A Rapid Molecular Detection Tool for Toxigenic M1_UK  Streptococcus pyogenes

Abstract

Following the implementation of a range of nonpharmaceutical intervention strategies in response to the coronavirus disease 2019 (COVID-19) pandemic in 2020, transmission and incidence of various invasive bacterial diseases declined in several countries, including invasive Streptococcus pyogenes infections [1–3]. However, with the waning of the COVID-19 pandemic and easing of these interventions by the end of 2022, health authorities from the United States (US) and several European countries reported an unprecedented surge in serious S pyogenes infections disproportionately among children [4, 5], with case numbers several-fold higher relative to pre–severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic levels [6–16]. Streptococcus pyogenes severe and invasive infections have high attendant morbidity and mortality rates and require prompt antibiotic treatment. Although the extent to which altered pathogen virulence, host susceptibility, and/or environmental factors have contributed to this surge in S pyogenes infections is not known, a coincidence with an increase in the incidence of common respiratory viral infections, such as respiratory syncytial virus (RSV) and influenza, and an increase in viral coinfection has been found in some investigations [7, 12, 15, 17, 18]. Unfortunately, it is difficult to assess the true burden of this outbreak since not all S pyogenes infections are notifiable diseases worldwide.

Streptococcus pyogenes is a strictly human-adapted pathogen and causative agent of a broad spectrum of diseases accounting for hundreds of millions of infections each year. It is among the leading causes of infectious disease–related deaths worldwide [19], estimated to be responsible for 500 000 fatalities annually [20]. To date, there are no licensed vaccines for prevention of S pyogenes infection, and the global burden of S pyogenes diseases is widely recognized as a major public health concern [21]. Streptococcus pyogenes is classified into >200 emm types [22], but the modern-day S pyogenes emm1 genotype M1_global is disproportionately represented in invasive infections in economically developed countries, causing an increase in the incidence and severity of infections since the mid-1980s [23–27]. Comprehensive population phylogenomic analysis is consistent with the acquisition of a streptococcal pyrogenic exotoxin A (SpeA) encoding phage being one of the key evolutionary events that preceded the emergence of and likely contributed to the fitness and epidemiological success of the M1_global genotype [28]. After 3 decades of clinical dominance in high-income settings, a new emm1 genotype entered the S pyogenes emm1 population in 2008 and has since been expanding both geographically and in proportion of S pyogenes emm1 infections, steadily replacing the “ancestral” M1_global genotype in multiple countries [6, 29–34]. This newly emergent emm1 clade (designated M1_UK) is defined by 27 single-nucleotide polymorphisms (SNPs) and is characterized by significantly increased SpeA expression levels relative to M1_global in vitro [29, 33, 35]. Although all seasonal surges in S pyogenes infections are polyclonal in nature, being caused by strains of multiple emm types, recent epidemiological investigations from multiple countries have shown M1_UK strains to be overrepresented in scarlet fever and invasive infections [6, 14, 29, 34]. The inherit virulence and documented epidemic capacity of the M1_global genotype warrants close surveillance of the spread and incidence rate of the M1_UK variant as a cause of increased S pyogenes infection rate.

Molecular tests currently available for the detection of S pyogenes from throat swab specimens include rapid antigen detection tests and nucleic acid amplification tests, but the gold standard diagnostic test remains bacterial culture [36]. However, while of clinical value, further characterization, such as emm typing and whole genome sequencing (WGS), is required for active surveillance. With the emergence of both multidrug-resistant strains and new toxigenic group A Streptococcus variants that may increase transmissibility or virulence [21], there is an urgent need for new rapid and accurate detection methods, particularly for prospective longitudinal surveillance. Real-time polymerase chain reaction (PCR) is now routinely used for the detection and typing of various microbial pathogens and proved instrumental in the management of the SARS-CoV-2 outbreak [37]. Compared with the current sequencing approaches used for S pyogenes surveillance, a rapid, affordable, and easily deployable molecular assay for the detection of important strains, such as M1_UK, has substantial utility. An allele-specific PCR method has recently been developed to detect M1_UK-specific SNPs and help support surveillance efforts of this toxigenic clone in the United Kingdom (UK) [38].

We recently reported that a single SNP (+5 G > C) in the 5′ transcriptional leader sequence of the transfer-messenger RNA gene ssrA is driving increased SpeA expression in M1_UK [33]. Herein, an allele-specific real-time PCR assay targeting this ssrA SNP was developed for the detection of M1_UK strains. The assay was evaluated using a panel of 51 clinical isolates comprising 15 different emm types (Table 1) and showed very high specificity and sensitivity. The method presented here establishes the M1_UK-specific ssrA SNP as a suitable target for efficient and rapid identification of M1_UK isolates and presents a powerful molecular tool for global surveillance of M1_UK.

