High-Performance Liquid

Pharmacology, University of Arkansas for Medical Sciences, Little Rock, Arkansas 72205 ... The two pairs of HNE-derived diastereomeric 1,N2-propanodeo...
4 downloads 0 Views 179KB Size
Download PDF
Chem. Res. Toxicol. 1997, 10, 1259-1265

1259

Synthesis and 32P-Postlabeling/High-Performance Liquid Chromatography Separation of Diastereomeric 1,N2-(1,3-Propano)-2′-deoxyguanosine 3′-Phosphate Adducts Formed from 4-Hydroxy-2-nonenal Ping Yi,† DeJin Zhan,† Victor M. Samokyszyn,‡ Daniel R. Doerge,† and Peter P. Fu*,† National Center for Toxicological Research, Jefferson, Arkansas 72079, and Department of Pharmacology, University of Arkansas for Medical Sciences, Little Rock, Arkansas 72205 Received June 11, 1997X

4-Hydroxy-2-nonenal (HNE), a major electrophilic byproduct of lipid peroxidation, is mutagenic and cytotoxic. The two pairs of HNE-derived diastereomeric 1,N2-propanodeoxyguanosine 3′-monophosphate adducts were synthesized from reaction of HNE with 2′deoxyguanosine 3′-monophosphate. After HPLC separation, these adducts were characterized by UV-visible absorption and negative ion electrospray ionization MS/MS analysis. To further characterize the structures, these adducts were dephosphorylated to the corresponding HNEmodified deoxyguanosine adducts and their HPLC retention times and UV spectra were compared with those of the synthetic standards prepared from reaction of HNE with 2′-deoxyguanosine. Separation of these adducts by 32P-postlabeling/HPLC was developed. Reaction of HNE with calf thymus DNA resulted in only one pair of diastereomeric adducts, with one adduct predominantly formed with a modification level of 1.2 ( 0.5 adducts/107 nucleotides.

Introduction It has now been established that chemical carcinogens exert their tumorigenic activity through a genotoxic or a nongenotoxic mechanism (1). One major nongenotoxic mechanism is induction of lipid peroxidation and subsequent formation of endogenous DNA adducts (2-13). Aerobic oxidation of lipid can be initiated as the result of metabolism of endogenous or exogenous substrates (14-17). Lipid peroxidation results in the formation of volatile aldehydic byproducts, such as malondialdehyde (MDA),1 formaldehyde, acetaldehyde, acrolein, crotonaldehyde, and 4-hydroxy-2-nonenal (HNE). A number of these products are highly cytotoxic (14-18), cause DNA strand breaks (18, 19) and DNA-protein cross-linking (3), form endogenous DNA adducts (2, 4-12, 18, 20-26), and induce mutation in bacterial systems and tumors in experimental animals (6, 17, 27-31). The endogenous DNA adducts formed from MDA (6, 9, 12, 13), formaldehyde, acrolein (8, 22, 23), crotonaldehyde (8, 21-24), 4-hydroxy-2-hexenal (2, 4, 5, 20), and HNE (4, 5, 20) have been reported. MDA modification of DNA results in the formation of the DNA adduct 3-(2′-deoxy-β-D-erythropentofuranosyl)pyrimido[1,2-a]purin-10(3H)-one. This adduct has been detected in DNA isolated from human liver, breast, and white blood cells (6, 12), and is a * To whom correspondence should be addressed. Tel: (870) 5437207. Fax: (870) 543-7136. E-mail: [email protected]. † National Center for Toxicological Research. ‡ University of Arkansas for Medical Sciences. X Abstract published in Advance ACS Abstracts, October 15, 1997. 1 Abbreviations: HNE, 4-hydroxy-2-nonenal; HNE-adducts 1 and 2, 3-(2′-deoxy-β-D-erythro-pentofuranosyl)-5,6,7,8-tetrahydro-(8R)-hydroxy-(6S)-[1(R and S)-hydroxyhexanyl]pyramido[1,2-a]purin-10(3H)one; HNE-adduct 3, 3-(2′-deoxy-β-D-erythro-pentofuranosyl)-5,6,7,8tetrahydro-(8S)-hydroxy-(6R)-[1(R or S)-hydroxyhexanyl]pyramido[1,2a]purin-10(3H)-one; HNE-adduct 4, 3-(2′-deoxy-β-D-erythro-pentofuranosyl)-5,6,7,8-tetrahydro-(8S)-hydroxy-(6R)-[1(S or R)-hydroxyhexanyl]pyramido[1,2-a]purin-10(3H)-one; MDA, malondialdehyde.

S0893-228x(97)00100-8 CCC: $14.00

Figure 1. Structures and abbreviations of the four diastereomeric 1,N2-propanodeoxyguanosine adducts generated by reaction of HNE with deoxyguanosine. Since the absolute configurations of these four diastereomeric adducts have not been determined, the two structures shown in the figure may be reversed.

promutagenic lesion in Escherichia coli. (32). Thus, formation of this adduct may increase the risk of cancer. HNE, an R,β-unsaturated aldehyde, has been found in both rodent and human tissues and is the most cytotoxic lipid peroxidation product (2). Consequently HNE is suspected to be associated with a wide variety of toxicological effects (17-19, 33-38), and both HNE and MDA are the lipid peroxidation products that have been mostly studied in human tissues. To determine whether HNE causes genotoxic effects, including mutagenicity and tumorigenicity, it is necessary to develop methodologies for detection of HNE adducts formed with nuclear DNA. Winter et al. reported that Michael addition of HNE to deoxyguanosine resulted in the formation of two pairs of diastereomeric 1,N2-propanodeoxyguanosine adducts (20) (Figure 1). Alternatively, HNE can first be enzymatically epoxidized to form the oxirane 2,3-epoxy-HNE metabolite, which then binds to DNA (2, 4, 5). Chung and coworkers have shown that binding of 2,3-epoxy-HNE to DNA generates a 1,N2-ethenodeoxyguanosine adduct (4). © 1997 American Chemical Society

