Advertisement
Biology of Human Tumors

Dysregulation of miR-212 Promotes Castration Resistance through hnRNPH1-Mediated Regulation of AR and AR-V7: Implications for Racial Disparity of Prostate Cancer

Yijun Yang, Dingwu Jia, Hogyoung Kim, Zakaria Y. Abd Elmageed, Amrita Datta, Rodney Davis, Sudesh Srivastav, Krzysztof Moroz, Byron E. Crawford, Krishnarao Moparty, Raju Thomas, Robert S. Hudson, Stefan Ambs and Asim B. Abdel-Mageed
DOI: 10.1158/1078-0432.CCR-15-1606 Published April 2016

Abstract

Purpose: The causes of disproportionate incidence and mortality of prostate cancer among African Americans (AA) remain elusive. The purpose of this study was to investigate the mechanistic role and assess clinical utility of the splicing factor heterogeneous nuclear ribonucleoprotein H1 (hnRNP H1) in prostate cancer progression among AA men.

Experimental Design: We employed an unbiased functional genomics approach coupled with suppressive subtractive hybridization (SSH) and custom cDNA microarrays to identify differentially expressed genes in microdissected tumors procured from age- and tumor grade–matched AA and Caucasian American (CA) men. Validation analysis was performed in independent cohorts and tissue microarrays. The underlying mechanisms of hnRNPH1 regulation and its impact on androgen receptor (AR) expression and tumor progression were explored.

Results: Aberrant coexpression of AR and hnRNPH1 and downregulation of miR-212 were detected in prostate tumors and correlate with disease progression in AA men compared with CA men. Ectopic expression of miR-212 mimics downregulated hnRNPH1 transcripts, which in turn reduced expression of AR and its splice variant AR-V7 (or AR3) in prostate cancer cells. hnRNPH1 physically interacts with AR and steroid receptor coactivator-3 (SRC-3) and primes activation of androgen-regulated genes in a ligand-dependent and independent manner. siRNA silencing of hnRNPH1 sensitized prostate cancer cells to bicalutamide and inhibited prostate tumorigenesis in vivo.

Conclusions: Our findings define novel roles for hnRNPH1 as a putative oncogene, splicing factor, and an auxiliary AR coregulator. Targeted disruption of the hnRNPH1-AR axis may have therapeutic implications to improve clinical outcomes in patients with advanced prostate cancer, especially among AA men. Clin Cancer Res; 22(7); 1744–56. ©2015 AACR.

Translational Relevance

African Americans (AA) remain to be at high risk of prostate cancer morbidity and mortality than any other ethnic groups in the United States. The current study demonstrates aberrant expression and correlation of the splicing factor hnRNPH1 with androgen receptor (AR) expression and disease progression in AA men compared with Caucasian Americans. Dysregulation of the putative tumor suppressor miR-212, and to a lesser extent miR-22, prompts aberrant expression of hnRNPH1, AR, and its spliced variant AR-V7 in prostate tumor cells. Molecular targeting of hnRNPH1 abrogated hormone resistance and inhibited growth of castration-resistant prostate cancer cells in vivo. Our findings suggest that differences in disease biology might contribute to racial differences in disease progression. In addition to its potential clinical utility as biomarkers for disease progression, targeted disruption of hnRNPH1–AR axis may have therapeutic implications to improve clinical outcomes in patients with advanced prostate cancer, especially among AA men.

Introduction

Aberrant expression of androgen receptor (AR) is implicated in initiation and development of castration-resistant prostate cancer (1). On the basis of their physical interactions and ability to modulate transcription, a repertoire of intermediary transcriptional protein complexes (coactivators and corepressors) are recruited by AR to modify chromatin and facilitate transcription of androgen-regulated genes (AGR) in a cell type–specific manner (2). Notably, the differential expression and pathophysiologic significance of these cofactors in disease progression remain elusive.

Current evidence suggests that African Americans (AA) are twice as likely to develop and die of prostate cancer compared with their Caucasian Americans (CA) counterparts (3). Earlier onset of the disease, high disease volume, aggressive metastatic disease, and low survival rate are evident among AA males (4). The racial disparity has been attributed to differences in tumor growth rates and disease aggressiveness and is associated with worst outcomes in advanced disease in AA men (5). Other contributing factors include ancestry predispositions, lifestyle variables and aberrations in AR, genetic and genomic variables implicated in development of tumorigenesis and disease progression (5). Although the disproportionate incidence and mortality cannot be explained by genetic, socioeconomic, and environmental factors, chromosome 8q24 has been implicated in enhancing susceptibility but not in the aggressiveness of prostate cancer in AA men (6). While the concept has been proposed (7), little attention was focused on unraveling the underlying biologic mechanisms involved in prostate cancer racial disparities.

As a residual scaffolding of the nucleus to which DNA sequences and actively transcribed genes are anchored (8), the nuclear matrix (NM), has recently sparked a surge of interest as the molecular underpinning of cancer-specific markers (9). The hnRNP family has more than 30 members of ubiquitously expressed NM proteins (10). The hnRNPs complex with heterogeneous nuclear RNA (hnRNA) and modulate pre-mRNA biogenesis, metabolism, and transport (11). The hnRNPH/F is a subfamily of hnRNPs encoded by different genes into subtype-naïve forms, including hnRNPH (hnRNPH1), hnRNPH' (hnRNP H2), hnRNPF, and hnRNP2H9 (12). These proteins possess a modular and highly conserved structure encompassing two glycine-rich auxiliary domains and two or three repeats of RNA binding quasi-RNA recognition motif (qRRM). The hnRNP H/F members regulate both inhibitory and stimulatory alternative splicing of target genes through binding cognate G-rich intronic and exonic sequences in proximity to the polyadenylation sites (13).

The hnRPH1 gene encodes a splicing regulator, which is aberrantly expressed in a number of human cancers, including head and neck, hepatocellular, pancreatic, colon, and laryngeal carcinomas (14). As a bona fide component of the NM (15), the functional significance of hnRNPH1 is relatively unknown and only recently has evidence emerged related to its biologic function. Notably, hnRNPH1 was shown to assemble into sequences in association with other nuclear matrix proteins to regulate DNA replication, RNA transcription, and steroid hormone action (16). As part of the intronic and exonic splicing enhancer complex, hnRNPH1 acts in splicing regulation of the neuronal c-src pre-mRNA (17) and β-tropomyosin pre-mRNA in rat striated muscle cells (18). The rapid reduction of hnRNPH1 transcripts in cells undergoing differentiation underscores a potential role for this NM protein in inhibition of cell differentiation (19). However, the functional significance of hnRNPH1 in cancer development and progression has not been elucidated.

In the current study, we demonstrate that downregulation of miR-212 is potentially responsible for aberrant expression of hnRNPH1 in prostate tumor cells, especially among AA men. hnRNPH1 in turn promotes prostate cancer tumorigenesis by regulating transcription and activation of AR and its splice variant AR-V7 in conjunction with coactivator SRC-3, also known as nuclear receptor coactivator-3 (NCoA-3).

Materials and Methods

Patients and prostate cancer specimens

Fresh, flash-frozen specimens were obtained from age (50–60 years) and tumor grade–matched (Gleason score 6) AA and CA prostate cancer patients. All patients received no prior therapy, presented with palpable prostate tumors, and underwent radical prostatectomy. Part of the specimens were excised, embedded in Tissue-Tek OCT Compound (Jed Pella Inc.), snap-frozen in liquid nitrogen, and stored at −80°C. In addition, histopathologic sections were made to confirm prostate cancer staging and grading. The study was conducted on the basis of an approved protocol by the Institutional review board (IRB) at Tulane University (New Orleans, LA).

Construction of LCM-based racially stratified prostate cancer–specific SSH cDNA libraries and arrays

First-strand cDNA synthesis and suppressive subtractive hybridization (SSH) analysis was performed using SMART PCR cDNA Synthesis and PCR-Select cDNA Subtraction Kits (Clontech Laboratories, Inc.). Details of laser capture microdissection (LCM), RNA preparation, suppressive subtractive hybridization (SSH), and custom construction of the cDNA arrays are described in Supplementary Methods.

