Monte Carlo simulation of base and nucleotide excision repair of clustered DNA damage sites. I. Model properties and predicted trends.

DNA is constantly damaged through endogenous processes and by exogenous agents, such as ionizing radiation. Base excision repair (BER) and nucleotide excision repair (NER) help maintain the stability of the genome by removing many different types of DNA damage. We present a Monte Carlo excision repair (MCER) model that simulates key steps in the short-patch and long-patch BER pathways and the NER pathway. The repair of both single and clustered damages, except double-strand breaks (DSBs), is simulated in the MCER model. Output from the model includes estimates of the probability that a cluster is repaired correctly, the fraction of the clusters converted into DSBs through the action of excision repair enzymes, the fraction of the clusters repaired with mutations, and the expected number of repair cycles needed to completely remove a clustered damage site. The quantitative implications of alternative hypotheses regarding the postulated repair mechanisms are investigated through a series of parameter sensitivity studies. These sensitivity studies are also used to help define the putative repair characteristics of clustered damage sites other than DSBs.

DNA repair of a single UV photoproduct in a designed nucleosome.

Eukaryotic DNA repair enzymes must interact with the architectural hierarchy of chromatin. The challenge of finding damaged DNA complexed with histone proteins in nucleosomes is complicated by the need to maintain local chromatin structures involved in regulating other DNA processing events. The heterogeneity of lesions induced by DNA-damaging agents has led us to design homogeneously damaged substrates to directly compare repair of naked DNA with that of nucleosomes. Here we report that nucleotide excision repair in Xenopus nuclear extracts can effectively repair a single UV radiation photoproduct located 5 bases from the dyad center of a positioned nucleosome, although the nucleosome is repaired at about half the rate at which the naked DNA fragment is. Extract repair within the nucleosome is >50-fold more rapid than either enzymatic photoreversal or endonuclease cleavage of the lesion in vitro. Furthermore, nucleosome formation occurs (after repair) only on damaged naked DNA (165-bp fragments) during a 1-h incubation in these extracts, even in the presence of a large excess of undamaged DNA. This is an example of selective nucleosome assembly by Xenopus nuclear extracts on a short linear DNA fragment containing a DNA lesion.

Aberrant mobility phenomena of the DNA repair protein XPA.

The DNA repair protein XPA recognizes a wide variety of bulky lesions and interacts with several other proteins during nucleotide excision repair. We recently identified regions of intrinsic order and disorder in full length Xenopus XPA (xXPA) protein using an experimental approach that combined time-resolved trypsin proteolysis and electrospray ionization interface coupled to a Fourier transform ion cyclotron resonance (ESI-FTICR) mass spectrometry (MS). MS data were consistent with the interpretation that xXPA contains no post-translational modifications. Here we characterize the discrepancy between the calculated molecular weight (31 kDa) for xXPA and its apparent molecular weight on SDS-PAGE (multiple bands from approximately 40-45 kDa) and gel filtration chromatography ( approximately 92 kDa), as well as the consequences of DNA binding on its anomalous mobility. Iodoacetamide treatment of xXPA prior to SDS-PAGE yielded a single 42-kDa band, showing that covalent modification of Cys did not correct aberrant mobility. Determination of sulfhydryl content in xXPA with Ellman's reagent revealed that all nine Cys in active protein are reduced. Unexpectedly, structural constraints induced by intramolecular glutaraldehyde crosslinks in xXPA produced a approximately 32-kDa monomer in closer agreement with its calculated molecular weight. To investigate whether binding to DNA alters xXPA's anomalous migration, we used gel filtration chromatography. For the first time, we purified stable complexes of xXPA and DNA +/- cisplatin +/- mismatches. xXPA showed at least 10-fold higher affinity for cisplatin DNA +/- mismatches compared to undamaged DNA +/- mismatches. In all cases, DNA binding did not correct xXPA's anomalous migration. To test predictions that a Glu-rich region (EEEEAEE) and/or disordered N- and C-terminal domains were responsible for xXPA's aberrant mobility, the molecular weights of partial proteolytic fragments from approximately 5 to 25 kDa separated by reverse-phase HPLC and precisely determined by ESI-FTICR MS were correlated with their migration on SDS-PAGE. Every partial tryptic fragment analyzed within this size range exhibited 10%-50% larger molecular weights than expected. Thus, both the disordered domains and the Glu-rich region in xXPA are primarily responsible for the aberrant mobility phenomena.

