Characterization and Analysis of Clustered Regularly Interspaced Short Palindromic Repeats (CRISPRs) in Pandemic and Non-Pandemic Vibrio parahaemolyticus Isolates from Seafood Sources

Vibrio parahaemolyticus is one of the significant seafood-borne pathogens causing gastroenteritis in humans. Clustered regularly interspaced short palindromic repeats (CRISPR) are commonly detected in the genomes of V. parahaemolyticus and the polymorphism of CRISPR patterns has been applied as a genetic marker for tracking its evolution. In this work, a total of 15 pandemic and 36 non-pandemic V. parahaemolyticus isolates obtained from seafood between 2000 and 2012 were characterized based on hemolytic activity, antimicrobial susceptibility, and CRISPR elements. The results showed that 15/17 of the V. parahaemolyticus seafood isolates carrying the thermostable direct hemolysin gene (tdh+) were Kanagawa phenomenon (KP) positive. The Multiple Antibiotic Resistance (MAR) index ranged between 0.1 and 0.4, and 45% of the isolates have an MAR index ≥ 0.2. A total of 19 isolates were positive for CRISPR detection, including all tdh+ trh− isolates, two of tdh− trh+, and each of tdh+ trh+ and tdh− trh−. Four spacer types (Sp1 to Sp4) were identified, and CRISPR-positive isolates had at least one type of spacer homolog to the region of Vibrio alginolyticus megaplasmid. It is of interest that a specific CRISPR profile and spacer sequence type was observed with correlations to the hemolysin genotype (tdh/trh). Thus, these provide essential data on the exposure of foreign genetic elements and indicate shared ancestry within different genotypes of V. parahaemolyticus isolates.


Introduction
Vibrio parahaemolyticus is a Gram-negative halophilic bacterium which belongs to the family Vibrionaceae. It is an oxidase-positive, facultative anaerobic bacterium, similar to other members in the genus Vibrio, present in marine or estuarine environments [1]. Many V. parahaemolyticus strains are pathogenic and can cause gastroenteritis in humans due to consumption of raw or undercooked seafood [2]. It was discovered as a common cause of foodborne diseases in Japan in 1950 and is responsible for the world's worst seafood-associated diarrhoea after new pandemic strains emerged in 1996 [3]. The virulence genes, tdh and trh, encoded for the thermostable direct hemolysin (TDH) and TDH-related hemolysin (TRH), respectively, are considered as virulence factors associated with V. parahaemolyticus hemolysis and cytotoxicity activity in the host, and have been used as pandemic group-specific markers together with group-specific toxRS PCR (GS-PCR) to

Detection of Hemolysin Genes
A single colony of V. parahaemolyticus was inoculated into tryptic soy broth (TSB) with 1% NaCl and incubated at 30 • C with shaking overnight. Genomic DNA was extracted using a boiling method, as described previously [19]. Briefly, the 1 mL of broth culture was centrifuged at 10,000× g for 5 min, washed twice, resuspended in 1 mL of sterile distilled water, and boiled for 10 min. The boiled culture was centrifuged at 20,000× g at 4 • C, 10-fold diluted, and used as templates to detect the tdh and trh genes. The virulence genes detection of V. parahaemolyticus performed by PCR targeted to the tdh gene using a forward primer (5 -GGTACTAAATGGCTGACATC-3 ) and a reverse primer (5 -CCACTACCACTCTCATATGC -3 ) to detect a 251 bp gene fragment and the trh gene using a forward primer (5 -GGCTCAAAATGGTTAAGCG-3 ) and a reverse primer (5 -CATTTCCGCTCTCATATGC-3 ) to detect a 250 bp gene fragment [20]. The PCR reaction was carried out with a reaction mixture consisting of 1.5 mM MgCl 2 , 0.2 mM dNTPs, 0.4 µM of each primer, 0.025 U of GoTaq DNA polymerase (Promega, Madison, WI, USA), and 2.0 µL DNA templates in a 20 µL volume. The PCR reactions were performed with a T100 TM Thermal Cycler (Bio-Rad, Hercules, CA, USA). The PCR process included initial denaturation at 96 • C for 5 min, followed by 35 cycles of denaturation at 94 • C for 1 min, annealing at 55 • C for 1 min, extension at 72 • C for 1 min, and a final extension at 72 • C for 7 min. Electrophoresis was performed on a 1% agarose gel and the PCR products were detected using a UV transilluminator.

