bioRxiv preprint doi: https://doi.org/10.1101/683029; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

1

2

3 Diversity in natural transformation frequencies and regulation across species

4

5

6 Chelsea A. Simpson1, Ram Podicheti2, Douglas B. Rusch2, Ankur B. Dalia1, Julia C. van

7 Kessel1*

8

9

10 1. Department of Biology, Indiana University, Bloomington, IN 47405

11 2. Center for Genomics and Bioinformatics, Indiana University, Bloomington, IN 47405

12

13

14

15 * Corresponding author:

16 Email: [email protected]

17 Telephone: 812-856-2235

18 Fax: 812-856-5710

19

20 Running title: Natural transformation in

21

22 Keywords: competence, natural transformation, vibrio

23 bioRxiv preprint doi: https://doi.org/10.1101/683029; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

24 Abstract

25 In marine Vibrio species, chitin-induced natural transformation enables to take up DNA

26 from the external environment and integrate it into their genome via homologous recombination.

27 Expression of the master competence regulator TfoX bypasses the need for chitin induction and

28 drives expression of the genes required for competence in several Vibrio species. Here, we

29 show that TfoX expression in two Vibrio campbellii strains, DS40M4 and NBRC 15631, enables

30 high frequencies of natural transformation. Conversely, transformation was not achieved in the

31 model quorum-sensing strain V. campbellii BB120 (previously classified as ).

32 Surprisingly, we find that is not required for transformation in V. campbellii

33 DS40M4. This result is in contrast to that requires the quorum-sensing regulator

34 HapR to activate the competence regulator QstR. However, similar to V. cholerae, QstR is

35 necessary for transformation in DS40M4. To investigate the difference in transformation

36 frequencies between BB120 and DS40M4, we used previously studied V. cholerae competence

37 genes to inform a comparative genomics analysis coupled with transcriptomics. BB120 encodes

38 homologs of all known competence genes, but most of these genes were not induced by ectopic

39 expression of TfoX, which likely accounts for the non-functional natural transformation in this

40 strain. Comparison of transformation frequencies among Vibrio species indicates a wide

41 disparity among even closely related strains, with Vibrio vulnificus having the lowest functional

42 transformation frequency. We show that ectopic expression of both TfoX and QstR is sufficient

43 to produce a significant increase in transformation frequency in Vibrio vulnificus.

44

45 Significance

46 Naturally transformable or competent bacteria are able to take up DNA from their environment,

47 a key method of horizontal gene transfer for acquisition of new DNA sequences. Our research

48 shows that Vibrio species that inhabit marine environments exhibit a wide diversity in natural

49 transformation capability ranging from non-transformable to high transformation rates in which bioRxiv preprint doi: https://doi.org/10.1101/683029; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

50 10% of cells measurably incorporate new DNA. We show that the role of regulatory systems

51 controlling the expression of competence genes (e.g., quorum sensing) is conserved among

52 closely related species but differs throughout the genus. Expression of two key transcription

53 factors, TfoX and QstR, are necessary and sufficient to stimulate high levels of transformation in

54 Vibrio campbellii and recover low rates of transformation in Vibrio vulnificus. bioRxiv preprint doi: https://doi.org/10.1101/683029; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

55 Introduction

56 Natural transformation is a physiological state in which bacteria are able to take up

57 extracellular DNA from the environment, transport it across the cell envelope, and integrate it

58 into their genome via homologous recombination. In marine Vibrio species, competence is

59 induced by growth on the chitinous exoskeletons of crustaceans (1, 2). This process has been

60 best studied in Vibrio cholerae (reviewed in (2)). Insoluble polysaccharide chitin is broken down

61 by secreted extracellular chitinases (3). Soluble chitin oligosaccharides ultimately induce

62 expression of the master competence regulator TfoX (4), which activates expression of

63 numerous components of the competence machinery required to take up extracellular DNA. In

64 addition to the chitin-sensing system, V. cholerae cells must also have a functional quorum-

65 sensing system for natural transformation to be successful (1). Quorum sensing, the process of

66 cell-cell communication, allows bacterial cells to respond to changes in population density and

67 alter gene expression (5). The quorum-sensing systems in Vibrio species rely on detection of

68 extracellular that are sensed by membrane-bound sensor kinases, which shuttle

69 phosphate to or away from the core response regulator LuxO at low cell density or high cell

70 density, respectively (5). At the end of this phosphorylation cascade at high cell density, the

71 master transcription factor called LuxR is activated, which controls expression of hundreds of

72 genes (6). LuxR is the name of the master transcription factor in Vibrio campbellii (previously

73 called Vibrio harveyi), whereas the homologs in other Vibrio species have different names:

74 HapR (V. cholerae), SmcR (Vibrio vulnificus), and OpaR (Vibrio parahaemolyticus) (7). In V.

75 cholerae, DhapR mutants are not competent, and this is due to HapR regulation of various

76 competence genes, including comEA, comEC, qstR, and dns (1, 8, 9). HapR directly activates

77 qstR (encoding a transcriptional regulator) and represses dns (encoding an extracellular DNase)

78 (8). QstR subsequently activates downstream genes required for DNA uptake and integration,

79 such as comEA, comEC, and comM (10). The requirement for HapR for competence can be

80 circumvented if qstR and tfoX expression are induced and the dns gene is deleted (10). bioRxiv preprint doi: https://doi.org/10.1101/683029; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

81 However, in wild-type V. cholerae, functional chitin-sensing and quorum-sensing systems are

82 required for competence.

83 Growth on chitin is sufficient to induce competence in several Vibrio species, including

84 V. cholerae, V. vulnificus, and V. parahaemolyticus (1, 11-13). Anecdotally, chitin-dependent

85 natural transformation has not been observed in other species. In these cases, it may be that

86 other environmental signals are required to induce competence or that laboratory conditions for

87 chitin sensing are not sufficient. Further, multiple studies have identified different transformation

88 capabilities among V. cholerae strains (14). However, because the chitin-sensing system drives

89 expression of the core competence regulator TfoX, overexpression of Tfox is sufficient to

90 bypass the requirement for chitin and induce competence in V. cholerae, V. vulnificus, V.

91 parahaemolyticus, and Vibrio natriegens (1, 12, 15-17).

