Overview of MicroRNAs (MiRNAs)
MiRNAs are small non-coding endogenous RNA molecules that vary in length from 18 to 25 nucleotides. There are numerous dysregulated miRNAs that are implicated in the pathogenesis of cancer. MiRNAs play a role in the expression of up to 30% of human genes . MiRNAs can be up- or down-regulated. The up-regulation of oncogenic miRNAs (also known as oncomirs) and the down-regulation of tumour suppressor miRNAs are demonstrated in many malignancies [2-5]. MiRNAs are located within the introns and exons of protein-coding genes and they are also located in the intergenic region. Transcription of the intronic and exonic miRNAs appears to occur in parallel with transcription of a host gene. However, transcription of miRNAs located at the intergenic regions appear to be transcribed independently .
MiRNAs regulate gene expression at both the transcriptional and post-transcriptional level. A single miRNA can have several mRNA targets that are involved in the oncogenic process. However, a single mRNA can be regulated by many different miRNAs that bind to its sites . Dysregulation of miRNAs have been associated with the development of cancer and not surprisingly up to 50% of miRNA genes are located in cancer-related genomic locations .
The biosynthesis of miRNA is a complicated process. The transcription of a miRNA gene begins with initiation of transcription by RNA Polymerase II. This results in the formation of a large molecule, pri-miRNA . The enzyme Drosha and the nuclear protein DGCR8 then cleaves a miRNA precursor (pre-miRNA), which is exported from the nucleus to the cytoplasm by Exportin5 [10, 11]. Drosha and DGCR8 form a large complex, which is known as the microprocessor complex . In the cytoplasm the pre-miRNA is then cleaved by Dicer and an RNA transactivator binding protein into a small dsRNA duplex(miRNA:miRNA*). This duplex molecule is composed of a mature miRNA and a complementary strand (miRNA*). The mature miRNA is subsequently incorporated in a further complex, named RNA-induced silencing complex (RISC). The miRNA* strand is then usually eliminated by cleavage. The mature miRNA molecule is then free to interact with various mRNA targets, resulting in either cleavage or inhibiting protein synthesis .
Expression Profiling of Circulating MiRNA in Patients with Prostate Cancer
MiRNAs are an ideal molecule for a blood-based biomarker for the detection of cancer, as they are dysregulated in carcinogenesis and are highly stable in both tissue and in blood samples [5, 14-16]. Various studies have documented the differential expression of miRNAs in the circulation of patients with cancer when compared with healthy controls, making miRNA an ideal non-invasive biomarker [17-19]. Mitchell et al.  also confirmed that tumour-derived miRNAs are present in the circulation (present in both serum and plasma) at sufficient levels to be used as a suitable biomarker. MiRNAs are resistant to endogenous ribonuclease (RNAse) activity as well as variations in temperature and pH [14, 20].
The exact mechanisms whereby miRNAs are released into the circulation have been debated. One potential theory is that miRNAs simply leak into the circulation via tissue degradation in a passive, energy-independent process. However, there is increasing evidence that miRNAs are actively secreted into the circulation within exosomes and microvesicles and perhaps may even be selected to be transferred to distant cells [19, 21, 22]. MiRNAs are also present within Ago complexes in the circulation. MiRNAs in the circulation contained within microvesicles or within Ago complexes may originate from different cell types and may actually reflect a tissue-specific miRNA expression profile .
There has been wide variability in results when miRNAs from patients' serum has been used to differentiate between those with or without prostate cancer. Mitchell et al.  analysed serum samples from a cohort of 50 individuals, 25 patients with metastatic prostate cancer and 25 male, age-matched controls. This revealed increased expression of miR-100, -125b, -141, -143 and -296 in the serum of the metastatic prostate cancer group. MiR-141 had the greatest differential expression in the prostate cancer group in comparison with the control group. In fact, it had a 46-fold overexpression. Serum levels of miR-141 could detect individuals with prostate cancer with a high accuracy, with 100% specificity and 60% sensitivity with an area under the curve (AUC) of 0.907. This identified a blood-based PCR analytical tool for the detection of prostate cancer. Brase et al.  later identified miR-141 to be upregulated and could be used to differentiate between metastatic and localised prostate cancer. However, Mahn et al.  encountered difficulties with detecting miR-141 in the circulation, this is surprising given that Mitchell et al.  and Brase et al.  found it to be expressed early in prostate cancer sera.
