Free access
Research Article
25 March 2020

Reduced In Vitro Susceptibility of Streptococcus pyogenes to β-Lactam Antibiotics Associated with Mutations in the pbp2x Gene Is Geographically Widespread


Recently, two related Streptococcus pyogenes strains with reduced susceptibility to ampicillin, amoxicillin, and cefotaxime, antibiotics commonly used to treat S. pyogenes infections, were reported. The two strains had the same nonsynonymous (amino acid-substituting) mutation in the pbp2x gene, encoding penicillin-binding protein 2X (PBP2X). This concerning report led us to investigate our library of 7,025 genome sequences of type emm1, emm28, and emm89 S. pyogenes clinical strains recovered from intercontinental sources for mutations in pbp2x. We identified 137 strains that, combined, had 37 nonsynonymous mutations in 36 codons in pbp2x. Although to a lesser magnitude than the two previously published isolates, many of our strains had decreased susceptibility in vitro to multiple beta-lactam antibiotics. Many pbp2x mutations were found only in single strains, but 16 groups of two or more isolates of the same emm type had an identical amino acid replacement. Phylogenetic analysis showed that, with one exception, strains of the same emm type with the same amino acid replacement were clonally related by descent. This finding indicates that strains with some amino acid changes in PBP2X can successfully spread to new human hosts and cause invasive infections. Mapping of the amino acid changes onto a three-dimensional structure of the related Streptococcus pneumoniae PBP2X suggests that some substitutions are located in regions functionally important in related pathogenic bacterial species. Decreased beta-lactam susceptibility is geographically widespread in strains of numerically common emm gene subtypes. Enhanced surveillance and further epidemiological and molecular genetic study of this potential emergent antimicrobial problem are warranted.


Generations of microbiologists, physicians, and others with strong interests in infectious diseases have been taught that Streptococcus pyogenes (group A streptococcus [GAS]) is universally susceptible to beta-lactam antimicrobial agents (1). Although the molecular basis for this resilient phenotype is unknown, given the global disease burden of greater than 700 million cases annually (2), universal susceptibility to these agents has been fortunate. Recently, Vannice et al. (3) identified two clonally related and epidemiologically linked strains of rare type emm43.4 S. pyogenes that had 8-fold higher MICs for ampicillin and amoxicillin and 3-fold higher MICs for cefotaxime, indicating decreased susceptibility to these antibiotics. The two strains had an identical single nonsynonymous (amino acid-altering) mutation in the pbp2x gene, encoding penicillin-binding protein 2X (PBP2X). This mutation confers a threonine-to-lysine replacement at amino acid 553 (Thr553Lys), a polymorphism that was not found in susceptible strains of type emm43.4. The authors suggested that the Thr553Lys replacement may be a first step toward S. pyogenes evolving resistance to beta-lactam antibiotics. The genomes of these two strains were sequenced as part of an outbreak investigation being done by Public Health—Seattle & King County in collaboration with the Centers for Disease Control and Prevention (CDC). Standard analysis conducted by the CDC GAS genome sequencing method includes identification of features potentially contributing to antibiotic resistance, including PBP2X variants (4). Through this process, the pbp2x missense mutations associated with decreased antibiotic susceptibility reported by Vannice et al. (3) were detected. The identification of these two strains is concerning and may signal a substantial public health problem because beta-lactams remain the frontline treatment globally for the majority of GAS infections.
To assess the potentially unrecognized broader extent of this inauspicious discovery, we felt compelled to interrogate our library of 7,025 genome sequences of type emm1, emm28, and emm89 S. pyogenes clinical isolates from intercontinental sources for nonsynonymous mutations in pbp2x. Bioinformatic analysis identified 137 strains with 37 amino acid changes at 36 sites in pbp2x that could alter MIC values for beta-lactam antibiotics. A subset of strains with pbp2x mutations was analyzed for beta-lactam MIC values using the gradient method (Etest strips). Our results indicate that clinical isolates with pbp2x mutations associated with small decreases in beta-lactam susceptibility in this common human-specific pathogen are more widespread than appreciated. Enhanced surveillance and fuller epidemiological and molecular genetic study of this potentially emergent antimicrobial problem are warranted.


S. pyogenes strains and whole-genome sequence data.

The emm1 (n = 3,615), emm28 (n = 2,095), and emm89 (n = 1,315) S. pyogenes strains that we studied have been described in our previous publications (58). The genome sequence data generated with Illumina instruments have previously been deposited in publicly available databases in the National Center for Biotechnology Information Sequence Read Archive (BioProject accession numbers PRJNA236767, PRJNA434389, PRJNA287922, and PRJNA387243). Nucleotide polymorphisms in the pbp2x gene in these strains were identified by bioinformatics methods that have been extensively described previously (5).

Antibiotic susceptibility determinations.

Forty-two strains with nonsynonymous mutations in pbp2x were tested for potential decreased susceptibility to penicillin by plating on tryptic soy agar supplemented with 6-ng/ml penicillin G (benzylpenicillin) or 15-ng/ml ampicillin (9, 10). These strains represent a diverse array of organisms with distinct pbp2x mutations from emm1, emm28, and emm89 organisms. Six reference strains lacking pbp2x mutations (wild-type [WT] strains, consisting of one strain of emm1, three strains of emm28, and two strains of emm89) were included as PBP2X consensus wild-type comparators. The reference strains are emm type and genetic clade matched and have the most common allele representative of their genetic background for global transcriptional regulators of known virulence factors, and several have been extensively studied both in vitro and in animal virulence experiments. The plates were incubated overnight at 37°C in 5% CO2, and growth was assessed as present or absent. MIC values for six beta-lactam antibiotics (ampicillin, penicillin G, cefotaxime, cefoxitin, ceftazidime, and meropenem) were determined by the gradient method (Etest strips; bioMérieux) using standard clinical laboratory procedures. MIC values were scored independently by three investigators. Some strains were also tested for penicillin G and ampicillin susceptibility by broth microdilution in Todd-Hewitt broth supplemented with 0.5% yeast extract (THY). Liquid cultures were incubated overnight at 37°C in 5% CO2, and growth was determined by determination of the optical density at 600 nm.

Phylogenetic analysis of whole-genome sequence data.

The phylogeny among the strains was inferred by neighbor joining based on concatenated sequential core chromosomal single nucleotide polymorphisms (SNPs) by methods described previously (5). To constrain inferences to predominantly vertically acquired SNPs, regions of recombination were predicted based on entire core genome sequences using the Gubbins algorithm, and putatively horizontally acquired SNPs were excluded. Clades of related strains were defined using hierarchical Bayesian analysis of population structure (hierBAPS), also as previously described (5).

Construction of isogenic strain MGAS27213-PBP2X-WT.

Isogenic strain MGAS27213-PBP2X-WT was constructed by replacing the naturally occurring mutant pbp2x gene (Pro601Leu) of clinical isolate MGAS27213 with the wild-type allele encoding Pro601 using procedures previously described (11). Briefly, wild-type pbp2x of strain MGAS27566 was amplified by PCR using primers pbp2x-1 (5′-GTGAATACATGCGATAGGAGAACTCCAG-3′) and pbp2x-2 (5′-CAATTGTACATTGATTCGCCAACTAAGTC-3′). The PCR amplicon was cloned into suicide vector pBBL740 and then transformed into parental strain MGAS27213, encoding the mutant pbp2x allele (Pro601Leu). Whole-genome sequencing of isogenic strain MGAS27213-PBP2X-WT confirmed that the mutant pbp2x allele was replaced by the wild-type pbp2x allele and that the constructed strain lacked spurious mutations.

