has been suggested that the relatively low frequency of killing that was observed may have been due to the common use of non-marine a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 bacteria as model prey [9]. Later, experiments that used more ecologically relevant model prey (e.g. isolates from pelagic seawater, marine particles, and coral) found that killing occurred in > 50% of the competition assays [9,10]. Interestingly, these studies also showed that some of the model prey were able to survive the challenge against select predatory bacteria that had killed other bacteria, suggesting that those surviving prey possessed defense mechanisms [9,10].
In addition to chemical-mediated bacterial antagonism, marine bacteria possess and use a variety of contact-dependent killing mechanisms [11][12][13]. In this work, we focused on one such mechanism that is well-characterized and carried by many gram-negative bacteria, the type VI secretion system (T6SS) [12,14]. The T6SS is a nanomachine that is capable of killing eukaryotic and bacterial prey by directly injecting toxic effector proteins into them, which then carry out a variety of lethal functions [15][16][17][18]. Generally, the needle-like apparatus is assembled in stages, and once complete, it resembles an inverted bacteriophage tailspike [19]. Upon assembly initiation, a transmembrane baseplate is formed to anchor the system to the cell envelope [20]. VgrG and PAAR-domain-containing effector proteins are then recruited to the baseplate to form a needle-like tip and serve as the nucleation site for the formation of an Hcp protein tube [21][22][23]. A sheath comprised of VipA and VipB subunits then assembles around the tube and when the sheath contracts the Hcp/VgrG/PAAR complex is propelled into adjacent target cells [24][25][26][27]. Assisting with the extracellular secretion of these effector molecules are the proteins VasK and VasF, which are believed to be associated with the membrane-associated complexes [28]. The ATPase ClpV then disassembles the sheath, and possibly the entire apparatus, within seconds after "firing" the T6SS [29]. Bacteria carrying a functional T6SS exhibit a remarkable ability to efficiently kill their bacterial prey, which can lead to the displacement of host associated commensals [30], intraspecific competition during host colonization [31], community phase separation [32], and possibly intraguild predation [33]. Conversely, some bacteria have developed mechanisms to resist T6SS attack. For example, a recent study demonstrated that the production of exopolysaccharide (EPS) by Vibrio cholerae can act as a unidirectional barrier to protect itself from T6SS-mediated predators [34]. Furthermore, it has been documented that some bacteria possess immunity genes against various effectors, which is also how bacterial predators protect themselves against their own T6SS effectors [35][36][37]. In addition to passive resistance mechanisms, bacteria such as P. aeruginosa can sense exogenous T6SS attacks and retaliate with a T6SS of their own [38].
Despite the growing number of T6SS studies, relatively little is known about the effectiveness of T6SS deployment against different marine bacteria. It was found that the marine bacterium and human pathogen, Vibrio cholerae, is capable of using its T6SS to kill species such as V. communis, V. harveyi, Pseudoalteromonas phenolica, and Aeromonas sp. [39,40]. Other Vibrio species, such as V. parahaemolyticus, V. alginolyticus, and V. fischeri were also shown to have functional T6SSs, however the known scope of their marine prey is restricted to three Vibrio species (V. cholerae, V. natriegens, and select strains of V. fischeri) due to a limited number of studies [31,41,42]. We considered that further exploration into the range of marine bacteria that are susceptible to the T6SS should increase our understanding of the types of bacteria that a specific T6SS can kill, while also helping to inform microbial ecologists on select types of bacteria, and ultimately the mechanisms, that provide resistance to T6SS attack. Such knowledge may prove useful in understanding marine microbial community dynamics and has already been posited to be an important consideration for the development of antimicrobials and probiotics [34]. Here, we challenged a number of phylogenetically diverse marine bacterial types, including members of the phyla Proteobacteria and Bacteroidetes against T6SS attack from V. cholerae strain 2740-80. The presented results are broadly discussed in the context of marine microbial ecology, which includes implications for human health, aquaculture, and coral disease research.

