Fig 2: HCF-1 promotes mitosis by regulating CDC42 expression.HCF-1 and CDC42 depletion induced chromosome misalignment and multinucleation in HeLa cells. a A schematic graph of synchronization and cell phenotype detection. Cell transfection and the first 15 h thymidine block were performed the same time, followed by 9 h of thymidine release and the second 15 h thymidine block. Cells were collected 12 h after thymidine release for chromosome misalignment detection and 16 h after thymidine release for multinucleate phenotype detection. b, d, f, h The quantification is acquired via counting the ratio of the number of mutated cells to the number of normal cells in a confocal resolvable plane in a random field of view. b, d Cells were transfected with shRNA-producing plasmids. GFP was expressed in successfully transfected cells. Cells were counter-stained with β-tubulin antibody. Scale bar: 10 µm. c Quantification of chromosome misalignment after HCF-1 and CDC42 knockdown. The ratio of chromosome misalignment was 6.2, 50 and 58.3% in transfected groups, respectively (n = 15). e Quantification of multinucleation after HCF-1 and CDC42 knockdown. The ratio of chromosome misalignment was 2.5, 17.2 and 14.3% in transfected groups, respectively (n = 10). f, h) Cells were transfected with shHCF-1 and CDC42F28L vectors. Both GFP and mCherry were expressed in successfully transfected cells. Cells were counter-stained with β-tubulin antibody. Scale bar: 10 µm. g Quantification of chromosome misalignment in the co-expression cells. The ratio of chromosome misalignment was 3.4, 59.5, 4.3 and 4.2% in transfected groups, respectively (n = 15). i Quantification of multinucleation in the co-expression cells. The ratio of chromosome misalignment was 2.5, 21.6, 4.5 and 5.4% in transfected groups, respectively (n = 10). Results are expressed as mean ± SD. (***p < 0.001. Student’s t-test). Replicates and supplemental data showed similar phenotypes and variation trends.

Fig 3: HCF-1 regulates expression of Pdx-1.(A) Western blot analysis of Pdx-1 protein levels in HCF-1 knockdown cells. (B) Relative Pdx1 protein levels as determined by densitometry analysis. Pdx1 levels were normalized to β-actin levels. Results are pooled from 5 experiments and represent mean +/− SEM. **denotes a p-value = 0.03, ***denotes a p-value = 0.05 (Student’s t-test). (C) RT-qPCR analysis of Pdx1, Ins1 and Ins2 transcript levels in HCF-1 knockdown cells. The mean normalized RNA level for each gene in sicontrol treated cells was set to 1. The data shown are pooled from three independent experiments and are represented as mean +/− SEM. *denotes a p-value <0.05 relative to cells treated with sicontrol siRNA (Student’s t-test).

Fig 4: ASXL3 directly interacts with BRD4 in small cell lung cancer cells. a The GFP-fusion proteins were purified from HEK293T cells transfected with GFP-tagged ASXL1/2/3. The purified proteins were subjected to mass spectrometry analysis. Peptide numbers of BAP1 complex subunits and BRD4 pulled down by GFP-ASXLs were shown. b BAP1, HCF1, FOXK1, and BRD4 levels in HEK293T cells 24 h after transfection with plasmids expressing GFP or GFP-ASXL1–3. Immunoprecipitation (IP) from whole cell lysates were performed with antibodies against the GFP epitope, followed by immunoblotting (IB) with antibodies against the indicated proteins. HSP90 was used as negative control, n = 3. c IP of endogenous ASXL3 of NCI-H1963 cells with two different homemade antibodies followed by IB for BAP1 and BRD4; IgG was used as negative control, n = 3. d IP of endogenous BRD4 from NCI-H1963 cells followed by IB for BRD4, ASXL3, and BAP1; IgG was used as negative control, n = 3. LC, light chain. e Nuclear extract from NCI-H1963 SCLC cells was subjected to size exclusion chromatography, and then protein levels of ASXL3, BAP1, and BRD4 were determined by western blot analysis. BRD2 and BRD3 were used as controls, n = 2. f Schematic diagram depicting the BBM depletion of human ASXL3 protein. g BAP1 and BRD4 levels in HEK293T cells 24 h after transfection with plasmids expressing GFP, GFP-ASXL3-FL, or GFP-ASXL3-ΔBBM. IP from whole cell lysates was performed with antibodies against the GFP epitope, followed by IB with antibodies against the BAP1 and BRD4, n = 3. h 293T cells were transfected with Flag-tagged BRD4 truncations (Additional file 2: Figure S3B) together with GFP-tagged ASXL3. Whole-cell lysates were used for IP with Flag antibody followed by IB for GFP. CDK9 was used as positive control, n = 3. i Recombinant His-tagged BRD4-ET domain and GST-tagged ASXL3-BBM was purified from Escherichia coli. The in vitro binding assay was performed to determine the direct interaction between BRD4-ET domain and ASXL3-BBM, n = 3. j Alignment by CLUSTALW analysis shows the similarity between ASXL3-BBM and BRD4 binding motif within other BRD4 binding proteins

Fig 5: Inhibition of BAP1 reduces ASXL3 protein stability in small cell lung cancer cells.A Three different human SCLC cell lines (NCI-H1963, NCI-H748, and NCI-H1882) were treated with iBAP-II (0, 10, and 20 μM) for 24 h. The protein levels of ASXL1, ASXL2, ASXL3, BAP1, HCFC1, FOXK1, and FOXK2 were determined by western blot. β-tubulin was used as the internal control. Human SCLC cell line NCI-H1963 cells were treated with 50 μg/ml cycloheximide (CHX) at different times in the presence of either DMSO or iBAP-II. The ASXL3 protein levels were determined by western blot (B) and further quantified by ImageJ (C), n = 2. D The log2 fold change of the expression levels of genes nearby ASXL3 peaks grouped as TSS and nonTSS regions annotated by Homer in ASXL3-depleted cells, and cells treated with either iBAP or iBAP-II, n = 2. E Gene Set Enrichment Analysis (GSEA) of iBAP-II gene signature enrichment in ASXL3 or BAP1-depleted conditions. A total of 383 downregulated genes in iBAP-II (20 μM) versus DMSO treatment (adj.p < 0.01, logFC < −1) is defined as the iBAP signature gene set. GSEA of iBAP signature genes in ASXL3-depleted cells compared with control cells by shRNA. GSEA of iBAP signature genes in BAP1-depleted cells compared with control cells by sgRNA. F Representative track example that shows BAP1 occupancy in NCI-H1963 cells treated with either DMSO or iBAP-II. G Pathway analysis shows the most significant signaling pathways enriched in genes nearest to the loss of BAP1 peaks (n = 1999) induced by iBAP-II treatment. H ChIP-Seq analysis representing 70% (2470/3529) of all BAP1 peaks from BAP1 chromatin occupancy being collectively reduced upon iBAP treatment. Occupancy and log2FC heatmaps of BAP1 in DMSO and iBAP-II conditions are centered on these BAP1 peaks (n = 2470). When comparing the same ranking of peaks, BAP1 occupancy also decreased from ASXL3 depletion via shRNA. Peaks were sorted based on the mean signals from DMSO and iBAP-II samples in each row. I The average plot shows BAP1 levels at ASXL3-occupied TSS (left) and nonTSS (right) regions. J The box plot shows the H2AK119Ub levels between DMSO and iBAP-II treated NCI-H1963 cells.