The BAG-1 CTCF DNA-binding sites are critical to the regulation of BAG-1 gene expression. A, genetic analysis of the 5′BAG-1 upstream regulatory region. Eight separate 5′BAG-1 promoter deletion mutants were constructed and sequenced to determine the correct bp deletions. These 5′ deletion mutants were subsequently cotransfected with CMV-β-gal into HCT116 cells, harvested after 36 h, and processed for luciferase and β-gal assay. The CpG island is shown in gray. B, internal deletion analysis of the BAG-1 promoter CpG island. The three CTCF DNA-binding sites are shown as light spotted boxes. We constructed a series of internal deletion mutants in the BAG-1 promoter that systematically removed various parts of the CpG island, including the two 5′and one 3′CTCF DNA-binding sites. The CpG island deletion mutant plasmids and CMV-β-gal were cotransfected into HCT116 cells. Results are presented as luciferase/β-galactosidase relative units and as relative fold induction over the sham-treated control. All results are the mean of at least three separate experiments all done in duplicate. The results of individual transfections varied by <25 %. Error bars around data points represent 1 SD about the mean. Statistical significance was established by Student’s t test compared with transfected wild-type HCT116 cells. *, P < 0.05.

Chromatin immunoprecipitation assay to determine CTCF and BORIS binding to the BAG-1 promoter. A, ChIP analysis to determine CTCF binding to the BAG-1 promoter in HCT116-based DNMT somatic knockout cell lines. HCT116, DNMT1^−/−, DNMT3B^−/−, and DKO cells were exposed to formaldehyde to cross-link protein-DNA interactions. Chromatins were immunoprecipitated with anti-CTCF antibody. CTCF binding was quantified using real-time PCR with primers that covered the BAG-1 CpG island. Experimental results are the mean of at least three separate experiments. B, ChIP analysis to determine CTCF binding to the BAG-1 promoter in HCT116-based cells overexpressing DNMT1, DNMT3B, or both. HCT116, DNMT1⁺, DNMT3B⁺, and DKI cells as described above. C, EMSA analysis of protein complex binding to the second BAG-1 CTCF DNA (oligo)–binding site. Nuclear lysates from HCT116, DNMT1^−/−, DNMT3B^−/−, DNMT1⁺, DNMT3B⁺, and DKI cells were incubated with a ³²P-labeled CTCF-binding site oligo on ice for 30 min, followed by separation on a nondenaturing polyacrylmide gel. Sections of fluorograms from native gels, obtained using a TYPHOON Phosphorimager, as shown. Arrows, position of the CTCF shifted band and the free probe. D, ChIP analysis to determine BORIS binding to the BAG-1 promoter in HCT116-based cells overexpressing DNMT1, DNMT3B, or both. Chromatins were immunoprecipitated with anti-BORIS antibody, and BORIS binding was quantified by real-time PCR as described above. ChIP experiments are presented as fold enrichment relative units calculated as the mean ± SD fold enrichment of immunoprecipitated DNA relative to IgG control, which represents ChIP specificity, done in duplicate for each experiment. Error bars around data points represent 1 SD about the arithmetic mean and statistical significance was established by Student’s t test compared with transfected wild-type HCT116 cells. *, P < 0.05; **, P < 0.01.

CTCF decreases and BORIS increases BAG-1 promoter activity in transient assays in HCT116 cells. Cotransfection of either the pBAG-1-Luc (A) or 3×-CTCF-luciferase (B) reporter plasmids with a CTCF expression plasmid (pCMV-CTCF). Cotransfection of either the pBAG-1-Luc (C) or 3x-CTCF-luciferase (D) reporter plasmid transfected into HCT116 cells with a BORIS expression plasmid (pCMV-BORIS). Cells were also transfected with CMV-β-gal, and results are presented as luciferase/β-galactosidase relative units as the relative fold change compared with sham-treated HCT116 control; 1 μg (+) and 2 μg (++) of CMV-CTCF or CMV-BORIS expression plasmid. The total amount of pCMV-based expression vector DNA was held constant at 6.0 μg per transfection. The results are presented as relative fold induction over the sham-treated control. All results are the mean of at least three separate experiments, all done in duplicate. The results of individual transfections varied by <25%. Error bars around data points represent 1 SD about the arithmetic mean. Statistical significance was established by Student’s t test compared with transfected wild-type HCT116 cells. *, P < 0.05.

