Mild hyperthermia inhibits homologous recombination, induces BRCA2 degradation, and sensitizes cancer cells to poly (ADP-ribose) polymerase-1 inhibition.

Defective homologous recombination (HR) DNA repair imposed by BRCA1 or BRCA2 deficiency sensitizes cells to poly (ADP-ribose) polymerase (PARP)-1 inhibition and is currently exploited in clinical treatment of HR-deficient tumors. Here we show that mild hyperthermia (41-42.5 °C) induces degradation of BRCA2 and inhibits HR. We demonstrate that hyperthermia can be used to sensitize innately HR-proficient tumor cells to PARP-1 inhibitors and that this effect can be enhanced by heat shock protein inhibition. Our results, obtained from cell lines and in vivo tumor models, enable the design of unique therapeutic strategies involving localized on-demand induction of HR deficiency, an approach that we term induced synthetic lethality.

Cyclopentenyl cytosine has biological and anti-tumour activity, but does not enhance the efficacy of gemcitabine and radiation in two animal tumour models.

Cyclopentenyl cytosine (CPEC), targetting the de novo biosynthesis of cytidine triphosphate (CTP), increases the cytotoxicity of gemcitabine (2',2'-difluoro-2'-deoxycytidine, dFdC) alone and in combination with irradiation in several human tumour cells in vitro. We investigated whether CPEC enhances the therapeutic ratio of gemcitabine and irradiation in human pancreatic BxPC-3 xenografts and in rat syngeneic L44 lung tumours. These models were selected because gemcitabine and radiation are used to treat both pancreatic and lung cancer patients and both models differ in growth capacity and in gemcitabine-induced radiosensitisation. A profound dose-dependent CTP-depletion was observed after a single injection of CPEC in both tumour tissue and in normal jejunum. In both models, CPEC alone induced a slight but significant tumour growth delay. The combination of CPEC with gemcitabine, at time intervals that showed CTP-depletion after CPEC, enhanced neither tumour growth delay nor toxicity as compared to gemcitabine alone. In addition, no beneficial effect of CPEC was observed in combination with gemcitabine and radiation. These results suggest that CPEC and gemcitabine alone as well as in combination with radiation target a similar cell population in both tumour models. In conclusion, future clinical development of CPEC as a modulator of gemcitabine combined with radiation is unlikely.

G0 cell cycle arrest alone is insufficient for enabling the repair of ionizing radiation-induced potentially lethal damage.

The repair of ionizing radiation-induced potentially lethal damage (PLD) is suggested to be important for the clinical response to radiotherapy. PLD repair is usually studied in quiescent cultures prepared by growing cells to confluence with an accumulation of cells in G(0) phase of the cell cycle, but the biological pathways enabling PLD repair are still unknown. In this study, we examined whether the controlled expression of two different inducers of G(0) cell cycle arrest, the human tumor suppressor gene growth arrest specific 1 (GAS1) in murine fibroblasts and the forkhead transcription factor FOXO3a in human colon carcinoma cells, is sufficient to enable PLD repair. We found that GAS1 and FOXO3a induced a cell cycle arrest in G(0) phase with a concomitant reduction of proliferation of log-phase cells. In both cell systems, this cell cycle arrest in G(0) phase did not enable PLD repair in log-phase cells. Significant PLD repair was found in all confluent cultures that showed similar cell cycle distributions, while GAS1 and FOXO3a in confluent cells did not influence PLD repair. No differences were found in cell cycle re-entry after replating cells with different capacities for PLD repair. Our data suggest that the induction of G(0) cell cycle arrest and the reduction of proliferation are not sufficient to enable PLD repair.

Chromosome fragments have the potential to predict hyperthermia-induced radio-sensitization in two different human tumor cell lines.

