Methylseleninic acid downregulates hypoxia-inducible factor-1α in invasive prostate cancer

Authors


Abstract

Alternative strategies are needed to control growth of advanced and hormone refractory prostate cancer. In this regard, we investigated the efficacy of methylseleninic acid (MSeA), a penultimate precursor to the highly reactive selenium metabolite, methylselenol, to inhibit growth of invasive and hormone refractory rat (PAIII) and human (PC-3 and PC-3M) prostate cancer cells. Our results demonstrate that MSeA inhibits PAIII cell growth in vitro as well as reduces weights of tumors generated by PAIII cells treated ex vivo. A significant reduction in the number of metastatic lung foci by MSeA treatment was also noted in Lobund-Wistar rats. The PAIII cells along with PC-3, DU145 and PC-3M cells undergo apoptosis after MSeA treatments in both normoxia and hypoxia. Treatment of metastatic rat and human prostate cancer cell lines with MSeA decreased hypoxia-inducible factor-1α (HIF-1α) levels in a dose-dependent manner. Additionally, HIF-1α transcription activity both in normoxic and hypoxic conditions is reduced after MSeA treatment of prostate cancer cells. Furthermore, VEGF and GLUT1, downstream targets of HIF-1α, were also reduced in prostate cancer cells after MSeA treatment. Our study illustrates the efficacy of MSeA in controlling growth of hormone refractory prostate cancer by downregulating HIF-1α, which is possibly occurring through stabilization or increase in prolyl hydroxylase activity.

In men with advanced prostate cancer, hormone therapy is accepted as the initial treatment of choice and produces good responses in most patients. However, many patients relapse and become resistant to further hormone manipulation. Radical prostatectomy is recommended for treatment of localized prostate cancer. Unfortunately, local recurrences occur in up to one-third of patients by 5 years after surgery.1 Furthermore, a combination of hormone therapy with radical prostatectomy in patients with localized disease resulted in 5-year disease-free survival of 64%.2

Effective radiation therapy requires the presence of oxygen.3 As a result of rapid mitotic growth and clonal expansion, tumor cells are usually forced away from vessels beyond effective diffusion distance of oxygen and thrive remarkably well within a hypoxic environment. This tumor hypoxia may lead to resistance to radiation and chemotherapy.4 Hypoxia within solid tumors induces hypoxia-inducible factor (HIF)-1α,5 and its elevation in prostate cancer cells may attribute to enhanced growth rates, survival and increased metastatic potential.6

HIF-1 is a heterodimer composed of an inducible α-subunit that confers the sensitivity to oxygen and a constitutively expressed β-subunit, aryl hydrocarbon receptor nuclear translocator.7 Under normoxic conditions after a post-translational hydroxylation by prolyl hydroxylase (PHD), HIF-1α interacts with tumor suppressor von-Hippel-Lindau protein and is rapidly degraded via ubiquitin-dependent proteasome pathway.8 Hypoxia induces a rapid increase in HIF-1α protein stability and transcriptional activity,9 resulting in the activation of target genes involved in erythropoiesis, glycolysis and angiogenesis.10 HIF-1α also enhances expression of genes coding for growth factors, growth factor receptors, components of apoptotic pathway and cell cycle regulators.10–12 With a shift from traditional cytotoxic chemotherapy toward targeted therapies, development of new targeted chemical compounds as well as noninvasive targeted strategies toward HIF-1α is becoming increasingly attractive.

Several antitumor agents including EF24 (curcumin analog),13 glucosamine,14 apigenin,15 quercetin,16 NS398 (Cox-2 inhibitor),17 doxorubicin (anthracyclin),18 trichostatin A (histone deacetylase inhibitor)19 and rapamycin (mTOR inhibitor)20 have been shown to inhibit HIF-1α in numerous cancer cell types including those of the prostate. These agents may lead to decreased HIF-1α DNA binding, decreased HIF-1α synthesis or decreased HIF-1α transactivation.21 A hallmark behind recent strategies of prostate cancer treatment is identifying chemotherapeutic agents with less toxicity or fewer side effects.

