Abstract: MicroRNAs (miRNAs) are small non-coding RNAs that negatively regulate gene expression. Recent findings indicate that miRNAs are dysregulated in human tumors, suggesting a potential role for these molecules in the pathogenesis of cancer. Thus far, only a limited number of studies have investigated miRNA expression in prostate cancer. Results from these studies indicate that miRNA expression profiles may distinguish carcinoma from non-neoplastic specimens and further classify tumors according to androgen dependence. In addition, a prognostic significance was attributed to specific miRNAs as predictors of clinical recurrence following radical prostatectomy. For a handful of miRNAs, for which a widespread dysregulation in prostate cancer was consistently found, functional investigation has been pursued in prostate cancer experimental models to establish the rationale for the development of miRNA-based therapies. A better understanding of the role exerted by specific miRNAs in the development and progression of prostate cancer is needed, as is a precise definition of their targets relevant to the disease. However, based on available findings, a possible role for miRNAs in the management of prostate cancer as novel biomarkers and new therapeutic targets or intervention tools can be envisioned.
Prostate Cancer and microRNAs
Prostate cancer is the most frequently diagnosed cancer in men in Western countries (Damber and Aus, 2008). At present, diagnosis of prostate cancer derives from serum prostate-specific antigen (PSA) measurement, digital rectal examination, and histopathological evaluation of prostate needle biopsies. Organ-confined prostate cancer is treated with surgery (radical prostatectomy) or radiation therapy, and an increase in PSA level usually represents the first sign of recurrence (Damber and Aus, 2008). Androgen deprivation is the current therapy of choice for patients with metastatic disease. Unfortunately, patients almost invariably experience disease progression and develop hormone-refractory disease (Damber and Aus, 2008). Eradication of advanced prostate cancer still represents an unsolved clinical problem, making the development of alternative treatment approaches highly desirable.
Due to the high variability in the course of prostate cancer, novel biomarkers are strongly needed at the time of diagnosis to facilitate treatment planning. Currently used prognostic tools (nomograms) exclusively rely on pathological and clinical parameters (Damber and Aus, 2008). In this context, understanding the molecular alterations that distinguish the progressive from non-progressive disease will allow not only the identification of novel biomarkers to be included in clinical nomograms for improving their predictive power, but will also provide mechanistic information for the discovery of new therapeutic targets and the design of tailored therapeutic interventions.
MicroRNAs (miRNAs) are endogenous small (about 22 nucleotides long) non-coding RNA molecules shown to have a key role in the regulation of gene expression. miRNA genes are transcribed as hairpin-containing primary miRNAs, processed into approximately 70-nucleotide precursors, exported to the cytoplasm and cleaved into mature single-stranded miRNAs. Binding of mature miRNAs to at least partially complementary sequences mostly situated in the 3′-untranslated regions of target mRNA leads to mRNA cleavage, decay, or inhibition of translation (Garzon et al., 2006). It is estimated that 30% of human genes are under the control of miRNAs. miRNAs have also been shown to participate in almost all pivotal biological processes (Garzon et al., 2006).
Expression profiling studies using microarrays and other methods have shown that miRNAs are dysregulated in a wide variety of human cancers. In some instances, the expression of selected miRNAs or specific miRNA signatures was found to correlate with diverse clinico-pathological features and to be able to predict patient clinical outcome and/or response to treatment (Garzon et al., 2006). Such findings have highlighted the potential of miRNAs as new diagnostic or prognostic/predictive biomarkers. Moreover, the role of miRNAs functioning as oncogenes and tumor suppressors, as emerged from functional studies in experimental models (Garzon et al., 2006), has generated great interest in their possible use as novel targets or tools for anticancer therapies.
miRNA Expression Studies in Prostate Cancer: A Possible Role for miRNAs as Novel Diagnostic and Prognostic Biomarkers?
