Novel small molecule IL‐6 inhibitor suppresses autoreactive Th17 development and promotes T_reg development

Summary

Multiple sclerosis (MS) is the leading cause of non‐traumatic neurological disability in the United States in young adults, but current treatments are only partially effective, making it necessary to develop new, innovative therapeutic strategies. Myelin‐specific interleukin (IL)‐17‐producing T helper type 17 (Th17) cells are a major subset of CD4 T effector cells (T_eff) that play a critical role in mediating the development and progression of MS and its mouse model, experimental autoimmune encephalomyelitis (EAE), while regulatory T cells (T_reg) CD4 T cells are beneficial for suppressing disease. The IL‐6/signal transducer and activator of transcription 3 (STAT‐3) signaling pathway is a key regulator of Th17 and T_reg cells by promoting Th17 development and suppressing T_reg development. Here we show that three novel small molecule IL‐6 inhibitors, madindoline‐5 (MDL‐5), MDL‐16 and MDL‐101, significantly suppress IL‐17 production in myelin‐specific CD4 T cells in a dose‐dependent manner in vitro. MDL‐101 showed superior potency in suppressing IL‐17 production compared to MDL‐5 and MDL‐16. Treatment of myelin‐specific CD4 T cells with MDL‐101 in vitro reduced their encephalitogenic potential following their subsequent adoptive transfer. Furthermore, MDL‐101 significantly suppressed proliferation and IL‐17 production of anti‐CD3‐activated effector/memory CD45RO⁺CD4⁺ human CD4 T cells and promoted human T_reg development. Together, these data demonstrate that these novel small molecule IL‐6 inhibitors have the potential to shift the T_eff : T_reg balance, which may provide a novel therapeutic strategy for ameliorating disease progression in MS.

Introduction

Multiple sclerosis (MS) is a chronic central nervous system (CNS) disease characterized by inflammation‐mediated demyelination and neuronal degeneration. It is the leading cause of non‐traumatic neurological disability in the United States in young adults; thus, most patients suffer from the effects of MS for most of their adult life. Although the precise etiology of MS is still unknown, myelin‐specific IL‐17‐producing Th17 cells are a major subset of CD4 T effector cells (T_eff) that play a critical role in mediating the formation of acute MS lesion and disease development and progression in MS 1 and its mouse model, experimental autoimmune encephalomyelitis (EAE). IL‐6 induces the differentiation of myelin‐specific Th17 cells that are highly encephalitogenic in adoptively transferred EAE 2, 3, 4. IL‐6^–/– mice are completely resistant to EAE induction 5, 6, 7, 8, while injection of recombinant IL‐6 induces severe EAE in those IL‐6^–/– mice 6, suggesting that IL‐6 is essential for the development of autoreactive Th17 cells and EAE development. Meanwhile, the IL‐6/signal transducer and activator of transcription 3 (STAT‐3) pathway is a key signaling pathway blocking the development of inducible regulatory T cells (iT_reg), which are beneficial for MS and EAE by dampening the encephalitogenic T_eff cells. IL‐6, signaling through STAT‐3, completely abrogates the de‐novo induction of iT_reg cells 9, 10. As a result, dysregulated IL‐6/STAT‐3 signaling skews the T_eff : T_reg balance toward an enhanced T_eff response, favoring the development of autoimmunity.

The dysregulation of IL‐6/STAT‐3 signaling plays a significant role in the pathogenesis of MS and other autoimmune diseases. IL‐6 mRNA and protein levels were found to be elevated in the CNS of MS patients 11, 12 and B cells from MS patients secret significantly more IL‐6 than healthy controls (HC) 13. CD4 T cells from MS patients have significantly more IL‐6 receptors (IL‐6R) than HCs 14, and the expression of phosphorylated STAT‐3 (pSTAT‐3) in peripheral blood mononuclear cells (PBMC) from relapsing–remitting MS (RRMS) patients strongly correlates with MS disease activity 15, suggesting that IL‐6 is critical for MS pathogenesis, and suppressing the elevated IL‐6/STAT3 signaling pathway in MS patients may serve as an innovative therapeutic approach for MS patients.

Anti‐IL‐6/STAT‐3 signaling pathway antibodies have been tested in CNS autoimmunity and other autoimmune diseases. Administration of the αIL‐6 monoclonal antibody (mAb) or αIL‐6Rα mAb significantly reduced EAE development, both in actively induced EAE and in the adoptive transfer model of EAE 16, 17. However, like all the peptide/protein‐based therapies, monoclonal antibody‐based therapies have the disadvantage of infusion reactions 18 and the generation of blocking antibodies against the drug. In contrast, small molecule compounds typically offer improved bioavailability and manufacturing features over protein‐based drugs. Therefore, in this study we have determined the effects of three novel small molecule IL‐6 inhibitors, MDL‐5, MDL‐16 and MDL‐101, in the inhibition of Th17 development, promoting iT_reg development and suppressing the encephalitogenic potential of myelin‐specific CD4 T cells in the EAE model of MS.

