Quorum sensing governs collective dendritic cell activation in vivo

Abstract

Dendritic cell (DC) activation by viral RNA sensors such as TLR3 and MDA-5 is critical for initiating antiviral immunity. Optimal DC activation is promoted by type I interferon (IFN) signaling which is believed to occur in either autocrine or paracrine fashion. Here, we show that neither autocrine nor paracrine type I IFN signaling can fully account for DC activation by poly(I:C) in vitro and in vivo. By controlling the density of type I IFN-producing cells in vivo, we establish that instead a quorum of type I IFN-producing cells is required for optimal DC activation and that this process proceeds at the level of an entire lymph node. This collective behavior, governed by type I IFN diffusion, is favored by the requirement for prolonged cytokine exposure to achieve DC activation. Furthermore, collective DC activation was found essential for the development of innate and adaptive immunity in lymph nodes. Our results establish how collective rather than cell-autonomous processes can govern the initiation of immune responses.

Keywords: cytokine signaling; dendritic cell activation; quorum sensing; type I IFN.

Figure 1. Different models of cell‐cell communication through cytokine

Schemes illustrating distinct modes of cellular communication potentially contributing to DC activation. In the autocrine model, cells respond to their own cytokine production, while in the paracrine model, cells respond to the cytokine produced by their close neighbors. Finally, during quorum sensing, the cellular response is collective but critically depends on the density of secreting cells. In this model, a low density of secreting cells may not achieve detectable activation while a high density of cytokine‐producing cells triggers collective activation.

Figure 2. Cellular density influences dendritic cell activation by poly(I:C) in vitro

1. A, B

   BMDCs were plated at different densities and stimulated with poly(I:C) for 24 h or left unstimulated. DC activation was evaluated based on surface expression of CD86 and CD40 molecules as measured by flow cytometry. (A) Representative dot plots showing CD86 and CD40 expression. (B) Quantification of CD86 and CD40 upregulation as a fold change in geometric mean fluorescence intensity (gMFI) over unstimulated controls. Data are pooled from 3 independent experiments.

2. C

   MutuDCs were plated at different cell densities and stimulated with poly(I:C) or left unstimulated. The graph shows the upregulation of CD86 as a fold change in gMFI over unstimulated cells. Data are pooled from 4 independent experiments.

3. D

   Flow cytometric quantification of CD86 and CD40 upregulation on WT (blue) or Ifnar ^−/− (green) BMDCs expressed as a fold change in gMFI over unstimulated controls.

4. E

   CD86 and CD40 expression was quantified on BMDCs plated at high density (2 × 10⁵ cells/cm²) and stimulated with poly(I:C) in the presence (red dots) or absence (blue dots) of anti‐IFNAR1 blocking antibody. Data are pooled from 2 independent experiments.

5. F

   CD86 and CD40 expression was quantified on BMDCs plated at low density (2 × 10³ cells/cm²) and stimulated with poly(I:C) (dark blue dots), IFN‐α (light blue dots) or a combination of both (green dots). Data are representative of 2 independent experiments.

6. G

   IFN‐α (left) and IFN‐β (right) production was quantified by ELISA for BMDCs plated at different densities and stimulated or not with poly(I:C). Dotted lines represent the ELISA threshold of detection. Data are pooled from 2 independent experiments.

Data information: Black lines indicate mean values. Dotted lines in B, C, D, E, and F represent a fold change equal to 1. (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, ns; non‐significant, one‐way ANOVA (B, C, and F) and unpaired t‐test (D, E, and G)).

Figure 3. The density of type I IFN‐producing cells regulates DC activation by poly(I:C) in vitro

1. A

   IFN‐α (left) and IFN‐β (right) production by WT and Irf3 ^−/− Irf7 ^−/− BMDCs plated at high density (2 × 10⁵ cells/cm²) and stimulated with poly(I:C) (blue dots) or left unstimulated (gray dots). Dotted lines represent the ELISA threshold of detection. Data are representative of 2 independent experiments.

