Inhibition of lipogenesis and induction of apoptosis by valproic acid in prostate cancer cells via the C/EBPα/SREBP-1 pathway

Background: Lipid metabolism reprogramming is now accepted as a new hallmark of cancer. Hence, targeting the lipogenesis pathway may be a potential avenue for cancer treatment. Valproic acid (VPA) emerges as a promising drug for cancer therapy, however, the underlying mechanisms are not yet fully understood. This study aimed to investigate the effects and mechanisms of VPA on cell viability, lipogenesis, and apoptosis in human prostate cancer PC-3 cells. Methods: The effects of VPA on the viability and migration of PC-3 cells were investigated using MTT cell viability assay and wound-healing assay. Oil-Red O staining was used to examine lipid droplets, and DAPI staining assay and Annexin V-FITC and PI double-staining assay were used to measure the extent of cell apoptosis. Quantitative real-time PCR and Western blotting were used to determine the expression of lipogenesis and apoptosis genes. Statistical and analytical data were analyzed with SPSS 17.0 Software, and statistical signicance was set to * P < 0.05, ** P < 0.01, and *** P < 0.001 levels. Results: The results showed that VPA signicantly reduced lipid accumulation and induced apoptosis of PC-3 cells. Moreover, the expression of CCAAT/enhancer-binding protein α (C/EBPα), as well as sterol regulatory element binding protein 1 (SREBP-1) and its downstream effectors, including fatty acid synthase (FASN), acetyl CoA carboxylase 1 (ACC1), and antiapoptotic B cell lymphoma 2 (Bcl-2), markedly decreased in PC-3 cells after VPA administration. Mechanistically, the overexpression of C/EBPα rescued the levels of SREBP-1, FASN, ACC1, and Bcl-2, enhanced lipid accumulation and attenuated apoptosis of VPA-treated PC-3 cells. Conversely, knockdown of C/EBPα by siRNA further decreased lipid accumulation, enhanced apoptosis, and reduced the levels of SREBP-1, FASN, antisense: 5′-AUUGUCACUGGUCAGCUCCdTdT-3′. and SREBP-1 in PC-3 cells treated with various concentrations of VPA was analyzed by Western blotting analysis. The levels of C/EBPα and SREBP-1 were quanti ﬁ ed by densitometry and normalized with those of β-Actin. and PC-3 cells were transfected with C/EBPα-expressing plasmid, followed by treatment with or without 2.0 mM VPA for 48 h. The levels of C/EBPα, SREBP-1, FASN, and Bcl-2 were measured by Western blotting analysis. The fold-change was calculated based on a densitometric analysis of the band intensities. were transfected empty were transfected followed VPA The SREBP-1, measured transfected Statistically signicant differences indicated as fellows: NS (no signicance),

Histone deacetylases are crucially involve in regulating the etiology and progression of prostate cancer.
Valproic acid (VPA), a histone deacetylase inhibitor, has been used as an anticonvulsant and moodstabilizing drug for more than 40 years. Previous study indicates that VPA suppresses adipogenesis and decreases the expression levels of peroxisome-proliferator-activated receptor γ, CCAAT/enhancer-binding protein α (C/EBPα), and FASN, all of which are key regulators of adipogenesis in adipocytes [8].
Treatment of ob/ob mice with VPA for 14 days results in decreased hepatic fat accumulation [9]. Recent studies have shown that VPA can suppress the malignancy of various cancers including prostate cancer, glioblastoma, and melanoma because it inhibits tumor growth and metastasis, induces differentiation and apoptosis, and enhances chemotherapy sensitivity [10][11][12]. Although a study indicates that VPA pretreatment suppresses PCa cell viability [13], its potential roles and, more importantly, the underlying mechanisms of its actions have not been extensively studied.
The present study aimed to gain insights into the antitumor mechanisms of VPA in PCa cells. The results showed that VPA inhibited cell viability in PC-3 cells through suppressing lipid accumulation and inducing cell apoptosis. It involved the key regulators of lipogenesis, including C/EBPα, sterol regulatory element binding protein-1 (SREBP-1) and their target genes. These ndings might provide insights for PCa therapy targeting lipogenesis and apoptosis.
MTT cell viability assay PC-3 cells were seeded in a 96-well plate at a density of 2 × 10 4 cells/well and incubated overnight, followed by treatment with the indicated concentrations of VPA. Then 10 μL of 5 mg/mL MTT (purchased from Amresco Inc., OH, USA) in phosphate-buffered saline (PBS) was added to each well and the cells were incubated in the dark for 4 h at 37˚C. The cell culture medium was replaced with 150 μL of dimethyl sulfoxide, and the mixture was stirred for 10 min. The optical density (OD) values for samples were measured using a Pan-wavelength microplate reader at 570 nm. Each test included a blank containing a complete culture medium without cells. Triplicate wells were used for each sample, and the experiments were repeated at least three times to get means and standard deviations.