Several epidemiologic and experimental studies as well as clinical intervention trials have supported the hypothesis that enhanced selenium status reduces the risk of prostate cancer.22–28 The majority of the inhibitory effects of selenium compounds involve apoptosis as a critical event, and several effective organic selenium compounds including seleno-keto acids have been reviewed on the basis of their apoptotic potential in several cancer models including prostate cancer.29, 30 A promising anticancer agent methylseleninic acid (MSeA) has been shown to be effective in inhibiting prostate cancer growth in vitro and in TRAMP model.31, 32 However, the potential of MSeA as a therapeutic agent has not been tested in humans, and the underlying mechanism(s) of selenium compounds including MSeA have not been studied in hypoxic conditions.

The PAIII model of invasive prostate cancer in Lobund-Wistar (LW) rats shows areas of hypoxia as clearly identified by pimonidazole staining. Hypoxia is a key regulatory factor in prostate cancer, and compromised oxygenation of tumor tissue plays a role in the aggressive nature of these tumors in LW rats.33 Furthermore, transient epithelial cell hypoxia has been shown to rapidly increase HIF-1α and its downstream targets such as VEGF in rat prostate tumors.34

Our investigation tested the hypothesis that MSeA inhibits HIF1α and its downstream targets involved in growth of invasive prostate cancer under hypoxic conditions.

Abbreviations

DMOG: dimethyloxallyl glycine; DMSO: dimethylsulfoxide; GLUT1: glucose transporter 1; HIF-1α: hypoxia-inducible factor-1 alpha; LW: Lobund-Wistar; MSeA: methylseleninic acid; MTT: 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; PAIII: prostate adenocarcinoma cells; PHD: prolyl hydroxylase; VEGF: vascular endothelial growth factor

Material and Methods

In vivo growth of PAIII cells

All animal studies were approved by the University of Notre Dame Institutional Animal Care and Use Committee and were conducted in a facility accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International. PAIII cells were derived from a freshly isolated subcutaneous tumor growing in LW rat. These cells were treated with MSeA (5 and 10 μM) and vehicle (saline) control for 90 min ex vivo in DMEM high glucose medium at 37°C. After two washes in sterile phosphate-buffered saline (PBS), a trypan blue exclusion test was performed, and more than 99% cells were found to be viable at that time. An equal number of viable cells (1 × 106) for each treatment were injected subcutaneously into the right flank of 2-month-old LW rats (250-g body weight) in 0.2 ml DMEM. Eight rats were injected for each ex vivo treatment. Twenty-eight days later, all rats were sacrificed, and the tumors were excised and weighed.

Lung foci experiment

One million PAIII cells were injected (i.v) through the tail vein into two groups of five (2-month-old) LW rats each. After 2 weeks of rest period, MSeA (1.5 mg/kg/day) was administered by oral gavage to one group of rats for 5 days a week. The control group of rats received an equivalent volume of normal saline. Two weeks later, all rats were sacrificed, and the number of subpleural pulmonary metastatic foci on surface of all lungs was assessed macroscopically.

Cell lines and treatments

PAIII cells were maintained in DMEM high glucose medium (Invitrogen, Carlsbad, CA); PC-3 cells (obtained from ATCC) were grown in F-12K medium (ATCC, Manassas, VA) and PC-3M cells (obtained from Dr. Fidler at M.D. Anderson Cancer Center, Houston, TX) and DU145 cells (obtained from ATCC) were maintained in minimum essential medium (ATCC, Manassas, VA). All the media were supplemented with 10% FBS and 1% penicillin–streptomycin solution. All the cell types were routinely passaged weekly and maintained in 5% CO2 at 37°C. Before treatments, cells were incubated with 0.1% FBS overnight. Next day, cells were either replenished with fresh medium containing 0.1 or 10% FBS along with MSeA and other agents (1 mM DMOG, 10 μM MG132, 50–200 μM selenomethionine, 100–200 μM Se-methylselenocysteine and 2.5–5 μM selenite). For normoxia conditions, cells were treated in incubators, accessible to ambient air (∼21% O2 levels). For experiments performed under hypoxic conditions, cells were placed in MIC-101 Modular Incubator Hypoxia Chamber (Billups-Rothenberg, Del Mar, CA) and perfused with a premixed standard gas containing 1% oxygen, 5% CO2 and 94% nitrogen (GTS-WELCO, Reading, PA). The airspace within the chamber was perfused twice for 4 min at a flow rate of 20–25 l/min at 2 psi with the premixed gas, and the sealed chamber was placed in a 37°C incubator.

