Hypoxia upregulates the gene expression of mitochondrial aconitase in prostate carcinoma cells

in Journal of Molecular Endocrinology

Hypoxia induces metabolic alteration in cancer cells by stabilizing hypoxia-inducible factor 1α (HIF-1α (HIF1A)), which regulates the bioenergetic genes of glycolysis and lipid metabolic pathways. However, the target genes of hypoxia-induced metabolic alterations in the prostate remain uncertain. Mitochondrial aconitase (mACON) (ACONM) is an enzyme that is central to carbohydrate and energy metabolism and is responsible for the interconversion of citrate to isocitrate as part of the citric acid cycle in the human prostate. We evaluated the effects of the molecular mechanisms of hypoxia on mACON gene expression in PC-3 and LNCaP human prostate carcinoma cells. Immunoblotting assays revealed that hypoxia modulated mACON and lactate dehydrogenase A (LDHA) protein expression, while these effects were attenuated when HIF-1α was knocked down. Hypoxia induced fatty acid synthase (FASN) in PC-3 cells while hypoxia blocked FASN gene expression in LNCaP cells after 24-h incubation. Results of real-time RT-qPCR, immunoblotting, and transient gene expression assays revealed that hypoxia treatment or co-transfection with H IF-1α expression vector enhanced gene expression of mACON, implying that hypoxia modulated mACON at the transcriptional level. Hypoxia-induced mACON promoter activity is dependent on the DNA fragment located at −1013 to −842 upstream of the translation initiation site. l-mimosine, an iron chelator, stabilized HIF-1α but downregulated mACON gene expression, suggesting that iron chelation blocked the hypoxia-induced mACON gene expression. These results suggest that hypoxia dysregulates the expressions of LDHA, FASN, and mACON genes, and the hypoxia-induced mACON gene expression is via the HIF-1α-dependent and iron-dependent pathways in prostate carcinoma cells.

Abstract

Hypoxia induces metabolic alteration in cancer cells by stabilizing hypoxia-inducible factor 1α (HIF-1α (HIF1A)), which regulates the bioenergetic genes of glycolysis and lipid metabolic pathways. However, the target genes of hypoxia-induced metabolic alterations in the prostate remain uncertain. Mitochondrial aconitase (mACON) (ACONM) is an enzyme that is central to carbohydrate and energy metabolism and is responsible for the interconversion of citrate to isocitrate as part of the citric acid cycle in the human prostate. We evaluated the effects of the molecular mechanisms of hypoxia on mACON gene expression in PC-3 and LNCaP human prostate carcinoma cells. Immunoblotting assays revealed that hypoxia modulated mACON and lactate dehydrogenase A (LDHA) protein expression, while these effects were attenuated when HIF-1α was knocked down. Hypoxia induced fatty acid synthase (FASN) in PC-3 cells while hypoxia blocked FASN gene expression in LNCaP cells after 24-h incubation. Results of real-time RT-qPCR, immunoblotting, and transient gene expression assays revealed that hypoxia treatment or co-transfection with HIF-1α expression vector enhanced gene expression of mACON, implying that hypoxia modulated mACON at the transcriptional level. Hypoxia-induced mACON promoter activity is dependent on the DNA fragment located at −1013 to −842 upstream of the translation initiation site. l-mimosine, an iron chelator, stabilized HIF-1α but downregulated mACON gene expression, suggesting that iron chelation blocked the hypoxia-induced mACON gene expression. These results suggest that hypoxia dysregulates the expressions of LDHA, FASN, and mACON genes, and the hypoxia-induced mACON gene expression is via the HIF-1α-dependent and iron-dependent pathways in prostate carcinoma cells.

Introduction

Metabolic anomalies in altered forms of flux through key metabolic pathways are primary hallmarks of many malignant tumors including prostate tumors (Bertilsson et al. 2012, Costello & Franklin 2012). Most cancer cells fulfill their energy needs primarily via glycolysis, which is regulated by p53, c-Myc, and hypoxia-inducible factor 1α (HIF-1α (HIF1A); Yeung et al. 2008). Under certain hypoxic conditions, anaerobic glycolysis, instead of aerobic glycolysis, which increases lactate accumulation or lipid metabolism pathways or both, enhances fatty acid synthase (FASN) expression and provides metabolic advantages for the tumor cells (Hu et al. 2011). Tumor cells engaged in glycolysis do not convert their pyruvate completely into lactate when the availability of oxygen to the cell is insufficient to support cellular oxidative metabolism. Rather, a measurable fraction of the pyruvate metabolized in the tricarboxylic acid cycle provides energy and precursors for the biosynthetic pathway that consumes tricarboxylic acid cycle intermediates (DeBerardinis 2008). In prostate tissue, the low-citrate phenotype reflects a hypoxia-defense adaptation and citrate levels are important to defining the oncogenic condition (Costello & Franklin 2005). The malignant prostate cells are more O2 limited than in the normal phenotype and become even more O2 limited as their malignancy state advances (Hochachka et al. 2002). The hypoxia adaptation in the prostate alters HIF-1α levels and energy metabolism (Bourdeau-Heller & Oberley 2007, Higgins et al. 2009). However, the metabolic alternations in the prostate carcinoma cells induced by hypoxia are still not well delineated.

Mitochondrial aconitase (aconitase hydratase, EC4.2.1.3; mACON) is regarded as the key enzyme in citrate oxidation in human prostate epithelial cells and its abnormal expression is implicated in tumorigenesis of the prostate (Costello & Franklin 1994, Juang 2004a, Bertilsson et al. 2012). The mACON gene is located in chromosome 22q13.2 and its expression in the prostate is regulated by androgen, prolactin, cholesterol, iron, zinc, and p53 (Juang 2004b,c, Juang et al. 2004, Feng et al. 2005, Tsui et al. 2006,2011). Several researchers have suggested that hypoxia modulates the metabolic phenotype of human prostate carcinoma cells (Hochachka et al. 2002, Higgins et al. 2009); however, there is no study so far that discusses the modulation of hypoxia and its impact on mACON gene expression.

Fatty acid metabolism rather than glycolysis is reported as the energy fuel for prostate carcinoma cells (Liu 2006). The glycolysis–citrate–lipogenesis pathways in providing the synthetic and bioenergetic requirement are essential for the growth and proliferation of prostate tumor cells (Costello & Franklin 2012). FASN, a key regulator of de novo lipid biosynthesis, is highly overexpressed in prostate carcinoma cells and constitutes a therapeutic target in this disease (Baron et al. 2004). Previous studies of ectopic expression of FASN in prostate cells or transgenic expression of FASN in mice indicated that FASN acts as an oncogene of prostate cancer (Migita et al. 2009). However, the effect of hypoxia on gene expression of FASN in prostate cells is still not clearly understood.

It is well known that many human cancers have higher lactate dehydrogenase A (LDHA) levels than normal tissues, and the inhibition of LDHA blocks tumor progression (Le et al. 2010). The serum LDH level has been regarded as a prognostic factor for prostate cancer, and a high serum LDH level means a poor outcome for prostate cancer patients (Yamada et al. 2011). Hypoxia induces expression of LDHA, which converts pyruvate into lactate in several cell types (Semenza et al. 1996). Increased levels of LDH within the cells are believed to be largely due to the upregulation by the transcription factor HIF-1α, as an earlier study explored hypoxia response elements in the promoter of the LDHA gene (Semenza 2007). Nonetheless, no report has yet described the regulation of LDHA by hypoxia in prostate carcinoma cells.