Group-Specific PCR (GS-PCR)
To investigate pandemic isolates of V. parahaemolyticus, GS-PCR was carried out using a forward primer (5 -TAATGAGGTAGAAACA-3 ) and a reverse primer (5 -ACGTAACGGG CCTACA-3 ) to detect a 651 bp amplicon of the toxRS sequence of the new O3:K6 clone [3]. In each reaction of 20 µL, 0.2 µM of each primer, 0.125 mM dNTPs, 1.5 mM MgCl 2 , 4 µL of 5× reaction buffer, 0.025 U of GoTaq DNA polymerase (Promega, Madison, WI, USA), and 2.5 µL DNA templates were used. The amplification conditions consisted of an initial denaturation at 96 • C for 5 min, followed by 25 cycles of denaturation at 96 • C for 1 min, annealing at 45 • C for 2 min, extension at 72 • C for 3 min, and a final extension at 72 • C for 7 min. Electrophoresis was performed on a 1% agarose gel and the PCR products were detected using a UV transilluminator.

Determination of Hemolytic Activity
For a quantitative analysis of hemolytic activity, V. parahaemolyticus isolates were cultured in TSB with 1% NaCl and incubated at 30 • C with shaking overnight. After incubation, bacterial culture was transferred into a microcentrifuge tube and washed twice with phosphate buffer saline solution (PBS). Then, bacterial suspension was adjusted to 0.5 McFarland standard. A total of 190 µL of each strain were mixed with 10 µL of packed human red blood cells (HRBCs). Packed HRBCs were obtained after centrifugation at 4000 rpm for 10 min at 4 • C. After centrifugation, RBCs were washed 3 times with PBS and incubated at 37 • C for 5 h. Distilled water and PBS were added to prepared HRBCs as the positive and negative control, respectively. After incubation, the supernatant was obtained by centrifuge at 10,000 rpm for 10 min at 4 • C and the amount of hemoglobin released from the lysed HRBCs was measured at 540 nm [21]. The tests were performed in duplicate. The percentage of hemolysis was calculated using the following formula: % Hemolysis = OD540 of sample − OD540 of negative control OD540 of positive control − OD540 of negative control × 100

Kanagawa Phenomenon (KP) Assay
For the Kanagawa phenomenon (KP) assay, bacteria were grown on Wagatsuma medium (BAM Media M178) containing 5% HRBCs. Plates were incubated at 37 • C for 35 h, and hemolytic zones around the colonies were observed. Isolates producing a clear hemolytic zone around the colonies were identified as Kanagawa phenomenon-positive, those with no hemolytic zone were identified as Kanagawa phenomenon-negative [22].

Antimicrobial Susceptibility Test
All isolates of V. parahaemolyticus were tested for susceptibility using a standard disk diffusion assay. Antibiotics will be used in this test according to Clinical and Laboratory Standards Institute (CLSI) guidelines [23], including ampicillin (10 µg), gentamicin (10 µg), erythromycin (15 µg), sulfonamide (300 µg), tetracycline (30 µg), ciprofloxacin (5 µg), sulfamethoxazole (25 µg), chloramphenicol (30 µg), norfloxacin (10 µg). The distribution of inhibition zone diameters was interpreted based on the Clinical and Laboratory Standards Institute (CLSI) guidelines. The Multiple Antibiotic Resistance (MAR) index was calculated using the formula MAR = a/b, where "a" is the number of antibiotics to which the test isolate was resistant, and "b" is the total number of antibiotics tested. A value greater than 0.2 indicates that the isolates were isolated from high-risk sources [24].

Determination of CRISPR Sequences
The presence of CRISPR element in V. parahaemolyticus was detected using the PCR method described previously [15]. Briefly, PCR was carried out using PCR mixture containing 1.5 mM MgCl 2 , 0.25 mM dNTPs, 0.25 µM of forward primer (5 -ATGCATTCCAAAGCT ACCACTC-3 ) and reverse primer (5 -GCCTACCAGATAGCAAGTGTCC-3 ), 0.025 U of Takara Ex Taq DNA polymerase (Takara Biochemicals, Tokyo, Japan), and 2.0 µL of DNA templates in a 20 µL volume. The PCR process included initial denaturation at 94 • C for 1 min, followed by 30 cycles of denaturation at 94 • C for 1 min, annealing at 50 • C for 1 min, extension at 72 • C for 1 min, and a final extension at 72 • C for 5 min. The PCR products were purified using GenepHlow TM Gel/PCR Kit (Geneaid, New Taipei, Taiwan) and sequenced (Macrogen, Seoul, Korea).

