Controlled rotation of optically trapped microscopic particles.

We demonstrate controlled rotation of optically trapped objects in a spiral interference pattern. This pattern is generated by interfering an annular shaped laser beam with a reference beam. Objects are trapped in the spiral arms of the pattern. Changing the optical path length causes this pattern, and thus the trapped objects, to rotate. Structures of silica microspheres, microscopic glass rods, and chromosomes are set into rotation at rates in excess of 5 hertz. This technique does not depend on intrinsic properties of the trapped particle and thus offers important applications in optical and biological micromachines.

Does single-dose cell resistance to the radio-mimetic zeocin correlate with a zeocin-induced adaptive response in Chlamydomonas reinhardtii strains?

This study aimed to test whether a correlation exists between single-dose resistance to zeocin and the ability to develop a zeocin-induced adaptive response (AR) in Chlamydomonas reinhardtii strains. Three genotypes were used: wild type (WT) strain 137C and two strains (H-3 and AK-9-9), which are highly resistant to radiation based on survival studies. Based on a micro-colony assay, the strains could be arranged according to their single-dose resistance to zeocin as follows: AK-9-9 > H-3 > 137C. However, zeocin induced a similar level of DSB in strains AK-9-9, H-3 and 137C. The radio- and zeocin-resistant strains AK-9-9 and H-3 showed higher DSB rejoining capacity than the WT strain 137C, suggesting that DSB rejoining can at least partly account for different cell survival. Both WT and radio-resistant strains develop zeocin-induced AR involving increased DSB rejoining. The radio- and zeocin-resistant strains AK-9-9 and H-3 again showed higher DSB rejoining capacity than the WT strain 137C. The higher resistance of strains H-3 and AK-9-9 did not abrogate their ability to adapt, albeit with a smaller magnitude as compared to the WT strain. The obtained results characterize new radio-resistant C. reinhardtii strains, which enrich the collection of resistant C. reinhardtii strains.

A role for topoisomerase II alpha in the formation of radiation-induced chromatid breaks.

Chromatid breaks in cells exposed to low dose irradiation are thought to be initiated by DNA double-strand breaks (DSB), and the frequency of chromatid breaks has been shown to increase in DSB rejoining deficient cells. However, the underlying causes of the wide variation in frequencies of G2 chromatid breaks (or chromatid 'radiosensitivity') in irradiated T-lymphocytes from different normal individuals and cancer cases are as yet unclear. Here we report evidence that topoisomerase II alpha expression level is a factor determining chromatid radiosensitivity. We have exposed the promyelocytic leukaemic cell line (HL60) and two derived variant cell lines (MX1 and MX2) that have acquired resistance to mitoxantrone and low expression of topoisomerase II alpha, to low doses of gamma-radiation and scored the induced chromatid breaks. Chromatid break frequencies were found to be significantly lower in the variant cell lines, compared with their parental HL60 cell line. Rejoining of DSB in the variant cell lines was similar to that in the parental HL60 strain. Our results indicate the indirect involvement of topoisomerase II alpha in the formation of radiation-induced chromatid breaks from DSB, and suggest topoisomerase II alpha as a possible factor in the inter-individual variation in chromatid radiosensitivity.

Telomere length abnormalities in mammalian radiosensitive cells.

Telomere lengths in radiosensitive murine lymphoma cells L5178Y-S and parental radioresistant L5178Y cells were measured by quantitative fluorescence in situ hybridization. Results revealed a 7-fold reduction in telomere length in radiosensitive cells (7 kb) in comparison with radioresistant cells (48 kb). Therefore, it was reasoned that telomere length might be used as a marker for chromosomal radiosensitivity. In agreement with this hypothesis, a significant inverse correlation between telomere length and chromosomal radiosensitivity was observed in lymphocytes from 24 breast cancer patients and 5 normal individuals. In contrast, no chromosomal radiosensitivity was observed in mouse cell lines that showed shortened telomeres, possibly reflecting differences in radiation responses between primary cells and established cell lines. Telomere length abnormalities observed in radiosensitive cells suggest that these two phenotypes may be linked.

Adaptive response of a new radioresistant strain of Chlamydomonas reinhardtii and correlation with increased DNA double-strandbreak rejoining.

