(CANCERRESEARCH46, 2821-2826, June 19861
Influence of Hypoxia and an Acidic Environment on the Metabolism and Viability
of Cultured Cells: Potential Implications for Cell Death in Tumors'
Daniela Rotin,2 Brian Robinson,
and Ian F. Tannock3
Departments ofMedical BiophysicsfD. R., I. F. T.jand BiochemistryfB.
R.J, University ofToronto, andDivision ofPhysics[D. R., L F. T.J and Department of Medicine
II. F. T.J,OntarioCancerInstitute,500SherbourneStreet,Toronto,Ontario,CanadaM4X 1K9
a hypoxic environment, provided that the supply of other nu
trients is maintained. Studies of spheroids (i.e., multicellular
Hypoxia and an acidic environment are known to occur in regions of
aggregates of cells which may grow in culture) have demonstra
solid tumors and might be involved in the causation of necrosis. The
ted the formation of a necrotic center in the absenceof severe
viability and energy metabolism of cells in tissue culture were therefore
hypoxia (7). The thickness of the viable rim of spheroids ap
investigated under hypoxic and/or acidic conditions. Acute exposure of
pears to depend on limited diffusion ofglucose as well as oxygen
Chinese hamster ovary (CHO) cells or human bladder cancer MGH-UI
(8, 9), and other factors probably contribute to cell death.
cells to hypoxia plus low pH (6.5 to 6.0) was cytotoxic in a time- and
pH-dependentmanner, survivingfraction was reducedto -@10@
following
Hypoxic cells are dependent on anaerobic glycolysis to supply
a 6-h exposure to hypoxia at pH 6.0. There was no effect on viability their energy requirements; indeed, hypoxia is known to increase
whenaerobic CHO cells were exposed for 6 h at pH 6.0, or wheneither the activity of all I 1 glycolytic enzymes (10). Anaerobic glycol
cell line was rendered hypoxicfor 6 h at pH 7.0; MGH-U1 cells showed ysis leads to hydrolysis of ATP (1 1) and to the accumulation of
slight sensitivity to acidic pH in air. Decrease in viabilityof CHO cells lactic acid, with a consequent decreasein pH. The averagepH
incubatedunderacidconditionswasobservedoverthe rangeof oxygen
in a variety of tumors was found to be about 0.5 units lower
concentrations from 0.2 to 0.05%, similar to the range which causes
change in cellular sensitivity to radiation. Glucose consumption and than that of the surrounding normal tissues [(pH 6.5 to 6.9 and
7.0 to 7.5, respectively (12—15)1,
but pH values of 6.0 or lower
lactate production by both cell lines were inhibited at low pH under both
have been recorded in tumors (16). Values of pH are likely to
aerobic and hypoxic conditions. Cellular adenosine triphosphate
(AlP)
ABSTRACT
levels and the energy charge I(ATP + ½adenosine diphosphate)/(aden
osine monophosphate + adenosine diphosphate + A1P)I of CHO cells
vary in different
regions of tumors, and lower than average
hypoxia at pH 6.0 but were not influenced by hypoxia or acid pH alone.
values would be expected to occur in areasof hypoxia.
Despite the limited capacity of cells to withstand acidifica
tion, the role oflow pH in contributing to cell death in hypoxic
Inhibition
regions
were reduced by about 85 and 25%, respectively, after a 6-h exposure to
of glycolysis
by incubation
of CHO cells under hypoxic con
ditions in the absence of glucose(at pH 7.0) led to a larger fall in cellular
AlP and energy charge, but cell survival fell to only @@I0_2
at 6 h. These
results demonstrate
that hypoxia and an acid environment interact to
cause marked toxicity. A decrease in energy charge of the cells may
contribute
to loss of viability,
but additional
mechanisms
appear
to be
involved.
of tumors
has not been addressed.
In a preliminary
report (17) we showed that the combined effect of acidity and
hypoxia could lead to a marked fall in the viability of cultured
cells. The purpose of the present work was to extend these
initial studies and to analyze characteristics of the interaction
between hypoxia and an acidic environment which may lead to
changesin cell survival and cellular energy metabolism.
INTRODUCTION
MATERIALS
Cell death occurs in most solid tumors, but underlying mech
anisms remain poorly understood.
in a tumor
is sometimes
found
Cells. CHO4 cells and the human bladder cancer cell line MGH-U1
(obtained from Dr. G. Prout, Urology Research Laboratory, Massachu
The edge ofa necrotic region
to be parallel
AND METhODS
to a neighboring
blood vessel (1—3),suggesting that one cause of cell death may
be limited diffusion of essential nutrients to the tumor cells or
setts General Hospital, Boston, MA) were maintained in complete a-
inadequate
Both cell lines were free of Mycoplasma.
