DNGR-1 signalling limits dendritic cell activation for optimal antigen cross-presentation

Abstract

Innate immune receptors often induce activation of conventional dendritic cells (cDCs) and enhance antigen (cross-)presentation, favouring immune responses. DNGR-1 (CLEC9A), a receptor expressed by type 1 cDCs (cDC1s) and implicated in immune responses to viruses and cancer, recognises F-actin exposed on dead cell remnants and promotes cross-presentation of associated antigens. Here, we show that recruitment of phosphatase SHIP1, a process governed by a single amino acid residue adjacent to the signalling motif of the receptor, partly explains how DNGR-1 fails to trigger cDC1 activation in vitro. Substituting this residue converts DNGR-1 into an activating receptor but decreases induction of cross-presentation of dead cell-associated antigens. Introducing the reverse mutation into the related receptor Dectin-1 impairs its activation capacity while enhancing its ability to promote cross-presentation. These findings reveal a functional trade-off in receptor signalling and suggest that DNGR-1 has evolved to prioritise antigen cross-presentation over cellular activation, possibly to minimise inflammatory responses to dead cells.

Keywords: Activation; CLEC9A; Cross-presentation; DNGR-1; cDC1.

Figure 1. DNGR-1 signalling results in phagosomal rupture and cross-presentation of dead cell-associated antigens but does not activate cDC1s.

(A) ELISA for IFN-γ release from OT-I effector T (T_E) cells co-cultured with wild-type (WT) or DNGR-1 deficient Clec9a knock-in Cre (C9^KI-Cre) bone marrow-FLT3L cultured (BM-FLT3L) cDC1s incubated with ovalbumin (OVA)-dead cells (left) or SIINFEKL peptide (right). Mean ± SD from biological duplicates is plotted. (B) Uptake of Cell Tracker-Deep Red (CT-DR)-labelled dead cell debris by WT or C9^KI-Cre BM-FLT3L cDC1s assessed by flow cytometry. Plotted as phagocytic index (% CT-DR⁺ cells x CT-DR MFI of CT-DR⁺ cells/arbitrary unit) mean ± SD from n = 3. (C) Schematic of WT (C9), W155A-W250A (C9(2WA)), or Y7F (KO/C9(Y7F)) DNGR-1 transduced into DNGR-1 knockout (KO) splenic cDC1 line MuTuDC1940 (MuTuDCs). Intracellular cytoplasmic domain (ICD); transmembrane domain (TM); extracellular domain (ECD). (D) ELISA for IFN-γ release from OT-I T_E cells co-cultured with C9 KO, KO/C9, KO/C9(2WA), or KO/C9(Y7F) MuTuDCs incubated with OVA-dead cells (left), SIINFEKL peptide (right), or (E) DNGR-1 ligand (DNGR-1L)-OVA coupled beads. Mean ± SD from biological (D) quadruplet or (E) duplicates is plotted. All lines are plotted even when they cannot be seen because of superimposition. (F) Uptake of CT-DR-labelled dead cell debris as in (B) by C9 KO or KO/C9 MuTuDCs. Cytochalasin D (CD) co-culture was included as a negative control. (G, H) C9 KO or KO/C9 MuTuDCs transduced with lysenin-mCherry fusion protein were co-cultured with α-DNGR-1 IgG coupled beads and assessed by confocal microscopy. (G) Representative images, scale bar = 5 µm. Beads not internalised marked by α-rat IgG staining. (H) Quantification of lysenin-mCherry⁺ phagosomes per cells in field of view (Lysenin index), bars indicate mean ± SEM. (I) Absorbance of β-galactosidase activity from B3Z-DNGR-1-SYK reporter cells stimulated ± DNGR-1L (left) or plate-bound α-DNGR-1 IgG (right). Mean ± SEM of four replicates. (J) Confocal microscopy of KO/C9 MuTuDCs treated as in (G). (K, L) WT, C9^KI-Cre BM-FLT3L cDC1s or (M, N) C9 KO MuTuDCs or cells reconstituted with indicated receptors were cultured overnight ± indicated stimuli and assessed in triplicate by flow cytometry for (K, M) surface or (N) intracellular protein expression, or (L) release of IL-12 p40 by ELISA in cultured supernatants. Cells treated with 200 U/mL IFN-α or 20 nM DNGR-1L (K) or 10 µg/mL Poly(I:C) or 20 nM DNGR-1L (L). Mean ± SEM of each group from biological triplicates (K, L) or pooled duplicates from two independent experiments (M, N) is plotted. MFI mean fluorescence intensity. Data are representative of two (A, B, F–H, J–N) or ≥ three (D, E, I) independent experiments. Data were analysed using Tukey-corrected two-way ANOVA (A–F, K–N) or unpaired t test (H). Significant values comparing against C9^KI-Cre BM-FLT3L cDC1 (A), C9 KO MuTuDCs (D, E, H), or untreated samples (K–N) are plotted. (A, D, E) ****P < 0.0001, (H) ***P = 0.0002, (K) ****P < 0.0001, (L) **P = 0.0020, ****P < 0.0001, (M) ***P = 0.0004, ****P < 0.0001, (N) ****P < 0.0001. See also Fig. EV1. Source data are available online for this figure.

