pyruvate dehydrogenase pdh-e1b controls tumor progression by … · metabolism and chemical biology...

Metabolism and Chemical Biology

Pyruvate Dehydrogenase PDH-E1b ControlsTumor Progression by Altering the MetabolicStatus of Cancer CellsRyo Yonashiro1, Kayoko Eguchi1, Masaki Wake2, Norihiko Takeda2, and Koh Nakayama1

Abstract

Downregulation of pyruvate dehydrogenase (PDH) is crit-ical for the aberrant preferential activation of glycolysis incancer cells under normoxic conditions. Phosphorylation-dependent inhibition of PDH is a relevant event in thisprocess, but it is not durable as it relies on PDH kinases thatare activated ordinarily under hypoxic conditions. Thus, itremains unclear how PDH is durably downregulated in cancercells that are not hypoxic. Building on evidence that PDHactivity depends on the stability of a multi-protein PDHcomplex, we found that the PDH-E1b subunit of the PDHcomplex is downregulated to inhibit PDH activity under con-ditions of prolonged hypoxia. After restoration of normoxicconditions, reduced expression of PDH-E1b was sustained

such that glycolysis remained highly activated. Notably,PDH-E1b silencing in cancer cells produced a metabolic statestrongly resembling the Warburg effect, but inhibited tumorgrowth. Conversely, enforced exogenous expression of PDH-E1b durably increased PDH activity and promoted the malig-nant growth of breast cancer cells in vivo. Taken together, ourresults establish the specific mechanism through which PDHacts as an oncogenic factor by tuning glycolytic metabolism incancer cells.

Significance: This seminal study offers a mechanistic explana-tion for why glycolysis is aberrantly activated in normoxic cancercells, offering insights into this long-standing hallmark of cancertermed theWarburg effect. Cancer Res; 78(7); 1592–603.�2018 AACR.

IntroductionCells use atmospheric oxygen to efficiently produce energy.

When cells confront oxygen-limited environments, they adapt bytriggering the hypoxic response, which alters cellular functionsincluding metabolism, respiration, vascularization, and erythro-poiesis (1). Hypoxia-inducible factor (HIF) plays a critical role inthis process by inducing multiple genes involved in the acuteresponse to these conditions (2). We previously showed thatCREB and NFkB become activated and play important physio-logic roles during prolonged hypoxia (3). Although the acutehypoxic response induces drastic changes in cellular status, thechronic response may serve to maintain the resultant status overthe course of long-term hypoxia.

Many types of cancer cells exhibit a specific type of irregularmetabolism characterized by high dependence on glycolysis forenergy. This "Warburg effect" defined as aerobic glycolysis, ischaracterized by cancer phenotypes such as high glycolytic rateand elevated lactate production under normoxia (4). Hypoxia

also shifts cellular metabolism into a glycolytic mode andincreases lactate production. Because tumor cells are oftenexposed to hypoxia under physiologic conditions, it is possiblethat their sustained hypoxic metabolism is a direct cause of theWarburg effect; however, the underlying mechanism remainsincompletely understood.

Previously, we showed that PHD3 forms a large complex inresponse to hypoxia (5). Pyruvate dehydrogenase (PDH), anenzyme that catalyzes conversion of pyruvate into acetyl-CoA, isa component of this complex (6). PDH consists of five majorsubunits, PDH-E1a, PDH-E1b, PDH-E2, PDH-E3, and PDH-E3BP, which in mammals form a complex larger than 5,000 kDa(7). Interaction of PHD3 with PDH stabilizes the PDH complexand plays a role in maintaining its enzymatic activity (6). Underhypoxic conditions, depletion of PHD3 destabilizes the complex,leading to a reduction in PDH activity. Consistent with this,PHD3�/� cells contain reduced levels of PDH complex and havediminished PDH activity.

Inhibition of PDH activity is mediated by phosphorylationof PDH-E1a by pyruvate dehydrogenase kinase 1 (PDK1), akinase induced under hypoxic conditions (8, 9). This mecha-nism is important for the inhibition of PDH under hypoxia,leading to primarily glycolytic cellular metabolism, and is alsoconsidered critical for induction of the Warburg effect in cancercells (10). However, PDH is phosphorylated primarily underhypoxic conditions, when expression of PDK is high, andbecomes dephosphorylated upon reoxygenation, thus shiftingcellular metabolism back to the normal mode. Because theWarburg effect is defined as aerobic glycolysis, it is possible thatmechanisms other than phosphorylation are responsible forPDH inhibition under normoxia. Here, we describe a previ-ously unknown mechanism of PDH inhibition, mediated bydownregulation of the PDH-E1b subunit.

1Oxygen Biology Laboratory, Medical Research Institute, Tokyo Medical andDental University (TMDU), Bunkyo-ku, Tokyo, Japan. 2Department of Cardio-vascular Medicine, Graduate School of Medicine, The University of Tokyo,Bunkyo-ku, Tokyo, Japan.

Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).

Corresponding Author: Koh Nakayama, Oxygen Biology Laboratory, MedicalResearch Institute, Tokyo Medical and Dental University (TMDU), 1-5-45 Yush-ima, Bunkyo-ku, Tokyo 113-8510, Japan. Phone/Fax: 813-5803-4815; E-mail:[email protected]

doi: 10.1158/0008-5472.CAN-17-1751

�2018 American Association for Cancer Research.

CancerResearch

Cancer Res; 78(7) April 1, 20181592

on August 7, 2020. © 2018 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst February 7, 2018; DOI: 10.1158/0008-5472.CAN-17-1751

http://crossmark.crossref.org/dialog/?doi=10.1158/0008-5472.CAN-17-1751&domain=pdf&date_stamp=2018-3-15

http://cancerres.aacrjournals.org/

Materials and MethodsCell culture

MCF7, MDA-MB-231, and SKBR3 breast cancer cells, and HeLacells were obtained from ATCC and cultured in DMEM (high-glucose; Wako) containing 10% FBS and antibiotics. Mouseembryonic fibroblasts were maintained in DMEM supplementedwith 10%FBS, 0.1mmol/L nonessential amino acids, 0.2mmol/L2-mercaptoethanol, and antibiotics. Mycoplasma contaminationwas monitored with PCR or DAPI staining. Cell reauthenticationtest was not performed. All the cell lines were obtained between2007and2012, and a frozen vial of each cell linewas thawed every2–3 months (or approximately 25 passages.)

Hypoxic treatmentCells were treated under hypoxic conditions (1% O2 and 5%

CO2, balanced with N2) in a hypoxia workstation (HirasawaWorks). An oxygen sensor was used to regulate the oxygenconcentration inside the workstation, which was maintained at1% throughout the experiment (MC-8G-S, Iijima Electrics).

Reagents and antibodiesThe following antibodies were used: anti-b-actin (Sigma

Japan); anti-HIF-1a (Novus Biologicals); anti-HIF-2a (NovusBiologicals); anti-CREB (Cell Signaling Technology Japan);anti-phosphoS133-CREB (Cell Signaling Technology); anti-PDH-E1a (Abcam); anti-PDH-E1b (Abcam); anti-PDH-E2(Abcam), anti-PDH-E3 (Bio-Rad), anti-E3BP (Santa Cruz Bio-technology), anti-PINK1 (Novus Biologicals), anti-cytochrome C(Cell Signaling Technology), anti-Tom20 (BD Biosciences), anti-Tim23 (BD Biosciences), anti-b-tubulin (Wako), and anti-GAPDH (Wako). PDH complex purification beads were pur-chased from Abcam. MG132 and chloroquine were purchasedfrom Sigma.

Trypan blue exclusion assayCells were stained with 0.5% Trypan blue solution after nor-

moxia, hypoxia, or reoxygenation treatment. Stained andunstained cells were counted using hemocytometer.

Mitochondrial imaging with fluorescence microscopyCells were stained with Mitotracker (Thermo Fisher Scientific)

for 30 minutes after normoxic or hypoxic treatment for 72 hours,and fixed with 4% paraformaldehyde in PBS. Samples weremounted with Vectashield containing DAPI (Vector Laborato-ries), and imagedwith LSM510META confocalmicroscope (ZeissJapan).

