Analysis of Streptococcus dysgalactiae subspecies equisimilis gene transcripts during experimental primate necrotizing myositis

Abstract

Streptococcus dysgalactiae subspecies equisimilis (SDSE) is a gram-positive bacterial pathogen capable of causing various infections in humans. Recently, isolates of SDSE emm type stG62647 have emerged as a cause of severe invasive infections, including necrotizing myositis. However, the molecular processes underlying these infections remain poorly understood. To address this gap, we performed RNAseq analysis to examine SDSE gene transcript levels during experimental necrotizing myositis infection in non-human primates, animals phylogenetically closely related to humans. We analyzed the transcriptomes of two related SDSE stG62647 human isolates (MGCS36044 and MGCS36089) during necrotizing myositis infection in six non-human primates. The transcriptome from in vitro growth in nutrient-rich media differed considerably from that of SDSE bacteria grown in vivo in experimental necrotizing myositis, with 254 genes differentially expressed, indicating extensive genetic adaptation in infected skeletal muscle. Notably, we observed a marked upregulation of ihk-irr genes encoding a two-component regulatory system that promotes evasion of phagocytosis and resistance to killing by human polymorphonuclear leukocytes in Streptococcus pyogenes. Similarly, genes comprising the sag operon encoding the streptolysin S cytolytic toxin virulence factor demonstrated very high transcript abundance in vivo. Additionally, we present evidence that a 40-nt deletion in fasB alters expression of ska, encoding streptokinase. Collectively, our data provide new insights into the SDSE genes transcribed in vivo, thereby enhancing our understanding of the molecular basis of pathogen and primate host interactions. The SDSE genes identified in this study offer promising targets for future studies on molecular pathogenesis and therapeutic interventions.IMPORTANCEStreptococcus dysgalactiae subspecies equisimilis (SDSE) has emerged as an increasingly important bacterial pathogen causing serious invasive infections in humans worldwide. Despite its clinical importance, the mechanisms through which SDSE causes infections remain poorly understood, and no licensed vaccine currently exists. SDSE can cause necrotizing myositis, an infection with high morbidity and mortality. We used a primate infection model and bacterial transcriptome analysis to gain new understanding of the molecular events contributing to SDSE pathogenesis in necrotizing myositis. Our results provide extensive new information about the transcriptome of SDSE in vivo and reveal numerous potential targets for future therapeutic and vaccine research.

Keywords: RNAseq; Streptococcus dysgalactiae; emerging clone; pathogenesis; primate.

Fig 1

Principal component analysis of SDSE gene expression during in vitro and in vivo growth. Strains were grown in vitro in quadruplicate to mid-exponential (ME) and early stationary (ES) phases and in vivo as single replicates in skeletal muscle tissue from two non-human primates (NHPs), NHPs 15 (MGCS36044) and 14 (MGCS36089). These two NHPs were selected for analysis based on the highest number of SDSE genes with mapped reads. Top panel: growth in vitro. Thirteen genes were excluded due to low expression. (A) MGCS36044 (1,973 genes) grown at ME (blue) and ES (red). (B) MGCS36089 (1,984 genes) grown at ME (blue) and ES (red). For both isolates, ME replicates clustered together and separated from the ES replicates along principal component 1 (PC1) (C and D). MGCS36044 (blue) and MGCS36089 (red) grown at ME (C) and ES (D). A total of 1,954 genes were shared between both isolates. Bottom panel: in vivo vs. in vitro growth. A total of 1,662 genes (MGCS36044) and 1,630 (MGCS36089) had mapped reads in vivo. (E and F) MGCS36044 grown in vivo (green) and in vitro (blue) at ME and ES. (G and H) MGCS36089 grown in vivo (green) and in vitro (red) at ME and ES. There was a clear separation between replicates grown in vivo and in vitro along PC1.

Fig 2

Abundance and genome coverage of the MGCS36044 and MGCS36089 genes during in vivo infections in six NHPs. Individual NHP identifiers are shown below each graph. Two NHPs were excluded from analysis due to an insufficient number of reads mapping to each strain. (A) The number of MGCS36044 genes (left panel) and MGCS36089 (right panel) with mapped reads are shown for the six NHPs. (B) Percent genome coverage was calculated for 1,986 (MGCS36044) and 1,997 (MGCS36089) genes, excluding those encoding rRNAs, tRNAs, and disrupted genes.

Fig 3

Pairwise comparisons of MGCS36044 and MGCS36089 transcriptomes from biopsies. Transcript abundance ranks were calculated based on normalized counts. The obtained R² values are shown. Black lines indicate trend lines for each comparison. (A) SDSE strain comparison. Biopsies from NHP 15 (infected with MGCS36044) were compared to biopsies from NHP 14 (infected with MGCS36089). These two NHPs had yielded the highest number of SDSE genes with mapped reads, and the results correspond to 1,535 genes shared by both strains. (B) Comparison between biopsies from two NHPs infected with the same strain. Biopsies from NHPs 15 and 13 yielded the highest and second highest numbers of MGCS36044 genes with mapped reads, respectively. The results correspond to 1,097 SDSE shared genes. (C) Strain-specific R² values corresponding to all samples. For each strain, the NHP sample with the largest number of genes with mapped reads [NHP 15 for MGCS36044 and NHP 14 for MGCS36089] was compared to all other samples. The number of shared genes and the corresponding R² values are indicated. The comparisons involving the highest number of shared genes (*) yielded the highest R² values.

Fig 4

Expression of SDSE putative virulence genes during in vivo infection. Only data from NHPs with the highest number of genes with mapped reads are shown. (A) MGCS36044 in NHP 15. Fifty-four putative virulence genes out of a total of 1,662 genes had mapped reads. (B) MGCS36089 in NHP 14. Fifty-two putative virulence genes out of a total of 1,631 had mapped reads.

Fig 5

Differentially expressed MGCS36044 genes during in vivo growth in NHPs compared to in vitro growth. A total of 537 and 370 MGCS36044 genes from NHP 15 were differentially expressed in vivo relative to in vitro at ME (blue) and ES (green), respectively. A total of 254 genes (orange) were common to both categories. The fold change value cutoff and adjusted P-value cutoff were 2 and ≤0.05, respectively.

Fig 6

Organization of the fasBCAX operon and associated mutations. (A) Genes and corresponding gene products in the atypical fasBCA two-component system, which encodes two histidine kinases, in strains MGCS36044 and MGCS36089. In both strains, the operon is oriented toward the chromosomal origin of replication (ori). From left to right: (i) fasX, fas operon regulatory RNA; (ii) fasA, response regulator; (iii) fasC, histidine kinase; and (iv) fasB, histidine kinase. fasB from MGCS36089 contains a 40 nt internal deletion (from nt. 899 to 938, inclusive), which results in a frameshift at codon 300 of 447 total codons. (B) Summary of mutations in fasB or fasC corresponding to ska transcript abundance (in RPKMs) and virulence in a mouse model of necrotizing myositis (59). Data are shown for four strains lacking mutations in vir and MTR genes and three strains with mutations in either fasB or fasC. WT, wild-type allele.