Gene array analysis

Total RNA isolated from LCM-procured matched normal epithelium and tumor cells from flash-frozen sections of AA and CA patients was amplified using the MessageAmp aRNA Kit according to the manufacturer's instructions (Ambion Inc.). The details of microarray analysis are provided in Supplementary Methods.

Tissue microarray analysis

Differential expression of hnRNPH1 in prostate cancer of AA men and CA men was validated by immunohistochemistry (IHC) using an ethnicity-based tissue microarray analysis (TMA-4; NCI Cooperative Prostate Cancer Tissue Resource). The TMA-4 patients' characteristics, clinical annotations, and analysis are shown in Supplementary Methods and Supplementary Table S1.

Cell culture and plasmids

The prostate cancer cell lines, authenticated by short tandem repeat (STR) profiling, were used within 6 months after receipt from ATCC. Description of cell maintenance is included in the Supplementary Information. The human hnRNPH1 expression plasmid, pcDNA3.1/V5-His-TOPO-HNRPH1, was constructed by subcloning a cDNA of the hnRNPH1 gene (Genbank accession # BC001348) into a pcDNA3.1/V5-His-TOPO expression kit in accordance with the manufacturer's instructions (Invitrogen Life Technologies). cDNAs derived from hnRNPH1-expressing MDA-PCa-2b cells was used as a template to amplify a 1373 bp fragment encompassing the ORF of the gene with 30 PCR cycles using a primer set (sense, 5′-GTAAGAGACGATGTTGGG-3′; antisense 5′-GCTCCTTGGTTACCTATGC-3′) a high-fidelity platinum Taq DNA polymerase (Life Technologies) at 94°C for 1 minute (denaturing), 53°C for 1 minute (annealing), and 70°C for 2 minutes (extension). The target sequence was amplified by PCR and fused in-frame into a pcDNA3.1/V5-HisTOPO TA expression plasmid to generate pCMV-hnRNPH1. Insertion and orientation of DNA were verified by colony PCR, restriction digest map, and PCR amplification using hnRNPH1-specific sense primer and the plasmid flanking BGH reverse primer. The pCMV-AR, a human AR-FL expression plasmid, was kindly provided by Dr. X.-B. Shi (University of California at Davis, Davis, CA). The supra PSA/pGL3-luc (psPSA-luc), a luciferase reporter gene driven by truncated PSA promoter sequences encompassing AREs, was a gift from Dr. L.W. Chung (Cedars-Sinai Medical Center, Los Angeles, CA). The control pcDNA3.1+ plasmid was from Addgene.

Cell growth and drug sensitization analysis

The details of prostate cancer cell growth in response to drugs, hormones, and transfections are included in the Supplementary Methods.

Quantitative PCR and miRNA analysis

The qPCR and miRNA analyses were performed as previously described (20, 21). Analysis details and primers used for gene expression (Supplementary Table S2) and miRNAs (Supplementary Table S3) are provided in the Supplementary Methods.

Immunoblot, co-immunoprecipitation, and reporter assays

The assays were performed as previously described (21). Experimental details are furnished in Supplementary Information.

Electrophoretic mobility shift assay

Details of the gel retardation assays are provided in the Supplementary Methods and Supplementary Table S4 as we previously described (22).

Chromatin immunoprecipitation assay

The chromatin immunoprecipitation (ChIP) assays were performed using primer sets flanking AREs within the promoter (ARE I and II) and the enhancer element (ARE III) of PSA gene (23). In addition, ChIP analysis was performed by PCR primer sets (Supplementary Table S5) flanking exons B and H as well as exons D and E encompassing ARE-1 and ARE-2, respectively, of the AR gene (23). Experimental details are included in Supplementary Methods.

Animal studies

Animal studies were performed in accordance with the Institution Animal Care and Use Committee at Tulane University (New Orleans, LA). The growth of LNCaP and C4-2B cells expressing hnRNPH1 or hnRNPH1 shRNA was monitored in SCID mice as described in Supplementary Methods.

Statistical analysis

For microarray analysis, the intensity of each hybridization signal was evaluated photometrically by integrator software (GeneTAC) and normalized to the average signals of the housekeeping genes. Raw microarray measurements were typically normalized, and the background adjusted intensities were then log-transformed to reduce the dynamic range, achieve normality, and make the datasets from different hybridizations comparable. Fluorescence intensities of the two channels were balanced using within-array and between-array normalization methods. Within-array normalization allows for the comparison of the Cy3 and Cy5 channels while the normalized between-array compares the gene expression levels across slides or arrays. Linear regression of the two channels and the log ratio against average intensity (MA plots) were used for within-array normalization. Box plot method for between-array normalization was used for comparing the distributions of log intensities or log ratios of genes on different arrays. For each array, the spot replicates of each gene were merged and expressed as median ratios ± SD. The ratios were log-transformed and normalized using the local intensity–dependent algorithm.

The evaluation of differential gene expression in AA and CA groups was approached as a collection of tests for each gene of the “null hypothesis” of no difference or alternatively as estimating the probability that a gene shows differential expression using a two-sided t test statistic criterion with multiple testing adjustments and an overall level of significance of 5%. Genes with significant differential expression in tumor cells of AA men were reported in order of increasing P value after a Bonferroni adjustment procedure employed. All statistical analysis tests were performed with the Statistical Analysis Software 9.1 (SAS Institute) and graphs were plotted using R software (The R Foundation for Statistical Computing). ANOVA and Fisher exact test were employed to determine whether there were significant differences between AA men and CA men on age at diagnosis, age at prostatectomy, Gleason score, and final score of hnRNPH1. Kaplan–Meier method was used to construct disease recurrence curves and to compare months to PSA recurrence-free using log-rank test. Correlation between hnRNPH1 and clinical parameters were examined using Pearson correlation.

Analysis of hnRNPH1 expression in prostate cancer datasets was analyzed with BRB array tools. Raw CEL files were preprocessed using MAS.5, median array normalized, and filtered for probe sets with default settings. Log-transformed triplicate data was averaged and used for expression analysis. The hnRNPH1 expression data was extracted as log2 median centered intensity and compared using Student t test, one-way ANOVA, and Pearson correlation at significance level of P < 0.05.

Results

hnRNPH1 expression correlates with AR expression and disease progression among AA men

A functional genomics approach is schematically depicted (Supplementary Fig. S1A and S1B). The SSH and construction of race-based libraries (∼700 cDNAs/group) were completed as shown in Supplementary Fig. S2. Following image acquisition and normalization (24), the degree of variability and reproducibility among analyzed samples of various datasets was assessed. Statistical linear regression of Cy3 against Cy5 and the linear regression of log ratio against average intensity (MA plots) were used for within-array normalization (Supplementary Fig. S3A). Box plot method for between-array normalization was used to compare the distributions of log intensities or log ratios of genes on different arrays (Supplementary Fig. S3B).

Our custom cDNA microarray screening and sequence analysis (Supplementary Fig. S3C) revealed the differential expression of hnRNPH1 (P < 0.0001) in microdissected prostate cancer cells procured from AA men in comparison with CA men. In an independent biracial cohort (n = 12/group), twice as high hnRNPH1 transcripts were detected in microdissected prostate tumor cells relative to matched normal epithelium in AA, but not the CA, men (P < 0.001; Fig. 1A). This finding was validated by interrogating three independent prostate cancer cohorts procured from Affymetrix NCBI Geo and Oncomine gene expression datasets. In the first biracial GSE17386 dataset (25), hnRNPH1 gene expression is significantly upregulated (P < 0.0005; 1.76-fold) in prostate tumors in AA men compared with CA men (Fig. 1B). In the second dataset by Wallace and colleagues encompassing 20 normal glands and 69 tumor tissues (26), the hnRNPH1 transcripts levels are significantly elevated in tumors versus normal epithelium and to a greater extent in AA (P > 0.0001) tumors than in CA tumor cells (P < 0.018; Fig. 1C and D). Analysis of this dataset (26) shows aberrant hnRNPH1 gene expression in tumors with positive extraprostatic extension and high Gleason sum score (Fig. 1E and F). A similar finding was observed in an independent dataset by Glinsky and colleagues (ref. 27; Fig. 1G). Furthermore, a direct correlation of hnRNPH1 and AR expression (P < 0.0001) and possible coderegulation of these genes in prostate tumor cells were observed in two interrogated datasets (refs. 26, 27; Fig. 1I and J). We further validated this observation in microdissected tumor cells from an independent cohort of AA men (n = 13) and CA men (n = 17; Supplementary Table S6). The results depicted selective twice as high coexpression (P < 0.005) of hnRNPH1 (Fig. 1K) and AR (Fig. 1L) in AA men than CA men as well as a positive correlation (r = 0.8323; P < 0.005) between AR and hnRNPH1 transcripts in prostate tumor cells (Fig. 1M). Collectively, data from microdissected tumors and two independent cohorts support correlation between AR and hnRNPH1 expression in prostate tumors, especially among AA men.