Bacterial strains
The phylogeny, description, and source of each isolate that was used in the challenge assays are contained in Tables 1 and 2. Prior to our experiments, each of the marine bacterial challengers listed in Table 2 was plated onto rifampicin containing media to generate spontaneous rifampicin mutants (R r ). Single R r colonies for each isolate were picked, streaked purified, and confirmed resistant to rifampicin before storage in 25% glycerol at -80˚C. For challenge assays, the isolates were grown with Zobell 2216E at 25˚C. Autoclaved Zobell 2216E media was prepared by amending 0.22 μm-filtered seawater with 5g of peptone and 1g of yeast extract liter -1 , while plates contained an additional 15g of agar liter -1 [43]. The following concentrations of antibiotics were used where appropriate: streptomycin, 100 μg/ml; rifampicin, 50 μg/ml; ampicillin 100 μg/ml (Sigma-Aldrich; St. Louis, MO, USA).
For mutagenesis, V. coralliilyticus strains were grown in a modified glycerol artificial seawater (GASW) media supplemented with 50 mM Tris-Base (Sigma-Aldrich) (GASW-Tris) and the pH adjusted to 8.3 with HCl prior to autoclaving to prevent acidification of the media and incubated at 27˚C [55], unless otherwise stated. For solid media, 15 g/l of agar (Teknova; Hollister, CA, USA) was added prior to autoclaving. All E. coli strains were grown in LB-Miller at 37˚C, unless otherwise stated. Antibiotics for selection with E. coli were used at the following concentrations unless otherwise stated: kanamycin, 50 μg/ml; streptomycin, 25 μg/ml; spectinomycin, 50 μg/ml; and chloramphenicol, 15 μg/ml (Sigma-Aldrich). Antibiotics for selection with V. coralliilyticus were used at the following concentrations unless otherwise stated: ampicillin, 200 μg/ml; streptomycin, 50 μg/ml; spectinomycin, 100 μg/ml; and chloramphenicol,  . Bacterial cultures were washed with either ASW (GASW lacking glycerol, tryptone, or yeast extract) or phosphate buffered saline (PBS) for Vibrio and E. coli strains, respectively. Plasmid construction. All of the plasmids that were used are listed in (S1 Table), and the DNA oligonucleotide primers are listed in (S2 Table). The plasmid pBU226 is a suicide vector used to create a clean deletion of the vtpR homolog in OCN008 except for the first and last 18 nucleotides. Genomic DNA from OCN008 was used as template for PCR with the primer pairs 008-vtpR-up-EcoRI-F and Vcor-vtpR-up-OEX-R and Vcor-vtpR-down-OEX-F and 008-vtpR-down-XbaI-R to amplify regions up-and downstream of vtpR, respectively. The resulting PCR product was cloned as an EcoR1/XbaI fragment into the same sites in pSW4426T to create pBU226. Unless otherwise stated, all suicide plasmids were screened using PCR and Sanger sequencing using the primer pair pSW4426T-MCS-F and pSW4426T-MCS-R to confirm successful cloning.
The plasmid pBU247 is a suicide vector used to create a clean deletion of the vasK homolog in OCN008 except for the first and last 18 nucleotides. OCN008 genomic DNA was used as template for PCR with the primer pairs 008vasK-up-EcoR1-F and 008vasK-up-OEX-R and 008vasK-down-OEX-F and 008vasK-down-XbaI-R. The resulting PCR product was cloned as an EcoR1/XbaI fragment into the same sites in pSW4426T to create pBU247. The plasmid pBU266 is a suicide vector used to create a clean deletion of the vtpA homolog in OCN008 except for the first and last 18 nucleotides. OCN008 genomic DNA was used as template for PCR with the primer pairs vtpA-up-EcoRI-F and vtpA-up-OEX-R and vtpAdown-OEX-F and vtpA-down-SpeI-R. The resulting PCR product was cloned as an EcoR1/ SpeI fragment into the same sites in pSW4426T to create pBU266.
The plasmid pBU267 is a suicide vector used to create a clean deletion of the vtpB homolog in OCN008 except for the first and last 18 nucleotides. OCN008 genomic DNA was used as template for PCR with the primer pairs vtpB-up-SpeI-F and vtpB-up-OEX-R and vtpB-down-OEX-F and vtpA-down-SpeI-R. The resulting PCR product was cloned as a SpeI fragment into the XbaI site in pSW4426T that had been previously dephosphorylated with FastAP Thermosensitive Alkaline Phosphatase (Thermo Fisher Scientific) to create pBU266.
