Isolation and characterization of a novel phage belonging to a new genus against Vibrio parahaemolyticus

V. parahaemolyticus 1.1997^T brought from Guangdong marine pathogenic Vibrio company in China was used as the host bacterium for phage isolation. It was incubated in rich organic (RO) medium (1 M peptone, 1 M yeast extract, and 1 M sodium acetate in artificial seawater, pH 7.5) at 30℃ with a shaking speed of 160 rpm min^− 1 [24]. The sewage samples used for phage isolation were collected from the seafood markets in Guangzhou, China (23.10°N, 113.33°E), and filtered through a 0.22-µm membrane (Millipore, Massachusetts, USA) to remove bacteria and large particles. The sewage samples were added into an exponentially growing host bacterial culture to allow plaque formation using the double-layer agar method [25]. A clear individual plaque was collected and suspended in storage medium (SM; 8 mM MgSO₄, 50 mM Tris-HCl, and 100 mM NaCl, pH 7.5). After purifying at least five times, the well-separated plaque was collected and stored in SM at 4 °C.

To obtain high-titer phage suspension, a purified phage plaque was inoculated into bacterial culture and amplified overnight, followed by centrifugation at 12,000 × g for 10 min. The supernatant was filtered through a 0.22-µm membrane to remove cell fragments. The filtrate was precipitated with polyethylene glycol 8000 (10% w/v) overnight. Then, phage pellets were obtained through centrifugation (10,000 × g, 60 min, 4 °C) and resuspended in SM. The phage particles were subjected to cesium chloride solutions (ρ = 1.3, 1.5, 1.7 g mL^− 1) for purification and centrifuged at 200,000 × g at 4 °C for 24 h using an Optima L-100 XP ultracentrifuge (Beckman Coulter, CA, USA). The purified phage particles were dialyzed through 30-kDa superfilters (Millipore, Bedford, MA, USA).

Phage morphology was characterized by TEM. Briefly, approximately 20 µL of phage suspension was added onto the surface of a copper grid to adsorb in darkness for 30 min. Then, the phage sample was negatively stained with phosphotungstic acid (1%, pH 7.0) for 20 min and dried for 30 min. The phage sample was examined using a JEM-2100 transmission electron microscope (JEOL, Tokyo, Japan).

A total of 35 Vibrio strains, including 6 strains of V. parahaemolyticus and 29 strains of other species (V. alginolyticus, V. antiquarius, V. azureus, V. campbellii, V. caribbeanicus, V. chagasii, V. cholerae, V. fortis, V. harveyi, V. inhibens, V. neocaledonicus, V. owensii, V. rotiferianus, V. tubiashii, V. variabilis and V. xuii), were used to determine the lytic host range of the phage isolated in this study with the double-layer agar method (Table S1). The bacteria strains were grown on the RO agar medium plates and single colony was collected and then incubated in RO liquid medium at 30℃ with a shaking speed of 160 rpm min^− 1. The log-phase bacteria strains (incubated about 2–10 h) were mixed with the phage and kept in the dark for 20–30 min for infection. After infection, the phage-host mixtures were added with molten RO agar medium and poured onto solid agar plates, then incubated overnight at 28 °C. All bacteria strains have been tested at least three times. The presence of plaques on a bacterial lawn was checked to determine phage infection of the host bacterium.

A one-step growth curve was determined to study the infectivity and replication ability of the phage. Briefly, phages were added to an early log-phase host culture (V. parahaemolyticus 1.1997^T) at a multiplicity of infection of 0.03 and incubated for 20 min at room temperature in the dark. Then, phages that were not adsorbed were removed by centrifugation (8,000 × g, 4 °C, 5 min), and the cell pellets were washed and resuspended in 100 mL of RO medium. The phage suspension was incubated at 28 °C with a shaking of 160 rpm min^− 1. Subsamples were obtained at 10-min intervals and assayed by the double-layer agar method. The burst size was calculated as the ratio between the number of phages before and after the burst [26]. The experiment of one-step growth curve has been conducted with three replications.

