[CANCER RESEARCH 47, 2407-2412, May 1, 1987]

Role of Hypoxia in Anticancer Drug-induced Cytotoxicity for Ehrlich Ascites Cells1

Vicram Gupta,2 and John J. Costanzi

Division of Hematology-Oncology, Department of Internal Medicine, The University of Texas Medical Branch, Galveston, Texas 77550

ABSTRACT

A review of the literature on the effect of hypoxia on in vitro drugsensitivity had suggested that there was consistently more cytotoxicityunder hypoxic conditions for the drugs misonidazole and mitomycin Cwhile there was much conflicting data for the drugs Adriamycin andbleomycin. We have examined the effect of oxygen on the cellularresponse of Ehrlich's ascites tumor cells to the drugs mitomycin C,

misonidazole, Adriamycin, and bleomycin. Significant differences wereobserved when we compared the cytotoxicity of mitomycin C and misonidazole as a function of oxygen concentration. For the drugs Adriamycinand bleomycin no differential effects of oxygen were noted for a 1-h drugexposure with hypoxia while some differences were noted only for bleomycin for an 8-h drug exposure time. Because differences dependent onoxygen concentration were observed for some drugs but not others at thesame experimental conditions, and as indicated by a review of theliterature, it is suggested that some of the conflicting data in the literaturewith respect to some of these drugs may be cell-line dependent. Othervariables which may also be of importance were the duration of drugexposure time in hypoxia and cell density. The observed oxygen concentration-dependent changes in cell survival for the Ehrlich cells with drugsexamined could not be explained on the basis of changes in drug-inducedcellular oxygen consumption.

INTRODUCTION

Oxygen microenvironment in tumors is heterogeneous (1,2).It is known that cell killing by radiation is dependent uponoxygen concentration and that hypoxic cells are resistant toradiation (3). Hypoxic cells may also be resistant to drugsbecause of drug delivery problems secondary to inadequatevasculature (4) and/or intrinsic effects of hypoxia on cell response to drugs (5). There is some evidence that oxygen concentration may play a significant role in drug-induced cellkilling (5). Drugs such as mitomycin C and misonidazole areknown to be selectively metabolized under hypoxic conditionsto active compounds (6, 7) while drugs such as Adriamycin andbleomycin may be cytotoxic through oxygen radical-mediatedmechanisms (8, 9). Pre-, during, and postdrug treatment influences of varying oxygen concentrations have all been describedin the literature (10,11). The effect of oxygen concentration onthe cytotoxicity of misonidazole has been established (12, 13)and similar data with mitomycin C have been recently published(14). However, the in vitro data are conflicting with the exception of misonidazole. Because of the conflicting data summarized in Table 1 we have examined the effect of oxygen concentration on cell survival for the anticancer agents mitomycin C,misonidazole, Adriamycin, and bleomycin.

MATERIALS AND METHODS

Tumor line used for the experiments was a tetraploid Ehrlich's ascites

tumor (Mason Research Institute, Worcester, MA) which was main-

Received4/9/86; revised 12/2/86; accepted2/4/87.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 inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1Supported in part by USPHS Grant CA-32938, National Cancer Institute(NCI), and Cancer Center Grant CA-17701, NCI.

1To whom requests for reprints should be addressed, at Division of Hematol-ogy, Oncology/Internal Medicine Department, Clay Hall, H82, University ofTexas Medical Branch, Galveston, TX 77550.

tained by i.p. injection of 1 x IO7cells weekly in ICR white mice. Cells

from animals bearing 6-8-day old ascites tumors were used.In vitro studies on the effect of oxygen on drug-induced cell survival

were performed as follows. EAT3 cells (1-5 x IO6) were suspended in10 ml of Hams FIO media in 25-ml Erlynmyer glass flasks. The flaskswere sealed with rubber sleeve stoppers and fitted with two 21-gaugeneedles for an inlet and outlet port. The flasks were gassed with oxygenmixture of interest at flow rates of 1 liter/min for 1.5 h at 37*C on an

Orbital Shaker, model 3520 (Lab-Line Instruments, Inc.) at 1800 rpm.This is similar to the method reported by Mohindra and Rauth (15).Gas mixtures used contained 0-20% O2, 5% CO2, and balance N2(Linde Corp., Houston, TX). Following this prééquilibraiion of oxygen,drugs were added directly without opening the flask by using a 21-gauge needle in a 0.1 -ml volume. The effect of both short and prolongeddrug exposure time during hypoxia on drug cytotoxicity was examinedin some cases. For the 1-h drug exposures a cell density of 1 x 105/ml

was used without humidification. Measurement of volume of mediafailed to reveal any significant effect of evaporation. For the 8-h drugexposure under hypoxia, a cell density of 5 x KlVnil was used withhumidification as this time period would have resulted in significantevaporation. In some instances cells were exposed to a humidifiedpreincubation atmosphere of 5% O2 (oxic) or N2 (hypoxic) for 16.5 hbefore drug incubation for l h at the stated oxygen atmosphere in orderto determine whether the duration of hypoxia before drug exposureinfluences cytotoxicity. After drug incubation with continuous gassing,the cells were centrifuged, washed with 10 ml of media, and resuspendedin fresh Hams FIO media for plating. The colony assay for the EATcells has been described previously (32). Tumor cells were suspendedin Hams FIO media containing 15% fetal calf serum, 1% penicillin-streptomycin, and 0.3% agar and plated in 35-mm l'etri dishes in a 2-