Materials and methods

Animals

C57BL/6, B10.PL and Swiss Jackson Laboratory (SJL)/J mice were purchased from the Jackson Laboratory and bred in a specific pathogen‐free animal facility at Ohio State University (OSU) Wexner Medical Center. B10.PL mice transgenic for the myelin basic protein (MBP) Ac1‐11‐specific T cell receptor (TCR) chains Vα2.3 or Vβ8.2 (TCR transgenic mice) 19 were also bred in a specific pathogen‐free animal facility at OSU Wexner Medical Center. All animal protocols were approved by the OSU Institutional Animal Care and Use Committee.

Human subjects

Blood samples from healthy individuals were purchased from American Red Cross and no institutional review board (IRB) permission is required. Peripheral blood mononuclear cells (PBMCs) isolated over a Ficoll gradient were stored in liquid nitrogen until further use.

In‐vitro culture of splenocytes from TCR transgenic mice

Splenocytes were prepared from naive 5–10‐week‐old Vα2.3/Vβ8.2 TCR transgenic mice that are specific for MBP Ac1‐11 and cultured in 24‐well plates at 2 × 10⁶ cells/well with irradiated B10.PL splenocytes (6 × 10⁶ cells/well). Cells were activated with MBP Ac1‐11 (10 µg/ml) and different combinations of cytokines to differentiate T_eff helper cells. Cytokine concentrations were as follows: 0·5 ng/ml IL‐12, 25 ng/ml IL‐6 and 1 ng/ml transforming growth factor (TGF)‐β1. MDL analogs were dissolved in dimethylsulfoxide (DMSO) and added at different concentrations during the in‐vitro culture of splenocytes from TCR transgenic mice.

Trypan blue exclusion assay

Splenocytes from TCR transgenic mice that are specific for MBP Ac1‐11 were cultured with MBP Ac1‐11 and different concentrations of MDL‐5 or MDL‐16 for 24, 48 and 72 h. DMSO was used as a vehicle control. Cell viability was assessed by Trypan blue dye exclusion after 24, 48  and 72 h exposure to the compounds. The splenocytes were mixed with 0·4% w/v freshly made Trypan blue and counted on a hemocytometer. The percentage viability is calculated by dividing the number of viable (unstained) cells by the number of all counted cells (stained and unstained).

EAE induction

Immunization. Eight to 10‐week‐old naive C57BL/6 mice were subcutaneously injected (s.c.) over four sites in the flank with 200 µg myelin oligodendrocyte glycoprotein (MOG) 35–55 peptide (CSBio, Menlo Park, CA, USA) in an emulsion with complete Freund’s adjuvant (CFA) (Difco, Detroit, MI, USA). Pertussis toxin (List; 200 ng) was injected i.p. at the time of immunization and 48 h later. The mice were evaluated daily for clinical signs of EAE. For administration of MDL‐101 in vivo (Fig. 4i), MDL‐101 stock solution (dissolved in DMSO) were diluted in 10% DMSO/5% Tween 80/10% polyethylene glycol (PEG) 400/75% physiological saline (0·9%) and injected intraperitoneally (i.p.) into mice at 50 mg/kg.

Figure 4.

