Sources of artifact in measurements of 6mA and 4mC abundance in eukaryotic genomic DNA

BACKGROUND

Directed DNA methylation on N6-adenine (6mA), N4-cytosine (4mC), and C5-cytosine (5mC) can potentially increase DNA coding capacity and regulate a variety of biological functions. These modifications are relatively abundant in bacteria, occurring in about a percent of all bases of most bacteria. Until recently, 5mC and its oxidized derivatives were thought to be the only directed DNA methylation events in metazoa. New and more sensitive detection techniques (ultra-high performance liquid chromatography coupled with mass spectrometry (UHPLC-ms/ms) and single molecule real-time sequencing (SMRTseq)) have suggested that 6mA and 4mC modifications could be present in a variety of metazoa.

RESULTS

Here, we find that both of these techniques are prone to inaccuracies, which overestimate DNA methylation concentrations in metazoan genomic DNA. Artifacts can arise from methylated bacterial DNA contamination of enzyme preparations used to digest DNA and contaminating bacterial DNA in eukaryotic DNA preparations. Moreover, DNA sonication introduces a novel modified base from 5mC that has a retention time near 4mC that can be confused with 4mC. Our analyses also suggest that SMRTseq systematically overestimates 4mC in prokaryotic and eukaryotic DNA and 6mA in DNA samples in which it is rare. Using UHPLC-ms/ms designed to minimize and subtract artifacts, we find low to undetectable levels of 4mC and 6mA in genomes of representative worms, insects, amphibians, birds, rodents and primates under normal growth conditions. We also find that mammalian cells incorporate exogenous methylated nucleosides into their genome, suggesting that a portion of 6mA modifications could derive from incorporation of nucleosides from bacteria in food or microbiota. However, gDNA samples from gnotobiotic mouse tissues found rare (0.9-3.7 ppm) 6mA modifications above background.

CONCLUSIONS

Altogether these data demonstrate that 6mA and 4mC are rarer in metazoa than previously reported, and highlight the importance of careful sample preparation and measurement, and need for more accurate sequencing techniques.

A conserved AU-rich element in the 3' untranslated region of bcl-2 mRNA is endowed with a destabilizing function that is involved in bcl-2 down-regulation during apoptosis

The control of mRNA stability is becoming recognized as a crucial point of gene expression regulation. A common element responsible for mRNA decay modulation is the adenine- and uracil-rich element that is found in the 3' untranslated region of numerous mRNAs subjected to fast expression changes in response to various stimuli. Previously we identified a post-transcriptional regulation level for the antiapoptotic bcl-2 gene, which could be involved in t(14;18) lymphoma-associated bcl-2 overexpression. Here we demonstrate that bcl-2 mRNA is endowed with an adenine- and uracil-rich element (ARE) characterized by high evolutionary conservation not only among all chordates examined, but even between chordates and the nematode Caenorhabditis elegans (ced-9 gene). As for other well-established destabilizing AREs, the insertion of the bcl-2 ARE downstream from stable beta-globin mRNA causes an enhanced decay of the beta-globin transcript, which proves its functional role. This possibility is corroborated by the fact that the pathway leading to the modulating activity of bcl-2 ARE is influenced by PKC, since the addition of DAG and TPA markedly attenuated the bcl-2 ARE destabilizing potential. Conversely, it is noteworthy that when C(2)-ceramide is added to the culture medium as the apoptotic agent, the beta-globin transcript harboring the bcl-2 ARE undergoes a dramatic increase in decay. This observation clearly indicates that the destabilizing function of bcl-2 ARE is enhanced by apoptotic stimuli and suggests that this element could be involved in a post-transcriptional mechanism of bcl-2 down-regulation during apoptosis. The half-life of the mRNA of bcl-2 in Jurkat cells is prolonged by PKC stimulation and shortened by C(2)-ceramide addition, strongly supporting the view that bcl-2 mRNA stability plays a physiological role in modulating bcl-2 expression, particularly in its down-regulation during apoptosis. Thus, this element becomes a new candidate for mediating those bcl-2 gene expression changes-from apoptosis-associated down-regulation to tumor-associated overexpression-observed thus far that profoundly influence single cell fate and tissue homeostasis.

