Formation of the 1, N2-Glyoxal Adduct of Deoxyguanosine by

Automated synthesis, characterization, and structural analysis of oligonucleotide C-3′-radical precursors. Georges Lahoud , Jesse Fancher , Sanda Gr...
3 downloads 0 Views 89KB Size
Download PDF
Chem. Res. Toxicol. 2001, 14, 1247-1253

1247

Formation of the 1,N2-Glyoxal Adduct of Deoxyguanosine by Phosphoglycolaldehyde, a Product of 3′-Deoxyribose Oxidation in DNA Mohamad Awada and Peter C. Dedon* Division of Bioengineering and Environmental Health, 56-787, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, Massachusetts 02139 Received April 10, 2001

Oxidation of deoxyribose in DNA results in the formation of a variety of electrophilic products that have the potential to react with nucleobases to form adducts. We now report that 2-phosphoglycolaldehyde, a model for the 3′-phosphoglycolaldehyde residue generated by 3′oxidation of deoxyribose in DNA, reacts with dG and DNA to form the diastereomeric 1,N2glyoxal adducts of dG, 3-(2-deoxy-β-D-erythro-pentofuransyl)-6,7-dihydro-6,7-dihydroxyimidazo[1,2-a]purine-9(3H)-one. The glyoxal adducts were the predominant species formed under biological conditions (pH 7.4 and 37 °C), with several minor fluorescent adducts, including 1,N6-ethenoadenine. The adducts were fully characterized by HPLC, mass spectrometry, and UV and NMR spectroscopy. The reaction of 2-phosphoglycolaldehyde with dG occurred with a rate constant of 10-6 M-1 s-1 compared to the rate constants of 0.08 and ∼10-9 M-1 s-1 for the reactions of glyoxal and glycolaldehyde with dG, respectively. The kinetic results rule out contamination of 2-phosphoglycolaldehyde preparations with glyoxal as the basis for our observations. The rate constant for the formation of glyoxal from 2-phosphoglycolaldehyde (10-8 s-1) is consistent with glyoxal generation being the rate-limiting step in the formation of dG adducts in reactions with 2-phosphoglycolaldehyde. Mechanistic studies were also undertaken to define the basis for the different oxidation states of glyoxal and 2-phosphoglycolaldehyde. Although 2-phosphoglycolaldehyde produced a weak ESR signal consistent with generation of hydroxyl radicals and it caused DNA strand breaks at high concentrations, the formation of the glyoxal adducts of dG was insensitive to radical quenchers (e.g., sorbitol) and independent of molecular oxygen. In contrast, the formation of glyoxal-dG adducts with glycolaldehyde was dependent on molecular oxygen and quenched by sorbitol, and the glycolaldehyde-glyoxal rearrangement produced a strong ESR signal characteristic of alkyl radicals. These observations are consistent with a model in which glyoxal is generated from 2-phosphoglycolaldehyde by a nonradical, oxygen-independent mechanism that is currently under investigation. Our results provide a mechanistic basis for the observation by Murata-Kamiya et al. [(1995) Carcinogenesis 16, 2251-2253] that oxidation of DNA with the Fe(II)-EDTA complex results in the formation of the glyoxal adducts of dG.

Introduction The sugar-phosphate backbone of DNA is highly vulnerable to oxidation by a variety of radical-mediated species (1). The initial product of such an oxidation event is a carbon-based sugar radical that leads to degradation of the deoxyribose and the generation of a variety of electrophilic products. One of these products is the phosphoglycolaldehyde residue that forms by oxidation of the 3′-position of deoxyribose. The presence of this residue was demonstrated by Barton and co-workers using rhodium(III) complexes (2). As shown in Scheme 1, the formation of the 3′-phosphoglycolaldehyde is accompanied by base propenoic acid, which is analogous to the 3′-phosphoglycolate and base propenal products of 4′-oxidation (1). We have previously demonstrated that the 4′-oxidation product, base propenal, and the 5′oxidation product, butenedialdehyde, react with dG and dC, respectively, to form potentially mutagenic adducts * To whom correspondence should be addressed. Telephone: (617) 253-8017. Fax: (617) 258-0225. E-mail: [email protected].

Scheme 1. 3′-Oxidation of Deoxyribose Generates the Phosphoglycolaldehyde Residue

(3, 4). We now report that 2-phosphoglycolaldehyde reacts with dG to form the 1,N2-glyoxal adducts (3a,b in Scheme 2). Glyoxal (2) is a ubiquitous oxidation product of lipids and carbohydrates, and it represents one of the advanced glycation end products involved in the pathology of diabetes and aging (5, 6). This dicarbonyl reacts with guanine bases in DNA to form the two diastereomers of

10.1021/tx0155092 CCC: $20.00 © 2001 American Chemical Society Published on Web 08/07/2001

1248

Chem. Res. Toxicol., Vol. 14, No. 9, 2001

Scheme 2. Reaction of 2-Phosphoglycolaldehyde with dG To Form Two Diastereomeric Adducts

3-(2-deoxy-β-D-erythro-pentofuransyl)-6,7-dihydro-6,7-dihydroxyimidazo[1,2-a]purine-9(3H)-one (3a,b in Scheme 2), and it has been shown to be mutagenic in bacterial and mammalian cells (7-9). The repertoire of glyoxal sources has recently been extended by Murata-Kamiya et al. with the demonstration that glyoxal and its 1,N2adducts of dG are generated in DNA exposed to the oxygen radical-forming Fe(II)-EDTA complex (10). The results of our studies suggest that glyoxal is generated from 3′-phosphoglycolaldehyde residues in oxidized DNA by a nonradical, oxygen-independent pathway.

