Inactivation of the HIF-1A/PDK3 Signaling Axis Drives Melanoma Toward Mitochondrial Oxidative Metabolism and Potentiates the Therapeutic Activity of Pro-Oxidants

Published OnlineFirst August 3, 2012; DOI: 10.1158/0008-5472.CAN-12-0979

Cancer Therapeutics, Targets, and Chemical Biology Research

Inactivation of the HIF-1a/PDK3 Signaling Axis Drives Melanoma toward Mitochondrial Oxidative Metabolism and Potentiates the Therapeutic Activity of Pro-Oxidants

Jerome Kluza1, Paola Corazao-Rozas1, Yasmine Touil1, Manel Jendoubi1, Cyril Maire1, Pierre Guerreschi1, Aurelie Jonneaux1, Caroline Ballot1,Stephane Balayssac4, Samuel Valable2,3, Aurelien Corroyer-Dulmont2,3, Myriam Bernaudin2,3, Myriam Malet-Martino4, Elisabeth Martin de Lassalle1, Patrice Maboudou5, Pierre Formstecher1, Renata Polakowska1, Laurent Mortier1, and Philippe Marchetti1,5

Abstract Cancer cells can undergo a metabolic reprogramming from oxidative phosphorylation to glycolysis that allows them to adapt to nutrient-poor microenvironments, thereby imposing a selection for aggressive variants. However, the mechanisms underlying this reprogramming are not fully understood. Using complementary approaches in validated cell lines and freshly obtained human specimens, we report here that mitochondrial respiration and oxidative phosphorylation are slowed in metastatic melanomas, even under normoxic conditions due to the persistence of a high nuclear expression of hypoxia-inducible factor-1a (HIF-1a). Pharmacologic or genetic blockades of the HIF-1a pathway decreased glycolysis and promoted mitochondrial respiration via specific reduction in the expression of pyruvate dehydrogenase kinase-3 (PDK3). Inhibiting PDK3 activity by dichloroacetate (DCA) or siRNA-mediated attenuation was sufficient to increase pyruvate dehydrogenase activity, oxidative phosphorylation, and mitochondrial reactive oxygen species generation. Notably, DCA potentiated the antitumor effects of elesclomol, a pro-oxidative drug currently in clinical development, both by limiting cell proliferation and promoting cell death. Interestingly, this combination was also effective against BRAF V600E-mutant melanoma cells that were resistant to the BRAF inhibitor vemurafenib. Cotreatment of melanomas with DCA and elesclomol in vivo achieved a more durable response than single agent alone. Our findings offer a preclinical validation of the HIF-1/PDK3 bioenergetic pathway as a new target for therapeutic intervention in metastatic melanoma, opening the door to innovative combinations that might eradicate this disease. Cancer Res; 72(19); 5035–47. 2012 AACR.

Introduction (ROS; refs. 2, 3). It is generally postulated that the cellular Over the last decades, huge efforts devoted to comprehend effects of ROS depend on the level at which ROS are produced. melanoma biology led to the identification of new targets for Controlled production of ROS participates in the promotion antimelanoma therapy (1). Apart from targeting the oncogenic and progression of melanoma (3), whereas higher ROS gener- mutations present in many (but certainly not all) melamomas, ation displays cell-damaging effects (4). Constitutive produc- another promising strategy is to exploit biochemical particu- tion of the ROS weakens melanoma cells that are closer to the larities of melanoma cells. One biochemical hallmark of mel- point where cell death can occur. Hence, melanoma cells show anoma is the generation of excessive reactive oxygen species increased sensitivity to ROS-induced death as compared with melanocytes and to other tumors (5). According to this view, elesclomol, a pro-oxidant molecule that targets the mitochon- Authors' Affiliations: 1Unit 837 Equipe 4 Inserm and FacultedeM edecine, drial electron transport chain (ETC; ref. 6), has been evaluated Universite de Lille II 1 Place, Verdun Cedex; 2CNRS, UMR ISTCT 6301, CERVOxy Group; GIP CYCERON, Caen Cedex; 3Universite de Caen in clinical trials for the treatment of metastatic melanomas and Basse-Normandie, Normandie; 4Laboratoire SPCMIB, UMR CNRS 5068 has shown encouraging clinical responses. Intriguingly, clinical Universite Paul Sabatier, Toulouse Cedex; and 5Centre de Bio-Pathologie, favorable responses occurred in a subset of patients distin- Plate-forme de Biotherapie, Banque de Tissus, CHRU Lille, France guished by low serum lactate dehydrogenase (LDH; ref. 7). Note: Supplementary data for this article are available at Cancer Research Thus, serum LDH can be considered as the predictor of Online (http://cancerres.aacrjournals.org/). response to elesclomol. At a cellular level, elesclomol requires J. Kluza and P. Corazao-Rozas contributed equally to this work. a functioning ETC to induce ROS-mediated melanoma cell Corresponding Author: Philippe Marchetti, INSERM U 837 Facultede death (6). In this context, unraveling the regulatory mechan- Medecine 1, Place Verdun F-59045, Lille Cedex, France. Phone: 33-3-20- isms essential for enhancing ROS production is fundamental 62-69-52; Fax: 33-3-20-62-68-84; E-mail: [email protected] for improving the efficiency of pro-oxidants in melanoma. doi: 10.1158/0008-5472.CAN-12-0979 Approximately, 60% to 90% of cancers display a metabolic 2012 American Association for Cancer Research. profile, the so-called Warburg phenotype, characterized by

www.aacrjournals.org 5035

Kluza et al.

theirdependenceuponglycolysisasthemajorsourceof study was obtained from the local Person's Protection Com- energy, irrespective of the oxygen level (8). According to the mittee, and all patients provided informed consent. The sam- Warburg effect, pyruvate, the end product of glycolysis, is ples came from patients who had to be treated surgically for mainly converted into lactate by LDH-A that is upregulated cutaneous melanoma metastases. in transformed cells, rather than oxidized in mitochondria. It seems conceivable that melanoma cells mainly rely on Cell lines glycolysis for energetic needs based on the following reasons: HL60, A375, and SKMel-28 were obtained from American (i) the glycolytic phenotype of melanoma cell lines has been Type Culture Collection over the past 2 years. The human recently identified by metabolic profiling (9); (ii) metastatic melanoma cell lines HBL, LND, MM074, and MM079 were melanomas are characterized by their particularly high avidity established in the Laboratory of Oncology and Experimental for 2[18F]fluoro-2-deoxy-D-glucose (18FDG) clinically detect- Surgery (Institut Bordet, Universite libre de Bruxelles, Brussels, able on positron emission tomography (PET; ref. 10); (iii) the Belgium) from lymph node metastases of patients with mel- isoenzyme LDH-5, the more effective in the conversion of anoma and were kindly provided by G. Ghanem in 2009. Mel- pyruvate to lactate, is detected on histologic sections and in 4M was established in the laboratory by Dr. Polakowska in blood sera from patients with highly metastatic melanoma (for 2009. All cells were passaged within 6 months of thawing. The review; ref. 11); and (iv) hypoxia-inducible factor-1a (HIF-1a), a identity of cell lines was regularly examined by morphologic master regulator of metabolism required for the active adap- criteria and the presence of differentiation markers (last test in tation of cancer cells to hypoxic conditions, is overexpressed in March 2012). They were expressing at least one of the mela- human melanoma (12, 13). In addition to hypoxia, HIF-1a has nocyte differentiation markers (TYR, TYRP1, TYRP2/DCT, also been found stabilized in normoxic conditions in melano- Melan-A/MART-1, gp100/pmel17, S100B) as routinely assessed ma cells. For instance, melanoma antigen-11 regulates HIF-1 by Western blotting/immunofluorescence. The identity of accumulation by inhibition of prolyl hydroxylase activity (14). HBL, LND, and Mel-4M was also confirmed by karyotyping Endothelin or the antiapoptotic protein, Bcl-2, induces HIF-1a and array comparative genomic hybridization testing (IRCL accumulation (15, 16). Genetic alterations, such as Micro- and Laboratoire de genetique, CHRU, Lille, France). phthalmia-Associated Transcription Factor (MITF) germline All cell lines were cultured at 37 C under 5% CO2 in RPMI mutation (17) or the oncogenic V600E BRAF mutation (18) medium (Gibco-BRL, Life Technologies) supplemented with overexpress HIF-1a and enhance transcriptional activity of 10% fetal calf serum (Gibco-BRL), antibiotics, and 1 mmol/L downstream genes. sodium pyruvate (Gibco-BRL). Cells were transfected with However, recent studies indicate that melanomas do not siRNA targeting PDK-3 (sc-39029, Santa Cruz Biotechnology) adopt a bona fide Warburg phenotype, as glutamine stimulates or a nontargeting control siRNA (sc-37707) using Lipofecta- mitochondrial metabolism to favor melanoma growth (9, 19). It mine 2000 (Life Technologies). PDK3 overexpression in mel- is well established that, under hypoxic conditions, HIF-1a anoma cell lines was obtained by transfection with pJP1520- activates the expression of glycolytic enzymes and glucose PDK3 vector [provided through central repository DNASU transporters and downregulates mitochondrial activity and (http://dnasu.asu.edu; ref. 23)] DCA, and/or elesclomol were ROS production through several distinct mechanisms in a added 48 hours after transfection. HBL cells lacking mitochon- context-specific manner (20). HIF-1a has also been involved drial DNA (HBL r0) were generated as published (24). The A375 in the "glutamine addiction" of cancer cells (21). cell line, which harbors the BRAFV600E mutation, was treated Our objective was to investigate the influence of metabolic with vemurafenib to obtain the resistant subline (A375-R) as pathways on mitochondrial ROS production, particularly in described in the Supplementary material. melanoma under normoxic conditions. We hypothesized that HIF-1a can also prevent the onset of oxidative stress via HIF-1a shRNA downregulation of mitochondrial respiration, and therefore Three HIF-1a short hairpin RNA (shRNA) clones (see Sup- can itself represent a factor of resistance to pro-oxidative drugs plementary material for sequences) were obtained by limiting in melanomas. dilution. Activation of HIF-1a was quantified by a DNA-binding ELISA kit (TransAM Active Motif). Materials and Methods Chemicals Determination of glucose, lactate, and ATP All chemicals were purchased from Sigma–Aldrich except Extracellular glucose and lactate levels were measured on CM-H2DCFDA from Life Technologies and YC-1) from Cayman a SYNCHRON LX20 Clinical system (Beckman Coulter). For Chemical Company (Interchim). Elesclomol, provided by Synta assessment of intracellular ATP, the Celltiter Glow Assay Kit Pharmaceuticals Corp., was prepared as described (6). The (Promega) was used. combined effect of dichloroacetate (DCA) and elesclomol was analyzed by the combination index isobologram (22). Cytofluorometric analysis Cell viability and cell cycle were assessed after staining with Patients and tissue samples propidium iodide (PI). The detection of ROS was determined Cutaneous metastases were obtained from 4 melanoma using the CM-H2DCFDA or hydroethidium probe following patients (see Supplementary Table S1) at the Clinique de current protocols (25) and conducted on a FACS Canto II Dermatologie, CHRU (Lille, France). Ethical approval for this cytofluorometer (Beckton Dickinson).

