Cancer Letters 364 (2015) 44–58
Contents lists available at ScienceDirect
Cancer Letters
j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / c a n l e t
Original Articles
Epithelial–mesenchymal transition induces similar metabolic
alterations in two independent breast cancer cell lines
Yuvabharath Kondaveeti a, Irene K. Guttilla Reed b, Bruce A. White a,*
a
b
Department of Cell Biology, University of Connecticut Heath Center, Farmington, CT 06030, USA
Department of Biology, University of St. Joseph, West Hartford, CT 06117, USA
A R T I C L E
I N F O
Article history:
Received 24 February 2015
Received in revised form 18 April 2015
Accepted 20 April 2015
Keywords:
Epithelial–mesenchymal transition
Metabolic reprogramming
Aerobic glycolysis
Warburg effect
Breast cancer
A B S T R A C T
Epithelial–mesenchymal transition (EMT) induces invasive properties in epithelial tumors and promotes metastasis. Although EMT-mediated cellular and molecular changes are well understood, very little
is known about EMT-induced metabolic changes. HER2-positive BT-474 breast cancer cells were induced
to undergo a stable EMT using mammosphere culture, as previously described by us for the ERαpositive MCF-7 breast cancer cells. Two epithelial breast cancer cell lines (BT-474 and MCF-7) were compared
to their respective EMT-derived mesenchymal progeny (BT-474EMT and MCF-7EMT) for changes in metabolic pathways including glycolysis, glycogen metabolism, anabolic pathways and gluconeogenesis. Both
EMT-derived cells displayed enhanced aerobic glycolysis along with the overexpression of specific glucose
transporters, lactate dehydrogenase isoforms, monocarboxylate transporters and glycogen phosphorylase isoform. In contrast, both EMT-derived cells suppressed the expression of crucial enzymes in anabolic
pathways and gluconeogenesis. STAT3, a transcription factor involved in tumor initiation and progression, plays a role in the EMT-related changes in the expression of specific enzymes and transporters. This
study provides a broad overview of similar metabolic changes induced by EMT in two independent breast
cancer cell lines. These metabolic changes may provide novel therapeutic targets for metastatic breast
cancer.
© 2015 Elsevier Ireland Ltd. All rights reserved.
Abbreviations: EMT, epithelial–mesenchymal transition; OXPHOS, oxidative phosphorylation; ROS, reactive oxygen species; GATA3, GATA binding protein 3; ZEB1,
zinc finger E-box binding homeobox 1; ZEB2, zinc finger E-box binding homeobox
2; SLUG, zinc-finger binding transcription factor sanil2; ER, estrogen receptor; PR,
progesterone receptor; HER2, amplified ERBB2 oncogene; GLUT1, glucose transporter 1; GLUT3, glucose transporter 3; GLUT12, glucose transporter 12; HK1,
hexokinase 1; HK2, hexokinase 2; HK3, hexokinase 3; GPI, glucose-6-phosphate isomerase; PFKM, phosphofructokinase, muscle; PFKL, phosphofructokinase, liver; PFKP,
phosphofructokinase, platelet; ALDOA, aldolase A; ALDOB, aldolase B; ALDOC, aldolase C; TPI1, triosephosphate isomerase 1; GAPDH, glyceraldehyde-3-phosphate
dehydrogenase; PGK1, phosphoglycerate kinase 1; PGK2, phosphoglycerate kinase
2; PGAM1, phosphoglycerate mutase 1; PGAM2, phosphoglycerate mutase 2; ENO1,
enolase 1; ENO2, enolase 2; PKM1, pyruvate kinase, muscle 1; PKM2, pyruvate kinase,
muscle 2; LDHA, lactate dehydrogenase A; LDHB, lactate dehydrogenase B; LDHC,
lactate dehydrogenase C; MCT1, monocarboxylate transporter 1; MCT2,
monocarboxylate transporter 2; MCT4, monocarboxylate transporter 4; GYS1, glycogen synthase, muscle; GYS2, glycogen synthase, liver; PYGL, glycogen phosphorylase,
liver; PYGB, glycogen phosphorylase, brain; PYGM, glycogen phosphorylase, muscle;
G6PD, glucose-6-phosphate dehydrogenase; TKT, transketolase; TALDO1, transaldolase
1; GFPT1, glutamine–fructose-6-phosphate transaminase 1; PHGDH, phosphoglycerate dehydrogenase; PC, pyruvate carboxylase; PCK1, phosphoenolpyruvate
carboxykinase 1; PCK2, phosphoenolpyruvate carboxykinase 2; FBP1, fructose-1,6bisphosphatase 1; FBP2, fructose-1,6-bisphosphatase 2; G6PC, glucose-6-phosphatase,
catalytic subunit; IL6, interleukin 6; JAK2, janus kinase 2; STAT3, signal transducer
and activator of transcription 3.
* Corresponding author. Tel.: +1 8606792811; fax: +1 8606791269.
E-mail address: bwhite@uchc.edu (B.A. White).
http://dx.doi.org/10.1016/j.canlet.2015.04.025
0304-3835/© 2015 Elsevier Ireland Ltd. All rights reserved.
Introduction
Worldwide, breast cancer accounted for 521,000 global
deaths in 2012, according to the World Health Organization. In the
US, the National Cancer Institute predicts 232,670 new cases
along with 40,000 deaths from breast cancer in 2014. Breast
cancer is composed of several subtypes with distinct molecular
characteristics, clinical behaviors and treatment options. The
most common subtypes of invasive breast carcinoma are estrogen
receptor/progesterone receptor (ER/PR)-positive and human
epidermal growth factor receptor 2 (HER2)-positive [1]. These subtypes are composed of cells that although neoplastically transformed,
still display a significant degree of epithelial organization
including basal–apical polarity and E-cadherin-mediated adherent junctions. These ER/PR-positive and HER2-positive subtypes
are less invasive and confer a more favorable diagnosis compared
to triple-negative (ER-, PR- and HER2-) subtype [2]. Nevertheless,
these subtypes can eventually give rise to distant metastatic
lesions [3].
Metastasis is a complex process that involves a series of events
including localized stromal invasion, intravasation, transport through
circulation, extravasation and colonization. The early phase of
Y. Kondaveeti et al./Cancer Letters 364 (2015) 44–58
localized stromal invasion by epithelial tumor cells involves the
downregulation of intercellular adhesions, loss of apicobasal polarity, and cytoskeletal reorganization that results in a mesenchymal
phenotype, coincident with enhanced motility and invasiveness, and
the acquisition of cancer stem cell properties [4]. These changes constitute the epithelial–mesenchymal transition (EMT), a process that
is orchestrated by the re-expression of embryonic transcription
factors, including ZEB1, ZEB2, SLUG and SNAIL, in response to genetic
damage coupled to multiple signals within the microenvironment, including the hormone/growth factors, degree of oxygenation,
inflammation and extracellular matrix components [5,6]. As EMT
is an early step in the metastatic process, it represents an important target in the development of anti-metastatic adjuvant therapies.
However, the fact that EMT is induced by multiple factors predicts
that complete prevention of EMT is probably not possible. Thus, it
is important to better define the characteristics of EMT-derived
cancer cells that may provide therapeutic targets to inhibit metastasis. Here, we examined metabolic pathways in EMT-derived breast
cancer cells.
Normal cells with healthy mitochondria metabolize glucose into
lactate only under hypoxic conditions. In contrast, tumor cells typically import more glucose and metabolize it into lactate even in
the presence of oxygen. This process is referred to as ‘aerobic glycolysis’ or the ‘Warburg effect’ [7]. A high glycolytic rate
confers the following growth advantages for tumor cells: faster synthesis of ATP as compared to mitochondrial oxidative
phosphorylation (OXPHOS); synthesis of ATP independently of
oxygen; generation of fewer reactive oxygen species (ROS); and
support of cell proliferation by providing increased quantities of
glycolytic intermediates, several of which are precursors for biosynthetic pathways [8,9]. It is important to note that the TCA cycle
and OXPHOS still contribute to ATP production and TCA cyclerelated anabolic pathways to some extent in neoplastic cancer cells.
An increasing body of research suggests that aerobic glycolysis is
just one component of the global changes in the metabolism of
tumor cells compared to normal cells, referred to as ‘metabolic reprogramming’ [10].
The vast majority of studies on metabolic reprogramming have
been performed in the setting of neoplastic transformation. Considerably less is known about metabolic reprogramming in the
context of metastatic transformation. Since EMT promotes metastatic transformation of epithelial tumor cells, we examined whether
EMT induces additional metabolic changes in breast cancer cell lines.
We previously reported that culturing ERα-positive, epithelial MCF-7
breast cancer cells in prolonged mammosphere culture induced EMT
and generated a stable population of mesenchymal cancer cells
(termed MCF-7EMT cells herein) [11]. Compared to parental MCF-7
cells, MCF-7EMT cells are highly motile, generate larger tumors in vivo
and display cancer stem cell phenotype characterized by CD44hi/
CD24lo expression. In the current study, we similarly induced EMT
through mammosphere culture in HER2-positive, epithelial BT474 breast cancer cell line and generated stable mesenchymal BT474EMT cells with CD44hi/CD24low expression. We compared the
metabolic pathways of EMT-derived mesenchymal BT-474EMT and
MCF-7EMT cells with their parental epithelial BT-474 and MCF-7 cells
in order to determine EMT-induced metabolic reprogramming. Here
we show for the first time that EMT induces multiple metabolic
changes including enhanced aerobic glycolysis with increased expression of specific transporters and enzymes related to glycolysis.
Surprisingly, EMT was also associated with a significant suppression of crucial enzymes within some anabolic side pathways and
gluconeogenesis that would otherwise extract carbons from glycolysis or promote the flux of carbons against the flow of the
glycolytic pathway. These novel findings identify several metabolic components as potential therapeutic targets for metastatic breast
cancer.