Figure 2. An isoleucine residue adjacent to the hemITAM motif restrains the ability of DNGR-1 to activate cDC1s.

(A) Aligned DNGR-1 (CLEC9A, C9; left) or Dectin-1 (CLEC7A, C7; right) cytoplasmic domain sequences from indicated species. HemITAM sequence is bolded in red. Consensus is represented by an asterisk (*) and similarity with a dot (.). (B) Aligned hemITAM motifs of DNGR-1 and Dectin-1 from Mus musculus. The triacidic motif from Dectin-1 is also depicted with the corresponding sequence from DNGR-1. (C) Schematic of chimeric receptor comprised of the intracellular cytoplasmic domain (ICD) of Dectin-1 fused with the transmembrane (TM) and extracellular domain (ECD) of DNGR-1 (C7::C9), or I6G DNGR-1 (C9(I6G)) transduced into C9 KO MuTuDC (left). Right side depicts WT Dectin-1 (C7), or chimeric receptor constructs using the TM and ECD of Dectin-1 fused with the ICD of WT (C9::C7), I6G (C9(I6G)::C7), or Y7F (C9(Y7F)::C7) DNGR-1 transduced into RAW 264.7 cells. (D, E) Representative flow cytometric analysis of surface proteins from C9 KO MuTuDCs or those reconstituted with indicated receptors and stimulated overnight ± with DNGR-1L. (E) Quantification of flow cytometric analysis of surface proteins from MuTuDCs in (D), as well as CCL22 in cultured supernatants from those cells. Mean ± SD from biological duplicates is plotted. (F) Representative histograms and quantification of flow cytometric analyses of surface marker MFI or TNF-α in cultured supernatants from RAW 264.7 cells ectopically expressing indicated receptors or transduced with empty vector (EV) stimulated overnight ± with zymosan depleted (Zym-D). Mean ± SD from biological duplicates is plotted. (G) RNAseq volcano plot of differentially expressed genes between 12.5 nM DNGR-1L stimulated versus untreated (Control) KO/C9 (left) or KO/C9(I6G) MuTuDCs (right). N = 3 per group. (H) RT-qPCR analysis of indicated genes from MuTuDCs treated with 12.5 nM DNGR-1L over time. Mean ± SD from biological duplicates is plotted. (I) Gene set enrichment analysis (GSEA) of Reactome signalling pathways identified in DNGR-1L-stimulated KO/C9(I6G) MuTuDCs versus KO/C9 MuTuDCs from (G). Data are representative of two (F, H), or three (D, E) independent experiments. See also Fig. EV2. Source data are available online for this figure.

Figure 3. cDC1 glycolytic switch is restricted during DNGR-1 signalling.