Attempts to investigate dysregulation of miRNAs in prostate cancer and to establish a connection with the onset and progression of the disease have been mainly based on the analysis of miRNA expression profiles in prostate cancer versus normal/benign tissues. Global miRNA profiling data from seven independent studies are available. Initially, Lu et al. (2005) analyzed the expression of 217 miRNAs in different types of human cancer, including prostate cancer specimens (6 tumors and 8 normal tissues) and showed a general down-regulation of miRNAs in tumors compared with normal tissues. A subsequent large-scale microarray analysis conducted on different tumor histotypes, including prostate cancer specimens (56 carcinomas and 7 normal tissues from non-cancer individuals), identified a common miRNA signature for solid tumors that was mainly composed of over‑expressed miRNAs (Volinia et al., 2006). A prostate cancer-specific miRNA signature was first obtained by Porkka et al. (2007) on 9 prostate cancer and 4 benign prostatic hyperplasia samples. Differential expression of 51 miRNAs (37 down- and 14 upregulated) was found between cancer and nonmalignant tissues. Hierarchical clustering of tumors by their miRNA expression was shown to accurately separate prostate cancer from benign prostatic hyperplasia specimens and further classify the tumors according to their androgen dependence (hormone‑naive versus hormone‑refractory). Ozen et al. (2008), who analyzed the expression of 480 miRNAs in 16 clinically localized prostate cancer and 10 benign prostatic hyperplasia samples, found widespread down‑regulation of miRNAs in cancer, including miR‑125b, miR‑145, and let‑7c, for which expression data were confirmed by quantitative reverse-transcriptase polymerase chain reaction (qRT-PCR). A genome-wide expression profiling of miRNAs and mRNAs, performed by Ambs et al. (2008) in 60 primary prostate cancer samples and 16 surrounding non-tumor tissues, showed that key components of miRNA processing and several genes carrying miRNA coding sequences in their introns, such as miR-106b-25 cluster and its host gene MCM7, were significantly up‑regulated in tumors. In a further study carried out in a series of 57 primary prostate cancers, the same research group found a differential expression of 19 miRNAs in tumors with and without perineural invasion, with high-level expression of miR-224 in perineural cancer cells (Prueitt et al., 2008).
Tong et al. (2009) carried out the first paired analysis of miRNA expression on micro-dissected malignant and uninvolved areas of 40 prostatectomy specimens. Five miRNAs (miR-23b, miR-100, miR-145, miR-221, and miR-222) were found to be significantly down‑regulated in malignant compared with normal tissues. Very recently, Schaefer et al. (2010) analyzed miRNA expression in matched tumor and adjacent normal tissues obtained from 76 patients and identified 15 differentially expressed miRNAs: miR-16, miR-31, miR-125b, miR-145, miR-149, miR-181b, miR-184, miR-205, miR-221, and miR-222 were down‑regulated whereas miR-96, miR-182, miR-182*, miR‑183, and miR-375 were up‑regulated. Expression of 5 miRNAs correlated with Gleason score (miR-31, miR-96, and miR-205) or pathological tumor stage (miR-125b, miR-205, and miR-222). In addition, miR-205 was found to be the best discriminating miRNA, providing a correct overall classification of 72%.
Taking advantage of the stability of tumor-derived miRNAs in circulating blood, Mitchell et al. (2008) evaluated the expression of miR-141 in a case-control cohort of serum samples collected from 25 individuals with metastatic prostate cancer and 25 healthy age-matched male control individuals. They found a remarkably higher expression of miR-141, as detected by qRT‑PCR, in prostate cancer than in control samples. Moreover, its serum levels were shown to be able to detect individuals with cancer with a high accuracy (60% sensitivity and 100% specificity).
