Transcriptional coactivator SUB1 expression in prostate cancer

The expression profiling and transcriptome sequence analysis showed upregulation of SUB1 in metastatic prostate cancer (Figure 2a). Moreover, The Cancer Genome Atlas (TCGA) data show that SUB1 is overexpressed in metastatic prostate adenocarcinoma (Figure 2b). To validate this observation, we performed qPCR using RNA from multiple prostate cancer and benign tissues and confirmed increased expression of SUB1 in metastatic prostate cancer tissues relative to benign prostate samples (Figure 2c). Immunoblot analysis using SUB1-specific antibody (Figure 2d) indicated SUB1 protein overexpression. Similarly, elevated levels of SUB1 protein was observed in metastatic prostate cancer cell lines relative to benign cell lines (Figure 2e). Additionally, we investigated SUB1 protein expression in a large number of prostate cancer samples by immunohistochemical analysis, which showed weak or no reactivity in many benign tissues but stronger staining in aggressive prostate cancer tissue and metastatic prostate tumors (Figure 2f).

Open in a separate window

Figure 2

SUB1 is overexpressed in aggressive prostate cancer. (a) SUB1 gene expression from next-generation RNA sequencing (RPKM (log2)) data from benign, prostate carcinoma (PCa) and metastatic prostate cancer (MET) tissues. (b) Expression of SUB1 in normal prostate, primary tumor and metastatic tumor samples from TCGA. (c) qPCR of SUB1 using RNA from benign, PCa and MET tissues. (d) SUB1 protein expression by immunoblot analysis of prostate tissue extracts using SUB1 antibody. β-Actin was used as a loading control. (e) Immunoblot analysis of SUB1 in prostate cancer cell lines. GAPDH was used as a loading control. (f) Immunohistochemical analysis of SUB1 in benign prostate epithelia (top), primary PCa (middle) and metastatic PCa (bottom).

SUB1 expression is essential for prostate cancer cell proliferation and invasion

To determine the functional significance of SUB1 expression in prostate cancer, we perturbed SUB1 levels in prostate cells and investigated the effect of this modulation on cell proliferation, migration and invasion. We used both transient RNA interference and stable knockdown strategies targeting SUB1 in aggressive prostate cancer cell lines DU145 and PC3 and hormone-responsive LnCaP and VCaP cells. The efficiency of SUB1 knockdowns were confirmed by immunoblot (Figure 3a) and qPCR (Supplementary Figures S3a–g) analyses. We observed significant decrease in cell proliferation upon transient knockdown of SUB1 compared with control cells transfected with non-T small interfering RNAs (siRNAs) (Figure 3a). Next, we tested cell motility after stable SUB1 knockdown in prostate cancer cells using wound healing assay. The efficiency of SUB1 stable knockdowns were confirmed by immunoblot (Supplementary Figure S4a). SUB1 knockdown showed a wider wound area 24h post wound generation relative to control cells, the delayed time to heal indicating an inability of SUB1 knockdown cells to migrate (Supplementary Figures S4b and c). Additionally, SUB1 knockdown in DU145 and PC3 reduced the invasive potential of these cells as assessed by Boyden chamber Matrigel invasion assay (Figure 3b). Further we tested for colony formation after transient knockdown of SUB1 and colonies were quantified (Figures 3c and d). Taken together, these observations demonstrate the involvement of SUB1 in the proliferation, migration, invasion and colony formation of prostate cancer cells in vitro.

Open in a separate window

Figure 3

SUB1 is essential for prostate cancer cell proliferation and invasion. (a) Transient knockdown of SUB1 in prostate cancer cell lines reduces prostate cancer cell proliferation. Immunoblot analysis of protein using lysates from prostate cancer cell lines DU145, PC3, LnCaP and VCaP treated with two specific and independent SUB1 siRNA duplexes. β-Actin was used as a loading control. Cell proliferation assay of these cells transfected with either SUB1 siRNA duplex or Non-T siRNA control. (b) Knockdown of SUB1 reduces DU145 and PC3 cell invasion. Boyden chamber Matrigel invasion assay was performed using DU145 or PC3 cells in which SUB1 was transiently knocked down using two independent SUB1 siRNA duplexes. Non-T siRNA-treated cells served as control. Invaded cells were stained with crystal violet and the absorbance was measured at 560nm. (c) Representative images of colony formation assay. (d) Quantification of colonies formed. The colony formation efficiency was significantly reduced in both SUB1 siRNA1- and 2-treated cells as compared with untreated and Non-T siRNA controls.