1260 Chem. Res. Toxicol., Vol. 10, No. 11, 1997

Under certain experimental conditions, both oxiranedependent and Michael addition pathways occur and 1,N2-ethenodeoxyguanosine and 1,N2-propanodeoxyguanosine adducts are formed (4). Douki and Ames have reported the development of an HPLC-electrochemical detection assay for detection of the diastereomeric 1,N2propanodeoxyguanosine adducts formed from HNE and other R,β-unsaturated carbonyls (7). The 32P-postlabeling technique has been shown to be one of the most sensitive analytical methodologies for detection of carcinogen-modified DNA adducts. However, at present, detection of HNE-derived 1,N2-propanodeoxyguanosine using 32P-postlabeling techniques has not been reported. In this paper, we report the synthesis of the diastereomeric 1,N2-propanodeoxyguanosine 3′-monophosphate adducts from reactions of HNE with calf thymus DNA and with 2′-deoxyguanosine 3′-monophosphate and the separation of these adducts by 32Ppostlabeling/HPLC.

Materials and Methods Caution: HNE has been determined to be highly cytotoxic and mutagenic in Salmonella. Therefore, appropriate safety procedures should be followed when working with this compound. Materials. Amberlyst-15 was purchased from Aldrich Chemical Co. (Milwaukee, WI). 2′-Deoxyguanosine 3′-phosphate (sodium salt) (3′-dGMP), proteinase K, ribonuclease A type XIIA, micrococcal nuclease, nuclease P1, spleen phosphodiesterase, and calf thymus DNA (sodium salt, type I) were purchased from Sigma Chemical Co. (St. Louis, MO). Polynucleotide kinase (T4 PNK) was purchased from U.S. Biochemical Corp. (Cleveland, OH). All solvents used were HPLC grade. HNE-dimethyl acetal was prepared according to the procedure of Gree et al. (39). Synthesis of HNE-Derived 1,N2-Propanodeoxyguanosine Adducts. A solution of HNE-dimethyl acetal (50 mg, 250 µmol) in 1 mL of acetone and 15 µL of H2O was stirred with 10 mg of Amberlyst-15 at ambient temperature for 90 min. The resulting HNE was reacted with 50 mg (200 µmol) of 2′deoxyguanosine in 10 mL of phosphate buffer (25 mM, pH 7.5) with stirring at 37 °C for 48 h. After cooling, the mixture was filtered through a 0.45-µm filter and extracted twice with chloroform to remove the unreacted HNE, and the solvent was removed under reduced pressure. The residue was redissolved in methanol/water (20/80, v/v) for HPLC separation using a Waters µBondapack C18 (Waters Millipore, Milford, MA) column (250 × 10 mm), with 20% methanol in water at 2 mL/min to remove the unmodified deoxyguanosine and then with 50% methanol in water to collect the HNE-adducts. The HNEadduct fractions were further purified using the same column with a 40-min linear gradient of 25-50% methanol in water at a flow rate of 2 mL/min. Chemical Reaction of 3′-dGMP with HNE and Adduct Identification. HNE (40 mg, 0.25 mmol), obtained from hydrolysis of HNE-dimethyl acetal, was dissolved in 1 mL of acetone and reacted with 50 mg (0.144 mmol) of 3′-dGMP in 10 mL of 25 mM phosphate buffer (pH 7.5) at 37 °C for 48 h. The mixture was filtered and extracted twice with chloroform to remove the unreacted HNE. After removal of solvent under reduced pressure, the residue was redissolved in water for HPLC purification using a Waters µBondapack C18 column (250 × 10 mm). The column was first eluted with water at 2 mL/ min, to remove the unmodified 3′-dGMP, and then eluted isocratically with 30% methanol in water to collect the fractions containing HNE-modified 3′-dGMP adducts. The chromatographic peaks containing the HNE-modified adducts were further purified using the same column along with an 80-min linear gradient of water to 40% methanol in water at a flow rate of 2 mL/min. For identification of the HNE-modified 3′-dGMP adducts, each of the isolated adducts (about 20 nmol) in 30 µL of 50 mM

