Transcriptomic analysis of human brain microvascular endothelial cells exposed to laminin binding protein (adhesion lipoprotein) and Streptococcus pneumoniae

Streptococcus pneumoniae invades the CNS and triggers a strong cellular response. To date, signaling events that occur in the human brain microvascular endothelial cells (hBMECs), in response to pneumococci or its surface adhesins are not mapped comprehensively. We evaluated the response of hBMECs to the adhesion lipoprotein (a laminin binding protein—Lbp) or live pneumococci. Lbp is a surface adhesin recently identified as a potential ligand, which binds to the hBMECs. Transcriptomic analysis was performed by RNA-seq of three independent biological replicates and validated with qRT-PCR using 11 genes. In total 350 differentially expressed genes (DEGs) were identified after infection with S. pneumoniae, whereas 443 DEGs when challenged with Lbp. Total 231 DEGs were common in both treatments. Integrative functional analysis revealed participation of DEGs in cytokine, chemokine, TNF signaling pathways and phagosome formation. Moreover, Lbp induced cell senescence and breakdown, and remodeling of ECM. This is the first report which maps complete picture of cell signaling events in the hBMECs triggered against S. pneumoniae and Lbp. The data obtained here could contribute in a better understanding of the invasion of pneumococci across BBB and underscores role of Lbp adhesin in evoking the gene expression in neurovascular unit.

RNA-seq analysis. Gene expression in the challenged hBMECs (Lbp or S. pneumoniae) was evaluated by RNA-seq. Initial quality checks for RNA isolated from induced-hBMECs showed no sign of degradation (Supplementary information Fig. S2), while all cDNA libraries had optimal fragment size between 150 and 300 nt (Supplementary information Fig. S3). Total 9 cDNA libraries were generated from 3 biological replicates as follows: non-induced hBMECs (NC1 to NC3), hBMECs induced with S. pneumoniae (SP1 to SP3) or hBMECs induced with recombinant adhesion lipoprotein (Lbp-1 to Lbp-3). hBMECs exposed to recombinant Lbp yielded an average raw reads of 1.35 × 10 7 , while 1.52 × 10 7 raw reads per sample were obtained for hBMECs induced with S. pneumoniae (Supplementary information Fig. S4). In total, 11,398 genes for each sample were mapped (Supplementary dataset 1.1 and 1.2).
Differentially expressed genes (DEGSs) and validation. Data analysis revealed a total of 443 DEGs (346 up-regulated genes and 97 down-regulated) for hBMECs induced with Lbp and 350 differentially expressed genes (DEGs) (266 up-regulated genes and 84 down-regulated) for hBMECs exposed to S. pneumoniae (Fig. 1A1-A4; Supplementary datasets 1.3 and 1.4). The log 2 fold change values (LogFC), resulting from averaging three independent biological replicates for each condition, ranged between 6.25 and − 2.92 for hBMECs infected with S. pneumoniae and 7.26 to − 3.13 for hBMECs induced with Lbp (Fig. 1B). We noticed a consistent differential expression in most of the DEGs identified from both conditions. Genes expression changed in the same trend, except one gene (btc) which was up-regulated (LogFC = 2.53) in hBMECs induced with Lbp and down-regulated (LogFC = − 1.56) in S. pneumoniae infected cells (Fig. 1A4). A total of 231 DEGs were observed to be common in both treatments (189 up-regulated, 41 down-regulated and gen btc mentioned above) (Fig. 1A1,A2,A4; Supplementary dataset 1.5).
To validate the results obtained from the RNA-seq analysis, a subset of 11 representative DEGs was analyzed with qRT-PCR (Table 1). Results from qRT-PCR were consistent to those obtained from RNA-seq with a high correlation, which was evaluated by the Pearson Correlation Coefficient (r = 0.994, p < 0.05; for hBMECs exposed to S. pneumoniae and r = 0.975, p < 0.05; for hBMECs induced with Lbp, Fig. 2). Thus, confirming reliability of data derived from RNA-seq analysis. www.nature.com/scientificreports/ multiple omics-derived candidate datasets and automatically highlights the expression of genes participating in the statistically significant pathways identified by the server (  [33][34][35] . It is important to note that the pathways identifiers and pathway names assigned by each server may differ (Supplementary information Table S1).