92 The type strain of Vibrio campbellii called BB120 (or ATCC BAA-1116) is a model

93 system for studying quorum sensing in Vibrio species (18). This strain was historically called

94 Vibrio harveyi until a recent comparative genomics analysis reclassified it as V. campbellii (19).

95 However, while BB120 is highly studied, the competence of this strain and others in this species

96 was undetermined. Here, we show that the DS40M4 and NBRC 15631 environmental isolates

97 of V. campbellii are highly competent following TfoX overexpression, whereas the lab strain

98 BB120 and another environmental isolate called HY01 are not able to undergo natural

99 transformation either via chitin induction or TfoX overexpression. We compare multiple Vibrio

100 species and show that wide variation in transformation frequencies exists among strains in this

101 genus.

102

103 Results and Discussion

104

105 Induction of natural transformation via TfoX overexpression in V. campbellii DS4M04 and NBRC

106 15361 strains bioRxiv preprint doi: https://doi.org/10.1101/683029; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

107 Anecdotal reports and unpublished experiments have shown that the V. campbellii type

108 strain BB120 lacks the ability to undergo natural transformation. To formally test this, we

109 assayed for transformation of plasmid DNA into BB120 using both chitin-dependent and -

110 independent methods to induce competence (17, 20). For chitin-dependent transformations,

111 cells are incubated with powdered chitin from shrimp shells and DNA and plated with antibiotic

112 selection. For chitin-independent transformations, a plasmid expressing V. cholerae tfoX via an

113 IPTG-inducible promoter (pMMB67EH-tfoX-kanR, Table S2) is used to induce competence in

114 cells, and the cells are incubated with DNA prior to plating for antibiotic selection. When we

115 performed both types of transformation experiments with BB120, we did not recover antibiotic-

116 resistant colonies from either method, whereas high transformation frequencies were obtained

117 with the same plasmid DNA introduced into positive control strains (V. natriegens for chitin-

118 independent, Fig. S1A; V. cholerae for chitin-dependent, data not shown).

119 Because strains of V. cholerae exhibit differing natural transformation frequencies, we

120 hypothesized that the absence of competence in BB120 may not be indicative of competence in

121 other V. campbellii isolates. We therefore assayed natural transformation in three V. campbellii

122 environmental isolates with sequenced genomes: DS40M4, NBRC 15631, and HY01 (19, 21-

123 23). DS40M4 was isolated from the Atlantic Ocean near the west coast of Africa (24), HY01 was

124 isolated from a bioluminescent shrimp in Thailand (22), and NBRC 15631 (also called CAIM 519

125 and ATCC 25920) was isolated from seawater in Hawaii (19). These species cluster together in

126 the Harveyi clade in the Vibrio genus, with DS40M4 and NBRC 15631 being the most closely

127 related (Fig. 1A). Incubation on chitin was unable to induce cells to take up the plasmid DNA in

128 any of the V. campbellii strains (data not shown). However, both the DS40M4 and NBRC 15631

129 isolates were able to undergo transformation of plasmid DNA using the chitin-independent

130 method in which TfoX was ectopically expressed (Fig. S1A). Transformation was dependent on

131 TfoX expression because strains containing the empty vector control did not produce

132 transformants (Fig. S1A). bioRxiv preprint doi: https://doi.org/10.1101/683029; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

133 We next assayed transformation of linear DNA (referred to as transforming DNA, tDNA).

134 We generated linear recombination products targeting the luxO gene in each V. campbellii strain

135 containing an antibiotic resistance cassette and 3 kbp of DNA homologous to the regions

136 flanking luxO for each strain (DluxO::specR substrates). High transformation frequencies were

137 obtained for both DS4M04 and NBRC 15631 with their respective tDNAs that are dependent on

138 the inducible expression of tfoX (Fig. 1B). We tested various parameters for transformation,

139 including the number of cells in the reaction, the amount of time tDNA was incubated with cells

140 before outgrowth, and the amount of tDNA added to the reaction (Fig. S1B, S1C, S1D). Using

141 these optimized methods, we still did not observe transformation with either the BB120 strain or

142 the HY01 strain (Fig. 1B). We note that the DS40M4 strain is close to or as efficient as V.

143 natriegens at transformation with linear tDNA (Fig. 1B), which has thus far been reported as the

144 Vibrio strain displaying the highest transformation frequencies (17). The lack of transformation

145 observed in the highly pathogenic HY01 strain compared to the high frequencies observed in

146 the oceanic isolate DS40M4 underscores a possible distinction in conservation of competence

147 functions. As proposed in a recent study (10), it is possible that vibrios that are adapted to

148 specific niches, such as those of a host organism for pathogenic strains, might have lost the

149 ability to take up DNA if it were no longer beneficial. Conversely, natural transformation might be

150 more advantageous for strains existing free-living in the ocean.

151

152 Assessment of quorum-sensing phenotypes in V. campbellii strains

153 Because quorum sensing activates genes required for natural transformation in V.

154 cholerae, we wanted to determine whether DS40M4 or NBRC 15631 contain functional quorum-

155 sensing systems. Using chitin-independent natural transformation via tfoX induction, we

156 generated DluxO and DluxR mutations in both the DS40M4 and NBRC 15631 strain

157 backgrounds. In the model strain BB120, a DluxO mutant results in a constitutively expressed bioRxiv preprint doi: https://doi.org/10.1101/683029; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

158 master quorum-sensing transcription factor LuxR, producing high levels of and

159 mimicking a high cell density phenotype (25). Conversely, a DluxR mutant of BB120 is unable to

160 produce bioluminescence (26). We compared the phenotypes of wild-type, DluxO, and DluxR

161 strains for BB120, DS40M4, and NBRC 15361 by assessing expression of the quorum-sensing

162 bioluminescence genes, luxCDABE. We were unable to monitor bioluminescence because

163 DS40M4 and NBRC 15631 do not contain all of the luxCDABE bioluminescence genes like

164 BB120 (they only encode luxB homologs) and do not bioluminesce (data not shown). We

165 therefore used a GFP reporter plasmid in which the BB120 bioluminescence promoter

166 (PluxCDABE) is transcriptionally fused to gfp. In BB120, a DluxR strain exhibits a 16-fold decrease

167 in GFP expression compared to wild-type, whereas a DluxO mutant exhibits similar GFP levels

168 as wild-type (Fig. 2A). In the DS40M4 strains, we observed similar GFP expression to the

169 analogous BB120 strains (Fig. 2B). We complemented the DluxR deletion strain in DS40M4 and

170 showed that GFP expression was restored to wild-type levels (Fig. S2A).

