The Emergence and Impact of the M1_UK Lineage on Invasive Group A Streptococcus Disease in Aotearoa New Zealand

Abstract

A toxigenic variant of Streptococcus pyogenes serotype M1 (emm1) is responsible for the recent unprecedented notifications of scarlet fever and invasive group A streptococcal (iGAS) infection in the United Kingdom [1, 2]. Since its emergence in 2008, this variant (known as M1_UK) has become the dominant emm1 lineage in many countries, reportedly expanding and associated with disease surges throughout the northern hemisphere and Australia [1–3].

M1_UK is differentiated from earlier global emm1 lineages (M1_global) based on the presence of 27 chromosomal single-nucleotide polymorphisms (SNPs). Having evolved via intermediate sublineages, M1_UK is characterized by its increased production of the streptococcal pyrogenic exotoxin A (SpeA) superantigen [1]. Notably, the immune response to enhanced SpeA expression by M1_UK has not been described.

This retrospective genomic analysis describes the prevalence of M1_UK in invasive emm1 strains over the past decade in Aotearoa New Zealand (NZ), which has a high incidence of iGAS in comparison with countries that have a similar human development index [4]. We provide a novel description of the SpeA antibody response in children with M1_UK and M1_global pharyngitis and compare the prevalence of M1_UK in the community vs invasive disease.

METHODS

Bacterial Strains

The Institute of Environmental Science and Research’s Invasive Pathogens Laboratory received 417 group A Streptococcus (GAS) isolates collected from patients between January 2013 and November 2023 that were assigned emm1 based on standard emm-typing procedures (https://www.cdc.gov/strep-lab/php/group-a-strep/emm-typing.html). Isolates were referred from diagnostic laboratories around the country for passive (nonnotifiable) surveillance and included if received by 6 December 2023. All emm1 isolates were analyzed to determine the proportion that belong to the M1_UK lineage.

Genomic Analysis

Isolates were determined to be M1_UK by a combination of allele-specific polymerase chain reaction (AS-PCR) and whole genome sequencing. Heat-killed suspensions from the earliest emm1 isolates (2013–2014, n = 88) were screened for rofA, gldA, and pstB SNPs characteristic of the M1_UK lineage by a recently reported AS-PCR [5]. Genotypically characterized strains were used as controls (Supplementary Data 1). This method confirmed the genotype of 34 of 88 isolates. Isolates with non–wild type AS-PCR results or that failed to amplify (54/88) were sequenced with the remainder of the emm1 isolates not subjected to AS-PCR (2015–November 2023, 328/329; Supplementary Data 1). Genomic DNA was extracted, and libraries were prepared by the plexWell 96 Library Preparation Kit (SeqWell) and sequenced on an Illumina NextSeq with 150–base pair paired-end chemistry.

Genomes were assembled with SKESA (https://github.com/ncbi/SKESA), and emm type was confirmed in silico (https://github.com/MDU-PHL/emmtyper). Genome assemblies were subjected to in silico PCR that detected wild type or SNP alleles in the rofA, gldA, and pstB genes, as in the AS-PCR, to differentiate the M1_global, M1_UK, and M1_intermediate lineages [5]. Phylogenetic analyses were performed on the 377 NZ invasive emm1 genomes and compared with 539 publicly available global genomes from 2004 to 2020 [3], including 4 historical NZ isolates (Supplementary Data 2). Additionally, 59 community emm1 genomes were included that had been obtained from a study of school-aged children conducted from 2018 to 2019 in Auckland [6]. Overall 15,087 SNPs relative to M1_global MGAS5005 were used to construct maximum likelihood phylogeny via IQ-TREE2 version 2.0.6 (https://github.com/iqtree/iqtree2) with model selection and 2000 ultrafast bootstrapping. Alignment was generated by snippy version 4.6.0 (https://github.com/tseemann/snippy) and recombinant corrected with gubbins version 3.2.2 [7]. Phylogeny was visualized by Microreact (https://microreact.org/).

