3-BrPA (3-bromopyruvate) is an alkylating agent with anti-tumoral activity on hepatocellular carcinoma. This compound inhibits cellular ATP production owing to its action on glycolysis and oxidative phosphorylation; however, the specific metabolic steps and mechanisms of 3-BrPA action in human hepatocellular carcinomas, particularly its effects on mitochondrial energetics, are poorly understood. In the present study it was found that incubation of HepG2 cells with a low concentration of 3-BrPA for a short period (150 μM for 30 min) significantly affected both glycolysis and mitochondrial respiratory functions. The activity of mitochondrial hexokinase was not inhibited by 150 μM 3-BrPA, but this concentration caused more than 70% inhibition of GAPDH (glyceraldehyde-3-phosphate dehydrogenase) and 3-phosphoglycerate kinase activities. Additionally, 3-BrPA treatment significantly impaired lactate production by HepG2 cells, even when glucose was withdrawn from the incubation medium. Oxygen consumption of HepG2 cells supported by either pyruvate/malate or succinate was inhibited when cells were pre-incubated with 3-BrPA in glucose-free medium. On the other hand, when cells were pre-incubated in glucose-supplemented medium, oxygen consumption was affected only when succinate was used as the oxidizable substrate. An increase in oligomycin-independent respiration was observed in HepG2 cells treated with 3-BrPA only when incubated in glucose-supplemented medium, indicating that 3-BrPA induces mitochondrial proton leakage as well as blocking the electron transport system. The activity of succinate dehydrogenase was inhibited by 70% by 3-BrPA treatment. These results suggest that the combined action of 3-BrPA on succinate dehydrogenase and on glycolysis, inhibiting steps downstream of the phosphorylation of glucose, play an important role in HepG2 cell death.

INTRODUCTION

Cancer cells depend on large quantities of energy to maintain elevated rates of proliferation. In this regard, specific adaptations in energy-yielding pathways are observed, including activation of glycolysis and down-regulation of oxidative phosphorylation, also known as the Warburg effect [1,2], a metabolic signature of many tumour cells. Accordingly, the effects of specific drugs designed to interfere with the energy-yielding pathways of many types of cancer have been investigated [35]. This is the case of the analogue of pyruvate/lactate, 3-BrPA (3-bromopyruvate), an alkylating agent able to react with thiol (SH) and hydroxy (OH) groups of several enzymes [610].

Studies with HCC (hepatocellular carcinoma) cells have demonstrated that the main blocking site of 3-BrPA is the mt-HK (hexokinase) type II (mitochondrial HK II), which causes an impairment of glycolytic flux [11,12]. In addition, 3-BrPA seems to affect mitochondrial respiration, although the exact mechanism of mitochondrial collapse was not identified [11]. Subsequent studies have confirmed 3-BrPA as an agent able to cause ATP depletion in tumour cells, with no apparent effect on nontransformed cells [1315]. As a result, 3-BrPA has been considered a potent agent in the treatment against the proliferation of HCC [5,13,14]. In addition to 3-BrPA effects on energy homoeostasis, few studies have suggested that this compound induces cell death either through activation of the mitochondrial pathway of apoptosis or necrosis [5,12,15,16]. Given that the precise mechanisms by which 3-BrPA affects mitochondrial physiology and bioenergetics are still unknown, a detailed characterization of the pharmacological action of this anti-tumoral agent is crucial to the development of more efficient treatments.

In the present study it is demonstrated for the first time that brief exposure of a HCC cell line, HepG2, to 3-BrPA at a micromolar concentration leads to a severe impairment of mitochondrial respiratory function by specifically affecting SDH (succinate dehydrogenase) activity. An interesting observation is that when cells are forced to rely on mitochondrial oxidative phosphorylation rather than glycolysis, 3-BrPA effects on respiratory parameters appear to be more pronounced. Moreover, we show that 3-BrPA, at the concentrations used, inhibits glycolytic flux in a way not dependent on mt-HK II inhibition, but due to enzyme inhibition downstream of the phosphorylation of glucose. The metabolic implications of such results to HepG2 cell death are discussed in an attempt to better understand the action of 3-BrPA in vivo.

EXPERIMENTAL

Cell culture

HepG2, a human HCC cell line, was obtained from American Type Culture Collection and grown in MEM (minimal essential medium) with 5 mM glucose, supplemented with 10% (v/v) FBS (foetal bovine serum), 0.22% sodium bicarbonate and 0.2% Hepes (pH 7.4) at 37 °C in a humidified incubation chamber with 5% CO2. For viability assays, cells were grown on 24-well plates. For all other assays, cells were grown on plastic Petri dishes. In each assay, cells were seeded at a density of 105 cells/ml. Cells were sub-cultured every 2–4 days and were used for experiments when they were nearly 95% confluent.

Cell viability

Cells were incubated with different concentrations of 3-BrPA at 37 °C for 30, 60, 90 and 180 min. After each treatment, cells were washed twice with PBS and cell viability was assessed using the MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide] assay as described previously [17] and cross-checked by the Trypan Blue dye-exclusion assay.

Animal care and use

The experimental protocols using animals were approved by the Committee for Ethics in Animal Research of the Universidade Federal do Rio de Janeiro in compliance with the Brazilian College for Animal Experimentation.

Preparation of cellular extracts

HepG2 cells were disrupted by liquid nitrogen addition and mixed in a regular lysis buffer containing 10 mM Tris/HCl (pH 7.0), 20 mM NaF, 1 mM DTT (dithiothrietol), 250 mM sucrose, 5 mM EDTA, 1 mM PMSF, 10 μM leupeptin and 1 μM pepstatin A. The suspension was centrifuged at 100 g for 5 min at 4 °C, and the resulting supernatant was used for measurement of the recovered activities of HK, PGI (phosphoglucose isomerase), PFK (phosphofructokinase), GAPDH (glyceraldehyde-3-phosphate dehydrogenase), PGK (phosphoglycerate kinase) and PK (pyruvate kinase) after 3-BrPA treatment [18].

Preparation of HK isoforms

Mitochondrial fractions for measurements of type II mt-HK (HepG2 cells) activities were obtained by centrifugation of the cellular extracts at 10000 g for 15 min at 4 °C. The resulting pellets were resuspended in lysis buffer and used for enzymatic assays [18]. Mitochondrial fractions for measurements of type I mt-HK (mouse brain) activities were obtained as previously described [19]. GK (glucokinase) activity was measured in soluble fractions of mouse liver. Briefly, liver was excised and homogenized with a Potter–Elvehjem-type homogenizer in a medium containing 20 mM triethanolamine (pH 7.5), 5 mM EGTA, 100 mM glucose, 100 mM KCl, 1 mM DTT, 5% glycerol, 5 mM EDTA, 0.25% sodium azide and 0.5 mM PMSF. The homogenate was centrifuged at 100000 g for 40 min. Ammonium sulfate (45% saturation) was added to the supernatant and centrifuged at 15000 g for 10 min. Ammonium sulfate (65% saturation) was added to the resulting supernatant and centrifuged at 15000 g for 10 min. The final pellet was suspended in a small volume of medium and used for enzymatic assays.

Measurements of enzymatic activities

Recovery of glycolytic enzyme activities after treatment with 3-BrPA

The activity of the glycolytic enzymes HK, PGI, PFK, GAPDH, PGK and PK were measured in cellular extracts after cell pre-treatment with 3-BrPA. Culture medium was replaced with fresh medium (MEM) containing 150 μM 3-BrPA and cells were incubated for 30 min at 37 °C. Enzymatic activities were measured using the following reaction media: (a) for HK, 10 mM Tris/HCl (pH 7.4), 5 mM MgCl2, 0.5 mM β-NADP+, 1 unit/ml GAPDH (Leuconostoc mesenteroides), 0.1% Triton X-100, 1 μg/ml antimycin A, 1 μM rotenone, 2 mM sodium azide, 2 mM ATP and 50–100 μg/ml protein [20]; (b) for PGI, 50 mM Tris/HCl (pH 7.4), 0.5 mM β-NAD+, 1 unit/ml GAPDH (Leuconostoc mesenteroides), 1 mM fructose 6-phosphate and 50–100 μg/ml protein [21]; (c) for PFK, 50 mM Tris/HCl (pH 7.4), 5 mM MgCl2, 5 mM ammonium sulfate, 1.5 units/ml aldolase, 4 units/ml triose phosphate isomerase, 4 units/ml α-glycerophosphate dehydrogenase, 0.25 mM β-NADH, 0.1 mM ATP and 50–100 μg/ml protein [22]; (d) for GAPDH, 50 mM Tris/HCl (pH 7.4), 2 mM MgCl2, 1 mM ATP, 1 mM EDTA, 13 units/ml PGK, 0.25 mM β-NADH and 50–100 μg/ml protein [23]; (e) for PGK, 50 mM Tris/HCl (pH 7.4), 2 mM MgCl2, 1 mM ATP, 0.2 mM β-NADH, 5 units/ml GAPDH and 50–100 μg/ml protein [23]; and (f) for PK, 50 mM imidazole buffer (pH 7.4), 2 mM MgCl2, 100 mM KCl, 0.5 mM β-NADH, 1 mM ADP, 5 units/ml lactate dehydrogenase and 50–100 μg/ml protein [24]. The reactions were started by the addition of 1 mM glucose (for HK), 1 mM fructose 6-phosphate (for PGI and PFK), 5 mM 3-phosphoglycerate (for GAPDH), 5 mM 3-phosphoglycerate (for PGK) or 1 mM phosphoenolpyruvate (for PK). For all of the reactions, the baseline was followed for 3 min.