171 The results with NBRC 15631 were different than BB120. The DluxO mutant exhibited

172 higher levels of GFP compared to the DluxR mutant, as expected (Fig. 2C). Curiously, the wild-

173 type NBRC 15361 strain consistently yielded low levels of GFP, even though the cells reached

174 similar densities as the wild-type strains of BB120 and DS40M4 (Fig. 2C). This phenotype

175 suggests that the NBRC 15631 cells do not produce and/or sense (s) but retain

176 functional luxO and luxR genes that are epistatic to autoinducer sensing. Strikingly, neither

177 DS40M4 nor NBRC 15631 encode a LuxM homolog, which synthesizes AI-1 in BB120. DS40M4

178 and NBRC 15631 both encode close homologs of BB120 CqsS and LuxPQ, which bind CAI-1

179 and AI-2, respectively (98-99% amino acid identity). Conversely, LuxN, which binds AI-1, is less

180 well conserved in DS40M4 and NBRC 15631 at 63-64% identity. There is high sequence

181 conservation in the C-terminal receiver domain portion of LuxN, whereas the N-terminal region

182 that binds autoinducers is not conserved with BB120. These data suggest that the DS40M4 and bioRxiv preprint doi: https://doi.org/10.1101/683029; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

183 NBRC 15631 LuxN histidine kinases may recognize a different autoinducer than BB120.

184 However, if this is the case, these two strains likely recognize the same molecule because the

185 LuxN proteins are 99% conserved between DS40M4 and NBRC 15631. To determine if NBRC

186 15631 responds to an autoinducer from another strain, we incubated NBRC 15631 cells in

187 media supplemented with cell-free supernatants from various Vibrio strains (including DS40M4).

188 The GFP levels in the supernatant-treated cells were low and not significantly different than

189 untreated NBRC 15631 cultures (data not shown). Importantly, we observed a high

190 transformation frequency in NBRC 15631 DluxO that was similar to DS40M4. Thus, we suspect

191 that there is a defect in regulation in the quorum-sensing pathway of NBRC 15631 that results in

192 a low-cell-density-like state but can be bypassed by deletion of luxO. In this state, low levels of

193 LuxR would be produced, which is likely the cause for the lower transformation frequencies in

194 wild-type NBRC 15631 compared to DS40M4. From these data, we conclude that DS40M4 and

195 NBRC 15631 have conserved LuxR and LuxO functions.

196

197 Quorum sensing is not required for transformation in V. campbellii DS40M4

198 In V. cholerae, HapR expression is required for natural transformation (8, 27). In the

199 absence of hapR, deletion of dns results in modest increases in transformation (28). A hapR

200 mutation can be bypassed by over-expression of qstR to produce transformants, though at a

201 reduced rate compared to wild-type (8). Combination of qstR expression and deletion of dns in a

202 hapR mutant results in maximal transformation frequencies (10). To determine if the luxR genes

203 of DS40M4 and NBRC 15631 are required for natural transformation, we assayed

204 transformation frequencies in DluxR and DluxO mutants compared to wild-type for each V.

205 campbellii strain using chitin-independent transformation. As a control, we performed the

206 analogous experiment in V. cholerae with DhapR and DluxO mutants. Strikingly, we observed

207 that a DluxR mutant in the DS40M4 strain is capable of natural transformation, though at a ~60- bioRxiv preprint doi: https://doi.org/10.1101/683029; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

208 fold reduced frequency compared to wild-type (Fig. 3A, 3B). The DS40M4 DluxO mutant

209 exhibited a similar transformation frequency as wild-type DS40M4, which has also been

210 observed in V. cholerae (1). This is due to maximal expression of LuxR/HapR in the DluxO

211 background. The DluxR NBRC 15631 mutant did not yield colonies above the limit of detection

212 (Fig. 3B). However, the DluxO NBRC 15631 strain had >100-fold higher levels of transformation

213 than the wild-type NBRC 15631 strain and was similar to the frequencies observed in the

214 DS40M4 DluxO strain (Fig. 3B). This result indicates that the presumed high levels of LuxR

215 present in the DluxO strain restore transformation frequencies in NBRC 15631. From these

216 data, we conclude that quorum-sensing control of competence via LuxR regulation is not

217 required for transformation in V. campbellii DS40M4.

218

219 Conservation of competence genes and gene expression between BB120 and DS40M4

220 To investigate the differences in transformation frequencies between the BB120 and

221 DS40M4 V. campbellii strains, we performed a comparative genomics analysis of the known V.

222 cholerae competence genes against the genes present in V. campbellii DS40M4 and BB120.

223 We generated a list of 47 V. cholerae genes based on published data supporting a role for the

224 gene product and/or Tn-Seq data suggesting that mutants lacking these genes exhibit differing

225 phenotypes in natural transformation assays. Our analyses show that BB120 and DS40M4 both

226 encode homologs of all known genes that play a role in competence in V. cholerae (Fig. 4A,

227 Table S4). Conservation of amino acid identity ranges from 25% to 95% with a median value of

228 73%. However, the vast majority of genes that have low amino acid conservation with V.

229 cholerae are still highly conserved between BB120 and DS40M4. For only two genes, pilA and

230 VC0860 (a minor pilin gene), there is low conservation among the three strains. For example,

231 the pilA gene in BB120 shares 44% amino acid identity with V. cholerae pilA and only 57%

232 identity with DS40M4 pilA (Fig. 4A). PilA sequences vary widely among environmental V. bioRxiv preprint doi: https://doi.org/10.1101/683029; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

233 cholerae strains (29), and it was recently shown that differences in PilA protein sequences

234 among strains enable V. cholerae cells to discriminate between each other, which leads to

235 decreased cell aggregation (30). Thus, it is unsurprising that the PilA sequence differs between

236 all three vibrio strains (Fig. 4A). However, the VC0860 minor pilin protein is required for

237 competence in V. cholerae (31), and thus the low conservation of this gene between BB120 and

238 DS40M4 may reduce or eliminate transformation.

239 Because it appears that BB120 encodes all known genes required for competence at

240 some degree of conservation, we next questioned whether the expression of these genes was

241 sufficient for transformation. We performed RNA-seq comparing gene expression in the

242 presence or absence of tfoX induction in both BB120 and DS40M4 strains and analyzed the

243 reads per kilobase of transcript per million mapped reads (RPKM). As expected, induction of

244 TfoX expression resulted in upregulation of the genes required for DNA uptake and integration

245 (comEC, comEA, comM, comF, and dprA) and pilus structure (pilABCD, pilMNOPQ, minor pilins

246 VC0857-0861) in DS40M4 (Fig. 4B). However, for most of these genes in BB120, there is no

247 significant change in expression +/- tfoX induction (Fig. 4B). Further, even if an increase in gene

248 expression was observed in BB120 with tfoX induction, it was modest compared to the large

249 changes in gene expression in DS40M4 with tfoX induction. Collectively, these data suggest

250 that expression levels of competence genes in BB120 are not sufficient for transformation. It is

251 also possible that the pilA and VC0860 homologs in BB120 are not functional, due to their lower

252 conservation with DS40M4 homologs.

