Diverse Aquatic Animal Matrices Play a Key Role in Survival and Potential Virulence of Non-O1/O139 Vibrio cholerae Isolates

Vibrio cholerae can cause pandemic cholera in humans. The waterborne bacterium is frequently isolated from aquatic products worldwide. However, current literature on the impact of aquatic product matrices on the survival and pathogenicity of cholerae is rare. In this study, the growth of eleven non-O1/0O139 V. cholerae isolates recovered from eight species of commonly consumed fish and shellfish was for the first time determined in the eight aquatic animal matrices, most of which highly increased the bacterial biomass when compared with routine trypsin soybean broth (TSB) medium. Secretomes of the V. cholerae isolates (draft genome size: 3,852,021–4,144,013 bp) were determined using two-dimensional gel electrophoresis (2DE-GE) and liquid chromatography-tandem mass spectrometry (LC-MS/MS) techniques. Comparative secretomic analyses revealed 74 differential extracellular proteins, including several virulence- and resistance-associated proteins secreted by the V. cholerae isolates when grown in the eight matrices. Meanwhile, a total of 8,119 intracellular proteins were identified, including 83 virulence- and 8 resistance-associated proteins, of which 61 virulence-associated proteins were absent from proteomes of these isolates when grown in the TSB medium. Additionally, comparative genomic and proteomic analyses also revealed several strain-specific proteins with unknown functions in the V. cholerae isolates. Taken, the results in this study demonstrate that distinct secretomes and proteomes induced by the aquatic animal matrices facilitate V. cholerae resistance in the edible aquatic animals and enhance the pathogenicity of the leading waterborne pathogen worldwide.


