Streptococcus pyogenes Lineage ST62/emm87: The International Spread of This Potentially Invasive Lineage

Abstract

Streptococcus pyogenes is known to be associated with a variety of infections, from pharyngitis to necrotizing fasciitis (flesh-eating disease). S. pyogenes of the ST62/emm87 lineage is recognized as one of the most frequently isolated lineages of invasive infections caused by this bacterium, which may be involved in hospital outbreaks and cluster infections. Despite this, comparative genomic and phylogenomic studies have not yet been carried out for this lineage. Thus, its virulence and antimicrobial susceptibility profiles are mostly unknown, as are the genetic relationships and evolutionary traits involving this lineage. Previously, a strain of S. pyogenes ST62/emm87 (37–97) was characterized in our lab for its ability to generate antibiotic-persistent cells, and therapeutic failure in severe invasive infections caused by this bacterial species is well-reported in the scientific literature. In this work, we analyzed genomic and phylogenomic characteristics and evaluated the virulence and resistance profiles of ST62/emm87 S. pyogenes from Brazil and international sources. Here we show that strains that form this lineage (ST62/emm87) are internationally spread, involved in invasive outbreaks, and share important virulence profiles with the most common emm types of S. pyogenes, such as emm1, emm3, emm12, and emm69, which are associated with most invasive infections caused by this bacterial species in the USA and Europe. Accordingly, the continued increase of ST62/emm87 in severe S. pyogenes diseases should not be underestimated.

1. Introduction

1.1. Streptococcus pyogenes Infections

Streptococcus pyogenes, also known as group A Streptococcus (GAS), can be associated with various infections, such as pharyngitis, impetigo, cellulitis, and abscesses. This bacterium is an important reemerging pathogen associated with an increasing number of invasive diseases such as septicemia, streptococcal toxic shock syndrome (STSS), and necrotizing fasciitis [1]. Repeated GAS infections may lead to autoimmune sequelae, including rheumatic fever. With the establishment of surveillance programs in many developed countries, data on the incidence and severity of S. pyogenes diseases has been accumulating, allowing scientists to have access to important epidemiological data contributing to a better understanding and control of these diseases [2]. In New Zealand, the overall incidence of infections caused by invasive group A Streptococcus (iGAS) during a 6-year period was 5.6 per 100,000 (95%CI 4.1–7.4) [3]. In Idaho, USA, from 2008 to 2019, the incidence increased from 1.04 to 4.76 cases/100,000, and two outbreaks were identified [4].

M protein, a surface immunodominant antigen of S. pyogenes, is an important determinant of bacterial virulence. This protein is encoded by the emm gene, whose hypervariable region is the base for the emm typing of S. pyogenes strains. GAS can be classified into more than 240 emm types and 1000 subtypes, whose distribution patterns can vary regionally [5].

In France, from 2009 to 2017, a total of 61 different types of emm were identified, and the most frequent were emm28 (16%), emm89 (15%), emm1 (14%), and emm4 (8%), representing >50% of circulating GAS strains. However, other types of emm (emm44, emm66, emm75, emm83, emm87) emerged continually. Furthermore, among emerging types, emm75 and emm87 showed increased prevalence with persistent annual incidence, demonstrating the risk of clonal expansion [6]. A study on iGAS, conducted from January 2000 to May 2017, in Europe and the USA, demonstrated that the most identified types were emm1, emm28, emm89, emm3, emm12, emm4, and emm6, and together they represented approximately 50–70% of total GAS strains in North America. The emm1 type prevailed in most surveys conducted in the United States, followed by emm12, emm28, and emm3. However, in the Thunder Bay District of Canada, unusual types predominated, and the most prevalent were emm87 (12.3%), emm82 (10.8%), emm1, emm101, and emm83 (9.2% each), and emm114 (7.7%) [7].