Wound-healing assay
The cell migration was examined using the wound-healing assay. PC-3 cells in each group were seeded into a 6-well plate. When the cells reached 60%-80% con uence, a wound was built across the cell monolayer using a BioClean 1000 μL plastic pipette tip. The cells were rinsed with PBS three times and incubated with fresh medium containing different dosages of VPA for another 4 days or 2.0 mM VPA for another 2, 4, and 6 days. The cell proliferation and migration into the wound area were photographed under an inverted microscope (Leica, Wetzlar, Germany) with 100× magni cation. The relative migration rate was calculated using the following formula: relative migration rate = (distance between the gap at 0 h -distance between the gap at each time point)/distance between the gap at 0 h × 100%.

Oil-Red O staining
The adherent PC-3 cells grown on a glass coverslip at the bottom of a 6-well plate were xed in 4% paraformaldehyde for 15 min and then analyzed with an Oil-Red O stain kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, China) according to the supplier's instructions. After being rinsed in 60% isopropanol to remove unbound dye and counterstained with hematoxylin, the samples were taken photo of under an Olympus BX53 microscope (Olympus Corporation, Tokyo, Japan) equipped with a DP72 microscope digital camera and Image-Pro Plus 7.0 software. In order to quantify lipid accumulation, Oil-Red O was extracted with 100% isopropanol, and the light absorbance of the solution was measured at 520 nm.
Then they were stained with 10 μg/mL DAPI (purchased from Roche Corporation, Basel, Switzerland) at 37°C in the dark for 10 min. The nuclear morphology was viewed under ultraviolet light, and the images were captured using an inverted uorescence microscope (Leica, Wetzlar, Germany). Apoptotic cells were identi ed according to characteristic changes, including nuclear condensation, fragmentation, and presence of apoptotic bodies [14].
Annexin V-FITC and PI double-staining assay PC-3 cells (4 × 10 5 per well) were seeded in a 6-well plate overnight and treated with various concentrations of VPA for 4 days. In order to measure the extent of cell apoptosis, an Annexin Vuorescein isothiocyanate (FITC) and PI double-staining apoptosis detection kit was used according to the manufacturer's instructions (Sungene Biotech, Tianjin, China). In brief, the cells were harvested and suspended in 1× binding buffer. Then, 100 μL of the cell suspension was incubated with 5 μL of Annexin V-FITC for 10 min, followed by incubation with 5 μL of PI solution for another 5 min. The labeled cells were then assessed by ow cytometry and analyzed using Cellquest 6.0 (BD Biosciences, NJ, USA) [15].

Statistical analysis
The experiments were repeated at least three times, and data were analyzed using SPSS (version 17.0; SPSS, Inc., IL, USA). All data were expressed as the mean ± S.D. A statistical analysis between two groups was performed using the Student¢s t-test. One-way analysis of variance was used for the bar plots containing three or more groups. * P < 0.05, ** P < 0.01, and *** P < 0.001 indicate signi cant differences, whereas the nonsigni cant difference is denoted by NS.