The first evidence of a possible prognostic relevance of miRNAs in prostate cancer was obtained by Tong et al. (2009) in a series of samples from 40 patients who underwent radical prostatectomy. Through the analysis of miRNA expression profiles, they identified 16 miRNAs associated with biochemical recurrence within 2 years of surgery and validated two over-expressed miRNAs (miR-135b and miR-194) by qRT-PCR. More recently, Spahn et al. (in press) analyzed miR-221 expression by qRT-PCR in tumor specimens from 92 patients who underwent radical prostatectomy and showed that miR-221 was down-regulated in samples from patients who experienced clinical recurrence. In particular, miR-221 expression levels were shown to be significant predictors of clinical recurrence in both univariate and multivariate analyses in the presence of PSA levels, pathological stage, and Gleason score as covariates. Finally, Schaefer et al. (2010) evaluated the prognostic significance of a panel of miRNAs, as detected by qRT-PCR, in a series of radical prostatectomy specimens from 75 prostate cancer patients and found that the recurrence‑free interval was significantly decreased in patients with high miR-96 expression in the tumor samples. In addition, Cox regression analysis showed that miR-96 remained a prognostic indicator alone with Gleason score in the prediction model for disease recurrence. This prognostic information was then confirmed in an independent tumor sample set consisting of 79 radical prostatectomy specimens (Schaefer et al., 2010).
miRNA Functional Studies in Prostate Cancer: Towards the Development of miRNA-based Therapies
For a handful of miRNAs, for which profiling studies have provided convincing evidence of a widespread dysregulation in prostate cancer, functional investigation has also been pursued (Figure 1). Specifically, the consequences of the modulation of their expression on the phenotype of experimental models of the disease have been assessed to verify whether these miRNAs could be considered as therapeutic targets or tools. In this context, it has been shown that over-expressed miRNAs can be efficiently silenced by chemically modified antisense oligonucleotides (named antagomirs when conjugated with cholesterol). Conversely, the expression of specific miRNAs that are down-modulated in cancer can be achieved by using synthetic miRNA mimics or expression vectors carrying miRNA genes (Gandellini et al., 2009a).
A putative oncogenic function was proposed for the miR-106b-25 cluster and miR-32 (Ambs et al., 2008), which were shown to directly repress the tumor-suppressors E2F1, p21/WAF1, and Bim in prostate cancer cell lines (Figure 1). Consistently, infection of 22Rv1 cells with a lentivirus encoding the miR-106b-25 cluster significantly inhibited caspase-3/7 activation induced by doxorubicin and etoposide, thus confirming an antiapoptotic role for these miRNAs.
More and more frequently, prostate cancer-related miRNAs were described as tumor suppressors, which is not surprising given that most dysregulated miRNAs in prostate cancer seem to be down-regulated (Gandellini et al., 2009a). Bonci et al. (2008) demonstrated that, in human prostate, the miR-15a-miR-16-1 cluster can control cell survival, proliferation, and invasion by suppressing cyclin D1 and WNT3A (Figure 1). Efficient reconstitution of miR-15a and miR-16-1 expression was achieved by injecting a lentiviral vector expressing the miRNA precursor into LNCaP human prostate cancer xenografts in mice and resulted in marked tumor regression (Bonci et al., 2008). In addition, the delivery of antagomirs specific for miR-15a and miR-16 was able to induce hyperplasia in normal mouse prostate and promote the growth and invasiveness of untransformed RWPE-1 prostate cells, which became tumorigenic in immunodeficient NOD-SCID mice (Bonci et al., 2008). Injection of miR-16 with atelocollagen via the tail vein was also shown to significantly inhibit the growth of prostate tumors in bone, likely due to the suppression of cyclin-dependent kinases CDK1 and CDK2 (Takeshita et al., 2010).
Growth-modulatory functions have also been reported for miR-23b, miR-145, and miR-100, as their ectopic expression in LNCaP cells significantly reduced the proliferative potential (Tong et al., 2009). However, an extensive investigation of the oncogenes targeted by these miRNAs in prostate cancer is still lacking.
It is our opinion that miR-205 presently represents one of the most relevant miRNAs in prostate cancer. We showed that restoration of miR-205 in aggressive prostate cancer cells induced marked morphological changes and cytoskeleton rearrangements, which were compatible with a reverse transition from a mesenchymal to an epithelial state (Gandellini et al., 2009b). In fact, miR-205 up‑regulated E‑cadherin expression (and that of diverse cell‑cell adhesion molecules) and its localization at the membrane and repressed several factors known to be involved in the acquisition of a motile and invasive behavior (i.e., interleukin-6, enhancer of zeste homolog 2, caveolin-1, and metalloproteinase-2). We hypothesized that these events were driven by the concurrent suppression of specific putative miR‑205 targets, such as N-chimerin, ErbB3, E2F1, E2F5, ZEB2, and protein kinase Cε (Figure 1). For the latter, we demonstrated its direct role in regulating epithelial-to-mesenchymal transition. Overall, our findings suggested a crucial role for miR-205 in maintaining the epithelial organization of human prostatic tissue.
