Transcription-coupled DNA repair is genomic context-dependent

EvoLSTM: context-dependent models of sequence evolution using a sequence-to-sequence LSTM

Motivation

Accurate probabilistic models of sequence evolution are essential for a wide variety of bioinformatics tasks, including sequence alignment and phylogenetic inference. The ability to realistically simulate sequence evolution is also at the core of many benchmarking strategies. Yet, mutational processes have complex context dependencies that remain poorly modeled and understood.

Results

We introduce EvoLSTM, a recurrent neural network-based evolution simulator that captures mutational context dependencies. EvoLSTM uses a sequence-to-sequence long short-term memory model trained to predict mutation probabilities at each position of a given sequence, taking into consideration the 14 flanking nucleotides. EvoLSTM can realistically simulate mammalian and plant DNA sequence evolution and reveals unexpectedly strong long-range context dependencies in mutation probabilities. EvoLSTM brings modern machine-learning approaches to bear on sequence evolution. It will serve as a useful tool to study and simulate complex mutational processes.

Availability and implementation

Code and dataset are available at https://github.com/DongjoonLim/EvoLSTM.

Supplementary information

Supplementary data are available at Bioinformatics online.

Blinded by the UV light: how the focus on transcription-coupled NER has distracted from understanding the mechanisms of Cockayne syndrome neurologic disease

Cockayne syndrome (CS) is a devastating neurodevelopmental disorder, with growth abnormalities, progeriod features, and sun sensitivity. CS is typically considered to be a DNA repair disorder, since cells from CS patients have a defect in transcription-coupled nucleotide excision repair (TC-NER). However, cells from UV-sensitive syndrome patients also lack TC-NER, but these patients do not suffer from the neurologic and other abnormalities that CS patients do. Also, the neurologic abnormalities that affect CS patients (CS neurologic disease) are qualitatively different from those seen in NER-deficient XP patients. Therefore, the TC-NER defect explains the sun sensitive phenotype common to both CS and UVsS, but cannot explain CS neurologic disease. However, as CS neurologic disease is of much greater clinical significance than the sun sensitivity, there is a pressing need to understand its molecular basis. While there is evidence for defective repair of oxidative DNA damage and mitochondrial abnormalities in CS cells, here I propose that the defects in transcription by both RNA polymerases I and II that have been documented in CS cells provide a better explanation for many of the severe growth and neurodevelopmental defects in CS patients than defective DNA repair. The implications of these ideas for interpreting results from mouse models of CS, and for the development of treatments and therapies for CS patients are discussed.

Chromium (VI) induces both bulky DNA adducts and oxidative DNA damage at adenines and guanines in the p53 gene of human lung cells

Chromium (VI) [Cr(VI)], a ubiquitous environmental carcinogen, is generally believed to induce mainly mutagenic binary and ternary Cr(III)-deoxyguanosine (dG)-DNA adducts in human cells. However, both adenine (A) and guanine (G) mutations are found in the p53 gene in Cr exposure-related lung cancer. Using UvrABC nuclease and formamidopyrimidine glycosylase (Fpg), and ligation-mediated PCR methods, we mapped the distribution of bulky DNA adducts (BDA) and oxidative DNA damage (ODD) in the p53 gene in Cr(VI)-treated human lung cells. We found that both BDA and ODD formed at 2'-deoxyadenosine (dA) and dG bases. To understand the causes for these Cr-induced DNA damages, we mapped the distribution of BDA adducts and ODD in the p53 gene DNA fragments induced by Cr(III), Cr(VI) and Cr(V), the three major cellular Cr forms. We found that (i) dA at -CA- is a major Cr(VI) binding site followed by -GG- and -G-. Cr(VI) does not bind to -GGG-, (ii) Cr(VI)-DNA binding specificity is distinctly different from the Cr(III)-DNA binding in which -GGG- and -GG- are preferential sites, (iii) Cr(V) binding sites include all of Cr(VI) and Cr(III)-DNA binding sites and (iv) Cr(VI) and Cr(V) induce Fpg-sensitive sites at -G-. Together, these results suggest that Cr(VI) induction of BDA and ODD at dA and dG residues is through Cr(V) intermediate. We propose that these Cr(VI)-induced BDA and ODD contribute to mutagenesis of the p53 gene that leads to lung carcinogenesis.

