Two hundred-eleven GAS isolates were used to test the emergence of persister cells to β-lactam antibiotics. These isolates belong to a convenience collection obtained from infected patients and colonized individuals, in different Brazilian cities, from different clinical sites (Supplementary Table 1). The majority of these isolates were from outpatient cases of symptomatic oropharyngeal infections, and obtained from 1978 to 1997. Clonality were previously analyzed by pulsed-field gel electrophoresis (PFGE) for roughly half of these isolates, which displayed extensive genetic diversity (Melo et al., 2003). These GAS were identified by routine methods and confirmed by latex agglutination tests (Streptococcal Grouping Kit; Oxoid, Basingstoke, Hampshire, United Kingdom). Minimal inhibitory concentration (MIC) for all antimicrobials used in this study, except azithromycin and ethidium bromide (EtBr), was previously determined for this GAS collection (Melo et al., 2003). Since all GAS isolates analyzed were equally able to produce persisters under the experimental model used, to get some insights into the molecular mechanisms associated with antimicrobial persisters in GAS we randomly choose the GAS strain 37–97 among the isolates of this collection whose PFGE patterns were previously determined. This strain showed sequence type (ST) 62, emm 87 and was isolated from symptomatic oropharynx infection case, in 1997, in the outpatient clinic of the Hospital de Puericultura Martagão Gesteira, Rio de Janeiro, RJ. Additionally, nine other GAS isolates were chosen from the convenience collection based on diverse PFGE patterns, different clinical sources, and susceptibility to all antimicrobials tested (Supplementary Table 2). These nine isolates were used as control in the phenotypic tests to detect persisters to validate the results obtained for the representative strain 37–97. Pure cultures of the 211 GAS isolates analyzed were obtained from lyophilized stocks. One tube of each isolate was opened and after reconstitution, cultures was stored at −80°C in brain heart infusion (w/v) with 0.5% (w/v) of yeast extract and 18% (v/v) glycerol.

MIC determinations for azithromycin (Azi; Sigma, St. Louis, MO, United States) and ethidium bromide (EtBr; Sigma) were done using the agar dilution method as recommended by the Clinical & Laboratory Standards Institute (CLSI, 2021) with concentrations ranging from 0.06 to 4 μg/mL and 0.015 to 4 μg/mL; respectively. Two biological experiments were performed (N = 2).

The model used in this study to generate persisters was based on previous work done with Staphylococcus aureus strains (Novais et al., 2020). In this system (here called biofilm like-environment) high bacterial load is inoculated in order to allow the formation of an initial bacterial film on the smooth surface of a cellophane membrane placed onto agar media containing antibiotics to mimic bacterial agglomeration found in some environments such as those encountered in biofilms. Persisters were indirectly detected in the system containing antibiotic by CFU counting (Orman and Brynildsen, 2015; Yu et al., 2019). To prepare the bacterial inoculum, GAS isolates (n = 211) were grown in Todd Hewitt broth containing 0.5% (w/v) of yeast extract (THB-Y) at 37°C/6 h in order to reach the exponential phase. After centrifugation, the pellet was adjusted (∼ 1–2 × 10¹⁰ colony forming unit-CFU/mL) using the same broth. To form a bacterial film, a 100-μL volume (∼ 2–4 × 10⁷ CFU/cm²) was homogeneously spread on the surface of a cellophane membrane placed onto THB-Y agar containing 5% defibrinated sheep blood (BAB) and supplemented with 0.005–8 μg/mL penicillin (Pen; Wyeth-Whitehall Ltda, Itapevi, SP, Brazil) or 0.25–4 μg/mL cephalexin (Cep; Sigma). After 37°C/18 h, persisters were removed from the cellophane membranes at the highest drug concentration in which growth was detected for CFU counting. To test whether defibrinated sheep blood interfered with the analysis, the experiments were also performed in the absence of blood. Antimicrobial susceptible control cells were obtained exactly as described above but using inoculum size adjusted to concentrations recommended by CLSI (∼10⁶CFU/plate; condition that does not allow the generation of persisters). Four biological experiments were performed with two technical replicates each. CFU determinations were carried out for the representative strain 37-97. Two CFU determinations were carried out for each dilution (N = 4).

