Epigenetically altered miR-193b targets cyclin D1 in prostate cancer

1417 Introduction Prostate cancer (PC) is the second most frequently diagnosed cancer type and the sixth leading cause of cancerrelated deaths in males worldwide [1]. Localized PC can be cured with prostatectomy and/or radiation therapy. However, there is no curative treatment for advanced and castrationresistant disease [2]. The use of prostatespecific antigen (PSA) for the screening of asymptomatic men for prostate cancer is known to reduce the diseasespecific mortality, but screening is associated with overdiagnosis [3]. Therefore, there is a strong interest to find additional markers and therapeutic targets for the diagnosis and treatment of prostate cancer. MicroRNAs (miRNAs) could serve as such markers or as drug targets [4]. ORIGINAL RESEARCH Epigenetically altered miR193b targets cyclin D1 in prostate cancer Kirsi M. Kaukoniemi1,2,a, Hanna E. Rauhala1,a, Mauro Scaravilli1,2, Leena Latonen1,2, Matti Annala1, Robert L. Vessella3, Matti Nykter1, Teuvo L. J. Tammela4 & Tapio Visakorpi1,2 1Institute of Biosciences and Medical Technology - BioMediTech, University of Tampere, Tampere, Finland 2Fimlab Laboratories, Tampere University Hospital, Tampere, Finland 3Department of Urology, University of Washington, Seattle, Washington 4Department of Urology, University of Tampere and Tampere University Hospital, Tampere, Finland © 2015 The Authors. Cancer Medicine published by John Wiley & Sons Ltd. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. Keywords Cyclin D1, micro-RNA, prostate cancer Correspondence Tapio Visakorpi, Institute of Biosciences and Medical Technology, University of Tampere, Tampere FI-33014, Finland. Tel: +358 50 318 5829; Fax: +358 3 364 1291; E-mail: [email protected] Funding Information The research leading to these results was funded by Tampere University Doctoral Programme in Biomedicine and Biotechnology. In addition, grant support has been received from the Sigrid Juselius Foundation, the Finnish Cultural Foundation, Pirkanmaa Regional fund, the Academy of Finland, the Cancer Society of Finland, the Medical Research Fund of Tampere University Hospital, and the European Community’s Seventh Framework Programme ProspeR (FP7/2007-2013) under grant agreement no. HEALTH-F2-2007-201438. Received: 14 January 2015; Revised: 13 May 2015; Accepted: 27 May 2015 Cancer Medicine 2015; 4(9):1417–1425 doi: 10.1002/cam4.486 aEqually contributing authors. Abstract MicroRNAs (miRNA) are important regulators of gene expression and often differentially expressed in cancer and other diseases. We have previously shown that miR193b is hypermethylated in prostate cancer (PC) and suppresses cell growth. It has been suggested that miR193b targets cyclin D1 in several malignancies. Here, our aim was to determine if miR193b targets cyclin D1 in prostate cancer. Our data show that miR193b is commonly methylated in PC samples compared to benign prostate hyperplasia. We found reduced miR193b expression (P < 0.05) in stage pT3 tumors compared to pT2 tumors in a cohort of prostatectomy specimens. In 22Rv1 PC cells with low endogenous miR193b expression, the overexpression of miR193b reduced CCND1 mRNA levels and cyclin D1 protein levels. In addition, the exogenous expression of miR193b decreased the phosphorylation level of RB, a target of the cyclin D1CDK4/6 pathway. Moreover, according to a reporter assay, miR193b targeted the 3’UTR of CCND1 in PC cells and the CCND1 activity was rescued by expressing CCND1 lacking its 3’UTR. Immunohistochemical analysis of cyclin D1 showed that castrationresistant prostate cancers have significantly (P = 0.0237) higher expression of cyclin D1 compared to hormonenaïve cases. Furthermore, the PC cell lines 22Rv1 and VCaP, which express low levels of miR193b and high levels of CCND1, showed significant growth retardation when treated with a CDK4/6 inhibitor. In contrast, the inhibitor had no effect on the growth of PC3 and DU145 cells with high miR193b and low CCND1 expression. Taken together, our data demonstrate that miR193b targets cyclin D1 in prostate cancer. Cancer Medicine Open Access 1418 © 2015 The Authors. Cancer Medicine published by John Wiley & Sons Ltd. K. M. Kaukoniemi et al.miR-193b Targets Cyclin D1 in PC miRNAs are small, ~22 nucleotide long noncoding RNA molecules that were first discovered in Caenorhabditis elegans [5]. Since their discovery, miRNAs have been found to be important regulators of gene expression in other organisms as well [5, 6]. miRNAs modulate gene expression by binding to a complementary sequence in the 3’ untranslated region (3’UTR) of target messenger RNAs (mRNAs). The binding of a miRNA leads to the degradation of the mRNA molecule or alternatively, to the suppression of mRNA translation, depending on the level of complementary binding [7]. Because one miRNA can have multiple targets and one mRNA molecule can be targeted by many