Cellular radiosensitivity, assessed by loss of clonogenicity, has been shown to correlate with the number of radiation-induced chromosomal aberrations. Also an increased radiosensitivity by hyperthermia has been shown to correlate with an increase in chromosomal aberrations. Therefore, determination of the number of chromosomal aberrations might be used as an assay to predict the radiosensitivity of tumors pre-treated with hyperthermia at clinically relevant temperatures. The use of premature chromosome condensation combined with fluorescent in situ hybridisation (PCC-FISH) has been shown to be clinically applicable. Therefore, the use of chromosomal aberrations as determined with PCC-FISH for the prediction of hyperthermia-induced radio-sensitization in human tumor cells was investigated. Confluent cultures of SW-1573 (human lung carcinoma) and RKO (human colorectal carcinoma) cells were treated with 1 h 41 degrees C or 43 degrees C hyperthermia prior to gamma-irradiation. Clonogenic cell survival and induction of chromosomal aberrations (unrejoined chromosomal fragments and translocations), by PCC-FISH, were studied at 24 h after treatment. Pre-treatment with hyperthermia at 41 degrees C for 1 h enhanced the radiosensitivity of RKO cells but not of SW-1573 cells. Increasing the temperature to 43 degrees C for 1 h enhanced the radiosensitivity of SW-1573 cells. When radio-sensitization was observed, a significant increase in the number of unrejoined chromosomal fragments was found but the frequency of translocations was not increased. Hyperthermia-induced radio-sensitization is correlated with an increase in unrejoined chromosomal fragments. This suggests that determination of the number of chromosomal fragments after hyperthermia and radiation treatment might be used for the prediction of combined treatment response in cancer patients.

Negligible radiation protection of endothelial cells by vascular endothelial growth factor.

Glioblastoma multiforme (GBM) is a radioresistant tumor. Tumor neoangiogenesis is an important mechanism for tumor sustenance. Angiogenesis is primarily mediated by vascular endothelial growth factor (VEGF), and earlier studies have suggested that VEGF protects human umbilical vein endothelial cells (HUVECs) against high doses of radiation. We tried to extend these findings to other endothelial cell lines and clinically relevant irradiation doses. Therefore, four different endothelial cell lines (HUVEC-C, primary HUVEC-P, an immortalized HUVEC cell line: EC-RF24, and bovine retina endothelial cells: BREC) were cultured without or with recombinant human VEGF165 (rhVEGF165). Cells were irradiated with gamma-rays from a 137Cs-source. Radiosensitivity was determined by proliferation or clonogenic assay. Apoptosis was assayed by flow cytometric determination of the sub-G1 population or by counting nuclear fragmentation. We found that the biologically active rhVEGF165 was able to improve clonogenic survival of HUVEC-C after 2 and 5 Gy. However, rhVEGF165 could not significantly alter the radiosensitivity of all cell lines studied in proliferation assays. rhVEGF165 only slightly reduced apoptosis in HUVEC-C after 3 Gy. In conclusion, the radioprotective effect from rhVEGF165 was found on different endothelial cell lines after clinically relevant radiation doses was negligible. We therefore hypothesize that the high VEGF-levels found in GBM in vivo do not reduce the radiosensitivity of endothelial cells, which is thought to contribute to the strong radioresistance of the tumor vasculature.

Effect of 41 degrees C and 43 degrees C on cisplatin radiosensitization in two human carcinoma cell lines with different sensitivities for cisplatin.

The effect of trimodality treatment consisting of hyperthermia, cisplatin and radiation was investigated in two cell lines with different sensitivities to cisplatin. Hyperthermia treatment was performed for 1 h at 41 degrees C and 43 degrees C in order to compare the effects of the two temperatures. Clonogenic assays were performed with cisplatin-sensitive SiHa human cervical carcinoma and cisplatin-resistant SW-1573 human lung carcinoma cell lines. Cells were treated with various combinations of hyperthermia, cisplatin and radiation. Radiation was performed after 1 h of simultaneous hyperthermia and cisplatin treatment. Cisplatin exposure was for 1 h or continuous without refreshment of the cisplatin-containing medium. SiHa cells were more sensitive to cisplatin than SW-1573 cells. Hyperthermia at 41 degrees C decreased survival in SW-1573 cells but was not cytotoxic in SiHa cells. Hyperthermia at 43 degrees C decreased survival dramatically in both cell lines with SiHa being the most sensitive. The addition of hyperthermia at 41 degrees C and 43 degrees C to cisplatin treatment led to enhanced cell kill in both cell lines compared with cisplatin alone. Radiosensitization was observed after continuous but not after 1 h of cisplatin treatment. Hyperthermia at 43 degrees C increased radiosensitivity whereas hyperthermia at 41 degrees C did not. A combination of 41 degrees C hyperthermia with continuous cisplatin treatment had an additive effect on SW-1573 cells but enhanced cisplatin radiosensitivity of SiHa cells. In SW-1573 cells trimodality treatment using 43 degrees C hyperthermia enhances cisplatin radiosensitivity. We conclude that hyperthermia at 43 degrees C enhances cisplatin-induced radiosensitization in both cisplatin-sensitive and -resistant cell lines. Hyperthermia at 41 degrees C was also able to increase cisplatin-induced radiosensitivity but only in the cisplatin-sensitive SiHa cell line.

