Modular small RNA drives the emergence of virulence traits and environmental trade-offs in Vibrio cholerae

Deepak Balasubramanian; Qinqin Pu; Cole Crist; Salvador Almagro-Moreno

doi:10.1038/s41467-026-73137-6

. 2026 May 14;17:6440. doi: 10.1038/s41467-026-73137-6

Modular small RNA drives the emergence of virulence traits and environmental trade-offs in Vibrio cholerae

Deepak Balasubramanian ^1,², Qinqin Pu ¹, Cole Crist ², Salvador Almagro-Moreno ^1,^2,^3,^✉

PMCID: PMC13376802 PMID: 42135309

Abstract

The molecular drivers and fitness trade-offs controlling the transition of environmental bacteria into human pathogens remain enigmatic. Here, we describe a small RNA (sRNA) with a unique modular structure that shapes the evolution of toxigenic Vibrio cholerae, the agent of cholera. The sRNA comprises a variable 5’ module located within the ompU ORF and a conserved 3’ region downstream. This atypical location confers a distinct bimodular structure to the OmpU-encoded sRNA (OueS), generating allelic variants with unique functions. Unlike environmental counterparts, the OueS allele from toxigenic strains controls phenotypes essential during host colonization in a domain-dependent manner: a) inhibits biofilm formation by suppressing RyhB, b) confers resistance against intestinal bacteriophages by activating the CBASS system, and c) drives mucus penetration. Toxigenic OueS is essential for successful intestinal colonization and acts as a functional surrogate of the master virulence regulator ToxR, controlling over 84% of its regulome. However, this allele leads to reduced competitive fitness during colonization of natural reservoirs when compared to environmental ones. Finally, we determined that the acquisition of the ompU ORF is modular and controlled by exogenous conditions, providing critical insights into the evolution of toxigenic V. cholerae and the emergence of pathogenic traits in bacteria.

Subject terms: Pathogens, Molecular evolution, Bacterial genetics

In this work, authors describe a modular small RNA that drives the emergence of virulence traits in the cholera pathogen. It controls key colonization phenotypes, acting as the functional surrogate of a major virulence regulator, and is associated with environmental fitness trade-offs.

Introduction

The emergence of human pathogens is one of the most pressing public health concerns of modern times. Pathogen emergence is a complex and multifactorial phenomenon that culminates with the organism acquiring the ability to colonize and harm the human host^1–3. A successful infectious process necessitates accomplishing numerous steps and overcoming critical host-associated hurdles. These include avoidance of host defense mechanisms and phage predation, the correct spatiotemporal expression of virulence genes in response to host signals, and, ultimately, proliferation and transmission. Despite its critical relevance, many essential questions regarding this complex phenomenon remain unanswered. For instance, the sole gain of laterally acquired virulence genes cannot fully explain the transition of environmental strains into human pathogens, as typically only a limited number of clades within a species eventually emerge to cause disease. To date, the specific molecular drivers and fitness trade-offs that enable certain strains within a population to undergo this process remain enigmatic. Identifying the elements associated with the process of pathogen emergence is critical for successful disease management and control⁴.

Vibrio cholerae is the etiological agent of the severe diarrheal disease cholera and remains a major scourge in places with limited access to clean drinking water^5,6. The disease is contracted via the consumption of food or water contaminated with toxigenic strains of V. cholerae⁵. Following ingestion of planktonic or biofilm-associated cells, V. cholerae must overcome several host defenses, such as acidic stomach pH, bile acids, and antimicrobial peptides^7–10. Successful colonization also requires stringent downregulation of biofilm formation, as it triggers a strong immune response, paired with upregulation of flagella-mediated motility to swim through the mucus layer towards the epithelium^11–13. A complex regulatory cascade mediated by the virulence regulator ToxR facilitates adaptation to the harsh conditions within the host and, along with other regulators, the coordinated and timely expression of virulence genes^14–17. ToxR is a transmembrane protein with a cytoplasmic DNA-binding domain that controls the expression of hundreds of genes, however, the regulator only binds directly to the promoters of 39 genes^15,18–21. Interestingly, about 40% of the promoters that ToxR binds to are located in horizontally acquired pathogenicity islands, specifically, VSP-I, VSP-II, VPI-1 and VPI-2 islands²¹. Among the genes directly regulated by ToxR are those encoding the cholera toxin (CT), the source of the watery diarrhea associated with cholera, and, together with TcpP, the expression of ToxT, a master virulence regulator^18,22. ToxR also directly and reciprocally modulates the expression of two outer membrane proteins (OMPs): OmpU and OmpT. Specifically, ToxR represses ompT expression, whereas it activates transcription of ompU^19,23,24.

OmpU is highly expressed during infection, making up to 60% of the outer membrane, and is essential for host survival and intestinal colonization^19,25,26. The porin is associated with numerous virulence-adaptive phenotypes such as bile resistance, and tolerance to organic acids and cationic peptides (e.g., polymyxin B or the host-derived antimicrobial peptide P2)^7,8,19,26,27. Beyond resistance to cell-damaging agents, OmpU has also been associated with several more complex phenotypes. For instance, OmpU exerts immunomodulatory effects such as induction of host tolerance to LPS and suppression of the host immune response by inducing M1 polarization of macrophages^28,29. Furthermore, OmpU serves as the receptor of bacteriophage ICP2, a phage isolated from stool samples of cholera patients³⁰. We recently determined that OmpU is also associated with biofilm formation in V. cholerae, as ompU deletion mutants form 3-4-fold more robust biofilms than wild-type strains³¹. Interestingly, moonlighting roles critical to infection-related processes appear to be a common feature among OMPs in bacterial pathogens. For instance, OMPs have been associated with activation of host signal transduction pathways, inhibition of host cell apoptosis, adhesion and signaling, and modulation of immune responses^32–36. To date, the mechanisms by which porins like OmpU modulate such diverse virulence-associated phenotypes remain poorly understood.

Despite encompassing over 200 serogroups, only V. cholerae strains from serogroups O1 and O139 can cause cholera in humans^37–39. Strains from these two serogroups are phylogenetically related and confined to a single clade containing all toxigenic V. cholerae called the pandemic cholera group (PCG)⁴⁰. To elucidate the evolutionary origins and emergence of toxigenic V. cholerae, we previously developed a comparative genomics framework³¹. Our analyses of environmental and toxigenic strains of V. cholerae identified the presence of allelic variations in core genes that circulate in environmental populations and confer preadaptations to host colonization, enhancing their pandemic potential³¹. These genes encode what we term virulence adaptive polymorphisms (VAPs) and are connected with the emergence of toxigenic strains of V. cholerae, explaining their limited phylogenetic distribution³¹. Interestingly, we found that ompU in toxigenic V. cholerae strains encodes VAPs, and the virulence traits associated with the porin are allelic-dependent, as these traits are absent from strains with ompU alleles that do not encode them^31,41. The molecular mechanisms controlling these allelic-dependent emergent phenotypes remain unaddressed.

Here, we describe a small RNA (sRNA) with a unique modular structure that contributes to the evolution of toxigenic strains of V. cholerae. The sRNA comprises a highly variable 5’ module located within the ompU ORF and a conserved 3’ region downstream from the gene. The bimodular nature of the OmpU-encoded sRNA (OueS) generates allelic variants that differentially contribute to both the emergence of virulence potential in toxigenic strains in a domain-dependent manner, and the associated fitness trade-offs between human infection and environmental survival. Our findings provide critical insights into the evolution of pandemic V. cholerae and reveal specific molecular mechanisms and fitness costs associated with the emergence of pathogenic traits in bacteria.

Results

OmpU represses biofilm formation in V. cholerae and modulates its transcriptome

OmpU plays a major role in V. cholerae pathogenesis. However, it remains unclear how an outer membrane porin can modulate so many disparate virulence-related phenotypes^7,8,19,26,27. We previously demonstrated that deletion of ompU in toxigenic V. cholerae leads to more robust biofilm formation³¹. To dissect this phenotype, we ectopically expressed ompU from an inducible plasmid in an in-frame isogenic ompU deletion mutant in V. cholerae C6706, an El Tor strain from the seventh cholera pandemic (∆ompU-pOmpU^C6706). Expression of ompU decreases biofilm formation in the ∆ompU mutant back to wild-type (WT) levels, indicating that OmpU represses biofilm formation in V. cholerae (Fig. 1A). On the other hand, expression of an environmental ompU allele from a strain isolated from the Great Bay Estuary in New Hampshire, V. cholerae GBE1114⁴², either ectopically or through allelic exchange in the isogenic C6706 background, does not timely repress biofilm formation, suggesting this phenotype is allelic-dependent (Fig. 1A)³¹. Interestingly, biofilm suppression is critical in the early stages of intestinal colonization as biofilm components elicit a strong immune response, leading to clearance of V. cholerae strains that do not timely repress it⁴³. We found it quite puzzling that the expression of a porin localized in the outer membrane can control a phenotype that is central to the infectious process.

To gain insights into the mechanisms behind this marked phenotype, we examined the transcriptional changes associated with OmpU-mediated biofilm repression. We performed RNA-Seq analyses of V. cholerae biofilms produced by WT, ∆ompU and ∆ompU-pOmpU^C6706 strains. Our data reveals a drastic shift in the transcriptome of the ∆ompU mutant strain compared to WT (Supplementary Fig. S1), with differential regulation of 270 genes, of which 181 genes are repressed by OmpU and 89 are activated (absolute FC ≥ 2.0, q ≤ 0.01) (Supplementary Data Table 1). Ectopic expression of ompU^C6706 in the ∆ompU background results in a shift of the transcriptome, relatively closer to WT, and the recovery in expression of the dysregulated genes (Supplementary Fig. S1 and Supplementary Data Table 1). GO enrichment analysis reveals that OmpU primarily represses genes involved in iron metabolism and acquisition, specifically in vibriobactin metabolism, siderophore transport, and iron ion homeostasis (q ≤ 0.05) (Supplementary Fig. S1 and Supplementary Data Table 1). On the other hand, OmpU activates genes involved in the defense response to phages and symbionts, as well as those involved in heat response and protein folding (q ≤ 0.05) (Supplementary Fig. S1 and Supplementary Data Table 1).

Upon misfolding, OmpU activates the cell envelope stress response via the release of sigma factor RpoE in the cytosol⁴⁴ (Fig. 1B). To address the potential role of RpoE in the transcriptomic changes associated with OmpU, we examined the biofilm transcriptome of an ∆rpoE mutant strain (Fig. 1C). By normalizing the ∆ompU and ∆rpoE datasets to the WT, we isolated the RpoE-dependent changes in gene expression from additional transcriptional shifts unique to ompU. This comparison demonstrates that although there is a natural overlap between the two datasets, a large percentage of the transcriptomic changes associated with ompU expression during biofilm formation appear to be independent of RpoE activation (Fig. 1C and Supplementary Data Table 2). Specifically, OmpU modulates the expression of 270 genes during biofilm formation, and 144 of those genes (53.3%) are not associated with RpoE (Fig. 1C and Supplementary Data Table 2). Overall, our data indicate that OmpU represses biofilm formation in an allele-dependent manner and has a significant impact on the V. cholerae transcriptome.

The ompU ORF encodes a small RNA

Given the large transcriptomic changes associated with OmpU extending beyond the activation of RpoE, we considered the possibility of the ompU gene coding for regulatory sequences within its ORF besides the porin. A genome-wide screening in V. cholerae indicated potential unconfirmed sRNAs encoded within the 3’ end of the ompU gene among thousands of others across the genome⁴⁵. To examine whether the ompU ORF potentially encodes an internal sRNA, we performed northern blot analysis with probes targeting the 3’ termini of ompU and the 5’ termini as a control (Fig. 1D). We reasoned that a probe targeting the sRNA locus within ompU should result in two bands corresponding to both the ompU mRNA and the potential sRNA. The northern blot data shows that a probe specific for the 5’ end of the ompU ORF (675,488–675,527 nt) results in a ~1100 nt band corresponding to the expected size of the mRNA transcript (Fig. 1D). On the other hand, in addition to the ~1.1 kb mRNA band, smaller size bands of ~200 nt appear with a probe targeting the 3’ end of the ompU ORF (674,837–674,876 nt) potentially corresponding to a sRNA ( ~160 nt) (Fig. 1D). Hfq is a RNA binding protein that plays an integral role in their stability and target recognition⁴⁶. Interestingly, the lower bands are not detectable in an ∆hfq mutant, strengthening the possibility of an Hfq-dependent sRNA being encoded within the 3’ end of the ompU ORF (Fig. 1D).

To define the 5’ and 3’ termini of the potential sRNA encoded within the ompU ORF, we performed rapid amplification of cDNA ends (RACE) assays. Given the overabundance of the ompU transcript and the overlapping nature of the mRNA and the potential sRNA, our initial 5’ RACE studies yielded only the termini of ~1100 nt ompU mRNA transcript. To circumvent this, total RNA from the WT strain cultured in LB was size fractionated on a MOPS agarose gel and the 100–500 nt fraction, encompassing the potential sRNA region, was extracted and used as a template for RACE studies. Assays on this size-fractionated RNA led to the identification of a sRNA corresponding to 166 nt that we termed ompU-encoded small RNA (OueS) (Fig. 1E). Surprisingly, even though the majority of the sRNA (112 nt) is encoded within the ompU ORF before its 3’ terminus, 54 nt are located downstream of the stop codon in the 3’ UTR (674,939 – 674,774 on chromosome I), making the OueS locus quite atypical as it overlaps an ORF and its 3’ UTR, with an identical 3’ end as the ompU mRNA (Fig. 1E). Computational prediction of the secondary structure of OueS reveals a four stem-loop structure, with the terminal stem-loop ending in a poly-U tail typical of a Rho-independent terminator and a potential Hfq binding site, further corroborating our finding in Fig 1F^46,47 (Fig. 1F). Monitoring ompU and OueS expression at different stages of growth by northern blotting indicate that the expression of the mRNA and sRNA are uncoupled in the early stages of growth and drastically decrease after 9 h (Fig. 1G–I). Specifically, the expression of ompU remains constant and high during the first 6 h of growth, whereas that of OueS increases overtime and peaks at the early stationary phase (6 h) with a reduction in sRNA levels at later growth phases (Fig. 1G). Interestingly, the environmental ompU allele GBE1114, which does not lead to biofilm suppression, encodes a distinct allele and structural variant of OueS (Fig. 1J, K). Expression analysis of size-fractionated RNA (as above) from V. cholerae GBE1114 grown in LB medium shows that OueS^GBE1114 is expressed at levels comparable to those in V. cholerae C6706 (Supplementary Fig. S2A). Interestingly, even though the toxigenic and environmental OueS alleles are 166 nt in length, the predicted secondary structures of the two sRNAs drastically differ in the 5’ region encoded within the ompU ORF. In contrast, the downstream region that harbors the Hfq binding motif⁴⁷ is identical between the two sRNA alleles, suggesting a bimodular architecture of OueS (Figs. 1F, J, K). The non-canonical location of OueS, overlapping an ORF and its UTR, and its potentially modular nature makes this a unique sRNA that might contribute to the phenotypic diversity of V. cholerae^48–50. Overall, our results indicate that the ompU ORF in V. cholerae encodes both the OmpU porin and the 5’ terminus of a 166 nt sRNA that we termed OueS.

OueS represses biofilm formation in V. cholerae

Next, we examined whether the ompU-mediated biofilm repression seen in Fig. 1A is mediated by OueS. We ectopically expressed OueS in the ∆ompU mutant (∆ompU-pOueS) and determined that the strain recovers the WT phenotype and displays reduced biofilm formation when compared to ∆ompU, similar to the ectopic expression of OmpU (Fig. 1L). This finding indicates that OueS is responsible for the ompU-mediated biofilm repression in V. cholerae (Fig. 1L). To examine the specificity of OueS-mediated biofilm repression, first, we ectopically expressed a non-specific (NS) region outside of the OueS locus located towards the 5’ end of the ompU ORF (675,307–675,521). The NS region had no effect on biofilm formation compared to the ∆ompU strain (Fig. 1L). Subsequently, we planned on generating mutations within OueS to potentially nullify its effect. However, the 5’ end of OueS is encoded within the ompU ORF, and thus, mutations affecting OueS would also affect the OmpU coding region. To ensure that the mutations within the sRNA do not disrupt the protein coding sequence, we took advantage of the codon wobble and mutated every third base in the 166 bp OueS locus, disrupting the sRNA secondary structure while preserving the OmpU amino acid sequence (Fig. 1J), what we termed OueS full-length mutant (FL-mut). We integrated OueS FL-mut at the native OueS locus, generating a strain encoding a WT OmpU porin and a mutated OueS sRNA (C6706 OmpU^mut). The OmpU porin produced by this strain is functional and stable, as determined by resistance to bile (Fig. 1M) and OmpU immunoblots (Fig. 1N), respectively, indicating that the porin function remains unaffected due to OueS disruption. The OmpU^mut strain formed higher biofilm levels, similar to ∆ompU, further indicating that biofilm repression is mediated by OueS and not OmpU (Fig. 1L). In addition, we generated ectopic expression constructs of three OueS wobble mutants: OueS 5’-mut harboring mutations exclusively in the OueS region that lies within the ompU ORF, and OueS 3’-mut with mutations only in the OueS region downstream of ompU, and the OueS FL-mut with mutations across the entire length of the sRNA. All three mutant versions of OueS disrupted the predicted secondary structures of the sRNA while the protein coding sequence remains unaltered (Fig. 1J and Supplementary Fig. S2B–E). Ectopic expression of either of the three OueS mutant sRNAs (5’-mut, 3’-mut and FL-mut) in the ∆ompU strain does not lead to suppression of biofilm formation (Fig. 1L). Additionally, unlike the OmpU and OueS alleles from toxigenic strain C6706, ectopic expression of the environmental OueS^GBE1114 does not lead to biofilm suppression (Fig. 1L). Overall, our results demonstrate that OueS leads to stringent biofilm repression in V. cholerae, this phenotype is dependent on the OueS allele and that the sRNA has a unique and possibly bimodular structure that divides its ompU-encoded 5’ end from its 3’ terminus.

OueS is responsible for ompU-dependent transcriptomic changes during biofilm formation

Given the effect of OueS on biofilm repression, we hypothesized that expression of OueS is responsible for the altered transcriptional profile associated with OmpU (Fig. 1C and Supplementary Fig. S1). To address this, we examined the transcriptomes of ∆ompU-pOueS^C6706 cells harvested from biofilms. Comparison of the biofilm transcriptomes of ∆ompU-pOueS^C6706 to those of the WT and ∆ompU strains reveals that OueS expression restores the shift in the transcriptome similar to ectopic ompU expression (Supplementary Fig. S1A). Further, we compared the differentially regulated genes (absolute FC ≥ 2.0, q ≤ 0.01) of two pairwise comparisons: WT Vs ∆ompU (270 genes) and ∆ompU-pOueS^C6706 Vs ∆ompU (267 genes). There are 132 genes (48.89%) in the overlap region whose expression is modulated by OueS, 101 of which are repressed and 31 activated (Supplementary Data Table 3). Analysis of the expression values of these 132 genes between the two pair-wise comparisons reveals that the direction of the fold change (activation Vs repression) as well as the expression levels are very similar in 131 of the 132 genes between the two groups (Pearson Correlation Coefficient = 0.9998; Supplementary Data Table 3). GO enrichment analysis on the genes repressed by OueS reveal that 63 of the 101 OueS repressed genes are enriched in the GO Biological Process database, and 17 of the 31 activated ones are enriched. Similar to the OmpU transcriptome, among the OueS-repressed genes there was a significant positive enrichment (q < 0.05) of genes belonging to the vibriobactin biosynthetic processes and siderophore transmembrane transport (Fig. 2A). Specifically, in addition to several genes involved in iron uptake such as the tonB operon (VC1543-VC1548) and hut-tonB1 gene cluster (VCA0907-VCA0914), genes of the rbm gene cluster (VC0929-VC0932) shown previously to play a role in biofilm formation⁵¹ are repressed by OueS^C6706 expression (Supplementary Data Table 3). Overall, our data indicate that OueS is responsible for changes in the transcriptome during biofilm formation associated with OmpU.

Fig. 2 — A Functional gene ontology (GO Biological Processes) categorization of the genes repressed by OueS during biofilm formation was performed by the Panther Overrepresentation Test^108,109. OueS represses categories related to iron uptake. B Biofilm formation of strains expressing mutant versions of OueS was examined under iron-replete (dark gray bars) and iron-depleted (white bars) conditions by exogenous addition of 100 µM FeCl₃ or 100 µM 2,2’-bipyridyl, respectively. Data was compared to growth media for baseline iron conditions (light gray bars). n ≥ 6 biological replicates. C Venn diagram indicating the number of genes that OueS alleles interact with based on MAPS analyses. The number of genes that OueS^C6706 uniquely interacts with is indicated in blue. D Read mapping plot for RyhB locus. E OueS represses expression of *ryhB*. Normalized read counts (transcripts per kilobase million, TPM) from the biofilm transcriptome data was normalized to the expression of *VC0799*, whose expression remains stable between the strains (TPM range 0.96–1.68). Transcriptome data from biological replicates are shown. Significance was calculated using an unpaired T test. **** = p < 0.0001. F Northern blots demonstrate increased levels of RyhB in a ∆*ompU* mutant. The presence of RyhB is drastically reduced upon expression of OueS^C6706. G Alignment indicating the specific location of OueS domains and mutant construction. H Domains in the 5’ locus of OueS^C6706 were replaced with the corresponding 5’ domains of the OueS^GBE1114 allele in an isogenic *V. cholerae* C6706 P_ompU-OueS^C6706 background. Total RNA was isolated from cells cultured under iron-limiting conditions, size fractionated, and OueS expression was quantified using qRT-PCR. OueS expression levels were comparable across all strains. Expression was normalized to the *recA* using the 2^-∆∆cT method. n = 3 biological replicates (I) OueS domains are differentially required for biofilm repression. Northern blots demonstrate increased levels of RyhB in (J) individual and (K) combination domain mutants with biofilm derepression. The presence of RyhB is drastically reduced in strains expressing domains that repress biofilm formation. Representative blot of three biological replicates are shown. Data are presented as mean ± s.e.m. Source data are provided as a Source Data file.