miRNAs, miRNAs form a complex level of regulation of gene expression. In addition to genetic changes (chromosomal rearrangements, deletions, amplifications, and mutations), epigenetic events such as aberrant promoter hypermethylation, global hypomethylation, and posttranscriptional histone modification may also cause the dysregulation of miRNAs in cancer [7]. Several miRNAs are deregulated in prostate cancer and have been shown to affect apoptosis, cell cycle, intracellular signaling, DNA repair, adhesion/migration, and androgen signaling (reviewed in [4]). We previously showed that hsamiR193b3p (aka miR193b) may function as an epigenetically regulated tumor suppressor in prostate cancer [8]. Using bisulfite sequencing we showed that miR193b is hypermethylated in some PC cell lines, with the 22Rv1 cell line being the most heavily methylated resulting in the loss of miR193b expression in the cells. Transient transfection of premiR193b into 22Rv1 cells caused significant growth reduction due to a decreased fraction of cells in Sphase of the cell cycle. There are several suggested target genes for miR193b in different cancers, for example, CCND1, ETS1, and KIT [9–16]. The cyclin D1encoding gene, CCND1, is targeted by miR193b in hepatocellular carcinoma [10], melanoma [11], and pancreatic cancer [14]. Cyclin D1 is frequently aberrantly expressed in cancer. It regulates the expression of genes that are involved in DNA replication and the DNA damage checkpoint. The best known function of cyclin D1 is most likely its catalytic function as a regulatory partner for the cyclindependent kinases (CDKs) 4 and 6. In response to cellular stimuli, cyclin D1 activates CDKs 4 and 6, which in turn phosphorylate various proteins, the principal substrate being retinoblastoma (RB) protein. When phosphorylated, RB releases the E2F transcription factor, which activates genes that are necessary for DNA synthesis and cell cycle progression from the G1 to S phase. In addition to its catalytic function, cyclin D1 also has noncatalytic functions, with the major function being transcriptional regulation [17]. In prostate cancer, cyclin D1 has been shown to function as a corepressor to androgen receptor (AR) [18–20]. Because cyclin D1 is a suggested target of miR193b in other cancers, we aimed to study whether it is also a target in prostate cancer. In addition, our goal was to confirm the hypermethylation of miR193b in clinical prostate cancer samples. Materials and Methods Cell lines, xenografts, and clinical samples The prostate cancer cell lines 22Rv1, PC3, LNCaP, and DU145 were obtained from the American Type Cell Collection (Manassas, VA). LAPC4 and VCaP cell lines were provided by Dr. Charles Sawyers (University of California at Los Angeles, Los Angeles, CA) and Dr. Jack Schalken (Radboud University Nijmegen Medical Center, Nijmegen, the Netherlands), respectively. All cell lines were cultured under recommended conditions. The 17 PC LuCaP xenografts were provided by one of the authors (R.L.V.). All clinical samples were obtained from Tampere University Hospital (TAUH, Tampere, Finland). Freshfrozen tissue samples, including 10 benign prostate hyperplasias (BPH), 26 untreated prostatectomy specimens, and nine castrationresistant tumors (CRPC), were used to study miR193b methylation, whereas 78 hormonally untreated PC prostatectomy specimens (Table S1) were used to study miR193b expression. BPH samples were obtained from transurethral resection (TURP) or (cysto) prostatectomies from patients with BPH or bladder cancer. Untreated cancer samples were obtained from prostatectomies and CRPC samples from TURP. The samples used in this study were histologically examined to contain >70% cancerous or hyperplastic tissue. For cyclin D1 immunohistochemical analysis, a total of 267 formalinfixed prostate cancer specimens (198 from prostatectomies and 69 CRPC samples) were used to construct tissue microarrays (TMAs; Table S2). Ethics Committee of Tampere University Hospital and National Authority for Medicolegal Affairs have approved the use of clinical tumor material. DNA and RNA extraction from clinical samples For the methylation analysis, freshly frozen tissue blocks were cut into 10 × 20micrometer sections using a cryotome. DNA was isolated using an AllPrep RNA/DNA minikit (Qiagen, Valencia, CA) according to the manufacturer’s protocol. For miR193b expression analysis and arrays, RNA was isolated using Trizol® reagent (Invitrogen, Life Technologies Corporation, Carlsbad, CA) according to the manufacturer’s protocol. 