(A) GSEA of Reactome metabolic pathways identified in C9 KO MuTuDCs reconstituted with C9 or C9(I6G) ± DNGR-1L stimulation. (B) GSEA of Reactome glucose metabolism and glycolysis pathways in KO/C9(I6G) MuTuDCs from (A). (A, B) Data derived from experiment in Fig. 2G. (C) Extracellular acidification rate (ECAR, indicator of glycolysis) measured at baseline and after ± 30 nM DNGR-1L injection of C9 KO MuTuDCs reconstituted with indicated receptors. N = 3–6 per group. Data normalised to baseline measurement immediately after injection with stimuli and shown as % of baseline. Mean ± SEM is plotted. (D, E) ECAR at 2 h post-treatment of (D) MuTuDCs treated as in (C) or (E) RAW 264.7 cells ectopically expressing indicated receptors or transduced with EV stimulated ± Zym-D. N = 3–6 per group. Data normalised as in (C). Mean ± SEM is plotted. (F) Schematic of major metabolic pathways downstream of glucose catabolism. PEP phosphoenolpyruvate, PKM2 pyruvate kinase M2, S7P sedoheptulose 7-phosphate, 3-PG 3-phosphoglycerate, OAA oxaloacetate, α-KG α-ketoglutarate, TCA tricarboxylic acid, PPP pentose phosphate pathway. (G) Heatmap showing metabolites in untreated KO/C9 MuTuDCs or KO/C9, KO/C9(I6G), or KO/C7::C9 MuTuDCs treated with 12.5 nM DNGR-1L (log2 fold change of levels at 60 min compared to 0 min post-stimulation). F6P fructose 6-phosphate, F1,6BP fructose 1,6-bisphosphate, DHAP dihydroxyacetone phosphate, SAM S-adenosyl methionine, 2-HG 2-hydroxyglutarate. Colour bars along the right side of the graph correspond to schematic in (F). (H) Fractional labelling of metabolites in same groups described and treated as in (G) cultured with uniformly-labelled U-¹³C-glucose introduced at the time of stimulation. Mean ± SEM from five biological replicates is plotted. (I, J) Left, glycolytic capacity (maximum ECAR after rotenone and antimycin A injection) and right, ECAR and oxygen consumption rate (OCR) of MuTuDCs treated for <1 h with (I) C3K (PKM2 inhibitor) or (J) DASA-58 (PKM2 agonist). (I, J) Data are normalised to baseline measurement immediately after injection with PKM2 drugs and shown as % of baseline. Mean ± SEM is plotted (n = 4 per treatment and 9 for untreated samples). (K) Flow cytometric analysis of % I-A/E^hi CD86^hi cells of indicated MuTuDCs stimulated ± 12.5 nM DNGR-1L in the presence of DMSO (Control), 10 µM C3K, or 40 µM DASA-58. Data are normalised to control condition for each cell line. Mean ± SEM from triplicate measurements is plotted. Data are representative of two independent experiments (C–E, I–K). Data were analysed using Tukey-corrected two-way ANOVA (C–E, H) or one-way ANOVA (I–K). Significant values comparing against untreated samples are plotted (C–E, H–K). (C) ****P < 0.0001, (D) **P = 0.0048, ****P < 0.0001, (E) **P = 0.0063, ***P = 0.0004, ****P < 0.0001, (H) *P = 0.0362, ****P < 0.0001, (I) **P = 0.0032, ****P < 0.0001, (J) ***P = 0.0005, ****P < 0.0001, (K) *P = 0.0309, ***P = 0.0006, ****P < 0.0001. See also Fig. EV3. Source data are available online for this figure.

Figure 4. SHIP1 tempers cDC1 activation by DNGR-1.

(A) Schematic of pull-down performed against the intracellular cytoplasmic domain (ICD) of C9, C9(I6G), or C9 and C9(I6G) 7Y-phosphorylated peptides incubated with lysates from C9 KO MuTuDCs. (B) Heatmap of label-free quantification (LFQ) intensities from samples outlined in (A) analysed by mass spectrometry. (C) Western blot analysis of C9 KO MuTuDCs or those reconstituted with C9 or C9(I6G) treated with 12.5 nM DNGR-1L. Representative images (left) and densitometry analyses of indicated proteins normalised to β-actin signal (x10). Densities represent ≥ three independent experiments shown as mean ± SEM. Ubq-SYK ubiquitinated-SYK pY352. (D) Heatmap of surface protein MFIs assessed by flow cytometry in indicated MuTuDC lines edited with control guides (gControl) or guides targeting SHIP1, SHP-1, or SHP-2 cultured overnight ± 12.5 nM DNGR-1L. Data shown are relative to untreated samples. (E) Confocal microscopic analysis of SHIP1 pY1021 in KO/C9, KO/C9(I6G), and KO/C9(2WA) MuTuDCs co-cultured with α-DNGR-1 IgG coupled beads. Representative images (left). SHIP1 pY1021⁺ phagosomal MFI were binned and plotted against their relative distribution with a fitted curve (top right). Violin plot quantification of SHIP1 pY1021⁺ phagosomes (bottom right) of internalised beads (SHIP1 pY1021 index). Data are representative of two independent experiments (D, E). Data were analysed using Tukey-corrected two-way ANOVA (C) or unpaired t test (E). Only significant values observed between KO/C9 and KO/C9(I6G) treated samples plotted (C, E). (C) SHIP1 pY1021 10 min *P = 0187, 30 min *P = 0.0390; SHP-2 pY542 *P = 0.0338; SHP-2 pY580 10 min ***P = 0.0002, 30 min **P = 0.0038, 60 min **P = 0.0028; p38 pT180/Y182 10 min *P = 0.0475, **P = 0.0064, 60 min ****P < 0.0001, (E) **P = 0.0042. See also Fig. EV4. Source data are available online for this figure.

