MiR-148a Attenuates Paclitaxel Resistance of Hormone-refractory, Drug-resistant Prostate Cancer PC3 Cells by Regulating MSK1 Expression*

MicroRNAs are involved in cancer pathogenesis and act as tumor suppressors or oncogenes. It has been recently reported that miR-148a expression is down-regulated in several types of cancer. The functional roles and target genes of miR-148a in prostate cancer, however, remain unknown. In this report, we showed that miR-148a expression levels were lower in PC3 and DU145 hormone-refractory prostate cancer cells in comparison to PrEC normal human prostate epithelial cells and LNCaP hormone-sensitive prostate cancer cells. Transfection with miR-148a precursor inhibited cell growth, and cell migration and invasion, and increased the sensitivity to anti-cancer drug paclitaxel in PC3 cells. Computer-aided algorithms predicted mitogen- and stress-activated protein kinase, MSK1, as a potential target of miR-148a. Indeed, miR-148a overexpression decreased expression of MSK1. Using luciferase reporter assays, we identified MSK1 as a direct target of miR-148a. Suppression of MSK1 expression by siRNA, however, showed little or no effects on malignant phenotypes of PC3 cells. In PC3PR cells, a paclitaxel-resistant cell line established from PC3 cells, miR-148a inhibited cell growth, and cell migration and invasion, and also attenuated the resistance to paclitaxel. MiR-148a reduced MSK1 expression by directly targeting its 3′-UTR in PC3PR cells. Furthermore, MSK1 knockdown reduced paclitaxel-resistance of PC3PR cells, indicating that miR-148a attenuates paclitaxel-resistance of hormone-refractory, drug-resistant PC3PR cells in part by regulating MSK1 expression. Our findings suggest that miR-148a plays multiple roles as a tumor suppressor and can be a promising therapeutic target for hormone-refractory prostate cancer especially for drug-resistant prostate cancer.

MicroRNAs (miRNAs) 2 are small non-coding RNAs composed of about 22-24 nucleotides and control protein expres-sion through translational inhibition or mRNA degradation by binding to the 3Ј-untranslated region (3Ј-UTR) of target mRNAs (1). miRNAs regulate a number of biological processes such as development, proliferation, differentiation, and apoptosis. Aberrant expression of miRNA has been reported in a variety of cancers, some of which have been shown to act as tumor suppressors or oncogenes (2).
MiR-148a expression is down-regulated in human breast cancer and undifferentiated gastric cancer (3,4). DNA methylation-associated silencing of miR-148 expression is identified in human cancer cell lines established from lymph node metastasis of colon, melanoma, and head and neck cancer, suggesting its role for the development of metastasis (5). Direct targets of miR-148a so far reported include transcription growth factor-␤-induced factor 2 (TGIF2), DNA (cytosine-5-)-methyltransferase 3␤ (DNMT3b) and pregnane X receptor (PXR) (5)(6)(7). However, the functional roles and target genes of miR-148a in prostate cancer have not yet been documented.
Mitogen-and stress-activated kinase 1 (MSK1), also known as ribosomal protein S6 kinase, 90kDa, polypeptide 5 (RPS6KA5), is a serine/threonine kinase that serves as a downstream target of extracellular signal-regulated kinase (ERK) or p38 mitogen-activated protein kinase in response to various stimuli including epidermal growth factor (EGF), phorbol ester (TPA), UV-irradiation, and anisomycin (8 -11). Activated MSK1 phosphorylates chromatin-related proteins such as histone H3 and HMG-14 and transcription factors such as CREB, ATF1, NF-B, and ER81 (8,(11)(12)(13)(14)(15). MSK1 is also involved in transcriptional activation of genes including immediate early genes c-fos and c-jun and inflammatory gene IL-6 (16 -23). Relevant to cancer is the report that MSK1 is required for tumor promoter-induced cell transformation of JB6 Cl41 mouse epidermal skin cells (24). The detailed mechanisms underlying regulation of MSK1 expression and its functional roles in cancer are poorly understood.
Most patients with prostate cancer initially respond to androgen ablation, but eventually the tumor becomes refractory to the treatment. Although patients with hormone-refractory prostate cancer are usually treated with taxane anti-cancer drugs such as docetaxel and paclitaxel, the clinical outcome is not satisfactory. There is an urgent need to develop novel treat-ments to enhance the chemosensitivity in patients with hormone-refractory, drug-resistant prostate cancer.