Detoxification: a novel function of BRCA1 in tumor suppression?

Our studies found that BRCA1 levels negatively correlate with DNA adducts induced by Benzo(a)pyrene (BaP). Pulse-chase experiments showed that the increase in BaP-induced DNA adducts in BRCA1 knockdown cells may not be associated with BRCA1’s function in nucleotide excision repair activity; rather, it may be associated with its function in modulating transcriptional regulation. BRCA1 knockdown in MCF-10A cells significantly attenuated the induction of CYP1A1 following BaP treatment indicating that the increase in BaP-induced adducts in BRCA1 knockdown cells is not CYP1A1 dependent. However, our study shows that BRCA1 defective cells may still be able to biotransform BaP by regulating other CYP enzymes, including CYP1B1. Knockdown of BRCA1 also severely affected the expression levels of two types of uridine diphosphate glucorunyltransferase (UGT1A1 and UGT1A9) and NRF2. Both UGTs are known as BaP-specific detoxification enzymes, and NRF2 is a master regulator of antioxidant and detoxification genes. Thus, we concluded that the increased amount of BaP-induced DNA adducts in BRCA1 knockdown cells is strongly associated with its loss of functional detoxification. Chromatin immunoprecipitation assay revealed that BRCA1 is recruited to the promoter/enhancer sequences of UGT1A1, UGT1A9, and NRF2. Regulation of UGT1A1 and UGT1A9 expression showed that the induction of DNA adducts by BaP is directly affected by their expression levels. Finally, overexpression of UGTs, NRF2, or ARNT significantly decreased the amount of BaP-induced adducts in BRCA1-deficient cells. Overall, our results suggest that BRCA1 protects cells by reducing the amount of BaP-induced DNA adducts possibly via transcriptional activation of detoxification gene expression.

Recognition and incision of Cr(III) ligand-conjugated DNA adducts by the nucleotide excision repair proteins UvrABC: importance of the Cr(III)-purine moiety in the enzymatic reaction

Hexavalent chromium [Cr(VI)] is an ubiquitous environmental contaminant and a well-known etiological agent of human lung cancer. Inside human cells, Cr(VI) is reduced to Cr(III), which can conjugate with amino acids, ascorbic acids, and glutathiones in the cytoplasm. Conjugated and unconjugated Cr(III) can enter the nucleus to form adducts with DNA and electrostatically interact with the phosphate group of DNA. It has been found that in both human and Escherichia coli systems, Cr(III) ligand-conjugated DNA ternary adducts are efficiently repaired by the nucleotide excision repair (NER) pathway. In contrast, DNA adducts formed by unconjugated Cr(III) with DNA are repaired significantly less efficiently by the NER system. These results raise the possibility that the NER system repairs Cr(III) ligand-conjugated DNA adducts and biadducts such as Cr(III)-guanine-phosphate adducts but not Cr(III)-phosphate adducts. To test this hypothesis, we determined the cutting efficiency and the mode of cutting of DNA modified with tannin-conjugated Cr(III) by the E. coli NER enzymes UvrABC. Tannin compounds, gallic acid (GA), and ethyl gallate (EGA) can reduce Cr(VI) to Cr(III) to form Cr(III)-GA 2 and Cr(III)-EGA 2, respectively, which can interact with a single guanine or adenine base but not with the DNA phosphate backbone. We found that UvrABC is able to incise Cr(III)-GA 2- and Cr(III)-EGA 2-modified plasmid DNA, and the amount of incision increased as a function of tannin concentration used for modifications. In contrast, UvrABC nuclease does not incise GA- and EGA-modified plasmid DNA. Mapping the sequence specificity of Cr(III)-GA 2- and Cr(III)-EGA 2-DNA formation in the human p53 gene sequence by UvrABC nuclease cutting, we found that the sequence specificity for both adducts is the same but is much more selective than Cr(III)-guanine-DNA adducts. Together, these results suggest that NER proteins from E. coli recognize the purine-Cr(III) adduct but not the Cr(III)-backbone phosphate complex.

