3.1 The pathogenicity of strain VSc190401
The diseased half-smooth tongue sole naturally infected by strain VSc190401 showed an apparent abdominal lump (Figure 1A). After dissection, effusion flowed out from the abdominal cavity. The internal organs exhibited serious hyperemia, and the intestinal tract became thin and transparent, which was filled with a large amount of effusion and white pus (Figure 1B). The mortality rate of this case was more than 40%.
During the artificial challenge test, the fish in the negative control group and blank control group remained healthy and showed no symptoms. The half-smooth tongue sole artificially infected in the experiment group with strain VSc190401 also showed the same symptoms as the naturally infected fish, such as effusion in the abdominal cavity, internal organ hyperemia, and thin and transparent intestine. The artificially infected turbot did not appear abdominal lump, while the organ hyperemia and intestinal inflammation were observed. Artificial infection confirmed that the V. scophthalmi strain VSc190401 had strong pathogenicity to fish.
Figure 1 The symptoms of diseased fish with natural infection and artificial infection V. scophthalmi strain VSc190401.
A: Naturally infected half-smooth tongue sole had obvious abdominal lump. B: Naturally infected half-smooth tongue sole showed serious internal organ hyperemia and enteritis. C: Artificially infected half-smooth tongue soles showed the same symptoms of abdominal limp. D: Artificially infected half-smooth tongue soles appeared the same internal organ lesions as the naturally infected case. E: Artificially infected turbot did not appear abdominal lump. F: Artificially infected turbot showed serious organ hyperemia and enteritis. Bar = 2 cm.
3.2 Genomic information of strain VSc190401
After assembly, the size of the whole genome of V. scophthalmi strain VSc190401 was 3,541,838 bp, including two circular chromosomes (Chr I 3,286,294 bp and Chr II 202,664 bp) with an approximate GC content of 45.00% and 45.37%, respectively, and two plasmids (plasmid I 24,538 and plasmid II 28,342 bp) with an approximate GC content of 43.32% and 43.41%, respectively. The strain contained 3,185 coding genes, among which Chr I consisted of 2,943 CDSs, 104 tRNA genes, and 37 rRNA genes. Chr II contained 188 CDSs. Plasmid I contained 32 CDSs, and plasmid II contained 22 CDSs. Figure 2 shows the genome information.
Figure 2 Circular genome maps of V. scophthalmi strain VSc190401
Note: The outermost circle is the identification of genome size. The second and the third circle are the CDSs on the positive and negative strands, respectively, and different colors indicate the different functional annotations of CDSs in the COG database. The fourth circle is rRNA and tRNA. The fifth circle is the GC content, the red part outside indicates that the GC content of the region is higher than the average GC content of the whole genome, the blue part inward indicates that the GC content of the region is lower than the average GC content of the whole genome, and the higher peak value means the greater difference from the average GC content. The innermost circle is the GC-Skew value, and its algorithm is , which can assist to determine the leading strand and lagging strand. In general, the leading strand GC-skew > 0 and the lagging strand GC-skew < 0. The green part outside means GC-skew > 0, the orange part inward means GC-skew < 0, and the higher peak value means larger value. The legend circle1 is the functional classification in the COG database, and the legend circle2 is a different RNA classification.
3.3 Phylogenetic analyses
ANI analysis showed that V. scophthalmi strain VSc190401 was clustered with V. scophthalmi VS-12 and VS-05 in the whole genome of eight selected strains, and there were some genetic differences (Figure 3). The annotation information in the NCBI database indicated that V. scophthalmi VS-12 and VS-05 strains were all isolated from Japanese flounder cultured in Korea, containing three and two plasmids in addition to two chromatins, respectively. Based on the ANI analysis, the whole genome of V. scophthalmi strain VSc190401 was found to be genetically different compared with Korean isolate strains VS-12 and VS-05, suggesting that there were some differences in host diversity or pathogenicity between Chinese and Korean isolates.
Figure 3 Phylogenetic tree analysis based on ANI values of V. scophthalmi strain VSc190401 and the complete genomes of eight bacterial strains downloaded from the NCBI database. The scale represents genetic distance.
