The Dlt and LiaFSR systems derepress SpeB production independently in the Δpde2 mutant of Streptococcus pyogenes

The second messenger molecule, c-di-AMP, plays a critical role in pathogenesis and virulence in S. pyogenes. We previously reported that deleting the c-di-AMP phosphodiesterase gene pde2 severely suppresses SpeB production at the transcriptional level. We performed transposon mutagenesis to gain insight into the mechanism of how Pde2 is involved in SpeB regulation. We identified one of the genes of the dlt operon, dltX, as a suppressor of the SpeB-null phenotype of the Δpde2 mutant. The dlt operon consists of five genes, dltX, dltA, dltB, dltC, and dltD in many Gram-positive bacteria, and its function is to incorporate D-alanine into lipoteichoic acids. DltX, a small membrane protein, is a newly identified member of the operon. The in-frame deletion of dltX or insertional inactivation of dltA in the Δpde2 mutant restored SpeB production, indicating that D-alanylation is crucial for the suppressor phenotype. These mutations did not affect the growth in lab media but showed increased negative cell surface charge and enhanced sensitivity to polymyxin B. Considering that dlt mutations change cell surface charge and sensitivity to cationic antimicrobial peptides, we examined the LiaFSR system that senses and responds to cell envelope stress. The ΔliaR mutation in the Δpde2 mutant also derepressed SpeB production, like the ΔdltX mutation. LiaFSR controls speB expression by regulating the expression of the transcriptional regulator SpxA2. However, the Dlt system did not regulate spxA2 expression. The SpeB phenotype of the Δpde2ΔdltX mutant in higher salt media differed from that of the Δpde2ΔliaR mutant, suggesting a unique pathway for the Dlt system in SpeB production, possibly related to ion transport or turgor pressure regulation.

The model organism used in this study, S. pyogenes or Group A Streptococcus (GAS) is an obligate human pathogen that causes diverse diseases from mild superficial infections to severe invasive, toxigenic, or post-streptococcal autoimmune sequelae (Walker et al., 2014).GAS is still a significant public health concern in developed and developing countries.Approximately 700 million people worldwide suffer from pharyngitis caused by GAS each year.Inadequate treatment or repeated GAS infections develop non-suppurative autoimmune sequelae, acute rheumatic fever (ARF), which can further damage the heart and cause rheumatic heart diseases (RHD) showing high mortality (Carapetis et al., 2005).Approximately 320,000 deaths occurred in 2015 globally due to RHD (Watkins et al., 2017).S. pyogenes expresses an array of cell wall-associated and secreted virulence factors essential for causing various GAS diseases.Despite the long-standing knowledge of GAS diseases, an effective commercial vaccine against GAS is still unavailable.
In S. pyogenes, c-di-AMP regulates cellular activities and virulence factor expression, but the underlying mechanisms are largely unknown.Our previous studies have shown that misregulation of c-di-AMP homeostasis by the deletion of the cdi-AMP synthase gene dacA or a phosphodiesterase gene pde2 leads to the loss of the transcription of the virulence factor SpeB (Fahmi et al., 2019;Faozia et al., 2021).The SpeB null phenotype observed in the DdacA mutant is mediated via the regulation of the potassium transporter KtrAB (Faozia et al., 2021).However, the mechanism by which the Dpde2 mutant exhibits the SpeB-null phenotype remains unknown.Through transposon mutagenesis, the first gene in the dlt operon, dltX, was identified as a suppressor of the SpeB-null phenotype of the Dpde2 mutant.The primary function of the dlt operon is to incorporate D-alanine ester into teichoic acids, resulting in a decreased negative surface charge and increased resistance against the host's cationic antimicrobial peptides (CAMPs) (Peschel et al., 1999;Kamar et al., 2017).
Given the role of the dlt operon in modulating cell surface charge, we also investigated whether or not cell envelope stress influences SpeB production in the Dpde2 mutant.Gram-positive bacteria possess a conserved cell envelope stress response regulatory system called LiaFSR, which senses and responds to cell envelope stress induced by CAMPs (Lin et al., 2020).LiaFSR influenced SpeB production in the Dpde2 mutant.However, our data revealed that the Dlt and LiaFSR systems independently regulate SpeB production in the Dpde2 mutant.

