Comparative genomics analysis of Streptococcus agalactiae reveals that isolates from cultured tilapia in China are closely related to the human strain A909

Abstract

Background: Streptococcus agalactiae, also referred to as Group B Streptococcus (GBS), is a frequent resident of the rectovaginal tract in humans, and a major cause of neonatal infection. In addition, S. agalactiae is a known fish pathogen, which compromises food safety and represents a zoonotic hazard. The complete genome sequence of the piscine S. agalactiae isolate GD201008-001 was compared with 14 other piscine, human and bovine strains to explore their virulence determinants, evolutionary relationships and the genetic basis of host tropism in S. agalactiae.

Results: The pan-genome of S. agalactiae is open and its size increases with the addition of newly sequenced genomes. The core genes shared by all isolates account for 50 ~ 70% of any single genome. The Chinese piscine isolates GD201008-001 and ZQ0910 are phylogenetically distinct from the Latin American piscine isolates SA20-06 and STIR-CD-17, but are closely related to the human strain A909, in the context of the clustered regularly interspaced short palindromic repeats (CRISPRs), prophage, virulence-associated genes and phylogenetic relationships. We identified a unique 10 kb gene locus in Chinese piscine strains.

Conclusions: Isolates from cultured tilapia in China have a close genomic relationship with the human strain A909. Our findings provide insight into the pathogenesis and host-associated genome content of piscine S. agalactiae isolated in China.

Figure 1

The distribution of accessory gene number and genes functional classification of 15 Streptococcus agalactiae strains. (a) Number of conserved accessory genes in each strain. All noncore genes in each strain were classified into different levels of conservation according to the number of strains. Different conservations are presented by various colors. (b) Comparison of COG functional categories between core and accessory genes. The Y-axis indicates the percentage of genes in a particular functional category relative to the genes of all COG categories. Only assigned COG functional genes were taken into account. (c) COG functional categories of 122 core genes. From the analysis of the core genome, 122 core genes have been deleted or suffered loss of function in the ST260-553 fish strains. The Y-axis indicates the percentage of genes in a particular functional category relative to the genes of all COG categories. Unclassified COG functional genes were taken into account.

Figure 2

Diversity of the CRISPR1 locus in 13 Streptococcus agalactiae strains. Spacers were identified by the CRISPRtionary program, attributing a number to each spacer [19]. Repeats are not shown and spacers are represented in arbitrarily colored boxes. Strains with similar spacer arrays are arranged together and their names are given on the left. Single spacers appear in white boxes framed in black; identical spacers shared by different strains are shown with a same color background with black numbers; homologous spacers in the same strain are represented using the same color background and white numbers. The special spacers are underlined in red.

Figure 3

Comparison of the CDs of the Prophage derived from Streptococcus agalactiae GD201008-001 (a), A909 (b) and ZQ0910 (c). The detailed prophage views of three strains were produced using the online software PHAST (http://phast.wishartlab.com/index.html). Different colors represent various phage elements. The putative internal attL site and attR site of A909 prophage are labeled in white boxes framed in red.

Figure 4

Core and pan-genome calculations for 15 S. agalactiae strains. (A) S. agalactiae core genome. Each point represents the number of conserved genes between genomes. They are plotted as functions of strain number (x). For each x, circles are the 15!/[(x-1)!(15-x)] values from the different strain combinations. Squares are the averages of such values. The blue line represents the least-squares fit of the function C(x) = Ac x ^–tc + yc. The best fit was obtained with correlation r² = 0.960 for Ac = 1021士80, tc = 0.28, yc = 1140士8. (B) S. agalactiae pan-genome. Numbers of genes are calculated for all possible combinations and plotted as a function of strain numbers (x). The red line demonstrates the exponential model based on the mean value of pan genes. The deduced pan-genome size P(x) = As*x^(ts) + ys. The best fit was obtained with correlation r² = 0.999 for As = 726士2, ts = 0.562, ys = 1284士7. (C) Number of new predicted gene clusters identified by the sequencing additional genomes. The curve is fitted to the function S(x) = As*x^(ts)-As*(x-1)^(ts).

Figure 5

Phylogenetic tree showing the relationship among 15 S. agalactiae strains based on 1114 orthologous gene clusters. A neighbor-Joining (NJ) phylogenetic tree based on the 1,114 single copy orthology clusters of the 15 strains.

Figure 6

Whole genome alignment between S. agalactiae GD201008-001 and A909 and Chinese piscine isolates specific island. The genomes of GD201008-001 and A909 were compared with each other using progressive MAUVE with default parameters, and the colinearity of the genomes is shown. Human strain A909 has two additional prophage sequences (prophage LambdaSa 03 (548,964 bp-585,049 bp), prophage LambdaSa 05 (2,069,932 bp-2,086,164 bp)), PI-1 type pilus(711,448 bp -720,303 bp) ) and lmb-scpB locus (1,293,687 bp -1,303,948 bp), but lacked the Chinese piscine S. agalactiae specific genes island (1,992,339 bp -2,002,579 bp in GD201008-001), which only matched the region of Streptococcus anginosus SK52 = DSM 20563 with high level amino acid identity (78-98%).