Deletion of Wiskott-Aldrich syndrome protein triggers Rac2 activity and increased cross-presentation by dendritic cells

Abstract

Wiskott-Aldrich syndrome (WAS) is caused by loss-of-function mutations in the WASp gene. Decreased cellular responses in WASp-deficient cells have been interpreted to mean that WASp directly regulates these responses in WASp-sufficient cells. Here, we identify an exception to this concept and show that WASp-deficient dendritic cells have increased activation of Rac2 that support cross-presentation to CD8(+) T cells. Using two different skin pathology models, WASp-deficient mice show an accumulation of dendritic cells in the skin and increased expansion of IFNγ-producing CD8(+) T cells in the draining lymph node and spleen. Specific deletion of WASp in dendritic cells leads to marked expansion of CD8(+) T cells at the expense of CD4(+) T cells. WASp-deficient dendritic cells induce increased cross-presentation to CD8(+) T cells by activating Rac2 that maintains a near neutral pH of phagosomes. Our data reveals an intricate balance between activation of WASp and Rac2 signalling pathways in dendritic cells.

Figure 1. Der p 2 induces skin pathology in WASp KO mice.

(a) Whole-skin sections (10 μm) from 4 mm² punch biopsies of back skin from day 50 were stained with hematoxylin and eosin. Epidermal thickening is indicated in μm. (b) Langerhans cells in epidermis by histology. Epidermal sheets labelled with Langerin (CD207, green) for identification of Langerhans cells. The mean number of Langerhans cells per mm² of epidermis at day 50 is indicated. (c,d) DCs and T cells in dermis by histology. (c) The mean number of total Langerin⁺ (including Langerhans and dermal DCs, green), CD11c⁺EpCAM⁺ (mature Langerhans cells, CD11c in red and EpCAM in blue) DCs and (d) CD4⁺CD3⁺ (red and green, respectively) and CD8β⁺ (blue) T cells per mm² is indicated. Examples of counted cells are magnified in the white boxes. (a–d) Bar represents mean value and each dot represents one mouse (a,b) or one picture (c,d). Results are a pool of two separate experiments. (a–d) WT unchallenged n=3–4; WKO unchallenged n=3–6; WT Der p 2 n=3–9; WKO Der p 2 n=4–10. Scale bar, 50 μm. (e) DCs and T cells in skin by flow cytometry analysis. Absolute numbers of cells in the back skin at day 50 from unchallenged and Der p 2-challenged WT and WASp KO mice as measured by flow cytometry. WT unchallenged n=4; WKO unchallenged n=5; WT Der p 2 n=6; WASP KO Der p 2 n=8. (f) DC egress from the skin. Imiquimod was applied on the ear and 48 h later, MHCII^highDEC205⁺CD8⁻ DCs and CCR7⁺DEC205⁺CD8⁻ DCs were analysed by flow cytometry in dLNs. WT unchallenged n=4; WKO unchallenged n=4; WT imiquimod n=4; WASP KO Imiquimod n=4. (a–f) Results are representative of two separate experiments. *P<0.05; **P<0.01 as calculated by the unpaired Student's t-test. LC, Langerhans cells; WT, wild type; WKO, WASp KO.

Figure 2. Der p 2 induces expansion of WASp KO CD8⁺IFNγ⁺ T cells.

