Title: <scp>TRIM33</scp> drives prostate tumor growth by stabilizing androgen receptor from Skp2‐mediated degradation
Abstract: Article4 July 2022free access Source DataTransparent process TRIM33 drives prostate tumor growth by stabilizing androgen receptor from Skp2-mediated degradation Mi Chen Mi Chen orcid.org/0000-0003-0851-5116 Cancer Centre, University of Macau, Taipa, Macau SAR Centre for Precision Medicine Research and Training, University of Macau, Taipa, Macau SAR MoE Frontiers Science Center for Precision Oncology, University of Macau, Taipa, Macau SAR Faculty of Health Sciences, University of Macau, Taipa, Macau SAR Contribution: Conceptualization, Validation, Investigation, Visualization, Methodology Search for more papers by this author Shreyas Lingadahalli Shreyas Lingadahalli orcid.org/0000-0002-9609-6486 Cancer Centre, University of Macau, Taipa, Macau SAR Centre for Precision Medicine Research and Training, University of Macau, Taipa, Macau SAR MoE Frontiers Science Center for Precision Oncology, University of Macau, Taipa, Macau SAR Faculty of Health Sciences, University of Macau, Taipa, Macau SAR Contribution: Validation, Investigation, Methodology Search for more papers by this author Nitin Narwade Nitin Narwade orcid.org/0000-0002-1368-307X Cancer Centre, University of Macau, Taipa, Macau SAR Centre for Precision Medicine Research and Training, University of Macau, Taipa, Macau SAR MoE Frontiers Science Center for Precision Oncology, University of Macau, Taipa, Macau SAR Faculty of Health Sciences, University of Macau, Taipa, Macau SAR Contribution: Investigation, Visualization Search for more papers by this author Kate Man Kei Lei Kate Man Kei Lei Cancer Centre, University of Macau, Taipa, Macau SAR Centre for Precision Medicine Research and Training, University of Macau, Taipa, Macau SAR MoE Frontiers Science Center for Precision Oncology, University of Macau, Taipa, Macau SAR Pilot Laboratory, University of Macau, Taipa, Macau SAR Institute of Translational Medicine, University of Macau, Taipa, Macau SAR Search for more papers by this author Shanshan Liu Shanshan Liu orcid.org/0000-0002-3696-9417 Xuzhou Medical University, Xuzhou, China Contribution: Investigation Search for more papers by this author Zuxianglan Zhao Zuxianglan Zhao Cancer Centre, University of Macau, Taipa, Macau SAR Centre for Precision Medicine Research and Training, University of Macau, Taipa, Macau SAR MoE Frontiers Science Center for Precision Oncology, University of Macau, Taipa, Macau SAR Faculty of Health Sciences, University of Macau, Taipa, Macau SAR Contribution: Visualization Search for more papers by this author Yimin Zheng Yimin Zheng orcid.org/0000-0002-0394-9735 Cancer Centre, University of Macau, Taipa, Macau SAR Centre for Precision Medicine Research and Training, University of Macau, Taipa, Macau SAR MoE Frontiers Science Center for Precision Oncology, University of Macau, Taipa, Macau SAR Faculty of Health Sciences, University of Macau, Taipa, Macau SAR Contribution: Visualization Search for more papers by this author Qian Lu Qian Lu Xuzhou Medical University, Xuzhou, China Contribution: Supervision Search for more papers by this author Alexander Hin Ning Tang Alexander Hin Ning Tang Department of Pathology, The University of Hong Kong, Hong Kong, Hong Kong SAR Contribution: Supervision, Validation Search for more papers by this author Terence Chuen Wai Poon Terence Chuen Wai Poon orcid.org/0000-0003-4723-6910 Cancer Centre, University of Macau, Taipa, Macau SAR Centre for Precision Medicine Research and Training, University of Macau, Taipa, Macau SAR MoE Frontiers Science Center for Precision Oncology, University of Macau, Taipa, Macau SAR Faculty of Health Sciences, University of Macau, Taipa, Macau SAR Pilot Laboratory, University of Macau, Taipa, Macau SAR Institute of Translational Medicine, University of Macau, Taipa, Macau SAR Contribution: Supervision Search for more papers by this author Edwin Cheung Corresponding Author Edwin Cheung [email protected] orcid.org/0000-0001-8034-2833 Cancer Centre, University of Macau, Taipa, Macau SAR Centre for Precision Medicine Research and Training, University of Macau, Taipa, Macau SAR MoE Frontiers Science Center