A novel approach to DNA damage assessments: measurement of the thymine glycol lesion

A different approach to the measurement of DNA damage has been developed based on the fact that many lesions can be excised from DNA in the form of modified dinucleoside monophosphates. Liquid chromatography-tandem mass spectrometry (LC-MS/MS) is used in conjunction with isotopically labeled internal standards to quantify the lesion. The method has several advantages, including high sensitivity for the detection of dinucleoside monophosphates. The method was applied to the measurement of the 5,6-dihydroxy-5,6-dihydrothymine (thymine glycol) lesion in the DNA of mouse fibroblast cells exposed in culture to various treatments including ionizing radiation, UVC light and buthionine sulfoximine (BSO), an inhibitor of glutathione synthesis. The application of the method to the measurement of other DNA lesions is discussed.     

Singlet oxygen-induced DNA damage

Singlet oxygen, hydrogen peroxide, hydroxyl radical and hydrogen peroxide are the reactive oxygen species (ROS) considered most responsible for producing oxidative stress in cells and organisms. Singlet oxygen interacts preferentially with guanine to produce 8-oxo-7,8-dihydroguanine and spiroiminodihydantoin. DNA damage due to the latter lesion has not been detected directly in the DNA of cells exposed to singlet oxygen. In this study, the singlet oxygen-induced lesion was isolated from a short synthetic oligomer after exposure to UVA radiation in the presence of methylene blue. The lesion could be enzymatically excised from the oligomer in the form of a modified dinucleoside monophosphate. Using liquid chromatography-tandem mass spectrometry (LC-MS/MS), the singlet oxygen lesion was detected in the form of modified dinucleoside monophosphates in double-stranded DNA and in the DNA of HeLa cells exposed to singlet oxygen. Pentamer containing the singlet oxygen-induced lesion and an isotopic label was synthesized as an internal standard for quantifying the lesion and served as well as for correcting for losses of product during sample preparation.     

Assessment of DNA damage at the dimer level: measurement of the formamide lesion

UVC-radiation-induced DNA damage was measured in mouse fibroblast cells using liquid chromatography-tandem mass spectrometry (LC-MS/MS) in conjunction with isotopically labeled internal standards. The thymine glycol and formamide lesions were assayed in the form of modified dinucleoside monophosphates. The 8-oxo-7,8-dihydroguanine lesion was measured as the modified nucleoside. DNA damage in cells treated with tirapazamine was also measured. Tirapazamine is a chemotherapeutic agent that acts via a free radical mechanism. The two agents, UVC radiation and tirapazamine, produce markedly different profiles of DNA damage, reflecting their respective mechanisms of action. Both agents produce significant amounts of thymine glycol and formamide damage, but only the former produced a measurable amount of the 8-oxo-7,8-dihydroguanine lesion. The merits of measuring DNA damage at the dimer level are discussed.     

Detection and characterization of formamido lesions in DNA by liquid chromatography-mass spectrometry

DNA X-irradiated in oxygenated aqueous solution produces the formamido lesion from the breakdown of pyrimidine nucleosides. This pyrimidine breakdown product inhibits the hydrolysis by nuclease P1 of the phosphoester bond 3' to the damaged nucleoside. Consequently, the lesion can be obtained from an enzymatic digest of the DNA as a modified dinucleoside monophosphate in which the 5' nucleoside contains the lesion. In this form, the formamido lesion can be detected with good sensitivity by liquid chromatography-mass spectrometry (LC-MS). Nucleosides that have lost the base moiety also inhibit nuclease P1. Together, the formamido and abasic lesions account for all of the substantial peaks in the LC-MS ion current profile.     

