Ongoing Activity of RNA Polymerase II Confers Preferential Repair of Nitrogen Mustard-induced N-Alkylpurines in the Hamster Dihydrofolate Reductase Gene I

Recently, it has been demonstrated that nitrogen mustard-induced Nalkylpurines are excised rapidly from actively transcribing genes, while they persist longer in noncoding regions and in the genome overall. It was suggested that transcriptional activity is implicated as a regulatory element in the efficient removal of lesions. By treating cells or not with the transcription inhibitor a-amanitin, we have explored whether ongoing activity of RNA polymerase II was coordinately related to proficient repair of nitrogen mustard-induced alkylation products in the actively transcribed dihydrofolate reductase gene in the Chinese hamster ovary Bl l cells. Nuclear run-off transcription analysis verified that ~-amanitin completely and selectively inhibited transcription by RNA polymerase II. At the drug exposure examined, nitrogen mustard induced DNA damage capable of a complete transcription termination in the RNA polymerase II-transcribed dihydrofolate reductase gene and reduced 28S rDNA transcription by a factor of 7.9. The transcription activity did partially recover following reincubation in drug-free medium; this recovery was about 34 and 76% of ribosomal 28S gene transcripts and dihydrofolate reductase gene transcripts, respectively, after 6 h of repair incubation, a-Amanitin significantly inhibited the removal of nitrogen mustard-induced N-alkylpurines in the 5'-half of the essential, constitutively active dihydrofolate reductase gene, while no effect of a-amanitin was observed on the lesion removal from a noncoding region 3'-flanking to the gene and from the genome overall. In the actively transcribed gene region, about 77% of N-alkylpurines were removed 21 h following drug exposure of cells not treated with a-amanitin and about 47% in 21 h in a-amanitin treated cells. The global semiconservative replication seemed unaffected by the a-amanitin treatment. From these results we suggest that gene-specific repair of nitrogen mustard-induced N-alkylpurines is dependent on ongoing activity of the transcribing RNA polymerase II. The findings are discussed in terms of the current ideas about the mechanism of preferential DNA repair. I N T R O D U C T I O N Recent development of various techniques to examine the formation and the repair of various DNA lesions in individual genes has advanced our knowledge of the DNA repair processes in a wide array of organisms (see Ref. 1 for review). Especially in the case of UV light-induced cyclobutane pyrimidine dimers, ample evidence has been presented that repair of this lesion occurs preferentially in transcriptionally active genes (2-5). Also, the efficiency of DNA repair may be affected by the nature of the lesions, its location within the genome, the transcriptional activity, and the conformation of the chromatin structure of the damaged DNA regions. DNA base damage is known to undergo a complex excision repair mechanism involving several specific enzymes (6). Considerable differences in the mode of repair exist among different monofunctional alkylating agents and even among agents which cause similar types of DNA lesions. DNA base methylation introduced with dimethyl sulfate Received 7/28/93; accepted 11/1/93. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 Supported by the Danish Cancer Society (Grant 92-030). 2 To whom correspondence should be addressed. seem to be randomly removed from the genome independently of its location within the genome (7, 8), while a correlation between transcriptional activity and availability to lesion removal of methylnitrosourea-induced DNA methylation has been demonstrated in the same hamster cells (9), a rat insulinoma cell line (10), and cultured human cells (11). Although alkylation by SNl-alkylating agents (e.g., methylnitrosourea) has greater sequence selectivity than Sy2-alkylating agents (e.g., dimethyl sulfate) in vitro (12), it is not clear how this would result in differences in the mode of repair. It is likely that the observed differences in repair of methylpurines may depend on the degree of chromatin distortion invoked by these monofunctional alkylating agents and thereby the recognition enzymes involved. Alternatively, it may reflect the degree of transcription-terminating lesions induced by the DNA adducts. A particularly intriguing problem is to establish the relationship, if any, between DNA repair and transcription. HN23 is a highly electrophilic anticancer agent which alkylates DNA bases mainly at N 7guanine and to a much lesser extent at other positions (13-15). HN2 bears two alkylating groups per molecule which together allow the formation of cross-links between paired DNA strands, between guanines in the same strand, or between DNA and protein (16-20). The biological effect of HN2 may depend on preferential reaction at certain genomic locations. Recently, we described a general approach to analyze gene-specific damage and repair of N-alkylpurines; the methodology is especially suitable for measuring NT-guanine alkylation upon exposure in vivo and measures the production of both DNA monoadducts and cross-links (8). We have shown that HN2-mediated N-alkylpurines are excised rapidly from actively transcribing genes but persist longer in an inactive gene region and a noncoding region of the hamster genome (8, 21). Also, the much less frequent HN2mediated DNA interstrand cross-links have recently been reported to be preferentially repaired (22, 23). Finally, by using the same general technique for the measurements of N-alkylpurines (8), but probing the membranes with single-stranded RNA probes (riboprobes), we observed a slight bias toward repair in the transcribed strand after exposure to HN2. Thus, the phenomenon of preferential repair of HN2induced N-alkylpurines resemble that seen for the removal of UV light-induced cyclobutane pyrimidine dimers, although some quantitative differences may occur (1). The biological advantage of preferential repair is evident, but the mechanism by which it occurs is still largely unknown. The observation that repair of HN2-induced alkylation products occurs nonrandomly in the genome suggests that the transcription complex could be actively involved in the recognition and removal of HN2-induced N-alkylpurines. The lesions generated by HN2 capable of blocking transcription at the NT-guanine alkylation sites in DNA (24) may be an important signal: stalled RNA polymerase could direct repair enzymes to the transcription-terminating lesion. To determine whether ongoing transcription is required for the preferential repair of HN2-induced DNA lesions, we have used o~-amanitin to inhibit transcription by RNA polymerase II in HN2-treated CHO cells, a-Amanitin, the bi3 The abbreviations used are: HN2, nitrogen mustard, bis(2-chloroethyl)methylamine; CHO, Chinese hamster ovary; DHFR, dihydrofolate reductase.