253 Over-expression of V. cholerae tfoX in V. campbellii DS40M4 produced a similar

254 expression profile for the 47 competence genes to the profile produced in V. cholerae upon tfoX

255 induction (32). One exception to this is that we observed a ~10-fold increase in pilT expression

256 with tfoX induction, as compared to V. cholerae that showed a similar level of pilT expression in

257 the presence or absence of tfoX induction (32). Another exception is that tfoX induction in

258 DS40M4 resulted in a ~20-fold increase in cytR expression in DS40M4, whereas in a similar bioRxiv preprint doi: https://doi.org/10.1101/683029; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

259 experiment in V. cholerae, cytR expression is only increased ~4-fold (32). Although the reason

260 for these differences is unclear, it minimally indicates that there are slight differences in the TfoX

261 regulons of V. campbellii DSS40M4 and V. cholerae. Overall, our observation that DS40M4 up-

262 regulates >20 genes similar to V. cholerae in the presence of TfoX suggests high conservation

263 of the regulatory network controlling competence.

264 We noted that the fold-changes in the qstR and cytR transcript levels are larger in

265 DS40M4 +/- tfoX induction as opposed to BB120 +/- tfoX induction. This is important because

266 both QstR and CytR are transcriptional regulators; CytR is a transcription factor that positively

267 regulates competence genes, among others, in V. cholerae (33). The levels of cytR in BB120

268 with tfoX induction are similar to those in DS40M4 with tfoX induction, suggesting that low CytR

269 expression is not likely the cause for lack of transformation in BB120 (Fig. 4B). Conversely, the

270 levels of qstR in BB120 are lower than in DS40M4 under tfoX induction conditions (Fig. 4B).

271 Though we cannot compare RPKM values for genes between strains, by normalizing to other

272 genes such as recA of each strain, we can see that the levels of qstR are lower in BB120 (Fig.

273 4B). Because qstR is required for expression of essential competence genes (e.g., comEA,

274 comEC) in V. cholerae (10), the low levels of qstR expression in BB120 might be one reason for

275 its inability to take up DNA.

276

277 Expression of QstR in DS40M4 drives maximum levels of natural transformation in the absence

278 of LuxR

279 We next sought to examine the mechanism behind the observation that the DluxR strain

280 of DS40M4 retains transformation capability. We hypothesized that this may be due to either 1)

281 higher levels of qstR expression in DS40M4 than in V. cholerae (in the absence of LuxR/HapR),

282 or 2) poorly functional or poorly expressed dns in DS40M4 compared to V. cholerae. To formally

283 test whether increased levels of qstR would increase transformation frequencies, we cloned the

284 qstR gene from DS40M4 downstream of tfoX under control of an IPTG-inducible promoter to bioRxiv preprint doi: https://doi.org/10.1101/683029; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

285 form a synthetic operon: Ptac-tfoX-qstR (pCS32). We observed a ~20-fold increase in

286 transformation frequency in DS40M4 DluxR with the expression of qstR, which brought the level

287 in this background to the wild-type frequency (Fig. 5A), suggesting that Dns activity may not

288 play a role in transformation frequency as seen in V. cholerae. Deletion of qstR in DS40M4

289 abolishes transformation, much like in V. cholerae (Fig. 5B). These data indicate that QstR

290 expression is epistatic to LuxR. From these data, we conclude that TfoX and QstR are both

291 necessary and sufficient to initiate natural transformation in V. campbellii DS40M4 in the

292 absence of LuxR. Unfortunately, we did not observe any transformation in BB120 or HY01

293 under induction of both tfoX and qstR (data not shown).

294

295 Natural transformation frequencies vary widely across Vibrio species

296 The notable differences in transformation frequencies between strains of V. campbellii

297 led us to investigate the frequencies of natural transformation in other vibrios. Varied

298 transformation frequencies have been reported for other Vibrio species in the literature, with V.

299 natriegens having the highest level of transformation frequency (10, 12, 16, 17, 34). However,

300 several of these experiments were performed in different labs with different conditions, such as

301 different tfoX genes, tDNA homology lengths, tDNA quantity, media, and outgrowth time. To

302 formally assay frequencies of transformation with consistent experimental conditions, we tested

303 transformation using both chitin-dependent and –independent methods in multiple Vibrio

304 species: V. campbellii, V. parahaemolyticus, V. cholerae, and V. vulnificus. The tDNA substrates

305 were generated to target the gene encoding the LuxR/HapR homolog in each strain: luxR,

306 opaR, hapR, and smcR, respectively. The only difference in experimental conditions was the

307 use of media containing higher concentrations of salt required for viability in V. campbellii, V.

308 vulnificus, and V. parahaemolyticus. In our experiment, only V. cholerae was able to undergo

309 natural transformation when chitin was used to induce competence (Fig. S2B). This is in bioRxiv preprint doi: https://doi.org/10.1101/683029; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

310 contrast to other studies that have successfully used crab shell pieces to induce transformation

311 in V. vulnificus and V. parahaemolyticus (12, 13). However, another group was unable to induce

312 competence in V. parahaemolyticus with long-chain chitin polymers (16). Thus, it is important to

313 note that we only used chitin from shrimp shells and did not test the effect of soluble chitin

314 oligosaccharides or crab shells, which may produce better transformation frequencies.

315 When we stimulated competence by overexpressing TfoX (chitin-independent), all

316 strains except V. campbellii BB120 and HY01 were able to take up the tDNA and produce

317 recombinants (Fig. 6A). V. vulnificus exhibited the lowest transformation frequency, and V.

