EPCR deficiency ameliorates inflammatory arthritis in mice by suppressing the activation and migration of T cells and dendritic cells

Abstract

Objectives: Endothelial protein C receptor (EPCR) is highly expressed in synovial tissues of patients with RA, but the function of this receptor remains unknown in RA. This study investigated the effect of EPCR on the onset and development of inflammatory arthritis and its underlying mechanisms.

Methods: CIA was induced in EPCR gene knockout (KO) and matched wild-type (WT) mice. The onset and development of arthritis was monitored clinically and histologically. T cells, dendritic cells (DCs), EPCR and cytokines from EPCR KO and WT mice, RA patients and healthy controls (HCs) were detected by flow cytometry and ELISA.

Results: EPCR KO mice displayed >40% lower arthritis incidence and 50% less disease severity than WT mice. EPCR KO mice also had significantly fewer Th1/Th17 cells in synovial tissues with more DCs in circulation. Lymph nodes and synovial CD4 T cells from EPCR KO mice expressed fewer chemokine receptors CXCR3, CXCR5 and CCR6 than WT mice. In vitro, EPCR KO spleen cells contained fewer Th1 and more Th2 and Th17 cells than WT and, in concordance, blocking EPCR in WT cells stimulated Th2 and Th17 cells. DCs generated from EPCR KO bone marrow were less mature and produced less MMP-9. Circulating T cells from RA patients expressed higher levels of EPCR than HC cells; blocking EPCR stimulated Th2 and Treg cells in vitro.

Conclusion: Deficiency of EPCR ameliorates arthritis in CIA via inhibition of the activation and migration of pathogenic Th cells and DCs. Targeting EPCR may constitute a novel strategy for future RA treatment.

Keywords: CIA; RA; Th cells; cytokines; dendritic cells; endothelial protein C receptor.

Figure 1.

The incidence and severity of CIA in EPCR KO and WT mice. Male EPCR KO and WT mice were immunized with chicken type II collagen on days 0 and 21. Clinical signs of arthritis were examined after the second immunization (arthritis induction), until day 28, when mice were euthanized for histological evaluation. (A) Arthritis incidence and average clinical scores in mice (n = 25 mice/group). (B) Histological scores in mice (n = 20 mice/group). SCI: synovitis cell infiltrate; CPGL: cartilage proteoglycan loss; CE: cartilage erosion; VI: vascular invasion from subchondral bone. Data are shown as mean (s.d.). *P < 0.05, **P < 0.01 vs WT. (C) Representative knee joints from mice with CIA stained with toluidine blue. Arrows in upper panel indicate damaged cartilage and arrows in lower panel indicate synovial invasion. Scale bar: 200 µm

Figure 2.

PC/aPC expression/activity in EPCR KO and WT mice. PC/aPC expression/activity in synovial tissues or plasma from normal mice (male, 7 weeks old) or mice with CIA at day 28 after arthritis induction detected by immunostaining (n = 10) or aPC activity assay. (A) Representative image of PC/aPC expression in synovial tissues, detected by immunostaining. Arrows indicate growth plates in normal mice. Scale bar: 200 µm. (B, C) aPC activity in synovial tissues (n = 10) or plasma from normal (n = 5) or CIA mice (n = 20), measured by aPC activity assay. Data are shown as mean (s.d.). aPC activity in synovial tissues is expressed as ng/100 mg tissue and in plasma as ng/ml of recombinant human aPC (units). *P < 0.05 vs WT

Figure 3.

