The energy blockers 3-bromopyruvate and lonidamine: effectson bioenergetics of brain mitochondria

Lara Macchioni & Magdalena Davidescu & Rita Roberti &Lanfranco Corazzi

Received: 16 June 2014 /Accepted: 26 August 2014 /Published online: 7 September 2014# Springer Science+Business Media New York 2014

Abstract Tumor cells favor abnormal energy production viaaerobic glycolysis and show resistance to apoptosis, suggest-ing the involvement of mitochondrial dysfunction. The differ-ences between normal and cancer cells in their energy metab-olism provide a biochemical basis for developing new thera-peutic strategies. The energy blocker 3-bromopyruvate (3BP)can eradicate liver cancer in animals without associated tox-icity, and is a potent anticancer towards glioblastoma cells.Since mitochondria are 3BP targets, in this work the effects of3BP on the bioenergetics of normal rat brain mitochondriawere investigated in vitro, in comparison with the anticanceragent lonidamine (LND). Whereas LND impaired oxygenconsumption dependent on any complex of the respiratorychain, 3BP was inhibitory to malate/pyruvate and succinate(Complexes I and II), but preserved respiration from glycerol-3-phosphate and ascorbate (Complex IV). Accordingly, al-though electron flow along the respiratory chain and ATPlevels were decreased by 3BP in malate/pyruvate- andsuccinate-fed mitochondria, they were not significantly influ-enced from glycerol-3-phosphate- or ascorbate-fed mitochon-dria. LND produced a decrease in electron flow from allsubstrates tested. No ROS were produced from any substrate,with the exception of 3BP-induced H2O2 release from succi-nate, which suggests an antimycin-like action of 3BP as aninhibitor of Complex III. We can conclude that 3BP does notabolish completely respiration and ATP synthesis in brainmitochondria, and has a limited effect on ROS production,confirming that this drug may have limited harmful effects onnormal cells.

Keywords Brain mitochondria . Bromopyruvate .

Lonidamine . Respiration

Introduction

Mitochondria of nerve tissue contribute about 90% of the ATPproduced through oxidative phosphorylation (Rolfe andBrown 1997). ATP utilization occurs mainly in the cytosol,providing chemical energy for supporting various cellularfunctions, including phospholipid metabolism, protein synthe-sis, neurotransmitter cycling, and ion transport across cellularmembranes. A significant amount of ATP in the brain is spentfor maintaining and restoring the transmembrane Na+/K+ gra-dients (Du et al. 2008). Glial cells, which far outnumberneurons in the brain, are very active in brain, providing ho-meostasis maintenance, feedback to neurons, and regulation ofsynaptic plasticity. In addition, glial cells play a vital role in thebrain by removing from the extracellular space the by-productsof neuronal activity and by secreting substances that are crucialfor neuronal functions (Burdakov and Ashcroft 2002).

Originating from glial cells and their precursors, glioblas-toma is the most aggressive form of malignant brain tumors.In contrast to normal cells, tumor cells rely preferentially onanaerobic glycolysis for ATP generation, evenwhen oxygen isavailable (Warburg 1956; Pedersen 2007). Evidence of thelow propensity of glioblastoma cells to respire comes from thelow respiratory rate in the presence of a glucose-based medi-um (Macchioni et al. 2011a).

In GL15 glioblastoma cells, the “energy blocker” 3-bromopyruvate (3BP) mediates autophagy and cardiolipindegradation, leading cells to viability loss (Davidescu et al.2012). This finding extends the potent anticancer power of3BP to brain cancer. Indeed, 3BP can eradicate liver cancer inanimals and exerts, at the same time, minimal toxicity towardsnormal hepatic parenchyma (Ko et al. 2001; Geschwind 2002;

L. Macchioni :M. Davidescu :R. Roberti : L. Corazzi (*)Department of Experimental Medicine, Section of Physiology andBiochemistry, University of Perugia, Via S. Gambuli, 106132 Perugia, Italye-mail: lanfranco.corazzi@unipg.it

J Bioenerg Biomembr (2014) 46:389–394DOI 10.1007/s10863-014-9577-5

Ko et al. 2004; Vali et al. 2007). Likely, 3BP inhibits hepato-cellular carcinoma by acting preferentially against hexokinase2 (Hk-2), which is highly expressed in liver carcinomas (Koet al. 2001). In addition, some other mitochondrial enzymesmay also be targeted by this drug (Pereira da Silva et al. 2009).Considering that 3BP is a strong and non-specific alkylatingagent (Shoshan 2012), a complete understanding of the mech-anism involved in its actions remains to be clarified.