Prior to experiments, exponentially growing MGH-Ul cells were
detached from their flasks using 0.025% trypsin and 0.01% EDTA,
washed, and resuspended in fresh medium. CHO cells were maintained
routinely in culture flasks and were transferred and grown in spinner
removal
of catabolites
by the vascular
medium (18) supplemented with antibiotics and 10% FCS. Cultures
were reestablished from frozen stock at approximately 3-mo intervals.
system. The
distance between capillaries and the edge of a necrotic region
(typically
100 to 200 @zm)(1—3)is consistent
with estimates
of
the diffusion distance for oxygen in tissue (4). Evidence from
radiobiological
experiments
that
most
solid tumors
contain
hypoxic cells (4) and the observation that the distance separat
ing necrotic regions from blood vesselswas smaller in tumors
of mice that were placed in an oxygen-deficient environment
than in mice breathing air (5) are consistent with the hypothesis
that hypoxia may contribute to cell death.
It is unlikely that hypoxia is the sole cause of cell death in
solid tumors. We and others (6) have observed that cultured
cells may survive for periods of 24 h or longer when placed in
Received 8/6/85; revised 12/9/85, 2/24/86; accepted 2/27/86.
The costs of publication of this article were defrayed in part by the payment
of page charges. This article must therefore be hereby marked advertisement in
accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
I This
study was supported
by a research
grant
from
the National
Cancer
culture for a few days prior to their use in experiments.
Cell Survival Experiments. A suspension of exponentially growing
cells was washed and diluted in a-medium plus 5% DFCS to provide a
concentration of l0@cells/mi. Volumes of 10 ml were added to small
glassvials and stirred continuously at 37C. A humidified gas mixture
of 5% CO2 in air or nitrogen (<10 ppm 02), or in a specified interme
diate oxygen concentration, flowed through the vials, as described
previously (19). At appropriate times, 0.5 ml ofthe cell suspension was
removed by passing the long needle of a syringe through the gas outlet.
The cells were counted, and appropriate dilutions were plated in amedium + 10% FCS in triplicate Petri dishes. Colonies were stained
and counted 9 to 13 days later. In all experiments, acidic conditions
were achieved by adding appropriate amounts of sodium bicarbonate
(i.e., less than the usual concentration
of 26 mM) to bicarbonate-free
a-
medium (+ 5% DFCS) to obtain the desired pH. In experiments in
Institute of Canada and by Grant CA 36913 from the National Cancer Institute,
NIH.
2 Recipient
of
a research
studentship
from
the
National
Cancer
Institute
of
4 The
Canada.
3 To
whom
requests
for
reprints
should
be
abbreviations
used
are:
CHO
cells,
Chinese
hamster
ovary
cells
DFCS,
dialyzed fetal calf serum; HPLC, high-performance liquid chromatography;
TBAP, tetrabutylammonium phosphate; FCS, fetal calf serum.
addressed.
2821
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1986 American Association for Cancer Research.
POTENTIAL
IMPLICATIONS
OF HYPOXIA
AND AN ACIDIC ENVIRONMENT
FOR TUMOR
CELL DEATH
which cells were incubated in the absence of glucose, cells were washed
once with glucose-free medium prior to initiation of incubation.
BiochemicalAssays.TheconcentrationofL-lactate,pyruvate,ando
A new set ofstandards
a standard curve was constructed from the peak area of the Standards.
glucose in the incubation medium was measured with a spectrophotom
confirmed by the addition of a known quantity of AMP, ADP, and
eter (Cary 219, Varian) using commercial kits (Sigma, St. Louis, MO).
AlP (standard
Energy charge
ATP)J (22).
In the presence of excess NAD and lactate dehydrogenase, lactate is
converted to pyruvate and NADH. The increase in absorbance at 340
nm due to the formation of NADH is directly proportional to lactate
concentration.Whenthereversereactionis carriedout, thedecreaseof
solution) to a duplicate
was calculated as [(AlP
sample, as shown
+ ½ADP)/(AMP
in Fig. 5E.