Analysis of CRISPR and CRISPR-Virulence Typing
CRISPR patterns, including the direct repeats (DRs) and spacers, were analyzed using the CRISPRfinder server available at https://crispr.i2bc.paris-saclay.fr/Server/ (accessed on 14 April 2020). The RNA secondary structure of repeats was predicted using RNA fold [25].
The homology analysis of spacer sequences was identified by submitting the spacer sequence to BLASTN (somewhat similar sequences). Additionally, CGView was used to visualize the positions of protospacer on foreign genetic elements [26]. ProgressiveMauve was used for multiple genome alignments to visualize conserved genomic regions [27].
CRISPR-positive isolates of V. parahaemolyticus were analyzed based on CRISPR sequences combined with virulence genes, tdh and trh. The dendrogram was constructed using BioNumerics 7.6 software (Applied Maths, Saint-Martens-Latem, Belgium) with the UPGMA algorithm, using the Dice similarity coefficient [15].

Statistical Analyses
The correlation analyses were analyzed by Pearson's correlation test, using Statistical Package for Social Sciences (SPSS v. 17, IBM, Armonk, NY, USA). The correlation coeffi-cient (r-value) was used to interpret the correlation and p-value ≤ 0.05 was considered statistically significant in comparative data.

Detection of Virulence Genes and Hemolytic Activity
In this study, 51 isolates of V. parahaemolyticus isolated from seafood sources were grouped based on the presence of tdh, trh, and the pandemic marker genes detected by GS-PCR ( Table 1). The percentages of hemolysis observed among V. parahaemolyticus isolates were varied and the correlation between hemolytic activity and the presence of the tdh gene was not found (r = 0.036, p = 0.802). The correlation between hemolytic activity and the presence of the trh gene was not done due to the small sample size. However, the greatest hemolytic activity (98%) was found in one of the tdh+ trh+ isolates ( Figure 1). Furthermore, among 20 tdh+ isolates, 18 isolates exhibited β-hemolytic activity on Wagatsuma blood agar.
quences combined with virulence genes, tdh and trh. The dendrogram was constructed using BioNumerics 7.6 software (Applied Maths, Saint-Martens-Latem, Belgium) with the UPGMA algorithm, using the Dice similarity coefficient [15].

Statistical Analyses
The correlation analyses were analyzed by Pearson's correlation test, using Statistical Package for Social Sciences (SPSS v. 17, IBM, Armonk, NY, USA). The correlation coefficient (r-value) was used to interpret the correlation and p-value ≤ 0.05 was considered statistically significant in comparative data.

Detection of Virulence Genes and Hemolytic Activity
In this study, 51 isolates of V. parahaemolyticus isolated from seafood sources were grouped based on the presence of tdh, trh, and the pandemic marker genes detected by GS-PCR (Table 1). The percentages of hemolysis observed among V. parahaemolyticus isolates were varied and the correlation between hemolytic activity and the presence of the tdh gene was not found (r = 0.036, p = 0.802). The correlation between hemolytic activity and the presence of the trh gene was not done due to the small sample size. However, the greatest hemolytic activity (98%) was found in one of the tdh+ trh+ isolates ( Figure 1). Furthermore, among 20 tdh+ isolates, 18 isolates exhibited β-hemolytic activity on Wagatsuma blood agar.

Characterization of Antimicrobial Susceptibility
All V. parahaemolyticus isolates from seafood were susceptible to chloramphenicol, whereas the highest resistance was recorded in response to ampicillin (100%) and followed by erythromycin (43.1%). It is of interest that 74.5% and 31.3% of the isolates demonstrated intermediate resistance to ciprofloxacin and norfloxacin, respectively ( Table 2). An MAR index of 0.2 or above was shown by 23 isolates (45%) ( Table 1). No significant correlation between the MAR index and the presence of the tdh gene (r = 0.096; p = 0.053), trh gene (r = 0.023; p = 0.871), or hemolytic activity (r = 0.112; p = 0.436) was observed.