Novel small molecule interleukin (IL)‐6 inhibitor madindoline (MDL)‐101 significantly suppresses Th17 development and encephalitogenic potential of myelin‐specific CD4 T cells. (a–c) Splenocytes from naive T cell receptor (TCR) transgenic mice were activated with myelin basic protein (MBP) Ac1‐11 plus transforming growth factor (TGF)‐β1 and IL‐6 for 3 days in the presence of different concentrations of MDL‐101. (a) IL‐17 production in CD4 T cells was determined by intracellular flow cytometric analysis. Cells were gated on CD4⁺CD44⁺ cells. Flow data are representative of three independent experiments. (b) Group means were calculated and compared to the group treated with dimethylsulfoxide (DMSO) by one‐way analysis of variance (anova). (c) IL‐17 in supernatant was determined by enzyme‐linked immunosorbent assay (ELISA) and compared with anova. (d) Splenocytes from TCR transgenic mice were cultured with rmIL‐6 for 30 min in the presence of 5 or 15 μM of MDL‐101. Phosphorylated signal transducer and activator of transcription 3 (pSTAT‐3) was determined by phospho flow cytometry. Cells were gated on CD4⁺ cells. Flow data represent three independent experiments. Group means were calculated and compared with one‐way anova. (e) Splenocytes from TCR transgenic mice were cultured with MBP Ac1‐11 and different concentrations of MDL‐101 for 24, 48 and 72 h. Cells were collected for cell death determination by using Trypan blue exclusion assay. (f,g) Splenocytes from naive TCR transgenic mice were activated with MBP Ac1‐11 and IL‐6 for 3 days in the presence of MDL‐101 (10 μM). DMSO was used as vehicle control. At the end of 3 days’ culture, the cells were collected and injected intraperitoneally (i.p.) into naive B10PL recipient mice. (f) IL‐17 in supernatant was determined by ELISA and compared with unpaired t‐test. (g) Experimental autoimmune encephalomyelitis (EAE) was monitored in recipient mice. Disease incidence (sick mice/total mice) is indicated in parentheses. A statistically significant difference was considered to be P < 0·05, as determined by Mann–Whitney U‐test. Data represent two independent experiments. (h) Splenocytes from Swiss Jackson Laboratory (SJL)/J mice that were immunized with proteolipid protein (PLP) 139–151 (day 20 post‐immunization) were isolated and activated with PLP 139–151 and IL‐6 for 3 days in the presence of MDL‐101 or DMSO (10 μM). Then the cells were collected and injected i.p. into naive SJL/J recipient mice. EAE development was monitored and compared by Mann–Whitney U‐test. (i) Naive B6 mice were immunized with MOG 35–55. The mice were then injected i.p. with MDL‐101 (50 mg/kg) or vehicle control (DMSO) for 7 days starting on day 10 after immunization. EAE is monitored. All error bars denote standard error of the mean (s.e.m.). *P < 0·05; ***P < 0·001; ****P < 0·0001.

Adoptive transfer. Splenocytes were isolated from naive 5–10‐week‐old Vα2.3/Vβ8.2 TCR transgenic mice or immunized SJL/J mice (days 14–21 post‐immunization) and activated with MBP Ac1‐11 (10 μg/ml) or proteolipid protein (PLP) 139–151 (10 μg/ml), plus IL‐6 (25 ng/ml) and MDL‐101 or DMSO (10 μM). After 72 h, the cells were washed with PBS and injected (6–10 × 10⁶/mouse) i.p. into naive B10.PL or SJL/J mice.

The mice were evaluated daily for clinical signs of EAE. Mice were scored on a scale of 0–6: 0, no clinical disease; 1, limp/flaccid tail; 2, moderate hind limb weakness; 3, severe hind limb weakness; 4, complete hind limb paralysis; 5 quadriplegia or premoribund state; and 6, death.

Enzyme‐linked immunosorbent assay (ELISA)

ELISA was performed to detect the expression of human and mouse IL‐17 in supernatant. Purified αmIL‐17 or αhIL‐17 (BD Biosciences, San Jose, CA, USA) was diluted in 0·1 M NaHCO3 (pH 8.2) at 2 µg/ml. Immunolon II plates (Dynatech Laboratories, El Paso, TX, USA) were coated with 50 µl of primary antibodies per well and incubated overnight at 4°C. The plates were washed twice with phosphate‐buffered saline (PBS)/0·05% Tween 20. The plates were blocked with 200 µl of 1% bovine serum albumin (BSA) in PBS per well for 2 h. The plates were washed twice with PBS/0·05% Tween 20, and 100 µl of supernatants were added in duplicate. The plates were incubated overnight at 4°C and washed four times with PBS/0·05% Tween 20. Biotinylated rat anti‐mouse or anti‐human secondary antibody (BD Biosciences) was diluted in PBS/1% BSA; 100 µl of 1 µg/ml biotinylated antibody was added to each well, and plates were incubated at room temperature for 1 h. The plates were washed six times with PBS/0·05%Tween 20, and 100 µl avidin‐peroxidase was added at 2·5 µg/ml and incubated for 30 min. The plates were washed eight times with PBS/0·05% Tween 20, and 100 µl ABTS substrate containing 0·03% H₂O₂ or TMB substrate was added to each well. The plate was monitored for 10–20 min for color development and read at A405. A standard curve was generated from a cytokine standard, and the cytokine concentration in the samples was calculated.

Intracellular staining and flow cytometric analysis

Flow cytometric analysis was performed to evaluate the expression of surface markers and IL‐17 production in CD4 T cells, as previously described 2. Briefly, splenocytes were activated with myelin‐antigen for 3 days. Cells were then collected, washed and resuspended in staining buffer (1% BSA in PBS). The cells were incubated with mAbs to the cell‐surface markers for 30 min at 4°C. After washing twice with staining buffer, cells were fixed and permeabilized using Cytofix/Cytoperm solution (BD Biosciences) for 20 min at 4°C. Cells were stained for IL‐17 for 30 min at 4°C; 80 000–100 000 live cell events were acquired on a FACSCantoII (BD Biosciences) and analyzed using FlowJo software (Tree Star, Inc., Ashland, OR, USA). Peridinin chlorophyll (PerCP)‐αCD4, Pacific Blue‐αCD44 and phycoerythrin (PE)‐αIL‐17 were purchased from BD Biosciences.

