Valproic acid (VPA) is an established drug in the long-term therapy of seizure disorders. Recently, VPA has been associated with anticancer activity, an effect thought to be mediated through the inhibition of cellular histone deacetylase 1. We investigated the effect of various doses of VPA (0, 1.2, and 5.0 mmol/L) administered either acutely or chronically on histone acetylation, p21 gene expression, androgen receptor expression, prostate-specific antigen (PSA) expression, and cell survival and proliferation in prostate cancer cell lines. We also studied the effect of chronic VPA on tumor xenograft growth in vivo. Our results show that acute treatment (3 days) VPA can increase net histone H3 acetylation and up-regulate p21, AR, and cytosolic PSA expression. Interestingly, the effects on AR and PSA are reversed with chronic teatment. In addition, acute VPA reduces cell survival but has no effect on the subsequent proliferation of surviving cells following drug withdrawal. However, when VPA is chronically administered (10-14 days) to prostate cancer cells, even lower doses of VPA result in marked decreases in the net proliferation rate, correlating with increased caspase-2 and caspase-3 activation. These effects are evident in both androgen receptor-positive (LNCaP and C4-2) and androgen receptor-negative (DU145 and PC3) prostate cancer cells. Moreover, chronic VPA treatment results in statistically significant reduction of tumor xenograft growth in vivo. We conclude that acute treatment has nominal effects on prostate cancer cell survival and proliferation, but chronic VPA results in profound decreases in proliferation, independently of androgen regulation. (Cancer Res 2006; 66(14): 7237-44)

Since the advent of prostate-specific antigen (PSA) testing, prostate cancer diagnosis and treatment have resulted in a “stage migration” with the most recent data suggesting improved outcomes (1) with early treatment. However, when primary treatment fails, the disease develops along an intractable course for which no long-term effective therapy exists (2). Removal of androgens may cause a significant acute reduction in cancer burden, but invariably, the cancer develops mechanisms to cope with the lack of normal concentrations of androgens (3). It is the androgen-independent prostate cancers that ultimately lead to patient death (4, 5). One approach to treating advanced prostate cancer has been the evaluation and development of various modulators of epigenetic transcriptional regulation. Acetylation and deacetylation of histones play an important role in the epigenetic regulation of gene expression (6). There are at least two classes of enzymes involved in determining the acetylation state of histones, histone acetyl transferases and histone deacetylases (HDAC), and altered states of these complexes have been associated with various malignancies (7).

HDAC inhibitors are emerging as an exciting new class of potential anticancer agents for the treatment of solid and hematologic malignancies. In recent years, an increasing number of structurally diverse HDAC inhibitors that inhibit proliferation and induce differentiation and/or apoptosis of tumor cells in culture and in animal models have been identified. Despite the large number of HDAC inhibitors that have been described, little is known about their target enzyme specificity. There is increasing evidence to support a distinct biological role for each of the HDAC enzymes, and it is likely that inhibition of specific members of the HDAC family will have specific functional consequences on gene expression, cell cycle regulation, proliferation, differentiation, and apoptosis.

Several HDAC inhibitors have shown impressive antitumor activity in vivo with remarkably little toxicity in preclinical studies and are currently in phase I clinical trials (8). Our results suggest that acute treatment with valproic acid (VPA) leads to growth inhibition and cell death in prostate cancer cells; however, the antiproliferation effect ceases following withdrawal of drug. More importantly, when VPA is administered chronically, a profound decrease in cell proliferation results in vitro with significant reduction in tumor volume at clinically relevant doses in vivo. The ideal application for this type of therapeutic effect would be in patients who have had biochemical recurrence of prostate cancer after radical prostatectomy. Because the tumor burden at the time of PSA recurrence is low and the number of doublings required until death is high, even minimal prolongation of the tumor doubling time, multiplied over many tumor doublings, could result in significant delay in progression. Our findings suggest that chronic VPA may be of potential therapeutic benefit in the treatment of prostate cancer.

Cell lines and chemical. Human prostate cancer cell lines LNCaP, PC3, and DU145 were obtained from American Type Culture Collection (Manassas, VA), and the C4-2 line was obtained from Dr. Leland Chung. The cells were grown in RPMI 1640 with l-glutamine (Cellgro, Herndon, VA) supplemented with 10% heat-inactivated fetal bovine serum (FBS; Life Technologies, Inc., Carlsbad, CA), 5 μg/mL ciprofloxacin hydrochloride (U.S. Biological, Swampscott, MA), and 50 μg/mL gentamicin (Quality Biological, Inc., Gaithersburg, MD). Cells were allowed to grow until 80% to 90% confluent and harvested with 0.05% trypsin/0.53 mmol/L EDTA (Cellgro) before each subsequent passage. VPA (1 mol/L; VPA sodium salt; Sigma, St. Louis, MO) stock was made in PBS and filter sterilized through a 0.22-μm filter.

