Divergent androgen regulation of unfolded protein response pathways drives prostate cancer

• DOI: 10.15252/emmm.201404509
• Publication Date: 2015-04-12T00:21:41Z
• Authors: Xia Sheng; Yke Jildouw Arnol...; Margrethe Storm; Martina Tesikova; Hatice Zeynep Nen...; Sen Zhao; Ladan Fazli; Paul Rennie; Bjørn Risberg; Håkon Wæhre; Håvard Danielsen; Ian G Mills; Yang Jin; Gökhan Hotamisligil; Fahri Saatcioglu
• Keywords: Male; 0301 basic medicine; 570; Medicine (General); UPR; QH426-470; Protein Serine-Threonine Kinases; 03 medical and health sciences; R5-920; SDG 3 - Good Health and Well-being; androgen receptor; Cell Line, Tumor; Endoribonucleases; Genetics; Humans; /dk/atira/pure/sustainabledevelopmentgoals/good_health_and_well_being; name=SDG 3 - Good Health and Well-being; Research Articles; Cell Proliferation; androgens; prostate cancer; 3. Good health; Gene Expression Regulation; Receptors, Androgen; Androgens; Unfolded Protein Response; /dk/atira/pure/sustainabledevelopmentgoals/good_health_and_well_being; ER stress

Research Article11 April 2015Open Access Source Data Divergent androgen regulation of unfolded protein response pathways drives prostate cancer Xia Sheng Department Biosciences, University Oslo, Norway Search for more papers by this author Yke Jildouw Arnoldussen Margrethe Storm Martina Tesikova Hatice Zeynep Nenseth Sen Zhao Ladan Fazli The Vancouver Prostate Centre, Vancouver, BC, Canada Paul Rennie Bjørn Risberg Institute Cancer Genetics and Informatics, Oslo Hospital, Division Pathology, Håkon Wæhre Surgery, Center Biomedicine, Håvard Danielsen Ian G Mills Centre Molecular Medicine Norway, Urology, Prevention, Research, Radium Yang Jin Gökhan Hotamisligil Complex Diseases, Harvard School Public Health, University, Boston, MA, USA Fahri Saatcioglu Corresponding Author Information Sheng1,‡, Arnoldussen1,12,‡, Storm1,‡, Tesikova1, Nenseth1, Zhao1, Fazli2, Rennie2, Risberg3,4, Wæhre3,5,6, Danielsen3,6,7, Mills8,9,10, Jin1, Hotamisligil11 1,3 1Department 2The 3Institute 4Division 5Division 6Center 7Department 8The 9Department 10Department 11Department 12Present address: Biological Chemical Work Environment, National Occupational ‡These authors contributed equally to work *Corresponding author. Tel: [phone]; Fax: [phone]; E-mail: [email protected] EMBO Mol Med (2015)7:788-801https://doi.org/10.15252/emmm.201404509 PDFDownload PDF article text main figures. Peer ReviewDownload a summary the editorial decision process including letters, reviewer comments responses feedback. ToolsAdd favoritesDownload CitationsTrack CitationsPermissions Figures & Info Abstract (UPR) is homeostatic mechanism maintain endoplasmic reticulum (ER) function. UPR activated various physiological conditions as well in disease states, such cancer. As androgens regulate secretion development normal drive (PCa) growth, they may affect pathways. Here, we show that canonical are directly divergently regulated PCa cells, through receptor (AR), which critical survival. AR bound gene regulatory sites IRE1α branch, but simultaneously inhibited PERK signaling. Inhibition arm profoundly reduced cell growth vitro tumor formation preclinical models vivo. Consistently, expression were correlated human PCa, spliced XBP-1 was significantly upregulated compared with prostate. These data establish genetic switch orchestrated regulates suggest targeting signaling have therapeutic utility PCa. Synopsis Androgens generate divergent whereby activated, one inhibited. Genetic depletion or small molecule inhibition tumor. upregulate signaling, downregulate UPR. effects on mediated direct binding increases transcription XBP-1S target genes. specimens increased tissue. genetically inhibitors dramatically reduces vivo, suggesting serve potential