Lonidamine (LND), first synthesized and designed as anantispermatogenic drug (Corsi and Palazzo 1976), exhibitsantiglycolytic activity, most likely acting through the inhibi-tion of the mitochondrial bound Hk-2. In contrast to 3BP,LND exerts low structural reactivity. It targets mitochondriaand induces apoptosis via a direct effect on the PT pore,although the molecular mode of action is still unclear(Ravagnan et al. 1999).

Considering the above, we thought it would be of interestto evaluate the in vitro response of normal brain mitochondriato treatments with 3BP and LND. We report that 3BP andLND target selectively the respiratory chain complexes, with-out abolishing completely respiration and ATP synthesis.Although the electron flow along the respiratory chain wasaffected, ROS production was restricted to 3BP, due to itseffect on Complex III.

Material and methods

Chemicals Carbonyl cyanide 3-chlorophenylhydrazone(CCCP), 4-(2-hydroxyethyl)-1piperazine ethansulfonic acid(HEPES), 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazoli-um bromide (MTT) , N ,N ,N ′ ,N ′ - t e t r ame thy l -p -phenylenediamine dihydrochloride (TMPD), ascorbate,bromopyruvate (3BP), lonidamine (LND), ADP (K+ salt),and respiratory substrates were from Sigma, Milan, Italy.Amplex Ultra Red Reagent was from Molecular Probes(Invitrogen, Italy).

Ethic statement Rats were housed at the Laboratory AnimalResearch Centre of Perugia University. The experimentalprocedures were carried out in accordance with EuropeanDirectives, approved by the Institutional Animal Care andBioethics Committee of University of Perugia. Efforts weremade to minimize animal stress/discomfort.

Mitochondria Highly pure mitochondria were prepared fromrat brain cortex (CD, 2 months old, Charles River, Italy) aspreviously reported (Monni et al. 2000) and resuspended in20 mM Hepes, 70 mM sucrose, 100 mM KCl, 1 mMMgCl2,0.1 mM EGTA and 2 mM KH2PO4 (pH 7.4, buffer A).Mitochondria were metabolically active, with a respiratorycontrol ratio (state 3 to state 4) of 5–6 (Lai et al. 1985).

E f f e c t o f 3 B P a n d LND o n m i t o c h o n d r i a lrespiration According to previous reports describing the ef-fects of 3BP on mitochondria, 100, 200, and 400 μM concen-trations were tested with this drug (Rodrigues-Ferreira et al.2012; Dell’Antone 2009). For LND, 200, 400, or 800 μMconcentrations were selected, based on cell culture studies(Floridi et al. 1981). Mitochondria (0.3 mg protein) werepre-incubated with 3BP or LND for 10 min at 30 °C. Themixtures were transferred into a thermostatic water-jacketedchamber and respiration was started by addition of state 3respiratory substrates (3.0 mM malate/1.5 mM pyruvate,5.0 mM succinate, 3 mM glycerol-3-phosphate, or 0.1 mMTMPD/1 mM ascorbate, and 0.1 mM ADP/5 mM phosphate)in buffer A (0.4 ml final volume), at 30 °C. Oxygen wasdetermined fluorimetrically as previously described(Macchioni et al. 2011b). To assess maximal oxygenconsumption, respiring mitochondria were uncoupledby adding 5 μM CCCP. In parallel experiments, ATPlevels were measured after 15 min of incubation withrespiratory substrates.

MTT reduction assay The reduction of the synthetic electronacceptor MTT was determined to assess the activity of elec-tron flow along the respiratory chain (Schönfeld and Reiser2006). Mitochondria (0.1 mg protein) were pre-incubated inthe presence of 200 μM 3BP or 400 μM LND for 10 min at30 °C. Incubation was continued for 10 min at 30 °C byadding state 3 respiratory substrates (see above) in 0.5 ml ofbuffer A. MTT was then added at 0.1 mg/ml and incubationwas continued for 20 min at 30 °C. Samples were centrifuged,supernatants removed, and pellets resuspended in 1 ml ofabsolute ethanol. MTT reduction was evaluated by measuringthe absorbance at 595 nm.

