Table of Contents  
Year : 2021  |  Volume : 23  |  Issue : 2  |  Page : 188-196

TGF-β1-regulated miR-3691-3p targets E2F3 and PRDM1 to inhibit prostate cancer progression

1 Department of Pathology, School of Biology and Basic Medical Sciences, Soochow University, Suzhou 215123, China Department of Pathology, Tongren Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai 200336, China Collaborative Innovation Center of Clinical Immunology between Soochow University and Sihong People's Hospital, Sihong 223900, China Department of Pathology, Sihong People's Hospital, Sihong 223900, China Laboratory Animal Research Center, Soochow University School of Medicine, Suzhou 215123, China Department of Surgery, The First Affiliated Hospital of Soochow University, Suzhou 215006, China Department of Pathology, The Second Affiliated Hospital of Soochow University, Suzhou 215004, China Suzhou Key Laboratory of Tumor Microenvironment and Pathology, Soochow University, Suzhou 215006, China

Date of Submission25-Dec-2019
Date of Acceptance07-Aug-2020
Date of Web Publication06-Nov-2020

Correspondence Address:
Shou-Li Wang
Department of Pathology, School of Biology and Basic Medical Sciences, Soochow University, Suzhou 215123; Suzhou Key Laboratory of Tumor Microenvironment and Pathology, Soochow University, Suzhou 215006
Yong-Sheng Zhang
Department of Pathology, The Second Affiliated Hospital of Soochow University, Suzhou 215004
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/aja.aja_60_20

Rights and Permissions

Transforming growth factor-β1 (TGF-β1) acts as a tumor promoter in advanced prostate cancer (PCa). We speculated that microRNAs (miRNAs) that are inhibited by TGF-β1 might exert anti-tumor effects. To assess this, we identified several miRNAs downregulated by TGF-β1 in PCa cell lines and selected miR-3691-3p for detailed analysis as a candidate anti-oncogene miRNA. miR-3691-3p was expressed at significantly lower levels in human PCa tissue compared with paired benign prostatic hyperplasia tissue, and its expression level correlated inversely with aggressive clinical pathological features. Overexpression of miR-3691-3p in PCa cell lines inhibited proliferation, migration, and invasion, and promoted apoptosis. The miR-3691-3p target genes E2F transcription factor 3 (E2F3) and PR domain containing 1, with ZNF domain (PRDM1) were upregulated in miR-3691-3p-overexpressing PCa cells, and silencing of E2F3 or PRDM1 suppressed PCa cell proliferation, migration, and invasion. Treatment of mice bearing PCa xenografts with a miR-3691-3p agomir inhibited tumor growth and promoted tumor cell apoptosis. Consistent with the negative regulation of E2F3 and PRDM1 by miR-3691-3p, both proteins were overexpressed in clinical PCa specimens compared with noncancerous prostate tissue. Our results indicate that TGF-β1-regulated miR-3691-3p acts as an anti-oncogene in PCa by downregulating E2F3 and PRDM1. These results provide novel insights into the mechanisms by which TGF-β1 contributes to the progression of PCa.

Keywords: E2F transcription factor 3; miR-3691-3p; PR domain containing 1, with ZNF domain; prostate cancer; transforming growth factor-β1

How to cite this article:
Hu YM, Lou XL, Liu BZ, Sun L, Wan S, Wu L, Zhao X, Zhou Q, Sun MM, Tao K, Zhang YS, Wang SL. TGF-β1-regulated miR-3691-3p targets E2F3 and PRDM1 to inhibit prostate cancer progression. Asian J Androl 2021;23:188-96

How to cite this URL:
Hu YM, Lou XL, Liu BZ, Sun L, Wan S, Wu L, Zhao X, Zhou Q, Sun MM, Tao K, Zhang YS, Wang SL. TGF-β1-regulated miR-3691-3p targets E2F3 and PRDM1 to inhibit prostate cancer progression. Asian J Androl [serial online] 2021 [cited 2022 Oct 7];23:188-96. Available from:

Yue-Mei Hu, Xiao-Li Lou
These authors contributed equally to this work.

  Introduction Top

Prostate cancer (PCa) is a clinically heterogeneous multifocal and highly aggressive disease.[1] Although early stage PCa is clinically manageable, the evolution of PCa to a hormone-independent disease is invariably associated with advanced metastasis, which limits the therapeutic options.[2] Combinations of methods are currently employed to aid in the diagnosis and prognosis of PCa, including serum prostate-specific antigen (PSA) levels and the Gleason score.[3],[4] However, these methods carry some limitations, such as frequent underestimation of the Gleason score and the fact that PSA levels can be elevated in conditions other than PCa.[5],[6] Thus, there is a crucial need to improve our understanding of the molecular mechanisms underlying PCa to assist in the development of novel diagnostic and prognostic markers and treatment strategies.

MicroRNAs (miRNA) are small (approximately 20 nucleotides) noncoding RNAs that play important roles in cell physiology by regulating mRNA expression and stability. Many miRNAs have been shown to function as classical oncogenes or tumor suppressors[7] and are deregulated in various cancer subtypes,[8],[9],[10] including PCa.[11] For example, miR-1 and miR-31, which are downregulated in PCa, are considered to function as tumor suppressors through regulation of Notch3 and cyclin dependent kinase 1 (CDK1) mRNAs.[12] Understanding the roles of the downregulated miRNAs and their target genes in PCa progression could identify new options for the therapeutic use of miRNA mimics (agomirs) or inhibitors (antagomirs) in PCa. Indeed, we previously showed that inhibition of the expression of miR-450b-5p by the cytokine transforming growth factor-β1 (TGF-β1) reverses the differentiation of rhabdomyosarcoma via effects on the miRNA target gene.[13]

