Fig 2: High HDAC7 expression was correlated with poor NSCLC prognosis, and HDAC7 overexpression promoted NSCLC proliferation. a Kaplan–Meier survival analysis of HDAC7 expression from the Kaplan–Meier plotter database. b Western blots of HDAC7 expression in tumor (T) and paired adjacent nontumor (N) tissues from 12 NSCLC patients. c Representative HDAC7 IHC images of NSCLC and adjacent non-tumor tissues. Scale bar, 200 μm and 20 μm (inset), respectively. IHC analysis (d) and Kaplan–Meier survival analysis (e) of HDAC7 expression in 319 NSCLC patients. f The growth curve of indicated NSCLC cells. g Representative images and analysis of EdU incorporation assay. EdU-positive cells (red) and Hoechst 33342–stained cells (blue). Scale bar, 100 μm (inset). h Representative images and analysis of colony formation assay. i Representative western blotting of HDAC7, CDK2 and Cyclin A2 in indicated NSCLC cells. j Representative western blotting of whole cell lysates (WCL) at indicated times. A549 cells were synchronized in G0/G1 by serum starvation followed by serum re-addition. k Gross photograph of subcutaneous xenograft tumors, tumor growth curve, changes of nude mice body weight, and subcutaneous tumor weight in each group. l Representative western blotting of HDAC7, CDK2, and Cyclin A2 in subcutaneous tumor tissues. All the data are expressed a mean ± SD. *P < 0.05. LV, lentivirus

Fig 3: SAHA inhibits WNT5A-mediated ESCC cell migration and invasion.A Representative western blots showing the levels of FLAG-WNT5A, HDAC7, SNAIL, E-cadherin, and vimentin in WNT5A-overexpressing EC109 and KYSE450 cells after treatment with SAHA for 48 h. β-actin was used as an internal control. B, C Representative images and statistical analysis of wound healing assays and Transwell migration/invasion assays using WNT5A-overexpressing ESCC cells after treatment with SAHA (1.25 μM) for 48 h. The migratory ability in wound healing assays was expressed as the mean scratch area. The initial scratch area (0 h) was set to 100%. *P < 0.05 compared with the LV-Control group, #P < 0.05 compared with the LV-Control + SAHA group, and &P < 0.05 compared with the LV-WNT5A group. LV lentivirus.

Fig 4: HADC7 is the deacetylase for ALKBH5 acetylation at K235.a HADC7 overexpression reduced ALKBH5 acetylation at K235. HeLa cells were transfected with the indicated plasmids together with the ALKBH5-HA vector; ALKBH5-HA was IPed using anti-HA antibody, and K235 acetylation in the IPed ALKBH5-HA was determined using the anti-pan acetylated lysine antibody. b, c ALKBH5 interacted with HDAC7. The ALKBH5-HA and HDAC7-FLAG plasmids were cotransfected into HeLa cells, and the HDAC7-FLAG (b) and ALKBH5-HA (c) complexes were co-IPed using anti-FLAG and anti-HA antibodies, respectively. ALKBH5-HA and HDAC7-FLAG were detected. d Silencing of HDAC7 increased the endogenous K235 acetylation of ALKBH5. HeLa cells were transfected with anti-HDAC7 siRNAs, and K235 acetylation was determined. e HDAC7 overexpression decreased the endogenous K235 acetylation of ALKBH5. HeLa cells were transfected with the indicated dose of HDAC7 plasmid, and K235 acetylation was determined. f HDAC7 directly deacetylated ALKBH5 at K235 in a dose-dependent manner in the in vitro deacetylation reaction. Immunopurified wild-type ALKBH5 containing K235-acetylated ALKBH5 stimulated by KAT8 overexpression was incubated with the indicated dose of immunopurified HADC7, and ALKBH5 acetylation at K235 was determined using anti-Ac K235 and anti-pan acetylated lysine antibodies. Source data are provided as a Source Data file.

Fig 5: ALKBH5 acetylation at K235 is critical for the RNA m6A demethylation activity of ALKBH5.a, b K235 acetylation of ALKBH5 decreased the cellular mRNA m6A levels. The wild-type ALKBH5 and its mutant K235R and K235Q plasmids were transfected into ALKBH5 KO HeLa cells, and the cellular mRNA m6A level was determined by dot blotting (a) and quantified by LC‒MS/MS analysis (b) (n = 3, two-tailed unpaired Student’s t test, mean ± SD). c, d Wild-type ALKBH5, but not the K235R mutant, directly demethylated m6A in the m6A-RNA oligos in the in vitro demethylation reaction. Immunopurified wild-type ALKBH5 and its mutant K235R (c) or recombinant wild-type ALKBH5 and its mutant K235R (d) were incubated with m6A RNA oligos; the m6A level was determined by dot blotting or LC‒MS/MS assays. e Cumulative distribution curve for the m6A peak abundance in NC, WT and K235R cells. f Distribution of m6A peaks in the 5′ UTR, CDS, stop codon and 3′ UTR in NC, WT and K235R cells. g Top consensus motif identified by HOMER with m6A peaks in NC, WT and K235R cells. h The indicated ALKBH5 vectors together with the KAT8 plasmid were cotransfected into ALKBH5 KO HeLa cells, and the cellular RNA m6A level was determined. i, j Recombinant wild-type ALKBH5 or its K235R mutant was incubated with m6A RNA oligos after recombinant wild-type ALKBH5 or its K235R mutant was treated with immunopurified KAT8 (i) or immunopurified HDAC7 (j), and the m6A level was determined. Source data are provided as a Source Data file.