Materials and Methods Materials. Deoxyribonucleosides, calf thymus DNA, diethylenetriaminepentaacetic acid (DETAPAC), 5,5-dimethyl-1-pyrroline N-oxide (DMPO), o-phenylenediamine, 2,3-dimethylquinoxaline, and all other reagents were purchased from Sigma Chemical Co. (St. Louis, MO). HPLC solvents were purchased from EM Sciences (Cincinnati, OH), and NMR solvents were from Aldrich Chemical Co. (St. Louis, MO) and Cambridge Isotope Laboratories (Andover, MA). Plasmid Giga purification kits were purchased from Qiagen (Valencia, CA). Chelex cationexchange resin was purchased from Bio-Rad (Hercules, CA). Instrumentation. Mass Spectra were recorded on HewlettPackard 5989B electrospray ionization (ESI) and Finnigan TSQ 7000 electron ionization mass spectrometers. 1H and 13C NMR spectra were recorded on 300 and 400 MHz Varian and Bruker spectrometers. ESR spectra were recorded on an EPR Bruker EMX spectrometer outfitted with 13 in. magnets, an ER 4102ST cavity, and a Klystron microwave source producing X-band (9.19.9 GHz) radiation. Products were analyzed by reversed phase (C18) HPLC using a Hewlett-Packard model 1100 HPLC system with a model 104A diode array detector. Preparation and Purification of pUC19 Plasmid DNA. Plasmid pUC19 was isolated from DH5R Escherichia coli using Qiagen kits as described elsewhere (11). The DNA pellet, washed with 80% ethanol and dried under vacuum, was dissolved in Chelex-treated potassium phosphate buffer (100 mM, pH 7.4). The DNA was exhaustively dialyzed against Chelex-treated potassium phosphate buffer containing 1 mM DETAPAC for 12 h at 4 °C to remove trace metals and then against Chelextreated potassium phosphate buffer for an additional 12 h at 4 °C. Synthesis of 2-Phosphoglycolaldehyde (1). 2-Phosphoglycolaldehyde was synthesized as described by Urata et al. (12) and Lee et al. (13). Briefly, D,L-glycerol phosphate (10 mmol) and sodium periodate (11 mmol) were dissolved in 150 mL of H2O at ambient temperature. The pH of the solution was adjusted to 6.0 with 1 M HCl, and the solution was stirred for 1.5 h. The reaction was quenched with 2 mmol of ethylene glycol to consume the unreacted periodate. The pH was readjusted to 7.0 with 1 M NaOH, and 22 mmol of BaCl2 was added to the solution. After 1 h at 0 °C, the white precipitate of barium iodate was collected by centrifugation and discarded. Ethanol (4 volumes) was added to the supernatant, and the barium salt of 2-phosphoglycolaldehyde was collected by centrifugation and dried under vacuum to remove water and ethanol. It was redissolved in water, and the barium ions were removed by