1.2. The Increased Emergence of ST62/emm87 Lineage

Some studies have correlated certain types of emm with more severe clinical manifestations and disease. Despite the prevalence of specific types of emm types in invasive diseases, such as emm1, recent isolates of emm87 have been associated with family- and hospital-acquired outbreaks of invasive S. pyogenes infections. Due to this pattern, it has been proposed that these GAS isolates are highly transmissible [8,9]. In fact, other studies have shown that the prevalence of emm87 has increased among iGAS infections in different countries, such as Portugal, Spain, and France [6,10,11], being one of the five most frequent emm types in Europe and thirteenth in the USA [7].

Recent studies with an emm87 knockout mutant of the S. pyogenes strain 20161436, isolated from an invasive disease in Minnesota, USA, showed that the Δemm87 knockout was less cytotoxic to THP-1 macrophage-like cells and that purified M87 showed increased dose-dependent cytotoxicity. Furthermore, mature 1L-1β release was decreased in Δemm87-infected macrophages, indicating that M87 can trigger mature 1L-1β release. Additionally, the loss of M87 was found to cause an increase in S. pyogenes clearance. The emm87 knockout showed reduced survival in whole blood and neutrophil killing assays, although it did not sensitize S. pyogenes cells to killing by serum, to which it was intrinsically resistant. Furthermore, infection with the emm87 knockout mutant led to a greater neutrophil oxidative burst compared with S. pyogenes wild-type (WT) and emm87-complemented strains. Finally, in a systemic mouse model, mortality was reduced in the Δemm87 compared with WT and emm87-complemented strains. Thus, it has been suggested that emm87 is involved in the pathogenicity of GAS by modulating the interaction between S. pyogenes and innate immune cells [12].

1.3. Therapeutic Failures and the ST62/emm87 Lineage of S. pyogenes

Although GAS remains universally susceptible to penicillin, an important aggravating factor of S. pyogenes infections, including iGAS, is therapeutic failure, which can occur even when the appropriate antimicrobial is used to treat such infections. Treatment failure has remained a problem for years and can occur during the treatment of relatively simple illnesses such as pharyngitis to more severe invasive diseases. Treatment failure in invasive infections can have high failure rates. It was estimated in 2005 that the burden of iGAS diseases is unexpectedly high, with at least 663,000 new cases and 163,000 deaths per year [13], and there is no evidence that this rate is decreasing. Several mechanisms have been implicated in therapeutic failures, including the “Eagle” effect, tolerance, decreased penetration of the antibiotic into biofilms, and the development of persister cells [14]. However, the latter seems to be the most accepted mechanism currently to explain failures in different bacterial species to various types of antimicrobials. Persisters are dormant cells that exhibit inhibition of protein synthesis, metabolism, and impairment of cell division so that they do not respond to the action of antimicrobials that require active bacterial replication for action. Recently, using proteomic and other molecular approaches, we demonstrated that the S. pyogenes strains 37–97 (ST62/emm87) can form persister cells that are refractory to different types of drugs, including β-lactams and clindamycin, indicated for the treatment of iGAS [15]. In addition to impaired cell growth and protein inhibition, persisters of ST62/emm87 background also showed an increased efflux capacity that could also be involved in the observed antimicrobial refractoriness.

1.4. Study Aims

Despite the increasing number of S. pyogenes sequence submissions to the Sequence Read Archive (SRA)—5 in 2009 to 47,936 in August 2023—most strains were from Europe and North America [16,17]. In the GenBank, 2241 genome sequences of S. pyogenes from different lineages are available, of which only 270 are completely closed genomes [18]. Of the total of 2241 genomes, only 71 are from Brazil. In addition, to the best of our knowledge, there is no study on the comparative genomics of the ST62/emm87 lineage. Thus, to gain insights into the pathogenesis and evolution of this lineage, we performed comparative genomic and phylogenomic analyses using the whole genome sequence (WGS) of the 37–97 strain and other ST62/emm87 genomes of international origin, deposited at the National Center for Biotechnology Information (NCBI) databases, such as GenBank and SRA, including those related to severe outbreaks.