ml volume. The dishes were then incubated in humidified ModulatorIncubator Chambers (Billup-Rothenberg Inc., Del Mar, CA) andflushed with 5% O2, 5% CO2, and 90% N2. A 5% O2 atmosphere wasused for the plating because we had previously found there was nogrowth of the EAT cells at 20% ( ).-in the absence of antioxidants (32).Seven to 8 days later, plates were scored for colony formation (>50cells).

Oxygen measurements in solution were performed using a DiamondElectrotech, Inc. (Ann Arbor, MI) oxygen measuring system whichconsisted of an amplifier unit (no. 1201), calibration cell (no. 1251),and Clarke type oxygen electrode. Oxygen measurements were madein separate experiments with and without cells similar to the drugexperiments except that phosphate buffered saline was used. For eachexperiment a separate calibration curve using N2 and 0.1, 1, 5, 10, and21% O2 was performed. The current was recorded as a function of timeand the oxygen concentration calculated from the calibration curve.Using this system, oxygen level below 0.1 % cannot be measured accurately. The data was plotted according to an equation (Equation A)described by Whillans and Rauth (33):

(Cm - CJ = (Cw - C.)e-*" (A)

where C\,\ is the oxygen tension in solution at time t. f H is the initialoxygen tension in solution, C, is the oxygen concentration at equilibrium, and *i (s ') is a physical constant independent of oxygen concen

tration and depletion rate. From the measured values of oxygen consumption rate (R) and k,, one can calculate the effect of oxygendepletion on the oxygen concentration at equilibrium using the relationship C. = CM - R/ki (33) where C[girepresents oxygen concentra

tion in the gas phase.Oxygen consumption studies were performed on a Gilson Oxygraph

located in the Division of Radiation Research, University of Texas

1The abbreviation used is: EAT, Ehrlich ascites tumor cells.

2407

on May 4, 2020. © 1987 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

HYPOXIA AND DRUG CYTOTOXICITY

Table 1 Summary of literature on the effect of oxygen concentration on drug cytotoxicily

Drug

Preferentiallytoxic to

oxic cellsPreferentially toxic

to hypoxic cells No difference

MisonidazoleMetronidazoleMitomycin C

ProcarbazineActinomycin DVincristineStreptonigrinBleomycin

Adriamycin

BisantreneDTICCyclophosphamideBCNU

Cisplatin

5-FluorouracilMethotrexateEtoposideAlkeranNitrogen mustard

CHO"(12)EMT6(13)

CHO and HeLa (15)EMT6 (10, 16-19), CHO (14, 20), V79,

KHT, HeLa (20), S180 (17), EAT*,and P3884

EMT6 (10)EMT6 (10, 18)EMT6 (10)EMT6 (17)EMT6 (10)V79 (22)CHO (23)EAT*

B14 FAF28 (25)V79 (26)CHO (23)WHF1B(27)L1210, HTC (28)

EMT6 (31)

EMT6(10, 17)S180(I7)

CHO (29)V79 (30)

EMT6 (18)

EMT608)EMT6 (18)

CHO (16)WiDR and HEC-1A (19)HTC (19)

EMT6 (21)EMT6 (21)

EMT6 (24)

EMT6(21)EAT*

CHO (23)CHO (23)EMT6(10, 21)S 1X0(17)EMT6 (10, 21)CHO (23)EMT6 (10, 21)EMT6(10, 21)

°CHO, Chinese hamster ovary; DTIC, 5-(3,3-dimethyl-l-triereno)imidazole-4-carboxamide; BCNU, l,3-bis(2-chloroethyl)-l-nitrosourea; EMT6, mouse mammarytumor cells; V79, B14 FAF28, Chinese hamster, S180, KHT, mouse sarcoma cells; P388 and 1.1210. mouse leukemic cells; HeLa, WiDR, HEC-1 A, human epithelialcancer cell lines; HTC, human tumor biopsy cells.

* This study.

Medical Branch, Galveston, TX, through the courtesy of Dr. NeilI loud I. Oxygen consumption rates at 21% O2 were determined withand without the drugs Adriamycin, bleomycin, and mitomycin C at SOMg/ml directly inside the oxygen consumption chamber. In addition,mitomycin C (3 /ig/ml)- and misonidazole (1000 ¿tg/ml)-treatedcellsat 0.01% (>., for l h were washed free of drug, resuspended in media,and the <>, consumption rates were determined.

