Repair of thymine glycol by hNth1 and hNeil1 is modulated by base pairing and cis-trans epimerization

Oxidation of thymine yields 5,6-dihydroxy-5,6-dihydrothymine (thymine glycol. Tg) which, as cis 5S,6R and 5R,6S 2'-deoxyribonucleoside diastereoisomers (dTg1, dTg2), are in equilibrium with their trans 5S,6S and 5R,6R epimers. The stereoselective excision of Tg from DNA by the mammalian orthologs of E. coli DNA N-glycosylase/AP lyases Nth and Nei was reported using substrates in which Tg opposed adenine. Since we showed that Tg is the major product of oxidation of 5-methylcytosine, we asked if the opposing purine influenced stereospecific enzymatic excision. The human ortholog hNth1 released Tg2 much more rapidly than Tg1 regardless of the opposing purine. In contrast, hNeil1 released Tg non-stereoselectively, but the rate of excision was much greater when Tg opposed guanine. Remarkably, the kinetics of excision of Tg by hNth1 and hNeil1 were biphasic, describing a double exponential curve which yielded two rate constants. We suggest that the greater rate constant describes the rate of enzymatic excision of Tg. The smaller rate constant represents the equilibrium constant for the cis and trans epimerization of dTg1 and dTg2 in high molecular weight DNA. Thus, only one of the epimers of dTg1 and dTg2 are enzymatically processed but it is not yet known whether it is cis or trans. Thus, base excision repair of Tg in mammals is mediated by at least two DNA N-glycosylase/AP lyases which are affected by the nature of the diastereoisomer of dTg, the rate of cis-trans epimerization of each diastereoisomer, and the nature of the opposing purine.

Substrate specificity of human endonuclease III (hNTH1). Effect of human APE1 on hNTH1 activity

Base excision repair of oxidized pyrimidines in human DNA is initiated by the DNA N-glycosylase/apurinic/apyrimidinic (AP) lyase, human NTH1 (hNTH1), the homolog of Escherichia coli endonuclease III (Nth). In contrast to Nth, the DNA N-glycosylase activity of hNTH1 is 7-fold greater than its AP lyase activity when the DNA substrate contains a thymine glycol (Tg) opposite adenine (Tg:A) (Marenstein, D. R., Ocampo, M. T. A., Chan, M. K., Altamirano, A., Basu, A. K., Boorstein, R. J., Cunningham, R. P., and Teebor, G. W. (2001) J. Biol. Chem. 276, 21242-21249). When Tg is opposite guanine (Tg:G), the two activities are of the same specific activity as the AP lyase activity of hNTH1 against Tg:A (Ocampo, M. T. A., Chaung, W., Marenstein, D. R., Chan, M. K., Altamirano, A., Basu, A. K., Boorstein, R. J., Cunningham, R. P., and Teebor, G. W. (2002) Mol. Cell. Biol. 22, 6111-6121). We demonstrate here that hNTH1 was inhibited by the product of its DNA N-glycosylase activity directed against Tg:G, the AP:G site. In contrast, hNTH1 was not as inhibited by the AP:A site arising from release of Tg from Tg:A. Addition of human APE1 (AP endonuclease-1) increased dissociation of hNTH1 from the DNA N-glycosylase-generated AP:A site, resulting in abrogation of AP lyase activity and an increase in turnover of the DNA N-glycosylase activity of hNTH1. Addition of APE1 did not abrogate hNTH1 AP lyase activity against Tg:G. The stimulatory protein YB-1 (Marenstein et al.), added to APE1, resulted in an additive increase in both activities of hNTH1 regardless of base pairing. Tg:A is formed by oxidative attack on thymine opposite adenine. Tg:G is formed by oxidative attack on 5-methylcytosine opposite guanine (Zuo, S., Boorstein, R. J., and Teebor, G. W. (1995) Nucleic Acids Res. 23, 3239-3243). It is possible that the in vitro substrate selectivity of mammalian NTH1 and the concomitant selective stimulation of activity by APE1 are indicative of selective repair of oxidative damage in different regions of the genome.

Expression of the human DNA glycosylase hSMUG1 in Trypanosoma brucei causes DNA damage and interferes with J biosynthesis

In kinetoplastid flagellates such as Trypanosoma brucei, a small percentage of the thymine residues in the nuclear DNA is replaced by the modified base beta-D-glucosyl-hydroxymethyluracil (J), mostly in repetitive sequences like the telomeric GGGTTA repeats. In addition, traces of 5-hydroxymethyluracil (HOMeUra) are present. Previous work has suggested that J is synthesised in two steps via HOMedU as an intermediate, but as J synthesising enzymes have not yet been identified, the biosynthetic pathway remains unclear. To test a model in which HOMeUra functions as a precursor of J, we introduced an inducible gene for the human DNA glycosylase hSMUG1 into bloodstream form T.brucei. In higher eukaryotes SMUG1 excises HOMeUra as part of the base excision repair system. We show that expression of the gene in T.brucei leads to massive DNA damage in J-modified sequences and results in cell cycle arrest and, eventually, death. hSMUG1 also reduces the J content of the trypanosome DNA. This work supports the idea that HOMeUra is a precursor of J, freely accessible to a DNA glycosylase.

Targeted deletion of mNth1 reveals a novel DNA repair enzyme activity

DNA N-glycosylase/AP (apurinic/apyrimidinic) lyase enzymes of the endonuclease III family (nth in Escherichia coli and Nth1 in mammalian organisms) initiate DNA base excision repair of oxidized ring saturated pyrimidine residues. We generated a null mouse (mNth1(-/-)) by gene targeting. After almost 2 years, such mice exhibited no overt abnormalities. Tissues of mNth1(-/-) mice contained an enzymatic activity which cleaved DNA at sites of oxidized thymine residues (thymine glycol [Tg]). The activity was greater when Tg was paired with G than with A. This is in contrast to Nth1, which is more active against Tg:A pairs than Tg:G pairs. We suggest that there is a back-up mammalian repair activity which attacks Tg:G pairs with much greater efficiency than Tg:A pairs. The significance of this activity may relate to repair of oxidized 5-methyl cytosine residues (5meCyt). It was shown previously (S. Zuo, R. J. Boorstein, and G. W. Teebor, Nucleic Acids Res. 23:3239-3243, 1995) that both ionizing radiation and chemical oxidation yielded Tg from 5meCyt residues in DNA. Thus, this previously undescribed, and hence novel, back-up enzyme activity may function to repair oxidized 5meCyt residues in DNA while also being sufficient to compensate for the loss of Nth1 in the mutant mice, thereby explaining the noninformative phenotype.