In the present study, we demonstrated that miR-148a is down-regulated in hormone-refractory prostate cancer cells compared with normal and hormone-sensitive cancer cells. Then, we investigated effects of ectopic expression of miR-148a on malignant phenotypes of hormone-refractory prostate cancer PC3 cells and found that miR-148a plays multiple tumor suppressive roles. We also identified MSK1 as a direct target of miR-148a, which, however, did not mediate miR-148a effects in PC3 cells. In paclitaxel-resistant PC3 cells, miR-148a inhibited malignant phenotypes including paclitaxel -resistance and reduced MSK1 expression via acting on its 3Ј-UTR. Finally, we proved that miR-148 attenuates paclitaxel-resistance by regulating MSK1 expression. We thus provided evidence that miR-148a has potential as a novel therapeutic target for treatment of hormone-refractory prostate cancer especially for drug-resistant prostate cancer.

EXPERIMENTAL PROCEDURES
Cell Culture-PrEC normal human prostate epithelial cells were obtained from Clonetics and cultured as recommended by the supplier. LNCaP, PC3, and DU145 human prostate cancer cells were purchased from ATCC and cultured in RPMI 1640 medium supplemented with 10% fetal bovine serum. Paclitaxelresistant cancer cells, PC3PR, were established from parental PC3 cells by stepwise increases of paclitaxel in the culture medium. First, PC3 cells were incubated with 5 nM of paclitaxel for 2 days. Then, the culture medium was changed to fresh medium without paclitaxel. After several passages with 5 nM paclitaxel, cells were exposed to 10 nM paclitaxel. The same procedure was repeated until resistant cells viable at 500 nM paclitaxel were obtained. HEK293 human embryonic kidney cells were obtained from ATCC and cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum.
RNA Extraction and Quantitative Reverse Transcriptase-PCR-Total RNA containing miRNA was extracted using the miRNeasy Mini kit (Qiagen, Hilden, Germany). Target miRNA was reverse transcribed to cDNA by a gene-specific RT primer using the TaqMan MicroRNA Reverse Transcription kit (Applied Biosystems, Foster City, CA). TaqMan MicroRNA Assay was performed with Premix Ex Taq (Takara, Shiga, Japan) using the Thermal Cycler Dice Real Time System (Takara). The relative quantification value of the target, normalized to a control, was calculated by the comparative Ct methods. 18 S rRNA was used as an internal control gene. To determine the MDR1 mRNA levels, semi-quantitative RT-PCR was performed using the High Capacity cDNA Reverse Transcription kit (Applied Biosystems) and Takara Taq (Takara).
Western Blot Analysis-Whole cell lysates were separated on SDS-PAGE and transferred to polyvinylidene difluoride membranes. After blocking with 5% skimmed milk for 1 h, membranes were incubated overnight at 4°C with a primary antibody in TBS-T and then reacted with a horseradish peroxidase-conjugated secondary antibody for 1 h. Immunoreactive proteins were detected with the ECL Plus Western blotting Detection System (GE Healthcare, Waukesha, WI) and visual-ized using the LAS-4000 Lumino-Image Analyzer (Fujifilm, Tokyo, Japan). Antibodies against MSK1 and MDR1 were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Anti-␤-actin was obtained from Sigma-Aldrich.
Transfection-Cells seeded onto 6-well plates were transfected with hsa-miR-148a precursor (Ambion, Austin, TX) or siGENOME SMARTpool Human MSK1 (Thermo Scientific Dharmacon, Waltham, MA) using Lipofectamine 2000 (Invitrogen, Carlsbad, CA) and cultured in Opti-MEM I reduced serum medium for 8 h. Then, culture medium was changed to normal medium. As a negative control, Pre-miR miRNA precursor negative control 1 and 2 (Ambion) or Stealth RNAi Negative Control Medium GC Duplex (Invitrogen) was used.
Cell Growth Assay-Transfected cells were re-seeded onto 12-well plates at 5 ϫ 10 4 cells/well and cultured for 5 days. The number of viable cells was determined using the trypan blue dye exclusion assay.
Colony Formation Assay-Transfected cells were re-seeded onto 6-well plates at 400 cells/well. Two weeks later, colonies were fixed with 100% methanol for 15 min and stained with crystal violet for 20 min. After taking photographs, the number of colonies with diameter more than 1.5 mm was counted.
Cell Migration Assay-Transfected cells in 6-well plates were cultured until cells reached confluency. Cell layers were wounded using a 1-ml tip and cultured for 24 h. Photographs were taken at time 0 and 24 h, and the wound area was determined using Photoshop software. The area of cell migration was determined by subtracting the area free of cells at 24 h from that at 0 h.
