Effects of Dimethyl Sulfoxide and Thiourea Upon Intercalator-Induced DMA Single-Strand Breaks in Mouse Leukemia (L1210) Cells

(CANCER RESEARCH 43, 5718-5724, December 1983]

Effects of Dimethyl Sulfoxide and Thiourea upon Intercalator-induced DMA Single-Strand Breaks in Mouse Leukemia (L1210) Cells

Yves Pommier,1 Leonard A. Zwelling, Michael R. Mattern, Leonard C. Erickson, Donna Kerrigan, Ronald Schwartz, and Kurt W. Kohn

Laboratory of Molecular Pharmacology. Developmental Therapeutics Program, Division of Cancer Treatment, National Cancer Institute, NIH, Bethesda. Maryland 20205

ABSTRACT bleomycin, generate free radical species which produce DNA breaks (5, 6), and it has been proposed that some intercalating The free radical scavengers, dimethyl sulfoxide (MezSO) and agents also break DNA by a free radical mechanism. In particular, thiourea, were used to assess the role of free radicals in the ADM2 can generate free radicals during NADPH and O2 reaction production of intercalator-induced DNA breaks and cytotoxicity with reducÃaseenzymes (2, 3, 14), and DNA strand breaks can in mouse leukemia L1210 cells. Both agents decreased X-ray be generated by this mechanism (4). break production, and this decrease was comparable in magni Free radical formation depends on reductive reactions which tude to the degree of inhibition of X-ray-induced cell killing. By are most likely to occur in those intercalating agents, incuding contrast, Me2SO increased the DNA breaks produced by the ADM, which contain quinone groups. However, several interca intercalators, Adriamycin, 5-iminodaunorubicin, and 4'-(9-acridi- lating agents, such as ellipticine, 2-Me-9-OH-E+, 5-ID, or m- nylamino)methanesulforHn-anisidide. This was not due to an AMSA, do not contain quinone groups and exhibit relatively little enhancement of Adriamycin or 4'-(9-acridinylamino)methane- free radical production (19, 23) yet still stimulate the formation sulfon-m-anisidide uptake by Me2SO. Strand break production of protein-associated DNA strand breaks in cells (27,33,36,37). by intercalators was decreased by thiourea. This was not due to Thus, intercalating agents could produce DNA strand breaks by an inactivation of the intercalators or to a decrease of Adriamycin 2 distinct mechanisms, but it is not dear whether the free radical or 4'-{9-acridinylamino)methanesulfon-/n-anisidide uptake by thi mechanism may make an appreciable contribution in living cells. ourea. Experiments using nudeokJ sedimentation to assess the Therefore, we investigated the effects of the free radical DNA linking number and domain size from cells treated with scavengers, Me2SO and thiourea, on intercalator-induced DNA Me2SO and thiourea indicated that these chemicals alter chro- breakage and cytotoxicity. These compounds reduce the extent matin structure in a fashion which may account for effects on of DNA breakage in cells exposed to ionizing radiation or hydro intercalator-induced DNA scission. The alterations in intercalator- gen peroxide (5, 22). In the course of these experiments, we induced DNA scission were not accompanied by corresponding found unexpectedly that Me2SO actually increased the interca alterations in cytotoxicity, thus dissociating intercalator-induced lator-induced DNA breakage. We propose that this phenomenon strand break production from lethality and the mechanism of X- may involve an effect on chromatin structure. ray break production. MATERIALS AND METHODS INTRODUCTION Materials. [2-'4C]Thymidine (58 mCi/mmol) and [mef/jy/-3H]thymidine A variety of DNA intercalating agents cause single- and double- (20 Ci/mmol) were purchased from New England Nuclear, Boston, Mass. strand DNA scission in mammalian cells (25-27, 33-37). This MejSO and thiourea were purchased from Fisher Scientific Co., Fairtawn, N. J. and from Eastman Organic Chemicals, Rochester, N. Y., respec DNA scission differs from the DNA breakage produced by other tively. ADM (NSC 123127) and m-AMSA (NSC 249992) were obtained DNA damaging agents in several respects, (a) The intercalator- from the Drug Synthesis and Chemistry Branch, Division of Cancer induced DNA breaks are associated with an approximately stoi- Treatment, National Cancer Institute; m-AMSA was dissolved in 100% chiometric amount of covalently bound protein which dissociates M62SO at 10 mM, and ADM was dissolved in glass distilled water at 1 at the same time the breaks are reseated (27, 36). (b) The mg/ml. 5-ID (NSC 254681) was a gift from Dr. Robert I. Glazer, Applied formation of the DNA breaks is saturable and is inhibited at low Pharmacology Section, Laboratory of Medical Chemistry and Biology, temperature but wilt occur subsequently in the presence of drug National Cancer Institute, who obtained the compound from Dr. E. Acton, Stanford Research Institute. 5-ID was dissolved in glass distilled water when the temperature is raised (36). (c) The formation and at 1 mM. 2-Me-9-OH-E+ was a gift from Dr. J. B. Le Pecq, Laboratoire resealing of the breaks is not accompanied by stimulation of de Pharmacologie MolÃ©culaireau Centre National de la Recherche Scien poly(adenosine diphosphoribose) synthesis (35). (d) The break tifique, Institut Gustave-Roussy, Villejuif, France, and was dissolved in resealing can occur in isolated nuclei and permeabilized cells in glass distilled water at 8.25 mw. [14C]m-AMSA (19.6 mCi/mmol) and the absence of nucleoside triphosphates (24, 35). These findings [MC]ADM (17.2 mCi/mmol) were synthesized by SRI International, Menlo taken together suggest a mechanism of DNA scission coordi Park, Calif., and were obtained through the Chemical Resources Section, nated with protein-DNA binding having its origins in an enzyme National Cancer Institute. [14C]m-AMSA and [14C]ADM were dissolved constituent of the cell nucleus, perhaps a topoisomerase (27, in 100% Me2SO and in glass-distilled water, respectively. ADM, 5-ID, m- 36). 2The abbreviations used are: ADM, Adriamycin; 2-Me-9-OH-E+, 2-methyl-9- Some DNA damaging agents, such as ionizing radiation and hydroxyellipticinium; 5-ID, 5-iminodaunorubicin; m-AMSA, 4'-(9-acridinylamino)methanesulfon-m-aniskJ"tde; Me^O, dimethyl sulfoxide; [14C]m-AMSA, 4'-(9- acridinyl-[9-'4C]amino)methanesulfon-m-anisidide; [14C]ADM, Adriamycin hydro- 1To whom requests for reprints should be addressed, at Building 37, Room chloride-(14-"C]; RPMI 1630, Roswell Park Memorial Institute tissue culture Me 5D17, 9000 Rockvilte Pike, Bethesda, Md. 20205. dium 1630; DPC, DNA-protein cross-links; SSB, single-strand breaks; SDS, sodium Received May 26,1983; accepted August 25,1983. dodecyl sulfate.