Identification of intrinsic order and disorder in the DNA repair protein XPA.

The DNA-repair protein XPA is required to recognize a wide variety of bulky lesions during nucleotide excision repair. Independent NMR solution structures of a human XPA fragment comprising approximately 40% of the full-length protein, the minimal DNA-binding domain, revealed that one-third of this molecule was disordered. To better characterize structural features of full-length XPA, we performed time-resolved trypsin proteolysis on active recombinant Xenopus XPA (xXPA). The resulting proteolytic fragments were analyzed by electrospray ionization interface coupled to a Fourier transform ion cyclotron resonance mass spectrometry and SDS-PAGE. The molecular weight of the full-length xXPA determined by mass spectrometry (30922.02 daltons) was consistent with that calculated from the sequence (30922.45 daltons). Moreover, the mass spectrometric data allowed the assignment of multiple xXPA fragments not resolvable by SDS-PAGE. The neural network program Predictor of Natural Disordered Regions (PONDR) applied to xXPA predicted extended disordered N- and C-terminal regions with an ordered internal core. This prediction agreed with our partial proteolysis results, thereby indicating that disorder in XPA shares sequence features with other well-characterized intrinsically unstructured proteins. Trypsin cleavages at 30 of the possible 48 sites were detected and no cleavage was observed in an internal region (Q85-I179) despite 14 possible cut sites. For the full-length xXPA, there was strong agreement among PONDR, partial proteolysis data, and the NMR structure for the corresponding XPA fragment.

Photoaffinity labeling of mouse fibroblast enzymes by a base excision repair intermediate. Evidence for the role of poly(ADP-ribose) polymerase-1 in DNA repair.

To examine the interaction of mammalian base excision repair (BER) enzymes with DNA intermediates formed during BER, we used a novel photoaffinity labeling probe and mouse embryonic fibroblast cellular extracts. The probe was formed in situ, using an end-labeled oligonucleotide containing a synthetic abasic site; this site was incised by apurinic/apyrimidinic endonuclease creating a nick with 3'-hydroxyl and 5'-reduced sugar phosphate groups at the margins, and then a dNMP carrying a photoreactive adduct was added to the 3'-hydroxyl group. With near-UV light (312 nm) exposure of the extract/probe mixture, six proteins were strongly labeled. Four of these include poly(ADP-ribose) polymerase-1 (PARP-1) and the BER participants flap endonuclease-1, DNA polymerase beta, and apurinic/apyrimidinic endonuclease. The amount of the probe cross-linked to PARP-1 was greater than that cross-linked to the other proteins. The specificity of PARP-1 labeling was examined using various competitor oligonucleotides and DNA probes with alternate structures. PARP-1 labeling was stronger with a DNA representing a BER intermediate than with a nick in double-stranded DNA. These results indicate that proteins interacting preferentially with a photoreactive BER intermediate can be selected from the crude cellular extract.

An integrated confocal and magnetic resonance microscope for cellular research.

Complementary data acquired with different microscopy techniques provide a basis for establishing a more comprehensive understanding of health and disease at a cellular level, particularly when data acquired with different methodologies can be correlated in both time and space. In this Communication, a brief description of a novel instrument capable of simultaneously performing confocal optical and magnetic resonance microscopy is presented, and the first combined images of live Xenopus laevis oocytes are shown. Also, the potential benefits of combined microscopy are discussed, and it is shown that the a priori knowledge of the high-resolution optical images can be used to enhance the boundary resolution and contrast of the MR images.

Nucleotide excision repair in oocyte nuclear extracts from Xenopus laevis.

