Spermidine biosynthesis in Saccharomyces cerevisae: polyamine requirement of a null mutant of the SPE3 gene (spermidine synthase)

The Saccharomyces cerevisiae SPE3 gene, coding for spermidine synthase, was cloned, sequenced, and localized on the right arm of chromosome XVI. The deduced amino acid sequence has a high similarity to mammalian spermidine synthases, and has putative S-adenosylmethionine binding motifs. To investigate the effect of total loss of the SPE3 gene, we constructed a null mutant of this gene, spe3delta, which has no spermidine synthase activity and has an absolute requirement for spermidine or spermine for the growth. This requirement is satisfied by a very low concentration of spermidine (10(-8) M) or a higher concentration of spermine (10(-6) M).

Benzo[a]pyrene diol epoxide-DNA cis adduct formation through a trans chlorohydrin intermediate

Alkylation of DNA by 7r,8t-dihydroxy,9t,10t-epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene (anti-BPDE) forms mainly trans adducts (with respect to the C-9/10 positions). We recently described a halide-catalyzed pathway that preferentially generates cis adducts and now report that the trans chlorohydrin of anti-BPDE (trans-BPDCH) is an intermediate in the chloride-catalyzed reaction. trans-BPDCH was synthesized, and both it and anti-BPDE were reacted with deoxyadenosine as a model DNA nucleophile. The stereochemistry and yields of deoxyadenosine adducts were determined as a function of chloride concentration. In the absence of salt, the fraction of cis adducts obtained from anti-BPDE and trans-BPDCH are 0.33 and 0.67, respectively. Adding sodium chloride increases the fraction of cis adducts (and consequently decreases the fraction of trans adducts), with the midpoint of the increase for both substrates at approximately 35-40 mM chloride. The chloride-dependent curves for BPDE and BPDCH converge at 1 M chloride, where the fraction of cis adducts is 0.88. Chloride also increases the total yield of cis adducts with either substrate, whereas the yield of trans adducts from the chlorohydrin is not significantly changed. These results support a mechanism by which chloride ion undergoes nucleophilic addition to the benzylic C-10 position of anti-BPDE. This generates a trans halohydrin that alkylates DNA with inversion of configuration to form a cis adduct. This pathway may have biological significance because chlorohydrins could form in serum or in cells with relatively high intracellular concentrations of chloride.

Molecular cloning and functional expression of cDNAs encoding a human Na+-nucleoside cotransporter (hCNT1)

We report identification of a new human nucleoside transporter protein by molecular cloning and functional expression of its cDNA. Previously, we used expression selection in Xenopus oocytes to isolate a cDNA from rat jejunal epithelium encoding the pyrimidine-selective Na+-dependent nucleoside transporter rCNT1 (Q.-Q. Huang, S. Y. M. Yao, M. W. L. Ritzel, A. R. P. Paterson, C. E. Cass, and J. D. Young. J. Biol. Chem. 269: 17757-17760, 1994). cDNAs for a human homologue of rCNT1, designated hCNT1, have been isolated from human kidney by hybridization cloning and reverse transcriptase polymerase chain reaction amplification strategies. hCNT1 was 83% identical to rCNT1 in amino acid sequence and exhibited the transport characteristics of an Na+-dependent nucleoside transporter with selectivity for pyrimidine nucleosides and adenosine when expressed in Xenopus oocytes. Deoxyadenosine, which undergoes net renal secretion, and guanosine were poor permeants. hCNT1 did, however, transport 3'-azido-3'-deoxythymidine. This is the first demonstration that members of the CNT family exist in human cells and provides evidence of their involvement in the renal transport of physiological nucleosides and nucleoside drugs. The hCNT1 gene was mapped to chromosome 15q25-26.

Metabolism of adenosine and deoxyadenosine by human erythrocytes and CCRF-CEM leukemia cells

Human lymphocytes lacking adenosine deaminase die and T-cell leukemias are killed by deoxycoformycin (dCf), an inhibitor of adenosine deaminase, due to impaired metabolism of dAdo. The initial metabolism of exogenous adenosine (Ado) and deoxyadenosine (dAdo) has been compared in human erythrocytes and CCRF-CEM leukemia cells and the data obtained have been simulated using kinetic constants obtained in vitro for the enzymes involved. Cells were mixed with 3H-labelled Ado and dAdo, samples were taken at 3 sec intervals and progress curves for the 3H-labelled metabolites formed were determined by quantitative two-dimensional thin layer chromatography. Erythrocytes rapidly take up Ado and the predominant metabolite after 60 sec is hypoxanthine (Hyp), while for dAdo, deoxyinosine (dIno) predominates. By contrast, leukemia cells convert to Ado predominantly to AMP, while dAdo is converted first to Hyp and the to AMP. The presence of dCf had little effect upon Ado metabolism by induced accumulation of dAdo. Erythrocytes rapidly degrade Ado and dAdo to Hyp, although the phosphorolysis of dIno is relatively slow. Human CCRF-CEM leukemia cells convert most of the Ado or dAdo to AMP after 60 sec. For dAdo, the sequence of reactions would be dAdo-->dIno-->Hyp-->IMP-->sAMP-->AMP. dCf does not significantly affect the conversion of Ado-->AMP, but dCf blocks AMP accumulation from dAdo, consistent with the reaction sequence shown above. A computer model has been developed for the metabolism of Ado and dAdo, but some of the kinetic constants determined in vitro for this model do not pertain to intact cells.

2'-deoxyadenosine induces apoptosis in rat chromaffin cells

We show here that 2'-deoxyadenosine (2'-dAdo) but not adenosine was toxic to chromaffin cells of 3-4-week-old rat adrenal glands. More than 75% of the cells plated in culture gradually died over a 3-day period in the presence of 100 microM 2'-dAdo plus 3 microM deoxycoformycin (DCF). Morphological observations together with bisbenzimide staining and terminal deoxynucleotidyl transferase-mediated nick and labeling showed membrane blebbing, shrinkage of cell bodies, chromatin condensation, and DNA fragmentation, suggesting apoptosis-like cell death by 2'-dAdo. Lethal effects of 2'-dAdo were potentiated by DCF, a drug that inhibits adenosine deaminase. 2'-dAdo-prompted cell death was not prevented by inhibitors of nucleoside transporter (3 microM dilazep or 1 microM nitrobenzylthioinosine), precursors of pyrimidine nucleotide biosynthesis (300 microM uridine or 100 microM 2'-deoxycytidine), or 5 mM nicotinamide. Cells incubated with 2'-dAdo (100 and 300 microM) showed a three- and ninefold, respectively, increase in content of dATP, a product known to be an inhibitor of ribonucleotide reductase, an enzyme essential for DNA synthesis. Formation of dATP was completely prevented by iodotubercidin (ITu), a drug that inhibits phosphorylation of 2'-dAdo to dATP by nucleoside kinase. It is interesting that nanomolar concentrations of ITu also completely protected chromaffin cells from 2'-dAdo lethality. Our study demonstrates for the first time that mammalian adrenal chromaffin cells undergo apoptotic cell death by a natural nucleoside and suggests that this model could be used to study apoptosis and cell function.

Inhibition of the human immunodeficiency virus type 1 integrase by guanosine quartet structures

An oligonucleotide (T30177) composed entirely of deoxyguanosine and thymidine has previously been shown to fold upon itself in the presence of potassium into a highly stable four-stranded DNA structure containing two stacked deoxyguanosine quartets (G4s). T30177 also protects host cells from the cytopathic effects of human immunodeficiency virus type 1 (HIV-1). We report that this G4 oligonucleotide is the most potent inhibitor of HIV-1 integrase identified to date, with IC50 values in the nanomolar range. Both the number of quartets formed and the sequence of the loops between the quartets are important for optimal activity. T30177 binds to HIV-1 integrase without being processed and blocks the binding of the normal viral DNA substrate to the enzyme. The normal DNA substrate was not able to compete off T30177 binding to HIV-1 integrase, indicating a tight binding of G4s to the enzyme. Experiments with truncated HIV-1 integrases indicate that the N-terminal region containing a putative zinc finger is required for inhibition by T30177 and that T30177 binds better to full-length or deletion mutant integrases containing the zinc finger region than to a deletion mutant consisting of only the central catalytic domain. The N-terminal region of integrase alone is able to bind efficiently to T30177, but not the linear viral DNA substrate, in the presence of zinc. Hence, G4s represent the first class of compounds that inhibit HIV-1 integrase by interacting with the enzyme N-terminal domain. The greater inhibitory potency of T30177 in buffer containing magnesium versus manganese suggests that divalent metal ion coordination along the phosphodiester backbone may play a role in the inhibitory activity. T30177 inhibited HIV-2 integrase with similar potency as HIV-1 but inhibited feline and simian immunodeficiency virus integrases at higher concentrations, suggesting selectivity can be achieved. We propose that novel AIDS therapies could be based upon guanosine quarters as inhibitors of HIV-1 integrase.

