Nitrofurans as Radiosensitizers of Hypoxic Mammalian Cellscancerres.aacrjournals.org/content/canres/32/12/2616.full.pdf · binding of the label from nitrofurazone-14 C to bovine serum

[CANCER RESEARCH 32, 2630-2632, December 1972]

Nitrofurans as Radiosensitizers of Hypoxic Mammalian Cells

J. D. Chapman, A. P. Reuvers, J. Borsa, A. Petkau, and D. R. McCalla1

Medical Biophysics Branch, Whiteshell Nuclear Research Establishment, Atomic Energy of Canada Limited, Pinawa, Manitoba, Canada /J. D. C.,A. P. R..J. B., A. P.], and Department of Biochemistry, McMaster University, Hamilton, Ontario, Canada ¡D.R. M./

SUMMARY

Our studies showed that various nitrofuran derivatives haveexcellent radiosensitizing properties in hypoxic Chinesehamster cells at concentrations at which there is no effect onplating efficiency. The radiosensitizing effect is primarily adose-modifying effect, with nitrofurantoin, nitrofurazone, andnifuroxime (each at 500 ¿M) in complete medium givingenhancement ratios of 1.65, 2.0, and 2.2, respectively. Inparallel studies, the radiosensitizing effect of molecular oxygen(air-saturated conditions) was found to be dose modifying,with an enhancement ratio of ~2.9. Selective

radiosensitization in hypoxia has been demonstrated inproliferating populations of cells in every phase of the cellcycle, as well as in nonproliferating populations which werecontact inhibited at the time of irradiation. In all cellpopulations tested, these compounds did not alter radio-sensitivity in air-saturated conditions.

The permeability of phospholipid membranes tonitrofurazone in 0.1 M NaCl has been measured, and apermeability constant (K) of 1.5 (±0.7) X IO"5 cm/sec

obtained. No diffusion time lag was observed. In the presenceof 0.1 M NaCl and 0.5 mM MgCl2, the permeability constantwas increased to 2.5 (±0.4) X IO"5 cm/sec, and a diffusion lag

time of 12 ±3 min was obtained. Comparison of this constantwith permeability constants for other molecules suggests thatthe phospholipid membrane presents no great barrier to thepenetration of this compound.

Hypoxia is shown to enhance the rate of radiation-inducedbinding of the label from nitrofurazone-14 C to bovine serum

albumin, DNA, polynucleotides, and Chinese hamster cells.Mechanisms of radiosensitization are discussed in relation tothe observed binding and the antibacterial mode of action ofthe nitrofurans. The potential application of this class ofcompounds in the radiotherapy of tumors in which hypoxia issuspected is discussed in relation to their knownpharmacological properties and current clinical usage.

INTRODUCTION

It has been suggested that the concentration of oxygendissolved in tissues at the time of irradiation may be a factor indetermining the success of radiotherapy for some tumors (17,

' Supported by grants from the National Cancer Institute of Canada.Received May 9, 1972; accepted August 22, 1972.

18). Subsequent experiments indicated that cells in solidtumors irradiated in vivo are inactivated by ionizing radiationaccording to a multicomponent survival curve that may berelated to a variation in oxygen concentration across thetumor mass (22, 36). If the radiosensitivity of hypoxic cellswithin solid tumors could be made to coincide with that of theoxygenated cells of the same tumor, the success of currentradiotherapy regimes would be greatly enhanced. Attempts tomodify the radiosensitivity of suspected hypoxic centers intumors by oxygen have been frustrated by difficulties and ingeneral have not been successful. Kaplan (23) has stated,regarding clinical trials of hyperbaric oxygen radiotherapy,that "Unfortunately these randomized clinical trials have to

date failed to reveal any significant improvement in survival or,in most instances, even in the frequency of local eradication ofseveral types of cancer treated under hyperbarix oxygen . .. ."

Another approach to the problem of hypoxia in radiotherapywould be the use of a different chemical radiosensitizer.Advantages could possibly be gained if the sensitizer were notrapidly metabolized by tissue so that concentrations displayingradiosensitizing potential could be established in the vicinityof the hypoxic cells during the radiation treatment.

Bridges (9) has reviewed the topic of chemicalradiosensitization, and one of the most promising classes ofradiosensitiziers for application in tumor therapy appears to bethe electron-affinic type described by Adams et al. (1-4).PNAP2 was shown to be strongly electron affinic (1) and,

subsequently, was shown to be an effective radiosensitizer ofhypoxic mammalian cells (2, 13). A study with analogs ofPNAP (10) has demonstrated that the N02 substituent of thebenzene ring is essential for demonstration ofradiosensitization in mammalian cells. Furthermore, the extentof radiosensitization could, to some degree, be correlated withthe electronegativity of the nitrobenzene analogs tested.Studies have been extended to some nitrofuran derivativeswhich are known to be more electronegative than theircorresponding nitrobenzene derivatives (39) and which havethe attraction of established clinical acceptability (32).Preliminary experiments indicated that nitrofuran derivativesare excellent radiosensitizers of hypoxic mammalian cells (37).In this paper, we report additional radiobiological datademonstrating the radiosensitizing effectiveness of 3nitrofuran derivatives, along with data pertaining to thetoxicity, permeability, and mechanism of radiosensitization ofthese compounds.

'The abbreviations used are: PNAP, p-nitroacetophenone; poIy(A),poly(adenylate); poly(C), poly(cytidylate); poly(I), poly(inonsinate);poly(U), poly(uridylate); HBSS, Hanks' balanced salt solution; TCA,

trichloroacetic acid; ER, enhancement ratio.

