Introduction

Prostate cancer (PCa) is the most frequently diagnosed cancer and the second leading cause of cancer death among men in developed countries.1 Most patients are initially responsive to androgen-deprivation therapy, but their cancers eventually become resistant to androgen-deprivation therapy and progress to castration-resistant prostate cancer (CRPC). Currently, CRPC is difficult to treat, and most clinical trials for advanced PCa have shown limited benefits, with disease progression and metastasis to the bone or other sites.2, 3 Therefore, understanding the molecular mechanisms of CRPC and metastatic pathways underlying PCa using currently available genomic approaches would help to improve therapies for and prevention of the disease.

The discovery of non-coding RNAs (ncRNAs) in the human genome was an important conceptual breakthrough in the post-genome sequencing era.4 Improving our understanding of ncRNAs is necessary for continued progress in cancer research. MicroRNAs (miRNAs) are endogenous small ncRNA molecules (19–22 bases in length) that regulate protein-coding gene expression by repressing translation or cleaving RNA transcripts in a sequence-specific manner. Currently, 2042 human mature miRNAs are registered at miRBase release 19.0 (http://microrna.sanger.ac.uk/). miRNAs are unique in their ability to regulate multiple protein-coding genes. Bioinformatic predictions indicate that miRNAs regulate >30–60% of the protein-coding genes in the human genome.5, 6

Emerging evidence has demonstrated that miRNAs are aberrantly expressed in many human cancers and have significant roles in the initiation, development and metastasis of these cancers.7, 8 It is believed that normal regulatory mechanisms can be disrupted by the aberrant expression of tumor suppressive or oncogenic miRNAs in cancer cells. Therefore, identification of aberrantly expressed miRNAs is the first step toward elucidating miRNA-mediated oncogenic pathways. Based on this, we identified miRNA expression signature in PCa using clinical specimens and found that miR-145 expression was significantly downregulated in PCa.9 We also reported that miR-145 suppresses cell proliferation, migration and invasion by targeting FSCN1.10, 11 Other studies have reported that the miR-143/145 cluster is significantly downregulated in PCa and have suggested its important association with oncogenesis and bone metastasis in PCa.12, 13

The aim of the present study was to investigate the functional significance of the miR-143/145 cluster and to identify the molecular targets regulated by this cluster in PCa cells. Genome-wide gene expression data and in silico database analysis showed that Golgi membrane protein 1 (GOLM1) was a promising candidate target of the miR-143/145 cluster. GOLM1, also known as Golgi phosphoprotein 2 (GOLPH2) and Golgi protein 73 (GP73), is a newly identified Golgi-associated protein. Although the functional roles of GOLM1 are currently unclear, GOLM1 has been shown to interact with other proteins through its transmembrane protein domain.14 The overexpression of GOLM1 has been reported in several types of cancers, including PCa.15, 16, 17 Elucidation of the molecular targets regulated by the tumor-suppressive miR-143/145 cluster will provide new insights into the potential molecular mechanisms of PCa oncogenesis and metastasis and will facilitate the development of novel therapeutic strategies for the treatment of PCa.

Materials and methods

Clinical prostate specimens

Clinical specimens were obtained from patients at the Teikyo University Chiba Medical Center from 2008 to 2013. All the patients had elevated levels of prostate-specific antigen and had undergone transrectal prostate needle biopsy. Non-cancerous prostate tissue (n=24) was obtained from patients who were negative for malignancy without indurations on the prostate. Prostate tumor tissues (n=28) contained 90–100% malignant cells in the biopsy cores. Patients’ characteristics are shown in Table 1. Before tissue collection, all patients provided written informed consent of tissue donation for research purposes. The protocol was approved by the Institutional Review Board of Teikyo University.

Table 1 Patient characteristics
Full size table

Cell culture

PC3 and DU145 cells, human PCa cells obtained from the American Type Culture Collection (Manassas, VA, USA), were maintained in RPMI-1640 medium supplemented with 10% fetal bovine serum in a humidified atmosphere of 5% CO2 and 95% air at 37 °C.

RNA extraction

Total RNA was isolated using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s protocol. RNA concentrations were determined spectrophotometrically, and molecular integrity was checked by gel electrophoresis. RNA quality was confirmed using an Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA).