Awada and Dedon passing the solution over a column of Dowex AG50X8 (H+ form, Bio-Rad). The pH of the solution was then readjusted to 7.0 with 1 M NaOH, and the solution was stored at -80 °C. Purity was verified by NMR spectroscopy. 1H NMR (D2O): δ 5.18 (t, 1, H1), 3.88 (m, 2, H2). 13C NMR: δ 90.7 (C-1), 68.2 (C-2). Reaction of 2-Phosphoglycolaldehyde with 2′-Deoxyguanosine. 2′-Deoxyguanosine (10 mM) was reacted with 2-phosphoglycolaldehyde (100 mM) in potassium phosphate buffer (100 mM, pH 7.4, 37 °C) for 24 h. Products were collected by HPLC using a C18 column (Haisil HL, 5 µm, 250 mm × 4.6 mm) with a 0 to 20% (20 min) acetonitrile gradient in 50 mM ammonium acetate (pH 7.0). The products eluted at 13.36 and 13.48 min. UV spectroscopy: 215, 248 (λmax), 272 (sh) nm. 1H NMR (D2O): δ 7.82 (s, 1, H1′), 6.12 (t, 1, H1′), 5.78 (s, 1, H10), 5.21 (s, 1, H11), 4.44 (dt, 1, H4′), 4.01 (m, 1, H5′), 4.45-4.75 (m, 2, H3′ and H5′), 2.20-2.80 (m, 2, H2′). ESI mass spectrometry: MH+ 326. The 1,N2-dG adducts of glyoxal were prepared by mixing 100 mg of dG with 400 µL of 40% glyoxal solution in 20 mL of potassium phosphate buffer (100 mM, pH 7.4), and incubating the resulting solution at 37 °C for 2 h. The pure adduct was collected by HPLC using a C18 semipreparative column (Haisil HL, 5 µm, 250 mm × 10 mm) and isocratic elution in 5% acetonitrile in water. Reaction of 2-Phosphoglycolaldehyde with Calf Thymus DNA. 2-Phosphoglycolaldehyde (300 mg, 2 mmol) was incubated overnight with calf thymus DNA (2.1 mg, 6.5 µmol of nucleotides) in 10 mL of potassium phosphate buffer (100 mM, pH 7.4) at 37 °C. The treated DNA was passed twice over a Sephadex G-25 column to remove 2-phosphoglycolaldehyde and then depurinated in 0.1 N HCl at 90 °C for 15 min. After neutralization with ammonium hydroxide (10% in H2O), the DNA was precipitated with ethanol and the supernatant containing free purine adducts was removed. The adducts were isolated by HPLC as described earlier (5.3 min retention time). UV spectroscopy: 215, 248 (λmax), 282 (sh) nm. ESI-MS: MH+ 210. 1H NMR (D2O): δ 7.90 (s, 1, H8), 5.82 (s, 1, H10), 5.23 (s, 1, H11). Treatment of Plasmid pUC19 with 2-Phosphoglycolaldehyde. Damage reactions were carried out in 100 µL volumes containing 10 µg of plasmid pUC19 DNA and 2-phosphoglycolaldehyde (0-100 mM) in Chelex-treated potassium phosphate buffer (100 mM, pH 7.4) for 24 h at 37 °C in the presence of or absence of sorbitol (250 mM). To assess DNA strand breakage by glyoxal, pUC19 (10 µg) was incubated with 0.1 mM glyoxal in a 100 µL reaction volume for 24 h at 37 °C. Following the incubation, 5 µL of the treated plasmid DNA (500 ng) was mixed with 5 µL of 20% Ficoll 400 loading buffer containing 0.1% bromophenol blue, 100 mM EDTA, and 1% SDS, and loaded onto a 1% agarose gel (Tris-borate-EDTA buffer; 14). Plasmid topoisomers were resolved at 3 V/cm for 3 h. Subsequently, the gels were stained with ethidium bromide (0.5 µg/mL) for 30 min, and the DNA was visualized by UV transillumination. The fluorescent DNA bands were quantified in digital images recorded with a CCD camera. ESR Studies. Reaction mixtures contained 50 mM glycolaldehyde or 1 M 2-phosphoglycolaldehyde with 100 mM DMPO in potassium phosphate buffer (100 mM, pH 7.4). Following incubation for 10 min at 37 °C, ESR spectra were recorded on an EPR Bruker EMX spectrometer operating at 9.510 GHz and ambient temperature. The ESR spectrometer settings were as follows: field set, 3390 G; field scan, 100 G; modulation frequency, 100 kHz; modulation amplitude, 2.0 G; time constant, 1 s; microwave power, 20 mW; and receiver gain, 30 dB. Anaerobic Reactions. For reactions under anaerobic conditions, a freeze-pump-thaw procedure described by Gates and co-workers (15) was employed to remove dissolved gases from the reaction mixtures. Briefly, Pyrex tubing (5 mm outer diameter, 0.8 mm wall thickness) was cut into 6-8 in. lengths and sealed at one end with an oxygen-enriched propane torch. Samples were placed in the tubes and frozen in liquid nitrogen, and a vacuum was applied to the frozen samples for 15 min via

Formation of the 1,N2-Glyoxal Adduct of dG

Chem. Res. Toxicol., Vol. 14, No. 9, 2001 1249

connection to a stopcock with rubber vacuum tubing. The stopcock was then closed and the sample thawed at ambient temperature. Following five repetitions of the freeze-pumpthaw cycle, the tubing was sealed with a propane torch and the sample incubated overnight at 37 °C. The products were then analyzed as described earlier. Reaction Kinetics. The rate of glyoxal formation from 2-phosphoglycolaldehyde was defined by derivatization of glyoxal with o-phenylenediamine as described by Murata-Kamiya et al. (10). Briefly, 2-phosphoglycolaldehyde (1 M) was incubated in Chelex-treated potassium phosphate buffer (100 mM, pH 7.4) at 37 °C. At various times, 100 µL of the reaction mixture was removed and combined with 10 µL of 20 µg/µL o-phenylenediamine and 0.5 µg/µL 2,3-dimethylquinoxaline (internal standard), followed by incubation for 1.5 h at ambient temperature. The quinoxaline product was extracted with ethyl acetate, evaporated to dryness, and dissolved in 100 µL of 50% methanol in water. The glyoxal was quantified as the quinoxaline derivative of o-phenylenediamine by HPLC using a reverse phase column (Phenomenex ODS, 5 µm, 4.6 mm × 250 mm) and an isocratic solvent system consisting of 25% acetonitrile in 50 mM ammonium acetate (pH 7.0). The kinetics of formation of the glyoxal adducts of dG were determined with 2-phosphoglycolaldehyde (10-1000 mM) or glyoxal (0.75 mM) and 3-10 mM dG in Chelex-treated potassium phosphate buffer (100 mM, pH 7.4) with 1 mM uracil as an internal standard. The reaction mixture was incubated at 37 °C, and at various times, an aliquot was removed and the glyoxal adduct quantified by C18 reversed phase HPLC with a 0 to 20% (20 min) acetonitrile gradient in 50 mM ammonium acetate (pH 7.0).