We have developed efficient DNA repair extracts derived from the unusually large nuclei of Xenopus oocytes. These extracts use nucleotide excision repair (NER) to completely remove bulky adducts from DNA. There is very little or no synthesis on control, undamaged DNA, indicating the extracts do not have significant nonspecific nuclease activity, and repair of cyclobutane pyrimidine dimers (CPDs) occurs in the dark, indicating that NER, and not photolyase, is responsible for CPD repair. The extracts can be inactivated with antibodies specific to repair proteins and then repair activity can be restored by adding purified recombinant protein. Here we describe detailed protocols for preparing Xenopus nuclear repair extracts.

The oxidative DNA lesion 8,5'-(S)-cyclo-2'-deoxyadenosine is repaired by the nucleotide excision repair pathway and blocks gene expression in mammalian cells.

Xeroderma pigmentosum (XP) patients with inherited defects in nucleotide excision repair (NER) are unable to excise from their DNA bulky photoproducts induced by UV radiation and therefore develop accelerated actinic damage, including cancer, on sun-exposed tissue. Some XP patients also develop a characteristic neurodegeneration believed to result from their inability to repair neuronal DNA damaged by endogenous metabolites since the harmful UV radiation in sunlight does not reach neurons. Free radicals, which are abundant in neurons, induce DNA lesions that, if unrepaired, might cause the XP neurodegeneration. Searching for such a lesion, we developed a synthesis for 8,5'-(S)-cyclo-2'-deoxyadenosine (cyclo-dA), a free radical-induced bulky lesion, and incorporated it into DNA to test its repair in mammalian cell extracts and living cells. Using extracts of normal and mutant Chinese hamster ovary (CHO) cells to test for NER and adult rat brain extracts to test for base excision repair, we found that cyclo-dA is repaired by NER and not by base excision repair. We measured host cell reactivation, which reflects a cell's capacity for NER, by transfecting CHO and XP cells with DNA constructs containing a single cyclo-dA or a cyclobutane thymine dimer at a specific site on the transcribed strand of a luciferase reporter gene. We found that, like the cyclobutane thymine dimer, cyclo-dA is a strong block to gene expression in CHO and human cells. Cyclo-dA was repaired extremely poorly in NER-deficient CHO cells and in cells from patients in XP complementation group A with neurodegeneration. Based on these findings, we propose that cyclo-dA is a candidate for an endogenous DNA lesion that might contribute to neurodegeneration in XP.

Tight correlation between inhibition of DNA repair in vitro and transcription factor IIIA binding in a 5S ribosomal RNA gene.

UV-induced photoproducts (cyclobutane pyrimidine dimers, CPDs) in DNA are removed by nucleotide excision repair (NER), and the presence of transcription factors on DNA can restrict the accessibility of NER enzymes. We have investigatigated the modulation of NER in a gene promoter using the Xenopus transcription factor IIIA (TFIIIA)-5S rDNA complex and Xenopus oocyte nuclear extracts. TFIIIA alters CPD formation primarily in the transcribed strand of the 50 bp internal control region (ICR) of 5S rDNA. During NER in vitro, CPD removal is reduced at most sites in both strands of the ICR when TFIIIA is bound. Efficient repair occurs just outside the TFIIIA-binding site (within 10 bp), and in the absence of 5S rRNA transcription. Interestingly, three CPD sites within the ICR [+56, +75 (transcribed strand) and +73 (non-transcribed strand)] are repaired rapidly when TFIIIA is bound, while CPDs within approximately 5 bases of these sites are repaired much more slowly. CPDs at these three sites may partially displace TFIIIA, thereby enabling rapid repair. However, TFIIIA is not completely displaced during NER, at least at sites outside the ICR, even though the NER complex could be sterically hindered by TFIIIA. Such inefficient repair of transcription factor binding sites could increase mutation frequency in regulatory regions of genes.

Extended X-ray absorption fine structure evidence for a single metal binding domain in Xenopus laevis nucleotide excision repair protein XPA.