Phospho flow cytometry was used to detect pSTAT‐3 expression in CD4 T cells. Briefly, splenocytes from TCR transgenic mice were activated with IL‐6 for 20–30 min. Then the cells were collected and fixed using BD Cytofix™ fixation buffer for 10–12 min at 37°C. The cells were washed with staining buffer (2% FBS in PBS) and then permeabilized using ice‐cold Perm buffer III (Phosflow; BD Biosciences) for 30 min at 4°C. After washing twice with staining buffer, cells were stained for surface markers and pSTAT‐3 with antibody diluted in staining buffer for 1 h at room temperature. Cells were washed twice prior to flow cytometric analysis. Fluorescein isothiocyanate (FITC) αmCD4 (Biolegend, San Diego, CA, USA), AF647 αpSTAT‐3 (BD Biosciences) were used for staining. Approximately 100 000 cell events were acquired on a FACSCanto II (BD Biosciences) and analyzed using FlowJo software version 10.1 (Tree Star, Inc.).

Human CFSE‐based proliferation assay

PBMCs were isolated from healthy controls. Carboxyfluorescein succinimidyl ester (CFSE) (Invitrogen, Carlsbad, CA, USA) was added to the human PBMCs to a final concentration of 0·5 μM, followed by a 15‐min incubation at 37°C in the dark. Cells were then washed three times with PBS and plated with 15 ng/ml of rhIL‐6 in 48‐well plates that have been coated with anti‐human CD3. Different concentrations of MDL‐101 were added to the culture. DMSO was used as the vehicle control. Day 5 after activation, cells were collected and stained for CD4 (Biolegend) and CD45RO (BD Biosciences); 80 000–100 000 live cell events were acquired on a FACSCantoII (BD Biosciences) and analyzed using FlowJo software (Tree Star, Inc.).

In‐vitro human T_reg induction and flow cytometric analysis of human T_regs

Inducible T_regs (iT_regs) were generated by culturing the human PBMCs on 24‐well plates coated with 1 μg/ml of anti‐human CD3/CD28 in the presence of 2 U/ml rhIL‐2, 1 ng/ml TGF‐β1, 15 ng/ml rhIL‐6 and 2·5 nM all trans‐retinoic acid (RA) for 72 h at 37°C. Different concentrations of MDL‐101 were added to the culture. DMSO was used as the vehicle control. After 72 h of culture, the cells were initially stained with antibodies for CD4 (BD Biosciences), CD45RA (BioLegend) and CD25 (BD Biosciences), followed by intracellular staining for forkhead box protein 3 (FoxP3) (eBioscience). The data were analyzed using FlowJo software (Tree Star, Inc.). The cells were initially gated on CD4⁺CD45RA⁺ naive CD4 T cells, and this population was further analyzed for CD25 and FoxP3 expression so that only the T_regs that differentiated from the naïve CD4 T cell population were quantitated.

Statistical analysis

GraphPad software (GraphPad Prism Software, Inc., San Diego, CA, USA) was utilized for statistical analysis. A statistically significant difference in EAE clinical scores was considered to be P < 0·05, as determined by Mann–Whitney U‐test. The Mann–Whitney U‐test is non‐parametric, and therefore accounts for the fact that EAE scores are ordinal and not interval‐scaled. ELISA and quantitated flow data comparisons were performed using two‐tailed unpaired Student’s t‐tests. The quantitated data comparisons with three or more groups were performed using one‐way analysis of variance (anova). Differences with P < 0·05 were considered significant.

Results

Computer‐aided design of novel small molecule IL‐6 inhibitors, MDL‐5, MDL‐16 and MDL‐101

IL‐6 binds to IL6‐Rα and GP130’s D2/D3 domains to form IL‐6/IL‐6Rα/GP130 heterotrimer. Two heterotrimers form an active heterohexameric signaling complex, leading to downstream STAT‐3 activation. Targeting the GP130 D1 domain offers unique drug design selectivity, as other cytokine receptors that utilize GP130 either do not require the D1 domain or do not require hexamerization, reducing potential toxicity and side effects. The natural product madindoline A (MDL‐A, Fig. 1) 20 binds to the D1 domain of GP130 21 and antagonizes GP13022. However, MDL‐A is a weak binder and hard to synthesize. By using a fragment‐based design approach called multiple ligand simultaneous docking (MLSD) 23, 24, three novel MDL analogs, MDL‐5, MDL‐16 and MDL‐101, were derived from MDL‐A (Fig. 1) and show inhibition of IL‐6 signaling by the down‐regulation of STAT‐3 phosphorylation and STAT‐3 nuclear translocation in cancer cell lines 25. Synthesis, chemical characterization and detailed structural information are presented in the patent 25.

