Biomedicine & Pharmacotherapy 98 (2018) 173–179

Contents lists available at ScienceDirect

Biomedicine & Pharmacotherapy

journal homepage: www.elsevier.com/locate/biopha

Glucose-dependent growth arrest of leukemia cells by MCT1 inhibition: T

Feeding Warburg’s sweet tooth and blocking acid export as an anticancer
strategy
⁎
Aleksandra I. Pivovarovaa, Gordon G. MacGregorb,
a
Alabama College of Osteopathic Medicine, 445 Health Sciences Blvd, Dothan, AL 36303, United States
b
Department of Biological Sciences, University of Alabama in Huntsville, 301 Sparkman Dr, Huntsville, AL 35899, United States

A R T I C L E I N F O A B S T R A C T

Keywords: This study aims to investigate the utilization of The Warburg Effect, cancer’s “sweet tooth” and natural greed for
Warburg effect glucose to enhance the effect of monocarboxylate transporter inhibition on cellular acidification. By simulating
Monocarboxylate transporter hyperglycemia with high glucose we may increase the effectiveness of inhibition of lactate and proton export on
MCT1 the dysregulation of cell pH homeostasis causing cell death or disruption of growth in cancer cells. MCT1 and
MCT4
MCT4 expression was determined in MCF7 and K562 cell lines using RT-PCR. Cell viability, growth, intracellular
AR-C155858
pH and cell cycle analysis was measured in the cell lines grown in 5 mM and 25 mM glucose containing media in
Cariporide
Cell-cycle arrest the presence and absence of the MCT1 inhibitor AR-C155858 (1 μM) and the NHE1 inhibitor cariporide (10 μM).
Hyperglycemia The MCT1 inhibitor, AR-C155858 had minimal effect on the viability, growth and intracellular pH of MCT4
expressing MCF7 cells. AR-C155858 had no effect on the viability of the MCT1 expressing K562 cells, but de-
creased intracellular pH and cell proliferation, by a glucose-dependent mechanism. Inhibition of NHE1 on its
own had a no effect on cell growth, but together with AR-C155858 showed an additive effect on inhibition of cell
growth. In cancer cells that only express MCT1, increased glucose concentrations in the presence of an MCT1
inhibitor decreased intracellular pH and reduced cell growth by G1 phase cell-cycle arrest. Thus we propose a
transient hyperglycemic-clamp in combination with proton export inhibitors be evaluated as an adjunct to
cancer treatment in clinical studies.

1. Introduction members, from which only a few have been described as proton-linked
short-chain monocarboxylic acid transporters [6,7]. They are MCT1-
The German physiologist, Otto Warburg observed that cancer cells MCT4, and the concentration gradient of protons and mono-
utilize a higher amount of glucose compared to noncancerous cells [1]. carboxylates such as pyruvate, lactate, and ketone bodies dictates the
Not only do cancer cells consume more glucose, but the metabolic direction in which that substrate will be transported [8]. Due to the
pathway they employ to completely metabolize glucose also differs importance of MCTs in monocarboxylate and proton transport and their
from the pathway used by normal cells. Cancerous tissue and cells rely high expression levels in cancer cells [9], MCTs are now being con-
on pyruvate reduction to lactate for energy production even in the sidered in cancer prognosis [10]. The levels of MCT1 and MCT4 have
presence of oxygen [2]. This phenomenon was first described by Otto been noticeably upregulated in breast, cervical, colorectal, and gastric
Warburg, and is referred to as “the Warburg effect” [3]. To avoid a cancers, along with some glioblastomas [11]. Consequently these two
decrease in intracellular pH due to increased metabolic activity, lactate representatives of the MCT family have become important novel targets
molecules and protons are quickly and efficiently exported out of the for cancer therapy [12–15].
cell [4,5]. This function is performed by proton-linked mono- AR-C155858 is a potent MCT1 inhibitor and was first created by
carboxylate transporters, although there are multiple proton export AstraZeneca to function as an immunosuppressant to block the pro-
pathways in the plasma membrane including the sodium-hydrogen liferation of T-lymphocytes [16], but eventually the drug found its
exchanger (NHE), sodium bicarbonate transporters (NBC), chloride application in cancer treatment [17]. AR-C155858 binds to and inhibits
bicarbonate exchangers (AE) and others (Fig. 1). MCT1, but has no effect on MCT4 function up to a concentration as high
The monocarboxylate transporter (MCT) family includes 14 as 10 μM [16,17]. The viability of cells expressing MCT4 should not be

⁎
Corresponding author.
E-mail address: [email protected] (G.G. MacGregor).

https://doi.org/10.1016/j.biopha.2017.12.048
Received 20 November 2017; Received in revised form 5 December 2017; Accepted 13 December 2017
0753-3322/ © 2017 Elsevier Masson SAS. All rights reserved.
A.I. Pivovarova, G.G. MacGregor Biomedicine & Pharmacotherapy 98 (2018) 173–179

were purchased from (Life Technologies™, Carlsbad, CA). Cells were

passaged every 3–5 days by trypsinization and grown at 37 °C in a
humidified 5% CO2 incubator. K562 cells were cultured in a similar
Glucose
2 Pi + manner, except they were a suspension culture and did not need to be
2 NAD+ 2 ADP
trypsinized. Cells were passaged with a 1:3 to 1:5 split every 3–5 days.
For experiments, cells were used directly from the T-25 flasks or plated
2 NADH + 2 H+ + 2 ATP ADP + Pi + H+ on 24 well plates. The MCT1 inhibitor AR-C155858 and the NHE1 in-
Cell hibitor cariporide (obtained from Tocris Biosciences, Minneapolis, MN)
2 Pyruvate Work were dissolved as a stock solutions of either 1 mM or 10 mM in DMSO
[H+] (ATCC, Manassas, VA) and stored at −20 °C. The AR-C155858 and
2 NAD+
H+ cariporide were made up at a working concentrations of 1 μM and
2 Lactate H+ Na+ H+ 10 μM respectively which would expose the cells to a final DMSO
Lactate H+ concentration of 0.1 or 0.2% DMSO.