1419 © 2015 The Authors. Cancer Medicine published by John Wiley & Sons Ltd. miR-193b Targets Cyclin D1 in PCK. M. Kaukoniemi et al. Microarrays miRNA and mRNA expression data from cell lines was produced using Agilent Tehcnologies’ (Santa Clara, CA) miRNA microarray system (version 1 array containing 470 human and 64 viral miRNAs) [8] and Whole Human Genome Kit (4 × 44k) (Agilent Tehcnologies, Santa Clara, CA), respectively. For LuCaP xenografts Affymetrix (Santa Clara, CA) HG U133 Plus 2.0 microarray was used. The arrays were done according to the manufacturer’s protocols. Cell transfection and cotransfection To determine the effect of miR193b expression on cyclin D1, RB, and pRB levels, 22Rv1 cells were transfected with a final concentration of 5 nmol/L premiR193b or scrambled control using INTERFERin transfection reagent (Polyplustransfection, Illkirch, France) according to the manufacturer’s instructions. The cells were collected 3 days after transfection. miRNA precursors were purchased from Ambion (Applied Biosystems/Ambion, Austin, TX). For the Cyclin D1 rescue and cell cycle experiment, 22Rv1 cells were cotransfected with 10 μg of plasmid, pCMVCCND1 lacking a 3’UTR (Plasmid 19927; Addgene, Cambridge, MA), or the control plasmid pCMV6ACGFP (OriGene Technologies, Rockville, MD) along with either premiR193b or premiRscramble at a final concentration of 10 nmol/L, using jetPRIME transfection reagent (Polyplustransfection). The cells were collected 2 days after transfection. Flow cytometric analysis Cells were harvested by trypsinization, washed with phosphatebuffered saline (PBS) and fixed in cold 70% EtOH. Ethanol was aspirated, and the cells were washed and rehydrated with PBS. Staining was performed with propidium iodide (Sigma-Aldrich, St. Louis, MO), and cell cycle profiles were analyzed with Accuri C6 Flow cytometer. Luciferase reporter assay For the luciferase reporter assay, the pSGG3UTR plasmid was purchased from SwitchGear Genomics (Menlo Park, CA). The plasmid contains a luciferase gene fused with the 3’UTR of CCND1. 22Rv1 cells were cotransfected with the 3’UTR plasmid, a control plasmidcontaining Renilla luciferase and premiR193b or premiRscramble using Lipofectamine™ 2000 transfection reagent (Invitrogen) according to the manufacturer’s instructions. Firefly and Renilla luciferase activities were measured 24 h after transfection using the DualGlo Luciferase Assay System (Promega, Madison, WI). Renilla luciferase values were used for data normalization. The luciferase assay was performed in quadruplicate, and repeated four times. Western Blot Nuclear and cytoplasmic proteins for RB and pRB western blots and total protein for cyclin D1 western blots were isolated as described previously [21, 22]. A total of 20 μg of proteins were separated on 8% (RB, pRB) and 10% (Cyclin D1) SDSPAGE gels and blotted to a PVDF membrane (ImmobilonP, Millipore Corp., Billerica, MA). The membranes were incubated with primary antibodies against phosphoRb (Ser795, 1:1000, Cell Signaling, Danvers, MA), Rb (C15: sc50, 1:500; Santa Cruz Biotechnology, Inc., Dallas, TX) or cyclin D1 (clone SP4, 1:100; Dako, Glostrup, Denmark) and with antibodies against fibrillarin (C13C3, 1:4000; Cell Signaling) or actin (pan AB5 clone ACTN05, 1:400; Lab Vision Corp., Fremont, CA). After secondary antibody incubation (antirabbitHRP for phosphoRB, RB, cyclin D1 and fibrillarin, and antimouseHRP for Actin; 1:3000, Dako), the proteins were visualized by autoradiography. Immunohistochemistry Antibodies against cyclin D1 (dilution 1:100, clone SP4; Dako) and Ki67 (dilution 1:500, MM1; Leica Biosystems Newcastle Ltd., Newcastle upon Tyne, UK) were used with a Power Vision+ PolyHRP IHC kit (ImmunoVision Technologies Co., Hillsborough, CA) according to the manufacturer’s instructions and as described by Leinonen et al. [23]. The slides were scanned with an Aperio ScanScope XT scanner (Leica Microsystems GmbH, Wetzlar, Germany). The virtual microscope [24, 25] and the ImmunoRatio web application [26] were used to score the cyclin D1staining intensity (0–1 = negative and weak, 2 = moderate, and 3 = strong) in a blinded fashion. Quantitative realtime RTPCR For miR193b qRTPCR, a TaqMan microRNA Assay (Applied Biosystems, Foster City, CA) and the CFX96 qRTPCR detection system (BioRad Laboratories Inc., Hercules, CA) were used according to the manufacturers’ recommendations. miRNA expression was normalized to RNU6B expression. For CCND1 qRTPCR, firststrand complementary DNA synthesis was performed from total RNA using AMV reverse transcriptase (Finnzymes Inc., Espoo, Finland) according to the manufacturer’s instructions. The expression of CCND1 was measured with Maxima SYBR Green (Fermentas Inc., Burlington, ON, Canada) and the CFX96 qRTPCR detection system (BioRad Laboratories Inc.) and was normalized to the β - Actin 1420 © 2015 The Authors. Cancer Medicine published by John Wiley & Sons Ltd. K. M. Kaukoniemi et al.miR-193b Targets Cyclin D1 in PC reference gene. The primer sequences used for the CCND1 and βActin qRTPCR were: CyclinD1for 5’- CCCTCGGTG GGTCCTACTTCAA3’, CyclinD1rev 5’- TGGCATTTTGG AGAGGAAGT3’, and Bactinf4 5’- TGGGACGACATGGAG AAAAT3’, Bactinr4 5’- AGAGGCGTACAGGGATAGCA3’. miR193b methylation analysis Methylated DNA was enriched using MethylMiner™ Methylated DNA