Figure 1.

Chemical structures of small molecule interleukin (IL)‐6 inhibitors, madindoline (MDL)‐A, MDL‐5, MDL‐16 and MDL‐101.

Novel small molecule IL‐6 inhibitors, MDL‐5 and MDL‐16, inhibit IL‐17 production in myelin‐specific CD4 T cells

IL‐6 drives the differentiation of IL‐17‐producing Th17 cells, which are highly encephalitogenic in the EAE model of MS 2, 3, 4. Therefore, we determined whether the novel small molecule IL‐6 inhibitors, MDL‐5 and MDL‐16, suppress the development of myelin‐specific Th17 cells in vitro. Splenocytes from naive TCR transgenic mice that are specific for MBP Ac1‐11 were activated with MBP Ac1‐11 and IL‐6/TGF‐β1 for 3 days in the presence of different concentrations of MDL‐5 or MDL‐16. DMSO was used as a vehicle control. Both MDL‐5 and MDL‐16 inhibit IL‐17 production in myelin‐specific CD4 T cells in a dose‐dependent manner (Figs. 2 and 3a). At 25 and 35 μM, MDL‐5 significantly suppresses IL‐17 production in activated myelin‐specific CD4 T cells (Fig. 2). Similarly, MDL‐16 suppresses IL‐17 production in a dose‐dependent manner (Fig. 3). The IL‐17 in the supernatant was significantly lower with two concentrations of MDL‐16 (25 and 35 μM) (Fig. 3c). IL‐6 signals through STAT‐3 in CD4 T cells. To confirm that MDL‐16 suppresses the IL‐6 signaling pathway, pSTAT‐3 expression was determined by flow cytometric analysis in MDL‐16‐treated myelin‐specific CD4 T cells. Splenocytes from TCR transgenic mice were stimulated directly ex vivo with IL‐6 for 30 min in the presence of different concentrations of MDL‐16. MDL‐16 inhibited pSTAT‐3 expression in a dose‐dependent manner (Fig. 3d). The flow cytometric analysis of IL‐17 production was gated on viable activated CD4⁺CD44⁺ cells, which excluded the possibility that the suppressive effects were caused by MDL cellular toxicity. However, we also performed Trypan blue exclusion testing to determine the viability of myelin‐specific T cells cultured with different concentrations of MDL‐5 (Fig. 2d) and MDL‐16 (Fig. 3e). MDL‐5 has no significant cellular toxicity on myelin‐specific CD4 T cells after 72 h culture at all the concentrations tested (Fig. 2d). Similarly, MDL‐16, at concentrations of 10 and 20 μM, has no significant cellular toxicity after 24, 48 or 72 h culture (Fig. 3e), whereas 40 μM of MDL‐16 shows cellular toxicity on myelin‐specific CD4 T cells after 72 h culture (Fig. 3e). These data suggest that MDL‐5 and 16 suppressed pSTAT‐3 downstream of IL‐6R signaling, resulting in diminished Th17 cell responses.

Figure 2.

Madindoline (MDL)‐5 significantly inhibits interleukin (IL)‐17 production in myelin‐specific CD4 T cells. (a–c) Splenocytes from naive T cell receptor (TCR) transgenic mice that are specific for myelin basic protein (MBP) Ac1‐11 were activated with MBP Ac1‐11 plus transforming growth factor (TGF)‐β1 and IL‐6 for 3 days, in the presence of different concentrations of MDL‐5. (a) IL‐17 production in CD4 T cells was determined by intracellular flow cytometric analysis. Cells were gated on CD4⁺CD44⁺ cells. Flow data are representative of three independent experiments. (b) Group means were calculated and compared to the group treated with dimethylsulfoxide (DMSO) by one‐way analysis of variance (anova). (c) IL‐17 in supernatant was determined by enzyme‐linked immunosorbent assay (ELISA) and compared with one‐way anova. (d) Splenocytes from T cell receptor (TCR) transgenic mice were cultured with MBP Ac1‐11 and different concentrations of MDL‐5 for 72 h. Cells were collected for cell death determination by using Trypan blue exclusion assay. Data are representative of three independent experiments. DMSO was used as a vehicle control. All error bars denote standard error of the mean (s.e.m.). **P < 0·01; ***P < 0·001.

Figure 3.

