METTL16 is highly expressed in HSPCs among vertebrate

(A) Uniform manifold approximation and project (UMAP) visualization of 21 hematopoietic cell clusters from the fetal liver of humans. HRA002414. BMSC bone marrow stromal cells. (B) Dot plot showing the expression of METTL genes in hematopoietic cell clusters in (A). (C) UMAP visualization of 22 hematopoietic cell clusters from the fetal liver of mice. CRA006858. LMP lymphoid-primed multipotent progenitors, EBP eosinophil/basophil progenitors. (D) Dot plot showing the expression of METTL genes in hematopoietic cell clusters in (C). (E) UMAP visualization of 27 hematopoietic cell clusters from the caudal hematopoietic tissue (CHT) of zebrafish. GSE146404. (F) Dot plot showing the expression of METTL genes in hematopoietic cell clusters in (E).

HSPC deficiency initiates in the CHT of mettl16^−/− embryos.

(A–H) The whole-mount in situ hybridization (WISH) assay shows the expression of HSPC marker cmyb in the CHT of siblings and mettl16 mutants from 2 to 5 dpf. Numbers at the bottom right indicate the number of embryos with similar staining patterns among all embryos examined. n = 3 independent experiments. The black arrowheads indicate cmyb⁺ cells in the CHT. Scale bars, 100 μm. (I) Quantification of WISH in (A–H). ‘n’ indicates the number of individuals analyzed for cmyb⁺ cells in the CHT of siblings and mettl16 mutants, respectively, from 2 to 5 dpf. (J) Live imaging showing the number of EGFP-positive HSPCs in the CHT region of siblings/Tg (cmyb: EGFP) and mettl16^−/−/Tg (cmyb: EGFP) zebrafish from 2 to 5 dpf. Numbers at the bottom right indicate the number of embryos with similar staining patterns among all embryos examined. n = 3 independent experiments. Scale bars, 20 μm. (K) Quantification of live imaging in (J). ‘n’ indicates the number of individuals analyzed for GFP⁺ cells in the CHT of siblings and mettl16 mutants at 3 and 4 dpf, respectively. (L) Expression of differentiation markers β-globin (erythroid), hbae3 (erythroid), mpx (myeloid), l-plastin (myeloid), and rag1 (lymphoid) in sibling and mettl16^−/− embryos at 5 dpf by WISH. Numbers at the bottom right indicate the number of embryos with similar staining patterns among all embryos examined. n = 3 independent experiments. The black arrowheads indicate differentiated blood cells. Scale bars, 100 μm. (M) qPCR analysis showing the expression of erythroid cell markers (upper) and myeloid cell markers (lower) in mettl16 mutants at 5 dpf. n ≥ 15 per group, performed with three biological replicates. Data information: In (I,K,M), data were represented as mean ± SEM. *adjusted P < 0.05, **adjusted P < 0.01, ***adjusted P < 0.001, **** adjusted P < 0.0001, n.s. non-significant, Student’s t-test. Source data are available online for this figure.

Depletion of Mettl16 inhibits HSPC proliferation through G1/S cell cycle arrest.

(A) Double immunostaining of cmyb: EGFP and EDU showing the number of proliferating HSPCs in the CHT of siblings and mettl16 mutants from 2 dpf to 5 dpf. Numbers at the bottom right indicate the number of embryos with similar staining patterns among all embryos examined. n = 3 independent experiments. The white arrowheads indicate proliferating HSPCs. Scale bars, 40 μm. (B) Double immunostaining of cmyb: EGFP and PCNA (upper panels) or pH3 (lower panels) showing the number of proliferating HSPCs in the CHT of siblings and mettl16 mutants at 5 dpf. Numbers at the bottom right indicate the number of embryos with similar staining patterns among all embryos examined. n = 3 independent experiments. Scale bars, 40 μm. (C, D) Flow analysis showing the cell cycle of HSPCs of mettl16-deficient zebrafish at 3 dpf. n ≥ 200 per group, performed with three biological replicates. (E) Flow analysis showing the cell cycle of HSPCs of mettl16-deficient zebrafish at 5 dpf. n ≥ 200 per group, performed with three biological replicates. Data information: In (A, B, D), data were represented as mean ± SEM, * adjusted P < 0.05, ** adjusted P < 0.01, *** adjusted P < 0.001, **** adjusted P < 0.0001, n.s. non-significant, Student’s t-test. Source data are available online for this figure.