Enrichment Kit (Invitrogen) according to the manufacturer’s instructions. Briefly, 2 μg of genomic DNA was fragmented by sonication. Methylated DNA was enriched by binding to magnetic beads coated with the methylCpGbinding domain of the human MBD2 protein (MethylCpG Binding Domain Protein 2) and eluted as a single fraction using 2 M NaCl. Finally, the DNA was ethanolprecipitated and resuspended in 50 μL of DNasefree water. For qPCR, iQTM SYBR® Green supermix and CFX96 qRTPCR detection system (BioRad Laboratories Inc.) were used. The sequences of the primers used in the qPCR were: miR193b_DMR_f1 5’- TGGCGTTTCTGG TTTCTCTT3’ and miR193b_DMR_r2 5’- CGCACCTTTTCTCCTCAT TT3’. Each sample was run in duplicate. The methylation status of miR193b was calculated from the elutionfraction signal in relation to the total qPCR signal. Growth curves For growth analysis with the CDK4/6 inhibitor PD0332991 (Selleck Chemicals, Houston, TX), the 22Rv1 cells were seeded in 24well plates in quadruplicate for each inhibitor concentration. The following day, the cells were imaged (day 0), and then treated with 0 nmol/L, 100 nmol/L, 500 nmol/L, or 2000 nmol/L inhibitor. The medium was replaced every other day with fresh medium containing the inhibitor. The cells were imaged daily with an Olympus IX71 microscope with the OASIS automation control system and Surveyor imaging software version 5.5.5.26 (Objective Imaging Ltd., Cambridge, UK). Cell growth was analyzed by measuring the cell surface area with an inhouse macro and ImageJ software (NIH, Bethesda, MD). The experiments were repeated three times. Statistical analysis An unpaired t test was used to calculate the significant difference in cell cycle analysis and in luciferase activity between the premiRscramble and premiR193b transfected samples as well as the difference in miR193b expression between differentially staged prostatectomy tumors. The paired t test was used to calculate the Pvalue of the growth curve assays on the last day of the experiment. Fisher’s exact test was used to assess the significant difference in miR193b methylation in clinical samples. Fisher’s exact, chisquare, Mann–Whitney U, and unpaired t tests were used to analyze the association between clinicopathologic variables. A Pvalue <0.05 was considered significant. Spearman’s correlation was used to study the correlation of miR193b and CCND1 expression in cell lines and xenograft samples in microarray analysis. Results We have previously shown that miR193b is strongly hypermethylated in 22Rv1 cells and moderately methylated in VCaP cells but is not methylated in LAPC4, LNCaP, DU145, PC3, EP156T, or PrEC cells, as determined by bisulfite sequencing [8]. In addition, we sequenced five untreated and four CRPC clinical specimens and found increased methylation [8]. Here, we wanted to confirm hypermethylation of miR193b in clinical specimens. We recently performed genomewide methylation analysis of methylation by MeDIPsequencing (unpubl. data) and found that the most differentially methylated genomic region in proximity to miR193b is 16:14396975–14397475 (GRCh37), which is located 349 bp upstream of the miR193b gene. Here, we measured the methylation of miR193b in that region in BPH (n = 10), PC (n = 26) and CRPC (n = 9) samples using MethylMinerqPCR. The samples were classified as methylated if they showed more than 20% methylation. According to our analysis, miR193b was significantly more methylated in cancer than in BPH (P < 0.0001; Fig. 1A). Next, we measured the expression of miR193b in a larger set (n = 78) of prostatectomy samples and found that the expression was lower (P < 0.05) in stage pT3 tumors than in stage pT2 tumors (Fig. S1). However, we did not find an association between miR193b expression and the Gleason score or progressionfree survival (data not shown). Because CCND1 is a suggested target of miR193b target in hepatocellular and pancreatic carcinoma and melanoma [10, 11, 14], we aimed to determine if it is also a target in prostate cancer. The expression levels of miR193b and CCND1 were obtained using expression microarrays of PC cell lines and xenograft samples. The array data showed a negative correlation between miR193b and CCND1 expression in the cell lines (Fig. 1B, r = −0.7714) as well as in the xenograft samples (Fig. 1C, r = −0.2522). We then studied the expression of CCND1 mRNA and protein levels in 22Rv1 cells transiently transfected with premiR193b. A marked reduction in the CCND1 expression at both the mRNA (Fig. 2A) and protein (Fig. 2B) level was detected in cells transfected with premiR193b compared to the scrambled premiR transfected cells. Similar to 22Rv1 cells, the Cyclin D1 protein expression was diminished also in 1421 © 2015 The Authors. Cancer Medicine published by John Wiley & Sons Ltd. miR-193b Targets Cyclin D1 in PCK. M. Kaukoniemi et al. VCaP cells transiently transfected with premiR193b (Fig. S2). In addition, a marked