318 natriegens and V. campbellii DS40M4 have the highest frequencies (Fig. 1B, Fig. 6A). Similar to

319 a previously published comparison, V. parahaemolyticus had a similar transformation frequency

320 to V. cholerae (35). Because expression of both tfoX and qstR were sufficient to increase

321 transformation frequency in DS40M4 in the absence of luxR, we hypothesized that expression

322 of these two transcription factors might increase transformation frequency in other vibrios. We

323 introduced the same Ptac-tfoX-qstR plasmid pCS32 into each Vibrio strain tested above and

324 assayed for transformation frequency. We observed a >2-log increase in transformation in V.

325 vulnificus (Fig. 6A), though no significant increases in the other vibrios tested. We suspect that

326 ectopic expression of qstR is likely compensating for low levels of qstR expression in V.

327 vulnificus. Of note, this is a drastic improvement of transformation frequency in this species,

328 which is otherwise too low for multiplex genome editing by natural transformation (MuGENT)

329 (34). Our observed variations in transformation frequencies across Vibrio species may be due to

330 either differences in expression and/or function of Dns and/or varying levels of QstR in different

331 strains.

332

333 Quorum sensing is not required for transformation in V. parahaemolyticus

334 To determine if the lack of requirement for LuxR for competence extends beyond V.

335 campbellii, we tested transformation in V. vulnificus and V. parahaemolyticus in the presence or bioRxiv preprint doi: https://doi.org/10.1101/683029; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

336 absence of the LuxR-type quorum-sensing regulator in each species. Deletion of the smcR

337 gene in V. vulnificus abolished transformation, similar to what is observed in V. cholerae (Fig.

338 6B). However, deletion of the opaR gene in V. parahaemolyticus did not eliminate

339 transformation, but decreased it ~30-fold, similar to what is observed in V. campbellii DS40M4

340 (Fig. 6C, Fig. 3A). Minimally, these data show that the regulatory factors (e.g., QstR,

341 LuxR/HapR) controlling transformation are conserved in V. cholerae, V. parahaemolyticus, and

342 V. campbellii DS40M4, but the stringency of dependence on these factors varies among

343 species. These differences may also be reflective of their evolutionary relatedness; V.

344 parahaemolyticus and V. campbellii are more closely related to each other than to V. cholerae

345 and V. vulnificus (Fig. 1A). It may also reflect niche-specific requirements for DNA uptake that

346 differ in free-living strains compared to host-associated strains.

347

348 Concluding Remarks

349 In this study, we identified two strains of V. campbellii that are capable of natural

350 transformation in the presence of tfoX expression. We observed a wide diversity in

351 transformation frequencies, not only among campbellii strains but across Vibrio species, some

352 which are highly competent and some that are incapable of transformation under tested

353 laboratory conditions. Our results show that the connection between quorum sensing and

354 competence is linked but not dependent in some Vibrio strains. Unfortunately, we were

355 unsuccessful at restoring competence to V. campbellii BB120. It is likely that the limitations to

356 transformation in BB120 are multifactorial, including a potentially defective PilA and/or VC0860

357 minor pilin homolog, decreased gene expression of genes required for pilus assembly, DNA

358 uptake, and DNA integration upon induction of tfoX, and low expression of QstR. It is also not

359 clear why so many competence genes failed to induce upon TfoX expression, whereas this

360 method of inducing competence was highly successful in V. campbellii DS40M4 and NBRC

361 15631 and other Vibrio species. We postulate that either V. cholerae TfoX does not function bioRxiv preprint doi: https://doi.org/10.1101/683029; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

362 properly in BB120 or that BB120 lacks regulatory control of the core competence genes from

363 another transcription factor, possibly one that is not known in V. cholerae yet. The only BB120

364 gene known to be controlled by TfoX that responded with its induction was qstR, suggesting that

365 TfoX might function in BB120 but perhaps at a reduced capability. This is in accordance with the

366 finding that the V. fischeri tfoX is slightly decreased in the ability to cross-activate the qstR of V.

367 cholerae (35). One additional possibility is that there is a regulatory feedback loop that

368 downregulates transcription levels of structural genes in the absence of a functional pilus (i.e., if

369 the BB120 PilA is not functional), similar to the homeostatic regulation of flagellin in Bacillus

370 (36). Future experiments are required to dissect the broken pieces of the BB120 competence

371 regulatory and functional networks to synthetically generate competence in this model

372 organism.

373

374 Materials and Methods

375

376 Bacterial Strains and Media

377 All bacterial strains and plasmids used are listed in Table S1. When necessary, media

378 was supplemented with carbenicillin (100 µg/ml), kanamycin (50 µg/ml for E. coli, 100 µg/ml for

379 vibrios), spectinomycin (200 µg/ml), chloramphenicol (10 µg/ml), and/or trimethoprim (10 µg/ml).

380 Plasmids were transferred from E. coli to Vibrio strains by conjugation on LB plates.

381 Exconjugants were selected on LB or LM plates with polymyxin B at 50 U/ml and the

382 appropriate selective antibiotic. For descriptions of growth media used for each strain, see SI

383 Materials and Methods.

384

385 Molecular Methods bioRxiv preprint doi: https://doi.org/10.1101/683029; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

386 All PCR products were synthesized using Phusion HF polymerase (New England

387 Biolabs). Sequencing of constructs and strains was performed at Eurofins Scientific. Cloning

388 procedures and related protocols are available upon request. Agarose gels (0.8-2%) were used

389 to resolve DNA products. Oligonucleotides used in the study are listed in Table S3.

390

391 Synthesis of Linear tDNAs

392 Linear tDNAs were generated by splicing-by-overlap extension (SOE) PCR as previously

393 described by Dalia et al. (15). For details, see SI Materials and Methods.

394

395 Natural Transformation

396 Chitin-independent transformations were performed following the protocol established in

397 Dalia et al. (17). Chitin-dependent transformations were performed following the protocol

398 established in Dalia et al. (20). For details, see SI Materials and Methods. Following natural

399 transformation, strains containing the correct target mutation were identified via colony PCR

400 with a forward and reverse detection primer (Table S3).

401

402 GFP assays

403 Vibrio strains were first cured of the pMMB67EH-tfoX-kanR plasmid by serial inoculation

404 (3-4 times) in the absence of antibiotic selection. The pCS19 reporter plasmid was conjugated

405 into strains. Overnight cultures were diluted 1:5,000 into fresh media and incubated at 30°C

406 shaking overnight. GFP fluorescence and OD600 were measured in black 96-well plates using

407 the BioTek Cytation 3 Plate Reader.