It is conceivable that reduction of miR-205 expression levels could represent an oncogenic event that drives the progression towards a cell phenotype with enhanced invasive properties, which presumably favor metastasis. Consistent with this hypothesis, metastasis-derived prostate cancer cell lines express almost undetectable levels of miR-205 and tumors from patients with lymph node dissemination of the disease are characterized by lower miR‑205 expression than node-negative patients (Gandellini et al., 2009b). Given that miR-205 down‑regulation has been reported also in other cancers of epithelial origin, such as breast carcinoma (Wu and Mo, 2009), therapies based on the restoration of this miRNA could be investigated for a variety of tumors for which a tumor-suppressive function of miR‑205 could be documented. However, it is not always possible to generalize this concept.
In some instances, miRNAs described as therapeutic targets or tools in a specific tumor histotype may not be equally valuable for others. For example, miR-21 has been proposed as an oncogenic miRNA as well as a therapeutic target in different human cancers. Strikingly, we observed that miR-21 knockdown is not sufficient per se to affect the proliferative and invasive potential or the chemo- and radiosensitivity profiles of prostate cancer cells (Folini et al., 2010). In addition, miR-21 was not differently expressed in carcinomas and matched normal tissues obtained from patients subjected to radical prostatectomy (Folini et al., 2010). In general, it is our opinion that the potential role of a given miRNA as therapeutic target should always be characterized in the context of the specific disease entity. In addition, the use of experimental models that often do not perfectly recapitulate the features of clinical tumors represents a limiting factor in the definition of the actual role of specific miRNAs in prostate cancer.
Even if endowed with oncogenic or tumor-suppressive roles in prostate cancer cell lines, some miRNAs might not be crucial for the pathogenesis of clinical prostate cancer. For example, oncogenic functions have been described in experimental models for miR-125b and miR-221/222 (Gandellini et al., 2009a), and an over-expression of these miRNAs could be expected in prostate cancer. Surprisingly, several reports have instead indicated that miR-125b and miR‑221/222 are more often down-regulated in prostate carcinomas than in non-neoplastic tissues (Gandellini et al., 2009a). For this reason, an extensive characterization of both the expression and the function of candidate miRNAs in a given disease is warranted before translating these molecules into therapy.
Future Implications for Personalized Prostate Cancer Care
The current understanding of the role exerted by specific miRNAs in the initiation and progression of prostate cancer is still at an early stage, and only scanty evidence that causatively links dysregulation of an miRNA to disease development has been obtained. In addition, a consensus on which miRNAs are specific biomarkers for prostate cancer has not yet been reached owing to only partially overlapping of the results collected thus far, mainly from high-throughput screening in clinical samples. Observed discrepancies may depend on differences in case series examined, selection of tumor specimens (which appears particularly relevant because of the heterogeneous and multifocal nature of the malignancy), RNA isolation protocols, and detection platforms. In this regard, the achievement of a general consensus on the techniques and controls to be used and the conduction of studies on larger case series of paired neoplastic and normal tissues for prostate cancer patients, with associated pathobiological and clinical information, are warranted for the proper validation of specific miRNAs as new prostate cancer biomarkers. However, the evidence that miRNAs are highly stable in formalin-fixed tissues (Li et al., 2007) and that their expression levels can be directly measured not only in prostatectomies but also in core biopsies (Mattie et al., 2006) strongly suggests their potential to contribute to the diagnosis of prostate cancer and, eventually, to classify high‑ and low‑risk patients. In addition, the possibility to detect cancer-related miRNAs in blood (Mitchell et al., 2008) represents the rationale for setting up noninvasive systems for early prostate cancer detection and disease monitoring and could be of potential relevance for promoting active surveillance strategies in patients with low‑risk disease.