Quantitative real-time reverse transcriptase–PCR (RT-PCR)

First-strand cDNA was synthesized from 1 μg total RNA using a High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, USA). Gene-specific PCR products were assayed continuously using a 7900-HT Real-Time PCR System (Applied Biosystems) according to the manufacturer’s protocol. The initial PCR step consisted of a 10-min hold at 95 °C, followed by 40 cycles consisting of a 15-s denaturation at 95 °C and a 1-min annealing/extension at 63 °C. TaqMan probes and primers for GOLM1 (P/N: Hs00213061_m1) and GAPDH (P/N: Hs02758991_g1) as an internal control were obtained from Applied Biosystems (Assay-On-Demand Gene Expression Products). The expression levels of miR-143 (Assay ID: 002249) and miR-145 (Assay ID: 002278) were analyzed by TaqMan quantitative real-time PCR (TaqMan MicroRNA Assay; Applied Biosystems) and normalized to the expression of RNU48 (Assay ID: 001006). The delta delta Ct method was used to calculate the relative quantities of target genes. All reactions were performed in triplicate, and each assay included negative control reactions that lacked cDNA.

Transfections with mature miRNA and small interfering RNA (siRNA)

The following mature miRNA species were used in this study: Pre-miR miRNA Precursor for hsa-miR-143 (Product ID: PM10883) and hsa-miR-145 (Product ID: PM11480; Applied Biosystems). The following siRNAs were used: Stealth Select RNAi siRNA, si-GOLM1 (Cat no. HSS181966 and HSS181968; Invitrogen), and negative control miRNA/siRNA (P/N: AM17111; Applied Biosystems). RNAs were incubated with OPTI-MEM (Invitrogen) and Lipofectamine RNAiMax reagent (Invitrogen) as described previously. The transfection efficiency of miRNA in PC3 and DU145 cells was confirmed based on downregulation of TWF1 (PTK9) mRNA following transfection with miR-1 as previously reported.

Cell proliferation, migration and invasion assays

Cells were transfected with 10 nM miRNA or siRNA by reverse transfection and plated in 96-well plates at 3 × 103 cells per well. After 72 h, cell proliferation was determined with the XTT assay using a Cell Proliferation Kit II (Roche Molecular Biochemicals, Mannheim, Germany) as previously reported.18, 19

Cell migration was evaluated with a wound-healing assay. Cells were plated in six-well plates, and the cell monolayers were scraped using a P-20 micropipette tip. The initial gap length (0 h) and the residual gap length at 24 h after wounding were calculated from photomicrographs.

A cell invasion assay was carried out using modified Boyden chambers containing Transwell-precoated Matrigel membrane filter inserts with 8-μm pores in 24-well tissue culture plates (BD Biosciences, Bedford, MA, USA) at 2 × 105 cells per well. Cells were transfected with 10 nM miRNA or siRNA by reverse transfection and plated in 10-cm dishes at 8 × 105 cells per dish. After 48 h, the cells were collected, and 2 × 105 cells were added to the upper chamber of each migration well. Cells were allowed to invade for 48 h. After gentle removal of the non-migratory cells from the filter surface of the upper chamber, the cells that invaded into the lower chamber were fixed and stained with Diff-Quick (Sysmex Corporation, Kobe, Japan). The number of cells that migrated to the lower surface was determined microscopically by counting four areas of constant size per well.

All experiments were performed in triplicate.

Western blotting

Cells were harvested 72 h after transfection, and lysates were prepared. Fifty micrograms of protein lysate was separated on Mini-PROTEAN TGX gels (Bio-Rad, Hercules, CA, USA) and transferred to polyvinylidene difluoride membranes. Immunoblotting was performed with rabbit anti-GOLM1 antibodies (1:500; HPA010638, Sigma-Aldrich, St Louis, MO, USA) and anti-GAPDH antibodies (1:1000; ab8245, Abcam, Cambridge, UK), used as an internal loading control. Membranes were washed and incubated with anti-mouse immunoglobulin G horseradish peroxidase-linked antibodies (No. 7076, Cell Signaling Technology, Danvers, MA, USA). Complexes were visualized with an Immun-Star WesternC Chemiluminescence Kit (Bio-Rad), and the expression levels of these genes were evaluated by the ImageJ software provided by National Institutes of Health (NIH), Bethesda, MD, USA (ver.1.44; http://rsbweb.nih.gov/ij/).

