This article is mentioned in:

Introduction

microRNAs (miRNAs) are recognized as a class of small (19–25 nt) non-coding RNAs that play important roles in gene regulation by partially or fully complementary matching with the 3′-untranslated region (UTR) of target mRNAs, and trigger a transcriptional and/or post-transcriptional suppression of the target gene (1,2). They are involved in numerous physiological functions such as cell differentiation, migration, proliferation, apoptosis and senescence (3). Recently, it was reported that more than 1,000 human miRNAs frequently target regions related to cancer development (4), and this complex collection could be further classified as oncogenic, tumor-suppressive, or context-dependent miRNAs (5).

miR-330 was first discovered by Weber (6), and described as a downregulated miRNA in tumors by Gaur et al(7), when the authors compared the differential expression patterns of 241 miRNAs between 13 normal and 59 tumor-derived cell lines. This finding was further confirmed by Ruike et al(8) who investigated the miRNA and mRNA expression associations across 16 human cell lines. mir-330 is located on chromosome 19q13, a fragile point which was reportedly associated with tumor aggressiveness in prostate cancer (9,10). To date, investigations concerning the biological functions of miR-330 in cancer are sparse. Recently, Qu et al(11) reported its oncogenic role in human glioblastoma by regulating the SH3GL2 gene. In prostate cancer, however, it was reported as a tumor-suppressor by Lee et al(12), who demonstrated its significantly lower expression and inverse correlation with E2F1 in both prostate cancer cell lines and cancer specimens. They further discovered its proapoptotic role through E2F1-mediated suppression of Akt phosphorylation.

Specificity protein 1 (Sp1), the essential member of the Sp/Krupel-like (KLF) transcription factor family, is regarded as ubiquitous in cells and participates in the tumorigenesis of a variety of cancers including prostate (13). It plays an important role in tumor progression, including cell proliferation, angiogenesis, differentiation, apoptosis, migration and invasion (14), representing itself as an ideal target for cancer treatment. Here, we report the post-transcriptional regulation of Sp1 by miR-330. Our investigation elucidated the tumor-suppressive role of miR-330 in prostate cancer in an anti-migratory and -invasive manner.

Materials and methods

Reagents

All the RNA duplexes were synthesized by GenePharma Co., Ltd. (Shanghai China). The sense sequences of the duplexes were as follows: miR-330a mimic (termed as miR-330), 5′-GCAAAGCACACGGCCUGCAGAGA-3′; negative control RNA (termed as NC), 5′-ACUACUGAGUGACA GUAGA-3′; small interfering RNA for Sp1 (termed as siRNA), 5′-AAAGCGCUUCAUGAGGAGUGA-3′ (15). All of the primers for quantitative real-time PCR (qPCR) were synthesized by Sangon Biotech (Shanghai China), whose sequences were as follows: GAPDH forward, 5′-AAGGTGAAGGTCG GAGTCA-3′ and reverse, 5′-GGAAGATGGTGATGGGATT T-3′; Sp1 forward, 5′-AGTTCCAGACCGTTGATGGG-3′ and reverse, 5′-GTTTGCACCTGGTATGATCTGT-3′; matrix metalloproteinase-2 (MMP2) forward, 5′-TACAGGATCATT GGCTACACACC-3′ and reverse, 5′-GGTCACATCGCTCC AGACT-3′; MMP9 forward, 5′-TGTACCGCTATGGTTAC ACTCG-3′ and reverse, 5′-GGCAGGGACAGTTGCTTCT-3′. The primary antibodies were, anti-GAPDH (Epitomics, Burlingame, CA, USA), anti-Sp1, anti-MMP2 (both from Santa-Cruz Biotechnology, Santa Cruz, CA, USA) and anti-MMP9 (Epitomics).

Cell culture and transfection

The human prostate cancer cell lines DU145 and PC3 and the normal prostate epithelial cell line RWPE were purchased from Shanghai Institute of Cell Biology, Chinese Academy of Sciences. Cells were cultured in RPMI-1640 medium supplemented with 10% heat-inactivated fetal bovine serum (FBS) (Gibco-BRL, Grand Island, NY, USA), penicillin (50 U/ml), and streptomycin (50 μg/ml) in a humidified atmosphere containing 5% CO2 maintained at 37°C. For miR-330 overexpression, we chose miR-330a mimic transfection, which is efficient and economical. Briefly, cells were seeded in a 6-well plate at a density of 20–40×104 cells/well in medium without antibiotics. At ~60–70% confluency, transfection was performed using Lipofectamine™ 2000 reagent purchased from Invitrogen Life Technologies (Carlsbad, CA, USA) according to the manufacturer’s instructions. The transfection efficiency was assessed by qPCR.