2616 CANCER RESEARCH VOL. 32

Research. on August 23, 2019. © 1972 American Association for Cancercancerres.aacrjournals.org Downloaded from

NitrofuransasCellularRadiosensitizers

MATERIALS AND METHODS

Chinese hamster cell line V79-379-A was routinely grown insuspension culture consisting of minimal essential mediumwith spinner salts (Grand Island Biological Co., Grand Island,N. Y.) supplemented with antibiotics and fetal calf serum (7%by volume: Becton and Dickinson Ltd., Clarkson, Ontario,Canada). Cells grown in this manner at 37° had a

doubling time of ~10 hr and could be transferred from spinner

culture to monolayer culture (attached to glass) and backagain with ease.

5-Nitro-2-furaldehyde-semicarbazone (nitrofurazone), 7V-(5-nitro-2-furfurylidine-l-aminohydantoin (nitrofurantoin), andnitrofurazone-formyl-14C (specific activity, 31.1 juCi/mg)

were generously provided by Norwich Pharmacal Co.,Norwich, N. Y. ^«fz-5-nitro-2-furaldoxime (nifuroxime) wasobtained from Aldrich Chemical Co., Milwaukee, Wis. Bovineserum albumin (Fraction V), salmon sperm DNA (highlypolymerized), poly(A), and poly(C) were obtained from SigmaChemical Co., St. Louis, Mo. Poly(I) and poly(U) werepurchased from Schwarz/Mann, Orangeburg, N. Y. In all ofour cellular experiments, the drugs were dissolved in completegrowth medium unless specified to the contrary.

The radiation source, radiation quality, and dose rate in acellular monolayer have been described elsewhere (13).

Toxicity studies with cells growing in vitro were performedto determine both the cytostatic and cytocidal action of thesedrugs. The cytostatic effect was measured by determining theeffect of various concentrations of each nitro furan on net cellproliferation over a 24-hr growth period. These experimentswere performed with cells growing attached to Petri dishes,and the data were reduced to a percentage of normalproliferation. We determined the cytocidal effect of eachnitrofuran derivative by exposing small numbers of hamstercells attached to Petri dishes to various concentrations of eachdrug for various lengths of time. The ability of cells treated sothat they would subsequently proliferate in drug-free mediumand form normal colonies in 5 to 7 days was assayed. Thecellular proliferative capacity (or colony-forming ability) wasthe end point used in all radiobiological experiments, as well.

We prepared cells for irradiation by diluting growingcultures and plating an appropriate number of cells in glassPetri dishes (60 X 15 mm) in 3 ml of medium. Cells werepermitted to attach and grow at 37°for 2 to 3 hr. The cell

number per dish was adjusted so that, depending uponirradiation conditions and dose, approximately 100 coloniesper dish would be evident after 6 days of incubation at 37°for

hamster cells. Prior to irradiation, the medium in which thecells had attached was removed from the dishes and replacedwith 0.75 ml of complete medium with or without drug. Theair-tight chamber used to hold the Petri dishes in the X-rayfield and degassing procedures used to establish hypoxia havebeen described elsewhere (13).

Synchronously growing populations of Chinese hamstercells were established by a mitotic selection procedure (12).Cells growing in spinner culture were plated in 16-oz bottles (3to 4 X IO6 cells/bottle) and were incubated overnight at 37°.

The monolayers were then shaken at hourly intervals, and the2nd or 3rd population selected was used. Radiation doses for

synchronously growing cells were adjusted so that the resultinglevel of survival was approximately the same for cellsirradiated under the various conditions. Whole survival curveswere also constructed for cells in the most sensitive and mostresistant parts of the cell cycle.

We prepared confluent cultures by plating ~106 cells into

glass Petri dishes in an adequate volume of medium andincubating them until the cells were observed to have reachedconfluence. Contact-inhibited cells (at confluence for ~48 hr)were irradiated in a manner identical with that described forasynchronously growing cell monlayers. The cells were coveredwith 0.75 ml of medium (with and without sensitizer) in airand in nitrogen environments. After irradiation, the cells weretrypsinized, counted, and diluted, and appropriate numberswere plated into plastic Petri dishes. The plating efficiency ofcells handled in this manner was always greater than 75%.

Reconstituted phospholipid membranes, prepared byprocedures previously described (33, 34), were used as modelsfor an examination of the permeability of membranes to thenitrofurazone-14 C radiosensitizer. Through the use of aperfusion technique (34), all of the nitrofurazone-14C that

diffused through the phospholipid membrane was collected inthe perfusate as a function of time. The nitrofurazone-14C

activity in the volume of perfusate collected each hr wasmeasured by liquid scintillation counting, as previouslydescribed (34).

Solutions containing 3.2 mg bovine serum albumin andabout IO6 cpm (20 jug) of nitrofurazone-14C in 1.25 ml of0.053 M phosphate buffer, pH 7.2, were irradiated in 25-mlsuction flasks with -y-rays from a 137Cs source. The dose rate

as determined by a Victoreen dosimeter was 330 R/min.Hypoxie conditions were achieved by blowing H2 O-saturatedN2 over the solution for 10 min prior to radiation, as well asduring radiation.

After irradiation, the solution was applied to a SephadexG-25 column and eluted with water. Protein-containingfractions were located by their UV absorbance. One ml fromthe peak tube was added to 15 ml of dioxane fluor andcounted in a Nuclear-Chicago Unilux 1 counter.

Polynucleotides (0.6 mg) were dissolved in 0.015 M sodiumcitrate:0.15 M NaCl buffer (pH 7.0), irradiated, andchromatographed as described previously. The concentrationof polynucleotide in the peak tube was determined from Aj60measurements after acid hydrolysis.