Definitive identification of mammalian 5-hydroxymethyluracil DNA N-glycosylase activity as SMUG1

Purification from calf thymus of a DNA N-glycosylase activity (HMUDG) that released 5-hydroxymethyluracil (5hmUra) from the DNA of Bacillus subtilis phage SPO1 was undertaken. Analysis of the most purified fraction by SDS-polyacrylamide gel electrophoresis revealed a multiplicity of protein species making it impossible to identify HMUDG by inspection. Therefore, we renatured the enzyme after SDS-polyacrylamide gel electrophoresis and assayed slices of the gel for DNA N-glycosylase activity directed against 5hmUra. Maximum enzymatic activity was identified between molecular mass markers 30 and 34 kDa. Protein was extracted from gel slices and subjected to tryptic digestion and analysis by mass spectrometry. Analysis revealed the presence of 11 peptides that were homologous or identical to the sequence of the recently characterized human single-stranded monofunctional uracil DNA N-glycosylase (hSMUG1). The cDNA of hSMUG1 was isolated and expressed as a recombinant glutathione S-transferase fusion protein that was shown to release 5hmUra with 20x the specific activity of the most purified bovine fraction. We conclude that hSMUG1 and HMUDG are the same protein.

Stimulation of human endonuclease III by Y box-binding protein 1 (DNA-binding protein B). Interaction between a base excision repair enzyme and a transcription factor

Human endonuclease III (hNth1) is a DNA glycosylase/apurinic/apyrimidinic (AP) lyase that initiates base excision repair of pyrimidines modified by reactive oxygen species, ionizing, and ultraviolet radiation. Using duplex 2'-deoxyribose oligonucleotides containing an abasic (AP) site, a thymine glycol, or a 5-hydroxyuracil residue as substrates, we found the AP lyase activity of hNth1 was 7 times slower than its DNA glycosylase activity, similar to results reported for murine and human 8-oxoguanine-DNA glycosylase, which are also members of the endonuclease III family. This difference in rates contrasts with the equality of rates found in Escherichia coli and Saccharomyces cerevisiae endonuclease III homologs. A yeast two-hybrid screen for potential modulators of hNth1 activity revealed interaction with the damage-inducible transcription factor Y box-binding protein 1 (YB-1), also identified as DNA-binding protein B (DbpB). The in vitro addition of His(6)YB-1 to hNth1 increased the rate of DNA glycosylase and AP lyase activity. Analysis revealed that YB-1 affects the steady state equilibrium between the covalent hNth1-AP site Schiff base ES intermediate and the noncovalent ES intermediate containing the AP aldehydic sugar and the epsilon-amino group of the hNth1 active site lysine. This equilibrium may be a checkpoint in modulating hNth1 activity.

Excessive base excision repair of 5-hydroxymethyluracil from DNA induces apoptosis in Chinese hamster V79 cells containing mutant p53

We have demonstrated previously that the toxicity of 5-hydroxymethyl-2'-deoxyuridine (hmdUrd) to Chinese hamster fibroblasts (V79 cells) results from enzymatic removal of large numbers of hydroxymethyluracil residues from the DNA backbone [Boorstein,R. et al. (1992) Mol. Cell. Biol., 12, 5536-5540]. Here we report that a significant portion of the hmdUrd-induced cell death that is dependent on DNA base excision repair in V79 cells is apoptosis. Incubation of V79 cells with pharmacologically relevant concentrations of hmdUrd resulted in the characteristic changes of apoptosis as measured by gel electrophoresis, flow cytometry and phase contrast microscopy. However, hmdUrd did not induce apoptosis in V79mut1 cells, which are deficient in DNA base excision repair of 5-hydroxymethyluracil (hmUra). Apoptosis was not prevented by addition of 3-aminobenzamide, which inhibits synthesis of poly(ADP-ribose) from NAD, indicating that apoptosis was not the direct consequence of NAD depletion. Pulsed field gel electrophoresis indicated that hmdUrd treatment resulted in high molecular weight (2.2-4.5 Mb) DNA double-strand breaks prior to formation of internucleosomal ladders in V79 cells. Simultaneous measurement of DNA strand breaks with bromodeoxyuridine/terminal deoxynucleotidyl transferase-fluorescein isothiocyanate labeling and of cell cycle distribution indicated that cells with DNA strand breaks accumulated in late S/G(2) and that hmdUrd-treated cells underwent apoptosis after arrest in late S/G(2) phase. Our results indicate that excessive DNA base excision repair results in the generation of high molecular weight DNA double-strand breaks and eventually leads to apoptosis in V79 cells. Thus, delayed apoptosis following DNA damage can be a consequence of excessive DNA repair activity. Immunochemical analysis showed that both V79 and V79mut1 cells contained mutant p53, indicating that apoptosis induced by DNA base excision repair can be independent of p53.

