Loss of EGFR signaling regulated miR-203 promotes prostate cancer bone metastasis and tyrosine kinase inhibitors resistance.

Activation of EGFR signaling pathway leads to prostate cancer bone metastasis; however, therapies targeting EGFR have demonstrated limited effectiveness and led to drug resistance. miR-203 levels are down-regulated in clinical samples of primary prostate cancer and further reduced in metastatic prostate cancer. Here we show that ectopic miR-203 expression displayed reduced bone metastasis and induced sensitivity to tyrosine kinase inhibitors (TKIs) treatment in a xenograft model. Our results demonstrate that the induction of bone metastasis and TKI resistance require miR-203 down regulation, activation of the EGFR pathway via altered expression of EGFR ligands (EREG and TGFA) and anti-apoptotic proteins (API5, BIRC2, and TRIAP1). Importantly, a sufficient reconstitution of invasiveness and resistance to TKIs treatment was observed in cells transfected with anti-miR-203. In prostate cancer patients, our data showed that miR-203 levels were inversely correlated with the expression of two EGFR ligands, EREG and TGFA, and an EGFR dependent gene signature. Our results support the existence of a miR-203, EGFR, TKIs resistance regulatory network in prostate cancer progression. We propose that the loss of miR-203 is a molecular link in the progression of prostate cancer metastasis and TKIs resistance characterized by high EGFR ligands output and anti-apoptotic proteins activation.

samples from 25 with independent prostate tumors used in qRT-PCR analyses were collected form Wan Fang Hospital, Taipei Medical University. RNA was extracted from dissected tissue containing greater than 70% tumor cell content.

MicroRNA Luciferase Assay
RasB1 cells were transfected with 1μg of human SUZ12, AREG, EREG, and TGFA 3'UTR reporter and 1μg of precursor miR's encoding empty vector or miR-203 precursor. The psiCHECK-2 vectors contain both firefly and Renilla luciferase reporters. Cell extracts were prepared 24 hours after EGF (10µM) or CI1033 (10nM) treatment and the luciferase and Renilla activities were measured using the Dual Luciferase Reporter Assay System (Promega, WI).
Renilla luciferase activities were calculated as the mean ± SEM after normalization to firefly luciferase activities. Three independent experiments were done in triplicate.

FACS Analysis
Promoter functional analysis using FACS and relative MFI (median fluorescent intensity) value was measured as described [1,2]. Cells were treated with or without EGF (10µM), and CI1033 (10nM) for 48 hours. The MFI (median fluorescent intensity) value for RFP was measured by FACS using FACSDiva software and normalized to the value of the vehicle. Promoter functional analysis using FACS and relative MFI value was measured as described [3]. Mean fluorescent intensities were determined from the first peak of fluorescence (which occurs in >80% of the cells since conditions are used to optimize single lentivirus incorporation) and represent single copy integration.

Chromatin Immunoprecipitation (ChIP)
ChIP assays were performed using the EZ magna ChIP A kit (Millipore, CA) with a modified protocol. Cells were treated with or without EGF (10µM) for 24 hours. Cultured cells (1X10 7 ) were cross-linked with 1% formaldehyde at RT for 15 minutes. The fixation was quenched with glycine, and cells were washed twice with cold PBS containing complete protease inhibitor (Roche, CA). Cell pellets were resuspended in cell lysis buffer and incubated on ice for 15 minutes. Nuclei were collected by centrifugation at 10,000 rpm at 4°C for 10 minutes and resuspended in nuclei lysis buffer. Chromatin was sheared using a sonicator (Branson Sonifier 250, Teltow, Germany) with a microtip in 20-second burst followed by 1 minutes of cooling on ice for a total sonication time of 5 minutes per sample. The procedure results in DNA fragment sizes of 100-300 bp. Sheared chromatin was divided to perform immunoprecipitation with rabbit IgG antibody (Santa Cruz Biotechnology, CA) or primary antibody at 4°C overnight.
Immunoprecipitation, washing, elution, reverse cross-linking, and DNA purification steps were performed according to Millipore's protocol. Quantitative PCR was performed in triplicate with 2μl of eluted chromatin. ChIP antibodies and PCR primers are listed in Supplemental Tables (Table S5).

Immunohistochemistry (IHC) staining
Immunohistochemistry (IHC) was performed using EREG antibodies from R&D (R&D Systems, MN) at 1:60 dilution. In general, unstained sections were deparaffinized and rehydrated. Antigen retrieval was performed using Target Antigen Retrieval Solution (DAKO, CA) and autoclave for 10 minutes. For IHC, endogenous peroxidase was blocked using a 3% hydrogen peroxide solution.
All sections were blocked with Cyto Q Background Buster Reagent (Innovex BioSciences, CA).

Primary antibodies were incubated overnight at 4°C in Antibody Diluent with Background
Reducing Components (DAKO, CA). Secondary antibody, 1:250 HRP labeled anti mouse/rabbit (Vector laboratories, CA), incubation was performed at room temperature for 30 minutes and bound peroxidase detected using the ABC Peroxidase Kit (Vector laboratories, CA) and DAB (DAKO, CA). All IHC slides were counterstained with hematoxylin. For histomorphometric analysis of tissue sections, microscopic images were collected under 200X magnification using an Axioplan microscopy system (Zeiss, Thornwood, New York).

Clinical outcome and correlation analyses using human data sets
We used mRNA expression data from a public human prostate cancer data set [4,5]. The study was conducted under MSKCC Institutional Review Board approval on 28 normal, 151 primary, and 19 metastatic samples. The expression data (and resulting z-scores) were log2 normalized.
Additionally, microRNA expression was determined for 113 tumors and 28 matched normal samples with Agilent microRNA V2 arrays. KRAS (Broad Institute), metastasis [6], and EGFR signaling [7] responsive gene signatures were used to determine correlations with miR-203 levels. Gene sets were scored by summing the expression Z score per tumor within the cohort.
Tumors were mean stratified by miR-203 expression and the mean Z scores were determined in each group.

Statistical Analysis
In vivo animal results and clinical outcome analysis are expressed as plots showing the median and box boundaries extending between 25th to 75th percentiles, with whiskers down to the minimum and up to the maximum value. All in vitro data were presented as means ± S.E.M. Statistical calculations were performed with GraphPad Prism (GraphPad Software, Inc.) analysis tools.
Differences between individual groups were analyzed by one way or two way ANOVA test.
Bonferroni's post test was used for comparisons among 3 or more groups. Log-rank test was used for survival curve analysis. P values less than 0.05 were considered statistically significant.