Substrate specificity of Escherichia coli MutY protein

The MutY protein of Escherichia coli removes mismatched deoxyadenine residues from DNA. In this study, duplex oligodeoxynucleotides containing modified bases are used as model substrates for this enzyme. In contrast to a recent report [Lu, A.-L., et al. (1995) J. Biol. Chem. 270, 23582], dA:8-oxo-dG appears to be the preferred natural substrate for MutY, as evidenced by the specificity constants (kcat/Km) for dA:8-oxo-dG and dA:dG of 39 600 x 10(-6) and 383 x 10(-6) (min-1 nM-1), respectively. kcat for the duplex containing dA:dG was highest at lower pH; the rate of cleavage for the duplex containing dA:8-oxo-dG was unaffected over a pH range of 5.5-8.0. The presence of an 8-oxo function in dG increased significantly the rate of removal of dA from all substrates tested. Replacement of dA by rA reduced the specificity constant of dA:8-oxo-dG to 294 x 10(-6) (min-1 nM-1), whereas replacement of dA by 2'-O-methyladenosine virtually abolished enzymatic activity. Modifications of the dG moiety generally were better tolerated than those of dA; however, introduction of a methyl ether at the 6 position of dG produced a noncleavable substrate and replacement of dG by 2'-O-methylguanosine generated a substrate with a low specificity constant. Rates of cleavage of duplexes containing dA:dC and dA:tetrahydrofuran were three orders of magnitude lower than the reference substrate. Duplexes containing a carbocyclic analog of dA were not cleaved. A model is proposed to explain the recognition of DNA substrates by MutY and the catalytic properties of this enzyme.

1,N6-ethenodeoxyadenosine, a DNA adduct highly mutagenic in mammalian cells

1,N6-Ethenodeoxyadenosine (epsilon dA) is one of four exocyclic DNA adducts produced by chloroethylene oxide and chloroacetaldehyde, reactive metabolites of vinyl chloride, a human carcinogen. epsilon dA has also been detected in DNA of the liver of humans and untreated animals, suggesting its formation from endogenous sources. The mutagenic potential of epsilon dA was studied using a single-stranded shuttle vector system in several E. coli strains and in simian kidney cells (COS7). This vector system enables quantitative analysis of translesional synthesis past a site-specifically placed DNA adduct in both hosts owing to the lack of the complementary strand. In experiments with five strains of E. coli, a very limited number of targeted mutations (one epsilon dA-->T, one epsilon dA-->dC, and two epsilon dA-->single base deletion) were observed among 756 transformants in hosts preirradiated with UV; no targeted mutations were observed among 563 transformants in nonirradiated hosts. These results indicate that nonmutagenic base pairings of epsilon dA:T are the almost exclusive events in E. coli. In COS7 cells, the frequency of targeted mutations was 70%, consisting of epsilon dA-->dG (63%), epsilon dA-->T (6%), and epsilon dA-->dC (1%), indicating that the insertion of dCMP opposite the adduct is predominant. When compared with the results for 3,N4-ethenodeoxycytidine (epsilon dC), which was studied previously in the same system [Moriya et al. (1994) Proc. Natl. Acad. Sci. U.S.A. 91, 11899-11903], the results of this study indicate that the intrinsic mutagenic potency of epsilon dA is comparable to that of epsilon dC in mammalian cells.

Lipid peroxidation-induced putative malondialdehyde-DNA adducts in human breast tissues

The etiology of the majority of human breast cancers is unknown; however, oxidative stress and lipid peroxidation have been suggested to play a role in breast carcinogenesis. To address this possibility, DNA adducts induced by malondialdehyde (MDA), an end product of lipid peroxidation, were analyzed in surgical specimens of normal breast tissues of 51 breast cancer patients using the nuclease P1-enhanced version of the 32P-postlabeling assay. Normal breast tissue samples from 28 noncancer patients receiving reduction mammoplasty served as controls. Two previously characterized putative MDA-deoxyadenosine (dA) and one MDA-deoxyguanosine adduct were detected in all tissue samples examined. Normal breast tissues from cancer patients exhibited significantly higher levels of the putative MDA adducts [median (42.5) and range (2.2-202.8) of relative adduct labeling x 10(9) values] than those found in noncancer controls (median, 15.67; range, 2.4-382.1; P = 0.0001, Mann-Whitney U test). Ten of the 51 cancer patients and 1 of the 28 controls were found to contain the putative MDA adducts at the level of > 1/10(7) nucleotides, a frequency comparable to that found in human liver. Age and body mass did not significantly influence the levels of these adducts. However, the presence of a previously detected benzo(a)pyrene-like DNA adduct in the breast tissues was associated with higher levels of the putative MDA-dA adducts in cancer patients (P = 0.012). The level of the putative MDA-dA adducts was significantly lower in smokers and former smokers compared to nonsmokers among cases after adjusting for age, body mass index, and status of the benzo(a)pyrene-like adduct (P = 0.009). Tumor tissues (n = 11) displayed significantly lower levels of the putative MDA adducts (median, 10.2; range, 5.3-20.6) than their corresponding normal adjacent tissues (median, 25.5; range, 10.5-138; P < 0.01). These findings provide evidence that lipid peroxidation products can accumulate in human breast tissues and reach relatively high levels in the breast tissues of women with breast cancer. There seems to be an interaction between these endogenous DNA modifications and carcinogen exposure-induced DNA adducts. Detection and quantitation of the putative MDA-DNA adducts may potentially be a useful tool in the understanding of breast cancer etiology.

Formation and accumulation of DNA ethenobases in adult Sprague-Dawley rats exposed to vinyl chloride

DNA ethenobases are promutagenic lesions formed by carcinogens such as vinyl chloride (VC). Their formation was investigated in 6-week old, male Sprague-Dawley rats exposed to 500 p.p.m. VC by inhalation (4 h/day, 5 days/ week) for 1, 2, 4 or 8 weeks and in 7- and 14-week old, matched control animals. 1,N6-Ethenoadenine (epsilon A) and 3, N4-ethenocytosine (epsilon C) deoxyribonucleotides were analysed by immunoaffinity purification and 32P-postlabelling. This postlabelling method was compared with a radio-immunoassay method, which yielded similar results. Background levels of ethenobases were found in DNA from the liver, lungs, kidneys and circulating lymphocytes of unexposed, control rats. In the liver, the following background molar ratios of ethenobase to parent base in DNA were detected (mean values x 10(-8)): epsilon A/A, 0.04-0.05; epsilon C/C, 0.06-0.07. In the lungs, kidneys and circulating lymphocytes, background levels of epsilon A and epsilon C ranged from 1.7 to 4.2 x 10(-8) and from 4.8 to 11.2 x 10(-8), respectively. Following a 5-day exposure to VC, a significant increase of epsilon A and epsilon C was measured in hepatic DNA from rats sacrificed immediately after treatment. Further, a dose-dependent increase of both etheno adducts was observed in liver DNA of VC-treated rats. Compared to the 5-day exposure, approximately 4-fold higher levels of epsilon A and epsilon C were observed in the liver of animals after 8 weeks of exposure. In contrast, there was an accumulation of epsilon C but not of epsilon A in lungs and kidneys. In circulating lymphocytes, no significant increase of ethenobase levels above control values was observed after 2 months of exposure to VC. Both etheno adducts were found to be persistent in liver DNA, after 2 months following the termination of VC exposure. These results further support the notion that DNA etheno-bases are critical lesions in VC-induced carcinogenesis. The possible contribution of lipid peroxidation products that also yield ethenobases, on the formation and persistence of these DNA adducts, remains to be clarified.

Stabilization of triple-stranded oligonucleotide complexes: use of probes containing alternating phosphodiester and stereo-uniform cationic phosphoramidate linkages

Pyrimidine oligonucleotides containing alternating anionic and stereo-uniform cationic N-(dimethylamino-propyl)phosphoramidate linkages [e.g. d(T+T-)7T, d(T+T-)2(T+C-)5T and (U'+U'-)7dT, where U' is 2'-O-methyluridine)] are shown to bind to complementary double-stranded DNA segments in 0.1 M NaCl at pH 7 to form triple-stranded complexes with the pyrimidine.purine.pyrimidine motif. For each of the sequences investigated, one stereoisomer bound with higher affinity, and the other stereoisomer with lower affinity, than the corresponding all-phosphodiester oligonucleotide. The stereoisomer of d(T+T-)7T that interacted weakly with a dT.dA target in 0.1 M NaCl formed a novel dA.dA.dT triple-stranded complex with poly(dA) or d(Al5C4A15) in 1 M NaCl; in contrast, the stereoisomer that bound strongly to the dT.dA target failed to form a dA.dA.dT triple-stranded complex.

Characterization of an N6-oxopropenyl-2'-deoxyadenosine adduct in malondialdehyde-modified DNA using liquid chromatography/electrospray ionization tandem mass spectrometry

Malondialdehyde (MDA), a product of lipid peroxidation, causes mutations in bacterial and mammalian cells and cancer in rats. MDA reacts with deoxynucleosides in vitro and the monomeric adduct of MDA with deoxyguanosine (M1G-dR) is the major adduct. M1G-dR has been detected in rat and human liver. Random mutagenesis studies with MDA-modified DNA and recent 32P-postlabeling studies indicate that in addition to M1G-dR, adducts to deoxyadenosine may also be formed. We have utilized liquid chromatography coupled with electrospray ionization tandem mass spectrometry to characterize an N6-oxopropenyl-2'-deoxyadenosine adduct (M1A-dR) in calf DNA modified with MDA.