Cell growth inhibition (MTT assay)

Rat and human prostate cancer cells were plated at a density of 10,000 cells/well for 24 hr in 96 well plates. The following day, the cells were incubated with increasing doses of MSeA for 24 hr. Untreated cells served as controls. 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT, 50 μg/well) was added for 4 hr at 37°C in phenol-free DMEM plain medium in the dark. The MTT was aspirated, DMSO solution was added to dissolve the crystals and absorbance was read at 570 nm.35

DNA-binding assay for HIF-1α

After treatment of cells grown under normoxic and hypoxic conditions with MSeA, samples were collected by scraping and washed in 1× PBS that was supplemented with phosphatase inhibitors provided with the nuclear extraction kit (Cayman Chemicals, Ann Arbor, MI). Briefly, the pelleted cells were suspended in hypotonic buffer and incubated for 15 min on ice to swell. A 10% NP40 solution was added to lyse cells. Cytosolic fraction was removed after 30 sec of high-speed centrifugation at 14,000g. The nuclear pellet was resuspended in ice-cold extraction buffer containing protease and phosphatase inhibitors, and tubes were vortexed for 15 sec. Tubes were placed on a shaker, rocked for 15 min and then centrifuged at 14,000g for 10 min at 4°C. Protein content in the cytosolic supernatant and pelleted nuclear fractions was measured by Coomassie Plus (Bradford) Protein Assay Kit (Thermo Scientific, Rockford, IL). HIF-1α DNA-binding activity in the nuclear fractions was performed using transcription factor assay kit (Cayman Chemical, Ann Arbor, MI) following the manufacturer's instructions. Briefly, equal amounts of nuclear fractions (30 μg) from MSeA-treated cells were reacted with precoated consensus dsDNA sequence containing the HIF-1α response element (HRE) in 96-well plate for 1 hr at room temperature. After washing five times with wash buffer, HIF-1α antibody (1:100) was added in binding buffer for 1 hr at room temperature. Then, the plate was washed five times with the wash buffer, and 1:00 dilution of goat anti-rabbit HRP-conjugated antibody was incubated in the wells for 1 hr at room temperature. After another set of five washes in wash buffer, 100 μl of developing solution was added with agitation for 30 min. A 100 μl of stopping solution per well was added, and the plate was read at 450 nm in a SPECTRAmax® PLUS384 plate reader (Molecular Devices Corporation, Sunnyvale, CA). The HIF-1α DNA-binding activity was performed in duplicate for each concentration of MSeA.

Immunoblotting

Cells were harvested by scraping and washed with PBS containing protease inhibitors. Nuclear and cytosolic proteins were extracted as described above. Equal amounts of protein (50 μg) were separated on 8 or 10% SDS-PAGE gels depending upon the protein being studied and transferred to PVDF membranes. Primary antibodies against cleaved PARP (Cell Signaling Technology, Danvers, MA), HIF-1α (R&D Systems, Minneapolis, MN), HIF-1β (Cell Signaling, Danvers, MA), Lamin B and β-actin (Santa Cruz Biotechnology, Santa Cruz, CA) were reacted at 1:1,000 with blots. Anti-mouse and anti-rabbit secondary antibodies (Cell Signaling, Danvers, MA) were used at a dilution of 1:3,000. Band expressions were developed using Pierce ECL reagents (Thermo Scientific, Rockford, IL), and densities were analyzed using VisionWorks™ software (UVP, Upland, CA). All Western blotting experiments were repeated two to three times.