5718 CANCER RESEARCH VOL. 43

AMSA, and 2-Me-9-OH-E+ were stored frozen in stock solutions. [log (f,/ro)] [tog (flo/ro)] ' Cell Labeling, Irradiation, and Drug Treatment. L1210 mouse leu kemia cells were grown in suspension culture in RPMI 1630 supple mented with 15% fetal calf serum. Stock cultures were maintained in were PB is the DNA break frequency produced by the X-ray (300 or 1000 static bottles without antibiotics and were used to initiate suspension rad-equivalents), and r,, ro, and Ã±oare the retention (16) of DNA from drug-treated, untreated, and 300- or 1000-rad-treated [14C]thymidine- cultures. Cultures used to assess drug effects were in exponential growth phase with a doubling time of 13 to 15 hr. labeled cells. Retention was evaluated at the time corresponding to retention of 0.35 of the [3H]DNA in the high-sensitivity assay or 0.60 of Cellular DNA was radioactively labeled in exponentially growing cells by incubation with [2-uC]- or [mef/7y/-3H]thymidine for 20 hr at 37Â°.For the [3H]DNA in the low-sensitivity assay. The exact choice of this end experiments in which DNA SSB and DPC were measured by alkaline point was not critical, since the elution kinetics were nearly first-order elution, [14C]thymidine labeling of DNA was at 0.01 Â¿tCi/ml,and the DNA following all drug treatments. Results are expressed as "rad-equivalents," of cells used as internal standard (16) was labeled with [mef/?y/-3H]- indicating that the elution rate of DNA from drug-treated cells is equal to thymidine(0.1 //Ci/ml; 10"6 M unlabeled thymidine added). In experiments that produced by that X-ray dose. in which DNA SSB were measured by alkaline sucrose sedimentation, DNA-Protein Cross-Linking. Drug-treated (or untreated control) [14C]- and [3H]thymidine labeling concentrations were increased 20- and [14C]thymidine-labeled cells were X-irradiated at ice temperature (4Â°)with 10-fold, respectively, as compared to alkaline elution experiments. In the 3000 rads. These cells were combined with an equal number (approxi mately 5 x 105) of [3H]thymidine-labeled cells which had received no nucleoid sedimentation experiments, cellular DNA was labeled with [mef/)y/-3H]thymidine (0.4 Â¿tCi/ml;10"6 M unlabeled thymidine added). In drug treatment and had been concurrently irradiated with 3000 rads in all cases, radioactive label was removed by centrifugation prÃorto drug ice. Cells were deposited on a 2-^m polyvinyl chloride filter (Type BS; treatment or irradiation of the cells. Millipore) and lysed with a solution consisting of 2 M NaCI, 0.2% sarkosyl, L1210 cells at a concentration between 1 and 1.5 x I06/ml in iced and 0.04 M disodium EDTA, pH 10.0 (5 ml). This lysis solution was RPM11630 plus 15% fetal calf serum were irradiated with either a 137Cs removed by washing the filter with 0.04 M EDTA, pH 10.0 (3 ml). DNA or a 200-kV X-ray source as described previously (16, 36). Cells were elution was performed with tetrapropylammonium hydroxide-EDTA, pH maintained at ice temperature until they were assayed either by alkaline 12.1, at a pump speed of 0.03 to 0.04 ml/min. Fractions (6 ml) were elution or alkaline sucrose sedimentation. All drug treatments were at collected every 3 hr for 15 hr. Samples were processed, and data were 37Â°. Cells were incubated with Me2SO (0.64 M = 5% v/v) or thiourea computed as described previously (16). DPC frequencies (Pâ€ž)were (0.1 M) for 10 min and then with intercalators plus Me2SO or thiourea for calculated using the bound to one terminus model of Ross ef al. (27) as: 30 min. Reactions were stopped by adding excess iced medium to cells P. = [(1 - T1 - (1 - ror1] PB in drug-containing medium followed by centrifugation and resuspension twice in iced drug-free medium. Under these conditions, neither Me2SO where PB is the break frequency (in rad-equivalents) produced by 3000 nor thiourea was cytotoxic. rads, and ro and r are the retention of DNA on the filter from 3000-rad- Isolation of L1210 Cell Nuclei. The procedure of nuclei isolation has irradiated [3H]- or [14C]thymidine-labeled cells, respectively. The degree been described previously (12, 24). L1210 mouse leukemia cells were to which r exceeds ro is a measure of DPC. centrifuged and resuspended in nuclei buffer (150 mw NaCI, 1 mw Alkaline Sucrose Sedimentation. After determination of the cell