Activator Role of the Pneumococcal Mga-Like Virulence Transcriptional Regulator

Abstract

Global transcriptional regulators that respond to specific environmental signals are crucial in bacterial pathogenesis. In the case of the Gram-positive pathogen Streptococcus pneumoniae (the pneumococcus), the sp1800 gene of the clinical isolate TIGR4 encodes a protein that exhibits homology to the Mga “stand-alone” response regulator of the group A Streptococcus. Such a pneumococcal protein was shown to play a significant role in both nasopharyngeal colonization and development of pneumonia in murine infection models. Moreover, it was shown to repress the expression of several genes located within the rlrA pathogenicity islet. The pneumococcal R6 strain, which derives from the D39 clinical isolate, lacks the rlrA islet but has a gene (here named mga_Spn) equivalent to the sp1800 gene. In this work, and using in vivo approaches, we have identified the promoter of the mga_Spn gene (Pmga) and demonstrated that four neighboring open reading frames of unknown function (spr1623 to spr1626) constitute an operon. Transcription of this operon is under the control of two promoters (P1623A and P1623B) that are divergent from the Pmga promoter. Furthermore, we have shown that the Mga_Spn protein activates the P1623B promoter in vivo. This activation requires sequences located around 50 to 120 nucleotides upstream of the P1623B transcription start site. By DNase I footprinting assays, we have also demonstrated that such a region includes an Mga_Spn binding site. This is the first report on the activator role of the pneumococcal Mga-like protein.

INTRODUCTION

During infection, pathogenic bacteria must be able to survive in different niches of their hosts. This adaptation requires a coordinated regulation in the expression of many virulence and metabolic genes. Global transcriptional regulators that respond to specific environmental signals (response regulators) are key elements in such regulatory networks. One example is the Mga protein of the Gram-positive (G+) bacterium Streptococcus pyogenes (group A Streptococcus [GAS]), which causes a broad spectrum of diseases in humans (3). To date, very little is known about how Mga (multiple gene regulator of GAS) is able to sense changes in the environment. However, its role in pathogenesis has been studied in detail (12, 23). During exponential growth, Mga activates directly the transcription of several virulence genes, including its own gene. These Mga-regulated genes encode proteins that enable the bacterium to colonize specific host tissues and evade the host immune response. In addition, a transcriptome analysis revealed that Mga activates or represses, likely in an indirect way, the expression of various genes involved in the transport and utilization of sugars and other metabolites (30). Homologues of Mga have been identified in several G+ pathogens, including Streptococcus dysgalactiae (5, 38), Streptococcus pneumoniae (9), and Bacillus anthracis (35).

S. pneumoniae (the pneumococcus) remains a main cause of morbidity and mortality worldwide. It usually resides in the nasopharynx of healthy individuals. However, when the immune system weakens, S. pneumoniae can cause serious diseases, such as pneumonia, meningitis, and septicemia (15, 37). The genomic sequence of the TIGR4 strain (a serotype 4 clinical isolate) revealed that about 5% of this genome is composed of insertion sequences that may contribute to genome rearrangements through uptake of foreign DNA (34). Signature-tagged mutagenesis experiments in TIGR4 led to the identification of several genes associated with virulence (8). One of them was the sp1800 gene, which is highly conserved in the pneumococcal strains whose genomes have been totally or partially sequenced. The sp1800 gene encodes a protein named MgrA (Mga-like repressor A) due to its homology to the Mga response regulator of GAS (9). MgrA (493 amino acids) was shown to play a significant role in both nasopharyngeal colonization and development of pneumonia in murine infection models (9). Furthermore, microarray experiments showed that MgrA is able to repress the expression of several genes located within the rlrA pathogenicity islet (9). In contrast to the sp1800 gene, the rlrA islet has been found in a small number of pneumococcal strains (27), indicating that it might not be the main target of the MgrA regulator (9). This fact suggested that novel MgrA-regulated genes could be identified in work involving different pneumococcal strains and/or under different bacterial growth conditions. Indeed, there is evidence that some response regulators influence the transcriptional profile in a different manner depending on the bacterial strain and/or serotype (10, 11, 22, 30).

The genome of the pneumococcal R6 strain, which derives from the D39 clinical isolate (serotype 2), has been totally sequenced. Unlike the TIGR4 strain, R6 and D39 lack the rlrA pathogenicity islet (14, 20, 34). The spr1622 gene (here named mga_Spn) of the R6 strain encodes a protein (Mga_Spn) that differs from the MgrA regulator in two amino acid residues. In the present work, we have identified the promoter of the mga_Spn gene (Pmga). Upstream of this promoter there are four open reading frames (ORFs) of unknown function (spr1623 to spr1626) that are highly conserved in the TIGR4 strain. We have demonstrated that these ORFs are transcribed into a single polycistronic mRNA molecule from two promoters (P1623A and P1623B) that are divergent from the Pmga promoter. Moreover, unlike previous studies in TIGR4 (9), we show here that Mga_Spn activates the P1623B promoter in vivo. This activation requires sequences that are recognized by a His-tagged Mga_Spn protein in vitro. Hence, our findings show, for the first time, that the pneumococcal Mga-like regulator has a positive effect on gene expression.

MATERIALS AND METHODS

Bacterial strains, oligonucleotides, and plasmids.

The S. pneumoniae R6 strain was used (14). To construct the R6Δmga mutant strain, gene replacement by homologous recombination was carried out. A 1,165-bp DNA fragment that contained the pC194 cat gene (chloramphenicol resistance) (13) was flanked by R6 DNA sequences (543 bp and 605 bp). In the R6 genome, such DNA sequences flank the mga_Spn gene (promoter plus coding sequence). The cat cassette generated in vitro was used to transform competent R6 cells. Selection of transformants resistant to chloramphenicol (1.5 μg/ml) led to the isolation of the R6Δmga strain. Dye terminator sequencing at Secugen (CIB, Madrid, Spain) confirmed that R6Δmga lacks the chromosomal region that spans coordinates 1596826 to 1598431.