Direct effect of 3-BrPA on enzyme preparations

The direct effect of 3-BrPA on three isoforms of HK and on SDH was evaluated by incubating the enzyme preparations with different concentrations of 3-BrPA. The activities of mt-HK I, mt-HK II and GK were measured using an adaptation of a radiometric assay described previously [25]. The assay was performed at 37 °C in 0.4 ml of reaction medium containing 20 mM Tris/HCl (pH 7.5), 5 mM MgCl2, 1 mM [γ-32P]ATP (3000 c.p.m./nmol of ATP), 10 mM glucose (mt-HK types I and II) or 100 mM glucose (GK), 1 mM sodium orthovanadate, 5 mM sodium azide and different concentrations of 3-BrPA. Duplicates were performed for all assays and blanks were obtained in the absence of glucose. The reaction was started by the addition of protein (0.1 mg/ml) and quenched by the addition of 1 ml of activated charcoal in 0.1 M HCl. After centrifugation (1000 g for 15 min), 0.4 ml aliquots of supernatant were added to liquid-scintillation vials and counted. The resulting radioactivity corresponds to [32P]G6P (glucose 6-phosphate) formed from the HK-catalysed reaction. The activity of SDH was determined spectrophotometrically using DCPIP (2,6-dichlorophenol-indophenol) as an artificial electron acceptor and succinate as the substrate [26]. The assay was performed at room temperature (25 °C) in 1.0 ml of reaction medium containing 20 mM phosphate buffer (pH 7.2), 0.1% Triton X-100, 4 mM sodium azide, 5 mM succinate, 50 μM DCPIP and 2.5–300 μM 3-BrPA. Blanks were obtained in the absence of succinate. The reaction was started by adding 0.1 mg of HepG2 mitochondrial fraction (prepared as described for the mt-HK activity using the same buffer, but without DTT) and the reduction of DCPIP was monitored for 3 min at 600 nm. The SDH activity was calculated using the molar absorption coefficient of reduced DCPIP (21.0 mM−1·cm−1). SDH activity in the presence of 3-BrPA is presented as a percentage of the control activity.

Metabolomic analyses of HepG2 cells by NMR

Culture medium was replaced by fresh glucose-free DMEM (Dulbecco's modified Eagle's medium; Invitrogen) supplemented with 5 mM D-[U-13C]glucose and 0.2% Hepes. Following this, cells were incubated for 30 min at 37 °C in the absence or in the presence of 150 μM 3-BrPA. Following the incubation period, medium was removed and ∼8×107 cells were suspended in glucose-free DMEM containing 10% (v/v) deuterium oxide to a final volume of 600 μl. One-dimensional 13C spectra of the HepG2 cellular metabolites was obtained at 5 °C using 45o pulses with a repetition time of 0.6 s, 16000 complex points, ∼20000 scans and a spectral width of 200 p.p.m. The free-induction decays were zero-filled to 16384 points and apodized with exponential multiplication using line broadening of 20 Hz. One-dimensional 31P spectra were obtained using 45o pulses with a repetition time of 0.6 s, 16000 complex points, ∼2000 scans and a spectral width of 120 p.p.m. The free-induction decays were zero-filled to 16384 points, and apodized with exponential multiplication using line broadening of 10 Hz. Spectra of HepG2 cell metabolites were acquired with a Bruker DRX 400 MHz using a broadband inverse detection probe (BBI). Spectral processing and analysis was performed using Topspin 2.0. Analysis and assignment of the metabolites were obtained using the Human Metabolome Database v1.0 using the 13C one-dimensional spectra search mode with a tolerance of 0.25 p.p.m. [27].

Detection of lactate in the culture medium

Culture medium was replaced with fresh medium containing 150 μM 3-BrPA and cells were incubated at 37 °C. Two different culture media were used: GFM (glucose-free medium; DMEM), which contains only glutamine as an oxidizable substrate, and GM {GFM supplemented with 5 mM glucose or 5 mM D-[U-13C]glucose (for NMR assays)}. Both media were supplemented with 0.2% Hepes. At the times indicated, samples of 50 or 100 μl from culture medium were collected to evaluate lactate accumulation as described below.

Enzymatic assay

Lactate measurement was performed in a hydrazine/glycine buffer (pH 9.2), containing 5 mg/ml β-NAD+ and 15 units/ml lactate dehydrogenase to a final volume of 200 μl. The absorbance due to formation of NADH was monitored in a microplate reader (SpectraMax M5, Molecular Devices) for 30 min at 340 nm and was correlated with the presence of lactate on samples from a standard curve [28].

Detection by NMR

One-dimensional 13C spectra for the kinetics of lactate cell export were obtained at 5 °C using 45o pulses with a repetition time of 0.6 s, 16000 complex points, 2028 scans and a spectral width of 200 p.p.m. The free-induction decays were zero-filled to 16384 points and apodized with exponential multiplication using line broadening of 10 Hz. Spectra were acquired and analysed as described for metabolomic analysis as described above.

Oxygen consumption of intact and digitonin-permeabilized HepG2 cells

Oxygen consumption rates were measured polarographically in an oximeter using a Clark-type oxygen electrode (Oxytherm, Hansatech Instruments) or using high-resolution respirometry (Oroboros Oxygraph-O2K). The electrode was calibrated between 0 and 100% saturation with atmospheric oxygen at 37 °C. To measure the direct effect of 3-BrPA concentrations on mitochondrial respiration, HepG2 cells were removed from culture dishes through trypsinization and approx. 2.5–5×106 cells were added to a respiration medium containing 0.25 M mannitol, 0.1% fatty-acid-free BSA, 10 mM MgCl2 and 10 mM KH2PO4 (pH 7.2) as previously described [29], followed by the sequential addition of 0.005% digitonin, 10 mM succinate (complex II substrate), 100 μM ADP (ATP synthesis substrate), 0.05–5 mM 3-BrPA and 1 μM FCCP (carbonyl cyanide p-trifluoromethoxyphenylhydrazone).

For measurements of mitochondrial respiration following treatment with 3-BrPA, HepG2 cells were pre-incubated with 150 μM 3-BrPA for 30 min at 37 °C, with GM or GFM. After the incubation period, media were removed and cells were either suspended in the same culture medium used for the pre-incubation (for measurements of oxygen consumption of intact cells), or in the respiration medium described above (for evaluation of respiratory complexes of permeabilized cells). Basal consumption, oligomycin-independent respiration (proton leak) and FCCP-stimulated respiration (maximum respiration) were assessed in intact HepG2 cells as described previously [30]. 3-BrPA effects on respiratory complexes of 0.005% digitonin-permeabilized HepG2 cells were performed following the addition of different substrates/modulators as 10 mM pyruvate+10 mM malate (complex I-linked substrates), 10 mM succinate, 1 μM rotenone, 50 μM ADP, 1 μg/ml oligomycin and 1 μM FCCP, or titration from 250–750 nM FCCP.

Protein determination

The protein concentration in the samples was determined as described by Lowry et al. [31].

Statistical analysis

Statistical analyses were performed using Origin® 7.5 (OriginLab). All results are expressed as means±S.E.M. for n independent experiments. Statistical significance was determined using a Student's t test. Differences were considered statistically significant for P<0.05 or P<0.01.

RESULTS

3-BrPA affects HepG2 viability in a dose- and time-dependent manner

Figure 1(A) shows that cell incubation with 10–100 μM 3-BrPA for up to 60 min caused a small 10% decrease in viability. On the other hand, 50 μM 3-BrPA incubation for 180 min caused a significant 50% decrease in HepG2 cell viability (P<0.05). Higher concentrations of 3-BrPA (1–5 mM) were effective to promote a more than 30% reduction in viability at all times tested (P<0.05). These results were confirmed by a Trypan Blue exclusion assay (Figure 1B) and suggest that, depending on the concentration and period of exposure to 3-BrPA, this compound may act on different intracellular targets with distinct affinities and reactivity.

