Mycobacteria-responsive sonic hedgehog signaling mediates programmed death-ligand 1- and prostaglandin E2-induced regulatory T cell expansion

Abstract

CD4(+)CD25(+)FoxP3(+) regulatory T cells (Tregs) are exploited by mycobacteria to subvert the protective host immune responses. The Treg expansion in the periphery requires signaling by professional antigen presenting cells and in particularly dendritic cells (DC). However, precise molecular mechanisms by which mycobacteria instruct Treg expansion via DCs are not established. Here we demonstrate that mycobacteria-responsive sonic hedgehog (SHH) signaling in human DCs leads to programmed death ligand-1 (PD-L1) expression and cyclooxygenase (COX)-2-catalyzed prostaglandin E2 (PGE2) that orchestrate mycobacterial infection-induced expansion of Tregs. While SHH-responsive transcription factor GLI1 directly arbitrated COX-2 transcription, specific microRNAs, miR-324-5p and miR-338-5p, which target PD-L1 were downregulated by SHH signaling. Further, counter-regulatory roles of SHH and NOTCH1 signaling during mycobacterial-infection of human DCs was also evident. Together, our results establish that Mycobacterium directs a fine-balance of host signaling pathways and molecular regulators in human DCs to expand Tregs that favour immune evasion of the pathogen.

Figure 1. Opposing role of SHH and NOTCH1 signaling during M. bovis BCG-induced Treg expansion.

(a,b) Immature DCs were cultured in GM-CSF and IL-4 alone (Med) or along with pharmacological inhibitors of SHH signaling pathway like Cyclopamine (SMO inhibitor), WNT signaling pathway like IWP-2 (a WNT secretion inhibitor) or NOTCH signaling like γ-secretase inhibitor (GSI) for 1 h followed by infection with M. bovis BCG (MOI 1:10) for 24 h. After extensive wash, DCs were co-cultured with autologous CD4⁺ T cells. CD4⁺CD25⁺FoxP3⁺ Tregs were analyzed by flow cytometry. (a) Representative dot blot of 6 independent experiments is shown. (b) Percentage of CD4⁺CD25⁺FoxP3⁺ cells in the DC-CD4⁺ T cell co-cultures (mean ± SEM, n = 8). (c) In a similar set up as panels (a,b) percentage of IFN-γ⁺CD4⁺ cells in the DC-CD4⁺ T cell co-cultures (mean ± SEM, n = 7). (d) T cell cytokines IL-2, TNF-α and IFN-γ were analyzed in the cell-free culture supernatants of DC:T cell co-culture (mean ± SEM, n = 6–7) by cytokine bead array. Med, Medium. *P < 0.05; **P < 0.005; ***P < 0.001 (one-way ANOVA followed by Turkey’s multiple-comparisons test).

Figure 2. M. bovis BCG-induced proinflammatory responses are moderately NOTCH dependent and suppressed by SHH.

DCs were cultured with GM-CSF and IL-4 and left untreated (Med) or infected with M. bovis BCG alone or after 1 h pretreatment with the indicated inhibitors for 24 h. (mean ± SEM, n = 6–10) (a) Surface expression of maturation markers CD83, HLA-DR, CD40, CD86 and CD80 were examined by flow cytometry. Data is represented as % positive cells or MFI. (b) Cell-free supernatants from the above-said experiment were assessed for secretion of IL-6, TNF-α, IL-12p70 by cytokine bead array (mean ± SEM, n = 5). Med, Medium. *P < 0.05; **P < 0.005; ***P < 0.001 (one-way ANOVA followed by Turkey’s multiple-comparisons test).

Figure 3. PI3K-mTOR-NF-κB pathway mediates M. bovis BCG-induced SHH signaling and Treg expansion.

(a,b) Five-day-old differentiated immature DCs were infected with 1:10 MOI of M. bovis BCG or M. tuberculosis H37Ra for 6 h. Transcript (a) and protein (b) levels of various SHH signaling markers were determined using quantitative real time RT-PCR and immunoblotting respectively. (c,d) Expression analysis of SHH signaling markers in immature DCs pretreated with the indicated pharmacological inhibitors for 1 h prior to 6 h infection with M. bovis BCG. (e) DCs infected with M. bovis BCG alone or after pretreatment of LY294002 (PI3K inhibitor), Rapamycin (mTOR inhibitor) or BAY 11-7085 (NF-κB inhibitor) were co-cultured with autologous CD4⁺ T cells. Percentage of CD4⁺CD25⁺FoxP3⁺ T cells (mean ± SEM, n = 4) were analyzed by flow cytometry. All RT-PCR data represents the mean ± SEM from at least 3 independent experiments and all blots are representative of 3 independent experiments. Images have been cropped for presentation; full-size blot is shown in Supplementary Fig. S1. Med, Medium. *P < 0.05; **P < 0.005; ***P < 0.001 (one-way ANOVA followed by Turkey’s multiple-comparisons test).