DNA solutions (0.15 M NaCl solution:citrate buffer) wereirradiated in the presence of nitrofurazone and then weredeproteinized by shaking with chloroform:isoamyl alcohol(9:1) after the addition of sodium dodecyl sulfate (1%). Thedeproteinized step was repeated (usually 3 times) until theaqueous phase reached constant activity. The solution wasthen chromatographed as described above. The amount ofprotein present in the polynucleotide preparations wasdetermined by the folin procedure of Lowry et al (28), withbovine serum albumin as a reference.

We demonstrated radiation-induced binding to hamster cellsby preparing a standard suspension of cells andnitrofurazone-14C in HBSS and irradiating them in a 7-cell(Atomic Energy of Canada Ltd.; dose rate,~18,000 rad/min).

Cells growing asynchronously in spinner culture were

DECEMBER 1972 2617

Research. on August 23, 2019. © 1972 American Association for Cancercancerres.aacrjournals.org Downloaded from

Chapman, Reuvers, Borsa, Petkau, andMcCalla

harvested by centrifugation (700 X g for 10 min), washed oncein HBSS, and resuspended at a final concentration of 10 6

cells/ml in HBSS that contained 100 juMnitrofurazone-formyl-14C (specific activity, 31.1 ¿/Ci/mg).

Aliquots (0.05 ml) were dispensed into the bottom of testtubes and were allowed to equilibrate with environments of airor pure nitrogen which were circulated over the droplets priorto and during irradiation. After irradiation, 1 ml of ice-cold 5%TCA was added to each test tube, and the samples were held inan ice bath for 30 min. The acid precipitates were collected onmembrane filters, washed with ice-cold 5% TCA, dried, andcounted in toluene base liquid scintillator in a Nuclear-ChicagoMark II liquid scintillation spectrometer.

RESULTS

In Vitro Toxicity of Nitrofuran Derivatives. We measured thecytostatic effect of each nitrofuran derivative by exposinghamster cells growing at 37°in complete growth medium to

various concentrations of each drug for 24 hr and assaying theincrease in cell number. Chart 1 shows growth inhibitionresponses for nitrofurazone, nitrofurantoin, and nifuroxime,and the concentrations of drug resulting in 50% inhibition ofproliferation are 44, 72, and 44 yU, respectively. Thecytocidal effect of various concentrations of the nitrofuranswas measured, and the data are shown in Chart 2. It isapparent that nifuroxime is at least 2 to 3 times more toxic tohamster cells in culture than are nitrofurazone andnitrofurantoin. In all of the subsequent experiments reportedin this paper, the time of exposure of the the nitrofurans tocells was always less than 30 min so that, even at very highconcentrations of the drugs, negligible reduction in platingefficiency was measured in parallel toxicity runs.

Characterization of Nitrofuran Radiosensitization of Cells.To estimate the radiosensitizing effectiveness of potentialradiosensitizers of hypoxic cells, we routinely use a rapidscreening technique based on a single dose of irradiation. Chart

zoP 120

CE

ïÃ1̄00gj

S 80Q.

< 60

Ki 40

o.0 20a«

O

—¿�I—

•¿�NITROFURAZONE

A NITROFURANTOIN-t- NIFUROXIME

-4IO'3 IO

CONCENTRATION OF DRUG [M]

IO"

Chart 1. Cytostatic effect of some nitrofuran derivatives onproliferating Chinese hamster cells expressed as % of normalproliferation versus concentration of drug in complete growth medium.

IOC

oz

-Ul

o

IO

NITROFURAZONEBNITROFURANTOIN NIFUROXIME

[x 125/iM •¿�250/iM D

0 5 10 15 0 5 10 15 0 5 10 15TIME OF EXPOSURE TO DRUGS AT 37° (HOURS)

Chart 2. Cytocidal effect of some nitrofuran derivatives onproliferating Chinese hamster cells expressed as subsequent platingefficiency after various times of exposure to drugs at 37°.

g(EIk

O

feo

10-'-

•¿�NITROFURAZONEA NITROFURANTOIN+ NIFUROXIME

VA/W'R/W\/\A/ WV!-tf- Kr» IO-4 IO'3

CONCENTRATION OF DRUG [M]

Chart 3. Radiosensitizing effectiveness of various concentrations ofsome nitrofuran derivatives expressed as ratios of surviving fraction(sensitized/unsensitized) in hypoxia after 1650 rads as a function ofsensitizer concentration. The sensitizing effectiveness of oxygen inair-saturated conditions is indicated on the chart. •¿�,furfuraldehydesemicarbazone.

3 shows the ratio of surviving fractions for hamster cellsirradiated with 1650 rads in hypoxia in the presence andabsence of various concentrations of each drug tested. All

concentrations of the nitrofurans tested showedradiosensitizing ability which increased with concentration.The concentrations of nifuroxime, nitrofurazone, andnitrofurantoin that give a "ratio of surviving fractions" of 0.10(equivalent to an enhancement ratio of ~1.7 for purelydose-modifying effects) were 100, 200, and 600 juM,respectively. The radiosensitizing effectiveness of

2618 CANCER RESEARCH VOL. 32

Research. on August 23, 2019. © 1972 American Association for Cancercancerres.aacrjournals.org Downloaded from

Nitrofurans as Cellular Radiosensitizers

furfuraldehyde semicarbazone has been measured over asimilar concentration range, and the data are also shown inChart 3. It is evident that the introduction of a N02 group onthe furan ring greatly enhances the radiosensitizingeffectiveness of that structure. The radiosensitizingeffectiveness of oxygen in air-saturated conditions is shown inthe same plot, and the extent of radiosensitization bynitrofurazone and nifuroxime at 500 juM (and greater)approaches that of oxygen.

