
Full text loading...
Category: Microbial Genetics and Molecular Biology
Heterogeneity of Nucleotide Excision Repair in Eukaryotic Genomes, Page 1 of 2
< Previous page | Next page > /docserver/preview/fulltext/10.1128/9781555816704/9781555813192_Chap10-1.gif /docserver/preview/fulltext/10.1128/9781555816704/9781555813192_Chap10-2.gifAbstract:
A definitive understanding of nucleotide excision repair (NER) in eukaryotes must ultimately embrace the process as it transpires in living cells and must accommodate many cellular biological observations. Some aspects of chromatin structure are considered in the context of the influence of such structure on the accessibility of sites in DNA to NER. The influence of transcription on NER efficiency is then discussed by two different mechanisms. First, chromatin structure is remodeled in a transcribed gene during transcription and in some regions of active genes. NER proteins can gain better access to damage as a consequence of this remodeling. Second, NER preferentially occurs on the transcribed strand of active genes in a transcription-coupled NER (TC-NER) process that is a specific response to blockage of RNAPII at a damaged site. The molecular details of the coupling of transcription blockage to TC-NER remain to be elaborated in detail. Genetic studies suggest several mechanisms that result in preferential NER of transcribed strands of active genes. As more is learned about the protein and nucleic acid participants of this process, rapid progress in our understanding of TC-NER is anticipated.
Full text loading...
(A) The nucleosome core particle (left) consists of 1.5 turns of DNA wrapped around a histone octamer comprising two molecules each of the core histones H2A, H2B, H3, and H4. Also shown (right) is a representation of the nucleosome, consisting of the core particle plus the linker DNA between two core particles and bound histone H1 (shown in grey). (B) Representation of the position of the contacts made between core histones (indicated by their numbers) and DNA as it wraps around the histone octamer. (Adapted from reference 216 .)
The crystal structure of the nucleosome core particle provides a molecular-scale roadmap of the histone octamer (two copies each of histones H2A, H2B, H3, and H4) and its interactions with 146 bp of DNA wrapped around the protein complex. One striking feature is the nonuniform shape of the DNA and the regular variation in double-helix geometry that suggests why nucleosomes reside in defined positions on DNA and require active remodeling to change chromatin structure. The N-terminal histone tails, which are subject to extensive post-translational modifications ( Fig. 10–4 ), protrude from the cracks between gyres of the DNA. (Adapted from reference 105 .)
Levels of folding of DNA and nucleosomes that give rise to highly condensed chromatin.
Twin supercoil domain model. (A) A transcription complex together with a nascent mRNA molecule is shown moving ca.10 bp along the DNA duplex. This movement creates a single positive supercoil in front of the polymerase and a negative supercoil behind it. (B) Histones are shown transferring (as intact octamers or partial assemblies). This transfer is facilitated by the fact that negatively supercoiled DNA preferentially binds histone octamers compared with positively supercoiled DNA. An additional two nucleosome cores are shown more remote from the polymerase, each storing one negative supercoil associated with the organization of two turns of DNA. (Adapted from reference 166 .)
Histone modifications on the nucleosome core particle. The nucleosome core particle showing six of the eight core histone N-terminal tail domains and two C-terminal tails is illustrated. Sites of post-translational modification are indicated by colored symbols that are defined in the key (lower left); acK, acetyl lysine; meR, methylarginine; meK, methyllysine; PS, phosphorylserine; uK, ubiquitinated lysine. Residue numbers are shown for each modification. Note that H3 lysine 9 can be either acetylated or methylated. The C-terminal (C-term) tail domains of one H2A molecule and one H2B molecule are shown (dashed lines), with sites of ubiquitination at H2A lysine 119 (most common in mammals) and H2B lysine 123 (most common in yeast). Modifications are shown on only one of the two copies of histones H3 and H4, and only one tail is shown for H2A and H2B. Sites marked by arrows are susceptible to cutting by trypsin in intact nucleosomes. The diagram is a compendium of data from various organisms, some of which may lack particular modifications (e.g., there is no H3meK9 in S. cerevisiae). (Adapted from reference 183 .)