Figure 4. M. bovis BCG-induced PD-L1 and COX-2 are SHH signaling dependent.

(a,b) Quantitative real time RT-PCR for expression analysis of PD-L1, PD-L2, PTGS2/COX-2 and ELISA from cell-free supernatants to estimate the secretion of PGE₂ on (a) infection of immature DCs with M. bovis BCG or M. tuberculosis H37Ra for 12 h or (b) infection of indicated pharmacological inhibitor treated immature DCs with M. bovis BCG for 12 h. (c) DCs were cultured in GM-CSF and IL-4 alone (Med) or with pharmacological inhibitor followed by M. bovis BCG as indicated for 24 h. Surface expression of PD-L1 as % positive cells was analyzed by flow cytometry (mean ± SEM, n = 8). (d,e) DCs were treated as explained above. Immunoblotting for COX-2 from total cell lysate (d) and ELISA for measuring PGE₂ in the cell-free supernatant (e). (f–i) Immature human DCs were transiently transfected with NT or SHH siRNA. 48 h post transfection, cells were infected with M. bovis BCG for 24 h to assess the surface expression of PD-L1 by flow cytometry (mean ± SEM, n = 4) (f) or 12 h to estimate COX-2 protein by immunoblotting (g) and PGE₂ in the cell-free supernatant by ELISA (h). Validation of SHH siRNA was performed by immunoblotting for SHH in the siRNA-transfected DCs (i). All RT-PCR and ELISA data represents the mean ± SEM from at least 3 independent experiments and all blots are representative of 3 independent experiments. Images have been cropped for presentation; full-size blot is shown in Supplementary Fig. S1. Med, Medium; NT, Non-targeting. *P < 0.05; **P < 0.005; ***P < 0.001 (one-way ANOVA followed by Turkey’s multiple-comparisons test).

Figure 5. PD-L1 and COX-2 mediate M. bovis BCG-induced Treg expansion.

(a) Inhibition of either PD-L1 or COX-2, partially inhibit the BCG-induced Treg expansion. γ-irradiated BCG-stimulated DCs were incubated with anti-PD-L1 blocking antibody or isotype antibody (upper panel). Alternatively, DCs were pretreated with DMSO or NS-398 (COX-2 inhibitor) before stimulation with BCG (lower panel). After extensive wash, DCs were co-cultured with autologous CD4⁺ T cells for five days. Representative dot-plots showing frequency of CD4⁺CD25⁺FoxP3⁺ Tregs were presented. The Treg response induced by DCs cultured in medium alone is represented by ‘Med’. (b,c) Inhibition of both PD-L1 and COX-2 in DCs significantly suppress the ability of BCG to expand Tregs. DCs were pretreated with DMSO or NS-398 before stimulation with BCG. After extensive washing, these DCs were incubated with anti-PD-L1 blocking antibody or isotype antibody and co-cultured with autologous CD4⁺ T cells for five days. CD4⁺CD25⁺FoxP3⁺ Tregs were analyzed by flow cytometry. Representative dot-plots showing frequency of CD4⁺CD25⁺FoxP3⁺ Tregs were presented (b). (c) Percentage of CD4⁺CD25⁺FoxP3⁺ cells in the DC-CD4⁺ T cell co-cultures (mean ± SEM, n = 5). *P < 0.05 (one-way ANOVA followed by Holm-Sidak’s multiple comparisons test).

Figure 6. Bi-functional role of SHH signaling to regulate PD-L1 and COX-2 expression.