In order to study the relationship between radioresistance and the adaptive response, we aimed to produce a new strain of Chlamydomonas reinhardtii with characteristics of high radioresistance coupled with a protoplast structure typical for the genus, and the cell-wall-less phenotype to facilitate rapid cell lysis in DNA double-strand break (DSB) assays. The adaptive response of the new strain was investigated using clonogenic and DSB assays. Strain H-3 was derived by mating a radioresistant strain (AK-9-9) with the cell-wall-less mutant CW15 strain and selecting for radioresistance by clonogenic assay. The random amplification of polymorphic DNA (RAPD) molecular marker system was used to evaluate genetic polymorphisms between H-3 and other related C. reinhardtii strains. DSB were estimated using constant-field electrophoresis. Of several mutant strains tested, strain H-3 was shown to be most radioresistant on the basis of dose to give a 90% lethality (LD90) rate and dose to give a 99% lethality rate (LD99). In addition to its high radioresistance and thinner cell wall as compared with that of the other parental strain AK-9-9, H-3 also expressed a radiation-induced adaptive response measured by clonal survival when given a priming dose before a test dose. DSB were also rejoined more rapidly in cells exposed to a priming dose 4 h previously. It is concluded from split-dose experiments that the already highly radioresistant strain H-3 is further capable of 'over recovery' or adaptation to radiation exposure. Accelerated DSB rejoining in cells given a priming dose may underlie the cellular adaptive response in this organism.

Progress towards understanding the nature of chromatid breakage.

The wide range of sensitivities of stimulated T-cells from different individuals to radiation-induced chromatid breakage indicates the involvement of several low penetrance genes that appear to link elevated chromatid breakage to cancer susceptibility. The mechanisms of chromatid breakage are not yet fully understood. However, evidence is accumulating that suggests chromatid breaks are not simply expanded DNA double-strand breaks (DSB). Three models of chromatid breakage are considered. The classical breakage-first and the Revell "exchange" models do not accord with current evidence. Therefore a derivative of Revell's model has been proposed whereby both spontaneous and radiation-induced chromatid breaks result from DSB signaling and rearrangement processes from within large looped chromatin domains. Examples of such rearrangements can be observed by harlequin staining whereby an exchange of strands occurs immediately adjacent to the break site. However, these interchromatid rearrangements comprise less than 20% of the total breaks. The rest are thought to result from intrachromatid rearrangements, including a very small proportion involving complete excision of a looped domain. Work is in progress with the aim of revealing these rearrangements, which may involve the formation of inversions adjacent to the break sites. It is postulated that the disappearance of chromatid breaks with time results from the completion of such rearrangements, rather than from the rejoining of DSB. Elevated frequencies of chromatid breaks occur in irradiated cells with defects in both nonhomologous end-joining (NHEJ) and homologous recombination (HR) pathways, however there is little evidence of a correlation between reduced DSB rejoining and disappearance of chromatid breaks. Moreover, at least one treatment which abrogates the disappearance of chromatid breaks with time leaves DSB rejoining unaffected. The I-SceI DSB system holds considerable promise for the elucidation of these mechanisms, although the break frequency is relatively low in the cell lines so far derived. Techniques to study and improve such systems are under way in different cell lines. Clearly, much remains to be done to clarify the mechanisms involved in chromatid breakage, but the experimental models are becoming available with which we can begin to answer some of the key questions.

Lack of spontaneous and radiation-induced chromosome breakage at interstitial telomeric sites in murine scid cells.

Interstitial telomeric sites (ITSs) in chromosomes from DNA repair-proficient mammalian cells are sensitive to both spontaneous and radiation-induced chromosome breakage. Exact mechanisms of this chromosome breakage sensitivity are not known. To investigate factors that predispose ITSs to chromosome breakage we used murine scid cells. These cells lack functional DNA-PKcs, an enzyme involved in the repair of DNA double-strand breaks. Interestingly, our results revealed lack of both spontaneous and radiation-induced chromosome breakage at ITSs found in scid chromosomes. Therefore, it is possible that increased sensitivity of ITSs to chromosome breakage is associated with the functional DNA double-strand break repair machinery. To investigate if this is the case we used scid cells in which DNA-PKcs deficiency was corrected. Our results revealed complete disappearance of ITSs in scid cells with functional DNA-PKcs, presumably through chromosome breakage at ITSs, but their unchanged frequency in positive and negative control cells. Therefore, our results indicate that the functional DNA double-strand break machinery is required for elevated sensitivity of ITSs to chromosome breakage. Interestingly, we observed significant differences in mitotic chromosome condensation between scid cells and their counterparts with restored DNA-PKcs activity suggesting that lack of functional DNA-PKcs may cause a defect in chromatin organization. Increased condensation of mitotic chromosomes in the scid background was also confirmed in vivo. Therefore, our results indicate a previously unanticipated role of DNA-PKcs in chromatin organisation, which could contribute to the lack of ITS sensitivity to chromosome breakage in murine scid cells.