Cell Invasion Assay-Transfected cells were re-seeded onto Matrigel Invasion Chamber (Becton Dickinson, Franklin Lakes, NJ) at 1 ϫ 10 5 cells/well. Twenty-four hours later, cells invaded through the Matrigel membrane were fixed with 100% methanol and stained with crystal violet. After taking photographs, the number of invaded cells was counted.
Statistical Analysis-Statistical significance was determined by the Student's t test.

MiR-148a Was Down-regulated in PC3 and DU145 Cells-
Previous works (3,4) showed that miR-148a expression is down-regulated in several types of cancer. We investigated expression levels of miR-148a in PrEC normal prostate epithelial cell line, LNCaP hormone-sensitive prostate cancer cell line and PC3 and DU145 hormone-refractory prostate cancer cell lines. MiR-148a expression was significantly decreased in PC3 and DU145 cells compared with PrEC and LNCaP cells (Fig.  1A). We previously reported that miR-34a expression is downregulated in PC3 and DU145 cells because of functional defects of tumor suppressor gene p53 (25). Because the expression pattern of miR-148a among these four cell lines was similar to that of miR-34a, we tested if p53 could regulate the expression of miR-148a. In PC3 cells, over-expression of p53 resulted in upregulation of miR-34a but not miR-148a, indicating that miR-148a expression is likely to be regulated by p53-independent mechanisms (Fig. 1B).
Ectopic Expression of miR-148a Inhibited Cell Growth in PC3 Cells-To study whether decreased expression of miR-148a contributes to malignant phenotypes of prostate cancer cells, we first evaluated effects of ectopic expression of miR-148a on cell growth in PC3 cells. Cell growth assay revealed that miR-148a significantly inhibited cell growth at late phase (days 4 and 5 after transfection) ( Fig. 2A). Phase contrast microscopic analysis showed that intercellular contacts appeared to be inhibited at late phase in PC3 cells transfected with miR-148a precursor (Fig. 2B). To further confirm the ability of miR-148a to inhibit cell growth, we performed colony formation assay and found that the number of colonies formed was reduced in PC3 cells transfected with miR-148a compared with those transfected with a control precursor (Fig. 2C). These results suggest that miR-148a inhibits cell growth of PC3 cells.

Ectopic Expression of miR-148a Attenuated Cell Migration and Invasion in PC3
Cells-It has been recently reported that miR-148a inhibits cell motility of cell lines established from lymph node metastasis of colon, melanoma, and head and neck cancer (5). Here we examined whether ectopic expression of miR-148a could inhibit cell migration of PC3 cells using cell migration assay. The area of migration during 24 h was lower in PC3 cells expressing miR-148a than in those expressing a negative control precursor (Fig. 3A). To evaluate miR-148a effects on the invasive ability, we performed the invasion assay using the Matrigel Invasion Chamber. The number of invaded cells was significantly decreased in PC3 cells transfected with miR-148a precursor compared with those transfected with a negative control precursor (Fig. 3B). These results suggest that ectopic expression of miR-148a attenuates cell migration and invasion of PC3 cells.
Ectopic Expression of miR-148a Increased the Sensitivity to Paclitaxel of PC3 Cells-To our knowledge, little is known about the effects of miR-148a on the sensitivity to anti-cancer drugs. We studied whether the sensitivity to paclitaxel could be altered by miR-148a in PC3 cells. In the absence of paclitaxel, miR-148a showed no effect on proliferation of PC3 cells at the time points examined (Fig. 3C, left panel), which was consistent with the data shown in Fig. 2A. In the presence of 20 nM paclitaxel at which concentration no cell growth was observed, miR-148a decreased cell viability of PC3 cells (Fig. 3C, right panel), which was associated with increased apoptosis as determined by Hoechst 33342 staining (data not shown). These results show that miR-148a can increase the sensitivity to paclitaxel of hormone-refractory PC3 cells.