45
Materials and methods
Cell culture
BT-474 and MCF-7 cell lines were obtained from the American Type Culture Collection (Manassas, VA). EMT in BT-747 cells was induced using prolonged
mammosphere culture method as described earlier [11]. BT-474 and MCF-7 and their
corresponding EMT-derived BT-474EMT and MCF-7EMT cells were all cultured in DMEM/
F-12 supplemented with 10% heat inactivated FBS (Gibco, Grand Island, NY) and 1×
MycoZap™ Plus-CL antibiotic (Lonza, Walkersville, MD). BT-474 and MCF-7 cells were
additionally supplemented with 1× Insulin–Transferrin–Selenium solution (Gibco,
Grand Island, NY). All cells were cultured in a humidified incubator maintained at
5% CO2 and 37 °C.
Flow cytometry
BT-474 and BT-474EMT cells were trypsinized and 1 million cells were plated in
FACS tubes. Cells were washed with FACS buffer (5% FBS in 1× PBS). Cells were labeled
with CD24-PE/Cy7 and CD44-Alexa Fluor 647 conjugated antibodies (BioLegend, San
Diego, CA) in FACS buffer by incubating at 4 °C for 30 minutes in the dark. Cells were
washed twice with FACS buffer and then labeled with viability dye eFluor® 506
(eBioscience, San Diego, CA) for 20 minutes. Cells were washed again and analyzed
on BD LSR II flow cytometer (BD Biosciences, San Jose, CA). Unstained, isotype and
single antibody controls were also performed for each cell line. Only live cells were
included in the data analysis using FlowJo software (Ashland, OR). Flow cytometry
analysis was performed twice for each cell type.
Glucose uptake and lactate production
BT-474 and MCF-7 cells (200,000 cells/well) and BT-474EMT and MCF-7EMT cells
(50,000 cells/well) were plated in regular growth media. Cells were allowed to attach
for 24-hours and washed with PBS. 5 mM glucose medium was added and cells were
cultured for another 24 hours. Glucose and lactate concentrations in the culture
medium were determined by fluorometric based Glucose Assay Kit and Lactate Assay
Kit (BioVision, Inc., Milpitas, CA) according to the vendor’s instructions using Synergy
2 Multi-Mode Microplate Reader (BioTek, Winooski, VT). The amount consumed or
produced by cells was determined by comparing the concentration in the medium
incubated without cells and then normalized to cell number. These assays were performed twice with three experimental replicates for each cell type.
Conditional media and drugs
RPMI 1640 Medium Modified without l-Glutamine, without Amino acids and
Glucose (US Biological, Swampscott, MA) was supplemented with 2.5 mM l-glutamine,
1× MEM essential and non-essential amino acid mixtures, 10% heat inactivated FBS,
15 mM HEPES (Gibco, Grand Island, NY); 0.5 mM sodium pyruvate, chemically defined
lipid mixture 1, 15 mM sodium bicarbonate (Sigma, St. Louis, MO) and 1× MycoZap™
Plus-CL antibiotic (Lonza, Walkersville, MD). To this base medium, 20 mM d-galactose
or 5 mM d-glucose (Sigma, St. Louis, MO) was added to make galactose or 5 mM
glucose media, respectively. 2-Deoxy-d-glucose (Sigma, St. Louis, MO) and metformin
(Tocris Bioscience, Bristol, UK) solutions were made using sterile distilled water and
oligomycin A (Tocris Bioscience, Bristol, UK) solution was made using ethanol.
2-Deoxy-d-glucose, metformin and oligomycin A were used at concentrations of 5 mM,
10 mM and 5 μM, respectively.
Growth and viability assays
For a 6-day growth curve, all four cell lines were plated at a density of 10,000
cells/well in a 6-well plate in regular growth media. Cells were counted on days 2,
4 and 6 using TC20 Automated Cell Counter (Bio-Rad, Hercules, CA). For growth analysis (in 20 mM galactose, 2-deoxy-d-glucose, metformin and oligomycin A), BT474 and MCF-7 cells were plated at a density of 300,000 cells/well and BT-474EMT
and MCF-7EMT cells were plated at a density of 50,000 cells/well in a 6-well plate
in the regular growth media. Cells were allowed to attach for 24 hrs, washed with
PBS and medium of interest was added. Cells were cultured for 3 days and then live
cells were counted using TC20 Automated Cell Counter (Bio-Rad, Hercules, CA) by
trypan blue exclusion method. Viability assays were performed three times for each
cell type.
Real-time qPCR and primers
In all assays, three separate RNA samples for each cell type extracted on different days or independent experiments were analyzed. Total RNA was isolated using
TRIzol reagent (Ambion RNA, Carlsbad, CA). DNase treatment was performed using
TURBO DNA-free kit (Ambion RNA, Carlsbad, CA). cDNA was made from 1 μg of total
RNA using iSCRIPT cDNA synthesis kit (Bio-Rad, Hercules, CA). 50 ng cDNA was used
to perform real-time PCR using SYBR green based SsoFast EvaGreen Supermix (BioRad, Hercules, CA). Gene expression was normalized to TATA-box binding protein
(TBP) using 2−ΔCt method. Relative expression (2−ΔCt) values less than 0.001 were considered extremely low/undetectable; between 0.001 and 0.01 were considered very
46
Y. Kondaveeti et al./Cancer Letters 364 (2015) 44–58
low; and between 0.01 and 0.1 were considered low. Specificity of the primer sets
was confirmed by melting curve analysis. Primer sequences are listed in
Supplementary Table S1.
Western blot and antibodies
Western blot analysis was performed three times using three independent protein
samples for each cell type isolated on different days. Protein lysates were prepared
in RIPA buffer (TEKnova, Hollister, CA) containing Halt Protease and Phosphatase Inhibitor Cocktail (Thermo Scientific, Rockford, IL). Protein concentrations were measured
using BCA assay (Thermo Scientific Pierce, Rockford, IL) and 10 μg of protein was
run on 10% SDS-PAGE. Proteins were transferred to nitrocellulose membrane and
blocked with 5% BSA in 1× TBST. Primary and secondary antibody incubations were
done according to the vendor’s instructions. Blots were labeled using Amersham ECL
prime (GE Healthcare Life Sciences, Buckinghamshire, UK) and imaged using G:box
imaging system (SynGene, Cambridge, UK). Primary antibodies to G6PD (1:2500),
TKT (1:1000), GFPT1 (1:500), PHGDH (1:2500) LDHA (1:10,000), PCK2 (1:2500), STAT3
(1:2500) and phospho-STAT3 (1:500) were purchased from Cell Signaling Technology (Danvers, MA); TALDO1 (1:2500) was purchased from Life Technologies (Carlsbad,
CA); FBP1 (1:2500) was purchased from Sigma-Aldrich (St. Louis, MO); β-actin (1:2500)
was purchased from Abcam (Cambridge, MA); LDHB (1:10,000) was purchased from
OriGene (Rockville, MD).
Clustered image map (CIM)/heat map
Real-time qPCR analysis of glycolytic enzymes in all four cell lines was performed as described above. The relative expression values (2−ΔCt values) were used
to generate heat map using CIMminer, an online service developed by Genomics and
Bioinformatics group.
STAT3 inhibitor treatment
BT-474EMT and MCF-7EMT cells were plated at a density of 100,000 cells/well in a
6-well plate in the regular growth media. Cells were allowed to attach for 24 hrs, washed
with PBS and regular growth medium containing S3I-201 (Tocris Bioscience, Bristol,
UK) at final concentrations of 40 μM, 60 μM and 80 μM was added. After 24 hrs, RNA
from the cells was isolated and real-time qPCR was performed as described above.
These experiments were performed two times, each with duplicate samples.
A
BT-474
B
Statistical analysis
Ordinary one-way ANOVA followed by Sidak’s multiple comparison post-test or
two-way ANOVA was performed to determine statistical significance. P < 0.05 was
considered to be significant. Statistical analysis was performed using GraphPad Prism
software (La Jolla, CA).
Results
Serial mammosphere culture induces EMT in BT-474 cells
In the current study, we induced EMT in the HER2-positive BT474 breast cancer cell line using 3-dimensional mammosphere
culture as previously described by us [11]. During the first two weeks,
BT-474 cells formed tightly adherent spheroids, but after the third
week, spheroids exhibited a looser configuration (Fig. 1A). Upon
transfer of dispersed spheroids from the fifth week back to standard 2-dimensional monolayer cell culture, the cells, termed BT474EMT, displayed a stable mesenchymal phenotype (Fig. 1A). Flow
cytometry analysis characterized epithelial BT-474 cells as CD44lo/
CD24hi, whereas BT-474EMT cells as CD44hi/CD24lo, a putative breast
cancer stem cell phenotype (Fig. 1B). Both the mesenchymal BT474EMT and MCF-7EMT cells (also referred to as ‘EMT-derived cancer
cells’) proliferated more rapidly than the corresponding epithelial
BT-474 and MCF-7 cells (also referred to as ‘parental cancer cells’)
in standard 2-dimensional culture (Fig. 1C). Epithelial markers,
E-cadherin and GATA3, were highly expressed in the parental cancer
cells, but suppressed in EMT-derived cancer cells (Fig. 2A). In contrast, the mesenchymal marker, vimentin (Fig. 2B), and the EMTassociated transcription factors, ZEB1, ZEB2 and SLUG (Fig. 2C), were
robustly expressed in the EMT-derived cancer cells only. EMTderived cancer cells lost the expression of hormone receptors, ER,
Week 1 - 2
BT-474
BT-474EMT
Week 3 - 5
C
Growth Curve
Viable Cell Number
1.5
CD24
BT-474EMT
106
MCF-7EMT
BT-474EMT
1.0 106
****
MCF-7
BT-474
****
5.0 105
0.0
CD44
Day 0
Day 2
Day 4
Day 6
Fig. 1. BT-474 cells cultured in mammosphere conditions undergo stable EMT. (A) Images showing morphological changes in BT-474 cells in mammosphere culture for
5 weeks. BT-474 cells grown in the monolayer culture (left), images from different weeks in the mammosphere culture (middle) and BT-474EMT in the monolayer culture
(right). (B) Flow cytometry analysis showing the expression of CD24 and CD44 surface markers in epithelial BT-474 and mesenchymal BT-474EMT cells. (C) Viable cell counts
for BT-474 and MCF-7 (shown in blue lines); and BT-474EMT and MCF-7EMT (shown in green lines) at days 2, 4 and 6 of culture after plating at equal density. Note that the
growth curves of BT-474 and MCF-7 are superimposed on each other. Results are presented as mean ± S.E.M. ****p < 0.0001 (two-way ANOVA).