PBP2X structure modeling.

The crystal structure of PBP2X from Streptococcus pneumoniae (PDB accession number 1RP5, chain A) was used to map the location of the amino acid substitutions relative to the active site of the transpeptidase domain. This structure was used because the structure of S. pyogenes PBP2X has not been determined. The two PBP2X proteins are well conserved in both amino acid sequence (54.1% identical, 82.1% similar) and structural fold, and PBP2X from S. pneumoniae has been well studied by several investigators (1216). The S. pyogenes amino acid substitutions were mapped onto the S. pneumoniae PBP2X structure using the Chimera program (17). Chimera was also used to align PBP2X with PBP3 from Pseudomonas aeruginosa (PDB accession number 6UN3) and PBP2a from Staphylococcus aureus (PDB accession number 1VQQ, chain A) to assign the role of each residue in relation to PBP2X within S. pyogenes.


Identification of PBP2X amino acid replacements.

To test the hypothesis that the S. pyogenes strains in our international collection of human clinical isolates contained polymorphisms in the pbp2x gene, we interrogated the population genomic data generated in our previous studies of emm1, emm28, and emm89 organisms (58). The vast majority of these strains were recovered from a normally sterile site of patients with invasive infections, such as bacteremia and necrotizing fasciitis. Among the 7,025 whole-genome sequences examined, we identified 137 strains that in the aggregate had 37 nonsynonymous mutations altering 36 codons of the 2,259-nucleotide pbp2x gene. We also identified 161 strains with a synonymous single nucleotide polymorphism (that is, a silent mutation that would not alter the amino acid sequence of PBP2X) at 10 positions, each in a separate codon of pbp2x. Thus, 79% of SNP sites resulted in an amino acid replacement, a significantly greater percentage than expected by chance alone (for the 48 pbp2x alleles in the cohort by the Nei-Gojobori method, the ratio of rates of nonsynonymous/synonymous site substitutions [Ka/Ks] = 1.49; Fisher exact test, P = 5.27e−41). This elevated percentage of nonsynonymous mutations is consistent with the effect of positive selection acting on pbp2x. Among the strains with nonsynonymous mutations, with a single exception, each of the 137 strains had only one amino acid replacement relative to the consensus wild-type PBP2X sequence. The exception was an emm28 strain (MGAS28532) recovered in the United States that had a unique combination of two contiguous amino acid replacements (Phe599Tyr and Gly600Asp) in PBP2X (Table 1). Of note, none of the 7,025 isolate sequences interrogated had an insertion or deletion mutation in pbp2x, indicating that the peptidoglycan transpeptidase function of PBP2X is essential. This finding is consistent with the results of saturating transposon mutagenesis screens, which, during library generation, also failed to recover strains with integrations in pbp2x (18, 19).
TABLE 1 PBP2X amino acid replacement strains and PBP2X wild-type comparator strainsa
Strain (MGAS no.)emm typeAmino acid substitutionCountryYrCapsuleResults on agar with:MIC (μg/ml)
Penicillin G at 6 ng/mlAmpicillin at 15 ng/mlPenicillin GbAmpicillinCefotaximeCefoxitinCeftazidimeMeropenem
22211Wild typeAustralia1998PositiveNegativeNegative<0.0160.0160.0231.50.190.006
238771Wild typeCanada2002Positive  0.012*0.016    
247911Wild typeFinland2006Positive  0.012*0.016    
786728Wild typeCanada1991Negative  0.012*0.016    
835728Wild typeFinland1996Negative  0.012*0.016    
1077828Wild typeCanada1998Negative  0.012*0.016    
1078328Wild typeCanada1998Negative  0.012*0.016    
1105228Wild typeFinland2000Negative  0.012*0.016    
2796128Wild typeUnited States2005NegativeNegativeNegative<0.0160.0160.02310.190.006
2842628Wild typeUnited States1999NegativeNegativeNegative<0.0160.0160.0321.50.190.004
2873728Wild typeUnited States2012NegativeNegativeNegative<0.0160.0160.0321.50.190.003
2890528Wild typeUnited States2004Negative  0.012*0.016    
2353089Wild typeItaly1997weak  0.012*0.016    
2656889Wild typeUnited States1996Positive  0.012*0.016    
2664589Wild typeUnited States2009Positive  0.012*0.016    
2684489Wild typeUnited States2008NegativeNegativeNegative<0.0160.0160.0231.50.190.006
2754589Wild typeFinland2010weak  0.012*0.016    
2756689Wild typeFinland2011NegativeNegativeNegative<0.0160.0160.0321.50.190.006
2831528Asp52GlyUnited States1998NegativeNegativeNegative<0.0160.0160.02310.190.004
2832928Asp52GlyUnited States1998Negative        
2843328Asp52GlyUnited States2003Negative        
2889428Asp52GlyUnited States2004Negative        
2869228Ser92PheUnited States2008Negative        
2663789Met171IleUnited States2009Negative        
2663989Met171IleUnited States2009Negative        
2666789Met171IleUnited States2009PositiveNegativeNegative<0.0160.0160.0231.50.250.006
2741389Asp233AsnUnited States2011Negative        
2841528Val281IleUnited States2003Negative        
2877228Gly288SerUnited States2012Negative        
2877328Gly288SerUnited States2012NegativePositivePositive0.0160.0230.0321.50.250.006
2686089Gly288SerUnited States2003PositivePositivePositive0.0160.0230.04720.380.008
2692989Gly288SerUnited States2003PositivePositivePositive0.0160.0230.0161.50.190.006
2693289Gly288SerUnited States2003PositivePositivePositive0.0160.0230.0321.50.250.006
2743889Gly288SerUnited States2006PositivePositivePositive0.0160.0230.0471.50.250.008
2871128Thr323MetUnited States2011Negative        
2836728Met342IleUnited States2003Negative        
2703389Met342IleUnited States2008PositivePositivePositive0.0230.0230.03230.250.006
2878228Thr461ProUnited States2012NegativePositiveNegative0.0160.0160.0161.50.1250.006
2689989Gln462HisUnited States2003NegativeNegativeNegative<0.0160.0160.0231.50.190.006
2808828Gly521SerUnited States2010NegativeNegativeNegative<0.0160.0160.0231.50.190.004
2853228Phe599Tyr and Gly600AspUnited States2006NegativePositivePositive0.0230.0320.0231.50.190.012
2798228Gly600AspUnited States2005Negative        
2816528Gly600AspUnited States1998Negative        
2833628Gly600AspUnited States2000Negative        
2838028Gly600AspUnited States2003Negative        
2842528Gly600AspUnited States1999NegativePositivePositive0.0160.0230.0321.50.190.006
2879228Gly600AspUnited States2000NegativePositivePositive0.0160.0230.0321.50.190.006
2714389Gly600AspUnited States2012NegativePositivePositive0.0160.0230.0321.50.1250.006
2732689Gly600AspUnited States2013NegativePositivePositive<0.0160.0160.02310.0940.004
2683789Pro601LeuUnited States2011Negative        
2721389Pro601LeuUnited States2012NegativePositivePositive0.0320.0470.0641.50.380.012
2730889Pro601SerUnited States2013NegativeNegativeNegative<0.0160.0160.0321.50.250.003
2731689Pro601LeuUnited States2013NegativePositivePositive0.0320.0470.0641.50.380.016
2745389Pro601LeuUnited States2010NegativePositivePositive0.0320.0470.0641.50.380.012
2832328Gly647AspUnited States2001Negative        
2674389Val662IleUnited States1999Negative        
255991Glu695AspUnited States2010PositiveNegativeNegative<0.0160.0160.0231.50.190.006
256031Glu695AspUnited States2010Positive        
2700989Lys730ArgUnited States2007Negative        
Summary of available data for 137 emm1, emm28, and emm89 S. pyogenes strains with amino acid replacements in PBP2X and 18 PBP2X wild-type comparator strains.
Penicillin G MIC assays were done using test strips with either of two dose ranges, consisting of the standard range of 0.016 to 256 μg/ml or a low range 0.002 to 32 μg/ml; assays done using the low-dose-range strips are indicated (*).
German Democratic Republic, the former East Germany.
The analysis identified four sites that had the same amino acid replacement (Gly288Ser, Met342Ile, Gly600Asp, and Pro601Leu) present in multiple emm types (Fig. 1). In each case, these amino acid replacements were represented among strains of types emm28 and emm89 (Fig. 1). Additionally, these four replacements were present among multiple isolates within a single emm type. The finding of the same replacements both in multiple emm types and in multiple isolates of the same emm type strongly suggests that these changes have been selected by exposure to beta-lactam antibiotics. In contrast, there was no example of sharing of amino acid replacements between type emm1 strains and either emm28 or emm89 strains. Despite the emm1 cohort comprising the greatest number of isolates (n = 3,615), it had a lower frequency of nonsynonymous SNP sites (n = 8) than either the emm28 (2,095 isolates and 21 nonsynonymous sites) or emm89 (1,315 isolates and 12 nonsynonymous sites) cohort. Moreover, the nonsynonymous SNP sites among the emm1 isolates differed in distribution compared to their distribution in the emm28 and emm89 isolates, being among the emm1 isolates somewhat less prevalent in the middle (i.e., the transpeptidase domain) and more prevalent toward the 3′ end of pbp2x (i.e., the penicillin-binding protein and serine/threonine kinase-associated [PASTA] domains).
FIG 1 Location of PBP2X amino acid replacements identified among the 7,025 genomes of emm1, emm28, and emm89 clinical isolates. Amino acid replacements identified in multiple strains are highlighted in yellow, with superscripts denoting the number of strains. Replacements identified in both emm28 and emm89 strains are in bold and enclosed in brackets. Replacements associated with reduced susceptibility to one or more of the beta-lactam antibiotics tested under the in vitro conditions analyzed are shown in red. The dimerization, transpeptidase, and PASTA domains are indicated.
An alignment of PBP2X of S. pyogenes, S. pneumoniae, and Streptococcus agalactiae shows that the Gly residue at position 288 and the Met residue at position 342 are conserved among the three species (Fig. 2). Of note, Met342 is located in the conserved SXXK motif, containing the transpeptidase activity catalytic Ser residue.
FIG 2 Aligned streptococcal PBP2X sequences. PBP2X of S. pyogenes strain MGAS5005 (GenBank accession number AAZ51984.1), S. pneumoniae (PBP reference sequence, GenBank accession number WP_050265832), and S. agalactiae (PBP reference sequence, GenBank accession number WP_134808185) were aligned by use of the ClustalW program. To facilitate comparisons, for each species, every 10th amino acid is in red. Amino acids of the consensus sequence are highlighted to indicate conserved domains, as indicated in the inset at the lower right. The three key conserved motifs (SXXK, SXN, and KSGT) of the transpeptidase are shown in red and bold below the aligned sequences.

