Inhibition of METTL3 in macrophages provides protection against intestinal inflammation

Abstract

Inflammatory bowel disease (IBD) is prevalent, and no satisfactory therapeutic options are available because the mechanisms underlying its development are poorly understood. In this study, we discovered that increased expression of methyltransferase-like 3 (METTL3) in macrophages was correlated with the development of colitis and that depletion of METTL3 in macrophages protected mice against dextran sodium sulfate (DSS)-induced colitis. Mechanistic characterization indicated that METTL3 depletion increased the YTHDF3-mediated expression of phosphoglycolate phosphatase (PGP), which resulted in glucose metabolism reprogramming and the suppression of CD4+ T helper 1 (Th1) cell differentiation. Further analysis revealed that glucose metabolism contributed to the ability of METTL3 depletion to ameliorate colitis symptoms. In addition, we developed two potent small molecule METTL3 inhibitors, namely, F039-0002 and 7460-0250, that strongly ameliorated DSS-induced colitis. Overall, our study suggests that METTL3 plays crucial roles in the progression of colitis and highlights the potential of targeting METTL3 to attenuate intestinal inflammation for the treatment of colitis.

Keywords: Colitis; Glucose metabolism; Macrophage; N6-methyladenosine modification.

Fig. 1

METTL3-deficient mice are resistant to DSS-induced colitis. A Immunofluorescence staining for CD68 (red) and METTL3 (green) and DAPI staining of nuclei (blue) in healthy human colon tissue and IBD tissue. Scale bars: 30 μm. B Mettl3 mRNA expression in the colons of mice with DSS-induced colitis (n = 6 independent experiments). C Immunofluorescence staining for F4/80 (red) and METTL3 (green) and DAPI staining of nuclei (blue) in colon tissues from Mettl3^fl/fl or Mettl3^fl/flLyz2^Cre mice with DSS-induced colitis. Scale bars: 30 μm. The body weight (D) and colon length (E) of DSS-treated Mettl3^fl/fl and Mettl3^fl/flLyz2^Cre mice were measured during the development of colitis (Mettl3^fl/fl group, n = 9 mice; Mettl3^fl/flLyz2^Cre group, n = 10 mice). F Representative images of H&E-stained colon sections from DSS-treated Mettl3^fl/fl and Mettl3^fl/flLyz2^Cre mice. Scale bars: 100 µm. The histological scores of the colons were determined (Mettl3^fl/fl group, n = 9 mice; Mettl3^fl/flLyz2^Cre group, n = 10 mice). G qRT-PCR analysis of Tnf-α and Il-1β in the colon of DSS-challenged Mettl3^fl/fl and Mettl3^fl/flLyz2^Cre mice (n = 3 independent experiments). H Plasma FITC-dextran concentrations following oral gavage (250 mg/kg body weight) in Mettl3^fl/fl and Mettl3^fl/flLyz2^Cre mice (n = 4 mice per group). I Colon sections from DSS-treated Mettl3^fl/fl and Mettl3^fl/flLyz2^Cre mice were stained with antibodies against Ki67 and ZO-1. J Survival curve of Mettl3^fl/fl and Mettl3^fl/flLyz2^Cre mice after they were challenged with 2.5% DSS in the drinking water. The data are expressed as the means ± SDs. P values were determined by a two-tailed t-test (B, D, E, F, G, H) and the Gehan–Breslow–Wilcoxon test (J). *P ≤ 0.05, ***P < 0.001. NS (nonsignificant) indicates P > 0.05