OueS from toxigenic strains suppresses biofilm formation by directly repressing the iron-responsive sRNA RyhB

We analyzed the OueS-regulated genes for insights into the molecular basis behind the stringent biofilm repression associated with the toxigenic allele of the sRNA. The top functional categories of genes repressed by OueS^C6706 are associated with iron uptake and metabolism (Fig. 2A and Supplementary Data Table 3). Interestingly, iron is important for biofilm formation^52–54. Consequently, we examined the interplay between iron availability and biofilm formation, and their association with OueS. Addition of exogenous iron to the media results in increased biofilm formation across all the strains, overriding the suppression exerted by OueS (dark gray bars, Fig. 2B and Supplementary Fig. S3A). On the other hand, depletion of the native iron in the media using the iron chelator 2,2’-bipyridyl results in drastically reduced biofilm formation in all strains, closely resembling the phenotype associated with OueS^C6706 production (white bars, Fig. 2B and Supplementary Fig. S3A). These changes during biofilm formation are not due to growth defects of the strains under these conditions (Supplementary Fig. S3B–D). We hypothesized that OueS^C6706 suppresses biofilm formation by repressing one or more iron-uptake regulators. To define the target profile of OueS, we performed MS2-affinity purification coupled to RNA-sequencing (MAPS)⁵⁵. MAPS was conducted under iron-limiting conditions as above⁵². Under these conditions, OueS^C6706 bound to 61 RNA targets ( ≥ 2-fold enrichment, q-value < 0.01), (Table 1 and Supplementary Data Table 4). Since biofilm repression is allelic-dependent, we performed parallel MAPS analyses using strains expressing the environmental allele OueS^GBE1114. Furthermore, as a negative control, we examined the target genes of the OueS^mut variant as any binding would be non-specific, thus removing any false positives. Comparative analyses revealed 43 targets uniquely bound by OueS^C6706 (≥ 2-fold enrichment, q-value < 0.01) (Fig. 2C and Table 1). These genes include the iron-responsive regulator RyhB, which promotes biofilm formation by modulating iron uptake⁵², providing a potential mechanistic basis for OueS-mediated biofilm repression. Unlike internal controls, direct binding to RyhB was observed exclusively for OueS^C6706 (Fig. 2D and Supplementary Fig. S3E). Furthermore, biofilm transcriptome analyses indicate that the transcript numbers of ryhB are significantly increased upon deletion of ompU and decrease upon expression of OueS^C6706 (Fig. 2E). On the other hand, OueS^C6706 does not affect the major iron uptake repressor fur (VC2106), overall, making RyhB a strong potential target of OueS (Fig. 2E)⁵⁴. Next, we performed northern blots to examine the levels of RyhB in different V. cholerae OueS mutant strains (Fig. 2F)⁵². Our results show that strains encoding OueS^C6706 exhibit highly reduced levels of RyhB (Fig. 2F). This repression is relieved in strains coding for the non-functional OueS, OmpU^mut, or the environmental allele OueS^GBE1114, corroborating our biofilm, MAPS and transcriptomic data (Fig. 2F). Overall, our results indicate that the inhibition of biofilm formation meditated by OueS involves the direct suppression of RyhB in an allelic-dependent manner.

Table 1.

RNA targets of OueS^C6706 identified by MAPS

Locus Tag	Function	Fold enrichment (tagged/untagged)	q-value
VC0061	thiamin biosynthesis protein ThiC	3.40	0.00252
VC0091	O-methyltransferase- related protein	5.08	0.00001
VC0107	Iron-responsive sRNA, RyhB	2.23	0.00000
VC0170	peptide ABC transporter, ATP-binding protein	2.60	0.00250
VC0329	DNA-directed RNA polymerase, beta	2.80	0.00077
VC0432	malate dehydrogenase	6.84	0.00000
VC0476	D-erythrose 4-phosphate dehydrogenase	4.48	0.00149
VC0583	quorum-sensing regulator HapR	3.34	0.00000
VC0608	iron(III) ABC transporter, periplasmiciron-compound-binding protein	2.11	0.00301
VC0784	sodium/alanine symporter	5.98	0.00000
VC0941	serine hydroxymethyltransferase, GlyA	6.98	0.00001
VC1014	RnfD-related protein	4.44	0.00717
VC1056	recombination protein RecR	5.40	0.00002
VC1219	phenylalanyl-tRNA synthetase, alpha chain	4.44	0.00717
VC1334	conserved hypothetical protein	5.97	0.00611
VC1442	cytochrome c oxidase, subunit CcoN	2.16	0.00375
VC1618	multidrug resistance protein, putative	2.40	0.00010
VC1730	DNA topoisomerase I	3.64	0.00961
VC1877	tetraacyldisaccharide 4‘-kinase, lpxK	2.53	0.00910
VC1888	hemolysin-related protein	2.37	0.00000
VC2095	phosphoglucomutase	6.50	0.00333
VC2209	nonribosomal peptide synthetase VibF	2.21	0.00014
VC2337	transcriptional regulator, LacI family	5.97	0.00611
VC2352	NupC family protein	2.29	0.00283
VC2369	sensor histidine kinase FexB	2.53	0.00080
VC2512	conserved hypothetical protein	2.99	0.00233
VC2535	pmbA protein	3.40	0.00252
VC2586	ribosomal protein L14	14.20	0.00132
VC2663	molecular chaperone groEL_1	4.68	0.00000
VC2698	aspartate ammonia-lyase	2.86	0.00027
VC2699	C4-dicarboxylate transporter, anaerobic	3.40	0.00252
VC2738	phosphoenolpyruvate carboxykinase	2.05	0.00498
VC2770	ATP synthase F0, A subunit	2.45	0.00143
VCA0025	transporter, NadC family	2.94	0.00004
VCA0052	hypothetical protein	5.87	0.00007
VCA0090	conserved hypothetical protein	5.17	0.00225
VCA0276	Glycine cleavage system P protein GcvP, authentic frameshift	4.15	0.00019
VCA0707	regulatory protein UhpC, putative	5.87	0.00007
VCA0814	agmatinase	2.60	0.00250
VCA0909	oxygen-independent coproporphyrinogen IIIoxidase, putative	3.27	0.00066
VCA0914	hemin ABC transporter, permease protein,putative	4.76	0.00080
VCA0977	ABC transporter, ATP-binding protein	13.49	0.00001
VCA1073	bifunctional proline dehydrogenase/pyrroline-5-carboxylate dehydrogenase	2.78	0.00013

Open in a new tab

OueS domains determine RyhB repression and biofilm formation

Sequence comparison of the clinical OueS^C6706 and the environmental allele OueS^GBE1114 reveal that the differences between the two alleles are confined to four discrete domains in the 5’ module of the sRNA, what we referred to as domains 1-4 (D1-D4), whereas the 3’ module is fully conserved (Fig. 2G). This natural modular divergence enabled us to test whether discrete sequence modules encode separable regulatory functions within the sRNA, analogous to protein domain architecture, using evolutionarily sampled variants rather than synthetic modifications. To assess the functional contribution of these domains in RyhB repression and biofilm formation, we generated a series of chimeric OueS variants where we substituted the individual domains and their combinations from OueS^C6706 for those of OueS^GBE1114. Our qRT-PCR analyses show that all chimeric sRNAs are expressed at similar levels as WT (Fig. 2H). Phenotypic analyses reveal that substitution of D2 or D3 alone was sufficient to abolish biofilm repression ( ~3-fold higher than WT) (Fig. 2I). Consequently, replacement of both domains together (D2,3) or substitution of all four domains (D1-4) yield a phenotype indistinguishable from OueS^GBE1114 (Fig. 2I). In contrast, replacement of D1 or D4 had no effect on OueS^C6706 ability to repress biofilm formation (Fig. 2I). Surprisingly, other domain combinations such as D2,4, D3,4 or D2,3,4 consistently led to WT levels of biofilm repression suggesting potential epistatic interactions between the domains (Fig. 2I). Finally, using Northern blots we determined the associations between these domains and RyhB repression. Individual domain mutants exhibit repression levels of RyhB that are consistent with those expected based on the biofilm phenotypes, with strains expressing D2 or D3 chimeras showing derepression of RyhB, whereas those expressing D1 or D4 exhibit WT OueS^C6706 levels of the iron-responsive sRNA (Fig. 2J). Furthermore, domain combination mutants show a similar pattern of RyhB repression as predicted based on biofilm data, with those domain mutants that repress biofilm showing WT OueS^C6706 levels of RyhB, and those forming robust biofilm formation exhibiting ∆ompU levels of the sRNA (Fig. 2K). Together, our data demonstrate that the 5’ module of OueS encodes specific domains that are responsible for the emergence of virulence-associated phenotypes.

Toxigenic OueS confers resistance against intestinal bacteriophages by activating the CBASS defense system

Our transcriptomic analyses of genes repressed by OueS successfully led us to identify the molecular mechanisms associated with biofilm repression by the sRNA. Next, we took a similar approach and examined the genes activated by the toxigenic allele of OueS to identify other potential phenotypes associated with the sRNA and its molecular mechanisms. Functional enrichment analysis of OueS^C6706 reveals a strong overrepresentation of genes involved in the ‘defense response to virus’ and ‘defense response to symbionts’ categories (Fig. 3A and Supplementary Data Table 3). Several subcategories belonging to viral defense pathways were the most significantly enriched functional categories (FDR < 0.001), indicating a non-random association with OueS^C6706 regulation (Fig. 3A and Supplementary Data Table 3). These observations led us to hypothesize that toxigenic OueS controls defense mechanisms against bacteriophages. Phages play a critical role in shaping V. cholerae intestinal infections and cholera outbreak dynamics^30,56–58. In order to test whether OueS modulates susceptibility to bacteriophages, we tested the survival of several OueS mutant strains in the presence of intestinal bacteriophages ICP1, ICP2 and ICP3^30,59,60. These phages were isolated from stool samples from cholera patients and utilize either OmpU (ICP2) or the O1 antigen of LPS (ICP1 and ICP3) as receptors^30,59,60. Phage ICP1 failed to generate plaques in the WT strain, likely due to phase variations in the poly(A) tract of the wbeL gene, as previously reported⁶¹. On the other hand, both ICP2 and ICP3 effectively formed plaques on the WT (Fig. 3B, C). As expected, the ∆ompU strain is resistant to ICP2 infection due to the absence of its OmpU receptor^30,60 (Fig. 3B). Ectopic expression of OueS^C6706 results in a marked reduction in plaquing efficiency for ICP2 ( ~1-log) (Fig. 3B). On the other hand, expression of OueS^mut or OueS^GBE1114 does not lead to a reduction in plaquing efficiency (Fig. 3B). Similarly, expression of OueS^C6706 results in a marked reduction in plaquing efficiency for ICP3 ( ~2-log), whereas expression of OueS^mut or OueS^GBE1114 does not lead to a reduction in plaquing efficiency (Fig. 3B). Together, our results demonstrate that toxigenic OueS confers allele-specific resistance to intestinal phages.

Fig. 3 — A Functional GO analysis^108,109 of OueS activated genes reveals enrichment of viral and symbiont defense categories. Plaque assays reveal that ectopic expression of the toxigenic OueS^C6706 allele in the WT strain confers resistance to *V. cholerae* bacteriophages (B) ICP2 and (C) ICP3, in contrast to its environmental counterpart OueS^GBE1114. D Read mapping plot for *glyA* locus. E Schematic of the cyclic oligonucleotide-based anti-phage signaling system (CBASS) operon: *capV* (VC0178)*, dncV* (VC0179)*, cap2* (VC0180), and *cap3* (VC0181). The genomic coordinates pertaining to *V. cholerae* El Tor N16961 genome are indicated. F OueS activates expression of the CBASS operon genes. TPM data was analyzed as in Fig. 2E from biological replicate transcriptome datasets. Deletion of genes in the CBASS operon abolishes the effect of OueS^C6706 on resistance against bacteriophages (G) ICP2 and (H) ICP3. We examined the role of OueS in the suppression of three potential repressors of the CBASS system. Plaque assays reveal that deletion of *glyA* in a ∆*ompU* background leads to resistance against bacteriophages (I) ICP2 and (J) ICP3. K Northern blots demonstrate decreased levels of *glyA* upon expression of OueS^C6706 but not OueS^GBE1114 or OueS^mut. Representative blot of three biological replicates are shown. L All 9 domain mutants are sensitive to ICP3, indicating that all of the OueS^C6706 domains are essential for protection against this intestinal bacteriophage. n ≥ 5 biological replicates for all efficiency of plating data. 2way ANOVA with Tukey’s multiple comparisons test p-value: * < 0.05, ** < 0.01, *** < 0.001, **** < 0.0001. Data are presented as mean ± s.e.m. p-values are in comparison to the WT, unless specified. Source data are provided as a Source Data file.

Small RNAs predominantly act as repressors of gene expression^62,63. Thus, we hypothesized that OueS^C6706 promotes phage resistance by repressing a negative regulator of phage defense systems. We interrogated the OueS MAPS dataset to identify candidate OueS^C6706 targets linked to phage resistance. The one potential candidate we uncovered is glyA (VC0941), which encodes a serine hydroxy methyltransferase and is a metabolic regulator of folate biosynthesis⁶⁴ (fold enrichment = 6.98, q-value 0.00001) (Fig. 3D and Table 1). GlyA is essential for folate synthesis, and reduced folate levels activate the cyclic oligonucleotide-based antiphage signaling system (CBASS) encoded on the Vibrio Seventh Pandemic Island 1 (Fig. 3E)⁶⁵. Consistent with this model, biofilm transcriptome analyses reveal that expression of the CBASS effector genes capV, dncV, and cap2 is reduced in the ∆ompU strain but restored to WT levels upon expression of OueS^C6706 (Fig. 3F). To determine the role of CBASS on OueS-mediated phage resistance, we constructed deletion mutants for each of these CBASS components and examined the effect of OueS^C6706 expression in these strains. Expression of OueS^C6706 in the ∆capV, ∆dncV, and ∆cap2 backgrounds does not reduce the efficiency of plating for ICP2, as seen when expressed in the WT background (Fig. 3G). Similarly, expression of OueS^C6706 in those CBASS deletion mutants does not reduce the plaquing efficiency of ICP3 (Fig. 3H). As expected, expression of OueS^GBE1114 in the CBASS mutants does not lead to a reduction in plating efficiency for either of the phages (Fig. 3G, H). Overall, our data indicate that the toxigenic allele of OueS confers resistance against intestinal bacteriophages ICP2 and ICP3 by activating the CBASS defense system.

OueS represses glyA conferring resistance against phages

Our MAPS analyses reveal glyA as an allele-specific target of OueS^C6706 with high confidence (fold enrichment = 6.98, q-value 0.00001) (Fig. 3D and Table 1). Next, we constructed deletion mutants of glyA in the WT and ∆ompU background and examined their resistance profiles against intestinal bacteriophages upon OueS expression. Deletion of glyA in the the ∆ompU background yields no plaques as it is the phage receptor (Fig. 3I). In contrast, in the WT background, glyA deletion confers resistance against ICP2 regardless of which allele of OueS is being expressed or whether OueS is functional (OueS^mut) (Fig. 3I). Similarly, ∆glyA and ∆ompU∆glyA strains exhibit reduced efficiency of plating of phage ICP3 irrespective of which allele of OueS is expressed (Fig. 3I). Our data indicates that GlyA acts downstream of OueS as a repressor of phage resistance against ICP2 and ICP3 and its removal relieves the need of OueS^C6706 expression to confer resistance. Next, we performed northern blot analyses to examine the potential direct interaction between OueS and GlyA (Fig. 3K). The blots reveal a reduction in glyA levels in strains expressing OueS^C6706 (Fig. 3K). On the other hand, expression of OueS^GBE1114 allele or OueS^mut does not alter the levels of glyA (Fig. 3K). Overall, our data shows that toxigenic OueS represses GlyA, conferring resistance against phages.

Finally, we determined the role of OueS domains in phage resistance. To do this, as in Fig. 2, we expressed the 9 chimeric OueS mutants in the ∆ompU background. Since the ∆ompU strain is naturally resistant to the ICP2 phage as it lacks its receptor, we focused our analyses on ICP3 (Fig. 3L). In contrast to biofilm repression, expression of any of the chimeric OueS individual or combination variants leads to efficiency of plating similar to OueS^GBE1114. Our data indicate that all four OueS domains within the 5’ module of OueS^C6706 are required to confer resistance against ICP3 (Fig. 3L).

OueS controls ToxR-mediated phenotypes

ToxR directly regulates the expression of ompU⁶⁶. Consequently, we investigated the potential regulatory relationship between ToxR and OueS. In order to test whether OueS expression is controlled by ToxR, we performed Northern blot analyses in a ∆toxR strain. Similar to a ∆ompU strain, OueS is undetectable in a ∆toxR background, indicating that the virulence regulator controls its expression (Fig. 4A). Next, we explored how this regulatory connection between ToxR and OueS leads to phenotypic outcomes. First, we examined the association between ToxR and the sRNA during biofilm formation. We determined that deletion of ToxR increases biofilm production similar to a ∆ompU mutant (Fig. 4B). Ectopic expression of either the ompU^C6706 ORF or OueS^C6706 in the ∆toxR strain restores biofilm repression to WT levels (Fig. 4B). This OueS-mediated repression is specific, as the WT phenotype cannot be rescued in a ∆toxR::OmpU^mut strain or by ectopic expression of the OueS 5’-mut, 3’-mut or FL-mut alleles or the NS region within the ompU ORF (Fig. 4B). Next, we examined the molecular mechanisms mediating biofilm suppression in ∆toxR background by studying the relationship of OueS and ToxR via RyhB (Fig. 4C). Northern blot analyses reveal that even in the ∆toxR background, OueS^C6706 represses RyhB in an allele-specific manner, indicating that ToxR-associated biofilm suppression is mediated by OueS-dependent RyhB repression (Fig. 4C). Pairwise comparisons of biofilm transcriptome data from WT and ∆toxR-pOueS^C6706 relative to ∆toxR reveal that 42.69% of the ToxR regulated genes during biofilm formation overlap with the OueS transcriptome (155 of 363 genes; absolute FC ≥ 2.0, q ≤ 0.01; Supplementary Data Table 5). The directionality of expression (activation Vs repression) in 149 of these 155 OueS-regulated genes (96%) is identical between the WT and ∆toxR-pOueS^C6706 strains, strongly suggesting that OueS restores a significant portion of the regulatory changes observed with the loss of ToxR function in biofilms (Supplementary Data Table 5). Overall, these findings indicate that ToxR controls expression of OueS, which regulates ToxR-mediated biofilm regulation via RyhB repression, highlighting a close connection between the sRNA and ToxR regulome.

Fig. 4 — A Northern blot analysis reveals that OueS expression is controlled by ToxR. 5S rRNA served as a loading control; representative of n = 3 biological replicates. B Ectopic expression of OueS in ∆*toxR* mutant represses biofilm formation. Strains harbor either the empty vector or encode OmpU or OueS variants. n ≥ 6; two-tailed T test p-value **** ≤ 0.0001. C Northern blots demonstrate increased RyhB levels in a ∆*toxR* mutant, which are drastically reduced upon expression of OueS^C6706 (n = 3). D PCA plot of transcriptome data shows that ∆*toxR* strains expressing OueS^C6706 cluster with WT, unlike OueS^GBE1114. E OueS^C6706 (dark gray slice) modulates the expression of 84.7% (493 genes) of the ToxR virulence regulon (581 genes) (absolute FC ≥ 2.0; q ≤ 0.01), whereas OueS^GBE1114 modulates 13.4% (dark outer circle). These targets exclude the *ctxAB* and *tcp* operons. F Functional enrichment analysis under AKI conditions shows that ToxR-activated categories (light green) are primarily modulated by OueS (dark green). G Motility assays in ∆*toxR* background. Expression of the OueS^C6706 restores motility to WT levels. n ≥ 6 biological replicates; two-tailed unpaired T test, **** p-value < 0.0001. H, I Mucin penetration assays (H) OueS^C6706 is essential for motility through a mucus barrier. I OueS domains show differential roles in penetration, suggesting epistatic interactions. Significance via one-way ANOVA with Tukey’s test; n ≥ 6; **** p-value < 0.0001. J Functional enrichment analysis of ToxR-repressed genes. Categories associated with nutrient utilization and metabolism are largely regulated by OueS (dark red). K Phenotypic microarrays (AUC) showing growth activated (green) or repressed (red) by specific compounds. Data reveal that ToxR utilizes OueS to repress precursors of carbon metabolism and nucleotide biosynthesis precursors via OueS. L Growth curves of mutant strains in the presence of OueS-repressed nutrient sources (n = 3 biological replicates). All data are mean ± s.e.m. Source data are provided as a Source Data file.

