BAP18 coactivates androgen receptor action and promotes prostate cancer progression

Abstract

BPTF associated protein of 18 kDa (BAP18) has been reported as a component of MLL1-WDR5 complex. However, BAP18 is an uncharacterized protein. The detailed biological functions of BAP18 and underlying mechanisms have not been defined. Androgen receptor (AR), a member of transcription factor, plays an essential role in prostate cancer (PCa) and castration-resistant prostate cancer (CRPC) progression. Here, we demonstrate that BAP18 is identified as a coactivator of AR in Drosophilar experimental system and mammalian cells. BAP18 facilitates the recruitment of MLL1 subcomplex and AR to androgen-response element (ARE) of AR target genes, subsequently increasing histone H3K4 trimethylation and H4K16 acetylation. Knockdown of BAP18 attenuates cell growth and proliferation of PCa cells. Moreover, BAP18 depletion results in inhibition of xenograft tumor growth in mice even under androgen-depletion conditions. In addition, our data show that BAP18 expression in clinical PCa samples is higher than that in benign prostatic hyperplasia (BPH). Our data suggest that BAP18 as an epigenetic modifier regulates AR-induced transactivation and the function of BAP18 might be targeted in human PCa to promote tumor growth and progression to castration-resistance.

Figure 1.

BAP18 enhances AR-induced transactivation in Drosophila. (A) Expression ARAF-1 in third instar eye imaginal discs driven by GMR-GAL4 driver and carrying an ARE-GFP reporter in pericentric heterochromatin were crossed with lines harboring gain of function (UAS-C17orf49, UAS-FLAG-CG33695) or CG33695 loss of function (BL10680, BL21629) mutants as indicated. The effect of BAP18 over-expression and mutations on ARAF-1 induced transactivation was evaluated by levels of GFP expression (upper panels). GMR-GAL4 expressed ARAF-1 was detected with anti-AR (N-20) antibody (middle panels) and merged images are shown in the lower panel. Quantification of GFP expression revealed by color intensity is shown at the bottom. *P < 0.05, ***P < 0.001. Scale bars, 50 μm. (B–E) Polytene chromosomes from the third instar larvae of flies carrying UAS-FLAG-CG33695 and GMR-GAL4 expression plasmids were dissected. The location of CG33695 on chromosomes was examined with anti-FLAG antibody (B and B′). Polytene were stained with anti-dCBP (C and C′) and with DAPI to visualize DNA (D and D′). E and E′, merged images. (F) Multiple-sequence alignments of the BAP18 from different species. The amino acid sequences of BAP18 in various species were extracted from the NCBI website. Multiple sequence alignments were performed using Vector NTI software (Life Technologies). Residues identical in all four species are highlighted in yellow; residues conserved are highlighted in blue; residues that are conserved but not identical are highlighted in green. Residues identical in the four species are highlighted in yellow; residues identical in Homo sapiens and Mus musculus are highlighted in blue; residues that are conserved but not identical are highlighted in green. Drosophila melanogaster CG33695 and Human sapiens BAP18 share 58% similarity. Human sapiens and Macaca mulatta BAP18 share 99%, while Human sapiens and Mus musculus BAP18 share 95% similarity. The red frame indicates the SANT domain. EMBL/RefSeq accession numbers are as follows: Homo sapiens (NP_777553.1), Macaca mulatta (AFE65909.1), Mus musculus (NP_665701.1) and Drosophila melanogaster (NP_001027248.1).

Figure 2.

BAP18 interacts with AR in vivo. (A) Schematic diagram of full-length BAP18 and BAP18 truncation mutants used in this study. Numbers indicate amino acid position. (B) Co-immunoprecipitation (Co-IP) assay showing the interaction between BAP18 and AR in vivo. HEK293 cells were transfected with FLAG-BAP18 and AR expression plasmids with or without DHT (10⁻⁸ M). After 48 h, cell lysates were immunoprecipitated with anti-FLAG antibody or with IgG as a control. Precipitated proteins were examined by Western blotting using antibodies against FLAG or AR. Input represents 5% of the total cell extract used for each immunoprecipitation. (C) BAP18 interacts with AR and N-terminal AR variants in 22Rv1 cells. Cell lysates of 22Rv1 cells under the same transfection condition were subjected to Co-IP probed with FLAG antibody or IgG. (D) HEK293 cells were co-transfected with GFP-BAP18 or a series of GFP-BAP18 mutation expression plasmids and AR as indicated with or without DHT (10⁻⁸ M) for 48 h. IP performed using anti-GFP and whole-cell extracts (input) were analyzed by immunoblotting using the indicated antibodies. (E) HEK293 cells were co-transfected with FLAG-BAP18 expression plasmids and AR and treated with or without DHT (10⁻⁸ M). Then cells were stained with TOPRO3 to visualize the nucleus (blue), anti-FLAG (red), anti-AR (green). The localization of the two proteins was analyzed by a fluorescence microscope. Scale bars, 50 μm. (F) 22Rv1 cells were transfected with plasmids expressing for GFP-BAP18 or two truncated mutants and treated with or without DHT (10⁻⁸ M). The endogenous AR were immunostained with anti-AR (red), whereas BAP18 and two truncates localizations were detected by direct GFP fluorescence. Blue is the DAPI stain of the nucleus. Scale bars, 10 μm.