Madindoline (MDL)‐16 significantly inhibits phosphorylated signal transducer and activator of transcription 3 (pSTAT‐3) expression and interleukin (IL)‐17 production in myelin‐specific CD4 T cells. (a–c) Splenocytes from naive T cell receptor (TCR) transgenic mice were activated with myelin basic protein (MBP) Ac1‐11 plus transforming growth factor (TGF)‐β1 and IL‐6 for 3 days in the presence of different concentrations of MDL‐16. (a) IL‐17 production in CD4 T cells was determined by intracellular flow cytometric analysis. Cells were gated on CD4⁺CD44⁺ cells. Flow data are representative of three independent experiments. (b) Group means were calculated and compared to the group treated with dimethylsulfoxide (DMSO) by one‐way analysis of variance (anova). (c) IL‐17 in supernatant was determined by enzyme‐linked immunosorbent assay (ELISA) and compared with one‐way anova. (d) Splenocytes from T cell receptor (TCR) transgenic mice were cultured with rmIL‐6 for 30 min in the presence of different concentrations of MDL‐16. pSTAT‐3 was determined by phospho flow cytometry. Cells were gated on CD4⁺ cells. Flow data represent two independent experiments. (e) Splenocytes from TCR transgenic mice were cultured with MBP Ac1‐11 and different concentrations of MDL‐16 for 24, 48 and 72 h. Cells were collected for cell death determination by using Trypan blue exclusion assay. All error bars denote standard error of the mean (s.e.m.). *P < 0·05; ***P < 0·001; ****P < 0·0001.

Novel MDL analog MDL‐101 suppresses Th17 development and encephalitogenic potential of myelin‐specific CD4 T cells

To improve the suppressive effects on IL‐17 production, a novel MDL analog MDl‐101 was designed and tested. Splenocytes from TCR transgenic mice were activated with MBP Ac1‐11 for 3 days, plus different concentrations of MDL‐101. MDL‐101 suppressed IL‐17 production in myelin‐specific CD4 T cells in a dose‐dependent manner (Fig. 4a). MDL‐101 significantly suppresses IL‐17 production in myelin‐specific CD4 T cells at a concentration of 15 μM (Fig. 4), while MDL‐5 (Fig. 2) and MDL‐16 (Fig. 3) show no suppression of IL‐17 production at 15 μM. pSTAT‐3 expression was also significantly suppressed by MDL‐101 at both lower doses (5 and 15 μM; Fig. 4d). Trypan blue exclusion assay showed no significant cellular toxicity at all the concentrations used at 24, 48  or 72 h of culture (Fig. 4e). To determine whether MDL‐101 suppresses the encephalitogenic potential of myelin‐specific CD4 T cell, splenocytes from TCR transgenic mice were activated with MBP Ac1‐11 and IL‐6 for 3 days in the presence of 10 μM of MDL‐101 or vehicle control (DMSO). IL‐17 production was significantly lower in MDL‐101‐treated cells compared to vehicle control (DMSO)‐treated cells (Fig. 4f). The cells were injected i.p. into naive B10PL mice. EAE severity in the mice receiving MDL‐101‐treated myelin‐specific CD4 T cells was significantly lower than those receiving vehicle control‐treated cells (Fig. 4g and Table 1), suggesting that MDL‐101 suppressed T cell encephalitogenicity. To make certain that the suppressive effect of MDL‐101 on encephalitogenic potential is not specific to MBP Ac1‐11 or the major histocompatibility complex (MHC) (H‐2^u) of B10PL mice, we performed a similar experiment by adoptively transferring splenocytes from immunized SJL/J mice (H‐2^s) that were activated with PLP 139–151 plus IL‐6 for 3 days ex vivo in the presence of MDL‐101 or DMSO. EAE development was significantly lower in SJL mice receiving MDL‐101‐treated CD4 T cells compared to vehicle control treated cells (Fig. 4h and Table 1), further confirming that MDL‐101 suppressed the encephalitogenic potential of myelin‐specific CD4 T cells. To evaluate the in‐vivo efficacy of MDL‐101 treatment, MDL‐101 (50 mg/kg) was injected i.p. daily into immunized C57BL/6 mice from days 10–16 post‐immunization, when 60–80% of mice developed clinical signs of EAE. However, there is no difference in EAE severity between the MDL‐101‐treated and vehicle control‐treated groups (Fig. 4i). A preliminary PK study shows that the half‐life of MDL‐101 in vivo is very short (data not shown). These data demonstrated that MDL‐101 had improved potency in suppressing Th17 development and suppressed encephalitogenic potential of myelin‐specific Th17 cells in vitro. However, more drug optimization is needed to improve in‐vivo pharmacokinetic properties for therapeutic treatment in vivo.

Table 1.