Genome-wide gene expression and in silico analyses for the identification of genes regulated by the miR-143/145 cluster

To gain further insight into which genes were affected by the miR-143/145 cluster, we performed a combination of in silico and genome-wide gene expression analyses. First, genes regulated by the miR-143/145 cluster were listed using the TargetScan database as described previously.18, 19 To investigate the expression status of candidate of miR-143/145 cluster-target genes in PCa clinical specimens, we examined the gene expression profiles in the GEO database (accession number: GSE29079). In this expression study, tissue samples from 47 PCa with clinical high-risk tumors were included, and none of the patients had been treated with neo-adjuvant treatments, such as radiation, cytotoxic and endocrine therapies.20

To identify molecular pathways regulated by miR-143/145 cluster in PCa, upregulated genes which were targeted by miR-143/145 cluster were analyzed in terms of Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway categories using the GENECODIS program (http://genecodis.cnb.csic.es/).

Plasmid construction and dual-luciferase reporter assay

Partial wild-type sequences of the GOLM1 3′ untranslated region (UTR) or those with a deleted miR-143 target site (position 1610–1617 of the GOLM1 3′UTR) and deleted miR-145 target sites (positions 5–12 and 260–267 of the GOLM1 3′UTR) were inserted between the XhoI–PmeI restriction sites in the 3′UTR of the hRluc gene in the psiCHECK-2 vector (C8021; Promega, Madison, WI, USA). The protocol for vector construction was described previously.18, 19 Sequences of the oligonucleotides are described in Supplementary Table S1. The synthesized DNA was cloned into the psiCHECK-2 vector. PC3 cells were transfected with 50 ng of the vector and 10 nM miR-143 or miR-145 using Lipofectamine 2000 (Invitrogen). The activities of firefly and renilla luciferases in cell lysates were determined with a dual-luciferase assay system (E1910; Promega). Normalized data were calculated as the ratio of renilla/firefly luciferase activities.

Immunohistochemistry

A tissue microarray of 39 prostate tumors and 8 normal prostate tissues was obtained from US Biomax, Inc. (PR956; Rockville, MD, USA). Detailed information on all tumor specimens can be found at http://www.biomax.us/index.php. The patients’ backgrounds and clinicopathological characteristics are summarized in Supplementary Table S2. The tissue microarray was immunostained with an UltraVision Detection System (Thermo Scientific) following the manufacturer’s protocol. Primary rabbit polyclonal antibodies against GOLM1 (Sigma-Aldrich) were diluted 1:500. The slides were treated with biotinylated goat antibodies. Diaminobenzidine hydrogen peroxidase was the chromogen, and counterstaining was performed with 0.5% hematoxylin. Immunostaining was evaluated according to a scoring method described previously.18, 21 Each case was scored on the basis of the intensity and area of staining. The intensity of staining was graded on the following scale: 0, no staining; 1+, mild staining; 2+, moderate staining; and 3+, intense staining. The area of staining was evaluated as follows: 0, no staining of cells in any microscopic fields; 1+, <30% of cells stained positive; 2+, 30–60% of cells stained positive; 3+, >60% of cells stained positive. A combined staining score (intensity + extension) of <2 indicated low expression, a score between 3 and 4 indicated moderate expression, and a score between 5 and 6 indicated high expression.

Statistical analysis

The relationships between two groups and the numerical values obtained by real-time RT-PCR were analyzed using the paired t-test. The relationship among three variables and numerical values was analyzed using the Bonferroni-adjusted Mann–Whitney U test. All analyses were performed using Expert StatView software (version 4, SAS Institute Inc., Cary, NC, USA).

Results

The expression levels of miR-143 and miR-145 in PCa specimens

We evaluated the expression levels of miR-143 and miR-145 in normal prostate tissues (n=24) and PCa tissues (n=28). Prostate-specific antigen levels in patients with normal prostate tissues were measured to have a median value of 8.7 ng ml−1 (range 4.3–22 ng ml−1), whereas PCa patients had high prostate-specific antigen values, with a median of 182 ng ml−1 (3.4–3750 ng ml−1). PCa patients had progressive disease with N1 (67.8%) and M1 (64.2%) according to TNM (Tumor, Node, Metastatis) classification (Table 1).

The expression levels of miR-143 and miR-145 were determined by RT-PCR. The expression levels of miR-143 (Figure 1a) and miR-145 (Figure 1b) were significantly lower in PCa tissues compared with normal tissues (P=0.00148 and P<0.0001, respectively). The expression of miR-143 was highly correlated with that of miR-145 (correlation coefficient, 0.898; P<0.0001; Figure 1c).