Peroxynitrite mediated oxidation of purine bases of nucleosides and isolated DNA

Reaction of nitric oxide with superoxide anion produces the highly reactive species peroxynitrite (ONOO-). This compound has been shown to be a strong oxidant of lipids and proteins. However, no data are available on its effect on DNA, with the exception of the induction of strand breaks. We report the result of studies on the reactions of peroxynitrite with the adenine and guanine moieties of nucleosides and isolated DNA. The samples were analyzed for 8-oxo-7,8-dihydro-2'-deoxyguanosine (8-oxo-dGuo),2,2-diamino-4-[(2-deoxy-beta-D-erythro-pentofuranosyl) amino]-5-(2H)-oxazolone (oxazolone) and 8-oxo-7,8-dihydro-2'-deoxyadenosine (8-oxo-dAdo). The effects of peroxynitrite treatment were compared with those of ionizing radiation in aerated aqueous solution, chosen as a source of hydroxyl radicals. At the nucleoside level, both oxidizing conditions led to the formation of oxazolone and 8-oxo-dAdo. In addition, evidence was provided for the formation of the 4R* and 4S* diastereoisomers of 4-hydroxy-8-oxo-4, 8-dihydro-2'-deoxyguanosine. The latter dGuo oxidation products were chosen as markers of the release of singlet oxygen (1O2) upon reaction of peroxynitrous acid with hydrogen peroxide. Oxidation of purine bases was then studied within isolated DNA. A significant increase in the level of 8-oxo-dGuo, oxazolone and 8-oxo-dAdo was observed within double stranded DNA upon exposure to gamma-radiation. Oxazolone and 8-oxo-dAdo were formed upon peroxynitrite treatment but no significant increase in the amount of 8-oxo-dGuo was detected. These results showed that peroxynitrite exhibits oxidizing properties toward purine moieties both in nucleosides and isolated DNA. However, the significant differences in the oxidative damage distribution within DNA observed after exposure to gamma radiation by comparison with peroxynitrite treatment questions the involvement of hydroxyl radicals as the main oxidizing species released by decomposition of peroxynitrous acid.

Copper-dependent formation of miscoding etheno-DNA adducts in the liver of Long Evans cinnamon (LEC) rats developing hereditary hepatitis and hepatocellular carcinoma

Formation of etheno-DNA adducts in the liver was investigated in Long Evans cinnamon (LEC) rats, a Long Evans strain with hereditary abnormal copper metabolism, which develop spontaneous hepatitis and later hepatocellular carcinoma. Using an ultrasensitive immunoaffinity/32P-postlabeling assay (J. Nair et al., Carcinogenesis, 16: 613-617, 1995), the etheno adducts 1,N6-ethenodeoxyadenosine (epsilon dA) and 3,N4-ethenodeoxycytidine (epsilon dC) were measured in the liver of 7-, 18-, 30-, and 87-week-old LEC rats. Levels were highest in the liver of 18-week old rats 85 +/- 17 (epsilon dA) and 85 +/- 30 (epsilon dC) adducts per 10(9) parent nucleotides, and the increase in the levels of etheno adducts was age dependent. Age-matched Long Evans agouti rats, a tumor-free sibling line of LEC rats, had much lower levels of both etheno adducts. Etheno adduct levels in LEC rats were well correlated with the hepatic copper levels, and peak adduct levels coincided with the age of commencement of fulminant hepatitis. Our results demonstrate for the first time a copper- and age-dependent formation of highly miscoding etheno-DNA adducts in the liver of LEC rats. These adducts are formed from lipid peroxidation products (F. El-Ghissassi et al., Chem. Res. Toxicol., 8: 273-283, 1995) and thus could arise in the liver of LEC rats from oxygen radicals generated by copper-catalyzed Fenton-type reactions. Etheno-DNA adducts along with other oxidative DNA base damages may thus be involved in liver carcinogenesis in LEC rats.

Assessment of tumor cell proliferation using [18F]fluorodeoxyadenosine and[18F]fluoroethyluracil

This study was to develop radiofluorinated ethyluracil (FEU) and deoxyadenosine analogues (FAD) for noninvasive assessment of tumor proliferative potential by positron emission tomography (PET). 5-(2-Fluoroethyl)uracil ([18F]FEU) was prepared by treating 2,4-dimethoxy-5-(2-hydroxyethyl)pyrimidine with K18F, followed by hydrolysis with HBr. Fluorodeoxyadenosine ([18F]FAD) was prepared by treating a triacetylated analogue of adenosine with K18F. In vitro cell proliferation assay of [18F]-FEU was performed using human peripheral blood mononucleus cells. Tissue distributions were studied in breast tumor-bearing rats at 0.5, 1, 2 and 4 h along with autoradiography at 45 min postinjection. PET imaging studies were conducted in VX-2 tumor-bearing rabbits. In vitro assay indicated that [18F]FEU incorporated into DNA/RNA during cell proliferation. Tumor-to-tissue count density ratios of [18F]FAD and [18F]-FEU increased as a function of time. [18F]FAD had higher tumor-to-nontumor tissue count density ratios than [18F]FEU. Autoradiograms of [18F]FEU and [18F]FAD, and PET images of [18F]FEU, showed that the tumors could be well visualized. The results suggest that [18F]FEU and [18F]FAD have potential use in evaluating tumor cell proliferation by PET.

Impact of the stereochemistry of benzo[a]pyrene 7,8-dihydrodiol 9,10-epoxide-deoxyadenosine adducts on resistance to digestion by phosphodiesterases I and II and translesion synthesis with HIV-1 reverse transcriptase

Spatial orientations of bulky DNA adducts can influence the extent of resistance to digestion by exonucleases and translesion synthesis by HIV-1 reverse transcriptase (HIV-1 RT). In order to determine how different diastereomers of benzo[a]pyrene 7,8-dihydrodiol 9,10-epoxide (BPDE)-adducted DNAs influence the activity of these enzymes, 11-mer and 33-mer oligodeoxyribonucleotides were synthesized bearing site-specific and stereospecific BPDE adducts at adenine N6 on position two of the human N-ras codon 61. Phosphodiesterase I, which hydrolyzes DNA in the 3'-->5' direction, exhibited greater resistance opposite the lesion with C10-R BPDE-adducted templates than the corresponding C10-S adducts. However, the opposite stereoselective resistance to digestion was observed with phosphodiesterase II, which hydrolyzes DNA in the 5'-->3' direction. These results are complemented by the in vitro replication pattern exhibited with HIV-1 RT. Primer extension reactions under conditions defining single encounters between polymerase and substrate revealed adduct-dependent termination one base 3' to each of the lesions. When experimental conditions were altered to permit multiple encounters, HIV-1 RT was able to replicate past the damaged site on four of the six adducted templates, exhibiting little pausing opposite the lesion. Analyses of the replication pattern past these lesions revealed two general categories of replication blockage, which, like the exonucleolytic digestion data, were also based on the C10-R and C10-S configuration of the stereoisomers. Thus, the chirality of BPDE-dA adducts modulates enzymatic functions. Furthermore, the (+)- and (-)-anti-trans-BPDE-dA modified templates exhibited the most facile bypass, while the (+)- and (-)-anti-cis-BPDE adducts were most blocking.

Purine deoxynucleoside metabolism in human melanoma cells with a high spontaneous mutation rate

A human melanoma cell line (MM96L) had a spontaneous mutation rate at the HGPRT locus of approx. 7 times normal. The cells had elevated dATP and dGTP pools, lacked purine nucleoside phosphorylase (PNP) and were sensitive to killing by deoxyadenosine, deoxyinosine and related purines but not to inosine or hypoxanthine. Four other melanoma cell lines exhibited a range of nucleoside sensitivities and dNTP pool sizes. Failure of intact MM96L cells to degrade exogenous deoxyadenosine and deoxyinosine to hypoxanthine was confirmed by NMR of culture medium. Normal melanocytes were PNP+ and were insensitive to deoxyinosine. Comparison of the metabolites of [14C]deoxyinosine from MM96L and a PNP+ cell line of similar doubling time (HeLa) showed that both cell types produced 14C-labelled guanine and adenine nucleotides, with [14C]dATP and [14C]dADP being found in MM96L. This indicates that human sAMP synthetase or a similar enzyme catalyses the conversion of dIMP to dAMP, the resultant elevation of dATP causing base misincorporation and a mutator phenotype.

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.

Thiyl radicals in ribonucleotide reductases

The ribonucleoside triphosphate reductase (RTPR) from Lactobacillus leichmannii catalyzes adenosylcobalamin (AdoCbl)-dependent nucleotide reduction, as well as exchange of the 5' hydrogens of AdoCbl with solvent. A protein-based thiyl radical is proposed as an intermediate in both of these processes. In the presence of RTPR containing specifically deuterated cysteine residues, the electron paramagnetic resonance (EPR) spectrum of an intermediate in the exchange reaction and the reduction reaction, trapped by rapid freeze quench techniques, exhibits narrowed hyperfine features relative to the corresponding unlabeled RTPR. The spectrum was interpreted to represent a thiyl radical coupled to cob(II)alamin. Another proposed intermediate, 5'-deoxyadenosine, was detected by rapid acid quench techniques. Similarities in mechanism between RTPR and the Escherichia coli ribonucleotide reductase suggest that both enzymes require a thiyl radical for catalysis.