Results Products of the Reaction of 2-Phosphoglycolaldehyde with dG and DNA. HPLC analysis of a 24 h reaction between dG and 2-phosphoglycolaldehyde revealed two closely eluting products as shown in Figure 1A (peaks a and b). Analysis of the product pair by positive ion electrospray ionization (ESI) mass spectrometry revealed a single protonated molecular ion at m/z 326 (Figure 1B). 1H NMR spectroscopy of the adduct pair showed the expected resonances for dG with two additional signals between 5 and 6 ppm (Figure 1C). All of these results are identical to those obtained with products formed in a reaction of glyoxal with dG (data not shown) and with the observations of Loeppky and co-workers (16). We therefore conclude that the products formed in a reaction of dG with 2-phosphoglycolaldehyde consist of the two diastereomeric 1,N2-cyclized dG derivatives 3a and 3b shown in Scheme 2. Although there are four possible diastereomeric forms of the glyoxal-dG adduct (two trans and two cis; see Scheme 2), the NMR spectrum revealed an absence of coupling between the C-H protons of the glyoxal moiety, which is consistent with a dihedral angle of ∼90° between the hydrogen atoms and thus a trans stereochemistry. This is in agreement with published results for the glyoxal-dG adducts (16). The glyoxal-dG adducts were found to be stable under acidic and neutral pH conditions but degraded at alkaline pH (data not shown). Furthermore, the presence of sorbitol, a radical scavenger, did not affect the yield or rate of formation of the glyoxal adducts of dG in reactions with 2-phosphoglycolaldehyde (data not shown). We next characterized the products formed in a reaction of 2-phosphoglycolaldehyde with DNA. Following a 24 h incubation at pH 7.0 and 37 °C, calf thymus DNA treated with 2-phosphoglycolaldehyde was subjected to

Figure 1. Chromatographic and spectral analysis of adducts formed in a reaction of 2-phosphoglycolaldehyde with dG. (A) HPLC chromatogram of 2-phosphoglycolaldehyde after incubation with dG (pH 7.4, 37 °C) for 24 h. Peaks a and b represent the glyoxal-dG adducts, and peak c represents dG. The inset shows the UV absorption spectrum of the glyoxal-dG adduct (peaks a and b). (B) ESI-MS characterization of the glyoxaldG adducts. (C) 1H NMR spectrum and structural assignments of the glyoxal-dG adducts.

acid depurination and the products were resolved by HPLC as shown in Figure 2A. A single peak appearing at 5 min in the chromatogram (peak a, Figure 2A) contained a product(s) that yielded a single protonated molecular ion at m/z 210 by positive ion ESI mass spectrometry (Figure 2B) and a 1H NMR spectrum similar to that obtained with the dG adducts with the exception of the absence of the deoxyribose resonances (Figure 2C). Again, identical products were obtained in a reaction of glyoxal with DNA (data not shown). Reaction Kinetics. One explanation for the observed formation of the glyoxal-dG adducts from 2-phosphoglycolaldehyde is contamination of 2-phosphoglycolaldehyde with glyoxal. To rule out this possibility, we compared the kinetics of formation of the dG adducts with those of glyoxal and 2-phosphoglycolaldehyde; examples of reaction profiles are shown in Figure 3. The reaction of glyoxal with dG was rapid, with a second-

1250

Chem. Res. Toxicol., Vol. 14, No. 9, 2001

Figure 2. Chromatographic and spectral analysis of adducts formed in a reaction of 2-phosphoglycolaldehyde with calf thymus DNA. (A) HPLC chromatogram of depurination products from calf thymus DNA treated with 2-phosphoglycolaldehyde (pH 7.4, 37 °C) for 24 h: peak a, glyoxal-G adducts; peak b, guanine; and peak c, adenine. The inset shows the UV absorption spectrum of the glyoxal-G adducts. (B) ESI-MS characterization of the glyoxal-G adducts. (C) 1H NMR spectrum and structural assignments of the glyoxal-G adducts.