In addition to rhabdomyosarcoma, TGF-β1 plays an essential role in promoting several adenocarcinomas, including PCa,[14] colorectal cancer,[15] pancreatic ductal adenocarcinoma,[16] and lung adenocarcinoma.[17] Moreover, the progression of PCa to an androgen-independent state is accompanied by changes in the effects of growth factor signaling pathways,[18] including an increase in TGF-β1 production that promotes metastasis.[19],[20] TGF-β1 has been shown to modulate the expression of several miRNAs, such asmiR-224, miR-15a and miR-16,[21],[22] in androgen-independent PCa. Therefore, we hypothesized that overexpression of TGF-β1-downregulated miRNAs might reverse the malignant phenotype in PCa, similar to the effects of overexpressing miR-450b-5p in rhabdomyosarcoma[13] and miR-196a-3p in breast cancer.[23] According to large amounts of literature reports, TGF-β1 plays essential role in the regulation of different adenocarcinomas, including procreate cancer,[14] colorectal cancer,[15] pancreatic ductal adenocarcinoma[16] and lung adenocarcinoma.[17]

In the present study, we first screened a dataset of miRNAs differentially expressed in TGF-β1-deficient and wild-type human colorectal adenocarcinoma cells (GSE53337), and we identified five miRNAs that were significantly differentially regulated by TGF-β1 (miR-4723-3p, miR-324-3p, miR-4313, miR-196a-3p, and miR-3691-3p). We then examined their expression in TGF-β1-treated human PCa cell lines and clinical specimens; identified miR-196a-3p as a TGF-β1-inhibited miRNA in PCa; and investigated the expression and function of miR-196a-3p as a potential tumor suppressor using in vitro assays, a mouse xenograft model, and human clinical PCa specimens.

  Materials and Methods Top

Cell lines and cell culture

The human prostate carcinoma cell lines PC-3 and DU145 were obtained from the Shanghai Cell Bank of the Chinese Academy of Sciences (Shanghai, China) and were maintained in RPMI 1640 basic medium or Dulbecco's Modified Eagle Medium supplemented with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin. All cell culture reagents were purchased from Gibco (Burlington, ONT, Canada). The cell lines were maintained in a humidified 5% CO2 incubator at 37°C.

Clinical samples

All pairs of primary PCa and benign prostatic hyperplasia (BPH) tissue samples were collected from patients seen at the Second Affiliated Hospital of Soochow University (Suzhou, China), from January, 2001 to December, 2011. The boundary between the adjacent noncancerous and cancerous tissues was at least 1.5 cm, and the identities of both tissue types were verified by pathologists. Tissues were sliced into 10-μm-thick sections using a cryostat microtome, placed in 1.5 ml microtubes (Corning, Tewksbury, MA, USA), and stored at −80°C until analysis. Signed informed consent was obtained from all patients and the study was approved by the Clinical Research Ethics Committee of the Second Affiliated Hospital of Soochow Hospital (JD-LK-2019-076-01).

Oligonucleotide synthesis and transfection

For overexpression studies, miR-3691-3p mimic and a control mimic sequence were purchased from RiboBio (Guangzhou, China). PC-3 and DU145 cells were seeded into 6-well plates at 105 cells per well in medium without antibiotics, incubated overnight, and transfected with mimics (100 nmol l−1 final concentration) using Lipofectamine™ 2000 (Thermo Fisher Scientific, Carlsbad, CA, USA) according to the manufacturers' protocol. After 24 h or 48 h, the cells were harvested, washed, and used for experiments. For RNA interference studies, small interfering RNAs (siRNAs) specific for the miR-3691-3p target genes E2F transcription factor 3 (E2F3) and PR domain containing 1, with ZNF domain (PRDM1) were designed using BLOCK-iT RNAi Designer (Invitrogen, San Diego, CA, USA) and synthesized by RiboBio. After confirmation of the specificity and efficacy of target gene knockdown by western blot analysis, three siRNAs were selected for the experiments: E2F3 siRNA, GCACTACGAAGTCCAGATA; PRDM1 siRNA, GGACCTCGATGACTTTAGA; and control scrambled siRNA, UUUTGATCAUTGATGAAA. PCa cells were transfected with siRNAs (30 nmol l−1) using the same method as for miRNA mimics.

Reverse transcription-quantitative polymerase chain reaction (RT-qPCR)

RNA was extracted from patient tissues samples or PCa cell lines using TRIzol (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's protocol. Primers for amplification of mRNAs were designed and synthesized by GeneCopoeia (Guangzhou, China). RNA was reverse transcribed using the RevertAid™First Strand cDNA Synthesis Kit (Thermo Fisher Scientific). RT-qPCR reactions were performed on an ABI PCR 7500 Real-Time System (Applied Biosystems, Foster City, CA, USA) with U6 as an internal control. For miRNA amplification, cDNA was synthesized with a miRNA-specific stem-loop primer and qPCR was performed with the following primers: U6 forward 5'-GCTTCGGCAGCACATATACTAAAAT-3', U6 reverse: 5'-CGCTTCACGAATTTGCGTGTCAT-3'; Has-miR-3691-3p 5'-GGCACCAAGTCTGCGTCAT-3'; Has-miR-4313 5'-GAAAGCCCCCTGGCCC-3'; Has-miR-196a-3p 5'-GGAACGGCAACAAGAAACT-3'; Has-miR-324-3p 5'-GAAACTGCCCCAGGTGC-3'; and Has-miR-4723-3p 5'-GAAACCCTCTCTGGCTCCTC-3'. RT-qPCR cycling conditions were 10 min at 95°C followed by 40 cycles of 95°C for 15 s and 60°C for 60 s. Relative quantification of miRNA levels was performed using the comparative cycle threshold (Ct) method.

Cell viability assay

Cell viability was measured at 1 day, 3 day, and 5 days using the 3-(4,5-dimethylthiazol-2-yl) 2, 5-diphenyltetrazolium bromide (MTT) assay. Briefly, PCa cells were seeded into 96-well plates at 104 cells per well and incubated at 37°C for the indicated times. Aliquots of 20 μl per well of 5 mg ml−1 MTT (Sigma-Aldrich, St. Louis, MO, USA) were added to the cells and the plates were incubated for an additional 4 h at 37°C. The supernatant was removed, and the formazan crystals were dissolved by the addition of 150 μl per well dimethyl sulfoxide. Absorbance at 570 nm was measured using a microplate reader (Thermo Scientific, Thermo Electron Co., Carlsbad, CA, USA).