Fig. 2

METTL3 in macrophages facilitates DSS-induced colitis. A–C WT mice were irradiated, and the bone marrow in these mice was then reconstituted with bone marrow cells from Mettl3^fl/fl or Mettl3^fl/flLyz2^Cre mice for a period of 8 weeks; the mice were then treated with 2.5% DSS for 7 d. Mouse body weights were recorded every day during colitis development (A). Representative images of the colons and data for the colon length in Mettl3^fl/fl and Mettl3^fl/flLyz2^Cre mice after DSS administration (B). Images of H&E-stained colon sections are shown, and the histological score for colitis was calculated (C). n = 8 mice per group. Scale bars: 100 µm. D–F Mettl3^fl/fl and Mettl3^fl/flLyz2^Cre mice challenged with DSS were intraperitoneally injected with control or clodronate liposomes. The development of colitis was monitored by assessing the change in body weight relative to the initial body weight on day 0 (D), the colon length (E), H&E staining of the colon and the histological score (F). (n = 5 mice per group) Scale bars: 100 µm. The data are expressed as the means ± SDs. P values were determined by a two-tailed t-test (A–F). **P < 0.01, ***P < 0.001. NS (nonsignificant) indicates P > 0.05

Fig. 3

METTL3-deficient macrophages suppress Th1 differentiation. A Flow cytometric analysis of the frequencies of CD4⁺IFN-γ⁺, CD4⁺IL4⁺, CD4⁺FoxP3⁺ and CD4⁺IL17A⁺ cells in the colons of Mettl3^fl/fl and Mettl3^fl/flLyz2^Cre mice (n = 6 mice per group). B Percentages of CD4⁺IFN-γ⁺, CD4⁺IL4⁺, CD4⁺FoxP3⁺ and CD4⁺IL17A⁺ cells cocultured with Mettl3^fl/fl or Mettl3^fl/flLyz2^Cre BMDMs (BMDMs:T cells, 1:2) for 96 h (n = 6 independent experiments). The data are expressed as the means ± SDs. P values were determined by a two-tailed t-test (A, B). **P < 0.01, ***P < 0.001. NS (nonsignificant) indicates P > 0.05

Fig. 4

METTL3 promotes the degradation of Pgp mRNA in an m6A-YTHDF3-dependent manner. A Venn diagram analysis comparing upregulated proteins (fold change >1.2, P < 0.05) and upregulated genes (fold change >1.5, P < 0.05) after METTL3 depletion in BMDMs. B Integrative Genomics Viewer (IGV) tracks showing the distribution of m6A peaks on the Pgp and Spr transcripts in Mettl3^fl/fl and Mettl3^fl/flLyz2^Cre BMDMs according to MeRIP-seq data. C Analysis of Pgp and Spr expression in Mettl3^fl/fl and Mettl3^fl/flLyz2^Cre BMDMs by qRT-PCR (n = 3 independent experiments). D Immunoblot analysis of the PGP protein abundance in Mettl3^fl/fl and Mettl3^fl/flLyz2^Cre BMDMs. E Immunofluorescence imaging of F4/80 and PGP in the colons of Mettl3^fl/fl and Mettl3^fl/flLyz2^Cre mice challenged with DSS (n = 5 independent experiments) Scale bars: 100 µm. F MeRIP-qPCR analysis was used to assess Pgp m6A modifications (n = 3 independent experiments). G Pgp expression in Mettl3^fl/fl and Mettl3^fl/flLyz2^Cre BMDMs treated with actinomycin D at the indicated time points was measured via qRT-PCR (n = 3 independent experiments). H RIP-qPCR analysis of the binding of YTHDF3 to Pgp mRNA in Mettl3^fl/fl and Mettl3^fl/flLyz2^Cre BMDMs (n = 3 independent experiments). I qRT-PCR was used to measure the mRNA expression levels of Pgp mutants in 293T cells treated with actinomycin D at the indicated time points (n = 3 independent experiments). qRT-PCR (J) and immunoblot (K) analyses of the expression of different Pgp mutants in Mettl3^fl/fl and Mettl3^fl/flLyz2^Cre BMDMs (n = 3 independent experiments). qRT-PCR (L) and western blot (M) analyses of Pgp mutant expression in 293T cells transfected with NC siRNA or YTHDF3 siRNA (n = 3 independent experiments). The data are expressed as the means ± SDs. P values were determined by a two-tailed t-test (C, E, F, G, H, I, J, L). *P ≤ 0.05, **P < 0.01, ***P < 0.001. NS (nonsignificant) indicates P > 0.05