The objectives of this study were to determine the regulatory effects of hypoxia on the expressions of LDHA, FASN, and mACON in human prostate carcinoma cells. The mechanism of the hypoxic effect on mACON gene expression was also investigated.

Materials and methods

Cell culture and chemicals

PZ-HPV-7, CA-HPV-10, LNCaP, PC-3, and DU145 cell lines were obtained from the Bioresource Collection and Research Center (BCRC, Taiwan) and were maintained as described previously (Tsui et al. 2012). Cells were incubated under normoxia using a standard CO2 incubator at 37 °C in a humidified atmosphere with 5% CO2 and 95% room air (21% O2) until the cells grew to 70–80% confluence in RPMI-1640 medium with 10% FCS. On the day of hypoxic treatment (1% O2, 5% CO2, and 94% N2), growth medium was switched to fresh medium with 10% FCS that had been pre-equilibrated in the hypoxic incubator (APM-30D, Astec, Fukuoka, Japan). Cells were then incubated in the hypoxic incubator for another 24 h. l-mimosine, chetomin, and ferric ammonium citrate were purchased from Sigma Chemical Company, FCS was purchased from HyClone (Logan, UT, USA), and RPMI-1640 medium was purchased from Gibco, Invitrogen Corporation.

Immunoblotting assay

Equal quantities of cell extract (40 μg) were loaded onto 12% SDS–PAGE and analyzed using an electrochemiluminescence detection system and were viewed using the ChemiGenius image capture system (Syngene, Cambridge, UK). Polyclonal mACON antiserum was prepared as described previously (Tsui et al. 2011). The blot membranes were probed with 1:500 HIF-1α (610958, BD Biosciences, San Jose, CA, USA), 1:3000 diluted β-actin antiserum (I-19, Santa Cruz Biotechnology), 1:100 diluted vascular endothelial growth factor (VEGF; A-20, Santa Cruz Biotechnology), 1:500 diluted LDHA (E-9, Santa Cruz Biotechnology), 1:1000 diluted FASN (A-5, Santa Cruz Biotechnology), 1:1000 diluted heme oxygenase 1 (HO-1; Hsp32, Stressgen, Victoria, BC, Canada), 1:100 p53 (DO-1, Santa Cruz Biotechnology), and 1:12 500 diluted mACON antiserum. The intensities of the different bands were analyzed using the GeneTools program of ChemiGenius (Syngene).

Lactate secretion assay

Cells were incubated under normoxic or hypoxic conditions in 0.5 ml RPMI medium with 10% FCS in a 24-well plate (1×105 cells/well) for 24 h. Following incubation, the supernatants of the conditioned media from each well were collected for lactate secretion assays using a lactate assay kit (K627-100, Bio Vision, Milpitas, CA, USA) according to the protocol described by the manufacturer. The lactate level in each sample was adjusted by the concentration of protein in the whole cell extract, which was measured using a bicinchoninic acid protein assay kit (Pierce Protein Research, Rockford, IL, USA).

HIF-1α expression vector

The human HIF- cDNA (MGC: 10483) vector was purchased from Invitrogen. Human HIF- cDNA was linearized by cutting with BamHI and XbaI and ligated with the overexpression vector pcDNA3 (Invitrogen) as described previously (Chung et al. 2011).

Knockdown HIF-1α

LNCaP and PC-3 cells were transduced with HIF- shRNA lentiviral particles (sc-35561-V, Santa Cruz Biotechnology) as described previously (Chung et al. 2012). The cells (LN-HIF-1αsi and PC-HIF-1αsi) were selected by incubation with 10 μg/ml puromycin dihydrochloride for at least five generations. The mock-transfected LNCaP (LN-COLsi) and PC-3 (PC-COLsi) cells were transduced with control shRNA lentiviral particles (sc-10808-V, Santa Cruz Biotechnology) and were clonally selected in the same manner as the HIF-1α knockdown cells.

Real-time RT-PCR

Total RNA was isolated using the Trizol reagent, and cDNA was synthesized using the Superscript III preamplification system (Invitrogen). Real-time PCR (qPCR) was performed as described previously (Chung et al. 2012). FAM dye-labeled TaqMan MGB probes and PCR primers were purchased for human mACON (HS00426616_g1) from Applied Biosystems. For the internal positive control, β-actin (HS016060665_g1) was used with a FAM reporter dye-labeled TaqMan MGB probe. The amplification conditions were as follows: 40 cycles at 95 °C for 15 s and 60 °C for 1 min. Mean cycle threshold (Ct) values for mACON were normalized against the β-actin control probe to calculate ΔCt values using StepOne software v2.0 (Applied Biosystems). All reactions were preformed in triplicate, and each experiment was conducted on at least three independent occasions.

Reporter vectors, transient transfection, and reporter assay

The reporter vectors containing the 5′-flanking region of the mACON gene were constructed as described previously (Tsui et al. 2011). LNCaP cells or PC-3 cells were plated onto 24-well plates 1 day before transfection. Cells were transiently transfected using TransFast transfection reagent (Promega BioScience) as described previously (Tsui et al. 2012). The luciferase activity was adjusted for transfection efficiency using the normalization control plasmid pCMVSPORTβgal (Invitrogen). For the transient co-transfection of HIF-1α experiments, cells were transfected with the same amount of plasmid in each well by adding pcDNA3 vector to eliminate the variable degrees of efficiency of reporter activities.

Statistical analysis

Results are expressed as the means±s.e.m. of at least three independent replications of each experiment. Statistical significance was determined using Student's t-test and one-way ANOVA using the SigmaStat program for Windows, version 2.03 (SPSS, Inc.).

Results

The immunoblotting assays revealed that hypoxia, in comparison to normoxia, significantly upregulated the protein levels of HIF-1α in both LNCaP and PC-3 cells (Figs 1A and 2A). Results also showed that hypoxia upregulated mACON, LDHA, HO-1 (HMOX1), and VEGF (VEGFA) protein levels (Fig. 1A). Similar results were found in the hypoxia-treated PC-3 cells (Fig. 2A). The quantitative analysis was done by determining the intensity of each band for different genes and β-actin from three independent experiments (Figs 1B and 2B). Interestingly, the quantitative results indicated that FASN levels were downregulated in LNCaP cells by 39% after 24 h of hypoxic culture in comparison with LNCaP cells incubated under normoxic condition (Fig. 1B). However, we did not find a similar phenomenon in the PC-3 cells. Hypoxic treatment significantly increased FASN levels (1.53-fold) in PC-3 cells (Fig. 2B).

