Journal: Cell Death & Disease

Article Title: The RBPJ/DAPK3/UBE3A signaling axis induces PBRM1 degradation to modulate the sensitivity of renal cell carcinoma to CDK4/6 inhibitors

doi: 10.1038/s41419-022-04760-6

Figure Lengend Snippet: a–c Using the PBRM1 and UBE3A antibodies to performed the IP assay. Western blotting analysis the whole-cell lysates (WCL) of 293T ( a ), 786-O ( b ), and ACHN ( c ) cells. d Western blotting analysis of UBE3A proteins in 786-O whole-cell lysates pulled down by GST-EV or GST-PBRM1 recombinant proteins. Asterisks indicated the corresponding protein band of GST-EV and GST-PBRM1. e A schematic diagram depicting a set of GST-UBE3A recombinant protein constructs. f Western blotting analysis of PBRM1 proteins in 786-O whole-cell lysates pulled down by GST-EV or GST-UBE3A recombinant proteins. Asterisks indicated the corresponding protein band of GST-EV and GST-UBE3A recombinant proteins. g, h 786-O and ACHN cells were infected with indicates shRNAs for 72 h. Cells were harvested for western blotting analysis ( g ) and RT-qPCR assay ( h ). Statistical significance was determined by one-way ANOVA followed by Tukey’s multiple comparisons test. Data presented as mean ± SEM with three replicates ( n = 3). ns not significant. i – k IHC analysis of the tissue microarray with a cohort of patients with renal cell carcinoma by using the UBE3A and PBRM1 antibodies. The typical images of IHC were shown in ( i ). Heatmap showing the IHC score of PBRM1 and UBE3A in ( j ). Correlation analysis of the IHC score of PBRM1 and UBE3A proteins in ( k ).

Article Snippet: The antibodies used as follows: UBE3A (10344-1-AP, Proteintech; 1:1000 dilution), PBRM1 (12563-1-AP, Proteintech; 1:500 dilution), DAPK3 (2928, Cell signaling technology, 1:1000 dilution), RBPJ (14613-1-AP, Proteintech; 1:1000 dilution); P21 (10355-1-AP, Proteintech; 1:1000 dilution); GAPDH (10494-1-AP, Proteintech; 1:10000 dilution).

Techniques: Western Blot, Recombinant, Construct, Infection, Quantitative RT-PCR, Microarray

Journal: Cell Death & Disease

Article Title: The RBPJ/DAPK3/UBE3A signaling axis induces PBRM1 degradation to modulate the sensitivity of renal cell carcinoma to CDK4/6 inhibitors

doi: 10.1038/s41419-022-04760-6

Figure Lengend Snippet: a A schematic diagram depicted that UBE3A contained a consensus DAPK phosphorylation motif which was adjacent to the PKA phosphorylation site. b Western blotting analysis the whole-cell lysates (WCL) of 293T cells. c Western blotting analysis the WCL 786-O and ACHN cells. d Western blotting analysis of UBE3A proteins in 786-O whole-cell lysates pulled down by GST-EV or GST-DAPK3 recombinant proteins. e Western blotting analysis of DAPK3 proteins in 786-O whole-cell lysates pulled down by GST-EV or GST-UBE3A recombinant proteins. f , g 786-O cells were transfected with indicated plasmids. Twenty-four hours post transfection, cells were harvested for Western blotting analysis ( f ) and RT-qPCR analysis ( g ). Statistical significance was determined by one-way ANOVA followed by Tukey’s multiple comparisons test. Data presented as Mean ± SEM with three replicates ( n = 3). ns not significant. h , i 786-O cells were transfected with indicated shRNAs. Seventy-two hours post infection, cells were harvested for western blotting analysis ( h ) and RT-qPCR analysis ( i ). Statistical significance was determined by one-way ANOVA followed by Tukey’s multiple comparisons test. Data presented as Mean ± SEM with three replicates ( n = 3). ns not significant. j 786-O cells were infected with indicated shRNAs. After 72 h, cells were treated with CHX, and cells were collected for western blot analysis at different timepoints. k 786-O cells were transfected with indicated plasmids. After 24 h, cells were treated with CHX, and cells were collected for western blot analysis at different timepoints. The GAPDH was recognized as the loading control. The protein level of PBRM1 was first normalized to loading control. The normalized values were further normalized to the values in 0 h group. Immunoblots (IB) are representative of results from two independent experiments ( n = 2). Statistical significance was determined by multiple student’s t -test at the time point of 5, 10, 15 h. Data presented as Mean ± SEM with two replicates. Ns not significant; *** P < 0.001. Data presented as Mean ± SEM with two replicates. Ns not significant; *** P < 0.001. l 786-O cells were infected with the indicated shRNAs. After 72 h, cells were collected for western blotting after treatment with MG132 for 8 h.

Article Snippet: The antibodies used as follows: UBE3A (10344-1-AP, Proteintech; 1:1000 dilution), PBRM1 (12563-1-AP, Proteintech; 1:500 dilution), DAPK3 (2928, Cell signaling technology, 1:1000 dilution), RBPJ (14613-1-AP, Proteintech; 1:1000 dilution); P21 (10355-1-AP, Proteintech; 1:1000 dilution); GAPDH (10494-1-AP, Proteintech; 1:10000 dilution).

Techniques: Phospho-proteomics, Western Blot, Recombinant, Transfection, Quantitative RT-PCR, Infection, Control

Journal: Cell Death & Disease

Article Title: The RBPJ/DAPK3/UBE3A signaling axis induces PBRM1 degradation to modulate the sensitivity of renal cell carcinoma to CDK4/6 inhibitors

doi: 10.1038/s41419-022-04760-6

Figure Lengend Snippet: DAPK3 competed with PKA to bind with UBE3A and enhance the PBRM1 degradation in renal cancer cells. PBPJ transcriptionally regulated DAPK3 expression and then promoted UBE3A-mediated degradation of PBRM1. Then, PBRM1 increased the p21 expression and sensitized renal cancer cells to CDK4/6 inhibitors. In combination with RBPJ inhibitors, CDK4/6 inhibitors synergistically enhanced renal cancer cells.