reduction was detected in phosphoRB protein level in nuclear protein fractions (Fig. 2C). To confirm that miR193b binds to the 3’UTR - region of the CCND1 gene in prostate cancer cells, we performed a luciferase reporter assay. A pSGGluciferase plasmid containing CCND1 3’UTR at the 3’ end of the luciferase gene was transfected together with either premiR193b or premiRscramble into 22Rv1 cells which express low levels of miR193b endogenously. The inhibition of luciferase activity was observed when the cells were transfected together with the plasmid and premiR193b (P < 0.05) but not with premiRscramble (Fig. 3A), confirming that miR193b targets the 3’UTR of CCND1. To evaluate the significance of CCND1 as a miR193b target, a rescue experiment was performed. We used the phosphorylation of RB as a measure of cyclin D1 activity. PhosphoRB levels decreased in cells transfected with premiR193b. When the cells were cotransfected with premiR193b and the pCMVCCND1 plasmid lacking the 3’UTR of CCND1, no such reduction in phosphoRB levels was observed (Fig. 3B). Similarly, transient premiR193b transfection caused a reduction in the number of cells in S and G2/M phase fractions of cell cycle, whereas there was no difference in cells cotransfected with premiR193b and CCND1 lacking 3’UTR (Fig. S3). Because cyclin D1 regulates the activity of CDK4/6, we treated PC cell lines with the CDK4/6 inhibitor PD0332991 at different concentrations and measured the effect on prostate cancer cell growth. 22Rv1 and VCaP cells, which express low levels of miR193b and high levels of CCND1, Figure 1. miR193b is methylated in cancer samples and has inverse expression pattern compared to CCND1. (A) miR193b methylation in clinical samples (BPH, PC, and CRPC) was assessed by MethylMinerqPCR. miR193b and CCND1 expressions were studied by miRNA and mRNA microarray (B) in prostate cancer cell lines and (C) xenograft samples. Spearman correlation coefficiencies are given. BPH, benign prostate hyperplasias; CRPC, castrationresistant tumors; PC, prostate cancer. AB C Figure 2. miR193b overexpression reduces the expression of CCND1, the level of Cyclin D1 protein and phosphorylation level of retinoblastoma (RB). 22Rv1 prostate cancer cells were transfected with premiR193b or premiRcontrol. (A) Expression levels of CCND1 mRNA was measured using qRTPCR. Western blot analysis was used to detect protein expression of (B) Cyclin D1 from total proteins and (C) phosphorylation level of RB protein from nuclear protein fraction. Actin and fibrillarin antibodies were used as loading controls. A BC 1422 © 2015 The Authors. Cancer Medicine published by John Wiley & Sons Ltd. K. M. Kaukoniemi et al.miR-193b Targets Cyclin D1 in PC showed significant growth suppression at a 500 nmol/L concentration (P < 0.05; Fig. 4A, B), whereas the various concentrations of the inhibitor had little effect on the growth of DU145 (Fig. 4C), LNCaP LAPC4 or PC3 cells, which express high levels of miR193b and low levels of CCND1 (Fig. S4A–C). Figure S4D shows that the vehicle, dimethylsulfoxide (DMSO), has no effect on the cell growth of 22Rv1 cells at 0.4% concentration. To study the expression of cyclin D1 in clinical samples, we performed an immunohistochemical analysis on TMAs. The analysis showed that the CRPC samples expressed higher levels of cyclin D1 compared to PC samples (P = 0.0237; Table 1). In prostatectomy samples, the cyclin D1 staining intensity was not associated with Gleason score, pT stage, or diagnostic PSA levels. However, there was strong positive association between cyclin D1 and the proliferation marker Ki67 (P < 0.0001) in the prostatectomy cohort. CRPC cells expressing high levels of cyclin D1 had increased Ki67 values, although the association was not statistically significant. Discussion Here, we confirmed the hypermethylation of miR193b gene in prostate cancer. The hypermethylation of miR193b leads to reduced expression, as we have previously shown [8]. Because miRNAs regulate the expression of proteincoding genes, the key question is which proteins are targeted by miR193b in prostate cancer. CCND1 is a target in hepatocellular and pancreatic carcinoma, as well as in melanoma [10, 11, 14]. We have previously shown that the overexpression of miR193b reduces the proliferation of prostate cancer cells due to a decreased number of cells in Sphase of the cell cycle [8], suggesting that Figure 4. CDK 4/6 inhibitor PD0332991 suppresses the growth of (A) 22Rv1 and (B) VCaP cells but not (C) DU145 cells. Cells were treated with 0, 100, 500, and 2000 nmol/L concentrations of the inhibitor and growth was followed for 5 to 6 days. Each concentration was done in quadruplicates and each experiment was done in triplicates, averages from experiments ±SEM are shown. Pvalues of growth differences between different concentrations on day 5 or 6 were calculated using paired ttest, *Pvalue <0.05, **Pvalue <0.01. A B C Figure 3. miR193b