Results
The Dlt system impacts the expression of the SpeB virulence factor in the Pde2deficient mutant of S. pyogenes Our previous study demonstrated that deleting a c-di-AMP phosphodiesterase gene, pde2, abolished SpeB production at the transcriptional level in S. pyogenes (Fahmi et al., 2019).We employed transposon mutagenesis to identify potential genes involved in the regulation of SpeB production in the Dpde2 mutant.We screened ~3,000 colonies and isolated 24 transposongenerated mutants that showed SpeB activity similar to the wild type.DNA sequencing was performed using primers binding to a transposon sequence to determine the transposon insertion sites of these mutants.This process was successful for 17 mutants.Interestingly, all 17 mutants had a transposon insertion either in the promoter region of or in the dltX gene.These mutants were not clones because they all have different transposon insertion sites.The dlt operon is highly conserved in Gram-positive bacteria, encoding gene products that incorporate D-alanine ester into teichoic acid molecules (Peschel et al., 1999;Neuhaus and Baddiley, 2003).In S. pyogenes, the dlt operon comprises six genes, dltXABCDE.The first gene, dltX, encodes a membrane-associated small protein consisting of 47 amino acids (Figure 1).To confirm that dltX inactivation is responsible for derepression of the SpeB production in the Dpde2 strain, we deleted dltX in the Pde2-deficient mutant and evaluated the SpeB activity of this Dpde2DdltX mutant.Like the transposongenerated mutants, the dltX deletion in the Dpde2 background restored SpeB activity comparable to that of the wild type (Figures 2A, B).To investigate further, we disrupted dltA by inserting a plasmid in the Dpde2 background and evaluated its SpeB activity.The resultant Dpde2WdltA mutant does not express all the genes downstream of dltA in the operon due to a polar effect.We chose dltA because dltX is too small to use this gene disruption method.The strain also produced SpeB at a level equivalent to that of the Dpde2DdltX mutant (Figure 2B).When we added the dltX gene back to the Dpde2DdltX, the SpeB phenotype of the Dpde2DdltX(pdltX) strain was almost the same as that of the Dpde2 mutant (Figure 2D).These results indicate that the Dlt system affects SpeB production in the Dpde2 mutants.The single gene deletion mutants, the DdltX mutant and WdltA mutant, showed SpeB activity similar to that of the wild type (Figure 2C).

Deletion of the dltX gene increases the negative surface charge of S. pyogenes
Previous research indicates that D-alanylation defects in lipoteichoic acids (LTA) significantly alter the surface charge of GAS (Kristian et al., 2005).In this study, we investigated the role of DltX, a new member of the dlt operon in GAS, in modulating cell surface charge.The surface charges of strains were examined using the cationic protein cytochrome c.The DdltX and Dpde2DdltX mutants bound more cytochrome c than the wild-type strain (Figure 3), indicating that the absence of DltX leads to an increase in negative surface charge.The amount of the cytochrome c bound to the Dpde2 mutant was a little bit higher than that bound to the wild type, but the difference was much smaller than those between the DdltX mutants and the wild type (Figure 3), suggesting that the absence of Pde2 minimally affects the expression of the dlt operon.
Deleting dltX causes increased sensitivity to the cationic antimicrobial peptide polymyxin B D-alanylation carried out by the Dlt system decreases the affinity of cationic antimicrobial peptides (CAMPs) by reducing Inactivation of the dltX or dltA gene restores the SpeB activity of the Dpde2 mutant.The activity of the secreted protease SpeB is shown on protease indicator plates.Strains were grown overnight and spotted (2µl) onto protease indicator agar plates after serial dilution.SpeB activity displays a clear zone around the spotted cells after incubation.The strains' names are shown above each image, and the dilution degrees of the spotted cultures are indicated at the left side of the first image.Plates were incubated anaerobically at 37°C for 24 -48 h.Since pLZ12Km has the same kanamycin resistance gene as the transposon used for the screening, it was used as the control for kanamycin addition to the media.To construct the dltXcomplemented plasmid, pdltX, the p7INT plasmid was used.The following strains were tested: the wild type (WT), pde2 deletion mutant (DPde2), a transposon-generated mutant (DPde2TnDltX), dltX deletion mutant (DDltX), pde2 and dltX double deletion mutant (DPde2DDltX), dltA insertional disruption mutant (WDltA), pde2 deletion and dltA insertional disruption mutant (DPde2WDltA), and dltX-complemented Dpde2DdltX strain (DPde2DDltX(pDltX).This image is representative of many. the negative surface charge on bacterial cells (Peschel et al., 1999;Kristian et al., 2005).We investigated the effect of dltX gene deletion in S. pyogenes on the resistance to a CAMP polymyxin B (PMB). S. pyogenes strains were cultured with varying concentrations of PMB, and the minimum inhibitory concentrations (MICs) of PMB were measured.The MIC of PMB for both the wild type and Dpde2 mutant was 50 µg/ml.In contrast, the MIC for the strains with dltX deletion, the DdltX and Dpde2DdltX mutants, decreased to 10 µg/ml.These results indicate that the dltX deletion in S. pyogenes increases the susceptibility of the bacteria to PMB.