Time course of the effects of 3-BrPA on HepG2 cell viability

Figure 1
Time course of the effects of 3-BrPA on HepG2 cell viability

HepG2 cells in culture were incubated with different concentrations of 3-BrPA (50–5000 μM) for 30 (●), 60 (○), 90 (■) and 180 (Δ) min. Following the incubation procedure, cell viability was accessed by MTT assay (A) or Trypan Blue exclusion assay (B) as described in the Experimental section. Values represent means±S.E.M. (n=5).

Figure 1
Time course of the effects of 3-BrPA on HepG2 cell viability

HepG2 cells in culture were incubated with different concentrations of 3-BrPA (50–5000 μM) for 30 (●), 60 (○), 90 (■) and 180 (Δ) min. Following the incubation procedure, cell viability was accessed by MTT assay (A) or Trypan Blue exclusion assay (B) as described in the Experimental section. Values represent means±S.E.M. (n=5).

3-BrPA affects lactate production by HepG2 cells

Glucose availability is an important factor controlling energy production, which is important for cancer cell survival. It has been proposed that 3-BrPA inhibits glycolytic flux and that this inhibition is associated with a decrease in cell viability [11,14]. To check whether this is the case for HepG2 cells, lactate production was followed in cells incubated with 150 μM 3-BrPA for up to 60 min (Figure 2), a condition that caused a 25% decrease in cell viability (Figure 1). In order to also evaluate the contribution of the utilization of endogenous substrates, such as glycogen (bypassing HK reaction), on lactate production, cells were incubated with glucose-supplemented (GM-cells) or glucose-free (GFM-cells) medium (Figure 2). As expected, GFM-cells presented a 16-fold decrease (P<0.05) in the rate of lactate formation compared with GM-cells (Figures 2A and 2B). Treatment of GM-cells with 3-BrPA led to a 3-fold decrease (P<0.05) in lactate production, whereas the absence of glucose in the medium resulted in a 1.5-fold decrease in lactate production. The addition of antimycin A, an inhibitor of oxidative phosphorylation, forces cells to rely only on glycolysis for ATP production. In this situation, it is expected that lactate formation is stimulated, mimicking the Pasteur effect [32]. Indeed, addition of antimycin A to GM-cells caused a 3-fold increase (P<0.05) in lactate production (Figure 2A), indicating that cells were already near maximum glycolytic capacity. An intriguing finding was that no net increase in lactate production was observed when antimycin A was added to GM-cells treated with 3-BrPA (Figure 2A). Figure 2(B) shows a magnification of results presented in Figure 2(A) of the effects of 3-BrPA and antimycin A on lactate production by GFM-cells. In contrast with GM-cells, antimycin A induced an 8.5-fold increase (P<0.05) in the rate of lactate production when glucose was withdrawn from the medium (Figure 2B). An interesting finding was that, even in the presence of antimycin A, no increase in lactate production was observed after treatment with 3-BrPA (Figure 2B).

Lactate production by HepG2 cells in the presence of 3-BrPA

Figure 2
Lactate production by HepG2 cells in the presence of 3-BrPA

(A) HepG2 cells in culture were incubated in the absence (open symbols) or in the presence (closed symbols) of 150 μM 3-BrPA in glucose-supplemented medium (GM-cells) (circles) or in glucose-free medium (GFM-cells) (triangles) for up to 60 min as described in the Experimental section. At 30 min of incubation (indicated by arrows) 2 μg/ml antimycin A (Ant A) was added to the culture. Medium samples were removed at the times indicated and used to measure lactate production. (B) Results obtained in the absence of glucose presented in (A), showing a magnification of the antimycin A effect on lactate production. Values represent means±S.E.M. (n=4). Broken lines show lactate production if antimycin A had not been added.

Figure 2
Lactate production by HepG2 cells in the presence of 3-BrPA

(A) HepG2 cells in culture were incubated in the absence (open symbols) or in the presence (closed symbols) of 150 μM 3-BrPA in glucose-supplemented medium (GM-cells) (circles) or in glucose-free medium (GFM-cells) (triangles) for up to 60 min as described in the Experimental section. At 30 min of incubation (indicated by arrows) 2 μg/ml antimycin A (Ant A) was added to the culture. Medium samples were removed at the times indicated and used to measure lactate production. (B) Results obtained in the absence of glucose presented in (A), showing a magnification of the antimycin A effect on lactate production. Values represent means±S.E.M. (n=4). Broken lines show lactate production if antimycin A had not been added.

3-BrPA targets HepG2 cell glycolysis: inhibition of GAPDH and 3-PGK, but not of HK

Since it has been demonstrated that 3-BrPA has an important inhibitory effect on mt-HK type II of tumour cells [11], we evaluated whether mt-HK inhibition of HepG2 cells would be responsible for the observed effects on lactate production. The direct effect of 3-BrPA on mt-HK activity is shown in Figure 3(A). Unexpectedly, only a 20% decrease in mt-HK activity was observed when cells were incubated for 20 min with 3-BrPA at concentrations as high as 5 mM (P<0.05) (Figure 3A). To check whether the inhibition of mt-HK activity was dependent on the time of incubation with 3-BrPA, G6P production was followed for up to 60 min (Figure 3B). The results showed that 150 μM 3-BrPA did not affect the time course of the reaction and 5 mM of the compound promoted a 30% inhibition (3.01±0.3 to 2.1±0.1 μmol of G6P/mg of protein) of mt-HK activity (P<0.05). Since 3-BrPA reacts with SH or/and OH groups of many enzymes [610], differences in the reactivity of these groups in HK from distinct tissues could explain the differences between the results of the present study and those shown in the literature [11]. For that reason, the inhibition pattern of different HK isoforms by 3-BrPA (mt-HK I and GK) was evaluated. It becomes apparent that GK presented higher sensitivity for 3-BrPA than mt-HK I or II (Figure 3A). The half-maximum inhibition of GK was obtained with 300–400 μM 3-BrPA, whereas for the other two HK isoforms, 5 mM 3-BrPA only slightly affected their activity (20% inhibition) (P<0.05). These results indicate that mt-HK II from HepG2 cells is unlikely to be an important target of 3-BrPA at the conditions that promoted inhibition of the glycolytic flux (Figure 2).

Effects of 3-BrPA on mt-HK II activity and on the recovery of glycolytic enzyme activities in HepG2 cells

Figure 3
Effects of 3-BrPA on mt-HK II activity and on the recovery of glycolytic enzyme activities in HepG2 cells

(A) Activities of mt-HK type I from mouse brain (▲), type II from HepG2 cells (○) and GK from mouse liver (●) were measured in the presence of different concentrations of 3-BrPA (0.01–5 mM) as described in the Experimental section. (B) Time course of mt-HK II activity in the absence (○), 150 μM (▲) or 5 mM (●) 3-BrPA. (C) Cells in culture were pre-incubated with 150 μM 3-BrPA for 30 min and the activities of HK, PGI, PFK, GAPDH, PGK and PK were measured as described in the Experimental section. Open bars, control cells; closed bars, cells pre-incubated with 3-BrPA. The relative activity of each enzyme is related to the control condition. The maximal activities (m-units) of controls were: HK=33±4; PGI=1040±27; PFK=35±3; GAPDH=515±41; PGK=637±15; and PK=654±39. Values represent means±S.E.M. (n=4).

Figure 3
Effects of 3-BrPA on mt-HK II activity and on the recovery of glycolytic enzyme activities in HepG2 cells

(A) Activities of mt-HK type I from mouse brain (▲), type II from HepG2 cells (○) and GK from mouse liver (●) were measured in the presence of different concentrations of 3-BrPA (0.01–5 mM) as described in the Experimental section. (B) Time course of mt-HK II activity in the absence (○), 150 μM (▲) or 5 mM (●) 3-BrPA. (C) Cells in culture were pre-incubated with 150 μM 3-BrPA for 30 min and the activities of HK, PGI, PFK, GAPDH, PGK and PK were measured as described in the Experimental section. Open bars, control cells; closed bars, cells pre-incubated with 3-BrPA. The relative activity of each enzyme is related to the control condition. The maximal activities (m-units) of controls were: HK=33±4; PGI=1040±27; PFK=35±3; GAPDH=515±41; PGK=637±15; and PK=654±39. Values represent means±S.E.M. (n=4).

