Significance of dimers to the size of newly synthesized DNA in UV-irradiated Chinese hamster ovary cells.

DNA synthesized after UV irradiation is smaller than that in unirradiated cells even when pulse-labeling times are increased to compensate for the overall reduction in the rate of DNA replication. By isolating newly replicated DNA, incubating it with dimer-specific endonuclease from Micrococcus luteus, and analyzing it on alkaline sucrose gradients, we have been able to demonstrate that this DNA is synthesized in segments corresponding in size to the interdimer distance on the parental strand. In addition, the same DNA analyzed on neutral gradients shows no reduction in molecular weight as a result of UV irradiation and/or endonuclease digestion. Our data are thus inconsistent with the presence of "gaps" in newly synthesized DNA opposite the dimers on the parental strand. We suggest that if such gaps are produced as a result of delayed synthesis around dimers, they are filled before the growing point reaches the next dimer.     

Translesion synthesis is the main component of SOS repair in bacteriophage lambda DNA.

Agents that interfere with DNA replication in Escherichia coli induce physiological adaptations that increase the probability of survival after DNA damage and the frequency of mutants among the survivors (the SOS response). Such agents also increase the survival rate and mutation frequency of irradiated bacteriophage after infection of treated bacteria, a phenomenon known as Weigle reactivation. In UV-irradiated single-stranded DNA phage, Weigle reactivation is thought to occur via induced, error-prone replication through template lesions (translesion synthesis [P. Caillet-Fauquet, M: Defais, and M. Radman, J. Mol. Biol. 117:95-112, 1977]). Weigle reactivation occurs with higher efficiency in double-stranded DNA phages such as lambda, and we therefore asked if another process, recombination between partially replicated daughter molecules, plays a major role in this case. To distinguish between translesion synthesis and recombinational repair, we studied the early replication of UV-irradiated bacteriophage lambda in SOS-induced and uninduced bacteria. To avoid complications arising from excision of UV lesions, we used bacterial uvrA mutants, in which such excision does not occur. Our evidence suggests that translesion synthesis is the primary component of Weigle reactivation of lambda phage in the absence of excision repair. The greater efficiency in Weigle reactivation of double-stranded DNA phage could thus be attributed to some inducible excision repair unable to occur on single-stranded DNA. In addition, after irradiation, lambda phage replication seems to switch prematurely from the theta mode to the rolling circle mode.     

Perturbations in simian virus 40 DNA synthesis by ultraviolet light.

Perturbations of Simian Virus 40 (SV40) DNA replication by ultraviolet (UV) light during the lytic cycle in permissive monkey CV-1 cells resemble those seen in host cell DNA replication. Formation of Form I DNA molecules (i.e. completion of SV40 DNA synthesis) was more sensitive to UV irradiation than synthesis of replicative intermediates or Form II molecules, consistent with inhibition of DNA chain elongation. The observed amounts of [3H]thymidine incorporated in UV-irradiated molecules could be predicted on the assumption that pyrimidine dimers are responsible for blocking nascent DNA strand growth. The relative proportion of labeled Form I molecules in UV-irradiated cultures rapidly increased to near-control values with incubation after 20 or 40 J/m2 of light (0.9--1.0 or 1.8--2.0 dimers per SV40 genome, respectively). This rapid increase and the failure of Form II molecules to accumulate suggest that SV40 growing forks can rapidly bypass many dimers. Form II molecules formed after UV irradiation were not converted to linear (Form III) molecules by the dimer-specific T4 endonuclease V, suggesting either that there are no gaps opposite dimers in these molecules or that T4 endonuclease V cannot use Form II molecules as substrates.     

Cell cycle-dependent strand bias for UV-induced mutations in the transcribed strand of excision repair-proficient human fibroblasts but not in repair-deficient cells.