Plasma cytokines and anti-CII antibodies and tissue Th1/2/Th17/Treg cells in mice with CIA. (A) IL-17, IFN-γ, MMP-3, TGF-1β and anti-CII antibodies IgG1, IgG2a and IgG2b in plasma from EPCR KO and WT mice with CIA at day 28 after arthritis induction, measured by ELISA (n = 20). (B) Gate strategies for Th and Treg cell detection by flow cytometry. (C) Th1, Th2, Th17 and Treg cells and (D) chemokine receptors CXCR3 (CD183), CXCR5 (CD185), CCR6 (CD196) and CCR7 (CD197) on CD3⁺CD4⁺ T cells from blood, spleen, lymph node (LN) and synovium of WT and EPCR KO mice at day 28 after arthritis induction, detected by flow cytometry (n = 6). Data are shown as mean (s.d.). *P < 0.05, **P < 0.01 vs WT

Figure 4.

Lymphocyte proliferation and Th phenotypes in WT or EPCR KO spleen cells in vitro. Spleen cells isolated from WT or EPCR KO mice were labelled with CFSE or treated with mouse EPCR blocking antibody RCR16 or non-blocking antibody RCR20 (5 µg/ml), aPC (10 µg/ml) and LPS (100 ng/ml) for 24 h. Lymphocyte proliferation or Th phenotypes were detected by flow cytometry. (A) Lymphocyte proliferation evaluated by the intense fluorescence of CFSE after a 5 day incubation. (B, C) The percentages of Th1, Th2, Th17 and Treg cells in CD3⁺CD4⁺ T cells within spleen cells. (D) The percentage of IFN-γ (Th1), IL-4 (Th2) and IL-17 (Th17) producing CD3⁺CD4⁺ T cells. (E) EPCR expression by CD3⁺CD4⁺ T cells or Th1/Th2/Th17/Treg cells. (F) Gating strategies of EPCR on T cells. Data are shown as mean (s.d.) (n = 4). *P < 0.05, **P < 0.01 vs controls

Figure 5.

The in vitro and in vivo effects of EPCR on DCs. (A) The gating strategies for flow cytometry detection of DCs and DC subsets and the levels of CD40, CD80, CD86 and IAb on DCs within blood, lymph nodes (LN), spleen and synovium (Syn) from WT or EPCR KO mice with CIA. (B) DCs and their subsets cDCs, mDCs and pDCs within blood, LN, spleen and Syn from WT or EPCR KO mice with CIA and (C) levels of CD40, CD80, CD86 and IAb expression on DCs in these tissues (n = 5 mice). (D) Levels of CD40, CD80, CD86 and IAb on bone marrow (BM)-generated WT or EPCR KO mature BMDCs in response to RCR16 treatment for 24 h (n = 3). Data are shown as mean (s.d.). *P < 0.05, **P < 0.01. (E) MMP-2 and MMP-9 in the culture supernatants of BMDCs and spleen cells after 24 h incubation, detected by zymography. ST: MMP-2 and MMP-9 standard

Figure 6.

The effect of EPCR on immune cells from WT or EPCR KO mice with CIA and the effect of blocking EPCR on PBMCs from patients with RA. (A-C) The levels of immune cells including CD3, CD4 and CD8 T cells, NK cells, NK T cells (NK/T), monocytes (Mono), myeloid-derived suppressor cells (MDSCs) and B cells within blood, lymph node (LN), spleen and synovium (Syn) from WT or EPCR KO mice (n = 5 mice) with CIA, detected by flow cytometry. (D) EPCR expression on CD3 and CD4 T cells within PBMCs from patients with RA or matched HCs (n = 6). (E) The percentages of Th1, Th2, Th17 and Treg cells in CD3⁺CD4⁺ T cells within RA PBMCs treated with human EPCR blocking antibody RCR252 (R252), non-blocking antibody RCR92 (R92), TNF inhibitor ADA and ETA (all 10 µg/ml) for 24 h, detected by flow cytometry (n = 4). (F) MMP-2 and MMP-9 and IFN-γ in culture supernatants of RA PBMCs after 24 h treatment (n = 4). MMP-2 and MMP-9 were detected by zymography and IFN-γ by ELISA. Data are shown as mean (s.d.). ST: MMP-2 and MMP-9 standard; Con: control. *P < 0.05, **P < 0.01, ***P < 0.001 vs WT, HCs or Con