2. Results and Discussion

Resistance gene analysis using the Comprehensive Antibiotic Resistance Database (CARD) revealed only two resistance genes: lmrP and mefE. The lmrP gene is predicted to encode a major facilitator superfamily (MFS) LmrP protein that has been associated with increased resistance to the antibiotic classes lincosamides, streptogramins, tetracyclines, and macrolides [19]. The mefE gene has been reported in other Gram-positive species, such as Streptococcus pneumoniae, and plays an important role in macrolide resistance [20]. The analysis of 58 genomes included in this study (Supplementary Materials Table S1) shows the same resistance profile for all ST62/emm87 genomes. The exception was the outgroup strain NS6033, which carries the tetM gene. These findings demonstrate that strains of the lineage ST62/emm87 exhibit no significant variation in terms of antimicrobial resistance compared with strains of other lineages of S. pyogenes isolated from various countries. As far as antimicrobial resistance is concerned, S. pyogenes has continued to be highly susceptible to almost all classes of antibiotics. Resistance to macrolides (and related compounds) and tetracyclines alone is commonly found among S. pyogenes [21].

The analyses of the virulence gene profile of strain 37–97 from Brazil revealed the presence of 20 known virulence-related genes: fbp54, which encodes a fibronectin/fibrinogen-binding protein; hasABC, hyaluronan synthase A, B, and C; hylP, hyaluronoglucosaminidase-phage-associated; ideS/mac, Ig protease IdeS domain-containing protein; lmb, laminin-binding protein; mf/spd, streptodornase B; mf2, deoxyribonuclease (DNase) Mf2; mf3, DNase Mf3; scpA, C5a peptidase; ska, streptokinase A; slo, streptolysin O; smeZ, streptococcal mitogenic exotoxin Z; speB, streptopain; speC, exotoxin type C; speG, exotoxin type G; speJ, exotoxin type J; speK, exotoxin type K; and ssa, streptococcal superantigen SSA. Analysis of the other ST62/emm87 genomes from NCBI of international origin showed that some virulence genes were highly conserved among these strains and include hasABC, ideS/mac, lmb, mf/spd, ska, slo, smeZ, speB, speG, and speK (Figure 1).

The presence of these various virulence-related genes is consistent with the variety of diseases (from mild to invasive infections and scarlet fever) caused by the S. pyogenes of the ST62/emm87 lineage (Supplementary Materials Table S1). It is well known that the presence of some virulence factors, especially streptococcal pyrogenic exotoxins (SPEs), is involved in the intense inflammatory response and tissue destruction in S. pyogenes severe infections; therefore, characterization of the virulome for specific lineages is of special importance [22,23].

For the most divergent genomes of strains GCH145, NCTC12065, NS6033, 44079V1S1, and 20018V1I1—which clustered into a completely independent clade and were therefore chosen as outgroups for the phylogenetic tree—the hylP, mf2, speC, and ssa genes are absent; however, these four genes were detected in all other ST62/emm87 genomes analyzed (Figure 1 and Figure 2). These isolates were recovered mainly from localized infections, except for one that was obtained from a sterile site and one isolate whose clinical site was not reported (Supplementary Materials Table S1).

In the studied collection (n = 58), 16 genomes were from invasive diseases that occurred in the USA (n = 14), Germany (n = 1), and Denmark (n = 1). It is important to note that these invasive isolates were distributed in different clades of the phylogenetic tree (Figure 2).

The virulence genes are highly conserved among the analyzed genomes, except for hylP, which was lost in the invasive isolates TSPY6, TSPY342, TSPY578, TSPY1009, and TSPY1074, and for polymorphisms found in the scpA gene. These data suggest that HylP hyaluronidase is probably not essential for invasive S. pyogenes infections. In fact, studies with invasive strains of S. pyogenes emm89 from Japan also detected the absence of hylP [24].