+ ADP +
Effect of Hypoxia and Acidity on Cell Survival. Because of
the conversion of glucose to glucose-6-phosphate by ATP in the pros
the different buffering capacity of media containing different
ence of hexokinase, coupled with the subsequentreduction of NADP
to NADPH; the absorbanceof NADPH was measuredat 340 nm. The
in allexperiments(unless
day), although
RESULTS
absorbance at 340 nm due to disappearance of NADH is proportional
to pyruvate concentration. Determination of glucose levels is based on
initial glucose concentration
was run for each experiment(each
all the standard curves obtained were similar. Peak identification was
concentrations
of bicarbonate,
medium pH was monitored
over
the 6-h duration of an experiment in which CHO cells (at I x
otherwise stated)
106 cells/ml)
was 5.6 mM.
were incubated
at various initial values of pH and
The cellularconcentrationof ATP, ADP, and AMP wasmeasured gassed with air or N2 (each with 5% CO2). The results show
that medium pH increased following the addition of cells, but
by HPLC. The following chemicals were used. Acetonitrile and
KH2PO4(HPLC grade) were obtained from Fisher (Fair Lawn, NJ);
TBAP from Waters Associates (Milford, MA); trichlorotrifluoroethane
(Freon), from Matheson (Whitby, Ontario); and perchloric acid, from
Baker (Phillipsburg, NJ). All other chemicals were purchased from
Sigma (St. Louis, MO). For determination of adenine nucleotides,
l.0
samples were prepared for analysis as described by Bump et aL (20).
Briefly, iO@cells were centrifuged (1000 rpm, for 5 mm, at 4C), washed
once (for 5 mm, including centrifugation
time at Ot) with 10 ml of
ice-cold phosphate-buffered
saline, deproteinized
for 15 mm in 1 ml of
‘0-I
ice-cold perchioric acid (0.4 M), and sedimented, and the supernatant
was stored at —70Cfor up to 1 wk.
Perchloric acid was removed by extraction with 1 ml of 20% trieth
ylamine in Freon, followedby four extractions with I ml ofcold Freon.
All extractions
were carried out on ice. For preparation
of standard
solutions, increasing concentrations ofAMP, ADP, and AlP dissolved
in ice-cold PBS were added to 1 ml of ice-cold perchloric
acid (0.4 M)
for 15 mm. Standard solutions were then treated in a manner identical
to the experimental samples. For HPLC analyses, 5 to 10 Miof the
i0-@
extract (from the samples or the standard solutions) were Injected into
the HPLC instrument(WatersAssociates,Milford, MA; Model 440
absorbance detector), using a reverse phase Nova Pal C1@column with
isocratic elution and a buffercontaining 65 mM KH2PO., 1 mi@iThAP,
and 5% acetonitrile at pH 3.2 (21). The flow rate was I mi/mm, and
absorbancewas measured at 254 nm. The area under each peak was
calculated
by an integrator
(Data Moduk
730@,Waters Associates),
(j
and
i0-@
V:,1
,@
icr5
PH
adjusted
Cells added
gassing started
Incubation time I h)
incubation time (h)
Fig. 2. Plating efficiencyof CHO (A)and MGH-Ul cells (B) incubatedin air
Fig. 1. pH of media measured during incubation ofCHO cells (1 x 10' cells/
(closed symbols) or hypoxia (N,, open symbols) for up to 6 h at the indicated
values of medium pH. Mean and range for triplicate plates are plotted. Data are
ml) with constant gassingwith either air(+5% CO@J(cIasedsymboIs)or
N,(+5%
representativeoftwo experimentsperformedwith MGH-Ul cellsand more than
@
CO2) (open symbols). The values indicated at the left side ofeach cure represent
ten experiments using CHO celk •and 0, pH 7.0
@
the initial pH ofthe mediumbeforethe additioe olcells and initiation ofgassing@ pH 6.25; A and
and 0, pH 6.5; V and V,
pH 6.0.
2822
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1986 American Association for Cancer Research.
@
.
POTENTIAL
it decreased
gradually
IMPLICATIONS
after initiation
OF HYPOXIA
ofgassing,
AND AN ACIDIC ENVIRONMENT
reaching values
that were close to the initial medium pH (Fig. 1). The stability
of medium pH over the course of 6-h incubation with cells was
greater at lower pH (6.0 to 6.5) than at higher pH (-.‘7.0)
and
was also greater under hypoxic conditions than in air.