Analysis of CRISPR Repeat Sequences and Patterns
A total of 19/51 (37.3%) of V. parahaemolyticus seafood isolates were positive for CRISPR (Table 1). Overall, the CRISPR element was significantly more prevalent in the tdh+ isolates and occurred less in the tdh− isolates (r = 0.832; p < 0.001). Two or three 28-bp direct repeats (DRs) were found and interspaced by spacers of 31 and/or 32 bp. (Table 3). Although the CRISPR repeat sequence is highly conserved, the polymorphisms within the terminal repeats were observed in this study ( Table 3). The terminal repeats differed in a single nucleotide in the central region and 7 bases in the terminal region of predicted RNA secondary structures. However, these do not affect the based pairing of RNA structure ( Figure 2).

Spacer Sequence Analysis
The spacer sequences analysis revealed 4 different CRISPR spacer types which were designated as Sp1 to Sp4 (Table 4). Sp1, Sp2, and Sp3 shared the best matched with 100, 96.7, and 87.1% homology to fragments of V. alginolyticus plasmid pL300 (Table 4). For

Spacer Sequence Analysis
The spacer sequences analysis revealed 4 different CRISPR spacer types which were designated as Sp1 to Sp4 (Table 4). Sp1, Sp2, and Sp3 shared the best matched with 100, 96.7, and 87.1% homology to fragments of V. alginolyticus plasmid pL300 (Table 4). For Sp4, no homology to foreign genetic elements was detected. It is of interest that the spacer type was found to be correlated with tdh/trh profile and at least one type of spacer homolog to the V. alginolyticus plasmid (Sp1 to Sp3) was shared by all CRISPR-positive V. parahaemolyticus isolates (Table 3). Further homology analysis of the spacer sequences against publicly available V. parahaemolyticus genomes showed the association of spacer type and tdh/trh profile, although virulence genes of some isolates were unknown (Table 5). These hits are located on a CRISPR array, as they were detected nearby cas gene(s), and some hits have also been addressed as CRISPR region.

Spacer Origin
To further examine the finding of extrachromosomal origin of CRISPR spacers, the homology found among spacers was compared. This study found that Sp1, Sp2, and Sp3 sequences were matched with 6 mega-plasmid sequences of V. alginolyticus isolates, previously obtained from pipefish caught in the Kiel-Fjord, Germany, which ranged between 280 and 300 kbp in size [28]. Additionally, CRISPR spacers were mapped to represent Kiel alginolyticus megaplasmid, pL300, to locate the position of spacers. This study found that Sp1 to Sp3 spacers were distributed around pL300 ( Figure 3). Sp1, Sp2, and Sp3 spacers were matched with the protospacer sequences of genes encoding for the modulator of FtsH protease Hfl, hypothetical protein, and DNA methyltransferase, respectively.
The alignment between pL300 sequences and other plasmids was performed to identify plasmid identity. Apart from V. alginolyticus isolates from Kiel Fjord, this megaplasmid was highly homolog (98.6% identity; E-value of 0.0) to the MYPK1 plasmid of V. alginolyticus strain GS (accession number CP054703.1). Few similarities to other plasmid sequences of Enterobacteria, including Escherichia coli, Klebsiella pneumoniae, Kluyvera ascorbata, and Citrobacter sp. were observed (data not shown). The progressiveMauve alignment results showed the conservation regions among related ancestor plasmids ( Figure S1). The conserved CDSs found among these plasmids included the essential genes required for plasmid conjugal transfer and integration (Table S1). A closer look at the protospacer regions among the four plasmids is shown in Figure 4. sequences were matched with 6 mega-plasmid sequences of V. alginolyticus isolates, previously obtained from pipefish caught in the Kiel-Fjord, Germany, which ranged between 280 and 300 kbp in size [28]. Additionally, CRISPR spacers were mapped to represent Kiel alginolyticus megaplasmid, pL300, to locate the position of spacers. This study found that Sp1 to Sp3 spacers were distributed around pL300 (Figure 3). Sp1, Sp2, and Sp3 spacers were matched with the protospacer sequences of genes encoding for the modulator of FtsH protease Hfl, hypothetical protein, and DNA methyltransferase, respectively. The alignment between pL300 sequences and other plasmids was performed to identify plasmid identity. Apart from V. alginolyticus isolates from Kiel Fjord, this megaplasmid was highly homolog (98.6% identity; E-value of 0.0) to the MYPK1 plasmid of V. alginolyticus strain GS (accession number CP054703.1). Few similarities to other plasmid sequences of Enterobacteria, including Escherichia coli, Klebsiella pneumoniae, Kluyvera ascorbata, and Citrobacter sp. were observed (data not shown). The progressiveMauve alignment results showed the conservation regions among related ancestor plasmids ( Figure  S1). The conserved CDSs found among these plasmids included the essential genes required for plasmid conjugal transfer and integration (Table S1). A closer look at the protospacer regions among the four plasmids is shown in Figure 4.