Biological processes identified in the Lbp-induced hBMECs.
Lbp is one of the promising surface proteins of S. pneumoniae, which was previously identified as a potential ligand of hBMECs in our laboratory 13 .
In this study, we analyzed the gene expression of hBMECs exposed to this protein by RNA-seq. The functional analysis performed by Reactome server identified 14 pathways with FDR < 0.05 (Table 3). Seven out of those 14 pathways coincided with pathways identified in the independent analysis of S. pneumoniae-induced hBMECs transcriptome (Supplementary dataset 2.9) and several DEGs involved in those 7 pathways were common in both treatments (Supplementary dataset 1.9). Below we have described the signaling pathways altered exclusively in the Lbp-induced transcriptome. Incubation of Lbp with hBMECs markedly evoked expression of genes related to biological pathways such as "Interferon signaling" (R-HSA-913531) and "Interferon-γ signaling" (R-HSA-877300), which accounted for 21 DEGs and 13 DEGs, respectively (Table 3). During the bacterial infection, interferon signaling plays a central role in mounting the immune response. We identified up-regulation of 19 DEGs participating in the interferon signaling. Among these genes, we observed slight up-regulation of genes encoding for guanylate-binding protein 1 to 5 (GBP1, LogFC 1.10; GBP2, LogFC 1.61; GBP3, LogFC 1.08; GBP4, LogFC 1.30 and GBP5, LogFC 2.78), which are described as cytosolic "glue trap" capturing cytosolic Gram-negative bacteria (Supplementary dataset 1.10). Moreover, Lbp induced expression of IFI35 (LogFC 1.31) and OAS1 (LogFC 1.31), which activate macrophages to release pro-inflamatory cytokines via NFκβ. Lbp also induced expression of a group of interferon-induced IFIT proteins (IFIT1, LogFC 1.09; IFIT2, LogFC 1.02 and IFIT3, LogFC 1.09). IFIT3 acts an inhibitor of cellular processes such as cell migration, proliferation exhibiting an antiproliferative activity.
On the other side, Lbp evoked genes participating in the "Extracellular matrix organization" pathway (R-HSA-1474244, 25 DEGs), which together with two pathways containing small number of DEGs ("Type I hemidesmosome assembly" -R-HSA-446107 and "Assembly of collagen fibrils and other multimeric structures" -R-HSA-2022090) were related to maintenance of cell and tissue structures and integrities. The microvessels in brain consist of specialized microvascular endothelial cells interconnected by tight junctions surrounded by extracellular matrix (ECM) components. Thus, down-regulation of genes encoding ECM-stabilizing proteins and compounds (COL12A1, LogFC − 1.  www.nature.com/scientificreports/ Several pathological conditions such as exposure to pathogens induce cellular senescence. In this study, Lbp evoked expression of 8 genes involved in the signaling pathway "Senescence-Associated Secretory Phenotype" (R-HSA-2559582). Senescence is an irreversible proliferation arrest that was induced by Lbp, which was evidenced by the down-regulation of CDK6 (LogFC − 1.29) participating in the control of cell cycle and proliferation and by the up-regulation of genes such as histone H2 (HIST1H2AC, LogFC 1.49; HIST2H2BE, LogFC 1.38 and

Biological processes identified in the S. pneumoniae-induced hBMECs. Streptococcus pneumo-
niae has the ability to cross the BBB and generate strong inflammatory reactions that contributes substantially to the brain damage. Functional analysis of transcriptome revealed 8 pathways (statistical significance, FDR < 0.05) altered in hBMECs during infection ( Table 2; Supplementary dataset 1.6). 5 out of 8 pathways were related to signaling mediated by cytokines such as IL-4, IL-13 and IL-10. 