However, Lodes et al.  identified a unique miRNA signature for prostate cancer based on an extraction technique from serum samples. In an analysis comprising of 13 samples (five from patients with prostate cancer, eight from age-matched controls) 15 upregulated miRNAs were identified. Brase et al.  investigated the expression profile of miRNAs in the progression of prostate cancer from organ-confined disease to metastatic disease. After an initial screening study of 21 patients (14 with localised prostate cancer, seven with metastatic prostate cancer), subsequent validation studies were performed with serum samples to investigate the expression profile of five dysregulated miRNAs. MiR-375 was useful to identify those patients with metastatic disease but also a positive lymph node status. MiR-141 was also identified as upregulated in the serum of patients with higher grade tumours.
MiR-21 has previously been identified as an oncomir, being upregulated in several cancers. A study by Zhang et al.  into the expression levels of miR-21 and disease progression reported many interesting results. The serum of 56 patients were included in that study; 20 with localised prostate cancer, 20 with androgen-dependent prostate cancer, 10 with hormone-refractory prostate cancer (HRPC) and six with BPH. Patients with HRPC expressed higher levels of miR-21. Of the patients with HRPC who subsequently received chemotherapy, those who were resistant to chemotherapy had higher levels of miR-21 than those who were sensitive to chemotherapy, but there was no validation of these results in a separate group. Although numbers in the study were small, this identifies the potential of miRNAs as not only a biomarker for diagnosis, but also for disease progression and for response to treatment.
Yaman Agaoglu et al.  investigated the expression profile of three miRNAs; miR-21, -141 and -221 in plasma. In all, 71 patients were included in the study, 51 with prostate cancer and 20 healthy controls. There was a significant upregulation of miR-21 and miR-221. Levels of all three miRNAs were upregulated in the patients with metastatic disease as compared with those with localised prostate cancer, again confirming that miRNAs have the potential to be used as biomarkers for diagnosis and disease progression. Zheng et al.  also identified that miR-221 was upregulated in patients with prostate cancer. In addition, miR-221 was also found to be significantly elevated in patients with androgen-dependent prostate cancer.
Moltzahn et al.  also identified a miRNA signature that could be used to diagnose prostate cancer and also correlate with disease progression in the serum of 48 patients, 36 with prostate cancer and 12 controls. Five miRNAs were identified to be upregulated and four down-regulated. Three miRNAs, miR-93, -106a and -24 were significantly dysregulated in patients with metastatic disease again identifying that there is a miRNA expression profile unique to both localised prostate cancer and also to metastatic prostate cancer.
Heneghan et al.  analysed the expression profile of patients with various malignancies including prostate, colon, breast, renal and melanoma. The study showed a significant down-regulation of the tumour suppressors miR-145 and -155 in prostate cancer. Whilst the down-regulation of these two miRNAs were also seen in several of the malignancies included in the study, a significant upregulation of let-7a in prostate and breast cancer, but an upregulation of miR-195 was exclusive to breast cancer. Using three miRNAs in combination; miR-195, -155 and let-7a could be used a biomarker for breast cancer with a sensitivity of 94%. This process of using a combination of miRNAs to identify a miRNA signature unique to a particular cancer has the potential to be used as a blood-based biomarker across other malignancies.
Mahn et al.  analysed the expression of four miRNAs in 37 patients with localised prostate cancer, eight with metastatic prostate cancer, 18 with BPH and 20 healthy controls. Three of these four miRNAs were significantly upregulated in patients with prostate cancer compared with patients with BPH, miR-26a, -195 and let-7i. MiR-26a could differentiate prostate cancer from BPH with an AUC of 0.703, yet PSA could differentiate between the two clinical conditions with a superior AUC of 0.834. However, in patients with an elevated PSA level, miR-26b could differentiate those with cancer from those with BPH with an AUC of 0.918, this included 37 patients with prostate cancer and seven patients with BPH. These three miRNAs along with the fourth miRNA, miR-32, could be used in combination to further enhance their diagnostic potential with an AUC of 0.758. Of 10 patients who underwent a radical prostatectomy, the expression levels of miR-26a and miR-195 were significantly reduced at postoperative day 7, returning to normal expression levels. This further highlights the diagnostic potential of the overall expression profile of many oncomirs and tumour-suppressor miRNAs. This is not the first time that expression levels of miRNAs have returned to normal for patients undergoing oncological surgery. This has previously been described in breast cancer and patients with colorectal cancer [18, 32].