The plasmid pBU270 is a replicative vector used to express a wild type copy of vasK to complement the OCN008 vasK mutant. OCN008 genomic DNA was used as template for PCR with the primer pair vasK-XbaI-F and vasK-XbaI-R. The resulting PCR product was cloned as an XbaI fragment into the same site in pBU246 that had been previously dephosphorylated to create pBU270.
The plasmid pBU271 is a replicative vector used to express a wild type copy of vtpR to complement the OCN008 vtpR mutant. OCN008 genomic DNA was used as template for PCR with the primer pair 008-vtpR-SacI-F and 008-vtpR-XbaI-R. The resulting PCR product was cloned as a SacI/XbaI fragment into the same sites in pBU246 to create pBU271.
Mutant creation. All V. coralliilyticus suicide vectors were introduced using tri-parental conjugations with E. coli as previously described [55]. Donor and recipient strains were grown overnight with the appropriate antibiotics and DAP or DT as required (Sigma-Aldrich). Overnight cultures were diluted 1:1000 in fresh culture medium without antibiotics, grown to an optical density measured at 600 nm (OD 600 ) of 0.4, and then one ml washed three times with either ASW or PBS for Vibrio or E. coli strains, respectively. The strains were then combined, resuspended in ASW to a total volume of 50 μl, and spotted onto GASW-DEX plates supplemented with DAP and DT. Conjugation spots were incubated at 30˚C for 15 h before being resuspended in ASW, washed three times with ASW, diluted, and plated onto GASW-DEX supplemented with chloramphenicol, but lacking DAP or DT, at 27˚C. Chloramphenicolresistant colonies, were streaked for isolation on GASW-DEX with spectinomycin and streptomycin, the colonies were then screened for the presence of the suicide vector integrated into the chromosome using colony PCR and the primers pSW4499-cat-F and pSW4499-oriT-R. Colonies of Vibrio with the integrated plasmid were grown for 15 h in GASW-DEX broth. Cultures were washed with ASW three times, diluted, and plated onto GASW-ARA to isolate mutants with a clean deletion of the target gene. Mutants were confirmed using PCR and primers specific to the gene being mutated.
Challenge assays. Bacterial isolates were grown in liquid Zobell 2216E media overnight, washed, diluted 1:10 into fresh media, and grown for approximately 3 h. The cultures were then concentrated to an OD 600 of 10 via centrifugation at 8,600 x g for 5 minutes. Predator and challenger were mixed 1:1 (v:v; 10 μl total) and 5 μl aliquots of the co-cultures were spotted onto Zobell 2216E agar. We note that the starting colony forming units (CFUs) for each V. cholerae strain was~1.3x10 7 mL -1 in the competition assays. Starting CFUs were not determined for the other isolates. Challenge assays and V. cholerae monocultures (controls) were incubated for 4 h at 25˚C. The cells were then re-suspended, serial-diluted, and plated onto antibiotic selection media to recover and enumerate the surviving predator and challenger. Each challenge was independently repeated three times (biological replicates, n = 3), and each biological replicate consisted of three technical replicates. To ensure that our results were consistent and that the data interpretation was standardized, each bacterial challenger was screened against the same batch-culture of the predator. Statistical difference between the mean +/-SD of treatments was determined by two-tailed t-tests using GraphPad Prism version 7.0 (GraphPad Software, Inc.) and is described within each figure legend.