Figure 3.

BAP18 enhances AR-induced transactivation and knockdown BAP18 inhibits AR target gene expression. (A–B) The transcriptional activity of AR was increased by overexpression of BAP18 both in 22Rv1 and LNCaP cells. Cells were transfected with full length BAP18 or its truncated mutants as indicated in the absence or presence of ligand (DHT). BAP18 enhances AR mediated transactivation, and both of N- terminus and C-terminus are required for its function. The expression levels of BAP18 truncated mutations were detected by Western blotting. (C) BAP18 enhances AR, ARAF-1 or AR-V7 mediated transactivation. The plasmids expressing the full-length or truncated AR were transfected into cos-7 cells with or without FLAG-BAP18, and the cells were treated with or without DHT (10⁻⁸M). After 24 h of DHT, cell lysates were subjected to luciferase assay. Relative luciferase units shown are the mean value at least three times. (D) BAP18 enhances AR or AR-V7-induced tranasactivation on MMTV-tk-luc reporter. Cos-7 cells were co-transfected with MMTV-tk-luc and pRL-TK together with the indicated expression plasmids with or without DHT (10⁻⁸M). (E) BAP18 enhances AR or AR-V7-induced tranasactivation on PSA-tk-luc reporter. The similar luciferase assays were performed as above with PSA-tk-luc in cos-7 cells. (F) 22Rv1 cells were harvested 48 h after transfection with three different siRNAs (#1, #2 and #3) targeting BAP18 or control siRNA (siCtrl). The expression of BAP18 was measured by Western blotting with anti-BAP18 antibodies. (G) 22Rv1 and LNCaP cells were infected with shRNA Lentivirus against BAP18 (shBAP18) or shRNA Lentivirus control (shCtrl). Western blotting analysis data are shown. (H) RNA-seq analysis profile of shBAP18 or shCtrl after treatment with or without DHT in 22Rv1 cells. Heat map displays the altered expression of DHT-induced genes upon BAP18 knockdown. Fold change is indicated at right. (I) Heat map representation of BAP18 knockdown on the expression of various genes in the absence or presence of DHT. Bold letters, widely studied AR target genes. (J) Venn diagram showing DHT-regulated genes in shCtrl and shBAP18 cells in the presence of DHT or vehicle. (K) Biological processes of DHT-induced transcripts in shCtrl and shBAP18 cells as revealed by KEGG pathway analysis.

Figure 4.

BAP18 facilitates the recruitment of MLL1 subcomplex to the promoters of AR target genes. (A–B) Real-time PCR (RT-PCR) analysis showing the effect of BAP18 knockdown on activation of 17 AR target genes. 22Rv1 or LNCaP cells with knocking down BAP18 expression by shBAP18 as indicated were harvest after treatment with or without DHT (10⁻⁸M) for 24 h. Total RNA was analyzed by Real-time quantitative PCR (RT-qPCR). Levels of all mRNAs were normalized to that of GAPDH mRNA. Statistical significance of differences between experimental groups was assessed t-test. Error bars represent mean ± SD. *P < 0.05; **P < 0.01; and ***P < 0.001. (C) 22Rv1 cells were transfected with FLAG-BAP18 for 40 h and treated with DHT (10⁻⁸M) for an additional 8 h and subjected to ChIP assay using antibodies specific to AR, BAP18, MOF, H4K16ac, H3K4me3 and H3K27me3. The immunoprecipitated DNA fragments were PCR amplified using primers flanking the promoter region of PSA gene (PSA-ARE I/II). (D) ChIP assay were performed using indicated antibodies in 22Rv1 cells with knocking down BAP18 expression by shBAP18 with or without DHT (10⁻⁸M) and analyzed by qPCR. The data are representative of more than two independent experiments.