408

409 RNA-seq bioRxiv preprint doi: https://doi.org/10.1101/683029; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

410 Strains were inoculated in 5 ml LBv2 and grown overnight shaking at 250 RPM at 30°C.

411 Each strain was back-diluted to an OD600= 0.005 in fresh LBv2 (uninduced control samples) or

412 LBv2 with 100 µM IPTG (induced samples). Cultures were grown shaking at 250 RPM at 30°C

413 until they reached OD600 =~1.8. Cells were collected by centrifugation at 3700 RPM at 4°C for

414 10 min. The supernatant was removed, the pellets flash frozen in liquid N2, and stored at -80°C.

415 RNA was isolated using the Qiagen RNeasy Mini Kit following the manufacturers protocol and

416 treated with DNase (Ambion). RNA-seq was performed at the Center for Genomics and

417 Bioinformatics at Indiana University. The results were deposited in the National Center for

418 Biotechnology Information Gene Expression Omnibus database (NCBI GEO). For details on

419 RNA sequencing and data analyses, see SI Materials and Methods.

420

421 Comparative Genomics and Phylogenetics

422 Reciprocal BLASTs between DS40M4 and BB120 were performed using NCBI BLASTP

423 ver. 2.7.1. following re-annotation of Genbank sequences with Prokka ver. 1.12 (37). To

424 generate the phylogenetic tree, a multiple sequence alignment was performed with Genbank

425 sequences using MUSCLE ver. 3.8.31 (38), input into the program RAxML ver. 8.2.12 (39), and

426 a best scoring maximum likelihood tree was constructed. For details, see SI Materials and

427 Methods.

428

429 Acknowledgments

430

431 We thank Triana Dalia for guidance on SOE PCR. We also thank Evan Schneider for technical

432 assistance. We thank Margo Haywood for the DS40M4 strain and Varaporn Vuddhakul for the

433 HY01 strain. This work was supported by National Institutes of Health grants R35GM124698 to

434 JVK and R35GM128674 to ABD. bioRxiv preprint doi: https://doi.org/10.1101/683029; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

435

436 References

437 1. Meibom KL, Blokesch M, Dolganov NA, Wu CY, & Schoolnik GK (2005) Chitin induces 438 natural competence in Vibrio cholerae. Science 310(5755):1824-1827. 439 2. Sun Y, Bernardy EE, Hammer BK, & Miyashiro T (2013) Competence and natural 440 transformation in vibrios. Mol Microbiol 89(4):583-595. 441 3. Hayes CA, Dalia TN, & Dalia AB (2017) Systematic genetic dissection of chitin 442 degradation and uptake in Vibrio cholerae. Environ Microbiol 19(10):4154-4163. 443 4. Yamamoto S, et al. (2011) Identification of a chitin-induced small RNA that regulates 444 translation of the tfoX gene, encoding a positive regulator of natural competence in 445 Vibrio cholerae. J Bacteriol 193(8):1953-1965. 446 5. Rutherford ST & Bassler BL (2012) Bacterial quorum sensing: its role in virulence and 447 possibilities for its control. Cold Spring Harbor perspectives in medicine 2(11). 448 6. van Kessel JC, Rutherford ST, Shao Y, Utria AF, & Bassler BL (2013) Individual and 449 Combined Roles of the Master Regulators AphA and LuxR in Control of the Vibrio 450 harveyi Quorum-Sensing Regulon. J Bacteriol 195(3):436-443. 451 7. Ball AS, Chaparian RR, & van Kessel JC (2017) Quorum sensing gene regulation by 452 LuxR/HapR master regulators in vibrios. J Bacteriol. 453 8. Lo Scrudato M & Blokesch M (2013) A transcriptional regulator linking quorum sensing 454 and chitin induction to render Vibrio cholerae naturally transformable. Nucleic Acids Res 455 41(6):3644-3658. 456 9. Antonova ES, Bernardy EE, & Hammer BK (2012) Natural competence in Vibrio 457 cholerae is controlled by a nucleoside scavenging response that requires CytR- 458 dependent anti-activation. Mol Microbiol 86(5):1215-1231. 459 10. Jaskolska M, Stutzmann S, Stoudmann C, & Blokesch M (2018) QstR-dependent 460 regulation of natural competence and type VI secretion in Vibrio cholerae. Nucleic Acids 461 Res 46(20):10619-10634. 462 11. Pollack-Berti A, Wollenberg MS, & Ruby EG (2010) Natural transformation of Vibrio 463 fischeri requires tfoX and tfoY. Environ Microbiol 12(8):2302-2311. 464 12. Gulig PA, Tucker MS, Thiaville PC, Joseph JL, & Brown RN (2009) USER friendly 465 cloning coupled with chitin-based natural transformation enables rapid mutagenesis of 466 Vibrio vulnificus. Appl Environ Microbiol 75(15):4936-4949. 467 13. Chen Y, Dai J, Morris JG, Jr., & Johnson JA (2010) Genetic analysis of the capsule 468 polysaccharide (K antigen) and exopolysaccharide genes in pandemic Vibrio 469 parahaemolyticus O3:K6. BMC microbiology 10:274. 470 14. Dalia AB, Seed KD, Calderwood SB, & Camilli A (2015) A globally distributed mobile 471 genetic element inhibits natural transformation of Vibrio cholerae. Proc Natl Acad Sci U 472 S A 112(33):10485-10490. 473 15. Dalia AB, Lazinski DW, & Camilli A (2014) Identification of a membrane-bound 474 transcriptional regulator that links chitin and natural competence in Vibrio cholerae. mBio 475 5(1):e01028-01013. 476 16. Chimalapati S, et al. (2018) Natural Transformation in Vibrio parahaemolyticus: a Rapid 477 Method To Create Genetic Deletions. J Bacteriol 200(15). 478 17. Dalia TN, et al. (2017) Multiplex Genome Editing by Natural Transformation (MuGENT) 479 for Synthetic Biology in Vibrio natriegens. ACS Synth Biol 6(9):1650-1655. 480 18. Bassler BL, Wright M, Showalter RE, & Silverman MR (1993) Intercellular signalling in 481 Vibrio harveyi: sequence and function of genes regulating expression of luminescence. 482 Mol Microbiol 9(4):773-786. bioRxiv preprint doi: https://doi.org/10.1101/683029; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

