Mutagenic SOS repair of site-specific psoralen damage in plasmid pBR322

Inefficient replication reduces RecA-mediated repair of UV-damaged plasmids introduced into competent Escherichia coli

Transformation of Escherichia coli with purified plasmids containing DNA damage is frequently used as a tool to characterize repair pathways that operate on chromosomes. In this study, we used an assay that allowed us to quantify plasmid survival and to compare how efficiently various repair pathways operate on plasmid DNA introduced into cells relative to their efficiency on chromosomal DNA. We observed distinct differences between the mechanisms operating on the transforming plasmid DNA and the chromosome. An average of one UV-induced lesion was sufficient to inactivate ColE1-based plasmids introduced into nucleotide excision repair mutants, suggesting an essential role for repair on newly introduced plasmid DNA. By contrast, the absence of RecA, RecF, RecBC, RecG, or RuvAB had a minimal effect on the survival of the transforming plasmid DNA containing UV-induced damage. Neither the presence of an endogenous homologous plasmid nor the induction of the SOS response enhanced the survival of transforming plasmids. Using two-dimensional agarose-gel analysis, both replication- and RecA-dependent structures that were observed on established, endogenous plasmids following UV-irradiation, failed to form on UV-irradiated plasmids introduced into E. coli. We interpret these observations to suggest that the lack of RecA-mediated survival is likely to be due to inefficient replication that occurs when plasmids are initially introduced into cells, rather than to the plasmid's size, the absence of homologous sequences, or levels of recA expression.

Third strand-mediated psoralen-induced correction of the sickle cell mutation on a plasmid transfected into COS-7 cells

A significant level of correction of the mutation responsible for sickle cell anemia has been achieved in monkey COS-7 cells on a plasmid containing a beta-globin gene fragment. The plasmid was treated in vitro with a nucleic acid 'third strand' bearing a terminal photoreactive psoralen moiety that binds immediately adjacent to the mutant base pair. Following covalent attachment of the psoralen by monoadduct or diadduct formation to the mutant T-residue on the coding strand, the treated plasmid was transfected into the cells, which were then incubated for 48 h to allow the cellular DNA repair mechanisms to remove the photoadducts. Upon re-isolation and amplification of the transfected plasmid, sickle cell mutation correction, as determined by sequence analysis of both complementary strands, was established in a full 1%. This result encourages extension of the approach to correct the mutation directly on the chromosome.

Asymmetric recognition of psoralen interstrand crosslinks by the nucleotide excision repair and the error-prone repair pathways

Psoralen is an asymmetric photoreactive intercalator with a furane and a pyrone side. When intercalated at 5'-TpA-3' sites and upon UVA irradiation, the psoralen can react with the thymine residues on both strands, introducing an interstrand crosslink. Using psoralen-coupled triple-helix-forming oligonucleotides, psoralen interstrand crosslinks can be site-specifically introduced in the coding sequence of URA3, a yeast auxotrophic marker carried on plasmid vectors. In addition, crosslinks introduced via a triple-helix-forming oligonuleotide are oriented with the furane side of the psoralen associated with a specific strand of the target sequence. Here, the transformation efficiency, the mutation frequency and the mutational spectra of site-specifically placed and oriented crosslinks were examined in yeast cells. We found that the nature of the targeted mutations depended on the crosslink orientation: bypass of the pyrone-adducted thymine yielded T-->A or T-->C substitutions and A insertions, while bypass of the furane-adducted thymine yielded T-->G substitutions and G insertions. Thus, the structure of the damage strongly influences the choice of the nucleotide incorporated during translesion synthesis. In addition, the observed pattern of mutagenesis suggests a coupling to transcription, similar to the one observed in mammalian cells. Finally, the substitutions affected only the coding strand when the pyrone link of the psoralen crosslink was on this strand, whereas they affected both strands when the pyrone link was on the transcribed strand, suggesting that the incision preference of psoralen crosslinks, which has been observed with purified uvrABC proteins in bacteria, is conserved in live eucaryotic cells.