targets CCND1 3’UTR. (A) Luciferase experiment was performed in 22Rv1 cells cotransfected with pSGGplasmid containing CCND1 3’UTR, Renilla luciferase plasmid, and premiR193b or premiRcontrol. Values were normalized against Renilla luciferase activity. The means of four experiments ±SEM are shown. *Pvalue <0.05 (B) Western blot analysis of pRB in 22Rv1 cells transfected with pCMVCCND1 plasmid lacking CCND1 3’UTR together with miRscramble or miR193b. Fibrillarin antibody was used as loading control for nuclear proteins. A B 1423 © 2015 The Authors. Cancer Medicine published by John Wiley & Sons Ltd. miR-193b Targets Cyclin D1 in PCK. M. Kaukoniemi et al. CCND1 could also be a target for miR193b in prostate cancer. We showed that the expressions of miR193b and CCND1 are inversely correlated in prostate cancer cell lines and xenografts. Subsequently, we demonstrated reduced mRNA and protein expression of CCND1 in 22Rv1 cells transiently transfected with premiR193b. Using a reporter assay, we confirmed that miR193b targets the 3’UTR of the CCND1 gene in 22Rv1 prostate cancer cells. Concordantly, Chen et al. [11]. and Xu et al. [10]. have previously shown by luciferase assay that miR193b targets the CCND1 3’UTR in Malme3M malignant melanoma and HepG2 hepatocellular carcinoma cells. In addition, we performed a rescue experiment, in which we cotransfected miR193b into 22Rv1 cells with and without the pCMVCCND1 plasmid lacking the 3’UTR of CCND1. These results show that CCND1 is a bona fide target of miR193b in prostate cancer cells. To assess the relevance of cyclin D1 targeting by miR193b in prostate cancer cells, we studied the activity of the cyclinD1–RB pathway in the regulation of the G1/S transition in the cell cycle. The transfection of 22Rv1 cells with miR193b reduced the level of phosphoRB, in accordance with the known function of cyclin D1 in the regulation of RB phosphorylation. In addition, the phosphorylated RB protein levels were rescued in pCMVCCND1/miR193b cells to those of the control cells. Finally, when prostate cancer cell lines were treated with the CDK4/6 inhibitor PD0332991, the cell lines responded to the treatment according to miR193b expression/methylation status and cyclin D1 levels. Those with low miR193b and high cyclin D1 (22Rv1, VCaP) responded to the drug with growth inhibition, while the others did not. These results demonstrate that the downregulation of cyclin D1 by miR193b is functionally relevant for prostate cancer cell growth. To assess the clinical significance of cyclin D1, we stained TMAs and found that the expression of cyclin D1 is higher in CRPC (n = 69) compared to hormonenaïve PC (n = 198). Previously, Drobnjak et al. [27]. showed increased cyclin D1 staining in 22 CRPC bone metastases compared to 86 primary PC tumors. In our prostatectomy samples, the expression was strongly associated with proliferation but not with Gleason score, pathological status, PSA value, or age at diagnosis. This is in line with previous studies showing that cyclin D1 expression is associated with the proliferation marker Ki67 but not with other clinicopathological variables [27–31]. Because it is known that one gene can be targeted by several miRNAs and that cyclin D1 is a key regulator of the cell cycle G1/S transition, it is not surprising that CCND1 has been suggested to be a target of more than one miRNA in prostate cancer. Bonci et al. reported that miR15a and 161 interact directly with the 3’UTR of CCND1 and reduce cyclin D1 protein expression, thereby reducing prostate cell proliferation [32]. However, the function of CCND1 targeting by these three miRNAs is altered by different mechanisms in prostate cancer. miR193b is epigenetically regulated and silenced by methylation [8], whereas the function of miRs 15a and 161 is disrupted by deletions in the encoding chromosomal region 13q14 [32]. We have previously reported a homozygous deletion of the miR15a and 161 locus in prostate cancer, although the frequency is relatively low [33]. In addition to CCND1, there are several other suggested targets for miR193b, such as YWHAZ, PLAU (aka uPA), and KIT in different cancer types [9–16]. Xie et al. [34]. showed that the knockdown of cystic fibrosis transmembrane conductance regulator (CFTR) led to the suppression of miR193b expression. They also showed that the forced overexpression of miR193b completely abrogated elevated urokinasetype plasminogen activator (uPA) activity after CFTRknockdown in PC3 cells, suggesting that the tumorsuppressing effect of CFTR is mediated through the miR193buPA axis. However, it should be noted that miR193b expression is low in 22Rv1 and VCaP cells [8], which do not express uPA [35, 36], suggesting that uPA is not the major target of miR193b in prostate cancer. In conclusions, we have demonstrated that the overexpression of cyclin D1 in prostate cancer is driven, at least partly, by the reduced expression of miR193b. The