Of the 50 genomes for which clinical sources were reported, 17 (34.00%) showed higher levels of polymorphism in the scpA gene. Of these 17 isolates, a total of 43.75% (n = 7/16) were from invasive diseases, while only 29.41% (n = 10/34) were from non-invasive diseases (p = 0.3181). C5a peptidase (ScpA) is a serine protease that specifically degrades the C5a complement component. It has been suggested that this protease inhibits chemotaxis and may also play a role in host immune evasion [25,26,27]. In Streptococcus agalactiae, polymorphisms in scpB (scpA homolog) can lead to functional differences in protein synthesis [28]. However, the role of scpA gene polymorphism in the pathogenesis of S. pyogenes diseases needs to be evaluated experimentally.

It is notable that Bra001, Bra002, and Bra047 strains from Brazil were obtained from patients who presented with scarlet fever. These strains shared identical virulence patterns with the Brazilian strain 37–97 (isolated from a pharyngitis case), except that strain Bra047 lacks mf2 and speC genes (Supplementary Materials Table S1). These data seem to indicate that the presence of mf2 and speC may not be critical for the pathogenesis of scarlet fever in this S. pyogenes lineage. Corroborating these findings, variation in the content of some genes, including speC, was also evident for a global population of scarlet fever strains displaying type emm1 [29]. In contrast to the invasive S. pyogenes, none of the scarlet fever strains show significant polymorphisms in the scpA gene. The speC gene was absent in 4 of 13 pharyngitis isolates, the mef2 gene in 5 isolates, and the hylP gene in 2 isolates. Ten isolates were collected from skin/soft tissue infections. Among them, 3 lack the hylP gene, and 1 (isolate 44079V1S1) lacks the hylP gene and also the mef2, mef3, speC, speJ, and ssa genes.

In the phylogenetic tree, 51 of the 58 ST62/emm87 genomes analyzed were grouped into 3 main clusters, marked in yellow, blue, and red colors (Figure 2). Strain 37–97 from Brazil was grouped in the blue clade, which clustered genomes from EUA, Canada, Sweden, and New Zealand whose strains were collected from 1997 to 2015. The Brazilian strains Bra002 and Bra047 were grouped in the yellow clade; however, the Bra001 and Bra045 strains were in a more basal position outside the main clusters. It is remarkable that strains from all clades were able to cause conditions ranging from uncomplicated pharyngitis to more complicated invasive diseases (Supplementary Materials Table S1 and Figure 2).

When the virulence profile of strain 37–97 was compared with hypervirulent strains of S. pyogenes from the USA and Canada [30,31,32,33] that exhibited different types of emm and multilocus sequence typing (MLST) (strains M1T15448, ST28/emm1; 1838, ST15/emm3; MGAS9429, ST36/emm12; and MGAS15252, ST172/emm59), we found that fbp54, hasABC, ideS/mac, lmb, mf/spd, scpA, ska, slo, smeZ, speB, and speG genes were the only virulence-related genes present concomitantly among these hypervirulent strains, and all these genes were present in the strain 37–97 from Brazil. As previously suggested, other factors, including host genetics, may be important for the establishment of invasive diseases. However, more studies are necessary to test this hypothesis [34,35]. In addition, minor genomic variations may contribute to bacterial hypervirulence.

It has been observed that some specific mutations can discriminate between GAS strains isolated from invasive and non-invasive diseases. A mutation that inactivates the two-component virulence repressor system CovRS may facilitate a more severe disease caused by GAS [8,36,37]. For example, the genomes of the ST62/emm87 isolates, namely TSPY807, TSPY808, TSPY809, TSPY810, TSPY811, and TSPY816, were obtained from GAS isolates that caused a fatal intrafamily cluster of severe invasive disease in four siblings. An 8-bp duplication (insertion) was previously detected in the CovS-encoding gene of these TSPY genomes, resulting in a reading frame shift and a premature stop codon at position 28 compared with wild-type covS [8]. Because CovR suppresses the production of the antiphagocytic hyaluronic acid capsule, high-level production of the capsule is likely essential for the hypervirulent phenotype induced by CovRS inactivation [8,37].