RESULTS

Measurements of oxygen tension in solution without cellswith time were used to determine the physical constant k¡according to data in Fig. 1. A kt value of 0.55 min"' with a

half-time of 1.26 min for equilibration was obtained. The former value is considerably higher than previously reported systems (33, 34) and reflects the high degree of gas exchangeoccurring in our system. This can be explained on the basis ofgeometry of vessel used (33, 34), and the high rate of shakingcausing sufficient dispersion of the fluid over the glass surfaceand thereby decreasing the depth of the fluid.

The respiratory rate of the EAT cells at 37°Cwas 1.312 ±0.179 nmol O2/min/106 cells. This is similar to the previouslyquoted respiratory rate of about 0.1% O2/min/106 cells for

these cells (34, 35).From the experimental values of k\ and R, the theoretical

oxygen level in solution at equilibrium (C„)was calculated fora cell density of 1 x 105/ml. Calculated C«,values for C{g]of 20,10, 5, 1, 0.1, and 0.01% O2 were 19.985, 9.985, 4.985, 4.985,0.985,0.085, and 0% O2, respectively. The latter value may notbe correct as the oxygen consumption rate changes in the 0.01-0.1% O2 range (12, 35, 36). At 0.01% O2 the oxygen consumption rate for EAT cells decreases to about 25% of initial value(35). This would result in a C. value of 0.0064% O2 for a 0.01 %O2 atmosphere. At the cell density of 1 x 105/ml used experi-

1.0

0.1oW

u

0.01234

TIME (MIN)

Fig. 1. Data shown is a semilogarithmic plot of Equation A from three separateexperiments.

4 V. Gupta, unpublished observations.

mentally in Figs. 2 and 4, we were unable to detect any significant differences in the measured oxygen levels with or withoutcells for a 1% O2 atmosphere. Similar measurements at 0.1%O2 were not attempted because this represents the limit of ouroxygen measuring system.

The effect of oxygen concentration on the cytotoxicity ofmitomycin C for the EAT cells is shown in Fig. 2. There wasan increase in the cytotoxicity of mitomycin C as the oxygenconcentration was decreased to 0.01% O2. At the higher drugconcentrations of mitomycin C (2 and 3 ng/ml) there was noapparent difference in the cytotoxicity at 0.1% or 0.01% O2.

2408on May 4, 2020. © 1987 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

HYPOXIA AND DRUG CYTOTOXICITY

0.000101 23456

/ng/ml MITOMYCIN-C

Fig. 2. Surviving fraction of 1 x 10*/ml EAT cells treated with mitomycin Cfor l h after preincubation with different oxygen concentrations for 1.5 h. Ban,mean ±SE of three to five experiments.

5 9§SSIE

10.01 0.1 1.0 10 100

% OXYGEN CONCENTRATION

Fig. 3. Relative drug concentration of mitomycin C necessary as a function ofoxygen concentration to achieve equivalent cell killing. Data extrapolated fromFig. 2. The data are expressed as a ratio of drug concentration necessary to reducecell survival to 0.31(0,,) for the various oxygen levels when compared to 0.01%O,.

There was also no difference in survival curves between 0.01%O: or N2 atmosphere (data not shown).

The relative increase in drug concentration of mitomycin Cnecessary to achieve equivalent reduction of cell survival isplotted in Fig. 3 based on the data in Fig. 2. The half-maximalincrease in drug concentration to achieve equivalent cell killingoccurs at 4.0% O2. This is different than what is observed withradiation-induced cell killing where cell survival decreases asthe oxygen concentration is increased with a half-maximal valueof approximately 0.5% O2 (3).

These data with mitomycin C are in contrast to the influenceof oxygen concentration on the cytotoxic effects of misonida-zole in Fig. 4. There is an apparent stringent requirement forextreme hypoxia for the induction of the cytotoxicity of mison-idazole. Differences between survival curves of misonidazoleand mitomycin C are apparent for N2 and 0.01 and 0.1% ().-atmosphere (Figs. 2 and 4). Similar experiments with Adria-mycin and bleomycin (Figs. 5 and 6) failed to reveal significantdifferences in the cytotoxicity between the N2 and 20% O2atmosphere for a 1-h drug exposure. However, significant differences in cell survival were observed for a more prolongeddrug exposure (8 h) during hypoxia for bleomycin (Fig. 7) butnot Adriamycin (Fig. 8) under similar experimental conditions.EAT cells exposed for 16.5 h to a 5% O2 or N2 atmospherebefore treatment with 1 itg/ml of Adriamycin for 1 h underhypoxic and oxic conditions failed to reveal significant differences in cell survival (Table 2). The cell survival values obtainedare comparable to that shown in Fig. 5 for the same drugconcentration but for a much shorter duration of preincubationhypoxia. For cells exposed to a 5% O2 or N2 atmosphere for

o6<ocu.O

ce.co

0.0001

0.01 -

0.001

O 100 500 750 1000

¿ig/mlMISONIDAZOLE

Fig. 4. Same as in Fig. 2 except that misonidazole was used. Points, mean ofquadruplicate plates from one experiment.

gGcelici

Ioc

tn

1.0

0.1

001

0001

0.0001

20% 02

0.5 1.0 1.5 2.0

¿/g/mlADRIAMYCIN

3.0

Fig. 5. Surviving fraction of l x KI'm I EAT cells treated with Adriamycin

for 1 h after preincubation in 20% ( >..or N.. atmosphere for 1.5 h. Points, mean±SE of three to six experiments.