Figure 5. An activatory DNGR-1 exhibits compromised ability to induce cross-presentation.

(A–C) ELISA for IFN-γ release from OT-I T_E cells co-cultured with C9 KO MuTuDCs or those reconstituted with indicated receptors incubated with (A) DNGR-1L-OVA-coupled beads, (B) OVA-dead cells, or (C) SIINFEKL peptide. Mean ± SD from biological (B, C) duplicates or (A) triplicates are plotted. (D, E) Uptake of DNGR-1L-coupled beads by C9 KO MuTuDCs or those reconstituted with indicated receptors assessed by flow cytometry. (D) Representative plot depicting the use of post-uptake streptavidin staining to distinguish MuTuDCs that have internalised (in) biotinylated-DNGR-1L coupled beads from those attached to surface DNGR-1 (out). (E) Representative histogram of cells treated 20:1 with beads (left) and phagocytic index (% internalised beads x bead MFI/arbitrary unit) mean ± SD from biological duplicates (right). (F–H) Confocal microscopic analysis of lysenin-mCherry fusion protein-expressing C9 KO MuTuDCs or those reconstituted with indicated receptors co-cultured with α-DNGR-1 IgG-coupled or isotype IgG-coupled beads. Beads not internalised marked by α-rat IgG staining. (F) Violin plot quantification of internalised beads per cell per field of view (n = 20). (G) Violin plot of lysenin-mCherry⁺ phagosomes per cells in field of view (Lysenin index). (H) Representative images of lysenin-mCherry⁺ phagosomes from indicated MuTuDCs, scale bar = 5 µm. (I–K) ELISA for IFN-γ release from OT-I T_E cells co-cultured with KO/C9(I6G) or SHIP1 sufficient (guide control; gControl) or deficient KO/C9 MuTuDCs incubated with (I) DNGR-1L-OVA-coupled beads, (J) OVA-dead cells, or (K) SIINFEKL peptide. Mean ± SD from (I–K) biological duplicates or quadruplets (KO/C9 SHIP1 KO) are plotted. (L) Uptake of DNGR-1L-coupled beads by KO/C9 SHIP1 sufficient or deficient MuTuDCs assessed by flow cytometry. Phagocytic index plotted with mean ± SD from biological triplicates. Data are representative of two (F–L) or ≥ three (A–C) independent experiments. Data were analysed using Tukey-corrected two-way ANOVA with only significant values observed between KO/C9 and KO/C9(I6G) or KO/C7::C9 MuTuDCs plotted in (A–C, E–G) or between KO/C9 gControl and KO/C9(I6G) or KO/C9 SHIP1 KO MuTuDCs in (I–K). (A, B, G, I, J) ****P < 0.0001. See also Fig. EV5. Source data are available online for this figure.

Figure 6. A “DNGR-1-like” Dectin-1 receptor displays reduced ability to induce activation but gains cross-presentation capacity.