FACS analysisFor cell-cycle analysis, cells were cultured under normoxic,

hypoxic (72 hours), or reoxygenated (6 hours) conditions, andfixed with cold 70% ethanol. After washing with PBS twice, cellswere treated with RNase at 37�C for 1 hour followed by stainingwith propidium iodide, and analyzed by FACSCalibur. For theapoptosis assay, cells were incubated with Annexin-V FITC (Naca-laiTesque) for 15minutes on ice after the treatment, and analyzedby FACSCalibur.

Western blottingCells were harvested on ice and lysed in lysis buffer [50mmol/L

Tris-HCl (pH8.0), 150mmol/LNaCl, 1mmol/L EDTA, 1%Triton

X-100, 0.1mg/mLPMSF, and 2mg/mL leupeptin). Total cell lysateswere subjected to SDS-PAGE (50 mg of total lysate/lane), and thentransferred to nitrocellulose membranes (PALL).

qPCRTotal RNA was isolated from cultured cells using an RNeasy kit

(Qiagen). First-strand cDNA synthesis was performedwith 2 mg oftotal RNA using the PrimeScript II kit (Takara Bio). SynthesizedcDNA was used for PCR analysis using SsoFast EvaGreen Super-mix (Bio-Rad) on a CFX96 real-time PCR detecting system (Bio-Rad). Relative expression levels were calculated using the DCt

method and normalized against b-actin or CPH internal control.Primer sequences are listed in Supplementary Table S1.

siRNACells were transfected with negative control siRNA (#45-2001,

Invitrogen), HIF-1a siRNA (Invitrogen), HIF-2a siRNA (Invitro-gen), LONP siRNA (Sigma), CLPP siRNA (Sigma), or AFG3L2siRNA (Sigma) using Lipofectamine RNAiMAX transfectionreagent (Invitrogen). Theday after transfection, cellswere culturedin normoxic or hypoxic conditions for 24–48 hours, and thenprocessed to extract total RNA or protein.

Colony assayAssay plates (6-well plate) were prepared by pouring DMEM/

10% FBS containing 0.5% agar. Single-cell suspensions werehomogeneously mixed with DMEM/10% FBS containing 0.3%agar and layered over on the bottom agar. MCF7 cells andMB231cells were grown for 2 weeks and 4 weeks, respectively, to formcolonies. Colonies were counted in three different fields.

Mitochondrial purificationCells were suspended with mitochondrial isolation buffer

[10 mmol/L HEPES (pH 7.4), 200 mmol/L D-Mannitol,70 mmol/L Sucrose, 1 mmol/L EDTA] and disrupted with adounce homogenizer. After centrifugation of postnuclearsupernatant, isolated mitochondria and cytosolic fraction weresubjected to Western blot analysis.

Metabolome analysisMB231 cells were treated under normoxic or hypoxic (1% O2)

conditions for 24 hours and then subjected to metabolomeanalysis. The analysis was conducted using capillary electropho-resis time-of-flight mass spectrometry (CE-TOFMS) for cationanalysis and CE-tandem mass spectrometry (CE-MS/MS) foranion analysis as described previously (11, 12). Briefly,CE-TOFMS analysis was carried out using an Agilent CE capillaryelectrophoresis system equipped with an Agilent 6210 time-of-flight mass spectrometer (Agilent Technologies). The spectrom-eter scanned fromm/z 50 to 1,000 (11), and peaks were extractedusing the MasterHands automatic integration software (KeioUniversity; ref. 13) and MassHunter Quantitative AnalysisB.04.00 (Agilent Technologies). Principal component analysis(PCA) was performed using SampleStat.

Measurement of the oxygen consumption rateThe oxygen consumption rates (OCR) were measured using an

XF24 Extracellular Flux Analyzer (Seahorse Biosciences) placed inthe InvivO2hypoxic chamber (Ruskinn Technology).MB231 cellsor MCF7 cells were seeded in XF24 V7 Cell Culture microplateat a density of 4 � 104 cells/well. For hypoxic and reoxygenation

Downregulation of PDH-E1b Attenuates the Tumor Growth

www.aacrjournals.org Cancer Res; 78(7) April 1, 2018 1593




experiments, mediumwas first changed to the assaymedium, andthen theOCRwasmeasured for 12 hours [2-hour normoxia, then8-hour hypoxia (1% O2), and then 2-hour reoxygenation]. Forprolonged hypoxic experiments, cells were initially treated withhypoxia (1% O2) in growing medium for 42 hours. Then, themedium was changed to the assay medium pre-equilibratedto hypoxia (note that this process transiently exposes cells tonormoxic condition), and the OCR was measured for 6 hours(until 48-hour hypoxia). The timings of hypoxic treatment aredepicted in the figure. The assay medium comprised 25 mmol/Lglucose, 143 mmol/L NaCl, 0.8 mmol/L MgSO4, 1.8 mmol/LCaCl2, 5.4 mmol/L KCl, 0.9 mmol/L NaH2PO4 (pH 7.4) and 1�Glutamax Supplement (Thermo Fisher Scientific).

Lactate assayMCF7orMB231 cells were cultured under normoxic or hypoxic

conditions, and intracellular lactate levels or lactate levels in themedium were measured using a Lactate Assay Kit (Biovision).

Cell proliferation analysisGrowth of control, PDH-E1a, or PDH-E1b-KDMCF7 cells was

monitored in medium containing different concentrations ofglucose. Cells were cultured in DMEM [with two differentglucose concentrations, low (1.0 g/L) or high (4.5 g/L), Wako]supplemented with 10% FBS. Cells were cultured for up to96 hours, and cell numbers were counted every 24 hours usinga hemocytometer.

Glucose assayCells were grown in low glucose DMEM supplemented with

10% FBS. Culture media were collected, and glucose concentra-tions were measured by Glucose Assay Kit (Abcam).

PDH assayPDH activity was measured using the PDH enzyme activity

dipstick assay kit (Abcam). Briefly, active PDH complex wascaptured from the cell lysate using an anti-PDH antibody, andits activity was measured in a colorimetric assay (6).

Cancer genomic data analysisA cancer genomic datasets analysis tool, cBioPortal (14, 15),

was used to assess amplification of PDH genes from differenthuman cancer genomic datasets. Briefly, five genes encoding thePDH complex (PDHB, PDHA1, DLAT, DLD, and PDHX) weresearched for in the entire 167 cancer database available in the tool.Datasets showing amplification of PDH genes were sorted, andcases showing PDHB amplification plus amplification of any ofthe five PDHgeneswere counted, and expressed as a percentage oftotal case. The analysis was performed using the database datedOct 28, 2017.

Tumor formation assay in nude miceCells were suspended at 1 � 107 cells/mL in a 1:1 mixture of

medium and Matrigel (Corning), and 100 mL of cell suspensionwas injected into the mammary fat pad of a BALB/cAJcl-nu/numouse (4-week-old female, CLEA). Tumor size was measuredweekly, and mice were sacrificed 5–10 weeks after injection tocollect tumors. Protein and RNA were extracted from tumorsand used for subsequent analyses. All animal experiments wereperformed in accordance with a protocol approved by the TokyoMedical and Dental University Animal Care Committee.

Statistical analysisAll experiments were performed at least three times, and the

mean and standard deviation (SD) are shown for each experi-ment. Data were analyzed using Student t test, and P valuessmaller than 0.05 were considered significant.

ResultsHypoxia treatment decreases tumor-forming ability of cancercells

As hypoxia promotes malignant transformation of tumors, wetested whether hypoxic pretreatment is sufficient to enhancetumor growth. Breast tumor cell lines, MB231 and MCF7, wereexposed to normoxia or hypoxia for 72 hours, and then inocu-lated into nude mice to generate xenografts. Pretreated cellsshowed normal cellular responses to prolonged hypoxia withCREB phosphorylation and HIF-2a expression (Fig. 1A). Cellswere viable (Fig. 1B; Supplementary Fig. S1A and S1B), andexhibited a normal cell-cycle profile without any signs of celldeath in normoxia, hypoxia, and upon reoxygenation (Fig. 1C).The colony-forming ability of hypoxic cells in soft agar was lowerthan that of normoxic cells; this effect was maintained afterreoxygenation (Fig. 1D). Although both cell lines formed tumorsin recipient mice, pretreatment with hypoxia reduced the tumor-forming ability of both cell lines (Fig. 1E and F).