HRE luciferase reporter assay

PC-3M cells were plated in 96-well plates in 10% FBS. The next day, medium was changed to 0.1% serum medium. Cells in each well were cotransfected with 0.2 μg of HRE-expressing plasmid p2.136 and 0.02 μg of pSV-renilla using Lipofectamine 2000 transfection reagent (Invitrogen, Carlsbad, CA). After 18-hr incubation, medium was replaced with medium containing 10% serum. The next day, cells were treated with varying doses of MSeA for overnight (16 hr) in the hypoxic chamber. After treatment, cells were incubated in passive lysis buffer (Promega, Madison, WI) at room temperature for 10 min. Luciferase assays were performed using dual luciferase system (Promega, Madison, WI), and numerical ratio of relative light units (RLU) from firefly luciferase activity was normalized to RLU from renilla luciferase. Transfections were performed in triplicate.

Quantitative real-time PCR

Total RNA was isolated from cells treated with MSeA (2.5–20 μM) using TRIzol reagent (Gibco BRL, Rockville, MD) according to the manufacturer's instructions. Isolated RNA was dissolved in RNase-free water. cDNA synthesis was performed with the Superscript™ First Strand Synthesis System (Invitrogen, Carlsbad, CA) according to the manufacturer's instructions using oligo(dT) as the 3′ primer. PCR was performed using the RT2 SYBR Green Master Mix (Superarray Bioscience Corporation, Frederick, MD). QuantiTect Primers specific for VEGF, GLUT1, HIF-1α and GAPDH Qiagen (Valencia, CA) were used at a final concentration of 100 nM in 25 μl PCR reactions. cDNA negative controls were run for each target gene. GAPDH expression determined for each sample was used to normalize expression of the target genes. Relative expressions were depicted as percent of the normalized untreated control. For VEGF, GLUT1 and HIF-1α, thermocycling conditions were initiated with a 10-min 95°C activation step followed by 40 cycles of 94°C for 15 sec, 56°C for 30 sec and 72°C for 30 sec. For GAPDH, thermocycling conditions were initiated with a 10-min 95°C activation step followed by 40 cycles of 95°C for 15 sec, 62°C for 30 sec and 72°C for 45 sec. Reactions were run in duplicate, and experiments were repeated three times. Relative expressions were calculated using the ΔΔCt method.

VEGF ELISA

After treatments with MSeA (2.5–10 μM), spent medium was collected to measure VEGF levels using the Quantikine kit (R&D Systems, Minneapolis, MN) according to the manufacturer's instructions.

Statistical analysis

An analysis of variance was performed to compare tumor weights and lung foci within the MSeA and saline vehicle treatment groups. Student's t-test was used to compare the vehicle-treated cells with various concentrations of MSeA in all the in vitro experiments including MTT assay, VEGF and HIF-1α DNA-binding ELISAs. p < 0.05 was considered statistically significant for these parameters.

Results

MSeA inhibits growth of PAIII tumors and metastatic lung foci in LW rats

The rats were observed by a board-certified veterinarian on a daily basis for these experiments, and no differences in body weight, appetite or physical appearance of animals were noted during or at the end of the experiment. A significant (p < 0.001) reduction in tumor weights of PAIII cells treated ex vivo with MSeA was observed in a dose-dependent manner (Fig. 1a). Furthermore, an oral administration of 1.5 mg MSeA/kg/day for 2 weeks showed a significant (p < 0.005) drop in PAIII-induced metastatic lung foci in LW rats when compared to normal saline-fed controls (Fig. 1b). These results demonstrated efficacy of MSeA against invasive prostate cancer in the rodent model.

Figure 1.

(a) Growth inhibition of MSeA-treated PAIII cells in LW rats. PAIII cells were pretreated with increasing doses of MSeA (ex vivo) and injected s.c into LW rats and tumor weights measured 28 days later as described in Material and Methods. Data show a significant reduction in tumor weights of MSeA-treated PAIII cells in a dose-dependent manner (N = 8, p < 0.001). (b) Reduced subpleural pulmonary metastatic foci in MSeA-treated LW rats. PAIII cells were injected i.v through tail vain in LW rats. After 14 days, MSeA at a dose of 1.5 mg/kg/day was administered by oral gavage for 2 weeks. Control rats received normal saline. Data show that number of lung foci in the MSeA-treated group was significantly reduced (N = 5, p < 0.005).

