Regulation and Consequence of Serine Catabolism in Streptococcus pyogenes

Abstract

The Gram-positive bacterium Streptococcus pyogenes (also called group A Streptococcus [GAS]), is found strictly in humans and is capable of causing a wide variety of infections. Here we demonstrate that serine catabolism in GAS is controlled by the transcriptional regulator Spy49_0126c. We have designated this regulator SerR (for serine catabolism regulator). Microarray and transcriptional reporter data show that SerR acts as a transcriptional repressor of multiple operons, including sloR and sdhBA. Purified recombinant SerR binds to the promoters of both sloR and sdhB, demonstrating that this regulation is direct. Deletion of serR results in a lower culture yield of the mutant than of the wild type when the strains are grown in defined medium unless additional serine is provided, suggesting that regulation of serine metabolism is important for maximizing bacterial growth. Deletion of sloR or sdhB in the ΔserR mutant background restores growth to wild-type levels, suggesting that both operons have roles in serine catabolism. While reports have linked sloR function to streptolysin O expression, transport experiments with radiolabeled l-serine reveal that the sloR operon is required for rapid acquisition of serine, suggesting a novel role for this operon in amino acid metabolism.

INTRODUCTION

The Gram-positive bacterium Streptococcus pyogenes, also known as group A Streptococcus (GAS), is a strictly human pathogen that poses large health and economic burdens worldwide (2). The most familiar illness associated with GAS is acute pharyngitis, commonly known as “strep throat,” but S. pyogenes also causes a wide range of other noninvasive and invasive infections, including pyoderma, toxic shock syndrome, and necrotizing fasciitis, as well as postinfection sequelae, such as acute rheumatic fever, rheumatic heart disease, and glomerulonephritis (9). The wide range of infections caused by GAS requires the bacterium to survive at various locations within the human host, including on the skin, in the mucosal lining of the nasopharynx, and in the blood, with each site posing different physiological difficulties that GAS must overcome in order to survive. In vitro cultures in saliva, blood, and amniotic fluid have demonstrated the high plasticity of the GAS transcriptome in response to various host environments (16, 39, 43), suggesting that GAS uses adaptation to changing environments via differential expression of stress and nutrient acquisition genes to ensure survival.

The GAS genome is predicted to encode an average of 13 two-component systems and more than 100 transcription factors whose activity depends on environmental and bacterial factors yet to be discovered (34). It has become increasingly clear that the regulatory system of S. pyogenes is extremely complex, consisting of numerous parallel and converging regulatory networks partaking in the regulation of both virulence and metabolic genes (21). For example, CcpA is a transcriptional regulator that mediates carbon catabolite repression (CCR) in Gram-positive bacteria (48), while the stand-alone regulator CodY responds to branched-chain amino acid (BCAA) concentrations, as well as to GTP levels in at least one species, in order to modulate BCAA biosynthesis accordingly (27). Importantly, in addition to regulating genes involved in carbon source utilization and BCAA biosynthesis, both CcpA and CodY alter virulence factor expression in GAS (20, 26, 40, 41), attesting to the importance of coordinated regulation of metabolism and virulence in this organism.

Amino acids play various roles within the cell, contributing to both bacterial growth and survival. As the building blocks of peptides and proteins in all organisms, amino acids are essential nutrients that must be produced by the cell or scavenged from the environment in order to permit growth. Group A Streptococcus is auxotrophic for 15 of the 20 essential amino acids (10), including l-serine, and thus is highly dependent on amino acids obtained from the host organism in the form of free amino acids or peptides (35). Apart from allowing for protein synthesis, amino acid availability has substantial effects on gene expression in S. pyogenes through the modulation of transcriptional regulator activity (27) and riboswitch formation (45, 49). Some amino acids, such as l-arginine, also play an important role in pH resistance in many bacteria, including GAS (8). Despite the importance of amino acids for various cellular processes and the polyauxotrophy of S. pyogenes for most amino acids, little is understood regarding the transport of free amino acids or the regulation of genes involved in free amino acid uptake in GAS. In fact, we have been unable to find any reports experimentally identifying a transporter responsible for the uptake of any free amino acid in S. pyogenes. A previous report provided evidence that a common transporter may exist for l-isomers of alanine, serine, threonine, and glycine in GAS (36); however, the transporter itself was never identified. By the same token, information regarding the direct regulation of amino acid degradation in general is lacking for this organism.

In this study we aimed to characterize the role of the sloR operon in S. pyogenes, especially with regard to the upstream putative transcriptional regulator. The sloR gene (spy49_0128) was previously linked to the expression of streptolysin O (SLO), an important GAS virulence factor (13, 23, 42, 46); however, the exact nature of this effect was not elucidated. Our studies reveal a novel role for this operon in the regulation of l-serine catabolism. Our data demonstrate that the SloR operon and the divergently transcribed transcriptional regulator, which we have designated SerR, are directly involved in l-serine utilization in GAS. Furthermore, regulation of l-serine catabolism is demonstrated to be important under conditions where l-serine is not overly abundant, since increased degradation of l-serine resulting from the absence of SerR can lead to restricted cell growth.

MATERIALS AND METHODS

Bacterial strains and media.

All S. pyogenes strains used in this study were derivatives of the sequenced serotype M49 strain NZ131 (Table 1) (31). All GAS strains were maintained on Todd-Hewitt medium with 2% (wt/vol) yeast extract (THY; Difco) plates with antibiotics as needed at the following concentrations: chloramphenicol (Cm), 3 μg/ml; erythromycin (Em), 0.5 μg/ml; spectinomycin (Spc), 100 μg/ml. GAS liquid cultures were grown statically in one of the following types of medium: THY, C-medium (25), or chemically defined medium (CDM) (47) containing 0.5% glucose and supplemented with l-asparagine to 100 mg/liter. For all experiments, overnight cultures were grown at 30°C in the media indicated in the figures and the text, diluted to an optical density at 600 nm (OD₆₀₀) of ≈0.01 into fresh media of the same type, and incubated at 37°C for the duration of the experiment. All plasmids were constructed using Escherichia coli DH10β (Invitrogen) or BH10C (18) as the host, and all E. coli strains were grown in Luria broth at 30°C with shaking, with antibiotics as needed at the following concentrations: chloramphenicol, 10 μg/ml; erythromycin, 500 μg/ml; spectinomycin, 100 μg/ml.

Table 1.