Toxigenic OueS modulates 84.7% of the ToxR regulome

ToxR controls the expression of hundreds of genes, however, it only directly binds to 39 promoters²¹. To date, the reason behind this remains puzzling. Given the molecular and phenotypic connection that we uncovered above, we hypothesized that ToxR-expression of OueS might be responsible for the modulation of some of these genes. First, in order to test this and further elucidate the relationship between OueS and ToxR and the role of the sRNA during the infectious process, we compared their transcriptomes under virulence-inducing conditions (AKI)^67,68. These in vitro conditions lead to the activation of the virulence cascade and the production of the major virulence factors associated with cholera symptoms, allowing us to examine the relationship of ToxR and OueS in a host-contextualized manner^67,68. Principal component analysis (PCA) of the RNA-Seq data shows a major shift in the transcriptome of the ∆toxR strain compared to WT by modulating the expression of 581 genes (400 activated, 181 repressed) (Fig. 4D and Supplementary Data Table 6). Ectopic expression of OueS^C6706 in the ∆toxR mutant (∆toxR-pOueS^C6706) results in a drastic shift in the transcriptome that closely resembles the WT (Fig. 4D and Supplementary Fig. S4). Interestingly, expression of the environmental allele OueS^GBE1114 in the ∆toxR strain does not restore the transcriptome to WT levels, indicating that this overlap between the ToxR and OueS transcriptomes is allele specific (Fig. 4D and Supplementary Fig. S4). To examine the extent of OueS^C6706 complementation, we determined the ratio of the fold change between the ToxR and OueS regulons [(toxR/WT) / (toxR-pOueS^C6706/WT)]. Our analysis reveals that OueS modulates the expression of at least 493 of the 581 genes within the ToxR regulon (84.7%) under AKI conditions (Fig. 4E, Supplementary Fig. S4 and Supplementary Data Table 7). Importantly, OueS^C6706 does not modify the expression of ctxAB or the TCP operon, which are activated by ToxR^18,22 (Fig. 4E, Supplementary Fig. S4 and Supplementary Data Table 7). Finally, to determine whether the OueS-mediated modulation of the ToxR regulome is associated with the emergence of toxigenic clones of V. cholerae, we examined the transcriptome of ∆toxR strains ectopically expressing the OueS^GBE1114 allele under AKI conditions. Similar pairwise comparisons [(toxR/WT) Vs (toxR-pOueS^GBE1114/WT)] reveal that, unlike the toxigenic allele, OueS^GBE1114 only modulates 13.4% (78 of 581 genes) of the ToxR regulome under AKI conditions (Fig. 4E, Supplementary Fig. S4 and Supplementary Data Table 7). Furthermore, OueS^C6706 also modulates the OmpU transcriptome under virulence-inducing conditions (Supplementary Information). Overall, the data shows that toxigenic OueS modulates the expression of 84.7% of the ToxR regulome, potentially acting as its functional surrogate.

OueS controls ToxR-activated mucin penetration in a domain-dependent manner

We performed enrichment analysis of the 581 genes of the ToxR regulome to determine the specific functional categories of genes modulated by OueS^C6706. The analyses reveal that OueS controls the ToxR-mediated upregulation of genes associated primarily with chemotaxis, flagellar synthesis and phosphorelay signaling (Fig. 4F and Supplementary Data Table 7). These include several gene clusters involved in chemotaxis (VC1394-1406, VCA1086-VCA1097; cheV (VC1602 and VC2202)), flagellar biosynthesis (VC2120-2144), motility-associated killing (VCA0880-VCA0883), and two-component systems (VC0069, VC1638-VC1639, VC1795) (Supplementary Data Table 7). Among other functions, V. cholerae uses motility and chemotaxis to penetrate the mucosal layer and adhere to the intestinal epithelium, which facilitates colonization^11,69. To initially elucidate the role of OueS^C6706 in this complex phenotype, first, we performed motility assays in soft agar supplemented with bicarbonate to simulate virulence-inducing conditions^67,68. Interestingly, ∆toxR and ∆ompU strains exhibit reduced motility compared to the WT, indicating a role for ToxR and OmpU in this process (Fig. 4G). Specifically, the WT strain generates a motility zone of 60.1 mm, whereas in ∆toxR it is reduced to 45.3 mm (Fig. 4G). Expression of the toxigenic allele of OueS in the ∆toxR background restores WT motility (61.2 mm), whereas strains expressing an environmental allele of OueS, OueS^GBE1114, show similar motility zones as ∆toxR (45.1 mm) (Fig. 4G). Finally, mutant versions of OueS^C6706 expressed either ectopically or in cis also show a reduced motility zone similar to ∆toxR (46.1 and 46.5 mm, respectively) (Fig. 4G). Overall, our results show that ToxR upregulates motility in V. cholerae via toxigenic OueS. Next, we investigated whether OueS^C6706-mediated motility translates into enhanced mucin penetration, a motility-dependent phenotype with a complex regulation that is essential during the infectious process^11,69–73. To do this, we used the column-based assays with porcine mucin, which have previously been used to examine mucin penetration in V. cholerae^11,71. Our data shows that compared to the WT (1.37% of input CFUs), ∆ompU and ∆toxR strains show reduced mucin penetration, 0.03% and 0.08%, respectively (Fig. 4H). This is the first time that ToxR has been shown to mediate mucin penetration. Expression of the toxigenic OueS^C6706 allele in the ∆toxR background leads to increased mucin penetration resembling WT levels (1.96%) (Fig. 4H). On the other hand, expression of OueS^GBE1114 or OueS^mut does not revert to WT phenotype, and strains exhibit mucin penetration levels similar to ∆toxR (0.02% and 0.04%, respectively) (Fig. 4H). Taken together, our data indicate that ToxR regulates a critical phenotype for intestinal colonization, mucin penetration, and the toxigenic allele of OueS orchestrates this. To shed light into the mechanistic basis of this complex phenotype, we investigated the role of OueS domains in mucin penetration (Fig. 4I). Substitution of OueS domains D1 or D2 led to a 0.07% and 0.05%, respectively of mucin penetration that resemble that of ∆toxR strains or OueS^GBE1114 (Fig. 4I). Replacement of domains D3 (1.13%) or D4 (1%) had no effect on mucin penetration like the WT, however, combinations of these domains in the form of D3,4 (0.09%) or all four domains (D1-4, 0.03%), lead to a phenotype similar to ∆toxR (Fig. 4I). Surprisingly, other domain combinations (D2,3, D2,4, or D2,3,4) led to unexpected patterns of mucin penetration (Fig. 4I). Specifically, despite D3 and D4 individually not altering mucin penetration, when combined with changes in D2 they yield to opposite outcomes: D2,3 resembles ∆toxR (0.03%) whereas D2,4 maintains WT levels (1.07%) of mucin penetration (Fig. 4I). Furthermore, counterintuitively, strain D2,3,4 exhibits WT levels of mucin penetration (1.4%) (Fig. 4I). These unexpected outcomes suggest complex epistatic interactions between the OueS domains controlling the emergence of this phenotype. Overall, our data indicate that ToxR controls mucin penetration via OueS in a domain-dependent manner.

ToxR mediates repression of nutrient utilization in an OueS-dependent manner

Functional enrichment analysis reveals that the genes downregulated by ToxR under AKI culture conditions are primarily involved in metabolic processes (Fig. 4J and Supplementary Data Table 6). Interestingly, OueS^C6706 largely restores the expression patterns in these GO categories (dark red), indicating that ToxR repression of these genes is mainly mediated via the sRNA (Fig. 4J and Supplementary Data Table 7). These include several metabolism-associated genes, including a fructose uptake system (VCA0517-VC0519) and an anaerobic glycerol phosphate dehydrogenase operon (VCA0747-VCA0749) (Supplementary Data Table 7). Given these transcriptional patterns, we hypothesized that ToxR might be involved in differential nutrient utilization in V. cholerae. In order to test this, we performed phenotypic microarrays of V. cholerae C6706 WT and ∆toxR strains. Based on the transcriptome data, we tested the ability of the strains to utilize 190 different carbon sources (Biolog Microplates PM1, 2 A), 59 sources of phosphorous, and 35 sources of sulfur (Biolog Microplate 4 A), as these plates contain compounds that are intermediaries in glucose metabolic pathways or are involved in translational processes (Fig. 4K). We compared the growth of the WT and ∆toxR mutant strains under these conditions and selected compounds where the ∆toxR strain exhibits at least a 1.5-fold greater area under the curve (AUC) than the WT. We did not identify any compounds that fit these criteria in the PM1 and PM2A plates (Supplementary Data Table 8). On the other hand, 21 compounds in the PM4A plate fit the criteria where the ∆toxR strain demonstrates a more robust growth than the WT strain, indicating that ToxR represses utilization of these compounds (Supplementary Data Table 8). Specifically, in agreement with the transcriptome data, 19 % of those compounds (4 of 21) are directly involved in pathways related to the metabolism of carbon compounds⁷⁴ (Fig. 4K). These include compounds that are involved in glucose and glycogen metabolism and hexosamine biosynthesis such as 2-phosphoglyceric acid, 3-phosphoglyceric acid, glucose-1-phosphate, and glucosamine-6-phosphate (∆toxR/WT AUC range 1.5-1.6) (Fig. 4K, L). Furthermore, the ∆toxR strain exhibits significantly enhanced growth in 7 nucleotide monophosphate compounds (33 %) that serve as precursors for nucleotide biosynthesis and cyclic dinucleotides and act as secondary messengers in bacteria⁷⁵ (Fig. 4K, 4L). These include 2’- 3’- and 5’ monophosphates of adenosine, guanine, uridine, thymidine and cytidine (∆toxR/WT AUC range 1.5 – 1.7; Supplementary Data Table 8). Given the extensive overlap between the ToxR and OueS^C6706 transcriptomes (Fig. 4D, E), we hypothesized that the sRNA is involved in regulating the utilization of these nutrient sources. To test this hypothesis, we examined the growth patterns of strains expressing either the toxigenic (C6706) or environmental (GBE1114) allele of OueS in the ∆toxR background in the same Microplates. Interestingly, ∆toxR strains expressing either allele of OueS display reduced growth in the compounds involved in both carbon utilization and nucleotide biosynthesis when compared to the parental ∆toxR strain, exhibiting a similar growth pattern as WT (OueS / WT AUC range 0.8 – 1.4) (Fig. 4K, L and Supplementary Data Table 8). Overall, our phenotypic and molecular data demonstrate that toxigenic OueS acts as a functional surrogate of ToxR, controlling ~84% of its regulome and regulates several novel ToxR-mediated phenotypes such as biofilm repression, mucin penetration, and repression of carbon utilization.

OueS is expressed from the ompU promoter

To identify a potential promoter of OueS, we used biofilm repression as a functional readout to initially monitor sRNA expression. First, we generated mutant strains where we cloned OueS together with varying lengths of its upstream region potentially harboring the promoter and integrated it at the native ompU locus in V. cholerae C6706 (Fig. 5A). We did so in 10 bp increments (P₀, P₁₀, P₂₀, … P₁₀₀) from zero bp upstream of OueS (P₀) to 100 bp upstream (P₁₀₀) and monitored biofilm formation of these 10 strains (Fig. 5A, B). All the constructs repress biofilm formation similar to the WT, indicating that OueS does not have a native promoter within ompU (Fig. 5A, B). One likely scenario that explains this data is that OueS is co-transcribed with the ompU mRNA transcript from the ompU promoter. To verify this, we generated an ompU promoter mutant by deleting the 300 bp region immediately upstream of the ompU start codon, which encodes both the promoter and transcription start site⁷⁶, generating the strain ∆P_ompU, which retains an intact ompU and OueS coding region (Fig. 5A). Biofilm formation by ∆P_ompU resembles the ∆ompU strain, suggesting a lack of OueS expression (Fig. 5B). To corroborate these findings, we performed northern blot analyses of ∆P_ompU to examine its role in OueS production. Similar to ∆ompU, ∆toxR or ∆hfq strains, OueS is undetectable in the ∆P_ompU strain (Figs. 5C, 1D). Overall, our results demonstrate that OueS expression is controlled by ToxR via the ompU promoter (Fig. 5C).

Fig. 5 — A Schematic of mutant constructs to identify the location of the OueS promoter. B Deletion of *ompU* promoter leads to derepression of biofilm formation. n = 6 biological replicates. Two-tailed unpaired T test p-value **** < 0.0001. C Northern blot analysis reveals that OueS expression is controlled via the *ompU* promoter. 5S rRNA probes are used as a loading control. D *V. cholerae* P_ompU-OueS shows similar OueS expression levels as WT. Expression was measured using qRT-PCR with OueS-specific primers, normalized to *recA*. n = 3 biological replicates (E) *V. cholerae* P_ompU-OueS represses biofilm formation at similar levels as WT. n ≥ 7 biological replicates. Two-tailed unpaired T test p-value **** <0.0001. Data are presented as mean ± s.e.m. F Expression of the toxigenic allele of OueS leads to successful intestinal colonization of *V. cholerae*. *V. cholerae* strains were orogastrically inoculated in 3–5-day old CD-1 mice to perform competition assays. Data is represented as competitive indices between the test strain and a *V. cholerae* C6706∆*lacZ* control strain. G Exogenous addition of s-IgAs and ICP3 bacteriophages highlights the role of OueS in the context of a host-relevant scenario. Strains that cannot repress biofilm and protect against phage predation via OueS exhibit a drastic colonization defect logs larger than in untreated mice. Data from untreated mice from panel (F) shown for comparison. n = 6 per strain. In panels (F, G), p-values indicate significance compared to the WT strain, unless otherwise specified; one-way ANOVA * < 0.01, ** < 0.001, *** < 0.0001, **** < 0.00001. Source data are provided as a Source Data file.

OueS is essential for successful intestinal colonization

OueS is essential for phenotypes associated with successful host colonization, such as biofilm repression or resistance against intestinal phages and controls a large percentage of the ToxR regulon. To examine whether OueS plays a direct role in intestinal colonization, we performed competition assays using the infant mouse model of infection^69,77,78. Inducer-dependent ectopic expression of the sRNA is not feasible in animal models. Nonetheless, our data demonstrates that both expression of OueS from directly from P_ompU (P_ompU-OueS^C6706, ~ 1-fold) and ectopic expression from an inducible promoter (pBAD-OueS^C6706, ~8-fold) are sufficient to recapitulate the WT biofilm phenotype (Fig. 5D, E and Supplementary Fig. S5), supporting the use of the P_ompU-driven strains for in vivo intestinal colonization assays. Additionally, we generated isogenic strains encoding mutant OueS variants directly driven by P_ompU at the native ompU locus, similar to the ectopic expression variants generated in Fig. 1. As above, we mutated every third nucleotide of the OueS sRNA locus that lies either exclusively within the ompU ORF (OueS 5’-mut), in the 3’ UTR (OueS 3’-mut) or across the full length of the sRNA (OueS FL-mut). The competitive fitness during intestinal colonization of these strains was determined against a V. cholerae C6706∆lacZ strain^31,69,78. The ability of the various mutants to colonize the intestine relative to the WT strain was calculated by determining their competitive indices (CI). As expected, ∆ompU demonstrates a ~1-log decrease in the CI of the small intestine (CI 0.06 – 0.14; Fig. 5F). Interestingly, the P_ompU-OueS^C6706 strain rescues the colonization deficiency associated with the loss of OmpU and exhibits CI similar to WT (CI 1.05 – 1.29) (Fig. 5F). These changes in colonization are specific to OueS as strains expressing the OueS 5’-mut (CI 0.15 – 0.44), 3’-mut (CI 0.17 – 0.48), FL-mut (CI 0.11 – 0.29) or OmpU^mut (CI 0.14 – 0.20) cannot restore WT CIs (Fig. 5F). Finally, the mutant with a deletion in the ompU promoter (∆P_ompU) shows a decrease in CI of the small intestine similar to the ∆ompU strain (CI 0.06 – 0.17; Fig. 5F).

The bottlenecks associated with phenotypes controlled by toxigenic OueS, e.g., biofilm repression and resistance against intestinal bacteriophages, are largely absent in infant mice. We hypothesized that the true role of OueS during intestinal colonization under host-relevant scenarios would surface in the presence those barriers. For this, we examined the fitness of OueS mutants after exposure to 0.5 mg/mL secretory IgAs (sIgA) in mice preexposed to 10⁷ PFU ICP3^43,79. sIgA has been shown to impede the colonization of V. cholerae strains that do not repress biofilm in a timely fashion; however, they are not produced in infant mice at levels that are representative to concentrations in the human gut⁴³. Similarly, bacteriophages like ICP3 were isolated from stool samples from patients with cholera and are naturally absent in our animal model⁸⁰. Our data shows that under host-relevant conditions, ∆ompU demonstrates a ~4-log decrease in the CI of the small intestine, a reduction that is 3-logs larger than in infant mice without sIgAs and ICP3 (CI 0.00009 - 0.003; Fig. 5G). Strikingly, expression of OueS^C6706 fully restores WT colonization levels (CI 0.79 - 1.06; Fig. 5G). This rescue in phenotype is specific as expression of OueS^mut (CI 0.0003 - 0.002) leads to a similar decrease in colonization as ∆ompU (Fig. 5G). Taken together, our results show that the ompU ORF encodes a sRNA, OueS, that is controlled by the master virulence regulator ToxR via the ompU promoter and is essential for successful intestinal colonization under host-relevant conditions.

OueS alleles exhibit a modular structure

Our data demonstrates that specific allelic variations of OueS are essential for several virulence-associated phenotypes to emerge, as evidenced by the large differences between the toxigenic allele OueS^C6706 and the environmental one (OueS^GBE1114). Interestingly, the sequence and structural divergences between those two alleles suggest a possible modular nature of the sRNA, as all the differences are confined to the 5’ end of the sRNA, the region encoded within the 3’ end of the ompU ORF (Fig. 1). To initially examine the potential modular nature of OueS, we examined its diversity within the V. cholerae species. Recently, we surveyed the natural genetic variability of ompU and identified, from over 1800 sequences analyzed, 41 alleles of the gene forming 14 major clades⁴¹. Sequence analysis of the OueS locus from these 14 clades reveals 11 alleles of OueS, as 3 of the 14 are identical (Supplementary Fig. S6). Subsequently, we analyzed the sequence diversity across the 11 OueS alleles (Fig. 6A). Consistent with our findings between OueS^C6706 and the environmental allele OueS^GBE1114, the 5’ region of OueS shows extensive variation among the different OueS alleles specifically in the region between nucleotides ~30 to ~100 (Fig. 6A, B). On the other hand, the 3’ region of OueS remains highly conserved (Fig. 6A, B). Two major insertion events give rise to distinct OueS sequence groups. Crucially, these insertions and the other sequence variations are confined to the 5′ region of OueS, which lies within the ompU ORF. Specifically, alleles (GBE0428, IRLE0081, GBE0917, GBE1116, IRLE0079) harbor either a ~30-nt or 18-nt insertion within the 5′ region, accompanied by shared and unique point mutations (Fig. 6B). The second group of alleles (FORC_055, IRLE0082, GBE1194) contain a distinct ~9-nt insertion and individual point mutations, again restricted to the 5′ region (Fig. 6B). In contrast, the 3′ region of OueS shows remarkable conservation with only three unique point mutations present in three of the 11 alleles, and no insertions or structural disruptions occur in this region (Fig. 6B). Phylogenetic analyses indicate that the alleles fall into several distinct clades, from which we selected representative alleles from each clade to perform subsequent phenotypic analyses (Fig. 6C). The modeled secondary structures of these OueS alleles show extensive structural variability at the 5’ region of the sRNA and a conserved stem-loop structure at the 3’ end with a poly-U tail that bears the signature of a Hfq-binding site (Fig. 6D). Overall, the diverse sequence and structural patterns of OueS suggest a modular nature of the sRNA, where the structural variation in the 5’ domain may contribute to regulatory specificity, while the conserved 3’ domain maintains core sRNA stability.

Modular heterogeneity in OueS contributes to pathogenic potential of Vibrio cholerae strains

Our results indicate that OueS has marked sequence and structural variations confined to its 5’ end (Fig. 6A, B), and, based on our analyses of OueS^C6706 and OueS^GBE1114, these variations lead to diverse phenotypic outcomes (Figs. 2, 3 and 4). We hypothesized that the sRNA has a modular structure that allows some alleles to differentially confer advantages during intestinal colonization, enhancing the pathogenic potential of toxigenic strains. To examine this, we first determined the distribution of the toxigenic OueS^C6706 allele in the V. cholerae species. Understanding this would highlight whether this is a preadaptation that existed before host colonization and present in numerous environmental strains, or likely evolved with the host and is confined exclusively to strains within the pandemic cholera group⁴¹. To address this, we first dereplicated over 1800 V. cholerae genomes deposited in public databases, which led to 378 clades⁴¹ (Fig. 6E). We then examined the presence of OueS^C6706 in those dereplicated clades and found that strains encoding the clinical allele of OueS are not confined to the PCG but are scattered across the phylogenetic tree, suggesting it evolved as a preadaptation to virulence (Fig. 6E, Supplementary Table S1). Besides the alleles highlighted in Fig. 6C, to examine whether a toxigenic-like OueS from a phylogenetically distant environmental strain confer the same properties to V. cholerae C6706, we selected the allele from the environmental strain V. cholerae IRLE0074, which we isolated previously from the Indian River Lagoon in Florida⁴¹ (Fig. 6E). Subsequently, to examine the association of these alleles (Fig. 6C, E) with the emergence of virulence traits in V. cholerae, we constructed six isogenic mutant strains in V. cholerae C6706 encoding the above described OueS alleles driven from P_ompU at the native ompU locus (P_ompU-OueS allele). These mutants cover the landscape of sequence and structural diversity of the sRNA (Fig. 5). To test whether their transcription levels were consistent in the toxigenic background, we examined their expression via qRT-PCR under AKI conditions (Fig. 6F). Our results indicate that all the alleles driven from P_ompU in the isogenic C6706 background are expressed at levels similar to OueS^C6706 (Fig. 6F).

