Transcriptional Regulation of a Bacteriophage Encoded Extracellular DNase (Spd-3) by Rgg in Streptococcus pyogenes

The Streptococcus pyogenes transcriptional regulator Rgg controls the expression of virulence-associated genes encoded both within the core genome and within horizontally transmissible DNA such as temperate bacteriophage. Previously, we showed that Rgg binds to the non-coding DNA upstream of the bacteriophage gene encoding an extracellular DNase Spd-3. In the current study, we further characterized Rgg-mediated regulation of spd-3 expression. Two spd-3 transcripts were identified by northern blotting. The 5′ ends were 27 and 594 nucleotides upstream of the start codon as determined with primer extension analysis and 5′ RACE (rapid amplification of c-DNA ends), respectively. Results obtained with gel shift assays showed that purified Rgg bound specifically to non-coding DNA containing the promoters of both transcripts. Transcriptional fusion analyses confirmed the presence of Rgg-repressible promoters within these DNA regions. In addition, repression was associated with direct DNA binding by Rgg as determined with chromatin immunoprecipitation (ChIP) coupled with quantitative PCR (qPCR). The results show that the chromosomally encoded transcriptional regulator, Rgg, directly represses both bacteriophage promoters controlling the expression of Spd-3. The results provide new information regarding the regulation of prophage encoded virulence factors of S. pyogenes and highlight the complex evolutionary history of S. pyogenes and temperate bacteriophage.


Introduction
Streptococcus pyogenes causes several human diseases ranging in severity from self-limiting pharyngitis to life-threatening necrotizing fasciitis and streptococcal toxic shock syndrome [1]. The virulence of the pathogen varies temporally over the course of decades due to changes in both the pathogen and human immunity [2,3]. The determination of the genome sequences of several isolates revealed a theoretically limitless pan-genome, or mobilome, comprised of mobile genetic elements (MGE) including bacteriophages, integrative conjugative elements, transposons, and insertion sequences [3][4][5]. MGEs significantly alter the composition of the chromosome. For example, between two and eight complete or partial bacteriophage genomes are present in the genomes of S. pyogenes and bacteriophage DNA can account for up to 12% of the chromosome [6,7]. Often, the bacteriophage encode virulence factors including superantigens [8] and extracellular nucleases [9,10], which profoundly influence interactions between the pathogen and its human host. Thus, chromosomal heterogeneity, including variation in the number and types of prophage within a chromosome, is responsible for much of the genetic diversity observed among clinical isolates and contributes to the clinical and temporal variation in the outcome of human colonization with S. pyogenes [7,[11][12][13].
S. pyogenes can produce up to four extracellular DNases [14,15]. One (MF-1/DNaseB) is chromosomally encoded and is adjacent to rgg (also known as ropB [16]), which encodes a global transcriptional regulator [17]. The remainder are encoded by prophage. Therefore, the number of extracellular DNases potentially produced by an isolate varies depending on the prophage content of the chromosome. Extracellular DNases have long been thought to be important in liquefying pus to promote bacterial dissemination. In addition, prophage-encoded extracellular DNase degrades bacterial DNA following induction of the lytic phase, which can further reduce viscosity at the site of colonization and promote transmission of the progeny virions to new bacterial hosts [9]. More recently, the prophage encoded extracellular nuclease Sda1 was found to degrade neutrophil extracellular traps (NETs) [18]. NETs are a component of innate immunity and are composed of a scaffold of neutrophil-derived chromatin and antimicrobial peptides, which entraps and kills microbes, including S. pyogenes [19]. Degradation of bacterial DNA also decreases macrophage mediated killing of S. pyogenes due to decreased stimulation of toll-like receptor 9, which recognizes unmethylated CpG-rich DNA [20]. Thus, extracellular DNase's promote pathogen dissemination and survival, although some appear to be relatively more important than others [18,21]. Importantly, the exoproteins contribute to virulence in both mouse models of invasive infection [20] and in a cynomolgus macaque model of pharyngitis [21].
Inactivation of the gene encoding the transcriptional regulator Rgg increased expression of both SdaB (Spy49_1692c; Mf-1) and Spd-3 (Spy49_1455) in the post-exponential phase of growth [17]. Subsequently, we found that Rgg binds to non-coding prophage DNA upstream of spd-3 [24]. The purpose of the current study was to characterize further the role of Rgg in the regulation of the prophage encoded DNase Spd-3.