Nucleotide excision repair (NER) is an important cellular mechanism, conserved from bacteria to humans, responsible for eliminating multiple types of structurally distinct DNA lesions from the genome. The protein XPA appears to play a central role in NER, recognizing and/or verifying damaged DNA and recruiting other proteins, including RPA, ERCC1, and TFIIH, to repair the damage. Sequence analysis and genetic evidence suggest that zinc, which is essential for DNA binding, is associated with a C4-type motif, C-X2-C-X17-C-X2-C. Sequence analysis suggests that a second, H2C2-type zinc-binding motif may be present near the C-terminal. Seventy percent of the amino acid sequence of Xenopus laevis XPA (xXPA) is identical to human XPA and both putative zinc-binding motifs are conserved in all known XPA proteins. Electrospray ionization-mass spectroscopy data show that xXPA contains only one zinc atom per molecule. EXAFS spectra collected on full-length xXPA in frozen (77 K) 15% glycerol aqueous solution unequivocally show that the zinc atom is coordinated to four sulfur atoms with an average Zn--S bond length of 2.33 +/- 0.02 A. Together, the EXAFS and mass spectroscopy data indicate that xXPA contains just one C4-type zinc-binding motif.

DNA polymerases alpha and beta are required for DNA repair in an efficient nuclear extract from Xenopus oocytes.

Xenopus oocytes and an oocyte nuclear extract efficiently repair the bulky DNA lesions cyclobutane pyrimidine dimers,(6-4) photoproducts, and N-acetoxy-2-aminofluorene (AAF) adducts by an excision repair mechanism. Nearly all (>95%) of the input damaged DNA was repaired within 5 h in both injected cells and extracts with no significant incorporation of label into control undamaged DNA. Remarkably, more than 10(10) cyclobutane pyrimidine dimers or(6-4) photoproducts are repaired/nuclei. The extracts are free from nuclease activity, and repair is independent of exogenous light. Both the high efficiency and DNA polymerase requirements of this system appear to be different from extracts derived from human cells. We demonstrated a requirement for DNA polymerases alpha and beta in repair of both photoproducts and AAF by inhibiting repair with several independent antibodies specific to either DNA polymerases alpha or beta and then restoring repair by adding the appropriate purified polymerase. Repair is inhibited by aphidicolin at concentrations specific for blocking DNA polymerase alpha and dideoxynucleotide triphosphates at concentrations specific for inhibiting DNA polymerase beta.

Arrested DNA replication in Xenopus and release by Escherichia coli mutagenesis proteins.

Xenopus oocytes and oocyte nuclear extracts repair ultraviolet photoproducts on double-stranded (ds) DNA and replicate single-stranded (ss) to ds DNA. M13 ss DNA molecules containing cyclobutane pyrimidine dimers were maintained but not replicated in Xenopus oocytes yet were replicated in progesterone-matured oocytes. The replication arrest functioned only in cis. The replication arrest was alleviated by injection into oocytes of messenger RNAs encoding the prokaryotic mutagenesis proteins UmuD'C or MucA'B. These results may help explain how cells stabilize repair or replication events on DNA with unrepairable lesions.

Angiogenin is a cytotoxic, tRNA-specific ribonuclease in the RNase A superfamily.

Angiogenin is a 14.4-kDa human plasma protein with 65% homology to RNase A that retains the key active site residues and three of the four RNase A disulfide bonds. We demonstrate that recombinant angiogenin functions as a cytotoxic tRNA-specific RNase in cell-free lysates and when injected into Xenopus oocytes. Inhibition of protein synthesis by angiogenin correlates with degradation of endogenous oocyte tRNA. Exogenous, radiolabeled tRNA is also hydrolyzed by angiogenin, whereas oocyte rRNA and mRNA are not detectably degraded by angiogenin. Protein synthesis was restored to angiogenin-injected oocytes by injecting the RNase inhibitor RNasin plus total Xenopus or calf liver tRNAs, thereby demonstrating that the tRNA degradation induced by angiogenin was the sole cause of cytotoxicity. A similar tRNA-reversible inhibition of protein synthesis was seen in rabbit reticulocyte lysates. Angiogenin therefore appears to be a specific cellular tRNase, whereas five homologues in the RNase A superfamily lack angiogenin's specificity for tRNA. One of these homologues purified from human eosinophils, eosinophil-derived neurotoxin, nonspecifically degrades oocyte RNA similar to RNase A and is also cytotoxic at very low concentrations.

