[CANCER RESEARCH 52, 4441-4447, August 15. 1992]

Regulation of Intracellular pH in Tumor Cell Lines: Influence ofMicroenvironmental Conditions1

Michael J. Boyer2 and Ian F. Tannock3

Departments of Medicine and Medical Biophysics, Ontario Cancer Institute and University of Toronto, Toronto, Ontario M4X 1K9, Canada

ABSTRACT

The effect of microenvironmental factors on the regulation of intra-<i-lInlar pH (pH¡)in MGH Ul cells and EMT-6 cells was studied usingthe fluorescent pH probe 2',7'-bis-(2-carboxyethyl)-5-(and-6)-carboxy-fluorescein. Na*/H* exchange and Na*-dependent Cl /HCO3 ex

change were found to be present in both cell types. The activity of bothexchangers was dependent on pi I¡.with low levels of activity at neutralpH and an increase in activity as pi 1,fell. The level of extracellular pH(pi I..) also influenced the operation of the exchangers, with a fall inactivity as pHe was reduced over the range 7.4-6.6. This effect was moremarked for the Na*-dependent Cl /HCO3 exchanger than for the

Na /I I antiporter, suggesting that under conditions of reduced pi I. theNa /I I antiporter is the major mechanism for regulation of pi I,. Neither 6 h of radiobiological hypoxia nor variations in the extracellular[Ca2+l over the range 1-6 IBMhad an effect on the regulation of pH,,while extracellular lactate (5-10 HIM)caused a small, concentration-dependent decrease in the combined activity of both exchangers. Weconclude that under the microenvironmental conditions found in someregions of tumors, Na /II exchange may be the major method ofregulation of pi I,.

INTRODUCTION

The level of pH¡4of mammalian cells is regulated within the

narrow range that is compatible with normal cellular function.At least three membrane-based transport systems are involvedin the regulation of pH¡(1,2). The Na+/H+ exchanger is a Mr

110,000 glycoprotein which is ubiquitous in mammalian cells(3). This antiport catalyzes the electroneutral exchange of Na+and H* across the cell membrane and is inhibited by amiloride

and its analogues (4, 5). Under physiological conditions, theexchanger operates to export H+ from the cell, although under

the appropriate ionic conditions it may operate in reverse; thisproperty has been exploited in the selection of variant cells thatlack the exchanger (6).

Aniónexchange also contributes to the regulation of pH¡andat levels of pH¡close to neutrality may be more active thanNa+/H+ exchange in some cell lines (7). Two different aniónexchangers exist, the cation-independent Cl~/HCOj exchanger and the Na+-dependent Cl~/HCOj exchanger. Underphysiological conditions, the cation-independent Cl~/HCOjexchanger allows Cl~ ions to enter the cell in exchange for

HCOj ions and therefore acts to decrease pH¡following cyto-plasmic alkalinization (8-10). By contrast, the Na+-dependentCl~/HCOj exchanger acts to protect cells from cytoplasmic

acidification (8-10). Both of these exchangers are inhibited bythe stilbene derivative DIDS (8-10).

Received 2/19/92; accepted 6/3/92.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.

1Supported by Grant CAS 1033 from the NIH.2 M. J. B. is supported by the Medical Research Council of Canada.3 To whom requests for reprints should be addressed, at Room 741. Ontario

Cancer Institute. 500 Sherbourne Street, Toronto. ON M4X 1K9. Canada.4 The abbreviations used are: pH¡,intracellular pH; pH,., extracellular pH;

/ÕHC03-bicarbonate intracellular buffering capacity; DIDS, 4-4'-diisothiocyano-stilbene-2,2'-disulfonic acid; EIPA, ethylisopropylamiloride; BCECF, 2',7'-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein; NMG, iV-methyl-D-glucamine.

The microenvironment in some regions of solid tumors differs from that of normal tissues (11, 12). The levels of pHewithin tissues may be measured by the use of microelectrodes,and such measurements have demonstrated that pHe is, onaverage, 0.5 pH units lower in solid tumors than in normaltissues (11, 12). By contrast, data obtained through the use of31P nuclear magnetic resonance spectroscopy, which measures

predominantly pH¡,suggest that pH¡within solid tumors iseither normal or close to normal (11). This implies that cellswithin solid tumors are capable of regulating the level of pH¡,despite lower-than-normal levels of pHc. Regions of reducedoxygen tension may also be present within tumors (11-13).These regions arise as a result of inadequate development oftumor vasculature and may correspond with regions of reducedpHc. The combination of hypoxia and reduced pHc may beresponsible, in part, for the high rate of spontaneous cell deathseen in some solid tumors (14).

Manipulation of pH¡within tumor cells has been proposed asa potential anticancer strategy (15). Furthermore, a reducedlevel of pHj sensitizes cells to the cytotoxic effects of hyperther-mia (16, 17). One major obstacle to these antitumor strategiesis the ability of cells to regulate their pH¡.In order to betterunderstand what factors influence the regulation of pH¡,wehave examined the influence of microenvironmental conditionssuch as might exist within solid tumors on the operation of boththe Na+/H+ antiport and the Na+-dependent Cl~/HCOj ex

changer in tumor cell lines.

MATERIALS AND METHODS

Cells. Experiments were performed with murine EMT-6 cells andthe human bladder carcinoma cell line MGH Ul. Cells were maintained routinely in «medium with 5% fetal calf serum, and new cultures, free of Mycoplasma, were reestablished from frozen stock every 3months.