Results
Effects of VPA on PC-3 cell viability The incidence of PCa exhibits an increasing trend in recent years. Previous studies indicated that VPA is a potential suppressor of the growth of PCa cells [10,13,16]. To test whether the growth inhibition was dose dependent, PC-3 cells were treated with different concentrations of VPA (ranging from 0.5 to 10.0 mM) for 4 days, followed by the MTT cell viability assay. The result showed that the growth of PC-3 cells was slow after VPA administration. The growth ability decreased in a dose-dependent manner, ranging from 7.52% for the low VPA concentration (0.5 mM) up to 76.7% for the high VPA concentration (10.0 mM) compared to the control cells (Fig. 1a, left). Meanwhile, PC-3 cells were treated with 2.0 mM VPA over a time course of 1, 2, 4, and 6 days. As shown in Figure 1a (right), VPA inhibited the growth of PC-3 cells signi cantly in a time-dependent manner, ranging from 9.73% for 1 day up to 64.4% for 6 days compared to the control cells. The inhibitory effect of VPA on PC-3 cell migration was con rmed by the wound-healing assay. Consistently, similar results were observed, and showed that VPA markedly delayed the wound closure of PC-3 cells (Fig. 1b and 1c). Altogether, the results demonstrate that VPA remarkably prevents the PC-3 cell viability.

Effects of VPA on lipogenesis in PC-3 cells
Cancer is a disorder characterized by increased metabolic activity leading to enhanced cell growth and proliferation. Alterations in lipid metabolism are one of the main features in cancer cells [5,17,18]. Therefore, the functional effects of VPA on lipogenesis in PC-3 cells were determined in this study. PC-3 cells were treated with 0, 0.5, 1.0, 2.0, and 5.0 mM VPA for 4 days, followed by Oil-red O staining and Western blotting analysis. The results showed that VPA decreased lipid deposition in PC-3 cells in a dosedependent manner, as evidenced by Oil-red O staining (Fig. 2a, b). The expression levels of FASN and ACC1, which are the key enzymes of de novo lipogenesis, were signi cantly suppressed in a dosedependent manner when PC-3 cells were cultured in a medium containing VPA (Fig. 2c, d).
Effects of VPA on apoptosis in PC-3 cells VPA is now a promising anticancer compound due to its pharmacological effects on lipogenesis and apoptosis [8][9][10]. Therefore, the degree of apoptosis for VPA-treated PC-3 cells was analyzed by DAPI staining assay and Annexin V-FITC and PI double-staining assay. DAPI staining showed that the nuclear morphology of VPA-treated cells, including the rippled surface of nuclei, chromatin condensation, and nuclear fragmentation, presented more and brighter blue uorescence compared to the control cells, suggesting the typical characteristics of apoptosis (Fig. 3a). To further investigate the apoptosis of PC-3 cells induced by VPA, the apoptotic cells were quanti ed with a ow cytometer using Annexin V-FITC and PI double-staining. As shown in Figure 3b and 3c, the percentage of apoptotic cells induced by VPA exhibited a signi cant increase compared to the control cells (from 8.89% to 22.73%). Consistent with these results, Western blotting analysis showed that the expression level of Bcl-2 in VPA-treated PC-3 cells signi cantly decreased in a dose-dependent manner (Fig. 3d, e). Collectively, the results above suggest that VPA suppresses the growth of PC-3 cells by inhibiting lipogenesis and inducing apoptosis.
C/EBPα regulated lipogenesis and apoptosis-related genes in VPA-treated PC-3 cells C/EBPα and SREBP-1 play pivotal roles in lipid metabolism by inducing the transcription of genes related to lipogenesis [8,19]. Recent studies revealed that C/EBPα and SREBP-1 are markedly upregulated in human cancers, providing the mechanistic link between lipid metabolism alterations and malignancies [18,20,21]. To identify the metabolic changes occurring during VPA-induced antitumor progression, the expression levels of C/EBPα and SREBP-1 were determined by Western blotting analysis. As shown in Figure 4a and 4b, the expression levels of C/EBPα and SREBP-1 were signi cantly suppressed in PC-3 cells treated with VPA in a dose-dependent manner. Furthermore, the ectopic expression of C/EBPα inverted the downregulation of SREBP-1, FASN, ACC1, and Bcl-2 caused by VPA, as shown by Western blotting analysis (Fig. 4c, d). The siRNA strategy was used to speci cally knockdown C/EBPα and Western blotting analysis result con rmed the dramatic C/EBPα downregulation in PC-3 cells. As expected, the decrease of protein levels of SREBP-1, FASN, ACC1, and Bcl-2 was ampli ed in PC-3 cells transfected with siRNA sequences speci cally against C/EBPα (in short, Si-C/EBPα) compared to the cells transfected with scrambled control siRNA (in short, Si-Control) (Fig. 4e, f). These results indicate that C/EBPα plays an important role in expression of SREBP-1, FASN, ACC1, and Bcl-2 in VPA-treated PC-3 cells. Hence it is presumed that the C/EBPα/SREBP-1 pathway might be involved in the suppression of lipogenesis and induction of apoptosis in PC-3 cells after VPA administration.