Bacterial strains and plasmids used in this study

Strain or plasmid	Description or genotype; phenotypea	Reference or source
Strains
E. coli
BL21(DE3)	E. coli host for protein expression	Invitrogen
DH10B	F−mcrA Δ(mrr-hsdRMS-mcrBC) φ80lacZΔM15 ΔlacX74 recA1 endA1 araD139 deoR Δ(ara leu)7697 galU galK rpsL nupG λ−	Invitrogen
BH10C	ΔpncB variant of DH10B	18
S. pyogenes
NZ131	Wild-type M49 S. pyogenes isolate	42a
BNL100	NZ131 but ΔserR::aad9; Spcr	This study
BNL102	NZ131 but ΔsloR::aad9; Spcr	This study
BNL105	NZ131::pBL105; Emr	This study
BNL106	BNL100::pBL105; Spcr Emr	This study
BNL114	NZ131 but ΔserR::aad9 ΔsloR::cat; Spcr Cmr	This study
BNL116	NZ131 but ΔsdhB::cat; Cmr	This study
BNL117	NZ131 but ΔserR::aad9 ΔsdhB::cat; Spcr Cmr	This study
BNL141	NZ131::pBL106; Emr	This study
BNL142	BNL100::pBL106; Spcr Emr	This study
BNL143	BNL102::pBL106; Spcr Emr	This study
BNL144	BNL114::pBL106; Spcr Cmr Emr	This study
BNL158	NZ131::p7Int; Emr	This study
BNL159	BNL100::p7Int; Spcr Emr	This study
BNL160	BNL102::p7Int; Spcr Emr	This study
BNL161	BNL114::p7Int; Spcr Cmr Emr	This study
BNL162	BNL100::pBL109; Spcr Emr	This study
BNL163	BNL102::pBL110; Spcr Emr	This study
BNL164	BNL114::pBL110; Spcr Cmr Emr	This study
Plasmids
pET-15b	Protein expression vector; Ampr	Novagen
pFED760	Temperature-sensitive pG+host9 plasmid (ISS1 deleted); Emr	29
p7INT	pUC18-derived streptococcal integration vector; Emr	32
pBL100	pFED760 containing the upstream and downstream regions of serR flanking aad9 at the PstI site; Emr Spcr	This study
pBL101	pFED760 containing the upstream and downstream regions of sloR flanking aad9 at the PstI site; Emr Spcr	This study
pBL102	pFED760 containing the upstream and downstream regions of sloR flanking cat at the PstI site; Emr Cmr	This study
pBL103	pFED760 containing the upstream and downstream regions of sdhB flanking cat at the PstI site; Emr Cmr	This study
pBL104	pET-15b vector containing serR between the BamHI and NdeI sites; Ampr	This study
pBL105	p7INT containing the PsloR-luxAB fusion at the EcoRI site; Emr	This study
pBL106	p7INT containing the PsdhAB-luxAB fusion between the BamHI and EcoRI sites; Emr	This study
pBL109	p7INT containing serR under the control of its native promoter between the BamHI and EcoRI sites; Emr	This study
pBL110	p7INT containing sloR under the control of its native promoter between the BamHI and XhoI sites; Emr	This study

^aCm, chloramphenicol; Em, erythromycin; Spc, spectinomycin; Amp, ampicillin.

Construction of serR, sdhB, and sloR mutants and complementation vectors.

All plasmids and oligonucleotide primers (Integrated DNA Technologies) used in this study are listed in Tables 1 and 2, respectively. Mutants were constructed by allelic exchange using a method described previously (11). This method uses a temperature-sensitive pG⁺host9 plasmid (pFed760) (29) to facilitate the deletion of target genes via homologous recombination. Five hundred-base-pair regions upstream and downstream of the targeted gene(s) were amplified by PCR (Phusion DNA polymerase; Finnzymes) using S. pyogenes NZ131 DNA as the template and the following primer pairs: for serR, BL1/BL2 and BL3/BL4; for sloR, BL5/BL6 and BL7/BL8; for sdhB, BL9/BL10 and BL11/BL12. The restriction sites included within these primers created a PstI site at the 5′ end and an XhoI site at the 3′ end of the upstream flanking region (UFR) and a BglII site at the 5′ end and a PstI site at the 3′ end of the downstream flanking region (DFR). The primers used for amplification of a spectinomycin resistance cassette (aad9), BL13/BL14, from pLZ12Spec (19) and a chloramphenicol resistance cassette (cat), BL15/BL16, from pEVP3 (7) included restriction enzyme sites that created XhoI and BglII sites at the ends of the resistance cassettes. PCR fragments digested with XhoI and/or BglII (all restriction enzymes were purchased from New England Biolabs) were ligated in vitro (T4 DNA ligase; New England Biolabs), and the desired UFR-aad9-DFR or UFR-cat-DFR fusions were amplified using the outside primer pair BL1/BL4, BL5/BL8, or BL9/BL12. The amplified fusion constructs were digested with PstI and were ligated into pFed760 at the PstI site by in vitro ligation to generate the deletion plasmids (pBL100 to pBL103). Following allelic exchange, deletion of the targeted gene(s) in Cm-resistant or Spc-resistant colonies that were Em sensitive was confirmed by PCR using mutant chromosomal DNA as the template.

Table 2.

Oligonucleotide primers used in this study

Function and name	Sequence (5′–3′)a	Usage
Gene deletion
BL1	GCGTGCTGCAGTAATAGTGCAGATTCTTTTG	UFR of serR
BL2	GCGTGCTCGAGCCAAATCACCATAAAA	UFR of serR
BL3	GCGTGAGATCTGTCAAACTTCCTATATCTATCTTC	DFR of serR
BL4	GCGTGCTGCAGTCGACTTCAATCAAAGCTTGCAAAG	DFR of serR
BL5	GCGTGCTGCAGAATCACCTTAATAATAATTT	UFR of sloR
BL6	GCGTGCTCGAGAATCTATTCCACCAATCTAT	UFR of sloR
BL7	GCGTGAGATCTTAAACATCTTTTTGACAGGAC	DFR of sloR
BL8	GCGTGCTGCAGCATAAATTGTTCTAGCCTCTTGTTC	DFR of sloR
BL9	GCGTGCTGCAGGACACCAATAACATCATAACCTTGC	UFR of sdhB
BL10	GCGTGCTCGAGAGGCAACTCCGTTTTTTATTTCTATGT	UFR of sdhB
BL11	GCGTGAGATCTAAAGGAAAGCTATGTTTTATACTATTGAAGAACTTGT	DFR of sdhB
BL12	GCGTGCTGCAGATTTTAGCATTTAATTCGTTTACAGCTATAGCATTTC	DFR of sdhB
BL13	GCGTGAGATCTTCGATTTTCGTTCGT	aad9
BL14	GCGTGCTCGAGTTAGAATGAATATTT	aad9
BL15	GCGTGCTCGAGGCGAAAAAGGAGAAGTCGGTTCAGAAA	cat
BL16	GCGTGAGATCTCGGTATCGATAAGCTTGATGAAAATTTGTTTG	cat
Reporter fusion
BL17	GCGTGGAATTCGTCAAACTTCCTATATCTATCTTCTTG	sloR promoter
BL18	GCGTGGCGGCCGCATACGAACCTCCTCATTGATAATAT	sloR promoter
BL19	GCGTGGGATCCAAAAATTCTCCCATCAATAATAAGATAGAAAAAGA	sdhBA promoter
BL20	GCGTGGCGGCCGCAGGCAACTCCGTTTTTTATTTCTATGTTTATTATA	sdhBA promoter
BL25	GCGTGGCGGCCGCATTAATCACCAAAAAGGAATAGAGT	luxAB
BL27	GCGTGGAATTCGCCTTTAATTTTATTATGGT	luxAB
His-SerR purification
BL29	GCGTGGGATCCTTATTGATCGGCTTCAATTTTTTTAAGG	serR
BL30	GCGTGCATATGGATAAAGAAACGCTAAACTACTGGA	serR
EMSA
BL31	GCGTGGAATTCACAATTGACCTGTCACTGGATTAA	sloR promoter
BL32	5′ 6-FAM–ACAATTGACCTGTCACTGGATTAA	sloR promoter
BL33	5′ 6-FAM–AGGCAACTCCGTTTTTTATTTCTATGTTTATTATA	sdhBA promoter
BL34	5′ 6-FAM–ACAGAAGCTCCTTTAAGATAGTTATTAGTAGCTGTC	serS promoter
BL35	GTGTTTAAAGACCTCTCATGGGCAAAT	rRNA promoter
BL36	CAGGTTTCTCATAGCCTGTCAACTACTTTT	rRNA promoter
BL37	56-FAM–CAGGTTTCTCATAGCCTGTCAACTACTTTT	rRNA promoter
Complementation constructs
BL38	GCGTGGGATCCGTCAAACTTCCTATATCTATCTTCTTG	sloR promoter
BL39	AATAATAATATCCATATACGAACCTCCTCATTGATAATATAGTTAAATTT	sloR promoter
BL40	TGAGGAGGTTCGTATATGGATATTATTATTGGAACAAGTCTTTTGATTCTT	sloR
BL41	GCGTGCTCGAGATCTTGTTAAAAGTCCTGTCAAAAA	sloR
BL42	GCGTGGAATTCATACGAACCTCCTCATTGATAATATA	serR