Western immunoblotting. Seventy percent to 80% confluent T150 flasks of LNCaP, PC3, or DU145 were treated with medium containing 0, 1.2, and 5.0 mmol/L VPA for 48 and 72 hours. The cells were harvested with 0.05% trypsin/0.53 mmol/L EDTA, washed in PBS, and resuspended in 100 μL Mammalian Protein Extraction Reagent (Pierce, Rockford, IL). The BCA protein assay kit (Bio-Rad, Hercules, CA) was used to determine total protein concentration and purified bovine serum albumin (Sigma) to generate the standard curve. Concentrated proteins were separated on a 12% Tris-HCl polyacrylamide gel (Bio-Rad) and transferred to polyvinylidene difluoride membrane (Bio-Rad). The membrane was blocked for an hour in blocking buffer [100 mmol/L Tris-HCl (pH 7.5), 150 mmol/L NaCl, 0.1% Tween 20] with 5% nonfat dry milk and then incubated with rabbit antiacetylated histone H3 (Upstate, Charlottesville, VA) overnight followed by anti-rabbit IgG peroxidase conjugate (Sigma) for 1.5 hours at room temperature. Immunoreactive bands were detected using the enhanced chemiluminescence plus Western blotting detection system (Amersham Biosciences, Piscataway, NJ) according to the manufacturer's instructions. Monoclonal anti-β-actin in mouse (Sigma) and anti-mouse IgG-peroxidase (Sigma) were used to detect β-actin in the same blots. Anti-Cip1/WAF1/p21 mouse monoclonal IgG (Upstate) and anti-mouse IgG peroxidase (Sigma) were used to test p21 expression.

Cell proliferation assay. Prostate cancer cell lines LNCaP, PC3, and DU145 were seeded at 1 × 104 per well in 96-well culture plates and incubated overnight with RPMI 1640 containing 10% FBS. The cells were then treated with medium containing VPA (0, 1.2, and 5.0 mmol/L) for 3 days. The fraction of cells surviving after acute VPA treatment was determined using the Cell Proliferation kit I (Roche, Basel, Switzerland) as per manufacturer's recommendation. To determine the proliferation of the viable cells after acute VPA treatment, LNCaP, PC3, and DU145 cells were seeded and treated in the same manner as above. After the 3-day VPA treatment, cells were washed with PBS and medium changed to RPMI 1640 containing 10% FBS. Viable cells were detected using the Cell Proliferation kit I at 24-hour intervals for 3 days to calculate cell proliferation.

Caspase assay. Prostate cancer cell lines LNCaP, PC3, and DU145 were treated with medium containing 0, 1.2, and 5.0 mmol/L VPA for 1, 3, and 14 days. The cells were harvested by 0.05% trypsin/0.53 mmol/L EDTA. Caspase activity was assessed by measuring conversion of a fluorogenic substrate specific for each caspase subtype with ApoAlert Caspase Assay Plates (BD Biosciences, San Jose, CA) as per manufacturer's recommendation. The plates were read using the CytoFluorII Series 4000 (Applied Biosystems, Foster City, CA; excitation, 380 nm; emission, 460 nm).

Flow cytometry. The apoptosis assay was done using APO-BRDU kit for measuring apoptosis by dual-color flow cytometry (Chemicon International, Temecula, CA). Prostate cancer cell lines LNCaP, PC3, and DU145 were seeded in T150 flasks and incubated overnight with RPMI 1640 containing 10% FBS. The cells were then treated with medium containing 0, 1.2, and 5.0 mmol/L VPA for 3 days. The cells were harvested by 0.05% trypsin/0.53 mmol/L EDTA and fixed with 1% (w/v) paraformaldehyde (Sigma) for 30 to 60 minutes, washed twice with PBS, and resuspended and incubated in the DNA labeling solution for 60 minutes at 37°C. The cells were rinsed with rinse buffer twice and incubated with the fluorescein-PRB-1 antibody solution in the dark for 30 minutes at room temperature followed by propidium iodide (PI)/RNase A solution in the dark for 30 minutes at room temperature. The cells were analyzed by flow cytometry within 3 hours. Both FITC and PI were excited at 488 nm. The emission from FITC was collected on FL1 (530/40 nm bandpass filter), whereas the emission from PI was collected on FL4 (670/30 nm bandpass for linear PI) and FL5 (610/20 nm for log PI). Doublet discrimination was done using pulse width versus forward scatter method. Data are plotted as fold induction of apoptosis using untreated cells as a reference. Statistical analysis was not done on this experiment.

Establishment of tumor xenografts. Human prostate cancer cell lines LNCaP, C4-2, and DU145 were grown to 80% to 90% confluence and harvested. Cells were resuspended in 1× phosphate-buffered saline (pH 7.4; BioSource, Rockville, MD), mixed 1x with Matrigel (BD Biosciences, Palo Alto, California), and injected (1 × 106 per injection) s.c. into the lateral flanks of male athymic nu/nu mice. Once palpable tumors were established, animals were randomized into control and treatment arms with the latter receiving 0.4% VPA in drinking water. Tumor volumes were measured thrice weekly until animals were sacrificed, and tumors were harvested on day 35.

Statistical analysis. Data are presented as means ± SE. All statistical analyses were done on an IBM-compatible computer, running GraphPad Prism 4.0, on Windows XP. All error bars shown represent the SE. All experiments were done with at least three replicates and as many as 20 replicates. Statistical significance was calculated by the Student's t test for paired comparisons, when appropriate, by Wilcoxon rank-sum test, or by repeated measures two-way ANOVA with post-hoc testing for comparison of dose treatment effects, where applicable. P < 0.05 was considered statistically significant. Because there are over 100 paired comparisons (including post-hoc testing), Ps are reported only as P < 0.05 when significant or P < 0.01 when highly significant. Comparisons lacking significance are specifically denoted in the legends and text.

