Abstract
OBJECTIVE Numerous microRNAs (miRNAs) are deregulated in human cancers. The experimental evidence supports that miRNAs plays a role in the initiation and progression of human malignancies. The present study was undertaken to evaluate the differential expression of 6 miRNAs as biomarker for early detection of prostate cancer, and then to determine whether the expression profiling of these miRNAs could predict the prognosis of prostate cancer.
METHODS The expression profilings of these 6 miRNAs were investigated using the method of locked nucleic acid (LNA)-modified oligonucleotide in situ hybridization (ISH). And the technology of tissue microarray (TMA) was employed using the formalin-fixed, paraffin-embedd (FFPE) specimens taken from 52 patients with prostate carcinoma (PCa) and 38 patients with benign prostatic hyperplasia (BPH).
RESULTS The rates of positive expression for 6 miRNAs (miR-15b, miR-16, let-7g, miR-96, miR-182 and miR-183) were 26.92%, 15.38%, 15.38%, 67.31%, 61.54% and 71.15% in the specimens of prostate cancer, and 57.89%, 76.32%, 68.42%, 44.74%, 31.58%, 47.37% in the tissues of benign prostatic hyperplasia, respectively. The expressions of all 6 miRNAs between the prostate cancer and benign prostatic hyperplasia tissues were significantly different (P < 0.05). The positive rate of these 6 miRNAs was significantly related to the Gleason Grading of prostate cancer (P < 0.01). There was no significant correlation between the expression of these miRNAs and age and the concentration of serum PSA of the patient (P > 0.05). We also found that the expression of miR-15b, miR-96 and miR-182 correlated with clinical stages of tumor (P < 0.05). The expression of miR-96 correlated with lobus prostatae of tumor invasion (P < 0.01), but the expressions of the remaining five miRNAs were not correlated with that (P > 0.05). In addition, the expression of miR-15b was negatively related to that of miR-96, miR-182 and miR-183, respectively (P < 0.01, r < 0.00). There was a positive correlation among the expressions of miR-96, miR-182 and miR-183 in prostate cancer (P < 0.01, r > 0.00). The expression of miR-16 was positively related to that of miR-let-7g (P < 0.01, r > 0.00).
CONCLUSION The results suggest that miRNA expression profiling could have relevance to the biological and clinical behavior of prostate cancer, and they might be important biomarkers for early detection and prognostic assessment of prostate cancer.
keywords
Introduction
Prostate cancer is one of the most common malignant diseases in western countries, and its incidence has always being increased in the past few years in China. Early diagnosis and therapy of prostate carcinoma contribute to the good prognosis of patients. Although some theories have been proposed to explain the development of prostate cancer[1,2], the exact mechanism is still uncertain. MicroRNAs (miRNAs) are a class of naturally occurring small noncoding RNA with 18 to 24 nucleotides in length that negatively regulate expression of protein-coding genes at the posttranscriptional level through binding to complementary target mRNAs. These small molecules have a great impact on our understanding of gene regulation mechanisms because miRNAs play a pivotal role in cell growth, differentiation, apoptosis and carcinogenesis[3-6]. According to the accumulating evidence, the aberrant expression of numerous miRNAs in human cancers are involved in the development and progression of cancer. Many miRNAs have been identified in various types of tumor, which show that different sets of miRNAs are usually deregulated in different cancers and they may act as oncogenes or tumor suppressor genes. A better understanding of the molecular mechanism of miRNAs in prostate cancer may lead to new ideas for diagnosis and therapies for PCa. The purpose of this study was to examine the expression of 6 miRNAs in benign prostate hyperplasia and prostate cancer and further evaluate their clinically pathological significance and potential value in early diagnosis of PCa. Correlations among the expressions of 6 miRNAs in the tissues of prostate cancer were also examined and discussed in this study.