(a,b) LN and spleen T cells by flow cytometry. Absolute numbers of (a) total and (b) effector/memory (CD44^hiCD62L⁻), CD4⁺ and CD8⁺ T cells from day 50 spleens and dLNs from unchallenged and Der p 2-challenged WT and WASp KO mice on Balb/c background as measured by flow cytometry. (c) In vitro stimulation of spleen cells. Total splenocytes from unchallenged or Der p 2-challenged mice at day 50 were either unstimulated or stimulated with PMA plus ionomycin for 4 h or Der p 2 for 48 h (c). Absolute numbers of total CD4⁺IFNγ⁺ and CD8⁺IFNγ⁺ T cells after Der p 2 and PMA plus ionomycin stimulation as measured by flow cytometry. (a–c) Bar represents mean value and each dot represents one mouse. (a,b) Results are a pool of two separate experiments and (c) representative of two separate experiments. (a,b) WT unchallenged n=4–9; WKO unchallenged n=6–10; WT Der p 2 n=10–11; WKO Der p 2 n=8. (c) WT unchallenged n=3–7; WKO unchallenged n=5–8; WT Der p 2 n=5–6; WKO Der p 2 n=6. *P<0.05; **P<0.01 as calculated by the unpaired Student's t-test. WT, wild type; WKO, WASp KO.

Figure 3. L. major induces increased number of WASp KO CD8⁺IFNγ⁺ T cells.

(a) Ear infiltration of cells. (a) Ears from WT and WASp KO control or L. major infected mice on Balb/c background after 6 weeks. (b) Absolute numbers in ear of total MHCII^hiCD11c⁺ DCs; total CD4⁺CD3⁺ and CD8⁺CD3⁺ T cells, measured by flow cytometry. (c–e) dLN infiltration of cells. Absolute numbers in dLN of total MHCII^hi DCs; total CD4⁺CD3⁺ and CD8⁺CD3⁺ T cells; CD4⁺/CD8⁺ T-cell ratio; IFNγ⁺CD4⁺CD3⁺ and IFNγ⁺CD8⁺CD3⁺ cells, measured by flow cytometry. (a–e) Bar represents mean value and each dot represents one ear or dLNs. Results from week 2 and week 6 are representative of two separate experiments. WT control n=3–4; WKO control n=4; WT L. major 2 weeks n=6; WASp KO L. major 2 weeks n=6; WT L. major 6 weeks n=10; WASp KO L. major 6 weeks n=7. *P<0.05; **P<0.01 as calculated by the unpaired Student's t-test. FMO, fluorescence minus one (negative control for MHC class II and CD103); WT, wild type; WKO, WASp KO.

Figure 4. DC-specific WASp deletion induces increased CD8⁺ T cells.

(a) Schematic representation of the targeting strategy. Mice containing the WAS allele flanked by loxP sites were bred with CD11c-Cre mice to generate DC/cWKO mice on C57Bl/6 background. (b,c) WASp expression as determined by (b) western blotting and (c) flow cytometry of CD11c⁺CD8⁺, CD11c⁺CD8⁻, CD4⁺CD3⁺ and CD8⁺CD3⁺ cells from spleen. (d,e) Flow cytometry analysis of total CD4⁺CD3⁺ and CD8⁺CD3⁺ T cells in the (d) LNs and (e) spleen and (f) total effector/memory (CD44^hiCD62L⁻) CD4⁺CD3⁺ and CD8⁺CD3⁺ T cells in LNs and spleen. (c) Bar represents mean±s.d. of WT n=3; WKO n=1; DC/cWKO n=3. (d,f) WT n=6–7; DC/cWKO n=6. The data is representative of (b) one, (c) two, (d,e) four and (f) two separate experiments. *P<0.05; **P<0.01 as calculated by the unpaired Student's t-test. DC/cWKO, WASp^fl/flCD11c^Cre/wt; fl, floxed (LoxP flanked); WT, wild type; WKO, WASp KO.

Figure 5. Increased cross-presentation by WASp KO DCs.

