Glatiramer acetate (GA; Copaxone) is a copolymer therapeutic that is approved by the Food and Drug Administration for the relapsing-remitting form of multiple sclerosis. Despite an unclear mechanism of action, studies have shown that GA promotes protective Th2 immunity and stimulates release of cytokines that suppress autoimmunity. In this study, we demonstrate that GA interacts with murine paired Ig-like receptor B (PIR-B) on myeloid-derived suppressor cells and suppresses the STAT1/NF-κB pathways while promoting IL-10/TGF-β cytokine release. In inflammatory bowel disease models, GA enhanced myeloid-derived suppressor cell–dependent CD4+ regulatory T cell generation while reducing proinflammatory cytokine secretion. Human monocyte-derived macrophages responded to GA by reducing TNF-α production and promoting CD163 expression typical of alternative maturation despite the presence of GM-CSF. Furthermore, GA competitively interacts with leukocyte Ig-like receptors B (LILRBs), the human orthologs of PIR-B. Because GA limited proinflammatory activation of myeloid cells, therapeutics that target LILRBs represent novel treatment modalities for autoimmune indications.

Glatiramer acetate (GA) is a Food and Drug Administration–approved peptide-based drug for patients with the relapsing-remitting form of multiple sclerosis (MS) and ameliorates models of autoimmunity, including autoimmune encephalomyelitis (1), colitis (2), and graft-versus-host disease (3, 4). Originally designed to compete for peptide binding to MHC class II (MHC-II) (5, 6), GA was hypothesized to act as a decoy Ag to promote anergy or tolerance (7, 8). In several studies, GA showed preferential activation of Th2 lineage, enhanced regulatory T cell (Treg) function, and limited CD8+ T cell effector function (9). Later studies showed that GA directly promotes alternative activation of monocytes regardless of T cell myelin recognition (10). Thus, GA-induced type II monocytes were sufficient to ameliorate outcomes in autoimmune encephalomyelitis models. Others demonstrated that GA inhibited inflammatory cytokine secretion by monocytes and dendritic cells (1113), suggesting that GA has therapeutic benefit, in part, by suppressing innate immunity.

Classically activated (M1) macrophages generate a proinflammatory response to LPS, IFN-γ, and GM-CSF and activate NF-κB, STAT1, and STAT5, respectively. Conversely, alternatively activated (M2) macrophages are immunoregulatory in response to IL-4, IL-13, and M-CSF and preferentially activate STAT3 and STAT6 (14). In mice, monocytic Ly6C+ myeloid-derived suppressor cells (MDSC) display similar plasticity and exhibit M1/M2 dichotomy (15). MDSC play important roles in the regulation of the immune response during infection, malignancy, transplantation, and various immune disorders (1618). PIR-B is a murine receptor whose expression is restricted to B cells and myeloid cells. PIR-B binds MHC class I (MHC-I) as well as angiopoietin-like (ANGPTL) family members (1921). PIR-B signaling promotes SHP1/2 phosphatase activity, resulting in attenuated cell activation, suppressed classical immunity, and as we previously showed, the development and function of MDSC (22). Because GA has been implicated in MHC interactions, we hypothesized that GA may alter signaling downstream of the PIR-B on MDSC.

GA (Teva Neuroscience), anti–PIR-A/B (6C1; BD Biosciences), and anti–PIR-B (R&D Systems, Santa Cruz Biotechnology) were used. GA-FITC was synthesized by desalting GA using Zeba columns (Pierce) followed by FITC conjugation (Pierce). PHA, anti–Gr-1, anti-CD115, anti-F4/80, anti-CD11b, and isotype-matched Abs were purchased from eBioscience. MCA26 (23) and Lewis lung carcinoma cells (no. CRL-1642; American Type Culture Collection) were s.c. implanted in female BALB/c and C57BL/6 mice, respectively. Wild-type (WT) mice were obtained from the Jackson Laboratory (Bar Harbor, ME). Colonies of hemagglutinin (HA)-TCR transgenic (Tg) mice (24), and PIR-B–deficient mice (22) were established from the mice generously provided by Drs. C.A. Bona (Icahn School of Medicine at Mount Sinai, New York City, NY) and T. Takai (Tohoku University, Sendai, Japan). Animals were handled in accordance with the institutional animal guidelines.

MCA26 tumor-bearing BALB/c mice were used to generate murine MDSC in vivo as previously described (23). Briefly, mice with tumor sizes >10 × 10 mm2 were sacrificed, and MDSC were enriched from total bone marrow cells by Percoll gradient centrifugation (GE Healthcare) as previously reported (25). Cells banding at 50–60% were labeled with anti-CD115 PE (eBioscience), followed by magnetic bead positive selection using anti-PE microbeads (Miltenyi Biotec). For in vitro generation of MDSC, mouse bone marrow cells were cultured with 50 ng/ml GM-CSF and 50 ng/ml IL-6 for 5 d.

Suppressive activity of MDSC was assessed in a peptide-mediated proliferation assay of TCR Tg T cells (26). Total splenocytes (1 ☓ 105) from HA-TCR mice were cocultured with serial dilutions of irradiated MDSC in the presence of HA peptide (110SFERFEIFPKE120; Washington Biotechnology) at 0.5 μg/ml. Proliferation was detected by [3H]thymidine incorporation and measured by 1450 MicroBeta TriLux liquid scintillation counter (Wallac). Treg induction assays were performed as described previously (22). Total splenocytes (4 ☓ 106) from HA-TCR mice were cocultured with irradiated MDSC (1 ☓ 106) in the presence of HA peptide and GA (50 μg/ml) where indicated. The percentage of Treg was assessed using FACSCanto II or FACSFortessa (BD Biosciences) multiparameter flow cytometer 5 d later using fluorochrome-conjugated anti-CD4, anti-CD25, and anti-Foxp3 Abs (eBioscience). FACS data were analyzed using FlowJo software (Tree Star).

Culture supernatants were collected for measurement of mouse or human cytokines (IL-6, IL-10, IL-23, IFN-γ, TNF-α, TGF-β1, and IL-17A) by Ready-SET-Go! ELISA kit (eBioscience) according to the manufacturer’s recommendations.

MDSC were stimulated with GA for 2 h, followed by treatment with vehicle, LPS (100 ng/ml), or IFN-γ (20 ng/ml) for an additional 30 min. Total cell lysates were prepared for immunoblotting. p–NF-κB, p-p38, p-STAT1, p-SHP1, β-actin, tubulin (Cell Signaling Technology), FITC (Thermo Fisher Scientific), and PIR-B (R&D Systems) detection Abs were used according to the manufacturer’s recommendations. For immunoprecipitation, total cell lysates from CD115+ MDSC were incubated with PBS vehicle control or GA-FITC, followed by anti-FITC Ab detection and protein G bead pulldown according to the manufacturer’s recommendations (Thermo Fisher Scientific). The precipitates and total lysates were analyzed by immunoblot as previously described (22).