16.5 h (Table 2) differences in the plating efficiency of drug-untreated cells, in cell cycle distribution by flow cytometry, orin uptake of initiated thymidine were not observed (data notshown).

The effects of cell density and pH on bleomycin cytotoxicitywere determined because these variables may affect the cytotoxicity of bleomycin if different cell densities are used underhypoxic and oxic conditions. As can be seen from Table 3, therewere significant differences with the relative survival fractionincreasing with cell density. We were unable to demonstrate asignificant effect of pH on bleomycin cytotoxicity over a rangeof 6.49-7.22 (data not shown).

In order to determine whether some of the differences in thecell survival curves as a function of oxygen concentration withthe drugs mitomycin C and misonidazole were due to changesin drug-induced cell oxygen consumption rates, measurementsof oxygen consumption were performed under conditions similar to that utilized in the experiments. We were unable todemonstrate any significant effects. In addition, we were alsounable to demonstrate significant changes in cellular oxygenconsumption after the addition of extremely high drug concen-

2409

on May 4, 2020. © 1987 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

HYPOXIA AND DRUG CYTOTOXICITY

20% Oj

N,

0.0120 40 60 80

¿jg/mlBLEOMYCIN

100

Fig. 6. Same as in Fig. S except bleomycin was used. Points, mean ±SE ofthree experiments.

0.01

20% 02

4 6

EXPOSURE TIME (MRS)

Fig. 8. Same as in Fig. 7 except that 0.1 Mg/ml of Adriamycin was used.Points, mean ±SE of four experiments.

001246

EXPOSURE TIME (MRS)

Fig. 7. Surviving fraction of 5 x lOVml EAT cells treated with 10 jig/mlbleomycin for 2-8 h drug exposure time after initial l.S-h preincubation time at20% ( ), or Nj. Points, mean ±SE of four experiments. pH measurements underoxic and hypoxic conditions for the 8-h time period varied from 7.13 to 7.22 and7.13 to 7.09, respectively. Differences at 8-h point only were significant (/' <

O.OS3)by analysis of variance.

trations (50 Mg/ml) of Adriamycin, bleomycin, and mitomycinC. These O2 cell consumption rates were 12.78 ±2.14, 13.00±2.12, and 14.30 ±2.65 nmol O2/min/107 cells, respectively,

when compared to a control respiratory rate of 13.12 ±1.79nmol O2/min/107 cells.

DISCUSSION

The drug misonidazole was first studied because there areconsistent data in the literature demonstrating its selectivetoxicity towards hypoxic cells in vitro. In addition, the oxygenconcentration-dependent cytotoxicity of misonidazole had beenestablished and had indicated that it is most toxic at oxygenconcentrations less than 0.1% O2 (12, 13). The data presentedin Fig. 4 contimi these observations and indicate that theexperimental conditions used in the present study were suffi-

Table 2 Effect of prolonged hypoxia on Adriamycin cytotoxicityEAT cells were exposed to the stated O2 atmosphere for 16.5 h with continuous

gassing and then treated with Adriamycin (1 Mg/ml) l h at the same atmosphere.

Experiment1rOxygen concentration5%O2

N25%O2N2pH6.83

6.496.696.50Surviving

fraction0.010

0.0310.0420.036

" No changes in cell cycle distribution by flow cytometry or uptake of initiated

t(umiliine at end of experiment for control cells.

Table 3 Effect of cell density on bleomycin cytotoxicityCells were treated with 50 Mg/rnl bleomycin for l h at 20% O2. At cell densities

of 1 x lO'-lO" and 1 x IO1 the pH was 7.12 and 6.86, respectively, at end of

experiment.

Cell densityRelative increase in

survival fraction1 x IO41 x IO51 x 10*1 x IO7

11.21 ±0.40«

1.68 ±0.084.35 ±2.00

" Mean ±SE of three to four experiments.

cient to demonstrate biological differences in the oxygen concentration-dependent cytotoxicity of anticancer agents.

The data in Figs. 2 and 3 clearly demonstrate that thecytotoxicity of mitomycin C for the EAT cells is highly oxygenconcentration dependent, being more toxic to hypoxic thanwell-oxygenated cells. For the EAT cells there was very littledifference in the cytotoxicity of mitomycin C in the 0.0-0.1%O2 range. These data are in contrast to recently published databy Marshall and Rauth (14). These investigators observed thatmost of the change in toxicity of mitomycin C under hypoxicconditions occurred in the 0.001-1% O2 range for Chinesehamster ovary cells. Moreover, these differences were apparentafter a 5-h drug exposure with very little difference observed at1 h. in contrast to our data with a 1-h drug exposure time. Anexamination of cell survival curves from other studies wouldalso suggest that quantitative differences in the effect of mitomycin C under hypoxic and oxic conditions are in some instances a function of duration of drug exposure time (16, 20).In this regard it has previously been shown that the duration ofhypoxia before drug exposure is not an important variable formitomycin C cytotoxicity (37). Some work published by Ludwig