Reaction of epichlorohydrin with 2'-deoxynucleosides: characterization of adducts

Epichlorohydrin (ECH) is a simple 3-carbon epoxide of industrial importance and thus has the potential for human exposure in the workplace. It has been shown to be genotoxic in several systems and is a compound capable of reacting with biological nucleophiles. This study details the products formed from the reaction of ECH with 2'-deoxynucleosides at pH 7 and 37 degrees C for 6 h. Reaction with 2'-deoxyguanosine yielded 7-(3-chloro-2-hydroxypropyl) guanine (7-CHP-Gua) resulting from alkylation at N-7 of 2'-deoxyguanosine followed by depurination. Two unusual adducts were also partially characterized which resulted from further reaction of 7-CHP-Gua with another molecule of ECH to yield 1,7-bis(3-chloro-2-hydroxypropyl)guanine (1,7-bis-CHP-Gua) which could then cyclize with the exocyclic amino group to yield 1,N2-(2-hydroxypropano)-7-(3-chloro-2-hydroxypropyl) guanine (1,N2-HP-7-CHP-Gua). Reaction with 2'-deoxyadenosine gave only one product, namely 1,N6-(2-hydroxypropano)-2'-deoxyadenosine (1,N6-HP-dAdo). The reaction of 2'-deoxythymidine with ECH also yielded one product which was identified as 3-(3-chloro-2-hydroxypropyl)-2'-deoxythymidine (3-CHP-dThd). A 3-(3-chloro-2-hydroxypropyl)-2'-deoxyuridine (3-CHP-dUrd) product was isolated from the reaction of ECH with 2'-deoxycytidine. This product most likely resulted from the deamination of an initially formed 3-(3-chloro-2-hydroxypropyl) -2'-deoxycytidine (3-CHP-dCyd), a phenomenon which we have previously reported to occur during the reaction of 2'-deoxycytidine with other aliphatic epoxides. Evidence is also presented that 3-CHP-dUrd is converted to 3-(2,3-dihydroxypropyl)-2'deoxyuridine (3-DHP-dUrd) under physiological conditions, with a half-life of 213 h. Reaction of ECH with calf thymus DNA (pH 7.0, 37 degrees C, 3 h) resulted in the formation of 7-CHP-Gua (200 nmol/mg DNA.

Modelling of purine nucleoside metabolism during mouse embryonic development: relative routes of adenosine, deoxyadenosine, and deoxyguanosine metabolism

The individual activities for adenosine kinase, deoxyadenosine kinase, adenosine deaminase, deoxyguanosine kinase, and purine nucleoside phosphorylase were determined during days 7 to 13 of mouse embryonic development. Adenosine deaminase increased 74-fold between days 7 and 9; deoxyadenosine kinase increased 5.4-fold during the same interval. Adenosine kinase, deoxyguanosine kinase, and purine nucleoside phosphorylase exhibited less than 2-fold changes in activity between days 7 and 13. Using Michaelis constants for each enzyme and the maximal velocities determined from enzyme assay, the relative routes of adenosine and deoxyadenosine metabolism via phosphorylation or deamination were modeled as a function of nucleoside concentration for days 7 through 13. For days 7 and 8, phosphorylation of adenosine is the principle route of metabolism at physiological concentrations. A switch occurred at day 9 and following where deamination is at least 5-fold greater than phosphorylation at all substrate concentrations. Deoxyadenosine phosphorylation was at most 10% of deamination at day 7 and then declined to less than 1% for days 9 to 13. Phosphorolysis was the principle route of deoxyguanosine metabolism through the 7 to 13 day period. Thus catabolism rather than phosphorylation was the principle pathway for purine deoxynucleoside metabolism during this period.

Rapid radioassay for metabolites of adenosine and deoxyadenosine in erythrocytes

A radioassay has been developed to quantify the uptake and initial metabolism of adenosine (Ado) or deoxyadenosine (dAdo) by human erythrocytes. Cell suspension and [3H]Ado are mixed at 3-s intervals with a novel dual-syringe apparatus, and uptake and metabolism of Ado is stopped by centrifuging the cells through a dibutylphthalate layer into perchloric acid. The neutralized cell extract is analyzed by two-dimensional chromatography on poly(ethyleneimine)-cellulose plates by two procedures using combinations of solvents optimised for the separation of nucleosides and nucleobases, and for nucleotides derived from the exogenous [3H]Ado.

Stereoselective activation of dibenzo[a,l]pyrene to (-)-anti (11R,12S,13S,14R)- and (+)-syn(11S,12R,13S,14R)-11,12-diol-13,14-epoxides which bind extensively to deoxyadenosine residues of DNA in the human mammary carcinoma cell line MCF-7

Dibenzo[a,l]pyrene (DB[a,l]P) is an environmental contaminant and a very potent carcinogen. DB[a,l]P exceeds the carcinogenic potency of both benzo[a]pyrene and 7,12-dimethylbenz[a]anthracene in rodent bioassays. Previous studies demonstrated that DB[a,l]P is metabolized to DB[a,l]P-11,12-diol-13,14-epoxide (DB[a,l]PDE) in the human mammary carcinoma cell line MCF-7. In the present study the major DNA adducts formed in DB[a,l]P-treated MCF-7 cells have been identified through the use of 33P-postlabeling. TLC and HPLC. DB[a,l]P is metabolically activated in MCF-7 cells to form large amounts of three major DNA adducts and smaller amounts of three other adducts. The three major DNA adducts are with deoxyadenosine: two are formed by reaction of (+)-syn-DB[a,l]PDE (11S,12R,13S,14R), the third by reaction of (-)-anti-DB[a,l]PDE (11R,12S,13S,14R). The results demonstrate that DB[a,l] is stereoselectively metabolized in MCF-7 cells to form one enantiomer of each diol epoxide diastereomer; (+)-syn-DB[a,l]PDE and (-)-anti-DB[a,l]PDE. The high extent of binding of these diol epoxides to deoxyadenosine in DNA of MCF-7 cells may help to explain the very high carcinogenic potency of DB[a,l]P and suggests that DB[a,l]P could also pose a carcinogenic threat to humans.

Nucleoside triphosphate donors for nucleoside kinases: donor properties of UTP with human deoxycytidine kinase

The reported higher efficiency of UTP, relative to ATP, as phosphate donor for deoxycytidine kinase (dCK), has been extended and found to apply to both dCyd and dAdo as acceptors. UTP as phosphate donor was shown to follow strictly Michaelis kinetics, with Km = 1 microM, in striking contrast to ATP, which exhibits marked negative cooperativity (Hill coef. = 0.7) with a several-fold higher Kmapp = 15 microM. Phosphate transfer was followed directly with use of mixtures of [gamma-32P]ATP and cold UTP as donors, or with 3H-labeled acceptors and cold donors. With equimolar concentrations of ATP and UTP (50 microM or 1 mM each), and dCyd or dAdo as acceptor, only minimal phosphate transfer occurred from ATP (3-10%). With a 6:1 ratio of ATP:UTP, hence exceeding the intracellular ratio, phosphate transfer from ATP increased, but still did not exceed 25-40% with either dCyd or dAdo as acceptor. Moreover, relative ATP transfer is dependent on the dCyd concentration. We conclude that the major intracellular phosphate donor for dCK is not ATP, but UTP. Preliminary data for human thymidine kinases (TK1 and TK2) exhibit quite different behaviour. The foregoing, together with literature data, are highly relevant to in vitro studies on the properties of the nucleoside kinases, and to the design of chemotherapeutically active nucleoside analogues.

Synthetic nucleosides and nucleotides. XXXV. Synthesis and biological evaluations of 5-fluoropyrimidine nucleosides and nucleotides of 3-deoxy-beta-D-ribofuranose and related compounds

1-O-Acetyl-2,5-di-O-p-chlorobenzoyl-3-deoxy-D-ribofuranose (1), derived from the antibiotic cordycepin was coupled with trimethylsilylated derivatives (2a-c) of N4-propionylcytosine, N4-p-toluoyl-5-fluorocytosine and 5-fluorouracil in the presence of trimethylsilyl trifluoromethanesulfonate (TMS-triflate) to give fully acylated nucleosides (3a-b and 3d, respectively). Selective removal of the N4-propionyl group of 3a by treatment with hydrazine hydrate gave 2',5'-di-O-p-chlorobenzoyl-3'-deoxycytidine (4). Deamination of 4 with sodium nitrite in trifluoroacetic acid afforded 2',5'-di-O-p-chlorobenzoyluridine (3c) in good yield. Compounds 3a-d were saponified to give free 3'-deoxycytidine (5a), 5-fluoro-3'-deoxycytidine (5b), 3'-deoxyuridine (5c), and 5-fluoro-3'-deoxyuridine (5d), respectively. These 3'-deoxyribonucleosides (5a-d) were then converted to corresponding 5'-monophosphate and further phosphorylated to the 5'-triphosphates by the phosphoroimidazolidate method. The nucleosides (5a-d) were examined for growth-inhibitory effects on mouse leukemic L5178Y cells, and their IC50 values (microgram/ml) were 1.8, 33, 6.5, and 18, respectively. On the other hand, the antiviral activities of these compounds on a rhabdovirus, infectious hematopoietic necrosis virus (IHNV), were moderate (IC50 = 100-500 micrograms/ml in CHSE-214 cells). The 5'-triphosphates showed remarkable inhibitory effects on DNA polymerase beta and DNA polymerase alpha-primase purified from testes of the cherry salmon, Oncorhynchus masou, but not on common DNA polymerase alpha from same source.

