Skip to main content
  • Loading metrics

Allelic diversity uncovers protein domains contributing to the emergence of antimicrobial resistance

  • Trudy-Ann Grant,

    Roles Data curation, Formal analysis, Investigation, Methodology, Validation, Visualization, Writing – original draft, Writing – review & editing

    Affiliation Burnett School of Biomedical Sciences, University of Central Florida, Orlando, Florida, United States of America

  • Mario López-Pérez,

    Roles Formal analysis, Investigation, Methodology, Visualization, Writing – original draft, Writing – review & editing

    Affiliations Burnett School of Biomedical Sciences, University of Central Florida, Orlando, Florida, United States of America, Evolutionary Genomics Group, División de Microbiología, Universidad Miguel Hernández, San Juan, Alicante, Spain

  • Jose Manuel Haro-Moreno,

    Roles Formal analysis, Investigation, Writing – review & editing

    Affiliations Burnett School of Biomedical Sciences, University of Central Florida, Orlando, Florida, United States of America, Evolutionary Genomics Group, División de Microbiología, Universidad Miguel Hernández, San Juan, Alicante, Spain

  • Salvador Almagro-Moreno

    Roles Conceptualization, Data curation, Formal analysis, Funding acquisition, Methodology, Project administration, Resources, Supervision, Visualization, Writing – original draft, Writing – review & editing

    Affiliation Burnett School of Biomedical Sciences, University of Central Florida, Orlando, Florida, United States of America


Antimicrobial resistance (AMR) remains a major threat to global health. To date, tractable approaches that decipher how AMR emerges within a bacterial population remain limited. Here, we developed a framework that exploits genetic diversity from environmental bacterial populations to decode emergent phenotypes such as AMR. OmpU is a porin that can make up to 60% of the outer membrane of Vibrio cholerae, the cholera pathogen. This porin is directly associated with the emergence of toxigenic clades and confers resistance to numerous host antimicrobials. In this study, we examined naturally occurring allelic variants of OmpU in environmental V. cholerae and established associations that connected genotypic variation with phenotypic outcome. We covered the landscape of gene variability and found that the porin forms two major phylogenetic clusters with striking genetic diversity. We generated 14 isogenic mutant strains, each encoding a unique ompU allele, and found that divergent genotypes lead to convergent antimicrobial resistance profiles. We identified and characterized functional domains in OmpU unique to variants conferring AMR-associated phenotypes. Specifically, we identified four conserved domains that are linked with resistance to bile and host-derived antimicrobial peptides. Mutant strains for these domains exhibit differential susceptibility patterns to these and other antimicrobials. Interestingly, a mutant strain in which we exchanged the four domains of the clinical allele for those of a sensitive strain exhibits a resistance profile closer to a porin deletion mutant. Finally, using phenotypic microarrays, we uncovered novel functions of OmpU and their connection with allelic variability. Our findings highlight the suitability of our approach towards dissecting the specific protein domains associated with the emergence of AMR and can be naturally extended to other bacterial pathogens and biological processes.

Author summary

Antimicrobial resistance (AMR) is one of the major threats to global health. To date, tractable approaches that decipher how AMR emerges within a bacterial population remain limited. Here we developed an approach that uses genetic diversity from environmental populations to decode emergent phenotypes such as AMR. Specifically, we examined naturally occurring allelic variants of an outer membrane porin, OmpU, in Vibrio cholerae and established associations between genotype and phenotype. Using this approach, we identified and characterized the functional domains in OmpU unique to variants conferring AMR-associated phenotypes. Our perspective towards disentangling the emergence of AMR can be naturally extended to other proteins and bacterial pathogens.


Antimicrobial resistance (AMR) is a complex phenomenon fueling the arms race to combat emerging and re-emerging pathogens. By 2050, it is estimated that 10 million individuals will die annually due to infections caused by pathogens that exhibit AMR [1]. For instance, resistance to many currently used antimicrobials such as ciprofloxacin or ampicillin has emerged in up to 80% of Klebsiella pneumoniae and Escherichia coli strains, increasing the likelihood of treatment failures [2]. Current approaches to combat AMR include the development of antimicrobial cocktails and the discovery of novel or modified ones, however, the rapid adaptations to new therapeutics by pathogenic strains demands a deeper understanding of the molecular mechanisms and evolutionary paths associated with the emergence of this threat [3, 4]. Furthermore, given the complexity of this problem, approaches that recognize the relevance of the environment that pathogenic agents inhabit and closely align with the "One Health" paradigm are imperative to effectively address AMR [58].

Gram-negative bacteria pose one of the greatest challenges to our ability to overcome AMR due to their highly selective outer membrane embedded with proteins that hinder the effectiveness of antimicrobial compounds [911]. One group of membrane proteins commonly associated with AMR are outer membrane porins (OMPs). OMPs facilitate the translocation of molecules within the cell and are often essential to pathogenesis as regulators of this selective barrier [1215]. OMPs modulate transport of small molecules, with selectivity based upon charge and size, allowing for the intake of essential nutrients, or preventing internalization of antimicrobials [1620]. Furthermore, alteration in levels of OMP expression, mutations in the porin constriction region, or expression of alternative porins can confer antimicrobial resistance [15, 18, 2126]. To date, how these critical adaptations for virulence evolve remains poorly understood.

Vibrio cholerae, a comma-shaped Gram-negative bacterium, is the etiological agent of the severe diarrheal disease cholera. Interestingly, the toxigenic strains that can cause the disease in humans belong to one phylogenetically confined group within the species, the pandemic cholera group (PCG) [2729]. Recently, we determined that strains from PCG encode allelic variations in core genes that confer preadaptations for virulence [27]. These allelic variations termed virulence adaptive polymorphisms (VAPs), naturally circulate in environmental populations of V. cholerae including non-toxigenic strains [27]. OmpU is a major porin in V. cholerae that can make up to 60% of its outer membrane under certain conditions and is associated with antimicrobial resistance and intestinal colonization, among other phenotypes [3033]. Recently we determined that the gene encoding OmpU in toxigenic strains potentially encodes VAPs and exhibits allelic diversity in environmental populations [27]. Interestingly, unlike the ompU allele encoded by toxigenic strains, the allele from the environmental isolate V. cholerae GBE1114 does not confer virulence-adaptive properties such as bile resistance and leads to impaired intestinal colonization [27]. We hypothesize that specific VAP-encoding domains within the toxigenic ompU allele confer these emergent virulence-adaptive properties.

Given that these traits are contained within the variable regions of these two divergent alleles, in this article, we reasoned that these natural allelic variations could also provide us with an ideal model system to identify the specific genetic variations that contribute to AMR and the emergence of virulence-adaptive traits. A preliminary sequence survey exposed large differences between the toxigenic and the GBE1114 alleles making it unfeasible to select any particular set of residues (74.7% identity). To narrow down the potential regions associated with the emergence of AMR, first, we examined the natural variability of ompU uncovering 41 alleles of the porin out of over 1600 sequences analyzed. These alleles form two phylogenetic clusters with distinct evolutionary paths and striking diversity. We selected 7 representative alleles from each cluster encompassing the diversity of ompU and constructed 14 isogenic mutant strains where we exchanged the toxigenic allele for an environmental one. We exposed the ompU mutant strains to host antimicrobials (e.g. bile) to determine how genotypic variations affect phenotypic outcomes. Conservation and structural analyses of each residue within the alleles from strains that exhibit resistant phenotypes versus sensitive ones led to the identification and characterization of four conserved domains among alleles that conferred bile resistance. A mutant strain where we exchanged the four specific domains from the sensitive allele into the clinical one exhibit a phenotype similar to an ompU deletion strain in the presence of bile and the host antimicrobial peptide, P2. Finally, to determine novel potential roles of OmpU, we performed a high-throughput screening using phenotypic microarrays and compared the differential survival and growth of V. cholerae N16961 (WT) and a ΔompU mutant. We identified clinically relevant antibiotics where resistance is associated with OmpU. Overall, our assessment of natural allelic variations uncovered functional protein domains associated with AMR, shedding light on the evolutionary processes leading to the emergence of this phenomenon.


ompU forms two major phylogenetic clusters

OmpU plays a crucial role in several virulence-associated traits in V. cholerae such as bile tolerance, antimicrobial peptide resistance, and intestinal colonization [30, 31, 33]. We recently determined that only some alleles of ompU confer these observed phenotypes [27]. In the presence of 0.4% whole bile, V. cholerae N16961 (WT) exhibits an average 25% survival as measured by colony-forming unit (c.f.u) post-treatment, while a ΔompU mutant strain shows only 0.16% survival (p<0.0001) [27]. An isogenic V. cholerae N16961 mutant strain encoding the ompU allele from the environmental strain V. cholerae GBE1114 exhibits 1.4% survival (p = 0.02 versus ΔompU) [27]. This allele-dependent phenotype indicates the presence of unique functional residues in the clinical allele of ompU leading to the emergence of bile resistance. To identify these functional residues, we first performed a sequence alignment of the N16961 (resistant) and the GBE1114 (sensitive) proteins (Fig 1A). The protein sequences share 74.7% identity with a plethora of mutations including 3 insertions ranging from 1-11 amino acid (aa) per event, 8 deletions, and 30 non-synonymous mutations, relative to WT, overall precluding us from directly identifying potential candidates within clinical OmpU leading to AMR.