(A, B) Flow cytometric analyses of surface proteins and ELISA of indicated proteins released in culture supernatants from (A) RAW 264.7 cells ectopically expressing indicated receptors or transduced with EV stimulated overnight ± Zym-D or (B) C9 KO MuTuDCs reconstituted with indicated receptors stimulated overnight ± DNGR-1L. Data shown as mean ± SEM from biological triplicates. Note that the TNF-α data in (A) are plotted in two different ways (absolute versus relative concentration to untreated controls) to emphasise the similarity between C7 and C9(I6G)::C7. (C) Uptake of DNGR-1L-coupled beads by C9 KO MuTuDCs or those reconstituted with indicated receptors assessed by flow cytometry. Plotted as phagocytic index (% internalised beads x bead MFI of bead⁺ cells/arbitrary unit) mean ± SD from biological duplicates. (D, E) C9 KO MuTuDCs reconstituted with C7::C9 or C7(G14I)::C9 receptors co-cultured with DNGR-1L-OVA coupled beads. Single internalised bead⁺ cells were subsequently sorted and co-cultured with OT-I T_E cells. (D) Schematic of experiment (left) and representative flow plots from pre- and post-sort enrichment (right). (E) ELISA for IFN-γ released from OT-I T_E cells co-cultured with MuTuDCs from (D) or MuTuDCs loaded with exogenous SIINFEKL peptide. Mean ± SD from biological duplicates is plotted. Data are representative of two (D, E) or ≥ three (A–C) independent experiments. Data were analysed using Tukey-corrected two-way ANOVA (A–C, E). Significant values comparing against untreated controls (A, B) or between KO/C7(G14I)::C9 (E) are plotted. (A) H2-D^d **P = 0.0043, ***P = 0.0006, ****P < 0.0001; CD83 *P = 0.0374, ***P = 0.0002, ****P < 0.0001; TNF-α **P = 0.0017 (C7), P = 0.0067 (C9(Y7F)::C7, ***P = 0.0004 (C7(G14I)), P = 0.0006 (C9::C7), ****P < 0.0001, (B) CD86 *P = 0.0469, ****P < 0.0001; CCR7 **P = 0.0097, ***P = 0.0004, ****P < 0.0001; CCL17 ****P < 0.0001; CCL22 *P = 0.0217, **P = 0.0018, ****P < 0.0001, (C, E) ****P < 0.0001. See also Fig. EV6. Source data are available online for this figure.

Figure EV1. Assessment of DNGR-1 stimulation on cDC1 activation (related to Fig. 1).

(A) Representative flow cytometry gating strategy for analyzing cDCs derived from bone marrow-FLT3L cultures (BM-FLT3L) before and after XCR1⁺ MACS enrichment. (B) DNGR-1 expression by WT BM-FLT3L cDC1 and cDC2 compared to C9^KI-Cre cDC1. (C–E) WT, C9^KI-Cre BM-FLT3L cDC1s or (F, G) C9 KO MuTuDCs or those reconstituted with indicated receptors were cultured overnight ± designated stimuli and assessed for surface expression of the specified markers by flow cytometry. (C) H2-K^b expression from biological duplicates combined from two experiments with mean ± SEM plotted (left) and representative histograms (right). (D) Representative histograms of staining for indicated markers. (E) Cells were cultured overnight on uncoated plates or plates coated with different concentrations of α-DNGR-1 IgG (clone 1F6 or 7H11, as indicated) and assessed for indicated surface marker expression by flow cytometry. MFI values from biological duplicates pooled from two independent experiments with mean ± SEM is plotted. (F) Representative histograms of staining for indicated markers. (G) H2-K^b expression from biological duplicates pooled from two independent experiments with mean ± SEM is plotted. Data are representative of two (A–G) independent experiments. (C, E, G) Data were analysed using Tukey-corrected two-way ANOVA with significant values comparing against untreated samples plotted. (C) ****P < 0.0001.

Figure EV2. Cell lines to analyse DNGR-1 function (related to Fig. 2).

