Induction of DNA double-strand breaks by restriction enzymes in X-ray-sensitive mutant Chinese hamster ovary cells measured by pulsed-field gel electrophoresis.

This investigation was designed to determine whether the cytotoxic effects of different restriction endonucleases are related to the number and type of DNA double-strand breaks (DSBs) they produce. Chinese hamster ovary (CHO) K1 and xrs-5 cells, a radiosensitive mutant of CHO K1, were exposed to restriction endonucleases HaeIII, HinfI, PvuII and BamHI by electroporation. These enzymes represent both blunt and sticky end cutters with differing recognition sequence lengths. The number of DSBs was measured by pulsed-field gel electrophoresis (PFGE). Two forms of PFGE were employed: asymmetric field-inversion gel electrophoresis (AFIGE) for measuring the kinetics of DNA breaks by enzyme digestion and clamped homogeneous gel electrophoresis (CHEF) for examining the size distributions of damaged DNA. The amount of DNA damage induced by exposure to all four restriction enzymes was significantly greater in xrs-5 compared to CHO K1 cells, consistent with the reported DSB repair deficiency in these cells. Since restriction endonucleases produce DSBs alone as opposed to the various types of DNA damage induced by X rays, these results confirm that the repair defect in this mutant involves the rejoining of DSBs. Although the cutting frequency was directly related to the length of the recognition sequence for four restriction enzymes, there was no simple correlation between the cytotoxic effect and the amount of DNA damage produced by each enzyme in either cell line. This finding suggests that the type or nature of the cutting sequence itself may play a role in restriction enzyme-induced cell killing.

Biological dosimetry of absorbed radiation by C-banding of interphase chromosomes in peripheral blood lymphocytes.

Induction of premature chromosome condensation enables direct observation of radiation-induced cytogenetic damage in non-stimulated, interphase, human peripheral blood lymphocytes. This phenomenon can be explored in radiation protection for biological dosimetry in instances of accidental exposure to ionizing radiation. Quantification of an exposure by means of this approach has been limited so far mainly to the analysis of chromosome fragments. This limitation is due to the fact that conventional Giemsa staining of prematurely condensed chromosomes (PCCs) does not allow visualization of the centromeric regions and, as a result, the identification of dicentrics, centric rings and acentric fragments. In the present report a C-banding procedure, refined to avoid swelling and chromosome distortion of freshly prepared PCCs spreads, is used to identify such aberrations in non-stimulated human lymphocytes. The method allows immediate banding of the centromeric regions and enables scoring of aberrations within a time interval (3-4 h after blood sample withdrawal) that is only a fraction of that normally required when cells stimulated to proliferate are analysed at metaphase. The dose-response for dicentrics and centric rings measured in interphase lymphocytes was found to be similar to that obtained at metaphase. Measurement of dicentrics and centric rings in prematurely condensed chromosomes of human lymphocytes would provide valuable information on radiation dose estimates, especially in cases of extreme urgency.

Induction by H2O2 of DNA and interphase chromosome damage in plateau-phase Chinese hamster ovary cells.

The induction by H2O2 of DNA breaks, DNA double-strand breaks (DSBs), and interphase chromatin damage and their relationship to cytotoxicity were studied in plateau-phase Chinese hamster ovary (CHO) cells. Damage in interphase chromatin was assayed by means of premature chromosome condensation (PCC); DNA DSBs were assayed by nondenaturing filter elution (pH 9.6), and DNA breaks by hydroxyapatite chromatography. Cells were treated with H2O2 in suspension at 0 degrees C for 30 min and treatment was terminated by the addition of catalase. Concentrations of H2O2 lower than 1 mM were not cytotoxic, whereas concentrations of 40 and 60 mM reduced cell survival to 0.1 and 0.004, respectively. An induction of DNA breaks that was dependent on H2O2 concentration was observed at low H2O2 concentrations that reached a maximum at approximately 1 mM; at higher H2O2 concentrations induction of DNA breaks either remained unchanged or decreased. Damage at the chromosome level was not evenly distributed among the cells, when compared to that expected based on a Poisson distribution. Three categories of cells were identified after exposure to H2O2: cells with intact, control-like chromosomes, cells showing chromosome fragmentation similar to that observed in cells exposed to ionizing radiation, and cells showing a loss in the ability of their chromatin to condense into chromosomes under the PCC reaction. The fraction of cells with fragmented chromosomes, as well as the number of excess chromosomes per cell, showed a dose response similar to that of DNA DSBs, reaching a maximum at 1 mM and decreasing at higher concentrations. The results indicate that induction of DNA and chromosome damage by H2O2 follows a complex dependence probably resulting from a depletion of reducing equivalents in the vicinity of the DNA. Reducing equivalents are required to recycle the transition metal ions that are needed to maintain a Fenton-type reaction. The absence of cell killing at H2O2 concentrations that yielded the maximum amount of DNA and chromosome damage suggests that this damage is nonlethal and repairable. It is suggested that lethal DNA and chromosome damage is induced at higher concentrations of H2O2 where cell killing is observed by an unidentified mechanism.

