Ni2+-Dependent and PsaR-Mediated Regulation of the Virulence Genes pcpA, psaBCA, and prtA in Streptococcus pneumoniae

Previous studies have shown that the transcriptional regulator PsaR regulates the expression of the PsaR regulon consisting of genes encoding choline binding protein (PcpA), the extracellular serine protease (PrtA), and the Mn2+-uptake system (PsaBCA), in the presence of manganese (Mn2+), zinc (Zn2+), and cobalt (Co2+). In this study, we explore the Ni2+-dependent regulation of the PsaR regulon. We have demonstrated by qRT-PCR analysis, metal accumulation assays, β-galactosidase assays, and electrophoretic mobility shift assays that an elevated concentration of Ni2+ leads to strong induction of the PsaR regulon. Our ICP-MS data show that the Ni2+-dependent expression of the PsaR regulon is directly linked to high, cell-associated, concentration of Ni2+, which reduces the cell-associated concentration of Mn2+. In vitro studies with the purified PsaR protein showed that Ni2+ diminishes the Mn2+-dependent interaction of PsaR to the promoter regions of its target genes, confirming an opposite effect of Mn2+ and Ni2+ in the regulation of the PsaR regulon. Additionally, the Ni2+-dependent role of PsaR in the regulation of the PsaR regulon was studied by transcriptome analysis.


Introduction
Streptococcus pneumoniae, an encapsulated bacterium is a common cause of otitis media, bacterial meningitis, bacteremia, and pneumoniae, leading to millions of death every year, particularly in developing countries [1][2][3]. Although, S. pneumoniae has an asymptomatic association within the human nasopharyngeal cavity [4], it has also the ability to spread to other sites in the human body to cause severe infections [5][6][7]. The survival of S. pneumoniae in different niches inside the human body might depend on the availability of macro-and micro-nutrients on the respective infection sites. Metal ions are an integral part of nutrients, and play a vital role in the regulation of many cellular processes in S. pneumoniae [8][9][10]. The deprivation or excess of metal ions may result in impaired growth of bacterial cells [11]. Therefore, proper regulation of metal homeostasis is important for the survival of S. pneumoniae. For this in the Results section. Standard deviation was calculated from three independent replicates of each sample.

Quantitative real time (qRT)-PCR experiments
For qRT-PCR, S. pneumoniae D39 wild-type was grown in CDM with and without the addition of 0.3 mM Ni 2+ and harvested at mid-exponential growth phase. RNA was isolated as described before [16]. Additionally, RNA was treated with DNase I (RNase-free) (Thermo Fisher Scientific, St. Leon-Rot, Germany) for 60 min at 37°C to remove any DNA contamination. qRT-PCR was performed in triplicates as described before [16]. The transcription level of the target genes was normalized to gyrA transcription using the relative expression software tool [36].

Inductively coupled plasma-mass spectrometry (ICP-MS) analysis
To measure the intracellular concentrations of metal ions, S. pneumoniae D39 was grown till OD 600 = 0.

DNA Microarray Analysis
To observe the impact of the psaR deletion on the transcriptome of S. pneumoniae in the presence of Ni 2+ , S. pneumoniae D39 wild-type and its isogenic psaR mutant (RW100) [9] were grown in two biological replicates in CDMchelex with 0.3 mM of NiSO 4. (H 2 O) 6 . Cells were harvested at the mid-exponential growth phase. Further experiments were performed essentially as described before [37]. DNA microarray data were analyzed by using the MicroPrep software package as described before [38]. To identify differentially expressed genes a Bayesian p-value <0.001 and a fold-change cut-off of 2 were applied. The DNA microarray data have been deposited to Gene Expression Omnibus (GEO) with accession number GSE73818.