RNA isolation and qPCR

Cells were harvested 48 h after transfection, and were treated with RNAiso for small RNA and RNAiso Plus for total RNA. The extracted small RNA and total RNA were transcribed into cDNA with the One Step PrimeScript miRNA cDNA Synthesis kit and PrimeScript RT reagent kit (Takara, Japan), respectively. For quantification, qPCR was performed using SYBR Premix Ex Taq (Takara) with the 7500 Fast Sequence Detection System (Applied Biosystems, Foster City, CA, USA). After the Ct values of the target genes were subtracted by that of the internal control GAPDH, the relative expression level of each target gene was obtained using the 2−ΔΔCt method.

Apoptosis assay

Apoptosis was assessed by the Annexin V-FITC kit (Invitrogen Life Technologies). Both cell lines were transfected for 48 h, and were harvested and stained with Annexin V-FITC and PI according to the manufacturer’s instructions. Cell samples were analyzed by BD LSR II Flow Cytometry System with FACSDiva software (BD Biosciences, Franklin Lakes, NJ, USA) and apoptotic cell fractions were determined.

Cell cycle analysis

Briefly, cells were harvested 48 h after transfection, washed with phosphate-buffered saline (PBS), detrypsinized, resuspended and finally fixed in 75% ethanol at −20°C for 24 h after removal of the supernatant. Subsequently, cells were washed with PBS, and incubated with DNA Prep Stain (Beckman Coulter, Fullerton, CA, USA) for 30 min. We used the BD LSR II Flow Cytometry System with FACSDiva software to detect cell cycle distribution, which was analyzed by ModFit LT software package and presented as the percentage of cell counts in every cycle for the G1, S and G2 phases.

Western blot analysis

Cells were harvested ~48 h after treatment, and lysed in RIPA buffer; the isolated protein was quantified using the Bradford protein assay (Bio-Rad, Hercules, CA, USA) and separated with equivalent loading on 10% SDS-polyacrylamide gels subsequently transferred to nitrocellulose membranes (Millipore, Bedford, MA, USA). The membranes were blocked with 5% non-fat milk and incubated with the primary antibody with the recommended dilution rate at 4°C for at least 12 h, followed by incubation with the corresponding horseradish peroxidase (HRP)-conjugated secondary antibody at an appropriate dilution in TBS-Tween at room temperature for 1 h. Finally, Enhanced Chemiluminescence Plus reagent (Millipore) was used to detect the bound secondary antibody.

Transwell assay

Millicell Transwell inserts (8-μm) (Millipore) were placed in a 24-well plate (Corning Incorporated, Corning, NY, USA) with 600 μl serum-free RPMI-1640 medium in the lower chamber. The oligo-treated cells were harvested, counted and seeded (6–8×104 cells/well) to the upper chamber. For the invasion assay, the wells were coated with Matrigel (BD Biosciences, Canada) diluted with serum-free RPMI-1640 medium in a 1:8 ratio before seeding. After incubation at 37°C with 5% CO2 for 48 h, the inserts were withdrawn and fixed in cold methanol for 10 min and then stained with 0.5% crystal violet in PBS. After 15 min of incubation, the inserts were washed thoroughly in water and the cells on the upper surfaces of the membranes were removed, and then dried overnight. The membranes were excised and fixed on slides with neutral resin, and the cells fixed to the bottom were observed using an inverted microscope.