^aRestriction sites are underlined. The complementary sequences in the sloR complementation primers are set in boldface. 5′ 6-FAM, 6-carboxyfluorescein added at the 5′ end.

Complementation vectors were constructed as follows. serR with its native promoter was amplified using primer pair BL29/BL42 and was cloned into p7INT between the BamHI and EcoRI sites to generate pBL109. The sloR promoter and the sloR gene were amplified using primer pairs BL38/BL39 and BL40/BL41, respectively. Primers BL39 and BL40 contain sequences complementary to each other, allowing for overlap extension PCR between the two amplification products, resulting in an in-frame fusion between sloR and its promoter while eliminating the spy49_0127 gene. The sloR-promoter fusion was ligated into p7INT between the BamHI and XhoI sites to generate pBL110. p7INT integrates into a neutral site in the GAS chromosome (attB) (32); thus, the plasmids were transformed into GAS by electroporation, and transformants were selected for on THY-Em plates. Chromosomal DNA was purified from all complemented strains for use in PCR confirmation of vector integration.

Construction of luxAB transcriptional fusions.

Transcriptional fusions of the sloR and sdhBA promoter regions to luxAB were generated as follows. The sloR and sdhBA promoter regions were amplified by PCR from S. pyogenes NZ131 DNA using primer sets BL17/BL18 and BL19/BL20, respectively. The luxAB genes were amplified from pCN59 (3) using primers BL25 and BL27. These primers include restriction enzyme sites that create NotI sites at the 3′ ends of the promoter regions and at the 5′ end of the luxAB operon. Following digestion of all products with NotI, the promoter fragments were fused to luxAB by in vitro ligation. The desired fusion products were amplified using the ligation reaction products as the template and primers BL17 or BL19 and BL27. These primers include restriction enzyme sites that create EcoRI sites at both ends of the P_sloR-luxAB fusion product or BamHI and EcoRI sites at the 5′ and 3′ ends of the P_sdh-luxAB fusion product, respectively. Following digestion with EcoRI and/or BamHI, the desired fusion products were gel purified and ligated into similarly digested p7INT to generate pBL105 and pBL106. All plasmid-generating ligation reaction products were transformed into E. coli DH10β or BH10C, and cells were plated onto Luria broth plates with erythromycin to select for transformants. Following transformation of reporter constructs into GAS, the integration of the reporters into the chromosome was confirmed by PCR.

Luciferase reporter assay.

Overnight cultures of GAS strains carrying transcriptional fusions were diluted to an OD₆₀₀ of 0.01 in fresh medium and were grown at 37°C without agitation. Throughout growth, the optical transmittance of the culture at 600 nm was measured using a Spectronic 20D spectrophotometer (Milton Roy). Concurrently, 100-μl aliquots of cultures were placed in a 96-well dish and were exposed to decyl aldehyde vapors for 35 s (Acros Organics). Prior to the measurement of luminescence, the plate lid was removed, and luminescence readings were performed within 1 min. All luminescence measurements were performed using a Microbeta Plus liquid scintillation counter, model 1450 (Wallac). All luciferase data shown are representative of experiments performed in triplicate.

Purification of RNA and microarray analysis.

NZ131 and the isogenic serR deletion mutant (BNL100) were grown to mid-logarithmic phase in C-medium at 37°C without shaking. RNA was harvested from three replicate cultures using a RiboPure-Bacteria kit (Ambion) according to the manufacturer's instructions, and the quality of the RNA was confirmed by agarose gel electrophoresis. RNA concentrations were determined using a NanoDrop 1000 spectrophotometer (Thermo Scientific). Reverse transcription using Superscript II reverse transcriptase (RT) and random hexamer primers (both from Life Technologies) was performed with 5 μg total bacterial RNA incorporating amino-allyl dUTP (Sigma). RNA was degraded using sodium hydroxide, and cDNA was purified using a PCR purification kit (Qiagen). cDNA was then incubated with cyanine-3 ester (Amersham Pharmacia) and was washed again using the PCR purification kit, and dye incorporation was measured using a NanoDrop instrument. All hybridization to S. pyogenes whole-genome microarrays (NimbleGen) was performed in accordance with the NimbleGen hybridization protocol. NimbleScan was used to create raw data files, and Partek was used for RMA normalization. Labeling, array hybridization, and quantitation were conducted as a service by The W. M. Keck Center for Comparative and Functional Genomics at the University of Illinois.

Expression and purification of recombinant SerR.