Complete survival curves were constructed to demonstratethe radiosensitizing effectiveness of each of the drugs at 500MM in complete medium, and the data are shown in Chart 4.The radiosensitizing effect of the nitrofurans as well asmolecular oxygen is primarily dose modifying. The data inChart 4 are from an experiment performed on the same cellpopulation; ER's as calculated from the £)0values of the

curves for radiosensitization by nitrofurantoin, nitrofurazone,nifuroxime, and molecular oxygen (air-saturated conditions),are 1.66, 1.95, 2.06, and 2.85, respectively. Small variations inER are observed from day to day for any 1 concentration ofdrug, and the average ER's determined from 3 independent

experiments for each of the sensitizing conditions mentionedabove are 1.65, 2.0, 2.2, and 2.9. At 500 MM,these drugs hadno effect on the radiosensitivity of hamster cells inair-saturated environments. What was previously reported (37)as a slight radioprotective effect of the nitrofurans in air has

io"-

o01

O

erin

10-2

IO '3

lililí1

a NITROGEN 2970 radsA ImM NITROFURANTOIN 1615 rads•¿�500/JLM NITROFURAZONE 1615 radsx AIR ä 1055 rods

6 8 10 12 14

TIME AFTER SELECTION 8 PLATING (HOURS)

Chart 5. Radiosensitivity of Chinese hamster cells that were selectedat mitosis and irradiated at hourly intervals throughout the 1st cellcycle. Cells received the amounts of irradiation shown on the chart.

ERNITROGEN I 00

NITROFURANTOIN I 66NITROFURAZONE 1.95

NIFUROXIME 2.06AIR 2.85

1000 2000 3000 4000

RADIATION DOSE (rods)

Chart 4. Survival curves demonstrating the radiosensitizingeffectiveness in hypoxia of 500 pM nifuroxime, nitrofurazone, andnitrofurantoin. The hypoxic and air-saturated responses of the same cellpopulation are also shown.

since been shown to result from a small amount (0.3%) ofdimethyl sulfoxide that had been used initially to dissolve thedrugs prior to dilution in complete medium.

Whole survival curves for hypoxic hamster cells irradiated inthe presence of various concentrations of nifuroxime showedthe radiosensitizing effect to be dose modifying at eachconcentration, much like the radiosensitizing effectiveness ofoxygen as measured by Elkind et al. (15). Significantradiosensitization (ER, l .22) was measured with nifuroxime at10 MM, a concentration of drug which is nontoxic by bothcriteria used to estimate cellular toxicity (see Charts 1 and 2).

The radiosensitization of mammalian cells by oxygen hasbeen shown to be independent of position within the cell cycle(13, 24, 25), and this is also true for the electron-affinicradiosensitizer PNAP (13). Experiments were performed todetermine the cell cycle dependency of the radiosensitizingeffectiveness of nitrofurazone and nitrofurantoin. Cellsselected at mitosis and plated were irradiated at hourlyintervals during the 1st cell cycle. Chart 5 shows the survivingfractions from an experiment in which cells were irradiatedwith 1055 rads in air, 2970 rads in N2, and 1615 rads in N2 inthe presence of either 500 MM nitrofurazone or 1 mMnitrofurantoin. The cell cycle response for all of these casesshows the same general shape; cells in Gt were most sensitiveand cells in late S were resistant. Nitrofurazone appears to beslightly more effective in radiosensitizing G( than inradiosensitizing S cells. This increased effect on G! cells isdemonstrated in Chart 6, which displays whole survival curvesfor G i cells (2 hr after selection and plating) and S cells (8 hrafter selection and plating). The ER's for 500 MM

nitrofurazone in this run were 2.18 for GÃŒcells and 1.86 for Scells.

DECEMBER 1972 2619

Research. on August 23, 2019. © 1972 American Association for Cancercancerres.aacrjournals.org Downloaded from

Chapman, Reuvers, Borsa, Petkau, andMcCalla

1000 2000 3000 O 1000 2000 3000 4OOO

RADIATION DOSE (rods)

Chart 6. Survival curves of G, cells (2 hr after mitotic selection) andlate-S-phase cells (8 hr after mitotic cells) irradiated in hypoxia (a), inhypoxia with 500 MMnitrofurazone (•),in air (X), and in air with 500MMnitrofurazone (o).

Hypoxie cells in tumors are likely to be part of thenonproliferating or slowly proliferating population. Stationaryphase cultures at confluence have been studied as an in vitromodel system more closely approximating nonproliferatingcells in vivo (7, 19, 26, 27). Our hamster cell line, whenbrought to a nonproliferating state at contact inhibition, isblocked in a postmitotic, pre-DNA-synthetic part of the cellcycle (11), and it has a radiation sensitivity similar to that ofGj cells in synchronously growing cultures if the cells areirradiated as confluent cultures and immediately dispersed bytrypsin for an assay of colony-forming ability (J. D. Chapman,unpublished results). Confluent cultures were irradiated in airand in hypoxia, with and without 500 juMnitrofurazone; thesurvival data are shown in Chart 7. An ER of 1.82 fornitrofurazone in hypoxia was obtained. This value is notsignificantly different from those measured in proliferating cellpopulations.

The charts that show cellular radiosensitivity (Chart 3 to 7)were constructed with the use of data from single-cellpopulations, and the number of colonies counted for eachsurvival determination was such that the error for each was lessthan 7%(near the size of the symbols on the charts).