Y. Kondaveeti et al./Cancer Letters 364 (2015) 44–58
2
****
C
100
80
60
15
10
5
**
0
MCF-7 MCF-7EMT
0.4
2
1
Relative mRNA levels
Relative mRNA levels
****
***
D
****
****
0.1
BT-474 BT-474EMT
0.4
****
***
0.2
MCF-7 MCF-7EMT
BT-474 BT-474EMT
PR
0.02
****
40
35
30
25
20
0.4
0.3
0.2
0.1
****
0.0
0.00
MCF-7 MCF-7EMT
MCF-7 MCF-7EMT
HER2
40
0.04
MCF-7 MCF-7EMT
0.0
BT-474 BT-474EMT
Relative mRNA levels
Relative mRNA levels
0
SLUG
0.2
ER
BT-474 BT-474EMT
100
0.6
0.3
MCF-7 MCF-7EMT
0.06
****
200
MCF-7 MCF-7EMT
0.0
0
BT-474 BT-474EMT
****
300
ZEB2
ZEB1
4
3
****
BT-474 BT-474EMT
Relative mRNA levels
****
BT-474 BT-474EMT
Vimentin
****
400
120
Relative mRNA levels
4
Relative mRNA levels
Relative mRNA levels
6
0
B
GATA3
E-cadherin
8
****
BT-474 BT-474EMT
MCF-7 MCF-7EMT
Relative mRNA levels
A
47
35
30
25
20
0.6
0.4
0.2
****
****
0.0
BT-474 BT-474EMT
MCF-7 MCF-7EMT
Fig. 2. BT-474EMT and MCF-7EMT cells show EMT markers and loss of hormone receptors. Real-time qPCR analysis of (A) epithelial markers: E-cadherin and GATA3; (B)
mesenchymal marker, vimentin; (C) EMT transcription factors: ZEB1, ZEB2 and SLUG; (D) hormone/growth factor receptors: ER, PR and HER2 in BT-474, BT-474EMT, MCF-7
and MCF-7EMT cells. Results are presented as mean ± S.E.M. **p < 0.01; ***p < 0.001; ****p < 0.0001 (ordinary one-way ANOVA).
PR and HER2, compared to parental cancer cells and resemble triplenegative breast cancer cells (Fig. 2D).
BT-474EMT and MCF-7EMT cells display enhanced aerobic glycolysis
and glycolytic dependency
The two EMT-derived cancer cells (BT-474EMT and MCF-7EMT) were
then compared to their corresponding parental cancer cells (BT474 and MCF-7) in order to examine the metabolic changes induced
by EMT. We first measured glucose uptake and lactate production
rates in all four cell lines in order to assess the relative glycolytic
rate. Glucose uptake was 1.8-fold higher in BT-474EMT cells (p < 0.001)
and 1.5-fold higher in MCF-7EMT cells (p < 0.01) as compared to BT474 and MCF-7 cells, respectively (Fig. 3A). Further, the enhanced
glucose uptake in EMT-derived cancer cells was matched by a significant increase in lactate production (Fig. 3A). Thus, both EMTderived cancer cells demonstrated enhanced glycolytic rate compared
to the parental cancer cells.
In order to determine the relative dependency on glycolysis for
growth of all four cell lines, we cultured them in a medium in which
glucose was replaced with 20 mM galactose. In this medium, the
growth of the EMT-derived cancer cells was more severely affect-
ed compared to the parental cancer cells (Fig. 3B). We also examined
the relative sensitivity of each cell line to the glycolytic inhibitor,
2-deoxyglucose. Consistent with the above finding, 2-deoxyglucose
showed a significantly more pronounced effect on the growth of
EMT-derived cancer cells compared to parental cancer cells (Fig. 3B).
In contrast, inhibition of oxidative phosphorylation with either
metformin (complex I inhibitor) or oligomycin A (ATP synthase inhibitor) suppressed the growth of the two parental cancer cells
significantly more than the EMT-derived cancer cells (Fig. 3C). Hence,
increased glycolysis in association with elevated glycolytic dependency and decreased OXPHOS dependency indicates that EMTderived cancer cells have enhanced aerobic glycolysis compared to
the parental cancer cells.
Enhanced aerobic glycolysis in BT-474EMT and MCF-7EMT cells is
associated with changes in the expression of specific transporters
and enzymes
Increased glucose uptake is associated with elevated expression of
GLUT3 and GLUT12 in BT-474EMT and MCF-7EMT cells
Glucose transport across the plasma membrane is the first ratelimiting step for glycolysis. We measured the mRNA levels of the
48
Y. Kondaveeti et al./Cancer Letters 364 (2015) 44–58
A
Glucose Uptake
Lactate Production
15
**
****
10
5
µmol / 106 cells / day
µmol / 106 cells / day
15
0
5
0
BT-474 BT-474EMT
B
MCF-7 MCF-7EMT
BT-474 BT-474EMT
58.4%
61%
40
****
81.5%
****
30
80.4%
20
10
0
BT-474 BT-474EMT
C
MCF-7 MCF-7EMT
Cell number (% of Control)
70
60
50
80
40
****
20
***
**
51.2%
57.8%
66.1%
71%
30
20
10
0
BT-474 BT-474EMT
****
92.7%
90.6%
0
BT-474 BT-474EMT
MCF-7 MCF-7EMT
Oligomycin A
50
40
42.3%
44.5%
60
Metformin
60
MCF-7 MCF-7EMT
2-Deoxyglucose
80
Cell number (% of Control)
Cell number (% of Control)
Cell number (% of Control)
Galactose
70
****
****
10
MCF-7 MCF-7EMT
****
37.4%
****
39.8%
70
60
50
40
30
77.6%
74.3%
20
10
0
BT-474 BT-474EMT
MCF-7 MCF-7EMT
Fig. 3. BT-474EMT and MCF-7EMT cells exhibit enhanced aerobic glycolysis and glycolytic dependency. (A) Glucose uptake and lactate production rates in BT-474, BT474EMT, MCF-7 and MCF-7EMT cells. (B) Viable cell numbers represented as percent decrease relative to control after culturing cells for 3 days in the media containing 20 mM
galactose without glucose (relative to DMEM/F12), and media containing 5 mM glucose with 5 mM 2-deoxyglucose (relative to 5 mM glucose media). (C) Viable cell numbers
represented as percent decrease relative to control after culturing cells for 3 days in the media containing 10 mM metformin and 5 μM oligomycin A. Results are presented
as mean ± S.E.M. **p < 0.01; ***p < 0.001; ****p < 0.0001 (ordinary one-way ANOVA).
following glucose transporters (GLUTs), GLUT1, GLUT3, and GLUT12,
all of which have been shown to be upregulated in cancer. Both
GLUT3 and GLUT12 expressions were significantly elevated in both
EMT-derived cancer cells as compared to the parental cancer cells
(Fig. 4A). In BT-474EMT cells, the mRNA levels for GLUT3 and GLUT12
increased 94-fold and 18-fold, respectively. In MCF-7EMT cells, GLUT3
and GLUT12 mRNA levels increased 4.5-fold and 3-fold, respectively. Increased GLUT1 expression has been reported in numerous
cancers. GLUT1 expression was relatively robust in both parental
cancer cells, but was not significantly altered in BT-474EMT cancer
cells as compared to BT-474 cells, and was modestly (1.3-fold) elevated in the MCF-7EMT cells as compared to MCF-7 cells (Fig. 4A).
Imported glucose is ‘activated’ for metabolism and prevented from
leaving the cells by phosphorylation to glucose-6-phosphate by hexokinases (HKs). HK1 mRNA levels dropped by 50% and 20% in BT474EMT and MCF-7EMT cells, respectively (Fig. 4B). HK2 mRNA levels
decreased by 30% in BT-474EMT cells, but increased 2.3-fold in MCF7EMT cells (Fig. 4B). HK3 expression was essentially absent in all the
four cell lines (data not shown). Although somewhat lower in the
EMT-derived cancer cells, HK1 and HK2 mRNA levels remained mod-
erately abundant, at about an order of magnitude higher than GLUT3
and GLUT12 mRNA levels. In summary, these findings indicate that
the significant increase in glucose uptake in EMT-derived cancer cells
is due to an upregulation of GLUT3 and GLUT12.
Increased lactate production in BT-474EMT and MCF-7EMT cells is
associated with upregulation of lactate dehydrogenases and specific
lactate transporters
Pyruvate is converted into lactate or vice versa by tetrameric isozymes formed by lactate dehydrogenase isoforms, LDHA and LDHB.
The protein levels of LDHA and LDHB were increased by 3- to 7-fold
in EMT-derived cancer cells as compared to the parental cancer cells
(Fig. 5A). The LDHB mRNA was significantly increased in both BT474EMT and MCF-7EMT cells, and LDHA mRNA was significantly
increased only in MCF-7EMT cells (Fig. 5A). Of note, BT-474 cells displayed a high level of LDHA mRNA, but a very low level of LDHA
protein, indicating translational or posttranslational regulation of
LDHA expression in these cells (Fig. 5A). LDHC expression was undetectable in the four cell lines (data not shown).
Y. Kondaveeti et al./Cancer Letters 364 (2015) 44–58
GLUT1
**
0.2
0.1
0.0
****
0.4
0.3
***
0.2
0.1
B
MCF-7 MCF-7EMT
BT-474 BT-474EMT
HK1
**
0.4
0.3
**
0.2
0.1
MCF-7 MCF-7EMT
BT-474 BT-474EMT
MCF-7 MCF-7EMT
HK2
4
Relative mRNA levels
8
Relative mRNA levels
0.5
0.0
0.0
BT-474 BT-474EMT
6
4
GLUT12
GLUT3
0.5
Relative mRNA levels
Relative mRNA levels
0.3
Relative mRNA levels
A
49
****
2
0
3
*
2
1
0
BT-474 BT-474EMT
MCF-7 MCF-7EMT
BT-474 BT-474EMT
MCF-7 MCF-7EMT
Fig. 4. BT-474EMT and MCF-7EMT cells show upregulation of glucose transporters, GLUT3 and GLUT12. Real-time qPCR analysis of (A) glucose transporters: GLUT1, GLUT3
and GLUT12; (B) hexokinases: HK1 and HK2 in BT-474, BT-474EMT, MCF-7 and MCF-7EMT cells. Results are presented as mean ± S.E.M. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001
(ordinary one-way ANOVA).