Adduct-specific monoclonal antibodies for the measurement of cisplatin-induced DNA lesions in individual cell nuclei

The anticancer drug cisplatin executes its cytotoxic activity via formation of intra- and interstrand crosslinks in DNA. The relative contribution of structurally defined cisplatin adducts to induce apoptosis and the cellular processing of these lesions is still poorly understood mostly due to the lack of sensitive analytical tools for in vivo studies. Here we describe a new method to establish and characterize monoclonal antibodies (Mab) for structurally defined DNA adducts. The two major reaction products of cisplatin, the guanine–guanine (Pt-[GG]) and adenine–guanine (Pt-[AG]) intrastrand crosslinks are recognized by Mab R-C18 and R-B3, respectively. Both antibodies were employed in an immuno-cytological assay allowing the quantification of drug-induced lesions in individual cell nuclei at clinically relevant doses. Analyzing various tissues of cisplatin-treated C57Bl/6 mice the accumulation of Pt-(GG) was highest in kidney tubular cells compared with 30, 50 and 90% lower levels in kidney stroma, liver and peripheral blood cells, respectively. Adduct kinetics revealed that wild type mouse cells remove up to 80% of the crosslinks in contrast to their complete persistence in nucleotide excision repair-deficient (XPC^−/−) mice. The aptitude of the immunoassay for human molecular dosimetry studies was demonstrated by measuring adduct levels in tumor biopsies from patients treated with cisplatin.

Observation and prevention of an artefactual formation of oxidized DNA bases and nucleosides in the GC-EIMS method

Gas chromatography coupled to electron impact mass spectrometry (GC-EIMS) analysis following hydrolysis of DNA is a widely used assay for the detection of oxidized nucleobases and nucleosides. However, evidence was recently provided for an oxidation of guanine residues of hydrolysed DNA during the silylation prior to GC-EIMS analysis. This reaction accounts for the overestimation of the yield of 8-oxo-7,8-dihydroguanine by GC-EIMS. In the present work, we showed that adenine, cytosine, thymine and thymidine also give rise to oxidized derivatives during the derivatization. This was inferred from the measurement of the amount of 5-formyluracil, 5-hydroxymethyluracil, 5-hydroxycytosine (5-OHCyt), 8-oxo-7,8-dihydroadenine (8-OxoAde) and 5-hydroxymethyl-2'-deoxyuridine (5-HMdUrd) in a series of experiments based on the use of purified bases and nucleosides. Isotopically labelled oxidized bases and 5-HMdUrd were used as internal standards to control the quantitative aspect of the silylation reaction. Support for an artefactual oxidation of hydrolysed DNA was provided by the comparison of the amount of 8-OxoAde and 5-OHCyt detected within native and gamma-irradiated DNA by HPLC-EC and GC-EIMS. To prevent the artefactual formation of oxidized bases during the silylation, an approach based on an HPLC prepurification was developed to remove the precursors of the oxidized bases measured in the DNA sample. The HPLC/GC-EIMS assay was successfully applied to the quantitation of 8-OxoAde and 5-OHCyt in calf thymus DNA. In addition, the detection of the dose-dependent formation of 5-HMdUrd in isolated DNA exposed to ionizing radiation was achieved using the same approach.

RNAs of influenza A, B, and C viruses

The nucleic acids of influenza A, B, and C viruses were compared. Susceptibility to nucleases demonstrates that influenza C virus, just as influenza A and B viruses, possesses single-stranded RNA as its genome. The base compositions of the RNAs of influenza A, B, and influenza C virus are almost identical and comparative analysis on polyacrylamide gels shows that the genome of influenza C/GL/1167/54 virus, like that of the RNAs of influenza A and B viruses, is segmented. Eight distinct RNA bands were found for influenza A/PR/8/34 virus and for influenza B/Lee/40 virus. The RNA of influenza C/GL/1167/54 virus separated into at least four segments. The total molecular weights of the RNA of influenza A/PR/8/34 and B/Lee/40 virus were calculated to be 5.29 X 10(6) and 6.43 X 10(6), respectively. A minimum value of 4.67 X 10(6) daltons was obtained for influenza C/GL/1167/54 virus RNA. The data suggest that influenza C viruses are true members of the influenza virus group.