To understand the efficiency of bacterial inactivation, a lytic activity for the phage against V. parahaemolyticus were assessed at different multiplicity of infections (MOIs) by monitoring the OD₆₀₀ (optical density measurements at a wave-length of 600 nm) using a microplate reader (Synergy H1, Bio-Tek, USA). Briefly, freshly prepared host culture was added to the a 96-well plate before infection with high-titer phage suspension at various MOIs (0.001, 0.01, 0.1, 1 and 10). Wells inoculated with RO medium or V. parahaemolyticus only was served as control groups. The plate was placed in the instrument and incubated at 30℃ with orbital shaking. The OD₆₀₀ of the cultures were monitored in real time and recorded every 30 min. The assay was conducted with eight replications for each treatment.

Phage DNA was obtained by using the phenol–chloroform extraction method. Briefly, approximately 1 mL of high-titer phage suspension was treated with proteinase K, sodium dodecyl sulfate (10% w/v), and ethylenediaminetetraacetic acid (EDTA; pH 8.0) and incubated at 55 °C for 3 h. The digested sample was purified with phenol/chloroform/isoamyl alcohol (25:24:1 v:v:v) and chloroform/isoamyl alcohol (24:1 v:v) to remove any debris. Next, the DNA pellet was washed with pre-cooled 70% ethanol, air-dried at room temperature, dissolved in Tris-EDTA buffer (10 mM Tris-HCl, 1 mM EDTA, pH 8.0), and stored at 4 °C. The phage genome was sequenced and assembled using the Illumina HiSeq 4000 platform with a 150-bp paired-end DNA library and Velvet software (v.1.2.03) [27].

The phage termini and packaging mechanism were identified by the online service PhageTerm on a Galaxy-based server (https://​galaxy.​pasteur.​fr) [28]. The GeneMarkS online server (http://​topaz.​gatech.​edu/​GeneMark/​genemarks.​cgi) was used to identify putative open reading frames (ORFs) [29]. The functions of ORFs were annotated by BLASTP online against the non-redundant (nr) database of the National Center for Biotechnology Information (NCBI), with a cut-off e-value of < 10^− 5, and the results were checked manually. Putative transfer RNA (tRNA) genes were detected with tRNAscan-SE (http://​lowelab.​ucsc.​edu/​tRNAscan-SE/​) [30]. Virulence factors and antibiotic resistance-encoding genes were searched using the Virulence Factor Database (VFDB, http://​www.​mgc.​ac.​cn/​VFs/​main.​htm) and the Comprehensive Antibiotic Resistance Database (CARD, https://​card.​mcmaster.​ca/​analyze/​rgi) webservers, respectively [31, 32]. In addition, genome-based life cycle classification was also carried out using an online platform called PhageAI (https://​phage.​ai/​). The complete genome data was deposited in the GenBank database with accession number OP793884.

Viral CONTigs Automatic Clustering and Taxonomy v.2.0 (vConTACT2) was used to compare the isolated phage against the Prokaryotic Viral RefSeq 207 database using whole genome gene-sharing profiles, and related phages were identified by genome pairs with a similarity score > 1 [33]. The whole phage sequence was also submitted to NCBI BLASTN to search forsimilar sequences. To analyze the evolutionary relationships of the phage, complete amino acid profiles of phages were submitted to the Virus Classification and Tree Building Online Resource (VICTOR, https://​ggdc.​dsmz.​de/​victor.​php) for phylogenetic tree construction [34, 35]. All pairwise comparisons of the amino acid sequences were conducted using the Genome-BLAST Distance Phylogeny (GBDP) method with the settings recommended for prokaryotic viruses [34, 36]. Branch support was inferred from 100 pseudo-bootstrap replicates each, and the tree was rooted at the midpoint and visualized with ggtree [37, 38]. In addition, intergenomic nucleotide sequence similarity and aligned genome fractions within the imported phages were plotted with the Virus Inter-genomic Distance Calculator (VIRIDIC) using default parameters [39].

