Development of thermotolerance in mouse fibroblast LM cells with modified membranes and after procaine treatment.

Mouse fibroblast LM cells have been modified with respect to the content of polyunsaturated fatty acyl (PUFA) chains of the membrane phospholipids. The membranes of the modified cells were enriched in PUFA chains and were more fluid as compared to the normal cells, as judged by fluorescence polarization measurements. The thermosensitivity of the PUFA-substituted cells was enhanced. Thermotolerance in the PUFA-substituted fibroblasts could be induced to the same extent as in the nonsubstituted cells. The thermosensitivity in both the PUFA and the nonsubstituted fibroblasts could be enhanced by the treatment of procaine. Procaine could inhibit the triggering as well as the induction of thermotolerance. It is supposed that the mechanism of heat sensitization by procaine is different from the mechanism of preventing thermotolerance induction. The clinical implications of this finding are discussed.

Reduced DNA break formation and cytotoxicity of the topoisomerase II drug 4'-(9'-acridinylamino)methanesulfon-m-anisidide when combined with hyperthermia in human and rodent cell lines.

The interaction between hyperthermia and the anticancer drug 4'-(9'-acridinylamino)methanesulfon-m-anisidide (mAMSA) was studied in the human HeLa S3 and the rodent Ehrlich ascites tumor cell line. For both cell lines it was found that hyperthermia preceding the drug treatment reduced the extent of mAMSA induced DNA breakage as well as mAMSA cytotoxicity. Formation and resealing of mAMSA induced DNA break formation was found to be related to cytotoxicity. Hyperthermic protection for the action of mAMSA was found not to be a result of changed permeability for the drug. The data also do not support the possibility that heat has caused inactivation of the putative target enzyme of mAMSA, topoisomerase II. It is suggested that the hyperthermic protection for the mAMSA drug action is due to a hyperthermic alteration of the chromatin organization, especially at topoisomerase II target sequences that are found to be enriched in the nuclear matrix (P.N. Cockerill and W.T. Garrard. Cell, 44: 273-282, 1986). We show here that heat has caused an alteration of protein binding to the nucleus that seems related to the hyperthermic inhibition of mAMSA induced DNA break induction. It is concluded that preheating cells before treatment with mAMSA should not be used, at least not in this sequence, in cancer therapy.

Goralatide (AcSDKP) selectively protects murine hematopoietic progenitors and stem cells against hyperthermic damage.

Hyperthermic purging procedures may be improved by methods that selectively inhibit the proliferative activity of normal hematopoietic progenitors and stem cells, since active proliferation of these subsets is accompanied by increased heat sensitivity. For this reason, bone marrow cells from CBA/H mice were incubated with Goralatide (tetrapeptide Acetyl-N-Ser-Asp-Lys-Pro), a well-known inhibitor of normal hematopoietic progenitor cells to enter the S phase of the cell cycle. Subsequently, the cell suspensions were heat-treated at 43 degrees C for up to 90 minutes. After an exposure of 8 hours to 10(-9) M Goralatide, the number of CFU-GM cells in S phase decreased from 30 to 10%, resulting in an almost 10-fold increase in survival after 90 minutes at 43 degrees C. No effect on the primitive subsets could be detected because of their quiescent cell cycle state in normal bone marrow. To investigate the potential vulnerability of these subsets for Goralatide, bone marrow cells from 5-fluorouracil (5-FU)-pretreated mice were used. 5-FU induced increase in proliferative activity of the CFU-S-12 (day-12 colony-forming units-spleen), and the stem cell with marrow repopulating ability could be abolished by an incubation period with 10(-9) M Goralatide for 16 and 24 hours, respectively. Hence, this decrease in proliferative activity confers a decrease in hyperthermic sensitivity for the primitive hematopoietic subsets. The cytotoxic effect of the incubation on the absolute number of the hematopoietic progenitors and stem cells was <10%. Goralatide treatment (10(-8), 10(-9), and 10(-10) M) up to 24 hours had no effect on the growth kinetics and cell cycle distribution and consequently on the hyperthermic sensitivity of L1210 cells. Based on these results, it can be concluded that Goralatide will have a positive effect on the survival of hematopoietic progenitors and stem cells after hyperthermia and may lead to a gain in the therapeutic window of this purging modality.

Studies on the hyperthermic sensitivity of the murine hematopoietic stem cell compartment. I. Heat effects on clonogenic stem cells and progenitors.