Colitis was induced in C57BL/6 mice by feeding plain water or water containing 3.5% dextran sulfate sodium (DSS) (Sigma-Aldrich) for 7 d, and the body weight was followed. Colitis groups were injected daily with PBS (0.1 ml), GA (150 μg/mouse, s.c.), and two adoptive transfers of MDSC (5 × 106 cells per mouse) on days 1 and 6 where indicated. All animals were graded daily with observations expressed as a clinical score. The clinical score was determined by blinded independent analysis of mice. Weight loss of 0, 1–5, 5–10, 10–20, and >20% received a score of 0, 1, 2, 3, and 4, respectively. For stool consistency, normal pellets, soft stool, and liquid stool received scores of 0, 2, and 4, respectively. Bleeding from the anus described as absent, present, or gross presence received scores of 0, 2, and 4, respectively. The average of the three scores was used to calculate clinical score. On day 12, mice were sacrificed and the entire colons were removed. Colon length was measured, and pathology scores were determined from 10% formalin-fixed H&E-stained histology sections. Pathology of the large intestine was scored by blinded independent analysis. Grade 0 was assigned when no changed were observed; grade 1 indicated minimal infiltration of the lamina propria and/or milk mucosal hyperplasia; grade 2 indicated mild infiltration of the lamina propria with extension into the submucosa, focal erosions, mild hyperplasia, and visible mucin depletion; grade 3 indicated mild to moderate infiltration of the lamina propria and submucosa, transmural ulceration, and moderate mucosal hyperplasia and mucin depletion; grade 4 indicated transmural infiltration with ulceration, increased hyperplasia and mucin depletion, and multifocal crypt necrosis; grade 5 indicated transmural infiltration with ulceration, widespread crypt necrosis, and loss of intestinal structure and glands.

PBMC were isolated from healthy human donor blood buffy coats (New York Blood Center) by Lymphoprep gradient centrifugation (STEMCELL Technologies), according to the manufacturer’s instructions. Myeloid cells were positively selected from PBMC using anti-human CD33 microbeads (Miltenyi Biotec), then cultured in complete RPMI 1640 medium for 5 d in the presence of GA and 50 ng/ml M-CSF (PeproTech) or 50 ng/ml GM-CSF (PeproTech) where indicated. Five days later, viable cells were analyzed by flow cytometry using anti-CD33, anti-CD14, anti-CD16, anti-CD163, anti-CD206, anti-CD86, and isotype-control matched Abs (eBioscience). An aliquot of the 5-d culture was restimulated with LPS (Sigma-Aldrich) at 50 ng/ml for 16 h, followed by ELISA analysis of the supernatant.

2B4 cells expressing NFAT-GFP and DAP12 transgenes were a gift from H. Arase (Osaka University, Suita, Japan). Primers containing an in-frame XhoI restriction site flanking the extracellular domain of LILRB1, LILRB2, LILRB3, and LILRB4 genes were used for PCR amplification from cDNA plasmids (GE Dharmacon). LILRB extracellular domain was cloned into pMXs-Puro retroviral vector (gift from H. Arase) containing an N-terminal SLAM signal peptide and C-terminal murine PILR transmembrane domain separated by XhoI cloning site. pMXs-Puro LILRB1, LILRB2, LILRB3, LILRB4 retroviral constructs were generated by XhoI (New England Biolabs) digestion of PCR fragment and XhoI plus calf intestinal phosphatase (New England Biolabs) digestion of pMXs-Puro vector. DNA fragments were PCR purified or gel-extraction purified (Qiagen), followed by T4 DNA Ligase (New England Biolabs) ligation and transformation into DH5α Competent Cells (Life Technologies) according to manufacturers’ recommendations. pMXs-Puro LILRB-transduced 293 MMLV packaging cell lines were used to generate infectious virus to transduce 2B4 cells over 48 h. Positive transduced cells were isolated using PE-conjugated anti–LILRB1-, anti–LILRB2-, anti–LILRB3-, and anti–LILRB4 (eBioscience)-labeled cells, followed by FACS sorting (BD FACSAria). Nontransduced, LILRB1-, LILRB2-, and LILRB3-transduced 2B4 reporter cells were stimulated with anti-CD3 (eBioscience), anti-LILRB1 (R&D Systems), anti-LILRB2 (BioLegend), or anti-LILRB3 (R&D Systems). LILRB4 Ab (9F4) was internally developed by hybridoma cloning following LILRB4 immunization of mice. Abs were coated onto 96-well tissue culture plates, and cells were stimulated 16 h in the presence of GA followed by measurement of GFP by flow cytometry.

Statistical analyses were performed using Student t test. The results are presented as mean ± SD. A p value <0.05 was considered to be statistically significant.

Because GA was previously shown to limit TNF-α and IL-12 release in innate immune cells, we investigated whether GA suppresses M1-like MDSC maturation. Monocytic MDSC isolated from WT tumor-bearing mice were stimulated ex vivo in the presence of IFN-γ or LPS. Two hours of GA pretreatment was sufficient to suppress STAT1, p38, and NF-κB phosphorylation (Fig. 1A). GA suppressed LPS-induced TNF-α and IL-6 production while enhancing IL-10 and TGF-β regulatory cytokines (Fig. 1B). We previously showed that PIR-B modulates MDSC polarization by inhibiting LPS and IFN-γ mediated activation of MAPK, NF-κB, and STAT1 (22). Because GA variably binds murine MHC-I and MHC-II (6), we hypothesized that GA may alter myeloid cell activation by modulating PIR-B signaling. PIR-B deficiency skews MDSC maturation to an M1 phenotype by enhancing TNF-α secretion while lowering IL-10 (22). We observed that ex vivo GA-treated WT MDSC produced less TNF-α and more IL-10; no significant change in cytokine levels was observed in GA-treated PIR-B−/− MDSC (Fig. 1C), and GA treatment did not alter expression of costimulatory or inhibitory receptors (data not shown). To determine whether GA altered the myeloid cell phenotype by binding PIR-B, we analyzed interactions between GA and PIR-B. FITC-conjugated GA specifically stained WT MDSC, similarly to anti–PIR-B or FLAG-tagged recombinant ANGPTL2, a natural ligand of PIR-B (21). Unconjugated GA blocked this interaction in WT but not PIR-B−/− MDSC (Fig. 1D). The interaction of GA with PIR-B was further confirmed by coimmunoprecipitation. Anti-FITC magnetic beads bound to FITC-conjugated GA pulled down PIR-B from WT MDSC lysates (Fig. 1E). PIR-B contains multiple cytoplasmic ITIM motifs known to associate with SHP-1 (27). Immunoblotting for phosphorylated SHP-1 (Y564) on anti–PIR-B immunoprecipitated WT MDSC lysates demonstrated acute activation of SHP-1 on PIR-B in response to GA but not from PIR-B−/− MDSC lysate (Fig. 1F). We concluded that GA binds PIR-B on MDSC to inhibit classical activation of proinflammatory cytokine release.