Exposure of CHO cells to either hypoxia or low pH (in the
range of pH 6.0 to 6.5) for up to 6-h had no effect on plating
efficiency, but the combination of both conditions led to a
marked decreasein cell survival. Loss ofviability was dependent
on duration of exposure (Fig. LI) and on pH (Fig. 3A). For
MGH-Ul cells (Fig. 2B), exposure to hypoxia alone for up to
6 h has no effect on viability, but exposure of aerobic cells at
pH 6.0 wascytotoxic. The combinationof hypoxiaand acidic
pH also resulted
in a marked
decrease
in plating
efficiency of
MGH-Ul cells. Both types of cell showed no change in their
ability to exclude trypan blue following 6-h incubation under
I
I
I
I
I
l.0
Air
i0@
FOR TUMOR CELL DEATH
any of the above conditions.
The oxygen sensitivity of pH-dependent cell killing of CHO
cells is shown in Fig. 38. At pH 6.2 ±0.2 there was a rapid
decline in plating efficiency as the oxygen concentration was
decreased from about 0.2% (0.7 mm of Hg) to about 0.05%
(0.2 mm of Hg).
Effect of Hypoxia and Acidity on Energy Metabolism.
Exper
iments were undertaken to determine whether the observed
decreases in cell survival could be correlated with changes in
cellular energy metabolism. In both CHO and MGH-U1 cells,
glucose consumption and lactate production were inhibited by
low pH, with almost complete cessation of glycolysis at pH 6.0
(Fig. 4). Glucose consumption and lactate production by CHO
cells were always higher under hypoxic conditions than in air
(Fig. 4A), whereas they were similar for MGH-Ul cells under
aerobic and hypoxic conditions (Fig. 4B). Net pyruvate removal
from medium by both cell lines was not affected by pH and was
slightly higher in air than in hypoxia (Fig. 4).
The influence of various conditions on levels of AlP and
energy charge in CHO cells is summarized in Table 1. This
table was constructed following HPLC analysis of levels of
AMP, ADP, and ATP, as demonstrated in Fig. 5. Table 1A
shows that, after 6-h incubation, energy charge was reduced
below the normal value of 0.85 only in cells incubated under
hypoxic conditions at low pH. The dependenceof this fall in
energy charge on time of incubation and on pH is shown in
Fig. 6. The cellular concentrationof ATP showeda larger
:@
cr4
‘0_s
N2
4
72
I
I
70
6.8
I
I
6.6
6.4
Medium pH
I
I
6.2
6.0
E
®
Oxygen
effect
Air, p1170
3.0
I.0
N2,pH7O
AarpH6.0
N2,pH6O
®MGH-UI
25
20
‘5
I0@'
,02
I
10
1.
‘
0.i
0.01
% 0@ fl gas mIxture
Fig. 3. A, the influenceofpH on the plating efficiencyofCHO cells incubated
under aerobic or hypoxic conditions for 6 h. Mean and range of triplicate plates
of a typical experiment are plotted. B, the influenceof oxygenconcentration on
the plating efficiency of CHO cells incubated at low pH. Points, mean of three
experiments carried out at pH 6.2 ±0.2; bars, SE.
@i@kLLh
AirpH7O
N2,pH7.0
Air, p1165
N2,pH6.5
Air, pHSO
N2,pH6O
Fig. 4. Lactate production, glucose consumption, and net pyruvate consump
tion in CHO (A) and MGH-U1 (B) cells. Values are expressed as mM concentra
tion measured in cell-free medium following incubation of 1 x l0@cells/mI for 6
h at the indicated values of pH, under aerobic or hypoxic conditions. Represent
ative data are shown for two experiments with MGH-Ul and for ten experiments
using CHO cells. Initial glucoseand pyruvateconcentrationsin the mediumwere
5.6 mat and 1 mM, respectively.
2823
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1986 American Association for Cancer Research.
POTENTIAL
IMPLICATIONS
OF HYPOXIA
AND AN ACIDIC ENVIRONMENT
Table I RelaiiveconcentrationsoJATP and enerij@chargeofCHO cells
CELL DEATH
A. 5,@l @f25,@M standord
incubatedfor 6 h under the in,dicated conditions
Energy charge = (ATP + 1/2ADP)/(AMP + ADP + ATP).
Incubation
condition%
chargeA.Effect
FOR TUMOR
of ATP@Energy
of hypoxia and low pHL@
Air, pH 7.0
N2,pH 7.0
115
Air, pH 6.0
0.88
0.87
99
140.85
0.64cB.Effect
N2,pH 6.0100
of .y@irradiationd
Control
0.88C.Effect
-y-irradiated (30 Gy)100
0.000
I 130.86
ofglucose depletione
sompleN2,PH7.0,
N2,pH 7.0, —glucose
N3, pH 6.0, —glucose2.5
a ATP
levels
are expressed
0.50.56
as percentage
ofcontrols,
®
Glucose
0.45
to account
for any possible
loss of ATP during preparation ofcells prior to extraction with perchloric acid.
b Means
of two
to four
experiments.