Subsequently, we examined the role of modular heterogeneity of the OueS allelic mutants during intestinal colonization. Specifically, we determined the competitive indices of the isogenic OueS mutants using the infant mouse model of infection. Our assays reveal a heterogeneous colonization landscape mediated by the various OueS alleles. Mutants expressing the OueS allele from the environmental strain GBE1194 exhibit a CI higher to the WT strain (CI 1.4 – 1.5) whereas OueS^IRLE0074 bearing 100% sequence identity to OueS^C6706 resembles the WT strain (CI 0.8 – 1.2) (Fig. 6G). On the other hand, isogenic mutants encoding OueS from environmental strains GBE1114, GBE0428, IRLE0081, and GBE1116 exhibit a decrease in colonization (CI 0.45 – 0.56) (Fig. 6G). Next, we examined the colonization ability of the strain expressing the environmental OueS^GBE1114 allele after preexposure to sIgA and in the presence of ICP3 phage, as above (Fig. 6H). Unlike expression of OueS^C6706 that restored colonization ability of an ∆ompU strain, expression of OmpU^GBE1114 (CI 0.0002 - 0.006) or OueS^GBE1114 (CI 0.001 - 0.007) leads to a similar decrease in colonization as ∆ompU (Fig. 6H). Overall, our data indicates that a) the OueS allele encoded by toxigenic strains is a virulence preadaptation since some environmental strains (e.g., GBE1194, IRLE0074) encode alleles that also lead to successful colonization, and b) OueS is important for the emergence of pathogenic potential in V. cholerae, as strains like GBE1114 encode alleles that lead to defective intestinal colonization.

V. cholerae can chromosomally exchange the ompU locus in a modular fashion

To further elucidate the modular nature of OueS, we determined whether it is acquired as part of a unit within the ompU ORF (Fig. 7). For this, we adapted a chitin-induced natural transformation protocol to monitor ompU allele uptake⁸¹. We used locus-specific qRT-PCR to examine whether isogenic mutant strains encoding the environmental allele of ompU can acquire the toxigenic one (Fig. 7A). As donor DNA, we amplified the toxigenic ompU^C6706 allele with 1 kb flanking regions. As the recipient strain, we used the isogenic C6706::ompU^GBE1114. We cultured it in artificial sea water with chitin flakes (0.5% final) to induce natural competence and added donor DNA. Following incubation, we removed extracellular DNA, isolated genomic DNA and performed qRT-PCR targeting ompU^C6706 (Fig. 7B). The cycle threshold for DNA alone (blue line) is cT 31.2, whereas for V. cholerae C6706::ompU^GBE1114 alone is 30.5 (green line). qRT-PCR targeting ompU^C6706 reveal a pronounced shift in amplification kinetics in samples containing both chitin and donor DNA shifting to cT 23.2 (orange line) (Fig. 7B). This corresponds to a ~7-cycle increase in detectable ompU^C6706 signal (Fig. 7B). Next, we quantified DNA uptake efficiency by calculating transformation frequencies in the presence of chitin (Fig. 7C). Transformation frequencies (ratio of output/input CFUs) increase by 3-logs relative to control conditions from 10^-6 (limit of detection) to 10^-3 (Fig. 7C). This suggests that around 1 cell out of every 1000 has acquired the toxigenic allele of ompU. To corroborate the acquisition of the ompU^C6706 and its stable chromosomal replacement of the environmental ompU^GBE1114 allele, first, we exposed transformed cultures to 0.4% bile and plated to obtain individual colonies (Fig. 7D). Next, we screened 120 colonies using ompU^C6706-specific primers and identified three that were positive. Sequencing of the ompU locus of the three individual transformants revealed that all three underwent genomic replacement of the environmental ompU^GBE1114 allele with the toxigenic ompU^C6706 allele, what we termed revertants (Fig. 7D and Supplementary Fig. S7). Our results indicate that V. cholerae can chromosomally exchange the ompU ORF in a modular fashion.

Modular acquisition of toxigenic ompU leads to the emergence of virulence-associated traits

Next, we determined whether acquisition of the toxigenic ompU ORF leads to the emergence of the virulence traits associated with OmpU and OueS. First, we tested the survival of three independent revertant strains (Rev) in the presence of bile (Fig. 7E). Our results indicate that, unlike ∆ompU and C6706::ompU^GBE1114 (input strain), the revertant strains exhibit tolerance to 0.4% bile similar to the WT (Fig. 7E). Similarly, mucin penetration assays show that the revertants recover the WT phenotype, unlike the ∆ompU and C6706::ompU^GBE1114 that show over a 95% reduction (Fig. 7F). Similarly, revertant strains exhibit biofilm formation (Fig. 7G), resistance to the host-derived antimicrobial peptide P2 (Fig. 7H), and acidic pH tolerance (Fig. 7I) similar to WT. Finally, we examined the ability of revertant strains to compete against the WT using the infant mouse model of intestinal colonization. The three strains that acquired the toxigenic allele of the ompU ORF exhibit CIs comparable to WT (avg CI 0.88 - 1.6), whereas the parental strain C6706::ompU^GBE1114 and the ∆ompU strain (avg CI 0.08 – 0.1) display 1-log decrease in intestinal colonization (Fig. 7J). Overall, our results demonstrate that the modular acquisition of the toxigenic ompU ORF lead to the emergence of virulence traits in V. cholerae.

Environmental drivers control the modular acquisition of the ompU ORF

It is possible that exogenous drivers alter the modular acquisition rates of the toxigenic ompU ORF in V. cholerae populations. To test this, we modified the artificial seawater (ASW) media used during natural transformation experiments to reflect some exogenous conditions encountered by the bacterium in the aquatic environment or the digestive tract (Fig. 7K). Specifically, we examined the frequency of acquisition at different pH, in the presence of bile or antimicrobial peptides, or exposed to ICP3 bacteriophages. Our results indicate that pH plays a dual role in the acquisition of the ompU ORF. Specifically, pH 6.5 resulted in a 6.5-fold increase in transformation frequency compared to pH 7.5, which is the original pH of ASW (Fig. 7K). On the other hand, no measurable DNA uptake could be detected at pH 8.5, suggesting a fitness disadvantage for the acquisition of the toxigenic allele under alkaline pH (Fig. 7K). Exposure to 0.4% bile further enhances DNA uptake at pH 6.5 (11.1-fold), suggesting an additive effect of the two conditions (Fig. 7K). In addition to the antimicrobial peptide P2 to the ASW media leads to a 16.5-fold increase in transformation frequency (Fig. 7K). Interestingly, in the presence of bacteriophage ICP3, there was no significant difference in the DNA uptake frequency compared to ASW alone (Fig. 7K). Overall, our data shows that exogenous conditions modulate the acquisition of the ompU ORF.

Diverse transcriptomes lead to convergent phenotypes in OueS

Small RNAs regulate gene expression by binding target mRNAs, and differences in sequence or structure could potentially account for variations in their ability to direct colonization. However, we observed no correlation between the competitive index (CI) of the clinical and environmental OueS alleles and their size (166–196 nucleotides) or predicted secondary structure, suggesting that these features likely do not explain their differences in colonization ability^82,83 (Figs. 5, 6). Nonetheless, sequence variation analysis unequivocally reveals that OueS is bimodular, comprising of variable ompU-encoded 5’ modules and a conserved 3’ one, the former contributing to the emergence of pathogenic traits in V. cholerae (Figs. 5, 6). To correlate OueS modules with their transcriptional profiles, we used the infant mouse intestinal colonization data to classify the OueS alleles into toxigenic-like (GBE1194, IRLE0074) and environmental-like (GBE1114, GBE0428, GBE1116, IRLE0181) variants (Fig. 6). Analysis of the AKI transcriptomes of representative toxigenic-like (C6706, GBE1194) and environmental-like (GBE1114, GBE1116) OueS allele-expressing strains identified 96 genes (57 repressed and 39 activated) exclusively regulated by the toxigenic-like OueS alleles (Supplementary Fig. S8 and Supplementary Data Table 9). However, we did not find significant functional enrichment among the 96 genes, indicating that no single group of genes or pathway is over-represented in the dataset. This suggests a complex regulatory landscape, potentially involving regulatory redundancies and compensatory networks associated with the emergence of these phenotypes.

Allelic variability of OueS leads to fitness trade-offs between human infection and environmental survival

V. cholerae is a natural inhabitant of aquatic biomes and can be found associated with reservoirs that critically affect disease transmission⁸⁴. To examine the role of OueS in V. cholerae environmental survival and the potential fitness trade-offs associated with its toxigenic allele, we determined the CIs of OueS mutants during the colonization of two model environmental reservoirs established in the lab: the cyanobacteria Microcystis aeruginosa and the crustacean Artemia salina (Fig. 8A, C)⁸⁴. We performed competition assays in both model systems following a similar approach as above, where we compared the colonization potential of test strains (lacZ⁺) against a V. cholerae C6706 ∆lacZ control strain. Our results indicate that OmpU is important for colonization of both M. aeruginosa (Fig. 8B) and A. salina (Fig. 8D) as an ∆ompU mutant exhibits a ~0.6-log reduction in its CI (0.39 and 0.43, respectively). Expression of toxigenic OueS (OueS^C6706) leads to a significant increase in CI in both the cyanobacteria (CI 0.75) and crustacean (CI 0.76) model systems when compared to ∆ompU (Fig. 8B, D). Interestingly, the CI of strains expressing an environmental OueS allele (OueS^GBE1114) closely resembles that of the WT strain (CI 1.05 – 1.06) in both reservoirs (Fig. 8B, D). Furthermore, mutant strains expressing the environmental OmpU allele (OmpU^GBE1114), which encodes both the environmental porin and sRNA, outcompete the WT strain encoding its toxigenic versions in both reservoirs (CI 1.51 - 1.67) (Fig. 8B, D). Overall, our results indicate that OmpU is important for colonization of environmental reservoirs in an OueS-dependent manner and its toxigenic allele leads to fitness trade-offs when compared to its environmental counterpart.

Fig. 8 — A DIC and fluorescent (GFP + TRITC) images of *V. cholerae* C6706 in co-culture with *Microcystis aeruginosa*. B Competition assays of *V. cholerae* strains with *M. aeruginosa*. C Fluorescence microscopy image of *Artemia salina* infected with *V. cholerae* C6706 constitutively expressing GFP shows colonization of its gut space. D Competition assays of *V. cholerae* strains in *A. salina*. CIs in these reservoirs uncover the fitness trade-offs associated with the toxigenic OueS allele and the emergence of virulence traits in *V. cholerae*. Images are representative of n > 3 biological replicates. * p < 0.05, ** p < 0.005, ***p < 0.0005, **** p < 0.0001 determined by ordinary one-way ANOVA with Tukey’s multiple comparisons test and are presented as mean ± s.e.m. Source data are provided as a Source Data file.

Discussion

Emerging and re-emerging infectious diseases pose immense health and economic risks⁸⁵. To date, the evolutionary drivers that facilitate certain strains within a population to develop virulence traits and emerge as human pathogens remain largely unknown. This critically hinders our ability to develop effective strategies (e.g., policies, therapeutics, vaccines, etc.) against facultative pathogens in a timely fashion. Our previous results show that the presence of allelic variations in core and acquired genes explains the emergence of virulence traits in toxigenic V. cholerae isolates^31,40,41. The gene coding for the major outer membrane protein OmpU encodes these allelic variations in the form of virulence adaptive polymorphisms (VAPs) and plays an integral role in pathogen emergence^31,40,41. In this study, we demonstrate that the ompU ORF encodes a sRNA with a unique modular structure that plays a central role in the emergence of virulence traits in V. cholerae (Fig. 9). In contrast to other characterized 3’ UTR-derived sRNAs identified in V. cholerae, the majority of the 5’ end of the OueS locus is encoded within the ompU ORF and is highly diverse, whereas its 3’ terminus is conserved and lies in the UTR of the ORF⁵⁰. This atypical location confers a bimodular structure to OueS ultimately generating allelic variants that contribute to the emergence of virulence-associated phenotypes in some strains, a property that is unique among known sRNAs. Investigating the prevalence of this configuration will shed light on the role of sRNAs in the evolution of bacterial pathogens^48–50,86. The expression of OueS is driven from P_ompU and is regulated by ToxR, potentially generating one single transcript. It remains to be determined how the combined ompU-OueS transcript is processed to generate both the porin and sRNA, which could be mediated by ribonucleases such as RNase E or others^48,87. This dual output of the porin and the sRNA may be facilitated by the coupling of transcription-translation that allows for peptide synthesis before transcript processing by the RNase(s) or by the regulatory modulation of RNase activity on the combined ompU transcript. Interestingly, OueS expression is decoupled from ompU production in early stages of growth, peaking during the early stationary phase and drastically decreasing at later growth stages (Fig. 1). This phenomenon strongly points towards post-transcriptional regulation of the ompU transcript leading to modulation of OueS levels. In this study, using a controlled ectopic system (to capture subtle effects) and chromosomal promoter strains (to mimic realistic expression windows), we show that OueS function is consistent across a diverse expression gradient (Supplementary Fig. S5). Findings that elucidate whether the expression and processing rates of OueS are modulated in response to specific host signals during the infection process will provide critical insights into the mechanisms governing V. cholerae spatiotemporal colonization dynamics.

Fig. 9 — A OueS is a modular sRNA with four major domains. The 5’ module of OueS encodes four variable domains that are responsible for the emergence of essential virulence traits in *V. cholerae,* such as biofilm repression, resistance against intestinal bacteriophages, and mucin penetration. Our data uncovers potential epistatic interactions between these domains, leading to non-linear emergence of virulence phenotypes. B Schematic of OueS regulation. The virulence regulator ToxR controls the expression of hundreds of genes, including the *ompU* ORF via P_ompU. The ORF encodes both a) the OmpU porin (blue) and b) the OueS small RNA (red), which encompasses its 3’ terminus and UTR. OueS is expressed via P_ompU and acts as a functional surrogate of ToxR by modulating the expression of ~ 85% of its regulome. Created in BioRender. Balasubramanian, D. (2026) https://BioRender.com/pratvezC The *ompU* ORF encodes a modular porin and sRNA. *V. cholerae* encodes numerous alleles of *ompU* with distinct modular versions of OmpU and OueS structurally and functionally. Modularity in the porin and the sRNA enables the emergence of phenotypes essential for host colonization, such as antimicrobial resistance or stringent biofilm regulation.

We determined that the OueS allele from toxigenic strains controls phenotypes essential for host colonization and survival. For instance, it inhibits biofilm formation by directly repressing RyhB, a well-characterized sRNA controlled by Fur (Fig. 2)^52,54. Interactions between sRNAs to modulate gene expression is being increasingly recognized as a critical component governing cell morphology and behavior^88–92. For instance, expression of the major stress response regulator rpoS in Escherichia coli is reciprocally modulated by interactions between two sRNAs, ArcZ and CyaR⁹¹. Similarly, interactions between sRNAs modulate cell division and ethanol tolerance in some non-pathogens^88,89. To our knowledge, our finding that OueS regulates RyhB in an allele-dependent manner is the first identified example of an sRNA-sRNA interaction that directly affects host colonization and the emergence of pathogenic traits in bacteria. Recent high-throughput studies demonstrate potential to identify interactions between sRNAs on a whole genome scale, such as RNA sponges^90,92, opening avenues to unravel fascinating aspects of the intricate regulatory network in bacteria.

Bacteriophages are a key driver in the emergence of pandemic V. cholerae and play a critical role in its intestinal survival and colonization, environmental prevalence, and seasonality of cholera epidemics^{30,56–58,79,93}. Our data demonstrates that expression of the toxigenic allele of OueS confers resistance to two phages intimately associated with cholera dynamics: ICP2 and ICP3, both of which were isolated from stool samples of cholera patients (Fig. 3)⁸⁰. We determined that OueS mediates phage resistance via activation of the CBASS abortive infection system by directly repressing glyA expression (Fig. 3). The CBASS system was previously shown to confer resistance to circulating variants of the ICP3 phage, and thus, our findings provide a molecular and evolutionary mechanism explaining this phenotype⁹⁴. Interestingly, the OmpU porin, which is encoded in the ompU ORF together with OueS, is the cell receptor for ICP2⁸⁰. Our data reveals an ingenious mechanism developed by the cell to defend against phage infections by co-transcribing an sRNA that confers phage resistance along with the porin that functions as the receptor. Given how commonly phages utilize porins as cell receptors, we contend that examining the ubiquitousness of this arrangement will unveil fascinating nuances and novel mechanisms used by bacteria to resist viral infections. A cocktail of ICP phages have been proposed as a therapeutic intervention to prevent host colonization by the bacterium⁷⁹. Therefore, elucidating the associated resistance mechanisms will shed new light in the arms race between bacteria, phages and their receptors, which can have direct implications in preventing infections and cholera outbreaks.

The transcriptional regulator ToxR plays a key role in virulence and other physiological processes by modulating the expression of hundreds of genes in V. cholerae^20,21. Surprisingly, ToxR only binds directly to the promoter of 39 genes²¹. Here, we provide a mechanism that helps elucidate this conundrum, as 84% of the ToxR regulome can be explained by the expression of toxigenic OueS via activation of the ompU promoter by ToxR. For instance, we show that OueS mediates novel regulatory effects of ToxR such as biofilm repression and mucin penetration, two phenotypes that play a crucial role during host colonization^11,69,95 (Figs. 4, 5). It is possible that this is not an isolated example and other sRNAs act as functional surrogates of virulence regulators, suggesting a unique mechanism among bacterial pathogens. Elucidating this will shed light on the regulatory cascades of other DNA-binding regulators and provide novel downstream targets for therapeutic interventions. Interestingly, OueS does not modulate expression of CT or TCP, indicating a compartmentalized role of the sRNA within the ToxR regulome (Fig. 4). Furthermore, the transcriptome of the environmental OueS^GBE1114 allele does not overlap with the ToxR regulome, which highlights the close relationship between the evolution of the sRNA and the emergence of virulence traits in V. cholerae.

Our data shows that OueS is critical for successful intestinal colonization and complements the loss of the ompU ORF (Fig. 5). Interestingly, this phenomenon is not widespread among the different environmental alleles of OueS or exclusive to the toxigenic one (Fig. 6). Taken together, our analyses suggest that the OueS allele encoded by toxigenic strains is a preadaptation to virulence as a) it confers a competitive advantage during colonization and is responsible for virulence-associated traits (e.g., OueS^C6706 provides both), b) some environmental alleles of the gene do not confer these competitive advantages (e.g., OueS^GBE1114 or OueS^IRLE0181), c) other environmental alleles such as OueS^GBE1194 do, and d) exact same versions of the toxigenic allele are present in environmental strains and lead to a similar phenotype (OueS^IRLE0074) (Fig. 6).

V. cholerae is a natural inhabitant of aquatic biomes and is often associated with reservoirs that contribute to their environmental survival^84,96. We uncovered a fitness trade-off associated with the expression of OueS alleles and colonization of environmental reservoirs. Our results show that the toxigenic allele of OueS leads to fitness costs during environmental survival, as demonstrated by the strain encoding the environmental allele exhibiting a higher CI during colonization of both A. salina and M. aeruginosa (Fig. 8). It is tempting to hypothesize that these selective pressures function as bottlenecks restricting the emergence of pathogenic traits in environmental strains of V. cholerae. We recently uncovered that allelic variations of several core and acquired genes act as a critical bottleneck for cholera emergence^31,40,41. Interestingly, in many cases, the individual effects of these mutations during colonization were limited, however, they exhibit a synergistic and non-linear outcome when combined⁴⁰. We predict that a similar scenario might be at play regarding the fitness trade-offs of virulence traits. These fitness costs might explain the limited distribution of toxigenic isolates of V. cholerae and other human pathogens. Future research that delineates how the various biotic and abiotic factors in the estuarine environment influence V. cholerae colonization dynamics will be critical to understand the evolution of its virulence traits and host transmission potential.