Ratio of the amount of repair synthesis per unit DNA in MNase-sensitive and -resistant regions as a function of repair time after UV irradiation of cells. Note that very soon after UV irradiation, the repair synthesis label is distributed largely in MNase-sensitive regions of the genome. However, at later times the label is distributed more uniformly between MNase-sensitive and -resistant regions ( 147 ).
Transitions in chromatin structure that take place during NER of transcriptionally silent DNA. (A) UV radiation-induced lesions are formed throughout chromatin, with a slight modulation depending on local chromatin structure. (B) The XPC-RAD23B and/or DDB complex recognizes a DNA distortion. (C) Chromatin-remodeling complexes are directed to the site and change chromatin structure to allow sufficient binding of other NER factors in an ATP-dependent manner. (D) An open preincision complex is formed on ATP hydrolysis after recruitment of TFIIH, XPG, XPA, and RPA. (E) The endonucleases XPG and ERCC1-XPF incise the damaged strand 3’ and 5’ to the lesion, respectively. (F) A DNA polymerase holoenzyme fills in the gap by repair synthesis. (G) Redistribution of nucleosomes occurs, or reassembly of nucleosomes, which might be mediated by CAF1 and/or chromatin-remodeling complexes. DNA ligase I seals the nicks, perhaps in DNA already assembled into nucleosomes. (Adapted from reference 188 .)
Both the monofunctional psoralen 4’-(hydroxy-methyl)-4,5’,8-trimethylpsoralen (HMT) (A) and AAF (B) adducts are removed more slowly from α-DNA sequences of African green monkey cells than from bulk DNA. (Adapted from references 88 [panel A] and 226 [panel B].)
The methylation of DNA following NER of UV damage in confluent and in logarithmic-phase human fibroblasts is slower and less complete than in undamaged cells undergoing normal semiconservative DNA replication ( 77 ).
Effect of NER on the methylation of DNA. Repair of damage immediately in advance of a replication fork is shown on the left. If excision is initiated close to a methylated controlling sequence (A), a methylated base may be removed (B) and replaced (by repair synthesis) with a nonmethylated base (C). Replication of this region before remethylation can occur gives rise to a nonmethylated DNA duplex, which is not a substrate for the maintenance methylase. The other hemimethylated DNA duplex is normally methylated (D). Repair of damage close to a site of methylation immediately after replication is shown on the right. Excision of damage before the daughter strands have been methylated (A’ and B’) gives rise to a nonmethylated DNA molecule (C ‘), which is not a substrate for the maintenance methylase.
Detection of NER in specific gene sequences. (A) Cells are exposed to a DNA-damaging agent, which produces base adducts for which a specific detection method is available that can convert the damage to DNA strand breaks. In the example shown, DNA is damaged by UV radiation, thereby generating CPD that can be cleaved with a PD-DNA glycosylase/AP lyase such as T4 denV. Some of the cells are lysed immediately, while others are allowed to carry out NER in the presence of the nucleoside analog bromodeoxyuridine. DNA is extracted, purified, and treated with a restriction enzyme to generate fragments. (B) The restricted DNA is subjected to equilibrium density centrifugation in CsCl gradients to separate replicated (HL) DNA fragments from unreplicated (LL) DNA. (C) The unreplicated (LL) DNA is then treated with T4 denV or left untreated, forming strand breaks at the sites of CPD. (D) The DNA fragments are resolved by gel electrophoresis, transferred to a membrane, and probed with a 32P-labeled fragment derived from a gene of interest. Following autoradiography, DNA fragments not exposed to enzyme (—) yield an autoradiographic signal. In contrast, DNA isolated from cells immediately after exposure to UV radiation is degraded by the T4 denV (+) and yields little or no autoradiographic signal. The intensity of the autoradiographic signal after incubation (24 h) shows the extent to which CPD were removed by repair. The ratio of the intensity of the signal in the treated and untreated samples for a given time point indicates the fraction of DNA molecules that were free of damage (P0). The average number of damaged sites per fragment, S, can be derived using the Poisson equation (S = — ln P0). (Adapted from reference 15 .)