Reagents. EIPA was synthesized by Aldrich (Milwaukee, WI), aspreviously described (18). DIDS was from ICN Biomedicals (St. Laurent, PQ, Canada). BCECF acetoxymethyl ester was from MolecularProbes (Eugene, OR). All other reagents were from Sigma (St. Louis,MO).

Solutions. Unless otherwise indicated, all solutions were nominallyHCOj-free. NaCl solution contained 140 miviNaCl, 5 miviKC1, 5 miviglucose, 1 miviCadi, and 1 m\i MgCI2, buffered to the indicated pHwith 20 mm 2-(A'-morpholino)ethanesulfonic acid/Tris. NaHCO3 solu

tion contained 25 HIMNaHCO3, 115 HIMNaCl, and other componentsidentical to those in NaCl solution. All solutions containing NaHCO_,were prepared in advance, but without the NaHCO3; this was addedimmediately before use. NMG and KC1 solutions were prepared byisoosmotic replacement of NaCl with jV-methyl-D-glucamine and KC1,respectively; the other components were identical to those describedabove for the NaCl solution. The composition of other solutions used insome experiments is indicated in the text or in figure legends.

Measurement of pH¡.Measurements of pH¡were made using thefluorescent dye BCECF. Cells grown as a monolayer on glass coverslipswere loaded with BCECF. The coverslip was then placed into a cuvetusing a specially designed holder aligned at an angle of 30°to the

excitation beam of a Perkin Elmer LS3 fluorometer (Perkin Elmer,Mississauga, ON, Canada). The holder also served as a cap for the

4441

on April 9, 2021. © 1992 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

REGULATION OF INTRACELLULAR pH IN TUMOR CELLS

cuvet, minimizing the loss of CO2. The cuvet was equipped with aperfusion system to allow the exchange of the buffer surrounding thecells. Exchanges were made with a volume of buffer at least 10 timesgreater than the volume contained within the cuvet. The temperature ofthe solution in the cuvet was controlled precisely, and all experimentswere carried out at 37°C.

Within the range of pH¡6-7.5, fluorescence intensity at 525 nm(following excitation at 495 nm) is linearly related to pH¡(19). At threetime points during each experiment (prior to intracellular acidification,following intracellular acidification, and at the end of the experiment),fluoresence intensity was measured at the same emission wavelengthbut with an excitation wavelength of 440 nm. Following excitation atthis wavelength, fluorescence intensity is independent of pH¡and depends only on the amount of BCECF present. This allowed the calculation of a fluorescence ratio that was independent of the amount ofBCECF present and thus took into account any loss of cells fromcoverslips or leakage of BCECF from the cells. Calibration of fluoresence measurements was performed using the ionophore nigericin, in asolution containing 140 imi K ' . as described previously (20).

Cytoplasmic Acidification and Recovery. Intracellular acidificationwas achieved by two different methods. In some experiments, nigericin,2 Mg/ml in NMG solution, was used to produce intracellular acidification to a level of 6.4-6.6. In other experiments, cells were placed in KC1solution containing NH4C1 for 30 min. Acidification, to a level determined by the concentration of NH4»used, was then produced by exchanging the NH4Cl-containing solution with a NH4Cl-free solution(21).

Cytoplasmic acidification was carried out in Na+- and HCOj-freebuffer. In experiments designed to measure the activity of the Na+/H+

exchanger, this buffer was replaced after intracellular acidification byNa+-containing, HCOj-free buffer. By contrast, in those experimentswhere the action of the Na+-dependent C1~/HCO5 exchanger was beinginvestigated, the buffer was replaced by Na+- and HCOj-containing

buffer, with either amiloride or EIPA present. This procedure was alsoused to measure the combined activity of both exchangers, but in theseexperiments, amiloride and EIPA were absent.

Following the change in extracellular buffer, the maximal rate of theresulting intracellular alkalinization was measured. In experimentswhere the influnce of pH¡on the operation of the exchangers wasexamined, the initial rate of alkalinization, rather than the maximalrate, was measured. The results of experiments were converted into H+

efflux by multiplying the observed rates of change of pH¡by the totalbuffering capacity (-¡Vsee below) using the formula

H+ efflux = rate of change of pH¡•¿�ß^

where ßT= ßt+ &HCOÕand ß\is the intrinsic intracellular buffering

capacity. For experiments performed in the absence of bicarbonate,0HC03 was 0.

Calculation of Intracellular Buffering Capacity. Buffering capacity isthe capacity of a cell to buffer changes in pi I, following addition orremoval of H+ and is defined as A [H+]/A pH¡(22). In order to measure

intrinsic (nonbicarbonate) intracellular buffering capacity, cells wereexposed for 5 min to HCOj-free NMG solution (pH 7.3) containing 3HIMammonium chloride, followed by replacement of the extracellularfluid with NH4-free NMG solution. The resulting fall in pH¡was measured and used to calculate intrinsic buffering capacity using the formula described by Roos and Boron (22). The pKa of ammonium chloride was taken as 9.3. Briefly, it was assumed that NH, was freelydistributed across the cell membrane, and the intracellular concentration of Ml., at a given value of pi I, and pi I., was calculated from theequation

L = ([NH4Cl]e •¿� •¿�10<"Hi"K>)

The intrinsic buffering capacity was then calculated using the formula

(3,= A [NH4+]i / A pH¡

where A [NH4.]¡and A pH¡are the changes that resulted from the

removal of all extracellular [NH4C1]. Measurements of intrinsic buffering capacity were made at the resting level of pH¡only, since valueshave been shown to be constant over the range of pH¡6.4-7.2 (23).