MSeA inhibits growth and induces apoptosis in normoxia and hypoxia

Prostate tumors in LW rats have hypoxic regions that cause clonal expansion of cells that eventually develop resistance to several anticancer agents. A goal of our study was to examine whether PAIII cells derived from these hypoxic regions will be sensitive to MSeA-induced growth inhibition and cell death under normoxic and hypoxic conditions. PAIII cells were treated with increasing concentrations of MSeA for 24 hr, and significant growth inhibition was observed in a dose-dependent manner (Fig. 2a) in both normoxia and hypoxia. The IC50 for MSeA was less than 5 μM for both conditions. These cells subsequently initiate apoptosis as demonstrated by increased levels of cleaved PARP in a dose-dependent fashion in normoxia (Fig. 2b). By contrast, under hypoxic conditions, the invasive human prostate cancer cells were more responsive to effects of MSeA in a dose-dependent manner (Fig. 2a). In a similar fashion, after treatments with MSeA (5–20 μM) under both normoxic and hypoxic conditions, the human prostate cancer cell lines (PC-3, DU145 and PC-3M) initiated apoptosis as observed by cleaved-PARP analysis (Fig. 2c), and the effect was more robust under hypoxic conditions.

Figure 2.

Inhibition of prostate cancer cells in vitro by MSeA. (a) Invasive rodent and human prostate cancer cells were treated with MSeA (2.5–20 μM) for 24 hr either in normoxia (N, 21% O2) or in hypoxia (H, 1% O2) conditions as described in Material and Methods. All the prostate cancer cell types are growth inhibited (MTT assay) in a dose-dependent manner, *p < 0.05. (b) PAIII cells undergo apoptosis (cleaved PARP) in a dose-dependent manner. (c) PC-3, DU145 and PC-3M cells in hypoxia seemed to be more sensitive to MSeA-induced apoptosis when compared to cells treated in normoxia.

MSeA inhibits HIF-1α protein expression

A time-course experiment revealed that HIF-1α in PAIII cells was stabilized within 2 hr of hypoxia when compared to normoxia (Fig. 3a, upper panel), and this status of HIF-1α was maintained throughout the hypoxic exposure. Of interest in both PC-3 cells (Fig. 3a, lower panel) and in PC-3M cells (data not shown), HIF-1α levels appeared to be constitutively elevated in normoxia, and no change was observed during hypoxia. After 2-hr MSeA treatment, PAIII cells in a dose-dependent fashion reduced the levels of HIF-1α protein (Fig. 3b); however, no change was observed in HIF-1β expression.

Figure 3.

HIF-1α expression in PAIII, PC-3 and PC-3M cells and its modulation by MSeA in the presence of growth factors and varying oxygen levels. (a) PAIII and PC-3 cells were incubated in hypoxia (1% O2) for time durations shown. Elevation in HIF-1α expression was noted in PAIII cells after 1 hr of hypoxia, whereas in PC-3 cells HIF-1α was activated even in normoxia (21% O2). (b) After a 2-hr MSeA treatment of PAIII cells, HIF-1α levels were reduced dose-dependently without affecting HIF-1β levels. Lamin B was measured as the loading control for the nuclear fractions. (c) PC-3M cells were stimulated with 10% FBS following an overnight starvation (0.1% FBS) in the presence and absence of MSeA in both normoxia and hypoxia. After a 2-hr MSeA treatment of PC-3M cells inhibits both growth factor (FBS) and hypoxia-induced HIF-1α protein expression.

In PC-3M cells, MSeA dose-dependently downregulated HIF-1α protein levels in normoxia (Fig. 3c, upper panel). Similarly in hypoxia, MSeA downregulated HIF-1α protein expression with and without serum stimulation (Fig. 3c, lower panel). However, in the presence of 10% FBS, the PC-3M cells were more resistant to 5 μM MSeA-mediated decrease in HIF-1α in hypoxia. There was no appreciable change in HIF-1β levels after MSeA treatments.