Identification of spd-3 Transcripts
As an initial step to characterize the regulation of spd-3 expression, northern blotting was done using RNA isolated during the post-exponential phase of growth from both the wild-type S. pyogenes strain NZ131 and an rgg mutant. Two distinct transcripts were detected and both were more abundant in the mutant strain compared to the wild-type strain (Fig. 1), which was consistent with our previous finding that Rgg represses spd-3 expression [17]. The more abundant transcript was approximately 925 bp in length and accounted for 65% of the transcript signal, as determined by densitometry.

Mapping the spd-3 Transcriptional Start Sites
Primer extension analysis was used to determine the 59 termini of the two transcripts. Extension with primer spd3PEc_96 (Table 1) showed a transcript that originated 27 bp upstream of the predicted spd-3 open reading frame (ORF) (Fig. 2). This origin, coupled with a putative transcriptional terminator 91 bp down-stream of the spd-3 ORF, predicted a 919 bp transcript, which corresponded to the more abundant approximately 925 bp transcript detected by northern blotting (Fig. 1). A variety of primers were used in attempts to identify the start site of the longer transcript by using primer extension; however, we were unable to do so, possibly due to secondary structure formed within the large untranslated region. As an alternative approach, 59 RACE was used and the results showed that the 59 terminus was 594 bp upstream of the spd-3 start codon (Fig 3). The results predicted a 1,487 bp transcript, which also correlated with the size of the larger transcript identified with northern blotting (Fig. 1). In addition, the start of transcription coincided with the non-coding DNA region previously shown to be bound by Rgg in vivo during the exponential phase of growth [24]. Analyses of the DNA proximal to the transcriptional start sites revealed the presence of putative -10 and -35 RNA polymerase binding sites (Fig. 3). The two transcriptional start sites were designated P 1 and P 2 (Fig. 3).

Rgg Binds Specifically to DNA Containing the P1 and P2 Promoters
ChIP coupled with DNA genechips (ChIP-chip) and gel-shift assays previously showed that Rgg binds to P 1 [25]; however, we did not previously detect binding to P 2 . Therefore, we re-analyzed Rgg binding to both sites by using the more sensitive ChIP-qPCR procedure. As expected, DNA containing P 1 was enriched by approximately 9-fold in strain SA5, which encodes an Rgg-Myc fusion protein used to facilitate immunoprecipitation, compared to the control strain (Fig. 4). In addition, the P 2 region was enriched nearly 75-fold (Fig. 4). The results show that Rgg binds to both the P 1 and P 2 DNA (Fig. 4). Binding near P2 was further evaluated with gel-shift assays and the results confirmed that Rgg specifically binds to DNA containing P 2 (Fig. 5).

Rgg Directly Regulates spd-3 Expression
To characterize the two spd-3 promoters further, prophage DNA regions bound by Rgg (P1 and P2) were cloned adjacent to a promoterless firefly luciferase (luc) gene present in the shuttle plasmid pKSM720 [26]. The recombinant plasmids were used to transform both the wild-type and the rgg mutant strains to determine if expression from P1 and P2 was regulated by Rgg. The use of the plasmid based reporter system also allowed us to measure promoter activity apart from the prophage, thus avoiding confounding factors potentially associated with prophage induction, such as variation in gene copy number. Promoter activity was detected with P 2 DNA and the presence of Rgg in the wildtype strain decreased transcription, consistent with other data indicating that Rgg represses spd-3 transcription (Table 2). Similarly, a fragment containing both spd-3 promoters (P1 and P2), showed significantly more activity compared to the fragment containing only P2 (Table 2). Again, more promoter activity was detected in the rgg mutant strain, consistent with Rgg-dependent repression of spd-3 expression by direct binding to the promoter regions (Table 2).

Rgg Acts in cis to regulate spd-3 Promoters
Finally, we used the transcriptional reporter system and ChiP-qPCR to determine if Rgg bound in cis to regulate expression of the luc gene. One PCR primer was specific to luc and a second was upstream of P 1 . The results showed that Rgg bound directly to the plasmid DNA to repress expression of luc (Fig. 6).