INTRODUCTION
Vibrio cholerae is a Gram-negative bacterium that is found growing in brackish coastal waters and estuaries worldwide (Yen and Camilli, 2017). Ingestion of water or aquatic products contaminated with toxic V. cholerae can cause cholera in humans (Ramamurthy et al., 2020). In the Indian subcontinent, the description of a disease resembling cholera has been mentioned in Sushruta Samita, estimated to have been written between ∼400 and 500 BC (Siddique and Cash, 2014). In 1817, the first pandemic of cholera spread from India to several other regions of the world (Lippi et al., 2016). Consequently, six additional major pandemics have occurred, the latest of which originated in Indonesia in the 1960s and was still ongoing (Rabaan, 2019). It was estimated that cholera caused ∼95,000 deaths annually in endemic countries, such as India, Ethiopia, and Nigeria (Ali et al., 2015). To date, V. cholerae isolates have been classified into more than 200 serogroups (Zmeter et al., 2018), of which serogroups O1 and O139 can cause cholera outbreaks. Nevertheless, clinical infections with non-O1/O139 V. cholerae isolates have been reported (Zmeter et al., 2018). These pathogens can cause intestinal, extra-intestinal diseases, and even death (Zmeter et al., 2018). For example, a pathogenic and non-O1/O139 V. cholerae AM-19226 strain colonized human intestinal epithelial cells by a Type III secretory system (T3SS), disrupted homeostasis, and caused diarrhea disease (Miller et al., 2016). Virulence-related proteins have been identified in non-O1/O139 V. cholerae isolates, e.g., a cholix toxin (Chx), a multifunctional autoprocessing repeats-in-toxin (MARTX), a GlcNAc binding protein A (GbpA), a V. cholerae cytolysin (VCC), and an outer membrane protein OmpU (Ramamurthy et al., 2020). For instance, intraperitoneal injection of Chx in mice can cause a lethal hemorrhagic inflammatory and cytotoxic response in the liver (Ogura et al., 2017). Non-O1/O139 V. cholerae isolates exerted cell rounding via MARTX that induced the depolymerization of actin fibers of host cells (Fullner and Mekalanos, 2000). The binding of V. cholerae by GbpA led to increased mucus production, which drew more bacteria for better colonization of the host intestinal surface (Rothenbacher and Zhu, 2014). VCC was a membrane-damaging protein toxin with potent cytolytic/cytotoxic activity against a wide range of eukaryotic cells (Kathuria and Chattopadhyay, 2018), while OmpU was a key adhesion protein and an important virulence factor for the successful colonization of Vibrio species into the host (Ganie et al., 2021). Therefore, the identification of virulence determinants in non-O1/O139 V. cholerae isolates is imperative to ensure food safety systems and human health.
Several types of secretory systems have been identified in pathogenic bacteria (Ratner et al., 2017). Secreted proteins play very important roles in bacterial signaling, cell to cell communication, and survival in the environment and hosts (Stastna and Van Eyk, 2012). For instance, cholera toxin (CT), a major toxin of V. cholerae, is secreted by a Type II secretory system (T2SS) (Rasti and Brown, 2019). V. cholerae AM-19226 that lacks CT and toxin coregulated pilus (TCP), a type IV pilus required for V. cholerae pathogenesis, can cause enterotoxicity by T3SS pathogenic islands similar to that in pathogenic Vibrio parahaemolyticus (Megli and Taylor, 2013;Rivera-Cancel and Orth, 2017). Bacterial Type VI secretory system (T6SS) can also secret virulence factors (Miller, 2013), such as a hemolysin co-regulatory protein (Hcp) and a valine-glycine repeat protein G (VgrG) (Li et al., 2019). For example, three VgrG alleles have been reported in V. cholerae: VgrG-1 exhibited actin cross-linking activity with cytotoxic effect on eukaryotic cells; VgrG-2 was essential in regulating bacterial movement and biofilm formation; and VgrG-3 showed an antibacterial function by hydrolyzing cell wall of Gram-negative bacteria (Sha et al., 2013;Arteaga et al., 2020).
Two-dimensional gel electrophoresis (2D-GE) is one of the most versatile and widely used techniques to study the proteomics of a biological system (Meleady, 2018). The combination of 2D-GE with liquid chromatography-tandem mass spectrometry (LC-MS/MS) can provide basic information from protein identity to sample heterogeneity analysis. For instance, Mir et al. identified 150 differentially regulated proteins of virulent Salmonella enterica serovar Typhi in a model host of Caenorhabditis elegans using 2D-GE analysis (Mir et al., 2021). Monteiro et al. (2016) identified several differential cytoplasmic proteins between drug-resistant Escherichia coli 5A, 10A, 12A, and 23B strains and non-resistant E. coli L137 using 2D-GE combined with MS analysis, including a betalactamase Tem, an outer membrane protein A, and a betalactamase Toho-2. Zhu et al. (2020) obtained secretomes of 12 V. parahaemolyticus isolates and identified 8 virulence-associated proteins using 2D-GE and LC-MS/MS techniques, including a superoxide dismutase, a maltoporin, an outer membrane channel TolC, an enolase, an elongation factor Tu, a polar flagellin B/D, a transaldolase, and a flagellin C. Recently, Shan et al. (2021) determined secretomes and proteomes of 20 V. cholerae isolates by 2D-GE and LC-MS/MS analyses, and identified 11 extracellular and 22 intracellular virulence-associated proteins.
In our previous studies, 2D-GE conditions have been optimized recently (Zhu et al., 2020). Secretomes of V. parahaemolyticus isolates recovered from 12 species of edible aquatic animals were characterized, and differentially secreted proteins were identified using 2D-GE and LC-MS/MS techniques (Zhu et al., 2020). China is the largest producer, exporter, and consumer of aquatic products worldwide. The total aquatic product output was ∼65,490,200 tons in China in 2020, which accounted for nearly 60% of the world's total output (National Bureau of Statistics, http://www.stats.gov.cn/, accessed on 20 August 2021). Recently, several V. cholerae isolates were recovered from 36 species of edible aquatic animals and identified by our research group . Secretomes and proteomes of 20 V. cholerae isolates incubated in the routine trypsin soybean broth (TSB) medium were determined using 2D-GE and LC-MS/MS techniques (Shan et al., 2021). Based on our previous studies, we, therefore, asked whether various aquatic product matrices could affect the survival and pathogenicity of V. cholerae isolates of aquatic animal origins. Thus, the major objectives of this study were: (1) to determine the growth of eleven non-O1/O139 V. cholerae isolates in eight types of commonly consumed fish and shellfish matrix media; (2) to employ 2D-GE and LC-MS/MS techniques to obtain secretomes and proteomes of these V. cholerae isolates when incubated in the matrix media; and (3) to identify virulence-and resistanceassociated proteins of the V. cholerae isolates induced by the matrices. To the best of our knowledge, this study was the first to investigate the impact of various aquatic animal matrices on the survival, secretomes, and proteomes of V. cholerae isolates.
The results of this study will support the increased demand for new vaccine targets for food safety control of V. cholerae contamination in edible aquatic animals.

