3.1 Characterization of bacteria
3.1.1 Morphological structure of bacteria
Strain BTXS2 presented as typical large green single colony with smooth surface and flat edges on Vibrio chromogenic medium (Figure 1A); The morphology of the bacteria was observed using an optical microscope after Gram staining, and the results showed BTXS2 was a Gram-negative bacterium with a short rod-like shape (Figure 1B); After observing under SEM, the BTXS2 was short rod-shaped (0.5-0.8×1.4-2.0 μm) with a terminal flagellum (Figure 1C and D). The shape of the bacteria was similar to that of V.parahaemolyticus reported by references [27, 31], but no flagellum was observed in their results. It may be due to the glutaraldehyde fixation and centrifugation process that damaged the flagella and caused it to fall off. According to the morphological observation, it was preliminarily identified as V.parahaemolyticus.
3.1.2 Physiology, Biochemistry and Antibiotic resistance of BTSX2
The carbon source utilization and enzymatic results of the bacteria were compared with Bergey's Manual [32], and the result showed that BTSX2 was indeed a strain of V.parahaemolyticus. It can use 44 of 71 types of carbon source including eleven sugars in the BIOLOG plate (Table S2). BTSX2 has a great advantage in competition with other bacterial species due to its ability to use a broad range of carbon sources. The strain BTXS2 can utilize lactose and inositol but cannot utilize L-aspartic acid, which is not consistent with the result reported by Heinemeyer [33]. It indicates the specificity of the BTXS2 in carbon source utilization, which may affect and determine the survival and distribution of the BTXS2. The enzymatic reaction of BTXS2 showed positive results for alkaline phosphatase, esterase (C4), lipid esterase (C8), lipase (C14), and naphthol-AS-BI-phosphohydrolase, indicating that it has a high metabolic diversity (Table S3). The excellent physiological, biochemical and metabolic capabilities of BTXS2 provide guarantee for its widespread in estuaries and oceans.
The antibiotic sensitivity tests showed that the BTXS2 was resistant to penicillin and vancomycin, but sensitive to other antibiotics tested in the experiment (Table 1), indicating that the V.parahaemolyticus BTXS2 isolated from the environment is resistant to multiple antibiotics. It should be noted that the BTXS2 has acquired resistance to the last-line antibiotic vancomycin, which is rarely reported [34]. And it may increase the cost of treatment of antibiotic-resistant V.parahaemolyticus based on the fact that vancomycin is often used to clinical treatment of antibiotic-resistant bacteria.
3.1.3 The effect of pH and salinity on bacterial growth
As shown in Figure 2A, different initial pH values had no obvious effect on the growth of V.parahaemolyticus BTXS2 except for pH 5.0. In the exponential growth phase, the growth rates at pH 6.0 and 7.0 were almost the same, followed by that at pH 8.0 and pH 9.0. And in the stationary phase, the numbers of BTXS2 could reach an equal value at different pH. The BTXS2 entered the exponential growth phase later at pH 5.0 compared with that at other pH, but it reached the stable phase almost at the same time as pH 9.0. The growth properties of BTXS2 at different pH were inconsistent with the strains isolated by Yaashikaa, et al. [27] and Soto-Rodriguez, et al. [35] which grew best at pH 9.0 and pH 5.0 respectively. The BTXS2 isolated from Bohai Coast had good stable performance in the natural water environment, enhancing the possibility of its spread in different water bodies.
The growth of BTXS2 in 1.8%, 2.6% and 3% NaCl LB culture had no statistically significant difference (Figure 2B). At the end of exponential growth, the growth of bacteria in 1.8% NaCl LB was slightly faster than 2.6% and 3%, and reached the stable phase slightly earlier. The growth of V.parahaemolyticus in 0.5% NaCl LB culture was relatively slow, and began to grow rapidly after 230 min, which was consistent with the results reported by Soto-Rodriguez, et al. [35]. It shows that the BTXS2 could grow well at salinities of estuary, offshore and seas. The wide tolerance of the BTXS2 may enhance the survival and spread within an ecological niche [36].