Introduction an essential organelle folding impacts key functions cell, lipid biosynthesis, calcium homeostasis (Hetz, 2012). Different pathological interfere capacity ER, leads accumulation misfolded proteins, named ER stress (for review, see Tabas Ron, 2011). In attempt cope stress, several intracellular signal transduction pathways, collectively termed (UPR), activated. aims increase lumen, thus decreasing load cell. If unsuccessful, however, apoptotic death results. at least three well-conserved sensors ER-localized transmembrane receptors: pancreatic kinase-like kinase (PERK), activating factor 6 (ATF6), inositol-requiring 1 (IRE1) (Tabas model, unstressed these proteins held inactive state chaperones, inhibit their activity. Upon chaperones dissociate from receptors bind lumen allowing oligomerization membrane, translocation ATF6α Golgi where it cleaved into active factor. This then specific turn orchestrates adaptation stress. Endoplasmic has been linked many chronic diseases, diabetes, neurodegeneration, cancers, proinflammatory reviews, Hotamisligil, 2010; Clarke et al, 2014). cancer, wide range cytotoxic hypoxia, pH changes, nutrient deprivation trigger activation help cells Thus, setting could be cytoprotective important role especially tumors arising secretory case multiple myeloma Tsai Weissman, 2010). For example, IRE1α-X-box-binding (XBP-1) pathway shown promote xenograft (Romero-Ramirez 2004) loss sensitized oxidative (Liu 2009). Transgenic mice studies splicing occurs during primary number breast (Spiotto However, under certain conditions, lead apoptosis, example c-Jun N-terminal (JNK), thereby inhibiting Furthermore, can impact autophagy mitophagy, two processes impart survival benefit (Kroemer Maes Agostinis, paradoxical: involved adaptive also initiate Liu Ye, 2011; Vandewynckel 2013; There limited information date. Global profiling experiments androgen-treated changes some stress-associated genes, N-myc downstream-regulated (NDRG1), disulfide isomerase-related (PDIR), homocysteine-responsive ER-resident ubiquitin-like domain member (HERPUD1), oxygen-regulated 150 (ORP150) (Segawa 2002). models, indicated downregulation branches high-grade PIN Nkx3.1:Pten mutant mice, mouse model Expression ER-associated molecules, HERPUD1 NDRG1, samples patients 2002), GRP78 levels associated greater risk recurrence worse (Pootrakul 2006; Daneshmand 2007). highly hormonal signals, particular via initiation progression Bluemn Nelson, We postulated developed ways engage hormonally sustain tissue integrity support tumorigenesis. induce unique profile coordinately translational implications. Results Given function would burden hypothesized affected To assess this, investigated possible concordance between set 190 tumors. (Supplementary Fig S1 Supplementary Table S1). stratified groups according status, ARLow (n = 60), ARmedium 70), ARhigh assessed expression. 1A, S2). profiles prominent ERN1 (IRE1), degradation-enhancing alpha-mannosidase-like (EDEM1), ATF6 DNAJC3 (or 58 kDa interferon-induced kinase, P58IPK), presented separately ease evaluation (Fig 1B). validated using independent patient cohort S3), Figure 1. Correlation cohorts Possible correlation AR- UPR-associated global available TCGA Adenocarcinoma 190) (http://www.cbioportal.org/public-portal/index.do). Tumors status groups, ARlow 60). Pearson's analysis R software heatmap. significant differences genes (A), EDEM1, ATF6, (P58IPK), presented. P-values different given. Download figure PowerPoint activate branch vivo next whether LNCaP cells. time-dependent manner upon administration 2A). principal target, XBP-1S, similar 2B). contrast, there very marginal, change unspliced S2A). established DnaJ 4 (ERdj4), P58IPK, ribosome-associated membrane (RAMP4), robustly treatment S2B–E). studied CWR22. withdrawal castration, CWR22 regress due decrease situ (Wainstein 1994). decreased castration up 72 h followed back basal 120 2C). after reaching approximately 40% 2D), whereas XBP-1U not 2E). level phosphorylated IRE1α, total 2F). 