Modularity has become a central concept in biology, stemming from the idea that biological systems are intricately linked networks that are typically divided into sets of interacting modular subsystems⁹⁷. Modularity spans many scales in biological systems and provides a cohesive mechanistic framework to explain biological complexity^98–101. Recently, we determined that the porin OmpU contains modular protein domains that uniquely contribute to the emergence of antimicrobial resistance in certain alleles⁴¹. Here, we unveil a new layer in the complex modular nature of the ompU ORF and demonstrate that OueS also displays modular characteristics and encode domains that give rise to virulence-associated traits (Figs. 7, 9). The 5’ module of OueS encodes four variable domains that are responsible for the emergence of essential virulence traits in V. cholerae, such as biofilm repression, resistance against intestinal bacteriophages, or mucin penetration. Our data uncovers potential epistatic interactions between these domains, leading to non-linear effects controlling the emergence of virulence phenotypes. Furthermore, OueS is a bimodular sRNA comprising of a) a variable 5’ module encoded within the ompU gene that contributes to diversity and emergence potential, and b) a highly conserved 3’ module downstream from the ompU ORF that likely plays a role in sRNA stability and target recognition by binding Hfq. This structural arrangement likely positions OueS at the intersection of two evolutionary forces: diversifying selection on the coding sequence of ompU and the downstream regulatory function of the sRNA. Variability in the 5’ module, shaped by changes in the ompU ORF, can introduce functional diversity into OueS without compromising the core regulatory architecture housed in the 3’ region. We contend that this modular structure provides flexibility for evolutionary exploration while preserving essential sRNA functions. Taken together, our results indicate that the ompU ORF encodes both a porin and a sRNA, and their modular nature contributes to the emergence of phenotypes essential for host colonization such as antimicrobial and phage resistance, stringent biofilm regulation, and colonization of environmental reservoirs (Figs. 7, 9). Our results support a scenario where a limited network of highly variable modular genes like ompU could lead to phenotypic heterogeneity in a population, including the emergence of virulence-associated traits. Elucidating the molecular processes and interaction dynamics between these genes will yield critical insights into the evolution of facultative pathogens. Overall, our findings provide mechanistic insights into the emergence of pathogenic traits in bacteria, unraveling the hidden genetics and fitness costs associated with pathogen emergence.

Methods

Ethics statement

All animal procedures were conducted in accordance with the guidelines and protocols approved by the Institutional Animal Care and Use Committee (IACUC) at the University of Central Florida (#IPROTO202300049) and St. Jude Children’s Research Hospital (#3311). The facility is certified with the AAALAC International Accreditation.

Bacterial strains, media, and growth conditions

All V. cholerae strains used in this study (Supplementary Data Table 10) are derivatives of the seventh pandemic El Tor biotype strain C6706, first isolated from a patient in Peru in 1961¹⁰², except for V. cholerae GBE1114, an environmental isolate from the Great Bay Estuary, New Hampshire⁴². E. coli S17 λ-pir strains harboring a chromosomally integrated RP4 plasmid and tra genes¹⁰³ were used for conjugation of the deletion constructs into V. cholerae as well as for routine cloning purposes. Phage propagation was performed on V. cholerae AC6169, a BSL1 derivative of the El Tor strain E7946. The following media were used for bacterial cultivation: Lysogeny broth (LB) broth (10 g l⁻¹ tryptone, 5 g l⁻¹ yeast extract, 5 g l⁻¹ NaCl); Biofilm assays were performed in tryptone broth (tryptone 10 g l⁻¹, NaCl 5 g l⁻¹); AKI broth was used for virulence-inducing conditions^67,68 (15 g l⁻¹ Bacto^TM peptone, 4 g l⁻¹ yeast extract, 5 g l⁻¹ NaCl, 0.03% w/v sodium bicarbonate). Plates were used at a final concentration of 1.5% agar. Top agar for phage assays contained LB broth + 0.7% agar. Where necessary, the following antibiotics or supplements were used: streptomycin 1000 µg ml⁻¹, kanamycin 45 µg ml⁻¹, polymyxin B 50 µ ml⁻¹, carbenicillin 50 µg ml⁻¹, 2,2’-bipyridyl 100 µg ml⁻¹, FeCl₃ 100 µg ml⁻¹, 5-bromo-4-chloro-3-indolyl-ß-D-galactopyranoside (X-gal) 40 µg ml⁻¹, and arabinose (0.01%). Reagents were procured from Sigma.

For routine culture, E. coli and V. cholerae were grown in LB broth at 37 °C with aeration (200 rpm). Biofilm and growth curve assays were performed in 96-well flat-bottom, tissue culture-treated plates (CytoOne) incubated at 37 °C.

Cloning, vectors, and strain construction

In-frame deletion mutants of ompU, toxR, ryhB, and lacZ were constructed via homologous recombination¹⁰⁴. Briefly, approximately 500 bp fragments flanking the gene of interest to be deleted were PCR amplified, cloned as BglII-NotI (upstream fragment) or NotI-EcoRI (downstream fragment) in tandem into the V. cholerae suicide vector pKAS154¹⁰⁴, and electroporated into E. coli S17λ-pir¹⁰³. Deletion constructs of genes of CBASS (capV, dncV, cap2) were generated using the HiFi DNA Assembly (NEB) to prevent polar effects of the deletions within the operon. E. coli strains harboring the deletion constructs were then conjugated with V. cholerae C6706, and allelic exchange was carried out with appropriate antibiotic selection, as described¹⁰⁴. Positive clones were first screened by colony PCR and then confirmed by sequencing the deletion loci from genomic DNA using flanking primers. The ompU promoter mutant ∆P_ompU, was generated in a similar manner by amplifying the regions immediately flanking the 300 bp locus right before the ompU start codon, generating a deletion construct in pKAS154, followed by allelic exchange into C6706, as described above. The OmpU^mut strain with an intact OmpU amino acid sequence but a mutated OueS and driven by P_ompU was generated by making wobble mutations across the entire length of the OueS sequence (Supplementary Fig. S2). The fragment was then synthesized commercially (Twist Biosciences) along with 500 bp of the ompU downstream region and cloned as a BglII-EcoRI fragment into pKAS154. Allelic exchange with WT C6706 was then performed as described above.

Ectopic OueS-expressing clones and non-specific ompU locus clones were generated by PCR amplification from the appropriate genomic DNA template using specific primers listed in Supplementary Data Table 10. The amplicons were then cloned as EcoRI-HindIII fragments into pBAD22, an expression vector that utilizes an arabinose-based induction system (P_BAD) to drive expression of the cloned fragment. The OueS domain mutants were synthesized commercially (Twist Biosciences) and subsequently cloned into pBAD22, as above. After the introduction into E. coli by electroporation and sequence confirmation of the inserts, the plasmids were electroporated into the appropriate V. cholerae strains, followed by selection on carbenicillin-containing LB plates. For strains harboring the empty vector, plasmid pBAD22 was electroporated into the appropriate V. cholerae strain and selected as above. Ectopic ompU-expressing clones were generated in a similar manner.

Chromosomal promoter fusions were constructed as described above for the deletion constructs. The ~500 bp region upstream of the ompU ORF, including P_ompU was first cloned as an XbaI-SacI fragment into pKAS154 to generate plasmid pDBS467. Subsequently, the various OueS constructs: OueS 5’-mut, 3’-mut, FL-mut fragments (commercially synthesized; Twist Biosciences), and the various OueS alleles (amplified from the respective genomic DNA) were cloned into pDBS467 downstream of P_ompU as a SacI-EcoRI fragment. Allelic exchange was then performed as described above.

All deletion and complementation constructs in E. coli were Sanger sequenced before introduction into V. cholerae. Genomic DNA was isolated from all V. cholerae deletion and chromosomal fusion strains, and the region(s) of interest amplified using flaking primers for confirmation by Sanger sequencing.

RNA isolation, transcriptome and qRT-PCR studies

For Northern blots and RACE studies, RNA was isolated using the hot-phenol extraction method at the appropriate growth stage and conditions^105–107. Briefly, the cell pellets were harvested and resuspended in cell lysis solution (0.5% SDS, 20 mM sodium acetate, 10 mM EDTA), followed immediately by the addition of 500 µL of acid-phenol:chloroform (Ambion) preheated to 65 °C and vortexed. After incubation at 65 °C for 10 min, the samples were centrifuged at 21,000 x g for 10 min and the aqueous layer extracted twice with acid-phenol:chloroform. The aqueous phase was then extracted twice with chloroform and precipitated overnight at − 80 °C with 2.5 volumes of ice-cold absolute ethanol. The total RNA was then pelleted and washed with cold 70% ethanol, pellets dried and resuspended in 100 µl RNase-free water containing 40 U murine RNase inhibitor (New England Biolabs). Following quantification by Nanodrop, 50 µg of the total RNA was DNase-treated (Ambion RNase-free DNase) following the manufacturer protocols. The RNA was reextracted as above and precipitated with 1/10^th volume 3 M sodium acetate and cold absolute ethanol, followed by a cold 70% ethanol wash. Final resuspension of the dried RNA pellets was in 50 µL of RNase-free water containing RNase inhibitor (as above). Size fractionation for RACE studies was performed by running 10 µg of hot phenol-isolated RNA on a precast 1.25% MOPS agarose gel (Lonza) in 1x MOPS buffer (Quality Biological) with the Riboruler RNA ladders (Thermo) for size determination. Gel extraction was performed using the Zymoclean gel RNA recovery kit (Zymoresearch).

For transcriptome studies, RNA was isolated using a commercial kit (DirectZol RNA Miniprep kit, Zymo Research) following the manufacturer instructions. DNase treatment, rRNA depletion, library preparation and Illumina NextSeq2000 sequencing (2 × 51 bp reads) was performed at SeqCenter, LLC (Pittsburgh, PA). Demultiplexing, quality control and adapter trimming was performed with bcl-convert (v3.9.3). CLC Genomics Workbench (Qiagen) was used for subsequent data processing. Briefly, paired Illumina reads were mapped to the El Tor reference genome (GenBank AE003852.1- Chromosome I; AE003853.1- chromosome II) using the following settings to generate the transcripts per million bases (TPM) values: mismatch cost 2, insertion / deletion cost 3, length / similarity fraction 0.8, global strand-specific alignment. Subsequent pairwise analysis between two datasets of interest (absolute fold change ≥ 2.0 filtered on average expression for FDR correction ≤ 0.01) were used for downstream analysis. Functional enrichment analysis of the transcriptome data was performed by the Panther Overrepresentation Test against the Biological processes dataset of the GO Ontology Database (Released 07-01-2022)^108,109.

For qRT-PCR studies, total RNA was isolated using the hot phenol method and size fractionated (for OueS expression, as above). Subsequent cDNA synthesis was performed with 500 ng total RNA per sample using the SuperScript III Reverse Transcriptase with random hexamers (Invitrogen), per manufacturer protocols with 1 U Murine RNase Inhibitor (NEB). The qRT-PCR assay was performed using the PowerUp SYBR Green Master Mix (Applied Biosystems) following manufacturer instructions with 25 ng of input cDNA per reaction on a QuantStudio 5 cycler (Applied Biosystems). The biological triplicate data was normalized using recA gene expression as the housekeeping control and compared to expression in the WT strain. The primers used are listed in Supplementary Data Table 10.

Biofilm assays

The ability of the strains to form biofilms on surfaces was determined using the crystal violet-based assays in 96-well polystyrene plates, optimized from published protocols¹¹⁰. Briefly, the strains from LB plates were grown in tryptone broth with carbenicillin for 14–16 h at 37 °C with aeration. The cultures were then diluted 1:500 in fresh tryptone broth supplemented with carbenicillin and arabinose, and 200 µL dispensed in wells of the 96-well plates. After static incubation for 24h at 37 °C, the cultures were poured off, and the plates gently washed three times with water to remove unattached cells. The surface-attached cells were then stained with 225 µL per well of crystal violet (0.1% in water) for 20 minutes at room temperature and washed thoroughly with water. The attached stain was subsequently eluted with 225 µL per well of 50% acetic acid for 15 min and quantified by reading the absorbance in a Tecan Sunrise plate reader at 595 nm. To examine the effect of iron on biofilm formation, TB media was supplemented with either 100 µM FeCl₃ (MP Biomedicals) for iron repletion or 100 µM 2, 2’-bipyridyl (Sigma) to deplete residual iron. Growth and processing of biofilms was performed as described above. For transcriptome studies, 24 h biofilms were harvested by scrapping off attached cells into fresh media and used for total RNA isolation.

Northern blot assays

For northern blot assays, total RNA isolated using the hot-phenol method (see above) were used. Briefly, 6 - 10 µg RNA per lane was first resolved on a 1.2% agarose gel (Lonza) in 1 x MOPS buffer (Quality Biological) and transferred to BrightStar Plus positively charged nylon membranes (Invitrogen) by electroblotting (Trans-Blot SD transfer cell, BioRad). Following transfer, the membranes were UV crosslinked (UV Crosslinker, Fisherbrand), prehybridized for 5 h (ULTRAhyb Ultrasensitive Hybridization buffer, Invitrogen) in roller bottles in a UVP Hybridizer Oven (Analytik Jena) at 42 ˚C. Commercially synthesized 5’ biotinylated probes (10 pmoles of 40-mer probes; Eurofins Genomics or IDT) for the RNA targets were used for hybridization for 15 h at 42 ˚C, washed with low- and high-stringency buffers (Invitrogen), and developed using the Chemiluminescence Nucleic Acid Detection kit (Thermo), following manufacturer instructions. Developed blots were imaged on a Bio-Rad ChemiDoc MP imaging system.

Rapid Amplification of cDNA Ends (RACE)

The 5’ and 3’ termini of the sRNA OueS were determined by the RACE assay (SMARTer RACE 5’/3’ kit, Takara Bio USA), adapted for non-poly-A-tailed RNA. Kit components were used, unless otherwise mentioned. Briefly, total RNA was separated on a 1.2% MOPS agarose gel, and the 100–500 nt region was gel extracted (Zymoclean Gel RNA recovery kit). This size fractionated RNA was used as the template for both 5’ and 3’ RACE studies.

For 5’ RACE, first-strand cDNA synthesis was performed by random priming, followed by annealing the SMARTer IIA oligonucleotide to the 5’ ends of the newly synthesized cDNA fragments. This was followed by PCR amplification using a forward primer specific to the 5’-attached oligonucleotide and a reverse primer within the ompU gene (DB196: 5’ GATTACGCCAAGCTTtcgtaacgtagaccgatagccagttcgtct 3’). The fragments were then cloned non-directionally using the A-overhangs generated during the final PCR step and electroporated into E. coli Stellar Competent cells (Clontech) and selected on carbenicillin plates. Plasmids from positive clones were then sequenced to identify the 5’ terminus of OueS. For 3’ RACE, the size-fractionated total RNA was poly-A-tailed using the E. coli Poly(A) polymerase following manufacturer instructions (New England Biolabs). cDNA was then synthesized using a primer containing a complementary poly-T sequence and a 5’ overhang (DB164: 5’ GGGGGGGGAATTCTTTTTTTTTTTT 3’). The newly synthesized cDNA was used as template for a second- strand synthesis using an ompU-specific primer that binds inside the 3’ end of the ompU ORF (DB195: 5’ GATTACGCCAAGCTTccgctcttacatctcttaccagttcaatctgctag 3’). The amplicons were then gel purified, and PCR amplified using the 3’ ompU-specific primer and a primer targeting the 5’ overhang of the poly-T primer (DB164) used in the first step of the process. The resulting amplicons were then cloned non-directionally into vectors with a T-overhang and electroporated into E. coli. Subsequently, plasmids from positive clones were sequenced to define the 3’ terminus of OueS.

Prediction of sRNA secondary structure

The secondary structures of the WT and mutant versions and alleles of OueS were determined using mFold on the UNAFold web server (RNA Folding Form V2.3)¹¹¹ using default parameters. Structures with the least predicted ∆G values were selected.

MS2 affinity purification coupled with RNA-sequencing (MAPS)

MAPS was performed to identify RNA targets of OueS modified from established methods¹¹². Briefly, 100 mL of ∆ompU strains expressing ms2 aptamer-tagged OueS^C6706, OueS^GBE1114, or OueS^mut, and the respective untagged versions were cultured under iron-limiting conditions (LB + 100 µM bipyridyl) and 0.01% arabinose at 37 °C, 220 rpm for 16 h. Cultures were rapidly chilled, harvested by centrifugation at 8000 x g for 10 min at 4 °C, resuspended in 15 mL Buffer A (20 mM Tris-HCl, 150 mM KCl, 1 mM MgCl₂, 1 mM DTT, 1 mM PMSF) and lysed using a French Press (Glen Mills). Lysates were clarified by centrifugation at 16000 x g for 30’ at 4 °C and immediately used for affinity purification. The phage MS2 coat protein fused to maltose-binding protein (MS2-MBP) was purified using Ni-affinity chromatography (HisPur Ni-NTA Purification Kit, Thermo Scientific) and immobilized on amylose resin (New England Biolabs), following manufacturer protocols. The clarified lysates were incubated with the pre-equilibrated resin to capture ms2-tagged OueS-RNA complexes, followed by extensive washing to remove nonspecific interactions. Bound complexes were eluted with buffer A + 150 mM maltose, and RNA was extracted using acid phenol-chloroform. Enrichment of OueS in pulldown fractions was confirmed by qRT-PCR using OueS-specific primers targeting the 3′ region of the sRNA. Purified RNA was DNase-treated (RNase-free DNase, Invitrogen) and library preparation was performed using Illumina’s Stranded Total RNA Prep Ligation with Ribo-Zero Plus kit and 10-bp unique dual indices, followed by sequencing on NovaSeq X Plus (at SeqCenter LLC). Sequencing reads were mapped to the V. cholerae genome, and transcripts enriched in ms2-tagged samples relative to untagged controls were identified using default settings (CLC Genomics Workbench). Targets were then filtered for significance ( ≥ 2.0, q-value < 0.01) and the datasets of the three sRNA variants were compared in order to define OueS^C6706 allele-specific target profiles.

OueS conservation mapping, phylogenetic distribution, and sequence alignment

The conservation mapping of OueS alleles, as well as the identification of the clinical-like OueS alleles in the dereplicated genomes, were performed as described in our previous publication⁴⁰. The sequences of the 11 OueS alleles were used to identify the median sequence identity and plotted as described⁴⁰. The maximum likelihood phylogenetic tree of SNPs constructed from our previous analysis of 1804 V. cholerae genomes⁴⁰ was used to map the distribution of the clinical OueS alleles among the 378 dereplicated clades.

Sequence alignment of the OueS alleles were performed using the default progressive alignment algorithm on CLC Genomics Workbench (Qiagen) or Geneious v7.1.3.

Growth kinetics under iron limitation

Overnight LB-grown cultures were diluted 1:1000 in LB broth with or without supplementation with FeCl₃ (iron-replete) or 2,2’-bipyridyl (iron-deplete). 200 µL aliquots of the cultures dispensed in flat-bottom 96-well plates (tissue culture-treated, CytoOne) were then monitored for OD₅₉₅ every 30 minutes for 24 hours at 37 °C in a Tecan Sunrise plate reader. Each biological replicate had at least six technical replicates, and the data for each condition were averaged across technical and biological replicates, corrected to the baseline (LB) absorbance values.

Bacteriophage assays

Phages ICP2 and ICP3 were propagated on V. cholerae AC6169, a modified non-toxigenic El Tor strain that has a phase-locked expression of the O1 antigen (A. Camilli). Briefly, log phase cells (OD600 of 0.5-0.7) in LB broth were infected at a multiplicity of infection of 0.1 with the respective phages and allowed propagation for 3–5 h, 200 RPM at 37 ˚C. Phage lysates were harvested by centrifugation and filtered through a 0.2 µ filter before use. Top agar overlays were performed using media containing the indicator host strain supplemented with carbenicillin and arabinose, and the phage dilutions were spot titrated. The efficiency of plating (EOP) was calculated for ICP2 and ICP3 lysates: EOP = (titer on test strain / titer on WT C6706).

Motility assays

Motility was evaluated using established protocols¹¹³ on AKI media with 0.03% (w/v) sodium bicarbonate supplemented with 0.3% agar, and carbenicillin and arabinose for plasmid maintenance and induction, respectively. Diameter measurements were taken 24 h post-inoculation after incubation at 37 ˚C (n ≥ 6).

Phenotypic microarrays

Biolog phenotype microarrays were performed following established protocols¹¹⁴. Briefly, V. cholerae was suspended from fresh plates to 42% transmittance in 1x Biolog IF-0a media (Biolog turbidimeter, 20-mm diameter tube). The inoculum was generated by diluting the cell suspension in IF-0a media plus Biolog Redox Dye Mix D (Biolog, Hayward, CA), supplemented with carbenicillin and arabinose (final transmittance of 85 %). The assays were performed in Biolog Microplates containing 190 sources of carbon (PM1 and PM2A), 59 sources of phosphorus and 35 sources of sulfur (PM4A). The plates were inoculated with 100 μl/well of the cell suspension mixture and incubated at 37 ˚C with shaking in a TECAN Sunrise microplate reader (Tecan, US). Optical density was measured at 595 nm every hour for 24 h. The data was analyzed using Microsoft Excel and GraphPad Prism. As per manufacturer recommendations, the assays were performed in duplicate.

