An IL-23-STAT4 pathway is required for the proinflammatory function of classical dendritic cells during CNS inflammation

Abstract

Although many cytokine pathways are important for dendritic cell (DC) development, it is less clear what cytokine signals promote the function of mature dendritic cells. The signal transducer and activator of transcription 4 (STAT4) promotes protective immunity and autoimmunity downstream of proinflammatory cytokines including IL-12 and IL-23. In experimental autoimmune encephalomyelitis (EAE), an animal model of multiple sclerosis (MS), Stat4^-/- mice are resistant to the development of inflammation and paralysis. To define whether STAT4 is required for intrinsic signaling in mature DC function, we used conditional mutant mice in the EAE model. Deficiency of STAT4 in CD11c-expressing cells resulted in decreased T cell priming and inflammation in the central nervous system. EAE susceptibility was recovered following adoptive transfer of wild-type bone marrow-derived DCs to mice with STAT4-deficient DCs, but not adoptive transfer of STAT4- or IL-23R-deficient DCs. Single-cell RNA-sequencing (RNA-seq) identified STAT4-dependent genes in DC subsets that paralleled a signature in MS patient DCs. Together, these data define an IL-23-STAT4 pathway in DCs that is key to DC function during inflammatory disease.

Keywords: EAE; STAT4; cytokines; dendritic cells.

Fig. 1.

STAT4 in T cells and CD11c⁺-expressing cells, but not in Lyz2-expressing cells, is required for EAE development. (A and B) EAE clinical and cumulative scores of WT and Stat4^fl/flCD4^Cre mice. (C) Representative flow plots of the frequencies and scatter plots of the absolute numbers of infiltrating CD4⁺ T cells in the CNS of WT and Stat4^fl/flCD4^Cre mice. (D and E) EAE clinical and cumulative scores of WT and Stat4^fl/flLyz2^Cre mice post-EAE induction. (F) Representative flow plots of the frequencies and scatter plots of the absolute numbers of infiltrating CD4⁺ T cells in the CNS of WT and Stat4^fl/flLyz2^Cre mice. (G and H) EAE clinical and cumulative scores of WT and Stat4^fl/flCD11c^Cre mice post-EAE induction. (I) Representative flow plots of the frequencies and scatter plots of the absolute numbers of infiltrating CD4⁺ T cells in the CNS of WT and Stat4^fl/flCD11c^Cre mice. Data represent mean ± SEM (n = 5) and represent two to three independent experiments. One-way ANOVA with Tukey post hoc test was used to compare means in (A), (D), and (F), and an unpaired t test was used to compare means in the remaining plots. ^**P < 0.01, ^***P < 0.001, and ^****P < 0.0001. ns, indicates not significant.

Fig. 2.

STAT4 is activated in CD11c⁺ cells in the inflamed CNS and regulates cDC2 infiltration. (A) Scheme of pSTAT4 staining in CD11c⁺ cells of the lumbar sections at the peak of EAE. (B) Immunofluorescence staining of pSTAT4 and CD11c in the spinal cord lumbar sections of EAE mice at day 15 postimmunization, EAE clinical score 3.5. (C) Representative flow plots and scatter plots of the frequencies of cDC2 cells in the CNS of WT and Stat4^fl/flCD11c^Cre at day 16 post-EAE immunization. (D) Representative histograms and scatter plot of MHCII^high+ frequencies in the CNS of WT and Stat4^fl/flCD11c^Cre at day 16 postimmunization. (E–G) Representative histograms and scatter plots of gMFI of the costimulatory markers (CD86, CD80, and CD40) on the surface of cDC2s in the CNS. Data are pooled from two to three independent experiments (n = 16 or n = 10). An unpaired t test was used to compare means in all the plots. ^***P < 0.001 and ^****P < 0.0001. gMFI, geometric mean fluorescence intensity.

Fig. 3.

Transferring WT CD45.1 CD11c⁺ cDCs renders Stat4^fl/flCD11c^Crerecipient mice EAE susceptible. (A) Scheme of adoptive transfer of WT CD45.1-BMDCs and CD45.2-BMDCs to Stat4^fl/flCD11c^Cre recipient mice. BMDCs were adoptively transferred to the recipient mice on days −1, 5, 3, 7, and 9 of EAE. (B) Representative flow plots of the transferred (MHCII⁺CD26⁺CD11c⁺CD11b⁺CD172α⁺) BMDCs. (C) EAE clinical score and (D) scatter plot of the cumulative score of the recipient mice after transferring CD45.1 or CD45.2-BMDCs (n = 10 or 7 per mouse group). (E and F) Representative flow plots and scatter plots of the frequencies of CD4⁺ T cells in the CNS of the WT and Stat4^fl/flCD11c^Cre recipient mice. (G and H) Representative histograms and scatter plots of the frequencies of MHCII⁺ cells in WT mice and Stat4^fl/flCD11c^Cre recipient mice. (I) Representative flow plot of the frequencies of CD45.1-cDC2s in the recipient Stat4^fl/flCD11c^Cre in the CNS. (J) Bar plot of the frequencies of cDC2s in WT mice and the Stat4^fl/flCD11c^Cre. White represents transferred WT CD45.1 BMDC-cDC2s. Gray represents the frequencies of CD45.2-cDC2s in the CNS. (K and L) cDCs in the CNS were defined as MHCII⁺CD11c⁺CD11b⁺CD26⁺ and stained for CD45.1 to define donor and recipient populations. Percentage of donor and recipient cDC2 in the CNS of WT (K) or Stat4^fl/flCD11c^Cre mice (L). In (E–J) (n = 9 or 7 per mouse group), (K and L) (n = 5). Data represent mean ± SEM and represent two independent experiments. One-way ANOVA with Tukey multiple comparison test was used to compare means in all the plots. ^**P < 0.01. ns, indicates not significant. BM: bone marrow, BMDCs: bone marrow–derived DCs. Illustration in (A) was created with BioRender.