Permeability of Model Membranes to Nitrofurazone. Chart8a illustrates the results obtained in a 7-hr experiment for a0.06-sq cm membrane in 0.1 M NaCl. Q(t) is the time integralof the nitrofurazone-14 C counts. It is seen that Q(t) increases

linearly with time. From the slope of the line, the amount ofradiosensitizer permeating in a unidirectional manner throughthe membrane per unit time, dQ/dt, is determined. Thepermeability constant K (cm/sec) is obtained as previouslydescribed (34) from the equation

dQ/dt = K-A-&c

where A is the area of membrane in sq cm and Ac is the

concentration difference of nitrofurazone-14 C across themembrane (*;105 cpm/cu cm). In 0.1 M NaCl, the

permeability constant was determined to be 1.5 (±0.7) X10~5 cm/sec, and there was no diffusion time lag, indicating

that the drug partitions readily into the lipid membrane phase.In 0.1 M NaCl and 0.5 mM MgCl2, the Q(t) versus t curves(Chart 8¿>)were basically similar, except that the value of Kwas somewhat higher [=2.5 (±0.4)X IO"5 cm/sec], and a

o NITROGEN•¿�N2+500/iM NITROFURAZONE

AIR

10 -

0 1000 2000 3000 4000

RADIATION DOSE (rods)

Chart 7. Survival curves for cells irradiated while contact inhibited inhypoxia, in hypoxia with 500 fiM nitrofurazone, and in air-saturatedconditions.

(o)

O I M NOCIx

k-09x IO"cm/sec

(b)

01 M NOCI ,05mM MgCI2 /

OI234567OI23456

TIME ( HOURS)

Chart 8. Passage of nitrofurazone-"C through a reconstituted

phospholipid membrane as a function of time. The membrane area (inboth a and b) was 0.06 sq cm.

2620 CANCER RESEARCH VOL. 32

Research. on August 23, 2019. © 1972 American Association for Cancercancerres.aacrjournals.org Downloaded from

Nitrofurans as Cellular Radiosensitizers

Q 30 -

40 60DOSE ( K rods )

Chart 9. Radiation-induced binding of "C from labeled

nitrofurazone to protein, o and n, Experiment 1; •¿�and •¿�,Experiment2;krads, kilorads.

diffusion time lag of 12 ±3 min was demonstrated. Thediffusion time lag obtained by extrapolating the steady stateportion of the curve to the time axis is a measure of the timerequired for the nitrofurazone to reach an equilibriumdistribution within the different regions of the membrane,particularly the interfaces. The values of K are about 1 orderof magnitude less than the permeability constant for water(34), and they suggest that the radiosensitizer should readilypenetrate the membranes of individual cells.

Radiation-induced Binding of Nitrofurazone-14 C. Chart 9shows the binding of 14C to protein when solutions of labelednitrofurazone and serum albumin were irradiated with -y-rays.

The amount bound increases with dose and is greater underhypoxia than in air by a factor of 4 to 5 at 20,000 rads.Irradiation of a nitrofurazone solution, followed by theaddition of protein, did not result in binding that wassignificantly greater than that found in unirradiated controls.

The radiation-induced binding experiments were performedwith radiation doses approximately 10 times greater than thedose range used to characterize the cellular radiosensitizingeffects. This dose range was chosen primarily because of therelatively low specific activity of the labeled sensitizer.Nevertheless, the linear increase in sensitizer bound withincreasing radiation dose (Charts 9 and 10) indicates that theobserved binding characteristics can be expected to hold forthe biological dose range as well.

Table 1 shows that there is extensive radiation-inducedbinding of 14C from labeled nitrofurazone to DNA and to

polynucleotides. Under the conditions of our hypoxicexperiments, 1.6 to 6.4 million counts were bound per mmoleof nucleotide, representing of the order of 1 molecule ofnitrofurazone per 5,000 nucleotide units. With the exceptionof poly(A), about 15 times as much label was bound underhypoxic conditions as were bound under aerobic conditions at20,000 rads.

The protein content of poly(l), poly(U), and poly(A) wasbelow 0.25%, so binding to contaminating protein would nothave amounted to more than about 4,500 cpm/mmole

nucleotide. The poly(C) contained 2.5% protein which mighthave bound up to 45,000 cpm/mmole nucleotide. Even thisvalue is so far below the total amount bound to thenucleotides that protein binding can be considered to benegligible.

Similarly, little radioactivity was bound to polymers inunirradiated controls or when the polymer was added topreviously irradiated nitrofurazone solutions.

Hamster cells were suspended in HBSS containing 100 ¿iMnitrofurazone-14 C and irradiated in air and hypoxia. Chart 10shows the amount of 14C label from the sensitizer bound to

the TCA-insoluble fraction of whole cells as a function ofradiation dose. Radiation-induced binding in hypoxia is

6 800-

01 2345

RADIATION DOSE (x 18,000 rads)

Chart 10. Radiation-induced binding of "C from labelednitrofurazone to Chinese hamster cells. X, hypoxic conditions; »,air-saturated conditions.

Table 1Binding of ' 'C from labeled nitrofurazone to polynucleotides

after irradiation (20,000 rads)

AmountboundCompoundDNAPoly

(I)Poly(A)Poly(C)Poly(U)Nitrofurazoneinput(cpm

X10~5/1.25

ml)3.252.32.22.32.3(cpm

X10'5/mmolenucleotide)inAir2.6°3.72.61.83.2Hypoxia37°64162748N2:air14176.21515

°Assuming an average molecular weight of 310 for the nucleotides

of DNA.

DECEMBER 1972 2621

Research. on August 23, 2019. © 1972 American Association for Cancercancerres.aacrjournals.org Downloaded from

Chapman, Reuvers, Borsa, Petkau, and McCalla

enhanced by a factor of 7.0 to 7.5, as compared withoxygenated conditions. This enhancement ratio for binding towhole cells is greater than that measured for binding to proteinand is less than that for DNA, suggesting that a variety ofcellular molecules probably are reacting with thenitrofurazone-14 C.