Lactate is transported across the cell membrane by the monocarboxylate transporters (MCTs). MCT1 and MCT2 are primarily involved in pyruvate/lactate import, whereas MCT4 is primarily
involved in lactate export. MCT1 expression was very low in BT474 and BT-474EMT cells, and declined to very low levels in MCF7EMT cells compared to MCF-7 cells (Fig. 5B). In contrast, MCT2 mRNA
levels increased 70-fold in BT-474EMT cells as compared to BT-474
cells. MCT2 mRNA was relatively abundant in MCF-7 cells, and increased modestly in MCF-7EMT cells (Fig. 5B). MCT4 mRNA levels were
elevated 11-fold and 15-fold, respectively, in BT-474EMT and MCF7EMT cells compared to BT-474 and MCF-7 cancer cells (Fig. 5B). In
summary, enhanced production and export of lactate in EMTderived cancer cells are likely driven by an upregulation of LDHA
and LDHB enzymes, and an upregulation of MCT2 and MCT4
transporters.
Expression of glycolytic enzymes was relatively robust and showed
little change in BT-474EMT and MCF-7EMT cells
The mRNA levels of the isoforms of all enzymes within the glycolytic pathway were also assayed by qPCR in EMT-derived and their
respective parental cancer cells. Overall, the relative expression levels
of glycolytic enzymes, or of at least one isoform, were high in all
the four cell lines (Fig. 6). Also, the overall expression of glycolytic
enzymes either did not change, or decreased somewhat in the EMTderived cancer cells compared to the corresponding parental cancer
cells.
Given the limitations inherent in using the enzyme mRNA level
as an approximation of enzyme activity, these findings indicate that
the EMT-induced increase in aerobic glycolysis was due to significant changes at the very beginning (i.e., GLUT transporters) and very
end (i.e., LDHs, MCT transporters) of an otherwise active glycolytic pathway that was already present within the parental cancer
cells.
BT-474EMT and MCF-7EMT cells upregulate glycogen phosphorylase
liver isoform, which breaks down glycogen
Glycogen serves as a fuel reserve and provides glucose for the
cells under low energy conditions. Glycogen metabolism was assessed by measuring the expression of the enzymes that synthesize
glycogen (glycogen synthases: GYS1 and GYS2) and that break down
glycogen (glycogen phosphorylases: PYL, PYB and PYM). GSY1 mRNA
was downregulated in BT-474EMT cells compared to BT-474 cells, its
expression did not significantly changed in MCF-7 EMT cells
compared to MCF-7 cells (Fig. 7A). Although, GYS2 mRNA was
upregulated in EMT-derived cancer cells, its expression levels are
10-fold lower compared to GYS1 (Fig. 7A). In contrast, the liver
isoform of glycogen phosphorylase (PYGL) showed a dramatic increase (i.e., 10- to 200-fold) in the EMT-derived cancer cells compared
to parental cancer cells (Fig. 7B). Other glycogen phosphorylase
isoforms, PYGB and PYGM, were expressed at very low levels compared to PYGL in EMT-derived cancer cells (Fig. 7B). These data
indicate that the breakdown of glycogen, as opposed to its storage,
is maintained in the EMT-derived cancer cells in order to provide
carbons for glycolysis.
BT-474EMT and MCF-7EMT cells display decreased expression of crucial
enzymes in some anabolic side pathways
Glycolysis not only generates ATP rapidly from a typically abundant fuel, but also supplies carbons to anabolic side pathways
involved in the synthesis of macromolecules that are needed for rapid
cell proliferation. Given the faster growth rate of the EMT-derived
cancer cells, we hypothesized that the anabolic side pathways would
also be enhanced in addition to glycolysis in EMT-derived cancer
cells. Because the first enzymes within these anabolic side pathways determine the flux into the respective pathway, we examined
50
Y. Kondaveeti et al./Cancer Letters 364 (2015) 44–58
LDHB
LDHA
150
****
150
100
50
Relative mRNA levels
Relative mRNA levels
200
100
50
0
0
LDHA
LDHB
b-actin
b-actin
BT-474 BT-474EMT MCF-7 MCF-7EMT
B
BT-474 BT-474EMT MCF-7 MCF-7EMT
MCT1
MCT2
3
2.0
1.5
1.0
0.5
****
0.0
BT-474 BT-474EMT
MCF-7 MCF-7EMT
Relative mRNA levels
2.5
Relative mRNA levels
****
****
MCT4
****
2
1
****
0
8
Relative mRNA levels
A
****
6
4
****
2
0
BT-474 BT-474EMT
MCF-7 MCF-7EMT
BT-474 BT-474EMT
MCF-7 MCF-7EMT
Fig. 5. BT-474EMT and MCF-7EMT cells show upregulation of lactate dehydrogenases along with specific monocarboxylate transporters. Real-time qPCR analysis (top)
and western blot analysis (bottom) of (A) lactate dehydrogenases: LDHA and LDHB; (B) real-time qPCR analysis of monocarboxylate transporters, MCT1, MCT2 and MCT4 in
BT-474, BT-474EMT, MCF-7 and MCF-7EMT cells. Western blots are representative of 3 independent sets of samples. Results are presented as mean ± S.E.M. ****p < 0.0001 (ordinary one-way ANOVA).
the expression of first enzymes in three pathways: (1) the pentose
phosphate pathway (PPP), (2) the hexosamine biosynthetic pathway
(HBP) and (3) the serine biosynthesis pathway (SBP).
Oxidative PPP is suppressed in BT-474EMT and MCF-7EMT cells
The pentose phosphate pathway (PPP) provides cells with NADPH,
a cofactor used for reductive biosynthesis and maintenance of cellular redox potential, and with ribose-5-phosphate, which is essential
for nucleic acid synthesis. The reactions in PPP are divided into two
pathways: (1) the oxidative pathway, which utilizes glucose-6phosphate as a precursor, and (2) non-oxidative pathway, which
utilizes fructose-6-phosphate and glyceraldehyde-3-phosphate as
precursors. Oxidative PPP provides cells with NADPH, and the nonoxidative PPP provides cells with ribose-5-phosphate. Glucose-6phosphate dehydrogenase (G6PD) is the first enzyme in the
irreversible oxidative PPP. Transketolase (TKT) and transaldolase
(TALDO1) are the key enzymes in the reversible non-oxidative PPP.
The mRNA and protein levels of G6PD were strongly decreased in
EMT-derived cancer cells as compared to the corresponding parental cancer cells (Fig. 8A). Although the TKT mRNA levels did not
change, the corresponding protein levels were significantly decreased in EMT-derived cancer cells as compared to parental cancer
cells (Fig. 8A). Both mRNA and protein levels of TALDO1 remained
unaltered in EMT-derived cancer cells as compared to parental cancer
cells (Fig. 8A). These findings indicate that EMT-derived cancer cells
suppress oxidative PPP by essentially silencing the expression of
G6PD. The EMT-derived cancer cells also decrease the non-oxidative
PPP to a lesser degree by downregulating the expression of TKT.
HBP is not significantly altered in BT-474EMT and MCF-7EMT cells
HBP utilizes fructose-6-phophate to generate UDP-GlucNAc, which
is used for making glycoproteins and glycolipids. The first enzyme
in this pathway is glutamine–fructose-6-phosphate transaminase
1 (GFPT1). GFPT1 mRNA levels were significantly increased in MCF7EMT cells but remain unchanged in BT-474EMT cells, and GFPT1 protein
levels were not altered in EMT-derived cancer cells compared to parental cancer cells (Fig. 8B). These findings indicate that the
hexosamine biosynthetic pathway remained intact in EMT-derived
cancer cells compared to parental cancer cells.
SBP is suppressed in BT-474EMT and MCF-7EMT cells
SBP utilizes the glycolytic intermediate 3-phosphoglycerate to
generate l-serine, which is a precursor used for the biosynthesis
of other amino acids, proteins, nucleotides, and complex lipids. Phosphoglycerate dehydrogenase (PHGDH) catalyzes the first, irreversible
reaction in this pathway, while consuming NAD+. Thus, the SBP competes with glycolysis for both carbons and a rate-limiting cofactor.
The mRNA and protein levels of PHGDH were strongly decreased
in EMT-derived cancer cells as compared to parental cancer cells
(Fig. 8B). This indicates that EMT-derived cancer cells suppress SBP
by downregulating the expression of PHGDH.
BT-474EMT and MCF-7EMT cells display decreased expression of
enzymes in gluconeogenesis
Carbons from non-carbohydrate precursors can be diverted for
glucose synthesis through gluconeogenesis. We examined the
Y. Kondaveeti et al./Cancer Letters 364 (2015) 44–58
BT-474
BT-474EMT
MCF-7
51
MCF-7EMT
GAPDH
GPI
PFKP
ALDOC
ENO2
PGAM2
ALDOB
PGK2
PFKL
PFKM
PKM1
ENO1
ALDOA
PGK1
PKM2
TPI1
634.39
90.38
121.82
66.15
21.03
48.07
4.07
5.35
1.81
0.01
0.74
0.00
PGAM1
Fig. 6. BT-474EMT and MCF-7EMT cells display no significant changes in glycolytic enzymes. Expression of glycolytic enzymes shown as a clustered image map (heat map)
in BT-474, BT-474EMT, MCF-7 and MCF-7EMT cells using relative expression values (2−ΔCt values) from real-time qPCR analysis. Enzyme abbreviations: GAPDH: glyceraldehyde3-phosphate dehydrogenase; GPI: glucose-6-phosphate isomerase; PFKP: phosphofructokinase, platelet; ALDOC: aldolase C; ENO2: enolase 2; PGAM2: phosphoglycerate
mutase 2; ALDOB: aldolase B; PGK2: phosphoglycerate kinase 2; PFKL: phosphofructokinase, liver; PFKM: phosphofructokinase, muscle; PKM1: pyruvate kinase, muscle 1;
ENO1: enolase 1; ALDOA: aldolase A; PGK1: phosphoglycerate kinase 1; PKM2: pyruvate kinase, muscle 2; TPI1: triosephosphate isomerase 1; PGAM1: phosphoglycerate
mutase 1.
expression of gluconeogenic enzymes and their isoforms: pyruvate carboxylase (PC), phosphoenolpyruvate carboxykinase isoforms
1 (PCK1; cytoplasmic) and 2 (PCK2; mitochondrial), the two cytoplasmic isoforms of fructose-1,6-bisphosphatase (FBP1 and FBP2),
and glucose-6-phosphatase catalytic subunit (G6PC). PC expression was significantly decreased in MCF-7EMT cells but not in BT474EMT cells compared to their respective parental cancer cells
(Fig. S1). The expressions of PCK2 and FBP1 were significantly suppressed at both mRNA and protein levels in the EMT-derived cancer
cells compared to parental cancer cells (Fig. 8C). mRNA levels of both
PCK1 and FBP2 were very low in parental cancer cells and their levels
were further decreased in EMT-derived cancer cells (Fig. S1). As expected for non-gluconeogenic cell types, G6PC mRNA was
undetectable in all four cell lines (data not shown). These data indicate that gluconeogenesis is suppressed in EMT-derived cancer
cells due to the loss of key enzymes, PCK2 and FBP1.