To study the effect of nucleotide excision repair on the spectrum of mutations induced in diploid human fibroblasts by UV light (wavelength, 254 nm), we synchronized repair-proficient cells and irradiated them when the HPRT gene was about to be replicated (early S phase) so that there would be no time for repair in that gene before replication, or in G1 phase 6 h prior to S, and determined the kinds and location of mutations in that gene. As a control, we also compared the spectra of mutations induced in synchronized populations of xeroderma pigmentosum cells (XP12BE cells, which are unable to excise UV-induced DNA damage). Among the 84 mutants sequenced, base substitutions predominated. Of the XP mutants from S or G1 and the repair-proficient mutants from S, approximately 62% were G.C----A.T. In the repair-proficient mutants from G1, 47% were. In mutants from the repair-proficient cells irradiated in S, 71% (10 of 14) of the premutagenic lesions were located in the transcribed strand; with mutants from such cells irradiated in G1, only 20% (3 of 15) were. In contrast, there was no statistically significant difference in the fraction of premutagenic lesions located in the transcribed strand of the XP12BE cells; approximately 75% (24 of 32) of the premutagenic lesions were located in that strand, i.e., 15 of 19 (79%) in the S-phase cells and 9 of 13 (69%) in the G1-phase cells. The switch in strand bias supports preferential nucleotide excision repair of UV-induced damage in the transcribed strand of the HPRT gene.     

Structure of Replicating Simian Virus 40 Deoxyribonucleic Acid Molecules

Properties of replicating simian virus 40 (SV40) deoxyribonucleic acid (DNA) have been examined by sedimentation analysis and by direct observation during a lytic cycle of infection of African green monkey kidney cells. Two types of replicating DNA molecules were observed in the electron microscope. One was an open structure containing two branch points, three branches, and no free ends whose length measurements were consistent with those expected for replicating SV40 DNA molecules. A second species had the same features as the open structure, but in addition it contained a superhelix in the unreplicated portion of the molecule. Eighty to ninety per cent of the replicative intermediates (RI) were in this latter configuration, and length measurements of these molecules also were consistent with replicating SV40 DNA. Replicating DNA molecules with this configuration have not been described previously. RI, when examined in ethidium bromide-cesium chloride (EB-CsCl) isopycnic gradients, banded in a heterogeneous manner. A fraction of the RI banded at the same density as circular SV40 DNA containing one or more single-strand nicks (component II). The remaining radioactive RI banded at densities higher than that of component II, and material was present at all densities between that of supercoiled double-stranded DNA (component I) and component II. When RI that banded at different densities in EB-CsCl were examined in alkaline gradients, cosedimentation of parental DNA and newly replicated DNA did not occur. All newly replicated DNA sedimented more slowly than did intact single-stranded SV40 DNA, a finding that is inconsistent with the rolling circle model of DNA replication. An inverse correlation exists between the extent of replication of the SV40 DNA and the banding density in EB-CsCl. Under alkaline conditions, the parental DNA strands that were contained in the RI sedimented as covalently closed structures. The sedimentation rates in alkali of the covalently closed parental DNA decreased as replication progressed. Based on these observations, some possible models for replication of SV40 DNA are proposed.     

Gaps and forks in DNA replication: Rediscovering old models

Most current models for replication past damaged lesions envisage that translesion synthesis occurs at the replication fork. However older models suggested that gaps were left opposite lesions to allow the replication fork to proceed, and these gaps were subsequently sealed behind the replication fork. Two recent articles lend support to the idea that bypass of the damage occurs behind the fork. In the first paper, electron micrographs of DNA replicated in UV-irradiated yeast cells show regions of single-stranded DNA both at the replication forks and behind the fork, the latter being consistent with the presence of gaps in the daughter-strands opposite lesions. The second paper describes an in vitro DNA replication system reconstituted from purified bacterial proteins. Repriming of synthesis downstream from a blocked fork occurred not only on the lagging strand as expected, but also on the leading strand, demonstrating that contrary to widely accepted beliefs, leading strand synthesis does not need to be continuous.     

Pyrimidine dimers block simian virus 40 replication forks.

UV light produces lesions, predominantly pyrimidine dimers, which inhibit DNA replication in mammalian cells. The mechanism of inhibition is controversial: is synthesis of a daughter strand halted at a lesion while the replication fork moves on and reinitiates downstream, or is fork progression itself blocked for some time at the site of a lesion? We directly addressed this question by using electron microscopy to examine the distances of replication forks from the origin in unirradiated and UV-irradiated simian virus 40 chromosomes. If UV lesions block replication fork progression, the forks should be asymmetrically located in a large fraction of the irradiated molecules; if replication forks move rapidly past lesions, the forks should be symmetrically located. A large fraction of the simian virus 40 replication forks in irradiated molecules were asymmetrically located, demonstrating that UV lesions present at the frequency of pyrimidine dimers block replication forks. As a mechanism for this fork blockage, we propose that polymerization of the leading strand makes a significant contribution to the energetics of fork movement, so any lesion in the template for the leading strand which blocks polymerization should also block fork movement.     