The influence of 2-chloro-2'-deoxyadenosine on metabolism of deoxyadenosine in human primary CNS lymphoma

The effects of 2-chloro-2'-deoxyadenosine (2CdA) on the activity of enzymes important for the metabolism of deoxyadenosine were studied in lysates prepared from human primary central nervous system (CNS) lymphomas and normal human lymphocytes. Strong inhibition (approximately 100%) of the phosphorylation of deoxyadenosine to its deoxynucleotide phosphate derivatives was produced in both systems in the presence of 2CdA, which was phosphorylated concomitantly to 2-chloro-2'-deoxyAMP. Interestingly, 2CdA was also found to be an inhibitor of the deamination of both deoxyadenosine (over 50%) and AMP (70%). These findings add to our understanding of the mechanisms of toxicity of this drug, especially considering that 2CdA is resistant to deamination by adenosine deaminase. These results challenge the existing theories of 2CdA toxicity, which have been limited to the formation of phosphate derivatives of 2CdA. The present in vitro studies have demonstrated that 2CdA also inhibits both phosphorylation and deamination of deoxyadenosine (dAdo), suggesting that its mechanism of toxicity includes a block in dAdo metabolic pathways. This has important implications for the perturbation of cell methylation, a functionality associated with, for example, apoptosis.

Large scale synthesis of p-benzoquinone-2'-deoxycytidine and p-benzoquinone-2'-deoxyadenosine adducts and their site-specific incorporation into DNA oligonucleotides

Benzene is a carcinogen in rodents and a cause of bone marrow toxicity and leukemia in humans. p-Benzoquinone (p-BQ) is one of the stable metabolites of benzene, as well as of a number of drugs and other chemicals. 2'-Deoxycytidine (dC) and 2'-deoxyadenosine (dA) were allowed to react with p-BQ in aqueous solution at pH 7.4 and 4.5. The yields were considerably higher at pH 4.5 than at pH 7.4, as indicated by HPLC analysis. The desired products were isolated by column chromatography on silica gel or cellulose. Identification was done by FAB-MS, 1H NMR, and UV spectroscopy. The reaction of p-BQ with dC and dA at pH 4.5 produced the exocyclic compounds 3-hydroxy-1,N4-benzetheno-2'-deoxycytidine (p-BQ-dC), and 9-hydroxy-1,N6-benzetheno-2'-deoxyadenosine (p-BQ-dA), respectively, in a large scale and high yield. These adducts have been previously made in a microgram scale as the 3'-phosphate for 32P-postlabeling studies of their incidence in DNA. The p-BQ-dC and p-BQ-dA adducts have, in addition to the two hydroxyl groups of deoxyribose, one newly formed hydroxyl group at the C-3 or C-9 of the exocyclic base of each product respectively. Incorporation of these adducts into oligonucleotides as the phosphoramidite requires the protection of all three hydroxyl groups in these compounds. The exocyclic hydroxyl on the base has been successfully protected by acylation after protecting the 5'- and the 3'-hydroxyl groups of the sugar moiety with a 4,4'-dimethoxytrityl group and a cyanoethyl N,N-diisopropylphosphoramidite group, respectively. For the first time, to our knowledge, the fully protected phosphoramidites of p-BQ-dC and p-BQ-dA were prepared and incorporated site-specifically into a series of oligonucleotides. The coupling efficiency was very high (> 98%). However, deprotection of the DNA oligomers with ammonia produced only 50% of the desired oligomers containing the adduct. In contrast, when 10% of 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) in methanol at room temperature was used, only the desired oligomers were detected by HPLC. Thus, by deprotecting the oligomers with methoxide ions (DBU/methanol) and avoiding the use of ammonia, a high yield of modified DNA was obtained. After purification of these oligomers by HPLC, they were hydrolyzed enzymatically and analyzed by HPLC, which confirmed the base composition and the incorporation of the adducts. The mass spectroscopic analysis of the DNA oligomers was confirmed by electrospray MS. These oligomers are now under investigation for their biochemical properties.

Substrate specificity of Ty1 integrase

Integration of the Saccharomyces cerevisiae retrotransposon Ty1 requires the element-encoded integrase (IN) protein, which is a component of cytoplasmic virus-like particles (VLPs). Using purified recombinant Ty1 IN and an oligonucleotide integration assay based on Ty1 long terminal repeat sequences, we have compared IN activity on substrates having either wild-type or altered donor ends. IN showed a marked preference for blunt-end substrates terminating in an A:T pair over substrates ending in a G:C pair or a 3' dideoxyadenosine. VLP activity on representative substrates also showed preference for donor strands which have an adenosine terminus. Staggered-end substrates showed little activity when nucleotides were removed from the end of the wild-type donor strand, but removal of one nucleotide from the complementary strand did not significantly diminish activity. Removal of additional nucleotides from the complementary strand reduced activity to minimal detection levels. These results suggest that the sequence specificity of Ty1 IN is not stringent in vitro. The absence of Ty1 IN-mediated 3' dinucleotide cleavage, a characteristic of retroviral integrases, was demonstrated by using selected substrates. In addition to the forward reaction, both recombinant IN and VLP-associated IN carry out the reverse disintegration reaction with long terminal repeat-based dumbbell substrates. Disintegration activity exhibits sequence preferences similar to those observed for the forward reaction.

Nucleoside 2-deoxyribosyltransferase. Pre-steady-state kinetic analysis of native enzyme and mutant enzyme with an alanyl residue replacing Glu-98

Nucleoside 2-deoxyribosyltransferase catalyzes cleavage of a 2'-deoxyribosylnucleoside (A) to a nucleobase (P) with deoxyribosylation of the enzyme. Substrates quenched the intrinsic fluorescence of native enzyme (E) and a catalytically inactive mutant enzyme (E98A enzyme). The time courses of these reactions were analyzed in terms of the following scheme where EX is the 2-deoxyribosyl ester of Glu-98. [formula: see text] The initial complexes between E and dAdo, dGuo, dIno, and dCyd or those between EX and the corresponding nucleobases were formed in a rapid equilibrium step. Native enzyme and E98A enzyme bound 2'-deoxyribosylnucleosides with similar affinities (k-1/k1). From a comparison of the time-dependent fluorescence changes associated with the reaction of native enzyme or E98A enzyme with these substrate, the kinetic step for 2-deoxyribosylation of Glu-98 was identified (k2 and k-2). dThd and dUrd quenched the fluorescence of native enzyme in a biphasic process. The late phase of this reaction was associated with 2-deoxyribosylation of Glu-98. The pre-steady-state kinetic constants calculated from fluorescence quenching data for dAdo and Cyt were consistent with the experimental values for the steady-state kinetic coefficients and the equilibrium constant of the reaction.

Detection of deoxyadenosine-4-aminobiphenyl adduct in DNA of human uroepithelial cells treated with N-hydroxy-4-aminobiphenyl following nuclease P1 enrichment and 32P-postlabeling analysis

To characterize the DNA adducts in human uroepithelial cells (HUC) exposed to 4-aminobiphenyl and its proximate N-hydroxy metabolites, we used 32P-postlabeling analyses following butanol extraction of the DNA hydrolysates. Using this method, we identified N-(deoxyguanosin-3',5'-bisphospho-8-yl)-4-aminobiphenyl (pdGp-ABP) as a major adduct and N-(deoxyadenosin-3',5'-bisphospho-8-yl)-4-aminobiphenyl (pdAp-ABP) as a minor adduct in an immortalized non-tumorigenic cell line of HUC following exposure to N-hydroxy-4-aminobiphenyl (N-OH-ABP). Towards characterization of pdAp-ABP, we postlabeled the synthetic N-(deoxyadenosin-3'-phospho-8-yl)-4-aminobiphenyl (dAp-ABP) adduct to generate pdAp-ABP and determined its chromatographic (TLC and HPLC) properties and sensitivity to nuclease P1 digestion. In contrast to pdGp-ABP, which was cleaved to the corresponding 5'-monophosphate by nuclease P1, the pdAp-ABP adduct was unaffected when incubated with nuclease P1 under similar conditions. To test whether nuclease P1 digestion could be adopted for enrichment of the dAp-ABP adduct in HUC samples, postlabeling analyses were carried out after butanol extraction following nuclease P1 digestion of the DNA hydrolysate. Under these conditions, the pdAp-ABP adduct was detected in DNA from HUC E7 cells treated with N-OH-ABP and in calf thymus DNA reacted with N-OH-ABP under acidic (pH 5.0) conditions. These data indicate that pdGp-ABP and pdAp-ABP adducts are generated in HUC E7 on treatment with N-OH-ABP and that nuclease P1 enrichment may provide a method for qualitative and quantitative analyses of the pdAp-ABP adduct in DNA.