2410

on May 4, 2020. © 1987 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

HYPOXIA AND DRUG CYTOTOXICITY

et al. (19) would suggest that the enhanced cytotoxicity ofmitomydn C under hypoxic conditions may be cell line dependent. These workers were unable to demonstrate a significant effect of mitomycin C under hypoxic conditions at clinically achievable drug concentrations for human tumor biopsyspecimens, WiDR and HEC-1A human tumor cell lines, whiledemonstrating a significant effect for EMT6 and S180 murinetumor cells. These data, if confirmed, may have significantclinical implications for the use of mitomycin C to selectivelykill hypoxic tumor cells.

The increased metabolism of mitomycin C under hypoxicconditions to alkylating intermediates is well known and formsthe basis for the increased cytotoxicity under hypoxic conditions(5, 6). Although one can demonstrate that rate of formation ofalkylating intermediates is higher under hypoxic than aerobicconditions, there is no quantitative correlation between theseintermediates and the extent of cell killing under hypoxicconditions (16). These data would suggest that factors otherthan, or in addition to, the formation of alkylating intermediates may influence cytotoxicity of mitomycin C under hypoxicconditions. Such reasons may explain the differences in theoxygen concentration dependency curves of mitomycin C between our data and that of Marshall and Rauth (14). In fact,Ludwig et al. (19) in their study were unable to demonstratethe enhanced disappearance of mitomycin C under hypoxicconditions for the human cell line WiDR suggesting that theenzymes responsible for enhanced metabolism under hypoxicconditions may not be present in some human tumor cell lines.

Because hypoxic cells are radiation resistant, the use ofmitomycin C as an adjunct to selectively kill hypoxic cells hasbeen proposed for clinical use (5). However, animal studieshave failed to demonstrate any significant selective cytotoxicityof mitomycin C for hypoxic cells in vivo (20, 38, 39). Theoxygen concentrations in normal and tumor tissues are heterogeneous and reported to be in the range of 2-5% O2 and 0.1-5% O2, respectively (1, 2,40). Estimates of the hypoxic fractionin tumors range from 0-100% (41). These observations and ourdata indicating significant cytotoxicity of mitomycin C at oxicand intermediate levels of hypoxia may help to explain thedifficulty in trying to demonstrate a selective effect of mitomycin C on hypoxic cells in vivo.

Our data raise the question why there are differences in theoxygen concentration dependency between misonidazole andmitomycin C. Misonidazole is metabolized under hypoxic conditions by a nitroreductase (5). The nitroreduction is highlysensitive to inhibition by rather low concentrations of oxygenand the extent of inhibition appears to be drug dependent (42).The metabolism of mitomycin C to reactive alkylating intermediates is enhanced under hypoxic conditions (14, 16). Thesedifferences in drug metabolism between mitomycin C and misonidazole may help to explain the differences in the oxygenconcentration-dependent cytotoxicities observed with thesedrugs. Differences in the oxygen concentration dependencybetween mitomycin C and misonidazole for the EAT cellscannot be explained on the basis of changes in drug-inducedoxygen consumption since we did not observe significantchanges in these measurements.

The duration of hypoxia and drug exposure time (25, 26),cell density (25, 43), pH (25, 30), and proliferative status (44)are some variables which may influence the response of cells todrugs under hypoxic conditions. The conflicting data publishedin the literature (Table 1) would suggest that the cytotoxicityof bleomycin is enhanced under oxygenated conditions for 1-h(10), 4-h (22), and 4-8-h (23) drug exposure time. There aretwo studies where no differences were observed for a 1-2-hdrug exposure time (23, 24). Our data clearly indicate that forthe EAT cells the drug exposure time in hypoxia was important

for the expression of cytotoxic effects of bleomycin. Such cellswere more resistant than oxygenated cells with time. We foundthat the EAT cells were also more resistant to bleomycin as thecell density increased. This is similar to the effects of cell densityon Adriamycin cytotoxicity (25, 43). Since some investigatorshave used different cell densities under oxic and hypoxic conditions, it is possible that this may explain some of the quantitative differences in the literature. Our data and that of theliterature does not explain why hypoxic cells should be moreresistant to bleomycin than oxygenated cells with time. Theobserved differences are difficult to explain on the basis of acellular oxygen radical-mediated mechanism. This is supportedby the fact that differences in the cytotoxic effects of bleomycinfor a 1-h drug exposure time or significant changes in cellularoxygen consumption rates at rather high drug concentrationsof bleomycin were not observed. The cell survival differenceswith bleomycin after 8-h exposure to hypoxia cannot also beexplained on the basis of changes in proliferative characteristicssince EAT cells exposed to 16.5 h of hypoxia (Table 2) did notdisplay significant changes in cell cycle distribution or uptakeof tritiated thymidine.