The oligonucleotides used are listed in Table 1. Plasmids pAS, pAST, and pAS-T2T1rrnB (here named pAST2), which are based on the pMV158 replicon, were used (32). They carry the tetL gene (tetracycline resistance). Plasmid pDL287, a pVA380-1 derivative that carries a kanamycin resistance gene, was also used (21). To construct pAS-Pmga, a 170-bp region of the R6 genome (promoter Pmga) was amplified by PCR using the PrSp1 and PrSp2 primers. The amplified DNA was digested with BamHI, and the 142-bp digestion product was inserted into the BamHI site of pAS. In pAS-Pmga, gfp expression is under the control of the Pmga promoter. To construct pAST-PAB and pAST2-Pmga, a 333-bp region of the R6 DNA was amplified with the PmgaSac and PABSac primers. After SacI digestion, the 301-bp restriction fragment (promoters P1623A, P1623B, and Pmga) was cloned into the SacI site of pAST (pAST-PAB; gfp expression under the control of the P1623A and P1623B promoters) and pAST2 (pAST2-Pmga; gfp expression under the control of the Pmga promoter). To construct pAST-PABΔ84, a 246-bp region of the R6 DNA (promoters P1623A and P1623B) was amplified with the PABSac and PABΔ84Sac primers. The PCR product was digested with SacI, and the 216-bp restriction fragment was inserted into the SacI site of pAST. To construct pAST-PABΔ153, a 177-bp region of the R6 DNA (promoters P1623A and P1623B) was amplified with the PABSac and PABΔ153Sac primers. After SacI digestion, the 146-bp restriction fragment was cloned into the SacI site of pAST. In pAST-PABΔ84 and pAST-PABΔ153, gfp expression is under the control of the P1623A and P1623B promoters. Construction of the pDLPsulA::mga plasmid was as follows. (i) Amplification of a 189-bp region of the R6 DNA (promoter PsulA) (32) was done using the PsulNde and PsulCla primers. The PCR-synthesized DNA was digested with NdeI, generating the 172-bp PsulA fragment. (ii) Amplification of a 1,650-bp region of the R6 DNA (promoterless mga_Spn gene) was done using the mgaNde and mgaCla primers. After digestion with NdeI, the 1,636-bp restriction fragment was ligated to the 172-bp PsulA fragment (PsulA::mga fusion gene). (iii) Amplification of the PsulA::mga fusion gene was performed with the PsulCla and mgaCla primers. After digestion with ClaI, the 1,777-bp restriction fragment was cloned into the ClaI site of plasmid pDL287 (21).

Table 1.

Oligonucleotides used in this work

Name	Sequencea (5′ to 3′)
1622A	AGTTCCTGATTGTATTCCCT
1622C	GATTCTGTATTCACGCCCTC
1622D	TTCTAATTGCCTATGACTTTTTTTAG
C1622D	CTAAAAAAAGTCATAGGCAATTAGA
INTgfp	CATCACCATCTAATTCAACAAG
PErpoE	GCCCAGCAAATACTTCTAATTCC
ASTtetL	GAGGGCAGACGTAGTTTATAGGG
1623A	GAGGGCGTGAATACAGAATC
1623B	CGTAAATTTACATGAACAGTTGGG
1623C	GGAGGGTAGGCAGTGTTGTGATC
1626A	GCACCTTCTACAGCGTCTTTAGCG
PDA	GTGATTTTACCTGCCAAGAGACC
PDB	GAAAAGTCAATTATTTCGATTG
PrSp1	ATAAATTATCGGATCCAACCTCTTGC
PrSp2	GAATTTGATTCTGGATCCACGCCCTC
PmgaSac	CTTTATAAATTATGAGCTCAAACCTCTTGC
PABSac	ATATCAAAAAATCGAGCTCTTTGATTATTAC
PABΔ84Sac	ATTTCGTATAAGAGCTCTACGGAGACAATATA
PABΔ153Sac	GAATACAGAATCGAGCTCAAGTCTAAAG
PsulNde	CAAGGATTTTCATCATATGATTTTTC
PsulCla	ACTGATTGTTAATCGATTTGCTTTCTGT
mgaNde	TGCAAGAGGTTTCATATGATAATTTATAAAG
mgaCla	GTACATTTTTCTTAATCGATTGAAGGTCTTTTC
1622Nde	GAGAGAAAGATACATATGAGAGATTTA
1622Xho-His	TTTTGTTATTTTCTCGAGCTCATCTAATCG
1622H	CGGATTAAACCTCTTGCAATTATACC
1622I	CAAATTCTTTAATTGTTGCTATTA

^aRestriction sites are in bold, and the base changes that generate restriction sites are underlined.

Growth and transformation of bacteria.

S. pneumoniae was grown in AGCH medium (17, 32) supplemented with 0.2% yeast extract and 0.3% sucrose. For plasmid-harboring cells, the medium was supplemented with tetracycline (1 μg/ml) and/or kanamycin (50 μg/ml). Experiments were performed at 37°C. Procedures for competence development and transformation of S. pneumoniae were reported previously (19).

Isolation of DNA and RNA.

Genomic DNA from S. pneumoniae was prepared as described previously (17). For small-scale preparations of plasmid DNA, the High Pure plasmid isolation kit (Roche Applied Science) was used (32). The Aurum Total RNA minikit (Bio-Rad) was used to isolate total RNA from S. pneumoniae. Cells were grown to an optical density at 650 nm (OD₆₅₀) of 0.3. Then, cultures were processed as specified by the supplier, except that cells were resuspended in buffer L (50 mM Tris-HCl, pH 7.6, 1 mM EDTA, 50 mM NaCl, 0.1% deoxycholate) and incubated at 30°C for 10 min. DNA and RNA concentrations were determined using a NanoDrop ND-1000 spectrophotometer (Bio-Rad).

PCR conditions.

The Phusion High-Fidelity DNA polymerase (Finnzymes) and the Phusion HF buffer were used. Reaction mixtures (50 μl) contained 5 to 30 ng of template DNA, 20 pmol of each primer, 200 μM each deoxynucleoside triphosphate (dNTP), and 1 unit of DNA polymerase. PCR conditions were reported previously (32). PCR products were purified with the QIAquick PCR purification kit (Qiagen).

Primer extension of total RNA.