PDH-E1b expression is reduced in prolonged hypoxiaHypoxia is typically associated with alterations in cellular

metabolism. As pretreatment of breast cancer cells with hypoxia(72 hours) reduced their ability to form tumors, we searched for apossible cause by focusing on metabolic enzymes. Prolongedexposure to hypoxia for up to 72 hours resulted in a decrease inPDH-E1b protein expression in MCF7 andMB231 cells (Fig. 2A),as well as in breast cancer SKBR3 cells, HeLa cells, and mouseembryonic fibroblasts (Supplementary Fig. S2A–S2C). Theseresults indicate that this phenomenon is a general cellularresponse that is conserved across multiple cell types. It is mainlyPDH-E1b that was affected by hypoxia, and PDH-E1a, PDH-E3,and PDH-E3BP were also reduced to some extent, whereasPDH-E2 was not affected (Fig. 2A). Hypoxic treatment inducedmitochondrial fission and mitophagy in some of MCF7 cells;however, mitochondrial structure and content in both cell lineswere mostly intact after 72 hours of hypoxic treatment (Fig. 2Band C). These results suggest that reduced expression of PDH-E1bis an active process that takes place in the mitochondria. Impor-tantly, the reduction in the PDH-E1b expression was maintainedeven after reoxygenation for 24 hours, when both PDH-E1aphosphorylation and HIF-2a expression decreased (Fig. 2D).Accordingly, we observed a decrease in PDH enzymatic activitywhen cells were exposed to hypoxia for 48 hours, that wasmaintained after 24 hours of reoxygenation (Fig. 2E). Further-more, PDH activity remained inactive after reoxygenation for upto 48 hours when the cells were pretreated with hypoxia for 72hours (Fig. 2E). These results indicate that downregulation ofPDH-E1b plays a role inmaintaining glycolytic metabolism uponreoxygenation, and that the effect is sustained for longer periodswhen the cells are pretreated with hypoxia for longer times.

HIF-1 regulates PDH-E1b expression in hypoxiaNext, we investigated the mechanism underlying PDH-E1b

downregulation under prolonged hypoxia. qPCR analysis of

Yonashiro et al.

Cancer Res; 78(7) April 1, 2018 Cancer Research1594




hypoxic cultured cells revealed that PDH-E1b expression wasreduced, whereas PDH-E1a expression was largely unaffected, atthe mRNA level (Fig. 3A). Because HIF is a key regulator of thehypoxic response, we next investigated whether HIF is involved inthis process. Knockdown (KD) of HIF-1a increased PDH-E1bexpression in normoxia and hypoxia (Fig. 3B, lanes 1 and 2; 4 and5; Supplementary Fig. S3). Furthermore, knockdown of HIF-1ainduced the expression of PDH-E1b mRNA to some extent,whereas a typical HIF-1 target gene GLUT1 was efficiently down-regulated (Fig. 3C). Moreover, HIF-1a KD induced PDH activityin normoxic MCF7 cells when its activity is not inhibited byphosphorylation (Fig. 3D). A similar effect was observed in HeLaand MB231 cells (Supplementary Figs. S4A–S4D and S5A–S5D).However, increase of PDH-E1b mRNA by the depletion of HIF-1was observed only under hypoxic conditions. These results indi-cate that PDH-E1b is commonly regulated at the level of protein ina HIF-1–dependent manner both under normoxic and hypoxicconditions, whereas a regulation at the level ofmRNA is limited tohypoxic conditions. Knockdown of HIF-2a also moderatelyincreased PDH-E1b expression in hypoxic MCF7 and MB231cells, which is likely due to functional overlap with HIF-1a(Fig. 3B; Supplementary Fig. S5A). However, this effect was notseen inHeLa cells. Taken together, these results suggest that HIF-1

is the main factor responsible for downregulating PDH-E1b,resulting in a reduction in its enzymatic activity inMCF7, MB231,andHeLa cells.We also examined the induction of transcriptionalrepressors under hypoxic conditions. Transcriptional repressors,DEC1, DEC2, and Cited2 were induced markedly in MCF7 andHeLa cells, whereasDEC1 andCited2were induced inMB231 cells(Supplementary Fig. S6A–S6C). Thus, these factors could also beinvolved in downregulation of PDH-E1b at the mRNA levelduring prolonged hypoxia.

We next examined the molecular mechanism by whichPDH-E1b expression is reduced in hypoxic cells. AlthoughmRNAregulation was involved as demonstrated above, regulation at theprotein level was also indicated by the observation that PDH-E1bwas almost undetectable in PDH-E1a-KD cells, even though thePDH-E1b mRNA level was normal (Supplementary Fig. S7A–S7C). To investigate the cause of the depletion, we tested inhibi-tors of the proteasome and lysosome, the two major proteolyticpathways in the cell. Neither class of compound inhibited thedecrease in PDH-E1b level (Supplementary Fig. S8A and S8B).Because PDH proteins localize in the mitochondrial matrix, andLONP is induced in hypoxia and controls energy metabolism inmitochondria (16), we next focused on three mitochondrialmatrix proteases: LONP, CLPP, and i-AAA (AFG3L2). Depletion

A

Norm Hypo

B C

pCREB

CREB

β-Actin

37

37

50

Norm

Hy

po

100 HIF-2 α

D E

MCF7

MB231

Norm Hypo Reoxy

1.8±0.2

Trypan blue–positive cells (%)

3.3±1.0 1.9±1.0

1.6±0.6 2.6±1.0 2.3±0.6

MCF7

MB231

Norm Hypo Reoxy

0.7±0.2

DNA Fragmentation (%)

1.5±1.3 1.4±0.9

0.8±0.2 1.3±0.6 1.2±0.4

MCF7

MB231

Norm Hypo Reoxy

80

40

0Norm Hypo Reoxy

Num

ber o

f col

onie

s MCF7MB231

Norm Hypo Reoxy

40

20

0

******

**

G1: 69%S: 21%

G2–M:10%

G1

SG2–M

G0–G1: 52%S: 39%

G2–M: 9%

G0–G1: 56%S: 28%

G2–M:16%

250

0

150

50

FL2-A200 400400 600 800

250

0

150

50

FL2-A200 400 600 800

250

0

150

50

FL2-A200 400 600 800

G1: 74%S: 14%

G2–M: 12%

G1

SG2–M

G0–G1: 55%S: 27%

G2–M: 18%

G0–G1: 39%S: 28%

G2–M: 33%

250

0

150

50

FL2-A200 400 600 800

250

0

150

50

FL2-A200 400 600 800

250

0

150

50

FL2-A200 400 600 800

NormHypo

Week 3 Week 4 Week 5

*

*

*

0

120

180

60

Cou

nts

Cou

nts

F


**

**

*

0

25

50NormHypo

Tum

or v

olum

e (m

m3 )

Tum

or v

olum

e (m

m3 )

Figure 1.

Pretreatment with hypoxia decreases the ability of cancer cells to form tumors. A, Expression of HIF-2a and pCREB under prolonged hypoxia. MB231 cellswere pretreated with normoxia or hypoxia for 72 hours, and then proteins were extracted and subjected to Western blotting. Blotting was performed withantibodies against phospho-CREB, CREB, HIF-2a, and b-actin. B, Viability of cells exposed to prolonged hypoxia. MCF7 cells or MB231 cells were treated withnormoxia or hypoxia for 72 hours, or reoxygenated for 6 hours after exposure to hypoxic conditions. Treated cells were subjected to a Trypan blue exclusionassay or stained with DAPI. The percentages of stained cells and fragmented nuclei are shown. � indicates the SD. C, Cell-cycle distribution of MCF7 andMB231 cells treated with prolonged hypoxia. Normoxic, hypoxic (72 hours), or reoxygenated (6 hours) cells were subjected to cell-cycle analysis by flow cytometry.D, Pretreated cells were collected and subjected to a colony formation assay in soft agar. The number of colonies formed was counted and is shown on the graph.E and F, MB231 cells (E) or MCF7 cells (F) pretreated with normoxia (norm) or hypoxia (hypo) for 72 hours were inoculated into the mammary fat pad offemale nude mice (n ¼ 8). Images of tumors and tumor volumes (mm3) are shown on the graph. � , P < 0.05; �� , P < 0.02.






of LONP, CLPP, or AFG3L2 restored the PDH-E1b expression,which was reduced in hypoxic-treated cells, suggesting that theseproteases are involved in degrading PDH-E1b in MCF7, MB231,andHeLa cells (Fig. 3E, lanes 5–8; Supplementary Figs. S4D, S5D,and S9).