Purification of Strep-tagged PsaR and Electrophoretic mobility shift assays
The overexpression and purification of C-terminally Strep-tagged PsaR was achieved in L. lactis NZ9000 essentially as described before [9,39]. Electrophoretic mobility shift assays (EMSAs) were performed essentially as described previously [10]. In short, PCR products of PpcpA, PpsaB, PprtA, and PadcR were labeled with [γ- 33

Results
Ni 2+ -dependent expression of the PsaR regulon in S. pneumoniae In a previous study, we have shown that, like Zn 2+ , Co 2+ also induces the expression of the PsaR regulon, while addition of Mn 2+ causes repression of the PsaR regulon [21]. The PsaR regulon comprises the psa operon (psaBCA), encoding Mn 2+ -dependent ABC transporters, pcpA, encoding a choline binding protein and prtA, encoding a serine protease. In this study, we decided to explore the impact of Ni 2+ on the expression of the PsaR regulon. To investigate the impact of Ni 2+ on the expression of the PsaR regulon, cells were grown in CDM with either 0 or 0.3 mM Ni 2+ , and qRT-PCR was performed. qRT-PCR data revealed that the expression of pcpA, psaBCA, and prtA was highly upregulated in the presence of 0.3 mM Ni 2+ compared to 0 mM Ni 2+ (Table 3), suggesting the putative role of Ni 2+ in the regulation of the PsaR regulon.
To further verify the role of Ni 2+ in the regulation of the PsaR regulon in S. pneumoniae, the D39 wild-type strain containing either PpcpA-lacZ, PpsaB-lacZ, or PprtA-lacZ was grown in CDMchelex and CDMchelex-Mn 2+ (CDMchelex without Mn 2+ ) with the addition of 0, 0.1, 0.3 or 0.5 mM Ni 2+ , and β-galactosidase assays were performed. Our β-galactosidase data (Miller Units) revealed that the expression of the PpcpA-lacZ, PpsaB-lacZ, and PprtA-lacZ increased significantly with increasing concentrations of Ni 2+ in CDMchelex and CDMchelex-Mn 2+ (Table 4). However, the expression of these transcriptional lacZ-fusions was much higher in CDMchelex-Mn 2+ compared to CDMchelex due to the unavailability of Mn 2+ in CDMchelex-Mn 2+ . This data indicates that the expression of pcpA, psaBCA, and prtA is regulated by Ni 2+ and in agreement with our qRT-PCR analysis data mentioned above.

PsaR mediates expression of the PsaR regulon in the presence of Ni 2+
To check, whether the observed Ni 2+ -dependent high expression of the PsaR regulon is mediated by the Mn 2+ / Zn 2+ / Co 2+ -responsive transcriptional regulator PsaR, the psaR mutant strain (RW100) containing PpcpA-lacZ, PpsaB-lacZ, and PprtA-lacZ were grown in Table 3. The relative expression of prtA, psaB, psaC, psaA, and pcpA genes was normalized with the housekeeping gene gyrA. The log 2 fold increase is relative to the expression in the D39 wild-type grown in CDMchelex with 0.3 mM Ni 2+ to that with 0 mM Ni 2+ . Standard deviation of three independent replications is given in parentheses. CDMchelex with 0, 0.1, 0.3 or 0.5 mM Ni 2+ . The expression of PpcpA-lacZ, PpsaB-lacZ, and PprtA-lacZ was highly derepressed in the psaR mutant. We did not observe significant difference in the expression of PpcpA-lacZ, PpsaB-lacZ, and PprtA-lacZ in the psaR mutant strain at different concentrations of Ni 2+ (Table 4), indicating that PsaR mediates the Ni 2+ -dependent expression of the PsaR regulon.
To analyze the impact of psaR deletion on the global gene expression of S. pneumoniae and find more targets of PsaR in the presence of Ni 2+ , transcriptome of psaR mutant strain was compared with S. pneumoniae D39 wild-type strain grown in CDMchelex with 0.3 mM Ni 2+ . The expression of psaR was significantly downregulated, confirming the inactivation of psaR in the psaR deletion strain. The expression of pcpA, psaBCA, and prtA was highly upregulated in the psaR mutant (Table 5). This data further confirms our β-galactosidase data mentioned above indicating Ni 2+ -dependent derepression of the PsaR regulon. We did not find any new target of PsaR in the presence of Ni 2+ . Notably, an operon (spd_0616-spd_618) encoding amino acid ABC transporter proteins was downregulated in our transcriptomic analysis, but in our β-galactosidase assay we did not observe any activity of the respective promotor of this operon in the psaR mutant (Data not shown here).