Both replication bypass fidelity and repair efficiency influence the yield of mutations per target dose in intact mammalian cells induced by benzo[a]pyrene-diol-epoxide and dibenzo[a,l]pyrene-diol-epoxide

Mutations induced by polycyclic aromatic hydrocarbons (PAH) are expected to be produced when error-prone DNA replication occurs across unrepaired DNA lesions formed by reactive PAH metabolites such as diol epoxides. The mutagenicity of the two PAH-diol epoxides (+)-anti-7,8-dihydroxy-9,10-epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene (BPDE) and (+/-)-anti-11,12-dihydroxy-13,14-epoxy-11,12,13,14-tetrahydrodibenzo[a,l]pyrene (DBPDE) was compared in nucleotide excision repair (NER) proficient and deficient hamster cell lines. We applied the (32)P-postlabelling assay to analyze adduct levels and the hprt gene mutation assay for monitoring mutations. It was found that the mutagenicity per target dose was 4 times higher for DBPDE compared to BPDE in NER proficient cells while in NER deficient cells, the mutagenicity per target dose was 1.4 times higher for BPDE. In order to investigate to what extent the mutagenicity of the different adducts in NER proficient cells was influenced by repair or replication bypass, we measured the overall NER incision rate, the rate of adduct removal, the rate of replication bypass and the frequency of induced recombination in the hprt gene. The results suggest that NER of BPDE lesions are 5 times more efficient than for DBPDE lesions, in NER proficient cells. However, DBPDE adducts block replication more efficiently and also induce 6 times more recombination events in the hprt gene than adducts of BPDE, suggesting that DBPDE adducts are, to a larger extent, bypassed by homologous recombination. The results obtained here indicate that the mutagenicity of PAH is influenced not only by NER, but also by replication bypass fidelity. This has been postulated earlier based on results using in vitro enzyme assays, but is now also being recognized in terms of forward mutations in intact mammalian cells.

Lack of strand-specific repair of UV-induced DNA lesions in three genes of the archaeon Sulfolobus solfataricus

In all organisms, specialized systems are devoted to repair of DNA lesions induced by exposure to UV light. In both Eucarya and Bacteria, UV-induced pyrimidine dimers in the transcribed strand of active genes are repaired at a faster rate compared to the non-transcribed strand and the rest of the genome. Preferential repair of transcribed strands requires the Transcription-Repair Coupling Factor in Escherichia coli and the CSA and CSB proteins in humans. These factors are needed for coupling of transcription to nucleotide excision repair (NER), a major pathway for repair of UV-induced lesions. Whereas transcription-coupled NER (TC-NER) is an evolutionary conserved process, not all active genes show preferential repair of transcribed strands. The existence of a NER pathway in the Archaea has not been demonstrated directly, yet it is suggested by the presence and properties of homologues of NER nucleases and helicases. However, none of the proteins responsible for the lesion recognition steps or for TC-NER has been found in archaeal genomes. Moreover, the kinetics of gene or strand-specific repair has never been investigated in any organism of this domain. We have analysed the kinetics of repair of UV-induced DNA damage in the transcribed and non-transcribed strands of three genes of the hyperthermophilic archaeon Sulfolobus solfataricus. We found that in all three genes the two strands are repaired with the same efficiency with each other and with the genome in general, thus providing no evidence of strand bias or transcription coupling of the repair process in the genes analysed. Further studies will be required to test the existence of a transcription-coupled repair pathway in other archaeal genes and to elucidate the mechanism of UV lesion recognition and repair in Archaea.