Detection of DNA double-strand breaks in synchronous cultures of CHO cells by means of asymmetric field inversion gel electrophoresis.

A pulsed field gel electrophoresis technique, asymmetric field inversion gel electrophoresis (AFIGE), was used to evaluate induction by X-rays of DNA damage in CHO cells. The fraction of DNA activity released from the plug (FAR) was used as a measure for the amount of radiation-induced DNA damage, predominantly DNA double-strand breaks (dsb) (Stamato and Denko 1990), and was determined at various stages of growth and phases of the cell cycle in a range of doses between zero and 70 Gy. The FAR per unit dose fluctuated throughout the cell cycle and was high for cells irradiated in G1; it decreased as cells entered S and reached a minimum in the middle of this phase. The FAR per unit dose increased again as cells progressed towards the end of S, and reached values in G2 similar to those measured in G1. When damage was introduced into DNA by means of 125I decay similar fluctuations in the FAR per decay were observed throughout the cell cycle, suggesting that the variations in the FAR per unit of radiation dose observed throughout the cell cycle do not derive from alterations in the induction of dsb. The fluctuations in the FAR per unit dose throughout the cell cycle were quantitatively similar to the fluctuations in the fraction of activity eluted in irradiated cells assayed by the non-unwinding filter elution assay throughout the cycle (Okayasu et al. 1988), and suggest that both techniques respond to similar DNA replication-associated alterations of the biophysical and/or biochemical properties of the DNA molecule. It is concluded that caution needs to be exercised before differences observed in the FAR between different cell lines or between various phases of the cell cycle after exposure to a given dose of radiation are interpreted as suggesting differences in the induction of DNA dsb.

Measurement of DNA double-strand breaks in CHO cells at various stages of the cell cycle using pulsed field gel electrophoresis: calibration by means of 125I decay.

Experiments were performed to calibrate a recently developed pulsed field gel electrophoresis assay, the asymmetric field inversion gel electrophoresis (AFIGE), for the measurement of double-strand breaks (dsb) in the DNA of mammalian cells. Calibration was carried out by means of 125I decay accumulation under conditions preventing repair, and is based on the observation that each 125I decay in the DNA produces approximately one dsb. Iodine was incorporated into DNA in the form of 5'-iododeoxyuridine and decay accumulation was allowed in cells kept frozen at -80 degrees C. Since widely different DNA damage dose-response curves were obtained in cells exposed to X-rays in various phases of the cell cycle, calibration was performed using synchronized populations of cells that were allowed to accumulate DNA damage in G1, G1/S, mid-S, and G2 + M. For this purpose the fraction of activity (in DNA) released from the plug (FAR) was measured and correlated to the number of 125I decays accumulated during the elapsed period of time. Fluctuations in the FAR per 125I decay were observed throughout the cell cycle that were similar to those previously reported for the FAR per Gy of X-rays. Comparison of the FAR per 125I decay with the FAR per Gy gave an induction of 21 +/- 3, 31 +/- 3, 21 +/- 3 and 26 +/- 8 dsb per Gy per diploid DNA complement for G1, G1/S, S, and G2 + M cells, respectively. The results suggest that the observed fluctuations in the FAR per Gy throughout the cycle reflect cell-cycle-associated differences in the physicochemical properties of the DNA molecules that alter their electrophoretic mobility, rather than variations in the induction of dsb per Gy, i.e. the sensitivity of the assay fluctuates throughout the cycle. We propose that similar phenomena underlie the observed fluctuations throughout the cell cycle in the fraction of activity eluted per Gy in the non-unwinding filter elution assay. 125I decays accumulated at 4 degrees C in partly purified DNA from cells embedded in agarose plugs and lysed immediately, gave FAR identical to those obtained with cells kept frozen. This finding suggests that indirect effects do not significantly contribute to DNA damage induction by 125I decay, and indicate that calibration of electrophoresis techniques for dsb measurements can be carried out using this simplified protocol.

Radiosensitivity throughout the cell cycle and repair of potentially lethal damage and DNA double-strand breaks in an X-ray-sensitive CHO mutant.