Wound-healing migration assay

DU145 and PC-3 cells were transfected for 24 h, seeded into 6-well plates, and grown to confluency. A sterile pipette tip was then used to scratch a wound across the cell monolayer, and the wells were incubated at 37°C for 48 h. The cells were photographed at 0 h, 24 h, and 48 h and the wound width was measured. The wound-healing rate was calculated using data from five high-magnification visual fields for each well.

Transwell invasion assay

DU145 and PC-3 cell lines were transfected for 24 h, washed, resuspended in serum-free medium, and added at 5 × 104 cells per well to the upper chamber of Transwell chambers (catalog# 354481, Corning, Tewksbury, MA, USA). Medium containing 10% FBS was added to the lower chamber. The plates were incubated at 37°C for 24 h, and the invaded cells present on the lower side of the membrane were stained with crystal violet. Cells in five visual fields per well were counted for each condition.

Apoptosis assay

DU145 and PC-3 cells were transfected for 24 h and 48 h, respectively, and washed. Samples of 5 × 104 to 5 × 105 cells were stained with propidium iodide (PI) and annexin V (Annexin V-FITC-PI kit; Beyotime, Shanghai, China) according to the manufacturer's protocol. Stained cells were immediately analyzed using a BD FACSVerse (Becton Dickinson, Franklin Lakes, NJ, USA).

Western blot analysis

Cells were transfected for 48 h and lysed using RIPA buffer (catalog# 9806S, CST, Danvers, MA, USA). The samples were centrifuged, and proteins in the supernatants were separated by sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis with 10% or 12% gels and then transferred to PVDF membranes (Millipore, Billerica, MA, USA). The membranes were blocked by incubation with skim milk(1:5000; Boster, Wuhan, China) and then incubated for 12 h at 4°C with primary antibodies against E2F3 (catalog# ab50917; Abcam, Cambridge, UK), PRDM1 (catalog# ab106766, Abcam), cyclin-dependent kinase 17 (CDK17, catalog# ab159068, Abcam), or β-actin (catalog# ab8227, Abcam). After washing, the membranes were incubated for 2 h at 37°C with anti-rabbit, anti-mouse, or anti-goat IgG secondary antibodies. Electronic chemical Laboratory (ECL) detection kit was used for the signal development (Merck Millipore, Darmstadt, Germany).

Mouse xenograft model

Female athymic nude mice, 4 weeks–6 weeks of age (Model Animal Research Center of Soochow University, Suzhou, China) were randomly assigned to groups of 5 per condition. PC-3 or DU145 cells (107 per mouse) were injected into the right mammary fat pads mouse of each mouse. When tumors reached approximately 50 mm3 in volume, control or miR-3691-3p agomirs (designed by RiboBio) were injected into the tumors (1 nmol per injection, three times per week) for 2 weeks. Tumors were then excised and volumes were calculated as V (mm3) = 0.5 × x2 × y, where x and y represent the tumor width and length, respectively. Tumor samples were sectioned and processed for analysis. All experimental protocols were approved by the Institutional Animal Care and Use Committee of Soochow University (ECSU-201800099).

Immunohistochemistry staining (IHC)

Paraffin-embedded sections (3 μm–5 μm thick) of xenograft tumors were immunostained following general protocols. Briefly, sections were boiled in 10 mmol l−1 citrate buffer for 10 min, incubated with normal rabbit serum to block nonspecific binding, and then immunostained with primary antibodies against TGF-β1 (catalog# 3C11, Santa Cruz Biotechnology, Santa Cruz, CA, USA), E2F3 (catalog# ab50917, Abcam), PRDM1 (catalog# bs-6466R, Bioss, Beijing, China), Ki-67 (catalog# bs-23103R, Bioss), or cleaved caspase-3 (catalog# ab2302, Abcam). The sections were then incubated with biotinylated anti-rabbit secondary antibody (Beyotime) followed by 3,3'diaminobenzidine tetrahydrochloride (Beyotime). The slides were counterstained with hematoxylin. Protein expression levels were semi-quantified using the total immunoreactive score (IRS). Three independent pathologists blindly reviewed each slide and categorized the staining extent and intensity on a four-point (0–3) scale. Extent: 0, no positive cells; 1, ≤25% positive cells; 2, 26%–50% positive cells; or 3, >50% positive cells. Intensity: 0, negative; 1, weak; 2, moderate; or 3, strong. The extent and intensity scores were multiplied to give a total IRS between 0 and 9. IRS of 6–9 and 0–4 were defined as high and low protein expression, respectively.

Statistical analyses

Statistical analysis was performed using SPSS version 12.0 for Windows (SPSS, Chicago, IL, USA). All in vitro assays were performed at least three times, each in triplicate. Continuous variables are expressed as mean ± standard deviation (s.d.) of three independent experiments. Survival was analyzed using Kaplan-Meier and Cox regression methods. P < 0.05 was considered statistically significant.