Figure 1
Figure 1

Hypoxia modulates protein expression of mACON, FASN, and LDHA in LNCaP cells. (A) LNCaP cells were cultured under hypoxic or normoxic conditions for 24 h. The expression of HIF-1α, mACON, FASN, LDHA, HO-1, VEGF, and β-actin were determined by immunoblotting assay. (B) The quantitative analysis was done by determining the intensity of each band for target genes and β-actin from three independent experiments. Data are presented as the fold-induction (±s.e.m., n=3) of the relative density of the target gene/β-actin in relation to the normoxia treatment group. (C) Mitochondria were extracted and the enzymatic activity of mACON was analyzed in the LNCaP cells. Data are presented as the mean percentage (±s.e.m., n=6) of mACON enzymatic activity relative to that of the normoxia-treated group. (D) Supernatants of the conditioned medium were collected and the lactate levels were analyzed in the LNCaP cells. Data are presented as the mean percentage (±s.e.m., n=6) of lactate secretion relative to that of the normoxia-treated cell group; +P<0.01.

Citation: Journal of Molecular Endocrinology 51, 1; 10.1530/JME-13-0090

Figure 2
Figure 2

Hypoxia modulates protein expression of mACON, FASN, and LDHA in PC-3 cells. (A) PC-3 cells were treated and then the expression of proteins were analyzed as described in Fig. 1A. (B) The quantitative analysis and data presentation are described as in Fig. 1B (*P<0.05,+P<0.01). (C) Mitochondria were extracted and the enzymatic activity of mACON was analyzed in the PC-3 cells. (D) Supernatants of the conditioned medium were collected and the lactate levels were analyzed in the PC-3 cells. Data are presented as the mean percentage (±s.e.m., n=6) of mACON enzymatic activities and lactate secretion relative to that of the normoxia-treated cell group (+P<0.01).

Citation: Journal of Molecular Endocrinology 51, 1; 10.1530/JME-13-0090

Because we found that hypoxia upregulated mACON protein levels in both LNCaP and PC-3 cells, we sought to determine whether its enzymatic activities were induced by hypoxia. Results revealed that hypoxia, in comparison with normoxia, significantly upregulated 1.41- and 1.44-fold mACON enzymatic activities respectively in LNCaP and PC-3 cells (Figs 1C and 2C). We further determined the cytosolic aconitase (cACON) activities in the cells. However, results showed that hypoxia has no significant enzymatic effects on cACON in either LNCaP (P=0.148) or PC-3 (P=0.213) cells (Supplementary Figure S1, see section on supplementary data given at the end of this article).

We analyzed the lactate levels in the supernatant of the culture media after hypoxic or normoxic culture for 24 h. Results showed that hypoxic treatment enhanced lactate secretion in both LNCaP (2.34-fold) and PC-3 (2.11-fold) cells after 24 h of hypoxic culture in comparison with PC-3 cells incubated under normoxia (Figs 1D and 2D).

As we found the divergent effect of hypoxia on the FASN expression in LNCaP and PC-3 cells, we continued to determine the effects of hypoxia on FASN, mACON, LDHA, and HIF-1α protein levels in other prostate cell lines including PZ-HPV-7, CA-HPV-10, and DU145 cells. The results of immunoblotting assays revealed that hypoxia significantly upregulated protein levels of HIF-1α, FASN, mACON, and LDHA in PZ-HPV-7 and CA-HPV-10 cells; however, hypoxia downregulated minimally but significantly protein expressions of FASN and mACON while upregulated HIF-1α and LDHA protein expressions on DU145 cells (Fig. 3).

Figure 3
Figure 3

Hypoxia modulates protein expression of mACON, FASN, and LDHA in PZ-HPV-7, CA-HPV-10, and DU145 cells. Cells were cultured under hypoxic or normoxic conditions for 24 h. The expression of HIF-1α, FASN, LDHA, mACON, and β-actin were determined by immunoblotting assay. (B) The quantitative analysis and data presentation are described as in Fig. 1B (*P<0.05,+P<0.01).

Citation: Journal of Molecular Endocrinology 51, 1; 10.1530/JME-13-0090

In order to determine that the hypoxic dysregulation of mACON, LDHA, VEGF, and FASN in prostate carcinoma cells was HIF-1α dependent, we knocked down HIF-1α in LNCaP (LN-HIF-1αsi) and PC-3 (PC-HIF-1αsi) cells by transduction with HIF- shRNA lentiviral particles. The immunoblotting assays revealed that hypoxia, in comparison with normoxia, significantly upregulated (P<0.01) protein levels of HIF-1α, mACON, LDHA, and VEGF in mock-transfected LNCaP (LN-COLsi) and PC-3 (PC-COLsi) cells. However, the increase in protein expressions of the genes under hypoxia was significantly attenuated (P<0.01) when the HIF-1α gene was knocked down in the LNCaP (LN-HIF-1αsi) and PC-3 (PC-HIF-1αsi) cells (Fig. 4A and C). The quantitative analysis was done by determining the intensity of each band for different genes and β-actin from three independent experiments (Fig. 4B and D). As expected, in comparison with normoxia, hypoxia significantly downregulated (P<0.01) protein levels of FASN in the mock-transfected LNCaP (LN-COLsi) cells but upregulated protein levels of FASN in the mock-transfected PC-3 (PC-COLsi) cells. Moreover, dysregulation of FASN protein levels was attenuated when the HIF-1α gene was knocked down from both (LN-HIF-1αsi and PC-HIF-1αsi) cells.

Figure 4
Figure 4

HIF-1α knockdown attenuates the activation of hypoxia on gene expression of mACON in prostate carcinoma cells. The mock-knockdown LNCaP (LN-COLsi) and the HIF-1α-knockdown LNCaP cells (LN-HIF-1αsi) (A) and the mock-knockdown PC-3 (PC-COLsi) and HIF-1α-knockdown PC-3 cells (PC-HIF-1αsi) (C) were cultured under normoxic (N) or hypoxic (HP) conditions for 24 h. The expressions of HIF-1α, mACON, FASN, LDHA, VEGF, and β-actin were determined by immunoblotting assay. Quantitative results were analyzed by determining the intensity of each band for the target genes (white bars, FASN; black bars, mACON; gray bars, VEGF; white coarse bars, LDHA, and gray cyclone bars, HIF-1α) and β-actin (B and D). Data are presented as the fold-induction (±s.e.m., n=3) of the relative density of the target gene/β-actin in relation to the mock-knockdown normoxia-treated group (*,#,+P<0.01).

Citation: Journal of Molecular Endocrinology 51, 1; 10.1530/JME-13-0090

As our results showed that hypoxic treatment upregulated mACON protein levels in both LNCaP and PC-3 cells, we determined the effects of the hypoxia mimetic, l-mimosine, on gene expression of mACON. In contrast to the results of hypoxic treatments, l-mimosine treatment downregulated mACON protein levels in LNCaP cells even though l-mimosine indeed stabilized HIF-1α (Fig. 5A). Interestingly, mACON protein levels were upregulated to 1.55-fold under hypoxia in comparison with normoxia. The quantitative results indicated that 800 μM l-mimosine induced a 1.89-fold increase in HIF-1α protein levels but downregulated mACON levels by 25% after 24 h of incubation in comparison with solvent control treatment groups (Fig. 5B). Results of transient gene expression assays revealed that 200, 400, and 800 μM l-mimosine downregulated the promoter activity of the human mACON gene by 20.1, 24, and 20% respectively (Fig. 5C). The immunoblotting assays showed that l-mimosine did not affect P53 (TP53) expression in LNCaP cells, indicating that decreasing mACON gene expression by l-mimosine is not dependent on the p53 pathway (Fig. 5D).