FIGURE 1.

GA restricts classical activation in MDSC via interaction with PIR-B. (A) WT MDSC treated with GA (5 μg/ml) for 2 h were stimulated with LPS (100 ng/ml) or IFN-γ (20 ng/ml) for 30 min. Immunoblot of lysates detected the phosphorylation of NF-κB, p38, and STAT1. (B) WT MDSC were incubated with GA in the absence or presence of LPS (100 ng/ml) for 30 h. Secreted TNF-α, IL-6, IL-10, and TGF-β were quantified by ELISA. (C) CD115+ MDSC from Lewis lung carcinoma tumor-bearing WT or PIR-B−/− hosts were incubated in the absence (white bars) or presence of GA (gray bars) for 24 h. Secreted TNF-α and IL-10 were measured by ELISA. Data (B and C) represent mean ± SD (*p < 0.05, **p < 0.005, ***p < 0.0005, calculated by Student t test) (D) MDSC isolated from WT or PIR-B KO hosts were preincubated in the absence (blue line) or presence of GA (red line), washed with cold PBS, and stained for surface PIR-B, GA-FITC, or ANGPTL2-FLAG. FACS data are gated on forward scatter–area (FSC-A), side scatter–area (SSC-A), forward scatter–width (FSC-W), and DAPI cells. Histogram data are overlaid FACS staining controls (gray filled). (E) Lysates from WT splenocytes incubated with GA or GA-FITC were immunoprecipitated using anti-FITC Ab capture. Precipitates were separated by SDS-PAGE followed by Western blot using indicated Abs. (F) WT and PIR-B KO MDSC were treated with GA at time 0, 10, and 30 min, followed by anti–PIR-B immunoprecipitation. Precipitate was separated by SDS-PAGE, followed by probing for phosphorylated SHP-1 and anti–PIR-B.

FIGURE 1.

GA restricts classical activation in MDSC via interaction with PIR-B. (A) WT MDSC treated with GA (5 μg/ml) for 2 h were stimulated with LPS (100 ng/ml) or IFN-γ (20 ng/ml) for 30 min. Immunoblot of lysates detected the phosphorylation of NF-κB, p38, and STAT1. (B) WT MDSC were incubated with GA in the absence or presence of LPS (100 ng/ml) for 30 h. Secreted TNF-α, IL-6, IL-10, and TGF-β were quantified by ELISA. (C) CD115+ MDSC from Lewis lung carcinoma tumor-bearing WT or PIR-B−/− hosts were incubated in the absence (white bars) or presence of GA (gray bars) for 24 h. Secreted TNF-α and IL-10 were measured by ELISA. Data (B and C) represent mean ± SD (*p < 0.05, **p < 0.005, ***p < 0.0005, calculated by Student t test) (D) MDSC isolated from WT or PIR-B KO hosts were preincubated in the absence (blue line) or presence of GA (red line), washed with cold PBS, and stained for surface PIR-B, GA-FITC, or ANGPTL2-FLAG. FACS data are gated on forward scatter–area (FSC-A), side scatter–area (SSC-A), forward scatter–width (FSC-W), and DAPI cells. Histogram data are overlaid FACS staining controls (gray filled). (E) Lysates from WT splenocytes incubated with GA or GA-FITC were immunoprecipitated using anti-FITC Ab capture. Precipitates were separated by SDS-PAGE followed by Western blot using indicated Abs. (F) WT and PIR-B KO MDSC were treated with GA at time 0, 10, and 30 min, followed by anti–PIR-B immunoprecipitation. Precipitate was separated by SDS-PAGE, followed by probing for phosphorylated SHP-1 and anti–PIR-B.

Close modal

Multiple pathways including inhibition of T cell activation and induction of Treg have been proposed to explain how MDSC with M2-like properties suppress adaptive immunity (16). We ascertained whether GA treatment alters an effect of MDSC on T cell maturation and activation. To study Ag-specific T cell responses, we stimulated splenocytes from mice expressing an HA-specific Tg TCR on CD4+ T cells with HA peptide in the presence of MDSC. MDSC-mediated suppression of T cell proliferation was unaffected by the presence of GA (Fig. 2A), but GA further promoted MDSC-dependent CD25+Foxp3+ Treg conversion (Fig. 2B). We also found that GA had no influence on MDSC-mediated T cell suppression when PIR-B knockout (KO) MDSC were cocultured with total splenocytes (Supplemental Fig. 1A). Analyses of cytokine profiles indicated that MDSC and GA both contributed to reduction of IL-17 (Fig. 2C). In the presence of MDSC, GA specifically enhanced IL-10 and TGF-β while reducing IL-23 and IL-6 (Fig. 2D), thus favoring conditions that suppress Th17 maturation but enhance Treg induction. We conclude that the Treg-promoting properties of GA result, in part, from altered cytokine release from innate immune cells. These findings also provide a potential mechanism for how GA ameliorates MS.

FIGURE 2.

GA enhances MDSC-mediated Treg induction. (A) Splenocytes from CD4 HA-TCR Tg mice (SPL) were cocultured with HA peptide and indicated ratios of MDSC purified from MCA26-bearing BALB/c mice in the absence (white bars) or presence of GA (5 μg/ml) (black bars). T cell proliferation was assessed by [3H]thymidine incorporation. (B) Splenocytes (SPL) were cultured with HA peptide. Where indicated, MDSC were cocultured at a ratio of 4:1 in the presence or absence of GA. SPL cultures were stained for CD4+CD25+Foxp3+ Treg after 5 d. FACS data are gated on forward scatter–area (FSC-A), side scatter–area (SSC-A), forward scatter–width (FSC-W), DAPI, and CD3+CD4+ cells. (C) Secreted IL-17A from cultures in (B) was determined by ELISA. (D) Purified MDSC were stimulated with GA for 24 h. Secreted IL-10, TGF-β, IL-6, and IL-23 were quantified by ELISA. Data (C and D) represent mean ± SD (*p < 0.05, **p < 0.01, calculated by Student t test).

FIGURE 2.