VPA increases histone H3 acetylation. Three prostate cancer cell lines were treated with VPA (0, 1.2, and 5 mmol/L) for 48 and 72 hours, and then H3 acetylation was assessed by Western blot. Shown in Fig. 1A, all three cell lines show significant increases in acetylated H3 compared with the β-actin control at 48 hours. The effect was dose dependent and seemed to be maximal for LNCaP and PC3 at 72 hours and DU145 at 48-hour post-treatment. For reasons not completely understood, levels of acetylated H3 in DU145 remained consistently high at 72 hours in the absence of VPA. It is possible that this phenomenon may be a reflection of cell overgrowth given the rapid proliferation rate of DU145 cells.

Figure 1.

Western blot analysis of acetylated histone H3 and p21WAF/CIP1 expression in VPA-treated cells. A, VPA treatment induces a dose-dependent H3 acetylation in prostate cancer cells. LNCaP, PC3, and DU145 cells were treated with 0, 1.2, and 5.0 mmol/L VPA for 48 and 72 hours. In LNCaP and PC3 cells, H3 acetylation is induced 13- and 21-fold at 72 hours, respectively, whereas, in DU145 cells, the induction is 21-fold at 48 hours of 5.0 mmol/L VPA treatment. The relative fold increase was determined by scanning densitometry of the Western blot normalized to β-actin. B, VPA treatment causes induction in p21WAF/CIP1 expression. In LNCaP, PC3, and DU145 cells, levels of p21 is increased by 12.5-, 21.5-, and 48.5-fold, respectively, after 48 hours of 5.0 mmol/L VPA treatment.

Figure 1.

Western blot analysis of acetylated histone H3 and p21WAF/CIP1 expression in VPA-treated cells. A, VPA treatment induces a dose-dependent H3 acetylation in prostate cancer cells. LNCaP, PC3, and DU145 cells were treated with 0, 1.2, and 5.0 mmol/L VPA for 48 and 72 hours. In LNCaP and PC3 cells, H3 acetylation is induced 13- and 21-fold at 72 hours, respectively, whereas, in DU145 cells, the induction is 21-fold at 48 hours of 5.0 mmol/L VPA treatment. The relative fold increase was determined by scanning densitometry of the Western blot normalized to β-actin. B, VPA treatment causes induction in p21WAF/CIP1 expression. In LNCaP, PC3, and DU145 cells, levels of p21 is increased by 12.5-, 21.5-, and 48.5-fold, respectively, after 48 hours of 5.0 mmol/L VPA treatment.

Close modal

Treatment of malignant cells with HDACI frequently results in increased p21WAF/CIP1 expression and hence was investigated as secondary confirmation of HDACI activity (9). As predicted, a dose-dependent increase p21WAF/CIP1 expression was shown in all three cell lines (Fig. 1B); however, in contrast to the acetylated H3 data, p21WAF/CIP1 was maximal at 48 hours for LNCaP and 72 hours for PC3 and DU145. This suggests that H3 acetylation is not causal for p21 elevation but rather that VPA up-regulates both independently in a dose-dependent fashion.

Effect of acute VPA treatment on cell survival. Previous studies of other HCACIs in prostate cancer have shown significant therapeutic effect (1012). Hence, we sought to determine if VPA had similar activity. Cells were treated with VPA for 3 days and evaluated by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay to evaluate drug effect on cell viability. Results are shown in Fig. 2 as percentage cell viability compared with untreated control. In all three cell lines, VPA decreased cell viability compared with untreated in a dose-dependent, statistically significant manner, although the effect was generally modest and evident in DU145 only at the highest dose level (5 mmol/L VPA). The data that reach statistical significance (P < 0.05, Students t test) when compared with untreated cells are marked with an asterisk.

Figure 2.

The effect of acute VPA treatment on prostate cancer cell viability. LNCaP, PC3, and DU145 cells were treated with 0, 1.2, and 5.0 mmol/L VPA for 72 hours, and the surviving fraction was assayed for by MTT. VPA treatment affects cell viability in a dose-dependent fashion. Comparison of treated versus untreated groups was done by Student's t test (*, P < 0.01 for all comparison relative to untreated cells).

Figure 2.

The effect of acute VPA treatment on prostate cancer cell viability. LNCaP, PC3, and DU145 cells were treated with 0, 1.2, and 5.0 mmol/L VPA for 72 hours, and the surviving fraction was assayed for by MTT. VPA treatment affects cell viability in a dose-dependent fashion. Comparison of treated versus untreated groups was done by Student's t test (*, P < 0.01 for all comparison relative to untreated cells).

Close modal

Previous analyses of different HDACIs have shown apoptosis as the major mechanism of cell death in prostate cancer cells, although potency of this stimulation varies with different cell lines (13, 14). To elucidate whether apoptosis is involved in the cancer cell killing effect of VPA, LNCaP, PC3, and DU145 cells were evaluated by terminal deoxynucleotidyl transferase–mediated dUTP nick end labeling (TUNEL) assays, and the effect was measured by flow cytometry (Supplementary Fig. S1A). Data are shown as scatter plots with fold induction of apoptotic index as insets. These data show that apoptosis is indeed activated in a dose-dependent fashion for all three cell lines; however, maximal induction was <15% even at the highest dose (5 mmol/L VPA), and the effect in DU145 was only significant at the highest dose.