Preferential in vivo DNA repair of melphalan-induced damage in human genes is greatly affected by the local chromatin structure

To investigate the molecular mechanisms of action of the nitrogen mustard melphalan in patients treated for multiple myeloma, the in vivo induction and repair of melphalan-induced DNA damage was measured in genes with different transcriptional activity (b-actin>p53>N-ras>d-globin) from leukocytes of 20 multiple myeloma patients following chemotherapeutic administration of high-dose melphalan (200mg/m(2)) and autologous blood stem cell transplantation. Heterogeneous repair was found among the studied genes. The extent of repair was always in the order: b-actin>p53>N-ras>d-globin, correlating with the gene transcriptional state. Similar findings were obtained using peripheral blood mononuclear cells (PBMC) from healthy volunteers following in vitro treatment with melphalan, indicating that these results are not malignant disease-specific. Following in vitro treatment of PBMC from healthy volunteers with alpha-amanitin, an inhibitor of RNA polymerase II that can also induce condensation of chromatin structure, a significant inhibition of the removal of melphalan-induced damage in the three active genes but not in the silent d-globin gene was found, suggesting that transcription and/or chromatin structure may play important roles in the preferential DNA repair. When the in vivo DNA damage formation and repair in multiple myeloma patients following chemotherapeutic administration of melphalan was measured in the two strands of the active genes, no strand bias was found, indicating that the global genome repair subpathway of nucleotide excision repair may play a crucial role in the repair of these adducts. These results were also confirmed in PBMC from healthy volunteers following in vitro treatment with melphalan. Using micrococcal nuclease digestion of nuclei isolated from PBMC of multiple myeloma patients before the chemotherapeutic treatment, as well as from PBMC of healthy volunteers, we probed the chromatin structure in each gene and found that the "looseness" of the chromatin structure correlated with the levels of the gene-specific repair, being again in the order: b-actin>p53>N-ras>d-globin. To conclude, the in vivo gene-specific repair of melphalan-induced damage in humans is greatly affected by the local chromatin structure.

Codon 12 region of mouse K-ras gene is the site for in vitro binding of transcription factors GATA-6 and NF-Y

Codon 12 of the K-ras gene is a generally recognized example of a mutational hot spot. By the approach of gel retardation and specific antibodies, a double-stranded oligonucleotide corresponding to the codon 12 region of the mouse K-ras gene (from 20 to 50 bp with respect to the exon 1 start) was found to be a site for cooperative binding of the transcription factors GATA-6 and NF-Y. GATA-6 and NF-Y were selectively activated with lung carcinogens 3-methylcholanthrene and nitrosoethylurea in mice of strains susceptible to lung tumorigenesis but not in animals of resistant strains. The interaction of GATA-6 and NF-Y with the codon 12 region of the K-ras gene is suggested to be involved in the mechanism of lung carcinogenesis.

Sequence specificity of Cr(III)-DNA adduct formation in the p53 gene: NGG sequences are preferential adduct-forming sites

Hexavalent chromium [Cr(VI)] is a known etiological factor in human lung cancer. Cr(VI) exposure-related lung cancer has a high mutation incidence in the p53 gene. Upon intake in human cells Cr(VI) is reduced to Cr(III), which is able to conjugate with amino acids and consequently form either binary Cr(III)-DNA or ternary Cr(III)-amino acid-DNA adducts. Both binary and ternary Cr(III)-DNA adducts are mutagenic. We have found that the Escherichia coli nucleotide excision enzyme UvrABC nuclease is able to incise Cr(III)- and Cr(III)-histidine-modified plasmid DNA and the extent of incision is proportional to the amount of Cr(III)-DNA adducts in the plasmid. In order to determine the role of Cr(III)-DNA adducts in the mutagenesis of the p53 gene in human cancer using the UvrABC nuclease incision method, we have mapped the Cr(III)-DNA distribution in PCR DNA fragments amplified from exons 5, 7 and 8 of the p53 gene. We have found that the sequence specificities of Cr(III)-DNA and Cr(III)-histidine-DNA adducts in the p53 gene sequence are identical and that both types of adducts are preferentially formed at -NGG- sequences, including codons 245, 248 and 249, the mutational hotspots in human lung cancer. It has been found that Cr(III)-DNA adducts induce mainly G to T mutations. Therefore, these results suggest that Cr(III)-DNA adduct formation contributes to the p53 gene mutations in lung carcinogenesis.