5-chloro-2'-deoxyuridine cytotoxicity results from base excision repair of uracil subsequent to thymidylate synthase inhibition

The lack of a phenotypic alteration of 5-hydroxymethyluracil (hmUra) DNA glycosylase (hmUDG) deficient Chinese hamster V79mut1 cells exposed to DNA-damaging agents known to produce hmUra has raised the question whether there might be DNA substrates other than hmUra for hmUDG. Based on the structural similarity between 5-chlorouracil (ClUra) and hmUra and the observations that 5-chloro-2'-deoxyuridine (CldUrd) induces base excision repair (BER) events, we asked whether hmUDG or some other DNA BER enzyme is responsible for the removal of ClUra from DNA. An in vivo flow cytometry assay with FITC-anti-BrdUrd (which cross-reacts with CldUrd) showed that exogenous CldUrd is incorporated into DNA. However, both in vivo and in vitro experiments indicated that ClUra is not excised from DNA by hmUDG or other DNA glycosylase activities. The absence of removal of ClUra by hmUDG raised the question whether DNA strand breaks occurred subsequent to thymidylate synthase inhibition, leading to deoxyuridine incorporation, followed by cleavage of uracil from DNA by uracil DNA glycosylase (UDG). An in vivo thymidylate synthase activity assay in V79 cells demonstrated that CldUrd treatment inhibits thymidylate synthase as effectively as 5-fluoro-2'-deoxyuridine (FdUrd) treatment. Uracil, a known UDG inhibitor, partially reverses the cytotoxic effects of CldUrd on V79 cells, thus confirming that CldUrd induced cytotoxicity is a result of UDG activity. Our results demonstrated that while CldUrd is not directly repaired from DNA, its cytotoxicity is directly due to the UDG removing uracil subsequent to inhibition of thymidylate synthase by CldUMP.

Lack of phenotypic alteration of hmUra-DNA glycosylase-deficient hamster cells exposed to DNA-damaging agents

V79mut1 cells are resistant to the toxic effects of 5-hydroxymethyl-2'-deoxyuridine (hmdUrd) and are deficient in the DNA repair enzyme hydroxymethyluracil-DNA glycosylase (hmUDG). We have therefore proposed that the toxicity of hmdUrd results from the repair of the lesion from DNA. In order to clarify the biological role of hmUDG, we have determined whether the repair-deficient cells showed resistance or sensitivity to the toxic or mutagenic effects of other DNA-damaging agents. Cells were exposed to hmdUrd, ionizing or ultraviolet radiation, to the alkylating agent MNNG, and to oxidative stress produced by hypoxanthine/xanthine oxidase, glucose/glucose oxidase, nitric oxide donor SNAP, or to H2O2. The V79mut1 cells did not show increased mutagenesis in response to hmdUrd. Relative to the V79 parent cells, the V79mut1 cells were not markedly altered in sensitivity to oxidizing agents and ionizing radiation (which produce hmdUra in DNA). The repair-deficient cells wee equally sensitive as the parent V79 cells to DNA damage induced by ultraviolet radiation or by MNNG. No significant differences were seen between the parent and the repair-deficient cells in terms of synthesis of poly(ADP-ribose) in response to damage or in their sensitization to 3-aminobenzamide. Thus, the loss of the 5-hydroxymethyluracil (hmUra)-DNA glycosylase activity in mammalian cells in culture confers no obvious deleterious effect on cell survival or mutagenicity in response to a wide range of DNA damage. These studies indicate that the major lesion known to be repaired by hmUra-DNA glycosylase, an hmUra residue replacing thymine, is produced in cells only in small quantities as the result of exposure to common DNA-damaging agents. These results raise the possibility that hmUra-DNA glycosylase may have evolved to respond to other lesions than hmUra residues formed from the oxidation of thymine.

Cloning and expression of the cDNA encoding the human homologue of the DNA repair enzyme, Escherichia coli endonuclease III

We previously purified a bovine pyrimidine hydrate-thymine glycol DNA glycosylase/AP lyase. The amino acid sequence of tryptic bovine peptides was homologous to Escherichia coli endonuclease III, theoretical proteins of Saccharomyces cerevisiae and Caenorhabditis elegans, and the translated sequences of rat and human 3'-expressed sequence tags (3'-ESTs) (Hilbert, T. P., Boorstein, R. J., Kung, H. C., Bolton, P. H., Xing, D., Cunningham, R. P., Teebor, G. W. (1996) Biochemistry 35, 2505-2511). Now the human 3'-EST was used to isolate the cDNA clone encoding the human enzyme, which, when expressed as a GST-fusion protein, demonstrated thymine glycol-DNA glycosylase activity and, after incubation with NaCNBH3, became irreversibly cross-linked to a thymine glycol-containing oligodeoxynucleotide, a reaction characteristic of DNA glycosylase/AP lyases. Amino acids within the active site, DNA binding domains, and [4Fe-4S] cluster of endonuclease III are conserved in the human enzyme. The gene for the human enzyme was localized to chromosome 16p13.2-.3. Genomic sequences encoding putative endonuclease III homologues are present in bacteria, archeons, and eukaryotes. The ubiquitous distribution of endonuclease III-like proteins suggests that the 5,6-double bond of pyrimidines is subject to oxidation, reduction, and/or hydration in the DNA of organisms of all biologic domains and that the resulting modified pyrimidines are deleterious to the organism.

The p53 status of Chinese hamster V79 cells frequently used for studies on DNA damage and DNA repair

Chinese hamster lung fibroblast V79 cells have been widely used in studies of DNA damage and DNA repair. Since the p53 gene is involved in normal responses to DNA damage, we have analyzed the molecular genetics and functional status of p53 in V79 cells and primary Chinese hamster embryonic fibroblast (CHEF) cells. The coding product of the p53 gene in CHEF cells was 76 and 75% homologous to human and mouse p53 respectively, and was 95% homologous to the Syrian hamster cells. The V79 p53 sequence contained two point mutations located within a presumed DNA binding domain, as compared with the CHEF cells. Additional immunocytochemical and molecular studies confirmed that the p53 protein in V79 cells was mutated and nonfunctional. Our results indicate that caution should be used in interpreting studies of DNA damage, DNA repair and apoptosis in V79 cells.

