Pneumococcal metabolic adaptation and colonization is regulated by the two-component regulatory system 08

Streptococcus pneumoniae two-component regulatory systems (TCS) enable adaptation and ensure its maintenance in host environments. This study deciphers the impact of the TCS08 on pneumococcal gene expression and its role in metabolic and pathophysiological processes. Transcriptome analysis and real-time PCR demonstrated a regulatory effect of the TCS08 on genes involved mainly in environmental information processing, intermediary metabolism, and colonization by S. pneumoniae D39 and TIGR4. Striking examples are genes of the fatty acid biosynthesis, arginine-deiminase system, and psa operon encoding the manganese ABC transport system. In silico analysis confirmed that TCS08 is homologous to Staphylococcus aureus SaeRS and a SaeR-like binding motif is displayed in the promotor region of pavB, the upstream gene of the tcs08 operon encoding a surface-exposed adhesin. Indeed, PavB is regulated by the TCS08 as confirmed by immunoblotting and surface abundance assays. Similarly, Pilus-1 of TIGR4 is regulated by TCS08. Finally, in vivo infections using the acute pneumonia and sepsis models showed a strain dependent effect. Loss of function of HK08 or TCS08 attenuated D39 virulence in lung infections. The RR08 deficiency attenuated TIGR4 in pneumonia, while there was no effect on sepsis. In contrast, lack of HK08 procured a highly virulent TIGR4 phenotype in both pneumonia and sepsis infections. Taken together, these data indicate the importance of TCS08 in pneumococcal fitness to adapt to the milieu of the respiratory tract during colonization. IMPORTANCE Streptococcus pneumoniae interplays with its environment by using 13 two-component regulatory systems and one orphan response regulator. These systems are involved in the sensing of environmental signals thereby modulating pneumococcal pathophysiology. This study aimed to understand the functional role of genes subject to control by the TCS08. The identified genes play a role in transport of compounds such as sugars or amino acids. In addition, the intermediary metabolism and colonization factors are modulated by TCS08. Thus, TCS08 regulates genes involved in maintaining pneumococcal physiology, transport capacity and adhesive factors to enable optimal colonization, which represents a prerequisite for invasive pneumococcal disease.


INTRODUCTION
encode for RR08 and HK08) is highly homologous to the SaeRS system of 99 Staphylococcus aureus (24), where it has been associated with the regulation of genes 100 encoding α-hemolysin (hla), coagulase (coa), fibronectin (Fn) binding proteins and 20 101 other virulence factors (25)(26)(27). Interestingly, the SaeRS system of S. aureus has been 102 shown to respond to sub-inhibitory concentrations of α-defensins and high 103 concentrations of H2O2, suggesting a sensing mechanism responsive to host immune 104 system molecules and membrane alterations (26,27). In pneumococci, a previous study 105 on TCS08 has revealed its importance for pneumococcal virulence (11). Moreover, two 106 reports have shown a regulatory effect of the pneumococcal TCS08 on the rlrA 107 pathogenicity islet (pilus-1 or PI-1) and the cellobiose phosphotransfer system (PTS) 108 (24, 28). Hence, the initial information available about this system suggests its 109 involvement in pneumococcal adaptation, fitness and virulence. Nevertheless, its target 110 genes and its precise role in pneumococcal pathogenicity are yet to be defined.

113
Influence of TCS08 on pneumococcal growth behavior in chemically-defined 114 medium 115 To investigate the effect of loss of function of TCS08 components on 116 pneumococcal fitness, nonencapsulated S.p. D39 and TIGR4 parental strains and their 117 isogenic mutants were cultured in a chemically-defined medium (CDM). All strains 118 presented a similar growth pattern and reached similar cell densities in the stationary 119 phase, with the exception of the TIGR4ΔcpsΔrr08 mutant (Fig. 1). A steeper logarithmic 120 phase was detected in the rr08 mutant in TIGR4 ( Fig. 1A and 1E). Additionally, the 121 calculated growth rates of the different mutants in both D39 and TIGR4 strains 122 suggested a significant reduction in the generation time of the rr08 mutant in TIGR4 123 (Fig. 1A). The observed behavior among the TCS08 mutants in the CDM used in this 124 study may point to strain-dependent specific effects. 125 126 Impact of TCS08 on TIGR4 gene expression 127 The initial screening for the effects of TCS08 inactivation on gene expression was 128 conducted by microarrays using RNA samples extracted from TIGR4∆cps and its 129 isogenic rr08-, hk08-, and tcs08-mutants grown in CDM. Genes presenting significant  SP_0198 and SP_0899 (29). These 172 proteins contain conserved lipobox motifs and are therefore also thought to be surface-173 exposed and might be involved in unknown fitness related processes. 174 175 TCS08 is involved in the regulation of metabolic functions of S. pneumoniae 176 Results obtained by the microarray screening suggested a regulatory effect of 177 TCS08 in the expression of genes involved in the uptake and transport of essential 178 nutrients for S.p. TIGR4 such as arginine and manganese ( Fig. 2 and Table S1). These 179 metabolites/ ions are transported into the cell via specific ABC transporter systems. Of 180 particular interest is the arginine-deiminase system (ADS), which is essential for arginine 181 uptake and utilization in pneumococci. All genes of the arcABCDT operon displayed 182 important changes in their expression in the absence of the RR or the HK08.