A detailed understanding of the functions exerted by specific miRNAs in prostate cancer and the precise identification of their key targets relevant to the disease will be instrumental for the development of miRNA-based therapeutics. Since miRNAs have the potential to modulate a cohort of gene networks, they might become therapeutically relevant in a “one-hit multitarget” context against prostate cancer. The possibility to use miRNA agonists or antagonists, in order to restore or inhibit the function of down‑regulated onco-suppressive miRNAs or up‑regulated oncogenic miRNAs, respectively, has been already successfully demonstrated in experimental tumor models. However, before translating experimental research advances into clinical practice, important issues mainly related to the development of new approaches for the in vivo delivery of miRNA-modulating molecules need to be addressed. In this context, the use of delivery systems displaying carrier‑defined specificity (e.g., cell specific immunoliposomes targeting antigens highly expressed on the prostate cancer cell membrane) might represent a suitable approach for controlled delivery of therapeutic molecules to relevant tissues, thereby avoiding toxic side effects. It is noteworthy that, although there are no ongoing miRNA-based clinical trials in cancer patients, a phase I trial in healthy male volunteers is ongoing (NCT00688012, ClinicalTrial.gov) on a locked nucleic acid oligomer targeting miR-122 (SPC3649), which has been developed as a new therapeutic approach for hepatitis C infection.
In the therapeutic setting, interference with prostate cancer-specific miRNAs could be exploited not only to produce a direct anticancer effect but also to improve the response of tumor cells to conventional anticancer agents, since it has been recently suggested that specific miRNAs may contribute to chemoresistance in different human cancer cell lines and that the modulation of their expression increases cell sensitivity to cytotoxic drugs with different action mechanisms as well as to ionizing radiation. In addition, altered expression of specific miRNAs could provide information about sensitivity or resistance of individual tumors to different treatments before starting therapy (response prediction). Again, changes in the expression of specific miRNAs during therapy might offer a tool for monitoring treatment success (response control).
Although a considerable investment in new research is required before a possible use of miRNAs as novel biomarkers and new therapeutic targets or intervention tools can be formulated, the rapid expansion of the field, together with the promising results obtained in other tumor types, suggest an important impact of these molecules for the management of prostate cancer patients in the future.
(Corresponding author: Nadia Zaffaroni, Ph.D., Department of Experimental Oncology and Molecular Medicine, Fondazione IRCCS Istituto Nazionale Tumori, Via Venezian 1, 20133 Milano, Italy.)
Ambs S, Prueitt RL, Yi M, Hudson RS, Howe TM, Petrocca F, Wallace TA, Liu CG, Volinia S, Calin GA, et al. Genomic profiling of microRNA and messenger RNA reveals deregulated microRNA expression in prostate cancer. Cancer Res 68(15):6162-70, 2008.
Bonci D, Coppola V, Musumeci M, Addario A, Giuffrida R, Memeo L, D’Urso L, Pagliuca A, Biffoni M, Labbaye C, et al. The miR-15a-miR-16-1 cluster controls prostate cancer by targeting multiple oncogenic activities. Nat Med 14(11):1271-7, 2008.
Damber JE, Aus G. Prostate cancer. Lancet 371:1710-21, 2008.
Folini M, Gandellini P, Longoni N, Profumo V, Callari M, Pennati M, Colecchia M, Supino R, Veneroni S, Salvioni R, et al. miR-21: an oncomir on strike in prostate cancer. Mol Cancer 9(1):12, 2010.
Gandellini P, Folini M, Zaffaroni N. Towards the definition of prostate cancer-related microRNAs: where are we now? Trends Mol Med 15(9):381-90, 2009a.
Gandellini P, Folini M, Longoni N, Pennati M, Binda M, Colecchia M, Salvioni R, Supino R, Moretti R, Limonta P, et al. miR-205 Exerts tumor-suppressive functions in human prostate through down-regulation of protein kinase Cepsilon. Cancer Res 69(6):2287-95, 2009b.