2.2. RNA isolation and cDNA generation and RT-PCR analysis

Fig. 1. Schematic diagram of Warburg metabolism and proton export pathways in a
cancer cell. Cancer cells rely heavily on glycolysis for energy production, even in the An appropriate number of cells, usually 5 × 106–1 × 107 cells were
presence of oxygen and produce lactate as a byproduct, this was first discovered by Otto
taken from suspension culture or trypsinized off a T-25 flask, pelleted
Warburg and is known as The Warburg Effect. In the case of Warburg metabolism, me-
tabolic acidosis results when the rate of ATP hydrolysis from cell work, exceeds the rate at by centrifugation at 1000 × g for 5 min and the supernatant discarded.
which ATP is produced via glycolysis. Hence, the biochemical mechanism of proton ac- Essentially the RNA isolation protocol was similar to the Qiagen
cumulation is not lactate production, but ATP hydrolysis. As the rate of ATP hydrolysis RNeasy® kit instructions (catalog No. 74104). To make cDNA, about 1
exceeds all other reactions, the rate of proton release eventually exceeds metabolic proton μg of total cellular RNA was thawed on ice, and cDNA according to the
buffering and protons will accumulate inside the cell and cytosolic acidification results. QuantiTect Reverse Transcription Kit protocol (Qiagen Catalogue No.
Tumor cells prevent intracellular acidification by having multiple proton export pathways
205311). The cDNA was then quantified using the Nanodrop spectro-
including the Na+/H+ exchanger 1 (NHE1), the vacuolar H+-ATPase, and the proton
coupled monocarboxylate transporters (MCT1 and MCT4). We will focus on the mono-
photometer and frozen at −20 °C until use for real-time PCR analysis.
carboxylate transporters MCT1 and MCT4 due to their high expression and upregulation Real-time PCR was performed using a 7500 Fast Real-time PCR system
in many cancers, and potential as an anti-cancer pharmaceutical target. The concept for (Life technologies™, Carlsbad, CA). RT-PCR was set up in a 96-well
this figure was unashamedly taken from Robergs et al. 2004 [38]. plate in a reaction volume of 25 μL per well 2x Fast SYBR® Green DNA
polymerase (Catalog number 4385617). All MCT oligos were purchased
affected by the MCT1 inhibitor AR-C155858, due to the presence of the from Qiagen Inc. (Valencia, CA). SDHA oligos were synthesized by
additional lactate and proton coupled transporter, MCT4. However cells Eurofins MWG Operon LLC (Huntsville, AL), SDHA_F: TCTGCACTCTG
that only express MCT1 (such as K562 cells) would fail to export pro- GGGAAGAAG and SDHA_R: CAAGAATGAAGCAAGGGACA. The PCR
tons in the presence of AR-C155858 leading to an intracellular acid- efficiency was determined for the target genes MCT1, MCT4 and the
ification. reference gene SDHA by PCR analysis of serial dilutions of cDNA. The
The current theories on targeting the Warburg Effect of cancer threshold crossing point values (CT) were linearly correlated with the
usually refer to exploiting their reliance on glycolysis [18] and tar- logarithmic value of the DNA amount. The slope of this line provided
geting one of the several enzymes involved glucose metabolism. Our the PCR efficiency number for the gene under the given parameters
theory is the exact opposite, we are proposing to let cancer’s sweet (primers used and PCR-protocol).
tooth and greed for sugar become its Achilles’ heel. By increasing the To determine the fold change of MCT1 and MCT4 expression over
amount of available glucose, we are promoting its uptake and utiliza- the reference gene SDHA we used the quantification method and
tion by cancer cells (due to reliance on the Warburg effect), causing an guidelines of Pfaffl 2001 [19]. The efficiencies of all three genes ex-
increase in their metabolic rate and a rise in generated lactate and amined in this study were close to the ideal efficiency of two, and we
protons. These protons need to be extruded from the cell to maintain a simplified our analysis and assigned Etarget a value of two. The threshold
physiological intracellular pH. We propose to supplement glucose to crossing RT-PCR cycle (Ct) for reference and target genes were obtained
hyperglycemic levels followed by inhibition of lactate and proton from the RT-PCR cycler software. The reference gene SDHA showed
transport with the MCT1 inhibitor AR-C155858. We hypothesize these constant expression over different glucose concentrations, therefore no
cells will show an increased intracellular acidification with subsequent normalization was needed and a simplified equation could be used,
cell distress and deleterious effects such as inhibition of growth or Fold change over SDHA = (Etarget)ΔCt(ref−target).
death.
2.3. K562 and MCF7 (GFP) cell counting, viability assay and cell cycle
assay
2. Materials and methods
MCF7 (GFP) and K562 cells were cultured in media containing
2.1. Cell culture different concentrations of glucose with or without MCT1 inhibitors.
Cells were then isolated for counting, viability and cell cycle assays
The MCF7 (GFP) human breast cancer cell lines was obtained from using a Tali® cell counter (Life technologies™, Carlsbad, CA). Cell cycle
Cell Biolabs, INC (San Diego, CA). The non-fluorescent MCF7 and K562 progression was measured by quantification of cellular DNA content. As
cells, the human myelogenous leukemia cell line were obtained from cells progress through the cell cycle, the amount of DNA doubles. This
American Type Culture Collection, Manassas, VA, and were a kind gift doubling can be tracked and used to determine the cell cycle phase (G1,
of Dr. Eric Mendenhall (University of Alabama in Huntsville). All cell S, and G2/M). This procedure was performed using manufacturer in-
lines were cultured in Dulbecco's Modified Eagle Medium (DMEM), structions (Tali® Cell Cycle Kit - Catalog no. A10798). The cell number/
containing no phenol red, no glucose and no pyruvate. DMEM was fluorescence data from the Tali® was analyzed with FCS Version 5 (De
supplemented with 10% fetal calf serum, 100 Units/ml of penicillin and Novo, Glendale, CA) and Multicycle AV (Phoenix Flow Systems, Inc.,
100 μg/mL of streptomycin solution, glutamine (2 mM), non-essential San Diego, CA) using the mathematical fitting models of Kallioniemi
amino acids (1x) and the appropriate amount of glucose to give either et al. [20]. The proportion of cells in the G1, S and G2 phases was
5 mM or 25 mM final glucose concentrations. All cell culture reagents determined and plotted.