Materials and Methods
Clinical materials
Fifty two patients with prostate cancer and 38 patients with benign prostate hyperplasia selected in the study from 2001 to 2005 had underwent the TURP (transurethral resection, prostate) at the Department of Urinary Surgery, the First Affiliated Hospital of Anhui Medical University, China. The formalin-fixed, paraffin-embedded (FFPE) specimens taken from the patients were obtained from the Department of Pathology, Anhui Medical University. The median age of the patients was 71 years old (ranging from 55 to 85). Thirty eight pieces of FFPE tissues taken from the patients with benign prostate hyperplasia were used as controls. All FFPE tissues of BPH and PCa were available for series section, H&E staining and reexamination. The clinicopathological data of 52 prostate cancer patients are shown in Table 1.
Clinicopathological data of prostate cancer patients.
The tumor types of 52 cases of prostate cancer are all adenocarcinoma. We performed clinical stage of the tumors according to Jewett-Whitmore-Prout staging system.
Main reagents
Digoxigenin-labeled oligonucleotide probes for 6 miRNAs were custom-synthesized by Integrated DNA Technologies (Coralville, IA) for in situ hybridization. All of the miRNAs and their corresponding oligonucleotide probes are described in Table 2.
miRNAs,chromosomal location and probe sequences.
In situ hybridization reagents were purchased from BOSTER Co. (Wuhan, China), Sigma Co. (USA) and Zhong Shan Co. (Beijing, China).
Preparation of tissue microarrays (TMAs)
All FFPE specimens were sectioned into 4 μm in thickness and stained with H&E for identification of BPH or PCa tissues. Tissues used in the arrays were verified by a board of certified pathologists who marked the areas of BPH or PCa of each tissue. Then we used a manual tissue puncher (D = 1.5 mm) to collect the marked areas of the tissues from paraffin blocks and put them in the receptor block holders to make tissue chips using the array maker. The newly arrayed paraffin blocks were melted in a oven at 55°C for 5~10 min so as to hold the tissue cores together. Multiple and consecutive thin sections were cut from the receptor blocks and layered on a glass slide, and then dried in the oven at 90°C. The sections were hybridized in situ with digoxigenin-labeled miRNA probes. We used 52 specimens of prostate cancer and selected 3 areas of each type of morphology to make a tissue arrays at the tissue spots of 51, 51, 54 on 3 glass slides, respectively (52 × 3 = 156 punches). As compared, we used 38 specimens of benign prostatic hyperplasia and selected 2 areas of each type of morphology to create a tissue chip at the tissue spots of 52, 24 on 2 glass slides, respectively (38 × 2 = 76 punches).
Staining procedure of in situ hybridization and evaluation of staining
In situ hybridization is an important tool for analyzing gene expression and this method has been used for many years in our laboratory. The tissue array sections were treated with 3% hydrogen peroxide and 10% pepsin (diluted with 3% citric acid), respectively after deparaffinization and rehydration. The sections were pre-hybridized at 37°C for 4 h with a prehybridization solution (Boster Co., China). Next, the sections were incubated in 100 μl hybridization solution/section containing 1 μl denatured probe and 400 μl dilution of oligonucleotide probe (Boster Co., China) at 43°C for 16 h~20 h. The slides were washed at 37°C in 2 × SSC (5 min, 3 times), 0.5 × SSC (5 min, 3 times) and 0.2 × SSC (5 min, 3 times), respectively. In the procedure of color reaction, the slides were incubated in sheep serum at 37°C for 30 min and then incubated with a mouse anti-digoxigenin antibody at 37°C for 60 min. After washed with PBS, the color was developed in DAB (Zhong Shan Co., China) for 15 min~30 min identified by occasional observation. Then the counterstaining of slides was conducted with hematoxylin followed by a sealing procedure with neural gum.