Molecular spectrum of mutations induced by 5-hydroxymethyl-2'-deoxyuridine in (CHO)-PL61 cells

We have utilized (CHO)-PL61 cells to characterize the mutations produced in mammalian cells by exogenous treatment with the nucleoside 5-hydroxymethyl-2'-deoxyuridine (hmdUrd). HmdUrd is incorporated into DNA as a thymidine analogue and is removed by the repair enzyme hmUra-DNA glycosylase. PL61 cells are hprt(-) and contain adjacent single copies of the Escherichia coli gpt and neo genes (gpt+, neo+) separated by 2 kb, rendering the cells thioguanine sensitive (TGs) and geneticin resistant (G418r). Cells were exposed to hmdUrd and the colonies resistant to thioguanine or thioguanine and G418 were selected. Selection in thioguanine alone (TGr/gpt(-)) allows the growth of all gpt(-) mutants (small, intermediate and large deletions/insertions and point mutations) while selection in thioguanine and G418 (TGr/gpt(-), G418r/neo+) prevents survival of colonies containing vary large deletions of the gpt gene that include the neo gene. To confirm the types of mutation at the molecular level, the gpt gene was amplified from mutants' genomic DNA by PCR, and the amplified DNA was sequenced directly by the dideoxy method. Our study showed that 4 microM hmdUrd induced mutations to TGr/gpt(-) at a rate 3-4 times that of control, but showed no marked increase in mutation to TGr/gpt(-), G418r/neo+. The predominant type of hmdUrd induced mutation in the thioguanine resistant cells at the gpt locus was complete loss of the gpt gene resulting from a large deletion. Background mutations were generally point mutations or small insertion/deletion mutations. We propose that hmdUrd induces large/intermediate deletions as a major type of mutations in mammalian cells as a consequence of DNA repair, and not as a result of misincorporation or mispairing, suggesting that base excision repair by itself can lead to large deletion mutagenesis.

Purification of a mammalian homologue of Escherichia coli endonuclease III: identification of a bovine pyrimidine hydrate-thymine glycol DNAse/AP lyase by irreversible cross linking to a thymine glycol-containing oligoxynucleotide

We purified a homologue of the Escherichia coli DNA repair enzyme endo nuclease III 5000-fold from calf thymus which, like endonuclease III, demonstrates DNA-glycosylase activity against pyrimidine hydrates and thymine glycol and AP lyase activity (DNA strand cleavage at AP sites via beta-elimination). The functional similarity between the enzymes suggested a strategy for definitive identification of the bovine protein based on the nature of its enzyme-substrate (ES) intermediate. Prokaryotic DNA glycosylase/AP lyases function through N-acylimine (Schiff's base) ES intermediates which, upon chemical reduction to stable secondary amines, irreversibly cross link the enzyme to oligodeoxynucleotides containing substrate modified bases. We incubated endonuclease III with a 32P- labeled thymine glycol-containing oligodeoxynucleotide in the presence of NaCNBH3. This resulted in an increase in the apparent molecular weight of the enzyme by SDS-PAGE. Phosphorimaging confirmed irreversible cross linking between enzyme and DNA. Identical treatment of the most purified bovine enzyme fraction resulted in irreversible cross linking of the oligodeoxynucleotide to a predominant 31 kDa species. Amino acid analysis of the 31 kDa species revealed homology to the predicted amino acid sequence of a Caenorhabditis elegans 27.8 kDa protein which, in turn, has homology to endonuclease III. The translated amino acid sequences of two partial 3' cDNAs, from Homo sapiens and Rattus sp., also demonstrate homology to the C. elegans and bovine sequences suggesting a homologous family of endonuclease III-like DNA repair enzymes is present throughout phylogeny.

Comparison of the effects of UV irradiation on 5-methyl-substituted and unsubstituted pyrimidines in alternating pyrimidine-purine sequences in DNA

We previously demonstrated the UV-induced formation of cytosine hydrate in DNA and its deamination product, uracil hydrate, via their release from the DNA backbone by the DNA glycosylase activity of Escherichia coli endonuclease III. Subsequently, endonuclease III-mediated release of thymine hydrate from UV-irradiated poly(dA-dT) was reported. Therefore, we asked whether 5-methylcytosine residues in DNA underwent photohydration and deamination to thymine hydrate in analogy to UV-induced deamination of cytosine. An alternating DNA copolymer containing 5-methylcytosine was irradiated with UVC and incubated with endonuclease III. No 5-methylcytosine hydrate was released. Instead, UV-induced nonenzymatic release of 5-methylcytosine occurred. Similarly, incubation of UV-irradiated poly(dA-dT) with endonuclease III did not release thymine hydrate; nonenzymatic release of thymine occurred. Nonenzymatic release of 5-methylpyrimidines was oxygen dependent, enhanced by ferric ion and inhibited by free radical scavengers. In contrast, photohydration of cytosine was oxygen independent, and only small amounts of cytosine were nonenzymatically released. Thus, 5-methylpyrimidine residues within alternating Pu-Py sequences in DNA do not undergo photohydration, but instead undergo cleavage of their N-glycosyl bonds yielding abasic (AP) sites. The inability to repair such AP sites may explain the UV sensitivity of E. coli xthnfo mutants, which lack AP endonuclease activity. We suggest that N-glycosyl bond cleavage is mediated by radical species formed via transfer of an electron from UV-excited triplet 5-methylpyrimidines to ground state oxygen and/or ferric ions.

Oxidative damage to 5-methylcytosine in DNA

Exposure of pyrimidines of DNA to ionizing radiation under aerobic conditions or oxidizing agents results in attack on the 5,6 double bond of the pyrimidine ring or on the exocyclic 5-methyl group. The primary product of oxidation of the 5,6 double bond of thymine is thymine glycol, while oxidation of the 5-methyl group yields 5-hydroxymethyluracil. Oxidation of the 5,6 double bond of cytosine yields cytosine glycol, which decomposes to 5-hydroxycytosine, 5-hydroxyuracil and uracil glycol, all of which are repaired in DNA by Escherichia coli endonuclease III. We now describe the products of oxidation of 5-methylcytosine in DNA. Poly(dG-[3H]dmC) was gamma-irradiated or oxidized with hydrogen peroxide in the presence of Fe3+ and ascorbic acid. The oxidized co-polymer was incubated with endonuclease III or 5-hydroxymethyluracil-DNA glycosylase, to determine whether repairable products were formed, or digested to 2'-deoxyribonucleosides, to determine the total complement of oxidative products. Oxidative attack on 5-methylcytosine resulted primarily in formation of thymine glycol. The radiogenic yield of thymine glycol in poly(dG-dmC) was the same as that in poly(dA-dT), demonstrating that 5-methylcytosine residues in DNA were equally susceptible to radiation-induced oxidation as were thymine residues.

