- Email: [email protected]
The SOS-dependent upregulation of uvrD is not required for efficient nucleotide excision repair of ultraviolet light induced DNA photoproducts in Escherichia coli
Mutation Research 485 (2001) 319–329
The SOS-dependent upregulation of uvrD is not required for efficient nucleotide excision repair of ultraviolet light induced DNA photoproducts in Escherichia coli David J. Crowley∗ , Philip C. Hanawalt Department of Biological Sciences, Stanford University, Stanford, CA 94305-5020, USA Received 18 September 2000; received in revised form 26 January 2001; accepted 29 January 2001
Abstract We have shown previously that induction of the SOS response is required for efficient nucleotide excision repair (NER) of the major ultraviolet light (UV) induced DNA lesion, the cyclobutane pyrimidine dimer (CPD), but not for repair of 6-4 photoproducts (6-4PP) or for transcription-coupled repair of CPDs [1]. We have proposed that the upregulation of cellular NER capacity occurs in the early stages of the SOS response and enhances the rate of repair of the abundant yet poorly recognized genomic CPDs. The expression of three NER genes, uvrA, uvrB, and uvrD, is upregulated as part of the SOS response. UvrD differs from the others in that it is not involved in lesion recognition but rather in promoting the post-incision steps of NER, including turnover of the UvrBC incision complex. Since uvrC is not induced during the SOS response, its turnover would seem to be of great importance in promoting efficient NER. Here we show that the constitutive level of UvrD is adequate for carrying out efficient NER of both CPDs and 6-4PPs. Thus, the upregulation of uvrA and uvrB genes during the SOS response is sufficient for inducible NER of CPDs. We also show that cells with a limited NER capacity, in this case due to deletion of the uvrD gene, repair 6-4PPs but cannot perform transcription-coupled repair of CPDs, indicating that the 6-4PP is a better substrate for NER than is a CPD targeted for transcription-coupled repair. © 2001 Elsevier Science B.V. All rights reserved. Keywords: Excision repair; UvrD; Helicase II; Transcription-coupled repair; SOS response; UV damage
1. Introduction The UvrD protein, also known as DNA helicase II, plays multiple roles in maintaining genomic stability in Escherichia coli. UvrD is a 3 → 5 DNA helicase that is required for methyl-directed mismatch repair [2] and been implicated in DNA replication and ∗ Corresponding author. Present address: Biology Department, College of Liberal Arts, Mercer University, 1400 Coleman Avenue, Macon, GA 31207, USA. Tel.: +1-478-301-2250; fax: +1-478-301-2802. E-mail address: crowley [email protected] (D.J. Crowley).
recombination [3]. In addition, UvrD is involved in the post-incision steps of nucleotide excision repair (NER), the versatile DNA repair process responsible for removal of a variety of lesions from DNA. Following lesion recognition and dual incisions mediated by the UvrA, UvrB, and UvrC proteins, UvrD promotes excision of the damaged oligonucleotide and releases UvrC protein from the DNA. DNA polymerase I subsequently fills in the resulting gap and dissociates UvrB from the DNA. DNA ligase completes the repair reaction by sealing the nick between the newly synthesized DNA patch and the parental DNA strand. UvrD-mediated turnover of UvrC and, subsequently,
0921-8777/01/$ – see front matter © 2001 Elsevier Science B.V. All rights reserved. PII: S 0 9 2 1 - 8 7 7 7 ( 0 1 ) 0 0 0 6 8 - 4
320
D.J. Crowley, P.C. Hanawalt / Mutation Research 485 (2001) 319–329
UvrB is necessary for promoting the catalytic rate of NER that characterizes this process in vivo [2]. Many mutant alleles of uvrD have been isolated and their pleiotropic phenotypes have been studied extensively. As would be expected for cells defective in any step of NER, uvrD mutants are UV sensitive. However, their sensitivity is not as great as that of uvrA, uvrB, or uvrC mutants that are completely defective in NER. Similarly, unlike uvrA, uvrB, or uvrC mutants, liquid holding recovery [4] and host cell reactivation [5,6] do occur in uvrD mutants, albeit to a reduced extent compared to that in wildtype cells. These observations suggest that uvrD mutants perform limited NER, a fact that was confirmed when repair of UV damage was measured in uvrD mutants [6,7]. The studies showed that the removal of thymine dimers, as measured by two dimensional chromatography, was greatly attenuated, but not eliminated, in uvrD mutants. We previously demonstrated that induction of the SOS response led to efficient overall NER of the major UV-induced lesion, the cyclobutane pyrimidine dimer (CPD), but had no effect on the rate of overall repair of 6-4 photoproducts (6-4PP) [1]. The subpathway of NER termed transcription-coupled repair, in which lesions in the transcribed strand of active genes are repaired in preference to those in the nontranscribed strand or genome overall, was also unaffected by SOS induction. To explain these results, we proposed that upregulation of the SOS proteins UvrA, UvrB, and/or UvrD enhanced cellular NER capacity and we showed that the kinetics of UvrA and UvrB upregulation correlated with efficient global NER of CPDs. Since removal of 6-4PPs and transcription-blocking CPDs were unaffected by this NER upregulation, we additionally proposed that these lesions are preferred targets for constitutive NER complexes, and thus, do not require upregulation of cellular repair capacity for their efficient repair. Based on the fact that uvrD mutants perform limited NER and that uvrD is an SOS-regulated gene induced with kinetics similar to those of the other SOS regulated NER genes, uvrA and uvrB [2,8], we designed experiments to test these hypotheses. First, we measured the rates and extents of overall repair of CPDs and 6-4PPs, and specific repair of transcription-blocking CPDs in uvrD mutants to determine if the latter two were preferred substrates for NER in cells in which the repair capacity was limited. We then asked whether
SOS-dependent upregulation of UvrD was required for achieving wildtype levels of overall NER of CPDs. To do this, we transformed a uvrD deletion mutant with a plasmid carrying the wildtype uvrD gene under control of a foreign promoter. This construct produces a level of UvrD protein similar to that present in uninduced wildtype cells, but the level is not increased following induction of the SOS response. We measured the rates and extents of repair of UV lesions from the genome overall and from the induced lactose operon to determine the role of UvrD, if any, in promoting SOS-dependent NER.
2. Materials and methods 2.1. Bacterial strains and plasmids The E. coli K-12 strain HL108 is a thyA deoC derivative of W3110 [9]. Strain HL952 is a derivative of HL108 that carries a uvrA::Tn10 mutant allele. Strain HL1030 is a derivative of HL108 and was the recipient of the uvrD294::kan allele by P1 transduction using strain SK4090 as a donor [10]. HL1031 and HL1032 are derivatives of HL1030 that were transduced with mutant alleles (Tn10 insertions) of uvrA and uvrC, respectively. HL1054 is a derivative of HL108 and was the recipient of the uvrD::tet allele from the donor strain GE1752 uvrD::tet [3]. HL1054 was selected by screening for tetracycline resistant transductants on LB plates supplemented with low concentrations of tetracycline (5 g/ml). HL1056 was made by transforming HL1054 with the uvrD+ plasmid pET9d-H2wt (kindly provided by S. Matson from UNC-Chapel Hill), selecting for kanamycin resistance, and screening for UV resistance. This plasmid contains the entire wildtype uvrD coding region under control of the T7 RNA polymerase promoter [3]. The transformation was performed by electroporation. 2.2. UV survival assay Cells were grown at 37◦ C in either Davis minimal salts supplemented with 0.4% glucose and 10 mg/ml thymine or in LB. Stationary phase cultures of approximately equal cell density were streaked on LB plates using sterile cotton swabs. Plates were irradiated with
D.J. Crowley, P.C. Hanawalt / Mutation Research 485 (2001) 319–329
254 nm ultraviolet light in incremental doses. Plates were incubated overnight at 37◦ C. 2.3. Cell growth and DNA preparation Cells were grown at 37◦ C in Davis minimal salts supplemented with 0.4% glucose and 10 mg/ml thymine. To label the DNA, cells were grown in medium containing 1 Ci/ml of 3 H-thymine in addition to the nonradioactive thymine. Cultures were grown to saturation and subcultured in fresh medium supplemented with 1 mM isopropyl -D-thiogalactoside (IPTG). Cultures were grown to mid log phase (approximately 3 × 108 cells per milliliter) as measured by OD600 . Cells were collected by filtration on 0.45 M Millipore filters, washed with prewarmed Davis medium, and resuspended in Davis medium containing 1 mM IPTG. Cells were UV-irradiated at the doses indicated and placed in a flask containing growth supplements. Samples of the culture were removed at various times and mixed with an equal volume of ice cold NET (100 mM NaCl, 10 mM Tris pH 8.0, 10 mM EDTA) buffer. Cells were pelleted by centrifugation at 4◦ C and resuspended in TE pH 8.0. Cells were lysed by addition of lysozyme (to 1 mg/ml) and RNaseA (to 100 g/ml) and incubation for 15 min at 37◦ C. Proteinase K (to 100 g/ml) and Sarkosyl (to 0.5%) were then added and the mixture was incubated at 50◦ C for 1 h. The DNA was extracted with phenol:chloroform and precipitated with 2.5 M ammonium acetate and two volumes of 95% ethanol. Purified DNA was resuspended in TE pH 8.0. A portion of each DNA sample was incubated with SstII and ApaI restriction enzymes according to manufacturers’ instructions (Gibco-BRL). The remaining DNA was lightly sonicated using a Branson sonifier and the concentration was determined by fluorometry using Hoechst 33258 [11]. The radioactivity in 3 H-labeled DNA was quantified by scintillation spectrometry. 2.4. Repair of CPDs in the lactose operon The frequency of CPDs in the individual strands of the lactose operon was determined using an established method [12,13]. DNA restricted with ApaI and SstII was quantified as described above and 300 ng of each sample was treated or mock-treated with T4
321
endonuclease V in NET buffer containing 1 mg/ml bovine serum albumin. Samples were electrophoresed overnight in alkaline gels containing 1% agarose. DNA was transferred to Hybond N+ membranes by Southern blotting and hybridized with 32 P -labeled probes. Strand-specific RNA probes were generated according to the protocol of Promega, using plasmid pZH10 [13] as a template. Detection was performed using a BioRad phosphorimager and its associated Molecular Analyst software. The frequency of CPDs per 6.6 kb restriction fragment was calculated from the percentage of fragments with no CPDs (zero class) using the Poisson expression (−ln of the zero class = average number of dimers per fragment). 2.5. Immunoassay for overall NER The repair of CPDs and 6-4PPs was measured using an immunoassay. Following denaturation by boiling, 50 ng (CPD) or 500 ng (6-4PP) of each DNA sample (irradiated with 40 J/m2 ) were loaded in triplicate onto a Hybond N+ membrane using a slot blot apparatus. When lower doses of UV were used, higher amounts of DNA were loaded onto the membranes. At 8 J/m2 , 250 ng (CPD) of each sample were loaded; at 4 J/m2 , 500 ng (CPD) of each sample were loaded. The membrane was incubated for 2 h in the presence of mouse antibodies against either CPDs (TDM-2) or 6-4PPs (64M-2) diluted 1:2000 in PBS (antibodies were a generous gift of Mori et al. [14]. Horseradish peroxidase-conjugated secondary antibodies were used at a dilution of 1:5000 and detected using the enhanced chemiluminescence detection system (Amersham) and subsequent phosphorimager analysis (BioRad). Following detection, the amount of 3 H-labeled DNA loaded in each slot was confirmed by scintillation counting.