The ThermoScript reverse transcriptase enzyme (Invitrogen) was used. Primers were ³²P labeled at the 5′ end with [γ-³²P]ATP (3,000 Ci/mmol; Perkin Elmer) and T4 polynucleotide kinase (New England BioLabs). Nonincorporated nucleotide was removed using MicroSpin G-25 columns (GE Healthcare). In assays with nonradiolabeled primers, [α-³²P]dATP (3,000 Ci/mmol; Hartmann) was used in the extension reactions. Reaction mixtures (20 μl) contained ∼2 μg of total RNA and 1 to 2 pmol of primer. To anneal the primer with the transcript, samples were incubated at 65°C for 5 min. Extension reactions were carried out at 58°C for 60 min. After heating at 85°C for 5 min, samples were ethanol precipitated and dissolved in loading buffer (80% formamide, 1 mM EDTA, 10 mM NaOH, 0.1% bromophenol blue, 0.1% xylene cyanol). cDNA products were analyzed by sequencing gel (8 M urea–6% polyacrylamide) electrophoresis. Dideoxy-mediated chain termination sequencing reactions were run in the same gel. Labeled products were visualized using Fujifilm Image Analyzer FLA-3000. The intensity of the bands was quantified using the QuantityOne software (Bio-Rad).

RT-PCR assays.

For reverse transcription-PCR (RT-PCR) assays, for first-strand cDNA synthesis, 20 pmol of primer was annealed to ∼1.5 μg of total RNA. The mixture was incubated with 15 units of ThermoScript reverse transcriptase (Invitrogen) at 55°C for 45 min. PCRs were then carried out using cDNA as the template (10% of the first-strand reaction), 20 pmol of each primer, and the Phusion High-Fidelity DNA polymerase (see “PCR conditions” above). To rule out the presence of genomic DNA in the RNA preparation, the same reactions were performed in the absence of the reverse transcriptase. As a positive control, PCRs were performed with genomic DNA. PCR products were analyzed by agarose (0.8%) gel electrophoresis. Gels were stained with ethidium bromide, and DNA was visualized using a Gel-Doc system (Bio-Rad).

Fluorescence assays.

Plasmid-carrying cells were grown to an OD₆₅₀ of 0.3. Fluorescence intensity was measured as reported earlier (32) using a Thermo Scientific Varioskan Flash instrument (excitation at 488 nm and emission at 515 nm). In each case, three independent cultures were analyzed. The fluorescence corresponding to 200 μl of phosphate-buffered saline (PBS) buffer without cells was around 0.03 arbitrary units.

Western blots.

Cells were grown to an OD₆₅₀ of 0.3. The protocol used to prepare whole-cell extracts was described previously (32). Total proteins were separated by SDS-polyacrylamide (10%) gel electrophoresis. Proteins were transferred electrophoretically to Immun-blot polyvinylidene difluoride (PVDF) membranes (Bio-Rad) using a Mini Trans-Blot (Bio-Rad) as reported previously (32). Membranes were probed with polyclonal antibodies against His-tagged Mga_Spn. Antigen-antibody complexes were detected using antirabbit horseradish peroxidase (HRP)-conjugated antibodies, the Immun-StarTM HRP substrate kit (Bio-Rad), and the luminescent image analyzer LAS-3000 (Fujifilm Life Science). The intensity of the bands was quantified using the QuantityOne software (Bio-Rad).

Overproduction and purification of Mga_Spn-His.

Gene mga_Spn was engineered to encode a His-tagged Mga_Spn protein (Mga_Spn-His). A 1,512-bp region of the R6 genome was amplified by PCR using the 1622Nde and 1622Xho-His oligonucleotides, which include single restriction sites for NdeI and XhoI, respectively (Table 1). The amplified DNA was digested with both enzymes, and the 1,481-bp digestion product was cloned into the pET24b vector (Novagen), which enables a C-terminal His₆ tag fusion. Escherichia coli BL21(DE3) cells harboring plasmid pET24b-mga_Spn-His were grown at 37°C in tryptone-yeast extract medium containing kanamycin (30 μg/ml). When the culture reached an OD₆₀₀ of 0.45, isopropyl-β-d-thiogalactopyranoside (IPTG) was added (1 mM). After 25 min, cells were incubated with rifampin (200 μg/ml) for 60 min. Cells were then sedimented, washed twice with buffer V-His (10 mM Tris-HCl, pH 7.6, 5% glycerol, 300 mM NaCl, 5 mM β-mercaptoethanol), and stored at −80°C. The cell pellet was concentrated (40×) in buffer V-His containing an EDTA-free protease inhibitor cocktail (Roche). Cells were disrupted by passage through a French pressure cell, and the whole-cell extract was centrifuged to remove cell debris. Imidazole (10 mM) was added to the clarified extract, which was loaded onto a HisTrap HP column (GE Healthcare) preequilibrated with buffer V-His containing 10 mM imidazole. After washing with the same buffer, Mga_Spn-His was eluted with buffer V-His containing 250 mM imidazole. Fractions containing Mga_Spn-His were identified by Coomassie-stained SDS-polyacrylamide (10%) gels, pooled, and dialyzed against buffer P (20 mM Tris-HCl, pH 7.6, 5% glycerol, 250 mM NaCl, 1 mM EDTA, 1 mM dithiothreitol [DTT]). The protein preparation was concentrated by filtering through a 10-kDa-cutoff membrane (Macrosep; Pall), loaded onto a HiLoad Superdex 200 gel filtration column (Amersham), and subjected to fast-pressure liquid chromatography (FPLC) (Biologic DuoFlow; Bio-Rad). Fractions containing Mga_Spn-His were pooled, concentrated, and stored at −80°C. Protein concentration was determined using a NanoDrop ND-1000 spectrophotometer (Bio-Rad).

DNase I footprinting assays.