5036 Cancer Res; 72(19) October 1, 2012 Cancer Research

Targeting PDK to Enhance Oxidative Stress in Melanoma

Figure 1. A, colony-forming assays of cells in the presence of glucose or glutamine (mean SD; n ¼ 3, t-test, , P < 0.01). B, cell-cycle analysis of cell lines treated with 10 mg/mL oligomycin, 5 mmol/L rotenone, or 75 mmol/L bromopyruvate for 24 hours. Numbers represent the percentage of sub-G1 cells. C, cells were treated with oligomycin for 4 hours then collected for ATP measurement (mean SD, n ¼ 3). D, growing, the rate of glucose and lactate released by cells (0.9 106 cells/mL) in RPMI medium was calculated by dividing the difference of amount of glucose/lactate present in the supernatant between the 2 consecutive time points (2–6 hours) by the number of hours elapsed (mean SD, n ¼ 5). E, oxygen consumption of intact cells. Data are representative of 5 experiments. F, proportions of mitochondrial oxygen consumption due to proton leak (respiration not modiﬁed by oligomycin) and ATP turnover (the respiration inhibited by oligomycin). Oxygen consumption was determined as in E (mean SD, n ¼ 3). G and H, respiration rate (G) and oxygen consumption (H) in human skin metastasis and peritumoral tissue samples from 4 different patients ( , patient 1 and patient 2; , patient 4; and , patient 3). Horizontal lines represent median values (paired t-test, , P < 0.001). I, coronal images of PET/CT scan patient (patient 2; left) show multiple foci of abnormal FDG uptake. Right, axial image of one metastasis with laterothoracic localization. After metastasis abalation, 2 samples were obtained from sites with different SUV values then prepared for oxygen consumption. www.aacrjournals.org Cancer Res; 72(19) October 1, 2012 5037

Kluza et al.

A B DAPI HIF-1 Overlay HBL HBL A375 LND A375 Normoxia Hypoxia HIF-1 kd2 HIF-1 kd3 Scrambled HIF-1 kd1 HBL –DFO 1.00 0.530.15 0.23 HIF-1 kd2 Scrambled HIF-1 kd2 HIF-1α Scrambled YC-1 100 μ mol/L Control YC-1 10 μ mol/L YC-1 100 μ mol/L Control YC-1 10 μ mol/L Control YC-1 10 μ mol/L YC-1 100 μ mol/L +DFO HIF-1α HIF-1α 1.00 0.400.290.65 1.00 0.59 2.17 0.88 LND 1.00 0.53 0.45 1.00 1.12 0.65 1.00 1.05 0.75 Lamin B Lamin B Lamin B

HL60

C DEHBL ) ) HIF-1 kd1 HIF-1 kd2 HIF-1 kd3 2 0.2 –1 Scrambled 0.1 500 HIF-1 kd1 HIF-1 kd2 0 HIF-1 kd3

Scrambled HIF-1 kd1 HIF-1 kd2 HIF-1 kd3 400 −0.1 HK2 − 0.2 1.00 0.56 0.19 0.44 300 −0.3 LDH-A 1.00 0.43 0.36 0.36 −0.4 200 Actin

−0.5 Lactate release (mg.mL 0246810 HK2 HK2 HK2 mRNA level fold change (log fold mRNA level EGF EGF EGF HIF2 HIF2 HIF2 PFKp PFKp PFKp LDHA LDHA LDHA PGK1 PGK1 PGK1 VEGF VEGF VEGF PFKm PFKm PFKm Time (h) (ratio HBL HIF-1kd cell lines vs. HBL scrambled) HBL HIF-1kd cell lines vs. (ratio GLUT1 GLUT1 GLUT1 ALDOA ALDOA ALDOA GAPDH GAPDH GAPDH

F G Control zVAD 2DG 2DG + zVAD Control zVAD 2DG 2DG + zVAD Control zVAD 2DG 2DG + zVAD Control zVAD 2DG 2DG + zVAD Intact PARP High Cleaved PARP Scrambled HIF-1 kd1 HIF-1 kd2 HIF-1 kd3

Scrambled HIF-1 kd1 HIF-1 kd2 HIF-1 kd3 18 (counts/ID/g) Scrambled HIF-1 kd1 HIF-1 kd2 HIF-1 kd3 F-FDG uptake HBL Control

SUV: 0.52 SUV: 0.31 SUV: 0.22 SUV: 0.24 Low

2DG 10 mmol/L DNA content (PI)

HI2.0 J Scrambled (2 mmol/L Gln) 0.4 1.1 ATP (fentolmol.cells Glutamine Glutamine Scrambled (0 mmol/L Gln) Respiration

) HIF-1 kd2 (2 mmol/L Gln) 60 ATP 6 HIF-1 kd2 (0 mmol/L Gln) 0.5 cells)

–6 0.3 0.9 40 1.0 .10 –1 2 mmol/L 0.1 mmol/L 0 mmol/L 2 mmol/L 0.1 mmol/L 0 mmol/L 0.2 0.7 20 respiration .sec

0.5 2 OXPHOS-related –1 Respiration rate Respiration 0 0.1 0.5 ) Cell numbers (×10 Cell numbers 0

024487296 (pmol O

Time (h) Control HIF-1 kd1 HIF-1 kd2 HIF-1 kd3 HIF-1 kd1 HIF-1 kd2 HIF-1 kd3 Scrambled Scrambled HIF-1 kd2 Scrambled YC-1 10 μ mol/L