GA enhances MDSC-mediated Treg induction. (A) Splenocytes from CD4 HA-TCR Tg mice (SPL) were cocultured with HA peptide and indicated ratios of MDSC purified from MCA26-bearing BALB/c mice in the absence (white bars) or presence of GA (5 μg/ml) (black bars). T cell proliferation was assessed by [3H]thymidine incorporation. (B) Splenocytes (SPL) were cultured with HA peptide. Where indicated, MDSC were cocultured at a ratio of 4:1 in the presence or absence of GA. SPL cultures were stained for CD4+CD25+Foxp3+ Treg after 5 d. FACS data are gated on forward scatter–area (FSC-A), side scatter–area (SSC-A), forward scatter–width (FSC-W), DAPI, and CD3+CD4+ cells. (C) Secreted IL-17A from cultures in (B) was determined by ELISA. (D) Purified MDSC were stimulated with GA for 24 h. Secreted IL-10, TGF-β, IL-6, and IL-23 were quantified by ELISA. Data (C and D) represent mean ± SD (*p < 0.05, **p < 0.01, calculated by Student t test).

Close modal

We next assessed the therapeutic efficacy of administering GA in combination with MDSC. Mice given DSS-supplemented water develop acute colitis, and GA treatment can alleviate symptoms (28). Adoptive transfer of MDSC significantly decreased disease onset and GA treatment in combination with MDSC achieved the best efficacy in suppressing the development of inflammatory bowel disease (IBD) (Fig. 3A). Similar observations were made when assessing treatment efficacy based on colon length (Fig. 3B) and pathological scores (Fig. 3C, Supplemental Fig. 2A). Furthermore, we found that mice treated with MDSC plus GA had significantly higher TGF-β and IL-10 and significantly lower IFN-γ and IL-17A in the colon when compared with untreated DSS IBD mice (Supplemental Fig. 2B). Cells isolated from the mesenteric lymph nodes of day 12 DSS IBD mice were analyzed for their reactivity ex vivo. Upon restimulation with PHA, T cell proliferation was decreased in mice treated with MDSC and further decreased when combined with GA compared with the control group (Fig. 3D). Mice treated with MDSC had a higher percentage of CD4+CD25+FoxP3+ Treg in mesenteric lymph nodes. This effect was enhanced by treatment with GA (Fig. 3E). The results suggest that GA helps to sustain a functional M2-like MDSC to ameliorate IBD.

FIGURE 3.

GA enhances MDSC-dependent suppression of DSS-induced colitis. (A) C57BL/6 mice were fed with water containing 3% DSS from day 0 to day 7 to induce colitis. Mice receiving GA were administered a daily injection from day 1 to day 12. Mice receiving MDSC were given i.v. adoptive transfer of MDSC on days 1 and 6. Clinical scores determining the severity of colitis were determined based on stool consistency, rectal bleeding, and weight loss. The data represent the mean changes in clinical scores for each group (five to eight mice per group) from three separate experiments. (B) Mice from (A) were sacrificed on day 12, and colon length was measured. (C) Colons from (B) were fixed and prepared for H&E staining. Inflammation and tissue infiltration were graded blindly for pathological score. (D) Mesenteric lymph node (MLN) cells from the mice in (A) were isolated and restimulated with vehicle or PHA. [3H]Thymidine incorporation was used to measure T cell proliferation. (E) MLN cells from the mice in (A) were stained for CD4, CD25, and Foxp3 to detect Treg. FACS data are gated on forward scatter–area (FSC-A), side scatter–area (SSC-A), forward scatter–width (FSC-W), DAPI, and CD3+CD4+ cells. Representative dot plots of the MLN cells are shown. (F) C57BL/6 mice were fed water containing 2% DSS from day 0 to day 7 to induce colitis. Mice receiving GA were administered daily injections on days 1–9. Mice receiving MDSC were given i.v. adoptive transfer of WT or PIR-B KO MDSC on days 1 and 4. Body weight was determined every day. (G) Mice from (F) were sacrificed on day 9, and colon length was measured. (H) MLN cells from the mice in (A) were isolated and cultured for 3 d. [3H]Thymidine incorporation was used to measure T cell proliferation. Data represent mean ± SD (*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, calculated by Student t test).

FIGURE 3.

GA enhances MDSC-dependent suppression of DSS-induced colitis. (A) C57BL/6 mice were fed with water containing 3% DSS from day 0 to day 7 to induce colitis. Mice receiving GA were administered a daily injection from day 1 to day 12. Mice receiving MDSC were given i.v. adoptive transfer of MDSC on days 1 and 6. Clinical scores determining the severity of colitis were determined based on stool consistency, rectal bleeding, and weight loss. The data represent the mean changes in clinical scores for each group (five to eight mice per group) from three separate experiments. (B) Mice from (A) were sacrificed on day 12, and colon length was measured. (C) Colons from (B) were fixed and prepared for H&E staining. Inflammation and tissue infiltration were graded blindly for pathological score. (D) Mesenteric lymph node (MLN) cells from the mice in (A) were isolated and restimulated with vehicle or PHA. [3H]Thymidine incorporation was used to measure T cell proliferation. (E) MLN cells from the mice in (A) were stained for CD4, CD25, and Foxp3 to detect Treg. FACS data are gated on forward scatter–area (FSC-A), side scatter–area (SSC-A), forward scatter–width (FSC-W), DAPI, and CD3+CD4+ cells. Representative dot plots of the MLN cells are shown. (F) C57BL/6 mice were fed water containing 2% DSS from day 0 to day 7 to induce colitis. Mice receiving GA were administered daily injections on days 1–9. Mice receiving MDSC were given i.v. adoptive transfer of WT or PIR-B KO MDSC on days 1 and 4. Body weight was determined every day. (G) Mice from (F) were sacrificed on day 9, and colon length was measured. (H) MLN cells from the mice in (A) were isolated and cultured for 3 d. [3H]Thymidine incorporation was used to measure T cell proliferation. Data represent mean ± SD (*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, calculated by Student t test).

Close modal

To confirm whether PIR-B is necessary for the therapeutic efficacy of GA in combination with MDSC, PIR-B KO MDSC were transferred to mice during DSS administration. Adoptive transfer of PIR-B KO MDSC with GA exhibited no suppressive efficacy on the development of IBD (Fig. 3F, 3G, Supplemental Fig. 3). Rather, PIR-B KO MDSC transfer induced slightly more severe IBD development with reducing the body weight and shorter colon length (Fig. 3F, 3G, Supplemental Fig. 3). Further, T cell proliferation was not decreased in mice treated with PIR-B KO MDSC, whereas WT MDSC treatment significantly decreased T cell proliferation (Fig. 3H). Thus, these results indicate that PIR-B is required for GA to sustain an M2-like phenotype in MDSC and to ameliorate IBD.