A 222-bp region of the R6 genome (coordinates 1598298 to 1598519) was amplified by PCR using the 1622I and 1622H primers. One of the primers was previously ³²P labeled at the 5′ end using [γ-³²P]ATP (3,000 Ci/mmol; PerkinElmer) and T4 polynucleotide kinase. Reaction mixtures (10 μl) contained 30 mM Tris-HCl, pH 7.6, 1.2 mM DTT, 0.2 mM EDTA, 1 mM CaCl₂, 10 mM MgCl₂, 50 mM NaCl, 1% glycerol, 4 nM ³²P-labeled 222-bp DNA, and different concentrations of Mga_Spn-His. After 20 min at room temperature, 0.04 units of DNase I (Roche Applied Science) was added for 5 min at the same temperature. Reactions were stopped with 1 μl of 250 mM EDTA. Then, 4 μl of loading buffer (80% formamide, 1 mM EDTA, 10 mM NaOH, 0.1% bromophenol blue, and 0.1% xylene cyanol) was added. Samples were heated at 95°C for 5 min and loaded onto 8 M urea–6% polyacrylamide gels. Dideoxy-mediated chain termination sequencing reactions using the 222-bp fragment and either the 1622I or the 1622H oligonucleotide were run in the same gel. Labeled products were visualized using a Fujifilm Image Analyzer FLA-3000 or by autoradiography. The intensity of the bands was quantified using the QuantityOne software (Bio-Rad).

RESULTS

Transcription of the mga_Spn gene in pneumococcal R6 cells.

The complete genome sequence of S. pneumoniae R6 has been published (14) (GenBank accession number AE007317.1). The ATG codon at coordinate 1598270 is likely the translation start site of the spr1622 gene (here named mga_Spn), since it is preceded by a putative Shine-Dalgarno sequence (5′-AAAGAGAGAAAG-3′) (Fig. 1) that complies with the reported consensus sequence for pneumococcus (Chang Bioscience, San Francisco, CA). Translation from this ATG codon would produce a protein of 493 residues (Mga_Spn). EMBOSS needle global sequence alignment (31) of Mga_Spn and the Mga regulator (530 residues) encoded by the M6_Spy1720 gene of S. pyogenes MGAS10394 revealed 42.6% similarity and 21.4% identity.

Fig 1.

Genetic map of the region spanning coordinates 1596789 to 1600589 of the R6 genome (14). The gene spr1622 has been named mga_Spn in this work. For each ORF, the coordinates of the predicted start and stop codons are indicated. The nucleotide sequence of the region spanning the start codon of the mga_Spn gene (coordinate 1598270) and the start codon of the spr1623 ORF (coordinate 1598960) is shown. The putative Shine-Dalgarno sequence (SD) of mga_Spn is indicated. The main sequence elements (−35 box, −10 box, and extended [ext] −10 box) of the promoters identified in this work (Pmga, P1623B, and P1623A), as well as the transcription start sites (+1 position), are indicated.

To examine whether the mga_Spn gene was transcribed, RT-PCR experiments were carried out (Fig. 2). The 1622A oligonucleotide was used as a primer for extension on total RNA isolated from R6 cells. The cDNA products were further amplified by PCR using either the 1622A and C1622D or the 1622A and 1622C primers. As controls, PCRs were performed using total RNA (negative control) or genomic DNA (positive control) as the template. With the 1622A and C1622D primers, a PCR product that migrated at the position expected for a 1,023-bp DNA was synthesized. Such a product was not visualized in the negative control. With the 1622A and 1622C primers, no PCR products were detected. However, a product with the mobility expected for a 1,221-bp fragment was synthesized in the positive control. Therefore, transcription of the mga_Spn gene was initiated at a site located downstream of coordinate 1598452. Sequence analysis of the region spanning this coordinate and the translation start codon of the mga_Spn gene revealed the existence of a putative promoter (here named Pmga) (Fig. 1). It has a consensus −10 hexamer (5′-TATAAT-3′) and a consensus −10 extension (5′-TGTG-3′) and shows a 3/6 match at the −35 hexamer (5′-ATGCTA-3′) (consensus residues are shown in bold). The −35 and −10 elements are separated by 16 nucleotides (nt). The features of the Pmga promoter indicate that it would be likely recognized by a housekeeping σ factor similar to Escherichia coli σ⁷⁰ (7).

Fig 2.

Transcription of mga_Spn in vivo. RT-PCR assays were performed using RNA from R6 cells. (Left) The positions of the oligonucleotides used (1622A, 1622C, and C1622D) are shown. (Right) RT-PCRs (lanes R) were subjected to agarose (0.8%) gel electrophoresis. RT-PCRs without adding the reverse transcriptase were performed as negative controls (lanes N). The sizes of PCR-amplified DNA fragments (1 and 2) using genomic DNA as the template (lanes P, positive control) are indicated. The sizes (in bp) of DNA fragments (lane M) used as molecular weight markers (HyperLadder I, Bioline) are indicated on the right of the gel.

We next performed primer extension assays using RNA from R6 cells and the 1622D primer, which is complementary to the C1622D oligonucleotide (Fig. 2). No cDNA products were detected (not shown), which indicated that the amount of mga_Spn transcripts in the RNA preparation was small. To amplify the signal, a 136-bp chromosomal region (coordinates 1598440 to 1598305), which contained the putative Pmga promoter, was inserted into the BamHI site of the pAS vector (plasmid pAS-Pmga; Fig. 3). This site is located upstream of a promoterless gfp allele that encodes a variant of the green fluorescent protein (GFP) (32). The intensity of fluorescence in cultures of cells carrying the pAS-Pmga plasmid was slightly higher (1.5-fold) than that in cultures of cells harboring the pAS vector, indicating that the 136-bp region contained a promoter signal. Also, primer extension assays were performed using RNA from cells carrying plasmid pAS-Pmga. As primer, the INTgfp oligonucleotide, which anneals to gfp transcripts, was used (Fig. 3). Two cDNA products of 120 and 121 nucleotides were detected, indicating that transcription of the gfp gene started at a site located 7 to 8 nucleotides downstream of the −10 element of the Pmga promoter. Thus, the Pmga promoter was functional under our bacterial growth conditions.

Fig 3.