32P-Postlabeling Assay for Free Radical-Induced Dna Damage: The Formamido Remnant of Thymine

A prominent lesion in DNA exposed to oxidative free radicals results from the degradation of thymine leaving a formamido remnant. A ³²P-postlabeling assay has been developed for the detection of the formamido lesion. The assay is based on the circumstance that the lesion prevents hydrolysis by nuclease PI of the phosphoester bond 3' to the damaged nucleoside. Thus, a nuclease PI plus acid phosphatase digest of DNA generates mostly nucleosides whereas the formamido lesion is rendered as a modified dinucleoside monophosphate. Dinucleoside monophosphates, but not nucleosides, are apt substrates for ³²P-postlabeling by polynucleotide kinase. The assay was applied to calf thymus DNA X-irradiated in oxygenated solution. The formamido lesion could be detected down to a dose of a few Gy.     

Sequence-dependent formation of intrastrand crosslink products from the UVB irradiation of duplex DNA containing a 5-bromo-2'-deoxyuridine or 5-bromo-2'-deoxycytidine

The replacement of thymidine with 5-bromo-2′-deoxyuridine (^BrdU) is well-known to sensitize cells to ionizing radiation and photoirradiation. We reported here the sequence-dependent formation of intrastrand crosslink products from the UVB irradiation of duplex oligodeoxynucleotides harboring a ^BrdU or its closely related 5-bromo-2′-deoxycytidine (^BrdC). Our results showed that two types of crosslink products could be induced from d(^BrCG), d(^BrUG), d(G^BrU), or d(A^BrU); the C(5) of cytosine or uracil could be covalently bonded to the N(2) or C(8) of its neighboring guanine, and the C(5) of uracil could couple with the C(2) or C(8) of its neighboring adenine. By using those crosslink product-bearing dinucleoside monophosphates as standards, we demonstrated, by using liquid chromatography-mass spectrometry/mass spectrometry (LC-MS/MS), that all the crosslink products described above except d(G[N(2)-5]U) and d(G[N(2)-5]C) could form in duplex DNA. In addition, LC-MS/MS quantification results revealed that both the nature of the halogenated pyrimidine base and its 5′ flanking nucleoside affected markedly the generation of intrastrand crosslink products. The yields of crosslink products were much higher while the 5′ neighboring nucleoside was a dG than while it was a dA, and ^BrdC induced the formation of crosslink products much more efficiently than ^BrdU. The formation of intrastrand crosslink products from these halopyrimidines in duplex DNA may account for the photosensitizing effects of these nucleosides.     

Analysis of DNA damage at the dinucleoside monophosphate level: application to the formamido lesion.

The dinucleoside monophosphates d(TpG), d(TpC), and d(TpT) were X-irradiated in oxygenated solution. In each case the modification of the dinucleoside in which the thymine base is degraded to a formamido remnant was observed as a principal product. The hydrolysis of the phosphoester bond of formamido-modified dinucleosides is much slower than that of the corresponding unmodified dinucleosides. This effect is also observable in the hydrolysis of irradiated DNA, where hydrolysis by nuclease P1 (plus acid phosphatase) generates the modified dinucleosides d(TFpN), TF being the modified thymidine. The total yield of the formamido lesion in all its forms, d(TFpN), exceeds the yield of any other base modification.     

Circular dichroism study of 2'-5' dinucleoside monophosphates modified with N-2-acetylaminofluorene.

The conformational properties of four 2′ – 5′ dinucleoside monophosphates modified with $\text{N}$-2-acetylaminofluorene have been studied by circular dichroism spectroscopy. Covalent binding of this chemical carcinogen at the C₈ position of guanosine in the 2′ – 5′ dinucleoside monophosphates induces striking changes in their circular dichroic spectra depending on their base sequence and composition. The changes in CD spectra, redshift of the extrema and change of their polarity, not observed in the spectra of corresponding 3′ – 5′ derivatives modified with $\text{N}$-2-acetylaminofluorene are correlated with the difference in the configuration of 2′ – 5′ and 3′ – 5′ dinucleoside monophosphates and discussed in respect to the intramolecular stacking interactions.     

Dinucleoside monophosphates. I. Optical properties and conformation in solution with one base charged.