483 19. Lin B, et al. (2010) Comparative genomic analyses identify the Vibrio harveyi genome 484 sequenced strains BAA-1116 and HY01 as Vibrio campbellii. Environ Microbiol Rep 485 2(1):81-89. 486 20. Dalia AB (2018) Natural Cotransformation and Multiplex Genome Editing by Natural 487 Transformation (MuGENT) of Vibrio cholerae. Methods Mol Biol 1839:53-64. 488 21. Colston SM, et al. (2019) Complete Genome Sequence of Vibrio campbellii DS40M4. 489 Microbiol Resour Announc 8(4). 490 22. Rattanama P, et al. (2009) Shrimp pathogenicity, hemolysis, and the presence of 491 hemolysin and TTSS genes in Vibrio harveyi isolated from Thailand. Dis Aquat Organ 492 86(2):113-122. 493 23. Baumann P, Baumann L, & Reichelt JL (1973) of marine bacteria: Beneckea 494 parahaemolytica and Beneckea alginolytica. J Bacteriol 113(3):1144-1155. 495 24. Dias GM, et al. (2012) Genome sequence of the marine bacterium Vibrio campbellii 496 DS40M4, isolated from open ocean water. J Bacteriol 194(4):904. 497 25. Bassler BL, Wright M, & Silverman MR (1994) Sequence and function of LuxO, a 498 negative regulator of luminescence in Vibrio harveyi. Mol Microbiol 12(3):403-412. 499 26. Swartzman E, Silverman M, & Meighen EA (1992) The luxR gene product of Vibrio 500 harveyi is a transcriptional activator of the lux promoter. J Bacteriol 174(22):7490-7493. 501 27. Lo Scrudato M & Blokesch M (2012) The regulatory network of natural competence and 502 transformation of Vibrio cholerae. PLoS Genet 8(6):e1002778. 503 28. Blokesch M & Schoolnik GK (2008) The extracellular nuclease Dns and its role in natural 504 transformation of Vibrio cholerae. J Bacteriol 190(21):7232-7240. 505 29. Aagesen AM & Hase CC (2012) Sequence analyses of type IV pili from Vibrio cholerae, 506 Vibrio parahaemolyticus, and Vibrio vulnificus. Microb Ecol 64(2):509-524. 507 30. Adams DW, Stutzmann S, Stoudmann C, & Blokesch M (2019) DNA-uptake pili of Vibrio 508 cholerae are required for chitin colonization and capable of kin recognition via sequence- 509 specific self-interaction. Nat Microbiol. 510 31. Seitz P & Blokesch M (2013) DNA-uptake machinery of naturally competent Vibrio 511 cholerae. Proc Natl Acad Sci U S A 110(44):17987-17992. 512 32. Borgeaud S, Metzger LC, Scrignari T, & Blokesch M (2015) The type VI secretion 513 system of Vibrio cholerae fosters horizontal gene transfer. Science 347(6217):63-67. 514 33. Watve SS, Thomas J, & Hammer BK (2015) CytR Is a Global Positive Regulator of 515 Competence, Type VI Secretion, and Chitinases in Vibrio cholerae. PLoS One 516 10(9):e0138834. 517 34. Dalia AB, McDonough E, & Camilli A (2014) Multiplex genome editing by natural 518 transformation. Proc Natl Acad Sci U S A 111(24):8937-8942. 519 35. Metzger LC, Matthey N, Stoudmann C, Collas EJ, & Blokesch M (2019) Ecological 520 implications of gene regulation by TfoX and TfoY among diverse Vibrio species. Environ 521 Microbiol. 522 36. Mukherjee S & Kearns DB (2014) The structure and regulation of flagella in Bacillus 523 subtilis. Annu Rev Genet 48:319-340. 524 37. Seemann T (2014) Prokka: rapid prokaryotic genome annotation. Bioinformatics 525 30(14):2068-2069. 526 38. Edgar RC (2004) MUSCLE: multiple sequence alignment with high accuracy and high 527 throughput. Nucleic Acids Res 32(5):1792-1797. 528 39. Stamatakis A (2014) RAxML version 8: a tool for phylogenetic analysis and post-analysis 529 of large phylogenies. Bioinformatics 30(9):1312-1313. 530

531 bioRxiv preprint doi: https://doi.org/10.1101/683029; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

532 Figure legends

533

534 Figure 1. Natural transformation of V. campbellii strains via tfoX expression. (A)

535 Phylogenetic tree of Vibrio strains based on comparison of amino acid sequences of 79 core

536 conserved genes in the genomes shown. (B) Chitin-independent transformation of V. campbellii

537 strains BB120, DS40M4, NBRC 15631, and HY01 compared to V. natriegens. Strains contained

538 either a plasmid expressing tfoX (pMMB67EH-tfoX-kanR) or an empty vector control

539 (pMMB67EH-kanR for V. campbellii strains or pMMB67EH-carbR for V. natriegens). Strains

540 were transformed with 300 ng of linear luxR::specR tDNA (for V. campbellii) or dns::specR tDNA

541 (for V. natriegens). LOD, limit of detection.

542

543 Figure 2. V. campbellii DS40M4 and NBRC 15631 encode functional LuxR and LuxO

544 proteins. The PluxCDABE-gfp reporter plasmid pCS019 was introduced into wild-type, DluxR, and

545 DluxO strains of V. campbellii BB120 (A), DS40M4 (B), and NBRC 15631 (C), and the GFP

546 fluorescence divided by OD600 was determined. Different letters indicate significant differences

547 (p < 0.001; one-way analysis of variation (ANOVA), followed by Tukey’s multiple comparison

548 test of log-transformed data; n = 3).

549

550 Figure 3. Natural transformation of V. campbellii DS40M4 does not require LuxR. (A, B)

551 Transformation frequencies using chitin-independent transformation of a DluxB::tmR substrate

552 (300 ng) into wild-type, DluxR, and DluxO strains of DS40M4 (A) and NBCR 15631 (B). (C)

553 Transformation frequencies using chitin-independent transformation of a Dvc1807::tmR substrate

554 (300 ng) into wild-type, DhapR, and DluxO strains of V. cholerae. LOD, limit of detection.