The data in the literature with respect to the effect of Adriamycin under hypoxic or oxic conditions is most confusing.Some data would suggest that chronically (23, 25, 26) but notacutely hypoxic cells (21, 23, 25, 26) are more resistant toAdriamycin. However, other data would indicate that Adriamycin is more cytotoxic under oxic conditions (10) or that thereare no significant differences in the effect of Adriamycin between acutely and chronically hypoxic cells (21). The reasonfor these differences in results is not clear.

In contrast to our data with bleomycin, we were unable todemonstrate any significant difference in the cytotoxicity ofAdriamycin between oxic and hypoxic conditions for 1-8 h ofdrug exposure time or for cells previously exposed to a 16.5-hpreincubation time in hypoxia (chronic hypoxia). It is possiblethat some of the differences may be related to the induction ofoxygen-regulated proteins which may be of importance in modulating the response to Adriamycin under hypoxic conditions(44, 45). In this regard it is interesting that prolonged exposureto metabolic inhibitors such as 2-deoxyglucose and 2,4-dinitro-phenol which results in increased cell survival to Adriamycin(46) may also be associated with induction of new cell proteinsynthesis (47-49). Because the EAT cells are already growingin vivo under conditions of radiobiological hypoxia (50), it ispossible that when these cells are used for in vitro drug responses that the relative resistance to Adriamycin reportedfollowing a period of prolonged hypoxia may not be observed.In an assessment of in vivo response of hypoxic tumor cells toAdriamycin, Tannock (51) concluded that hypoxic cells werespared by Adriamycin but that this was due to drug deliveryproblems.

We had previously demonstrated an effect of Superoxidedismutase on overcoming the effect of excess oxygen toxicityon the growth of EAT cells (32). Therefore, the effect of oxygenradical scavengers on Adriamycin-induced cell cytotoxicity wasstudied. We were unable to demonstrate any significant effectson Adriamycin cytotoxicity with the oxygen radical scavengersSuperoxide dismutase, catalase, dimethylsulfoxide, and vitaminE.5 Brox et al. have reported that they did not observe significant

differences in DNA strand breakage due to Adriamycin underhypoxic or oxic conditions (52). We have presented data indicating that differences in Adriamycin-induced cytotoxicity under hypoxic conditions or in cell oxygen consumption rateswere not present for EAT cells. These observations suggest thatoxygen radical-mediated mechanisms may not be important for

5V. Gupta, unpublished observations.

2411

on May 4, 2020. © 1987 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

HYPOXIA AND DRUG CYTOTOXICITY

some tumor cells for clinically achievable drug concentrationsof Adriamycin at the cell level.

ACKNOWLEDGMENTS

We thank Dr. Neil Howell for help with the oxygen consumptionmeasurements. Dr. James Hokanson, Biostatistical Services, CancerCenter, performed the statistical analysis. Richard Eberle providedexcellent technical assistance.

REFERENCES

1. Cater, D. B., and Silver, I. A. Quantitative measurements of oxygen tensionin normal tissues and in the tumors of patients before and after radiotherapy.Acta Radiol., 53:233-256, 1960.

2. Vaupel, P. W., Frinak, S., and Sicher, H. I. Heterogeneous oxygen partialpressure and pH distribution in C3H mouse mammary adenocarcinoma.Cancer Res., 41: 2008-2013, 1981.

3. Hall, E. J. Radiobiology for the Radiologist, p. 85. New York: Harper andRow, 1978.

4. Ozols, R. F., Locker, G. Y., Doroshpw, J. H., Grotzinger, K. R., Myers, C.E., and Young, R. C. Pharmacokinetics of Adriamycin and tissue penetrationin murine ovarian cancer. Cancer Res., 39: 3209-3214, 1979.

5. Kennedy, K. A., Teicher, B. A., Rockwell, S., and Salterelli, A. C. Thehypoxic tumor cell: a target for selective cancer chemotherapy. Biochem.Pharmacol., 22: 1-8, 1980.

6. Kennedy, K. A., Rockwell, S. A., and Sartorelli, A. C. Preferential activationof mitomycin C to cytotoxic metabolites by hypoxic tumor cells. Cancer Res.,40:2356-2360, 1980.

7. Josephy, P. D., Palcic, B., and Skarsgard, L. D. In vitro metabolism ofmisonidazole. Br. J. Cancer, 43:443-450, 1981.

8. Bachur, N. R., Gordon, S. L., and Gee, M. V. A general mechanism formicrosomal activation of quinone anti-cancer agents to free radicals. CancerRes., J«:1745-1750.1978.

9. Lown, J. W., and Sim, S. K. The mechanism of the bleomycin-inducedcleavage of DNA. Biochem. Biophys. Res. Commun., 77:1150-1157, 1977.

10. Teicher, B. A., Lazo, J. S., and Sartorelli, A. C. Classification of antineo-plastic agents by their selective toxicities toward oxygenated and hypoxictumor cells. Cancer Res., 41: 73-81, 1981.

11. Gupta, V., and Krishan, A. Effect of oxygen concentration on the growthand drug sensitivity of human melanoma cells in soft-agar clonogenic assay.Cancer Res., 42: 1005-1007, 1982.