A least squares analysis of the titration properties of several dinucleoside monophosphates enables calculation of the pK's for protonation. These pK's are used to resolve the spectral properties of dinucleoside monophosphates with one base charged from the apparent spectral properties of a dinucleoside monophosphate in aqueous solution. This method is applied to dinucleoside monophosphates containing adenosine and/or cytidine. Results of CD, nmr, and CD-temperature dependence measurements are presented. The results indicate that singly protonated dimers of these nucleosides stack as do their unprotonated analogs. It is suggested that this is true for all dimers with one base charged.     

Preparation and enzymatic hydrolysis of dinucleoside monophosphates and DNA modified with aromatic residues

The following procedures have been used to prepare fifteen modified dinucleoside monophosphates: (a) bisulfite-catalyzed transamination with aniline to give an N4-phenylcytidine (CPh), (b) bisulfite-catalyzed transamination with beta-naphthylamine to give an N4-beta-naphthylcytidine (CbetaN), (c) alkylation with 7-bromomethylbenz[a] anthracene to afford a 7(benz[a]anthryl-7-methyl)guanosine (GMBA), and (d) reaction with N-acetoxy-2-acetylaminofluorene to give an 8-(N-2-fluorenylacetamido)guanosine (GAAF). The compounds prepared were A-CPh, CPh-A, CPh-G, U-CPh, CPh-U, A-CbetaN, CbetaN-A, G-CbetaN, CbetaN-G, U-CbetaN, CbetaN-U, GMBA-U, U-GMBA, GAAF-U, and U-GAAF. All of the modified compounds were hydrolyzed to the expected monomers with venom and spleen exonucleases. Hydrolysis by micrococcal nuclease was inhibited in the following cases: A-CPh, A-CbetaN, U-GMBA, and U-GAAF. The first three reactions above were applied to denatured calf thymus DNA to prepare modified DNA samples containing from 0.3 to 2.0% bound aromatic residues. The modified nucleic acids were completely hydrolyzed to nucleosides by the combination of venom exonuclease, deoxyribonuclease I and alkaline phosphatase. The same results were obtained with a combination of spleen exonuclease, deoxyribonuclease II, and alkaline phosphatase. Hydrolysis of the modified nucleic acids by micrococcal nuclease and alkaline phosphatase afforded primarily nucleosides, with some dinucleoside monophosphates. The amount of the latter did not exceed that found in the hydrolysis of control DNA, however. Other workers have observed inhibition of enzymatic hydrolysis of nucleic acids modified by aromatic carcinogens. We postulated that their results may have been caused by cross-links, which were avoided in our studies.     

Conformation of acetylaminofluorene and aminofluorene modified guanosine and guanosine derivatives

Acetylaminofluorene and aminofluorene modified Guo, GMP, d(GpA) and d(ApG) have been studied by circular dichroism and ¹H nuclear magnetic resonance. Aminofluorene modified Guo is preferentially in the $\text{anti}$ conformation and acetylaminofluorene modified Guo in the $\text{syn}$ conformation. It is proposed that the $\text{anti}$ conformation of aminofluorene modified Guo is stabilized by an intra molecular hydrogen bond between the NH group of aminofluorene residue and the 5′-OH group of the sugar. The results on the modified dinucleoside monophosphates are analyzed according to this hypothesis.     

32P-postlabeling of acrolein-deoxyguanosine adducts in DNA after nuclease P1 digestion.

In order to study the relationship between the level of acrolein-DNA adducts and their biological effects, sensitive methods are needed to quantitate DNA adducts. 32P-postlabeling is one such method that has been widely used and we have adapted the technique to detect acrolein-deoxyguanosine adducts. Adducts formed by the reaction of acrolein and deoxyguanosine-3'-monophosphate were isolated by HPLC. Based on their UV spectra and cochromatography with standards after dephosphorylation with acid phosphatase, these adducts were identified as the nucleotide equivalents of cyclic 1,N2-propanodeoxyguanosine adducts formed by acrolein that have been described by Chung et al. [15]. As nucleotides, the adducts were good substrates for polynucleotide kinase-mediated transfer of phosphate from ATP and were able to be detected by 32P-postlabeling. These adducts were resistant to the activity of nuclease P1 and dinucleoside monophosphates in the form d(G*pN) where G* is the acrolein-guanine adduct also resisted digestion by nuclease P1. Digestion of DNA by nuclease P1 and acid phosphatase resulted in the conversion of normal nucleotides to nucleosides and selective enrichment of the adducts as dinucleoside monophosphates. Using nuclease P1/acid phosphatase digestion, followed by 32P-postlabeling and TLC separation, levels of the two adducts in acrolein-treated DNA were found to be about 6185 and 19,222 nmol/mol.     

