Vibrio parahaemolyticus Isolates from Asian Green Mussel: Molecular Characteristics, Virulence and Their Inhibition by Chitooligosaccharide-Tea Polyphenol Conjugates

Fifty isolates of Vibrio parahaemolyticus were tested for pathogenicity, biofilm formation, motility, and antibiotic resistance. Antimicrobial activity of chitooligosaccharide (COS)-tea polyphenol conjugates against all isolates was also studied. Forty-three isolates were randomly selected from 520 isolates from Asian green mussel (Perna viridis) grown on CHROMagarTM Vibrio agar plate. Six isolates were acquired from stool specimens of diarrhea patients. One laboratory strain was V. parahaemolyticus PSU.SCB.16S.14. Among all isolates tested, 12% of V. parahaemolyticus carried the tdh+trh− gene and were positive toward Kanagawa phenomenon test. All of V. parahaemolyticus isolates could produce biofilm and showed relatively strong motile ability. When COS-catechin conjugate (COS-CAT) and COS-epigallocatechin-3-gallate conjugate (COS-EGCG) were examined for their inhibitory effect against V. parahaemolyticus, the former showed the higher bactericidal activity with the MBC value of 1.024 mg/mL against both pathogenic and non-pathogenic strains. Most of the representative Asian green mussel V. parahaemolyticus isolates exhibited high sensitivity to all antibiotics, whereas one isolate showed the intermediate resistance to cefuroxime. However, the representative clinical isolates were highly resistant to nine types of antibiotics and had multiple antibiotic resistance (MAR) index of 0.64. Thus, COS-CAT could be used as potential antimicrobial agent for controlling V. parahaemolyticus-causing disease in Asian green mussel.


Introduction
Vibrio parahaemolyticus has become a serious foodborne pathogen and raised public health concern in Thailand, China, Japan, and other Asian countries [1]. V. parahaemolyticus present in aquatic products contributes to significant economic losses across the entire supply chain [2]. It is a member of the genus Vibrio from family Vibrionaceae, a Gram-negative, marine halophilic bacterium that naturally inhabits in global coastal waters, sediment and various types of marine animals [3] such as fish, shrimp, crab, clam, oyster [4][5][6][7], and mussel [8]. V. parahaemolyticus can also be transmitted to humans when consuming contaminated raw or poorly cooked seafood products [9,10]. This bacterium contributes to an acute gastroenteritis, which includes nausea, diarrhea, vomiting, fever, and chills. It can also cause severe symptoms in children, the elderly, and immunocompromised patients [11,12]. The pathogenic V. parahaemolyticus often infects and causes disease by using several virulence factors. Adhesions, thermostable direct hemolysin (TDH), and TDHrelated hemolysin (TRH) are the most important virulence factors in this bacterium [13]. was adopted for selection of V. parahaemolyticus isolates. These two agar plates were used for confirmation. The result from CHROMagar TM Vibrio plates was mainly used for further experiments. Forty-three isolates with different colony colors from the total 520 of isolates on CHROMagar TM Vibrio agar plate were randomly selected and then characterized and specified by the MALDI-Biotyper ® system (microflex LT; Bruker Daltonik GmbH, Bremen, Germany) [25]. Halophilism was also performed using NaCl-tryptone broth (T 1 N 0 , T 1 N 3 , T 1 N 6 , T 1 N 8 , and T 1 N 10 ).
plates were used for confirmation. The result from CHROMagar TM Vibrio plates w mainly used for further experiments. Forty-three isolates with different colony colors fr the total 520 of isolates on CHROMagar TM Vibrio agar plate were randomly selected a then characterized and specified by the MALDI-Biotyper ® system (microflex LT; Bru Daltonik GmbH, Bremen, Germany) [25]. Halophilism was also performed using Na tryptone broth (T1N0, T1N3, T1N6, T1N8, and T1N10).
Fifty isolates used in this study were molecularly identified and confirmed for parahaemolyticus. Forty-three isolates were retrieved from the Asian green mussel, wh the remaining six isolates were isolated from the stool specimens of diarrhea patients fr Songklanagarind Hospital, Faculty of Medicine and one laboratory strain of V. parahaem lyticus PSU.SCB.16S.14 was gifted by the Food Safety Laboratory, Prince of Songkla U versity, Hat Yai, Thailand.

