Integrin α3 promotes TH17 cell polarization and extravasation during autoimmune neuroinflammation

Abstract

Multiple sclerosis (MS) is an autoimmune disease of the central nervous system (CNS) caused by CNS-infiltrating leukocytes, including T_H17 cells that are critical mediators of disease pathogenesis. Although targeting leukocyte trafficking is effective in treating autoimmunity, there are currently no therapeutic interventions that specifically block encephalitogenic T_H17 cell migration. Here, we report integrin α3 as a T_H17 cell-selective determinant of pathogenicity in experimental autoimmune encephalomyelitis. CNS-infiltrating T_H17 cells express high integrin α3, and its deletion in CD4⁺ T cells or Il17a fate-mapped cells attenuated disease severity. Mechanistically, integrin α3 enhanced the immunological synapse formation to promote the polarization and proliferation of T_H17 cells. Moreover, the transmigration of T_H17 cells into the CNS was dependent on integrin α3, and integrin α3 deficiency enhanced the retention of CD4⁺ T cells in the perivascular space of the blood-brain barrier. Integrin α3-dependent interactions continuously maintain T_H17 cell identity and effector function. The requirement of integrin α3 in T_H17 cell pathogenicity suggests integrin α3 as a therapeutic target for MS treatment.

Fig. 1.. Integrin α3 is exclusively expressed by Th17 cells.

(A) Volcano plot showing DEGs in Junb^+/+ CD4-Cre (WT) and Junb^fl/fl CD4-Cre (Junb KO) Th17 cells polarized in vitro for 48 h. Genes with FC > 2 and FDR < 0.05 are in orange, and select migration-related genes are labeled and highlighted in blue. Triangles indicate genes with FDR < 10⁻²⁰ or Log₂(FC) > 5 or < −5. Expression of integrin α3 in 72 h polarization cultures was detected by (B) qPCR (n = 4 mice/subset, pooled from 3 independent experiments), (C) Western blot (representative of 3 independent experiments), and (D) flow cytometry (n = 7–8 mice/subset, pooled from 4 independent experiments). KO Ctrl, negative staining control is Itga3-deficient Th1 culture (Itga3^fl/fl CD4-Cre). mRNA expression is presented relative to Actb expression and normalized over naïve CD4⁺ T cells. (E) Integrin α3 expression for day 6 cultures of human (h) naïve CD4⁺ T cells polarized as indicated. n = 3 donors/subset, pooled from 3 independent experiments. (F) Bar plot of RPKM expression values of Itga3 transcripts in TF-knock-out (TF KO, Batf^−/−, Irf4^−/−, Stat3^fl/fl CD4-Cre, Rorc(t)^GFP/GFP) vs wild-type (WT) 48 h Th17 cell polarization cultures (differential expression, FDR < 10⁻⁵). (G) ChIP-seq tracks for TFs and p300 peaks enriched at Itga3-proximal putative enhancers (shaded green) in 48 h Th17 cell cultures. gMFI presented as fold over Naïve (D) or Isotype (E). Data are summarized as mean ± SEM. One-way ANOVA test. *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001; ns, not significant. Datasets for (f) and (g) from GSE40918 and GSE98414.

Fig. 2.. CNS-infiltrating Th17 cells express integrin α3 during EAE pathogenesis.

Flow cytometric analysis of integrin α3 expression on CD4⁺ T cell subsets isolated from the spinal cord of EAE-induced Itga3^fl/fl CD4-Cre (KO^CD4) or wildtype control mice (A) at the onset (day 13, n = 5–6 mice/genotype) or (B) at the peak (day 14, n = 5 mice/genotype). CD4⁺ T cell subsets: Th17, IL-17A⁺ IFN-γ⁻; Th17.1, IL-17A⁺ IFN-γ⁺; Th1, IL-17A⁻ IFN-γ⁺; Treg, IL-17A⁻ Foxp3⁺. (C) Flow cytometric analysis of integrin α3 expression on ZG⁺ or ZG⁻ IFN-γ-producing CD4⁺ T cells isolated from the spinal cord of WT^Il17a R26^ZG or KO^Il17a R26^ZG (KO Ctrl) mice at the peak of EAE (day 17, n = 5–6 mice/genotype). EAE inductions in (A, B, and C) were performed independently, once with biological replicates. (D) Plot showing select migration-related DEGs in CSF CD4⁺ T cells versus blood CD4⁺ T cells from MS patients. Dashed lines represent FDR = 0.05 and fold change = −1.5 or 1.5. RNA-seq data is replotted from (64). gMFI presented as fold over KO (A and B). Data are summarized as mean ± SEM (A and B) or represented by lines connecting data from the same mouse (C). Unpaired Student’s t-test (A, B, and C) and paired Student’s t-test (C). *, p < 0.05; **, p < 0.01; ***, p < 0.001; ns, not significant.