Multiple mechanisms control chromosome integrity after replication fork uncoupling and restart at irreparable UV lesions

DNA replication forks pause in front of lesions on the template, eventually leading to cytotoxic chromosomal rearrangements. The in vivo structure of damaged eukaryotic replication intermediates has been so far elusive. Combining electron microscopy (EM) and two-dimensional (2D) gel electrophoresis, we found that UV-irradiated S. cerevisiae cells uncouple leading and lagging strand replication at irreparable UV lesions, thus generating long ssDNA regions on one side of the fork. Furthermore, small ssDNA gaps accumulate along replicated duplexes, likely resulting from repriming events downstream of the lesions on both leading and lagging strands. Translesion synthesis and homologous recombination counteract gap accumulation, without affecting fork progression. The DNA damage checkpoint contributes to gap repair and maintains a replication-competent fork structure. We propose that the coordinated action of checkpoint, recombination, and translesion synthesis-mediated processes at the fork and behind the fork preserves the integrity of replicating chromosomes by allowing efficient replication restart and filling the resulting ssDNA gaps.     

Analysis of single-stranded DNA stability and damage-induced strand loss in mammalian cells using SV40-based shuttle vectors

The fate and stability of fully or partially single-stranded DNA molecules transfected into mammalian cells have been analysed. For this, we constructed a simian virus 40 (SV40)-based shuttle vector containing the f1 bacteriophage replication origin in the two possible orientations (pi SVF1-A and pi SVF1-B). This vector contains the SV40 origin of replication, the late viral genes and DNA sequences for replication and selection in Escherichia coli. It also carries the lacO sequence, which permits the analysis of plasmid stability. Single-stranded DNA from pi SVF1-A and pi SVF1-B were produced in bacteria and annealed in vitro to form a heteroduplex molecule. We showed that, in monkey kidney COS7 cells, single-stranded vectors replicate to form duplex molecules. After transfection of the three forms of molecules (single-stranded, heteroduplex or double-stranded), replicated DNA was rescued in E. coli. Vector stability was analysed by checking for plasmid rearrangements and screening for lacO mutants. The single-stranded pi SVF1 has a lower rearrangement level, while the spontaneous mutation frequency (on the lacO target) is in the same range as for the double-stranded vector. In contrast, the level of spontaneous mutagenesis is higher for the heteroduplex than for the single- and double-stranded forms. In addition, we found that replication of heteroduplex with one strand containing ultraviolet light-induced lesions yields progeny molecules in which the irradiated strand is mostly lost. This result indicates for the first time the specific loss of the damaged strand in mammalian cells.     

recA-dependent DNA repair processes

UV-radiation-induced lesions in DNA result in the formation of: (1) excision gaps (i.e. a lesion is excised, leaving a gap), (2) daughter-strand gaps (i.e. a lesion can be skipped during replication, leaving a gap), and (3) double-strand breaks (i.e. the DNA strand opposite a gap can be cut). In Escherichia coli, the recA gene product is involved in repairs of all three types of lesions--repair of daughter-strand gaps (2) and double-strand breaks (3) constitutes post-replication repair. The evidence suggests, furthermore, that recA-dependent repair of excision gaps (1) produced in DNA replicated prior to UV irradiation (pre-replication repair) appears to occur by similar mechanisms.     

Effects of UV irradiation on the fate of 5-bromodeoxyuridine-substituted bacteriophage T4 DNA.