LiaFSR influences SpeB production in the Dpde2 mutant
The LiaFSR gene regulatory system, which is composed of a membrane-bound repressor protein (LiaF), a sensor kinase (LiaS), and a response regulator (LiaR), functions to detect and responds to cell envelope stress induced by CAMPs (Arias et al., 2011;Lin et al., 2020).Given that the degree of D-alanylation of teichoic acids affects cell envelope charge and both the Dlt and LiaFSR systems respond to CAMPs, we investigated the role of the LiaFSR system in SpeB regulation in the Dpde2 mutant.We deleted the response regulator liaR in the Dpde2 background and measured SpeB activity.The Dpde2DliaR mutant derepressed SpeB production like the Dpde2DdltX mutant, although at a lower level (Figure 4).This result indicates that LiaFSR also influences SpeB production in the Dpde2 mutant.
The Dlt system and LiaFSR are not linked in the speB regulation in the Dpde2 mutant Since SpeB production was also influenced by LiaFSR in the Dpde2 mutant (Figure 4), and defects in D-alanylation may result in the cell envelope stress by altering cell surface charge, we investigated if SpeB activity restoration by dlt mutation in the Dpde2 background was through the LiaFSR system.LiaFSR responds to cell envelope stressors by regulating the expression of spxA2 in Gram-positive bacteria (Baker et al., 2020;Sanson et al., 2021).For example, the treatment of the cell wall stressor vancomycin (0.5 µg/ml) stimulates LiaFSR, which enhances the transcription of spxA2 in S. pyogenes (Lin et al., 2020).We measured spxA2 transcript levels in cells with or without vancomycin treatment (Figure 5).As expected, all the strains, except the liaR deletion mutants, showed an increased amount of spxA2 transcript when treated with vancomycin (Figure 5A).Moreover, significant changes of the spxA2 transcript level in the Dpde2, DdltX, and Dpde2DdltX mutants compared to that of the wild type was not observed when they were treated with vancomycin (Figure 5B).A negative control, tetracycline (1 µg/ml), a protein synthesis inhibitor, showed no significant change (less than twofold) of the spxA2 transcript level.These results suggest that neither c-di-AMP levels in cells nor D-alanylation of teichoic acids affect the expression of spxA2.
The deletion of dltX, not liaR, in the Dpde2 mutant increases the level of cellular c-di-AMP The amounts of c-di-AMP levels of strains were measured.Both the dltX and liaR deletion in the wild type did not change cellular cdi-AMP level, but all pde2 deletion mutants, the Dpde2, Dpde2DdltX, and Dpde2DliaR mutants produced increased amounts of c-di-AMP (Figure 6A).These increases agree to previous studies since the deletion of a c-di-AMP phosphodiesterase gene, pde2 or gdpP increases cellular c-di-AMP level in S. pyogenes (Fahmi et al., 2019).Interestingly, dltX deletion in the Dpde2 mutant increased c-di-AMP amount further, but liaR deletion in the Dpde2 mutant did not affect c-di-AMP production.
K + transport capacity and cellular c-di-AMP level show an inverse relationship in S. pyogenes We investigated if the impaired capability of K + transport affects cellular c-di-AMP levels.The DktrB strain, the mutant with the deletion of the high-affinity K + transporter gene, produced a high amount of c-di-AMP, even more than that of the Dpde2 mutant (Figure 6B).The DktrA strain, the mutant with the deletion of K + transport inhibitor gene, produced a lower amount of c-di-AMP level than that of the wild type.These results indicate that the relationship between K + transport capability and c-di-AMP production is inversely proportional in S. pyogenes.
The SpeB phenotype of the Dpde2DdltX mutant was not changed in high salt media, unlike the wild type and Dpde2DliaR mutant.We examined the SpeB activities of strains in high-salt media.Regular C medium used to make protease detection plates contains 10 mM K 2 HPO 4 .To increase salt concentration, we doubled or tripled the amount of K 2 HPO 4 .In these higher salt C media, the SpeB activities of the wild type and Dpde2DliaR mutants were reduced (Figure 7).However, the SpeB activities of the Dpde2DdltX mutant did not change in higher salt media.Thus, DltX and LiaR respond differently to a high osmolarity condition in the Dpde2 background.The DdltX and DliaR mutants still showed SpeB activity in high salt conditions.