Fig. 3.. Integrin α3 expression is induced by the IL-6-Stat3 pathway.

(A) Flow cytometric analysis of integrin α3 expression by CD4⁺ T cells cultured for 72 h in the presence of IL-6 (0–5.0 ng/ml). n = 3 mice, representative of 3 independent experiments. (B) qPCR analysis of Itga3 transcript in Stat3^+/+ CD4-Cre and Stat3^fl/fl CD4-Cre naïve CD4⁺ T cells cultured for 18 h in the presence of IL-6 (10 ng/ml). Naïve CD4⁺ T cells from Itga3^fl/fl CD4-Cre mice serve as a negative Itga3 control. mRNA expression is presented relative to Actb expression and normalized over Stat3^+/+ CD4-Cre naïve CD4⁺ T cells (Naïve). n = 4 mice/genotype, pooled from 3 independent experiments. ND, not detectable. (C) Flow cytometric analysis of integrin α3 in naïve CD4⁺ T cells cultured for 72 h in the presence of indicated cytokines (IL-6, 10 ng/ml; IL-1β, 20 ng/ml; IL-23, 25 ng/ml). n = 3 mice, representative of 3 independent experiments. (D) Flow cytometric analysis of integrin α3 expression in pTh17 cells cultured for 72 h in the presence of TGF-β (0–0.6 ng/ml). n = 7 mice, pooled from 2 independent experiments. r, Pearson’s correlation coefficient calculated for cultures treated with 0–0.5 ng/mL of IL-6 (A) and with 0–0.6 ng/mL of TGF-β (D). Naïve CD4⁺ T cells isolated from Itga3^fl/fl CD4-Cre mice and cultured without cytokines served as a negative staining control (KO^CD4 Ctrl) (A and D). gMFI presented as fold over KO^CD4 cells cultured under identical conditions (A, C, and D). Data are summarized as mean ± SEM. Unpaired Student’s t-test (B) or one-way ANOVA test (C). *, p < 0.05; **, p < 0.01; ***, p < 0.001; ns, not significant.

Fig. 4.. Deletion of integrin α3 in CD4⁺ T cells attenuates EAE.

(A) Clinical score of EAE from WT (n = 10) and KO^CD4 (n = 11) mice, pooled from 2 independent experiments. The area under the curve (AUC) values calculated from EAE clinical scores were compared. (B) Representative flow cytometric analysis of CD4⁺ T cells isolated from the spinal cords of EAE-induced mice at disease peak (day 16). (C) The frequencies and numbers of CD4⁺ T cell subsets in the spinal cords of EAE-induced mice at disease peak (day 16, n = 5–6 mice/genotype, representative of 2 independent experiments). CD4⁺ T cell subsets: Th17, IL-17A⁺; Th1, IFN-γ⁺; Treg, Foxp3⁺. (D) Spinal cord infiltration of CD4⁺ T cells at disease peak (day 16) imaged by confocal microscopy. n = 3 mice/genotype, representative of 2 independent experiments. (Left) Schematic image of perivascular cuff depicts the accumulation of leukocytes in the perivascular space between basement membranes. The perivascular space is surrounded by the endothelial and the parenchymal basement membranes that are visualized by pan-laminin (green). (Right) Representative immunofluorescence images selected from three different regions of spinal cord sections. CD4, red; pan-laminin, green. Perivascular spaces in the immunofluorescence images are marked by yellow arrowheads. (E) Clinical score of EAE from WT^Il17a R26^ZG (n = 17) and KO^Il17a R26^ZG mice (n = 12) mice, pooled from 2 independent experiments. (F) Clinical score of passive EAE induced by adoptive transfer of in vitro cultured WT or KO^CD4 2D2 pTh17 cells into Tcra^−/− mice (WT pTh17 cells, n = 8; KO^CD4 pTh17 cells, n = 5, representative of 2 independent experiments). (G) Clinical score of passive EAE induced by adoptive transfer of in vitro cultured WT or KO^CD4 2D2 Th1 cells into Tcra^−/− mice (WT Th1 cells, n = 7; KO^CD4 Th1 cells, n = 5, representative of 2 independent experiments). Data are summarized as mean ± SEM. Mann-Whitney U-test or Unpaired Student’s t-test (D). *, p < 0.05; **, p < 0.01; ***, p < 0.001; ns, not significant.