In order to search for the 3-BrPA target in HepG2 cell glycolysis, the activities of different enzymes from the ATP-consuming and ATP-producing phases of this pathway were measured. After incubation of HepG2 cells with 150 μM 3-BrPA for 30 min, the recovered activities of HK, PGI and PFK were not altered by the treatment. On the other hand, activities of the enzymes involved in the glycolytic ATP synthesis, GAPDH, PGK and PK, were significantly modified (Figure 3C). The activities of GAPDH and PGK were strongly suppressed by 3-BrPA treatment (90 and 70% of control activities respectively, P<0.01). Conversely, the recovery of PK activity was almost twice that of control (P<0.01) (Figure 3C).

3-BrPA treatment affects the glycolytic metabolite profile

According to the results showing the inhibition of specific glycolytic enzymes (Figure 3C), it can be assumed that 3-BrPA treatment may compromise glycolysis downstream of the glyceraldehyde 3-phosphate formation step. To further characterize this metabolic failure, the intermediates of glycolysis in HepG2 cells treated with 3-BrPA were analysed non-invasively by NMR spectroscopy. For this purpose, 5 mM D-[U-13C]glucose was added to GFM in the presence or in the absence of 150 μM 3-BrPA for 30 min. The 13C-NMR spectra of the cells revealed that the assigned glycolytic metabolites such as sugar phosphates (fructose 1,6-bisphosphate, ribose 5-phosphate, G6P, fructose 6-phosphate, glucose 1-phosphate and 6-phosphoglucoronate) and triose phosphates (glyceraldehyde 3-phosphate and dihidroxyacetone phosphate) (see 13C chemical shift from 66 to 105 p.p.m. in Figure 4A) are clearly presented in large excess after treatment with 3-BrPA. Additionally, it is evident that [13C]2-phosphoglycerate derived from D-[U-13C]glucose is decreased in 3-BrPA-treated HepG2 cells (Figure 4A). These spectral profiles are in agreement with the inhibition of the glycolytic steps involved in ATP synthesis, i.e. GAPDH and PGK (Figure 3C), indicating that these enzymes are the main targets of 3-BrPA in HepG2 cells. This hypothesis was reinforced by the decreased levels of 2-acetolactate and amino acids (L-aspartate, L-asparagine, L-alanine and L-glutamate) (see 13C chemical shift from 30 to 55 p.p.m. in Figure 4A). Unexpectedly, higher amounts of intracellular L-lactate was observed in HepG2 cells treated with 3-BrPA (Figure 4A). The inhibition of lactate production (and export) by 3-BrPA in HepG2 cells was further confirmed using NMR spectroscopy analysis (Figure 4B). The most abundant 13C-labelled metabolite detected in the culture medium was [13C]lactate, which steadily accumulated.

NMR spectroscopy analysis of HepG2 cell metabolites treated with 3-BrPA

Figure 4
NMR spectroscopy analysis of HepG2 cell metabolites treated with 3-BrPA

HepG2 cells in culture were pre-incubated with 150 μM 3-BrPA for 30 min. (A) Spectra highlight the enrichment of informative metabolic intermediates. Top spectrum, control cells; bottom spectrum, cells pre-incubated with 3-BrPA. (B) Culture medium samples were removed at the times indicated and subjected to [13C]lactate determination. The numbers indicate specific 13C-containing chemicals of the following intermediates: (1) fructose 1,6-bisphosphate, (2) ribose 5-phosphate, (3) G6P, (4) glyceraldehyde 3-phosphate, (5) dihidroxyacetone phosphate, (6) fructose 6-phosphate, (7) glucose 1-phosphate, (8) 6-phosphoglucoronate, (9) 2-phosphoglycerate, (10) L-aspartate, (11) L-asparagine, (12) L-alanine, (13) L-glutamate, (14) L-lactate, (15) 2-acetolactate.

Figure 4
NMR spectroscopy analysis of HepG2 cell metabolites treated with 3-BrPA

HepG2 cells in culture were pre-incubated with 150 μM 3-BrPA for 30 min. (A) Spectra highlight the enrichment of informative metabolic intermediates. Top spectrum, control cells; bottom spectrum, cells pre-incubated with 3-BrPA. (B) Culture medium samples were removed at the times indicated and subjected to [13C]lactate determination. The numbers indicate specific 13C-containing chemicals of the following intermediates: (1) fructose 1,6-bisphosphate, (2) ribose 5-phosphate, (3) G6P, (4) glyceraldehyde 3-phosphate, (5) dihidroxyacetone phosphate, (6) fructose 6-phosphate, (7) glucose 1-phosphate, (8) 6-phosphoglucoronate, (9) 2-phosphoglycerate, (10) L-aspartate, (11) L-asparagine, (12) L-alanine, (13) L-glutamate, (14) L-lactate, (15) 2-acetolactate.

Rapid (∼20 min) 31P spectra of HepG2 cells were acquired immediately before the 13C spectra acquisition. The spectra showed accumulation of phosphomonoesters in the cells treated with 3-BrPA (results not shown). 31P spectra are consistent with the accumulation of the phosphorylated hexoses detected by 13C NMR.

3-BrPA strongly affects mitochondrial oxygen consumption in HepG2 cells

Although it has been suggested that mitochondrial respiration is affected by 3-BrPA [11], there are no available data in the literature with regard to the mechanisms involved in this process. Direct effects of 3-BrPA on mitochondrial respiration of HepG2 cells were evaluated after cell permeabilization and 3-BrPA titration. Figures 5(A) and 5(B) show that when respiration was supported by succinate, oxygen consumption of digitonin-permeabilized HepG2 cells was strongly inhibited by 3-BrPA addition. When 400 μM of the compound was added, state 3 respiration (ADP-stimulated respiration coupled to ATP synthesis) was completely inhibited (Figures 5A and 5B). The IC50 value for 3-BrPA inhibition was 150 μM (Figure 5B). The inhibition of oxygen consumption by 3-BrPA is probably due to a blockage of electron transport in the ETS (electron transport system), since addition of a proton ionophore (FCCP), that would fully stimulate respiration if the inhibition was at the level of the ADP↔ATP recycling system [ANT (adenine nucelotide transporter)/FoF1-ATP synthase], was not able to increase oxygen consumption (Figure 5A). No effect of the same concentration of 3-BrPA was observed in the azide-sensitive ATPase activity (results not shown).

Titration of mitochondrial oxygen consumption of HepG2 permeabilized cells by 3-BrPA

Figure 5
Titration of mitochondrial oxygen consumption of HepG2 permeabilized cells by 3-BrPA

(A) Representative record of mitochondrial oxygen consumption of HepG2-permeabilized cells under different regimes. Cells were added to a standard respiration buffer and permeabilized by the addition of 0.005% digitonin as described in the Experimental section. Arrows indicate the addition of substrates/modulators as follows: 10 mM succinate, 200 μM ADP and 3-BrPA to obtain the final concentration as indicated, and 1 μM FCCP. (B) Rates of oxygen consumption in the presence of different concentrations of 3-BrPA. Values represent means±S.E.M. (n=4).

Figure 5
Titration of mitochondrial oxygen consumption of HepG2 permeabilized cells by 3-BrPA

(A) Representative record of mitochondrial oxygen consumption of HepG2-permeabilized cells under different regimes. Cells were added to a standard respiration buffer and permeabilized by the addition of 0.005% digitonin as described in the Experimental section. Arrows indicate the addition of substrates/modulators as follows: 10 mM succinate, 200 μM ADP and 3-BrPA to obtain the final concentration as indicated, and 1 μM FCCP. (B) Rates of oxygen consumption in the presence of different concentrations of 3-BrPA. Values represent means±S.E.M. (n=4).

Pre-incubation with 3-BrPA affects mitochondrial respiratory parameters of intact and permeabilized HepG2 cells

In addition to the direct effects of 3-BrPA on mitochondria, the respiratory parameters of HepG2 cells were also assessed after pre-incubation with this compound (Figures 6–8). As for lactate production, cells were incubated with glucose-supplemented (GM-cells) or glucose-free (GFM-cells) medium to analyse possible differences in the effects of 3-BrPA when glycolysis and oxidative phosphorylation were differentially stimulated. Figure 6(A) shows that GFM-cells pre-incubated with 150 μM 3-BrPA for 30 min presented a 50% decrease in basal respiration (3.1±0.19 to 1.6±0.13 nmol of O2·min−1·10−6 cells) and FCCP-stimulated maximum respiration (3.5±0.22 to 1.9±0.3 nmol of O2·min−1·10−6 cells) (P<0.01). Additionally, no difference was observed in the respiratory rate not coupled to ATP synthesis, the proton leak (oligomycin-independent respiration) (Figure 6A). A different situation was observed in GM-cells (Figure 6B), where 3-BrPA caused only a 20% inhibition of basal respiration (2.1±0.12 to 1.6±0.1 nmol of O2·min−1·10−6 cells) (P<0.05) and no difference was detected on FCCP-stimulated maximum respiration. Curiously, despite the apparent small effect of 3-BrPA treatment on the basal respiration of GM-cells, oligomycin-independent respiration was almost 2-fold higher than that measured on control cells (0.6±0.1 to 1.1±0.1 nmol of O2·min−1·10−6 cells) (P<0.05) (Figure 6B). These results indicate that, depending on the carbon source utilized, 3-BrPA has distinct effects on the energy metabolism of HepG2 cells.