High-pressure-liquid-chromatographic and fluorimetric methods for the determination of adenine released from ribosomes by ricin and gelonin

The high fluorescence of adenine-containing compounds after reaction with chloroacetaldehyde was used to measure the adenine released from rat liver and Artemia salina ribosomes by the action of ricin A chain and gelonin, two ribosome-inactivating proteins (RIPs) that share the same mechanism of action, consisting in the hydrolysis of the N-glycosidic bond of A-4324 of 28 S rRNA. Two methods were employed: (i) h.p.l.c. of the chloroacetaldehyde-reactive material released by RIPs; h.p.l.c. associated with a fluorescence detector allows the identification of adenine and its dosage at quantities as low as 2 ng; (ii) the direct fluorimetric measurement of the material that had reacted with chloroacetaldehyde. The amount of adenine released increases when ribosomes are pretreated in conditions that lead to their dissociation into subunits. Adenine protects ribosomes from the inhibition by ricin A-chain. When ribosomes were incubated with ricin A-chain in the presence of [14C]adenine no incorporation of radioisotope in ribosomes was observed, indicating that neither exchange nor reversal reactions occurred. A binding of [14C]adenine to ricin A chain was not detected by equilibrium dialysis.

Translocation specificity of the Tn3 element: characterization of sites of multiple insertions

247 independent events involving insertion of the TN3 transposable element into a 4 kb constructed plasmid (pTU4) of partially known DNA sequence were studied by restriction endonuclease mapping, and 65 of these insertion sites were examined further by DNA sequence analysis. Our results show that the previously proposed regional specificity for Tn3 insertion is associated with a strong preference for AT-rich segments as insertion sites. Moreover, multiple insertions of the Tn3 occurred at certain AT-rich nucleotide positions, and 23 of 26 independent insertion events at a single nucleotide position were found to be in the same orientation. A region of the recipient plasmid showing major homology with the terminal 18 bp of Tn3 was identified in the vicinity of an 11 nucleotide segment that included three insertional hot spots and 36 independent insertions. Our results indicate that the site and orientation of insertion of Tn3 are at least partly determined by the primary nucleotide sequence of the recipient genome, and suggest that insertional hot spots may result from the combined effects of AT richness plus homology of the recipient genome with the terminal sequences of Tn3.

Sequence-specific interaction of Hoechst 33258 with the minor groove of an adenine-tract DNA duplex studied in solution by 1H NMR spectroscopy

The interaction of Hoechst 33258 with the minor groove of the adenine-tract DNA duplex d(CTTTTGCAAAAG)2 has been studied in both D2O and H2O solutions by 1D and 2D 1H NMR spectroscopy. Thirty-one nuclear Overhauser effects between drug and nucleotide protons within the minor groove of the duplex, together with ring-current induced perturbations to the chemical shifts of basepair and deoxyribose protons, define the position and orientation of the bound dye molecules. Two drug molecules bind cooperatively and in symmetry related orientations at the centre of the 5'-TTTT and 5'-AAAA sequences with the binding interactions spanning only the four A-T basepairs. The positively charged N-methylpiperazine moieties point towards the centre of the duplex while the phenol groups are disposed towards the 3'-ends of the sequence. Resonance averaging is apparent for both the D2/D6 and D3/D5 phenol protons and D2"'/D6"' and D3"'/D5"' of the N-methylpiperazine ring and is consistent with these groups being involved in rapid rotation or ring-flipping motions in the bound state. Interstrand NOEs between adenine H2s and deoxyribose H1' are consistent with a high degree of propeller twisting of the A-T basepairs at the binding site of the aromatic benzimidazole and phenol rings of Hoechst. The data imply that the minor groove is particularly narrow with many contacts between the complementary curved surfaces of the drug and DNA indicating that strong van der Waals interactions, involving the floor and the walls of the minor groove, stabilize the complex. In our model the NH groups of the benzimidazole rings are positioned to make a pair of bifurcated hydrogen bonds with the adenine N3 and thymine O2 on the floor of the minor groove.