A proteomic analysis was done to assess protein differential expression between cells grown in biofilm-like environment (condition that allows generation of persisters) and GAS susceptible cells (inoculum size adjusted to ∼10^6sCFU/plate, condition that does not promote antibiotic persistence). Bacterial cells from strain 37–97 were collected from the cellophane membrane, suspended in phosphate buffered saline (PBS) (140 mM NaCl; 2.7 mM KCl; 8 mM Na₂HPO₄, and KH₂PO₄ 1.5 mM; pH 7.2) using vigorous shaking, and adjusted to OD_{600 nm} = 0.4. Pellet was washed twice, resuspended in PBS and lysed with 106 μm beads (Sigma) in a Bio101 Fast Prep system (BioSavant, Qbiogene, Carslbad, CA, United States) using six cycles (5 speeds/30 s pulse). After centrifugation, the protein concentration was estimated using a Qubit 2.0 (Invitrogen Life Technologies, CA, United States), and lysates diluted in sodium dodecyl sulfate polyacrylamide Gel (SDS-PAGE) sample buffer (1:1, v/v) (Laemmli, 1970). Proteins were separated using a 12.5% SDS-PAGE gel electrophoresis, and individual bands were isolated from the gels. All procedures used for the treatment of gel slices and trypsin digestion were performed as previously described (Shevchenko et al., 1996). The resulting peptides were desalted using an in-house reverse-phase microcolumn (POROS R2 resin, Applied Biosystems, Carlsbad, CA, United States) and dried by vacuum centrifugation (Rodrigues et al., 2011). Peptides were solubilized in 20 μL of 0.1% (v/v) formic acid (FA), and 10 μL were injected into a trap column (Opti-Pak C18, Waters, Milford, MA, United States). Liquid chromatography separation was performed using a reverse-phase capillary column (nanoEase C18, 100 mm × 100 μm, Waters) connected to a nano-HPLC system (Waters UPLC, Waters). The eluted peptides were introduced into an ESI-Q-TOF-MS/MS (Q-TOF Micro, Waters) controlled by MassLynx software (Version 4.1, Waters). Mass spectra (MS) were collected in the 50–2,000 m/z range, and the three most abundant ions (charges +2, +3, and +4) were submitted for collision-induced dissociation (CID) using argon gas at 13 psi and 18–45 V. The raw data were converted to a peak list using the ProteinLynx Global software (version 4.0, Waters). Protein identification was considered valid if at least one peptide with minimum of 10 amino acids was observed with a maximum error tolerance of 50 ppm and Mascot score ≥ 46 (p ≤ 0.05). The GenBank (Acc) access number, locus tag, and gene and protein names were determined using BLASTp.¹ In addition, Uniprot BLAST analysis² was performed in order to identify homologs in S. pyogenes MGAS10750. Only e-values ≤ 1.0 e–3 were considered in the database search.

The increase in ethidium bromide (EtBr) MIC values is highly sensitive and specific in identifying efflux-proficient strains in S. aureus (Patel et al., 2010). Therefore, we evaluated the occurrence of EtBr-refractory cells in the biofilm-like environment. Different EtBr concentrations (0.015–4.0 μg/mL) were added to BAB agar that was covered with cellophane membranes. High-bacterial load was placed onto the surface of cellophane membranes to produce a biofilm-like environment as described before. After 18 h incubation (37°C), GAS cells were recovered from the cellophane membranes at the highest EtBr concentration in which growth was detected for CFU determinations. Controls were performed exactly as above but with susceptible cells (∼10⁶ CFU/plate). Four biological experiments were performed with two technical replicates each. CFU determinations were carried out for the representative strain 37–97. Two CFU determinations were carried out for each dilution (N = 4).

These experiments were performed to assess if the persistence observed for GAS cells was associate to β-lactams only, or to universal mechanisms as those involving efflux pumps. The ability of GAS strain 37–97 to become refractory to different antimicrobials was tested in concentrations ranging from 0.01 to 4 μg/mL erythromycin (Ery; Sigma), 0.06–4 μg/mL azithromycin (Azi; Sigma), 0.01–1 μg/mL clindamycin (Cli; Sigma), 0.25–16 μg/mL chloramphenicol (Chl; Sigma), or 0.125–16 μg/mL tetracycline (Tet; Sigma). Plates were examined after incubation for 18 h at 37°C. Controls were also performed exactly as above but with susceptible cells (∼10⁶ CFU/plate). Additionally, for control purposes, these experiments were also done with additional nine GAS isolates (Supplementary Table 2), using the highest concentration of antibiotic in which bacterial growth was detected for the representative strain 37–97. For each antimicrobial tested, two to six biological experiments were performed with two technical replicates each. CFU determinations were carried out for the representative strain 37–97. Two CFU determinations were carried out for each dilution. Ery (N = 2), Azi (N = 4), Cli (N = 6), Chl (N = 6), Tet (N = 6).

