The genes in 499 PCa and 52 adjacent healthy tissues from the TCGA database were analyzed using the edgeR software package (http://www.bioconductor.org/packages/release/bioc/ html/edgeR.html). Next, 156 genes identified to be significantly associated with the survival of the patients were screened. The selected 156 genes were analyzed using the DAVID available online database (https://david.ncifcrf.gov/). The functions of the top seven core genes on the cell were investigated using Gene Ontology (GO) analysis (http://www.geneontology.org/). ASF1B was selected for subsequent experiments.

Between June 2015 and September 2017, 37 samples of PCa tissues and healthy adjacent tissues (mean age, 65.5±7 years) were obtained from patients with PCa who were admitted to The First Affiliated Hospital of Xinxiang Medical University (Xinxiang, China). Patients were diagnosed with PCa by biopsy or pathology. Patients who also had other malignant tumors, coronary heart disease or diabetes were excluded. All patients signed informed consent for the use of their tissues in the present study. This research was approved by the Ethics Committee of The First Affiliated Hospital of Xinxiang Medical University.

Human prostatic hyperplasia epithelial cell line (BPH) and PCa cell lines (PC-3, DU145, LNCap, VcaP and C4-2) were purchased from Beijing Zeping Technology Co., Ltd. (Beijing, China). PC-3 is derived from a bone metastatic site, and LNCap is derived from a left supraclavicular lymph node metastatic site. As a subline of LNCap, C4-2 exhibits androgen- independent growth in association with skeletal metastasis. DU145 is derived from a brain metastatic site and VcaP is derived from a vertebra metastatic site. BPH cells were cultured in Dulbecco's modified Eagle's medium (DMEM; Beijing Solarbio Science & Technology Co., Ltd., Beijing, China) with 10% fetal bovine serum (FBS; Thermo Fisher Scientific, Inc., Waltham, MA, USA) and penicillin-streptomycin mixed solution. PC-3, DU145, LNCap, VcaP and C4-2 cells were cultured in RPMI-1640 (Beijing Solarbio Science & Technology Co., Ltd., Beijing, China) with 10% FBS and 100X penicillin-streptomycin mixed solution. All cells were maintained in a humidified incubator with 5% CO₂ at 37°C.

The human anti-silencing function 1 (ASF1)B-target small interfering (si)RNA and unspecific scrambled siRNA vectors were designed and synthesized by Nanjing Kehao Biotechnology Co., Ltd. (Nanjing, China). LNCap and C4-2 cells were seeded at a density of 1×10⁴ cells/well into 6-well plates. The cells were transfected with siRNA-ASF1B (1 µg) or siRNA-negative control vector (1 µg) using Hieff Trans™ Liposomal Transfection reagent (Shanghai Yusheng Biotechnology Co., Ltd., Shanghai, China) following the manufacturer's protocol. The cells were incubated at 37°C for 12 h. Next, the cells were collected and prepared for the subsequent experiments.

Tissues and cells of total RNA were extracted using RNA extraction kit (Promega Corporation, Madison, WI, USA). A total of 1 µg of RNA was transcribed to cDNA using TianScript cDNA Synthesis kit (Tiangen Biotech Co., Ltd., Beijing, China). Reaction conditions were as follows: 85°C for 5 min, and at 4°C for 5 min. SYBR^® Premix Ex Taq™ II kit (Takara, Beijing, China) was used to amplify cDNA under reaction conditions as follows: Predegeneration at 92°C for 5 min; denaturation at 92°C for 15 sec; annealing at 58°C for 30 sec for 30 cycles; and extension at 72°C for 35 sec. GAPDH was used as the loading control. The 2^−ΔΔCq method was applied to compare gene expression levels. The primers used are listed in Table I.