2 out of 8 pathways were related with cell immune response and one was identified as the pathway ATF4 ( Table 2). The biological pathways "Immune system" (R-HSA-168256), "Cytokine signaling in immune system" (R-HSA-1280215) and "Signaling by interleukins" (R-HSA-449147) accounted for the largest number of participating DEGs 77, 52 and 36, respectively ( Table 2)   www.nature.com/scientificreports/ 1.11), which contribute in the internal formation of protein filaments that stabilize the cytoskeleton. Furthermore, invasion of S. pneumoniae to hBMECs evoked the over-regulation of genes implicated in cell activation, differentiation and proliferation contributing to S. pneumoniae clearance. S. pneumoniae is recognized by antigenpresenting cells via Toll-like receptors inducing the activation of factor nuclear kappa and leading the subsequent regulation of lymphocyte populations. Several DEGs implicated in these processes such as the HLA class I histocompatibility antigen (HLA-F, LogFG 1.03), nuclear factor kappa beta 2 (NFKB2, LogFG 1.32), nuclear factor kappa beta 1A (NFKBIA, LogFG 1.40), some interleukins such as IL-23A (LogFC 1.36), IL-1β (LogFC 1.42), Table 3. DEGs involved in the biological processes, categorized by Reactome server (transcriptomes induced after exposure of hBMECs to Lbp). , which is a member of the NOD-like receptors intracellular located that plays a role on innate immunity by detecting intracellular pathogen-associated patterns. We also observed a markedly overexpression of genes participating in the production of granulocytes such as granulocyte colonystimulating factor (CSF3, LogFC 3. Gene ontology of hBMECs transcriptomes induced by S. pneumoniae and Lbp. Functional analysis was investigated on the DEGs of both treatments (S. pneumoniae and Lbp) using PaintOmics server. Analysis revealed that 12.75% DEGs participated in cellular processes, 20.13% DEGs were involved in environmental information processing, 2.68% DEGs were related with genomic information processing, 6.71% DEGs were associated with human diseases, 35.58% DEGs play a role in metabolism and 22.15% DEGs were part of the organismal system (Fig. 3A). www.nature.com/scientificreports/ Biological processes (signaling pathways) activated in hBMECs after infection with S. pneumoniae and induced with Lbp were investigated. Despite seven signaling pathways were identified as common pathways in both transcriptomes (by Reactome), we could not assure that those pathways involving DEGs were statistically significant. To this regard, functional analysis seeking pathways altered by S. pneumoniae and Lbp was performed using PaintOmics. Functional analysis of biological processes revealed 12 statistically significant signaling pathways (Fisher combined p-value, p < 0.05) ( Table 4). 8 out of 12 pathways were involved in the host immune response against S. pneumoniae (Table 4

Deletion of Lbp gene altered gene expression in hBMECs in contrast to wild type S. pneumoniae.
To corroborate the role of Lbp in evoking the cell response we challenged the cells with Lbp deletion mutant (SPΔLbp) and the expression of selected genes was compared with the expression levels induced by wild type S. pneumoniae. Similarly, to rule out that the cell response to Lbp challenge, observed in this study, was not merely because of the exogenous protein originated from E.coli, we challenged the cells with enhanced green fluorescent protein (eGFP overexpressed in E.coli, a benign protein control) and the cell response was compared with the gene expression induced by Lbp. A DNA cassette (5′-351 bp upstream of Lbp-bla gene encoding beta lactamase-300 bp downstream to Lbp gene-3′) constructed to delete Lbp by homologous recombination is presented in Supplementary information Fig. S11. S. pneumoniae was transformed with the cassette and transformants were selected in the presence of carbenicillin. Replacement of Lbp by bla in SPΔLbp confirmed with sequencing is presented in Supplementary information Fig. S11.
Expression of VCAM-1, SAA2, SAA1, IL-8, CCL2 and IL1RL1 in hBMECs was significantly decreased (p < 0.01, unpaired two-tailed t test, Fig. 5) in SPΔLbp compared to wild type S. pneumoniae. While, downregulation of CPA4, CNN1, DHRS2, CDK6 and COL12A1 observed in hBMECs induced with wild type S. pneumoniae was abolished in case of SPΔLbp (Fig. 5). These results indicate that Lbp plays an important role in inducing the signalling events in hBMECs. It is noteworthy that, deletion of Lbp did not abolish the differential expression of the genes (mainly upregulated), however significant reduction in the expression was observed. This result may indicate a minor contribution of other adhesins (e.g. CbpA, RrgA, Nan, etc.) in the induction of cell signaling in comparison with Lbp, which exerts a remarkable impact.