order rate constant of 0.08 M-1 s-1 (Figure 3A). In contrast, the formation of glyoxal adducts in a reaction of 2-phosphoglycolaldehyde with dG occurred more slowly, with a second-order rate constant of 10-6 M-1 s-1 (Figure 3B). These results suggest that preparations of 2-phosphoglycolaldehyde were not contaminated with glyoxal and that a glyoxal equivalent was formed during a reaction of 2-phosphoglycolaldehyde with dG. To determine if the rate-limiting step for the overall reaction between 2-phosphoglycolaldehyde and dG was indeed the generation of glyoxal, we incubated 2-phosphoglycolaldehyde at 37 °C and monitored the formation of glyoxal at various times using o-phenylenediamine (10). As shown in the example in Figure 3C, the rate constant for glyoxal formation was 10-8 s-1. If this is assumed to be the pseudo-first-order rate constant for adduct formation, then at the 10 mM dG concentration used in the reaction of 2-phosphoglycolaldehyde with dG, the secondorder rate constant for adduct formation would be 10-6 M-1 s-1, as we observed earlier. It is thus likely that the rate-limiting step in glyoxal adduct formation in the

Awada and Dedon

Figure 3. Kinetics of glyoxal formation from 2-phosphoglycolaldehyde and of the formation of glyoxal-dG adducts in reactions of dG with glyoxal and 2-phosphoglycolaldehyde. (A) Disappearance of glyoxal in a reaction of dG (3.75 mM) with glyoxal (0.75 mM). (B) Disappearance of dG in a reaction of 2-phosphoglycolaldehyde (300 mM) with dG (10 mM) in potassium phosphate buffer (100 mM, pH 7.4) at 37 °C. (C) Formation of glyoxal from 1 M 2-phosphoglycolaldehyde in potassium phosphate buffer (100 mM, pH 7.4) at 37 °C. Glyoxal was determined as the quinoxaline derivative of o-phenylenediamine (10) with 2,3-dimethylquinoxaline as an internal standard.

reaction of dG with 2-phosphoglycolaldehyde is generation of glyoxal. Interestingly, glycolaldehyde was found to react with dG to form the glyoxal adduct with a rate constant of ∼2 × 10-9 M-1 s-1 (data not shown), too slow to account for the formation of glyoxal adducts from 2-phosphoglycolaldehyde. Mechanism of Glyoxal Formation from 2-Phosphoglycolaldehyde. The formation of the glyoxal-dG adducts from the reaction of 2-phosphoglycolaldehyde with dG requires an oxidation event. We first investigated the role of free radical chemistry in adduct formation, given the potential for monosaccharides such as glycolaldehyde to undergo autoxidation under biological conditions to generate carbon-centered free radicals and reactive oxygen species (17-19), as illustrated for 2-phosphoglycolaldehyde in Scheme 3. The generation of free radical species by 2-phosphoglycolaldehyde was studied

Formation of the 1,N2-Glyoxal Adduct of dG

Chem. Res. Toxicol., Vol. 14, No. 9, 2001 1251

Scheme 3. Hypothetical Oxygen-Dependent Mechanism of Glyoxal Formation from 2-Phosphoglycolaldehyde by Analogy to Glycolaldehyde (18)

by electron spin resonance (ESR) with the DMPO spin trap and by DNA damage analysis. As a positive control for the ESR studies, the DMPO spin trap (100 mM) was incubated in a 50 mM solution of glycolaldehyde at pH 7.4 and 37 °C. The resulting alkyl radicals produced an intense ESR spectra of DMPO-R as shown in Figure 4A, where g ) 2.00597, aN ) 22.96 G, and aH ) 16.01 G. On the other hand, the ESR spectrum of a 1 M solution of 2-phosphoglycolaldehyde with DMPO exhibited a signal of relatively low intensity (Figure 4B) assigned to the hydroxyl radical derivative of DMPO (DMPO-OH; spectrum simulated in Figure 4C), where g ) 2.00597, aN ) 15.20 G, and aH ) 14.63 G. We also used DNA strand break analysis to characterize the free radicals generated by 2-phosphoglycolaldehyde. In our assay system, radical-mediated DNA strand breaks convert supercoiled plasmid pUC19 (form I) to a nicked, relaxed form (II), and both forms are quantified following resolution by agarose gel electrophoresis (20). As shown in Figure 5, increasing concentrations of 2-phosphoglycolaldehyde caused increasing levels of strand breaks in the plasmid DNA, with 100 mM 2-phosphoglycolaldehyde damaging ∼30% of the plasmid molecules (∼17 nM nicked plasmid). The addition of sorbitol, a radical scavenger, completely inhibited the strand breaks produced by 100 mM 2-phosphoglycolaldehyde (lane 7, Figure 5), though it did not inhibit adduct formation as noted earlier, which is consistent with the ESR evidence for hydroxyl radical generation by 2-phosphoglycolaldehyde. However, 0.1 mM glyoxal, roughly the amount expected to be generated by 100 mM 2-phosphoglycolaldehyde in a 24 h incubation, did not cause strand breaks (lane 8, Figure 5). Finally, we investigated the role of molecular oxygen in the formation of glyoxal adducts in reactions of dG with 2-phosphoglycolaldehyde and glycolaldehyde. Reactions were performed under anaerobic conditions using a freeze-pump-thaw technique (15) with subsequent analysis of adduct formation by HPLC as described earlier. While the formation of glyoxal adducts from glycolaldehyde was completely inhibited under anaerobic conditions, adduct formation by 2-phosphoglycolaldehyde was unaffected (data not shown).

