Fig 2: Loss of METTL3 impairs the YTHDF1-mediated translation of SPRED2.a Metagene profiles of the m6A distribution across the transcriptome in WT and KO BMDMs. b Consensus sequence motif for m6A methylation identified in WT and KO BMDMs. c Overlap between m6A-downregulated genes (fold-change >4 and P value < 0.00001) in BMDMs and MAPK pathway-related functional genes. d Scatter plots showing m6A-related changes in cellular transcript levels in KO BMDMs versus WT BMDMs. e m6A enrichment in Spred2 mRNA in WT and KO BMDMs, as determined by m6A-RIP-qPCR. n = 3 independent experiments. f qRT-PCR (left panel) of SPRED2 expression in WT and KO BMDMs (n = 3 mice per group). The expression of SPRED2 in BMDMs (middle panel) and TAMs (right panel) was determined by immunoblotting. The blots are representative of n = 3 independent experiments. g Spred2 mRNA stability in WT and KO BMDMs treated with actinomycin D at the indicated times determined by qRT-PCR (n = 3 mice per group). h A representative polysome gradient profile for WT and KO BMDMs. i Analysis of Spred2 mRNA in the non-ribosomal fraction (<40S), 40S, 60S, 80S and polysomes of KO BMDMs compared to those of control cells. n = 3 independent experiments. j Schematic representation of mutations in the conserved coding sequence (CDS) for the investigation of the roles of m6A in SPRED2 expression. k FLAG-SPRED2-CDS WT, FLAG-SPRED2-CDS mut1, FLAG-SPRED2-CDS mut2 or FLAG-SPRED2-CDS mut1/2 was transfected into BMDMs for 24 h. Protein expression was measured by western blotting. The blots are representative of n = 3 independent experiments. l RNA immunoprecipitation (RIP) analysis of the interaction of Spred2 mRNA transcripts with YTHDF1 in WT and KO BMDMs. Enrichment of Spred2 was measured by qRT-PCR. n = 3 independent experiments. Data are means ± SD. P values were determined by hypergeometric test (b) and two-tailed t-test (c, e–g, i and l). *P ≤ 0.05 and **P < 0.01. NS (non-significant) means P > 0.05. The source data are provided as a Source data file.

Fig 3: Low SPRED2 gene expression correlates with poor survival of patients with breast tumors. (A) METABRIC dataset analysis showing the mutation types of SPRED2 in ERa+ breast cancer patients. (B) COSMIC dataset analysis showing the SPRED2 methylation status in breast cancer patients. Of the total number of 1414 samples in this dataset, all but one of the 545 samples are hypermethylated at position chr2:65313277 (located in the 3'UTR and coinciding with a CTCF binding site). (C–G) Kaplan–Meier plots for the overall survival (OS) and the relapse free survival (RFS) of all (C,D), ERa+ (E), tamoxifen-treated ERa+ (F), and ERa– (G) breast cancer patients, classified as tumors expressing high levels (red line) and low levels (black line) of SPRED2 mRNA.

Fig 4: SPRED2 deficiency enhances cell proliferation, migration, and induces resistance to 4-OHT in ERa+ BC cells. (A,B) Bar graphs showing the RT-qPCR quantitation of SPRED2 mRNA levels in cells transduced with viral particles for the expression of two different shRNAs targeting SPRED2 or of a negative control shRNA in MCF7-V (shScrambled) (A) and of the empty vector in T47D (shEV) (B) cells. (C) Relative proliferation of MCF7-V cells expressing shScrambled and shSPRED2, measured with a crystal violet assay in monolayer culture. The numbers of cells for each of them are standardized to the corresponding values at 24 h set to 1. (D) Cell migration assay with MCF7-V cells with and without SPRED2 knockdown. Representative images of the scratch wound areas, outlined with magenta, at the time of the wounding (0 h) and after 24 h. Images of all four replicates (including these) are presented in Figure S2C,D. (E) Bar graphs showing the means of closure for MCF7-V cells. The extent of closure was analyzed with the software ImageJ. (F) Cell growth analyzed by crystal violet staining of MCF7-V cells with and without SPRED2 knockdown and treated with increasing doses of 4-OHT. (G) Dose–response curves of MCF7-V cells treated with increasing doses of 4-OHT. WT, parent MCF7-V cells. (H) Cell growth analyzed by crystal violet staining of T47D cells treated with increasing doses of 4-OHT. (I) Dose–response curves of T47D cells treated with increasing doses of 4-OHT; for panels (F–I), n = 3 independent experiments in triplicates. All error bars represent the standard errors of the means (mean ± SEM). Asterisks indicate significant differences compared to the samples with shScrambled (MCF7-V) or shEV (T47D) (ns for p > 0.05, * p = 0.05, ** p = 0.01, *** p = 0.001 and **** p = 0.0001). Statistical significance was determined with a one-way ANOVA (A), a two-tailed unpaired t-test (B,E), and a two-way ANOVA with Bonferroni multiple comparison test for all other panels.

Fig 5: Mettl3 depletion in macrophages impairs PD-1 blockade therapeutic efficacy in B16 tumours.a–f B16 cells were intravenously injected into WT and KO mice. An anti-PD-1 Ab was injected on days 0, 3, 6, 9, 12. Bioluminescence was measured at the indicated time points, and representative images are shown (a) and were quantified (b) (n = 6 mice per group). Lungs were dissected and photographed (c), and representative HE staining images of lung sections are shown (d). Lung metastatic nodules (e) (n = 6 mice per group) and overall survival (f) were observed (WT, KO or WT + PD-1 Ab group, n = 6 mice; KO + PD-1 Ab group, n = 7 mice). g Schematic showing that Mettl3 ablation in myeloid cells impairs the translation of SPRED2 mediated by YTHDF1, which facilitates tumour progression by regulating cytokine responses and enhancing tumour-promoting macrophage and Treg infiltration. Data are shown as the mean ± SD. P values were determined by a two-tailed t-test (b, e) and the Gehan–Breslow–Wilcoxon test (f). *P ≤ 0.05, **P < 0.01 and ***P < 0.001. The source data are provided as a Source data file.