The MS Remyelinating Drug Bexarotene (an RXR Agonist) Promotes Induction of Human Tregs and Suppresses Th17 Differentiation In Vitro

Abstract

The retinoid X receptor agonist bexarotene promotes remyelination in patients with multiple sclerosis. Murine studies have also demonstrated that RXR agonists have anti-inflammatory effects by enhancing the ability of all-trans-retinoic acid (atRA) to promote T-regulatory cell (Treg) induction and reduce Th17 differentiation in vitro. By stimulating human naïve CD4 T-cells in the presence of Treg or Th17 skewing cytokines, we show that bexarotene also tips the human Treg/Th17 axis in favor of Treg induction, but unlike murine cells this occurs independently of atRA and retinoic acid receptor signaling. Tregs induced in the presence of bexarotene express canonical markers of T-regulation and are functionally suppressive in vitro. Circulating Treg numbers did not increase in the blood of trial patients receiving bexarotene; we believe this is because Treg induction is likely to occur within tissues. These findings lend support to developing RXR agonists as treatments of autoimmune diseases, in particular multiple sclerosis.

Keywords: T cells; Th17 & Tregs cells; autoimmunity; multiple sclerosis; retinoid X receptor (RXR); retinoids.

Figure 1

RXR agonists promote human iTreg differentiation independent of RAR signaling. (A) Gating strategy and example dot plots showing iTreg induction following culture of naïve T cells for 7 days in iTreg skewing conditions +/- bexarotene (1 μM), +/- atRA (100 nM) or both (B) Summary data for n=6 experiments (C, D) percent iTreg induction over a range of atRA and bexarotene concentrations. (E) Co-culture with the RAR antagonist AGN reverses increased iTreg induction seen in the presence of atRA, but not Bexarotene. (F, G) The RXR agonists 9CisRA and NRX194204 also promote human iTreg differentiation in vitro. Shown are mean values ± SEM. Significance calculated using a repeated measures one-way ANOVA. Šidák’s correction was performed. *p < 0.05, **p < 0.01, ***p < 0.001 ns, not significant.

Figure 2

Comparison of iTregs vs donor-matched nTregs. (A) Percent suppression of CD3/CD28 stimulated Teff in the presence of iTregs or iTregs induced in the presence of bexarotene (two-tailed ANOVA, ns). (B) Phenotypic characteristics of iTregs, induced +/-bexarotene, +/-atRA or both vs. donor-matched nTregs stained immediately (ex-vivo) or following culture for 7 days (5 μM CD3, 1 μM CD28, 1 ng/ml IL-2). Shown are mean values ± SEM. iTreg, n = 4. ex vivo iTreg, n=4. nTreg + stim, n = 3. Significance was calculated with one-way ANOVA, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 (Tukey’s correction). (C) FOXP3 TSDR methylation, determined by bisulfite sequencing, of iTregs and iTregs induced in the presence of bexarotene vs. donor-matched nTregs. Owing to X inactivation FOXP3 TSDR demethylation is ∼50% lower in females compared to males. Plots from two representative donors (one male, one female) are shown; total donors n = 6.

Figure 3

Bexarotene significantly reduces the differentiation of CD4+ naive lymphocytes into Th17 IL-17a+ cells in vitro. (A) Representative dot plots showing intracellular Th17 and IFN-γ following culture of human naiïve CD4 cells in Th17 skewing conditions +/- bexarotene, +/-atRA or both. (B) Summary data showing percent IL-17+ cells as a percent of total live CD4 cells at the end of culture (n = 7). (C) Supernatant IL-17a as measured at day 7 (n = 7). Shown are mean values ± SEM. Significance was calculated with repeated measures one-way ANOVA. Šidaák’s correction was performed. **p < 0.01, ****p < 0.0001.

Figure 4

Peripheral blood immune phenotyping of patients randomized to receive bexarotene on the CCMR-One trial. Samples were analyzed at baseline, then at months 2, 4 and 6 following treatment. (A) Frequency of CD3+CD4+ cells as a percent of total lymphocytes. (B) Frequency of CD25hiCD127−FOXP3+ Tregs as percent of CD4+ cells. (C) Frequency of Helios+ cells as a percent of CD25hiCD127−FOXP3+ Tregs. Shown are mean values ± SEM, n=5. (D) Absolute Treg counts as defined by CD4+CD25hiCD127low surface phenotype. (E) Representative longitudinal plots of Treg gating and quantification. FOXP3 and HELIOS expression for these populations are also shown. (F) FOXP3 TSDR methylation of flow sorted, circulating CD4+CD25hi CD127low Tregs pre- and post-bexarotene (n=5; two representative donors are shown). (G) Frequency IL-17a+ cells, (H) IL-17a supernatant concentration and (I) absolute Th17 cell counts (as measured by IL-17a positivity). (J) Representative longitudinal plots of intracellular IL17 and IFN-γ in CD3+CD4+ cells, from patients pre- and post-bexarotene, following stimulation with PMA + Ionomycin + Brefeldin A. Each sample was run in duplicate. For further information regarding CCMR-One subject demographics, refer to Table 1.