SpeA Serology

Sera were obtained from the community GAS infection study described previously (ethics approval 17/NTA/262 [8]). Convalescent samples from children with GAS-positive pharyngitis caused by an emm1 strain (n = 36) [6] or non-emm1 strains (n = 15 infected with emm53 or emm81) were tested for anti-SpeA antibodies, as were matched healthy children with negative throat swab and no evidence of a recent GAS infection (n = 23). Convalescent samples were collected at least 2 weeks after the initial positive swab to ensure seroconversion. Antibodies were quantified by an immunoassay in which SpeA, expressed and purified as described [9], was covalently coupled to MagPlex microspheres (beads) via the xMAP Antibody Coupling Kit (Luminex; Thermo Fisher) following published protocols [10]. Sera were diluted 1:1000, and net median fluorescence intensity was determined by subtracting background values from no-serum control wells.

RESULTS

M1_UK Is the Dominant Invasive emm1 Lineage

Passive surveillance data show that the M1_UK lineage was present in NZ in 2013 but as a low proportion (<2%) of all invasive emm1 isolates that year (Figure 1A). The proportion of M1_UK isolates increased from 2015, outcompeting the M1_global lineage from 2019, and is now the dominant iGAS emm1 lineage in NZ. Between 2020 and 2022, coinciding with the COVID-19 pandemic, fewer iGAS isolates were referred annually to the Institute of Environmental Science and Research, and the proportion of emm1 isolates decreased (Supplementary Figure 1). In 2023 emm1 proportions returned to prepandemic levels, with M1_UK being the dominant emm1 lineage (88%). Furthermore, the proportion of M1_UK increased to 12.6% of all invasive isolates in 2023 to November, up from 0.2% 10 years earlier.

Figure 1.

A, Proportion of emm1 lineages: M1_global and M1_intermediate combined (blue) and M1_UK (red) as a proportion of all passive surveillance invasive emm1 isolates from New Zealand collected from patients by year (January 2013–November 2023). All M1_intermediate strains in this study are the M1_13SNPs sublineage (Supplementary Data 1). B, Maximum likelihood phylogeny of 377 New Zealand invasive group A streptococcal isolates, 539 global context isolates [3], and 59 community emm1 isolates from Auckland school-aged children [6]. Terminal nodes are colored by location, and the outer ring indicates emm1 lineage. The inner ring shows invasive and community isolates from the Auckland 2018–2019 subset. Scale bar: number of substitutions per site. C, Anti-SpeA antibody responses in sera from children with pharyngitis (M1_UK, n = 13; M1_global, n = 23; non-emm1, n = 15) vs healthy controls (n = 23). The non-emm1 group combines data from children with emm81 (open diamonds) and emm53 (closed diamonds) pharyngitis. Horizontal lines show median values. Statistical significance was assessed by the Kruskal-Wallis test with Dunn test for multiple comparisons. ***P < .001. ****P < .0001. MFI, mean fluorescence intensity.

Open in new tabDownload slide

Phylogenetic analysis of NZ isolates as compared with global context isolates shows that there have been multiple introductions of the M1_UK lineage into NZ over the last decade (Figure 1B). The largest NZ M1_UK clade likely has phylogenetic origins in the United Kingdom, and a smaller, more recently expanding subclade is related to isolates from Queensland, Australia.

Community and Invasive M1_UK

In 2018 and 2019 combined, 16 of 36 (44%) invasive emm1 isolates referred from the Auckland region belonged to the M1_UK lineage; the remainder belonged to the M1_global and M1_intermediate lineages (Supplementary Data 2). During a similar period, 20 of 59 (34%) Auckland community emm1 isolates collected from children with pharyngitis were M1_UK, which was not significantly different (χ² test, P > .05). Phylogenetic analysis of this 2018–2019 Auckland subset did not show clustering by infection type, suggesting that community and invasive isolates are genomically related (Figure 1B).