  Results Top

TGF-β1-regulated expression of miR-3691-3p in PCa cell lines and tissues

To identify TGF-β1-regulated miRNAs in adenocarcinomas, we first screened a dataset of miRNAs in a TGF-β1-deficient human colorectal adenocarcinoma cell line compared with the parental cell line (GSE53337) and identified five significantly differentially expressed miRNAs; miR-4723-3p, miR-324-3p, miR-4313, miR-196a-3p, and miR-3691-3p. To determine whether the same miRNAs were regulated by TGF-β1 in human PCa cells, we performed RT-qPCR analysis of DU145 and PC-3 human PCa cell lines incubated with or without TGF-β1 (10 ng ml−1) for 24 h. As shown in [Figure 1]a, four of the miRNAs, the exception being miR-4313, were downregulated in TGF-β1-treated PC-3 cells compared with the untreated controls (fold change [FC] = 0.60; P < 0.05 for all four), whereas in TGF-β1-treated DU145 cells, miR-4313 (FC = 0.12; P < 0.01) and miR-3691-3p (FC = 0.15; P < 0.01) were markedly decreased by TGF-β1, miR-324-3p and miR-4723-3p were slightly decreased (both FC = 0.55; P < 0.05), and miR-196a-3p was marginally and insignificantly increased (FC=1.27). BPH is a known risk factor for PCa,[24],[25],[26] and the majority of PCa patients have a history of chronic prostatitis. Therefore, we also examined expression of the five miRNAs in five paired samples of PCa and BPH specimens. As shown in [Figure 1]b, miR-3691-3p expression was marked lower in the primary PCa samples compared with BPH samples (FC = 0.05; P < 0.01), miR-196a-3p, miR-324-3p, and miR-4723-3p were slightly lower in PCa (FC = 0.82), and excepted miR-4313 (FC = 0.04; P < 0.01). We also analyzed 100 pairs of PCa and adjacent noncancerous tissues and again observed that miR-3691-3p was significantly decreased in PCa tissues compared with control tissues (P < 0.001; [Figure 1]c). These data prompted us to select miR-3691-3p for further investigation. Whether the decrease in miR-3691-3p express in PCa tissues is related to disease progression requires further analysis.
Figure 1: Expression levels of miR-3691-3p in PCa cell lines and tissues. (a) RT-qPCR analysis of five miRNAs in PC-3 and DU145 cells incubated with or without 5 ng ml−1 TGF-β1 for 24 h. (b) RT-qPCR analysis of five TGF-β1-regulated miRNAs in five paired samples of human PCa and BPH tissues. (c) RT-qPCR analysis of miR-3691-3p expression in 100 paired samples of PCa and normal prostate tissues. Data are presented as the mean ± s.d. of three independent experiments. (d) PCa patients were dichotomized into high and low miR-3691-3p expression groups using the median expression level as the cut-off value. Kaplan–Meier curves show BCR-free survival. Data are presented as the mean ± s.d. of one experiment, representative of three independent experiments. *P < 0.05; **P < 0.01; ***P < 0.001. RT-qPCR: real-time quantitative polymerase chain reaction; TGF-β1: transforming growth factor beta 1; PCa: prostate cancer; BPH: benign prostatic hyperplasia; BCR: biochemical recurrence; s.d.: standard deviation.

Click here to view

Correlation between miR-3691-3p expression and prognosis

To assess the relationship between miR-3691-3p expression level and clinicopathological and prognostic features in PCa, we assigned the 100 patients to groups of low and high miR-3691-3p expression using the median relative expression level of miR-3691-3p in PCa tissues as the cut-off. We found that lower miR-3691-3p expression was significantly associated with advanced pathological stage (P = 0.0251), positive lymph node metastasis (P = 0.003), high preoperative PSA level (P = 0.0019), and positive angiolymphatic invasion (P = 0.0017). However, there was no statistically significant association between miR-3691-3p expression and other features, including Gleason score and age [Table 1].
Table 1: Correlation between miR.3691.3p expression and clinicopathological features in prostate cancer patients

Click here to view

The prognostic impact of miR-3691-3p in PCa was analyzed by constructing Kaplan–Meier survival curves. Patients with lower miR-3691-3p expression level had a significantly shorter biochemical recurrence (BCR) - free survival after radical prostatectomy than did patients with higher miR-3691-3p levels (P = 0.0113, log-rank test; [Figure 1]d.

Taken together, these data demonstrate that the decreased miR-3691-3p expression level in PCa adversely affects prognosis.

Effect of miR-3691-3p overexpression on the behavior of PCa cell lines

To investigate how reduced miR-3691-3p expression affects PCa behavior, we transiently transfected a miR-3691-3p mimic or control sequence into PC-3 and DU145 cells, verified that miR-3691-3p was overexpressed by qPCR [Supplementary Figure 1 [Additional file 1]], and then analyzed the effects of miR-3691-3p upregulation on the proliferation, migration, invasion, and apoptosis of PCa cells in vitro. The MTT proliferation assay revealed a reduction in viability on day 5 in cells transfected with the miR-3691-3p mimic compared with the control sequence for both DU145 cells (P < 0.001) and PC-3 cells (P < 0.05; [Figure 2]a). Moreover, transfection with the miR-3691-3p mimic increased the apoptotic rate of PCa cells, as detected by annexin V-FITC/PI staining and flow cytometry [Figure 2]b. Thus, the mean percentage of miR-3691-3p-overexpressing DU145 cells in early apoptosis (annexin V-positive/PI-negative) and late apoptosis (annexin V-positive/PI-positive) was 18.3% and 10.1%, respectively, compared with 4.5% and 1.2%, respectively, for control DU145 cells [Figure 2]b. Similarly, the mean percentage of miR-3691-3p-overexpressing PC-3 cells in early and late apoptosis was 7.7% and 11.3%, respectively, compared with 3.8% and 4.5%, respectively for the control cells [Figure 2]b. We next analyzed the effect of miR-3691-3p overexpression on PCa cell migration and invasion using wound-healing and Transwell assays, respectively. Notably, DU145 and PC-3 cells transfected with the miR-3691-3p mimic displayed markedly and significantly inhibited migration (P < 0.001; [Figure 2]c) and invasion (P < 0.001; [Figure 2]d) compared with cells transfected with the control sequence.
Figure 2: Effect of miR-3691-3p overexpression on the behavior of PCa cell lines. DU145 and PC-3 cells were transfected with a control or miR-3691-3p mimic and then analyzed. (a) Cell proliferation was analyzed by the MTT assay on days 1, 3 and 5 after transfection. (b) Apoptosis was analyzed using an Annexin V-FITC/PI flow cytometry assay. Representative plots of DU145 cells (upper panels) and PC-3 cells (lower panels) are shown on day 2 after transfection. (c) Representative images (upper panels) and quantification (lower panels) of migration on a wound-healing assay of PCa cells on day one and ady two after transfection. (d) Transwell invasion assay of PCa cells after transfection. Invasion was quantified by counting five independent symmetrical visual fields under the microscope. Data are presented as the mean ± s.d. of one experiment, representative of three independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001. PCa: prostate cancer; MTT: methyl thiazolyl tetrazolium; FITC: fluorescein isothiocyanate; PI: propidium iodide; OD: optical density; s.d.: standard deviation.