V. cholerae Isolates and Culture Conditions
The non-O1/O139 V. cholerae strains used in this study ( Table 1) were isolated and characterized by Su and Chen (2020), and stored at −80 • C freezer in our laboratory at Shanghai Ocean University, Shanghai, China. V. cholerae isolates ( Table 1) were inoculated in TSB (Beijing Luqiao Technology Co., Ltd., Beijing, China) (pH 8.5, 3% of NaCl) or aquatic product matrix media (see below), and incubated at 37 • C, respectively. Growth curves of V. cholerae isolates were determined using Bioscreen Automatic Growth Curve Analyzer (BioTek Instruments, Inc., Winooski, VT, USA). Bacterial cells grown to the mid-logarithmic phase without shaking were collected by centrifugation for extracellular protein extraction, or to the late logarithmic phase shaking at 180 rpm for intracellular protein extraction as described previously (Zhu et al., 2020).

Preparation of Aquatic Product Matrix Media
The eight species of edible aquatic animals included 4 species of fish: Aristichthys nobilis, Carassius auratus, Ctenopharyngodon idellus, and Parabramis pekinensis; and 4 species of shellfish: Mactra antiquata, Mactra quadrangularis Deshayes, Perna viridis, and Paphia undulata ( Table 1) Aquatic product matrix media were prepared according to the method described by Wang et al. (2017) with minor modifications. Briefly, aliquots of 200 g (wet weight) of fish meat samples (fish skin was removed) were cut into 2 × 2 × 2 cm pieces using sterile knives, while ∼100 g (wet weight) of shellfish meat samples (shell was removed) were washed with sterile water. The processed samples were collected in sterile sealed plastic bags (Maojie, Nanjing, China), and stored at −20 • C for 4 days. Subsequently, the samples were transferred to 4 • C for 3 days to collect the matrix leach solution. After freezing and thawing twice, the collected matrix leach solution was centrifuged at 10,000 g for 20 min at 4 • C, and the supernatant was filtered through a.22-µm-pore membrane filter (Millipore, Bedford, MA, USA). The sterile filtrate derived from each of the eight types of aquatic animal samples was stored at −80 • C and used as matrix media ( Table 1). The matrix leaching rate was calculated by the percentage of leaching solution mass to the processed sample mass. All tests were performed in triplicates.

Measurement of Crude Protein, Carbohydrate, and Fat Components
Protein concentrations of the matrix media were measured using Bradford Protein Assay Kit (Shanghai Sangon Biological Engineering Technology and Service Co., Ltd., Shanghai, China) according to the manufacturer's instructions, and serum albumin was used as the standard protein. Crude fat contents were determined using Automatic Soxhlet Fat Extraction System (Gerhardt, Bonn, Germany) (Shin et al., 2013). Carbohydrate contents were measured by the phenol-sulfuric acid method using Synergy 2 Multifunctional Microplate Reader (BioTek, Maricopa, USA) (Albalasmeh et al., 2013). The pH was measured using an electronic pH meter (Mettler Toledo FiveEasy Plus, Shanghai, China). All tests were performed in triplicates.

2D-GE Analysis
Extracellular proteins of the V. cholerae isolates were extracted according to the method described previously (Zhu et al., 2020). Intracellular proteins were extracted using Bacterial Protein Extraction Kit (Sangon, Shanghai, China) containing protease inhibitors, following the manufacturer's instructions.
After hydration, IEF and SDS-PAGE analyses, the gels were stained using Protein Stains K (Sangon, Shanghai, China) according to the manufacturer's instructions. Silver-stained gels were scanned, and protein spot detection, matching, and quantitative intensity analysis were performed using PDQuest Advanced 8.0.1 software (Bio-RAD, Hercules, USA), as described previously (Zhu et al., 2020;Shan et al., 2021).

LC-MS/MS Analysis
The LC-MS/MS analysis was carried out at HooGen Biotech, Shanghai, China using Q Executive Mass Spectrometer (Thermo Fisher Scientific (TFS), Waltham, MA, USA) coupled with Easy nLC 1200 Chromatography System with the same parameters described previously (Zhu et al., 2020). The automated peptide identification using UniProt V. cholerae 89344 20210707 databases (download in July 2021) in Mascot version 2.2 server (Matrix Science, London, United Kingdom), as described previously (Zhu et al., 2020;Shan et al., 2021).
Genome circular maps of the eleven V. cholerae isolates were constructed based on the reference genome of V. cholerae MS6 (GenBank accession number: NZ_AP014524.1), which was retrieved from the Genome Database of National Center for Biotechnology Information (NCBI) (http://www.ncbi.nlm. nih.gov/genome). The Blastcluster software (http://www.ncbi. nlm.nih.gov/) was used for the pangenome analysis. Common and strain-specific genes were predicted using OrthoMCL software (http://orthomcl.org/common/downloads/software/).