Promoter Pmga is functional in vivo. Primer extension reactions were carried out on total RNA isolated from R6 cells harboring plasmid pAS-Pmga. The gfp gene carries translation initiation signals optimized for prokaryotes (SD) (25). The tetL gene confers resistance to tetracycline. The main sequence elements of the Pmga promoter (gray boxes) and the ATG initiation codon of the gfp gene (black box) are indicated. BamHI sites are underlined. The asterisks indicate the 3′ ends of the cDNA products synthesized using the INTgfp primer. The sizes of the cDNA products (lane R) are indicated in nucleotides on the right of the gel. Dideoxy-mediated chain termination sequencing reactions using pLS1 DNA (19) and the F-pLS1 primer (5′-TGCTGGCAGGCACTGGC-3′; coordinates 802 to 818 of pLS1) were used as DNA size markers (lanes A, C, G, and T).

Transcription of the spr1623-spr1626 operon in pneumococcal R6 cells.

Upstream of the mga_Spn gene there are four ORFs (spr1623 to spr1626) that appeared to be organized into an operon (Fig. 1). The mga_Spn gene and the putative operon would be divergently transcribed. The ATG initiation codon of the mga_Spn gene (coordinate 1598270) and the ATG initiation codon of the spr1623 ORF (coordinate 1598960) are separated by 689 bp. To investigate whether the putative operon was transcribed, RT-PCR assays were performed (Fig. 4). First, the 1623B oligonucleotide was used as a primer for extension on total RNA isolated from R6 cells. The resulting cDNA was amplified by PCR using either the 1623B and 1623C or the 1623B and 1623A primers. With the 1623B and 1623C primers, a PCR product that migrated at the position expected for a 695-bp DNA fragment was synthesized (Fig. 4). With the 1623B and 1623A primers, no PCR products were visualized. Nevertheless, such primers amplified an 892-bp region using genomic DNA as the template. Next, we performed RT-PCR assays using the 1626A primer for cDNA synthesis (Fig. 4). Amplification of the cDNA with the 1626A and 1623C primers generated a product that moved at the position expected for a 1,917-bp fragment. Collectively, these results indicated that the four ORFs (spr1623 to spr1626) were transcribed into a polycistronic mRNA molecule from a site(s) located between coordinates 1598433 and 1598630. Sequence analysis of this region predicted a promoter sequence (here named P1623A; Fig. 1) that has a canonical −10 hexamer (5′-TATAAT-3′) and a near-consensus −35 hexamer (5′-TTGACT-3′). The spacing between the two sequence elements is 17 nucleotides.

Fig 4.

The spr1623 to spr1626 ORFs constitute an operon. RT-PCR assays were performed using RNA from R6 cells. The positions of the oligonucleotides used (1623A, 1623B, 1623C, and 1626A) are shown. RT-PCRs (lanes R) were analyzed by agarose (0.8%) gel electrophoresis. As negative controls (lane N), RT-PCRs were carried out without adding the reverse transcriptase. The sizes of PCR-amplified DNA fragments (1, 2, and 3) using genomic DNA as the template (lanes P, positive control) are indicated. Lanes M, DNA fragments used as molecular weight markers (in bp) (HyperLadder I, Bioline).

To analyze whether the P1623A promoter was functional in vivo, we performed primer extension assays using the PDA oligonucleotide (Fig. 5). Two cDNA products, of 106 and 191 nucleotides, were detected, which would correspond to transcription initiation events at coordinates 1598592 (P1623A promoter) and 1598507, respectively. This result indicated that the pneumococcal RNA polymerase recognized not only the P1623A promoter but also a promoter sequence (named P1623B) that has a consensus −10 hexamer (5′-TATAAT-3′) but lacks a −35 element (Fig. 1). The functionality of the P1623B promoter was confirmed further by primer extension using the PDB oligonucleotide (cDNA product of 60 nucleotides) (Fig. 5).

Fig 5.

The spr1623-spr1626 operon is transcribed from promoters P1623A and P1623B. Primer extension assays were performed using RNA from R6 cells and the PDA (left gel) or PDB (right gel) primer. The sizes of the cDNA products (lanes R) are indicated in nucleotides on the right of the gels. A, C, G, and T sequence ladders were used as DNA size markers. They were prepared using M13mp18 DNA (39) and the −40 M13 primer (5′-GTTTTCCCAGTCACGAC-3′) (left gel) or a PCR-amplified DNA fragment from the E. faecalis V583 genome and the Rev primer (5′-GATTTCTTCAATTTGTTCCATC-3′) (right gel). The asterisks in the scheme below the gels indicate the transcription start sites identified in this study.

Mga_Spn activates the P1623B promoter in vivo.

To investigate whether Mga_Spn influenced the activity of a particular promoter in vivo, we constructed a pneumococcal strain designed to overproduce Mga_Spn. First, we constructed the PsulA::mga fusion gene, in which the Pmga promoter of the mga_Spn gene was replaced with the promoter of the pneumococcal sulA gene (PsulA) (18, 32). The fusion gene was then inserted into pDL287 (21), generating the pDLPsulA::mga recombinant. Compared to R6 plasmid-free cells (Fig. 6, lane 1) or R6 cells carrying pDL287 (not shown), the amount of Mga_Spn increased ∼8-fold in cells harboring pDLPsulA::mga (Fig. 6, lane 3). By primer extension, we analyzed the effect of the Mga_Spn overproduction on the activity of the P1623A and P1623B promoters located on the bacterial chromosome. We used a mix of oligonucleotides radioactively labeled at the 5′ end: PDA (Fig. 5) and PErpoE, which anneal to spr1623 and rpoE transcripts, respectively. The rpoE gene (spr0437 in the R6 genome) encodes the delta subunit of the RNA polymerase and was used as an internal control. As shown in Fig. 7, using RNA from R6/pDL287 cells (lane 1; low levels of Mga_Spn), three cDNA products of 106 nt (P1623A promoter), 191 nt (P1623B promoter), and 231 nt (PrpoE promoter) were synthesized. Unlike the 231-nt product, the amounts of the 106-nt and 191-nt cDNAs increased 2.6-fold and 4.5-fold, respectively, when RNA from R6/pDLPsulA::mga cells was used (lane 2; overproduction of Mga_Spn). Therefore, overproduction of Mga_Spn led to activation of promoters P1623A and P1623B, although the effect appeared to be greater on the activity of promoter P1623B.

Fig 6.