3.1.4 Phylogeny analysis
After comparing the 16S rRNA sequence of BTXS2 with the NCBI nucleotide database, nine sequences belonging to V.parahaemolyticus were selected and compared with the 16S rRNA sequence to finally form phylogenetic tree (Figure 3). The results showed that the conservative sequence of BTXS2 was not significantly different from other V.parahaemolyticus. The strain of V.parahaemolyticus similar to BTXS2 widely distributed all over the world such as America, China, Japan and so on, and they are widely distributed in fishponds, shrimp ponds and sediments [37]. The 16S rRNA sequence of BTXS2 isolated from Tianjin estuary was the most homologous to ATCC 17802 that was a standard strain of BTXS2, and it was clustered under the same branch as FROC, indicating that they had a strong genetic relationship. Based on their pathogenicity, it shows that BTXS2 poses a security threat to humans all over the world.
3.1.5 Genome analysis of BTXS2
After sequencing and assembly optimization, the genome sequence was submitted to NCBI. The DNA size was 5068431 bp, which included two chromosomes and a plasmid. The G+C content was 45.32%, 45.65 and 67.99%, respectively. A total of 6182 genes were predicted to be the coding genes of the newly sequenced genome. Using RepeatMasker (Version 4.0.5) software to predict scattered repetitive sequences, a total of 384 scattered repetitive sequences and 138 tandem repetitive sequences were obtained. Sequencing analysis revealed that BTXS2 contained 131 tRNAs, 37 rRNAs and 15 sRNAs. The distribution of 11 gene islands is shown in the Figure S1. The islands contain virulence genes encoding putative type III secretion systems, type VI secretion systems and superoxide dismutase, as well as streptomycin resistance genes (Table S4). The annotation with the GO database revealed that BTXS2 contained functional genes related to virus structure and reproduction (Table S5), indicating that there were phage-related genes on the genome. Compared with the PHI database, it was found that the BTXS2 had a variety of pathogenic genes that threatened human health, such as causing human food poisoning, fever, diarrhea, gastritis, and other diseases. And there were also pathogenic genes that cause skin infections and sepsis in Danio rerio and Penaeus monodon (Table S6 and Figure S2 and S3). The BTXS2 may pose a great threat and negative impacts on human health and fish in the estuaries and coasts.
The antibiotic resistance genome were found in the gene sequence, and the result of PCR showed that the BTXS2 contained antibiotic resistance genes for β-lactamases (Figure S4), which commonly presented in various V.parahaemolyticus isolated from other studies[9, 16, 38]. A resistance gene related to the efflux pump mechanism was found in the PCR results of BTXS2, which might play an important role in antibiotic resistance. The antibiotic resistance gene for quinolones was also founded in BTXS2, but it is interesting that it did not show resistance in the antibiotic susceptibility test, which may be due to the fact that the gene is not expressed in the BTXS2. Gene alas, an alanyl-tRNA synthetase conferring resistance to novobiocin, was first founded in the V.parahaemolyticus genome, which was not reported in other V.parahaemolyticus, and it would bring greater challenges to the treatment of V.parahaemolyticus.
3.2 Characterization of phage
3.2.1 VB_VpP_BT-1011 morphology
Phage VB_VpP_BT-1011 could form a single-character plaque with a diameter of about 3 mm on double-layer plates (Figure 4A). After staining, the results of TEM showed that the phage was an isometric eicosahedral structure with a total length of 54 ± 1 nm, a tail length of 23 ± 0.5 nm, and a head length of 31 ± 0.5 nm (Figure 4B). VB_VpP_BT-1011 belonged to Mycoviridae family. Compared with the V.parahaemolyticus myophage found in other studies, VB_VpP_BT-1011 was smaller in size and had longer tail fibers[24, 39, 40]. The process of phage adsorption (Figure 4C) and cell membrane damage (Figure 4D) were observed.