Bicarbonate buffering capacity was calculated as 2.3 [HCO5]¡(21).Since HCOj is in equilibrium with CO2, which is freely permeableacross the cell membrane, the value of [HCOj]¡is dependent on pH¡,pHe, and the amount of CO2/HCOj in the medium (provided that thisis assumed to be constant). The value of [HCOjli was calculated fromknowledge of these variables using the Henderson-Hasselbalch equation. However, since [HCO3]¡changes with variations in pHi, /3Hco3also varies. Thus values of /3Hco3 were calculated for the differentcombinations of pH¡and pHe that existed during maximum pH¡recovery in different experiments. The total buffering capacity was the sum ofintrinsic buffering capacity and bicarbonate buffering capacity.

Generation and Assessment of Hypoxia. Some experiments werecarried out under hypoxic conditions. Six h prior to the experiment,coverslips were placed in a cuvet, and the medium was replaced with«-mediumthat had been bubbled with N2 for 1 h. This medium wasbubbled in the cuvet for a further 20 min and then kept under a N2atmosphere. All subsequent manipulations were carried out using solutions bubbled with N2 for 1 h, and the coverslip was not removed fromthe cuvet at any time prior to the experiment.

In order to ensure that radiobiological hypoxia was present, cellsgrown on coverslips and handled as described above were irradiatedwith 10 Gy. They were then trypsinized, seeded into culture dishescontaining regular a-medium plus 5% fetal calf serum, and grown in ahumidified environment containing 5% CO2 and air. Colonies werecounted 9-14 days later. Survival of these cells was compared to that of

cells handled identically except under aerobic conditions. There was a100-fold greater killing of aerobic cells as compared to hypoxic cells,implying that radiobiological hypoxia was present (data not shown).

RESULTS

Mechanisms of pH¡Regulation. The resting level of pH¡inMGH Ul cells incubated in HCOj-containing medium was7.21 (SEM = 0.02; n = 4 experiments), while in HCOj-freemedium it was 7.25 (SEM = 0.02; n = 4 experiments). Theseresults were not sigificantly different (P = 0.10). In EMT-6cells, the corresponding values were 7.06 (SEM = 0.02; n = 4experiments) and 7.19 (SEM = 0.06; n = 4 experiments) (P =0.16). Initial experiments demonstrated that cells from bothcell lines possessed Na+/H+ exchange activity and Na+-depen-dent C1-/HCOT exchange activity (Table 1). Na+/H+ exchangeactivity was dependent on the presence of extracellular Na+ and

could be inhibited by the compound amiloride or its more potent analogue, EIPA. Activity of the Na+/H+ exchanger in

MGH Ul cells was decreased to approximately 15% of controlvalues by 0.1 mivi amiloride, while EIPA (10 UM) decreasedNa+/H+ exchange in EMT-6 cells by 95%.

Operation of the Na+-dependent Cl~/HCOj exchanger alsorequired the presence of extracellular Na+ and could be inhib

ited by the stilbene derivative DIDS. In both cell lines, DIDS(0.1 HIM) resulted in a decrease in the activity of the Na"1"-dependent Cl~/HCOj exchanger to less than 5% of controlvalues. When an outwardly directed Cl~ gradient was producedby replacing most extracellular Cl~ with gluconate, an increase

in the maximal rate of operation of the exchanger was observedin both cell lines (Table 1).

In experiments performed at pHe 7.4, activity of the Na^/H"1"

exchanger was markedly dependent on the level of pH¡(Fig. 1),although the threshold value of pH¡below which the exchangerbecame active was higher in MGH Ul cells (~pH¡7.0) than inEMT-6 (~pH¡6.7). Activation of the exchanger was maximal atpH¡6.4-6.8 in MGH Ul cells and pH¡6.4-6.6 in EMT-6 cells.At levels of pH¡lower than 6.4, there was a reduction in the

4442

on April 9, 2021. © 1992 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

REGULATION OF INTRACELLULAR pH IN TUMOR CELLS

Table 1 Transmembrane hydrogen ion flux in MGH VI and EMT-6 cells during recovery from intracellular acidificationValues are the mean (±SEM) of at least three experiments. Acidification took place in Na* and HCO3 free medium, which was then replaced by the test solution

indicated (see "Materials and Methods" for composition). Efflux rates are the product of the maximal rate of pi 1, change and buffering capacity. Sodium gluconatesolution had the same composition as Nal K "O, solution, but with NaCl replaced by sodium gluconate. Composition of the other test solutions is described in"Materials and Methods."

Test solution Exchanger evaluated

H+ efflux (mm min'1)

MGH Ul EMT-6

NaCl solutionNaCl solution + amiloride or EIPANaHCO3 solutionNaHCO3 solution + amiloride or EÕPANaHCO3 solution + amiloride + DIDSNa gluconate solution + amilorideNa*/H+

Inhibitor independent activityNa+/H+ & Na+-dep HCO3/C(-Na+-dep HCO3/C1Inhibitor independent activityNa*-dep HCO3/C1-, Cl~ gradient5.43

±0.390.78 ±0.05"

14.44 ±0.617.22 ±0.63"0.32 ±0.25"- c

10.36 ±0.25°3.91

±0.140.20 ±0.03*7.26 ±0.412.86 + 0.31*0.44 ±0.13*-c4.40 ±0.29*"

Amiloride, 100 Mm.*EIPA, 10 MM.CDIDS, 100 MM.