for Precision Oncology, University of Macau, Taipa, Macau SAR Faculty of Health Sciences, University of Macau, Taipa, Macau SAR Contribution: Conceptualization, Resources, Supervision, Investigation, Methodology Search for more papers by this author Mi Chen Mi Chen orcid.org/0000-0003-0851-5116 Cancer Centre, University of Macau, Taipa, Macau SAR Centre for Precision Medicine Research and Training, University of Macau, Taipa, Macau SAR MoE Frontiers Science Center for Precision Oncology, University of Macau, Taipa, Macau SAR Faculty of Health Sciences, University of Macau, Taipa, Macau SAR Contribution: Conceptualization, Validation, Investigation, Visualization, Methodology Search for more papers by this author Shreyas Lingadahalli Shreyas Lingadahalli orcid.org/0000-0002-9609-6486 Cancer Centre, University of Macau, Taipa, Macau SAR Centre for Precision Medicine Research and Training, University of Macau, Taipa, Macau SAR MoE Frontiers Science Center for Precision Oncology, University of Macau, Taipa, Macau SAR Faculty of Health Sciences, University of Macau, Taipa, Macau SAR Contribution: Validation, Investigation, Methodology Search for more papers by this author Nitin Narwade Nitin Narwade orcid.org/0000-0002-1368-307X Cancer Centre, University of Macau, Taipa, Macau SAR Centre for Precision Medicine Research and Training, University of Macau, Taipa, Macau SAR MoE Frontiers Science Center for Precision Oncology, University of Macau, Taipa, Macau SAR Faculty of Health Sciences, University of Macau, Taipa, Macau SAR Contribution: Investigation, Visualization Search for more papers by this author Kate Man Kei Lei Kate Man Kei Lei Cancer Centre, University of Macau, Taipa, Macau SAR Centre for Precision Medicine Research and Training, University of Macau, Taipa, Macau SAR MoE Frontiers Science Center for Precision Oncology, University of Macau, Taipa, Macau SAR Pilot Laboratory, University of Macau, Taipa, Macau SAR Institute of Translational Medicine, University of Macau, Taipa, Macau SAR Search for more papers by this author Shanshan Liu Shanshan Liu orcid.org/0000-0002-3696-9417 Xuzhou Medical University, Xuzhou, China Contribution: Investigation Search for more papers by this author Zuxianglan Zhao Zuxianglan Zhao Cancer Centre, University of Macau, Taipa, Macau SAR Centre for Precision Medicine Research and Training, University of Macau, Taipa, Macau SAR MoE Frontiers Science Center for Precision Oncology, University of Macau, Taipa, Macau SAR Faculty of Health Sciences, University of Macau, Taipa, Macau SAR Contribution: Visualization Search for more papers by this author Yimin Zheng Yimin Zheng orcid.org/0000-0002-0394-9735 Cancer Centre, University of Macau, Taipa, Macau SAR Centre for Precision Medicine Research and Training, University of Macau, Taipa, Macau SAR MoE Frontiers Science Center for Precision Oncology, University of Macau, Taipa, Macau SAR Faculty of Health Sciences, University of Macau, Taipa, Macau SAR Contribution: Visualization Search for more papers by this author Qian Lu Qian Lu Xuzhou Medical University, Xuzhou, China Contribution: Supervision Search for more papers by this author Alexander Hin Ning Tang Alexander Hin Ning Tang Department of Pathology, The University of Hong Kong, Hong Kong, Hong Kong SAR Contribution: Supervision, Validation Search for more papers by this author Terence Chuen Wai Poon Terence Chuen Wai Poon orcid.org/0000-0003-4723-6910 Cancer Centre, University of Macau, Taipa, Macau SAR Centre for Precision Medicine Research and Training, University of Macau, Taipa, Macau SAR MoE Frontiers Science Center for Precision Oncology, University of Macau, Taipa, Macau SAR Faculty of Health Sciences, University of Macau, Taipa, Macau SAR Pilot Laboratory, University of Macau, Taipa, Macau SAR Institute of Translational Medicine, University of Macau, Taipa, Macau SAR Contribution: Supervision Search for more papers by this author Edwin Cheung Corresponding Author Edwin Cheung [email protected] orcid.org/0000-0001-8034-2833 Cancer Centre, University of Macau, Taipa, Macau SAR Centre for Precision Medicine Research and Training, University of Macau, Taipa, Macau SAR MoE Frontiers Science Center for Precision Oncology, University of