In vitro RNA editing-like activity in a mitochondrial extract from Leishmania tarentolae

A mitochondrial extract from Leishmania tarentolae directs the incorporation of uridylate (U) residues within the pre-edited domain of synthetic cytochrome b (CYb) and NADH dehydrogenase subunit 7 mRNA. This has several characteristics of an in vitro RNA editing activity, but no direct evidence for involvement of guide RNAs was obtained. Inhibition by micrococcal nuclease suggests a requirement for some type of endogenous RNA. The limitation of internal U-incorporation to the pre-edited region in the CYb mRNA and the inhibition by deletion or substitution of both mRNA anchor sequences for CYb gRNA-I and -II could be consistent either with a gRNA-mediated process or a secondary structure-mediated process. A low level of incorporation of [alpha-32P]CTP occurs at the same sites as UTP. Internal U-incorporation activity is selectively inhibited by heterologous RNAs, suggesting an involvement of low affinity RNA-binding proteins which can be competed by the added RNA.

Mimosine inhibits viral DNA synthesis through ribonucleotide reductase

The plant amino acid mimosine (beta-N(3-hydroxy-4-pyridone)-alpha-amino propionic acid) is toxic for animals and reversibly inhibits the replication of animal cells in vitro. We have found that mimosine inhibits the DNA synthesis of a variety of DNA viruses, including adenovirus, parvovirus, and papovavirus and the replication of vaccinia and herpes simplex virus 1 in cells in culture. However, mimosine did not inhibit the replication of SV40 DNA in a cell-free system. Because mimosine inhibition of viral DNA synthesis was reversible by iron in the medium, we hypothesized that mimosine was inhibiting ribonucleotide reductase through its capacity to chelate the iron required in the R2 subunit of this enzyme. In support of this hypothesis, we found that mimosine-treated cells had a reduction in the pools of dGTP and dATP and that mimosine inhibited ribonucleotide reductase in vitro in an iron-dependent manner.

Methylphosphonodiester substitution near the conserved CA dinucleotide in the HIV LTR alters both extent of 3'-processing and choice of nucleophile by HIV-1 integrase

We present evidence suggesting that the 3'-processing activity of HIV-1 integrase is dramatically affected by electrostatic and/or steric perturbations 3' to the conserved CA dinucleotide. When the phosphodiester bond 3' to the scissile phosphodiester is replaced by a methylphosphonodiester linkage, 3'-processing decreases by two orders of magnitude. This block of cleavage can be somewhat overcome by increasing the pH of the reaction. Labeling of the substrates at the 3'-end revealed blockage of water and glycerol, but stimulation of the viral DNA 3'-hydroxyl, acting as the nucleophile with the methylphosphonodiester substrate. Interestingly, a circular trinucleotide was formed using the phosphodiester and methylphosphonodiester substrates when the terminal nucleotide was 3'-deoxyadenosine but not 2'-deoxyadenosine. Mutagenesis of the enzyme active site has previously been shown to alter the choice of nucleophile in the 3'-processing reaction. Taken together, the results in this study suggest that 'mutagenesis' of the DNA backbone can also alter the choice of nucleophile.

(E)-5',6'-didehydro-6'-deoxy-6'-fluorohomoadenosine: a substrate that measures the hydrolytic activity of S-adenosylhomocysteine hydrolase

(E)-5',6'-Didehydro-6'-deoxy-6'-fluorohomoadenosine (EDDFHA), which is a poor substrate for the oxidative activity of S-adenosyl-L-homocysteine (AdoHcy) hydrolase and thus a poor mechanism-based inhibitor was shown to be a good substrate for the hydrolytic activity of this enzyme. Incubation of EDDFHA with AdoHcy hydrolase (NAD+ form) produces a large molar excess of hydrolytic products [e.g., fluoride ion, adenine (Ade) derived from chemical degradation of homoadenosine 6'-carboxaldehyde (HACA), and 6'-deoxy-6'-fluoro-5'-hydroxyhomoadenosine (DFHHA)] accompanied by a slow irreversible inactivation of the enzyme. The enzyme inactivation was shown to be time-dependent, biphasic, and concomitant with the reduction of the enzyme-bound NAD+ (E.NAD+) to E-NADH. The reaction of EDDFHA with AdoHcy hydrolase was shown to proceed by three pathways: pathway a, water attack at the 6'-position of EDDFHA and elimination of fluoride ion results in the formation of HACA, which degrades chemically to form Ade; pathway b, water attack at the 5'-position of EDDFHA results in the formation of DFHHA; and pathway c, oxidation of EDDFHA results in formation of the NADH form of the enzyme (inactive) and 3'keto-EDDFHA, which could react with water at either the C5' or C6' positions. The partition ratios among the three pathways were determined to be k3':k6':k5' = 1:29:79 with one lethal event (enzyme inactivation) occurring every 108 nonlethal turnovers.(ABSTRACT TRUNCATED AT 250 WORDS)

Translesional synthesis on DNA templates containing site-specifically placed deoxyadenosine and deoxyguanosine adducts formed by the plant carcinogen aristolochic acid

Synthetic oligonucleotides (18-mers) containing either a single deoxyadenosine residue or a single deoxyguanosine residue were treated with aristolochic acid I (AAI) or aristolochic acid II (AAII), the main components of the plant carcinogen aristolochic acid (AA). These reactions resulted in the formation of site-specifically adducted oligonucleotides containing the two known AAI-DNA adducts (dA-AAI, dG-AAI) or the two known AAII-DNA adducts (dA-AAII, dG-AAII) at position 15 from the 3' end. Using HPLC chromatography, the oligonucleotides were purified and subsequently shown to contain the adducts of interest by 32P-postlabelling. The adducted oligonucleotides were used as templates in primer (11-mer) extension reactions catalysed by modified bacteriophage T7 DNA polymerase (Sequenase). Regardless of the type of DNA adduct examined, DNA synthesis was blocked predominantly (80-90%) at the nucleotide 3' to each adduct, although primer extension to the full length of the template was noted with unmodified control templates. However, 15 nucleotide products, indicating blocking of DNA synthesis after incorporation of a nucleotide opposite the adduct and translesional synthesis products were formed in all cases in different amounts, depending on the adduct structure. When a 14-mer primer together with high dNTP concentrations was used to examine nucleotide incorporation directly across from the four different purine adducts we found that the deoxyadenosine adducts (dA-AAI and dA-AAII) allowed incorporation of dAMP and dTMP equally well, whereas the deoxyguanosine adducts (dG-AAI and dG-AAII) allowed preferential incorporation of dCMP. Molecular dynamic simulations showed that the aristolactam moiety of all adducts exhibit a strong stacking, with the adenine residue at the 3' end of the 14-mer primer. These studies demonstrate that all AA purine adducts provide severe blocks to DNA replication and that the guanine adducts may not be very efficient mutagenic lesions. In contrast, the translesional bypass past adenine adducts of the aristolochic acids suggests a mutagenic potential resulting from dAMP incorporation by polymerase. AT-->TA transversion mutations would be the mutagenic consequences of AA adenine adducts, which are consistent with the activating mutations of c-ras genes found in AA-induced tumours of rodents.

Alpha-deoxyadenosine, a major anoxic radiolysis product of adenine in DNA, is a substrate for Escherichia coli endonuclease IV

Oligonucleotides containing a unique alpha-deoxyadenosine or tetrahydrofuran (a model abasic site) were synthesized using phosphoramidite chemistry. Repair enzymes from Escherichia coli, including endonucleases III, IV, and VIII, exonuclease III, formamidopyrimidine N-glycosylase, and deoxyinosine 3'-endonuclease, as well as UV dimer N-glycosylases from T4 (den V) and Micrococcus luteus, were examined for their ability to recognize alpha-deoxyadenosine and tetrahydrofuran. In agreement with prior studies, a tetrahydrofuran-containing oligonucleotide was a substrate for endonuclease IV and exonuclease III, but not for the other repair enzymes. However, an oligonucleotide containing alpha-deoxyadenine was a substrate only for endonuclease IV. Competitive inhibition studies with both substrates confirmed that the activity recognizing alpha-deoxyadenine was endonuclease IV and not a possible contaminant in the endonuclease IV preparation. Using E. coli extracts, the activity that recognized alpha-deoxyadenine was dependent on nfo, the structural gene of endonuclease IV, further substantiating that endonuclease IV is the enzyme that recognized alpha-deoxyadenine. Kinetic measurements indicated that alpha-deoxyadenosine was as good a substrate for endonuclease IV as tetrahydrofuran; the Km and Vmax values for both substrates were similar. Using substrates that were labeled at either the 3'- or 5'-terminus, endonuclease IV was shown to hydrolyze the phosphodiester bond 5' to either alpha-deoxyadenosine or tetrahydrofuran, leaving the lesion, alpha-deoxyadenosine or tetrahydrofuran, on the 5'-terminus of the nicked site. The ability of endonuclease IV to recognize alpha-deoxyadenosine suggests that endonuclease IV is able to recognize a new class of DNA base lesions that is not recognized by other DNA N-glycosylases and AP endonucleases.