MSeA inhibits HIF-1α transcription

HIF-1α is an important transcription factor that binds to a number of genes via the HRE and activates their transcription.37 We investigated the effect of MSeA on activity of HIF-1α by studying DNA binding using ELISA and using HRE luciferase reporter assay. Using an ELISA-based assay, PAIII cells showed a two-fold increase in DNA-binding activity of HIF-1α after a 2-hr exposure to hypoxia when compared to normoxia (Fig. 4a, left panel). Furthermore, MSeA treatment was able to significantly decrease the DNA-binding activity of HIF-1α in a dose-dependent manner in these cells. While in PC-3 cells, the DNA-binding activity was higher in normoxia when compared to cells in hypoxia. In addition, MSeA significantly reduced the DNA-binding activity of HIF-1α in these cells in a dose-dependent manner under both normoxic and hypoxic conditions (Fig. 4a, right panel).

Figure 4.

(a) DNA-binding activity of HIF-1α in PAIII and PC-3 cells. Cells were treated with increasing concentrations of MSeA in normoxia and hypoxia conditions for 2 hr, and nuclear extracts were assayed for DNA-binding activity as described in Material and Methods. PC-3 cells displayed greater HIF-1α DNA-binding activity both in normoxia and hypoxia when compared to PAIII cells. MSeA was able to reduce the DNA-binding activity in a dose-dependent manner, *p < 0.05, **p < 0.01. (b) HRE reporter gene activity in PC-3M cells. Cells transfected with HRE-2.1 and RLTK plasmids were treated with MSeA, selenomethionine, Se-methylselenocysteine and selenite for 16 hr in hypoxia. There was a significant reduction after 10-μM MSeA treatment, *p < 0.05.

An overnight treatment of luciferase-expressing PC-3M cells with MSeA demonstrated a dose-dependent reduction in luminescence (relative RLU) when compared to untreated controls in hypoxia (Fig. 4b). Other forms of selenium such as selenomethionine, Se-methylselenocysteine and sodium selenite, however, did not show any significant changes in the HRE reporter activity in PC-3M cells when compared to untreated controls.

MSeA inhibits downstream targets of HIF-1α

As MSeA demonstrated a clear inhibitory effect on HIF-1α protein expression, its DNA-binding activity and the HRE reporter gene activity, we next examined whether expression of downstream targets of HIF-1α was reduced. Initial studies tested the relative mRNA expression of HIF1α, VEGF and GLUT1 in PAIII cells treated with increasing doses of MSeA in hypoxia using real-time PCR. Figure 5a shows a dose-dependent decrease in HIF-1α and its downstream targets, VEGF and GLUT1, in a single experiment. On repeating the experiment in triplicate, the results indicated that MSeA reduces the expression of VEGF mRNA in a dose-dependent manner (Fig. 5b; p < 0.05 for 10 μM dose). Similarly, when PC3 and PC-3M cells were treated with MSeA under both normoxic and hypoxic conditions for 24 hr, the secreted VEGF levels were reduced significantly (p < 0.05) after MSeA treatments (Fig. 5c).

Figure 5.

MSeA influences HIF-1α and its downstream targets in prostate cancer cells. Cells were treated with increasing doses of MSeA (2.5–10 μM) for 24 hr in hypoxia or in normoxia. (a) Real-time PCR from a single experiment showed that MSeA reduces HIF-1α, VEGF and GLUT1 mRNA expression levels in PAIII cells under hypoxia. (b) Upon repeating the experiment in triplicate, MSeA was found to reduce relative VEGF mRNA expression levels in PAIII cells in a dose-dependent manner under hypoxic conditions, *p < 0.05. (c) All the MSeA treatments (5–20 μM) significantly reduced secretory VEGF levels (p < 0.05) when compared to respective controls in both PC-3 and PC-3M cells under normoxia (N) and hypoxia (H).

