P. L. OLIVE AND R. E. DURAND ... scribed previously (Sutherland and Durand, .... (dotted line is control distribution before incubation at room temperature).
Br. J. Cancer (1978) 37, Suppl. III, 124
ACTIVATION OF RADIOSENSITIZERS BY HYPOXIC CELLS P. L. OLIVE AND R. E. DURAND From the Department of Human Oncology, Wisconsin Clinical Cancer Center, 420 N. Charter Street, Madison, Wisconsin, U.S.A. 53706
Summary.-Hypoxic cells can metabolize nitroheterocyclic compounds to produce toxic intermediates capable of affecting the survival of neighbouring oxygenated cells. Mutagenesis experiments with E. coli WP-2 343 (deficient in nitroreductase) indicated that reduction of nitroheterocyclics outside bacteria causes killing and mutations within bacteria, presumably due to the transfer of the "active" specie(s). Using animal tissue slices to reduce nitrofurans, cultured L-929 cells incubated under aerobic conditions were far more sensitive to the toxic and DNA damaging effects of these drugs. Transfer of the active species also occurs in a tissue-like environment in multicell spheroids where the presence of a hypoxic central core served to convert the nitroheterocyclics to intermediates which also damaged the neighbouring oxygenated cells.
NITROHETEROCYCLIC compounds can mimic the effects of oxygen in sensitizing hypoxic mammalian cells to radiation damage. Extensive periods of incubation of cells with most nitrocompounds results in cell killing due to reduction of the nitro group to toxic intermediates. Unfortunately, those nitroheterocyclics that are in theory the best radiosensitizers are also the most likely to undergo metabolic reduction with subsequent loss of radiosensitizing power, since the same principle of electron affinity determines both properties. The mechanism of radiosensitization of hypoxic cells by electron affinic compounds is fairly well understood at a radiation chemical level (Adams, 1973). However, reduction of nitroheterocyclics to toxic intermediates may also alter the radiation response of tissues by preferentially damaging hypoxic or nutrientdepleted cells (Olive and McCalla, 1977), by combining with SH groups (Willson and Searle, 1975), or by altering cellular respiration resulting in changes in the size of the hypoxic population (Biaglow and Durand, 1976). Thus, metabolism of nitroheterocyclics has been shown to play an increasingly important role in their mode of action as radiosensitizers in vivo.
Experiments described below indicate that nitroheterocyclics reduced outside cells can influence the survival of cultured cells growing both as single cells or as spheroids. MATERIALS AND METHODS
Nitroheterocyclics.-Nitrofurans were obtained from Norwich Pharmacal Co., Norwich, N.Y. 5-methyl-3-(5'-nitrofuryl)pyrazole was obtained from ABIC Chemical Labs, Ramat-Gan, Israel. Niridazole and metronidazole were obtained from Dr J. Biaglow, Ro-07-0582 (misonidazole) from Dr C. Smithen, and the nitropyrrole, NP-6, from Dr J. Raleigh. Mutagenesis experiments.-E. coli WP2 343 (UVRA-, Try-, nfr-) was isolated by Dr D. R. McCalla (McCalla and Voutsinos, 1975). E. coli 343 was grown overnight in nutrient broth and transferred to fresh broth for 2 h before the experiment. About 109 bacteria were treated with nitroheterocyclics in 1% DMSO in minimal Davis buffer with 2% nutrient broth containing 1 ml S-9 mix (Ames et at., 1973). After 2 h incubation at 37°C, bacteria were centrifuged and resuspended in buffer for plating in triplicate on minimal agar plus 2% nutrient broth. Measurement of DNA damage.-Mouse L-929 fibroblasts were maintained and treated as previously described (Olive and McCalla,
125
RADIOSENSITIZATION OF HYPOXIC CELLS
1977). DNA damage was determined by measuring the proportion of DNA in singlestranded form using the technique of hydroxylapatite chromatography (Ahnstrom and Edvardsson, 1974). Spheroid experiments.-Chinese hamster V79 cells were grown as spheroids as described previously (Sutherland and Durand, 1976). The Sta-Put technique for cell separation based on size has been described by Miller and Phillips (1969) and applied to spheroids by Durand (1975). Cellularity (the number of cells per spheroid) was determined for each population of spheroids, and cell survival data as well as the sedimentation profile were corrected for cell loss. Incorporation of 3H-thymidine during the incubation period with the drug (002 pC/ml, 18 mC/ mmol) was used as a measure of cell growth. RESULTS AND DISCUSSION
4
z
a
ci a z
mJ
C,) Id
-j
0
a
TIME (HOURS) FIG. 1.-DNA damage by nitroheterocydes. Cultured mouse L-929 fibroblasts were incubated with 0 * 3 mM nitroheterocycles in buffer under anaerobic conditions. A decrease in double-stranded DNA is a measure of DNA damage. -0-, metronidazole (E1/2-495mv); - x -, NP-6 (E1/2-410 mv); - A -, misonidazole (-395 mV); - 0-, niridazole (-395 mV); -7-, nitrofurylpyrole (-358 mV); -A-, nitrofurazone (-282 mV).