Discussion
Streptococcus pneumoniae invades the neurovascular unit by crossing the BBB via transcellular route or by disrupting the tight junctions, which enhances the barrier instability 36,37 . Transcellular translocation of S. pneumoniae is facilitated by interaction of surface bacterial compounds with the host's cell receptors. Hitherto, several surface proteins such as CbpA, RrgA, Nan and Ply have shown to promote pneumococcal invasion of brain endothelial cells [38][39][40][41] . Recently, we also have identified pneumococcal protein ligands that interact with hBMECs, among them Lbp showed a strong interaction with endothelial cells 13 . Similarly to our previous results indicating the surface exposure of pneumococcal Lbp protein, Lbp of S. pyogenes has been described as a membrane protein and the interaction with laminin was shown to be mediated via zinc bonds 13,42 . Thus, in this study using RNA-seq technology we attempted to decipher the cell signaling events triggered in hBMECs upon adhesion of Lbp. In addition, an integrative functional analysis (DEGs induced by S. pneumoniae and Lbp) was executed for a better understanding of S. pneumoniae invasion across BBB. To confirm that the effect observed was caused by Lbp and not by contaminants originated from E.coli, we evaluated the cell response against eGFP that was produced using the same E.coli expression system. Although we observed a small alteration in the gene expression in eGFP challenged cells, the cell response caused by Lbp protein was significantly higher (Fig. 5). In addition, the significant reduction in cell response to SPΔLbp compared with wild type S. pneumoniae indicates that Lbp is one of the major surface proteins that evokes signaling events in endothelial cells of brain microvasculature.
In this study, genes evoked in both treatments (S. pneumoniae and Lbp) showed same trend of expression, thus it is tempting to speculate that this consistency might be due to the similarity of signaling events induced by S. pneumoniae and Lbp. The common signaling pathways observed in both treatments as well as pathways specifically induced either by S. pneumoniae or Lbp provided us insight into the plausible role of adhesion lipoprotein. The molecular events evoked in both conditions are discussed below. To facilitate the comprehension of the gene expression, LogFC value is referred in parenthesis for some genes.
Entry of pneumococcus to the brain is facilitated by adhesins expressed on the cell surface 18,19 . We found up-regulation of ICAM-1, VCAM-1 and NRCAM in both transcriptomes (Supplementary dataset 2.1) and slight up-regulation of PECAM-1 (SP 1.02) in the S. pneumoniae-induced cells. PECAM-1 has been previously described to interact with the lectin-like domain of the pneumococcal neuraminidase A (NanA) contributing to the attachment of bacteria to the BBB 39,43 . In the receptor-mediated translocation, choline-binding protein (CbpA) recognizes the pIgR molecule facilitating entry of S. pneumoniae to the neurovascular unit 44 . We observed induced expression of several cell surface receptors in both transcriptomes (e.g. EFNA-1: SP 1.12, Lbp 1.37; FLT-1: SP 1.09, Lbp 1.51) participating in the PI3K-AKT signaling pathway, which is activated after cellular stimuli.