Types of Places Collected Number of Sample Suratthani
Local market 2 Asian green mussel farm 6 Natural habitat 4 Trang Local market 2 Asian green mussel farm 4 Natural habitat 2 Songkhla Local market 8

Polymerase Chain Reaction (PCR) Assay
Twenty microliters of glycerol stock of V. parahaemolyticus isolates (n = 50) was in ulated into 5 mL of Luria-Bertani (LB) broth (Merck, Burlington, MA, USA) containing NaCl (w/v) and incubated at 37 °C for 16-18 h, followed by centrifugation (8000 × g fo min). The genomic DNA was isolated using a PureLink TM Genomic DNA Mini Kit (In trogen, Thermo Fisher Scientific, Waltham, MA, USA). DNA concentration was measur with the aid of a NanoDrop spectrophotometry (Thermo Fisher Scientific, Waltham, M USA). PCR primers were synthesized via Integrated DNA Technologies (Singapore c Singapore) as shown in Table 2. PCR reaction mixture comprised 5 µL of 4 × Taq P Mastermix (QIAGEN, Germantown, MD, USA), 2 µL of genomic DNA (50 ng/µL), 0.5 of primer pair solution (10 µM each), and 12 µL of Rnase free water. PCR was amplif under the selected conditions: pre-denaturation at 95 °C for 2 min, 30 cycles  Fifty isolates used in this study were molecularly identified and confirmed for V. parahaemolyticus. Forty-three isolates were retrieved from the Asian green mussel, while the remaining six isolates were isolated from the stool specimens of diarrhea patients from Songklanagarind Hospital, Faculty of Medicine and one laboratory strain of V. parahaemolyticus PSU.SCB.16S.14 was gifted by the Food Safety Laboratory, Prince of Songkla University, Hat Yai, Thailand.

Polymerase Chain Reaction (PCR) Assay
Twenty microliters of glycerol stock of V. parahaemolyticus isolates (n = 50) was inoculated into 5 mL of Luria-Bertani (LB) broth (Merck, Burlington, MA, USA) containing 3% NaCl (w/v) and incubated at 37 • C for 16-18 h, followed by centrifugation (8000 × g for 5 min). The genomic DNA was isolated using a PureLink TM Genomic DNA Mini Kit (Invitrogen, Thermo Fisher Scientific, Waltham, MA, USA). DNA concentration was measured with the aid of a NanoDrop spectrophotometry (Thermo Fisher Scientific, Waltham, MA, USA). PCR primers were synthesized via Integrated DNA Technologies (Singapore city, Singapore) as shown in Table 2. PCR reaction mixture comprised 5 µL of 4 × Taq PCR Mastermix (QIAGEN, Germantown, MD, USA), 2 µL of genomic DNA (50 ng/µL), 0.5 µL of primer pair solution (10 µM each), and 12 µL of Rnase free water. PCR was amplified under the selected conditions: pre-denaturation at 95 • C for 2 min, 30 cycles for denaturation at 95 • C for 5 s, annealing at 58 • C for 15 s, and extension at 72 • C for 10 s, and ending extension at 72 • C for 5 min [26]. PCR products were finally determined using 2% agarose gel electrophoresis.

Preparation of COS-Tea Polyphenol Conjugates Using Free Radical Grafting Method
COS-CAT and COS-EGCG conjugates were prepared using free radical grafting method as tailored by Mittal et al. [21]. First, pH of COS solution (1%, w/v) was adjusted to 5.0 using acetic acid (1 M). Simultaneously, 1 M H 2 O 2 (4 mL) containing 0.10 g ascorbic acid were incubated (40 • C, 10 min) to generate hydroxyl radicals. Both solutions were then mixed and the mixture was incubated at room temperature for 1 h with continuous stirring. CAT and EGCG (10%, w/w of COS) were then added into the mixture and incubated for 24 h in dark, at room temperature. With dialysis against distilled water, the unbound CAT and EGCG were removed. COS-CAT and COS-EGCG conjugate powders were obtained after lyophilization of dialysates.