The serR gene was amplified from NZ131 genomic DNA using primers BL29 and BL30 containing BamHI and NdeI restriction sites, respectively. Following BamHI/NdeI digestion, the PCR product was cloned into the pET15b vector (Novagen), which adds a 6-histidine tag to the amino terminus of the protein, to generate pBL104. pBL104 was then transformed into E. coli BL21(DE3) (Novagen) by electroporation and was grown in Luria broth with 100 μg/ml ampicillin at 30°C with shaking. At an OD₆₀₀ of 0.8, protein expression was induced with 0.5 mM isopropyl-β-d-thiogalactopyranoside (IPTG) at 16°C with shaking for 16 h. Following induction, cells were pelleted by centrifugation and were then resuspended in cold buffer A (20 mM sodium phosphate buffer [pH 7.6], 0.5 M NaCl, 20 mM imidazole, 20 mM β-mercaptoethanol [BME], 1 μg/ml DNase I, and protease inhibitor cocktail). Cells were lysed by sonication, and the soluble and insoluble fractions of the cell lysates were separated by centrifugation at 14,000 rpm in a Sorvall SA-600 rotor. Solubility studies indicated that recombinant SerR was found in inclusion bodies; thus, the insoluble pellet was resuspended in cold buffer B (8 M urea, 20 mM sodium phosphate buffer [pH 7.6], 0.5 M NaCl, 20 mM imidazole, and 20 mM BME) to denature SerR_His. After a 30-min incubation at 30°C with periodic vortexing, the remaining insoluble fraction was removed by centrifugation. SerR_His was purified from the denatured, soluble fraction by Ni²⁺ affinity chromatography using a 1-ml HisTrap HP column (GE Healthcare) and was eluted in cold buffer C (8 M urea, 20 mM sodium phosphate buffer [pH 7.6], 0.5 M NaCl, 500 mM imidazole). The purity of SerR_His was estimated to be >95% as determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) with Coomassie blue staining. Denatured SerR_His was refolded in vitro by buffer exchange using a Slide-A-Lyzer cassette (Thermo Scientific) and was dialyzed against cold buffer D (20 mM Tris-HCl [pH 8.0], 100 mM KCl, 2.5 mM MgCl₂, 0.5 mM EDTA [pH 8.0], 1 mM dithiothreitol [DTT], protease inhibitor cocktail) at 4°C for 3.5 h twice in succession. The sample was then dialyzed against buffer E (20 mM Tris-HCl [pH 8.0], 100 mM KCl, 2.5 mM MgCl₂, 0.5 mM EDTA [pH 8.0], 1 mM DTT, 25% glycerol, protease inhibitor cocktail) at 4°C overnight. Following buffer exchange, all precipitated SerR_His was removed by centrifugation, and the final soluble SerR_His concentration was determined by a Bradford assay using Coomassie protein assay reagent and albumin standards (Thermo Scientific). Protein aliquots were stored at −80°C until use.

EMSAs.

All DNA fragments for electrophoretic mobility shift assays (EMSAs) were amplified from NZ131 chromosomal DNA by PCR using the primers shown in Table 2 and were purified by gel purification. Select primers included 6-carboxyfluorescein (FAM) fluorescent tags (Integrated DNA Technologies) at their 5′ ends. The following EMSA probes were amplified with the primers given in parentheses: P_sloR (BL17/BL32), P_sdh (BL19/BL33), P_serS (BL21/BL34), and P_ssRNA (BL35/BL37). Unlabeled probes were amplified using primers BL31, BL20, and BL36 in place of BL32, BL33, and BL37, respectively. For EMSAs, varying concentrations of SerR_His (in buffer E) were incubated with 20 mM HEPES (pH 7.9), 100 mM KCl, 12.5 mM MgCl₂, 0.2 mM EDTA (pH 8.0), 0.5 mM dithiothreitol, 50 μg/ml salmon sperm DNA, 0.001 U/μl poly(dI·dC), 100 μg/ml bovine serum albumin, 0.5 mM CaCl₂, and 12% (vol/vol) glycerol at room temperature for 15 min prior to the addition of 10 nM probe and competitor DNA if applicable. Following the addition of the probe, the reaction mixtures were incubated at room temperature for an additional 30 min and were then fractionated on 5% native polyacrylamide gels buffered with 20 mM potassium phosphate (pH 7.5) for 90 min at 110 V and 4°C. If competitor DNA was included in the reaction mixture, it was added simultaneously with the labeled probe at a 10× molar excess. All gel shifts were detected by fluorescence imaging using a Typhoon PhosphorImager (GE Life Sciences).

Metabolite analysis.

NZ131 and isogenic mutant strains were grown in THY at 37°C for 3.5 h until mid-log phase. Cells were treated with hyaluronidase and were immediately pelleted by centrifugation, washed once with 20 mM Na₂HPO₄-NaH₂PO₄–1 mM MgCl₂ (pH 7.0) buffer, and then resuspended at approximate concentrations of 10⁸ CFU/ml in 20 mM Na₂HPO₄-NaH₂PO₄–1 mM MgCl₂ (pH 7.0) buffer containing 50 mM l-serine. Cells were incubated statically at 37°C for 4 h, at which point cells were pelleted by centrifugation, and the cell-free supernatant was collected for metabolite analysis. Viable CFU counts were determined at 0 h and 4 h by plating 10-fold dilutions of cells on THY to ensure equivalent numbers of cells for different cultures and to confirm that the cells remained viable over the course of the experiment. The concentrations of ammonia in supernatants were determined by using an ammonia assay kit (catalog no. 11 112 732 035; R-Biopharm) according to the manufacturer's instructions but scaled down for use in a 96-well plate. Analysis was performed using a Synergy 2 microplate reader (BioTek).

l-Serine transport analysis.

NZ131 and isogenic mutant strains were grown in CDM at 37°C until the OD₆₀₀ reached approximately 0.3. Linezolid was added to all cultures at a final concentration of 5 μg/ml, and the cultures were incubated for 20 min at 37°C to allow protein synthesis to subside. A 0.5-μCi portion of l-[U-¹⁴C]serine (60 mCi/mmol; MP Radiochemicals) was added to 1-ml aliquots of each culture, and starting 4 min after the addition of l-[U-¹⁴C]serine, 200-μl samples were removed every 10 min for a total of 30 min and were filtered on 0.45-μm-pore-size membrane filters (Millipore) by vacuum filtration. Cells were immediately washed with 5 ml phosphate-buffered saline (PBS) at room temperature. Filters were air dried at 37°C and were placed in scintillation vials containing 5 ml scintillation fluid cocktail (Ultima Gold; Perkin-Elmer). Radioactivity was assessed using a Beckman LS 6000 IC scintillation counter.

Microarray data accession number.

All the microarray data generated in this study have been deposited in the NCBI Gene Expression Ominibus (GEO) database (http://www.ncbi.nlm.nih.gov/geo/) under accession number GSE24860.

RESULTS

Deletion of spy49_0126c results in aberrant growth compared to the growth of wild-type GAS.