YC-1 100 μ mol/L HBL HBL

Figure 2. A, confocal immunoﬂuorescence analysis of HIF-1a expression (green) in melanoma cell lines. Magniﬁcation, 630. DAPI, 40, 6-diamidino- 2-phenylindole. B, immunoblots for HIF-1a accumulation in nuclear fraction from: cells maintained in normoxia or treated with 500 mmol/L desferrioxamine (DFO) for 18 hours (right); cells treated with YC-1 for 18 hours (left); cells cultured in hypoxic conditions (1% O2) for 24 hours (middle). Densitometric values of proteins normalized on lamin expression are expressed. C, quantitative real-time PCR analysis of HIF-1a target genes (mean SD, n ¼ 4). D, HK2 and LDH expression by immunoblot analysis. Densitometric values of proteins normalized on actin expression are expressed. E, lactate was measured in the supernatant (mean SD, n ¼ 5). F, immunoblot analysis of PARP cleavage in cells treated with 10 mmol/L 2-DG for 48 hours and/or the pan-caspase inhibitor

z-VAD.fmk (50 mmol/L); cell-cycle analysis of cells treated with 10 mmol/L 2-DG for 48 hours. Numbers represent the percentage of sub-G1 cells. G, microPET scan of SCID mice carried scrambled or HIF-1a shRNA HBL melanoma xenograft (circle). An intensity scale is represented. Radioactivity is also seen in kidney (k). The SUVs are indicated. H, colony-forming assays of cells in the presence of glucose or glutamine and counted after trypan blue exclusion. I, respiration rates of HBL cells exposed to YC-1 for 18 hours and HIF-1a shRNA cells. Oxygen consumption was determined as in Fig. 1D. J, proportion of OXPHOS-related respiration was obtained by the difference between routine and oligomycin-inhibited respirations normalized with maximum mitochondrial respiratory capacity. ATP levels were also determined (mean SD, n ¼ 5).

5038 Cancer Res; 72(19) October 1, 2012 Cancer Research

Targeting PDK to Enhance Oxidative Stress in Melanoma

PCR analysis Histology The mRNA was quantified by quantitative real-time PCR Tumor specimens were stained with H&E. For in situ using a protocol optimized for Lightcycler 480 detector (Roche determination of cell proliferation or apoptosis, sections were Applied Science; ref. 25). The primer sequences are available analyzed with an antibody to Ki-67 or terminal deoxynucleo- upon request. The transcript level in each sample was nor- tidyl transferase–mediated dUTP nick end labeling (TUNEL; In malized to that of 18S rRNA. The relative expression of target Situ Cell Death Detection Kit, Roche). As a marker of ROS- mRNAs was analyzed using the Pfafll method (25). mediated protein oxidation, protein carbonyls were detected by the dinitrophenyl hydrazine (DNPH) method using Oxyblot Metabolite profiles Protein Oxidation Detection Kit (Millipore). Metabolites were extracted (26), and profiles were obtained using the 1H-NMR method as previously described (25). In vivo study All procedures with animals were conducted according to Immunoblot analysis the Institutional guidelines. Immunodeficient female severe Immunoblotting was conducted as indicated (25). To detect combined immunodeficient (SCID) mice, 6- to 8-week-old, ETC proteins, monoclonal antibodies (OXPHOS complexes kit, under isoflurane anesthesia were injected with 2 106 HBL MitoSciences; 1:1,000) were used. Otherwise, the primary anti- cells, mixed (1:1 volume) with BD Matrigel Basement Mem- bodies used are described in Supplementary materials. brane Matrix. When tumors reached 400 mL, the mice were divided into 4 groups: control group (n ¼ 4): mice were treated Chromatin immunoprecipitation with saline with the same schedule as the treated animals; Cells were cross-linked in 1% formaldehyde for 8 minutes elesclomol group (n ¼ 6): mice were treated with elesclomol (20 and processed for chromatin immunoprecipitation (ChIP) mg/kg, i.v. injection for 5 d/wk); DCA group (n ¼ 6): mice were assay using the HIF-1a ExactaChIP Chromatin IP Kit (R&D treated with DCA (75 mg/L added to the drinking water); DCA/ Systems). Capture of the DNA fragments was tested by PCR elesclomol group (n ¼ 6) mice were treated with the combi- using PDK3 primers previously reported (27). nation of DCA and elesclomol. For patients' tumors implanted in mice, fresh tumor samples were minced into small pieces, mixed (1:1 volume) with Matrigel, and then injected into the Measurement of oxygen consumption fl Oxygen consumption was monitored with Oxygraph oxygen ank of SCID mice as described earlier. electrodes (Hansatech Instruments Ltd.; ref. 25). Cell viability Imaging microPET scan analysis was checked by PI staining. For melanoma specimens, oxygen – consumption assays were conducted using the respirometry The PET experiments (authorization N 14 55) were carried system Oxygraph-2k (Oroboros Instruments). Data were nor- out under anesthesia. Images were acquired on a microPET malized to the dry weight of the specimens. Siemens Inveon Preclinical system. The emission scan lasting 1 hour was initiated after injection of 18FDG (370 MBq/kg) through the tail vein. To quantify the 18F-FDG uptake on the PDH activity and phosphoPDH detection last frame (corresponding to 40–60 min), the measured tissue The determination of pyruvate dehydrogenase (PDH) activ- activity concentration [counts (kBq)/mL] was divided by the ity and phosphorylation of PDH were conducted using the PDH injected activity in kBq per gram of body weight (kBq/g) to give Assay Kit (MSP18, Mitosciences) and the PhosphoPDH In-Cell a standardized uptake value (SUV). ELISA Kit Colorimetric (MSP48, Mitosciences), respectively. Statistical analysis Transmission electron microscopy Results were analyzed using GraphPad Prism version 5.00 Cells were prepared for transmission electron microscopy as (GraphPad Software). The Student t test was used to compare described (25). data sets. Statistical significance was set at P < 0.05.

Table 1. Quantiﬁed levels of mitochondrial TCA cycle–related metabolites, glutamine, and lactate in HBL scrambled vs. HBL HIF-1 shRNA

Protein (nmol/mg) Citrate Succinate Fumarate Malate Glutamine Lactate HBL scrambled 2.2 0.7 5.4 0.7 0.7 0.2 13.7 1.9 8.1 3.0 332 16 HBL HIF-1 kd1 4.1 0.7a 7.2 0.4a 1.4 0.2a 21.7 2.2a 14.3 3.8a 184 18a HBL HIF-1 kd2 5.8 1.0a 5.7 0.3 1.0 0.2a 16.8 1.9a 17.7 3.4a 172 19a HBL HIF-1 kd3 6.6 0.8a 6.1 0.4a 1.6 0.3a 19.3 1.6a 23.1 3.0a 185 13a

NOTE: 1H NMR spectra of the metabolome for each of the individual cell cultures when cells reached a subconﬂuent state of 70% to 80%. n ¼ 10, results are mean SD. aP < 0.05 versus scrambled.

www.aacrjournals.org Cancer Res; 72(19) October 1, 2012 5039

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 2012 American Association for Cancer Research. 5040 Downloaded from acrRs 21)Otbr1 2012 1, October 72(19) Res; Cancer rti xrsinlvl fPK.D hPasyi B el eti omxao rae ih500 with treated or normoxia in kept cells HBL in nonspeci assay used ChIP controls D, Reaction PDK3. of levels expression protein 3. Figure al. et Kluza H D A F Respiration rate pmol O .sec–1.mg–1 Serine phosphorylation of PDH

,shmtcoeve fprvt xdto nmtcodi ihihbtr sdi hssuy ,mN xrsinlvl fPK- sfrs C, isoforms. PDK1-4 of levels expression mRNA B, study. this in used inhibitors with mitochondria in oxidation pyruvate of overview schematic A, 2 12 16 20 (% relative to control HBL cells) 0 4 8 100 20 40 60 80 0

DCA Vehicle Scrambled Scrambled cancerres.aacrjournals.org Scrambled + DFO Time (min) HIF-1 kd1 Published OnlineFirstAugust3,2012;DOI:10.1158/0008-5472.CAN-12-0979

6420 HIF-1 kd2 HIF-1 kd2 HIF-1 kd3 HIF-1 kd2 + DFO HBL Input Control IgG HIF-1α Ab ﬁ

g nioisadPRcnutdb sn hl eoi N ipt.Cpue N a nlzdb C using PCR by analyzed was DNA Captured (input). DNA genomic whole using by conducted PCR and antibodies IgG c +0.5 mmol/L DCA +1 mmol/L DCA 12108 +5 mmol/L DCA E +10 mmol/L DCA S300 S293 S232 +20 mmol/L DCA PDH activity (OD 450 nm) I 0.1 0.2 0.3

Respiration rate 0 –1 –1 0 15 10 5 0 pmol O .sec .mg 1.00 on September 27, 2021. © 2012American Association for Cancer Research. 2 12 16 Scrambled HIF-1 kd3 HIF-1 kd2 HIF-1 kd1 HBL Scrambled 0 4 8 B Time (min)