174
A.I. Pivovarova, G.G. MacGregor Biomedicine & Pharmacotherapy 98 (2018) 173–179

2.4. Intracellular pH measurements 3.2. The MCT1 inhibitor AR-C155858 did not alter the viability of MCF7
cells, but had a small effect on cell number
Cells were removed from culture and resuspended in physiological
experimental solution (145 mM NaCl, 4.5 mM KCl, 1 mM CaCl2, 1 mM Inhibition of the MCT1 transporter with 1 μM AR-C155858 did not
MgCl2, 5 mM glucose, 20 mM HEPES, pH 7.4. Cells were loaded with produce a change in cell viability in MCF7 cells grown either in 5 mM or
10 μM BCECF-AM (Tocris Biosciences, Minneapolis, MN) and left for 25 mM glucose containing media (F (3, 20) = 1.57, P = 0.2278), data
15 min in the dark. Cells were then aliquoted, centrifuged to form a cell not shown. The number of cells grown in the culture media containing
pellet and resuspended into either calibration solutions or experimental 5 mM glucose did not change with the addition of 1 μM AR-C155858
solutions. The BCECF fluorescence was calibrated with valinomycin/ (Fig. 2C) and was 118.1 ± 9.1% (n = 6, P = 0.075) of control. When
nigericin 10 μM (Molecular Probes, Carlsbad, CA) in buffers of pH 5.5, the media glucose concentration was increased to 25 mM, addition of
pH 6.5 or pH 7.5. For investigational measurements cells were re- the MCT1 inhibitor 1 μM AR-C155858 decreased cell growth slightly to
suspended in physiological experimental solutions with additions of 86.1 ± 4.3% of control (n = 6, P = 0.0083).
25 mM glucose and combinations of 1 μM AR-C155858 and 10 μM
cariporide. Iso-osmolarity was maintained with the addition of 20 mM 3.3. The MCT1 inhibitor AR-C155858 showed a small increase in K562 cell
mannitol. Cell loaded with BCECF were excited at 488 nm and emission viability and decreased K652 cell number by a glucose enhanced mechanism
was recorded at 500–535 nm on a Zeiss LSM 500 confocal microscope.
The fluorescence intensity of ten randomly selected cells was averaged There was a small increase in K562 cell viability in both glucose
for each data point [21]. solutions when cells were grown in the presence of 1 μM AR-C155858
(data not shown). The number of cells grown in culture media sup-
plemented with 5 mM glucose in the presence of the MCT1 inhibitor
2.5. Statistical analysis and graph plotting decreased to 77.1 ± 2.6% (n = 6) from control, a change of
22.9 ± 2.6% (P < 0.0001). When the glucose concentration was in-
For the RT-PCR experiments, biological replicates were taken one to creased to 25 mM (Fig. 2D) addition of the MCT1 inhibitor 1 μM AR-
three samples per week, one to four weeks apart. For the viability, cell- C155858 reduced cell growth even more to 57.9 ± 2.2% (n = 6), a
cycle and AR-C155858 inhibitor studies, cells were defrosted and pas- decrease of 42.1 ± 2.2% (P < 0.0001) from control.
saged a minimum of three times in the different glucose concentration
media. For the BCECF intracellular pH fluorescence experiments, bio- 3.4. Inhibition of the MCT1 transporter with AR-C155858 caused a
logical replicates were from the subsequent passage of multiple dif- glucose-dependent intracellular acidification in K562 cells
ferent cell cultures. All data were plotted and statistical analysis per-
formed using Prizm 7.03 graph suite of software (GraphPad Software, Addition of the MCT1 inhibitor AR-C155858 had no effect on the
Inc., La Jolla, CA). All data are shown as mean ± SEM, with the intracellular pH of MCF7 cells (Fig. 3A). The resting pH of MCF7 cells
number of experiments, n, shown in parenthesis. Statistical analysis was bathed in 5 mM glucose containing solution was 6.93 ± 0.16 (n = 5)
performed using Student’s two-tailed t-test for unpaired samples or one- which did not change and was 6.85 ± 0.14 (P = 0.697) 30 min after
way ANOVA, if the calculated P-value was below the threshold chosen the addition of the MCT1 inhibitor, AR-C155858 (1 μM). Bathing the
for statistical significance (P < 0.05), then the two measured phe- MCF7 cells in 25 mM glucose solution with or without 1 μM AR-
nomena are considered different. C155858, also did not alter the resting pH from the 5 mM glucose bath
solution (F (3, 16) = 0.1281, P = 0.9420).
K562 cells bathed in 5 mM glucose solution had a resting in-
3. Results tracellular pH of 7.01 ± 0.14 which acidified to 6.24 ± 0.17
(P = 0.0085, n = 5) after addition of the MCT1 inhibitor 1 μM AR-
3.1. The quantification and glucose dependence of monocarboxylate C155858 (Fig. 3B). In high-glucose (25 mM) solution, inhibition of
transporter expression in the MCF7 and K562 cell lines MCT1 caused the intracellular pH to acidify even more (Fig. 3B, E) to
5.79 ± 0.01 (P = 0.0002, n = 5). We hypothesize that this in-
The expression level of mRNA for monocarboxylate transporters, tracellular acidification in the K562 cells causes cell-cycle arrest and
MCT1 and MCT4 was measured in the weakly invasive, estrogen posi- inhibits cell growth (Fig. 3E and F).
tive breast cancer cell line, MCF7, and in the myelogenous leukemia
line, K562, using real-time PCR (RT-PCR). 3.5. Inhibition of MCT1 had no effect on the cell cycle populations of MCF7
MCF7 cells grown in 5 mM glucose showed 0.02 ± 0.01 (n = 5) cells but produced a cell-cycle arrest in K562 cells
fold less expression of MCT1 than the reference gene, succinate dehy-
drogenase complex, subunit A (SDHA). The values did not change sig- Cell-cycle analysis was only performed in 25 mM glucose containing
nificantly when MCF7 cells were grown in 25 mM glucose, 0.03 ± 0.01 media, as it showed the greatest change in cell number (Fig. 2D), and
(n = 5, P = 0.810). MCT4 levels were higher in MCF7 cells, being hence should demonstrate the biggest change in the cell-cycle phase
19.50 ± 3.56 (n = 5) fold higher than the reference SDHA expression, populations. In MCF7 cells, the percentage of cells in G1 phase was
which did not change upon increasing glucose concentration to 25 mM, high, being about 76.97 ± 1.6% (n = 6), in the control cells, which
21.98 ± 1.86 (n = 5, P = 0.553). In our hands MCF7 cells pre- decreased to 65.18 ± 1.40% (n = 6), in the cells grown in the presence
dominantly only express MCT4 (Fig. 2A). of 1 μM AR-C155858 (Fig. 3C). There was no difference in the popu-
K562 cells grown in 5 mM glucose showed 42.9 ± 14.0 (n = 5) fold lation of cells in the S or G2 phase grown with and without the in-
more expression of MCT1 than the reference gene SDHA (Fig. 2B). This hibitor. Hence, MCT1 inhibition in MCF7 cells does not cause G1 cell-
increased to 108.3 ± 25.4 (n = 4, P = 0.0482) fold over the reference cycle arrest.
gene SDHA when the media glucose concentration was increased to In K562 cells, the percentage of cells in G1 phase was
25 mM glucose. MCT4 levels were not dependent on glucose con- 26.46 ± 2.60% (n = 6) in the control cells, which increased to
centration, and were 1.3 ± 1.0 (n = 5) times higher than the reference 39.29 ± 1.29% (n = 6) in the G1 phase of cells grown in the presence
gene SDHA in 5 mM glucose concentration and 1.3 ± 0.4 (n = 4, of 1 μM AR-C155858. This was an increase of 12.83 ± 2.90%
P = 0.9746) higher than SDHA gene expression in 25 mM glucose (P = 0.0013) of the total cell population (Fig. 3D). There was no dif-
containing media (Fig. 2B). Therefore K562 cells predominantly express ference in the S phase of cells grown with and without 1 μM AR-
MCT1. C155858, 36.56 ± 1.95% (n = 6), versus 36.51 ± 3.75% (n = 6),