Macau, Taipa, Macau SAR Faculty of Health Sciences, University of Macau, Taipa, Macau SAR Contribution: Conceptualization, Resources, Supervision, Investigation, Methodology Search for more papers by this author Author Information Mi Chen1,2,3,4,†, Shreyas Lingadahalli1,2,3,4,9,†, Nitin Narwade1,2,3,4,†, Kate Man Kei Lei1,2,3,5,6, Shanshan Liu7, Zuxianglan Zhao1,2,3,4, Yimin Zheng1,2,3,4, Qian Lu7, Alexander Hin Ning Tang8, Terence Chuen Wai Poon1,2,3,4,5,6 and Edwin Cheung *,1,2,3,4 1Cancer Centre, University of Macau, Taipa, Macau SAR 2Centre for Precision Medicine Research and Training, University of Macau, Taipa, Macau SAR 3MoE Frontiers Science Center for Precision Oncology, University of Macau, Taipa, Macau SAR 4Faculty of Health Sciences, University of Macau, Taipa, Macau SAR 5Pilot Laboratory, University of Macau, Taipa, Macau SAR 6Institute of Translational Medicine, University of Macau, Taipa, Macau SAR 7Xuzhou Medical University, Xuzhou, China 8Department of Pathology, The University of Hong Kong, Hong Kong, Hong Kong SAR 9Present address: Vancouver Prostate Centre, Vancouver, BC, Canada † These authors contributed equally to this work *Corresponding author. Tel: +853 88224992; E-mail: [email protected] EMBO Reports (2022)23:e53468https://doi.org/10.15252/embr.202153468 PDFDownload PDF of article text and main figures. Peer ReviewDownload a summary of the editorial decision process including editorial decision letters, reviewer comments and author responses to feedback. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Abstract Androgen receptor (AR) is a master transcription factor that drives prostate cancer (PCa) development and progression. Alterations in the expression or activity of AR coregulators significantly impact the outcome of the disease. Using a proteomics approach, we identified the tripartite motif-containing 33 (TRIM33) as a novel transcriptional coactivator of AR. We demonstrate that TRIM33 facilitates AR chromatin binding to directly regulate a transcription program that promotes PCa progression. TRIM33 further stabilizes AR by protecting it from Skp2-mediated ubiquitination and proteasomal degradation. We also show that TRIM33 is essential for PCa tumor growth by avoiding cell-cycle arrest and apoptosis, and TRIM33 knockdown sensitizes PCa cells to AR antagonists. In clinical analyses, we find TRIM33 upregulated in multiple PCa patient cohorts. Finally, we uncover an AR-TRIM33-coactivated gene signature highly expressed in PCa tumors and predict disease recurrence. Overall, our results reveal that TRIM33 is an oncogenic AR coactivator in PCa and a potential therapeutic target for PCa treatment. Synopsis TRIM33 is an oncogenic coactivator of AR-dependent transcription in prostate cancer. TRIM33 exerts its coactivator activity by stabilizing AR from Skp2-mediated ubiquitination and proteasomal degradation. TRIM33 coactivates AR-mediated transcription in prostate cancer (PCa) by stabilizing AR from Skp2-mediated ubiquitination and proteasomal degradation. TRIM33 is highly expressed in PCa tumors and is required for prostate tumor growth. TRIM33 and AR target genes define an AR-TRIM33-coactivated gene signature that is highly expressed in PCa tumors. Patients with a high expression of this gene signature show a higher risk of disease recurrence. Introduction Prostate cancer (PCa) is the most common noncutaneous cancer and is responsible for the second-highest cancer-related death among men (Siegel et al, 2020). Androgen receptor (AR) occupies a central position in the development and progression of PCa (Heemers & Tindall, 2009; Massie et al, 2011). Thus, androgen deprivation therapy (ADT) has been widely used as the primary treatment for advanced PCa (Miller et al, 2019). While patients demonstrate a favorable initial response to ADT, in most cases, the disease will inevitably recur and progress to a fatal stage of PCa termed castration-resistant prostate cancer (CRPC), which is also driven by AR signaling (Feldman & Feldman, 2001; Decker et al, 2012; Grasso et al, 2015). After activation by androgen, AR translocates to the nucleus, binds to specific AR binding sites (ARBSs), and recruits multiple coregulators in a highly regulated manner (Bennett