K HIF-1 kd1

Skin metastasis Peritumoral tissue HIF-1 kd2 0.850.640.77

1.00 1.00 mRNA level fold change (log ) HIF-1 kd3 2 Control (ratio HBL HIF-1 kd cell lines vs. HBL scrambled) 0.5 1.0 Time (min) 2

0.57 0.59 Control 6420 0.5 mmol/L DCA

0.5 mmol/L PDK4 PDK3 PDK2 PDK1 HIF-1 kd3 HIF-1 kd2 HIF-1 kd1

0.43 0.65 1 mmol/L 0.1 0.2 0.3 0.520.570.821.00 DCA 1 mmol/L 5 mmol/L 0 PDHE1α PDHE1α PhosphoS293 0.55 0.24 YC-1 100 YC-1 50 YC-1 10 Control 2 mmol/L Time (min) 51015 510150 0 51015 1412108 Actin LDH-A HIF-1α Lamin A μmol/L μmol/L μmol/L G

J Respiration rate –1 –6 (pmol O2.sec .10 cells) 15 20 25 30 35

Control + DCA C m PDK3 o/ F o 8hus B d evda control. as served kd2 HBL hours. 18 for DFO mol/L 0.1 0.2 0.3 0.4 0.5 Actin 0 HL 60 HBL HBL (–Gln) HBL (+Gln) DCA (mmol/L) siPDK3 siCo 1.00 ρ 0 Time (min) Respiration rate 3210 Scrambled –1 –6 (pmol O .sec .10 cells) 0.48

2 HBL 20 40 60

0 HIF-1 kd1 .00.41 0.70 HIF-1 kd2

mock PDK3 MOCK 7654 Control PDK3 Actin PDK3 HIF-1 kd3 1.00 Respiration rate 0.1 0.2 0.3 0.4 (pmol O .sec–1.10–6 cells) 0 Control

2 HBL 0.84 20 40 60 acrResearch Cancer DCA 10 mmol/L 5 mmol/L 1 mmol/L 0.5 mmol/L 0.1 mmol/L HBL Co. μ 0 YC-1 10 mol/L Time (min) 0.30 YC-1 100 μmol/L siControl 1.00 DCA HBL

** siControl siPDK3 0.64 siPDK3 Published OnlineFirst August 3, 2012; DOI: 10.1158/0008-5472.CAN-12-0979

Targeting PDK to Enhance Oxidative Stress in Melanoma

Results data (12, 32), HIF-1a was expressed under normoxia in nuclei Metabolic signature of metastatic melanoma cells of melanoma cells (Fig 2A and B and Supplementary Fig. S1B) We first examined the effects of glucose and glutamine and of metastatic melanomas from patients (Supplementary fl a shortage on cell growth. In low glucose medium, HBL cells Fig. S1A). The in uence of hypoxia on selective HIF-1 expres- dramatically reduced their ability to proliferate. The absence of sion was moderate [Fig. 2A and ref. (32)]. We then used shRNA a glutamine also limited HBL cell growth (Fig. 1A). Moreover, targeting HIF-1 mRNA for knockdown of gene expression melanoma cells were partially resistant to cell death induced (Supplementary Fig. S1C and S1D and Fig. 2B) in the HBL cell a by the mitochondrial inhibitors (rotenone or oligomycin), line selected for its elevated level of HIF-1 (Fig. 2A and B). HIF- a fi whereas maximum cell death was achieved when cells were 1 knockdown cells resulted in signi cant decreased expres- incubated with bromopyruvate to inhibit glycolysis. Converse- sion of key glycolytic genes and corresponding proteins, such fi ly, the promyelocytic cell line, HL60, was critically dependent as hexokinase II, the rst step rate-limiting enzyme of glycol- fi on OXPHOS for survival [Fig. 1B and (28, 29)]. All melanoma ysis and LDH-A, which converts pyruvate into lactate, the nal cell lines studied had a cell death profile similar to that of product of glycolysis (Fig. 2C and D). In addition, a prominent mitochondrial DNA–depleted melanoma cells (HBL r0), which decrease in lactate release was observed in shRNA cells a lack functional ETC (24). Inhibition of ATP synthase with (Fig. 2E), and HIF-1 knockdown cells were relatively resistant to death induced by the glycolytic inhibitor, 2-DG (Fig. 2F). oligomycin had almost no impact on ATP in melanoma cells, 18 whereas in HL60, ATP level was highly sensitive to inhibition by Comparison of FDG accumulation in tumor xenografts oligomycin (Fig. 1C). Thus, metastatic melanoma cells, such as showed a sharply higher activity in control cells compared a respiratory deficient r0 cells, generate almost all ATP via with HIF-1 knockdown cells supporting the crucial role of a glycolysis despite abundant oxygen, a situation known as the HIF-1 in the glucose addiction of melanoma (Fig. 2G). More- a Warburg effect (30). Confirming this phenotype, a high glyco- over, HIF-1 knockdown cells were resistant to the antipro- lytic activity as measured by glucose consumption and lactate liferative effect of glutamine shortage (Fig. 2H) suggesting that a production characterized melanoma cells (Fig. 1D). Melanoma HIF-1 also contributes to the dependence of melanoma cells cell lines also exhibited a significantly lower oxygen consump- on glutamine. In comparison with control cells, HBL cells a tion rate than HL60 (Fig. 1E). Moreover, the proportion of lacking HIF-1 showed a 2-fold increase in mitochondrial melanoma respiration used to produce ATP (ATP turnover) oxygen consumption (Fig. 2I and Supplementary Fig. S2A). was far less important than that of HL60 (Fig. 1F). These results Similar mitochondrial effects were observed when cells were a were confirmed in metastatic melanoma specimens from incubated with YC-1 (Fig. 2I), an established HIF-1 inhibitor a patients (see Supplementary Table S1 and Supplementary Fig. (ref. 33 and Fig. 2B). Relative to HBL scrambled cells, HIF-1 fi S1A) compared with the surrounding tissues (Fig. 1G and H). shRNA cells presented signi cantly higher levels of tricarbox- Finally, we evaluated oxygen consumption in melanoma cells ylic acid (TCA) intermediates, accumulation of glutamine, and fi 1 isolated from cutaneous metastasis (patient 2) according to lower contents of intracellular lactate as quanti ed by H-NMR the intensity of 18FDG uptake, expressed as SUVs (Fig. 1I). (Table 1). Derepression of mitochondrial respiration in cells a fi Specimen taken from high-SUV area (inner zone) showed weak lacking HIF-1 signi cantly enhanced the fraction of respira- respiration rate and low ATP turnover as compared with tion that is used for ATP synthesis, and therefore they pro- peripheral specimen from low SUV area (Fig. 1I). Thus, the duced more ATP than scrambled shRNA cells (Fig. 2J). By metabolic phenotype of human melanoma is characterized by accomplishing these major effects on glycolysis, glutamine a high rates of aerobic glycolysis, dependence on glutamine, and metabolism, and mitochondrial respiration, HIF-1 drives the low mitochondrial activity. main metabolic changes observed in melanoma. A HIF-1a–dependent PDK3 pathway controls HIF-1a is critical for the metabolic phenotype of mitochondrial respiration metastatic melanoma We then evaluated the mechanisms by which HIF-1a down- As HIF-1 is a critical regulator of metabolism in cells under regulates mitochondrial OXPHOS. We found no change in the hypoxic conditions (31), we studied the role of HIF-1a in shape and number of mitochondria (Supplementary Fig. S2B– melanoma cells maintained in normoxia. Confirming previous S2D). Unlike previous description (34), the respiratory chain

primers speciﬁc for PDK3. E, PDH activity was determined as described in Materials and Methods. Data are representative microplate-recorded data from 3 experiments. F, HBL cells treated with DCA for 4 hours and HBL HIF-1a shRNA cells were tested in cell-ELISA kit or immunoblotted to detect phosphorylated PDH and total PDHEA1a. Results are expressed as percentage of control (untreated) HBL cells (mean SD, n ¼ 3; , P < 0.05 and , P < 0.01). G, respiration rates of HBL (without glutamine), HBLr0, and HL-60 cells treated with DCA for 15 minutes (left) or in PDK3 siRNA HBL cells (right). Bottom, HBL cells were transiently transfected with empty vector (Mock) or full-length human PDK3 cDNA (PDK3) and respiration was assessed as in Fig. 1E in cells treated with 5 mmol/L DCA for 30 minutes. H, mitochondrial respiration in tissues from HBL tumor-bearing mice was assessed as in Fig. 1G following ex vivo addition of DCA (1.2 mmol/L every 3 minutes, arrows). I, respiration rate in melanoma metastasis to the skin (black) and peritumoral tissue samples (white) from patient 1 following the sequential addition of DCA (0.6 mmol/L, arrows). J, partial 1H-NMR spectra of aqueous extracts from HBL cells treated with DCA. Results showed a 38% 24% increase for fumarate (Fum), 42% 7% for malate (~), 40% 10% for succinate (Succ), 36% 8% for citrate (*), 105% 12% for glutamate (Glu), and 68% 18% for glutamine (Gln) in DCA-treated HBL cells compared with untreated cells. Spectra are representative of 10 experiments. K, HBL cells were treated with DCA for 18 hours, then the nuclear/cytosolic fractions were subjected to immunoblotting for HIF-1a or LDH-A.