Plasmid construction and luciferase reporter assay

For investigating the possible target genes of miR-330, we used the target prediction program Targetscan (http://www.targetscan.org/). As the program predicted, the potential target gene Sp1 has 2 putative binding sites in its 3′-UTR. Therefore, we designed 2 pairs of primers for amplification from the cDNA of RWPE cells: Site 1 forward, 5′-TGCAAGGTAGCATGGGTCCAAGAG-3′ and reverse, 5′-TGGTGGGAAGCCAAGACAACA-3′; Site 2 forward, 5′-GTCGCAGCAGTAGCTTTGGGGA-3′ and reverse, 5′-ACTTAGGGCAGTTGAGAGGCAGA-3′. The products were then inserted into the pmirGLO Dual-Luciferase miRNA target expression vector (Promega Corporation, Madison, WI, USA) between the SacI and SalI sites (named pGL-WT1 and pGL-WT2). All the insertions were confirmed by sequencing. HEK293T cells were plated in a 24-well plate and cultured for 24 h. Co-transfection was performed with 50 nM of either the miR-330 mimic or NC oligo and 100 ng wild-type reporter plasmid. The relative luciferase activity was measured by the Dual-Luciferase reporter assay system (Promega) 24 h after transfection.

Statistical analysis

All the graphical data were constructed by GraphPad Prism version 5 for Windows. Data were analyzed by use of SPSS 15.0 software, and are expressed as means ± standard deviation (SD) for each group. The t-test or one-way ANOVA was used to make comparisons; P<0.05 was considered to indicate a statistically significant result.

Results

miR-330 is downregulated in prostate cancer cells

To confirm the expression levels of miR-330 in the human prostate cancer cell lines, mRNAs were extracted from DU145, PC3 and RWPE cells, and then transcribed into the template for qPCR quantification analysis. We used U6 RNA level as an internal control. Compared with the RWPE cells, both cancer cell lines exhibited significantly decreased expression of miR-330 when normalized to U6 [DU145 downregulated by 0.5-fold (P<0.05), and PC3 by 0.7-fold (P<0.01)] (Fig. 1). This result indicates that miR-330 is downregulated in prostate cancer cells.

Figure 1 Quantification of miR-330 in prostate cancer cell lines. The relative expression level of miR-330 was decreased by 0.5-fold in the DU145 cells and 0.3-fold in PC3 cells when compared with miR-330 in the normal epithelial cell line.

Cell migration is inhibited by miR-330 in prostate cancer

To determine whether miR-330 affects cell migratory capacity, cells with miR-330 overexpression were subjected to a Transwell assay (FBS-induced migration and Matrigel invasion test). After 24 to 48 h, the migrating cells were fixed, stained and observed microscopically. As the representative micrographs clearly demonstrate, miR-330 overexpression led to potent inhibition of cell migration and invasion (Fig. 2).

Figure 2 Cell motility was inhibited by miR-330 in prostate cancer. The representative micrographs (magnification, ×100) clearly showed the potent influence of miR-330 on cell invasion and migration capability.

Neither apoptosis nor cell cycle arrest is induced by miR-330 in prostate cancer cells

To investigate the biological function of miR-330 we performed flow cytometry, as previous reports indicate a pro-apoptotic role of miR-330. Both cell lines were treated by either miR-330 or NC for 48 h, and flow cytometric analysis was carried out. There was an insignificant difference in the percentage of apoptotic cells between the miR-330 and NC pretreatment groups (DU145, 7.37±2.65 vs. 5.87±1.95; P>0.05 and PC3, 5.47±1.48 vs. 6.70±2.09; P>0.05) (Fig. 3A). Likewise, cell cycle analysis did not reveal any obvious change in cell cycle distribution in G1, S, G2 phases between both cell groups (P>0.05) (Fig. 3B). Therefore, miR-330 did not affect apoptosis or cell cycle arrest in the prostate cancer cells, at least during our observation period.

Figure 3 Neither apoptosis nor cell cycle arrest is induced by miR-330 in prostate cancer cells. Apoptosis and cell cycle distribution of DU145 and PC3 cells were examined by flow cytometry 48 h after transfection. (A) Quadrant (Q)4 represents early apoptotic cells (Annexin V+/PI2), and Q2 late apoptotic cells (Annexin V+/PI+); the apoptotic cell population was calculated by the sum of these two percentages, and represented as the means ± SD (P>0.05). (B) The cell populations at the G1, S, G2 phases were indicated by the representative histograms. There was a similar cellular distribution at each cell cycle phase between the NC group and the mimic group (P>0.05).