et al, 2010). These coregulators then form a productive AR transcriptional complex that ultimately determines the transcriptional outcome of AR. Coregulators act at the promoter and enhancer regions of AR target genes and display vital roles in controlling the AR-mediated transcription process through specific and distinct ways (Heemers & Tindall, 2007; Sung & Cheung, 2014). Androgen receptor coregulators are divided into four main types based on their functional characteristics: (i) pioneer factors (e.g., FOXA1) bind to chromatin before androgen stimulation to modulate AR genome occupancy; (ii) collaborative factors (e.g., ERG) facilitate AR binding by recruiting other cofactors to modify the epigenetic state of the surrounding chromatin; (iii) coactivators (e.g., p160/SRC) enhance AR transcriptional activity by acting as bridging factors or chromatin modifiers; and (iv) corepressors (e.g., NCoRs) suppress AR transcriptional activity (Sung & Cheung, 2014). The importance of coregulators in defining the AR transcription output implies a critical role for these regulatory proteins in pathologies linked to aberrant AR action, such as PCa. Indeed, accumulating evidence shows that aberrant expression or improper regulation of these coregulators alter the AR transcriptional landscape, promoting PCa tumorigenesis and progression (Wang et al, 2009; Sharma et al, 2013; Mills, 2014; Pomerantz et al, 2015). Hence, identifying all the essential components of the AR coregulator complex is necessary for understanding the molecular mechanism of androgen-regulated transcription and discovering new promising pharmacological targets in this pathway. In this study, we utilized a proteomics approach called rapid immunoprecipitation of endogenous proteins (RIME) to systematically identify all coregulator proteins of the AR interactome in PCa cells. We functionally characterized one of these coregulators, TRIM33. We demonstrate that TRIM33 is a potent coactivator of AR transcriptional activity. Mechanistically, we show TRIM33 promotes PCa growth by preventing AR from Skp2-mediated protein degradation. Finally, we identified an AR-TRIM33 coregulatory gene signature that is overexpressed in PCa, essential for disease progression, and predicts recurrence-free survival. Results TRIM33 is a novel AR-interacting protein To obtain an unbiased and comprehensive catalog of AR-interacting proteins in PCa, we performed a series of RIME analyses (Mohammed et al, 2016) with anti-AR antibody and IgG (control) on dihydrotestosterone (DHT)-stimulated LNCaP and VCaP cells. We determined high confident AR interactors by applying the following stringent criteria: (i) interacting proteins are present in three biological replicates, (ii) for each interacting protein, the ratio of normalized label-free quantification (LFQ) intensity between the AR pull-down and the IgG pull-down > 5, and (iii) the number of unique peptides representing each interacting protein in the AR pull-down > 4. Based on these criteria, we identified 134 and 177 AR-interacting proteins in LNCaP and VCaP cells, respectively (Dataset EV1). Overlapping the AR-interacting proteins from LNCaP and VCaP cells revealed 91 common proteins between the two cell lines (Fig EV1 and Dataset EV1). When we categorized the AR-interacting proteins according to their biological processes, we found they are enriched for functional roles such as DNA binding, chromatin binding, RNA binding, and enzyme binding (Fig 1A). Our AR-RIME on PCa cells captured many well-characterized AR coactivators (e.g., EP300, NCOAs, and PARP1), corepressors (e.g., NCoRI, SMRT, and HDACs), and components of the large chromatin-modifying complex SWI/SNF (e.g., SMCA5, SMARCB1, SMARCD2, and SMARCC2; Sung & Cheung, 2014). We also found numerous AR collaborative factors, including NKX3-1, HOXB13, GRHL2, and pioneer factors FOXA1 and GATA2 (Sung & Cheung, 2014). While we were able to identify many known components of the AR coregulatory complex, approximately 50% of the hits from our RIME analysis are novel AR-interacting proteins. Among the top-ranked hits that appear in both LNCaP and VCaP cells is TRIM33 (Fig 1A and B), a member of the tripartite motif-containing family of proteins involved