Detection of Mga_Spn in pneumococcal cell extracts by Western blotting using polyclonal antibodies against His-tagged Mga_Spn. Total proteins from R6 cells (lane 1), R6Δmga cells (lane 2), and pDLPsulA::mga-carrying cells (lane 3) were separated by SDS-PAGE. His-tagged Mga_Spn protein (6 ng) (lane 4) and prestained proteins (Invitrogen) (not shown) were run in the same gel.

Fig 7.

Mga_Spn mediates activation of the P1623B promoter. Primer extension reactions were carried out using total RNA from R6/pDL287 (lanes 1, 5, and 6), R6/pDLPsulA::mga (lane 2), R6Δmga/pDL287 (lane 3), or R6Δmga/pDLPsulA::mga (lane 4) cells. Used as primers were 5′-labeled oligonucleotides: a mix of the PDA and PErpoE primers (lanes 1, 2, 3, and 4), primer PDA (lane 5), or primer PErpoE (lane 6). The sizes (in nucleotides) of the cDNA products are indicated on the left of the gel: 106 nt for the P1623A promoter, 191 nt for the P1623B promoter, and 231 nt for the PrpoE promoter. Sequence ladders were used as DNA size markers (lanes A, C, G, and T). They were prepared using a PCR-amplified DNA fragment from the E. faecalis V583 genome and the Fw primer (5′-CGTTTGAGCAATATAATCGTTTG-3′).

We next constructed an R6 derivative, named R6Δmga, in which the chromosomal region spanning the coordinates 1596826 and 1598431 was replaced with the cat gene (chloramphenicol resistance) of plasmid pC194 (13). This mutant strain lacks the mga_Spn gene (including the Pmga promoter) but conserves the P1623A and P1623B promoter sequences (Fig. 1). As expected, R6Δmga cells did not synthesize Mga_Spn (Fig. 6, lane 2). By primer extension, we examined the activity of the chromosomal P1623A and P1623B promoters in R6Δmga cells carrying either pDL287 (absence of Mga_Spn) (Fig. 7, lane 3) or pDLPsulA::mga (overproduction of Mga_Spn) (lane 4). Again, a mix of the 5′-labeled PDA and PErpoE primers was used. In contrast to what was observed for R6/pDL287 cells (lane 1; low levels of Mga_Spn), the 191-nt product (P1623B) was not detected in R6Δmga/pDL287 cells (lane 3; absence of Mga_Spn), although no changes were found in the amounts of the 106-nt (P1623A) and 231-nt (PrpoE) cDNAs. Thus, in the absence of Mga_Spn the activity of the P1623B promoter decreased without affecting the activity of the P1623A promoter. However, unexpectedly, the activity of the P1623B promoter on the R6Δmga genome did not change in the presence of pDLPsulA::mga (lane 4; overproduction of Mga_Spn). These results suggested that the genome of the R6Δmga strain lacked not only the mga_Spn gene (including the Pmga promoter) but also a site required for Mga_Spn-mediated activation of promoter P1623B (see below).

Mapping the site required for Mga_Spn-mediated activation of the P1623B promoter.

The promoter-probe vector pAST (32) (Fig. 8A) carries a multiple cloning site between the T1-T2 tandem terminators of the E. coli rrnB rRNA operon and a promoterless gfp allele. Moreover, the T1-T2 terminators (T1T2rrnB region) are located downstream of the tetL gene, which confers resistance to tetracycline (19). Transcription of the tetL gene terminates efficiently at the T1T2rrnB region (32). To delimit the site required for Mga_Spn-mediated activation of the P1623B promoter, a deletion analysis was carried out. Three chromosomal regions were inserted independently into the SacI site of pAST (Fig. 8A): (i) the PAB region (coordinates 1598304 to 1598600; pAST-PAB); (ii) the PABΔ84 region (coordinates 1598388 to 1598600; pAST-PABΔ84); and (iii) the PABΔ153 region (coordinates 1598457 to 1598600; pAST-PABΔ153). In these constructions, gfp expression was under the control of both promoters, P1623A and P1623B. Thus, the promoter activity of each chromosomal region was evaluated by fluorescence assays (Fig. 8A). The promoter activity of the PAB and PABΔ84 regions was 2-fold higher in R6 cells (low levels of Mga_Spn) than in R6Δmga cells (absence of Mga_Spn). However, the promoter activities of the PABΔ153 region were similar in the two genetic backgrounds. These results indicated that the region spanning the coordinates 1598388 and 1598457 contained sequences that were required for Mga_Spn-mediated activation of promoters P1623A and/or P1623B.

Fig 8.

Fluorescence assays. (A) Activity of the P1623A (PA) and P1623B (PB) promoters. The promoter-probe vector pAST was described previously (32). The positions of the tetL (tetracycline resistance) and gfp (green fluorescence protein) genes are indicated. The T1T2 box represents the tandem terminators T1 and T2 of the E. coli rrnB rRNA operon. Gray boxes represent DNA fragments from the R6 genome. (B) Activity of the Pmga promoter. Plasmid pAST2 was described (named pAS-T2T1rrnB) by Ruiz-Cruz et al. (32). Compared to pAST, it carries the T1T2rrnB region inserted in the opposite orientation (box T2T1). The position of promoter Pmga is shown. The intensity of fluorescence (arbitrary units) corresponds to 0.8 ml of culture (OD₆₅₀ = 0.3). In each case, three independent cultures were analyzed.

The promoter activity of the PABΔ84 and PABΔ153 regions was further examined by primer extension (Fig. 9). A mix of the 5′-labeled INTgfp and ASTtetL primers was used. They anneal to gfp and tetL transcripts, respectively. The tetL gene of pAST was used as an internal control. Using RNA from R6 cells (low levels of Mga_Spn) harboring pAST-PABΔ84 (Fig. 9, lane 2), three cDNA products of 102 nt (PtetL promoter), 111 nt (P1623A promoter), and 196 nt (P1623B promoter) were synthesized. Unlike the 102-nt and 111-nt cDNAs, the amount of the 196-nt cDNA decreased 5-fold when RNA from R6 cells harboring pAST-PABΔ153 was used (lane 3). Thus, in R6 cells (low levels of Mga_Spn), deletion of the region that spans the 1598388 and 1598457 coordinates (Fig. 8A) reduced the activity of the P1623B promoter but not the activity of the P1623A promoter. The specific decrease in the activity of promoter P1623B was also observed in R6Δmga cells (absence of Mga_Spn) carrying either PABΔ84 (Fig. 9, lane 4) or PABΔ153 (lane 5). These results demonstrated that Mga_Spn was able to activate, directly or indirectly, the P1623B promoter in vivo. This activation required sequences located within the region spanning coordinates 1598388 to 1598457 (Fig. 1 and 8A). Such a 70-bp region maps between the Pmga and P1623B divergent promoters, just 50 bp upstream of the P1623B transcription start site (coordinate 1598507).