Chromosome damage induced by nanomolar concentrations of bleomycin in porated mammalian cells.

We have examined chromosome damage caused by a wide range of bleomycin (BLM) concentrations in Chinese hamster ovary (CHO-K1) cells reversibly porated by the bacterial cytotoxin streptolysin-O (SLO). Chromosome damage was measured using the micronucleus cytokinesis block technique (employing cytochalasin-B). Treatment of exponentially growing cells with 0.045 IU/mL SLO for 5 min resulted in up to a thousand-fold and a million-fold increase in biological effectiveness, compared to treatment in the absence of SLO for 24 hr and 5 min, respectively. Increases in micronuclei of 4-5 times background level were observed after only 5 min exposure to the drug in the presence of SLO at doses as low as 100 pg/mL (approximately 70 pmol/L). These results indicate that the use of SLO may facilitate the treatment of cells with BLM for periods of time resembling acute exposure to ionizing radiations.

Independent forms of potentially lethal damage fixed in plateau-phase Chinese hamster cells by postirradiation treatment in hypertonic salt solution or araA.

Plateau-phase Chinese V79 hamster cells were sequentially treated after exposure to gamma rays in medium made hypertonic by the addition of sodium chloride (370 mM) and with various concentrations of 9-beta-D-arabinofuranosyladenine (araA) to study their combined effect on fixation of potentially lethal damage (PLD). A 10-min treatment in hypertonic medium fixed an extensive amount of PLD and caused a decrease in D0 from 1.8 to 1.2 Gy without significantly affecting Dq. Subsequent treatment with araA caused further fixation of PLD but resulted in a specific, concentration-dependent reduction in Dq from 4.9 to 1.6 Gy after a 4-h exposure to 150 microM araA. A 30-min treatment in hypertonic medium reduced not only Do (from 1.8 to 1.0 Gy) but also Dq (from 4.9 to 2.7 Gy). Subsequent treatment with araA in this case affected only the residual shoulder, reducing it to 1.6 Gy after a 4-h treatment with 100 microM araA, a value similar to that obtained after treatment with araA of cells exposed to salt for only 10 min. When the repair of PLD fixed by a 10-min treatment with salt was measured by delaying its postirradiation application in the presence of various amounts of araA, a small decrease in the repair rate was observed but no significant effect on the relative increase in survival. Qualitatively similar results were obtained for repair of PLD sensitive to araA after a 10-min treatment in hypertonic medium. These results suggest the radiation induction of forms of PLD with different sensitivity to fixation by postirradiation treatments. araA is proposed to fix a form of PLD termed alpha-PLD, the repair of which takes place within 4-6 h and which causes the formation of the shoulder in the survival curve of cells plated immediately after irradiation. Short treatments in hypertonic medium (less than 10 min) are proposed to fix a form of PLD termed beta-PLD, the repair of which takes place within 1 h and leads to restoration of the slope to values equal to those obtained in the survival curve of cells plated immediately after irradiation. However, longer treatments in hypertonic medium also affect Dq and thus also alpha-PLD. Repair of beta-PLD was not significantly affected by araA and repair of alpha-PLD was not significantly affected by short hypertonic treatment, thus indicating the independence of the two forms of PLD.(ABSTRACT TRUNCATED AT 400 WORDS)

Higher-order chromatin structure-dependent repair of DNA double-strand breaks: modeling the elution of DNA from nucleoids.