Depletion of PDH-E1a/E1b diminishes PDH activity andincreases lactate production

Because PDH-E1b expression decreases while PDH activity isrepressed under prolonged hypoxia, we mimicked this situationby depleting PDH-E1b expression with shRNA. For this purpose,MCF7 and MB231 cells were infected with retrovirus carryingshRNA targeting control, PDH-E1a or PDH-E1b (#1 and #2).Knockdown of both genes was efficient (Fig. 4A). Importantly,knockdown of PDH-E1b decreased expression of PDH-E1a, andvice versa; however, downregulation at the mRNA level wasspecific for the shRNA target, and did not affect the other subunit(Supplementary Fig. S7A), indicating that this coregulatory effectoccurred at the protein level. The effect was not inhibited by

proteasome or lysosome inhibitors (Supplementary Fig. S7Band S7C). These results indicate that a coregulatory machinery,distinct from major degradative pathways, synchronizes theexpression of PDH-E1a and PDH-E1b in cells. Notably, expres-sion of other PDH subunits (PDH-E2, PDH-E3, and PDH-E3BP)was not altered in PDH-E1a- or PDH-E1b-KD cells (Fig. 4A). PDHactivity was almost completely inhibited in both types of shRNA-expressing cells (Fig. 4B).

PDH catalyzes the conversion of pyruvate to acetyl-CoA, thusconnecting glycolysis to the TCA cycle. Inhibition of PDH activityshould suppress this conversion, resulting in accumulation ofpyruvate, leading in turn to the conversion of pyruvate into lactateby lactate dehydrogenase. The culture medium of PDH-E1a-KDand PDH-E1b-KD cells rapidly became more acidic (yellow inPhenol red–containing medium) than that of control cells platedat a similar density, reflecting elevated production of lactate in theknockdown cells (Fig. 4C). Measurement of intracellular lactatelevels in these cells confirmed that PDH-KD cells contained higherlactate concentrations (Fig. 4C).

A B C

75 PDH-E2

37 PDH-E1α

PDH-E1β37

50β-Actin

0 24 48 72 48+

re24

0 24 48 72 48+

re24MB231MCF7

HIF-2α

HIF-1α

100

100

PDH-E2

PDH-E1α

PDH-E1β

β-Actin

0 24 48 72 0 24 48 72MCF7 MB231

7537

37

50

Norm

Hypo

72

h

Hypo 48 h Hypo 72 h

PDH Assay

0

5

10

0

5

10

Rel

ativ

e P

DH

act

ivity

** ** **** *

h

h

p-PDH-E1α

p-PDH-E1α

D E

Norm Hypo

MCF7

MB231

β-tubulin

Tom20

Tim23

cytC

MCF7MB231N H N H

PINK175

25

25

15

37

50

PDH-E1β

PDH-E350

PDH-E3BP50

Hypo

Hypo

Figure 2.

Expression of PDH-E1b decreases under prolonged hypoxia. A, MCF7 or MB231 cells were treated with hypoxia for the indicated time and then subjectedto Western blotting. Blotting was performed with antibodies against PDH-E1b, PDH-E1a, PDH-E2, PDH-E3, PDH-E3BP, and b-actin. p-PDH-E1a, phosphorylatedPDH-E1a. B, MCF7 or MB231 cells were treated with normoxia or hypoxia for 72 hours and then stained with MitoTracker. Representative images of stainedmitochondria under each condition are shown. Scale bar, 20 mm.C,MCF7 orMB231 cells treatedwith normoxic or hypoxic conditions for 72 hourswere harvested andsubjected to Western blotting. Blotting was performed with antibodies against Tom20 (mitochondrial outer membrane), cytochrome C (intermembrane space),Tim23 (innermembrane), PDH-E1b (matrix), PINK1, and b-tubulin. N, normoxia; H, hypoxia.D,MCF7 or MB231 cells were treatedwith hypoxia for the indicated times,followed by 24 hours of reoxygenation (re). Cells were then subjected to Western blotting with antibodies specific for PDH-E1b, PDH-E1a, PDH-E2, HIF-1a, HIF-2a,and b-actin. p-PDH-E1a, phosphorylated PDH-E1a. E, MCF7 cells were treated with hypoxia, followed by reoxygenation (re) for the indicated times and thenharvested. PDH activitywas thenmeasured. Representative image of dipsticks used for the PDH assay (top) and relative PDH activity (bottom) are shown. � , P <0.05and �� , P < 0.02.

Yonashiro et al.





Inhibition of PDH expression induces glycolytic metabolismHigh lactate production is characteristic of cancer cells exhibit-

ing the Warburg effect. Accordingly, we tested the metabolic stateof PDH-E1a-KD and PDH-E1b-KD MB231 cells by measuringtheir OCR using a flux analyzer and lactate concentration in themedium using a Lactate Assay Kit. The ratio of the OCR tolactate production rate under normoxic conditions was around1300pmol/min/mmol/L in control cells, but only one-half of thatvalue in PDH-KD cells (Fig. 4D), suggesting that PDH-KD cells arein a metabolic state resembling the Warburg effect (Fig. 4D;Supplementary Fig. S10A and S10B). Exposure to hypoxia for 8hours reduced the OCR and increased the lactate production ratein both control and PDH-KD cells, resulting in comparablemetabolic states (Fig. 4D). Reoxygenation for 2 hours did notchange the OCR/lactate production ratio (Fig. 4D). Furthermore,prolonged exposure to hypoxia for 48 hours also resulted inOCR/

lactate production ratios that were comparable among these threecell types, indicating that they all utilize a similar glycolyticmetabolism under prolonged hypoxic conditions. Similarchanges in the OCR/lactate production ratio were observed inanother breast cancer cell line, MCF7, when examined under thesame hypoxic conditions (Supplementary Figs. S10B and S11).Reoxygenation for 2 hours moderately recovered the OCR inMCF7 cells, but not in MB231 cells, suggesting that some addi-tional time is required to fully recover the OCR once cells areexposed to hypoxia extensively (Supplementary Fig. S10A andS10B). To further characterize the metabolic state of PDH-KDcells, we performedmetabolome analysis to compare control andPDH-E1b-KD MB231 cells cultured under normoxic or hypoxicconditions for 24 hours. As PDH catalyzes conversion of pyruvateto acetyl-CoA, PDH-E1b-KD cells accumulated pyruvate, con-tained less acetyl-CoA, and more lactate than control cells

A

PDH-E1α PDH-E1β

B

DC

HypoNorm

PDH-E1β

siRNA

Norm Hypo

E

MCF7 MB231R

elat

ive

PD

H a

ctiv

ity

**

0

1

2

0 24 48 72 0 24 48 720

1

2

0 24 48 72 0 24 48 72MCF7 MB231

****

****

h

Rel

ativ

e ex

pres

sion

**

Rel

ativ

e ex

pres

sion

siRNA

0

1

2

0

1

2

siRNA

Rel

ativ

e ex

pres

sion

GLUT1** ** **

β-Actin

PDH-E1β

PDH-E1α

Cont

rol

LONP

CLPP

AFG

3L2

Cont

rol

LONP

CLPP

AFG

3L2

siRNA

Hypo Norm

37

37

50

10 9.5 9.7 12 7.0 9.9 13 11

1 2 3 4 5 6 7 8

PDH-E1β37

HIF-2α

PDH-E1α

HIF-1α

100

37

100

β-Actin50

Cont

rol

HIF-

1αHI

F-2α

Cont

rol

HIF-

1αHI

F-2αsiRNA

1 2 3 4 5 6

Hypo Norm

#

10 12.9 8.7 8.1 14.6 12.8

0

5

10

Con Con

Con Con

Con Con

Figure 3.