Opposite effect of Ni 2+ and Mn 2+ in the regulation of the PsaR regulon
Previous studies showed that the PsaR-mediated expression of the PsaR regulon depends on the balance between Mn 2+ , Co 2+ and/ or Zn 2+ [9,21,31]. In this study, we observed that the expression of the PsaR regulon was highly derepressed in response to various Ni 2+ concentrations. Therefore, we decided to explore the influence of Ni 2+ and Mn 2+ together on the expression of the PsaR regulon. The expression of PpcpA-lacZ, PpsaB-lacZ, and PprtA-lacZ in S. pneumoniae D39 wild-type was measured at different concentrations of Ni 2+ and Mn 2+ in CDMchelex and CDMchelex-Mn 2+ (Table 6). β-galactosidase data (Miller units) showed that high expression of PpcpA-lacZ, PpsaB-lacZ, and PprtA-lacZ at 0.1 or 0.3 mM of Ni 2+ was nullified by the addition of 0.02 or 0.05 mM Mn 2+ (Table 6). However, Mn 2+ repression was higher in CDMchelex compared to CDMchelex-Mn 2+ . This might be due to the fact that CDMchelex contains 5-7 μM of Mn 2+ which is enough to cause the repression of the PsaR regulon [21]. These results suggest that the Mn 2+ -dependent repression of the PsaR regulon is derepressed by the addition of Ni 2+ . Ni 2+ counteracts the Mn 2+ -PsaR interaction with PpcpA, PpsaBCA, and PprtA To find out whether the observed opposite effects of Ni 2+ and Mn 2+ on the expression of pcpA, psaBCA, and prtA are mediated by the direct DNA binding activity of the PsaR protein, the effects of these metal ions on the binding of PsaR-Strep tag to 33 P-labeled promoters of pcpA, psaB, and prtA were studied in vitro. The promotor region of phtB was used as a negative control. Due to the metal-ion chelating ability of EDTA, we decided to exclude it from all buffers used to perform EMSAs. PsaR-Strep tag was not able to bind with the promoter regions of pcpA, psaB, and prtA without the addition of any metal ion (Fig 1A, 1B and 1C. Lane 2) which is in agreement with the previous study [9]. First of all, we checked the DNA binding activity of PsaR-Strep to the promoter regions of pcpA, psaB, and prtA with different concentrations of Mn 2+ . We observed that 0.05 and 0.1 mM Mn 2+ were able to stimulate the binding of PsaR--Strep tag to the promoter region of pcpA. However, only 0.1 mM Mn 2+ was able to stimulate the binding of PsaR-Strep to the promoter regions of psaB and prtA (Fig 1A, 1B and 1C. Lane 4). No binding of PsaR-Strep to the promoter regions of psaB, and prtA was observed at 0.05 mM Mn 2+ (Fig 1A, 1B and 1C. Lane 3). Interestingly, no shift in the promoter regions of pcpA, psaB, and prtA was observed with 0.2 or 0.4 mM Ni 2+ (Fig 1A, 1B and 1C. Lane 5 and 6), suggesting that Ni 2+ does not stimulate the binding of PsaR with pcpA, psaB, and prtA promoters. Previously, it has been shown that Zn 2+ binds to the PsaR in such a way which leads to the inactivation of Mn 2+ -PsaR interaction with the promoter regions of pcpA, psaB, and prtA [9]. We hypothesized that like Zn 2+ , Ni 2+ also interferes in the Mn 2+ -dependent binding of PsaR--Strep to the promoter regions of pcpA, psaB, and prtA. Therefore, we decided to explore the influence of Ni 2+ on the in vitro Mn 2+ -PsaR-Strep tag interaction. Interestingly, the binding of PsaR to all three promoters in the presence of Mn 2+ was impaired with the addition of Ni 2+ (Fig 1 Lanes 7-10). This data suggests that the Mn 2+ -PsaR interaction with pcpA, psaB, and  prtA promoters is competed away in the presence of Ni 2+ , indicating a direct role of Ni 2+ in the regulation of the PsaR regulon through PsaR.