Phylogenetic analysis of strains of the same emm type with the identical PBP2X amino acid replacement using whole-genome sequence data.

We identified 16 instances in which two or more strains of the same emm type had the identical amino acid replacement (Table 1 and Fig. 1). In general, strains of the same emm type with the identical pbp2x nonsynonymous mutation were identified in only one country, although a few exceptions to this were identified (Table 1; see Discussion). We tested the hypothesis that the strains with the same amino acid change were clonally related. This matter is important to address for public health and basic science reasons, because if these organisms are clonally related, it is unambiguous evidence that they can disseminate successfully to new human hosts and cause infections. Phylogenetic analysis of whole-genome sequence data showed that, with one exception, strains of the same emm type with the identical amino acid replacement are closely related, likely as a consequence of descent from a common progenitor (Fig. 3). Exceptions were the 22 emm28 strains with a Gly600Asp replacement (Fig. 3). These findings indicate multiple independent evolutionary origins of the Gly600Asp polymorphism, that is, multiple episodes of evolutionary convergence. We note that the single strain with the combined Phe599Tyr and Gly600Asp replacement was very closely related to two strains having only the single Gly600Asp change. This phylogenetic relationship suggests that the Phe599Tyr amino acid change was acquired (likely by selection) after the Gly600Asp change occurred in a progenitor. Consistent with this idea, the dual-amino-acid-replacement strain was isolated in 2006, years after the genetically related emm28 strains with only the Gly600Asp replacement were initially found.
FIG 3 Genetic relationships among S. pyogenes emm1, emm28, and emm89 clinical isolates. Phylogenies were inferred by neighbor joining based on core chromosomal SNPs. Isolates with nonsynonymous SNPs in pbp2x are colored according to the amino acid replacements in PBP2X, as shown in the keys. Clades of more closely related strains are shown by shapes (e.g., circles and squares), as indicated. (A) Relationships among 3,615 emm1 strains; (B) relationships among 2,095 emm28 strains; (C) relationships among 1,315 emm89 strains.

Association of PBP2X amino acid replacements with decreased susceptibility to beta-lactam antibiotics.

We next tested the hypothesis that the PBP2X amino acid replacements are associated with decreased susceptibility to beta-lactam antibiotics. Strains were streaked onto tryptic soy agar plates supplemented with 6-ng/ml of penicillin G or 15-ng/ml ampicillin, and the plates were incubated overnight. These concentrations were previously determined to be minimally inhibitory for S. pyogenes (9, 10). Wild-type strains that lacked pbp2x mutations did not grow after overnight incubation in the presence of these beta-lactam antibiotics. In contrast, many strains with pbp2x nonsynonymous mutations grew well on both antibiotic-containing media, including organisms with the Gly288Ser, Met342Ile, Phe599Tyr plus Gly600Asp, Gly600Asp, and Pro601Leu amino acid replacements (Table 1; Fig. 1 and 3). Of note, in contrast to the five strains with the Pro601Leu replacement, the single emm89 strain (MGAS27308) with the Pro601Ser change did not grow in the presence of either antibiotic under the plating conditions tested. Similarly, none of the 10 emm1 strains representing 8 different amino acid replacements grew under the antibiotic conditions tested. The data are consistent with the hypothesis of an association between some naturally occurring pbp2x mutations and decreased susceptibility to these beta-lactams in some genetic backgrounds. We next used Etest strips to determine the MICs for penicillin G and found that many strains with pbp2x mutations had decreased susceptibility to this agent, as tested in this fashion, whereas all 18 wild-type comparator strains lacking pbp2x mutations were fully susceptible (Table 1).
It is well-known that the same PBP2X amino acid replacement can confer divergent phenotypes of susceptibility to different beta-lactam antibiotics. Thus, we next performed MIC susceptibility testing with five additional beta-lactam antibiotics (ampicillin, cefotaxime, cefoxitin, ceftazidime, and meropenem) using the Etest gradient method. We found that, compared to the wild-type control strains, many PBP2X mutant strains had reduced susceptibility to one or more beta-lactam antibiotics (Table 1). The Etest MIC results for penicillin G and ampicillin were confirmed for some strains using broth microdilution and penicillin G and ampicillin agar (Table 1; Fig. 4; see also Fig. S1 in the supplemental material). Of note, strains with the Pro601Leu amino acid change, which occurred in both emm28 and emm89 strains, had the highest MIC measurements with all beta-lactam antibiotics tested except cefoxitin (Table 1). Specifically, the penicillin G MICs for strains with the Pro601Leu change ranged from 23 ng/ml to 32 ng/ml, approximately 4- to 5-fold higher than those for strains with the wild-type PBP2X (Table 1). To unambiguously demonstrate that the Pro601Leu PBP2X amino acid replacement was responsible for the altered MICs, we created an isogenic strain containing the wild-type pbp2x gene in place of the naturally occurring mutant pbp2x allele (encoding Pro601Leu). As expected, the isogenic Pro601 strain (i.e., the PBP2X consensus wild-type engineered derivative strain) was more susceptible to beta-lactam antibiotics than the naturally occurring parental strain with the Pro601Leu substitution (Fig. 4). Also, the strain with both the Phe599Tyr and Gly600Asp amino acid replacements had MIC measurements that were equal to or greater than those for strains with only the Gly600Asp change, suggesting that this dual amino acid replacement may have an additive effect on MICs.
FIG 4 Beta-lactam antibiotic susceptibility assays. (A and B) Shown is the growth of emm89 PBP2X wild-type strain MGAS27556 (left) in comparison with that of PBP2X Pro601Leu amino acid replacement strain MGAS27316 (right) on medium supplemented with 6-ng/ml penicillin G (A) or 15-ng/ml ampicillin (B). (C and D) Graphed is the MIC dilution growth of emm1, emm28, and emm89 PBP2X wild-type and amino acid replacement variant strains in THY broth supplemented with penicillin G (C) or ampicillin (D). (E and F) Graphed is the growth of emm89 PBP2X Pro601Leu replacement strain MGAS27213 and its isogenic PBP2X wild-type engineered derivative in THY broth supplemented with penicillin G (E) or ampicillin (F). All THY broth growth experiments were done in quadruplicate, and the results are given as the mean ± SD. *, significant differences in growth, as determined by Student's t test at a P value of <0.05. OD600, optical density at 600 nm.