Fig. 5.. Integrin α3 is required for the polarization and expansion of Th17 cells in vivo.

(A) Schematic of EAE induction in Tcra^−/− mice and adoptive transfer of naïve CD4⁺ T cells. Naïve CD4⁺ T cells obtained from WT 2D2 (CD45.1⁺) and KO^CD4 2D2 (CD45.1⁻) mice were mixed at a 1:1 ratio and adoptively transferred into EAE-induced Tcra^−/− mice. CD4⁺ T cells were loaded with proliferation dye prior to the adoptive transfer. CD4⁺ T cells in iLNs were analyzed at (C) 24 h, (D) 60 h, and (E) 7 days post EAE induction. (B) Plot showing the percentage of KO^CD4 2D2 CD4⁺ T cells within total CD4⁺ T cells in iLNs of EAE-induced or naïve Tcra^−/− mice. n = 3–4 mice, representative of 2 independent experiments. (C) Flow cytometric analysis of CD62L and CD44 expression on CD4⁺ T cells isolated from iLNs of Tcra^−/− mice at 24 h post EAE induction. n = 6 mice, pooled from 2 independent experiments. (D) Flow cytometric analysis of TF expression in CD4⁺ T cells isolated from iLNs of Tcra^−/− mice at 2.5 days post EAE induction. n = 7 mice, representative of 2 independent experiments. Horizontal lines indicate mean values. G0–6, generations 0–6. (E) Flow cytometric analysis of CD4⁺ T cells isolated from iLNs of Tcra^−/− mice at 7 days post EAE induction. n = 4 mice, representative of 2 independent experiments. CD4⁺ T cell subsets: Th17, IL-17A⁺; Th1, IFN-γ⁺; Treg, Foxp3⁺. Data are summarized as mean ± SEM. Mean values (B) or data acquired from the same recipient mouse (C-E) are connected by lines. One-way ANOVA test (B) and paired Student’s t-test (C-E). *, p < 0.05; **, p < 0.01; ****, p < 0.0001; ns, not significant.

Fig. 6.. Integrin α3 promotes T cell stimulation by facilitating T cell-APC interaction.

Flow cytometric analysis of recombinant integrin α3β1 (rVLA-3) binding to (A) cells isolated from iLNs of EAE-induced mice (day 7, n = 5 mice, representative 2 independent experiments) or (B) CD11c⁺ BMDCs (n = 4 mice, pooled from 2 independent experiments). T cells, MHCII⁻ CD3ε⁺; B cells, MHCII⁺ CD3ε⁻ CD19⁺; Dendritic cells, MHCII⁺ CD3ε⁻ CD19⁻ CD11c⁺. Cells incubated without rVLA-3 were used as a negative staining control (Ctrl). WT or KO^CD4 2D2 pTh17 cells were co-cultured with MOG_35–55 peptide-pulsed BMDCs for 30 min (C and D) or 5 min (E). (C) The percentage of 2D2 pTh17 cells making conjugations with BMDCs. n = 3 mice/genotype, pooled from 3 independent experiments. (D) The immunological synapse formation between 2D2 pTh17 cells and BMDCs (WT 2D2 pTh17 cells, n = 115; KO^CD4 2D2 pTh17 cells, n = 90, pooled from 4 independent experiments). (E) Flow cytometric analysis of p-ERK1/2 in 2D2 pTh17 cells. Pooled from 4 independent experiments. Data are summarized as mean ± SEM. Unpaired Student’s t-test (A-D) or two-way ANOVA test (E). *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001; ns, not significant.

