In general mammalian cells recover from DNA synthesis inhibition by ultraviolet light (u.v.) before most of the pyrimidine dimers have been removed from the genome. This is a complex phenomenon whose biological significance has not been fully assessed. In Chinese hamster V79 cells this recovery seems to be directly coupled to an enhanced rate of double-stranded DNA elongation. The presence of the DNA polymerase it inhibitor, aphidicolin, after u.v. irradiation produces two different responses. At low concentration, sufficient to inhibit 95 % of DNA replication but having no effect on excision repair, the drug has no effect on the recovery. This shows that ongoing replicative DNA synthesis is not required for recovery. At higher concentrations of aphidicolin, sufficient to block excision repair, the recovery phenomenon was prevented. The recovery was also prevented by actinomycin D at a concentration that inhibits 60 % of RNA synthesis. In quantitative autoradiography experiments in which previously irradiated cells were fused with unirradiated cells the nuclei of the latter exhibited a higher resistance to inhibition by u.v. than nuclei from non-fused cells. These results indicate that: (1) even the low repair rate exhibited by V79 cells (relative to human cells) is important for recovery; although most of the dimers remain in the V79 genome after recovery of DNA synthesis, either the removal of lesions from some important region of chromatin or the activity of the repair process itself is important for the recovery; (2) the recovery mechanism is induced and depends on RNA synthesis and the production of specific factors.

Finally, we have observed that cells previously treated with fluorodeoxyuridine become more resistant to inhibition by u.v. After irradiation these cells replicate DNA faster than untreated cells. Since it has been shown that this drug activates unused origins of replication in Chinese hamster cells, reducing the average replicon size, we assume that the acquired resistance has to do with the operation of a larger number of smaller replicons. This may also be the mechanism whereby recovery from inhibition occurs after u.v. irradiation.

Damage to DNA elicits several biological responses. Besides the immediate mobilization to eliminate the burden of lesions the cells exhibit a slower response, whose kinetics are compatible with an inducible process, and which consists ultimately of acquiring tolerance to the remaining lesions (Meneghini et al. 1981). This tolerance is reflected by a recovery from DNA synthesis inhibition (Meneghini et al. 1981), an enhanced capability to reactivate infecting virus whose genome has been damaged (Bockstahler & Lytle, 1970; Sarasin & Hanawalt, 1978), and an increasing resistance to lethal injuries produced by ultraviolet light (u.v.) (Domon & Rauth, 1973; Todd, 1973; Menck & Meneghini, 1982). At present we do not know whether these phenomena are causally related. From the molecular point of view the recovery from DNA synthesis inhibition is interesting because it may reveal the occurrence of inducible tolerance mechanisms in the eukaryotic cell perhaps resembling the SOS mechanism in bacteria. It is well established that pyrimidine dimers arrest the replication fork movement (Meneghini et al. 1981), and the recovery might be related to an inducible mechanism that somehow overcomes the block. Such recovery could also be related to induced mutagenesis and transformation. To study recovery of DNA synthesis, many investigators have used the method of DNA sedimentation in alkaline sucrose gradients (Meyn & Humphrey, 1971; Lehmann & Kirk-Bell, 1972; D’Ambrosio & Setlow, 1976). However, this technique has its drawbacks; it does not permit a direct analysis of the rate of DNA elongation because of the formation of daughter-strand gaps (Lehmann, 1972; Cordeiro-Stone et al. 1979). Moreover, possible artefacts may arise from an expanded pool of longer DNA strands after u.v. irradiation, which remain as a source of further chain elongation at later pulse-labelling times (Painter, 1978). The method of DNA-fibre autoradiography has been used to measure directly the elongation rate of double-stranded DNA (Edenberg, 1976; Doniger, 1978; Dahle et al. 1980). In Chinese hamster cells, this technique has revealed that, along with a recovery from inhibition by u.v. of [1H]thymidine incorporation, there occurs a recovery in the rate of elongation of double-stranded DNA (Dahle et al. 1980). To speed up analysis of recovery, we have used a technique based on CsCl gradient centrifugation of DNA pulse-labelled with BrdUrd (bromodeoxyuridine), which also measures the rate of elongation of double-stranded DNA (Cordeiro & Meneghini, 1973). We found that recovery of [1H] thymidine incorporation in V79 cells is accompanied by a recovery of DNA elongation rate (Ventura & Meneghini, 1984). Rodent cells have been used in many of these experiments because, in spite of a limited excision repair capacity, they recover from DNA synthesis inhibition, suggesting that other mechanisms are involved. However, even in human cells excision repair, though it appears to be necessary (Moustacchi et al. 1979), is not sufficient for recovery (Lehmann et al. 1979). More recently, studies with Chinese hamster ovary (CHO) cells defective in excision repair have confirmed that the low, wild-type repair ability is required for recovery (Griffiths & Ling, 1985). However, another u.v.-sensitive CHO mutant, normal in excision repair but defective in postreplication repair, has been shown to be defective in DNA synthesis recovery (Collins & Waldren, 1982). Thus again excision repair does not seem to be sufficient for recovery.

Several points remain obscure. There are doubts about whether the recovery is an inducible process, dependent on previous transcription and protein synthesis. In addition, the mechanism of recovery itself is unknown. In principle, it can be ascribed to a larger number of active replicons or to an acquired capacity of the replication machinery to bypass a lesion (Park & Cleaver, 1979; Meneghini & Mello- Filho, 1983). In both cases the net result would be a faster overall DNA elongation rate.