Effect of pre-incubation with 3-BrPA on oxygen consumption of intact HepG2 cells

Figure 6
Effect of pre-incubation with 3-BrPA on oxygen consumption of intact HepG2 cells

HepG2 cells in culture were pre-incubated with 150 μM 3-BrPA for 30 min in the absence (GFM-cells) (A) or presence (GM-cells) (B) of glucose as described in the Experimental section. Oxygen consumption was measured in the same culture medium used in the incubation step without FBS. The histogram represents the rates of oxygen consumption under different regimes (basal respiration, BASAL; FCCP-induced maximum respiration, 1 μM FCCP; and oligomycin-independent respiration, 1 μg/ml OLIGO). Open bars, control cells; closed bars, cells pre-incubated with 3-BrPA. Values represent means±S.E.M. (n=5). **P<0.01 and *P<0.05, significantly different from control cells.

Figure 6
Effect of pre-incubation with 3-BrPA on oxygen consumption of intact HepG2 cells

HepG2 cells in culture were pre-incubated with 150 μM 3-BrPA for 30 min in the absence (GFM-cells) (A) or presence (GM-cells) (B) of glucose as described in the Experimental section. Oxygen consumption was measured in the same culture medium used in the incubation step without FBS. The histogram represents the rates of oxygen consumption under different regimes (basal respiration, BASAL; FCCP-induced maximum respiration, 1 μM FCCP; and oligomycin-independent respiration, 1 μg/ml OLIGO). Open bars, control cells; closed bars, cells pre-incubated with 3-BrPA. Values represent means±S.E.M. (n=5). **P<0.01 and *P<0.05, significantly different from control cells.

To analyse in more detail the effects of 3-BrPA on mitochondrial function and ETS complex activities, oxygen consumption was measured on digitonin-permeabilized cells using high-resolution respirometry (Figures 7 and 8). Following incubation with 150 μM 3-BrPA for 30 min, GFM-cells presented a significant decrease in complex I (7.8±1.4 to 3.8±0.6 pmol of O2·s−1·10−6 cells) and succinate-driven respiration (54±9.3 to 29.6±3.5 pmol of O2·s−1·10−6 cells) (P<0.05) (Figures 7C and 8A). Total electron transport capacity after FCCP titrations was significantly decreased (84.6±8.3 to 44.4±9 pmol of O2·s−1·10−6 cells) (P<0.05) (Figures 7C and 8A), whereas oxygen flow related to proton leak was not significantly altered (Figures 7C and 8A), as observed with intact cells (Figure 6A). Different effects of 3-BrPA treatment on respiratory parameters were observed in GM-cells when compared with GFM-cells (Figures 7D and 8B). 3-BrPA was able to significantly decrease only the oxygen flow induced by succinate (35.7±5.7 to 15±4.4 pmol of O2·s−1·10−6 cells) (P<0.05). In a similar manner to the respiratory flow observed for intact GM-cells (Figure 6B), 3-BrPA did not affect maximum respiration flow induced by FCCP, and promoted a 3-fold increase in proton leak respiration of permeabilized cells (4.4±0.95 to 14.2±4.2 pmol of O2·s−1·10−6 cells) (P<0.05) (Figures 7D and 8B). Moreover, the respiratory control induced by ADP was delayed after 3-BrPA treatment in GFM-cells either with pyruvate/malate or succinate as respiratory substrates (Figure 7C); however, in GM-cells, 3-BrPA treatment did not change the response to ADP when pyruvate/malate was used to induce respiration, but the respiratory control induced by ADP was slowed when succinate was the respiratory substrate (Figure 7D).

Representative records of oxygen concentration (A and B) or oxygen flow (C and D) of permeabilized HepG2 cells pre-incubated with 3-BrPA

Figure 7
Representative records of oxygen concentration (A and B) or oxygen flow (C and D) of permeabilized HepG2 cells pre-incubated with 3-BrPA

HepG2 cells in culture were pre-incubated with 150 μM 3-BrPA for 30 min in the absence (GFM-cells) (A and C) or presence (GM-cells) (B and D) of glucose as described in the Experimental section. Oxygen consumption was measured using high-resolution respirometry on standard respiration buffer after the addition of cells and 0.005% digitonin. Arrows indicate the addition of substrates/modulators as follows: 10 mM pyruvate+malate (Pyr/Mal), 50 μM ADP, 1 μM rotenone (Rot), 10 mM succinate (Succ), 1 μg/ml oligomycin (Oligo) and FCCP as indicated. Continuous traces, control cells; broken traces, cells pre-incubated with 3-BrPA.

Figure 7
Representative records of oxygen concentration (A and B) or oxygen flow (C and D) of permeabilized HepG2 cells pre-incubated with 3-BrPA

HepG2 cells in culture were pre-incubated with 150 μM 3-BrPA for 30 min in the absence (GFM-cells) (A and C) or presence (GM-cells) (B and D) of glucose as described in the Experimental section. Oxygen consumption was measured using high-resolution respirometry on standard respiration buffer after the addition of cells and 0.005% digitonin. Arrows indicate the addition of substrates/modulators as follows: 10 mM pyruvate+malate (Pyr/Mal), 50 μM ADP, 1 μM rotenone (Rot), 10 mM succinate (Succ), 1 μg/ml oligomycin (Oligo) and FCCP as indicated. Continuous traces, control cells; broken traces, cells pre-incubated with 3-BrPA.

Rates of oxygen consumption of permeabilized HepG2 cells pre-incubated with 3-BrPA

Figure 8
Rates of oxygen consumption of permeabilized HepG2 cells pre-incubated with 3-BrPA

HepG2 cells in culture were pre-incubated with 150 μM 3-BrPA for 30 min in the absence (GFM-cells) (A) or presence (GM-cells) (B) of glucose as described in the Experimental section. Oxygen consumption was measured using high-resolution respirometry on standard respiration buffer after the addition of cells and 0.005% digitonin. The following substrate and/or modulator concentrations were added: 10 mM pyruvate+malate (Pyr/Mal), 50 μM ADP, 1 μM rotenone (Rot), 10 mM succinate (Succ), 1 μg/ml oligomycin (Oligo) and 1 μM FCCP. St4(CI) and St4(CII) mean state 4 obtained when complex I (Pyr/Mal) or complex II (Succ) were stimulated respectively. Open bars, control cells; closed bars, cells pre-incubated with 3-BrPA. Values represent means±S.E.M. (n=5). *P<0.05, significantly different from control cells.

Figure 8
Rates of oxygen consumption of permeabilized HepG2 cells pre-incubated with 3-BrPA

HepG2 cells in culture were pre-incubated with 150 μM 3-BrPA for 30 min in the absence (GFM-cells) (A) or presence (GM-cells) (B) of glucose as described in the Experimental section. Oxygen consumption was measured using high-resolution respirometry on standard respiration buffer after the addition of cells and 0.005% digitonin. The following substrate and/or modulator concentrations were added: 10 mM pyruvate+malate (Pyr/Mal), 50 μM ADP, 1 μM rotenone (Rot), 10 mM succinate (Succ), 1 μg/ml oligomycin (Oligo) and 1 μM FCCP. St4(CI) and St4(CII) mean state 4 obtained when complex I (Pyr/Mal) or complex II (Succ) were stimulated respectively. Open bars, control cells; closed bars, cells pre-incubated with 3-BrPA. Values represent means±S.E.M. (n=5). *P<0.05, significantly different from control cells.

SDH as a target of 3-BrPA in the mitochondria of HepG2 cells

The results presented in Figures 7 and 8 point to a preferential inhibition of mitochondrial respiratory complex II (SDH) activity after 3-BrPA treatment in both GM- and GFM-cells. To check this hypothesis, the direct effect of 3-BrPA on SDH activity in the HepG2 mitochondrial fraction was measured (Figure 9A). Accordingly, the results confirmed that SDH from HepG2 cells was inhibited by 3-BrPA, presenting an IC50 value between 10 and 20 μM.

