The effect of ubiquitin-specific peptidase 21 on proliferation, migration, and invasion in DU145 cells

Bioinformatic data

RNA-seq data from The Cancer Genome Atlas-Prostate Adenocarcinoma (TCGA-PRAD), which included 51 normal samples and 481 tumor samples, were downloaded from the Genomic Data Commons (https://portal.gdc.cancer.gov/). Survival and clinical data were downloaded from the University of California Santa Cruz Xena (USCS Xena) (http://xena.ucsc.edu/). Complete clinical and sequencing survival data were available for all 481 patients. The Gene Expression Omnibus (GEO) data were downloaded from the National Center for Biotechnology Information (https://www.ncbi.nlm.nih.gov/geo/). Sample GSE32571 contained 59 prostate cancer samples and 39 matched benign tissue samples, whereas GSE70768 contained 113 prostate tumor samples and 73 matched benign tissue samples. The USP21 RNA levels of PRAD tissues and adjacent normal tissues from TCGA or GEO (GSE70768 and GSE32571), as well as those of different stages, were analyzed using edgeR from R (version 3.30.3).

Kaplan-Meier survival curve analysis

Patients were sorted into TCGA-PRAD groups with USP21 high-expression or USP21 low-expression. The surv CUTpoint function from the survMiner package was used to obtain the best segmentation values. Kaplan–Meier survival curves for overall survival (OS), progression-free interval (PFI), disease-free interval (DFI), and disease-specific survival (DSS) were plotted using the R survminer package (version 0.4.8).

Gene set enrichment analysis

The patients from the TCGA-PRAD group were divided into high-and low-expression groups according to the median values of USP21. Gene set enrichment analysis (GSEA) software (version 4.2.1) was used to study cancer-related pathways associated with USP21 levels in PRAD cells. Pearson’s correlation analysis was used to rank genes. The six genes with the highest correlation were selected for each pathway, and R (version 4.0.5) and ggplot2 (version 3.3.5) software were used to compare the mRNA levels of these genes under high- or low- USP21 expression levels.

Immune infiltration analysis

Tumor Immune Estimation Resource (TIMER) (https://cistrome.shinyapps.io/timer/) was used to analyze the effect of USP21 expression level on the immune infiltration of prostate cancer cells and the effect of USP21 copy level on the infiltration of other immune cells. The relationship between USP21 expression and methylation or copy number was analyzed using Gene Set Cancer Analysis (http://bioinfo.life.hust.edu.cn/GSCA/#/).

Cell culture

The human PRAD DU145 cell line was purchased from IMMOCELL (Xiamen, China) and was maintained in Minimum Essential Medium (Thermo Scientific, Logan, UT) containing 10 % (v/v) fetal bovine serum, 100 g/mL streptomycin, and 100 U/mL penicillin in an incubator at 37 °C with 5 % CO₂.

Construction of plasmids

Plasmids expressing short hairpin (sh) USP21 (shUSP21-1 and shUSP21-2, Gene ID:27,005) or a scrambled control (shNC), and USP21 expression plasmid (USP21 OE), were created using the pLV-sh-puro and pcDNA3.3 N-3HA vectors, which were purchased from XIAMEN Anti-HeLa Biological Technology Trade Co. Ltd (Xiamen, China). Appropriate primer sequences were developed according to the specifications of the vectors and are detailed in Table 1.

Quantitative PCR

Using an RNA isolation kit (Omega) and a reverse transcription kit (Vazyme), RNA was extracted from the cells and reverse transcribed in accordance with the manufacturer’s recommendations. The Real-Time PCR System (Bio-Rad Laboratories, Inc.) and SYBR qPCR kit (Vazyme) were used to determine the mRNA expression levels of the target genes (Vazyme). The quantitative PCR (qPCR) primers are listed in Table 2.

Western blotting

Proteins were extracted from the cells using a kit (Beyotime), then lysed in ice-cold RIPA buffer (Beyotime). Finally, SDS-PAGE (10 %) was used to separate the proteins per lane. Proteins were transferred onto polyvinylidene difluoride membranes after separation. After blocking with 5 % skim milk, the membranes were incubated with USP21 (cat. no. 17856-1-AP, 1:1,000; Proteintech) and beta actin antibodies (cat. no. 20536-1-AP, 1:1,000; Proteintech) for 2 h at room temperature. This was followed by incubation with anti-rabbit immunoglobulin (cat. no. SA00001-2; 1:10,000; Proteintech) conjugated with horseradish peroxidase for 1 h at room temperature. A chemiluminescence kit (Thermo Fisher Scientific) was used to visualize the membranes. We used ImageJ v1.48 (National Institutes of Health) for densitometry.