12

!•X 6aI 4

V 2 -

a)

l ' \ ' l ' l ' \

6.2 6.4 6.6 6.8 7.0 7.2

Intracellular pH

2 6.4 6.6 6.8 7.0 7.2

Intracellular pHFig. 1. Rate of H+ efflux due to the activity of the Na+/H+ exchanger (o) and

the Na+-dependent Ch/HCCrj exchanger (A) ¡nMGH Ul cells (a) and EMT-6

cells (A).After acidification with NH4C1 to the pH¡indicated, extracellular bufferwas replaced with solution at pH 7.4 and with composition as described in"Materials and Methods." Intracellular alkalinization over the first minute was

then measured. Each point represents a separate experiment.

activity of the exchanger in both cell lines. The observedchanges in activity of the exchanger could be due to a directeffect of pH¡;alternatively, changes in the transmembrane pHgradient that accompany changes in pH¡could account for theresults (see below).

The level of pH¡also influenced the operation of the Na+-dependent C1~/HCO J exchanger with maximum rates of oper

ation in the range of 6.6-7.0 when pHe was 7.4 (Fig. 1). In bothcell lines, at levels of pH¡close to neutrality, the activity of theNa+-dependent Cl~/HCOj exchanger was greater than that ofthe Na+/H+ antiport.

The activity of the Na+-dependent Cl~/HCOj exchanger in

MGH Ul cells was also examined as a function of the extracellular [HCOj] (Fig. 2) and [Na"1"](data not shown). The ex

changer obeyed Michaelis-Menten kinetics with respect to bothof these substrates. For HCOä, Vmaxwas 8.7 IÎIMH+/min, andAn, was 6.0 mM HCOj. For Na+, the value of Fmaxwas 9.9 HIMH+/min, and Km was 76 HIMNa"1";thus the exchanger is essentially independent of [Na"1"]under any conditions that might beencountered in vivo ([Na+] > 110 HIM).

Effects of pHe on the Operation of the Exchangers. Experiments were undertaken to determine the effect of reduced levelsof pHc, such as might be found in regions of solid tumors, onthe operation of both the Na^-dependent Cl~/HCOj exchangerand the Na^/H"1" antiport. In MGH Ul cells following intra

cellular acidification to pH¡6.4-6.5, there was a decrease in thecombined rate of operation of both exchangers as pHe wasreduced over the range 7.4-6.6 (Fig. 3a). This decrease wasfound to be due largely to a decrease in the rate of operation of

I6

O

s/f

'!)

101

i

20Extracellular30[HC03]1

i

40(mM)i 50

Fig. 2. Rate of H+ efflux by the Na+-dependent C1~/HCO5 exchanger in MGH

Ul cells, as a function of extracellular [HCOj]. Cells were acidified to pH¡6.4-6.6with nigericin in NMG solution, and the extracellular buffer was then replacedwith NaHCO3 solution, pH 7.4, containing 100 MMamiloride and with [HCOj] asindicated. The maximum rate of the ensuing alkalinization was measured. Points,mean of at least three experiments; bars, SE.

4443

on April 9, 2021. © 1992 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

REGULATION OF INTRACELLULAR pH IN TUMOR CELLS

16 -i

6.4 6.6 6.8 7.0 7.2 7.4

Extracellular pH7.6

8

6

ÃœJ 2

b)

6.4 6.6 6.8 7.0 7.2 7.4Extracellular pH

7.6

cE 6

X

UJ•¿�f

a:

4

2

e)

6.4 6.6 6.8 7.0 7.2 7.4 7.6

Extracellular pHFig. 3. a and c, rate of H+ efflux due to combined activity of both exchangers

(0) or of the Na"7H+ exchanger (D) and the Na*-dependent Cl /HCOj exchanger (A) operating independently, as a function of pH, in MGH Ul (a) andEMT-6 (c) cells. After intracellular acidification to pH¡6.4-6.6 with nigericin inNMG solution, extracellular buffer was replaced with solutions at the pH indicated and with composition as described in "Materials and Methods." The max

imal rate of alkalinization was measured. Points, mean of at least three experiments; bars, SE. b, rate of H+ efflux due to activity of the Na+-dependent Cl~/

HCOj exchanger in MGH Ul cells, as a function of pHe. After acidification topli, 6.4-6.6 with nigericin in NMG solution, extracellular buffer was replacedwith NaHCOj solution that had been prepared in order to maintain [11('( ) ,|,. at

25 mM at the pH indicated (A) or with regular NaHCO3 solution (A). Themaximal rate of alkalinization was measured. Points, mean of at least threeexperiments; bars, SE.

the Na+-dependent Cl~/HCOj exchanger. As pHe falls, the

concentration of HCOj ions also falls as predicted by theHenderson-Hasselbalch equation. Solutions were therefore pre

pared at different pHe, but [HCOj] was maintained at 25 mM.Reduction of the activity of the exchanger was noted even whensuch correction was made, although the magnitude of the decrease was smaller (Fig. 3b). Activity of the Na+/H+ exchanger

was also reduced at levels of pHe below 7.0, but the decrease wasless than that of the Na+-dependent Cl"/HCOj exchanger (Fig.

3fl). Similar changes were noted in EMT-6 cells (Fig. 3c).The decrease in activity of both exchangers at reduced pHe

could be due to a decrease in the H+ gradient across the cell

membrane; alternatively, low levels of pHc might directly influence the operation of the exchangers, in a manner analogous tothat of pH¡.In an attempt to differentiate between these twopossibilities, experiments were carried out with MGH Ul cellswhere the same transmembrane H+ gradient was produced at

different levels of pHe by acidifying the cytoplasm to either pH6.5 or 7.0 (Fig. 4).