Discussion
Temperate bacteriophage can ferry virulence-associated genes encoding a variety of toxigenic and enzymatic exoproteins among bacteria. In S. pyogenes, prophages encode superantigens, which are responsible for toxic shock syndrome; extracellular DNases, which assist in the pathogen's escape from the innate immune response; and surface exposed proteins, which have not been characterized. Despite the importance of the gene products to human diseases caused by S. pyogenes, comparatively little is known about the regulation of their expression. Here, we examined the role of Rgg in the repression of the bacteriophage encoded extracellular DNase Spd-3. Two spd-3 transcripts were identified and the 59 termini determined. The abundance of both transcripts was elevated in an rgg mutant strain and Rgg bound to DNA containing the promoters of both transcripts, indicating that Rgg binding represses expression. To analyze regulation outside the context of the prophage, which can excise from the chromosome and thereby increase spd-3 copy number, a plasmid-based reporter system was used to measure spd-3 promoter activity. The results confirmed that Rgg repressed expression from both promoters. Finally we showed with ChIP-qPCR that Rgg bound to DNA in cis to effect transcription of the reporter gene. The results show that the chromosomally encoded regulator Rgg represses expression of the prophage-encoded virulence gene spd-3 expression by direct binding to prophage promoters.
Rgg regulates dozens of genes in strain NZ131 in a growthphase associated manner, including many known to contribute to virulence [16,27,28]. It does so, in part, by changing its DNA binding specificity in response to glycolytic flux via direct proteinprotein interaction with LacD.1 [29]. In addition, Rgg binds to hydrophobic peptides, which modulates regulatory activity [30,31]. The genes controlled by Rgg are functionally diverse and include both cell-associated proteins and soluble exoproteins  associated with virulence [17,25]. For example, Rgg represses expression of the cell wall-associated antiphagocytic M protein in the exponential phase of growth and activates expression of a secreted cysteine protease SpeB in the post-exponential phase [16,27,32]. The results of the current study, together with results from previous studies, show that Rgg coordinates the expression of both prophage and chromosomally-encoded virulence genes in response to environmental cues. Consistent with these findings, Rgg contributes to the outcome of S. pyogenes infection as assessed with animal models and as observed in human epidemics of invasive disease [33] [34].
Excision of prophage from the chromosome is associated with increased gene copy number and often with simultaneous expression of prophage-encoded virulence genes. In some instances the repressor of the lytic phase also directly represses expression the prophage-encoded virulence genes. For example, expression of the prophage-encoded Shiga toxin (StxAB) in Escherichia coli is primarily controlled by a prophage repressor that also controls lysogeny [35,36]; thus stxAB expression is dependent on prophage induction. In contrast, expression of the prophageencoded cholera toxin (Ctx) is controlled by several chromosomally encoded regulators and is not dependent on induction of the  encoding prophage [37]. Thus two paradigms have emerged from the study of the regulation of prophage encoded virulence factors.
In S. pyogenes, expression of prophage encoded virulence genes, including those encoding superantigens, has been associated with the induction of prophage [4,8,9]; however, induction does not appear to be necessary for expression [4]. Moreover, expression of the phage-encoded extracellular DNase Sdn (SpyM3_1409) decreases following mitomycin C induction of prophage, despite an increase in sdn copy number [4]. In this study, we showed that Rgg directly represses the promoters controlling spd-3 expression, even in the absence of prophage induction, as determined by using a plasmid based reporter system. Although our investigation focused on the regulation of spd-3 expression, the results are consistent with a model in which the expression of prophage encoded virulence factors in S. pyogenes is not dependent on induction the lytic cycle and is controlled, at least in part, by chromosomally encoded regulatory loci. Given the relevance of prophage encoded superantigens and secreted DNases to human disease, additional investigation into the regulation of other prophage-encoded virulence factors is warranted.
Phenotypic variation is a hallmark of many pathogens, including S. pyogenes, and can result in heterogeneous clinical outcomes of infection. We previously showed that Rgg binds to non-coding DNA upstream of a prophage integrase/excisionase (Spy49_0746c) to repress expression [17,24]. Increased expression of the integrase/exicisionase in an rgg mutant strain was associated with a decrease in the frequency of prophage excision from the chromosome [17]. The current study extends the idea that Rgg regulates specific prophage encoded genes, which alters the phenotype of the pathogen. Given the tremendous variation in the number and composition of bacteriophage among different isolates of S. pyogenes [7], we speculate that the direct regulation of MGEs by Rgg contributes, directly or indirectly, to the variation in the Rgg regulon observed among various isolates of S. pyogenes [38] and potentially the clinical outcome of human infection.