Article Snippet: The antibodies used as follows: UBE3A (10344-1-AP, Proteintech; 1:1000 dilution), PBRM1 (12563-1-AP, Proteintech; 1:500 dilution), DAPK3 (2928, Cell signaling technology, 1:1000 dilution), RBPJ (14613-1-AP, Proteintech; 1:1000 dilution); P21 (10355-1-AP, Proteintech; 1:1000 dilution); GAPDH (10494-1-AP, Proteintech; 1:10000 dilution).

Techniques: Expressing

Journal: Molecular Cancer Research

Article Title: Loss of PBRM1 Alters Promoter Histone Modifications and Activates ALDH1A1 to Drive Renal Cell Carcinoma

doi: 10.1158/1541-7786.mcr-21-1039

Figure Lengend Snippet: Figure 2. Loss of PBRM1 results in significantly gained H3K4me3 peaks across the epigenome. A, Heatmap of H3K4me3 signal (RPKM) at 259 lost (left) and 1,420 gained (right) H3K4me3 peaks with PBRM1 knockdown in 786-O cells. Two independent PBRM1 shRNA lines are shown compared with the non-targeting shRNA control (“C”) line. B, H3K4me3 ChIP-seq track showing an example of a gained H3K4me3 peak in both PBRM1 shRNA lines (red) compared with the control line (blue). Two H3K4me3 peaks at other locations are included to show scale and specificity of the change. C, TRAP motif analysis of top transcription factor motifs enriched at open chromatin regions (as determined by ATAC-seq) at the 1,420 gained H3K4me3 loci. D, Metagene plot of ATAC-seq signal (RPKM) atthe 1,420 gained (top) and 259 lost (bottom) H3K4me3 peaks in 786-O PBRM1 shRNA cells (red) and control cells (blue).

Article Snippet: We would also like to acknowledge the Transgenic Mouse Core at Columbia University for helping make the Pbrm1 fl/flmouse, and the Genomics Core Facility at the Department of Genetics and Genomic Sciences for performing the RNA-seq library preparation and sequencing.We thank the Steven Johnson laboratory for providing the ALDH1A1-HA vector through Addgene.

Techniques: Knockdown, shRNA, Control, ChIP-sequencing

Journal: Molecular Cancer Research

Article Title: Loss of PBRM1 Alters Promoter Histone Modifications and Activates ALDH1A1 to Drive Renal Cell Carcinoma

doi: 10.1158/1541-7786.mcr-21-1039

Figure Lengend Snippet: Figure 3. Gained H3K4me3 peaks following PBRM1 loss are associated with a retinoicacid signature. A, GSEA validation of derived gene signatures associated with 259 lost and 1,420 gained H3K4me3 peaks. B and C, Enrichr pathway analysis of the gained H3K4me3 peak gene signature, using the (B) KEGG and (C) Reactome pathway databases. D, H3K4me3 ChIP-seq tracks at the ALDH1A1 locus in PBRM1 shRNA cells (red) and non-targeting control cells (blue). E, RNA-seq expression of ALDH1A1 in PBRM1 shRNA cells (red) and non-targeting control cells (blue) at the indicated FBS concentration.

Article Snippet: We would also like to acknowledge the Transgenic Mouse Core at Columbia University for helping make the Pbrm1 fl/flmouse, and the Genomics Core Facility at the Department of Genetics and Genomic Sciences for performing the RNA-seq library preparation and sequencing.We thank the Steven Johnson laboratory for providing the ALDH1A1-HA vector through Addgene.

Techniques: Biomarker Discovery, Derivative Assay, ChIP-sequencing, shRNA, Control, RNA Sequencing, Expressing, Concentration Assay

Journal: Molecular Cancer Research

Article Title: Loss of PBRM1 Alters Promoter Histone Modifications and Activates ALDH1A1 to Drive Renal Cell Carcinoma

doi: 10.1158/1541-7786.mcr-21-1039

Figure Lengend Snippet: Figure 4. PBRM1 deficiency results in higher ALDH1A1 expression that increases tumorigenic potential. Western blot analysis of ALDH1A1 protein levels in (A) 786-O cells and (B) A-704 cells. For (C) 786-O and (D) A-704 cells, the ALDEFLUOR assay was used to measure the percent of cells that were ALDE- FLUOR-positive. n ¼ 3/line, error bars represent SEM, from independent experiments, and statistical testing comparing all column means was performed using ordinary one-way ANOVA with Tukey’s multiple comparisons test. E, Colony formation in soft-agar of 786-O control and PBRM1 shRNA #1 cells. Increasing doses of DEAB were mixed in with soft-agar at the time of plating. n ¼ 3/line at each dose, error bars represent SEM, from independent experiments, and statis- tical testing comparing the 0 mmol/L treated condition versus higher doses within a cell line was performed using an ordinary two-way ANOVA with the Tukey’s multiple comparisons test, with no significant differences among the control cells. F and G, Tumorsphere assays in 786-O control and PBRM1 shRNA #2 lines. F, vehicle control or DEAB (15 mmol/L) was added to tumorsphere media at the time of plating. G, cells were transfected with non-targeting siRNA (C) or 1 of 3 ALDH1A1-targting siRNAs (#2, #5, or #7) 24 hours before plating for tumorsphere assay. H, 786-O control cells were transfected with empty-vector control plasmid (pcDNA-EV) or a plasmid expressing human influenza hemagglutinin (HA)-tagged ALDH1A1 (pcDNA- ALDH1A1-HA). Bottom, Western blot analysis comparing ALDH1A1 levels in 786-O control cells transfected with EV- control (left lane) or ALDH1A1-HA (middle lane), or non- transfected PBRM1 shRNA #2 cells (right lane). Top, tumor- sphere formation in 786-O control cells transfected with the indicated plasmids. For F–H, n ¼ 3/condition, from indepen- dent experiments; for F–G, statistical testing was performed using an ordinary two-way ANOVA with the Sidak method to correct for multiple comparisons between conditions within cell lines; for H, an unpaired t test was performed.