Effects of SREBP-1a and SREBP-1c on expression of lipogenesis and apoptosis-related genes in VPAtreated PC-3 cells
The changes in cell metabolism and growth are closely linked through SREBPs. The SREBP family consists of three subtypes: SREBP-1a, SREBP-1c, and SREBP-2. SREBP-1a and SREBP-1c are derived from a single gene using of alternative transcription start sites. In general, SREBP-1c seems to act more speci cally on the genes involved in fatty acid synthesis, while SREBP-1a affects a large number of genes involved in the regulation of lipid metabolism, cell proliferation, differentiation, and death [19][20][21][22]. Importantly, our previous study indicates that C/EBPα enhances SREBP-1a activation at the transcriptional level [23]. Hence the present study examined the effects of C/EBPα on the transcriptional levels of SREBP-1a and SREBP-1c in PC-3 cells in the presence of VPA. PC-3 cells were transfected with C/EBPα-expressing plasmid or Si-C/EBPα for 48 h in the presence of VPA (2.0 mM), and the transcriptional levels of SREBP-1a and SREBP-1c were tested by qRT-PCR analysis. The results demonstrated that ectopic expression of C/EBPα neutralized the suppression of SREBP-1a and SREBP-1c activation, while Si-C/EBPα exerted an inhibitory effect and enhanced suppression in PC-3 cells exposed to VPA (Fig. 5a). In addition, the ectopic expression of SREBP-1a rescued the expression of FASN and ACC1 as well as the expression of Bcl-2 in VPA-treated cells (Fig. 5b, c). Different from SREBP-1a, the ectopic expression of SREBP-1c rescued the expression of FASN and ACC1 but had no effect on Bcl-2 expression in VPA-treated cells (Fig. 5d, e).
Effects of the C/EBPα/SREBP-1 pathway on lipogenesis and apoptosis in VPA-treated PC-3 cells Subsequently, the effects of C/EBPα and SREBP-1 on PC-3 cell viability, lipogenesis, and apoptosis were examined. MTT cell viability assay exhibited an apparent increase in C/EBPα-or SREBP-1-expressing cells and a sharp decrease in Si-C/EBPα cells compared to the control cells (Fig. 6a). The ectopic expression of C/EBPα or SREBP-1 led to a signi cant increase in lipid accumulation in PC-3 cells treated with VPA compared to the control cells. The loss of lipids and smaller lipid droplets as a result of C/EBPα knockdown was also con rmed by Oil-red O staining (Fig. 6b, c). The ectopic expression of C/EBPα or SREBP-1 partially suppressed VPA-induced apoptosis in PC-3 cells, whereas Si-C/EBPα enhanced it (Fig.  6d, e). Taken together, it is concluded that VPA treatment signi cantly decreases lipid accumulation and induces apoptosis as a result of the inhibition of C/EBPα as well as SREBP-1 and its target gene expression. VPA-triggered suppression of the C/EBPα/SREBP-1 pathway might provide mechanistic insights and offer a novel therapeutic strategy for PCa.