The combined activity of both exchangers increased as thetransmembrane H+ gradient increased and, for a given gradient,

was greater at pH¡7.0 than pH¡6.5 (Fig. 4a) (i.e., greater athigher values of pHe). After acidification to pH¡7.0, recovery ofpH¡was due mainly to the operation of the Na+-dependentCl~/HCOj exchanger (Fig. 4b). By contrast, after cells wereacidified to pH¡6.5, both the Na+/H+ exchanger and the Na+-dependent Cl~/HCOj exchanger were active (Fig. 4c), and their

relative contribution to pH regulation depended on the trans-membrane pH gradient.

Effects of Hypoxia and Extracellular Calcium Concentration.Since regions of tumors which have reduced levels of pHc maybe hypoxic, experiments were performed to examine the abilityof cells to regulate pH¡under hypoxic conditions. After 6 h ofhypoxia (confirmed by resistance to radiation), the viability ofcells was not reduced (data not shown). The combined rate ofoperation of both the Na+/H+ antiport and the Na+-dependentCl~/HCOj exchanger did not differ from values obtained for

aerobic EMT-6 or MGH Ul cells (Fig. 5a).Over the range 1-6 mM, extracellular [Ca2+] did not affect the

ability of MGH Ul cells to recover from an acid load (Fig. 5b).Effects of Extracellular Lactate Concentration. The presence

of láclatein the extracellular buffer caused a Na+-independent

alkalinization of the cytoplasm, after intracellular acidification,presumably because lactate can enter the cell and combine withcytoplasmic H+. This effect could be blocked partly by themonocarboxylate transport inhibitor, a-hydroxy cinnamic acid(Fig. 6a). The magnitude of the lactate-induced recovery of pH¡was small, however (<15% for 5 HIMlactate), when compared tothe rate of operation of the Na+/H+ exchanger and the Na+-dependent Cl~/HCOj exchanger.

In the presence of Na+ and HCOj, extracellular lactate resulted in a small, concentration-dependent decrease in the rateof cytoplasmic alkalinization (Fig. 6b). The presence of a-hy-droxy cinnamic acid resulted in a further slight fall in H+ efflux.

Even in the presence of 10 HTMlactate, pH¡recovery proceededat more than 80% of control values.

Lactate thus had two separate effects. In the absence of Na+

and HCOj it produced a small intracellular alkalinization (Fig.6a). In the presence of these ions, it caused a decrease in thecombined rate of operation of the Na+/H+ and Na+-dependentCl~/HCOj exchangers, although the alkalinizing effect of

lactate was still present (compare plots with and without

on April 9, 2021. © 1992 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

.5 20

6

10

UJ

+3:

REGULATION OF INTRACELLULAR pH IN TUMOR CELLS

16

È 12

14

12

I ' I ' 1 ' 1 r-

0.0 0.2 0.4 0.6 0.8 1.0 1.2

Gradient

b)

—¿�T-

Uj

0.0 0.2 0.4 O.B

Gradient0.8 1.0

0.0 0.2

GradientFig. 4. a, rate of H+ efflux due to the combined activity of both the Na+/H+

and the Na+-dependent CI~/HCOa exchangers in MGH Ul cells, expressed as afunction of the transmembrane pH gradient (pHc-pH¡).After acidification to pH¡6.5 ('"•)or pH¡7.0 (*) extracellular buffer was replaced with NaHCOj of varying

pi I,., to produce the indicated gradient. The maximal rate of alkalinization wasmeasured. Points, mean of at least three experiments; bars, SE. b and c, rate of H+efflux due to the combined activity of the Na+/H+ exchanger (•,D) and theNa+-dependent Cl /HCOj exchanger (A, A) in MGH Ul cells, expressed as afunction of the transmembrane pH gradient (pHc-pH¡).After acidification to pH¡6.5 (c) or 7.0 (ft) extracellular buffer was replaced with solutions with compositionas described in "Materials and Methods" and of varying pHe. to produce the

indicated gradient. The maximal rate of alkalinization was measured. Points.mean of at least three experiments; hut-., SE.

^

S

m

8

18

Aerobic Hypoxie

MGH U1

b)

Aerobic Hypoxie

EMT-6

,5

,E

UJ

9

0 —¿�l ' 1— —¿�I—

02468Extracellular [Ca2+ ] (mM)

Fig. 5. a, rate of H+ efflux due to the combined activity of both the Na*/H*and the Na+-dependent CI'/HCOj exchangers in aerobic (D) and hypoxic (•)MGH Ul and EMT-6 cells. Results are the mean and SE of at least four experiments, b. rate of H+ efflux due to the combined activity of both the Na+/H * andthe Na+-dependent C1~/HCO5 exchangers in MGH Ul cells, expressed as afunction of extracellular [Ca2+]. After acidification to pH¡6.4-6.6, extracellularbuffer was replaced with NaHCOj solution containing the indicated concentration of Ca2^. The maximal rate of alkalinization was measured. Points, mean ofat least three experiments: bars, SE.

a-hydroxy cinnamic acid in Fig. 6b). The net effect was a small(<20%) fall in the combined activity of the exchangers.