Chitin-induced natural transformation and transformation frequency

Natural transformation was performed using a chitin-dependent protocol modified from established methods⁸¹. Briefly, V. cholerae C6706::ompU^GBE1114 was grown overnight at 30 °C in LB broth and subcultured into fresh LB to mid-log phase (OD₆₀₀ 0.6). Cells were pelleted and resuspended in 0.5 x Instant Ocean (IO; 7 g/L sea salts) to an OD₆₀₀ of 1.0. Transformation mixtures were prepared by combining 150 µL of sterile chitin slurry (8 g shrimp shell chitin suspended in 150 mL 0.5 x IO), 750 µL 0.5 x IO, and 100 µL of resuspended cells in sterile 2.0 mL microcentrifuge tubes. Cultures were incubated static at 30 °C for 16–24 h to induce competence (with pH adjustment and supplementation with bile, 20 µg/mL P2, or phage ICP3 added as required). Following incubation, approximately 550 µL of supernatant was carefully removed without disturbing the chitin, and 3 µg of WT V. cholerae C6706 ompU DNA was added. Reactions were further incubated static overnight at 30 ˚C. The following day, the supernatant was gently removed, and the chitin was washed extensively with 0.5 × IO to eliminate extracellular DNA. To collect cells attached to the chitin, 1 mL of 0.5 x IO was added and vortexed vigorously. Positive transformants were recovered by treating cells with 0.4% bile for 1.5 h, plated to obtain single colonies on LB plates, screened by colony PCR and subsequent sequencing of the ompU locus. Transformation frequencies were determined by treating parallel reactions with DNase I for 30 min prior to DNA extraction. Genomic DNA was isolated, and qPCR was performed using WT ompU-specific primers. Quantification was based on a standard curve generated from gDNA samples isolated from cultures containing known numbers of bacterial cells.

Mucin penetration

Previously established protocols were followed¹¹. Briefly, mucin columns were prepared by adding 1% mucin from porcine stomach (Sigma) into 1-ml syringes. Mid-log phase bacterial cultures (100 µl) in PBS were carefully loaded onto the top of the mucin columns and incubated at 37 °C for 30 min to allow settling. 500 µL fractions were then collected from the bottom of the columns. Bacterial counts, represented as CFU/mL were determined by serial dilution and plating on LB agar.

P2 and low pH survival assays

Assays were performed as described previously⁴¹. Overnight cultures were subcultured 1:100 into fresh LB and grown to an optical density at 600 nm (OD₆₀₀) of 0.5. Cells were washed twice and resuspended in LB. The resuspended cultures were adjusted to either LB at pH 5.5 or LB supplemented with 200 µg/mL P2 and incubated at 37 °C with shaking for 1.5 h. Colony-forming units (CFU) were determined by plating serial dilutions on LB agar.

Infant mouse competition assays

For in vivo animal experiments, three-to-five-day-old CD-1 ® IGS mice (Strain code 022; Charles River Laboratories) of both genders were randomly selected from mixed litters for all animal experiments. Mice were housed under the following conditions: temperature 68˚ - 72 ˚F, humidity 40–60%, under a 12 h light-dark cycle. Infant mice were housed with the mothers, monitored under the care of full-time staff, and weaned 2 h prior to infection.

Intestinal colonization competition assays in three-to-five day-old infant mice were performed with LB-grown lacZ⁺ cells of the WT, deletion mutants, and OueS alleles competed against an isogenic WT strain harboring a lacZ deletion, essentially as described^31,69,78. The colonization efficiencies of the different strains in the small intestine harvested at ~22 hours post infection are represented as competitive indices⁶⁹. Briefly, strains were inoculated from fresh LB plates into LB broth, incubated for 12 h at 37 °C with aeration, and diluted 1:1000 fold in fresh LB broth. The test (lacZ⁺) and control (lacZ^-) strains were mixed 1:1 to obtain ~10⁶ CFU/mL. 50 µL of the bacterial mixture, corresponding to ~10⁵ CFUs, were then used to intragastrically inoculate 3–5-day-old infant mice and maintained at 30 ˚C for 22 h. The exact input CFU numbers were determined by plating dilutions on LB-Sm-X-Gal plates. Approximately 22 h post infection, the mice were euthanized, and their small intestines transferred into 4 mL LB-10% glycerol stored on ice. The samples were then processed for CFU by homogenizing the tissues, serially diluting in 1x PBS (pH 7.4) and plating appropriate dilutions on LB-Sm-X-Gal plates. In vitro competition experiments were performed in parallel by inoculating 50 µL of the input CFU in LB-Sm broth and incubating for 22 h at 37 ˚C with aeration. Samples were then serially diluted and plated as for the in vivo samples above. Results are presented as the competitive index (CI), which is the ratio of the intestinal lacZ⁺ CFU (blue colonies) to lacZ^- CFU (white colonies), normalized to the input CFU ratio and in vitro competition CFU ratio.

For experiments involving sIgA and ICP3, previously established protocols were followed^43,79. Briefly, 3-day old mice were orogastrically dosed with phage ICP3 (10⁷ PFU/50 µL dose) diluted in 2.5% sodium bicarbonate for 3 hours. The bacterial inoculum was prepared by exposing V. cholerae cells prepared as above to human-derived secretory IgA (sIgA; Millipore Sigma) at a final concentration of 0.5 mg/mL for 1 h at 37 ˚C. Post-exposure, cells were diluted to achieve an infectious dose of 10⁵ CFU/50 µL and infected as above.

Colonization of environmental reservoirs

Bacterial inoculum for infections of the environmental hosts were prepared essentially as described for the mice colonization assays. Infections for fluorescent visualization were performed with a GFP-tagged V. cholerae C6706 strain, and images obtained using a Carl Zeiss Axio Observer 7 Inverted Microscope. a) Artemia salina. Briefly, eggs of Artemia salina were hatched in artificial sea water supplemented with 2% sodium chloride (ASW), incubated at room temperature and aeration for 48 hours. Infections were performed in 96-well flat-bottom plates with each well containing 200 µL ASW with 20–30 individual crustaceans and 10⁷ CFU/mL of bacterial mixture. After 48 h incubation at room temperature, samples were washed twice in ASW and either homogenized for determining CIs (as above) or used directly for microscopy. b) Microcystis aeruginosa. We followed modifications of previously established protocols¹¹⁵: Microcystis aeruginosa (UTEX LB 2385) were incubated in flasks containing BG-11 media (Gibco) under a 12 h light-dark cycle with aeration (100 rpm) for about 14 days. The cyanobacterial cells then were centrifuged, resuspended in media and co-cultured at a final OD₇₀₀ of 0.6 with 10⁵ CFU/mL V. cholerae cells in a 200 µL reaction volume. CIs were determined after 24 h incubation by plating dilutions on LB-Sm-X-Gal plates, as above.

Statistical analyses

All data were analyzed for statistical significance using Prism 10 software (GraphPad Prism), except for the transcriptome and MAPS data that were analyzed within the CLC Genomics Workbench (Qiagen). Unless otherwise indicated, all experiments were performed with at least five independent biological replicates (n ≥ 5) and the aggregate data (mean or median ± s.d.) are shown.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Supplementary information

Supplementary Information^{(3.9MB, pdf)}

41467_2026_73137_MOESM2_ESM.pdf^{(34.5KB, pdf)}

Description of Additional Supplementary Files

Supplementary Dataset 1^{(26.1KB, xlsx)}

Supplementary Dataset 2^{(105KB, xlsx)}

Supplementary Dataset 3^{(19.1KB, xlsx)}

Supplementary Dataset 4^{(21.3KB, xlsx)}

Supplementary Dataset 5^{(25.5KB, xlsx)}

Supplementary Dataset 6^{(36.6KB, xlsx)}

Supplementary Dataset 7^{(135.5KB, xlsx)}

Supplementary Dataset 8^{(38.5KB, xlsx)}

Supplementary Dataset 9^{(153.3KB, xlsx)}

Supplementary Dataset 10^{(23.9KB, xlsx)}

Supplementary Dataset 11^{(82.3KB, xlsx)}

Supplementary Dataset 12^{(36.8KB, xlsx)}

Reporting Summary^{(981.6KB, pdf)}

Transparent Peer Review file^{(571.4KB, pdf)}

Source data

source data^{(15MB, xlsx)}

Acknowledgements

The authors would like to thank the anonymous reviewers for their thoughtful comments. We thank Dr. Andrew Camilli (Tufts University School of Medicine) for generously providing the ICP phages and the host V. cholerae AC6163 strain, Dr. Mario López-Pérez (Universidad Miguel Hernandez) for his comments on the manuscript, and Alicia Campos- López for her technical assistance.

Author contributions

S.A.M. designed the research. D.B., Q.P., and C.C. performed research. D.B., Q.P., C.C., and S.A.M. analyzed data and wrote the manuscript.

Peer review

Peer review information

Nature Communications thanks Bjoern Voss, Matthew Waldor, and the other anonymous reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Funding

S.A.M discloses support for the research of this work from National Science Foundation (NSF) CAREER award (#2045671), Burroughs Wellcome Investigator in the Pathogenesis of Infectious Disease (#1021977), and ALSAC at St. Jude Children’s Research Hospital. D.B., Q.P., and C.C. declare no relevant funding.

Data availability

The transcriptome data generated in this study have been deposited in the NCBI GEO database under accession numbers GSE272767 (GSM8410380 – GSM8410418; https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE272767) and GSE272640 (GSM8408188 - GSM8408195; https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE272640). Source data are provided in this paper.

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

The online version contains supplementary material available at 10.1038/s41467-026-73137-6.