Large-scale cDNA transfection screening for genes related to cancer development and progression

A large-scale assay was performed by transfecting 29,910 individual cDNA clones derived from human placenta, fetus, and normal liver tissues into human hepatoma cells and 22,926 cDNA clones into mouse NIH 3T3 cells. Based on the results of colony formation in hepatoma cells and foci formation in NIH 3T3 cells, 3,806 cDNA species (8,237 clones) were found to possess the ability of either stimulating or inhibiting cell growth. Among them, 2,836 (6,958 clones) were known genes, 372 (384 clones) were previously unrecognized genes, and 598 (895 clones) were unigenes of uncharacterized structure and function. A comprehensive analysis of the genes and the potential mechanisms for their involvement in the regulation of cell growth is provided. The genes were classified into four categories: I, genes related to the basic cellular mechanism for growth and survival; II, genes related to the cellular microenvironment; III, genes related to host-cell systemic regulation; and IV, genes of miscellaneous function. The extensive growth-regulatory activity of genes with such highly diversified functions suggests that cancer may be related to multiple levels of cellular and systemic controls. The present assay provides a direct genomewide functional screening method. It offers a better understanding of the basic machinery of oncogenesis, including previously undescribed systemic regulatory mechanisms, and also provides a tool for gene discovery with potential clinical applications.

Single-Stranded Breaks in DNA but Not Oxidative DNA Base Damages Block Transcriptional Elongation by RNA Polymerase II in HeLa Cell Nuclear Extracts

Transcription and repair of many DNA helix-distorting lesions such as cyclobutane pyrimidine dimers have been shown to be coupled in cells across phyla from bacteria to humans. The signal for transcription-coupled repair appears to be a stalled transcription complex at the lesion site. To determine whether oxidative DNA lesions can block correctly initiated human RNA polymerase II, we examined the effect of site-specifically introduced oxidative damages on transcription in HeLa cell nuclear extracts. We found that transcription was blocked by single-stranded breaks, common oxidative DNA lesions, when present in the transcribed strand of the transcription template. Cyclobutane pyrimidine dimers, which have been previously shown to block transcription both in vitro and in vivo, also blocked transcription in the HeLa cell nuclear transcription assay. In contrast, the oxidative DNA base lesions, 8-oxoguanine, 5-hydroxycytosine, and thymine glycol did not inhibit transcription, although pausing was observed with the thymine glycol lesion. Thus, DNA strand breaks but not oxidative DNA base damages blocked transcription by RNA polymerase II.

Effects of genomic context and chromatin structure on transcription-coupled and global genomic repair in mammalian cells

It has been long recognized that in mammalian cells, DNA damage is preferentially repaired in the transcribed strand of transcriptionally active genes. However, recently, we found that in Chinese hamster ovary (CHO) cells, UV-induced cyclobutane pyrimidine dimers (CPDs) are preferentially repaired in both the transcribed and the non-transcribed strand of exon 1 of the dihydrofolate reductase (DHFR) gene. We mapped CPD repair at the nucleotide level in the transcriptionally active DHFR gene and the adjacent upstream OST gene, both of which have been translocated to two chromosomal positions that differ from their normal endogeneous positions. This allowed us to study the role of transcription, genomic context and chromatin structure on repair. We found that CPD repair in the transcribed strand is the same for endogenous and translocated DHFR genes, and the order of repair efficiency is exon 1 > exon 2 > exon 5. However, unlike the endogenous DHFR gene, efficient repair of CPDs in the non-transcribed strand of exon 1 is not observed in the translocated DHFR gene. CPDs are efficiently repaired in the transcribed strand in endogenous and translocated OST genes, which indicates that efficient repair in exon 1 of the non-transcribed strand of the endogenous DHFR gene is not due to the extension of transcription-coupled repair of the OST gene. Using micrococcal nuclease digestion, we probed the chromatin structure in the DHFR gene and found that chromatin structure in the exon 1 region of endogenous DHFR is much more open than at translocated loci. These results suggest that while transcription-coupled repair is transcription dependent, global genomic repair is greatly affected by chromatin structure.