Discussion Glyoxal is one of several endogenously produced R-oxoaldehydes that are thought to play a role in the patho-

Figure 4. ESR spectra of reaction mixtures containing 50 mM glycolaldehyde (A) or 1 M 2-phosphoglycolaldehyde (B) in potassium phosphate buffer (100 mM, pH 7.4) at 37 °C, with 100 mM DMPO. ESR spectra were recorded as described in Materials and Methods. Panel C represents a simulated spectrum of the DMPO-OH adduct for comparison to the ESR spectrum in panel B.

Figure 5. Plasmid DNA cleavage by 2-phosphoglycolaldehyde and glyoxal. Supercoiled plasmid pUC19 was incubated with 2-phosphoglycolaldehyde or glyoxal, in the presence or absence of radical quenchers, in potassium phosphate buffer (100 mM, pH 7.4) at 37 °C: lane 1, untreated pUC19; lanes 2-6, pUC19 with 1, 3, 10, 30, and 100 mM 2-phosphoglycolaldehyde, respectively; lane 7, 100 mM 2-phosphoglycolaldehyde and 250 mM sorbitol; and lane 8, 0.1 mM glyoxal.

physiology of diabetes and aging (21). The demonstration almost four decades ago that glyoxal reacts with guanine in DNA (22) and more recent observations of the mutagenicity of glyoxal (7-9) suggest that the dicarbonyl compound may play a role in carcinogenesis. While glyoxal is best known as a product of lipid and carbohydrate oxidation (see, for example, ref 21), it has also been shown to arise from other endogenous and exogenous sources such as the metabolism of 2-hydroxyethyl nitrosamines (16, 23) and from oxidation of DNA by Fenton chemistry (10). We have now identified 2-phosphoglycolaldehyde, a model for the residue formed by 3′-oxidation of deoxyribose, as a possible source of glyoxal and the 1,N2-glyoxal-dG adducts that arise during DNA oxidation. The results of our studies reveal that, under biological conditions of temperature and pH, 2-phosphoglycolaldehyde reacts with dG to form the two diastereomeric 1,N2glyoxal-dG adducts. The results of kinetic studies shown in Figure 3 confirm that glyoxal is not simply a contaminant of preparations of 2-phosphoglycolaldehyde and further that glycolaldehyde is not the source of the

1252

Chem. Res. Toxicol., Vol. 14, No. 9, 2001

Scheme 4. Proposed Mechanism of the Phosphate-Phosphonate Rearrangement Leading to the Generation of Glyoxal from 2-Phosphoglycolaldehyde

glyoxal. The formation of glyoxal-dG adducts occurs almost five orders of magnitude more slowly with 2-phosphoglycolaldehyde than with glyoxal (10-6 vs 0.08 M-1 s-1). Instead, we have shown that glyoxal arises from 2-phosphoglycolaldehyde mainly by an oxygen-independent, nonradical pathway. The formation of glyoxal from 2-phosphoglycolaldehyde must involve an oxidation step. On the basis of studies with glycolaldehyde (18), we initially assumed that glyoxal formation from 2-phosphoglycolaldehyde would involve an oxygen-dependent, radical-mediated pathway such as that shown in Scheme 3. The ESR spectrum for the glycolaldehyde-glyoxal rearrangement confirmed the formation of an alkyl radical as expected for oxidation of its ene-diol tautomer (18). However, the ESR spectrum for 2-phosphoglycolaldehyde revealed only low levels of hydroxyl radical (Figure 4) that may have been responsible for the observed plasmid nicking caused by high concentrations of 2-phosphoglycolaldehyde (Figure 5). Furthermore, in contrast to glycolaldehyde, radical scavengers such as sorbitol and ethanol did not affect the formation of glyoxal-dG adducts with 2-phosphoglycolaldehyde, and the reaction was found to be independent of oxygen. Thus, we have unequivocally ruled out a glycolaldehyde-like, oxygen-dependent, radical-mediated mechanism for glyoxal formation from 2-phosphoglycolaldehyde. A more likely mechanism for the formation of glyoxal from 2-phosphoglycolaldehyde involves a phosphatephosphonate rearrangement such as that shown in Scheme 4. The proposal for the transformation posits that 2-phosphoglycolaldehyde exists in equilibrium with a tautomer (1a) that can cyclize to give a pentavalent phosphorus species (1b). This intermediate then forms a phosphonium enolate structure (1c) that gives rise to glyoxal (2) and phosphonic acid upon rearrangement. Literature reports support the proposed mechanism. For example, in the reverse of our proposed reaction, Kikroyannidis et al. have shown that phosphonic acid diesters add to glyoxal to form 2-phosphoglycolaldehyde derivatives via a pentavalent phosphorus or R-hydroxyphosphonate intermediate (24, 25). Our results suggest that this reaction is reversible in aqueous medium, which yields glyoxal and phosphonate. While further studies are necessary to establish the phosphate-phosphonate rearrangement as the mechanism responsible for glyoxal formation from 2-phosphoglycolaldehyde, preliminary experiments using gas chromatography and mass spectrometry have revealed the formation of phosphonate in incubations of 2-phosphoglycolaldehyde solutions at 37 °C while 31P NMR studies demonstrated the presence of a chemical