A study by Galloway-Peña and colleagues [38] examined the effects of CovRS inactivation on acapsular strains of serotype M4. Two strains were used: a wild-type strain (M4-SC-1) and a naturally occurring CovS-inactivated strain (M4-LC-1) that contains an 11-bp covS insertion. Strain M4-LC-1 exhibited increased expression of surface proteins belonging to the Mga regulon, since Mga inactivation led to the reversal of this effect. Furthermore, only the M4-LC-1 strain showed upregulation of these surface proteins and several others, while the M4-SC-1 strain did not. Notably, the M4-LC-1 strain was more virulent in a mouse model of bacteremia, despite inducing fewer skin lesions in the same model. Thus, the findings demonstrate the importance of the covS inactivation mechanism in the virulence of S. pyogenes, shedding light on the mechanisms of hypervirulence and pathogenicity of this bacterium [38]. It is important to mention that in addition to the six strains TSPY807-811 and TSPY816 mentioned previously [8], we found in the collection studied here two other GAS strains (INFECT5034_SPY and Bra047) that have the covS gene inactivated. The natural covS mutation in Bra047 (isolated from a case of scarlet fever in Brazil) was due to an 11-bp insertion (5′ AGAAAATGCAG 3′) in this gene, and INFECT5034_SPY from Sweden (isolated from blood) had exactly the same 8-bp insertion (5′ CTTTTTTT 3′) found in strains TSPY807-811 and TSPY816.

Previous studies using different lineages of GAS strains implicated in streptococcal toxic shock syndrome (STSS) indicated differences in the ST62/emm87 strains with respect to gene content (presence of speK) and genetic polymorphisms for dpiB, which encodes a positive regulator of maeE encoding the malic enzyme. However, the exact role played by each genetic variation, in conjunction with the host response, in the pathogenesis of STSS and other severe infections caused by GAS remains unclear [39]. In our analysis using NGAS743 as a reference, all ST62/emm87 genomes analyzed from invasive and non-invasive GAS carried the speK gene with 100% identity and coverage, except GCH145_1, which showed 99.86% identity and 100% coverage. Regarding dpiB polymorphism, only TSPY6, TSPY57-58, TSPY157, TSPY539, TSPY578, TSPY646, SPY2015, and SPYORA1394A showed variability. Thus, only 3 of the 16 ST62/emm87 strains classified as iGAS present this polymorphism. Consequently, we were unable to associate the presence of the dpiB polymorphism with the severity of the infections.

Strain 37–97 from Brazil presented four regions consistent with known complete bacteriophages: PHAGE_Strept_315.3_NC_004586 (41), PHAGE_Strep_315.2_NC_004585 (18), PHAGE_Strep_315.4_NC_004587 (32), and PHAGE_Strep_315.4_NC _004 587 (36). The average GC content of each was 38.16%, 37.25%, 39.08%, 38.59%, and all int genes were confirmed via UNIPROT blast submission (http://www.uniprot.org; accessed on 10 July 2023) and showed 100% identity. The genomic organization of these phages includes genes dedicated to regulation, DNA replication, DNA restriction, genome packaging, phage morphology, and cell lysis. These phages also carried virulence-associated genes such as hyaluronidase (UNIPROT: A0A4D6BC74), C5a peptidase (UNIPROT: Q99Z24), and superantigen A (UNIPROT: A0A4D6AEG5).

The distribution of these phages in the genomes of the other 57 GAS analyzed was variable. Only 13 of the 57 genomes carried these four bacteriophages. Although most of these genomes (carrying these four bacteriophages) were grouped in the yellow clade together with the Brazilian GAS 37–97 (n = 8/14). Genomes carrying these four phages were also found distributed in all clades, except in the outgroups. A total of 30 genomes grouped into different clades contained at least two of these bacteriophages [PHAGE_Strep_315.2_NC_004585 (18) and PHAGE_Strept_315.3_NC_004586 (41)]. Taken together, these data suggest a horizontal and promiscuous dissemination of these two phages among ST62/emm87 strains. Phage-related genes may constitute approximately 50% of the accessory genes of S. pyogenes [16]. However, despite playing a key role in the evolution of S. pyogenes lineages and virulence diversity, their role in the genomic variability of S. pyogenes is still incompletely understood [40].