MSK1 Was a Direct Target of miR-148a in PC3 Cells-To identify a direct target of miR-148a, we searched for potential targets of miR-148a using the TargetScan software. MSK1 was one of the predicted targets with the highest score, which contained three putative miR-148a binding sites within its 3Ј-UTR (Fig. 4A). Other computer-aided algorithms, PicTar and miRanda also predicted MSK1 as a putative target of miR-148a. Furthermore, DNA sequences of three potential miR-148a binding sites predicted within the human MSK1 3Ј-UTR were highly conserved among species including chimpanzee and rhesus (Fig. 4B). These in silico data suggested that three putative binding sites could be involved in regulation of MSK1 expression by miR-148a. Western blot analysis showed that the expression levels of miR-148a in LNCaP, PC3, DU145 cells were higher than that in PrEC cells (Fig. 4C). Combined with the data shown in Fig. 1A, there was an inverse correlation between the expression of MSK1 and miR-148a in PrEC, PC3, and DU145 cells except for LNCaP cells. In LNCaP cells, MSK1 expression may be also regulated by mechanisms other than those involving miR-148a. As shown in Fig. 4D, miR-148a precursor significantly decreased the MSK1 protein level in PC3 cells on days 2, 3, 4, and 5 after transfection, suggesting that MSK1 could be a direct target of miR-148a. To prove this possibility, we constructed a firefly luciferase reporter gene containing the MSK1 3Ј-UTR with three potential target sites (MSK1 3Ј-UTR WT) and a mutant version of the reporter gene in which all these sites were mutated (MSK1 3Ј-UTR MT). Introduction of miR-148a precursor into PC3 cells decreased relative luciferase activity of the reporter gene containing the MSK1 3Ј-UTR WT compared with a control precursor, while that of the reporter gene containing the MSK1 3Ј-UTR MT was not affected by miR-148a (Fig. 5A). Similar results were obtained in other cell line, HEK293 human embryonic kidney cells (Fig. 5B). To determine contribution of each of three putative miR-148a binding sites (site 1, 2, or 3) to the MSK1 3Ј-UTR activity, we generated MSK1 3Ј-UTR firefly luciferase reporter genes in which each site was mutated (MSK1 3Ј-UTR MT1, MT2, or MT3, respectively). Suppression of luciferase activity by miR-148a was seen with the MSK1 3Ј-UTR MT2 reporter gene, which was almost similar to that with the WT reporter gene (Fig. 5C). However, miR-148 showed little or no effects on luciferase activity with the MT1 and MT3 reporter genes. In all,

MiR-148a Attenuates Paclitaxel Resistance by MSK1 Regulation
these results suggest that miR-148a may directly bind to the sites 1 and 3 of the MSKI 3Ј-UTR and inhibit its protein expression.
Knockdown of MSK1 Showed Little or No Effect on Malignant Phenotypes of PC3 Cells-To investigate whether effects of miR-148a on various malignant phenotypes of PC3 cells could be mediated by MSK1, we performed its knockdown experiments. As shown in Fig. 6A, MSK1 siRNA almost completely diminished its protein expression in PC3 cells on days 2, 3, 4, and 5 after transfection. Cell growth assay showed that introduction of MSK1 siRNA inhibited cell growth on days 2 (p Ͻ 0.01), 3 (p Ͻ 0.01), and 4 (p Ͻ 0.05) after transfection (Fig. 6B), but the inhibitory effect was rather marginal. Also, MSK1 knockdown had little or no effects on colony formation, cell migration and invasion, and the sensitivity to paclitaxel of PC3 cells (supplemental Fig. S1, A-D, respectively). Taken together, it is suggested that MSK1 does not play an obvious role in mediating the miR-148a effects in PC3 cells.
Establishment of Paclitaxel-resistant Cells from PC3 Cells-To study the roles of miR-148a on the resistance to anti-cancer drug, we established a PC3PR paclitaxel-resistant cell line from PC3 cells. The IC 50 of PC3PR cells was 2577.3 nM, whereas that of parental PC3 cells was 8.6 nM (Fig. 7A), indicating that the resistance to paclitaxel was markedly increased in PC3PR cells. The expressions of MDR1 protein and mRNA, which is involved in the export of drugs and drug resistance, were significantly increased in PC3PR cells compared with parental PC3 cells (Fig. 7B).