Selective hydrolysis of damaged DNA by nuclease P1

The turnover rates for hydrolysis by nuclease P1 of the 16 unmodified dideoxynucleoside monophosphates were measured. In addition, the turnover rates were measured in a variety of dideoxynucleoside monophosphates containing free radical-induced base modifications. The modified bases included cis-5,6-dihydroxy-5,6-dihydrothymine (thymine glycol), 5,6-dihydrothymine, 5-hydroxymethyuracil, 8-hydroxyguanine, 5-hydroxy-5-methylhydantoin and the formamido remnant which can be derived from either a thymine or a cytosine base. The turnover rate for dinucleoside monophosphates containing 4,8-dihydro-4-hydroxy-8-oxo-guanine modifications, which are induced by singlet oxygen, were also measured. A model was devised for the hydrolysis of DNA by nuclease P1 which uses the observed turnover rates as parameters. The model predicts the abundance of monomers and dimers as hydrolysis proceeds. Whereas the level of monomers increases monotonically, the level of each dimer first increases and then falls off. There are advantages to phosphorylating dimers, as compared with monomers, using polynucleotide kinase. Consequently this model may be of interest in connection with ³²P-postlabeling applied to the measurement of DNA damage in nuclease P1 partial hydrolysates of DNA.     

Ultraviolet radiation-mediated damage to cellular DNA

Emphasis is placed in this review article on recent aspects of the photochemistry of cellular DNA in which both the UVB and UVA components of solar radiation are implicated individually or synergistically. Interestingly, further mechanistic insights into the UV-induced formation of DNA photoproducts were gained from the application of new accurate and sensitive chromatographic and enzymic assays aimed at measuring base damage. Thus, each of the twelve possible dimeric photoproducts that are produced at the four main bipyrimidine sites can now be singled out as dinucleoside monophosphates that are enzymatically released from UV-irradiated DNA. This was achieved using a recently developed high-performance liquid chromatography-tandem mass spectrometry assay (HPLC-MS/MS) assay after DNA extraction and appropriate enzymic digestion. Interestingly, a similar photoproduct distribution pattern is observed in both isolated and cellular DNA upon exposure to low doses of either UVC or UVB radiation. This applies more specifically to the DNA of rodent and human cells, the cis-syn cyclobutadithymine being predominant over the two other main photolesions, namely thymine-cytosine pyrimidine (6-4) pyrimidone adduct and the related cyclobutyl dimer. UVA-irradiation was found to generate cyclobutane dimers at TT and to a lower extent at TC sites as a likely result of energy transfer mechanism involving still unknown photoexcited chromophore(s). Oxidative damage to DNA is also induced although less efficiently by UVA-mediated photosensitization processes that mostly involved 1O2 together with a smaller contribution of hydroxyl radical-mediated reactions through initially generated superoxide radicals.     

Photooxidation of d(TpG) by riboflavin and methylene blue. Isolation and characterization of thymidylyl-(3',5')-2-amino-5-[(2-deoxy-beta-D- erythro-pentofuranosyl)amino]-4H-imidazol-4-one and its primary decomposition product thymidylyl-(3',5')-2,2-diamino-4-[(2-deoxy-beta-D- erythro-pentofuranosyl)amino]-5(2H)-oxazolone.