183
Interestingly, these changes were not consistent in both mutants as the Δrr08 stain 184 displayed a significant downregulation of this operon while the hk08 mutant showed an 185 upregulation ( Fig. 2 and table S1). Additionally, no significant effects were observed for 186 the arc operon in the Δtcs08 mutant. Analysis by qPCR partially confirmed the initial 187 findings on the expression of the arc operon and demonstrated a strain-dependent effect 188 for these genes. Indeed, the expression of the arginine deiminase gene arcA was only 189 significantly increased in the rr08 and hk08 in TIGR4 (Fig. 3A), whereas no 190 differences were found in D39 (Fig. 3B). Furthermore, the arginine-ornithine antiporter 191 arcD (30, 31) presented a similar expression to arcA in TIGR4 and D39 TCS08 mutants, 192 however the changes were not significant (Fig. 3). An additional key player in the 193 pneumococcal fitness is the psa operon. This operon plays a role in the uptake of 194 manganese and in the response to oxidative stress in the pneumococci. The analysis by 195 microarray showed a significant increase of 2-fold in the expression of the psa operon 196 for the hk08 mutant in the TIGR4 strain ( Fig. 2 and Table S1). Conversely, no 197 statistically important effects were observed in the psa operon in the rr08 and tcs08 198 mutants in the same strain ( Fig. 2 and Table S1). Validation of the microarray data by 199 qPCR discovered a significant increase in the expression of psaA in the rr08 mutant of 200 D39. Surprisingly, the microarray data for the psa operon could not be confirmed by 201 qPCR in TIGR4 (Fig. 3A).

202
Immunoblot analyses of pneumococci cultured in CDM were carried out to 203 elucidate the effect of TCS08 components on the protein levels of selected candidates 204 from D39 and TIGR4 based on gene expression data (Fig. 4). For the ADS system, the

207
On the contrary, the loss of the HK08 in TIGR4 resulted in a 2-fold lower abundance of 208 ArcA (Fig. 4A). The remaining rr08 and tcs08 mutants in both strains showed non-209 significant effects in the protein levels of ArcA. Interestingly, the results obtained for the 210 ArcA protein in the absence of the HK08 in both strains did not reflect the transcriptome 211 (2-fold upregulation) or qPCR results. In the case of PsaA, the immunoblot analysis 212 confirmed a significantly higher expression of 1.5-fold in the TIGR4 hk08 (Fig. 4A), 213 correlating with the microarray data (Fig 2).

214
In a complementary approach, the surface abundance of PsaA was examined by a 215 flow cytometric approach (Fig. 5). For D39 a non-significant increase in the surface 216 abundance of PsaA was measured in mutants lacking both TCS08 components. The 217 low effect of the TCS08 on PsaA observed for surface abundance correlates with the 218 immunoblot ( Fig. 4 and 5). Similarly, the increased surface abundance of PsaA in TIGR4 219 mutants lacking the HK08 (Fig. 5) correlated with the immunoblot and microarray 220 analysis. The adhesins PavB and PI-1 were shown to be regulated in the TIGR4 strain by 224 our initial microarray analysis ( Fig. 2 and Table S1) and confirmed by qPCR.

225
Interestingly, pavB is a gene upstream of the 5' region of the tcs08 operon presenting 226 properties of a sortase-anchored adhesin. PavB has been shown to interact with various 227 extracellular matrix proteins and probably also directly with a cellular receptor (32, 33), with an upregulation in mutants lacking the HK08 by at least 2-fold ( Fig. 3). Moreover, 234 the absence of both components of the TCS08 leads to a significantly reduced 235 expression the pilus-1 in TIGR4. While, no significant effect was seen for pavB in either 236 the rr08 or tcs08 mutant in neither D39 or TIGR4 strains at the gene expression level.