Figure 5.

BAP18 interacts with MLL1 subcomplex. (A) BAP18, AR, MLL1 and MOF are recruited together to cis-regulatory elements of PSA in the absence or presence of DHT. ChIP/re-ChIP experiments were performed using specific antibodies against BAP18, AR, MLL1, MOF or IgG as indicated. DNA eluted from unprecipitated chromatin was used as input. Chromatin samples were analyzed by gel electrophoresis. (B) 22Rv1 cells were treated with shBAP18 or shCtrl, and immunoprecipitated using anti-AR antibodies or IgG. Precipitated protein complex were analyzed by Western blotting with antibodies as indicated. Input represents 5% of the total cell extract used for each immunoprecipitation. (C and D) LNCaP cells were incubated with or without DHT, transfected with BAP18 expression plasmids or siRNA against BAP18 (siBAP18). Immunoprecipitate generated with anti-FLAG or anti-AR antibody was subjected to Western blotting with the indicated antibodies. (E) AR-dependent transactivation requires BAP18, MLL1 or Ash2L in cells. Cos-7 cells were co-transfected with the AR together with BAP18, MLL1, Ash2L or MOF expression plasmid as indicated. The total amount of the transfected DNA was kept constant with the empty vector.

Figure 6.

Knockdown of BAP18 inhibits tumor growth of prostate cancer xenograft. (A) Growth curve of the stable knockdown by shBAP18 in 22Rv1 cells. Total cell numbers were counted on days 0, 2, 4, 6 and 8. Data were means ± SD of three independent experiments performed in triplicates. *P < 0.05. (B) Representative microphotographs of clone formation experiments of stable 22Rv1 cells. A total of 1 × 10⁴ cells were seeded in 3.5-cm dishes for 7 days. Colonies were stained by crystal violet. (C) 22Rv1 cells were transfected with siAR or siCtrl for 4 h, then transfected with BAP18 overexpression plasmids or vector in the presence of DHT or not. At 72 h after transfection, microscopic analysis of the cells were performed. Scale bars, 50 μm. (D) Photograph showing the xenograft tumor in male SCID mice with shCtrl (left) or shBAP18 (right) stable 22Rv1 cells. Tumors were dissected from mice at the 28th day post injection. (E–F) Dissected tumors were measured. The average volumes or weights and standard errors were calculated for each group. *P < 0.05. (G) The mRNA samples from tumor xenograft mice were isolated, and qPCR for the indicated mRNA was performed. Data were expressed as means ± SD (n = 8). Statistical significance was determined using t-test, *P < 0.05, **P < 0.01. (H–I) Representative photograph showing the xenograft tumor in male SCID mice treated castration injected with shCtrl (left) or shBAP18 (right) stable 22Rv1 cells. (J) The average tumor volume of shCtrl and shBAP18 were plotted over days after tumor cell injected. Data points represent the mean of 14 tumors from 7 mice. Bars represent standard deviation of the mean, P < 0.001. (K) Four weeks later tumor weights were statistically analyzed using t-test, P < 0.001.

Figure 7.

BAP18 mRNA and protein expression level in clinical prostatectomy specimens. (A) Expression of BAP18 in different prostate cancer cell lines. (B and C) All fresh tissue specimens were from patient undergoing radical prostatectomy without any hormonal therapy. Expression of BAP18 protein (B) and mRNA (C17orf49) (C) were detected in PCa or BPH. (D) Representative tissue stained with an antibody against BAP18. Stronger staining of BAP18 was seen in PCa (Gleason Grade [GG] 3, 4 and 5) than that in BPH (magnification 10x; Scale bars: 200 μm). Rectangle represents magnification image (magnification, 40x; scale bars: 50 μm). (E) BAP18 expression in PCa samples with different Gleason Grades (GG = 3, 4 and 5) compared with BPH (n = 33). BAP18 has obviously higher staining in PCa samples compared with that in BPH. In PCa, no significant correlation between the intensity of BAP18 immunostaining and different Gleason Grades could be observed. Statistical significance of the differences between experimental groups was assessed u-test, *P < 0.05, **P < 0.01. (F) Presentation for the percentage of PCa or BPH samples with low, middle or high expression of BAP18. (G) Schematic representation of the key role of BAP18 in up-regulation of AR-induced transactivation and promotion of prostate cancer progression.