Bacterial strains and plasmids used in this study are listed in Table 1. S. pneumoniae D39 was grown as described previously (Kloosterman et al., 2006a; Afzal et al., 2014). For β-galactosidase assays, derivatives of S. pneumoniae D39 were grown in a chemically defined medium (CDM) (Kloosterman and Kuipers, 2011) supplemented either with 0.03 or 50 mM cysteine. CDM was prepared without cysteine. For selection on antibiotics, the medium was supplemented with the following concentrations of antibiotics: tetracycline: 2.5 μg/ml for S. pneumoniae; ampicillin: 100 μg/ml for Escherichia coli and erythromycin: 0.25 μg/ml for S. pneumoniae and 120 μg/ml for E. coli. All bacterial strains used in this study were stored in 10% (v/v) glycerol at -80°C. For PCR amplification, chromosomal DNA of S. pneumoniae D39 (Lanie et al., 2007) was used as a template. Primers used in this study are based on the sequence of the S. pneumoniae D39 genome and are listed in Table 2.

A markerless cmbR mutant (MA1000) was constructed in S. pneumoniae D39 using pORI280, as described before (Kloosterman et al., 2006a). Primer pairs cmbR-1/cmbR-2 and cmbR-3/cmbR-4 were used to generate PCR fragments of the left and right flanking regions of cmbR. The integrity of the cmbR mutant (MA1000) was further confirmed by PCR and DNA sequencing.

Chromosomal transcriptional lacZ-fusions to the spd-0150, metE, gshT, spd-0618, tcyB, metA, and yvdE promoters were constructed in our previous study (Afzal et al., 2016). These constructs were further introduced into the D39 ΔcmbR (MA1000) strain resulting in strains MA1002-06, respectively. Transcriptional lacZ-fusion to the yvdE promoter was constructed in pPP2 (Halfmann et al., 2007) with primer pairs mentioned in Table 2 resulting in pMA1001. This construct was further introduced into the D39 wild-type and the D39 ΔcmbR (MA1000) strains resulting in strains MA1001 and MA1007, respectively. The following sub-clones of Pspd-0150, PmetE, Pspd-0618, and PmetA were made in pPP2 (Halfmann et al., 2007) using the primer pairs mentioned in Table 2: Pspd-0150-M (mutation in the cmbR site), PmetE-M (mutation in the cmbR site), Pspd-0618R1-M (mutation in the cmbR site 1), Pspd-0618R2-M (mutation in the cmbR site 2), and PmetA-TER (termination of the cmbR site), resulting in plasmids pMA1002-06, respectively. These constructs were introduced into the S. pneumoniae D39 wild-type, resulting in strains MA1008-12, respectively. All plasmid constructs were checked for the presence of the insert by PCR and DNA sequencing.

β-galactosidase assays were performed as described before (Israelsen et al., 1995; Halfmann et al., 2007) using cells that were harvested in the mid-exponential growth phase, and grown in CDM supplemented either with 0.03 or 50 mM cysteine.

Microarray analysis was performed as described before (Afzal et al., 2015a; Shafeeq et al., 2015). For DNA microarray analysis of S. pneumoniae in the presence of cysteine, the transcriptomes of S. pneumoniae D39 wild-type, grown in replicates in CDM with 0.03 mM cysteine, was compared to that grown in CDM with 50 mM cysteine and harvested at respective mid-exponential growth phases. For the identification of differentially expressed genes, a Bayesian p-value of <0.001 and a fold-change cut-off >1.5 was applied. RNA isolation was performed as described before (Afzal et al., 2015a). All other procedures regarding the DNA microarray experiments and data analysis were performed as previously described (Shafeeq et al., 2011a,b; Afzal et al., 2015b). Microarray data have been submitted to GEO under the accession number GSE89458.

For quantitative RT-PCR, S. pneumoniae D39 wild-type and D39 ΔcmbR were grown in replicates in CDM supplemented with either 0.03 mM or 50 mM cysteine. RNA isolation was done as described before (Afzal et al., 2015a). First, strand cDNA synthesis was performed on RNA (Shafeeq et al., 2011b). cDNA (2 μl) was amplified in a 20 μl reaction volume that contained 3 pmol of each primer (Table 2) and the reactions were performed in three technical replicates on two biological replicates of RNA (Shafeeq et al., 2011b). The transcription level of specific genes was normalized to gyrA transcription, amplified in parallel with gyrA-F and gyrA-R primers. The results were interpreted using the comparative CT method (Schmittgen and Livak, 2008). Differences in expression of twofold or greater relative to control were considered as significant.