All
groups
were
incubated
in the
presence
I.
01
ofglucose (5.6 mM).Multipleestimatesofenergy charge undercontrol conditions
are highly reproducible with a range of0.82 to 0.88.
C Energy
d Cells
e Means
charge
were
at
the
incubated
of three
beginning
in standard
experiments.
of
the
experiments
a-medium
Control
values
was
plus
for
5%
0.85.
DFCS
relative
(pH
7.4).
concentrations
of
ATP are those observed in cells incubated in N2, pH 7.0, with glucose present
(5.6 mM).
decrease than energy charge (Table 1A), since there was a
concomitant reduction in AMP levels under hypoxic and acidic
conditions (data not shown).
Changes in energy metabolism of cells exposed to hypoxia
and low pH might causecell death or might simply be an early
manifestation of loss of viability. We therefore studied the rate
of glycolysis and energy charge following an independent
method of causing cell death. A lethal dose of 30 Gy of ‘y—ir
radiation reduced the surviving fraction ofCHO cells below 1.3
x lO@ but had no effect on their lactate production (Fig. 7) or
energy charge (Table 1B) for at least 6 h.
Effect of Glucose and Pyruvate on Survival and Energy Me
0.005 c. 5,@lof somple N2,PH7.0, eGlucose
I
@ILjt\
@0.ooo
N2,PH6.0, ®Glucose
tabolism of CHO Cells. To investigate further whether changes
in energy metabolism might be causally related to cell death,
we incubated cells in a hypoxic environment in the absence of
glucose. We therefore removed the major substrate for glycolyis
and prevented oxidative phosphorylation. As expected, these
conditions led to almost total inhibition of lactate production
(Fig. 8) and to a fall in energy charge to an even lower value
than that observed under hypoxic conditions at pH 6.0 (Fig. 5;
Table 10. However, cell survival fell only to about 1O_2when
cells were incubated for 6 h under hypoxic conditions at pH 7.0
in the absence ofglucose (Fig. 9); this is a much higher surviving
fraction
than
was observed
when
the cells were
incubated
at pH
6.0 under hypoxic conditions in the presence of glucose (5.6
mM).
Removal of glucose from the medium also led to a lower cell
survival under hypoxic conditions at pH 6.0 (Fig. 9) than in the
presenceof glucose. Removal of pyruvate had a similar (albeit
smaller) effect to lower survival of hypoxic cells at pH 6.0, but
it had no effect at pH 6.5 or 7.0 (data not shown).
0.0I0
The low
.E. 5,@l mixture of 2/3
sample
B
@:
and /3 5MMstandard
0.
0.005
Fig. 5. HPLC chromatograms showing the separation of AMP, ADP, and
ATP. The procedure for HPLC analysis is described in “Materials
and Methods.―
A, separation of standard solution containing 25 Mmol of each nucleotide; B,
separation of nucleotides extracted from 10' CHO cells following incubation for
6 h at pH 7.0 under hypoxia, with normal (5.6 mM) glucose concentration; C, as
in B, but cells were incubated in medium lacking glucose; D, as in B, but cells
@
were incubated at pH 6.0 E, peaks were identified by analysis of the HPLC
@
profile of a mixture containing ¾of an extract from a 5 @M
standard solution
(containing 5 @imolofeach nucleotide) and ½ofan extract from sample (B). The
numbers indicated by the peaks of AMP, ADP, and ATP represent the exact
elution time, as printed by the integrator.
0.000
0
2
4
./*,
8
Time (mm)
2824
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1986 American Association for Cancer Research.
@
-@
@
@@oI
POTENTIAL
@
IMPLICATIONS
Hypoxia
09
@
08
@
07
OF HYPOXIA AND AN ACIDIC ENVIRONMENT
FOR TUMOR CELL DEATH
‘
•_•
•\
@“O2hrs.
. 6 hrs.