3. Results To study the effect of a complete loss of UvrD activity on the repair of UV-induced photoproducts, we constructed strain HL1030, which carried the uvrD294::kan allele in an otherwise repair proficient background [10]. Cells with mutant alleles of uvrD, including similar deletion mutants, have been shown to be significantly UV sensitive [15]. The level of UV
322
D.J. Crowley, P.C. Hanawalt / Mutation Research 485 (2001) 319–329
Fig. 1. The moderate UV sensitivity of uvrD mutants is due to low levels of NER. Cultures of the wildtype strain HL108 (right) and its isogenic mutants HL952 (uvrA, second from right), HL1030 (uvrD, second from left), and HL1031 (uvrA uvrD, left) were streaked on an LB plate and irradiated in increments with the indicated doses of 254 nm UV.
sensitivity was not as great as that produced in cells defective in either uvrA, uvrB, or uvrC [5,6], supporting other studies that showed that uvrD mutants could perform limited excision repair of UV-induced damage in vivo [6,7]. We confirmed the intermediate UV sensitivity of uvrD mutants using a UV survival plate assay and comparisons between isogenic wildtype, uvrD, uvrA, and uvrA uvrD mutant strains (Fig. 1). The uvrD mutant was significantly more UV sensitive than the wildtype strain, which showed no detectable loss of viability up to 40 J/m2 UV. However, the uvrD mutant was much more resistant than the isogenic uvrA mutant, with survivors apparent after treatment with up to 16 J/m2 UV. Transduction of a uvrA mutant allele (or a uvrC mutant allele, data not shown) into the uvrD deletion mutant eliminated its resistance, demonstrating
that this resistance is likely due to a low cellular capacity for Uvr-dependent excision repair of UV damage. The residual UvrABC-dependent excision that occurs in uvrD mutants has been reported in earlier papers to be repair of CPDs [6,7]. These repair studies were done by measuring loss of thymine dimer content over time by two-dimensional paper chromatography, a rather insensitive method that not only measures CPD content but also, presumably, 6-4PP content. We sought to determine more carefully the rate and extent of removal of both CPDs and 6-4PPs from the genome overall using antibodies raised against the two respective lesions (Fig. 2). Using a dose of 40 J/m2 , we found that only 6-4PPs were removed from the genome, albeit at a greatly reduced rate compared to that in the isogenic uvrD+ strain (see Fig. 5B for uvrD+ rate). No detectable repair of CPDs occurred up to 90 min after UV (data not shown). Irradiation with 40 J/m2 UV induces approximately 4000 CPDs and 1000 6-4PPs in the DNA of an exponentially growing cell with two chromosomes [1], meaning that about 550 6-4PPs were removed in the first 60 min after UV in the uvrD mutants, at what appeared to be a constant rate (Fig. 2). No repair of either CPDs or 6-4PPs occurred in the uvrA uvrD or uvrC uvrD double mutants (data not shown), confirming that the limited repair in the uvrD mutants was mediated by UvrABC proteins. Although we found no detectable repair of CPDs in the genome overall at a dose of 40 J/m2 UV, we wanted to see whether transcription-coupled repair of these lesions occurred in the uvrD mutants at this dose. Our previous work had shown that 6-4PPs and transcription-blocking CPDs (i.e. substrates for transcription-coupled repair) are preferred substrates for NER in cells with a limited repair capacity [1]. Therefore, we felt it was possible that transcription-coupled repair of CPDs might be occurring in the uvrD mutants at a rate comparable to that of the overall removal of 6-4PPs and that the limited number of total CPDs repaired in this manner were undetectable in the antibody assay due to the high background of unrepaired genomic CPDs. Surprisingly, we found that this was not the case, as both the nontranscribed and transcribed strands of the induced lactose operon showed little or no loss of T4 endonuclease sensitive sites up to 60 min after UV (Fig. 3), mirroring the results from the CPD antibody assay
D.J. Crowley, P.C. Hanawalt / Mutation Research 485 (2001) 319–329
323
Fig. 2. The 6-4PPs, but not the CPDs, are repaired in uvrD mutants irradiated with 40 J/m2 UV. (A) Typical immunoblots generated from experiments using anti-CPD (left, squares) or anti-6-4PP (right, diamonds) monoclonal antibodies on DNA isolated from strain HL1030 uvrD at the indicated time points after irradiation with 40 J/m2 UV. (B) Graphical representation of data generated from multiple immunoblots of DNA from three independent experiments. Each error bar represents one standard deviation calculated from the averages of the independent experiments. Error bars not shown are obscured by the data point. Symbols as in part A.
(Fig. 2). We conclude that 6-4PPs are the preferred substrate for repair in uvrD mutants, which have a very low capacity for NER. Since a dose of 40 J/m2 was lethal to the uvrD deletion mutants (Fig. 1), it was clear that this level of UV damage saturated the limited repair capacity of these cells. We measured the removal of CPDs from the genome of uvrD deletion mutants irradiated with
either 8 or 4 J/m2 UV, doses at which we detected significant survival by our plating assay (Fig. 1). At 8 J/m2 , 800 CPDs are induced in the DNA (and 200 6-4PPs, which cannot be detected using the antibody assay) and approximately half were removed 1 h after UV (Fig. 4). Therefore, about 400 CPDs were removed from these cells in the first hour after UV. At 4 J/m2 , 400 CPDs are induced in each cell (about
324
D.J. Crowley, P.C. Hanawalt / Mutation Research 485 (2001) 319–329
Fig. 3. Transcription-coupled repair of CPDs is not detectable in uvrD mutants irradiated with 40 J/m2 UV. (A) Southern blots showing maintenance of T4 endonuclease V (T4 endo V) sensitive sites over time in the two strands of the IPTG-induced lactose operon from HL1030 uvrD mutants. The average number of CPDs was measured in each strand of the 6.6 kb restriction fragment in DNA isolated at the indicated times after irradiation with 40 J/m2 UV. (B) Graphical representation of data collected from Southern blots. Each point represents the average repair calculated from three independent biological experiments. (䉭) Transcribed strand; (䊊) nontranscribed strand.
