Cutting Edge: MicroRNA-223 Regulates Myeloid Dendritic Cell–Driven Th17 Responses in Experimental Autoimmune Encephalomyelitis Free

Blood-derived myeloid cells make up a significant proportion of the infiltrating perivascular leukocytes in CNS lesions in multiple sclerosis (MS) (1) and its animal model, experimental autoimmune encephalomyelitis (EAE) (2, 3). These APCs have the ability to phagocytize myelin debris and to secrete cytokines that promote the expansion and polarization of CD4⁺ Th1 and Th17 T cells critical in the initiation and progression of EAE (2).

Ly6c^hiCD11b⁺CD11c⁺ monocyte-derived dendritic cells (mo-DCs) are a prominent component of CNS infiltrates cells in EAE (2, 3). mo-DCs arise from monocytes following their migration into inflamed tissues (4). Myeloid DCs (mDCs) are responsible for initiating epitope spreading during EAE within the CNS, inducing the differentiation of Th17 cells (2, 5, 6). In addition, Ly6C^hi dendritic cells are critical for Th17 polarization (7).

MicroRNAs (miRNAs) are small noncoding RNAs of 17–25 nt that regulate gene expression by inducing mRNA degradation or interfering with mRNA translation (8). More than 60% of protein-coding genes are regulated by miRNAs (9). miRNAs regulate various biological processes, including immune cell lineage commitment, differentiation, maturation, and maintenance of immune homeostasis (10), and are involved in the pathogenesis of cancer and inflammatory and autoimmune diseases (11).

Expression studies on tissues from MS patients identified miR-223 as one of the top upregulated miRNAs (12, 13). miR-223 expression, which is mainly confined to myeloid cells (14), was shown to regulate the proliferation and activation of neutrophils (15) and M1 macrophages (16). We demonstrate that miR-223 regulates EAE pathogenesis by controlling mDC-induced activation of pathogenic Th17 cells.

Female wild-type (WT) C57BL/6 and miR-223^−/− mice were purchased from The Jackson Laboratory and housed in the Northwestern Center for Comparative Medicine specific pathogen–free facility. Identical results were obtained using littermate or WT C57BL/6 mice as controls.

Eight- to ten-week-old female mice were injected s.c. with 100 μl an emulsion containing 200 μg Mycobacterium tuberculosis H37Ra (Difco, Detroit, MI) and 200 μg MOG_35–55 (Genemed, San Francisco, CA) distributed over three sites on the flank. For transfer EAE, spleen and inguinal lymph nodes collected on day 8 post-MOG_35–55 were reactivated with 20 μg/ml MOG_35–55 and 10 ng/ml recombinant murine IL-12 (R&D Systems) for 72 h. A total of 20 × 10⁶ cells was transferred to recipient mice. In both models, 200 ng pertussis toxin (List, Campbell, CA) was administered i.p on days 0 and 2. Two or three independent experiments were performed, with at least five mice/group. Clinical signs of EAE were assessed daily, as previously described (2).

CNS-immune cells were isolated by Percoll gradient centrifugation from homogenized combined brain and spinal cords and enumerated as previously described (2).

FcRs were blocked using anti-mouse CD16/32 (0.25 μg; eBioscience) and cells were stained for 30 min at 4°C using the specified Abs (BD Biosciences or eBioscience). Cytokine expression was determined in activated T cells (18 h with 1 μg/ml ionomycin and 20 ng/ml PMA) and APCs (18 h with 10 ng/ml Escherichia coli 0111:B4 LPS) (Sigma) in the presence of 2 μg/ml brefeldin A (Sigma). Cells were acquired on a BD Canto II and analyzed using BD FACSDiva version 6.1.

Purified CD4⁺ T cells (≥90%) were cultured with CD4⁻ WT or knockout (KO) splenocytes (as APCs) plus α-CD3 (1 μg/ml) in Th1-promoting (200 U/ml IL-2, 10 ng/ml IL-12, 1 μg/ml anti–IL-4) or Th17-promoting (10 ng/ml TGF-β, 50 ng/ml IL-6, 10 ng/ml IL-23, 1 μg/ml anti–IFN-γ, 1 μg/ml anti–IL-4, 1 μg/ml anti–IL-2) conditions and analyzed for cytokine expression by flow cytometry.

CD11b⁺Ly6C^hi monocytes were purified (>95%) from bone marrow via magnetic cell sorting (Miltenyi Biotec).

CD11b⁺ myeloid cells, B220⁺ B cells, and CD90⁺ T lymphocytes were purified using magnetic cell sorting (Miltenyi Biotec). Gene expression quantification was performed as previously described (15).

Statistical analyses were performed using GraphPad Prism 5.0. Single comparisons of two means were analyzed by the Student t test. For multiparametric data, two-way ANOVA with a Bonferroni posttest was used. The p values < 0.05 were considered significant.