Bacterial Strains and Culture Conditions
S. pyogenes strain NZ131 (serotype M49) was isolated from a patient with acute post-streptococcal glomerulonephritis (Table 3) [22]. NZ131 and its genetic derivatives including rgg-(32) and SA5 and have been previously described (24). S. pyogenes strains were grown with Todd-Hewitt broth (Becton Dickinson, Spark, MD) containing 0.2% (wt/vol) yeast extract at 37uC in a 5% CO 2 atmosphere without agitation. E. coli DH5a was grown with Luria-Bertani medium at 37uC with agitation. When necessary, antibiotics was added to the growth media at the following concentrations: carbenicillin at 100 mg/ml for E. coli; spectinomycin at 100 mg/ml for both E. coli and S. pyogenes; erythromycin at 2.5 mg/ml for S. pyogenes; kanamycin at 50 mg/ml for E. coli and 500 mg/ml for S. pyogenes.

DNA Manipulation
To isolate plasmid DNA from E. coli, either the QIAprep Spin Miniprep Kit (Qiagen, Valencia, CA) or Maxi/Midi prep purification systems (Qiagen) was used. DNA fragments were PCR amplified with GoTaq DNA polymerase (Promega, Madison, WI) and the amplified DNA was separated by using agarose gel electrophoresis and purified using the SpinPrep Gel DNA kit (EMD Milliopore, Darmstadt, Germany). DNA sequencing to confirm various constructs was done at Iowa State University (Ames, IA).
The S. pyogenes strains containing the transcriptional fusion plasmids were grown with THY broth to the exponential phase of growth (A 600 ,0.35) and luciferase activity was measured according to manufacturer's instructions (Promega).

Electrophoretic Mobility Shift Assays (EMSA)
An Rgg-maltose binding protein fusion protein (Rgg-MBP) was expressed in E. coli and purified as previously described [24]. Noncoding DNA upstream of spd-3 (corresponding to nucleotides 21 to 2457 bp relative to the spd-3 start codon) was amplified by using primers spd3fwd1 and spd3rev1 primers and NZ131 genomic DNA as a template ( Table 1). The fragment was isolated by using agarose gel electrophoresis, purified, and cloned into pGEM-T-easy (Promega, Madison, WI) to create pSA29. As a non-specific control of DNA binding, a similarly sized fragment was similarly prepared by using groEL2fwd and groEL2rev primers, which are specific to the groEL ORF ( Table 1). The fragments were excised from pGEM-T easy, gel purified, dephosphorylated, and end labeled with [c 32 P] ATP using polynucleotide kinase. Different amounts of Rgg-MBP were incubated in 25 ml of binding buffer (25 mM Tris-Cl, pH 7.5, 0.1 mM EDTA, 75 mM NaCl, 1 mM dithiothreitol, 10% glycerol, and 0.5 mg/ml of calf thymus DNA) at room temperature for 20 min. Competition experiments were conducted by including unlabeled DNA prior to protein addition. The reaction mixtures were separated with a 6% nondenaturing polyacrylamide gel. The gels were dried, exposed to an Amersham Biosciences storage