SUB1 modulates gene expression in prostate cancer

To evaluate SUB1-mediated effects in prostate cancer progression, we performed gene expression analysis using RNA from SUB1 knockdown prostate cell lines. We identified multiple genes that were modulated upon SUB1 knockdown including PLK1, C-MYC, BUB1B and CDKN1B, among others (Figure 4a). PLK1 and C-MYC are known to have a role in cell proliferation, invasion and metastasis.^{41, 42} We validated the activation of CDKN1B and the downregulation of PLK1, C-MYC and BUB1B, both at the mRNA and protein levels upon SUB1 knockdowns (Supplementary Figures S5a–h, Figure 4b and Supplementary Figures S6a–c) and observed induction of these genes upon SUB1 overexpression in RWPE cells (Figure 4c). TCGA data show that PLK1 and BUB1B are overexpressed in metastatic prostate adenocarcinoma (Supplementary Figures S7a and b). Additionally, we validated both PLK1 mRNA and protein levels across benign, prostrate carcinoma and metastatic prostate cancer tissues by qPCR and immunoblot analysis, respectively (Supplementary Figures S7c and d), which confirmed the direct correlation between SUB1 and PLK1. Additionally, SUB1 and PLK1 mRNA levels are correlated in prostate cancer cell lines (Supplementary Figures S7e and f). Further, we generated stable SUB1-expressing RWPE cells using lentivirus (Supplementary Figure S8). Overexpression of SUB1 resulted in increased cell proliferation (Supplementary Figure S8 and Figure 4d) and invasion (Figure 4e). Furthermore, SUB1 overexpression elevated PLK1 and C-MYC expression, and reduced CDKN1B expression (Figure 4c and Supplementary Figure S6c), showing SUB1 dysregulation triggers alterations in critical oncogenes and tumor suppressors in prostate cancer. To verify the role of PLK1 in cell proliferation and invasion, we treated stable SUB1-overexpressing RWPE cells with PLK1 siRNA SMARTpool or PLK1 inhibitor (volasertib (BI6727)) and analyzed Myc-DDK-tagged SUB1 and PLK1 (Figure 4d inset). Both PLK1 knockdown and PLK1 inhibitor decreased cell proliferation (Figure 4d) and Matrigel invasion (Figure 4e). Thus, these consolidated observations underscore a downstream role for PLK1, C-MYC and CDKN1B in SUB1-mediated prostate cancer cell proliferation and invasion.

Open in a separate window

Figure 4

SUB1 modulates PLK1 expression in prostate cancer cells. (a) Heatmap of select genes in stable SUB1 knockdown PC3 cells. (b) Immunoblot analysis showing SUB1, PLK1 and C-MYC in prostate cancer cells after transient knockdown of SUB1. (c) Immunoblot analysis showing Myc-DDK-tagged SUB1, PLK1 and C-MYC in stable RWPE-SUB1- (Myc-DDK-tagged) overexpressing cells. (d and e) Immunoblot, cell proliferation and Matrigel invasion assays were performed using lenti-lacZ or lenti-SUB1 cells. Untreated cells, non-T siRNA, PLK1-specific siRNA SMARTpool-treated cells or in the presence of PLK1 inhibitor (volasertib (BI6727)) were used. Invaded cells were stained with crystal violet and the absorbance was measured at 560nm. Inset, photomicrographs of invaded cells.

Earlier studies suggest the importance of SUB1 in regulating transcription in vivo. For example, SUB1 enhances transcriptional activation by the activators GCN5 and HAP4 in yeast,⁴³ and stimulates transcription in vitro with diverse kinds of activators, including SMYD3,³² VP16,^{24, 44} BRCA-1⁴⁵ and octomer transcription factor-1,⁴⁶ possibly by facilitating assembly of the preinitiation complex. First, we compared motif occurrences within 500bp upstream regions of downregulated genes and undifferentially expressed genes (when SUB1 is knocked down) using a computer program called CLOVER.⁴⁷ Next, to validate the binding of SUB1 to PLK1, BUB1B and C-MYC promoters, we conducted chromatin immunoprecipitation (ChIP) assays using commercially available anti-SUB1 or -DDK antibodies in stable RWPE cells overexpressing lacz or Myc-DDK-tagged SUB1. As expected, SUB1 is enriched at PLK1, C-MYC and BUB1B promoter regions (Figures 5a–c). These data demonstrate that PLK1, C-MYC and BUB1B promoters are transcriptionally activated by SUB1 in prostate cancer.