Disruption of Mettl16 invokes the alteration of m⁶A in cell cycle genes.

(A) Schematic of Mettl16-WT and two Mettl16 mutants, including exclusive cytoplasmic Mettl16 (del-NLS) and catalytic-dead Mettl16 (PP180/181AA). (B, C) qPCR (B) and WISH (C) analysis showing the decreased expression of HSPC marker cmyb in mettl16-deficient zebrafish at 4 dpf was restored by Mettl16-WT, but not Mettl16-del-NLS and Mettl16-PP180/181AA. n ≥ 10 per group, performed with three biological replicates (B). Numbers at the bottom right indicate the number of embryos with similar staining patterns among all embryos examined, n = 3 independent experiments (C). Scale bars, 100 μm (C). (D) Quantification of WISH in (C). ‘n’ indicates the number of individuals analyzed for cmyb⁺ cells in the CHT at 4 dpf, respectively. (E) m⁶A enrichment in RNAs in the CHT of siblings and mettl16-deficient zebrafish at 3 dpf. The m⁶A/A level of siblings was set to one. n ≥ 20 per group, performed with three biological replicates. (F) Volcano plots reveal the differential m⁶A peaks of total RNAs in mettl16^−/− mutants compared with siblings at 3 dpf. The significantly increased and decreased m⁶A peaks are highlighted in red and blue, respectively (P ≤ 0.05, |log₂FC | ≥1). Statistical analysis was performed by a hypergeometric test. (G) Gene ontology analysis of the m⁶A peaks that decrease upon mettl16 knockout. Statistical analysis was performed by a hypergeometric test. (H) Pie chart depicting the annotations of the m⁶A peaks that decrease upon mettl16 knockout. (I) Pie chart depicting the percent of the significantly decreased m⁶A peaks which have a UACAGAGAA highly similar motif upon mettl16 knockout. (J) Heatmap showing differentially expressed genes in mettl16^−/− mutants compared with siblings at 3 dpf. (K) GSEA plot showing the general downregulation of genes involved in the regulation of cell proliferation in mettl16^−/− mutants compared with siblings. Statistical analysis was performed by permutation test using GSEA software. Data information: In (B, D, E), data were represented as mean ± SEM, *adjusted P < 0.05, ** adjusted P < 0.01, **** adjusted P < 0.0001, n.s. non-significant, one-way ANOVA analysis with post hoc test of Tukey’s multiple comparison correction (B, D), Student’s t-test (E). Source data are available online for this figure.

mybl2b is a functionally essential target of Mettl16 in HSPC development.

(A) Venn diagram showing the overlap between hypomethylated transcripts and differentially expressed genes in mettl16^−/− mutants compared with siblings. (B) Integrative Genomics Viewer (IGV) tracks displaying MeRIP-seq (upper panels) and RNA-seq (lower panels) read distribution in mybl2b mRNA of siblings and mettl16^−/− mutants. The red arrow at the bottom of the tracks indicates the position of the m⁶A peak. (C, D) qPCR analysis showing that the mRNA expression level of mybl2b, trpa1b, and ntrk2b in mettl16^−/− embryos at 3 dpf and 4 dpf. n ≥ 15 per group, performed with three biological replicates. (E, F) qPCR analysis showed the mRNA expression level of mybl2b in mettl16 morphants at 3 dpf and 4 dpf. n ≥ 15 per group, performed with three biological replicates. (G) m⁶A enrichment in mybl2b mRNA in mettl16 deficient zebrafish at 3 dpf by meRIP-qPCR. n ≥ 200 per group, performed with three biological replicates. (H, I) WISH (H) and qPCR (I) analysis showed that the decreased expression of HSPC marker cmyb in mettl16 morphants at 4 dpf was restored by mybl2b mRNA co-injection. Numbers at the bottom right indicate the number of embryos with similar staining patterns among all embryos examined, n = 3 independent experiments (H). Scale bars, 100 μm (H). n ≥ 10 per group, performed with three biological replicates (I). (J, K) Flow analysis of the cell cycle showing the increase of HSPCs in the G0/G1 phase in mettl16 morphants was restored by mybl2b mRNA co-injection. n ≥ 200 per group, performed with three biological replicates. Data information: In (C–I, K), data were represented as mean ± SEM, *adjusted P < 0.05, ** adjusted P < 0.01, *** adjusted P < 0.001, **** adjusted P < 0.0001, n.s. non-significant, Student’s t-test (C–G), one-way ANOVA analysis with post hoc test of Tukey’s multiple comparison correction (H, I, K). Source data are available online for this figure.