The proteins of tissues and cells were collected using RIPA lysate buffer (Beijing Solarbio Science & Technology Co., Ltd.). Protein concentration was detected using a bicinchoninic acid assay. Proteins (25 µg/lane) were separated using 8% SDS-PAGE and were then transferred onto a polyvinylidene fluoride membrane. The membrane was blocked with 5% skimmed milk at room temperature 1.5 h. Next, the membrane was incubated with the primary antibodies (Table II) at 4°C for 24 h. The membrane was then bound to the donkey anti-mouse IgG (cat. no. ab150109; 1:6,000), goat anti-mouse IgG (cat. no. ab6785; 1:6,000) (both from Abcam, Cambridge, UK) and donkey anti-rabbit IgG (cat. no. NL004; 1:5,000; R&D Systems, Inc., Minneapolis, MN, USA) secondary antibodies at 37°C for 1 h. The signal was visualized using enhanced chemiluminescence (Sangon Biotech Co., Ltd., Shanghai, China). Quantity One 4.6.2 software (Bio-Rad Laboratories, Inc., Hercules, CA, USA) was used to perform densitometry.

Cell viability was examined using a CCK-8 kit (Beijing Solarbio Science & Technology Co., Ltd.) following the manufacturer's protocol. In brief, LNCap and C4-2 cells were plated in 96-well plates (1.5×10³ cell/well) in an incubator for 24 h. Cells were exposed to PBS (control), unspecific scrambled siRNA plasmid (mock) and ASF1B-target siRNA (siRNA-ASF1B) for 12, 24 and 48 h. Next, CCK-8 solution was added to the cells and incubated for 4 h. Absorbance at 450 nm was recorded and evaluated using a microplate reader.

Cell apoptosis was examined using an Annexin V-fluorescein isothiocyanate (FITC)/propidium iodide (PI) apoptosis detection kit (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany). LNCap and C4-2 cells were inoculated in 6-well plates (3×10⁴ cells/well) in the incubator for 24 h. All cells were treated as aforementioned. Next, cells were digested by 0.25% trypsin for 2 min at room temperature. The trypsin was discarded and cells were re-suspended in Annexin V binding buffer. The Annexin V-FITC and PI were added to cells and incubated for 20 min in the dark at room temperature. BD FACSAria I/II flow cytometry was performed to measure cell apoptosis and analysis was performed using BD CellQuest Pro 3.3 software (both from BD Biosciences, San Jose, CA, USA).

Briefly, LNCap and C4-2 cells were treated as aforementioned. The PI and RNAse mix was added to the cells in the dark at 37°C for 20 min. Cell cycle was analyzed using flow cytometry.

The colony formation of cells was determined using crystal violet staining (Beijing Solarbio Science & Technology Co., Ltd.). Briefly, LNCap and C4-2 cells were inoculated in 6-well plates (3×10⁴ cell/well) and incubated for 24 h. All cells were treated as aforementioned. Cells were then fixed with 100% methanol for 10 min at room temperature. Next, cells were stained by 0.1% crystal violet for 5 min at room temperature. Finally, the cell colonies were observed under an inverted microscope (IX73; Olympus Corporation, Tokyo, Japan; magnification, ×2.5). The number of the colonies containing >50 cells was counted to calculate the colony formation as follows: (Number of colonies/number of cells) ×100%.

IBM SPSS Statistics 20.0 software (IBM, Corp., Armonk, NY, USA) was used for statistical analysis. Data are presented as the mean ± standard deviation. One-way analysis of variance followed by Tukey's text was performed to compare the differences among groups. The differences between tumor and healthy adjacent tissues were calculated using a paired Student's t-test. Kaplan Meier survival analysis with a log-rank test was performed to compare the overall survival rate of patients grouped by high and low ASF1B expression. The association between the expression of ASF1B and the clinicopathological features of patients with PCa were assessed using the Chi-square test. P<0.05 was considered to indicate a statistically significant difference. The experiments were independently repeated at least three times.