100 6-4PPs) and after 40 min, about 70% had been removed from the genome. No signal was detected after 60 min, indicating that over 90% of the initial 400 CPDs were removed within the first hour after UV. Again, no repair was detected at either dose when the double mutants uvrA uvrD or uvrC uvrD were used (data not shown). Combining the results from the three different doses, and assuming that overall repair of 6-4PPs was complete in 1 h at the lower doses, we found that 500–600 lesions were repaired (in a UvrA- and UvrC-dependent manner) in uvrD deletion
mutants in 1 h regardless of the initial level of DNA damage. UvrD is upregulated as part of the SOS response along with the lesion recognition proteins, UvrA and UvrB. We had previously shown that the SOS-dependent upregulation of UvrA and UvrB correlated with efficient global NER of CPDs [1]. UvrD is upregulated three- to six-fold after UV [16–20] and its induction occurs with kinetics similar to those of UvrA and UvrB, based on similarities in the strength of LexA repression and the SOS box sequence [8]. To
D.J. Crowley, P.C. Hanawalt / Mutation Research 485 (2001) 319–329
325
Fig. 4. Repair of CPDs is detectable in uvrD mutants irradiated with low UV doses. Monoclonal antibodies specific for CPDs were used in an immunoassay with DNA isolated from HL1030 uvrD mutants after irradiation with 40 J/m2 ((䊐) from Fig. 2), 8 J/m2 (䉫), or 4 J/m2 (䊊). The 8 and 4 J/m2 points represent the average repair calculated from at least two immunoblots of samples from a single experiment.
test whether the SOS-dependent induction of UvrD was necessary for efficient global repair of CPDs, we transformed a uvrD deletion strain with a plasmid carrying the wildtype coding region of uvrD under control of the T7 phage promoter. This construct has been shown by direct immunological methods to produce approximately 350 UvrD molecules in cells that do not possess T7 RNA polymerase, just slightly below the wildtype level of 500 UvrD molecules per uninduced cell [3]. Since the uvrD gene in this construct is not under LexA control, the basal level of UvrD should be maintained in cells that are otherwise proficient in SOS induction. We considered the possibility that this plasmid, which is derived from pBR322, increases in copy number as part of the SOS response [21]. An increase in plasmid copy number after UV could mimic SOS upregulation of UvrD despite the lack of direct SOS induction of the plasmid borne uvrD+ gene. However, we have observed no significant change in plasmid copy number up to 60 minutes after UV treatment in these transformants (data not shown). Fig. 5A shows results from a UV survival plating assay similar to that in Fig. 1. The uvrD deletion in this case was marked by a tetracycline resistance element but retained the same level of UV sensitivity
Fig. 5. An SOS noninducible uvrD+ allele complements the UV sensitivity of a uvrD deletion mutant and facilitates efficient NER. (A) Cultures of the wildtype strain HL108 (top), the isogenic uvrD mutant HL1054 (middle), and the uvrD+ transformant HL1056 (bottom) were streaked on an LB plate and irradiated in increments with the indicated doses of 254 nm UV. (B) Monoclonal antibodies specific for CPDs (䊐) or 6-4PPs (䉫) were used in an immunoassay with DNA isolated from HL1056 cells (solid lines and symbols) after irradiation with 40 J/m2 . Dotted lines and empty symbols represent data from the isogenic wildtype strain HL108 [1] and are included for comparison. (C) Quantitative southern hybridization of strand-specific RNA probes to the IPTG-induced lactose operon was performed on DNA isolated from HL1056 cells (solid lines and symbols). The average number of CPDs was measured in each strand of the 6.6 kb restriction fragment in DNA isolated at the indicated times after UV irradiation with 40 J/m2 . (䊊) Transcribed strand; (䉭) nontranscribed strand. Dotted lines and empty symbols represent data from similar experiments using the isogenic wildtype strain HL108 [1] and are included for comparison.
326
D.J. Crowley, P.C. Hanawalt / Mutation Research 485 (2001) 319–329
previously demonstrated using the uvrD kanamycin deletion (Fig. 1). When the uvrD+ plasmid was transfected into these cells, UV resistance was enhanced greatly, although the transformants remained slightly more UV sensitive than the isogenic wildtype cells (compare growth at doses of 53, 66, and 80 J/m2 ). Overall NER (Fig. 5B) and transcription-coupled repair of CPDs (Fig. 5C) were measured in the transformants and the results are presented with the isogenic wildtype repair profiles for comparison [1]. The transformants showed a slightly lower rate of removal of both 6-4PPs and CPDs, but did achieve wildtype levels of repair 40 min after UV (Fig. 5B). The rate and extent of CPD repair in the transformants were greater than those of lexA3(Ind-) mutants that are unable to induce the SOS response and upregulate NER capacity [1]. The transformants were also able to perform transcription-coupled repair of CPDs, although both the transcribed and nontranscribed strands of the induced lactose operon showed an initial lag in repair of CPDs compared to that in wildtype cells (Fig. 5C). Twenty minutes after UV, the repair of CPDs in both strands corresponded with those of the wildtype. Consistent with our immunoassay results (Fig. 5B), the rate of repair of the nontranscribed strand, which is an indicator of the global repair rate [1], was similar to that of wildtype cells.
4. Discussion Our previous work demonstrated that induction of the SOS response in E. coli enhances NER capacity [1]. This enhanced NER capacity is necessary for efficient global repair of the principal UV-induced lesion, the CPD, but is not needed for the rapid rate of global repair of 6-4PPs or the efficient repair of CPDs in the transcribed strand of an active gene. In this study, we have continued to explore the relationship between the induction of the SOS response and the efficiency of NER using mutants deleted for the SOS-inducible NER gene, uvrD. We asked two questions: (1) what are the targets for the limited NER that occurs in uvrD mutants, and (2) is UvrD upregulation necessary for promoting wildtype levels of NER, specifically for the overall repair of CPDs?
4.1. Limited NER in uvrD mutants is targeted to 6-4PPs Based on our previous studies, we predicted that cells with a limited capacity for NER would target 6-4PPs and transcription-blocking CPDs for repair, limiting the removal of CPDs in nontranscribed regions of the genome [1]. Since uvrD mutants were reported to have a low level of NER capacity [6,7], we felt that we could use these cells to test this hypothesis. Using a dose of 40 J/m2 , which induces approximately 4000 CPDs and 1000 6-4PPs in the DNA of a log phase bacterium [1], we found that in the first 90 min after UV, only 6-4PPs were detectably removed from the uvrD deletion mutants (Fig. 2). No repair of CPDs was detected in the genome overall or in the nontranscribed strand of the induced lactose operon (Figs. 2 and 3). Perhaps a very small fraction of CPDs located in the transcribed strand of the lactose operon were removed within the first hour after UV (Fig. 3). The lack of significant repair of CPDs in the induced lactose operon of the uvrD deletion mutant is consistent with the results of another study [12]. Based on the number of lesions induced in a cell and the amount of repair calculated from our assays, we estimated that roughly 550 6-4PPs and perhaps a small number of CPDs were removed from the DNA of the uvrD deletion mutants after a dose of 40 J/m2 UV. These results clearly show that a 6-4PP is a better substrate for NER than is a transcription-blocking CPD, illustrating that the recognition and repair of 6-4PPs is even more efficient than transcription-coupled repair of CPDs in vivo. Two lower doses were employed to test whether global repair of CPDs could occur in uvrD mutants. Indeed, CPDs were removed at both 8 J/m2 (which produces 800 CPDs and 200 6-4PPs) and 4 J/m2 (which produces 400 CPDs and 80 6-4PPs) (Fig. 4). Interestingly, the number of lesions removed in both cases (assuming complete removal of 6-4PPs) was equivalent to the 500–600 lesions we calculated to be removed from the uvrD cells in the first hour after irradiation with 40 J/m2 . The fact that a fixed number of DNA lesions was removed in these cells independent of UV dose is not surprising given the absolute requirement for UvrD in promoting the turnover of UvrC in vitro. Biochemical studies have shown that only a limited number of incisions can take place when limiting concentrations of UvrC are present in reactions with-
D.J. Crowley, P.C. Hanawalt / Mutation Research 485 (2001) 319–329
out UvrD [22]. Only when UvrD is added to these reactions does incision proceed catalytically [19,23–25]. Based on these biochemical results, the number of UvrC molecules present in the cell should determine the amount of incision that could take place in cells without functional UvrD. Based on our data, we would, thus, expect that approximately 500–600 molecules of UvrC were present in each cell, a number that is in agreement with a recent immunological determination of 500 UvrC molecules per cell [26]. If approximately 500 UvrC molecules are indeed present in each cell, we explain our 40 J/m2 repair results as follows. Since approximately 400 UvrB molecules are present in each cell [1] and UvrA loading of UvrB at damage sites is catalytic [22], we propose that most of the 400 UvrB molecules are rapidly loaded at 6-4PPs, a favored NER substrate [27], by UvrA dimers shortly after irradiation. Available UvrC molecules then bind the UvrB–DNA complexes and incisions are made rapidly [28]. As determined both in vitro [19,23,24] and in vivo [7,28], UvrBC-mediated incisions persist in the absence of functional UvrD, due to the inability of cells to promote UvrC turnover, excision, and DNA polymerase I gap filling. Why then do we detect loss of these lesions if excision has not taken place? We considered that perhaps we only see “repair” of these lesions when we purify the DNA and denature it prior to loading it onto slot blots. The incised, lesion-containing oligonucleotides may be too small to be retained on a membrane or may be lost during phenol extraction, making them undetectable once we assay with lesion-specific antibodies. However, given that incision proceeds at a normal rate initially in uvrD mutants [7,28], it is difficult to understand why, if this hypothesis is true, we did not observe a rapid “repair” soon after UV that plateaus at later time points. A more likely explanation is that excision and perhaps UvrBC turnover does take place in uvrD mutants, but at greatly reduced rates. It has been shown that DNA polymerase I will promote excision of the damaged oligonucleotide and complete repair in the absence of UvrD in vitro [29]. If this is true in vivo, then the rate of DNA polymerase I gap filling in the absence of UvrD may determine the slow rate of repair we measured and may explain why repair does not plateau shortly after UV. This slow turnover may also explain why more lesions were removed (i.e. 550–600) than there were UvrB or UvrC molecules in the cell.
