Objective Genomic structures and functional variations of Pseudomonas plecoglossicida strains isolated from diseased Larimichthys crocea were studied, in order to provide a theoretical foundation for developing effective prevention and control strategies.

Methods Whole-genome resequencing was performed on 8 P. plecoglossicida strains isolated from diseased large yellow croakers across different times and locations. Along with ATCC® 700383™ as the reference strain, the PacBio and Illumina sequencing platforms were applied on the strains. The complete circular genomes were assembled to annotate functions associated with the genes. Comparative genomic analysis was conducted on all isolated strains including BCH2017050402, which represented the pathogen of visceral white spot disease in large yellow croaker. Pan-genome analysis was performed on 13 P. plecoglossicida genomes using bioinformatics tools.

Results The complete genome sequences of 8 P. plecoglossicida strains isolated from diseased large yellow croaker ranged from 5469808 bp to 5525584 bp containing 62.66%–62.76% GC contents and encoded 4997–5045 predicted genes with a single circular chromosome but no plasmids. BCH2017050402 had a genome of 5.52 Mb (5525533 bp), containing 5045 genes and 108 non-coding RNAs (ncRNAs). The functional annotation identified 3814 genes with Clusters of Orthologous Group(COG) assignments, 4702 genes mapped to Kyoto Encyclopedia of Genes and Genomes (KEGG) metabolic pathways, 19 genomic islands, and 460 virulence factor genes. It also harbored diverse protein secretion systems, including 4 Type III secretion system (T3SS)- and 10 Type VI secretion system (T6SS)-related proteins. The reference ATCC® 700383™ had a genome of 5443146 bp with 4954 predicted genes organized as a single circular chromosome without plasmids. The comparative genomic analysis on BCH2017050402 and ATCC® 700383™ found 6727 SNPs between them, which included 2012 non-synonymous mutations in the coding regions and 51 InDels within CDS containing 18 insertions and 33 deletions. There were no large-scale structural rearrangements, but multiple small variations, 4 large insertions, and several deletions. High genomic synteny and structural similarity were observed between BCH2017050402 and other isolates from the fish in Ningde (i.e., FFH2014050403, FDH2015122403, CZH2019040302, FFH2013032002, and PQLYC4). In contrast, an inverted genomic segment containing one translocation and one translocation-inversion region, along with 12 non-synonymous mutations in coding regions, were found different from the isolate XSDHY-P of Zhejiang-origin. However, compared with P. putida NBRC14164, the synteny was low. On 13 P. plecoglossicida genomes, the core accounted for 88.83% of the pan-genome, with 94 strain-specific genes identified.

Conclusion The genomic sequences of the 8 P. plecoglossicida strains isolated from diseased L. crocea were extremely similar. Nonetheless, significant divergences existed between the strains of different origins and closely related species. Using comparative and pan-genomic approaches, this study characterized the genomics of the large yellow croaker-originated P. plecoglossicida and unveiled the structural variations in a background of highly conserved core genome as well as the distinctive profiles of virulence and antimicrobial resistance genes.