H2O2 measurement Generation of H2O2 in mitochondria wasmeasured as fluorescence of resorufin (λex 550 nm, λem585 nm) produced byH2O2-dependent Amplex Red oxidationunder the catalysis of horseradish peroxidase, by using aShimadzu RF-5000 spectrofluorometer equipped with tem-perature control and magnetic stirrer device. Mitochondria(0.1 mg protein) were pre-incubated with 200 μM 3BP or400 μM LND for 10 min at 30 °C, then transferred into acuvette in the presence of horseradish peroxidase (2.5 U/ml)and Amplex Red reagent (1 μM). Respiration was started byadding respiratory substrates (3.0 mM malate/1.5 mM pyru-vate, 5.0 mM succinate, or 3.0 mM glycerol-3-phosphate, and0.1 mM ADP/5.0 mM phosphate) in buffer A. Calibrationsignals were generated with known amounts of H2O2 duringeach experiment.

ATP determination ATP levels were quantified by the ATPBioluminescent Assay (FLAA, Sigma) by using a calibratedATP standard curve.

390 J Bioenerg Biomembr (2014) 46:389–394

Statistical analyses The results, expressed as means±SD,were analyzed for statistical significance by Student’s t-test.p-values<0.01 were considered significant.

Results and discussion

Effect of 3BP and LND on oxygen consumption ofmitochondria 3BP and LND are primarily consideredantiglycolytics, but they also target mitochondria. We investi-gated the direct effects of these drugs on mitochondrial respi-ration by using purified rat brain mitochondria as experimen-tal model. In the resting state (no substrates), the respiration ofcontrol mitochondria was low, but its rate increased upon

addition of respiratory substrates, phosphate, and ADP (respi-ratory state 3) (Fig. 1). Due to respiratory control, respirationrate declined, indicating coupling to ATP synthesis. Couplingwas lost when CCCP was added, producing maximal oxygenconsumption of isolated mitochondria (not shown). The ki-netics of oxygen consumption through Complex II from suc-cinate was influenced by 3BP and LND in a concentration-

Flu

ore

scen

ce q

uen

chin

g(a

rbit

rary

un

its)

Time (min)

malate/pyruvate glyc-3-P

Control

LND

3BP LND

3BP

Control

succinate

Flu

ore

scen

ce q

uen

chin

g(a

rbit

rary

un

its)

Time (min)

Control

100

200

400

270

250

230

210

190

170

0 161284

succinate

Control

200

400

800

Time (min)

a

b

Time (min)

TMPD/Asc

Control

3BP

LND

Time (min)

270

250

230

210

190

170

0 161284

270

250

230

210

190

170

0 161284

270

250

230

210

190

170

0 161284

270

250

230

210

190

170

0 161284

Fig. 1 Effect of 3BP and LNDonoxygen consumption ofmitochondria. a Mitochondriawere pre-incubated with 3BP(100–400 μM) or LND (200–800 μM) for 10min at 30 °C, thenrespiration was started by adding5.0 mM succinate and 0.1 mMADP/5 mM phosphate in bufferA. b Mitochondria were pre-incubated as in a with 200 μM3BP or 400 μM LND, then res-piration was started with 3.0 ma-late/1.5 mM pyruvate, or 3 mMglycerol-3-phoshate, or 0.1 mMTMPD/1 mM ascorbate, and0.1 mM ADP/5 mM phosphate.The time-dependent oxygen con-sumption was monitored as fluo-rescence quenching. The freshmedium contained about 95 nmoloxygen in 0.4 ml. Representativetraces of three independent ex-periments are shown

Table 1 Initial rate of oxygen consumption of mitochondria

nmol×min−1×mg protein−1

Substrate Control 3BP (200 μM) LND (400 μM)

malate/pyruvate 2.22±0.31 0.99±0.08 0.75±0.08

succinate 10.03±1.22 3.61±0.42 5.69±0.48

glycerol-3-phosphate 4.23±0.41 3.50±0.51 0.25±0.06

TMPD/ascorbate 8.73±0.11 8.52±0.09 4.43±0.38

See legend to Fig. 1

0

20

40

60

80

100

120

MT

T (

per

cen

t of

con

tro

l)

Control3BPLND

**

*

*

*

*

Fig. 2 Effect of 3BP and LND on MTT reduction. Mitochondria werepre-incubated with 200 μM 3BP or 400 μM LND for 10 min at 30 °C,then respiration was started by adding state 3 respiratory substrates. After10 min respiration, MTT reduction was evaluated by measuring theabsorbance at 595 nm. Values are the mean±S.D. of three independentexperiments; *p<0.01