555 Different letters indicate significant differences (p < 0.05; one-way analysis of variation

556 (ANOVA), followed by Tukey’s multiple comparison test of log-transformed data; n = 4). bioRxiv preprint doi: https://doi.org/10.1101/683029; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

557

558 Figure 4. Comparative genomics and transcriptomics analyses of BB120 and DS40M4

559 competence genes. Genes required for DNA uptake and integration previously determined in

560 V. cholerae were identified in BB120 and DS40M4 using reciprocal BLAST analyses. Genes are

561 organized based on function. The locus tags correspond to V. cholerae genes; corresponding

562 locus tags for BB120 and DS40M4 are in Table S4. (A) The chart indicates the amino acid

563 identity shared between V. cholerae, BB120, and DS40M4, which is shown in each bar and by

564 the color scale. (B) The chart indicates the RPKM values derived from RNA-seq data comparing

565 either BB120 or DS40M4 transcript levels in the presence or absence of tfoX induction (strains

566 contain plasmid pMMB67EH-tfoX-kanR).

567

568 Figure 5. QstR and TfoX are necessary and sufficient to bypass LuxR for natural

569 transformation. (A) Transformation frequencies using chitin-independent transformation of a

570 DluxB::tmR substrate (300 ng) into wild-type and DluxR strains of DS40M4, either with the

571 pMMB67EH-tfoX-kanR plasmid (ptfoX) or pCS32 plasmid (ptfoX-qstR). (B) Transformation

572 frequencies using chitin-independent transformation of a DluxB::specR substrate (300 ng) into

573 wild-type, DluxR, DqstR, and DluxR DqstR strains of DS40M4, all containing pMMB67EH-tfoX-

574 kanR. Different letters indicate significant differences (p < 0.01; one-way analysis of variation

575 (ANOVA), followed by Tukey’s multiple comparison test of log-transformed data; n = 3).

576

577 Figure 6. Natural transformation frequencies vary in Vibrio species. (A) Chitin-independent

578 transformations in each of the listed Vibrio species using SpecR linear tDNAs (300ng) targeting

579 the luxR homolog in each strain in the presence of a plasmid expressing either tfoX alone

580 (pMMB67EH-tfoX-kanR) or both tfoX and qstR (pCS32). Asterisks indicate significant

581 differences (p < 0.0001; two-way analysis of variation (ANOVA), followed by Sidak’s multiple bioRxiv preprint doi: https://doi.org/10.1101/683029; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

582 comparison test of log-transformed data; n = 3). (B, C) Transformation frequencies using chitin-

583 independent transformation of a DpomB::tmR substrate (300 ng) into wild-type and DsmcR V.

584 vulnificus strains expressing tfoX (pMMB67EH-tfoX-kanR) (B) or wild-type and DopaR V.

585 parahaemolyticus strains expressing tfoX (pMMB67EH-tfoX-kanR) (C). The asterisk indicates p

586 < 0.05 in an unpaired, two-tailed t-test.

587 bioRxiv preprint doi: https://doi.org/10.1101/683029; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Figure 1

A Vibrio campbellii HY01 Vibrio parahaemolyticus RIMD 2210633

Vibrio natriegens NBRC 15636 (ATCC 14048)

Vibrio cholerae N16961

Vibrio vulnificus ATCC 27562

Vibrio fischeri ES114

Vibrio campbellii NBRC 15631 (ATCC 25920, CAIM 519)

Vibrio campbellii DS40M4

Vibrio campbellii BB120 (ATCC BAA-1116) 0.01 B 100 10-1 10-2 10-3 10-4

-5 10 LOD -6 LOD LOD LOD LOD 10 LOD LOD 10-7 10-8 Transformation Frequency (CFUs/mL) ptfoX - ptfoX - ptfoX - ptfoX - ptfoX - BB120 DS40M4 NBRC HY01 V. nat. 15631 bioRxiv preprint doi: https://doi.org/10.1101/683029; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Figure 2

BB120 DS40M4 NBRC 15631 A B a C b 30000 30000 a 30000 600 600 600 20000 a a 20000 20000

10000 10000 10000 GFP/OD GFP/OD GFP/OD

b b a a 0 0 0 WT DluxR DluxO WT DluxR DluxO WT DluxR DluxO bioRxiv preprint doi: https://doi.org/10.1101/683029; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Figure 3

100 100 100 A DS40M4 B NBRC 15631 C V. cholerae a -1 10-1 10-1 a 10 c 10-2 10-2 10-2 10-3 10-3 10-3 a 10-4 10-4 10-4 LOD 10-5 10-5 b 10-5 LOD -6 -6 -6 10 10 10 b 10-7 10-7 10-7 -8 -8 -8 Transformation Frequency (CFUs/mL)

10 Transformation Frequency (CFUs/mL) Transformation 10Frequency (CFUs/mL) 10 WT DluxR DluxO WT DluxR DluxO WT DhapR DluxO bioRxiv preprint doi: https://doi.org/10.1101/683029; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Figure 4

A B Gene conservation Gene expression

BB120 DS40M4 BB120 BB120 BB120 DS40M4 DS40M4 vs vs vs - + tfoX - + tfoX V. cholerae V. cholerae DS40M4

chitin 59 utilization

inhibition 38 32

41 integration DNA uptake

56

65

40 pilus

52 57

30 38 regulation

67

20 40 60 80 100 0 50 100 150 200 Color key values Color key values bioRxiv preprint doi: https://doi.org/10.1101/683029; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Figure 5

0 a A 100 DS40M4 B 10 DS40M4 a -1 10-1 a 10 -2 -2 10 10 b -3 10-3 10 b -4 10-4 10 -5 10-5 10 -6 -6 10 c 10 c LOD -7 LOD 10-7 10 -8

-8 Transformation Frequency10 (CFUs/mL)

Transformation 10Frequency (CFUs/mL) WT DluxR DluxR WT DluxR DqstR DluxR ptfoX ptfoX ptfox- DqstR qstR bioRxiv preprint doi: https://doi.org/10.1101/683029; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Figure 6

PtfoXptfoX A 101 ns PtfoX-qstRptfoX-qstR ns 100 ns 10-1 * 10-2 ns 10-3 10-4 10-5 10-6 10-7 10-8 NBRC 15631 DS40M4 V. para V. cholerae V. vulnificus Transformation Frequency (CFUs/mL) V. campbellii

B 101 V. vulnificus C 101 V. parahaemolyticus 100 100 * 10-1 10-1 10-2 10-2 10-3 10-3 10-4 10-4 10-5 10-5 LOD 10-6 10-6 10-7 10-7 10-8 -8

Transformation10 Frequency (CFUs/mL) Transformation Frequency (CFUs/mL) WT DsmcR WT DopaR