MATERIALS AND METHODS

Bacterial Strains and Growth Conditions

All S pyogenes strains were routinely grown at 37°C on 5% horse blood agar overnight, then inoculated into Todd-Hewitt broth supplemented with 1% yeast extract and grown statically to the desired growth phase. The 51 S pyogenes clinical isolates used in this study were obtained from Pathology Queensland (Australia), the statewide public pathology service, and collected between 2016 and 2021. Nucleic acid extracts from an additional panel of non–S pyogenes bacterial and fungal species (n = 20, Table 2) were tested for specificity of the multiplex M1_UK real-time PCR assay.

Multiplex M1_UK Real-time PCR Assay Design

The allele-specific oligonucleotide probes used in this study to identify M1_UK  S pyogenes strains target the +5 G > C SNP in the 5′ leader sequence of the transfer-messenger RNA gene ssrA (Supplementary Methods Table) [33]. Locked nucleic acids were included in the design to increase mismatch discrimination of this SNP. While the sequence targeted by the allele-specific probes is unique to S pyogenes, we sought to include an additional target to confirm the presence of S pyogenes through detection of the conserved S pyogenes–specific gene speB (GenBank accession number L26125) [39]. The ssrA and speB primers and probes were designed using available sequence data from GenBank, along with M1_UK WGS data [33]. DNA was extracted for proof-of-concept testing of the multiplex M1_UK real-time PCR assay (Supplementary Methods). For blinded testing of isolates from Pathology Queensland, isolates were cultured on 5% horse blood agar overnight at 37°C and microwave-prepared as previously described [40]. The reaction mixture of duplex ssrA-speB assay consisted of 10 μL SensiFAST Probe No-ROX PCR Master Mix, 0.5 μM each of ssrA and speB forward and reverse primers, 0.5 μM each of ssrA wild-type and ssrA M1_UK probes, 0.25 μM speB probe, and 3 μL of template, made up to a final reaction volume of 20 μL using DNase-free water. Amplification and detection were performed on the ABI 7500 Fast Real-Time PCR System, using 2-step cycling conditions as follows: 95°C for 5 minutes, followed by 40 cycles of 95°C for 10 seconds and 60°C for 30 seconds. Acquisition was on FAM (non-M1_UK  ssrA), HEX (speB), and Cy5 (M1_UK  ssrA) channels at the 60°C stage.

Illumina Genome Sequencing

Whole genome sequencing of 51 isolates from pharyngeal swabs from Pathology Queensland (Australia) was performed by the Australian Genome Research Facility (Brisbane, Australia) on an Illumina NovaSeq with 150 bp paired-end reads. Details of genome assembly and analyses are provided in the Supplementary Methods.

Virulence Gene Screen

The virulence genes speA, spd1, speC, and ssa were PCR amplified from genomic DNA of S pyogenes emm1 strains using either MangoTaq DNA polymerase (Meridian Bioscience) or the KAPA HiFi PCR kit (Roche) according to the manufacturer's instructions, with primers listed in the Supplementary Methods Table. PCR products were purified using the Wizard SV Gel and PCR Clean-Up System (Promega), followed by 1%–2% agarose gel electrophoresis. DNA bands were visualized using the GelDoc XR+ Imaging System (Bio-Rad).

Quantitative Real-time PCR and Western Blot Analysis

Gene expression was examined using standard real-time PCR, and protein abundances were determined by Western blot analysis as described in the Supplementary Methods.

Ethics and Informed Consent

This study was approved by the Children's Health Queensland Hospital and Health Service Human Research Ethics Committee (HREC/22/QCHQ/85249) and was ratified by The University of Queensland Human Research Ethics Committee. Informed consent was not needed for the use of the clinical isolates in this study, as these were obtained during routine pathology services. No identifying patient information can be provided for these isolates.

Statistical Analysis

All statistical analysis was completed using Prism software (GraphPad; version 10.0.1). Significance was calculated using 1-way analysis of variance with Dunnett multiple comparisons post hoc test. A P value <.05 was considered to be statistically significant. Confidence intervals (CIs) for assay sensitivity and specificity were calculated using the Wilson method.