The main limitation of this study is related to the small number of genomes analyzed. This is due to the fact that genomes from invasive and serious GAS infections are rare and that relatively few strains of this lineage cause invasive infections in most countries. Furthermore, most isolates are from the USA and are related to outbreaks; thus, there is a low diversity of strains sequenced. Additionally, the data obtained from the NCBI BioSample regarding the type of infection is limited and sometimes incomplete, which may have influenced the definition of invasive and non-invasive infections.

Severe infections caused by S. pyogenes are relatively rare compared with mild to moderate infections, but they are associated with a high mortality rate and may increase in prevalence. The lineage of GAS may be critically important for its ability to cause disease. S. pyogenes strains of the ST62/emm87 lineage are spread in different countries as agents of pharyngitis, scarlet fever, skin/soft tissue infections (SSTI), streptococcal toxic shock syndrome (STSS), and invasive diseases, with increasing frequency and ability to cause outbreaks/infection clusters. Its potential to cause highly lethal invasive diseases and to develop mutations (such as those occurring in covS) that can lead to hypervirulence is a cause for concern. Importantly, these strains share many of the known virulence-related genes, including those frequently detected in hypervirulent invasive strains. It is possible for a bacterial strain to transition from a less virulent form to a more virulent form. However, the genetic mechanisms that lead to strains of S. pyogenes within the same lineage causing everything from simple pharyngitis to toxic shock syndrome and necrotizing fasciitis remain to be better elucidated. The possibility of the generation of persisters by isolates of this lineage as a potential cause of therapeutic failure is an additional threat, especially in serious invasive diseases including STSS and flesh-eating diseases [15].

3. Materials and Methods

3.1. Comparative Genomics

We performed a retrospective genomic study of S. pyogenes strains of the ST62/emm87 lineage. We selected all 16 assembled and annotated ST62/emm87 genomes deposited in GenBank, which were downloaded from the Genome Tree Report (https://ncbi.nlm.nih.gov/tools/treeviewer/; accessed on 15 April 2022) using the completely closed genomes of the NGAS743 strain (ST62/emm87) as reference. Additionally, we downloaded other genomes from the SRA database that were selected based on previous studies [8] and to cover different countries. A total of 41 genome reads (fastq) deposited in the NCBI Sequence Read Archive—SRA files (https://www.ncbi.nlm.nih.gov/sra; accessed on 15 April 2022) were downloaded and the adapter sequences and low-quality sequences were trimmed with TrimGalore version 0.6.4 (https://github.com/FelixKrueger/TrimGalore; accessed on 17 April 2022). The assembly was performed using Spades incorporated into the PATRIC platform (https://www.patricbrc.org/; accessed on 17 April 2022). The genome of the S. pyogenes strain 37–97 (GenBank accession: 37–97S; NZ_CP041408.1), sequenced by us [15], was annotated using RAST (Rapid Annotation Using Subsystem Technology) [41]. Previously, 37–97 genome sequence was trimmed using BBDuk Trimmer (version 1.0; accessed on 14 January 2019), and genome assembly was carried out using Newbler v3.0 [42]. Scaffolds were aligned against a reference genome (S. pyogenes strain NGAS743; GenBank Accession: CP007560) using Cross Match (version 0.990329; http://www.phrap.org/phrap.docs/phrap.html, accessed on 10 February 2019). Intra-scaffold and inter-scaffold gaps resulting from repetitive sequences were resolved by in silico gap filling [15].

The clinical source reported in this study was obtained from the NCBI BioSample. The genomes of S. pyogenes strains obtained from necrotizing fasciitis (SPYORA1394A and SPY2015) and streptococcal toxic shock syndrome (STSS) (TSPY58) were included in the invasive infection group. One isolate obtained from blood and another collected from a sterile site were not incorporated into the group of invasive infection because there was no accurate information about the clinical condition of the correspondent patients.