J Bioenerg Biomembr (2014) 46:389–394 391

dependent manner (Fig. 1a). At 15 min incubation time, about50 % and 30 % respiration was preserved in the presence of100 and 200 μM 3BP, respectively. It is worth to note that100 μM3BP is a concentration higher than that producing celldeath in GL15 glioblastoma cells (Macchioni et al. 2011a). Inthe presence of LND, about 70 % and 55 % respiration waspreserved at 400 and 800 μM, respectively. These are theconcentrations usually employed in cell culture experiments(Floridi et al. 1981; Ravagnan et al. 1999). In the subsequentexperiments, 200 μM 3BP and 400 μM LND were used.Table 1 summarizes the initial rate of oxygen consumption,

as calculated from kinetic studies. 3BP inhibited respirationfrom malate/pyruvate (Complex I), but not significantly fromglycerol-3-phosphate and TMPD/ascorbate (Complex IV).Contrarily, LND was inhibitory from any substrate, althoughstill significant respiratory activity was detected from succi-nate and TMPD/ascorbate (Table 1 and Fig. 1b).

Electron transfer through the respiratory chain The reducingpower of the electron flow along the respiratory chain wasassessed by MTT reduction, a process occurring at distinctsites of the electron transport chain (Schönfeld and Reiser2006). This approach allows the identification of sites where3BP or LND could interfere with the electron flow. In mito-chondrial preparations respiring in state 3, the pre-treatmentwith 3BP produced a decrease of the reducing power fromsubstrates feeding Complexes I and II, but not from glycerol-3-phosphate and ascorbate, whereas LND always produced adecrease ofMTT reduction (Fig. 2). These data, in accordancewith oxygen consumption, indicate that 3BP, contrarily toLND, partly preserves the mitochondrial respiratory activity.

ATP synthesis In coupled mitochondria, the amount of newlysynthesized ATP is related to the rate of oxygen consumption.In agreement with the high oxygen consumption measuredfrom succinate (Fig. 1a), ATP levels were the highest insuccinate-fed control mitochondria (Fig. 3). 3BP and LNDpre-treatment of mitochondria reduced ATP levels to 15–20%of control from both malate/pyruvate and succinate. The in-hibitory effect of 3BP on succinate-driven ATP synthesis was

ControlBrPALND

0

5

10

15

AT

P (

nm

ol/m

g p

rote

in)

**

*

*

Control3BPLND

Fig. 3 Effect of 3BP and LND on ATP levels. Mitochondria were pre-incubated with 200 μM 3BP or 400 μM LND for 10 min at 30 °C, thenrespiration was started by adding state 3 respiratory substrates. ATP levelswere measured after 15 min of incubation. Values are the mean±S.D. offour independent experiments; * p<0.01

30 sec 30 sec 30 sec

30 sec 30 sec 30 sec

Control 200µM 3BP 400µM LND

malate/pyruvate malate/pyruvate malate/pyruvate

succinate succinate succinate

0.177 0.146 0.135

0.146

0.3540.149

30 sec 30 sec 30 sec

glyc-3-P0.468 654.0024.0glyc-3-Pglyc-3-P

Fig. 4 H2O2 release bymitochondria. Mitochondria werepre-incubated with 200 μM 3BPor 400 μM LND and respirationwas started by adding respiratorysubstrates, as described in Mate-rial and methods. H2O2 releasewas evaluated as Amplex Redfluorescence increase. Represen-tative traces of three independentexperiments are shown. Numbersalong the traces are in nmol H2O2/min/mg protein. Calibration sig-nals were generated with knownamounts of H2O2 during eachexperiment

392 J Bioenerg Biomembr (2014) 46:389–394

also observed previously in isolated liver mitochondria(Dell’Antone 2009). ATP synthesis from glycerol-3-phosphate and ascorbate was not significantly influenced bythe drugs (Fig. 3). Interestingly, we can conclude that, in 3BPtreated mitochondria, glycerol-3-phosphate and ascorbate canfeed respiration, which is also coupled to the ATP synthesis.

