The present study and all related procedures were approved (approval no. 2021-1142) by the Ethics Committee of Zhejiang University School of Medicine Second Affiliated Hospital (Hangzhou, China). Prior to conducting the study, informed consent forms were obtained from all patients with PCa participating in the study.

JASPARA database (https://jaspar.elixir.no/) was used to predicted sequences of transcription factor AR binding to the promoter of SENPs genes.

The survival analysis of SENPs was obtained from the GEPIA 2.0 database (http://gepia.cancer-pku.cn/), including overall survival (OS) and disease-free survival (DFS) in patients with cancer. Patients were divided into two groups according to the tumor types and the levels of SENPs from The Cancer Genome Atlas (TCGA) database (https://www.cancer.gov/ccg/research/genome-sequencing/tcga), either lower or higher than the mean level. P<0.05 was considered to indicate a significantly significant difference. The 95% confidence interval (CI) is presented as dotted line, hazards ratio (HR).

LNCaP cells (cat. no. CRL-1740) were purchased from the American Type Culture Collection, cells were cultured in RPMI-1640 media (cat. no. PYG0006; Wuhan Boster Biological Technology, Ltd.) supplemented with 10% fetal bovine serum (FBS) in 37°C and 5% CO₂ atmosphere. To mimic patients undergoing ADT treatment, androgen-independent cell lines, had already been generated: LNCaP-AI cells derived from androgen-sensitive LNCaP cells cultured under androgen-depleted conditions (7). LNCaP-AI cells were cultured in RPMI-1640 medium with 10% dextran-charcoal stripped FBS (CS-FBS; Gibco; Thermo Fisher Scientific, Inc.).

Enzalutamide (cat. no. MDV3100) was purchased from Selleck Chemicals and dihydrotestosterone (DHT; cat. no. A8380) from MilliporeSigma. Antibody against GAPDH (cat. no. 5174) was obtained from Cell Signaling Technology, Inc. Anti-human SENP5 (cat no. 19529-1-AP) antibody was purchased from Proteintech Group, Inc. Secondary antibodies [HRP-conjugated sheep anti-rabbit antibodies (cat. no. AP510P) or HRP-conjugated sheep anti-mouse antibodies (cat. no. AC111P)] for western blotting were obtained from MilliporeSigma. DyLight 488 AffiniPure Goat Anti-Rabbit IgG (H+L) for immunofluorescence were obtained from Abbkine Scientific Co., Ltd.

Immunofluorescence staining with SENP5 was performed as previously described (8).

LNCaP and LNCaP-AI cells were seeded in six-well plates (5×10⁵/well) and transfected with the siRNAs (100 nM) using Lipofectamine 2000^® according to the manufacturer's instructions (Thermo Fisher Scientific, Inc.). The siRNAs used in the study were as follows: siSENP5 #1: 5′-CCAACACTTGTGCATTCTGAA-3′; siSENP5 #2: 5′-CCTTACCAGAACATCGTTCTA-3′; siAR: 5′-GGAACTCGATCGTATCATTGC-3′; negative control: 5′-UUCUCCGAACGUGUCACGUTT-3′. Cell transfection was performed for 6 h at 37°C, after which the medium was replaced. After 48 h, the cells were collected for subsequent experiments.

Total protein of cells was extracted with RIPA buffer (Beyotime Institute of Biotechnology), and western blot analysis was performed as previously described (9).

LNCaP and LNCaP-AI cells were transfected with siScr or siSENP5 for 48 h. Cell proliferation assay was performed as previously described (9).

RNA isolation and qPCR were performed as previously described (8). Primers used were as follows: SENP5 forward, 5′-CTTTAGGTCAGGCCAATGGTC-3′ and reverse, 5′-CAGCAGCCGTAACAAAAGCC-3; and GAPDH forward, 5′-AACAGCCTCAAGATCATCAGCA-3′ and reverse, 5′-CATGAGTCCTTCCACGATACCA-3′.

ChIP-qPCR was performed as previously described (10). Cells (5×10⁷) were collected, followed by ChIP assays with anti-AR (Cell Signaling Technology, Inc.; cat. no. 5153) or IgG (Proteintech Group, Inc.; cat no. 30000-0-AP). qPCR analysis was carried out on ChIPed and input DNA. Data are presented as percent of input DNA, and error bars represent S.D. for technical duplicates of the qPCR analysis. Primer sequences to detect the AR binding site along the SENP5 promoter were as follows: SENP5-ChIP-F1: TGGGGCGGGTAAGACATAGA; SENP5-ChIP-R1: CCATTCCAGCTCTGGACGTT.

Data were analyzed with GraphPad Prism 8 (Dotmatics). The values were shown as the mean ± S.D. for triplicate experiments and the statistical differences were calculated by one-way ANOVA analysis of variance with Dunnett's test or Newman-Keuls test and Student's two-tailed t-test. P<0.05 was considered to indicate a statistically significant difference.

In order to clarify the expression and potential clinical values of SENPs in PCa, firstly, the mRNA expression level of SENPs were analyzed in PCa and normal tissues using the TCGA database. As revealed in Fig. 1A, the expression level of SENPs were almost no difference between PCa and normal tissues. However, interestingly, significant co-expression between SENPs and AR was found (Fig. 1B), except for SENP3, strongly indicating a potential regulatory relationship between SENPs and AR. It has been previously reported that AR can transcriptionally regulate SENP1 and SENP2 (6). Therefore, it was hypothesized that AR may also be able to transcriptionally regulate other genes in the SENP family. Using the public online tool JASPARA, an interaction was found between the promoter regions of SENPs genes and the transcription factor AR (Table SI, Table SII, Table SIII, Table SIV, Table SV, Table SVI, Table SVII). These data indicated that SENPs are potential downstream target genes of transcription factor AR.

Despite there was almost no difference in the expression level of SENPs between PCa and normal tissues, it was attempted to further clarify whether the expression of SENPs is related to the prognosis of PCa. Using the TCGA database, it was found that the mRNA expression of SENP5 was significantly negatively associated with OS (Fig. 2A) and DFS (Fig. 2B) in PCa, while other genes in the SENP family were not markedly associated with OS and DFS (Fig. S1), indicating that high expression of SENP5 is associated with poor prognosis in PCa. Furthermore, it was found that the protein level of SENP5 was markedly higher in PCa tissues compared with normal adjacent tissues through immunofluorescence analysis (Fig. 2C and D). In addition, SENP5 silencing significantly inhibited the proliferation activity of LNCaP and LNCaP AI cells (Fig. 2E and F). These results suggested that SENP5 may act as an oncogene in PCa.

To further illuminate the regulatory mechanism between AR and SENP5, PCa cell lines (LNCaP and LNCaP-AI) were treated with MDV3100 and DHT respectively. DHT, the most potent androgen, is usually synthesized in the prostate from testosterone secreted by the testis, which acts as the activating ligand of AR (11). Furthermore, MDV3100 is an antagonist of AR that is widely used in the treatment of patients with PCa (12). Compared with the control group, qPCR results showed that the mRNA level of SENP5 was significantly reduced after MDV3100 treatment (Fig. 2G). On the contrary, the mRNA level of SENP5 significantly increased after DHT treatment (Fig. 2G). Consistent with the mRNA results, similar results were also obtained at the protein level (Fig. 2H). In addition, it was further confirmed that AR could transcriptionally activate SENP5 expression using CHIP-qPCR (Fig. 2I). These results further confirmed that SENP5 is indeed a direct downstream target gene of AR.