We have carried out a series of experiments designed to characterize the impact of UV irradiation (260 nm) on 5-bromodeoxyuridine-labeled (heavy) T4 bacteriophage, both before and after infection of Escherichia coli. In many respects, these effects differ greatly from those previously described for non-density-labeled (light) phage. Moreover, our results have led us to propose a model for a novel mechanism of host-mediated repair synthesis, in which excision of UV-damaged areas is followed by initiation of replication, strand displacement, and a considerable amount of DNA replication. UV irradiation of 5-bromodeoxyuridine-labeled phage results in single-stranded breaks in a linear, dose-dependent manner (1.3 to 1.5 breaks per genomic strand per lethal hit). This damage does not interfere with injection of the phage genome, but some of the UV-irradiated heavy phage DNA undergoes additional intracellular breakdown (also dose dependent). However, a minority (25%) of the injected parental DNA is protected, maintaining its preinjection size. This protected moiety is associated with a replicative complex of DNA and proteins, and is more efficiently replicated than is the parental DNA not so associated. Most of the progeny DNA is also found with the replicative complex. The 5-bromodeoxyuridine of heavy phage DNA is debrominated by UV irradiation, resulting in uracil which is removed by host uracil glycosylase. Unlike the simple gap-filling repair synthesis after infection with UV-irradiated light phage, the repair replication of UV-irradiated heavy phage is extensive as determined by density shift of the parental label in CsC1 gradients. The newly synthesized segments are covalently attached to the parental fragments. The repair replication takes place even in the presence of chloramphenicol, a protein synthesis inhibitor, suggesting it is host mediated. Furthermore, the extent of the repair replication is greater at higher doses of UV irradiation applied to the heavy phage. This abundant synthesis results ultimately in dispersion of the parental sequences as short stretches in the midst of long segments of newly synthesized progeny DNA. Together, the extensive replication and the resulting distribution pattern of parental sequences, without significant solubilization of parental label, are most consistent with a model of repair synthesis in which the leading strand displaces, rather than ligates to, the encountered 5' end.     

Replicative intermediates in UV-irradiated simian virus 40

We have used Simian virus 40 (SV40) as a probe to study the replication of UV-damaged DNA in mammalian cells. Viral DNA replication in infected monkey kidney cells was synchronized by incubating a mutant of SV40 (tsA58) temperature-sensitive for the initiation of DNA synthesis at the restrictive temperature and then adding aphidicolin to temporarily inhibit DNA synthesis at the permissive temperature while permitting pre-replicative events to occur. After removal of the drug, the infected cells were irradiated at 100 J/m2 (254 nm) to produce 6-7 pyrimidine dimers per SV40 genome, and returned to the restrictive temperature to prevent reinitiation of replication from the SV40 origin. Replicative intermediates (RI) were labeled with [3H]thymidine, and isolated by centrifugation in CsCl/ethidium bromide gradients followed by BND-cellulose chromatography. The size distribution of daughter DNA strands in RI isolated shortly after irradiation was skewed towards lengths less than the interdimer spacing in parental DNA; this bias persisted for at least 1 h after irradiation, but disappeared within 3 h, by which time the size of the newly-synthesized DNA exceeded the interdimer distance. No significant excision of dimers from parental strands in either replicative intermediates or Form I (closed circular) DNA molecules was detected. These data are consistent with the hypothesis that replication forks are temporarily blocked by dimers encountered on the leading strand side of the fork, but that daughter strand continuity opposite dimers is eventually established. Evidence was obtained for the generation at late times after irradiation, of Form I molecules in which the daughter DNA strands contain dimers. Thus DNA strand exchange as well as trans-dimer synthesis may be involved in the generation of supercoiled Form I DNA from UV-damaged SV40 replicative intermediates.     

Inhibition and recovery of DNA synthesis in human cells after exposure to ultraviolet light

The inhibition of DNA synthesis in normal human cells by UV is a complex function of fluence because it has several causes. At low fluences, inhibition of replicon initiation is most important. This is made clear by the fact that it occurs to a lesser degree in cells from patients with ataxia telangiectasia (AT). Assuming that only leading strand synthesis is blocked by UV-induced lesions, single lesions between replicons in parental strands for leading strand synthesis inhibit DNA synthesis by acting as temporary blocks until they are replicated by extension of the lagging strand of the adjacent replicon. A more severe inhibition occurs when two lesions are induced between adjacent growing replicons, because one in four possible configurations may result in a long-lived unreplicated region (LLUR). In the absence of excision repair, these may eventually be replicated by activation of an otherwise unused origin within the LLUR. The frequency of LLURs increases steeply with fluence. Activation of normally unused origins to replicate LLURs may facilitate recovery from inhibition of DNA synthesis, but repair of lesions is probably more important. In excision-repair-defective cells, an LLUR without an origin to initiate its replication may be a lethal lesion.     