A possible relationship between the repair of DNA double-strand breaks (DSBs) and their distribution within higher-order chromatin (Johnston and Bryant, Int. J. Radiat. Biol. 66, 531-536, 1994) has recently been demonstrated. Radiosensitive cells deficient for components of the DNA-dependent protein kinase DSB repair pathway exhibited a particular failure in the rejoining of DSBs occurring as multiples within looped DNA structures. Here, a Poisson-based model of induction of DSBs and elution of DNA from residual nuclear structures is presented. By applying this model to cells of a panel of human and rodent cell lines, a mean of 1.6 Mbp for the size of the relevant looped structures was obtained. Such large chromatin structures are of the same magnitude as those observed by functional mapping of interphase and mitotic chromosome structure, nucleoid sedimentation and the "replicon clusters" apparent during DNA replication. This work supports the hypotheses that (1) such structures are critical targets for induction of DSBs and (2) the distribution of damage within these domains may be a factor in the response and sensitivity of mammalian cells to ionizing radiation.

G2 chromatid aberrations: kinetics and possible mechanisms.

Chromatid breaks and exchanges are induced by radiation in G2 mammalian cells. Breaks are at a maximum number at about 30 min after irradiation and decrease apparently exponentially with time between irradiation and sampling. Few breaks are observed immediately following exposure, probably as a result of selection of mitotic cells where chromosomes are condensed and there is consequently a lack of time for expression of damage. The change in frequency of breaks with time, from 30 min after radiation exposure and onwards, can be interpreted in two possible ways: either in terms of a repair process or in terms of a change in radiosensitivity through G2. However, our results with an inhibitor of repair of DNA double-strand breaks (ara A) and with "transient hypothermia" which extends the G2 phase, argue for an interpretation based on rejoining of chromatid breaks, possibly reflecting the repair of a subclass of dsb. Data from experiments with irradiated and restriction endonuclease treated radiosensitive mutant rodent lines indicate that enhanced levels of conversion of dsb into chromosomal aberrations may be largely independent of repair rates of bulk dsb. In CHO cells and in human lymphocytes exchanges initially increase rapidly with time and then remain at a constant frequency, supporting the notion of a uniform chromosomal radiosensitivity throughout most of G2 and providing further evidence that the mechanism for mis-joining broken chromatids (leading to exchanges) is different from that for rejoining of chromatoid breaks. Ratios of breaks to exchanges were found to vary in different cell lines and at different times during treatment with inhibitors or at altered temperatures, possibly (in different cell lines) indicating different levels of enzymes involved in misjoining, but suggesting that the mechanisms of chromosomal rejoining and misjoining are independent, at least to some degree.

Effects of the nucleoside analogues alpha-ara A, beta-ara A and beta-ara C on cell growth and repair of both potentially lethal damage and DNA double strand breaks in mammalian cells in culture.

The effects of the DNA polymerase inhibitors a-ara A and beta-ara C have been compared with those of beta-ara A on the endpoints of cell growth, repair of x-ray induced potentially lethal damage (PLD) and repair of x-ray induced DNA double strand breaks, a-ara A was found to have no effects on any of the endpoints studied. beta-ara C inhibited cell growth and DNA double strand breaks repair more strongly than beta-ara A but it was less effective in inhibiting repair of PLD. The inhibition of PLD repair by beta-ara C resulted in a diminution of the shoulder width of the x-ray survival curve, a result similar to that previously found for beta-ara A. The effectiveness of beta-ara C was enhanced when cells were treated with the drug in fresh medium rather than under plateau phase conditions. The lower effectiveness of beta-ara C, when compared with beta-ara A in causing expression of PLD, is interpreted in terms of a difference in the ability of the two drugs to cause fixation or misrepair of the DNA double strand breaks during or after treatment with the drugs.

Use of restriction endonucleases to study relationships between DNA double-strand breaks, chromosomal aberrations and other end-points in mammalian cells.

Some of the cellular effects of radiation, such as mutations, chromosomal aberrations and cell killing, can be mimicked by inducing 'pure' double-strand breaks (dsb) in DNA of cells with restriction endonucleases (RE), although the chemical structure of the ends of dsb induced by RE are likely to differ from those induced by X-rays. Chromosomal aberrations are induced by treatment of cells with a variety of RE at all stages of the cell cycle. The frequency with which RE induce dsb in the DNA may be one factor determining the number of aberrations induced. However, the structure of the dsb generated may also determine the frequencies of aberrations induced. RE which generate 'cohesive-ended' dsb in the DNA have been shown to induce lower frequencies of aberrations than those causing 'blunt-ended' dsb, when inactivated Sendai virus is used to permeabilize cells. Other methods, involving a hypertonic shock to the treated cells, have led to results in which there is little or no difference in the effectiveness between the two types of dsb. It is argued here that the use of treatments which cause a hypertonic shock may influence the frequencies of aberrations induced.