In this study, we aimed to investigate the role of the sloR operon in GAS, in view of its previously reported effect on streptolysin O expression (33, 37). sloR is the second gene in an operon that also encodes a predicted member of the YjgF family (spy49_0127), as well as a divergently transcribed putative transcriptional regulator (spy49_0126c) whose function was briefly examined with regard to SLO expression but has never been investigated directly (Fig. 1 A) (33, 37). A temperature-sensitive plasmid containing the upstream and downstream regions of the spy49_0126c gene flanking a spectinomycin resistance cassette (plasmid pBL100 [Table 1]) was used to delete this putative regulator from the chromosome by allelic exchange, generating strain BNL100 (Fig. 1A). Initial characterization of BNL100 included examination of growth compared to that of the parental strain in a variety of growth media. No difference in growth was detected when strains were grown in glucose- and peptide-rich Todd-Hewitt broth with yeast extract (THY) (Fig. 1B), in agreement with previous reports (33). However, when BNL100 was grown in more nutrient-poor media, growth defects became apparent. In the peptide-rich, glucose-low C-medium, BNL100 grew at a rate similar to that of the wild type until the OD₆₀₀ of the culture reached approximately 0.3, at which point the growth rate of the mutant slowed. The growth of BNL100 remained slower than that of the wild type for the remainder of the growth curve, but similar culture yields were reached with the two strains (Fig. 1C). In contrast, in chemically defined medium (CDM), which is rich in glucose but has only free amino acids and no peptides, BNL100 showed a sizeable growth yield defect. Although their initial growth rates were similar to those of wild-type GAS, BNL100 cultures never reached cell densities as high as those of the wild type (Fig. 1D). A single copy of the spy49_0126c gene at a neutral location in the chromosome fully complemented this growth defect (Fig. 1D).

Fig. 1.

Deletion of spy49_0126c affects GAS growth in culture media that differ in nutrient availability. (A) Schematic of the sloR operon in wild-type NZ131 and in the BNL100 mutant, in which spy49_0126c (light shaded arrow) is replaced with the aad9 cassette (open arrow) under the control of the Spc^r promoter, providing spectinomycin resistance. NZ131 chromosomal DNA and spectinomycin cassette DNA are represented by solid and dashed lines, respectively. (B and C) Growth curves of NZ131 (▴) and BNL100 (▪) in THY (B) and C-medium (C). (D) Growth curves of BNL158 (wild type with empty vector) (♦), BNL159 (ΔserR with empty vector) (•), and BNL162 (ΔserR with pBL109) (▿) in CDM. The results shown are representative of three independent experiments.

Deletion of spy49_0126c results in the enhanced expression of multiple operons, some of which are associated with l-serine utilization.

Given the annotation of spy49_0126c as a putative transcriptional regulator and the aberrant growth profiles of BNL100, we sought to identify putative targets of regulation by the spy49_0126c gene product. Microarray analysis was carried out using RNA harvested at mid-log phase from the wild-type and BNL100 strains grown in C-medium. This medium was used in order to enhance the likelihood that the genes responsible for aberrant growth would be identified while allowing for sufficient growth of both strains without the risk of BNL100 entering stationary phase, which might have occurred had CDM been used. Microarray analysis revealed >5-fold-higher expression of four annotated operons by BNL100 than by the parental strain (Table 3). Two of these operons consisted of genes encoded divergently from spy49_0126c, including sloR (spy49_0128) and subunits of a putative V-type Na⁺-ATP synthase (spy49_0130 to -0137). The other two operons comprised a total of three genes, all linked to the utilization of l-serine. serS (spy49_1358) encodes a putative seryl-tRNA synthetase responsible for loading l-serine to cognate uncharged tRNAs. The expression of serS has been reported to be under the control of T-box regulation in S. pyogenes (49), a mechanism in which the concentration of charged tRNAs determines the structure of a riboswitch present in the leader region of the regulated gene, resulting in the formation of a terminator or antiterminator structure that alters the gene expression. sdhBA (spy49_1794 and -1795) encode the putative β and α subunits of l-serine dehydratase, a catabolic enzyme that converts l-serine to ammonium and pyruvate (EC 4.3.1.17). These enzymes are widespread in bacteria (14), and their sequence homology is highly variable (17). sdhBA from GAS are 38% and 41% identical, respectively, to the l-serine dehydratase enzyme subunits from Peptostreptococcus asaccharolyticus, which have demonstrated enzymatic activity (17). Sequence analysis indicates that the GAS β subunit contains the C-terminal ACT domain, including the putative l-serine binding site of the enzyme (Conserved Domain Database, NCBI) (28). Additionally, an increase in sdhBA expression in a ropB (rgg) mutant was accompanied by increased degradation of l-serine with concomitant ammonia production (4, 5). Given this information, in addition to the results presented here, we find it reasonable to conclude that the sdhBA genes encode an l-serine dehydratase that is functional in GAS. No regulator has been shown to directly control the expression of sdhAB in this organism.

Table 3.

Genes upregulated >5-fold in the Δspy49_0126c mutant vs NZ131 by microarray analysis

Gene ID	Operon name	Putative function	Fold changea
Spy49_0127-0128	Spy49_0127, sloR	Putative endoribonuclease, SLO regulator	14.1, 16.8
Spy49_0129-0137	Spy49_0129, ntpI to ntpD	Hypothetical protein, V-type Na+-dependent ATP synthase	15.3–24.5
Spy49_1358	serS	Seryl-tRNA synthetase	6.5
Spy49_1794-1795	sdhBA	l-Serine dehydratase	7.6, 9.11

^aValues for operons containing more than two genes are given as a range for the entire operon.

Expression of sloR and sdhAB is increased in the absence of spy49_0126c.

To validate the microarray results (Table 3), reporter fusions linking the promoter regions of either sloR or sdhBA to the luxAB genes, which encode luciferase, were integrated in single copy into the GAS chromosome. Strains carrying the reporter fusions were grown in THY, and luxAB expression was assessed over the entire growth curves of the bacteria. THY was used in these experiments in order to certify that differences in gene expression levels resulted from the lack of spy49_0126c rather than from differences in growth between strains. Light expression profiles revealed that both sloR expression and sdhAB expression were 5- to 10-fold higher in BNL100 than in the parental strain over the entire course of growth (Fig. 2A and B). A luciferase reporter fused to the promoter of the recA gene, a housekeeping gene expressed during exponential phase (30) but not predicted to be regulated by SerR, showed no difference in expression between strains, demonstrating that the upregulation of sloR and sdhAB was not due to a general increase in gene expression in BNL100 (data not shown). These data, in addition to the microarray results, indicate that the spy49_1026c gene product represses the transcription of both the sloR and sdhAB operons throughout the exponential growth of S. pyogenes. Given that the spy49_0126c gene product represses genes involved in l-serine utilization, we have termed the gene serR (for serine catabolism regulator).

Fig. 2.