Yi et al. phosphate buffer (pH 7.0) was dephosphorylated yielding the corresponding HNE-deoxyguanosine adduct by incubation with 200 U of alkaline phosphatase (Sigma, type Vl) at 37 °C for 40 min. Chemical Reaction of DNA with HNE. A solution of HNE, obtained from hydrolysis of HNE-dimethyl acetal (13 mg, 65 µmol) in 300 µL of acetone, purified calf thymus DNA (55 mg) in 30 mL of 25 mM phosphate buffer (pH 7.5), and 1 drop of toluene was incubated at 37 °C for 3 days. After incubation, the incubation mixture was ice-cooled and extracted with 30 mL of chloroform twice. The DNA was precipitated by the addition of 60 mL of ice-cold ethanol and 300 µL of 5 N NaCl, washed with 70% ethanol, and redissolved in water. The DNA adducts were then analyzed by 32P-postlabeling/HPLC. 32P-Postlabeling/TLC and 32P-Postlabeling/HPLC Separation of the Diastereomeric 1,N2-Propanodeoxyguanosine [γ-32P]-3′,5′-Bisphosphate Adducts. The modified DNA (20 µg, 1 µg/µL) in water, from reaction of HNE with calf thymus DNA, was hydrolyzed to the corresponding deoxyribonucleoside 3′-monophosphates at 37 °C for 3 h with 80 µL of a mixture containing micrococcal nuclease (1.2 units) and spleen phosphodiesterase (0.2 unit) in 20 mM sodium succinate and 10 mM calcium chloride (pH 6.0) and treated with nuclease P1 (10 µg) at 37 °C for 30 min following the procedure of Reddy and Randerath (40). Each of the HNE-deoxyguanosine 3′-monophosphate adducts, either from enzymatic digestion of DNA as described above or from direct synthesis, was dissolved in 10 µL of 10 mM BisTris buffer (pH 7.1) and γ-32P-phosphorylated with 10 µL of PNK mix containing 200 µCi of [γ-32P]ATP (sp act. 3000-4000 Ci/ mmol), 10 U of T4 polynucleotide kinase, and 2 µL of 10× kinase buffer (200 mM bicine, pH 9.0; 100 mM dithiothreitol; 100 mM magnesium chloride; 11 mM spermidine, in distilled water) by incubating at 37 °C for 40 min. The residual [γ-32P]ATP was destroyed by incubation with 5 mU of potato apyrase at 37 °C for 30 min. (1) 32P-Postlabeling/TLC Separation. The resulting [γ-32P]3′,5′-bisphosphate adducts were each applied to a prewashed 20- × 20-cm PEI cellulose TLC plate (Macherey-Nagel, Duren, Germany). For D1 separation, the plate was developed in 3.5 M ammonium formate (pH 3.5) to the top of the plate. The upper one-fourth part of the plate was then excised, and the remaining plate (15 cm) was developed (D2 direction) in 0.5 M Tris-HCl (pH 8.0), 4.5 M urea, and 0.8 M LiCl. After drying, the plate was autoradiographed with intensifying screen using DuPont Cronex films for 2 h, and radioactivity of the spots on the PEI plate was quantitated by Cerenkov counting. The relative adduct levels (RAL) were calculated following Reddy’s procedure (40). (2) 32P-Postlabeling/HPLC Separation. The [γ-32P]-3′,5′bisphosphate adducts and the adduct mixture were each separated by HPLC on a combination of two Burdick and Jackson Cl8 reversed-phase columns (4.6 × 250 mm) eluted at 0.6 mL/min with solvent system as follows: a 40-min linear gradient from 3% to 7% methanol in 50 mM NaH2PO4, pH 4.5; a 30-min linear gradient from 7% to 35% methanol in 50 mM NaH2PO4, pH 4.5; then isocratically for an additional 60 min. To avoid interference by the high radioactivity of the free 32Pphosphate which eluted at ca. 7 min, the FLO-ONE radioactivity detector (Radiomatic Instruments, Tampa, FL) was turned on at 10 min after each injection was started. The scintillation fluid flow rate was 1.8 mL/min. Instrumentation. HPLC analysis was performed with a Waters Associates, Inc. instrument consisting of two model 510 pumps, a model 660 solvent programmer, a model U6K injector, and a Hewlett-Packard 1040A detection system. For radiochromatography analysis, a Radiomatic FLO-ONE\Beta A-515 system in line with the HPLC system was used, or a model 328 Golden Retriever fraction collector (Isco Inc., Lincoln, NE) was used to collect HPLC eluent at 30-s or 1-min intervals, and the radioactivity was measured with a Searle Analytic Mark III model 6881 liquid scintillation counter. UV spectra were determined with a Beckman model DU-65 spectrophotometer.

4-Hydroxy-2-nonenal-Modified DNA Adducts

Chem. Res. Toxicol., Vol. 10, No. 11, 1997 1261

1H

NMR spectra were recorded at 500.13 MHz and at approximately 28 °C on a Bruker AM500 spectrometer (Bruker Instruments, Billerica, MA). The samples were dissolved in methanol-d4. DMSO-d6 was also used as solvent in one case, and D2O was used as solvent in separate experiments. Chemical shifts are reported in ppm downfield from TMS by assigning the residual proton signal of the solvent to 3.30 and 2.49 ppm for methanol-d4 and DMSO-d6, respectively. No data using D2O as solvent is reported. General conditions for data acquisition and processing have previously been reported (41). The data point resolution was 0.215 Hz/point after zero-filling. Chemical shifts and coupling constants of the resonances of adducts 1&3 and 3 were determined as a difference (if any) from those of adduct 4 from interactive computer graphics displays. Coupling constants for resonances H1-H4 were determined from a 5-spin computer simulation using the Bruker program PANIC. Other measurements were first-order. Mass spectra were obtained using a Quattro II triple quadrupole mass spectrometer (Micromass, Altrincham, U.K.) equipped with an electrospray ionization interface from continuous infusion of solutions (ca. 5 µg/mL) in 20% aqueous methanol containing 1 mM ammonium acetate. Product ion scans were collected over the range of m/z 10-550 (1-s cycle time), using a sampling cone potential of 50 V, argon as the collision gas at a pressure of 4 × 10-3 mbar, and collision energies of 12-50 eV. Alternatively, on-line LC/MS was performed using full scans for positive ions (m/z 100-600 with a cycle time of 1 s). A Nucleosil C18 column (2 × 250 mm; Keystone Scientific, Bellfonte, PA) was eluted using a mobile phase of 50% aqueous methanol at a flow rate of 0.2 mL/min. The entire HPLC column effluent was delivered into the atmospheric pressure ion source.