We set about testing some of these points, using several metabolic inhibitors. The use of inhibitors involves some risks because of their lack of absolute specificity, but it can reveal aspects that otherwise would be difficult to assess. We took care to ascertain that, upon removal of the inhibitors, the level of DNA synthesis in unirradiated cells returned to the level of untreated cells, and also that the drug had no secondary effects on chromatin structure. The three main areas of investigation were: (1) the importance of the limited excision repair in V79 cells for recovery, (2) the inducibility of recovery; (3) the activation of unused replication origins during recovery.

Cells

V79 Chinese hamster fibroblasts were kindly provided by Dr M. Taylor from Indiana University. A clone (C-l) was used throughout the experiments. Cells were routinely grown in Dulbecco’s modified Eagle’s medium, pH 7-0, supplemented with 10% foetal calf serum, 472 units ml-1 penicillin and 94 µg ml-1 streptomycin. The cells were kept in a 5% CO2- humidified atmosphere at 37°C. For irradiation, cultures were rinsed with phosphate-buffered saline (PBS), and exposed in this same solution to 254nm u.v. from a low-pressure mercury lamp at a dose rate of 0-5 J m-2s-1. The rates of semiconservative replication and RNA synthesis were determined by labelling cultures for 20-30 min with 5µCiml-1 of [1H]thymidine (dThd) (72Ci mmol-1) or 5µCi ml-1 [1H]uridine (27·1 Ci mmol). The cultures were rinsed twice with PBS and once with 5% trichloroacetic acid for 10 min at 5°C. After two more washings with trichloroacetic acid the cells were rinsed with 95 % ethanol and treated with 1 · 5 ml of 0 · 3 M-NaOH for 2 h at 37°C for lysis. Measurements of A260 and radioactivity in this solution gave the final values of DNA synthesis in cts min-1 unit-1 A260.

CsCl density gradient and neutral sucrose density gradient centrifugation

The methods described previously were followed (Ventura & Meneghini, 1984). These two techniques were used to measure the values of D and Mn, respectively, which were in turn used to determine the relative rate of elongation of double-stranded DNA. D represents the fraction of the BrdUrd-pulse-labelled DNA that remained within the density range of the unsubstituted DNA after CsCl centrifugation, and Mn is the number average molecular weight (Ventura & Meneghini, 1984).

Autoradiography and cell fusion

The cells were first synchronized in medium containing 2 mM-hydroxyurea for 14 h before releasing and pulse-labelling for 30min with 10 µ Ci ml-1 of [1H]dThd (cells D). They were then irradiated with 5Jm-2 of 254 nm u.v. or sham-irradiated, trypsinized and transferred to vials containing cells growing on slides (cells R). After 9 h the mixed cells were fused with polyethylene glycol (Pontecorvo et al. 1977) incubated for a further 3h and irradiated with 5 J m-2 or sham- irradiated. One hour later the cells were pulse-labelled with 0· 5µCi ml-1 of [1H]dThd for 30min after which they were washed, fixed and exposed to AR-10 Kodak stripping-film. After developing, the cells were stained with Toluidine Blue and grains in R cell nuclei were scored. Under these conditions D nuclei appeared heavily labelled and R nuclei contained from 40-90 grains. Separate experiments (not shown) indicated that the number of grains is proportional to the pulse-labelling time.

Nucleoid sedimentation

Cells were pre-labelled for 17h with 0·3µCiml-1 of [1H]dThd and 10-6M-dThd. After experimental treatment the cells were trypsinized and resuspended in 3 ml of PBS, centrifuged for 15 min at 600 g at 4°C and resuspended in 50 µl of PBS containing 10 mM-EDTA; 250 µ l of lysis solution containing 2 · 28M-NaCl, 24 mM-EDTA, 0· 6% Triton X-100 and 10mM-Tris-HCl, pH 8·0, was added to the cell suspension, the mixture was layered over 4 ·5 ml of a 15 % to 30 % sucrose gradient containing l-9M-NaCl, 10mM-EDTA, lOmM-Tris’HCl, pH 8-0, and left for 15 min before centrifugation at 20°C for 30 min at 10 000 rev. min-1 in a SW-50.1 Beckman rotor. Fractions (0-2 ml) were collected on strips of Whatman paper 17. After drying the paper strips were cut, placed in vials with 5-0 ml of PPO-POPOP-toluene and their radioactivity was counted in a scintillation spectrometer.

In u.v.-irradiated mammalian cells DNA synthesis is inhibited mainly because of the arrest of the replication fork by the pyrimidine dimer, u.v. irradiation also prevents initiation of DNA synthesis but this is mainly observed at very low doses (Kaufmann & Cleaver, 1981). Fig. 1 shows that in V79 cells the inhibition of [1H]dThd incorporation produced by a dose of 5 J m-2 is overcome by 7—lOh. This recovery cannot be attributed to the accumulation of cells in S phase by u.v. irradiation. Though there is a slight increase in the proportion of cells in.S’ phase from 4-6 h after u.v. (rising from 60-70 %), by 9-10h, when recovery has been fully attained, the percentage of u.v.-irradiated cells in S phase is the same as in the control population (results not shown). We have shown that recovery of [1H]dThd incorporation is linked to a recovery in the rate of elongation of double-stranded DNA, as measured by centrifugation of DNA in CsCl density gradients (Ventura & Meneghini, 1984). Thus, it seems appropriate to designate this phenomenon as recovery of DNA synthesis from inhibition by u.v. irradiation.

Fig. 1.

Recovery of DNA synthesis from inhibition by u.v. and effect of aphidicolin. Cells were irradiated with 5 J m-2 and incubated for the indicated periods of time with medium with (•) or without (○) 1 µ gml-1 of aphidicolin. The cells were further incubated for 1 h in normal medium and pulse-labelled with [1H]dThd for 20min. Bars indicate deviation for duplicate. ‘Control synthesis’ relates to incorporation of [1H]dThd in unirradiated cultures with or without aphidicolin as appropriate.