www.aacrjournals.org Cancer Res; 72(19) October 1, 2012 5041

Kluza et al.

complex IV and the expression of ETC proteins were not Fig. S4). In the r0 clone, due to its lack of basal (Fig. 1E) and affected by HIF-1a expression (Supplementary Fig. S2E and DCA-induced respiration (Fig. 3G), the pro-oxidant capacities S2F). Four different PDH kinases (PDK1–4) are known to of DCA were markedly diminished compared with HBL paren- phosphorylate the E1 moiety of the PDH complex (PDH) at tal cells (Fig. 4D). The complex I inhibitor, rotenone, dramat- 3 serine sites, thereby decreasing PDH activity and preventing ically decreased DCA-induced ROS generation in the same pyruvate from undergoing mitochondrial oxidation [Fig. 3A manner as the classic antioxidants, vitamin C and N-acetyl and ref. (35)]. HIF-1a knockdown correlated with the reduced cysteine (NAC; Fig. 4E). It must be noted that apocynin, a expression of PDK3 mRNA without significantly affecting the classic inhibitor of NADPH oxidase, which has also predom- related isoforms (Fig. 3B). Results were confirmed at the inant antioxidant capacities (36), reduced DCA-stimulated protein level (Fig. 3C and Supplementary Fig. S1B). Chromatin ROS production with about 40% ROS inhibition. The inhibitor immunoprecipitation assay showed the direct binding of HIF- of complex III, antimycin A, also reduced ROS production when 1a to the PDK3 promoter in living cells (Fig. 3D). In agreement combined with DCA, whereas the complex IV and V inhibitors with the aforementioned data on PDK3 downregulation, PDH had no noticeable effects (Fig. 4E). This would indicate that activity was enhanced in HIF-1a knockdown cells or in mel- mitochondrial ROS promoted by DCA originated from com- anoma cells incubated in the presence of YC-1 (Fig. 3E and plex I and/or III. Altogether, our results suggest that the HIF- Supplementary Fig. S3). Increased PDH activity was also 1a/PDK3 axis protects melanoma from mitochondrial ROS observed in cells transfected with siRNA directed against PDK3 production. or in the presence of the pyruvate dehydrogenase kinase (PDK) inhibitor, DCA (Fig. 3E and Supplementary Fig. S3). Dephos- Inhibition of the PDK3-dependent HIF-1a pathway phorylation of PDH in HIF-1a knockdown cells or in HBL cells promotes antimelanoma activity of elesclomol treated with DCA (Fig. 3F) correlated with the activation of Finally, we investigated whether mitochondrial ROS gener- PDH. ation promoted by DCA and elesclomol had potent antimela- Accordingly, exposure of HBL cells (Fig. 3G), HBL tumors noma activities. The combination of DCA and elesclomol from SCID mice (Fig. 3H), or freshly resected cutaneous inhibited cell viability more than each treatment did alone metastases (Fig. 3I) to DCA resulted in a dose-dependent (Fig. 5A–C and Supplementary Fig. S5). In congruence with increase of mitochondrial respiration regardless of the pres- these findings, PDK3 inhibition rendered HBL cells highly ence or absence of glutamine (Fig. 3G). No such effects were sensitive to elesclomol-induced cell death (Supplementary Fig. observed in HBL r0 cells or in the OXPHOS-dependent cell line S5). Selected concentration curves indicate a synergistic effect HL60 (Fig. 3G, top left). DCA-induced increase in respiration for HBL cells when DCA was combined with elesclomol (Fig. was abolished by overexpression of PDK3 (Fig. 3G, bottom) 5A). Interestingly, combination of DCA and elesclomol was suggesting that PDK3 is an important target for DCA. More- sufficient to induce oxidative cell death in A375 melanoma cells over, DCA decreased the nuclear level of HIF-1a as well as its with acquired resistance to vemurafenib (Fig. 5D). target LDH-A (Fig. 3K). There was a similar trend toward To explore the therapeutic usefulness of pharmacologic increased mitochondrial respiration in melanoma cells trans- combination, we sought to determine whether DCA treatment fected with PDK3 siRNA (Fig. 3G, top right), comparable with sensitized HBL xenografts to elesclomol in vivo. DCA or ele- that observed for cells lacking HIF-1a (Fig. 2I). We noted a sclomol alone reduced growth of HBL xenografts (Fig. 5E). In significant elevation in several TCA cycle intermediates, glu- comparison, the combination of DCA and elesclomol was more tamine, and a low lactate content in DCA-treated cells (Fig. 3J) effective in delaying tumor growth than single agent alone. mirroring the phenotype observed in HIF-1a kd cells (Table 1). Very similar results were obtained in a preclinical model Thus, a HIF-1a/PDK3 axis contributes to the control of mito- consisting to evaluate the response of human tumor fragments chondrial respiration observed in melanoma. grafted into SCID mice to drug combination (Fig. 5F). Treat- ment of mice with the combination of DCA and elesclomol was Inhibition of the HIF-1a–dependent PDK3 pathway more effective in accumulating oxidatively modified proteins, enhances mitochondrial ROS generation reducing cell proliferation, and inducing apoptosis of HBL We next decided to investigate whether the enhanced xenografts than single agent alone (Fig. 5G). Altogether, these mitochondrial respiration would lead to increased mitochon- data suggest that the antimelanoma activity of elesclomol, drial ROS generation. We observed a significant increase in the although effective alone, can be significantly enhanced by DCA. ROS levels when PDK3 activity was inhibited by siRNA or DCA (Fig. 4A and Supplementary Fig. S4). Incubating PDK3 siRNA- or DCA-treated HBL cells with distinct ROS inducers further Discussion increased ROS levels. The pro-oxidative effect of DCA was Our results show that melanoma cells use mainly glycolysis, maintained after glutamine withdrawal (Fig. 4A). Remarkably, and not mitochondria, for energy production and cell prolif- DCA had no additional effect on ROS production in PDK3 eration. In agreement with other reports (9, 19), we observed siRNA knockdown cells, and PDK3 overexpression impeded that glutamine also supports melanoma cell growth. Glucose DCA-induced increase in ROS generation confirming PDK3 as and glutamine metabolic pathways are able to compensate a potential target for DCA (Fig. 4B). Consistently, exposure of from one another to promote cancer cell survival (37, 38). cells lacking HIF-1a to the ROS inducers also generated more However, unlike the model originally stipulated by Warburg ROS than cells expressing HIF-1a (Fig. 4C and Supplementary (39), melanoma cells maintain functional mitochondria. Thus,

5042 Cancer Res; 72(19) October 1, 2012 Cancer Research

Targeting PDK to Enhance Oxidative Stress in Melanoma

A 200 300 +NAC siCo Gln 160 siPDK3 w/o gln 200 DCFDA (%) DCFDA

DCFDA (%) DCFDA 120 2 2

80 100 40 Relative increase in Relative Relative increase in Relative 0 0 MFI of CM-H MFI of CM-H DCA Menadione N-EM N-EM

Figure 4. A, cells were treated with Control Elesclomol 10 mmol/L menadione, 5 mmol/L N- Elesclomol EM, or 300 nmol/L elesclomol for 8 Menadione hours and PDK3 was inhibited by siRNA (left) or by DCA (0.5 mmol/L 50 B 400 for 8 hours). When indicated, HBL 40 cells were cultured in the presence or