327
If the more widely cited number of 10–20 UvrC molecules per cell [2] is correct, there must be some turnover of these molecules to account for the levels of repair we observed in the uvrD mutants. Perhaps other cellular factors are able to provide alternative or “back-up” mechanisms for promoting excision and turnover of UvrBC in the absence of UvrD. One obvious candidate is the Rep helicase, which can form a heterodimer with UvrD that may play a role in DNA replication [30]. However, it has been shown that Rep cannot effectively substitute for UvrD function in NER in vitro [24]. Perhaps one of the nine remaining helicases in E. coli can perform the function of UvrD to some extent and help promote catalytic NER in vivo. Another candidate for the post-incision steps of NER is exonuclease VII, considering that a secondary xseA mutation increased the excision deficiency of a polAex mutant strain [31]. Regardless of the factor(s) involved, the number of lesions removed and the constant rate of their repair suggests that some turnover of NER proteins is likely to occur in the absence of functional UvrD in vivo. 4.2. SOS-dependent upregulation of UvrD is not required for efficient overall NER Three NER proteins are upregulated as part of the SOS response: UvrA, UvrB, and UvrD. We had proposed that the rapid SOS-dependent upregulation of UvrA and UvrB after UV was necessary for promoting efficient overall NER of the abundant, yet poorly recognized, CPDs [1]. In this study, we attempted to test whether UvrD induction was also necessary for the enhanced repair capacity we have observed in SOS proficient cells. We introduced a plasmid carrying an SOS-noninducible uvrD+ allele into a uvrD deletion mutant and measured overall NER of 6-4PPs and CPDs as well as transcription-coupled repair of CPDs and compared the results to wildtype cells (Fig. 5). Our results indicate that the uvrD (UvrD+ ) cells were nearly as efficient as wildtype cells in repairing both UV lesions, indicating that SOS upregulation of UvrD levels is not important for promoting the SOS-dependent repair of CPDs. Previous studies using this plasmid have shown that it produces slightly lower than wildtype levels of UvrD in transformed cells [3]. This may account for the initial lag in repair of CPDs detected in both strands
328
D.J. Crowley, P.C. Hanawalt / Mutation Research 485 (2001) 319–329
of the lactose operon and the slight UV sensitivity of the transformants compared to the isogenic wildtype (Fig. 5A). Since repair is restored to essentially wildtype levels by 20 min after UV, we believe that the upregulation of SOS proteins, likely UvrA and/or UvrB, promotes efficient repair and that basal levels of UvrD, expressed from a plasmid or from the chromosome, are sufficient to support this rate of NER.
Acknowledgements The authors would like to thank Ann Ganesan and C. Allen Smith for helpful discussions and critical reading of this manuscript. We would also like to thank Steve Matson for his generous gifts of the uvrD+ plasmid and deletion strains and Toshio Mori for his gift of CPD and 6-4PP antibodies. This work was supported by a Cellular and Molecular Biology traineeship GM07276 and Outstanding Investigator Grant CA44349 from the National Cancer Institute, NIH.
References [1] D.J. Crowley, P.C. Hanawalt, Induction of the SOS response increases the efficiency of global nucleotide excision repair of cyclobutane pyrimidine dimers, but not 6-4 photoproducts, in UV-irradiated Escherichia coli, J. Bacteriol. 180 (1998) 3345–3352. [2] E. Friedberg, G.C. Walker, W. Siede, DNA Repair and Mutagenesis, American Society of Microbiology, Washington, DC, 1995. [3] J.W. George, R.M. Brosh Jr., S.W. Matson, A dominant negative allele of the Escherichia coli uvrD gene encoding DNA helicase II. A biochemical and genetic characterization, J. Mol. Biol. 235 (1994) 424–435. [4] M.S. Tang, K.C. Smith, The effects of lexA101, recB21, recF143 and uvrD3 mutations on liquid-holding recovery in ultraviolet-irradiated Escherichia coli K12 recA56, Mutat. Res. 80 (1981) 15–25. [5] R.H. Rothman, A.J. Clark, Defective excision and postreplication repair of UV-damaged DNA in a recL mutant strain of E. coli K-12, Mol. Gen. Genet. 155 (1977) 267–277. [6] N.B. Kuemmerle, W.E. Masker, Effect of the uvrD mutation on excision repair, J. Bacteriol. 142 (1980) 535–546. [7] C.A. Van Sluis, I.E. Mattern, M.C. Paterson, Properties of uvrE mutants of Escherichia coli K12.I. Effects of UV irradiation on DNA metabolism, Mutat. Res. 25 (1974) 273–279.
[8] G.C. Walker, in: F.C. Neidhardt (Ed.), Escherichia coli and Salmonella: Cellular and Molecular Biology, Vol. 1, ASM Press, Washington, DC, 1996, pp. 1400–1416. [9] B.J. Bachmann, in: F.C. Neidhardt (Ed.), Escherichia coli and Salmonella: Cellular and Molecular Biology, Vol. 2, ASM Press, Washington, DC, 1996, pp. 2460–2488. [10] G. Zhang, E. Deng, L. Baugh, S.R. Kushner, Identification and characterization of Escherichia coli DNA helicase II mutants that exhibit increased unwinding efficiency, J. Bacteriol. 180 (1998) 377–387. [11] C.F. Brunk, K.C. Jones, T.W. James, Assay for nanogram quantities of DNA in cellular homogenates, Anal. Biochem. 92 (1979) 497–500. [12] I. Mellon, G. Champe, Products of DNA mismatch repair genes mutS and mutL are required for transcription-coupled nucleotide-excision repair of the lactose operon in Escherichia coli, Proc. Natl. Acad. Sci. U.S.A. 93 (1996) 1292–1297. [13] I. Mellon, P.C. Hanawalt, Induction of the Escherichia coli lactose operon selectively increases repair of its transcribed DNA strand, Nature 342 (1989) 95–98. [14] T. Mori, et al., Simultaneous establishment of monoclonal antibodies specific for either cyclobutane dimers and (6-4) photoproducts from the same mouse immunized with ultraviolet-irradiated DNA, Photochem. Photobiol. 54 (1991) 225–232. [15] B.K. Washburn, S.R. Kushner, Construction and analysis of deletions in the structural gene (uvrD) for DNA helicase II of Escherichia coli, J. Bacteriol. 173 (1991) 2569–2575. [16] H.A. Arthur, P.B. Eastlake, Transcriptional control of the uvrD gene of Escherichia coli, Gene 25 (1983) 309–316. [17] H.M. Arthur, D.R. Cavanagh, P.W. Finch, P.T. Emmerson, Regulation of the Escherichia coli uvrD gene in vivo, J. Bacteriol. 169 (1987) 3435–3440. [18] A.M. Easton, S.R. Kushner, Transcription of the uvrD gene of Escherichia coli is controlled by the LexA repressor and by attenuation, Nucleic Acids Res. 11 (1983) 8625–8640. [19] K. Kumura, M. Sekiguchi, Identification of the uvrD Gene Product of Escherichia coli As DNA Helicase II and Its Induction by DNA-damaging agents, J. Biol. Chem. 259 (1984) 1560–1565. [20] E.C. Siegel, The Escherichia coli uvrD gene is inducible by DNA damage, Mol. Gen. Genet. 191 (1983) 397–400. [21] E. Bertrand-Burggraf, P. Oertel, M. Schnarr, M. Daune, M. Granger-Schnarr, Effect of induction of SOS response on expression of pBR322 genes and on plasmid copy number, Plasmid 22 (1989) 163–168. [22] D.K. Orren, C.P. Selby, J.E. Hearst, A. Sancar, Post-incision steps of nucleotide excision repair in Escherichia coli. Disassembly of the UvrBC–DNA complex by helicase II and DNA polymerase I, J. Biol. Chem. 267 (1992) 780–788. [23] A.T. Yeung, W.B. Mattes, L. Grossman, Protein complexes formed during the incision reaction catalyzed by the Escherichia coli UvrABC endonuclease, Nucleic Acids Res. 14 (1986) 2567–2582. [24] I. Husain, B. Van Houten, D.C. Thomas, M. Abdel-Monem, A. Sancar, Effect of DNA polymerase I and DNA helicase II on the turnover rate of UvrABC excision nuclease, Proc. Natl. Acad. Sci. U.S.A. 82 (1985) 6774–6778.
D.J. Crowley, P.C. Hanawalt / Mutation Research 485 (2001) 319–329 [25] P.R. Caron, S.R. Kushner, L. Grossman, Involvement of helicase II (uvrD gene product) and DNA polymerase I in excision mediated by the UvrABC protein complex, Proc. Natl. Acad. Sci. U.S.A. 82 (1985) 4925–4929. [26] C.G. Lin, O. Kovalsky, L. Grossman, DNA damage-dependent recruitment of nucleotide excision repair and transcription proteins to the Escherichia coli inner membranes, Nucleic Acids Res. 25 (1997) 3151–3158. [27] D.L. Svoboda, C.A. Smith, J.S. Taylor, A. Sancar, Effect of sequence, adduct type, and opposing lesions on the binding and repair of ultraviolet photodamage by DNA photolyase and (A)BC excinuclease, J. Biol. Chem. 268 (1993) 10694–10700.
329
[28] R. Ben-Ishai, R. Sharon, in: K. Seeberg, E. Kleppe (Eds.), Chromosome Damage and Repair, Plenum Press, New York and London, 1981. [29] B. Van Houten, Nucleotide excision repair in Escherichia coli, Microbiol. Rev. 54 (1990) 18–51. [30] I. Wong, M. Amaratunga, T.M. Lohman, Heterodimer formation between Escherichia coli Rep and UvrD proteins, J. Biol. Chem. 268 (1993) 20386–20391. [31] J.W. Chase, W.E. Masker, J.B. Murphy, Pyrimidine dimer excision in Escherichia coli strains deficient in exonucleases V and VII and in the 5 → 3 exonuclease of DNA polymerase I, J. Bacteriol. 137 (1979) 234–242.