ROS synthesis Low levels of ROS species are physiologicallysynthesized in normal respiration. Indeed, the mitochondrialelectron transport chain is the main cellular source of oxygensuperoxide anion (O2

−), a reactive oxygen species that isquickly dismutated to H2O2 by the mitochondrial superoxidedismutase (Macchioni et al. 2011b). ROS increase in mito-chondria dysfunction, thus contributing to deleterious cellinjury. As 3BP and LND influence the electron flow alongthe respiratory chain, we investigated whether ROS synthesiswas also affected. To study this issue, mitochondria wereincubated by feeding respiratory chain with the electronsderived from oxidation of the Complex I (malate/pyruvate)or Complex II (succinate) substrates, and of glycerol-3-phos-phate, in the presence of ADP and phosphate (respiratory state3). The slopes of fluorescent traces in Fig. 4 show the rate ofH2O2 release from mitochondria respiring under condition ofATP synthesis. The extent of H2O2 released depended on thefed substrate, the highest value being from glycerol-3-phosphate (0.468 nmol/min/mg protein) and the lowest fromsuccinate (0.146 nmol/min/mg protein). 3BP and LND did notinfluence H2O2 produced from malate/pyruvate and fromglycerol-3-phosphate, whereas from succinate only 3BP en-hanced the release of H2O2 up to 0.354 nmols/min/mg protein(Fig. 4).

A tight link between succinate-supported reversed electrontransfer and ROS production has been reported (Liu et al.2002). Indeed, mitochondria incubated in respiratory state 4(lack of ADP) and supported by succinate produce a physio-logically relevant ROS amount (0.6–0.7 nmol/min/mg pro-tein) (Liu et al. 2002; Tretter et al. 2007; Macchioni et al.2011b). ROS apparently occurr at Complex I, since rotenonestops their production. Succinate-dependent ROS increase isalso observed in the presence of antimycin (Complex IIIinhibitor), suggesting that blocking downstream of ComplexII is responsible for ROS production (Liu et al. 2002;Macchioni et al. 2011b). Therefore, we can speculate thatthe increased ROS production from succinate in 3BP treatedmitochondria is an antimycin-like action of 3BP as an inhib-itor of Complex III.

Conclusions 3BP and LND do not abolish respiration ofnormal brain mitochondria, at least when assayed fromComplexes II and IV; moreover, the increase of ROS synthesiswas restricted to 3BP from succinate, suggesting that thesedrugs may have limited harmful effects on normal cells,

although effects exerted outside mitochondria cannot be ruledout. This is in agreement with the finding that 3BP exertsminimal toxicity towards normal hepatic parenchyma (Valiet al. 2007). 3BP is very rapid in killing GL15 glioblastomacells already at low concentration (60–80 μM, 1–2 h), incontrast to LND (400 μM, 48 h) (our unpublished observa-tions). Therefore, we can state that, at concentrations effectivein killing glioblastoma cells, 3BP does not abolish the mito-chondrial functions of normal tissue. Recent clinical applica-tions (Ko et al. 2012) support 3BP as effective cancer thera-peutic that, when formulated properly, acts without associatedtoxicity.

Acknowledgments Mr Carlo Ricci is thanked for skilful technicalassistance.

References

Burdakov D, Ashcroft FM (2002) Shedding new light on brain metabo-lism and glial function. J Physiol 544:334

Corsi G, Palazzo G (1976) 1-Halobenzyl-1H-indazole-3-carboxylicacids. A new class of antispermatogenic agents. J Med Chem 19:778–783

Davidescu M, Sciaccaluga M, Macchioni L, Angelini R, Lopalco P,Rambotti MG, Roberti R, Corcelli A, Castigli E, Corazzi L (2012)Bromopyruvate mediates autophagy and cardiolipin degradation tomonolyso-cardiolipin in GL15 glioblastoma cells. J BioenergBiomembr 44:51–60

Dell’Antone P (2009) Targets of 3-bromopyruvate, a new, energy deplet-ing, anticancer agent. Med Chem 5:491–496

Du F, Zhu XH, Zhang Y, Friedman M, Zhang N, Ugurbil K, Chen W(2008) Tightly coupled brain activity and cerebral ATP metabolicrate. Proc Natl Acad Sci USA 105:6409–6414

Floridi A, Paggi MG, D’Atri S, De Martino C, Marcante ML, SilvestriniB, Caputo A (1981) Effect of lonidamine on the enrgy metabolismof Ehrlich ascites tumor cells. Cancer Res 41:4661–4666

Geschwind JF, Ko YH, Torbenson MS, Magee C, Pedersen PL (2002)Novel therapy for liver cancer: direct intraarterial injection of apotent inhibitor of ATP production. Cancer Res 62:3909–3913

