GBS-SBG - GBS Serotyping by Genome Sequencing

Abstract

Group B Streptococcus (GBS; Streptococcus agalactiae) is the most common cause of neonatal meningitis and a rising cause of sepsis in adults. Recently, it has also been shown to cause foodborne disease. As with many other bacteria, the polysaccharide capsule of GBS is antigenic, enabling its use for strain serotyping. Recent advances in DNA sequencing have made sequence-based typing attractive (as has been implemented for several other bacteria, including Escherichia coli, Klebsiella pneumoniae species complex, Streptococcus pyogenes, and others). For GBS, existing WGS-based serotyping systems do not provide complete coverage of all known GBS serotypes (specifically including subtypes of serotype III), and none are simultaneously compatible with the two most common data types, raw short reads and assembled sequences. Here, we create a serotyping database (GBS-SBG, GBS Serotyping by Genome Sequencing), with associated scripts and running instructions, that can be used to call all currently described GBS serotypes, including subtypes of serotype III, using both direct short-read- and assembly-based typing. We achieved higher concordance using GBS-SBG on a previously reported data set of 790 strains. We further validated GBS-SBG on a new set of 572 strains, achieving 99.8% concordance with PCR-based molecular serotyping using either short-read- or assembly-based typing. The GBS-SBG package is publicly available and will hopefully accelerate and simplify serotyping by sequencing for GBS.

Keywords: Group B Streptococcus serotypes; genome sequencing; molecular epidemiology; short read typing.

Fig. 1.

Genetic organization of the GBS cps locus and serotyping database strategies. (a) Coloured arrows at the top depict the genetic organization of the serotype II sequence. Arrows are drawn to scale, as indicated by the axis at the top. Alignments for the other nine serotype sequences (indicated by Roman numerals on the left) are shown in grey bars below. Vertical red lines indicate single nucleotide polymorphisms relative to the serotype II sequence. Black bars indicate genes that are absent from a given serotype. Orange arrows indicate genes that are unique to a given serotype, and dotted lines indicate where they fall relative to the serotype II reference. All sequences used for this figure panel are taken from Kapatai et al. [54]. The black lines demarcated by diamonds at each end at the bottom of the panel indicate the region of the cps locus spanned by the serotype sequences in the Sheppard database, Kapatai database, and GBS-SBG databases, as indicated to the right of each line. (b) Coloured arrows depict the genetic organization of the serotype III sequence from Kapatai et al. 2017 [54], using the same scale bar as for (a). Dotted lines between the panels show the relative alignment of the serotype II and serotype III reference sequences. Grey bars and vertical red lines indicate alignment and SNPs for the serotype III subtypes (indicated on the left). The yellow box indicates the region in which SNPs reported by Kong et al. [23] are located.

Fig. 2.

Phylogenetic relationship of serotype reference sequences in the GBS-SBG database. A maximum likelihood tree is shown; a total of 13 serotype sequences were used to plot an unrooted binary tree. The scale bar for branch length at the bottom indicates the number of substitutions per site. There were a total of 11911 positions in the alignment. The labels for serotypes Ia and subtypes of serotype III are highlighted in blue for ease of visualization.