Discussion
In a previous study, we observed that the transcription of the virulence factor gene speB ceases when the c-di-AMP phosphodiesterase pde2 gene is deleted in S. pyogenes (Fahmi et al., 2019).In this current research, we investigated the regulatory mechanisms and found that the Dlt system and LiaFSR regulate SpeB production through different mechanisms in the Dpde2 mutant.
The dlt operon, found in nearly all Gram-positive bacteria, is comprised mostly of four to five genesdlt(X)ABCD.This operon is primarily responsible for incorporating D-alanine esters into teichoic acids.In previous studies, this D-alanylation is entirely abolished when any of the dlt core genes (dltA to dltD) is inactivated in bacteria, including S. aureus, S. pyogenes, S. pneumoniae, Enterococcus faecalis, and Group B Streptococcus (Peschel et al., 1999;Boyd et al., 2000;Abachin et al., 2002;Kristian et al., 2005;Koprivnjak et al., 2006;Kovaćs et al., 2006).Recently, dltX has been discovered as an essential gene for D-alanylation in B. thuringiensis, but the specific role of the small membrane-associated protein DltX in this process remains unstudied (Kamar et al., 2017).Our study also suggests that DltX is essential for D-alanylation in S. pyogenes because the dltX deletion strain showed the same phenotype as the strain with a core dlt gene disruption.The mutation in the dlt operon in bacteria leads pleiotropic effects, including increased negative cell surface charge, increased sensitivity to CAMPs, enhanced autolysis, increased acid sensitivity, decreased biofilm formation, reduced adhesion to epithelial cells, and expression of altered virulence properties (Peschel et al., 1999;Boyd et al., 2000;Abachin et al., 2002;Kristian et al., 2005;Koprivnjak et al., 2006;Kovaćs et al., 2006).
D-alanylation of teichoic acid is a crucial bacterial defense mechanism against cationic antimicrobial peptides (CAMPs) (Kamar et al., 2017).The gram-positive thick cell wall plays a significant role in shielding against the host's antimicrobial peptides (AMPs).More than 60% of the total mass of the cell wall is made up of negatively charged teichoic acids (TAs) (Neuhaus and Baddiley, 2003;Silhavy et al., 2010).TAs are commonly found as wall teichoic acids (WTAs) or lipoteichoic acids (LTAs).WTAs are linked to peptidoglycan, and LTAs are anchored to the cytoplasmic membrane via their glycolipid moiety (Fischer, 1988;Fischer et al., 1990).The critical components of TAs are disaccharide anchors and phosphodiester-linked polyglycerol phosphate or polyribitol phosphate, which contributes to its net negative surface charge (Kojima et al., 1985;Bera et al., 2007).The positively charged AMPs (e.g., cathelicidins, defensin, etc.) are electrostatically attracted by the negatively charged bacterial surface and damage the bacterial cell membrane.Bacteria can modify their cell surface charge by incorporating positively charged residues to counteract CAMPs.This modification can be carried out by adding L-lysine to phosphatidylglycerol mediated by the mprF gene or incorporating D-alanine ester on free hydroxyls of the repeating sugar mediated by dlt operon as has been observed in firmicutes (Peschel et al., 1999;Abachin et al., 2002;Kristian et al., 2005;Weidenmaier and Peschel, 2008;Cox et al., 2009;Saar-Dover et al., 2012).In Streptococcus, D-alanylation appears to be the primary mechanism of CAMP resistance.D-alanine incorporation most likely modifies the conformation of LTAs, increasing cell wall density and decreasing CAMP penetration (Saar-Dover et al., 2012).In S. pyogenes, the dlt operon consists of six genes dltXABCDE.DltX is a small protein with 47 amino acids.The absence of the dltA gene, essential for D-alanylation, significantly lowers the expression of GAS virulence factors, including M protein and SIC (streptococcal inhibitor of complement) protein (Cox et al., 2009).S. pyogenes cell envelope TAs are mostly lipoteichoic acids.The loss of lipoteichoic acid D-alanylation by dltA mutation displays an increased negative surface charge, enhanced susceptibility to AMP, lysozyme, and neutrophil killing, in addition to decreased adhesion and invasion into the human pharyngeal epithelial cells (Kristian al., 2005).Our findings demonstrate that along with c-di-AMP, the Dlt system influences the production of the virulence factor SpeB in S. pyogenes.Inactivation of dltA or dltX in the Dpde2 background restores SpeB production to the wild-type level (Figure 2B).Also, this study shows that the DdltX mutant displays increased negative surface charge (Figure 3) and is more susceptible to cationic antimicrobial peptide polymyxin B than the wild type.These findings suggest that the dltX gene deletion causes D-alanylation defects, which increase negative surface charge and are more attracted to cationic polymyxin B. Similar results were observed in B. thuringiensis, where dltX is essential for D-alanylation (Kamar et al., 2017).The SpeB derepression by dlt mutation in the Dpde2 mutant appears not to be linked to the net negative surface charge.Even though the loss of the pde2 gene drastically reduces speB expression, the surface charge of the Dpde2 mutant was not changed (Figure 3).This could suggest that the deletion of the pde2 gene does not affect D-alanylation of teichoic acid.As expected, the Dpde2DdltX