Because CCCP inhibits proton motive force and also the transcription of some transport associated genes (Baron and Rolain, 2018), the ability of this compound to inhibit the formation of persister cells was tested. The CCCP (Sigma) at 100 μM final concentration (Clancy et al., 1996) was incorporated to BAB (covered or not with a cellophane membrane) supplemented with 8 μg/mL penicillin (MIC = 0.01 μg/mL), 4 μg/mL cephalexin (MIC = 0.5), 4 μg/mL erythromycin (MIC = 0.12 μg/mL), 4 μg/mL azithromycin (MIC = 0.12 μg/mL), 1 μg/mL clindamycin (MIC = 0.01 μg/mL), 16 μg/mL chloramphenicol (MIC = 1.0 μg/mL), 16 μg/mL tetracycline (MIC = 0.12 μg/mL), or 4 μg/mL EtBr (MIC = 0.06 μg/mL). BAB plates without CCCP were used to control bacterial growth. High bacterial load was placed onto the surface of cellophane membranes or uncovered BAB plates. After 18 h incubation (37°C), all plates were examined and GAS cells recovered from the cellophane membranes at the highest drug concentration in which growth was detected for CFU determinations. Controls were performed exactly as above but with susceptible cells (∼10⁶ CFU/plate). Three biological experiments were performed with two technical replicates each for the representative strain 37–97. Two CFU determinations were carried out for each dilution (N = 3).

GAS persister cells of the strain 37–97 recovered from the cellophane membranes covering BAB plates with 8 μg/mL penicillin were subjected to successive passages (up to 500 generations) on BAB without antibiotics. After passaging, bacterial growth was adjusted to concentrations recommended by CLSI (∼10⁶ CFU/plate), and the penicillin MIC was determined using the agar dilution method (CLSI, 2021). Two biological experiments were performed (N = 2).

For total DNA preparation, penicillin-persister (8 μg/mL penicillin plates; MIC 0.01 μg/mL) and -susceptible cells of the strain 37–97 were recovered from cellophane membranes. An aliquot of the cell suspension was inoculated in THB-Y (1:200 dilution). After incubation (37°C/18 h), DNA was obtained using the Wizard Genomic DNA Purification Kit (Promega; Madison, WI, United States). Genomic libraries were prepared using the Nextera XT kit (Illumina, San Diego, CA, United States) and sequenced on an Illumina HiSeq (125 pb reads). Reads were trimmed using BBDuk Trimmer (version 1.0) and genome assembly was carried out using Newber v3.0 (Margulies et al., 2005). Scaffolds were aligned against a reference genome (S. pyogenes strain NGAS743; Acc: CP007560) using cross match (version 0.990329).³ Intra-scaffold and inter-scaffold gaps resulting from repetitive sequences were resolved by in silico gap filling. Any remaining gaps in the genomic sequence from penicillin-persister cells of the 37–97 strains (37–97P) were filled with “N” with estimated sizes based on the complete sequence of the susceptible cells of the strain 37–97 (37–97S). The sequenced genomes were annotated using RAST 2.0v (Overbeek et al., 2014). Taxonomic analysis was performed by calculating average nucleotide identity (ANI) for whole genomes using OrthoANIu tool.⁴ Multi locus sequence typing (MLST) was performed for the genome sequences using MLST 2.4.0 software.⁵ Differences in single nucleotide polymorphisms (SNPs) between samples 37–97S and 37–97P were evaluated using cross match with parameter discrep lists. The generated list was compared to the Newbler assembly ace file and genome annotation. SNPs were verified by resequencing on an ABI 3730 DNA Analyzer (Life Technologies—Applied Biosystem; Carlsbad, CA, United States). Reactions were performed using the BigDye Terminator v3.1 Cycle Sequencing Kit in 36-cm capillaries with POP7 polymer according to the manufacturer’s instructions.