Effects of mutational loss of nucleoside kinases on deoxyadenosine 5'-phosphate/deoxyadenosine substrate cycle in cultured CEM and V79 cells

The functions of a deoxynucleoside kinase and a deoxynucleotidase can give rise to substrate cycles in which the two enzymes catalyze in opposite directions the irreversible interconversion of a deoxynucleoside 5'-monophosphate (dNMP) and its deoxynucleoside. Earlier evidence showed that pyrimidine dNMP cycles occur in cultured cells and participate in the regulation of the size of dNMP pools there by affecting the transport of deoxyribonucleosides across the cell membrane. Here, we apply an isotope flow method using labeled adenine as precursor of dAMP and DNA to quantify deoxyadenosine excretion as a measure of the catabolic activity of a putative dAMP/deoxyadenosine cycle. A comparison of human CEM lymphoblasts and hamster V79 fibroblasts, including mutant cells lacking kinases for the phosphorylation of deoxyadenosine, shows a much lower deoxyadenosine excretion in CEM cells (0.05% of dATP synthesized by reduction of ADP) as compared with V79 cells (4% of dATP). Mutational loss of deoxycytidine kinase increases these values to 0.3% in CEM cells and to 10% in V79 cells. This strongly suggests the presence of a dAMP/deoxyadenosine cycle in both CEM and V79 cells. Additional loss of adenosine kinase only marginally affects deoxyadenosine excretion in CEM cells. The small excretion of deoxyadenosine (also in the absence of both kinases) demonstrates that in CEM cells the in situ activity of the deoxynucleotidase affecting the dAMP/deoxyadenosine substrate cycle is very low and that the cycle has mainly an anabolic function there.

Adenosylmethionine-dependent synthesis of the glycyl radical in pyruvate formate-lyase by abstraction of the glycine C-2 pro-S hydrogen atom. Studies of [2H]glycine-substituted enzyme and peptides homologous to the glycine 734 site

The active form of pyruvate formate-lyase (PFL) from Escherichia coli contains a glycyl radical in position 734 of the polypeptide chain which is produced post-translationally by pyruvate formate-lyase-activating enzyme (PFL activase) using S-adenosylmethionine (AdoMet) and dihydroflavodoxin as co-substrates (Wagner, A.F. V., Frey, M., Neugebauer, F.A., Schäfer, W., and Knappe, J. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 996-1000). Studying radical synthesis with [2-2H]glycine-labeled PFL, we have now found stoichiometric incorporation of a 2H atom into the 5'-deoxyadenosine (dAdo) co-product via mass and NMR spectroscopic analyses. Furthermore, a series of peptides homologous to the Gly-734 site of PFL have been synthesized for analyzing recognition determinants of PFL activase. Peptides that proved active as substrates (monitored by [14C]dAdo formation from [14C]AdoMet) were also competitive inhibitors of PFL conversion to the radical form. In the sequence of the standard peptide Arg-Val-Ser-Gly-Tyr-Ala-Val, which corresponds to amino acid residues 731-737 of PFL, the Gly residue was replaceable by D-Ala (actually displaying enhanced efficiency), whereas a normal Ala totally abolished the interaction with PFL activase. Our results show that the radical in pyruvate formatelyase is produced by stereospecific abstraction of the pro-S hydrogen of glycine 734 by the 5'-dAdo radical generated in the active center of PFL activase. Gly-734 is probably located in a beta-turn segment of the protein.

Activation of apoptosis in early mouse embryos by 2'-deoxyadenosine exposure

Adenosine deaminase (ADA) catalyzes the irreversible hydrolytic deamination of adenosine and deoxyadenosine to nontoxic derivatives. The importance of this reaction in the female reproductive tract of mice is suggested by pronounced utero-placental expression of ADA, and by embryolethality of the potent ADA-inhibitor deoxycoformycin (dCF) on day 7-8 of gestation. The present study investigated the effects of dCF, adenosine, and deoxyadenosine on the mouse neurula. Morphological cell death was monitored by the acridine orange reaction (AOR), and biochemical cell death by internucleosomal DNA cleavage (IDC). A strong AOR appeared in day 7-8 embryos between 3 and 4.5 hr post-exposure to dCF in utero; there was no apparent effect on day 6 or day 9 embryos. Most embryonic tissues were responsive, although the heart and extraembryonic membranes were resistant. Up to 75% of the embryonic chromatin was degraded in a regular pattern in concert with the AOR. Immediate activation of "whole-body" apoptosis was reproduced in short-term whole embryo culture with 0.1 mM deoxyadenosine in the presence of 0.01 mM dCF. This was not activated by exposure to dCF alone nor to adenosine; however, high concentrations of adenosine completely blocked the response to deoxyadenosine, whereas niacinamide inhibited the AOR without changing IDC. The cytotoxic effect of deoxyadenosine was correlated with an expansion of embryonic dATP pools determined by high-performance liquid chromatography analysis. The results suggest that deoxyadenosine is the embryotoxic metabolite which accumulates in the antimesometrium of pregnant mice treated with dCF. Exposure to this metabolic toxin activates apoptosis in day 7-8 embryos through an adenosine-sensitive, NAD-dependent mechanism.

Novel splicing, missense, and deletion mutations in seven adenosine deaminase-deficient patients with late/delayed onset of combined immunodeficiency disease. Contribution of genotype to phenotype

We examined the genetic basis for adenosine deaminase (ADA) deficiency in seven patients with late/delayed onset of immunodeficiency, an underdiagnosed and relatively unstudied condition. Deoxyadenosine-mediated metabolic abnormalities were less severe than in the usual, early-onset disorder. Six patients were compound heterozygotes; 7 of 10 mutations found were novel, including one deletion (delta 1019-1020), three missense (Arg156 > His, Arg101 > Leu, Val177 > Met), and three splicing defects (IVS 5, 5'ss T+6 > A; IVS 10, 5'ss G+1 > A; IVS 10, 3'ss G-34 > A). Four of the mutations generated stop signals at codons 131, 321, 334, and 348; transcripts of all but the last, due to delta 1019-1020, were severely reduced. delta 1019-1020 (like delta 955-959, found in one patient and apparently recurrent) is at a short deletional hot spot. Arg156 > His, the product of which had detectable activity, was found in three patients whose second alleles were unlikely to yield active ADA. The oldest patient diagnosed was homozygous for a single base change in intron 10, which activates a cryptic splice acceptor, resulting in a protein with 100 extra amino acids. We speculate that this "macro ADA," as well as the Arg156 > His, Arg101 > Leu, Ser291 > Leu, and delta 1019-1020 products, may contribute to mild phenotype. Tissue-specific variation in splicing efficiency may also ameliorate disease severity in patients with splicing mutations.

Multiple magnesium ions in the ribonuclease P reaction mechanism

The ribozyme ribonuclease (RNase) P cleaves precursor transcripts to produce the mature 5'-end of tRNAs. This hydrolysis reaction has a divalent cation requirement that is primarily catalytic, rather than structural; RNase P can be considered a metalloenzyme. Kinetic analysis shows that the RNase P catalytic mechanism has a cooperative dependence upon Mg2+ concentration. At least three Mg2+ ions are required for optimal activity, suggesting a multiple metal ion mechanism. The 2'-OH at the site of substrate cleavage may act as a ligand for a catalytically important Mg2+: deoxyribose substitution reduces the apparent number of Mg2+ bound from three to two and increases the apparent dissociation constant for Mg2+ from the micromolar to the millimolar range. In addition to these cation effects, the deoxyribose substitution reduces the rate of catalysis by 3400-fold; substitution with 2'-O-methyl at the cleavage site reduces the catalytic rate 10(6)-fold. If we presume no significant conformational effects of the substitutions, these results suggest that the 2'-OH serves as hydrogen-bond donor. The kinetic analysis of the catalytic mechanism is based upon the characterization of the pH dependence of the reaction. There is a hyperbolic (saturable) dependence on hydroxide concentration, with the half-maximal rate achieved at pH 8.0-8.5. The rate of the cleavage step is about 200 min-1 at pH 8.0, which is 500-fold faster than the steady-state parameter kcat.

Effect of 5'-deoxy-5'-isobutylthioadenosine on formation and release of adenosine from neonatal and adult rat ventricular myocytes

1. Studies in rat polymorphonuclear leucocytes have suggested that 5'-deoxy-5'-isobutylthioadenosine (IBTA), an inhibitor of the IMP-selective cytosolic 5'-nucleotidase, may be used to test its role in adenosine formation in intact cells. We investigated adenosine formation in neonatal and adult rat cardiomyocytes. 2. 2-Deoxyglucose (30 mM) with oligomycin (2 micrograms/ml) induced a 90-100% fall in ATP concentration in 10 min in neonatal and 60 min in adult heart cells. Adenosine accumulation was substantially increased, accounting for 13% of the fall in ATP concentration in neonatal cells and 56% in adult cells. 3. Anti-(rat liver ecto-5'-nucleotidase) serum did not inhibit adenosine accumulation. Furthermore, dipyridamole (10 microM), a nucleoside-transport blocker, inhibited by 80% the appearance of the newly formed adenosine in the medium, showing that adenosine is produced intracellularly by both adult and neonatal-rat myocytes in response to inhibition of oxidative metabolism. 4. IBTA (3 mM) inhibited by 80% the appearance of adenosine in the medium, but did not inhibit total adenosine accumulation by neonatal-rat myocytes and only modestly inhibited total adenosine accumulation by adult myocytes. 5. IBTA, like dipyridamole, inhibited incorporation of extracellular adenosine (10 microM) into neonatal and adult ventricular myocyte nucleotides by 60-70%. Transport of IBTA (100 microM) into the cells did not appear to be inhibited by dipyridamole (30 microM). 6. We conclude that IBTA acted primarily to inhibit adenosine release from myocytes. The small effect on adenosine formation rates implies that the IMP-selective cytosolic 5'-nucleotidase plays a minor role in this tissue.