DISCUSSION

The experiments reported here demonstrate that pH¡regulation in MGH Ul human bladder carcinoma and EMT-6 murine

mammary sarcoma cells is the result of the combined action ofboth the Na+/H+ antiport and the Na+-dependent Cl^/HCOjexchanger. Hypoxia and alterations in extracellular [Ca2+] had

no effect on the activity of these exchangers, while low levels ofpHe and elevations in extracellular láclate ion concentraliondecreased their rate of operation. These conditions may exist insome regions of solid tumors (11).

The Na+/H+ exchanger appears to be present in all mamma

lian cells (5, 22, 23). In addition to contributing to the recoveryof pH; following an acid load, a role for this anliporter inmitogenesis and volume regulalion has been suggested (4, 24).Our results demonstrate that this antiporter is inactive at valuesof pH¡close to the normal resting pH¡,increases in activity as

4445

on April 9, 2021. © 1992 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

REGULATION OF INTRACELLULAR pH IN TUMOR CELLS

3 -

n ,

LU•¿�*•a:

I

Control Laciaie1mM

Laciate5mM

Lac\a\e10mM

16

c1 12

8 -

2 4 6 8 10

Extracellular [Láclate] (mM)

12

Fig. 6. a, rate of Na* and HCOj-independent H* efflux in MGH \J\ cells

exposed to different extracellular concentrations of láclatein NMG solution afteracidification to pH¡6.4-6.6. Experiments were performed in the presence (•)andabsence (o) of a-hydroxy cinnamic acid. Results are the mean and SE error of atleast three experiments, b, rate of H+ efflux due to the combined activity of boththe NaVH* and the Na+-depcndent CI~/HCO5 exchangers in MGH U\ cells,

expressed as a function of extracellular láclateconcentration, in the presence (*)and absence ( ) of a-hydroxy cinnamic acid. After acidification to [ill, 6.4-6.6,the extracellular buffer was replaced with NaHCOj solution containing the indicated concentration of láclate.The maximal rate of alkalinization was measured.Points, mean of at least three experiments; oars, SE.

pHj falls, and once again becomes inactive at very low levels ofpH¡.The fall in activity observed at pH¡below 6.2 is of uncertain physiological significance, however, since the levels of pH¡involved rarely occur in vivo, even within the acidic areas ofsolid tumors. Data compiled by Wike-Hooley et al. (12) showthat less than 5% of measurements of pHe in human solidtumors have values below 6.2, and viable cells in these acidregions almost certainly maintain higher values of pH¡.

Na+-dependenl CT/HCOa exchange activity has been dem

onstrated previously in several mammalian cell lines, althoughnot every cell line possesses this exchanger (9, 10, 25-27).Although both cell lines tested by us possessed Na+-dependentCl~/HCOj exchange activity, there were substantial differences

in the contributions to pH¡regulation that were due to theoperation of this exchanger. In MGH Ul cells at pH¡below 7.0(when the Na^/H* antiport is active), up to 65% of total H+efflux was bicarbonate dependent; in EMT-6 cells, by contrast,

Na+-dependent Cl~/HCOj exchange accounted for only one-third of the H+ efflux below pH¡6.8.

Activity of the Na+-dependenl Cl~/HCOj exchanger was dependent on pH¡.As for the Na+/H+ antiport, the activity of theNa+-dependent Cl'/HCOj exchanger decreased as the normal

resting value of pH¡was approached. In both cell lines, theNa+-dependenl Cl'/HCOj exchanger remained active athigher levels of pH¡than the Na+/H+ antiporter. This resultsuggests that the Na+-dependenl Cl~/HCOj exchanger may be

involved with the regulation of pH¡at levels close to the normalresting pH¡and is in keeping with previous findings (7). However, in MGH Ul, resting pH¡was no higher in the presence ofHCOj than in its absence, while in EMT-6 it was slightly lower.The presence and activity of a Na^-independenl anión ex

changer, as has been described in many other cell lines, probably accounts for this apparent paradox (8-10).

As pHc was reduced, the ability of cells to recover from anacid load was impaired. At levels of pHe below 7.0, there was amuch greater reduction in the activity of the Na+-dependentCr/HCOj exchanger than that of the Na+/H+ antiport. Since

the pHe within regions of solid tumors is often in the range of6.5-7.0 (11, 12), our findings indicate that the Na+/H+ ex

changer may play the major role in pH¡regulation in cellssituated in these regions. This result may also explain the observation that variant MGH Ul cells, lacking the Na+/H+ an-

tiport, were unable to grow tumors in vivo, despite possessingNa+-dependent Cl~/HCOj exchange activity (28).

It is not clear from our results whether low pHe itself or theresulting decrease in the transmembrane H* gradient is respon

sible for the decrease in activity of both of the exchangers underconditions of extracellular acidity. Previous studies have demonstrated that increased extracellular [H+] (lower pHe) resultsin a decrease in the activity of the Na+/H+ antiport (5, 23, 29).

The results of these studies are consistent with direct competition of H+ for the external Na+ binding site of the transporter

(5).We observed a decrease in the rate of operation of the Na+-

dependent Cl~/HCOj exchanger as the level of pHe was re

duced, and the fall in activity of this transporter was of greatermagnitude than the fall in activity of the Na+/H+ antiporter.

Competition of extracellular protons for binding sites on thetransporter may contribute to the observed fall in activity in amanner analogous to that of the Na+/H+ exchanger. Our re

sults also indicate, however, that the fall in extracellular[HCOj] that accompanies the reduction in pHe plays a part inthe overall decrease in activity.