Awada and Dedon

shift consistent with a pentavalent phosphorus intermediate such as 1b (M. Awada and P. C. Dedon, manuscript in preparation). Our findings may shed light on the recent observation by Murata-Kamiya et al. (10) that glyoxal is a more abundant product of DNA oxidation by the Fe(II)-EDTA complex than is 8-oxo-dG. They proposed that the carbons of glyoxal are derived from the C-4′ and C-5′ positions of deoxyribose (10). This is consistent with our observation that glyoxal forms from phosphoglycolaldehyde, a product of 3′-hydrogen atom abstraction of the deoxyribose ring of DNA (2). Barton and co-workers demonstrated that photoreactive rhodium(III) complexes bind in the major groove of DNA and oxidize the 3′-position of deoxyribose to produce a 3′-phosphoglycolaldehyde-terminated fragment with the release of base propenoic acid (2). This product spectrum is analogous to that of the 3′-phosphoglycolic acid and base propenal arising from C4′-deoxyribose oxidation produced by enediyne and bleomycin antibiotics (reviewed in ref 26). Using o-benzyloxyhydroxylamine derivatization and GC/MS, we have also identified 3′-phosphoglycolaldehyde in DNA treated with the Fe(II)-EDTA complex (M. Awada and P. C. Dedon, manuscript in preparation). On the basis of deuterium kinetic isotope effects, Tullius and co-workers have confirmed that the Fe(II)-EDTA complex is capable of oxidizing the 3′-position of deoxyribose, though the reactivity of this position appears to be lower than the reactivities of the 4′- and 5′-positions that have greater solvent accessibility in double-stranded DNA (1). We conclude that 2-phosphoglycolaldehyde is capable of modifying DNA to form the 1,N2-glyoxal-dG adducts identical to those produced in a reaction between dG and glyoxal. Although ESR and pUC19 plasmid nicking experiments revealed that 2-phosphoglycolaldehyde generates a low level of hydroxyl radicals in solution, the insensitivity of the dG adduction reaction to oxygen and radical scavengers suggests that the mechanism of adduct formation proceeds mainly by an oxidation-independent pathway possibly involving a phosphatephosphonate rearrangement. The relevance of this reaction to DNA chemistry is currently being investigated using reagents that produce 3′-phosphoglycolaldehyde residues after 3′-hydrogen abstraction from DNA.

Acknowledgment. We thank Dr. John Wishnok for help with mass spectrometry (Mass Spectrometry Laboratory, Division of Bioengineering and Environmental Health, Massachusetts Institute of Technology), Drs. Jeff Simpson and Mark Wall for help with NMR and ESR experiments (Spectroscopy Laboratory, Department of Chemistry, Massachusetts Institute of Technology; supported in part by NIH Grant ISI0RR13886-01). This work was supported by NIH Grant GM59790.

References (1) Pogozelski, W. K., and Tullius, T. D. (1998) Oxidative strand scission of nucleic acids: Routes initiated by hydrogen abstraction from the sugar moiety. Chem. Rev. 98, 1098-1107. (2) Sitlani, A., Long, E. C., Pyle, M., and Barton, J. K. (1992) DNA photocleavage by phenanthrenequinone diimine complexes of rhodium(III): Shape-selective recognition and reaction. J. Am. Chem. Soc. 114, 2303-2312. (3) Dedon, P. C., Plastaras, J. P., Rouzer, C. A., and Marnett, L. J. (1998) Indirect mutagenesis by oxidative DNA damage: Formation of the pyrimidopurinone adduct of deoxyguanosine by base propenal. Proc. Natl. Acad. Sci. U.S.A. 95, 11113-11116.