DNA base excision repair of 5-hydroxymethyl-2'-deoxyuridine stimulates poly(ADP-ribose) synthesis in Chinese hamster cells

5-Hydroxymethyl-2'-deoxyuridine (hmdUrd) is incorporated into DNA as a thymidine analog resulting in extensive substitution of thymine residues with 5-hydroxymethyluracil (hmUra) residues. These hmUra residues are then subject to excision by action of hmUra-DNA glycosylase. 3-Aminobenzamide (3AB), an inhibitor of poly(ADP-ribose) synthesis, is toxic to cells that incorporate and repair hmdUrd. To demonstrate that incorporation and repair of hmdUrd stimulates synthesis of poly(ADP-ribose) from intracellular NAD, V79 hamster cells were treated with hmdUrd and intracellular NAD levels were measured. Following hmdUrd treatment, NAD levels fell markedly (80-90%) within 4 h and remained low for at least 10 h, before partially recovering by 24 h. The degree of NAD lowering was dose dependent and paralleled net hmdUrd incorporation. The NAD lowering was largely prevented by concurrent treatment with 4 mM 3AB. No effects on NAD levels were seen following treatment with deoxythymidine or bromodeoxyuridine, which are incorporated into DNA but, in contrast to hmdUrd, are not repaired. When the incorporation of hmdUrd into DNA was blocked with hydroxyurea or aphidicolin, no NAD lowering was seen. HmdUrd also did not produce lowering of NAD concentrations in mutant cell strains deficient in the ability either to incorporate hmdUrd into DNA or to repair hmdUrd from DNA. These results demonstrate that synthesis of poly(ADP-ribose) resulted directly from the incorporation into DNA of the nucleoside hmdUrd and its subsequent repair. These results unequivocally demonstrate that the initiation of normal DNA base excision repair by itself, and not DNA damage per se, is a sufficient stimulus for the induction of poly(ADP-ribose) synthesis.

Toxicity of camptothecin to Chinese hamster cells containing 5-hydroxymethyl-2'-deoxyuridine in their DNA

5-Hydroxymethyl-2'-deoxyuridine (hmdUrd) is incorporated into the DNA of V79 Chinese hamster cells as an analogue of thymidine. Incorporated residues are then recognized and excised by hmUra-DNA glycosylase (hmUDG). The removal of large numbers of hmUra residues and subsequent strand breakage is cytotoxic, as has been demonstrated by our finding that a mutant cell line, which is deficient in this enzyme, is resistant to hmdUrd (Boorstein et al., 1992a). In order to determine whether topoisomerase I plays a role in hmUDG initiated base excision repair, V79 cells and repair deficient V79mut1 cells were exposed to combinations of hmdUrd and the topoisomerase I inhibitors camptothecin (CPT), CPT-11, and beta-lapachone. Treatment of V79 cells with hmdUrd followed by non-toxic concentrations of camptothecin or CPT-11 showed significant enhancement of the baseline cytotoxicity of the hmdUrd alone. In contrast, camptothecin and CPT-11 had no effect in combination with hmdUrd in the V79mut1 cells. Non-toxic concentrations of beta-lapachone, which inhibits topoisomerase I by a different mechanism than camptothecin and CPT-11, produced no synergistic toxicity in V79 cells. Neither camptothecin nor CPT-11 inhibited removal of hmdUrd from hmdUrd treated cells, nor did they affect hmdUrd-induced poly(ADP-ribose) synthesis. Camptothecin did not alter the cell cycle distribution of either hmdUrd treated cells or untreated cells at concentrations sufficient to cause synergistic toxicity with hmdUrd. Results from our study indicate that the utility of topoisomerase I inhibitors may be enhanced by sensitizing cells with hmdUrd initiated repair activity which arrests cells in S-phase and produces DNA lesions that are further converted into lethal damage by camptothecin.

Effect of pH and temperature on the stability of UV-induced repairable pyrimidine hydrates in DNA

UV irradiation of cytosine yields 6-hydroxy-5,6-dihydrocytosine (cytosine hydrate) whether the cytosine is in solution as base, nucleoside, or nucleotide or on the DNA backbone. Cytosine hydrate decomposes by elimination of water, yielding cytosine, or by irreversible deamination, yielding uracil hydrate, which, in turn, decomposes by dehydration yielding uracil. To determine how pH and temperature affect these decomposition reactions, alternating poly(dG-[3H]dC) copolymer was irradiated at 254 nm and incubated under different conditions of pH and temperature. The cytosine hydrate and uracil hydrate content of the DNA was determined by the use of Escherichia coli endonuclease III, which releases pyrimidine hydrates from DNA by virtue of its DNA glycosylase activity. Uracil content was determined by using uracil-DNA glycosylase. The rate of decomposition of cytosine hydrate to cytosine was determined at 4 temperatures at pH 3.1, 5.4, and 7.4. The Ea was determined from the rates by using the Arrhenius equation and proved to be the same at pH 5.4 and 7.4, although the decomposition rate at pH 5.4 was faster at all temperatures. At pH 3.1, the Ea was reduced. These results suggest that the dehydration reaction is affected by two discrete protonations, most probably of the N-3 and the OH group of C-6 of cytosine hydrate. The deamination of cytosine hydrate to uracil hydrate was maximal at pH 3.1 at all temperatures. The doubly protonated cytosine hydrate probably is the common intermediate for both competing decomposition reactions, explaining why cytosine hydrate is prone to deamination at acid pH.(ABSTRACT TRUNCATED AT 250 WORDS)

A mammalian cell line deficient in activity of the DNA repair enzyme 5-hydroxymethyluracil-DNA glycosylase is resistant to the toxic effects of the thymidine analog 5-hydroxymethyl-2'-deoxyuridine

We isolated a mutant mammalian cell line lacking activity for the DNA repair enzyme 5-hydroxymethyluracil-DNA glycosylase (HmUra-DNA glycosylase). The mutant was isolated through its resistance to the thymidine analog 5-hydroxymethyl-2'-deoxyuridine (HmdUrd). The mutant incorporates HmdUrd into DNA to the same extent as the parent line but, lacking the repair enzyme, does not remove it. The phenotype of the mutant demonstrates that the toxicity of HmdUrd does not result from substitution of thymine in DNA by HmUra but rather from the removal via base excision of large numbers of HmUra residues in DNA. This finding elucidates a novel mechanism of toxicity for a xenobiotic nucleoside. Furthermore, the isolation of this line supports our hypothesis that the enzymatic repairability of HmUra derives not from its formation opposite adenine via the oxidation of thymine, but rather from its formation opposite guanine as a product of the oxidation and subsequent deamination of 5-methylcytosine.