Dissecting transcription-coupled and global genomic repair in the chromatin of yeast GAL1-10 genes

Transcription-coupled repair (TCR) and global genomic repair (GGR) of UV-induced cyclobutane pyrimidine dimers were investigated in the yeast GAL1-10 genes. Both Rpb9- and Rad26-mediated TCR are confined to the transcribed strands, initiating at upstream sites approximately 100 nucleotides from the upstream activating sequence shared by the two genes. However, TCR initiation sites do not correlate with either transcription start sites or TATA boxes. Rad16-mediated GGR tightly correlates with nucleosome positioning when the genes are repressed and are slow in the nucleosome core and fast in linker DNA. Induction of transcription enhanced GGR in nucleosome core DNA, especially in the nucleosomes around and upstream of the transcription start sites. Furthermore, when the genes were induced, GGR was slower in the transcribed regions than in the upstream regions. Finally, simultaneous deletion of RAD16, RAD26, and RPB9 resulted in no detectable repair in all sites along the region analyzed. Our results suggest that (a). TCR may be initiated by a transcription activator, presumably through the loading of RNA polymerase II, rather than by transcription initiation or elongation per se; (b). TCR and nucleosome disruption-enhanced GGR are the major causes of rapid repair in regions around and upstream of transcription start sites; (c). transcription machinery may hinder access of NER factors to a DNA lesion in the absence of a transcription-repair coupling factor; and (d). other than GGR mediated by Rad16 and TCR mediated by Rad26 and Rpb9, no other nucleotide excision repair pathway exists in these RNA polymerase II-transcribed genes.

The contribution of cis-elements to disease-associated repeat instability: clinical and experimental evidence

Alterations in the length (instability) of gene-specific microsatellites and minisatellites are associated with at least 35 human diseases. This review will discuss the various cis-elements that contribute to repeat instability, primarily through examination of the most abundant disease-associated repetitive element, trinucleotide repeats. For the purpose of this review, we define cis-elements to include the sequence of the repeat units, the length and purity of the repeat tracts, the sequences flanking the repeat, as well as the surrounding epigenetic environment, including DNA methylation and chromatin structure. Gender-, tissue-, developmental- and locus-specific cis-elements in conjunction with trans-factors may facilitate instability through the processes of DNA replication, repair and/or recombination. Here we review the available human data that supports the involvement of cis-elements in repeat instability with limited reference to model systems. In diverse tissues at different developmental times and at specific loci, repetitive elements display variable levels of instability, suggesting vastly different mechanisms may be responsible for repeat instability amongst the disease loci and between various tissues.

Preferential carcinogen-DNA adduct formation at codons 12 and 14 in the human K-ras gene and their possible mechanisms

