Transcription-coupled repair is inducible in hamster cells

Abstract

In mammalian cells, the rate of nucleotide excision repair of UV dimers is heterogeneous throughout the genome, with repair occurring more rapidly in the transcribed strand of active genes than in the genome overall. This repair pathway is termed transcription-coupled repair (TCR) and is thought to permit the rapid resumption of RNA synthesis following UV irradiation. To evaluate the inducibility of the TCR process, we examined the repair of UV-induced cyclobutane pyrimidine dimers (CPDs) at the level of the gene following exposure of hamster cells to a sub-lethal UV fluence, 3 h prior to a higher dose. Repair was detected by a well-established technique allowing quantification of CPDs at the level of a specific strand by Southern blot hybridization. Here, we show that prior low-dose irradiation clearly enhanced the early rate of CPD removal in the transcribed strand of the active DHFR gene. Furthermore, the RNA synthesis recovery following UV exposure was stimulated by the priming UV dose. Thus, we provide evidence for an inducible TCR response to CPDs in hamster cells. This pathway is independent of the p53 activation, since the hamster cell line that we used expresses high levels of mutant p53 protein.

INTRODUCTION

DNA is constantly exposed to damaging agents from a wide variety of endogenous and exogenous sources. Several highly conserved enzymatic pathways have evolved to repair these different lesions. The most versatile and ubiquitous mechanism for DNA repair is the nucleotide excision repair (NER) pathway by which mammalian cells remove bulky lesions, which include UV-induced cyclobutane pyrimidine dimers (CPDs) and 6-4 pyrimidine–pyrimidone photoproducts (for a review see 1). The rate of repair of CPDs is heterogeneous throughout the genome, with CPDs being more efficiently removed from actively transcribed genes than from non-coding regions (2). In particular, the transcribed strand (TS) of an active gene is repaired more rapidly than the non-transcribed strand (NTS) (3). This repair process has been termed transcription-coupled repair (TCR) and is thought to permit the rapid resumption of RNA synthesis following UV irradiation (4).

Until recently, it was commonly accepted that the mammalian excision repair system occurs constitutively. In particular, the existence of an inducible SOS response well characterized in bacteria has not been demonstrated in eukaryotic cells. However, reports have suggested that NER is inducible in mammalian cells. It has been shown that carcinogen treatment enhances the excision repair capacity of repair-proficient mammalian cells, as assessed by host cell reactivation (HCR) assay (5). Also, cell extracts from UV-irradiated human cells exhibit greater NER extent than extracts from unirradiated cells in an in vitro assay using damaged plasmid DNA (6). Pretreatment with thymidine dinucleotides of UV-irradiated human skin cells leads to an enhancement of DNA repair as measured by both HCR assay and immunoassay to determine the repair rate of CPDs (7). Furthermore, other findings using HCR assays have suggested that TCR may be inducible (8,9). However, all these results are based on the assumption that enhanced reactivation of a virus or a reporter gene is representative of cellular mechanisms. A recent report has indeed addressed this question and showed that NER plays an important role in determining the level of expression of a UV-irradiated reporter gene introduced into unirradiated cells but did not provide evidence for TCR at the level of this reporter gene (10). This suggests that biochemical reactions involved in the HCR assay are not fully identical to those responsible for TCR.

Since no previous work clearly demonstrated that TCR is inducible in mammalian cells, we tested this hypothesis in hamster cells by monitoring removal of CPDs from endogenous genes. At the level of the genome, TCR is usually assessed by quantifying the removal of CPDs from the TS of an active gene using Southern hybridization of DNA treated with T4 endonuclease after separation on alkaline agarose gels. This method was initially used to demonstrate gene- and strand-specific repair of CPDs from the active DHFR gene in mammalian cells (2,3). To evaluate the inducibility of the TCR pathway, we examined the repair of CPDs from the DHFR gene following exposure of hamster cells to a sub-lethal UV fluence, 3 h prior to a higher UV dose. Here, we provide evidence for an inducible TCR response to CPDs in hamster cells.

MATERIALS AND METHODS

Cell culture and UV irradiation

The methotrexate-resistant Chinese hamster cell line CHO K1 B11 containing a 50-fold amplification of the DHFR gene (11) was grown in Ham’s F12 medium without glycine, hypoxanthine and thymidine, supplemented with 10% fetal bovine serum, 2 mM glutamine and penicillin–streptomycin in humidified 5% CO₂, 95% air at 37°C. Actively growing cells were used in all experiments. Irradiation of monolayer cells was performed without medium using a germicidal light (254 nm). Induced cells were first irradiated with 2 J/m² in PBS and then incubated in fresh supplemented medium for 3 h, prior to the second irradiation with 20 J/m². Uninduced cells were only irradiated at 20 J/m².