References

1.Balasubramanian, D., Lopez-Perez, M., Grant, T. A., Ogbunugafor, C. B. & Almagro-Moreno, S. Molecular mechanisms and drivers of pathogen emergence. Trends Microbiol30, 898–911 (2022). [DOI] [PubMed] [Google Scholar]
2.Diard, M. & Hardt, W. D. Evolution of bacterial virulence. FEMS Microbiol. Rev.41, 679–697 (2017). [DOI] [PubMed] [Google Scholar]
3.Martinez, J. L. Bacterial pathogens: from natural ecosystems to human hosts. Environ. Microbiol.15, 325–333 (2013). [DOI] [PubMed] [Google Scholar]
4.Bonneaud, C., Weinert, L. A. & Kuijper, B. Understanding the emergence of bacterial pathogens in novel hosts. Philos. Trans. R. Soc. Lond. B Biol. Sci.374, 20180328 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Kanungo, S., Azman, A. S., Ramamurthy, T., Deen, J. & Dutta, S. Cholera. Lancet399, 1429–1440 (2022). [DOI] [PubMed] [Google Scholar]
6.Balasubramanian, D., Murcia, S., Ogbunugafor, C. B., Gavilan, R. & Almagro-Moreno, S. Cholera dynamics: lessons from an epidemic. J. Med. Microbiol. 70, 10.1099/jmm.0.001298 (2021). [DOI] [PubMed]
7.Provenzano, D., Schuhmacher, D. A., Barker, J. L. & Klose, K. E. The virulence regulatory protein ToxR mediates enhanced bile resistance in Vibrio cholerae and other pathogenic Vibrio species. Infect. Immun.68, 1491–1497 (2000). [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Mathur, J. & Waldor, M. K. The Vibrio cholerae ToxR-regulated porin OmpU confers resistance to antimicrobial peptides. Infect. Immun.72, 3577–3583 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Nalin, D. R. et al. Cholera, non-vibrio cholera, and stomach acid. Lancet2, 856–859 (1978). [DOI] [PubMed] [Google Scholar]
10.Sack, G. H. Jr. et al. Gastric acidity in cholera and noncholera diarrhoea. Bull. World Health Organ47, 31–36 (1972). [PMC free article] [PubMed] [Google Scholar]
11.Liu, Z. et al. Mucosal penetration primes Vibrio cholerae for host colonization by repressing quorum sensing. Proc. Natl. Acad. Sci. USA105, 9769–9774 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Freter, R., O’Brien, P. C. & Macsai, M. S. Role of chemotaxis in the association of motile bacteria with intestinal mucosa: in vivo studies. Infect. Immun.34, 234–240 (1981). [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Faruque, S. M. et al. Transmissibility of cholera: in vivo-formed biofilms and their relationship to infectivity and persistence in the environment. Proc. Natl. Acad. Sci. USA103, 6350–6355 (2006). [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Brown, R. C. & Taylor, R. K. Organization of tcp, acf, and toxT genes within a ToxT-dependent operon. Mol. Microbiol.16, 425–439 (1995). [DOI] [PubMed] [Google Scholar]
15.DiRita, V. J., Parsot, C., Jander, G. & Mekalanos, J. J. Regulatory cascade controls virulence in Vibrio cholerae. Proc. Natl. Acad. Sci. USA88, 5403–5407 (1991). [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Krukonis, E. S., Yu, R. R. & Dirita, V. J. The Vibrio cholerae ToxR/TcpP/ToxT virulence cascade: distinct roles for two membrane-localized transcriptional activators on a single promoter. Mol. Microbiol.38, 67–84 (2000). [DOI] [PubMed] [Google Scholar]
17.Murley, Y. M., Carroll, P. A., Skorupski, K., Taylor, R. K. & Calderwood, S. B. Differential transcription of the tcpPH operon confers biotype-specific control of the Vibrio cholerae ToxR virulence regulon. Infect. Immun.67, 5117–5123 (1999). [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Miller, V. L., Taylor, R. K. & Mekalanos, J. J. Cholera toxin transcriptional activator ToxR is a transmembrane DNA binding protein. Cell48, 271–279 (1987). [DOI] [PubMed] [Google Scholar]
19.Provenzano, D. & Klose, K. E. Altered expression of the ToxR-regulated porins OmpU and OmpT diminishes Vibrio cholerae bile resistance, virulence factor expression, and intestinal colonization. Proc. Natl. Acad. Sci. USA97, 10220–10224 (2000). [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Bina, J. et al. ToxR regulon of Vibrio cholerae and its expression in vibrios shed by cholera patients. Proc. Natl. Acad. Sci. USA100, 2801–2806 (2003). [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Kazi, M. I., Conrado, A. R., Mey, A. R., Payne, S. M. & Davies, B. W. ToxR antagonizes H-NS regulation of horizontally acquired genes to drive host colonization. PLoS Pathog.12, e1005570 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Miller, V. L. & Mekalanos, J. J. Synthesis of cholera toxin is positively regulated at the transcriptional level by ToxR. Proc. Natl. Acad. Sci. USA81, 3471–3475 (1984). [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Miller, V. L. & Mekalanos, J. J. A novel suicide vector and its use in construction of insertion mutations: osmoregulation of outer membrane proteins and virulence determinants in Vibrio cholerae requires ToxR. J. Bacteriol.170, 2575–2583 (1988). [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Wibbenmeyer, J. A., Provenzano, D., Landry, C. F., Klose, K. E. & Delcour, A. H. Vibrio cholerae OmpU and OmpT porins are differentially affected by bile. Infect. Immun.70, 121–126 (2002). [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Chakrabarti, S. R., Chaudhuri, K., Sen, K. & Das, J. Porins of Vibrio cholerae: purification and characterization of OmpU. J. Bacteriol.178, 524–530 (1996). [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Provenzano, D., Lauriano, C. M. & Klose, K. E. Characterization of the role of the ToxR-modulated outer membrane porins OmpU and OmpT in Vibrio cholerae virulence. J. Bacteriol.183, 3652–3662 (2001). [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Merrell, D. S., Bailey, C., Kaper, J. B. & Camilli, A. The ToxR-mediated organic acid tolerance response of Vibrio cholerae requires OmpU. J. Bacteriol.183, 2746–2754 (2001). [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Sakharwade, S. C. & Mukhopadhaya, A. Vibrio cholerae porin OmpU induces LPS tolerance by attenuating TLR-mediated signaling. Mol. Immunol.68, 312–324 (2015). [DOI] [PubMed] [Google Scholar]
29.Khan, J., Sharma, P. K. & Mukhopadhaya, A. Vibrio cholerae porin OmpU mediates M1-polarization of macrophages/monocytes via TLR1/TLR2 activation. Immunobiology220, 1199–1209 (2015). [DOI] [PubMed] [Google Scholar]
30.Seed, K. D. et al. Evolutionary consequences of intra-patient phage predation on microbial populations. Elife3, e03497 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Shapiro, B. J., Levade, I., Kovacikova, G., Taylor, R. K. & Almagro-Moreno, S. Origins of pandemic Vibrio cholerae from environmental gene pools. Nat. Microbiol.2, 16240 (2016). [DOI] [PubMed] [Google Scholar]
32.Bernardini, M. L., Sanna, M. G., Fontaine, A. & Sansonetti, P. J. OmpC is involved in invasion of epithelial cells by Shigella flexneri. Infect. Immun.61, 3625–3635 (1993). [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Chevalier, S. et al. Structure, function and regulation of Pseudomonas aeruginosa porins. FEMS Microbiol. Rev.41, 698–722 (2017). [DOI] [PubMed] [Google Scholar]
34.Devoy, A. F., Dyet, K. H. & Martin, D. R. Stability of PorA during a meningococcal disease epidemic. J. Clin. Microbiol.43, 832–837 (2005). [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Galdiero, M. et al. Porins from Salmonella enterica serovar Typhimurium activate the transcription factors activating protein 1 and NF-kappaB through the Raf-1-mitogen-activated protein kinase cascade. Infect. Immun.70, 558–568 (2002). [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Massari, P., King, C. A., Ho, A. Y. & Wetzler, L. M. Neisserial PorB is translocated to the mitochondria of HeLa cells infected with Neisseria meningitidis and protects cells from apoptosis. Cell Microbiol.5, 99–109 (2003). [DOI] [PubMed] [Google Scholar]
37.Baine, W. B. et al. Epidemiology of cholera in Italy in 1973. Lancet2, 1370–1374 (1974). [DOI] [PubMed] [Google Scholar]
38.Cvjetanovic, B. & Barua, D. The seventh pandemic of cholera. Nature239, 137–138 (1972). [DOI] [PubMed] [Google Scholar]
39.Devault, A. M. et al. Second-pandemic strain of Vibrio cholerae from the Philadelphia cholera outbreak of 1849. N. Engl. J. Med.370, 334–340 (2014). [DOI] [PubMed] [Google Scholar]
40.López-Perez, M. et al. Allelic variations and gene cluster modularity act as nonlinear bottlenecks for cholera emergence. Proc. Natl. Acad. Sci. USA122, e2417915122 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Grant, T. A., Lopez-Perez, M., Haro-Moreno, J. M. & Almagro-Moreno, S. Allelic diversity uncovers protein domains contributing to the emergence of antimicrobial resistance. PLoS Genet.19, e1010490 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Schuster, B. M. et al. Ecology and genetic structure of a northern temperate Vibrio cholerae population related to toxigenic isolates. Appl. Environ. Microbiol.77, 7568–7575 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Hsiao, A., Liu, Z., Joelsson, A. & Zhu, J. Vibrio cholerae virulence regulator-coordinated evasion of host immunity. Proc. Natl. Acad. Sci. USA103, 14542–14547 (2006). [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Mathur, J., Davis, B. M. & Waldor, M. K. Antimicrobial peptides activate the Vibrio cholerae sigmaE regulon through an OmpU-dependent signalling pathway. Mol. Microbiol.63, 848–858 (2007). [DOI] [PubMed] [Google Scholar]
45.Liu, J. M. et al. Experimental discovery of sRNAs in Vibrio cholerae by direct cloning, 5S/tRNA depletion and parallel sequencing. Nucleic Acids Res.37, e46 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
46.Updegrove, T. B., Zhang, A. & Storz, G. Hfq: the flexible RNA matchmaker. Curr. Opin. Microbiol.30, 133–138 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
47.Sauer, E. & Weichenrieder, O. Structural basis for RNA 3’-end recognition by Hfq. Proc. Natl. Acad. Sci. USA108, 13065–13070 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
48.Davis, B. M. & Waldor, M. K. RNase E-dependent processing stabilizes MicX, a Vibrio cholerae sRNA. Mol. Microbiol.65, 373–385 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]
49.Papenfort, K., Forstner, K. U., Cong, J. P., Sharma, C. M. & Bassler, B. L. Differential RNA-seq of Vibrio cholerae identifies the VqmR small RNA as a regulator of biofilm formation. Proc. Natl. Acad. Sci. USA112, E766–E775 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
50.Hoyos, M., Huber, M., Forstner, K. U. & Papenfort, K. Gene autoregulation by 3’ UTR-derived bacterial small RNAs. Elife 9, 10.7554/elife.58836 (2020). [DOI] [PMC free article] [PubMed]
51.Fong, J. C. & Yildiz, F. H. The rbmBCDEF gene cluster modulates development of rugose colony morphology and biofilm formation in Vibrio cholerae. J. Bacteriol.189, 2319–2330 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]
52.Mey, A. R., Craig, S. A. & Payne, S. M. Characterization of Vibrio cholerae RyhB: the RyhB regulon and role of ryhB in biofilm formation. Infect. Immun.73, 5706–5719 (2005). [DOI] [PMC free article] [PubMed] [Google Scholar]
53.Gao, H. et al. Fur represses Vibrio cholerae biofilm formation via direct regulation of vieSAB, cdgD, vpsU, and vpsA-K transcription. Front. Microbiol.11, 587159 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
54.Davies, B. W., Bogard, R. W. & Mekalanos, J. J. Mapping the regulon of Vibrio cholerae ferric uptake regulator expands its known network of gene regulation. Proc. Natl. Acad. Sci. USA108, 12467–12472 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
55.Lalaouna, D. et al. A 3’ external transcribed spacer in a tRNA transcript acts as a sponge for small RNAs to prevent transcriptional noise. Mol. Cell58, 393–405 (2015). [DOI] [PubMed] [Google Scholar]
56.Faruque, S. M. et al. Seasonal epidemics of cholera inversely correlate with the prevalence of environmental cholera phages. Proc. Natl. Acad. Sci. USA102, 1702–1707 (2005). [DOI] [PMC free article] [PubMed] [Google Scholar]
57.Nelson, E. J. et al. Transmission of Vibrio cholerae is antagonized by lytic phage and entry into the aquatic environment. PLoS Pathog.4, e1000187 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
58.Zahid, M. S. et al. Effect of phage on the infectivity of Vibrio cholerae and emergence of genetic variants. Infect. Immun.76, 5266–5273 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
59.Beckman, D. A. & Waters, C. M. Vibrio cholerae phage ICP3 requires O1 antigen for infection. Infect. Immun.91, e0002623 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
60.Lim, A. N. W., Yen, M., Seed, K. D., Lazinski, D. W. & Camilli, A. A tail fiber protein and a receptor-binding protein mediate ICP2 bacteriophage interactions with Vibrio cholerae OmpU. J. Bacteriol.203, e0014121 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
61.Seed, K. D. et al. Phase variable O antigen biosynthetic genes control expression of the major protective antigen and bacteriophage receptor in Vibrio cholerae O1. PLoS Pathog.8, e1002917 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
62.Gottesman, S. & Storz, G. Bacterial small RNA regulators: versatile roles and rapidly evolving variations. Cold Spring Harb Perspect Biol 310.1101/cshperspect.a003798 (2011). [DOI] [PMC free article] [PubMed]
63.Storz, G., Vogel, J. & Wassarman, K. M. Regulation by small RNAs in bacteria: expanding frontiers. Mol. Cell43, 880–891 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
64.Severin, G. B. et al. Activation of a Vibrio cholerae CBASS anti-phage system by quorum sensing and folate depletion. mBio14, e0087523 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
65.Cohen, D. et al. Cyclic GMP-AMP signalling protects bacteria against viral infection. Nature574, 691–695 (2019). [DOI] [PubMed] [Google Scholar]
66.Crawford, J. A., Kaper, J. B. & DiRita, V. J. Analysis of ToxR-dependent transcription activation of ompU, the gene encoding a major envelope protein in Vibrio cholerae. Mol. Microbiol.29, 235–246 (1998). [DOI] [PubMed] [Google Scholar]
67.Iwanaga, M. & Yamamoto, K. New medium for the production of cholera toxin by Vibrio cholerae O1 biotype El Tor. J. Clin. Microbiol.22, 405–408 (1985). [DOI] [PMC free article] [PubMed] [Google Scholar]
68.Iwanaga, M. et al. Culture conditions for stimulating cholera toxin production by Vibrio cholerae O1 El Tor. Microbiol. Immunol.30, 1075–1083 (1986). [DOI] [PubMed] [Google Scholar]
69.Millet, Y. A. et al. Insights into Vibrio cholerae intestinal colonization from monitoring fluorescently labeled bacteria. PLoS Pathog.10, e1004405 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
70.Frederick, A., Huang, Y., Pu, M. & Rowe-Magnus, D. A. Vibrio cholerae Type VI activity alters motility behavior in mucin. J. Bacteriol. 202, 10.1128/jb.00261-20 (2020). [DOI] [PMC free article] [PubMed]
71.Liu, Z. et al. Vibrio cholerae represses polysaccharide synthesis to promote motility in mucosa. Infect. Immun.83, 1114–1121 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
72.Silva, A. J., Pham, K. & Benitez, J. A. Haemagglutinin/protease expression and mucin gel penetration in El Tor biotype Vibrio cholerae. Microbiology149, 1883–1891 (2003). [DOI] [PubMed] [Google Scholar]
73.Xu, J. et al. The role of morphological adaptability in Vibrio cholerae’s motility. mBio16, e0246924 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
74.Nothaft, H. & Szymanski, C. M. Protein glycosylation in bacteria: sweeter than ever. Nat. Rev. Microbiol.8, 765–778 (2010). [DOI] [PubMed] [Google Scholar]
75.Goncheva, M. I., Chin, D. & Heinrichs, D. E. Nucleotide biosynthesis: the base of bacterial pathogenesis. Trends Microbiol.30, 793–804 (2022). [DOI] [PubMed] [Google Scholar]
76.Sperandio, V. et al. Cloning and characterization of the gene encoding the OmpU outer membrane protein of Vibrio cholerae. Infect. Immun.64, 5406–5409 (1996). [DOI] [PMC free article] [PubMed] [Google Scholar]
77.Klose, K. E. The suckling mouse model of cholera. Trends Microbiol.8, 189–191 (2000). [DOI] [PubMed] [Google Scholar]
78.Matson, J. S. Infant mouse model of Vibrio cholerae infection and colonization. Methods Mol. Biol.1839, 147–152 (2018). [DOI] [PubMed] [Google Scholar]
79.Yen, M., Cairns, L. S. & Camilli, A. A cocktail of three virulent bacteriophages prevents Vibrio cholerae infection in animal models. Nat. Commun.8, 14187 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
80.Seed, K. D. et al. Evidence of a dominant lineage of Vibrio cholerae-specific lytic bacteriophages shed by cholera patients over a 10-year period in Dhaka, Bangladesh. mBio2, e00334–00310 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
81.Meibom, K. L., Blokesch, M., Dolganov, N. A., Wu, C. Y. & Schoolnik, G. K. Chitin induces natural competence in Vibrio cholerae. Science310, 1824–1827 (2005). [DOI] [PubMed] [Google Scholar]
82.Bardill, J. P. & Hammer, B. K. Non-coding sRNAs regulate virulence in the bacterial pathogen Vibrio cholerae. RNA Biol.9, 392–401 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
83.Perez-Reytor, D. et al. Role of non-coding regulatory RNA in the virulence of human pathogenic vibrios. Front. Microbiol.7, 2160 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
84.Almagro-Moreno, S. & Taylor, R. K. Cholera: Environmental reservoirs and impact on disease transmission. Microbiol. Spectr. 1, 10.1128/microbiolspec.oh-0003-2012 (2013). [DOI] [PubMed]
85.Bloom, D. E., Black, S. & Rappuoli, R. Emerging infectious diseases: A proactive approach. Proc. Natl. Acad. Sci. USA114, 4055–4059 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
86.Djapgne, L. & Oglesby, A. G. Impacts of small RNAs and their chaperones on bacterial pathogenicity. Front. Cell Infect. Microbiol11, 604511 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
87.Bandyra, K. J. & Luisi, B. F. RNase E and the high-fidelity orchestration of RNA metabolism. Microbiol. Spectr 6, 10.1128/microbiolspec.RWR-0008-2017 (2018). [DOI] [PMC free article] [PubMed]
88.Grutzner, J. et al. sRNA-mediated RNA processing regulates bacterial cell division. Nucleic Acids Res.49, 7035–7052 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
89.Han, R. et al. Multiple small RNAs interact to co-regulate ethanol tolerance in Zymomonas mobilis. Front. Bioeng. Biotechnol.8, 155 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
90.Iosub, I. A. et al. Hfq CLASH uncovers sRNA-target interaction networks linked to nutrient availability adaptation. Elife 9, 10.7554/elife.54655 (2020). [DOI] [PMC free article] [PubMed]
91.Kim, W. & Lee, Y. Mechanism for coordinate regulation of rpoS by sRNA-sRNA interaction in Escherichia coli. RNA Biol.17, 176–187 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
92.Liu, F. et al. In vivo RNA interactome profiling reveals 3’UTR-processed small RNA targeting a central regulatory hub. Nat. Commun.14, 8106 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
93.Silva-Valenzuela, C. A. & Camilli, A. Niche adaptation limits bacteriophage predation of Vibrio cholerae in a nutrient-poor aquatic environment. Proc. Natl. Acad. Sci. USA116, 1627–1632 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
94.O’Hara, B. J., Alam, M. & Ng, W. L. The Vibrio cholerae Seventh Pandemic Islands act in tandem to defend against a circulating phage. PLoS Genet.18, e1010250 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
95.Silva, A. J. & Benitez, J. A. Vibrio cholerae biofilms and cholera pathogenesis. PLoS Negl. Trop. Dis.10, e0004330 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
96.Vezzulli, L., Pruzzo, C., Huq, A. & Colwell, R. R. Environmental reservoirs of Vibrio cholerae and their role in cholera. Environ. Microbiol. Rep.2, 27–33 (2010). [DOI] [PubMed] [Google Scholar]
97.Hatleberg, W. L. & Hinman, V. F. Modularity and hierarchy in biological systems: Using gene regulatory networks to understand evolutionary change. Curr. Top. Dev. Biol.141, 39–73 (2021). [DOI] [PubMed] [Google Scholar]
98.Litvin, O., Causton, H. C., Chen, B. J. & Pe’er, D. Modularity and interactions in the genetics of gene expression. Proc. Natl. Acad. Sci. USA106, 6441–6446 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
99.Hartwell, L. H., Hopfield, J. J., Leibler, S. & Murray, A. W. From molecular to modular cell biology. Nature402, C47–C52 (1999). [DOI] [PubMed] [Google Scholar]
100.Montoya, J. M., Pimm, S. L. & Sole, R. V. Ecological networks and their fragility. Nature442, 259–264 (2006). [DOI] [PubMed] [Google Scholar]
101.Krause, A. E., Frank, K. A., Mason, D. M., Ulanowicz, R. E. & Taylor, W. W. Compartments revealed in food-web structure. Nature426, 282–285 (2003). [DOI] [PubMed] [Google Scholar]
102.Thelin, K. H. & Taylor, R. K. Toxin-coregulated pilus, but not mannose-sensitive hemagglutinin, is required for colonization by Vibrio cholerae O1 El Tor biotype and O139 strains. Infect. Immun.64, 2853–2856 (1996). [DOI] [PMC free article] [PubMed] [Google Scholar]
103.Strand, T. A., Lale, R., Degnes, K. F., Lando, M. & Valla, S. A new and improved host-independent plasmid system for RK2-based conjugal transfer. PLoS ONE9, e90372 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
104.Skorupski, K. & Taylor, R. K. Positive selection vectors for allelic exchange. Gene169, 47–52 (1996). [DOI] [PubMed] [Google Scholar]
105.Rio, D. C., Ares, M. Jr., Hannon, G. J. & Nilsen, T. W. Purification of RNA by SDS solubilization and phenol extraction. Cold Spring Harb. Protoc.2010, pdb prot5438 (2010). [DOI] [PubMed] [Google Scholar]
106.Rio, D. C., Ares, M. Jr., Hannon, G. J. & Nilsen, T. W. Removal of DNA from RNA. Cold Spring Harb. Protoc.2010, pdb prot5443 (2010). [DOI] [PubMed] [Google Scholar]
107.Rio, D. C., Ares, M. Jr., Hannon, G. J. & Nilsen, T. W. Ethanol precipitation of RNA and the use of carriers. Cold Spring Harb. Protoc.2010, pdb prot5440 (2010). [DOI] [PubMed] [Google Scholar]
108.Mi, H. et al. Protocol update for large-scale genome and gene function analysis with the PANTHER classification system (v.14.0). Nat. Protoc.14, 703–721 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
109.Thomas, P. D. et al. PANTHER: Making genome-scale phylogenetics accessible to all. Protein Sci.31, 8–22 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
110.O’Toole, G. A. Microtiter dish biofilm formation assay. J. Vis. Exp.10.3791/2437 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
111.Zuker, M. Mfold web server for nucleic acid folding and hybridization prediction. Nucleic Acids Res.31, 3406–3415 (2003). [DOI] [PMC free article] [PubMed] [Google Scholar]
112.Carrier, M. C., Laliberte, G. & Masse, E. Identification of new bacterial small RNA targets using MS2 affinity purification coupled to RNA sequencing. Methods Mol. Biol.1737, 77–88 (2018). [DOI] [PubMed] [Google Scholar]
113.Reddi, G., Pruss, K., Cottingham, K. L., Taylor, R. K. & Almagro-Moreno, S. Catabolism of mucus components influences motility of Vibrio cholerae in the presence of environmental reservoirs. PLoS ONE13, e0201383 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
114.Mackie, A. M., Hassan, K. A., Paulsen, I. T. & Tetu, S. G. Biolog phenotype microarrays for phenotypic characterization of microbial cells. Methods Mol. Biol.1096, 123–130 (2014). [DOI] [PubMed] [Google Scholar]
115.Islam, M. S. et al. Role of cyanobacteria in the persistence of Vibrio cholerae O139 in saline microcosms. Can. J. Microbiol.50, 127–131 (2004). [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Information^{(3.9MB, pdf)}

41467_2026_73137_MOESM2_ESM.pdf^{(34.5KB, pdf)}

Description of Additional Supplementary Files

Supplementary Dataset 1^{(26.1KB, xlsx)}

Supplementary Dataset 2^{(105KB, xlsx)}

Supplementary Dataset 3^{(19.1KB, xlsx)}

Supplementary Dataset 4^{(21.3KB, xlsx)}

Supplementary Dataset 5^{(25.5KB, xlsx)}

Supplementary Dataset 6^{(36.6KB, xlsx)}

Supplementary Dataset 7^{(135.5KB, xlsx)}

Supplementary Dataset 8^{(38.5KB, xlsx)}

Supplementary Dataset 9^{(153.3KB, xlsx)}

Supplementary Dataset 10^{(23.9KB, xlsx)}

Supplementary Dataset 11^{(82.3KB, xlsx)}

Supplementary Dataset 12^{(36.8KB, xlsx)}

Reporting Summary^{(981.6KB, pdf)}

Transparent Peer Review file^{(571.4KB, pdf)}

source data^{(15MB, xlsx)}

Data Availability Statement

[CR1] 1.Balasubramanian, D., Lopez-Perez, M., Grant, T. A., Ogbunugafor, C. B. & Almagro-Moreno, S. Molecular mechanisms and drivers of pathogen emergence. Trends Microbiol30, 898–911 (2022). [DOI] [PubMed] [Google Scholar]

[CR2] 2.Diard, M. & Hardt, W. D. Evolution of bacterial virulence. FEMS Microbiol. Rev.41, 679–697 (2017). [DOI] [PubMed] [Google Scholar]

[CR3] 3.Martinez, J. L. Bacterial pathogens: from natural ecosystems to human hosts. Environ. Microbiol.15, 325–333 (2013). [DOI] [PubMed] [Google Scholar]

[CR4] 4.Bonneaud, C., Weinert, L. A. & Kuijper, B. Understanding the emergence of bacterial pathogens in novel hosts. Philos. Trans. R. Soc. Lond. B Biol. Sci.374, 20180328 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR5] 5.Kanungo, S., Azman, A. S., Ramamurthy, T., Deen, J. & Dutta, S. Cholera. Lancet399, 1429–1440 (2022). [DOI] [PubMed] [Google Scholar]

[CR6] 6.Balasubramanian, D., Murcia, S., Ogbunugafor, C. B., Gavilan, R. & Almagro-Moreno, S. Cholera dynamics: lessons from an epidemic. J. Med. Microbiol. 70, 10.1099/jmm.0.001298 (2021). [DOI] [PubMed]

[CR7] 7.Provenzano, D., Schuhmacher, D. A., Barker, J. L. & Klose, K. E. The virulence regulatory protein ToxR mediates enhanced bile resistance in Vibrio cholerae and other pathogenic Vibrio species. Infect. Immun.68, 1491–1497 (2000). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR8] 8.Mathur, J. & Waldor, M. K. The Vibrio cholerae ToxR-regulated porin OmpU confers resistance to antimicrobial peptides. Infect. Immun.72, 3577–3583 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR9] 9.Nalin, D. R. et al. Cholera, non-vibrio cholera, and stomach acid. Lancet2, 856–859 (1978). [DOI] [PubMed] [Google Scholar]

[CR10] 10.Sack, G. H. Jr. et al. Gastric acidity in cholera and noncholera diarrhoea. Bull. World Health Organ47, 31–36 (1972). [PMC free article] [PubMed] [Google Scholar]

[CR11] 11.Liu, Z. et al. Mucosal penetration primes Vibrio cholerae for host colonization by repressing quorum sensing. Proc. Natl. Acad. Sci. USA105, 9769–9774 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR12] 12.Freter, R., O’Brien, P. C. & Macsai, M. S. Role of chemotaxis in the association of motile bacteria with intestinal mucosa: in vivo studies. Infect. Immun.34, 234–240 (1981). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR13] 13.Faruque, S. M. et al. Transmissibility of cholera: in vivo-formed biofilms and their relationship to infectivity and persistence in the environment. Proc. Natl. Acad. Sci. USA103, 6350–6355 (2006). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR14] 14.Brown, R. C. & Taylor, R. K. Organization of tcp, acf, and toxT genes within a ToxT-dependent operon. Mol. Microbiol.16, 425–439 (1995). [DOI] [PubMed] [Google Scholar]

[CR15] 15.DiRita, V. J., Parsot, C., Jander, G. & Mekalanos, J. J. Regulatory cascade controls virulence in Vibrio cholerae. Proc. Natl. Acad. Sci. USA88, 5403–5407 (1991). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR16] 16.Krukonis, E. S., Yu, R. R. & Dirita, V. J. The Vibrio cholerae ToxR/TcpP/ToxT virulence cascade: distinct roles for two membrane-localized transcriptional activators on a single promoter. Mol. Microbiol.38, 67–84 (2000). [DOI] [PubMed] [Google Scholar]

[CR17] 17.Murley, Y. M., Carroll, P. A., Skorupski, K., Taylor, R. K. & Calderwood, S. B. Differential transcription of the tcpPH operon confers biotype-specific control of the Vibrio cholerae ToxR virulence regulon. Infect. Immun.67, 5117–5123 (1999). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR18] 18.Miller, V. L., Taylor, R. K. & Mekalanos, J. J. Cholera toxin transcriptional activator ToxR is a transmembrane DNA binding protein. Cell48, 271–279 (1987). [DOI] [PubMed] [Google Scholar]

[CR19] 19.Provenzano, D. & Klose, K. E. Altered expression of the ToxR-regulated porins OmpU and OmpT diminishes Vibrio cholerae bile resistance, virulence factor expression, and intestinal colonization. Proc. Natl. Acad. Sci. USA97, 10220–10224 (2000). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR20] 20.Bina, J. et al. ToxR regulon of Vibrio cholerae and its expression in vibrios shed by cholera patients. Proc. Natl. Acad. Sci. USA100, 2801–2806 (2003). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR21] 21.Kazi, M. I., Conrado, A. R., Mey, A. R., Payne, S. M. & Davies, B. W. ToxR antagonizes H-NS regulation of horizontally acquired genes to drive host colonization. PLoS Pathog.12, e1005570 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR22] 22.Miller, V. L. & Mekalanos, J. J. Synthesis of cholera toxin is positively regulated at the transcriptional level by ToxR. Proc. Natl. Acad. Sci. USA81, 3471–3475 (1984). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR23] 23.Miller, V. L. & Mekalanos, J. J. A novel suicide vector and its use in construction of insertion mutations: osmoregulation of outer membrane proteins and virulence determinants in Vibrio cholerae requires ToxR. J. Bacteriol.170, 2575–2583 (1988). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR24] 24.Wibbenmeyer, J. A., Provenzano, D., Landry, C. F., Klose, K. E. & Delcour, A. H. Vibrio cholerae OmpU and OmpT porins are differentially affected by bile. Infect. Immun.70, 121–126 (2002). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR25] 25.Chakrabarti, S. R., Chaudhuri, K., Sen, K. & Das, J. Porins of Vibrio cholerae: purification and characterization of OmpU. J. Bacteriol.178, 524–530 (1996). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR26] 26.Provenzano, D., Lauriano, C. M. & Klose, K. E. Characterization of the role of the ToxR-modulated outer membrane porins OmpU and OmpT in Vibrio cholerae virulence. J. Bacteriol.183, 3652–3662 (2001). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR27] 27.Merrell, D. S., Bailey, C., Kaper, J. B. & Camilli, A. The ToxR-mediated organic acid tolerance response of Vibrio cholerae requires OmpU. J. Bacteriol.183, 2746–2754 (2001). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR28] 28.Sakharwade, S. C. & Mukhopadhaya, A. Vibrio cholerae porin OmpU induces LPS tolerance by attenuating TLR-mediated signaling. Mol. Immunol.68, 312–324 (2015). [DOI] [PubMed] [Google Scholar]

[CR29] 29.Khan, J., Sharma, P. K. & Mukhopadhaya, A. Vibrio cholerae porin OmpU mediates M1-polarization of macrophages/monocytes via TLR1/TLR2 activation. Immunobiology220, 1199–1209 (2015). [DOI] [PubMed] [Google Scholar]

[CR30] 30.Seed, K. D. et al. Evolutionary consequences of intra-patient phage predation on microbial populations. Elife3, e03497 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR31] 31.Shapiro, B. J., Levade, I., Kovacikova, G., Taylor, R. K. & Almagro-Moreno, S. Origins of pandemic Vibrio cholerae from environmental gene pools. Nat. Microbiol.2, 16240 (2016). [DOI] [PubMed] [Google Scholar]

[CR32] 32.Bernardini, M. L., Sanna, M. G., Fontaine, A. & Sansonetti, P. J. OmpC is involved in invasion of epithelial cells by Shigella flexneri. Infect. Immun.61, 3625–3635 (1993). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR33] 33.Chevalier, S. et al. Structure, function and regulation of Pseudomonas aeruginosa porins. FEMS Microbiol. Rev.41, 698–722 (2017). [DOI] [PubMed] [Google Scholar]