Garzon R, Fabbri M, Cimmino A, Calin GA, Croce CM. MicroRNA expression and function in cancer. Trends Mol Med 12(12):580-7, 2006.
Li J, Smyth P, Flavin R, Cahill S, Denning K, Aherne S, Guenther SM, O’Leary JJ, Sheils O. Comparison of miRNA expression patterns using total RNA extracted from matched samples of formalin-fixed paraffin-embedded (FFPE) cells and snap frozen cells. BMC Biotechnol 7:36, 2007.
Lu J, Getz G, Miska EA, Alvarez-Saavedra E, Lamb J, Peck D, Sweet-Cordero A, Ebert BL, Mak RH, Ferrando AA, Downing JR, et al. MicroRNA expression profiles classify human cancers. Nature 435(7043):834-8, 2005.
Mattie MD, Benz CC, Bowers J, Sensinger K, Wong L, Scott GK, Fedele V, Ginzinger D, Getts R, Haqq C. Optimized high-throughput microRNA expression profiling provides novel biomarker assessment of clinical prostate and breast cancer biopsies. Mol Cancer 5:24, 2006.
Mitchell PS, Parkin RK, Kroh EM, Fritz BR, Wyman SK, Pogosova-Agadjanyan EL, Peterson A, Noteboom J, O’Briant KC, Allen A, et al. Circulating microRNAs as stable blood-based markers for cancer detection. Proc Natl Acad Sci U S A 105:10513-8, 2008.
Ozen M, Creighton CJ, Ozdemir M, Ittmann M. Widespread deregulation of microRNA expression in human prostate cancer. Oncogene 27(12):1788-93, 2008.
Porkka KP, Pfeiffer MJ, Waltering KK, Vessella RL, Tammela TL, Visakorpi T. MicroRNA expression profiling in prostate cancer. Cancer Res 67(13):6130-5, 2007.
Prueitt RL, Yi M, Hudson RS, Wallace TA, Howe TM, Yfantis HG, Lee DH, Stephens RM, Liu CG, Calin GA, Croce CM, Ambs S. Expression of microRNAs and protein-coding genes associated with perineural invasion in prostate cancer. Prostate 68:1152-64, 2008.
Schaefer A, Jung M, Mollenkopf HJ, Wagner I, Stephan C, Jentzmik F, Miller K, Lein M, Kristiansen G, Jung K. Diagnostic and prognostic implications of microRNA profiling in prostate carcinoma. Int J Cancer 126(5):1166-76, 2010.
Spahn M, Kneitz S, Scholz CJ, Nico S, Rüdiger T, Ströbel P, Riedmiller H, Kneitz B. Expression of microRNA-221 is progressively reduced in aggressive prostate cancer and metastasis and predicts clinical recurrence. Int J Cancer, in press, doi: 10.1002/ijc.24715.
Takeshita F, Patrawala L, Osaki M, Takahashi RU, Yamamoto Y, Kosaka N, Kawamata M, Kelnar K, Bader AG, Brown D, et al. Systemic delivery of synthetic microRNA-16 inhibits the growth of metastatic prostate tumors via down-regulation of multiple cell-cycle genes. Mol Ther 18(1):181-7, 2010.
Tong AW, Fulgham P, Jay C, Chen P, Khalil I, Liu S, Senzer N, Eklund AC, Han J, Nemunaitis J. MicroRNA profile analysis of human prostate cancers. Cancer Gene Ther 16(3):206-16, 2009.
Volinia S, Calin GA, Liu CG, Ambs S, Cimmino A, Petrocca F, Visone R, Iorio M, Roldo C, Ferracin M, et al. A microRNA expression signature of human solid tumors defines cancer gene targets. Proc Natl Acad Sci U S A 103(7):2257-61, 2006.
Wu H, Mo YY. Targeting miR-205 in breast cancer. Expert Opin Ther Targets 13(12):1439-48, 2009.
[Discovery Medicine, 9(46):212-218, March 2010.]