3.4 Functional annotation
COG annotation results showed that 2,648 genes were annotated into 22 types of genes, accounting for 83.14% of total genes in V. scophthalmi strain VSc190401. The number of each type of gene was as follows: one A-type gene (RNA processing and modification), two B-type genes (chromatin structure and dynamics), 138 C-type genes (energy production and conversion), 36 D-type genes (cell cycle control, cell division, chromosome partitioning), 199 E-type genes (amino acid transport and metabolism), 68 F-type genes (nucleotide transport and metabolism), 128 G-type genes (carbohydrate transport and metabolism), 104 H-type genes (coenzyme transport and metabolism), 59 I-type genes (lipid transport and metabolism), 167 J-type genes (translation, ribosomal structure and biogenesis), 135 K-type genes (transcription), 192 L-type genes (replication, recombination and repair), 161 M-type genes (cell wall/membrane/envelope biogenesis), 53 N-type genes (cell motility), 118 O-type genes (posttranslational modification, protein turnover, chaperones), 147 P-type genes (inorganic ion transport and metabolism), 28 Q-type genes (secondary metabolites biosynthesis, transport and catabolism), 687 S-type genes (function unknown), 140 T-type genes (signal transduction mechanisms), 99 U-type genes (intracellular trafficking, secretion, and vesicular transport), and 41 V-type genes (defense mechanisms) (Figure 4). Supplementary Table 1 lists all COG functional annotation information.
Figure 4 COG functional annotation of CODs in the whole genome of V. scophthalmi strain VSc190401.
The functional annotation results in the GO database showed that 2,298 genes were annotated into three types of genes, which accounted for 72.15% of total genes of V. scophthalmi strain VSc190401. Among them, 1,327 genes were related to the cellular component, 1,850 genes were related to the molecular function, and 1,795 were related to the biological process. The GO terms with the highest numbers of genes in the classification of the biological process were oxidation-reduction process (174 genes, 5.46%) and regulation of transcription/DNA-templated (141 genes, 4.43%). The GO terms with the highest numbers of genes in the classification of cellular components were integral components of membrane (641 genes, 20.13%), cytoplasm (336 genes, 10.55%), and plasma membrane (91 genes, 2.86%). The GO terms with the highest numbers of genes in the classification of molecular function were ATP binding (281 genes, 8.82%), DNA binding (244, 7.66%), metal ion binding (127 genes, 3.99%), and transcription factor activity/sequence-specific DNA binding (82 genes, 2.57%). More information was shown in Figure 5. Supplementary Table 2 lists all GO functional annotation information.
Figure 5 GO functional annotation of CODs in the whole genome of V. scophthalmi strain VSc190401.
The results of the KEGG pathway analysis showed that 1,915 genes were annotated into 196 known metabolic pathways. The metabolic pathway with the largest number of genes was the biosynthesis of amino acids, containing 106 genes, followed by two-component system (96 genes), carbon metabolism (83 genes), ABC transporters (81 genes), and purine metabolism (65 genes). Cluster analysis showed that 196 metabolic pathways were categorized into six classifications of cellular processes, metabolism, human diseases, genetic information processing, organismal systems, and environmental information processing, and the numbers of genes in these six classifications were 235, 1348, 112, 219, 47 and 268, respectively (Figure 6). The 235 genes in the classification of cellular processes could be divided into four categories, and most of them were clustered into cell motility (74 genes) and cellular community-prokaryotes (135 genes). The 1,348 genes in the classification of metabolism were divided into 12 categories. The categories with the largest gene number included global and overview maps (242 genes), carbohydrate metabolism (256 genes), amino acid metabolism (202 genes), and metabolism of cofactors and vitamins (149). The 112 genes in the classification of human diseases were clustered into 10 categories, and the categories with the largest gene number were drug resistance: antimicrobial (37 genes), and infectious diseases: bacterial (27 genes) and neurodegenerative diseases (13 genes). The 219 genes in the classification of genetic information processing were clustered into four categories. The categories with the largest gene number were translation (79 genes), replication and repair (90 genes), and folding, sorting and degradation (46 genes). The 47 genes in the classification of organismal systems were divided into eight categories. Among them, the top three categories with the largest gene number were the immune system (10 genes), aging (11 genes), endocrine system (nine genes), and environmental adaptation (seven genes). The environmental information was divided into two categories, including signal transduction (115 genes) and membrane transport (153 genes). Supplementary Table 3 shows all KEGG pathway annotation information.
Figure 6 The KEGG pathway annotation results of V. scophthalmi strain VSc190401. The ordinate indicates the level 2 KEGG pathway classification, and the abscissa indicates the number of genes under the annotation of this classification. Different column colors represent the level 1 KEGG pathway classification. The rightmost bar indicates the number of genes under different level-1 classifications. Since the same gene may be annotated into multiple level-2 classifications, the number of genes classified by level 1 will be de-redundant.
Through the IslandViewer online system, 10 genomic islands were predicted to be contained in the whole genome of the strain Vsc190401. They were all located on chromosome I. The longest genomic island was 34,105 bp, and the shortest one was 7,883 bp. Supplementary Table 4 and Supplementary Figure 2 illustrate their detailed information.
3.5 Prediction of virulence genes of the strain VSc190401
To date, there are few studies on virulence factors of V. scophthalmi, and no specific virulence genes have been reported. A total of 334 potential virulence genes were predicted in the whole genome of the strain VSc190401. Moreover, 175 genes had annotation information in VFDB databases, including 107 offensive virulence genes, 31 defensive virulence genes, seven virulence-associated regulation genes, and 30 non-specific virulence genes (Table 1). Supplementary Table 5 shows detailed information.