mutant with restored SpeB production bound almost twice more cationic cytochrome c than the amount bound to the wild type, probably because of the lack of D-alanylation (Figure 3).Hyyryläinen et al., 2007;Lin et al., 2020).In S. pyogenes, the sensor kinase LiaS and the repressor protein LiaF are colocalized in the cell membrane.They are involved in microdomain (Exportal) formation, and the disruption of GAS Exportal by CAMPs activates the LiaFSR system (Lin et al., 2020).Mutation in LiaFSR alters SpeB activity; DliaR mutation significantly increases speB transcription through SpxA2 (Sanson et al., 2021).In study, DliaR in the Dpde2 mutant derepressed SpeB production, but the derepression degree was less than that of DdltX (Figure 4).The RNA polymerase binding protein SpxA is a transcriptional regulator commonly found in firmicutes (Port et al., 2017;Lin et al., 2020).Several gram-positive bacteria have two highly similar SpxA paralogs, SpxA1 and SpxA2 (Nakano et al., 2005;Port et al., 2017).SpxA1 mainly responds to oxidative stress, and SpxA2 regulates numerous cellular activities such as cell division, cell wall homeostasis, fatty acid biosynthesis, virulence regulation, biofilm formation, as well as cell envelope stress regulation (Kajfasz et al., 2010;Baker et al., 2014;Kajfasz et al., 2015;Baker et al., 2020).SpxA1 and SpxA2 exist in S. pyogenes and have an opposite effect on virulence regulation or stress response (Port et al., 2017).The DspxA1 mutant is highly attenuated and shows enhanced PMB resistance, while DspxA2 becomes hypervirulent and more sensitive to PMB (Port et al., 2017).SpxA2 negatively regulates speB expression (Port et al., 2017).GAS spxA2 transcription is highly dependent on the LiaFSR system, similar to other firmicutes (Nakano et al., 2003;Baker et al., 2020;Sanson et al., 2021).Our qRT-PCR data confirmed that LiaR is essential for the spxA2 gene expression (Figure 5).The spxA2 transcript level was significantly reduced in the DliaR mutants compared to the wild type or the Dpde2 mutant in the presence of the cell wall stressor vancomycin (Figure 5).The transcription of spxA2 did not change in the Dpde2, DdltX, or Dpde2DdltX mutant compared to the wild type in the presence of vancomycin (Figure 5).These findings indicate that SpeB restoration of the Dpde mutant by dlt mutation is not through the LiaFSR-regulated gene spxA2.We observed no significant change in the liaR transcript level in the Dpde2 or DdltX mutant compared to that of the wild type or the Dpde2 mutant.Also, the transcription of dltX was not altered in the Dpde2 or DliaR mutant relative to the wild type or the Dpde2 mutant (Unpublished data).Thus, the Dlt system and LiaFSR are probably not linked in regulating SpeB production in the Dpde2 mutant.
It was previously shown that cellular c-di-AMP level controls K + transporter activity in S. pyogenes (Faozia et al., 2021).This study examined whether or not K + transport capability change affects cellular c-di-AMP levels.When the high-affinity transporter gene ktrB was deleted, cells produced a high amount of c-di-AMP.However, less c-di-AMP was generated when the transporter regulator (inhibitor) gene ktrA was deleted.Thus, K + transport capacity has an inverse relationship to cellular c-di-AMP level in S. pyogenes.Cellular c-di-AMP level is changed by environmental osmolarity alteration in several gram-positive bacteria, including Lactococcus lactis, Lactobacillus plantarum, Listeria monocytogenes, and S. aureus (Pham et al., 2018).When the bacteria encounter high osmolarity conditions, cellular c-di-AMP level decreases rapidly.Under a low osmolarity condition, cellular c-di-AMP levels in these bacteria increase (Pham et al., 2018).Thus, these results, including ours, demonstrate that bacteria alter cellular c-di-AMP levels based on environmental or cellular conditions and change cellular activities based on c-di-AMP levels.However, the molecular mechanism of how these bacteria sense environmental or cellular conditions to change c-di-AMP level has yet to be elucidated.
The Dpde2DdltX mutant produces a higher amount of c-di-AMP than that of the Dpde2 or Dpde2DliaR mutant (Figure 6).Since cellular c-di-AMP level can be altered by osmolarity or turgor pressure change (Pham et al., 2018), we examined the SpeB phenotypes of strains in different salt concentration conditions.The single gene deletion mutants, the DdltX mutant and DliaR mutant, behaved similarly; they showed identical SpeB phenotypes regardless of salt concentration change.However, the Dpde2DdltX mutant and Dpde2DliaR mutant behaved differently.The Dpde2DliaR mutant showed reduced SpeB production, but the SpeB production of the Dpde2DdltX mutant did not decrease in higher salt media.This indicates that the DltX and LiaFSR systems affect SpeB production through different pathways in the Dpde2 background.
In summary, our results show that both the Dlt and LiaFSR systems affect SpeB production in the Dpde2 background.The LiaFSR system controls speB expression through the transcriptional regulator SpxA2.We could not pinpoint the regulatory pathway the Dlt system involves, but salt transport or turgor pressure regulation might be involved.We also proved that the small membrane protein DltX is essential for the D-alanylation of teichoic acids in S. pyogenes.