in transcriptional regulation, cell growth, and tumorigenesis (Hatakeyama, 2011). We validated the association between AR and TRIM33 in coimmunoprecipitation assays (Fig 1C). We also performed in vitro GST pull-down assays with bacterially expressed and purified proteins, GST-AR immobilized to GSH beads, and his-MBP-TRIM33. Our results show that immobilized GST-AR could pull down his-MBP-TRIM33, suggesting AR interacts with TRIM33 directly (Fig 1D). Figure 1. TRIM33 is a novel protein interactor of AR A. MS-ARC graphs showing AR-associated proteins with DHT stimulation in LNCaP cells (left) and VCaP cells (right). The proteins were clustered according to their molecular function. The length of each line represents the ratio of normalized LFQ intensity between the AR pull-down and the IgG pull-down. B. Peptide coverage of TRIM33 for the corresponding RIME experiment. C. Validation of the AR and TRIM33 interaction by coimmunoprecipitation assays. Endogenous AR protein in LNCaP cells (left) or VCaP cells (right) were immunoprecipitated and detected for TRIM33 by immunoblotting. IgG served as a negative control. D. In vitro GST pull-down assays were performed with bacterially expressed and purified GST or GST-tagged AR immobilized onto GSH beads. Immobilized GST protein beads were used to pull down purified his-MBP (left) or his-MBP-tagged TRIM33 (right). Download figure Download PowerPoint Click here to expand this figure. Figure EV1. Profiling proteins associated with chromatin-bound AR by RIME in PCa cellsVenn diagrams showing the overlap of proteins identified in LNCaP and VCaP cells from all replicates. Download figure Download PowerPoint An interplay between AR and TRIM33 cistromes in androgen signaling To begin determining whether TRIM33 is a coregulator of AR in androgen-dependent transcription, we first asked whether TRIM33 is recruited to AR binding sites (ARBSs). To do this, we performed ChIP-seq of TRIM33 in LNCaP cells. Overall, we identified 7,532 and 12,431 TRIM33 binding sites (T33BSs) before and after DHT treatment, respectively (Fig 2A). Besides increasing the number of T33BS, DHT also enhanced the average binding intensity of TRIM33 (Fig 2B). By overlapping the T33BSs with ARBSs from our previous AR ChIP-seq dataset (Tan et al, 2012), we identified 3,067 AR-TRIM33 colocalized binding sites (Fig EV2A). TRIM33 binding intensity at colocalized sites was also enhanced upon DHT stimulation (Fig 2C). Similar to the genomic distribution profile of AR (Tan et al, 2012), AR and TRIM33 are mainly co-occupied in distal intergenic and intronic regions (Fig 2D). Moreover, motif analysis of the AR-TRIM33 shared cistrome revealed significant enrichment for the AR binding motif (ARE; Fig 2E), suggesting TRIM33 is recruited to these binding sites by AR. To determine whether TRIM33 potentially plays a role in AR-dependent transcription, we searched for TRIM33 binding at ARBSs associated with model androgen-regulated genes. As shown in Fig 2F and G, DHT strongly stimulated TRIM33 binding at ARBSs of KLK2, KLK3, and FKBP5. Moreover, depleting AR abolished TRIM33 binding at these ARBSs without affecting TRIM33 protein level (Figs 2G and EV2B), further supporting that AR recruits TRIM33. Together, our results suggest an interplay between the AR and TRIM33 cistromes in androgen-regulated transcription. Figure 2. The interplay between TRIM33 and AR cistromes in PCa A. Venn diagram overlapping EtOH- and DHT-treated TRIM33 cistromes in LNCaP cells. B. Comparison between tag densities of TRIM33-binding sites (T33BSs) from EtOH- and DHT-treated LNCaP cells. The tag density distribution is represented in the form of a violin plot. The inner boxplot shows the first and third quartiles and is split by the medians, whiskers extending a 1.5-fold interquartile range beyond the box. C. Average tag densities of AR (left) and TRIM33 (right) binding at 3,067 shared binding sites from EtOH- and DHT-treated LNCaP cells (upper panel). Heatmaps show AR (left) and TRIM33 (right) normalized binding intensity at shared binding sites (lower panel). D. Pie chart illustrating the genomic distribution of AR-TRIM33 shared binding sites on chromatin. The assessed binding distribution included intron, distal