Fig 9.

Genomic region needed for Mga_Spn-mediated activation of the P1623B promoter. Primer extension reactions were carried out using total RNA from R6/pAST-PABΔ84 (lanes 1 and 2), R6/pAST-PABΔ153 (lane 3), R6Δmga/pAST-PABΔ84 (lane 4), or R6Δmga/pAST-PABΔ153 (lane 5) cells. Used as primers were 5′-labeled oligonucleotides: a mix of the INTgfp and ASTtetL primers (lanes 2, 3, 4, and 5) or the ASTtetL primer (lane 1). The sizes (in nucleotides) of the cDNA products are indicated on the left of the gel: 102 nt for the PtetL promoter, 111 nt for the P1623A promoter, and 196 nt for the P1623B promoter. Dideoxy-mediated chain termination sequencing reactions using pAST DNA and the INTgfp primer were run in the same gel (lanes A, C, G, and T).

Mga_Spn does not influence the activity of promoter Pmga in vivo.

Plasmid pAST2 (named pAS-T2T1rrnB in reference 32) carries a multiple cloning site upstream of the promoterless gfp gene (Fig. 8B). Compared to the promoter-probe vector pAST, pAST2 carries the T1-T2 terminators (T1T2rrnB region) inserted in the opposite orientation (T2T1rrnB region). The T2T1rrnB region functions as a transcriptional terminator signal, although it is not as efficient as the T1T2rrnB region (32). The site required for Mga_Spn-mediated activation of the P1623B promoter is located around 80 to 150 nucleotides upstream of the Pmga transcription start site (coordinate 1598308). This fact suggested that Mga_Spn might also influence the activity of the Pmga promoter in vivo. To test this hypothesis, the PAB chromosomal region (coordinates 1598304 to 1598600), which carries the P1623A, P1623B, and Pmga promoters, was inserted into the SacI site of pAST2, generating the pAST2-Pmga recombinant (Fig. 8B). In this construction, gfp expression is under the control of the Pmga promoter. Fluorescence assays showed that the activities of the Pmga promoter located on pAST2-Pmga were similar in R6 and R6Δmga cells. Therefore, Mga_Spn did not influence the activity of the Pmga promoter under our bacterial growth conditions.

Mga_Spn-His binds to a site located upstream of promoter P1623B.

To determine whether Mga_Spn was able to interact with the P1623B promoter region, we performed DNase I footprinting assays with a His-tagged Mga_Spn protein (Mga_Spn-His). This variant of Mga_Spn carries six additional His residues at the C-terminal end. We used a 222-bp DNA fragment (coordinates 1598298 to 1598519 of R6) that contained the 70-bp region (1598388 to 1598457) known to be required for Mga_Spn-mediated activation of the P1623B promoter (Fig. 10C). Such a DNA fragment was radioactively labeled either at the 5′ end of the coding strand (Fig. 10A) or at the 5′ end of the noncoding strand (Fig. 10B). On the coding strand and in the presence of Mga_Spn-His, the region spanning the −52 and −90 positions relative to the transcription start site of the P1623B promoter was protected against DNase I digestion. On the noncoding strand, changes in the DNase I sensitivity (diminished cleavages) were observed from −57 to −79 and from −83 to −102. Moreover, the −82 and −104 positions were slightly more sensitive to DNase I cleavage. We conclude that Mga_Spn interacts with sequences located between the positions −52 and −102 relative to the P1623B transcription start site (Fig. 10C). Such sequences are included within the region shown to be required for Mga_Spn-mediated activation of the P1623B promoter. Hence, the Mga_Spn regulator activates directly the P1623B promoter.

Fig 10.

DNase I footprints of Mga_Spn-His-DNA complexes. The 222-bp DNA fragment (coordinates 1598298 to 1598519) was labeled at the 5′ end of either the coding (A) or the noncoding (B) strand. The labeled DNA (4 nM) was incubated with the indicated concentrations of Mga_Spn-His. Dideoxy-mediated chain termination sequencing reactions were run in the same gel (lanes A, C, G, and T). Densitometer scans corresponding to DNA without protein (gray line) and DNA with protein (black line; 240 nM in panel A and 200 nM in panel B) are shown. The Mga_Spn-His-protected regions are indicated with brackets. Arrowheads indicate positions that are slightly more sensitive to DNase I cleavage. The indicated positions are relative to the transcription start site of the P1623B promoter. (C) Nucleotide sequence of the region that spans coordinates 1598509 to 1598380 of the R6 genome. It includes the transcription start site of the P1623B promoter (coordinate 1598507), the region required for Mga_Spn-mediated activation of the P1623B promoter (1598457 to 1598388), and the site recognized by Mga_Spn-His (brackets).