Residual late radiation damage in mouse stromal tissue assessed by the tumor bed effect.

Irradiation of murine subcutaneous stroma before implantation of tumor cells leads to retarded tumor growth. This effect is called Tumor Bed Effect (TBE) and can be used to assess the sensitivity of stromal tissue to radiation. We tested the ability of stromal tissue to recover from X-ray-induced damage as a function of the time interval between X-irradiation and implantation of tumor cells over a period of 195 days. We also assessed the effects of a second test treatment of X-irradiation before implantation to assess residual damage by the first radiation treatment. The tumor bed effect in C57Bl10xDBA2 mice observed after X-ray treatment and implantation of M8013 cells (from a transplantable mouse mammary carcinoma) declines with the time that elapses between X-rays and implantation. Implantation of tumor cells 195 days after initial irradiation of 10 or 20 Gy resulted in a considerably smaller TBE. The half-time of the decay is estimated as about 50 days. The extent of the recovery was then tested in two-fraction experiments, with radiation fractions separated by intervals of 30 or 180 days. In the experiment with re-irradiation at an interval of 30 days after the first radiation dose of 20 Gy hardly any recovery was observed, whereas at an interval of 180 days a considerable recovery was observed. We presume that the recovery in TBE that was observed a long time after the irradiation results from a proliferative stimulus to endothelial cells which takes place during the post-irradiation period. The proliferative response leads to cell death of the X-ray damaged endothelial cells and thereafter these are replaced by healthy cells.

Histone deacetylase inhibitor BL1521 induces a G1-phase arrest in neuroblastoma cells through altered expression of cell cycle proteins.

Histone deacetylase inhibitors (HDACi) have been discovered as potential drugs for cancer treatment. The effect of BL1521, a novel HDACi, on the cell cycle distribution and the induction of apoptosis was investigated in a panel of MYCN single copy and MYCN amplified neuroblastoma cell lines. BL1521 arrested neuroblastoma cells in the G1 phase and induced up to 30% apoptosis. Downregulation of CDK4, upregulation of p21(WAF1/CIP1) and an increase of hypophosphorylated retinoblastoma protein were observed, indicating a possible mechanism for the cell-cycle arrest. BL1521 also induced downregulation of p27, which may underlie the observed induction of apoptosis.

Mismatch repair proficiency is not required for radioenhancement by gemcitabine.

PURPOSE: Mismatch repair (MMR) proficiency has been reported to either increase or decrease radioenhancement by 24-h incubations with gemcitabine. This study aimed to establish the importance of MMR for radioenhancement by gemcitabine after short-exposure, high-dose treatment and long-exposure, low-dose treatment. METHODS AND MATERIALS: Survival of MMR-deficient HCT116 and MMR-proficient HCT116 + 3 cells was analyzed by clonogenic assays. Mild, equitoxic gemcitabine treatments (4 h, 0.1 microM vs. 24 h, 6 nM) were combined with gamma-irradiation to determine the radioenhancement with or without recovery. Gemcitabine metabolism and cell-cycle effects were evaluated by high-performance liquid chromatography analysis and bivariate flow cytometry. RESULTS: Radioenhancement after 4 h of 0.1 microM of gemcitabine was similar in both cell lines, but the radioenhancement after 24 h of 6 nM of gemcitabine was reduced in MMR-proficient cells. No significant differences between both cell lines were observed in the gemcitabine metabolism or cell-cycle effects after these treatments. Gemcitabine radioenhancement after recovery was also lower in MMR-proficient cells than in MMR-deficient cells. CONCLUSION: Mismatch repair proficiency decreases radioenhancement by long incubations of gemcitabine but does not affect radioenhancement by short exposures to a clinically relevant gemcitabine dose. Our data suggest that MMR contributes to the recovery from gemcitabine treatment.

TR3 orphan receptor is expressed in vascular endothelial cells and mediates cell cycle arrest.

