Cytosolic Recognition of Viral Infection Switches Non-Plasmacytoid Dendritic Cells Into
3
Supplementary figure legends
Supplementary Fig. 1: Expression of TLR3 is restricted to CD8a+ DC. cDNA from sorted DC subsets was amplified by semi-quantitative (a) or quantitative (b) PCR using primer sets specific for mouse TLR3 or TLR7. The amount of cDNA per sample was adjusted with respect to the b-actin signal (a) or normalised against GAPDH (b). (c) Sorted spleen CD8a+ DC and PDC were cultured overnight with the indicated concentrations of poly I:C and R848. Levels of IL-6 (top panels) and IL-12 p40 (bottom panels) were measured by ELISA.
Supplementary Fig. 2: Phagocytosis by CD8a+ DC is unaffected by poly I:C within target cells. (a) CD11c-enriched splenocytes were cultured for 3h alone or with polyI:C-loaded or mock-treated Vero cells, which had been exposed to UV-light and stained with PKH26. Association with Vero cells was analysed by flow cytometry after gating on CD11c+ CD8a+ DC. (b-c) Confocal analysis of uptake by sorted CD11c+ CD8a+ DC cocultured for 2 hours with PKH26-labelled Vero cells (red). DC were stained with APC-conjugated anti-CD11c (b, blue) or anti-CD11c-FITC (c, green). Data in (c) show 3D-image reconstruction of deconvolved confocal stacks through X-Y-Z projection (left) and semi-transparent volume rendering (right) to confirm internalisation.
Supplementary Fig. 3: Unirradiated cells loaded with poly I:C promote activation of CD8a+ DC. CD8a+ DC were cultured at a ratio of 1:5 with either syngeneic (C57BL/6) splenocytes , allogeneic (BALB/c) splenocytes or xenogeneic cells (Vero) treated as indicated. IL-6 was measured by ELISA after overnight culture.
Supplementary Fig. 4: CD8a+ DC are highly resistant to infection by EMCV or SFV. Vero cells or CD11c-enriched spleen DC were infected in parallel with EMCV or SFV-OVA at the indicated MOI. After 6-7h, cells were fixed, stained with anti-EMCV or anti-OVA and analysed by flow cytometry. Dot plots show EMCV (a) or OVA (b) staining in Vero cells or CD8a+ DC (gated on CD8a/CD11c-double positive cells). Numbers represent the percentage of cells within the indicated gate.
Supplementary Fig. 5: Quantitation of cell-associated OVA. (a) Vero cells were electroporated in the absence (left panel) or presence of OVA (middle panel) or with OVA + poly I:C (right panel), fixed, stained with anti-OVA and analysed by flow cytometry. (b, c) Vero cells were loaded with OVA + poly I:C or infected with SFV-OVA, as indicated. Mock-treated cells were used as a negative control. OVA protein was determined in cell lysates by ELISA. Data shown are the average of triplicate samples ±1SD. Results are representative of 2 experiments.
Supplementary Fig. 6: Induction of CTL cross-priming with dsRNA-bearing cells. (a) Naive C57BL/6 recipient mice were immunised with Vero cells loaded with or without OVA ± poly I:C or Vero cells infected with SFV-OVA, as indicated and injected with OVA peptide pulsed target cells on day 6 post-immunisation. Splenocytes were isolated the following day and analysed for Thy1.2+ tetramer+ CD8+ cells (top panel) and presence of target cells (bottom panel) by flow cytometry. Dot plots (upper panel) represent Thy1.2+ cells and numbers indicate the percentage of tetramer+ cells among CD8+ T cells. Histograms (lower panel) represent cells gated on CD45.1 and show CFSE profiles of sub-populations pulsed with different concentrations of OVA peptide, as indicated. The percentage of cells within each peak and calculated % killing are shown. (b) C57BL/6 mice were given OT-I T cells and then immunised with Vero cells electroporated with OVA ± poly I:C or OVA-loaded cells in combination with anti-CD40 mAb. CTL killing was assessed on day 14. Histogram shows percentages of killed target cells pulsed with either 10 nM or 100 nM OVA peptide.
Supplementary Fig. 7: Tlr3-/- chimeric mice do not show a generic defect in CTL cross-priming. Chimeric mice (3 animals per group) reconstituted with Tlr3-/- or Tlr3+/+ control bone marrow were injected with OT-I T cells and challenged the following day with OVA-loaded Vero cells in combination with CpG 1668+D19 (25µg each/mouse) and anti-CD40 mAb (50µg/mouse). 13 days later, mice received CFSE-labeled targets and were sacrificed 24h later. Absolute numbers of OT-I T cells (left panel) and CTL killing of target cells pulsed with OVA peptide (right panel) are shown for Tlr3-/- (filled circles) and WT chimeras (open circles).