DNA damage, repair and chromosomal damage.

An important question in radiobiology is the relationship between primary DNA damage and chromosomal aberrations. What determines the chromosomal aberration frequency, especially in radiosensitive cells? Much evidence points to the double-strand break (dsb) as the critical lesion, however there is controversy over whether it is the initial induction, repair or residual dsb which determine of the level of expression of chromosome damage. The picture is further complicated by the fact that chromosome damage can be measured at several levels e.g. at metaphase, as micronuclei and as prematurely condensed chromosomes. Differential frequencies of chromosome damage are measured in different cell lines. Repair and residual dsb may play a role in metaphase aberrations when cells are exposed in G1, but in irradiated G2 cells the differential frequencies do not depend on repair of dsb or on the residual level of dsb since a difference in the cell lines is observed at short intervals after irradiation, and in radiosensitive cell lines where there is no deficiency in the repair of dsb, e.g. ataxia telangiectasia cells. Thus, at least in G2 cells, a mechanism involving 'conversion' of dsb into chromatid breaks is proposed. There are a number of possible reasons for high conversion of dsb into chromatid breaks including altered chromatin structure, high chromosome condensation rates and covalent closure of chromosome ends.

The signal model: a possible explanation for the conversion of DNA double-strand breaks into chromatid breaks.

PURPOSE: To present and evaluate the 'signal' model for the formation of radiation-induced chromatid breaks. CONCLUSIONS: Chromatid breaks in human cells represent the apparent interstitial loss of up to about 40 Mbp of DNA, difficult to account for as single lesions under the classical 'breakage-and-reunion' hypothesis. If breakage-first resulted from two interacting DNA double-strand breaks (dsb) with the loss or displacement of the intervening fragment, a dose-squared relationship would be predicted for chromatid breaks. However, the relationship between chromatid break frequency and dose for human cells is linear. The alternative 'exchange' model of Revell is based on the principle of the interaction of two initiating lesions, thus also predicting a dose-squared relationship for chromatid 'breaks'. The signal model explains the conversion of dsb into chromatid breaks on the assumption that a single dsb generates a signal which triggers the cell to initiate a recombinational exchange involving a large loop of chromatin. Incomplete exchanges would be observed as chromatid breaks. Possible candidates for the signalling molecule(s) are DNA protein kinase (DNA PK) and the ATM protein.

A component of DNA double-strand break repair is dependent on the spatial orientation of the lesions within the higher-order structures of chromatin.

By the use of a modified neutral filter elution procedure variations in the repair of DNA dsb have been observed between the ionizing radiation sensitive mutant xrs-5 and the parent cell line CHO-K1. Conventional neutral filter elution requires harsh lysis conditions to remove higher-order chromatin structures which interfere with elution of DNA containing dsb. By lysing cells with non-ionic detergent in the presence of 2 mol dm-3 salt, histone-depleted structures that retain the higher-order nuclear matrix organization, including chromatin loops, can be produced. Elution from these structures will only occur if two or more dsb lie within a single-looped domain delineated by points of attachment to the nuclear matrix. Repair experiments indicate that in CHO cells repair of dsb in loops containing multiple dsb are repaired with slow kinetics whilst dsb occurring in loops containing single dsb are repaired with fast kinetics. Xrs-5 cells are defective in the repair of multiply damaged loops. This work indicates that the spatial orientation of dsb in the higher-order structures of chromatin are a possible factor in the repair of these lesions.

Responses of radiosensitive repair-proficient cell lines to restriction endonucleases.

Radiosensitive mutant mammalian cell lines fall into two categories: (1) those exhibiting a deficiency in the rejoining of dsb, e.g. Chinese hamster xrs and XR1, and murine scid cells; and (2) those exhibiting apparently normal rejoining of bulk dsb, e.g. hamster irs mutants and cells from ataxia-telangiectasia individuals. Cells of both types also show hypersensitivity to restriction endonucleases when applied by cell poration techniques. These data are reviewed, and new data are presented for Pvu II treatment of the radiosensitive dsb repair-proficient Chinese hamster VC4 mutant, which has been reported to have normal cellular and chromosomal sensitivity to restriction endonucleases and neutrons. We find that VC4 is hypersensitive to blunt-ended dsb generated by PvuII. We conclude that the enhanced sensitivity of this and other repair-proficient mutants to radiation and restriction endonucleases results from a dsb processing defect leading to abnormal conversion of dsb into chromosomal aberrations.