Antibody Response to SpeA in Children With Pharyngitis

The level of anti-SpeA antibodies was significantly higher in children with culture-positive GAS pharyngitis caused by M1_UK or M1_global as compared with healthy controls (P < .001). However, there was no significant difference (P > .01) in anti-SpeA antibody levels in children infected with non-emm1 strains as compared with healthy controls or those with M1_UK or M1_global infections (Figure 1C).

DISCUSSION

M1_UK has outcompeted its global counterpart and is now the dominant emm1 lineage associated with invasive disease in many countries [2, 3]. This study shows that NZ has a similar pattern, where M1_UK is now the dominant invasive emm1 lineage referred to the national reference laboratory for passive surveillance. Phylogenetic analyses indicate that multiple M1_UK introductions have occurred over the last decade with evidence of subsequent local clonal expansion. This echoes the multiple international introductions and subsequent establishment of specific GAS sequence clusters observed in the Auckland community [6] and highlights the influence of globally expanding GAS lineages on the NZ population structure. The high level of genomic similarity between circulating community M1_UK strains and those referred for surveillance in this study indicates that the community is a source of invasive infection in NZ. This supports recent evidence from the United Kingdom that community transmission of M1_UK pharyngeal strains can lead to invasive infection [2].

M1_UK is associated with increased SpeA expression in vitro [1], although it is unclear how this correlates with immune responses following clinical infection. For children with pharyngitis, the antibody response to SpeA was not significantly different between M1_UK and M1_global infections in this study, suggesting that any increase in SpeA expression is not associated with a concurrent increase in antigen-specific antibodies. However, it remains possible that differences in the magnitude of anti-SpeA antibodies may be present in M1_UK invasive disease, where SpeA production might be expected to play a more central role in pathogenesis. Of note, some children infected with non-emm1 strains had elevated SpeA antibodies, which may be a result of cross-reactivity between SpeA and other superantigens expressed by the infecting strain types (emm53 and emm81 [6]) or prior infections with SpeA-producing strains.

Concurrent with the COVID-19 pandemic, this study shows a decrease in the proportion of referred emm1 iGAS isolates. This is in keeping with the hypothesis that public health measures, including the use of masks, decreased transmission of strains typically associated with pharyngitis, such as emm1, rather than those more typically associated with skin infections [11]. The increase in emm1 iGAS isolates referred in 2023 may reflect a return to prepandemic numbers. However, given the surges in invasive disease and scarlet fever driven by M1_UK overseas [1, 2], ongoing national surveillance of this variant remains warranted.

In summary, genomic surveillance has traced the emergence of M1_UK in NZ, including multiple international introductions, followed by local clonal expansion. The genomic similarity between community and invasive strains suggests that measures that effectively prevent GAS pharyngitis in schoolchildren, such as timely diagnosis, antibiotic treatment, and future vaccines, may also reduce iGAS infections.

Supplementary Data

Supplementary materials are available at Open Forum Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.

Notes

Acknowledgments. We thank diagnostic laboratories in Aotearoa New Zealand for their contributions to surveillance of invasive group A Streptococcus. We also thank Awanui Labs for collection and storage of specimens as part of the community study in Auckland. Genomic sequencing was conducted by the Next Generation Sequencing Laboratory at The Institute of Environmental Science and Research, funded by the New Zealand Ministry of Health and Genomics Aotearoa (ID 2004).

Author contributions. N. J. M. conceived the study. N. J. M., J. B., A. A., T. P., R. H. W., and P. E. C. provided resources. A. H., A. T., A. V., J. B., C. S., J. M., N. L., P. S., and X. R. performed experimental work and/or acquired data. A. V., N. L., P. S., and X. R. analyzed the data. A. V. and N. J. M. wrote the first draft of the manuscript; all authors revised it and approved the submitted version.

Data availability. Reads for isolates sequenced for this study are available in the NCBI Sequence Read Archive under BioProjects PRJNA1100230 and PRJNA1104651.

Financial support. This work was supported by funding from Manatū Hauora, Ministry of Health New Zealand (contract 370383-00) awarded to N. J. M., A. A., and R. H. W. as part of a Strep A Vaccine Initiative.

References

Author notes