Sp1 is post-transcriptionally downregulated by miR-330

To further investigate the underlying mechanism of miR-330-induced metastatic inhibition, Sp1 was selected for qPCR and western blot analysis, by the aid of a bioinformatic tool. The result showed that following miR-330 overexpression, Sp1 expression was inhibited only at the protein level rather than at the mRNA level, which suggests a post-transcriptional regulation. Expression of other metastasis-related molecules such as MMP2 and MMP9 was also markedly downregulated at the mRNA and protein levels (Fig. 4A) (P<0.05), which verified the involvement of miR-330 in the regulation of metastasis.

Figure 4 miR-330 post-transcriptionally downregulates specificity protein 1 (Sp1) and is inversely correlated with it. (A) Marked reduction in MMP2 and MMP9 mRNA expression was observed in both prostate cancer cell lines (*P<0.05; **P<0.01), while Sp1 was nearly unaffected (P>0.05). Western blotting in parallel indicated the prominent decrease in Sp1, MMP2 and MMP9 at the protein level. (B) An elevated expression level of Sp1 was found in the prostate cancer cell lines compared with that in the RPWE cells (*P<0.05; **P<0.01).

The basic expression levels of Sp1 were also examined and compared between the DU145 and PC3 cells and normal epithelial RWPE cells. Sp1 expression was obviously elevated in the cancer cells and was ~2-fold increased when compared with the level in the RWPE cells (P<0.05) (Fig 4B). Considering the above data and the inverse correlation with miR-330 (Fig. 1), we surmised that Sp1 is very likely the target of miR-330.

Inhibition of cell motility by miP-330 is associated with downregulation of Sp1

For further investigation of the possible novel target Sp1, knockdown technique was employed to evaluate its biological function in vitro. A small interfering RNA (siSp1) designed to knock down the Sp1 gene was used to treat the cancer cells compared with NC. The Transwell assay clearly demonstrated that Sp1 knockdown caused significant suppression of the migratory and invasive capability in much the same pattern as miR-330 overexpression (Fig. 5B). In the mechanistic study, Sp1 knockdown, as anticipated, had a similar effect as miR-330 overexpression; MMP2 and MMP9 expression was prominently decreased at both the mRNA and protein levels; a drastic decrease in Sp1 expression indicated efficiency of knockdown (Fig. 5A).

Figure 5 Knockdown of specificity protein 1 (Sp1) mimics miR-330 overexpression. (A) qPCR and western blotting showed a significant decrease in Sp1, MMP2 and MMP9 in both prostate cancer cell lines (*P<0.05; **P<0.01) after knockdown of Sp1. (B) Representative micrographs (magnification, ×100) of the Transwell assay showed notable impairment of migration and invasion capacity of the cancer cells by Sp1 knockdown.

Sp1 is a novel target of miR-330

We further determined the relationship between miR-330 and Sp1. As the bioinformatic tool revealed, two putative target sites were identified, which starts from the 231 base position and the 4035 base position on the 3′UTR sequence respectively (Fig. 6A). Two 400 bp-long fragments around these target sites were cloned downstream of the firefly luciferase of pmirGLO Dual-Luciferase miRNA target expression vector, and termed pGL-WT1 and pGL-WT2, respectively. An individual plasmid with either miR-330 or the NC duplex was used to co-transfect HEK293T cells. The luciferase activity in the groups treated with miR-330 was deceased, when compared with the control; the combination of pGL-WT1 and miR-330 resulted in a more prominent change with the activity value ~40% (P<0.01) of the NC group than pGL-WT2 (P<0.05) (Fig. 6B). These findings indicate that miR-330 inhibited Sp1 expression through direct binding of the 3′-UTR of its transcript.

Figure 6 Specificity protein 1 (Sp1) is a novel target of miR-330. (A) The schematic illustration represents two potential miR-330-targeted sites on 3′ UTR of the Sp1 transcript, as predicted by the online bioinformatic tool Targetscan (http://www.targetscan.org/). The ‘231–237’ and ‘4035–4041’ sites are denoted as ‘Site 1’ and ‘Site 2’, respectively. (B) HEK293T cells were co-transfected with 50 nmol of either miR-330 or NC and 100 ng of pGL-WT1 or pGL-WT2, and 24 h later, cells were lysed and assessed for luciferase activity. The relative activity was represented as normalization of the firefly value to the Renilla value and expressed as means ± SD.