The SOS-dependent upregulation of uvrD is not required for efficient nucleotide excision repair of ultraviolet light induced DNA photoproducts in Escherichia coli David J. Crowley∗ , Philip C. Hanawalt Department of Biological Sciences, Stanford University, Stanford, CA 94305-5020, USA Received 18 September 2000; received in revised form 26 January 2001; accepted 29 January 2001
Abstract We have shown previously that induction of the SOS response is required for efficient nucleotide excision repair (NER) of the major ultraviolet light (UV) induced DNA lesion, the cyclobutane pyrimidine dimer (CPD), but not for repair of 6-4 photoproducts (6-4PP) or for transcription-coupled repair of CPDs [1]. We have proposed that the upregulation of cellular NER capacity occurs in the early stages of the SOS response and enhances the rate of repair of the abundant yet poorly recognized genomic CPDs. The expression of three NER genes, uvrA, uvrB, and uvrD, is upregulated as part of the SOS response. UvrD differs from the others in that it is not involved in lesion recognition but rather in promoting the post-incision steps of NER, including turnover of the UvrBC incision complex. Since uvrC is not induced during the SOS response, its turnover would seem to be of great importance in promoting efficient NER. Here we show that the constitutive level of UvrD is adequate for carrying out efficient NER of both CPDs and 6-4PPs. Thus, the upregulation of uvrA and uvrB genes during the SOS response is sufficient for inducible NER of CPDs. We also show that cells with a limited NER capacity, in this case due to deletion of the uvrD gene, repair 6-4PPs but cannot perform transcription-coupled repair of CPDs, indicating that the 6-4PP is a better substrate for NER than is a CPD targeted for transcription-coupled repair. © 2001 Elsevier Science B.V. All rights reserved. Keywords: Excision repair; UvrD; Helicase II; Transcription-coupled repair; SOS response; UV damage
1. Introduction The UvrD protein, also known as DNA helicase II, plays multiple roles in maintaining genomic stability in Escherichia coli. UvrD is a 3 → 5 DNA helicase that is required for methyl-directed mismatch repair [2] and been implicated in DNA replication and ∗ Corresponding author. Present address: Biology Department, College of Liberal Arts, Mercer University, 1400 Coleman Avenue, Macon, GA 31207, USA. Tel.: +1-478-301-2250; fax: +1-478-301-2802. E-mail address: crowley [email protected] (D.J. Crowley).
recombination [3]. In addition, UvrD is involved in the post-incision steps of nucleotide excision repair (NER), the versatile DNA repair process responsible for removal of a variety of lesions from DNA. Following lesion recognition and dual incisions mediated by the UvrA, UvrB, and UvrC proteins, UvrD promotes excision of the damaged oligonucleotide and releases UvrC protein from the DNA. DNA polymerase I subsequently fills in the resulting gap and dissociates UvrB from the DNA. DNA ligase completes the repair reaction by sealing the nick between the newly synthesized DNA patch and the parental DNA strand. UvrD-mediated turnover of UvrC and, subsequently,
0921-8777/01/$ – see front matter © 2001 Elsevier Science B.V. All rights reserved. PII: S 0 9 2 1 - 8 7 7 7 ( 0 1 ) 0 0 0 6 8 - 4
320
D.J. Crowley, P.C. Hanawalt / Mutation Research 485 (2001) 319–329
UvrB is necessary for promoting the catalytic rate of NER that characterizes this process in vivo [2]. Many mutant alleles of uvrD have been isolated and their pleiotropic phenotypes have been studied extensively. As would be expected for cells defective in any step of NER, uvrD mutants are UV sensitive. However, their sensitivity is not as great as that of uvrA, uvrB, or uvrC mutants that are completely defective in NER. Similarly, unlike uvrA, uvrB, or uvrC mutants, liquid holding recovery [4] and host cell reactivation [5,6] do occur in uvrD mutants, albeit to a reduced extent compared to that in wildtype cells. These observations suggest that uvrD mutants perform limited NER, a fact that was confirmed when repair of UV damage was measured in uvrD mutants [6,7]. The studies showed that the removal of thymine dimers, as measured by two dimensional chromatography, was greatly attenuated, but not eliminated, in uvrD mutants. We previously demonstrated that induction of the SOS response led to efficient overall NER of the major UV-induced lesion, the cyclobutane pyrimidine dimer (CPD), but had no effect on the rate of overall repair of 6-4 photoproducts (6-4PP) [1]. The subpathway of NER termed transcription-coupled repair, in which lesions in the transcribed strand of active genes are repaired in preference to those in the nontranscribed strand or genome overall, was also unaffected by SOS induction. To explain these results, we proposed that upregulation of the SOS proteins UvrA, UvrB, and/or UvrD enhanced cellular NER capacity and we showed that the kinetics of UvrA and UvrB upregulation correlated with efficient global NER of CPDs. Since removal of 6-4PPs and transcription-blocking CPDs were unaffected by this NER upregulation, we additionally proposed that these lesions are preferred targets for constitutive NER complexes, and thus, do not require upregulation of cellular repair capacity for their efficient repair. Based on the fact that uvrD mutants perform limited NER and that uvrD is an SOS-regulated gene induced with kinetics similar to those of the other SOS regulated NER genes, uvrA and uvrB [2,8], we designed experiments to test these hypotheses. First, we measured the rates and extents of overall repair of CPDs and 6-4PPs, and specific repair of transcription-blocking CPDs in uvrD mutants to determine if the latter two were preferred substrates for NER in cells in which the repair capacity was limited. We then asked whether
SOS-dependent upregulation of UvrD was required for achieving wildtype levels of overall NER of CPDs. To do this, we transformed a uvrD deletion mutant with a plasmid carrying the wildtype uvrD gene under control of a foreign promoter. This construct produces a level of UvrD protein similar to that present in uninduced wildtype cells, but the level is not increased following induction of the SOS response. We measured the rates and extents of repair of UV lesions from the genome overall and from the induced lactose operon to determine the role of UvrD, if any, in promoting SOS-dependent NER.
2. Materials and methods 2.1. Bacterial strains and plasmids The E. coli K-12 strain HL108 is a thyA deoC derivative of W3110 [9]. Strain HL952 is a derivative of HL108 that carries a uvrA::Tn10 mutant allele. Strain HL1030 is a derivative of HL108 and was the recipient of the uvrD294::kan allele by P1 transduction using strain SK4090 as a donor [10]. HL1031 and HL1032 are derivatives of HL1030 that were transduced with mutant alleles (Tn10 insertions) of uvrA and uvrC, respectively. HL1054 is a derivative of HL108 and was the recipient of the uvrD::tet allele from the donor strain GE1752 uvrD::tet [3]. HL1054 was selected by screening for tetracycline resistant transductants on LB plates supplemented with low concentrations of tetracycline (5 g/ml). HL1056 was made by transforming HL1054 with the uvrD+ plasmid pET9d-H2wt (kindly provided by S. Matson from UNC-Chapel Hill), selecting for kanamycin resistance, and screening for UV resistance. This plasmid contains the entire wildtype uvrD coding region under control of the T7 RNA polymerase promoter [3]. The transformation was performed by electroporation. 2.2. UV survival assay Cells were grown at 37◦ C in either Davis minimal salts supplemented with 0.4% glucose and 10 mg/ml thymine or in LB. Stationary phase cultures of approximately equal cell density were streaked on LB plates using sterile cotton swabs. Plates were irradiated with
D.J. Crowley, P.C. Hanawalt / Mutation Research 485 (2001) 319–329
254 nm ultraviolet light in incremental doses. Plates were incubated overnight at 37◦ C. 2.3. Cell growth and DNA preparation Cells were grown at 37◦ C in Davis minimal salts supplemented with 0.4% glucose and 10 mg/ml thymine. To label the DNA, cells were grown in medium containing 1 Ci/ml of 3 H-thymine in addition to the nonradioactive thymine. Cultures were grown to saturation and subcultured in fresh medium supplemented with 1 mM isopropyl -D-thiogalactoside (IPTG). Cultures were grown to mid log phase (approximately 3 × 108 cells per milliliter) as measured by OD600 . Cells were collected by filtration on 0.45 M Millipore filters, washed with prewarmed Davis medium, and resuspended in Davis medium containing 1 mM IPTG. Cells were UV-irradiated at the doses indicated and placed in a flask containing growth supplements. Samples of the culture were removed at various times and mixed with an equal volume of ice cold NET (100 mM NaCl, 10 mM Tris pH 8.0, 10 mM EDTA) buffer. Cells were pelleted by centrifugation at 4◦ C and resuspended in TE pH 8.0. Cells were lysed by addition of lysozyme (to 1 mg/ml) and RNaseA (to 100 g/ml) and incubation for 15 min at 37◦ C. Proteinase K (to 100 g/ml) and Sarkosyl (to 0.5%) were then added and the mixture was incubated at 50◦ C for 1 h. The DNA was extracted with phenol:chloroform and precipitated with 2.5 M ammonium acetate and two volumes of 95% ethanol. Purified DNA was resuspended in TE pH 8.0. A portion of each DNA sample was incubated with SstII and ApaI restriction enzymes according to manufacturers’ instructions (Gibco-BRL). The remaining DNA was lightly sonicated using a Branson sonifier and the concentration was determined by fluorometry using Hoechst 33258 [11]. The radioactivity in 3 H-labeled DNA was quantified by scintillation spectrometry. 2.4. Repair of CPDs in the lactose operon The frequency of CPDs in the individual strands of the lactose operon was determined using an established method [12,13]. DNA restricted with ApaI and SstII was quantified as described above and 300 ng of each sample was treated or mock-treated with T4
321
endonuclease V in NET buffer containing 1 mg/ml bovine serum albumin. Samples were electrophoresed overnight in alkaline gels containing 1% agarose. DNA was transferred to Hybond N+ membranes by Southern blotting and hybridized with 32 P -labeled probes. Strand-specific RNA probes were generated according to the protocol of Promega, using plasmid pZH10 [13] as a template. Detection was performed using a BioRad phosphorimager and its associated Molecular Analyst software. The frequency of CPDs per 6.6 kb restriction fragment was calculated from the percentage of fragments with no CPDs (zero class) using the Poisson expression (−ln of the zero class = average number of dimers per fragment). 2.5. Immunoassay for overall NER The repair of CPDs and 6-4PPs was measured using an immunoassay. Following denaturation by boiling, 50 ng (CPD) or 500 ng (6-4PP) of each DNA sample (irradiated with 40 J/m2 ) were loaded in triplicate onto a Hybond N+ membrane using a slot blot apparatus. When lower doses of UV were used, higher amounts of DNA were loaded onto the membranes. At 8 J/m2 , 250 ng (CPD) of each sample were loaded; at 4 J/m2 , 500 ng (CPD) of each sample were loaded. The membrane was incubated for 2 h in the presence of mouse antibodies against either CPDs (TDM-2) or 6-4PPs (64M-2) diluted 1:2000 in PBS (antibodies were a generous gift of Mori et al. [14]. Horseradish peroxidase-conjugated secondary antibodies were used at a dilution of 1:5000 and detected using the enhanced chemiluminescence detection system (Amersham) and subsequent phosphorimager analysis (BioRad). Following detection, the amount of 3 H-labeled DNA loaded in each slot was confirmed by scintillation counting.