Results and Discussion Synthesis of HNE-Derived 1,N2-Propanodeoxyguanosine Adducts. In this paper, we report the synthesis of diastereomeric 1,N2-propanodeoxyguanosine 3′-monophosphate adducts from reaction of HNE with 2′deoxyguanosine 3′-monophosphate and then the separation of these adducts by 32P-postlabeling/HPLC. To assist in structural assignment of the HNE-derived diastereomeric 1,N2-propanodeoxyguanosine 3′-monophosphate adducts, as a first step, the known HNE-modified 1,N2propanodeoxyguanosine adducts were prepared as standards by reaction of HNE with 2′-deoxyguanosine according to the procedure of Winter et al. (20) with modification. The reaction mixture was purified by reversed-phase HPLC (Figure 2). The chromatographic peaks eluting at 26.5-27 min (assigned as peak 1&2), 30.5 min (peak 3), and 32.2 min (peak 4) had similar UVvisible absorption (Figure 3) and mass spectra (not shown). On-line electrospray ionization LC/MS was used to generate positive ion mass spectra, which demonstrated ions corresponding to M + H+ ) m/z 424, M + Na+ ) m/z 446, M + K+ ) m/z 462, BH2+ ) m/z 308, and BH + Na+ ) m/z 330 for all peaks (not shown). These assignments are consistent with those derived from FAB/ MS previously reported by Winter et al. (20). This suggested that these adducts were the HNE-derived diastereomeric 1,N2-propanodeoxyguanosine adducts. The structures were further characterized by 1H NMR spectral analysis (Table 1) and by comparison of the 1H NMR data with those previously published by Winter et al. (20). It was readily apparent that each adduct contained a 2′deoxyribose moiety. Peak 1&2 appeared to exist as a mixture of two diastereomers of unequal population (adducts 1&2). Because of the very small quantity of material and the spectral complexity, no attempt was made to determine the NMR spectral parameters of the apparent minor component. The major component will

Figure 2. HPLC profiles of the diastereomeric 1,N2-propanodeoxyguanosine adducts formed from (A) chemical reaction of HNE with deoxyguanosine and dephosphorylation of each of the isolated diastereomeric 1,N2-propanodeoxyguanosine 3′-monophosphate adducts labeled as the (B) HNE-adducts 1&2, (C) HNE-adduct 3, and (D) HNE-adduct 4 as shown in Figure 4.

Figure 3. UV-visible absorption spectra of the diastereomeric 1,N2-propanodeoxyguanosine adducts labeled as the HNEadducts 1&2 (s) HNE-adduct 3 (- - -), and HNE-adduct 4 (‚‚‚) as shown in Figure 2.

be arbitrarily referred to here as adduct 1 and the minor component as adduct 2. A complete and accurate set of

1262 Chem. Res. Toxicol., Vol. 10, No. 11, 1997

Yi et al.

Table 1. Proton Magnetic Resonance Spectral Data of DNA Adducts from Reaction of 4-Hydroxy-2-nonenal with Deoxyguanosinea (Chemical shift in ppm) assignment

adduct 1b

adduct 3

adduct 4f

1 2a 2b 3 4 5-8 9 G8 1′ 2′ 2′′ 3′ 4′ 5′ e 5′′ e

6.415 1.736 2.108 3.74c NDd NDd 0.926 7.907 6.238 2.687 2.326 4.515 3.980 3.776 3.721

6.400 1.603 2.185 3.612 3.464 1.28-1.45 0.926 7.914 6.233 2.669 2.336 4.510 3.986 3.779 3.707

6.400 1.602 2.185 3.612 3.464 1.28-1.45 0.926 7.916 6.233 2.678 2.331 4.513 3.989 3.779 3.710

Figure 4. HPLC profile of the diastereomeric 1,N2-propanodeoxyguanosine 3′-monophosphate adducts formed by reaction of HNE with 2′-deoxyguanosine 3′-monophosphate.

a Spectral parameters are reported with methanol-d as solvent, 4 except for footnote f. Selected coupling constants (J) in Hz for adduct 4 are as follows: J1-2a, 2.9; J1-2b, 2.7; J2a-2b, 13.5; J2a-3, 12.5; J2b-3, 3.8; J3-4, 7.5; J1′-2′, 7.7; J1′-2′′, 6.0. The corresponding coupling constants for adducts 1 and 3 were indistinguishable from those in adduct 4, where measurable. b Sample was a mixture. Chemical shifts of main species are reported. c Estimate based on homonuclear decoupling. d Not determined. e No distinction between methylene protons is implied. f Selected chemical shifts were also measured and assigned with DMSO-d6 as solvent. The chemical shifts (δ) in ppm are as follows: δ 1-OH, 6.64; N-H, 7.07; 3′-OH, 5.24; 5′-OH, 4.95. The assignment of the N-H resonance is tentative.