Cell proliferation

Ten thousand DU145 cells per well were planted in 96-well plates, and were transfected with 0.2 μg of shUSP21, shNC, vector, or USP21 OE plasmid.

At 24, 48, and 72 h after transfection, cells were treated with methylthiazolyldiphenyl-tetrazolium bromide (MTT) (Beyotime Biotechnology, Shanghai, China) for 4 h. Cells were agitated for 10 min with 100 µL of dimethylsulfoxide (DMSO) (added after the supernatant had been discarded), in order to dissolve the crystals. An absorbance reader (SpectraMax, San Francisco, CA, USA) was used to measure cell viability at 490 nm.

For EdU assay, 48 h after transfection, MEM with 50 µM EdU from Guangzhou RiboBio Co., Ltd. was incubated with the cells for 4 h at 37 °C. Next, the cells were fixed with 4 % formaldehyde for 20 min and incubated with 2 mg/mL glycine for 5 min at the same temperature. The cells were treated with EdU (Guangzhou RiboBio Co., Ltd.) for 20 min at 28 °C after being exposed to 0.5 % Triton X-100 for 10 min. The nuclear DNA was stained for 10 min at room temperature with DAPI (5 μg/mL). Fluorescence imaging was performed using a microscope (Motic).

Colony formation assay

After the DU145 cells in 6-well plates were transfected with 5 μg of shUSP21, shNC, vector, or USP21 OE plasmid for 24 h, they were uniformly dispersed into 2.0 × 10³ cells per well. After 2 weeks, the cells underwent a 15-min methanol fixation process. They were then stained with 0.5 % crystal violet (Yeasen, Catalog Number: 60505ES25) in phosphate-buffered saline (PBS) for 15 min. Finally, the colonies were imaged and the numbers of colonies were counted using Image J 1.52v (NIH, Bethesda, MD, USA).

Cell cycle assay

Three hundred thousand cells per well were grown in 6-well plates and were transfected with 5 μg of shUSP21, shNC, vector, or USP21 OE plasmid for 48 h. The cells were then collected and fixed in 70 % ethanol. The fixed cells were washed using PBS before being incubated at 37 °C for 30 min in 0.2 % Triton X-100 and 10 μg/mL RNase. After being treated with 7-AAD (Beyotime) for 30 min in the dark, the cells were examined using the NovoCyte setup and NovoExpress^® software 1.4.1.

Apoptosis assay

A total of 1 × 10⁶ DU145 cells per well were seeded in 6-well plates and transfected with 5 μg of shUSP21, shNC, vector, or USP21 OE plasmid. After 48 h, the cells were centrifuged at 200 g for 5 min. After washed with PBS, the cells were stained using an Annexin V-FITC/PI kit, in accordance with the manufacturer’s instructions. The data were examined using the NovoCyte and NovoExpress^® software 1.4.1.

Cell migration and invasion assay

From a 6-well plate, DU145 cells were transfected with 5 μg of shUSP21, shNC, vector, or USP21 OE plasmid for 24 h. Then 5 × 10⁴ DU145 cells in 100 μL Dulbecco’s Modified Eagle Medium (MEM) (without Fetal Bovine Serum [FBS]) were implanted into the upper chambers of Transwell plates (Corning, Catalog number: #3422, 8 μm aperture). In the lower chambers of the Transwell plates, 500 μL of MEM containing 10 % FBS were plated. Migrated cells were stained with 0.5 % crystal violet after 24 h of incubation. In three random fields, Image J 1.52v (NIH, Bethesda, MD, USA) was used to count the number of migrating cells. For the invasion assay, the top chambers’ membranes were pre-coated with Matrigel that had been 8 times diluted.

Statistical analysis

The R Survival package was subjected to a log-rank test. Mann-Whitney U and Wilcoxon tests were performed for non-parametric data. Student’s t-test was used for parametric data to identify significant differences between the two groups. In GraphPad Prism 8.2.1, Tukey’s test and one-way ANOVA were performed to determine which differences across groups were significant. The threshold for statistical significance was set at p<0.05. Data are expressed as mean ± standard deviation.

USP21 mRNA level is upregulated in PARD

USP21 mRNA expression was considerably higher in patients with PRAD and at stages T3 and T4 than in those without and at stages T1 and T2 (Figure 1A). In addition, there was no significant difference in expression levels of USP21 mRNA in the residual tumors (Figure 1A). These findings suggest that USP21 contributes to the development of PRAD. Analysis of high-throughput RNA sequencing data from TCGA-PRAD and GEO showed that the expression level of USP21 mRNA was considerably higher in tumor tissues than in non-tumor tissues (Figure 1B). The OS, PFI, and DFI of patients with PRAD with high USP21 levels were worse than those of patients with low USP21 levels; however, there was no significant difference in DSS (Figure 1C). Thus, these individuals have poor prognosis when USP21 mRNA expression is upregulated in PRAD tissues. Therefore, USP21 mRNA, which is highly expressed in tissues of patients with PRAD, may be closely linked to the development of PRAD.