RESULTS

Allele-Specific Real-time PCR

Our recent finding that a single genetic mutation in the 5′ transcriptional leader sequence of ssrA is required for increased SpeA expression [33], a defining feature of M1_UK [29], prompted us to develop a multiplexed real-time PCR screening assay targeting the ssrA SNP (+5 G > C) for genetic distinction between M1_UK and other emm types. While the sequence targeted by ssrA-specific primers and probes is highly conserved and almost exclusive to S pyogenes (determined by BLASTn analysis, data not shown), we included an additional target to detect the speB gene as a second species-specific control. As a proof-of-concept, we first tested our multiplex M1_UK real-time PCR assay on a set of previously characterized M1_global (5448, HKU425, HKU488; all non-M1_UK) and M1_UK (SP1380, SP1448, and SP1449) isolates [33, 41, 42]. Using genomic DNA as a template, the assay accurately detected all M1_UK isolates (Table 2). To further evaluate the ability of our real-time PCR assay to identify M1_UK isolates, we performed blinded testing on 51 clinical isolates collected in Queensland, Australia, between 2016 and 2021 (Table 1). This time, however, a simple and time saving microwave-based approach was used to extract DNA from these isolates as a template for the assay to test utility for routine clinical practice [40]. Of the 51 isolates, 7 isolates were positive for the ssrA M1_UK probe (Table 2). To validate these findings, we performed WGS on the 51 isolates, which comprised 15 different emm types, with emm1 (11/51), emm12 (10/51), and emm3 (7/51) being the most common emm types (Supplementary Results Table 1). However, the proportion of emm1 was higher among invasive isolates compared to noninvasive isolates (5/9 vs 6/42). Of the 11 emm1 isolates, 7 were M1_UK (SP1511, SP1512, SP1522, SP1527, SP1532, SP1534, and SP1544), 3 were M1_global (SP1516, SP1525, and SP1550), and a single isolate was a less common intermediate M1_13SNPs strain lacking the ssrA SNP (SP1538) [29, 33, 35]. The prevalence of M1_UK in invasive S pyogenes infections was higher than in noninvasive infections (4/5 vs 3/6), which were all recovered from patients with scarlet fever (Table 1). To further confirm specificity of the assay, an additional panel of non–S pyogenes bacterial and fungal species (n = 20) isolate nucleic acid extracts was screened for potential cross-reactivity. As expected, no amplification was observed for any of these species (Table 2). Both sensitivity and specificity of the multiplex assay were 100% (sensitivity: 95% CI, 72.2%–100%; specificity: 95% CI, 94.6%–100%).

We also performed limit-of-detection studies to evaluate the analytical sensitivity of our assay using nucleic acid extracts from representative strains of the non-M1_UK (M1_global, 5448) [42] and M1_UK (SP1448) genotypes [33]. Following droplet digital PCR quantification, limit-of-detection testing from 100 – 1 target copies/reaction showed a lower limit of detection of 5 target copies/reaction for all 3 targets in the multiplex M1_UK real-time PCR assay.

Phylogeny

To investigate the evolutionary relationship between the 11 emm1 S pyogenes isolates in this clinical isolate panel, we performed a phylogenetic analysis of the genome sequences using a previous genomic framework of 737 global emm1 strains representative of the circulating emm1 population [33]. The 11 emm1 S pyogenes isolates were scattered throughout the phylogeny, with the 7 M1_UK genotypes clustering within 3 distinct sublineages, indicating persistence of multiple M1_UK sublineages in Queensland (Figure 1). The single M1_13SNPs isolate (SP1538) was related to the previously defined intermediate emm1 isolates (M1_inter), indicating heterogeneity in the emm1 population regarding invasive and noninvasive presentations. Moreover, examination of the accessory genome content revealed that 4 of the 7 M1_UK strains (SP1527, SP1532, SP1534, and SP1544) and the intermediate M1_13SNPs isolate SP1538 harbored bacteriophage-encoded toxin genes encoding superantigens SSA and SpeC, and DNase Spd1 (Figure 1). Prevalence of these toxin genes has been repeatedly found in contemporary S pyogenes strains causing scarlet fever [41, 43–45]; however, this extended toxin profile has thus far only been detected in Australian M1_UK strains [33]. All toxin gene profiles were confirmed by PCR (Figure 2A).

Figure 1.