Titration and threshold curves of SDH activity in HepG2 mitochondria

Figure 9
Titration and threshold curves of SDH activity in HepG2 mitochondria

(A) 3-BrPA titration curve. SDH activity was measured by a spectrophotometric method using DCPIP as an artificial electron acceptor as described in the Experimental section. The reaction was started with the addition of HepG2 mitochondrial fraction (0.01 mg/ml) in the absence or 2.5–300 μM of 3-BrPA. Values are expressed as a percentage of control activity. Control activity: 9.1±1.1 nmol of DCPIP·min−1·mg of protein−1. (B) Threshold curves. The percentage of respiratory rate as a function of the percentage of SDH inhibition by 3-BrPA. Each point comes from the experimental titration curves and represents the percentage of the respiratory rate as a function of the percentage of the SDH activity for the same 3-BrPA concentration.

Figure 9
Titration and threshold curves of SDH activity in HepG2 mitochondria

(A) 3-BrPA titration curve. SDH activity was measured by a spectrophotometric method using DCPIP as an artificial electron acceptor as described in the Experimental section. The reaction was started with the addition of HepG2 mitochondrial fraction (0.01 mg/ml) in the absence or 2.5–300 μM of 3-BrPA. Values are expressed as a percentage of control activity. Control activity: 9.1±1.1 nmol of DCPIP·min−1·mg of protein−1. (B) Threshold curves. The percentage of respiratory rate as a function of the percentage of SDH inhibition by 3-BrPA. Each point comes from the experimental titration curves and represents the percentage of the respiratory rate as a function of the percentage of the SDH activity for the same 3-BrPA concentration.

The SDH inhibition profile may not be strictly correlated with oxidative phosphorylation inhibition and, in fact, may vary among different cell types [33]. This phenomenon of biochemical threshold can be demonstrated by threshold curves, which may indicate either a deficiency or an ‘excess of enzyme activity’ or a ‘buffering effect by the metabolic network’ of a given complex of oxidative phosphorylation responses. Figure 9(B) shows a threshold curve of the rates of oxygen consumption and the degree of SDH inhibition attained by 3-BrPA. It becomes apparent that at the level of 50% inhibition of SDH activity, no significant decrease in HepG2 mitochondrial respiration was observed; however, above this degree of SDH inhibition, a large fall in the respiratory rate was detected. More importantly, at 150 μM 3-BrPA, which corresponds to 70% inhibition of SDH activity, respiration was inhibited in 60% (Figure 9B).

DISCUSSION

HCC is an important health problem, being one of the most common fatal cancers worldwide [34]. The search for an efficient therapy led to the proposal of a new approach which consists of the direct intra-arterial administration of 3-BrPA to the tumour. The treatment with 3-BrPA eradicates HCC without affecting healthy surrounding tissues of the host [13,14]. The beneficial properties of this agent rely on its capacity to inhibit cellular energy machinery systems, i.e. glycolysis and oxidative phosphorylation [11]; however, the understanding of the specific metabolic steps and mechanisms of 3-BrPA action in human HCC cell lines, particularly its effects on mitochondria bioenergetics, are very limited. In the present study, we could establish the time course of 3-BrPA effects on the viability of HepG2 cells, which indicated that, depending on the concentration, period of incubation and the substrates oxidized by the cells, 3-BrPA affects different metabolic pathways.

Cancer cells are believed to present an accelerated glycolysis, even in the presence of sufficient amounts of oxygen [1,2]. HepG2 cells follow this pattern, since only 3-fold stimulation on lactate production was observed when oxidative phosphorylation was inhibited by antimycin A addition (Figure 2A). In non-transformed cell lines, the stimulatory effect on lactate production by antimycin A may be as high as 10-fold [32]. In the presence of 3-BrPA, the amount of lactate produced was significantly decreased, indicating an inhibitory effect on glycolysis (Figure 2A). The 70% inhibition of glycolysis could not be explained by mt-HK inhibition (Figures 2A, 3A and 3B). Other candidate target enzymes for 3-BrPA action could be GAPDH and PGK, which, in a previous study, were immediately inhibited by 3-BrPA in a competitive manner [35]. Indeed, in the present work we demonstrated that treatment of HepG2 cells with 150 μM 3-BrPA by short periods of incubation is sufficient to promote an irreversible inhibition of both GAPDH and PGK activities by more than 70% (Figure 3C), despite the fact that cell viability is almost 100% (Figure 1). The inhibition of GAPDH and PGK activities dramatically altered the steady-state levels of glycolytic intermediates in HepG2 cells (Figure 4A). It is remarkable that the ATP-consuming reactions of glycolysis (HK and PFK) were not affected by 3-BrPA, but the downstream ATP-producing steps of this pathway (GAPDH and PGK), together with oxidative phosphorylation, were severely inhibited (Figures 3, 4 and 8). The glycolytic ATP-consuming steps (i.e. HK and PFK) would facilitate the bioenergetic collapse of tumour cells by accelerating the intracellular ATP depletion. It should be mentioned that, besides the bioenergetic alterations, 3-BrPA-treated cells showed increased levels of intracellular lactate (Figure 4A) which would affect the internal pH of the cell. The reasons for this increase in intracellular lactate are not yet understood, but it might be speculated that lactate transport across cellular membranes may be affected by 3-BrPA treatment, since this compound shares the same transporter as lactate.

Since no stimulation of glycolysis was observed following the addition of antimycin A, the rate of lactate production by HepG2 cells appeared to be on its maximum flux, which indicates that 3-BrPA also affects mitochondrial function. This is more clearly observed when lactate production was measured in GFM-cells (Figure 2). In this situation, the carbon sources for lactate synthesis are glycogen and mainly glutamine oxidation, as demonstrated previously [36]. Since replacement of glucose by glutamine induces a maximum stimulation of oxidative phosphorylation, the observed decrease in lactate formation due to 3-BrPA treatment indicates a decrease in glutaminolysis and, ultimately, a decreased mitochondrial oxidative capacity. In addition, one can hypothesize that inhibition of glutamate dehydrogenase [37], SDH [6] and malic enzyme [9], all of them involved in lactate production from glutamine by 3-BrPA, could also contribute to the inhibitory effects shown in Figure 2(B).

The results presented in Figures 5–8 show in more detail the effects of 3-BrPA on mitochondrial respiratory parameters. Pre-incubation with 3-BrPA caused a significant decrease in the rate of oxygen consumption either in intact (Figure 6) or permeabilized (Figures 7 and 8) HepG2 cells, indicating that 3-BrPA may be reacting irreversibly on SH groups of the ETS under cell-culture conditions, in accordance with previous in vitro assays shown for both SDH and PDH (pyruvate dehydrogenase) [6,10]. Unexpectedly, the degree of inhibition of respiration by 3-BrPA depends on the carbon source supplied to cells (Figures 2 and 6–8), glucose or glutamine, both of which are rapidly consumed during proliferation of tumours. Accordingly, in results with GFM-cells, we observed significant alterations in mitochondrial bioenergetics. The inhibition of ETS by 3-BrPA may be related to an imbalance of cellular redox state and to the absence of substrates, turning the cellular enzymes more reactive to 3-BrPA [7,10]. On the other hand, in the presence of glucose (GM-cells), the inhibition of complex-I linked substrates (pyruvate/malate) was not apparent (Figures 7D and 8B), indicating the protective effect of this substrate. Despite the distinct effects on complex-I respiratory activity, it is important to emphasize that, under both conditions, complex II-driven respiration, i.e. SDH activity, was significantly reduced (Figures 7C, 7D and 8).

An interesting finding was that the proton leak was increased in the GM-cells (Figures 6B and 8B). This proton leak may reflect the permeability transition pore formed in the mitochondrial membrane, as observed in apoptosis or necrosis [384141]. One possibility is that the inhibition of glycolysis at the levels of the GAPDH and PGK reactions (Figure 3C) increased the intracellular levels of G6P (Figure 4A), which leads to mt-HK dissociation from the mitochondrial membrane and favours the binding of pro-apoptotic proteins to VDACs (voltage-dependent anion channels) [42]. In fact, the results from 13C-NMR spectra show that the level of G6P is higher in HepG2-treated cells than in control cells (Figure 4A). Actually, it was demonstrated that 3-BrPA promotes the dissociation of mt-HK from VDACs derived from HCC Huh-BAT and MH134 cells [16]. On the other hand, one can speculate that, in GFM-cells, mt-HK dissociation does not occur since G6P levels are low (Figure 4A) and proton leak is decreased (Figures 6–8). Additionally, we cannot exclude the possibility that a reaction of 3-BrPA with crucial cysteine residues (e.g. Cys56) located at the ANT may lead to the formation of a permeability transition pore [43], contributing to the increase in proton leak in GM-cells.