0.6
@,0.5
7.0
6.6
6.2
5.8
5.4
5.0
Medium pH
I
©Hypoxia,
pH6.0
L@ Q9
•NN
08
.@
07
06
@
0
0.5
0
2
4
ncubation time ( h I
@
o ----i-
,
,-
6
8
‘I',,
0
6
time ( h
Fig. 8. Lactate production by CHO cells incubated with or without 5.6 mM
glucose (gk), under air (closed symbols) or N, (open symbols) at pH 7.0 or 6.0.
Values represent mean lactate concentration of three measurements carried out
on cell-free media following incubation of 1 x 10' cells/mI for up to 6 h. •and
6
Fig. 6. Reduction of energy charge of hypoxic CHO cells as a function of
medium pH, after 2-h or 6-h incubation (A), and incubation time at pH 6.0 (B).
Points, mean calculated from the concentration of adenine nucleotides measured
by FIPLC in one to four experiments.
‘
4
Incubation
0, pH 7.0, with glucose;U and 0, pH 7.0, without glucose;V and V, pH 6.0 with
glucose; A and
@,
pH 6.0, without glucose.
XRT(30Gy@4
2
4
6
8
20
22
24
Incubot@on
time ( h
Fig. 7. Lactate productionby CHO cellsat intervalsafter 30-Gy‘y-irradiation
and by unirradiated control cells. Data represent values for three measurements
oflactate concentration in cell-free medium following incubation of 1 x l0@cells/
ml at pH 7.4. The experiment was repeated, and similar results obtained.
production of lactate under hypoxic conditions at pH 6.0 was
further inhibited by removal of glucose (Fig. 8) or pyruvate
(data not shown) from the medium.
DISCUSSION
The present experiments have demonstrated a marked de
crease in survival of two types of cells when they were incubated
under hypoxic conditions at low pH. Hypoxia and an acidic
environment
are known
to occur in regions
4
Incubation time(h)
of solid tumors,
and our results suggestthat these conditions may contribute to
the process of natural cell death which occurs in solid tumors.
Lethal effects are observed in the range of pH 6.0 to 6.5, and
although this range is lower than average measurements of pH
in most
solid tumors
(12—16), regional
hypoxia
with consequent
production of lactic acid is likely to lead to tumor regions with
values of pH in this range. Oxygen sensitivity of the lethal
Fig. 9. Plating efficiencyof CHO cells assessedin an experiment identicalto
that described in Fig. 8. Symbols are as in Fig. 8. Data represent mean and range
of values for triplicate plates. Similar results were obtained in two replicate
experiments.
effect of low pH occurs over a similar range of PO2 to that
which influences radiation sensitivity, although it seems un
likely that these effects occur through related mechanisms.
2825
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1986 American Association for Cancer Research.
POTENTiAL IMPLICATIONS OF HYPOXIA AND AN ACIDIC ENVIRONMENT FOR TUMOR CELL DEATH
In both CHO and MGH-U1 cells, glucose consumption and
lactate production were suppressed as the pH of the medium
was lowered from 7.0 to 6.0, confirming earlier reports that
glycolysis is inhibited at low pH (23). This effect is probably
due to inactivation of phosphofructokinase, the main regulatory
enzyme in glycolysis, at low pH (24-26). Thus, one possible
explanation for the cell death observed under conditions of
hypoxia and acidity is that cells incubated in hypoxia (a condi
tion in which respiration
is inhibited)
REFERENCES
1. Thomlinson, R. H., and Gray, L. H. The histological structure of some
human lung cancers and the possible implications for radiotherapy. Br. J.
Cancer, 9: 539—549,
1955.
2. Tannock, I. F. The relation betweencellproliferationand the vascularsystem
in a transplanted
kinetics and radiobiology. Br. J. Cancer, 51: 407—413,1985.
4. Tannock, I. F. Oxygen diffusion and the distribution of cellular radiosensi
tivity in tumours. Br. J. Radiol., 45: 515—524,1972.
5. Tannock, I. F., and Steel, G. G. Tumor growth and cell kinetics in chronically
hypoxic animals. J. Nati. Cancer Inst., 45: 123—133,1970.
6. Shrieve, D. C., Deen, D. F., and Harris, J. W. Effects ofextreme hypoxia on
the growth and viability of EMT6/SF mouse tumor cells in vitro. Cancer
Res.,43:3521—3527,
1983.
7. Carlsson, J., Stalnacke, C. G., Acker, H., Haji-Karim, M., Nilson, S., and
the levels of
recovery appears to depend on ion exchange pumps (e.g., Na']
of drugs
Larsson, B. The influence ofoxygen on viability and proliferation in cellular
spheroids. lnt. J. Radiat. Oncol. Biol. Phys., 5: 2011—2020,
1979.