CRISPR-Virulence Typing
In this study, a combination between CRISPR spacer sequence types and tdh/trh virulence genes of V. parahaemolyticus was used to create a dendrogram. CRISPR-virulence typing profiles of V. parahaemolyticus were organized into 4 CRISPR-virulence types (CVTs) ( Figure 5). The isolates within the same CVT cluster composed identical virulence profiles and spacer content.
The pandemic isolates with tdh+ trh− genotype was classified in the CVT1 cluster, which contained spacer type 1 and 4, and this was identical to that observed in the reference strain of V. parahaemolyticus RIMD 2210633. The tdh− trh+ and tdh− trh− isolates were classified in the CTV2 and CTV3 clusters, respectively. The differences between CVT1, CVT2, and CVT3 clusters were due to deletion or addition of a single spacer. It is of interest that the tdh+ trh+ isolate in the CVT4 cluster did not have any close association with others. Nevertheless, all 4 clusters shared spacers matched with pL300 of V. alginolyticus, suggesting that these isolates encounter this plasmid before they are diverged.

CRISPR-Virulence Typing
In this study, a combination between CRISPR spacer sequence types and tdh/trh virulence genes of V. parahaemolyticus was used to create a dendrogram. CRISPR-virulence typing profiles of V. parahaemolyticus were organized into 4 CRISPR-virulence types (CVTs) ( Figure 5). The isolates within the same CVT cluster composed identical virulence profiles and spacer content.

CRISPR-Virulence Typing
In this study, a combination between CRISPR spacer sequence types and tdh/trh virulence genes of V. parahaemolyticus was used to create a dendrogram. CRISPR-virulence typing profiles of V. parahaemolyticus were organized into 4 CRISPR-virulence types (CVTs) ( Figure 5). The isolates within the same CVT cluster composed identical virulence profiles and spacer content.