Figure 5
Figure 5

l-mimosine downregulates the expression of human mACON gene via iron chelation. (A) Immunoblots of HIF-1α, mACON, and β-actin using lysates from varying concentrations of l-mimosine-treated LNCaP cells. (B) The quantitative analysis and data presentation are described as in Fig. 1B (*P<0.05,+P<0.01). (C) The mACON reporter vector-transfected LNCaP cells were treated with varying concentrations of l-mimosine as indicated for 24 h. Data are presented as the mean percentage±s.e.m. (n=6) of the luciferase activity induced by l-mimosine relative to the control solvent group (*P<0.05). (D) Immunoblotting for P53 and β-actin using lysates from varying concentrations of l-mimosine-treated LNCaP cells. (E) Immunoblotting for HIF-1α, mACON, and β-actin using lysates from varying concentrations of l-mimosine-treated PC-3 cells. (F) PC-3 cells were treated with 400 μM l-mimosine and/or 100 μg/ml ferric ammonium citrate for 24 h. The expressions of mACON and β-actin were determined by immunoblotting assays. (G) The mACON reporter vector-transfected PC-3 cells were treated with l-mimosine (400 μM) and/or ferric ammonium citrate (100 μg/ml) for 24 h. Data are presented as the mean percentage±s.e.m. (n=6) of the luciferase activity in relation to the control solvent-treated group (*P<0.05; C, control solvent; L, l-mimosine; F, ferric ammonium citrate).

Citation: Journal of Molecular Endocrinology 51, 1; 10.1530/JME-13-0090

Immunoblotting results also showed that l-mimosine treatment upregulated HIF-1α but downregulated mACON protein levels in PC-3 cells (Fig. 5E). We determined whether blocking of mACON gene expression by l-mimosine was due to the iron chelation nature of l-mimosine as mACON is an iron-upregulated gene. The immunoblotting assays revealed that ferric ammonium citrate, an iron donor, increased mACON gene expression, and this increase was blocked after co-treatment with l-mimosine in PC-3 cells (Fig. 5F). Similar results were found in the transient gene expression assays, which indicated that l-mimosine blocked the ferric ammonium citrate-induced increases in mACON promoter activity in PC-3 cells (Fig. 5G).

The RT-qPCR analyses showed that expression of the mACON gene was significantly induced by hypoxia in the LNCaP cells. Hypoxia induced 5.1±0.7-fold increases in mACON expression after incubation for 12 h, while hypoxia (vs normoxia) did not significantly increase mACON mRNA levels after 24-h incubation (Fig. 6A). The transient gene expression assay revealed that hypoxia, in comparison with normoxia, upregulated the promoter activity of the mACON gene in LNCaP cells by about twofold. Transient overexpression of HIF-1α significantly induced mACON promoter activities in LNCaP cells under both normoxic and hypoxic conditions (Fig. 6B). Results of immunoblotting assays showed that hypoxia induced 5.8±0.5-, 2.2±0.1-, 1.5±0.1-, and 3.6±0.4-fold increases in HIF-1α, LDHA, mACON, and VEGF protein levels respectively. The 100 nM chetomin co-treatment did not significantly block the HIF-1α protein levels; however, co-treatment with chetomin attenuates hypoxia-induced LDHA, mACON, and VEGF protein expressions. Co-treatment with chetomin attenuates hypoxia-blocked FASN protein expression in LNCaP cells (Fig. 6C). Induction of transient gene expression using a 5′-deletion assay indicated that increased mACON promoter activities induced by HIF-1α is dependent on the region −1013 to −841 upstream of the translational initiation site of the mACON gene (Fig. 6D).

Figure 6
Figure 6

Hypoxia modulates the gene expression of mACON of prostate carcinoma cells. (A) LNCaP cells were cultured under normoxic (black bars) or hypoxic (white bars) conditions for various time periods. The mACON mRNA levels were determined by RT-qPCR. The mRNA levels of normoxia-treated cells for 3 h were set at 1. Data are presented as the mean-fold (±s.e.m., n=3) of the mRNA levels in relation to the normoxia-treated group for 3 h (*P<0.01). (B) The mACON reporter vector-transfected LNCaP cells were co-transfected with varying concentrations of HIF-1α expression vector and then incubated in normoxic (black bars) or hypoxic (white bars) condition for another 24 h. Data are presented as the mean percentage (±s.e.m., n=6) of the luciferase activity in relation to the mock-transfected group (*P<0.05,+P<0.01). (C, left) LNCaP cells were cultured under normoxic (N), hypoxic (HP), or hypoxia with 100 nM of chetomin (HP+C) conditions for 24 h. The expressions of HIF-1α, mACON, LDHA, FASN, VEGF, and β-actin were determined by immunoblotting assay (C, right). The protein levels of normoxia-treated cells were regarded as 1. Data are presented as the fold-induction (±s.e.m., n=3) of protein levels of the hypoxia-treated group in relation to the normoxia-treated group (*P<0.05) and hypoxia with chetomin-treated group in relation to the hypoxia-treated groups (+P<0.05). (D) Nested deletion constructs of mACON reporter vectors were co-transfected with control expression vector (pcDNA3; white bars) or HIF-1α expression vector (black bars) into LNCaP cells. The SV40 promoter represents pGL3-promoter vector (Promega Bioscience). Data are presented as the mean percentage ±s.e.m. (n=6) of the luciferase activity of the reporter vectors induced by the HIF-1α expression vector relative to the luciferase activity associated with group that was co-transfected with control expression vector (+P<0.01).

Citation: Journal of Molecular Endocrinology 51, 1; 10.1530/JME-13-0090

Discussion

Hypoxia is regarded as a potent tumor-induced shield against destruction in prostate cancer (Marignol et al. 2008). The first response of oxygen limitation in the cells is to coordinate the regulation of glycolytic enzyme genes in order to maintain ATP synthesis (Warburg 1956, Kroemer & Pouyssegur 2008, Hu et al. 2011). One of the major hypoxia adaptations to prevent excessive reactive oxygen species is to induce expression of LDHA, which converts pyruvate into lactate (Semenza et al. 1996). In this study, we demonstrated that hypoxia induced LDHA protein expression in LNCaP and PC-3 prostate carcinoma cells in support of the theory of increase in the ‘anoxic glycolytic pathway’ in prostate carcinoma cells when tumor cells are exposed to hypoxia proposed by Costello & Franklin (2005). Our results showed that hypoxia enhanced lactate secretion in prostate carcinoma cells, which is consistent with an other report (Higgins et al. 2009). Our results further showed that HIF-1α knockdown in prostate carcinoma cells attenuated the activation of LDHA protein expression by hypoxia; moreover, chetomin blocked hypoxia-induced LDHA. These results are in agreement with previous reports, indicating an essential binding site for HIF-1α in the promoter of the human LDHA gene (Semenza 2007).