Analysing the effects of 3-BrPA on other HK isoforms, the results of the present study showed that GK, the major isoform in normal liver, was much more sensitive to 3-BrPA inhibition (Figure 3A) when compared with the mitochondrial isoforms found in tumour cells. GK is characterized by a high susceptibility to SH group oxidation [44] and a strong sensitivity towards SH group reactive reagents [45,46]. The SH group of the cysteine residues has been shown to modulate enzyme activity [47]. In contrast with GK, low-Km HKs show an approx. 100-fold lower sensitivity towards inhibition of enzyme activity by SH group reagents, such as alloxan [46]. The results in Figure 3(A) show a weaker inhibition by 3-BrPA treatment of HepG2 mt-HK II when compared with mice liver GK. Thus it may be concluded that 3-BrPA acts like a SH group reagent following a very similar pattern to those SH group reagents described previously for GK and low-Km HKs [47]. Previous results have shown that 3-BrPA almost completely inhibits mt-HK II from HCC [11]. The apparent contrasting results shown in Figure 3 and those described previously [11] may be explained, in part, by the protocol used to assay mt-HK activities in the present study, in which the reactions were started by the addition of enzymes in a medium containing ATP, glucose and 3-BrPA. In the previous study, 5 mM 3-BrPA was added to mt-HK preparations before the addition of glucose [11]. It had been shown that cysteine residues of the catalytic site of GK and mt-HK II are much less sensitive to SH group reagents while in contact with their substrates [47].

In conclusion, it seems that 3-BrPA may act on several cellular enzymes and, based on the threshold inhibition curve and lactate production (Figures 2, 4B and 9B) we propose that GAPDH, PGK and SDH are important targets of this compound. Therefore the mutual effects of 3-BrPA on glycolysis and mitochondria may be responsible for the decrease in viability and death of HCC cells. We believe that this information will help to design new analogues and compounds that might contribute to more effective treatment of HCC.

We are grateful to Dr Erich Gnaiger for kind technical advice with regard to the use of the high-resolution respirometry in O2K Oroboros. We also would like to thank the unknown reviewers for their valuable comments in improving the paper.

Abbreviations

     
  • ANT

    adenine nucelotide transporter

  •  
  • 3-BrPA

    3-bromopyruvate

  •  
  • DCPIP

    2,6-dichlorophenol-indophenol

  •  
  • DMEM

    Dulbecco's modified Eagle's medium

  •  
  • DTT

    dithiothreitol

  •  
  • ETS

    electron transport system

  •  
  • FBS

    foetal bovine serum

  •  
  • FCCP

    carbonyl cyanide p-trifluoromethoxyphenylhydrazone

  •  
  • GAPDH

    glyceraldehyde-3-phosphate dehydrogenase

  •  
  • GFM

    glucose-free medium

  •  
  • GK

    glucokinase

  •  
  • GM

    GFM supplemented with 5 mM glucose or 5 mM D-[U-13C]glucose

  •  
  • G6P

    glucose 6-phosphate

  •  
  • HCC

    hepatocellular carcinoma

  •  
  • HK

    hexokinase

  •  
  • MEM

    minimal essential medium

  •  
  • mt-HK

    mitochondrial HK

  •  
  • MTT

    3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide

  •  
  • PFK

    phosphofructokinase

  •  
  • PGI

    phosphoglucose isomerase

  •  
  • PGK

    phosphoglycerate kinase

  •  
  • PK

    pyruvate kinase

  •  
  • SDH

    succinate dehydrogenase

  •  
  • VDAC

    voltage-dependent anion channel

FUNDING

This work was supported by the Conselho Nacional de Desenvolvimento Científico e Tecnológico [grant number 480378/2004-5] and the Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro [grant number E-26/111.911/2008].