237
On the protein level, quantifications were performed by immunoblotting (Fig. 4) and 238 the levels of surface abundance were evaluated by flow-cytometry (Fig. 5). For PI-1, the 239 backbone protein RrgB was used as representative. Immunoblot analysis and flow-240 cytometry indicated higher protein levels and surface abundance, respectively, in 241 mutants lacking HK08 and RR08. These results are in line with gene expression 242 analyses. Importantly, the lower protein levels of RrgB in the absence of both TCS08 243 components correlated with the downregulation measured by qPCR and transcriptomics 244 ( Fig. 4A and 5A). For PavB, immunoblots revealed a high impact on PavB amounts in 245 the different mutants with a 2-fold increase in the absence of HK08 in D39 and even 10-246 fold in TIGR4. In contrast, the lack of either the RR08 or both components of the TCS08 247 procured a 2-fold decrease of PavB in both D39 and TIGR4 (Fig. 4). Similar, the surface 248 abundance of PavB was higher in the hk08-mutant and lower in the rr08-and tcs08-249 mutants as indicated by flow cytometry (Fig. 5). Importantly, these data fit with the gene 250 expression analysis of the mutants by microarrays.

259
To assess the impact of the TCS08 or its individual components on pneumococcal 260 colonization, lung infection or sepsis, CD-1 mice were intranasally or intraperitoneally 261 infected with bioluminescent wild-type strains (D39 or TIGR4) and corresponding 262 isogenic mutants. In D39, intranasal infections with mutants lacking either the HK08 or 263 both components of the TCS08 increased the survival time of mice, thus the mutants 264 were attenuated and represent a less virulent phenotype ( Fig. 7B and F). The rr08-265 mutant of D39 showed no differences in developing lung infections ( Fig. 7B and F). In 266 the sepsis model no differences between the wild-type of D39 and its isogenic mutants 267 were observed (Fig. 7H). Strikingly and in contrast to D39 infections, the acute 268 pneumonia and sepsis infection models indicated a higher virulence potential of TIGR4 269 bacteria lacking the HK08. On the contrary, the loss of the RR08 in the TIGR4 genetic  The role of a subset of pneumococcal TCS in competence, physiology, and 284 virulence has been characterized providing an initial understanding of their specific 285 regulons (10, 12, 37). As such, TCS08 of S. pneumoniae has been initially identified and 286 suggested to be important for virulence (11,12,37). Nevertheless, the mechanism 287 underlying its effect on pathophysiological processes has not been elucidated before. A

302
The staphylococcal SaeRS TCS is known to be essential for the virulence of  Interestingly, we observed the most predominant regulation for genes participating in 311 environmental information processing and intermediary metabolism ( Fig. 2 and table 312 S1). The genes grouped in these two categories are annotated as part of ABC 313 transporters, phosphotransferase systems, and lipid biosynthesis and were found to be 314 localized all along the pneumococcal genome (Fig. S3). The genes found to be 315 regulated by TCS08 share an important feature, namely their localization and/or activity 316 in the pneumococcal membrane. Additionally, several of the different PTS and ABC 317 transporters regulated by TCS08 are involved in fitness and virulence of this pathogen.

318
Hence, the effect of the TCS08 is more pronounced in the colonization phase of the 319 pneumococcal life cycle. This is for example the case for the neuraminidase NanA,  ArcA. Remarkably, compared to the respective wild-type strains ArcA occurred at higher 331 protein levels in all mutants of D39 and the TIGR4 mutant lacking both the HK08 and 332 RR08 (2-fold), while ArcA had lower protein levels in the TIGR4 mutants lacking either 333 HK08 or RR08 (2-fold). However, only the opposite effect of deletion of hk08 on the 334 ArcA level was statistically significant. This is a further proof that the ADS in D39 and 335 TIGR4 is differentially regulated as has been shown before for the stand-alone regulator 336 ArgR2. There, the arc operon showed a constitutive expression in D39, while in TIGR4 337 gene expression was upregulated by ArgR2 (31). 338 It is essential that pneumococci activate their metabolic inventory when colonizing 339 their host to ensure adaptation and fitness. As such, our results point to a role of TCS08 340 in the fine tuning of colonization and metabolic homeostasis as exemplified by the level 341 of change in the expression of pavB, and the genes of the pilus-1, fab, and arc operons.