MSeA influences PHD activity

We next investigated whether MSeA acted by regulating hydroxylation of the specific prolyl moiety (P564) by the oxygen-sensing enzyme, HIF-α PHD, and thus affected HIF1-α degradation through ubiquitination. To simulate hypoxia under normoxic conditions, we inhibited PHD activity using two distinct methods, i.e., exposure of cells to cobalt chloride (CoCl2), a hypoxia mimetic, or treatment with dimethyloxallyl glycine (DMOG), a competitive inhibitor of PHD. In brief, CoCl2 induces the hypoxia response pathway possibly by inhibiting PHD through a competitive occupancy of the ferrous ion in the enzyme active site, inhibiting hydroxylation and thus stabilizing HIF1-α. In a similar manner, DMOG inhibits PHD by competitive interaction with α-ketoglutarate, another of the cofactors required by PHD for activity.

When cells were pretreated with 10 μM MSeA followed by incubation with 100 μM CoCl2, HIF-1α DNA binding was reduced compared to cells first treated with CoCl2 and then with MSeA. The latter treatment demonstrated no change in DNA binding (Fig. 6a). When cells were coadministered 10 μM MSeA and 100 μM CoCl2, MSeA-treated cells exhibited downregulation of HIF-1α expression (Fig. 6b). To determine whether MSeA directly influenced PHD activity in PC-3M cells, cultures were treated with 1 mM DMOG in the presence of 10 μM MSeA. HIF-1α expression did not change, suggesting that MSeA may be influencing hydroxylation of HIF-1α in prostate cancer cells and possibly causing degradation of HIF-1α (Fig. 6c). In a separate experiment, a 2-hr pretreatment with 10 μM MG132, a proteasome inhibitor, followed by 2 hr of MSeA (5 and 10 μM) treatment of PC-3M cells showed an accumulation of HIF-1α in nuclear fraction in a dose-dependent manner (Fig. 6d), further strengthening our finding that MSeA may be enhancing PHD activity in these invasive prostate cancer cells.

Figure 6.

MSeA may influence PHD activity in prostate cancer cells. (a) PAIII cells were treated with 100 μM CoCl2 in the presence or absence of 2.5 and 5 μM MSeA. Nuclear fractions were subjected to HIF-1α DNA-binding activity. Less of HIF-1α gets stabilized in nucleus if cells are pretreated MSeA. (b) Two-hour treatment of 10 μM MSeA reduced the HIF-1α levels in 100 μM CoCl2-treated PC-3M cells. (c) PC-3M Cells were treated with 1 mM DMOG and incubated in the absence or presence of 10 μM MSeA for 2 hr. MSeA seemed to affect the HIF-1α by possibly interacting at the PHD domains. (d) PC-3M cells treated with 10 μM MG132 and incubated in the presence or absence of 5 and 10 μM MSeA in hypoxia showed a dose-dependent increase in HIF-1α expression. *p < 0.05 compared to control.

Discussion

Our study demonstrates that MSeA inhibits the expression and activity of HIF-1α in invasive rat and human prostate cancer cells. MSeA has been shown to be an effective anticancer agent in several in vitro and in vivo prostate cancer models and is known to induce growth arrest and apoptosis.29, 38, 39 In our study, ex vivo MSeA treatment of PAIII cells resulted in significant reduction in the tumor weights in LW rats. More importantly, upon oral MSeA treatment, the number of metastatic lung foci was significantly reduced. These in vivo data demonstrate, for the first time, that MSeA can potentially reduce the metastatic spread of prostate cancer cells to the lungs.