Figure 1:

USP21 RNA is upregulated in PRAD tissues. (A) Association between USP21 expression and disease status, residual tumor, and clinical T stage in PRAD tissues from the TCGA-PRAD group. (B) PRAD tissues of patients from the TCGA-PRAD group (far left), GSE70768 (middle), and GSE32571 (rightmost) groups were compared with adjacent normal tissue. (C) In patients with PRAD from the TCGA-PRAD group, the relationship between USP21 mRNA expression and OS, PFI, DSS, or DFI was assessed using Kaplan-Meier survival analysis. NT: non-tumor tissues, PRAD: prostate adenocarcinoma, OS: overall survival, PFI: progression-free interval, DSS: disease-specific survival, DFI: disease-free interval, ns: not significant, *p<0.05, **p<0.01, ***p<0.001.

Silencing USP21 suppresses cell proliferation

To further explore the function of USP21 in PRAD, we knocked down USP21 in DU145 cells. The results showed that shUSP21-1 and shUSP21-2 successfully reduced the levels of USP21 in DU145 cells (Figure 2A and B). The proliferation of DU145 cells after USP21 knockout was significantly lower than in those without USP21 knockout, which indicated that USP21 downregulation inhibited cell proliferation (Figure 2C). Additionally, USP21-knockdown cells had a considerably lower EdU⁺ rate than control cells, which suggested that USP21 deletion reduced the proportion of EdU⁺ cells (Figure 2D). Moreover, USP21 silencing suppressed colony formation (Figure 2E). Cell cycle assay demonstrated that USP21-knockdown enhanced the proportion of cells in the G₀/G₁ phase (Figure 2F).

Figure 2:

Low expression levels of USP21 suppresses proliferation. (A, B) the DU145 cells in the 6-well plate were transfected with 5 μg of shUSP21 or shNC plasmid for 48 h, and the expression of USP21 was analyzed by qPCR (A) and western blot (B), n=3. (C,D) after DU145 cells in the 96-well plate were transfected with 0.2 μg of shUSP21 or shNC plasmid for 24, 48, and 72 h, or, for 48 h, the cells were analyzed using MTT (C) and EdU (D) assay. (E) After DU145 cells in the 6-well plate were transfected with 5 μg of shUSP21 or shNC plasmid for 24 h, they were analyzed via colony formation experiments. (F) After DU145 cells in the 6-well plate were transfected with 5 μg of shUSP21 or shNC plasmid for 48 h, they were analyzed using 7-AAD. Sh: short hairpin, NC: Negative control, OD: optical density, ns: not significant, *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001.

Decreased USP21 levels of DU145 cells resulted in increase of apoptosis and reduced migration and invasion abilities

After USP21-knockdown, there was a dramatic increase in the proportion of apoptotic cells according to an Annexin V/PI staining experiment performed using flow cytometry (Figure 3A). These observations indicated that a decrease in USP21 expression upregulates apoptosis. Moreover, the number of migratory and invading DU145 cells was considerably reduced after the downregulation of USP21 (Figure 3B). This phenomenon suggested that knockdown of USP21 led to a decrease in the migration ability, colonization of new spaces, and invasiveness of DU145 cells.

Figure 3:

The capacity of DU145 cells to migrate and invade was reduced when USP21 was downregulated because apoptosis increased. (A) After DU145 cells in the 6-well plate were transfected with 5 μg of shUSP21 or shNC plasmid for 48 h, they were analyzed using annexin V-FITC and propidium iodide (PI) staining. (B) After DU145 cells in the 6-well plate were transfected with 5 μg of shUSP21 or shNC plasmid for 24 h, they were analyzed using Transwell experiments with or without Matrigel to demonstrate how USP21 silencing affects the capacity of DU145 cells to migrate (upper) or invade (lower). **p<0.01, ***p<0.001.

Overexpressing USP21 enhances proliferation

As shown in Figure 4A and B, the USP21 expression plasmid enhanced the expression of USP21 in cells. Proliferation, colony formation, and cell cycle alterations of the tested cells were subjected to MTT and EdU assays, crystal violet staining, and PI staining, respectively. The formation of cell colonies was accelerated and cell proliferation improved after USP21 (Figure 4C–E). In addition, the G₀/G₁ phase of the cell cycle was halted by overexpression of USP21 (Figure 4F). Therefore, DU145 cell proliferation was more rapid when USP21 was expressed at higher levels.