Figure 1
figure 1

Expression levels of miR-143 and miR-145. (a, b) miR-143 and miR-145 expression levels in clinical PCa specimens and PCa cell lines (PC3 and DU145) compared with non-cancerous tissues. RNU48 was used as the internal control. (c) Correlation between the relative expression level of miR-143 (x-axis) and that of miR-145 (y-axis). Levels are plotted as a scatterplot.

Full size image

Effects of miR-143 and miR-145 transfection on cell proliferation, migration and invasion in PCa cells

Next, the expression levels of miR-143 and miR-145 in DU145 and PC3 cells were determined by RT-PCR and were compared with those of normal prostate tissues. miR-143 and miR-145 were significantly downregulated in DU145 and PC3 cells as compared with normal prostates (Figures 1a and b).

To examine the functional roles of miR-143 and miR-145, we performed gain-of-function studies using miRNA transfection into DU145 and PC3 cells. XTT assays revealed significant inhibition of cell proliferation in DU145 and PC3 cells transfected with miR-143 and miR-145 in comparison with mock-transfected cells and control transfectants (cell proliferation, percentage of mock: DU145 cells 68.8%±3.7%, 53.4%±1.1%, 100.0%±1.3% and 93.7%±1.7%, respectively, P<0.0001; PC3 cells 90.4%±5.7%, 60.0%±1.2%, 100.0%±3.3% and 103.2%±4.0%, respectively, P=0.0969 and P<0.0005; Figure 2a).

Figure 2
figure 2

Effects of transfection with mature miR-143 and miR-145 on RCC cell lines. (a) Cell proliferation was determined with XTT assays in PC3 and DU145 cell lines after 72 h of transfection with 10 nM miR-143, miR-145, miR-control or mock transfection. (b) Cell migration activity was determined by the wound-healing assay. (c) Cell invasion activity was determined by Matrigel invasion assay. *P<0.01; **P<0.0001.

Full size image

Moreover, wound-healing assays demonstrated significant inhibition of cell migration in DU145 and PC3 cells transfected with miR-143 and miR-145 in comparison with mock-transfected cells and control transfectants (cell migration, percentage of mock: DU145 cells 64.2%±5.2%, 59.0%±10.0%, 100.0%±7.1% and 88.7%±8.8%, respectively, P=0.0047 and P=0.0016; PC3 cells 68.9%±4.5%, 10.7%±4.9%, 100.0%±5.8% and 103.6%±7.7%, respectively, P=0.0013 and P<0.0001; Figure 2b).

Similarly, Matrigel invasion assays revealed that transfection with these miRNAs reduced cell invasion. Indeed, the number of invading cells was significantly decreased in DU145 and PC3 cells transfected with miR-143 and miR-145 in comparison with mock-transfected cells and control transfectants (percentage of invading cells: DU145 cells 59.7%±7.3%, 12.1%±2.8%, 100.0%±4.9% and 114.1%±3.3%, respectively, P<0.0001; PC3 cells 90.8%±3.2%, 17.7%±1.9%, 100.0%±6.7% and 103.6%±4.2%, respectively, P=0.1539 and P<0.0001; Figure 2c).

Identification of targets genes regulated by the miR-143/145 cluster in PCa

To gain further insight into the molecular mechanisms and pathways regulated by the tumor-suppressive miR-143/145 cluster in PCa, we performed a combination of gene expression and in silico analyses. The strategy of selection of miR-143/145 cluster-regulated genes is shown in Figure 3. The TargetScan program showed that 1053 genes had putative target sites for both miR-143 and miR-145 in their 3′UTR regions. To confirm the expression levels of these genes in clinical prostate tissues, GEO database (GEO accession number: GSE29079) analysis was performed. Among the 1053 genes, 347 genes were upregulated in 47 PCa specimens compared with 48 non-cancerous prostate tissues. Expression data for the 44 genes (>1.5 fold change) are summarized in Table 2, and the other upregulated genes are shown in Supplementary Table S2.

Figure 3
figure 3

Flow chart of the strategy for analysis of miR-143/145 cluster target genes. A total of 1053 genes were selected as putative miR-143/145 cluster target genes by TargetScan database analysis. We then analyzed the expression levels of these 1053 genes by using available data sets of GEO (GSE 29079). The analyses showed that 44 genes were significantly upregulated in PCa specimens compared with normal specimens.

