ACSS3 represses prostate cancer progression through downregulating lipid droplet-associated protein PLIN3

Abstract

Current endocrine therapy for prostate cancer (PCa) mainly inhibits androgen/androgen receptor (AR) signaling. However, due to increased intratumoural androgen synthesis and AR variation, PCa progresses to castration-resistant prostate cancer (CRPC), which ultimately becomes resistant to endocrine therapy. A search for new therapeutic perspectives is urgently needed. Methods: By screening lipid metabolism-related gene sets and bioinformatics analysis in prostate cancer database, we identified the key lipid metabolism-related genes in PCa. Bisulfite genomic Sequence Polymerase Chain Reaction (PCR) (BSP) and Methylation-Specific Polymerase Chain Reaction (PCR) (MSP) were preformed to detect the promoter methylation of ACSS3. Gene expression was analyzed by qRT-PCR, Western blotting, IHC and co-IP. The function of ACSS3 in PCa was measured by CCK-8, Transwell assays. LC/MS, Oil Red O assays and TG and cholesterol measurement assays were to detect the levels of TG and cholesterol in cells. Resistance to Enzalutamide in C4-2 ENZR cells was examined in a xenograft tumorigenesis model in vivo. Results: We found that acyl-CoA synthetase short chain family member 3 (ACSS3) was downregulated and predicted a poor prognosis in PCa. Loss of ACSS3 expression was due to gene promoter methylation. Restoration of ACSS3 expression in PCa cells significantly reduced LD deposits, thus promoting apoptosis by increasing endoplasmic reticulum (ER) stress, and decreasing de novo intratumoral androgen synthesis, inhibiting CRPC progression and reversing Enzalutamide resistance. Mechanistic investigations demonstrated that ACSS3 reduced LD deposits by regulating the stability of the LD coat protein perilipin 3 (PLIN3). Conclusions: Our study demonstrated that ACSS3 represses prostate cancer progression through downregulating lipid droplet-associated protein PLIN3.

Keywords: ACSS3; PLIN3; lipid droplet; prostate cancer.

Figure 1

ACSS3 was downregulated and predicted poor prognosis in PCa. (A) By screening lipid metabolism‐related gene sets in five independent PCa databases, four genes were selected; three of these genes (ACSS3, ACSF2 and CLU) were found to be downregulated and one gene (FABP5) was found to be upregulated (│Fold change│>1.5, p < 0.05). (B) Kaplan-Meier analysis in the TCGA and Taylor databases was conducted to determine whether the OS, DFS and BRFS of patients were associated with ACSS3 gene expression in tumors. (C) According to in-depth analysis of the TCGA and Taylor databases, low ACSS3 expression was positively correlated with tumor stage, metastasis, recurrence and Gleason score. Student's t test; *, p < 0.05; **, p < 0.01; ***, p < 0.001. (D) ACSS3 protein expression was analyzed using PCa tissue arrays from 107 normal tissues and 371 tumor tissues. ACSS3 protein levels were significantly lower in PCa tissues than in normal tissues. Low ACSS3 expression was positively correlated with tumor stage, grade and Gleason score. Student's t test; *, p < 0.05; **, p < 0.01; ***, p < 0.001.

Figure 2

ACSS3 significantly reduced the abnormal lipid accumulation in PCa cells. (A) GSEA of ACSS3 mRNA and PCa signaling pathways. FDR < 25% and p < 0.05 were considered significant. (B-C) ACSS3-overexpressing or ACSS3-knockout PCa cell lines were generated by transfecting overexpressing lentivirus and sgRNA (CRISPR-Cas9). Oil Red O staining was conducted as a visual indicator of intracellular lipids in PCa. The relative diameters of LDs and TG and cholesterol contents were measured as quantitative indicators of lipid accumulation in PCa cells. The results represent the mean ± SEM from 3 independent experiments. (D) An LC/MS lipidomic assay was conducted to detect intracellular lipids in C4-2 cells with or without stable overexpression of ACSS3 (n = 6). Only statistically significant changes (Student's t test, p < 0.05, VIP > 1) are presented, indicated by the red and green colors in the heat map. Downregulated KEGG pathways for lipid metabolism based on the results from the LC/MS lipidomic assay.

Figure 3

ACSS3 promoted ER stress-mediated cell apoptosis. (A) The microarray assays were conducted in C4-2 cells with stable overexpression of ACSS3 (n = 3). GO analysis showed that ACSS3 overexpression significantly upregulated ER stress-, unfolded protein- and misfolded protein-related signaling pathways. (B) GSEA of ACSS3 mRNA and PCa signaling pathways. FDR < 25% and p < 0.05 were considered significant. (C-D) qRT-PCR and Western blot analysis were performed to validate the microarray data. (E) ER Track imaging indicated ER stress expansion in ACSS3-overexpressing 22RV1 cells. The bar is 10 μm. (F) Ultrastructural analysis by transmission electron microscopy (TEM) confirmed the presence of dilated and irregularly shaped rough ER in ACSS3-overexpressing C4-2 and 22RV1 cells or ACSS3-knockout cells. The bar is 0.5 μm. (G) TUNEL assays were performed to detect cell apoptosis. The results represent the mean ± SEM from 3 independent experiments. Student's t test; *, p < 0.05; **, p < 0.01; ***, p < 0.001.