Confirmation of BAG gene family expression. A, validation of microarray data using real-time RT-PCR assays in HCT116, DNMT1^−/−, DNMT3B^−/−, and DKO cells. Total RNA was isolated and purity checked, and levels of BAG-1, BAG-3, and BAG-4 RNA were determined using Taqman probes. Results are presented as BAG relative RNA levels compared with control HCT116 cells. All experiments were normalized to GAPDH. All results are the mean of at least three separate experiments. Error bars represent 1 SD about the arithmetic mean. Statistical significance was established by Student’s t test compared with wild-type HCT116 cells. *, P < 0.05; **, P < 0.01. B, BAG-1 and BAG-3 protein levels in the HCT116 DNMT somatic knockout cells. Nuclear extracts were isolated from HCT116, DNMT1^−/−, DNMT3B^−/−, and DKO cells, and 20 μg of protein were separated by SDS-PAGE, transferred onto nitrocellulose, and immunoblotted using either an anti–BAG-1 or anti–BAG-3 antibody. Blots were reprobed with actin antibody to ensure equal protein loading. C, promoter region analyses of the BAG gene family. The 5′ upstream regulatory regions of BAG-1, BAG-3, and BAG-4 were aligned, and potential cis-acting DNA-binding/enhancer sites and potential CpG islands were identified using software available at Entrez Genome and Genetics Computer Group. D, analysis of the BAG-1 promoter. The 5′ promoter region of BAG-1 was cloned upstream of the luciferase gene (pBAG-1-Luc) and transfected into HCT116, DNMT1^−/−, DNMT3B^−/−, DKO, DNMT1⁺, Rat-1, or Rat-1-DNMT1⁺ cells. Cells were also transfected with CMV-β-gal, and results are presented as luciferase/β-galactosidase relative units and represent the relative fold change compared with sham-treated HCT116 control. All results are the mean of at least three separate experiments all done in duplicate. The results of individual transfections varied by <25%. Error bars around data points represent 1 SD about the arithmetic mean. Statistical significance was established by Student’s t test compared with transfected wild-type HCT116 cells. *, P < 0.05; **, P < 0.01.

Effect of cellular DNA methyltransferase status on BAG-1 promoter chromatin remodeling. Analysis of DNA methylation in HCT116 somatic cell knockouts and DNMT-overexpressing cell lines. A, the methylation status of the CpG island in the BAG-1 upstream promoter region in HCT116, DNMT1^−/−, DNMT3B^−/−, DKO, DNMT1^−/−, and DNMT3B^−/−, and DKI cell lines were analyzed by bisulfite genomic sequencing. BAG-1 gene expression is linked to histone methylation. HCT116, DNMT1⁺, DNMT3B⁺, DNMT1^−/−, DNMT3B^−/−, and DKO cells were harvested, followed by ChIP analysis using either a histone H3 dimethyl K4 or a dimethyl K9 antibody (Abcam), as described (20). Primers to the first and second CTCF-binding sites (B) and third CTCF-binding (C) were subsequently used in quantitative RT-PCR. Diagonal-lined column, alterations of H3-K4 methylation; dark-spotted columns, changes of H3-K9 methylation. All results for ChIP experiments are presented as fold enrichment relative units calculated as the mean ± SD fold enrichment of immunoprecipitated DNA relative to IgG control, which represents ChIP specificity, done in duplicate for each experiment. All results are the mean of at least three separate experiments. Error bars around data points represent 1 SD about the mean. Statistical significance was established by Student’s t test compared with transfected wild-type HCT116 cells. *, P < 0.05.

shRNA knockdown of BORIS decreases BAG-1 expression and alters chromatin status. A, shRNA knockdown of BORIS decreases intracellular protein levels. HCT116 cells were transfected with either vector only (Cont) or a BORIS hairpin loop shRNA, and cells were harvested, separated by SDS-PAGE, transferred onto nitrocellulose, and immunoblotted using an anti-BORIS antibody. B, shRNA knockdown of BORIS decreases BORIS binding to the BAG-1 CpG island. HCT116 cells were transfected with either vector only or BORIS shRNA and harvested for ChIP analysis with an anti-BORIS antibody, as described above. C, shRNA knockdown of BORIS alters BAG-1 promoter chromatin status. Control vector or BORIS shRNA–transfected HCT116 cells were harvested to determine histone H3-K4 (dark columns) or H3-K9 (check light gray columns) methylation status via ChIP analysis, using either a histone H3 dimethyl K4 or H3 dimethyl K9 antibody (Abcam), as described above. D, shRNA knockdown of BORIS decreases BAG-1 expression. HCT116 cells were transfected with either vector only or BORIS shRNA. Total RNA was isolated and purity checked, and levels of BAG-1 RNA were determined using real-time PCR with Taqman probes. All results are the mean of at least three separate experiments. Results are presented as BAG-1 relative RNA levels, wherein 1.0 represents BAG-1 RNA level in control HCT116 cells. All experiments were normalized to GAPDH. Error bars represent 1 SD about the arithmetic mean. For ChIP experiments (B and C), results are presented as fold enrichment relative units calculated as the mean ± SD fold enrichment of immunoprecipitated DNA relative to IgG control, which represents ChIP specificity, done in duplicate for each experiment. For all experiments, error bars around data points represent 1 SD about the arithmetic mean and statistical significance was established by Student’s t test compared with transfected wild-type HCT116 cells. *, P < 0.05.