Triplex-mediated, in vitro targeting of psoralen photoadducts within the genome of a transgenic mouse

Light-activated psoralens can covalently modify DNA and are widely used to study nucleic acid secondary structure and mutagenesis. Sequence specificity can be added to the photoaddition reaction by attaching the psoralen to an oligonucleotide designed to recognize a double-stranded DNA binding site through formation of a triple helix. We have previously used this strategy to study targeted psoralen modification of a triplex binding site within the bacterial supF gene carried in viral genomes. In the present work we report the targeting of psoralen photoadducts in vitro to a specific site in the genome of a transgenic mouse. Both 10 base and 16 base oligonucleotide-psoralen conjugates were capable of sequence-specific modification of genomic mouse DNA, while a truncated 8 base conjugate was not. Light activation was necessary, and a dose dependence was demonstrated for target site modification and mutagenesis. The 10 base conjugate rapidly found its target, with sequence-specific binding occurring after just 10 min incubation in the presence of mouse DNA. The ability to target psoralen photoadducts within mammalian genomes may prove useful in the study of chromatin structure and DNA repair. Moreover, this work may lead to potential in vivo applications of targeted psoralen modification.

8-Methoxypsoralen DNA interstrand cross-linking of the ribosomal RNA genes in Tetrahymena thermophila. Distribution, repair and effect on rRNA synthesis

The distribution and repair of 8-methoxypsoralen-DNA interstrand cross-links in the ribosomal RNA genes (rDNA) in Tetrahymena thermophila have been studied in vivo by Southern blot analysis. It is found that the cross-links at a density of < or = 1/2 x 10(4) base pairs (bp) are distributed equally between three domains (terminal spacer, transcribed region and central spacer) as defined by restriction enzyme analysis (BamHI and ClaI). It is furthermore shown that a dosage resulting in approximately one cross-link per rDNA molecule (21 kbp, two genes) is sufficient to block RNA synthesis. Finally, it is shown that the cross-links in the rDNA molecules are repaired at equal rate in all three domains within 24 h and that RNA synthesis is partly restored during this repair period. The majority of the cells also go through one to two cell divisions in this period but do not survive.

Psoralens cleave pBR322 DNA under ultraviolet radiation

Supercoiled (SC) pBR322 was used to probe the recent claim that 5-geranoxylpsoralen (5-GOP) did not photoreact with DNA. Contrary to expectations, 5-GOP was found to damage DNA in the presence of UV-A through two competing pathways: (a) single strand breaks, identified by the conversion of supercoiled into open circular and linear DNA, and (b) cross-linking, revealed by the fluence-dependent decrease in the extent of denaturation of the double stranded supercoiled DNA to single stranded circular DNA. In addition, a fluence-dependent modification reduced the ability of the restriction enzyme EcoR I to linearize the photosensitized DNA, and alkali-labile lesions were generated. Psoralen, 5-methoxypsoralen, and 8-methoxypsoralen, which are well-known to undergo cycloaddition to DNA, had a more pronounced effect on supercoiled DNA. Single strand breaks occurred more readily than with 5-GOP, and the surviving SC form remaining had reduced electrophoretic mobility in agarose gels. In all cases, the DNA damage was more prominent when oxygen was absent.

Psoralen damage-induced plasmid recombination in Saccharomyces cerevisiae: dependence on RAD1 and RAD52

Photoreaction with psoralen, a DNA-crosslinking reagent, induces mitotic recombination in the yeast Saccharomyces cerevisiae. Psoralen damage-induced recombination was studied with non-replicating plasmids, which transform yeast cells by undergoing recombination events with chromosomal DNA. When plasmid DNA was photoreacted with psoralen in vitro and transformed into yeast cells, transformation was stimulated by psoralen modification in a dose-dependent manner. The stimulation by psoralen damage requires RAD52 gene function and is partially dependent on RAD1. Analysis of transformants indicates that plasmid integration occurs at the homologous chromosomal loci. Multiple tandem integrations are common in repair-proficient cells, with more than 20 copies of integrated plasmid seen in some transformants. Multiple integration depends on RAD1 function; only 9% of rad1 transformants, compared to 80% of RAD transformants, contained multiple plasmid copies, while 52% of the rad1 transformants were produced by gene conversion.

DNA damage by 8-methoxypsoralen plus near ultraviolet light (PUVA) and its repair in Escherichia coli: genetic analysis

Mutants of Escherichia coli, hyper-resistant and sensitive to 8-methoxypsoralen plus near ultraviolet light (PUVA) have been isolated and studied. Results show that a mutation, located at 57.2 min on the linkage map of E. coli, is responsible for the hyper-resistant phenotype. It is also responsible for the synthesis of a 55-kdal protein in high concentrations. In a wild-type cell the synthesis of this enzyme is inducible by mitomycin C. There are indications that the mutation may have occurred in a regulatory gene, puvR, and as a result the operon, including a putative puvA gene (the structural gene for the synthesis of the 55-kdal protein), is expressed constitutively. A model for the control of the PUV operon is proposed.