Materials and methods
Bacterial strains and media S. pyogenes HSC5 (emm genotype 14) (Hanski et al., 1992;Port et al., 2013) was employed for all experiments, including strain construction.Molecular cloning experiments utilized Escherichia coli DH5a or TOP10 (Invitrogen), which was cultured in Luria-Bertani broth.The routine culture of S. pyogenes employed Todd-Hewitt medium (BBL) supplemented with 0.2% yeast extract (Difco) (THY medium), and cells were grown at 37°C in sealed tubes without agitation.Unless otherwise indicated, C medium (Lyon et al., 1998) was used to grow S. pyogenes for SpeB activity assay and RNA preparation for real-time qRT-PCR.Bacto agar (1.4%, w/v; Difco) was added to make solid media.Cultures on solid media were incubated under the anaerobic condition created by a commercial product (e.g., GasPak; catalog no.260678; BBL).When appropriate, antibiotics were added to the media at the following concentrations if they are not specified: kanamycin, 50µg/ml for E. coli and 500µg/ml for S. pyogenes; erythromycin, 500µg/ml for E. coli and one µg/ml for S. pyogenes.

Manipulation of DNA
Plasmid DNA was isolated via a commercial kit (e.g., Gene Elute plasmid miniprep kit; Sigma) and used to transform S. pyogenes or E. coli as described previously (Caparon et al., 1991).Enzymes for DNA cloning and PCR were used according to the manufacturers' recommendations.Chromosomal DNA was purified from S. pyogenes using a commercial kit (e.g., GenElute bacterial genomic DNA kit; Sigma).

Transposon mutagenesis
For the transposon mutagenesis, TnWKm2, a Tn4001 derivative containing a kanamycin resistance determinant, was employed (Kang et al., 2012).Briefly, the purified plasmid containing TnWKm2 introduced to the Dpde2 mutant using electroporation (voltage: 2100V, Capacitor: 25 uF, Resistance: 200 Ohms).The colonies with kanamycin resistance were patched on the protease indicator plates, and the strains showing the wild-type level protease activity were selected.The transposon insertion sites in those strains were identified by sequencing of chromosomal DNA with a primer binding to a transposon sequence.The sequencing data were compared with the NCBI genomic database (http://www.ncbi.nlm.nih.gov/BLAST/) to identify transposon-insertion sites.