We next attempted to determine which pathway, intrinsic or extrinsic, was involved in the fraction of cells undergoing apoptosis. To this end, direct measurement of various cellular caspase activities (Supplementary Fig. S1B) after acute VPA treatment was done. In this experiment, the cells were treated with medium containing 0, 1.2, and 5.0 mmol/L VPA for 72 hours. Of note, subsequent studies to evaluate the effect of acute VPA on caspase activation are done at 24 hours to avoid duplication of data. The cells were then harvested, and caspase activity was assessed. LNCaP showed the most pronounced increase in caspase-2 and caspase-3 activities, and DU145 showed significant increases in caspase-2 and caspase-3 activities at the higher dose (5 mmol/L) only. PC3 had nominal increases (not significant) in caspase activity after acute treatment with VPA. In all cases, the enzymatic caspase activity was modest and seemed to favor the intrinsic pathway of apoptosis activation (mitochondrial based). These findings are consistent with other HDACIs, which also have been found to favor intrinsic apoptosis pathways (15).

Cellular proliferation after acute VPA treatment. Inasmuch as the majority of the cells treated after VPA survived treatment, we next sought to determine if these cells behaved any differently phenotypically. In particular, we examined whether the proliferation rate was different in surviving prostate cancer cells after acute VPA treatment. Cells were treated for 72 hours with various doses of VPA and then replated (without VPA), and the proliferation was assessed at several time points by MTT as indicated in Fig. 3A-C. The data are plotted as fold cell growth, as the various starting points of the different cells varied according to the amount of cell death, which occurred initially from the VPA treatment. In these experiments, normalization to cell number was necessary because a fraction of viable cells underwent cell death following replating, resulting in unequal cell numbers. Of those cells that survived acute treatment with VPA, the rate of cell growth was found to be identical for VPA-treated and untreated cells in all three cell lines. These data indicate that, although VPA treatment may result in death of a subpopulation of prostate cancer cells, those that survived maintained the same proliferative potential as cells that had never been treated.

Figure 3.

The effect of acute VPA treatment on subsequent cell proliferation. LNCaP, PC3, and DU145 cells were treated with 0, 1.2, and 5.0 mmol/L VPA for 72 hours, washed, and grown in medium without VPA. The proliferation of these cells was monitored by MTT assay. Results are for LNCaP (A), PC3 (B), and DU145 (C). Comparison of groups was done by repeated measures of ANOVA with post-hoc testing revealed no significant differences between treated and untreated groups. Cells that received acute treatment with VPA retain exactly the same proliferation rate, as cells that were never treated.

Figure 3.

The effect of acute VPA treatment on subsequent cell proliferation. LNCaP, PC3, and DU145 cells were treated with 0, 1.2, and 5.0 mmol/L VPA for 72 hours, washed, and grown in medium without VPA. The proliferation of these cells was monitored by MTT assay. Results are for LNCaP (A), PC3 (B), and DU145 (C). Comparison of groups was done by repeated measures of ANOVA with post-hoc testing revealed no significant differences between treated and untreated groups. Cells that received acute treatment with VPA retain exactly the same proliferation rate, as cells that were never treated.

Close modal

Cellular proliferation decreases after chronic VPA treatment. We also sought to determine whether the proliferation rate was different in surviving prostate cancer cells after chronic VPA treatment. Here, we measured cellular proliferation in cultured cells with chronic VPA administration (typically 2-week long cultures). Cells were expanded into serially larger flasks, and total cell counts were measured to maintain an environment that would sustain log-phase cell growth. Although we initially attempted to sustain cultures beyond 2 weeks, the amount of cell death was so significant in all tested cell lines that we were unable to consistently salvage any viable cells beyond that time. As shown in Fig. 4, proliferation of the tested cell lines with chronic VPA administration was profoundly diminished. In all cases, the higher dose of VPA (5 mmol/L) not only decreased the rate of proliferation but actually caused continued cell death so that few viable cells were present at the end of the two week experiment. Of the chronically treated cells that underwent cell death via apoptosis, there were marked increases in caspase-2 and caspase-3 activities compared with acute treatment (Supplementary Fig. S2A-C). For PC3, caspase-8 and caspase-9 activities were also elevated. These findings are consistent with the observation that PC3 phenotypically behaves differently to HDACIs in the activation of apoptosis (13). The effect of chronic treatment with 5 mmol/L VPA in DU145 was so profound that we were unable to harvest sufficient cellular material to do a caspase activity assay (despite multiple attempts to scale up the starting cell number). However, given the fact that caspase-2 and caspase-3 activities were so elevated in DU145 when treated with 5 mmol/L VPA at early time points, it is likely that the intrinsic pathway was still dominant at later time points.

Figure 4.

Chronic VPA treatment reduces cell proliferation. LNCaP (A), PC3 (B), and DU145 (C) cells were expanded continuously in the presence of 0, 1.2, and 5.0 mmol/L VPA for up to 14 days. The number of total viable cells was directly counted by trypan blue exclusion staining at the time of passage and medium change. Comparison of groups was done by repeated measures of ANOVA with post-hoc testing. (treated versus untreated over the entire time course; *, P < 0.01).