Article Snippet: We would also like to acknowledge the Transgenic Mouse Core at Columbia University for helping make the Pbrm1 fl/flmouse, and the Genomics Core Facility at the Department of Genetics and Genomic Sciences for performing the RNA-seq library preparation and sequencing.We thank the Steven Johnson laboratory for providing the ALDH1A1-HA vector through Addgene.

Techniques: Expressing, Western Blot, Control, shRNA, Transfection, Plasmid Preparation

Journal: Molecular Cancer Research

Article Title: Loss of PBRM1 Alters Promoter Histone Modifications and Activates ALDH1A1 to Drive Renal Cell Carcinoma

doi: 10.1158/1541-7786.mcr-21-1039

Figure Lengend Snippet: Figure 5. With PBRM1 deficiency, ARID2 is more highly expressed and remains bound to other SWI/SNF subunits. Western blot analysis of the indicated PBAF complex subunits in (A) 786-O cells and (B) A-704 cells. Immunoprecipitation (IP) experiments in 786-O cells for (C) BRG1 and (D) ARID2. Matched isotype IgG was used for control IPs. Inputs are aliquots taken from pre-cleared nuclear extracts before the IPs were performed. E, Western blot analysis of heavy fractions (#1–10, out of 24 total) from glycerol gradient fractionation of nuclear extracts from 786-O cells. The PRC2 protein EZH2 is shown for comparison.

Article Snippet: We would also like to acknowledge the Transgenic Mouse Core at Columbia University for helping make the Pbrm1 fl/flmouse, and the Genomics Core Facility at the Department of Genetics and Genomic Sciences for performing the RNA-seq library preparation and sequencing.We thank the Steven Johnson laboratory for providing the ALDH1A1-HA vector through Addgene.

Techniques: Western Blot, Immunoprecipitation, Control, Fractionation, Comparison

Journal: Molecular Cancer Research

Article Title: Loss of PBRM1 Alters Promoter Histone Modifications and Activates ALDH1A1 to Drive Renal Cell Carcinoma

doi: 10.1158/1541-7786.mcr-21-1039

Figure Lengend Snippet: Figure 6. ARID2 is required for increased ALDH1A1 expression and tumorsphere formation, whereas BRG1 and SNF5 are dispensable. A, C, and D, Western blot analysis exploring effects of knocking down (A) ARID2, (C) SNF5, and (D) BRG1 in 786-O control and PBRM1 shRNA #1 cells. A non-targeting siRNA was used as a control (labeled “C”). B and E, Tumorsphere assays in 786-O con- trol and PBRM1 shRNA #2 cells. Cells were transfected with non-targeting siRNA (labeled “C”) or targeting siRNAs 24 hours before plating for the tumorsphere assay. n ¼ 3–6/condition, from independent experiments; statistical testing was per- formed using an ordinary two-way ANOVA with Tukey’s multiple comparisons test between conditions within cell lines.

Article Snippet: We would also like to acknowledge the Transgenic Mouse Core at Columbia University for helping make the Pbrm1 fl/flmouse, and the Genomics Core Facility at the Department of Genetics and Genomic Sciences for performing the RNA-seq library preparation and sequencing.We thank the Steven Johnson laboratory for providing the ALDH1A1-HA vector through Addgene.

Techniques: Expressing, Western Blot, Control, shRNA, Labeling, Transfection

Journal: bioRxiv

Article Title: PBRM1-Dependent PBAF Targeting is Required for EMT and Metastasis in Breast Cancer

doi: 10.1101/2025.10.19.683137

Figure Lengend Snippet: (A) Schematic of NMuMG cells treatment with TGFβ1 and micrographs of associated morphological changes (4X magnification). (B and C) Representative transwell invasion assay images (B) and bar plot (C) of absorbance quantification of NMuMG sgCt and sg Pbrm1 cells with and without TGFβ1 treatment. n=3 technical replicates. Data are represented as mean ± SD. (D) Immunoblots of Pbrm1, E-Cadherin, and Vimentin in lysates of NMuMG vector control and sh Pbrm1 cells with TGFβ1 treatment at the indicated concentrations and incubation times. (E) Relative cell counts of NMuMG control and sh Pbrm1 cells with and without TGFβ1 treatment, normalized to untreated cells. Representative graph, n=3 biological replicates. Data are represented as mean ± SD. (F) Relative cell counts of NMuMG sgCt and sg Pbrm1 cells with and without TGFβ1 treatment, normalized to untreated cells. Representative graph, n=3 biological replicates. Data are represented as mean ± SD. (G) Representative flow cytometry density dot plots of Edu and PI staining in fixed NMuMG sgCt and sg Pbrm1 cells with and without 72h TGFβ1 treatment. The gating strategy for cells in different cell cycle stages (G0/G1, S, or G2/M) is indicated with boxes. (H) Bar plot of the percentage of cells in different cell cycle stages in NMuMG sgCt and sg Pbrm1 cells with TGFβ1 treatment for the indicated incubation times. Gating was performed as in (G). n=3 biological replicates. Data are represented as mean ± SD. (I) Representative flow cytometry density dot plots of AnnexinV and PI staining in unfixed NMuMG sgCt and sg Pbrm1 cells with and without 5d TGFβ1 treatment. The gating strategy for the percentage of live (PI-)/dead (PI+) and apoptotic (Annexin V+)/non-apoptotic (Annexin V-) cells is indicated with quadrants. (J) Bar plot of percentage of PI-live cells (left) and Annexin V+ apoptotic cells (right) in NMuMG sgCt and sg Pbrm1 cells with TGFβ1 treatment for the indicated incubation times. Gating was performed as in (I). n=3 biological replicates. Data are represented as mean ± SD. (K) Immunoblots of Pbrm1 and Cleaved-PARP (C-PARP) from lysates of NMuMG sgCt and sg Pbrm1 cells with and without TGFβ1 treatment for the indicated time periods. Statistical comparison was done using multiple unpaired t-tests with Holm-Sidak correction. *: p < 0.05, **: p < 0.01, ***: p < 0.001, ****: p < 0.0001