The heat sensitivity at 42 degrees, 43 degrees and 44 degrees C of various hematopoietic subsets in murine bone marrow (MRA, CFU-S-12, CFU-S-8, CFU-GM, BFU-E and CFU-E) was investigated in order to determine whether there is a relationship between heat sensitivity and the position of cells within the stem cell hierarchy. The results show that the primitive stem cell with marrow repopulating ability (MRA) is extremely heat-resistant compared with the most differentiated hematopoietic progenitor (colony-forming unit-erythroid [CFU-E]). The proliferative activity of the hematopoietic subsets was determined from the number of cells killed by hydroxyurea (HU). It is demonstrated that there is a progressive increase in the proportion of hematopoietic subset cells in S-phase with maturation. The various heat sensitivities among the different hematopoietic subsets appear to be related to their proliferative activity. This relationship may have relevance to the clinical application of hyperthermia as a purging modality.

Studies on the hyperthermic sensitivity of the murine hematopoietic stem cell compartment. II. Heat effect on donor stem cells with long-term repopulating ability.

Variations in hyperthermic sensitivity among different hematopoietic progenitor and stem cell populations of the bone marrow have been previously described for clonogenic subsets responsible for short-term hematopoiesis. However, less is known of the heat sensitivity of more primitive stem cells capable of long-term repopulation in irradiated recipients. In the present study, control and heat-treated (60 minutes at 43 degrees C) donor bone marrow cells from congenic B6-Gpi-1a mice were transplanted at different cell doses (10(4), 10(5), 10(6), and 10(7) nucleated cells) in pre-irradiated (6 Gy) B6-Gpi-1b mice. The development and levels of donor marrow engraftment were determined from blood Gpi phenotyping, and the bone marrow dose required for equivalent long-term engraftment at 20 weeks provided an estimate of the surviving fraction corresponding to primitive stem cells of long-term repopulating ability (LTRA). Comparison with previous bone marrow cell survival values demonstrates that LTRA cells are less sensitive to hyperthermic treatment than other hematopoietic subsets, confirming a relationship between the heat sensitivity and the hierarchical structure of the hematopoietic stem cell compartment.

Intracellular free calcium concentrations in cell suspensions during hyperthermia.

The aim of this study was to explore the possibility that heat-induced alterations in calcium homeostasis are the cause of hyperthermic cell killing. Therefore, the intracellular free calcium concentration ([Ca2+]i) was determined spectrofluorometrically, using the fluorescent calcium probe fura-2/acetoxy methylester (AM), at both physiological and hyperthermic temperatures in cell suspensions from six different tumor cell lines. For all cell lines fura-2 leakage appears to contribute to a change in the fluorescence signal and hence leads to a false indication of an increase in [Ca2+]i, especially at the hyperthermic temperature. Two methods were introduced that circumvent this problem and results in true values of [Ca2+]i. Also, measurements of [Ca2+]i in single cells using a fluorescent microscopical technique (not affected by dye leakage) were used for comparison. All three approaches show that a hyperthermic treatment that kills > 90% of the cells does not lead to changes in the [Ca2+]i in most cell lines. Therefore, heat-induced alterations of calcium homeostasis cannot be considered the general cause for hyperthermic cell killing.

Characterization of oxygen-tolerant Chinese hamster ovary cells. II. Energy metabolism and antioxidant status.

Further characteristics of an oxygen-tolerant variant of Chinese hamster ovary cells (CHO-99) capable of stable proliferation at 99% O2/1% CO2, and O2 level that is lethal to the parental line (CHO-20), are described. Previous work has revealed that CHO-99 cells have 2- to 4-fold increased activities of superoxide dismutases, catalase and glutathione peroxidase, and substantially increased relative volumes of mitochondria and peroxisomes. To document possible additional mechanisms of O2 tolerance we compared CHO-20 cells growing at 20% O2 (normoxia) and CHO-99 cells at 99% O2 (normobaric hyperoxia). We show the following: (1) the estimated total (oxidative and glycolytic) ATP production in CHO-99 cells was 36% decreased. ATP production through oxidative phosphorylation was 52% lower in CHO-99 cells, while the relative contribution from glycolysis was increased from 6% to 30%. The ATP content was 29% lower in CHO-99 cells, the adenylate energy charge being also significantly decreased, indicating that energy production through oxidative phosphorylation is compromised in CHO-99 cells. Cyanide-resistant respiration was 4-fold higher in CHO-99 cells, probably reflecting, at least partly, the increased peroxisomal activity in these cells. (2) The level of reduced glutathione was several fold increased in CHO-99 cells, oxidized glutathione being unaltered; (NADPH + NADP+) levels were elevated 2.7-fold, while the ratio of NADPH to NADP+ was increased almost two-fold. These changes were associated with a 50% increased metabolism of glucose through the hexose monophosphate pathway. (3) No evidence was obtained for an increased steady-state level of endogenous lipid peroxidation in CHO-99 cells, in spite of a 50% increased content of polyunsaturated fatty acids in the phospholipid fraction.