Nature of the Complementary Strands Synthesized in Vitro upon the Single-Stranded Circular DNA of Bacteriophage øX174 after Ultraviolet Irradiation

This paper describes experiments intended to decide whether UV lesions in DNA act as absolute blocks to chain elongation by the Escherichia coli DNA polymerase or only slow down the polymerization process. Ultraviolet (UV)-irradiated, single-stranded (SS) circular DNA of bacteriophage øX174 was used as template for the polymerase in a reaction mixture in vitro, under conditions allowing synthesis of not more than one complementary strand per template molecule. The mean length of the newly synthesized complementary strands (as determined by velocity sedimentation in alkaline CsCl gradients), as well as the over-all template activity (as measured by deoxyadenosine monophosphate [dAMP] incorporation) was found to decrease with the number of biologically lethal hits sustained by the irradiated templates. With the increase of time or temperature of reaction, the net synthesis of complementary strands increased (as a consequence of increased initiation), but their mean length remained constant. The mean length of synthesized strands was greater than would be expected if all biologically lethal hits were to block the polymerization process. The lethal hits which serve as blocking lesions are inferred to be pyrimidine dimers because it is possible to obtain synthesis of full-length complementary strands if, when heat-denatured, UV-irradiated, double-stranded replicative form (RF II) DNA of bacteriophage øX174 is used as a template, it is pretreated with yeast photoreactivating enzyme (YPRE) in presence of visible light.     

Recombination between Homologous Chromosomes Induced by Unrepaired UV-Generated DNA Damage Requires Mus81p and Is Suppressed by Mms2p

DNA lesions caused by UV radiation are highly recombinogenic. In wild-type cells, the recombinogenic effect of UV partially reflects the processing of UV-induced pyrimidine dimers into DNA gaps or breaks by the enzymes of the nucleotide excision repair (NER) pathway. In this study, we show that unprocessed pyrimidine dimers also potently induce recombination between homologs. In NER-deficient rad14 diploid strains, we demonstrate that unexcised pyrimidine dimers stimulate crossovers, noncrossovers, and break-induced replication events. The same dose of UV is about six-fold more recombinogenic in a repair-deficient strain than in a repair-proficient strain. We also examined the roles of several genes involved in the processing of UV-induced damage in NER-deficient cells. We found that the resolvase Mus81p is required for most of the UV-induced inter-homolog recombination events. This requirement likely reflects the Mus81p-associated cleavage of dimer-blocked replication forks. The error-free post-replication repair pathway mediated by Mms2p suppresses dimer-induced recombination between homologs, possibly by channeling replication-blocking lesions into recombination between sister chromatids.     

Mechanisms for recF-dependent and recB-dependent pathways of postreplication repair in UV-irradiated Escherichia coli uvrB.

The molecular mechanisms for the recF-dependent and recB-dependent pathways of postreplication repair were studied by sedimentation analysis of DNA from UV-irradiated Escherichia coli cells. When the ability to repair DNA daughter strand gaps was compared, uvrB recF cells showed a gross deficiency, whereas uvrB recB cells showed only a small deficiency. Nevertheless, the uvrB recF cells were able to perform some limited repair of daughter strand gaps compared with a "repairless" uvrB recA strain. The introduction of a recB mutation into the uvrB recF strain greatly increased its UV radiation sensitivity, yet decreased only slightly its ability to repair daughter strand gaps. Kinetic studies of DNA repair with alkaline and neutral sucrose gradients indicated that the accumulation of unrepaired daughter strand gaps led to the formation of low-molecular-weight DNA duplexes (i.e., DNA double-strand breaks were formed). The uvrB recF cells were able to regenerate high-molecular-weight DNA from these low-molecular-weight DNA duplexes, whereas the uvrB recF recB and uvrB recA cells were not. A model for the recB-dependent pathway of postreplication repair is presented.     

