Identification of Novel AR-Targeted MicroRNAs Mediating Androgen Signalling through Critical Pathways to Regulate Cell Viability in Prostate Cancer

et al. (2013) Identification of Novel AR-Targeted MicroRNAs Mediating Androgen Signalling through Critical Pathways to Regulate Cell Viability in Prostate Cancer. PLoS ONE 8(2): e56592. doi:10.1371/journal.pone.0056592 Identification of Novel AR-Targeted MicroRNAs Mediating Androgen Signalling through Critical Pathways to Regulate Cell Viability in Prostate Cancer Wenjuan Mo 0 Jiyuan Zhang 0 Xia Li 0 Delong Meng 0 Yun Gao 0 Shu Yang 0 Xuechao Wan 0 Caihong Zhou 0 Fenghua Guo 0 Yan Huang 0 Stefano Amente 0 Enrico V. Avvedimento 0 Yi Xie 0 Yao Li 0 Ming Tat Ling, Queensland University of Technology, Australia 0 1 State Key Laboratory of Genetic Engineering, Institute of Genetics, School of Life Science, Fudan University , Shanghai , China , 2 Department of Biology, University of Naples ''Federico II'' , Naples , Italy , 3 Department of Molecular Medicine and Biotechnology, Universita` degli Studi ''Federico II'' , Naples , Italy MicroRNAs (miRNAs) have been recognized as significantly involved in prostate cancer (PCa). Since androgen receptor (AR) plays a central role in PCa carcinogenesis and progression, it is imperative to systematically elucidate the causal association between AR and miRNAs, focusing on the molecular mechanisms by which miRNAs mediate AR signalling. In this study, we performed a series of time-course microarrays to observe the dynamic genome-wide expressions of mRNAs and miRNAs in parallel in hormone-sensitive prostate cancer LNCaP cells stimulated by androgen. Accordingly, we introduced Response Score to identify AR target miRNAs, as well as Modulation Score to identify miRNA target mRNAs. Based on theoretical identification and experimental validation, novel mechanisms addressing cell viability in PCa were unravelled for 3 miRNAs newly recognized as AR targets. (1) miR-19a is directly up-regulated by AR, and represses SUZ12, RAB13, SC4MOL, PSAP and ABCA1, respectively. (2) miR-27a is directly up-regulated by AR, and represses ABCA1 and PDS5B. (3) miR-133b is directly upregulated by AR, and represses CDC2L5, PTPRK, RB1CC1, and CPNE3, respectively. Moreover, we found miR-133b is essential to PCa cell survival. Our study gives certain clues on miRNAs mediated AR signalling to cell viability by influencing critical pathways, especially by breaking through androgen's growth restriction effect on normal prostate tissue. - Funding: This work is supported in part by two grants 31071142 and 31171246 from the National Natural Science Foundation of China, and by Shanghai Key Laboratory of Intelligent Information Processing of China (No. IIPL-2010-002). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. . These authors contributed equally to this work. MicroRNAs (miRNAs) are 20,24 nt endogenous proteinnonencoding RNAs, and have emerged as a major class of regulatory molecules involved in mammal embryonic development and pathogenesis [1]. Recently, an increasing number of studies have pointed out that miRNAs play strong roles in prostate cancer (PCa) initiation, progression and metastasis [1,2,3]. Prostate is dependent on androgens for growth and development, meanwhile its normal tissue is controlled by certain growth restriction mechanisms to avert androgen-induced over-growth, it is imperative to reveal how the androgen receptor (AR) mediates these actions and breaks through growth restriction for guiding PCa carcinogenesis. Thus, we attempted to systematically identify miRNAs that bridge the pathways from AR stimulation to cellular phenotypic effect in PCa. Presently, several reports identified miRNAs in AR signalling in the prostate cancer [4,5,6,7]. miR-21 was directly up-regulated by AR in androgen-responsive PCa cells [6], due to AR binding on the defined promoter. J. Ribas et al. further found inhibition of miR-21 can diminish androgen-induced PCa cell proliferation, and miR-21 was sufficient for androgen-dependent tumours to overcome castration-induced growth arrest [6]. miR-125b was direct stimulated by AR, and promoted androgen-independent PCa growth by repressing the expression of Bak1 which regulated apoptotic signalling in PCa [7]. J. Ribas et al. [6] performed microarray analysis for miRNA expression in two androgendependent PCa cell lines LNCaP and LAPC-4 to find ARregulated target miRNAs. However, it cannot clearly distinguish the direct and indirect targets of AR since the miRNA expression profile was obtained at 72 h after androgen stimulation. Recently, K. Takayama et al. have performed a genome-wide screening of AR target genes by integrating CAGE and ChIP-chip analysis to identify AR binding sites (ARBSs) in the human genome in LNCaP cells [4]. They determined genome-wide ARBSs in the 100 kb vicinity of miRNA genes. Based on the chromosome binding, K. Takayama et al. provided useful information for elucidating miRNA-mediated AR signalling network. However, under the special biological conditions, not all targets identified by ChIP-chip analysis are the real targets of AR; among all the ARtargeted miRNAs, the critical miRNAs contributing to AR signalling, may not be found out only through chromosomebinding analysis. On the other hand, to study how miRNA mediates AR signalling, it is necessary to identify miRNAs target mRNAs. Seed-sequence-based predictions of miRNA target, such as TargetScanS, miRanda and miRDB databases, provide necessary informative clues; application of these prediction database in the specific context can identify miRNAs actual targets. In most cases for animals, although miRNAs and target mRNAs are not completely base-matched, miRNAs can still cause target mRNA degradation via exonucleases or P-body [8]. Namely, the expression change of target mRNA can mainly reflect miRNAs regulation. Therefore, it is ideal to simultaneously observe expressions of both miRNAs and mRNAs in a time-series manner in order to efficiently identify miRNAs regulation on target in a tissue-specific context. Recently, V. Jayaswal et al. [9] have provided a dynamic data simultaneously observing miRNA and mRNA expressions in a myeloma cell line U266. Based on the matched miRNA-mRNA time-course data, they calculated odds statistic [9] for each miRNA-mRNA pair obtained from sequencebased prediction. Moreover, miRNA expression change may not necessarily produce an instantaneous change in target mRNA expression [9], this time-lag effect ought to be considered when determining miRNA regulation. In V. Jayaswal et al.s method [9], the significance of odds statistic was not assessed by false discovery rate which is the standard for assessing significance, and time-lags with different intervals for representing miRNAs delayed effect were equally treated that is not in line with actual circumstance. Therefore, an improved algorithm for accurately identifying miRNA target in a specific conte (...truncated)