Unlike other organs, the prostate produces, accumulates, and releases large amounts of citrate (Costello & Franklin 1994). The study indicates that prostate carcinoma cells metabolize cytoplasmic citrate via mACON and FASN to enhance their metastatic behavior (Mycielska et al. 2006). FASN is a key metabolic enzyme that catalyzes the synthesis of palmitate from the condensation of malomy-CoA and acetyl-CoA de novo and plays a central role in energy homeostasis by converting excess carbon intake into fatty acids for storage. Elevated expression of FASN is linked with prostate tumorigenesis and progression (Migita et al. 2009, Flavin et al. 2011). Costello & Franklin (2005) proposed a ‘glycolysis–citrate–lipogenesis’ metabolic pathway indicating that a net increase in lipid biosynthesis and incorporation of citrate for lipogenesis is essential for proliferation in tumor cells. Our study showed that hypoxia enhances FASN gene expression in highly metastatic prostate carcinoma PC-3 cells. Another study also indicated hypoxia upregulated FASN gene expression in breast cancer cells (Furuta et al. 2008). However, we found that hypoxia decreased FASN gene expression in the human prostate carcinoma LNCaP cells, which is consistent with a previous report of hypoxia decreasing FASN mRNA in LNCaP cells (Higgins et al. 2009). LNCaP cells, in comparison with PC-3 cells and P53 wild-type cells are androgen-dependent and express a more oxidative phenotype, as assessed by energy metabolism and metabolic gene expression (Higgins et al. 2009, Tsui et al. 2011). Results from the present study indicate that the effect of hypoxia on FASN gene expression seems to be dependent on cell type, although the mechanisms of divergent effects of hypoxia on FASN in prostate carcinoma cells are still unknown.

The issue of adaptive response to hypoxia suggests that the limitation of cellular oxygen availability is the direct cause of the failure of the tricarboxylic acid cycle activity and citrate oxidation (Semenza et al. 1996). Nonetheless, other hypotheses suggest that hypoxia augments mACON in the Krebs cycle, leading to a gain of two thirds of the potential energy available from either glucose or fat-derived acetyl-CoA flux into the tricarboxylic acid cycle in the prostate (Hochachka et al. 2002, Costello & Franklin 2005). Our results demonstrate that hypoxia enhances the gene expression of mACON in prostate carcinoma cells in a HIF-1α-dependent manner. Hypoxia enhances gene expression of LDHA, FASN, and mACON in PC-3 cells, which suggests that hypoxia not only enhances the conversion of pyruvate into lactate but also promotes citrate in prostate carcinoma cells for both oxidation for energy production and export for lipogenesis. Our study is the first report to reveal HIF-1α-induced mACON gene expression at the transcriptional and translational levels in human prostate carcinoma cells, although the detailed mechanisms of HIF-1α's effect on mACON gene expression in prostate carcinoma cells still require further investigation. Interestingly, a study using the human embryonic kidney 293 cells indicated that hypoxia increased cACON but did not regulate mACON protein accumulation or activity (Schneider & Leibold 2003). The other study indicated that hypoxia (0.2% O2) for 48 h decreased total aconitase activity in primary human pulmonary arterial endothelial cells (Chang et al. 2009). However, our results showed that hypoxia enhances mACON gene expression but does not affect the enzymatic activity of cACON in either PC-3 or LNCaP cells. Therefore, it seems that hypoxia-induced mACON gene expression is specific to prostate cells.

l-mimosine is a prolyl 4-hydroxylase inhibitor, which culminates in the stabilization of the transcriptional activator HIF-1α and increases expressions of prostate-specific antigen, B-cell translocation gene 2, and N-myc downstream-regulated gene 1 gene products that mediate hypoxic adaptation in vitro in human prostate carcinoma cells (Chung et al. 2011, 2012). Our results indicate that l-mimosine induces increases in HIF-1α protein levels but reduces mACON gene expression. An early report indicated that l-mimosine induced P53 expression in human RKO colorectal cancer cells (Ji et al. 1997), while our recent study indicated that P53 downregulates mACON gene expression (Tsui et al. 2011). Interestingly, results of immunoblotting assays in the present study imply that l-mimosine does not increase P53 protein expression in wild-type P53-expressing LNCaP cells, suggesting that the decrease in mACON gene expression by l-mimosine may not occur through the p53 pathway. l-mimosine also acts as iron chelator, which reversibly blocks prostate cell proliferation at the late G1 phase (Chung et al. 2012). Our results indicate that l-mimosine blocks the increase in mACON gene expression by the iron donor, ferric ammonium citrate, suggesting that inhibition of mACON by l-mimosine is due to iron deprivation even though HIF-1α induces mACON gene expression. These results are in agreement with our previous study, which indicated that mACON is not only an iron-dependent enzyme but is also modulated by iron at both the transcriptional and translation levels (Juang 2004b).

HO-1, the inducible isoform of the rate-limiting enzyme in heme degradation, has cytoprotective effects against oxidative damage (Gozzelino et al. 2010). Immunohistochemical analysis of prostate tissue reveals significant elevation of HO-1 expression in epithelial cancer cells compared with those of normal or benign prostate tissues (Alaoui-Jamali et al. 2009, Li et al. 2011). Although the functions of HO-1 in prostate carcinoma cells are still not well delineated, our results show that hypoxia induces HO-1 expression in prostate (LNCaP and PC-3) cells in vitro. These results are consistent with previous research on the human microvascular endothelial cell line HMEC-1 (Ockaili et al. 2005).

Results of RT-qPCR and transient gene expression assays in the present study indicated that modulation of hypoxia in mACON occurs at the transcriptional levels. Moreover, our immunoblotting data showed that chetomin, an epidithiodioxopiperazine known to disrupt the HIF-1α pathway (Kung et al. 2004), attenuates hypoxia-induced mACON expressions, suggesting that hypoxia-induced mACON gene expression is via the HIF-1α-dependent pathway in prostate carcinoma cells. Induction of transient gene expression by the 5′-deletion assay indicated that enhancement of HIF-1α on mACON promoter activity is dependent on the −1013 to −841 region upstream of the translational initiation site of the mACON gene. However, we did not find the conserved DNA sequence (5′-RCGTG-3′) of HIF-1α response element (Wang et al. 1995) but the putative Egr1 and Oct4 binding sites in the mACON promoter region (−1013 to −841). Studies indicated that hypoxia indirectly regulates target gene expression by modulating microRNAs, chromatin-modifying enzymes, and transcription factors including EGR1 and OCT4 (POU5F1) (Zhang et al. 2007, Semenza 2012). However, we cannot rule out the possibility that the consensus hypoxia response elements may locate at the 5′-upstream of the DNA fragment (−1013 to +38), which we used in this study, as an other study indicated that 60% of HIF-1 binding sites are located >2.5 kb upstream of the transcription start site (Schodel et al. 2011). Whether mACON gene expression by hypoxia enhances directly and/or indirectly via the HIF-1α-dependent pathway remains to be further elucidated.