DCFDA 300 absence of glutamine before ROS 2 30 determination (mean SD, n ¼ 4; 200 t-test, , P < 0.05 and , P < 0.01). 20 B, left, cells were treated as before 100 ROS determination; right, HBL cells 10

were transiently transfected with MFI of CM-H 0 0 empty vector (Mock) or full-length % cells producing ROS human PDK3 cDNA (PDK3) and ROS DCA Mock production was assessed in cells PDK3 Control treated with 5 mmol/L DCA for 12 siPDK3 hours (mean SD, n ¼ 4; t-test, , P < 0.05 and , P < 0.01). C, cells + DCA Mock siPDK3 + DCA

were treated as in A and HIF-1 was PDK3 + DCA ov. inhibited by 100 mmol/L YC-1 or by shRNA (right), then ROS generation ﬂ C 100 100 was measured by ow cytometry Untreated +NAC HBL scrambled +NAC n ¼ P < (mean SD, 3; t-test, , 0.05 80 YC-1 100 80 HIF-1 kd2 and , P < 0.01). D, HBL and HBL r0 μmol/L cells were incubated with 5 mmol/L

DCFDA (%) DCFDA 60 (%) DCFDA 60 DCA and/or 10 mmol/L NAC for 12 2 2 hours before cytoﬂuorometric 40 40 analysis. The mean ﬂuorescence intensity (MFI) values are shown. 20 20 Relative increase in Relative E, cells were preincubated for 30 increase in Relative 0 0 MFI of CM-H minutes with 1 mmol/L rotenone, MFI of CM-H 1 mmol/L antimycin A (AA), sodium m N-EM azide (NaN3), 2 g/mL oligomycin, N-EM Control 100 mmol/L apocynin, or vitamin C Control m Elesclomol (100 mol/L) and NAC (10 mmol/L) Elesclomol Menadione Menadione then incubated for 8 hours with 0.5 mmol/L DCA. Results are percentages of ROS inhibition DEControl DCA DCA + NAC 100 calculated in comparison with DCA- 102 505 97 n ¼ stressed cells (mean SD, 3). 80

HBL 60

212 314 207 Inhibition of 20

DCA-induced ROS (%) DCA-induced ROS 0 HBL ρ 0 3 AA VitC NAC NaN CM-H2DCFDA CM-H2DCFDA CM-H2DCFDA Apocynin Rotenone Oligomycin

www.aacrjournals.org Cancer Res; 72(19) October 1, 2012 5043

Kluza et al.

ABC 60 100 5 6 h Control 24 h 80 4 40 60 3 DCA 40 2 Control Elesclomol 20 Elesclomol 1 % Cell viability 20

Elesclomol + NAC DCA (mmol/L) % of dead cells (PI-negative cells) (PI-negative DCA + 0 0 0 0.05 0.1 0.5 1 5 0246810121416 Elesclomol DCA (mmol/L) Elesclomol (nmol/L) 0 500 1,000 1,500 0 Colony number HBL spheroids in multicellular DCA DCA + Control Elesclomol D G Elesclomol Control DCA DCA + Control Vemurafenib DCA Elesclomol Elesclomol 90

60 Elesclomol Elesclomol + DCA 30 A375-R A375

% DNPH-positive cells % DNPH-positive 0 H2DCFDA DCA DCA + Protein oxidation Control Elesclomol (anti-DNPH Ab staining) Elesclomol E Control DCA Control 90 3,000 Control Elesclomol

DCA 60 2,000 DCA Elesclomol Elesclomol + DCA 30 Elesclomol Elesclomol 1,000

+ DCA cells % Ki-67–positive 0 Tumor volume ( μ L) volume Tumor Elesclomol DCA DCA + DCA + 0 Cell proliferation Control 0 10 20 30 40 Elesclomol Days (anti-Ki-67 Ab staining) Elesclomol Control DCA F 2,500 Control 90 DCA 2,000 Elesclomol DCA + 60 Elesclomol 1,500 Elesclomol Elesclomol + DCA 30 1,000

Tumor volume ( μ L) volume Tumor 500 0 % TUNEL-positive % TUNEL-positive cells

0 DCA DCA + 32 37 42 47 Control Cell death Elesclomol Days post tumor cell Elesclomol (TUNEL staining) implantation

Figure 5. A, left, combination effect of DCA plus elesclomol on HBL cells incubated for 8 hours before assessment of cell viability. Cells were pretreated with 10 mmol/L NAC for 1 hour (mean SD, n ¼ 3). Right, IC50 isobologram of the combination treatments. A plot under the line indicates a synergistic combination. B, colony-forming ability of HBL cells treated with 0.5 mmol/L DCA and/or 300 nmol/L elesclomol. Two weeks later crystal violet–stained colonies were counted. C, HBL spheroids were grown for 15 days then treated as in B. or #, P < 0.05 and ## or , P < 0.01. D, effects of the combination DCA þ elesclomol on vemurafenib-resistant melanoma cells. The A375 parental cell line and the A375-R vemurafenib-resistant cells were exposed to 1 mmol/L DCA

and/or 300 nmol/L elesclomol for 8 hours or 300 nmol/L vemurafenib for 48 hours and submitted to CM-H2DCFDA staining. The gray peak represents cells not exposed to drugs. Numbers refer to the percentage of cell death determined by PI staining cells from parallel culture. E, the tumor-bearing mice were divided into 4 treatment groups: control, DCA alone, elesclomol alone, or the combination of elesclomol and DCA (mean SD, n ¼ 4to6), P < 0.05. F, xenografted tumor fragments from patient 1 were treated (indicated by arrow) following the protocol described in E. G, sections from the tumors of HBL-injected mice were stained with an antidiphenylhydrazine antibody for detection of carbonylated proteins as a marker of irreversible oxidative damage, with an anti- Ki67 antibody to assess proliferation and by TUNEL assay to detect apoptotic cell death. Insert corresponds to positive control for TUNEL (mean SD, n ¼ 4, , P < 0.05 vs. control).