Adoptively transferred experimental autoimmune encephalomyelitis (EAE) development in naive B10.PL or Swiss Jackson Laboratory (SJL) mice adoptively transferred with myelin‐specific CD4 T cells that were treated with 10 μM of madindoline (MDL)‐101 or vehicle control

	Groups	Number of mice	Incidence of EAE	Mean area under the curve
B10PL	Control	17	10/17 (59%)	15·83 ± 3·62
	MDL‐101	17	5/17 (29%)	5·95 ± 4·59
SJL	Control	13	13/13 (100%)	50·27 ± 3·34
	MDL‐101	16	13/16 (81%)	25·84 ± 4·29

MDL‐101 suppresses human Th17 development and promotes T_reg development of human CD4 T cells

To determine whether MDL‐101 suppresses the effector function of human CD4 T cells, human PBMCs were activated with αhCD3 for 3–5 days in the presence of different concentrations of MDL‐101 or vehicle control (DMSO). T cell proliferation was determined by CFSE proliferation assay, while IL‐17 production in supernatants was determined by ELISA. As the activation of myelin‐reactive CD4 T cells in MS patients is independent of CD28‐mediated co‐stimulation 26, 27, 28, we activated human PBMCs with αhCD3 to preferentially activate memory CD45RO⁺CD4⁺ T cells that do not require CD28‐mediated co‐stimulation. As shown in Fig. 5a,b, MDL‐101 significantly suppressed the proliferation of anti‐CD3‐activated CD45RO⁺CD4⁺ T cells at all three concentrations (5, 15 and 25 μM). Furthermore, MDL‐101 significantly suppressed IL‐17 production of human CD4 T cells (Fig. 5c,d) (n = 6). These data suggest that MDL‐101 has the capacity to inhibit the effector function of human CD4 T cells.

Figure 5.

Novel small molecule interleukin (IL)‐6 inhibitor madindoline (MDL)‐101 significantly suppresses Th17 development and promotes regulatory T cell (T_reg) development of human CD4 T cells. (a–d) Peripheral blood mononuclear cells (PBMCs) from six healthy controls (HCs) were labeled with carboxyfluorescein succinimidyl ester (CFSE) and activated with αhCD3 for 5 days in the presence of different concentrations of MDL‐101. Dimethylsulfoxide (DMSO) was used as vehicle control. CFSE was determined by flow cytometric analysis (a,b). Supernatant was collected and IL‐17 was determined by enzyme‐linked immunosorbent assay (ELISA) (c,d). (a) Representative flow plot of CFSE⁺‐proliferating CD4 T cells of one HC. Cells were gated on CD45RO⁺CD4⁺cells. (b) % of proliferating CFSE⁺CD45RO⁺CD4⁺ T cells in each treated group of each patient were summarized and compared by one‐way analysis of variance (anova) (n = 6). (c) Representative IL‐17 ELISA data from one HC. (d) IL‐17 in each treated group of each patient were compared with one‐way anova (n = 6). (e,f) PBMCs from nine HCs were activated with αhCD3/CD28 plus recombinant human transforming growth factor (rhTGF)‐β1, rhIL‐2 and RA for 3 days, in the presence of different concentrations of MDL‐101 or DMSO. CD25⁺forkhead box protein 3 (FoxP3)⁺CD45RA⁺CD4⁺ T_regs were determined by intracellular staining. (e) Representative flow plot of CD25⁺FoxP3⁺CD45RA⁺CD4⁺ T_regs in MDL‐101 or DMSO‐treated group from one HC. Cells were gated on CD45RA⁺CD4⁺ T cells. (f) Group means were calculated and compared with one‐way anova (n = 9). All error bars denote standard error of the mean (s.e.m.). *P < 0·05; ***P < 0·001; ****P < 0·0001.

As the IL‐6/STAT‐3 pathway is a key signaling pathway blocking the development of iT_reg 9, 10, we also determined whether MDL‐101 promotes the development of human iT_regs. Human PBMCs were activated with αhCD3/CD28 for 3 days in the presence of TGF‐β, IL‐2, IL‐6 and trans‐retinoic acid. The percentage of iT_regs from naive CD4⁺CD45RA⁺ T cells in one PBMC sample is shown in Fig. 5e. After 72 h of culture under iT_reg differentiating conditions, the total number of CD25⁺FoxP3⁺ T_regs in the CD4⁺CD45RA⁺ population increased from 11% in the control group to 14% in the group treated with 5 μM of MDL‐101 and 21% in the group treated with 15 μM of MDL‐101. The T_reg data from all the PBMC samples were summarized and compared (Fig. 5f), demonstrating that MDL‐101 has the capacity to promote human T_reg development.