Biofilm assays. The ability of the V. coralliilyticus strains to produce a biofilm was measured using a modified crystal violet assay [56]. Cultures of V. coralliilyticus were initially grown overnight (approximately 15 h) in GASW-Tris. The cultures were then diluted 1:1000 into fresh media. In a 24-well plate, one ml aliquots of the diluted cultures were placed to each well (four replicates per strain). The plates were then incubated in a humidified incubator at 28˚C for 48 h. After incubation, the liquid cultures were then carefully aspirated using a pipette while being careful not to disrupt the sides of the wells. One ml of ASW was then added to each well and then removed via a pipette. This washing process was repeated two more times. Into each well, one ml of a 0.1% crystal violet solution was then added and incubated at room temperature for 15 min. The crystal violet solution was then poured off and the wells were again washed three times with ASW. The plates were then dried overnight with their lids off and inverted. After drying, one ml of a 30% acetic acid solution was added to each well, incubated at room temperature for 15 min, and then 500 μl of each well was transferred to a new 24-well plate. The absorbance of each well was measured at 550 nm in a plate reader (Epoch Microplate Spectrophotometer). One ml aliquots of sterile GASW-Tris processed in an identical manner as the bacterial cultures served as the blank.

Select Ɣ-Proteobacteria exhibited susceptibility to T6SS attack by V. cholerae
To test the efficacy of V. cholerae T6SS deployment against marine bacteria, we challenged a suite of marine isolates from different environmental and phylogenetic backgrounds ( Table 2) against V. cholerae with an active T6SS (T6SS + ), or its isogenic T6SS knockout mutant (T6SS -) derivative that was created and confirmed in a previous study (see Table 1). Colony forming unit recoveries were reduced by~90% for eight out of the 15 isolates in the challenges against T6SS + V. cholerae in comparison to the challenges against T6SS -V. cholerae (p < 0.01, twotailed t-test), indicating that those isolates were susceptible to T6SS attack (Fig 1B). The eight susceptible isolates were all Ɣ-Proteobacteria, including three members of the order Vibrionales (V. harveyi, Vibrio sp. SWAT-3, and V. shilonii), a bacterial group that has been previously reported to contain marine species that are sensitive to V. cholerae's T6SS [39]. The other susceptible isolates were from the orders Alteromonadales (Alteromonas Alt-SIO and Pseudoalteromonas flavipulchra), Oceanospirillales (Halomonas sp. 73), and Pseudomonadales (Pseudomonas sp. and Psychrobacter aquimaris). To our knowledge, this is the first report of marine bacterial susceptibility to a T6SS from the three aforementioned orders.
Surprisingly, four of the isolates were killed or inhibited when challenged against both T6SS + and T6SS -V. cholerae. These isolates were from the phyla Bacteroidetes (Aequorivita sp. 97 and Roseivirga sp. 121) and Ɣ-Proteobacteria (P. flavipulchra and Pseudomonoas sp. 2) (Fig 1B). Importantly, we note that these strains appear to have suffered no mortality when grown in monoculture under identical conditions, and that the monoculture recoveries werẽ 90% higher for each of the four strains in comparison to their recovery after the challenge against T6SS + or T6SS -V. cholerae (p < 0.01, two-tailed t-test; S1 Fig). While the mechanisms that led to their significant CFU reduction when co-cultured with V. cholerae were not further explored here, we offer several scenarios that might explain our observations: (1) the challenged isolates grew slower in co-culture, (2) the challenged isolates were outcompeted for resources, and/or (3) the challenged isolates were killed, either by toxic byproducts of metabolism or an alternative inhibitory mechanism used by V. cholerae.
Interestingly, we also observed that when T6SS + and T6SS -V. cholerae were each challenged against Pseudoalteromonas flavipulchra it resulted in death or inhibition for both V. cholerae strains (no CFUs were recovered, Fig 1A). These results may be explained in part by P. flavipulchra's highly antagonistic nature which has been demonstrated to inhibit the growth of a number of marine bacteria via the release of inhibitory chemical(s) [10]. This species is also known to produce L-amino acid (lysine or glycine) oxidases that are capable of hydrolyzing amino acids present within cells or in the growth media to produce hydrogen peroxide [57]. These enzymes are bactericidal to a wide range of isolates and can be autotoxic [58][59][60]. Although P. flavipulchra achieved a CFU recovery of~1.5 x 10 7 mL -1 after 4 h monoculture incubation (S1 Fig), it is possible that production of these toxic compounds could be triggered by the presence of V. cholerae when grown in co-culture. Such scenarios, in combination, or with any exacerbating effects that V. cholerae may exert in the co-culture, could explain the observed loss of CFUs for both competing species.