Fig. 1.

Recovery of DNA synthesis from inhibition by u.v. and effect of aphidicolin. Cells were irradiated with 5 J m-2 and incubated for the indicated periods of time with medium with (•) or without (○) 1 µ gml-1 of aphidicolin. The cells were further incubated for 1 h in normal medium and pulse-labelled with [1H]dThd for 20min. Bars indicate deviation for duplicate. ‘Control synthesis’ relates to incorporation of [1H]dThd in unirradiated cultures with or without aphidicolin as appropriate.

Recovery and excision repair

By 9h after 5 J m-2 only 23 % of the dimers have been excised, as measured by sites sensitive to the Microccus luteus u.v. endonuclease (results not shown). This result seems to make less likely any role for excision repair in the recovery of DNA synthesis in V79 cells. However, Griffiths & Ling (1985) showed that CHO cells that are completely defective in excision repair do not recover from inhibition by u.v. We decided therefore to determine whether the low level of repair exhibited by repair- competent V79 cells was important for the recovery. Aphidicolin, an inhibitor of DNA polymerase ex is an inhibitor of excision repair (Ciarrocchi et al. 1979). At a concentration of 1 µ g ml-1, aphidicolin reduced [1H]thymidine incorporation to less than 5 % of control levels in V79 cells, but had no effect on excision repair of dimers as determined by the nucleotide sedimentation technique. DNA strand breaks accumulate in u.v.-irradiated cells incubated with inhibitors of repair, and in preparations of nucleoids this is reflected in a lasting reduction in sedimentation rate under neutral conditions (Mattern et al. 1982). Fig. 2 shows that in exponentially growing V79 cells only high concentrations of aphidicolin (50 and 100 µ gmD1) effectively block DNA repair. We have found that upon removal of this drug, even at such high concentration, there occurs a reversal of replicative DNA synthesis inhibition to 80% of the control in 1 h (results not shown). Incubation with aphidicolin at a concentration of 1 qgmD1 for up to 10 h has no effect on the recovery of DNA synthesis (Fig. 1). However, at a concentration of l00 µgml-1 the drug prevented recovery (Fig. 3). These results argue in favour of the idea that recovery does not depend on continuous DNA synthesis, but does require normal excision repair. It is interesting to note that the inhibition of RNA synthesis by u.v. irradiation, a phenomenon that is only detected at relatively high u.v. doses, is also followed by recovery, but with faster kinetics than DNA synthesis recovery (Fig. 4). It is also clear that l00 µg ml-1 of aphidicolin prevents recovery of RNA synthesis. At 1 µg ml-1 aphidicolin has no effect on this recovery (results not shown). A similar prevention of RNA synthesis recovery was observed with 1 mM-arabinocytidine (araC), which is also (Fig. 2) an efficient inhibitor of excision repair. However, neither 1 mM-araC nor a mixture of 20 µ M-araC plus 2 mM-hydroxyurea, an effective combination of excision repair inhibitors (Cleaver, 1982), could be used to test for prevention of DNA synthesis recovery, since removal of these compounds did not bring a prompt reversal of DNA synthesis inhibition, as was the case for aphidicolin.

Fig. 2.

Relative sedimentation distance of nucleoids after u.v. irradiation and the effects of araC and aphidicolin (APH). The cells were irradiated with 20 J m-2 of u.v. and left for 30 min or 8 h before nucleoid preparation. In some samples aphidicolin or araC at the indicated concentrations was present during the 8-h period.

Fig. 2.

Relative sedimentation distance of nucleoids after u.v. irradiation and the effects of araC and aphidicolin (APH). The cells were irradiated with 20 J m-2 of u.v. and left for 30 min or 8 h before nucleoid preparation. In some samples aphidicolin or araC at the indicated concentrations was present during the 8-h period.

Fig. 3.

Effect of inhibition of excision repair on DNA synthesis recovery. The cells were irradiated with 5 J m-2 and incubated for the indicated times with (•) or without (○) 100 µgml-1 of aphidicolin. The cells were further incubated for Ih in normal medium and pulse-labelled with [1H]dThd for 30min. Bars indicate deviation for duplicate.

Fig. 3.

Effect of inhibition of excision repair on DNA synthesis recovery. The cells were irradiated with 5 J m-2 and incubated for the indicated times with (•) or without (○) 100 µgml-1 of aphidicolin. The cells were further incubated for Ih in normal medium and pulse-labelled with [1H]dThd for 30min. Bars indicate deviation for duplicate.

Fig. 4.

Effect of inhibition of excision repair on RNA synthesis recovery. Cells were incubated for 30 min with 100 Llg ml-1 of aphidicolin or normal medium, irradiated with 15 J m-2 and incubated for the indicated periods of time with (•) or without (○) 100 µ g ml-1 of aphidicolin. At the end of these periods they were pulse-labelled for 40 min with [1H]uridine. Bars represent deviation for duplicate.

Fig. 4.

Effect of inhibition of excision repair on RNA synthesis recovery. Cells were incubated for 30 min with 100 Llg ml-1 of aphidicolin or normal medium, irradiated with 15 J m-2 and incubated for the indicated periods of time with (•) or without (○) 100 µ g ml-1 of aphidicolin. At the end of these periods they were pulse-labelled for 40 min with [1H]uridine. Bars represent deviation for duplicate.