Fig 1. OmpU forms two highly diverse phylogenetic clusters.

(A) Sequence alignment of the ompU alleles from strains V. cholerae N16961 and GBE1114. (B) OmpU has diverged into two major phylogenetic clusters. Red arrow indicates the clinical allele from N16961. Green and purple shading represent Cluster 1 and 2 respectively. (C) Survival of V. cholerae mutant strains encoding representative ompU alleles in the presence of 0.4% whole bile. Despite the large genetic differences between the alleles, strain survival in bile leads to convergent phenotypes. Error bars represent the standard deviation of the mean from at least three (N≥3) independent replicates. Statistical comparisons were performed using one-way ANOVA analyses followed by a Dunnett’s multiple comparison test. All constructs were compared to the WT unless otherwise stated. *p<0.05, **p<0.01, ***p<0.001.

To determine the diversity of OmpU and subsequently correlate it with AMR, we obtained all sequences of the ompU gene in V. cholerae genomes from public databases, including environmental strains of V. cholerae in our collection [27, 29, 34]. V. cholerae N16961 OmpU was used as a reference to search for all homologs obtaining a total of 1620 sequences. After a previous clustering of the sequences (100% identity), to eliminate the clonality bias of V. cholerae, we generated a phylogenetic tree to examine the evolutionary history of OmpU (Fig 1B). The analysis reveals two distinct major clusters composed of 28 clades (Cluster 1; C1) and 13 clades (Cluster 2; C2), respectively (Fig 1B). Alleles from toxigenic strains of V. cholerae (clinical allele) were confined to one clade within the larger cluster C1. Given this allelic variability, we hypothesized that we could compare the phenotype associated with different alleles and infer genotype-phenotype correlations to discern the functional residues that differentiate the resistant from the sensitive ones. To make this approach tractable we reduced the number of alleles tested to 14 by prioritizing distribution of clades and genotype patterns based on both alignment and phylogeny (Table 1). Based on these criteria, we selected 7 alleles within C1, with nucleotide identities ranging from 96% (closest to clinical; IRLE0074) to 75% (furthest; IRLE0079), strengthening the scenario in which the clinical allele of ompU emerged as a preadaptation in V. cholerae populations prior to host colonization [27, 35]. On the other hand, the nucleotide identities of alleles from C2 range from 72% (closest to clinical; IRLE0082) to 69% (furthest; GBE1116). The environmental ompU allele of strain GBE1114, which is sensitive to bile, is located in C2. In comparison to GBE1114, the nucleotide identities of alleles present in C2 range from 89% (closest to GBE1114) to 79% (furthest). From C1 we included the ompU alleles from the following V. cholerae strains (number between brackets indicates clade number): IRLE0074 (1), GBE0658 (2), IRLE0181 (3), IRLE0079 (4), IRLE0081 (5), GBE0428 (6), and GBE0917 (7), and from C2 IRLE0082 (8) GBE1116 (9), GBE1194 (10), RIMD 2214396 (11), FORC 073 (12), IDH-06787 (13), and GBE1114 (14) (Table 1, S1 Fig, S1 Appendix).

Naturally occurring allelic variations of OmpU lead to phenotypic diversity in the presence of bile

To examine the phenotypes associated with the different representative ompU alleles, we constructed 14 isogenic mutants, in the genomic background of V. cholerae N16961 each encoding one of them (Table 1). First, we examined the growth of the strains in LB media and the stability of OmpU using immunoblots (S2 Fig). We detected no growth defect among the mutant strains and found that the mutations did not affect the protein stability (S2 Fig). Subsequently, we examined the survival of these strains in the presence of bile (Fig 1C). We observed that the overall resistance pattern in the presence of bile broadly correlates with tree topology, with several strains from C1 exhibiting a resistance phenotype like WT whereas the ones from C2 (closer to GBE1114) show a sensitive one (Fig 1C). Specifically, strains ompU1, ompU2, ompU3, and ompU5 from C1 display similar resistance to bile as the WT strain. Of note, ompU4, ompU6 and ompU7 warrant some further investigations as they show a decrease in survival in the presence of bile, with ompU6 exhibiting a phenotype similar to some strains from C2 (ompU8 and ompU10) (Fig 1C). Strains encoding alleles from C2 exhibit on average ~1-log decrease in survival compared to the WT strain (Fig 1C). Those encoding alleles from clades 9, 11 and 12 are the most sensitive to bile when compared to WT and their survival is similar to the strain encoding the allele from GBE1114 (clade 14).

Genotype to phenotype associations uncover potential domains linked with bile resistance

Phenotypic characterization of ompU alleles reveal conserved resistance patterns stemming from diverse genotypes (Fig 1). To determine functional domains unique to alleles that confer resistance to bile (resistant alleles), we first created groups of resistant and sensitive alleles based on statistical significance. Using this criterion, we generated two groups with distinct phenotypes: A) resistant (ompU1, ompU2, ompU3, and ompU5): alleles that displayed no statistically significant difference in survival when compared to WT and B) sensitive (ompU9, ompU11, and ompU12): alleles that displayed no statistically significant difference in survival when compared to GBE1114 (clade 14). These two groups with similar phenotypes but diverse genotypes provided an ideal tool to identify the functional residues contributing to bile resistance in clinical OmpU. We performed a Scorecons analysis of the resistant and sensitive groups to assist the identification of potential functional residues unique to resistant alleles (S3 Fig). Based on our analysis, we identified a total of 244/350 (ca. 70%) conserved positions at the protein level (= 1) and 106 positions where residues were not conserved (<1). Of the 106 variable positions, 82 (75%) were distributed in large groups (≥4 or more consecutive variable or inserted residues) within external loops or internal modules of the porin based on published annotations [36, 37]. Conversely, the remaining 24 (25%) were mainly single amino acid changes or variable groups distributed throughout the hydrophobic transmembrane ß-sheets embedded within the outer membrane. The distribution of these large variable groups of residues in external modules of OmpU further suggests that environmental pressures play a role in selecting emergent adaptive mutations to tolerate adverse conditions. To increase the probability of identifying functional domains conferring resistance to host antimicrobials, we focused primarily on amino acid positions where residue variability occurred in clusters. Based on this parameter, we identified seven aa regions of interest (S1 Fig): 1) 31–43, 2) 102–114, 3) 153–158, 4) 193–203, 5) 264–272, 6) 293–305 and 7) 336–338 relative to the WT allele.

After examining each aligned column and narrowing down to clusters of non-conserved amino acids, we then sought to identify regions unique to resistant alleles. Using the criteria described in the materials and methods section, we examined the seven identified regions and of these, regions 1 (31-43aa), and 3 (153-158aa) showed a clear discrimination between the sensitive and resistant alleles. Conservation scores for regions 4 (193–203) and 7 (336–338) were variable for both resistant and sensitive alleles, however, based on the criteria outlined, we observed a trend towards non-conservative for this region with a higher number of nonsynonymous residues occurring in sensitive alleles. Regions 2 (102-114aa), 5 (264-272aa) and 6 (293-305aa) have positions that are so variable that no pattern could be identified even for resistant alleles limiting their relevance as potential domains associated with the emergence antimicrobial resistance in OmpU. Thus, they were excluded from further analysis. Overall, our systematic genotype to phenotype association and comparative sequence analyses led us to identify regions 1, 3, 4 and 7, which we will term N-terminal coil (NTC), L3 region (L3R), L4-loop (L4) and VAS respectively, for subsequent analyses (Fig 2A, S4 Fig).

Fig 2. Comparative visualization of domains within OmpU associated with bile resistant phenotypes.

(A). Multiple sequence alignment of ompU alleles resistant and sensitive to whole bile. Conserved residues are indicated by dots and those absent are indicated by gray boxes. (B-E) Comparative architecture of OmpU domains associated with the bile resistant phenotype. Domains are color coded and visualized in OmpU N16961 and GBE1114. Top and slabbed view of (B, C) N16961 and (D, E) GBE1114, respectively. The identified domains are colored as follows: NTC (purple), L3R (orange), L4 (blue) and VAS (green).

3D visualization and structural analysis of identified protein domains

OmpU is a homotrimer that is distributed throughout the outer membrane [38]. Each monomer consists of 16 ß-sheets embedded within the outer membrane connected by 8 short turns on the periplasmic side and 8 loops on the external side [36] (Fig 2B). From the extracellular side, there are eight loops, termed L1-8, connecting these ß-sheets. Of these, six extend into the extracellular space (L1, L4-8) and are exposed to fluctuating environmental conditions, one (L2) overlaps its neighboring monomer and facilitates stabilization of the trimer and the last (L3) folds into the porin and forms a constriction region across the porin that interacts with amino acids on the inner barrel wall [36, 37] (Fig 2B). On the periplasmic side of the porin, at the N-terminal region, a coil extends inward into the porin and interacts with L3 [37]. We examined whether the four identified domains translate into structural differences in OmpU using the protein visualization software Chimera [39]. In order to do this, we generated a predicted OmpU protein model of the GBE1114 allele and examined individual and superimposed OmpU protein models of N16961 (OmpUN16961) and GBE1114 (OmpUGBE1114). As identified in our analyses above, structural comparisons reveal that the NTC domain that pushes inward into OmpUN16961 [40] is absent from OmpUGBE1114 (Fig 2B and 2D). Secondly, amino acid residues present in the L3R domain identified above, are located on the inward L3 loop, and differ in charge and hydrophobicity between the two alleles (Fig 2C and 2E). OmpUN16961 encodes charged residues (E, and K), whereas OmpUGBE1114 does not encode any charged residues (Fig 2A). Lastly, domains L4R and VAS are located on loops L4 and L8 respectively that protrude into the extracellular space (Fig 2C and 2E). Interestingly, overlay of both porins reveal an extended L4 in OmpUN16961 when compared to OmpUGBE1114. Additionally, the VAS domain is hydrophobic in OmpUN16961 whereas in OmpUGBE1114 it is basic (Fig 2A). Overall, the identified domains translate to several structural differences that may alter protein-protein interactions, modulate open and closed conformation states, or modify overall stabilization of the porin in unfavorable conditions.