(a) The recruitment of GLI1 at human PTGS2 and PD-L1 promoter upon infection with M. bovis BCG for 12 h in immature DCs was evaluated by ChIP assay. (b,c) DCs were infected with M. bovis BCG or M. tuberculosis H37Ra alone (b) or with pharmacological inhibitors of SHH pathway (c) and quantitative real time RT-PCR analysis was performed on total RNA isolated using indicated miRNA-specific primers. (d) Putative miR-324-5p and miR-338-5p binding sites in the 3′UTR of PD-L1. (e) THP-1 cells were transfected with WT PD-L1 3′UTR or miR-324-5pΔ miR-338-5pΔ PD-L1 3′UTR constructs with miR-324-5p or miR-338-5p mimics as indicated. Transfected THP-1 cells were further stimulated with M. bovis BCG as indicated and luciferase assay was performed. (f,g) DCs transfected with control or miR-324-5p and miR-338-5p mimics were infected with M. bovis BCG for 24 h (f) or 18 h (g) as indicated. While surface expression of PD-L1 was evaluated by flow cytometry (mean ± SEM, n = 3) (f), protein levels of PD-L1 in the total cell lysate was assessed by immunoblotting (g). All RT-PCR and luciferase data represents the mean ± SEM from 3 independent experiments. Images have been cropped for presentation; full-size blot is shown in Supplementary Fig. S1. Med, Medium; NC, Negative control. *P < 0.05; **P < 0.005; ***P < 0.001 (Student’s t-test for Panel a, one-way ANOVA followed by Turkey’s multiple-comparisons test for others).

Figure 7. M. bovis BCG-induced NOTCH1 signaling in DCs regulates PI3K-mTOR-NF-κB pathway.

(a) Quantitative real time RT-PCR for assessing NOTCH1 signaling markers, HES1, NOTCH1, JAG1 and JAG2 on infection of immature DCs with M. bovis BCG or M. tuberculosis H37Ra for 6 h. (b) Activation of NOTCH1 signaling was determined by immunoblotting for NICD in DCs infected with M. bovis BCG or M. tuberculosis H37Ra. (c,d) Transcript (c) and intracellular domain (d) of NOTCH2-4 on M. bovis BCG stimulation was analyzed by quantitative real time RT-PCR (c) or immunoblotting (d). (e) NOTCH1 signaling activation was assessed by immunoblotting for NICD in DCs pretreated with the indicated pharmacological inhibitors and infected with M. bovis BCG. (f) Immunoblotting for evaluating PI3K-mTOR-NF-κB pathway activation using DCs infected with M. bovis BCG with or without GSI (NOTCH signaling inhibitor). (g) DCs were cultured with GM-CSF and IL-4 and left untreated (Med) or infected with M. bovis BCG alone or after 1 h pretreatment with the indicated inhibitors for 24 h. Surface expression of maturation markers CD83, HLA-DR, CD40, CD86 and CD80 were examined by flow cytometry (mean ± SEM, n = 7–10). Data is represented as % positive cells or MFI ((g), top panels). Cell-free supernatants from the above-said experiment were assessed for secretion of IL-6, TNF-α, IL-12p70 by cytokine bead array ((g), lower panels) (mean ± SEM, n = 5–7). All RT-PCR data represents the mean ± SEM from at least 3 independent experiments and all blots are representative of 3 independent experiments. Images have been cropped for presentation; full-size blot is shown in Supplementary Fig. S2. Med, Medium. *P < 0.05; **P < 0.005; ***P < 0.001 (one-way ANOVA followed by Turkey’s multiple-comparisons test).

Figure 8. Counter-regulation of NOTCH1 and SHH pathway functions during M. bovis BCG infection of DCs.

(a,b) Transcript (a) and surface (b) expression of PD-L1 in DCs infected with M. bovis BCG alone or after 1 h pretreatment of GSI by quantitative real time RT-PCR and flow cytometry (mean ± SEM, n = 7) respectively. (c–e) Immature DCs were pretreated with SHH signaling inhibitors and infected with M. bovis BCG for 6 h. NOTCH1 signaling markers, NOTCH1, JAG1 and JAG2 transcripts were analyzed using quantitative real time RT-PCR (c), NICD by immunoblotting (d) and densitometric analysis of panel D (e). (f) Model: schematic representation of the obtained results. All RT-PCR and densitometry data represents the mean ± SEM from at least 3 independent experiments and all blots are representative of 3 independent experiments. Images have been cropped for presentation; full-size blot is shown in Supplementary Fig. S2. Med, Medium. *P < 0.05; **P < 0.005; ***P < 0.001 (one-way ANOVA followed by Turkey’s multiple-comparisons test).