Figure 1.
Figure 1.

Selective expression and correlation of hnRNPH1 with AR expression, prostate cancer progression, and recurrence in AA men. A, qRT-PCR analysis of hnRNPH1 gene expression relative to the β-actin in LCM-procured normal prostate epithelium (NE) and tumor cells (T) of an independent cohort of AA and CA (n = 12/group). B, analysis of gene expression of Affymetrix GSE17386 dataset (NCBI Geo; Oncomine) denotes a significant expression of hnRNPH1 in prostate tumors in AA men in comparison with CA men. C, analysis of the second dataset reveals significant elevation of hnRNPH1 transcripts levels in prostate tumor cells (T) in comparison with the normal prostate epithelium (NE). D, analysis of second database indicates significant hnRNPH1 gene expression in prostate tumor cells compared with the normal epithelium in both groups, but to a greater extent in AA men than CA men. E–G, aberrant hnRNPH1 gene expression correlates with extraprostatic extension (EPE) and high Gleason sum score. H, hnRNPH1 gene expression increases in 5-year recurrent tumors in comparison with nonrecurrent tumors. I and J, positive correlation between hnRNPH1 and AR gene expression in prostate tumors in comparison with the normal epithelium. Affymetrix Human Genome U133 Plus 2.0 array datasets (26, 27) AR and hnRNPH1 expression data were extracted as log2 median and compared using Pearson correlation. K and L, qRT-PCR analysis demonstrates coexpression of hnRNPH1 and AR transcripts, respectively, in microdissected tumor cells relative to their matched normal epithelium in an independent cohort of AA men (n = 13) than CA men (n = 17). Gene expression is significant (P < 0.05) relative to normal adjacent tissue (a) and to CA men (b). M, regression analysis depicts a positive correlation (r = 0.8323; P < 0.005) between AR and hnRNPH1 transcript levels shown in K and L.

An ethnicity-based TMA analysis, encompassing 150 prostate tumor cores each from AA and CA men, corroborated the selective expression and predominant nuclear immunoreactivity of hnRNPH1 protein in tumor cells in both groups (Fig. 2A–C), compared with BPH (Fig. 2D–F) and normal epithelium (Fig. 2G–I). The antibody specificity was validated with negative staining of prostate tumors (without primary antibody; Fig. 2J) and positive weak and intense staining of hnRNPH1 in AR-null PC-3 (Fig. 2K) and AR-expressing LNCaP cell cores (Fig. 2L), respectively. Importantly, histoscore analysis revealed significant nuclear immunostaining in prostate tumor cells in AA men compared with CA men (Fig. 2M). In addition, of the TMA clinicopathologic variables studied (Supplementary Table S1), the hnRNPH1 expression was found to positively correlate with Gleason score in both groups, but with significant (P < 0.01) correlation in Gleason 6–7 tumors in AA men compared with CA men (Fig. 2N).

Figure 2.
Figure 2.

Ethnicity-based TMA-4 analysis reveals prostate tumor cell nuclear immunoreactivity and correlation with disease progression in AA men. A–C, a representative prostate adenocarcinoma core depicting intense nuclear immunoreactivity to hnRNPH1 (arrow) in comparison with stroma (arrowhead). D–I, representative cores of benign prostatic hyperplasia (d–f) and normal prostate epithelium (g–i) demonstrating weak nuclear immunoreactivity (arrow) in the epithelial cells in comparison with the adjacent stroma (arrowhead). J, validation of the hnRNPH1 antibody specificity with negative staining of prostate tumor cores in the absence of primary antibody. K and L, weak and intense immunostaining of hnRNPH1 in cell cores of AR-naïve PC-3 and AR-expressing LNCaP, respectively. M, histoscore of hnRNPH1 in tumors (T) relative to the normal epithelium (NE) in AA (n = 148) and CA men (n = 152) expressed as mean ± SEM. N, hnRNPH1 histoscore stratified by Gleason scores. *, denotes significant difference at P < 0.05 in comparison with controls.

Aberrant coexpression of hnRNPH1 and AR are direct targets of miR-22 and miR-212 in prostate tumors and prostate cancer cell lines

Next we investigated the underlying mechanisms by which hnRNPH1 is transcriptionally upregulated in prostate tumor cells. We pursued in silico analysis in a number of miRNA-predicting targets programs such as TargetScan, miRNA and Microcosm Targets (Supplementary Table S7) to uncover putative miRNAs that may target hnRNPH1 transcripts in prostate tumor cells. On the basis of sequence analysis algorithms, hnRNPH1 emerged as one of the high rated targets for mirR-22, miR-212, miR-132, and miR-495. qRT-PCR validation analysis revealed significant (P < 0.05) downregulation of miR-22 and miR-212 in LCM-procured prostate tumor cells relative to matched normal prostate epithelium, especially among AA men compared with CA men (Fig. 3A and B). In contrast, miR-132 and miR-495 were downregulated in prostate tumor cells regardless of race (Supplementary Fig. S4). miRNA mimics assays were performed to determine the functional significance of miR-22 and/or miR-212 in modulating transcripts levels of hnRNPH1, AR, and its spliced variant AR-V7 in C4-2B cells. Ectopic restoration of miR-22 and/or miR-212 (Fig. 3C and D) was coupled with a significant reduction in hnRNPH1 transcripts in C4-2B cells (Fig. 3E). However, miR-212 mimics alone or in combination with miR-22 exhibited a robust suppressive effect (>90%) compared with miR-22 (50%). Likewise, the miR-212–targeted inhibition of the splicing factor hnRNPH1 was coupled with an 80% reduction in all AR transcripts (AR-FL and AR-Vs) compared with miR-22 (40%) in C4-2B cells, as measured by qRT-PCR primers targeting the NTD (Fig. 3F). The results were also confirmed by conventional PCR (Fig. 3G). Together our data suggest that miR-212, and to a lesser extent miR-22, may play a pivotal role in regulating the endogenous levels of hnRNPH1 and all AR transcripts in prostate tumor cells.

Figure 3.
Figure 3.

Expression and functional significance of miR-212 and miR-22 in the regulation of hnRNPH1 and AR and its splice variants in prostate cancer cells. A and B, in silico analysis of different miRNA predicting targets programs (TargetScan, miRNA and Microcosm Targets) predicted hnRNPH1 as a target for miR-22 and miR-212 among others. qRT-PCR analysis of hnRNPH1-associated miR-22 and miR-212 expression in prostate tumors derived from AA (n = 13) and CA men (n = 17) relative to the normal epithelium in both groups. The data are expressed as fold change relative to U6 as an internal control. *, significance at P < 0.05 relative to controls. #, a significant difference in hnRNPH1 levels in prostate tumor cells procured from CA men compared with AA men. C and D, transfection efficiencies of miR-22 and/or miR-212 in C4-2B cells, respectively, as measured by qRT-PCR relative to cells transfected with nontargeting control miRNA (control miR). Ectopic expression of either miR-22 and/or miR-212 mimics significantly reduced expression of hnRNPH1 and AR-FL and all N-terminal containing AR variants (AR-nt) in C4-2B cells compared with control miR–transfected cells, as measured by qRT-PCR (E and F) and conventional PCR analyses (G). In addition to reduction of hnRNPH1 and AR-FL transcripts, the ectopic expression of miR-212 mimics significantly inhibited the AR splice variant AR-V7 mRNA in C4-2B cells when measured by qRT-PCR (H), conventional PCR (I) and Western blot (J) analyses. hnRNPH1 shRNA silencing causes reduction of AR-FL and AR-V7 mRNAs compared with control shRNA-transfected C4-2B cells as measured by qRT-PCR (K) and immunoblot (L) assay, as quantified relative to GAPDH by densitometric analysis (M).