KH2PO4, 5 mw MgCI;>, 1 mw [ethylenebis(oxyethenenitrilo)]tetraacetic number using a Coulter Counter Model ZBI (Coulter Electronics, Hialeah, acid, and 0.1 mw dithiothreitol, pH 6.4) at 4Â°.These cells were centifuged Fla.), a volume of medium containing 105 cells was centrifuged at 1000 again and resuspended in Vio volume ice-cold nuclei buffer. Then 9/io x g for 10 min at 0Â°.The medium was decanted, and the cell pellet was volume of 4Â°nuclei buffer containing 0.3% Triton X-100 was added and resuspended in 0.4 ml of ice-cold phosphate-buffered saline (0.15 M the mixture incubated for 10 min at 4Â°. The nuclei were pelleted by NaCI, 0.71 HIM KH2PO4, and 4.28 HIM K2HPO4). Cell lysates were centrifugation (1200 rpm for 5 min) and resuspended in nuclei buffer at prepared by adding the following components (11): 0.1 ml of [14C]- 37Â°. Nuclei were examined microscopically after staining with trypan thymidine-labeled cells; 0.1 ml of [3H]thymidine-labeled cells, 50 Â¡Aof a blue to confirm their permeability and the absence of cytoplasm, m- solution containing 10 mg of proteinase K per ml, 2% SDS, 5 mw CaCI2, AMSA treatments were performed for 30 min at 37Â°and stopped by a 1 IHM disodium EDTA, and 0.01 M Tris, pH 8; and 0.2 ml of 2% sarkosyl Vzodilution of treated nuclei in drug-free nuclei buffer at ice temperature. (Ciba Geigy, Ardsley, N. Y.). The tubes were rocked gently to mix the Alkaline Elution Assays. The alkaline elution methodology has been components. The lysate was incubated at 50Â°for 1 hr and layered gently described in detail in previous publications (16, 36). onto a 4 to 20% linear sucrose gradient with a wide-bore 1-ml pipet. DNA SSB. Treated (or untreated control) [14C]thymidine-labeled cells Alkaline sucrose gradients were prepared from stock solutions con (approximately 5 x 105) were mixed with an equal number of [3H]- taining: (a) 4% sucrose, 0.9 M NaCI, 0.05 M Na2EDTA, and 0.2 M NaOH; thymidine-labeled cells which had received no drug treatment and had and (b) 20% sucrose, 0.9 M NaCI, 0.05 M Na2EDTA, and 0.3 M NaOH. been concurrently irradiated to serve as internal standard cells (16). Cells The gradients were centrifuged in cellulose nitrate tubes in a Beckman were deposited gently on polycarbonate membrane filters (2-nm pore SW-40 Ti rotor at 11,000 rpm, at 20Â°for 16 hr (u2f = 7.53 x 1010 rad2 diameter; Nucleopore Corp., Pleasanton, Calif.) in a Swinnex 25 filter sec). The gradients were fractionated from the top by displacement with holder (Millipore Corp., Bedford, Mass.) and lysed with 0.1 M glycine, a solution of 60% sucrose/1 M NaCI. Fractions (0.5 ml) were collected in 0.025 M disodium EDTA, and 2% SDS (BDH BiomÃ©dical Ltd., Poole, scintillation vials and mixed with 3 ml of distilled water and 10 ml England), pH 10.0, plus proteinase K (0.5 mg/ml). Elution was performed Aquassure (New England Nuclear) containing 0.7% glacial acetic acid. with tetrapropylammonium hydroxide-EDTA-0.1% SDS, pH 12.1, using DNA sedimenting to the bottom of the tube was recovered by cutting off a peristaltic pump to control flow rate. For the high sensitivity assay, the tube bottom and washing it with 1 ml of 0.1 M NaOH/0.3% sarkosyl. elution was carried out at a pump speed of 0.03 to 0.04 ml/min (2 ml/ The samples were counted in a Packard Tri-Carb 2450 B liquid scintilla hr), and fractions were collected at 3-hr intervals over 15 hr as described tion spectrometer (Packard Instruments, Chicago, III.). Weight average previously (16). For the low-sensitivity assay (which was required to molecular weight was computed as described by Dingman and Kakunaga assay DNA break frequencies over 600 rad-equivalents), elution was (10). Induced break frequencies were computed as described previously carried out at a pump speed of 0.12 to 0.16 ml/min, and fractions were (11). collected at 5-min intervals for 30 min (36). The [3H]DNA internal standard Nucleoid Sedimentation. Following treatment, [3H]thymidine-labeled cells received 300 rads of X-irradiation in the high-sensitivity assay and cells were pelleted by centrifugation (1000 rpm for 5 min), and the pellets 1000 rads in the low-sensitivity assay. SSB frequency was calculated were resuspended at 2 x 105 cells/ml in 2-ml neutral lysis solution (0.1% by: Triton X-100, 0.02 M EDTA, and 0.01 M Tris, pH 8). After 15 min of