It is well known that pneumococcus experiences autolysis releasing bacterial components to the extracellular milieu, which in turn activates and enhances host immune response 45 . We found that several signaling pathways (e.g. Cytokine Signaling in Immune system, Signaling by Interleukins, Chemokine receptors bind chemokines, Interferon alpha/beta signaling, etc.) involved in cell immune response were induced independently in both hBMECs transcriptomes. Moreover, the integrative analysis showed participation of DEGs in common pathways (e.g. Cytokine-cytokine receptor interaction, Chemokine signaling pathway, TNF signaling pathway, NF-kappa B signaling pathway, etc.). S. pneumoniae is recognized by antigen-presenting cells via pattern recognition receptors. Immune activation occurs through toll-like receptor-2 (TLR-2) by recognition of pathogen-associated molecular patterns 46 . From the integrative analysis, we observed that TLR-2 was evoked in both transcriptomes. TLR-2-dependent pathway is enough to cause inflammation by stimulating the production of pro-inflammatory cytokines and chemokines 46,47 . In this study, both treatments induced the over expression of several cytokines such as IL-24, IL-1A and IL-23A and some chemokines (e.g. CXCL1, CXCL2, CXCL3, CXCL5, CXCL6) to activate white blood cells and to promote leukocyte chemotaxis for inducing pleocytosis. Early immune response is associated www.nature.com/scientificreports/ mentioned) trigger the leukocyte activation and recruitment, participate as cell adhesion molecules (ICAM-1, VCAM-1 and SELE) and involve in the intracellular negative signaling (BCL3, SOCS3 and TNFAIP3) regulating the cytokine signaling. Additionally, NOD-like receptor 2 (playing a role in this pathway) was evoked in both treatments. NOD-like receptors are the molecules which detect pathogen-associated molecular patterns 53 . Specifically, NOD-2 has been described to sense muramyl peptides from pneumococcal peptidoglycan to enhance innate immune response 54 . Moreover, NOD-2 is described to activate a NF-κβ-dependent pro-inflammatory gene expression 55 . In this study, the NF-κβ signaling pathway was activated involving up-regulation of BIRC3 for activation of NFKBI and further stimulation of TRAF1 and NFKB2 to induce and regulate the signal transduction during inflammation. Activation of NOD2 is also described to occur after delivery of pneumococcal peptidoglycan fragments produced during the lysozyme-dependent digestion of S. pneumoniae in macrophages (phagocytosis) 56 . Additionally, we identified an increased expression of integrins (e.g. ICAM-1) facilitating the adhesion of macrophages to the microvascular epithelium to contribute with the pneumococcal clearance. The integrative analysis, performed with the total of DEGs identified in the S. pneumoniae-induced and Lbp-induced transcriptomes, revealed the activation of the phagosome pathway. Up-regulation of C1R, C1S and C3 indicate activity of the complement cascade and up-regulation of Cathepsin S (CTSS) indicates active degradation of antigenic proteins to peptides for presentation to the MHC class II for further lymphocyte stimulation. Overall, S. pneumoniae and Lbp activated biological processes that contribute to the bacterial clearance by classic innate and adaptive immune responses. It is noteworthy that, Lbp evoked Interferon signaling by up-regulating expression of interferon-induced guanylate-binding proteins encoding by GBP1-5, which have been described to act as a cytosolic "glue trap" capturing the bacteria 57 . Additionally, Lbp activated the non-canonical NF-κβ signaling pathway for enhancing the lymphocyte adhesion to the microvascular epithelium via ICAM-1. It has been reported that some pathogens activate this pathway 58 . Furthermore, we observed that Lbp evoked biological processes involved in the cell structure organization such as the extracellular matrix organization, Type-I hemidesmosome assembly, and assembly of collagen fibrils and other multimeric structures. Although, these three processes were not statistically significant in S. pneumoniae treatment, several genes participating in cell structure maintenance such as FLNA, FLNB, BST2 were down regulated. In the following lines, these three processes are discussed together for practical purposes.
Extracellular matrix (ECM) is formed by structural components providing stability to endothelial cells and is involved in tissue homeostasis preventing bacterial invasion 59 . We identified over expression of genes stimulating production of laminins (LAMB3) and collagen (COL7A1, COL25A1). Additionally, DEGs encoding proteins regulating tight junction assembly (CD321), stimulating collagen production (ADAMTS3) as well as promoting adhesion to the endothelial cells (FBLN5) were up-regulated suggesting an active remodeling of ECM. Proteolytic cleavage of the ECM is required for the bacterial invasion into host tissues. Pathogenic bacteria recruit host proteases or inhibit expression of genes encoding structural compounds 60 . Lbp caused down-regulation of genes encoding collagen type XII (COL12A1 − 1.80) and integrins (ITGB4 − 1.31) as well as essential genes for the biogenesis of connective tissue (LOXL1). Additionally, Lbp evoked expression of the collagenase MMP1 (1.78), which is involved in the breakdown of ECM, which in turn may contribute to invasion of S. pneumoniae to the neurovascular unit. Furthermore, Lbp had effect on the type I hemidesmosome assembly and the assembly of collagen fibrils pathways. Type I hemidesmosomes are specialized junctional complexes connecting cytoskeleton of the cells to the ECM, while assembly of collagen fibrils refers the molecular mechanisms involved in the formation of higher collagen structures within the extracellular space 61,62 . Thus, these pathways play a role in the maintenance of tissue structure and integrity preserving the mechanical and physical properties of tissues. Lbp evoked expression of genes encoding laminin and laminin extracellular glycoproteins (LAMB3 and LAMC2) and caused down-regulation of gene encoding integrin (ITGB4) and plectin (PLEC), which interacts with molecules from the cell cytoskeleton providing support and strength to the cell. Overall, these findings suggest increased ECM instability facilitating permeability of BBB favoring the streptococcal invasion.
