A regulatory loop involving the cytochrome P450-soluble epoxide hydrolase axis and TGF-β signaling

Abstract

Fatty acid metabolites, produced by cytochrome P450 enzymes and soluble epoxide hydrolase (sEH), regulate inflammation. Here, we report that the transforming growth factor β (TGF-β)-induced polarization of macrophages to a pro-resolving phenotype requires Alk5 and Smad2 activation to increase sEH expression and activity. Macrophages lacking sEH showed impaired repolarization, reduced phagocytosis, and maintained a pro-inflammatory gene expression profile. 11,12-Epoxyeicosatrienoic acid (EET) was one altered metabolite in sEH^-/- macrophages and mimicked the effect of sEH deletion on gene expression. Notably, 11,12-EET also reduced Alk5 expression, inhibiting TGF-β-induced Smad2 phosphorylation by triggering the cytosolic translocation of the E3 ligase Smurf2. These findings suggest that sEH expression is controlled by TGF-β and that sEH activity, which lowers 11,12-EET levels and promotes TGF-β signaling by metabolizing 11,12-EET to prevent Alk5 degradation. Thus, an autocrine loop between sEH/11,12-EET and TGF-β1 regulates macrophage function.

Keywords: Biochemistry; Cell biology; Molecular biology; Omics; Transcriptomics.

Graphical abstract

Figure 1

Consequence of impaired sEH activity on the resolution of inflammation and sEH expression in macrophages (A) Cells present in peritoneal fluid from wild-type and sEH^−/− mice up to 6 days after a single intraperitoneal injection of zymosan A (10 mg/kg). Neutrophils, Res. MØ, resident macrophages; Rec. MØ, recruited macrophages; DCs, dendritic cells; n = 5–6 animals per group. (B and C) Ephx2 RNA (B) and sEH protein (C) in murine bone-marrow-derived macrophages under basal conditions M0 (Sol) and after polarization to the M1, M2a, or M2c states; n = 6 mice per group. M2c macrophages from sEH^−/− mice were included as a negative control. (D) Concentration-dependent effects of TGF-β1 (48 h) on sEH expression in M1-polarized murine macrophages; n = 5 mice per group. (E) sEH activity in polarized macrophages; n = 5 independent experiments. M2c-polarized macrophages from sEH^−/− mice were included as a negative control. Error bars = SEM, (A) Two-way ANOVA followed by Sidak’s multiple comparisons test; (B–E) one-way ANOVA and Tukey’s multiple comparisons test. ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.

Figure 2

Consequence of impaired sEH activity on macrophage phagocytosis and polarization (A and B) pHrodo Red Zymosan uptake by M2c-polarized macrophages from (A) wild-type (WT) and sEH^−/− (−/−) mice or (B) from WT M2c macrophages treated with solvent or an sEH inhibitor (sEH-I, t-AUCB, 5 μmol/L); scale bar = 200 μm; n = 12 mice per group. (C) Oxidized LDL (ox-LDL) uptake by M2c-polarized macrophages from WT and sEH^−/− mice; scale bar = 50 μm; n = 10 mice per group. (D) Expression marker genes in bone-marrow-derived macrophages from wild-type (WT) and sEH^−/− (−/−) mice under basal conditions (M0) and following M1, M2a, and M2c polarization; n = 5–6 mice per group. Error bars = SEM, (A–C) Student’s t test; (D) two-way ANOVA followed by Sidak’s multiple comparisons test. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗∗p < 0.0001.

Figure 3

Role of ALK5-Smad2 pathway in the regulation of sEH expression by TGF-β1 (A) Effect of TGF-β1 (10 ng/mL) on the phosphorylation of SMAD2 (p-SMAD2) and SMAD1/5 (p-SMAD1/5) in M1-polarized murine macrophages; n = 5–6 mice per group. (B) Expression of Snail and Id3 in M2c-polarized macrophages; n = 5–6 mice per group. (C) Consequence of ALK1 inhibition (ALK1i) with LDN193189 (1 nmol/L) and ALK5 inhibition (ALK5i) with SD208 (500 nmol/L) on the TGF-β1-induced increase in sEH expression in murine macrophages; n = 5–6 mice per group. (D) Effect of TGF-β1 (10 ng/mL, 24 h) on sEH promoter activity in HEK cells; n = 6 independent experiments. (E) ChIP-qPCR analysis of SMAD2 binding to human sEH promoter in M1-polarized human macrophages stimulated with TGF-β1 (10 ng/mL, 24 h); n = 5–6 donors. (F) Effect of decoy oligonucleotides directed against SMAD2 and SP1 on the TGF-β1-induced increase of sEH expression in murine M1 macrophages; n = 5–6 mice per group. (G and H) Expression of Ephx2 (g) and sEH (h) in M2c-polarized macrophages from wild-type (WT) and SMOC1^+/− (+/−) littermates; n = 6 mice per group. Error bars = SEM, (A) One-way ANOVA followed by Turkey’s multiple comparisons test; (B, C, E, G, and H) Student’s t test; (D and F) two-way ANOVA followed by Sidak’s multiple comparisons test. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.