Cysteine is one of the most important amino acids for bacteria. It is also present in human blood plasma at concentration of 0.03 mM (Lopez, 2013). To study the impact of cysteine on the transcriptome of S. pneumoniae D39 wild-type, we performed microarray comparison of S. pneumoniae D39 grown in CDM with 0.03–50 mM cysteine. 0.03 mM concentration was chosen, as this is the concentration of cysteine in human blood plasma (Lopez, 2013). 50 mM concentration of cysteine is normally used to prepare CDM. A number of genes/gene clusters were differentially regulated under our tested conditions (Table 3). The expression of spd-0447-49 and spd-1098-99 was altered under our tested conditions. These genes belong to the glutamine regulon and their expression has been reported to be downregulated in the presence of a nitrogen source (Kloosterman et al., 2006b). The expression of important metal-related genes (prtA, psaBC and spd-1402) was downregulated under our tested conditions. These genes belong to the PsaR regulon and repressed by transcriptional regulator PsaR in the presence of manganese (Johnston et al., 2006). These genes have been shown to have role in virulence of pneumococcus (Kloosterman et al., 2008). Therefore, it might be interesting to further explore the role of cysteine in the regulation of these genes.

The expression of putative methionine/cysteine transport and biosynthesis pathway genes (spd-0150, metQ, spd-0431, metEF, gshT, spd-0618, fhs, tcyB, metB-csd, metA, spd-1898, yvdE, and cysK) was significantly upregulated in the presence of 0.03 mM cysteine. The role of methionine in regulation of these genes has been demonstrated in our recent study (Afzal et al., 2016). Furthermore, we showed that a transcriptional regulator CmhR acts as a transcriptional activator of fhs, folD, metB, metEF, metQ, and spd-0431. However, regulatory mechanism of spd-0150, gshT, spd-0618, tcyB, metA, spd-1898, yvdE, and cysK is not explored. Therefore, we decided to further explore the role of cysteine in the regulation of these genes.

To confirm our microarray results and to study the expression of spd-0150, metQ, spd-0431, metEF, gshT, spd-0618, fhs, tcyB, metB-csd, metA, spd-1898, yvdE, and cysK under limiting cysteine concentration in CDM, we performed quantitative RT-PCR on these genes. Our quantitative RT-PCR results demonstrated that the expression of these genes was increased significantly in CDM with 0.03 mM cysteine, when compared to 50 mM (Figure 1). These data not only confirms our microarray results mentioned above, but also suggests a direct role of these genes in cysteine transport and biosynthesis.

There are three LysR-type transcriptional regulators in different bacteria, which have been shown to be involved in the regulation of sulfur amino acids (Sperandio et al., 2005, 2010). S. pneumoniae also has two LysR-type transcriptional regulators (CmhR and CmbR), which are proposed to be involved in the regulation of the sulfur amino acids (Novichkov et al., 2010). Our recent study has revealed the regulatory mechanism of CmhR in S. pneumoniae and demonstrates that CmhR acts as a transcriptional activator of the fhs, folD, metB-csd, metEF, metQ, and spd-0431 in the presence of methionine (Afzal et al., 2016). The presence of CmbR (putative Cysteine Methionine Biosynthesis Regulator) in S. pneumoniae suggests its involvement in the regulation of cysteine-responsive genes. Therefore, we decided to further study the role of transcriptional regulator CmbR in the regulation of cysteine transport and biosynthesis genes.