Quantitative Reverse Transcription-Polymerase Chain Reaction Assay
The qRT-PCR assay was performed according to the method described previously (Zhu et al., 2020). The 16S RNA was used as the internal reference gene (Zhu et al., 2020). All RT-PCRs were performed using CFX96 Real-Time PCR Detection System (Bio-Rad, Hercules, USA). In this study, all tests were performed in triplicate. The data were analyzed using SPSS statistical analysis software version 17.0 (SPSS Inc., Armonk, NY, USA).

Genome Features of the Eleven V. cholerae Isolates Originating in Edible Aquatic Animals
Phenotypes of the eleven V. cholerae isolates were characterized (Su and Chen, 2020), which were recovered from eight species of commonly consumed fish and shellfish ( Table 1). Draft genome sequences of these isolates ranged from 3,852,021 to 4,144,013 bp, with G+C contents from 47.37 to 47.7% (Shan et al., 2021). Remarkably, among the annotated 3,394-3,717 genes, ∼524-559 were predicted as virulence-related genes against the VFDB in the eleven V. cholerae genomes, which referred to adhesion, invasion, damage of host cells and tissues, regulation of virulence, motility, biofilm formation, intracellular and extracellular survival, or nutrient acquisition (Webb and Kahler, 2008). Moreover, ∼186-207 genes were predicted as resistance-related genes; ∼322-347 genes coded for proteins with signal peptides; and 829-885 genes for transmembrane transporters in the eleven V. cholerae genomes. Additionally, several CRISPR-Cas repeat arrays (n = 2-6), and prophage gene clusters (n = 0-3), ranging from 9,415 to 33,111 bp, were identified (Supplementary Table S1), implying possible horizontal gene transfer during the genome evolution of these V. cholerae isolates. The genome circular maps of the eleven V. cholerae isolates were constructed (Figure 1).
Comparative genomic analyses revealed that V. cholerae b9-50 isolate, originating from the shellfish M. antiquata, had the largest genome size (4,144,013 bp), whereas V. cholerae L10-6 isolate, from the fish A. nobilis, was the smallest (3,852,021 bp). The maximum number of virulence-related genes (n = 559) were identified in V. cholerae J9-62 isolate from the fish C. auratus, whereas V. cholerae Q6-10 isolate from the fish C. idellus had the minimum (n = 524). Interestingly, V. cholerae J9-62 also carried the maximum number of resistance-related genes (n = 207), while V. cholerae N3-6, N4-21, N8-88, and Q6-10 isolates, which originated from the shellfish P. undulata, P. viridis, M. quadrangularis Deshayes, and fish C. idellus, respectively, contained the minimum (n = 186). V. cholerae b9-50 genome The larger and smaller chromosomes of the V. cholerae genomes, respectively. V. cholerae MS6 was used as a reference genome (GenBank accession number: NZ_AP014524.1). Circles from the inward to outside represented GC-skew (values more than zero in purple and less than zero in green), GC content, predicted protein-coding genes of the reference genome, and eleven V. cholerae genomes, respectively. carried 3 prophage gene clusters (total length of 61,553 bp), whereas none was identified in V. cholerae L10-6, and Q10-54 from the fish C. idellus.