We observed a significant delay in the onset and severity of MOG_35–55-induced EAE in miR-223–KO versus WT mice (Fig. 1A), concomitant with a reduction in CNS infiltration of CD45^hiCD11b⁺Ly6G⁻CD11c⁺ mDCs and CD45^hiCD11b⁻CD3⁺ CD4⁺ T cells (Fig. 1B). The numbers of macrophages/monocytes, neutrophils, B cells, and plasmacytoid DCs (pDCs) did not differ (Supplemental Fig. 1A). We observed a significant reduction in mo-DCs (Ly6^hi) in the CNS of miR-223–KO mice (Fig. 1C) and reduced percentages and numbers of CD4⁺ IL-17⁺, CD4⁺ GM-CSF⁺, L-17⁺/IFN-γ⁺, and IL-17⁺/GM-CSF⁺ cells (Fig. 1D, Supplemental Fig. 1B), confirming the importance of Ly6C^hi mDCs in CNS inflammation (2, 6, 17).

CD11b⁺ myeloid cells had the highest basal level of miR-223 expression (Fig. 2A), and expression is upregulated during EAE, with comparable levels at disease onset and peak. miR-223 was not expressed in B and T cells. In transfer EAE, disease onset is delayed and severity is significantly reduced in miR-223 recipients of WT MOG_35–55-specific CD4⁺ T cells (Fig. 2B). Collectively, these results indicate a critical role for miR-223 in peripheral and CNS myeloid APCs required for EAE initiation. Interestingly, the fact that miR-223 controls astrocyte expression of CXCL12, a chemokine that attracts CD4⁺ T cells and myeloid cells (18), may provide a mechanistic basis for the reduced disease severity and immune cell infiltration observed in KO mice.

PD-L1, but not PD-L2, expression was significantly higher on mDCs and bone marrow monocytes (Fig. 3A, Supplemental Fig. 2A) and CD40 and CD80 expression was enhanced on splenic macrophage/monocytes, but not pDCs, of MOG_35–55-primed KO mice (Supplemental Fig. 2B). These results are in direct accordance with the fact that miR-223 regulates STAT1 expression (19), which is a crucial regulator of PD-L1 expression on mDCs (20). Spleen mDCs from KO mice expressed significantly lower levels of IL-1β, IL-23p19, and IL-6, but levels of IL-12p40 and TNF were not different than controls (Fig. 3B, Supplemental Fig. 2C). Splenic macrophages/monocytes from miR-223–KO mice had increased expression of TNF, but there were no differences in cytokine expression in splenic pDCs. These data suggest that mDCs from miR-223–KO mice have a reduced capacity to activate T cells as a result of the PD-L1 increase and, in particular, Th17 cells because of the decreased expression of Th17-promoting cytokines (IL-1β, IL-6, and IL-23). These findings are particularly interesting because miR-223 regulates Roquin-1 expression by binding to the 3′ untranslated region (21). Roquin-1 is an important regulator of expression of Th17-related cytokines and Th17 differentiation (22), and its loss leads to overproduction of IL-6 and increases Th17 cell differentiation. Mice expressing an M199R mutation in the Roquin protein (Roquin^san/san mice) develop autoimmune pathologies (23), and Roquin-1 was shown to regulate innate and inflammatory responses in autoimmunity (24). In addition, miR-223 regulates genes involved in the inflammasome pathway, cytokine–cytokine receptor interactions, and signaling via the TLR, JAK-Stat, FcεRI, NOD-like receptor, and NF-κB pathways (25). We are currently assessing which of these mechanisms are involved in miR-223 regulation of mDCs.

Naive CD4⁺ T cells from WT mice were cultured with anti-CD3 and CD4⁻ splenic APCs from naive WT or miR-223–KO mice in Th1- or Th17-driving conditions, and T cell differentiation was assessed at days 1, 3, and 5. In Th1-skewing conditions, although the percentage of IFN-γ⁺ cells was reduced at day 3 with KO APCs, WT and KO APCs induced equivalent numbers of Th1 cells at day 5 (Fig. 4A). In contrast, KO APCs were significantly less capable of inducing differentiation of IL-17–producing and GM-CSF–producing Th17 cells, with virtually no GM-CSF, a cytokine reported to be essential in the pathogenic process of EAE (26), produced at day 5. The low levels of IFN-γ observed in Th17 conditions confirm the specific skewing of the Th17 differentiation.

Collectively, our results uncover a new role for miR-223 in mDCs and emphasize the important role that it plays in the development of CNS inflammation by its ability to regulate induction of Th17 responses by controlling levels of Th17-polarizing cytokines, including IL-1β, IL-6, and IL-23. Additionally, miR-223 may participate in the regulation of monocyte differentiation into mo-DCs in inflamed environments. In the absence of miR-223, C57BL/6 mice exhibit significantly fewer clinical signs of EAE and associated peripherally derived CNS-infiltrating inflammatory cells, confirming the importance of this miRNA in the disease pathology and identifying a potential new autoimmune regulatory target.

We thank Dr. Eva Gottwein and Marc Manzano from Northwestern University for insightful discussions.

The authors have no financial conflicts of interest.