3. Results To study the effect of a complete loss of UvrD activity on the repair of UV-induced photoproducts, we constructed strain HL1030, which carried the uvrD294::kan allele in an otherwise repair proficient background [10]. Cells with mutant alleles of uvrD, including similar deletion mutants, have been shown to be significantly UV sensitive [15]. The level of UV
322
D.J. Crowley, P.C. Hanawalt / Mutation Research 485 (2001) 319–329
Fig. 1. The moderate UV sensitivity of uvrD mutants is due to low levels of NER. Cultures of the wildtype strain HL108 (right) and its isogenic mutants HL952 (uvrA, second from right), HL1030 (uvrD, second from left), and HL1031 (uvrA uvrD, left) were streaked on an LB plate and irradiated in increments with the indicated doses of 254 nm UV.
sensitivity was not as great as that produced in cells defective in either uvrA, uvrB, or uvrC [5,6], supporting other studies that showed that uvrD mutants could perform limited excision repair of UV-induced damage in vivo [6,7]. We confirmed the intermediate UV sensitivity of uvrD mutants using a UV survival plate assay and comparisons between isogenic wildtype, uvrD, uvrA, and uvrA uvrD mutant strains (Fig. 1). The uvrD mutant was significantly more UV sensitive than the wildtype strain, which showed no detectable loss of viability up to 40 J/m2 UV. However, the uvrD mutant was much more resistant than the isogenic uvrA mutant, with survivors apparent after treatment with up to 16 J/m2 UV. Transduction of a uvrA mutant allele (or a uvrC mutant allele, data not shown) into the uvrD deletion mutant eliminated its resistance, demonstrating
that this resistance is likely due to a low cellular capacity for Uvr-dependent excision repair of UV damage. The residual UvrABC-dependent excision that occurs in uvrD mutants has been reported in earlier papers to be repair of CPDs [6,7]. These repair studies were done by measuring loss of thymine dimer content over time by two-dimensional paper chromatography, a rather insensitive method that not only measures CPD content but also, presumably, 6-4PP content. We sought to determine more carefully the rate and extent of removal of both CPDs and 6-4PPs from the genome overall using antibodies raised against the two respective lesions (Fig. 2). Using a dose of 40 J/m2 , we found that only 6-4PPs were removed from the genome, albeit at a greatly reduced rate compared to that in the isogenic uvrD+ strain (see Fig. 5B for uvrD+ rate). No detectable repair of CPDs occurred up to 90 min after UV (data not shown). Irradiation with 40 J/m2 UV induces approximately 4000 CPDs and 1000 6-4PPs in the DNA of an exponentially growing cell with two chromosomes [1], meaning that about 550 6-4PPs were removed in the first 60 min after UV in the uvrD mutants, at what appeared to be a constant rate (Fig. 2). No repair of either CPDs or 6-4PPs occurred in the uvrA uvrD or uvrC uvrD double mutants (data not shown), confirming that the limited repair in the uvrD mutants was mediated by UvrABC proteins. Although we found no detectable repair of CPDs in the genome overall at a dose of 40 J/m2 UV, we wanted to see whether transcription-coupled repair of these lesions occurred in the uvrD mutants at this dose. Our previous work had shown that 6-4PPs and transcription-blocking CPDs (i.e. substrates for transcription-coupled repair) are preferred substrates for NER in cells with a limited repair capacity [1]. Therefore, we felt it was possible that transcription-coupled repair of CPDs might be occurring in the uvrD mutants at a rate comparable to that of the overall removal of 6-4PPs and that the limited number of total CPDs repaired in this manner were undetectable in the antibody assay due to the high background of unrepaired genomic CPDs. Surprisingly, we found that this was not the case, as both the nontranscribed and transcribed strands of the induced lactose operon showed little or no loss of T4 endonuclease sensitive sites up to 60 min after UV (Fig. 3), mirroring the results from the CPD antibody assay
D.J. Crowley, P.C. Hanawalt / Mutation Research 485 (2001) 319–329
323
Fig. 2. The 6-4PPs, but not the CPDs, are repaired in uvrD mutants irradiated with 40 J/m2 UV. (A) Typical immunoblots generated from experiments using anti-CPD (left, squares) or anti-6-4PP (right, diamonds) monoclonal antibodies on DNA isolated from strain HL1030 uvrD at the indicated time points after irradiation with 40 J/m2 UV. (B) Graphical representation of data generated from multiple immunoblots of DNA from three independent experiments. Each error bar represents one standard deviation calculated from the averages of the independent experiments. Error bars not shown are obscured by the data point. Symbols as in part A.
(Fig. 2). We conclude that 6-4PPs are the preferred substrate for repair in uvrD mutants, which have a very low capacity for NER. Since a dose of 40 J/m2 was lethal to the uvrD deletion mutants (Fig. 1), it was clear that this level of UV damage saturated the limited repair capacity of these cells. We measured the removal of CPDs from the genome of uvrD deletion mutants irradiated with
either 8 or 4 J/m2 UV, doses at which we detected significant survival by our plating assay (Fig. 1). At 8 J/m2 , 800 CPDs are induced in the DNA (and 200 6-4PPs, which cannot be detected using the antibody assay) and approximately half were removed 1 h after UV (Fig. 4). Therefore, about 400 CPDs were removed from these cells in the first hour after UV. At 4 J/m2 , 400 CPDs are induced in each cell (about
324
D.J. Crowley, P.C. Hanawalt / Mutation Research 485 (2001) 319–329
Fig. 3. Transcription-coupled repair of CPDs is not detectable in uvrD mutants irradiated with 40 J/m2 UV. (A) Southern blots showing maintenance of T4 endonuclease V (T4 endo V) sensitive sites over time in the two strands of the IPTG-induced lactose operon from HL1030 uvrD mutants. The average number of CPDs was measured in each strand of the 6.6 kb restriction fragment in DNA isolated at the indicated times after irradiation with 40 J/m2 UV. (B) Graphical representation of data collected from Southern blots. Each point represents the average repair calculated from three independent biological experiments. (䉭) Transcribed strand; (䊊) nontranscribed strand.