Fig. 5

METTL3 targets the PPP to mediate the pathogenesis of colitis. A NADPH production in Mettl3^fl/fl and Mettl3^fl/flLyz2^Cre BMDMs was measured. The lactate levels in the culture media of Mettl3^fl/fl and Mettl3^fl/flLyz2^Cre BMDMs were also measured (n = 6 independent experiments). qRT-PCR analysis of the mRNA expression of glycolysis-related (B) and PPP-related genes (C) in Mettl3^fl/fl and Mettl3^fl/flLyz2^Cre BMDMs (n = 3 independent experiments). D Mettl3^fl/fl and Mettl3^fl/flLyz2^Cre BMDMs were incubated with ¹³C-glucose for 24 h before the samples were analyzed via LC–MS. The fractional enrichment of ¹³C-glucose-derived intracellular PPP and TCA cycle metabolites is shown. The designations m + 0, m + 1, etc. are related to the labeled fractions (¹³C₀, ¹³C₁, etc.) (n = 3 independent experiments). E The OCR profiles of Mettl3^fl/fl and Mettl3^fl/flLyz2^Cre BMDMs were monitored with a Seahorse XF24 analyzer (n = 3 independent experiments). F–I DSS-treated Mettl3^fl/fl and Mettl3^fl/flLyz2^Cre mice were intraperitoneally injected with the control treatment or 6-AN. The development of colitis was monitored by evaluating the change in body weight relative to the initial body weight on day 0 (F), the colon length (G), H&E staining of the colon (H), and the histological score (H). Scale bars: 100 µm. Mettl3^fl/fl group, n = 7 mice; other groups, n = 8 mice. The percentage of CD4⁺IFN-γ⁺ cells was analyzed via flow cytometry (I) (n = 6 independent experiments). The data are expressed as the means ± SDs. P values were determined by a two-tailed t-test (A–H). *P ≤ 0.05, **P < 0.01, ***P < 0.001. NS (nonsignificant) indicates P > 0.05

Fig. 6

Identification of METTL3 inhibitors through structure-based virtual screening and validation assays. A Pyramid flowchart showing the approach used to identify METTL3 inhibitors in the Chemdiv2019 database. B Docking poses of the top 63 hits within the catalytic region of the METTL3 protein. C The effects of the top 63 compounds on the enzymatic activity of METTL3 were investigated (n = 3 independent experiments). D 2D ligand interaction diagrams of F039-0002/METTL3 and 7460-0250/METTL3. E Surface plasmon resonance (SPR) sensorgram showing the binding of F039-0002 and 7460-0250 to the METTL3 protein. F IC50 values of F039-0002 and 7460-0250 for inhibiting the enzymatic activity of the METTL3/METTL14 protein complex in vitro (n = 3 independent experiments)

Fig. 7

METTL3 inhibitors protect against DSS-induced colitis. A, B Wild-type mice challenged with DSS were intraperitoneally injected with the control treatment, F039-0002 or 7460-0250. The development of colitis was monitored by assessing the colon length (A), H&E staining of the colon (B), and the histological score (B). n = 9 mice per group. Scale bars: 100 µm. C qRT-PCR and western blot analysis of changes in the PGP mRNA and protein levels in BMDMs after F039-0002 or 7460-0250 treatment (n = 3 independent experiments). D Flow cytometric analysis of CD4⁺IFN-γ⁺ Th1 cells in mice challenged with DSS and administered the control treatment, F039-0002 or 7460-0250 (n = 6 independent experiments). The data are expressed as the means ± SDs. P values were determined by a two-tailed t-test (A–D). *P ≤ 0.05, ***P < 0.001

Fig. 8

Schematic showing that Mettl3 depletion in macrophages increases the expression of PGP, reprograms glucose metabolism and Th1 cell infiltration, and subsequently attenuates colitis in mice. Two novel METTL3 inhibitors, namely, F039-0002 and 7460-0250, showed strong efficacy in inhibiting DSS-induced colitis