Inducibility of recovery

Using a split-dose protocol, D’Ambrosio & Setlow (1976) observed that two doses of u.v. separated by an interval had significantly less effect on the rate of DNA elongation than a single u.v. dose equivalent to the total of the split doses. We followed similar protocols to see whether the first dose has some effect on the kinetics of recovery after a second dose. Fig. 5 shows that this is not the case, the degree of inhibition and the kinetics of recovery being the same whether or not the cells had been previously irradiated. Similar responses were obtained when the first dose was diminished or the time between the two doses was reduced. These results of [1H]dThd incorporation are corroborated by measurements of the elongation of double-stranded DNA by means of CsCl density gradient centrifugation (Fig. 6). As previously reported (Ventura & Meneghini, 1984) a dose-dependent decrease in elongation is observed initially, followed by a recovery 8 h later. A second dose then brings the elongation rate down to a level similar to that observed after a single dose. These results suggest that the phenomenon of recovery is not inducible. However, we have now carried out experiments that suggest quite the contrary. In the experiment shown in Fig. 7 we irradiated the cells with 5 J m-2 and followed the recovery in the presence of actinomycin D at a concentration of 0·05 µgml-1. At this concentration the antibiotic has only a negligible effect on DNA synthesis while it reduces RNA synthesis to 40% of the control level. It is clear that the recovery phenomenon is strongly inhibited under these conditions. We found a similar effect by using 10-3 M-cycloheximide. However, this compound has a strong inhibitory effect on DNA synthesis, which is not reversed upon its removal. We used a different approach to test the inducibility of recovery. Cells, previously u.v. irradiated (D), were fused with unirradiated cells (R) and, after u.v. irradiation of the fused cells, the extent of DNA synthesis in the R nuclei was measured by quantitative autoradiography. In this experiment D cells were heavily prelabelled with [1H]dThd so as to distinguish their nuclei from those of the R cells, which were much more lightly labelled. The results in Fig. 8 show that, of the various experimental protocols, the only one resulting in a significant difference between the amounts of DNA synthesis in the R nuclei in R+D fused cells compared with unfused R cells was when both R and D were irradiated. In this case DNA synthesis was elevated in the fused cells. This is consistent with the hypothesis that factors produced in previously irradiated D cells are transferred to R nuclei, which, upon irradiation, recover faster from DNA synthesis inhibition.

Fig. 5.

Effect of a previous u.v. dose on the recovery of DNA synthesis after a second dose. Cells were previously irradiated with 0 (○) or 5 J m-2 (•) and 20 h later irradiated again with 5 J m-2. After the indicated times they were pulse-labelled for 30 min with [1H]dThd. Bars indicate deviation for duplicate

Fig. 5.

Effect of a previous u.v. dose on the recovery of DNA synthesis after a second dose. Cells were previously irradiated with 0 (○) or 5 J m-2 (•) and 20 h later irradiated again with 5 J m-2. After the indicated times they were pulse-labelled for 30 min with [1H]dThd. Bars indicate deviation for duplicate

Fig. 6.

Effect of a previous u.v. dose on the degree of inhibition of DNA elongation after a second dose. Cells were irradiated with the indicated doses and after the indicated times pulse-labelled for 1 h in medium containing 10-5M-BrdUrd, 39µ Ciml-1 of [1H]dThd (5×10-7M), 10-5M-uridine and 10-6M-FdUrd. In the sample represented by the rightmost bar, the cells were irradiated twice with 5 J m-2, separated by 8 h, and 1 h after the second dose they were incubated in the medium above. The DNAs were extracted and submitted to CsCl density gradient centrifugation or sucrose gradient centrifugation for determinations of D and Mn, respectively (Ventura & Meneghini, 1984). Mn/D for control cells is regarded as 100 %. The deviation for duplicate samples is indicated.

Fig. 6.

Effect of a previous u.v. dose on the degree of inhibition of DNA elongation after a second dose. Cells were irradiated with the indicated doses and after the indicated times pulse-labelled for 1 h in medium containing 10-5M-BrdUrd, 39µ Ciml-1 of [1H]dThd (5×10-7M), 10-5M-uridine and 10-6M-FdUrd. In the sample represented by the rightmost bar, the cells were irradiated twice with 5 J m-2, separated by 8 h, and 1 h after the second dose they were incubated in the medium above. The DNAs were extracted and submitted to CsCl density gradient centrifugation or sucrose gradient centrifugation for determinations of D and Mn, respectively (Ventura & Meneghini, 1984). Mn/D for control cells is regarded as 100 %. The deviation for duplicate samples is indicated.

Fig. 7.

Action of actinomycin D on DNA synthesis recovery. Cells were incubated for 30 min prior to irradiation with 5 J m-2 and for the indicated periods of time after irradiation in medium with (•) or without (○) 0 · 05 µ ml-1 of actinomycin D. At the end they were pulse-labelled for 30 min with [1H]dThd. Bars indicate deviation for duplicate.

Fig. 7.

Action of actinomycin D on DNA synthesis recovery. Cells were incubated for 30 min prior to irradiation with 5 J m-2 and for the indicated periods of time after irradiation in medium with (•) or without (○) 0 · 05 µ ml-1 of actinomycin D. At the end they were pulse-labelled for 30 min with [1H]dThd. Bars indicate deviation for duplicate.

Fig. 8.

Cell fusion to test for inducibility of the recovery phenomenon. Donor cells were irradiated with 0 (D) or 5 J m-2 (Du v) and 9h later fused to receptor cells (R). Three hours later the fused cells were irradiated with 0 (R) or 5 J m-2 (Ru.v.) and 1 h later they were pulse-labelled with [1H]dThd for 30 min. After developing, the autoradiography grains on the R cell nuclei were counted in both fused and isolated cells. The standard deviation of the mean of 50 nuclei is indicated.