Full size image
Table 2 Identification of target genes regulated by the miR-143/145 cluster in PCa
Full size table

Furthermore, those 347 genes were classified KEGG pathways using GeneCodis program to identify molecular pathways regulated by miR-143/145 cluster in PCa. Four signaling pathways (endocytosis, oocyte meiosis, protein processing in endoplasmic reticulum and ubiquitin-mediated proteolysis) were identified in this analysis (Supplementary Tables S3A–E).

GOLM1 as a target of post-transcriptional repression by the miR-143/145 cluster in PCa cells

The mRNA and protein expression levels of GOLM1 were markedly downregulated in miR-143- and miR-145-transfected PC3 and DU145 cells in comparison with mock-transfected cells and control transfectants (Figures 4a and b). Therefore, we next performed a luciferase reporter assay to determine whether GOLM1 mRNA had functional target sites for miR-143 and miR-145. We used three vectors encoding the partial sequences of the 3′UTR of GOLM1 mRNA (Figure 4c, Supplementary Table S1). The luminescence intensity was significantly decreased in the presence of two sites targeted by miR-143 or miR-145 (positions 1610–1617 and 5–12, respectively; Figure 4c, middle and lower panels). In contrast, the luminescence intensity was not decreased by transfection with miR-145 at position 260–267 or with deleted sequences of either miRNA (Figure 4c, middle panel). These data suggested that miR-143 and miR-145 bound directly to specific sites in the 3′UTR of GOLM1 mRNA.

Figure 4
figure 4

GOLM1 expression was suppressed by miR-143 and miR-145 transfection at both the mRNA and protein levels in DU145 and PC3 cells. (a) GOLM1 mRNA expression 72 h after transfection with miR-143 and miR-145. GAPDH expression was used for normalization. **P<0.0001. (b) GOLM1 protein expression 72 h after transfection with miR-143 and miR-145. GAPDH was used as a loading control. (c) miR-143 and miR-145 binding sites in the 3′UTR of GOLM1 mRNA. Luciferase reporter assay using the three vectors encoding putative miR-145 target sites at positions 5–12 and 260–267 and the putative miR-143 target site at position 1610–1617 for both wild type (WT) and deletion (DEL). Renilla luciferase values were normalized to firefly luciferase values. *P<0.01.

Full size image

Effects of GOLM1 knockdown on cell proliferation, migration and invasion in PCa cell lines

The expression of GOLM1 mRNA was upregulated in DU145 and PC3 cells compared with normal prostate tissues (data not shown). Therefore, to examine the functional role of GOLM1 in PCa, we performed loss-of-function studies using transfection with two different siRNA sequences targeting GOLM1 in DU145 and PC3 cells. The expression of GOLM1 mRNA and protein was markedly repressed in si-GOLM1 transfectants (Figures 5a and b).

Figure 5
figure 5

GOLM1 mRNA and protein expression levels were suppressed by si-GOLM1 transfection in DU145 and PC3 cells. (a) GOLM1 mRNA expression 72 h after transfection with si-GOLM1. GAPDH expression was used for normalization. **P<0.0001. (b) GOLM1 protein expression 72 h after transfection with si-GOLM1. GAPDH was used as a loading control. The ratio of GOLM1/GAPDH expression was evaluated using ImageJ software (ver. 1.43; http://rsbweb.nih.gov/ij/index.html).

Full size image

XTT assays revealed that transfection with si-GOLM1 inhibited cell proliferation in PC3 cells compared with mock-transfected cells and si-control transfectants; however, no significant inhibition was observed in DU145 cells (cell proliferation, percentage of mock: DU145 cells 89.5%±2.3%, 92.7%±2.9%, 100.0%±1.9% and 98.5%±1.7%, respectively, P=0.0033 and P=0.031; PC3 cells 78.0%±1.2%, 81.5%±0.9%, 100.0%±1.3% and 92.0%±1.4%, respectively, P<0.0001, Figure 6a).

Figure 6
figure 6

Effects of GOLM1 silencing by si-GOLM1 transfection on the PCa cells. (a) Cell proliferation was determined by the XTT assay. (b) Cell migration activity was determined by the wound-healing assay. (c) Cell invasion activity was determined by the Matrigel invasion assay. **P<0.0001, *P<0.01.