Phylogenetic analysis of 748 global emm1 Streptococcus pyogenes. Maximum-likelihood phylogenetic tree derived from 3482 single-nucleotide polymorphism sites from a 1 626 524 bp core genome alignment relative to the 5448 M1_global reference genome. The position of the 11 emm1 isolates sequenced in this study are indicated by black tips and annotated by strain name (colored by genotype: M1_global [blue], M1_inter [purple], and M1_UK [red]). A global database of 737 emm1 isolates was used for context [23] with comparative reference isolates referred to in this study are annotated in black. Continent of isolation and clinical sample type are colored as shown in the legend. Carriage of key virulence and antimicrobial resistance determinants are indicated as per legend with probable referring to fragmented gene assemblies preventing accurate in silico prediction. Abbreviation: iGAS, invasive Group A Streptococcus.

Open in new tabDownload slide

Figure 2.

Maintenance of M1_UK sublineages with an extended toxin repertoire and detection of CovR A¹¹¹V mutants in M1_UK. A, Carriage of scarlet fever–associated toxins in 11 emm1 pharyngeal isolates was confirmed by polymerase chain reaction (PCR). B, Quantitative real-time PCR of indicated virulence genes in M1_UK (SP1380, SP1511, and SP1512) versus M1_global (5448) genotypes. Data from 3 biological replicates are presented as mean values ± standard error of the mean. Statistical significance was assessed using 1-way analysis of variance with Dunnett multiple comparisons post hoc test against the 5448 control group (****P < .0001, ***P < .001, **P < .01, *P < .05; ns, not significant). C, Western blot analysis of SpeA, PepO, SpeB (pro- and mature forms), and SLO in culture supernatants and SpyCEP in cell wall extracts from indicated emm1 strains grown to late-logarithmic phase of growth. D, Western blot analysis of SpeB (pro- and mature forms) in overnight culture supernatants. The Australian M1_UK isolate SP1380 was used as a control [33].

Open in new tabDownload slide

Two of the 4 invasive M1_UK isolates (SP1511 and SP1512) displayed a large mucoid colony morphology indicative of increased hyaluronic acid capsule expression, a phenotype frequently observed in hypervirulent CovR/CovS-inactivated clinical isolates [46]. Sequence analysis of covR/S confirmed the presence of a mutation in covR that changes amino acid 111 from alanine to valine in the phylogenetically related isolates SP1511 and SP1522. This CovR A¹¹¹V substitution was previously reported in a lethal invasive emm81 isolate and shown to alter phosphorylation-dependent binding of CovR to target promoter regions, including the capsule (hasA) and S pyogenes cell envelope protease (SpyCEP, scpC) promoters [47]. Increased expression of the CovR/S-controlled endopeptidase PepO can be used as a bacterial marker for the identification of isolates with defects in the CovR/CovS 2-component regulatory system [48], which is responsible for the “characteristic” downregulation of the broad-spectrum cysteine protease SpeB in covR/S mutants [49]. To characterize the CovR A¹¹¹V mutation in M1_UK, we first compared gene expression of the marker genes speA, pepO, speB, hasA, and scpC during late-exponential growth phase (optical density at 600 nm of approximately 0.8) in SP1511 and SP1512 against 2 representative M1_global strains, 5448 and HKU488, using standard real-time PCR (Figure 2B). As expected, speA gene expression was significantly upregulated in both M1_UK isolates. Other upregulated genes included pepO, hasA, and scpC, confirming previous observations in emm81 [47]. In addition, expression of the pore-forming toxin gene slo was strongly increased in SP1511 and SP1512. No definitive role for the CovR A¹¹¹V mutation in speB gene regulation could be established, but real-time PCR data showed a trend toward lower expression levels in SP1511 and SP1512. Findings were verified by Western blot analysis using all 11 emm1 S pyogenes isolates from this study, which also confirmed increased SpeA production in the 7 M1_UK strains (Figure 2C). Interestingly, while the 40 kDa SpeB zymogen form (pro-SpeB) showed comparable abundance in all isolates, mature SpeB (mSpeB) was absent in both covR mutant strains, even after 24 hours of incubation (Figure 2D). The biological significance of autocatalytic processing of pro-SpeB is still unclear, but it may affect substrate specificities and disease-relevant phenotypes.

DISCUSSION

Herein, we present a rapid detection tool for M1_UK  S pyogenes, confirming the maintenance of M1_UK strains in Queensland, Australia. This tool has significant utility for surveillance of this toxigenic variant, given the dramatic and synchronous rise in the incidence of severe S pyogenes infections reported in several countries since the end of 2022. The factors contributing to this surge remain incompletely understood; however, recent epidemiological studies suggest a direct link between the expansion of the M1_UK clone and increased invasive S pyogenes activity [12, 14, 15, 34]. High circulation of respiratory viruses, such as RSV and influenza, following the relaxation of public health interventions designed to limit transmission of SARS-CoV-2 may have further increased the health risk of streptococcal coinfection [7, 12, 17, 18].