Relative location of amino acid changes in the PBP2X three-dimensional structure.

To assess the potential consequence of the identified amino acid replacements on PBP2X, we mapped the location of the changes on a crystal structure available for S. pneumoniae PBP2X (Fig. 5). The structure of this protein has been well studied by several investigators because of its importance in beta-lactam resistance in this common human pathogen (1214, 16). The variant amino acids at positions 342, 599, 600, and 601 mapped to regions known to influence structure-function relationships (20, 21). This determination of influence was derived from overlaying PBP2X from S. pneumoniae with the clinically relevant and structurally similar P. aeruginosa PBP3 (PDB accession number 6UN3) and S. aureus PBP2a (PDB accession number 1VQQ, chain A). Recently, it was discovered that residues on the bottom of the α-8 helix of PBP3 are essential in forming an aromatic pocket (20) comprised of Tyr532 and Phe533. This aromatic pocket is key in binding and stabilizing the side chains of beta-lactam antibiotics. In PBP2X, a similar, conserved aromatic pocket is formed with the neighboring His594 and Tyr595 residues. The amino acid substitutions at residues 599, 600, and 601 that we observed in the clinical S. pyogenes isolates studied here are located directly above this aromatic pocket (Fig. 5). Substitutions at these positions may perturb binding interactions between the beta-lactam antibiotic and PBP2X and thereby decrease the acylation efficiency of the antibiotics, leading to reduced susceptibility. It is noteworthy that the Pro601Leu substitution was associated with decreased penicillin G susceptibility but that the Pro601Ser substitution was not. The serine side chain is relatively small and hydrophilic, whereas that of leucine is larger and hydrophobic, and both differ from proline, a secondary amino acid (i.e., an imino acid). Thus, both serine and leucine could potentially perturb the PBP2X structure and beta-lactam binding interactions but to a different extent, possibly leading to the differences in susceptibility observed.
FIG 5 Location of S. pyogenes PBP2X substitutions relative to the X-ray crystallography structure of PBP2X from S. pneumoniae (PDB accession number 1RP5, chain A). (A) Variant sites influencing structure-function. Illustrated as spheres on the S. pneumoniae PBP2X ribbon diagram are the key amino acid replacement sites associated with reduced beta-lactam susceptibility from the S. pyogenes clinical isolates, with the relative amino acid positions being labeled. Shown in red is the transpeptidase catalytic serine residue (S. pneumoniae residue 337 = S. pyogenes residue 340). The amino acids depicted are those of the S. pneumoniae PBP2X. (B) All variant sites. The relative positions of all 36 amino acid replacement sites observed among the 7,025 sequenced strains are shown as blue spheres, with the catalytic serine being shown in red. (C) Aromatic pocket. Illustrated is a surface representation showing the aromatic pocket proposed to be involved in binding and stabilizing beta-lactam side chains. The His594 and Tyr595 residues lining the pocket are shown in blue.
Similarly, the Met342Ile substitution is noteworthy because residue 342 is located directly within the active-site pocket of PBP2X near the catalytic Ser337. In PBP2X, a methionine cluster is conserved within the active site (15). Thus, the Met342Ile substitution present in the clinical S. pyogenes isolate may disrupt the conserved methionine cluster and thereby perturb binding and acylation of the enzyme by beta-lactam antibiotics. Decreased acylation efficiency, in turn, would explain the reduced beta-lactam susceptibility associated with clinical S. pyogenes isolates containing amino acid variants at position 342. In contrast, the Gly288Ser substitution is not located in a position to directly impact substrate binding or acylation (Fig. 5). Indicative of this, the Cα of residue 288 is 17.5 Å from the Oγ of the catalytic Ser337 in the S. pneumoniae PBP2X structure. Regardless, the substitution could detrimentally impact enzyme dynamics or stability to alter function. A more definitive conclusion on how the Gly288Ser substitution alters enzyme structure and function awaits further work.


Here we report on 137 strains of S. pyogenes from intercontinental sources that have 37 amino acid replacements at 36 sites in the PBP2X protein, some of which correlate with decreased susceptibility to beta-lactam antibiotics under the conditions tested. Importantly, none of the mutations that we identified resulted in resistance in vitro to any of the six beta-lactams studied (as defined by CLSI), and none approached the level for ampicillin or cefotaxime MICs described for the PBP2X substitution Thr553Lys. This substitution evidently conferred an MIC at the CLSI-determined breakpoint for nonsusceptibility to ampicillin. However, isogenic mutant strains were not constructed to prove that the mutant allele of pbp2x was solely responsible for the altered MIC value. This is an important point, because the two strains described by Vannice et al. (3) also contained a Ser79Phe amino acid replacement in the topoisomerase ParC. Substitutions in ParC can confer resistance to fluoroquinolone antibiotics. In principle, substitutions in ParC could produce a slowed growth phenotype potentially contributing to the altered beta-lactam MICs observed. As described by Vannice et al. (3), all five emm43.4 strains that they analyzed also had a Thr236Ala amino acid replacement in a gene annotated “glycoside hydrolase family 25.” The only known enzymatic activity of the family 25 glycoside hydrolases is that of a lysozyme muramidase. Members of this protein family participate in peptidoglycan remodeling, and thus, in principle, the Thr236Ala substitution might also contribute to the altered beta-lactam MIC. Clearly, much more work using isogenic mutant strains is required to deconvolute the role of specific amino acid replacements in these proteins in the observed altered MICs.
Our research was stimulated by the recent description of two clonally related type emm43.4 S. pyogenes strains with the same Thr553Lys amino acid replacement in PBP2X associated with altered MICs of beta-lactam antibiotics (3). We note that the Thr553Lys change likely reflects very recent antimicrobial selection, in view of the described course of treatment of the two host patients. Our work was made possible in part by the availability of 7,025 genome sequences from geographically dispersed strains of types emm1, emm28, and emm89 that we previously generated for molecular pathogenesis, population genomic, and epidemiological purposes. This unique resource permitted us to rapidly identify strains with mutations in the pbp2x gene by bioinformatic methods and subsequently assess the beta-lactam susceptibility phenotypes by standard clinical microbiology methods. Our findings indicate that the decreased beta-lactam susceptibility associated with some PBP2X amino acid polymorphisms in this pathogen is geographically widespread and has arisen multiple times independently over many years in contemporary epidemic clones of serotype emm1, emm28, and emm89 GAS (Fig. 2). The data support the interpretation that the nonsynonymous mutations have been selected by exposure to beta-lactam antibiotics used during treatment of infections caused by S. pyogenes. However, inasmuch as S. pyogenes can be carried asymptomatically in the upper respiratory tract or other anatomic site, it is also possible that selection occurred during antibiotic treatment of an asymptomatic carrier for an infection caused by another organism. The potential for reduced beta-lactam susceptibility to confer an advantage during human infections (either invasive or nonsystemic infection, such as pharyngitis or localized skin and soft tissue infections) or asymptomatic carriage remains untested. More study is required to address these important issues.