The major initial product of riboflavin- and methylene blue-mediated photosensitization of 2'-deoxyguanosine (dG) in oxygen-saturated aqueous solution has previously been identified as 2-amino-5-[(2-deoxy-beta-D-erythro-pentofuranosyl)amino] 4H-imidazol-4-one (dlz). At room temperature in aqueous solution dlz decomposes quantitatively to 2,2-diamino-4-[(2-deoxy-beta-D-erythro- pentofuranosyl)amino]-5(2H)-oxazolone (dZ). The data presented here show that the same guanine photooxidation products are generated following riboflavin- and methylene blue-mediated photosensitization of thymidylyl-(3',5')-2'-deoxyguanosine [d(TpG)]. As observed for the monomers, the initial product, thymidylyl-(3',5')-2-amino-5-[(2-deoxy- beta-D-erythro-pentofuranosyl)amino]-4H-imidazol-4-one [d(Tplz)], decomposes in aqueous solution at room temperature to thymidylyl-(3',5')-2,2-diamino-4- [(2-deoxy-beta-D-erythro-pentofuranosyl)amino]-5(2H)-oxazolone [d(TpZ)]. Both modified dinucleoside monophosphates have been isolated by HPLC and characterized by proton NMR spectrometry, fast atom bombardment mass spectrometry, chemical analyses and enzymatic digestions. Among the chemical and enzymatic properties of these modified dinucleoside monophosphates are: (i) d(Tplz) and d(TpZ) are alkali-labile; (ii) d(Tplz) reacts with methoxyamine, while d(TpZ) is unreactive; (iii) d(Tplz) is digested by snake venom phosphodiesterase, while d(TpZ) is unaffected; (iv) relative to d(TpG), d(TpZ) and d(Tplz) are slowly digested by spleen phosphodiesterase; (v) d(Tplz) and d(TpZ) can be 5'-phosphorylated by T4 polynucleotide kinase. The first observation suggests that dlz and dZ may be responsible for some of the strand breaks detected following hot piperidine treatment of DNA exposed to photosensitizers.     

Photooxidation of d(TpG) by phthalocyanines and riboflavin. Isolation and characterization of dinucleoside monophosphates containing the 4R* and 4S* diastereoisomers of 4,8-dihydro-4-hydroxy-8-oxo-2''-deoxy-guanosine.

Phthalocyanine mediated photosensitization of 2'-deoxyguanosine (dG) in oxygen saturated aqueous solution has previously been shown to result in the addition of molecular oxygen to the guanine base generating the 4R* and 4S* diastereoisomers of 4,8-dihydro-4-hydroxy-8-oxo-2'-deoxyguanosine (dO) (the asterisk denotes unambiguous assignment of the 4R and 4S diastereoisomers). The data presented here show that the same guanine modified bases are generated in a 1:1 ratio when thymidylyl-(3',5')-2'-deoxyguanosine (d(TpG)) is similarly photo-oxidized. These modified dinucleoside monophosphates, labelled d(TpO)-A and -B, have been isolated by high performance liquid chromatography and characterized by proton NMR spectrometry, fast atom bombardment mass spectrometry, and enzymatic digestions. Photosensitization in D2O instead of H2O leads to an increase in the rate of d(TpO) formation that is consistent with a type II (singlet oxygen) reaction mechanism. Three interesting properties of these modified dinucleoside monophosphates are: i) the rate of their digestion with spleen phosphodiesterase is greatly reduced relative to d(TpG), ii) they are not digested by snake venom phosphodiesterase, and iii) they are stable to 1.0 M piperidine at 90 degrees C for 30 min. The latter observation indicates that 4,8-dihydro-4-hydroxy-8-oxoguanine is not a base lesion responsible for the strand breaks observed following hot piperidine treatment of DNA exposed to type II photosensitizers or chemically generated singlet oxygen.     