2. armsLNCaP cultured treated R1881 times. A, B. mRNA quantitative PCR (qPCR). Controls vehicle 84 100. represent mean triplicate, bars SE. ranged 1.66 × 10−5 0.025, 48 *P 0.013 respect vehicle-treated unpaired Student's t-test. C–E. xenografts grown nude collected times castration. value t 0 Columns each time point, 5.7 10−10 0.0002. post-castration 3.17 10−6 F, G. Protein Western blot analysis. representative experiments. online figure. 2 [emmm201404509-sup-0005-SDataFig2.pptx] differentially pathway. results eIF2α phosphorylation inhibits translation, alleviating overload. Both 2G). p-eIF2α rapidly exposure confirming addition general synthesis, promotes translation subset ATF4 (Holcik Sonenberg, 2005). Whereas S3A), addition, CHOP, downstream ATF4, S3B) level, its Altogether, indicate selectively UPR, while branch. Supporting obtained VCaP another androgen-responsive line S4A). natural dihydrotestosterone (DHT) induced S4B), physiological. determine third pathway, targets reagents present assay above S2A), only slightly treatment. Similarly, gene, GRP78, modestly (approximately 2-fold) S3C). lesser degree arm. apoptosis One Kinase (JNK) (Urano 2000; Nishitoh Since IRE1α-XBP-1S generally proliferative effects, JNK induces (Lorenzo Saatcioglu, 2008), determined activation. UV-induced stimulation S4C). consistent previous findings showing block triggered inducers, thapsigargin (e.g., Lorenzo 2008). knockdown effect viability. S5A, coincided cleavage caspase-3 S5B). cleavage, whole IRE1α-XBP-1 viability Together above, proapoptotic affects require AR. siRNA-mediated resulted 60–70% reduction 3A), completely blocked R1881-induced control siRNA 3B). prevented androgen-induced 3C), 3E). CHOP 3D) without affecting S6E). other targets, RAMP4, ERdj4, S6A–D) underscoring importance Similar 3F, quantification S6F). Taken together, required selective 3. influences membersAfter starvation, transfected (CTRL) siRNA. Cells h. A. qPCR CTRL Bars SE 0.001 indicating difference siRNA- siRNA-transfected paired B–E. Same points stimulation. F. top label Representative blots shown. 3 [emmm201404509-sup-0006-SDataFig3.pptx] Direct sequences examined chromatin immunoprecipitation sequencing (ChIP-seq) (Massie 2011) suggested binds validate observations, performed individual ChIP efficiently loaded classical (Brookes 1998) PSA enhancer 4A). Two predicted ChIP-Seq showed 6- 8-fold enrichment RAMP4 EDEM1 had 2.5- 4-fold 4B). vicinity 4. (A) (B) (C) 24 fixed, described Materials Methods antibody. experiment duplicate. Error < 0.01 0.04 C (control) both explored growth. Knockdown led proliferation 5A). Ectopic restored 5B) XBP-1S. 5. factors either (5 nM) starved 2% CT-FCS medium days before measured CCK-8 assay. graph triplicate repeated SD 6.6 4.5 comparison Ctrl against XBP-1, respectively, rescues defect siIRE1α-transfected 5 nM Lipofectamine RNAiMax reagent. day transfection, vector (Empty) Flag-XBP-1S (XBP-1S). Three harvested being applied triplicate. 0.02, **P 8.54 10−7 C, D. clonogenic Control LN-Scr (Scr), LN-shIRE1 (shIRE1), LN-shXBP1 (shXBP-1) weeks. colonies formed stained crystal violet photographed. extent area covered quantified Gene Tools (SynGene). SEM. 4.38 3.39 10−20 E. Growth xenografted mice. expressing shRNA (LN-shIRE1), XBP1 (LN-shXBP-1), (LN-Scr) subcutaneously implanted flanks male (6 per group). Tumor size points. pictures harvest 0.03 shIRE1 week 7, P 0.02 shXBP-1 8 PCNA immunostaining animals bearing LN-shIRE1, LN-shXBP-1, Sca