mechanism for miR193b suppression is the Table 1. Association of clinicopathological variables with Cyclin D1 expression. Variable Cyclin D1 expression P Negative (0/1) Positive (2–3) Prostatectomy specimens, n (%) 74 (37) 124 (63) Locally recurrent CRPCs, n (%)120 (29) 49 (71) 0.0237 Prostatectomy specimens Gleason score, n (%)1 <7 24 (33) 46 (37) 7 38 (52) 60 (48) >7 11 (15) 19 (15) 0.8337 pT Stage, n (%)2 pT2 54 (75) 88 (72) pT3 18 (25) 35 (28) 0.6216 PSA ng/mL (mean ± SD)320.0 ± 31.5 14.3 ± 11.3 0.9786 Age (mean ± SD)462.6 ± 5.2 63.2 ± 4.9 0.4167 Ki67 (mean ± SD) 7.1 ± 7.0 13.6 ± 14.5 <0.0001 Locally recurrent CRPCs, n (%)3 Ki67 (mean ± SD) 13.2 ± 9.3 20.7 ± 15.8 0.0950 1Chisquare test. 2Fisher’s exact test. 3Mann–Whitney Utest. 4Unpaired t test. 1424 © 2015 The Authors. Cancer Medicine published by John Wiley & Sons Ltd. K. M. Kaukoniemi et al.miR-193b Targets Cyclin D1 in PC hypermethylation of the DMR upstream of the miR193b gene. Additional studies are warranted to translate these findings to clinical benefits. For example, the restoration of miR193b expression could theoretically reduce the proliferation of prostate cancer cells. Alternatively, the loss of miR193b expression could indicate the sensitivity of prostate cancer cells to cyclin D1 inhibition. Acknowledgments We thank P. Martikainen and M. Vakkuri for the skillful technical assistance. The research leading to these results was funded by Tampere University Doctoral Programme in Biomedicine and Biotechnology. In addition, grant support has been received from the Sigrid Juselius Foundation, the Finnish Cultural Foundation, Pirkanmaa Regional fund, the Academy of Finland, the Cancer Society of Finland, the Medical Research Fund of Tampere University Hospital, and the European Community’s Seventh Framework Programme ProspeR (FP7/20072013) under grant agreement no. HEALTHF22007201438. Conflict of Interest None declared. References 1. Jemal, A., F. Bray, M. M. Center, J. Ferlay, E. Ward, and D. Forman. 2011. Global cancer statistics. CA Cancer J. Clin. 61:69–90. 2. Knudsen, K. E., and W. K. Kelly. 2011. Outsmarting androgen receptor: creative approaches for targeting aberrant androgen signaling in advanced prostate cancer. Expert. Rev. Endocrinol. Metab. 6:483–493. 3. Schroder, F. H., J. Hugosson, M. J. Roobol, T. L. Tammela, M. Zappa, V. Nelen, et al. 2014. Screening and prostate cancer mortality: results of the European Randomised Study of Screening for Prostate Cancer (ERSPC) at 13 years of followup. Lancet 384:2027–2035. 4. Catto, J. W., A. Alcaraz, A. S. Bjartell, R. De Vere White, C. P. Evans, S. Fussel, et al. 2011. MicroRNA in prostate, bladder, and kidney cancer: a systematic review. Eur. Urol. 59:671–681. 5. Lee, C., and V. Ambros. 2001. An extensive class of small RNAs in Caenorhabditis elegans. Science 294:862–864. 6. Lagos-Quintana, M., R. Rauhut, W. Lendeckel, and T. Tuschl. 2001. Identification of novel genes coding for small expressed RNAs. Science 294:853–858. 7. Di Leva, G., and C. M. Croce. 2010. Roles of small RNAs in tumor formation. Trends Mol. Med. 16:257–267. 8. Rauhala, H. E., S. E. Jalava, J. Isotalo, H. Bracken, S. Lehmusvaara, T. L. Tammela, et al. 2010. miR193b is an epigenetically regulated putative tumor suppressor in prostate cancer. Int. J. Cancer 127:1363–1372. 9. Li, X. F., P. J. Yan, and Z. M. Shao. 2009. Downregulation of miR193b contributes to enhance urokinasetype plasminogen activator (uPA) expression and tumor progression and invasion in human breast cancer. Oncogene 28:3937–3948. 10. Xu, C., S. Liu, H. Fu, S. Li, Y. Tie, J. Zhu, et al. 2010. MicroRNA193b regulates proliferation, migration and invasion in human hepatocellular carcinoma cells. Eur. J. Cancer 46:2828–2836. 11. Chen, J., H. E. Feilotter, G. C. Pare, X. Zhang, J. G. Pemberton, C. Garady, et al. 2010. MicroRNA193b represses cell proliferation and regulates cyclin D1 in melanoma. Am. J. Pathol. 176:2520–2529. 12. Chen, J., X. Zhang, C. Lentz, M. Abi-Daoud, G. C. Pare, X. Yang, et al. 2011. miR193b regulates Mcl1 in melanoma. Am. J. Pathol. 179:2162–2168. 13. Gao, X. N., J. Lin, L. Gao, Y. H. Li, L. L. Wang, and L. Yu. 2011. MicroRNA193b regulates cKit protooncogene and represses cell proliferation in acute myeloid leukemia. Leuk. Res. 35:1226–1232. 14. Ikeda, Y., E. Tanji, N. Makino, S. Kawata, and T. Furukawa. 2012. MicroRNAs associated with mitogenactivated protein kinase in human pancreatic cancer. Mol. Cancer Res. 10:259–269. 15. Leivonen, S. K., A. Rokka, P. Ostling, P. Kohonen, G. L. Corthals, O. Kallioniemi, et al. 2011. Identification of miR193b targets in breast cancer cells and systems biological analysis of their functional impact. Mol. Cell Proteomics 10:M110.005322. 16. Hu, H., S. Li, J. Liu, and B. Ni. 2012. MicroRNA193b modulates proliferation, migration, and invasion of nonsmall cell lung cancer cells. Acta Biochim. Biophys. Sin. (Shanghai) 44:424–430. 