Fig. 8.

Cell fusion to test for inducibility of the recovery phenomenon. Donor cells were irradiated with 0 (D) or 5 J m-2 (Du v) and 9h later fused to receptor cells (R). Three hours later the fused cells were irradiated with 0 (R) or 5 J m-2 (Ru.v.) and 1 h later they were pulse-labelled with [1H]dThd for 30 min. After developing, the autoradiography grains on the R cell nuclei were counted in both fused and isolated cells. The standard deviation of the mean of 50 nuclei is indicated.

The possible role of utilization of new replication origins in the recovery phenomenon

Park & Cleaver (1979) proposed that when replication forks are blocked by pyrimidine dimers the arrest of DNA replication could be relieved by chain growth from adjacent unblocked forks. According to this idea, initiation of replication at origins that are not normally utilized would permit a recovery from DNA synthesis inhibition. Taylor (1977) has shown that when Chinese hamster cells are held at the beginning of S phase by the thymydilate synthase inhibitor, fluorodeoxyuridine (FdUrd), new origins are activated and the inter-origin distance is diminished. We decided to submit cells to FdUrd treatment to see whether they became more resistant to u.v. irradiation, as might be expected. The experiment depicted in Fig. 9 shows that exposure to FdUrd for increasing time before irradiation does in fact bring about a resistance of DNA synthesis to inhibition by u.v. In this experiment the resistance was maximum after 6h of exposure of cells to FdUrd; at this time [1H]dThd incorporation in irradiated cells was 89% of that in unirradiated cells, whereas with no FdUrd the relative incorporation was only 51 %. After removal of FdUrd and before the [1H]dThd pulse-labelling the cells were exposed for 1 h to 10-5M-dThd in an attempt to restore the TTP pool, depleted by the FdUrd treatment. An alteration in [1H]dThd incorporation does not necessarily indicate altered DNA synthesis, especially when the TTP pool has been disturbed. However, it was confirmed that a real increase in DNA synthesis takes place on incubation with FdUrd in experiments in which the cells were pre-labelled for 26 h with [1H]dThd, treated with FdUrd and labelled with cold BrdUrd (Fig. 10). The area of the hybrid peak represents the real amount of template DNA that was replicated, whereas the shift represents the concentration of BrdUrd in the TTP pool. This latter parameter is the same in the four profiles, whereas DNA synthesis was more than doubled in FdUrd-treated cells. This is not due only to an accumulation of cells in N phase, which changed from 68 % in untreated cells to 91 % in treated cells. From the areas of the hybrid DNA peaks it can be calculated that the inhibition of DNA synthesis by u.v. was 28 % in untreated cells and 10 % in FdUrd-treated cells.

Fig. 9.

Effect of FdUrd on inhibition of DNA synthesis by u.v. The cells were incubated for the indicated times with 10-6M-FdUrd, irradiated with 0 (○) or 5 J m-2 (•) of u.v. incubated for 1 h in 10-:>M-dThd and pulse-labelled for 30 min with [1H]dThd.

Fig. 9.

Effect of FdUrd on inhibition of DNA synthesis by u.v. The cells were incubated for the indicated times with 10-6M-FdUrd, irradiated with 0 (○) or 5 J m-2 (•) of u.v. incubated for 1 h in 10-:>M-dThd and pulse-labelled for 30 min with [1H]dThd.

Fig. 10.

Stimulation of DNA synthesis by previous FdUrd treatment. The cells were labelled for 13 h with 0·2 µCimr1 [1H]dThd plus 10-6M-dThd. Culture was rinsed in PBS, and incubated for 6h in 10-6M-FdUrd. This medium was removed, the cells were irradiated with 5 J m-2 and incubated for 1 h in 10-5 M-dThd. At the end the medium was replaced by new medium containing 10-5 M-BrdUrd and after 2 h the DNA was extracted for DNA centrifugation in a CsCl density gradient. The fractions of radioactivity shifted to the hybrid position in relation to total counts were: A (control), 0 · 085; B (u.v.- irradiated, untreated), 0 · 062; C (treated with FdUrd, unirradiated), 0 · 190; D (treated with FdUrd, u.v.-irradiated), 0 · 171.

Fig. 10.

Stimulation of DNA synthesis by previous FdUrd treatment. The cells were labelled for 13 h with 0·2 µCimr1 [1H]dThd plus 10-6M-dThd. Culture was rinsed in PBS, and incubated for 6h in 10-6M-FdUrd. This medium was removed, the cells were irradiated with 5 J m-2 and incubated for 1 h in 10-5 M-dThd. At the end the medium was replaced by new medium containing 10-5 M-BrdUrd and after 2 h the DNA was extracted for DNA centrifugation in a CsCl density gradient. The fractions of radioactivity shifted to the hybrid position in relation to total counts were: A (control), 0 · 085; B (u.v.- irradiated, untreated), 0 · 062; C (treated with FdUrd, unirradiated), 0 · 190; D (treated with FdUrd, u.v.-irradiated), 0 · 171.

The resistance to inhibition by u.v. detected in the experiments of Figs 9, 10 was observed over a range of u.v. doses up to 12 J m-2. Moreover, the resistance disappeared after a prolonged period (18 h) in which the cells were kept in the presence of 10-5 M-dThd. We carried out the experiment shown in Fig. 11 to see whether the phenomenon of resistance can be ascribed to an increased rate of double-stranded DNA elongation. What the results in Fig. 11 seem to indicate is that in spite of a higher level of DNA synthesis in FdUrd-treated versus control cells, revealed in the experiments of Figs 9, 10, the overall rate of double-stranded DNA elongation in FdUrd-treated cells is no greater than in control cells. However, in u.v.-irradiated cells the situation is different; with FdUrd present there is a significantly higher rate of DNA elongation than in untreated cells, indicating a coupling of resistance to inhibition to faster DNA elongation in u.v.-irradiated cells.