(a) Immune synapse. Enriched CD8⁺ and CD8⁻ DCs from Flt3L tumour cell-injected mice on C57Bl/6 background were pulsed with ovalbumin and incubated with OT-I CD8⁺ T cells. Percentage of synapses was measured by counting the number of conjugates with polarized actin (red) and microtubule organizing center γ-tubulin (green) towards the synapse and divided by the total number of cell conjugates. (b) CD8⁺ T-cell proliferation with SIINFEKL peptide. Enriched CD8⁺ and CD8⁻ DCs from Flt3L tumour cell-injected mice on C57Bl/6 background were incubated overnight with 2 μg ml⁻¹ SIINFEKL peptide and co-cultured with CFSE-labelled OT-I (Vβ5.1/5.2⁺) CD8⁺ T cells for 72 h. Total number of OT-I CD8⁺ T cells is indicated. (c) CD8⁺ T-cell proliferation with ovalbumin. Equal numbers of FACS-sorted splenic CD8⁺ DCs and CD8⁻ DCs from wild-type and WASp KO mice on C57Bl/6 background were incubated overnight with ovalbuminm, co-cultured with CFSE-labelled OT-I (Vβ5.1/5.2⁺) CD8⁺ T cells, and proliferation determined at 72 h. Total number of OT-I CD8⁺ T cells is indicated. (d) Ovalbumin uptake. DCs were incubated with soluble ovalbumin-Alexa594 to assess uptake of ovalbumin. (e) Ovalbumin degradation. DCs were incubated with soluble DQ-ovalbumin to assess the capacity to process antigen. Note that increased DQ-ovalbumin mean fluorescence intensity indicates increased degradation. (f) Ovalbumin acidification. DCs were incubated with soluble pH rodo-ovalbumin. Note that increased pH rodo-ovalbumin mean fluorescence intensity indicates decreased pH value. (g) CD8⁺ DCs and CD8⁻ DCs from wild-type and WASp KO mice were incubated overnight with ovalbumin. The presentation of SIINFEKL peptide on MHC class I H-2K^b molecules was assessed by flow cytometry and fold increase in expression was determined using the MFI value acquired for 0 μg ml⁻¹ ovalbumin set to 1 (dotted line) (a) A total of 39–113 conjugates per mouse was analyzed. (a–g) Bar represents mean±s.d. of WT n=3; WKO n=3 per group. The data is representative of (a,b,g) two experiments, (c) four experiments and (d–f) three separate experiments. *P<0.05; **P<0.01 as calculated by the unpaired Student's t-test. MFI, mean fluorescence intensity; WT, wild type; WKO, WASp KO.

Figure 6. The phagosomal compartment in WASp KO DCs.

Wild-type and WASp KO CD8⁺ and CD8⁻ DCs from mice on C57Bl/6 background were incubated with ovalbumin-coated latex beads overnight. (a) Ovalbumin-bead uptake by DCs. (b) CD8⁺ T-cell proliferation with ovalbumin-coated beads. FACS-sorted DCs that had taken up one ovalbumin-coated bead were co-cultured with CFSE-labelled OT-I (Vβ5.1/5.2⁺) CD8⁺ T cells, and proliferation determined as CFSE dilution at 72 h. (c) DCs were incubated with latex beads coupled with pH-sensitive (FITC) and pH-insensitive (Alexa647) dyes. FITC and Alexa647 intensities were measured at the specified time points and the pH was determined as described in the materials and methods. (d,e) Ovalbumin-bead acidification. (d) DCs were incubated with pH rodo-ovalbumin-coated beads to assess acidification of antigen in phagocytic vesicles. Note that increased pH rodo mean fluorescence intensity translates into decreased pH value. (e) DCs were pre-treated with NH₄Cl to abolish acidification before addition of pH rodo-ovalbumin beads. (b–e) Bar represents mean±s.d. of WT n=3; WKO n=3. The data are representative of (b,d) three and (c,e) two separate experiments. *P<0.05; **P<0.01 as calculated by the unpaired Student's t-test. WT, wild type; WKO, WASp KO.

Figure 7. WASp KO CD8⁻ DCs activate Rac2 and ROS production.