DECEMBER 1983 5719

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1983 American Association for Cancer Research. Y. Pommier et al. incubation at room temperature, 1 to 2 x 105 cells were layered onto preformed 15 to 30% neutral sucrose gradients [36-ml total volume 3000 containing 1.9 M NaCI, 0.1 M Tris, and 0.01 M EDTA, pH 8; modification - of the procedure of Cook and Brazell (8)]. For experiments in which ethidium bromide titration of DMA supercoiling was carried out, the 15 2000 and 30% neutral sucrose solutions contained, in addition, various con IÃœ400 centrations (0 to 40 ng/ml) of ethidium bromide (Sigma Chemical Co., Â«<-o tS m M 1000 St. Louis, Mo.). The samples were centrifuged after an additional 45-min 3 â€” incubation in a Beckman L8-80 ultracentrifuge (SW 28 rotor; 20Â° at 15,000 to 17,000 rpm for 1.5 to 2.5 hr). Equivalent weight fractions were collected from the bottom of the tubes, and the trichloroacetic acid- 400 800 1000 2000 3000 insoluble radioactivity contained in each fraction was determined by liquid ADMINISTERED y- OR X-RAY DOSE (Rads) scintillation spectrometry, as described previously (21). The distance Chart 1. Inhibitory effects of Me2SO and thiourea on the DNA SSB produced sedimented (peak fraction) by "control" nucleoids (one sample/run, which by X-rays. Me2SO (0.64 M) and thiourea (0.1 M) were added to cell suspensions at 37Â°, 10 min prior to -y-irradiation W or X-irradiation (B), and were kept in the consisted of nucleoids from cultures that were not treated with Me2SO medium during irradiation at ice temperature. DNA SSB frequency was assayed or thiourea) was set equal to 100%, and the percentage of control by alkaline elution with deproteinization at a pump speed of 2 ml/hr (A) or 8 ml/hr nucleoid sedimentation was calculated for each sample by dividing the (B). Control, â€¢;Me2SO,A; thiourea, O. nucleoid sedimentation distance (peak fraction) of the treated sample by that of the control and multiplying this ratio by 100% (21). Produced by Intercalating Agents. The SSB frequency pro Transport Studies. This technique was similar to that used previously duced by a 30-min treatment of L1210 cells with ADM or 5-ID by Zwelling ef al. (35). Exponentially growing L1210 cells were concen trated to approximately 107/ml in RPM11630 plus 15% fetal calf serum. was approximately proportional to the drug concentration of up After a 30-min incubation with the radioactively labeled intercalating to 8.7 and 5 /Â¿M,respectively(Chart 2, upper panels). By contrast, the capacity of m-AMSA to produce SSB tended to level off at drugs in the presence or the absence of Me2SO (0.64 M) or thiourea (0.1 M), which had been added to the medium 10 min before, 1 ml of the cell concentrations greater than 3 pM. However, the magnitude of suspension was layered above 0.6 ml of silicone oil (Versilube F50; the maximum m-AMSA-induced SSB frequency was much General Electric Co., Waterford, N. Y.) in a microcentrifuge tube and greater than that produced by the anthracyclines (Chart 2). In centrifuged for 1 min at 12,000 x g. The bottom of the tube was cut off the case of 2-Me-9-OH-E+, the apparent SSB frequency tended and placed into a liquid scintillation counting vial. The cell pellet was to level off at lower values than in the case of m-AMSA (Chart dispersed in 1.5 ml of phosphate-buffered saline and solubilized by the 2, lower panels). addition of 1.5 ml of 0.4 M NaOH and incubated overnight at 37Â°.The Me2SO (0.64 M) increased the DNA SSB frequency produced sample was prepared for scintillation counting by adding 3 ml of water by ADM, 5-ID, and m-AMSA at all the concentrations used (Chart and 10 ml of Aquassure (New England Nuclear) containing 0.7% glacial 2). The shapes of the DNA SSB frequency curves observed for acetic acid. cells treated with Me2SO were similar to those observed in The amount of extracellular fluid trapped in the cell pellet was esti mated by the use of [14C]inulin (New England Nuclear) to be approxi untreated cells. In the case of 2-Me-9-OH-E+, Me2SO decreased mately 1 to 3 Ml/107 cells. Small corrections were applied on the basis of the DNA SSB frequency (Chart 2, lower right panel). Thiourea this value and the determined radioactivity in the supernatant. (0.1 M), on the other hand, reduced the DNA SSB frequency Chemical Inactivation Studies. ADM (1.72 mw). m-AMSA (0.1 mw), produced by all 4 intercalators (Chart 2). or 2-Me-9-OH-E+ (10 mw) was added to RPM11630 plus 15% fetal calf The enhancement of the m-AMSA-induced DNA strand break serum containing either Me2SO (0.64 M) or thiourea (0.1 M). These age by Me2SO was dependent on Me2SO concentration, with mixtures were incubated at 37Â°for 40 min under continuous agitation 0.64 M giving the maximum effect without significant toxicity by and then assayed as stock solutions for their capacity to induce breaks itself (Chart 3). Alternatively, thiourea reduced DNA breakage in in cellular DNA as described previously. They were compared in the a concentration-dependent manner; 0.1 M was the most efficient same experiments with stock solutions containing neither Me2SO nor noncytotoxic thiourea concentration (Chart 3). thiourea. Their effects were also compared with those produced by low Lack of Effect of Me2SO or Thiourea on the Cellular Uptake concentrations of Me2SO or thiourea, which were identical to the final of ADM and m-AMSA. The uptake of [14C]ADM by L1210 cells concentrations of Me2SO or thiourea achieved by the dilution of the Me2SO- or thiourea-containing stock drug solutions. was concentration-dependent and tended to level off slightly at Cell Survival. Cells were treated as above and assayed for colony- concentrations above 7 MM- The uptake of [14C]m-AMSA was forming ability in soft agar by the method of Chu and Fisher (7). strictly proportional to the drug concentration in the 1 to 5 ^M range as reported previously (34). Neither Me2SO (0.64 M) nor thiourea (0.1 M) modified the uptake of ADM or m-AMSA (Chart RESULTS 4). Effects of Me2SO and Thiourea on DNA SSB Produced by Formation and Reseating of DNA SSB Produced by m- Ionizing Radiation. The DNA SSB frequency produced by y- or AMSA in the Presence of Me2SO or Thiourea. The SSB pro X-radiation as quantified by alkaline elution was proportional to duced by m-AMSA form and reverse rapidly at 37Â°(36). Within the dose (Chart 1). At the doses chosen, neither Me2SO (0.64 M 5 to 10 min of treatment, a plateau was reached (Chart 5). = 5% v/v) nor thiourea (0.1 M) modified the elution kinetics of Removal of m-AMSA by washing the cells twice in medium at DNA from untreated mouse leukemia L1210 cells, indicating that 37Â°was followed by a rapid disappearance of SSB (Chart 5). neither agent produced DNA SSB. Both Me2SO and thiourea Addition of Me2SO in the presence of m-AMSA immediately reduced the number of DNA-strand breaks produced by ionizing increased the DNA SSB frequency. This effect was rapidly radiation. This finding is consistent with the free radical scaven reversible after Me2SO removal (while m-AMSA remained). The ger activity of Me2SO and thiourea. opposite phenomenon was produced by thiourea (Chart 5). Effects of Me2SO and Thiourea on DNA Strand Breaks When Me2SO or thiourea was added to the medium just after

5720 CANCER RESEARCH VOL. 43

ADM m-AMSA

S x 5 H' d =â€¢3 8

DRUG CONCENTRATION I .Ml Chart 4. Dependence of cellular uptake of [14C]ADM and [14C]m-AMSA on drug concentration in the absence (â€¢)orin the presence of Me2SO (A) or thiourea (O). L1210 cells (107/ml) were incubated with Me2SO (0.64 M) or thiourea (0.1 M) for 10 min prior to and during treatment with [14C]ADM (0.015 /iCi/ml) or ["C]m-AMSA (0.01 /iCi/ml) for 30 min at 37Â°.