17. Musgrove, E. A., C. E. Caldon, J. Barraclough, A. Stone, and R. L. Sutherland. 2011. Cyclin D as a therapeutic target in cancer. Nat. Rev. Cancer 11:558–572. 18. Knudsen, K. E., W. K. Cavenee, and K. C. Arden. 1999. Dtype cyclins complex with the androgen receptor and inhibit its transcriptional transactivation ability. Cancer Res. 59:2297–2301. 19. Reutens, A. T., M. Fu, C. Wang, C. Albanese, M. J. McPhaul, Z. Sun, et al. 2001. Cyclin D1 binds the androgen receptor and regulates hormonedependent signaling in a p300/CBPassociated factor (P/ CAF)- dependent manner. Mol. Endocrinol. 15:797–811. 20. Petre, C. E., Y. B. Wetherill, M. Danielsen, and K. E. Knudsen. 2002. Cyclin D1: mechanism and consequence of androgen receptor corepressor activity. J. Biol. Chem. 277:2207–2215. 1425 © 2015 The Authors. Cancer Medicine published by John Wiley & Sons Ltd. miR-193b Targets Cyclin D1 in PCK. M. Kaukoniemi et al. 21. Jalava, S. E., K. P. Porkka, H. E. Rauhala, J. Isotalo, T. L. Tammela, and T. Visakorpi. 2009. TCEB1 promotes invasion of prostate cancer cells. Int. J. Cancer 124:95–102. 22. Urbanucci, A., B. Sahu, J. Seppala, A. Larjo, L. M. Latonen, K. K. Waltering, et al. 2012. Overexpression of androgen receptor enhances the binding of the receptor to the chromatin in prostate cancer. Oncogene 31:2153–2163. 23. Leinonen, K. A., T. T. Tolonen, H. Bracken, U. H. Stenman, T. L. Tammela, O. R. Saramaki, et al. 2010. Association of SPINK1 expression and TMPRSS2: ERG fusion with prognosis in endocrinetreated prostate cancer. Clin. Cancer Res. 16:2845–2851. 24. Tuominen, V. J., and J. Isola. 2009. The application of JPEG2000 in virtual microscopy. J. Digit. Imaging 22:250–258. 25. Tuominen, V. J., and J. Isola. 2010. Linking wholeslide microscope images with DICOM by using JPEG2000 interactive protocol. J. Digit. Imaging 23:454–462. 26. Tuominen, V. J., S. Ruotoistenmaki, A. Viitanen, M. Jumppanen, and J. Isola. 2010. ImmunoRatio: a publicly available web application for quantitative image analysis of estrogen receptor (ER), progesterone receptor (PR), and Ki67. Breast Cancer Res. 12:R56. 27. Drobnjak, M., I. Osman, H. I. Scher, M. Fazzari, and C. Cordon-Cardo. 2000. Overexpression of cyclin D1 is associated with metastatic prostate cancer to bone. Clin. Cancer Res. 6:1891–1895. 28. Murphy, C., M. McGurk, J. Pettigrew, A. Santinelli, R. Mazzucchelli, P. G. Johnston, et al. 2005. Nonapical and cytoplasmic expression of interleukin8, CXCR1, and CXCR2 correlates with cell proliferation and microvessel density in prostate cancer. Clin. Cancer Res. 11:4117–4127. 29. Aaltomaa, S., M. Eskelinen, and P. Lipponen. 1999. Expression of cyclin A and D proteins in prostate cancer and their relation to clinopathological variables and patient survival. Prostate 38:175–182. 30. Kallakury, B. V., C. E. Sheehan, R. A. Ambros, H. A. Fisher, R. P. Jr Kaufman, and J. S. Ross. 1997. The prognostic significance of p34cdc2 and cyclin D1 protein expression in prostate adenocarcinoma. Cancer 80:753–763. 31. Comstock, C. E., M. P. Revelo, C. R. Buncher, and K. E. Knudsen. 2007. Impact of differential cyclin D1 expression and localisation in prostate cancer. Br. J. Cancer 96:970–979. 32. Bonci, D., V. Coppola, M. Musumeci, A. Addario, R. Giuffrida, L. Memeo, et al. 2008. The miR15amiR161 cluster controls prostate cancer by targeting multiple oncogenic activities. Nat. Med. 14:1271–1277. 33. Porkka, K. P., E. L. Ogg, O. R. Saramaki, R. L. Vessella, H. Pukkila, H. Lahdesmaki, et al. 2011. The miR15amiR161 locus is homozygously deleted in a subset of prostate cancers. Genes Chromosom. Cancer 50:499–509. 34. Xie, C., X. H. Jiang, J. T. Zhang, T. T. Sun, J. D. Dong, A. J. Sanders, et al. 2013. CFTR suppresses tumor progression through miR193b targeting urokinase plasminogen activator (uPA) in prostate cancer. Oncogene 32:2282–2291, 2291.e1–7. 35. Helenius, M. A., K. J. Savinainen, G. S. Bova, and T. Visakorpi. 2006. Amplification of the urokinase gene and the sensitivity of prostate cancer cells to urokinase inhibitors. BJU Int. 97:404–409. 36. Prensner, J. R., M. K. Iyer, O. A. Balbin, S. M. Dhanasekaran, Q. Cao, J. C. Brenner, et al. 2011. Transcriptome sequencing across a prostate cancer cohort identifies PCAT1, an unannotated lincRNA implicated in disease progression. Nat. Biotechnol. 29:742–749. Supporting Information Additional supporting information may be found in the online version of this article: Table S1. Description of clinical samples used to study miR193b expression. Table S2. Description of clinical samples used in immunohistochemistry. Figure S1. miR193b expression in clinical samples. Figure S2. miR193b overexpression reduces the expression of CCND1 in protein level in VCaP cells. Figure S3. CCND1 lacking 3’ UTR is able to rescue cell cycle effects of miR193b. Figure S4. The effect of CDK 4/6 inhibitor PD0332991 and dissolvent dimethylsulfoxide (DMSO) on growth of (A) LAPC4, (B) LNCaP, (C) PC3, and (D) 22Rv1 cells.