Alternative polyadenylation variants of the RNA binding protein, HuR: abundance, role of AU-rich elements and auto-Regulation

The RNA-binding protein, HuR, is involved in the stabilization of AU-rich element-containing mRNAs with products that are involved in cell-cycle progression, cell differentiation and inflammation. We show that there are multiple polyadenylation variants of HuR mRNA that differ in their abundance, using both bioinformatics and experimental approaches. A polyadenylation variant with distal poly(A) signal is a rare transcript that harbors functional AU-rich elements (ARE) in the 3'UTR. A minimal 60-nt region, but not a mutant form, fused to reporter-3'UTR constructs was able to downregulate the reporter activity. The most predominant and alternatively polyadenylated mature transcript does not contain the ARE. HuR itself binds HuR mRNA, and upregulated the activity of reporter from constructs fused with ARE-isoform and the HuR ARE. Wild-type tristetraprolin (TTP), but not the zinc finger mutant TTP, competes for HuR binding and upregulation of HuR mRNA. The study shows that the HuR gene codes for several polyadenylation variants differentially regulated by AU-rich elements, and demonstrates an auto-regulatory role of HuR.

Repeating units of Xenopus laevis oocyte-type 5S DNA are heterogeneous in length

The restriction enzymes Hind III and Hae III cleave Xenopus laevis 5S DNA at one and three sites, respectively, in each repeating unit of approximately 700 base pairs. The cleavage sites for both enzymes have been located within the repeating unit by denaturation mapping of the restriction fragments. The Hind III products and one of the Hae III fragments are variable in length, indicating heterogeneity in the length of the repeating unit in 5S DNA. This length heterogeneity is confined to the major A + T-rich spacer region. Repeating units differ from each other by discrete quanta of approximately 15 base pairs. The A + T-rich spacer has been shown to consist largely of tandem subrepeats of just this size (Brownlee, Cartwright, and Brown, 1974). We suggest that the repeat-length heterogeneity is due to variable numbers of these subrepeats in the spacer regions of the major repeating units.

Oxidative damage to proteins, lipids, and DNA in cortical brain regions from patients with dementia with Lewy bodies

Dementia with Lewy bodies (DLB) forms the second most common pathological subgroup of dementia after Alzheimer's disease. The present study compares the levels of oxidative damage to proteins, lipids, and DNA bases in cortical brain areas from patients with DLB with levels in matched control tissues. Overall, there was a trend for protein carbonyl levels to be increased in all areas, but a significant difference was found only in the parietal and temporal lobes. No differences were observed in the levels of lipid peroxidation. Measurement of products of damage to DNA bases showed increased levels of thymine glycol, 8-hydroxyguanine, 2,6-diamino-4-hydroxy-5-formamidopyrimidine, 5-hydroxycytosine, 5-hydroxyuracil, 5-hydroxymethyluracil, and xanthine. Xanthine levels were increased in the DLB group in the parietal, occipital, and temporal lobes, indicating that peroxynitrite or other deaminating species may be involved. The finding of increased protein carbonyls and increased DNA base products in cortical regions from DLB patients indicates that oxidative stress may play a role in DLB.