(A) Analysis of the indicated surface marker expression (top, middle; flow cytometry) or CCL22 released into cultured supernatants (bottom; ELISA) from C9 KO MuTuDCs reconstituted or not with the indicated receptors and stimulated overnight ± DNGR-1L. Mean ± SEM from biological replicates pooled from two independent experiments (left) and representative flow cytometry profiles (right) are plotted. Data here are partly represented in Fig. 2D,E and are relative to average of untreated controls to better emphasise the response to DNGR-1L. Dotted line represents 1. (B) Flow cytometric analysis of surface DNGR-1 (C9) expression in C9 KO MuTuDCs or those reconstituted or not with the indicated receptors. Cell lines were established after sorting for equal expression of DNGR-1. (C) Flow cytometric analysis of surface Dectin-1 (C7) expression by parental RAW 264.7 cells or cells ectopically expressing the indicated receptors or transduced with empty vector (EV). Cell lines were established after sorting for equal expression of Dectin-1. (B, C) Representative histograms (left) and mean MFI ± SEM (right) are plotted from biological (B) quintuplets or (C) triplicates. (D) Analysis of the indicated surface marker expression (top, middle; flow cytometry) or TNF-α released into cultured supernatants (bottom; ELISA) from RAW 264.7 cells ectopically expressing the indicated receptors or transduced with EV and stimulated overnight ± Zym-D. Mean ± SEM from biological replicates pooled from two independent experiments (left) and representative flow cytometry profiles (right) are plotted. Data here are partly represented in Fig. 2F and are relative to average of untreated controls to better emphasise the response to Zym-D. Dotted line represents 1. Data are representative of two (D), or three (A–C) independent experiments. Data were analysed using Tukey-corrected two-way ANOVA with significant values comparing against untreated samples plotted (A, D). (A) I-A/E^hi CD86^hi *P = 0.0117, ****P < 0.0001; CCR7 *P = 0.0279 (KO/C7::C9), P = 0.0134 (KO/C9(I6G)), ****P < 0.0001; CCL22 ****P < 0.0001, (D) H2-D^d **P = 0.0011, ****P < 0.0001; CCR7 *P = 0.0319, **P = 0.0040, ****P < 0.0001; TNF-α **P = 0.0033, ***P = 0.0004, ****P < 0.0001.

Figure EV3. Assessment of metabolism changes induced by DNGR-1 signalling (related to Fig. 3).

(A, B) Oxygen consumption rates (OCR) measured at 2 h after ± DNGR-1L injection of (A) C9 KO MuTuDCs or those reconstituted with the indicated receptors or (B) RAW 264.7 cells ectopically expressing the indicated receptors or EV injected ± Zym-D, n = 3–6 per group. Data normalised to baseline measurement immediately after injection with stimuli. Mean ± SEM relative to untreated samples is plotted. (C) Fractional labelling of glycolytic (top) or tricarboxylic acid cycle (bottom) metabolites in C9 KO MuTuDCs reconstituted with indicated receptors and stimulated ± 12.5 nM DNGR-1L cultured with uniformly-labelled U-¹³C-glucose introduced at the time of stimulation. Data shown as mean ± SEM from five biological replicates. (D) Extracellular acidification rate (ECAR) and OCR of MuTuDCs treated for 2 h with C3K (PKM2 inhibitor) or DASA-58 (PKM2 agonist). Data normalised as in (A). Mean ± SEM is plotted (n = 4 per treatment and 9 for untreated samples). Data are representative of two (A, B, D) independent experiments. Data were analysed using Tukey-corrected two-way ANOVA with significant values comparing against untreated controls plotted (A–D). (C) (top) *P = 0.0128, **P = 0.0084, ***P = 0.0006; (bottom) *P = 0.0136, ****P < 0.0001, (D) ****P < 0.0001.

Figure EV4. SHIP1 inhibition rescues DNGR-1 mediated cDC1 activation. (related to Fig. 4).