DNA polymerase beta and DNA synthesis in Xenopus oocytes and in a nuclear extract.

The identities of the DNA polymerases required for conversion of single-strand (ss) M13 DNA to double-strand (ds) M13 DNA were examined both in injected Xenopus laevis oocytes and in an oocyte nuclear extract. Inhibitors and antibodies specific to DNA polymerases alpha and beta were used. In nuclear extracts, inhibition by the antibody to polymerase beta could be reversed by purified polymerase beta. The polymerase beta inhibitors, dideoxythymidine triphosphate (ddTTP) and dideoxycytidine triphosphate (ddCTP), also blocked DNA synthesis and indicated that polymerase beta is involved in the conversion of ssDNA to dsDNA. These results also may have particular significance for emerging evidence of an ssDNA replication mode in eukaryotic cells.

Cell surface and intracellular functions for ricin galactose binding.

The role of the two galactose binding sites of ricin B chain in ricin toxicity was evaluated by studying a series of ricin point mutants. Wild-type (WT) ricin and three ricin B chain point mutants having mutations in either 1) the first galactose binding domain (site 1 mutant, Met in place of Lys-40 and Gly in place of Asn-46), 2) the second galactose binding domain (site 2 mutant, Gly in place of Asn-255), or 3) both galactose binding domains (double site mutant containing all three amino acid replacements formerly stated) were expressed in Xenopus oocytes and then reassociated with recombinant ricin A chain. The different ricin B chains were mannosylated to the same extent. Cytotoxicity of these toxins was evaluated when cell entry was mediated either by galactose-containing receptors or through an alternate receptor, the mannose receptor of macrophages. WT ricin and each of the single domain mutants was able to kill Vero cells following uptake by galactose containing receptors. Lactose blocked the toxicity of each of these ricins. Site 1 and 2 mutants were 20-40 times less potent than WT ricin, and the double site mutant had no detectable cytotoxicity. WT ricin, the site 1 mutant, and the site 2 mutant also inhibited protein synthesis of mannose receptor-containing cells. Ricin can enter these cells through either a cell-surface galactose-containing receptor or through the mannose receptor. By including lactose in the cell medium, galactose-containing receptor-mediated uptake is blocked and cytotoxicity occurs solely via the mannose receptor. WT ricin, site 1, and site 2 mutants were cytotoxic to macrophages in the presence of lactose with the relative potency, WT greater than site 2 mutant greater than site 1 mutant. The double site mutant lacked cytotoxicity either in the absence or presence of lactose. Thus, even for mannose receptor-mediated toxicity of ricin, at least one galactose binding site remains necessary for cytotoxicity and two galactose binding sites further increases potency. These results are consistent with the model that the ricin B chain galactose binding activity plays a role not only in cell surface binding but also intracellularly for ricin cytotoxicity.

Cytotoxic potential of ribonuclease and ribonuclease hybrid proteins.

Pancreatic RNase injected into Xenopus oocytes abolishes protein synthesis at concentrations comparable to the toxin ricin yet has no effect on oocyte protein synthesis when added to the extracellular medium. Therefore RNase behaves like a potent toxin when directed into a cell. To explore the cytotoxic potential of RNase toward mammalian cells, bovine pancreatic ribonuclease A was coupled via a disulfide bond to human transferrin or antibodies to the transferrin receptor. The RNase hybrid proteins were cytotoxic to K562 human erythroleukemia cells in vitro with an IC50 around 10(-7) M whereas greater than 10(-5) M native RNase was required to inhibit protein synthesis. Cytotoxicity requires both components of the conjugate since excess transferrin or ribonuclease inhibitors added to the medium protected the cells from the transferrin-RNase toxicity. Compounds that interfere with transferrin receptor cycling and compartmentalization such as ammonium chloride decreased the cytotoxicity of transferrin-RNase. After a dose-dependent lag period inactivation of protein synthesis by transferrin-RNase followed a first-order decay constant. In a clonogenic assay that measures the extent of cell death 1 x 10(-6) M transferrin-RNase killed at least 4 logs or 99.99% of the cells whereas 70 x 10(-6) M RNase was nontoxic. These results show that RNase coupled to a ligand can be cytotoxic. Human ribonucleases coupled to antibodies also may exhibit receptor-mediated toxicities providing a new approach to selective cell killing possibly with less systemic toxicity and importantly less immunogenicity than the currently employed ligand-toxin conjugates.