OmpU-mediated resistance to bile requires specific domains present in the clinical allele

To determine whether the domains we identified above play a role in OmpU-mediated AMR, we generated four individual mutant constructs that encode the GBE1114 sequence of each identified domain in the background of V. cholerae N16961. For ease of classification, each mutant was assigned a binary code based on the domains encoded, where ‘1’ represents the WT domain from N16961 and ‘0’ represents the domain from GBE1114. This was done for each domain and the strain classification is as follows: ompU0111 (NTC), ompU1011 (L3R), ompU1101 (L4) and ompU1110 (VAS). In addition, to determine whether a synergistic effect exists among the domains, we constructed an isogenic mutant strain encoding an ompU allele with each of the four domains exchanged with the corresponding sequence from GBE1114. This allele was assigned the binary code ‘0000’ (null). As previously done, first we performed growth analysis of the mutant strains in LB and found that they do not exhibit any growth defect in rich media when compared to the WT (S5 Fig). Immunoblots indicate that OmpU is expressed in the isogenic mutant strains and exhibit protein stability (Fig 3A). Subsequently, we examined the survival of the mutants in the presence of 0.4% whole bile (Fig 3B). Interestingly, two individual domain mutants, ompU0111 (NTC) and ompU1011 (L3R), show the greatest susceptibility to bile with survival rates of 9.6% and 4.5% respectively (Fig 3B). The ompU1101 mutant exhibits 17% survival rate, whereas the ompU1110 mutant shows no survival defect, and displays similar survival as the WT. The ompU0000 “null” mutant has an average survival rate of 3.4%, closely resembling the phenotype of the strain encoding the allele from GBE1114 (Fig 3B). Interestingly, the ompU0000 mutant shows similar survival in the presence of bile to the single domain mutant ompU1011. Overall, our results indicate that, independently, the NTC and L3R domains contribute to the ability of OmpU to confer bile resistance.

Fig 3. OmpU domain mutants exhibit differential resistance to 0.4% bile.

(A) Immunoblot analysis of individual OmpU domain mutants. (B) Survival of individual and naturally occurring ompU domain combination mutants in the presence of 0.4% whole bile. (C) Immunoblot analysis of naturally occurring OmpU domain mutants. Error bars represent the standard deviation of the mean from at least three (N≥3) independent replicates. Statistical comparisons were performed using one-way ANOVA analyses followed by a Dunnett’s multiple comparison test. All constructs were compared to the WT unless otherwise stated. *p<0.05, **p<0.01, ***p<0.001.

Next, we examined the diversity of naturally-occurring domain combinations in our representative dataset of ompU alleles to unearth potential synergistic effects between the domains. We identified four naturally occurring domain combinations and classified each using the established binary nomenclature. The domains combinations are as follows: ompU0100 (GBE0917), ompU0010 (IDH-06787), ompU0110 (GBE1116), and ompU1100 (IRLE0081). Next, as described above, we constructed mutant strains encoding those domain combinations and after growth analysis, we determined that there was no significant growth difference when compared to the WT. Subsequently, the stability of OmpU was verified by immunoblots in the mutants (Fig 3C) and then their bile resistance profiles examined. Upon exposure to 0.4% whole bile, all mutants exhibit a decrease in bile resistance with survival ranging from 6 to 12% (Fig 3B). While we determined that L3R is crucial for resistance to bile, the mutant ompU0100 reveals that presence of only the clinical L3R domain cannot recover the bile-resistant phenotype, further suggesting a synergistic effect between the domains. Secondly, strains encoding alleles with the clinical L4 domain (ompU0010) or the L3R and L4 domains (ompU0110), display a minor increase in survival when compared with ompU0100 (Fig 3C). Lastly, the strain encoding both the clinical NTC and L3R domain (ompU1100) shows the highest resistance to bile (12%) when compared to other domain combinations (Fig 3C). Our results indicate that individually both NTC and L3R domains are crucial for increased resistance to bile (Fig 3C). However, the synergistic effect of all domains appears to be essential for AMR in the clinical OmpU to emerge.

OmpU domain mutants exhibit differential survival upon exposure to various antimicrobials

OmpU, in toxigenic V. cholerae, is known to confer resistance to cationic antimicrobial peptides (CAMPs) such as P2 or polymyxin B (PB) and provides tolerance to organic acid [32, 33, 41]. To determine whether these OmpU-associated phenotypes are allelic- and domain-dependent, we phenotypically characterized the ompUGBE1114 and ompU0000 domain mutants in the presence of P2 and PB, and when subjected to organic acid challenge.

Antimicrobial peptide P2.

After exposure to 200μg P2, WT has a 91% survival rate whereas the ΔompU mutant exhibits a survival rate of 7.5%, comparable to previously published studies [33]. Interestingly, the mutant ompUGBE1114 shows 12% survival, and the “null” domain mutant ompU0000 23%. (Fig 4A). All these strains exhibit a statistically significant difference in survival when compared to WT (p<0.001), indicating that survival in the presence of P2 is both allelic and domain-dependent (Fig 4A). Next, we determined the role of the individual domains in resistance to P2. Surprisingly, we found that all the domain mutants show a decrease in survival ranging between 43–52% compared to WT (p<0.0001) (Fig 4). A potential synergistic effect between the domains might explain their difference with ompU0000.

Fig 4. Role of protein domains in OmpU-associated phenotypes.

Survival of V. cholerae N16961 ompU mutant strains in the presence of (A) 200ug P2, (B) polymyxin B (450UmL-1) and (C) organic acid (0.005X). Error bars represent the standard deviation of the mean from at least three (N≥3) independent replicates. Statistical comparisons were performed using one-way ANOVA analyses followed by a Dunnett’s multiple comparison test. All constructs were compared to the WT unless otherwise stated. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001.

Polymyxin B (PB).

Secondly, in the presence of PB, as previously determined [27], ΔompU show a drastic decrease in survival (11.3%) compared to WT (92.9%). Unlike for P2, neither ompUGBE1114 nor ompU0000 show a statistically significant change in survival compared to WT (Fig 4B). As shown in Fig 1C, there is phenotypic diversity in the alleles tested in the presence of bile, thus, we considered that it would be possible that resistance to PB might be allelic-dependent yet not associated with the specific allele from strain GBE1114. To examine this, we exposed the other 13 mutant strains encoding the different ompU alleles to PB and measured their survival (Fig 4B). Overall, we did not find major differences between the mutant strains and the WT, certainly none of them resembling the ΔompU mutant (Fig 4B). Strains with the ompU allele from clades 4, 9, and 12 show the lowest survival in the presence of PB (between 42% and 55%) but not enough to warrant strong evidence that PB resistance is allelic-dependent. Furthermore, there is no clear pattern that associates these strains with a given cluster.

Organic acid

On the other hand, even though neither ompUGBE1114 (C14) nor ompU0000 show a decrease in survival compared to WT when cells were exposed to an organic acid cocktail (propionic acid, butyric acid, and acetic acid), the role of OmpU in organic acid tolerance appears to be allelic-dependent (Fig 4C). Specifically, strains from clades 9, 10 and 11 show between 14% and 20% survival in organic acid, similar to ΔompU (12%), whereas the WT show an average survival of 50% (Fig 4C). All the strains with increased sensitivity to organic acid belong to Cluster 2. Overall, our results uncover complex associations between the different alleles and domains of ompU and their ability to confer phenotypes associated with resistance to host antimicrobials.

OmpU confers resistance to Rifamycin SV

OmpU is ubiquitous in the outer membrane of V. cholerae and, as our results indicate, appears to have more nuances than expected in the form of allele-associated domains [27]. Based on this, we considered that it is possible that the porin plays other roles in V. cholerae survival beyond those already known. In order to test this, we screened for potential novel functions of OmpU by performing high-throughput assays using phenotypic microarrays (Fig 5A). We tested for sensitivity to chemicals and antimicrobials (960 compounds and conditions), growth and survival at different pH (96 conditions) and osmolytes (96 conditions). We compared the growth of the WT and ΔompU mutant strains under these conditions and identified thirteen antimicrobial compounds in the presence of which the WT strain exhibits at least two-fold greater area under the curve (AUC) compared to the ΔompU mutant (Fig 5A). Of particular interest, four of these compounds are antibiotics that have been used in clinical settings: rifamycin, oxytetracycline, cinoxacin and troleandomycin [4246]. Subsequently, we performed in-vitro survival assays in the presence of these antibiotics to corroborate the OmpU-dependent phenotypes observed in the phenotypic microarrays and determine whether resistance to these compounds is allelic-dependent. Troleandomycin was not commercially available and could not be included in our in vitro assays. Given that the concentration of antimicrobials provided in the phenotypic microarray plates are proprietary information, first we performed titrations of each antimicrobial to determine if we could replicate the growth difference in a survival assay. Furthermore, to examine the role of allelic variations on survival in the presence of these antimicrobials, a large enough cut-off for the difference between the WT and ΔompU was deemed necessary. For this, we followed a cutoff in which a) the WT displays at least 25% survival under that antibiotic concentration, and b) at that concentration the ΔompU mutant exhibits at least a 5-fold decrease in survival.