DISCUSSION

Most of the data presented in this paper pertain to the toxicand radiosensitizing properties of some nitrofuran derivativesin an in vitro mammalian cell system. This cell system isconvenient for rapidly estimating the merits of compoundsselected as potential radiosensitizers. It is hoped that thesedata will provide a useful basis upon which to constructexperimental designs for the radiotherapy of solid tumors inanimals.

All radiosensitizers of the "electron-affinic type" (1) will

probably be toxic to mammalian cells at high concentrations.One aim of our research has been to define drug concentrationlimits within which good selective radiosensitization ofhypoxic cells can be demonstrated and which show notoxicity. Regarding the cy tosta tic effect of the nitrofurans,those tested inhibit cell proliferation at concentrations 6 to 9times lower than that of PNAP (13). This result agrees with aprevious observation that cellular toxicity as estimated bygrowth inhibition increases with increasing electronegativitiesof drugs (10). Cytocidal studies (Chart 2) with thesenitrofuran derivatives show that cells can tolerate exposures tothese drugs for a few hr at concentrations that completelyinhibit growth without any subsequent effect on the ability ofthe cells to proliferate and form normal colonies. These studiespredict that either 500 fiM concentrations of nitrofurazoneand nitrofurantoin or 125 pM concentrations of nifuroximecan be tolerated by cells for exposures of up to 5 hr. The orderof toxicity of these 3 nitrofuran derivatives, as determined bygrowth inhibition (cytostatic) and plating efficiency(cytocidal), is identical with the order established for chronicand acute toxicity of these drugs in mice (32).

The radiosensitizing effectiveness of these drugs inmammalian cells is greater than that of most radiosensitizerscharacterized to date (2, 5, 9, 13, 31). At concentrations thatdo not interfere with cell proliferation (< 20 ^iM), bothnitrofurazone and nifuroxime radiosensitize cells with ER's of

between 1.2 and 1.4. This extent of radiosensitization issignificant and would be expected to enhance the effectivenessof present radiotherapy regimes if all hypoxic cells within atumor could be sensitized to this extent (see Ref. 22). Theobservation that nitrofurazone can sensitize bothnonproliferating and proliferating cells in every phase of thecell cycle demonstrates that the radiosensitizing mechanism isnot influenced to any great extent by the metabolic state ofthe cell or specific metabolism in distinctive parts of the cellcycle. On the basis of the characterizations presented in thispaper, the nitrofurans act in a manner similar to those of otherelectron-affinic radiosensitizers (2, 13) and are intermediatebetween PNAP and molecular oxygen in effectiveness.

Studies with nitrofuran derivatives have shown that plasmaconcentrations of up to 50 juM can be detected between 30

and 60 min after a dose that is well below the acutely toxiclevel (32) has been administered, p.o., to rats. The resultspresented in this paper show that significant levels ofradiosensitization (ER's of 1.3 to 1.5) could be realized with

nitrofuran concentrations of this magnitude in the vicinity ofhypoxic cells. In clinical studies in man, urinaryconcentrations of nitrofurantoin up to and exceeding 1 mMhave been detected after its administration p.o. (8, 38). Theestablishment of high concentrations of nitrofurantoin in theurine prior to and during irradiation may very well result in amore successful radiotherapy of bladder tumors. Radiationstudies with hypoxic cells with doses of 500 rads and less(doses more in line with normal radiotherapy procedures)show the radiosensitizing effect of the nitrofurans to be dosemodifying, even on the shoulder portion of the survival curves.

The demonstration that nitrofurazone can readily penetratethe bimolecular lipid membrane structure correlates with theobservation that radiation-induced damage in the DNA ofhamster cells (as estimated by single-strand-break analysis) isenhanced for cells irradiated in hypoxia in the presence ofthese nitrofuran derivatives (14). If we assume that themechanism of radiosensitization involves short-lived, freeradical species, then the drug would have to be in the vicinityof the nucleus and DNA at the time or irradiation in order tohave an enhancing effect at that level. Studies with hamstercells at 37°indicate that the maximum radiosensitizing effect

is achieved with the drug (500 /¿Mnitrofurazone) in contactwith the cells only 5 min prior to irradiation. These datademonstrate that the nitrofurans probably penetrate the cellmembrane rapidly and enhance radiation-induced damagewithin the cell. The possibilities that only a few nitrofuranmolecules at some site in the cell are required for the effect orthat the radiation target for cell lethality is at the cellmembrane cannot be excluded on the basis of this work.

The radiation-induced binding of the label ofnitrofurazone-14 C to macromolecules and to whole cells and

the enhancement of this binding by the absence of oxygensuggest that sensitizer binding may be involved inradiosensitization. While this enhanced binding is nonspecific,in that it has been demonstrated for protein, DNA, andpolynucleotides, considerably more enhancement (14- to17-fold) is observed with DNA, poly(I), poly(U), and poly(C)than with protein. With poly(A) and protein, enhancements of4- to 5-fold are seen when oxygen is excluded. Sinceradiation-induced strand breakage of hamster cell DNA isenhanced in hypoxia by the nitrofurans to an extent similar tothat for cell killing (14), it is possible that enhanced binding tocellular DNA may result in enhanced radiation-induceddamage in the DNA and, ultimately, in enhanced cell killing.Recent studies on the antibacterial mode of action ofnitrofurazone have shown that activation of this drug byreductive metabolism in bacteria leads to the binding ofnitrofurazone-14 C to the acid-insoluble fraction (29) and tothe formation of single-strand breaks in DNA (30). Theenzymatic activation of the nitrofurans by bacteria and thesubsequent binding, DNA strand breakage, and lethality seemto be parallel to the radiation-enhanced effects obtained withthe same compounds.