Characterization of adenosine deaminase (ADA)-negative B-lymphoblastoid cells cocultured with ADA-positive fibroblasts

A cell line, BAD05, derived from B lymphocytes of an adenosine deaminase (ADA; EC 3,5,4,4)-deficient patient could not proliferate in a serum-free medium containing 100 mumol/l deoxyadenosine. When BAD05 was cultured with ADA-positive fibroblasts, the proliferation of BAD05 was improved. BAD05 cell density increased when the initially mixed ratio of fibroblasts/BAD05 was 1/10 or higher, but decreased when the ratio was 1/20 or lower. Deoxyadenosine concentrations in the medium and ATP and deoxyATP (dATP) levels in the BAD05 were measured after 4 hours of coculture at initial BAD05 cell densities of 1 x 10(5) and 1 x 10(6) cells/ml. Deoxyadenosine concentrations in the medium decreased as the density of fibroblasts increased. The dATP level decreased as the mixed ratio rose. The ratio of fibroblasts/BAD05 rather than the cell density of fibroblasts had a larger effect on the dATP levels in BAD05. Under our experimental conditions, ADA-negative cells proliferated well when the ratio of ADA-positive cells/ADA-negative cells was over 1/10.

Deoxyribonucleoside triphosphate pools and thymidine chemosensitization in human T-cell leukemia

Thymidine kills cells by depleting dCTP stores. The present experiments tested whether deoxycytidine, by replenishing dCTP pools, could prevent thymidine cytotoxicity and thymidine's enhancement of carboplatin killing in two human T-cell acute leukemia cell lines. MOLT3 and JM cells were exposed to combinations of thymidine, deoxycytidine, and carboplatin and then assessed for survival, the magnitude of thymidine-carboplatin chemosensitization, and changes in deoxyribonucleoside triphosphate pools. For both cell lines, deoxycytidine (up to 144.5 micrograms/ml x 24 h) completely restored dCTP pools but only partially protected against thymidine cytotoxicity (100-1000 micrograms/ml x 24 h) and thymidine-carboplatin sensitization (up to 60 micrograms carboplatin/ml during the last hour of thymidine). This contrasts with complete protection in prior studies using other cell types. Thymidine alone markedly increased dTTP and dGTP pools and decreased dCTP; dATP pools underwent a sharp decline which has not been observed before in any cell line. In subsequent studies 0.0336-137.3 micrograms deoxyadenosine/ml partially prevented cytotoxicity and carboplatin sensitization by 300 micrograms thymidine/ml. Together, deoxycytidine and deoxyadenosine completely prevented thymidine-carboplatin sensitization even though dATP and dCTP pools were not entirely returned to normal. These findings are discussed in regard to the unusual sensitivity of T-cell malignancies to thymidine toxicity, mechanisms of cytotoxicity and chemosensitization by thymidine, and the possibility of thymidine selectively sensitizing T-cell malignancies to killing by alkylating agents.

Transition-state discrimination by adenosine deaminase from Aspergillus oryzae

Adenosine deaminase from Aspergillus oryzae resembles mammalian adenosine deaminases in its ability to catalyze the hydrolytic removal of many substituents from C-6, and in the chirality at C-6 of the active isomer of the transition-state-analogue inhibitor 6-hydroxymethyl-1,6-dihydropurine ribonucleoside. The 5'-OH group of adenosine has been found to contribute a factor of 5.10(4) to transition-state stabilization by calf intestinal adenosine deaminase, and crystallographic observations suggest that a zinc-histidine 'bridge' is formed between the 6-OH and the 5'-OH groups of the substrate in the transition state for its deamination. The present paper describes experiments indicating that this bridge is not present during the action of adenosine deaminase from Aspergillus oryzae. We find (1), that the fungal enzyme catalyzes deamination of adenosine and 5'-deoxyadenosine with kcat/Km values that are almost identical; (2), that the Ki value of the transition-state-analogue inhibitor 2'-deoxycoformycin is much higher for the fungal enzyme (2.7.10(-9) M) than for the mammalian enzyme (2.10(-12) M) and (3), that this difference in binding affinities arises mainly from a difference in rates of enzyme-inhibitor association. Thus, the onset of inhibition was markedly slower for the fungal enzyme (kon = 1.3.10(4) M-1 s-1) than for the calf intestinal enzyme (kon = 2.6.10(6) M-1 s-1). Effects of chelating agents and divalent cations suggest that the fungal enzyme, like other deaminases for adenosine and cytidine, contains essential zinc.

Different relationships between cellular adenosine or 3'-deoxyadenosine phosphorylation and cellular adenine ribonucleotide catabolism may be obtained

Treatment of BALB/c-3T3 mouse fibroblasts with 3'-deoxyadenosine led to a rapid accumulation of 3'-deoxyadenosine phosphates and the kinetics of this process has been determined. Concomitant with accumulation of these compounds, the adenine ribonucleotide pool was reduced. The kinetics of the two processes suggested that they were tightly coupled. The inhibitory effect of relatively high concentrations of coformycin indicated that IMP was an intermediate in the catabolic pathway. Similar experiments with Ehrlich ascites tumor cells were performed in Ringer-Hepes solution at pH 6.5 or 7.5 and with varying concentrations of orthophosphate. The experiments were performed with cells where ATP was [3H]-labeled. This allowed the determination of the catabolism of adenine ribonucleotides to labeled nucleosides under conditions where added adenosine was phosphorylated. The results showed that at low phosphate concentration (5.8 mM) at pH 6.5 adenosine may be phosphorylated at a rate that was completely balanced to the concomitant catabolism of adenine ribonucleotides; that is, there was apparently a tight kinetic coupling between anabolism of adenosine and catabolism of adenine ribonucleotides. With 3'-deoxyadenosine a corresponding effect was obtained although the apparent coupling between phosphorylation of 3'-deoxyadenosine and catabolism of adenine ribonucleotides was not complete. When experiments were performed at the same pH but at high concentration of phosphate (45 mM) there was in contrast no coupling between the two processes; that is, ATP was present in constant amounts while 3'-deoxyadenosine phosphates accumulated at a high rate. In experiments with adenosine under these conditions there was still some although a relatively limited degree of apparent coupling between phosphorylation of adenosine and catabolism of adenine ribonucleotides. In both lines of cells used and with both adenosine and 3'-deoxyadenosine, the main products of the catabolism of adenine ribonucleotides were inosine and hypoxanthine. With 3'-deoxyadenosine there was in addition (about 20%) formation of xanthosine, suggesting that IMP dehydrogenase had also been activated. These results lead to the suggestion that adenosine (or 3'-deoxyadenosine) may be phosphorylated in two ways. 1) Phosphorylation may depend on an adenosine kinase unrelated to catabolism of adenine ribonucleotides. 2) Phosphorylation may be tightly coupled to catabolism of adenine ribonucleotides. A nucleoside phosphotransferase may catalyze the transfer of a phosphoryl group from IMP to adenosine (or 3'-deoxyadenosine) to form AMP (or 3'-dAMP) and inosine, a process that may be tightly coupled to an AMP deaminase reaction. The IMP formed in the latter reaction may not be released but transferred to the phosphotransferase.(ABSTRACT TRUNCATED AT 400 WORDS)

Activation of the anaerobic ribonucleotide reductase from Escherichia coli by S-adenosylmethionine

The anaerobic ribonucleoside triphosphate reductase from Escherichia coli reduces CTP to dCTP in the presence of a second protein, named dA1, and a Chelex-treated boiled extract of the bacteria, named RT. The reaction requires S-adenosylmethionine, NADPH, dithiothreitol, ATP, and Mg2+ and K+ ions. It occurs only under anaerobic conditions. We now show that the overall reaction occurs in two steps. The first is an activation of the reductase by dA1 and RT and requires S-adenosylmethionine, NADPH, dithiothreitol, and possibly K+ ions. In the second step, the activated reductase reduces CTP to dCTP with ATP acting as an allosteric effector. During activation, S-adenosylmethionine is cleaved reductively to methionine + 5'-deoxyadenosine. This step is inhibited strongly by S-adenosylhomocysteine and various chelators. The activation of the anaerobic reductase shows a considerable similarity to that of pyruvate formate-lyase (Knappe, J., Neugebauer, F. A., Blaschkowski, H. P., and Gänzler, M. (1984) Proc. Natl. Acad. Sci. U.S.A. 81, 1332-1335).