In conclusion, we suggest that hypoxia enhances not only the ‘anoxic glycolytic pathway’ by inducing LDHA gene expression, the ‘glycolytic-lipogenic pathway’ by inducing FASN gene expression, but also the ‘promoter citrate oxidation pathway’ by increasing mACON gene expression. The greater induction of LDHA than FASN and mACON by hypoxia further suggests that the major metabolic adaptation of hypoxia is the ‘anoxic glycolytic pathway’ in prostate carcinoma cells. The modulation of hypoxia in LDHA, FASN, and mACON is via the HIF-1α-dependent pathway in prostate carcinoma cells. Although both l-mimosine and hypoxia treatments stabilize HIF-1α protein expression, hypoxia enhances expression of mACON at the transcriptional level, while l-mimosine blocks mACON expression by iron chelation. The hypoxia-induced mACON gene expression is dependent on the DNA fragment (−1013 to −841) upstream of the translation initiation site of the mACON gene. Hypoxia-induced mACON gene expression is via the HIF-1α-dependent and iron-dependent pathways in prostate carcinoma cells.

Supplementary data

This is linked to the online version of the paper at http://dx.doi.org/10.1530/JME-13-0090.

Declaration of interest

The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.

Funding

This Research was supported by Chang Gung Memorial Hospital (CMRP-D190543, -D190613, -G3B1891, and -G392143) and Taiwan National Science Council (101-2314-B-182A-099-MY3 and NSC 101-2320-B-182-002).

References

  • Alaoui-JamaliMABismarTAGuptaASzarekWASuJSongWXuYXuBLiuGVlahakisJZ2009A novel experimental heme oxygenase-1-targeted therapy for hormone-refractory prostate cancer. Cancer Research6980178024. (doi:10.1158/0008-5472.CAN-09-0419).

    • Search Google Scholar
    • Export Citation
  • BaronAMigitaTTangDLodaM2004Fatty acid synthase: a metabolic oncogene in prostate cancer?Journal of Cellular Biochemistry914753. (doi:10.1002/jcb.10708).

    • Search Google Scholar
    • Export Citation
  • BertilssonHTessemM-BFlatbergAVisetTGribbestadIAngelsenAHalgunsetJ2012Changes in gene transcription underlying the aberrant citrate and choline metabolism in human prostate cancer samples. Clinical Cancer Research1832613269. (doi:10.1158/1078-0432.CCR-11-2929).

    • Search Google Scholar
    • Export Citation
  • Bourdeau-HellerJOberleyTD2007Prostate carcinoma cells selected by long-term exposure to reduced oxygen tension show remarkable biochemical plasticity via modulation of superoxide, HIF-1α levels, and energy metabolism. Journal of Cellular Physiology212744752. (doi:10.1002/jcp.21069).

    • Search Google Scholar
    • Export Citation
  • ChangSYZhangYYHemannCMahoneyCEZweierJLLoscalzoJ2009MicroRNA-210 controls mitochondrial metabolism during hypoxia by repressing the iron–sulfur cluster assembly proteins ISCU1/2. Cell Metabolism10273284. (doi:10.1016/j.cmet.2009.08.015).

    • Search Google Scholar
    • Export Citation
  • ChungLCTsuiKHFengTHLeeSLChangPLJuangHH2011Curcumin provides potential protection against the activation of hypoxia and prolyl 4-hydroxylase inhibitors on prostate-specific antigen expression in human prostate carcinoma cells. Molecular Nutrition & Food Research5516661676. (doi:10.1002/mnfr.201100328).

    • Search Google Scholar
    • Export Citation
  • ChungLCTsuiKHFengTHLeeSLChangPLJuangHH2012l-mimosine blocks cell proliferation via upregulation of B-cell translocation gene 2 and N-myc downstream regulated gene 1 in prostate carcinoma cells. American Journal of Physiology. Cell Physiology302C676C685. (doi:10.1152/ajpcell.00180.2011).

    • Search Google Scholar
    • Export Citation
  • CostelloLCFranklinRB1994The bioenergetic theory of prostate malignancy. Prostate25162166. (doi:10.1002/pros.2990250308).

  • CostelloLCFranklinRB2005“Why do tumour cells glycolyse”?: from glycolysis through citrate to lipogenesis. Molecular and Cellular Biochemistry28018. (doi:10.1007/s11010-005-8841-8).

    • Search Google Scholar
    • Export Citation
  • CostelloLCFranklinRB2012The genetic/metabolic transformation concept of carcinogenesis. Cancer Metastasis Reviews31123130. (doi:10.1007/s10555-011-9334-8).

    • Search Google Scholar
    • Export Citation
  • DeBerardinisRJ2008Is cancer a disease of abnormal cellular metabolism? New angles on an old idea. Genetics in Medicine10767777. (doi:10.1097/GIM.0b013e31818b0d9b).

    • Search Google Scholar
    • Export Citation
  • FengTHTsuiKHJuangHH2005Cholesterol modulation of the expression of mitochondrial aconitase in human prostatic carcinoma cells. Chinese Journal of Physiology4893100.

    • Search Google Scholar
    • Export Citation
  • FlavinRZadraGLodaM2011Metabolic alterations and targeted therapies in prostate cancer. Journal of Pathology223283294. (doi:10.1002/path.2809).

    • Search Google Scholar
    • Export Citation
  • FurutaEPaiSKZhanRBandyopadhyaySWatabeMMoYYHirotaSHosobeSTsukadaTMiuraK2008Fatty acid synthase gene is up-regulated by hypoxia via activation of Akt and sterol regulatory element binding protein-1. Cancer Research6810031011. (doi:10.1158/0008-5472.CAN-07-2489).

    • Search Google Scholar
    • Export Citation
  • GozzelinoRJeneyVSoaresMP2010Mechanisms of cell protection by heme oxygenase-1. Annual Review of Pharmacology and Toxicology50323354. (doi:10.1146/annurev.pharmtox.010909.105600).

    • Search Google Scholar
    • Export Citation
  • HigginsLHWithersHGGarbensALoveHDMagnoniLHaywardSWMoyesCD2009Hypoxia and the metabolic phenotype of prostate cancer cells. Biochimica et Biophysica Acta178714331443. (doi:10.1016/j.bbabio.2009.06.003).

    • Search Google Scholar
    • Export Citation
  • HochachkaPWRupertJLGoldenbergLGleaveMKozlowskiP2002Going malignant: the hypoxia-cancer connection in the prostate. BioEssays24749757. (doi:10.1002/bies.10131).

    • Search Google Scholar
    • Export Citation
  • HuHDengCYangTChenYNiceECHuangCWeiY2011Proteomics revisits the cancer metabolome. Expert Review of Proteomics84505553. (doi:10.1586/epr.11.31).

    • Search Google Scholar
    • Export Citation
  • JiCMarnettLJPietenpolJA1997Cell cycle re-entry following chemically-induced cell cycle synchronization leads to elevated p53 and p21 protein levels. Oncogene1527492753. (doi:10.1038/sj.onc.1201441).

    • Search Google Scholar
    • Export Citation
  • JuangHH2004aModulation of mitochondrial aconitase on the bioenergy of human prostate carcinoma cells. Molecular Genetics and Metabolism81244252. (doi:10.1016/j.ymgme.2003.12.009).