175
A.I. Pivovarova, G.G. MacGregor Biomedicine & Pharmacotherapy 98 (2018) 173–179

Fig. 2. Expression levels of MCT1 and MCT4 and

effect of MCT1 inhibition on MCF7 and K562 cell
growth in culture. (A) MCF7 cells express about 975
times more MCT4 than MCT1 and expression levels
were not altered by glucose concentration. (B) K562
cells express about 33 fold more MCT1 than MCT4.
MCT1 values are plotted on the left axis while MCT4
values are plotted on the right axis. The level of
MCT1 expression was dependent on glucose and in-
creased 2.5 fold when the media glucose concentra-
tion was increased to 25 mM glucose (n = 4). MCT4
levels were not altered by glucose concentration. (C)
There was no change in MCF7 cell number when
cells were grown in 5mM glucose containing media
in the presence of 1 μM AR-C155858. There was a
small decrease in MCF7 cell number when cells were
grown in 25 mM glucose containing media in the
presence of 1 μM AR-C155858. (D) The number of
K562 cells grown in 5 mM glucose containing 1 μM
AR-C155858 media showed a decrease of
22.8 ± 2.6% (P = 0.0001) over control, and a de-
crease of 42.1 ± 2.2% (P = 0.0001) over control
when grown in the presence of 25 mM glucose con-
taining media.

respectively (P = 0.9908, ns). K562 cells in grown in the presence 1 μM 4. Discussion