A comparison of the relative activities of 8 radiosensitizers in the SOS chromotest

Misonidazole, and RSU 1069 and 6 of its analogues are all reported to show increased cytotoxicity towards hypoxic cells compared to oxic cells. DNA is considered to be the target through which these drugs exert their cytotoxic activity. Therefore we monitored induction of the SOS response in uvrABC excinuclease proficient and deficient strains of E. coli, under oxic and hypoxic conditions, as an indirect method of assessing the activity of these drugs towards DNA in a biological system. This was done using the SOS chromotest which utilizes E. coli strains which possess a sfiA::lacZ fusion allowing induction of the SOS response to be monitored by assaying beta-galactosidase activity. All of the drugs tested here show some induction of the SOS response in both uvrABC excinuclease proficient and deficient strains. Data shown here suggests that the uvrABC excinuclease is important in the production of a SOS induction signal from RSU 1069-induced DNA lesions and that RSU 1069 may act as a crosslinking agent. The data also shows that SOS induction activity and toxicity do not necessarily correlate and that production of a SOS induction signal may occur via a different pathway for RSU 1069 than for its analogues.

Nucleotide excision repair in Escherichia coli

One of the best-studied DNA repair pathways is nucleotide excision repair, a process consisting of DNA damage recognition, incision, excision, repair resynthesis, and DNA ligation. Escherichia coli has served as a model organism for the study of this process. Recently, many of the proteins that mediate E. coli nucleotide excision have been purified to homogeneity; this had led to a molecular description of this repair pathway. One of the key repair enzymes of this pathway is the UvrABC nuclease complex. The individual subunits of this enzyme cooperate in a complex series of partial reactions to bind to and incise the DNA near a damaged nucleotide. The UvrABC complex displays a remarkable substrate diversity. Defining the structural features of DNA lesions that provide the specificity for damage recognition by the UvrABC complex is of great importance, since it represents a unique form of protein-DNA interaction. Using a number of in vitro assays, researchers have been able to elucidate the action mechanism of the UvrABC nuclease complex. Current research is devoted to understanding how these complex events are mediated within the living cell.

Incision by UvrABC excinuclease is a step in the path to mutagenesis by psoralen crosslinks in Escherichia coli

4,5',8-Trimethylpsoralen (psoralen) plus near UV light produces interstrand crosslinks and monoadducts in DNA, both of which are mutagenic. In Escherichia coli, crosslinks are incised by UvrABC excinuclease, an event that can lead to homologous recombination and repair. To determine whether UvrABC incision of crosslinks is a step in the path to mutagenesis as well as repair, the effect of DNA homologous to a target gene on a plasmid was determined. pSV2-gpt DNA was treated with psoralen and transformed into a pair of hosts: one was gpt+, the other was delta (gpt-lac)5. The DNA was extracted and transformed into a tester strain [delta (gpt-lac)5] in which Gpt- mutations in the plasmid were scored. The results show that psoralen-induced mutations were reduced to background levels by the presence of the gpt+ homolog in the host chromosome. delta gpt hosts that were constitutively induced for the SOS response yielded point mutations, whereas noninduced hosts yielded almost exclusively large deletions. Since crosslinks were estimated to be responsible for most of the mutations observed, we conclude that the premutagenic lesion of psoralen crosslinks is recombinagenic and therefore very likely to be the product of UvrABC incision.

Mutagenesis induced by site specifically placed 4'-hydroxymethyl-4,5',8-trimethylpsoralen adducts

Closed circular double stranded M13mp19 DNA containing a site-specifically placed HMT (4'-hydroxymethyl-4-5'-8-trimethylpsoralen) monoadduct or crosslink was synthesized in vitro. The damaged DNA were scored for loss of infectivity by transfection into repair proficient or deficient E. coli and into SOS induced E. coli. Mutant phages were detected by the loss of alpha-complementation between the viral and the host Lac Z genes or by the acquisition of resistance to kpn I digestion. Our results indicate that HMT mutagenesis is targeted and that deletion or transversion of the modified thymidine is the predominant sequence change elicited by a monoadduct or a crosslink. Transfection of the monoadducted DNA into a Uvr A deficient strain did not change the mutation pattern but did increase the respective mutation frequencies. Transfection of the crosslinked DNA into a SOS induced host resulted in the appearence of other types of mutations attributable to an increase in both targeted and untargeted mutations.