Open in a separate window

Figure 5

SUB1 transactivates gene expression by directly binding to specific motifs of their promoter regions. ChIP-PCR analysis for the SUB1 occupancy on (a) PLK1, (b) C-MYC and (c) BUB1B promoters in SUB1- or lacZ-overexpressing RWPE cells. ChIP was performed using antibodies against SUB1-, Myc-DDK-tagged SUB1 and a control IgG. Inset: Schematic representation of the respective genomic regions showing gene and amplicon positions.

Benign and tumor tissues

In this study, we used tissues from clinically localized prostate cancer patients who underwent radical prostatectomy. Samples were also obtained from androgen-independent metastatic prostate cancer patients from a rapid autopsy program through the University of Michigan Prostate SPORE Tissue Core as described previously.^{71, 72} The detailed clinical and pathological data are maintained in a secure relational database. The Institutional Review Board at the University of Michigan Medical School approved this study. Both radical prostatectomy series and the rapid autopsy program are part of the University of Michigan Prostate SPORE Tissue Core.

Gene expression from TCGA

The patients clinical data for prostate adenocarcinoma were downloaded using TCGA assembler.⁷³ However, downloaded data comprised of only tumor pathologic and node pathologic information. Thus, based on tumor pathologic and node pathologic data as per 'https://cancerstaging.org/references-tools/quickreferences/Documents/ProstateSmall.pdf', samples were categorized into primary and metastatic tumor. Afterwards, level3 TCGA RNA-seq data (including raw_read_count and scaled_estimate for each sample) for all primary tumor, metastatic tumor and matched normal samples were downloaded using TCGA assembler. Transcript per million values for each gene was obtained by multiplying scaled_estimate by 1000000. Boxplot was generated using R (https://cran.r-project.org/).

Immunohistochemistry

Benign and prostate cancer tissues were obtained from the radical prostatectomy series at the University of Michigan and from the Rapid Autopsy Program, both part of the University of Michigan Prostate SPORE programs, through appropriate informed consent. Institutional Review Board approval was obtained to procure and analyze the tissues used in this study. Immunohistochemistry was carried out to evaluate SUB1 expression using rabbit polyclonal antibody against SUB1 (Novus Biologicals, Littleton, CO, USA; cat. no. NBP1-82454). Immunohistochemistry was performed using an automated protocol developed for the DISCOVERY XT automated slide staining system (Ventana Medical Systems Inc., Tucson, AZ, USA) using Ultramap anti-rabbit HRP (cat. no. 760-4315; Ventana Medical Systems Inc.) and was detected using ChromoMap DAB (cat. no. 760-159; Ventana Medical Systems Inc.). Hematoxylin II (cat. no. 790-2208; Ventana-Roche, Tucson, AZ, USA) was used as the counterstain. The study pathologist Dr Kunju (PK) evaluated the immunohistochemical staining.

Immunoblot analyses

Antibodies used in the study are listed in Supplementary Table S1. All antibodies were used at dilutions optimized in our laboratory. For immunoblot analysis, 10μg protein samples were separated on a sodium dodecyl sulfate–polyacrylamide gel electrophoresis and transferred onto Immobilon-P PVDF membrane (EMD Millipore, Billerica, MA, USA). The membrane was incubated for 1h in blocking buffer (Tris-buffered saline, 0.1% Tween (TBS-T), 5% nonfat dry milk), followed by incubation overnight at 4°C with the primary antibody. After a wash with TBS-T, the blot was incubated with horseradish peroxidase-conjugated secondary antibody and signals were visualized by Luminata Crescendo chemiluminescence western blotting substrate as per the manufacturer's protocol (EMD Millipore).