OBJECTIVE: Endothelial cells play a pivotal role in vascular homeostasis. In this study, we investigated the function of the nerve growth factor-induced protein-B (NGFI-B) subfamily of nuclear receptors comprising the TR3 orphan receptor (TR3), mitogen-induced nuclear orphan receptor (MINOR), and nuclear orphan receptor of T cells (NOT) in endothelial cells. METHODS AND RESULTS: The mRNA expression of TR3, MINOR, and NOT in atherosclerotic lesions was assessed in human vascular specimens. Each of these factors is expressed in smooth muscle cells, as described before, and in subsets of endothelial cells, implicating that they might affect endothelial cell function. Adenoviral overexpression of TR3 in cultured endothelial cells resulted in decreased [3H]thymidine incorporation, whereas a dominant-negative TR3 variant that inhibits the activity of endogenous TR3-like factors enhanced DNA synthesis. TR3 interfered with progression of the cell cycle by upregulating p27Kip1 and downregulating cyclin A, whereas expression levels of a number of other cell cycle-associated proteins remained unchanged. CONCLUSIONS: These findings demonstrate that TR3 is a modulator of endothelial cell proliferation and arrests endothelial cells in the G1 phase of the cell cycle by influencing cell cycle protein levels. We hypothesize involvement of TR3 in the maintenance of integrity of the vascular endothelium.

Sensitivity to ionizing radiation and chemotherapeutic agents in gemcitabine-resistant human tumor cell lines.

PURPOSE: To determine cross-resistance to anti-tumor treatments in 2',2'difluorodeoxycytidine (dFdC, gemcitabine)-resistant human tumor cells. METHODS AND MATERIALS: Human lung carcinoma cells SW-1573 (SWp) were made resistant to dFdC (SWg). Sensitivity to cisplatin (cDDP), paclitaxel, 5-fluorouracil (5-FU), methotrexate (MTX), cytarabine (ara-C), and dFdC was measured by a proliferation assay. Radiosensitivity and radioenhancement by dFdC of this cell panel and the human ovarian carcinoma cell line A2780 and its dFdC-resistant variant AG6000 were determined by clonogenic assay. Bivariate flowcytometry was performed to study cell cycle changes. RESULTS: In the SWg, a complete deoxycytidine kinase (dCK) deficiency was found on mRNA and protein level. This was accompanied by a 10-fold decrease in dCK activity which resulted in the >1000-fold resistance to dFdC. Sensitivity to other anti-tumor drugs was not altered, except for ara-C (>100-fold resistance). Radiosensitivity was not altered in the dFdC-resistant cell lines SWg and AG6000. High concentrations (50-100 microM dFdC) induced radioenhancement in the dFdC-resistant cell lines similar to the radioenhancement obtained at lower concentrations (10 nM dFdC) in the parental lines. An early S-phase arrest was found in all cell lines after dFdC treatment where radioenhancement was achieved. CONCLUSIONS: In the dFdC-resistant lung tumor cell line SWg, the deficiency in dCK is related to the resistance to dFdC and ara-C. No cross-resistance was observed to other anti-tumor drugs used for the treatment in lung cancer. Sensitivity to ionizing radiation was not altered in two different dFdC-resistant cell lines. Resistance to dFdC does not eliminate the ability of dFdC to sensitize cells to radiation.

Induction of the early response protein EGR-1 in human tumour cells after ionizing radiation is correlated with a reduction of repair of lethal lesions and an increase of repair of sublethal lesions.

The role of EGR-1 in potentially lethal damage repair (PLDR) was studied. Induction of the early response protein EGR-1 and survival after ionizing radiation of two human tumour cell lines after culturing for 48 h in serum-deprived medium was investigated. The glioblastoma cell line (Gli-6) and a lung carcinoma cell line (SW-1573) were selected as these cell lines differ considerably in the degree of PLD repair after radiation. In both cell lines induction of EGR-1 protein was observed between 30-120 min after treatment with 10 Gy in serum-deprived cultures. In cells growing in medium with normal serum no induction of EGR-1 was observed. No difference in EGR-1 expression levels between the two cell lines was detected. Linear-Quadratic analysis of the survival curves showed a much larger difference between the values of alpha after immediate and delayed plated cells of the cultures in normal serum as compared to cells cultured in serum-deprived medium. The cells cultured in serum-deprived medium showed much larger difference between the values of beta. This indicates that induction of EGR-1 is correlated with a reduction of repair of lethal lesions (PLDRalpha) and with an increase of repair of sublethal lesions (PLDRbeta).