(A) Western blot analysis of C9 KO MuTuDCs reconstituted with C9 made deficient for target proteins using CRISPR/Cas9. Two guide (g) RNAs that target SHIP1 (gSHIP1), SHP-1 (gSHP-1), SHP-2 (gSHP-2) or scrambled sequences (gControl) were used to complex with recombinant Cas9 protein for nucleofection. (B) Flow cytometric analysis of surface marker MFIs detected from KO/C9, KO/C9(I6G) or KO/C9 SHIP1, SHP-1, or SHP-2 deficient (KO) MuTuDCs cultured overnight ± 12.5 nM DNGR-1L. Mean ± SEM relative to untreated (Control) samples from biological replicates pooled from two independent experiments is plotted. (C) Flow cytometric analysis of surface protein MFIs from C9 KO MuTuDCs reconstituted with C9 or (I6G) MuTuDCs stimulated ± 12.5 nM DNGR-1L alone or in the presence of SHIP1 inhibitor 3AC or SHP-2 inhibitor RMC-4550 overnight. Mean ± SEM relative to untreated (Control) samples from biological triplicates is plotted. (D, E) Extracellular acidification rate (ECAR, indicator of glycolysis) measured at baseline and after ± 10 or 30 nM DNGR-1L injection of C9 KO SHIP1 sufficient and deficient MuTuDCs reconstituted with indicated receptors. N = 4–9 per group. (D) Data normalised to baseline measurement immediately after injection with stimuli and shown as % of baseline. Mean ± SEM is plotted. (E) ECAR at 2 h post-treatment. Data normalised as in (C). Mean ± SEM is plotted. Data are representative of one (C) or two (A, B, D, E) independent experiments. Data were analysed using Tukey-corrected two-way ANOVA. Comparisons are indicated (B, C) or against untreated controls (D, E) with significant values plotted. (B) **P = 0.0011, ****P < 0.0001, (C) *P = 0.0177, **P = 0.0029, ****P < 0.0001, (D) **P = 0.0071, ****P < 0.0001, (E) **P = 0.0071, ****P < 0.0001.

Figure EV5. Cross-presentation of NP antigen and phagosomal rupture (related to Fig. 5).

(A, B) IFN-γ release from F5 T_E cells co-cultured with C9 KO MuTuDCs reconstituted or not with the indicated receptors and incubated with (A) NP-expressing dead cells, or (B) ASNENMDAM peptide. Mean ± SD from biological triplicates is plotted. (C) Confocal microscopic analysis of lysenin-mCherry fusion protein-expressing KO/C9 MuTuDCs co-cultured with α-DNGR-1 IgG-coupled beads. 0.5 µm increments of consecutive Z-slices of the same image in Fig. 5H. Arrow indicates ruptured area of lysenin-mCherry⁺ phagosome indicated by loss of mCherry signal. Scale bar = 2 µm. Data are representative of two independent experiments (A–C). Data were analysed using Tukey-corrected two-way ANOVA. Comparisons are indicated (A, B) with significant values plotted. (A) ****P < 0.0001.

Figure EV6. Cell lines to analyse Dectin-1 function (related to Fig. 6).

(A) Flow cytometric analysis of the indicated surface marker expression by RAW 264.7 cells ectopically expressing indicated receptors or transduced with EV and stimulated overnight ± Zym-D. Mean ± SEM from biological triplicates (left) and representative histograms (right) are plotted. Note that these data are the same data as in Fig. 6A, with bar graphs plotted here as MFI relative to untreated controls to better emphasise the response to Zym-D. Dotted line represents 1. (B) Flow cytometric analysis of surface Dectin-1 (C7) expression in RAW 264.7 cells ectopically expressing the indicated receptors or transduced with EV. Cell lines were established after sorting for equal expression of Dectin-1. (C) Flow cytometric analysis of surface DNGR-1 (C9) expression in C9 KO MuTuDCs reconstituted or not with the indicated chimeric receptors. Cell lines were established after sorting for equal expression of DNGR-1. (B, C) Representative histograms (left) and mean MFI ± SEM (right) are plotted from biological (B) quintuplets or (C) triplicates. (D) Flow cytometric analysis of surface marker expression by C9 KO MuTuDCs reconstituted or not with indicated receptors stimulated overnight ± DNGR-1L. Mean ± SEM from biological triplicates (left) and representative histograms (right) are plotted. CCR7 is shown relative to untreated controls and the dotted line represents 1. Note that these are data from the same experiments as in Fig. 6B, plotted differently. Data are representative of two (B, C), or three (A, D) independent experiments. Data were analysed using Tukey-corrected two-way ANOVA with significant values comparing against untreated samples plotted (A–D). (A) H2-D^d *P = 0.0405, **P = 0.0011, ***P = 0.0007, ****P < 0.0001; CD83 *P = 0.0374, ***P = 0.0001, ****P < 0.0001, (D) I-A/E^hi CD86^hi *P = 0.0352 (KO/C7::C9), P = 0.0468 (KO/C7(G14I)::C9), ****P < 0.0001; CCR7 *P = 0.0240, ***P = 0.0004, ****P < 0.0001.