Vibriophage vB_VpaM_R16F (hereafter R16F) was isolated from a sewage sample from Chigang seafood market, Guangzhou, China, by infecting V. parahaemolyticus 1.1997^T. R16F formed clear plaques around 2–3 mm in diameter after infection for 12 h (Fig. 1A). With extension of the incubation time, the phage plaques became larger (up to 7–8 mm in diameter) with a turbid halo on the host bacterial lawn (Fig. 1A). Halo formation has previously been described as an indicator of phage-associated exopolysaccharide depolymerization [40, 41]. The depolymerases encoded by phages can specifically cleave polysaccharides (e.g. capsular polysaccharides, exopolysaccharides or lipopolysaccharides) of the host bacteria and therefore provide a significant advantage for phages to adsorb onto their hosts [42, 43]. Such halo zones have been observed for various phages, including phages that infect Pseudomonas putida, Klebsiella pneumoniae, and V. alginolyticus [44‐46]. TEM revealed that R16F has a narrow neck or collar region, an icosahedral capsid (head diameter 83 ± 7 nm), and a complete tail with contracted sheath (49 ± 2–100 ± 3 nm). It belongs to the myovirus morphotype (Fig. 1B).

To further understand the lytic cycle of R16F, a one-step growth curve was constructed, which showed a short latent period for R16F, less than 10 min (Fig. 1C). The latent period is the minimum time it takes from phage adsorption to lysis of the host with the release of progeny virions. The burst size for R16F, defined as the average number of progeny virions liberated by one infected host cell at the completion of a growth cycle, estimated from the one-step growth curve, was 13 plaque-forming units (PFU) cell^− 1 (Fig. 1C). Short latent periods (< 30 min) and small burst sizes (ranging from 10 to 100 PFU cell^− 1) have been observed for other phages that infect V. parahaemolyticus [7, 10, 47‐54]. A trade-off between burst size and latent period has been reported in lytic phage generation because phages are released from infected cells at the cost of destruction of the machinery necessary to produce more phage progeny [55]. Thus, phages with short latent periods at the cost of their burst size may be viewed as specialists for propagation when bacteria are prevalent.

R16F exhibited a narrow lytic spectrum and only infected the original host V. parahaemolyticus 1.1997^T among the species tested (Table S1). R16F was insensitive to chloroform, suggesting that there was no lipid in the viral capsid. This result was consistent with the previous suggestion that lipids are rare among phages, occurring in < 4% of isolates [56].

Given that the efficiency of the phage in bacterial inactivation is one of the critical properties should be considered for potential candidates in phage therapy, lytic activity analysis of R16F was tested in this study. The killing curves showed that R16F was highly effective against V. parahaemolyticus 1.1997^T at all MOIs tested (Fig. 2). Moreover, in these samples, there was no observable difference in inhibitory effects after 10 h of incubation, even at the lowest MOI (MOI = 0.001). R16F exhibited inhibition of bacterial growth regardless of the value of the MOIs, suggesting that increasing phage loads would lead to a limited positive effect on host inactivation of the phage with strong lytic ability. Although few phages were applied in bacteria inhibition, V. parahaemolyticus 1.1997^T could still be lysed by R16F within a short time which might benefit from the short latent of the phage.

Whole genome sequencing revealed that the R16F genome is a 139,011-bp linear double-stranded DNA with a G + C content of 35.21%, which is lower than its host (45.3%). According to the results from PhageTerm, a fixed position was recognized on the DNA of R16F and 353-bp direct terminal repeats (DTR) were observed, suggesting that the DNA termini of R16F belong to the DTR class. A total of 260 ORFs were identified in the genome of R16F, of which only 57 (21.92%) have known functional domains, while the remaining 203 (70.77%) were annotated as hypothetical proteins (Fig. 3; Table S2). The functional ORFs were identified to encode proteins that can be arranged in functional categories including phage structural formation, genome packaging, nucleotide metabolism, integration, lysis, and other functions. Among these functional ORFs, about 12.69% (33/260) are conserved. The genome of R16F was predicted to encode two tRNAs (tRNA^Arg and tRNA^Met) carried anticodons TCT (serine) and CAT (histidine). No antibiotic resistance- or virulence factor-related gene was found in the genome of R16F.