Formation of the 1,N2-Glyoxal Adduct of dG (4) Gingipalli, L., and Dedon, P. C. (2001) Reaction of cis- and trans2-butene-1,4-dial with 2′-deoxycytidine to form stable oxadiazabicyclooctaimine adducts. J. Am. Chem. Soc. 123, 2664-2665. (5) Marnett, L. J. (1994) DNA adducts of R,β-unsaturated aldehydes and dicarbonyl compounds. IARC Sci. Publ. 125, 151-163. (6) Thornalley, P. J. (1998) Glutathione-dependent detoxification of R-oxoaldehydes by the glyoxalase system: involvement in disease mechanisms and antiproliferative activity of glyoxalase I inhibitors. Chem.-Biol. Interact. 111-112, 137-151. (7) Murata-Kamiya, N., Kamiya, H., Kaji, H., and Kasai, H. (1997) Glyoxal, a major product of DNA oxidation, induces mutations at G:C Sites on a shuttle vector plasmid replicated in mammalian cells. Nucleic Acids Res. 25, 1897-1902. (8) Murata-Kamiya, N., Kamiy, H., Kaji, H., and Kasai, H. (1997) Mutational specificity of glyoxal, a product of DNA oxidation, in the lacI gene of wild-type E. coli W3110. Mutat. Res. 377, 255262. (9) Murata-Kamiya, N., Kaji, H., and Kasai, H. (1997) Types of mutations induced by glyoxal, a major oxidative DNA-damage product, in Salmonella typhimuium. Mutat. Res. 377, 13-16. (10) Murata-Kamiya, N., Kamiya, H., Iwamoto, N., and Kasai, H. (1995) Formation of a mutagen, glyoxal, from DNA treated with oxygen free radicals. Carcinogenesis 16, 2251-2253. (11) LaMarr, W. A., Sandman, K. M., Reeve, J. N., and Dedon, P. C. (1997) Large scale preparation of positively supercoiled DNA using the archaeal histone HMf. Nucleic Acids Res. 25, 16601662. (12) Urata, H., and Akagi, M. (1993) A convenient synthesis of oligonucleotides with a 3′-phosphoglycolate and 3′-phosphoglycolaldehyde terminus. Tetrahedron Lett. 34, 4015-4018. (13) Lee, L. V., Poyner, R. R., Vu, M. V., and Cleland, W. W. (2000) Role of metal ions in the reaction catalyzed by L-ribulose-5phosphate 4-epimerase. Biochemistry 39, 4821-4830. (14) Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. G., Smith, J. A., and Struhl, K. (1989) Current Protocols in Molecular Biology, John Wiley and Sons, New York. (15) Kim, W., and Gates, K. S. (1997) Evidence for thiol-dependent production of oxygen radicals by 4-methyl-5-pyrazinyl-3H-1,2dithiole-3-thione (oltipraz) and 3H-1,2-dithiole-3-thione: possible relevance to the anticarcinogenic properties of 1,2-dithiole-3thiones. Chem. Res. Toxicol. 10, 296-301. (16) Loeppky, R. N., Cui, W., Goelzer, P., Park, M., and Ye, Q. (1999)

Chem. Res. Toxicol., Vol. 14, No. 9, 2001 1253

(17) (18)

(19) (20) (21) (22) (23)

(24)

(25)

(26)

Glyoxal-guanine DNA adducts: detection, stability and formation in vivo from nitrosamines. IARC Sci. Publ. 150, 155-168. Thornalley, P. J., and Stern, A. (1984) The production of free radicals during the autoxidation of monosaccharides by buffer ions. Carbohydr. Res. 134, 191-204. Thornalley, P., Wolff, S., Crabbe, J., and Stern, A. (1984) The autoxidation of glyceraldehyde and other simple monosaccharides under physiological conditions catalyzed by metal ions. Biochim. Biophys. Acta 797, 276-287. Wolff, S. P., Crabbe, M. J., and Thornalley, P. J. (1984) The autoxidation of glyceraldehyde and other simple monosaccharides. Experientia 40, 244-246. LaMarr, W. A., Sandman, K. M., Reeve, J. N., and Dedon, P. C. (1997) Differential effects of DNA supercoiling on radical-mediated DNA strand breaks. Chem. Res. Toxicol. 10, 1118-1122. Abordo, E. A., Minhas, H. S., and Thornalley, P. J. (1999) Accumulation of R-oxoaldehydes during oxidative stress: a role in cytotoxicity. Biochem. Pharmacol. 58, 641-648. Shapiro, R., and Hachmann, J. (1966) The reaction of guanine derivatives with 1,2-dicarbonyl compounds. Biochemistry 5, 27992807. Chung, F. L., and Hecht, S. S. (1985) Formation of the cyclic 1,N2glyoxal-deoxyguanosine adduct upon reaction of N-nitroso-2hydroxymorpholine with deoxyguanosine. Carcinogenesis 6, 16711673. Mikroyannidis, J. A., Tsolis, A. K., and Gourghiotis, D. J. (1982) Synthesis and chemical properties of substituted 2-hydroxy-2phosphonylethanals and 1,2-dihydroxy-1,2-bisphosphonylethanes. Phosphorus Sulfur Relat. Elem. 13, 279-289. Ramirez, F., Glaser, S. L., Bigler, A. J., and Pilot, J. F. (1969) Synthesis of sugar-like phosphates by the oxyphosphorane condensation. Reaction of glyoxal with trialkyl phosphites and preparation of phosphate esters of glycolaldehyde, R-hydroxy-βketo aldehydes, and hydroxymalonaldehyde chloride. J. Am. Chem. Soc. 91, 496-500. Dedon, P. C., and Goldberg, I. H. (1992) Free-radical mechanisms involved in the formation of sequence-dependent bistranded DNA lesions by the antitumor antibiotics bleomycin, neocarzinostatin, and calicheamicin. Chem. Res. Toxicol. 5, 311-332.

TX0155092