References

References
1
Warburg
O.
On respiratory impairment in cancer cells
Science
1956
, vol. 
124
 (pg. 
269
-
270
)
2
Matoba
S.
Kang
J. G.
Patino
W. D.
Wragg
A.
Boehm
M.
Gavrilova
O.
Hurley
P. J.
Bunz
F.
Hwang
P. M.
p53 regulates mitochondrial respiration
Science
2006
, vol. 
312
 (pg. 
1650
-
1653
)
3
Costantini
P.
Jacotot
E.
Decaudin
D.
Kroemer
G.
Mitochondrion as a novel target of anticancer chemotherapy
J. Natl. Cancer Inst.
2000
, vol. 
92
 (pg. 
1042
-
1053
)
4
Moreno-Sánchez
R.
Rodríguez-Enríquez
S.
Marín-Hernández
A.
Saavedra
E.
Energy metabolism in tumor cells
FEBS J.
2007
, vol. 
274
 (pg. 
1393
-
1418
)
5
Pedersen
P. L.
The cancer cell's ‘power plants’ as promising therapeutic targets: an overview
J. Bioenerg. Biomembr.
2007
, vol. 
39
 (pg. 
1
-
12
)
6
Sanborn
B. M
Felberg
N. T
Hollocher
T. C.
The inactivation of succinate dehydrogenase by bromopyruvate
Biochim. Biophys. Acta
1971
, vol. 
227
 (pg. 
219
-
231
)
7
Chang
G. G.
Hsu
R. Y.
Mechanism of pigeon liver malic enzyme: kinetics, specificity, and half-site stoichiometry of the alkylation of a cysteinyl residue by the substrate-inhibitor bromopyruvate
Biochemistry
1977
, vol. 
16
 (pg. 
311
-
320
)
8
Tunnicliff
G.
Ngo
T. T.
Mechanism of inactivation of brain glutamic decarboxylase by 3-bromopyruvate
Int. J. Biochem.
1978
, vol. 
9
 (pg. 
249
-
252
)
9
Satterlee
J.
Hsu
R. Y.
Duck liver malic enzyme: sequence of a tryptic peptide containing the cysteine residue labeled by the substrate analog bromopyruvate
Biochim. Biophys. Acta
1991
, vol. 
1079
 (pg. 
247
-
252
)
10
Korotchkina
L. G.
Showkat
A. M.
Patel
M. S.
Involvement of α-cysteine-62 and β-tryptophan-135 in human pyruvate dehydrogenase catalysis
Arch. Biochem. Biophys.
1999
, vol. 
369
 (pg. 
277
-
287
)
11
Ko
Y. H.
Pedersen
P. L.
Geschwind
J. F.
Glucose catabolism in the rabbit VX2 tumor model for liver cancer: characterization and targeting hexokinase
Cancer Lett.
2001
, vol. 
173
 (pg. 
83
-
91
)
12
Gwak
G.
Yoon
J.
Kim
K.
Lee
H.
Chung
J.
Gores
G.
Hypoxia stimulates proliferation of human hepatoma cells through the induction of hexokinase II expression
J. Hepatol.
2005
, vol. 
42
 (pg. 
358
-
364
)
13
Geschwind
J. F.
Ko
Y. K.
Torbenson
M. S.
Magee
C.
Pedersen
P. L.
Novel therapy for liver cancer: direct intra-arterial injection of a potent inhibitor of ATP production
Cancer Res.
2002
, vol. 
62
 (pg. 
3909
-
3913
)
14
Ko
Y. H.
Smith
B. L.
Wang
Y.
Pomper
M. G.
Rini
D. A.
Torbenson
M. S.
Hullihen
J.
Pedersen
P. L.
Advanced cancers: eradication in all cases using 3-bromopyruvate therapy to deplete ATP
Biochem. Biophys. Res. Commun.
2004
, vol. 
324
 (pg. 
269
-
275
)
15
Xu
R.
Pelicano
Y. Z
Carew
J. S
Feng
L.
Bhalla
K. N
Keating
M. J.
Huang
P.
Inhibition of glycolysis in cancer cells: a novel strategy to overcome compound resistance associated with mitochondrial respiratory defect and hypoxia
Cancer Res.
2005
, vol. 
65
 (pg. 
613
-
621
)
16
Kim
W.
Yoon
J. H.
Jeong
J. M.
Cheon
G. J.
Lee
T. S.
Yang
J. I.
Park
S. C.
Lee
H. S.
Apoptosis-inducing antitumor efficacy of hexokinase II inhibitor in hepatocellular carcinoma
Mol. Cancer Ther.
2007
, vol. 
6
 (pg. 
2554
-
2562
)
17
Van De Loosdrecht
A. A.
A tetrazolium-based colorimetric MTT assay to quantitate human monocyte mediated cytotoxicity against leukemic cells from cell lines and patients with acute myeloid leukemia
J. Immunol. Methods
1994
, vol. 
174
 (pg. 
311
-
320
)
18
Graham
J. M.
Rickwood
D.
Subcellular Fractionation
A Practical Approach
1997
Oxford
Oxford University Press
19
Maciel
E. N.
Kowaltowski
A. J.
Schwalm
F. D.
Rodrigues
J. M.
Souza
D. O.
Vercesi
A. E.
Wajner
M.
Castilho
R. F.
Mitochondrial permeability transition in neuronal damage promoted by Ca2+ and respiratory chain complex II inhibition
J. Neurochem.
2004
, vol. 
90
 (pg. 
1025
-
1035
)
20
da-Silva
W. S.
Gómez-Puyou
A.
de Gómez-Puyou
M. T.
Moreno-Sanchez
R.
De Felice
F. G.
de Meis
L.
Oliveira
M. F.
Galina
A.
Mitochondrial bound hexokinase activity as a preventive antioxidant defense: steady-state ADP formation as a regulatory mechanism of membrane potential and reactive oxygen species generation in mitochondria
J. Biol. Chem.
2004
, vol. 
279
 (pg. 
39846
-
39855
)
21
Gracy
R. W.
Tilley
B. E.
Phosphoglucose isomerase of human erythrocytes and cardiac tissue
Methods Enzymol.
1975
, vol. 
41
 (pg. 
392
-
400
)
22
Sapico
V.
Anderson
R. L.
Regulation of D-fructose 1-phosphate kinase by potassium ion
J. Biol. Chem.
1970
, vol. 
245
 (pg. 
3252
-
3256
)
23
Molina y Vedia
L.
McDonald
B.
Reep
B.
Brüne
B.
Di Silvio
M.
Billiar
T. R.
Lapetina
E. G.
Nitric oxide-induced S-nitrosylation of glyceraldehyde-3-phosphate dehydrogenase inhibits enzymatic activity and increases endogenous ADP-ribosylation
J. Biol. Chem.
1992
, vol. 
267
 (pg. 
24929
-
24932
)
24
Farrar
G.
Farrar
W. W.
Purification and properties of the pyruvate kinase isozyme M1 from the pig brain
Int. J. Biochem. Cell Biol.
1995
, vol. 
27
 (pg. 
1145
-
1151
)
25
Sola-Penna
M.
dos Santos
A. C.
Alves
G. G.
El-Bacha
T.
Faber-Barata
J.
Pereira
M. F.
Serejo
F. C.
Da Poian
A. T.
Sorenson
M.
A radioassay for phosphofructokinase-1 activity in cell extracts and purified enzyme
J. Biochem. Biophys. Methods
2002
, vol. 
50
 (pg. 
129
-
140
)
26
Robinson
K. M.
Lemire
B. D.
Flavinylation of succinate: ubiquinone oxidoreductase from Saccharomyces cerevisiae
Methods Enzymol.
1995
, vol. 
256
 (pg. 
34
-
51
)
27
Wishart
D. S.
Tzur
D.
Knox
C.
Eisner
R.
Guo
A. C.
Young
N.
Cheng
D.
Jewell
K.
Arndt
D.
Sawhney
S.
, et al. 
HMDB: the human metabolome database
Nucleic Acids Res.
2007
, vol. 
35
 (pg. 
D521
-
D526
)
28
Hamilton
S. D.
Pardue
H.L.
Quantitation of lactate by a kinetic method with an extend range of linearity and low dependence on experimental variables
Clin. Chem.
1984
, vol. 
30
 (pg. 
226
-
229
)
29
Sakurai
K.
Cederbaum
A. I.
Oxidative stress and cytotoxicity induced by ferric-nitrolotriacetate in HepG2 cells that express cytochrome P450 2E1
Mol. Pharmacol.
1998
, vol. 
54
 (pg. 
1024
-
1035
)
30
El-Bacha
T.
Midlej
V.
Pereira da Silva
A. P.
Silva da Costa
L.
Benchimol
M.
Galina
A.
Da Poian
A.T.
Mitochondrial and bioenergetic dysfunction in human hepatic cells infected with dengue 2 virus
Biochim. Biophys. Acta
2007
, vol. 
1772
 (pg. 
1158
-
1166
)
31
Lowry
O. H.
Rosebrough
N. J.
Farr
A. L.
Randall
R. J.
Protein measurement with the Folin phenol reagent
J. Biol. Chem.
1951
, vol. 
193
 (pg. 
265
-
275
)
32
Merlo-Pich
M.
Deleonardi
G.
Biondi
A.
Lenaz
G.
Methods to detect mitochondrial function
Exper. Gerontol.
2004
, vol. 
39
 (pg. 
277
-
281
)
33
Rossignol
R.
Malgat
M.
Mazat
J. P.
Letellier
T.
Threshold effect and tissue specificity. Implication for mitochondrial cytopathies
J. Biol. Chem.
1999
, vol. 
274
 (pg. 
33426
-
33432
)
34
Lopez
A. D.
Mathers
C. D.
Ezzati
M.
Jamison
D. T.
Murray
C. J.
Global and regional burden of disease and risk factors, 2001: systematic analysis of population health data
Lancet
2006
, vol. 
367
 (pg. 
1747
-
1757
)
35
Jones
A. R.
Gillan
L.
Milmlow
D.
The anti-glycolytic activity of 3-bromopyruvate on mature boar spermatozoa in vitro
Contraception
1995
, vol. 
52
 (pg. 
317
-
320
)
36
DeBerardinis
R. J.
Mancuso
A.
Daikhin
E.
Nissim
I.
Yudkoff
M.
Wehrli
S.
Thompson
C. B.
Beyond aerobic glycolysis: transformed cells can engage in glutamine metabolism that exceeds the requirement for protein and nucleotide synthesis
Proc. Natl. Acad. Sci. U.S.A.
2007
, vol. 
104
 (pg. 
19345
-
19350
)
37
Baker
J. P.
Rabin
B. R.
Effects of bromopyruvate on the control and catalytic properties of glutamate dehydrogenase
Eur. J. Biochem.
1969
, vol. 
11
 (pg. 
154
-
159
)
38
Miyamoto
S.
Murphy
A. N.
Brown
J. H.
Akt mediates mitochondrial protection in cardiomyocytes through phosphorylation of mitochondrial hexokinase-II
Cell Death Differ.
2008
, vol. 
15
 (pg. 
521
-
529
)
39
Armstrong
J. S.
Mitochondria: a target for cancer therapy
Br. J. Pharmacol.
2006
, vol. 
147
 (pg. 
239
-
248
)
40
Chiara
F.
Castellaro
D.
Marin
O.
Petronilli
V.
Brusilow
W. S.
Juhaszova
M.
Sollott
S. J.
Forte
M.
Bernardi
P.
Rasola
A.
Hexokinase II detachment from mitochondria triggers apoptosis through the permeability transition pore independent of voltage-dependent anion channels
PLoS ONE
2008
, vol. 
19
 pg. 
e1852
 
41
Tsujimoto
Y.
Shimizu
S.
Role of the mitochondrial membrane permeability transition in cell death
Apoptosis
2007
, vol. 
12
 (pg. 
835
-
840
)
42
Pastorino
J. G.
Shulga
N.
Hoek
J. B.
Mitochondrial binding of hexokinase II inhibits Bax-induced cytochrome c release and apoptosis
J. Biol. Chem.
2002
, vol. 
277
 (pg. 
7610
-
7618
)
43
Costantini
P.
Belzacq
A. S.
Vieira
H. L.
Larochette
N.
de Pablo
M. A.
Zamzami
N.
Susin
S. A.
Brenner
C.
Kroemer
G.
Oxidation of a critical thiol residue of the adenine nucleotide translocator enforces Bcl-2-independent permeability transition pore opening and apoptosis
Oncogene
2000
, vol. 
19
 (pg. 
307
-
314
)
44
Tippett
P. S.
Neet
K. E.
Interconversions between different sulfhydryl-related kinetic states in glucokinase
Arch. Biochem. Biophys.
1983
, vol. 
222
 (pg. 
285
-
298
)
45
Lenzen
S.
Brand
F. H.
Panten
U.
Structural requirements of alloxan and ninhydrin for glucokinase inhibition and of glucose for protection against inhibition
Br. J. Pharmacol.
1988
, vol. 
95
 (pg. 
851
-
859
)
46
Lenzen
S.
Freytag
S.
Panten
U.
Flatt
P. R.
Bailey
C. J.
Alloxan and ninhydrin inhibition of hexokinase from pancreatic islets and tumoural insulin-secreting cells
Pharmacol. Toxicol.
1990
, vol. 
66
 (pg. 
157
-
162
)
47
Tiedge
M.
Richter
T.
Lenzen
S.
Importance of cysteine residues for the stability and catalytic activity of human pancreatic β-cell glucokinase
Arch. Biochem. Biophys.
2000
, vol. 
375
 (pg. 
251
-
260
)

Author notes

1

This work is dedicated to Leopoldo de Meis on his 70th birthday