Examples of convergent evolution.

Several examples of convergent evolution to decreased beta-lactam susceptibility were identified in this analysis. Four instances of the presence of the same otherwise rare single amino acid polymorphism in strains of different emm types were found. We identified strains of emm28 and emm89 with each of the following amino acid replacements: Gly288Ser, Met342Ile, Gly600Asp, or Pro601Leu. Given that strains of emm28 and emm89 are very distantly related genetically, the only reasonable interpretation is that these polymorphisms arose independently as a consequence of convergent evolution, presumably due to selection following exposure to a beta-lactam antibiotic. Similarly, the occurrence of the Gly600Asp replacement in some emm28 strains that have not shared a recent common ancestor serves as another clear example of convergent evolution in pbp2x. As further evidence of convergent evolution, Chochua et al. also found the Pro601Leu amino acid change in multiple unrelated GAS lineages, including emm4 and emm75 isolates containing it as a single amino acid replacement and emm87 and emm89 isolates containing it in combination with a second substitution (4).

Why is there an apparent difference between emm1 strains and emm28 and emm89 strains?

The majority of strains that we identified with pbp2x nonsynonymous mutations were either type emm28 or type emm89 subclade 3 organisms. In addition, the linear location of PBP2X amino acid replacements in the emm1 strains differed from that in the emm28 and emm89 strains. We believe that there are several factors that may contribute to these differences. First, it is important to note that essentially all emm28 organisms do not produce hyaluronic acid capsule, as a consequence of having an insertion of an adenine nucleotide after nucleotide 219 in an A-T-rich region (7). This single nucleotide insertion severely truncates hasA, whose gene product is required for capsule biosynthesis. Similarly, emm89 organisms of subclade 3 fail to make hyaluronic acid capsule because they lack the hasABC operon, required for capsule biosynthesis (5, 22). The hyaluronic capsule, among other interactions with the host, contributes to the capacity of S. pyogenes to resist phagocytosis. It is possible that a relationship exists between the inability of an S. pyogenes strain to produce hyaluronic acid capsule and the likelihood of generating a strain that is extant and that has PBP2X amino acid changes that result in decreased beta-lactam susceptibility. Second, it is possible that the in vivo regulation and/or expression of pbp2x differs between strains of distinct clonal backgrounds. A third possibility is that, for unknown reasons, emm1 strains with PBP2X amino acid replacements are simply less fit in vivo than emm28 and emm89 strains. A fourth possibility is that the in vivo topology of PBP2X differs between emm1 strains and emm28 and emm89 strains, perhaps due to interaction with other currently unknown proteins. However, all of these ideas are speculative, and more study is required to address these important observations. Finally, we note that the possibilities described above are not mutually exclusive.

Relationship of our findings to those reported for other pathogenic beta-hemolytic streptococci.

Strains of Streptococcus agalactiae and Streptococcus dysgalactiae subsp. equisimilis with decreased susceptibility to beta-lactam antibiotics have been reported (2339). In the case of S. dysgalactiae subsp. equisimilis, four isolates cultured from the blood of three epidemiologically associated patients were reported to be resistant to penicillin and oxacillin (25). Whole-genome sequencing identified nonsynonymous mutations in PBP2X that were thought to be causally involved in the resistance phenotype. In particular, the investigators identified the occurrence of Thr341Pro and Gln555Glu amino acid replacements and noted that these two changes are located close to positions 337, 547, and 557, which are among the more prevalently found variant sites reported for penicillin-resistant S. pneumoniae. As described above, we identified decreased susceptibility in S. pyogenes strains with an amino acid change at position 342 (Met342Ile), and the altered amino acid reported by Vannice et al. was Thr553Lys (3).

Public health implications.

Our population genomic analysis indicates that S. pyogenes strains with nonsynonymous mutations in pbp2x are not exceedingly rare in the emm1, emm28, and emm89 organisms that we studied, being present in approximately 2% of the collection of 7,025 isolates studied. Although this relatively low frequency is fortunate, two facts give us pause. First, the great majority of the isolates that we previously characterized by whole-genome sequencing were cultured from patients with invasive episodes. Given the relative lack of strains causing pharyngitis in this sample, coupled with the well-documented treatment failures occurring among individuals with culture-positive S. pyogenes pharyngitis (4048), it is possible that analysis of large samples of strains from pharyngitis patients will identify a different percentage of organisms with pbp2x mutations associated with altered susceptibility to beta-lactam antibiotics. Second, we identified strains with PBP2X amino acid changes that are clearly clonally related based on phylogenetic analysis of whole-genome data. For example, we recovered clonally related type emm28 organisms with the Gly600Asp replacement from patients in Canada and five different states in the United States, indicating that they can successfully disseminate over geographic distances and cause infections. The same is true for the 27 clonally related strains of type emm1 with the Asp734Gly change, organisms causing invasive infections in Denmark, Sweden, and Iceland between 2002 and 2007 (Table 1). Taken together, our findings stress the importance of renewed efforts to monitor antimicrobial susceptibility rates and values in this pathogen on an ongoing basis, the need to formulate an efficacious human vaccine against S. pyogenes, and the need for expanded vaccine efforts, as noted by many (4951). Importantly, these needs also were highlighted in the report of a symposium held more than 2 decades ago dedicated to the topic of a lack of penicillin resistance in S. pyogenes (1).

What may the future hold?

Although some favor the idea that we are not at the beginning of the end of the universal susceptibility of S. pyogenes to beta-lactam antibiotics (52), we believe that there are multiple reasons to be less sanguine in the long term. First, our data show that several distinct pbp2x mutations associated with decreased susceptibility occur in S. pyogenes strains of multiple emm types. Second, in contrast to the otherwise rare emm43.4 strains reported by Vannice et al. (3), we identified PBP2X amino acid replacements in strains of types emm1, emm28, and emm89, which, in the aggregate, are common causes of S. pyogenes pharyngitis and invasive episodes in many countries. Third, some of the organisms with amino acid changes are clonally related and have been recovered in multiple geographic locations, in some cases many years apart. Thus, if strains with PBP2X amino acid replacements have decreased fitness, at least in some cases the deficit is not sufficient to prohibit the successful dissemination of some clonal progeny to new hosts and the capacity to cause serious human invasive infections. Fourth, the exchange of genetic material between S. pyogenes strains can produce progeny with altered phenotypes, such as enhanced virulence and increased antimicrobial agent resistance. Thus, in principle, there is a risk of gene flow of a mutant pbp2x gene to a susceptible strain, a process that could accelerate the spread of decreased susceptibility to beta-lactams or frank resistance in this global human pathogen.