Article Snippet: Constructs encoding codon-optimized ORFs for bacterial expression of human PBRM1 BD2 (addgene #39013), BD3 (addgene #39030), BD4 (addgene #39103), BD5 (addgene #38999) and tandem BD2-5 (SGC construct ID #PB1A-c080) were transformed in BL21(DE3) for BDs, and BL21 Rosetta2 (DE3) pLysS for tandem BD2-5.

Techniques: Transwell Invasion Assay, Western Blot, Plasmid Preparation, Control, Incubation, Flow Cytometry, Staining, Comparison

Journal: bioRxiv

Article Title: PBRM1-Dependent PBAF Targeting is Required for EMT and Metastasis in Breast Cancer

doi: 10.1101/2025.10.19.683137

Figure Lengend Snippet: (A) Heatmap representation of genes increased (upper) or decreased (lower) in NMuMG sgCt cells with TGFβ1 treatment at different time points. (B) Metagene plots and heatmaps of regions of differential increased (upper) or decreased (lower) accessibility in NMuMG sgCt cells with TGFβ1 treatment. The set of regions includes sites with differential accessibility at any time point of TGFβ1 treatment compared to untreated cells. (C and D) Comparison of changes in RNA expression with changes in accessibility in NMuMG sgCt cells with 48h TGFβ1 treatment relative to no treatment. (C) Venn diagram of overlaps of DEGs from RNA-seq and differentially accessible regions from ATAC-seq annotated to the nearest gene. Total number of genes in each condition is indicated in parentheses in the Venn diagram. (D) Scatter plot of correlation between DEGs from RNA-seq and differentially accessible regions from ATAC-seq annotated to the nearest gene. Each data point in the scatter plot represents the change in expression value of a single gene, x axis : differentially accessible regions associated with the nearest gene in ATAC-seq as FC, y axis : DEG in RNA-seq as log2FC. The degree of correlation was calculated using all DEGs. (E) Top overrepresented GO terms from pathway analysis using Enrichr on genes with increased expression and accessibility in NMuMG sgCt cells with 48h TGFβ1 treatment. Gene sets were defined using overlap analysis in (C). (F) Motif analysis using Homer on genes with increased expression and accessibility in NMuMG sgCt cells with 48h TGFβ1 treatment. Gene sets were defined using overlap analysis in (C). (G) Top overrepresented GO terms from pathway analysis using Enrichr on genes with decreased expression and accessibility in NMuMG sgCt cells with 48h TGFβ1 treatment. Gene sets were defined using overlap analysis in (C). (H) Motif analysis using Homer on genes with decreased expression and accessibility in NMuMG sgCt cells with 48h TGFβ1 treatment. Gene sets were defined using overlap analysis in (C). (I) Principal Component Analysis of the RNA-seq data of NMuMG sgCt and sg Pbrm1 cells with and without TGFβ1 treatment for the designated times. (J) Heatmap representation of all genes decreased in NMuMG sg Pbrm1 relative to sgCt cells at any timepoint of TGFβ1 treatment compared to untreated cells. Genes induced by TGFβ1 in sgCt but not in sg Pbrm1 cells are subsetted with a box. (K) Top overrepresented GO terms from pathway analysis using Enrichr on the highlighted subset of genes from (J). (L) Principal Component Analysis of the ATAC-seq data for NMuMG sgCt and sg Pbrm1 cells with and without TGFβ1 treatment where samples are color coded by treatment. (M and N) Metagene plots of all sites of increased (M) or decreased (N) accessibility in NMuMG sg Pbrm1 cells relative to sgCt cells in any treatment condition. Red lines depict the average enrichment in sg Pbrm1 cells and black lines depict the average enrichment in sgCt cells. Peak summits are aligned at the center. (O) Venn diagram of DEGs from RNA-seq and the nearest gene of differentially accessible regions from ATAC-seq in sg Pbrm1 relative to sgCt cells treated with TGFβ1 for 48h. Total number of genes in each condition is indicated in parentheses in the Venn diagram. (P) Scatter plot of differential accessibility from ATAC-seq with the change in expression of the nearest gene in sg Pbrm1 relative to sgCt cells treated with TGFβ1 for 48h. Each data point in the scatter plot represents the FC expression value of a single gene, x axis : DEG in RNA-seq as log2FC, y axis : differentially accessible regions associated with the nearest gene in ATAC-seq as FC. Correlation was calculated using all differentially accessible sites.

Article Snippet: Constructs encoding codon-optimized ORFs for bacterial expression of human PBRM1 BD2 (addgene #39013), BD3 (addgene #39030), BD4 (addgene #39103), BD5 (addgene #38999) and tandem BD2-5 (SGC construct ID #PB1A-c080) were transformed in BL21(DE3) for BDs, and BL21 Rosetta2 (DE3) pLysS for tandem BD2-5.