X-ray crystallographic analysis of a ternary intercalation complex between proflavine and the dinucleoside monophosphates CpA and UpG

We report the crystal-structure analysis of a complex involving the drug proflavine and the two dinucleoside monophosphates cytidylyl-3′,5′-adenosine (CpA) and uridylyl-3′,5′-guanosine (UpG). The planar drug molecule is intercalated between C ?G and U ?A Watson-Crick base pairs, in a double-helical fragmentlike arrangement. Sugar conformations at the 3′-ends of the two strands are dissimilar. The backbone conformations fall within the ranges of values noted previously for dinucleoside intercalation complexes, and some correlations involving these are noted. The separation of the two strands and the basic twist angle of 16°, compared to other reported complexes, are indicative of sequence-dependent effects of the drug binding.     

Zn2+ Promoted Hydrolysis of RNA

Abstract

The Zn²⁺ promoted hydrolysis of phosphodiester bonds of RNA has been studied by using a series of model compounds from dinucleoside monophosphates to oligo- and polynucleotides. The results will be discussed with respect to complex formation between the metal ion catalysts and substrates.     

Synthesis and properties of oligodeoxyribonucleotides containing an ethylated internucleotide phosphate

Internucleotide phosphotriesters comprise an important class of DNA lesions produced by carcinogenic alkylating agents. To avoid confusion resulting from the presence of other DNA lesions, synthetically prepared oligonucleotides containing ethylated internucleotide phosphates as the sole form of damage were employed to investigate several chemical and biochemical properties of DNA alkyl phosphotriesters. A total of four oligonucleotides were synthesised for this study, the dimers Tp(Et)T and pTp(Et)T and the decamer d-TpTpTp(Et)TpCpTpApTpTpT together with its unmodified analogue. The dimers were characterized by UV and phosphorus NMR spectroscopy and the decamers by two-dimensional homochromatography, alkali hydrolysis, and variable-temperature circular dichroism (CD). Alkali hydrolysis of the ethylated decamer produced strand breaks in approximately 75% of the molecules. This is in close agreement with data previously obtained for dinucleoside ethyl phosphotriesters and triesters in alkylated cellular DNA. Results from the CD study suggest that the ethyl substituent does not disrupt base stacking within the oligomer. The interactions of two enzymes with the alkylated oligonucleotides were examined. First, it was found that ethylation of the internucleotide phosphate renders TpT inactive as a substrate for T4 polynucleotide kinase, implying that a negative charge is required on the 3'-phosphate group of the nucleotide to be phosphorylated. Hence, postlabeling assays of DNA damage that depend upon enzymatic phosphorylation of modified 3'-nucleotides cannot be applied to dinucleoside alkyl phosphotriesters. Second, both decamers, when annealed to a single-stranded plasmid template, were able to prime DNA synthesis, catalyzed by Escherichia coli DNA polymerase I, with equal effectiveness. The use of this reaction as a means of site-specifically incorporating phosphotriesters into viral vectors is recognized.     

Quantification of cidofovir in human serum by LC-MS/MS for children

A new method for the quantification of cidofovir (CDV), an acyclic nucleotide analogue of cytosine with antiviral activity against a broad-spectrum of DNA viruses, in human serum, using high-performance liquid chromatography coupled to tandem mass spectrometry (LC-MS/MS) has been developed. A strong anion exchange (SAX) solid-phase extraction procedure was applied for the sample preparation. The tandem mass spectrometer was tuned in the multiple reaction monitoring mode to monitor the m/z 278.1-->234.9 and the m/z 288.1-->133.1 transitions for CDV and the internal standard 9-(2-phosphonylmethoxyethyl)guanine (PMEG), respectively, using negative electrospray ionization. The MS/MS response was linear over the concentration range from 78.125 ng/ml to 10,000 ng/ml, with a lower limit of quantification of 78.125 ng/ml. The intra- and inter-day precisions (relative standard deviation (%)) for CDV were less than 7.8% and the accuracies (% of deviation from nominal level) were within +/-12.1% for quality controls. The novel LC-MS/MS method allowed a specific, sensitive and reliable determination of CDV in human serum and was applied to investigate the yet unknown pharmacokinetic properties of CDV in a paediatric cancer patient.