Discussion

Targeting the IL‐6 signaling pathway has been tested and found to be successful in some autoimmune diseases. Tocilizumab, a humanized αIL‐6R monoclonal antibody, has been approved for the treatment of rheumatoid arthritis, Castleman’s disease and juvenile idiopathic arthritis 29, 30, 31, which demonstrates the feasibility and safety of systemic treatment using an anti‐IL‐6 signaling pathway reagent. However, its efficacy in MS patients has not been established. Like all the peptide/protein‐based therapies, monoclonal antibody‐based therapies have the disadvantage of infusion reactions 18 and the generation of blocking antibodies against the drug. In contrast, small molecule compounds typically offer improved bioavailability and manufacturing features over protein‐based drugs. The pathological hallmark of MS is the presence of inflammatory MS lesions in the CNS, suggesting that suppressing infiltrating myelin‐reactive T cells in CNS is critical for MS therapy. Monoclonal antibodies have limited CNS penetration because of their size, thus the concentration of monoclonal antibodies in the CNS could be substantially lower than their concentration in blood. Therefore, the suppression of CNS‐infiltrating T_eff cells and/or the promotion of T_reg development in the CNS may not be as efficient as in the periphery. Small molecules can penetrate the phospholipid membrane of the blood–brain barrier by passive or carrier‐mediated mechanisms 32 and directly exert their function in CNS‐infiltrating T_eff and T_reg cells, leading to better suppression of encephalitogenic T cells in the CNS and disease progression in patients with established MS. However, there are no small molecule compounds targeting IL‐6 commercially available for use in MS/EAE study. To our knowledge, this is the first attempt to modulate the IL‐6 signaling pathway using small molecules for CNS autoimmunity.

T_eff : T_reg balance is critical for the normal function of the human immune system and an increased T_eff : T_reg ratio favors autoimmunity. The IL‐6/STAT‐3 signaling pathway is critical for the highly encephalitogenic Th17 cells while blocking the development of iT_regs. Therefore, we hypothesize that small molecule IL‐6 inhibitors will suppress encephalitogenic Th17 cells and promote T_reg development. Our data show that all three MDL analogs, MDL‐5, MDL‐16 and MDL‐101, suppress the development of encephalitogenic Th17 cells in vitro. MDL‐101 showed improved potency in suppressing Th17 development in vitro compared to MDL‐5 and MDL‐16. Furthermore, MDL‐101 suppressed the encephalitogenic potential of Th17 cells following their subsequent adoptive transfer. More importantly, MDL‐101 suppressed proliferation and IL‐17 production of human CD4 T cells, as well as promoting the development of human T_regs, which will further dampen encephalitogenic CD4 T cells. These data suggest that MDL analogs have great therapeutic potential for MS treatment. However, the potency of MDL‐101 in suppressing Th17 cells and promoting the development of T_regs is still not ideal. Furthermore, therapeutic administration of MDL‐101 in EAE mice did not show suppression on disease severity in EAE mice, suggesting that the in‐vivo pharmacokinetic properties need to be improved. Together, our data suggest MDL‐101 is a good lead compound for further modification and development of small molecule IL‐6 inhibitors with the ultimate goal of treating MS.

As one of the most important cytokine signaling pathways regulating immune responses, blockade of the IL‐6 signaling pathway in the context of normal cytokine signaling may cause immunosuppression which could lead to increased susceptibility to infections, as demonstrated by the recurrent infections commonly seen in hyper‐IgE syndrome patients with STAT3 mutations 33, 34, 35, 36. However, as MS patients have abnormally elevated IL‐6/STAT‐3 signaling, normalizing elevated IL‐6/STAT‐3 signaling in MS patients may restore the T_eff : T_reg balance without severe immunosuppression or increased susceptibility to infections. In fact, there are several ongoing Phase I clinical trials of STAT inhibition for cancer treatment (ClinicalTrials.gov). Phase 1 trial data showed that the adverse events of OPB31121, a small molecule compound inhibiting STAT‐3 and STAT‐5 phosphorylation 37, were predominantly grades 1 or 2, including nausea, vomiting, diarrhea and fatigue 38. Given that STAT‐3 is transiently activated when necessary, controlled administration of IL‐6/STAT‐3 targeting drug is probably safe. Nevertheless, further studies are needed to fully investigate the possible side effects.

Disclosures

The authors declare that they have no competing interests with the contents in this paper.

Author contributions

Y. Y. designed research, analyzed the results and wrote the paper. S. I. A., E. E. K., M. C. G., M. F. F., E. Y. Z. and G. P. performed the experiments. W. P. provided support of mouse studies. C. L conceptualized the IL‐6 small molecule inhibitor design and mapped the structural design. N. J., V. K., L. M. and J. F. designed the synthetic routes and prepared the experimental agents. A. L. R. and M. K. R. helped with data discussion and manuscript review. All authors read and approved the final manuscript.