Fig. 11.

Rate of elongation of double-stranded DNA after FdUrd treatment. The cells were exposed to 10-6M-FdUrd for 6h and then u.v.-irradiated with 0 or 5 J m-2 of u.v. There followed a 1-h incubation in 10-5 M-dThd after which the cells were pulse-labelled for l h with 10-5 M-BrdUrd, 10-6M-FdUrd, 10-5 M-uridine and 39 µCiml-1 of [1H]dThd (5×10-7M). The DNAs were extracted and submitted to CsCl density gradients centrifugation and sucrose density gradient centrifugation for determination of D and Mn, respectively. The deviation for duplicates is indicated.

Fig. 11.

Rate of elongation of double-stranded DNA after FdUrd treatment. The cells were exposed to 10-6M-FdUrd for 6h and then u.v.-irradiated with 0 or 5 J m-2 of u.v. There followed a 1-h incubation in 10-5 M-dThd after which the cells were pulse-labelled for l h with 10-5 M-BrdUrd, 10-6M-FdUrd, 10-5 M-uridine and 39 µCiml-1 of [1H]dThd (5×10-7M). The DNAs were extracted and submitted to CsCl density gradients centrifugation and sucrose density gradient centrifugation for determination of D and Mn, respectively. The deviation for duplicates is indicated.

Chinese hamster V79 cells are inefficient in excision of pyrimidine dimers compared with normal human fibroblasts. However, it has been shown that in transcriptionally active regions of the Chinese hamster genome the excision of dimers is very efficient (Bohr et al. 1985; Smith, 1987). In human cells excision of dimers plays an important role in recovery from DNA synthesis inhibition by u.v. (Moustacchi et al. 1979). More recently, Griffiths & Ling (1985) have shown that mutants of CHO cells, in which the excision of dimers is virtually non-existent, are defective in the recovery phenomenon. In this chapter we have described experiments designed to determine the effect of excision repair inhibition on recovery. Our results clearly show that at concentrations that inhibit repair aphidicolin is a strong inhibitor of recovery as well, while at a concentration that inhibits only DNA synthesis the drug has no effect on recovery. These data sustain the hypothesis that at least some type of excision repair is required for the recovery of DNA synthesis. In this connection it is noteworthy that the recovery of RNA synthesis from u.v. inhibition is also prevented by inhibition of excision repair (Fig. 4), in agreement with the data of Mayne (1984). Because excision repair seems to be particularly efficient in transcriptionally active regions (Bohr et al. 1985) we are tempted to suggest that excision of dimers from these regions enables DNA synthesis to occur, which in turn is required for the recovery of DNA synthesis. An alternative hypothesis suggests that excision repair is important because of the alterations of chromatin organization that are required for the recovery phenomenon.

Our finding that the recovery of DNA synthesis is not dependent on ongoing DNA synthesis requires further comment. We have used aphidicolin to inhibit DNA synthesis, and our results are in agreement with those of Lehmann et al. (1979), who have used FdUrd for the same purpose in human cells. Griffiths et al. (1981) concluded that continuous DNA synthesis is required for the recovery phenomenon to occur in Chinese hamster cells, using hydroxyurea as an inhibitor of semiconservative replication. We have obtained similar results using hydroxyurea but we have also noticed that this drug relaxed DNA supercoiling in nucleoids. This is not the case for aphidicolin and FdUrd, which in contrast bring about additional condensation of the nucleoids (Fig. 2; and unpublished results). We therefore ascribe the inhibition of recovery of hydroxyurea to a secondary effect on chromatin structure and conclude that recovery is not dependent on ongoing DNA synthesis. In fact, the prevention of recovery by 100 µ gml-1 of aphidicolin may be due to a persistence of unsealed breaks in u.v.-irradiated cells.

Our results with actinomycin D and with fused cells indicate that transcription of DNA is required for a recovery of DNA synthesis. That u.v. irradiation elicits synthesis of new proteins and RNA has been shown by others (Miskin & Ben-Ishai, 1981; Schorpp et al. 1984). We have noticed that in V79 cells the gene c-fos is activated after u.v. irradiation with kinetics similar to those of the recovery phenomenon (unpublished results). These results are consistent with the observation that inhibition of excision repair prevents recovery of RNA synthesis, which in turn inhibits recovery of DNA synthesis. The question then arises of why this faster recovery does not occur in the split-dose experiment described above when it is apparent in the fusion experiment. As yet, we have no answer to this question, but must suppose that a difference exists between the nuclei analysed in the fused cells, which carry a burden of dimers resulting from 5 J m-2, and the nuclei from cells submitted to a split-dose irradiation, which carry a burden of dimers resulting from two irradiations with 5Jm-2. Under split-dose conditions a larger amount of putative inducible factors may be required for recovery to occur. In fact, the kinetics of recovery from a single dose are much slower when the dose is increased (Dahle et al. 1980; and our unpublished results). In contrast, in human cells splitting the dose does enhance the recovery from inhibition produced by the second dose (Moustacchi et al. 1979).