The S pyogenes emm1 serotype is a prevalent cause of pharyngitis and severe invasive infections since the emergence of the speA-positive M1_global genotype in the mid-1980s. In recent years, the M1_UK clone has rapidly replaced M1_global and has now become the dominant emm1 type in several countries [29, 30, 33]. The apparent fitness advantage(s) of M1_UK is a serious global health concern and underscores the need to enhance surveillance efforts by public health entities around the world. WGS has become an essential tool for epidemiological surveillance of infectious diseases and pathogens, but resource and capacity constrain its use in routine clinical practice. Detection of M1_UK is therefore currently restricted to those countries undertaking and reporting WGS. To help address this issue, an allele-specific PCR method has recently been developed to detect M1_UK-specific SNPs in the rofA, gldA, and pstB genes [38]. However, while this method can distinguish M1_UK from other emm1 strains and less common intermediate sublineages, it may be labor-intensive, requiring genomic DNA extraction and purification and agarose gel electrophoresis to visualize amplified products which, depending on band intensities, could be subjective. It is worth noting that although enhanced SpeA expression serves as a significant indicator of M1_UK virulence, it cannot alone account for the success of this lineage since rare M1_23SNPs intermediate strains and other emm types generate comparable or greater levels of SpeA. A potential shift in metabolism has been suggested to also play a role in the wider success of M1_UK [35].

The molecular tool presented in this study provides an improved detection method by targeting the SNP in ssrA of M1_UK in a single reaction, detected rapidly by real-time PCR. This study highlights the feasibility of such an approach for the direct detection of M1_UK, via assay validation using a genetically well-characterized set of clinical isolates. The ssrA primer target sequences are highly conserved in S pyogenes, and only a few sequence variants were detected in the ssrA PCR amplicon region (Supplementary Results Figure 1). Our assay enables clear discrimination between M1_UK and other emm types, including some intermediate strains (M1_13SNPs, but not M1_23SNPs). Real-time PCR is an established detection method offering greater accessibility and requiring fewer resources compared to WGS, allowing for implementation across larger geographical areas and with higher testing volumes. Moreover, when combined with the microwave-preparation method, the real-time PCR method described herein yields fast turnaround times from sample to result.

This study confirms persistence of M1_UK strains with an extended toxin repertoire (positive for ssa, speC, and spd1) in Australia [33]. We also detected a CovR A¹¹¹V mutation in the M1_UK genotype, which has previously only been reported in a single lethal emm81 isolate causing bacteremia and necrotizing fasciitis [47]. The expression of SpeB remains unchanged in these invasive M1_UK  covR mutants, which retain increased speA expression levels, but expression of other important virulence factors is highly upregulated. Intriguingly, covR mutations have not frequently been reported in emm1 strains compared with covS mutations that have been implicated in the invasiveness of modern emm1 isolates [50]. Additional studies are required to better understand how this covR mutation may impact the virulence potential of M1_UK. Nonetheless, it will be important to monitor the frequency of this covR SNP within the M1_UK genotype in both invasive and noninvasive isolates.

Supplementary Data

Supplementary materials are available at The Journal of Infectious Diseases online (http://jid.oxfordjournals.org/). Supplementary materials consist of data provided by the author that are published to benefit the reader. The posted materials are not copyedited. The contents of all supplementary data are the sole responsibility of the authors. Questions or messages regarding errors should be addressed to the author.

Notes

Author contributions. S. B. and J. A. T. were responsible for article conceptualization and writing of the original draft. S. B., S. D., A. J. H., O. M. B., M. R. D., and J. A. T. were responsible for article research, experimental design, and data analysis and curation. All authors were responsible for writing, editing, and reviewing of the manuscript, and all authors approved the final submission of the manuscript. J. A. T. was responsible for correspondence with co-authors, completion of required documentation, and submission of the manuscript.

Data availability. Whole genome sequencing data obtained in this study were submitted to the Sequence Read Archive database (https://www.ncbi.nlm.nih.gov/sra; BioProject accession PRJNA1111678).

Financial support. This work was supported by the Queensland Paediatric Infectious Diseases Sakzewski Research Laboratory. We also thank the National Health and Medical Research Council for their support.

References

Author notes