Figure 4

RNA-seq analysis and validation (A) Volcano plot showing differential gene expression in M2c macrophages from wild-type and sEH^−/− mice; n = 3 mice per group. Significantly regulated genes are indicated in blue (higher in sEH^−/−) and green (higher in wild-type). (B) Heatmap showing the top 50 differentially expressed genes (Z scores) as in (A). (C–E) Bone-marrow-derived monocytes from WT and sEH^−/− mice were differentiated to M0, M1, or M2c macrophages. Nlrp3 gene expression (C), Nlrp3 protein levels (D), and IL-1β levels in the cell supernatant (E); n = 4–5 mice per group. (F) Heatmap showing most differentially regulated inflammation- and resolution-related transcripts in cells from peritoneal fluid 6 days after zymosan administration; n = 4–5 mice per group. (G) Expression of Tlr2 and Mrc1 in M2c-polarized macrophages from wild-type (WT) and sEH^−/− (−/−) mice; n = 5–6 mice per group. (H) Tlr2 and Mrc1 in macrophages from wild-type mice polarized to M2c in the presence of solvent (Sol) or an sEH inhibitor (sEH-I); n = 5–6 mice per group. (I) Tlr2 and Mrc1 expression in M2c-polarized macrophages from sEH^−/− mice overexpressing either GFP, the wild-type sEH (WT), or a catalytically inactive sEH mutant (ΔEH); n = 5–6 mice per group. Error bars = SEM, (C–E) Two-way ANOVA followed by Tukey’s multiple comparisons test; (G and H) Student’s t test; (I) one-way ANOVA followed by Turkey’s multiple comparisons test. ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.

Figure 5

PUFA mediator profiling in human macrophages (A) PUFA mediator levels in M1- and M2c-polarized human-monocyte-derived macrophages; n = 6–7 donors per group. Red = significantly elevated in M2c, blue = significantly elevated in M1, gray = non-significant. (B) PUFA mediator levels in human macrophages polarized from M1 to M2c in the presence of solvent (Sol) or the sEH inhibitor (t-AUCB, 10 μmol/L); n = 6–7 donors per group. (C) Effect of the selected arachidonic acid (AA) and linoleic acid (LA) epoxide (E)-diol (D) pairs, i.e., 11,12-EET/DHET and 9,10-EpOME/DiHOME (all 1 μmol/L, 48 h) on Tlr2 and NLRP3 expression in M1-polarized human macrophages; n = 4 donors. (D) Impact of 11,12-EET (E) and 11,12-DHET (D; both 1 μmol/L, 48 h) on the expression of Tlr2 and Nlrp3. Experiments were performed in the absence and presence of the EET antagonist 14,15-epoxyeicosa-5(Z)-enoic acid (EEZE; 1 μmol/L); n = 5 donors. Error bars = SEM, (C and D) One-way ANOVA followed by Turkey’s multiple comparisons test. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.

Figure 6

Effect of altered sEH expression on TGF-β signaling (A) Snai1 expression in M2c macrophages from wild-type (WT) and sEH^−/− (−/−) mice; n = 4 animals per group. (B) TGF-β-stimulated phosphorylation of Smad2 in M1-polarized macrophages; n = 5 animals per group. (C) Expression of Alk5, Alk1, and TgfβrII in M1-polarized macrophages; n = 6 animals per group. (D and E) Impact of sEH inhibition (sEH-I) on the TGF-β (15 min)-induced phosphorylation of SMAD2 phosphorylation in M1 macrophages from wild-type mice (D) and expression of Alk5 (e); n = 5 animals per group. (F) Alk5 expression in M1 and M2c macrophages in the absence and presence of MG132; n = 5 mice per group. The image shows non-adjacent lanes from the same blot with the excised section indicated by a break. (G) Nuclear (n) and cytoplasmic (c) levels of Smurf2 in cells treated with Solvent (0.1% DMSO) or 11,12-EET (3 mmol/L) scale bar = 10 μm; n = 6 animals per group. (H) Consequence of siRNA-mediated Smurf2 knockdown on Alk5 expression in M1 macrophages; comparable results were obtained using an additional three animals per group. Error bars = SEM, A, C, D, E, and G) Student’s t test; (B and F) two-way ANOVA followed by Sidak’s multiple comparisons test. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.