The resistance of FdUrd pre-treated cells to DNA synthesis inhibition by u.v. is interesting and may shed some light on the recovery phenomenon. Brozmanova (1984) has observed a similar phenomenon in HeLa cells and has attributed it to a mechanism similar to that SOS mechanism described for bacteria. We prefer to think that it has to do with the induction of new origins of replication reported to occur in CHO cells exposed to FdUrd (Taylor, 1977). This would lead to a decrease in the average size of the replicon (Taylor, 1977) and, in fact, Cleaver et al. (1983) have reported that the initial inhibition of DNA synthesis after u.v. irradiation was dependent on replicon size in different cells, those with smaller replicons being more resistant to inhibition. This is the result expected if the arrest of a replication fork at a dimer is alleviated by another fork approaching from an adjacent origin.

According to the above model, a given fibre of DNA would replicate faster in FdUrd-treated cells, because the origins of replication are closer to each other. This has not been observed in unirradiated cells (Fig. 11). However, Taylor (1977) has reported that the activation of unused replicons by FdUrd in CHO cells was accompanied by a drop in the rate of fork movement, as if there were a compensation for the larger number of replicons. In the case of irradiated cells the effect of FdUrd pre-treatment is clearer (Fig. 11), probably because the activation of an unused replicon means the relief of a replication fork that otherwise would remain blocked at a dimer. The recovery phenomenon observed after u.v. irradiation may have a similar nature, reinitiation between blocked replication forks leading to recovery (Painter, 1985). Interestingly, nucleoids of cells that have been exposed either to u.v. or FdUrd a few hours earlier sediment faster than control nucleoids (our unpublished results). It is possible that a larger number of activated replicons causes more points of attachment of chromatin to the nuclear cage, rendering the nucleoids more condensed. It is attractive to think of reinitiation via usually silent origins as an SOS type of response in mammalian cells, requiring the synthesis of specific factors that are necessary as catalysts of DNA synthesis or as inducers of modification in the chromatin structure, for instance creating more points of attachment of chromatin and nuclear matrix. Excision-deficient Chinese hamster cells may be incapable of recovery either because even the low level of excision in wild-type cells is important for synthesis of the required RNA, or because it facilitates the attachment of the chromatin to the nuclear matrix.

We thank the Natural Product Branch of the N.C.L for making aphidicolin available to us. This work was supported by grants from FAPESP, FINEP and CNPq, Brazilian foundations of scientific support. A.M.V. and J.M.O. hold fellowships from FAPESP.