    • Search Google Scholar
    • Export Citation
  • JuangHH2004bModulation of iron on mitochondrial aconitase expression in human prostatic carcinoma cells. Molecular and Cellular Biochemistry265185194. (doi:10.1023/B:MCBI.0000044395.59739.1f).

    • Search Google Scholar
    • Export Citation
  • JuangHH2004cCyclic adenosine 3′,5′-monosphosphate mediate prolactin regulation of mitochondrial aconitase in human prostate carcinoma cells. Molecular and Cellular Endocrinology219141149. (doi:10.1016/j.mce.2004.01.001).

    • Search Google Scholar
    • Export Citation
  • JuangHHHsiehMLTsuiKH2004Testosterone modulates mitochondrial aconitase in the full-length human androgen receptor-transfected PC-3 prostatic carcinoma cells. Journal of Molecular Endocrinology33121132. (doi:10.1677/jme.0.0330121).

    • Search Google Scholar
    • Export Citation
  • KroemerGPouyssegurJ2008Tumor cells metabolism: cancer's Achilles' heel. Cancer Cell13472482. (doi:10.1016/j.ccr.2008.05.005).

  • KungALZabludoffSDFranceDSFreedmanSJTannerEAVieiraACornell-KennonSLeeJWangBWangJ2004Small molecule blockade of transcriptional coactivation of the hypoxia-inducible factor pathway. Cancer Cell63343. (doi:10.1016/j.ccr.2004.06.009).

    • Search Google Scholar
    • Export Citation
  • LeACooperCRGouwAMDinavahiRMaitraADeckLMRoyerREVander JagtDLSemenzaGLDangCV2010Inhibition of lactate dehydrogenase A induces oxidative stress and inhibits tumor progression. PNAS10720372042. (doi:10.1073/pnas.0914433107).

    • Search Google Scholar
    • Export Citation
  • LiYSuJDingZhangXZhangJYoshimotoMLiuSBijianKGuptaASquireJAAlaoui-JamaliMA2011PTEN deletion and heme oxygenase-1 overexpression cooperate in prostate cancer progression and are associated with adverse clinical outcome. Journal of Pathology22490100. (doi:10.1002/path.2855).

    • Search Google Scholar
    • Export Citation
  • LiuY2006Fatty acid oxidation is dominant bioenergetic pathway in prostate cancer. Prostate Cancer and Prostatic Diseases3230234. (doi:10.1038/sj.pcan.4500879).

    • Search Google Scholar
    • Export Citation
  • MarignolLCoffeyMLawlerMHollywoodD2008Hypoxia in prostate cancer: a powerful shield against tumour destruction?Cancer Treatment Reviews34313327. (doi:10.1016/j.ctrv.2008.01.006).

    • Search Google Scholar
    • Export Citation
  • MigitaTRuizSFornariAFiorentinoMPrioloCZadraGInazukaFGrisanzioCPalescandoloEShinE2009Fatty acid synthase: a metabolic enzyme and candidate oncogene in prostate cancer. Journal of the National Cancer Institute101519532. (doi:10.1093/jnci/djp030).

    • Search Google Scholar
    • Export Citation
  • MycielskaMEBroke-SmithTPPalmerCPBeckermanRNastosTErgulerKDjamgozMB2006Citrate enhances in vitro metastatic behaviours of PC-3M human prostate cancer cells: status of endogenous citrate and dependence on aconitase and fatty acid synthase. International Journal of Biochemistry & Cell Biology3817661777. (doi:10.1016/j.biocel.2006.04.008).

    • Search Google Scholar
    • Export Citation
  • OckailiRNatarajanRSalloumFFisherBJJonesDFlowlerAAIIIKukrejaRC2005HIF-1 activation attenuates postichemic myocardial injury: role for heme oxygenase-1 in modulating microvascular chemokine generation. American Journal of Physiology. Heart and Circulatory Physiology289H542H548. (doi:10.1152/ajpheart.00089.2005).

    • Search Google Scholar
    • Export Citation
  • SchneiderBDLeiboldEA2003Effects of iron regulatory protein regulation on iron homeostasis during hypoxia. Blood10234043411. (doi:10.1182/blood-2003-02-0433).

    • Search Google Scholar
    • Export Citation
  • SchodelJOikonomopoulosSRagoussisJPughCWRatcliffePJMoleDR2011High-resolution genome-wide mapping of HIF-binding sites by ChIP-seq. Blood117e207e217. (doi:10.1182/blood-2010-10-314427).

    • Search Google Scholar
    • Export Citation
  • SemenzaGL2007Oxygen-dependent regulation of mitochondrial respiration by hypoxia-inducible factor 1. Biochemical Journal40519. (doi:10.1042/BJ20070389).

    • Search Google Scholar
    • Export Citation
  • SemenzaGL2012Hypoxia-inducible factors: mediators of cancer progression and targets for cancer therapy. Trends in Pharmacological Sciences33207214. (doi:10.1016/j.tips.2012.01.005).

    • Search Google Scholar
    • Export Citation
  • SemenzaGLJiangBHLeungSWPassantinoRConcordetJPMairePGiallongoA1996Hypoxia response elements in the aldolase A, enolase 1, and lactate dehydrogenase A gene promoters contain essential binding sites for hypoxia-inducible factor 1. Journal of Biological Chemistry2713252932537. (doi:10.1074/jbc.271.51.32529).

    • Search Google Scholar
    • Export Citation
  • TsuiKHChangPLJuangHH2006Zinc blocks gene expression of mitochondrial aconitase in human prostatic carcinoma cells. International Journal of Cancer118609615. (doi:10.1002/ijc.21411).

    • Search Google Scholar
    • Export Citation
  • TsuiKHFengTHLinYFChangPLJuangHH2011p53 Downregulates the gene expression of mitochondrial aconitase in human prostate carcinoma cells. Prostate716270. (doi:10.1002/pros.21222).

    • Search Google Scholar
    • Export Citation
  • TsuiKHChangYLFengTHChungLCLeeTYChangPLJuangHH2012Growth differentiation factor-15 upregulates interleukin-6 to promoter tumorigenesis of prostate carcinoma PC-3 cells. Journal of Molecular Endocrinology49153163. (doi:10.1530/JME-11-0149).

    • Search Google Scholar
    • Export Citation
  • WangGLJiangBHRueEASemenzaGL1995Hypoxia inducible factor 1 is a basic-helix–loop–helix-PAS heterodimer regulated by cellular O2 tension. PNAS9255105514. (doi:10.1073/pnas.92.12.5510).

    • Search Google Scholar
    • Export Citation
  • WarburgO1956On respiratory impairment in cancer cells. Science124269270.

  • YamadaYNakamuraKAokiSTobiumeMZennamiKKatoYNishikawaGYoshizawaTItohYNakaokaA2011Lactate dehydrogenase, Gleason score and HER-2 overexpression are significant prognostic factors for M1b prostate cancer. Oncology Reports25937944. (doi:10.3892/or.2011.1154).

    • Search Google Scholar
    • Export Citation
  • YeungSJPanJLeeMH2008Roles of p53, MYC and HIF-1 in regulating glycolysis – the seventh hallmark of cancer. Cellular and Molecular Life Sciences6539813999. (doi:10.1007/s00018-008-8224-x).

