Characterization of two novel pyrogenic toxin superantigens made by an acute rheumatic fever clone of Streptococcus pyogenes associated with multiple disease outbreaks

Abstract

The pathogenesis of acute rheumatic fever (ARF) is poorly understood. We identified two contiguous bacteriophage genes, designated speL and speM, encoding novel inferred superantigens in the genome sequence of an ARF strain of serotype M18 group A streptococcus (GAS). speL and speM were located at the same genomic site in 33 serotype M18 isolates, and no nucleotide sequence diversity was observed in the 33 strains analyzed. Furthermore, the genes were absent in 13 non-M18 strains tested. These data indicate a recent acquisition event by a distinct clone of serotype M18 GAS. speL and speM were transcribed in vitro and upregulated in the exponential phase of growth. Purified SpeL and SpeM were pyrogenic and mitogenic for rabbit splenocytes and human peripheral blood mononuclear cells in picogram amounts. SpeL preferentially expanded human T cells expressing T-cell receptors Vbeta1, Vbeta5.1, and Vbeta23, and SpeM had specificity for Vbeta1 and Vbeta23 subsets, indicating that both proteins had superantigen activity. SpeL was lethal in two animal models of streptococcal toxic shock, and SpeM was lethal in one model. Serologic studies indicated that ARF patients were exposed to serotype M18 GAS, SpeL, and SpeM. The data demonstrate that SpeL and SpeM are pyrogenic toxin superantigens and suggest that they may participate in the host-pathogen interactions in some ARF patients.

FIG. 1.

Schematic of the speL-speM locus and PTSAg phylogeny. (A) Open arrows denote the direction of transcription of ORFs encoding SpeL, SpeM, and hypothetical proteins in strain MGAS8232 (33). Lines between ORFs represent intergenic spaces. The annealing positions of primers 117 and 1721R are shown. An asterisk marks the position of a 98-bp direct repeat located at the ends of the prophage-like element. (B) The dendrogram was made with amino acid sequences corresponding to predicted mature forms of GAS and staphylococcal PTSAgs. Bootstrap confidence levels (based on 1,000 repetitions) of >70 are shown. The scale bar represents the number of amino acid substitutions per 100 sites, and summation of the entire branch length predicts the number of amino acid substitutions that have occurred along a given lineage.

FIG. 2.

Sequence alignment and structural analysis of SpeL and SpeM. (A) Amino acid sequences of putative and proven GAS PTSAgs were aligned. Cysteine residues in a Cys-Cys loop (SpeA and SSA) and histidine and aspartate residues in putative or proven zinc-binding motifs are shown in red. Structural elements are shown above the sequences. PTSAgs with known structures are underlined. (B) SpeL and SpeM structural models were predicted based on SpeC. Domains 1 and 2, N and C termini, β-sheets, α-helices, and a zinc ion (green circle) are represented. Red, blue, and cyan represent amino acids aspartate/glutamate, arginine/lysine, and histidine, respectively.

FIG. 3.

In vitro analysis of speL and speM transcript levels. (A) Arrows denote points when bacteria were removed for RNA isolation and transcript analysis. The mean and standard error of the mean (SEM) of triplicate measurements are shown. (B) TaqMan assays were conducted with two independent RNA preparations to assess the relative quantity of gene-specific transcripts. The cDNA quantity of speL, speM, and speB was normalized to gyrA cDNA in each sample. The average fold difference in transcript quantity in the exponential phase relative to the stationary phase and the SEM are shown.

FIG. 4.

Functional analysis of purified rSpeL and rSpeM. (A) rSpeL and rSpeM were analyzed with SDS-PAGE and immunoblots. Lanes 1 to 4 contain 1.0 μg of purified rSpeA, rSpeC, rSpeL, and rSpeM, respectively. Molecular mass markers are shown. Proteins were analyzed with SDS-PAGE, and immunoblots were probed with specific antisera against rSpeA and rSpeC. (B) Rabbit splenocytes and human PBMCs were incubated with rSpeC, rSpeL, and rSpeM and pulsed with [³H]thymidine to assess mitogenicity. DNA was harvested after 24 h, and the counts per minute (cpm) were determined by scintillation counting. The mean of quadruplicate experiments and the SEM values are shown. (C) PBMCs from five human donors were stimulated with anti-CD3 antibody, rSpeL, and rSpeM to determine the TCR Vβ profile. Cells were stained with monoclonal antibodies against TCR Vβ chains and analyzed by flow cytometry. The percentage of T cells expressing the listed Vβ are shown. P values were determined with a Student t test (P < 0.01). Error bars represent the SEM.

FIG. 5.

Human serologic response of patients with ARF. (A) Sera obtained from 55 ARF patients were screened by ELISA against a library of overlapping 15-mer synthetic peptides corresponding to the N-terminal hypervariable region of M18 protein. Frequency of reactivity (the number of serum samples positive for a specific peptide divided by the total sample number) and positive immunoreactivity were calculated as described previously (11). Amino acid sequences of the synthetic peptides are shown. (B) The reactivity of human patient sera to rSpeL and rSpeM was determined by ELISA. Sera were obtained from patients with ARF (n = 55) and pharyngitis (n = 44) and from individuals with no documented case of GAS infection (NHS) (n = 20). The mean ELISA value for triplicate experiments and the SEM values are shown. Statistical differences between ARF and pharyngitis results and ARF and NHS results were determined with a Student t test (P < 0.05).