Deletion of spy49_0126c results in increased transcription from the promoters of target genes. Shown are growth curves (open symbols and dashed lines) and expression curves (filled symbols and solid lines) (relative light units [RLU], calculated as counts per minute/OD₆₀₀) of the wild-type (triangles) and Δspy49_0126c (squares) strains grown in THY. (A) Growth and expression curves of strains carrying the P_sloR-luxAB reporter (strains BNL105 and BNL106). (B) Growth and expression curves of strains carrying the P_sdh-luxAB reporter (strains BNL141 and BNL142). The results shown are representative of three independent experiments.

SerR binds directly to the sloR and sdhAB promoters.

To determine if the repression of the sloR, sdhBA, and serS operons was due to direct or indirect regulation by SerR, His₆-tagged SerR (SerR_His) was purified from E. coli and was incubated with fluorescently labeled DNA fragments of the sloR, sdhBA, and serS promoter regions. A fluorescently labeled DNA fragment of an rRNA promoter was used as a negative control. Electrophoretic mobility shift assay (EMSA) analysis showed shifts of both the sloR and sdhBA probes in the presence of SerR_His (Fig. 3 A and B), indicating that SerR bound directly to the promoter regions of these operons. These shifts could be interrupted by the addition of a cold sloR (Fig. 3A, 5th lane) or a cold sdhBA (Fig. 3B, 5th lane) competitor, but not by a cold competitor consisting of the rRNA promoter (Fig. 3A and B, 6th lanes), demonstrating the specificity of the binding interactions. Negligible binding was seen when SerR_His was incubated with the rRNA probe, even at the highest protein concentrations used (Fig. 3C). Interestingly, no shift was detected when SerR_His was incubated with a serS promoter probe (Fig. 3D), suggesting that the upregulation of serS expression in BNL100 is due to indirect regulation by SerR. This would be consistent with the idea that serS is under the control of T-box regulation. Increased transcription of serS would be observed if levels of l-serine in the cell were decreased, which is the expected result of increased l-serine dehydratase (sdhBA) expression in BNL100. Thus, transcription of the sloR and sdhBA operons is directly repressed by SerR, while repression of serS is likely an indirect effect of SerR via inhibition of l-serine degradation.

Fig. 3.

SerR_His binds to the promoter regions of sloR and sdhBA. The binding of SerR_His to labeled promoter regions of sloR (A), sdhBA (B), rRNA (C), and serS (D) was analyzed by EMSA. The concentration of SerR_His is given above each lane. All reaction mixtures included 10 nM labeled probe. Where indicated, a 10-fold molar excess of an unlabeled probe (specific) or an unlabeled rRNA probe (nonspecific) was included. The rRNA promoter (PrRNA) was included as a negative control.

The growth culture yield defect of BNL100 in CDM can be rescued by the addition of extra l-serine.

As mentioned previously, SdhBA is an l-serine dehydratase that catalyzes the degradation of l-serine into pyruvate and ammonia. Based on the upregulation of sdhBA in BNL100, it seemed likely that the growth yield defect of BNL100 in CDM could be the result of increased l-serine degradation and subsequent faster depletion from the medium. Given the polyauxotrophy of GAS for 15 amino acids (10), including l-serine, depletion of l-serine from the medium would inhibit further growth, since no l-serine would be available for protein synthesis. Thus, in essence, strain BNL100 could have an increased nutritional requirement for l-serine due to increased SdhBA expression, and the traditional CDM recipe (47) might not contain enough l-serine to satisfy this requirement. To test this hypothesis, extra l-serine and other l-amino acids tested individually were added to CDM to a final concentration of 600 mg/liter, 6 times that normally included in the medium, and the growth of wild-type GAS and BNL100 (ΔserR) was monitored. Whereas the addition of l-serine and other l-amino acids had no effect on the growth of the parental strain, NZ131 (data not shown), the addition of l-serine complemented the growth yield defect of BNL100, nearly doubling the final yield of the mutant culture (Fig. 4). l-Amino acids other than l-serine had no effect on the growth of BNL100 (Fig. 4). Furthermore, supplementation of CDM with l-serine complemented the culture yield defect of BNL100 in a concentration-dependent manner (data not shown). These data are consistent with the hypothesis that l-serine is limiting in CDM due to the upregulation of sdhBA and consequent l-serine depletion from the medium in BNL100 cultures, resulting in the growth yield defect. The addition of extra l-serine is likely better able to satisfy the nutritional requirement of BNL100 for l-serine than CDM alone, allowing the culture to reach greater cell densities. This is also consistent with the lack of a growth defect in THY (Fig. 1B), since the peptides present in this medium could provide sufficient serine to allow for greater culture growth.

Fig. 4.

Supplementation of CDM with l-serine rescues the reduced culture yield of BNL100. All strains were grown in CDM plus 0.5% glucose at 37°C. The medium was supplemented with selected amino acids at a final concentration of 600 mg/liter. Strains and amino acid supplementation are as follows: NZ131 (▴), BNL100 (□), and BNL100 with extra l-serine (○), l-alanine (▵), l-arginine (⋄), l-histidine (▿), or l-threonine (). The results shown are representative of three independent experiments.

Deletion of sdhB in BNL100 can complement the culture yield defect in CDM by preventing l-serine degradation.

If the increased activity of SdhBA was indeed responsible for early depletion of l-serine from the medium and the subsequent early cessation of growth of BNL100, we reasoned that deletion of sdhBA should fully restore the culture yields of BNL100 grown in CDM to wild-type levels. We deleted sdhB from the chromosome by allelic exchange in NZ131 and BNL100 to generate BNL116 and BNL117, respectively. In the sdhBA operon, the sdhB gene is located proximally to the promoter, followed by sdhA. The chloramphenicol resistance cassette used to replace the sdhB gene included a transcriptional terminator, thereby likely disrupting the expression of sdhA as well. NZ131 (wild type), BNL100 (ΔserR), BNL116 (ΔsdhB), and BNL117 (ΔserR ΔsdhB) were grown in CDM, and culture growth was assessed. Deletion of sdhB in NZ131 had no effect on the growth yield of the culture (Fig. 5 A, compare NZ131 to BNL116). However, deletion of sdhB in BNL100 was sufficient to fully restore growth culture yields to wild-type levels without the addition of extra l-serine (Fig. 5A, compare BNL100 to BNL117). These results support the hypothesis that increased sdhBA expression in BNL100 contributes to the reduced culture yield in CDM, likely by increasing the nutritional requirement for l-serine.

Fig. 5.

Deletion of sdhB rescues the reduced culture yield of BNL100 in CDM by preventing the degradation of l-serine. (A) Deletion of sdhB has no effect on the growth of NZ131 but restores the culture yield of BNL100 to wild-type levels when it is grown in CDM plus 0.5% glucose at 37°C. The strains shown are NZ131 (wild type) (filled bar), BNL100 (ΔserR) (light shaded bar), BNL116 (ΔsdhB) (open bar), and BNL117 (ΔserR ΔsdhB) (dark shaded bar). (B) Deletion of sdhB prevents the degradation of l-serine to pyruvate and ammonia. NZ131 (wild type) (filled bar), BNL100 (ΔserR) (shaded bar), and BNL117 (ΔserR ΔsdhB) (open bar) were harvested during mid-log phase and were resuspended to equal cell densities in 20 mM Na₂HPO₄-NaH₂PO₄–1 mM MgCl₂ (pH 7.0) buffer containing 50 mM l-serine. Ammonia concentrations in cell-free supernatants were determined after 4 h. The results shown are averages and standard errors of the means (error bars) for three independent experiments. Asterisks indicate a significant difference (**, P < 0.001) by one-way analysis of variance with Tukey's multiple-comparison test (GraphPad Prism, version 5.0).