Tissue tolerance of normal and surgically manipulated canine liver to intraoperative radiation therapy (IORT).

PURPOSE: The purpose of the study is to obtain dose guidelines for the delivery of intraoperative radiotherapy to the liver of patients with colorectal liver metastases. Following partial resection of the liver, a single high dose of 10, 20, 25, and 30 Gy intraoperative radiotherapy was applied to both the resection plane as well as a nonsurgically manipulated part of the liver of 25 beagles. The temporal sequence of histological and ultrastructural changes of these irradiated parts of the liver tissue was investigated. METHODS AND MATERIALS: The feasibility of delivering single large dose of intraoperative electron beam radiotherapy to the normal and partially hepatectomized liver was experimentally investigated in a canine study. RESULTS: There were no postoperative complications, no morbidity or mortality with a minimal follow-up of 1 year. Autopsy performed 3 months following irradiation showed only mild histopathological changes. One year following intraoperative radiotherapy more distinct histopathological changes consisting of capsular thickening, diffuse parenchymal fibrosis and subcapsular hepatocellular atrophy were found. The liver function remained intact. CONCLUSION: This study demonstrated that intraoperative radiotherapy to part of the liver in the canine model can be safely applied and doses up to 30 Gy are well tolerated.

Feasibility of measuring radiation-induced DNA double strand breaks and their repair by pulsed field gel electrophoresis in freshly isolated cells from the mouse RIF-1 tumor.

PURPOSE: To examine the technical feasibility of pulsed field gel electrophoresis (PFGE) as a predictive assay for the radioresponsiveness of tumors. Induction and repair of DNA double strand breaks (DSBs) in a freshly prepared cell suspension from a RIF-1 tumor (irradiated ex vivo) was compared with DSB induction and repair in exponentially growing RIF-1 cells in culture (irradiated in vitro). METHODS AND MATERIALS: A murine RIF-1 tumor grown in vivo was digested, and cells were exposed to x-rays (ex vivo) at doses of 1 to 75 Gy. DNA damage was measured using CHEF (clamped homogeneous electric fields) electrophoresis. Repair kinetics were studied at 37 degrees C for 4 h after irradiation. Radiosensitivity was determined by clonogenic assay, and cell cycle distributions by flow cytometry. For comparison, a trypsinized suspension of exponentially growing RIF-1 cells in vitro was run parallel with each ex vivo experiment. RESULTS: Induction of DSBs, expressed as % DNA extracted from the plug, was similar in the in vitro and ex vivo irradiated cells. Compared to repair rates in vitro cultured RIF-1 cells, repair kinetics in a freshly prepared cell suspension from the tumor were decreased, unrelated to differences in radiosensitivity. Differences in repair could not be explained by endogenous DNA degradation, nor by influences of enzymes used for digestion of the tumor. A lower plating efficiency and differences in ploidy (as revealed by flow cytometry) were the only reproducible differences between in vivo and in vitro grown cells that may explain the differences in repair kinetics. CONCLUSIONS: The current results do not support the idea that PFGE is a technique robust enough to be a predictive assay for the radiosensitivity of tumor cells.

Preservation of the rat parotid gland function after radiation by prophylactic pilocarpine treatment: radiation dose dependency and compensatory mechanisms.