On the basis of NCBI BLASTN analysis, phages qdvp001, VPMCC14, and PWH3a-P1 share the highest sequence identities (95.15%, 84.98%, and 88.33%, respectively) with R16F, but they share only 33%, 15%, and 12% query coverage. To investigate the evolutionary relationships and taxonomic status of R16F, the vConTACT2 database was used to detect other homologous phages. A total of 37 phages with a similarity score of > 1 were identified, including six vibriophages with high scores of > 40 (Table S3). All the phages detected by vConTACT2 and other phage sequences obtained from NCBI BLASTN were used to construct a phylogenetic tree based on pairwise comparisons of the amino acid sequences of R16F. According to the phylogenetic analysis, R16F was most closely related to vibriophage qdvp001 with only 67% pairwise sequence identity (Fig. 4A). To better estimate the similarity between phage genomes, all phages from the tree were imported into the VIRIDIC software and demarcated based on the default similarity thresholds of 95% for species and 70% for genus [39, 57]. The VIRIDIC results showed that the highest value of intergenomic similarity was only 34.9%, observed between R16F and qdvp001 (Fig. 4B). Therefore, R16F is suggested to belong to a novel genus. Overall, R16F is considered a novel phage with a genome that differs from all previously described phage sequences.

Phage R16F harbors a gene (ORF 21) encoding an ultraviolet damage endonuclease (UVDE), which is involved in DNA repair. UV light may cause damage or mortality to bacteria and phages [58]. In bacterial cells, UVDE has been defined as a repair protein that is capable of identifying and removing various DNA lesions including UV photoproducts as well as non-UV-induced DNA damage such as a basic sites, strand breaks, and gaps [59]. In addition to being found in the genomes of bacteria and fungi, UVDE is present in the genomes of some phages [60, 61]. UVDE of bacteriophage T4 was involved in the dark repair of UV light damage in T4-infected cells [62]. Furthermore, a conserved catalytic GIY-YIG domain of nucleotide excision repair endonucleases UvrC, Cho, and similar proteins, is encoded in the R16F genome (ORF 31). UvrA, UvrB, and UvrC mediate nucleotide excision repair following various types of structurally unrelated DNA damage and remove the complete oligonucleotide containing the damage [63‐67].

A gene (ORF 24) encoding a type II TA system HicB family antitoxin (HicB) was found in the genome of R16F. TA systems are composed of a two-gene cassette encoding a toxic protein and an antitoxin protein, the latter counteracting the toxicity of the former [68]. Type II TA systems appear to be the most abundant and diverse in bacterial and archaeal genomes, largely being found in mobile gene modules such as plasmids and phages, but also in bacterial chromosomes, and they are highly prone to horizontal gene transfer [68‐70]. Type II TA systems in bacteria are thought to be involved in diverse biological processes, including plasmid maintenance, phage inhibition, persistence, stress response, and biofilm formation [68, 71]. The hicAB system, a type II TA system, encoding toxin HicA and antitoxin HicB proteins, has been predicted to be associated with several functions, including RNA-binding, persister cell formation and involvement in extracytoplasmic stress responses [69, 72, 73]. On the basis of BLASTP analysis, almost all homologues of the HicB in R16F were from bacterial genomes, implying that the HicB in R16F most likely derived from bacteria via horizontal gene transfer. However, no gene encoding a HicA toxin protein was identified in R16F. Hitherto, intact TA systems have been found in prophages, such as prophages induced from Streptococcus suis, Mannheimia haemolytica and Acinetobacter. However, this is the first report of a component of a TA system in a lytic phage [74‐76]. A recent review suggested that phages have evolved defenses against TA systems by incorporating active TA systems [77]. The roles of TA systems in phage genomes, especially those of lytic phages, remain unclear, but they are speculated to perform functions in phage–bacteria co-evolution.