Click here to view

Prediction and expression of candidate miR-3691-3p target genes in PCa cells

We used the DIANA-MICROT (, Target Scan (, and MIRDB databases ( to predict potential target genes of miR-3691-3p. Among the 200 target genes identified, 9 were commonly identified by all three algorithms. Using miR-Oncology database screening, we selected three target genes known to be associated with the occurrence and development of cancer; namely, Cyclin Dependent Kinase 17(CDK17), PRDM1 and E2F3 [Figure 3]a. To validate these potential miR-3691-3p target genes, we performed western blot analysis of PC-3 cells transfected with the control mimic or miR-3691-3p-mimic. As shown in [Figure 3]b, E2F3 protein was present at significantly lower levels in miR-3691-3p mimic-expressing cells than in control DU145 and PC-3 cells. In contrast, PRDM1 expression was significantly decreased by the miR-3691-3p mimic only in DU145 cells, while CDK17 was not significantly affected by the miR-3691-3p mimic in either cell line [Figure 3]b. We next confirmed these findings by RT-qPCR analysis and found that transfection with the miR-3691-3p mimic significantly decreased E2F3 mRNA levels in both DU145 cells (P < 0.001) and PC-3 cells (P < 0.01); significantly decreased PRDM1 mRNA levels in DU145 cells but not PC-3 cells (P < 0.01); and did not significantly affect CDK17 mRNA levels in either cell line [Figure 3]c, which was consistent with the results of the western blot analyses. Based on these results, E2F3 and PRDM1 were considered to be likely target genes of miR-3691-3p in PCa cells. Finally, these results were further substantiated by demonstrating concentration-dependent decreases in E2F3 and PRDM1 expression levels in both cell lines transfected with 10 nmol l−1–100 nmol l−1 miR-3691-3p mimic [Supplementary Figure 2 [Additional file 2]].
Figure 3: Effect of knockdown of endogenous miR-3691-3p target genes on the behavior of PCa cells. (a) Of the 200 target genes identified from three bioinformatic algorithms (DIANA-MICROT, Target Scan, and MIRDB), 9 were identified by all three. miR-Oncology database screening identified CDK17, PRDM1, and E2F3 as potential miR-3691-3p target genes closely related to the occurrence and development of tumors. (b and c) Western blot analysis (b) and RT-qPCR analysis (c) of CDK17, E2F3, and PRDM1 protein and mRNA expression, respectively, in PC-3 and DU145 cells at 48 h after transfection with a control or miR-3691-3p mimic. (d–f) DU145 and PC-3 cell lines were transfected with a control, E2F3-targeting, or PRDM1-targeting siRNA, and then analyzed. (d) Cell proliferation was assessed using the MTT assay on days one, two, and three after transfection. (e) Wound-healing migration assay performed on day one and day two after transfection. (f) Transwell invasion assay performed after transfection. Data are presented as the mean ± s.d. *P < 0.05, **P < 0.01, ***P < 0.001. PCa: prostate cancer; CDK17: cyclin dependent kinase 17; RT-qPCR: real-time quantitative polymerase chain reaction; PRDM1: PR domain containing 1, with ZNF domain; MTT: methyl thiazolyl tetrazolium; OD: optical density; E2F3: E2F transcription factor 3; s.d.: standard deviation.

Click here to view

Effect of E2F3 and PRDM1 knockdown on the behavior of PCa cell lines

E2F3 and PRDM1 are both transcriptional regulators that influence biological processes such as the cell cycle and apoptosis. To determine whether E2F3 and PRDM1 mediate the effects of miR-3691-3p in PCa cells, we transfected cells with E2F3- or PRDM1-specific siRNAs and examined the effects on cell proliferation, invasion, and migration. We first confirmed effective inhibition of E2F3 and PRDM1 expression at both the mRNA level (P < 0.001; [Supplementary Figure 3 [Additional file 3]]a) and protein level [Supplementary Figure 3]b) by RT-qPCR and western blot analysis, respectively. Analysis of proliferation using the MTT assay showed that E2F3 and PRDM1 silencing suppressed the proliferation of both PCa cell lines [Figure 3]d, consistent with the effects of the miR-3691-3p mimic. The wound-healing migration assay showed that PC-3 cells expressing E2F3- or PRDM1-targeting siRNAs migrated shorter distances than control cells, while in DU145 cells, only PRDM1 silencing reduced cell migration [Figure 3]e. Last, the invasion capability of the two lines was significantly reduced by transfection with either E2F3 or PRDM1 siRNA compared with the control siRNA (P < 0.001; s[Figure 3]f). These data suggest that downregulation of miR-3691-3p and concomitant upregulation of E2F3 and PRDM1 may play a key role in the progression of PCa.]

Effect of miR-3691-3p agomir administration on PCa growth in a mouse xenograft model

To verify that our in vitro results with cell lines translated to the in vivo situation, we established a PCa xenograft mouse model. For this, PC-3 or DU145 cells were injected into the mammary pads of female nude mice, and tumors were allowed to develop. The mice were then injected intratumorally with a control sequence or a miR-3691-3p-agomir three times a week for 2 weeks. At the end of the experiment, the tumors were excised for analysis. As shown in [Figure 4]a, [Figure 4]b, [Figure 4]c, the volume and weight of PCa tumors injected with the miR-3691-3p agomir were smaller than the tumors injected with the control sequence, indicating that elevation of miR-3691-3p concentrations locally at the tumor site suppressed tumor cell growth. We also performed IHC staining of tumor sections to investigate the expression levels of E2F3, PRDM1, and two key proteins involved in cell proliferation (Ki67) and apoptosis (cleaved caspase-3). These analyses showed that miR-3691-3p agomir injection inhibited tumor expression of E2F3 and PRDM1 [Figure 4]d and the proliferation marker Ki67 and increased the expression of the apoptosis effector cleaved caspase-3 [Figure 4]e, consistent with the results of the in vitro analyses. These results were observed in tumors formed by both DU145 and PC-3 cell lines. Collectively, these data indicate that miR-3691-3p suppresses the growth of PCa cell lines both in vitro and in vivo.
Figure 4: Effect of miR-3691-3p agomir treatment on the growth of PCa xenografts in mice. Groups of female nude mice were injected with DU145 or PC-3 cells into the mammary fat pads. When the tumors reached 50 mm3 in volume, a control sequence or miR-3691-3p-agomir was injected intratumorally three times a week for 2 weeks. (a) Tumors were then removed and their appearance, (b) volume, and (c) weight were recorded. Immunohistochemical staining of (d) E2F3 and PRDM1 and (e) Ki67 and cleaved caspase-3. Scale bar: 100 μm. Data are presented as the mean ± s.d. *P < 0.05, **P < 0.01, ***P < 0.001. PCa: prostate cancer; E2F3: E2F transcription factor 3; PRDM1: PR domain containing 1, with ZNF domain; s.d.: standard deviation.