Downregulation of PDH-E1b occurs in a HIF-1–dependent manner. A, MCF7 or MB231 cells were treated with hypoxia for the indicated times, and expressionof PDH-E1a and PDH -E1b was monitored by qPCR (n ¼ 3). B, MCF7 cells were transfected with control, HIF-1a, or HIF-2a-targeting siRNA, and treated withnormoxia or hypoxia for 48 hours. Harvested cells were subjected to Western blotting. Blotting was performed with antibodies against PDH-E1b, PDH-E1a, HIF-1a,HIF-2a, and b-actin. #, a nonspecific band detected by the antibody. Bands in the PDH-E1b blot were quantified, and their relative intensities are represented by thenumbers below. C and D, MCF7 cells transfected with control, HIF-1a, or HIF-2a siRNA were treated with normoxia or hypoxia for 48 hours. Expression ofPDH-E1b and GLUT1 was monitored by qPCR (C), and PDH activity was measured by PDH assay (D). E, MCF7 cells were transfected with siRNA targeting threemitochondria matrix proteases (LONP, CLPP, and AFG3L2) and treated with normoxia or hypoxia for 48 hours. Harvested cells were subjected to Westernblotting with antibodies against PDH-E1b, PDH-E1a, and b-actin. Bands in the PDH-E1b blot were quantified, and their relative intensities are represented by thenumbers below. � , P < 0.05; �� , P < 0.02.






(Fig. 5A). Similar changes were observed in hypoxic treatedcells, although the levels of acetyl-CoA and lactate becamehigher in both control and PDH-KD cells. Higher lactate pro-duction is an indication that these cells use anaerobic metab-olism in response to hypoxia. Higher acetyl-CoA level mayindicate an active conversion of acetate into acetyl-CoA in thesecancer cells (17). Furthermore, PDH-E1b-KD cells showedelevated glycolysis under normoxic conditions, with higherlevels of glycolytic metabolites compared with control cellssuch as fructose-1,6-bisphosphate (F1,6P), 3-phosphoglycerate(3-PG), 2-phosphoglycerate (2-PG), and phosphoenolpyruvate(PEP), another indication of Warburg effect-like metabolism(Fig. 5B). The difference between control and the KD cells wasno longer detectable under hypoxic conditions since glycolyticmetabolism is induced both in control and KD cells. Further-more, the levels of citric acid and cis-aconitic acid generated bythe TCA cycle were reduced markedly in PDH-E1b-KD cells,although parts of the cycle beyond 2-oxoglutarate (2-OG)remained mostly normal under normoxic conditions(Fig. 5C). Under hypoxic conditions, levels of metabolitesformed beyond 2-OG in the TCA cycle were significantly higher

in KD cells than in control cells, which might implicate acti-vation of a glutamine synthesis pathway that would compen-sate for metabolites beyond 2-OG. All of the 116 metabolitesanalyzed are shown in Supplementary Table S2. Principalcomponent (PC) analysis was also performed on this data setto visualize differences between control and PDH-E1b-KD cells(Fig. 5D). The first PC distinguished between normoxia- andhypoxia-treated samples, whereas the second PC clearly sepa-rated PDH-E1b-KD cells from control cells. Glutathione (GSH),pyruvate, malic acid, and fumaric acid were the top fourcontributors to the second PC.

The ability to form tumors is reduced in PDH-KD cellsMany cancer cells exhibit glycolytic metabolism, which is often

linked to malignant transformation. Since PDH-KD cells alsoexhibited glycolytic metabolism, we examined their tumor-forming ability in immunodeficient nude mice. Two clones fromPDH-E1a shRNA and PDH-E1b shRNA #1, and a clone fromcontrol shRNA and PDH-E1b shRNA #2 were examined. BothPDH-E1a-KD and PDH-E1b-KD MCF7 cells formed significantlysmaller tumors than the control cells (Fig. 6A and B). Reduced

A B

Rel

ativ

e P

DH

act

ivity

Con E1α E1βshRNA

shRNANorm Hypo

C D

****

****

**

**

0

5

10

Con E1α E1β

**

shRNA

0

0.5

1

Con E1α E1β Con E1α E1βLact

ate

conc

entra

tion

(mm

ol/L

)

PDH-E1αPDH-E1β

PDH-E3BP

PDH-E2

PDH-E3

75

5050

37

50

37

Cont

rol

E1α

E1β

E1βshRNA

β-Actin

(i) Hypoxia and reoxygenationNorm Hypo Reoxy

2 h 8 h 2 h

(ii) Prolonged hypoxiaHypo

48 hO

CR

/Lac

tate

pro

duct

ion

(10

pmol

/min

/mm

ol/L

)Aerobic

Glycolytic

42 h

OC

R/L

acta

te p

rodu

ctio

n (p

mol

/min

/mm

ol/L

)

#1 #2

**

0

50

100

150

Hypo 48 h

ConE1αE1β

**

0

50

100

150

Norm 2 h Hypo 8 h Reoxy 2 h

ConE1αE1β

Figure 4.

PDH-E1b knockdown promotes lactate production and glycolytic metabolism. A, MB231 cells stably expressing shRNA against PDH-E1a or PDH-E1b (E1a-KD,E1b-KD) were established. Inhibition of PDH-E1a or PDH-E1b was examined by Western blotting with antibodies against PDH-E1b and PDH-E1a. PDH-E2, PDH-E3,PDH-E3BP, and b-actin blots are also shown. B, PDH activities in E1a-KD and E1b-KD MB231 cells were examined by PDH assay. C, Lactate production in E1a-KDand E1b-KD MB231 cells. Images of culture medium from control, E1a-KD, and E1b-KD cells are shown (top). Intracellular lactate concentrations were measuredin a lactate assay (n ¼ 3). D, The metabolic status of control, E1a-KD, and E1b-KD MB231 cells was analyzed using a flux analyzer. The oxygen consumptionrate (OCR) and lactate production rate were measured in two experimental settings (i) hypoxia and reoxygenation and (ii) prolonged hypoxia as in the scheme.The ratio of the OCR to the lactate production rate was plotted (n ¼ 6). The OCR and the lactate production (lactate levels in the medium) were measuredat the time points indicated by arrows. Arrowhead, the point of medium change to the assay medium in ii. � , P < 0.05; �� , P < 0.02.

Yonashiro et al.





expression of PDH-E1a or PDH-E1b in tumors was confirmed byqPCR andWestern blotting (Supplementary Fig. S12A and S12B).Similar reduction of tumor growthwas observed in another breastcancer cell line, MB231 (Fig. 6C). The gene expression profile wasaltered in tumors formed by PDH-E1a or PDH-E1b-KD cells,especially in regard to hypoxic target genes such as VEGF, GLUT1,and MMP1 (Fig. 6D). Because these cells exhibit reduced expres-sion of GLUT1, we assessed their sensitivity to glucose concen-tration by measuring their growth rates in two different culturemedia, high- and low-glucose. Although growth of control,PDH-E1a-KD, and PDH-E1b-KD cells did not differ significantlywhen they were cultured in high-glucose medium, growth wassignificantly slower in PDH-E1a-KD and PDH-E1b-KD cells inlow-glucose medium (Fig. 6E). Glucose was absorbed morerapidly from the culture medium by PDH-KD cells, possiblybecause these cells depend mostly on glycolysis for energy pro-duction; these cells require more glucose to produce the sameamount of ATP as control cells during growth (Fig. 6F). However,expression of GLUT1 and 4 was not higher in the KD cells than incontrol cells, which might imply that the GLUTs expressed in

PDH-KD cells are sufficient to uptake glucose at least in thisexperimental setting (Fig. 6G). Altogether, these results indicatethat PDH-E1a-KD and PDH-E1b-KD cells require a larger supplyof glucose to grow as rapidly as control cells.

Elevated tumor formation ability in PDH-E1b–expressing cellsNext, we established a MCF7-derived cell line that ectopically

expressed PDH-E1b (MCF7þE1b cells). These cells expressedhigher levels of PDH-E1b expression under both normoxiaand hypoxia (Fig. 7A), possibly because the degradative machin-ery involving mitochondria proteases is restricted by elevatedexpression of PDH-E1b under hypoxic conditions. Furthermore,expression of PDH-E1a was higher in these cells (Fig. 7A), whichis due in part to the induction of PDH-E1a at the mRNA level(Supplementary Fig. S13). This might be caused by the cellsbalancing the amount of PDH-E1a and PDH-E1b subunits; theexpression of one is associated with that of the other (Fig. 4A).Ectopically expressed PDH-E1b is processed properly, and local-ized in mitochondria to form the PDH complex (Fig. 7B; Sup-plementary Fig. S14). MCF7þE1b cells had higher PDH activity

A B

C

D

Glycolysis

**

*

** **

TCA cycle

-9

-6

-3

0

3

6

-15 -10 -5 0 5 10 15Principal component 1 (62.9%)

Prin

cipa

l com

pone

nt 2

(18.