hnRNPH1 silencing reduces expression of AR and its splice variant AR-V7 in prostate cancer cells

Because of their potential roles in prostate cancer progression, we examined whether aberrant expression of AR and its spliced variant AR-V7 transcripts in prostate cancer cells is attributed to downregulation of miR-212 or indirectly due to their mediated upregulation of the splicing factor hnRNPH1. The ectopic expression of miR-212 mimics suppressed hnRNPH1 by approximately 90% and AR-FL and AR-V7 by approximately 50% in C4-2B as evidenced by qRT-PCR (Fig. 3H), conventional PCR (Fig. 3I), and Western blot analysis (Fig. 3J) analyses. As shown in Fig. 3K, shRNA silencing of hnRNPH1 significantly reduced AR-FL and AR-V7 mRNAs by at least 50% in C4-2B cells by qRT-PCR. The findings, corroborated by immunoblot analysis (Fig. 3L and M), further attest to the positive correlation of aberrant expression of hnRNPH1 and AR with disease progression. Taken together, the results implicate that downregulation of miR-22 and/or miR-212 triggers endogenous accumulation of hnRNPH1 transcripts, which in turn upregulates transcription of AR-FL and AR-V7, potentially through transcriptional regulation and/or slicing activity in prostate tumor cells.

hnRNPH1 promotes prostate cancer cell growth and sensitizes prostate cancer cells to bicalutamide

Constitutive gene expression of hnRNPH1 was significantly (P < 0.01) abundant in the AR-expressing MDA-PCa-2b (6-fold) and C4-2B (4-fold) cells compared with the AR(−) PC-3 cells (Fig. 4A and B). Next, we examined the role of hnRNPH1 on AR-expressing prostate cancer cell proliferation. Following optimization of transfection (Fig. 4C and D) and silencing (Fig. 4E and F) efficiencies, we show that siRNA silencing of hnRNPH1 induced significant time-dependent growth inhibition in MDA-PCa-2b cells (Fig. 4G). In contrast, the growth kinetics was not affected in AR-naïve PC-3 cells (Supplementary Fig. S5). Moreover, we examined whether hnRNPH1 silencing influences the sensitivity of prostate cancer cells to bicalutamide in presence or absence of androgens. Regardless of hormone availability, hnRNPH1 silencing increases sensitivity (P < 0.05) of C4-2B and MDA-PCa-2b cells to BIC cytotoxicity (Fig. 4H and I). The growth inhibitory effect of C4-2B cells stably transfected with hnRNPH1 shRNA (Fig. 4J) was confirmed in an animal model. While the body weights remained unchanged (Fig. 4K), hnRNPH1 silencing caused significant (P < 0.05) growth inhibition of C4-2B tumor xenografts in vivo (Fig. 4L and M). In contrast, overexpression of hnRNPH1 stimulated the growth of the less tumorigenic AR-expressing LNCaP cells in comparison with control plasmid–transfected cells in vivo (Supplementary Fig. S6).

Figure 4.
Figure 4.

hnRNPH1 confers growth stimulation and hormone resistance through activation of AR in prostate cancer cells. A, qRT-PCR analysis of hnRNPH1 transcripts in AR-expressing (C4-2B and MDA-PCa-2b) and AR-null PC-3 cells. B, immunocytochemical analysis of hnRNPH1 protein expression in MDA-PCa-2b cells in presence of complete medium. C and D, optimization of siRNA silencing and transfection efficiencies in prostate cancer cells by GFP and siGLO Lamin A/C, respectively. E and F, endogenous mRNA and proteins levels of hnRNPH1, respectively, at 24 hours following siRNA transfection. G, assessment of growth inhibitory effects by a cell counting assay kit in hnRNPH1 siRNA-silenced MDA-PCa-2b cells cultured in complete medium for up to 120 hours. H and I, cell growth of MDA-PCa-2b and C4-2B cells, respectively, pretransfected with siControl or hnRNPH1 siRNA and cultured in RPMI containing charcoal-stripped serum and various concentrations of BIC with (+) or without (−) DHT for 24 hours (n = 3). J, immunoblot analysis of C4-2B cells stably transfected with control shRNA or hnRNPH1 shRNA plasmids. K and L, body weight and tumor volume in animals transplanted with C4-2B cell stable clones, respectively. M, resected tumor sizes of siControl-, or siRNPH1-transfected C4-2B cells. N and O, COS-7 and CV-1 cells, respectively, were cultured in charcoal-stripped FBS medium in the absence (ethanol) or presence of DHT and cotransfected with hnRNPH1, pCMV-AR, and psPSA-Luc plasmids. P, LNCaP cells cotransfected with hnRNPH1 and psPSA-Luc plasmids and cultured with or without DHT. Q, C4-2B cells cotransfected with non-target siRNA (siControl) or HnRNPHI siRNA (siRNPH1) and psPSA-Luc reporter and cultured with or without DHT. For normalization, all cells were cotransfected with five ng pRL-SV40. Activity was measured with the dual luciferase system, and the results were expressed as fold change of relative light units (RLU). * and **, significant difference at P < 0.05 and P < 0.01, respectively, in comparison with controls (n = 3).

hnRNPH1 confers androgen-dependent and independent transactivation of the AR

Expression of hnRNPH1 in COS-7 (Fig. 4N) and CV-1 (Fig. 4O) cells induced hormone-independent AR activation (P < 0.05) as opposed to controls. In contrast, 10 nmol/L dihydrotestosterone (DHT) induced nearly twice the level of AR activation following ectopic coexpression of hnRNPH1 and AR in COS-7 and CV-1 cells (Fig. 4N and O) as opposed to either factor alone (P < 0.05). Interestingly, DHT in the absence of AR significantly increased (P < 0.05) PSA promoter activity in hnRNPH1-transfected COS-7 and CV-1 cells (Fig. 4N and O). Similar findings were observed in prostate cancer cells. Transfection of LNCaP cells with hnRNPH1 augmented AR transactivation in a ligand-dependent manner (Fig. 4P). Conversely, silencing hnRNPH1 in C4-2B cells inhibited PSA promoter activity with or without DHT (Fig. 4Q).

hnRNPH1 physically interacts with AR and recruits SRC-3

The AR activation by hnRNPH1 prompted us to investigate whether these proteins physically interact in the AR-expressing C4-2B and MDA-PCa-2b cells. Nuclear lysates immunoprecipitated with anti-hnRNPH1 or anti-AR antibody demonstrate an increase in immunoblotted AR and hnRNPH1 levels, respectively, in comparison with control rabbit IgG, suggesting interaction between these factors (Fig. 5A), an effect that is further augmented in presence of DHT (Fig. 5B). In addition to existence in the absence of DHT, immunofluorescence analysis demonstrates that DHT stimulates the interacting proteins to colocalize to the nucleus in prostate cancer cells, (Fig. 5C). The interaction between hnRNPH1 and AR is associated with recruitment of SRC-3 (NCoA-3), but not NCoA-4, to the interacting protein complex as evidenced by weak (Fig. 5D) and strong (Fig. 5E) immunoreactivity in the cytoplasm and the nuclear lysates of C4-2B cells, respectively. Of note, the qRT-PCR analysis revealed that DHT increases expression of hnRNPH1 in C4-2B and MDA-PCa-2b cells (Fig. 5F and G). The transcriptional upregulation of hnRNPH1 by DHT is coupled with its nuclear localization of hnRNPH1 proteins in both cell lines (Fig. 5H and I). On the basis of its role in mRNA biogenesis and AR interaction, we examined whether the endogenous expression of hnRNPH1 modulates DHT-mediated transcription of AR and PSA in prostate cancer cells. Silencing of hnRNPH1 expression caused significant (P < 0.05) reduction in the basal transcripts of AR (Fig. 5J and K) and PSA (Fig. 5L and M) in presence or absence of DHT in MDA-PCa-2B and C4-2B cells, respectively. Our findings suggest that hnRNPH1 is potentially involved in the regulation of ARGs.

Figure 5.
Figure 5.