PURPOSE: To study the ability of a prophylactic pilocarpine administration to preserve the rat parotid gland function after unilateral irradiation with graded doses of X-rays. METHODS: The right parotid gland of male albino Wistar rats was irradiated with single doses of X-rays (10-30 Gy, at 1.5 Gy min(-1)). Pilocarpine (4 mg/kg) was administered intraperitoneally, 1 hour prior to irradiation. Saliva samples of both left and right parotid gland were collected by means of miniaturized Lashley cups 4 days before and 3, 7, 10, and 30 days after irradiation. The parotid salivary flow rate (microl/min) was used as a parameter for the assessment of parotid gland function. RESULTS: Our data confirm that a single prophylactic treatment of pilocarpine can attenuate radiation-induced loss of gland function. Surprisingly, the effect of pilocarpine was not restricted to the irradiated gland only. Pilocarpine also enhanced the flow rate in the contralateral, nonirradiated gland. The latter effect was found for all doses above 10 Gy and became apparent around 7 days after the radiation treatment. The effectiveness of pilocarpine to attenuate function loss in the irradiated gland decreased with increasing dose and was lost after single doses of 30 Gy. CONCLUSIONS: Our data provide direct evidence that increasing the compensatory potential of the nondamaged gland, at least in part, underlies the "radioprotective effect" of pilocarpine in case of unilateral radiation. The ability of pilocarpine to ameliorate the early radiation-induced impairment of the parotid gland function in the irradiated gland may therefore be dependent on the remaining number of functional cells, and thus on the volume of the gland that lies within the radiation portal.

Synthesis and cytotoxicity of novel lignans.

In this study the syntheses of 11 novel lignans are described. Their cytotoxicities are studied in GLC4, a human small cell lung carcinoma cell line, using the microculture tetrazolium (MTT) assay. Ten of these compounds were substituted with a menthyloxy group on the 5-position of the lactone. These compounds can easily be prepared in (novel) 'one-pot', three- or four-step syntheses. In addition, methods for controlling the stereogenic centers are described. Furthermore, five naturally occurring podophyllotoxin-related compounds were tested. The cytotoxicities of all lignan compounds, and of three non-lignan intermediates originating from the syntheses, were compared with the clinically applied anticancer agents etoposide, teniposide, and cisplatin. Most compounds showed moderate to high activities against GLC4, and two of the compounds containing a menthyloxy group showed activities comparable to the reference cytotoxic agents.

Radiation-induced inhibition of tumor growth as monitored by PET using L-[1-11C]tyrosine and fluorine-18-fluorodeoxyglucose.

The potential use of PET to monitor radiotherapeutic effects on tumors has been evaluated with L-[1-11C]tyrosine and 18FDG. Single x-ray doses of 10, 30, or 50 Gy have been applied to rhabdomyosarcoma tumors growing in the flank of rats. Dose-dependent reductions of tracer uptake were registered by PET 4 and 12 days after treatment. These later effects on tracer uptake appeared to correlate with changes in tumor volume. Therefore, PET using L-[1-11C]tyrosine and 18FDG is suitable to monitor kinetics of tumor growth and tumor regression after radiotherapy. Direct effect on tracer uptake was not observed within 8 hr after irradiation. This indicates that, using PET, early predictions on the outcome of radiotherapy are not possible. When combining a radiation treatment with hyperthermia, radiation-induced inhibition of tumor growth was clearly enhanced. Tracer uptake remained at the pretreatment value, possibly due to invasion of host cells. From these experiments, it can be concluded that it is difficult to monitor a combined treatment of radiation and hyperthermia by PET.

PET measurements of hyperthermia-induced suppression of protein synthesis in tumors in relation to effects on tumor growth.

Hyperthermia-induced metabolic changes in tumor tissue have been monitored by PET. Uptake of L-[1-11C]tyrosine in rhabdomyosarcoma tissue of Wag/Rij rats was dose-dependently reduced after local hyperthermia treatment at 42, 45, or 47 degrees C. Tumor blood flow, as measured by PET with 13NH3, appeared to be unchanged. The L-[1-11C]tyrosine uptake data were compared to uptake data of L-[1-14C]tyrosine and with data on the incorporation of L-[1-14C]tyrosine into tumor proteins. After intravenous injection, the 14C data were obtained from dissected tumor tissue. Heat-induced inhibition of the incorporation of L-[1-14C]tyrosine into tumor proteins tallied with the L-[1-11C]tyrosine uptake data. Heat-induced inhibition of amino acid uptake in the tumor correlated well with regression of tumor growth. It is concluded that PET using L-[1-11C]tyrosine is eligible for monitoring the effect of hyperthermia on tumor growth.

Loco-regional differences in pulmonary function and density after partial rat lung irradiation.