Click here to view

Expression of E2F3 and PRDM1 in PCa clinical specimens

Finally, we examined the expression levels of E2F3 and PRDM1 by IHC staining of 168 PCa tumors and 65 BPH tissues. The results showed that E2F3 and PRDM1 were mainly expressed in the nucleus of prostatic epithelial cells. The two proteins were both strongly expressed in cancer tissues but were present at lower levels in BPH tissues [Supplementary Figure 4 [Additional file 4]]. Quantification of protein expression showed strong E2F3 staining in a higher proportion of PCa tissues (31.0%, 52/168) than BPH tissues (10.8%, 7/65; P < 0.05; [Table 2]), and similarly, significantly more PCa tissues than normal tissues showed positive staining for PRDM1 (28.6% [48/168] vs 16.9% [11/65]; P < 0.01; [Table 2]). Thus, the observed elevated expression of E2F3 and PRDM1 target genes was consistent with the reduced expression of miR-3691-3p detected in PCa compared with BPH clinical specimens.
Table 2: Expression of E2F3 and PRDM1 in prostate cancer and normal prostate tissues

Click here to view

  Discussion Top

Previous studies have shown that aberrant expression of several miRNAs play a role in the occurrence, development, and progression of PCa,[27] which has implications for the development of novel diagnostic, prognostic, and therapeutic tools.[28] TGF-β1 is well established to be involved in the growth and progression of a variety of cancers, including those of the breast, colorectum, and prostate, and serum TGF-β1 protein and mRNA levels are significantly increased in these cancers.[29] TGF-β1, a member of the TGF-β superfamily,[30] plays a dual role in cancer in that it is generally tumor suppressive but promotes the progression of late-stage cancers.[31] The TGF-β1 signaling pathway can promote small mother against decapentaplegic (SMAD) protein binding to miRNA promoter genes,[32],[33] thereby altering the expression of numerous miRNAs.[34],[35],[36],[37]

We previously showed that TGF-β1-mediated inhibition of miR-450b-5p reversed the differentiation of rhabdomyosarcoma.[13] In the present study, we identified miR-3691-3p as the most significantly downregulated miRNA among those tested in primary PCa compared with BPH samples. Moreover, low miR-3691-3p expression correlated positively with advanced pathological stage, lymph node metastasis, and shorter BCR-free survival after radical prostatectomy, thereby supporting the potential application of miR-3691-3p as a prognostic indicator and possible therapeutic target in PCa.

Our data suggest that miR-3691-3p is an anti-oncogene. Overexpression of miR-3691-3p inhibited the proliferation of PCacells in vitro and in vivo and reduced metastatic behavior in vitro. In addition, we predicted and validated E2F3 and PRDM1 as miR-3691-3p target genes and demonstrated that knockdown of endogenous E2F3 and PRDM1 in PCa cell lines attenuated malignant behaviors such as proliferation and metastasis.

E2F3 is a member of the E2F transcription factor family and is essential for cell proliferation.[38] Many reports have demonstrated the importance of intricate networks between E2F3 and miRNAs in regulating the balance of proliferation, apoptosis, and metastasis in various cancers.[39] Deregulated E2F3 transcriptional activity is present in the vast majority of human cancers and has been clearly implicated in the dysregulation of cell cycle control, proliferation, and apoptosis.[40] E2F3 is also a key transcription factor in tumor-associated macrophages and influences both the tumor microenvironment and tumor cell metastasis.[41] Here, we demonstrated that E2F3 regulates the malignant behavior of PCa cell lines in vitro, and showed that its expression is downregulated by miR-3691-3p agomir treatment of PCa xenografts in a mouse model. However, the exact mechanisms by which miR-3691-3p modulates E2F3 expression and the progression of PCa requires further study.

The second miR-3691-3p target gene identified in this study was PRDM1 (also known as B lymphocyte-induced maturation protein [Blimp-1]), which is known to play critical roles in the development and differentiation of many cell types in the mouse and other model organisms.[42],[43] PRDM1 is implicated in malignancy through its interactions with the p53 tumor-suppressor pathway, suggesting that PRDM1 itself may have a role in tumor suppression.[44] Another study showed that PRDM1 inhibits SW620 colon cancer cell proliferation via inhibition of c-Myc.[45] In the present study, we demonstrated that PRDM1 was significantly overexpressed in PCa compared with BPH clinical tissues, and that PRDM1 silencing in PCa cell lines inhibited their proliferation, migration, and invasion ability. Thus, our results suggest that PRDM1 may act as an oncogene in the progression of PCa.

  Author Contributions Top

YMH and LXL performed experiments, analyzed, and interpreted data. BZL, LS and SW analyzed data and provided samples and clinical data. LW, XZ, QZ and MMS performed mouse experiments. KT and YSZ provided tissue samples and analyzed immunohistology data. YSZ and SLW designed experiment and wrote the paper. All authors read and approved the final manuscript.

  Competing Interests Top

All authors declare no competing interests.

  Acknowledgments Top

This study was supported by Shanghai Changning District Committee of Science and Technology (CNKW2016Y01), Shanghai Tongren Hospital Project (TRYJ201501), Suzhou Science and Technology Development Program (SYS201717), the Second Affiliated Hospital of Soochow University Advance Research Program of the Natural Science Foundation of China Grants (SDFEYGJ1705), Open project of Jiangsu State Key Laboratory of Radiation Medicine and Projection (GJS1963) and the Priority Academic Program Development of Jiangsu Higher Education Institutions.

Supplementary Information is linked to the online version of the paper on the Asian Journal of Andrology website.