AR-C155858 showed less cells in the G2 phase, decreasing from
36.98 ± 1.25% (n = 6), in control to 24.21 ± 3.39% (n = 6), in cells Targeting cancer by means of its abnormal physiology and meta-
grown with the MCT1 inhibitor (Fig. 3D). bolism has become a powerful strategy for cancer treatment. The in-
creased glucose dependence and its utilization by cancer cells creates
many valid targets for small molecule inhibitors. There are many
3.6. NHE1 plays a minimal role in proton extrusion from the high metabolic strategies for targeting this abnormal physiology of cancer [18,22].
proton load generated by Warburg metabolism However these “Targeting the Warburg Effect” strategies usually re-
volve around interfering with glycolysis, with a concomitant decrease
To determine the contribution of the sodium-hydrogen exchanger in ATP synthesis and cellular starvation [23–25].
(NHE1) in dissipating this metabolically produced high proton load, we However, our theory of targeting the Warburg effect as a cancer
measured the effect of the high affinity, NHE1 specific inhibitor, car- treatment, is the exact opposite. By targeting cancer’s sweet tooth,
iporide on K562 cell intracellular pH (Fig. 4A). The resting intracellular providing more glucose and stimulating glycolysis, we have shown that
pH of K562 cells bathed in high 25 mM glucose was pH 7.01 ± 0.06 inhibition of a proton export pathway (MCT1 in this case) can amplify
(n = 4). This decreased to pH 6.78 ± 0.04 (n = 4, P = 0.0197) after the intracellular acidosis produced by MCT1 inhibition and induce a
30 min in the presence of 10 μM cariporide. In the combined presence cell cycle arrest [26,27], slowing the cancer growth and potentially
of the NHE1 inhibitor and the MCT1 blocker (1 μM AR-C155858) the making it more sensitive to radiation [28,29], other chemotherapeutic
intracellular pH decreased further to pH 5.84 ± 0.08 (n = 4, drugs [30,31] or hyperthermia [32].
P < 0.0001). Growing cells with 10 μM cariporide had no effect on It has previously been shown that MCT1 inhibition by AR-C155858
K562 cell numbers (Fig. 4B and C), as after three days in culture with in Ras-transformed fibroblasts prevented cell growth that could be re-
10 μM cariporide the cell number was 96.2 ± 6.0% of control lieved by expression of an alternative proton export pathway MCT4
(P = 0.5498, n = 4). On the other hand culture with both the NHE1 [33]. A previous study has also shown that inhibition of the NHE1 in
and MCT1 inhibitors resulted in growth arrest and a 60% reduction in leukemia cells with an amiloride analogue acidifies the cell resulting in
cell growth (Fig. 4B and C) with the cell count at 39.6 ± 1.7% of cell-cycle arrest in the G1/S phase and that there was a critical pH of
control (P < 0.0001, n = 4). Cariporide had no effect on the growth of about 6.8 that is necessary for cell cycle arrest to occur [27]. Addition
MCF7 cells (data not shown) with cells cultured in 10 μM cariporide of two proton transport inhibitors including cariporide to MCF7 cells
counted at 102.3 ± 2.8% (n = 4, P = 0.4386) of control and cells for 24 hrs resulted in a small decrease in intracellular pH (about
grown in both 10 μM cariporide and 1 μM AR-C155858 showing no 0.18 pH units) and but no effect on cell growth over seven days in
change and measured at 98.6 ± 3.4% of control value (n = 4, culture [34]. These observations agree with our experimental results in
P = 0.6971). The high expression of MCT4 in MCF7 cells is sufficient to that small changes of intracellular pH (∼ 0.2 pH units) brought about
export excess protons and regulate intracellular pH in the presence of by inhibition of NHE1 have no effect of cell growth while larger drops
NHE1 and MCT1 inhibitors (Fig. 4C). K562 cells grown in the presence in intracellular pH that cross the pH 6.8 threshold will cause cell cycle
of both 10 μM cariporide and 1 μM AR-C155858 show a large arrest in arrest and inhibit cell growth. Proton transport inhibitors may only
cell growth which is almost exclusively produced by inhibiting proton have a small effect on cell resting pH due to highly redundant (Na+-H+,
efflux through MCT1 transporters (Fig. 4C, bottom panel). Na+-HCO3, MCT1–4) proton export pathways. However, proton
transport inhibitors may have a much larger effect on the ability of the
cell to maintain intracellular pH when challenged with increased

176
A.I. Pivovarova, G.G. MacGregor Biomedicine & Pharmacotherapy 98 (2018) 173–179

Fig. 3. Inhibition of the MCT1 transporter with AR-C155858 caused a glucose-dependent intracellular acidification and induces a G1 cell cycle arrest in MCT1-expressing K562 cells. (A)
Addition of the monocarboxylate transporter 1 inhibitor, AR-C155858 (1 μM) had no effect on the intracellular pH of MCF7 cells in either 5 mM or 25 mM glucose containing media (F (3,
16), P = 0.1281). (B) 1 μM AR-C155858 causes an intracellular acidification in K562 cells. (C) There is a small decrease in the percentage of MCF7 cells in G1 phase was in the presence of
1 μM AR-C155858. There was no difference in the S or G2 phases of cells grown with and without 1 μM AR-C155858 (n = 6). (D) The MCT1 inhibitor AR-C155858 induced G1 phase cell
cycle arrest in K652 cells. (E) An example of K562 cells fluorescence loaded with BCECF after 30 min in 25 mM Glucose solution (left panel). K562 cells measured 30 min after the addition
of the MCT1 inhibitor show decreased BCECF fluorescence (Right panel). (F) A schematic representation of the cell cycle showing G1 phase, S phase, G2, where the DNA has doubled and
M phase where mitosis occurs. We hypothesize that this dramatic intracellular acidification causes cell-cycle arrest in the G1 phase and inhibits cell growth.