STAT3 signaling pathway regulates EMT-associated changes in some
enzyme/transporter expression
The IL6/JAK2/STAT3 signaling pathway has been shown to induce
EMT in MCF-7 breast cancer cells, and to be active in CD44+/CD24−
breast cancer stem cells [12,13]. Moreover, IL6 can generate an
autocrine positive feedback loop in which STAT3 promotes the
nuclear retention and activation of NF-κB (RelA), which in turn stimulates IL6 expression [14]. We previously reported an upregulation
of IL6 in MCF-7EMT cells [11]. Thus, we examined whether the IL6/
JAK2/STAT3 pathway might be involved in the gene expression
changes observed in the EMT-derived cancer cells. In this study, we
observed that IL6 mRNA expression was very low in parental cancer
cells, but was highly expressed in the EMT-derived BT-474EMT cells
(a 175-fold increase) and MCF-7EMT cells (a 70-fold increase) (Fig. 9A).
On the other hand, JAK2 mRNA levels were detectable but did not
parallel the IL6 expression (Fig. 9A). Although STAT3 mRNA was elevated only in MCF-7EMT cells compared to MCF-7 cells, STAT3 protein
was significantly elevated in both EMT-derived cancer cells compared to parental cancer cells (Fig. 9A). In addition, the activated
form of STAT3, phospho-STAT3 (Tyr 705), was only detectable in
EMT-derived cancer cells (Fig. 9A). These data indicate that STAT3
signaling is active in EMT-derived cancer cells but not in parental
cancer cells.
In order to examine the role of STAT3 signaling in EMT-related
changes, all the four cell lines were treated with the selective STAT3
inhibitor, S3I-201. STAT3 inhibition reversed the EMT-associated
changes in the expression of ZEB1, MCT2 and PCK2 in a dosedependent manner in both EMT-derived cancer cells (Fig. 9B–D).
STAT3 inhibition had no effect on G6PD mRNA in both EMTderived cancer cells (Fig. S2A). PHGDH mRNA was not affected by
STAT3 inhibition in BT-474EMT cells, but displayed a dose-dependent
increase in MCF-7EMT cells (Fig. S2B).
In summary, STAT3 signaling is required for optimal expression of ZEB1 and MCT2. Moreover, STAT3 signaling appears to play
a role in EMT-related suppression of PCK2 in both BT-474EMT and
MCF-7EMT cells, and PHGDH in MCF-7EMT cells, but not BT-474EMT cells.
52
Y. Kondaveeti et al./Cancer Letters 364 (2015) 44–58
A
GYS2
GYS1
0.015
Relative mRNA levels
0.8
0.6
0.4
***
0.2
**
0.005
0.000
0.0
BT-474 BT-474EMT
B
BT-474 BT-474EMT
MCF-7 MCF-7EMT
MCF-7 MCF-7EMT
PYGM
PYGB
PYGL
15
0.015
1.5
****
****
10
5
0
Relative mRNA levels
Relative mRNA levels
***
0.010
Relative mRNA levels
Relative mRNA levels
1.0
1.0
**
0.5
MCF-7 MCF-7EMT
0.005
****
****
0.000
0.0
BT-474 BT-474EMT
0.010
BT-474 BT-474EMT
MCF-7 MCF-7EMT
BT-474 BT-474EMT
MCF-7 MCF-7EMT
Fig. 7. BT-474EMT and MCF-7EMT cells upregulate glycogen phosphorylase liver isoform, which breaks down glycogen. Real-time qPCR analysis of (A) glycogen synthase
isoforms: GYS1 and GYS2; (B) glycogen phosphorylase isoforms: PYGL, PYGB and PYGM in BT-474, BT-474EMT, MCF-7 and MCF-7EMT cells. Results are presented as mean ± S.E.M.
**p < 0.01; ***p < 0.001; ****p < 0.0001 (ordinary one-way ANOVA).
Further work is required to identify other EMT-related signaling pathways that specifically contribute to the changes in the expression
of other enzymes and transporters in BT-474EMT and MCF-7EMT cells.
Discussion
From previous work and the current study, we generated mesenchymal breast cancer cells, BT-474EMT and MCF-7EMT, by inducing
EMT in two substantially different epithelial breast cancer cell lines,
BT-474 and MCF-7, respectively. In conjunction with previous reports,
EMT-derived cancer cells display breast cancer stem cell phenotype characterized by CD44hi/CD24lo expression [15]. Moreover, EMTderived cancer cells proliferate more rapidly compared to the
parental cancer cells, a finding that is somewhat at odds with other
studies on mesenchymal cancer cells generated from EMT [16].
However, increased proliferation of EMT-derived cancer cells is supported by their resemblance with triple-negative breast cancer cells,
which both proliferate faster than hormone receptor-positive cells
and display a less-differentiated phenotype [17].
We then utilized the EMT-derived cancer cells and the corresponding parental cancer cells, to examine whether EMT is associated
with changes in metabolic pathways. To summarize our findings,
EMT-derived cancer cells display an increase in aerobic glycolysis,
as indicated by increased glucose uptake and lactate production rates,
and a higher dependency on glycolysis for growth. This indicates
that EMT-derived cancer cells highly depend on glycolysis for oxygen
independent growth and ATP synthesis. This EMT-related increase
in aerobic glycolysis does not appear to be driven by an overall increase in the expression of glycolytic enzymes, although these
enzymes are already expressed at a relatively high level in the epithelial breast cancer cell lines. Instead, the EMT-related increase
in aerobic glycolysis occurs through the upregulation of specific
glucose transporters, lactate dehydrogenases and lactate transporters. The findings on the downregulation of crucial enzymes in some
anabolic side pathways and gluconeogenesis in EMT-derived cancer
cells are surprising given previous studies that have emphasized the
need for anabolic pathways in cancer (see below). Nevertheless, these
findings consistently indicate that, at least in some contexts, rapidly
dividing EMT-derived cancer cells protect glycolytic carbons, probably at the expense of cell-autonomous independence from
extracellular nutrients.
The increased glucose uptake and lactate production rates, greater
sensitivity to glucose restriction (i.e., glucose-free galactose media)
and 2-deoxyglucose treatment, and decreased sensitivity to OXPHOS
inhibitors demonstrate that EMT-derived cancer cells have enhanced aerobic glycolysis and are highly glycolytic as compared to
parental cancer cells. These findings are supported by the recent
report that breast cancer stem cells rely on aerobic glycolysis and
are also highly sensitive to 2-deoxyglucose [18].
Enhanced expression of glucose transporters, particularly GLUT1,
has been reported for several types of cancers [19]. In the current
study, GLUT1 expression was readily detectable in parental cancer
cells, and showed little/no change in EMT-derived cancer cells. In
contrast, GLUT3 expression was significantly increased in EMTderived cancer cells. GLUT3 has a higher affinity and greater transport
capacity for glucose than GLUT1 and is upregulated in many cancer
cell lines and cancer tissues [20]. A recent study linked GLUT3 expression to tumor initiating capability of brain tumor stem cells [21].
With respect to EMT, GLUT3 expression is increased during EMT
induced through TGF-β or by ectopic expression of ZEB1 or SNAIL
in non-small cell lung cancer [22].
GLUT12 is a Class III non-canonical glucose transporter that was
identified in the breast cancer cell line, MCF-7, based on its homology to GLUT4 [23]. GLUT12 is co-expressed in skeletal muscle with
GLUT4 at about a 1:10 ratio [24]. However, the affinity and transport capacity of GLUT12 for glucose has not been well-established
[25]. In our study, GLUT4 expression was undetectable in all four
cell lines (data not shown), whereas GLUT12 expression was low
Y. Kondaveeti et al./Cancer Letters 364 (2015) 44–58
A
G6PD
TKT
6
4
****
***
30
Relative mRNA levels
8
4
3
2
1
0
10
0
TKT
TALDO1
b-actin
b-actin
BT-474 BT-474EMT
B
b-actin
BT-474 BT-474EMT
MCF-7 MCF-7EMT
0.3
0.2
0.1
MCF-7 MCF-7EMT
2.0
1.5
1.0
0.5
0.0
0.0
GFPT1
PHGDH
b-actin
b-actin
C
BT-474 BT-474EMT
2.5
Relative mRNA levels
***
0.4
BT-474 BT-474EMT
MCF-7 MCF-7EMT
PHGDH
GFPT1
0.5
Relative mRNA levels
20
0
G6PD
*
****
BT-474 BT-474EMT
MCF-7 MCF-7EMT
PCK2
MCF-7 MCF-7EMT
FBP1
6
20
4
2
****
**
Relative mRNA levels
Relative mRNA levels
TALDO1
5
Relative mRNA levels
Relative mRNA levels
10
2
53
15
10
****
****
5
0
0
PCK2
FBP1
b-actin
b-actin
BT-474 BT-474EMT
MCF-7 MCF-7EMT
BT-474 BT-474EMT
MCF-7 MCF-7EMT
Fig. 8. BT-474EMT and MCF-7EMT cells display decreased expression of crucial enzymes in some anabolic side pathways and gluconeogenesis. Real-time qPCR analysis
(top) and western blot analysis (bottom) of enzymes in (A) pentose phosphate pathway (G6PD, TKT, TALDO1), (B) hexosamine biosynthetic pathway (GFPT1) and serine
biosynthetic pathway (PHGDH), (C) gluconeogenesis (PCK2, FBP1) in BT-474, BT-474EMT, MCF-7 and MCF-7EMT cells. Western blots are representative of 3 independent sets
of samples. Results are presented as mean ± S.E.M. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001 (ordinary one-way ANOVA).