Heterochromatin study demonstrating the non-linearity of fluorometry useful for calculating genomic base composition

A novel procedure for calculating base-pair frequencies in whole genomes is reported. This has been developed during a study of the role of heterochromatin in microevolution. Closely related species of the Crepis praemorsa complex have similar karyotypes but for their heterochromatin. The changes in relative AT frequency between species have been attributed to heterochromatin sequences by in situ banding of chromosomes with two base-specific fluorochromes. The absolute genome size of species, measured by cytofluorometry, correlated positively with increased karyotypic heterochromatin, as did the proportion of AT bases in the DNA. However, the determination of base content has called for a curvilinear interpretation of data obtained with two base-specific fluorochromes (bisbenzimide Hoechst 33342 and mithramycin), in contrast to the commonly assumed but erroneous direct relationship between fluorescence intensity and base content. Essentially, the fluorochromes' requirements for a sequence of certain base-pairs lead to the notion of Coefficients of Overspecificity: the result is a simple formula for calculating the AT proportion in a genome relative to a reference species from cytometric data, taking account of ligand binding statistics. These statistics and probabilities of oligonucleotide binding are essentially the same.

Isotope Edited Internal Standard Method for Quantitative Surface-Enhanced Raman Spectroscopy

A new isotope edited internal standard (IEIS) method for quantitative surface-enhanced Raman spectroscopy (SERS) is demonstrated using rhodamine 6G (R6G-d0) and rhodamine 6G (R6G-d4) edited with deuterium. The reproducibility and accuracy of the IEIS method is investigated both under optical resonance (SERRS) and nonresonance (SERS) conditions. A batch-to-batch concentration measurement reproducibility of better than 3% is demonstrated over a concentration range of 200 pM-2 microM with up to a factor of 3 difference between the concentration of the analyte and its IEIS. The superior performance of the IEIS method is further illustrated by comparing results obtained using absolute SERS/SERRS intensity calibration (with no internal standard) or using adenine (rather than R6G-d4) as an internal standard for R6G concentration quantization. Potential biomedical gene expression and comparative proteomic applications of the IEIS method are discussed.

Synthesis and applications of RNAs with position-selective labelling and mosaic composition

Knowledge of the structure and dynamics of RNA molecules is critical to understanding their many biological functions. Furthermore, synthetic RNAs have applications as therapeutics and molecular sensors. Both research and technological applications of RNA would be dramatically enhanced by methods that enable incorporation of modified or labelled nucleotides into specifically designated positions or regions of RNA. However, the synthesis of tens of milligrams of such RNAs using existing methods has been impossible. Here we develop a hybrid solid-liquid phase transcription method and automated robotic platform for the synthesis of RNAs with position-selective labelling. We demonstrate its use by successfully preparing various isotope- or fluorescently labelled versions of the 71-nucleotide aptamer domain of an adenine riboswitch for nuclear magnetic resonance spectroscopy or single-molecule Förster resonance energy transfer, respectively. Those RNAs include molecules that were selectively isotope-labelled in specific loops, linkers, a helix, several discrete positions, or a single internal position, as well as RNA molecules that were fluorescently labelled in and near kissing loops. These selectively labelled RNAs have the same fold as those transcribed using conventional methods, but they greatly simplify the interpretation of NMR spectra. The single-position isotope- and fluorescently labelled RNA samples reveal multiple conformational states of the adenine riboswitch. Lastly, we describe a robotic platform and the operation that automates this technology. Our selective labelling method may be useful for studying RNA structure and dynamics and for making RNA sensors for a variety of applications including cell-biological studies, substance detection, and disease diagnostics.

Methylated purines in the deoxyribonucleic acid of various Syrian-golden-hamster tissues after administration of a hepatocarcinogenic dose of dimethylnitrosamine

1. DNA was extracted from livers, kidneys and lungs of Syrian golden hamsters at various times (up to 96h) after injection of a hepatocarcinogenic dose of [14C]dimethylnitrosamine. Purine bases were released from the DNA by mild acid hydrolysis and separated by Sephadex G-10 chromatography. 2. At 7h after dimethylnitrosamine administration liver DNA was alkylated to the greatest extent, followed by that of lung and kidney, the values for which were 8 and 3% respectively of those for liver. 3. The O6-methylguanine/7-methylguanine ratios were initially the same in all three organs and in the liver DNA of rats under similar conditions of dose. 4. O6-Methylguanine was the most persistent alkylated purine in all three hamster tissues. There was evidence for excision of 7-methyl-guanine, the highest activity for this being present in the liver. 5. Detectable amounts of the minor products 3-methyladenine, 1-methyladenine, 3-methylguanine and 7-methyladenine were present in most hamster tissues, and their individual rates of loss from liver DNA were determined. 6. Ring-labelling of the normal purines in DNA was highest in the liver, followed closely by the lung (80% of that in liver) whereas the kidney had very low incorporation (3% of that in liver). 7. The results are discussed with respect to the hepatotoxicity of dimethylnitrosamine, the miscoding potential of the various alkylation products and the induction of liver tumours in hamsters.