To identify the core gene, data from 52 healthy adjacent tissues and 499 PCa tissues were collected from the TCGA database. A total of 2,971 genes were investigated (Fig. 1A) and 156 genes were identified to be significantly associated with the survival rate of patients with PCa (Fig. 1B). A high expression of ASF1B was associated with the poor prognosis of patients with PCa (Fig. 1C). In addition, GO analysis demonstrated that seven core genes in PCa were significantly associated with cell cycle distribution and cell differentiation (Fig. 1D), the ASF1B gene was selected as the core gene in the present study. To determine the role of the expression of ASF1B in PCa, the pathological features of patients with PCa were analyzed. The data demonstrated the high ASF1B expression was significantly associated with tumor node metastasis (TNM) stage and metastasis; however, no significant association was identified between ASF1B expression and other pathological features (Table III; P>0.05).

RT-qPCR and western blotting were performed in order to examine the expression of ASF1B in the PCa and healthy adjacent tissues. It was demonstrated that the mRNA expression of ASF1B in cancer tissues were significantly higher compared with that in healthy adjacent tissues (Fig. 2A; P<0.01). A total of six patients from the cohort were randomly selected to measure the protein expression of ASF1B. The results revealed that there was at least a 2-fold increase in the protein expression of ASF1B in cancer tissues compared with that in the healthy adjacent tissues (Fig. 2B; P<0.01). Furthermore, the mRNA and protein levels of ASF1B in normal cells (BPH) were significantly lower compared with those in the PCa cells (PC-3, DU145, LNCap, VcaP and C4-2) (Fig. 2C and D; P<0.01). The expression levels of ASF1B in LNCap and C4-2 cells were the highest compared with those in other cell lines. Hence, LNCap and C4-2 cells were selected for subsequent studies.

The transfection efficiencies of the siRNA-ASF1B plasmid in LNCap and C4-2 cells were examined by RT-qPCR and western blotting. The results of the present study observed a significant decrease in ASF1B protein and mRNA expression (Fig. 3A and B; P<0.05) following the transfection of LNCap with the siRNA-ASF1B plasmid compared with the mock control group. Furthermore, the ASF1B protein and mRNA expression levels were reduced by siRNA-ASF1B transfection in C4-2 cells (Fig. 3C and D; P<0.05).

The viabilities of LNCap and C4-2 cells were examined using a CCK-8 assay. The CCK-8 data demonstrated that the viabilities significantly decreased in siRNA-ASF1B-transfected LNCap and C4-2 cells compared with the mock (Fig. 3E and F; P<0.01). Following transfection for 12 h, cell viability was significantly decreased with si-ASF1B. Thus, the si-ASF1B transfection duration was set as 12 h.

The data in the present study revealed that compared with the mock control, siRNA-ASF1B significantly enhanced the apoptotic and promoted the cell cycle arrest of LNCap cells (Fig. 4A and B; P<0.01). Additionally, when C4-2 cells were treated with siRNA-ASF1B, the apoptosis and the number of cells in G1 phase were significantly increased, whereas the number of cells in G2 and S phases was decreased significantly (Fig. 4C and D; P<0.01).

In order to identify the effect of siASF1B on the proliferation of LNCap and C4-2 cells, colony formation analysis was performed. The results revealed that the colony formation of LNCap and C4-2 cells was significantly attenuated by siRNA- ASF1B transfection compared with the mock group (Fig. 5; P<0.01).

The results demonstrated that the protein and mRNA levels of cyclin D1 and Bcl-2 were downregulated; however, the protein and mRNA levels of p53, caspase-3 and Bax were upregulated in siRNA-ASF1B-transfected LNCap cells, compared with the mock group (Fig. 6A and B; P<0.01). In addition, the protein and mRNA expression levels of these factors in C4-2 cells were in accordance with those observed in LNCap cells (Fig. 6C and D; P<0.01).

In order to further explore the possible mechanism of siASF1B in LNCap and in C4-2 cells, western blotting was performed to detect the protein expression levels of p-PI3K, PI3K, p-Akt and Akt. Western blotting data demonstrated that siRNA-ASF1B significantly reduced the phosphorylation levels of PI3K and Akt in LNCap and C4-2 cells by at least a third compared with the mock control group. However, the levels of PI3K and Akt remained stable among other groups. The proportions of p-PI3K/PI3K and p-Akt/Akt in siRNA-ASF1B were significantly reduced compared with those in the mock group (Fig. 7; P<0.01).