Enhanced chromosomal response of ataxia-telangiectasia cells to specific types of DNA double-strand breaks.

The chromosomal response of two ataxia-telangiectasia (A-T) lymphoblastoid cell lines (A-T-PA and A-T-KM) to restriction endonucleases (RE) is compared with that of a normal (N-SW) lymphoblastoid cell line. The RE used were PvuII (generating DNA double-strand breaks with blunt termini), BamHI (cohesive termini with 4 base, 5' overhangs) and PstI (cohesive termini with 4 base 3' overhangs). Chromatid aberrations were analysed in cells 5 h after treatment. Cells were porated using streptolysin O to allow entry of RE. Both A-T lines showed an enhanced frequency of chromatid breaks in G2 phase compared with normal cells in response to RE. The enhanced response of A-T cells was most marked in the case of PvuII treatment when the enhancement ratios were 2.5 and 4.2 for A-T-PA and A-T-KM respectively. However, the frequency of DNA double-strand breaks (dsb), measured by neutral filter elution, were considerably lower in A-T-PA cells than N-SW, due to a lower efficiency of poration. When A-T-PA cells were treated with streptolysin O at a higher concentration (0.3 Units/ml), a condition that apparently led to a similar level of poration in A-T-PA as in N-SW cells treated with 0.06 Units/ml as judged by the similar number of dsb induced in the two lines for a given PvuII concentration, the enhancement ratio for A-T-PA cells treated with PvuII increased from 2.5 to 5.8. BamHI and PstI were found to be less clastogenic in all three cell lines as found previously for Chinese hamster cells, although part of this effect may be due to a lower activity, particularly in the case of PstI. However, even at a 4-6-fold higher concentration, BamHI was still less clastogenic than PvuII. It is concluded that dsb with blunt termini are more clastogenic than those with cohesive termini. The results suggest that the chromosomal sensitivity of A-T cells may result from a defect causing a higher rate of conversion of dsb into chromatid aberrations.

Kinetics of chromatid aberrations in G2 ataxia-telangiectasia cells exposed to X-rays and ara A.

The cytogenetic effects of X-rays alone or in combination with 9-beta-D-arabinofuranosyladenine (ara A) were studied in an immortalized fibroblastic line of ataxia-telangiectasia (A-T) cells. The average length of G2 in this line was determined by autoradiographic labelling (labelled mitoses) to be approximately 5 h. Samples of A-T cells treated with or without ara A, 4 h prior to fixation were irradiated at 1/2-hourly intervals, from 1.5 h to 3.5 h before fixation and then examined for the presence of metaphase chromatid aberrations. It is postulated that the kinetics of disappearance (rejoining) of chromatid deletions with postirradiation incubation time reflects the underlying repair of dsb. This rejoining was found to be inhibited by ara A. Thus the frequency of deletions in the presence of ara A should represent the frequency of deletions in the absence of dsb repair. The rejoining kinetics for deletions in A-T was similar to that found in a previous study of normal human fibroblasts (Mozdarani and Bryant 1987). The number of deletions in X-irradiated A-T cells at 1.5 h before fixation was found to be higher by a factor of approximately 2 than that found previously in normals, indicating that in A-T a higher rate of conversion of dsb into chromatid deletions occurs. The frequency of exchanges induced in G2 A-T cells was similarly enhanced but, unlike the situation in normal cells, ara A was found to cause only a slight increase in this frequency.

The induction of DNA double-strand breaks in CHO cells by Pvu II: kinetics using neutral filter elution (pH 9.6).

Chinese hamster CHO K1 cells were treated with the restriction endonuclease Pvu II during electroporation and assayed for DNA double-strand breaks (dsb). Dsb were measured by the non-denaturing filter elution technique (pH 9.6) at various times up to 24 h after restriction endonuclease (RE) treatment. The frequency of dsb following electroporation in the presence of 200 units/ml Pvu II increased over the post-treatment incubation period. This was found not to be due to cell or DNA degradation, indicating that Pvu II remains active for at least 24 h inside the cell. We suggest that these kinetics of dsb result from a competition between incision (by Pvu II) and dsb repair.