A high concentration of Ni 2+ in the medium leads to Mn 2+ deficiency in the cells
To determine the cell-associated concentrations of metal ions, we performed an ICP-MS analysis on the cells grown in CDMchelex either with 0 or 0.3 mM of Ni 2+ . ICP-MS data revealed that the cells grown in the presence of 0.3 mM Ni 2+ accumulate 10-fold (P<0.01, One way ANOVA) more Ni 2+ (Fig 2A) compared to cells grown in the absence of Ni 2+ . No significant difference in the concentrations of other metal ions was observed in our ICP-MS analysis except for Mn 2+ . The concentration of Mn 2+ was reduced by 1.5-fold (P<0.01, One way ANOVA) in the presence of Ni 2+ (Fig 2A). This data indicates that high concentration of Ni 2+ leads to Mn 2+ deficiency in the cell. To study this in more details, we have checked the impact of various concentrations of Ni 2+ on the cell-associated Mn 2+ . Cells were grown in CDMchelex with the addition of 0.02 mM Mn 2+ , and 0, 0.1, 0.3 or 0.5 mM Ni 2+ . As expected, addition of Ni 2+ in medium leads to an increased cell-associated Ni 2+ concentration. The cell-associated Ni 2+ concentration was increased by 2-fold (P<0.01, One way ANOVA) at 0.1 mM Ni 2+ , 13-fold at 0.3 mM Ni 2+ , and 16-fold at 0.5 mM Ni 2+ when compared to 0 mM Ni 2+ (Fig 2B). ICP-MS analyses data further revealed that an increasing concentration of Ni 2+ leads to a decrease in the concentrations of Mn 2+ . The cell-associated concentration of Mn 2+ was decreased by 1.25-fold (P<0.01, One way ANOVA) at 0.1 mM Ni 2+ , 3.52-fold (P<0.01, One way ANOVA) at 0.3 mM Ni 2+ , and 7.4-fold (P<0.01, One way ANOVA) at 0.5 mM Ni 2+ ( Fig  2B) when compared to the Mn 2+ concentration at 0 mM Ni 2+ . Notably, the cell-associated concentration of other metal ions (Zn 2+ , Fe 2+ , and Co 2+ ) was not affected (Fig 2). This data demonstrate that Ni 2+ has ability to cause Mn 2+ starvation which ultimately leads to the high expression of the PsaR regulon in the presence of Ni 2+ .