Figure 4:

Upregulation of USP21 enhances proliferation. (A and B) the DU145 cells in the 6-well plate were transfected with 5 μg of vector or USP21 OE plasmid for 48 h, and the expression of USP21 was analyzed by qPCR (A) and western blot (B). n=3. (C and D) After DU145 cells in the 96-well plate were transfected with 0.2 μg of vector or USP21 OE plasmid for 24, 48, and 72 h, or for 48 h, the cells were analyzed using MTT (C) and EdU (D) assay. (E) After DU145 cells in the 6-well plate were transfected with 5 μg of vector or USP21 OE plasmid for 24 h, the cells were analyzed via colony formation experiment. (F) After DU145 cells in the 6-well plate were transfected with 5 μg of vector or USP21 OE plasmid for 48 h, the cells were analyzed using 7-AAD. *p<0.05, ***p<0.001, and ****p<0.0001.

Increasing USP21 augments migration and invasion ability of DU145 cells

Next, annexin V/PI labeling in conjunction with flow cytometry was employed to identify apoptosis in DU145 cells after USP21 overexpression. There had no discernible effect on the apoptosis of DU145 cells (Figure 5A) after overexpression of USP2. Laboratory Transwell assays confirmed that USP21 overexpression in DU145 cells increased their migration and invasion abilities (Figure 5B). Thus, USP21 overexpression enhanced DU145 cell migration and invasiveness.

Figure 5:

Upregulating USP21 causes DU145 cells to invade and migrate. (A) After DU145 cells in the 6-well plate were transfected with 5 μg of vector or USP21 OE plasmid for 48 h, the cells were analyzed using annexin V-FITC and PI staining. (B) After DU145 cells in the 6-well plate were transfected with 5 μg of vector or USP21 OE plasmid for 24 h, the cells were analyzed using Transwell experiment without or with Matrigel. ns: not significant, *p<0.05, **: p<0.01.

USP21 in PRAD was of the capacity to control cell cycle and adhesion molecules

GSEA was performed to examine the impact of USP21 mRNA expression on cell adhesion molecules and cell cycle pathways in patients with PRAD. As expected, USP21 expression was significantly correlated with the cell cycle and cell adhesion molecules (Figure 6A and B). Cell adhesion molecules were mostly negatively correlated with USP21, whereas most cell cycle-related molecules were positively correlated with USP21 (Figure 6A and B). SELP, CNTN1, ICAM1, and NEGR1 showed considerably higher mRNA expression after USP21 knockdown (Figure 6C and D). The mRNA levels of these genes decreased following USP21 overexpression (Figure 6C and D). Furthermore, silencing USP21 led to a decrease in the mRNA expression levels of MCM7, ORC6, PCNA, and ANAPC2, whereas USP21 upregulation promoted an increase in the mRNA levels of these genes (Figure 6C and D).

Figure 6:

USP21 in PRAD has the potential to regulate cell adhesion molecules and the cell cycle. (A, B) GSEA that USP21 may regulate cell adhesion molecules (A) and the cell cycle (B). (C, D) The DU145 cells in the 6-well plate were transfected with 5 μg of shNC, shUSP21, vector, or USP21 OE plasmid for 48 h. The mRNA levels of ANAPC2, CNTN1, ICAM1, MCM7, NEGR1, ORO6, PCNA, and SELP were analyzed by qPCR. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001.

Correlations between immune infiltration in PRAD and USP21 expression

Immune cells in the tumor microenvironment have been demonstrated to affect tumor development [13]. The results of the present study indicate that USP21 may promote the development of PRAD. Therefore, additional tests were conducted to determine whether USP21 expression was associated with immune infiltration. The data showed a correlation between USP21 expression and B cell, CD8+ T cell, CD4+ T cell, macrophage, neutrophil, and dendritic cell infiltration (Figure 7A). Subsequently, we investigated whether there was any connection between invading immune cells and USP21 copy number. Deletion of USP21 significantly increased the infiltration levels of neutrophils (Figure 7B), which suggested that USP21 may be able to encourage immunological infiltration in patients with PRAD. Moreover, USP21 copy number positively correlated with USP21 expression, whereas USP21 methylation negatively correlated with USP21 expression (Figure 7C).

Figure 7:

Immune infiltration in patients with PRAD is correlated with USP21 expression. (A) Correlation between USP21 expression level and immune cells that infiltrate PRAD as analyzed by TIMER. (B) Correlation between USP21 copy quantity and the infiltration level of immune cells in PRAD as analyzed using TIMER. (C) GSCA showing correlation between USP21 level and methylation or copy number variation.