Fig. 7.. Integrin α3 promotes the maintenance of Th17 cell identity.

(A) Schematic of EAE induction in Tcra^−/− mice post adoptive transfer of naïve CD4⁺ T cells. Naïve CD4⁺ T cells obtained from WT^Il17a R26^ZG 2D2 (CD45.1⁺) and KO^ll17a R26^ZG 2D2 (CD45.1⁻) mice were mixed at a 1:1 ratio and adoptively transferred into Tcra^−/− mice. EAE was induced on the following day. Organs were harvested at clinical score of 1.5 for the analysis of frequency and cytokine expression profiles of ZG⁺ cells (B-C). 7 days post immunization, ZG⁺ WT^Il17a 2D2 or ZG⁺ KO^Il17a 2D2 cells were FACS-sorted from iLNs to perform RNA-seq (D-G). (B) The frequency of ZG⁺ populations in CD4⁺ T cells of each genotype was analyzed for indicated tissues. n = 6 mice, representative of 2 independent experiments. (C) Cytokine expression profiles of ZG⁺ CD4⁺ T cells of each genotype were analyzed for indicated tissues. n = 6 mice, representative of 2 independent experiments. (D) Volcano plot showing DEGs in ZG⁺ KO^Il17a 2D2 vs ZG⁺ WT^Il17a 2D2 RNA-seq analysis. Genes considered significant (FC > 1.2 and FDR < 0.05) are in orange, and select genes are labeled and highlighted in blue. FDR capped at 10⁻¹⁵. Log₂ (FC) capped at −2.5 or 2.5. Triangles indicate genes with FDR < 10⁻¹⁵ or Log₂ (FC) > 2.5 or < −2.5. (E) Bar plots showing the RNA-seq expression level (TPM) of select genes. (F) GSEA plots enriched in various gene sets for ranked genes in ZG⁺ KO^Il17a 2D2 vs ZG⁺ WT^Il17a 2D2 RNA-seq. (G) Ingenuity pathway analysis for DEGs (FDR < 0.05) in ZG⁺ KO^Il17a 2D2 vs ZG⁺ WT^Il17a 2D2 RNA-seq. Dashed line represents p-value = 0.05. Data acquired from the same recipient mouse are connected by lines (B and C). Paired Student’s t-test (B and C). *, p < 0.05; **, p < 0.01; ns, not significant.

Fig. 8.. Integrin α3 promotes the infiltration of Th17 cells into the CNS.

(A) The percentage of WT^Il17a R26^ZG 2D2 (CD45.1⁺) and KO^Il17a R26^ZG 2D2 (CD45.1⁻) cells among ZG⁺ CD4⁺ T cells for the transfer experiment performed as in Fig. 6B was analyzed for indicated tissues. n = 6 mice, representative of 2 independent experiments. (B) The percentage of WT and KO^CD4 CD4⁺ T cells was analyzed for indicated tissues at day 14 post adoptive transfer of in vitro-cultured WT (CD45.1⁺) or KO^CD4 (CD45.1⁻) 2D2 pTh17 mixed at a 1:1 ratio (Recipient mice: Tcra^−/−, n = 6, clinical score: 1.5 ± 0.55, representative of 3 independent experiments). (C) Flow cytometric analysis of rVLA-3 binding to b.End3 cells (pooled from 3 independent experiments). (D) Fluorescent live cell images showing rVLA-3 binding to b.End3 cells, representative of 2 independent experiments. Phase contrast is overlaid to highlight cell morphology. (E) Summary of transwell migration assay showing the percentage of Th17 or pTh17 cells that migrated to the bottom chamber filled with complete RPMI 1640 media after 6 h. n = 4 mice/genotype, pooled from 4 independent experiments. (F) Summary of transwell migration assay showing the percentage of hTh17 cells that migrated to bottom chamber filled with complete RPMI 1640 media after 6 h. n = 3 donors, pooled from 3 independent experiments. Data acquired from the same recipient mouse are connected by lines (WT^Il17a R26^ZG 2D2 of A and B). Data are summarized as mean ± SEM. Unpaired Student’s t-test (B-D) or paired Student’s t-test (A and B: between different organs). Two-way ANOVA test (E and F). *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001; ns, not significant.