AR–hnRNPH1 interaction and transcriptional regulation of AR and PSA in prostate cancer cells. A, prostate cancer cell lysates cultured in complete medium were subjected to immunoprecipitation (IP) using anti-AR or anti-hnRNPH1 antibody, followed by immunoblotting (IB) with the indicated antibodies in a reversed order as shown. B, lysates of prostate cancer cells cultured in charcoal-stripped medium with or without DHT were analyzed for AR–hnRNPH1 interaction by co-immunoprecipitation (co-IP) analysis as shown above (n = 3). C, representative deconvolution photomicrographs (Leica DMRXA) depicting endogenous expression and colocalization of AR and hnRNPH1 in prostate cancer cells under DHT-treated and hormone-deprived conditions for 2 hours. Cells were fixed and stained with DAPI nuclear counterstain (blue) and then reacted with hnRNPH1- or AR-specific antibody followed by a secondary antibody conjugated with Alexa Fluor 488 (green) or Alexa Fluor 568 (red). Inset depicts hnRNPH1 not localized (white arrow) or weakly colocalizes with AR (yellow arrow) in the nucleus in the absence of DHT. In merged photomicrographs, DHT increases both expression and nuclear colocalization of hnRNPH1 and AR (green arrow) in prostate cancer cells. Scale bar, 10 μm. D and E, immunoblot analysis demonstrating weak cytosolic and strong nuclear coprecipitation of SRC-3 with AR or hnRNPH1 antibodies in C4-2B cells, respectively. The purity of nuclear and cytoplasmic fractions was assessed by TATA-binding protein (TBP), α-tubulin, GAPDH, and lamin B, whereas actin was used as a loading control (n = 3). F and G, qRT-PCR analysis of hnRNPH1 expression in MDA-PCa-2b and C4-2B cells, respectively, cultured in phenol red–free, charcoal-stripped media and transfected with siControl (nontarget siRNA) or hnRNPH1 siRNA (siRNPH1) with or without DHT. H and I, immunoblot analysis of PSA, AR, and hnRNPH1 in cytoplasmic and nuclear lysates of siRNPH1-silenced or siControl-transfected MDA-PCa-2b and C4-2B cells, respectively, with or without DHT. J–M, qRT-PCR analysis of AR and PSA transcripts in MDA-PCa-2b and C4-2B cells, respectively, cultured in phenol red–free, charcoal-stripped media and transfected with siControl or siRNPH1 with or without DHT. * and **, statistically significant difference at P < 0.05 and P < 0.01, respectively (n = 3).

hnRNPH1 primes AR binding to AREs on target genes in hormone-dependent and independent fashion

EMSA was employed to examine hnRNPH1 ability to modulate AR binding to three DIG-labeled ds oligonucleotides encompassing the proximal promoter ARE-I (−170) and ARE-II (−394) and the enhancer element ARE-III (−4258) of PSA gene (Fig. 6A). Nuclear extract (NE) proteins binding to all AREs on PSA gene were reduced (∼50%) upon addition of anti-hnRNPH1 antibody (Fig. 6B) in presence or absence of DHT (Fig. 6C) in MDA-PCa-2b cells. Moreover, siRNA silencing of hnRNPH1 reduced such ARE binding with and without DHT (Fig. 6D), attesting the possibility of hnRNPH1 binding to AR/ARE complex in vivo in a hormone-dependent and independent manner.

Figure 6.
Figure 6.

hnRNPH1 mediates hormone-dependent and independent AR binding to AREs in prostate cancer cells. A, schematic representation of three AREs (underlined) encompassing the proximal ARE-I (−170), ARE-II (−394) and the enhancer element ARE-III (−4258) on the PSA promoter. B, nuclear extract of MDA-PCa-2b cells cultured in the complete medium was used for EMSA analysis with labeled DIG-labeled oligonucleotides corresponding to PSA AREs (Supplementary Table S4) in the presence or absence of hnRNPH1 antibody. Specific AR-DNA binding was observed in all AREs (arrowhead), which was reduced by the molar excess of cognate unlabeled ARE oligo. Binding of hnRNPH1 to ARE complex was evident by supershift (arrow) upon addition of a specific hnRNPH1 antibody (n = 2). C, EMSA analysis of hnRNPH1 binding to PSA enhancer ARE-III domain in MDA-PCa-2b cells under DHT treated or deprived conditions. Note the addition of hnRNPH1 antibody markedly inhibited ARE-III binding under both hormone-naïve and induced conditions (n = 3). D, siRNA silencing of hnRNPH1 caused potent reduction of both hormone-naïve and induced ARE-III binding in MDA-PCa-2b cells. E, ChIP assay performed using anti-hnRNPH1 and PCR amplification of sequences flanking AREs of PSA gene (Supplementary Table S5) in presence or absence of DHT (n = 3). F, PCR-amplified exon B, in the DNA-binding domain (DBD), and exons D, E (containing ARE-1 and 2, respectively), and H in the hormone-binding domain (HBD) of AR gene. G, ChIP analysis of hnRNPH1 binding to exons B, D, E, and H of AR gene as influenced by DHT in prostate cancer cells. Input DNA and rabbit control IgG were used as controls (n = 3).

Next, we examined by ChIP analysis whether hnRNPH1 binds AR/ARE complex in prostate cancer cells. Compared with control IgG, hnRNPH1 was found to be part of auxiliary protein–binding complex in all PSA AREs examined in presence or absence of the hormone in both AR-expressing prostate cancer cells (Fig. 6E). Likewise, ChIP analysis demonstrates hnRNPH1 binding to AR on ARE-1 and ARE-2–containing exons D and E of the AR gene in both cell lines (Fig. 6F and G). Taken together, the results suggest a novel hormone-dependent and independent AR coactivation role for hnRNPH1 in prostate cancer cells.

Discussion

Several lines of evidence lend credence to the fact that prostate cancer transforms more rapidly from an indolent to an aggressive phenotype in AA than CA men (28). Differences in underlying biologic and immunobiologic mechanisms were proposed as a possible explanation of the disproportionate burden and progression of prostate cancer in AA men (6, 26). Elucidation of molecular events underlying the progression of prostate cancer among AA men has been hampered by the limitations inherent to both in vitro and in vivo experimental approaches. Our integrated unbiased functional genomics approach, encompassing a combined biracial LCM/SSH/custom prostate cancer–specific cDNA array system, is unique in its ability to detect in vivo gene expression profiles in age-, and Gleason score–matched AA and CA men. This study also highlights several potentially important hnRNPH1–AR axis related molecular mechanisms underlying prostate cancer progression, especially in AA men. The hnRNPH1 is one of the lesser known members of the hnRNP family in terms of its biologic functions. On the basis of quantitative analysis on LCM procured prostate cancer cells, we demonstrate, for the first time, high hnRNPH1 gene expression in prostate tumors of AA men and CA men, but with nearly twice high level in AA men with indolent disease (Gleason 6) compared with CA men. Although not stratified based on race, elevated expression of hnRNPH1 has been shown predominantly in the nuclei of several human cancers, including pancreatic, hepatocellular, and gastric carcinoma (14). The expression of hnRNPK, an hnRNP family member, correlates with Gleason score and poor prognosis in prostate cancer patients (29). Taken together, hnRNPH1 may have potential clinical utility as a biomarker, prognostic indicator, and/or therapeutic target in the management of prostate cancer, especially among AA men.

The underlying molecular mechanism(s) involved in transcriptional regulation of hnRNPs remain elusive (30). In the current study, we demonstrate aberrant expression of hnRNPH1 in AR-expressing, but not AR-null prostate cancer cells. Upregulation of hnRNPH1 transcripts by DHT in prostate cancer cells implicates that it may be a transcriptional target of DHT/AR in prostate tumor cells, presumably through putative ARE on the promoter region. In addition, our in vivo miRNA screening and functional miRNA mimics assays demonstrate that hnRNPH1 is a direct target of the tumor suppressors, miR-22 and miR-212, in prostate tumor cells. Thus, their downregulation triggers accumulation of hnRNPH1 protein in prostate tumor cells. Alternatively, transcriptional upregulation of hnRNPH1 may be regulated through promoter DNA elements, including E2F, AP1, acute myeloid leukemia, and c-myc (30). This notion is strengthened by the fact that hnRNPH1 transcripts are differentially upregulated in SV40 transformed cells in comparison with normal cells (13). Together, androgens and dysregulated miR-22 and miR-212 apparently perturb hnRNPh1 overexpression in prostate tumor cells.