Fig 5. Exploration of novel functions of OmpU.

(A) Phenotypic microarrays of V. cholerae WT and ΔompU strains. Area under the curve (AUC) was calculated to determine the total growth of the strains under exposure to diverse antimicrobial compounds. Only compounds with greater than 2-fold decrease in AUC for ΔompU strain are shown. Red rectangles indicate antibiotics with clinical relevance. All arrays were performed in duplicate. (B, C) Survival of V. cholerae N16961 and ΔompU in the presence of varying concentrations of (B) oxytetracycline and (C) cinoxacin. (D) Survival of ompU alleles in 7.5μg/mL rifamycin SV. Error bars represent the standard deviation of the mean from at least three (N≥3) independent replicates. Statistical comparisons were performed using one-way ANOVA analyses followed by a Dunnett’s multiple comparison test. All constructs were compared to the WT unless otherwise stated. *p<0.05. N≥3, *p<0.05.

Cinoxacin and oxytetracycline

Based on our titrations of cinoxacin and oxytetracycline, we could not identify a concentration that met our cut-off to infer an OmpU-dependent role in the survival against these antimicrobials (Fig 5B and 5C).

Rifamycin SV

First, we tested survival in the presence of rifamycin SV (RSV), an antimicrobial agent used to treat traveler’s diarrhea [47]. In order to identify a suitable concentration of RSV, we performed titrations and observed that 7.5μg/mL provided proper resolution between the WT and ΔompU strains. At this concentration, the WT strain shows an average survival of 44% whereas ΔompU exhibits 1.5% (Fig 5D, S6 Fig). This large survival range provided us with enough potential resolution to observe differences between the alleles. As for the conditions tested above, next, we examined the survival of the ompUGBE1114 (C14) and ompU0000 strains in the presence of RSV. We found that even though resistance to RSV appears to be allelic-dependent (ompUGBE1114 exhibits decreased survival), the domains that we found associated with bile and P2 survival do not play a role in this phenotype (ompU0000 exhibits WT-level of survival). Our results indicate that other domains within OmpU might be responsible for this phenotype (Fig 5D). Finally, we tested the survival of the other ompU allele mutant strains in the presence of RSV and found that four of them (ompU5, ompU6, ompU8, and ompU13) exhibit increased resistance to the antibiotic than the WT (Fig 5D). Overall, resistance to RSV is OmpU-dependent and no allele other than ompUGBE1114 display lower survival rate than the WT allele.


The development of antimicrobial resistance among bacterial pathogens is unarguably one of the most critical public health threats that we face and fuels the arms race to combat hospital and community-acquired infections [4850]. Furthermore, this phenomenon is closely linked to the process of pathogen emergence [5153]. Gram-negative pathogens pose a major obstacle towards tackling this as the presence of their highly selective outer membrane (OM) critically facilitate adaptations for AMR. Porins represent one of the major proteins present in the OM and are often directly implicated in AMR mechanisms. OmpU, a homotrimer porin that is distributed throughout up to 60% of the OM of V. cholerae, is directly associated with the emergence of the pathogen and confers resistance to numerous host antimicrobials [38, 54, 55]. Here, using the allelic diversity found in natural populations of V. cholerae, we developed a framework that allowed us to identify domains within the porin associated with the emergence and evolution of AMR in pathogenic strains of the bacterium. Examination of the distribution and variability of ompU alleles within V. cholerae reveals a highly diverse landscape in which two large clusters encompass all the alleles of the porin. This vast diversity of allelic variants suggests a scenario in which variations in this protein are strongly driven by environmental pressures and likely confer a set of phenotypes much wider than anticipated. It is known that OmpU acts as the receptor of phage ICP2, which directly interacts with domains that are present on extracellular loops of the porin (L4 and L8) [56, 57]. Nonetheless, the extensive variability of OmpU goes beyond the region of the porin that acts as the phage receptor. This indicates that, despite phages likely being major drivers in the selection of these variants, the porin must undergo other strong selective pressures that led to this uncanny diversity.

Bile is an anionic salt under the pH conditions present in the small intestine and requires facilitated transport to disrupt the cell integrity of enteric pathogens such as V. cholerae [55, 58, 59]. Upon exposure to bile, V. cholerae activates a signaling cascade leading to the upregulation of ompU and promoting its survival in the SI [30, 31, 58, 6062]. Our results reveal that the ability of the OmpU allele encoded by toxigenic strains to confer resistance to bile is due to four domains encoded within different regions of the porin. Phenotypic characterization of the four identified domains, reveals that two are the most crucial for bile resistance: the NTC and L3 domains. Protein modelling of the WT allele show that the NTC domain extends from the periplasm inwards into the porin and interacts with the inner loop 3 (L3), which traverses the upper region of the OmpU towards to the extracellular side. Domains analogous to NTC and L3 in other porins contribute to pore size and altered open and close conformation states [37, 6368]. Given our findings, it is likely that the presence of the N-terminal coil contributes to a smaller pore size [37] in the WT allele leading to exclusion of bile molecules in contrast to the sensitive alleles. In addition, charged residues on the L3 domain in the WT allele may facilitate longer closed confirmation states and prevent bile influx. Some alleles in C1 (ompU4, ompU6, and ompU7) display decreased survival in the presence of bile when compared the WT. Comparative sequence analysis reveals that ompU4 encodes an elongated external loop 6 (275–296 aa), a feature absent from all other alleles, and contains indels in loop 2 (103-111aa) which in similar porins facilitates the stability of the trimer [69, 70]. The allele ompU6 does not have any major indels when compared with the WT but encodes several point mutations that likely affect the overall conformations of the porin. Lastly, ompU7 lacks the crucial NTC which is likely contributing to the observed survival defect.

We previously proposed that OmpU from toxigenic strains evolved in the environment and serve as preadaptations to virulence in the context of the human host [27, 35]. Our results here further corroborate this scenario as there are numerous environmental non-toxigenic strains that encode alleles of OmpU conferring bile resistance (Fig 1C). To initially assess whether the domains associated with the bile resistance phenotype evolve through direct exposure to bile (pathoadaptations) or as predicted, were acquired prior to exposure to host-associated antimicrobials, we performed in vitro evolution experiments by exposing the isogenic mutant strains encoding the alleles ompUGBE1114 and ompU0000 to 0.4% bile (S7C Fig and S2 Appendix). Alignment of the OmpU alleles of three of these evolved strains indicate that OmpU does not exhibit changes over time suggesting that compensatory mutations in the background of V. cholerae led to this phenotype in the presence of 0.4% bile (S7C Fig). Future work should resolve whether these results might also be due to the stringency of the treatment or whether higher resolution (e.g., sample size) are necessary in order to identify more nuanced changes in the population. Nonetheless, taken together our results suggest that toxigenic OmpU has evolved as a preadaptation to virulence likely driven by phage predation or adaptations to other biotic or abiotic variables.

Cationic antimicrobial peptides (CAMPs) are natural peptide products that kill bacteria through membrane destabilization [7173]. CAMPs selectively target Gram-negative pathogens due to their affinity for negatively charged outer membranes [71]. To evade CAMPs such as P2 or polymyxin B, V. cholerae activates the envelope stress response in a sigma E-dependent manner [32, 33]. The initiation of this cascade is dependent on the presence of OmpU [32, 33]. Interestingly, our data reveals that OmpU-associated CAMP resistance is allelic and domain dependent only in the presence of the host-derived peptide P2. On the other hand, polymyxin B is an antimicrobial peptide produced by Paenibacillus polymyxa [74], a bacterium that can be found in aquatic habitats [75]. It is likely that recurrent exposure to this CAMP led to population bottlenecks that selected for alleles that confer widespread resistance. Given the differences in the observed survival pattern, this difference in mechanism of action as well as the potential role of other genes in counteracting these antimicrobials remains enigmatic and warrants further investigations.