Determination of Minimum Inhibitory Concentration (MIC) and Minimum Bactericidal Concentration (MBC)
MIC and MBC of COS-CAT and COS-EGCG toward V. parahaemolyticus isolates were measured following the guidelines of Clinical and Laboratory Standards Institute (CLSI). Overnight culture (18-24 h) of V. parahaemolyticus isolate was adjusted to a final concentration of 10 8 CFU/mL (corresponding to approximately 0.5 McFarland standard). The standardized suspension was then diluted by 200-fold with Mueller Hilton Broth (MHB) (Difco TM , Baltimore, MD, USA) supplemented with 3% NaCl (w/v), namely "diluent" to obtain the working concentration of 10 6 CFU/mL. The COS-CAT and COS-EGCG powders were dissolved and diluted with deionized water [28]. Stock solution was subjected to two-fold dilution to attain the highest concentrations of 2.048 mg/mL and the lowest concentration of 0.004 mg/mL. One-hundred microliters of bacterial suspension and 100 µL of COS-CAT/COS-EGCG working solutions were mixed in each well, and incubated for 24 h at 37 • C. Thereafter, 20 µL of resazurin (0.09%) solutions was added for each well, and further incubated for 3 h at 37 • C. Subsequently, the wells with no color change were scored as "above the MIC value". MBC was determined by plating directly the content of wells with concentration higher than the MIC value. Culture solution (10 µL) was pipetted from each well with no bacterial growth and dropped uniformly on a sterile MHA medium [29], followed by incubation (37 • C for 24 h). MBC was defined as the minimum concentration of COS-tea polyphenol conjugated solutions without colony formation. Positive control consisted of bacterial suspension and diluent, while negative control comprised MHB and diluent.

Biofilm Crystal Violet (CV) Staining
CV staining method was adopted [30]. Overnight cultures were diluted 50-fold using 200 µL of Oxoid TSB broth (Oxoid Ltd., Hampshire, England) containing 3% NaCl (w/v) (TSB-N) in 96-well plates (Corning Inc., Corning, NY, USA). Culture was allowed to proliferate at 37 • C for 48 h. The cultures were removed and the well with the adherent biofilm was gently washed with 200 µL of sterile phosphate buffered saline (PBS) for three times. Then, 200 µL of 0.1% crystal violet was used to stain the surface-attached cells for 15 min. After solution removal, the well was thoroughly washed with sterile H 2 O for three times. Bound dye in each well was solubilized using 200 µL of ethanol (Analytical grade ≥ 99.9% in pure, RCI Labscan™, Bangkok, Thailand). Absorbance at 570 nm (A 570 ) was measured.

Analysis of Swimming and Swarming Motility
Swimming and swarming abilities of V. parahaemolyticus isolates were examined on semi-solid swimming plated (TSB-N in the presence of 0.2% agar) and solid swarming plates (TSB-N containing 0.5% agar), respectively [30,31]. The overnight cultures were diluted 50-fold using 5 mL of TSB-N broth and cultured at 37 • C with continuous shaking (200 rpm) until A 570 reached 1.2-1.4. Those cultures were used for testing.

Kanagawa Phenomenon (KP) Test
KP test was performed as detailed by Zhang et al. [32]. First, 2 µL of the third-round cell cultures were inoculated onto Wagatsuma agar medium consisting of 5% rabbit red blood cells (RBCs). The radius from the inoculation place to the edge of β-hemolysin zone was detected after static incubation (37 • C for 24 h).

Antibiotic Susceptibility Testing
Antimicrobial susceptibility testing of eight V. parahaemolyticus isolated from clinical sample, Asian green mussel samples from different origins and laboratory strain were performed using the Sensititre TM microbroth dilution system (Trek Diagnostic Systems, Cleveland, OH, USA) [33]. Cultures were grown overnight on TSA supplemented with 2.5% NaCl (w/v) plates at 37 • C. The cultures were transferred to sterile demineralized 2.5% saline solution to obtain the turbidity, equivalent to that of 0.5 McFarland standard. One-hundred milliliters of each suspension were transferred into sterile cation-adjusted MHB, and broth solution (50 mL) was dispersed onto CML1FMAR custom MIC plates  [34]. Resistance breakpoints were also used [34]. Multiple antibiotic resistance (MAR) index was calculated as tailored by Krumperman [35], in which the following equation was used: where "a" is the number of antibiotics, to which the particular isolate was resistant and "b" is the total number of antibiotics tested.