Figure 4.

Chronic VPA treatment reduces cell proliferation. LNCaP (A), PC3 (B), and DU145 (C) cells were expanded continuously in the presence of 0, 1.2, and 5.0 mmol/L VPA for up to 14 days. The number of total viable cells was directly counted by trypan blue exclusion staining at the time of passage and medium change. Comparison of groups was done by repeated measures of ANOVA with post-hoc testing. (treated versus untreated over the entire time course; *, P < 0.01).

Close modal

Androgen withdrawal and VPA treatment effects are independent. Because two of the cell lines tested in these experiments (DU145 and PC3) lack appreciable expression of androgen receptor, it is clear that VPA-induced cell death does not require the androgen receptor. However, VPA may still modulate androgen receptor activity with downstream consequences. Hence, we interrogated LNCaP for the effect of VPA treatment on androgen receptor and PSA expression by Western blot analysis, shown in Supplementary Fig. S3A and B. Acute treatment with VPA results in a slight increase in androgen receptor and, consequently, PSA expression. However, chronic treatment with VPA results in a net decrease in androgen receptor and minimal decrease in PSA expression compared with β-actin. These results are consistent with the findings of Abdul et al., who evaluated the effect of VPA on interleukin-6 (IL-6) and PSA secretion in vitro, at a time when it was thought that VPA affected gene expression by blocking calcium channels (16). Although the proposed mechanism may have been in error, their results seem to have been accurate.

Because VPA seems to alter androgen receptor expression, we hypothesized that it might also alter the sensitivity of androgen-independent prostate cancers to androgen withdrawal. To test this hypothesis in vitro, we used the androgen-independent cell line C4-2, which is capable of sustained proliferation in the absence of exogenous androgen but is nonetheless stimulated to enhanced proliferation in the presence of androgen. For these experiments, androgen stimulation was accomplished by the addition of synthetic androgen, R1881 (methyltrienolone), to medium containing charcoal-stripped FBS. When androgen is acutely withdrawn, there is an initial decrease in cellular proliferation (Fig. 5). Treatment with VPA results in decreased cell survival. However, when VPA and androgen withdrawal are combined, the observed result is identical to the predicted result. Hence, in this experiment, the two treatments are additive not synergistic, implying that the two treatment effects occur through completely independent pathways. Because the decrease in survival and proliferation is notable in androgen receptor-negative cell lines, this pathway independence is not surprising. However, because the decreases in proliferation induced by VPA treatment alone is so profound, it is still possible that any conversion from the androgen-independent phenotype to the androgen-dependent phenotype was below the limits of detection by this approach.

Figure 5.

Effect of androgen withdrawal and VPA treatment on androgen-independent prostate cancer cell growth. C4-2 cells, an androgen-independent subline of LNCaP, were grown in medium containing charcoal-stripped FBS with or without 5 nmol/L R1881 and treated with 0 and 5.0 mmol/L VPA. Cell proliferation was measured by MTT. C4-2 cells were chosen for this experiment, as they are capable of proliferation in the absence of androgen and therefore could permit additional measurement of the effect of VPA treatment. Calculated response is plotted based on the assumption of additive effects of VPA and androgen withdrawal. There is no statistically significant difference between the observed and the calculated proliferation of the chronic VPA-treated cells with hormone withdrawal.

Figure 5.

Effect of androgen withdrawal and VPA treatment on androgen-independent prostate cancer cell growth. C4-2 cells, an androgen-independent subline of LNCaP, were grown in medium containing charcoal-stripped FBS with or without 5 nmol/L R1881 and treated with 0 and 5.0 mmol/L VPA. Cell proliferation was measured by MTT. C4-2 cells were chosen for this experiment, as they are capable of proliferation in the absence of androgen and therefore could permit additional measurement of the effect of VPA treatment. Calculated response is plotted based on the assumption of additive effects of VPA and androgen withdrawal. There is no statistically significant difference between the observed and the calculated proliferation of the chronic VPA-treated cells with hormone withdrawal.

Close modal

Chronic VPA treatment inhibits progression of tumor xenografts. To assess the effects of chronic VPA treatment on prostate cancer cell growth in vivo, we established tumor xenografts in male athymic nu/nu mice using both androgen-dependent and androgen-independent prostate cancer cell lines, including LNCaP, C4-2, and DU145. Animals in the treatment arm received 0.4% VPA in drinking water with peak serum VPA measurements ranging from 3 to 70 mg/L. Tumor volumes were measured until week 5 and compared using the Wilcoxon rank-sum test and 2-way ANOVA with post-hoc testing. Animals treated with VPA showed statistically significant reduction in tumor volume compared with controls in all cell lines with 40%, 60%, and 70% reductions in LNCaP, C4-2, and DU145, respectively (Fig. 6). Toxicity studies, including complete blood count, alanine aminotransferase, and aspartate aminotransferase revealed no difference between treated and control animals.

Figure 6.

The effect of chronic VPA on prostate cancer xenografts. LNCaP, C4-2, and DU145 xenografts were established in male athymic mice. Animals in the treatment arm received 0.4% VPA in drinking water. Tumor volumes were compared using Wilcoxon rank-sum tests and 2-way ANOVA with post-hoc testing. (*, P < 0.05).