Regions of hypoxia tend to develop within solid tumors as aresult of a poorly developed vasculature (13). Cells in hypoxicregions are dependent on anaerobic glycolysis for energy production and thus produce láclateas the end product of energymetabolism. We therefore carried out experiments to examinethe influence of hypoxia and láclateion concentration on theregulation of pH¡.Hypoxia for 6 h does not cause cell dealh anddid not result in any alteralion in ihe regulalion of pH¡.

Al láclaleion concentrations above 5 mM, Ihere was a smalldecrease in ihe combined rale of operalion of the Na+/H+ an-liporl and ihe Na+-dependenl Cl~/HCOj exchanger. The con

centration of láclalein ihe inlrerslilial fluid of an experimenlallumor model in the rat has been measured and found to be 17niM (30). However, the arterial láclale concenlralion in thismodel was high (10 mM) compared to values measured in pa-lienls wilh melasi al ¡ccolon cancer (<1 HIM)(31), and ihe láclale concentration in ihe interstilal fluid of human lumors is

4446

on April 9, 2021. © 1992 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

REGULATION OF INTRACELLULAR pH IN TUMOR CELLS

likely to be lower. Thus the impact of lactate on the regulationof pH¡is likely to be small.

Our experiments have examined the influence of acutechanges in the microenvironment on the regulation of pH¡intumor cells. Cells in some regions of solid tumors in vivo may beexposed chronically to such changes, including elevations ofpCO2 to values in excess of 80 mm Hg (32). High levels ofpCO2 such as this are associated with increased amounts ofHCOj, as well as with a decrease in pHe. Such an increase inthe [HCÛ3]e may enhance the activity of the Na+-dependentCl~/HCOj exchanger, although our results indicate that even if

[HCOj]e is maintained at a constant value, reduction in thelevel pHe causes a decrease in the activity of this exchanger.

Our findings have implications for the development of strategies designed to selectively kill tumor cells in an acidic microenvironment (15). Several such strategies have been devised,and to date these have focused on the use of an intracellularacidifying agent in combination with drugs that inhibit the operation of either the NaVH+ antiport or the Na+-dependentCl~/HCOj exchanger (11, 33). Similar approaches have also

been used to enhance the effects of hyperthermia (16). Ourfindings suggest that in the cells that are the target for this typeof therapy, Na+/H+ exchange is the major mechanism for the

regulation of pH¡.The use of agents such as amiloride or EIPAin conjunction with acidifying agents may result in the death ofcells in these regions. Since Na+-dependent Cl~/HCOj ex

change contributes relatively little to pH¡regulation in theseregions, the use of drugs to inhibit the operation of this exchanger might even reduce the therapeutic index, since the toxiceffects on tissues where pHe (and hence the operation of thisexchanger) is normal may be greater than the therapeutic effectsin the tumor. Furthermore, the findings that pH¡regulation iseither normal or only minimally impaired under conditions ofhypoxia or elevated extracellular lactate concentration implythat these environmental conditions, which are also present inthe acidic regions of solid tumors, are unlikely to interfere withthis antitumor strategy. Thus we conclude that mechanismswhich regulate pH¡are appropriate targets for tumor-selectivetherapy.

REFERENCES

1. Frelin, C, Vigne, P., Ladoux, A., and Lazdunski, M. The regulation of theintracellular pH in cells from vertebrates. Eur. J. Biochem., 174:3-14, 1988.

2. Madshus, I. H. Regulation of intracellular pH in eukaryotic cells. Biochem.J., 250: 1-8, 1988.

3. Sardet, C, Franchi, A., and Pouyssegur, J. Molecular cloning, primary structure, and expression of the human growth factor-activatable Na+/H+ anti-porter. Cell, 56: 271-280, 1989.

4. Grinstein, S., and Rothstein, A. Mechanisms of regulation of the Na+/H*exchanger. J. Membr. Biol., 90: 1-12, 1986.

5. Aronson, P. S. Kinetic properties of the plasma membrane Na+-H+ exchanger. Annu. Rev. Physiol., 47: 545-560, 1985.

6. Pouyssegur, J., Sardet, C., Franchi, A., I.'Allemain. G., and Paris, S. Aspecific mutation abolishing NaVH* antiport activity in hamster fibroblastsprecludes growth at neutral and acidic pH. Proc. Nati. Acad. Sci. USA, 81:4833-4837, 1984.

7. Tonnessen, T. I., Sandvig, K., and Olsnes, S. Role of Na+-H+ and CI^-HCOj

antiports in the regulation of cytosolic pH near neutrality. Am. J. Physiol.,25Ä.-C1I17-C1126, 1990.

8. Ganz, M. B., Boyarsky, G., Sterzel, R. B., and Boron, W. F. Arginine vaso-pressin enhances pH, regulation in the presence of HCO5 by stimulatingthree acid base transport systems. Nature (Lond.), 337: 648-651, 1989.

9. Tonnessen, T. I.. Ludi, J., Sandvig, K., and Olsnes, S. Bicarbonate/chlorideantiport in Vero cells: 1. Evidence for both sodium-linked and sodium-independent exchange. J. Cell. Physiol., 132: 183-191, 1987.

10. Cassel. D., Scharf, O., Rotman, M., Cragoe, E. J., Jr., and Katz, M. Characterization of Na*-linked and Na*-independent Cl /HCOj exchange systems in Chinese hamster lung fibroblasts. J. Biol. Chem., 263: 6122-6127,1988.

11. Vaupel, P., Kallinowski, F., and Okunieff, P. Blood flow, oxygen and nutrientsupply, and metabolic microenvironment of human tumors: a review. CancerRes., 49: 6449-6465. 1989.