As a negative control, sections were incubated with dilution of oligonucleotide probe instead of the probe solution. In addition, the quality and specificity of the miRNA probe were further verified using RNase-treated tissue sections as a negative control. The positive control of tissue was developed with known positive sections. A 5S-miRNA was transcribed from a kind of house-keeping genes and the corresponding probe was synthesized (the sequence is 5’-TTA GCT TCC GAG ATC AGA CG-3’), so for in situ hybridization of these 6 miRNAs, the 5S probe was used as the positive control of probe. The current evidence suggested that mature miRNAs principally located in cytoplasm, so positive miRNA was recognized as intensely stained in the cytoplasm of tumor cells. Labeling indexes were calculated the percentages of positive cells among more than 500 cells counted with a 5 high-power fields using 400 magnifications. The means that we chosen fields in TMA are shown in Fig. 1.
Schematic representation of the selection of field of microscope in TMA.
Scoring of miRNA-ISH staining was categorized as follows: negative cells stained with buffy particles in the cytoplasm accounted for less than 10%; positive cells stained with (what) accounted for more than 10%. The mature miRNAs with light stain were very little in cytoplasm and sometimes it was difficult to be detected, as a result, we considered the weakly positive signal as a positive signal. The results were evaluated according to the staining intensity of all tissue spots of a specimen on tissue microarrays (not including the exutive tissue spots) and staining results of all specimens were calculated by two pathologists for confirmation. The study was performed single-blindly, therefore, the investigators performed the in situ hybridization analyses without knowing the patients and clinical characteristics.
Statistical analysis
With SPSS 13.0 statistical software, the results were analyzed by chi-square (χ2 ) test and the method of Fisher’s exact probability. The relationships among the expressions of these 6 miRNAs were analyzed according to Spearman analysis. Significance was accepted at a value of P < 0.05.
Results
The expressions of 6 miRNAs in benign prostatic hyperplasia and prostate cancer
The analysis results of 52 samples of prostate cancer showed that 6 miRNAs (miR-15b, miR-16, let-7g, miR-96, miR-182 and miR-183) were positive as follows: 14 (26.92%), 8 (15.38%), 8 (15.38%), 35 (67.31%), 32 (61.54%) and 37 (71.15%), respectively, and regarding the 38 BPH tissues, the 6 miRNAs were positive as follows: 22 (57.89%), 29 (76.32%), 26 (68.42%), 17 (44.74%), 12 (31.58%) and 18 (47.37%), respectively. The expression pattern of miRNAs in cells was cytoplasmic buffy staining. In our research, staining of these 6 miRNAs was all detected mainly in the cytoplasm, and the staining was found in the nucleus of only a few tumor cells (Fig. 2). The expressions of all 6 miRNAs between the prostate cancer and BPH tissues showed a significant difference (P < 0.05, Table 3).
The expressions of 6 miRNAs in prostate cancer and benign prostatic hyperplasia.
Correlation between the expression of 6 miRNAs and clinicopathologic features
The relationship between the expressions of these miRNAs and the clinicopathologic factors of prostate cancer is summarized in Table 4A and 4B. The positive rate of the 6 miRNAs was significantly related to the Gleason Grading of prostate cancer (P < 0.01), and the positive rate of miR-15b, miR-16 and miR-let-7g were decreased with an increasing Gleason Grading of prostate cancer (P < 0.01), however, the expression of miR-96, miR-182 and miR-183 were increased with an increasing Gleason Grading (P < 0.01). The results of the study indicated that the expression of miR-15b was decreased as the clinical stage of tumor was increased (P < 0.05) and the positive rate of miR-96, miR-182 were increased with an increasing clinical stage of prostate cancer (P < 0.05). We also found that the expression of miR-96 correlated with lobus prostatae of tumor invasion (P < 0.01). There were no significant differences between the positive rate of the 6 miRNAs expression and age and the concentration of serum PSA of the patients (P > 0.05). These results implied that there was a positive correlation between the expression of miR-96, miR-182, miR-183 and the degree of malignancy of PCa (P < 0.05). With regard to miR-15b, miR-16 and let-7g, the findings implied that there was a negative correlation between their expressions and the malignant degree of PCa (P < 0.05).
Relationship between miR-15b, miR-16 and let-7g expressions and clinicopathologic factors.