8. Li, C. K. N. The glucosedistribution in 9L rat brain multiceiltumor spheroids
and its effect on cell necrosis. Cancer (Phila.), 50: 2066—2073,1982.
9. Freyer, J. P., and Sutherland, R. M. Effects of oxygen and glucose on
spheroid growth (abstract).Strahlentherapie, 160: 52—66,
1984.
10. Robin, E. D., Murphy, B. .1., and Theodore, J. Coordinate regulation of
glycolysis by hypoxia in mammalian cells. J. Cell. Physiol., 118: 287—290,
1984.
11. Buss, W. B., and Nuccitelli, R. Metabolic regulation via intracellular pH.
Am. J. Physiol.,246: R409—R438,
1984.
12. Eden, M., Haines, B., and Kahler, H. The pH of rat tumors measured in
vivo. J. NatI. Cancer Inst., 16: 541—556,1955.
13. Ashby, B. S. pH studies in human malignant tumours. Lancet, 2: 312—315,
1966.
14. Jahde, E., Rajewsky, M. F., and Baumgartl, H. pH distributions in trans
planted neural tumors and normal tissues of BDIX rats as measured with
pH microelectrodes. Cancer Res., 42: 1498-1504, 1982.
15. Jam, R. K., Shah, S. A., and Finney, P. L. Continuous noninvasive monitor
ing of pH and temperature in rat Walker 256 carcinoma during normogly
cemia and hyperglycemia.J. Natl. Cancer Inst., 73:429-436, 1984.
16. Jahde, E., and Rajewsky, M. F. Tumor-selectivemodification of cellular
microenvironmentin vivo'@
effect of glucose infusion on the pH in normal
and malignant rat tissues. Cancer Res., 42: 1505—1512,1982.
17. Rotin, D., and Tannock, I. F. The effect ofthe drug lonidamine on Chinese
hamster ovary cells in vitro and on experimental tumors. Int. J. Radiat.
Oncol. BioLPhys., 10: 1595—1598,
1984.
18. Stanners, C. P., Eliceiri, G. L, and Green, H. Two types of ribosome in
mouse-hamsterhybridcells. Nat. New Biol., 230: 52—54,
1971.
19. Mohind.ra,J. K., and Rauth, A. M. Increased cell killing by metronidazole
and nitrofurazoneof hypoxiccompared to aerobic mammaliancells. Cancer
Res., 36: 930—936,
1976.
20. Bump, E. A., Calderwood, S. K., Sawyer, J. M., and Brown, J. M. Role of
the adenylate energy charge in the response of Chinese hamster ovary cells
to radiation. Int. J. Radiat. Oncol. Biol. Phys., 10: 1411—1414,1984.
21. Darwish, A. A., and Prichard, R. K. Analysis ofribonucleotides by reverse
phase HPLC using ion pairing on radially compressed or stainless steel
columns.J. Chromatogr., 4: 1511—1524,
1981.
22. Atkinson, D. E. Cellular Energy Metabolismand Its Regulation.NewYork:
Academic Press, 1977.
23. Halperin, M. L., Connors, H. P., Relman, A. S., and Karnovsky, M. L.
Factors that control the effect of pH on glycolysisin leukocytes.J. BiOL
Chem., 244: 384—390,
1969.
24. Trivedi, B., and Danforth, W. H. Effect ofpH on the kinetics offrog muscle
phosphofructokinase. J. Biol. Chem., 241: 41 10—41
14, 1966.
25. Wilhelm,G., Schulz,J., and Hofmann, E. pH-dependenceofacrobic glycol
ysis in EhrIICh ascites tumour cells. FEDS Left., 17: 158—162,1971.
26. Kitajima, S., Sakakibara, R., and Uyeda, K. Significance of phosphorylation
ofphosphofructokinase. J. Biol.Chem., 258: 13292—13298,
1983.
27. Warburg, 0. On the origin of cancer cells. Science (Wash. DC), 123: 309—
314, 1956.
28. Gillies, R. J., Ogino, T., Shulman, R. G., and Ward, D. C. “Pnuclear
magneticresonanceevidencefor the regulationofintracellular pH by Ehrlich
H' and/or HCO,/C1)
located in the cell membrane. We are
currently investigating the effect on cell viability ofagents which
enhance intracellular acidification by interference with the nor
mal function of these ion exchange pumps.