Concluding comment.

To summarize, we used our library of 7,025 S. pyogenes genome sequences from strains of types emm1, emm28, and emm89 to identify amino acid-altering mutations in pbp2x. Some of the strains with amino acid replacements in PBP2X had decreased susceptibility under the laboratory conditions tested to some beta-lactam antibiotics, including the commonly used penicillin G. Although many pbp2x mutations occurred in only one or two strains, we found that some PBP2X amino acid replacements were present in multiple clonally related strains causing infections many years apart. Decreased susceptibility to beta-lactams in S. pyogenes is geographically widespread and exists in strains of numerically common emm gene subtypes. We recommend that increased basic science and translational research attention be applied to this potentially severe public health threat. For example, the availability of an efficacious human vaccine against S. pyogenes pharyngitis would significantly decrease the use of beta-lactam antibiotic agents globally. We believe that for diagnostic laboratories not currently routinely performing beta-lactam susceptibility testing, it is reasonable to consider doing so, perhaps by measuring the penicillin MIC.


This study was supported in part by the Fondren Foundation, the Houston Methodist Hospital and Research Institute, and National Institutes of Health grants AI139369 and AI146771 (to J.M.M.) and AI32956 (to T.P.).
We thank the dedicated staff of the Houston Methodist Hospital and Research Institute Clinical Microbiology Laboratory (including Oluwatobi Adelanwa, Edevelia Cornelius, Lily Guevara, and Patricia L. Cernoch) for assistance in performing the Etest MIC determinations; Concepcion C. Cantu, Matthew Ojeda Saavedra, Layne Pruitt, and Prasanti Yerramilli for technical assistance; Jari Jalava, Carita Savolainen-Kopra, and Outi Lyytikäinen for expert opinion; Kati Räisänen and Tuula Siljander for emm typing of strains; and the Finnish clinical microbiology laboratories for sending the laboratory notification and strains to THL.

Supplemental Material

File (jcm.01993-19-s0001.pdf)
ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.