Measurement of NER in individual DNA strands. After irradiation of growing cells at time zero, steps A, B, and C as shown in Fig. 10–11 are followed. DNA is then subjected to electrophoresis in denaturing gels to separate the two strands of the fragment. Repair in the transcribed strand is visualized by probing a membrane with a radioactively labeled single-stranded DNA probe, so that the number of remaining lesions can be calculated. After stripping the signal, the same membrane is probed with labeled DNA from the complementary, nontranscribed strand and a similar calculation is made. (Adapted from reference 156 .)
The transcribed strand (TS) of the RPB2 gene of S. cerevisiae is repaired faster than the nontranscribed strand (NTS) during incubation of cells for the times indicated after their exposure to UV radiation. When the same experiment was carried out with a strain with a temperature-sensitive (ts) rbp2 allele incubated at the restrictive temperature, the preferential repair of the transcribed strand was abolished. Removal of CPD lesions was measured as in the experiment in Fig. 10–12 . (Adapted from reference 159 .)
Relative speed of repair in active genes. Repair is faster on the transcribed strand (TS) than on the nontranscribed strand (NTS), but variations in the rate of repair are evident between regions of both strands. (A) In the transcribed strand, repair is very fast close to the site of transcription initiation. (B) In the nontranscribed strand, but not in the transcribed strand, repair is markedly affected by the presence of nucleosomes. The color code in panel B reflects this finer detail. (Adapted from reference 156 .)
Different fates of RNAPII at a damaged site in DNA as mediated by CSB and other factors. The gold circle indicates the active site of RNAPII and the gold triangles denote a DNA lesion on the transcribed strand. (A) RNAPII dissociation from a site of damage. Such dissociation could be achieved by employing a transcription release factor or the Swi/Snf-like activities of CSB. (B) RNAPII is moved away from the damaged site. Such displacement could conceivably be achieved by Swi/Snf-like activities of CSB or by the use of other activities such as that of SII. (C) Remodeling of the damaged DNA-RNAPII interface by CSB. (D) At certain lesions, bypass might be promoted by accessory factors, of which CSB is a candidate activity. (E) Damage-binding factors arriving prior to RNAPII might facilitate repair by keeping the polymerase at a distance. (F) Degradation of RNAPII stalled at a lesion could occur. One or more of these fates of RNAPII appears to facilitate recruitment of the NER apparatus to the lesion. (Adapted from reference 156 .)
Def1 confers UV sensitivity independently of NER in yeast. A rad14 deletion strain of S. cerevisiae, completely defective in NER, is further sensitized by a deletion of the def1 gene. This additional sensitivity may arise because the degradation of stalled RNAPII is impaired by the def1 mutation. (Adapted from reference 217 .)
Possible stepwise mechanism to resolve arrested RNAPII elongation. Two examples of arrested RNAPII complexes and ways in which they might be resolved are shown. On the right is an elongation complex arrested at a DNA lesion, and on the left is a complex arrested for other reasons (such as a spontaneous stall at sequence forming a secondary structure). General elongation factors such as SII (TFIIS) play an important role in resolution of the latter situation but have little effect on RNAPII stalled at DNA damage. By contrast, human CSB or yeast Rad26 can affect the resolution of both types of complexes. If such resolution fails, ubiquitin-related mechanisms come into play. These do not always involve RNAPII proteolysis, which might be a solution of last resort. (Adapted from reference 157 .)
Rad26-independent TC-NER in the galactoseinduced GAL10 gene. Repair of CPD in the transcribed strand of GAL10 is shown. Specific DNA restriction fragments were incised with T4 denV enzyme at sites of CPD and end labeled with 32P by a procedure similar to that shown in Fig. 8–5. Lanes in the autoradiograph of the gel are DNA samples from unirradiated (U) and irradiated cells following incubation for 0, 1, 2, and 4 h. Ovals on the left represent nucleosome positions (darkest, least variable; lightest, most variable). Roman numerals denote the positions of the four Gal4-binding sites in the upstream activating sequence. The gold vertical arrow and bars on the right indicate the major transcriptional start site and TATA box of the gene, respectively. The black arrow on the right indicates the sites where TC-NER initiates in this gene. No removal of CPD from the nontranscribed strand occurs in rad16 mutant cells in similar experiments. nt, nucleotides. (Adapted from reference 96 .)
Preferential removal of CPD from active genes in mammalian cells a
TC-NER of UV radiation-induced CPD from the DHFR gene of CHO cells a