12. Wike-Hooley, J. L., Haveman, J., and Reinhold, H. S. The relevance oftumour pH to the treatment of malignant disease. Radiother. Oncol., 2:343-366, 1984.

13. Tannock, I. F. Oxygen diffusion and the distribution of cellular radiosensi-tivity in tumours. Br. J. Radio!., 45: 515-524, 1972.

14. Rotin. D., Robinson, B., and Tannock, I. F. Influence of hypoxia and anacidic environment on the metabolism and viability of cultured cells: potential implications for cell death in tumors. Cancer Res., 46: 2821 -2826, 1986.

15. Tannock, I. F., and Rotin, D. Acid pH in tumors and its potential fortherapeutic exploitation. Cancer Res., 49: 4373-4384, 1989.

16. Lyons, J. C., Kim, G. E., and Song, C. W. Modification of intracellular pHand thermosensitivity. Radial. Res., 129: 79-87, 1992.

17. Chu, G. L., Wang, Z., Hyun, W. C, Pershadsinngh, M. J., Fulwyer, M. J.,and Dewey, W. C. The role of intracellular pH and its variance in low pHsensitization of killing by hyperthermia. Radiât.Res., 122: 288-293, 1990.

18. Cragoe, E. J.. Jr.. Woltersdorf, O. W., Jr., Bicking, J. B., Kwong, S. F., andJones, J. H. Pyrazine diuretics. II. A'-Amidino-3-amino-5-substituted 6-ha-lopyrazinecarboxamides. J. Med. Chem., 10: 66-75, 1967.

19. Rink, T. J., Tsien, R. Y., and Pozzan, T. Cytoplasmic pH and free Mg2+ inlymphocytes. J. Cell Biol., 95: 189-196, 1982.

20. Thomas, J. A., Buchsbaum, R. N., Zimniak, A., and Racker, E. IntracellularpH measurements in Ehrlich ascites tumor cells utilizing spectroscopicprobes generated in situ. Biochemistry, 18: 2210-2218, 1979.

21. Boron, W. F. Cellular buffering and intracellular pH. In: D. W. Seldin and G.Giebisch (eds.), The Regulation of Acid-Base Balance, pp. 33-56. New York:Raven Press, 1989.

22. Roos, A., and Boron, W. F. Intracellular pH. Physiol. Rev., 61: 296-434,1981.

23. Grinstein, S., Cohen, S., and Rothstein, A. Cytoplasmic pH regulation inthymic lymphocytes by an amiloride-sensitive Na*/H+ antiport. J. Gen.Physiol., 83: 341-369, 1984.

24. Grinstein, S., Rotin, D., and Mason, M. J. Na+/H+ exchange and growthfactor-induced cytosolic pH changes. Role in cellular proliferation. Biochim.Biophys. Acta, 988: 73-97, 1989.

25. Reinertsen, K. V., Tonnessen, T. I., Jacobsen, J., Sandvig, K., and Olsnes, S.Role of chloride/bicarbonate antiport in the control of cytosolic pH: cell-linedifferences in activity and regulation of antiport. J. Biol. Chem., 263:11117-11125, 1988.

26. Gleeson, D.. Smith, N. D., and Boyer, J. L. Bicarbonate-dependent and-independent intracellular pH regulatory mechanisms in rat hepatocytes. J.Clin. Invest., 84: 312-321, 1989.

27. Gillies, R. J.. and Martinez-Zaguilan, R. Regulation of intracellular pH inBALB/c 3T3 cells: bicarbonate raises pH via NaHCOj/HCl exchange andattenuates the activation of NaVH+ exchange by serum. J. Biol. Chem., 266:1551-1556, 1991.

28. Rotin, D., Steele-Norwood, D., Grinstein, S., and Tannock, I. Requirementof the Na+/H* exchanger for tumor growth. Cancer Res., 49:205-211, 1989.

29. Aronson, P. S., Suhm, M. A., and Nee, J. Interaction of external H* with theNa+-H+ exchanger in renal microvillus membrane vesicles. J. Biol. Chem.,25«:6767-6771, 1983.

30. Cullino, P. M. The internal milieu of tumors. Prog. Exp. Tumor Res., 8:1-25, 1966.

31. Holroyde, C. P., Axelrod, R. S., Skutches, C. L., Haff, A. C., Paul, P., andReichard, G. A. Lactate metabolism in patients with metastatic colorectalcancer. Cancer Res., 39: 4900-4904, 1979.

32. Cullino, P. M., Grantham. F. H., Smith, S. H., and Haggerty, A. C. Modifications of the acid-base status of the internal milieu of tumors. J. Nati.Cancer Inst., 34: 857-869, 1965.

33. Newell, K. J.. and Tannock, I. F. Reduction of intracellular pH as a possiblemechanism for killing cells in acidic regions of solid tumors: effects of rarbonylcyanide-3-chlorophenylhydrazone. Cancer Res., 49: 4477-4482, 1989.

4447

on April 9, 2021. © 1992 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

1992;52:4441-4447. Cancer Res   Michael J. Boyer and Ian F. Tannock  Microenvironmental ConditionsRegulation of Intracellular pH in Tumor Cell Lines: Influence of

  Updated version

  http://cancerres.aacrjournals.org/content/52/16/4441

Access the most recent version of this article at:

   

   

   

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

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

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

  Permissions

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

.http://cancerres.aacrjournals.org/content/52/16/4441To request permission to re-use all or part of this article, use this link

on April 9, 2021. © 1992 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from