Several isolates were resistant to V. cholerae's T6SS
Four out of the 15 challengers co-existed with V. cholerae, as these isolates went unaffected by V. cholerae's T6SS (Fig 1B) and did not kill either T6SS + or T6SS -V. cholerae in their respective assays (Fig 1A). Two of the co-existing isolates were from the phylum Bacteroidetes (Flavobacteria sp. BBFL7 and Salgentibacter sp. 1). Interestingly, some members of the Bacteroidetes have been shown to exhibit immunity against T6SS effector proteins [35]. This is relevant to marine microbial ecology since Bacteroidetes are commonly found as the predominant taxa on bacteria-rich marine particles [61]. Resistance to contact-dependent killing mechanisms, such as the T6SS, may help enable these taxa to colonize and proliferate in such environments. The other isolates that we found to co-exist with T6SS + V. cholerae were two closely related Ɣ-Proteobacteria (Pseudoalteromonas Tw7 and Pseudoalteromonas Tw2), which were evidently resistant or immune to V. cholerae's T6SS (Fig 1B). It is also noteworthy that CFU recovery for T6SS -V. cholerae was significantly reduced after the competition with Pseudoalteromonas Tw2 in comparison to the recovery of T6SS + V. cholerae (p < 0.01, two-tailed t-test; Fig 1A). Overall, of the 15 isolates tested, only V. coralliilyticus displayed the ability to resist V. cholerae's T6SS (Fig 1B) and to kill both T6SS + and T6SS -V. cholerae (p < 0.01, two-tailed t-test; Fig 1A) (further discussed below).
Collectively, our results demonstrate that marine bacteria from a range of different taxa were susceptible to the T6SS of V. cholerae 2740-80, and that conversely, a number of taxa were resistant to its T6SS. We have begun looking into the resistance mechanisms that were employed by the isolates in our study, starting with V. coralliilyticus, an important coral and oyster pathogen [46,48,[62][63][64][65]. It has been suggested that V. coralliilyticus is capable of altering a susceptible coral's microflora that is thought to protect their host from infection [66], however, no mechanisms have been proposed for how the pathogen accomplishes this. Similarly, V. coralliilyticus is able to dominate the bacterial communities within shellfish hatcheries,  ) were challenged against marine bacterial isolates in competition assays. V. cholerae strains were also grown in monoculture to serve as controls. Error bars represent the mean ±SD of three biological replicates. (A) Recovered CFUs mL -1 after each challenge assay are shown side-by-side for T6SS + and T6SS -V. cholerae. Asterisks denote statistically significant differences (two-tailed t-test) between the CFUs recovered for the indicated V. cholerae strain after the challenge assay in comparison to its recovery when grown in monoculture (far right of graph). (B) Brackets indicate a two-tailed t-test implemented to determine statistically significant differences between recovered CFUs mL -1 for each challenged isolate after its competition assay against either T6SS + or T6SS -V. cholerae. https://doi.org/10.1371/journal.pone.0227864.g001 Resilience and susceptibility of marine bacteria to T6SS attack by Vibrio cholerae and Vibrio coralliilyticus suggesting an effective mechanism for competition [67]. These observations could be explained in part by this pathogen's ability to defend itself against other bacteria, or to kill other bacteria, as we found in the challenge against V. cholerae.