Synthesis of the diastereomers of thymidine glycol, determination of concentrations and rates of interconversion of their cis-trans epimers at equilibrium and demonstration of differential alkali lability within DNA

5,6-dihydroxy-5,6-dihydrothymidine (thymidine glycol) is a major product of the reaction of thymidine with reactive oxygen species, including those generated by ionizing radiation. Thymidine glycol exists as 2 diastereomeric pairs by virtue of the chirality of the C(5) and C(6) atoms. A simple procedure is described for synthesizing and purifying each of the diastereomeric pairs separately. After brominating thymidine, the two trans 5-bromo-6-hydroxy-5,6-dihydrothymidine (thymidine bromohydrin) C(5) diastereomers were easily separated by High Performance Liquid Chromatography. Each thymidine bromohydrin was quantitatively converted to the corresponding diastereomeric thymidine glycol pair by reflux in aqueous solution. The concentrations at equilibrium of the cis (5S,6R),(5R,6S) and trans (5S,6S),(5R,6R) forms of the thymidine glycol diastereomers were determined and were 80% cis and 20% trans for the 5S pair and 87% cis and 13% trans for the 5R pair. At equilibrium, the rate of cis-trans epimerization of the two sets of diastereomers was essentially identical. The 5S diastereomeric pair was significantly more alkali labile than the 5R pair due to the higher concentration of the 5S trans epimer at equilibrium. This differential alkali lability was also manifest when the thymine glycol moiety was present in chemically oxidized poly(dA-dT).poly(dA-dT) indicating that the chemical differences between the diastereomeric pairs are preserved in DNA. These chemical differences may affect the biological properties of this important oxidative derivative of thymine in DNA.

Formation and stability of repairable pyrimidine photohydrates in DNA

Ultraviolet irradiation of poly(dG-dC) and poly(dA-dU) in solution produces pyrimidine hydrates that are repaired by bacterial and mammalian DNA glycosylases [Boorstein et al. (1989) Biochemistry 28, 6164-6170]. Escherichia coli endonuclease III was used to quantitate the formation and stability of these hydrates in the double-stranded alternating copolymers poly(dG-dC) and poly(dA-dU). When poly(dG-dC) was irradiated with 100 kJ/m2 of 254-nm light at pH 8.0, 2.2% of the cytosine residues were converted to cytosine hydrate (6-hydroxy-5,6-dihydrocytosine) while 0.09% were converted to uracil hydrate (6-hydroxy-5,6-dihydrouracil). To measure the stability of these products, poly(dG-dC) was incubated in solution for up to 24 h after UV irradiation. Cytosine hydrate was stable at 4 degrees C and decayed at 25, 37, and 55 degrees C with half-lives of 75, 25, and 6 h. Uracil hydrate produced in irradiated poly(dA-dU) was stable at 4 degrees C and at 25 degrees C and decayed with a half-life of 6 h at 37 degrees C and less than 0.5 h at 55 degrees C. Uracil hydrate and uracil were also formed in irradiated poly(dG-dC). These experiments demonstrate that UV-induced cytosine hydrate may persist in DNA for prolonged time periods and also undergo deamination to uracil hydrate, which in turn undergoes dehydration to yield uracil. The formation and stability of these photoproducts in DNA may have promoted the evolutionary development of the repair enzyme endonuclease III and analogous DNA glycosylase/endonuclease activities of higher organisms, as well as the development of uracil-DNA glycosylase.

Phylogenetic evidence of a role for 5-hydroxymethyluracil-DNA glycosylase in the maintenance of 5-methylcytosine in DNA

5-Hydroxymethyluracil (HmUra) is formed in DNA as a product of oxidative attack on the methyl group of thymine. It is also the product of the deamination of 5-hydroxymethylcytosine (HmCyt) which may be formed via oxidation of 5-methylcytosine (MeCyt). HmUra is removed from DNA by a DNA glycosylase which, together with HmCyt-DNA glycosylase, is unique among DNA repair enzymes in being present in mammalian cells but absent from bacteria and yeast. We found HmUra-DNA glycosylase activity in a wide variety of vertebrate and invertebrate animals (except Drosophila) and in protozoans. In most vertebrate organisms the highest specific activity was in nervous and immune system tissue. The phylogenetic distribution of HmUra-DNA glycosylase correlates with the presence of 5-methylcytosine (MeCyt) as a regulator of gene expression. This distribution of activity supports the contention that HmUra-DNA glycosylase aids in the maintenance of methylated sites in DNA.

UV-induced pyrimidine hydrates in DNA are repaired by bacterial and mammalian DNA glycosylase activities

Escherichia coli endonuclease III and mammalian repair enzymes cleave UV-irradiated DNA at AP sites formed by the removal of cytosine photoproducts by the DNA glycosylase activity of these enzymes. Poly(dG-[3H]dC) was UV irradiated and incubated with purified endonuclease III. 3H-Containing material was released in a fashion consistent with Michaelis-Menten kinetics. This 3H material was determined to be cytosine by chromatography in two independent systems and microderivatization. 3H-Containing material was not released from nonirradiated copolymer. When poly(dA-[3H]dU) was UV irradiated, endonuclease III released 3H-containing material that coeluted with uracil hydrate (6-hydroxy-5,6-dihydrouracil). Similar results are obtained by using extracts of HeLa cells. There results indicate that the modified cytosine residue recognized by endonuclease III and the mammalian enzyme is cytosine hydrate (6-hydroxy-5,6-dihydrocytosine). Once released from DNA through DNA-glycosylase action, the compound eliminates water, reverting to cytosine. This is consistent with the known instability of cytosine hydrate. The repairability of cytosine hydrate in DNA suggests that it is stable in DNA and potentially genotoxic.