carbon-proton coupling constants for the propanocontaining ring of adducts 1, 3, and 4 was determined with the aid of spectral simulation. The data (footnote a in Table 1) are in agreement with the general trends previously reported or predicted for substitution of the general type shown in Figure 1, including the trans orientation between the C1 and C3 substituents (20, 42). A comparison with previously reported NMR data is interesting. Winter et al. (20) have reported chemical shifts for the corresponding dG adducts of HNE in DMSO-d6 (20). Most of our work has been done using methanol-d4 as solvent. It is apparent however that their adducts 3 and 4 correspond to our adducts 3 and 4, except that there is a difference in the resonance assignments of the 1-OH and N-H protons when DMSO-d6 is used as solvent. These resonance assignments were previously used for part of the assignment of configurations. Based on analysis of results reported using achiral trans-2-hexenal (20), it is likely that the difference between adducts 3 and 4 is due to the asymmetric C4 atom, which arises from our use of racemic HNE. There are insufficient data to propose the specific structural difference between diastereomers 1 and 2 compared to diastereomers 3 and 4. The chemical shift differences between adducts 3 and 4 are small but significant, particularly for the H2′ and H2′′ resonances of the dG moiety. Demonstration of these NMR spectral differences is novel and provides further support for the presence of diastereomers. In so far as the chemical shift of H2′ (and H2′′) is dependent on the time-averaged glycosyl torsion angle (ref 43 and references therein), it is possible that these adducts differ in syn/anti population or in glycosyl torsion angle within these domains. Although such a difference must be small, it could have biological significance. Chemical shift data for adducts of HNE using D2O as solvent have previously been reported (7). We have dissolved our adducts in D2O for comparison but found

Figure 5. UV-visible absorption spectra of the diastereomeric 1,N2-propanodeoxyguanosine 3′-monophosphate adducts formed by reaction of HNE with 2′-deoxyguanosine 3′-monophosphate.

little agreement in the NMR spectral parameters with variation of up to (0.4 ppm. The reason for the discrepancy was not investigated. The first two adducts showed an unresolved chromatographic peak eluting at 26.5-27 min (Figure 2). Winter et al. reported a slightly better HPLC separation and assigned them as a pair of diastereomeric 1,N2-propanodeoxyguanosine adducts (20). Thus, based on the mass and NMR spectral analysis, it is concluded that the four diastereomeric 1,N2-propanodeoxyguanosine adducts were present. Synthesis of 1,N2-propanodeoxyguanosine 3′Monophosphate Adducts from HNE. After reaction of HNE with 3′-dGMP, the 3′-monophosphate adducts were separated by HPLC (Figure 4). The chromatographic peaks, which all had similar UV-visible absorption spectra (Figure 5), eluting at 55.7, 56.5, 60.0, and 63.0 min were collected for further structural identification. The peaks were analyzed using negative ion electrospray ionization MS/MS (see Figure 6), and similar spectra were observed for the different peaks. The product ion spectrum contained ions corresponding to the [M - H]- at m/z 502, the m/z 458 ion tentatively assigned as resulting from loss of an allyl alcohol moiety from the propano ring, GMP- at m/z 346, the deoxyribose moiety at m/z 195, the G- ion at m/z 150, and ions corresponding to H2PO4- and PO3- at m/z 97 and 79, respectively. These results suggested that these peaks consisted of the diastereomeric 1,N2-propanodeoxyguanosine 3′-monophosphate adducts.

4-Hydroxy-2-nonenal-Modified DNA Adducts

Chem. Res. Toxicol., Vol. 10, No. 11, 1997 1263

Figure 6. Negative ion electrospray ionization mass spectrum of the diastereomeric 1,N2-propanodeoxyguanosine 3′-monophosphate adducts formed by reaction of HNE with 2′-deoxyguanosine 3′-monophosphate.

The structures of these 3′-monophosphate adducts were further characterized by dephosphorylation to the corresponding HNE-modified deoxyguanosine adducts and comparison of their HPLC retention times and UV spectra with those of the synthetic standards (Figures 2 and 3). Thus, the 3′-monophosphate adducts contained in peaks 1&2, 3, and 4 were dephosphorylated to the corresponding deoxyguanosine adducts, and the HPLC elution profiles are shown in Figure 2B-D, respectively. These chromatographic peaks matched with the deoxyguanosine adduct standards labeled 1-4 in Figure 1, respectively. Therefore, based on the mass spectra of these 3′-monophosphate adducts (Figure 6) and the identical retention times of the dephosphorylated products and the HNE-deoxyguanosine adducts (Figure 2), it was determined that two pairs of diastereomeric 1,N2-propanodeoxyguanosine 3′-monophosphate adducts were formed from reaction of HNE with 2′-deoxyguanosine 3′-monophosphate. 32P-Postlabeling/TLC and 32P-Postlabeling/HPLC Separation of the Diastereomeric 1,N2-Propanodeoxyguanosine [γ-32P]-3′,5′-Bisphosphate Adducts. After phosphorylation of the 3′-monophosphate 1,N2propanodeoxyguanosine adducts with [γ-32P]ATP, the resulting 1,N2-propanodeoxyguanosine [γ-32P]-3′,5′-bisphosphate adducts were separated by TLC. The autoradiograms of 32P-postlabeling/TLC separation of the 1,N2-propanodeoxyguanosine 3′-monophosphates contained in peaks 1&2, 3, and 4 of the HPLC profile (Figure 4) are shown in Figure 7. Each resulted in a single TLC spot, but all spots exhibited similar chromatographic

Figure 7. Autoradiograms of TLC separation of 3′,5′-bisphosphates resulting from 5′-32P phosphorylation of (A) 2′-deoxyguanosine 3′-monophosphate and the diastereomeric 1,N2propanodeoxyguanosine 3′-monophosphates contained in (B) peaks 1&2, (C) peak 3, and (D) peak 4 of the HPLC profile shown in Figure 4.

