The human prostate cancer cell lines, LNCaP, DU145, 22RV1 and PC-3, were purchased from the American Type Culture Collection (ATCC; Manassas, VA, USA). The PC-3M cell line was purchased from the National Platform of Experimental Cell Resources for Sci-Tech (Beijing, China). Cells were cultured in RPMI-1640 (HyClone, Logan, UT, USA) containing 10% fetal bovine serum (FBS; Gibco, Carlsbad, CA, USA) with 1% antibiotics. The human prostate epithelial cell line RWPE-1 was purchased from ATCC, and was maintained in keratinocyte serum-free medium supplemented with 0.05 mg/ml bovine pituitary extract and 5 ng/ml epidermal growth factor (Invitrogen, Carlsbad, CA, USA). All cells were cultured at 37°C in a humidified incubator with 5% CO₂ and 95% O₂.

HMGN5 shRNA sequences and construction of the lentivirus were identical to those used in a previous study (23). PC-3 cells were seeded into 6-well plates and grown to 70% confluency. On the day of infection, the purified lentivirus (shCtrl or shHMGN5) was added to the cells at a multiplicity of infection (MOI) of 20 for 6 h, and cells were washed twice with medium. Successful knockdown of HMGN5 was analyzed by western blotting and real-time quantitative PCR (qPCR) after 72 h.

Human HMGN5 cDNA was obtained from the German Science Centre for Genome Research (RZPD), and the fragments containing the HMGN5 coding sequence were subcloned between the AgeI and NheI sites of the GV205 lentiviral vector (synthesized by GeneChem Corporation, Shanghai, China). The lentiviral construct was verified by standard DNA sequencing.

Total RNA from the cultured cells was isolated using TRIzol reagent (Invitrogen) following the manufacturer’s protocol. We reverse-transcribed total RNA (3 μg) using the reverse transcription system (Promega, Madison, WI, USA). qPCR was performed using SYBR-Green PCR Mix (Roche, Indianapolis, IN, USA) in an Applied Biosystems 7300 Fast Real-Time PCR system. The primer sequences for real-time qPCR were as follows: HMGN5, GCAGTCAGGCAGTGACTGCCTTCG (forward) and CCCTTTTCTGTGGCATCTTC (reverse); GAPDH, CAGTCAGCCGCATCTTCTTTT (forward) and GTGACCAGGCGCCCAATAC (reverse). Gene expression analysis was performed using the comparative ΔΔCt method; expression was normalized to GAPDH.

Protein lysate was prepared by homogenization in RIPA lysis buffer containing phosphatase and protease inhibitors. Western blot assay was performed according to previously described protocols (24). Anti-HMGN5 (Sigma, St. Louis, MO, USA), -ERK1/2, -pERK1/2, -GAPDH and -PARP (Cell Signaling Technology, Inc., Beverly, MA, USA) were used as primary antibodies, and either goat anti-mouse IgG or goat anti-rabbit IgG (Sigma) was used as secondary antibody. The membrane was visualized using the ECL detection system (GE Healthcare Biosciences, Piscataway, NJ, USA).

Cell proliferation was assessed using the CellTiter-Blue reagent (Promega) according to the manufacturer’s instructions. Cells (2,000–5,000/well) were seeded into a 96-well plate and cultured for 12 h before new medium containing different doses of gemcitabine was added; cells were cultured for an additional 48 h, then 20 μl of CellTiter-Blue reagent was added to each well. Plates were incubated for 4 h at 37°C, then fluorescence was recorded at an excitation wavelength of 560 nm and an emission wavelength range of 590 nm using a Labsystems Fluoroskan Ascent plate reader (Thermo Fisher Scientific, Boston, MA, USA).