Statistical Analyses
Completely randomized design (CRD) was used for the entire study. The experiments and analyses were conducted in triplicate. Data were subjected to one-way analysis of variance (ANOVA) and a least significant difference test was used. p < 0.05 was considered a significant difference.

Characteristics of V. parahaemolyticus Isolates
All fifty collected isolates from Asian green mussel samples, clinical and laboratory strains, were identified as V. parahaemolyticus based on their morphological and biochemical characteristics. Double-plating method was used to identify species involving TCBS and CHROMagar™ Vibrio agars, the selective media providing the direct colony-color-based identification of V. parahaemolyticus by specific color development of the particular colonies. Out of 26 Asian green mussel collected samples, V. parahaemolyticus was detected in all the samples on TCBS agar ( Figure 2A) and CHROMagar TM Vibrio agar ( Figure 2B). Colonies of fifty V. parahaemolyticus isolates appeared. All V. parahaemolyticus colonies were spherical, transparent, and bluish-green or green color on TCBS plates. On CHROMagar TM plates, the colonies of fifty isolates were round, smooth, flat, mauve or purple red or purplish cream colony in color (positive colony = mauve color). On CHROMagar TM Vibrio agar, the colony colors of 50 V. parahaemolyticus were varied. No.1 was a laboratory strain (PSU.SCB.16S.14); No. 2-44 were V. parahaemolyticus isolated from Asian green mussel; and No. 45-50 were V. parahaemolyticus isolated from clinical samples. Lee et al. [36] found that 4 (10.5%) of the 38 V. parahaemolyticus strains had white colonies on ChromoVP agar. Su et al. [37] documented that 5% of V. parahaemolyticus strains appeared as white colonies on Bio-Chrome Vibrio medium. High variability and differential colony colors were observed when culture-based techniques were used for seafood, clinical, and environmental samples. Hence, molecular confirmation must be conducted to ensure the accurate detection of V. parahaemolyticus. All the isolates were also confirmed by MALDI Biotyper ® analysis and thermolabile hemolysin encoded by the tlh gene as species marker. As shown in Figure 3A, 100% of the 49 isolates including a positive laboratory strain (PSU.SCB.16S.14) were tlhpositive. The salt tolerance test also showed that all the recovered strains required sodium ions for their growth in media supplemented with 1% NaCl up to 8%. The result was in agreement with that reported by Beleneva et al. [38]. However, the isolates should be collected from other provinces or different geographic locations to acquire more data, in which a variety and abundance of strains can be gained. and CHROMagar™ Vibrio agars, the selective media providing the direct colony-c based identification of V. parahaemolyticus by specific color development of the parti colonies. Out of 26 Asian green mussel collected samples, V. parahaemolyticus was det in all the samples on TCBS agar (Figure 2A) and CHROMagar TM Vibrio agar (Figur Colonies of fifty V. parahaemolyticus isolates appeared. All V. parahaemolyticus col were spherical, transparent, and bluish-green or green color on TCBS plates. On CH Magar TM plates, the colonies of fifty isolates were round, smooth, flat, mauve or p red or purplish cream colony in color (positive colony = mauve color). On CHROMa Vibrio agar, the colony colors of 50 V. parahaemolyticus were varied. No.1 was a labor strain (PSU.SCB.16S.14); No. 2-44 were V. parahaemolyticus isolated from Asian green sel; and No. 45-50 were V. parahaemolyticus isolated from clinical samples. Lee et al found that 4 (10.5%) of the 38 V. parahaemolyticus strains had white colonies on Chrom agar. Su et al. [37] documented that 5% of V. parahaemolyticus strains appeared as w colonies on Bio-Chrome Vibrio medium. High variability and differential colony c were observed when culture-based techniques were used for seafood, clinical, and ronmental samples. Hence, molecular confirmation must be conducted to ensure th curate detection of V. parahaemolyticus. All the isolates were also confirmed by MA Biotyper ® analysis and thermolabile hemolysin encoded by the tlh gene as species ma As shown in Figure 3A, 100% of the 49 isolates including a positive laboratory s (PSU.SCB.16S.14) were tlh-positive. The salt tolerance test also showed that all the r ered strains required sodium ions for their growth in media supplemented with 1% up to 8%. The result was in agreement with that reported by Beleneva et al. [38]. How the isolates should be collected from other provinces or different geographic locatio acquire more data, in which a variety and abundance of strains can be gained.