To determine if GA exerts a similar mechanism of action in human myeloid cells, we stimulated CD33+ human monocytes from healthy donors (n = 16) in the presence of LPS and GA. TNF-α secretion was suppressed whereas TGF-β was enhanced (Fig. 4A), suggesting that GA treatment could dampen the monocyte proinflammatory response. However, monocyte-released IL-10 was not significantly enhanced in contrast to murine MDSC. To determine if GA had more potent activity in suppressing inflammation from M1-skewed human macrophages, monocyte-derived macrophages (MDM) were generated by culturing human monocytes with GM-CSF and GA for 5 d, where indicated. In response to LPS restimulation, GA potently inhibited TNF-α secretion in M1-like MDM (Fig. 4B). Levels of detectable IL-10 were unchanged in response to GA (data not shown).

FIGURE 4.

GA inhibits monocyte classical activation via LILRB2/B3 receptor interaction. (A) Human monocytes from healthy donors (n = 16) were stimulated with LPS in the presence of vehicle or GA for 16 h. Secreted TNF-α was measured by ELISA. Data represent paired data for individual donors. (*p < 0.05, **p < 0.005, calculated by paired t test). (B) Human monocytes were matured in the presence of GM-CSF with titrated GA for 5 d. Macrophages were stimulated with LPS for 16 h, and supernatant was analyzed for TNF-α release by ELISA. M-CSF matured macrophages from the same donor stimulated with LPS are shown as a control. (C) Expression of CD163 and CD16 from 5-d human MDM maturation in the presence of GM-CSF with titrated GA. Each line represents a different donor. (D) 2B4 NFAT-GFP reporter cells stably transduced with LILRB1, LILRB2, LILRB3, or LILRB4 were cross-link activated using LILRB-specific plate-bound Abs at the EC50 for NFAT-GFP activation. Cells were stimulated in the presence of GA for 16 h, followed by flow cytometry to assess reporter GFP intensity. (E) NFAT-GFP mean fluorescence intensity (MFI) expression from 2B4 reporter cells from (D) are unstimulated (gray filled) or stimulated in the presence of vehicle control (black line) or 100 μg/ml GA (red line). FACS data are gated on forward scatter–area (FSC-A), side scatter–area (SSC-A), forward scatter–width (FSC-W), and DAPI cells.

FIGURE 4.

GA inhibits monocyte classical activation via LILRB2/B3 receptor interaction. (A) Human monocytes from healthy donors (n = 16) were stimulated with LPS in the presence of vehicle or GA for 16 h. Secreted TNF-α was measured by ELISA. Data represent paired data for individual donors. (*p < 0.05, **p < 0.005, calculated by paired t test). (B) Human monocytes were matured in the presence of GM-CSF with titrated GA for 5 d. Macrophages were stimulated with LPS for 16 h, and supernatant was analyzed for TNF-α release by ELISA. M-CSF matured macrophages from the same donor stimulated with LPS are shown as a control. (C) Expression of CD163 and CD16 from 5-d human MDM maturation in the presence of GM-CSF with titrated GA. Each line represents a different donor. (D) 2B4 NFAT-GFP reporter cells stably transduced with LILRB1, LILRB2, LILRB3, or LILRB4 were cross-link activated using LILRB-specific plate-bound Abs at the EC50 for NFAT-GFP activation. Cells were stimulated in the presence of GA for 16 h, followed by flow cytometry to assess reporter GFP intensity. (E) NFAT-GFP mean fluorescence intensity (MFI) expression from 2B4 reporter cells from (D) are unstimulated (gray filled) or stimulated in the presence of vehicle control (black line) or 100 μg/ml GA (red line). FACS data are gated on forward scatter–area (FSC-A), side scatter–area (SSC-A), forward scatter–width (FSC-W), and DAPI cells.

Close modal

CD163, a scavenger receptor for the hemoglobin–haptoglobin complex, is expressed on subsets of myeloid cells and macrophages. Its expression on the cell surface is correlated with anti-inflammatory responses and is an indicator of poor prognosis in a variety of cancers (29, 30). CD163 surface expression is enhanced in the presence of M-CSF, IL-6, and IL-10 and in response to glucocorticoids (31). Meanwhile, inflammatory signals including GM-CSF, LPS, IFN-γ, and TNF-α inhibit CD163. Despite the presence of GM-CSF in 5-d MDM cultures, GA rescued CD163 expression in some donors at various concentrations (Fig. 4C, Supplemental Fig. 4A). Expression of CD16 also correlated with CD163 in response to GA. Varying dose-dependent expression of CD163 may be due to donor-specific responsiveness to GA. Based on murine data with PIR-B, we further determined if GA could functionally bind to human orthologs of PIR-B. Unlike mice, humans express five receptors (LILRB1, LILRB2, LILRB3, LILRB4, and LILRB5) that are functionally related to, but distinct from, mouse PIR-B. Like PIR-B, LILRB1 and LILRB2 recognize human MHC-I (32) and ANGPTL family members (21), whereas LILRB3 and LILRB4 ligand binding remains poorly defined. To determine if GA interacts with LILRBs, we used the 2B4 NFAT-GFP reporter assay originally described in studies of the structurally related NK cell receptors (33) and subsequently adapted to study LILRB receptor activation. Cross-link activation of the LILRB extracellular domain stimulates DAP12-dependent calcium flux, leading to NFAT-dependent GFP transcription. We generated LILRB1-, LILRB2-, LILRB3-, and LILRB4-specific NFAT-GFP reporter cell lines to determine if GA could alter receptor cross-link–dependent activation. Plate-bound Abs to LILRB1, LILRB2, LILRB3, and LILRB4 were able to activate NFAT-GFP in a dose-dependent manner (Supplemental Fig. 4B). Interestingly, soluble GA competed with LILRB2 and LILRB3 and inhibited Ab-dependent cross-link activation, also in a dose-dependent manner (Fig. 4D, 4E). LILRB1 and LILRB4 activation was not significantly altered in response to GA as compared with the parental cell line positive control (TCR stimulation). Our data suggest that a peptide component of GA is capable of competing with Abs for LILRB2 and LILRB3 binding.

We have identified that murine PIR-B and human orthologs LILRB2 and LILRB3 are novel targets for GA, a Food and Drug Administration–approved immunotherapeutic that suppresses classical activation of monocytes and promotes M2-like monocytic MDSC populations. LILRB and PIR-B have been well known to regulate maturation and to suppress activation of myeloid cells, including MDSC, macrophages, mast cells, neutrophils, dendritic cells, and B cells (34, 35). LILRB and PIR-B are ITIM-containing receptors. ITIM-containing receptors have been known to bind to three groups of ligands: membrane-bound proteins, including MHC-I or HLA class I, extracellular matrix proteins, and soluble proteins such as FcγRIIB (35). PIR-B and LILRB2 have known ligands, including MHC-I, Angptls, β-amyloid, myelin inhibitors, and CD1D. No ligands have been identified for LILRB3 (35). GA has been known to associate with MHC-II molecules and compete with myelin Ags (9, 36). However, studies indicated that antagonism of MHC-II might not be the primary mechanism of GA (9, 37). We found that GA can be associated with PIR-B and its human orthologs LILRB2 and LILRB3 (Fig. 1E), identifying not only a potential new ligand for PIR-B/LILRB but also a novel mechanism of GA in the immune system.