Figure 6.

The effect of chronic VPA on prostate cancer xenografts. LNCaP, C4-2, and DU145 xenografts were established in male athymic mice. Animals in the treatment arm received 0.4% VPA in drinking water. Tumor volumes were compared using Wilcoxon rank-sum tests and 2-way ANOVA with post-hoc testing. (*, P < 0.05).

Close modal

The primary modes of treatment for prostate cancer are surgery and radiation, either of which can result in treatment failure, typically manifested as biochemical recurrence with a rising PSA. On average, biochemical recurrence of prostate cancer after radical prostatectomy predates radiographic evidence of disease by over 8 years and death by ∼13 years (17). Because the tumor burden at the time of PSA recurrence is low and the number of doublings required until death is high, even small changes in the rate of tumor growth would translate into profound clinical benefit. For instance, if 15 doublings of a tumor are required from the time of biochemical detection until death and the doubling time of the tumor is 10 months, then, a 15% increase in the doubling time would extend life by another 1.9 years. Given that most men with advanced prostate cancer are in the 6th or 7th decade of life, such an increase in survival may actually eliminate many prostate cancer deaths (in exchange for other unrelated causes of death). With these concepts in mind, attention has been focused on the discovery of so-called “differentiation agent” therapies so that advanced prostate cancer can be converted from a terminal disease to a chronic medical condition (18). In this article, we provide preliminary evidence that at least one HDAC inhibitor (VPA) may also be suitable for this purpose.

Several structurally diverse inhibitors of HDAC have been identified, many of which are natural products. HDAC inhibitors can induce growth arrest, differentiation, and/or apoptotic cell death in a wide variety of cultured transformed cells, including neuroblastoma (19), melanoma (20), leukemia (21), as well as breast, prostate, lung, ovary, and colon cancers (22). Induction of apoptosis by HDAC inhibitors has been reported in multiple human cancer cell lines, frequently via intrinsic pathway caspase activation (15, 23).

VPA (2-propylpentanoic acid) is an established drug in the long-term therapy of seizure and bipolar disorders (24). Recently, it has become evident that VPA is also capable of inhibiting HDACs (2527), resulting in altered gene expression in certain susceptible cells. The net result is that VPA inhibits cell growth, regulates differentiation, and may have effects on tumor invasion and angiogenesis (7, 8, 28). We show here that acute treatment with VPA may have limited activity on prostate cancer cell proliferation, whereas chronic administration of VPA has a more profound effect. This phenomenon may be explained by HDACI activation of cell death and cell cycle arrest in G1-G2 (29). These effects are not mutually exclusive, and it is possible that a given cell could be arrested in the cell cycle but then stimulated to reenter the cycle and activate death at a later time. In such a scenario, chronic administration of the HDACI would be predicted to have markedly better therapeutic effect with decreased cell proliferation, as many initially quiescent cells subsequently traverse through the cell cycle check points. This effect is likely not specific to VPA but rather may be present with any number of other HDACIs currently under investigation in solid epithelial malignancies.

One gene most consistently induced by HDAC inhibitors is the cyclin-dependent kinase inhibitor p21WAF/CIP1 (22), which plays an important role in the arrest of cell growth. Butyrate, trichostatin A, depsipeptide, oxamflatin, MS-27-275, and the hydroxamic acid-based HDACIs all induce p21WAF/CIP1 expression (30). Our data confirm these observations, as p21WAF/CIP1 gene expression was associated with an increased level of histone H3 acetylation after VPA treatment (Fig. 1). Although it is not possible to correlate the effects on H3 acetylation and p21 expression, it is clear that VPA causes increased expression of both in a dose-dependent manner. In addition, analysis of apoptosis markers, such as by TUNEL assay or direct caspase activity, shows that VPA initiates a modest apoptotic response through the preferential activation of the mitochondrial (intrinsic) pathway (Supplementary Figs. S1 and S2). Such results are consistent with previous findings of other HDACIs in prostate cancer cells (13). Furthermore, when VPA is given acutely (≤3 days), less than a 30% to 40% reduction in cell viability results. Importantly, those cells that survive the acute VPA treatment show completely normal growth patterns when compared with untreated controls (Fig. 3).

To further investigate the effects of chronic VPA on prostate cancer cells in vivo, we established xenograft models using LNCaP, C4-2, and DU145 cells. The results show statistically significant reduction in tumor volumes of VPA-treated animals versus untreated controls for all cell lines. Tumor reduction ranged from ∼40% to 70% at therapeutic doses of VPA. These effects support our in vitro work, which showed marked cell kill in both androgen-dependent and androgen-independent disease following the chronic administration of VPA. Given its ability to reduce tumor volume in vivo without significant toxicity, VPA may be particularly useful for slowing cancer progression in patients with biochemical recurrence, as the tumor burden is low, and the number of cell divisions required for measurable disease is high compared with most other malignancies.

Development of the androgen-independent phenotype can occur through a diverse variety of mechanisms ranging from deletion or inactivating mutations of the androgen receptor, mutation of androgen receptor resulting in altered activity, such as promiscuous activation with other ligands, or alteration of one or more of the numerous androgen receptor interacting factors, including IL-6 (31), activator protein, c-Jun, HER-2/neu, protein kinase A (32), mitogen-activated protein kinase, and activated Akt (33). Recently, Chen et. al. (34) showed in vitro that androgen receptor amplification alone could result in the development of the androgen-independent phenotype. Multiple other recent studies have implicated androgen receptor as having a role in most androgen-independent prostate cancers (31, 35, 36).