Comparison of RNases and toxins upon injection into Xenopus oocytes.

Several toxins abolish cellular protein synthesis by attacking specific sites in 28 S RNA. One of these toxins, alpha-sarcin, is an RNase that also cleaves nonspecifically on the 3' side of purines in deproteinized RNA. Several other RNases were injected into Xenopus oocytes, examined for their ability to abolish protein synthesis, and compared with alpha-sarcin and ricin. Surprisingly, pancreatic RNase A or B abolished oocyte protein synthesis at concentrations (approximately 0.03 nM) comparable to, or lower than, the amount of alpha-sarcin (approximately 2 nM) or ricin (approximately 0.07 nM) required to abolish protein synthesis. RNases S and T1 only inhibited oocyte protein synthesis when used at concentrations approximately 10 x higher than RNase A whereas RNases C, T2, U2, and nuclease P1 required concentrations approximately 100 times higher than RNase A to abolish protein synthesis. There was a direct correlation between the degradation of oocyte RNA and the inhibition of protein synthesis. The RNase inhibitors RNasin and Inhibit-Ace injected into the oocyte both prevented RNase A from hydrolyzing oocyte rRNA and abolishing protein synthesis. Enzymatically inactive oxidized RNase A did not inhibit protein synthesis when injected into the oocyte. None of the RNases or alpha-sarcin abolished protein synthesis when added to oocyte extracellular medium. Angiogenin is a human plasma protein that induces blood vessel formation in chick embryos, has 35% amino acid identity with RNase A, and cleaves 18 S and 28 S RNA in rabbit reticulocyte lysates (St. Clair, D. K., Rybak, S. M., Riordan, J. F. & Vallee, B. L. (1988) Biochemistry 27, 7263-7268, and references therein). Recombinant angiogenin injected into oocytes abolished protein synthesis, and this toxic effect was inhibited by RNasin but was not inhibited by Inhibit-Ace. Unlike RNase A and the other nucleases that hydrolyzed cellular rRNA, no cleavage of 18 or 28 S RNA by recombinant angiogenin was seen at concentrations 100 x greater than necessary to abolish protein synthesis. Recombinant angiogenin must selectively attack specific RNA(s) or another target in the cell.

Excision repair of UV-damaged plasmid DNA in Xenopus oocytes is mediated by DNA polymerase alpha (and/or delta).

We studied DNA repair by injecting plasmids containing random pyrimidine dimers into Xenopus oocytes. We demonstrated excision repair by recovering plasmids and analyzing them with T4 UV endonuclease treatment and alkaline agarose gel electrophoresis. The mechanism for excision repair of these plasmids appears to be processive, rather than distributive, since repair occurs in 'all or none' fashion. At less than 4-5 dimers/plasmid, nearly all repair occurs within 4-6 hours (approximately 10(10) dimers repaired per oocyte); the oocyte, therefore, has abundant repair activity. Specific antibodies and inhibitors were used to determine enzymes involved in repair. We conclude that DNA polymerase alpha (and/or delta) is required because repair is inhibited by antibodies to human DNA polymerase alpha, as well as by aphidicolin, an inhibitor of polymerases alpha (and/or delta). Repair was not inhibited by hydroxyurea, cytosine beta-D-arabinofuranoside, or inhibitors of topoisomerase II (novobiocin). Oocyte repair does not activate semi-conservative DNA replication, nor is protein synthesis required. Photoreactivation cannot account for repair because dimer removal is independent of exogenous light.