7%)

Con (norm)

E1β-KD (norm) E1β-KD (hypo)

Con (hypo)

***

PyruvateLactate

Acetyl-CoA

Con

cent

ratio

n (p

mol

/106 c

ells

)

ConKDHypo conHypo KD

0

100

200

300

400

G6P0

20

40

60

80

F6P0

1,000

2,000

3,000

F1,6P

0

15

30

2-PG0

100

200

300

3-PG0.0

30.0

60.0

90.0

120.0

PEP

0.0

1.0

2.0

3.0

Acetyl-CoA0

10,000

20,000

30,000

40,000

50,000

Lactate

0

1,000

2,000

3,000

Malic acid0

100

200

300

Succinic acid

0

1,000

2,000

3,000

4,000

Pyruvate

0

500

1,000

1,500

2,000

0

300

600

Fumaric acid

0

500

1,000

1,500

2,000

Citric acid0

15

30

Cis-aconitic acidC

once

ntra

tion

(pm

ol/1

06 cel

ls)

Con

cent

ratio

n (p

mol

/106 c

ells

)

** *

**

**

**

ND ND

**

**

*** **

Figure 5.

Enhanced glycolysis and reduced TCA cycle activation in PDH-E1b-KD cells. A–C, Metabolome analysis comparing control and E1b-KD MB231 cells. Cells weretreated with normoxia or hypoxia (1% O2) for 24 hours, and the concentrations of metabolites representative of pyruvate metabolism (A), glycolysis (B),and the TCA cycle (C) weremeasured (n¼ 3). �, P <0.05 and �� , P <0.02.D, Results of principal component analysis, shown as a plot of the first principal component(accounting for 62.9% of total variance) against the second principal component (accounting for 18.7% of total variance).






than the parental line under normoxia, but the activity sharplydropped when they were cultured under hypoxia (Fig. 7C).However, MCF7þE1b cells sustained high PDH-E1b levels evenunder hypoxia, and reoxygenation for 24 hours was sufficient torestore PDH activity close to the level in normoxia, in contrast tothe control cells, which still exhibited reduced activity (Fig. 7C).When these cells were transplanted into nude mice, MCF7þE1bcells formed significantly larger tumors than control cells (Fig. 7Dand E). Furthermore, MCF7þE1b cells pretreated with hypoxiainitially formed smaller tumors than their normoxic counterparts,but growth accelerated after 5 weeks, and by 8 weeks the tumorsderived from hypoxia-treated cells were as large as those derivedfrom normoxia-treated cells (Fig. 7E). This was in clear contrast toparental MCF7 cells, which exhibited persistently slower tumorgrowth when pretreated with hypoxia, even though the tumor-bearing animals were maintained for longer period (Fig. 1F;Supplementary Fig. S15A and S15B). Importantly, the hypoxictarget genes, GLUT1 and GLUT4, which are barely expressed in

parental cells, were induced in tumors derived from MCF7þE1bcells (Fig. 7F), whereas GLUT1 was expressed at low levels intumors derived from PDH-E1b-KD cells (Fig. 6D).

Expression of PDH-E1b in human cancersFinally, we looked for changes in the expression level of

PDH in human cancer patient samples using data fromcBioPortal (Fig. 7G; Supplementary Table S3; refs. 14, 15). Geneamplification of PDHs was detected in a variety of cancer types,including breast, esophagus-stomach, and lung. Prostate (adeno-carcinoma) showed a high rate of amplification, up to 50% (allfive PDH subunits combined), and bladder, melanoma, ovarian,and prostate (neuroendocrine) showed higher ratios of PDH-E1bamplification.

DiscussionOne of the major pathways so far reported to regulate the

PDH activity is mediated by phosphorylation of the PDH-E1a

A B C

Control

PDH-E1α

PDH-E1β

shRNA

0

0.4

0.8

1.2

0

0.5

1

Con E1α E1β Con E1α E1β0

0.5

1

1.5

Con E1α E1β

VEGF GLUT1 MMP1

0

50

100

150

0 24 48 72 96

Con hiPDH-E1α hiPDH-E1β hiCon lowPDH-E1α lowPDH-E1β low

** **** ** ** **

Rel

ativ

e ex

pres

sion

Cel

l num

ber (

×104

)

D

****

shRNA

shRNA

48 h

Glu

cose

(mm

ol/L

)

E F

**

*

****

*

Tum

or v

olum

e (m

m3 )

Tum

or v

olum

e (m

m3 )

shRNAMCF7 MB231

G

24 h

****

shRNA****

0

2

4

ConE1αE1β

0

300

600


ConE1αE1β

h

GLUT1 GLUT4

Rel

ativ

e ex

pres

sion

60

40

20

0Week 3 Week 4 Week 5 Week 6 Week 7 Week 8

ConE1α-1 E1α-2 E1β#1-1 E1β#1-2 E1β#2

**********

**********

**********

**********

**********

**********

3

0

1

2

0

0 h 48 h 0 h 48 h

ControlE1αE1β

1

2

**

*

**

shRNA

Figure 6.

PDH-KD cells exhibit reduced tumor formation ability. A, MCF7 cells stably expressing shRNA against PDH-E1a or PDH-E1b were inoculated into themammary fat pad of female nude mice (n ¼ 7). Images of tumors are shown. B, The size of each tumor was monitored weekly, and the volume of tumors (mm3)formed by two clones for shRNA (E1a, E1b#1) and a clone for shRNA (con, E1b#2) were plotted. C, Volume (mm3) of control, E1a-KD, and E1b-KD MB231tumors (n ¼ 6). D, Tumors from MB231 cells were collected at week 5 and subjected to qPCR analysis. Expression levels of VEGF, GLUT1, and MMP1 are shown. E,Growth rate of control, E1a-KD, and E1b-KD MCF7 cells. Cells were cultured in high-glucose (4.5 g/L; hi) or low-glucose (1 g/L; low) medium for up to 96 hours,and cell numbers were counted at the indicated time points (n ¼ 3). F, Glucose consumption by control, E1a-KD, and E1b-KD MCF7 cells. Cells were culturedfor 48 hours in low-glucosemedium and the amount of glucose remaining in themediumwasmeasured.G, Expression of glucose transporters in cells. Cells at 0- and48-hour time-points were collected in F and subjected to qPCR analysis. Expression of GLUT1 and GLUT4 is shown. � , P < 0.05; �� , P < 0.02.

Yonashiro et al.





subunit. Here, we demonstrated that PDH-E1b is downregu-lated under hypoxic conditions and constitutes another layer ofPDH inhibition in cells. Phosphorylation is responsible forquick and timely inhibition so that PDH activity can be alteredin a relatively short time, for example, during acute hypoxia. Incontrast, during long periods of hypoxia, it may be importantto assure inhibition of PDH by downregulating the protein.Thus, cells might utilize two independent systems to securelyregulate the activity of this central metabolic enzyme and adaptto both acute and prolonged phases of hypoxia. Importantly,the lower PDH activity in MCF7 cells was maintained even afterreoxygenation for up to 48 hours when they were pretreated

with hypoxia for 72 hours (Fig. 2E). Therefore, this novelmechanism of PDH regulation could serve as a basis forestablishing aerobic glycolysis in cancer cells.