The response to ionizing radiation of synchronized and plateau-phase populations of xrs-5 cells was studied at the DNA and cellular level. Induction and repair of DNA double-strand breaks (dsb) were measured by the non-unwinding filter elution technique. Biphasic survival curves were obtained throughout the cell cycle as well as in the plateau phase, suggesting the presence in the cultures of a radiation-sensitive and a radiation-resistant subpopulation. Delayed plating of plateau-phase xrs-5 cells did not significantly modify cell survival indicating a deficiency in the repair of potentially lethal damage (PLD). Post-irradiation treatment with araA further indicated a deficiency in the repair of a form of PLD sensitive to this drug in the radiation-sensitive subpopulation. The radiosensitivity of synchronized populations decreased as cells progressed through S. Cells in late S were found to be most resistant to radiation, whereas cells irradiated in G2 + M had a sensitivity similar to that of cells irradiated in G1. The decrease in radiosensitivity in S was tentatively attributed to an increase in the proportion of radioresistant cells in the cell population. Preliminary analysis of the results obtained, assuming the presence of two subpopulations showing an exponential response as a function of dose, indicated no change in the D0 values of the radiation-sensitive and the radiation-resistant component throughout the cell cycle. The D0 value of the radiation-resistant component was similar to the D0 of repair-proficient CHO cells. At no stage in the cell cycle were shoulder-type survival curves observed. Transient expression of the xrs-5 repair gene may underlie the observed variation of radiosensitivity throughout the cell cycle. This gene expression could occur either at random, in a subpopulation of cells independent of cell cycle phase, or in a specific manner related to gene replication and the hemimethylated state transiently found immediately thereafter. It is proposed that the resistant tail observed in the survival curves throughout the cell cycle is due to the former phenomenon, and the decreased radiosensitivity in late S to the latter. xrs-5 cells were found to be deficient in dsb repair in the plateau phase of growth, as well as in G1 and mid-S phase. The latter observation indicates that the reduction in radiation sensitivity in S is not associated with a modulation in the ability of cells to repair dsb.

Production and repair of chromosome damage in an X-ray sensitive CHO mutant visualized and analysed in interphase using the technique of premature chromosome condensation.

Production and repair of chromosome damage were studied in interphase xrs-5 cells by means of premature chromosome condensation (PCC). The results obtained were compared with those previously reported for CHO cells. Production of chromosome damage per unit of absorbed radiation dose was in xrs-5 cells larger by a factor of 2.6 than in CHO cells (5.2 breaks per cell per Gy). Changes in chromatin structure, associated with the radiation-sensitive phenotype of xrs-5 cells, that increase the probability of conversion of a DNA double-stand break (dsb) to a chromosome break are involved to explain this effect. Repair of chromosome breaks as measured in plateau-phase G1 cells was deficient in xrs-5 cells and the number of residual chromosome breaks was practically identical to the number of lethal lesions calculated from survival data. This observation suggests that non-repaired chromosome breaks are likely to be manifestations of lethal events in the cell. The yield of ring chromosomes scored after a few hours of repair was higher by a factor of three in xrs-5 compared with CHO cells. This increase in ring formation suggests an increase in the probability of misrepair of chromosome damage that may stem either from the reduced ability of xrs-5 cells to repair dsb, or from the higher production of chromosome fragments observed per cell and per Gy.

Effects of hyperthermia on chromatin condensation and nucleoli disintegration as visualized by induction of premature chromosome condensation in interphase mammalian cells.

The effects of hyperthermia on chromatin condensation and nucleoli disintegration, as visualized by induction of premature chromosome condensation in interphase mammalian cells, was studied in exponentially growing and plateau phase Chinese hamster ovary cells. Exposure to heat reduced the ability of interphase chromatin to condense and the ability of the nucleolar organizing region to disintegrate under the influence of factors provided by mitotic cells when fused to interphase cells. Based on these effects treated cells were classified in three categories. Category 1 contained cells able to condense their chromatin and disintegrate the nucleolar organizing region. Category 2 contained cells able to only partly condense their chromatin and unable to disintegrate the nucleolar organizing region. Category 3 contained cells unable to condense their chromatin and unable to disintegrate the nucleolar organizing region. The fraction of cells with nondisintegrated nucleoli increased with increasing exposure time at 45.5 degrees C and reached a plateau at almost 100% after about 20 min. Exponentially growing and plateau phase cells showed similar response. Recovery from the effects of heat on chromatin condensation and disintegration of the nucleolar organizing region depended upon the duration of the heat treatment. For exposures up to 15 min at 45.5 degrees C, a gradual reduction in the fraction of cells with nondisintegrated nucleoli was observed when cells were allowed for repair at 37 degrees C. However, only a very limited amount of repair was observed after a 30-min exposure to 45.5 degrees C. The repair times observed at the chromosome level were similar to those reported for the removal of excess protein accumulating in chromatin or the nuclear matrix, suggesting a causal relationship between the two phenomena. It is proposed that nuclear protein accumulation on chromatin or in the nuclear matrix reduces the accessibility of chromatin to enzymes responsible for the phosphorylation reactions necessary for chromatin condensation and disintegration of the nucleolus.