in parental cancer cells, and increased dramatically in EMTderived cancer cells. The expression of GLUT12 protein was
previously reported in the ERα-positive MCF-7 and T-47D cell lines,
but was not detectable in ERα-negative MDA-MB-231 and MDA-
MB-435 cell lines, suggesting that GLUT12 is specific to ERαpositive breast cancer cells [26]. In contrast, GLUT12 mRNA
expression was elevated in ERα-negative BT-474EMT and MCF-7EMT
cells. Thus, our findings indicate that ERα-independent,
54
Y. Kondaveeti et al./Cancer Letters 364 (2015) 44–58
A
JAK2
IL6
**
0.05
2.0
0.8
0.6
0.4
***
****
0.2
0.0
0.00
BT-474 BT-474EMT
**
1.5
1.0
0.5
0.0
BT-474 BT-474EMT
MCF-7 MCF-7EMT
Relative mRNA levels
****
0.15
0.10
STAT3
1.0
Relative mRNA levels
Relative mRNA levels
0.20
MCF-7 MCF-7EMT
STAT3
B
ZEB1
ZEB1
P-STAT3
1.5
1.5
1.0
****
****
0.5
****
Fold Change
Fold Change
b-actin
0.0
1.0
BT-474 BT-474EMT
****
****
0.5
60 µM
80 µM
MCF-7EMT 40 µM
C
MCT2
MCT2
*
***
0.5
****
Fold Change
Fold Change
80 µM
1.5
1.5
1.0
****
****
0.5
****
0.0
0.0
BT-474EMT 40 µM
60 µM
80 µM
MCF-7EMT 40 µM
S3I-201
D
80 µM
PCK2
5
****
Fold Change
4
***
3
60 µM
S3I-201
PCK2
5
Fold Change
60 µM
S3I-201
S3I-201
2
****
0.0
BT-474EMT 40 µM
1.0
MCF-7 MCF-7EMT
*
1
****
4
****
3
***
2
1
0
0
BT-474EMT 40 µM
60 µM
80 µM
MCF-7EMT 40 µM
S3I-201
60 µM
80 µM
S3I-201
Fig. 9. STAT3 regulates the expression of specific enzyme/transporter changes in BT-474EMT and MCF-7EMT cells. Real-time qPCR analysis of (A) pro-inflammatory cytokine,
IL6; non-receptor tyrosine kinase, JAK2; and a transcription factor, STAT3; and western blot analysis of STAT3 and phospho-STAT3 (bottom) in BT-474, BT-474EMT, MCF-7
and MCF-7EMT cells. Real-time qPCR analysis of (B) ZEB1; (C) MCT2; (D) PCK2 in BT-474EMT cells (left) and MCF-7EMT cells (right) following 24 hr treatment of S3I-201 (STAT3
inhibitor). Western blots are representative of 3 independent sets of samples. Results are presented as mean ± S.E.M. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001 (ordinary one-way ANOVA).
EMT-related signaling pathways enhance GLUT12 expression in
certain contexts.
The expression of HKs did not increase in a parallel manner to
GLUT3 and GLUT12. In fact, there was a general decrease of HK1
and HK2 in EMT-derived cancer cells (with one exception). Since
lactate production mirrored glucose uptake, we assume that the relative expression of both HKs, which was about 10-fold higher than
that of GLUT transporters, was sufficient to phosphorylate cytoplasmic glucose in order to initiate its metabolism through glycolysis
to lactate (note that competing pathways, glycogen synthesis and
Y. Kondaveeti et al./Cancer Letters 364 (2015) 44–58
the oxidative arm of the PPP, were downregulated in both EMTderived cancer cells). However, further work is needed to directly
assay the flux of glucose into glucose-6-phosphate. It will also be
worthwhile to examine whether the subcellular localization of HK1
and HK2 changes during EMT. Both HKs have been found to be associated with the outer mitochondrial membrane, which increases
their accessibility to ATP, but may also reduce their susceptibility
to allosteric inhibition by glucose-6-phosphate.
The other source of glucose for glycolysis is the breakdown of
glycogen. We observed here that the dominant glycogen synthase
isoform, GYS1, was downregulated in BT-474EMT, and not altered in
MCF-7EMT cells. In contrast, glycogen phosphorylase, PYGL, was dramatically upregulated in both EMT-derived cancer cells. A study
showed that PYGL is upregulated in response to hypoxia in different cancer cell lines, and its depletion impaired tumor growth due
to premature senescence via a ROS-dependent mechanism and
reduced input into the PPP [27]. This indicates that EMT-derived
cancer cells favor the breakdown of cellular glycogen, as opposed
to its synthesis and storage compared to parental cancer cells.
The increased glucose uptake rate in EMT-derived cancer cells
was accompanied by increased lactate production, and increased
protein levels of both LDHA and LDHB. Extracellular lactate has been
linked to various aspects of tumor progression and metastasis [28,29].
Lactate production also promotes continual glycolysis through the
regeneration of NAD+, and LDHA is the primary component of tetrameric LDH isozyme responsible for lactate production. Indeed,
LDHA is upregulated in many cancers including breast cancer and
shown to be important in tumor initiation and progression
[30,31]. In contrast, LDHB expression is absent due to promoter
hypermethylation in some forms of prostate and breast cancers
[32,33]. However, LDHB was shown to be specifically expressed at
high levels in triple-negative breast cancer compared to other subtypes of breast cancer and its expression is correlated with a poor
clinical outcome [34]. Another study reported high LDHB expression in highly glycolytic, basal-like (i.e., mesenchymal) breast cancers
and cell lines [35]. These studies and our findings collectively link
elevated LDHB expression to triple-negative and highly glycolytic
forms of breast cancer. These findings raise the questions of which
LDH tetrameric isozymes are formed in the EMT-derived cancer cells
upon LDHB induction, or whether LDHB has a non-canonical action
independently of LDHA in more aggressive forms of breast cancer.
In addition to LDH isoforms, monocarboxylate transporters, MCT2
and MCT4, were also upregulated in EMT-derived cancer cells. MCT2
is selectively expressed in colorectal cancer cell lines and knockdown of MCT2 induces mitochondrial dysfunction, ROS generation,
cell-cycle arrest and senescence [36]. MCT2 has higher affinity for
lactate and pyruvate than MCT1 [37]. The upregulation of MCT2 in
EMT-derived cancer cells may be essential for enhancing pyruvate
uptake and lactate export. On the other hand, MCT4 has lower affinity for pyruvate compared to lactate. This biochemical feature of
MCT4 is very important in avoiding loss of pyruvate from the cell
and exporting lactate. MCT4 is widely expressed in glycolytic tissues
such as white (Type IIb) skeletal muscle fibers, astrocytes and white
blood cells [37,38]. A recent study found that MCT4 is highly expressed in triple-negative breast cancers and its expression is
correlated with a poor clinical outcome [39]. These studies and our
findings show that elevated expression of MCT4 is a hallmark feature
of triple-negative breast cancer.
The expression of oxidative PPP enzyme, G6PD, was downregulated in EMT-derived cancer cells. G6PD overexpression has been
reported in numerous cancers and its activity has been linked to
increased cell proliferation [40]. In contrast to studies linking G6PD
to cancer and cell proliferation, G6PD expression was strongly suppressed by 70% in BT-474EMT cells and by 90% in MCF-7EMT cells. In
addition, TKT protein levels were decreased in both EMT-derived
cancer cells compared to parental cancer cells. Loss of G6PD and
55
downregulation of TKT in EMT-derived cancer cells would be expected to result in fewer carbons exiting glycolysis, thereby
supporting enhanced aerobic glycolysis in these cells. These data
indicate that the need for NADPH synthesis in EMT-derived cancer
cells is reduced presumably due to a reduced dependency on oxidative phosphorylation, which generates ROS.
The expression of PHGDH, which catalyzes the first irreversible reaction in SBP, consumes 3-phosphoglycerate and also NAD+,
the key cofactor that needs to be continually replenished in order
to maintain aerobic glycolysis. The PHGDH gene is either amplified or expressed at a high level in the majority of ERα-negative
breast cancers [41,42]. In contrast to these existing studies, we found
that PHGDH expression is strongly suppressed in both ERα-negative
EMT-derived cancer cells. Loss of PHDGH will cause a decrease in
the synthesis of serine in EMT-derived cancer cells. As a result, these
cells would be expected to depend on import of extracellular
serine. In agreement with this, EMT-derived cancer cells showed
upregulation of SLC38A4, a sodium-coupled neutral amino acid
transporter, which may enhance serine import (Fig. S3). Thus, the
downregulation of PHGDH is likely to promote glycolytic rate by both
decreasing the removal of carbons from glycolysis and by contributing to a more favorable NAD+/NADH ratio. Further work is needed
to examine whether PHGDH expression increases in response to decreased levels of extracellular serine, and whether the EMTderived cancer cells are sensitive to serine restriction.
The expression of hexosamine biosynthetic pathway enzyme,
GFPT1, remained unaltered in EMT-derived cancer cells compared
to parental cancer cells. This indicates that EMT-derived cancer cells
retain their ability to use glycolytic carbons (fructose-6-phosphate)
to maintain the hexosamine biosynthetic pathway and subsequent O-glycosylation and N-glycosylation reactions. Both types of
glycosylation are important for numerous signaling pathways that
have been linked to growth, EMT, and invasiveness [43,44].