Virulence Genes
All fifty isolates identified as V. parahaemolyticus by biochemical, MALDI-Biotyper ® system tests and confirmed by PCR were detected for the tdh and trh genes. DNA fragments of 269 and 500 bp in size were produced from the amplification of V. parahaemolyticus pathogenic tdh and trh genes, respectively ( Table 2). Six out of fifty (12%) samples were positive for the tdh gene (tdh + trh -) ( Figure 3B). However, the isolates of V. parahaemolyticus having both tdh + trh + and tdhtrh + were not detected in this study. Most V. parahaemolyticus clinical isolates had positive result for KP test (Figure 4), thus confirming the presence of hemolysin tdh and/or trh genes [39]. Most V. parahaemolyticus isolated from food and environment do not carry tdh and/or trh genes [39]. V. parahaemolyticus strains having tdh gene and strains possessing both tdh and trh genes were found at very low level in mussel [40,41]. Vibrio species was isolated from bivalves and the culture environments along the Gyeongnam coast in Korea [42]. One hundred and ninety isolates of V. parahaemolyticus from oyster, mussel, and ark shell were negative for the tdh virulence genes, while 18 (9.5%) isolates were positive for trh virulence genes. All strains were positive for the trh gene when isolated from only oyster samples [42]. No trh + V. parahaemolyticus strains was detected in warm climate, including Thailand. Rodriguez-Castro et al. [43] found that trh + strains were dominant in the cold water, whereas tdh+ V. parahaemolyticus disseminated in warm water. Only clinical V. parahaemolyticus strains carried the trh + genes. Bhoopong et al. [44] documented that only 0.5% (3/629) of the clinical V. parahaemolyticus isolates from the 63 patients in Thailand carried the trh gene alone, whereas 87.4% (550/629) and 7% (44/629) of the isolates possessed the tdh gene and both genes, respectively. Chen et al. [45] found that 93% and 1% of the 501 clinical V. parahaemolyticus isolates from southeastern China carried tdh and trh genes, respectively. Nevertheless, distributions of tdh + and/or trh + strains may vary, depending on detection method, sample sources and geographical origin [46].

Hemolytic Activity
KP test was used to determine hemolytic activity of the isolates on the Wagatsuma agar containing 5% RBCs as depicted in Figure 4. Based on KP, the pathogenic isolates of V. parahaemolyticus could be differentiated from non-pathogenic counterpart. When

Virulence Genes
All fifty isolates identified as V. parahaemolyticus by biochemical, MALDI-Biotyper ® system tests and confirmed by PCR were detected for the tdh and trh genes. DNA fragments of 269 and 500 bp in size were produced from the amplification of V. parahaemolyticus pathogenic tdh and trh genes, respectively ( Table 2). Six out of fifty (12%) samples were positive for the tdh gene (tdh + trh − ) ( Figure 3B). However, the isolates of V. parahaemolyticus having both tdh + trh + and tdh − trh + were not detected in this study. Most V. parahaemolyticus clinical isolates had positive result for KP test (Figure 4), thus confirming the presence of hemolysin tdh and/or trh genes [39]. Most V. parahaemolyticus isolated from food and environment do not carry tdh and/or trh genes [39]. V. parahaemolyticus strains having tdh gene and strains possessing both tdh and trh genes were found at very low level in mussel [40,41]. Vibrio species was isolated from bivalves and the culture environments along the Gyeongnam coast in Korea [42]. One hundred and ninety isolates of V. parahaemolyticus from oyster, mussel, and ark shell were negative for the tdh virulence genes, while 18 (9.5%) isolates were positive for trh virulence genes. All strains were positive for the trh gene when isolated from only oyster samples [42]. No trh + V. parahaemolyticus strains was detected in warm climate, including Thailand. Rodriguez-Castro et al. [43] found that trh + strains were dominant in the cold water, whereas tdh+ V. parahaemolyticus disseminated in warm water. Only clinical V. parahaemolyticus strains carried the trh + genes. Bhoopong et al. [44] documented that only 0.5% (3/629) of the clinical V. parahaemolyticus isolates from the 63 patients in Thailand carried the trh gene alone, whereas 87.4% (550/629) and 7% (44/629) of the isolates possessed the tdh gene and both genes, respectively. Chen et al. [45] found that 93% and 1% of the 501 clinical V. parahaemolyticus isolates from southeastern China carried tdh and trh genes, respectively. Nevertheless, distributions of tdh + and/or trh + strains may vary, depending on detection method, sample sources and geographical origin [46]. M48 and M58 from Asian green mussel exhibited weak hemolysis (Figure 4), none of them exhibited strong β-hemolysis. This weak hemolysis might be related with other virulence factors, apart from TDH orTRH. Strains, which produce few extracellular enzymes, could have the weak hemolysis [47]. Although no isolates showed β-hemolysis activity, the potential risk involved in consuming Asian green mussel must be taken into consideration because of its short generation time.