production of protons from cell work metabolic activity caused by a alternative pathway for lactate and protons to leave the cell. In K562
five-fold increase in the supply of glucose for glycolysis. cells, the MCT1 inhibitor produced no change in viability, but a 22%
In this study, the application of the MCT1 inhibitor AR-C155858 decrease in cell number after growth in 5 mM glucose containing
had no effect on MCF7 viability, cell number or intracellular pH in media. The inhibition of cell growth was glucose-dependent and in-
normal 5 mM (90 mg/dl) glucose or increased 25 mM glucose media (to creased to about 42% in K562 cells incubated in an increased 25 mM
mimic hyperglycemia 450 mg/dl). This absence of an effect of the glucose containing media when exposed to the MCT1 inhibitor AR-
MCT1 inhibitor can be explained by the large expression of the AR- C155858. Such a significant decrease in cell growth can be explained by
C155858 insensitive MCT4 transporter in MCF7 cells, providing an the majority of the lactate efflux being carried out by MCT1 in K562

177
A.I. Pivovarova, G.G. MacGregor Biomedicine & Pharmacotherapy 98 (2018) 173–179

Fig. 4. NHE1 has a minimum role in proton efflux

during high metabolic proton production and no ef-
fect on K562 cell growth. (A) The resting pH K562
cells decreased slightly upon addition of 10 μM car-
iporide. In the combined presence of the NHE1 in-
hibitor and the MCT1 blocker (1 μM AR-C155858)
the intracellular pH decreased further to pH
5.84 ± 0.08 (n = 4, P < 0.0001). (B) After three
days in culture with 10 μM cariporide the cell
number was 96.2 ± 6.0% of control (P = 0.5498,
n = 4). Culture with both the NHE1 and MCT1 in-
hibitors resulted in growth arrest and the cell count
at 39.7 ± 1.6% of control (P < 0.0001, n = 4). (C)
Top panel. Here we show a schematic cell diagram in
the presence of cariporide which will inhibit the
NHE1 transporter. The cell is still able to export
much of the proton load through the MCT1 trans-
porter and maintain a moderately high pH and grow
as normal. Bottom panel. In the presence of the
NHE1 and MCT1 inhibitors, the cell metabolism
produced proton load is prevented from leaving the
cell, causing intracellular acidification and growth
arrest. This effect is predominantly due to the MCT1
activity inhibition.

cells, and the lack of expression of MCT4. Essentially, a glucose-sti- the ones expressing MCT4, if the MCT4 was also blocked from per-
mulated intracellular acidification takes place in MCT1 expressing cells. forming its function. The development of MCT4 inhibitors is currently
This acidification induced a cell-cycle arrest in the G1 phase, reducing in early stage pre-clinical discovery. Furthermore, the increase of glu-
cell division and growth but not altering cell viability. Inhibition of the cose in culture media demonstrates the ability to stimulate glycolysis,
sodium-hydrogen exchanger NHE1 on its own had a small effect thereby inducing the state of acidosis through intracellular accumula-
(0.23 pH units) on intracellular resting pH but had no effect on cell tion of protons generated by increased production and utilization of
growth. The dominant membrane proton transport pathway in cells ATP. Hence, the combination of MCT inhibitors with an induced hy-
under a high glycolysis generated metabolic proton load appears to be perglycemic state is an efficient strategy of cancer treatment. It is not
mediated through the lactate transporters and not the sodium-hydrogen uncommon for cancer cells to acquire drug resistance in response to
exchanger. However, it appears that silencing the NHE1 and MCT1 therapy. However, with a major reliance on glycolysis it is highly un-
pathways produces an additive effect on cell growth (compare Fig. 2D likely that cancer cells will be able to reduce their dependence on
to Fig. 4B). glycolysis, making the inhibition of pH regulators an even more at-
Although this study concentrated on two particular proton-coupled tractive solution.
lactate transporters (MCT1 and MCT4), the range of transporters reg-
ulating proton efflux and maintaining intracellular pH is quite wide 5. Conclusions
(Fig. 1.). These include the Na+/H+ exchanger, Na+/HCO3 co-trans-
porter, the plasma membrane proton pump ATPase (V-ATPase), car- This current study extends the knowledge of the effects of MCT1
bonic anhydrases, anion exchangers and others [35,36]. Each one of inhibition on cancer cells by combining it with an induced hypergly-
these proton efflux transporters can become a potential drug target, cemic event, causing an amplified intracellular acidification and cell
alone or in combination. Furthermore, most of these pH regulatory cycle arrest. We have shown that our “feeding the Warburg effect”
transporters have FDA approved pharmacological inhibitors on the strategy induced cell growth arrest in the lab and we propose that a
market. Besides considering the use of proton efflux inhibitors alone, transient hyperglycemic clamp, in combination with MCT inhibitors
the application of these inhibitors could be expanded to combinations and other proton efflux inhibitors should be investigated in the clinic.
that target other pathways.
Recently, an immunofluorescence (IF) method for detection of Acknowledgements
MCT1 and MCT4 levels in circulating tumor cells has been developed,
emphasizing the importance of evaluating MCT1 levels prior to, as well We would like to thank Dr. Sireesh Appajosyula, PharmD (Raleigh,
as after the treatment with the MCT1 inhibitor AZD3965 (an analogue NC) and Dr. Mark Rosenberg, MD (Boca Raton, FL) for the idea of using
of AR-C155858) to assess the responsiveness of cancer to the drug [37]. hyperglycemia and monocarboxylate transporter inhibition as a po-
MCT4 expression levels are taken into consideration as a biomarker for tential cancer treatment strategy. This research did not receive any
resistance to AZD3965 treatment in cells that retain the ability to export specific grant from funding agencies in the public, commercial, or not-
lactate [37]. Undoubtedly, both transporters appear to have great po- for-profit sectors.
tential in cancer treatment and its prognosis. Although MCT4 expres-
sion levels were proposed to be considered as a biomarker for re- References
sistance, a more proactive approach would be to inhibit MCT4, along
with MCT1 for a greater degree of elimination of a lactate and a proton [1] O. Warburg, F. Wind, E. Negelein, The metabolism of tumors in the body, J. Gen.
efflux. Considering our results of inhibition of MCT1 in K562 cells, the Physiol. 8 (1927) 519–530.
[2] O. Warburg, On the origin of cancer cells, Science 123 (1956) 309–314.
same results could be achieved in almost any cancerous cell line, even