V. coralliilyticus evidently killed V. cholerae by using its own T6SS. We hypothesized that V. coralliilyticus may have (1) been intrinsically resistant to the T6SS attack by V. cholerae, (2) killed V. cholerae before itself was attacked, or (3) survived due to a combination of both scenarios. For the first hypothesis, we considered that the well-characterized proteolytic activity of V. coralliilyticus [63,65,[68][69][70] might provide resistance to V. cholerae's T6SS by degrading the T6SS apparatus or effector proteins, or by killing V. cholerae directly. To test this, knockout mutants were created using V. coralliilyticus strain OCN008 which had clean deletions of the quorum sensing regulator vtpR, which regulates protease activity in this species [71], as well as the protease-encoding genes vtpA and vtpB individually and in combination (vtpAB). These four mutant strains were challenged against T6SS + and T6SS -V. cholerae in competition assays. We found that the ΔvtpR mutant recovery was reduced by nearly one logfold in the challenge against T6SS + V. cholerae in comparison to the challenge against T6SS -V. cholerae (p < 0.001, two-tailed t-test, Fig 2B) and that it was unable to kill V. cholerae (Fig 2A). However, all three V. coralliilyticus protease mutants (ΔvtpA, ΔvtpB, and ΔvtpAB) resisted V. cholerae's T6SS (Fig 2B) and retained their ability to kill V. cholerae at levels equivalent to the wild-type strain (p < 0.0001, two-tailed t-test; Fig 2A). Therefore, we concluded that while VtpR did play a role in the success of V. coralliilyticus survival against T6SS + V. cholerae, the tested proteases were unlikely to be the mechanism that protected V. coralliilyticus from T6SS attack and were not responsible for the observed killing of V. cholerae.
As a homolog of the V. cholerae quorum sensing regulator HapR, VtpR is believed to regulate a wide range of physiological functions [71]. Recently, Strutzmann and Blokesch (2016) reported that mutations that inactivated HapR resulted in reduced T6SS activity for V. cholerae [72]. We considered that if V. coralliilyticus carried a functional T6SS that was regulated in part by VtpR, then the deletion of vtpR in OCN008 may have diminished or eliminated T6SS expression in our experiments, explaining our observation that the ΔvtpR strain was unable to kill V. cholerae. Our hypothesis that V. coralliilyticus had a T6SS was partially supported by Kimes et al. (2011) who previously observed needle-like structures within V. coralliilyticus and found upregulated expression of predicted T6SS-associated proteins at temperatures that correlated with increased virulence [73]. However, leading up to this study it was unknown if V. coralliilyticus had a functional T6SS that could kill microbial prey.
To investigate if V. coralliilyticus was using a T6SS in the competition assays, we created a V. coralliilyticus vasK deletion mutant (T6SS -) and challenged it against T6SS + or T6SS -V. cholerae. The ΔvasK mutant had a deletion in a gene predicted to encode a homolog of VasK, which is essential for V. cholerae T6SS function [28]. In accordance with our hypothesis, we found that both T6SS + and T6SS -V. cholerae survived the challenge assays (Fig 3A), demonstrating that the killing of V. cholerae by V. coralliilyticus did require the vasK gene. Moreover, the T6SS -V. coralliilyticus mutant was susceptible to T6SS attack by V. cholerae (p < 0.0001, two-tailed t-test ; Fig 3b), demonstrating that V. coralliilyticus was not inherently resistant to the V. cholerae T6SS. Genetic complementation of the T6SS -V. coralliilyticus strain restored V. coralliilyticus protease-mutant challenge assays. V. coralliilyticus OCN008, protease-mutant derivatives (ΔvtpA, ΔvtpB, and ΔvtpAB), and the ΔvtpR mutant were challenged against T6SS + V. cholerae (black bars) or T6SS -V. cholerae (grey bars). V. cholerae strains were also grown in monoculture to serve as controls. Error bars represent the mean ±SD of three biological replicates. (A) V. cholerae CFU recovery. Asterisks denote statistically significant differences (two-tailed t-test) between the CFUs recovered for the indicated V. cholerae strain after the challenge assay in comparison to its recovery when grown in monoculture (far right of graph). (B) Brackets indicate a two-tailed t-test implemented to determine statistically significant differences between recovered CFUs mL -1 for each tested V. coralliilyticus isolate after its competition assay against either T6SS + or T6SS -V. cholerae.