It has been suggested that down-regulation of androgen receptor expression may have potential therapeutic benefit in the treatment of prostate cancer (35). Analysis of the effect of VPA on androgen receptor and PSA expression in LNCaP cells showed a modest initial induction followed by an eventual net decrease in total expression when compared with the β-actin control (Supplementary Fig. S3). This alteration in androgen receptor regulation suggests that, although VPA-induced cell death may not require androgen receptor, treatment with VPA still affects androgen regulation. To better assess this phenomenon, we studied the effect of chronic VPA administration on C4-2 cells, an androgen-independent subline of LNCaP. Removal of androgen stimulation causes a net decrease in proliferation as does treatment with VPA (Fig. 5). However, when the two are combined, the net effect is indistinguishable from the predicted sum of the two treatments. Hence, in this experiment, the two treatments seem additive rather than synergistic. Nonetheless, it should be noted that these data do not preclude the possibility that decreasing the expression level of androgen receptor by chronic VPA results in reversal of the androgen-independent phenotype in susceptible cells. Such determinations are better made in animal models or even in pilot clinical trials.

In summary, we show that chronic administration of VPA results in a marked decrease in proliferation of prostate cancer cells in vitro and significant reduction in tumor volume in vivo. We postulate that this enhanced activity results from capturing the resistant quiescent cells, which are temporarily cell cycle arrested, as they transition back into the cycle. The ideal application for this type of therapeutic effect would be in patients who have had biochemical recurrence of prostate cancer after radical prostatectomy, as they have the least tumor burden and largest number of tumor doublings required before progression to death. In such a scenario, even minimal prolongation of the tumor doubling time, multiplied over many tumor doublings, can result in significant delay in progression. Importantly, the androgen receptor is down-regulated with chronic VPA treatment, an effect that may have downstream consequences in the regulation of androgen sensitivity. These data provide compelling evidence that additional studies are warranted in evaluating the potential use of chronic VPA as a means of altering prostate cancer cell growth and progression.

Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/).

Q. Xia and J. Sung contributed equally to this work.

Grant support: NIH grant 2P50CA58236-09A, award number DAMD17-03-2-0033, Department of Defense Prostate Cancer Research Program, which is managed by the U.S. Army Medical Research and Material Command, and awards from the Prostate Cancer Foundation.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Harvey Clapp for financial support; Maggie Liu for technical assistance; Dr. Leland Chung for giving C4-2 cells; and William B. Nelson, Donald S. Coffey, Theodore DeWeese, Robert Vasella, Jonathan Simons, and Shawn Lupold for many helpful discussions.