Laminin-binding proteins of Streptococcus share high homology to AdcAII. AdcAII is a zinc-binding lipoprotein that is important for infectivity of S. pneumoniae. Expression of AdcAII depends on the concentration of Zn 2+ . Under lower concentrations of Zn 2+ the growth and colonization of S. pneumoniae seems to be affected. Interaction between laminin and AdcAII has been investigated previously by Brown et al. but no evidence of such interaction was found using dot blot experimets 63 . Previously, we have reported strong affinity between laminin binding protein (Spr0906) and BMECs 13 . In this study we have investigated the biological process altered in BMECs by Lbp but no evidence in the Zn-uptake was identified.
Finally, we identified induction of cell senescence by treatment of hBMECs with Lbp. Senescence is an irreversible proliferation arrest in which cell experiences changes in function, morphology and gene expression 64 . Licastro and Porcellini have suggested a close relation between CNS infections and senescence 65 . Lbp induced over expression of genes such as HIST1H2AC (responsible for DNA packaging) and CEBPB (stimulating transcription of CDKN2B), which reinforce the cell cycle arrest and down-regulation of CDK6 (cell division protein kinase 6, which is essential for cell cycle progression), respectively.

Conclusion
Streptococcus pneumoniae is an extracellular pathogen capable to invade the central nervous system and cause meningitis. Surface proteins of S. pneumoniae play a critical role in neuroinvasion. We recently reported interaction between Lbp with hBMECs. In this study, we comprehensive picture of the hBMECs response against Lbp as well as intact bacteria by using a high-throughput RNA-seq technology. Both Lbp and S. pneumoniae seem to induce host immune response with activation of cytokines and chemokines. Based on the signaling pathways identified from the bioinformatics analysis, Lbp seem to induce signaling pathways involved in the breakdown www.nature.com/scientificreports/ and remodeling of ECM as well as cell senescence. This investigation revealed overall cell signaling events occurred in the microvascular endothelium upon adhesion S. pneumoniae to the cell surface. Furthermore, it provides plausible importance of the Lbp adhesin of S. pneumoniae in evoking the signaling events in the endothelium of the brain microvascular, which may promote pneumococcal invasion to the brain.

Materials and methods
Culture of human brain microvascular endothelial cells. hBMECs (D3 cell line), were obtained from Merck/Millipore (Prague, Czech Republic). Details are in the supplementary information Method S1. Cells were either harvested for RNA isolation (non-induced cell control) or incubated with live S. pneumoniae or Lbp.
Bacterial strain. The neuroinvasive strain of S. pneumoniae (clinical isolate SPH) used in this study was isolated from the cerebrospinal fluid of a meningitis-suffering patient hospitalized in Louis Pasteur Hospital, Kosice, Slovakia. Cerebrospinal fluid was received in microbiological diagnostic laboratory to identify the causative agent of the meningitis. Use of the isolate SPH for experimental purpose was in accordance with the guidelines and regulations set by the ethical committee of the Hospital and diagnostic laboratory of the hospital. The isolate SPH was characterized by phenotyping (biochemical tests) and genotyping (sequencing of lytA and rpoB genes) in the hospital laboratory. S. pneumoniae was plated on Columbia agar blood base containing 5% (v/v) sheep blood and single isolated colony was grown as described in our previous publication 13 Table S3). Protein expression and purification were performed as described earlier 13 (details are in supplementary information Methods S2 and S3). Purity of recombinant Lbp was evaluated by SDS-PAGE and MALDI-TOF MS (details are presented in supplementary information Method S4). Protein concentration was measured by Bradford method and aliquots of purified proteins were stored at − 20 °C in 20% glycerol until use.