Multilocus sequence type (MLST) was performed using genome sequences submitted to MLST 2.0, hosted at the Center for Genome Epidemiology (CGE) (https://cge.cbs.dtu.dk/services/MLST/; accessed on 1 May 2022). For functional annotation of the total 58 analyzed genome sequences, the software ABRicate 0.9.8 (https://github.com/tseemann/abricate; accessed on 1 May 2022) was used with the following databases: antimicrobial resistance (Resfinder and CARD) and virulence (VFDB) genes.

Whenever necessary, manual curation was performed using UniProt (http://www.uniprot.org/; accessed on 5 May 2022) and/or the NCBI (https://blast.ncbi.nlm.nih.gov/Blast.cgi; accessed on 5 May 2022) databases. Additionally, the Map to Reference tool on the Geneious Prime platform version 2023.2.1 (Biomatters Inc.; Boston, MA, USA) and the BLAST command line (https://www.ncbi.nlm.nih.gov/books/NBK569856/; accessed on 5 September 2023) were used for further comparisons of genetic sequences grouped in a given phylogenetic cluster.

Flores et al. [8] described a mutation (8-pb insertion) in the covS response regulator from the two-component regulator CovRS in ST62/emm87 isolates, which may possibly increase the virulence potential of S. pyogenes. To test the presence of this insertion in other ST62/emm87 genomes analyzed here, the mutated sequence of the covS gene from TSPY809 (SRA accession: SRS1935365) and the covS sequence from MGAS5005 (reference genome used by Flores et al. [8]; GenBank accession: NC_007297; covS locus_tag: M5005_Spy0283) were used. Initially, covS sequences were searched in the studied genomes using the BLASTn command line, considering 100% identity and query coverage. Genomes that did not match were analyzed manually using Map to Reference on the Geneious Prime platform version 2023.2.1.

Additionally, Deniskin et al. [39] revealed a polymorphism in the dpiB gene (R170C mutation in protein sequence) and unique gene content (speK) among S. pyogenes ST62/emm87 strains involved in pediatric streptococcal toxic shock syndrome and suggested a possible contribution of these genetic profiles in the development of this syndrome by strains of this lineage. Thus, a strategy similar to that used to analyze the covS gene was carried out to investigate the dpiB polymorphism, but using the dpiB gene from the MGAS10750 genome as a reference (GenBank accession: NC_008024). Next, the dpiB gene sequence was aligned and compared using the Map to Reference tool on the Geneious Prime platform version 2023.2.1 with that of strain NGAS743 (Genbank accession: CP007560), which belongs to ST62/emm87, and this sequence was used as a reference to analyze the 58 genomes studied here. The searches were carried out using the BLASTn command line, with a threshold of 100% identity and coverage. Again, genomes that did not match 100% identity and query coverage were manually inspected using Map to Reference on the Geneious Prime platform version 2023.2.1.

The software Phaster (https://phaster.ca/; accessed on 29 May 2023) was used to search for bacteriophage content in the strain 37–97 genome, and only complete phages were considered in this analysis. The phages sequences found in the 37–97 genome were annotated using Geneious Prime software platform version 2023.2.1 to assess the presence of important genes, such as virulence-associated genes. Additional, sequences of the phages found in 37–97 genome were searched in the other 57 ST62/emm87 genomes of this study using the BLAST command line application (https://www.ncbi.nlm.nih.gov/books/NBK569856/; accessed on 20 September 2023).

3.2. Phylogenetic Tree

The tree was constructed with 58 S. pyogenes ST62/emm87 genome sequences available in the GenBank and SRA. These sequences were aligned using ROARY (https://github.com/sanger-pathogens/Roary; accessed on 23 February 2023) and the command “roary -e --mafft -p 8 *.gff” for core genome alignment. The maximum likelihood tree was achieved by employing RaxMLGUI [43] with the GAMMA distribution and GTR substitution model and default parameters with 1000 bootstrap replicates. The tool “Interactive Tree of Life” (iTOL) v.4 was used for tree visualizations and editions [44].