GA has been shown to reduce inflammatory cytokine production and M1-like maturation. GA reduced CD40, CD86, and IL-12 expression while enhancing expansion of CD14+CD16+ cells in PBMC from MS patients (38). Further, GA attenuated TNF-α–induced IRF-1 upregulation and NO production through inhibition of STAT1 activation (39). Consistent with the reported immunosuppressive effects on myeloid cell populations, we show that GA attenuates TNF-α and IL-6 release while enhancing TGF-β and IL-10 secretion in murine MDSC. GA treatment further correlated with enhanced SHP-1 phosphorylation of PIR-B, a mechanism we previously showed to promote MDSC M2-like maturation (22). GA promoted MDSC-dependent Treg generation and significantly enhanced the therapeutic effect of MDSC in murine colitis models. The in vivo tolerogenic effect of GA-treated MDSC might be due to both functional changes in MDSC and GA-treated MDSC-mediated Treg induction. MDSC and Treg are major cells having immunosuppressive function. The immunosuppressive function of MDSC in IBD is well known. Polymorphonuclear MDSC are accumulated in murine colitis models as well as in the peripheral blood of IBD patients, suggesting their critical roles in IBD (40). However, the roles of MDSC in IBD models is quite controversial because they can also function as proinflammatory myeloid cells. One of the reasons for the contradictory roles of MDSC is related to IL-17 production (41). Th17 cells have pathogenic effects in intestinal inflammation (41, 42). In our study, we found that MDSC reduced IL-17 production, and GA enhanced MDSC-mediated reduction of IL-17 production while increasing Treg expansion. Patients with IBD have been found to harbor reduced peripheral Treg (43). Treg can suppress inflammatory responses by inhibiting Th17 responses through regulation of TGF-β (44, 45). Thus, Treg are a promising target to treat intestinal inflammation, including Crohn disease and IBD (45). Because current drug therapies are unable to maintain long-term remission, GA will be a good candidate for clinical therapy in intestinal inflammatory diseases. We performed similar studies on human monocytes and MDM cultures and found that GA acutely suppressed TNF-α secretion. Interestingly, the efficacy of CD16 and CD163 upregulation by GA differed between donors, which is not surprising given the varying efficacies seen in MS patients. These human data suggest that GA can promote alternative macrophage activation. We provide further evidence to support the interaction between GA and human LILRB2 and LILRB3. Regulation or activation of the LILRB family of receptors may be responsible for GA-dependent alternative activation in human monocytes and macrophages. We suggest further investigation in this area to explore the use of LILRB-targeted interventions on human myeloid populations to improve upon the immunoregulatory changes observed in macrophage maturation in response to GA.

We thank Drs. B. Coakley, Lloyd Mayer, Sergio Lira, Huabao Xiong, and Zhang Yao for technical assistance and constructive discussion.

This work was supported in part by grants from the National Cancer Institute (NCI) (R01CA109322, R01CA127483, and R01CA208703 to S.-H.C., and R01CA140243 and R01CA188610 to P.-Y.P.) and the Kaohsiung Medical University Research Foundation (105KMUOR05 to S.-H.C.). NCI training grants were to W.v.d.T. (T32CA078207 and F32AI122715) and S.M. (T32GM062754). This work was supported by Methodist Research funds to S.-H.C. and P.-Y.P.

The online version of this article contains supplemental material.

Abbreviations used in this article:

ANGPTL

angiopoietin-like

DSS

dextran sulfate sodium

GA

glatiramer acetate

HA

hemagglutinin

IBD

inflammatory bowel disease

KO

knockout

MDM

monocyte-derived macrophage

MDSC

myeloid-derived suppressor cell

MHC-I

MHC class I

MHC-II

MHC class II

MS

multiple sclerosis

Tg

transgenic

Treg

regulatory T cell

WT

wild-type.