(a) Rac1/2 expression. Wild-type and WASp KO CD8⁺ and CD8⁻ DCs from mice on C57Bl/6 background were stained intracellularly with anti-Rac1 and anti-Rac2 antibodies and analysed by flow cytometry. Bar represents mean±s.d. of WT n=3; WKO n=3. (b) Rac1/2 localization. DCs were incubated with ovalbumin-coated beads for 2 h to allow phagocytosis, transferred to fibronectin-coated glass and stained for Rac1-FITC (green) and Rac2-Alexa555 (red) antibodies and analysed by confocal microscopy. Both panels show Rac1 and Rac2 staining to the left and bright field to the right, with the phagocytosed bead marked with an asterisk. (Left panel) Rac2 co-localization with the phagosome was calculated as: [(beads with Rac2)/(cells with beads)] × 100. Bars represent mean±s.d. of 3–4 mice; 7–16 pictures with total 19–119 cells per mouse. (right panel) The MFI from the middle of the cell towards the bead (a) or in the opposite direction (b) was measured using the ImageJ software. The (MFI a/MFI b) is shown as Rac2 intensity around the bead. Bars represent mean±s.d. of 3–4 mice; 3–4 pictures with total 11–21 cells per mouse. (c) Rac1/2 activity. Quantification of active GTP-bound Rac1/2 and GTP-bound Rac2. Bars represent mean±s.d. of 3–6 mice. (d) NADPH induced ROS production. DCs from wild type, WASp KO, Ncf1*, and WASp KO × Ncf1* mice on C57Bl/6 background were enriched and incubated with DHR-coated beads for 1–2 h and analysed by flow cytometry for ROS production. The dotted line indicates background DHR intensity upon DPI treatment. The data in (a) are representative of three and (b–d) of two separate experiments. *P<0.05; **P<0.01 as calculated by the unpaired Student's t-test. Scale bar, 5 μm. MFI, mean fluorescence intensity; WT, wild type; WKO, WASp KO.

Figure 8. WASp and the WASp-VCA domain in cross-presentation.

(a–c) Wild-type and WASp KO BM DCs from mice on C57Bl/6 background were analysed for capacity to acidify pH rodo-ovalbumin and induce proliferation of OT-I (Vβ5.1/5.2⁺) CD8⁺ T cells. (d–f) Wild-type BM DCs at day 6 of culture were treated with CK666 to inhibit activity of the Arp2/3 complex or treated with α-amanitin to inhibit polymerase II gene transcription. (g,h) WASp KO BM DCs at day 6 of culture were Amaxa transfected with wild-type WASp (GFP-WASpWT) or WASp lacking the VCA domain (GFP-WASpΔVCA) and sorted into GFP⁺ and GFP⁻ cells 6 h after transfection. (a) Expression of CD11c and CD8 on BM DCs at day 6 of culture and analysis of acidification using uptake of soluble pH rodo-ovalbumin and confocal microscopy. (b,e,h) DCs were incubated with pH rodo-ovalbumin-coated beads and gated for DCs that had taken up one bead and pH rodo intensity was analysed by flow cytometry. (c,f,i) DCs were incubated with 0.5 mg ml⁻¹ soluble ovalbumin and co-cultured with OT-I (Vβ5.1/5.2⁺) CD8⁺ T cells, and proliferation was assessed as total number of OT-I CD8⁺ T cells by flow cytometry. (d) Wild-type BM DCs were treated for 6 h with CK666 to inhibit Arp2/3 activity and α-amanitin to inhibit RNA polymerase II transcription. Note the drop in polymerized actin measured using phalloidin upon CK666 treatment and the reduced expression of cyclin A2 upon α-amanitin treatment. (g) Gating strategy for sorting of GFP⁺ and GFP⁻ WASp KO BM DCs. Negative control indicates non-transfected cells. The data in (a–c) are representative of three and (d–h) of two separate experiments. *P<0.05; **P<0.01 as calculated by the unpaired Student's t-test. Scale bar, 10 μm. WASpΔVCA, WASp lacking the verprolin-cofilin-acidic domain; WT, wild type; WKO, WASp KO.