The presence of cmbR in the S. pneumoniae genome suggests its involvement in the regulation of cysteine-responsive genes. cmbR codes for the putative transcriptional regulator CmbR, which belongs to the LysR family of proteins. CmbR is a homolog of a LysR-type regulator (also called FhuR) of Lactococcus lactis and Streptococcus mutans (Fernández et al., 2002; Sperandio et al., 2005, 2010). To study the role of CmbR in S. pneumoniae D39, we analysed the promoter regions of cysteine-regulated genes and predicted a 21-bp palindromic-like sequence (that has high homology with the FhuR binding site of L. lactis and S. mutans) in the promoter regions of spd-0150, metEF, gshT, spd-0618, tcyB, metA, and yvdE indicating that CmbR regulon in S. pneumoniae D39 is comprised of these genes. This DNA sequence might serve as the CmbR operator site in S. pneumoniae. The CmbR site present in the promoter regions of spd-0150, metEF, gshT, spd-0618, tcyB, metA, and yvdE is shown in Figure 2A. A weight matrix of these CmbR sites (5′-GYGATAAAAAWWAYTTATMAC-3′) was constructed (Figure 2B). Promoter regions of these genes were also examined in other streptococcal species (Streptococcus mitis, Streptococcus agalactiae, Streptococcus gallolyticus, Streptococcus gordonii, S. mutans, Streptococcus sanguinis, Streptococcus suis, and Streptococcus thermophilus) to check whether the CmbR site is also conserved in those streptococci. The CmbR site is highly conserved in these streptococci as well. Moreover, we constructed a phylogenetic tree of CmbR present in different streptococci, which shows that it is highly conserved in these streptococci (Figure 3).

To investigate the role of CmbR in the regulation of spd-0150, metEF, gshT, spd-0618, tcyB, metA, and yvdE, we performed β-galactosidase assays with the promoter lacZ-fusions that were constructed in the S. pneumoniae D39 wild-type and D39 ΔcmhR in CDM with 50 mM cysteine. We used CDM with 50 mM cysteine, because we assumed the role of CmbR as a transcriptional repressor of these genes in the presence of cysteine. The results of the β-galactosidase assays showed that the activity of Pspd-0150-lacZ, PmetE-lacZ, PgshT-lacZ, Pspd-0618-lacZ, PtcyB-lacZ, PmetA-lacZ, and PyvdE-lacZ increased significantly in D39 ΔcmbR compared to D39 wild-type in CDM with 50 mM cysteine (Figure 4). Increased expression of these promoters in D39 ΔcmbR indicates the role of CmbR as a transcriptional repressor of spd-0150, metEF, gshT, spd-0618, tcyB, metA, and yvdE.

To further confirm the role of CmbR as a transcriptional repressor of spd-0150, metEF, gshT, spd-0618, tcyB, metA, and yvdE, we performed quantitative RT-PCR on these genes in the presence of 50mM cysteine. Our quantitative RT-PCR results demonstrated that the expression of spd-0150, metEF, gshT, spd-0618, tcyB, metA, and yvdE increased significantly in D39 ΔcmbR compared to the D39 wild-type (Figure 5). These data provide further confirmation of our results that CmbR acts as a transcriptional repressor of spd-0150, metEF, gshT, spd-0618, tcyBC, metA, and yvdE.

To verify the functionality of the predicted CmbR operator site present in the promoter regions of the CmbR regulated genes (spd-0150, metEF, gshT, spd-0618, tcyBC, metA, and yvdE), we made transcriptional lacZ-fusions of Pspd-0150, PmetE, Pspd-0618, and PmetA, where conserved bases in the cmbR sites were mutated in Pspd-0150 (5′-CCCATAAAAAAAACTTATCAC-3′ to 5′-CCCGCGAAAAAAACTTATCAC-3′), PmetE (5′-CTTATAAGAATTACTAATAAC-3′ to 5′-CTTGCAAGAATTACTAGCAAC-3′), Pspd-0618 (R1: 5′-ATGGTAAAAAATCCTTATAAC-3′ to 5′-ATGGCCAAAAATCCTGCGAAC-3′ and R2: 5′-GCAATAAAAAATAGATATTAT-3′ to 5′-GCAGCGAAAAATAGAGCGTAT-3′), and terminated in PmetA (CmbR site is deleted), and β-galactosidase assays were performed on cells grown in CDM with 50 mM cysteine. β-galactosidase assays data revealed that mutating the CmbR operator site in Pspd-0150 and PmetE, and deletion of the CmbR operator site in PmetA led to significantly increased expression of these promoter lacZ-fusions in the presence of 50 mM cysteine (Figure 6). These data confirm that the predicted CmbR sites present in the promoter regions of these genes are functional and intact in S. pneumoniae D39. Two putative operator sites of CmbR are present in Pspd-0618 (R1 and R2). We mutated both sites individually. We could only observe derepression (caused by CmbR) of Pspd-0618 when operator site 1 (R1) was mutated and did not witness any change in the activity of Pspd-0618 due to mutations in operator site 2 (R2). This suggests that only operator site 1 (R1) is a functional operator site for CmbR in Pspd-0618.