To further test our hypothesis, we asked whether sdhBA expression correlated with ammonia excretion, since l-serine degradation by SdhBA should produce pyruvate and ammonia. We hypothesized that BNL100 would produce more ammonia than wild-type NZ131 due to increased expression of sdhBA, whereas BNL117 should produce less ammonia due to the lack of SdhBA. The results of our experiment supported this hypothesis. NZ131 (wild type), BNL100 (ΔserR), and BNL117 (ΔserR ΔsdhB) were harvested from mid-log-phase cultures, resuspended in a neutral buffer containing 50 mM l-serine, and incubated at 37°C to allow for the enzymatic degradation of l-serine and the excretion of ammonia. After 4 h, NZ131 supernatants contained 0.797 ± 0.078 mM ammonia whereas BNL100 supernatants contained >3-fold more ammonia (2.726 ± 0.0116 mM) (Fig. 5B). This indicated that increased expression of sdhBA correlated with increased ammonia production in the presence of l-serine, consistent with l-serine degradation by SdhBA. Additionally, deletion of sdhB (BNL117) resulted in the production of only 24.66 ± 0.59 μM ammonia after 4 h, >100 times less than the amount produced by BNL100, demonstrating that enhanced ammonia production by BNL100 in the presence of l-serine depended on the expression of sdhBA (Fig. 5B). These data provide further evidence that BNL100 has an increased nutritional requirement for l-serine due to increased expression of sdhBA and consequent l-serine degradation.

The sloR operon modulates the rate of l-serine consumption in GAS.

Unexpectedly, strain BNL161 (ΔserR ΔsloR), generated by deletion of sloR in BNL100, did not have a growth yield defect in CDM (Fig. 6 A). Previous publications have indicated roles for SloR activity in SLO regulation at both transcriptional and secretory steps (33, 37) but have never linked this gene to any metabolic role in the cell. The ability of the sloR deletion to complement the growth yield defect of BNL100 in CDM suggested that SloR plays a role in l-serine catabolism (Fig. 6A, BNL161 versus BNL100). Complementation with a single copy of sloR (BNL164) resulted in a statistically significant decrease in culture yield for the ΔserR ΔsloR double mutant but only partially complemented the phenotype compared to BNL100 (ΔserR) (Fig. 6A), bringing into question the polarity of the sloR mutation. We confirmed that sloR and downstream genes are coexpressed on a polycistronic message by amplifying the contiguous region spanning sloR to spy0148 using PCR where cDNA was the template source (results not shown). As such, we expect that ΔsloR mutations have polar effects on the downstream ATP synthase genes. Consequently, these downstream genes may also contribute to serine catabolism in GAS.

Fig. 6.

Deletion of sloR rescues the reduced culture yield of BNL100 in CDM by altering the rate of l-serine consumption in GAS. (A) Deletion of sloR rescues the reduced culture yield of BNL100 but does not affect the growth of NZ131, while complementation with a single copy of sloR partially restores the growth defect when serR is absent. Results are shown for strains NZ131 (wild type) (filled bar), BNL100 (ΔserR) (dark shaded bar), BNL160 (ΔsloR with empty vector) (open bar), BNL163 (ΔsloR with pBL110) (striped bar), BNL161 (ΔserR ΔsloR with empty vector) (light shaded bar), and BNL114 (ΔserR ΔsloR with pBL110) (checkered bar). (B) The sloR operon alters the rate of l-serine consumption in GAS. All strains were grown in CDM plus 0.5% glucose at 37°C until mid-log phase, at which point l-[U-¹⁴C]serine was added to all cultures. Radioactive counts contained within cells from 200-μl culture aliquots were measured at the indicated time points. Results are shown for strains NZ131 (wild type) (▴), BNL100 (ΔserR) (▪), and BNL114 (ΔserR ΔsloR) (⋄). The results shown are averages and standard errors of the means (error bars) for three independent experiments. Asterisks indicate a significant difference (**, P < 0.001) by one-way analysis of variance with Tukey's multiple-comparison test (GraphPad Prism, version 5.0).

We reasoned that the sloR operon may be acting to alter l-serine catabolism by transcriptional or posttranscriptional means. One possibility was that the sloR operon acts to enhance sdhBA transcription and that deletion of sloR would reduce SdhBA levels and thereby decrease the l-serine nutritional requirement in BNL114, allowing for higher culture yields. Using our sdhBA reporter fusion, we found that this was not the case. Light expression profiles revealed that expression from the sdhBA promoter is equivalently high in BNL114 and BNL100 (data not shown), indicating that SloR does not modulate l-serine catabolism via regulation of sdhBA transcription.

SloR is predicted to have nine transmembrane domains (TopPred, Mobyle server, Pasteur Institute) (33). Thus, we next hypothesized that the sloR operon was affecting l-serine catabolism by increasing the rate of l-serine transport into the cell. If this were the case, deletion of sloR in BNL100 would decrease the rate at which l-serine was transported into the cell, thereby disrupting the enhanced l-serine turnover we hypothesized to be present in strain BNL100. To test this hypothesis, we measured the ability of GAS strains to take up radiolabeled l-serine at a point in exponential growth prior to amino acid depletion, with the prediction that inactivation of the sloR operon would suppress the serine deficiency caused by the deletion of serR. l-[U-¹⁴C]serine transport experiments demonstrated that BNL100 (ΔserR) does indeed transport l-serine at a higher rate than that of wild-type NZ131 (Fig. 6B). After 30 min, BNL100 cells contained 19,675 ± 376 cpm, whereas NZ131 cells contained 1,260 ± 27 cpm, a difference of approximately 15-fold between the strains (Fig. 6B). In agreement with our hypothesis, BNL114 (ΔserR ΔsloR) transported l-serine into cells at rates comparable to those of NZ131, containing only 925 ± 80 cpm after 30 min (Fig. 6B). These data provide evidence that the sloR operon is involved in modulating l-serine consumption in S. pyogenes, a novel function for this operon.