Preparation of Eight Types of Aquatic Product Matrix Media
Leaching rates and initial pH values of 8 types of aquatic product matrix media were determined, and the results are shown in Figures 3A,B. After repeated the freezing and thawing of aquatic products twice at −20 • C for 4 days and 4 • C for 3 days, leaching rates of the 8-matrix media ranged from 18.68 to 28.06%, among which the highest one was observed from the shellfish M. antiquata matrix (28.06%), followed by the shellfish P. viridis (28.01%), and P. undulata (27.97%) matrices. Conversely, the leaching rate of the fish A. nobilis matrix was the lowest (18.68%), followed by the fish P. pekinensis (20.97%) and C. idellus (21.57%) matrices.
As shown in Figure 3B, pH values of the 8-matrix media ranged from 7.09 to 7.60, among which the highest pH was observed from P. pekinensis matrix (pH 7.60), whereas the pH of M. quadrangularis Deshayes matrix was the lowest (pH 7.09). results are presented in Figures 4A-C. Carbohydrate contents of the 8-matrix media ranged from 0.17 to 0.83‰, among which the highest one was P. undulata (0.83‰), followed by M. quadrangularis Deshayes (0.82‰) matrix. Conversely, the carbohydrate content of C. idellus matrix medium was the lowest (0.17‰). In addition, in the 4 types of fish matrix media, A. nobilis matrix contained higher carbohydrate content (0.40‰) than the 3 others (0.17-0.29‰); in the 4 types of shellfish matrix media, P. undulata matrix had the highest carbohydrate content (0.83‰), whereas P. viridis matrix the lowest (0.29‰) ( Figure 4A).
As shown in Figure 4C, notably, the fat content of M. quadrangularis Deshayes matrix was also the highest (0.61‰), whereas C. idellus matrix was the lowest (0.27‰) among the 8 types of aquatic product matrix media.

Distinct Secretomes of the V. cholerae Isolates Grown in Diverse Aquatic Product Matrices
Secretomes of the eleven V. cholerae isolates that were incubated in the eight types of aquatic product matrices were obtained by 2D-GE analysis (Figures 6A-K). Secretomic patterns produced by 3 independent 2D-GE experiments of each isolate were consistent (Figures not shown). Comparative secretomic analysis revealed 74 differential extracellular proteins (marked with different red numbers, Figure 6), which were secreted by the V. cholerae isolates when grown in the matrices and TSB medium, respectively. Amino acid sequences of each of these extracellular proteins were further determined by LC-MS/MS analysis.
Similarly, approximately two extracellular proteins were secreted by V. cholerae Q10-54 grown in C. idellus matrix (Figure 6F2), and five in the TSB medium ( Figure 6F1); one extracellular protein was secreted by V. cholerae B1-31 in P. pekinensis matrix (Figure 6A2), and five in the TSB medium ( Figure 6A1); one was secreted by V. cholerae Q6-10 in C. idellus matrix (Figure 6E2), and 15 in the TSB medium ( Figure 6E1).

Identification of Common and Differential Intracellular Proteins of the V. cholerae Isolates Grown in Diverse Aquatic Product Matrices
A total of 8,119 intracellular proteins were identified from the eleven V. cholerae isolates when grown in the eight types of aquatic product matrix media by LC-MS/MS analysis. Comparative proteomic analyses revealed approximately 209 common intracellular proteins shared among the isolates, approximately 160 of which were classified into three major GO categories (Supplementary Figure S1A). Given that multiple biological functions could be assigned for single identified protein, the metabolic processes were most abundant (84.38%, 135/160), followed by cellular processes (82.50%, 132/160), and catalytic activity (75.63%, 121/160). Conversely, intracellular proteins in the developmental process (0.63%, 1/160), reproduction (0.63%, 1/160), and reproductive process (0.63%, 1/160) showed the opposite patterns (Supplementary Figure S1A).

Identification of Common and Differential Intracellular Proteins of the V. cholerae Isolates Grown in Fish Matrices
There were approximately 142 intracellular proteins shared by 6 V. cholerae isolates (B1-31, B8-16, J9-62, L10-6, Q6-10, and Q10-54) of the fish origins when grown in the fish P. pekinensis, C. auratus, A. nobilis, and C. idellus matrices, twelve of which showed no match against the GO database. Functional classification of the other 130 common intracellular proteins into GO categories is shown in Supplementary Figure S2A.

Identification of Common and Differential Intracellular Proteins in the V. cholerae Isolates Grown in Shellfish Matrices
Comparative proteomic analyses also revealed ∼229 intracellular proteins shared by 5 V. cholerae isolates (b9-50, N3-6, N4-21, N8-56, and N8-88) of the shellfish origins when grown  Figure S3A). Similarly, there were 979 differential intracellular proteins produced by the 5 V. cholerae isolates grown in the shellfish matrices, approximately 469 of which showed no match against the GO database. The other 510 differential intracellular proteins were grouped into GO categories (Supplementary Figure S3B).