  References Top

Parnes HL, House MG, Tangrea JA. Prostate cancer prevention: strategies for agent development. Curr Opin Oncol 2013; 25: 242–51.  Back to cited text no. 1
Damber JE, Aus G. Prostate cancer. Lancet 2008; 371: 1710–21.  Back to cited text no. 2
Guichard G, Larre S, Gallina A, Lazar A, Faucon H, et al. Extended 21-sample needle biopsy protocol for diagnosis of prostate cancer in 1000 consecutive patients. Eur Urol 2007; 52: 430–5.  Back to cited text no. 3
Ling XH, Han ZD, Xia D, He HC, Jiang FN, et al. MicroRNA-30c serves as an independent biochemical recurrence predictor and potential tumor suppressor for prostate cancer. Mol Biol Rep 2014; 41: 2779–88.  Back to cited text no. 4
Utomo NB, Mochtar CA, Umbas R. Primary hormonal treatment in localized and locally advanced prostate cancer: effectiveness and survival predictive factors. Acta Med Indones 2012; 44: 10–5.  Back to cited text no. 5
Dasgupta S, Srinidhi S, Vishwanatha JK. Oncogenic activation in prostate cancer progression and metastasis: Molecular insights and future challenges. J Carcinog 2012; 11: 4.  Back to cited text no. 6
Ventura A, Jacks T. MicroRNAs and cancer: short RNAs go a long way. Cell 2009; 136: 586–91.  Back to cited text no. 7
Shah NR, Chen H. MicroRNAs in pathogenesis of breast cancer: implications in diagnosis and treatment. World J Clin Oncol 2014; 5: 48–60.  Back to cited text no. 8
Drusco A, Nuovo GJ, Zanesi N, Di Leva G, Pichiorri F, et al. MicroRNA profiles discriminate among colon cancer metastasis. PLoS One 2014; 9: e96670.  Back to cited text no. 9
D'Anzeo M, Faloppi L, Scartozzi M, Giampieri R, Bianconi M, et al. The role of micro-RNAs in hepatocellular carcinoma: from molecular biology to treatment. Molecules 2014; 19: 6393–406.  Back to cited text no. 10
Walter BA, Valera VA, Pinto PA, Merino MJ. Comprehensive microRNA profiling of prostate cancer. J Cancer 2013; 4: 350–7.  Back to cited text no. 11
Karatas OF, Guzel E, Suer I, Ekici ID, Caskurlu T, et al. miR-1 and miR-133b are differentially expressed in patients with recurrent prostate cancer. PLoS One 2014; 9: e98675.  Back to cited text no. 12
Sun MM, Li JF, Guo LL, Xiao HT, Dong L, et al. TGF-beta1 suppression of microRNA-450b-5p expression: a novel mechanism for blocking myogenic differentiation of rhabdomyosarcoma. Oncogene 2014; 33: 2075–86.  Back to cited text no. 13
Bello-DeOcampo D, Tindall DJ. TGF-betal/Smad signaling in prostate cancer. Curr Drug Targets 2003; 4: 197–207.  Back to cited text no. 14
Calon A, Espinet E, Palomo-Ponce S, Tauriello DV, Iglesias M, et al. Dependency of colorectal cancer on a TGF-beta-driven program in stromal cells for metastasis initiation. Cancer Cell 2012; 22: 571–84.  Back to cited text no. 15
David CJ, Huang YH, Chen M, Su J, Zou Y, et al. TGF-beta tumor suppression through a lethal EMT. Cell 2016; 164: 1015–30.  Back to cited text no. 16
Yu JR, Tai Y, Jin Y, Hammell MC, Wilkinson JE, et al. TGF-beta/Smad signaling through DOCK4 facilitates lung adenocarcinoma metastasis. Genes Dev 2015; 29: 250–61.  Back to cited text no. 17
Saraon P, Jarvi K, Diamandis EP. Molecular alterations during progression of prostate cancer to androgen independence. Clin Chem 2011; 57: 1366–75.  Back to cited text no. 18
Morimoto K, Tanaka T, Nitta Y, Ohnishi K, Kawashima H, et al. NEDD9 crucially regulates TGF-beta-triggered epithelial-mesenchymal transition and cell invasion in prostate cancer cells: involvement in cancer progressiveness. Prostate 2014; 74: 901–10.  Back to cited text no. 19
Thakur N, Gudey SK, Marcusson A, Fu JY, Bergh A, et al. TGFbeta-induced invasion of prostate cancer cells is promoted by c-Jun-dependent transcriptional activation of Snail1. Cell Cycle 2014; 13: 2400–14.  Back to cited text no. 20
Jin W, Chen F, Wang K, Song Y, Fei X, et al. miR-15a/miR-16 cluster inhibits invasion of prostate cancer cells by suppressing TGF-beta signaling pathway. Biomed Pharmacother 2018; 104: 637–44.  Back to cited text no. 21
Ottley E, Gold E. microRNA and non-canonical TGF-beta signalling: implications for prostate cancer therapy. Crit Rev Oncol Hematol 2014; 92: 49–60.  Back to cited text no. 22
Chen Y, Huang S, Wu B, Fang J, Zhu M, et al. Transforming growth factor-beta1 promotes breast cancer metastasis by downregulating miR-196a-3p expression. Oncotarget 2017; 8: 49110–22.  Back to cited text no. 23
Perletti G, Monti E, Magri V, Cai T, Cleves A, et al. The association between prostatitis and prostate cancer. Systematic review and meta-analysis. Arch Ital Urol Androl 2017; 89: 259–65.  Back to cited text no. 24
Jiang J, Li J, Yunxia Z, Zhu H, Liu J, et al. The role of prostatitis in prostate cancer: meta-analysis. PLoS One 2013; 8: e85179.  Back to cited text no. 25
Cai T, Santi R, Tamanini I, Galli IC, Perletti G, et al. Current knowledge of the potential links between inflammation and prostate cancer. Int J Mol Sci 2019; 20: 3833.  Back to cited text no. 26
Bonci D, Coppola V, Musumeci M, Addario A, Giuffrida R, et al. The miR-15a-miR-16-1 cluster controls prostate cancer by targeting multiple oncogenic activities. Nat Med 2008; 14: 1271–7.  Back to cited text no. 27
Schaefer A, Stephan C, Busch J, Yousef GM, Jung K. Diagnostic, prognostic and therapeutic implications of microRNAs in urologic tumors. Nat Rev Urol 2010; 7: 286–97.  Back to cited text no. 28
Gold LI. The role for transforming growth factor-beta (TGF-beta) in human cancer. Crit Rev Oncog 1999; 10: 303–60.  Back to cited text no. 29
Kingsley DM. The TGF-beta superfamily: new members, new receptors, and new genetic tests of function in different organisms. Genes Dev 1994; 8: 133–46.  Back to cited text no. 30
Massague J. TGFbeta in cancer. Cell 2008; 134: 215–30.  Back to cited text no. 31
Hata A, Davis BN. Control of microRNA biogenesis by TGFbeta signaling pathway - a novel role of Smads in the nucleus. Cytokine Growth Factor Rev 2009; 20: 517–21.  Back to cited text no. 32
Kong W, Yang H, He L, Zhao JJ, Coppola D, et al. MicroRNA-155 is regulated by the transforming growth factor beta/Smad pathway and contributes to epithelial cell plasticity by targeting RhoA. Mol Cell Biol 2008; 28: 6773–84.  Back to cited text no. 33
Butz H, Racz K, Hunyady L, Patocs A. Crosstalk between TGF-beta signaling and the microRNA machinery. Trends Pharmacol Sci 2012; 33: 382–93.  Back to cited text no. 34
Davis-Dusenbery BN, Chan MC, Reno KE, Weisman AS, Layne MD, et al. Down-regulation of Kruppel-like factor-4 (KLF4) by microRNA-143/145 is critical for modulation of vascular smooth muscle cell phenotype by transforming growth factor-beta and bone morphogenetic protein 4. J Biol Chem 2011; 286: 28097–110.  Back to cited text no. 35
Labbaye C, Testa U. The emerging role of MIR-146A in the control of hematopoiesis, immune function and cancer. J Hematol Oncol 2012; 5: 13.  Back to cited text no. 36
Zhang J, Zhang D, Wu GQ, Feng ZY, Zhu SM. Propofol inhibits the adhesion of hepatocellular carcinoma cells by upregulating microRNA-199a and downregulating MMP-9 expression. Hepatobiliary Pancreat Dis Int 2013; 12: 305–9.  Back to cited text no. 37
Wu L, Timmers C, Maiti B, Saavedra HI, Sang L, et al. The E2F1-3 transcription factors are essential for cellular proliferation. Nature 2001; 414: 457–62.  Back to cited text no. 38
Gao Y, Feng B, Lu L, Han S, Chu X, et al. MiRNAs and E2F3: a complex network of reciprocal regulations in human cancers. Oncotarget 2017; 8: 60624–39.  Back to cited text no. 39
Chen HZ, Tsai SY, Leone G. Emerging roles of E2Fs in cancer: an exit from cell cycle control. Nat Rev Cancer 2009; 9: 785–97.  Back to cited text no. 40
Trikha P, Sharma N, Pena C, Reyes A, Pecot T, et al. E2f3 in tumor macrophages promotes lung metastasis. Oncogene 2016; 35: 3636–46.  Back to cited text no. 41
Vincent SD, Dunn NR, Sciammas R, Shapiro-Shalef M, Davis MM, et al. The zinc finger transcriptional repressor Blimp1/Prdm1 is dispensable for early axis formation but is required for specification of primordial germ cells in the mouse. Development 2005; 132: 1315–25.  Back to cited text no. 42
Ohinata Y, Payer B, O'Carroll D, Ancelin K, Ono Y, et al. Blimp1 is a critical determinant of the germ cell lineage in mice. Nature 2005; 436: 207–13.  Back to cited text no. 43
Kucuk C, Iqbal J, Hu X, Gaulard P, De Leval L, et al. PRDM1 is a tumor suppressor gene in natural killer cell malignancies. Proc Natl Acad Sci U S A 2011; 108: 20119–24.  Back to cited text no. 44
Kang HB, Lee HR, Jee Da J, Shin SH, Nah SS, et al. PRDM1, a tumor-suppressor gene, is induced by Genkwadaphnin in human colon cancer SW620 cells. J Cell Biochem 2016; 117: 172–9.  Back to cited text no. 45