342
PavB belongs to a group of genes regulated by the TCS08 which are strongly 343 involved in colonization by its interactions with host proteins (32, 33). These group of 344 genes encode mostly for surface-exposed proteins associated to peptidoglycan 345 metabolism and adherence to host cells. These genes are found grouped clockwise 346 mostly in the first quarter of the pneumococcal genome, and transcription and replication 347 proceed into the same direction (Fig. S3). Interestingly, the regulation of the adhesins gene expression and protein abundance in the D39 strain. A minor but significant 357 differential pavB gene expression was measured by microarray analysis and qPCR for 358 TIGR4 ( Fig. 2 and 3). In contrast, PavB protein levels were significantly affected in all 359 mutants, with a 2-fold increase in the absence of the HK08 and a decrease in PavB in 360 mutants lacking either the RR08 or both components of the TCS08 as shown by 361 immunoblot and flow-cytometry ( Fig. 4 and 5).

362
The staphylococcal fibronectin binding protein FbnA is weakly regulated by the 363 SaeRS system of S. aureus (48), which in pneumococci correlates with the link found 364 between TCS08 and PavB/PI-1. A direct repeat sequence (TTTAAN7TTTAA), similar to 365 the imperfect SaeR binding site (GTTAAN6TTTAA) (49), can be found directly upstream 366 of pavB (Fig. 6) suggesting that the RR08 binds directly to the pavB promotor region. A 367 strong hint for the pavB gene regulation by the TCS08 is the higher abundance of PavB 368 in the absence of the HK08. Surprisingly, a conserved repeat sequence 369 TTTAAN14GTTAA was found close to the rlrA operon and could indicate an indirect 370 effect of the TCS08 in the regulation of the pilus-1 via its positive regulator RlrA (Table   371 S5). The in silico search for SaeR-like binding motifs among different TCS08 regulated 372 genes indicated the presence of a variation of this binding sequence for the cellobiose 373 and arc operons, while it was absent for the psa operon (Table S5). All of the genes 374 encoded in these operons have been reported to be under the regulation of CcpA-375 dependent stand-alone regulators (31, 50-52). Additionally, the psa operon has been 376 also shown to be under the regulation of the PsaR and TCS04 (PnpRS), which might be 377 interplaying with the TCS08 (53, 54). This suggests either a cooperative role or a 378 collateral effect of TCS08 and we hypothesize that the TCS08 acts as a membrane 379 stability sensor system.

380
The staphylococcal SaeRS was further reported to regulate proteases and being 381 involved in biofilm formation. Our microarray analysis showed an effect on the 382 expression for genes encoding a putative protease domain ( Fig. 2 and 3)  Interestingly, the SpdA, SpdB, and SpdC proteins have been reported to be involved in 387 the deposition and surface abundance of sortase-anchored proteins in S. aureus (55).

388
The gene expression of sp_0144 (TIGR4) presented an upregulation in the hk08 mutant 389 in TIGR4. It cannot be ruled out that the changes in SP_0144 also contribute to the 390 protein abundance demonstrated for PavB or PI-1 when the strains lack components of 391 the TCS08 (Fig. 3). In turn, changes in surface abundance of colonization factors will 392 interfere with the pneumococcal virulence and /or immune evasion. However, this 393 hypothesis was not evaluated in this study and needs experimental proof in a follow up 394 study.

395
Nasopharyngeal colonization by pneumococci requires adherence to host cells and 396 generates a foothold in the human host. Hence, the regulation of adhesins and ECM 397 binding proteins like PavB or PI-1 represents a successful strategy of the pathogen to 398 adapt to this host compartment. Similar, the sensing of human neutrophil peptides and 399 membrane disruption molecules is also essential to ensure a successful colonization 400 and immune escape phenotype. Our in vivo studies using pneumonia and sepsis murine 401 models confirmed the contribution of the pneumococcal TCS08 in colonization but also 402 virulence (Fig. 7). However, the effect is strain dependent, highlighting the role and 403 network of different stand-alone regulators and other regulatory systems of 404 pneumococci on the overall regulation of pneumococcal fitness and pathophysiology.