Gene expression analysis

Global gene expression data was generated using RNA isolated from SUB1 siRNA knockdown PC3 and non-target control cells in profiling analysis as well as in transcriptome sequencing analysis.⁶ Expression profiling was performed using the Agilent Whole Human Genome Oligo Microarray (Agilent, Santa Clara, CA, USA) and the analysis was performed according to the manufacturer's protocol. A bioconductor package ‘agilp'⁷⁴ was used to extract and normalize raw data from two-channel experiment arrays. Loess normalization was applied on each array. The gene expression profiling data has been deposited at gene expression omnibus (GSE74895). The differential expression of each gene was estimated by subtracting loess normalized log-2-transformed signal intensity of control sample from that of knockdown sample. Genes with absolute fold change of 2 were selected as differentially expressed genes. The heatmap.2 function of R package ‘gplots' was used to create the heat map. To measure mRNA expression levels, total RNA was isolated from prostate cancer cell lines and prostate cancer tissue samples using the RNeasy Mini Kit (Qiagen, Valencia, CA, USA). qPCR was performed as described.³⁷ All primers were synthesized by Integrated DNA Technologies (Coralville, IA, USA) and PCR reactions were performed in triplicate. Primer sequences are listed in Supplementary Table S2.

RNA interference and miRNA transfection

The siRNA duplexes (duplex 1, cat. no. SI04174835 and duplex 2, cat. no. SI00678580) used to inhibit SUB1 expression were purchased from Qiagen, and PLK1 SiGenome SMARTpool (cat. no. M-003290-01-0005) was purchased from GE Dharmacon (Lafayette, CO, USA). Precursors of respective microRNAs and negative controls were purchased from Ambion (Life Technologies). Transfections were performed either with oligofectamine or Lipofectamine RNAiMAX (Life Technologies). SUB1 shRNA (pGipz SUB1-shRNA 1 (V2LHS_85556) and SUB1-shRNA2 (V3LHS_331788)) were purchased from GE Dharmacon. Lentiviruses for these stable knockdowns were generated by the University of Michigan Vector Core. For miRNA transfection or RNA inference, we plated prostate cancer cells at 1 × 10⁵ cells per well in a 6-well plate, and after 12h, the cells were transfected either with siRNA duplexes or miRNAs. A second identical transfection was performed 24h later. Seventy-two hours after the first transfection, cells were harvested for RNA isolation or immunoblot analysis.

In vitro overexpression

SUB1 cDNA (Origene Technologies, Rockville, MD, USA; cat. no. RC204999; Myc-DDK tagged) was cloned into Gateway expression system (Life Technologies). To generate lentiviral constructs, PCR8-SUB1 (Myc-DDK tagged) was recombined with pLenti6/V5-Dest (Life Technologies) using LR Clonase II (Life Technologies). Lentiviruses were generated by the University of Michigan Vector Core. Benign immortalized prostate cells (RWPE) were infected with lentiviruses expressing SUB1 or lacZ, and stable clones were selected with 3.5μg/ml blasticidin (Santa Cruz Biotechnology Inc., Dallas, TX, USA).

Cell proliferation assays

Cell proliferation was measured by cell counting. For this, transient SUB1 and PLK1 knockdown and stable cells overexpressing SUB1 were used. After 72h of transfection using specific siRNA, the cells were trypsinized and seeded at a density of 10000 cells per well in 24-well plates (n=3). Non-T siRNA-treated cells were served as controls. The stable RWPE lacZ- or SUB1-overexpressing cells were plated at the same density as mentioned earlier, trypsinized and counted at specified time points by Z2 Coulter particle counter (Beckman Coulter, Brea, CA, USA). Each experiment has been performed with three replicates per sample.

Wound healing assay

DU145, PC3 scramble shRNA or SUB1 stable knockdown cells were seeded in 6-well plates in growth medium containing 10% fetal bovine serum and puromycin (10μg/ml) for DU145 and PC3, and then allowed to grow to confluent monolayer. For DU145 and PC3, the cells were serum starved for 12h and replenished with 10% fetal bovine serum-RPMI medium. The wound-induced migration was triggered by scraping the cells with a 200ul pipette tip, washed with Dulbecco's phosphate-buffered saline and replenished with respective medium. The wound was imaged immediately (0h) and at 24h with an inverted phase-contrast microscope under × 4 objective.