ACP (ORF 117 in phage R16F) serves as a ubiquitous and highly conserved transporter of acyl intermediates during the synthesis of fatty acids [78]. ACPs in bacteria are acyl donors for synthesis of various products, including endotoxin and acylated homoserine lactones, which are involved in quorum sensing. The unique nature of these growth and pathogenesis processes makes ACP-dependent enzymes a target for antibacterial agents [78]. In addition to being widespread in bacteria, homologs of ACP have been identified in several phages, such as V. helene phage 12B3, Erythrobacter phage vB_EliS-R6L, and some prophages [79, 80]. ACP encoded by phages was previously identified as an auxiliary metabolic gene and potentially involved in the biosynthesis of secondary metabolites in host cells [80].

The genome of phage R16F also harbors a phoH gene (ORFs 234 and 235). In Escherichia coli, gene transcription of the phosphate regulator is activated in response to phosphate restriction, and its products participate in the transportation and use of various forms of phosphate [81‐83]. The phoH gene has been detected in marine phages, including vibriophages, cyanophages, and roseophages [24, 84‐87]. Although many studies have hypothesized that phage-encoded phoH might play an important role in assisting host phosphorous metabolism, no concrete evidence has been reported to support this conjecture.

The biological and genomic characteristics are critical in evaluating phage fitness and identifying candidates for use in phage therapy. Several general properties should be considered, including bactericidal ability, efficiency, and whether the phage carries endotoxins [88]. Phages have often evolved to quickly suppress bacterial proliferation when the bacterial density is sufficiently high [55]. This phenomenon was observed here for R16F (latent period < 10 min, with small burst size). Moreover, the efficiency of bacterial inactivation was evaluated in various MOIs. Our findings demonstrated that R16F has highly effective inactivation in the prevention and control of V. parahaemolyticus. Therefore, R16F with strong lytic ability may be effective against Vibrio outbreaks in a short time. The narrow lytic spectrum (host range) of R16F would enable use of this phage to specifically target V. parahaemolyticus. Moreover, previous study suggested that various type of phage with narrow host specificity can be mixed in cocktails to control a range of pathogenic strains [48].

R16F was found to encode an endolysin (ORF 233) that produced at the later stage of phage replication and would be able to hydrolyze the bacterial peptidoglycan layer of the host cell wall from within, leading lysis of the host bacterium and release newly assembled phage particles [89]. In recent studies, phage-encoded endolysin are believed to have good performance to control foodborne pathogens including gram-negative bacteria and remove bacterial biofilms [42, 90]. The R16F-encoded endolysin showed 95.76% amino acid sequence identity with the endolysin encoded by phage qdvp001 (lysqdvp001). Lysqdvp001 and its homologues are highly divergent enzymes with superior lytic activity and a broader spectrum than their parent phages [51, 91]. Therefore, we infer that both vibriophage R16F and R16F-derived endolysin are promising for application in biocontrol.

No antibiotic resistance- or virulence factor-related gene was predicted in the genome of R16F using VFDB and CARD, indicating that R16F infection is unlikely to result in enhancement of Vibrio virulence or contamination of aquaculture environments with antibiotic resistance genes. Although R16F encodes a serine recombinase (ORF 52), no attachment sites on the phage or bacterial genomes (attP–attB) were observed in the 1000-nt sequences adjacent to either end of the serine recombinase-encoding gene. In addition, a PhageAI analysis revealed that the probability of phage R16F being virulent was found to be 91.53%. Therefore, R16F is considered safe at the genetic level as a biocontrol agent.