Ectopic Expression of miR-148a Attenuated Malignant Phenotypes of PC3PR Cells Including Paclitaxel Resistance-Here we tested if miR-148a could affect malignant phenotypes of PC3PR cells. Introduction of miR-148a precursor inhibited cell growth, colony formation, and cell migration and invasion of PC3PR cells (Fig. 8, A-D, respectively). In the absence of paclitaxel, miR-148a inhibited proliferation of PC3PR cells (Fig. 8E, left panel), which was consistent with the data shown in Fig. 8A. In the presence of 2 M paclitaxel, miR-148a decreased cell viability of PC3PR cells (Fig. 8E, right  panel), which was accompanied by increased apoptosis (data not shown). These results show that miR-148a can attenuate  1, 2, and 3). The position of the binding sites was numbered relative to the first nucleotide of the 3Ј-UTR. Mutations were introduced into the seed regions of all three binding sites as indicated by underline. B, interspecies conservation of putative miR-148a binding sites within the MSK1 3Ј-UTR is shown. Open box denotes perfectly conserved sequences between humans and other species. C, cell lysates were extracted from PrEC, LNCaP, PC3, and DU145 cells and subjected to Western blot analysis for MSK1. D, PC3 cells were transfected with a negative control or miR-148a precursor (10 nM) and cultured for 2, 3, 4, and 5 days. Cell lysates was harvested and subjected to Western blot analysis for MSK1 and ␤-actin.

malignant phenotypes of PC3PR cells including paclitaxel resistance.
Knockdown of MSK1 Attenuated Paclitaxel Resistance of PC3PR Cells-Finally, we investigated if miR-148a effects might be mediated by MSK1. As shown in Fig. 9A, MSK1 expression in PC3PR cells was higher than that in PC3 cells (left panel), but miR-148a expression was not decreased in PC3PR cells compared with PC3PR cells (right panel). These results suggest that up-regulation of MSK1 expression in PC3PR cells is not caused by miR-148a down-regulation, but by other unidentified mechanisms. However, ectopic expression of miR-148a decreased MSK1 protein levels (Fig. 9B) via acting on its 3Ј-UTR (Fig. 9C). MSK1 knockdown exhibited no effects on cell growth, colony formation, and cell migration and invasion of PC3PR cells (supplemental Fig. S2, A-D,  respectively). In the absence of paclitaxel, MSK1 knockdown showed no effect on proliferation of PC3PR cells (Fig. 9D,  left panel). In the presence of 2 M paclitaxel, miR-148a decreased cell viability of PC3PR cells (Fig. 9D, right panel). Collectively, these results indicate that miR-148a attenuates paclitaxel resistance of PC3PR cells through regulating MSK1 expression.

DISCUSSION
MiRNAs act as tumor suppressors or oncogenes in a variety of cancers including prostate cancer (26 -32). However, the functional roles of miRNAs in hormone-refractory prostate cancer are largely unknown. MiRNA expression profiling of prostate cancer tissues and cell lines has been recently documented (33)(34)(35)(36). Microarray data reported by Mattie et al. (33) revealed that miR-148a expression was lower in advanced pros-tatic tumor (Gleason score 8), prostatic lymph node metastasis, and PC3 cells than in pooled normal adjacent to tumor sample, transitional cell metaplasia and LNCaP cells. Porkka et al. (34) identified differentially expressed miRNAs between prostate carcinoma samples and benign prostatic hyperplasia (BPH), which showed that miR-148a was down-regulated in hormonerefractory carcinomas compared with BPH. Our microarray data also demonstrated that the levels of miR-148a in hormonerefractory PC3 and DU-145 cells were lower than those in PrEC normal prostate epithelial cells and hormone-sensitive LNCaP cells. 3 In an attempt to identify miRNA that behaves as a tumor suppressor and possesses potential as a therapeutic target for hormone-refractory prostate cancer, we focused on miR-148a in the present study. Using quantitative RT-PCR, we demonstrated that miR-148a expression was down-regulated in hormone-refractory PC3 and DU145 cells as compared with PrEC normal prostate epithelial cells and hormone-sensitive LNCaP cells. As revealed by transfection studies, the mechanism underlying the suppression of miR-148a expression did not involve defects of p53 in PC3 cells. In primary breast cancer specimens and cancer cell lines derived from lymph node metastasis, aberrant hypermethylation of the CpG islands has been found upstream of the miR-148a gene (3,5). It is thus possible that hypermethylation of the miR-148a promoter contributes to decreased expression of miR-148a in PC3 cells, which remains to be studied.  Introduction of miR-148a into head and neck cancer SIHN-011B cells induces inhibition of cell motility in vitro and also induces reduction of tumor growth and inhibition of metastasis formation in xenograft models (5). In the present study, we demonstrated that ectopic expression of miR-148a resulted in inhibition of cell growth, and cell migration and invasion in PC3 cells, suggesting that dysregulation of miR-148a expression may contribute to metastatic potential of PC3 cells.