The cell invasion assay was performed with 8-μm cell culture inserts (Millipore Corporation, Billerica, MA, USA) coated with 60 μl ECM gel (BD Biosciences, Bedford, MA, USA) mixed with RPMI-1640 serum-free medium in 1:4 dilution for 5 h at 37°C. Infected PC-3 cells (5×10⁴) were resuspended in 100 μl serum-free RPMI-1640 and placed into Matrigel, and the lower chamber of the Transwell was filled with 600 μl of RPMI-1640 containing 10% FBS before incubating cells at 37°C for 24 h. The Transwells were removed from the 24-well plates and fixed with 100% methanol for 30 min, and then stained with 0.1% crystal violet for 30 min. Non-invaded cells on the top of the Transwell were removed with a cotton swab. Five visual fields were chosen randomly under a light microscope at ×200 magnification, and invaded cells were counted. The data are expressed as means ± SD (standard deviation).

After infection for 24 h, cells (1,000/well) were seeded into a 6-well plate, and culture medium was changed every 3 days for 2 weeks. Colonies (50 or more cells/colony) were counted after staining with gentian violet.

The apoptosis assay was performed using the Apo-ONE^® Homogeneous Caspase-3/7 kit (Promega) according to the manufacturer’s instructions. Following infection for 24 h, the PC-3 cells (5,000/well) were seeded into a 96-well plate and cultured for 48 h, and then 100 μl of Apo-ONE Caspase-3/7 reagent was added to each well, and the plate was incubated at room temperature for 2 h before measuring the fluorescence of each well at an excitation wavelength of 480 nm and an emission wavelength range of 530 nm.

All data groups were analyzed by one-way ANOVA using SPSS 13.0 (SPSS, Inc., Chicago, IL, USA), and the graphs were generated in MS Excel. Data are expressed as means ± SD. A probability (P) value of <0.05 was assigned to indicate a statistically significant difference when compared with the control and is indicated by an asterisk in the figures.

HMGN5 expression was determined using real-time qPCR and western blot analysis in the prostate cancer cell lines, including LNCaP, DU145, 22RV1, PC-3 and PC-3M, as well as human prostate epithelial cell line RWPE-1. Both the mRNA and protein levels of HMGN5 were higher in the prostate cancer cell lines than the levels in the control human prostate epithelial cell line, and the HMGN5 protein level was correlated with its mRNA level (Fig. 1A and B).

To explore the function of HMGN5 in prostate cancer, we used a loss-of-function approach. PC-3 cells were infected with the shHMGN5 or shCtrl lentiviral vectors for 72 h; the shHMGN5 reduced HMGN5 expression at both the mRNA and protein levels (Fig. 2A). Cell proliferation assay results showed that the PC-3 cells with shHMGN5 exhibited a consistent decrease over the time course examined, with a decrease in proliferation of 30.03% at day 4 and 34.91% at day 5, compared with the PC-3 cells with shCtrl (Fig. 2B). To determine whether the downregulation of HMGN5 promoted PC-3 cell apoptosis, we used a homogeneous caspase-3/7 analysis, in which caspase-3/7 activity denoted the apoptosis level. The results showed that shHMGN5 induced the activity of caspase-3/7 compared with shCtrl in the PC-3 cells (Fig. 2C). Next, we assessed the role of HMGN5 in PC-3 cell invasion. The cell invasion assay indicated that there were fewer PC-3 cells in the shHMGN5-infected group when compared with the shCtrl group; the number of cells crossing the Matrigel was 60.3±6.4 in the shHMGN5 group vs. 119.4±10.8 in the shCtrl group (Fig. 2D). To further assess the tumor promoter functions of HMGN5 in prostate cancer cells, infected PC-3 cells were assessed with anchorage-dependent colony formation assays. After 2 weeks of culture, the number of colonies in the PC-3 cells infected with shHMGN5 was significantly less than that in the PC-3 with the shCtrl (Fig. 2E). These data provide evidence that HMGN5 promotes cell proliferation, invasion and clonogenicity, and antagonizes cell apoptosis.