  [Figure 1], [Figure 2], [Figure 3], [Figure 4]

  [Table 1], [Table 2]

This article has been cited by
1 Transforming growth factor-beta (TGF-) in prostate cancer: A dual function mediator?
Sepideh Mirzaei, Mahshid Deldar Abad Paskeh, Yalda Saghari, Ali Zarrabi, Michael R. Hamblin, Maliheh Entezari, Mehrdad Hashemi, Amir Reza Aref, Kiavash Hushmandi, Alan Prem Kumar, Navid Rabiee, Milad Ashrafizadeh, Saeed Samarghandian
International Journal of Biological Macromolecules. 2022;
[Pubmed] | [DOI]
2 A Potential Prognostic Marker PRDM1 in Pancreatic Adenocarcinoma
Bo Zhou, Jie Zhang, Hongda Zhu, Shugeng Wu, Alamgeer Yuchi
Journal of Oncology. 2022; 2022: 1
[Pubmed] | [DOI]
3 MicroRNA-195-5p is associated with cell proliferation, migration and invasion in prostate cancer and targets MIB1
Bin Chen, Guohui Bai, Xiaoyan Ma, Lulin Tan, Houqiang Xu
Oncology Reports. 2021; 46(6)
[Pubmed] | [DOI]
4 E2F3 promotes liver cancer progression under the regulation of circ-PRKAR1B
Guodong Liu,Xiwu Ouyang,Liansheng Gong,Lei Yao,Shiqing Liu,Jiarong Li,Qi Zhang,Yao Xiao
Molecular Therapy - Nucleic Acids. 2021; 26: 104
[Pubmed] | [DOI]
5 Fresh vs frozen testicular sperm for assisted reproductive technology in patients with non-obstructive azoospermia: A systematic review
Medhat Amer,Emad Fakhry
Arab Journal of Urology. 2021; : 1
[Pubmed] | [DOI]


Similar in PUBMED
 Related articles
Access Statistics
Email Alert *
Add to My List *
* Registration required (free)

  In this article
Materials and Me...
Author Contributions
Competing Interests
Article Figures
Article Tables

 Article Access Statistics
    PDF Downloaded304    
    Comments [Add]    
    Cited by others 5    

Recommend this journal