intergenic, exon, 5' and 3' UTRs, and promoter regions. E. Top motifs enriched from DHT-treated AR-TRIM33 shared peaks. P-values for enrichment over genomic background are shown. F. Genomic snapshots of AR and TRIM33 ChIP-seq peaks surrounding FKBP5, KLK3, and KLK2. G. Bar graphs showing the ChIP-qPCR results of AR and TRIM33 binding at FKBP5, KLK3, and KLK2 in LNCaP (top) and VCaP (bottom) cells with or without transient AR knockdown. The data are represented as a percentage of input chromatin immunoprecipitated. Data information: In (G), data are presented as mean ± SD (n = 3 independent experiments); *P < 0.05; **P < 0.01; ns = not significant (two-way ANOVA). Download figure Download PowerPoint Click here to expand this figure. Figure EV2. TRIM33 colocalizes with AR across the genome A. Venn diagram showing the overlap between the AR and TRIM33 cistromes in the presence of DHT. B. Western blot analysis of TRIM33 in LNCaP and VCaP cells with or without AR knockdown. Download figure Download PowerPoint TRIM33 is a coactivator of the AR transcriptional program Since TRIM33 interacts with AR and both factors are colocalized on chromatin, we next examined whether TRIM33 acts as a positive or negative coregulator of AR transcriptional activity. To do this, we performed luciferase reporter assays on LNCaP cells in which we depleted or overexpressed TRIM33 together with a luciferase reporter construct containing AR responsive elements upstream of a minimal TATA box. TRIM33 depletion significantly reduced AR transcriptional activity (Fig 3A), while overexpression led to a concentration-dependent increase in AR activity (Fig 3B). Moreover, depleting AR abrogated the positive transcriptional effect of TRIM33 (Fig 3C). We also assessed the effect of TRIM33 on endogenous androgen-regulated genes. Consistent with the transient transfection results, silencing TRIM33 reduced the expression of KLK2, KLK3, and FKBP5 in LNCaP and VCaP cells (Fig 3D), while overexpressing TRIM33 had the opposite effect (Fig 3E). Together, these findings suggest TRIM33 is a coactivator of AR-dependent transcription in PCa cells. Figure 3. TRIM33 enhances the AR transcriptional program A–C. AR-dependent luciferase assays were performed on LNCaP cells transfected with siTRIM33 or negative control (NC) (A), increasing amounts of TRIM33 plasmid (B), or TRIM33 plasmid and/or siAR (C). Western blot analysis shows the protein level of TRIM33 in LNCaP cells after TRIM33 knockdown (A) or overexpression (B). D. LNCaP cells (left) and VCaP cells (right) were transfected with the indicated siRNAs and treated with EtOH or 10 nM DHT for 12 h. Gene expression was measured by RT-qPCR after 48 h. Data were normalized against GAPDH. E. LNCaP (left) or VCaP (right) cells were transfected with the indicated amount of TRIM33 plasmid and treated with EtOH or 0.1 nM DHT for 12 h. Gene expression was measured by RT-qPCR after 48 h. Data were normalized against GAPDH. F. GSEA was performed to determine the enrichment of androgen-induced (upper panel) or -repressed gene set (lower panel) in siTRIM33-regulated gene set identified by RNA-seq in LNCaP cells treated with DHT. G. Gene expression of LNCaP cells treated with siTRIM33 or NC were profiled by RNA-seq. A total of 414 coinduced genes (androgen-induced and siTRIM33 suppressed) and 536 corepressed genes (androgen-suppressed and siTRIM33-induced) were identified and clustered across all samples. H. Functional enrichment analysis showing the main biological processes and pathways significantly enriched by AR-TRIM33-coactivated genes (P < 0.05, DAVID 6.8). Data information: In (A–E), data are presented as mean ± SD (n = 3 independent experiments); *P < 0.05; **P < 0.01; ns = not significant (two-way ANOVA). Download figure Download PowerPoint To determine the extent of TRIM33 coactivator activity on the global AR transcriptional program in PCa cells, we performed RNA-seq analysis on DHT-stimulated LNCaP cells with or without TRIM33 depletion. Overall, we identified 3,516 TRIM33 responsive genes (1,623 up and 1,893 down; Dataset EV2). We found that TRIM33-regulated genes are significantly associated with TRIM33 binding (Fig EV3A), suggesting TRIM33 can directly regulate transcription