Discussion
In this study, 51 seafood isolates of V. parahaemolyticus previously, obtained from various cities in Southern Thailand between 2000 and 2012, were characterized for their hemolytic characteristics, antibiotic resistance, and CRISPR. All experiments included four hemolytic genotypes, pandemic and non-pandemic, of V. parahaemolyticus isolates.
Although environmental and seafood isolates rarely contained tdh/trh, the presence of these isolates has been previously reported [29].
TDH-or TRH-producing V. parahaemolyticus are strongly associated with gastroenteritis. In this study, the result of a quantitative hemolysis assay against human blood demonstrated various levels of hemolytic activity exhibited by seafood isolates, and the highest hemolytic activity was detected in a tdh+ trh+ isolate. A previous study reported that temperature and culture density were the significant factors that affect the hemolysin activation of V. parahaemolyticus [30]. The gastroenteritis also showed a strong correlation with the KP, which is a type of β-hemolysis on the special blood agar and is induced by TDH [31]. A previous study reported that almost all environmental V. parahaemolyticus are tdh-negative and thus negative for KP [32]. In this work, two of the KP-negative V. parahaemolyticus isolates were positive for tdh gene. The previous study suggested that the isolates may possess a single copy of the tdh gene and therefore hemolysin was expressed at a low level [33]. Moreover, the hemolytic activity of V. parahaemolyticus may be phenotypically varied due to the sequence variation of hemolysin genes, which affects their promoter activities [6,34].
The increase in antibiotic-resistant bacteria is a major concern worldwide. Aquatic ecosystems could easily facilitate the exchange of mobile genetic elements among V. parahaemolyticus [35]. All V. parahaemolyticus isolates were resistant to ampicillin which is not recommended empirically to treat V. parahaemolyticus infection [36]. The ampicillin resistance found in this study correlates with other previous reports and this is not a new incident for ampicillin resistance among Vibrio spp. [37]. All V. parahaemolyticus isolates in this study were susceptible to chloramphenicol, which was the first truly broadspectrum, and a high susceptibility to tetracycline (96.3%), sulfonamide (96.3%), and trimethoprim/sulfamethoxazole (67.2%) was observed. Interestingly, 72.7% and 34.5% of the isolates showed intermediate resistance to ciprofloxacin and norfloxacin, respectively which are the antimicrobial agents recommended to treat human infections [38]. This result was consistent with other studies [39]. The high level of intermediate resistance to both antibiotics may also tend to become resistant. Additionally, four isolates (PSU635, PSU3858, PSU4446) were identified as multidrug resistant (MDR) as they resist at least one agent in three or more antimicrobial classes [40]. This could lead to untreatable illnesses and human health risks. The MAR index value observed in this study indicated the differences in the original sources of the isolates. In this study, nearly half of the V. parahaemolyticus isolates had an MAR index ≥ 0.2, which indicated that these isolates originated from high-risk sources that have been exposed to antibiotics, such as humans and aquaculture, and seafood might be considered as a possible vehicle for pathogen transmission. Previous studies have reported that the V. parahaemolyticus isolates from seafood showed an MAR index > 0.2 [41,42]. Thus, antimicrobial resistance should continue to be monitored to ensure seafood safety. CRISPR-Cas system is known as a bacterial adaptive immune system against conjugative plasmids, or phages. The virulence and genotypes of V. parahaemolyticus are reported to be associated with CRISPR/Cas evolution [17]. This study found that all pandemic V. parahaemolyticus isolates were identified as CRISPR positive, suggesting the association between CRISPR and the pathogenic evolution of pathogenic V. parahaemolyticus. Previous reports demonstrated high percentages of tdh+ (97.4%) and trh+ (47%) V. parahaemolyticus positive for CRISPR detection, with few reports demonstrating an association with the tdh− or trh− genotypes [15,17]. The result of this study corroborates earlier findings; however, we could detect CRISPR in two of the tdh− trh + and one of the tdh− trh− isolates. The presence of CRISPR in the tdh− isolates were also reported in a previous study [41]. The CRISPR loci detected in this study is located on the V. parahaemolyticus island-7 (VPAI-7) region of chromosome 2, which encodes the tdh gene and are closely associated with tdh+ V. parahaemolyticus [43][44][45], thus this may be the reason why this CRISPR loci is more prevalent in the tdh+ isolates. Additionally, no significant correlation between the presence of CRISPR and hemolytic activity, or antimicrobial resistance, was observed (p > 0.05).
In this study, the DRs detected in CRISPR loci were 28-bp in length and its sequences have also been reported [16]. These characteristics were identical to those found in the reference strain of V. parahaemolyticus RIMD 2210633. In 2019, Baliga et al. analyzed the DRs of 200 V. parahaemolyticus genomes available in the database and found that 92% possessed a single type of repeat (RU-1) and the minimal set of Cas proteins [16]. The presence of similar DRs was observed in our recent study. However, the CRISPR primers used in this study did not target the Cas genes. Thus, the minimalistic type of CRISPR-Cas system could not be addressed here. Although a previous study reported 3-5 repeat units in the CRISPR loci of V. parahaemolyticus, we demonstrated 2 repeat units in the tdh+ trh+ and the tdh− trh− isolates. The presence of 2 DRs have been reported among tdh+ trh+ isolates from clinical samples [15,17]. Most isolates available in the public databases possess tdh+ trh− genotype, which may acquire more spacers than others. Moreover, further studies are necessary to clarify the types of CRISPR-Cas systems and their activities in the defence system against foreign mobile genetic elements.
This study found that the sequences of CRISPR repeats found among V. parahaemolyticus isolates were not highly identical, especially at the terminal position, however, this does not affect the RNA secondary structures. The polymorphisms in terminal sequences among the isolates in different genotypes were also observed. The variations within the central and terminal repeat sequence were also observed in the CRISPR loci of Streptococcus thermophilus, which suggested that the sequence degeneracy at the 3 -end of the terminal repeat is probably due to the homologous recombination events of DRs [46]. A degenerate repeat is required for the polarized spacer acquisition process [47].
BLAST comparisons of spacers identified matches between spacers and foreign plasmid sequences, which demonstrated the horizontal transfer of V. alginolyticus plasmid to V. parahaemolyticus. Most of the spacer sequences found in this study were similar to those reported in other V. parahaemolyticus isolates from seafood sources [16]. We have previously discussed the similarity of CRISPR spacers found in clinical isolates of V. parahaemolyticus with the V. alginolyticus plasmids [15]. In 2020, all of these plasmids were published as the megaplasmids associated with the Kiel isolates of V. alginolyticus [28]. In continuation of our previous discussion, the spacer sequence and origin were characterized and analyzed in a recent study.
In this study, 3 CRISPR spacers (Sp1 to Sp3) match the coding sequences of V. alginolyticus megaplasmid pL300, and one of these spacers (Sp4) was unique, as it demonstrated no homology to any sequences in the databases. It is often difficult to find sequence homology to spacer sequences because of the limited number of sequence information databases of plasmids and prokaryotic viruses [48]. All V. parahaemolyticus isolates included at least one spacer perfectly matching some region of pL300. One of the homology searches revealed the perfect match between CRISPR spacer and plasmid region, encoded for DNA methyltransferase, an enzyme required to protect DNA from restriction digestion [49]. This gene has been previously reported to be specifically targeted by CRISPR array against the virus to prevent successful host infection and lysis [50]. It is therefore conceivable that V. parahaemolyticus has adapted to its own evolution by avoiding the integration of foreign DNA from V. alginolyticus. Moreover, the results obtained by Progressive Mauve analysis demonstrated that pL300 may originate via the insertion of multiple regions of foreign DNA from bacteria in family Enterobacteriaceae. These conserved regions were found to encode essential genes of conjugative plasmids [49]. In addition, the evolution of conjugative plasmids could be driven through the CRISPR preventive mechanism.
CRISPR spacers can protect against foreign plasmid integration and phage invasions. The influence of these mobile genetic elements on the phenotypes and genotypes of marine Vibrio have been reported. Previous studies demonstrated a strong association of filamentous phages, f237, with recent pandemic strains of V. parahaemolyticus. This phage possesses a unique ORF8 open reading frame that is found only in recent pandemic isolates, and the Zot (Zonula occludens toxin)-like toxin previously described in the V. cholerae is reported to be encoded in the ORF7 [51,52]. A match was found between the CRISPR spacer region of V. anguillarum PF7 with a zot-encoding prophage in V. anguillarum PF4, indicating the contribution of CRISPR to the phage resistance of this bacterium [53]. In this study, no homologous between the CRISPR spacer and any phage regions or endogenous genomic sequences located outside the CRISPR array were found.
The results of this study indicated the association between spacer groups and hemolysin genotypes. Thus, a combination between CRISPR spacer sequences and the presence of hemolysin genes was used to cluster the V. pahaemolyticus isolates. As expected, a close relationship between the tdh+ trh− and the tdh− trh+ isolates were found in this study. The tdh− trh+ isolates were reported to be involved significantly in the pre-pandemic outbreaks, whereas the tdh+ trh− isolates are involved in both pre-pandemic and pandemics [8]. The tdh− trh− isolate was presented as the possible origin of the tdh+ trh− V. parahaemolyticus and the tdh+ trh+ isolates were phylogenetically distant to others. Since fewer spacers are less efficient in the protection against foreign genetic elements, further evolutional study is needed to explain this situation. In addition, the position of a spacer in a CRISPR loci indicates the historical timeline of V. parahaemolyticus exposure to foreign genetic elements [54].