In the ras gene superfamily, codon 12 (-TGGTG-) of the K-ras gene is the most frequently mutated codon in human cancers. Recently, we have found that bulky chemical carcinogens preferentially form DNA adducts at codons 12 and 14 (-CGTAG-) in the K-ras gene in normal human bronchial epithelial (NHBE) cells. Furthermore, DNA adducts formed at codon 12 of the K-ras gene are poorly repaired compared with those at other codons including codon 14. These results suggest that targeted carcinogen-DNA adduct formation is a major reason for the observed high mutation frequency at codon 12 of the K-ras gene in human cancers. This preferential carcinogen-DNA adduct formation at codons 12 and 14 could result from effects of (1) primary sequences of these codons and their surrounding codons in the K-ras gene, (2) the chromatin structure, and/or (3) epigenetic factors such as C5 cytosine methylation or other DNA modifications at these codons and their surrounding codons. To distinguish these possibilities, we have introduced modifications with benzo[a]pyrene diol epoxide, N-hydroxy-2-aminofluorene, and aflatoxin B1 8,9-epoxide in (1) naked intact genomic DNA isolated from NHBE cells, (2) fragmented genomic DNA digested by restriction enzymes, and (3) in vitro synthesized DNA fragments containing the K-ras gene exon 1 sequence with or without methylation of the cytosines at CpG sites and the cytosines pairing with the guanines of codons 12 and 14. The distribution of carcinogen-DNA adducts in the K-ras gene was mapped at the nucleotide sequence level using the UvrABC nuclease incision method with or without the ligation-mediated polymerase chain reaction technique. We have found that carcinogens preferentially form adducts at codons 12 and 14 in the K-ras gene exon 1 in intact as well as in fragmented genomic DNA. In contrast, this preferential DNA adduct formation at codons 12 and 14 was not observed in PCR-amplified DNA fragments containing the K-ras gene exon 1 sequence. Methylation of the cytosine at the CpG site of codon 14, or the cytosine pairing with guanine of codon 14, greatly enhanced carcinogen-DNA adduct formation at codon 14 but did not affect carcinogen-DNA adduct formation at codon 12. Methylation of the cytosine pairing with the guanine of codon 12 also did not enhance carcinogen-DNA adduct formation at codon 12. Furthermore, we found that the cytosine at the CpG site of codon 14 is highly methylated in NHBE cells. These results suggest that cytosine methylation at the CpG site is the major reason for the preferential DNA damage at codon 14 and that epigenetic modification(s) other than cytosine methylation may contribute to the preferential DNA damage at codon 12 of the K-ras gene.

Transcription-coupled and transcription-independent repair of cyclobutane pyrimidine dimers in the dihydrofolate reductase gene

Using a ligation-mediated polymerase chain reaction technique, we have mapped the repair of ultraviolet light-induced cyclobutane pyrimidine dimers (CPDs) at the nucleotide level in exons 1, 2, and 5 of the dihydrofolate reductase (DHFR) gene in Chinese hamster ovary cells. We found that CPDs are preferentially repaired in the transcribed strand (T strand) and that the order of repair efficiency is exon 1 > exon 2 > exon 5. In the cells with a deletion of the DHFR gene encompassing the promoter region and the first four exons, CPDs are not repaired in the T strand of the residual DHFR gene. These results substantiate the idea that the preferential repair of CPDs in the T strand is transcription dependent. However, in the wild type gene we have found that CPDs are efficiently repaired in the nontranscribed strand (NT strand) of exon 1 but not in the NT strand of exons 2 and 5. Probing the chromatin structure of exons 1, 2, and 5 of the DHFR gene with micrococcal nuclease revealed that the exon 1 region is much more sensitive to micrococcal nuclease digestion than the exon 2 and exon 5 regions, suggesting that the chromatin structure in the exon 1 region is much more open. These results suggest that, although preferential repair of the T strand of the DHFR gene is transcription dependent, repair of the NT strand is greatly affected by chromatin structure.

Clusters of transcription-coupled repair in the human genome

A specialized nucleotide excision repair pathway known as transcription-coupled repair (TCR) counteracts the toxic effects of DNA damage in transcriptionally active genes. The clustering of active genes into gene-rich chromosomal domains predicts that the sites of TCR are unevenly distributed through the genome. To elucidate the genomic organization and chromosomal localization of TCR, we isolated DNA fragments encompassing TCR-mediated repair sites from UV-C irradiated xeroderma pigmentosum group C cells, which can only repair the transcribed strand of active genes. This DNA was used as a molecular probe to visualize TCR in normal metaphase spreads by reverse fluorescence in situ hybridization. Whereas DNA repair sites in normal human cells are evenly distributed through the genome, TCR is highly localized at specific chromosomal domains. Particularly, clusters of TCR sites were identified at early-replicating gene-rich bands and telomeric regions of several chromosomes. High gene-density chromosomes such as chromosome 19 and the GC-rich domains of several chromosomes (T bands) are preferential locations of TCR. Our results demonstrate that the intragenomic localization of TCR resembles the uneven distribution of the human transcriptome, CpG islands, and hyperacetylated histones, enforcing the basic link between DNA repair, transcription, and nuclear organization in a complex genome.