5044 Cancer Res; 72(19) October 1, 2012 Cancer Research

Targeting PDK to Enhance Oxidative Stress in Melanoma

we showed that inhibition of the HIF-1a/PDK3 axis is able to To best translate in vitro results into clinical application, restore respiration in mitochondria of melanoma cells. These we have chosen the combination of existing drugs (DCA findings confirm that cancer cells possess a highly adaptable and elesclomol) that are already in the clinic. Elesclomol is metabolism. Reduction in LDH-A activity (40) causes dere- one of the first mitochondrial-targeted drugs that has pression of mitochondrial respiration (25). Inhibition of PDK recently entered the clinical trials (7). Elesclomol does not favors mitochondrial oxidation and ROS production in lung act by specific targeting of mitochondrial proteins but most carcinoma (41). Forced oxidative metabolism interferes with likely by interfering with the electron flow to promote tumor cells proliferation and survival (25, 40, 41). Thereby, the mitochondrial ROS production (6). Thus, we have shown metabolic flexibility of cancer cells creates a new perspective that the DCA-dependent enhancement of the electron flow for the development of therapeutic approaches. impaired by elesclomol results in huge production of ROS. HIF-1a has been found highly expressed in melanoma cells DCA, previously used to treat mitochondrial diseases with- kept under normoxic conditions (12, 32). The elevated expres- out major toxicity, is currently undergoing clinical evalua- sion of HIF-1a is correlated with the advanced stage of tion for cancer (46). DCA irreversibly inhibits all PDK melanoma (12). Our results are consistent with the view that isoenzymes (47). We have shown that knocking down melanoma progression is associated with the apparition of HIF-1a markedly reduced PDK3 expression but not other changes in mitochondrial metabolism in a HIF-1a dependent isoenzymes under normoxic conditions (Fig. 3B). However, manner. Conversely, nonmetastatic primary melanomas upre- both PDK1 and PDK3 were upregulated under hypoxic gulate genes involved in mitochondrial oxidative phosphory- conditions (not shown), and it has been proven that PDK1 lation (42). prevents the overproduction of mitochondrial ROS in Our study defines the HIF-1a/PDK3 axis as a sensor for response to hypoxia (48, 49). Because solid tumors consist metabolic stress that regulates mitochondrial ROS level under of both normoxic and hypoxic regions, the use of a broad- normoxia. Overexpression of HIF-1a reduces the basal level of spectrum inhibitor of all 4 PDK isoenzymes, such as DCA, is ROS in hypoxic cancer cells (43). The isoenzyme PDK3 is of particular interest in vivo to fully reactivate mitochon- expressed in several cancer cells, such as neuroblastoma, colon dria. However, the main disadvantage of DCA, when used cancer (27), or leukemia (25). Hypoxic activation of the PDK3- alone, is that its in vivo antitumor activity is only apparent dependent HIF-1a pathway has been previously involved in at high concentrations (46). Our results suggest that DCA tumor progression and drug resistance (27). In agreement with canbealsousedincombinationregimentoboostthe our findings, the transcription profiling of human cancer cell mitochondrial electron flux and then potentiate the anti- lines confirmed the high expression of PDK3 in melanomas tumor activity of pro-oxidant drugs, such as elesclomol. (deposited in Array Express database, accession number E- Thus, our results offer proof-of-concept for using elesclomol MTAB-37). PDK3 is the only isoform reported to be not combined with DCA against metastatic melanoma includ- inhibited by a high concentration of pyruvate (44). Interest- ing those who have become resistant to BRAF inhibitors. ingly, high pyruvate and lactate concentrations result in In conclusion, inactivation of the HIF-1a/PDK3 axis pre- nonhypoxic HIF-1a stabilization (45). Thus, as suggested sents new therapeutic opportunities against metastatic (45), HIF-1a-mediated PDK3 expression may favor HIF-1a melanoma. stabilization independently of hypoxia in a self-amplifying loop that contributes to locking the melanoma cell metabolism into Disclosure of Potential Conflicts of Interest aerobic glycolysis. No potential conflicts of interest were disclosed. Our findings provide an attractive explanation for the pre- dictive value of LDH in patients treated with elesclomol (7). Authors' Contributions Because LDH-A is transcriptionally regulated by HIF-1a (Fig. Conception and design: J. Kluza, P. Corazao-Rozas, M. Jendoubi, A. Jonneaux, C. Ballot, S. Valable, R. Polakowska, L. Mortier, P. Marchetti 2C and D), serum LDH may be considered as an indirect Development of methodology: J. Kluza, P. Corazao-Rozas, Y. Touil, M. Jen- marker of the HIF-1a–mediated defective mitochondrial activ- doubi, A. Jonneaux, C. Ballot, L. Mortier Acquisition of data (provided animals, acquired and managed patients, ity. In that way, we propose that tumors from patients with provided facilities, etc.): J. Kluza, P. Corazao-Rozas, Y. Touil, M. Jendoubi, C. high LDH levels have also inactive mitochondria, a situation Maire, P. Guerreschi, A. Jonneaux, C. Ballot, S. Balayssac, S. Valable, A. Corroyer- that we found associated with resistance to elesclomol- Dulmont, M. Bernaudin, M. Malet-Martino, E. Martin de Lasalle, P. Maboudou in vitro Analysis and interpretation of data (e.g., statistical analysis, biostatistics, induced cell death (6). Under such conditions, inhibi- computational analysis): J. Kluza, P. Corazao-Rozas, M. Jendoubi, C. Maire, A. tion of HIF-1a and LDH expression by DCA (Fig. 3K) can Jonneaux, C. Ballot, S. Valable, L. Mortier contribute to restore sensitivity to elesclomol. Finally, we have Writing, review, and/or revision of the manuscript: J. Kluza, P. Guerreschi, S. Valable, M. Bernaudin, R. Polakowska, L. Mortier, P. Marchetti shown that combined inhibitors of the HIF-1a/PDK3 axis Administrative, technical, or material support (i.e., reporting or orga- along with pro-oxidant drugs have potential antimelanoma nizing data, constructing databases): J. Kluza, P. Guerreschi, L. Mortier a Study supervision: J. Kluza, P. Formstecher, R. Polakowska, L. Mortier, P. capacities. The hypoxia-independent activation of HIF-1 , Marchetti found in certain metastatic melanomas (such as those with activation of the MITF pathway or harboring BRAF mutation), Acknowledgments would confer particular sensitivity to this drug combination. The authors thank Dr. Aubert (CHRU, Lille) for immunostaining, Pr. Huglo Thus, in line with our results, this new therapeutic approach (CHRU, Lille) for PET-Scan analysis, Drs. Bates, Blackman, and Proia (Synta fi Pharmaceuticals Corp.) for providing elesclomol, Pr. Ghanem and Dr. Journe would be likely bene cial to patients progressing on BRAF (Institut Jules Bordet, Bruxelles) for melanoma cell lines and Thomas Cruz for inhibitors. technical help.

www.aacrjournals.org Cancer Res; 72(19) October 1, 2012 5045

Kluza et al.

Grant Support de France. P. Corazao-Rozas is a recipient of a CHRU Lille-Region Nord-Pas This work received a ﬁnancial support from INSERM, UNIVERSITE DE de Calais fellowship. ¸ The costs of publication of this article were defrayed in part by the payment of LILLE II, SocieteFrancaise de Dermatologie (P. Marchetti), Ligue contre le advertisement Cancer (Comite de l'Aisne; P. Marchetti and P. Formstecher), Societe page charges. This article must therefore be hereby marked in de Recherche Dermatologique (L. Mortier), BMS-Groupe de Cancerologie accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Cutanee (J. Kluza), and a special ﬁnancial support from the Association pour l'Etude des Anomalies Congenitales Neurodev of Pr. B. Poupard Received March 14, 2012; revised July 27, 2012; accepted July 27, 2012; (P. Guerreschi). J. Kluza received a fellowship from ARC and the Fondation published OnlineFirst August 3, 2012.

References 1. Fruehauf JP, Trapp V. Reactive oxygen species: an Achilles' heel of lipogenesis in hypoxic melanoma cells. Pigment Cell Melanoma Res melanoma? Expert Rev Anticancer Ther 2008;8:1751–7. 2012;25:375–83. 2. Yamaura M, Mitsushita J, Furuta S, Kiniwa Y, Ashida A, Goto Y, et al. 20. Semenza GL. Hypoxia-inducible factor 1: regulator of mitochondrial NADPH oxidase 4 contributes to transformation phenotype of mela- metabolism and mediator of ischemic preconditioning. Biochim Bio-

