Population Genomics of emm4 Group A Streptococcus Reveals Progressive Replacement with a Hypervirulent Clone in North America

Abstract

Clonal replacement is a major driver for changes in bacterial disease epidemiology. Recently, it has been proposed that episodic emergence of novel, hypervirulent clones of group A Streptococcus (GAS) results from acquisition of a 36-kb DNA region leading to increased expression of the cytotoxins Nga (NADase) and SLO (streptolysin O). We previously described a gene fusion event involving the gene encoding the GAS M protein (emm) and an adjacent M-like protein (enn) in the emm4 GAS population, a GAS emm type that lacks the hyaluronic acid capsule. Using whole-genome sequencing of a temporally and geographically diverse set of 1,126 isolates, we discovered that the North American emm4 GAS population has undergone clonal replacement with emergent GAS strains completely replacing historical isolates by 2017. Emergent emm4 GAS strains contained a handful of small genetic variations, including the emm-enn gene fusion, and showed a marked in vitro growth defect compared to historical strains. In contrast to other previously described GAS clonal replacement events, emergent emm4 GAS strains were not defined by acquisition of exogenous DNA and had no significant increase in transcript levels of nga and slo toxin genes via RNA sequencing and quantitative real-time PCR analysis relative to historic strains. Despite the in vitro growth differences, emergent emm4 GAS strains were hypervirulent in mice and ex vivo growth in human blood compared to historical strains. Thus, these data detail the emergence and dissemination of a hypervirulent acapsular GAS clone defined by small, endogenous genetic variation, thereby defining a novel model for GAS strain replacement. IMPORTANCE Severe invasive infections caused by group A Streptococcus (GAS) result in substantial morbidity and mortality in children and adults worldwide. Previously, GAS clonal strain replacement has been attributed to acquisition of exogenous DNA leading to novel virulence gene acquisition or increased virulence gene expression. Our study of type emm4 GAS identified emergence of a hypervirulent GAS clade defined by variation in endogenous DNA content and lacking augmented toxin gene expression relative to replaced strains. These findings expand our understanding of the molecular mechanisms underlying bacterial clonal emergence.

Keywords: M type; evolution; exotoxin; genomics; group A Streptococcus; population genetics; virulence.

FIG 1

Maximum likelihood phylogenetic reconstruction of 1,126 emm4 GAS disease isolates from North America and the United Kingdom. (A) Unrooted neighbor-joining phylogenetic tree based on 12,436 biallelic SNPs relative to the emm4 reference strain Duke (GenBank accession number CP031770) following correction for recombination (ClonalFrameML). Branch tips colored to indicate individual emm4 strain year of isolation and shape for subclade designation (BAPS). The dashed circle indicates single-locus variant (ST38) strains from the main (ST39) emm4 population. (B) Progressive replacement of historical (SC1) strains by emergent (SC3) over the 22-year period. The total number of isolates and geographic origin are indicated below bar graph. The red value for 2014 indicates that 32 isolates from the United Kingdom were included in the analysis.

FIG 2

Distribution of DNases, exotoxins, and macrolide-lincosamide-streptogramin (MLS) resistance within the emm4 population. A rooted maximum likelihood (ML) phylogram of 1,126 emm4 GAS disease strains is shown. Branch tip color and shape indicate year and subclade designation as in Fig. 1 and the associated legend. Subclades (SC1 to 3) are indicated by shaded boxes. DNase, exotoxin, and antimicrobial resistance gene content is indicated by vertical bars to the right of the ML phylogram and defined in the legend headings. Asterisks indicate nodes representing strains with completed genomes (see Table S3 in the supplemental material).

FIG 3

Highly polymorphic genetic loci identified within the emm4 GAS population after exclusion of recombinant loci. Genes of known or predicted regulators (red) are indicated.

FIG 4

Emergent emm4 GAS strains have altered in vitro growth and no increased SLO toxin production compared to historical strains. (A) Growth curves comparing historical (blue squares) to emergent (red circles) clade emm4 GAS strains in nutrient-rich (THY; solid) or glucose-deficient (C medium; open) conditions. Error bars represent the standard deviations of four strains grown in biological triplicate. Dashed lines indicate mid-exponential (ME) phase of growth. (B) Principal-component analysis of RNA-seq comparing emergent (red) and historical (blue) clade strains. Individual strains with replicate samples are labeled and indicated using dashed circles. PC 1, principal component 1. (C) Log₂ fold change in transcript levels derived from RNA-seq for selected genes derived from RNA-seq comparing four historical and four emergent emm4 GAS strains. Error bars represent standard deviations. (D) Transcript levels (qRT-PCR) of slo from emergent or historical GAS cells grown in nutrient-rich media and harvested at ME as defined in panel A. The asterisk indicates a P value of <0.005 (unpaired t test). (E) Western blot analysis using anti-Nga antibody (α-Nga) of culture supernatants (SN) from historical (blue text) or emergent (red text) emm4 GAS strains. Similar amounts of protein are demonstrated using anti-HPr. ctrl, control. (F) Transcript levels derived from qRT-PCR of slo from representative strains of emm1 (MGAS2221), emm3 (MGAS10870), emm4, emm28 (strain TSPY902), emm87 (strain TSPY1057), and emm89 (strain MSPY1). For emm4 GAS, transcript levels reflect four historical strains (in blue) and four emergent strains (in red). RNA was extracted from ME cultures as described in Methods. TaqMan qRT-PCR data for both panels D and F are means ± standard deviations of two biological replicates, with two technical replicates, done on two separate days. pos., positive; neg., negative.

FIG 5

In vivo mouse and ex vivo human blood virulence assays comparing emergent and historical clade emm4 GAS strains. (A) Cumulative Kaplan-Meier survival curves of mice (n = 7 per strains) infected with four historical (blue) or four emergent (red) emm4 GAS strains (2 × 10⁸ CFU). The P value of  <0.0001 was determined by the log rank test. Survival curves following infection with other dose ranges are shown in Fig. S4 in the supplemental material. (B) Survival of emergent (red) or historical (blue) strains following exposure to human blood ex vivo. Error bars represent standard deviations of four emergent and four historical emm4 GAS strains assayed in quadruplicate using three independent donors. The two asterisks indicate P < 0.05 (Mann-Whitney U test). Individual strain comparisons are shown in Fig. S4.