Colour junctions as predictors of radiosensitivity: X-irradiation combined with gemcitabine in a lung carcinoma cell line.

PURPOSE: To determine whether measurement of colour junctions by fluorescent in situ hybridisation (FISH) can predict radiosensitisation of gemcitabine (2'-2'-difluorodeoxycytidine). METHODS: Human lung carcinoma cells (SW-1573) were irradiated with X-rays with or without incubation of 10 n M gemcitabine for 24 h. Cell survival was determined with clonogenic assay. Colour junctions were measured by whole chromosome FISH of chromosomes 2 and 18 and were scored according to the PAINT method. RESULTS: A clear radiosensitisation by gemcitabine was observed on cell survival. A significant decrease in the number of colour junctions was observed after gemcitabine treatment compared with radiation treatment alone. The correlation between colour junction induction and cell survival was high for both with and without gemcitabine, but the gemcitabine-sensitised curve did not coincide with the non-sensitised curve. CONCLUSIONS: Gemcitabine-induced radiosensitisation is not predicted by induction of colour junctions in cultured SW-1573 cells. This reduces the clinical applicability of this predictive assay for radiotherapy in combination with gemcitabine.

Enhanced levels of deoxycytidine kinase and thymidine kinase 1 and 2 after pulsed low dose rate irradiation as an adaptive response to radiation.

After pulsed low dose rate irradiation the activities of deoxycytidine kinase and thymidine kinase 1 and 2 were increased 1.5-2-fold 6 h after treatment. Twenty-four hours after treatment the activities of these enzymes had returned to control levels. We presume that the increase of enzyme activities is part of an adaptive response to irradiation and that this increase could be an explanation for the increased survival in the initial part of the SW-1573 cell survival curve. The observation that not only S-phase specific thymidine kinase 1 but also mitochondrial thymidine kinase 2 increases, implies that both these enzymes play a role in an adaptive response of cells to irradiation.

Effects of gemcitabine on cell survival and chromosome aberrations after pulsed low dose-rate irradiation.

The radiosensitizing potential of gemcitabine (2',2'-difluoro-2'-deoxycytidine) was studied in combination with pulsed low dose-rate irradiation. The experiments were carried out with a human lung carcinoma cell line SW1573. These were irradiated at pulsed low dose rate (p-LDR); the average dose rate was 1 Gy/h. In the experiments with gemcitabine, this drug was applied for 24 h at a concentration of 10 nM prior to irradiation. The response of the cells to treatment was tested by using the standard clonogenic assay. Next to the cell-killing effects, damage to chromosomes was also assayed by using by whole chromosome Fluorescent In Situ Hybridization (FISH). Damage in chromosomes 2 and 18 was visualized by whole chromosome FISH and scored according to the PAINT method. A clear enhancement of the effects of radiation on cell survival was observed by preincubation of the cells with gemcitabine. The enhancement factor obtained from the p-LDR data was 1.7, which is much lower than the enhancement factor of 2.9 at high-dose rate. We did not observe a consistent increase in color junctions concomitant with radiosensitization. In chromosome 2, a small increase, and in chromosome 18, a decrease, in the number of color junctions was observed after radiation combined with gemcitabine compared to irradiation alone. These differences were not statistically significant. However, for the (unstable) acentric chromosome fragments from both chromosomes, significant changes were observed: In the case of chromosome 2, an increase, and in the case of chromosome 18, a decrease. So these results indicate that gemcitabine has no large and consistent effect on the repair of genomic lesions that induce secondary chromosome breaks. Although it is clear that gemcitabine-induced radiosensitization can be expected when it is combined with brachytherapy, as with radiation at a high-dose rate, the mechanism of radiosensitization is so far not evident, and further experiments will be needed to elucidate this.

Inhibition of homologous recombination by hyperthermia shunts early double strand break repair to non-homologous end-joining.