~~l T T T~ ~T T T 3000 DMSO In DMSO Ou- ui m-AMSA Removed O * * Â« 2000 KLU "5 Â» oÃ

If 01234S 020406080 100 P 1000 DRUG CONCENTRATION (HM) a Chart2. Effects of Me2SO and thiourea on intercalator-induced DNA SSB o frequency. DNA SSB frequency was measured in the absence (â€¢)orin the presence in of Me2SO (A) or thiourea (O). Cells were pretreated with Me;,SO (0.64 M)or thiourea (0.1 M) for 10 min prior to and during the intercalator treatments for 30 min at 37Â°. Cells were then washed at 4Â°by centrifugaron in iced RPM11630, and SSB were 20 40 60 80 assayed by alkaline elution using the protemase K method either at a pump speed TIME FOLLOWING DRUG ADDITION (Minutes) of 2 ml/hr in the case of ADM, 5-ID, and 2-Me-9-OH-E+, or at a pump speed of 8 ml/hr in the case of m-AMSA. Chart 5. Kinetics of formation and disappearance of SSB assayed by alkaline elution using proteinase K in L1210 cells exposed to 1 UM m-AMSA for 60 min at 37Â°. After 20 min, cells were centrifugea and resuspended in warm medium DMSO CONCENTRATION (M) containing either m-AMSA alone (â€¢)or m-AMSA plus Me2SO (0.64 M) (A) or thiourea (TU) (0.1 M) (O). At 40 min, both MejSO and thiourea were removed by 0.2 0.4 0.6 rapid centrifugation and resuspension of cells in m-AMSA containing medium. After 60 min, m-AMSA was removed (arrow), and DNA rejoining was studied in the absence (â€¢)orin the presence of Me2SO (0.64 M) (A) or thiourea (0.1 M) (O).

Equivalence of DNA SSB and DPC in the Presence of Me2SO and Thiourea. The SSB/DPC ratio for both m-AMSA (0.5 and 1 UM) and 5-ID (2 and 5 ^M) were in agreement with previous data (33, 36) (Table 1). Neither Me2SO nor thiourea modified this ratio, which is reflected by the fact that the SSB/ DPC ratios in the presence of Me2SO or thiourea were uniformly distributed among the overall values (mean, 0.69; range, 0.53 to 1.40) and that the S.D. value of the overall data (0.26) was relatively small as compared to their mean value (0.75) (Table 1). Effect of Me2SO on DNA SSB Produced by High Concentra tion of m-AMSA. Above 5000 rad-equivalents (4.5 breaks/106 nucleotides), the elution kinetics of DNA is very rapid, and most of the DNA Ã©lÃ»tesfromthe filter in the lysis and in the first 0.02 0.04 0.06 0.08 0.1 fractions, making accurate determination of the DNA SSB fre THIOUREA CONCENTRATION (M) quency difficult. Alkaline sucrose sedimentation assays, which Chart 3. Dependence of m-AMSA-induced DNA SSB on Me2SO or thiourea are accurate at these higher SSB frequencies, were, therefore, concentration. Cells were treated with the indicated concentrations of Me?SO (A) used to study the concentration dependence of m-AMSA-in or thiourea (O) for 10 min prior to and during m-AMSA treatment (1 JIM)for 30 min at 37Â°.Cells were then washed in iced RPMI 1630, and SSB were assayed by duced SSB at higher drug doses. DNA SSB frequency after alkaline elution using the proteinase K method at a pump speed of 8 ml/hr. treatment of L1210 cells with 2.5 /JM m-AMSA for 30 min were in good agreement with those determined by alkaline elution m-AMSA removal, no modification of the DNA rejoining kinetics using proteinase K (Charts 2 and 6). The stimulatory effect of was observed (f% of DNA SSB disappearance, 10 to 20 min) Me2SO was found even at 10 MMm-AMSA, when the m-AMSA (Chart 5). SSB tended to level off (Chart 6). The process by which m-

DECEMBER 1983 5721

Table 1 Relative production of DNA SSB and DPC in the absence or in the presence of MeÃ®SOCO.64u) or thiourea (0.1 u) SSB were assayed by alkaline elution using the proteinase K method. DPC were assayed by alkaline elution without proteinase K. SSB and DPC were determined on aliquots of the same cell suspension and are expressed in rad-equivalents.

equiva Intercalatortreatmentm-AMSA treatmentNone alents)1118.5 lents)799.4 teincross-links1.40 MM)m-AMSA(0.5 MezSO 1479.8 1919.0 0.77 ThioureaNone 748.21474.8 681.72442.1 1.100.60

MM)5-ID (1.0 MejSOThioureaNone201 1.4 2543.7 0.79 1083.3115.4 1555.3218.6 0.700.53

(2.0MM)5-ID Me2SO 337.4 549.1 0.61 m- AM SA CONCENTRATION L Mi ThioureaNone 117.6370.0 244.6616.7 0.480.60 Chart 6. DNA SSB frequencies after 30-min treatments with m-AMSA alone p, (5.0MM)MeanConcurrent â€¢)orin the presence of Me2SO (0.64 M) (D, O), assayed by alkaline sedimentation MejSO 782.3 986.1 P. O) or alkaline elution (â€¢,O).Cells were treated with Me.-SO at 10 min prior to 0.79 and during the 30-min m-AMSA treatment, and were washed twice at 4Â° by ThioureaSSB(rad-equiv277.1DPC(rad-413.5SSB/DNA-pro-0.670.75 centrifugation before being assayed by alkaline elution or alkaline sucrose sedi 0.26a* Â± mentation. Mean Â±S.D.