The AR mutation, aberrant expression, and transactivation are implicated in prostate cancer progression (31). Because of racial differences in the androgenic stimulation of the prostate, prostate cancer is detected at a younger age with 81% higher AR protein expression and progresses more rapidly in AA men compared with CA men (32). Notably, our study show in multiple independent cohorts from mined datasets that the hnRNPH1 expression in prostate tumors correlates with AR expression, Gleason score, extraprostatic extensions, disease recurrence, and poor clinical outcome, especially among AA men. We also demonstrate that hnRNPH1 transcripts are partially upregulated by DHT, suggesting a positive feedback loop between AR signaling and hnRNPH1. In prostate cancer cells, silencing of hRNPH1 causes downregulation of AR-FL and AR-V7 transcripts in prostate cancer cells, suggesting a role for this NM protein in transcription and/or splicing of AR in prostate tumor cells. The involvement of hnRNPH1 in alternative pre-mRNA splicing of non-AR targets has been documented (33). In addition, we also show that hnRNPH1 promotes the growth of AR-expressing prostate cancer cells in vitro and in vivo, potentially through transcriptional upregulation of AR-FL and AR-V7. Constitutively active AR splice variant AR-V7 requires full-length AR to mediate its growth-stimulatory action in CRPC cells (34). In agreement with other hnRNP subtypes, ectopic expression of hnRNPK enhances cell proliferation and anchorage-independent growth of breast cancer cells (35). Conversely, hnRNPA1 has been shown to inhibit PC cell growth through suppression of ARA54-enhanced AR transactivation (36), indicating that some hnRNP members may be mutually antagonistic in tumor cells. Together, our data suggest that aberrant expression of hnRNPH1 in prostate tumor cells contributes to disease progression.

The hnRNPH1 induces PSA promoter activity in AR-expressing cells in ligand-dependent and independent manner. AR-V7 lacks the ligand-binding domain and may mediate hnRNPH1 activation of PSA promoter in a ligand-independent manner. This notion is corroborated with the physical interaction between hnRNPH1 and AR in the nuclei of AR-expressing cells. In addition, the AR-independent induction of PSA promoter activity by hnRNPH1 in the presence of DHT suggests it may directly bind specific DNA sequences to regulate transcription. The hnRNPH1 is implicated as a trans-acting factor by direct binding to DNA sequences (37) and estrogen response element (38). This new role was also exhibited by other hnRNP members, including hnRNPA1 (39) and hnRNPK (40), primarily via binding DNA Matrix Attachment Regions (MAR), specific chromatin DNA sequences that interact with NM, and initiate transcription (41). Thus, hnRNPH1 binding to MARs may potentially modulate chromatin state and induce transcription of ARGs, possibly via modifications of RNA complexes and protein–protein interaction. The selective hnRNPH1 binding to ARE and/or MARS leads to AR-independent transcription of ARGs certainly warrants further investigations.

Coregulators interact directly or indirectly with AR and modulate its activity (42). Here we demonstrated nuclear colocalization and physical interaction between hnRNPH1 and AR in prostate cancer cells. In agreement with our findings, direct cell-free binding studies revealed high acceptor sites on NM of the prostate for AR binding (43). Regardless of hormone status, the hnRNPH1-AR binding was primarily detected in the PSA promoter and selective ARE-containing exons within the AR gene. In agreement, AR coactivator Tip60 is recruited to the promoter of the PSA in the absence of androgens (44). We also demonstrated recruitment of AR coactivator SRC-3 to the AR-hnRNPH1 complex. Indeed, AR coactivators SRC-1 and TIF-2 are upregulated in tissue specimens from patients who failed prostate cancer endocrine therapy and that their selective expression is coupled with enhanced activation of the AR signaling in tumor cells (45). Our findings suggest a new coactivation role for hnRNPH1 in AR transactivation in prostate cancer cells under hormone-induced and deprived conditions.

The nonsteroidal antiandrogen bicalutamide is often used as monotherapy or in combination with androgen deprivation therapy (46) for locally advanced or biochemically recurrent prostate cancer to prevent androgen-dependent activation of the AR and upregulation of ARGs (47) by binding to and accelerating degradation of the AR in tumor cells (48). Although this treatment regimen initially exhibits favorable responses, prostate cancer inevitably becomes refractory and develops resistance to bicalutamide (49). As a suppressor of AR transcription and activation, we demonstrate that hnRNP H1 silencing sensitizes prostate cancer cells to the bicalutamide-mediated growth arrest under DHT-deprived conditions, thus further augmenting AR-dependent growth inhibition by bicalutamide in prostate cancer cells. This effect was partially ameliorated by DHT in AR-expressing cells, suggesting that hnRNPH1 overexpression may interfere with response to hormonal therapy via transcriptional upregulation and amplification of AR signaling in prostate tumor cells. The hnRNPH1 is shown to abrogate apoptosis triggered by etoposide and the fluoropyrimidine anticancer drugs (50). Thus, further in vivo studies are required to validate whether hnRNPH1 targeting sensitizes prostate cancer to hormonal therapy.

In conclusion, our study paves the way for further understanding of the complex biology involved in prostate cancer progression. The results suggest selective expression of hnRNPH1 in a subset of prostate tumor cells may confer disease progression via enhancing transcription and activation of AR-FL and AR-V7 in both ligand-dependent and independent manner. The hnRNPH1–AR axis may represent a previously uncharacterized mechanism in prostate cancer patients with advanced disease, especially in AA men.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Authors' Contributions

Conception and design: Y. Yang, A.B. Abdel-Mageed

Development of methodology: Y. Yang, D. Jia, H. Kim, Z.Y. Abd Elmageed, A.B. Abdel-Mageed

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): Y. Yang, D. Jia, H. Kim, Z.Y. Abd Elmageed, A. Datta, R. Davis

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): Y. Yang, D. Jia, H. Kim, Z.Y. Abd Elmageed, S. Srivastav, K. Moroz, B. Crawford, R.S. Hudson, A.B. Abdel-Mageed

Writing, review, and/or revision of the manuscript: Y. Yang, D. Jia, H. Kim, Z.Y. Abd Elmageed, A. Datta, K. Moparty, S. Ambs, A.B. Abdel-Mageed

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): A. Datta, R. Thomas, S. Ambs

Study supervision: R. Thomas, A.B. Abdel-Mageed

Grant Support

The study was partially supported by grants from the American Cancer Society (RSGTCCE-116942; to A.B. Abdel-Mageed) and Department of Defense (W81XWH-09-1-0200; to A.B. Abdel-Mageed).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Acknowledgments

The authors thank Dr. L.W. Chung (Cedars-Sinai Medical Center, Los Angeles, CA) for providing the C4-2B cells and psPSA-luc plasmid, Dr. X.-B. Shi (University of California at Davis, Davis, CA) for providing the control pcDNA3.1 plasmid, and Ms. Jessica A. Daigle for editing the manuscript.

Footnotes

  • Note: Supplementary data for this article are available at Clinical Cancer Research Online (http://clincancerres.aacrjournals.org/).

  • Received July 6, 2015.
  • Revision received September 30, 2015.
  • Accepted October 11, 2015.