Discussion
Metabolic reprogramming not only promotes cancer cell plasticity but also provides novel insights for treatment strategies. Altered lipid metabolism is increasingly recognized as a signature of cancer cells, and the proteins involved in this process can be promising chemotherapeutic targets for cancer treatment [24][25][26][27]. A link between PCa progression and lipogenesis has been investigated in the past few decades. Several recent studies suggest that the activation of de novo lipogenesis and cholesterogenesis induces PCa cell proliferation and promotes PCa development and progression. Hence, targeting the lipogenesis pathway may be a promising avenue for PCa treatment [26][27][28][29].
VPA is known as a histone deacetylase inhibitor and an anticonvulsant and mood-stabilizing drug. It has been reported as a potent and promising anticancer drug candidate [10][11][12]. In the present study, the anticancer effects of VPA on human prostate cancer PC-3 cells were detected using MTT cell viability assay and wound-healing assay. The results indicated that VPA inhibited cell growth signi cantly in a dose-dependent and time-dependent manner. Moreover, VPA signi cantly inhibited lipogenesis and induced cellular apoptosis in PC-3 cells. Further assays were performed to analyze the underlying molecular mechanisms. C/EBPα and SREBP-1 are crucial factors controlling lipogenesis. It is reported that VPA suppresses the accumulation of intracellular triacylglycerol and decreases the expression level of C/EBPα [8]. In addition, a growing amount of evidence suggests that aberrant SREBP-1 activity can contribute to cancer [30]. A previous study indicates that SREBP-1 was overexpressed and involved in the transcriptional regulation of fatty acid synthesis through the altered expression of FASN in prostate cancer cells [19]. Hence, the present study focused on C/EBPα and SREBP-1. The results showed that the expression levels of C/EBPα and SREBP-1, as well as their target genes FASN, ACC1, and Bcl-2, decreased signi cantly after VPA treatment.
An adapted ''metabolic switch'' accompanies most physiological and pathological changes in cellular functions. Lipid metabolism is essential for cancer cells and is associated with the regulation of a variety of key cellular processes and functions [31][32][33]. The cellular metabolites can directly or indirectly regulate the apoptotic machinery. Metabolism is emerging as one of the key factors contributing to the dysregulation of apoptosis in cancer [34]. The suppression of C/EBPα inhibits cell proliferation by inducing G1-phase arrest and apoptosis [20]. Emerging evidences indicate that SREBP-1a couples lipid synthesis to cell progression and apoptosis [28,35]. A recent study shows that SREBP-1a regulates antiapoptotic factor apoptosis inhibitor 6 (Api6) besides regulating lipid metabolism in response to different nutrient levels. Consistently, the results revealed that the antiapoptotic Bcl-2 expression regulated by SREBP-1a might play a role in the apoptosis of VPA-treated PC-3 cells. The study suggests distinct effects of the two SREBP-1 isoforms, with SREBP-1a playing a more important role in VPAinduced apoptosis in PC-3 cells.
PCa is the most prevalent urological cancer and the heterogeneous nature of PCa is well known [36,37]. The castration-resistant prostate cancer (CRPC) is associated with a poor prognosis and the treatment of CRPC is still clinically di cult. The molecular targets for CRPC remain unclear and therapeutic approaches for patients with CRPC remain less well understood. A better understanding of the molecular biology of CRPC will lead to a dramatic increase in the treatment of patients. In the present study, the castration-resistant prostate cancer PC-3 cell line was used and the results demonstrate that VPA is a potential avenue for CRPC treatment.

Conclusions
Together, the present study indicates that VPA inhibits prostate cancer growth via the C/EBPα/SREBP-1 pathway targeting lipogenesis and apoptosis. Therefore, blocking histone deacetylases (HDAC) or fatty acid biosynthesis could be a potential treatment for PCa. Because of the cellular heterogeneity observed among prostate cancer patients, it would be more convincing if we use more prostate cancer cell lines (C4-2, 22Rv1 or DU145 etc.) to explore the effects and mechanism of VPA. Our results are expected to provide new insights into the treatment of prostate cancer, nd potential drug target proteins involved in dysregulated lipid metabolism and apoptosis, and explore the possibility of combined treatment targeting lipid metabolism and apoptosis in prostate cancer.   cells was analyzed by Western blotting analysis. β-Actin was used as a loading control. (e) The foldchange was calculated based on the densitometric analysis of the band intensities. All experiments were performed three times. Data are expressed as mean ± S.D. Statistically signi cant differences are indicated as fellows: NS (no signi cance), *P < 0.05, **P < 0.01, and ***P < 0.001.