3.2.2 Biological characteristics of VB_VpP_BT-1011
The latent time and burst size of the phage were determined by the one-step growth experiment. The one-step growth curve was divided into three stages, including the incubation period, the rising period and the stable period (Figure 5A). The one-step growth curve showed that the latent time of VB_VpP_BT-1011 was 30 min, and the rising period was 60 min. The burst size was 24 PFU/cell. Compared with the V.parahaemolyticus bacteriophage isolated by Matamp and Bhat [39] and Yang, et al. [41], VB_VpP_BT-1011 had a little longer latent time and a smaller burst size, but compared with four out of five phages isolated by Yu, et al. [42], they had a shorter latent time. The different latent time may be related to the specific recognition and assembly mechanism between host and phage. Moreover, mathematical modeling studies have demonstrated that both host quantity and quality are involved in determining the latent time [43, 44].
In the simulated water environment of different salinity, VB_VpP_BT-1011 could survive stably, and its survival rate after 24 h was higher than 84%. In addition, there was no significant difference in the survival rate under different salinity conditions (Figure 5B). VB_VpP_BT-1011 reached the highest survival rate (94.15%) at salinity of 1.8%. The survival rate under low-salinity conditions and high-salinity conditions were almost the same, indicating that the phage had good adaptability under the salinity conditions of natural water bodies.
The survival rate of VB_VpP_BT-1011 phage decreased after incubation in a simulated water environment in which pH values were ranging from 5.0 to 9.0 for 24 h, but there was no statistically significant difference in the survival rate under different pH (Figure 5C). After 24 h, the highest survival rate reached 83.45% at pH 7.0.
In the simulated freshwater environment with salinity of 0%, there was no significant difference in the effects of different temperatures on the survival rate of VB_VpP_BT-1011 (Figure 5D). And it reached the highest survival rate (94.54%) at 11oC. In the simulated seawater environment with salinity of 3%, there was no significant difference in the effect of the activity of VB_VpP_BT-1011 phage with the temperature ranging from 4oC to 37oC. After 24 h, it also reached the highest survival rate (96.28%) at 11oC. The conclusion can be drawn that the phage VB_VpP_BT-1011 could keep activity well throughout the year in freshwater and seawater. In the absence of the host, the bacteriophage VB_VpP_BT-1011 could exist stably in simulated freshwater and seawater. The results of the experiment that lasted for seven weeks showed that the survival rate of bacteriophages in simulated freshwater and seawater dropped by about 40% (Figure 5E). However, the survival rate of bacteriophages in the simulated seawater decreased more slowly than in the simulated freshwater environment, indicating that the bacteriophages have better adaptability to environments with higher NaCl concentrations.
Salinity, pH, and temperature are important factors affecting the survival and activity of bacteriophages [45]. In this study, VB_VpP_BT-1011 can survive stably under these environmental conditions, which shows its strong viability and the possibility of the phage's distribution and spread in various places. Thus, VB_VpP_BT-1011 may have the potential impact on the microbial community structure and ecological health in the natural environment since that bacteriophages occupy an important niche in the water environment [46].
3.2.3 Host Range
The results of spot incubation test showed that the phage could only infect BTXS2 (Table 2), indicating that phage VB_VpP_BT-1011 was highly specific to BTXS2, which was similar to phages in other studies [47, 48]. The specificity may be related to the phage-host interaction mechanism [49, 50]. In order to understand the host range of the phage better, more bacteria need to be tested. At the same time, it can be seen that the high specificity makes the use of this phage to treat V.parahaemolyticus more promising without affecting other microorganisms.
3.2.4 Phage genome analysis
The dsDNA genome size of the phage VB_VpP_BT-1011 was 43841 bp. The total G+C content was 42.77%. Three encoded tRNAs (starting aminoacyl-tRNA, arginine-carrying tRNA, and glycine tRNA) were found and a total of 72 coding genes were identified. A total of 75 CDSs were predicted using RAST, of which 15 were known phage functional genes, and the remaining 60 CDSs were predicted hypothetical proteins. A total of 268 ORFs were found using the ORF Finder online tool, and only 82 ORFs were found to have homology with the amino acid sequence of known proteins using the SMARTBLAST tool. The remaining 124 ORFs were speculative proteins and 62 ORFs with unknown functions, which might be due to the novelty of VB_VpP_BT-1011 and the lack of database information on environmental phages.