Relationship between miR-96, miR-182 and miR-183 expressions and clinicopathologic factors.
Correlations among the expressions of 6 miRNAs in prostate cancer tissue
The relationships among the expressions of these miRNAs were analyzed on the basis of expressions only in the cells of prostate cancer. We found that the expression of miR-15b was negatively related to that of miR-96, miR-182 and miR-183, respectively (P < 0.01, r < 0.00). There was a positive correlation among expressions of miR-96, miR-182 and miR-183 in PCa (P < 0.01, r > 0.00) and the expression of miR-16 was positively related to that of let-7g (P < 0.01, r > 0.00). No correlation between the expressions of the other miRNAs could be established. Statistical results are shown in Table 5.
Correlations among 6 miRNAs expression in prostate cancer.
Discussion
Prostate cancer is one of the most common cancers in western countries. The incidence rate of prostate cancer in Chinese male has been increased. Prostate specific antigen (PSA) and other biomarkers have been known as valuable tools for early detection and prognosis of prostate cancer, but their sensitivity and specificity are not very satisfied in clinical practice. MicroRNAs (miRNAs) are a recently discovered class of small noncoding RNAs. They have showed the ability to control fundamental cellular processes, such as differentiation of cells and timing of development of the organism[7], and the target genes of microRNA in most of mammalian are associated with transcription, signal transduction and tumorigenesis[8]. In addition, more than half of the human miRNAs genes are located in cancer-associated genomic regions or in fragile sites[9]. These results suggest that aberrations of miRNAs could be involved in various human diseases. Lu demonstrated that the miRNA expression profiles are surprisingly informative, reflecting the developmental lineage and differentiation state of the tumors[10], while the abrogation of differentiation is a hallmark of all human cancers. They also successfully classified the poorly differentiated tumors using expression profiles of miRNA. Liu further demonstrated that the expression of miRNAs is highly tissue specific[11]. Recently, miRNA expression profiling of prostate cancer and BPH samples was carried out and the results suggested that miRNA might act as signature specific for prostate cancer and hierarchical clustering of the prostate cancer samples by their miRNA expression and accurately separate the carcinomas from the BPH samples[12]. All the evidence suggests that miRNAs may play a more important role in the pathogenesis of a limited range of human cancers, including prostate cancer. Detection and functional analyses of miRNAs will provide us not only the biomarkers for early detection of prostate cancer but also increase our understanding of prostate carcinogenesis.
Recently, the expressions of some miRNAs selected from peripheral zone in benign tumor were analysed. The expression of prostate cancer tissues was analysed with Ozen’s group[13]. The results of the analyses indicated the widespread downregulation of miRNAs in prostate cancer tissues. The downregulated miRNAs included several miRNAs with proven target, whose proteins have been previously shown to be increased in prostate cancer, including RAS, E2F3. In addition, Shi et al.[14] found that miRNA (miR-125b) was overexpressed in most clinical samples of prostate cancer and the results from cultured cell lines of prostate cancer suggested BAK1 mRNA might be a direct target of miR-125b and the HER2-AR (androgen receptor) signal pathway might be involved in the up-regulation of miR-125b. These findings indicated that the mechanism of the aberrant expression of miRNAs might be extremely complicated and the expression profiles of miRNAs in cancer are protean. Different methods might give rise to varied results in different tumors. Although the precise mechanism hasn’t been fully understood, miRNAs seem to be a crucial factors for prostate oncogenesis.
In our study, the expressions of 6 miRNAs in 90 clinical samples of prostate tissue (52 prostate carcinomas and 38 BPH) were analysed to screen the miRNA signature in prostate cancer in order to use these potential biomarkers for early diagnosis or prediction of prognosis of prostate cancer. Over the past years, a number of different approaches to evaluate expressions of miRNAs have been described, however, there have been few reports of in situ hybridization for human miRNAs in cancers. So in the present study, in situ hybridization technique was selected as method for detection because this method can exactly localize the positive signal of miRNAs in cellular level. Furthermore, we applied archival, FFPE prostate tissues because miRNAs can survive in the condition of formalin fixation and paraffin embedding. The FFPE, archival tissues could be obtained easily.