AMSA produced DNA SSB, thus, was saturable with respect to *â€¢Â£2000 the drug concentration, and Me2SO raised the level of this plateau. CD Â§ Effects of Me2SO and Thiourea in Isolated L1210 Nuclei. Os 1500 Nuclei can be isolated from L1210 cells and the DNA effects of intercalators assayed by alkaline elution (12, 24). The isolation procedure (see "Materials and Methods") produced no more than - 1000 one DNA SSB/107 nucleotides in untreated cells. Irradiation of nuclei increased the slope of the DNA elution proportionally to the y- or X-ray dose (24). A similar pattern was seen after m- < 500 AMSA treatment, permitting the expression of DNA SSB in rad- lE equivalents. The yield of SSB produced by m-AMSA in isolated nuclei is 1 lower than in cells (24), but the curve displays apparent saturation m-AMSA CONCENTRATION (MM) above concentrations of 8 MM(Chart 7) as previously shown in Chart 7. Dependence of SSB frequency on m-AMSA concentration in isolated nuclei from L1210 cells. Nuclei were treated at 37Â°with m-AMSA in the absence whole cells (Chart 2). Treatment of nuclei with Me2SO or thiourea alone for 30 min at 37Â°had no effect on DNA elution as compared (â€¢)or in the presence of Me2SO (0.64 M) (A) or thiourea (0.1 M) (O) for 30 min. Drugs were removed by a Vio dilution of isolated nuclei in iced nuclei buffer. DNA to control. Me2SO slightly increased the DNA SSB produced by SSB were then assayed by alkaline elution using proteinase K at a pump speed of m-AMSA in isolated nuclei but to a lower extent than in whole 8 ml/hr. cells (Chart 2) and only at the lowest concentration of m-AMSA. However, the inhibitory effect of thiourea was as clear in isolated nucleoids from cells treated at 37Â°for 40 min with 0.64 M Me2SO nuclei as in whole cells. was similar to that of nucleoids from control cells (Chart 8), Effects of Me2SO and Thiourea on Nucleoid Sedimentation. indicating that Me2SO did not detectably change the overall Nucleoids consist of double-stranded DNA arranged in super- amount of compactness. However, these nucleoids differed from coiled loops that are probably maintained by the interaction of those from control cells in their response to ethidium bromide. DNA with nonhistone or structural proteins that resist dissocia The relaxing effect of ethidium bromide was similar up to 3 /tg/ tion by 1.9 M NaCI. Their rate of sedimentation in 15 to 30% ml, but its maximum relaxing effect was extended in that nu neutral sucrose gradients (pH 8) containing 1.9 M NaCI is a cleoids from Me2SO-treated cells relaxed more than those from function of their compactness (8, 28). The relative ability of the control cells (Chart 8), and the positive supercoiling ("rewinding") intercalating dye ethidium bromide to relax nucleoids by remov phase did not occur at concentrations of ethidium bromide lower ing negative supercoils from the DNA gives an estimate of in than 15 ^g/ml. In contrast, treatment of cells with 0.1 M thiourea vivo linking number (8, 9). The maximum relaxation of nucleoids for 40 min at 37Â°markedly compacted the nucleoids, producing (minimum compactness) from untreated cells was produced by a 46% increase in sedimentation rate (median, 150%; range, 112 ethidium bromide at 3 ^g/m\ (Chart 8). The increase in nucleoid to 169; 8 determinations). The ethidium bromide responses of sedimentation rate in gradients containing a high concentration nucleoids from cells treated with thiourea for 40 min at 37Â°were of ethidium bromide is thought to result from the introduction of also different from those of control cells. More ethidium bromide positive supercoils in nucleoid DNA. The sedimentation rate of (5 /Â¿g/ml)was required to relax competely the DNA of these

5722 CANCER RESEARCH VOL. 43

< O 0.5 1024680 2 40 2 404080 120 I x 3*â„¢!â„¢?â„¢!;?., DRUG CONCENTRATION I "' B. I ' i ' i 8 30

ETHIDIUM BROMIDE CONCENTRATION U/g/ml) Charts. Sedimentation of nucleoids from untreated cells (â€¢)and from cells treated for 40 min with Me.,SO (0.64 M) (A) or thiourea (0.1 M) (O) in neutral sucrose gradients (15 to 30% sucrose) containing various concentrations of ethidium bromide. Cells were lysed for 1 hr at 22Â°and then layered atop sucrose gradients just before ultracentrifugation. more compact structures (Chart 8). m-AMSA .'M.-4.HI I Relationship between DNA Breaks and Cell Survival. Both Me2SO and thiourea reduced the cytotoxicity of -y- or X-ray. ililil . I . I i I 0 0.5 10 0.2 0.4 0 0.5 10246 0.2 0.4 0.6 When the measured SSB frequency at each dose was plotted on the abscissa (Chart 1) instead of the administered X-ray dose MEASURED DMA SINGLE-STRAND BREAK FREQUENCY Ik Rad equivalents) (Chart 9A), it appeared that both Me^O and thiourea comparably Chart9. Survival of colony-forming ability of L1210 cells treated with X-ray in reduced the number of radiation-induced DNA strand breaks and ice or with ADM, 5-ID, m-AMSA, and 2-Me-9-OH-E+ for 30 min at 37Â° in the the cytotoxicity, indicating a possible mechanistic relationship absence (â€¢)orin the presence of Me2SO (0.64 M) (A) or thiourea (0.1 M) (O) (A). between cytotoxicity and the putative free radical-mediated DNA Drugs were removed by centrifugation in cold RPMI 1630 prior to soft agar colony formation assays. Relationship between SSB frequency and cytotoxicity in the scission. absence (â€¢)orin the presence of Me2SO (A) or thiourea (O) (B). Me2SO also decreased intercalator-induced cytotoxicity (Chart 9A) but, as previously shown, increased the SSB frequency radicals. Thus, these data argue against a general free radical produced by ADM, 5-ID, and m-AMSA (Chart 2). Therefore, mechanism of intercalator-induced DNA scission and suggest when the measured SSB frequency was plotted on the abscissa that the cytotoxicity of intercalators, which is reduced both by instead of the administered drug concentration, more DNA SSB Me2SO and thiourea, is not closely related to their DNA breaking were observed at a given level of cytotoxicity in the presence of activity (37). Our previous work has led us to propose that the intercalator- Me2SO (Chart 9B) indicating a dissociation between DNA SSB and cell lethality in the case of ADM, 5-ID, and m-AMSA. 2-Me- induced DNA scission may arise as an effect of drug intercalation. 9-OH-E+ differed from the other intercalato^ in that Me2SO Intercalator binding would trigger nuclear enzymes (27,36) such decreased the DNA breaks (Chart 2) and did not exhibit a as DNA topoisomerase, which would bind to and break DNA dissociation between DNA breaks and cytotoxicity (Chart 96, (13). right panel). Thiourea decreased both the DNA breaking activity Alteration of intracellular topoisomerase is not a likely expla (Chart 9) and the cytotoxicity of the 4 Â¡ntercalators(Chart 9). nation for these results because of the rapidity and the reversi bility of the Me2SO and thiourea effects (Chart 5). Direct altera DISCUSSION tions of the enzyme activity cannot be ruled out, since Me2SO has been shown to affect the activity of nuclear enzymes such The initial question posed in this study was whether free radical as DNA ligase, DNase I (29), DNA polymerase III (15), and DNA scavengers would inhibit the formation of DNA breaks in mam /J-polymerase (17). malian cells treated with intercalating agents. The ability of the Neither Me2SO nor thiourea alter cellular drug uptake (Chart free radical scavengers, Me2SO and thiourea, to inhibit the 4). However, either agent could alter drug-DNA binding. In the production of DNA breaks by ionizing radiation (1, 22) was case of Me. SO, enhanced DNA binding is unlikely, since Me2SO confirmed (Chart 1). Thiourea inhibited the production of protein- actually decreases cytotoxicity which presumably derives from associated DNA breaks by Â¡ntercalators,as would be expected DNA-drug interaction (Chart 9). In the case of the partial inhibition if it were scavenging free radicals. However, MezSO, which did of DNA scission and cytotoxicity of thiourea, however, this not itself produce DNA breaks (30, 32), increased the break hypothesis cannot be eliminated. frequencies produced by ADM, 5-ID, and m-AMSA; this is un If drug-DNA binding is altered, it may result from an alteration expected for drug-induced DNA breakage mediated by free of the chromatin target. Alteration in DNA linking number or in