178
A.I. Pivovarova, G.G. MacGregor Biomedicine & Pharmacotherapy 98 (2018) 173–179

[3] M.G. Vander Heiden, L.C. Cantley, C.B. Thompson, Understanding the Warburg (1994) 250–255.
effect: the metabolic requirements of cell proliferation, Science 324 (2009) [21] T.J. Rink, R.Y. Tsien, T. Pozzan, Cytoplasmic pH and free Mg2+ in lymphocytes, J.
1029–1033. Cell. Biol. 95 (1982) 189–196.
[4] A.P. Halestrap, N.T. Price, The proton-linked Monocarboxylate Transporter (MCT) [22] J.W. Kim, C.V. Dang, Cancer’s molecular sweet tooth and the Warburg effect,
family: structure, function and regulation, Biochem. J. 343 (1999) 281–299. Cancer Res. 66 (2006) 8927–8930.
[5] S.K. Parks, J. Chiche, J. Pouysségur, Disrupting proton dynamics and energy me- [23] J.R. Doherty, J.L. Cleveland, Targeting lactate metabolism for cancer therapeutics,
tabolism for cancer therapy, Nat. Rev. Cancer 13 (2013) 611–623. J. Clin. Invest. 123 (2013) 3685–3692.
[6] N. Draoui, O. Schicke, E. Seront, C. Bouzin, P. Sonveaux, O. Riant, O. Feron, [24] U.E. Martinez-Outschoorn, M. Peiris-Pagés, R.G. Pestell, F. Sotgia, M.P. Lisanti,
Antitumor activity of 7-aminocarboxycoumarin derivatives, a new class of potent Cancer metabolism: a therapeutic perspective, Nat. Rev. Clin. Oncol. 14 (2) (2017)
inhibitors of lactate influx but not efflux, Mol. Cancer Ther. 13 (2014) 1410–1418. 113, http://dx.doi.org/10.1038/nrclinonc.2017.1.
[7] A.P. Halestrap, D. Meredith, The SLC16 gene family – from Monocarboxylate [25] Y. Zhang, J.M. Yang, Altered energy metabolism in cancer: a unique opportunity for
(MCTs) to aromatic amino acid transporters and beyond, Pflügers Archiv. 447 therapeutic intervention, Cancer Biol. Ther. 14 (2013) 81–89.
(2004) 619–628. [26] J. Pouysségur, A. Franchi, G. L'Allemain, S. Paris, Cytoplasmic pH, a key determi-
[8] A.P. Halestrap, M.C. Wilson, The monocarboxylate transporter family-role and nant of growth factor-induced DNA synthesis in quiescent fibroblasts, FEBS Lett.
regulation, IUBMB Life 64 (2012) 109–119. 190 (1985) 115–119.
[9] R. Hussien, G.A. Brooks, Mitochondrial and plasma membrane lactate transporter [27] I.N. Rich, D. Worthington-White, O.A. Garden, P. Musk, Apoptosis of leukemic cells
and lactate dehydrogenase isoform expression in breast cancer cell lines, Physiol. accompanies reduction in intracellular pH after targeted inhibition of the Na+/H+
Genomics 43 (2011) 255–264. exchanger, Blood 95 (2000) 1427–1434.
[10] M. Eilertsen, S. Andersen, S. Al-Saad, Y. Kiselev, T. Donnem, H. Stenvold, [28] B.M. Bola, A.L. Chadwick, F. Michopoulos, K.G. Blount, B.A. Telfer, K.J. Williams,
I. Pettersen, K. Al-Shibli, E. Richardsen, L.T. Busund, R.M. Bremnes, P.D. Smith, S.E. Critchlow, L.J. Stratford, Inhibition of Monocarboxylate trans-
Monocarboxylate transporters 1–4 in NSCLC: MCT1 is an independent prognostic porter-1 (MCT1) by AZD3965 enhances radiosensitivity by reducing lactate trans-
marker for survival, PLoS One 9 (9) (2014) e105038, http://dx.doi.org/10.1371/ port, Mol. Cancer Ther. 13 (2014) 2805–2816.
journal.pone.0105038. [29] P. Sonveaux, F. Végran, T. Schroeder, M.C. Wergin, J. Verrax, Z.N. Rabbani, C.J. De
[11] C. Pinheiro, A. Longatto-Filho, J. Azevedo-Silva, M. Casal, F.C. Schmitt, F. Baltazar, Saedeleer, K.M. Kennedy, C. Diepart, B.F. Jordan, M.J. Kelley, B. Gallez, M.L. Wahl,
Role of monocarboxylate transporters in human cancers: state of the art, J. O. Feron, M.W. Dewhirst, Targeting lactate-fueled respiration selectively kills hy-
Bioenerg. Biomembr. 44 (2012) 127–139. poxic tumor cells in mice, J. Clin. Invest. 118 (2008) 3930–3942.
[12] I. Marchiq, J. Pouysségur, Hypoxia, cancer metabolism and the therapeutic benefit [30] R. Amorim, C. Pinheiro, V. Miranda-Gonçalves, H. Pereira, M.P. Moyer, A. Preto,