We determined that all individual domains played a role in resistance to P2. However, absence of no individual domain leads to a survival pattern similar to ompU0000 suggesting a synergistic effect among the domains leading to increased sensitivity in ompU0000. Computational modeling suggests that residues on N-terminal coil interact with the YDF domain to facilitate porin stabilization [37]. We hypothesize that this may function to expose the YDF domain upon envelope stress increasing the survival of clinical strains in the presence of host antimicrobials. Furthermore, the extended negatively charged L4 loop may act as an interacting site for positively charged P2 molecules and reduce its translocation into the cell. Also, interactions between charged residues on L3CR and the barrel wall are likely to affect open/close conformation states of the porin. Lastly, regions upstream of the C-terminal YDF domain are crucial for the stabilization of the anti-anti sigma factor RseB. This facilitates the first cleavage step in the envelope stress response signaling cascade, regulated by RpoE [76, 77]. Given these observed differences and the significance of OmpU to AMP resistance, we propose that these structural differences cooperatively contribute to activating RpoE upon exposure to P2. For some phenotypes tested, such as polymyxin B, neither allelic nor domain-dependent pattern was observed. This indicates that in the presence of these compounds, there may be compensatory events that offset its effects. On the other hand, survival in the presence of organic acid is allelic-dependent but appears to require the presence of ompU domains that are distinct from the ones required for bile or P2 survival. Future work will focus on elucidating these specific domains and examining how they might influence other functions of OmpU.

Our findings highlight the astonishing nuances of OmpU and its multifaceted role in the emergence and pathogenesis of V. cholerae. Adaptations leading to antimicrobial resistance are essential to the continued success of V. cholerae as a pathogen. Recent reports stress the critical importance of monitoring AMR in V. cholerae as the bacterium is gaining it at unprecedented rates [7880]. Examination of the natural variability of OmpU has provided us with a unique tool to determine how genotypic variations within a porin can foster emergent phenotypes in bacterial pathogens. To date, the potential downstream regulatory effects associated with the emergence of the identified domains remain unknown. Future work will address this and other critical questions that will take us closer to dissecting the complex AMR phenomenon.

Materials and methods

Bacterial strains and culture conditions

Strains used in this study (Table 1) were routinely grown aerobically on Luria Bertani (LB) media at 37°C for approximately 16h, unless otherwise stated. Antibiotic concentrations used for plasmid or strain selection are as follows, kanamycin (Kn) 45μg/mL, polymyxin B (PB) 50μg/mL, and streptomycin (Sm) 1mg/mL.

OmpU phylogenetic tree

All available V. cholerae genomes were downloaded from the National Center for Biotechnology Information (NCBI, release 244 July 2021). Then, the DNA sequence coding for the outer membrane protein OmpU (VC_0633) in V. cholerae O1 biovar El Tor str. N16961 was used as a reference to obtain all the homologous sequences in the genomes using DIAMOND [81]. Sequences were clustered with CD-HIT [82] and aligned using MUSCLE [83]. Maximum-likelihood tree was then constructed using RAxML [84] with the following parameters: “-f a” algorithm, 100 bootstrap replicates, PROTGAMMAJTT model.

Construction of mutant strains

V. cholerae constructs were all created in the N16961 background following previously published methods [27]. Briefly, for allelic exchange of each representative clades of ompU, the allele of interest was either PCR amplified from environmental strains in our collection (Table 1) or synthesized (Gene Universal) if strain was unavailable in our collection. In addition, we PCR amplified 500bp upstream of the 5’ end of V. cholerae N16961 ompU and 500bp downstream of the 3’ end of the gene. PCR or synthesized products were digested and ligated into the suicide vector pKAS154 using a four-segmented ligation. Ligation products were electroporated into Escherichia coli S17λpir and the products plated on LB supplemented with Kn to select for plasmid uptake. PCR amplicons of plasmid constructs were sequenced and compared to environmental allele for verification. Strains carrying the insert were conjugated with V. cholerae N16961 and selected for using a series of antibiotics outlined here [85]. PCR amplification was performed to screen for the ompU allele and potentially positive samples were sequenced and verified by sequence comparison.

Western blot analysis

Overnight cultures of the WT and mutant constructs were diluted to OD600nm 1.0, centrifuged, resuspended in 100μL of sample loading buffer and boiled for 5mins. Samples were diluted 1:5 in sample loading buffer and quantified using the Pierce BCA Protein Assay Kit (Thermo Scientific). Lysates were normalized to 100μg/mL in 1X Bolt LDS Sample Buffer (Invitrogen). A total of 2 μL of normalized lysates were loaded into precast 12% Tris-glycine gels (Invitrogen) alongside protein standards (SeeBlue Pre-stained Protein Standard, Invitrogen). Gels were run at room temperature for 150 min at 80 V in running buffer. Blots were transferred to nitrocellulose membranes using the iBlot system (Invitrogen). After the transfer, membranes were removed and placed in blocking buffer (TBST supplemented with 5% non-fat milk) and incubated overnight at 4°C. Subsequently, blocking buffer was removed and replaced with TBST buffer supplemented with 1:30,000 anti-OmpU rabbit primary antibody (GenScript) and incubated for 1h. Membranes were washed three times with TBST after which TBST supplemented with 1:10,000 of anti-rabbit secondary antibody (GenScript) was added and incubated for 30min at room temperature. After incubation with the secondary antibody, membranes were washed three times with TBS buffer. Reagents from the Pierce ECL kit were mixed according to manufacturer’s instructions and chemiluminescence was detected using BioRad ChemiDoc XRS system.

Sequence comparison and evaluation of resistant and sensitive ompU alleles

Multiple sequence alignments (S1 Fig) were generated using CLC Genomics workbench 21 and the conservation score for the residues in each aligned column calculated using the entropy-based method, Scorecons [86]. Based on the 7 entropic states in which amino acids exists, we examined the diversity of each residue within an aligned column and designated a score between 0 (not conserved) and 1 (conserved). To determine whether the residues of sensitive alleles influence conservation at each site, we subtracted conservation scores obtained for resistant alleles from those obtained for resistant and sensitive alleles. A value <0 indicates presence of non-synonymous residues at a given site. On the other hand, a site with a score equal to zero (0) would indicate that the residues are either conserved at this site or there is significant number of variations between both groups and would therefore not be of interest given that there is no exclusivity to either group. A score that is >0 would indicate that there is high conservation in the combined conservation comparison but differences in the resistant group. These sites would not be of interest given that the variability does not affect the survival pattern. Identification of naturally occurring domains. Based on the four identified domains, each representative allele was examined for changes in conservation score at each domain position when compared to the WT. Based on this criteria, four naturally occurring domain combinations were identified: ompU0100 (GBE0917), ompU0010 (IDH-06787), ompU0110 (GBE1116), and ompU1100 (IRLE0081).

OmpU structure visualization

Nucleotide sequences of the ompU alleles from N16961 and GBE1114 were translated to amino acid sequences using the online tool Expasy and uploaded to SWISS-MODEL to generate a protein model for each allele [87]. OmpU structures were visualized using UCSF ChimeraX [39].

Growth analysis

Overnight cultures were washed twice and diluted 1:100 in fresh LB. Aliquots of 200μl of each suspension was added to a sterile 96-well plate. Plates were incubated under aerobic conditions at 37°C for 15 hours with shaking. The optical density (OD) was measured at 595nm every hour using the Tecan Sunrise Microplate reader (Tecan). The experiment was carried out in triplicates, N = 3.

Survival assays

Overnight cultures were subcultured (1:100) in LB and grown to an optical density at 600nm (OD600nm) of 0.5. Cells were washed twice and resuspended in LB. Cultures were resuspended in LB or LB supplemented with 0.4% whole bile (Chem-Impex), 450 U/mL polymyxin B (Sigma), or 7.5μg/mL rifamycin SV (Sigma). Cultures were incubated for 1h at 37°C in a rotary shaker after which the colony forming units were calculated by plating serial dilutions on LB agar. For exposure to RSV, cultures were incubated for 30mins. Organic acid tolerance. Exposure to organic acid was performed using LB pH 4.5 supplemented with 0.005X organic acid cocktail containing propionic acid, butyric acid and acetic acid as previously described [41]. To investigate survival in the presence of the antimicrobial peptide P2, experiments were performed using protocol from reference 33 with minor modifications [33]. Briefly, overnight cultures of V. cholerae were subcultured (1:100) in LB and grown to an optical density at 600nm (OD600nm) of 0.5. Cells were washed twice with phosphate-buffered saline (PBS) pH 7.4 and resuspended in PBS pH 7.4. Aliquots of 100μL of culture were added to 400μL of PBS or PBS supplemented with P2 to achieve a final concentration of 200μg of P2. Cultures were incubated for 1.5h at 37°C in a rotary shaker after which the colony forming units were calculated by plating serial dilutions on LB agar. The percentage of bacterial survival was calculated by comparing the c.f.u./mL of the treated versus the untreated (N≥3). Titrations of oxytetracycline and cinoxacin were performed using concentrations ranging from .06–0.5μg/mL and 7.5–15μg/mL for rifamycin SV. Cultures were incubated for either 30 min (RSV and cinoxacin) or 1h (oxytetracycline) at 37°C in a rotary shaker after which the colony forming units were calculated by plating serial dilutions on LB agar. The percentage of bacterial survival was calculated by comparing the c.f.u./mL of the treated versus the untreated (N≥3).

Phenotypic microarrays

Biolog Phenotype MicroArrays for antimicrobial chemical sensitivity, pH and osmolarity (PM 9–20) were performed in accordance with the protocol from Mackie et al. [88]. Briefly, V. cholerae was suspended in 1x Biolog IF-0a media with 85% transmittance (Biolog turbidimeter, 20-mm diameter tube). Following, the cells were diluted 1:200 in IF-10 media plus Biolog Redox Dye Mix D to create the inoculum (Biolog, Hayward, CA). PM 9–20 were inoculated with 100 μl/well and incubated at 37°C with shaking. OD at 595 nm (OD595nm) was read every hour for 24h in a TECAN Sunrise microplate reader (TECAN US). Data was collected and analyzed by Magellan software (TECAN US). Biolog assays were performed in duplicate.