Challenge of hBMECs.
Monolayer of hBMECs was cultured on 6-well plates and incubated either with live S. pneumoniae cells (MOI 0.5/well) or with recombinant Lbp (approximately 1 nMol, 27 μg/well) or with culture medium (non-induced control) for 6 h at 37 °C under 5% CO 2 atmosphere. After incubation, culture media was removed, hBMECs were washed and cells were scrapped for RNA isolation.
RNA isolation from hBMECs. mRNA from hBMECs was isolated using RNeasy Mini Kit (Qiagen, Germany) according to manufacturer's instructions. During RNA isolation, samples were treated with DNAseI (Qiagen). RNA samples were eluted in RNase free water and quantified by nanodrop (Thermo Scientific). Integrity of RNA was monitored in 1% borax gel electrophoresis at 100 V and by capillary electrophoresis (Fragment analyzer, Advanced Analytical Technologies, Inc., USA). Samples were stored at − 80 °C until used for library preparation.
Preparation of RNA libraries. RNA libraries were prepared exactly as described before in our publication 66 .
In brief, 250 ng of RNA were reverse transcribed with oligodT primers for synthesis of the first strand cDNA using QuantSeq 3′ mRNA-Seq Library Prep Kit (Lexogen, Austria) as per manufacturer's instructions. RNA template was removed and second strand was synthetized by using random hexamer containing Illumina-compatible linker sequences at its 5′ end. Double strand DNA library was purified using magnetic beads provided in the kit. Each library was amplified by PCR using unique single indexing i7 primers to add complete adapter sequence required for cluster generation and to generate sufficient DNA for sequencing and quality control. The number of cycles in PCR for each library was determined using PCR Add-on kit for Illumina (Lexogen). Number of cycles used for library amplification were as follows: hBMECs induced with S. pneumoniae-20 cycles, hBMECs induced with Lbp-20 cycles and non-induced cells-17 cycles. Amplified libraries were purified using magnetic beads supplied in the kit. Quality of the library and length of the fragments were checked on fragment analyzer.
A set of 11 DEGs significantly up and down regulated in RNAseq were selected for validation. Primers used in qRT-PCR (Table 1) were designed using Geneious Pro software (Biomatters, USA). Reaction mix of qRT-PCR was composed of 6 ng of cDNA, 1 × qPCR GreenMaster with highROX (Jena Bioscience, Germany), gene specific primers (10 pMol each) and RNase free water up to total volume 20 μL. Each DEG was evaluated in triplicates. Amplification cycle was as follows: 95 °C-10 min, 40 × [95 °C-15 s, 50-60 °C-30 s (annealing temperature varied according to the primers used), 72 °C for 30 s. (signal capture)], melting curve 60 °C to 95 °C-0.3% temperature increment/s (StepOnePlus, Thermo Fisher Scientific, USA). The gene expression (ΔΔCt) was normalized to β-2-microtubulin (house-keeping gene) as described before 67 . ΔΔCt values were converted to logFC (http:// www. endme mo. com/ algeb ra/ log2. php). Correlation of expression values for DEGs obtained from RNA-seq and qRT-PCR was determined by calculating the Pearson correlation coefficient (r) using Graphpad Prism version 8. Comparative analysis of gene expression in hBMECs: SPΔLbp vs. wild type S. pneumoniae and GFP vs. Lbp. hBMECs were incubated either with SPΔLbp or wild type (both MOI 0.5/well) or eGFP or Lbp (both 1 nMol/well) for 6 h. Non-induced cells were kept as a negative control. RNA was isolated from induced and non-induced cells, reverse transcribed and used in qRT-PCR as described above. Please note that, eGFP was overexpressed and purified in the same way as Lbp. The logFC was calculated as described above. Unpaired t test (https:// www. graph pad. com/ quick calcs/ ttest1) was used to assess statistical difference between logFC observed in wild type vs. SPΔLbp or GFP vs. Lbp. The experiment was performed in biological triplicates.

Data availability
The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request. www.nature.com/scientificreports/