Full size image

Wound-healing assays demonstrated that transfection with the two siRNAs targeting GOLM1 significantly inhibited cell migration compared with mock-transfected cells and control transfectants in PC3 and DU145 cells (cell migration, percentage of mock: DU145 cells 68.5%±7.3%, 43.4%±4.6%, 100.0%±17.7% and 115.6%±4.5%, respectively, P=0.0389 and P=0.0008; PC3 cells 39.9%±4.7%, 27.8%±4.7%, 100.0%±2.1% and 105.1%±8.2%, respectively, P<0.0001; Figure 6b).

Similarly, Matrigel invasion assays demonstrated that the number of invading cells was significantly decreased si-GOLM1-1 and si-GOLM1-2 transfectants compared with mock-transfected cells and control transfectants, especially in PC3 cells (percentage of invading cells: DU145 cells 70.5%±5.4%, 80.9%±5.5%, 100.0%±4.4% and 127.1%±7.0%, respectively, P=0.0022 and P=0.0399; PC3 cells 39.7%±2.6%, 8.8%±1.0%, 100.0%±14.8% and 67.6%±9.2%, respectively, P<0.0001; Figure 6c).

Immunohistochemistry for detection of GOLM1 in a PCa tissue microarray

Next, we determined the expression levels of GOLM1 in PCa tissues and normal prostate tissues by immunohistochemical staining. GOLM1 was strongly expressed in several tumor lesions (Figures 7a and b), whereas no or low expression was observed in normal tissues (Figure 7c). Moreover, the expression score for GOLM1 protein was significantly higher in PCa tissues than in normal tissues (P=0.0244, Figure 7d). The patients’ backgrounds and clinicopathological characteristics are summarized in Supplementary Table S4. There were no significant correlations between in GOLM1 expression and the tested clinicopathological parameters (Gleason Score and stages, data not shown).

Figure 7
figure 7

Immunohistochemical staining of GOLM1 in PCa (n=39) and normal prostate tissues (n=8) by tissue microarray (left panel, original magnification: × 100; right panel, original magnification: × 400). (a) A strongly stained tumor lesion (Gleason Score 4+5=9, pT3bN1M0). (b) A strongly stained tumor lesion (Gleason Score 4+5=9, pT3bN0M0). (c) Negative staining in normal, non-malignant prostate tissues. (d) Quantification of GOLM1 expression. The expression of GOLM1 was upregulated in PCa specimens compared with normal prostate tissues (P=0.00244).

Full size image

Discussion

Recent studies of our miRNA expression signatures, including PCa, demonstrated that miR-143/145 cluster was frequently reduced in cancer tissues.9, 10, 18, 19 Therefore, it is suggested that the miR-143/145 cluster have an important tumor-suppressive function in human cancers. In this study, we first tested the expression levels of miR-143 and miR-145 in needle biopsy core tissues collected from patients with PCa and non-PCa. We confirmed the reduced expression of both miRNAs in PCa, and expression analysis showed positive correlations between miR-143 and miR-145. These data suggested that miR-143 and miR-145 are regulated by the same regulatory systems, and silencing of this miRNA cluster may be a key step in the oncogenesis and progression of PCa. Moreover, miR-143 and miR-145 were reported to inhibit stem cell characteristics of PC3 cells, resulting in the inhibition of bone invasion and tumorigenicity of PC3 cells in vivo.12 Others have reported that wild-type p53 suppresses the epithelial–mesenchymal transition and stemness of PC3 cells by modulating miR-145.13 Tumor-suppressor p53 has been reported to transcriptionally regulate miR-145, inducing its expression by interaction with a potential p53 response element at the miR-145 promoter.22 Interestingly, c-MYC is directly repressed by miR-145. These studies indicate that tumor-suppressive miR-145, as a new member of the p53 regulatory network, contributes to the direct linkage between p53 and c-MYC in human cancer pathways.23 Recent studies have shown that the miR-143/145 cluster is repressed by RAS responsive element-binding protein (RREB1) downstream of constitutively active KRAS.24 Thus, investigation of molecular targets and cancer pathways regulated by the miR-143/145 cluster is necessary for the clinical development of novel cancer therapeutics.

Researchers are interested in investigating the functional significance of the miR-143/145 cluster in cancer cells and identifying novel targets commonly regulated by these clustered miRNAs in human cancers, including PCa. Multiple reports have shown that miR-143 and miR-145 function as tumor suppressors, targeting a variety of oncogenic genes in several types of cancers.25, 26 Our recent data demonstrated that miR-143/145 cluster inhibited cancer cell migration and invasion in renal cell carcinoma cells via targeting hexokinase-2, suggesting that this cluster acts as a tumor suppressor in renal cell carcinoma.27 In this study, we also showed that miR-143/145 cluster function as a tumor suppressor in PCa. Therefore, identification of miR-143/145 cluster-regulated cancer pathways may provide new insights into the molecular mechanisms of human cancers.