The effects of covalent additions of a psoralen on transcription by E. coli RNA polymerase

Synthetic DNA substrates containing a site-specifically engineered psoralen monoadduct or diadduct were used to characterize the response of the E. coli RNA polymerase elongation complex to these lesions. The psoralen derivative HMT (4'-hydroxymethyl-4,5', 8-trimethylpsoralen) was site specifically placed into two synthetic double-stranded DNA fragments each of which contained an E. coli RNA polymerase promoter at one end. The HMT molecule was attached to the middle of the DNA fragments as either a furan-side monoadduct or an interstrand diadduct. Transcription off the HMT crosslinked DNA templates showed that E. coli RNA polymerase terminated at the HMT diadduct site, i. e., one nucleotide before the modified thymidine residue on the template strand. The furan-side monoadduct when on the template strand also blocked transcription by the polymerase. However, no effect on transcription was observed when the monoadduct was located on the non-template strand.

Suppressible base substitution mutations induced by angelicin (isopsoralen) in the Escherichia coli lacI gene: implications for the mechanism of SOS mutagenesis

Angelicin- plus near-UV-induced mutations were umuC dependent in Escherichia coli K-12. Angelicin, a monofunctional psoralen derivative, is believed to damage DNA almost exclusively at pyrimidine bases. To broaden our knowledge about the mutagenic specificity of SOS-dependent mutagens, we determined the mutational specificity of 233 suppressible lacI mutations induced by angelicin. More than 90% of the nonsense mutations arose via transversion substitutions. The three most frequently mutated sites were at A-T base pairs and accounted for more than one-third of all induced nonsense mutations. The two hottest sites were at the only occurrences of the 5'-TATA-3' tetranucleotide in lacI, a sequence expected to be a preferred binding site for a psoralen. Both A-T-to-T-A and A-T-to-C-G transversions were well induced by angelicin treatment, but the frequency of each transversion depended on the particular site. We also detected significant induction of transversion mutations at G-C sites. The induction of transversions by an SOS-dependent mutagen that generates lesions at pyrimidines supports the idea that DNA lesions influence the selection of bases that are incorporated via the process of SOS repair.

Lethal and mutagenic effects of 8-methoxypsoralen-induced lesions on plasmid DNA

The genotoxic effect of 8-methoxypsoralen damages (monoadducts and crosslinks) on plasmid DNA was studied. pBR322 DNA was treated with several concentrations of 8-methoxypsoralen plus fixed UVA light irradiation. After transformation into E. coli cells with different repair capacities (uvrA, recA and wild-type), plasmid survival and mutagenesis in ampicillin- and tetracycline-resistant genes were analysed. Results showed that crosslinks were extremely lethal in all 3 strains; indeed, it seemed that they were not repaired even in proficient bacteria. Monoadducts were also found to be lethal although they were removed to some extent by the excision-repair pathway (uvrA-dependent). Damaged plasmid DNA appeared to induce mutagenic repair, but only in the wild-type strain. In order to study the influence of the SOS response on plasmid recovery, preirradiation of the host cells was also performed. Preirradiation of the uvrA or wild-type strains significantly increased plasmid recovery. Consistent with the expectations of SOS repair, no effect was observed in preirradiated recA cells. Plasmid recovery in the excision-deficient strain was mainly achieved by the mutagenic repair of some fraction of the lesions, probably monoadducts. The greatest increase in plasmid recovery was found in the wild-type strain. This likely involved the repair of monoadducts and some fraction of the crosslinks. We conclude that repair in preirradiated repair-proficient cells is carried out mainly by an error-free pathway, suggesting enhancement of the excision repair promoted by the induction of SOS functions.