Supporting information

S2 Appendix. Bile directed in vitro evolution of isogenic ompU mutant strains.


S1 Fig. Multiple sequence alignment of ompU alleles.

Gray boxes represent missing residues and black dots represent conserved residues. Highlighted in black boxes are identified regions of interest numbered 1–7. Green and purple dots represent alleles from Cluster 1 and Cluster 2 respectively.


S2 Fig. Growth analysis and OmpU stability of Cluster 1 and Cluster 2 isogenic mutant strains.

(A) Growth analysis of isogenic mutant strains encoding environmental ompU alleles, N≥3. (B-C) Immunoblots of OmpU from representative clades. 200ng of whole cell lysates from strains encoding alleles from (B) Cluster 1 and (C) Cluster 2 were probed using α-OmpU antibodies.


S3 Fig. Scorecons analysis of ompU alleles exhibiting convergent phenotypes.

Scorecons result for alignments of bile resistant and sensitive allele groups. Conservation scores (y-axis) range between 0 (not conserved) and 1 (conserved) and are indicated by the pink dots. Amino acid location (x-axis) is relative to the WT allele (N16961).


S4 Fig. Sequence alignment and domain identification of ompU alleles exhibiting convergent phenotypes.

Gray boxes represent missing residues and black dots represent conserved residues. Highlighted in black boxes are confirmed regions of interest lettered A-D. Green and purple dots represent alleles from Cluster 1 and Cluster 2, respectively.


S5 Fig. Growth analysis of ompU domains mutant strains.

Strains were cultured overnight in LB at 37°C with aeration using the Tecan Sunrise microplate reader (Tecan) and optical density (OD) was measured hourly at 595 nm using the Magellan plate reader software (Tecan), N≥3.


S6 Fig. Titrations of Rifamycin SV.

Survival of V. cholerae N16961 WT and ΔompU in the presence of varying concentrations of rifamycin SV. Statistical comparisons were performed using two-way ANOVA analyses followed by a Bonferroni’s multiple comparison test. All constructs were compared to the WT unless otherwise stated, N≥3. *p<0.05, **p<0.01, ***p<0.001.


S7 Fig. In vitro evolution of OmpU isogenic mutant strains.

(A) Schematic of the methodology for the in vitro evolution experiments of V. cholerae OmpU isogenic mutant strains in the presence of 0.4% whole bile (Adapted from Levin-Reisman et al [89]). (B) Survival of V. cholerae isogenic mutant strains encoding ompU alleles ompUGBE1114 and ompU0000 over successive passages through 0.4% whole bile. n = 3. Statistical comparisons were performed using one-way ANOVA analyses followed by a Dunnett’s multiple comparison test. All constructs were compared to the WT unless otherwise stated. *p<0.05, **p<0.01, ***p<0.001. (C) Alignment of OmpU variants. Wild-type N16961 (WT), OmpU0000 (Null), OmpU0000 after bile passages (Null-Evo1, Null-Evo2, Null-Evo3), OmpUGBE1114 (GBE1114), OmpUGBE1114 after bile passages (GBE1114-Evo1, GBE1114-Evo2, GBE1114-Evo3).