100 6-4PPs) and after 40 min, about 70% had been removed from the genome. No signal was detected after 60 min, indicating that over 90% of the initial 400 CPDs were removed within the first hour after UV. Again, no repair was detected at either dose when the double mutants uvrA uvrD or uvrC uvrD were used (data not shown). Combining the results from the three different doses, and assuming that overall repair of 6-4PPs was complete in 1 h at the lower doses, we found that 500–600 lesions were repaired (in a UvrA- and UvrC-dependent manner) in uvrD deletion
mutants in 1 h regardless of the initial level of DNA damage. UvrD is upregulated as part of the SOS response along with the lesion recognition proteins, UvrA and UvrB. We had previously shown that the SOS-dependent upregulation of UvrA and UvrB correlated with efficient global NER of CPDs [1]. UvrD is upregulated three- to six-fold after UV [16–20] and its induction occurs with kinetics similar to those of UvrA and UvrB, based on similarities in the strength of LexA repression and the SOS box sequence [8]. To
D.J. Crowley, P.C. Hanawalt / Mutation Research 485 (2001) 319–329
325
Fig. 4. Repair of CPDs is detectable in uvrD mutants irradiated with low UV doses. Monoclonal antibodies specific for CPDs were used in an immunoassay with DNA isolated from HL1030 uvrD mutants after irradiation with 40 J/m2 ((䊐) from Fig. 2), 8 J/m2 (䉫), or 4 J/m2 (䊊). The 8 and 4 J/m2 points represent the average repair calculated from at least two immunoblots of samples from a single experiment.
test whether the SOS-dependent induction of UvrD was necessary for efficient global repair of CPDs, we transformed a uvrD deletion strain with a plasmid carrying the wildtype coding region of uvrD under control of the T7 phage promoter. This construct has been shown by direct immunological methods to produce approximately 350 UvrD molecules in cells that do not possess T7 RNA polymerase, just slightly below the wildtype level of 500 UvrD molecules per uninduced cell [3]. Since the uvrD gene in this construct is not under LexA control, the basal level of UvrD should be maintained in cells that are otherwise proficient in SOS induction. We considered the possibility that this plasmid, which is derived from pBR322, increases in copy number as part of the SOS response [21]. An increase in plasmid copy number after UV could mimic SOS upregulation of UvrD despite the lack of direct SOS induction of the plasmid borne uvrD+ gene. However, we have observed no significant change in plasmid copy number up to 60 minutes after UV treatment in these transformants (data not shown). Fig. 5A shows results from a UV survival plating assay similar to that in Fig. 1. The uvrD deletion in this case was marked by a tetracycline resistance element but retained the same level of UV sensitivity
Fig. 5. An SOS noninducible uvrD+ allele complements the UV sensitivity of a uvrD deletion mutant and facilitates efficient NER. (A) Cultures of the wildtype strain HL108 (top), the isogenic uvrD mutant HL1054 (middle), and the uvrD+ transformant HL1056 (bottom) were streaked on an LB plate and irradiated in increments with the indicated doses of 254 nm UV. (B) Monoclonal antibodies specific for CPDs (䊐) or 6-4PPs (䉫) were used in an immunoassay with DNA isolated from HL1056 cells (solid lines and symbols) after irradiation with 40 J/m2 . Dotted lines and empty symbols represent data from the isogenic wildtype strain HL108 [1] and are included for comparison. (C) Quantitative southern hybridization of strand-specific RNA probes to the IPTG-induced lactose operon was performed on DNA isolated from HL1056 cells (solid lines and symbols). The average number of CPDs was measured in each strand of the 6.6 kb restriction fragment in DNA isolated at the indicated times after UV irradiation with 40 J/m2 . (䊊) Transcribed strand; (䉭) nontranscribed strand. Dotted lines and empty symbols represent data from similar experiments using the isogenic wildtype strain HL108 [1] and are included for comparison.
326
D.J. Crowley, P.C. Hanawalt / Mutation Research 485 (2001) 319–329
previously demonstrated using the uvrD kanamycin deletion (Fig. 1). When the uvrD+ plasmid was transfected into these cells, UV resistance was enhanced greatly, although the transformants remained slightly more UV sensitive than the isogenic wildtype cells (compare growth at doses of 53, 66, and 80 J/m2 ). Overall NER (Fig. 5B) and transcription-coupled repair of CPDs (Fig. 5C) were measured in the transformants and the results are presented with the isogenic wildtype repair profiles for comparison [1]. The transformants showed a slightly lower rate of removal of both 6-4PPs and CPDs, but did achieve wildtype levels of repair 40 min after UV (Fig. 5B). The rate and extent of CPD repair in the transformants were greater than those of lexA3(Ind-) mutants that are unable to induce the SOS response and upregulate NER capacity [1]. The transformants were also able to perform transcription-coupled repair of CPDs, although both the transcribed and nontranscribed strands of the induced lactose operon showed an initial lag in repair of CPDs compared to that in wildtype cells (Fig. 5C). Twenty minutes after UV, the repair of CPDs in both strands corresponded with those of the wildtype. Consistent with our immunoassay results (Fig. 5B), the rate of repair of the nontranscribed strand, which is an indicator of the global repair rate [1], was similar to that of wildtype cells.
4. Discussion Our previous work demonstrated that induction of the SOS response in E. coli enhances NER capacity [1]. This enhanced NER capacity is necessary for efficient global repair of the principal UV-induced lesion, the CPD, but is not needed for the rapid rate of global repair of 6-4PPs or the efficient repair of CPDs in the transcribed strand of an active gene. In this study, we have continued to explore the relationship between the induction of the SOS response and the efficiency of NER using mutants deleted for the SOS-inducible NER gene, uvrD. We asked two questions: (1) what are the targets for the limited NER that occurs in uvrD mutants, and (2) is UvrD upregulation necessary for promoting wildtype levels of NER, specifically for the overall repair of CPDs?
4.1. Limited NER in uvrD mutants is targeted to 6-4PPs Based on our previous studies, we predicted that cells with a limited capacity for NER would target 6-4PPs and transcription-blocking CPDs for repair, limiting the removal of CPDs in nontranscribed regions of the genome [1]. Since uvrD mutants were reported to have a low level of NER capacity [6,7], we felt that we could use these cells to test this hypothesis. Using a dose of 40 J/m2 , which induces approximately 4000 CPDs and 1000 6-4PPs in the DNA of a log phase bacterium [1], we found that in the first 90 min after UV, only 6-4PPs were detectably removed from the uvrD deletion mutants (Fig. 2). No repair of CPDs was detected in the genome overall or in the nontranscribed strand of the induced lactose operon (Figs. 2 and 3). Perhaps a very small fraction of CPDs located in the transcribed strand of the lactose operon were removed within the first hour after UV (Fig. 3). The lack of significant repair of CPDs in the induced lactose operon of the uvrD deletion mutant is consistent with the results of another study [12]. Based on the number of lesions induced in a cell and the amount of repair calculated from our assays, we estimated that roughly 550 6-4PPs and perhaps a small number of CPDs were removed from the DNA of the uvrD deletion mutants after a dose of 40 J/m2 UV. These results clearly show that a 6-4PP is a better substrate for NER than is a transcription-blocking CPD, illustrating that the recognition and repair of 6-4PPs is even more efficient than transcription-coupled repair of CPDs in vivo. Two lower doses were employed to test whether global repair of CPDs could occur in uvrD mutants. Indeed, CPDs were removed at both 8 J/m2 (which produces 800 CPDs and 200 6-4PPs) and 4 J/m2 (which produces 400 CPDs and 80 6-4PPs) (Fig. 4). Interestingly, the number of lesions removed in both cases (assuming complete removal of 6-4PPs) was equivalent to the 500–600 lesions we calculated to be removed from the uvrD cells in the first hour after irradiation with 40 J/m2 . The fact that a fixed number of DNA lesions was removed in these cells independent of UV dose is not surprising given the absolute requirement for UvrD in promoting the turnover of UvrC in vitro. Biochemical studies have shown that only a limited number of incisions can take place when limiting concentrations of UvrC are present in reactions with-
D.J. Crowley, P.C. Hanawalt / Mutation Research 485 (2001) 319–329
out UvrD [22]. Only when UvrD is added to these reactions does incision proceed catalytically [19,23–25]. Based on these biochemical results, the number of UvrC molecules present in the cell should determine the amount of incision that could take place in cells without functional UvrD. Based on our data, we would, thus, expect that approximately 500–600 molecules of UvrC were present in each cell, a number that is in agreement with a recent immunological determination of 500 UvrC molecules per cell [26]. If approximately 500 UvrC molecules are indeed present in each cell, we explain our 40 J/m2 repair results as follows. Since approximately 400 UvrB molecules are present in each cell [1] and UvrA loading of UvrB at damage sites is catalytic [22], we propose that most of the 400 UvrB molecules are rapidly loaded at 6-4PPs, a favored NER substrate [27], by UvrA dimers shortly after irradiation. Available UvrC molecules then bind the UvrB–DNA complexes and incisions are made rapidly [28]. As determined both in vitro [19,23,24] and in vivo [7,28], UvrBC-mediated incisions persist in the absence of functional UvrD, due to the inability of cells to promote UvrC turnover, excision, and DNA polymerase I gap filling. Why then do we detect loss of these lesions if excision has not taken place? We considered that perhaps we only see “repair” of these lesions when we purify the DNA and denature it prior to loading it onto slot blots. The incised, lesion-containing oligonucleotides may be too small to be retained on a membrane or may be lost during phenol extraction, making them undetectable once we assay with lesion-specific antibodies. However, given that incision proceeds at a normal rate initially in uvrD mutants [7,28], it is difficult to understand why, if this hypothesis is true, we did not observe a rapid “repair” soon after UV that plateaus at later time points. A more likely explanation is that excision and perhaps UvrBC turnover does take place in uvrD mutants, but at greatly reduced rates. It has been shown that DNA polymerase I will promote excision of the damaged oligonucleotide and complete repair in the absence of UvrD in vitro [29]. If this is true in vivo, then the rate of DNA polymerase I gap filling in the absence of UvrD may determine the slow rate of repair we measured and may explain why repair does not plateau shortly after UV. This slow turnover may also explain why more lesions were removed (i.e. 550–600) than there were UvrB or UvrC molecules in the cell.