The nitrofurans have been used with good success as

2622 CANCER RESEARCH VOL. 32

Research. on August 23, 2019. © 1972 American Association for Cancercancerres.aacrjournals.org Downloaded from

Nitrofurans as Cellular Radioscnsitizers

antibacterial agents in the treatment of infections in humansand animals (6, 32). Friedgood and Green (16) reportedmarked atrophy of the testes following administration of p.o.doses of nitrofurazone to mice, as well as a retardation of thegrowth of a transplantable fibrosarcoma (16). This observationled to the experimental treatment with nitrofurazone ofprimary and metastatic seminomas in humans (20, 21, 35,40).Nitrofurazone chemotherapy was usually initiated after bothsurgery and radiotherapy had failed to arrest the metastaticinvasion of testicular cancer to the lungs and its subsequentgrowth. In only 1 case was nitrofurazone chemotherapy usedin parallel with an X-ray therapy regime which involved 4200R over a 34-day period (20), and regression of this pulmonarymétastases from embryonal carcinoma of the teste wasreported. This isolated case history leaves many questionsunanswered regarding whether the drug, or the X-rays, or acombination of both led to the observed regression. It is safeto say that the use of nitrofuran derivatives in conjunctionwith radiotherapy of tumors when hypoxia is suspected hasnot as yet been clinically evaluated. The demonstration in thispaper of the radiosensitizing potential of this class of drugswarrants a careful evaluation of their potential use in theradiotherapy of human cancers.

ACKNOWLEDGMENTS

We would like to thank our colleagues of the Medical BiophysicsBranch for their helpful discussions during the course of this work. Mrs.Rebecca Berry provided expert technical assistance with the14C-binding experiments.

REFERENCES

1. Adams, G. E. Molecular Mechanisms of Cellular Radiosensitizationand Protection, in: H. L. Moroson and M. Quintiliani (eds.),Radiation Protection and Sensitization, pp. 3-14. London: Taylor& Francis Ltd., 1970.

2. Adams, G. E., Asquith, J. C., Dewey, D. L., Foster, J. L., andWillson, R. L. Electron-affinic Sensitization. II. para-Nitro-acetophenone: A Radiosensitizer for Anoxic Bacterial and Mammalian Cells. Intern. J. Radiation Biol., 19: 575-585, 1971.

3. Adams, G. E., and Cooke, M. S. Electron-affinic Sensitization. I. AStructural Basis for Chemical Radiosensitizers in Bacteria. Intern.J. Radiation Biol., 15: 457-471, 1969.

4. Adams, G. E., and Dewey, D. L. Hydrated Electrons andRadiobiological Sensitization. Biochem. Biophys. Res. Commun.,12: 473-477, 1963.

5. Ashwood-Smith, M. J., Robinson, D. M., Barnes, J. H., and Bridges,B. A. Radiosensitization of Bacterial and Mammalian Cells bySubstituted Glyoxals. Nature, 270: 137-139, 1967.

6. Bailey, R. R., Gower, P. E., Roberts, A. P., and de Wardener, H. E.Prevention of Urinary-tract Infection with Low-doseNitrofurantoin. Lancet, 1112-1114, 1971.

7. Berry, R. J., Hall, E. J., and Cavanagh, J. Radiosensitivity and theOxygen Effect for Mammalian Cells Cultured in Vitro in StationaryPhase. Brit. J. Radiol., 4.J: 81-90, 1970.

8. Beutner, E. H., Petronio, J. J., Lind, H. E., Trafton, H. M., andCorreia-Branco, M. Nitrofurantoin, Rational in Vitro BacterialSensitivity Testing. Antibiot. Ann., 1954-55: 988-1001, 1955.

9. Bridges, B. A. Sensitization of Organisms to Radiation by

Sulfhydryl-binding Agents. Advan. Radiation Biol., 3: 123-176,1969.

10. Chapman, J. D., Raleigh, J. A., Borsa, J., Webb, R. G., andWhitehouse, R. Radiosensitization of Mammalian Cells byp-Nitroacetophenone. II. Effectiveness of Analogues. Intern. J.Radiât.Biol., 21: 475-482, 1972.

11. Chapman, J. D., Todd, P., and Sturrock, J. X-ray Survival ofCultured Chinese Hamster Cells Resuming Growth after PlateauPhase. Radiation Res., 42: 590-600, 1970.

12. Chapman, J. D., Webb, R. G., and Borsa, J. ATP Pool Levels inSynchronously Growing Chinese Hamster Cells. J. Cell Biol., 49:229-233, 1971.

13. Chapman, J. D., Webb, R. G., and Borsa, J. Radiosensitization ofMammalian Cells by p-Nitroacetophenone. I. Characterization inAsynchronous and Synchronous Populations. Intern. J. RadiationBiol., 79:561-573, 1971.

14. Dugle, D. L., Chapman, J. D., Gillespie, C. J., Borsa, J., Webb, R.G., Meeker, B. E., and Reuvers, A. P. Radiation Induced StrandBreakage in Mammalian Cell DNA. I. Enhancement of SingleStrand Breaks by Chemical Radiosensitizers. Intern. J. RadiationBiol., in press, 1972.

15. Elkind, M. M., Swain, R. W., Alescio, T., Sutton, H., and Moses, W.B. Oxygen, Nitrogen, Recovery and Radiation Therapy. In:Cellular Radiation Biology, M. D. Anderson Hospital and TumorInstitute at Houston, pp. 442-466. Baltimore: The Williams &WilkinsCo.,1965.

16. Friedgood, C. E., and Green, M. N. The Effect of Nitrofurazone onGrowth of Fibrosarcoma in Mice. Cancer Res., 10: 613-615,1950.