DNA repair assays

The method used to measure repair of CPDs in the DHFR gene was as described previously by Bohr and Okumoto (12). The initial level of damage (0 h) was determined immediately after the UV irradiation at 20 J/m². DNA repair was measured after different lengths of time (4, 8, 18 and 24 h) after damage. High molecular weight DNA was purified and restricted with KpnI restriction endonuclease. Detection of CPDs with T4 endonuclease V treatment was assessed in non-replicated DNA. Two 3 µg-aliquots of DNA from each repair time-point were treated in parallel, one with T4 endonuclease and the other mock-treated with endonuclease buffer alone for 30 min at 37°C. DNA samples were loaded onto 0.6% alkaline agarose gels and electrophoresed at 30 V for 16–18 h with buffer circulation. The DNA was transferred to nylon membrane (Boehringer Mannheim) in 0.4 N NaOH. Membranes were then hybridized with the genomic DNA probe pMB5, that recognizes a 14 kb fragment comprising the 5′ half of the DHFR gene. PMB5 was labeled with digoxigenin–dUTP, using a random primed DNA labeling kit (Boehringer Mannheim). Hybrids were detected by enzyme-linked immunoassay using an antibody-conjugate and subsequent enzyme-catalyzed color reaction (DIG Nucleic Acid Detection kit, Boehringer Mannheim). Hybridization to the fragment of interest was quantified by scanning densitometry of Southern blots, using the ImageQuaNT‘ analysis software (Molecular Dynamics).

The method used to measure strand-specific repair of CPDs in the DHFR gene was as described previously (3). The initial level of damage (0 h) was determined immediately after the 20 J/m² UV irradiation. Strand-specific repair was measured at different incubation times (2, 4, 8 and 24 h) after damage. Membranes were hybridized with strand-specific riboprobes, which detect either the TS or the NTS of the 14 kb KpnI fragment of the 5′ region of the DHFR gene (13). The riboprobes were labeled with digoxigenin-11-UTP, using a SP6/T7 Transcription kit (Boehringer Mannheim).

RNA synthesis recovery assay

The method used to measure RNA synthesis recovery was as described by Mayne and Lehmann (14) and by Troelstra et al. (15). Twenty-four hours before irradiation, cells were labeled with [¹⁴C]thymidine (0.025 µCi/mmol). At various times after UV irradiation, the medium was replaced by medium containing 5 µCi/ml [³H]uridine and cells were incubated for 15 min at 37°C. At the end of the labeling period, cells were washed, scraped from the dishes with 0.2 N NaOH and spotted onto GF/C Whatman glass filters soaked in 10% TCA. Filters were then washed with 96% EtOH, dried and counted in a liquid scintillation counter. ¹⁴C counts served as an internal control for the number of cells present. The measure of the RNA synthesis was taken from the ratio [³H] UV-irradiated cells/[³H] control cells.

RESULTS

Inducibility of CPDs repair in the DHFR gene in hamster cells

To determine whether repair of CPDs is inducible, we first compared gene-specific repair in UV-irradiated CHO B11 cells, either subjected or not to a previous sub-lethal UV dose. In our study, B11 cells were typically irradiated either with 20 J/m² in one dose or with two doses, one of 2 J/m² and, 3 h after, one of 20 J/m². Removal of CPDs at the level of the active DHFR gene was monitored over a 24-h period (Fig. 1). In this assay, the probe detects both strands of the DHFR gene and the quantity represents an average of the lesions in the two strands. Prior low-dose irradiation did not significantly change the initial pyrimidine dimer frequency (about 1.9 CPD/14 kb dhfr fragment), but it clearly enhanced the early rate of CPD removal at time-points 4 and 8 h (Fig. 1). The time required for 50% dimer removal was reduced from 12 to 7 h by prior UV irradiation of hamster cells. However, the extent of repair at 24 h was similar in induced and uninduced cells (Fig. 1). Thus, our results suggest that CPD repair is inducible in hamster cells.

Figure 1.