We hypothesize that identification of novel cancer pathways and target genes regulated by tumor-suppressive miRNAs is an important first step in understanding human oncogenesis. Based on this view, we performed a combination study of genome-wide gene expression analysis in cells transfected with tumor-suppressive miRNAs, in silico analysis and identification of novel miRNA-regulated cancer pathways and targets, as reported previously.17, 18 In the present study, we identified the GOLM1 as a miR-143/145 cluster-regulated molecular target and proved that miR-143/145 cluster directly binding 3′UTR of GOLM1. We also found that silencing of GOLM1 resulted in significant inhibition of migration and invasion in PCa cells. Furthermore, overexpression of GOLM1 was observed in PCa clinical specimens, as detected by immunohistochemistry. These data suggested that GOLM1 functioned as an oncogene, contributing to cancer cell migration and invasion. However, the mechanisms mediating the overexpression of GOLM1 in PCa and its role in cancer cell migration and invasion remain unclear.

The GOLM1/GP73/GOLPH2 protein is encoded by the GOLM1 gene, located on human chromosome 9q21.33. GOLM1 contains two transcript variants, and both variants encode the same protein.14 The expression of GOLM1 has been shown to be overexpressed in human PCa tissue,lung adenocarcinoma and hepatocellular carcinoma.15, 16, 17, 28, 29 Structurally, GOLM1 resembles a type II Golgi transmembrane protein, and studies have shown that GOLM1 processes proteins synthesized in the rough endoplasmic reticulum and assists in the transport of protein cargo through the Golgi apparatus.30 In PCa cells, GOLM1 is localized in the subapical cytoplasmic region, typical of Golgi distribution. Interestingly, the expression of GOLM1 is higher in the blood of patients with hepatocellular carcinoma and in the urine of patients with PCa than in healthy individuals, suggesting that GOLM1 is a useful tool for the diagnosis of human cancers.14, 15, 16, 17, 29

However, despite that a number of studies have demonstrated GOLM1 expression in cancers, we questioned whether GOLM1 contributed to the oncogenesis and metastasis of PCa. According to our immunohistochemistry data, there was no correlation between GOLM1 protein expression and tumor grade or stage. Even the gene expression data (GSE 29079) was not able to find clinical correlation. In contrast, another study reported the moderately increased expression of GOLM1 in prostatic intra-epithelial neoplasia, a precancerous condition.15 This data suggested that GOLM1 contributes to the oncogenesis and progression of PCa. The clinical study with expression status of GOLM1 and clinical features of PCa will be necessary in future.

Multiple reports had indicated that the Golgi apparatus is a key player in cell migration and acts as a hub for different signaling pathways, including the mitogen-activated protein kinase pathway, the mammalian target of rapamycin pathway, Rho family GTPases and cyclin-dependent kinases, through signaling crucial molecules to translocate to and from the Golgi, thereby playing a key role in cell migration and invasion.31, 32 Therefore, elucidation of the regulatory networks of the Golgi apparatus will provide important information for the development of new therapeutic strategies against cancer cell metastasis. Further research is needed to reveal the oncogenic functions of GOLM1 that are regulated by the tumor-suppressive miR-143/145 cluster in PCa. Therefore, improving our understanding of GOLM1 signaling, as mediated by the tumor-suppressive miR-143/145 cluster, should shed light on the mechanisms of PCa oncogenesis and metastasis and assist in the development of more effective therapeutic intervention for this disease.

In conclusion, downregulation of the miR-143/145 cluster is frequently observed in PCa clinical specimens. Restoration of these miRNAs significantly inhibited cancer cell migration and invasion, suggesting that miR-143/145 functioned as a tumor suppressor in PCa. To the best of our knowledge, this is the first report demonstrating that the tumor-suppressive miR-143/145 cluster directly regulated GOLM1 in PCa cells. Moreover, GOLM1 was upregulated in PCa clinical specimens and contributed to cancer cell migration and invasion, indicating that this protein functioned as an oncogene. The identification of novel target oncogenes regulated by the miR-143/145 cluster may lead to a better understanding of PCa oncogenesis and the development of new therapeutic strategies to treat this disease.