DISCUSSION

In pathogenic bacteria, global transcriptional regulators whose activity and/or intracellular concentration changes in response to external stimuli are crucial during the infection process. Some of these response regulators are associated with a membrane-bound sensor histidine kinase, the so-called two-component signal transduction systems (1, 29). Also, various “stand-alone” response regulators, whose sensory elements remain unidentified, have been implicated in the regulation of virulence gene expression (23). To this class of global regulators belongs the Mga protein of GAS (12). In the G+ bacterium S. pneumoniae, several two-component systems are known to contribute to its virulence, although to different extents depending on the strain and/or infection model used (26). Moreover, signature-tagged mutagenesis in the pneumococcal TIGR4 strain revealed that other putative transcriptional regulators might control the expression of specific virulence genes (8). It was the case of the sp1800 gene product, an Mga orthologue that was shown to act as a repressor of the rlrA pathogenicity islet (9). Here, we have performed a transcriptional analysis of the region that spans coordinates 1596789 to 1600589 of the pneumococcal R6 genome (14). Such a region contains the spr1622 gene (mga_Spn in this work), which is equivalent to the sp1800 gene of the TIGR4 strain, and four divergent ORFs (spr1623 to spr1626) that are highly conserved in TIGR4. We have identified the promoter of the mga_Spn gene (Pmga) and demonstrated that the four ORFs constitute an operon that is transcribed from two promoters (P1623A and P1632B). Furthermore, we have shown, for the first time, that the pneumococcal Mga-like protein is able to act as a transcriptional activator: (i) Mga_Spn activates the P1623B promoter in vivo and therefore the expression of the spr1623-spr1626 operon, (ii) this activation requires sequences located upstream of the P1623B promoter, and (iii) Mga_Spn interacts with such sequences in vitro. Hemsley et al. (9) reported that the sp1800 gene product of the TIGR4 strain did not affect transcription of the neighboring cluster of genes. This discrepancy with our results might be due to the use of different pneumococcal strains and/or to the use of different bacterial growth conditions.

The pneumococcal Mga_Spn regulator is homologous (42.6% similarity and 21.4% identity) to the Mga regulator of GAS. In Mga, two helix-turn-helix domains were mapped near the N terminus of the protein. Both of them are known to be required for DNA binding and transcriptional activation (24, 36). In addition, Mga is known to bind to regions located upstream of its target promoters. The position of the Mga binding site with respect to the start of transcription varies among the promoters tested (12). In the case of the Mga_Spn regulator, analysis of its amino acid sequence using the Pfam database (4) revealed that it also has two putative DNA-binding domains within the N-terminal region, the so-called HTH_Mga (residues 6 to 65) and Mga (residues 71 to 158) domains. Moreover, we have shown that Mga_Spn interacts with sequences located upstream of the P1623B promoter. Such sequences are also needed for Mga_Spn-mediated activation of the P1623B promoter. Thus, the Mga and Mga_Spn regulators might have similar DNA binding properties.

Using the BLASTN 2.2.25+ nucleotide sequence alignment program (40), we have analyzed the region that spans the P1623A and Pmga promoters of the R6 genome (Fig. 1), which includes promoter P1623B and the Mga_Spn binding site identified in this work. This analysis has revealed that such a region is identical to those in 10 pneumococcal strains whose genomes have been totally sequenced (R6, D39, TIGR4, G54, JJA, P1031, TCH8431/19A, CGSP14, Taiwan19F-14, and 70585). Furthermore, according to protein sequence database similarity searches, Mga_Spn is highly conserved among the above-described pneumococcal strains. Compared to R6 and D39, Mga_Spn has only two amino acid changes (I309M and V358I) in the TIGR4, G54, JJA, P1031, TCH8431/19A, and CGSP14 strains, three amino acid changes (C280Y, I309M, and V358I) in the Taiwan19F-14 strain, and three amino acid changes (I309M, V358I, and P450S) in the 70585 strain.

The pneumococcal R6 strain has an additional mga-like gene (spr1404). It is adjacent to the divergent spr1403 gene that encodes a collagen-like protein (PclA) (28). Both genes are absent in TIGR4 (2). Although the spr1404 gene product has homology to the Mga_Spn (59.3% similarity) and Mga (40% similarity) regulators, Paterson et al. (28) reported that single-deletion mutants lacking either spr1404 or spr1403 were not attenuated in a mouse model of invasive pneumonia. Thus, as pointed out by the authors, further work is required to elucidate whether the spr1404 gene has a significant role in pathogenesis.

The function of the spr1623-spr1626 operon is unknown. It encodes products of 188 (spr1623; hypothetical protein), 56 (spr1624; putative lipoprotein), 202 (spr1625; putative general stress protein 24), and 67 (spr1626; hypothetical protein) residues. According to the Protein Clusters database (16), the spr1624, spr1625, and spr1626 products are identical to those in the 10 pneumococcal strains mentioned above. However, unlike R6 and D39, the operon of the other strains has an additional ORF (named sp1801 in TIGR4). It is located upstream of the equivalent spr1623 ORF and would encode a product of 54 residues (hypothetical protein, putative transglycosylase associated protein). The absence of this ORF in R6 (and D39) is due to the deletion of one nucleotide between coordinates 1598751 and 1598752, which results in a truncated ORF that would encode a product of 20 amino acids.

Several observations suggest that the spr1623-spr1626 operon might play a role in virulence. First, the product (202 amino acids) of the spr1625 gene has homology (69% similarity) to the product of the Enterococcus faecalis gls24 gene (EF0080 in strain V583; 186 amino acids), which was shown to be a general stress-inducible gene involved in bile salts resistance (6). Also, it was shown to be important for virulence in a mouse peritonitis model (33). Second, Hemsley et al. (9) reported the characterization of a TIGR4 mutant strain (STM206) that carries a transposon inserted ∼300 bp upstream of the predicted translation start codon of the sp1800 gene. This mutant strain was attenuated for both nasopharyngeal carriage and lung infection in murine models. Moreover, it was much more affected in virulence than a mutant strain (AC1272) that carries a transposon inserted into the coding sequence of the sp1800 gene (mga_Spn gene in R6). Now, we have shown that the distances between the translation start codon of the mga_Spn gene (coordinate 1598270) and the transcription initiation sites of the spr1623-spr1626 operon are 323 nt (from the P1623A promoter; coordinate 1598592) and 238 nt (from the P1623B promoter; coordinate 1598507). Hence, the transposon in strain STM206 could be affecting the expression of the sp1801-sp1805 operon (spr1623-spr1626 operon in R6). If this were the case, the attenuation phenotype of the STM206 mutant strain (9) would indicate an important role of the operon in pneumococcal virulence.

In summary, this is the first report on the activator role of the pneumococcal Mga-like regulator (Mga_Spn). This regulatory protein activates the transcription of a four-gene operon from a site located upstream of the target promoter.