Motility Ability
V. parahaemolyticus has dual flagellar systems, i.e., a single polar flagellum for swimming in liquid and peritrichous lateral flagella for swarming on surfaces [48]. In the present study, swimming and swarming of clinical and Asian green mussel isolates were compared. Mobility abilities of 50 isolates could be classified into three levels: weak, medium, and strong, which respectively indicated that their mobilities were much lower, similar to and significantly higher than those of laboratory strains of V. parahaemolyticus PSU.SCB.16S.14. As shown in Figure 5A, all the 50 isolates were swimmers; 7 isolates were weak swimmers (< 15 mm); 26 isolates were moderate swimmers (< 30 mm); and 17 isolates were strong swimmers (> 30 mm). Similarly, all isolates were swarmers ( Figure 5B). Among all isolates, 43 isolates were moderates swarm cells; and 7 isolates were strong swarm cells. Thus, all isolates showed a relatively strong mobility. V. parahaemolyticus could move via propelling with the aid of flagella. Swimming and swarming behaviors are initial requirement for biofilm formation [49]. All V. parahaemolyticus isolates had relatively strong mobility, associated with their biofilm formation.

Biofilm Formation Capacity
The bacterial biofilm protects pathogens from environmental stress such as antimicrobial and increases disease severity in infected host [50][51][52]. The biofilm was formed by 50 isolates when tested using the CV staining ( Figure 5C). V. parahaemolyticus was able to form biofilms and attached to the surfaces of seafood [53]. Sun et al. [54] found that V. parahaemolyticus isolated from stool specimens of diarrhea patients exhibited biofilm formation. All clinical V. parahaemolyticus isolates were biofilm producers. Biofilm formation is governed by the source of isolates and cultural temperature. In general, pathogenic isolates produced more biofilms than non-pathogenic counterpart [55,56]. Optimum temperature for biofilm formation by V. parahaemolyticus was 37 °C [57]. In general, bacterial cells entrapped in biofilms are more resistant to harsh conditions [53].

Hemolytic Activity
KP test was used to determine hemolytic activity of the isolates on the Wagatsuma agar containing 5% RBCs as depicted in Figure 4. Based on KP, the pathogenic isolates of V. parahaemolyticus could be differentiated from non-pathogenic counterpart. When bacterium lyses human erythrocytes, a pore-forming toxin known as the thermostable direct hemolysin (TDH) was produced. As shown in Figure 4, no Asian green mussel isolates showed β-hemolysis, whereas all of clinical isolates exhibited β-hemolysis; the latter were isolated from the stool of patients. All the tdh + trh − isolates displayed a positive reaction as evidenced by a β-hemolysis zone surrounding the growth spot, whereas all the tdh − trh − isolates showed the negative reaction. Although four isolates namely M6, M13, M48 and M58 from Asian green mussel exhibited weak hemolysis (Figure 4), none of them exhibited strong β-hemolysis. This weak hemolysis might be related with other virulence factors, apart from TDH orTRH. Strains, which produce few extracellular enzymes, could have the weak hemolysis [47]. Although no isolates showed β-hemolysis activity, the potential risk involved in consuming Asian green mussel must be taken into consideration because of its short generation time.