in PCa cells. We also examined TRIM33-regulated genes together with DHT-regulated genes (1,240 up and 1,249 down, Dataset EV3). In gene set enrichment analyses (GSEA), androgen-induced genes are significantly enriched in genes downregulated by siTRIM33, while androgen-repressed genes are highly enriched in genes de-repressed by siTRIM33 (Fig 3F). These results are consistent with the above transient analyses and further support TRIM33 as an AR coactivator. Click here to expand this figure. Figure EV3. AR and TRIM33 collaboration is preferentially associated with activate transcription in PCa cells A. Venn diagram shows the association of TRIM33 binding sites with TRIM33-regulated genes. A two-tailed Fisher exact test (P = 2.5421E-21) was used to determine the significance of the overall association. B. Bar graphs show the number of coactivated (red) or corepressed (blue) genes associated with AR and TRIM33 cobinding sites determined by hypergeometric testing. The number of genes expected by chance is also plotted. Odds ratio (OR) was used to measure the association between categorical variables, with OR > 1 indicating a positive association. P-value was calculated by the Fisher's exact test. Download figure Download PowerPoint Next, we investigated the biological function of the AR-TRIM33 coregulated transcriptional program in PCa. Of the 1,240 DHT-upregulated genes, 414 were suppressed by TRIM33 depletion, including model AR-regulated genes, such as KLK2, KLK3, and FKBP5 (Fig 3G and Dataset EV4). As for the 1,249 DHT-downregulated genes, 536 were de-repressed by TRIM33 depletion (Fig 3G and Dataset EV4). In addition, we found that AR-TRIM33-coactivated but not corepressed genes are significantly associated with AR and TRIM33 cobinding, suggesting that AR and TRIM33 collaboration preferentially activate transcription (Fig EV3B). Functional enrichment analyses of the AR-TRIM33-coactivated genes show significant enrichment for biological processes associated with DNA metabolism, cell cycle, and cell mitosis, and pathways involved in prostate tumorigenesis (e.g., PI3K-Akt, MAPK FoxO, and ErbB signaling pathways; Fig 3H and Table EV1). In contrast, corepressed genes are enriched for biological processes related to cell proliferation and the native immune response (Appendix Fig S1 and Table EV1). Together, our results show that TRIM33 plays a critical role in shaping the AR transcriptome and suggests it promotes PCa growth by activating pro-mitotic genes. TRIM33 facilitates AR binding to chromatin To address the underlying mechanism of how TRIM33 upregulates the AR transcription program, we asked whether TRIM33 behaves as a coactivator by augmenting AR chromatin binding. To test this possibility, we examined the effect of global AR binding to chromatin before and after TRIM33 depletion. Our findings show that TRIM33 depletion significantly reduced the average binding intensity of AR at the 3,067 shared binding sites (Fig 4A), including the ARBSs associated with KLK2, KLK3, and FKBP5 (Fig 4B). We validated the effect of TRIM33 knockdown on ARBSs at these model genes in LNCaP and VCaP cells by ChIP-qPCR (Fig 4C). Together, these results indicate that TRIM33 is required for maximal AR chromatin binding. Figure 4. TRIM33 regulates the binding of AR on chromatin A. Line plot (upper panel) showing the average AR tag density of the 3,067 AR-TRIM33 shared binding sites in LNCaP cells treated with siTRIM33#1 (left) or siTRIM33#2 (right). Heatmap (lower panel) showing the normalized AR ChIP-seq binding intensities of the same AR-TRIM33 shared binding sites in the upper panel. B. Genomic snapshots of AR ChIP-seq peaks surrounding FKBP5, KLK3, and KLK2. C. ChIP-qPCR validation of AR binding sites after TRIM33 knockdown at FKBP5, KLK3, and KLK2 genes in LNCaP and VCaP cells. Data are represented as a percentage of input chromatin immunoprecipitated. Data are presented as mean ± SD (n = 3 independent experiments); **P < 0.01 (two-way ANOVA). Download figure Download PowerPoint TRIM33 stabilizes AR protein levels TRIM proteins have multiple functions, including the ability to stabilize their substrates (Wei et al, 2016; Fong et al, 2018). Thus, we speculated that TRIM33 might be working as