References

  1. 1.↵
    1. Chen CD,
    2. Welsbie DS,
    3. Tran C,
    4. Baek SH,
    5. Chen R,
    6. Vessella R,
    7. et al.
    Molecular determinants of resistance to antiandrogen therapy. Nat Med 2014;10:33–9.
    OpenUrl
  2. 2.↵
    1. Heemers HV,
    2. Tindall DJ
    . Androgen receptor (AR) coregulators: a diversity of functions converging on and regulating the AR transcriptional complex. Endocr Rev 2007;28:778–808.
    OpenUrlCrossRefPubMed
  3. 3.↵
    1. Siegel RL,
    2. Miller KD,
    3. Jemal A.
    Cancer Statistics. CA Cancer J Clin 2015;65:5–29.
    OpenUrlCrossRefPubMed
  4. 4.↵
    1. Powell IJ,
    2. Banerjee M,
    3. Novallo M,
    4. Sakr W,
    5. Grignon D,
    6. Wood DP,
    7. et al.
    Prostate cancer biochemical recurrence stage for stage is more frequent among African American than white men with locally advanced but not organ-confined disease. Urology 2000;55:246–51.
    OpenUrlCrossRefPubMed
  5. 5.↵
    1. Powell IJ,
    2. Bollig-Fischer A
    . Minireview: the molecular and genomic basis for prostate cancer health disparities. Mol Endocrinol 2013;27:879–91.
    OpenUrlCrossRefPubMed
  6. 6.↵
    1. Haiman CA,
    2. Patterson N,
    3. Freedman ML,
    4. Myers SR,
    5. Pike MC,
    6. Waliszewska A,
    7. et al.
    Multiple regions within 8q24 independently affect risk for prostate cancer. Nat Genet 2007;39:638–41.
    OpenUrlCrossRefPubMed
  7. 7.↵
    1. Epstein JI,
    2. Pizov G,
    3. Walsh PC
    . Correlation of pathologic findings with progression after radical retropubic prostatectomy. Cancer 1993;71:3582–93.
    OpenUrlCrossRefPubMed
  8. 8.↵
    1. Small D,
    2. Nelkin B,
    3. Vogelstein B
    . Nonrandom distribution of repeated DNA sequences with respect to supercoil loops and the nuclear matrix. Proc Natl Acad Sci U S A 1982;79:5911–5.
    OpenUrlAbstract/FREE Full Text
  9. 9.↵
    1. Leman ES,
    2. Getzenber RH
    . Nuclear structure as a source of cancer specific biomarekers. J Cell Biochem 2008;104:1988–93.
    OpenUrlCrossRefPubMed
  10. 10.↵
    1. Krecic AM,
    2. Swanson MS
    . hnRNP complexes: composition, structure, and function. Curr Opin Cell Biol 1999;11:363–71.
    OpenUrlCrossRefPubMed
  11. 11.↵
    1. Siomi MC,
    2. Eder PS,
    3. Kataoka N,
    4. Wan L,
    5. Liu Q,
    6. Dreyfuss G
    . Transporin-mediated nuclear import of heterogeneous nuclear RNP proteins. J Cell Biol 1997;138:1181–92.
    OpenUrlAbstract/FREE Full Text
  12. 12.↵
    1. Honoré B,
    2. Rasmussen HH,
    3. Vorum H,
    4. Dejgaard K,
    5. Liu X,
    6. Gromov P,
    7. et al.
    Heterogeneous nuclear ribonucleoproteins H, H′, and F are members of a ubiquitously expressed subfamily of related but distinct proteins encoded by genes mapping to different chromosomes. J Biol Chem 1995;270:28780–9.
    OpenUrlAbstract/FREE Full Text
  13. 13.↵
    1. Alkan SA,
    2. Martincic K,
    3. Milcarek C
    . The hnRNPs F and H2 bind to similar sequences to influence gene expression. J Biochem 2006;93:361–71.
    OpenUrl
  14. 14.↵
    1. Honoré B,
    2. Baandrup U,
    3. Vorum H
    . Heterogeneous nuclear ribonucleoproteins F and H/H′ show differential expression in normal and selected cancer tissues. Exp Cell Res 2004;294:199–204.
    OpenUrlCrossRefPubMed
  15. 15.↵
    1. Holzmann K,
    2. Korosec T,
    3. Gerner C,
    4. Grimm R,
    5. Sauermann G
    . Identification of human common nuclear-matrix proteins as heterogeneous nuclear ribonucleoproteins H and H′ by sequencing and mass spectrometry. Eur J Biochem 1997;244:479–86.
    OpenUrlPubMed
  16. 16.↵
    1. Barrach ER
    . Steroid hormone receptor localization in the nuclear matrix: interaction with acceptor sites. J Steroid Biochem 1987;27:115–21.
    OpenUrlCrossRefPubMed
  17. 17.↵
    1. Chou MY,
    2. Rooke N,
    3. Turck CW,
    4. Black DL
    . hnRNP H is a component of a splicing enhancer complex that activates a c-src alternative exon in neuronal cells. Mol Cell Biol 1999;19:67–77.
    OpenUrl
  18. 18.↵
    1. Chen CD,
    2. Kobayashi R,
    3. Helfman DM
    . Binding of hnRNP H to an exonic splicing silencer is involved in the regulation of alternative splicing of the rat beta-tropomyosin gene. Genes Dev 1999;13:593–606.
    OpenUrlAbstract/FREE Full Text
  19. 19.↵
    1. Liu J,
    2. Beqaj S,
    3. Yang Y,
    4. Honoré B,
    5. Schuger L
    . Heterogeneous nuclear ribonucleoprotein-H plays a suppressive role in visceral myogenesis. Mech Dev 2001;104:79–87.
    OpenUrlCrossRefPubMed
  20. 20.↵
    1. Ma L,
    2. Yang Y,
    3. Sikka SC,
    4. Kadowitz PJ,
    5. Ignarro LJ,
    6. Abdel-Mageed AB,
    7. et al.
    Adipose tissue-derived stem cell-seeded small intestinal submucosa for tunica albuginea grafting and reconstruction. Proc Natl Acad Sci U S A 2010;109:2090–5.
    OpenUrl
  21. 21.↵
    1. Abd Elmageed ZY,
    2. Yang Y,
    3. Thomas R,
    4. Ranjan M,
    5. Mondal D,
    6. Moroz K,
    7. et al.
    Neoplastic reprogramming of patient-derived adipose stem cells by prostate cancer cell-associated exosomes. Stem Cells 2014:32:983–97.
    OpenUrlCrossRefPubMed
  22. 22.↵
    1. Abdel-Mageed AB,
    2. Agrawal KC
    . Activation of nuclear factor kappa B: potential role in metallothionein-mediated mitogenic response. Cancer Res 1998;58:2335–8.
    OpenUrlAbstract/FREE Full Text
  23. 23.↵
    1. Cleutjens KB,
    2. van der Korput HA,
    3. van Eekelen CC,
    4. van Rooij HC,
    5. Faber PW,
    6. Trapman J
    . An androgen response element in a far upstream enhancer region is essential for high, androgen regulated activity of the prostate specific antigen promoter. Mol Endocrinol 1997;11:148–61.
    OpenUrlCrossRefPubMed
  24. 24.↵
    1. Smyth GK,
    2. Speed TP
    . Normalization of cDNA microarray data. Methods 2003;31:265–73.
    OpenUrlCrossRefPubMed
  25. 25.↵
    1. Timofeeva OA,
    2. Zhang X,
    3. Ressom HW,
    4. Varghese RS,
    5. Kallakury BV,
    6. Wang K,
    7. et al.
    Enhanced expression of SOS1 is detected in prostate cancer epithelial cells from African-American men. Int J Oncol 2009;35:751–60.
    OpenUrlPubMed
  26. 26.↵
    1. Wallace TA,
    2. Prueitt RL,
    3. Yi M,
    4. Howe TM,
    5. Gillespie JW,
    6. Yfantis HG,
    7. et al.
    Tumor immunobiological differences in prostate cancer between African-American and European-American men. Cancer Res 2008;68:927–36.
    OpenUrlAbstract/FREE Full Text
  27. 27.↵
    1. Glinsky GV,
    2. Glinskii AB,
    3. Stephenson AJ,
    4. Hoffman RM,
    5. Gerald WL
    . Gene expression profiling predicts clinical outcome of prostate cancer. J Clin Invest 2004:113:913–23.
    OpenUrlCrossRefPubMed
  28. 28.↵
    1. Powell IJ,
    2. Bock CH,
    3. Ruterbusch JJ,
    4. Sakr W
    . Evidence supports a faster growth rate and/or earlier transformation to clinically significant prostate cancer in black than in white American men, and influences racial progression and mortality disparity. J Urol 2010;183:1792–6.