Along with G6PD and PHGDH, the expressions of two key
gluconeogenic enzymes, PCK2 and FBP1, are also significantly repressed in the EMT-derived cancer cells. Loss of FBP1 expression
is shown to be a critical oncogenic event in EMT and basal-like breast
cancers [45]. This previous study also showed that loss of FBP1 inhibits oxygen consumption and ROS production by suppressing
mitochondrial complex I activity. Another recent study reported that
FBP1 is also uniformly depleted in over six hundred clear cell renal
cell carcinoma tumors [46]. In this latter study, re-expression of FBP1
was shown to diminish glycolysis and inhibit proliferation of clear
cell renal carcinoma cells. PCK2 is the mitochondrial isoform of phosphoenolpyruvate carboxykinase and almost no studies have
examined its role in cancer. However, a recent study reported that
PCK2 is a major downregulated protein in 5-fluorouracil resistant
colon cancers [47]. This report is consistent with our finding and
indicates that PCK2 suppression may be an important feature of more
aggressive cancers.
Two master regulators, c-MYC and HIF-1α, promote aerobic glycolysis by coordinated induction of GLUT1 and LDHA expressions
[48]. The findings that LDHA mRNA was upregulated only in MCF7 EMT cells, and that GLUT1 mRNA was not upregulated in BT474EMT cells indicate a minor, if any, role of c-MYC and HIF-1α in
EMT-associated metabolic changes reported herein. Since IL6/
STAT3 signaling has been shown to induce and maintain EMT
phenotype, we assessed the role of this pathway in the regulation
of EMT-associated metabolic changes. We observed here that STAT3
signaling is active in both EMT-derived cancer cells, possibly driven
by the significant upregulation of IL6 (Fig. 9A) or by constitutive activity of STAT3 [49], appears to be involved in the upregulation of
ZEB1 and MCT2, the suppression of PCK2 in both BT-474EMT cells
and MCF-7 EMT cells, and the suppression of PHGDH only in
MCF-7 EMT cells. Consistent with our findings, treatment with
oncostatin M, a member of the IL-6 family, resulted in the
56
EMT
BT-474 & MCF-7
BT-474EMT & MCF-7EMT
Glucose
Glucose
GLUT3
GLYCOGEN
GLUT12
GLYCOGEN
Glucose
GYS1
PYGB
NADP+
PYGL
NADPH
Glucose-6-P
Glucose-1-P
Glucose
GYS1
NADP+
NADPH
G6PD
G6PD
Glycolipids
Glu
Gln
Proteoglycans UDP-GlucNAc
Glycoproteins
G3P/E4P
Glycolipids
G3P/E4P
Fructose-6-P
Ribose-5-P
Nucleotides
TKT / TALDO1
GFPT1
NADPH
Glucose-6-P
Glucose-1-P
NADPH
Glu
Gln
Proteoglycans UDP-GlucNAc
Glycoproteins
G3P/E4P
G3P/E4P
Fructose-6-P
Ribose-5-P
Nucleotides
TKT / TALDO1
GFPT1
Fructose-1,6-bisP
NAD+
3-Phosphoglycerate
NADH
NAD+
Nucleotides
Serine Complex lipids
PHGDH
3-Phosphoglycerate
Proteins
NADH
PHGDH
PEP
PEP
NADH
Pyruvate
LDHA
Proteins
MCT2
NADH
NAD+
Pyruvate
Lactate
Nucleotides
Serine Complex lipids
NAD+
LDHA/B
Lactate
MCT4
Pyruvate
Pyruvate
PEP
PEP
PCK2
Oxaloacetate
PC
TCA Cycle
&
OXPHOS
Mitochondria
PCK2
Oxaloacetate
PC
TCA Cycle
&
OXPHOS
Mitochondria
Fig. 10. EMT-mediated changes in glucose metabolism. Diagram summarizing EMT-induced alterations in different pathways involved in glucose metabolism: glycolysis (shown in red arrows), gluconeogenesis (shown in
blue arrows), glycogen metabolism, oxidative and non-oxidative pentose phosphate pathways, hexosamine biosynthesis pathway and serine biosynthetic pathway (shown in green arrows). Enzymes involved in specific reactions are denoted in uppercase right next to the arrows in the color assigned to the specific pathway. Upregulated pathways are denoted with thick arrows and downregulated pathways are denoted with dotted arrows in
BT-474EMT and MCF-7EMT cells (right) compared to BT-474 and MCF-7 cells (left). Enzyme abbreviations: GYS1: glycogen synthase 1; PYGB: glycogen phosphorylase brain isoform; PYGL: glycogen phosphorylase liver isoform;
G6PD: glucose-6-phosphate dehydrogenase; TKT: transketolase; TALDO1: transaldolase 1; GFPT1: glutamine–fructose-6-phosphate transaminase 1; FBP1: fructose-1,6-bisphosphatase 1; PHGDH: phosphoglycerate dehydrogenase; LDHA/B: lactate dehydrogenase isoforms A and B; PC: pyruvate carboxylase (mitochondrial); PCK2: phosphoenolpyruvate carboxykinase 2 (mitochondrial). Metabolite abbreviations: G3P: glyceraldehyde-3-phosphate;
E4P: erythrose-4-phosphate; Gln: glutamine; Glu: glutamic acid; UDP-GlucNAc: UDP-N-acetylglucosamine; PEP: phosphoenolpyruvate. Names in boxes are general macromolecules whose synthesis relies on the product of the
anabolic side pathway shown.
Y. Kondaveeti et al./Cancer Letters 364 (2015) 44–58
FBP1
FBP1
Fructose-1,6-bisP
Y. Kondaveeti et al./Cancer Letters 364 (2015) 44–58
upregulation of ZEB1 via STAT3 activation in MCF-7 cells [50]. It is
of considerable interest to identify the constellation of signaling pathways that orchestrate all the observed changes in metabolic pathways
in EMT-derived cancer cells.
Conclusions
We have identified EMT-induced alterations related to glucose
metabolism in two independent and substantially different breast
cancer cell lines. We found notable differences in glycolysis and
glycolysis-associated pathways in EMT-derived cancer cells compared to their parental counterparts (Fig. 10). EMT-derived cancer
cells showed enhanced aerobic glycolysis and increased sensitivity to glycolytic inhibitor, 2-deoxyglucose. Importantly, this study
revealed enzymes/transporters such as GLUT3/12, LDHA/B, MCT2/4
and PYGL that are specifically upregulated in EMT-derived cancer
cells compared to parental cancer cells. These upregulated enzymes/
transporters may have potential for becoming therapeutic targets
against metastatic breast cancer. Another major finding of this study
is the loss of G6PD and PHGDH expressions in EMT-derived cancer
cells. These findings may be exploited by increasing oxidative stress,
or by modifying the diet to decrease serine to inhibit EMT-derived
cancer cell proliferation. Finally, suppression of STAT3 signaling reversed EMT-associated changes in the expression of ZEB1, MCT2,
and PCK2 in both EMT-derived cancer cell lines. This indicates that
the expression of certain enzymes/transporters in the EMT-derived
cancer cells is dependent on STAT3 signaling and can be targeted
by STAT3 inhibitors.
Authors’ contributions
YK and BAW designed the study and experiments, interpreted
the data, and wrote the manuscript. YK performed all experiments and analyzed the data. IKG generated MCF-7EMT cells using
prolonged mammosphere culture method and reviewed the manuscript critically for important intellectual content. All authors read
and approved the manuscript.
Acknowledgements
We would like to acknowledge the generous financial support
from the Department of Cell Biology and the Carole and Ray Neag
Comprehensive Cancer Center at the University of Connecticut Health
Center.
Conflict of interest
The authors declare that they have no competing interests.
Appendix: Supplementary material
Supplementary data to this article can be found online at
doi:10.1016/j.canlet.2015.04.025.
References
[1] G. Konecny, G. Pauletti, M. Pegram, M. Untch, S. Dandekar, Z. Aguilar, et al.,
Quantitative association between HER-2/neu and steroid hormone receptors
in hormone receptor-positive primary breast cancer, J. Natl. Cancer Inst. 95 (2)
(2003) 142–153.
[2] S. Cleator, W. Heller, R.C. Coombes, Triple-negative breast cancer: therapeutic
options, Lancet Oncol. 8 (3) (2007) 235–244.
[3] B. Weigelt, J.L. Peterse, L.J. van ‘t Veer, Breast cancer metastasis: markers and
models, Nat. Rev. Cancer 5 (8) (2005) 591–602.
[4] S. Valastyan, R.A. Weinberg, Tumor metastasis: molecular insights and evolving
paradigms, Cell 147 (2) (2011) 275–292.
[5] B. De Craene, G. Berx, Regulatory networks defining EMT during cancer initiation
and progression, Nat. Rev. Cancer 13 (2) (2013) 97–110.
57
[6] S. Lamouille, J. Xu, R. Derynck, Molecular mechanisms of epithelialmesenchymal transition, Nat. Rev. Mol. Cell Biol. 15 (3) (2014) 178–196.
[7] W.H. Koppenol, P.L. Bounds, C.V. Dang, Otto Warburg’s contributions to
current concepts of cancer metabolism, Nat. Rev. Cancer 11 (5) (2011) 325–
337.
[8] R.J. DeBerardinis, J.J. Lum, G. Hatzivassiliou, C.B. Thompson, The biology of
cancer: metabolic reprogramming fuels cell growth and proliferation, Cell Metab.
7 (1) (2008) 11–20.
[9] R.J. Deberardinis, N. Sayed, D. Ditsworth, C.B. Thompson, Brick by
brick: metabolism and tumor cell growth, Curr. Opin. Genet. Dev. 18 (1) (2008)
54–61.
[10] P.S. Ward, C.B. Thompson, Metabolic reprogramming: a cancer hallmark even
Warburg did not anticipate, Cancer Cell 21 (3) (2012) 297–308.
[11] I.K. Guttilla, K.N. Phoenix, X. Hong, J.S. Tirnauer, K.P. Claffey, B.A. White,
Prolonged mammosphere culture of MCF-7 cells induces an EMT and repression
of the estrogen receptor by microRNAs, Breast Cancer Res. Treat. 132 (1) (2012)
75–85.