  1. 1. O’Neill J. The Review on Antimicrobial Resistance 2016 [updated May 19, 2016]. Available from:
  2. 2. Antimicrobial resistance World Health Organization2021 [updated November 17, 2021]. Available from:
  3. 3. Reygaert WC. An overview of the antimicrobial resistance mechanisms of bacteria. AIMS Microbiol. 2018;4(3):482–501. Epub 20180626. pmid:31294229; PubMed Central PMCID: PMC6604941.
  4. 4. Holmes AH, Moore LS, Sundsfjord A, Steinbakk M, Regmi S, Karkey A, et al. Understanding the mechanisms and drivers of antimicrobial resistance. Lancet. 2016;387(10014):176–87. Epub 20151118. pmid:26603922.
  5. 5. Aslam B, Khurshid M, Arshad MI, Muzammil S, Rasool M, Yasmeen N, et al. Antibiotic Resistance: One Health One World Outlook. Front Cell Infect Microbiol. 2021;11:771510. Epub 20211125. pmid:34900756; PubMed Central PMCID: PMC8656695.
  6. 6. McEwen SA, Collignon PJ. Antimicrobial Resistance: a One Health Perspective. Microbiol Spectr. 2018;6(2). pmid:29600770.
  7. 7. Cabrera-Pardo JR, Lood R, Udekwu K, Gonzalez-Rocha G, Munita JM, Jarhult JD, et al. A One Health—One World initiative to control antibiotic resistance: A Chile—Sweden collaboration. One Health. 2019;8:100100. Epub 20190814. pmid:31649993; PubMed Central PMCID: PMC6804674.
  8. 8. Eriksen J, Bjorkman I, Roing M, Essack SY, Stalsby Lundborg C. Exploring the One Health Perspective in Sweden’s Policies for Containing Antibiotic Resistance. Antibiotics (Basel). 2021;10(5). Epub 20210503. pmid:34063697; PubMed Central PMCID: PMC8147834.
  9. 9. Breijyeh Z, Jubeh B, Karaman R. Resistance of Gram-Negative Bacteria to Current Antibacterial Agents and Approaches to Resolve It. Molecules. 2020;25(6). Epub 20200316. pmid:32187986; PubMed Central PMCID: PMC7144564.
  10. 10. Exner M, Bhattacharya S, Christiansen B, Gebel J, Goroncy-Bermes P, Hartemann P, et al. Antibiotic resistance: What is so special about multidrug-resistant Gram-negative bacteria? GMS Hyg Infect Control. 2017;12:Doc05. Epub 20170410. pmid:28451516; PubMed Central PMCID: PMC5388835.
  11. 11. Gupta V, Datta P. Next-generation strategy for treating drug resistant bacteria: Antibiotic hybrids. Indian J Med Res. 2019;149(2):97–106. pmid:31219074; PubMed Central PMCID: PMC6563750.
  12. 12. Sugawara E, Nikaido H. Pore-forming activity of OmpA protein of Escherichia coli. J Biol Chem. 1992;267(4):2507–11. pmid:1370823.
  13. 13. Kaeriyama M, Machida K, Kitakaze A, Wang H, Lao Q, Fukamachi T, et al. OmpC and OmpF are required for growth under hyperosmotic stress above pH 8 in Escherichia coli. Lett Appl Microbiol. 2006;42(3):195–201. pmid:16478504.
  14. 14. Iyer R, Moussa SH, Durand-Reville TF, Tommasi R, Miller A. Acinetobacter baumannii OmpA Is a Selective Antibiotic Permeant Porin. ACS Infect Dis. 2018;4(3):373–81. Epub 20171226. pmid:29260856.
  15. 15. Nikaido H. Molecular basis of bacterial outer membrane permeability revisited. Microbiol Mol Biol Rev. 2003;67(4):593–656. pmid:14665678; PubMed Central PMCID: PMC309051.
  16. 16. Delcour AH. Outer membrane permeability and antibiotic resistance. Biochim Biophys Acta. 2009;1794(5):808–16. Epub 2008/12/23. pmid:19100346; PubMed Central PMCID: PMC2696358.
  17. 17. Ude J, Tripathi V, Buyck JM, Soderholm S, Cunrath O, Fanous J, et al. Outer membrane permeability: Antimicrobials and diverse nutrients bypass porins in Pseudomonas aeruginosa. Proc Natl Acad Sci U S A. 2021;118(31). pmid:34326266; PubMed Central PMCID: PMC8346889.
  18. 18. Vergalli J, Bodrenko IV, Masi M, Moynie L, Acosta-Gutierrez S, Naismith JH, et al. Porins and small-molecule translocation across the outer membrane of Gram-negative bacteria. Nat Rev Microbiol. 2020;18(3):164–76. Epub 20191202. pmid:31792365.
  19. 19. Tuveri GM, Ceccarelli M, Pira A, Bodrenko IV. The Optimal Permeation of Cyclic Boronates to Cross the Outer Membrane via the Porin Pathway. Antibiotics (Basel). 2022;11(7). Epub 20220623. pmid:35884094; PubMed Central PMCID: PMC9311757.
  20. 20. Sun J, Rutherford ST, Silhavy TJ, Huang KC. Physical properties of the bacterial outer membrane. Nat Rev Microbiol. 2022;20(4):236–48. Epub 20211103. pmid:34732874; PubMed Central PMCID: PMC8934262.
  21. 21. Chen S, Zhang A, Blyn LB, Storz G. MicC, a second small-RNA regulator of Omp protein expression in Escherichia coli. J Bacteriol. 2004;186(20):6689–97. pmid:15466019; PubMed Central PMCID: PMC522180.
  22. 22. Gayet S, Chollet R, Molle G, Pages JM, Chevalier J. Modification of outer membrane protein profile and evidence suggesting an active drug pump in Enterobacter aerogenes clinical strains. Antimicrob Agents Chemother. 2003;47(5):1555–9. pmid:12709321; PubMed Central PMCID: PMC153306.
  23. 23. Bialek S, Lavigne JP, Chevalier J, Marcon E, Leflon-Guibout V, Davin A, et al. Membrane efflux and influx modulate both multidrug resistance and virulence of Klebsiella pneumoniae in a Caenorhabditis elegans model. Antimicrob Agents Chemother. 2010;54(10):4373–8. Epub 20100802. pmid:20679507; PubMed Central PMCID: PMC2944602.
  24. 24. Dam S, Pages JM, Masi M. Stress responses, outer membrane permeability control and antimicrobial resistance in Enterobacteriaceae. Microbiology (Reading). 2018;164(3):260–7. Epub 20180125. pmid:29458656.
  25. 25. De E, Basle A, Jaquinod M, Saint N, Mallea M, Molle G, et al. A new mechanism of antibiotic resistance in Enterobacteriaceae induced by a structural modification of the major porin. Mol Microbiol. 2001;41(1):189–98. pmid:11454211.
  26. 26. Thiolas A, Bornet C, Davin-Regli A, Pages JM, Bollet C. Resistance to imipenem, cefepime, and cefpirome associated with mutation in Omp36 osmoporin of Enterobacter aerogenes. Biochem Biophys Res Commun. 2004;317(3):851–6. pmid:15081418.
  27. 27. Shapiro BJ, Levade I, Kovacikova G, Taylor RK, Almagro-Moreno S. Origins of pandemic Vibrio cholerae from environmental gene pools. Nat Microbiol. 2016;2:16240. Epub 2016/12/20. pmid:27991885.
  28. 28. Mutreja A, Kim DW, Thomson NR, Connor TR, Lee JH, Kariuki S, et al. Evidence for several waves of global transmission in the seventh cholera pandemic. Nature. 2011;477(7365):462–5. Epub 20110824. pmid:21866102; PubMed Central PMCID: PMC3736323.
  29. 29. López-Pérez M, Grant T.-A., Haro-Moreno J. M, Zaragoza-Solas A, & Almagro-Moreno S. Cluster-driven evolution and modularity uncover paths of cholera emergence. bioRxiv. 2022.
  30. 30. Provenzano D, Klose KE. 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 U S A. 2000;97(18):10220–4. Epub 2000/08/16. pmid:10944196; PubMed Central PMCID: PMC27820.
  31. 31. Wibbenmeyer JA, Provenzano D, Landry CF, Klose KE, Delcour AH. Vibrio cholerae OmpU and OmpT porins are differentially affected by bile. Infect Immun. 2002;70(1):121–6. pmid:11748172; PubMed Central PMCID: PMC127639.
  32. 32. Mathur J, Waldor MK. The Vibrio cholerae ToxR-regulated porin OmpU confers resistance to antimicrobial peptides. Infect Immun. 2004;72(6):3577–83. Epub 2004/05/25. pmid:15155667; PubMed Central PMCID: PMC415678.
  33. 33. Mathur J, Davis BM, Waldor MK. Antimicrobial peptides activate the Vibrio cholerae sigmaE regulon through an OmpU-dependent signalling pathway. Mol Microbiol. 2007;63(3):848–58. Epub 2006/12/22. pmid:17181782.
  34. 34. Schuster BM, Tyzik AL, Donner RA, Striplin MJ, Almagro-Moreno S, Jones SH, et al. Ecology and genetic structure of a northern temperate Vibrio cholerae population related to toxigenic isolates. Appl Environ Microbiol. 2011;77(21):7568–75. Epub 2011/09/20. pmid:21926213; PubMed Central PMCID: PMC3209147.
  35. 35. Balasubramanian D, Lopez-Perez M, Grant TA, Ogbunugafor CB, Almagro-Moreno S. Molecular mechanisms and drivers of pathogen emergence. Trends Microbiol. 2022;30(9):898–911. Epub 20220302. pmid:35248462.
  36. 36. Li H, Zhang W, Dong C. Crystal structure of the outer membrane protein OmpU from Vibrio cholerae at 2.2 A resolution. Acta Crystallogr D Struct Biol. 2018;74(Pt 1):21–9. Epub 20180101. pmid:29372896; PubMed Central PMCID: PMC5786005.
  37. 37. Pathania M, Acosta-Gutierrez S, Bhamidimarri SP, Baslé A, Winterhalter M, Ceccarelli M, et al. Unusual Constriction Zones in the Major Porins OmpU and OmpT from Vibrio cholerae. Structure. 2018;26(5):708–21.e4. Epub 2018/04/12. pmid:29657131.
  38. 38. Chakrabarti SR, Chaudhuri K, Sen K, Das J. Porins of Vibrio cholerae: purification and characterization of OmpU. J Bacteriol. 1996;178(2):524–30. pmid:8550475; PubMed Central PMCID: PMC177687.
  39. 39. Pettersen EF, Goddard TD, Huang CC, Couch GS, Greenblatt DM, Meng EC, et al. UCSF Chimera—a visualization system for exploratory research and analysis. J Comput Chem. 2004;25(13):1605–12. pmid:15264254.
  40. 40. Pathania M, Acosta-Gutierrez S, Bhamidimarri SP, Basle A, Winterhalter M, Ceccarelli M, et al. Unusual Constriction Zones in the Major Porins OmpU and OmpT from Vibrio cholerae. Structure. 2018;26(5):708–21 e4. Epub 20180412. pmid:29657131.
  41. 41. Merrell DS, Bailey C, Kaper JB, Camilli A. The ToxR-mediated organic acid tolerance response of Vibrio cholerae requires OmpU. J Bacteriol. 2001;183(9):2746–54. Epub 2001/04/09. pmid:11292792; PubMed Central PMCID: PMC99489.
  42. 42. Otake H, Sato Y, Nakatani E, Hawke P, Takei S, Ogino A, et al. Oxytetracycline-hydrocortisone ointment reduces the occurrence of both dry socket and post-extraction pain after third molar extraction: An observational study. PLoS One. 2021;16(7):e0254221. Epub 20210702. pmid:34214141; PubMed Central PMCID: PMC8253388.
  43. 43. Elmes PC, Fletcher CM, Dutton AA. Prophylactic use of oxytetracycline for exacerbations of chronic bronchitis. Br Med J. 1957;2(5056):1272–5. pmid:13479695; PubMed Central PMCID: PMC1963159.