1
Murphy AM, McKiernan JM, Olsson CA. Controversies in prostate cancer screening.
J Urol
2004
;
172
:
1822
–4.
2
Diaz M, Patterson SG. Management of androgen-independent prostate cancer.
Cancer Control
2004
;
11
:
364
–73.
3
Navarro D, Luzardo OP, Fernandez L, Chesa N, Diaz-Chico BN. Transition to androgen-independence in prostate cancer.
J Steroid Biochem Mol Biol
2002
;
81
:
191
–201.
4
Kent EC, Hussain MH. The patient with hormone-refractory prostate cancer: determining who, when, and how to treat.
Urology
2003
;
62
:
S134
–40.
5
Scher HI, Zhang ZF, Nanus D, Kelly WK. Hormone and antihormone withdrawal: implications for the management of androgen-independent prostate cancer.
Urology
1996
;
47
:
61
–9.
6
Marks PA, Richon VM, Rifkind RA. Histone deacetylase inhibitors: inducers of differentiation or apoptosis of transformed cells.
J Natl Cancer Inst
2000
;
92
:
1210
–6.
7
Blaheta RA, Cinatl J, Jr. Anti-tumor mechanisms of valproate: a novel role for an old drug.
Med Res Rev
2002
;
22
:
492
–511.
8
Gottlicher M. Valproic acid: an old drug newly discovered as inhibitor of histone deacetylases.
Ann Hematol
2004
;
83
:
S91
–2.
9
Sowa Y, Orita T, Hiranabe-Minamikawa S, et al. Histone deacetylase inhibitor activates the p21/WAF1/Cip1 gene promoter through the Sp1 sites.
Ann N Y Acad Sci
1999
;
886
:
195
–9.
10
Carducci MA, Nelson JB, Chan-Tack KM, et al. Phenylbutyrate induces apoptosis in human prostate cancer and is more potent than phenylacetate.
Clin Cancer Res
1996
;
2
:
379
–87.
11
Ng AY, Bales W, Veltri RW. Phenylbutyrate-induced apoptosis and differential expression of Bcl-2, Bax, p53, and Fas in human prostate cancer cell lines.
Anal Quant Cytol Histol
2000
;
22
:
45
–54.
12
Lea MA, Shareef A, Sura M, desBordes C. Induction of histone acetylation and inhibition of growth by phenyl alkanoic acids and structurally related molecules.
Cancer Chemother Pharmacol
2004
;
54
:
57
–63.
13
Fronsdal K, Saatcioglu F. Histone deacetylase inhibitors differentially mediate apoptosis in prostate cancer cells.
Prostate
2005
;
62
:
299
–306.
14
Butler LM, Agus DB, Scher HI, et al. Suberoylanilide hydroxamic acid, an inhibitor of histone deacetylase, suppresses the growth of prostate cancer cells in vitro and in vivo.
Cancer Res
2000
;
60
:
5165
–70.
15
Peart MJ, Tainton KM, Ruefli AA, et al. Novel mechanisms of apoptosis induced by histone deacetylase inhibitors.
Cancer Res
2003
;
63
:
4460
–71.
16
Abdul M, Hoosein N. Inhibition by anticonvulsants of prostate-specific antigen and interleukin-6 secretion by human prostate cancer cells.
Anticancer Res
2001
;
21
:
2045
–8.
17
Pound CR, Partin AW, Eisenberger MA, Chan DW, Pearson JD, Walsh PC. Natural history of progression after PSA elevation following radical prostatectomy.
JAMA
1999
;
281
:
1591
–7.
18
Parnes HL, House MG, Kagan J, Kausal DJ, Lieberman R. Prostate cancer chemoprevention agent development: the National Cancer Institute, Division of Cancer Prevention portfolio.
J Urol
2004
;
171
:
S68
–75.
19
Ajamian F, Salminen A, Reeben M. Selective regulation of class I and class II histone deacetylases expression by inhibitors of histone deacetylases in cultured mouse neural cells.
Neurosci Lett
2004
;
365
:
64
–8.
20
Yoshida M, Furumai R, Nishiyama M, et al. Histone deacetylase as a new target for cancer chemotherapy.
Cancer Chemother Pharmacol
2001
;
48
:
S20
–6.
21
Huang H, Reed CP, Zhang JS, et al. Carboxypeptidase A3 (CPA3): a novel gene highly induced by histone deacetylase inhibitors during differentiation of prostate epithelial cancer cells.
Cancer Res
1999
;
59
:
2981
–8.
22
Kramer OH, Gottlicher M, Heinzel T. Histone deacetylase as a therapeutic target.
Trends Endocrinol Metab
2001
;
12
:
294
–300.
23
Medina V, Edmonds B, Young GP, et al. Induction of caspase-3 protease activity and apoptosis by butyrate and trichostatin A (inhibitors of histone deacetylase): dependence on protein synthesis and synergy with a mitochondrial/cytochrome c-dependent pathway.
Cancer Res
1997
;
57
:
3697
–707.
24
Loscher W. Basic pharmacology of valproate: a review after 35 years of clinical use for the treatment of epilepsy.
CNS Drugs
2002
;
16
:
669
–94.
25
Phiel CJ, Zhang F, Huang EY, et al. Histone deacetylase is a direct target of valproic acid, a potent anticonvulsant, mood stabilizer, and teratogen.
J Biol Chem
2001
;
276
:
36734
–41.
26
Thelen P, Schweyer S, Hemmerlein B, et al. Expressional changes after histone deacetylase inhibition by valproic acid in LNCaP human prostate cancer cells.
Int J Oncol
2004
;
24
:
25
–31.
27
Gottlicher M, Minucci S, Zhu P, et al. Valproic acid defines a novel class of HDAC inhibitors inducing differentiation of transformed cells.
EMBO J
2001
;
20
:
6969
–78.
28
Cinatl J, Jr., Kotchetkov R, Blaheta R, et al. Induction of differentiation and suppression of malignant phenotype of human neuroblastoma BE(2)-C cells by valproic acid: enhancement by combination with interferon-α.
Int J Oncol
2002
;
20
:
97
–106.
29
Marks PA, Richon VM, Miller T, Kelly WK. Histone deacetylase inhibitors.
Adv Cancer Res
2004
;
91
:
137
–68.
30
Mei S, Ho AD, Mahlknecht U. Role of histone deacetylase inhibitors in the treatment of cancer (review).
Int J Oncol
2004
;
25
:
1509
–19.
31
Grossmann ME, Huang H, Tindall DJ. Androgen receptor signaling in androgen-refractory prostate cancer.
J Natl Cancer Inst
2001
;
93
:
1687
–97.
32
Horvai AE, Xu L, Korzus E, et al. Nuclear integration of JAK/STAT and Ras/AP-1 signaling by CBP and p300.
Proc Natl Acad Sci U S A
1997
;
94
:
1074
–9.
33
Ghosh PM, Malik S, Bedolla R, Kreisberg JI. Akt in prostate cancer: possible role in androgen-independence.
Curr Drug Metab
2003
;
4
:
487
–96.
34
Chen CD, Welsbie DS, Tran C, et al. Molecular determinants of resistance to antiandrogen therapy.
Nat Med
2004
;
10
:
33
–9.
35
Santos AF, Huang H, Tindall DJ. The androgen receptor: a potential target for therapy of prostate cancer.
Steroids
2004
;
69
:
79
–85.
36
Zegarra-Moro OL, Schmidt LJ, Huang H, Tindall DJ. Disruption of androgen receptor function inhibits proliferation of androgen-refractory prostate cancer cells.
Cancer Res
2002
;
62
:
1008
–13.

Supplementary data