RNA Isolation and Northern Blotting
Overnight cultures of NZ131 and the rgg mutant were inoculated into 40-ml THY broth to an A 600 of 0.08. The cultures were grown to the post-exponential phase of growth (A 600 , 0.6). Total RNA was isolated as described previously [17]. The concentration and quality of RNA was assessed with an Agilent 2100 Bioanalyzer (Agilent, Palo Alto, CA) using an RNA 6000 Nano LabChip kit (Agilent). Fifteen micrograms of total RNA from each strain was separated with a 1.5% agarose-0.66 M formaldehyde gel in morpholinepropanesulfonic acid (MOPS) running buffer (20 mM MOPS, 10 mM sodium acetate, 2 mM EDTA; pH 7.0). RNA was blotted onto Hybond N + membranes (GE Healthcare Biosciences, Pittsburgh, PA.) with the Turboblotter alkaline transfer system (Schleicher & Schuell, Keene, N.H.), according to the manufacturer's instructions. RNA was fixed to the membrane by baking at 80uC for 30 min. The spd-3 coding DNA was amplified using spd3_cfwd and spd3_crev primers (Table 1), separated by agarose gel electrophoresis, the fragment excised from the gel, and purified as described above. Using the purified spd-3 coding DNA as a template, [a-32 P]dCTP radiolabelled probes were synthesized by the random-primed method (Ready-To-Go Labeling Kit; Pharmacia). Membranes were hybridized under aqueous conditions at 65uC with the radiolabelled probes. The blots were washed and exposed to an Amersham Biosciences storage phosphor screen and imaged with a Typhoon instrument.

Primer Extension Analysis
The 59 end of the spd-3 transcript was determined with 15 mg of RNA isolated from NZ131 and the rgg mutant using an AMV reverse transcriptase primer extension kit (Promega) according to manufacturer's instructions. The spd3PEc_96 primer (Table 1) was end-labeled with [c 32 P] ATP using polynucleotide kinase and the extension products were separated with a 6% polyacrylamideurea sequencing gel. The 59 end was mapped by comparison to a sequencing reaction generated with a SequiTherm EXCEL TM II DNA Sequencing Kit (Epicentre Biotechnologes, Madison, Wisconsin) with the end-labeled spd3PEc_96 and a DNA template that contained the entire non-coding DNA upstream of spd-3.

Rapid Amplification of c-DNA Ends (59-RACE)
The 59 end of the larger spd-3 transcript was determined using the 59 RACE System for Rapid Amplification of cDNA Ends, Version 2.0 (Invitrogen, Carlsbad, California) according to manufacturer's instructions. First strand cDNA was synthesized using 15 mg of total RNA, the gene specific primer spd3prx36, and Superscript III Reverse Transcriptase (Invitrogen). After the first strand cDNA synthesis, the mRNA template was removed with the RNase cocktail (Ambion, Austin, Texas). Single stranded cDNA was purified using a DNA Clean and Concentrator-5 kit (Zymo Research Irvine, California). A homopolymeric tail was added to the 39-end of the cDNA using recombinant TdT (Invitrogen) and dCTP. Five mL of dC-tailed single stranded c-DNA was amplified using abridged anchor primer (AAP-G) and nested prx primers ( Table 1). The amplified product was diluted 100-fold and 5 mL was used in a second-round of PCR amplification with abridged universal amplification primer (AUAP) and nested prx primers. The PCR products were separated by electrophoresis on a 1% agarose gel, purified, and sequenced at the Iowa State Sequencing facility.

ChIP-qPCR
A ChIP assay was performed using the experimental protocol previously described [24]. Briefly, cultures of S. pyogenes strains rgg - [32] and SA5 (encoding an Rgg-myc fusion protein [24]) were grown to an A 600 of approximately 0.6, which corresponds to the post-exponential phase of growth. The cultures were treated with 1% formaldehyde (w/v). DNA bound to Rgg-Myc was immunoprecipitated with a monoclonal antibody to Myc (Invitrogen, Carlsbad, CA). Control samples were similarly prepared from an NZ131 rgg mutant strain. Immunoprecipitated DNA was purified and specific regions quantitated with PCR using ABsolute SYBR Green ROX Mix (ABgene House, Surrey, United Kingdom). Primer pairs pspd3_IRfwd, spd3_378_XhoIrev and Spd3_237fwd, pspd3_IRrev were used to quantitate DNA containing P1 and P2, respectively ( Table 2). Primers pspd3_IRfwd and LucRev1 were used to measure DNA cloned upstream of the luc gene (Table 2). For the control region, primers groEL2fwd and groEL2rev were used (Table 2). Enrichment was normalized to the amount of non-specific groEL DNA in precipitated samples, as previously described [24].