To further investigate the function of HMGN5 in prostate cancer tumorigenesis, we constructed an HMGN5-expressing lentiviral vector, and the human prostate epithelial cell line RWPE-1 was engineered to stably express HMGN5. Western blot analysis showed that the HMGN5-transfected clones (RWPE-1-HMGN5) had a high level of HMGN5 protein compared with the GV205 (empty vector)-transfected controls (RWPE-1-Ctrl) (Fig. 3A). Then, the colony formation assay and cell invasion assay were performed, which mimic crucial events in tumorigenesis and cancer progression. We found that the RWPE-1-HMGN5 cells formed more, larger colonies compared with the RWPE-1-Ctrl cells in monolayer culture for 14 days (Fig. 3B). Furthermore, ectopic HMGN5 expression significantly increased the migration of the RWPE-1 cells in the cell invasion assay (Fig. 3C).

Abnormal activation of the MAPK signaling pathway plays an important role in prostate cancer tumorigenesis and progress; thus, we studied whether HMGN5 contributes to prostate cancer development by activating the MAPK signaling pathway. Western blot results showed that the level of pERK1/2 protein, which indicates the activity of the MAPK signaling pathway, was significantly increased in the RWPE-1 cells ectopically expressing HMGN5, and was strongly decreased in the PC-3 cells with HMGN5 knockdown when compared with the control (Fig. 3D). However, for ERK1/2, there was no difference in expression between the RWPE-1-HMGN5 and RWPE-1-Ctrl cells or the PC-3 with shHMGN5 and PC-3 with shCtrl cells (Fig. 3D). We observed the reverse effect following treatment with the ERK inhibitor U0126. As shown in Fig. 3B and C, the invasion and clonogenicity mediated by ectopic HMGN5 expression in the RWPE-1 cells was attenuated by U0126. These results demonstrated that HMGN5 contributes to prostate cancer tumorigenesis and progression by activating the MAPK signaling pathway.

Our previous study showed that gemcitabine downregulates HMGN5 mRNA levels, thus this next study focused on the relationship between the HMGN5 expression level and the response to gemcitabine in prostate cancer. First, we assessed the effect of gemcitabine on prostate cell lines with various levels of HMGN5 (Fig. 4A). Cell viability was detected by the CellTiter-Blue assay after 48 h of treatment with different doses of gemcitabine. Notably, cells with a higher level of HMGN5 expression (PC-3, PC-3M and LNCaP) showed more sensitivity to gemcitabine, and cells with a lower level of HMGN5 expression (DU145, 22RV1) showed less sensitivity to gemcitabine; cells with the lowest level of HMGN5 expression (RWPE-1) showed resistance to gemcitabine (Fig. 4A). Furthermore, downregulation of HMGN5 in the PC-3 cells with shHMGN5 resulted in the resistance of PC-3 cells to gemcitabine. PC-3 cells with shCtrl were sensitive to gemcitabine (Fig. 4B). Upregulation of HMGN5 in DU145 cells induced their sensitivity to gemcitabine (Fig. 4C). Next, we explored the mechanism of HMGN5-mediated sensitivity to gemcitabine. We first tested whether gemcitabine treatment downregulated the HMGN5 protein level in PC-3 cells. After 48 h of treatment with different doses of gemcitabine, western blotting showed that gemcitabine downregulated HMGN5 expression in a dose-dependent manner (Fig. 4D), meanwhile, consistent with HMGN5 expression, the MAPK signaling pathway marker, pERK1/2, was also downregulated by gemcitabine. Cleavage of PARP was upregulated by gemcitabine (Fig. 4D).

We established a gemcitabine-resistant PC-3 cell line (PC-3-R) by culturing PC-3 cells with a low level of gemcitabine for 3 months. Notably, the PC-3-R cell line expressed less HMGN5 protein and mRNA compared with its parental cells (Fig. 5). Taken together, these results indicate that HMGN5 may be a potential biomarker for the treatment of prostate cancer patients with gemcitabine.