Enhancement in repair of UV-induced CPDs in an active gene by pre-treatment with a low UV dose. (Top) Representative Southern blots of KpnI genomic fragments isolated from UV-irradiated induced and uninduced CHO B11 cells, not treated (–) or treated (+) with T4 endonuclease V and probed with DIG-labeled DHFR probe are shown. Cont, non-irradiated cells. (Bottom) Data obtained from scanning densitometry quantitation of Southern blots using the ImageQuaNT™ analysis software (Molecular Dynamics) are presented as mean values ± S.E. from five biological experiments with two to four sets of electrophoresis per experiment. The average number of CPDs per fragment was calculated by using the Poisson distribution formula: –Ln (fraction of fragment free of dimer), assuming the random distribution of CPDs in a homogeneous population of fragments (2). Student’s t-test showed that the repair of CPDs in induced cells was statistically different at time-points 4 h (P <0.0001) and 8 h (P <0.001) from repair of CPDs in uninduced cells.

To further characterize the inducible repair pathway, we analyzed the strand-specific repair of CPDs in UV-irradiated B11 cells, with and without a priming UV dose. Removal of CPDs was quantified at the level of the TS and the NTS of the DHFR gene over a 24-h period (Fig. 2). As expected, our results show that, without prior irradiation, 95% repair of the DHFR TS was completed in 24 h, while only 25% repair occurred in the NTS (Fig. 2). When cells were pre-irradiated with a low UV dose, CPD repair of the DHFR TS was markedly increased at the 2, 4 and 8 h time-points (Fig. 2). In contrast, there was no significant difference in repair of the NTS of the DHFR gene between induced and uninduced cells (Fig. 2). Thus, our data suggest that there is an inducible TCR pathway for UV-induced CPDs in hamster cells.

Figure 2.

Strand-specific repair of UV-induced CPDs in the DHFR gene in induced and uninduced cells. (Top) Representative Southern blots of KpnI genomic fragments isolated from UV-irradiated induced and uninduced CHO B11 cells, not treated (–) or treated (+) with T4 endonuclease V and hybridized with DIG-labeled probes for the TS and NTS of the DHFR gene are shown. (Bottom) Data were obtained from scanning densitometry quantitation of Southern blots, using the ImageQuaNT™ analysis software (Molecular Dynamics) and are presented as mean values ± S.E. from four biological experiments with two to four sets of electrophoresis per experiment.

Enhancement of total RNA synthesis recovery following UV exposure in hamster cells pre-treated by a low UV dose

It has been suggested that the role of TCR is to efficiently remove lesions blocking the transcription machinery. To determine whether RNA synthesis recovery following UV irradiation can be modulated by UV pre-treatment, we compared total RNA synthesis recovery in UV-irradiated B11 cells, with and without giving a low priming UV dose. We observed that the sub-lethal dose of 2 J/m² alone did not significantly inhibit RNA synthesis over a 3 h period (data not shown). This low UV dose should be sufficient to weakly inhibit RNA synthesis, but we were not able to visualize differences in RNA synthesis between irradiated and control cells. We then measured the RNA synthesis resumption over a 120 min period in B11 cells irradiated either with a typical dose of 20 J/m² or with 2 J/m², 3 h before the 20 J/m² irradiation (Fig. 3). Following 20 J/m² UV exposure, the immediate inhibition of RNA synthesis was weak (90–95% recovery), but the rate of RNA synthesis was reduced to 80% within 40 min post-incubation (Fig. 3). Prior low-dose UV irradiation significantly increased the RNA synthesis recovery to 95% within 2 h post-treatment, while no significant recovery was observed without a priming dose during this time (Fig. 3). Thus, our findings show that total RNA synthesis recovery early after UV irradiation is enhanced by pre-treatment in hamster cells. Finally, 5 h after UV exposure, recovery of RNA synthesis was complete in both induced and uninduced cells (data not shown). These results indicate that RNA synthesis recovered faster than the removal of CPDs occurred from the TS (Fig. 2).

Figure 3.

Faster RNA synthesis recovery following UV irradiation in B11 cells pre-treated with a low UV dose. CHO-B11 cells were irradiated with 20 J/m² (uninduced cells) or with 2 J/m² before 20 J/m² (induced cells) followed by a 0–120 min post-incubation period prior to a 15 min [³H] uridine pulse to label nascent RNA. The measure of the RNA synthesis was taken from the ratio [³H] UV-treated cells/[³H] unirradiated cells. The value 100% represents the c.p.m. obtained from the RNA of unirradiated cells. The initial post-incubation time (0 h) was determined after the 15 min [³H] uridine pulse. Data obtained from scintillation counting are presented as mean values ± S.E. from five biological experiments. Student’s t-test showed that the RNA recovery synthesis in induced cells was statistically different at time-points 60 min (P <0.05), 80 min (P <0.05) and 120 min (P <0.02) from RNA recovery synthesis in uninduced cells.