12. Taylor, Y. C., and Rauth, A. M. Oxygen tension, cellular respiration, andredox state variables influencing the cytotoxicity of the radiation sensitizermisonidazole. Radiât.Res., 91: 104-123, 1982.

13. Malcahy, R. T. Effect of oxygen on misonidazole chemosensitization andcytotoxicity in vitro. Cancer Res., 44: 4409-4413, 1984.

14. Marshall, R. S., and Rauth. A. M. Modification of the cytotoxic activity ofmitomycin C by oxygen and ascorbic acid in Chinese hamster ovary cells anda repair-deficient mutant. Cancer Res., 46: 2709-2713, 1986.

15. Mohindra. J. K.. and Rauth, A. Increased cell killing by metronidazole andnitrofurazone of hypoxic compared to aerobic mammalian cells. Cancer Res.,30:930-936, 1976.

16. Keyes, S. R., Fracasso, P. M., Heimbrook, D. C., Rockwell, S., Sugar, S. G.,and Sartorelli, A. C. Role of NADPH-cytochrome c reducíaseand DT-diaphorase in the biotransformation of mitomycin C. Cancer Res., 44:5638-5643, 1984.

17. Kennedy, K. A.. Siegfred, J. M.. Sartorelli, A. C., and Tritton, T. R. Effectsof anthracyclines on oxygenated and hypoxic tumor cells. Cancer Res., 43:54-59, 1983.

18. Shrieve, D. C., and Harris, J. W. Effects of gluththione depletion by buthion-ine sulfoximine on the sensitivity of EMT 6/SF cells to chemotherapy agentsor X-radiation. Int. J. Radiât.Oncol. Biol. Phys., 12: 1171-1174, 1986.

19. Ludwig, C. U., Peng, Y. M., Beaudry, J. N., and Salmon, S. S. Cytotoxicityof mitomycin C on clonogenic human carcinoma cells in not enhanced byhypoxia. Cancer Chemother. Pharmacol., 12: 146-150, 1984.

20. Rauth, A. M., Mohindra, J. K., and Tannock, I. F. Activity of mitomycin Cfor aerobic and hypoxic cells in vitro and in vivo. Cancer Res., 43: 4154-4158, 1983.

21. Harris, J. W., and Shrieve, D. C. Effects of adriamycin and X-rays on euoxicand hypoxic EMT-6 cells in vitro. Int. J. Radiât.Oncol. Biol. Phys., 5:1245-1248, 1979.

22. Roizin-Towle, L., and Hall, E. J. The effects of bleomycin on aerated andhypoxic cells in vitro, in combination with irradiation. Int. J. Radiât.Oncol.Biol. Phys., 5: 1491-1494, 1979.

23. Tannock, I., and Guttmann, I. Response of Chinese hamster ovary cells toanti-cancer drugs under aerobic and hypoxic conditions. Br. J. Cancer, 43:245-248, 1981.

24. Shrieve, D. C., and Harris, J. W. Effects of bleomycin and irradiation oneuoxic and hypoxic cells. Int. J. Radiât.Oncol. Biol. Phys. 5: 1495-1498,1979.

25. Born, R., and Eichholtz-Wirth, H. Effect of different physiological conditionson the action of adriamycin on Chinese hamster cells in vitro. Br. J. Cancer,¥¥.-241-246,1981.

26. Smith, E., Stratford, I. J., and Adams, G. E. Cytotoxicity of adriamycin onaerobic and hypoxic Chinese hamster V79 cells in vitro. Br. J. Cancer, 41:568-573, 1980.

27. Martin, W. M. C., and McNally, N. J. Cytotoxicity of adriamycin to tumourcells in vivo and in vitro. Br. J. Cancer, 42:881-889, 1980.

28. Ludwig, C. U., Bowden, G. T., Roberts, R. A., and Alberts, D. S. Reducedbisantrene-induced cytotoxicity and protein-associated DNA strand breaksunder hypoxic conditions. Cancer Treat. Rep., 68: 367-372, 1984.

29. Rauth, A. M., and Mohindra, J. K. Selective toxicity of 5-(3,3-dimethyl-l-triazeno)imidazole-4-carboxamide toward hypoxic mammalian cells. CancerRes., 41: 4900-4905, 1981.

30. Begg, A. C., Shrieve, D. C., Smith, K. A., and Terry, N. H. A. Effects ofhypoxia, pH, and growth stage on cell killing in Chinese hamster V79 cellsin vitro by activated cyclophosphamide. Cancer Res., 45: 3454-3459, 1985.

31. Teicher, B. A., Holden, S. A., and Rose, C. M. Effect of oxygen on thecytotoxicity and antitumor activity of etoposide. J. Nati. Cancer Inst., 75:1129-1133,1985.

32. Gupta, V., and Eberle, R. Modulation of tumor cell colony growth in softagar by oxygen and its mechanism. Br. J. Cancer, 42: 587-593, 1984.