Figure 4

ACSS3 regulated PLIN3 protein stability by affecting PLIN3 ubiquitination levels. (A) PLIN2 and PLIN3 mRNA levels were determined by qRT-PCR in C4-2 and 22RV1 cells with stable ACSS3 overexpression. The results represent the mean ± SEM from 3 independent experiments. Student's t test; ns, no significance. (B) PLIN2 and PLIN3 protein levels were determined by Western blot analysis in C4-2 and 22RV1 cells with stable ACSS3 overexpression, and 22RV1 cells with stable knockout of ACSS3. One representative experiment of 3 independent experiments is shown. (C) 22RV1 cells stably overexpressing ACSS3 were treated with cycloheximide (CHX, 10 μmol/L) to inhibit protein synthesis, and PLIN3 protein turnover was analyzed over time. Similarly, 22RV1 cells with stable knockout of ACSS3 were also treated with CHX. And statistic analysis. One representative experiment of 3 independent experiments is shown. (D) C4-2 and 22RV1 cells stably overexpressing ACSS3 were treated with proteasome inhibitor (MG132, 20 μM) or lysosome inhibitor (chloroquine, 50 μM). Western bolt was performed to detect the level of PLIN3. One representative experiment of 3 independent experiments is shown. (E) PLIN3‐related immunoprecipitation was used to assess the levels of ubiquitinated PLIN3 and AIP4 in C4-2 and 22RV1 cells stably overexpressing ACSS3, and 293T cells stably overexpressing ACSS3, PLIN3 and Myc-Ubquin. (F) PLIN3‐related immunoprecipitation was used to assess the levels of ubiquitinated PLIN3 and AIP4 in C4-2 and 22RV1 cells with stable knockout of ACSS3. One representative experiment of 3 independent experiments is shown.

Figure 5

PLIN3 rescued the ACSS3-mediated tumor cell suppression function. (A) C4-2 and 22RV1 cells with stable overexpression of ACSS3 and overexpression PLIN3. ACSS3 and PLIN3 protein levels were determined by Western blot analysis. (B) CCK-8 assays were performed to detect PCa cell proliferation. Con is a blank vector control for overexpressed ACSS3, and NC is a blank vector control for overexpressed PLIN3. The graphs show the mean ± SEM (n = 6 per group). (C-E) Migration, oil red staining and TG and cholesterol contents were performed to detect PCa cell migration, lipid accumulation, respectively. The results represent the mean ± SEM from 3 independent experiments. (F) Immunofluorescent was performed to detect the influence of ACSS3 in PLIN3 levels and LDs. The bar is 20 μm. (G) Western blot and densitometry and statistical analysis was performed to detect the protein levels of ER stress markers in C4-2 and 22RV1 cells with stable overexpression of ACSS3 and overexpression PLIN3. The results represent the mean ± SEM from 3 independent experiments. Student's t test; *, p < 0.05; **, p < 0.01; ***, p < 0.001; ns, no significance.

Figure 6

ACSS3 inhibited CRPC progression and reversed Enzalutamide resistance. (A-B) LNCaP, 22RV1 and C4-2 cells stably overexpressing ACSS3 were treated with cholesterol (10 µM). A CCK-8 assay was performed. The graphs show the mean ± SEM (n = 6 per group). (C) C4-2 cells with stable overexpression of ACSS3 and PLIN3 knockout. LC-MS was performed to evaluate the T (Testosterone) and DHT (Dihydrotestosterone) contents in the cells. The results represent the mean ± SEM from 3 independent experiments. (D-E) Oil red staining was performed to examine lipid accumulation. Relative diameters of LDs and TG contents and cholesterol contents were measured by image J. The results represent the mean ± SEM from 3 independent experiments. Western blot analysis was performed to examine the protein expression of ACSS3 and PLIN3 in C4-2 and C4-2-ENZR (C4-2-enzalutamide resistance), C4-2-ENZR cells stably overexpressing ACSS3 and the control cells. One representative experiment of 3 independent experiments is shown. (F) C4-2-ENZR and 22RV1 cells stably overexpressing ACSS3 were treated with enzalutamide (ENZ). A CCK8 assay was performed. One representative experiment of 3 independent experiments is shown. (G) NCG mice bearing C4-2-ENZR xenografts with stable ACSS3 overexpression were treated with vector control or enzalutamide (10 mg/kg p.o.) for approximately 7 weeks. Tumor volumes were measured. The graphs show the mean ± SEM (n = 6 per group) One-way ANOVA followed by Tukey's multiple comparison test, α = 0.05; *, p < 0.05; ***, p < 0.001; ns, no significance.

Figure 7

CRISPR/Cas9-ACSS3 knockout mice confirmed the role of endogenous ACSS3. (A) CRISPR/Cas9-ACSS3 knockout mice were generated. (B) IHC and HE staining demonstrated that loss of ACSS3 promoted the ratio of prostatic intraepithelial neoplasia (PIN) in the anterior prostates of transgenic mice. The graphs show the mean ± SEM (n = 10 mice per group). (C) IHC staining was performed to detect markers of cellular proliferation, ER stress, apoptosis and AR activity in the anterior prostates of transgenic mice. Oil red staining was performed to detect lipid accumulation in the anterior prostates of transgenic mice. The graphs show the mean ± SEM (n = 10 mice per group), Student's t test; **, p < 0.01; ***, p < 0.001; NS, no significance. (D) Proposed model illustrating the protective function of ACSS3/PLIN3-mediated LD elimination in PCa progression.