DISCUSSION

Complex regulatory networks have been shown to coordinate the expression of both virulence- and metabolism-associated genes in S. pyogenes. In this report, we characterize a transcriptional regulator that we have named SerR (for serine catabolism regulator). The serR gene (spy49_0126c) is divergently transcribed from a polycistronic operon containing the sloR gene, which has previously been shown to influence the expression of the virulence factor streptolysin O (33, 37). In this report, we describe a novel role for the sloR operon in l-serine utilization in GAS. Our data provide evidence that the sloR operon and sdhBA, the latter encoding an l-serine dehydratase, are repressed by SerR. Deletion of serR from the chromosome results in increased transcription of the sloR and sdhBA operons and an increased nutritional requirement for l-serine, leading to decreased culture yields in chemically defined medium (CDM) unless additional l-serine is provided. The negative transcriptional regulation of these genes by SerR is direct, as evidenced by the fact that purified SerR binds specifically to the sloR and sdhBA promoters in vitro. The SerR protein is predicted to have an N-terminal PER-ARNT-SIM (PAS) domain and a C-terminal helix-turn-helix that could facilitate this DNA binding (Pfam protein families database, Sanger Institute). Additionally, the decreased culture yield of the serR mutant in CDM can be rescued by the deletion of sdhB or sloR, presumably by preventing the rapid degradation of l-serine. To our knowledge, this is the first report identifying a transcriptional regulator that directly regulates the expression of genes involved in l-serine catabolism in S. pyogenes or in any other Gram-positive bacterium.

The physiological importance of regulation by SerR when l-serine is not overly abundant is clear in this study, since increased l-serine catabolism due to the absence of SerR results in restricted cell growth (Fig. 1B and Fig. 4, 5, and 6). However, the importance of l-serine catabolism to GAS physiology remains unclear. It seems paradoxical that S. pyogenes would maintain a mechanism for degrading l-serine considering that it is auxotrophic for this amino acid; however, serR, sloR, and sdhBA are conserved in all sequenced GAS genomes (http://www.microbesonline.org) (12). We hypothesize that l-serine degradation may contribute to GAS survival and pathogenesis in ways yet to be defined. During our investigation, we observed that NZ131 cells harvested at mid-log phase are less sensitive to low pH in the presence of l-serine than in its absence, while the presence of l-serine had no effect on the pH tolerance of a ΔsdhB mutant (our unpublished results), suggesting that l-serine catabolism may contribute to pH tolerance in GAS. We also observed that the sdhB mutant fails to aggregate and does not pellet readily compared to its wild-type counterpart (our unpublished data), suggesting that surface components, such as extracellular polysaccharide or membrane- and cell wall-associated proteins, are altered in the sdhB mutant. The mechanism(s) behind these physiological changes resulting from the deletion of sdhB remains to be elucidated, but it is clear that additional work will be needed to determine how l-serine catabolism and its regulation affect GAS physiology and pathogenicity.

The ability of the sloR mutation to rescue the growth phenotype of the serR mutant came as a surprise, since the sloR operon had not previously been linked to l-serine catabolism. However, the uptake studies using radiolabeled l-serine presented here provide evidence that the sloR operon influences the rate at which l-serine can be consumed by GAS. Two mechanisms by which SloR and the ATP synthase genes could be acting are the modulation of l-serine uptake, the gene products potentially being transporters themselves, or the modulation of SdhBA activity at a posttranscriptional level. Further experiments will be needed to precisely define the roles of SloR and the downstream ATP synthase in l-serine catabolism. It has been suggested previously that SloR acts as an environmental sensor (33). If this hypothesis is true, given the involvement of SloR in l-serine catabolism, we speculate that l-serine availability may be the signal for SloR, ultimately influencing SLO expression through an indirect mechanism. Furthermore, a previous study of S. pyogenes concluded that l-serine shares a common transporter with other neutral amino acids and that uptake may be driven via symport with protons (36). Thus, the putative ATP synthase composed of these downstream gene products might be involved in the generation and maintenance of gradients required for l-serine uptake.

Given that sloR is the second gene in a polycistronic operon, our sloR expression data concurrently demonstrate that the first gene in the operon, spy49_0127, is also negatively regulated by SerR. spy49_0127 is predicted to encode a YjgF family member. The YjgF family includes proteins found in all kingdoms of life; however, no definitive function for any of its members has been experimentally demonstrated (1). Interestingly, various reports have suggested roles for YjgF family members in amino acid metabolism, including roles in isoleucine biosynthesis and l-threonine deamination (6, 38). Crystallization of TdcF, a YjgF family member from E. coli, revealed that l-serine, in addition to l-threonine and 2-ketobutyrate, can bind in the cavity of TdcF, although l-serine did not fully occupy all of the binding sites (1). Whether the spy49_0127 gene product can in fact bind l-serine, as can TdcF, or perhaps binds an intermediate of l-serine catabolism remains to be determined. However, the cotranscription of spy49_0127 and sloR, the negative regulation by SerR, and the suggested roles for homologous family members in amino acid metabolism argue for future evaluation of the role of this YjgF family member in l-serine catabolism in GAS.

The distance between the sdhBA operon and serR in the genome initially seemed odd, given that the other genes regulated by SerR, with the exception of SerS, are in direct proximity to SerR. However, regulation of these genes by a common regulator may occur in at least one other bacterial species. Enterococcus faecalis has a gene homologous to serR, EF0096, that is divergently transcribed from an operon consisting of four putative genes: a sloR homologue, serS, and sdhBA (http://www.microbesonline.org) (12). These genes have been shown to be regulated by CcpA in E. faecalis (22), but regulation by the EF0096 gene product was not investigated. In many organisms, multiple regulators contribute to the expression of amino acid catabolic genes; thus, regulation by both CcpA and the SerR homologue may coordinate the expression of l-serine utilization genes in E. faecalis. This type of coordinated regulation likely occurs in GAS as well. Many different studies have identified transcriptional regulators that affect the expression of genes identified here as being involved in l-serine catabolism, although it should be noted that none of these regulators have been shown to directly bind the promoters of these genes. sdhB expression was increased in a ccpA covR double mutant in GAS, although no difference was seen in either single mutant (41), and deletion of the stand-alone regulator ropB (rgg) results in increased expression of sdhBA as well as of genes involved in the arginine demininase pathway (4, 5). spy49_0127 and sloR were upregulated both in a covR mutant (15) and in a vicR mutant (24), and sloR expression was upregulated at 29°C relative to 37°C (44). Interestingly, Streptococcus agalactiae and Streptococcus dysgalactiae have homologues of SerR that are transcribed divergently from homologues of SloR (http://www.microbesonline.org) (12). S. dysgalactiae also has a putative YjgF family member upstream of the sloR homologue, as is seen in GAS. Furthermore, the sdhBA operons in these species are located in regions of the genome comparable to those in GAS, just downstream of trmU in all species. Other streptococcal species do not have SerR or SloR homologues that we have been able to identify by BLAST analysis. However, for species that do have SerR and SloR homologues, including E. faecalis, we put forward the possibility that SerR and its homologues, in coordination with other regulatory proteins, act as negative transcriptional regulators to modulate l-serine acquisition and catabolism.