33. Whillans, D. W., and Rauth, A. M. An analysis of changes in oxygen tensionin stirred cellular suspensions under conditions of radiolytic and cellularconsumption. Cancer Clin. Trials, 3:63-67, 1980.

34. Whillans, D. W., and Rauth, A. M. An experimental and analytical study ofoxygen depletion in stirred cell suspensions. Radial. Res., 84:97-114, 1980.

35. Froese, G. The respiration of ascites tumor cells at low oxygen concentrations. Hindi iin. Biophys. Acta, 57: 509-519, 1962.

36. Wilson, D. F., Erecinska, M., Drown, C., and Silver, I. A. The oxygendependence of cellular energy metabolism. Arch. Biochem. Biophys., 795:485-493, 1979.

37. Rockwell, S., Kennedy, K. A., and Sartorelli, A. C. Mitomycin C as aprototype bioreductive alkylating agent: in vitro studies of metabolism andcytotoxicity. Int. J. Radiât.Oncol. Biol. Phys., 8: 753-755, 1982.

38. Rockwell, S., and Kennedy, K. A. Combination therapy with radiation andmitomycin C: preliminary results with EMT 6 tumor cells in vitro and invivo. Int. J. Radiât.Oncol. Biol. Phys., 5: 1673-1676, 1979.

39. Rockwell, S. Effects of mitomycin C alone and in combination with X-rayson EMT6 mouse mammary tumors in vivo. J. Nati. Cancer Inst., 71: 765-771, 1983.

40. Guyton, A. C. Function of the Human Body, pp. 232-233. Philadelphia:Saunders, 1969.

41. Moulder, J. E., and Rockwell, S. Hypoxic fractions of solid tumors: experimental techniques, methods of analysis, and a survey of existing data. Int. J.Radiât.Oncol. Biol. Phys., 10:695-712, 1984.

42. Rauth, A. M., McClelland, R. A., Michaels, H. B., and Battistella, R. Theoxygen dependence of the reduction of nitroimidazoles in a radiolytic modelsystem. Int. J. Radiât.Oncol. Biol. Phys., 10: 1323-1326, 1984.

43. Chambers, S. H., Bleehen, N. M., and Watson, J. V. Effect of cell densityon intracellular adriamycin concentration and cytotoxicity in exponentialand plateau phase EMT6 cells. Br. J. Cancer, 49: 301-306, 1984.

44. Drewinko, B., Patchen, M., Yang, L. Y., and Barlogie. B. Differential killingefficacy of twenty antitumor drugs on proliferating and nonproliferatinghuman tumor cells. Cancer Res., 41: 2328-2333, 1981.

45. Sciandra, J. J., Wilson, R. E., Michel, G. J., and Sutherland, R. M. Correlation of synthesis of oxygen regulated proteins with changes in adriamycinresistance. Thirty Fourth Annual Meeting of the Radiation Research Society(Abstract En-19), p. 125. 1986.

46. Colofiore, J. R., Ara, G., Berry, D., and Belli, J. A. Enhanced survival ofAdriamycin-treated Chinese hamster cells by 2-deoxyglucose and 2,4-dinitro-phenol. Cancer Res., 42: 3934-3940, 1982.

47. Rensing, L. Effects of 2,4-dinitrophenol and dinactin on heat-sensitive ec-dysone-specific puffs of drosophilia salivary gland chromosomes in vivo. CellDiffer., 2: 221-228, 1973.

48. Pouyssegur, J., Shin, R. P. C., and Pastan, I. Induction of two transformationsensitive membrane polypeptides in normal fibroblasts by a block in proteinsynthesis or glucose deprivation. Cell, //: 941-947, 1977.

49. Shin, R. P. C., Pouyssegur, J., and Pastan. I. Glucose depletion accounts forthe induction of two transformation-sensitive membrane proteins in Roussarcoma virus-transformed chick embryo fibroblasts. Proc. Nati. Acad. Sci.USA, 74: 3840-3844, 1977.

50. Deschner, E. E., and Gray, L. H. Influence of oxygen tension on X-rayinduced chromosomal damage in Ehrlich ascites tumor cells irradiated invitro and in vivo. Radiât.Res., 11: 115-146, 1959.

51. Tannock, I. Response of aerobic and hypoxic cells in a solid tumor toadriamycin and cyclophosphamide and interaction of the drugs with radiation. Cancer Res., 42:4921-4926, 1982.

52. Brox, L., Gowans, B., To, R., and Belch, A. The effect of anoxia on theanthracycline-induced DNA damage in the RPMI-6410 human lymphoblas-toid cell line. Can. J. Biochem., 60: 873-876, 1982.

2412

on May 4, 2020. © 1987 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

1987;47:2407-2412. Cancer Res   Vicram Gupta and John J. Costanzi  Ehrlich Ascites CellsRole of Hypoxia in Anticancer Drug-induced Cytotoxicity for

  Updated version

  http://cancerres.aacrjournals.org/content/47/9/2407

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/47/9/2407To request permission to re-use all or part of this article, use this link

on May 4, 2020. © 1987 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from