Strain construction
The generation of the pde2 deletion mutant has been described elsewhere (Fahmi et al., 2019).Other gene deletion mutations, DdltX and DliaR on chromosomal loci were generated by employing the shuttle vector with a temperature-sensitive replication origin, pJRS233 (Cho and Kang, 2013).Briefly, the target gene and ~1000 bp sequences immediately upstream and downstream were amplified by PCR.This PCR product was inserted into pJRS233 using the fast-cloning method (Li et al., 2011).The plasmid with a target gene deletion allele was obtained by inverse PCR.The target gene deletion plasmids, pDdltX and pDliaR, were created using this method.The primers used to generate the PCR products are listed in Table 1.pDdltX or pDliaR was used to replace each target gene in the wild type or the Dpde2 mutant by the gene deletion method that employs the temperaturesensitive replication origin, as described previously (Cho and Kang, 2013).The WdltA or Dpde2WdltA mutant was constructed by insertional disruption of the dltA gene through single homologous recombination.An internal region of the dltA gene was amplified by PCR and then inserted into the suicide vector pCIV2 (Okada et al., 1993;Lyon et al., 2001).The resultant plasmid, pWdltA, was used to transform HSC5 or the Dpde2 mutant into resistance to kanamycin.Since the suicide vector pCIV2 lacks the replication origin of S. pyogenes, the plasmid can only exist in S. pyogenes by integrating into the chromosome through homologous recombination.The fidelity of all genetic constructs was confirmed by PCR and/or DNA sequencing.
The dltX-complemented Dpde2DdltX(pdltX) strain was generated using the plasmid p7INT that inserts into the streptococcal chromosome (McShan et al., 1998).The DNA containing the dltX gene and its promoter region was amplified by PCR and inserted into p7INT through the fast-cloning method (Li et al., 2011).The resulting plasmid pdltX was transferred into the Dpde2DdltX mutant to make the Dpde2DdltX(pdltX) strain.

Determination of MIC to polymyxin B
The susceptibility of mutant strains to the cell membranetargeting antibiotic polymyxin B was monitored as follows.S. pyogenes cells grown in THY medium overnight were inoculated into fresh THY medium containing polymyxin B (100, 50, 10, 5, 1, 0.5, and 0 µg/ml).Cells were then grown in a 96-well plate overnight at 37°C, and the OD 600 of the overnight cultures (~18 h postinoculation) was measured to determine the final cell density.This experiment was performed in triplicate.

Cytochrome c binding assay
THY media (50 ml) were inoculated with 2-3% overnight cultures and cultivated to the early exponential phase.Cells were collected by centrifugation at 7000 g for 10 min, resuspended in 50 ml of chemically defined media (DMEM), and incubated overnight.Then, the cells were washed twice with 20 ml of morpholino propanesulfonic acid (MOPS) buffer (20 mM, pH 7), adjusted to the final cell OD 600 to 3 in 2 ml MOPS buffer with 0.2 mg/ml cytochrome c (Sigma-Aldrich, St. Louis, MO), and incubated for 10 minutes in a shaker at room temperature.As a control, 0.2 mg/ml cytochrome c was incubated in MOPS buffer under the same conditions without bacteria.After 10 min, cells were removed by centrifugation, and the cytochrome c content of the supernatants was quantified photometrically by measuring OD 530 .

qRT-PCR
Real-time qRT-PCR was conducted as described elsewhere (Cho and Kang, 2013).The primers for qRT-PCR are listed in Table 1.The gyrase A subunit gene (gyrA) was used as the internal reference gene to normalize the expression level of a specific transcript between samples (Kang et al., 2010).The reported data represent the means and standard errors from three independent assays performed on different days with new RNA samples.

Determination of gene expression under the influence of an antibiotic
Cells at the exponential phase (0.3 -0.4 of OD 600 ) were collected and incubated in a fresh medium with an antibiotic at 37°C for an hour.Then, cells were collected and lysed for RNA purification.

SpeB activity measurement using protease indicator plates
Overnight cultures in THY medium were serially diluted with fresh THY medium, and the diluted cells (2µl) were spotted onto protease indicator agar plates (C medium agar plates containing 2% skim milk).The plates were then incubated anaerobically at 37°C for 24 h to 48 h, and SpeB activity that displays a clear zone around the spotted cells was observed.