  44. 44. Finland M, Grigsby ME, Haight TH. Efficacy and toxicity of oxytetracycline (terramycin) and chlortetracycline (aureomycin); with special reference to use of doses of 250 mg. every four to six hours and to occurrence of staphylococcic diarrhea. AMA Arch Intern Med. 1954;93(1):23–43. pmid:13103847.
  45. 45. Scavone JM, Gleckman RA, Fraser DG. Cinoxacin: mechanism of action, spectrum of activity, pharmacokinetics, adverse reactions, and therapeutic indications. Pharmacotherapy. 1982;2(5):266–72. pmid:6763208
  46. 46. Weinberger M, Hudgel D, Spector S, Chidsey C. Inhibition of theophylline clearance by troleandomycin. J Allergy Clin Immunol. 1977;59(3):228–31. pmid:300083
  47. 47. DuPont HL, Petersen A, Zhao J, Mundt A, Jiang ZD, Miller S, et al. Targeting of rifamycin SV to the colon for treatment of travelers’ diarrhea: a randomized, double-blind, placebo-controlled phase 3 study. J Travel Med. 2014;21(6):369–76. pmid:25345982.
  48. 48. Micoli F, Bagnoli F, Rappuoli R, Serruto D. The role of vaccines in combatting antimicrobial resistance. Nat Rev Microbiol. 2021;19(5):287–302. Epub 20210204. pmid:33542518; PubMed Central PMCID: PMC7861009.
  49. 49. Sugden R, Kelly R, Davies S. Combatting antimicrobial resistance globally. Nat Microbiol. 2016;1(10):16187. Epub 20160927. pmid:27670123.
  50. 50. Plazak ME, Tamma PD, Heil EL. The antibiotic arms race: current and emerging therapy for Klebsiella pneumoniae carbapenemase (KPC)—producing bacteria. Expert Opin Pharmacother. 2018;19(18):2019–31. Epub 20181105. pmid:30346216.
  51. 51. Prestinaci F, Pezzotti P, Pantosti A. Antimicrobial resistance: a global multifaceted phenomenon. Pathog Glob Health. 2015;109(7):309–18. Epub 20150907. pmid:26343252; PubMed Central PMCID: PMC4768623.
  52. 52. Davies J, Davies D. Origins and evolution of antibiotic resistance. Microbiol Mol Biol Rev. 2010;74(3):417–33. pmid:20805405; PubMed Central PMCID: PMC2937522.
  53. 53. Ikhimiukor OO, Odih EE, Donado-Godoy P, Okeke IN. A bottom-up view of antimicrobial resistance transmission in developing countries. Nat Microbiol. 2022;7(6):757–65. Epub 20220530. pmid:35637328.
  54. 54. Benz R, Maier E, Chakraborty T. Purification of OmpU from Vibrio cholerae classical strain 569B: evidence for the formation of large cation-selective ion-permeable channels by OmpU. Microbiologia. 1997;13(3):321–30. pmid:9353751.
  55. 55. Simonet VC, Basle A, Klose KE, Delcour AH. The Vibrio cholerae porins OmpU and OmpT have distinct channel properties. J Biol Chem. 2003;278(19):17539–45. Epub 2003/02/28. pmid:12606562.
  56. 56. Lim ANW, Yen M, Seed KD, Lazinski DW, Camilli A. A Tail Fiber Protein and a Receptor-Binding Protein Mediate ICP2 Bacteriophage Interactions with Vibrio cholerae OmpU. J Bacteriol. 2021;203(13):e0014121. Epub 20210608. pmid:33875544; PubMed Central PMCID: PMC8316142.
  57. 57. Seed KD, Yen M, Shapiro BJ, Hilaire IJ, Charles RC, Teng JE, et al. Evolutionary consequences of intra-patient phage predation on microbial populations. Elife. 2014;3:e03497. Epub 20140826. pmid:25161196; PubMed Central PMCID: PMC4141277.
  58. 58. Almagro-Moreno S, Pruss K, Taylor RK. Intestinal Colonization Dynamics of Vibrio cholerae. PLoS Pathog. 2015;11(5):e1004787. Epub 2015/05/23. pmid:25996593; PubMed Central PMCID: PMC4440752.
  59. 59. Pagel M, Delcour AH. Effects of conjugated and unconjugated bile acids on the activity of the Vibrio cholerae porin OmpT. Mol Membr Biol. 2011;28(1):69–78. Epub 20101111. pmid:21067451.
  60. 60. Yang M, Liu Z, Hughes C, Stern AM, Wang H, Zhong Z, et al. Bile salt-induced intermolecular disulfide bond formation activates Vibrio cholerae virulence. Proc Natl Acad Sci U S A. 2013;110(6):2348–53. Epub 2013/01/22. pmid:23341592; PubMed Central PMCID: PMC3568309.
  61. 61. Rothenbacher FP, Zhu J. Efficient responses to host and bacterial signals during Vibrio cholerae colonization. Gut Microbes. 2014;5(1):120–8. Epub 2013/10/31. pmid:24256715; PubMed Central PMCID: PMC4049929.
  62. 62. Provenzano D, Lauriano CM, Klose KE. Characterization of the role of the ToxR-modulated outer membrane porins OmpU and OmpT in Vibrio cholerae virulence. J Bacteriol. 2001;183(12):3652–62. Epub 2001/05/24. pmid:11371530; PubMed Central PMCID: PMC95243.
  63. 63. Vasan AK, Haloi N, Ulrich RJ, Metcalf ME, Wen PC, Metcalf WW, et al. Role of internal loop dynamics in antibiotic permeability of outer membrane porins. Proc Natl Acad Sci U S A. 2022;119(8). pmid:35193963; PubMed Central PMCID: PMC8872756.
  64. 64. Sentausa E, Basso P, Berry A, Adrait A, Bellement G, Coute Y, et al. Insertion sequences drive the emergence of a highly adapted human pathogen. Microb Genom. 2019. Epub 2019/04/05. pmid:30946644.
  65. 65. Song W, Bajaj H, Nasrallah C, Jiang H, Winterhalter M, Colletier JP, et al. Understanding Voltage Gating of Providencia stuartii Porins at Atomic Level. PLoS Comput Biol. 2015;11(5):e1004255. Epub 20150508. pmid:25955156; PubMed Central PMCID: PMC4425685.
  66. 66. Ulmke C, Kreth J, Lengeler JW, Welte W, Schmid K. Site-directed mutagenesis of loop L3 of sucrose porin ScrY leads to changes in substrate selectivity. J Bacteriol. 1999;181(6):1920–3. pmid:10074088; PubMed Central PMCID: PMC93594.
  67. 67. Phale PS, Schirmer T, Prilipov A, Lou KL, Hardmeyer A, Rosenbusch JP. Voltage gating of Escherichia coli porin channels: role of the constriction loop. Proc Natl Acad Sci U S A. 1997;94(13):6741–5. pmid:9192635; PubMed Central PMCID: PMC21228.
  68. 68. Soares CM, Bjorksten J, Tapia O. L3 loop-mediated mechanisms of pore closing in porin: a molecular dynamics perturbation approach. Protein Eng. 1995;8(1):5–12. pmid:7770452.
  69. 69. Cowan SW, Schirmer T, Rummel G, Steiert M, Ghosh R, Pauptit RA, et al. Crystal structures explain functional properties of two E. coli porins. Nature. 1992;358(6389):727–33. pmid:1380671.
  70. 70. Phale PS, Philippsen A, Kiefhaber T, Koebnik R, Phale VP, Schirmer T, et al. Stability of trimeric OmpF porin: the contributions of the latching loop L2. Biochemistry. 1998;37(45):15663–70. pmid:9843370.
  71. 71. Lei J, Sun L, Huang S, Zhu C, Li P, He J, et al. The antimicrobial peptides and their potential clinical applications. Am J Transl Res. 2019;11(7):3919–31. Epub 20190715. pmid:31396309; PubMed Central PMCID: PMC6684887.
  72. 72. Brogden KA. Antimicrobial peptides: pore formers or metabolic inhibitors in bacteria? Nat Rev Microbiol. 2005;3(3):238–50. pmid:15703760.
  73. 73. Bahar AA, Ren D. Antimicrobial peptides. Pharmaceuticals (Basel). 2013;6(12):1543–75. Epub 20131128. pmid:24287494; PubMed Central PMCID: PMC3873676.
  74. 74. Trimble MJ, Mlynarcik P, Kolar M, Hancock RE. Polymyxin: Alternative Mechanisms of Action and Resistance. Cold Spring Harb Perspect Med. 2016;6(10). Epub 20161003. pmid:27503996; PubMed Central PMCID: PMC5046685.
  75. 75. Lal S, Tabacchioni S. Ecology and biotechnological potential of Paenibacillus polymyxa: a minireview. Indian J Microbiol. 2009;49(1):2–10. Epub 20090421. pmid:23100748; PubMed Central PMCID: PMC3450047.
  76. 76. Chaba R, Alba BM, Guo MS, Sohn J, Ahuja N, Sauer RT, et al. Signal integration by DegS and RseB governs the &#963; E-mediated envelope stress response in Escherichia coli. Proc Natl Acad Sci U S A. 2011;108(5):2106–11. Epub 20110118. pmid:21245315; PubMed Central PMCID: PMC3033255.
  77. 77. Sohn J, Sauer RT. OMP peptides modulate the activity of DegS protease by differential binding to active and inactive conformations. Mol Cell. 2009;33(1):64–74. pmid:19150428; PubMed Central PMCID: PMC2742306.
  78. 78. Kanungo S, Azman AS, Ramamurthy T, Deen J, Dutta S. Cholera. Lancet. 2022;399(10333):1429–40. pmid:35397865.
  79. 79. Das B, Verma J, Kumar P, Ghosh A, Ramamurthy T. Antibiotic resistance in Vibrio cholerae: Understanding the ecology of resistance genes and mechanisms. Vaccine. 2019. Epub 2019/07/06. pmid:31272870.
  80. 80. Mashe T, Domman D, Tarupiwa A, Manangazira P, Phiri I, Masunda K, et al. Highly Resistant Cholera Outbreak Strain in Zimbabwe. N Engl J Med. 2020;383(7):687–9. pmid:32786196.
  81. 81. Buchfink B, Xie C, Huson DH. Fast and sensitive protein alignment using DIAMOND. Nat Methods. 2015;12(1):59–60. Epub 20141117. pmid:25402007.
  82. 82. Li W, Godzik A. Cd-hit: a fast program for clustering and comparing large sets of protein or nucleotide sequences. Bioinformatics. 2006;22(13):1658–9. Epub 20060526. pmid:16731699.
  83. 83. Edgar RC. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res. 2004;32(5):1792–7. Epub 20040319. pmid:15034147; PubMed Central PMCID: PMC390337.
  84. 84. Stamatakis A. RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics. 2014;30(9):1312–3. Epub 20140121. pmid:24451623; PubMed Central PMCID: PMC3998144.
  85. 85. Skorupski K, Taylor RK. Positive selection vectors for allelic exchange. Gene. 1996;169(1):47–52. pmid:8635748.
  86. 86. Madera M. Profile Comparer: a program for scoring and aligning profile hidden Markov models. Bioinformatics. 2008;24(22):2630–1. Epub 20081009. pmid:18845584; PubMed Central PMCID: PMC2579712.
  87. 87. Waterhouse A, Bertoni M, Bienert S, Studer G, Tauriello G, Gumienny R, et al. SWISS-MODEL: homology modelling of protein structures and complexes. Nucleic Acids Res. 2018;46(W1):W296–W303. pmid:29788355; PubMed Central PMCID: PMC6030848.
  88. 88. Mackie AM, Hassan KA, Paulsen IT, Tetu SG. Biolog Phenotype Microarrays for phenotypic characterization of microbial cells. Methods Mol Biol. 2014;1096:123–30. pmid:24515365.
  89. 89. Levin-Reisman I, Ronin I, Gefen O, Braniss I, Shoresh N, Balaban NQ. Antibiotic tolerance facilitates the evolution of resistance. Science. 2017;355(6327):826–30. Epub 20170209. pmid:28183996.