behavior. Thus, 32P-postlabeling/TLC of a mixture of these 1,N2-propanodeoxyguanosine [γ-32P]-3′,5′-bisphosphate adducts resulted in a single large spot (data not shown). Therefore, 32P-postlabeling/TLC could not be utilized for resolution of HNE-derived 1,N2-propanodeoxyguanosine [γ-32P]-3′,5′-bisphosphate adducts. We next attempted to resolve the 3′,5′-bisphosphate adducts by reversed-phase HPLC. As shown in Figure 8A, the adduct mixture can be separated into three chromatographic peaks with retention times at 97, 103,

1264 Chem. Res. Toxicol., Vol. 10, No. 11, 1997

Yi et al.

an HPLC-electrochemical detection assay for detection of HNE diastereomeric 1,N2-propanodeoxyguanosine adducts (7). Thus, these analytical methods can complement each other for detecting and confirming the formation of these adducts. However, since the developed 32Ppostlabeling/HPLC method requires quite a long running time for analysis of each HNE-modified DNA sample (Figure 8), it warrants further study on the development of a more efficient HPLC separation.

Conclusion

Figure 8. Reversed-phase HPLC profiles of [32P]-3′,5′-bisphosphates resulting from incubation of HNE with (A) 2′-deoxyguanosine-3′-phosphate and (B) calf thymus DNA. The HNEmodified calf thymus DNA was enzymatically hydrolyzed to 3′monophosphates, and the 3′-monophosphates 32P-phosphorylated using polynucleotide kinase and [32P]ATP. See Materials and Methods for experimental conditions.

and 106 min, respectively, by using a combination of two Burdick and Jackson C18 reversed-phase columns (4.6 × 250 mm). When the 3′,5′-bisphosphates derived from the purified 3′-monophosphate 1,N2-propanodeoxyguanosine adducts contained in peaks 1 and 2 of Figure 4 were analyzed, the HPLC retention time matched with the peak at 97 min (data not shown). Similarly, the 3′monophosphate 1,N2-propanodeoxyguanosine adducts from peaks 3 and 4 of Figure 4 had retention times at 103 and 106 min, respectively, which were identical with those shown in Figure 8A. Consequently, these results indicate that 32P-postlabeling/HPLC can resolve the diastereomeric 1,N2-propanodeoxyguanosine [γ-32P]-3′,5′bisphosphate adducts derived from HNE. When only one Burdick and Jackson C18 reversedphase column was used, the adduct mixture also resolved into three chromatographic peaks; however, peaks 3 and 4 have no baseline separation (data not shown). The use of several other HPLC columns, including the Vydac 219 diphenyl column (250- × 4.6-mm i.d.), resulted in lesssatisfactory separation (data not shown). The HNE-modified DNA adducts obtained from reaction of HNE and calf thymus DNA were similarly analyzed by 32P-postlabeling/HPLC. As shown in Figure 8B, only adducts 3 and 4 were formed, with adduct 4 as the predominant adduct. Based on three measurements, the RAL value of the MDA-modified adduct 4 described above is 1.2 ( 0.5 adducts/107 nucleotides. This adduct level gave a signal/noise ratio of about 50-100; therefore, it is estimated that the detection limit (s/n ) 3) is between 1.2 adducts/108 nucleotides and 5 adducts/109 nucleotides. This level of sensitivity suggests that this methodology should be useful for detection of this adduct in biological samples. 32 P-Postlabeling/HPLC has been employed to detect other lipid peroxidation-derived endogenous DNA adducts, such as MDA, acrolein, and crotonaldehyde (8, 2124, 44). Chung and co-workers have also utilized 32Ppostlabeling/HPLC to analyze the 1,N2-ethenodeoxyguanosine adduct which is derived from 2,3-epoxy-HNE (4). Therefore, our 32P-postlabeling/HPLC method will facilitate the analysis of several lipid peroxidationdependent endogenous DNA adducts by using the same methodology. Douki and Ames have reported the use of

The HNE-derived diastereomeric 1,N2-propanodeoxyguanosine 3′-monophosphate adducts were synthesized from reactions of HNE with 2′-deoxyguanosine 3′-monophosphate and calf thymus DNA. Separation of these adducts by 32P-postlabeling/HPLC was developed. While reaction of HNE with 2′-deoxyguanosine 3′-monophosphate produces four 1,N2-propanodeoxyguanosine adducts, reaction of HNE with calf thymus DNA results in only one predominant adduct and one minor adduct, which indicates that binding of HNE to DNA is highly stereoselective.

Acknowledgment. We thank Dr. Frederick E. Evans for performing the NMR measurements and Ms. Evelyn Willingham for assistance in preparing this manuscript. We also thank the excellent constructive opinions suggested by the reviewers for revision of this paper. This research was supported in part by an appointment (Y.P.) to the Postgraduate Research Program at the National Center for Toxicological Research administered by the Oak Ridge Institute for Science and Education through an interagency agreement between the U.S. Department of Energy and the U.S. Food and Drug Administration.