Techniques: Comparison, RNA Expression, RNA Sequencing, Expressing

Journal: bioRxiv

Article Title: PBRM1-Dependent PBAF Targeting is Required for EMT and Metastasis in Breast Cancer

doi: 10.1101/2025.10.19.683137

Figure Lengend Snippet: (A) Schematic representation of the PBAF and cBAF complexes with shared subunits in grey, PBAF-specific subunits in red, and cBAF specific subunits in blue. The reader domains encoded by the different subunits and their respective functions are indicated. (B) Venn diagram of Phf10 and Smarca4 ChIP-seq peaks from NMuMG sgCt cells with Pbrm1 and Dpf2 ChIP-seq peaks from NMuMG cells (GSE249211). Total number of peaks identified in ChIP-seq for each protein is indicated in parentheses. (C) Metagene plots and heatmaps of ChIP-seq enrichment of Phf10, Pbrm1, Smarca4, and Dpf2 at Phf10 and Dpf2 ChIP-seq sites in NMuMG cells as described in (B). (D) Genomic feature distribution of the ChIP-seq peaks identified for Pbrm1, Phf10, Smarca4, and Dpf2 in NMuMG cells as described in (B). (E) Metagene plots and heatmaps of ChIP-seq enrichment of Phf10 H3K4me3, H3K4me1, and H3K27ac enrichment at Phf10 ChIP-Seq sites in sgCt cells as described in (B). (F) Heatmap representation of PBAF subunit abundance in NMuMG sg Pbrm1 cells with and without TGFβ1 treatment identified by Phf10 IP-MS. (G) (left)Venn diagram of Phf10 ChIP-seq peaks in NMuMG sgCt and sg Pbrm1 cells. Total number of Phf10 ChIP-seq peaks identified in each condition is indicated in parentheses. (right) Metagene plots and heatmaps of ChIP-seq enrichment of Phf10 in sgCt and sg Pbrm1 cells at Phf10 ChIP-seq sites in sgCt cells as described in (B). Peak summits are aligned at the center. (H) Venn diagram of Phf10 ChIP-seq peaks in untreated and TGFβ1-treated NMuMG sgCt cells. Total number of Phf10 ChIP-seq peaks identified in each condition is indicated in parentheses. (I) Metagene plots and heatmaps of Phf10 ChIP-seq enrichment in untreated and TGFβ1-treated NMuMG cells. Enrichment was plotted at Phf10 sites in untreated sgCt cells (top) and Phf10 sites unique to TGFβ-treated sgCt cells. Peak summits are aligned at the center. (J) (top) Venn diagram of Phf10 ChIP-seq peaks in NMuMG sgCt cells and sites of differential accessibility in sg Pbrm1 cells compared to sgCt cells. (bottom) The genomic feature distribution for sites of differential accessibility in sg Pbrm1 cells compared to sgCt cells (K) (top) Venn diagram of Phf10 ChIP-seq peaks in NMuMG sgCt cells with 48h TGFβ1 treatment and sites of differential accessibility in sg Pbrm1 cells compared to sgCt cells with 48h TGFβ1 treatment. (bottom) The genomic feature distribution for sites of differential accessibility in sg Pbrm1 cells compared to sgCt cells with 48h TGFβ1 treatment. (L) Metagene plots and heatmaps of Phf10 ChIP-seq and ATAC-seq enrichment at sites of differential accessibility in sg Pbrm1 cells compared to sgCt cells under untreated (left) and 48h TGFβ1-treated (right) conditions. (M) Genomic tracks of Phf10 ChIP-seq and ATAC-seq enrichment at the Tnfsf13b locus in sgCt and sg Pbrm1 cells in untreated (top) and 48h TGFβ1-treated (bottom) conditions.

Article Snippet: Constructs encoding codon-optimized ORFs for bacterial expression of human PBRM1 BD2 (addgene #39013), BD3 (addgene #39030), BD4 (addgene #39103), BD5 (addgene #38999) and tandem BD2-5 (SGC construct ID #PB1A-c080) were transformed in BL21(DE3) for BDs, and BL21 Rosetta2 (DE3) pLysS for tandem BD2-5.

Techniques: ChIP-sequencing, Protein-Protein interactions

Journal: bioRxiv

Article Title: PBRM1-Dependent PBAF Targeting is Required for EMT and Metastasis in Breast Cancer

