Pneumococal Surface Protein A (PspA) Regulates Programmed Death Ligand 1 Expression on Dendritic Cells in a Toll-Like Receptor 2 and Calcium Dependent Manner

Abstract

Pneumonia leads to high mortality in children under the age of five years worldwide, resulting in close to 20 percent of all deaths in this age group. Therefore, investigations into host-pathogen interactions during Streptococcus pneumoniae infection are key in devising strategies towards the development of better vaccines and drugs. To that end, in this study we investigated the role of S. pneumoniae and its surface antigen Pneumococcal surface protein A (PspA) in modulating the expression of co-stimulatory molecule Programmed Death Ligand 1 (PD-L1) expression on dendritic cells (DCs) and the subsequent effects of increased PD-L1 on key defence responses. Our data indicate that stimulation of DCs with PspA increases the surface expression of PD-L1 in a time and dose dependent manner. Characterization of mechanisms involved in PspA induced expression of PD-L1 indicate the involvement of Toll-Like Receptor 2 (TLR2) and calcium homeostasis. While calcium release from intracellular stores positively regulated PD-L1 expression, calcium influx from external milieu negatively regulated PD-L1 expression. Increase in PD-L1 expression, when costimulated with PspA and through TLR2 was higher than when stimulated with PspA or through TLR2. Further, knockdown of TLR2 and the intermediates in the TLR signaling machinery pointed towards the involvement of a MyD88 dependent pathway in PspA induced PD-L1 expression. Incubation of DCs with S. pneumoniae resulted in the up-regulation of PD-L1 expression, while infection with a strain lacking surface PspA failed to do so. Our data also suggests the role of PspA in ROS generation. These results suggest a novel and specific role for PspA in modulating immune responses against S. pneumoniae by regulating PD-L1 expression.

Fig 1. PspA upregulates expression of PD-L1 on DCs.

A, mouse bone marrow derived DCs were stimulated with indicated concentrations of PspA for 24h. B, DCs were stimulated with 15μg/ml PspA for indicated times. PD-L1 expression was monitored by flow cytometry. Bold lines represent expression levels in the presence of PspA while thin lines represent unstimulated controls. Data from one of three independent experiments are shown. The MFI values from three independent experiment (expressed as mean ± SD) are summarized as a bar graph in the right panel.

Fig 2. TLR2 and PspA synergistically stimulate PD-L1 up regulation.

For Panel A, mouse bone marrow derived DCs were stimulated with either 15 μg/ml PspA, 1 μg/ml Pam₃Csk₄, 0.2 μg/ml LPS or 1 μg/ml Immiquimod for 24h and surface expression of PD-L1 was monitored by flow cytometry. Dotted lines represent unstimulated controls while thin lines represent stimulations with PspA. Bold lines represent stimulations with indicated agonists to different TLRs. Panel B, DCs were stimulated with 15 μg/ml PspA in the presence or absence of ligands to different TLRs for 24h and PD-L1 expression was monitored by flow cytometry. Thin lines represent stimulations with PspA alone. Bold lines represent costimulations with PspA and indicated agonists to different TLRs. Data from one of three independent experiments are shown. In Panel C, PD-L1 expression is represented as bars indicating fold change in Relative Mean Fluorescence Intensity (MFI) for various groups. Bars represent mean ± SD of three independent experiments. ‘ns’ represents non-significant differences between compared groups.

Fig 3. PspA induces PD-L1 expression in a MyD88 dependent pathway.

Mouse bone marrow derived DCs were transfected with siRNAs against indicated molecules for 36h followed by stimulations with 1 μg/ml Pam₃Csk₄ and 15 μg/ml PspA for 24h. PD-L1 levels were monitored by flow cytometry. Dotted line represents unstimulated cells transfected with control siRNAs. Thin lines represent cells transfected with control siRNAs followed by stimulations with 1 μg/ml Pam₃Csk₄ and 15 μg/ml PspA. Bold lines represent cells transfected with specific siRNAs to indicated molecules followed by stimulations with 1 μg/ml Pam₃Csk₄ and 15 μg/ml PspA. Data from one of three independent experiments are shown. In Panel B, PD-L1 expression is represented as bars indicating fold increase in Relative Mean Fluorescence Intensity (MFI) for various groups. Bars represent mean ± S.D. of three independent experiments.

Fig 4. Route of calcium entry differentially regulates PspA induced PD-L1 expression.

Mouse bone marrow derived DCs were incubated with bio-pharmacological inhibitors to indicated molecules for 1h followed by stimulation with 15 μg/ml PspA for 24h. PD-L1 levels were monitored by flow cytometry. Dotted lines represent unstimulated cells. Thin lines represent cells stimulated with PspA. Bold lines represent cells treated with inhibitors to indicated molecules followed by stimulation with PspA. One of three independent experiments is shown. PD-L1 expression is represented as bar graph indicating fold increase in Relative Mean Fluorescence Intensity (MFI) for various groups. Bars represent mean ± SD of three independent experiments. For Panel B, mouse bone marrow derived DCs were incubated in the presence or absence of TMB-8 or EGTA for 1h followed by stimulations with 15 μg/ml PspA for 24h. Cells were incubated with phycoerythrin conjugated anti-mouse PD-L1 antibody. Merged images with DAPI (blue) and PD-L1 (red) staining are depicted. For Panel C total RNA was isolated from cells stimulated as indicated and PD-L1 transcript levels were measured by semi-quantitative RT-PCR. One of three independent experiments is shown.

Fig 5. Molecular sensors of intracellular calcium regulate PspA induced PD-L1 expression.

For Panel A, mouse bone marrow derived DCs were transfected with siRNAs against indicated molecules for 36h, followed by stimulation with 15 μg/ml PspA for 24h. PD-L1 levels were monitored by flow cytometry. Dotted lines represent unstimulated cells transfected with control siRNAs. Thin lines represent cells transfected with control siRNAs followed by stimulations with 15 μg/ml PspA. Bold lines represent cells transfected with siRNAs specific to the indicated molecules followed by stimulation with 15 μg/ml PspA. Data from one of three independent experiments is shown. In Panel B, PD-L1 expression is represented as bars indicating fold increase in Relative Mean Fluorescence Intensity (MFI) for various groups. In panel B, bars represent mean ± SD of three independent experiments. ‘ns’ represents non-significant differences between compared groups.

Fig 6. S. pneumoniae upregulates PD-L1 on DCs.

For Panel A, mouse bone marrow derived DCs were infected with wild type S. pneumoniae strain D39, R6 or JY2008 (see Materials and Methods) and PD-L1 levels were monitored by flow cytometry. Thin lines represent uninfected cells while bold lines represent cells incubated with indicated strain of S. pneumoniae. For Panel B, DCs were incubated with either TMB-8 or EGTA for 1h followed by incubation with D39 and PD-L1 levels were monitored by flow cytometry. Dotted line represents uninfected cells. Thin lines represent cells incubated with D39 while thick lines represent cells treated with indicated inhibitors followed by incubation with D39. Data from one of two independent experiments are shown. Bars in Panel C and D represent fold increase in Relative Mean Fluorescence Intensity (MFI; mean ± SD) for various groups. ns represents non-significant differences between compared groups.

Fig 7. PspA regulates apoptosis of DCs through PD-L1.

Mouse bone marrow derived DCs were transfected with siRNAs against PD-L1 for 36h, followed by stimulation with 15 μg/ml PspA for 24h. Cytoplasmic extracts were immunoblotted for indicated molecules. Numbers below the bands represent density of the band relative to unstimulated control. Data from one of two independent experiments are shown.