2. B

   CD86 and CD40 expression was quantified on WT (blue dots) and Irf3 ^−/− Irf7 ^−/− (orange dots) BMDCs plated at high density (2 × 10⁵ cells/cm²) and stimulated with poly(I:C) or left unstimulated. Data are pooled from 2 independent experiments.

3. C

   CD86 and CD40 expression on Irf3 ^−/− Irf7 ^−/− BMDCs plated at 2 × 10⁵ cells/cm² and stimulated with poly(I:C) (dark blue dots), IFN‐α (light blue dots), or a combination of both (green dots). Data are representative of 2 independent experiments.

4. D, E

   WT and Irf3 ^−/− Irf7 ^−/− BMDCs were mixed at different ratios for a total number of 2 × 10⁵ cells/cm² and stimulated with poly(I:C) or left unstimulated. (D) Representative dot plots showing CD86 and CD40 expression on WT (blue dots) and Irf3 ^−/− Irf7 ^−/− (orange dots) BMDCs. (E) Flow cytometric quantification of CD86 and CD40 expression on WT (blue dots) and Irf3 ^−/− Irf7 ^−/− (orange dots) BMDCs expressed as a fold change in gMFI over unstimulated controls.

5. F

   CD86 and CD40 expression on WT (blue dots) and Irf3 ^−/− Irf7 ^−/− (orange dots) BMDCs expressed as a fold change in the proportion of CD86⁺ CD40⁺ cells over unstimulated controls. Data are pooled from 2 independent experiments.

Data information: Black lines indicate mean values. Dotted lines in B, C, E, and F represent a fold change equal to 1. (**P < 0.01, ***P < 0.001, ****P < 0.0001, ns; non‐significant, unpaired t‐test (A, B, E, and F) and one‐way ANOVA (C)).

Figure 4. A quorum‐based mechanism governs DC activation in vivo

1. A

   CD86 and CD40 expression was quantified on WT (blue dots) and Irf3 ^−/− Irf7 ^−/− (orange dots) DCs in non‐draining (ndLN) and draining (dLN) popliteal lymph nodes of WT and Irf3 ^−/− Irf7 ^−/− mice injected with poly(I:C) or PBS. Data are pooled data from 2 independent experiments.

2. B

   Experimental set‐up. Irf3 ^−/− Irf7 ^−/− recipient mice were lethally irradiated and reconstituted with WT (GFP⁺) and Irf3 ^−/− Irf7 ^−/− (GFP^‐) bone marrow cells (BMCs) mixed at different ratios. Chimeras were injected in the footpad 4 weeks later with poly(I:C) and WT and Irf3 ^−/− Irf7 ^−/− DCs from non‐draining and draining popliteal lymph nodes were analyzed 24 h after poly(I:C) injection by flow cytometry.

3. C

   Microscopy images of sliced draining lymph nodes in CD11c‐eYFP:Irf3 ^−/− Irf7 ^−/− chimeras with 90:10 (left) and 5:95 (right) ratios. Type 1 IFN‐competent cells appear in green. Mosaic images were generated by acquisition of overlapping imaging fields. Scale bars represent 25 µm.

4. D, E

   CD86 and CD40 expression on WT (blue dots) and Irf3 ^−/− Irf7 ^−/− (orange dots) DCs from draining lymph nodes of mixed‐bone marrow chimeras with indicated WT:Irf3 ^−/− Irf7 ^−/− ratios. (D) Representative dot plots of CD86 and CD40 expression on WT (blue dots) and Irf3 ^−/− Irf7 ^−/− (orange dots) DCs of chimeras injected with PBS or poly(I:C). (E) Flow cytometric quantification of CD86 and CD40 expression on WT and Irf3 ^−/− Irf7 ^−/− DCs in chimeras injected with PBS or poly(I:C). Results from 5:95 and 10:90 chimeras were similar and pooled. Data are pooled from 3 independent experiments.

Data information: Each dot represents one mouse. Black lines indicate mean values. Dotted lines in A and E represent a fold change equal to 1. (**P < 0.01, ***P < 0.001, ****P < 0.0001, ns; non‐significant, unpaired t‐test).