Table 1 The annotation of virulence factors of the strain VSc190401 in VFDB databases
Virulence primary categories
|
Virulence Secondary categories
|
Gene numbers
|
Offensive virulence factors
|
Toxin
|
2
|
Offensive virulence factors
|
Secretion system
|
13
|
Offensive virulence factors
|
Adherence
|
81
|
Offensive virulence factors
|
Invasion
|
11
|
Defensive virulence factors
|
Cellular metabolism
|
1
|
Defensive virulence factors
|
Antiphagocytosis
|
21
|
Defensive virulence factors
|
Stress protein
|
9
|
Regulation of virulence-associated genes
|
Regulation
|
7
|
Nonspecific virulence factor
|
Iron uptake system
|
30
|
Among the offensive virulence genes, there were 81 genes related to adhesion, including flagella and pilus formation or motility exercise-related genes, such as Flg, Fle, Flh, Fli, Tcp, PilB/D/G/H/R/T/U, accessory colonization factor AcfB/D [26], adhesion proteins, Lap, OmpU [27] and so on. The results of comparative analysis to the NR database showed that the coverage of coding protein sequences was 96.44%-100%, and the identity was 65.8%-100%. There were 11 genes related to invasion, including flagella and T4SS genes. The coverage of their coding proteins was 56.14%-100%, and the identity was 50%-100% in the NR database. Moreover, 13 genes were related to secretion systems. Their sequence information was all similar to T3SS, T4SS, and T6SS of Gram-negative bacteria. The coverage of their coding proteins was 91.84%-100%, and the identity was 56.7%-100%. Two toxin genes were similar to Cya gene, the coding proteins of which were suspected as calmodulin-sensitive adenylate cyclase-haemolysin bifunctional protein [28]. The identity of their coding proteins was 98%-99.7%.
The defensive virulence genes included 21 antiphagocytosis-related genes, nine stress protein-related genes, and one cellular metabolism-related gene. The antiphagocytosis-related genes included two categories of capsule genes (Cpa, Cps) [29, 30] and alginate genes (algB/Q/U/R/Z, MucA) [31, 32]. The identity of their coding proteins was 85.6%-100%. The stress protein-related genes included superoxide dismutase enzyme gene (sodB), ATP-binding cassette transporter gene (MntABC), DNA repair protein gene (RecN), and respiratory metabolism gene (ClpCP) [33]. The identity of the protein sequence was 99%-100%. The cellular metabolism-related genes were isocitratelyase coding gene fragments, with a coverage of 100% and an identity of 100%.
Among the non-specific virulence genes, the most were ATP-binding cassette transporter genes (20 genes, 98.4%-100% protein sequence identity) and bacitracin-related genes (six genes, 90%-100% protein sequence identity). Moreover, iron uptake system-related genes, such as FbpABC and FeoAB (99.4%-100% protein sequence identity), were also found [34, 35].
The virulence-associated regulation genes contained two types of genes, RelA and PhoP. NR database comparative analysis results showed that the coverage of coding protein sequences was 100%, and their identity was 88.4%-100%. Other studies have confirmed that RelA and Phop gene regulate the synthesis of bacterial virulence factors as well as their primary and secondary metabolites, thus affecting the bacterial pathogenicity [36, 37]. In addition, the virulence-related gene sequences of ompA (identity 90%-100%), hemolysin (identity 96.7%-100%), beta-hemolysin/cytolysin (identity 98.2%-100%), enterobactin (identity 99.8%-100%), and T2SS (identity 97.2%-100%) were also found in the whole genome of the strain VSc190401.
Further analysis showed that the strain VSc190401 contained 36 secretion system-related genes, including one type I, 11 type II, six type IV, four type VI, 11 Sec-SRP pathway and three twin-arginine targeting (Tat) pathway. The analysis results also indicated that the strain VSc190401 only had a complete type II secretion system (the detailed information was shown in Supplementary Figure 2). This finding suggested that the type II secretion system was probably the only one product export pathway.