https://doi.org/10.1371/journal.pone.0227864.g002 the mutants' ability to kill V. cholerae (p < 0.0001, two-tailed t-test; Fig 3A) and its apparent resistance to T6SS-mediated attack (Fig 3B). Thus, the inability of the T6SS -V. coralliilyticus strain to kill V. cholerae, in combination with the mutant's susceptibility to T6SS-mediated killing by V. cholerae, strongly suggests that V. coralliilyticus used its own T6SS to kill V. cholerae in the challenge. Therefore, V. coralliilyticus evidently survived by winning in the 'quick draw', or by striking more effectively, as opposed to being resistant to attack. This contrasts with the previously described tit-for-tat interactions between V. cholerae and P. aeruginosa, in which, P. aeruginosa is described as intrinsically resistant to V. cholerae T6SS-mediated killing, while utilizing its own T6SS only in response to bacterial aggression [38]. Furthermore, given that the T6SS -V. coralliilyticus mutant was found to be susceptible to V. cholerae's T6SS, we were able to rule out the hypothesis that V. coralliilyticus employed other natural resistance mechanisms such as immunity to the toxic effector proteins or protective exopolysaccharide (EPS) "armor" that have been previously described [35,36]. Interestingly, the ΔvtpR strain, which was susceptible to T6SS + V. cholerae, was found to produce more EPS in comparison to the wild-type strain (p < 0.0001, Tukey's multiple comparisons test; S2 Fig) further suggesting that EPS production was not the protective mechanism for V. coralliilyticus in our study. In all, these results are the first evidence for a functional V. coralliilyticus T6SS, which we found to be effective at killing V. cholerae and required for resistance to T6SS-mediated killing by V. cholerae.
V. coralliilyticus T6SS has implications for coral and shellfish health. Intrigued by the ability of V. coralliilyticus OCN008 to resist V. cholerae's T6SS and to kill V. cholerae with its own T6SS, we conducted further experiments to determine if these characteristics were strainspecific. Four additional V. coralliilyticus strains including known coral and shellfish pathogens (OCN014, RE98, RE22, and BAA-450; Table 1) were challenged against T6SS + and T6SS -V. cholerae. Consistent with our initial result, we found that all four of the V. coralliilyticus strains were not affected by T6SS + V. cholerae ( Fig 4B) and that the survival of both T6SS + and T6SS -V. cholerae was reduced by > 99% (p < 0.0001, two-tailed t-test) by all of the tested V. coralliilyticus strains including OCN008, which was run alongside them ( Fig 4A). This suggested the presence of conserved mechanisms shared between the strains that offered protection to V. coralliilyticus and enabled each of the strains to kill V. cholerae. We surmise that the four V. coralliilyticus strains (OCN014, RE22, RE98, and BAA-450) that killed T6SS + and T6SS -V. cholerae (Fig 4A), like strain OCN008, carried a functional T6SS. While not explicitly tested here, that hypothesis is supported by predicted protein homology. Proteins required for T6SS functionality in V. cholerae shared 24-69% amino acid homology with predicted proteins in strain OCN008 (Table 3) and these proteins were present in the four other V. coralliilyticus strains (sharing 99-100% amino acid homology; Table 4).
These results could have a large impact on the understanding and treatment of coral and shellfish health. For example, strains OCN008, OCN014, and BAA-450 have been described as etiological agents of disease for multiple genera of coral [46,63,65], and strains RE98 and RE22 have been implicated in mass shellfish larvae mortalities [48,62,64]. We propose that the T6SS of V. coralliilyticus could be an important mechanism for the displacement of, and cholerae (grey bars). V. cholerae strains were also grown in monoculture to serve as controls. Error bars represent the mean ±SD of three biological replicates. (a) V. cholerae CFU recovery. Recovered CFUs mL -1 for V. cholerae strains after the challenge against T6SS -V. coralliilyticus were compared to their respective recovery when grown in monoculture (far right of graph) or to their recovery after the challenge against T6SS -/ pBU270 V. coralliilyticus. (b) V. coralliilyticus CFU recovery. Recovered CFUs for T6SS -V. coralliilyticus after the challenge against T6SS + V. cholerae was compared to the recovered CFUs when challenged against T6SS -V. cholerae. Recovered CFUs for T6SS -V. coralliilyticus after the challenge against T6SS + or T6SS -V. cholerae was also compared to the recovery of T6SS -/pBU270 V. coralliilyticus when challenged against T6SS + or T6SS -V. cholerae. Brackets indicate a two-tailed t-test.