    OpenUrlCrossRefPubMed
  29. 29.↵
    1. Barboro P,
    2. Repaci E,
    3. Rubagotti A,
    4. Salvi S,
    5. Boccardo S,
    6. Spina B,
    7. et al.
    Heterogeneous nuclear ribonucleoprotein K: altered pattern of expression associated with diagnosis and prognosis of prostate cancer. Br J Cancer 2009;100:1608–16.
    OpenUrlCrossRefPubMed
  30. 30.↵
    1. Carpenter B,
    2. MacKay C,
    3. Alnabulsi A,
    4. MacKay M,
    5. Telfer C,
    6. Melvin WT,
    7. et al.
    The roles of heterogeneous nuclear ribonucleoproteins in tumor development and progression. Biochem Biophys Acta 2006;1765:85–100.
    OpenUrlPubMed
  31. 31.↵
    1. Chen Y,
    2. Sawyers CL,
    3. Scher HI
    . Targeting the androgen receptor pathway in prostate cancer. Curr Opin Pharmacol 2008;8:440–8.
    OpenUrlCrossRefPubMed
  32. 32.↵
    1. Gaston KE,
    2. Kim D,
    3. Singh S,
    4. Ford OH III.,
    5. Mohler JL
    . Racial differences in androgen receptor protein expression in men with clinically localized prostate cancer. J Urol 2003;170:990–3.
    OpenUrlCrossRefPubMed
  33. 33.↵
    1. Russo A,
    2. Catillo M,
    3. Esposito D,
    4. Briata P,
    5. Pietropaolo C,
    6. Russo G
    . Autoregulatory circuit of human rpL3 expression requires hnRNP H1, NPM and KHSRP. Nucleic Acids Res 2011;39:7576–85.
    OpenUrlAbstract/FREE Full Text
  34. 34.↵
    1. Watson PA,
    2. Chen YF,
    3. Balbas MD,
    4. Wongvipat J,
    5. Socci ND,
    6. Viale A,
    7. et al.
    Constitutively active androgen receptor splice variants expressed in castration-resistant prostate cancer require full-length androgen receptor. Proc Natl Acad Sci U S A 2010;107:16759–65.
    OpenUrlAbstract/FREE Full Text
  35. 35.↵
    1. Mandal M,
    2. Vadlamudi R,
    3. Nguyen D,
    4. Wang RA,
    5. Costa L,
    6. Bagheri-Yarmand R,
    7. et al.
    Growth factors regulate heterogeneous nuclear ribonucleoprotein K expression and function. J Biol Chem 2001;276:9699–704.
    OpenUrlAbstract/FREE Full Text
  36. 36.↵
    1. Yang Z,
    2. Chang YJ,
    3. Miyamoto H,
    4. Yeh S,
    5. Yao JL,
    6. di Sant'Agnese PA,
    7. et al.
    Suppression of androgen receptor transactivation and prostate cancer cell growth by heterogeneous nuclear ribonucleoprotein A1 via interaction with androgen receptor coregulator ARA54. Endocrinology 2007;148:1340–9.
    OpenUrlCrossRefPubMed
  37. 37.↵
    1. Russo A,
    2. Siciliano G,
    3. Catillo M,
    4. Giangrande C,
    5. Amoresano A,
    6. Pucci P,
    7. et al.
    hnRNP H1 and intronic G runs in the splicing control of the human rpL3 gene. Biochim Biophys Acta 2010;1799:419–28.
    OpenUrlPubMed
  38. 38.↵
    1. Chen H,
    2. Hu B,
    3. Gacad MA,
    4. Adams JS
    . Cloning and expression of a novel dominant-negative-acting estrogen response element-binding protein in the heterogeneous nuclear ribonucleoprotein family. J Biol Chem 1998;273:31352–7.
    OpenUrlAbstract/FREE Full Text
  39. 39.↵
    1. Donev RM,
    2. Doneva TA,
    3. Bowen WR,
    4. Sheer D
    . HnRNP A1 binds directly to double-stranded DNA in vitro within a 36 bp sequence. Mol Cell Biochem 2002;233:181–5.
    OpenUrlCrossRefPubMed
  40. 40.↵
    1. Lynch M,
    2. Chen L,
    3. Ravitz MJ,
    4. Mehtani S,
    5. Korenblat K,
    6. Pazin MJ,
    7. et al.
    hnRNP K binds a core polypyrimidine element in the eukaryotic translation initiation factor 4E (eIF4E) promoter, and its regulation of eIF4E contributes to neoplastic transformation. Mol Cell Biol 2005;25:6436–53.
    OpenUrlAbstract/FREE Full Text
  41. 41.↵
    1. Barboro P,
    2. D'Arrigo C,
    3. Repaci E,
    4. Bagnasco L,
    5. Orecchia P,
    6. Carnemolla B,
    7. et al.
    Proteomic analysis of the nuclear matrix in the early stages of rat liver carcinogenesis: identification of differentially expressed and MAR-binding proteins. Exp Cell Res 2009;315:226–39.
    OpenUrlCrossRefPubMed
  42. 42.↵
    1. Wang X,
    2. Yang Y,
    3. Guo X,
    4. Sampson ER,
    5. Hsu CL,
    6. Tsai MY,
    7. et al.
    Suppression of androgen receptor transactivation by Pyk2 via interaction and phosphorylation of the ARA55 coregulator. J Biol Chem 2002;277:15426–31.
    OpenUrlAbstract/FREE Full Text
  43. 43.↵
    1. Barrack ER
    . The nuclear matrix of the prostate contains acceptor sites for androgen receptor. Endocrinology 1983;113:430–2.
    OpenUrlCrossRefPubMed
  44. 44.↵
    1. Halkidou K,
    2. Gnanapragasam VJ,
    3. Mehta PB,
    4. Logan IR,
    5. Brady ME,
    6. Cook S,
    7. et al.
    Expression of Tip60, an androgen receptor coactivator, and its role in prostate cancer development. Oncogene 2003;22:2466–77.
    OpenUrlCrossRefPubMed
  45. 45.↵
    1. Culig Z,
    2. Comuzzi B,
    3. Steiner H,
    4. Bartsch G,
    5. Hobisch A
    . Expression and function of androgen receptor coactivators in prostate cancer. J Steroid Biochem Mol Biol 2004;92:265–71.
    OpenUrlCrossRefPubMed
  46. 46.↵
    1. Blackledge G,
    2. Kolvenbag G,
    3. Nash A
    . Bicalutamide: a new antiandrogen for use in combination with castration for patients with advanced prostate cancer. Anticancer Drugs 1996;7:27–34.
    OpenUrlCrossRefPubMed
  47. 47.↵
    1. Furr BJ,
    2. Tucker H
    . The preclinical development of bicalutamide: pharmacodynamics and mechanism of action. Urology 1996;47:13–25.
    OpenUrlCrossRefPubMed
  48. 48.↵
    1. Waller AS,
    2. Sharrard RM,
    3. Berthon P,
    4. Maitland NJ
    . Androgen receptor localization and turnover in human prostate epithelium treated with the antiandrogen, Casodex. J Mol Endocrinol 2000;24:339–51.
    OpenUrlAbstract
  49. 49.↵
    1. Hodgson MC,
    2. Astapova I,
    3. Hollenberg AN,
    4. Balk SP
    . Activity of androgen receptor antagonist bicalutamide in prostate cancer cells is independent of NCoR and SMRT corepressors. Cancer Res 2007;67:8388–95.
    OpenUrlAbstract/FREE Full Text
  50. 50.↵
    1. Rauch J,
    2. O'Neill E,
    3. Mack B,
    4. Matthias C,
    5. Munz M,
    6. Kolch W,
    7. et al.
    Heterogeneous nuclear ribonucleoprotein H blocks MST2-mediated apoptosis in cancer cells by regulating A-Raf transcription. Cancer Res 2010;70:1679–88.
    OpenUrlAbstract/FREE Full Text
View Abstract
Back to top
Clinical Cancer Research: 22 (7)
April 2016
Volume 22, Issue 7

Sign up for alerts

View this article with LENS

Open full page PDF
Article Alerts
Email Article
Citation Tools
Dysregulation of miR-212 Promotes Castration Resistance through hnRNPH1-Mediated Regulation of AR and AR-V7: Implications for Racial Disparity of Prostate Cancer
Yijun Yang, Dingwu Jia, Hogyoung Kim, Zakaria Y. Abd Elmageed, Amrita Datta, Rodney Davis, Sudesh Srivastav, Krzysztof Moroz, Byron E. Crawford, Krishnarao Moparty, Raju Thomas, Robert S. Hudson, Stefan Ambs and Asim B. Abdel-Mageed
Clin Cancer Res April 1 2016 (22) (7) 1744-1756; DOI: 10.1158/1078-0432.CCR-15-1606
del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo

Jump to section

Advertisement