Figure 5. Optimal DC activation requires prolonged exposure to type I IFN

1. Experimental set‐up. BMDCs cultured at low density (2 × 10³ cells/cm²) were stimulated with poly(I:C) and IFN‐α was added at various time points. IFNAR signaling was interrupted using anti‐IFNAR1 blocking antibody at the indicated time point. BMDC activation was measured 24 h after poly(I:C) stimulation by flow cytometry.

2. Representative dot plots showing CD86 and CD40 expression by low‐density BMDCs stimulated with poly(I:C) for different periods of IFN‐α exposure.

3. Quantification of CD86 and CD40 upregulation by BMDCs expressed as a fold change in gMFI over unstimulated controls. Data are pooled from 2 independent experiments. Black lines indicate mean values. Dotted lines represent a fold change equal to 1. (**P < 0.01, ***P < 0.001, ****P < 0.0001, one‐way ANOVA).

Figure 6. Collective DC activation regulates inflammation and T‐cell immunity

1. A, B

   Monocyte recruitment in the draining lymph node requires collective DC activation (A) Representative dot plots showing the percentage of monocytes (CD11b⁺ Ly6C^high cells, black box) present in popliteal dLN of WT:Irf3 ^−/− Irf7 ^−/− chimeric mice injected with poly(I:C) (middle and right) or with PBS (left). CD11b⁺ Ly6C^low/int cells corresponded to neutrophils and were not included in the analysis. (B) Flow cytometric quantification of monocyte numbers in the lymph nodes of mice injected with poly(I:C) expressed as a fold change over PBS controls. Data are pooled from 2 independent experiments.

2. C, D

   Activation of NK cells in the draining lymph node requires collective DC activation. (C) Representative histograms showing granzyme B and CD69 expression by NK1.1⁺ Ly6C^low cells in dLN of 90:10 (dark blue) and 5:95 (gray) WT:Irf3 ^−/− Irf7 ^−/− chimeric mice injected with poly(I:C) or with PBS (black line). (D) Quantification of granzyme B intracellular content (left) and CD69 expression (right) in NK1.1⁺ Ly6C^low cells of chimeric mice injected with poly(I:C), expressed as a fold change in gMFI over PBS‐injected controls. Data are pooled from 2 independent experiments.

3. E, F

   Cross‐priming of CD8⁺ T cells requires DC activation and type I IFN. (E) Representative dot plots showing the percentage of CD44^high K^b‐OVA tetramer⁺ CD8⁺ T cells in the draining lymph node of WT and Irf3 ^−/− Irf7 ^−/− mice injected with PBS, OVA, or a mixture of OVA plus poly(I:C). (F) Flow cytometric quantification of CD44^high H‐2K^b‐OVA tetramer⁺ CD8⁺ T cells in the draining lymph node of WT mice injected with PBS (gray dots), OVA (black dots), OVA plus poly(I:C) (blue dots) and in draining lymph node of Irf3 ^−/− Irf7 ^−/− mice injected with OVA plus poly(I:C) (orange dots). Data are pooled from 2 independent experiments.

4. G, H

   Collective DC activation is essential for the generation of T‐cell responses. (G) Representative dot plots showing the percentage of CD44^high tetramer H‐2K^b‐OVA⁺ CD8⁺ T cells in the draining lymph node of chimeric mice injected with PBS or a mixture of OVA+poly(I:C). (H) Flow cytometric quantification of CD44^high H‐2K^b‐OVA tetramer⁺ CD8⁺ T cells in the draining lymph node of 90:10 and 5:95 (WT:Irf3 ^−/− Irf7 ^−/−) chimeric mice injected with PBS or OVA+poly(I:C). Data are pooled from 2 independent experiments.

Data information: Each dot represents one mouse. Black lines indicate mean values. Dotted lines indicate fold change equal to 1. (**P < 0.01, ***P < 0.001, ns; non‐significant, unpaired t‐test (B and D) and one‐way ANOVA (F and H)).