Discussion
Adherence to epithelial cells of human nasopharynx is the primary step of S. pneumoniae towards the pathogenesis [40]. The pneumococcal surface adhesion protein, PsaA and choline binding protein, PcpA are among those proteins that promote pneumococcal adherence in nasopharyngeal epithelial cells and colonization in mice [14,41,42]. Similarly, PrtA, a serine protease containing an LPXTG-anchor motif, is expressed on the surface of nearly all virulent pneumococcal strains and is required for full virulence in animal models [43,44]. The pcpA, psaBCA, and prtA genes comprise the PsaR regulon and their expression is regulated by transcriptional regulator PsaR [9]. The role of Mn 2+ , Zn 2+ , and Co 2+ in the regulation of pcpA, psaBCA, and prtA (PsaR regulon) has already been established [9,21,45]. In this study, we investigated the role of Ni 2+ on the expression of the PsaR regulon. The expression of the PsaR regulon was increased with the increasing concentrations of Ni 2+ and this increased expression of the PsaR regulon is directly linked with cell-associated Mn 2+ deficiency caused by a high concentration of Ni 2+ . Moreover, Mn 2+ and Ni 2+ have opposite regulatory effects on the expression of the PsaR regulon in S. pneumoniae. Where, Mn 2+ -binding represses the expression of the PsaR regulon, Ni 2+ derepresses the repression caused by Mn 2+ . Mn 2+ is an important transition metal ion that is a cofactor for many pneumococcal proteins which are involved in the colonization, virulence, and resistance to oxidative stress in S. pneumoniae [15]. Mn 2+ accumulation shows significant flexibility and cells can survive even at a 3% concentration of the normal accumulation level [45,46]. S. pneumoniae has a dedicated system for Mn 2+ transport (PsaBCA) that consists of two ABC transporters (PsaBC) and a cell surface salute binding protein (PsaA) [47][48][49]. Previous studies have shown that PsaA is not only important for virulence [14,41], but also has a direct role in the accumulation of cell associated Mn 2+ [48,49]. PsaA has the ability to bind Zn 2+ and Mn 2+ [46,48]. The binding affinity of PsaA to Zn 2+ is much higher compared to that of Mn 2+ , and PsaA-Zn 2+ interaction led to the~40% decrease in cell associated Mn 2+ accumulation [31,46]. Structural studies of PsaA have revealed that Cd 2+ can also bind to PsaA and ultimately results in the reduction of cell-associated Mn 2+ [30]. Recently, it was shown that PsaA can also bind to other d-block elements including Ni 2+ [50]. This unique property of PsaA to bind with different metal ions makes its role very important in the life style of S. pneumoniae. In our ICP-MS analysis, we observed a cell-associated Mn 2+ deficiency in the presence of relatively high concentrations of Ni 2+ . Therefore, based on our ICP-MS data, we can speculate that most likely Ni 2+ interacts with PsaA, which leads to Mn 2+ deficiency.
Biochemical studies of transcriptional regulator PsaR of S. pneumoniae showed that PsaR harbors two pairs of metal binding sites where Mn 2+ or Zn 2+ can bind [51]. Similarly, Mn 2+responsive regulators DtxR from Corynebacterium diphtheria and MntR from Bacillus subtilis, which are homologous of PsaR, also have two metal binding sites [52,53]. The binding of DtxR to the tox operon in C. diphtheria not only depends on the availability of Mn 2+ but also on Co 2+ , Fe 2+ , and Ni 2+ [54]. Similarly, The Mn 2+ -dependent DNA binding activity of MntR in B. subtilis is diminished in the presence of Ni 2+ , Zn 2+ , and Fe 2+ [55][56][57]. The metal responsive transcriptional regulators, ScaR of Streptococcus gordonii and SloR of Streptococcus mutants also belongs to DtxR family, and are homologous to PsaR [58][59][60]. Interestingly, the PsaR binding site is similar to the operator sequences of ScaR and SloR [61]. This might suggest that PsaR uses a similar mechanism of metal ion competition for regulatory metal ion homeostasis as other member of DxtR family regulators adopt.
It has been previously demonstrated that PsaR represses the expression of the PsaR regulon in the presence of Mn 2+ whereas Zn 2+ and Co 2+ relieved this repression [21,61]. Moreover, the in vitro studies of the interaction of PsaR to its target promotors showed that both Zn 2+ and Co 2+ could bind to PsaR in a different way [21]. When Zn 2+ interacts with PsaR, it relieves the PsaR interaction with the promoter regions of the PsaR regulon, whereas Co 2+ , just like Mn 2+ , stimulates the interaction of PsaR with the promoter regions of the PsaR regulon [9,21]. Here, we demonstrated that Mn 2+ -PsaR interaction leads to the binding of PsaR to the promoter regions of pcpA, psaBCA, and prtA which is an agreement with previous studies [9]. However, the Mn 2+ -PsaR interaction with pcpA, psaB, and prtA promoters was alleviated by the addition of Ni 2+ which suggests that the observed transcriptional response of the PsaR regulon is directly linked to the interaction of Ni 2+ and Mn 2+ on the PsaR-promoter interactions. In conclusion, we have shown that the interaction of PsaR to Ni 2+ plays a similar role as Zn 2+ , to induce derepression by PsaR in competition with Mn 2+ .