Horn DL, Zabriskie JB, Austrian R, Cleary PP, Ferretti JJ, Fischetti VA, Gotschlich E, Kaplan EL, McCarty M, Opal SM, Roberts RB, Tomasz A, Wachtfogel Y. 1998. Why have group A streptococci remained susceptible to penicillin? Report on a symposium. Clin Infect Dis 26:1341–1345.
Ralph AP, Carapetis JR. 2013. Group A streptococcal diseases and their global burden. Curr Top Microbiol Immunol 368:1–27.
Vannice K, Ricaldi J, Nanduri S, Fang FC, Lynch J, Bryson-Cahn C, Wright T, Duchin J, Kay M, Chochua S, Van Beneden C, Beall B. 20 October 2019. Streptococcus pyogenes pbp2x mutation confers reduced susceptibility to beta-lactam antibiotics. Clin Infect Dis doi:
Chochua S, Metcalf BJ, Li Z, Rivers J, Mathis S, Jackson D, Gertz RE, Jr, Srinivasan V, Lynfield R, Van Beneden C, McGee L, Beall B. 2017. Population and whole genome sequence based characterization of invasive group A streptococci recovered in the United States during 2015. mBio 8:e01422-17.
Beres SB, Kachroo P, Nasser W, Olsen RJ, Zhu L, Flores AR, de la Riva I, Paez-Mayorga J, Jimenez FE, Cantu C, Vuopio J, Jalava J, Kristinsson KG, Gottfredsson M, Corander J, Fittipaldi N, Di Luca MC, Petrelli D, Vitali LA, Raiford A, Jenkins L, Musser JM. 2016. Transcriptome remodeling contributes to epidemic disease caused by the human pathogen Streptococcus pyogenes. mBio 7:e00403-16.
Beres SB, Olsen RJ, Ojeda Saavedra M, Ure R, Reynolds A, Lindsay DSJ, Smith AJ, Musser JM. 2017. Genome sequence analysis of emm89 Streptococcus pyogenes strains causing infections in Scotland, 2010–2016. J Med Microbiol 66:1765–1773.
Kachroo P, Eraso JM, Beres SB, Olsen RJ, Zhu L, Nasser W, Bernard PE, Cantu CC, Saavedra MO, Arredondo MJ, Strope B, Do H, Kumaraswami M, Vuopio J, Grondahl-Yli-Hannuksela K, Kristinsson KG, Gottfredsson M, Pesonen M, Pensar J, Davenport ER, Clark AG, Corander J, Caugant DA, Gaini S, Magnussen MD, Kubiak SL, Nguyen HAT, Long SW, Porter AR, DeLeo FR, Musser JM. 2019. Integrated analysis of population genomics, transcriptomics and virulence provides novel insights into Streptococcus pyogenes pathogenesis. Nat Genet 51:548–559.
Nasser W, Beres SB, Olsen RJ, Dean MA, Rice KA, Long SW, Kristinsson KG, Gottfredsson M, Vuopio J, Raisanen K, Caugant DA, Steinbakk M, Low DE, McGeer A, Darenberg J, Henriques-Normark B, Van Beneden CA, Hoffmann S, Musser JM. 2014. Evolutionary pathway to increased virulence and epidemic group A Streptococcus disease derived from 3,615 genome sequences. Proc Natl Acad Sci U S A 111:E1768–E1776.
Gutmann L, Tomasz A. 1982. Penicillin-resistant and penicillin-tolerant mutants of group A streptococci. Antimicrob Agents Chemother 22:128–136.
Sakata H. 2013. Susceptibility and emm type of Streptococcus pyogenes isolated from children with severe infection. J Infect Chemother 19:1042–1046.
Zhu L, Olsen RJ, Nasser W, Beres SB, Vuopio J, Kristinsson KG, Gottfredsson M, Porter AR, DeLeo FR, Musser JM. 2015. A molecular trigger for intercontinental epidemics of group A Streptococcus. J Clin Invest 125:3545–3559.
Gordon E, Mouz N, Duee E, Dideberg O. 2000. The crystal structure of the penicillin-binding protein 2x from Streptococcus pneumoniae and its acyl-enzyme form: implication in drug resistance. J Mol Biol 299:477–485.
Mouz N, Di Guilmi AM, Gordon E, Hakenbeck R, Dideberg O, Vernet T. 1999. Mutations in the active site of penicillin-binding protein PBP2x from Streptococcus pneumoniae. Role in the specificity for beta-lactam antibiotics. J Biol Chem 274:19175–19180.
Mouz N, Gordon E, Di Guilmi AM, Petit I, Petillot Y, Dupont Y, Hakenbeck R, Vernet T, Dideberg O. 1998. Identification of a structural determinant for resistance to beta-lactam antibiotics in Gram-positive bacteria. Proc Natl Acad Sci U S A 95:13403–13406.
Pares S, Mouz N, Petillot Y, Hakenbeck R, Dideberg O. 1996. X-ray structure of Streptococcus pneumoniae PBP2x, a primary penicillin target enzyme. Nat Struct Biol 3:284–289.
Pernot L, Chesnel L, Le Gouellec A, Croize J, Vernet T, Dideberg O, Dessen A. 2004. A PBP2x from a clinical isolate of Streptococcus pneumoniae exhibits an alternative mechanism for reduction of susceptibility to beta-lactam antibiotics. J Biol Chem 279:16463–16470.
Pettersen EF, Goddard TD, Huang CC, Couch GS, Greenblatt DM, Meng EC, Ferrin TE. 2004. UCSF Chimera—a visualization system for exploratory research and analysis. J Comput Chem 25:1605–1612.
Le Breton Y, Belew AT, Valdes KM, Islam E, Curry P, Tettelin H, Shirtliff ME, El-Sayed NM, McIver KS. 2015. Essential genes in the core genome of the human pathogen Streptococcus pyogenes. Sci Rep 5:9838.
Zhu L, Olsen RJ, Beres SB, Eraso JM, Saavedra MO, Kubiak SL, Cantu CC, Jenkins L, Charbonneau ARL, Waller AS, Musser JM. 2019. Gene fitness landscape of group A streptococcus during necrotizing myositis. J Clin Invest 129:887–901.
Sacco MD, Kroeck KG, Kemp MT, Zhang X, Andrews LD, Chen Y. 2019. Influence of the alpha-methoxy group on the reaction of temocillin with Pseudomonas aeruginosa PBP3 and CTX-M-14 beta-lactamase. Antimicrob Agents Chemother 64:e01473-19.
Zapun A, Contreras-Martel C, Vernet T. 2008. Penicillin-binding proteins and beta-lactam resistance. FEMS Microbiol Rev 32:361–385.
Turner CE, Abbott J, Lamagni T, Holden MT, David S, Jones MD, Game L, Efstratiou A, Sriskandan S. 2015. Emergence of a new highly successful acapsular group A streptococcus clade of genotype emm89 in the United Kingdom. mBio 6:e00622-15.
Banno H, Kimura K, Seki T, Jin W, Wachino JI, Yamada K, Nagano N, Arakawa Y. 2018. High isolation rate and multidrug resistance tendency of penicillin-susceptible group B Streptococcus with reduced ceftibuten susceptibility in Japan. Eur J Clin Microbiol Infect Dis 37:1511–1519.
Dahesh S, Hensler ME, Van Sorge NM, Gertz RE, Jr, Schrag S, Nizet V, Beall BW. 2008. Point mutation in the group B streptococcal pbp2x gene conferring decreased susceptibility to beta-lactam antibiotics. Antimicrob Agents Chemother 52:2915–2918.
Fuursted K, Stegger M, Hoffmann S, Lambertsen L, Andersen PS, Deleuran M, Thomsen MK. 2016. Description and characterization of a penicillin-resistant Streptococcus dysgalactiae subsp. equisimilis clone isolated from blood in three epidemiologically linked patients. J Antimicrob Chemother 71:3376–3380.
Gaudreau C, Lecours R, Ismail J, Gagnon S, Jette L, Roger M. 2010. Prosthetic hip joint infection with a Streptococcus agalactiae isolate not susceptible to penicillin G and ceftriaxone. J Antimicrob Chemother 65:594–595.
Kamiya C, Kimura K, Doyama Y, Miyazaki A, Morimoto M, Banno H, Nagano N, Jin W, Wachino J, Yamada K, Arakawa Y. 2015. Ceftibuten-containing agar plate for detecting group B streptococci with reduced penicillin susceptibility (PRGBS). Diagn Microbiol Infect Dis 82:269–273.
Kimura K, Nagano N, Arakawa Y. 2015. Classification of group B streptococci with reduced beta-lactam susceptibility (GBS-RBS) based on the amino acid substitutions in PBPs. J Antimicrob Chemother 70:1601–1603.
Kimura K, Nagano N, Nagano Y, Wachino J, Suzuki S, Shibayama K, Arakawa Y. 2011. Predominance of sequence type 1 group with serotype VI among group B streptococci with reduced penicillin susceptibility identified in Japan. J Antimicrob Chemother 66:2460–2464.
Kimura K, Suzuki S, Wachino J, Kurokawa H, Yamane K, Shibata N, Nagano N, Kato H, Shibayama K, Arakawa Y. 2008. First molecular characterization of group B streptococci with reduced penicillin susceptibility. Antimicrob Agents Chemother 52:2890–2897.
Kimura K, Wachino J, Kurokawa H, Matsui M, Suzuki S, Yamane K, Nagano N, Shibayama K, Arakawa Y. 2013. High cephalosporin resistance due to amino acid substitutions in PBP1A and PBP2X in a clinical isolate of group B Streptococcus. J Antimicrob Chemother 68:1533–1536.
Koide S, Hayashi W, Taniguchi Y, Tanaka H, Kimura K, Nagano Y, Arakawa Y, Nagano N. 2019. Potential effect of selective pressure with different beta-lactam molecules on the emergence of reduced susceptibility to beta-lactams in group B streptococci. Microbiol Immunol 63:65–76.
Longtin J, Vermeiren C, Shahinas D, Tamber GS, McGeer A, Low DE, Katz K, Pillai DR. 2011. Novel mutations in a patient isolate of Streptococcus agalactiae with reduced penicillin susceptibility emerging after long-term oral suppressive therapy. Antimicrob Agents Chemother 55:2983–2985.
Metcalf BJ, Chochua S, Gertz RE, Jr, Hawkins PA, Ricaldi J, Li Z, Walker H, Tran T, Rivers J, Mathis S, Jackson D, Glennen A, Lynfield R, McGee L, Beall B, Active Bacterial Core Surveillance Team. 2017. Short-read whole genome sequencing for determination of antimicrobial resistance mechanisms and capsular serotypes of current invasive Streptococcus agalactiae recovered in the USA. Clin Microbiol Infect 23:574.e7–574.e14.
Nagano N, Kimura K, Nagano Y, Yakumaru H, Arakawa Y. 2009. Molecular characterization of group B streptococci with reduced penicillin susceptibility recurrently isolated from a sacral decubitus ulcer. J Antimicrob Chemother 64:1326–1328.
Nagano N, Nagano Y, Kimura K, Tamai K, Yanagisawa H, Arakawa Y. 2008. Genetic heterogeneity in pbp genes among clinically isolated group B streptococci with reduced penicillin susceptibility. Antimicrob Agents Chemother 52:4258–4267.
Nagano N, Nagano Y, Toyama M, Kimura K, Shibayama K, Arakawa Y. 2014. Penicillin-susceptible group B streptococcal clinical isolates with reduced cephalosporin susceptibility. J Clin Microbiol 52:3406–3410.
Nagano N, Nagano Y, Toyama M, Kimura K, Tamura T, Shibayama K, Arakawa Y. 2012. Nosocomial spread of multidrug-resistant group B streptococci with reduced penicillin susceptibility belonging to clonal complex 1. J Antimicrob Chemother 67:849–856.
Suzuki T, Kimura K, Suzuki H, Banno H, Jin W, Wachino J, Yamada K, Arakawa Y. 2015. Have group A streptococci with reduced penicillin susceptibility emerged? J Antimicrob Chemother 70:1258–1259.
Brook I. 2007. Overcoming penicillin failures in the treatment of group A streptococcal pharyngo-tonsillitis. Int J Pediatr Otorhinolaryngol 71:1501–1508.
Brook I. 2017. Treatment challenges of group A beta-hemolytic streptococcal pharyngo-tonsillitis. Int Arch Otorhinolaryngol 21:286–296.
Gastanaduy AS, Kaplan EL, Huwe BB, McKay C, Wannamaker LW. 1980. Failure of penicillin to eradicate group A streptococci during an outbreak of pharyngitis. Lancet ii:498–502.
Kim KS, Kaplan EL. 1985. Association of penicillin tolerance with failure to eradicate group A streptococci from patients with pharyngitis. J Pediatr 107:681–684.
Musser JM, Gray BM, Schlievert PM, Pichichero ME. 1992. Streptococcus pyogenes pharyngitis: characterization of strains by multilocus enzyme genotype, M and T protein serotype, and pyrogenic exotoxin gene probing. J Clin Microbiol 30:600–603.
Pichichero ME. 1992. Explanations and therapies for penicillin failure in streptococcal pharyngitis. Clin Pediatr (Phila) 31:642–649.
Pichichero ME, Casey JR. 2007. Systematic review of factors contributing to penicillin treatment failure in Streptococcus pyogenes pharyngitis. Otolaryngol Head Neck Surg 137:851–857.
Pichichero ME, Gooch WM, Rodriguez W, Blumer JL, Aronoff SC, Jacobs RF, Musser JM. 1994. Effective short-course treatment of acute group A beta-hemolytic streptococcal tonsillopharyngitis. Ten days of penicillin V vs 5 days or 10 days of cefpodoxime therapy in children. Arch Pediatr Adolesc Med 148:1053–1060.
Pichichero ME, Green JL, Francis AB, Marsocci SM, Murphy AM, Hoeger W, Noriega C, Sorrento A, Gootnick J. 1998. Recurrent group A streptococcal tonsillopharyngitis. Pediatr Infect Dis J 17:809–815.
Dale JB, Batzloff MR, Cleary PP, Courtney HS, Good MF, Grandi G, Halperin S, Margarit IY, McNeil S, Pandey M, Smeesters PR, Steer AC. 2016. Current approaches to group A streptococcal vaccine development. In Ferretti JJ, Stevens DL, Fischetti VA (ed), Streptococcus pyogenes: basic biology to clinical manifestations. University of Oklahoma Health Sciences Center, Oklahoma City, OK.
Steer AC, Carapetis JR, Dale JB, Fraser JD, Good MF, Guilherme L, Moreland NJ, Mulholland EK, Schodel F, Smeesters PR. 2016. Status of research and development of vaccines for Streptococcus pyogenes. Vaccine 34:2953–2958.
Vekemans J, Gouvea-Reis F, Kim JH, Excler J-L, Smeesters PR, O'Brien KL, Van Beneden CA, Steer AC, Carapetis JR, Kaslow DC. 2019. The path to group A streptococcus vaccines: World Health Organization research and development technology roadmap and preferred product characteristics. Clin Infect Dis 69:877–883.
Hanage WP, Shelburne SA. 20 October 2019. Streptococcus pyogenes with reduced susceptibility to beta-lactams: how big an alarm bell? Clin Infect Dis doi:

Information & Contributors


Published In

cover image Journal of Clinical Microbiology
Journal of Clinical Microbiology
Volume 58Number 425 March 2020
eLocator: e01993-19
Editor: Alexander J. McAdam, Boston Children's Hospital
PubMed: 31996443


Received: 3 December 2019
Returned for modification: 23 December 2019
Accepted: 15 January 2020
Published online: 25 March 2020


Request permissions for this article.


  1. population genomics
  2. bioinformatics
  3. antibiotic resistance
  4. whole-genome sequencing
  5. public health



James M. Musser
Center for Molecular and Translational Human Infectious Diseases Research, Houston Methodist Research Institute and Houston Methodist Hospital, Houston, Texas, USA
Clinical Microbiology Laboratory, Department of Pathology and Genomic Medicine, Houston Methodist Research Institute and Houston Methodist Hospital, Houston, Texas, USA
Department of Pathology and Laboratory Medicine, Weill Cornell Medical College, New York, New York, USA
Stephen B. Beres
Center for Molecular and Translational Human Infectious Diseases Research, Houston Methodist Research Institute and Houston Methodist Hospital, Houston, Texas, USA
Clinical Microbiology Laboratory, Department of Pathology and Genomic Medicine, Houston Methodist Research Institute and Houston Methodist Hospital, Houston, Texas, USA
Luchang Zhu
Center for Molecular and Translational Human Infectious Diseases Research, Houston Methodist Research Institute and Houston Methodist Hospital, Houston, Texas, USA
Clinical Microbiology Laboratory, Department of Pathology and Genomic Medicine, Houston Methodist Research Institute and Houston Methodist Hospital, Houston, Texas, USA
Randall J. Olsen
Center for Molecular and Translational Human Infectious Diseases Research, Houston Methodist Research Institute and Houston Methodist Hospital, Houston, Texas, USA
Clinical Microbiology Laboratory, Department of Pathology and Genomic Medicine, Houston Methodist Research Institute and Houston Methodist Hospital, Houston, Texas, USA
Department of Pathology and Laboratory Medicine, Weill Cornell Medical College, New York, New York, USA
Jaana Vuopio
Institute of Biomedicine, University of Turku, Turku, Finland
Turku University Hospital, Department of Clinical Microbiology, Turku, Finland
Hanne-Leena Hyyryläinen
Department of Health Security, Finnish Institute for Health and Welfare, Helsinki, Finland
Kirsi Gröndahl-Yli-Hannuksela
Institute of Biomedicine, University of Turku, Turku, Finland
Karl G. Kristinsson
Department of Clinical Microbiology, Landspitali University Hospital, Reykjavik, Iceland
Faculty of Medicine, School of Health Sciences, University of Iceland, Reykjavik, Iceland
Jessica Darenberg
Public Health Agency of Sweden, Solna, Sweden
Birgitta Henriques-Normark
Department of Microbiology, Tumor and Cell Biology, Karolinska Institutet, Stockholm, Sweden
Department of Laboratory Medicine, Division of Clinical Microbiology, Karolinska University Hospital, Stockholm, Sweden
Steen Hoffmann
Neisseria and Streptococcus Reference Laboratory, Bacteria, Parasites & Fungi, Infectious Disease Preparedness, Statens Serum Institut, Copenhagen, Denmark
Dominque A. Caugant
Division for Infection Control and Environmental Health, Norwegian Institute of Public Health, Oslo, Norway
Andrew J. Smith
College of Medical, Veterinary and Life Sciences, Glasgow Dental Hospital and School, University of Glasgow, Glasgow, Scotland
Scottish Microbiology Reference Laboratory, Glasgow, Scotland
Diane S. J. Lindsay
Scottish Microbiology Reference Laboratory, Glasgow, Scotland
David M. Boragine
Department of Biochemistry and Molecular Biology, Baylor College of Medicine, Houston, Texas, USA
Department of Pharmacology and Chemical Biology, Baylor College of Medicine, Houston, Texas, USA


Alexander J. McAdam
Boston Children's Hospital


Address correspondence to James M. Musser, [email protected].
James M. Musser, Stephen B. Beres, Luchang Zhu, and Randall J. Olsen are equally contributing co-first authors. The order of the authors is by mutual agreement.

Metrics & Citations




If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.

View Options

View options



Get Access

Buy Article
Journal of Clinical Microbiology Vol.58 • Issue 4 • ASM Journals Pay Per View, PPV 25
Journal Subscription
Journal of Clinical Microbiology
ASM members can purchase subscriptions to journals.
Join or renew

Figures and Media






Share the article link

Share with email

Email a colleague

Share on social media

American Society for Microbiology ("ASM") is committed to maintaining your confidence and trust with respect to the information we collect from you on websites owned and operated by ASM ("ASM Web Sites") and other sources. This Privacy Policy sets forth the information we collect about you, how we use this information and the choices you have about how we use such information.
FIND OUT MORE about the privacy policy