405
Such strain-dependent effects have been also shown for additional pneumococcal TCS 406 such as PnpRS and TCS09 (ZmpRS) (53, 56). Remarkably, a more virulent phenotype 407 was observed for the TIGR4 mutant lacking the HK08, while the TIGR4 deficient for the 408 RR08 displayed a decrease in virulence in the pneumonia model (Fig. 7A). In D39, the 409 opposite effect with a slight increase in survival was observed in the absence of the 410 HK08 in the same infection model (Fig. 7B). Additionally, the loss of function of both 411 TCS08 components in strain D39 resulted in a significant reduction in virulence in the 412 pneumonia model (Fig. 7B). Strikingly, this D39 attenuation was not observed in the 413 sepsis model. Similar, the TIGR4 rr08 mutant was also as virulent as the wild-type, 414 despite being attenuated in the pneumonia model ( Fig. 7G and 7H). In contrast, the 415 TIGR4Δhk08 mutant was significantly more virulent than the wild-type in the sepsis 416 model (Fig. 7 G). As such, our results suggest that the TCS08 is mostly involved in 417 bacterial fitness and regulation of adhesins required for a successful colonization. Such 418 striking difference between two representative pneumococcal strains may reflect their 419 different genomic background and the overall versatility of pneumococci.

420
Interesting pathophenotypes were observed in competitive mouse infections, i.e.

421
coinfections of the TIGR4 wild-type and its tcs08 isogenic mutants (Fig. S2). While the 422 pneumonia model showed an avirulent phenotype in the absence of the RR08, this 423 mutant revealed a higher competitive index when compared to its wild-type in the 424 coinfection assay in both, the nasopharyngeal and bronchoalveolar lavages, indicating 425 lower numbers of the wild-type in these host compartments. In addition, TIGR4 mutants 426 lacking either the HK08 and or both components of the TCS08 were apparently 427 outcompeted by the wild-type (Fig. S2) despite being more virulent than the wild-type as 428 indicated in the acute pneumonia model. A plausible explanation for this phenomenon 429 might be that the TIGR4 mutant lacking the HK08 is rapidly progressing from the 430 nasopharynx and lungs into the blood, and hence, low numbers are present in the 431 nasopharynx and lavage. Similar, the absence of the RR08 impairs progressing into the 432 blood and thus, higher numbers of the rr08-mutant are found in the nasopharynx.

433
Indeed, this pneumococcal behavior post-nasopharyngeal infection can also be 434 visualized in the bioluminescent images of the acute pneumonia model, in which the 435 mice infected with the strain lacking the HK08 rapidly developed pneumonia and sepsis 436 (Fig. 7A).

437
It is also important to mention here the mild impact of TCS08 on gene expression 438 alterations. This suggests a role for the TCS08 as a fine tuning and signal modulation 439 system, which is dependent on additional regulators. This hypothesis is supported by  However, a more thorough biochemically analysis would be needed to generate a 452 comprehensive regulatory map within pneumococcal regulators.

453
In conclusion, this study identified five main groups of genes influenced by the 454 pneumococcal TCS08 in a strain-specific manner. A high number of these genes 455 encode proteins involved in environmental signal processing, intermediary metabolism, 456 colonization or genetic information processing. Furthermore, most of the TCS08-457 regulated proteins are membrane-bound and involved in nutrient transport as well as 458 fatty acid biosynthesis. Additionally, surface-exposed PavB and PI-1 islet proteins 459 involved in adhesion to host components were confirmed to be controlled by the TCS08.

460
Thus, the HK08 of the TCS08 is probably sensing small molecules entering the 461 membrane compartment of pneumococci and adapts thereby the pneumococcus to the 462 specific environmental conditions during colonization.  Table S2.

467
Pneumococcal wild-type and isogenic tcs08 deletion mutants were grown on Columbia 468 blood agar plates (Oxoid) containing selection antibiotics (kanamycin 50 µg/ml and 469 erythromycin 5 µg/ml or spectinomycin 100 µg/ml) using an incubator at 37ºC, 5% CO2. The oligonucleotides and plasmid constructs used in this study are depicted in 482   Table S3 and Table S4. The isolation of pneumococcal chromosomal DNA was 483 achieved by using the standard phenol-chloroform extraction protocol. Briefly, 484 S. pneumoniae strains were cultured in blood agar for 6 hours, transferred to new blood 485 agar plates with antibiotics and grown for 10 hours at 37°C and 5% CO2. After   the insertion-deletion strategy was applied by amplifying 5' and 3' flanking regions of 502 rr08, hk08 and the full rr08-hk08 operon via PCR with specific primers. The genomic 503 fragments were cloned in a pGEM-t easy vector and transformed into E. coli DH5α and 504 further processed by inverse PCR using primers to delete the desired target gene and 505 replacing it with either spectinomycin (aad9) or erythromycin (erm R ) resistance gene 506 cassettes. To achieve the deletion of the desired regions, the inverse PCR products and 507 antibiotic cassettes were digested using specific restriction enzymes (Table S3). Finally, 508 the deleted gene fragments encompass the following regions in each mutant: Δhk08 (bp 509 29 to 953), Δrr08 (bp 100 to 644) and Δtcs08 (bp 128 of rr08 to bp 348 of hk08).