Depletion of HIF-1a increased PDH-E1b expression mainly atthe protein level in three different cell lines examined, indicatingthat the HIF-1 pathway plays a negative role on the PDH-E1bexpression. As HIF-1 induces LONP, induction of mitochondrialmatrix proteases might be one of the negative regulatorymechan-isms (Fig. 3E; Supplementary Figs. S4D and S5D; ref. 16). Of note,as knockdown of mitochondrial matrix proteases only moder-ately affected PDH-E1b under normoxic conditions, we cannotexclude a possibility that there might be some additional

A C

DNorm

Norm+E1β

Hypo+E1β

E

MCF7+E1β

PDH-E1β

GAPDH

37

37

N Hy reox

yN Hy re

oxy

MCF7

Norm Hypo ReoxyMCF7 MCF7+E1β

Norm Hypo Reoxy

GLUT1 GLUT4

****

0

10

20

30

MCF7 MCF7+E1β

NormHypoReoxy

**

Rel

ativ

e P

DH

act

ivity

**

0

2.5

5

0

5

10

**

Rel

ativ

e in

duct

ion

PDH-E1β

Flag-PDH-E1β

PDH-E1α

β-Actin

50

3750

Norm

Hypo

Norm

Hypo

MCF7 +E1β

37

50

37

MCF7

Flag-PDH-E1β

G

0

2.5

5**

PDH-E1β

Tim2325

37

50

37

50

C M C M

B

F

Flag-PDH-E1β

PDH-E1β

PDH-E1α

β-Actin

MCF7 +E1βMCF7

Rat

io o

f am

plifi

catio

n (%

)

0

25

50

ProstateAd.

0

5

10

15 PDH total

PDH-E1β

Con +E1β Con +E1β Con +E1β

3,000

2,000

1,000

0

Tum

or v

olum

e (m

m3 )

Control+E1β norm+E1β hypo

Week 3 Week 4 Week 5 Week 6 Week 7 Week 8

**

****

**

**

Esoph

agea

l

Esoph

.-stom

ach

Bladde

rLu

ng

Melano

ma

Stomac

h

Ovaria

nLiv

er

Sarcom

a

Prostat

e NE

Breast

Glioblas

toma

Testis

Uterine

Thymus

Pheoc

hromoc

yt.

Figure 7.

MCF7 cells ectopically expressing PDH-E1b exhibit elevated tumor formation ability. A, MCF7 cells stably expressing FLAG-tagged PDH-E1b (MCF7þE1b)were treated with normoxia (norm) or hypoxia (hypo) for 48 hours, harvested, and subjected toWestern blotting with antibodies against PDH-E1b, FLAG, PDH-E1a,or b-actin. B,Amitochondrial fractionwas prepared fromMCF7 andMCF7þE1b cells. Samples were subjected toWestern blotting with antibodies specific for FLAG,PDH-E1b, PDH-E1a, Tim23, or b-actin. C, cytoplasmic; M, mitochondria. C, Control or MCF7þE1b cells were exposed to normoxic, hypoxic (48 hours), orhypoxic (48 hours) plus reoxygenation (24 hours) conditions. Cell lysates were prepared and subjected to a PDH assay and Western blot analysis. Representativeimage of the dipsticks used in the assay and relative PDH activity are shown. Expression of PDH-E1b was detected using anti-PDH-E1b and anti-GAPDH antibodies(bottom right). D and E, Control (norm) and MCF7þE1b cells treated with normoxia (normþE1b) or hypoxia (hypoþE1b) for 72 hours were inoculated into themammary fat pad of female nude mice (n ¼ 8). Images of tumors are shown (D). Tumor size (mm3) was monitored weekly and plotted (E). F, Tumors werecollected at week 5 and subjected to qPCR analysis. Expression of GLUT1, GLUT4, and PDH-E1b was detected. G, Expression of the five main subunits of thePDH enzyme (PDH-E1a, PDH-E1b, PDH-E2, PDH-E3, and PDH-E3BP combined; gray bar) and PDH-E1b alone (black bar) in samples from human cancer patients.Cases showing amplification of total PDH or PDH-E1b, as a percentage of the total number of specimens from each organ. Ad, adenocarcinoma; NE, neuroendocrinetumor. � , P < 0.05; �� , P < 0.02.






machineries to regulate PDH-E1b at theprotein level in normoxia.As PDH-E1b promoter harbors a putative HRE, we used a ChIPassay to examine the possibility that HIF-1 binds to the promoterregion of PDH-E1b. Whereas interaction between HIF1 and theGAPDH HRE was clear, there was no interaction with the PDH-E1b promoter (Supplementary Fig. S16).

PDH-E1a and PDH-E1b are tightly associated in cells, and playa role in pyruvate dehydrogenation. Our data indicate that, whenone subunit is knocked down, expression of the other alsodecreases (Fig. 4A). We also showed that PDH-E1a becomestranscriptionally induced when PDH-E1b is ectopically expressed(Supplementary Fig. S13). Thus, cellular machinery mightcoregulate these two subunits at the mRNA and protein levels,which would strictly and precisely control enzymatic activity.Importantly, PDH-E3 and PDH-E3BP are also downregulatedby prolonged hypoxia, but are not altered in PDH-E1a andPDH-E1b-KD cells (Figs. 2A and 4A). These results indicatethat PDH-E3 and PDH-E3BP are not simply downregulated whenPDH-E1a and PDH-E1b are depleted, but there would rather bean active mechanism to downregulate these proteins in responseto hypoxia.

Although the molecular mechanism underlying the Warburgeffect is not completely understood, reports show that dysregula-tion of molecules functioning in glycolysis or the TCA cycle isinvolved (18). For example, EGFR signaling, which is oftenactivated in cancer cells, induces phosphorylation of pyruvatekinase PKM2. Phosphorylated PKM2 then activates transcriptionfactors in the nucleus to induce GLUT1 and LDH, which in turnplay critical roles in establishing the Warburg effect (19). Inhibi-tion of AMPK in cells also induces glycolytic metabolism (20).Patients carrying an AMPK mutation are often associated with amalignant form of cancer.

In this study, we showed that knockdown of a single PDH-E1subunit is sufficient to cause Warburg effect–type metabolismin breast cancer cell lines (Fig. 4D; Supplementary Fig. S11).However, it did not promote tumor growth of these cells; rather,PDH-depleted cells formed smaller tumors than control cells(Fig. 6A–C). This phenomenon could be in part due to reducedexpression of GLUT1, which would limit glucose uptake byPDH-KD tumors (Fig. 6D). In contrast, GLUT1 and GLUT4 werehighly induced in tumors derived fromcells ectopically expressingPDH-E1b (Fig. 7F). Upregulation of GLUTs might be critical forcancer cells that reside in a microenvironment with a limitedglucose supply.

The current understanding is that the Warburg effect creates afavorable metabolic state for tumor progression (21). However,our results indicate that sustained Warburg effect-type metabo-lism rather prevents cancer cells from growing. Dysregulation ofmitochondria metabolism and lack of oxidative phosphorylationreduces tumorigenicity in K-ras–driven lung cancer and breastcancer cells, respectively (22, 23). Moreover, whether a tumorrelies primarily on glycolysis or oxidative phosphorylation forenergy production is not necessarily related to tumormalignancy;rather it depends on the environmentwithin the organof origin orthe tissue to which the cancer cells have metastasized (24). Thus,flexible switching between glycolysis and oxidative phosphory-lation, depending on the circumstances that the tumor faces atany given time, might be required for progression. PDH mightcoordinate these two phases of tumor metabolism, therebypromoting tumor growth. However, it may be possible that thePDH-KD cells we generated cannot induce the same changes in

metabolites and/or gene expression that exist in tumor cellsexhibiting the Warburg effect, even though they, under normoxicconditions, consume less oxygen and produce high levels oflactate, both of which are hallmark criteria of the Warburg effect.

Importantly, even after establishing stable glycolyticmetabolism in prolonged hypoxic culture by downregulatingPDH-E1b, tumor growth can readily recover if PDH-E1b isconstitutively expressed at a higher level (Fig. 7E). As thetumor microenvironment is constantly changing from a nor-moxic to a hypoxic state and back again (25), this change willpotentially lead to repetitive up- and downregulation of PDH-E1b, which could alter the metabolic mode accordingly underpathophysiologic conditions. This could, in turn, promotetumor growth by providing tumor cells with the opportunityto use both glycolysis and oxidative phosphorylation to meettheir metabolic needs (26).

Patients harboring mutations in PDH exhibit severe lacticacidosis, which results in neurological disorders such as Leighsyndrome, which often causes lethality at a young age (27).However, the relationship between PDHmutation and the occur-rence of cancer is not well understood. A search of the cBioPortaldatabase revealed that PDH genes are amplified in a wide range ofcancer types (Fig. 7G). Thus, amplificationofPDH could representa tactical strategy by which cancer cells flexibly switch betweenaerobic and anaerobic metabolism, regardless of the oxygenenvironment to maintain a favorable metabolic state.