Effects of 5-hydroxymethyluracil and 3-aminobenzamide on the repair and toxicity of 5-hydroxymethyl-2'-deoxyuridine in mammalian cells

5-Hydroxymethyl-2'-deoxyuridine (HmdUrd), a cytotoxic analogue of thymidine, has been proposed for use as an anticancer agent. HmdUrd is incorporated into DNA and then removed at a rate of 30-40% per day. The removal of HmdUrd from DNA has been attributed to the action of 5-hydroxymethyluracil-DNA glycosylase (HmUra-DNA glycosylase). We demonstrated the release of [3H]HmUra into the growth medium of V79 Chinese hamster cells that had incorporated [3H]HmdUrd into their DNA. The amount of [3H]HmUra recovered from the growth medium was equal to the amount of [3H]HmdUrd lost from DNA. These experiments confirmed that the initial step of repair of HmUra in DNA was mediated by DNA glycosylase activity. A combination of HmUra and HmdUrd resulted in increased uptake of HmdUrd by cells and increased cytotoxicity. The increased incorporation of HmdUrd into DNA was not due to inhibition of repair. 3-Aminobenzamide, an inhibitor of poly(ADP-ribose) synthesis, was cytotoxic to cells which incorporated and repaired HmdUrd. The extent of toxicity was directly related to the number of HmUra residues in DNA. HeLa cells, known to be resistant to the toxic effects of HmdUrd, do not incorporate HmdUrd into their DNA. HeLa cells were resistant to the toxic effects of 3-aminobenzamide, confirming that the absence of HmdUrd in their DNA was not due to an accelerated rate of repair. These experiments indicate that the potential therapeutic antineoplastic properties of HmdUrd may be enhanced by using HmUra to increase the incorporation of HmdUrd into DNA and 3-aminobenzamide to interfere with repair of HmUra in DNA.

Mutagenicity of 5-hydroxymethyl-2'-deoxyuridine to Chinese hamster cells

5-Hydroxymethyluracil (HmUra) is formed from thymine in DNA through the action of ionizing radiation or reactive oxygen species generated by activated leukocytes. HmUra is removed from DNA by a specific DNA glycosylase, suggesting that it is also formed from endogenously generated reactive oxygen species and that its formation in DNA is potentially deleterious. To determine whether HmUra residues in DNA are mutagenic, hamster V79 cells were grown in the presence of 5-hydroxymethyl-2'-deoxyuridine (HmdUrd) which is incorporated into DNA, and mutagenicity at the ouabain- and thioguanine-resistant loci was determined. Levels of substitution ranged from 1/500 to 1/5,000 HmUra residues/thymine residues. There was slight mutagenicity at the thioguanine-resistant locus but none at the ouabain-resistant locus. The mutagenicity of HmdUrd, expressed as a function of HmUra substitution in DNA, was 1/30,000 in the hypoxanthine-guanine-phosphoribosyltransferase target gene. This low frequency indicates that the oxidation of thymine to HmUra in a preexisting AT base pair does not contribute significantly to the mutagenicity of ionizing radiation, because the yield of HmUra formed in DNA at mutagenic doses of radiation is too low. To determine whether repair of HmUra might be inhibited by ionizing radiation, cells were grown in medium containing HmdUrd and exposed to as much as 5 Gy of gamma-irradiation, and the removal of HmUra from DNA was measured. No inhibition of repair was noted. Preirradiation of cells neither accelerated the rate of repair nor raised the level of HmUra-DNA glycosylase activity, indicating that repair of HmUra was not induced by this type of oxidative stress. Although the mutagenicity of HmUra residues in DNA is low, even a rare mutation might be sufficiently deleterious to higher organisms to promote the development of HmUra-DNA glycosylase activity.

The repairability of oxidative free radical mediated damage to DNA: a review

Many DNA repair enzyme activities are present in both prokaryotic and eukaryotic organisms. Among these are DNA exo- and endonucleases and DNA glycosylases which remove oxidatively damaged portions of the DNA molecule, thereby initiating excision-repair. The existence of these enzymes may be taken as evidence that cellular DNA is continuously subject to endogenous oxidative stress. Many of the lesions introduced by ionizing and ultraviolet radiation are identical to those introduced into DNA by reactive oxygen species generated by activated white cells, and are substrates for the repair enzymes. The chemical nature of the lesions, their biologic effects, and the mechanism of their repairability are described.

Toxicity of 3-aminobenzamide to Chinese hamster cells containing 5-hydroxymethyluracil in their DNA

V79 cells incorporated 5-hydroxymethyl-2'-deoxyuridine (HmdUrd) into their DNA linearly over a wide range of concentrations and time. Cells grew normally when 0.03% of thymidine residues were replaced with HmdUrd. At this level of substitution, 5-hydroxymethyluracil (HmUra) was removed from DNA at a rate of 30-40%/24 h. Concentrations of HmdUrd in the growth medium which produced higher levels of substitution reduced survival and caused cells to delay their transit through S phase. However, the treatment of HmdUrd-containing cells with 3-aminobenzamide caused extensive cell death. At levels of HmdUrd substitution compatible with near 90% survival, the addition of 3-aminobenzamide, an inhibitor of poly (adenosine diphosphoribose) synthesis, killed over 90% of the cells. This toxicity was not due to inhibition of the removal of HmUra from DNA. Cells killed by this combination of agents arrested in the G2 phase of the cell cycle. We conclude that the toxicity of HmdUrd resulted primarily from the repair of the HmUra residue in DNA and not from any intrinsic toxicity of the HmUra residue itself. We also conclude that the cytotoxicity of 3-aminobenzamide resulted from interference with the completion of DNA repair following base (HmUra) excision. Since HmUra is also formed in DNA through the action of ionizing radiation, it may be among the components of radiation-induced DNA damage which sensitizes cells to 3-aminobenzamide.