Conclusions
An analysis of hemolysin genotypes in association with CRISPR loci can help to reveal the genome evolution of V. parahaemolyticus. Our characterization and in-depth analysis of CRISPR among seafood isolates of V. parahaemolyticus showed that these isolates exhibit various virulence phenotypes, and some isolates have become resistant to many antibiotics. More importantly, similar types of CRISPR repeats and spacer sequences were found in isolates harboring the same hemolysin genotypes (tdh and/or trh) and all CRISPRpositive isolates shared spacer(s) homolog to the similar plasmid of V. alginolyticus. In addition, the position of a spacer in a CRISPR loci could indicate an ancestry interaction of V. parahaemolyticus with a foreign genetic element. An in-depth analysis of CRISPR-Cas loci, combined with virulence genes, could provide important data for the evolutionary study of V. parahaemolyticus.

Supplementary Materials:
The following are available online at https://www.mdpi.com/article/ 10.3390/microorganisms9061220/s1, Figure S1: The multiple genome alignment by the progres-siveMauve between V. alginolyticus pL300 megaplasmid (top) and other plasmids from the blast hit results. Homologous genomes are indicated by the connected-with-lines collinear blocks, Table S1: Details regarding the conserved CDSs residing in the locally collinear blocks (LCBs) obtained from ProgressiveMauve analysis.

Conflicts of Interest:
The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.