Motility Ability
V. parahaemolyticus has dual flagellar systems, i.e., a single polar flagellum for swimming in liquid and peritrichous lateral flagella for swarming on surfaces [48]. In the present study, swimming and swarming of clinical and Asian green mussel isolates were compared. Mobility abilities of 50 isolates could be classified into three levels: weak, medium, and strong, which respectively indicated that their mobilities were much lower, similar to and significantly higher than those of laboratory strains of V. parahaemolyticus PSU.SCB.16S.14. As shown in Figure 5A, all the 50 isolates were swimmers; 7 isolates were weak swimmers (<15 mm); 26 isolates were moderate swimmers (< 30 mm); and 17 isolates were strong swimmers (> 30 mm). Similarly, all isolates were swarmers ( Figure 5B). Among all isolates, 43 isolates were moderates swarm cells; and 7 isolates were strong swarm cells. Thus, all isolates showed a relatively strong mobility. V. parahaemolyticus could move via propelling with the aid of flagella. Swimming and swarming behaviors are initial requirement for biofilm formation [49]. All V. parahaemolyticus isolates had relatively strong mobility, associated with their biofilm formation.

Antimicrobial Activity of COS-Tea Polyphenol Conjugates toward V. parahaemolyticus Isolates
Antimicrobial effects of COS-tea polyphenol conjugates on clinical and Asian green mussel V. parahaemolyticus isolates were examined. Antimicrobial activity was expressed

Biofilm Formation Capacity
The bacterial biofilm protects pathogens from environmental stress such as antimicrobial and increases disease severity in infected host [50][51][52]. The biofilm was formed by 50 isolates when tested using the CV staining ( Figure 5C). V. parahaemolyticus was able to form biofilms and attached to the surfaces of seafood [53]. Sun et al. [54] found that V. parahaemolyticus isolated from stool specimens of diarrhea patients exhibited biofilm formation. All clinical V. parahaemolyticus isolates were biofilm producers. Biofilm formation is governed by the source of isolates and cultural temperature. In general, pathogenic isolates produced more biofilms than non-pathogenic counterpart [55,56]. Optimum temperature for biofilm formation by V. parahaemolyticus was 37 • C [57]. In general, bacterial cells entrapped in biofilms are more resistant to harsh conditions [53].

Antimicrobial Activity of COS-Tea Polyphenol Conjugates toward V. parahaemolyticus Isolates
Antimicrobial effects of COS-tea polyphenol conjugates on clinical and Asian green mussel V. parahaemolyticus isolates were examined. Antimicrobial activity was expressed as MIC and MBC of COS-tea polyphenol conjugates against 50 V. parahaemolyticus isolates. COS-tea polyphenol conjugates showed the adverse effect on the growth of V. parahaemolyticus  [20,21]. Recently, Mittal et al. [21] reported that COS-CAT conjugate showed higher antioxidant and antimicrobial activities than COS and other COS-polyphenol conjugates. COS-CAT showed antimicrobial activity against both Gram-negative and Gram-positive bacteria. Antimicrobial activity of COS and polyphenols was linked to bacterial cell wall disintegration. Furthermore, changes in microbial DNA, mRNA, and protein synthesis via diffused COS and polyphenols could bring about the cell death [20,21]. Blueberry extract showed stronger antimicrobial effect on V. parahaemolyticus, which had no virulence genes than V. parahaemolyticus ATCC17802 (tdh − /trh + ) and ATCC 33847 (tdh + /trh − ), which had virulence genes. Increased virulence was associated with augmented antibiotic resistance [41]. COS-tea polyphenol conjugates as a bactericidal/bacteriostatic substance might have stronger antibacterial activity against a virulent strain. Antimicrobial activity of COS polyphenol conjugates against V. parahaemolyticus PSU.SCB.16S, MIC and MBC were 32 µg/mL and 64 µg/mL for the COS-CAT, respectively; and MIC and MBC of 64 µg/mL and 128 µg/mL were recorded for the COS-EGCG, respectively [21]. Gram-negative bacteria generally have hydrophilic thin outer membrane, comprising lipopolysaccharides. Therefore, they are susceptible to cellular lysis via COS and its polyphenol conjugates [58]. COS-tea polyphenol, especially COS-CAT, was a promising antimicrobial agent toward both spoilage and pathogenic bacteria. Sun et al. [59] reported that tolC gene expression was downregulated in V. parahaemolyticus F13. Although MIC and MBC of COS-CAT were higher than those of some antibiotics, it had the efficacy in inhibiting both pathogenic and non-pathogenic V. parahaemolyticus. Overall, nonpathogenic V. parahaemolyticus generally had more sensitivity to antibiotics than pathogenic V. parahaemolyticus. However, pathogenic and non-pathogenic V. parahaemolyticus were similarly susceptible to COS-CAT in the present study.