The 6 miRNAs we chose to analyze in this study had been described their altered in colorectal cancer, lung cancer and breast cancer[10]. In our study, among the differentially expressed miRNAs, miR-15b, miR-16 and miR-let-7g, which are all located at chromosome 3, were down-regulated in prostate cancer specimens as compared with those in BPH specimens. The remaining three, of miR-96, miR-182 and miR-183 localized in the same chromosomal region, 7q32.2, were up-regulated in prostate cancer, which suggested that these 6 miRNAs may potentially act as tumor suppressor genes or oncogenes in prostate carcinogenesis, respectively. According to the deregulation of miRNAs in prostate cancer and their occurring in cluster in chromosome, we also concluded that there might be some cancer-associated regions in chromosome 3 and 7.
Recently, miR-15b and miR-16 were found to be downregulated in human gastric cancer cells (SGC7901/VCR)[15]. The first evidence for miRNA involvement in human cancer came from a study by Calin et al.[16], their research found that there was downregulation of miR-15a and miR-16-1 in CLL samples and suggested a role of miR-15a and miR-16-1 as tumor suppressor genes. Subsequent investigations have shown that the putative target of these two miRNAs was oncogene BCL2 in the leukemic cell line MEG-01[17]. Although the two groups of miRNAs mentioned above were located at different chromosome, they might have the same target, such as BCL2, and this suggested that different miRNAs could play a certain role in the development of cancer through targeting the same gene.
Emerging evidence suggested that miRNA let-7 may control lung cancer development, or at least play a critical role in the pathogenesis of lung cancer. Takamizawa et al.[18] observed that the expression levels of let-7 were frequently reduced in both invisible spectro and a living body with lung cancer enrolled in the studies; more importantly, low levels of let-7 were significantly associated with shortened postoperative survival, independent of disease stage. Further research demonstrated that let-7 negatively regulates the expression of RAS and MYC by targeting their mRNAs for translation repression[19]. Both RAS and MYC have been implicated as key oncogenes in lung cancer. MiRNA let-7g is a member of the let-7 miRNA family, and its location (3p21.1-21.2) is in the minimal deleted regions which is involved in human cancers, such as lung cancer and breast cancer. We found the reduced expression of let-7g in prostate cancer compared with that in BPH. All these evidences suggest that let-7g regulation of some oncogenes as a mechanism of prostate oncogenesis.
In our study, the up-regulated miRNAs in prostate cancer samples including miR-96, miR-182 and miR-183 are located in the same chromosomal region, 7q32.2. The up-regulation of miR-182 was clearly observed in the cell lines of colorectal cancer and CHES1 protein was identified as a potential target of both miR-96 and miR-182[20]. CHES 1 is a member of the forkhead family of transcription factors that repress genes involved in apoptosis. Other members of this family, including FOXF2, FOXK2, FOXO1A, FOXO3A and FOXQ1, are also found as putative targets of miR-182, miR-183 and miR-96. It may be expected that gene targets of miR-182, miR-183 and miR-96 belong to the class of tumor suppressor genes. According to some published findings and the results in our study, we concluded that many tumor suppressor genes might be the direct targets of miR-96, miR-182, miR-183, however, we have not known other miRNAs about their effects in the development and progression of prostate cancer.