doi: 10.1101/2025.10.19.683137

Figure Lengend Snippet: (A) Volcano plot of TF consensus motifs enriched in regions of differential accessibility in NMuMG sg Pbrm1 relative to sgCt cells with 48h TGFβ1 treatment (FDR<0.001). (B) Metagene plots and heatmaps of ChIP-seq enrichment of Fosl2 (left), Fosb (center) and Atf3 (right) in sgCt and sg Pbrm1 cells in both untreated and 48h TGFβ1-treated conditions. Enrichment was plotted for corresponding TF binding sites compiled from all conditions. (C) Venn diagram of Phf10 ChIP-Seq sites with Fosl2 (left), Fosb (center) and Atf3 (right) ChIP-seq sites in sgCt cells with 48h TGFβ1 treatment. Total number of peaks identified in ChIP-seq for each protein is indicated in parentheses. To the right of each Venn diagram are the metagene plots and heatmaps of ChIP-seq enrichment of Phf10 with Fosl2 (left), Fosb (center) or Atf3 (right) at the Phf10 (top) and associated TF (bottom) binding sites. (D) Genomic feature distribution of the Fosl2, Fosb, and Atf3 ChIP-seq peaks. (E) Genomic tracks of Phf10 and Atf3 ChIP-seq enrichment in sgCt and sg Pbrm1 cells at Tnfsf13b locus in untreated (top) and 48h TGFβ1-treated (bottom) conditions. (F) Scatter plot of the change in expression for DEGs from sg Pbrm1 vs sgCt with 48h TGFβ1 plotted against DEGs from sh Atf3 vs shScr with 48h TGFβ1. Each data point in the scatter plot represents the log2FC expression value of a single gene in the indicated comparison, x axis : sh Atf3 vs shCt, y axis : sg Pbrm1 vs sgCt. The degree of correlation was calculated using all DEGs. (G) Venn diagram of overlap between DEGs in sg Pbrm1 vs sgCt and DEGs from sh Atf3 vs shScr, both with 48h TGFβ1 treatment. Total number of genes from each condition is indicated in parentheses. The top overrepresented GO terms for the set of genes increased by both sg Pbrm1 and sh Atf3 (top) or decreased by both sg Pbrm1 and sh Atf3 (bottom) were identified using Enrichr pathway analysis. (H) Absolute cell counts of control and sh Atf3 cells with and without 5 ng/mL TGFβ1 treatment. 0.9 million cells were seeded for each cell line and cell counts were taken on day 3 and 6. Representative graph, n=3 biological replicates. Data are represented as mean ± SD. (I) Heatmap representation of the differential accessibility of consensus TF motifs in NMuMG sg Pbrm1 cells with and without TGFβ1 treatment performed using diffTF analysis. Significant differences in weighted means between the groups are shown. (J) Intersection of Phf10 ChIP-seq peaks in NMuMG sgCt cells with Irf1 ChIP-seq peaks in 48h TGFβ1-treated NMuMG cells from a published Irf1 ChIP-Seq dataset (GSE141501). Total number of peaks identified in ChIP-seq for each protein is indicated in parentheses. (K) Metagene plots and heatmaps of ChIP-seq enrichment of Phf10 from 48h TGFβ1-treated NMuMG sgCt cells and Irf1 from 48h TGFβ1-treated NMuMG cells from a published dataset at Phf10 and Irf1 shared sites. (L) Genomic tracks of ChIP-seq enrichment of Phf10, Atf3, H3K14ac, and Irf1 in untreated and 48h-TGFβ1 treated NMuMG sgCt at Il15 locus. (M) Intersection of Phf10 ChIP-seq peaks in NMuMG sgCt cells with Snai1 ChIP-seq peaks in pBI3G mouse mesenchymal breast cancer cells from a published dataset (GSE61198). Total number of peaks identified in ChIP-seq for each protein is indicated in parentheses. (N) Metagene plots and heatmaps of ChIP-seq enrichment of Phf10 from 48h TGFβ1-treated NMuMG sgCt cells and Snai1 from pBI3G mouse mesenchymal breast cancer cells at Phf10 and Snai1 shared sites. (O) Genomic tracks of ChIP-seq enrichment at Snai2 locus of Phf10, Atf3, and H3K14ac in untreated and 48h-TGFβ1 treated NMuMG sgCt cells and Snai1 ChIP-Seq enrichment in pBI3G mouse mesenchymal breast cancer cells. *: p < 0.05, **: p < 0.01, ***: p < 0.001, ****: p < 0.0001

Article Snippet: Constructs encoding codon-optimized ORFs for bacterial expression of human PBRM1 BD2 (addgene #39013), BD3 (addgene #39030), BD4 (addgene #39103), BD5 (addgene #38999) and tandem BD2-5 (SGC construct ID #PB1A-c080) were transformed in BL21(DE3) for BDs, and BL21 Rosetta2 (DE3) pLysS for tandem BD2-5.

Techniques: ChIP-sequencing, Binding Assay, Expressing, Comparison, Control

Journal: bioRxiv

Article Title: PBRM1-Dependent PBAF Targeting is Required for EMT and Metastasis in Breast Cancer

doi: 10.1101/2025.10.19.683137

Figure Lengend Snippet: (A) Coomassie gel of the purified recombinant proteins BD2, BD3, BD4, BD5 and the tandem BD2-5 used for peptide and nucleosome binding assays. (B) Schematic representation of EpiCypher’s Captify™ assay. (C) Table of EC 50 values (nM) of the different BDs for the indicated peptides obtained using the ALPHA/dCypher assay. HP1 binding to H3K9me3 peptide is used as a positive control in this assay. (D) Binding curves of tandem BD2-5 with the indicated peptides obtained using the ALPHA/dCypher assay. EC 50 (nM) values are indicated next to the corresponding curves. (E) Table of the relative EC 50 values (nM) of the different BDs for nucleosomes bearing the indicated histone modifications obtained using the ALPHA/dCypher assay. HP1 binding to H3K9me3 nucleosomes is used as a positive control in this assay. (F) Binding curves of tandem BD2-5 for nucleosomes bearing the indicated peptides obtained using the ALPHA/dCypher assay. HP1 binding to H3K9me3 peptide is used as a positive control in this assay. EC 50 values (nM) obtained for positive binders are indicated in the legend. (G and H) Metagene plots and heatmaps of ChIP-seq enrichment of Phf10, H3K14ac, H3K18ac, and H3K27ac at Phf10 binding sites in untreated (H) and 48h TGFβ1-treated (I) sgCt cells. (I) Correlation matrix with r-values between Phf10 and H3K14ac, H3K18ac, H3K27ac, and H3K4me3 ChIP-seq enrichment in untreated and 48h TGFβ1-treated sgCt cells. (J) Metagene plots and heatmaps of ChIP-seq enrichment of Phf10, H3K14ac, H3K18ac, and H3K27ac in untreated and 48h TGFβ1-treated sgCt cells. The top heatmap is at Phf10 binding sites in untreated cells and the bottom is Phf10 binding sites only found in TGFβ1-treated cells. (K and L) Genomic tracks of ChIP-seq enrichment of Phf10, H3K14ac, H3K18ac, and H3K27ac in untreated (M) and 48h-TGFβ1 treated (M) sgCt cells at constitutive locus Cpne2 and an inducible locus Tnfsf13b .

Article Snippet: Constructs encoding codon-optimized ORFs for bacterial expression of human PBRM1 BD2 (addgene #39013), BD3 (addgene #39030), BD4 (addgene #39103), BD5 (addgene #38999) and tandem BD2-5 (SGC construct ID #PB1A-c080) were transformed in BL21(DE3) for BDs, and BL21 Rosetta2 (DE3) pLysS for tandem BD2-5.