Inhibition of repair of UV-damaged DNA by caffeine and mutation induction in Chinese hamster cells

The effect of caffeine on UV-irradiated Chinese hamster cells in vitro was studied on the cellular and molecular levels. Caffeine (1 mM) was shown to decrease the colony-forming ability and the frequencies of spontaneous and UV-induced mutations in Chinese hamster cells. The effect of caffeine in reducing the frequency of UV-induced mutations was demonstrated only if caffeine was present in the culture medium during the first post-irradiation cell division. Using alkaline sucrose gradient centrifugation, both parental and newly synthesized DNA in UV-irradiated and unirradiated cells were studied in the presence and absence of caffeine. Caffeine affected the sedimentation profile of DNA synthesized in UV-irradiated cells but not in unirradiated cells. Caffeine had no apparent effect on the incorporation of [³H]-thymidine into DNA of control or UV-irradiated cells, nor on the small amount of excision of UV-induced pyrimidine dimers. These results may be interpreted by a hypothesis that caffeine inhibits a certain S-phase specific, post-replication, dark-repair mechanism. The hamster and perhaps other rodent cells exposed to low doses of UV are capable of DNA replication, by-passing the non-excised pyrimidine dimers. This postulated repair process probably involves de novo DNA synthesis to seal the gaps in the nascent strand. This repair may be also responsible for the enzymatic production of mutations.     

Rapid and apparently error-prone excision repair of nonreplicating UV-irradiated plasmids in Xenopus laevis oocytes.

Repair of UV-irradiated plasmid DNA microinjected into frog oocytes was measured by two techniques: transformation of repair-deficient (delta uvrB delta recA delta phr) bacteria, and removal of UV endonuclease-sensitive sites (ESS). Transformation efficiencies relative to unirradiated plasmids were used to estimate the number of lethal lesions; the latter were assumed to be Poisson distributed. These estimates were in good agreement with measurements of ESS. By both criteria, plasmid DNA was efficiently repaired, mostly during the first 2 h, when as many as 2 x 10(10) lethal lesions were removed per oocyte. This rate is about 10(6) times the average for removal of ESS from repair-proficient human cells. Repair was slower but still significant after 2 h, but some lethal lesions usually remained after overnight incubation. Most repair occurred in the absence of light, in marked contrast to differentiated frog cells, previously shown to possess photoreactivating but no excision repair activity. There was no increase in the resistance to DpnI restriction of plasmids (methylated in Escherichia coli at GATC sites) incubated in oocytes; this implies no increase in hemimethylated GATC sites, and hence no semiconservative DNA replication. Plasmid substrates capable of either intramolecular or intermolecular homologous recombination were not recombined, whether UV-irradiated or not. Repair of Lac+ plasmids was accompanied by a significant UV-dependent increase in the frequency of Lac- mutants, corresponding to a repair synthesis error frequency on the order of 10(-4) per nucleotide.     

Transformation depending on intermolecular homologous recombination is stimulated by UV damage in transfected DNA

DNA-mediated gene transfer (DMGT) was performed in DNA repair-proficient and UV-hypersensitive, repair-deficient Chinese hamster ovary (CHO) cell lines using the UV-irradiated thymidine kinase gene from herpes simplex virus (HSV-TK). Transformation frequencies in repair-deficient CHO cell lines declined relative to repair-proficient cells with increasing UV damage in transfected DNA; approximately 3-fold higher UV fluence was required to inactivate 50% of irradiated HSV-TK plasmid molecules in repair-proficient cells. In cotransfection experiments performed with pairs of HSV-TK plasmids containing linker insertion mutations in TK coding sequences, moderate UV damage in plasmid DNA enhanced the yield of TK+ transformants resulting from homologous recombination between HSV-TK sequences up to 4-fold. These results suggest that UV damage in DNA can stimulate transformation of mammalian cells dependent on intermolecular DNA homology.     

Repair of DNA in replicated and unreplicated portions of the human genome.

Portions of the human genome that have replicated after ultraviolet light irradiation and those that remain unreplicated have both been examined for the distribution of pyrimidine dimers and the extent of repair replication following their removal. The data indicate that the number of unrepaired dimers and the extent of repair replication seen after their excision are equal in the replicated and unreplicated DNA. Furthermore, the daughter strand of replicated DNA is larger than the average interdimer distance found in the parental strand. Hence, DNA replication in normal human fibroblasts is clearly capable of getting past pyrimidine dimers, and a preferential repair of such lesions in DNA that is about to be or has been replicated does not operate to any visible extent in these cells.