PURPOSE: The purpose of this study was to explore regional differences in radiosensitivity of rat lung using lung function and computed tomography (CT) density as endpoints. METHODS: At first, CT scans were used to determine rat lung volumes. The data obtained enabled the design of accurate collimators to irradiate 50% of the total lung volume for the apex, base, left, right, mediastinal and lateral part of the lung. Male Wistar rats were irradiated with a single dose of 18 Gy of orthovoltage X-rays. Further rat thorax CT scans were made before and 4, 16, 26, and 52 weeks after irradiation to measure in vivo lung density changes indicative of lung damage. To evaluate overall lung function, breathing frequencies were measured biweekly starting 1 week before irradiation. RESULTS: Qualitative analysis of the CT scans showed clear density changes for all irradiated lung volumes, with the most prominent changes present in the mediastinal and left group at 26 weeks after radiation. Quantitative analysis using average density changes of whole lungs did not adequately describe the differences in radiation response between the treated groups. However, analysis of the density changes of the irradiated and non-irradiated regions of interest (ROI) more closely matched with the qualitative observations. Breathing frequencies (BF) were only increased after 50% left lung irradiation, indicating that the hypersensitivity of the mediastinal part as assessed by CT analysis, does not result in functional changes. CONCLUSIONS: For both BF and CT (best described by ROI analysis), differences in regional lung radiosensitivity were observed. The presentation of lung damage either as function loss or density changes do not necessarily coincide, meaning that for each endpoint the regional sensitivity may be different.

Plasma TGFbeta level in rats after hemithoracic irradiation.

Changes in TGF-beta plasma levels were observed 18 weeks after hemithoracic irradiation in rats. This coincides with an increase in the breathing frequency. being most pronounced between 22 and 28 weeks after irradiation. The correlation suggests a potential role of the circulating TGF-beta in the monitoring of localized radiation-induced lung injury.

Role of lipid peroxidation and DNA damage in paraquat toxicity and the interaction of paraquat with ionizing radiation.

Since the introduction of paraquat (PQ) as a herbicide in 1963, there have been many speculations concerning the critical lesion in PQ toxicity. Damage to membrane lipids might be an initial event leading to PQ-induced cell killing. The ability of PQ to induce lipid peroxidation was tested in liver homogenates of the mouse. Lipid peroxidation was indeed induced by PQ and shown to be dose dependent, starting to be significant at 2.5 mM. Subsequently, a possible correlation between lipid peroxidation and PQ-induced cell death was investigated in mouse fibroblasts (LM) and Ehrlich ascites tumour (EAT) cells using a clonogenic assay. It was found that in order to be cytotoxic PQ needs enzymatic activation (incubation at 37 degrees). In both cell lines, PQ-induced cell killing appeared to be dose dependent, starting at a dose of 0.5 mM. Supplementation of LM cells with the antioxidant vitamin E had no effect on PQ-induced cell killing and modification of the membranes of LM cells by incorporation of the polyunsaturated fatty acid 20:4 (arachidonic acid) did not sensitize the cells to PQ toxicity. PQ had no effect on the glutathione (GSH) level in EAT cells and complete GSH depletion by DL-buthionine-(SR)-sulfoximine could not sensitize the cells to PQ toxicity. In LM cells PQ-induced cell killing was enhanced after complete GSH depletion by DEM. This sensitization might, however, be attributed to the binding of DEM to proteins. From these results it seems unlikely that lipid peroxidation is the primary cause for PQ-induced cell killing. Another critical target in PQ toxicity is DNA. This possibility was investigated in EAT cells. PQ was found to induce DNA damage (detected by the alkaline unwinding assay) in the same dose range that caused cell death. A good correlation was obtained for cell killing after PQ treatment and DNA damage measured 2 hr after 37 degrees post-incubation. A proposed possible interaction between PQ and X-rays was also investigated. In EAT cells, X-ray-induced cell death was significantly enhanced by pre-incubation with PQ at doses of 0.5 mM and above. At the level of 10% survival an enhancement factor of 1.6 could be observed by treatment with 1 mM PQ when cell killing by PQ is not taken into account. Induction as well as processing of radiation-induced DNA damage seems to be unaffected by pre-incubation with PQ. The mechanism of radiosensitization by PQ is yet unclear.

Induction of DNA damage in Ehrlich ascites tumour cells by exposure to eupatoriopicrin.