noma cells by regulating G2–M cell cycle progression. Cancer Res phys Acta 2011;1813:1263–8. 2009;69:2647–54. 21. Wise DR, Ward PS, Shay JES, Cross JR, Gruber JJ, Sachdeva UM, 3. Ibanez~ I, Notcovich C, Policastro L, Duran H. Reactive oxygen species et al. Hypoxia promotes isocitrate dehydrogenase-dependent carbox- in the biology of melanoma. In: Tanaka Y, editor. Breakthroughs in ylation of a-ketoglutarate to citrate to support cell growth and viability. melanoma research. Rijeka (Croatia): InTech; 2011. p. 1–32. Proc Natl Acad Sci U S A 2011;108:19611–6. 4. Trachootham D, Alexandre J, Huang P. Targeting cancer cells by ROS- 22. Takeuchi H, Kondo Y, Fujiwara K, Kanzawa T, Aoki H, Mills GB, et al. mediated mechanisms: a radical therapeutic approach? Nat Rev Drug Synergistic augmentation of rapamycin-induced autophagy in malig- Discov 2009;8:579–91. nant glioma cells by phosphatidylinositol 3-kinase/protein kinase B 5. Fruehauf JP, Zonis S, al-Bassam M, Kyshtoobayeva A, Dasgupta C, inhibitors. Cancer Res 2005;65:3336–46. Milovanovic T, et al. Melanin content and downregulation of glutathione 23. Cormier CY, Mohr SE, Zuo D, Hu Y, Rolfs A, Kramer J, et al. Protein S-transferase contribute to the action of L-buthionine-S-sulfoximine structure initiative material repository: an open shared public resource on human melanoma. Chem Biol Interact 1998;111–112:277–305. of structural genomics plasmids for the biological community. Nucleic 6. Blackman RK, Cheung-Ong K, Gebbia M, Proia DA, He S, Kepros J, Acids Res 2010;38:D743–9. et al. Mitochondrial electron transport is the cellular target of the 24. Ballot C, Kluza J, Lancel S, Martoriati A, Hassoun SM, Mortier L, et al. oncology drug elesclomol. PLoS ONE 2012;7:e29798. Inhibition of mitochondrial respiration mediates apoptosis induced by 7. O'Day S, Gonzalez R, Lawson D, Weber R, Hutchins L, Anderson C, the anti-tumoral alkaloid lamellarin D. Apoptosis 2010;15:769–81. et al. Phase II, randomized, controlled, double-blinded trial of weekly 25. Kluza J, Jendoubi M, Ballot C, Dammak A, Jonneaux A, Idziorek T, et al. elesclomol plus paclitaxel versus paclitaxel alone for stage IV meta- Exploiting mitochondrial dysfunction for effective elimination of ima- static melanoma. J Clin Oncol 2009;27:5452–8. tinib-resistant leukemic cells. PLoS ONE 2011;6:e21924. 8. Kim J-W, Dang CV. Cancer's molecular sweet tooth and the Warburg 26. Beckonert O, Keun HC, Ebbels TMD, Bundy J, Holmes E, Lindon JC, effect. Cancer Res 2006;66:8927–30. et al. Metabolic profiling, metabolomic and metabonomic procedures 9. Scott DA, Richardson AD, Filipp FV, Knutzen CA, Chiang GG, Ronai for NMR spectroscopy of urine, plasma, serum and tissue extracts. Nat ZA, et al. Comparative metabolic flux profiling of melanoma cell lines: Protoc 2007;2:2692–703. beyond the Warburg effect. J Biol Chem 2011;286:42626–34. 27. LuCW,LinSC,ChenKF,LaiYY,TsaiSJ.Inductionofpyruvate 10. Berry JD, Cook GJR. Positron emission tomography in oncology. Br dehydrogenase kinase-3 by hypoxia-inducible factor-1 promotes Med Bull 2006;79–80:171–86. metabolic switch and drug resistance. J Biol Chem 2008;283: 11. Hersey P, Watts RN, Zhang XD, Hackett J. Metabolic approaches to 28106–14. treatment of melanoma. Clin Cancer Res 2009;15:6490–4. 28. Moreno-Sanchez R, Saavedra E, Rodríguez-Enríquez S, Gallardo- 12. Mills CN, Joshi SS, Niles RM. Expression and function of hypoxia Perez JC, Quezada H, Westerhoff HV. Metabolic control analysis inducible factor-1 alpha in human melanoma under non-hypoxic indicates a change of strategy in the treatment of cancer. Mitochon- conditions. Mol Cancer 2009;8:104. drion 2010;10:626–39. 13. Monsel G, Ortonne N, Bagot M, Bensussan A, Dumaz N. c-Kit mutants 29. Berridge MV, Herst PM, Tan AS. Metabolic flexibility and cell hierarchy require hypoxia-inducible factor 1alpha to transform melanocytes. in metastatic cancer. Mitochondrion 2010;10:584–8. Oncogene 2010;29:227–36. 30. Warburg O. On the origin of cancer cells. Science 1956;123:309–14. 14. Aprelikova O, Pandolfi S, Tackett S, Ferreira M, Salnikow K, Ward Y, 31. Denko NC. Hypoxia, HIF1 and glucose metabolism in the solid tumour. et al. Melanoma antigen-11 inhibits the hypoxia-inducible factor prolyl Nat Rev Cancer 2008;8:705–13. hydroxylase 2 and activates hypoxic response. Cancer Res 2009;69: 32. Kuphal S, Winklmeier A, Warnecke C, Bosserhoff A-K. Constitutive 616–24. HIF-1 activity in malignant melanoma. Eur J Cancer 2010;46: 15. Spinella F, Rosano L, Di Castro V, Decandia S, Nicotra MR, Natali PG, 1159–69. et al. Endothelin-1 and endothelin-3 promote invasive behavior via 33. Yeo E-J, Chun Y-S, Cho Y-S, Kim J, Lee J-C, Kim M-S, et al. YC-1: a hypoxia-inducible factor-1 in human melanoma cells. Cancer Res potential anticancer drug targeting hypoxia-inducible factor 1. J Natl 2007;67:1725–34. Cancer Inst 2003;95:516–25. 16. Iervolino A, Trisciuoglio D, Ribatti D, Candiloro A, Biroccio A, Zupi G, 34. Fukuda R, Zhang H, Kim J-W, Shimoda L, Dang CV, Semenza GL. HIF- et al. Bcl-2 overexpression in human melanoma cells increases angio- 1 regulates cytochrome oxidase subunits to optimize efficiency of genesis through VEGF mRNA stabilization and HIF-1-mediated tran- respiration in hypoxic cells. Cell 2007;129:111–22. scriptional activity. FASEB J 2002;16:1453–5. 35. Semenza GL. Regulation of cancer cell metabolism by hypoxia-induc- 17. Bertolotto C, Lesueur F, Giuliano S, Strub T, de Lichy M, Bille K, et al. A ible factor 1. Semin Cancer Biol 2009;19:12–6. sumoylation-defective MITF germline mutation predisposes to mela- 36. Heumuller€ S, Wind S, Barbosa-Sicard E, Schmidt HH, Busse R, noma and renal carcinoma. Nature 2011;480:94–8. Schroder€ K, et al. Apocynin is not an inhibitor of vascular NADPH 18. Kumar SM, Yu H, Edwards R, Chen L, Kazianis S, Brafford P, et al. oxidases but an antioxidant. Hypertension 2008;51:211–7. Mutant V600E BRAF increases hypoxia inducible factor-1 alpha 37. Yang C, Sudderth J, Dang T, Bachoo RM, Bachoo RG, McDonald JG, expression in melanoma. Cancer Res 2007;67:3177–84. et al. Glioblastoma cells require glutamate dehydrogenase to survive 19. Filipp FV, Scott DA, Ronai ZA, Osterman AL, Smith JW. Reverse TCA impairments of glucose metabolism or Akt signaling. Cancer Res cycle flux through isocitrate dehydrogenases 1 and 2 is required for 2009;69:7986–93.

5046 Cancer Res; 72(19) October 1, 2012 Cancer Research

Targeting PDK to Enhance Oxidative Stress in Melanoma

38. Cheng T, Sudderth J, Yang C, Mullen AR, Jin ES, Mates JM, et al. 44. Baker JC, Yan X, Peng T, Kasten S, Roche TE. Marked differences Pyruvate carboxylase is required for glutamine-independent growth of between two isoforms of human pyruvate dehydrogenase kinase. tumor cells. Proc Natl Acad Sci U S A 2011;108:8674–9. J Biol Chem 2000;275:15773–81. 39. Warburg O. On respiratory impairment in cancer cells. Science 45. Lu H, Forbes RA, Verma A. Hypoxia-inducible factor 1 activation by 1956;124:269–70. aerobic glycolysis implicates the Warburg effect in carcinogenesis. 40. Fantin VR, St-Pierre J, Leder P. Attenuation of LDH-A expression J Biol Chem 2002;277:23111–5. uncovers a link between glycolysis, mitochondrial physiology, and 46. Garber K. Oncology's energetic pipeline. Nat Biotechnol 2010;28: tumor maintenance. Cancer Cell 2006;9:425–34. 888–91. 41. Bonnet S, Archer SL, Allalunis-Turner J, Haromy A, Beaulieu C, 47. Bowker-Kinley MM, Davis WI, Wu P, Harris RA, Popov KM. Evidence Thompson R, et al. A mitochondria-Kþ channel axis is suppressed for existence of tissue-speciﬁc regulation of the mammalian pyruvate in cancer and its normalization promotes apoptosis and inhibits cancer dehydrogenase complex. Biochem J 1998;329:191–6. growth. Cancer Cell 2007;11:37–51. 48. Kim J-W, Tchernyshyov I, Semenza GL, Dang CV. HIF-1-mediated 42. Xu K, Mao X, Mehta M, Cui J, Zhang C, Xu Y. A comparative study of expression of pyruvate dehydrogenase kinase: a metabolic switch gene-expression data of basal cell carcinoma and melanoma reveals required for cellular adaptation to hypoxia. Cell Metab 2006;3: new insights about the two cancers. PLoS ONE 2012;7:e30750. 177–85. 43. Sasabe E, Tatemoto Y, Li D, Yamamoto T, Osaki T. Mechanism of HIF- 49. Papandreou I, Cairns RA, Fontana L, Lim AL, Denko NC. HIF-1 1alpha-dependent suppression of hypoxia-induced apoptosis in mediates adaptation to hypoxia by actively downregulating mitochon- squamous cell carcinoma cells. Cancer Sci 2005;96:394–402. drial oxygen consumption. Cell Metab 2006;3:187–97.

www.aacrjournals.org Cancer Res; 72(19) October 1, 2012 5047

Inactivation of the HIF-1α/PDK3 Signaling Axis Drives Melanoma toward Mitochondrial Oxidative Metabolism and Potentiates the Therapeutic Activity of Pro-Oxidants

Jérome Kluza, Paola Corazao-Rozas, Yasmine Touil, et al.

Cancer Res 2012;72:5035-5047. Published OnlineFirst August 3, 2012.

Updated version Access the most recent version of this article at: doi:10.1158/0008-5472.CAN-12-0979

Supplementary Access the most recent supplemental material at: Material http://cancerres.aacrjournals.org/content/suppl/2012/08/02/0008-5472.CAN-12-0979.DC1

Cited articles This article cites 48 articles, 18 of which you can access for free at: http://cancerres.aacrjournals.org/content/72/19/5035.full#ref-list-1

Citing articles This article has been cited by 7 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/72/19/5035.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/72/19/5035. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.