1
Teitelbaum
,
D.
,
A.
Meshorer
,
T.
Hirshfeld
,
R.
Arnon
,
M.
Sela
.
1971
.
Suppression of experimental allergic encephalomyelitis by a synthetic polypeptide.
Eur. J. Immunol.
1
:
242
248
.
2
Aharoni
,
R.
,
B.
Kayhan
,
R.
Arnon
.
2005
.
Therapeutic effect of the immunomodulator glatiramer acetate on trinitrobenzene sulfonic acid-induced experimental colitis.
Inflamm. Bowel Dis.
11
:
106
115
.
3
Arnon
,
R.
,
R.
Aharoni
.
2004
.
Mechanism of action of glatiramer acetate in multiple sclerosis and its potential for the development of new applications.
Proc. Natl. Acad. Sci. USA
101
(
Suppl. 2
):
14593
14598
.
4
Kerschensteiner
,
M.
,
C.
Stadelmann
,
G.
Dechant
,
H.
Wekerle
,
R.
Hohlfeld
.
2003
.
Neurotrophic cross-talk between the nervous and immune systems: implications for neurological diseases.
Ann. Neurol.
53
:
292
304
.
5
Fridkis-Hareli
,
M.
,
J. L.
Strominger
.
1998
.
Promiscuous binding of synthetic copolymer 1 to purified HLA-DR molecules.
J. Immunol.
160
:
4386
4397
.
6
Fridkis-Hareli
,
M.
,
D.
Teitelbaum
,
E.
Gurevich
,
I.
Pecht
,
C.
Brautbar
,
O. J.
Kwon
,
T.
Brenner
,
R.
Arnon
,
M.
Sela
.
1994
.
Direct binding of myelin basic protein and synthetic copolymer 1 to class II major histocompatibility complex molecules on living antigen-presenting cells--specificity and promiscuity.
Proc. Natl. Acad. Sci. USA
91
:
4872
4876
.
7
Neuhaus
,
O.
,
C.
Farina
,
A.
Yassouridis
,
H.
Wiendl
,
F.
Then Bergh
,
T.
Dose
,
H.
Wekerle
,
R.
Hohlfeld
.
2000
.
Multiple sclerosis: comparison of copolymer-1- reactive T cell lines from treated and untreated subjects reveals cytokine shift from T helper 1 to T helper 2 cells.
Proc. Natl. Acad. Sci. USA
97
:
7452
7457
.
8
Teitelbaum
,
D.
,
R.
Milo
,
R.
Arnon
,
M.
Sela
.
1992
.
Synthetic copolymer 1 inhibits human T-cell lines specific for myelin basic protein.
Proc. Natl. Acad. Sci. USA
89
:
137
141
.
9
Lalive
,
P. H.
,
O.
Neuhaus
,
M.
Benkhoucha
,
D.
Burger
,
R.
Hohlfeld
,
S. S.
Zamvil
,
M. S.
Weber
.
2011
.
Glatiramer acetate in the treatment of multiple sclerosis: emerging concepts regarding its mechanism of action.
CNS Drugs
25
:
401
414
.
10
Weber
,
M. S.
,
T.
Prod’homme
,
S.
Youssef
,
S. E.
Dunn
,
C. D.
Rundle
,
L.
Lee
,
J. C.
Patarroyo
,
O.
Stüve
,
R. A.
Sobel
,
L.
Steinman
,
S. S.
Zamvil
.
2007
.
Type II monocytes modulate T cell-mediated central nervous system autoimmune disease.
Nat. Med.
13
:
935
943
.
11
Vieira
,
P. L.
,
H. C.
Heystek
,
J.
Wormmeester
,
E. A.
Wierenga
,
M. L.
Kapsenberg
.
2003
.
Glatiramer acetate (copolymer-1, copaxone) promotes Th2 cell development and increased IL-10 production through modulation of dendritic cells.
J. Immunol.
170
:
4483
4488
.
12
Weber
,
M. S.
,
M.
Starck
,
S.
Wagenpfeil
,
E.
Meinl
,
R.
Hohlfeld
,
C.
Farina
.
2004
.
Multiple sclerosis: glatiramer acetate inhibits monocyte reactivity in vitro and in vivo.
Brain
127
:
1370
1378
.
13
Kim
,
H. J.
,
I.
Ifergan
,
J. P.
Antel
,
R.
Seguin
,
M.
Duddy
,
Y.
Lapierre
,
F.
Jalili
,
A.
Bar-Or
.
2004
.
Type 2 monocyte and microglia differentiation mediated by glatiramer acetate therapy in patients with multiple sclerosis.
J. Immunol.
172
:
7144
7153
.
14
Sica
,
A.
,
A.
Mantovani
.
2012
.
Macrophage plasticity and polarization: in vivo veritas.
J. Clin. Invest.
122
:
787
795
.
15
Umemura
,
N.
,
M.
Saio
,
T.
Suwa
,
Y.
Kitoh
,
J.
Bai
,
K.
Nonaka
,
G. F.
Ouyang
,
M.
Okada
,
M.
Balazs
,
R.
Adany
, et al
.
2008
.
Tumor-infiltrating myeloid-derived suppressor cells are pleiotropic-inflamed monocytes/macrophages that bear M1- and M2-type characteristics.
J. Leukoc. Biol.
83
:
1136
1144
.
16
Gabrilovich
,
D. I.
,
S.
Nagaraj
.
2009
.
Myeloid-derived suppressor cells as regulators of the immune system.
Nat. Rev. Immunol.
9
:
162
174
.
17
Yin
,
B.
,
G.
Ma
,
C. Y.
Yen
,
Z.
Zhou
,
G. X.
Wang
,
C. M.
Divino
,
S.
Casares
,
S. H.
Chen
,
W. C.
Yang
,
P. Y.
Pan
.
2010
.
Myeloid-derived suppressor cells prevent type 1 diabetes in murine models.
J. Immunol.
185
:
5828
5834
.
18
Garcia
,
M. R.
,
L.
Ledgerwood
,
Y.
Yang
,
J.
Xu
,
G.
Lal
,
B.
Burrell
,
G.
Ma
,
D.
Hashimoto
,
Y.
Li
,
P.
Boros
, et al
.
2010
.
Monocytic suppressive cells mediate cardiovascular transplantation tolerance in mice.
J. Clin. Invest.
120
:
2486
2496
.
19
Takai
,
T.
,
M.
Ono
.
2001
.
Activating and inhibitory nature of the murine paired immunoglobulin-like receptor family.
Immunol. Rev.
181
:
215
222
.
20
Masuda
,
A.
,
A.
Nakamura
,
T.
Maeda
,
Y.
Sakamoto
,
T.
Takai
.
2007
.
Cis binding between inhibitory receptors and MHC class I can regulate mast cell activation.
J. Exp. Med.
204
:
907
920
.
21
Zheng
,
J.
,
M.
Umikawa
,
C.
Cui
,
J.
Li
,
X.
Chen
,
C.
Zhang
,
H.
Huynh
,
X.
Kang
,
R.
Silvany
,
X.
Wan
, et al
.
2012
.
Inhibitory receptors bind ANGPTLs and support blood stem cells and leukaemia development. [Published erratum appears in 2012 Nature 488: 684.]
Nature
485
:
656
660
.
22
Ma
,
G.
,
P. Y.
Pan
,
S.
Eisenstein
,
C. M.
Divino
,
C. A.
Lowell
,
T.
Takai
,
S. H.
Chen
.
2011
.
Paired immunoglobin-like receptor-B regulates the suppressive function and fate of myeloid-derived suppressor cells.
Immunity
34
:
385
395
.
23
Huang
,
B.
,
P. Y.
Pan
,
Q.
Li
,
A. I.
Sato
,
D. E.
Levy
,
J.
Bromberg
,
C. M.
Divino
,
S. H.
Chen
.
2006
.
Gr-1+CD115+ immature myeloid suppressor cells mediate the development of tumor-induced T regulatory cells and T-cell anergy in tumor-bearing host.