517
The corresponding CSP was added and incubated at 37ºC for 15 minutes, followed by 518 the addition of the plasmid for transformation and a heat shock treatment of 10 minutes 519 on ice and 30 minutes at 30ºC, bacteria were allowed to grow for 2 hours at 37ºC and 520 plated on blood agar plates with the corresponding antibiotics. The resulting S. 521 pneumoniae D39 and TIGR4 tcs08-deficient mutants were screened by colony PCR and 522 real-time PCR (qPCR) (Fig. S1B). Stocks were generated in THY supplemented with 523 20% glycerol and stored at -80°C. Individual mutants for rr08 (sp_0083) and hk08 524 (sp_0084) as well as a Δtcs08 (sp_0083+sp_0084) mutant were confirmed by colony 525 PCR. For the analysis of the gene expression by microarray, TIGR4Δcps and its isogenic 530 rr, hk and tcs08 mutants were grown in CDM until an OD600nm of 0.35-0.4 in triplicate.

531
Bacterial cultures were then added to previously prepared tubes containing frozen killing 532 buffer (20mM Tris-HCL (pH 7.5), 5mM MgCl2, 20mM NaN3) and centrifuged for 5 533 minutes at 10,000 g. The supernatant was completely removed and the tubes containing 534 the pellets were immediately flash frozen in liquid nitrogen and stored at -80°C until the 535 next step. The pellets were processed for total RNA extraction using acid phenol-536 chloroform and DNase treatment to remove genomic DNA. The products were purified 537 using the RNA Clean-Up and Concentration kit (NORGEN BIOTEK CORP), the quality 538 of the RNA was determined by Agilent 2100 Bioanalyzer and the amount was quantified 539 using a NanoDrop ND-1000 (PeqLab). 5 µg of total RNA were subjected to cDNA 540 synthesis as described by Winter et al., (2011) (60). 100 ng of Cy3-labeled cDNA were 541 hybridized to the microarray following Agilent's hybridization, washing and scanning 542 protocol (One-Color Microarray-based Gene Expression Analysis, version 6.9.1). Data 543 were extracted and processed using the Feature Extraction software (version 11.5.1.1).

544
Further data analysis was performed using the GeneSpring software (version 14.8). A

545
Student´s t-test with p < 0.05, followed by a Benjamini and Hochberg false discovery 546 rate correction with q < 0.05 were performed for the analysis. GeneMatrix UNIVERSAL RNA purification kit (ROBOKLON). The RNA was checked for 552 quality and contamination by PCR and agarose gel electrophoresis. Next, cDNA 553 synthesis was carried out using the Superscript III reverse transcriptase (Thermofisher) 554 and random Hexamer primers (BioRad). The obtained cDNA was checked by PCR 555 using the same specific primers designed for the qPCR studies (Table S3)     search for RR08 binding motifs). Figure S1 depicts the genomic organization and 667 mutagenesis strategy used in this study, as well as the confirmation of the different 668 mutants by qPCR. Figure S2 presents the competitive index obtained from the 669 coinfection assays with the TIGR4 wildtype and its isogenic tcs08-mutants. Figure S3   670 illustrates the linear localization, orientation and category of the 159 genes obtained by 671 the microarray study. Figure S4 illustrates the protein alignment of S. aureus SaeRS and 672 pneumococcal TCS08.   Specific primers for the ribosomal protein S16 (sp_0775) were used as normalization isogenic tcs08-mutants, all cultured in CDM. The unpaired student´s t-test was applied 908 for the statistics and D39Δcps or TIGR4Δcps were used as reference accordingly. ns 909 indicates "no significant", * p-value<0.05 for n = 3 and the error bars indicate the SD.    time PCR (qPCR). Specific primers were used for the rr08 and hk08. Additionally, the 947 ribosomal protein S16 (sp_0775) was used as control.