A colorimetric assay for the quantitation of free adenine applied to determine the enzymatic activity of ribosome-inactivating proteins

Adenine quantitation is required for a variety of applications. To date, the prevalent method for quantifying free adenine, in a variety of applications, is the detection of fluorescent-derivatized adenine by HPLC. For the present study, we developed a high-throughput, nonradioactive, enzyme-based colorimetric adenine quantitation assay that is performed in one multireaction incubation step. The assay does not require adenine derivatization and is designed for microplates. The key step is the conversion of adenine to adenosine monophosphate by adenine phosphoribosyl transferase. Subsequent reactions finally produce three inorganic phosphate ions per adenine molecule. Phosphate is quantitated by the color-generating phosphorylysis of a particular purine derivate. Ribosome-inactivating proteins that release adenine from polynucleotides are often used to investigate intracellular protein trafficking and are important for the design of immunotoxins. We therefore used ricin, dianthin, saporin, and a variety of saporin fusion proteins to show that this method is suitable for quantifying adenine release using different substrates. The measured rate of adenine release and substrate specificity are comparable to those determined by HPLC and radioactive detection techniques.

Modified bases in the DNAs of unicellular eukaryotes: an examination of distributions and possible roles, with emphasis on hydroxymethyluracil in dinoflagellates

The occurrence of small amounts of one or more of several modified bases in the DNA of an organism is widespread in nature. Prominent among these bases are 5-methylcytosine, N6-methyladenine and 5-hydroxymethyluracil. All can be found in varying amounts in DNA of viral, prokaryotic and eukaryotic origin. In some organisms, modified nucleotides comprise a large fraction of DNA nucleotides and in others there is complete replacement of one of the common four nucleotides by a modified one. This article discusses the distributions and possible roles of the several modified bases found in prokaryote and eukaryote DNAs. Emphasis is given (1) methylcytosine in a broad variety of eukaryotes, (2) methyladenine in certain protozoa and protophyta and (3) hydroxymethyluracil in dinoflagellates. Attention is focused on the phenomenology and the possible consequences of the presence of hydroxymethyluracil in DNA.

Nonenzymatic template-directed synthesis on hairpin oligonucleotides. 3. Incorporation of adenosine and uridine residues

We have used [32P]-labeled hairpin oligonucleotides to study template-directed synthesis on templates containing one or more A or T residues within a run of C residues. When nucleoside-5'-phosphoro(2-methyl)imidazolides are used as substrates, isolated A and T residues function efficiently in facilitating the incorporation of U and A, respectively. The reactions are regiospecific, producing mainly 3'-5'-phosphodiester bonds. Pairs of consecutive non-C residues are copied much less efficiently. Limited synthesis of CA and AC sequences on templates containing TG and GT sequences was observed along with some synthesis of the AA sequences on templates containing TT sequences. The other dimer sequences investigated, AA, AG, GA, TA, and AT, could not be copied. If A is absent from the reaction mixture, misincorporation of G residues is a significant reaction on templates containing an isolated T residue or two consecutive T residues. However, if both A and G are present, A is incorporated to a much greater extent than G. We believe that wobble-pairing between T and G is responsible for misincorporation when only G is present.