BOCKSTAHLER
,
L. E.
&
Lytle
,
D. D.
(
1970
).
UV light enhanced reactivation of a mammalian virus
.
Biochem. biophys. Res. Commun. 4L, 184-189
.
Bohr
,
V. A.
,
Smith
,
C. A.
,
Okumoto
,
D. S.
&
Hanawalt
,
P. C.
(
1985
).
DNA repair in an active gene: removal of pyrimidine dimers from the DHFR gene of CHO cells is much more efficient than in the genome overall
.
Cell
40
,
359
369
.
Brozmanova
,
J.
(
1984
).
Stimulation of postirradiation DNA synthesis in ultraviolet irradiated HeLa cells by fluorodeoxyuridine
.
Neoplasma
31
,
169
173
.
Ciarrochi
,
G.
,
Jose
,
J. G.
&
Linn
,
S.
(
1979
).
Further characterization of a cell-free system for measuring replicative and repair synthesis with cultured fibroblasts and evidence for the involvement of polymerase or in DNA repair
.
Nucl. Acids Res.
7
,
1205
1219
.
Cleaver
,
J. E.
(
1982
).
Normal reconstruction of DNA supercoiling and chromatin structure in Cockayne syndrome during repair of damage from ultraviolet light
.
Am. J. hum. Gen.
34
,
566
575
.
Cleaver
,
J. E.
,
Kaufmann
,
W. K.
,
Kapp
,
L. N.
&
Park
,
S. D.
(
1983
).
Replication size and excision repair as factors in the inhibition and recovery of DNA synthesis from UV damage
.
Biochim. biophys. Acta T&), 207-215
.
Collins
,
A.
&
Waldren
,
C.
(
1982
).
Cell cycle kinetics and UV light survival in UV-1, a CHO mutant defective in post replication recovery..?
.
Cell Sci.
57
,
265
275
.
Cordeiro
,
M.
&
Meneghini
,
R.
(
1973
).
The rate of DNA replication in the polytene chromosome of Rhynchosciara angelae.
J. tnolec. Biol.
78
,
261
274
.
Cordeiro-Stone
,
M.
,
Schumacher
,
R. I.
&
Meneghini
,
R.
(
1979
).
Structure of replication fork in UV light irradiated human cells
.
Biophys. J.
27
,
287
300
.
Dahle
,
D.
,
Griffiths
,
T. D.
&
Carpenter
,
J. G.
(
1980
).
Inhibition and recovery of DNA synthesis in UV-irradiated Chinese hamster V-79 cells
.
Photochem. Photobiol.
32
,
157
165
.
D’Ambrosio
,
S. M.
&
Setlow
,
R. B.
(
1976
).
Enhancement of post-replication repair in Chinese hamster cells
.
Proc. natn. Acad. Sci. U.SA.
73
,
2396
2400
.
Domon
,
M.
&
Rauth
,
A. M.
(
1973
).
Cell cycle specific recovery from fractionated exposures to UV light
.
Radiat. Res.
55
,
81
92
.
DONIGER
,
J.
(
1978
).
DNA replication in UV light irradiated Chinese hamster cells: the nature of replicon inhibition and post-replication repair. J’, molec. Biol.
120
,
433
446
.
Edenberg
,
H. J.
(
1976
).
Inhibition of DNA replication by UV light
.
Biophys. J.
16
,
849
859
.
Griffiths
,
T. D.
,
Dahle
,
D. B.
,
Meechan
,
P. J.
&
Carpenter
,
J. G.
(
1981
).
Effects of inhibitors of DNA synthesis on the rate of DNA synthesis after exposure of mammalian cells to UV light
.
Biochim. biophys. Acta
656
,
55
61
.
Griffiths
,
T. D.
&
Ling
,
S. Y.
(
1985
).
Effect of ultraviolet light on thymidine incorporation, DNA chain elongation and replicon initiation in wild-type and excision deficient Chinese hamster ovary cells
.
Biochim. biophys. Acta
826
,
121
128
.
KAUFMANN
,
W. K.
&
Cleaver
,
J. E.
(
1981
).
Mechanisms of inhibition of DNA replication by ultraviolet light in normal human and xeroderma pigmentosum fibroblasts
.
J. molec. Biol.
149
,
175
187
.
Lehmann
,
A. R.
(
1972
).
Postreplication repair of DNA in UV-irradiated mammalian cells
.
J. molec. Biol.
66
,
319
337
.
Lehmann
,
A. R.
&
Kirk-Bell
,
S.
(
1972
).
Postreplication repair of DNA in UV-irradiated mammalian cells. No gaps in DNA synthesized late after UV-irradiation
.
Eur.J. Biochem.
31
,
438
445
.
Lehmann
,
A. R.
,
Kirk-Bell
,
S.
&
Mayne
,
L.
(
1979
).
Abnormal kinetics of DNA synthesis in UV-irradiated cells from patients with Cockayne’s syndrome
.
Cancer Res.
39
,
4237
4241
.
Mattern
,
M. R.
,
Paone
,
R. F.
& Day,
R. S.
Ill
(
1982
).
Eukaryotic DNA repair is blocked at different steps by inhibitors of DNA topoisomerases and of DNA polymerases or and (j. Biochim. biophys. Acta
697
,
6
13
.
Mayne
,
L. V.
(
1984
).
Inhibitors of DNA synthesis (aphidicolin and cytosine arabinoside/hydroxyurea) prevent the recovery of RNA synthesis after UV-irradiation
.
Mutat. Res.
131
,
187
191
.
Menck
,
C. F. M.
&
Meneghini
,
R.
(
1982
).
Recovery in survival capacity of UV-irradiated 3T3 mouse cells at Go cannot be solely dependent on the excision of pyridine dimers
.
Mutat. Res.
96
,
273
280
.
Meneghini
,
R.
,
Menck
,
C. F. M.
&
Schumacher
,
R. I.
(
1981
).
Mechanisms of tolerance to DNA lesions in mammalian cells
.
Q. Rev. Biophys.
14
,
381
432
.
Meneghini
,
R.
&
Mello-Filho
,
A. C.
(
1983
).
Rate of DNA synthesis in mammalian cells irradiated with ultraviolet light: a model based on the variations in the rate of movement of the replication fork and in the number of active replicons
.
J. theor. Biol.
100
,
359
372
.
Meyn
,
R. E.
&
Humphrey
,
R. M.
(
1971
).
DNA synthesis in UV-light irradiated Chinese hamster cells. Biophys. J.
11
,
295
301
.
MlSKIN
,
R.
&
Ben-Ishai
,
R.
(
1981
).
Induction of plasminogen activator by UV light in normal and xeroderma pigmentosum fibroblasts
.
Proc. natn. Acad. Sci. U.S.A.
78
,
6236
6240
.
Moustacchi
,
E.
,
Ehmann
,
U. K.
&
Friedberg
,
E. C.
(
1979
).
Defective recovery of semiconservative DNA synthesis in XP cells following split dose UV-irradiation
.
Mutat. Res.
62
,
159
171
.
Painter
,
R. B.
(
1978
).
Does UV light enhance postreplication repair in mammalian cells?
Nature, Lond.
275
,
243
245
.
Painter
,
R. B.
(
1985
).
Inhibition and recovery of DNA synthesis in human cells after exposure to ultraviolet light
.
Mutat. Res.
145
,
63
69
.
Park
,
S. D.
&
Cleaver
,
J. E.
(
1979
).
Post-replication repair: questions of its definition and possible alteration in XP cells strains
.
Proc. natn. Acad. Sci. U.S A.
76
,
3927
3931
.
Pontecorvo
,
G.
,
Riddle
,
P. N.
&
Hales
,
A.
(
1977
).
Time and mode of fusion of human fibroblasts treated with polyethylene glycol
.
Nature, Lond.
265
,
257
258
.
SARASIN
,
A. R.
&
Hana WALT
,
P.
C
. (
1978
).
Carcinogens enhance survival of UV-irradiated simian virus 40 in treated monkey kidney cells: induction of a recovery pathway
.
Proc. natn. Acad. Sci. U.S.A.
75
,
346
350
.
Schorpp
,
M.
,
Mallick
,
U.
,
Rahmsdorf
,
H. J.
&
Herllich
,
P.
(
1984
).
UV-induced extracellular factor from human fibroblasts communicates the UV response to nonirradiated cells
.
Cell
37
,
861
868
.
Smith
,
C. A.
(
1987
).
DNA repair in specific sequences in mammalian cells. J’
.
Cell Sci. Suppl
.
6
,
225
-
241
.
Taylor
,
J. H.
(
1977
).
Increase in DNA replication sites in cells at the beginning of S phase
.
Chromosoma
62
,
291
300
.
Todd
,
P.
(
1973
).
Fractionated UV light irradiation of cultured Chinese hamster cells
.
Radiat. Res.
55
,
93
100
.
Ventura
,
A. M.
&
Meneghini
,
R.
(
1984
).
Inhibition and recovery of the rate of DNA synthesis in V79 Chinese hamster cells following ultraviolet light irradiation
.
Mutat. Res.
131
,
81
88
.