Effects of Diverse Aquatic Product Matrices on Virulence-and Resistance-Associated Proteins Produced by the V. cholerae Isolates
Distinct secretomes and proteomes of the eleven V. cholerae isolates were induced by the eight types of aquatic animal matrices. When incubated in the TSB medium, only 7 virulenceand 5 resistance-associated extracellular proteins were secreted by these isolates (Shan et al., 2021). Nevertheless, when grown in the aquatic product matrices, 2 additional virulence-associated extracellular proteins were identified. Meanwhile, remarkably, there were 61 additional virulence-associated intracellular proteins were expressed by the V. cholerae isolates, which were absent from their proteomes derived from the TSB medium.
To confirm the identified proteins by 2D-GE and LC-MS/MS analyses, qRT-PCR assay was performed to detect the expression of randomly chosen differential proteins. The obtained data were generally consistent with those by the secretomic and proteomic analyses in this study (Supplementary Table S2).

DISCUSSION
The V. cholerae has been isolated from many species of aquatic animals Chen et al., 2021). In China, freshwater fish production was 25,863,823 tons in 2020 (National Bureau of Statistics, http://www.stats.gov.cn/, accessed on 20 August 2021), and contributed importantly to China's fishery production, the top four famous species of which included C. idellus, A. nobilis, C. auratus, and P. pekinensis. The shellfish production was 14,800,800 tons in 2020 in China, and the main species included M. antiquata, M. quadrangularis Deshayes, P. undulata, and P. viridis. Identification of risk factors in V. cholerae of aquatic product origins is crucial for assuring food safety systems, particularly in developing nations (Chen and Alali, 2018). To the best of our knowledge, this study was the first to determine the survival of V. cholerae isolates in the commonly consumed fish and shellfish matrices. Our results revealed that the 8 types of fish and shellfish matrices (except P. undulata) highly increased the biomass of the eleven V. cholerae isolates, when compared with the routine TSB medium, indicating that the matrices benefited the persistence of V. cholerae in the edible aquatic animals.
Based on the obtained draft genomes in our previous research (Shan et al., 2021), comparative genomic analyses demonstrated considerable genome variation among the V. cholerae isolates. Approximately putative 524-559 virulence-related genes, and 186-207 resistance-related genes were predicted in the eleven V. cholerae genomes. Balakhonov et al. (2015) reported 7 virulence genes and 60 antibiotic resistance genes in the draft genome of V. cholerae El Tor O1 N16961. Verma et al. (2019) identified 117 virulence genes in V. cholerae IDH06781 genome. Compared with the previous reports (Balakhonov et al., 2015;Verma et al., 2019), in this study, the eleven V. cholerae isolates of aquatic animal origins carried more virulence-and resistance-associated genes, which was consistent with their resistance phenotypes to multiple antibiotic drugs and heavy metals (Shan et al., 2021).
In this study, comparative secretomic analyses revealed several extracellular virulence-associated proteins secreted by the V. cholerae isolates when grown in the aquatic product matrices, most of which were different from those in the TSB medium (Shan et al., 2021). Moreover, they were specifically secreted by the V. cholerae isolates in different matrices. For example, the T2SS-related GspH family protein (Spots D2-3, H2-10, J2-10, and K2-6) was secreted by V. cholerae L10-6, N3-6, N8-56, and N8-88 isolates when incubated in A. nobilis, P. undulata, M. quadrangularis Deshayes, and M. quadrangularis Deshayes matrices, respectively. This protein can enhance the adhesion and motility of Pseudomonas aeruginosa (Nguyen et al., 2015). In this study, another T2SS-related icmF protein (Spot G2-10) was secreted by V. cholerae b9-50 grown in M. antiquata matrix. It has been reported that the icmF was involved in motion, adhesion, and conjugation in V. cholerae (Zusman et al., 2004). The chaperone protein Dnak (Spot G2-2), secreted by V. cholerae b9-50 in M. antiquata matrix, played a critical role in maintaining intracellular protein homeostasis and protecting cells from toxic stress in E. coli (Zuiderweg et al., 2017;Zwirowski et al., 2017). The flagellin (Spot D2-1), secreted by V. cholerae L10-6 and Q10-54 in A. nobilis and C. idellus matrices, respectively, was a key subunit protein of flagella and was convinced as a virulence factor that contributes to host cell adhesion and invasion (Hajam et al., 2017). This virulence-related protein was also secreted by the V. cholerae isolates when cultured in TSB medium.