DISCUSSION

Our results show that the pathway for CPDs is inducible in hamster cells. Repair was detected by a well-established technique allowing quantification of CPDs at the level of an endogenous gene by Southern blot hybridization (2,3). Pre-treatment of hamster cells with a low priming dose of UV light prior to a higher UV exposure enhances repair of CPDs in the TS of the active DHFR gene. These results suggest that TCR can be enhanced by activation of one or more signaling pathways. One known process involves the p53 protein. In response to DNA damage, the p53 tumor suppressor protein triggers multiple pathways, through transactivation of genes and protein–protein interactions, that regulate cell cycle arrest and apoptosis to maintain genomic integrity (for a review see 16). Recent conflicting studies have shown that p53 and/or p53-regulated gene products contribute to NER in human cells, modulating either overall DNA repair (17,18) or both global repair and TCR (19,20). However, the inducible pathway that we find in this report is independent of the p53 activation, since the CHO-K1 cell line that we used expresses high levels of mutant p53 protein that is non-inducible after irradiation and lacks G1 checkpoint function (21,22). Thus, it is likely that the signaling pathway required to induce TCR at the level of an active gene in CHO cells is independent of wild-type p53.

Pre-treatment of hamster cells with a low priming dose of UV light prior to higher UV exposure did not enhance repair of CPDs in the NTS of the active DHFR gene. These data are consistent with the deficiency in global genomic repair of CPDs in CHO cells (2), that lack the p53-regulated p48 gene product (18). Our findings do not rule out the possibility that overall repair may be inducible in cells expressing wild-type p53 protein.

It has been suggested that the role of TCR is to remove lesions blocking the transcription machinery from damaged DNA (4). Thus, it can be expected that the induction of TCR will enhance the recovery of RNA synthesis in UV-irradiated cells pre-treated with a low priming dose of UV. Our findings indeed show that total RNA synthesis recovery following UV irradiation is enhanced by pre-treatment in hamster cells. However, it is also expected that the recovery of RNA synthesis will follow removal of UV lesions from the TS. Nevertheless, we observed that RNA synthesis recovered faster than the removal of CPDs occurred from the TS. Data from Ljungman (23) support the hypothesis that the initially blocked RNA polymerases are somehow able to bypass photolesions prior to their removal. Resumption of RNA synthesis could then be accomplished either by removing the blocking lesions from the transcribed template or by lesion bypass resulting in translesion RNA synthesis. This may then explain why RNA synthesis resumed significantly faster than could be accounted for the repair of CPDs from the TS. The enhanced recovery of RNA synthesis and enhanced TCR in induced cells did not significantly affect UV survival in our study (data not shown). By contrast, decreased cell survival was observed in Cockayne’s syndrome (CS) cells that lack TCR (24) and an increased level of UV resistance has been correlated to an enhanced TCR in a stable UV resistant hamster cell line (25).

In conclusion, our results present strong evidence for an inducible TCR pathway for CPDs in hamster cells. A recent report has demonstrated inducible repair of thymine glycols in human cells (26). These findings and our study suggest that mammalian cells can elicit inducible DNA repair responses for both base and bulky DNA lesions. Additionally, our data show that wild-type p53 is not the signaling pathway required to induce TCR in CHO cells. It remains to be determined whether the observed inducible response reflects increased expression of repair genes directly involved in TCR, such as CS-A and CS-B genes or expression of other genes involved in pathways coupled to TCR, such as mismatch repair (MMR) genes (27). In yeast, several NER genes are subject to a modest transcriptional induction after UV irradiation (for a review see 28) and it was suggested that repair of CPDs is inducible (29). No transcription induction has so far been reported for the mammalian genes involved in TCR. An alternative hypothesis to explain the fastest repair of CPDs located in the TS of the pre-treated cells is that the proteins involved in the TCR machinery are already available near CPDs following the low UV dose. Although repair of UV lesions cannot be studied at such a low exposure of 2 J/m², it was shown that this UV dose is able to induce strand-specific mutations correlated to the strand bias for DNA repair (30). Since UV-damaged DNA and NER factors are recruited to the nuclear matrix soon after UV exposure (31–33), the fastest rate of TCR in pre-treated cells may then be due to an efficient recruitment of TCR components to this repair region. Additionally, MMR proteins, such as hMSH2, which is thought to act as general sensor for DNA damage (34) may then rapidly recruit other repair proteins involved in TCR. Future studies will define the underlying mechanisms responsible for this inducible response.