https://doi.org/10.1371/journal.pone.0227864.g003 protection against host-associated bacteria, as it attempts to colonize potential hosts. In contrast to the displacement of the host microflora, V. coralliilyticus might also use the T6SS to attack other host-associated organisms or the host's cells directly. Studies have suggested that during infections some strains of V. coralliilyticus kill the photosynthetic algal symbionts within coral cells (Symbiodinium spp.), resulting in coral bleaching [65,68]. Given that the V. cholerae T6SS has been shown to kill eukaryotic organisms, such as the amoeba Dictyostelium discoideum [14], it is tempting to speculate that the V. coralliilyticus T6SS may be capable of killing Symbiodinium. Moreover, while it is still unclear if V. coralliilyticus acts as an intracellular pathogen, a recent study has demonstrated that V. coralliilyticus can end up within coral cells and vesicles during infection [74]. It might be possible for V. coralliilyticus to respond in these environments with its T6SS to escape host vesicles, similar to how V. cholerae defends against predation by D. discoideum phagocytosis. Such mechanisms have been reported for the intracellular pathogens Francisella tularensis and Burkholderia pseudomallei, which are able to use their T6SS to escape vesicles and macrophages, or to spread from cell to cell [75,76]. We suggest that further investigations are warranted to better understand the role of T6SS deployment by V. coralliilyticus in coral and shellfish pathogenesis, which may aid in the protection of these environmentally and economically-important organisms.

Conclusion
The bacterial type VI secretion system is present in various gram-negative bacteria and is capable of killing microbial prey. Here, we challenged a diverse set of marine bacterial isolates against T6SS + V. cholerae and found high mortality rates for select members of the genus Vibrio. We also provided the first documentation for T6SS-mediated killing of marine Alteromonadales, Oceanospirillales, and Pseudomonadales. Additionally, a number of isolates from the Bacteroidetes and Ɣ-Proteobacteria phyla were found to be resistant to V. cholerae's T6SS, including the important marine pathogen, V. coralliilyticus. All five of the tested V. Five V. coralliilyticus strains (OCN008, RE014, RE98, RE22, and BAA-450) were challenged against T6SS + V. cholerae (black bars) and T6SS -V. cholerae (grey bars). V. cholerae strains were also grown in monoculture to serve as controls. Error bars represent the mean ±SD of three biological replicates. (a) V. cholerae CFU recovery. Asterisks denote statistically significant differences (two-tailed t-test) between the CFUs recovered for the indicated V. cholerae strain after the challenge assay in comparison to its recovery when grown in monoculture (far right of graph). (b) V. coralliilyticus CFU recovery. Brackets indicate a two-tailed t-test implemented to determine statistically significant differences between recovered CFUs mL -1 for each tested V. coralliilyticus isolate after its competition assay with either T6SS + or T6SS -V. cholerae.
https://doi.org/10.1371/journal.pone.0227864.g004 coralliilyticus strains killed V. cholerae, presumably via the use of their own T6SS. We propose that bacterial susceptibility and resistance to contact-dependent killing mechanisms, such as the T6SS, might be important for the structuring of marine microbial communities in high bacterial density environments. Future work will be required to test the ecological impacts of such mechanisms in situ which should be possible by using a molecular-based approach in combination with direct imaging techniques.

S1 Fig. Select marine isolates grown in monoculture.
Colony forming unit (CFU) recoveries of several marine isolates challenged against T6SS + V. cholerae or T6SS -V. cholerae (from Fig  1B) are compared to the isolates' recovery when grown in monoculture (two-tailed t-test). The monocultures were grown as a follow-up experiment on separate days from the challenge assays but conducted in accordance with the same protocol that was used for the challenge assays.
(TIFF) S2 Fig. V. coralliilyticus OCN008 biofilm assay. A crystal violet assay was conducted to measure the amount of biofilm produced by wild type V. coralliilyticus and the ΔvtpR and ΔvasK strains. Blank = marine broth with no V. coralliilyticus cells. Higher optical density (OD) values at 550nm indicates more biofilm (extracellular polysaccharide) production. Statistical differences between treatments are denoted by different letters (ordinary one-way ANOVA, α = 0.05; p < 0.0001, Tukey's multiple comparison test, a, b, c). (TIFF) S1