    • Search Google Scholar
    • Export Citation
  • ZhangPTchou-WongKMCostaM2007Egr-1 mediates hypoxia-inducible transcription of the NDRG1 gene through an overlapping Egr-1/Sp1 binding site in the promoter. Cancer Research6791259133. (doi:10.1158/0008-5472.CAN-07-1525).

    • Search Google Scholar
    • Export Citation
*

(K-H Tsui and L-C Chung contributed equally to this work)

If the inline PDF is not rendering correctly, you can download the PDF file here.

 

      Society for Endocrinology

Sept 2018 onwards Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 1026 257 15
PDF Downloads 260 111 5
  • View in gallery

    Hypoxia modulates protein expression of mACON, FASN, and LDHA in LNCaP cells. (A) LNCaP cells were cultured under hypoxic or normoxic conditions for 24 h. The expression of HIF-1α, mACON, FASN, LDHA, HO-1, VEGF, and β-actin were determined by immunoblotting assay. (B) The quantitative analysis was done by determining the intensity of each band for target genes and β-actin from three independent experiments. Data are presented as the fold-induction (±s.e.m., n=3) of the relative density of the target gene/β-actin in relation to the normoxia treatment group. (C) Mitochondria were extracted and the enzymatic activity of mACON was analyzed in the LNCaP cells. Data are presented as the mean percentage (±s.e.m., n=6) of mACON enzymatic activity relative to that of the normoxia-treated group. (D) Supernatants of the conditioned medium were collected and the lactate levels were analyzed in the LNCaP cells. Data are presented as the mean percentage (±s.e.m., n=6) of lactate secretion relative to that of the normoxia-treated cell group; +P<0.01.

  • View in gallery

    Hypoxia modulates protein expression of mACON, FASN, and LDHA in PC-3 cells. (A) PC-3 cells were treated and then the expression of proteins were analyzed as described in Fig. 1A. (B) The quantitative analysis and data presentation are described as in Fig. 1B (*P<0.05,+P<0.01). (C) Mitochondria were extracted and the enzymatic activity of mACON was analyzed in the PC-3 cells. (D) Supernatants of the conditioned medium were collected and the lactate levels were analyzed in the PC-3 cells. Data are presented as the mean percentage (±s.e.m., n=6) of mACON enzymatic activities and lactate secretion relative to that of the normoxia-treated cell group (+P<0.01).

  • View in gallery

    Hypoxia modulates protein expression of mACON, FASN, and LDHA in PZ-HPV-7, CA-HPV-10, and DU145 cells. Cells were cultured under hypoxic or normoxic conditions for 24 h. The expression of HIF-1α, FASN, LDHA, mACON, and β-actin were determined by immunoblotting assay. (B) The quantitative analysis and data presentation are described as in Fig. 1B (*P<0.05,+P<0.01).

  • View in gallery

    HIF-1α knockdown attenuates the activation of hypoxia on gene expression of mACON in prostate carcinoma cells. The mock-knockdown LNCaP (LN-COLsi) and the HIF-1α-knockdown LNCaP cells (LN-HIF-1αsi) (A) and the mock-knockdown PC-3 (PC-COLsi) and HIF-1α-knockdown PC-3 cells (PC-HIF-1αsi) (C) were cultured under normoxic (N) or hypoxic (HP) conditions for 24 h. The expressions of HIF-1α, mACON, FASN, LDHA, VEGF, and β-actin were determined by immunoblotting assay. Quantitative results were analyzed by determining the intensity of each band for the target genes (white bars, FASN; black bars, mACON; gray bars, VEGF; white coarse bars, LDHA, and gray cyclone bars, HIF-1α) and β-actin (B and D). Data are presented as the fold-induction (±s.e.m., n=3) of the relative density of the target gene/β-actin in relation to the mock-knockdown normoxia-treated group (*,#,+P<0.01).

  • View in gallery

    l-mimosine downregulates the expression of human mACON gene via iron chelation. (A) Immunoblots of HIF-1α, mACON, and β-actin using lysates from varying concentrations of l-mimosine-treated LNCaP cells. (B) The quantitative analysis and data presentation are described as in Fig. 1B (*P<0.05,+P<0.01). (C) The mACON reporter vector-transfected LNCaP cells were treated with varying concentrations of l-mimosine as indicated for 24 h. Data are presented as the mean percentage±s.e.m. (n=6) of the luciferase activity induced by l-mimosine relative to the control solvent group (*P<0.05). (D) Immunoblotting for P53 and β-actin using lysates from varying concentrations of l-mimosine-treated LNCaP cells. (E) Immunoblotting for HIF-1α, mACON, and β-actin using lysates from varying concentrations of l-mimosine-treated PC-3 cells. (F) PC-3 cells were treated with 400 μM l-mimosine and/or 100 μg/ml ferric ammonium citrate for 24 h. The expressions of mACON and β-actin were determined by immunoblotting assays. (G) The mACON reporter vector-transfected PC-3 cells were treated with l-mimosine (400 μM) and/or ferric ammonium citrate (100 μg/ml) for 24 h. Data are presented as the mean percentage±s.e.m. (n=6) of the luciferase activity in relation to the control solvent-treated group (*P<0.05; C, control solvent; L, l-mimosine; F, ferric ammonium citrate).

  • View in gallery

    Hypoxia modulates the gene expression of mACON of prostate carcinoma cells. (A) LNCaP cells were cultured under normoxic (black bars) or hypoxic (white bars) conditions for various time periods. The mACON mRNA levels were determined by RT-qPCR. The mRNA levels of normoxia-treated cells for 3 h were set at 1. Data are presented as the mean-fold (±s.e.m., n=3) of the mRNA levels in relation to the normoxia-treated group for 3 h (*P<0.01). (B) The mACON reporter vector-transfected LNCaP cells were co-transfected with varying concentrations of HIF-1α expression vector and then incubated in normoxic (black bars) or hypoxic (white bars) condition for another 24 h. Data are presented as the mean percentage (±s.e.m., n=6) of the luciferase activity in relation to the mock-transfected group (*P<0.05,+P<0.01). (C, left) LNCaP cells were cultured under normoxic (N), hypoxic (HP), or hypoxia with 100 nM of chetomin (HP+C) conditions for 24 h. The expressions of HIF-1α, mACON, LDHA, FASN, VEGF, and β-actin were determined by immunoblotting assay (C, right). The protein levels of normoxia-treated cells were regarded as 1. Data are presented as the fold-induction (±s.e.m., n=3) of protein levels of the hypoxia-treated group in relation to the normoxia-treated group (*P<0.05) and hypoxia with chetomin-treated group in relation to the hypoxia-treated groups (+P<0.05). (D) Nested deletion constructs of mACON reporter vectors were co-transfected with control expression vector (pcDNA3; white bars) or HIF-1α expression vector (black bars) into LNCaP cells. The SV40 promoter represents pGL3-promoter vector (Promega Bioscience). Data are presented as the mean percentage ±s.e.m. (n=6) of the luciferase activity of the reporter vectors induced by the HIF-1α expression vector relative to the luciferase activity associated with group that was co-transfected with control expression vector (+P<0.01).