Techniques: Purification, Recombinant, Binding Assay, Positive Control, ChIP-sequencing

Journal: bioRxiv

Article Title: PBRM1-Dependent PBAF Targeting is Required for EMT and Metastasis in Breast Cancer

doi: 10.1101/2025.10.19.683137

Figure Lengend Snippet: (A) Box plot of the mRNA expression of PBRM1 in normal vs tumor tissues of the TCGA human breast cancer dataset. Generated using GEPIA. (B) Kaplan-Meier curves displaying the effect of PBRM1 mRNA expression on overall survival in all breast cancer (left), non-metastatic breast cancer (center) and metastatic TNBC (right) patient samples. Curves generated using Km plotter. (C) Box plots of the mRNA expression of ARID1A in normal vs tumor tissues of the TCGA human breast cancer dataset. Generated using GEPIA. (D) Kaplan-Meier curves displaying the effect of ARID1A mRNA expression on overall survival in all breast cancer (left), non-metastatic breast cancer (center) and metastatic TNBC (right) patient samples. Curves generated using Km plotter. (E) Schematic representation of the syngeneic 4T1 fat pad model of breast cancer metastasis. 4T1 cells expressing luciferase are injected orthotopically into the inguinal fat pad, the primary tumor is surgically removed, and lung metastases are monitored using in vivo imaging. (F) (left) Immunoblots of whole cell extracts of 4T1 cells treated with shRNA or sgRNA against Pbrm1. (right) In vitro proliferation of 4T1 sh Pbrm1 and control cells. n=2 biological replicates. Two-way ANOVA was used for statistical comparison. Data are represented as mean ± SD. (G) Primary tumor growth in 4T1 sh Pbrm1 and control cells measured using vernier calipers at the indicated time points. Two-way ANOVA with multiple comparisons was used for statistical comparison. Data are represented as mean ± SEM. (H-I) Bioluminescent imaging of lung metastasis in 4T1 sh Pbrm1 and control cells (H) and sg Pbrm1 and sgCt cells (I) after removal of the primary tumor. Two-way ANOVA with multiple comparisons was used for statistical comparison. Data are represented as mean ± SEM. (J) H&E and Ki-67 stained IHC images of the lungs harvested at the end of the experiment described in (I). (K) (left) Immunoblots of whole cell extracts of NME cells treated with sgRNA against Pbrm1. (right) In vitro proliferation of NME sg Pbrm1 and control cells. n=2 biological replicates. Two-way ANOVA was used for statistical comparison. Data are represented as mean ± SD. (L) Scatter plot of lung weights from individual mice in the NME sgCt and sg Pbrm1 groups harvested at the end of experiment. Welch’s t-test was used for statistical comparison. Data are represented as mean ± SD. Images of representative lung halves bearing metastatic lung nodules from the two groups are shown. (M) Immunoblots of the other lung halves from (L), which were cultured in vitro under antibiotic selection. Whole cell extracts were used for immunoblotting. (N) (left) Immunoblots of whole cell extracts of Renca cells treated with shRNA or sgRNA against Pbrm1. (right) In vitro proliferation of Renca sg Pbrm1 and control cells. n=3 biological replicates. Two-way ANOVA was used for statistical comparison. Data are represented as mean ± SD. (O and P) Scatter plot of the number of lung nodules from individual mice in the Renca sh Pbrm1 and control groups (O), and sg Pbrm1 and sgCt groups (P) harvested at the end of experiment. Welch’s t-test was used for statistical comparison. Data are represented as mean ± SD. (Q) Immunoblots of the lung halves from (P), which were cultured in vitro under antibiotic selection. Whole cell extracts were used for immunoblotting. (R) Schematic representation of the syngeneic 4T1 fat pad model from of breast cancer metastasis from (E) using doxycycline-inducible Pbrm1 KD after primary tumor removal. (S) Immunoblots of whole cell extracts of 4T1 cells after doxycycline-induced expression of shRNA against Pbrm1. 2ug/ml doxycycline concentration was used for induction of hairpin expression. (T) Primary tumor growth in 4T1 ish Pbrm1 and control cells measured using vernier calipers at the indicated time points. Two-way ANOVA with multiple comparisons was used for statistical comparison. Data are represented as mean ± SEM. (U) Bioluminescent imaging of lung metastasis in 4T1 i shPbrm1 cells with and without doxycycline administration after removal of the primary tumor. Two-way ANOVA with multiple comparisons was used for statistical comparison. Data are represented as mean ± SEM. (V) Principal Component Analysis of the RNA-seq profile of 4T1 shCt and sh Pbrm1 cells with and without TGFβ1 treatment where samples are color coded by treatment. (W) Heatmap representation of genes increased decreased in 4T1 sh Pbrm1 cells with and without TGFβ1 treatment. (X) Top overrepresented GO terms from pathway analysis using Enrichr on the highlighted subset of genes from (C). (Y) Transwell invasion assay images and bar plot of absorbance quantification of 4T1 control and sh Pbrm1 cells with and without TGFβ1 treatment. n=3 technical replicates. Data are represented as mean ± SD. *: p < 0.05, **: p < 0.01, ***: p < 0.001, ****: p < 0.0001

Article Snippet: Constructs encoding codon-optimized ORFs for bacterial expression of human PBRM1 BD2 (addgene #39013), BD3 (addgene #39030), BD4 (addgene #39103), BD5 (addgene #38999) and tandem BD2-5 (SGC construct ID #PB1A-c080) were transformed in BL21(DE3) for BDs, and BL21 Rosetta2 (DE3) pLysS for tandem BD2-5.

