Sampling, isolation and identification of V. harveyi
During the epidemiological investigation of marine fish, thousands suspected pathogenic bacteria strains have been isolated from diseased marine fishes (incl. Trachinotus ovatus, Blotchy rock cod, Sparus macrocephlus, Hypoplectrus indigo, Lates niloticus, Lutjanus erythopterus, Plectropomus leopardus, and Rachycentron canadum). For isolation, the homogenate of lesion tissues (incl. spleen, tail, eye, liver, kidney, mouth, head, body surface, and intestine) have been screened with marine agar 2216E plate. Then the dominant clones have been selected for identification with the gene of 16 s rRNA. Subsequently, Vibrio spp. strains have been chosen for V. harveyi identification by MLSA of toxRVh and rctB [32, 64]. The phylogenetic tree was constructed from the concatenated sequences of toxRVh-rctB genes using Kimura 2-parameter model with the neighbor-joining method, bootstrapped 1,000 times via MEGA6.0 software. 29 Vibrio spp. strains including five V. harveyi strains sequences were downlated from National Coalition Building Institute (NCBI) database as the references. Sampling was done in South China (incl. Hainan or HN, Guangdong or GD, and Fujian or FJ) from 2011 and 2014. The primer pair sequences used for this study were shown in Additional file 1: Table S1.
Antibiotics Susceptibility Testing
Antibiotics susceptibility of V. harveyi isolates were test by disk diffusion method on Mueller-Hinton agar using commercial antibiotic disks (Oxoid) according to Clinical and Laboratory Standards Institute (CLSI) guidelines [65]. Escherichia coli ATCC 35218 were used as quality control strain in each run. 15 antibiotics including furazolidone 300 µg/disk, erythromycin 150 µg/disk, gentamicin 10 µg/disk, rifampicin 5 µg/disk, norfloxacin 10 µg/disk, ciprofloxacin 50 µg/disk, chloramphenicol 30 µg/disk, florfenicol 30 µg/disk, tetracycline 30 µg/disk, trimethoprim-sulfamethoxazole 23.75/1.25 µg/disk, amoxicillin 20 µg/disk, vancomycin 30 µg/disk, tobramycin 10 µg/disk, midecamycin 30 µg/disk, and doxycycline 300 µg/disk were tested. Interpretation of inhibition zones was carried out based on the manufacturers’ and CLSI M45-A guidelines [66].
Detection Of Resistance And Virulence Genes
DNA template preparation
Bacteria cultures were grown by streaking on marine agar 2216E plate at 28 °C for overnight. Then single clone was cultured with marine broth 2216E medium at 28 °C, 200 rpm for overnight. Bacterial lysates used as templates for the PCRs were prepared as follows: 1 mL overnight bacteria was centrifuged and resuspended with 100 µL sterilized water. Then it was cooled at -80 °C for 20 min and heated at 100 °C for 15 min immediately. After having been cooled at ice for 5 min, the suspension was centrifuged for 5 min at 12000 rpm. The supernatant was separated and used as template DNA for PCR assay.
Detection of typical V. harveyi virulence genes
Eight typical V. harveyi virulence genes (incl. four quorum sensing genes cqsS, luxN, luxP, and aphA, haemolysin gene vhh, serine protease encoding gene hflK, transcription regulator encoding genes luxR, and chitinase encoding genes chiA) were detected with the multiplex PCR method that we have validated before [67]. Briefly, the PCR reaction mixture contained Premix Taq (TaKaRa Taq Version 2.0 plus dye) (Takara, Japan) 38.0 µL, each primer 0.1 µM, template DNA 80 ng,and add sterilized water to 50 µL. The reaction was performed in an automatic thermal cycler (Bio-Rad, USA) under the following optimized cycling program: an initial denaturation step of 5 min at 95 °C; 35 cycles of denaturation at 94 °C for 30 s, annealing at 53 °C for 30 s, extension at 72 °C for 60 s; and a final extension at 72 °C for 10 min. The amplified PCR fragments were separated by 2.0% agarose gel which as electrophoresised with 0.5x TBE buffer at 120 V for 75 min. Reaction mixtures with sterilized water and X13SZ03 (V. harveyi 345 of which the completed genome was sequenced with accession numbers of CP025537.1 to CP025540.1) DNA as templates were served as a negative and positive controls, respectively.
Detection of atypical V. harveyi virulence genes
Six atypical V. harveyi virulence genes (incl. V. cholera virulence regulator gene toxRvc, V. cholera haemolysin gene hlyA, V. anguillarum flagella C subunit gene flaC, V. parahaemolyticus haemolysin tdh and trh, and V. vulnificus haemolysin vvh (Ruwandeepika et al., 2010; Lorenz et al., 2017) have been detected with single PCR. Briefly, the PCR reaction mixture contained Premix Taq (TaKaRa Taq Version 2.0 plus dye) (Takara, Japan) 12.5 µL, each primer 0.4 µM, template DNA 20 ng,and add sterilized water to 25 µL. The reaction was performed in an automatic thermal cycler (Bio-Rad, USA) under the following optimized cycling program: an initial denaturation step of 5 min at 95 °C; 35 cycles of denaturation at 94 °C for 30 s, annealing at Tm for 30 s, extension at 72 °C for 60 s / kb; and a final extension at 72 °C for 10 min. The amplified PCR fragments were separated by 1.0% agarose gel which as electrophoresised with 0.5x TBE buffer at 150 V for 30 min. A reaction mixture with sterilized water as templates was served as the negative control. For each atypical virulence gene, at least five PCR products were selected for sequence and homolog analysis using Blast (NCBI) when the positive candidates more than five. Otherwise, all the PCR products were selected for sequence and homolog analysis.
Detection of antibiotic resistance genes
According to antibiotics susceptibility results, the major resistance genes to quinolones (qnrSm, qnrAm, qnrBm, qepA, oqxA, and oqxB), tetracyclines (tetO, tetS, tetW, tetQ, tetB, and tetX), sulfonamides (sul 1, sul 2, and sul 3), macrolide (ermB, and ermC), beta-lactams (blaTEM, and blaSHV), and aminoglycosides [aac(6′)-Ib] were detected by single PCR with the same method as 2.3.3. In addition, an integrase gene (int) derived from the integrons of Vibrio species and the class 1 integrase gene intI1 were tested as well.
Statistical analysis
One-way ANOVA analysis (SNK test) was conducted to examine variations in average number of atypical virulence genes, antibiotic resistance, and antibiotic resistance genes among different locations. Pearson correlation coefficient was calculated to estimate the co-occurrence between the abundance of ARGs and the abundance of virulence genes, between the abundance of ARGs and antibiotic resistance, and between the genotype (absence of resistance gene) and phenotype (resistance to antibiotics). P < 0.05 was considered as significant difference. The Bray–Curtis distance matrix was calculated with the presence virulence genes for distance-based redundancy analysis (db-RDA) which was performed to determine the impact of antibiotic resistance on bacterial virulence. The above-mentioned statistical analyses were performed with PRIMER 6 & PERMANOVA+ [68] and IBM SPSS Statistics 19.0.