Fig. 4.

Mapping DC populations in the CNS at the peak of EAE. (A–G) scRNA-seq analysis of CNS DC populations from EAE mice at day 15 postimmunization. Total leukocytes were isolated from the CNS and pooled from 10 mice (WT) and 18 mice (Stat4^fl/flCD11c^Cre) before FACS-sorting for MHCII⁺CD11c⁺ all populations in WT and Stat4^fl/flCD11c^Cre mice. (A) Heatmap showing the differentially expressed genes in all the clusters. (B) The expression level of Dpp4 in DC clusters. (C) Dot plot showing the expression of DC signature markers in DC clusters. (D) Stacked bar plot showing cell numbers in DC clusters. (E) Violin plots depicted the expression of migratory genes in migratory (mig) cDCs clusters. (F) Heatmaps and (G) violin plots of differentially expressed genes between WT and Stat4^fl/flCD11c^Cre mice in mig cDC subsets. (H and I) Analysis of DC transcriptomes from published scRNA-seq data of human CSF samples from MS patients and controls as described in materials and methods (H) Heatmaps and (I) violin plots of differentially expressed genes between MS and control in human (h) cDC clusters. Max, maximum, Min, minimum.

Fig. 5.

STAT4 promotes cDC migration to the CNS. (A and B) scRNA-seq analysis of chemokine Ccl2, Ccl3, Ccl4, and Ccl5 expression levels in multiple populations in the CNS at the acute phase of EAE. (A) Dot plot showing chemokine expression levels in selected clusters between WT and Stat4^fl/flCD11c^Cre mice. (B) Heatmaps of the deferentially expressed chemokine genes between WT and Stat4^fl/flCD11c^Cre mice. (C and D) Total LNs were purified from the EAE mice at day 14 postimmunization and placed on the upper chamber of the Transwell plate. Chemokines (CCL2, CCL5, and CCL19) were added to the lower chamber. Migrated cells toward the chemokine gradient were counted and FACS-analyzed for DC markers. (C) Scheme of the migration assay. (D) Scatter plots of the total number of migrated cDCs (MHCII⁺CD11c⁺CD26⁺) from WT and Stat4^fl/flCD11c^Cre mice. Total leukocytes in the CNS of WT and Stat4^fl/flCD11c^Cre mice were purified at the EAE acute phase and analyzed for chemokine receptor (CCR2, CCR5, and CCR7) expressions on cDC2s. (E) Scatter plot of the chemokine receptor expressions on the surface of cDC2s (MHCII⁺CD11c⁺CD11b⁺CD26⁺) WT and Stat4^fl/flCD11c^Cre mice. Data represent mean ± SEM and represent two independent experiments (n = 6 or n = 5 per mouse group). An unpaired t test was used to compare means in all the plots. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001, and ^***P < 0.0001. ns, indicates not significant.

Fig. 6.

The IL-23-STAT4 pathway regulates CD11c⁺ cDCs in the CNS. (A) Scheme of Il23r^−/− GFP- or WT CD45-1 BMDCs adoptive transfer to the WT and Stat4^fl/flCD11c^Cre recipient mice. BMDCs were adoptively transferred to the recipient mice on days −1, 5, 3, 7, and 9 of EAE. (B) EAE clinical score and (C) scatter plot of a cumulative score of the recipient mice after transferring Il23r^−/− GFP or WT CD45.1 BMDCs. (D) Representative flow plots and (E) scatter plots of the frequencies of BMDC-GFP cDC2s in the CNS of the WT and Stat4^fl/flCD11c^Cre recipient mice. (F) Representative flow plots and (G) scatter plots of the frequencies of BMDC-CD45.1 cDC2s in the CNS of the WT and Stat4^fl/flCD11c^Cre recipient mice. Representative flow plots and scatter plots of the frequencies of the transferred (H) Il23r^−/−GFP cDC2 (I) or WT CD45.1 cDC2 BMDCs in the recipient Stat4^fl/flCD11c^Cre in the LNs. (J) Scatter plots of the frequencies and the total number of Il23r^−/− GFP or WT CD45.1 BMDCs in the recipient Stat4^fl/flCD11c^Cre in the LNs or (K) CNS. Data represent mean ± SEM and represent two independent experiments (n = 10 to 8 per mouse group). One-way ANOVA with Tukey multiple comparison test was used to compare means in all the plots. ^***P < 0.001 and ^****P < 0.0001. ns, indicates not significant. Illustration in (A) was created with BioRender.