Mettl16 regulates mRNA stability of mybl2b through m⁶A reader protein Igf2bp1

(A) Predicted secondary structures surrounding m⁶A peak in mybl2b mRNA. The UACAGAAAAA box was shown in red and the m⁶A modification site was shown in gray circle. (B) Native RIP of Mettl16 with mybl2b mRNA in wild-type zebrafish. n ≥ 200 per group, performed with three biological replicates. actin served as negative control. (C) qPCR analysis of embryos treated with actinomycin D for 4 and 8 h showed accelerated mybl2b mRNA degradation in mettl16 morphants compared to control. n ≥ 20 per group, performed with three biological replicates. (D) Native RIP of Igf2bp1 with mybl2b mRNA in wild-type zebrafish. n ≥ 200 per group, performed with three biological replicates. gapdh and thor were served as negative and positive control, respectively. (E, F) Validation of the knockdown (KD) efficiency of the shRNAs against METTL16 by qPCR (E) and western blot (F). n = 3 biological replicates. (G, H) Flow analysis showing increased G0/G1 phase in METTL16 knockdown K562 cells. n = 3 biological replicates. (I) m⁶A enrichment in MYBL2 mRNA in METTL16 knockdown K562 cells by meRIP-qPCR. n = 3 biological replicates. (J) qPCR analysis showing the mRNA expression level of MYBL2 in METTL16 knockdown K562 cells. n = 3 biological replicates. (K) qPCR analysis of K562 cells treated with actinomycin D for 4 and 8 h showing accelerated MYBL2 mRNA degradation in METTL16 knockdown K562 cells. n = 3 biological replicates. (L, M) Native RIP of METTL16 with MYBL2 mRNA in K562 (L) and HEK293 cells (M). n = 3 biological replicates. ACTIN and MALAT1 were served as negative and positive control, respectively. (N, O) Native RIP of IGF2BP1 with MYBL2 mRNA in K562 (N) and HEK293 cells (O). n = 3 biological replicates. GAPDH and SRF were served as negative and positive control, respectively. (P, Q) Validation of the knockdown (KD) efficiency of the siRNAs against IGF2BP1 via qPCR (P) and western blot (Q). n = 3 biological replicates. (R) qPCR analysis showing the mRNA expression level of MYBL2 in IGF2BP1 knockdown HEK293 cells. n = 3 biological replicates. (S) qPCR analysis of HEK293 cells treated with actinomycin D for 4 and 8 h showing accelerated MYBL2 mRNA degradation in IGF2BP1 knockdown HEK293 cells. n = 3 biological replicates. Data information: In (B–E, H–P, R, S), data were represented as mean ± SEM, *adjusted P < 0.05, ** adjusted P < 0.01, *** adjusted P < 0.001, **** adjusted P < 0.0001, n.s. non-significant, Student’s t-test (B, D, H–J, L, M, N, O, R), two-way ANOVA analysis with post hoc test of Tukey’s multiple comparison correction (C, K, S), one-way ANOVA analysis with post hoc test of Tukey’s multiple comparison correction (E, P). Source data are available online for this figure.