Cancer Res.
66
:
1123
1131
.
24
Bot
,
A.
,
S.
Casares
,
S.
Bot
,
H.
von Boehmer
,
C.
Bona
.
1998
.
Cellular mechanisms involved in protection against influenza virus infection in transgenic mice expressing a TCR receptor specific for class II hemagglutinin peptide in CD4+ and CD8+ T cells.
J. Immunol.
160
:
4500
4507
.
25
Kusmartsev
,
S. A.
,
Y.
Li
,
S. H.
Chen
.
2000
.
Gr-1+ myeloid cells derived from tumor-bearing mice inhibit primary T cell activation induced through CD3/CD28 costimulation.
J. Immunol.
165
:
779
785
.
26
Pan
,
P. Y.
,
G. X.
Wang
,
B.
Yin
,
J.
Ozao
,
T.
Ku
,
C. M.
Divino
,
S. H.
Chen
.
2008
.
Reversion of immune tolerance in advanced malignancy: modulation of myeloid-derived suppressor cell development by blockade of stem-cell factor function.
Blood
111
:
219
228
.
27
Bléry
,
M.
,
H.
Kubagawa
,
C. C.
Chen
,
F.
Vély
,
M. D.
Cooper
,
E.
Vivier
.
1998
.
The paired Ig-like receptor PIR-B is an inhibitory receptor that recruits the protein-tyrosine phosphatase SHP-1.
Proc. Natl. Acad. Sci. USA
95
:
2446
2451
.
28
Aharoni
,
R.
,
H.
Sonego
,
O.
Brenner
,
R.
Eilam
,
R.
Arnon
.
2007
.
The therapeutic effect of glatiramer acetate in a murine model of inflammatory bowel disease is mediated by anti-inflammatory T-cells.
Immunol. Lett.
112
:
110
119
.
29
Zaki
,
M. A.
,
N.
Wada
,
J.
Ikeda
,
H.
Shibayama
,
K.
Hashimoto
,
T.
Yamagami
,
Y.
Tatsumi
,
M.
Tsukaguchi
,
H.
Take
,
M.
Tsudo
, et al
.
2011
.
Prognostic implication of types of tumor-associated macrophages in Hodgkin lymphoma.
Virchows Arch.
459
:
361
366
.
30
Ino
,
Y.
,
R.
Yamazaki-Itoh
,
K.
Shimada
,
M.
Iwasaki
,
T.
Kosuge
,
Y.
Kanai
,
N.
Hiraoka
.
2013
.
Immune cell infiltration as an indicator of the immune microenvironment of pancreatic cancer.
Br. J. Cancer
108
:
914
923
.
31
Etzerodt
,
A.
,
S. K.
Moestrup
.
2013
.
CD163 and inflammation: biological, diagnostic, and therapeutic aspects.
Antioxid. Redox Signal.
18
:
2352
2363
.
32
Jones
,
D. C.
,
V.
Kosmoliaptsis
,
R.
Apps
,
N.
Lapaque
,
I.
Smith
,
A.
Kono
,
C.
Chang
,
L. H.
Boyle
,
C. J.
Taylor
,
J.
Trowsdale
,
R. L.
Allen
.
2011
.
HLA class I allelic sequence and conformation regulate leukocyte Ig-like receptor binding.
J. Immunol.
186
:
2990
2997
.
33
Arase
,
H.
,
E. S.
Mocarski
,
A. E.
Campbell
,
A. B.
Hill
,
L. L.
Lanier
.
2002
.
Direct recognition of cytomegalovirus by activating and inhibitory NK cell receptors.
Science
296
:
1323
1326
.
34
van der Touw
,
W.
,
H. M.
Chen
,
P. Y.
Pan
,
S. H.
Chen
.
2017
.
LILRB receptor-mediated regulation of myeloid cell maturation and function.
Cancer Immunol. Immunother.
66
:
1079
1087
.
35
Kang
,
X.
,
J.
Kim
,
M.
Deng
,
S.
John
,
H.
Chen
,
G.
Wu
,
H.
Phan
,
C. C.
Zhang
.
2016
.
Inhibitory leukocyte immunoglobulin-like receptors: immune checkpoint proteins and tumor sustaining factors.
Cell Cycle
15
:
25
40
.
36
Arnon
,
R.
,
M.
Sela
.
2003
.
Immunomodulation by the copolymer glatiramer acetate.
J. Mol. Recognit.
16
:
412
421
.
37
Aharoni
,
R.
,
P. G.
Schlegel
,
D.
Teitelbaum
,
O.
Roikhel-Karpov
,
Y.
Chen
,
R.
Arnon
,
M.
Sela
,
N. J.
Chao
.
1997
.
Studies on the mechanism and specificity of the effect of the synthetic random copolymer GLAT on graft-versus-host disease.
Immunol. Lett.
58
:
79
87
.
38
Chuluundorj
,
D.
,
S. A.
Harding
,
D.
Abernethy
,
A. C.
La Flamme
.
2017
.
Glatiramer acetate treatment normalized the monocyte activation profile in MS patients to that of healthy controls.
Immunol. Cell Biol.
95
:
297
305
.
39
Lu
,
H.
,
C.
Zeng
,
H.
Zhao
,
L.
Lian
,
Y.
Dai
.
2014
.
Glatiramer acetate inhibits degradation of collagen II by suppressing the activity of interferon regulatory factor-1.
Biochem. Biophys. Res. Commun.
448
:
323
328
.
40
Haile
,
L. A.
,
R.
von Wasielewski
,
J.
Gamrekelashvili
,
C.
Kruger
,
O.
Bachmann
,
A. M.
Westendorf
,
J.
Buer
,
R.
Liblau
,
M. P.
Manns
,
F.
Korangy
,
T. F.
Greten
.
2008
.
Myeloid-derived suppressor cells in inflammatory bowel disease: a new immunoregulatory pathway.
Gastroenterology
135
:
871
881, 881.e1–5
.
41
Kim
,
Y. J.
,
S. Y.
Chang
,
H. J.
Ko
.
2015
.
Myeloid-derived suppressor cells in inflammatory bowel disease.
Intest. Res.
13
:
105
111
.
42
Ostanin
,
D. V.
,
D.
Bhattacharya
.
2013
.
Myeloid-derived suppressor cells in the inflammatory bowel diseases.
Inflamm. Bowel Dis.
19
:
2468
2477
.
43
Eastaff-Leung
,
N.
,
N.
Mabarrack
,
A.
Barbour
,
A.
Cummins
,
S.
Barry
.
2010
.
Foxp3+ regulatory T cells, Th17 effector cells, and cytokine environment in inflammatory bowel disease.
J. Clin. Immunol.
30
:
80
89
.
44
Ueno
,
A.
,
H.
Jijon
,
R.
Chan
,
K.
Ford
,
C.
Hirota
,
G. G.
Kaplan
,
P. L.
Beck
,
M.
Iacucci
,
M.
Fort Gasia
,
H. W.
Barkema
, et al
.
2013
.
Increased prevalence of circulating novel IL-17 secreting Foxp3 expressing CD4+ T cells and defective suppressive function of circulating Foxp3+ regulatory cells support plasticity between Th17 and regulatory T cells in inflammatory bowel disease patients.
Inflamm. Bowel Dis.
19
:
2522
2534
.
45
Yamada
,
A.
,
R.
Arakaki
,
M.
Saito
,
T.
Tsunematsu
,
Y.
Kudo
,
N.
Ishimaru
.
2016
.
Role of regulatory T cell in the pathogenesis of inflammatory bowel disease.
World J. Gastroenterol.
22
:
2195
2205
.

The authors have no financial conflicts of interest.

Supplementary data