These 6 miRNAs have been identified in various types of tumor including prostate cancer, which show that different sets of miRNAs are usually deregulated in different cancers. According to the current view, expression profiling of some miRNAs, not expression level of one miRNA, shows the high accuracy in classifying some human tumors and reflecting the developmental lineage and differentiation state of the tumors, which suggest that the expression profiles of some miRNA could be specific for some tumors and might act as biomarker for early detection of some cancers[10]. The results in our study indicated that some miRNAs including miR-15b, miR-16, let-7g and miR-96, miR-182, miR-183 aberrantly expressed in prostate cancer samples revealed the miRNA signature of prostate cancer. So the expression profiling of these 6 miRNAs might be valuable for early diagnosis of prostate cancer. It has been reported that the miRNA signatures of different types of cancer could share some individual miRNAs[21]. Some differentially expressed miRNAs detected in this study have shown to be deregulated in other cancers, and this suggests that prostate cancer has a miRNA signature which is relatively specific for this type of cancer.
By analysing the relationship between the positive expressions of these 6 miRNAs and clinicopathologic characteristics, we revealed that there is a significant association between the expression of these miRNAs and Gleason Grading of prostate cancer (P < 0.01). The more advanced the Gleason grading is, the more the miR-96, miR-182 and miR-183 expressed, but less expression of miR-15, miR-16 and let-7g. Gleason Grading is a means widely used for prediction of prognosis of prostate cancer, so our report demonstrated that expression profiles of these miRNAs might reflect prognosis of prostate cancer and overexpression of miR-96, miR-182 and miR-183 may promote progression of prostate cancer. In our present study, we found that the expression of miR-15b was decreased as clinical stage of tumor was increased (P < 0.05) and the positive rate of miR-96, miR-182 was increased with an increasing clinical stage of prostate cancer (P < 0.05). These indicated that there is a relationship between the expressions of miR-15b, miR-96, miR-182 and tumor progression and metastasis. With regard to miR-96, we also found that the expression of miR-96 correlated with lobus prostatae of tumor invasion (P < 0.01). We concluded that the expression profiling of these 6 miRNAs might be informative in reflecting the degree of malignancy of prostate cancer and act as potential biomarkers for prognostic assessment of tumor, especially the expression of miR-96, which is regarded to have correlation with many clinicopathologic parameters of PCa.
miR-96, miR-182 and miR-183 were all up-regulated and they are located in the same chromosomal region, 7q32.2. Our research demonstrated that there was a positive correlation among expressions of miR-96, miR-182 and miR-183 in PCa (P < 0.01, r > 0.00, Table 5). The results suggested that miR-96, miR-182 and miR-183 occur in cluster and might play an incorporated role in oncogenesis, similar to a microRNA polycistron. Overexpression of this cluster may promote the development of prostate cancer. Although there was no significant relationship among the expressions miR-15b, miR-16 and let-7g which are all located at chromosome 3, we found that the expression of miR-15b was negatively related to those of miR-96, miR-182 and miR-183, respectively (P < 0.01, r < 0.00, Table 5). We concluded that miR-15b might act as tumor suppressor and interact with potential oncogenes (miR-96, miR-182 and miR-183) in the development and progression of prostate cancer. Furthermore, the expression of miR-16 was positively related to that of let-7g in PCa (P < 0.01, r > 0.00, Table 5). This result suggested that miR-16 and let-7g might play a role as a cluster in the tumorigenesis of tumor. These results stated above demonstrated that these 6 miRNAs and possibly many other miRNAs might synergistically play an important role in oncogenesis of prostate cancer, and expression profiling of miRNAs might be important biomarkers for early detection and prognostic assessment of prostate cancer.
Research is currently under way to define the exact mechanism of miRNAs in prostatic neoplastic progression. Although our group did not investigate the mechanism of these miRNAs in regulating the cell growth, apoptosis and transformation of prostate cancer and identify the potential targets of these miRNAs, we screened some miRNAs that differentially expressed in prostate cancer and identified them as biomarkers for early diagnosis or prediction of prognosis of prostate cancer. Extensive studies are now in progress aimed to reveal the relationship between the expression of miRNAs and development of PCa so that further move the basic research of miRNAs to the field of cancer biotechnology.
- Revision received November 3, 2008.
- Accepted January 12, 2009.
- Copyright © 2009 by Tianjin Medical University Cancer Institute & Hospital and Springer
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.