Fig. 6.

Model-genetic interaction between BRD4 and PR-DUB.3. a Whole-cell lysates were used for western blot with BAP1 and H2AK119Ub antibodies in human SCLC cell line NCI-H1963 transduced with either non-targeting CRISPR gRNA or BAP1 specific gRNA. HSP90 was used as an internal control, n = 3. b The scatter plot shows the overlap between ASXL3 and BAP1 targeted genes. c Pathway analysis by Metascape of genes that are positively correlated with both ASXL3 and BAP1 in NCI-H1963 cells. d The RNA-seq data for the expression of GRP and DDC genes in NCI-H1963 cells transduced with non-targeting shRNA or ASXL3-specific shRNAs, n = 2, two-tailed unpaired Student’s t test. **P < 0.01; *P < 0.05. e The RNA-seq data for the expression of GRP and DDC genes in NCI-H1963 SCLC cells transduced with non-targeting CRISPR gRNA or gRNA targeting BAP1 gene, n = 2. The Venn-diagram (f) and average plot (g) shows BAP1 peaks and occupancy in NCI-H1963 cells transduced with either shNONT or shASXL3. h Representative track example that shows BAP1 occupancy at enhancers between NCI-H1963 cells transduced with either shNONT or shASXL3. i A model of targeting the positive feedback within BRD4/ASXL3/BAP1 axis for novel SCLC therapy