Stimulation of the farnesoid X receptor promotes M2 macrophage polarization

Abstract

FXR is a key molecule that modulates anti-inflammatory activity in the intestinal-liver axis. Although FXR has pleiotropic functions including regulation of liver inflammation and activation of macrophages, it remains unclear whether it is involved in macrophage polarization. In this paper we demonstrated that stimulation of macrophages derived from the bone marrow using an FXR agonist activated polarization toward M2 but not M1 macrophages. The treatment of mice with chitin skewed macrophage polarization towards M2 macrophages, while co-treatment with an FXR agonist further promoted the polarization toward M2 macrophages in vivo. This skewed polarization towards M2 macrophages by an FXR agonist was accompanied by increased expression of signaling molecules related to the retinoic acid receptor. Inhibition of the retinoic acid receptor suppressed FXR agonist-mediated M2 macrophage polarization, indicating that this polarization was, at least, partly dependent on the retinoic acid receptor pathway. These data demonstrate that FXR has a role in polarization toward M2 macrophages and suggest a possible therapeutic potential of FXR agonists in M2 macrophage-related conditions.

Keywords: cell differentiation; farnesoid X receptor; inflammation; macrophage; tissue repair.

Figure 1

Activation of FXR in macrophages by GW4064 treatment selectively enhances M2- but not M1-associated genes. (A) mRNAs from peritoneal macrophages, BMDMs, splenic myeloid cells, dendritic cells, T cells, and B cells were extracted and converted to cDNA. Relative FXR mRNA gene (Nr1h4) expression in these cells was determined by real-time PCR. (B) Schematic representation of the M2 macrophage polarization protocol. Bone marrow cells were isolated from mice femurs and cultivated for 7 d in the presence of recombinant M-CSF to become BMDMs. M2 polarization was then induced for 1 d using 10 ng/mL of recombinant IL-4. (C) BMDMs were treated with 10 µM of GW4064 under M1 or M2 macrophage-inducing conditions and left untreated as a Neu condition for 1 d. RT-qPCR analysis for M2-related genes was then conducted. (D) Real-time PCR analysis for M1-related genes was performed by using the same samples used in (C). Each experiment was performed in triplicate and repeated three times. The results are representative data, expressed as mean +/- standard deviation. Analysis of variance (ANOVA) was used for the statistical analysis. **P < 0.01. Nr1h4, nuclear receptor subfamily 1 group H member 4; M-CSF, macrophage colony-stimulating factor; Neu, neutral.

Figure 2

Activation of FXR with GW4064 increased expression of M2-associated markers. (A) BMDMs were treated with or without 10 µM of GW4064 under the M2 condition from 1 to 3 d, and left untreated as a Neu condition and M2-associated gene expression then measured by real-time PCR. (B) ARG-1 protein (35 kDa) expression was examined by immunoblotting on day 2 of GW4064 treatment under M2 conditions, with relative ARG-1 expression normalized to β-actin (42 kDa) expression (left panel). The graph shows the relative expression of ARG-1 protein in GW4064-treated cells compared to that of the vehicle control group (right panel). (C) BMDMs were treated with 0, 5, or 20 µM of GW4064 under the M2 condition for 3 d. Flow cytometry analysis of CD11b⁺ F4/80⁺ CD206⁺ and CD206 expression on the cell surface was performed, with the results expressed as the percentage of CD206 in CD11b⁺ F4/80⁺ cells (upper panel) and MFI (lower panel). Each experiment was carried out in triplicate and repeated three times. The results of the representative data are shown as mean +/- SD. Analysis of variance (ANOVA) and the Student’s t test were used for the statistical analyses. MFI: mean fluorescence intensity. *P < 0.05, **P < 0.01.

Figure 3

Rarb and Cyb26b1 expression was increased by FXR activation with GW4064 in M2 macrophages. (A) BMDMs were treated with DMSO or 10 µM of GW4064 under M2-inducing conditions for 1 d. The mRNAs were isolated and checked for integrity and a DNA micro array then used to profile differential gene expression. The heatmap analysis shows profiling of gene alterations according to biological process annotations and pathway analysis. (B) A diagram showing that RARB and CYP26B1 are included in the retinoic acid signaling pathway. (C) Real-time PCR was performed to assess Rarb and Cyb26b1 mRNA gene expression in BMDMs treated with DMSO, 10 µM of GW4064, or 2 µM of LE540 under M2-inducing conditions for 1 d and left untreated as a Neu condition. Each experiment was performed in triplicate and repeated three times. The results of the representative data are shown as mean +/- SD. Analysis of variance (ANOVA) was used for the statistical analysis. ** P < 0.01. Rarb: Retinoic acid receptor beta; Cyp26b1: Cytochrome P450 family 26 subfamily B member 1.

Figure 4

Inhibition of the retinoic acid receptor by LE540 reduced expression of M2-related markers induced by GW4064-mediated FXR activation. BMDMs were treated with 2 and 10 μM of LE540 with or without 10 μM of GW4064 under M2 conditions for 1 d. Messenger RNA gene expression of Arg1 (A), Retnla (B) and Chil3 (C) was evaluated by RT-qPCR. (D) Immunoblotting analysis of ARG-1 (35 kDa) expression in BMDMs treated with 2 and 10 μM of LE540 with or without 10 μM of GW4064 under M2 conditions for 2 d and left untreated as a Neu condition (left panel). The expression level of ARG-1 was normalized to β-actin (42 kDa) expression, with the graph showing the relative expression of ARG-1 in BMDMs treated with or without LE540 in the presence of GW4064 compared to that observed with the control M2 condition (right panel). Each experiment was performed in triplicate and repeated three times. The results of the representative data are shown as mean +/- SD. Analysis of variance (ANOVA) was used for the statistical analysis. *P < 0.05, **P < 0.01.

Figure 5

Administration of a retinoic acid receptor inhibitor reduced GW4064-enhanced M2 macrophages in chitin-treated mice. (A) The schematic protocol to induce M2 peritoneal macrophages by injection of chitin particles and assess the effects of GW4064 and LE540 on M2 macrophage induction in vivo. The mice received an intraperitoneal injection of 50 µg of chitin to prime M2 induction for 3 hr and then either 30 mg/kg bodyweight of GW4064 or GW4064 together with 2 mg/kg bodyweight of LE540 were administered two times within a 24 hr interval. (B) Peritoneal macrophages were collected for further analysis. Messenger RNA expression of M2-associated genes in collected peritoneal macrophages (PMs) was analyzed by real-time PCR. (C) Flow cytometric analysis of CD206⁺ (M2) and CD11c⁺ (M1) from CD11b⁺ F4/80⁺ PMs. The frequency and total numbers of each population are shown in the lower panel. Five mice were used in each experimental group, except for the control group Neu in which four mice were used. Each experiment was repeated two times independently (n=4-5 mice/group), and the result of the representative data shown as mean +/- SE. Analysis of variance (ANOVA) was used for the statistical analysis. *P < 0.05, **P < 0.01.