Because cells in nutrient-deficient regions of tumors are
known to be resistant to radiation and to some of the conven
tionally usedanticancer drugs (31—33),
imitation or stimulation
of mechanisms which lead to natural tumor cell death might
have considerable potential in tumor therapy. Results of the
present study may aid in the selection and development
Br. J. Cancer, 22: 258—273,
bronchialand cervicalsquamouscellcarcinomas:inferencesfor their cellular
in which glycolysis is inhibited) might die ofenergy deprivation.
The observed reduction in ATP levels and in energy charge of
CHO cells under these conditions supports the view that energy
deprivation might contribute to cell death, especially since the
range of pH sensitivity (6.5 to 6.0) is similar for the decrease
phosphocreatine were not measured, this compound was not
likely to be a major source of energy in our cells, which were
not of muscle or nerve origin.
The absenceof an enhanced rate of glycolysis under hypoxic
conditions for MGH-U1 cells is consistent with earlier studies
(e.g., Ref. 27) which show that many tumor cells utilize glycol
ysis even in the presenceof oxygen. CHO cells, which are not
malignant, show the normal increase in rate ofglycolysis under
hypoxic conditions relative to air.
CHO cells which received a lethal dose of X-rays had a
normal rate ofglycolysis for 24 h and normal energy charge for
at least 6 h after irradiation. Cells that were lethally damaged
by exposure to hypoxia and low pH maintained intact mem
branes and excluded dye during the 6-h period of incubation,
similar to those treated with radiation. It seemsunlikely there
fore that the changes in metabolism observed within 6 h of
incubation under hypoxia and low pH were a nonspecific result
of cell death, although we cannot exclude this possibility with
certainty.
In agreement with earlier studies (28, 29), we have found that
inhibition of glycolysis by the strategy of removing glucose and
oxygen led to a decreaseof cell survival and reduction of ATP
levels and energy charge. In our experiments, this reduction in
cell survival was much smaller than that observed following
incubation ofcells at pH 6.0 (with glucose) under hypoxia. This
smaller cell kill could not be attributed to continuing glycolysis
from breakdown of glycogen stores, because the rate of lactate
production under hypoxic conditions was similar (and very low
Fig. 8) in cells incubated without glucose and in cells incubated
with glucose at pH 6.0. The observation that cell survival was
higher but energy charge was lower (Table 1; Fig. 9) in hypoxic
cells incubated in the absenceof glucose (at pH 7.0) than in
hypoxic cells incubated at pH 6.0 (with glucose) suggests that
mechanisms other than energy deprivation were contributing
to cell death. It is possible that hypoxia inhibited recovery from
intracellular acidification in cells placed in an acidic environ
ment, as shown for Ehrlich ascites tumor cells (28, 30). Such
tumour.
3. Moore, J. V., Hasleton, P. S., and Buckley, C. H. Tumourcords in 52 human
and low pH (a condition
of both energy charge and cell survival. Although
mouse mammary
1968.
ascites tumor cells. J. Cell Biol., 95: 24—28,1982.
29. Gerweck, L. E., Dahlberg, W. K., Epstein, L. F., and Shimm, D. S. Influence
of nutrient and energy deprivation on cellular response to single and frac
tionated heat treatment. RadiaL Res., 99: 573—581,
1984.
30. Bowen, J. W., and Levinson, C. W transport and the regulation of intracel
lular pH in Ehrlich tumor cells. J. Membrane BioL, 79: 7—18,1984.
31. Hill, R. P., and Stanley, J. A. The response of hypoxic B16 melanoma cells
to in vivo treatment with chemotherapeutic agents. Cancer Res., 35: 1147—
1153, 1975.
32. Sutherland, R. M., Eddy, H. A., Bareham, B., Reich, K., and Vanantwerp,
D. Resistance to Adriamycin in multicellular spheroids. tnt. J. Radiat. Oncol.
Biol. Phys., 5: 1225—1230,1979.
33. Tannock, I. Response of aerobic and hypoxic cells in a solid tumor to
Adriamycinand cyclophosphamideand interaction of the drugs with radia
with activity against nutritionally deprived tumor cells.
tion. Cancer Res., 42: 4921—4926,1982.
2826
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1986 American Association for Cancer Research.
Influence of Hypoxia and an Acidic Environment on the
Metabolism and Viability of Cultured Cells: Potential
Implications for Cell Death in Tumors
Daniela Rotin, Brian Robinson and Ian F. Tannock
Cancer Res 1986;46:2821-2826.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/46/6/2821
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1986 American Association for Cancer Research.