DNA damage in cultured L cells The amount of DNA damage produced in hypoxic L-929 cells was in approximate agreement with the electron affinity of these compounds (Fig. 1). However, L-929 cells incubated in buffer equilibrated with air were damaged when drug-reducing slices of mouse tissues were included, and DNA damage was accumulated at a rate proportional to the ability of that tissue TABLE.-Damage to L Cell DNA by to metabolize or "activate" the nitroNitrofurazone Reduced in Mouse Tissue furan (Table). Thus, activation occurring Slices within the hypoxic tissue slices resulted in Rate of Reduction % Double(jp&M/h/g the production of a toxic intermediate that stranded DNA tissue) affected survival of neighbouring L cells. ControlTissue 0 9-1 ± 2*0 15-0 ± 1 8 0-10 ± 0-02* Brain 10-8 ± 2-0 0-28 ± 0-02 Mutagenesis as a result of nitroheterocycle Mammary 12-0 ± 5 6 0-31 4 0-02 Heart activation 16 3 ± 2 8 0-58 ± 0-14 Lung 20-5 ± 1.0 1-86 ± 0-14 The importance of DNA as a target in Kidney 24 4 ± 9 7 2 04 ± 0 48 Srnall intestine 27 0 ± 2 0 2 90 ± 0*60 the actions of toxic intermediates pro- Liver duced outside bacteria by mouse livpr * Mean ± standard error of the mean. L cells were incubated for 15 min in 4 ml buffer homogenate is evident in bacterial muitawith air containing 0 5 g tissue genesis experiments. A bacterial tester equilibrated strain that was previously resistant to the slices and 0 * 25 mM nitrofurazone. mutagenic effects of nitrofurans (due to a deficiency of nitroreductase) was induced that the concentration of nitroheterocyclic to revert to tryptophan independence producing a given number of revertants when an activating system was included correlated roughly with their reduction in the incubation medium (Fig. 2). Note potential. -
126
P. L. OLIVE AND R. E. DURAND
t NITROFURANTOIN
t 4
c c
NITROFURAZONE
11
z
_E
DRUG CONCENTRATION (pg/ml) Fio. 2.-Mutations as a result of exogenous nitroreduction. E. coli WP2 343 were incubated with nitroheterocycles as described in the text. Mouse liver microsomes plus NADPH generating system served as a source of nitroreductase.
A)B)
12
C)
C) 0
@
.00
.)
C)
C),42, 101
v
0)535 0,. la >l C
°-1.
0
50
0
f
Q
,
.
100/0 50 100/0 RELATIVE DISTANCE SEDIMENTED
50
, 1OO
FiG. 3.-Cytotoxicity of misonidazole in spheroids: requirement for hypoxic cells. Spheroids varying in hypoxic fraction were incubated for 14-16 h with 5 mm misonidazole in medium containing 5% serum. Fractions of cells separated on the Sta-Put were analysed for % total cells (cell distribution), % viable cells (corrected for cell loss) and ct/min 3H-thymidine/cell. (A) 6-day old spheroids containing 4,000 cells/spheroid. (B) 17-day old spheroids with about 40,000 cells. (C) 19-day old spheroids incubated for 2 days at room temperature before incubation with misonidazole at 370 (dotted line is control distribution before incubation at room temperature). Solid lines are untreated spheroids; symbols are drug-treated spheroids.