Total RNA from penicillin-persisters and -susceptible cells obtained from 37–97 strain was prepared from a suspension of cells directly recovered from cellophane membranes as described in the item “Development of GAS-persister cells to β-lactams.” The RNeasy Mini kit (Qiagen; Germantown, MD, United States) was used for RNA preparation that was quantified by a Qubit 2.0 Fluorometer (Thermo Fisher Scientific Brasil; São Paulo, SP, Brazil). RNA quality was analyzed by gel electrophoresis. For some experiments, gene expression was also performed in presence of 100 μM CCCP. To test the effect of clindamycin in the expression of the efflux-associated locus MGAS10750_Spy1819, total RNA was prepared from GAS persister cells recovered from cellophane membranes on BAB plates containing 1 μg/mL clindamycin (MIC = 0.01 μg/mL). The real-time quantitative reverse transcriptase PCR (real-time RT-qPCR) was performed using Power SYBR Green RNA-to-CT^TM 1-Step Kit (Applied Biosystems) as recommended (“Guide to Performing Relative Quantitation of Gene Expression Using Real-Time Quantitative PCR”; Applied Biosystems). The rRNA 16S gene was used as an endogenous control. The calibrator sample was total RNA from susceptible cells of strain 37–97. The reaction was performed in a Step One^TM Real Time PCR System (Applied Biosystems). Data were analyzed using Step One Software 2.2 (Applied Biosystems). All primers were validated as recommended in the cited guide and listed in Supplementary Table 3. Three biological replicates were performed with three technical replicates each (N = 3).

To analyze the quantity of persister cells recovering in presence of β-lactams and other antimicrobial classes, one way ANOVA was applied followed by post hoc Tukey’s test for multiple comparisons. Two-tailed unpaired Student’s t-test was used to analyze most the binary experiments of gene expression. To analyze the hypothesis that the expression of efflux pump-associated genes increases in the persister GAS cells, one-tailed unpaired t-test was performed. All statistical tests were calculated using GraphPad Prism version 9.2.0 for Windows (GraphPad Software, La Jolla, CA, United States). In addition, to confront the null hypothesis, Scaled Jeffreys–Zellner–Siow (JZS) Bayes Factor for two-samples t-test was calculated to test the alternative hypothesis for r = 0.707 (Rouder et al., 2009).

Despite the susceptible MIC values for penicillin (MIC range = 0.0025–0.02 μg/mL; MIC50 and MIC90 = 0.01 μg/mL), persisters were detected for the 211 GAS isolates in all penicillin concentrations tested including those far above MIC and as high as 8 μg/mL. To observe a possible influence of defibrinated sheep blood on the formation of persisters, GAS strain 37–97 was inoculated on BAB plates with and without blood supplementation, both containing 8 μg/mL penicillin, covered or not with cellophane membranes. Persister cells were equally formed when high bacterial load was inoculated. The average detection of persister corresponded to 2.7% (p = 0.0022) of the total cell population grown in the absence of penicillin (6.0 ± 2.4 × 10¹⁰ CFU/mL) (Figure 1). When uncovered BAB plates were examined, persisters formed almost invisible (very tiny) hemolytic colonies, which returned to the normal size after passage in fresh media without antibiotics. Similar to the results obtained for penicillin, persisters could also arise on BAB plates containing 4 μg/mL cephalexin (MIC50 and MIC90 = 0.5 μg/mL). The mean percentage of persisters for 4 μg/mL cephalexin was 1.8% (p = 0.0016) of the total cell population grown in the absence of the drug (6.0 ± 2.4 × 10¹⁰ CFU/mL) (Figure 1).

Drug susceptibility could be reverted when persister cells were submitted to serial passaging on BAB plates without penicillin, with the antibiotic persistent cells returning to their original state of drug susceptibility (MIC = 0.01 μg/mL). To assess whether this persistence phenotype was actually induced by the biofilm-like environment or due to preexistent heterogeneous resistant subpopulations present in the high bacterial load provided by the heavy inoculum size (∼ 1–2 × 10⁹ CFU/100 μL), this inoculum was divided in 100 parts. To each part, 99.9 μL of THB-Y was added and the total 100 μL inoculated onto a cellophane membrane on the BAB plate containing 8 μg/mL penicillin. To control this experiment, the total inoculum (∼ 1–2 × 10⁹ CFU/100 μL) was also inoculated onto a cellophane membrane on the BAB plate with 8 μg/mL penicillin. After 18 h at 37°C, persisters were only generated in the environment of cell agglomeration of the control. No growth was detected in the 100 plates inoculated with part of the inoculum, clearly ruling out the presence of heteroresistant subpopulations in the GAS culture.