Antibiotic Susceptibility Profile of Different V. parahaemolyticus Isolates
Eight V. parahaemolyticus isolates represented V. parahaemolyticus PSU.SCB.16S.14 (Laboratory strain, VP), Asian green mussel from farm (M1 isolate), Asian green mussel from natural habitat (M42 isolate), Asian green mussel from local markets (M77, M91, M92, and M106 isolates), and clinical sample of stool specimens of diarrhea patients (HVP1 isolate) were used for testing. MBCs of COS-tea polyphenol conjugates against all V. parahaemolyticus (8 strains) were 1.024 mg/mL as shown in Table 3. Varying antibiotic susceptibility profiles with 21 antibiotics toward those eight isolates were noticeable (Table 4). Seven antibiotics namely ampicillin, cefoxitin, ceftriaxone, colistin, doripenem, ertapenem, and netilmicin did not show susceptible, intermediate, and resistant results because CLSI breakpoints of these antibiotics did not exist for V. parahaemolyticus [60]. Isolates tested were highly susceptible to antibiotics such as Amikacin (100%), ciprofloxacin (100%), gentamicin (100%), imipenem (100%), and levofloxacin (100%). However, HPV1 clinical isolate showed high resistance to amoxicillin/clavulanic acid, ampicillin/sulbactam, cefepime, cefotaxime, ceftazidime, cefuroxime, meropenem, piperacillin/tazobactam, and trimethoprim/sulfamethoxazole with MAR index of 0.64. This isolate was resistance to 9 antibiotics of 14 antibiotics tested. Most of the six Asian green mussel V. parahaemolyticus isolates in this study exhibited high sensitivity to all antibiotics, but M42 isolate exhibited intermediate resistance to cefuroxime. Elexson et al. [57] found that all V. parahaemolyticus isolates from cultured seafood products were resistant to penicillin and ampicillin. However, it has been discovered that the Asian green mussel cultivated in Thailand is frequently an open system culture in coasts and estuaries, in which antibiotics are not required. As a result, no drug resistant V. parahaemolyticus isolated from natural Asian green mussel farms and Asian green mussels sold in the local market was found in this study. Another health risk may arise with cross-contamination by shellfish to other seafoods in the market. To address the potential consequences of pathogenic V. parahaemolyticus in seafood, continuous monitoring of environmental and seafood samples, including mussel as well as tracking the source of clinical and environmental strains are still needed.

Conclusions
Fifty collected isolates including Asian green mussel samples, clinical and laboratory strains were identified as V. parahaemolyticus based on their morphological, biochemical, and molecular characteristics. They were all biofilm producers with strong motile ability. Only six isolates (12%) from the clinical sample were positive for the virulence tdh gene (tdh + trh − ) and had a positive result for the KP test. COS-CAT demonstrated the greatest bactericidal action against V. parahaemolyticus isolated from Asian green mussels and clinical samples with an MBC value of 1.024 mg/mL. In addition, V. parahaemolyticus isolated from Asian green mussel farms, natural habitat, and local markets showed no antibiotic resistance. Only the sample clinical isolates had a MAR value of 0.64 and were extremely resistant to nine kinds of antibiotics. Hence, to address the potential consequences of pathogenic V. parahaemolyticus in seafood, constant monitoring of environmental and seafood samples is still essential for food safety assurance.