Quantification of c-di-AMP in cell extracts by ELISA
The measurement of c-di-AMP concentration in cell extract was conducted as previously described (Fahmi et al., 2019).Briefly, S. pyogenes were grown to the exponential phase (OD600 = ~0.4) in 10 ml THY medium, washed three times with PBS, resuspended in 1 ml PBS, and lysed through PlyC treatment.The clear supernatant of the culture was collected in a fresh tube after centrifugation at 7,000 relative centrifugal force (rcf) at 4°C for 10 min.The cell lysates were then boiled for 10 min.Clear supernatants were collected after centrifugation and stored at -20°C until used to measure c-di-AMP concentration.The purified CabP protein was diluted to 50 µg/ml in coating buffer (50 mM Na2CO3, 50 mM NaHCO3, pH 9.6), and 100 µl of the solution was added to each well to coat the wells of a 96-well flat-bottom plate.The plates were sealed with plastic wrap and incubated overnight at 4°C.The coated wells were then washed three times with PBS containing 0.05% Tween 20 (PBST) and blocked with 5% bovine serum albumin (BSA).The cell extract samples were diluted (5 times) with 50 mM Tris buffer (pH 8).100µl of controls, standards, To create pDliaR and samples were added to the coated wells (in triplicate).The plates were incubated for 2 hrs at room temperature.Each well of the plates was washed three times with 200 µl PBST.Next, 100 µl of 0.1 µg/ml high-performance streptavidin (Thermo Fisher Scientific) in PBS was added and incubated for 1 hr.After wells were washed three times with PBST, 100 µl of the substrate (0.5 mg of o-phenylenediamine dihydrochloride [Sigma-Aldrich] in citrate buffer [pH 5] containing 20 µl H 2 O 2 ) was added to each well and incubated for 30 min at room temperature.Finally, the reactions were stopped with 100 µl of 2M H 2 SO 4 .The OD 492 of each well was measured using a plate reader.A standard curve was generated and used to measure the levels of c-di-AMP in samples.

Statistical testing
Each statistical test applied to the experiments was described in the figure legends.

FIGURE 1
FIGURE 1 Genetic organization of the dlt operon in S. pyogenes.The dlt operon consists of dltX, dltA, dltB, dltC, dltD, and dltE genes.Each arrow indicates an individual open reading frame and its orientation.dltX is co-expressed with the downstream genes in the dlt operon, whose expression is shown with a dotted arrow over the dlt genes.The proteins encoded by these open reading frames and their putative functions are shown.
FIGURE 3Deletion of dltX increases the negative surface charge of S. pyogenes.The amount of the cationic protein cytochrome c bound to S. pyogenes cell surface was measured to compare the negative surface charge of strains.The data are the means and standard errors of the means derived from three independent experiments.The significance of the difference between two strains was evaluated by a two-tailed unpaired t-test (*, P < 0.05; ***, P < 0.001; ****, P < 0.0001; ns, no significance).The following strains were tested: the wild type (WT), pde2 deletion mutant (DPde2), dltX deletion mutant (DDltX), pde2 dltX double deletion mutant (DPde2DDltX), dltX-complemented pde2 and dltX double deletion mutant [DPde2DDltX(pDltX)].

FIGURE 4
FIGURE 4The deletion of liaR partially restored the SpeB activity of the Dpde2 mutant.The clear zones on the protease indicator plates display the activity of SpeB secreted by S. pyogenes.The names of the strains used are shown above the image, and the dilution degrees of the overnight cultures are indicated at the left side of the picture.Plates were incubated anaerobically at 37˚C for 24 -48 h.The following strains were tested: the wild type (WT), pde2 deletion mutant (DPde2), pde2 and dltX double deletion mutant (DPde2DDltX), and pde2 and liaR double deletion mutant (DPde2DLiaR).This image is representative of many.
FIGURE 7The SpeB activity phenotype of DPde2DDltX did not change at a high salt condition, unlike DPde2DLiaR.The activity of the secreted protease SpeB is shown on protease indicator plates.Strains were grown overnight and spotted (2µl) onto protease indicator agar plates after serial dilution.SpeB activity displays a clear zone around the spotted cells after incubation.The strains' names are shown above each image, and the dilution degrees of the spotted cultures are indicated at the left side of the first image.Plates were incubated anaerobically at 37°C for 24 -48 h.The following strains were tested: the wild type (WT), pde2 deletion mutant (DPde2), dltX deletion mutant (DDltX), pde2 and dltX deletion mutant (DPde2DDltX), liaR deletion mutant (DLiaR), and pde2 and liaR deletion mutant (DPde2DLiaR).This image is representative of many.
shown 5′ to 3′.Uppercase sequences anneal to the HSC5 chromosome, and lowercase sequences anneal to plasmid sequences.b Mutagenesis primers were used for PCR reactions to amplify DNA segments used to construct plasmids for gene deletion or insertional gene disruption.c Analysis primers were used in real-time qRT-PCR to measure the level of gene transcription.d PCR product size.