Additionally, DNA samples from penicillin-persister (37–97P) and susceptible cells of the strain 37–97 (37–97S) underwent whole-genome sequencing (WGS). Both genomes have a GC content of 38.5% and 1.92 Mb in size. More details on the genome attributes for 37–97S (Acc: CP041408.1) and 37–97P (Acc: CP041615.1) are listed in the Supplementary Table 4. Both sequences were classified as ST62 by the MLST software. To calculate the ANI value, we used the genome sequence of a ST62 S. pyogenes, strain NGAS743, available in the GenBank (Acc: CP007560.1). The ANI value was 99.95% (coverage 37–97S = 78.66% and coverage NGAS743 = 78.83%). This value was higher than the optimal genome-wide ANI threshold for species delineation (ANI 95%; coverage 70%). WGS alignments generated in MAUVE showed high identity and perfect synteny of collinear blocks (Figure 2A). There was also no difference in the absence or presence of mobile genetic elements, genomic islands, or unique genes in the persister cells of the strain 37–97 (37–97P) compared with that of susceptible ones (37–97S). The ANI value for the genomes of 37–97S and 37–97P was 99.99% (coverage 37–97S = 99.89% and coverage 37–97P = 99.89%). Despite some differences in SNPs observed in the WGS, these could not be confirmed by Sanger resequencing of these regions, thus mutations were not associated with the emergence of persisters (Figure 2B) ruling out the phenomenon known as SCVs.

A total of 61 proteins were only detected in bacterial cells recovered from the biofilm-like environment of the strain 37–97, a condition that led to the formation of antimicrobial persisters (Supplementary Table 5). The most remarkable feature was the low frequency of L ribosomal proteins (LRP) in these GAS cells (3.3%). Seventy-nine proteins were only detected when GAS strain 37–97 was grown using an inoculum size recommended for MIC determination (Supplementary Table 6), a condition for susceptibility. However, the most frequently detected proteins under this condition were the LRP proteins (31.6%), which play essential roles in ribosome assembly and are crucial for protein synthesis and cell growth (Figure 3). These data clearly suggest a decrease in growth activity for cells grown in biofilm-like environments. Some multidrug resistance (MDR) efflux pump components were only detected under condition of cell agglomeration, including a protein associated with the periplasmic component of the efflux system that belongs to the root-nodulation-cell-division (RND) family (Uniprot access: Q1J790). Multiple sugar transport ATP-binding protein MsmK (Uniprot access: Q1J4L0) and the multidrug resistant ABC transporter ATP-binding and permease protein (Uniprot access: Q1J8L9) were also observed under this condition (Supplementary Table 5). A total of 128 proteins were detected in both conditions (Supplementary Table 7).

Efflux pump substrates (EtBr and different classes of antimicrobials) were used to assess the role of efflux pump activity in the formation of persisters. The MIC value of strain 37–97 for EtBr was 0.06 μg/mL. However, in the condition used to allow persister formation, growth was observed at concentrations of EtBr as high as 4 μg/mL, possibly indicating intense efflux activity. The percentage of EtBr-refractory cells recovered at concentration of 4 μg/mL was about 6% (p = 0.001) of the GAS cell population grown in biofilm-like environments in the absence of EtBr (5.5 ± 2.0 × 10¹⁰ CFU/mL) (Figure 4A).

Our data show that persisters were formed not only for the GAS representative strain 37–97 but similarly for the nine additional strains used as control, independent on the antibiotic classes analyzed, demonstrating that this phenomenon is a common feature in S. pyogenes. Persister cells were generated at MIC levels and at concentrations as high as 4 μg/mL erythromycin (MIC = 0.12 μg/mL), 4 μg/mL azithromycin (MIC = 0.12 μg/mL), 1 μg/mL clindamycin (MIC = 0.01 μg/mL), 16 μg/mL chloramphenicol (MIC = 1 μg/mL), and 16 μg/mL tetracycline (MIC = 0.12 μg/mL). The percentage of persisters recovered for 37–97 strain, considering all antimicrobials tested, ranged from 0.32 to 4.62% (p < 0.001) of the cell population grown in the absence of antimicrobials (5.5 ± 2.0 × 10¹⁰ CFU/mL) (Figure 4A).