The sequence of the phage was compared with the antibiotic resistance gene database, and the results showed that the phage did not have any antibiotic resistance gene. The bacteriophage VB_VpP_BT-1011 had the same modular genome structure as most dsDNA bacteriophages. RAST database was used to annotate the sequencing results, we found the phage recombinant protein NinG, DNA replication helicase protein DnaC/DnaI, recombinant functional protein, single-stranded DNA binding protein, phage-related endonuclease protein, phage terminal large subunit, phage major capsid protein and minor capsid protein genes (Figure 6A). The genes encoding these proteins play an important role in the structure of bacteriophage and the function of infection, recombination, and replication.
By comparing the genome of VB_VpP_BT-1011 with the genome of NCBI, ten similar phage genomes were selected to establish phylogenetic tree. The results showed that this phage had no significant similarity with other Vibrio phages in the NCBI database,but had a certain genetic relationship with the published Vibrio phages (Figure 6B). The phylogenetic tree analysis results based on the comparison of the whole genome showed that the phage was most similar to the Vibrio_phage_V039C genome. VB_VpP_BT-1011 and Vibrio_phage_V039C were in the same branch, and the relative distance was closer. In the course of evolution, horizontal gene transfer may lead to the widespread exchange of phage shared genes [51]. By comparison, it was found that both VB_VpP_BT-1011 and Vibrio_phage_V039C possessed large terminal subunits, phage capsids, single-stranded DNA binding proteins and DNA helicase genes. But they are significantly different from each other. Vibrio_phage_V039C contained a tape measure protein gene, HigA family antitoxin and Rha family regulatory protein, while the phage VB_VpP_BT-1011 isolated in this study contained a phage base plate protein, phage recombinant protein and VgrG protein. The difference in the types of genes leads to the difference in structure and function of each other, which will further affect their living habits and interaction with their hosts.
3.3 Inhibition of VB_VpP_BT-1011 on BTXS2
The result showed that the phage had antibacterial effect on V.parahaemolyticus BTXS2. When the MOIs were 0.00001-0.001, the higher the MOI was, the more obvious antibacterial effects would have on the host bacteria, which showed that MOI was an important factor affecting the antibacterial of phages. When the MOI was 0.01, the concentration of host bacteria increased first, but with the expansion of phage, the concentration of host bacteria began to decrease after 150 min. When the host bacteria were suppressed to a low concentration, a new resistant subgroup was selected and proliferated, resulting in the increase of bacterial concentration in the later period. When the MOIs were 0.1-1000, the growth of the host bacteria BTXS2 was inhibited, but the concentration of the host bacteria increased after 350 min, which might be related to bacterial resistance because bacteria could resist phage infection through a variety of mechanisms, including spontaneous mutation, restriction endonuclease modification system and adaptive immunity of CRISPR-Cas system [49]. Spontaneous mutation is an important mechanism for phage resistance and phage-host co-evolution [52] that may confer resistance to phage by changing the structure of the receptor on the surface of the bacteria [53], and it also determines the specificity of the phage[54]. Due to the selection of the anti-phage subgroup of BTXS2, they can regrow after the lysis of the bacterial population initially induced by the VB_VpP_BT-1011 [55]. In summary, it can be seen from the Figure 7 that the antibacterial effect of VB_VpP_BT-1011 on the BTXS2 was properly obvious, indicating that it has a strong antibacterial effect on the host. the antibacterial effect varied with different MOIs, but high MOI can effectively inhibit the growth of host bacteria, which provides possible theoretical data for the future use of bacteriophages in food and clinical applications to inhibit and kill pathogenic bacteria. This study provides foundation for further research on the relationship between V.parahaemolyticus related phages, and further research is needed on the diversity of bacteriophages and the mechanism of the antibacterial effect on bacteria especially pathogenic bacteria.