3.6 The drug resistance phenotype and genotype analysis of the strain VSc190401
A total of 38 antibiotics belonging to 10 categories were selected to test the antimicrobial phenotype of strain VSc190401 through the Kirby-Bauer disk diffusion method. The antibiotics of 10 categories were β-lactam, aminoglycosides, macrolides, tetracyclines, polypeptides, quinolones, sulfonamides, nitrofurans, amphenicols, and others. The results showed that the strain VSc190401 was resistant to all aminoglycosides (including neomycin, streptomycin, kanamycin, gentamicin, amikacin), macrolides (including erythromycin, azithromycin, clarithromycin, acetylspiramycin) and amphenicols (including chloramphenicol and florfenicol). It was sensitive to polypeptides (polymyxin B), quinolones (including pipemidic, nalidixic, fleroxacin, lomefloxacin, ciprofloxacin, ofloxacin, norfloxacin, enrofloxacin), sulfonamides (sulfamethoxazole) and nitrofurans (including furazolidone). In β-lactam antibiotics, the strain showed resistance to cefoperazone, ceftizoxime, cefotaxime, ceftriaxone, ceftazidime, cefradine and oxacillin, while it was sensitive to penicillin, ampicillin, cefalexin and cefazolin. In tetracyclines antibiotics, the strain showed resistance to doxycycline, while it was sensitive to minocycline and tetracycline. Among other antibiotics, the strain showed resistance to rifampicin, while it was sensitive to novobiocin.
The results of Blast analysis in the CARD database showed that 180 drug resistance genes belonging to 27 categories were found in the whole genome of the strain VSc190401 (Table 2). In terms of drug resistance phenotype and genotype correlation, the strain VSc190401 contained the streptomycin-resistant genes, gidB and vatB [38], and rifampin-resistant genes, rpoB [39]. The phenotype was consistent with the genotype. The novobiocin-resistant genes, alaS and cysB [40], nalidixic-resistant genes, gyrA/B and parC/E [41], tetracycline-resistant genes, tet31/34/35/B/R/S/T and adeR [42], ciprofloxacin-resistant genes, patA/B [43], and bpolymyxin B-resistant genes, PmrA/C/E, LpxA/C and rosB [44], were found in the whole genome, while the strain was sensitive to these drugs. The phenotype was inconsistent with the genotype. Besides, the multiple resistance genes and complex genes of different antibiotics were also found in the strain, such as multiple resistance genes, drrA [45], cpxA and ompR [46], multidrug efflux pump systems, including ABC transporter superfamily of proteins msbA [47]; two-component signal transduction system, EvgAS [48]; two-component regulatory system, BaeSR [49]; an activator of mtrCDE multidrug-resistance efflux pump, mtrA [50]; MexEF-Opr multidrug efflux systems [51]; two-component regulatory systems, VanRS [52] and ArlRS [53]. These results showed that the resistance mechanism of the strain VSc190401 might be complex. Phenotypes of multiple drug resistance without specific resistance genes suggested that there were multiple drug metabolism pathways. Supplementary Table 6 shows the annotation information of all drug resistance genes.
Table 2 The drug-resistance genes annotation of the strain VSc190401 in CARD databases
Drug resistance categories
|
Gene numbers
|
Drug resistance categories
|
Gene numbers
|
pleuromutilin antibiotic
|
3
|
penem
|
3
|
carbapenem
|
7
|
phenicol antibiotic
|
12
|
sulfonamide antibiotic
|
5
|
rifamycin antibiotic
|
4
|
aminoglycoside antibiotic
|
10
|
isoniazid
|
3
|
macrolide antibiotic
|
45
|
triclosan
|
11
|
glycopeptide antibiotic
|
4
|
acridine dye
|
10
|
tetracycline antibiotic
|
36
|
peptide antibiotic
|
14
|
monobactam
|
5
|
lincosamide antibiotic
|
1
|
diaminopyrimidine antibiotic
|
4
|
fluoroquinolone antibiotic
|
38
|
streptogramin antibiotic
|
5
|
nitroimidazole antibiotic
|
6
|
glycylcycline
|
5
|
penam
|
27
|
sulfone antibiotic
|
3
|
cephalosporin
|
10
|
aminocoumarin antibiotic
|
9
|
nybomycin
|
1
|
cephamycin
|
9
|
|
|
3.7 Predictive analysis of pathogen-host interaction between the strain VSc190401 and host
According to the annotation results of the PHI database, the strain VSc190401 contained 518 genes related to pathogen-host interaction. Among them, there were 346 genes related to reduced virulence, 44 genes related to loss of pathogenicity, 36 genes related to hypervirulence, 12 genes related to lethal factors, 11 genes related to effector, three genes related to chemical resistance, and two genes related to chemical sensitivity, and there were 108 genes with unaffected pathogenicity. The hypervirulence and effector genes were the key genes in correlation with pathogenicity. The hypervirulence of the strain VSc190401 included T3SS function genes, esaN and GdpX1, virulence regulatory factors, MorA, ccpE, RsmA, raxP and CdpR [54–57], two-component sensor kinase, BfiS [58], histidine kinase and response regulator two-component system, TcrX/Y [59], and iron ion transporters, feoB and pchD. The effector genes of the strain VSc190401 included T6SS components, VgrG and clpV [60], phospholipase D effector, LpdA [61], pleiotropic effector of virulence synthesis and pathogenicity attenuation, RpiRc [62]. Supplementary Table 7 lists the detailed annotation information of pathogen-host interaction genes