DECEMBER 1983 5723

DNA domain size, as defined by protein-bound loops (8,9), could 12. Filipski,J., and Kohn, K. W. Ellipticine-inducedprotein-associatedDNA breaks result in altered DMA-drug interaction which would alter the in isolated U 210 nuclei. Biochim. Biophys. Acta, 698: 280-286,1982. 13. Geliert, M. DNA topoisomerases.Ann. Rev. Biochem., 50: 879-910,1981. intercalator-dependent DMA breaking activity. Nucleoid sedimen 14. Manda,K., and Sato, S. Generationof free radicalsof quinonegroup-containing tation studies indicated that thiourea and Me2SO may alter anticancer chemicals in NADPH-microsomesystem as evidenced by initiation of sulfite oxidation. Gann, 66: 43-47, 1975. chromatin structure as reflected by changes in DNA linking 15. Heinze, J. E., and Cari, P. L. The effects of organic solvents on Escherichia number and/or domain (loop) size. coli DNA polymeraseIII. Biochim. Biophys. Acta, 402: 35-40, 1975. Thiourea produced 3 changes in nucleoid sedimentation: (a) 16. Kohn, K. W., Ewig, R. A. G., Erickson, L. C., and Zwelling, L. A. Measurement of DNA strand breaks and cross-links by alkaline elution. In: E. C. Friedberg an increase in sedimentation (in the absence of ethidium); (b) an and P. C. Hanawalt (eds.), DNA Repair. A Laboratory Manual of Research increase in ethidium concentration required to completely relax Techniques, pp. 379-401. New York: Marcel Dekker, 1981. the nucleoids; and (c) an increase in the sedimentation of nu- 17. Lacatena, R. M., Busiello, V., Girolamo, A. D., and Girolamo, M. D. DNA polymerase activities in Friend cells during the differentiation process. Cell cleoids that had been maximally relaxed with ethidium (minimum Differ., 10:109-116,1981. of the ethidium titration curve). 18. Lapeyre,J. N., and Bekhor, I. Effects of 5-bromo-2'-deoxyuridine and dimeth- ylsulfoxide on properties and structure of chromatin. J. Mol. Biol., 89: 137- The increased sedimentation in the absence of ethidium was 162, 1974. not due to a detectable increase in the protein content of 19. Lown, J. W., Chen, H. H., Plambeck, J. A., and Acton, E. M. Diminished nucleoids (28) (data not shown). It could be due to a decrease in SuperoxideaniÃ³ngenerationby reduced 5-iminodaunorubicinrelative to dau- norubicin and the relationship to cardiotoxicity of the anthracyclineantitumor DNA linking number and/or a decrease in domain (loop) size. agents. Biochem. Pharmacol.,28: 2563-2568,1979. The increase in ethidium concentration for complete relaxation 20. Luchnik, A. N., and Glaser, V. M. Decrease in the number of DNA topological turns during Friend erythroleukemia differentiation. Mol. Gen. Genet. 778- suggests a decrease in DNA linking number, whereas the in 459-463, 1980. crease in minimum sedimentation in ethidium titration suggests 21. Mattem, M. R., and Painter, R. B. Dependenceof mammalianDNA replication a decrease in domain size. Hence, both factors may contribute on DNA supercoiling. I. Effects of ethidium bromide on DNA synthesis in permeable Chinese hamster ovary cells. Biochim. Biophys. Acta, 563: 293- to the net effect of thiourea on chromatin structure. 305, 1979. The prominent effects of Me2SO were to decrease the sedi 22. Millar, B. C., Sapora, O., Fielden, E. M., and Laverock, P. S. The application of rapid lysis techniques in radiobidogy. IV. The effect of glycerol and DMSO mentation minimum in the ethidium titration curve, and to mark on Chinese hamster cells survival and DNA single-strand break production. edly increase the ethidium concentration required for complete RadiÃ¢t.Res.,86: 506-514,1981. relaxation (20) and "rewinding" of the nucleoids. This suggests 23. Mimnaugh, E. G., Trush, M. A., Ginsburg, E., and Gram, T. E. Differential effects of anthracycline drugs on rat heart and liver microsomal reduced that the most prominent effect of Me2SO is to increase the nicotinamide adenine dinucleotide phosphate-dependent lipid peroxidation. domain (loop) size. This increase in domain size could be due to Cancer Res., 42: 3574-3582, 1982. a reduction of the number of attachment points of nucleoid DNA 24. Pommier, Y., Kerrigan, D., Schwartz, R., and Zwelling, L. A. The formation and resealingof intercalator-inducedDNA strand breaks in isolated L1210 cell with proteins, which is consistent with previous work showing nuclei. Biochem. Biophys. Res. Commun., 707: 576-583, 1982. that Me2SO destabilizes DNA-chromatin protein interactions (18, 25. Ross, W. E., and Bradley, M. O. DNA double-strand breaks in mammalian cells after exposure to DNA intercalatingagents. Biochim. Biophys. Acta, 654: 31). 129-134,1981. Thus, thiourea and Me2SO produce prominent but different 26. Ross, W. E., Glaubiger,D. L., and Kohn, K. W. Protein-associatedDNA breaks effects on DNA winding and loop topology. These changes may in cells treated with Adriamycinor ellipticine.Biochim. Biophys. Acta, 579: 23- 30, 1978. reflect different alterations in chromatin structure which might 27. Ross, W. E., Glaubiger, D. L., and Kohn, K. W. Qualitative and quantitative influence the efficiency with which the activity of an enzyme aspects of intercalator-induced DNA strand breaks. Biochim. Biophys. Acta, responsible for intercalator-induced DNA breaks is expressed. 