Disclosure of Potential Conflicts of InterestN. Takeda reports receiving a commercial research grant fromDaiichi Sankyo

Company, Limited and Bayer Yakuhin, Ltd. No potential conflicts of interestwere disclosed by the other authors.

Authors' ContributionsConception and design:K. NakayamaAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.):R. Yonashiro, K. Eguchi, M. Wake, N. Takeda,K. NakayamaAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis):R. Yonashiro, K. Eguchi, M. Wake, N. Takeda,K. NakayamaWriting, review, and/or revision of the manuscript:R. Yonashiro, M. Wake,N. Takeda, K. NakayamaAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases):K. EguchiStudy supervision:K. Nakayama

AcknowledgmentsWe are grateful to Dr. Yoshio Miki for materials and Takahide Enomoto for

technical assistance.We also thankDrs. Johji Inazawa,Hiroshi Shibuya, HiroshiNishina and members of his laboratory for helpful suggestions. K. Nakayamawas supported by the Takeda Science Foundation, the Japan Foundation forApplied Enzymology, the KatoMemorial Bioscience Foundation, and the IchiroKanehara Foundation. This study was also supported by a Grant-in-Aid forScientific ResearchC (JSPSKAKENHIGrantNumber 15K08260), aGrant-in-Aidfor Scientific Research on Innovative Areas "Oxygen Biology" (Grant Number17H05523, MEXT, Japan), and the Integrated Research Projects on IntractableDisease Program, Medical Research Institute, TMDU.

The costs of publication of this articlewere defrayed inpart by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received June 12, 2017; revised November 11, 2017; accepted January 23,2018; published OnlineFirst February 7, 2018.

Yonashiro et al.





References1. Semenza GL. Targeting HIF-1 for cancer therapy. Nat Rev Cancer 2003;

3:721–32.2. Weidemann A, Johnson RS. Biology of HIF-1alpha. Cell Death Differ

2008;15:621–7.3. Nakayama K. cAMP-response element-binding protein (CREB) and

NF-kappaB transcription factors are activated during prolonged hypoxiaand cooperatively regulate the induction of matrix metalloproteinaseMMP1. J Biol Chem 2013;288:22584–95.

4. Warburg O. On the origin of cancer cells. Science 1956;123:309–14.5. Nakayama K, Gazdoiu S, Abraham R, Pan ZQ, Ronai Z. Hypoxia-induced

assembly of prolyl hydroxylase PHD3 into complexes: implications for itsactivity and susceptibility for degradation by the E3 ligase Siah2. Biochem J2007;401:217–26.

6. Kikuchi D, Minamishima YA, Nakayama K. Prolyl-hydroxylase PHD3interacts with pyruvate dehydrogenase (PDH)-E1beta and regulates thecellular PDH activity. Biochem Biophys Res Commun 2014;451:288–94.

7. Patel MS, Nemeria NS, Furey W, Jordan F. The pyruvate dehydrogenasecomplexes: structure-based function and regulation. J Biol Chem 2014;289:16615–23.

8. Kim JW, Tchernyshyov I, Semenza GL, Dang CV. HIF-1-mediated expres-sion of pyruvate dehydrogenase kinase: a metabolic switch required forcellular adaptation to hypoxia. Cell Metab 2006;3:177–85.

9. Papandreou I, Cairns RA, Fontana L, Lim AL, Denko NC. HIF-1 mediatesadaptation to hypoxia by actively downregulating mitochondrial oxygenconsumption. Cell Metab 2006;3:187–97.

10. Hitosugi T, Fan J, Chung TW, Lythgoe K, Wang X, Xie J, et al. Tyrosinephosphorylation of mitochondrial pyruvate dehydrogenase kinase 1 isimportant for cancer metabolism. Mol Cell 2011;44:864–77.

11. Ohashi Y, Hirayama A, Ishikawa T, Nakamura S, Shimizu K, Ueno Y, et al.Depiction of metabolome changes in histidine-starved Escherichia coli byCE-TOFMS. Mol Biosyst 2008;4:135–47.

12. Ooga T, Sato H, Nagashima A, Sasaki K, Tomita M, Soga T, et al. Meta-bolomic anatomy of an animal model revealing homeostatic imbalancesin dyslipidaemia. Mol Biosyst 2011;7:1217–23.

13. Sugimoto M, Wong DT, Hirayama A, Soga T, Tomita M. Capillary elec-trophoresis mass spectrometry-based saliva metabolomics identified oral,breast and pancreatic cancer-specific profiles. Metabolomics 2010;6:78–95.

14. Cerami E,Gao J,DogrusozU,Gross BE, Sumer SO, Aksoy BA, et al. The cBiocancer genomics portal: an open platform for exploring multidimensionalcancer genomics data. Cancer Discov 2012;2:401–4.

15. Gao J, Aksoy BA, Dogrusoz U, Dresdner G, Gross B, Sumer SO, et al.Integrative analysis of complex cancer genomics and clinical profiles usingthe cBioPortal. Sci Signal 2013;6:pl1.

16. Fukuda R, Zhang H, Kim JW, Shimoda L, Dang CV, Semenza GL. HIF-1regulates cytochrome oxidase subunits to optimize efficiency of respirationin hypoxic cells. Cell 2007;129:111–22.

17. Eales KL, Hollinshead KE, Tennant DA. Hypoxia andmetabolic adaptationof cancer cells. Oncogenesis 2016;5:e190.

18. Cairns RA, Harris IS, Mak TW. Regulation of cancer cell metabolism. NatRev Cancer 2011;11:85–95.

19. Yang W, Zheng Y, Xia Y, Ji H, Chen X, Guo F, et al. ERK1/2-dependentphosphorylation and nuclear translocation of PKM2 promotes the War-burg effect. Nat Cell Biol 2012;14:1295–304.

20. Faubert B, Boily G, Izreig S, Griss T, Samborska B, Dong Z, et al. AMPK is anegative regulator of the Warburg effect and suppresses tumor growth invivo. Cell Metab 2013;17:113–24.

21. Vander Heiden MG, Cantley LC, Thompson CB. Understanding the War-burg effect: the metabolic requirements of cell proliferation. Science2009;324:1029–33.

22. Fogal V, Richardson AD, Karmali PP, Scheffler IE, Smith JW, Ruoslahti E.Mitochondrial p32 protein is a critical regulator of tumor metabolism viamaintenance of oxidative phosphorylation. Mol Cell Biol 2010;30:1303–18.

23. Weinberg F, Hamanaka R, Wheaton WW, Weinberg S, Joseph J, Lopez M,et al. Mitochondrial metabolism and ROS generation are essential forKras-mediated tumorigenicity. Proc Natl Acad Sci U S A 2010;107:8788–93.

24. Dupuy F, Tabaries S, Andrzejewski S, Dong Z, Blagih J, Annis MG, et al.PDK1-dependent metabolic reprogramming dictates metastatic potentialin breast cancer. Cell Metab 2015;22:577–89.

25. Dewhirst MW, Cao Y, Moeller B. Cycling hypoxia and free radicals regulateangiogenesis and radiotherapy response. Nat Rev Cancer 2008;8:425–37.

26. Wallace DC. Mitochondria and cancer. Nat Rev Cancer 2012;12:685–98.27. Smeitink J, van den Heuvel L, DiMauro S. The genetics and pathology of

oxidative phosphorylation. Nat Rev Genet 2001;2:342–52.






2018;78:1592-1603. Published OnlineFirst February 7, 2018.Cancer Res Ryo Yonashiro, Kayoko Eguchi, Masaki Wake, et al. Altering the Metabolic Status of Cancer Cells

Controls Tumor Progression byβPyruvate Dehydrogenase PDH-E1

Updated version

10.1158/0008-5472.CAN-17-1751doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2018/01/26/0008-5472.CAN-17-1751.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/78/7/1592.full#ref-list-1

This article cites 27 articles, 9 of which you can access for free at:

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

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/78/7/1592To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-17-1751

http://cancerres.aacrjournals.org/content/suppl/2018/01/26/0008-5472.CAN-17-1751.DC1

http://cancerres.aacrjournals.org/content/78/7/1592.full#ref-list-1

http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://cancerres.aacrjournals.org/content/78/7/1592


pyruvate dehydrogenase pdh-e1b controls tumor progression by … · metabolism and chemical biology...

Documents