The adsorption of nucleotides and polynucleotides on montmorillonite clay

The binding of adenine derivatives to Na(+)-montmorillonite increases in the order 5'AMP, 3'-AMP, 5'ADP < adenosine < purine, adenine. With the exception of cytosine, cytosine derivatives bind less strongly than the corresponding adenine derivatives in the order 5'-CMP < cytidine < cytosine. There is little difference in the binding of uracil derivatives and these compounds bind less strongly than the corresponding adenine analogs. It is concluded that the adenine ring in adenine derivatives is protonated by the acidic montmorillonite surface and binding is a consequence of the electrostatic interaction between the protonated base and the negative charges on the surface of the montmorillonite. Different binding trends were observed with Cu(2+)-montmorillonite with AMP binding more strongly than adenosine and UMP binding more strongly than uridine. It is concluded that ligation to the Cu2+ is a major force in the binding of nucleotides to Cu(2+)-montmorillonite and are not readily washed from the clay. Factors contributing to the binding are discussed. Watson-Crick hydrogen bonding of 5'-AMP to poly(U) and 5'GMP to poly(C) was observed when the homopolymers are bound to the surface of the clay. No association of 5'-UMP to poly(U) bound to clay was detected. The possible role of montmorillonite clays in the prebiotic formation of RNA is discussed.

Measurement of oxidative DNA damage by gas chromatography-mass spectrometry: ethanethiol prevents artifactual generation of oxidized DNA bases

Analysis of oxidative damage to DNA bases by GC-MS enables identification of a range of base oxidation products, but requires a derivatization procedure. However, derivatization at high temperature in the presence of air can cause 'artifactual' oxidation of some undamaged bases, leading to an overestimation of their oxidation products, including 8-hydroxyguanine. Therefore derivatization conditions that could minimize this problem were investigated. Decreasing derivatization temperature to 23 degrees C lowered levels of 8-hydroxyguanine, 8-hydroxyadenine, 5-hydroxycytosine and 5-(hydroxymethyl)uracil measured by GC-MS in hydrolysed calf thymus DNA. Addition of the reducing agent ethanethiol (5%, v/v) to DNA samples during trimethylsilylation at 90 degrees C also decreased levels of these four oxidized DNA bases as well as 5-hydroxyuracil. Removal of guanine from hydrolysed DNA samples by treatment with guanase, prior to derivatization, resulted in 8-hydroxyguanine levels (54-59 pmol/mg of DNA) that were significantly lower than samples not pretreated with guanase, independent of the derivatization conditions used. Only hydrolysed DNA samples that were derivatized at 23 degrees C in the presence of ethanethiol produced 8-hydroxyguanine levels (56+/-8 pmol/mg of DNA) that were as low as those of guanase-pretreated samples. Levels of other oxidized bases were similar to samples derivatized at 23 degrees C without ethanethiol, except for 5-hydroxycytosine and 5-hydroxyuracil, which were further decreased by ethanethiol. Levels of 8-hydroxyguanine, 8-hydroxyadenine and 5-hydroxycytosine measured in hydrolysed calf thymus DNA by the improved procedures described here were comparable with those reported previously by HPLC with electrochemical detection and by GC-MS with prepurification to remove undamaged base. We conclude that artifactual oxidation of DNA bases during derivatization can be prevented by decreasing the temperature to 23 degrees C, removing air from the derivatization reaction and adding ethanethiol.

Heterogeneity of Epstein-Barr virus. III. Comparison of a transforming and a nontransforming virus by partial denaturation mapping of their DNAs

The DNAs of a transforming and a nontransforming Epstein-Barr virus strain, B95-8 AND P3HR-1, were compared by partial denturation mapping. B95-8 viral DNA showed a homogeneous denaturation pattern. In contrast, P3HR-1 viral DNA was heterogeneous, containing at least two classes of molecules, classified into groups A and B and present in a ratio of about 2:1 to 3:1. No evidence could be obtained that molecules from both groups A and B contain identical sequences present in different orientations as described for herpes simplex viral DNA. The majority of sequences present in B95-8 and in P3HR-1 viral DNA group A could be correlated by assuming that different sequences, about 12,000 base pairs long, were inserted or deleted, respectively, at different position of both viral genomes.