562:41-50,1979. 28. Roti Roti, J. L., and Painter.R. B. Effects of hyperthermiaon the sedimentation of nucleoids from HeLa cells in sucrose gradients. RadiÃ¢t.Res.,89:166-175, REFERENCES 1982. 29. Scher, B. M., Scher, W., Robinson, A., and Waxman, S. DNA ligase and 1. Antoku. S. Chemical protection against radiation-inducedDNA single-strand DNase activities in mouse erythroleukemic cells during dimethyl sulfoxide- break in cultured mammaliancells. RadiÃ¢t.Res.,65:130-138,1976. induced differentiation. Cancer Res., 42:1300-1306,1982. 2. Bachur, N. R . Gee, M. V., and Friedman, R. D. Nuclear catalyzed antibiotic 30. Scher, W., and Friend,C. Breakage of DNA and alterations in folded genomes free radicalformation. Cancer Res., 42:1078-1081,1982. by inducers of differentiation in Frienderythroleukemic cells. Cancer Res., 38: 3. Bachur, N. R., Gordon, S. L, and Gee, M. V. Anthracyclineantibiotic augmen 841-849, 1978. tation of microsomal electron transport and free radical formation. Mol. Phar- 31. Tanaka, M., Levy, J., Terada, M., Breslow, R., Rifkind, R. A., and Maries,P. macol., 13: 901-910, 1977. A. Induction of erythroid differentiation in murine virus erythroleukemia cells 4. Berlin, V., and Haseltine,W. A. Reductionof Adriamycinto a semiquinonefree by highly polar compounds. Proc. Nati. Acad. Sei. U. S. A., 72: 1003-1006, radical by NADPH cytochrome P450 reductase produces DNA cleavage in a 1975. reaction mediated by molecular oxygen. J. Biol. Chem., 256: 4747-4756, 32. Terada, M., Nudel, U., Fibach, E., Rifkind, R. A., and Marks, P. A. Changes in 1981. DNA associatedwith induction of erythroid differentiationby dimethyl sulfoxide 5. Bradley, M. O., and Erickson, L. C. Comparison of the effects of hydrogen in murine erythroleukemiacells. Cancer Res., 38: 835-840, 1978. peroxide and x-ray irradiation on toxicity, mutation, and DNA damage/repair 33. Zwelling, L. A., Kerrigan, D., and Michaels, S. Cytotoxicity and DNA strand in mammaliancells (V79). Biochim. Biophys. Acta, 654: 135-141,1981. breaks by 5-iminodaunorubicin in mouse leukemia L1210 cells: comparison with Adriamycin and 4'-(9-acridinylammo)methanesulfon-m-anisidide.Cancer 6. Burger, R. M., Peisach,J., and Horwitz, S. B. Activated Weomycin.A transient complex of drug, iron and oxygen that degrades DNA. J. Biol. Chem., 256: Res., 42:2687-2691,1982. 11636-11644,1981. 34. Zwelling, L. A., Kerrigan, D., Michaels, S., and Kohn, K. W. Cooperative 7. Chu, M. Y., and Fisher, G. A. Incorporation of cytosine-3Harabinosideand its sequestration of m-AMSA in L1210 cells. Biochem. Pharmacol., 37: 3269- effects on murine leukemia cells (L5178Y). Biochem. Pharmacol., 17: 753- 3277, 1982. 767, 1968. 35. Zwelling, L. A., Kerrigan, D., Pommier, Y., Michaels, S., Steren, A., and Kohn, 8. Cook, P. R., and Brazell, I. A. Supercoils in human DNA. J. Cell Sci., Ã•9:261- K. W. Formation and resealing of intercalator-induced DNA strand breaks in 279, 1975. permeabilized L1210 cells without the stimulated synthesis of poly(ADP- 9. Cook, P. R., and Brazell, I. A. Conformational constraints in nuclear DNA. J. ribose).J. Biol. Chem., 25: 8957-8963,1982. Cell Sci., 22:287-302, 1976. 36. Zwelling, L. A., Michaels, S., Erickson, L. C., Ungerieider,R. S., Nichols, M., 10. Dingman, C. W., and Kakunaga, T. DNA strand breaking and rejoining in and Kohn, K. W. Protein-associatedDNA strand breaks in L1210 cells treated response to ultraviolet light in normal human and xeroderma pigmentosum with the DNA intercalating agents, 4'-(9-acridinylamino)-methanesulfon-m-an- cells. Int. J. RadiÃ¢t.Biol.Relat. Stud. Phys. Chem. Med., 30: 55-66,1976. isidide (m-AMSA)and Adriamycin. Biochemistry, 20: 6553-6563,1981. 11. Erickson, L. C., Bradley, M. 0., and Kohn, K. W. Differential inhibition of the 37. Zwelling, L. A., Michaels, S., Kerrigan, D., Pommier, Y., and Kohn, K. W. rejoining of X-ray-induced DNA strand breaks in normal and transformed Protein-associated deoxyribonucleic acid strand breaks produced in mouse human fibroblasts treated with 1,3-bis(2-chloroethyl)-1-nitrosoureain vitro. leukemia L1210 cells by ellipticine and 2-methyl-9-hydroxyellipticinium. Cancer Res., 38. 672-677, 1978. Biochem. Pharmacol.,37: 3261-3267,1982.

5724 CANCER RESEARCH VOL. 43

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1983 American Association for Cancer Research. Effects of Dimethyl Sulfoxide and Thiourea upon Intercalator-induced DNA Single-Strand Breaks in Mouse Leukemia (L1210) Cells

Yves Pommier, Leonard A. Zwelling, Michael R. Mattern, et al.

Cancer Res 1983;43:5718-5724.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/43/12_Part_1/5718

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

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/43/12_Part_1/5718. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.