The sesquiterpene lactone eupatoriopicrin (EUP) from Eupatorium cannabinum L. has been shown to be cytotoxic in a glutathione (GSH)-dependent way. In order to assess possible DNA damage as a cause for cell death, the study reported was initiated. After 2 hr incubation of Ehrlich ascites tumour cells with EUP, the DNA damage, determined by the use of an alkaline DNA unwinding method, followed by hydroxylapatite column chromatography of degraded DNA, was observed at concentrations only slightly higher than those causing cell death in a clonogenic assay. The amount of EUP, requested to demonstrate DNA damage after a 24-hr post-incubation period lay within the concentration range that was effective in the clonogenic assay (1-10 micrograms/ml). Glutathione (GSH) depletion of the cells to about 99%, by use of buthionine sulphoximine (BSO), enhanced the extent of DNA damage. It is concluded that EUP-induced DNA damage may play a role in the observed cytotoxicity.

Cycloheximide- and puromycin-induced heat resistance: different effects on cytoplasmic and nuclear luciferases.

Inhibition of translation can result in cytoprotection against heat shock. The mechanism of this protection has remained elusive so far. Here, the thermoprotective effects of the translation inhibitor cycloheximide (CHX) and puromycin were investigated, using as reporter firefly luciferase localized either in the nucleus or in the cytoplasm. A short preincubation of O23 cells with either translation inhibitor was found to attenuate the heat inactivation of a luciferase directed into the cytoplasm, whereas the heat sensitivity of a nuclear-targeted luciferase remained unaffected. After a long-term CHX pretreatment, both luciferases were more heat resistant. Both the cytoplasmic and the nuclear luciferase are protected against heat-induced inactivation in thermotolerant cells and in cells overexpressing heat shock protein (Hsp)70. CHX incubations further attenuated cytoplasmic luciferase inactivation in thermotolerant and in Hsp70 overexpressing cells, even when Hsp70-mediated protection was saturated. It is concluded that protection by translation inhibition is unlikely due to an increase in the pool of free Hsps normally engaged in translation and released from the nascent polypeptide chains on the ribosomes. Rather, a decrease in nascent chains and thermolabile polypeptides may account for the heat resistance promoted by inhibitors of translation.

Enhanced selectivity of hyperthermic purging of human progenitor cells using Goralatide, an inhibitor of cell cycle progression.

Recurrence of leukemia is a major problem after autologous stem cell transplantation. One potential means of reducing this risk is to purge the autologous transplant in vitro by hyperthermia. We have demonstrated that after a hyperthermic treatment of 120 min at 43 degrees C, the leukemic progenitor cells (CFU-AML) are decreased by 5-log but the normal hematopoietic committed progenitor cells (CFU-GM, BFU-E and CFU-E) are reduced by only 1-log. Moreover, the hyperthermic sensitivity coincides with the stem cell hierarchy, ie CFU-GM are less heat sensitive than BFU-E, while CFU-E are the most sensitive. The impact of pretreatment with the tetrapeptide AcSDKP (Goralatide) on the proliferative activity and heat sensitivity of the normal and leukemic progenitor cells was determined. An incubation of 21 h at 37 degrees C with 10(-9) M Goralatide reduces the number of normal hematopoietic progenitor cells in S-phase and concomitantly decreases their hyperthermic sensitivity. This effect implies that the proliferative activity is the major determinant for the detected differences in hyperthermic sensitivity of the subsets in the normal hematopoietic stem cell compartment. In contrast, the cell cycle progression of leukemic progenitor cells is not affected and hence these cells are not protected from hyperthermia-induced cell killing after preincubation with Goralatide. Thus, the treatment with Goralatide increases the therapeutic window of hyperthermia and increases the potential value of this physical purging technique.

Effect of inhibitors of poly(ADP-ribose) polymerase on the heat response of HeLa S3 cells.

The purpose of this study was to investigate a possible involvement of poly(ADP-ribosyl)ation reactions in hyperthermic cell killing and hyperthermic DNA strand-break induction and repair in HeLa S3 cells. The inhibitors of poly(ADP-ribose) polymerase, 3-aminobenzamide (3AB) and 4-aminobenzamide (4AB), were used as tools in this study. Both inhibitors could sensitize the cells for hyperthermic cell killing equally well, although 3AB is known to be a more effective enzyme inhibitor. The heat sensitization at the level of cell killing could be reversed when the compounds were still present during a 4-h postincubation at 37 degrees C. More heat-induced DNA strand breaks were formed in the presence of 3AB and 4AB. Repair of strand breaks was inhibited during the postincubation at 37 degrees C. Thus the effect of 3AB and 4AB on DNA strand-break repair was different from the cited effect on cell survival. It is concluded that the sensitizing effect of 3AB and 4AB on hyperthermic cell killing is not caused by inhibition of poly(ADP-ribose) polymerase and is also not related to repair of DNA strand breaks.