References (1) Butterworth, B. (1990) Consideration of both genotoxic and nongenotoxic mechanisms in predicting carcinogenic potential. Mutat. Res. 239, 117-132. (2) Chen, H.-J. C., and Chung, F.-L. (1996) Epoxidation of trans-4hydroxy-2-nonenal by fatty acid hydroperoxides and hydrogen peroxide. Chem. Res. Toxicol. 9, 306-312. (3) Vaca, C. E., Wilhelm, J., and Harms-Ringdahl, M. (1988) Interaction of lipid peroxidation products with DNA. A review. Mutat. Res. 195, 137-149. (4) Sodum, R. S., and Chung, F.-L. (1988) 1,N2-Ethenodeoxyguanosine as a potential marker for DNA adduct formation by trans4-hydroxy-2-nonenal. Cancer Res. 48, 320-323. (5) Sodum, R. S., and Chung, F.-L. (1989) Structural characterization of adducts formed in the reaction of 2,3-epoxy-4-hydroxynonanal with deoxyguanosine. Chem. Res. Toxicol. 2, 23-28. (6) Chaudhary, A. K., Nokubo, M., Reddy, G. R., Yeola, S. N., Morrow, J. D., Blair, L. A., and Marnett, L. J. (1994) Detection of endogenous malondialdehyde-deoxyguanosine adducts in human liver. Science 265, 1580-1582. (7) Douki, T., and Ames, B. N. (1994) An HPLC-EC assay for 1,N2propano adducts of 2′-deoxyguanosine with 4-hydroxynonenal and other R,β-unsaturated aldehydes. Chem. Res. Toxicol. 7, 511-518. (8) Nath, R. G., and Chung, F.-L. (1994) Detection of exocyclic 1,N2propanodeoxyguanosine adducts as common DNA lesions in rodents and humans. Proc. Natl. Acad. Sci. U.S.A. 91, 7491-7495. (9) Wang, M.-Y., and Liehr, J. G. (1995) Induction by estrogens of lipid peroxidation and lipid peroxide-derived malonaldehyde-DNA adducts in male Syrian hamsters: role of lipid peroxidation in estrogen-induced kidney carcinogenesis. Carcinogenesis 16, 19411945. (10) Wang, M.-Y., and Liehr, J. G. (1995) Lipid hydroperoxide-induced endogenous DNA adducts in hamsters: Possible mechanism of lipid hydroperoxide-mediated carcinogenesis. Arch. Biochem. Biophys. 316, 38-46. (11) Li, D., Wang, M., Liehr, J. G., and Randerath, K. (1995) DNA adducts induced by lipid and lipid peroxidation products: Possible relationships to I-compounds. Mutat. Res. 344, 117-126.

4-Hydroxy-2-nonenal-Modified DNA Adducts (12) Vaca, C. E., Fang, J.-L., Mutanen, M., and Valsta, L. (1995) 32Ppostlabeling determination of DNA adducts of malonaldehyde in humans: Total white blood cells and breast tissue. Carcinogenesis 16, 1847-1851. (13) Ni, Y.-C., Kadlubar, F. F., and Fu, P. P. (1995) Formation of malondialdehyde-modified 2′-deoxyguanosinyl adduct from metabolism of chloral hydrate by mouse liver microsomes. Biochem. Biophys. Res. Commun. 216, 1110-1117. (14) Comporti, M. (1985) Biology of diseasesLipid peroxidation and cellular damage in toxic injury. Lab. Invest. 53, 599-623. (15) Esterbauer, H., and Cheeseman, H. (1987) Lipid peroxidation. II: Pathological implications. Chem. Phys. Lipids 45, 103-371. (16) Halliwell, B., and Gutteridge, J. M. C. (1989) Free Radicals in Biology and Medicine, Clarendon Press, Oxford. (17) Esterbauer, H., Zollner, H., and Schaur, R. J. (1990) Aldehydes formed by lipid peroxidation: Mechanisms of formation, occurrence, and determination. In Membrane Lipid Oxidation (VigoPelfrey, C., Ed.) pp 239-268, CRC Press, Boca Raton, FL. (18) Nishikawa, A., Sodum, R., and Chung, F.-L. (1992) Acute toxicity of trans-4-hydroxy-2-nonenal in Fisher 344 rats. Lipids 27, 5458. (19) Eckl, P. M., Ortner, A., and Esterbauer, H. (1993) Genotoxic properties of 4-hydroxyalkenals and analogous aldehydes. Mutat. Res. 290, 183-192. (20) Winter, C. K., Segall, H. J., and Haddon, W. F. (1986) Formation of cyclic adducts of deoxyguanosine with the aldehydes trans-4hydroxy-2-hexenal and trans-4-hydroxy-2-nonenal in vitro. Cancer Res. 46, 5682-5686. (21) Nath, R. G., Chen, H.-J. C., Nishikawa, A., Young-Sciame, R., and Chung, F.-L. (1984) A 32P-postlabeling method for simultaneous detection and quantification of exocyclic etheno and propano adducts in DNA. Carcinogenesis 15, 979-984. (22) Wilson, V. L., Foiles, P. G., Chung, F.-L., Povey, A. C., Frank, A. A., and Harris, C. C. (1991) Detection of acrolein and crotonaldehyde DNA adducts in cultured human cells and canine peripheral blood lymphocytes by 32P-postlabeling and nucleotide chromatography. Carcinogenesis 12, 1483-1490. (23) Smith, R. A., Williamson, D. S., Cerny, R. L., and Cohen, S. M. (1990) Detection of 1,N6-propanodeoxyadenosine in acroleinmodified polydeoxyadenylic acid and DNA by 32P-postlabeling. Cancer Res. 50, 3005-3012. (24) Chung, F.-L., Young, R., and Hecht, S. S. (1984) Formation of cyclic 1,N2-propanodeoxy-guanosine adducts in DNA upon reaction with acrolein or crotonaldehyde. Cancer Res. 44, 990-995. (25) Nair, J., Barbin, A., Guichard, Y., and Bartsch, H. (1995) 1,N6Ethenodeoxyadenosine and 3,N4-ethenodeoxycytidine in liver DNA from humans and untreated rodents detected by immunoaffinity/32P-postlabeling. Carcinogenesis 16, 613-617. (26) El Ghissassi, F., Barbin, A., Nair, J., and Bartsch, H. (1995) Formation of 1,N6-ethenodeoxyadenosine and 3,N4-ethenodeoxycytidine by lipid peroxidation products and nucleic acid bases. Chem. Res. Toxicol. 8, 278-283. (27) Feron, V. J., Til, H. P., de Vrijer, F., Woutersen, R. A., Cassee, F. R., and van Bladeren, P. J. (1991) Aldehydes: Occurrence, carcinogenic potential, mechanism of action and risk assessment. Mutat. Res. 259, 363-385. (28) Cajelli, E., Ferraris, A., and Brambilla, G. (1987) Mutagenicity of 4-hydroxynonenal in V79 Chinese hamster cells. Mutat. Res. 190, 169-171. (29) Marnett, L. J., Hurd, H. K., Hollstein, M. C., Levin, D. E., Esterbauer, H., and Ames, B. N. (1985) Naturally occurring carbonyl compounds are mutagens in Salmonella tester strain TA104. Mutat. Res. 148, 25-34.