Discussion

As a major health concern among males, prostate cancer is currently the second leading cause of cancer-related death in Western countries. Although reports of improved survival rates of localized or organ-confined tumors are increasingly common, confusion still lies in the problem of dealing with metastasis. Recent studies of metastasis-associated miRNAs have provided us with more insights into cancer pathogenesis (16,17). Watahiki et al(18) compared the miRNA profiling of a transplantable metastatic vs. a non-metastatic prostate cancer xenograft, and discovered 53 pro-metastatic miRNAs and 68 anti-metastatic miRNAs, some of which have been widely accepted as metastic miRs including miR-21, miR-331-3p, miR-205 and miR-203.

miR-330, a recently identified miRNA, has been rarely investigated to date. Although it has been regarded as a downregulated gene in many types of cancer cells, the biological functions are reportedly diverse and contradictory (11,12). Here, we first reported the anti-migratory and anti-invasive role of miR-330 in prostate cancer. The relatively lower expression level of miR-330 in the cancer cell lines PC3 and DU145 when compared with the normal epithelial cell line RWPE was consistent with Lee et al(12). For forced expression, we used miR-330 mimics, the analog of miR-330, via transfectant Lipofectamine 2000 for the sake of accessibility, and qPCR validated the high efficiency of transfection which favorably persisted for 4–5 days (data not shown). As for the functional study, the anti-migration and invasion capacity of miR-330 was clearly demonstrated by the Transwell assay, while its association with apoptosis or cell cycle arrest was not observed in our study. For further mechanistic investigation, we screened Sp1 as the metastasis-associated gene possibly targeted by miR-330, based upon in silico analysis and the inverse relationship between them. qPCR and western blotting revealed post-transcriptional inhibition of Sp1 and a significant decrease in downstream genes MMP2 and MMP9 by miR-330 overexpression; these phenomena were mimicked by Sp1 knockdown, which further strengthened the hypothesis that miR-330 plays an anti-migratory and -invasive role via Sp1 in prostate cancer. Finally, the luciferase reporter assay was performed to determined Sp1 as the target gene, and even pinpoint the pairing site ‘231–237’ in the Sp1 3′UTR as the major target region.

The implication of Sp proteins in tumorigenesis has been widely accepted, based on the increasing number of reports revealing ubiquitous overexpression of these transcription factors in various neoplastic tissues (19–22). Sp1 regulates thousands of genes including oncogenes and tumor-suppressor genes, via its direct interaction with the components of the basal transcription machinery, TAF4, TAF7 and the recruitment of TBP/TFIID (23–26). In prostate cancer, Sp1 was reported to regulate key genes including the androgen receptor (AR), c-Met, FAS, FLIP and TGF. MMP2 and MMP9 are two intensively studied members of the extracellular matrix-degrading gelatinase family, whose regulation reportedly depends on Sp1 via interaction of their GC-rich sequences in promoters with specific domains of Sp1 (27–29). Here, we confirmed the pivotal role of Sp1 in cell migration and invasion, and demonstrated its close relationship with these matrix metalloproteinases.

It should be noted that Lee et al(12) reported the anti-proliferative and pro-apoptotic role of miR-330 in prostate cancer cells through MTT and plate colony formation assays. We performed an apoptosis assay for fear that proliferation-related factors may interfere with our invasion assay. As the results indicated, both apoptosis and cell cycle analyses yielded an insignificant difference within our observation period. Considering the complex signaling network of miRNAs, we proposed that the pro-apoptotic and anti-proliferative roles of miR-330 may manifest themselves in a latent and moderate manner.

In summary, we identified miR-330 as an anti-metastatic miRNA in prostate cancer. By targeting the 3′UTR of Sp1, miR-330 post-transcriptionally suppressed the expression of Sp1 protein, and further decreased the levels of MMPs, and finally inhibited migration and invasion of prostate cancer cells. Admittedly, our conclusions were derived solely from an in vitro study. Thus, our findings require further confirmation using clinical specimens and in vivo experiments. Yet, they suggest that miR-330 may be a therapeutic target for prostate cancer metastasis.

Acknowledgements

This study was supported by a grant from the Zhejiang Provincial Natural Science Foundation of China (LY12H05006), the National Key Clinical Speciality Construction Project of China, the Key Medical Disciplines of Zhejiang Province and Zhejiang Provincial Natural Science Foundation of China (Z2090356).

References