Treatment of highly aggressive human prostate cancer cell lines (PC-3 and PC-3M) with MSeA resulted in a significant growth inhibition and induction of apoptosis. Our studies show that MSeA treatment produces greater robust effects in cells under hypoxic than under normoxic conditions especially at a physiological dose of 5 μM. Our data suggest that MSeA-induced apoptosis in hypoxia is not unique to PTEN-mutant (PC-3 and PC-3M) or PTEN-positive cells (PAIII and DU145). These observations collectively point to the effectiveness of MSeA against invasive prostate cancer growth that occurs under conditions of hypoxia. These results may be translatable to clinical treatment modalities because hypoxic microenvironments within solid tumors correlate with tumor invasiveness, metastasis and resistance to drug and radiation treatments.40

HIF-1α protein is overexpressed in primary and metastatic human prostate cancer as well as other common human cancers.41 Hypoxia induces a rapid increased stabilization and expression of HIF-1α protein resulting in enhanced transcriptional activity.9 The genes encoded by HIF-1α are highly supportive of a prosurvival environment and thus offer cytoprotection to aggressive tumor cells.10 In our study, MSeA treatment during hypoxia decreased HIF-1α protein expression in PAIII and PC-3M cells in a dose-dependent manner. However, PC-3M cells in the presence of serum seemed more resistant to MSeA treatment in hypoxia, suggesting that growth factor (serum)-induced signals such as PI3K, IGF-1 or EGFR may convey partial resistance. It is well established that HIF-1α is induced both by hypoxia and PI3K pathway in several cell types.20, 42, 43 Even though several reports including ours31, 44 showed that selenium compounds could influence PI3K/Akt pathway in various cancer cell lines, this interaction has yet to be addressed under hypoxic conditions.

DNA binding of HIF-1α is significantly reduced by MSeA in PAIII and PC-3 cells both in normoxia and hypoxia. Additionally, HRE activity during hypoxia in PC-3M cells is decreased after MSeA treatment. Other selenium compounds examined in our study, including selenomethionine, Se-methylselenocysteine and selenite, did not show significant changes in the HRE activity, suggesting that MSeA, perhaps acting as a precursor to methylselenol, exhibits specificity to redox-sensitive proteins particularly under hypoxia in PC-3M cells. In addition, PC-3M cells may be expressing low or no β-lyase responsible for generating methylselenol from Se-methylselenocysteine; this remains to be determined.

MSeA treatment of prostate cancer cells also resulted in reduction of VEGF and GLUT 1 gene expression as well as secreted VEGF. Both VEGF and GLUT 1 are downstream targets of HIF-1α and play important roles in HIF-1α-induced cancer invasion.45 An earlier report showed antiangiogenic properties of MSeA mainly through downregulation of VEGF.46 That observation and our findings are most likely occurring via HIF-1α reduction after MSeA treatment.

Treatment of cells with CoCl2 mimics hypoxia under normoxic conditions and thus induces HIF-1α.47 When prostate cancer cells were either pretreated with MSeA followed by CoCl2 exposure or cotreated with MSeA and CoCl2, a noticeable decrease in HIF-1α binding or protein expression, respectively, was observed. MSeA is possibly competing with CoCl2 and keeping PHD active and in turn allowing more HIF-1α degradation. Furthermore, MSeA in the presence of DMOG is not able to downregulate HIF-1α in PC-3M cells, again suggesting that MSeA may be influencing hydroxylation of HIF-1α and causing its degradation. Chintala et al.48 observed that MSeA targets HIF-1α in FaDu cells, in particular through elevation in PHD2 protein. In our study, MSeA may possibly be increasing the degradation of HIF-1α in cytosol and therefore allowing less stabilization of HIF-1α in the nucleus; this would result in lower VEGF and GLUT1 expression in cells and may reduce their survival.

In summary, our results show that MSeA inhibits HIF-1α expression and its activity along with HIF-1α downstream targets in more than one invasive prostate cancer cell line. These data raise the possibility of using MSeA for targeting HIF-1α in invasive prostate cancer therapy after we have established that MSeA reduces HIF-1α in vivo; these experiments will be part of future studies.

Acknowledgements

The authors thank Dr. Gregg L. Semenza for providing HRE-2.1 and RLTK plasmids. They thank Dr. Sailendra N. Nichenametla, Department of Public Health Sciences, Penn State College of Medicine, Hershey, Pennsylvania, for assistance in HRE luciferase reporter assay. This study was supported in part by NCI (J.T.P. and R.S.) and George Laverty Foundation (R.S.).

Ancillary