In S and G2 phase mammalian cells DNA double strand breaks (DSBs) can potentially be repaired by homologous recombination (HR) or non-homologous end-joining (NHEJ). Results of several studies suggest that these two mechanistically distinct repair pathways can compete for DNA ends. Because HR and NHEJ differ with respect to error susceptibility, generation of chromosome rearrangements, which are potentially carcinogenic products of DSB repair, may depend on the pathway choice. To investigate this hypothesis, the influence of HR and NHEJ inhibition on the frequencies of chromosome aberrations in G2 phase cells was investigated. SW-1573 and RKO cells were treated with mild (41 °C) hyperthermia in order to disable HR and/or NU7441/cisplatin to inactivate NHEJ and frequencies of chromosomal fragments (resulting from unrepaired DSBs) and translocations (products of erroneous DSB rejoining) were studied using premature chromosome condensation (PCC) combined with fluorescence in situ hybridization (FISH). It is shown here that temporary inhibition of HR by hyperthermia results in increased frequency of ionizing-radiation (IR)-induced chromosomal translocations and that this effect is abrogated by NU7441- or cisplatin-mediated inhibition of NHEJ. The results suggest that in the absence of HR, DSB repair is shifted to the error-prone NHEJ pathway resulting in increased frequencies of chromosomal rearrangements. These results might be of consequence for clinical cancer treatment approaches that aim at inhibition of one or more DSB repair pathways.

Cyclopentenylcytosine does not enhance cisplatin-induced radiosensitization in human lung tumour cells.

The search for agents that enhance the effect of ionizing radiation has been an object of study for decades. In this study, the sensitizing properties of cyclopentenylcytosine (CPEC) on radiation and cisplatin-induced radiosensitization in human squamous lung carcinoma cells were investigated. Human lung tumour SW-1573 cells (SWp, parental; SWg, gemcitabine-resistant) were incubated with CPEC and cisplatin and subsequently irradiated with different doses of γ-rays. Clonogenic survival was determined to measure the effectiveness of the treatments. CPEC (1 or 2 µM) treatment for 4 h decreased the plating efficiency to 75 and 50% in SWp and SWg cells, respectively. In the SWg cells, 0.1 and 1 µM CPEC for 4 h enhanced the cell killing effect of cisplatin. However, an increase was not noted in the SWp cells. Due to the moderate toxicity of 1 µM for 4 h, this CPEC dose was used in the radiosensitization experiments. However, CPEC neither radiosensitized the lung tumour cells nor enhanced the radiosensitizing effect of cisplatin. A 2-h incubation with 4 µM cisplatin also decreased the plating efficiency to 75-80% in the two cell lines. Using this cisplatin dose, radiosensitization was obtained in the two cell lines. Although cisplatin treatment clearly radiosensitized the lung tumour cells, CPEC treatment did not. Cisplatin-induced radiosensitization was also not enhanced by CPEC.

Angiogenesis inhibitor DC101 delays growth of intracerebral glioblastoma but induces morbidity when combined with irradiation.

The combination of irradiation with angiogenic inhibition is increasingly being investigated for treatment of glioblastoma multiforme (GBM). We investigated whether vascular endothelial growth factor receptor-2 (VEGFR-2) inhibitor DC101 affects morbidity and tumor growth in irradiated and non-irradiated intracerebral GBM-bearing mice, controlled with sham treatments. End-points were toxicity, morbidity and histology. Irradiation either or not combined, reduced tumor size strongly, whereas DC101 mono-treatment reduced tumor size by 64%. Irradiation delayed morbidity from 5.8 weeks in sham-treated mice to 10.3 weeks. Morbidity after combined treatment occurred after 5.9 weeks. Treatment with angiogenesis inhibitor DC101 delays tumor growth but it induces morbidity, by itself or combined with irradiation.

Clonogenic assay of cells in vitro.

Clonogenic assay or colony formation assay is an in vitro cell survival assay based on the ability of a single cell to grow into a colony. The colony is defined to consist of at least 50 cells. The assay essentially tests every cell in the population for its ability to undergo "unlimited" division. Clonogenic assay is the method of choice to determine cell reproductive death after treatment with ionizing radiation, but can also be used to determine the effectiveness of other cytotoxic agents. Only a fraction of seeded cells retains the capacity to produce colonies. Before or after treatment, cells are seeded out in appropriate dilutions to form colonies in 1-3 weeks. Colonies are fixed with glutaraldehyde (6.0% v/v), stained with crystal violet (0.5% w/v) and counted using a stereomicroscope. A method for the analysis of radiation dose-survival curves is included.