of targeting lactate/H++ symporters, J. Mol. Med. (Berl.) 94 (2016) 155–171. F. Baltazar, Monocarboxylate transport inhibition potentiates the cytotoxic effect of
[13] V. Miranda-Gonçalves, M. Honavar, C. Pinheiro, O. Martinho, M.M. Pires, 5-fluorouracil in colorectal cancer cells, Cancer Lett. 365 (2015) 68–78.
C. Pinheiro, M. Cordeiro, G. Bebiano, P. Costa, I. Palmeirim, R.M. Resi, F. Baltazar, [31] Z. Zhao, M.S. Wu, C. Zou, Q. Tang, J. Lu, D. Liu, Y. Wu, J. Yin, X. Xie, J. Shen,
Monocarboxylate transporters (MCTs) in gliomas: expression and exploitation as T. Kang, J. Wang, Downregulation of MCT1 inhibits tumor growth, metastasis and
therapeutic targets, Neuro. Oncol. 15 (2013) 172–188. enhances chemotherapeutic efficacy in osteosarcoma through regulation of the NF-
[14] C. Pinheiro, A. Longatto-Filho, L. Ferreira, S.M. Pereira, D. Etlinger, M.A. Moreira, κB pathway, Cancer Lett. 342 (2014) 150–158.
L.F. Jubé, G.S. Queiroz, F. Schmitt, F. Baltazar, Increasing expression of mono- [32] R.A. Coss, C.W. Storck, C. Daskalakis, D. Berd, M.L. Wahl, Intracellular acidification
carboxylate transporters 1 and 4 along progression to invasive cervical carcinoma, abrogates the heat shock response and compromises survival of human melanoma
Int. J. Gynecol. Pathol. 27 (2008) 568–574. cells, Mol. Cancer Ther. 2 (2003) 383–388.
[15] C. Pinheiro, A. Albergaria, J. Paredes, B. Sousa, R. Dufloth, D. Vieira, F. Schmitt, [33] R. Le Floch, J. Chiche, I. Marchiq, T. Naiken, K. Ilc, C.M. Murray, S.E. Critchlow,
F. Baltazar, Monocarboxylate transporter 1 is up-regulated in basal-like breast D. Roux, M.P. Simon, J. Pouysségur, CD147 subunit of lactate/H++ symporters
carcinoma, Histopathology 56 (2010) 860–867. MCT1 and hypoxia-inducible MCT4 is critical for energetics and growth of glyco-
[16] M.J. Ovens, A.J. Davies, M.C. Wilson, C.M. Murray, A.P. Halestrap, AR-C155858 is lytic tumors, Proc. Natl. Acad. Sci. U S A. 108 (2011) 16663–16668.
a potent inhibitor of monocarboxylate transporters MCT1 and MCT2 that binds to [34] P. Wong, H.W. Kleemann, I.F. Tannock, Cytostatic potential of novel agents that
an intracellular site involving transmembrane helices 7–10, Biochem. J. 425 (2010) inhibit the regulation of intracellular pH, Br. J. Cancer 87 (2002) 238–245.
523–530. [35] M. Damaghi, J.W. Wojtkowiak, R.J. Gillies, pH sensing and regulation in cancer,
[17] B. Nancolas, R.B. Sessions, A.P. Halestrap, Identification of key binding site residues Front. Physiol. 4 (2013) 370, http://dx.doi.org/10.3389/fphys.2013.00370.
of MCT1 for AR-C155858 reveals the molecular basis of its isoform selectivity, [36] H. Shen, E. Hau, S. Joshi, P.J. Dilda, K.L. McDonald, Sensitization of glioblastoma
Biochem. J. 466 (2015) 177–188. cells to irradiation by modulating the glucose metabolism, Mol. Cancer. Ther. 14
[18] S. Yeluri, B. Madhok, K.R. Prasad, P. Quirke, D.G. Jayne, Cancer's craving for sugar: (2015) 1794–1804.
an opportunity for clinical exploitation, J. Cancer Res. Clin. Oncol. 135 (2009) [37] S. Kershaw, J. Cummings, K. Morris, J. Tugwood, C. Dive, Optimisation of im-
867–877. munofluorescence methods to determine MCT1 and MCT4 expression in circulating
[19] M.W. Pfaffl, A new mathematical model for relative quantification in real-time RT- tumour cells, BMC Cancer 15 (2015) 387, http://dx.doi.org/10.1186/s12885-015-
PCR, Nucleic Acids Res. 29 (9) (2001) e45. 1382-y.
[20] O.P. Kallioniemi, T. Visakorpi, K. Holli, J.J. Isola, P.S. Rabinovitch, Automated peak [38] R.A. Robergs, F. Ghiasvand, D. Parker, Biochemistry of exercise-induced metabolic
detection and cell cycle analysis of flow cytometric histograms, Cytometry 16 acidosis, Am. J. Physiol. 287 (2004) 502–516.

179