Techniques: Expressing, Generated, Luciferase, Injection, In Vivo Imaging, Western Blot, shRNA, In Vitro, Control, Comparison, Imaging, Staining, Cell Culture, Selection, Concentration Assay, RNA Sequencing, Transwell Invasion Assay

Journal: Nature Communications

Article Title: Targeting dependency on a paralog pair of CBP/p300 against de-repression of KREMEN2 in SMARCB1-deficient cancers

doi: 10.1038/s41467-024-49063-w

Figure Lengend Snippet: a Localization of signals generated by SMARCB1, H3K27ac, and H3K4me3 CUT&RUN-seq, and RNA-seq around the KREMEN2 locus in JMU-RTK-2 + SMARCB1 and JMU-RTK-2 -SMARCB1 cells. b – g Enrichment of CUT&RUN signals for SMARCB1 ( b ), SMARCA4 ( c ), ARID1A ( d ), PBRM1 ( e ), SS18 ( f ), and GLTSCR1 ( g ) (relative to that of normal IgG) at the indicated regions distant from the transcription start site (TSS) of the KREMEN2 locus in JMU-RTK-2 +SMARCB1 and JMU-RTK-2 -SMARCB1 cells. Data are presented as the mean ± SD (standard deviation), n = 3 independent experiments. h Heatmap of KREMEN2 mRNA expression in SMARCB1-deficient cells (JMU-RTK-2, HS-ES-2R, and G402) transfected for 48 h with the indicated siRNAs. i Heatmap of KREMEN2 mRNA expression in SMARCB1- (JMU-RTK-2, HS-ES-2R, and G402) cells transfected for 48 h with the indicated siRNAs. j Expression of KREMEN2 mRNA (relative to that in siNT-transfected cells) in SMARCB1-deficient cell lines (JMU-RTK-2, HS-ES-2R, and G402) transfected for 48 h with the indicated siRNAs. Data are presented as the mean ± SD, n = 3 independent experiments. k Enrichment of CUT&RUN signals for the SMARCA1 (relative to that of normal IgG signal) at the indicated regions distant from the TSS of the KREMEN2 locus in JMU-RTK-2 +SMARCB1 and JMU-RTK-2 -SMARCB1 cells. Data are presented as the mean ± SD, n = 3 independent experiments. l , m Enrichment of CUT&RUN signals for the H3K27ac ( l ) and SMARCA1 ( m ) signals (relative to that of normal IgG signal) at the indicated regions distant from the TSS of the KREMEN2 locus in SMARCB1-deficient JMU-RTK-2 cells treated without or with 2 μM CP-C27 for 24 h. Data are presented as the mean ± SD, n = 3 independent experiments.

Article Snippet: Following trypsinization, 1 × 10 5 cells were harvested and subjected to CUT&RUN using 2 μL of an antibody specific for H3K4me3 (CST, 9751), H3K4me1 (CST, 5326), H3K27ac (CST, 8173), H3K27me3 (CST, 9733), CBP (Abcam, ab253202), p300 (CST, 54062), EZH2 (CST, 5246), SMARCB1 (CST, 91735), SMARCA4 (Abcam, ab110641), ARID1A (Abcam, ab182560), PBRM1 (CST, 89123), SS18 (CST, 21792), GLTSCR1 (CST, 45441), or SMARCA1 (CST, 12483), or normal IgG (CST, 66362), as described in the section “CUT&RUN-seq analysis”.

Techniques: Generated, RNA Sequencing, Standard Deviation, Expressing, Transfection

Journal: bioRxiv

Article Title: PBRM1-Dependent PBAF Targeting is Required for EMT and Metastasis in Breast Cancer

doi: 10.1101/2025.10.19.683137

Figure Lengend Snippet: (A) Coomassie gel of the purified recombinant proteins BD2, BD3, BD4, BD5 and the tandem BD2-5 used for peptide and nucleosome binding assays. (B) Schematic representation of EpiCypher’s Captify™ assay. (C) Table of EC 50 values (nM) of the different BDs for the indicated peptides obtained using the ALPHA/dCypher assay. HP1 binding to H3K9me3 peptide is used as a positive control in this assay. (D) Binding curves of tandem BD2-5 with the indicated peptides obtained using the ALPHA/dCypher assay. EC 50 (nM) values are indicated next to the corresponding curves. (E) Table of the relative EC 50 values (nM) of the different BDs for nucleosomes bearing the indicated histone modifications obtained using the ALPHA/dCypher assay. HP1 binding to H3K9me3 nucleosomes is used as a positive control in this assay. (F) Binding curves of tandem BD2-5 for nucleosomes bearing the indicated peptides obtained using the ALPHA/dCypher assay. HP1 binding to H3K9me3 peptide is used as a positive control in this assay. EC 50 values (nM) obtained for positive binders are indicated in the legend. (G and H) Metagene plots and heatmaps of ChIP-seq enrichment of Phf10, H3K14ac, H3K18ac, and H3K27ac at Phf10 binding sites in untreated (H) and 48h TGFβ1-treated (I) sgCt cells. (I) Correlation matrix with r-values between Phf10 and H3K14ac, H3K18ac, H3K27ac, and H3K4me3 ChIP-seq enrichment in untreated and 48h TGFβ1-treated sgCt cells. (J) Metagene plots and heatmaps of ChIP-seq enrichment of Phf10, H3K14ac, H3K18ac, and H3K27ac in untreated and 48h TGFβ1-treated sgCt cells. The top heatmap is at Phf10 binding sites in untreated cells and the bottom is Phf10 binding sites only found in TGFβ1-treated cells. (K and L) Genomic tracks of ChIP-seq enrichment of Phf10, H3K14ac, H3K18ac, and H3K27ac in untreated (M) and 48h-TGFβ1 treated (M) sgCt cells at constitutive locus Cpne2 and an inducible locus Tnfsf13b .

Article Snippet: Constructs encoding codon-optimized ORFs for bacterial expression of human PBRM1 BD2 (addgene #39013), BD3 (addgene #39030), BD4 (addgene #39103), BD5 (addgene #38999) and tandem BD2-5 (SGC construct ID #PB1A-c080) were transformed in BL21(DE3) for BDs, and BL21 Rosetta2 (DE3) pLysS for tandem BD2-5.

Techniques: Purification, Recombinant, Binding Assay, Positive Control, ChIP-sequencing