Interestingly, the antitoxins (Spots H2-1, J2-1, and K2-1) were secreted by some V. cholerae isolates when incubated in the shellfish matrices. In bacteria, toxin-antitoxin (T-A) systems can mediate cellular response to external stress by initiating processes, such as biofilm formation and programmed cell death (Álvarez et al., 2020). Recently, Janczak et al. reported that chromosomal localization of PemIK toxin-antitoxin system resulted in the loss of toxicity-characterization of pemIK (Sa1)-Sp from Staphylococcus pseudintermedius. The pemIK(Sa1)-Sp was homologous to the plasmid-encoded, highly toxic PemIKSa TA system in pathogenic Staphylococcus aureus (Janczak et al., 2020). The functionality of type II T-A systems of Gram-positive, strictly anaerobic, spore-forming pathogen Clostridioides difficile R20291 in human intestine has also been evaluated, of which MazEF and RelBE systems were functional in a heterologous expression system, and their corresponding toxins possessed an endoribonuclease activity (Álvarez et al., 2020).
In this study, comparative proteomic analyses revealed 83 putative intracellular virulence-associated proteins produced by the V. cholerae isolates when grown in diverse aquatic product matrices. Adhesion to host cells was necessary for pathogens to invade and persist in host epithelial and immune cells (Huang et al., 2019). In this study, adhesion-related virulence factors were produced by the V. cholerae isolates. For example, the OmpA (Spots S347), expressed by V. cholerae N3-6 in P. undulata matrix, was a key virulence factor mediating bacterial biofilm formation, eukaryotic cell infection, antibiotic resistance, and adherence to host cells in pathogenic Acinetobacter baumannii (Nie et al., 2020). Valeru et al. (2014) reported a regulating rule for OmpA in survival of V. cholerae and outer membrane vesicles as a potent virulence factor for this bacterium toward eukaryotes in the environment.
Bacterial chemotactic system is a typical coupling proteindependent signal transduction system and plays a key role in bacterial colonization and pathogenicity (Huang et al., 2019;Korolik, 2019). In this study, chemotaxis proteins CheW (Spots S219) and CheY (Spots S221) were expressed by V. cholerae b9-50 grown in M. antiquata matrix; CheA (Spots S593) expressed by V. cholerae N4-21 in P. viridis matrix; and a 3'3'-CGAMP-specific phosphodiesterase 2 (Spots S629) expressed by V. cholerae N4-21 in P. viridis matrix. The latter was initially identified in the seventh epidemic V. cholerae and involved in effective intestinal colonization and chemotactic regulation (Deng et al., 2018).
Virulence factors involved in invading and persisting in host cells were also identified in the V. cholerae isolates when incubated in the aquatic product matrices. For example, the DNA-binding protein HU (Spots S280), expressed by V. cholerae b9-50 in M. antiquata matrix, can regulate many cellular processes and the pathogenesis of bacteria, such as survival, stress response, virulence gene expression, and cell division (Martínez et al., 2015;Stojkova et al., 2019). The peptidase B, expressed by 6 V. cholerae isolates of the fish origins, was the main virulence factor in Leishmania spp. (Beyzay et al., 2017;Yao, 2020). The cyclic AMP receptor protein (Spots F263), expressed by V. cholerae B8-16 in P. pekinensis matrix, was an important transcription regulator of Yersinia pestis through direct or indirect mechanisms to regulate virulence and metal acquisition (Ritzert et al., 2019). Manneh-Roussel et al. (2018) reported that the cAMP receptor protein controled V. cholerae gene expression in response to host colonization. The hemolysin protein (Spots F527), expressed by V. cholerae J9-62 in C. auratus matrix, was considered as an important toxic factor of S. aureus, and V. cholerae toxicity (Liu et al., 2020;Wang et al., 2022).
Previous studies have indicated that virulence of marine pathogens was regulated by environmental stress conditions. For example, Zhou et al. recently revealed the role of pivotal virulence regulator ToxR in switching on the viable, but nonculturable state by sensing unfavorable environmental signals such as endogenous reactive oxygen species (hydrogen peroxide, H 2 O 2 ) in Vibrio alginolyticus, which infects humans and aquatic animals causing severe economic losses (Zhou et al., 2022). Yin et al. (2021) provided evidence for putative roles of two (p)ppGpp synthetase genes (relA and spoT) attributing environmental adaption and virulence regulation in V. alginolyticus. A lipid II flippase MviN was found to mediate the regulation of environmental osmotic pressure on esrB of the EsrA-EsrB twocomponent system to control the virulence in the marine fish pathogen Edwardsiella piscicida (Yin et al., 2020b). Shao et al. also found that the interplay between ferric uptake regulator Fur and horizontally acquired virulence regulator EsrB coordinated virulence gene expression in E. piscicida (Shao et al., 2021). A leucyl aminopeptidase PepA was found to bind to and negatively regulate esrB to control virulence in E. piscicida (Yin et al., 2020a). Additionally, Buchad and Nair (2021) reported a new mechanism in which small RNA (sRNA) SprX modulated S. aureus pathogenicity by regulating the regulator WalR of autolysins (Buchad and Nair, 2021). In this study, the aquatic product matrices possibly mediated environmental nutritional and/or osmotic pressure changes, which induced extracellular and intracellular virulence-associated proteins secretion and production in the V. cholerae isolates.