Ko YH, Pedersen PL, Geschwind JF (2001) Glucose catabolism in therabbit VX2 tumor model for liver cancer: characterization andtargeting hexokinase. Cancer Lett 173:83–91

Ko YH, Smith BL, Wang Y, Pomper MG, Rini DA, Torbenson MS,Hullihen J, Pedersen PL (2004) Advanced cancers: eradication in allcases using 3-bromopyruvate therapy to deplete ATP. BiochemBiophys Res Commun 324:269–275

KoYH, Verhoeven HA, LeeMJ, Corbin DJ, Vogl TJ, Pedersen PL (2012)A translational study “case report” on the small molecule “energyblocker” 3-bromopyruvate (3BP) as a potent anticancer agent: frombench side to bedside. J Bioenerg Biomembr 44:163–170

Lai JC, Sheu KF, Carlson KC Jr (1985) Differences in some of themetabolic properties of mitochondria isolated from cerebral cortexand olfactory bulb of the rat. Brain Res 343:52–59

Liu Y, Fiskum G, Schubert D (2002) Generation of reactive oxygenspecies by the mitochondrial electron transport chain. JNeurochem 80:780–787

Macchioni L, Davidescu M, Sciaccaluga M, Marchetti MC,Migliorati G,Coaccioli S, Roberti R, Corazzi L, Castigli E (2011a) Mitochondrial

J Bioenerg Biomembr (2014) 46:389–394 393

dysfunction and effect of antiglycolytic bromopyruvic acid in GL15glioblastoma cells. J Bioenerg Biomembr 43:507–518

Macchioni L, Davidescu M, Mannucci R, Francescangeli E, Nicoletti I,Roberti R, Corazzi L (2011b) H2O2 disposal in cardiolipin-enrichedbrain mitochondria is due to increased cytochrome c peroxidaseactivity. Biochim Biophys Acta 1811:203–208

Monni M, Corazzi L, Migliorati G, Roberti R (2000) Respiratory stateand phosphatidylserine import in brain mitochondria in vitro. JMembr Biol 173:97–105

Pedersen PL (2007) Warburg, me and hexokinase 2: Multiple discoveriesof key molecular events underlying one of cancer’s most commonphenotypes, the “Warburg Effect”, i.e., elevated glycolysis in thepresence of oxygen. J Bioenerg Biomembr 39:211–222

Pereira da Silva AP, El-Bacha T, Kyaw N, dos Santos RS, da-Silva WS,Almeida FCL, Da Poian AT, Galina A (2009) Inhibition of energy-producing pathways of HepG2 cells by 3-bromopyruvate. BiochemJ 417:717–726

Ravagnan L, Marzo I, Costantini P, Susin SA, Zamzami N, Petit PX,Hirsch F, Goulbern M, Poupon MF, Miccoli L, Xie Z, Reed JC,Kroemer G (1999) Lonidamine triggers apoptosis via a direct, Bcl-2-inhibited effect on the mitochondrial permeability transition pore.Oncogene 18:2537–2546

Rodrigues-Ferreira C, da Silva AP, Galina A (2012) Effect of the antitu-moral alkylating agent 3-bromopyruvate on mitochondrial respira-tion: role of mitochondrially bound hexokinase. J BioenergBiomembr 44:39–49

Rolfe DF, Brown GC (1997) Cellular energy utilization and molecularorigin of standard metabolic rate in mammals. Physiol Rev 77:731–758

Schönfeld P, Reiser G (2006) Rotenone-like action of the branched-chainphytanic acid induces oxidative stress in mitochondria. J Biol Chem281:7136–7142

Shoshan MC (2012) 3-Bromopyruvate: targets and outcomes. J BioenergBiomembr 44:7–15

Tretter L, Takacs K, Hegedus V, Adam-Vizi V (2007) Characteristics ofalfa-glycerophosphate-evoked H2O2 generation in brain mitochon-dria. J Neurochem 100:650–663

Vali M, Liapi E, Kowalski J, Hong K, Khwaja A, Torbenson MS,Georgiades C, Geschwind JF (2007) Intraarterial therapy with anew potent inhibitor of tumor metabolism (3-bromopyruvate): iden-tification of therapeutic dose and method of injection in an animalmodel of liver cancer. J Vasc Interv Radiol 18:95–101

Warburg O (1956) On respiratory impairment in cancer cells. Science124:269–270

394 J Bioenerg Biomembr (2014) 46:389–394