Whilst miRNAs have enormous potential as biomarkers for prostate cancer, work to elucidate their role as systemic markers of this disease remains in its infancy. To date there are a limited number of studies that have investigated circulating levels of miRNAs in men with prostate cancer, with study numbers varying from 13 to 83 patients. Studies range from comparing localised prostate cancer to controls, and some comparing patients with metastatic prostate cancer with controls, with only four studies having a validation cohort.
Currently RNA is being extracted from a various circulating media, including whole blood, serum and plasma using various different extraction techniques. There is no established consensus on what is the optimum media from which to isolate RNA. Pritchard et al.  have recently identified that a variety of circulating miRNAs are highly expressed in one or more blood cell types. They suggested that acceptable ranges for blood cell counts should be established for miRNAs that are vulnerable to blood cell effects. While most articles to date have focused on free or exosomal miRNAs extracted from either serum or plasma, Heneghan et al.  have highlighted an RNA extraction technique from whole blood using a trizol-based extraction technique that results in a higher yield of miRNAs than serum or plasma samples. The validity of this process was shown by the fact that there was no significant difference in the white cell count, haemoglobin or haematocrit levels between the cancer and control groups of patients in this article [32, 34]. With this lack of consistency about starting material and methodology across results they must be viewed with caution.
Urinary miRNAs are a promising tumour marker in urothelial cancer and miRNAs have also been identified in other bodily fluids, e.g. peritoneal fluid and saliva. This highlights the potential that these molecules have as potential biomarkers across a variety of media [35-37]. However, this is a relatively novel discovery and many further studies are required in this area to elucidate the true potential of urinary miRNAs to act as biomarkers for prostate cancer. Table 1 summarises studies investigating the expression profiles of circulation miRNAs in prostate cancer [14, 24-28, 30, 31, 38].
Table 1. Studies investigating the expression profiles of circulation miRNAs in prostate cancer.
|100, 125b, 141, 143, 296||–||50||25 metastaic PCa vs 25 controls||Yes||Serum||Mitchell et al.  2008|
|16, 92a, 103, 107, 197, 34b, 328, 485-3p, 486-5p, 92b, 574-3p, 636, 640, 766, 885-5p||–||13||5 PCa vs 8 controls||Yes||Serum||Lodes et al.  2009|
|9*, 141, 200b, 375, 516a||–||21||7 metastatic PCa vs 14 PCa||Yes||Serum||Brase et al.  2011|
|Let7a||145, 155||83||20 PCa vs 63 controls||No||Whole blood||Heneghan et al.  2010|
|21||–||56||50 PCa vs 6 BPH||No||Serum||Zhang et al.  2011|
|21, 221||–||71||51 PCa vs 20 controls||No||Plasma||Yaman Agaoglu et al.  2011|
|93, 106a, 874, 1207-5p, 1274a||24, 26b, 30c, 223||48||36 PCa vs 12 controls||No||Serum||Moltzahn et al.  2011|
|221||–||43||28 PCa vs 20 controls||No||Plasma||Zheng et al.  2011|
|26a, 195, let7i||–||83||45 PCa vs 38 controls||No||Serum||Mahn et al.  2011|
|141, 298, 346, 375|| ||50||25 metastatic PCa vs 25 controls||Yes||Serum||Selth et al. . 2011|
MiRNAs dysregulated in the circulation of men with prostate cancer are a potential biomarker for the diagnosis, prognosis and for monitoring response to therapy of prostate cancer. However, given the few studies, the lack of validation and the disparity of results, much larger scale clinical studies are necessary to determine if circulating miRNAs have the potential to act as biomarkers for prostate cancer in clinical practice.