Vibrio mimicus bacteria are native to aquatic environments but have the potential to cause diseases in animals and humans, such as gastroenteritis and cholera-like diarrhea (1). In Vibrio cholera bacteria, the main toxigenic factor is cholera toxin (CTX), which is encoded by ctxA and ctxB genes and is part of the bacteriophage CTXФ. The acquisition of CTXФ phage by V. cholerae bacteria was associated with the toxin coregulated pilus (TCP), which is involved in intestinal colonization and aggregation. This virulence factor is encoded in an operon in the Vibrio pathogenicity island 1 (VPI-1), and tcpA is the main structural subunit of that pilus. In addition to TCP, VPI-1 also harbors the acfA-D operon, which also plays a role in colonization (2).

In 2004, the largest documented foodborne outbreak of V. mimicus occurred in Thailand, in which 306 persons experienced symptoms including diarrhea, abdominal pain, and vomiting (3), but the virulome associated with these strains was not verified. In 2010, in the United States, a cluster of severe diarrheal diseases was caused by V. mimicus strains carrying CTX (4). In 2019, V. mimicus bacteria caused a seafood-associated outbreak in Florida (USA), in which the patients experienced severe diarrhea, although the strains were CTX-negative (5). However, the virulome of most genomes analyzed (n = 33) was not explored, leaving a gap regarding the association of the strains or lineages with virulence factors. To fill this gap, we analyzed 44 V. mimicus genomes, 35 from GenBank and 9 environmental genomes from Brazil and Japan that we sequenced by using an Illumina Hiseq 2500, assembled by using SPAdes 3.15.2 (https://github.com/ablab/spades), and then analyzed by using Abricate (https://github.com/tseemann/abricate) and the Comprehensive Antibiotic Resistance Database (https://card.mcmaster.ca) and the Virulence Factor Database (http://www.mgc.ac.cn/VFs/main.htm).

Figure

Figure. Maximum-likelihood phylogenetic tree of 44 Vibrio mimicusbacteria genomes from the United States and China. The best evolutionary model (general time reversible plus base frequencies plus ascertainment bias correction...

The phylogenetic analysis based on the core genome revealed clusters; the 2 main clusters had genomes that had been circulating in China and the United States for decades (Figure). These 2 clusters are characterized by distinct virulomes; 1 co-harbored ctxA, ctxB, ace, zot, TCP, and acfA-D (mainly clinical genomes), whereas the other did not have any of those genes. The lineage carrying those virulence genes has persisted for >3 decades (1980–2009), infecting persons in China and the United States. The other lineage, which lacks these virulence factors, also was identified in China (2020) and the United States (1977 [human source]). Another interesting cluster of genomes is the one that covers Brazil (1998 [animal source]) and the United States (2016 [environmental source]).

Our analysis indicates that V. mimicus lineages are disseminated and persist in distinct sources in space and time. In addition, another set of 3 related genomes (VM343, GCA_014525095.1, and GCA_009665195.1) also possessed CTX, TCP, or both, which suggests loss or partial acquisition of the pathogenicity islands of these elements. Of note, other genomes belonging to the same lineage (Figure) appear to have acquired the ctxA or ctxB genes from different sources; GCA_000175995.1, GCA_001471395.2, and GCA_009764005.1 (United States) possessed the ctxB2 genotype (El Tor [Australia]), whereas GCA_000222145.1 (China) had the ctxB1 genotype (classical [strain 569B]). Regarding the TCP cluster, analyses in blastn (https://blast.ncbi.nlm.nih.gov) revealed that all tcpA sequences, except for VM343, were identical and differed from the classical and El Tor genotypes. The tcpA allele carried by most genomes has also been characterized in nonpandemic clinical V. cholerae strains from the United States (6), whereas the tcpA allele of VM343 (Japan [environmental source]) is unique.

We identified several other virulence genes in the genomes (Appendix). We highlight the presence of the heat-stable enterotoxin gene (NAG-ST), identified throughout the phylogeny, and gene clusters related to exopolysaccharide production (vps) and the type III secretion system (T3SS) (vop, vsc, and vcr). We identified T3SS only in genomes that did not carry CTX, TCP, or both, including those from the 2019 outbreak in Florida (5), but 2 closely related genomes (1 of which was identified in the United States) co-carried the T3SS and NAG-ST genes. Because T3SS is a syringe-like protein secretion apparatus, the co-occurrence of this system with a diarrhea-associated toxin could increase the pathogenicity of these strains.

Our findings show that V. mimicus strains are spread throughout the world and that some of them carry a virulome comparable to that of V. cholerae bacteria. The virulome of environmental and clinical strains is apparently not heterogeneous (5), even with the analysis of the new environmental genomes. However, those findings may represent just the tip of the iceberg, given the bias regarding the locality of available genomes. Therefore, more genomic data must be generated to determine whether specialized clinical strains of V. mimicus exist, as they do for V. cholerae bacteria. Furthermore, the environmental and clinical genomes possessed a set of common virulence genes, suggesting that environmental strains have the potential to cause disease in humans. Because V. mimicus already possesses an intrinsic virulome, with the potential to cause disease, the acquisition of virulence determinants such as CTX, TCP, or both could specialize certain lineages, as revealed in our analysis by the clinical lineages that carry these determinants.

Top