17. Gray, L. H., Conger, A. D., Ebert, M., Hornsey, S., and Scott, O. C.A. Concentration of Oxygen Dissolved in Tissues at Time ofIrradiation as a Factor in Radiotherapy. Brit. J. Radiol., 26:638-648, 1953.

18. Gray, L. H., and Scott, O. C. A. Oxygen Tension and theRadiosensitivity of Tumors. Oxygen in the Animal Organism, pp.537-553. New York: Pergamon Press, Inc., 1964.

19. Hahn, G. M. Failure of Chinese Hamster Cells to Repair Sub-lethalDamage When X-irradiated in the Plateau Phase of Growth. Nature,2/7:741-742,1968.

20. Haltiwanger, E. Effect of Furacin on Testicular Tumor. J. Urol.,86: 125-126,1961.

21. Hayliar, B. L., O'Neal, A. H., Jr., and Dotterer, J. A. Metastatic

Testicular Tumor Treated with a Maintenance Dose ofNitrofurazone. J. Urol., 84: 565-568, 1960.

22. Hill, R. P., Bush, R. S., and Yeung, P. The Effect of Anaemia onthe Fraction of Hypoxie Cells in an Experimental Tumor. Brit. J.Radiol., 44: 299-304, 1971.

23. Kaplan, H. S. Radiobiology's Contribution to Radiotherapy:

Promise or Mirage? Failla Memorial Lecture. Radiation Res., 43:460-476, 1970.

24. Kruuv, J., and Sinclair, W. K. X-ray Sensitivity of SynchronizedChinese Hamster Cells Irradiated during Hypoxia. Radiation Res.,36: 45-54, 1968.

25. Legryr, G. A., and Hall, E. J. The Oxygen Effect and X-raySensitivity in Synchronously Dividing Cultures of Chinese HamsterCells. Radiation Res., 37: 161-172, 1969.

26. Little, J. B. Repair of Sub-lethal and Potentially Lethal RadiationDamage in Plateau Phase Cultures of Human Cells. Nature, 224:804-806, 1969.

27. Little, J. B. Differential Response of Rapidly and SlowlyProliferating Human Cells to X-irradiation. Radiology, 93:307-313, 1969.

28. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J.Protein Measurement with the Polin Phenol Reagent. J. Biol.Chem.,193: 265-275, 1951.

DECEMBER 1972 2623

Research. on August 23, 2019. © 1972 American Association for Cancercancerres.aacrjournals.org Downloaded from

Chapman, Reuvers, Borsa, Petkau, and McCalla

29. McCalla, D. R., Reuvers, A., and Kaiser, C. Mode of Action ofNitrofurazone. J. Bacteriol., 104: 1126-1134, 1970.

30. McCalla, D. R., Reuvers, A., and Kaiser, C. Breakage of BacterialDNA by Nitrofuran Derivatives. Cancer Res., 31: 2184-2188,

1971.31. Parker, L., Skarsgard, L. D., and Emmerson, P. T. Sensitization of

Anoxic Mammalian Cells to X-rays by Triacetoneamine JV-Oxyl.Survival and Toxicity Studies. Radiât.Res., 38: 493-500, 1969.

32. Paul, H. E., and Paul, M. F. The Nitrofurans-Chemo TherapeuticProperties. In: R. J. Schnitzer and F. Hawking (eds.), ExperimentalChemotherapy, Vol. 2, Part 1, pp. 307-370. New York: Academic

Press, Inc., 1964.33. Petkau, A., and Chelack, W. S. Permeability of a Thin Phospholipid

Membrane to Ions and Tobacco Mosaic Virus. Biochim. Biophys.Acta, 135: 812-824, 1967.

34. Petkau, A., and Chelack, W. S. Transient Permeability of a ModelLipid Membrane to Na22. Biochim. Biophys. Acta, 203: 34-46,

1970.

35. Pierce, G. B., Jr., Dixon, F. J., Jr., and Verney, E. L. A CriticalEvaluation of Furacin as an Antitesticular Agent. J. Urol., 84:569-574, 1960.

36. Powers, W. E., and Tolmach, L. J. A Multicomponent X-raySurvival Curve for Mouse Lymphosarcoma Cells Irradiated in Vivo.Nature, 797: 710-711, 1963.

37. Reuvers, A. P., Chapman, J. D., and Borsa, J.Nitrofurans-Potential Tools in Radiotherapy. Nature, 237:402-403, 1972.

38. Richards, W. A., Riss, E., Kass, E. H., and Finland, M.Nitrofurantoin-Clinical and Laboratory Studies in Urinary TractInfections. Arch. Intern. Med., 96: 437-450, 1955.

39. Sasaki, T. Polarographic Study of Nitrofuran Derivatives. II.Reduction Potentials of Nitrofuran Derivatives and NitrobenzeneAnalogs. Pharm. Bull. Tokyo, 2: 104-108, 1954.

40. Szczukowski, M. J., Daywitt, A. L., and Elrick, H. MetastaticTesticular Tumor Treated with Nitrofurazone. J. Am. Med. Assoc.,767: 1066-1068, 1958.

2624 CANCER RESEARCH VOL. 32

Research. on August 23, 2019. © 1972 American Association for Cancercancerres.aacrjournals.org Downloaded from

1972;32:2616-2624. Cancer Res   J. D. Chapman, A. P. Reuvers, J. Borsa, et al.   Nitrofurans as Radiosensitizers of Hypoxic Mammalian Cells

  Updated version

  http://cancerres.aacrjournals.org/content/32/12/2616

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

  [email protected] 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/32/12/2616To request permission to re-use all or part of this article, use this link

Research. on August 23, 2019. © 1972 American Association for Cancercancerres.aacrjournals.org Downloaded from