Chem. Res. Toxicol., Vol. 10, No. 11, 1997 1265 (30) Chung, F.-L., Chen, H.-J. C., Guttenplan, J. B., Nishikawa, A., and Hard, G. C. (1993) 2,3-Epoxy-4-hydroxynonenal as a potential tumor-initiating agent of lipid peroxidation. Carcinogenesis 14, 2073-2077. (31) Moriya, M., Zhang, W., Johnson, F., and Grollman, A. P. (1994) Mutagenic potency of exocyclic DNA adducts: Marked differences between Escherichia coli and Simian kidney cells. Proc. Natl. Acad. Sci. U.S.A. 91, 11899-11903. (32) Benamira, M., Johnson, K., Chaudhary, A., Bruner, K., Tibbetts, C., and Marnett, L. J. (1995) Induction of mutations by replication malondialdehyde-modified M13 DNA in Escherichia coli: Determination of the extent of DNA modification, genetic requirements for mutagenesis, and types of mutations induced. Carcinogenesis 16, 93-99. (33) Esterbauer, H., Schaur, R. J., and Zollner, R. (1991) Chemistry and biochemistry of 4-hydroxynonenal, malondialdehyde and related aldehydes. Free Radical Biol. Med. 11, 81-128. (34) Benedetti, A., Barbieri, L., Ferrali, M., Casini, A. F., Fulceri, R., and Comporti, M. (1981) Inhibition of protein synthesis by carbonyl compounds (4-hydroxyalkenals) originating from the peroxidation of liver microsomal lipids. Chem.-Biol. Interact. 35, 331-340. (35) Lapenna, D., Mezzetti, A., Gioia, S. D., Pierdomenico, S. D., Daniele, F., and Cuccurullo, F. (1995) Plasma copper and lipid peroxidation in cigarette smokers. Free Radical Biol. Med. 19, 849-852. (36) Grune, T., Michel, P., Sitte, N., Eggert, W., Albrecht-Nebe, H., Esterbauer, H., and Siems, W. G. (1997) Increased levels of 4-hydroxynonenal modified proteins in plasma of children with autoimmune diseases. Free Radical Biol. Med. 23, 357-360. (37) Romero, F. J., Romero, M. J., Bosch-Morell, F., Martinez, M. C., Medina, P., and Lluch, S. (1997) 4-Hydroxynonenal-induced relaxation of human mesenteric arteries. Free Radical Biol. Med. 23, 521-523. (38) Benedetti, A., Comporti, M., and Esterbauer, H. (1980) Identification of 4-hydroxynonenal as a cytotoxic product originating from the peroxidation of liver microsomal lipids. Biochem. Biophys. Acta 620, 281-296. (39) Gree, R., Tourban, H., and Carrie, R. (1986) Fumaraldehyde monodimethyl acetal: An easily accessible and versatile intermediate. Tetrahedron Lett. 27, 4983-4986. (40) Reddy, M. V., and Randerath, K. (1986) Nuclease P1-mediated enhancement of sensitivity of 32P-postlabeling test for structurally diverse DNA adducts. Carcinogenesis 7, 1543-1551. (41) Evans, F. E., Deck, J., and Howard, P. C. (1994) Structure of phenolic isomers of 2- and 3-nitrofluoranthene studied by oneand two-dimensional 1H NMR spectroscopy. Comparative analysis of mutagenicity. Chem. Res. Toxicol. 7, 352-357. (42) Eder, E., Hoffman, C., and Deininger, C. (1991) Identification and characterization of deoxyguanosine adducts of methyl vinyl ketone and ethyl vinyl ketone. Genotoxicity of the ketones in the SOS chromotest. Chem. Res. Toxicol. 4, 50-57. (43) Evans, F. E., Miller, D. W., and Beland, F. A. (1980) Sensity of the conformation of deoxyguanosine to binding at the C-8 position by N-acetylated and unacetylated 2-aminofluorene. Carcinogenesis 1, 955-959. (44) Fang, J.-L., Vaca, C. E., Valsta, L. M., and Mutanen, M. (1996) Determination of DNA adducts of malonaldehyde in human: effects of dietary fatty acid composition. Carcinogenesis 17, 1035-1040.

TX970100R