327
If the more widely cited number of 10–20 UvrC molecules per cell [2] is correct, there must be some turnover of these molecules to account for the levels of repair we observed in the uvrD mutants. Perhaps other cellular factors are able to provide alternative or “back-up” mechanisms for promoting excision and turnover of UvrBC in the absence of UvrD. One obvious candidate is the Rep helicase, which can form a heterodimer with UvrD that may play a role in DNA replication [30]. However, it has been shown that Rep cannot effectively substitute for UvrD function in NER in vitro [24]. Perhaps one of the nine remaining helicases in E. coli can perform the function of UvrD to some extent and help promote catalytic NER in vivo. Another candidate for the post-incision steps of NER is exonuclease VII, considering that a secondary xseA mutation increased the excision deficiency of a polAex mutant strain [31]. Regardless of the factor(s) involved, the number of lesions removed and the constant rate of their repair suggests that some turnover of NER proteins is likely to occur in the absence of functional UvrD in vivo. 4.2. SOS-dependent upregulation of UvrD is not required for efficient overall NER Three NER proteins are upregulated as part of the SOS response: UvrA, UvrB, and UvrD. We had proposed that the rapid SOS-dependent upregulation of UvrA and UvrB after UV was necessary for promoting efficient overall NER of the abundant, yet poorly recognized, CPDs [1]. In this study, we attempted to test whether UvrD induction was also necessary for the enhanced repair capacity we have observed in SOS proficient cells. We introduced a plasmid carrying an SOS-noninducible uvrD+ allele into a uvrD deletion mutant and measured overall NER of 6-4PPs and CPDs as well as transcription-coupled repair of CPDs and compared the results to wildtype cells (Fig. 5). Our results indicate that the uvrD (UvrD+ ) cells were nearly as efficient as wildtype cells in repairing both UV lesions, indicating that SOS upregulation of UvrD levels is not important for promoting the SOS-dependent repair of CPDs. Previous studies using this plasmid have shown that it produces slightly lower than wildtype levels of UvrD in transformed cells [3]. This may account for the initial lag in repair of CPDs detected in both strands
328
D.J. Crowley, P.C. Hanawalt / Mutation Research 485 (2001) 319–329
of the lactose operon and the slight UV sensitivity of the transformants compared to the isogenic wildtype (Fig. 5A). Since repair is restored to essentially wildtype levels by 20 min after UV, we believe that the upregulation of SOS proteins, likely UvrA and/or UvrB, promotes efficient repair and that basal levels of UvrD, expressed from a plasmid or from the chromosome, are sufficient to support this rate of NER.
Acknowledgements The authors would like to thank Ann Ganesan and C. Allen Smith for helpful discussions and critical reading of this manuscript. We would also like to thank Steve Matson for his generous gifts of the uvrD+ plasmid and deletion strains and Toshio Mori for his gift of CPD and 6-4PP antibodies. This work was supported by a Cellular and Molecular Biology traineeship GM07276 and Outstanding Investigator Grant CA44349 from the National Cancer Institute, NIH.
References [1] D.J. Crowley, P.C. Hanawalt, Induction of the SOS response increases the efficiency of global nucleotide excision repair of cyclobutane pyrimidine dimers, but not 6-4 photoproducts, in UV-irradiated Escherichia coli, J. Bacteriol. 180 (1998) 3345–3352. [2] E. Friedberg, G.C. Walker, W. Siede, DNA Repair and Mutagenesis, American Society of Microbiology, Washington, DC, 1995. [3] J.W. George, R.M. Brosh Jr., S.W. Matson, A dominant negative allele of the Escherichia coli uvrD gene encoding DNA helicase II. A biochemical and genetic characterization, J. Mol. Biol. 235 (1994) 424–435. [4] M.S. Tang, K.C. Smith, The effects of lexA101, recB21, recF143 and uvrD3 mutations on liquid-holding recovery in ultraviolet-irradiated Escherichia coli K12 recA56, Mutat. Res. 80 (1981) 15–25. [5] R.H. Rothman, A.J. Clark, Defective excision and postreplication repair of UV-damaged DNA in a recL mutant strain of E. coli K-12, Mol. Gen. Genet. 155 (1977) 267–277. [6] N.B. Kuemmerle, W.E. Masker, Effect of the uvrD mutation on excision repair, J. Bacteriol. 142 (1980) 535–546. [7] C.A. Van Sluis, I.E. Mattern, M.C. Paterson, Properties of uvrE mutants of Escherichia coli K12.I. Effects of UV irradiation on DNA metabolism, Mutat. Res. 25 (1974) 273–279.
[8] G.C. Walker, in: F.C. Neidhardt (Ed.), Escherichia coli and Salmonella: Cellular and Molecular Biology, Vol. 1, ASM Press, Washington, DC, 1996, pp. 1400–1416. [9] B.J. Bachmann, in: F.C. Neidhardt (Ed.), Escherichia coli and Salmonella: Cellular and Molecular Biology, Vol. 2, ASM Press, Washington, DC, 1996, pp. 2460–2488. [10] G. Zhang, E. Deng, L. Baugh, S.R. Kushner, Identification and characterization of Escherichia coli DNA helicase II mutants that exhibit increased unwinding efficiency, J. Bacteriol. 180 (1998) 377–387. [11] C.F. Brunk, K.C. Jones, T.W. James, Assay for nanogram quantities of DNA in cellular homogenates, Anal. Biochem. 92 (1979) 497–500. [12] I. Mellon, G. Champe, Products of DNA mismatch repair genes mutS and mutL are required for transcription-coupled nucleotide-excision repair of the lactose operon in Escherichia coli, Proc. Natl. Acad. Sci. U.S.A. 93 (1996) 1292–1297. [13] I. Mellon, P.C. Hanawalt, Induction of the Escherichia coli lactose operon selectively increases repair of its transcribed DNA strand, Nature 342 (1989) 95–98. [14] T. Mori, et al., Simultaneous establishment of monoclonal antibodies specific for either cyclobutane dimers and (6-4) photoproducts from the same mouse immunized with ultraviolet-irradiated DNA, Photochem. Photobiol. 54 (1991) 225–232. [15] B.K. Washburn, S.R. Kushner, Construction and analysis of deletions in the structural gene (uvrD) for DNA helicase II of Escherichia coli, J. Bacteriol. 173 (1991) 2569–2575. [16] H.A. Arthur, P.B. Eastlake, Transcriptional control of the uvrD gene of Escherichia coli, Gene 25 (1983) 309–316. [17] H.M. Arthur, D.R. Cavanagh, P.W. Finch, P.T. Emmerson, Regulation of the Escherichia coli uvrD gene in vivo, J. Bacteriol. 169 (1987) 3435–3440. [18] A.M. Easton, S.R. Kushner, Transcription of the uvrD gene of Escherichia coli is controlled by the LexA repressor and by attenuation, Nucleic Acids Res. 11 (1983) 8625–8640. [19] K. Kumura, M. Sekiguchi, Identification of the uvrD Gene Product of Escherichia coli As DNA Helicase II and Its Induction by DNA-damaging agents, J. Biol. Chem. 259 (1984) 1560–1565. [20] E.C. Siegel, The Escherichia coli uvrD gene is inducible by DNA damage, Mol. Gen. Genet. 191 (1983) 397–400. [21] E. Bertrand-Burggraf, P. Oertel, M. Schnarr, M. Daune, M. Granger-Schnarr, Effect of induction of SOS response on expression of pBR322 genes and on plasmid copy number, Plasmid 22 (1989) 163–168. [22] D.K. Orren, C.P. Selby, J.E. Hearst, A. Sancar, Post-incision steps of nucleotide excision repair in Escherichia coli. Disassembly of the UvrBC–DNA complex by helicase II and DNA polymerase I, J. Biol. Chem. 267 (1992) 780–788. [23] A.T. Yeung, W.B. Mattes, L. Grossman, Protein complexes formed during the incision reaction catalyzed by the Escherichia coli UvrABC endonuclease, Nucleic Acids Res. 14 (1986) 2567–2582. [24] I. Husain, B. Van Houten, D.C. Thomas, M. Abdel-Monem, A. Sancar, Effect of DNA polymerase I and DNA helicase II on the turnover rate of UvrABC excision nuclease, Proc. Natl. Acad. Sci. U.S.A. 82 (1985) 6774–6778.
D.J. Crowley, P.C. Hanawalt / Mutation Research 485 (2001) 319–329 [25] P.R. Caron, S.R. Kushner, L. Grossman, Involvement of helicase II (uvrD gene product) and DNA polymerase I in excision mediated by the UvrABC protein complex, Proc. Natl. Acad. Sci. U.S.A. 82 (1985) 4925–4929. [26] C.G. Lin, O. Kovalsky, L. Grossman, DNA damage-dependent recruitment of nucleotide excision repair and transcription proteins to the Escherichia coli inner membranes, Nucleic Acids Res. 25 (1997) 3151–3158. [27] D.L. Svoboda, C.A. Smith, J.S. Taylor, A. Sancar, Effect of sequence, adduct type, and opposing lesions on the binding and repair of ultraviolet photodamage by DNA photolyase and (A)BC excinuclease, J. Biol. Chem. 268 (1993) 10694–10700.
329
[28] R. Ben-Ishai, R. Sharon, in: K. Seeberg, E. Kleppe (Eds.), Chromosome Damage and Repair, Plenum Press, New York and London, 1981. [29] B. Van Houten, Nucleotide excision repair in Escherichia coli, Microbiol. Rev. 54 (1990) 18–51. [30] I. Wong, M. Amaratunga, T.M. Lohman, Heterodimer formation between Escherichia coli Rep and UvrD proteins, J. Biol. Chem. 268 (1993) 20386–20391. [31] J.W. Chase, W.E. Masker, J.B. Murphy, Pyrimidine dimer excision in Escherichia coli strains deficient in exonucleases V and VII and in the 5 → 3 exonuclease of DNA polymerase I, J. Bacteriol. 137 (1979) 234–242.