Matrigel invasion assay

Matrigel invasion assays were performed as described earlier.^{37, 75} Various test cells were seeded onto BD BioCoat Matrigel matrix (Corning Life Sciences, Tewksbury, MA, USA) in the upper chamber of a 24-well culture plate. The lower chamber containing respective medium was supplemented with 10% serum as a chemoattractant. After 48h, the noninvading cells and Matrigel matrix were gently removed with a cotton swab. Invasive cells located on the lower side of the chamber were stained with 0.2% crystal violet in methanol, air-dried and photographed using an inverted microscope (× 4). Invasion was quantified by colorimetric assay. For colorimetric assays, the inserts were treated with 150μl of 10% acetic acid and the absorbance measured at 560nm.

Colony formation assay

After 72h of transfection, untreated, Non-T siRNA/miRNA- or miR-101-treated cells were counted and seeded 800 cells per one well of 6-well plates (triplicate) and incubated at 37°C with 5% CO₂ for 7–10 days. Colonies were fixed with 10% (v/v) ethanol for 30min and stained with crystal violet (Sigma-Aldrich, St Louis, MO, USA) for 20min. Then, the photographs of the colonies were taken using Amersham Imager 600RGB (GE Healthcare Life Sciences, Pittsburgh, PA, USA). Colony quantification was carried out using ImageQuant TL Colony v.8.1 software (GE Healthcare Life Sciences).

miR reporter luciferase assays

Wild-type or mutant 3′-UTR of SUB1 were cloned into the pMIR-REPORT miRNA Expression Reporter Vector (Life Technologies). HEK293 cells were co-transfected with pre-miR-101 or controls and wild-type or mutant 3′-UTR-luc, as well as pRL-TK vector as an internal control for luciferase activity. Forty eight hours post-transfection, the cells were lysed and luciferase assays were conducted using the dual luciferase assay system (Promega, Madison, WI, USA). Each experiment was performed in triplicate.

ChIP assays

ChIP assays were carried out with respective antibodies (Supplementary Table S1) using the EZ-Magna ChIP Kit (EMD Millipore, Billerica, MA, USA) as described.³⁷ The primer sequences for the promoters analyzed are provided in Supplementary Table S3.

Chicken CAM assay

The CAM assay for local cell invasion, intravasation, metastasis and tumor (or xenograft) formation was performed as described previously.^{37, 48, 75} After 3 days of implanting the cells in each egg, lower CAM was harvested and extraembryonic tumors were isolated and weighed. For metastasis assay, the embryonic livers were harvested on day 18 of embryonic growth and analyzed for the presence of tumor cells by quantitative human Alu-specific PCR. Genomic DNA from lower CAM and livers were prepared using Puregene DNA purification system (Qiagen) and quantification of human Alu was performed as described.^{37, 48, 75} An average of eight eggs per group was used in each experiment.

Tumor xenograft model

All procedures involving mice were approved by the University Committee on Use and Care of Animals (UCUCA) at the University of Michigan and conform to all regulatory standards. To evaluate the role of SUB1 in tumor formation in vivo, we propagated stable SUB1 knockdown PC3 cells using two independent shRNAs and Non-T shRNA control cells, and inoculated 1 × 10⁶ cells subcutaneously into the dorsal flank of 5-week-old male athymic nude mice (n=8 for each group; Harlan Laboratories, Evigo Indianapolis, IN, USA). The tumor data obtained using scramble cells is the same as that used in an earlier study as SUB1 tumor xenograft study was conducted simultaneously using common control animals.³⁷ Tumor size was measured biweekly, and tumor volumes were calculated using the formula (π/6) (L × W2), where L is the length and W is the width of the tumor. After 5 weeks, mice from different groups were killed, and then the tumors were photographed, weighed and plotted.

We thank Twani Dhar, Sherin John, Sivakumar Nallasivam, Samuel Lee and Sai Akshaya Hodigere Balasubramanya for their assistance. We also thank the University of Michigan Vector Core for generating lentivirus. This work is supported, in part, by National Institutes of Health R01CA157845. SV is also supported by R01CA154980 and the University of Michigan Prostate Cancer SPORE Career Development award (to SV). Gene expression data has been deposited in the GEO database under accession number GSE74895

Author contributions

BVSKC and SV designed the experiments. BVSKC, MTG, SSP, ADR, SA, JS, SLC, XJ, NP and SV performed the experimental work. LPK and SD, DSC and MC performed statistical analysis. AMC was involved in fruitful discussions and suggestions. BVSKC and SV wrote the paper. All authors discussed the results and commented on the manuscript when needed.