In this study, comparative secretomic and proteomic analyses also revealed extracellular and intracellular resistance-associated proteins in the V. cholerae isolates when incubated in the aquatic product matrix media, indicating different molecular strategies, by which these isolates were resistant to antimicrobial drugs. For example, the beta-lactamase family protein (Spots H2-7, J2-4, and K2-5), secreted by V. cholerae N3-6, N8-56, and N8-88 isolates in P. undulata, and M. quadrangularis Deshayes matrices, played a key role in bacterial resistance to antibiotics (White et al., 2017). The ribosomal RNA large subunit methyltransferase I (Spots B2-1) was secreted by V. cholerae B8-16 in P. pekinensis matrix. Deletion of this methyltransferase I resulted in the increased susceptibility of Melissococcus plutonius to mirosamicin antibiotics (Takamatsu et al., 2018). On the other hand, the ABC transporter substrate binding protein (Spots T388, T490, and T549) was produced by V. cholerae N4-21 when grown in P. viridis matrix, which participated in cell detoxification, antibiotic, and drug efflux in M. tuberculosis and greatly affected the survival and development of many drugresistant strains (Cassio Barreto De Oliveira and Balan, 2020). Moreover, the TolC family outer membrane protein (Spots T446), produced by V. cholerae N4-21 in P. viridis matrix, was present in many pathogenic Gram-negative bacteria, including V. cholerae, E. coli, and P. aeruginosa, and formed an outer membrane channel that removed drugs and toxins from cells (Leong et al., 2014;Pattanayak et al., 2021). The BipA (Spots T119, T445, T797, D190 and D380) was expressed by V. cholerae b9-50, N4-21 and N8-56, and B8-16 and J9-62 isolates, which played an important role for virulence, antimicrobial resistance, and biofilm formation in P. aeruginosa (Gibbs and Fredrick, 2018). The SppA (Spots D540), expressed by V. cholerae J9-62 in C. auratus matrix, was originally described as a signal peptide peptidase, and later shown to be resistant to lantibiotics (Henriques et al., 2020).
Overall, this study was the first to investigate the impact of commonly consumed aquatic animal matrices on the survival and pathogenicity of V. cholerae isolates. The growth of the V. cholerae isolates was highly enhanced when incubated in most matrices compared with the routine TSB medium. Distinct secretomes and proteomes of the eleven V. cholerae isolates were induced by diverse aquatic animal matrices. Comparative secretomic analyses revealed 74 differential extracellular proteins, including several putative virulence-and resistance-associated extracellular proteins in the V. cholerae isolates, when grown in the matrices. Meanwhile, comparative proteomic analyses revealed 83 intracellular virulence-and 8 intracellular resistanceassociated proteins, of which 61 virulence-associated proteins were absent from proteomes of these isolates when grown in the TSB medium. Additionally, comparative genomic and proteomic analyses revealed a number of novel and strain-specific proteins with unknown function in the V. cholerae isolates. In future research, functions of these proteins together with the identified virulence-and resistance-associated proteins should be further investigated by cell and animal mode analysis. Taken together, the results in this study demonstrate that distinct secretomes and proteomes induced by the aquatic animal matrices facilitate V. cholerae persistence in the edible aquatic animals and enhance the pathogenicity of the leading waterborne pathogen worldwide.

DATA AVAILABILITY STATEMENT
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/Supplementary Material.

AUTHOR CONTRIBUTIONS
LY, YJ, XP, SQ, and LC participated in the design and/or discussion of the study. LY carried out the experiments and drafted the original manuscript. BZ and YX participated in the data analysis. LC revised the manuscript. All authors read and approved the final version to be published.

FUNDING
This study was supported by the grants from Shanghai Municipal Science and Technology Commission (No. 17050502200) and National Natural Science Foundation of China (No. 31671946).