Sta-Put experiments with spheroids Histological evidence indicates that hypoxic centres of large spheroids become necrotic after treatment with metro-
nidazole (Sutherland and Durand, 1976). However, measurements of cell loss and plating efficiencies indicate that the proportion of cells affected by toxic nitro-
RADIOSENSITIZATION OF HYPOXIC CELLS
heterocyclic intermediates is far greater than expected if metronidazole were killing hypoxic cells only. Using the Sta-Put cell separation technique, the effects of misonidazole were observed on populations of cells within spheroids containing different proportions of hypoxic and non-cyclic cells. In large spheroids (more than 25,000 cells), treated with 5 mM misonidazole or 0-5 mm nitrofurazone for 16 h, there was extensive cell loss (54%) and cell killing. However, in smaller spheroids (about 4,000 cells) with few hypoxic cells, there was much less cell killing (Fig. 3A) and spheroids with 5-8 cells were resistant to 10 mM misonidazole. Furthermore, in large spheroids depleted of hypoxic cells by incubation at 20° for 2 days before treatment with misonidazole at 370, there was no cell loss relative to the room temperature control, and cell killing was greatly decreased (Fig. 3C). These spheroids would not be expected to contain hypoxic cells at the start of treatment since incubation at room temperature reduces metabolism of external cells allowing reoxygenation of the spheroid. Also, irradiation of these spheroids indicated that the resistant "tail" on the survival curve was absent. Sta-Put data also suggested that hypoxic cells were not only preferentially sensitive to the toxic effects of the nitroimidazole, but their "activation" of nitroimidazoles caused killing of other, normally resistant populations within the spheroid. With successively higher concentrations of misonidazole, more of the cells bordering the hypoxic area were killed as suggested by a shift in the survival curve from 100% survival in most of the large cells to 100% survival in none of the external (aerobic) cells (Fig. 4). Cellularity also decreased with increasing concentrations although cells lysed apparently within the spheroid since significant cell loss after treatment for 16 h was not accompanied by a decrease in spheroid diameter. Early preferential killing and lysis of hypoxic cells would prevent further activation, and for longer incubation periods, external cells
127
% TOTAL CELLS % VIABLE CELLS 2.0 mM 100 5 -S-0 0~ ~~~~5
3.8 mM
loo
5
IN 50 4.2 mM
100o
5
0~~~~~5 5.0mM
100_
O 50 10.0 mm
50 100/0
50
100/0
5-
50
loo
5
0
50
100
RELATIVE DISTANCE SEDIMENTED FiG. 4.-Effect of concentration of misonidazole on cytotoxicity in spheroids: killing of aerobic cells. Sixteen to 18-day old spheroids were treated for 14-16 h with miisonidazole in medium containing 5% serum. Cells treated with 10 mM misonidazole were incubated for 3 days.
were also lost from the spheroid (as indicated by a decrease in spheroid diameter). Thus the Sta-Put data indicated that (1) hypoxic cells within spheroids were required for cell killing, (2) both aerobic and hypoxic cells were killed and (3) with increasing concentrations of misonidazole, toxicity became less specific for hypoxic cells. REFERENCES ADAMS, G. E. (1973) Chemical Radiosensitization of Hypoxic Cells. Br. Med. Bull., 29, 48. AHNSTROM, G. & EDVARDSSON, K. (1974) Radiationinduced Single-strand Breaks in DNA Determined by Rate of Alkaline Strand Separation and Hydroxylapatite Chromatography: An Alternative to Velocity Sedimentation. Inter. J. Radiat. Biol., 26, 493. AMES, B. N., DURSTON, W. E., YAMASAKI, E. & LEE, F. D. (1973) Carcinogens are Mutagens: A Simple Test System Combining Liver Homogenates for Activation and Bacteria for Detection. Proc. natn. Acad. Sci., U.S., 70, 2281.
128
P. L. OLIVE AND R. E. DURAND
BIAGLOW, J. E. & DURAND, R. E. (1976) The Effects of Nitrobenzene Derivatives on Oxygen Utilization and Radiation Response of an In vitro Tumor Model. Radiat. Re8., 65, 529. DURAND, R. E. (1975) Isolation of Cell Subpopulations from in vitro Tumor Models According to Sedimentation Velocity. Cancer Re8., 35, 1295. McCALLA, D. R. & VoUTSINos, D. (1974) On the Mutagenicity of Nitrofurans. Mutat. Re8., 26, 3. MILLER, R. G. & PHILLIPS, R. (1969) A Separation of Cells by Velocity Sedimentation. J. Cell. Phy8iol., 73, 191.
OLIVE, P. L. & MCCALLA, D. R. (1977) Cytotoxicity and DNA Damage to Mammalian Cells by Nitrofurans. Chem.-Biol. Interact., 16, 223. SUTHERLAND, R. M. & DURAND, R. E. (1976) Radiation Response of Multicell Spheroids, an In vitro Tumor Model. Curr. Top. Radiat. Re8. Quart., 11, 87. WILLSON, R. L. & SEARLE, A. J. F. (1975) Metronidazole Catalysed Reaction with Sulphydryl Groups and Tumor Radiosensitization. Nature, Lond., 255, 498.