DNA methyltransferases play an important role in mediating aberrant methylation patterns in human cancer cells, which is implicated in tumor initiation and progression (37). It has been recently reported that miR-148a directly targets and down-regulates DNA (cytosine-5-)-methyltransferase 3␤, DNMT3b (6). In PC3 cells, DNMT3b knockdown results in inhibition of cell growth, increase of apoptotic cells, and reduction of cell migration, whereas the invasive ability is not affected (38). These previous reports suggest that effects of miR-148a on cell growth and cell migration but not cell invasion may be mediated in part by down-regulation of DNMT3b in PC3 cells.   4). B, PC3PR cells were transfected with a negative control or miR-148a precursor (10 nM) and cultured for 2 days. Cell lysates were harvested and subjected to Western blot analysis for MSK1. C, MSK1 3Ј-UTR WT or MSK1 3Ј-UTR MT firefly luciferase reporter gene (20 ng) was transfected into PC3PR cells along with the Renilla luciferase expressing plasmid (20 ng) and either a negative control or miR-148a precursor (30 nM). Luciferase assay was performed 24 h after transfection. Relative luciferase activity was expressed as a ratio of firefly/Renilla luciferase activity. Data are expressed as mean Ϯ In addition to cell growth, and cell migration and invasion, we investigated effects of miR-148a on the sensitivity to anticancer drug paclitaxel. Introduction of miR-148a increased the chemosensitivity of PC3 cells, indicating that miR-148a can augment the sensitivity to paclitaxel in hormone-refractory PC3 cells that are sensitive to the anti-cancer drug. In clinical practice, it is of great importance to ameliorate the resistance of drug-resistant prostate cancer. To test the possibility that miR-148a could attenuate such a drug resistance, we established PC3PR paclitaxel-resistant cells from PC3 cells. PC3PR cells could survive in the presence of 500 nM paclitaxel and the IC 50 was higher by 300-fold in PC3PR cells than in parental PC3 cells. The expression level of MDR1 was markedly increased in PC3PR cells compared with PC3 cells. Transfection with miR-148a precursor attenuated paclitaxel resistance of PC3PR cells. Furthermore, miR-148a inhibited cell growth, and cell migration and invasion in PC3PR cells as observed in PC3 cells. Taken all together, our findings suggest that miR-148a can inhibit malignant phenotypes of hormone-refractory, drug-sensitive, and -resistant prostate cancer cells including paclitaxel resistance of drug-resistant cells.
MSK1 is a member of the p90 ribosomal protein S6 kinase family and is activated by Ras-MAPK and p38 stress kinase pathways (8 -11). MSK1 is implicated in TPA-and EGF-induced oncogenesis (24). In PC3 cells, RSK2, another member of the p90 ribosomal protein S6 kinase family, is up-regulated and involved in cell growth (39). We demonstrated that MSK1 is a direct target of miR-148a in both PC3 and PC3PR cells and hypothesized that some of the tumor suppressive effects of miR-148a might be mediated through down-regulation of MSK1 expression. In PC3 cells, suppression of MSK1 expression by siRNA showed little or no effects on cell growth, cell migration, and invasion and the sensitivity to paclitaxel. In PC3PR cells, MSK1 expression was elevated compared with PC3 cells, which, however, was not associated with miR-148a down-regulation. Intriguingly, despite the lack of effects on paclitaxel-sensitivity of PC3 cells, MSK1 knockdown decreased paclitaxel-resistance of PC3PR cells. These results imply that up-regulation of MSK1 expression and activation of the MSK1 signaling pathway may be involved in the development of paclitaxel-resistance in PC3PR cells. Nevertheless, additional direct target genes that mediate tumor suppressive effects of miR-148a as well as molecular mechanisms underlying attenuation of paclitaxel-resistance by MSK1 knockdown remain to be elucidated.
In conclusion, we demonstrated that miR-148a is down-regulated in PC3 and DU145 hormone-refractory prostate cancer cells and that miR-148a inhibits cell growth and cell migration and invasion in both PC3 and PC3PR cells. MiR-148a also augmented the chemosensitivity of PC3 cells and attenuated the chemoresistance of PC3PR cells. Thus, miR-148a plays multiple tumor suppressive roles in hormone-refractory, drug-sensitive and -resistant prostate cancer cells. Furthermore, we proved that MSK1 is a novel target gene of miR-148a in both PC3 and PC3PR cells and that miR-148 attenuates paclitaxelresistance of PC3PR cells by modulating MSK1 expression. Our findings suggest that miR-148a is a promising therapeutic tar-get for hormone-refractory prostate cancer especially for drugresistant prostate cancer.