5-Hydroxymethyluracil-DNA glycosylase activity may be a differentiated mammalian function

To determine the prevalence of the repair enzyme HMU-DNA glycosylase we assayed its activity in whole cell extracts of several bacterial species, the eukaryotic yeast Saccharomyces cerevisiae, mammalian cell lines and murine tissue. Enzyme activity was constitutively present in murine, hamster and human cell lines. It was not inducible by exposing cells to oxidative stress from ionizing radiation or by incubating cells with the 2'-deoxynucleoside of HMU, HMdU. In murine tissue, enzyme activity was highest in brain and thymus. HMU-DNA glycosylase activity was not detectable in bacteria or yeast nor could activity be detected after exposure of cells to H2O2. These results suggest that, in contrast to other DNA-repair enzymes, HMU-DNA glycosylase is a differentiated function limited to higher eukaryotic organisms.

Factors modifying 3-aminobenzamide cytotoxicity in normal and repair-deficient human fibroblasts

3-Aminobenzamide (3-AB), an inhibitor of poly(ADP-ribosylation), is lethal to human fibroblasts with damaged DNA. Its cytotoxicity was determined relative to a number of factors including the types of lesions, the kinetics of repair, and the availability of alternative repair systems. A variety of alkylating agents, UV or gamma irradiation, or antimetabolites were used to create DNA lesions. 3-AB enhanced lethality with monofunctional alkylating agents only. Within this class of compounds, methylmethanesulfonate (MMS) treatments made cells more sensitive to 3-AB than did treatment with methylnitrosourea (MNU) or methylnitronitrosoguanidine (MNNG). 3-AB interfered with a dynamic repair process lasting several days, since human fibroblasts remained sensitive to 3-AB for 36-48 hours following MMS treatment. During this same interval, 3-AB caused these cells to arrest in G2 phase. Alkaline elution analysis also revealed that this slow repair was delayed further by 3-AB. Human mutant cells defective in DNA repair differed in their responses to 3-AB. Among mutants sensitive to monofunctional alkylating agents, ataxia telangiectasia cells were slightly more sensitive to 3-AB than control cells, while Huntington's disease cells had a near-normal response. Among UV-sensitive strains, xeroderma pigmentosum variant (XPV) cells were more sensitive to 3-AB after MMS than were XP complementation group A (A) cells, which responded normally. Greater lethality with 3-AB could be dependent on inability of the mutant cells to repair damage by other processes.

3-Aminobenzamide is lethal to MMS-damaged human fibroblasts primarily during S phase

3-Aminobenzamide (3-AB) interferes with DNA repair and enhances lethality in growing MMS (methyl methane sulfonate)-treated human fibroblasts. This sensitivity to 3-AB disappears slowly; MMS-treated cells are sensitive to 3-AB for up to 36 hours (Boorstein and Pardee, 1984). Evidence is now presented that 3-AB potentiates the effects of MMS primarily during S phase. When cells were synchronized at the G1/S boundary, released, and then treated with MMS, 3-AB caused very substantial lethality in only 4 hours, and a 12-hour treatment gave maximum lethality. These cells also lost sensitivity to 3-AB within 12 hours of growth minus 3-AB. In contrast, MMS-treated quiescent (G0) cells did not lose sensitivity to 3-AB nor did 3-AB cause lethality during G0. Enhanced lethality occurred when damaged G0-arrested cells were subsequently allowed to proceed through S phase in the presence of 3-AB; this 3-AB sensitivity was removed only during growth in the absence of 3-AB. The lethality of 3-AB to a population of asynchronously cycling cells treated with MMS is thus the summation of effects on the cells as they traverse S phase. Aphidicolin prevented lethality of 3-AB to cells released from G1/S and treated with MMS. It also inhibited the loss of sensitivity to added 3-AB later. Correlation with the inhibition of DNA synthesis by this drug suggests that DNA synthesis is essential for the lethality enhancement by 3-AB. Cells treated first with MMS and then with 3-AB accumulated in G2. This G2 arrest depended on S-phase events and correlated with cell lethality. Cells treated with a nonlethal dose of MMS at the G1/S boundary were delayed briefly (3 hours) in their passage through S and G2. These cells, when also exposed to 3-AB, were delayed 6-9 hours in S and they became arrested in G2. There was no G2 arrest when 3-AB was added only after these cells had reached G2. Treatment with 3-AB during S phase thus resulted in both enhanced lethality and G2 arrest. 3-AB inhibited repair of DNA single-strand damage, shown by alkaline elution analysis, in both S-phase and quiescent cells. Aphidicolin inhibited disappearance of breaks and eliminated the difference between 3-AB-treated and untreated cells. Lethality did not correlate well with the measured single-strand damage.(ABSTRACT TRUNCATED AT 400 WORDS)

Beta-lapachone greatly enhances MMS lethality to human fibroblasts

beta-Lapachone is a naturally occurring tricyclic O-naphthoquinone. At microM concentrations it did not substantially affect viability, growth or DNA synthesis of cultured undamaged human fibroblasts. Cells exposed to minimally toxic concentrations of methyl methane sulfonate were strongly inhibited in these properties by beta-lapachone. The effects were not reversed by further incubation in the absence of beta-lapachone and were equal for initially quiescent or growing cells. Thus inhibitions were specific for damaged cells and did not involve replicative DNA synthesis. Inhibition of DNA strand break repair was demonstrated by alkaline elution, but unscheduled DNA synthesis was not inhibited. We propose that beta-lapachone inhibits a ligation step of DNA repair, in a manner perhaps similar to that reported for carbamoylating nitrosoureas. Other repair inhibitors differ significantly from beta-lapachone in their modes of action.

Coordinate inhibition of DNA synthesis and thymidylate synthase activity following DNA damage and repair

Two agents, 3-aminobenzamide (3-AB) and beta lapachone, that inhibit repair of mammalian cell DNA damaged by methyl methane sulfonate (MMS), also coordinately blocked both DNA replication (incorporation of 3H-thymidine) and thymidylate synthase (TS) activity. Aphidicolin also inhibited both 3H-TDR incorporation and TS in damaged cells, the former more strongly than the latter, in a manner not coordinated with lethality. It is proposed that the DNA lesions created by MMS and modified by repair inhibit semiconservative DNA synthesis by allosterically interacting with the DNA replication replitase complex, so as to block its overall function and also the activity of TS, one of its enzymes.