Action mechanism of ABC excision nuclease on a DNA substrate containing a psoralen crosslink at a defined position

Many carcinogenic as well as chemotherapeutic agents cause covalent linkages between complementary strands of DNA. If unrepaired, DNA crosslinks are blocks to DNA replication and transcription and therefore represent potentially lethal lesions to the cell. Genetic studies of Escherichia coli have demonstrated that the repair enzyme ABC excision nuclease, coded for by the three unlinked genes, uvrA, uvrB, and uvrC, plays a crucial role in DNA crosslink repair. To study the molecular events of ABC excision nuclease-mediated crosslink repair, we have engineered a DNA fragment with a psoralen-DNA interstrand crosslink at a defined position, digested this substrate with pure enzyme, and analyzed the reaction products on DNA sequencing gels. We find that the excision nuclease cuts only one of the two strands involved in the crosslink, incises the crosslink by hydrolyzing the ninth phosphodiester bond 5' and the third phosphodiester bond 3' to the furan-side thymine of the crosslink, and does not produce double-strand breaks at any significant level. Based on these data, we present a model by which ABC excision nuclease initiates crosslink repair in vivo.

Wavelength dependence for the photoreversal of a psoralen-DNA cross-link

We report an action spectrum for the photoreversal of a psoralen cross-link joining two self-complementary DNA oligonucleotides. The cross-link was formed between two thymines (T) on opposite strands of the DNA oligomers and 4'-(hydroxymethyl)-4,5',8-trimethylpsoralen (HMT). For comparison, we also present an action spectrum for the photoreversal of the isolated diadduct T-HMT-T. The wavelength dependence for the diadduct photoreversal parallels its absorption spectrum. Both the diadduct and the cross-linked DNA can be photoreversed by exposure to light with wavelengths between 240 and 313 nm. We did not observe photoreversal at 334 nm or above. At least two distinct absorption bands appear to contribute to photoreversal. We measured a quantum yield of 0.16 for photoreversal of the isolated diadduct at wavelengths between 240 and 266 nm. For wavelengths above 280 nm, the quantum yield is 0.30. We also observed a preferential photoreversal at the furan end of the psoralen in the T-HMT-T diadduct. In contrast, the cross-linked DNA oligonucleotides preferentially photoreversed at the pyrone end of the psoralen adduct. The rate constant for photoreversal of the cross-linked DNA is larger than that for the isolated diadduct at wavelengths below 300 nm.

Repair in E. coli of transforming plasmid DNA damaged by psoralen plus near-ultraviolet irradiation

Treatment of DNA with psoralen plus near-ultraviolet irradiation gives rise to both monoadducts and cross-links. We have examined the repair of plasmid NTP16 DNA treated in this way in vitro and then used to transform E. coli. Monoadducts are found to be potentially lethal, and can be repaired by uvr-dependent and recA-dependent pathways. The presence of a related resident plasmid in the transformed cells can enhance the survival of the incoming damaged NTP16 DNA. This effect is not recA-dependent, and a similar effect (designated "resident enhanced repair") has been observed previously with UV-irradiated plasmids of this particular incompatibility group. Removal of unbound psoralen from the plasmid DNA and exposure to further NUV is known to increase the ratio of cross-links to monoadducts, and we demonstrate that such cross-linked plasmid DNA is not readily repaired following transformation. However in the presence of homologous DNA (related resident plasmid) there is evidence for the repair, and hence uptake by the cell, of cross-linked DNA.

Mutagenesis of the lac promoter region in M13 mp10 phage DNA by 4'-hydroxymethyl-4,5',8-trimethylpsoralen

Double-stranded M13 phage DNA (M13 mp10 replicative form) was photoreacted with 4'-hydroxymethyl-4,5',8-trimethylpsoralen, using light of wavelength greater than 320 nm or greater than 390 nm to generate predominantly crosslinks or monoadducts, respectively. The damaged DNAs were scored for inactivation and mutagenesis after transfection into Escherichia coli. The appearance of light-blue or colorless plaques on indicator medium showed that mutation had occurred in the lac insert of the viral DNA. A comparison of the consequences of the two phototreatments with psoralen supports the idea that crosslinks are both more lethal and more mutagenic than monoadducts. Numerous mutant clones partially or totally deficient in beta-galactosidase were plaque-purified and amplified. The viral DNA of each clone was sequenced by the dideoxy chain-terminating procedure. All of the observed base-pair changes were mapped to the lac promoter region and consisted of 3 transition, 14 transversion, and 6 single base-pair frame-shift mutations. The predominant mutation was a T.A----G.C transversion.