[12] N.J. Sullivan, A.K. Sasser, A.E. Axel, F. Vesuna, V. Raman, N. Ramirez, et al.,
Interleukin-6 induces an epithelial-mesenchymal transition phenotype in
human breast cancer cells, Oncogene 28 (33) (2009) 2940–2947.
[13] G. Xie, Q. Yao, Y. Liu, S. Du, A. Liu, Z. Guo, et al., IL-6-induced epithelialmesenchymal transition promotes the generation of breast cancer stem-like
cells analogous to mammosphere cultures, Int. J. Oncol. 40 (4) (2012) 1171–
1179.
[14] H. Korkaya, G.I. Kim, A. Davis, F. Malik, N.L. Henry, S. Ithimakin, et al., Activation
of an IL6 inflammatory loop mediates trastuzumab resistance in HER2+ breast
cancer by expanding the cancer stem cell population, Mol. Cell 47 (4) (2012)
570–584.
[15] S.A. Mani, W. Guo, M.J. Liao, E.N. Eaton, A. Ayyanan, A.Y. Zhou, et al., The
epithelial-mesenchymal transition generates cells with properties of stem cells,
Cell 133 (4) (2008) 704–715.
[16] J.P. Thiery, H. Acloque, R.Y. Huang, M.A. Nieto, Epithelial-mesenchymal
transitions in development and disease, Cell 139 (5) (2009) 871–890.
[17] J.E. Visvader, Keeping abreast of the mammary epithelial hierarchy and breast
tumorigenesis, Genes Dev. 23 (22) (2009) 2563–2577.
[18] D. Ciavardelli, C. Rossi, D. Barcaroli, S. Volpe, A. Consalvo, M. Zucchelli, et al.,
Breast cancer stem cells rely on fermentative glycolysis and are sensitive to
2-deoxyglucose treatment, Cell Death Dis. 5 (2014) e1336.
[19] T. Yamamoto, Y. Seino, H. Fukumoto, G. Koh, H. Yano, N. Inagaki, et al., Overexpression of facilitative glucose transporter genes in human cancer, Biochem.
Biophys. Res. Commun. 170 (1) (1990) 223–230.
[20] R. Augustin, The protein family of glucose transport facilitators: it’s not only
about glucose after all, IUBMB Life 62 (5) (2010) 315–333.
[21] W.A. Flavahan, Q. Wu, M. Hitomi, N. Rahim, Y. Kim, A.E. Sloan, et al., Brain tumor
initiating cells adapt to restricted nutrition through preferential glucose uptake,
Nat. Neurosci. 16 (10) (2013) 1373–1382.
[22] M. Masin, J. Vazquez, S. Rossi, S. Groeneveld, N. Samson, P.C. Schwalie, et al.,
GLUT3 is induced during epithelial-mesenchymal transition and promotes
tumor cell proliferation in non-small cell lung cancer, Cancer Metab. 2 (2014)
11.
[23] S. Rogers, M.L. Macheda, S.E. Docherty, M.D. Carty, M.A. Henderson, W.C. Soeller,
et al., Identification of a novel glucose transporter-like protein-GLUT-12, Am.
J. Physiol. Endocrinol. Metab. 282 (3) (2002) E733–E738.
[24] C.A. Stuart, M.E. Howell, Y. Zhang, D. Yin, Insulin-stimulated translocation of
glucose transporter (GLUT) 12 parallels that of GLUT4 in normal muscle, J. Clin.
Endocrinol. Metab. 94 (9) (2009) 3535–3542.
[25] J. Pujol-Gimenez, J. Barrenetxe, P. Gonzalez-Muniesa, M.P. Lostao, The facilitative
glucose transporter GLUT12: what do we know and what would we like to
know?, J. Physiol. Biochem. 69 (2) (2013) 325–333.
[26] M.L. Macheda, S. Rogers, J.D. Best, Molecular and cellular regulation of
glucose transporter (GLUT) proteins in cancer, J. Cell. Physiol. 202 (3) (2005)
654–662.
[27] E. Favaro, K. Bensaad, M.G. Chong, D.A. Tennant, D.J. Ferguson, C. Snell, et al.,
Glucose utilization via glycogen phosphorylase sustains proliferation and
prevents premature senescence in cancer cells, Cell Metab. 16 (6) (2012)
751–764.
[28] F. Hirschhaeuser, U.G. Sattler, W. Mueller-Klieser, Lactate: a metabolic key player
in cancer, Cancer Res. 71 (22) (2011) 6921–6925.
[29] J.R. Doherty, J.L. Cleveland, Targeting lactate metabolism for cancer therapeutics,
J. Clin. Invest. 123 (9) (2013) 3685–3692.
[30] V.R. Fantin, J. St-Pierre, P. Leder, Attenuation of LDH-A expression uncovers a
link between glycolysis, mitochondrial physiology, and tumor maintenance,
Cancer Cell 9 (6) (2006) 425–434.
[31] Z.Y. Wang, T.Y. Loo, J.G. Shen, N. Wang, D.M. Wang, D.P. Yang, et al., LDH-A
silencing suppresses breast cancer tumorigenicity through induction of oxidative
stress mediated mitochondrial pathway apoptosis, Breast Cancer Res. Treat. 131
(3) (2012) 791–800.
[32] A. Leiblich, S.S. Cross, J.W. Catto, J.T. Phillips, H.Y. Leung, F.C. Hamdy, et al., Lactate
dehydrogenase-B is silenced by promoter hypermethylation in human prostate
cancer, Oncogene 25 (20) (2006) 2953–2960.
[33] N.J. Brown, S.E. Higham, B. Perunovic, M. Arafa, S. Balasubramanian, I. Rehman,
Lactate dehydrogenase-B is silenced by promoter methylation in a high
frequency of human breast cancers, PLoS ONE 8 (2) (2013) e57697.
[34] M.L. McCleland, A.S. Adler, Y. Shang, T. Hunsaker, T. Truong, D. Peterson, et al.,
An integrated genomic screen identifies LDHB as an essential gene for triplenegative breast cancer, Cancer Res. 72 (22) (2012) 5812–5823.
58
Y. Kondaveeti et al./Cancer Letters 364 (2015) 44–58
[35] J.B. Dennison, J.R. Molina, S. Mitra, A.M. Gonzalez-Angulo, J.M. Balko, M.G. Kuba,
et al., Lactate dehydrogenase B: a metabolic marker of response to neoadjuvant
chemotherapy in breast cancer, Clin. Cancer Res. 19 (13) (2013) 3703–3713.
[36] I. Lee, S.J. Lee, W.K. Kang, C. Park, Inhibition of monocarboxylate transporter 2
induces senescence-associated mitochondrial dysfunction and suppresses
progression of colorectal malignancies in vivo, Mol. Cancer Ther. 11 (11) (2012)
2342–2351.
[37] A.P. Halestrap, M.C. Wilson, The monocarboxylate transporter family – role and
regulation, IUBMB Life 64 (2) (2012) 109–119.
[38] A.P. Halestrap, The monocarboxylate transporter family – structure and
functional characterization, IUBMB Life 64 (1) (2012) 1–9.
[39] J. Doyen, C. Trastour, F. Ettore, I. Peyrottes, N. Toussant, J. Gal, et al., Expression
of the hypoxia-inducible monocarboxylate transporter MCT4 is increased in
triple negative breast cancer and correlates independently with clinical outcome,
Biochem. Biophys. Res. Commun. 451 (1) (2014) 54–61.
[40] P. Jiang, W. Du, X. Wang, A. Mancuso, X. Gao, M. Wu, et al., p53 regulates
biosynthesis through direct inactivation of glucose-6-phosphate dehydrogenase,
Nat. Cell Biol. 13 (3) (2011) 310–316.
[41] R. Possemato, K.M. Marks, Y.D. Shaul, M.E. Pacold, D. Kim, K. Birsoy, et al.,
Functional genomics reveal that the serine synthesis pathway is essential in
breast cancer, Nature 476 (7360) (2011) 346–350.
[42] J.W. Locasale, A.R. Grassian, T. Melman, C.A. Lyssiotis, K.R. Mattaini, A.J. Bass,
et al., Phosphoglycerate dehydrogenase diverts glycolytic flux and contributes
to oncogenesis, Nat. Genet. 43 (9) (2011) 869–874.
[43] K.E. Wellen, C.B. Thompson, A two-way street: reciprocal regulation of
metabolism and signalling, Nat. Rev. Mol. Cell Biol. 13 (4) (2012) 270–276.
[44] H.B. Ruan, J.P. Singh, M.D. Li, J. Wu, X. Yang, Cracking the O-GlcNAc code in
metabolism, Trends Endocrinol. Metab. 24 (6) (2013) 301–309.
[45] C. Dong, T. Yuan, Y. Wu, Y. Wang, T.W. Fan, S. Miriyala, et al., Loss of FBP1 by
Snail-mediated repression provides metabolic advantages in basal-like breast
cancer, Cancer Cell 23 (3) (2013) 316–331.
[46] B. Li, B. Qiu, D.S. Lee, Z.E. Walton, J.D. Ochocki, L.K. Mathew, et al., Fructose-1,6bisphosphatase opposes renal carcinoma progression, Nature 513 (7517) (2014)
251–255.
[47] J.W. Park, S.C. Kim, W.K. Kim, J.P. Hong, K.H. Kim, H.Y. Yeo, et al., Expression of
phosphoenolpyruvate carboxykinase linked to chemoradiation susceptibility
of human colon cancer cells, BMC Cancer 14 (2014) 160.
[48] M.G. Vander Heiden, L.C. Cantley, C.B. Thompson, Understanding the Warburg
effect: the metabolic requirements of cell proliferation, Science 324 (5930)
(2009) 1029–1033.
[49] M.Z. Kamran, P. Patil, R.P. Gude, Role of STAT3 in cancer metastasis and
translational advances, Biomed Res. Int. 2013 (2013) 421821.
[50] L. Guo, C. Chen, M. Shi, F. Wang, X. Chen, D. Diao, et al., Stat3-coordinated
Lin-28-let-7-HMGA2 and miR-200-ZEB1 circuits initiate and maintain oncostatin
M-driven epithelial-mesenchymal transition, Oncogene 32 (45) (2013) 5272–
5282.