Since the resistance-nodulation-division (RND) family of efflux pumps was one of the drug/proton antiporters detected in the proteome performed with cell grown under agglomeration condition, we used the pump inhibitor CCCP to dissipate the proton-motive force. Control plates with CCCP (100 μM) without antibiotic caused no effect on bacterial growth of 37–97. Despite the inhibition of chloramphenicol and clindamycin persisters by CCCP (Figure 4B), this compound did not inhibit the generation of persisters by β-lactams or other antimicrobials tested.

Of the 15 genes analyzed that were associated with the efflux pumps, seven showed some levels of upregulation in penicillin-persister cells compared with those of susceptible GAS cells (Figure 5A). Among these, genes of an operon associated with efflux pumps of the RND family showed increases of ≥4-fold, which included MGAS10750_Spy1817 (gene product: ABC transporter ATP binding protein; p = 0.0156), MGAS10750_Spy1818 (gene product: ABC transporter permease protein; p = 0.0088), and MGAS10750_Spy1819 (gene product: periplasmic component of efflux system, p < 0.001). An increase in transcripts >4-fold was also observed for a gene product annotated as belonging to a major facilitator superfamily, the multidrug resistance protein B (MGAS10750_Spy0495) (p = 0.04). Another gene upregulated was a homolog of the multiple sugar transport ATP-binding protein MsmK (MGAS10750_Spy1776), which displayed a 2.2-fold increase in expression levels. Despite this difference was not very expressive (p = 0.1490), protein of this same family was detected only on the proteome of cells grown in the biofilm-like environment. The loci MGAS10750_Spy0043 and MGAS10750_Spy1633 (norA homolog) showed about twofold increase (p < 0.001 and p = 0.0031, respectively, Figure 5A).

Because clindamycin was one of the antibiotics completely inhibited by CCCP, we also investigated the effect of this antibiotic in the overexpression of MGAS10750_Spy1819, which is part of the ABC transport operon. Our data showed an overexpression of about ninefold in these gene transcripts. It was observed that CCCP had simultaneously affected the transcript levels of this ABC operon and two gene homologs to ihk (MGAS10750_Spy1815) and irr (MGAS10750_ Spy1816) encoding the two-component regulatory system (TCS) Ihk/Irr, which are adjacent to and upstream this operon. Our data also demonstrated that both the operon and TCS were downregulated in the presence of CCCP (ABC operon: p = 0.03, p = 0.021, p < 0.001 and ihk/irr: p = 0.0057, p = 0.0389, respectively, Figure 5B). Similar to the downstream genes belonging to the ABC operon, ihk/irr homologs also displayed increased levels of transcripts (fourfold) for persisters formed in the absence of CCCP (p < 0.001 and p = 0.01), compared with the susceptible GAS cells (Figure 5B). These data suggest that the Ihk/Irr system could be acting as a regulator of this operon. Indeed, consistent with an increase in pump activity, genes (MGAS10750_Spy1765 and MGAS10750_Spy1120) annotated as belonging to the MarR and GntR families (pump negative transcriptional regulators) were downregulated in the persister cells (p = 0.0032 and p = 0.0137; Figure 5C).

Additionally, the expression of genes associated with protein biosynthesis and cell growth/division were evaluated. For all these genes, the transcript levels decreased, but for bcaT homolog (gene product: branched-chain-amino acid aminotransferase) this decrease was less than twofold. The guaA homolog (gene product: GMP synthesis [glutamine hydrolyzing]), which is involved in the GTP pathway, was twofold down-regulated (p = 0.016). Decreased expression was also observed for relA (gene product: GTP pyrophosphokinase; p = 0.003) and typA (gene product: GTP-binding protein TypA/BipA; p = 0.0314). Finally, the ftsA homolog, which is essential for cell division, was reduced threefold (p = 0.0187) (Figure 5D).

Among the genes associated with the stress conditions studied, which includes some genes related to oxidative stress, the majority was down-regulated in penicillin-persisters. An increase was only observed for a dpr homolog (gene product: hydrogen peroxide resistance regulator), which was about twofold (p = 0.03) more expressed compared with the susceptible GAS cells (Figure 5E). Finally, we examined the expression of three genes homologous to toxin-antitoxin (TA) systems found in the genome of S. pyogenes strain 37–97. Increased expression was observed for the hicA/B homologs (2.8- and 3.5-fold increase, respectively; p = 0.007 and p = 0.017, respectively) for persisters (Figure 5F).