[CR34] 34.Devoy, A. F., Dyet, K. H. & Martin, D. R. Stability of PorA during a meningococcal disease epidemic. J. Clin. Microbiol.43, 832–837 (2005). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR35] 35.Galdiero, M. et al. Porins from Salmonella enterica serovar Typhimurium activate the transcription factors activating protein 1 and NF-kappaB through the Raf-1-mitogen-activated protein kinase cascade. Infect. Immun.70, 558–568 (2002). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR36] 36.Massari, P., King, C. A., Ho, A. Y. & Wetzler, L. M. Neisserial PorB is translocated to the mitochondria of HeLa cells infected with Neisseria meningitidis and protects cells from apoptosis. Cell Microbiol.5, 99–109 (2003). [DOI] [PubMed] [Google Scholar]

[CR37] 37.Baine, W. B. et al. Epidemiology of cholera in Italy in 1973. Lancet2, 1370–1374 (1974). [DOI] [PubMed] [Google Scholar]

[CR38] 38.Cvjetanovic, B. & Barua, D. The seventh pandemic of cholera. Nature239, 137–138 (1972). [DOI] [PubMed] [Google Scholar]

[CR39] 39.Devault, A. M. et al. Second-pandemic strain of Vibrio cholerae from the Philadelphia cholera outbreak of 1849. N. Engl. J. Med.370, 334–340 (2014). [DOI] [PubMed] [Google Scholar]

[CR40] 40.López-Perez, M. et al. Allelic variations and gene cluster modularity act as nonlinear bottlenecks for cholera emergence. Proc. Natl. Acad. Sci. USA122, e2417915122 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR41] 41.Grant, T. A., Lopez-Perez, M., Haro-Moreno, J. M. & Almagro-Moreno, S. Allelic diversity uncovers protein domains contributing to the emergence of antimicrobial resistance. PLoS Genet.19, e1010490 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR42] 42.Schuster, B. M. et al. Ecology and genetic structure of a northern temperate Vibrio cholerae population related to toxigenic isolates. Appl. Environ. Microbiol.77, 7568–7575 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR43] 43.Hsiao, A., Liu, Z., Joelsson, A. & Zhu, J. Vibrio cholerae virulence regulator-coordinated evasion of host immunity. Proc. Natl. Acad. Sci. USA103, 14542–14547 (2006). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR44] 44.Mathur, J., Davis, B. M. & Waldor, M. K. Antimicrobial peptides activate the Vibrio cholerae sigmaE regulon through an OmpU-dependent signalling pathway. Mol. Microbiol.63, 848–858 (2007). [DOI] [PubMed] [Google Scholar]

[CR45] 45.Liu, J. M. et al. Experimental discovery of sRNAs in Vibrio cholerae by direct cloning, 5S/tRNA depletion and parallel sequencing. Nucleic Acids Res.37, e46 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR46] 46.Updegrove, T. B., Zhang, A. & Storz, G. Hfq: the flexible RNA matchmaker. Curr. Opin. Microbiol.30, 133–138 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR47] 47.Sauer, E. & Weichenrieder, O. Structural basis for RNA 3’-end recognition by Hfq. Proc. Natl. Acad. Sci. USA108, 13065–13070 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR48] 48.Davis, B. M. & Waldor, M. K. RNase E-dependent processing stabilizes MicX, a Vibrio cholerae sRNA. Mol. Microbiol.65, 373–385 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR49] 49.Papenfort, K., Forstner, K. U., Cong, J. P., Sharma, C. M. & Bassler, B. L. Differential RNA-seq of Vibrio cholerae identifies the VqmR small RNA as a regulator of biofilm formation. Proc. Natl. Acad. Sci. USA112, E766–E775 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR50] 50.Hoyos, M., Huber, M., Forstner, K. U. & Papenfort, K. Gene autoregulation by 3’ UTR-derived bacterial small RNAs. Elife 9, 10.7554/elife.58836 (2020). [DOI] [PMC free article] [PubMed]

[CR51] 51.Fong, J. C. & Yildiz, F. H. The rbmBCDEF gene cluster modulates development of rugose colony morphology and biofilm formation in Vibrio cholerae. J. Bacteriol.189, 2319–2330 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR52] 52.Mey, A. R., Craig, S. A. & Payne, S. M. Characterization of Vibrio cholerae RyhB: the RyhB regulon and role of ryhB in biofilm formation. Infect. Immun.73, 5706–5719 (2005). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR53] 53.Gao, H. et al. Fur represses Vibrio cholerae biofilm formation via direct regulation of vieSAB, cdgD, vpsU, and vpsA-K transcription. Front. Microbiol.11, 587159 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR54] 54.Davies, B. W., Bogard, R. W. & Mekalanos, J. J. Mapping the regulon of Vibrio cholerae ferric uptake regulator expands its known network of gene regulation. Proc. Natl. Acad. Sci. USA108, 12467–12472 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR55] 55.Lalaouna, D. et al. A 3’ external transcribed spacer in a tRNA transcript acts as a sponge for small RNAs to prevent transcriptional noise. Mol. Cell58, 393–405 (2015). [DOI] [PubMed] [Google Scholar]

[CR56] 56.Faruque, S. M. et al. Seasonal epidemics of cholera inversely correlate with the prevalence of environmental cholera phages. Proc. Natl. Acad. Sci. USA102, 1702–1707 (2005). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR57] 57.Nelson, E. J. et al. Transmission of Vibrio cholerae is antagonized by lytic phage and entry into the aquatic environment. PLoS Pathog.4, e1000187 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR58] 58.Zahid, M. S. et al. Effect of phage on the infectivity of Vibrio cholerae and emergence of genetic variants. Infect. Immun.76, 5266–5273 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR59] 59.Beckman, D. A. & Waters, C. M. Vibrio cholerae phage ICP3 requires O1 antigen for infection. Infect. Immun.91, e0002623 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR60] 60.Lim, A. N. W., Yen, M., Seed, K. D., Lazinski, D. W. & Camilli, A. A tail fiber protein and a receptor-binding protein mediate ICP2 bacteriophage interactions with Vibrio cholerae OmpU. J. Bacteriol.203, e0014121 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR61] 61.Seed, K. D. et al. Phase variable O antigen biosynthetic genes control expression of the major protective antigen and bacteriophage receptor in Vibrio cholerae O1. PLoS Pathog.8, e1002917 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR62] 62.Gottesman, S. & Storz, G. Bacterial small RNA regulators: versatile roles and rapidly evolving variations. Cold Spring Harb Perspect Biol 310.1101/cshperspect.a003798 (2011). [DOI] [PMC free article] [PubMed]

[CR63] 63.Storz, G., Vogel, J. & Wassarman, K. M. Regulation by small RNAs in bacteria: expanding frontiers. Mol. Cell43, 880–891 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR64] 64.Severin, G. B. et al. Activation of a Vibrio cholerae CBASS anti-phage system by quorum sensing and folate depletion. mBio14, e0087523 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR65] 65.Cohen, D. et al. Cyclic GMP-AMP signalling protects bacteria against viral infection. Nature574, 691–695 (2019). [DOI] [PubMed] [Google Scholar]

[CR66] 66.Crawford, J. A., Kaper, J. B. & DiRita, V. J. Analysis of ToxR-dependent transcription activation of ompU, the gene encoding a major envelope protein in Vibrio cholerae. Mol. Microbiol.29, 235–246 (1998). [DOI] [PubMed] [Google Scholar]

[CR67] 67.Iwanaga, M. & Yamamoto, K. New medium for the production of cholera toxin by Vibrio cholerae O1 biotype El Tor. J. Clin. Microbiol.22, 405–408 (1985). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR68] 68.Iwanaga, M. et al. Culture conditions for stimulating cholera toxin production by Vibrio cholerae O1 El Tor. Microbiol. Immunol.30, 1075–1083 (1986). [DOI] [PubMed] [Google Scholar]

[CR69] 69.Millet, Y. A. et al. Insights into Vibrio cholerae intestinal colonization from monitoring fluorescently labeled bacteria. PLoS Pathog.10, e1004405 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR70] 70.Frederick, A., Huang, Y., Pu, M. & Rowe-Magnus, D. A. Vibrio cholerae Type VI activity alters motility behavior in mucin. J. Bacteriol. 202, 10.1128/jb.00261-20 (2020). [DOI] [PMC free article] [PubMed]

[CR71] 71.Liu, Z. et al. Vibrio cholerae represses polysaccharide synthesis to promote motility in mucosa. Infect. Immun.83, 1114–1121 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR72] 72.Silva, A. J., Pham, K. & Benitez, J. A. Haemagglutinin/protease expression and mucin gel penetration in El Tor biotype Vibrio cholerae. Microbiology149, 1883–1891 (2003). [DOI] [PubMed] [Google Scholar]

[CR73] 73.Xu, J. et al. The role of morphological adaptability in Vibrio cholerae’s motility. mBio16, e0246924 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR74] 74.Nothaft, H. & Szymanski, C. M. Protein glycosylation in bacteria: sweeter than ever. Nat. Rev. Microbiol.8, 765–778 (2010). [DOI] [PubMed] [Google Scholar]

[CR75] 75.Goncheva, M. I., Chin, D. & Heinrichs, D. E. Nucleotide biosynthesis: the base of bacterial pathogenesis. Trends Microbiol.30, 793–804 (2022). [DOI] [PubMed] [Google Scholar]

[CR76] 76.Sperandio, V. et al. Cloning and characterization of the gene encoding the OmpU outer membrane protein of Vibrio cholerae. Infect. Immun.64, 5406–5409 (1996). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR77] 77.Klose, K. E. The suckling mouse model of cholera. Trends Microbiol.8, 189–191 (2000). [DOI] [PubMed] [Google Scholar]

[CR78] 78.Matson, J. S. Infant mouse model of Vibrio cholerae infection and colonization. Methods Mol. Biol.1839, 147–152 (2018). [DOI] [PubMed] [Google Scholar]

[CR79] 79.Yen, M., Cairns, L. S. & Camilli, A. A cocktail of three virulent bacteriophages prevents Vibrio cholerae infection in animal models. Nat. Commun.8, 14187 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR80] 80.Seed, K. D. et al. Evidence of a dominant lineage of Vibrio cholerae-specific lytic bacteriophages shed by cholera patients over a 10-year period in Dhaka, Bangladesh. mBio2, e00334–00310 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR81] 81.Meibom, K. L., Blokesch, M., Dolganov, N. A., Wu, C. Y. & Schoolnik, G. K. Chitin induces natural competence in Vibrio cholerae. Science310, 1824–1827 (2005). [DOI] [PubMed] [Google Scholar]

[CR82] 82.Bardill, J. P. & Hammer, B. K. Non-coding sRNAs regulate virulence in the bacterial pathogen Vibrio cholerae. RNA Biol.9, 392–401 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR83] 83.Perez-Reytor, D. et al. Role of non-coding regulatory RNA in the virulence of human pathogenic vibrios. Front. Microbiol.7, 2160 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR84] 84.Almagro-Moreno, S. & Taylor, R. K. Cholera: Environmental reservoirs and impact on disease transmission. Microbiol. Spectr. 1, 10.1128/microbiolspec.oh-0003-2012 (2013). [DOI] [PubMed]

[CR85] 85.Bloom, D. E., Black, S. & Rappuoli, R. Emerging infectious diseases: A proactive approach. Proc. Natl. Acad. Sci. USA114, 4055–4059 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR86] 86.Djapgne, L. & Oglesby, A. G. Impacts of small RNAs and their chaperones on bacterial pathogenicity. Front. Cell Infect. Microbiol11, 604511 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR87] 87.Bandyra, K. J. & Luisi, B. F. RNase E and the high-fidelity orchestration of RNA metabolism. Microbiol. Spectr 6, 10.1128/microbiolspec.RWR-0008-2017 (2018). [DOI] [PMC free article] [PubMed]

[CR88] 88.Grutzner, J. et al. sRNA-mediated RNA processing regulates bacterial cell division. Nucleic Acids Res.49, 7035–7052 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR89] 89.Han, R. et al. Multiple small RNAs interact to co-regulate ethanol tolerance in Zymomonas mobilis. Front. Bioeng. Biotechnol.8, 155 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR90] 90.Iosub, I. A. et al. Hfq CLASH uncovers sRNA-target interaction networks linked to nutrient availability adaptation. Elife 9, 10.7554/elife.54655 (2020). [DOI] [PMC free article] [PubMed]

[CR91] 91.Kim, W. & Lee, Y. Mechanism for coordinate regulation of rpoS by sRNA-sRNA interaction in Escherichia coli. RNA Biol.17, 176–187 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR92] 92.Liu, F. et al. In vivo RNA interactome profiling reveals 3’UTR-processed small RNA targeting a central regulatory hub. Nat. Commun.14, 8106 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR93] 93.Silva-Valenzuela, C. A. & Camilli, A. Niche adaptation limits bacteriophage predation of Vibrio cholerae in a nutrient-poor aquatic environment. Proc. Natl. Acad. Sci. USA116, 1627–1632 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR94] 94.O’Hara, B. J., Alam, M. & Ng, W. L. The Vibrio cholerae Seventh Pandemic Islands act in tandem to defend against a circulating phage. PLoS Genet.18, e1010250 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR95] 95.Silva, A. J. & Benitez, J. A. Vibrio cholerae biofilms and cholera pathogenesis. PLoS Negl. Trop. Dis.10, e0004330 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR96] 96.Vezzulli, L., Pruzzo, C., Huq, A. & Colwell, R. R. Environmental reservoirs of Vibrio cholerae and their role in cholera. Environ. Microbiol. Rep.2, 27–33 (2010). [DOI] [PubMed] [Google Scholar]

[CR97] 97.Hatleberg, W. L. & Hinman, V. F. Modularity and hierarchy in biological systems: Using gene regulatory networks to understand evolutionary change. Curr. Top. Dev. Biol.141, 39–73 (2021). [DOI] [PubMed] [Google Scholar]

[CR98] 98.Litvin, O., Causton, H. C., Chen, B. J. & Pe’er, D. Modularity and interactions in the genetics of gene expression. Proc. Natl. Acad. Sci. USA106, 6441–6446 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR99] 99.Hartwell, L. H., Hopfield, J. J., Leibler, S. & Murray, A. W. From molecular to modular cell biology. Nature402, C47–C52 (1999). [DOI] [PubMed] [Google Scholar]

[CR100] 100.Montoya, J. M., Pimm, S. L. & Sole, R. V. Ecological networks and their fragility. Nature442, 259–264 (2006). [DOI] [PubMed] [Google Scholar]

[CR101] 101.Krause, A. E., Frank, K. A., Mason, D. M., Ulanowicz, R. E. & Taylor, W. W. Compartments revealed in food-web structure. Nature426, 282–285 (2003). [DOI] [PubMed] [Google Scholar]

[CR102] 102.Thelin, K. H. & Taylor, R. K. Toxin-coregulated pilus, but not mannose-sensitive hemagglutinin, is required for colonization by Vibrio cholerae O1 El Tor biotype and O139 strains. Infect. Immun.64, 2853–2856 (1996). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR103] 103.Strand, T. A., Lale, R., Degnes, K. F., Lando, M. & Valla, S. A new and improved host-independent plasmid system for RK2-based conjugal transfer. PLoS ONE9, e90372 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR104] 104.Skorupski, K. & Taylor, R. K. Positive selection vectors for allelic exchange. Gene169, 47–52 (1996). [DOI] [PubMed] [Google Scholar]

[CR105] 105.Rio, D. C., Ares, M. Jr., Hannon, G. J. & Nilsen, T. W. Purification of RNA by SDS solubilization and phenol extraction. Cold Spring Harb. Protoc.2010, pdb prot5438 (2010). [DOI] [PubMed] [Google Scholar]

[CR106] 106.Rio, D. C., Ares, M. Jr., Hannon, G. J. & Nilsen, T. W. Removal of DNA from RNA. Cold Spring Harb. Protoc.2010, pdb prot5443 (2010). [DOI] [PubMed] [Google Scholar]

[CR107] 107.Rio, D. C., Ares, M. Jr., Hannon, G. J. & Nilsen, T. W. Ethanol precipitation of RNA and the use of carriers. Cold Spring Harb. Protoc.2010, pdb prot5440 (2010). [DOI] [PubMed] [Google Scholar]

[CR108] 108.Mi, H. et al. Protocol update for large-scale genome and gene function analysis with the PANTHER classification system (v.14.0). Nat. Protoc.14, 703–721 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR109] 109.Thomas, P. D. et al. PANTHER: Making genome-scale phylogenetics accessible to all. Protein Sci.31, 8–22 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR110] 110.O’Toole, G. A. Microtiter dish biofilm formation assay. J. Vis. Exp.10.3791/2437 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR111] 111.Zuker, M. Mfold web server for nucleic acid folding and hybridization prediction. Nucleic Acids Res.31, 3406–3415 (2003). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR112] 112.Carrier, M. C., Laliberte, G. & Masse, E. Identification of new bacterial small RNA targets using MS2 affinity purification coupled to RNA sequencing. Methods Mol. Biol.1737, 77–88 (2018). [DOI] [PubMed] [Google Scholar]

[CR113] 113.Reddi, G., Pruss, K., Cottingham, K. L., Taylor, R. K. & Almagro-Moreno, S. Catabolism of mucus components influences motility of Vibrio cholerae in the presence of environmental reservoirs. PLoS ONE13, e0201383 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR114] 114.Mackie, A. M., Hassan, K. A., Paulsen, I. T. & Tetu, S. G. Biolog phenotype microarrays for phenotypic characterization of microbial cells. Methods Mol. Biol.1096, 123–130 (2014). [DOI] [PubMed] [Google Scholar]

[CR115] 115.Islam, M. S. et al. Role of cyanobacteria in the persistence of Vibrio cholerae O139 in saline microcosms. Can. J. Microbiol.50, 127–131 (2004). [DOI] [PubMed] [Google Scholar]

PERMALINK

Modular small RNA drives the emergence of virulence traits and environmental trade-offs in Vibrio cholerae

Deepak Balasubramanian

Qinqin Pu

Cole Crist

Salvador Almagro-Moreno

Abstract

Introduction

Results

OmpU represses biofilm formation in V. cholerae and modulates its transcriptome

Fig. 1. The ompU ORF encodes a small RNA with allelic variants.

The ompU ORF encodes a small RNA

OueS represses biofilm formation in V. cholerae

OueS is responsible for ompU-dependent transcriptomic changes during biofilm formation

Fig. 2. OueS represses biofilm formation via direct supression of RyhB.

OueS from toxigenic strains suppresses biofilm formation by directly repressing the iron-responsive sRNA RyhB

Table 1.

OueS domains determine RyhB repression and biofilm formation

Toxigenic OueS confers resistance against intestinal bacteriophages by activating the CBASS defense system

Fig. 3. OueS activates the CBASS phage defense system by direcly repressing glyA confering resistance against intestinal bacteriophages.

OueS represses glyA conferring resistance against phages

OueS controls ToxR-mediated phenotypes

Fig. 4. OueS expression is controlled by ToxR acting as its functional surrogate.

Toxigenic OueS modulates 84.7% of the ToxR regulome

OueS controls ToxR-activated mucin penetration in a domain-dependent manner

ToxR mediates repression of nutrient utilization in an OueS-dependent manner

OueS is expressed from the ompU promoter

Fig. 5. OueS is expressed via the ompU promoter and leads to successful intestinal colonization of V. cholerae.

OueS is essential for successful intestinal colonization

OueS alleles exhibit a modular structure

Fig. 6. OueS is a modular sRNA that confers preadaptation to intestinal colonization.

Modular heterogeneity in OueS contributes to pathogenic potential of Vibrio cholerae strains

V. cholerae can chromosomally exchange the ompU locus in a modular fashion

Fig. 7. Acquisition of the ompU ORF is modular and controlled by environmental conditions.

Modular acquisition of toxigenic ompU leads to the emergence of virulence-associated traits

Environmental drivers control the modular acquisition of the ompU ORF

Diverse transcriptomes lead to convergent phenotypes in OueS

Allelic variability of OueS leads to fitness trade-offs between human infection and environmental survival

Fig. 8. OueS leads to fitness trade-offs between intestinal colonization and environmental survival.

Discussion

Fig. 9. The ompU ORF encodes a modular porin and sRNA.

Methods

Ethics statement

Bacterial strains, media, and growth conditions

Cloning, vectors, and strain construction

RNA isolation, transcriptome and qRT-PCR studies

Biofilm assays

Northern blot assays

Rapid Amplification of cDNA Ends (RACE)

Prediction of sRNA secondary structure

MS2 affinity purification coupled with RNA-sequencing (MAPS)

OueS conservation mapping, phylogenetic distribution, and sequence alignment

Growth kinetics under iron limitation

Bacteriophage assays

Motility assays

Phenotypic microarrays

Chitin-induced natural transformation and transformation frequency

Mucin penetration

P2 and low pH survival assays

Infant mouse competition assays

Colonization of environmental reservoirs

Statistical analyses

Reporting summary

Supplementary information

Source data

Acknowledgements

Author contributions

Peer review

Peer review information

Funding

Data availability

Competing interests

Footnotes

Supplementary information

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK